Virus Research 102 (2004) 1–2
Preface
In the last 2 years, we have witnessed the implementation
of a new technology that promises to revolutionize the way
post-genomic research in the mammalian cells is conducted,
and the field of virology will not lag behind in this revolution.
The new technology is based on a phenomenon known as
RNA interference (RNAi), RNA silencing, or postranscrip-
tional gene silencing, which is an evolutionarily conserved
mechanism found to operate in fungi, plants, and animals.
RNAi is a process that responds to double-stranded RNA
(dsRNA) by silencing the expression of the gene homolo-
gous to the dsRNA sequence.
RNAi is triggered in plants and invertebrates by long
dsRNA molecules which are processed by a dsRNA-specific
RNAse, named Dicer, into small fragments of 20–25 base
pairs known as short interfering RNAs (siRNAs). These
siRNAs selectively associate with a multiprotein complex
(RISC, for RNA-induced silencing complex), which un-
winds the siRNA duplex to produce a single-stranded RNA,
which further guides RISC to the target mRNA. This mRNA
is then cleaved by a ribonuclease component of RISC, silenc-
ing the expression of the cognate gene. SiRNAs can also be
generated endogenously from precursor hairpin RNA struc-
tures 60–90 nucleotides in length, encoded in the genome.
Those are processed by Dicer to yield single-stranded RNAs
that bind to RISC and arrest the translation of the target
mRNA without promoting its degradation. The small regu-
latory endogenous RNAs which act in this way and of which
hundreds have been described in animals and plants, are
known as microRNAs (miRNAs). Their sequences are not
absolutely complementary to the target mRNAs, as in the
case of the siRNAs, but contain mismatches.
The RNAi pathway was more difficult to be demonstrated
in mammalian cells in which the dsRNA fragments longer
than 30 bp activate components of the non-specific interferon
response, that mediates a generalized inhibition of gene ex-
pression through the non-specific shutdown of protein syn-
thesis and mRNA degradation. The recent observation that
synthetic 21-nt siRNA duplexes effectively inhibit mam-
malian endogenous gene expression in a sequence-specific
manner without activating the non-specific interferon re-
sponse, offered a great opportunity to study the func-
tion of vertebrate genes, and the genes of vertebrate
viruses.
RNAi has been proposed to play a central role in the reg-
ulation of many cellular processes at the translation level
0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.008
including transposon mobilization, silencing of transgenes,
regulation of some of the early stages of development, as
well as in the regulation of expression at the genome level.
In plants RNAi represents a defense mechanism against vi-
ral infections, and it might also have the same role in inver-
tebrate organisms. It is unclear whether RNAi serves as an
antiviral defense mechanism in mammals or not. However,
the observation that this system is functional in vertebrates
prompted several groups to explore the interaction between
virus infection and the RNA silencing machinery in mam-
malian cells.
The application of this technology in the field of animal
virology is just the beginning, but it has already proven
to constitute an insightful method for the characterization
of viral gene function. RNAi will be especially helpful
to study those viruses for which a reverse genetics sys-
tem has not been implemented. It is foreseeable that the
studies employing RNAi will soon extend to the analysis
of cellular genes that interact with the viral components
of virus replication cycles, and, more in the mid term,
to the analysis of viral gene function in complete animal
models.
There are also high hopes for applications of RNAi in
the treatment of viral diseases, as therapeutic or prophylac-
tic tools. This may be particularly relevant in the case of
viruses that establish chronic infections (e.g., human im-
munodeficiency virus, hepatitis B and C viruses) or latent in-
fections (e.g., herpes viruses). The main challenge for these
applications is the development of long-lasting, efficient and
tissue-specific methods to deliver the siRNAs to whole or-
ganisms.
The opening chapter reviews the role of siRNAs and
miRNAs in RNAi. The emphasis of this Special Issue is
on reviews which describes the effectiveness of RNAi to
silence gene expression in animal viruses. These include
viruses with different genome replication strategies, such
as positive, negative and double-stranded RNA viruses as
well as retroviruses (Chapters 2–8). Additional chapters
review RNAi applied to insect and plant virus replication
(Chapters 9–11), and two chapters deal with suppres-
sors of RNA silencing (Chapter 12) and application of
RNAi as antivirals in plants and animals (Chapter 13).
The last chapter (Chapter 14) outlines the strategies to
generate an RNA expression library to analyse a whole
genome.
2 Preface / Virus Research 102 (2004) 1–2
The topic of gene silencing is relatively new, and the
reader will find certain repetition in the citation of some
references and (to a lesser extent) in the text. We decided
not to edit this out. We considered that any harsh editing
would diminish the excitement for the reader to participate
in the explosion of high quality knowledge increase based on
relatively few but excellent papers in basic science. Finally,
we thank all the contributors for taking the time to share
their experiences in this emerging and exciting field.
Susana López∗
Carlos F. Arias
Dept. de Genetica del Desarrollo y Fisiologia Molecular
Instituto de Biotecnologia, Universidad Nacional
Autonoma de Mexico, Avenida Universidad 2001
Col. Chamilpa, Cuernavaca, Morelos 62210, Mexico
∗ Corresponding author. Tel.: +52-777-3291615
fax: +52-777-3172388
E-mail address: susana@ibt.unam.mx (S. López)
 Virus Research 102 (2004) 3–9

Derivation and function of small interfering RNAs and microRNAs

Bryan R. Cullen∗
Department of Molecular Genetics and Microbiology, Howard Hughes Medical Institute, Duke University Medical Center, Durham, NC 27710, USA

Abstract

Small interfering RNA (siRNA) duplexes are generally produced by Dicer cleavage of double-stranded RNAs of frequently exogenous
origin and can induce the cleavage and degradation of mRNAs bearing an identical sequence. In contrast, microRNAs (miRNAs) are encoded
within the eukaryotic genome as short RNA hairpin structures. While these pre-miRNAs are also processed by Dicer, mature miRNAs appear
to function primarily by inhibiting the translation of mRNAs bearing multiple, partially mismatched target sites. Nevertheless, recent data
argue that the posttranscriptional regulatory machinery utilized by siRNAs and miRNAs is largely or entirely identical. In this review, I will
discuss recent progress in unraveling the RNA processing pathway utilized for the biosynthesis of mature miRNAs and argue that this pathway
offers at least three distinct entry points for the functional expression of artificial siRNAs in vertebrate cells. While each of these entry points
offers distinct advantages and disadvantages, they all have the potential to induce the effective knock-down of specific genes either in cell
culture or in experimental animals.
© 2004 Elsevier B.V. All rights reserved.

Keywords: miRNA; RNA interference; RNA processing; siRNA

1. Introduction 2000). This cleavage is mediated by a processive cytoplas-

RNA interference (RNAi) was first defined in C. ele-
gans as a mechanism that induced the posttranscriptional
silencing of genes in response to the introduction of long
double-stranded RNAs (dsRNAs) of identical sequence (Fire
et al., 1998). It was soon realized that RNAi was closely
related to previously described, but poorly understood, gene
inactivation pathways in plants and certain fungi (reviewed
by Bernstein et al., 2001b). While long dsRNAs also proved
highly effective at inducing RNAi in Drosophila (Hammond
et al., 2000), initial efforts to demonstrate RNAi in verte-
brate cells were largely unsuccessful. This failure resulted
from the fact that long dsRNAs also activate the vertebrate
interferon response, a complex defense system, lacking in
non-vertebrate species, that leads to a global posttranscrip-
tional inhibition of gene expression (Sen, 2001).
Efforts to understand how RNAi works in plants and
invertebrates demonstrated that long dsRNAs are initially
cleaved into characteristic ∼21 nt dsRNAs, bearing 2 nt 3
overhangs, termed small interfering RNA (siRNA) duplexes
(Fig. 1) (Hamilton and Baulcombe, 1999; Hammond et al.,
∗ Tel.: +1-919-684-3369; fax: +1-919-681-8979.
E-mail address: culle002@mc.duke.edu (B.R. Cullen).
mic RNase III family enzyme called Dicer (Bernstein et al.,
2001a; Grishok et al., 2001; Hutvágner et al., 2001; Knight
and Bass, 2001). One strand of this siRNA duplex is then
selectively incorporated into a large protein complex termed
the RNA induced silencing complex or RISC (Hammond
et al., 2000; Nykänen et al., 2001). This RNA strand then
acts as a guide RNA to target RISC to homologous mR-
NAs (Martinez et al., 2002; Schwarz et al., 2003). Once
RISC is bound, a currently unidentified ribonuclease com-
ponent cleaves the target mRNA opposite the center of the
bound guide siRNA (Fig. 1). RISC is then released, to seek
additional mRNA targets, while the two fragments of the
mRNA are degraded by cellular exonucleases (Hutvágner
and Zamore, 2002).
A critical discovery, made initially in Drosophila and
subsequently confirmed in vertebrate cells, is that synthetic
siRNA duplexes can also program RISC and induce the in-
activation of specific target genes (Elbashir et al., 2001a,b).
This result was made possible in vertebrate cells by the
fact that the interferon response is activated by dsRNAs of
>30 bp (Manche et al., 1992), while siRNA duplexes bear
a double-stranded region of only ∼19 bp. This observa-
tion demonstrated that the RNAi pathway was indeed fully
active in mammalian cells and raised the possibility, now
confirmed, that RNAi could be used to perform reverse

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.009
4 B.R. Cullen / Virus Research 102 (2004) 3–9

Fig. 1. MicroRNA and siRNA processing and function. A proposed processing pathway for endogenously encoded miRNAs is shown at the center of
the figure. This proceeds from transcription through nuclear processing, nuclear export, cytoplasmic processing by Dicer and finally to incorporation into
RISC. This results in the initial primary miRNA transcript (pri-miRNA) being processed to a pre-miRNA and finally to a miRNA duplex intermediate,
one strand of which then serves as a substrate for RISC incorporation. The assembled RISC can then target mRNAs bearing a perfectly complementary
target site for degradation or can inhibit the translation of an mRNA that contains multiple, partly mismatched target sites. Entry points for designed
siRNAs are shown at the left of the figure. These include synthetic siRNA duplexes as well as short hairpin RNAs, transcribed from expression plasmids,
that closely mimic pre-miRNAs. Finally, artificial miRNA genes can also be used to express novel siRNAs. Long dsRNAs, which can generate siRNAs
in invertebrate and plant cells, activate the interferon response in vertebrate cells and are therefore not useful in the latter system.

genetic analyses in human cells (reviewed by Dykxhoorn
et al., 2003).
2. The discovery of microRNAs
As noted above, long dsRNAs are processed by Dicer to
give ∼21 bp siRNA duplexes, one strand of which is then
incorporated into RISC. Efforts in the Drosophila and C. el-
egans systems to clone these short non-coding RNAs based
on their size led to the surprising discovery that not only
these organisms, but also vertebrates, express over 200 ge-
nomically encoded ∼21 nucleotide RNAs (Lagos-Quintana
et al., 2001; Lau et al., 2001; Lee and Ambros, 2001). These
RNAs, which are now termed microRNAs (miRNAs), are
predicted to form part of one strand of an imperfect precur-
sor RNA hairpin structure of between 60 and 90 nucleotides
in length. This contrasts with siRNAs, which are invariably
derived from long dsRNAs. In addition, while miRNAs are
by definition of endogenous origin, siRNAs can be endoge-
nous or exogenous.
While the discovery of this large family of miRNAs was
unexpected, two miRNAs had in fact been known for several
years in C. elegans. These miRNAs, termed let-7 and lin-4,
emerged from mutational screens designed to identify genes
involved in regulating larval development in C. elegans
 B.R. Cullen / Virus Research 102 (2004) 3–9
(Lee et al., 1993; Reinhart et al., 2000). Both let-7 and lin-4
are 21 nt in length and both are initially transcribed as one
arm of an ∼70 nt stem-loop RNA precursor. Both let-7 and
lin-4 are expressed in a developmentally regulated manner
and both inhibit the expression of specific, developmentally
important mRNAs by forming duplexes with multiple con-
served sequence elements present in their 3 untranslated
region (3 UTR). However, unlike siRNAs, this interaction
specifically blocks the translation of these target mRNAs,
rather than inducing their degradation (Fig. 1) (Olsen and
Ambros, 1999). Interestingly, known target sequences for
let-7 and lin-4 bear central mismatches so that the bound
let-7 and lin-4 miRNAs would be predicted to form bulges
(Lee et al., 1993; Reinhart et al., 2000). Recent evidence
indicates that it is these mismatches that prevent target
mRNA cleavage by these miRNAs, and not some intrinsic
difference in miRNA versus siRNA function (see below).
Although many different miRNAs have been reported in
a range of species, very few have as yet been assigned a
function. Exceptions include not only let-7 and lin-4 in C.
elegans but also two miRNAs termed bantam and miR-14 in
Drosophila, miR-171 in Arabidopsis and miR-23 in human
cells (Llave et al., 2002; Brennecke et al., 2003; Kawasaki
and Taira, 2003; Xu et al., 2003). As in the case of let-7
and lin-4, these miRNAs are expressed in a developmen-
tally regulated manner and inhibit the expression of target
mRNAs bearing complementary sequences. Consistent with
the hypothesis that other miRNAs may also play a role in
development, many vertebrate and non-vertebrate miRNAs
have been found to be expressed in a tissue specific and/or
developmentally regulated manner (Lagos-Quintana et al.,
2001, 2002; Lau et al., 2001; Lee and Ambros, 2001). More-
over, mutational disruption of genes encoding proteins that
play a global role in miRNA biogenesis, such as Dicer, has
profound deleterious affects on the development of mutant
organisms (Grishok et al., 2001; Park et al., 2002).
3. Expression, processing and function of microRNAs
The emerging importance of miRNAs in regulating the
appropriate development and differentiation of multicellular
organisms, and the similarity of miRNAs to siRNAs, has
prompted efforts to more fully understand miRNA deriva-
tion and function. Although these studies remain incom-
plete, the pathway delineated in Fig. 1 is likely to represent
a fairly accurate overview of this process.
miRNAs are found either as individual miRNA genes or in
tandemly arrayed clusters that may suggest a functional re-
lationship analogous to a bacterial operon (Lagos-Quintana
et al., 2001; Lau et al., 2001). In any event, the miRNA
genes that have been studied in detail thus far, such as
human miR-30, appear to be transcribed as part of a sev-
eral hundred nucleotide long primary transcript termed a
pri-miRNA (Fig. 1) (Lee et al., 2002; Zeng and Cullen,
2003). In at least some cases, transcription is mediated by
5
RNA polymerase II, and it is therefore likely that these
pri-miRNAs are capped and polyadenylated. It remains pos-
sible that other pri-miRNAs might be transcribed by other
cellular polymerases.
The first step in miRNA processing is the excision
of the ∼70 nt miRNA hairpin intermediate, termed the
pre-miRNA, from the much longer pri-miRNA (Fig. 1) (Lee
et al., 2002; Zeng and Cullen, 2003). This cleavage, which
occurs largely or exclusively in the nucleus, is mediated by
an enzyme termed RNase III or Drosha, a member of the
family of large RNases that also includes Dicer (Lee et al.,
2003b). Analysis of the pre-miRNA cleavage product that
results from processing of the miR-30 pri-miRNA shows
that this cleavage both defines the 3 end of the miR-30
miRNA and leaves a 2 nt 3 overhang (Lee et al., 2002;
Zeng and Cullen, 2003). Importantly, this cleavage does
not occur at the base of the stem of the predicted 80 nt
miR-30 precursor RNA hairpin but rather ∼7 nt above the
base, resulting in a 63 nt pre-miRNA intermediate. The se-
quences that determine the sites of cleavage of the miR-30
pri-miRNA by RNase III remain to be defined. However,
it is clear that the helical nature of the basal region of the
miR-30 stem-loop precursor plays a critical role and that
insertions or deletions in the precursor stem can result in a
shift in the cleavage sites (Zeng and Cullen, 2003).
The next step in the formation of the mature miR-30
miRNA is export of the pre-miRNA to the cytoplasm. Re-
cent evidence (Yi et al., 2003) indicates that the factor re-
sponsible for nuclear export of pre-miRNAs is Exportin 5
(Exp5), a member of the karyopherin family of nucleocy-
toplasmic transport factors (Fig. 1) (reviewed by Cullen,
2003). Exp5 is known to mediate the nuclear export of a
range of other small non-coding RNAs, including the aden-
ovirus VA1 RNA and the human Y1 RNA (Gwizdek et al.,
2003). Like other karyopherin family members involved in
nuclear export, Exp5 is dependent on the GTP-bound form of
the Ran co-factor for specific binding to its export substrate
in the cell nucleus and is predicted to release its RNA cargo
in the cytoplasm after hydrolysis of Ran·GTP to Ran·GDP
by the cytoplasmic Ran GTPase activating protein. Although
pre-miRNAs are recognized as specific substrates for Exp5
mediated nuclear export (Yi et al., 2003), the characteristics
of the pre-miRNA that mediate this recognition remain to
be determined.
Dicer is a processive, ATP-dependent RNase that binds
with high affinity to the ends of dsRNAs bearing 2 nt 3
overhangs, i.e. the product that it normally generates during
processing of long dsRNAs (Zhang et al., 2002). Dicer then
cleaves both RNA strands ∼21 nt from the bound end. The
structure of the pre-miRNA, which consists of an RNA hair-
pin ending with a 2 nt 3 overhang (Zeng and Cullen, 2003),
is closely analogous to intermediates generated during pro-
cessing of long dsRNAs except that one end is closed by
a loop. Importantly, the fact that Dicer cleaves at a prede-
termined distance from the end of the pre-miRNA hairpin
means that the structure of the pre-miRNA predetermines
6 B.R. Cullen / Virus Research 102 (2004) 3–9
the sequence of the miRNA duplex intermediate that is gen-
erated by Dicer processing. This emphasizes the importance
of the initial, nuclear RNA processing event in determining
the final product of this pathway (Fig. 1).
Although it is now clearly established that Dicer process-
ing of pre-miRNAs gives rise to a duplex RNA intermediate,
this is generally short-lived and therefore hard to detect.
However, while the majority of miRNA precursors give rise
to only a single mature miRNA, a small number, including
miR-30, do give rise to mature miRNAs derived from both
strands (Zeng et al., 2002). This allows the structure of the
miRNA duplex intermediate to be clearly deduced. In fact,
this miRNA duplex is identical in structure to an siRNA
duplex, with the caveat that miRNA precursors frequently
contain one or two mismatches in the stem, while siRNAs
are generally perfectly double-stranded (Fig. 1).
The short half-life of the miRNA duplex intermediate
is apparently due largely to the fact that assembly of one
strand of the duplex into RISC is rapid and efficient. It is
believed that a helicase component of RISC binds to one
end of the duplex and unravels the double-strand, a process
that requires ATP (Nykänen et al., 2001). The strand whose
5 end is at the bound end of the duplex is then incorpo-
rated into RISC. This process is non-random, in that RISC
binding strongly favors the end of the miRNA duplex in-
termediate that is less tightly helical (Schwarz et al., 2003).
Therefore, it is commonly observed that the 5 end of the
miRNA strand incorporated into RISC is poorly base paired
in the predicted duplex intermediate, while the 5 end of
the excluded strand forms a strong basepair, such as an G:C
(Lagos-Quintana et al., 2001; Schwarz et al., 2003).
While the composition of the RISC complex remains to be
fully established, it is clear that RISC contains one or more
members of the Argonaut family of proteins (Hammond
et al., 2001; Carmell et al., 2002; Mourelatos et al., 2002).
Certain Argonaut proteins have been found to be essential
for miRNA function in C. elegans and for siRNA func-
tion in cultured human cells (Tabara et al., 1999; Doi et al.,
2003). However, their actual role, and whether different hu-
man Argonaut proteins have different roles, remains to be
determined.
As noted above, the C. elegans let-7 and lin-4 miRNAs
inhibit the expression of target mRNAs after interacting with
multiple imperfect target sites in the 3 UTR. Importantly,
this does not result in a marked drop in the cytoplasmic
level of the target mRNA (Olsen and Ambros, 1999). Inhi-
bition at the translational level has therefore been inferred,
even though the polysome profile of the target mRNAs is
largely unaffected. While the mechanism underlying this in-
hibition remains unclear, it appears to be cooperative in that
the binding of multiple RISC complexes is required (Doench
et al., 2003). Unlike mRNA cleavage by RISC, which al-
lows turnover of RISC (Hutvágner and Zamore, 2002), this
translational inhibition also appears to require stoichiomet-
ric levels of RISC and hence of the relevant miRNA (Zeng
et al., 2003). Translational inhibition is therefore less effec-
tive, particularly at lower levels of miRNA/siRNA expres-
sion, than is inhibition via mRNA cleavage. However, the
requirement for binding of multiple RISC complexes for ef-
fective mRNA translation inhibition does offer the potential
for coordinate regulation of mRNA expression by multiple,
distinct miRNAs.
Although it initially seemed possible that miRNAs and
siRNAs might be functionally distinct, with the former in-
ducing translational inhibition and the latter mRNA cleav-
age, this is no longer believed to be true. Endogenously
encoded or overexpressed human miRNAs have now been
shown to induce the cleavage of artificial mRNA substrates
bearing perfectly homologous target sites both in vitro and in
vivo (Hutvágner and Zamore, 2002; Zeng et al., 2003). Sim-
ilarly, artificial siRNAs have been shown to inhibit gene ex-
pression from mRNAs bearing imperfect targets without af-
fecting mRNA expression levels (Doench et al., 2003; Zeng
et al., 2003). Together, these data demonstrate that siRNAs
and miRNAs are able to program RISC in a functionally in-
distinguishable manner (Fig. 1). On the other hand, it does
remain possible that mRNA cleavage and mRNA transla-
tion inhibition are mediated by distinct forms of RISC, per-
haps containing different members of the Argonaut protein
family. However, in that case these alternate forms must be
equivalently programmable by both siRNAs and miRNAs.
4. Strategies for expression of designed siRNAs
The miRNA processing pathway delineated in Fig. 1 is
conserved in most, possibly all, higher eukaryotic cells and
can be used to promote the entry of exogenously encoded
or synthetic siRNAs into this highly effective posttranscrip-
tional pathway of gene inactivation. In many invertebrates,
long dsRNAs are an effective means of inducing gene
specific RNAi. However, as noted above, long dsRNAs
activate the interferon response in vertebrate cells and are
therefore not useful in these organisms. To overcome this
problem, it is possible to program RISC by transfection of
cells with synthetic siRNA duplexes that closely mimic the
miRNA duplex intermediate (Elbashir et al., 2001a). This
approach can result in the specific and effective destruction
of target mRNAs (Fig. 1). However, synthetic siRNAs can
be quite expensive and RNA transfection is not efficient in
all cell types of interest, particularly primary cells. More-
over, because only a finite amount of the siRNA duplex is
introduced into cells, inhibition is lost over a period of a
few days due to dilution in the growing cell culture and/or
degradation of the transfected siRNAs.
To overcome this problem, several groups have devel-
oped plasmids that can express short hairpin RNAs that
are structurally analogous to pre-miRNA hairpin interme-
diates, using promoters dependent on RNA polymerase
III (Brummelkamp et al., 2002; Paddison et al., 2002; Sui
et al., 2002). In a typical expression plasmid of this type,
the U6 or H1 promoter drives the expression of an RNA
 B.R. Cullen / Virus Research 102 (2004) 3–9
that is predicted to fold into a short RNA hairpin, with the
sequence of one side of the stem being complementary to
the mRNA target sequence of interest. Transcription termi-
nation is induced by a run of five thymidine residues, which
causes termination of transcription by RNA polymerase III
after the second thymidine residue. This allows the short
hairpin to acquire a 2 nt 3 overhang, exactly like authentic
pre-miRNAs, and recent evidence indicates that these short
hairpin RNAs are exported from the nucleus by Exp5, just
like an authentic pre-miRNA (Fig. 1) (Yi et al., 2003).
Subsequent work has refined the parameters for design
of short hairpin RNAs for maximal likelihood of successful
RNAi. Thus, it is now clear that stems of from 25 to 29 bp
in length give rise to better RNA processing and siRNA ex-
pression than do shorter stems (stems >29 bp in length can
induce the interferon response). Because Dicer cleaves ∼21
nt from the base of the short hairpin RNA, after binding to
the end of this RNA structure, it is the more basal sequence
of the stem that gives rise to the siRNA duplex intermedi-
ate (Lee et al., 2003a). To promote incorporation of the an-
tisense strand into RISC, it is important to ensure that the
projected 5 end of the antisense strand is poorly basepaired
(Schwarz et al., 2003). Even in the absence of these modifi-
cations, short hairpin RNA expression plasmids have proven
to be quite effective in inducing RNAi. However, these plas-
mids again have practical problems in terms of the ability
to effectively transfect cells of interest and in terms of the
transience of the inhibitory effect observed.
To overcome these problems, several groups have devel-
oped viral vectors, based on murine leukemia virus or hu-
man immunodeficiency virus that incorporate short hairpin
RNA expression cassettes (Barton and Medzhitov, 2002; Lee
et al., 2003a; Qin et al., 2003; Stewart et al., 2003). These
vectors can be used to efficiently infect target cells in culture
or in vivo and have been shown to give rise to long term ex-
pression of the encoded siRNA and, hence, to the long term
suppression of the target mRNA of interest. Most recently,
lentivirus-based short hairpin RNA expression vectors have
been used to effectively and stably suppress the expression
of genes in transgenic mice (Rubinson et al., 2003; Tiscornia
et al., 2003), thus hopefully heralding a new era of simple
reverse genetic experiments in experimental animals.
A final approach to the expression of siRNAs uses as its
entry point the first step in the expression pathway for en-
dogenous miRNAs (Fig. 1). Because pri-miRNAs contain
all the sequences required for miRNA processing in cis, it
should in principle be possible to switch the part of the
miRNA gene that encodes the mature miRNA, while leav-
ing the processing signals intact, and hence use the entire
miRNA processing machinery to produce an siRNA, or ar-
tificial miRNA, of interest. In fact, using the miR-30 gene
as a starting point, this has indeed been achieved using sev-
eral different siRNA sequences (Zeng et al., 2002; Zeng and
Cullen, 2003). A major advantage of this approach to siRNA
expression is that it relies on transcription via RNA poly-
merase II and thus has the potential to permit the use of
7
tissue-specific or developmentally-regulated promoters for
expression of particular siRNAs in vivo. In contrast, the
level of expression of siRNAs transcribed from vectors that
depend on RNA polymerase III is not easily controlled, al-
though this approach does permit the constitutive expression
of high siRNA levels.
5. Conclusions
Although RNAi plays a critical role in antiviral defense in
plants, and is important in inhibiting transposon activation
in invertebrates (Plasterk, 2002), it remains unclear whether
RNAi serves similar functional roles in vertebrate species.
What is clear is that miRNAs are expressed in all higher
eukaryotes and, at minimum, play a critical role in the post-
transcriptional regulation of genes during differentiation and
development. The pathway that mediates miRNA process-
ing and function is becoming increasingly well understood
and it is now apparent that this pathway can support several
different strategies for the introduction of functional, exoge-
nously produced siRNAs (Fig. 1). While it remains uncertain
whether RNAi plays any role in controlling virus infections
in animals, there is no question that the artificial induction
of RNAi can be used to target viral transcripts, or host
genes that produce co-factors critical for virus replication,
and thereby effectively block virus replication in culture.
Whether RNAi will lead to effective antiviral treatments in
the future remains to be seen. What is clear, however, is that
RNAi is an extremely powerful tool to identify and char-
acterize cellular factors that are required for effective virus
replication. In doing so, RNAi has the potential to identify
targets for chemotherapeutic intervention that would be
very difficult to define using more conventional approaches.
References
Barton, G.M., Medzhitov, R., 2002. Retroviral delivery of small interfering
RNA into primary cells. Proc. Natl. Acad. Sci. U.S.A. 99, 14943–
14945.
Bernstein, E., Caudy, A.A., Hammond, S.M., Hannon, G.J., 2001a. Role
for bidentate ribonuclease in the initiation step of RNA interference.
Nature 409, 363–366.
Bernstein, E., Denli, A.M., Hannon, G.J., 2001b. The rest is silence. RNA
7, 1509–1521.
Brennecke, J., Hipfner, D.R., Stark, A., Russell, R.B., Cohen, S.M.,
2003. Bantam encodes a developmentally regulated microRNA that
controls cell proliferation and regulates the proapoptotic gene hid in
Drosophila. Cell 113, 25–36.
Brummelkamp, T.R., Bernards, R., Agami, R., 2002. A system for stable
expression of short interfering RNAs in mammalian cells. Science
296, 550–553.
Carmell, M.A., Xuan, Z., Zhang, M.Q., Hannon, G.J., 2002. The argonaute
family: tentacles that reach into RNAi, developmental control, stem
cell maintenance, and tumorigenesis. Genes Dev. 16, 2733–2742.
Cullen, B.R., 2003. Nuclear RNA export. J. Cell Sci. 116, 587–597.
Doench, J.G., Petersen, C.P., Sharp, P.A., 2003. siRNAs can function as
miRNAs. Genes Dev. 17, 438–442.
8 B.R. Cullen / Virus Research 102 (2004) 3–9
Doi, N., Zenno, S., Ueda, R., Ohki-Hamazaki, H., Ui-Tei, K., Saigo, K.,
2003. Short-interfering-RNA-mediated gene silencing in mammalian
cells requires Dicer and eIF2C translation initiation factors. Curr. Biol.
13, 41–46.
Dykxhoorn, D.M., Novina, C.D., Sharp, P.A., 2003. Killing the messenger:
short RNAs that silence gene expression. Nat. Rev. 4, 457–467.
Elbashir, S.M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., Tuschl,
T., 2001a. Duplexes of 21-nucleotide RNAs mediate RNA interference
in cultured mammalian cells. Nature 411, 494–498.
Elbashir, S.M., Lendeckel, W., Tuschl, T., 2001b. RNA interference is
mediated by 21- and 22-nucleotide RNAs. Genes Dev. 15, 188–200.
Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver, S.E., Mello,
C.C., 1998. Potent and specific genetic interference by double-stranded
RNA in Caenorhabditis elegans. Nature 391, 806–811.
Grishok, A., Pasquinelli, A.E., Conte, D., Li, N., Parrish, S., Ha, I., Baillie,
D.L., Fire, A., Ruvkun, G., Mello, C.C., 2001. Genes and mechanisms
related to RNA interference regulate expression of the small temporal
RNAs that control C. elegans developmental timing. Cell 106, 23–34.
Gwizdek, C., Ossarch-Nazari, B., Brownawell, A.M., Doglio, A., Bertrand,
E., Macara, I.G., Dargemont, C., 2003. Exportin-5 mediates nuclear
export of minihelix-containing RNAs. J. Biol. Chem. 278, 5505–
5508.
Hamilton, A.J., Baulcombe, D.C., 1999. A species of small antisense RNA
in posttranscriptional gene silencing in plants. Science 286, 950–952.
Hammond, S.M., Bernstein, E., Beach, D., Hannon, G.J., 2000. An
RNA-directed nuclease mediates post-transcriptional gene silencing in
Drosophila cells. Nature 404, 293–295.
Hammond, S.M., Boettcher, S., Caudy, A.A., Kobayashi, R., Hannon, G.J.,
2001. Argonaute 2, a link between genetic and biochemical analyses
of RNAi. Science 293, 1146–1150.
Hutvágner, G., McLachlan, J., Pasquinelli, A.E., Bálint, É., Tuschl, T.,
Zamore, P.D., 2001. A cellular function for the RNA-interference
enzyme Dicer in the maturation of the let-7 small temporal RNA.
Science 293, 834–838.
Hutvágner, G., Zamore, P.D., 2002. A microRNA in a multiple-turnover
RNAi enzyme complex. Science 297, 2056–2060.
Kawasaki, H., Taira, K., 2003. Hes1 is a target of microRNA-23 during
retinoic-acid-induced neuronal differentiation of NT2 cells. Nature
423, 838–842.
Knight, S.W., Bass, B.L., 2001. A role for the RNase III enzyme DCR-1
in RNA interference and germ line development in Caenorhabditis
elegans. Science 293, 2269–2271.
Lagos-Quintana, M., Rauhut, R., Lendeckel, W., Tuschl, T., 2001. Iden-
tification of novel genes coding for small expressed RNAs. Science
294, 853–858.
Lagos-Quintana, M., Rauhut, R., Yalcin, A., Meyer, J., Lendeckel, W.,
Tuschl, T., 2002. Identification of tissue-specific microRNAs from
mouse. Curr. Biol. 12, 735–739.
Lau, N.C., Lim, L.P., Weinstein, E.G., Bartel, D.P., 2001. An abundant
class of tiny RNAs with probable regulatory roles in Caenorhabditis
elegans. Science 294, 858–862.
Lee, M.-T., Coburn, G.A., McClure, M.O., Cullen, B.R., 2003a. Inhi-
bition of HIV-1 replication in primary macrophages using Tat or
CCR5-specific small interfering RNAs expressed from a lentiviral
vector. J. Virol. 77, 11964–11972.
Lee, R.C., Ambros, V., 2001. An extensive class of small RNAs in
Caenorhabditis elegans. Science 294, 862–864.
Lee, R.C., Feinbaum, R.L., Ambros, V., 1993. The C. elegans hete-
rochronic gene lin-4 encodes small RNAs with antisense complemen-
tarity to lin-14. Cell 75, 843–854.
Lee, Y., Ahn, C., Han, J., Choi, H., Provost, P., Kim, S., Kim, V.N.,
2003b. The nuclear RNase III drosha initiates microRNA processing.
Nature 425, 415–419.
Lee, Y., Jeon, K., Lee, J.T., Kim, S., Kim, V.N., 2002. MicroRNA matu-
ration: stepwise processing and subcellular localization. EMBO J. 21,
4663–4670.
Llave, C., Xie, Z., Kasschau, K.D., Carrington, J.C., 2002. Cleavage
of scarecrow-like mRNA targets directed by a class of Arabidopsis
miRNA. Science 297, 2053–2056.
Manche, L., Green, S.R., Schmedt, C., Mathews, M.B., 1992. Interactions
between double-stranded RNA regulators and the protein kinase DAI.
Mol. Cell. Biol. 12, 5238–5248.
Martinez, J., Patkaniowska, A., Urlaub, H., Lührmann, R., Tuschl, T.,
2002. Single-stranded antisense siRNAs guide target RNA cleavage
in RNAi. Cell 110, 563–574.
Mourelatos, Z., Dostie, J., Paushkin, S., Sharma, A., Charroux, B., Abel,
L., Rappsilber, J., Mann, M., Dreyfuss, G., 2002. miRNPs: a novel
class of ribonucleoproteins containing numerous microRNAs. Genes
Dev. 16, 720–728.
Nykänen, A., Haley, B., Zamore, P.D., 2001. ATP requirements and small
interfering RNA structure in the RNA interference pathway. Cell 107,
309–321.
Olsen, P.H., Ambros, V., 1999. The lin-4 regulatory RNA controls de-
velopmental timing in Caenorhabditis elegans by blocking LIN-14
protein synthesis after the initiation of translation. Dev. Biol. 216,
671–680.
Paddison, P.J., Caudy, A.A., Bernstein, E., Hannon, G.J., Conklin, D.S.,
2002. Short hairpin RNAs (shRNAs) induce sequence-specific silenc-
ing in mammalian cells. Genes Dev. 16, 948–958.
Park, W., Li, J., Song, R., Messing, J., Chen, X., 2002. Carpel factory,
a Dicer homolog, and HEN1, a novel protein, act in microRNA
metabolism in Arabidopsis thaliana. Curr. Biol. 12, 1484–1495.
Plasterk, R.H.A., 2002. RNA silencing: the genome’s immune system.
Science 296, 1263–1265.
Qin, X.-F., An, D.S., Chen, I.S.Y., Baltimore, D., 2003. Inhibiting HIV-1
infection in human T cells by lentiviral-mediated delivery of small
interfering RNA against CCR5. Proc. Natl. Acad. Sci. U.S.A. 100,
183–188.
Reinhart, B.J., Slack, F.J., Basson, M., Pasquinelli, A.E., Bettinger, J.C.,
Rougvie, A.E., Horvitz, H.R., Ruvkun, G., 2000. The 21-nucleotide
let-7 RNA regulates developmental timing in Caenorhabditis elegans.
Nature 403, 901–906.
Rubinson, D.A., Dillon, C.P., Kwiatkowski, A.V., Sievers, C., Yang, L.,
Kopinja, J., Zhang, M., McManus, M.T., Gertler, F.B., Scott, M.L.,
Van Parijs, L., 2003. A lentivirus-based system to functionally silence
genes in primary mammalian cells, stem cells and transgenic mice by
RNA interference. Nat. Gen. 33, 401–406.
Schwarz, D.S., Hutvágner, G., Du, T., Xu, Z., Aronin, N., Zamore, P.D.,
2003. Asymmetry in the assembly of the RNAi enzyme complex. Cell
115, 199–208.
Sen, G.C., 2001. Viruses and interferons. Annu. Rev. Microbiol. 55, 255–
281.
Stewart, S.A., Dykxhoorn, D.M., Palliser, D., Mizuno, H., Yu, E.Y., An,
D.S., Sabatini, D.M., Chen, I.S.Y., Hahn, W.C., Sharp, P.A., Weinberg,
R.A., Novina, C.D., 2003. Lentivirus-delivered stable gene silencing
by RNAi in primary cells. RNA 9, 493–501.
Sui, G., Soohoo, C., Affar, E.B., Gay, F., Shi, Y., Forrester, W.C., Shi,
Y., 2002. A DNA vector-based RNAi technology to suppress gene
expression in mammalian cells. Proc. Natl. Acad. Sci. U.S.A. 99,
5515–5520.
Tabara, H., Sarkissian, M., Kelly, W.G., Fleenor, J., Grishok, A., Timmons,
L., Fire, A., Mello, C.C., 1999. The rde-1 gene, RNA interference,
and transposon silencing in C. elegans. Cell 99, 123–132.
Tiscornia, G., Singer, O., Ikawa, M., Verma, I.M., 2003. A general method
for gene knockdown in mice by using lentiviral vectors expressing
small interfering RNA. Proc. Natl. Acad. Sci. U.S.A. 100, 1844–1848.
Xu, P., Vernooy, S.Y., Guo, M., Hay, B.A., 2003. The Drosophila mi-
croRNA Mir-14 suppresses cell death and is required for normal fat
metabolism. Curr. Biol. 13, 790–795.
Yi, R., Qin, Y., Macara, I.G., Cullen, B.R., 2003. Exportin-5 Mediates
the nuclear export of pre-microRNAs and short hairpin RNAs. Genes
Dev. 17, 3011–3016.
 B.R. Cullen / Virus Research 102 (2004) 3–9 9

Zeng, Y., Cullen, B.R., 2003. Sequence requirements for micro
RNA processing and function in human cells. RNA 9, 112–
123.
Zeng, Y., Wagner, E.J., Cullen, B.R., 2002. Both natural and designed
micro RNAs can inhibit the expression of cognate mRNAs when
expressed in human cells. Mol. Cell 9, 1327–1333.
Zeng, Y., Yi, R., Cullen, B.R., 2003. MicroRNAs and small interfering
RNAs can inhibit mRNA expression by similar mechanisms. Proc.
Natl. Acad. Sci. U.S.A. 100, 5779–5784.
Zhang, H., Kolb, F.A., Brondani, V., Filipowicz, W., 2002. Human Dicer
preferentially cleaves dsRNAs at their termini without a requirement
for ATP. EMBO J. 21, 5875–5885.
 Virus Research 102 (2004) 11–17

RNA silencing in viral infections: insights from poliovirus

Maria-Carla Saleh1 , Ronald P. Van Rij1 , Raul Andino∗
Department of Microbiology and Immunology, University of California, 513 Parnassus Avenue, Box 0414, San Francisco, CA 94143-2280, USA

Abstract

RNA interference or RNA silencing is a dsRNA guided mechanism that mediates sequence specific degradation of RNA. The recent
demonstration that RNA interference can be used to inhibit virus replication has initiated an exciting field of research: first, as a potential
novel antiviral therapeutic approach and, second, as a tool for dissecting virus–host interactions. Here we review and discuss the current data
and perspectives on the use of RNA interference in the study of poliovirus as a model for positive strand RNA viruses.
© 2004 Elsevier B.V. All rights reserved.

Keywords: RNA interference; RNA silencing; siRNA; Poliovirus; Picornaviridae

1. Introduction Because of the extreme specificity and efficiency of the

Clearance of viral pathogens from a variety of hosts is of-
ten the result of the destruction of infected cells. However,
mechanisms that enable cell survival following infection
have been described (Guidotti and Chisari, 2001). Studies
in plants suggest that one of such mechanisms may be
based on RNA-induced gene silencing, now referred to as
RNA silencing or RNA interference (RNAi). This ancient
pathway is conserved among various species from different
kingdoms (fungi, animals and plants) and it seems to control
gene expression at transcriptional and post-transcriptional
levels (Bernstein et al., 2001b; Hammond et al., 2001b).
The post-transcriptional activity of the RNAi machinery
to degrade cytoplasmic RNA in a sequence specific manner
is key to its antiviral function. Biochemical studies indicate
that RNAi proceeds via a two-step mechanism. In the first
step, long dsRNAs are produced locally or taken up by
the cells and cleaved by the RNase III-like nuclease Dicer
(Bernstein et al., 2001a), which generates 21–23 nucleotide
duplex RNAs, called small interfering RNAs (siRNA). In the
second step, siRNAs are incorporated into a multicompo-
nent nuclease complex, the RNA-induced silencing complex
(RISC) (Hammond et al., 2001a). The antisense strand of the
duplex serves as a guide that directs RISC to recognize and
cleave cognate mRNAs (Martinez et al., 2002a) (Fig. 1b).
∗ Corresponding author. Tel.: +1-415-502-6358;
fax: +1-415-476-0939.
E-mail address: andino@itsa.ucsf.edu (R. Andino).
1 Author contributed equally to this review.
RNAi machinery, this mechanism has the ability to clear
plant and mammalian cells from viral infection (Gitlin and
Andino, 2003; Randall et al., 2003; Waterhouse et al., 2001).
2. Multiple functions of RNAi
The RNAi process is very efficient: a few dsRNA
molecules can trigger inactivation of a continuously tran-
scribed target mRNA for prolonged periods of time (Ruvkun,
2001). The RNAi-induced inactivation persists through cell
division and in some organisms can spread to untreated cells
and tissues; when the RNAi spreads into germ line cells it
can even be inherited by subsequent generations (Hammond
et al., 2001b). Although amplification and spreading of
RNA silencing have been demonstrated in plants, nema-
todes and in vitro in Drosophila extracts (Fire et al., 1998;
Lipardi et al., 2001; Lucas et al., 2001; Roignant et al.,
2003) they have not been observed in mammalian systems
(Chi et al., 2003; Stein et al., 2003).
A role for RNAi pathways in the regulation of endoge-
nous genes was suggested through the analysis of plants
and animals containing dysfunctional RNAi components.
For example, mutations in a component of RISC cause
pleiotropic developmental abnormalities in Arabidopsis
(Bohmert et al., 1998). A mutation in the Arabidopsis Dicer
orthologue causes defects in leaf development and overpro-
liferation of floral meristems (Jacobsen et al., 1999). Mu-
tations in Argonaute family members in Drosophila impact
normal development with drastic effects in neuronal devel-

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.010
12 M.-C. Saleh et al. / Virus Research 102 (2004) 11–17

(a) (b)
Substrate
for Dicer? dsRNA
Entry P1 P2 P3 AAAA

Dicer

Translation 5' 3'

siRNA

Target after siRNA Target RNA

Negative treatment?
strand 5' 3' (+)
synthesis RISC
3Dpol 5' (-)

(+) Degradation
(+) (+) of target RNA
Positive
strand 3' 5' (-)
synthesis

Packaging

Fig. 1. (a) Overview of poliovirus life cycle. After receptor mediated entry and uncoating of the virus, the viral RNA is translated in a cap-independent
fashion by the host cell machinery. The viral RNA is used as a template for negative strand synthesis by the virus encoded 3D polymerase. The negative
strand is subsequently used as a template for the production of positive viral RNA strands, which are packaged into new virions. (b) Overview of RNAi
pathway. The central component is dsRNA which is converted by Dicer into siRNA, small dsRNA of ∼21 nucleotides length with a two nucleotide
overhang at the 3 end. One strand is incorporated in the RISC complex and guides recognition and cleavage of the target RNA. Open questions with
respect to RNA silencing of positive strand RNA viruses include whether dsRNA intermediates produced during virus replication are a target for the
RISC complex, whether viruses encode proteins that inhibit the RNAi pathway and which of the RNA strands are best targeted after introduction of
synthetic siRNAs.

opment (Kataoka et al., 2001), and mutations in another
RISC component, piwi, result in defects in both germline
stem-cell proliferation and maintenance (Cox et al., 1998).
The emerging view is that RNA silencing is part of a so-
phisticated network of interconnected pathways for cellular
defense (pathogen resistance and stabilization of mobile ge-
netic elements), RNA surveillance (chromatin remodeling,
genome organization and stability) and development. In ad-
dition, RNAi may become a powerful tool to manipulate
gene expression experimentally. Here we review and discuss
the current data and perspectives on the use of RNAi in the
study of positive strand RNA viruses. Indeed RNAi opens
up exciting possibilities for virus research: first, as a poten-
tial novel antiviral therapeutic approach and, second, as a
tool for dissecting virus–host interactions.
3. Antiviral function of RNAi
with the RNA silencing machinery in Drosophila cells (Li
et al., 2002). Furthermore, many C. elegans mutants defi-
cient in RNAi show transposon mobilization with obvious
implications for genome stability (Ketting et al., 1999). A
key question that remains unanswered is whether RNAi has
a role as a natural antiviral defense in mammals. Five criti-
cal issues have recently been proposed to establish whether
RNAi plays a role in natural defense against viruses (Gitlin
and Andino, 2003):
(1) Generation of siRNA during the course of a natural in-
fection.
(2) Up-regulation of RNAi machinery components during
viral infection.
(3) Viral mechanisms to suppress or escape RNAi.
(4) Increased susceptibility to viral infection after mutation
or deletion of RNAi components.
(5) Local or systemic spread of an RNAi-based antiviral
response.

There is a good deal of genetic support for the importance
of RNAi in genome defense. In plants, RNAi is clearly
involved in the response to viruses: Arabidopsis strains
defective in post transcriptional gene silencing are more
susceptible to virus infections (Mourrain et al., 2000), and a
substantial number of plant viruses encode proteins that
counter silencing (Brigneti et al., 1998; Voinnet, 2001;
Voinnet et al., 1999). RNA silencing also appears to con-
tribute to antiviral defense in invertebrates as it was shown
by studies of flock house virus (FHV) and its interaction
Even though synthetic siRNA transfected into mammalian
cells can inhibit positive strand virus replication (Gitlin and
Andino, 2003; Kapadia et al., 2003; Randall et al., 2003;
Seo et al., 2003; Wilson et al., 2003), it is not clear whether
RNA silencing can be elicited naturally during these viral in-
fections. Although the replication intermediates may contain
long stretches of dsRNA, especially during negative strand
synthesis, it is not know whether they can be targeted for
processing by Dicer. Localization of dsRNA intermediates in
membranous and/or proteinous replication complexes may
 M.-C. Saleh et al. / Virus Research 102 (2004) 11–17 13

shield these intermediates from the RNAi machinery. An-
swers to these and other related issues have not been obtained
yet, but an exciting field of research has been opened up.
4. Poliovirus as a model system to study RNAi
Until the 1950s, poliovirus (PV), the etiologic agent of
poliomyelitis, crippled thousands of children every year.
Poliovirus is able to invade the nervous system, and can
cause total paralysis in a matter of hours. The virus enters
the body through the mouth and multiplies in the ton-
sils and the Peyer’s patches of the small intestine before
spreading to other sites by a viraemia. Initial symptoms
are fever, fatigue, headache, vomiting, and stiffness in the
neck and pain in the limbs. One in 200 infections leads
to irreversible paralysis (usually in the legs). Among those
paralyzed, 5–10% die when their breathing muscles become
immobilized (Minor, 1997).
After the introduction of effective vaccines in the late
1950s (inactivated polio vaccine (IPV)) and early 1960s
(oral polio vaccine (OPV)), poliomyelitis was brought un-
der control. More recently, thanks to a global effort to erad-
icate poliovirus, the disease has been eliminated from most
of the world, and only seven countries worldwide remain
polio-endemic (http://www.polioeradication.org). However,
as we face the final stages of the Poliovirus Global Erad-
ication Initiative it is important to develop effective anti-
poliovirus therapies.
Poliovirus has served as an excellent model to study
the molecular and cellular events that lead to viral replica-
tion. As a consequence, much is known about its biology,
life cycle and interactions with the host. Poliovirus is a
positive-stranded RNA virus, member of the family Picor-
naviridae. The genome comprises a long open reading frame
encoding a single polyprotein of more than 2000 amino
acids, which is processed into functional proteins by vi-
rally encoded proteases. The poliovirus polyprotein can be
divided into three major parts (P1–P3). The P1 region con-
tains the four structural proteins (VP1–VP4) that constitute
the capsid. Proteins derived from the P2 and the P3 region
are involved in viral replication. After receptor-mediated
entry, the viral genome is directly translated by the cellular
translation machinery. The same viral RNA is then used as

Fig. 2. Poliovirus replication can be suppressed by siRNA. (a) Pretreatment of HeLa cells with siRNA directed against the capsid region (siC) is able
to prevent plaque formation. Suppression of replication is not observed after treatment of cells with single stranded RNA (ssC(+) and ssC(−)), an
unrelated siRNA (directed against luciferase (siL)), or dsDNA of identical sequence of siC (C-DNA). (b) Western blot analysis with anti 3Dpol antibody
of poliovirus infected cells. siP indicates siRNA directed against the poliovirus polymerase region. (c) Northern blot analysis of poliovirus infected cells
at 6 and 8 h after infection. Reprinted from Gitlin et al. (2002) with permission.
14 M.-C. Saleh et al. / Virus Research 102 (2004) 11–17
template for negative strand RNA synthesis that results in
the formation of a double-stranded RNA replicative form
(RF). The negative strand RNA can in turn serve as a
template for synthesis of new positive strand RNA. Repli-
cation occurs in replication complexes tightly associated
with smooth membrane vesicles, which accumulate in the
cytoplasm during viral infection (Fig. 1a) (Rueckert, 1996).
Can RNAi suppress replication of this highly lytic virus?
In a recent study, Gitlin et al. (2002) used siRNA directed
against the poliovirus genome to show that RNAi can
effectively prevent viral replication in mammalian cells
(Fig. 2). Pretreatment of human and murine cells with
double-stranded siRNA to the poliovirus genome (capsid
and 3D polymerase) prevented plaque formation and re-
sulted in a 100-fold reduction of the progeny titer in one-step
growth curves. In these experiments, siRNAs promote
clearance of the virus from most infected cells. The RNAi
antiviral effect, for which the mechanistic detail remains to
be elucidated, is sequence-specific and not attributable to
either classical antisense mechanism or interferon response
effectors PKR and RNaseL.
This study yielded insights into the accessibility of the
viral genome to RNAi, and the ability of virus to escape
this defense mechanism. Several siRNAs targeting regions
in capsid and polymerase that are well conserved among
enteroviruses were effective in inhibiting viral replication.
In contrast, siRNA that targeted the 5 non-coding region
(5 UTR), also highly conserved among all picornaviruses,
failed to reduce virus replication (L. Gitlin and R. Andino,
unpublished observations). This result implies that some re-
gions of the 5 UTR are less accessible to the RNAi machin-
ery. Indeed, secondary and tertiary structure of the RNA, as
well as proteins binding to RNA elements of the 5 UTR,
could potentially shield the viral target RNA from the RNAi
machinery. Since siRNAs directed against the hepatitis C
virus IRES have been recently shown to effectively inhibit
replication of HCV replicons (Yokota et al., 2003), there
may be differences in the accessibility of different regula-
tory regions of different viral genomes. Notably, viruses can
eventually become resistant to synthetic short RNAs such
as those described by Gitlin et al. (2002). Escape mutants
emerged from cells treated with single siRNAs after po-
liovirus infection at a high multiplicity of infection. These
escape mutants presented subtle nucleotide changes in the
siRNA target sequences. It appears that a single mismatch
located approximately in the center of the siRNA nearly
abolishes silencing of poliovirus (Gitlin and Andino, unpub-
lished observations).
5. Therapeutic potential of RNAi as an antiviral agent
The susceptibility of viruses to RNAi in tissue culture
suggests that RNAi-based antiviral strategies might be fea-
sible also in entire mammals. Evidence that RNAi can be
induced in vivo has been obtained in a number of mouse
studies, in which injection of siRNA or plasmids encoding
small hairpin RNAs (shRNAs) directed against luciferase,
GFP, or a secreted form of human placental alkaline phos-
phatase (SEAP) could inhibit expression of these genes from
plasmids after high pressure tail injection (hydrodynamic
transfection) (Lewis et al., 2002; McCaffrey et al., 2002).
Silencing of endogenous genes was shown by downregu-
lation of GFP in a transgenic mouse (Lewis et al., 2002).
The first application of RNAi technology to prevent disease
was illustrated by protection of mice against antibody- or
concanavalin A-induced hepatitis by using siRNA directed
against fas protein (Song et al., 2003). Recently, shRNAs
against HBV mRNAs were shown to be able to suppress
virus replication from co-injected HBV expression plasmids
in hepatocytes (McCaffrey et al., 2003), providing proof of
principle for the use of RNAi-based antiviral strategies in
animal models.
Despite these promising results, significant hurdles exist
in the successful application of RNAi-based antiviral strate-
gies in a clinical setting. Several of these issues may be ad-
dressed using poliovirus as a model virus.
First, as shown in Gitlin et al. (2002), a limited number
of mismatches between siRNA and target RNA may already
abolish the suppressive action of the siRNA (Elbashir et al.,
2001; Martinez et al., 2002b). The error-prone nature of the
virally encoded RNA-dependent RNA polymerase will gen-
erate a spectrum of mutants in each replication cycle, and
it can be expected that escape mutants will rapidly arise
under incomplete suppression by siRNA. Indeed, point mu-
tations arose in the center of the target region of the po-
liovirus genome, when infections were performed at high
MOI (Gitlin et al., 2002). Although single mismatches have
been shown to render a siRNA ineffective (Elbashir et al.,
2001), they may be tolerated by the RNAi machinery to
some extent (Jacque et al., 2002). Simultaneously targeting
of multiple regions or of regions that constitute indispens-
able RNA structures for virus replication and thus do not
allow viral escape, might be approaches to circumvent vi-
ral escape, that can readily be explored using the poliovirus
model system. Alternatively, targeting essential cellular fac-
tors, such as CC-chemokine receptor 5 (CCR5) and CD4 in
the case of HIV-1 (Novina et al., 2002; Qin et al., 2003) or
co-administration with other antiviral drugs may limit the
possibility for viral escape.
A second aspect requiring further investigation is that
not all siRNAs are equally effective in the suppression of a
target RNA (Harborth et al., 2001). The sequence require-
ments for effective RNAi-based suppression have not been
well defined and efficient siRNAs need to be developed by
trial and error. It is likely that optimal siRNA sequences
are determined by the interaction of the siRNAs with both
multicomponent nuclease RISC and with the target RNA
sequence. Other open questions that apply specifically
to viral genomes as targets for RNAi, are whether there
is a differential susceptibility to RNAi between coding
and non-coding regions, and whether protein bound RNA
 M.-C. Saleh et al. / Virus Research 102 (2004) 11–17
structures are less accessible to the RNAi machinery. If
conserved secondary structures in the untranslated regions
of the virus are indeed susceptible, these might be the re-
gions of choice to develop siRNAs as mutations in these
structures would lead to loss of function.
Another aspect that is not yet properly defined is whether
it is possible to block viral replication once initiated. Effec-
tive inhibition of poliovirus was observed when cells were
transfected with siRNA before infection (Gitlin et al., 2002).
Although in theory siRNA could target newly synthesized
RNA strands of either positive and negative orientation,
replication of poliovirus on membranous vesicles and cou-
pling of RNA synthesis with virion assembly might shield
replicative intermediates from degradation by the RNAi ma-
chinery.
Virally encoded suppressors of the RNAi pathway have
been described in a variety of plant viruses (Brigneti et al.,
1998; Voinnet et al., 1999) and in FHV, which naturally in-
fects vertebrate and invertebrate hosts (Li et al., 2002). Thus
far, similar suppressors have not been reported in mam-
malian viruses. The presence of putative suppressors in spe-
cific mammalian virus families would limit the potential of
RNAi-based therapeutic strategies.
Finally, recent RNAi studies in animals rely heavily on
the efficient uptake and expression of siRNA in hepatocytes
(Liu et al., 1999; Song et al., 2003; Zhang et al., 1999).
Other methods to deliver shRNA by adenovirus-mediated
delivery or lentiviral vectors have been successful in vivo
as well (Tiscornia et al., 2003; Xia et al., 2002). However,
as viruses differ greatly in tissue and cell tropism, specific
targeting of siRNA expression systems to specific tissues
will be crucial for the success of RNAi-based therapeutic
approaches. These methods may therefore encounter similar
problems and limitations as gene therapy (Thomas et al.,
2003).
6. RNA silencing as a tool to study virus replication
and pathogenesis
RNA interference has been developed into a powerful
technique to manipulate gene expression experimentally,
which may benefit the study of virus replication. For
positive-stranded, nonsegmented viruses, such as poliovirus,
any siRNAs directed against its genome would result in
down regulation of the entire polyprotein. In contrast, for
negative-stranded and segmented RNA viruses, such as
paramyxoviruses and orthomyxoviruses, it is possible to
downregulate specific viral genes using RNAi. This pro-
vides the basis for a novel approach in which individual
viral proteins can be downregulated with RNAi to be re-
placed with modified proteins provided in trans (e.g., from
plasmid vectors).
Viruses depend on host factors at every stage of their
replication cycle. To establish the role of putative host fac-
tors in virus replication, RNAi-based downregulation may
15
be a powerful approach. Traditional knockout strategies are
labor intensive and may result in cellular toxicity, as can be
expected for proteins that play important roles in basic bio-
chemical processes in the host cell. For example, the poly
A binding protein has been implicated in positive-stranded
RNA virus replication, such as poliovirus (Herold and
Andino, 2001; Svitkin et al., 2001). However, the protein
is also involved in translation initiation and is essential
for cell growth in Saccharomyces cerevisiae (Sachs et al.,
1987). Transient, inducible and non-complete knock-down
by RNAi may not interfere with cell survival, but still
provide a window of opportunity to study the role of cel-
lular factors in virus replication in the cell. Furthermore,
complementation by an expression plasmid that encodes
the protein of interest in a mutated form could provide
additional information about sites of interaction.
For poliovirus, host factors involved in cap-independent
translation initiation and negative strand RNA synthesis have
begun to be defined through in vitro studies. Factors that have
been implied in these processes include canonical translation
initiation factors, poly A binding protein, poly(rC) binding
proteins 1 and 2, La antigen, and poly-pyrimidine tract bind-
ing proteins (reviewed in Andino et al., 1999; Belsham and
Sonenberg, 2000; Semler and Wimmer, 2002). Additional
interactions with host factors include the interaction with the
poliovirus receptor (CD155) for viral entry (Ren et al., 1990)
and possible interactions with proteins of the COPII system
leading to alterations in the secretory pathway to establish
membranous vesicles which function as sites for virus repli-
cation (Rust et al., 2001). It is likely that in the near future
approaches that employ downregulation of these factors us-
ing RNAi will yield insights in their specific function during
viral replication.
7. Conclusion
Much information about the mechanism and the use of
RNAi as a tool to manipulate gene expression has become
available in recent years. A better understanding of this
novel defense mechanism may result in effective therapeutic
approaches for intractable viral diseases that are currently
afflicting humans. However, although poliovirus (and other
viruses) can be targeted by artificially introduced siRNA
in vitro, still much research needs to be done to establish
the role of RNAi in natural virus infection, and several
critical issues need to be addressed, before RNAi-based
technology might be used for the treatment of viral
diseases.
Acknowledgements
This work was financially supported by an EMBO long
term fellowship to RvR and by NIH grant AI40085 to R.A.
We would like to thank L. Gitlin for helpful discussions.
16 M.-C. Saleh et al. / Virus Research 102 (2004) 11–17
References
Andino, R., Boddeker, N., Silvera, D., Gamarnik, A.V., 1999. Intracellular
determinants of picornavirus replication. Trends Microbiol. 7, 76–82.
Belsham, G.J., Sonenberg, N., 2000. Picornavirus RNA translation: roles
for cellular proteins. Trends Microbiol. 8, 330–335.
Bernstein, E., Caudy, A.A., Hammond, S.M., Hannon, G.J., 2001a. Role
for a bidentate ribonuclease in the initiation step of RNA interference.
Nature 409, 363–366.
Bernstein, E., Denli, A.M., Hannon, G.J., 2001b. The rest is silence. RNA
7, 1509–1521.
Bohmert, K., Camus, I., Bellini, C., Bouchez, D., Caboche, M., Benning,
C., 1998. AGO1 defines a novel locus of Arabidopsis controlling leaf
development. EMBO J. 17, 170–180.
Brigneti, G., Voinnet, O., Li, W.X., Ji, L.H., Ding, S.W., Baulcombe, D.C.,
1998. Viral pathogenicity determinants are suppressors of transgene
silencing in Nicotiana benthamiana. EMBO J. 17, 6739–6746.
Chi, J.T., Chang, H.Y., Wang, N.N., Chang, D.S., Dunphy, N., Brown,
P.O., 2003. Genomewide view of gene silencing by small interfering
RNAs. Proc. Natl. Acad. Sci. U.S.A. 100, 6343–6346.
Cox, D.N., Chao, A., Baker, J., Chang, L., Qiao, D., Lin, H., 1998. A
novel class of evolutionarily conserved genes defined by piwi are
essential for stem cell self-renewal. Genes Dev. 12, 3715–3727.
Elbashir, S.M., Martinez, J., Patkaniowska, A., Lendeckel, W., Tuschl, T.,
2001. Functional anatomy of siRNAs for mediating efficient RNAi in
Drosophila melanogaster embryo lysate. EMBO J. 20, 6877–6888.
Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver, S.E., Mello,
C.C., 1998. Potent and specific genetic interference by double-stranded
RNA in Caenorhabditis elegans. Nature 391, 806–811.
Gitlin, L., Andino, R., 2003. Nucleic acid-based immune system: the
antiviral potential of mammalian RNA silencing. J. Virol. 77, 7159–
7165.
Gitlin, L., Karelsky, S., Andino, R., 2002. Short interfering RNA confers
intracellular antiviral immunity in human cells. Nature 418, 430–434.
Guidotti, L.G., Chisari, F.V., 2001. Noncytolytic control of viral infections
by the innate and adaptive immune response. Annu. Rev. Immunol.
19, 65–91.
Hammond, S.M., Boettcher, S., Caudy, A.A., Kobayashi, R., Hannon, G.J.,
2001a. Argonaute2, a link between genetic and biochemical analyses
of RNAi. Science 293, 1146–1150.
Hammond, S.M., Caudy, A.A., Hannon, G.J., 2001b. Post-transcriptional
gene silencing by double-stranded RNA. Nat. Rev. Genet. 2, 110–119.
Harborth, J., Elbashir, S.M., Bechert, K., Tuschl, T., Weber, K., 2001.
Identification of essential genes in cultured mammalian cells using
small interfering RNAs. J. Cell Sci. 114, 4557–4565.
Herold, J., Andino, R., 2001. Poliovirus RNA replication requires genome
circularization through a protein–protein bridge. Mol. Cell 7, 581–591.
Jacobsen, S.E., Running, M.P., Meyerowitz, E.M., 1999. Disruption of an
RNA helicase/RNAse III gene in Arabidopsis causes unregulated cell
division in floral meristems. Development 126, 5231–5243.
Jacque, J.M., Triques, K., Stevenson, M., 2002. Modulation of HIV-1
replication by RNA interference. Nature 418, 435–438.
Kapadia, S.B., Brideau-Andersen, A., Chisari, F.V., 2003. Interference of
hepatitis C virus RNA replication by short interfering RNAs. Proc.
Natl. Acad. Sci. U.S.A. 100, 2014–2018.
Kataoka, Y., Takeichi, M., Uemura, T., 2001. Developmental roles and
molecular characterization of a Drosophila homologue of Arabidopsis
Argonaute1, the founder of a novel gene superfamily. Genes Cells 6,
313–325.
Ketting, R.F., Haverkamp, T.H., van Luenen, H.G., Plasterk, R.H., 1999.
Mut-7 of C. elegans, required for transposon silencing and RNA
interference, is a homolog of Werner syndrome helicase and RNaseD.
Cell 99, 133–141.
Lewis, D.L., Hagstrom, J.E., Loomis, A.G., Wolff, J.A., Herweijer, H.,
2002. Efficient delivery of siRNA for inhibition of gene expression
in postnatal mice. Nat. Genet. 32, 107–108.
Li, H.W., Li, W.X., Ding, S.W., 2002. Induction and suppression of RNA
silencing by an animal virus. Science 296, 1319–1321.
Lipardi, C., Wei, Q., Paterson, B.M., 2001. RNAi as random degradative
PCR: siRNA primers convert mRNA into dsRNAs that are degraded
to generate new siRNAs. Cell 107, 297–307.
Liu, F., Song, Y., Liu, D., 1999. Hydrodynamics-based transfection in
animals by systemic administration of plasmid DNA. Gene Ther. 6,
1258–1266.
Lucas, W.J., Yoo, B.C., Kragler, F., 2001. RNA as a long-distance infor-
mation macromolecule in plants. Nat. Rev. Mol. Cell Biol. 2, 849–857.
Martinez, J., Patkaniowska, A., Urlaub, H., Luhrmann, R., Tuschl, T.,
2002a. Single-stranded sntisense siRNAs guide target RNA cleavage
in RNAi. Cell 110, 563–574.
Martinez, L.A., Naguibneva, I., Lehrmann, H., Vervisch, A., Tchenio,
T., Lozano, G., Harel-Bellan, A., 2002b. Synthetic small inhibiting
RNAs: efficient tools to inactivate oncogenic mutations and restore
p53 pathways. Proc. Natl. Acad. Sci. U.S.A. 99, 14849–14854.
McCaffrey, A.P., Meuse, L., Pham, T.T., Conklin, D.S., Hannon, G.J.,
Kay, M.A., 2002. RNA interference in adult mice. Nature 418, 38–39.
McCaffrey, A.P., Nakai, H., Pandey, K., Huang, Z., Salazar, F.H., Xu, H.,
Wieland, S.F., Marion, P.L., Kay, M.A., 2003. Inhibition of hepatitis
B virus replication in mice by RNA interference. Nat. Biotechnol. 21,
639–644.
Minor, P.D., 1997. Poliovirus. In: Nathanson, N., Ahmed, R. (Eds.), Viral
Pathogenesis. Lippincott-Raven, Philadelphia, pp. 555–574.
Mourrain, P., Beclin, C., Elmayan, T., Feuerbach, F., Godon, C., Morel,
J.B., Jouette, D., Lacombe, A.M., Nikic, S., Picault, N., Remoue, K.,
Sanial, M., Vo, T.A., Vaucheret, H., 2000. Arabidopsis SGS2 and
SGS3 genes are required for posttranscriptional gene silencing and
natural virus resistance. Cell 101, 533–542.
Novina, C.D., Murray, M.F., Dykxhoorn, D.M., Beresford, P.J., Riess, J.,
Lee, S.K., Collman, R.G., Lieberman, J., Shankar, P., Sharp, P.A.,
2002. siRNA-directed inhibition of HIV-1 infection. Nat. Med. 8,
681–686.
Qin, X.F., An, D.S., Chen, I.S., Baltimore, D., 2003. Inhibiting HIV-1 in-
fection in human T cells by lentiviral-mediated delivery of small interf-
ering RNA against CCR5. Proc. Natl. Acad. Sci. U.S.A. 100, 183–188.
Randall, G., Grakoui, A., Rice, C.M., 2003. Clearance of replicating
hepatitis C virus replicon RNAs in cell culture by small interfering
RNAs. Proc. Natl. Acad. Sci. U.S.A. 100, 235–240.
Ren, R.B., Costantini, F., Gorgacz, E.J., Lee, J.J., Racaniello, V.R., 1990.
Transgenic mice expressing a human poliovirus receptor: a new model
for poliomyelitis. Cell 63, 353–362.
Roignant, J.Y., Carre, C., Mugat, B., Szymczak, D., Lepesant, J.A., An-
toniewski, C., 2003. Absence of transitive and systemic pathways al-
lows cell-specific and isoform-specific RNAi in Drosophila. RNA 9,
299–308.
Rueckert, R.R., 1996. Picornaviridae and their replication. In: Fields,
B.N., Knipe, D.M., Howley, P.M., et al. (Eds.), Fields Virology, 3rd
ed. Lippincott-Raven Publishers, Philadelphia, pp. 609–652.
Rust, R.C., Landmann, L., Gosert, R., Tang, B.L., Hong, W., Hauri, H.P.,
Egger, D., Bienz, K., 2001. Cellular COPII proteins are involved in
production of the vesicles that form the poliovirus replication complex.
J. Virol. 75, 9808–9818.
Ruvkun, G., 2001. Molecular biology. Glimpses of a tiny RNA world.
Science 294, 797–799.
Sachs, A.B., Davis, R.W., Kornberg, R.D., 1987. A single domain of yeast
poly(A)-binding protein is necessary and sufficient for RNA binding
and cell viability. Mol. Cell Biol. 7, 3268–3276.
Semler, B.L., Wimmer, E., 2002. Molecular Biology of Picornaviruses.
American Society for Microbiology Press, Washington.
Seo, M.Y., Abrignani, S., Houghton, M., Han, J.H., 2003. Small interfering
RNA-mediated inhibition of hepatitis C virus replication in the human
hepatoma cell line Huh-7. J. Virol. 77, 810–812.
Song, E., Lee, S.K., Wang, J., Ince, N., Ouyang, N., Min, J., Chen,
J., Shankar, P., Lieberman, J., 2003. RNA interference targeting Fas
protects mice from fulminant hepatitis. Nat. Med. 9, 347–351.
 M.-C. Saleh et al. / Virus Research 102 (2004) 11–17
Stein, P., Svoboda, P., Anger, M., Schultz, R.M., 2003. RNAi: mammalian
oocytes do it without RNA-dependent RNA polymerase. RNA 9, 187–
192.
Svitkin, Y.V., Imataka, H., Khaleghpour, K., Kahvejian, A., Liebig, H.D.,
Sonenberg, N., 2001. Poly(A)-binding protein interaction with elF4G
stimulates picornavirus IRES-dependent translation. RNA 7, 1743–
1752.
Thomas, C.E., Ehrhardt, A., Kay, M.A., 2003. Progress and problems with
the use of viral vectors for gene therapy. Nat. Rev. Genet. 4, 346–358.
Tiscornia, G., Singer, O., Ikawa, M., Verma, I.M., 2003. A general method
for gene knockdown in mice by using lentiviral vectors expressing
small interfering RNA. Proc. Natl. Acad. Sci. U.S.A. 100, 1844–1848.
Voinnet, O., 2001. RNA silencing as a plant immune system against
viruses. Trends Genet. 17, 449–459.
Voinnet, O., Pinto, Y.M., Baulcombe, D.C., 1999. Suppression of gene
silencing: a general strategy used by diverse DNA and RNA viruses
of plants. Proc. Natl. Acad. Sci. U.S.A. 96, 14147–14152.
17
Waterhouse, P.M., Wang, M.B., Lough, T., 2001. Gene silencing as an
adaptive defence against viruses. Nature 411, 834–842.
Wilson, J.A., Jayasena, S., Khvorova, A., Sabatinos, S., Rodrigue-Gervais,
I.G., Arya, S., Sarangi, F., Harris-Brandts, M., Beaulieu, S., Richard-
son, C.D., 2003. RNA interference blocks gene expression and RNA
synthesis from hepatitis C replicons propagated in human liver cells.
Proc. Natl. Acad. Sci. U.S.A. 100, 2783–2788.
Xia, H., Mao, Q., Paulson, H.L., Davidson, B.L., 2002. siRNA-mediated
gene silencing in vitro and in vivo. Nat. Biotechnol. 10, 1006–1010.
Yokota, T., Sakamoto, N., Enomoto, N., Tanabe, Y., Miyagishi, M.,
Maekawa, S., Yi, L., Kurosaki, M., Taira, K., Watanabe, M., Mizu-
sawa, H., 2003. Inhibition of intracellular hepatitis C virus replication
by synthetic and vector-derived small interfering RNAs. EMBO Rep.
4, 602–608.
Zhang, G., Budker, V., Wolff, J.A., 1999. High levels of foreign gene
expression in hepatocytes after tail vein injections of naked plasmid
DNA. Hum. Gene. Ther. 10, 1735–1737.
 Virus Research 102 (2004) 19–25

Interfering with hepatitis C virus RNA replication

Glenn Randall, Charles M. Rice∗
Laboratory of Virology and Infectious Disease, Center for the Study of Hepatitis C, The Rockefeller University,
1230 York Avenue, Box 64, New York, NY 10021, USA

Abstract

The emergence of RNA interference (RNAi) as a powerful tool for silencing gene expression has spurred considerable interest in its
experimental and therapeutic potential. RNAi is a cellular process of gene silencing in which small duplexes of RNA specifically target a
homologous sequence for cleavage by cellular ribonucleases. The introduction of 21–23 nucleotide RNA duplexes, termed small interfering
RNAs (siRNAs), into mammalian cells can specifically degrade homologous mRNAs. RNAi efficiently silences the expression of both cellular
and viral RNAs. A number of groups have demonstrated that siRNAs interfere with hepatitis C virus (HCV) gene expression and replication.
Additionally, cellular genes are efficiently silenced in the presence of replicating HCV. These studies lay the foundation for using RNAi as an
experimental tool for studying HCV replication and defining host genes that are significant for viral replication. The potential for RNAi as an
antiviral therapy remains less clear, as it will face many of the challenges that have hindered nucleic acid therapies in the past.
© 2004 Elsevier B.V. All rights reserved.

Keywords: Hepatitis C virus; RNA interference; Small interfering RNAs; Antiviral agents

1. RNA interference 2002). siRNAs are unwound, and single-stranded siRNA

Mechanisms of gene silencing by the targeting of RNAs
for degradation are conserved in plants, fungi, and animals.
While post-transcriptional gene silencing (PTGS) has been
described in plants for over a decade, the demonstration in
2001 that the closely related process of RNAi has activity
in mammalian cells has greatly expanded the size of its
scientific audience (Elbashir et al., 2001). It has since been
concluded that the molecular machinery of gene silencing
appears to be ancient and highly conserved (reviewed in
Bosher and Labouesse, 2000; Hannon, 2002). The mecha-
nism and functions of RNAi are described elsewhere in this
issue, so they will be only briefly summarized in this review.
It has been shown in Drosophila that larger double-stranded
RNAs (dsRNAs) are processed into short ∼22-mer siRNAs
by an RNase III-like enzyme, dicer (Bernstein et al., 2001).
These siRNAs associate with the RNA-induced silencing
complex (RISC) (Nykanen et al., 2001). RISC is a multi-
protein complex composed of members of the argonaute
family. In humans, there are seven identified argonaute
family members: hAgo1/eIF2C1, hAgo2/eIF2C2, hAgo3,
hAgo4, Hili, Hiwi, and Hiwi2 (reviewed in Carmell et al.,
∗ Corresponding author. Tel.: +1-212-327-7046;
fax: +1-212-327-7048.
E-mail address: ricec@rockefeller.edu (C.M. Rice).
then acts as a guide to substrate selection, leading to the
cleavage of a homologous target RNA molecule (Hammond
et al., 2000).
The demonstration of RNAi in mammalian cells was
complicated by the existence of the interferon (IFN)
signaling pathway. dsRNAs longer than 30 nucleotides
efficiently induce components of the IFN response, in-
cluding the dsRNA-activated protein kinase R (PKR),
2 ,5 -oligoadenylate synthetase, and RNase L (Manche
et al., 1992; Minks et al., 1979). These factors promote a
global inhibition of gene expression by inhibiting transla-
tion via the phosphorylation of translation initiation factor
eIF2␣ by PKR and degradation of RNAs by RNase L.
The induction of IFN was avoided by introducing chem-
ically synthesized 21-mer siRNAs into mammalian cells.
A siRNA homologous to the cellular mRNA encoding the
nuclear filament protein lamin A/C specifically silenced the
target gene expression in mammalian cells (Elbashir et al.,
2001). Silencing was effective, but transient, as the effect
diminished after 5 days. The transient nature of silencing is
likely due to the dilution of siRNAs by cell division. These
results demonstrated the activity of a gene silencing path-
way that is induced by dsRNA, similar to IFN responses.
Unlike IFN, however, RNAi initiates a sequence specific
silencing that is dependent on entirely different molecular
machinery.

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.011
20 G. Randall, C.M. Rice / Virus Research 102 (2004) 19–25

RNAi has been proposed to play a central role in the reg-
ulation of many cellular processes. It has been demonstrated
to silence the expression of transgenes and transposons in
plants. It has been implicated in development and stem cell
fate (reviewed in Carmell et al., 2002). RNAi also plays a
role in chromosome maintenance. In Saccharomyces pombe,
the disruption of RNAi resulted in a loss of silencing het-
erochromatic repeats at chromosomal centromeres (Volpe
et al., 2002). The related PTGS is a well-documented an-
tiviral system in plants (Al-Kaff et al., 1998; Dougherty
et al., 1994; Ruiz et al., 1998), and RNAi has been shown to
have antiviral function in insect cells (Li et al., 2002). These
processes silence the expression of transposons, repetitive
elements, and viruses (reviewed in Voinnet et al., 1999).
Viral inhibitors of RNAi have been isolated from plant
viruses (Brigneti et al., 1998; Li and Ding, 2001; Voinnet
et al., 2000) and also the animal virus, flock house virus
(Li et al., 2002; Lindenbach and Rice, 2002). It is currently
unclear whether RNAi is an innate antiviral defense in
mammals.
2. Hepatitis C virus
HCV is a major public health problem, with 170 mil-
lion chronically infected people throughout the world
(Anonymous, 1997). Chronically infected individuals are
a reservoir for new infections as well as being at risk for
progression to cirrhosis and hepatocellular carcinoma. Con-
sequently, HCV infection is the leading indicator of liver
transplant (Fishman et al., 1996). Antiviral therapy con-
sisting of peginterferon-alfa 2a-ribavirin shows sustained
response in only about half of the treated patients that are in-
fected with HCV genotypes 1 and 4 (Heathcote et al., 2000;
Zeuzem et al., 2000). These factors have led to a push to de-
velop new antiviral therapies. Additionally, the emergence
of drug resistance in RNA virus infections emphasizes the
need for multiple drug targets to limit viral escape.
HCV belongs to the family Flaviviridae, which includes
the flaviviruses, pestiviruses, hepaciviruses, and the GB
viruses (reviewed in Lindenbach and Rice, 2001). The
9.6-kb HCV RNA genome of positive (+) polarity consists

Fig. 1. Schematic representation of the HCV genome and siRNAs reported to silence HCV. (A) Shown is the HCV genome; with the 5 NTR, followed
by a large open reading frame that is translated and processed into at least 10 proteins (C, E1, E2, p7, NS2, NS3, NS4A, NS4B, NS5A, and NS5B) and
the 3 NTR. The numbers and arrows refer to the siRNA sites. (B) siRNAs that have been reported to successfully silence HCV. siRNAs are described
both by their sequence location (nucleotide) and the names given to them in the respective publications. The nucleotide number given is the position
of the 5 nucleotide of the siRNA within the HCV-Con1 sequence. The reported siRNA sequences, the gene that is targeted by the siRNA, and the
reference for each siRNA are given. Abbreviations: C, core; E, envelope glycoprotein; NS, nonstructural protein; NTR, nontranslated region.
 G. Randall, C.M. Rice / Virus Research 102 (2004) 19–25
of a 5 nontranslated region (NTR) containing an internal
ribosome entry site (IRES), a single open reading frame
of approximately 3000 codons, and a 3 NTR containing
a highly conserved 98-nucleotide sequence (Fig. 1A). The
IRES directs the translation of a polypeptide, which is sub-
sequently processed by host and viral proteases into at least
10 individual proteins. The structural proteins C (core),
E1, and E2 (envelope glycoproteins) are contained at the
N-terminus of the polyprotein, and are followed by p7
and the nonstructural proteins (NS) 2, 3, 4A, 4B, 5A, and
5B. NS 3–5B comprise a functional replication complex
that promotes transcription of a genomic (−) strand inter-
mediate. This serves as a template for production of (+)
strands that are either translated or packaged into virions as
genomic RNAs.
The structural proteins C, E1, and E2 compose the cap-
sid and constituents of the viral envelope. E1 and E2 are
thought to mediate HCV binding to the cell and viral en-
try. Enzymatic properties of HCV proteins are well defined;
NS5B is the RNA-dependent RNA polymerase, NS3 has
helicase and serine protease activity, and NS2-NS3 form a
distinct autoprotease. NS4A is a cofactor for NS3 proteol-
ysis. The functions of p7, NS2 following autoproteolysis,
NS4B, and NS5A are less clear. A number of studies have
attempted to define functions of viral proteins, particularly
NS5A and the core protein, by ectopic protein expression or
the identification of protein interaction partners (reviewed
in Tellinghuisen and Rice, 2002). The significance of these
putative functions and interactions for HCV replication was
not tested by genetic analysis, owing to the lack of an effi-
cient cell culture system.
Although HCV is notoriously difficult to grow in vitro,
systems have been developed that sustain efficient repli-
cation of HCV in cell culture. These are replicon-based
strategies in which the HCV IRES drives expression of the
neomycin phosphotransferase gene, while a heterologous
IRES from encephalomyocarditis virus (EMCV) promotes
translation of the HCV polypeptide (Blight et al., 2000;
Lohmann et al., 1999). HCV replicon replication promotes
continued expression of the neomycin phosphotransferase
gene that confers G418 resistance to cells. This allows the
establishment of cell populations that contain replicating
HCV. G418-resistant foci are also a means of quantifying
the number of cells that contain stably replicating HCV. The
replicon system is useful for dissection of HCV replication
and protein functions; however, the production of infectious
virus in vitro has yet to be documented. In addition to molec-
ular studies of HCV replication, replicons provide an excel-
lent system to evaluate HCV antiviral agents in cell culture
(reviewed in Randall and Rice, 2001).
3. HCV is sensitive to RNAi
While the initial demonstrations of RNAi in mammalian
cells showed silencing of cellular transcripts, virologists
21
quickly began to study the relevance of RNAi to their virus
of interest. The leap for hepatitis C virologists is relatively
small. As an RNA virus, the genome may be susceptible to
cleavage. Since HCV replication generates dsRNAs, HCV
may naturally activate RNAi, which would in turn serve as
an innate host defense against virus infection. Finally, a great
need exists for the development of effective HCV antiviral
therapies. The specific and effective nature of silencing by
RNAi are attractive therapeutic characteristics.
A number of groups have used the HCV replicon system
to evaluate the antiviral potential of RNAi (Kapadia et al.,
2003; Randall et al., 2003; Seo et al., 2003; Wilson et al.,
2003; Yokota et al., 2003). These studies have focused pri-
marily on two questions: (i) Does HCV interfere with RNAi?
(ii) Is HCV sensitive to RNAi? Regarding the first question,
it appears that HCV does not inhibit RNAi following the in-
troduction of siRNAs into cells containing replicating HCV.
Both cellular (lamin A/C) and viral (HCV) RNAs were effec-
tively silenced in the presence of replicating HCV (Randall
et al., 2003). The efficiencies of silencing lamin A/C expres-
sion in Huh-7.5 cells were similar either in the presence or
absence of replicating HCV RNAs. Since the introduction
of siRNAs into cells bypasses the initial processing steps of
larger dsRNAs into siRNAs by Dicer, it is possible that HCV
may interfere with an earlier step in the RNAi pathway. It
should be re-emphasized that so far no viruses that infect
mammals have been reported to encode inhibitors of RNAi.
A general consensus was reached that HCV RNAs, like
those of many other viruses, are indeed sensitive to RNAi.
siRNAs were designed to target cleavage of HCV RNAs
at various regions of the HCV genome (Fig. 1). The in-
troduction of siRNAs into cells containing HCV replicons
effectively silenced HCV expression in a number of differ-
ent assays. HCV antigen expression declined after treatment
with siRNAs, as assayed by either the level of replicon
reporter expression or by the levels of HCV protein expres-
sion in all studies. siRNAs triggered an exponential decline
in HCV RNA levels, with a maximal reduction achieved
and maintained after 4 days of a single siRNA treatment
(Randall et al., 2003). Strand-specific probing for HCV
RNAs revealed that siRNAs reduced the levels of both (+)
and (−) strand RNA (Wilson et al., 2003).
As shown in Fig. 1B, the siRNAs effectively targeted re-
gions in the 5 NTR, Core, NS3, NS4B, and NS5B. While
the relative efficiencies of individual siRNAs vary, it appears
that all regions of HCV RNA are accessible to the RNAi
machinery. It should be noted that not all siRNAs were ef-
fective. Success rates varied between studies, but it appears
that approximately half of the siRNAs were effective in at
least partially silencing HCV.
The efficiencies of silencing HCV varied depending on
the experimental conditions of each study, probably indica-
tive of the efficiency of siRNA introduction into cells. Under
optimal conditions, it was shown that siRNAs could clear
replicating HCV RNAs from >98% of cells in a series of two
experiments (Randall et al., 2003). In the first experiment,
22 G. Randall, C.M. Rice / Virus Research 102 (2004) 19–25
HCV NS5A expression was only detectable by immunoflu-
orescence in 1% of cells, as compared with cells treated
with an irrelevant siRNA. The second experiment examined
the formation of G418-resistant colonies following treatment
by siRNAs. As described earlier, G418-resistant colonies
form dependent upon HCV replication and the expression of
the neomycin phosphotransferase gene from HCV replicon
RNAs. Cells containing HCV replicons were treated with
siRNAs and grown in the absence of selection for 1 week.
G418 selection was then reapplied and G418 resistant foci
were counted 10 days later. Cells treated with HCV siR-
NAs formed G418 resistant foci in less than 2% of cells, as
compared with cells treated with irrelevant siRNAs. HCV
siRNAs did not affect cell viability, since treated cells grew
in the absence of selection. It was unclear whether the re-
maining G418 resistant cells received sufficient levels of the
siRNAs; or alternatively, if these cells contained viral RNAs
which escaped RNAi either due to mutation of viral se-
quences or mutations in the RNAi machinery. Nonetheless,
the ability of siRNAs to produce noncytolytic clearance of
HCV is attractive from a therapeutic viewpoint.
A hallmark of RNAi, as compared with IFN, is the
sequence-specific nature of RNAi gene silencing. siRNAs
that differed from their HCV target sequence by two or more
bases did not efficiently silence HCV. In one study, two
HCV genotype 1b replicons that differed from each other by
three bases within the siRNA target sequence were silenced
only by the exact homologous siRNA (Randall et al., 2003).
The introduction of siRNAs did not induce the expression
of IFN-induced genes MxA and PKR, and additionally,
siRNAs were more effective at reducing HCV RNA levels
than high doses (100 units/ml) of IFN-␣ (Kapadia et al.,
2003). It was not tested whether the addition of IFN-␣ and
HCV siRNAs showed an additive effect in reducing HCV
RNA levels. Taken together, these experiments showed that
HCV silencing by siRNAs was independent of, and perhaps
more effective than, the effects of IFN-␣.
The previously described experiments tested the ability of
siRNAs to clear replicating HCV RNAs from cells. It was
also demonstrated that the introduction of HCV siRNAs into
Huh-7 cells prior to HCV replicon RNAs precluded the es-
tablishment of HCV replication (Wilson et al., 2003). This
effect was transient, with efficient inhibition of replication
lasting for 5 days. In order to improve upon the transient
nature of silencing following siRNAs, a number of strate-
gies now exist for stable RNAi. These are based on vectors
containing a pol III promoter driving expression of either (i)
dual single stranded RNAs that could form a ds siRNA or
(ii) small hairpin RNAs (shRNAs) which are then processed
in vivo to yield a functional siRNA (Brummelkamp et al.,
2002; Paddison et al., 2002). It was examined whether HCV
RNAs could be silenced using vector-based stable silencing.
HCV antigen expression was partially silenced following
the transient introduction of the vectors into cells containing
replicating HCV (Yokota et al., 2003). In another study, sta-
ble cell populations were established that expressed vector
derived siRNAs (Wilson et al., 2003). The majority of cells
within this population were resistant to the establishment
of HCV replication. Recently, vectors that express shRNAs,
but not chemically synthesized siRNAs, have been shown
to induce interferon in a subset of constructs (Bridge et al.,
2003). This result is likely to be due to either the increased
length of dsRNA (siRNA stem plus an extended loop) or
aberrant transcription initiation or termination resulting in
long dsRNAs. The studies on HCV silencing by siRNA pro-
ducing vectors will need to be re-examined for the induction
of interferon. Assuming that these vectors silence HCV in
an interferon-independent manner, these results establish the
concept of creating HCV resistant cells. Additionally, the
stable expressing cell populations should provide a model
system to study viral escape from RNAi.
4. Prospects and questions for HCV and RNAi
Numerous questions follow these initial studies that
demonstrated the sensitivity of HCV to RNAi. Some of
these include the following: (i) Is RNAi an innate antiviral
pathway in mammals? (ii) What is the mechanism of HCV
clearance? (iii) How do we exploit RNAi to define the sig-
nificance of host interactions for the HCV life cycle? (iv)
What is the therapeutic potential of RNAi?
4.1. Is RNAi an innate antiviral pathway in mammals?
To date, it has not been demonstrated that RNAi is an
innate antiviral defense in mammals; or conversely, that
viruses express functions that inhibit mammalian RNAi. Ad-
ditionally, a large number of viruses have been reported to
be sensitive to RNAi (reviewed in Gitlin and Andino, 2003).
Since mammals have an immune system and interferon path-
way that perform antiviral roles, it is possible that RNAi
has evolved to perform other gene regulatory functions in
mammals. To determine whether RNAi is an innate antiviral
pathway, researchers need to address whether (i) HCV repli-
cation induces the expression or activity of components of
the RNAi pathway, (ii) replicating HCV is naturally targeted
by the RNAi machinery in the absence of experimentally
introduced siRNAs or shRNAs, (iii) inhibition of RNAi en-
hances HCV replication, and (iv) HCV encodes inhibitors
of the RNAi pathway.
4.2. What is the mechanism of anti-HCV action for siRNAs?
Replicating HCV RNAs are at least partially double
stranded and should in theory be susceptible to RNAi; how-
ever, the RNAs may be resistant to the RNAi machinery due
to either protection from ribonucleoproteins or a privileged
subcellular localization. The targeting of mRNAs, but not
genomic viral RNAs has been proposed as the mechanism
of siRNA clearance for both influenza virus and respiratory
syncytial virus (Bitko and Barik, 2001; Ge et al., 2003). It is
 G. Randall, C.M. Rice / Virus Research 102 (2004) 19–25
unclear whether the anti-HCV activity of siRNAs functions
by a similar mechanism. This is difficult to test in the case
of HCV, since the (+) strand can represent both genome
and message.
The majority of siRNAs that efficiently silence HCV were
designed to target the (+) strand, or both (+) and (−) strand.
In one study, a reduction of both (+) and (−) strand RNAs
was observed following silencing (Wilson et al., 2003). This
suggest that either (i) the HCV replicase and minus strand
RNAs in particular are sensitive to RNAi, or (ii) the HCV
replication complex is relatively unstable and that targeting
of the (+) strands indirectly results in a decrease in (−)
strands. One siRNA (286, Fig. 1B) has been described that
should target only (−) strands (Seo et al., 2003; Yokota et al.,
2003). When compared with other siRNAs in the study, it
was partially effective at silencing HCV. A 30% reduction
in HCV replicon reporter activity was observed as compared
with a 90% reduction for the optimal siRNA in this exper-
iment (Yokota et al., 2003). Although cleavage of the (−)
strand was not demonstrated, this result suggests that the
replicase complex may be at least partially accessible to siR-
NAs. It remains to be determined whether polysomal RNAs
and RNAs in the replication complex, as either templates or
nascent products, are similarly accessible to RNAi-mediated
cleavage.
4.3. How do we exploit RNAi to define the significance of
host interactions for the HCV life cycle?
HCV has an inordinately large number of reported
viral–host interactions that have no known significance in
the viral life cycle. Since HCV does not interfere with the
silencing of host genes, RNAi has significant potential as a
tool to investigate the significance of these interactions for
the HCV life cycle. The applicability of RNAi is currently
limited by the systems with which to characterize the HCV
life cycle. The previously described replicon system holds
great promise in characterizing the significance of host
genes for the translation and replication of HCV RNAs.
This approach would encompass host genes that were iden-
tified to interact with HCV RNAs and components of the
viral replicase complex (NS3, 4A, 4B, 5A, and 5B). Ad-
ditionally, cellular signaling pathways that intersect with
HCV can be similarly examined. For example, the mech-
anism of action for IFN-␣, -␤, or -␥ in the clearance of
HCV RNAs is currently unknown, both in vitro and in vivo.
IFN-induced genes can be silenced prior to the addition of
IFN to identify IFN-induced genes that perform effector
functions able to promote the clearance of HCV.
While systems that promote HCV assembly and egress
are still elusive, improved models to study HCV entry have
been described recently. Retroviral pseudotypes, in which
HIV genomes are enveloped in HCV glycoproteins, infect a
subset of liver cell lines in a HCV glycoprotein-dependent
manner (Bartosch et al., 2003; Hsu et al., 2003). This model
represents an advance compared with previous surrogate
23
systems to early stages of the HCV life cycle since HCV
glycoprotein binding and viral entry into cells can be stud-
ied. Earlier studies tested the binding efficiencies of solu-
ble, truncated versions of HCV glycoproteins either to cells,
or cellular proteins that were proposed to represent puta-
tive HCV receptors. This led to the identification of several
putative receptors, including CD81, low density lipid recep-
tor, scavenger receptor B1, DC-SIGN and L-SIGN (Agnello
et al., 1999; Gardner et al., 2003; Lozach et al., 2003;
Pileri et al., 1998; Pohlmann et al., 2003; Scarselli et al.,
2002). The significance of these putative receptors for HCV
glycoprotein-dependent entry can be tested by silencing the
receptor candidates with siRNAs and testing the ability of
these cells to be infected with HIV-HCV pseudotypes. Sim-
ilar approaches have confirmed the significance of CD4 and
CCR5 for HIV entry (Martinez et al., 2002; Novina et al.,
2002). RNAi will continue to be a powerful tool to probe
HCV–host interactions as systems to study other aspects of
the viral life cycle, such as viral assembly and egress, are
developed.
4.4. What is the therapeutic potential of RNAi?
The initial studies demonstrating that HCV is suscep-
tible to RNAi provide proof of principle for RNAi-based
anti-HCV therapy. However, many obstacles remain along
the journey from “bench to bedside.” These include issues
regarding efficacy, delivery, and resistance/viral escape. Effi-
cacy has traditionally been a stumbling block for nucleic acid
therapy strategies, whether they be naked nucleic acids or vi-
ral vectors designed to express the therapeutic nucleic acids.
RNAs are generally unstable in cells and patient sera. At-
tempts are underway to both stabilize and target chemically
synthesized RNAs in vivo without foregoing the efficacy of
these RNAs for silencing gene expression. One advantage to
HCV as a disease model is that the liver is relatively profi-
cient in the uptake of nucleic acids. Indeed, the liver has been
the organ of choice for demonstrating the potential of RNAi
therapy in vivo. Liver uptake of siRNAs in mice is efficient,
resulting in the silencing of the targeted mRNA sequence,
whether it is an HCV transgene or cellular genes (McCaffrey
et al., 2002). Silencing of Fas or caspase 8 by siRNAs in vivo
prevents the symptoms of hepatitis in mouse model systems
(Song et al., 2003; Zender et al., 2003). Recently, RNAi was
shown to inhibit hepatitis B virus replication in mouse liver
(McCaffrey et al., 2003). The anti-HCV potential of siRNAs
in a system that produces infectious virus has yet to be doc-
umented. Mice that contain chimeric mouse/human livers
support viral infection and replication and would provide a
model to test the efficacy of siRNAs against HCV in vivo,
in a fully infectious system (Mercer et al., 2001).
The specific silencing by RNAi without induction of non-
specific IFN responses is attractive from a therapeutic stand-
point. However, the specificity of silencing highlights the
possibility of viral escape mutants. A limited number of
highly conserved RNA sequences that should be constrained
24 G. Randall, C.M. Rice / Virus Research 102 (2004) 19–25
for viral escape exist in the HCV 5 and 3 NTRs. Mul-
tiple siRNAs targeting evolutionarily restricted HCV RNA
sequences may limit the emergence of escape mutants. Sta-
ble RNAi silencing vectors have been used to generate cell
lines that constitutively express high levels of HCV siRNAs
(Wilson et al., 2003). These cell lines could be used to ad-
dress the question of viral escape in cell culture since the
caveats associated with the transfection efficiency of chem-
ically synthesized siRNAs into cells are avoided. Variations
of these stable RNAi silencing vectors may also function in
gene therapy wherein numerous modes exist for expressing
therapeutic RNAs, such as antisense RNAs or ribozymes.
A goal would be the prevention of productive HCV infec-
tion by siRNA expression in liver cells. Successful treat-
ment may not require a high efficiency of siRNA delivery
since the liver has regenerative potential. Infected liver cells
that are killed by T cell-mediated cytolytic clearance would
gradually be replaced by HCV resistant cells. In theory, this
would result in a HCV resistant liver repopulated by cells
expressing HCV siRNAs.
5. Conclusions
Recent studies have solidified the use of siRNAs to inhibit
the expression of target RNAs, whether they are cellular or
viral in origin. RNAi as an “experimental tool” will undoubt-
edly be exploited to define HCV replication and pathogene-
sis. These experimental approaches can be applied to under-
standing HCV RNA replication and the significance of host
genes at various stages of the viral life cycle. A consensus
has been achieved that HCV is sensitive to the introduction
of siRNAs. Although RNAi can be activated to clear HCV
RNAs, it remains to be shown whether RNAi is a natural host
defense against HCV. Similarly, it will be interesting to see
whether HCV gene products interact with the RNAi machin-
ery. Finally, a major reason for studying HCV is the clini-
cal need for improved therapeutics. Despite many obstacles,
RNAi provides a new approach that, in addition to traditional
anti-viral therapeutics, may effectively treat HCV infection.
Acknowledgements
G.R. and C.M.R. are supported in part by grants from
the Public Health Service (CA57973 and AI40034) and the
Greenberg Medical Research Institute. G.R. is supported by
postdoctoral fellowship PF-02-016-01-MBC from the Amer-
ican Cancer Society. We thank Catherine Murray for critical
reading of the manuscript.
References
Agnello, V., Abel, G., Elfahal, M., Knight, G.B., Zhang, Q.X., 1999.
Hepatitis C virus and other Flaviviridae viruses enter cells via low
density lipoprotein receptor. Proc. Natl. Acad. Sci. U.S.A. 96, 12766–
12771.
Al-Kaff, N.S., Covey, S.N., Kreike, M.M., Page, A.M., Pinder, R., Dale,
P.J., 1998. Transcriptional and posttranscriptional plant gene silencing
in response to a pathogen. Science 279, 2113–2115.
Anonymous, 1997. World Health Organization—Hepatitis C: global preva-
lence. Wkly. Epidemiol. Rec. 72, 341–344.
Bartosch, B., Dubuisson, J., Cosset, F.L., 2003. Infectious hepatitis C
virus pseudo-particles containing functional E1-E2 envelope protein
complexes. J. Exp. Med. 197, 633–642.
Bernstein, E., Caudy, A.A., Hammond, S.M., Hannon, G.J., 2001. Role
for a bidentate ribonuclease in the initiation step of RNA interference.
Nature 409, 363–366.
Bitko, V., Barik, S., 2001. Phenotypic silencing of cytoplasmic genes
using sequence-specific double-stranded short interfering RNA and its
application in the reverse genetics of wild type negative-strand RNA
viruses. BMC Microbiol. 1, 34.
Blight, K.J., Kolykhalov, A.A., Rice, C.M., 2000. Efficient initiation of
HCV RNA replication in cell culture. Science 290, 1972–1974.
Bosher, J.M., Labouesse, M., 2000. RNA interference: genetic wand and
genetic watchdog. Nat. Cell. Biol. 2, E31–E36.
Bridge, A.J., Pebernard, S., Ducraux, A., Nicoulaz, A.L., Iggo, R., 2003.
Induction of an interferon response by RNAi vectors in mammalian
cells. Nat. Genet. 34, 263–264.
Brigneti, G., Voinnet, O., Li, W.X., Ji, L.H., Ding, S.W., Baulcombe, D.C.,
1998. Viral pathogenicity determinants are suppressors of transgene
silencing in Nicotiana benthamiana. EMBO J. 17, 6739–6746.
Brummelkamp, T.R., Bernards, R., Agami, R., 2002. A system for stable
expression of short interfering RNAs in mammalian cells. Science
296, 550–553.
Carmell, M.A., Xuan, Z., Zhang, M.Q., Hannon, G.J., 2002. The Arg-
onaute family: tentacles that reach into RNAi, developmental control,
stem cell maintenance, and tumorigenesis. Genes Dev. 16, 2733–2742.
Dougherty, W.G., Lindbo, J.A., Smith, H.A., Parks, T.D., Swaney, S.,
Proebsting, W.M., 1994. RNA-mediated virus resistance in transgenic
plants: exploitation of a cellular pathway possibly involved in RNA
degradation. Mol. Plant Microb. Interact. 7, 544–552.
Elbashir, S.M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., Tuschl,
T., 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference
in cultured mammalian cells. Nature 411, 494–498.
Fishman, J.A., Rubin, R.H., Koziel, M.J., Periera, B.J., 1996. Hepatitis C
virus and organ transplantation. Transplantation 62, 147–154.
Gardner, J.P., Durso, R.J., Arrigale, R.R., Donovan, G.P., Maddon, P.J.,
Dragic, T., Olson, W.C., 2003. L-SIGN (CD 209L) is a liver-specific
capture receptor for hepatitis C virus. Proc. Natl. Acad. Sci. U.S.A.
100, 4498–4503.
Ge, Q., McManus, M.T., Nguyen, T., Shen, C.H., Sharp, P.A., Eisen,
H.N., Chen, J., 2003. RNA interference of influenza virus production
by directly targeting mRNA for degradation and indirectly inhibiting
all viral RNA transcription. Proc. Natl. Acad. Sci. U.S.A. 100, 2718–
2723.
Gitlin, L., Andino, R., 2003. Nucleic acid-based immune system: the
antiviral potential of mammalian RNA silencing. J. Virol. 77, 7159–
7165.
Hammond, S.M., Bernstein, E., Beach, D., Hannon, G.J., 2000. An
RNA-directed nuclease mediates post-transcriptional gene silencing in
Drosophila cells. Nature 404, 293–296.
Hannon, G.J., 2002. RNA interference. Nature 418, 244–251.
Heathcote, E.J., Shiffman, M.L., Cooksley, W.G., Dusheiko, G.M., Lee,
S.S., Balart, L., Reindollar, R., Reddy, R.K., Wright, T.L., Lin, A.,
Hoffman, J., De Pamphilis, J., 2000. Peginterferon alfa-2a in patients
with chronic hepatitis C and cirrhosis. N. Engl. J. Med. 343, 1673–
1680.
Hsu, M., Zhang, J., Flint, M., Logvinoff, C., Cheng-Mayer, C., Rice,
C.M., McKeating, J.A., 2003. Hepatitis C virus glycoproteins mediate
pH-dependent cell entry of pseudotyped retroviral particles. Proc. Natl.
Acad. Sci. U.S.A. 100, 7271–7276.
 G. Randall, C.M. Rice / Virus Research 102 (2004) 19–25
Kapadia, S.B., Brideau-Andersen, A., Chisari, F.V., 2003. Interference of
hepatitis C virus RNA replication by short interfering RNAs. Proc.
Natl. Acad. Sci. U.S.A. 100, 2014–2018.
Li, W.X., Ding, S.W., 2001. Viral suppressors of RNA silencing. Curr.
Opin. Biotechnol. 12, 150–154.
Li, H., Li, W.X., Ding, S.W., 2002. Induction and suppression of RNA
silencing by an animal virus. Science 296, 1319–1321.
Lindenbach, B.D., Rice, C.M., 2001. Flaviviridae: the viruses and their
replication. In: Knipe, D.M., Howley, P.M. (Eds.), Fields Virology,
4th ed. Lippincott-Williams and Wilkins, Philadelphia, pp. 991–1041.
Lindenbach, B.D., Rice, C.M., 2002. RNAi targeting an animal virus:
news from the front. Mol. Cell. 9, 925–927.
Lohmann, V., Korner, F., Koch, J.O., Herian, U., Theilmann, L., Barten-
schlager, R., 1999. Replication of subgenomic hepatitis C virus RNAs
in a hepatoma cell line. Science 285, 110–113.
Lozach, P.Y., Lortat-Jacob, H., de Lacroix de Lavalette, A., Staropoli, I.,
Foung, S., Amara, A., Houles, C., Fieschi, F., Schwartz, O., Vire-
lizier, J.L., Arenzana-Seisdedos, F., Altmeyer, R., 2003. DC-SIGN
and L-SIGN are high affinity binding receptors for hepatitis C virus
glycoprotein E2. J. Biol. Chem. 278, 20358–20366.
Manche, L., Green, S.R., Schmedt, C., Mathews, M.B., 1992. Interactions
between double-stranded RNA regulators and the protein kinase DAI.
Mol. Cell. Biol. 12, 5238–5248.
Martinez, M.A., Gutierrez, A., Armand-Ugon, M., Blanco, J., Parera, M.,
Gomez, J., Clotet, B., Este, J.A., 2002. Suppression of chemokine
receptor expression by RNA interference allows for inhibition of
HIV-1 replication. AIDS 16, 2385–2390.
McCaffrey, A.P., Meuse, L., Pham, T.T., Conklin, D.S., Hannon, G.J.,
Kay, M.A., 2002. RNA interference in adult mice. Nature 418, 38–39.
McCaffrey, A.P., Nakai, H., Pandey, K., Huang, Z., Salazar, F.H., Xu, H.,
Wieland, S.F., Marion, P.L., Kay, M.A., 2003. Inhibition of hepatitis
B virus in mice by RNA interference. Nat. Biotechnol. 21, 639–644.
Mercer, D.F., Schiller, D.E., Elliott, J.F., Douglas, D.N., Hao, C., Rinfret,
A., Addison, W.R., Fischer, K.P., Churchill, T.A., Lakey, J.R., Tyrrell,
D.L., Kneteman, N.M., 2001. Hepatitis C virus replication in mice
with chimeric human livers. Nat. Med. 7, 927–933.
Minks, M.A., West, D.K., Benvin, S., Baglioni, C., 1979. Structural re-
quirements of double-stranded RNA for the activation of 2 ,5 -oligo(A)
polymerase and protein kinase of interferon-treated HeLa cells. J.
Biol. Chem. 254, 10180–10183.
Novina, C.D., Murray, M.F., Dykxhoorn, D.M., Beresford, P.J., Riess, J.,
Lee, S.K., Collman, R.G., Lieberman, J., Shankar, P., Sharp, P.A.,
2002. siRNA-directed inhibition of HIV-1 infection. Nat. Med. 3, 3.
Nykanen, A., Haley, B., Zamore, P.D., 2001. ATP requirements and small
interfering RNA structure in the RNA interference pathway. Cell 107,
309–321.
Paddison, P.J., Caudy, A.A., Bernstein, E., Hannon, G.J., Conklin, D.S.,
2002. Short hairpin RNAs (shRNAs) induce sequence-specific silenc-
ing in mammalian cells. Genes Dev. 16, 948–958.
Pileri, P., Uematsu, Y., Campagnoli, S., Galli, G., Falugi, F., Petracca,
R., Weiner, A.J., Houghton, M., Rosa, D., Grandi, G., Abrignani, S.,
1998. Binding of hepatitis C virus to CD81. Science 282, 938–941.
25
Pohlmann, S., Zhang, J., Baribaud, F., Chen, Z., Leslie, G.J., Lin, G.,
Granelli-Piperno, A., Doms, R.W., Rice, C.M., McKeating, J.A.,
2003. Hepatitis C virus glycoproteins interact with DC-SIGN and
DC-SIGNR. J. Virol. 77, 4070–4080.
Randall, G., Rice, C.M., 2001. Hepatitis C virus cell culture replication
systems: their potential use for the development of antiviral therapies.
Curr. Opin. Infect. Dis. 14, 743–747.
Randall, G., Grakoui, A., Rice, C.M., 2003. Clearance of replicating
hepatitis C virus replicon RNAs in cell culture by small interfering
RNAs. Proc. Natl. Acad. Sci. U.S.A. 100, 235–240.
Ruiz, M.T., Voinnet, O., Baulcombe, D.C., 1998. Initiation and mainte-
nance of virus-induced gene silencing. Plant Cell 10, 937–946.
Scarselli, E., Ansuini, H., Cerino, R., Roccasecca, R.M., Acali, S., Filo-
camo, G., Traboni, C., Nicosia, A., Cortese, R., Vitelli, A., 2002. The
human scavenger receptor class B type I is a novel candidate receptor
for the hepatitis C virus. EMBO J. 21, 5017–5025.
Seo, M.Y., Abrignani, S., Houghton, M., Han, J.H., 2003. Small interfering
RNA-mediated inhibition of hepatitis C virus replication in the human
hepatoma cell line Huh-7. J. Virol. 77, 810–812.
Song, E., Lee, S.K., Wang, J., Ince, N., Ouyang, N., Min, J., Chen,
J., Shankar, P., Lieberman, J., 2003. RNA interference targeting Fas
protects mice from fulminant hepatitis. Nat. Med. 9, 347–351.
Tellinghuisen, T.L., Rice, C.M., 2002. Interaction between hepatitis C
virus proteins and host cell factors. Curr. Opin. Microbiol. 5, 419–427.
Voinnet, O., Pinto, Y.M., Baulcombe, D.C., 1999. Suppression of gene
silencing: a general strategy used by diverse DNA and RNA viruses
of plants. Proc. Natl. Acad. Sci. U.S.A. 96, 14147–14152.
Voinnet, O., Lederer, C., Baulcombe, D.C., 2000. A viral movement
protein prevents spread of the gene silencing signal in Nicotiana
benthamiana. Cell 103, 157–167.
Volpe, T.A., Kidner, C., Hall, I.M., Teng, G., Grewal, S.I., Martienssen,
R.A., 2002. Regulation of heterochromatic silencing and histone H3
lysine-9 methylation by RNAi. Science 297, 1833–1837.
Wilson, J.A., Jayasena, S., Khvorova, A., Sabatinos, S., Rodrigue-Gervais,
I.G., Arya, S., Sarangi, F., Harris-Brandts, M., Beaulieu, S., Richard-
son, C.D., 2003. RNA interference blocks gene expression and RNA
synthesis from hepatitis C replicons propagated in human liver cells.
Proc. Natl. Acad. Sci. U.S.A. 100, 2783–2788.
Yokota, T., Sakamoto, N., Enomoto, N., Tanabe, Y., Miyagishi, M.,
Maekawa, S., Yi, L., Kurosaki, M., Taira, K., Watanabe, M., Mizu-
sawa, H., 2003. Inhibition of intracellular hepatitis C virus replication
by synthetic and vector-derived small interfering RNAs. EMBO Rep.
4, 602–608.
Zender, L., Hutker, S., Liedtke, C., Tillmann, H.L., Zender, S., Mundt, B.,
Waltemathe, M., Gosling, T., Flemming, P., Malek, N.P., Trautwein,
C., Manns, M.P., Kuhnel, F., Kubicka, S., 2003. Caspase 8 small
interfering RNA prevents acute liver failure in mice. Proc. Natl. Acad.
Sci. U.S.A. 100, 7797–7802.
Zeuzem, S., Feinman, S.V., Rasenack, J., Heathcote, E.J., Lai, M.Y., Gane,
E., O’Grady, J., Reichen, J., Diago, M., Lin, A., Hoffman, J., Brunda,
M.J., 2000. Peginterferon alfa-2a in patients with chronic hepatitis C.
N. Engl. J. Med. 343, 1666–1672.
 Virus Research 102 (2004) 27–35

Control of nonsegmented negative-strand RNA

virus replication by siRNA
Sailen Barik∗
Department of Biochemistry and Molecular Biology, College of Medicine, University of South Alabama,
307 University Blvd., Mobile, AL 36688, USA
I dedicate this article to my Ph.D. advisor and mentor, Professor N.C. Mandal, on the eve of his retirement from Bose Institute,
Calcutta; his love of science continues to be a guiding light

Abstract

Our laboratory provided the first proof-of-concept that double-stranded short interfering RNA (ds-siRNA) can act as potent and specific
antiviral agents. Designed against specific mRNAs of nonsegmented negative-stranded RNA (NNR) viruses, siRNAs abrogated expression
of the corresponding viral proteins, and generated the predicted viral phenotypes. Knockdown was demonstrated across different genera:
respiratory syncytial virus (RSV), a pneumovirus; vesicular stomatitis virus (VSV), a rhabdovirus; and human parainfluenza virus (HPIV),
a paramyxovirus. The targeted genes could have a wide range of functions, thus documenting the versatility of the technique. Interestingly,
antisense single-stranded siRNA (ss-siRNA) was also effective, albeit at a higher concentration. NNR viral genomic and antigenomic RNA,
which are encapsidated by nucleocapsid protein and serve as templates for viral RNA-dependent RNA polymerase, were resistant to siRNA.
Together, siRNAs offer complementary advantages over traditional mutational analyses that are difficult to perform in NNR viruses, and are
also an important new tool to dissect host–virus interactive pathways.
© 2004 Elsevier B.V. All rights reserved.

Keywords: Respiratory syncytial virus; Vesicular stomatitis virus; Parainfluenza virus; RNA interference; siRNA; Host–virus interaction

1. Introduction The specificity and efficacy of siRNA-mediated PTGS

RNA interference (RNAi), mediated by short interfering
RNA molecules (siRNA), is now recognized as a major
mechanism of post-transcriptional gene silencing (PTGS)
in all metazoan eukaryotes (Hannon, 2002; Pickford and
Cogoni, 2003). The application of the siRNA technology
in mammalian cell culture received a major impetus fol-
lowing the discovery that specific double-stranded siRNA
molecules in the 21-nucleotide range effectively degrade the
cognate mRNAs, and thus abrogate the expression of the
corresponding proteins (Elbashir et al., 2001). These siR-
NAs are apparently too short to activate the cellular “inter-
feron response” that is known to cause a general inhibition
of the cellular translation machinery (Elbashir et al., 2001;
Bitko and Barik, 2001b).
∗ Tel.: +1-251-460-6860; fax: +1-251-460-6127.
E-mail address: sbarik@jaguar1.usouthal.edu (S. Barik).
against intracellular cytoplasmic parasites was originally
documented using respiratory syncytial virus (RSV) as a
test model (Bitko and Barik, 2001b). Subsequently, a flurry
of exciting studies reproduced similar success in a variety of
important animal and plant viruses (See other articles of this
volume). The NNR virus family, in particular, encompasses
a large roster of members that infect practically every terres-
trial species and cause diseases of enormous consequence
in human and animal health (Banerjee et al., 1991; Palese
et al., 1996). Two major families of the NNR viruses are:
Paramyxoviridae and Rhabodoviridae. Clinically important
paramyxoviruses include human and bovine respiratory
syncytial viruses (RSV) and parainfluenza viruses (PIV),
measles virus, mumps virus, canine distemper virus, and the
encephalitis-causing Nipah virus. The rhaboviruses include
vesicular stomatitis (VSV) and rabies viruses. This review
summarizes progress in the use of siRNA in the phenotypic
knockdown of representative NNR viral genes, technical
considerations, and the extension of the technology to
studies of host–virus interaction.

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.012
28 S. Barik / Virus Research 102 (2004) 27–35

2. Nonsegmented negative-strand RNA (NNR) viruses
2.1. An overview of the NNR viral life cycle: transcription
versus replication
Historically, the early siRNA targets were all cellular mR-
NAs, transcribed in the nucleus by DNA-dependent RNA
polymerase (Cogoni and Macino, 2000). In contrast, the
vast majority of viral RNA genomes are transcribed ex-
clusively in the cytoplasm, catalyzed by virally encoded
RNA-dependent RNA polymerases (RdRP) (Barik et al.,
1990; O’Reilly and Kao, 1998; Ahlquist, 2002). Fig. 1 sum-
marizes the generalized scheme of transcription and replica-
tion of the NNR viruses; a more comprehensive treatise can
be found in earlier reviews (Banerjee et al., 1991; Banerjee
and Barik, 1992). A scholarly minireview by Ahlquist (2002)
has recently discussed these steps with special reference to
siRNA. The minimal RdRP of NNR viruses consists of the
large viral protein L and the accessory phosphoprotein P
(Banerjee and Barik, 1992; Barik et al., 1995). Depending
on the particular virus, optimal transcription may require ad-
ditional viral proteins, cytoskeletal proteins (such as tubulin
in VSV and actin in RSV) (Moyer et al., 1986; Burke et al.,
1998), and possibly additional cellular proteins (such as pro-
filin in RSV) (Burke et al., 2000; Bitko et al., 2003). The
viral genome is a single, linear, negative-sense (anti-mRNA
sense) RNA molecule, ranging roughly from 10 to 15 kb in
size in different NNR viruses, and is intimately wrapped
with the nucleocapsid protein N. The resultant nucleoprotein
complex, termed N-RNA template, associates itself with the
RdRP subunits, primarily L and P, and is finally packaged
inside the structural shell of the virion, mainly made up of
viral glycoproteins. The nucleocapsid inside the virion is,
therefore, transcription-ready, and thus, primary viral tran-
scription begins shortly after infection. A notable physical
property of the N-RNA template is its extreme resistance to
RNases, suggesting that the tightly bound N protein offers
protection. Only this N-RNA complex serves as the biolog-
ical template for the viral RdRP, which would not recognize
pure deproteinized genomic RNA.
The RdRP functions in two operationally distinct modes:
transcription and replication (Banerjee and Barik, 1992). In
the transcription mode, it copies the negative-sense genome
to produce mRNAs corresponding to each individual gene
by a stop–restart mechanism (Barr et al., 2002). The mR-
NAs have the features of standard eukaryotic mRNAs, i.e.,
they are 5 -capped and 3 -polyadenylated (Bisaillon and
Lemay, 1997; Barik, 1993). The transcription process ex-
hibits “polarity,” such that the genes most proximal to the
promoter (3 end of the negative-strand genome) are tran-
scribed most abundantly (Barik, 1992). This is also reflected
in the relative abundance of the corresponding proteins, and
fits the function of the different proteins while allowing
transcriptional economy. For example, the N gene, usually

Fig. 1. NNR viral life cycle (see text for details). A hypothetical NNR virus genome is shown for the purposes of illustration only. In specific viruses
some of these genes may be absent and others present. For example, VSV does not contain F, and RSV contains additional genes such as SH, HPIV-3
contains HN, etc. Note that in contrast to the mRNA, the full-length genomic and antigenomic RNA (− and + sense, respectively) are neither capped
nor polyadenylated, and are wrapped with N protein and hence, are not accessible to siRNA.
 S. Barik / Virus Research 102 (2004) 27–35 29

being the closest to the promoter, produces the most mRNA

and protein, which befits the stoichiometric need of N pro-
tein to wrap the large full-length genomic and antigenomic
(see below) RNA. The L gene, coding for the large subunit
of the RdRP, is usually the farthest (i.e., at the 5 end of the
negative-strand genome), and produces the least amounts
of mRNA. As the RdRP is only needed in catalytic quan-
tities, small amounts of the L mRNA suffice the functional
need.
The NNR viral mRNAs are translated by the host’s trans-
lation machinery. Since the virion particles do not pack-
age any mRNA (or any positive-sense RNA, for that mat-
ter), viral transcription is an obligatory pre-requisite to viral
protein synthesis. Transcription and translation eventually
generate more of the RdRP, boosting transcription further
(Fig. 1). Once sufficient N protein is synthesized, it binds to
the newly transcribed nascent RNA, and this somehow pro-
vokes the leading RdRP to switch from the transcription to
the replication mode, such that the RdRP now ignores the
intergenic stop signals, producing full-length complemen-
tary positive strand (mRNA-sense) RNA, called antigenome.
Like the genome, the antigenome is also encapsidated with
the N protein. In contrast to the genome, it is not tran- Fig. 2. Inhibition of RSV gene expression by siRNA (see Bitko and
Barik, 2001b for details). A549 cells were transfected with indicated
scribed, but simply serves as a replication template to pro-
concentrations of anti-P siRNA, and infected with RSV 16 h thereafter.
duce more of the full-length negative-strand genome. At the Lane ‘U’ indicates control, uninfected cells. Top: Immunoblot (Western)
end of the intracellular life cycle of the virus, the encapsi- of total cell extracts was performed with either rabbit anti-P or mono-
dated negative-strand genomes, along with other structural clonal anti-actin antibody (Boehringer-Mannheim), as indicated. Bottom:
proteins are packaged into virion particles. Viral protein synthesis in siRNA-treated cells was measured by standard
immunoprecipitation procedures. Cells were metabolically labeled with
All siRNAs used against NNR viruses in published pa- 35 S-(methionine plus cysteine) at 18 h post-infection, followed by im-
pers so far have been double-stranded and synthetic, and de- munoprecipitation with anti-RSV antibody, and analysis of the labeled
signed according to the recommendations of Elbashir et al. proteins by SDS–PAGE and autoradiography. ‘La’ represents treatment
(2001); this issue and its exceptions are further discussed with 100 nM anti-lamin A/C siRNA. The different viral protein bands are
in detail in Section 4.1. In most experiments, the cells were so indicated.
pre-transfected with the antiviral siRNA 10–24 h before in-
fection. 5 days, which was the longest time point in the experiment.
In fact, siRNA-treated infected cells were virtually indistin-
2.2. Silencing of viral RdRP guishable from uninfected ones (Bitko and Barik, 2001b).
On a clinical note, one can envisage two classes of an-
Based on the foregoing, silencing of the minimal essen- tivirals that would inhibit infection at even earlier steps, and
tial RdRP subunits, namely L and P, should lead to a nearly both have pros and cons.
total loss of all RNA synthesis, and this is what we have
observed in all three test viruses (RSV, VSV, and HPIV-3). (i) Inhibitors of virus adsorption: Recently, a novel aro-
RSV is shown as an example in Fig. 2, in which the ablation matic compound, dubbed RFI-641, has been shown to
of P protein by siRNA led to a drastic inhibition of overall inhibit RSV infection (Huntley et al., 2002) by blocking
viral gene expression (Bitko and Barik, 2001b). It is logi- the RSV glycoprotein, F, essential for fusion of the viral
cal to assume that primary transcription, which is catalyzed envelope with the host cell membrane (see Section 2.3).
by preformed viral RdRP, would be unaffected by siRNAs Complete exclusion of the virus, however, may affect
targeting the RdRP. However, in routine infections at low natural immunity, which is still the best form of pro-
m.o.i., primary transcription produces only small amounts tection against many NNR viruses, especially RSV and
of viral mRNA, and it is the secondary transcription that HPIV-3 (Crowe and Williams, 2003).
contributes to the bulk of viral mRNA and protein synthesis. (ii) Direct inhibitors of RdRP activity: The unique sequence
Thus, abolition of new RdRP synthesis by siRNA results in a features of NNR viral L and P proteins (Barik et al.,
strong and sustained antiviral effect at an early intracellular 1990; Delarue et al., 1990; Banerjee and Barik, 1992)
step in the NNR viral life cycle while still allowing a small and the absence of similar proteins in mammalian cells
amount of viral transcription and translation to occur. In our make them ideal candidates for structure-based drug de-
experience ex vivo, the antiviral effect persisted for at least sign. However, technical difficulties such as problems of
30 S. Barik / Virus Research 102 (2004) 27–35

recombinant expression and lack of three-dimensional

structures have hindered progress in this area. Thus, the
siRNA is a potential alternative for clinical intervention
of NNR virus infection.

As elaborated in the viral life cycle (Section 2.1; Fig. 1),

NNR virus growth generates both the negative-strand
genome and the positive-strand antigenome in the infected
cell. However, it has been shown that they are resistant to
siRNA (Bitko and Barik, 2001b). The most likely expla-
nation is that their obligatory encapsidation by N protein
protects them from annealing to the siRNA or to each other
(Fig. 1), just as it protects them from RNase action in vitro.
In any case, it is clear that the NNR viral mRNAs, and not
the genomes, are viable targets for RNA interference.

2.3. Silencing of viral accessory genes

There are a number of NNR viral proteins that are not

involved in viral RNA or protein synthesis, but take part
in important pre- or post-transcriptional aspects of the vi-
ral life cycle. For example, paramyxoviruses initiate infec-
tion by attaching to cell surface receptors and fusing vi-
ral and cell membranes. Viral attachment proteins, such
as hemagglutinin-neuraminidase (HN) or glycoprotein (G)
bind receptors, whereas fusion (F) proteins direct membrane
fusion (Dutch et al., 2000). The fusion protein, F, is gen-
erally essential for cell fusion (formation of syncytia), al-
though the relative importance of the attachment proteins
in fusion is under active study. Recombinant expression
has sometimes produced variable results. For instance, in
transient transfection experiments co-expression of all three
RSV glycoproteins—F, G, and SH—was required for opti-
mal fusion (Heminway et al., 1994a,b), whereas recombi-
nant RSV lacking SH formed fused cells as efficiently as
the wild type (Bukreyev et al., 1997). It is possible that the
exact outcome is dictated by the relative amounts of the re-
combinant proteins and/or the expression system used. The
siRNA approach, in contrast, obviates recombinant expres-
sion, and allows phenotypic analysis in a virus-infected cell
by removing one protein while leaving all others intact. This
has been illustrated by individually abrogating the F or the
SH protein in RSV-infected cells by siRNA. Knockdown of
F resulted in complete loss of syncytium (Bitko and Barik,
2001b), without affecting G or SH synthesis (unpublished
results). Reciprocally, abrogation of SH by siRNA had no
effect on F or G, and also had no appreciable effect on syn-
cytium (unpublished results). Together, these results validate
an essential role of F and the dispensability of SH in syn-
cytium formation by RSV.
In another example, a number of studies using reverse ge-
netics and variant virus strains have shown that both the fu-
sion (F) protein and hemagglutinin-neuraminidase (HN) are
essential for syncytium in HPIV-3 infected cells (Heminway
et al., 1994a,b; Dutch et al., 2000; Porotto et al., 2003). Cu-
riously, RSV encodes neither a hemagglutinin nor a neu-
Fig. 3. Inhibition of HPIV-3 syncytium formation by siRNA. A549 cell
monolayers were transfected with 50 nM of the following double-stranded
siRNA against the F or HN sequence of HPIV-3 (GenBank accession
numbers X05303 and Z26523, respectively): F: CCAUGAAAUGGA-
GAGCUGU; HN: CUCAGACUUGGUACCUGAC. Only the N19 portion
(Elbashir et al., 2001) of the sense strand of the siRNA is shown. Trans-
fection and subsequent virus infection was carried out as in Fig. 2, and
phase-contrast pictures were taken as described (Bitko and Barik, 2001b;
Bitko et al., 2003). Top: Note the large fused mass of cells (syncytium) at
the center of the panel labeled “HPIV-3,” and prevention of such fusion
by ds-siRNA (100 nM) targeting F and HN. Bottom: Immunoblot shows
loss of HN upon siRNA-treatment.
raminidase activity, but is obviously capable of forming syn-
cytia. Thus, we retested the role of the HN protein in syn-
cytium formation in wild type HPIV-3-infected cells using
siRNA. As shown in Fig. 3, abrogation of either F or HN by
the respective siRNA resulted in unfused cells, confirming
the need for both proteins for syncytium formation.
Although the vast majority of NNR viral mRNAs pro-
duce a single polypeptide, there are a few examples of mul-
tiple initiations. In a recent study, VSV matrix gene (M)
mRNA was shown to produce three proteins M1, M2, and
M3 (Jayakar and Whitt, 2002). M1 was the longest polypep-
tide initiating at the most 5 -proximal ATG; M2 and M3
were shorter, and produced by initiation at internal ATG’s in
the same reading frame. While the exact roles of these dif-
ferent proteins remain unclear, the M protein(s) have been
shown to inhibit host gene expression (Jayakar and Whitt,
2002; Kopecky and Lyles, 2003). Recent studies have sug-
gested that the M gene is also important for apoptosis of the
host cell, manifested by the characteristic rounding and de-
tachment of VSV-infected cells (Jayakar and Whitt, 2002;
Kopecky and Lyles, 2003). In an attempt to determine which
product of the M gene mRNA is involved in apoptosis and
 S. Barik / Virus Research 102 (2004) 27–35
Fig. 4. Inhibition of VSV cytopathic effect by ds-siRNA. Monolayers of
HEp-2 cells were transfected with double-stranded siRNA (80 nM) corre-
sponding to the following sequence of the VSV matrix (M) mRNA (Indi-
ana serotype; GenBank accession number X04452): AGGUAAGAAAU-
CUAAGAAA. Subsequent infection with VSV, phase-contrast microgra-
phy, and immunoblot were done as described in Fig. 3. The antibody
was made against a C-terminal peptide common to all three M proteins:
KFSDFREKALMFGLIVEEE (Jayakar and Whitt, 2002). Top: Note that
siRNA-treated cells look as if uninfected. Bottom: Immunoblot shows spe-
cific loss of the three M proteins, but not P, upon treatment with anti-M
siRNA.
cell-rounding, we targeted the region of the mRNA that is
unique to M1 and upstream of M2 and M3. Interestingly,
however, the siRNA abrogated the expression of all three
proteins (Fig. 4). Cell detachment was concomitantly inhib-
ited, demonstrating the role of the M gene in the process,
although we could not identify the exact protein involved.
The reason to explain the loss of all three proteins must
await further studies, but the most likely possibilities are:
(a) loss of one part of the mRNA made the rest susceptible
to degradation by RNases, and (b) loss of the 5 -cap and
lack of internal ribosome entry inhibited expression of the
downstream M2 and M3 ORFs.
2.4. Future directions
A variety of creative experiments can be designed by tar-
geting NNR viral genes with siRNA. A few suggestions are
offered here.
31
(a) Virus-resistant cells: It is possible to generate, by con-
stitutive or inducible expression of a transgene siRNA,
a cell line that would resist virus infection. Specialized
vectors have been constructed that drive siRNA expres-
sion from RNase III-based tRNA or snRNA (mainly
U6) gene promoter (Yu et al., 2002; Kawasaki and
Taira, 2003), and are marketed by commercial vendors.
This would establish a form of “intracellular immunity”
(Carmichael, 2002) that can be extended to produce
transgenic virus-resistant animals.
(b) Complementation experiments: As a rule, siRNAs lead
to the loss of the protein rather than its mutational al-
teration, and thus, cannot be used for structure–function
analysis. However, the phenotypic viral mutant can be
complemented by transient transfection of the corre-
sponding cDNA clone. The latter can be subjected to
site-directed mutagenesis, and the mutants that fail to
complement would define the amino acid residues im-
portant for function.
3. Cellular genes important in NNR virus–host
interaction
3.1. Profilin and RSV: a case in point
Silencing of cellular genes by siRNA can be used to ad-
dress the role of such genes in host–virus interaction. We
have recently applied this approach to determine the role
of profilin in the RSV life cycle (Bitko et al., 2003). Previ-
ous in vitro studies had shown that profilin stimulates RSV
transcription, but is not absolutely necessary for virus repli-
cation (Burke et al., 2000). Knockdown of profilin showed
multiple effects: (a) it had a minor effect on viral transcrip-
tion (and hence translation), supporting in vitro results; (b)
it strongly inhibited virion morphogenesis such that progeny
virions were not produced; (c) it inhibited formation of actin
stress fibers in cells where such fibers form in response to
RSV infection (e.g., HEp-2 and L4 cells); and (d) it abro-
gated fusion in all cells tested. These results establish an
essential role of profilin in the post-gene expression steps of
RSV growth, and in the regulation of the actin cytoskeleton
and the cell membrane in the infected cell.
3.2. Future directions
The ability to quickly and reversibly knock down inter-
active cellular and viral genes by siRNA will allow single
and combinatorial analysis of various interacting host–virus
pathways that would have been otherwise impossible. Two
examples are considered here.
(a) Genes dispensable for cellular viability but essential
for NNR virus: Although the loss of profilin was toler-
ated in cultured cells (Bitko et al., 2003), profilin gene
knockout led to embryonic lethality, suggesting its es-
sential role in early development. In a seminal study
32 S. Barik / Virus Research 102 (2004) 27–35
(Harborth et al., 2001) selected nuclear and membrane
proteins were targeted by siRNA to test their essential
need in a number of cell lines in culture. Two cytoskele-
ton/membrane regulatory proteins that were found to be
nonessential are Ena/VASP and zyxin. The validity of
this approach is borne out by the fact that Ena/VASP
and zyxin knockout mice are viable (Aszódi et al., 1999;
Hoffman et al., 2003). Thus, it would be exciting to de-
termine if such (relatively) nonessential cellular genes
are important in NNR viral life cycles. If they are, it
would not only enrich our fundamental knowledge of
host–virus interaction, but may lead to antivirals that
would specifically block such interactions without gross
toxicity to the host.
(b) Specific viral triggers of cellular pathways: NNR viruses
as a class activate a variety of cellular genes and sig-
naling pathways. These include cellular transcription
factors such as NF-␬B (Bitko and Barik, 1998; Helin
et al., 2001; Tian et al., 2002), protein kinases (Monick
et al., 2001; Pazdrak et al., 2002), and apoptosis (Bitko
and Barik, 2001a; Esolen et al., 1995; O’Donnell et al.,
1999; Cristina et al., 2001). The specific viral proteins
(single or multiple) that trigger these pathways can be
identified by siRNA knockdown.
4. The technique
In this section, we first describe the basic procedure
that we routinely use. To design siRNA, we generally fol-
low the original recommendations (Elbashir et al., 2001).
Interested readers can find a detailed description and al-
ternate siRNA designs in the Tuschl lab web page (http://
www.rockefeller.edu/labheads/tuschl/sirna.html). Briefly,
we look for a AA(N)19 TT sequence in the target mRNA
that is not too close to the ATG. This RNA and its
complementary strand is chemically synthesized and pur-
chased from commercial vendors (such as Dharmacon;
http://www.dharmacon.com). For transfection, various
amounts of the deprotected, desalted, double-stranded
siRNA (ds-siRNA) are transfected using OligofectAMINE
Reagent (Life Technologies) in OPTIMEM I (Life Tech-
nologies) (Bitko and Barik, 2001b). Care is taken to make
a homogeneous mixture of the transfection reagent and
RNA by heavy vortexing before addition to the culture. We
have recently switched to a newer reagent, TransIT-TKO® ,
marketed by Mirus (http://www.genetransfer.com), which
does not require the use of serum-free media. In our expe-
rience, optimization of the DNA to TKO ratio is advised
as excessive TKO may cause injury to some cell lines. The
virus is generally added 6–8 h after siRNA transfection if a
viral mRNA is targeted and 14–18 h after transfection if a
cellular mRNA is targeted.
In what follows, we describe some recent variations and
potential considerations in planning a siRNA experiment,
particularly in NNR viruses.
Fig. 5. Relative efficacy of single-stranded and double-stranded siRNAs.
The sequence of the siRNA, designed against RSV P, has been pub-
lished earlier (Bitko and Barik, 2001b). The symbols represent the fol-
lowing: (䊊), ds-siRNA; (䊐) ss-antisense siRNA; () same ss-siRNA but
5 -phosphorylated (Martinez et al., 2002). Transfection of siRNA and in-
fection with RSV were carried out as in Fig. 2. Residual P protein was
estimated by immunoblot, and plotted as percent of the untransfected
control against the negative logarithm of siRNA concentration in molarity
(M). Thus, 9 on the x-axis means 1 nM siRNA, 8 means 10 nM, and so
forth. The dotted line marks 50% loss.
4.1. Double-stranded versus single-stranded siRNA
Although the double-stranded siRNA is used with success
in cell culture, it is the antisense RNA strand that binds to
the target mRNA and forms the silencing complex, RISC. A
few recent studies have indeed documented the effectiveness
of single-stranded antisense RNA (ss-siRNA) (Tijsterman
et al., 2002; Martinez et al., 2002). In our hands, ss-siRNA
showed decent activity against NNR viruses, although it was
required in larger quantities to achieve a quantitatively sim-
ilar effect (Fig. 5). Using RSV P mRNA as the experimen-
tal target, the IC50 (amount of RNA causing 50% reduc-
tion of P protein) of ds-siRNA and ss-siRNA were about 10
and 75 nM, respectively. When the 5 end of the antisense
ss-siRNA was phosphorylated, the IC50 slightly improved to
50 nM (Fig. 5). As suggested earlier (Martinez et al., 2002),
we presume that this reflects the relative stability of the RNA
species in the order: dsRNA > 5 -phospho-ssRNA > ss-RNA.
4.2. Mismatched siRNA and position effect: pros and cons
As a rule, the siRNA is extremely stringent in its speci-
ficity such that a single nucleotide mismatch may abrogate
its function (Elbashir et al., 2001; Bitko and Barik, 2001b;
Chi et al., 2003; Semizarov et al., 2003). A few recent re-
ports have challenged this stringency (Hamada et al., 2002;
Jackson et al., 2003). In an extreme example, direct silenc-
ing of non-targeted genes containing as few as 11 contiguous
nucleotides of identity to the siRNA was observed (Jackson
et al., 2003). In our experience with NNR viruses, the toler-
ance to mismatch varies from one position to another on the
same NNR viral mRNA. In other words, a single nucleotide
mismatch at one site of the mRNA may prevent siRNA ac-
 S. Barik / Virus Research 102 (2004) 27–35
tion (Bitko and Barik, 2001b), whereas at another site on
the same mRNA, one mismatch may be largely tolerated
(unpublished results). The reason for the sequence effect is
currently unknown, but mismatch tolerance has raised is-
sues of specificity. However, this is probably not a major
concern with NNR viruses, since the viral genes have lit-
tle or no homology with one another or with cellular genes.
On the contrary, expansion of the target repertoire of the
siRNA may enable it to silence multiple mutational variants
of a given NNR virus that commonly occur due to the high
mutation rate of the RNA genome (Domingo and Holland,
1997). In any case, when designing a siRNA, a BLAST
search against the non-redundant nucleotide databases of the
GenBank should be routinely done to verify specificity.
Recent studies have also shown that the siRNA effect
can be strikingly position-dependent, i.e., siRNAs directed
against different regions of the same mRNA may exhibit
a wide range of silencing efficiency (Holen et al., 2002).
Again, the rules governing the position effect remain un-
known, but we have also observed this in silencing the NNR
virus.
Based on these considerations, it may be wise to design
multiple siRNAs against a target mRNA using the prescribed
design guidelines, test all of them over a range of concen-
trations, and then select the most effective one.
4.3. Abundant mRNAs and saturation of the silencing
machinery
In principle, multiple siRNAs can be transfected or ex-
pressed in the same cell to abrogate the respective NNR viral
proteins simultaneously. In our experience, this works, but
seems to have limits (unpublished result). Silencing of an
abundant NNR viral mRNA (for example, that of a promoter
proximal gene such as N) often leads to reduced silencing
of a promoter distal gene (such as L). The mechanism of
this is being investigated. Nonetheless, quantitative analy-
sis has shown that the siRNA effect may reach a plateau
with increasing siRNA concentration (Kamath et al., 2001;
Bitko and Barik, 2001b), suggesting that the RNAi-induced
silencing complexes (RISC) of the cell are saturable. Thus,
an abundant viral mRNA may sequester most of the cellular
RISC, leaving little for the less abundant ones. Conceivably,
such competition can also occur between cellular and vi-
ral mRNAs, thus potentially limiting host–virus interaction
studies.
5. Conclusions
Traditionally, structure–function analyses of RNA
genomes, including those of NNR viruses, have made use
of fortuitous mutants, natural variants, or chemically mu-
tagenized stocks (e.g., Crowe et al., 1994). Such mutations
are unpredictable and must be mapped by difficult com-
plementation analyses or sequencing. Because site-directed
33
mutagenesis requires a DNA template, direct mutational
analysis of selected NNR viral genes is also not possible.
These obstacles have been partially circumvented by the use
of cloned viral cDNA (Whelan et al., 1995; Lawson et al.,
1995; Pekosz et al., 1999; Marriott and Easton, 1999). De-
spite its revolutionary effect on NNR viral reverse genetics,
however, the cDNA-based strategy is not without limita-
tions. First, expression of recombinant viral genomic RNA
and all the viral proteins in the correct ratio that reflects
polarity remains a grueling task. As mentioned before, the
10–15 kb long RNA genome of the NNR viruses must have
specific sequences at the 5 and 3 termini for transcription
and replication and must be properly encapsidated by the
nucleocapsid protein (N) in order to be recognized by the
viral RdRP. Second, many NNR viral genomes and proteins
are currently expressed from vaccinia-based cDNA clones,
generally requiring super-infection by vaccinia virus, and
vaccinia itself modulates multiple cellular entities, includ-
ing MAP kinases and the actin cytoskeleton (de Magalhaes
et al., 2001). It is, therefore, virtually impossible to study
the interaction between cellular signaling pathways and
NNR viruses in cells that are super-infected by vaccinia.
Third, mutations in the recombinant DNA are permanent,
i.e., the mutational phenotype cannot be regulated at differ-
ent times in infection. For example, if a viral gene prod-
uct has dual essential functions—one early, and one late
in infection—the late function will remain undiscovered,
since the mutant virus will never proceed beyond the early
stage.
In conclusion, properly designed siRNA can be a quicker
and simpler alternative to cDNA-based reverse genetics of
NNR viruses. Its ability to simultaneously knock down viral
and cellular gene products offers particular advantages in
the dissection of host–virus signaling pathways. We predict
that the siRNA strategy will lead to major advances in these
areas in the immediate future.
Acknowledgements
The siRNA-related research in the author’s laboratory was
supported in part by NIH grants AI049682 from the National
Institute of Allergy and Infectious Diseases and EY013826
from the National Eye Institute. A computer program to
find the AA(N)19 TT sequences was kindly written and pro-
vided by Titus Barik (http://www.barikautomation.com).
Apology is offered to many whose original research had
advanced the field, but could not be cited due to space con-
straints. Nevertheless, the majority of the research articles
are cross-referenced in the recent reviews cited here.
References
Ahlquist, P., 2002. RNA-dependent RNA polymerases, viruses, and RNA
silencing. Science 296, 1270–1273.
34 S. Barik / Virus Research 102 (2004) 27–35
Aszódi, A., Pfeifer, A., Ahmad, M., Glauner, M., Zhou, X.H., Ny, L.,
Andersson, K.E., Kehrel, B., Offermanns, S., Fässler, R., 1999. The
vasodilator-stimulated phosphoprotein (VASP) is involved in cGMP-
and cAMP-mediated inhibition of agonist-induced platelet aggregation,
but is dispensable for smooth muscle function. EMBO J. 18, 37–48.
Banerjee, A.K., Barik, S., 1992. Gene expression of vesicular stomatitis
virus genome RNA. Virology 188, 417–428.
Banerjee, A.K., Barik, S., De, B.P., 1991. Gene expression of nonseg-
mented negative-strand RNA viruses. Pharmacol. Ther. 51, 47–70.
Barik, S., 1992. Transcription of human respiratory syncytial virus genome
RNA in vitro: requirement of cellular factor(s). J. Virol. 66, 6813–
6818.
Barik, S., 1993. The structure of the 5 terminal cap of the respiratory
syncytial virus mRNA. J. Gen. Virol. 74, 485–490.
Barik, S., McLean, T., Dupuy, L.C., 1995. Phosphorylation of Ser232
directly regulates the transcriptional activity of the P protein of human
respiratory syncytial virus: phosphorylation of Ser237 may play an
accessory role. Virology 213, 405–412.
Barik, S., Rud, E.W., Luk, D., Banerjee, A.K., Kang, C.Y., 1990. Nu-
cleotide sequence analysis of the L gene of vesicular stomatitis virus
(New Jersey serotype): identification of conserved domains in L pro-
teins of nonsegmented negative-strand RNA viruses. Virology 175,
332–337.
Barr, J.N., Whelan, S.P., Wertz, G.W., 2002. Transcriptional control of
the RNA-dependent RNA polymerase of vesicular stomatitis virus.
Biochim. Biophys. Acta 1577, 337–353.
Bisaillon, M., Lemay, G., 1997. Viral and cellular enzymes involved in
synthesis of mRNA cap structure. Virology 236, 1–7.
Bitko, V., Barik, S., 1998. Persistent activation of RelA by respiratory
syncytial virus involves protein kinase C, underphosphorylated Ikap-
paBbeta, and sequestration of protein phosphatase 2A by the viral
phosphoprotein. J. Virol. 72, 5610–5618.
Bitko, V., Barik, S., 2001a. An endoplasmic reticulum-specific
stress-activated caspase (caspase-12) is implicated in the apoptosis of
A549 epithelial cells by respiratory syncytial virus. J. Cell. Biochem.
80, 441–454.
Bitko, V., Barik, S., 2001b. Phenotypic silencing of cytoplasmic genes
using sequence-specific double-stranded short interfering RNA and its
application in the reverse genetics of wild type negative-strand RNA
viruses. BMC Microbiol. 1, 34.
Bitko, V., Oldenburg, A., Garmon, N.E., Barik, S., 2003. Profilin is re-
quired for viral morphogenesis, syncytium formation, and cell-specific
stress fiber induction by respiratory syncytial virus. BMC Microbiol.
3, 9.
Bukreyev, A., Whitehead, S.S., Murphy, B.R., Collins, P.L., 1997. Recom-
binant respiratory syncytial virus from which the entire SH gene has
been deleted grows efficiently in cell culture and exhibits site-specific
attenuation in the respiratory tract of the mouse. J. Virol. 71, 8973–
8982.
Burke, E., Dupuy, L., Wall, C., Barik, S., 1998. Role of cellular actin in
the gene expression and morphogenesis of human respiratory syncytial
virus. Virology 252, 137–148.
Burke, E., Mahoney, N.M., Almo, S.C., Barik, S., 2000. Profilin is required
for optimal actin-dependent transcription of respiratory syncytial virus
genome RNA. J. Virol. 74, 669–675.
Carmichael, G.G., 2002. Medicine: silencing viruses with RNA. Nature
418, 379–380.
Chi, J.T., Chang, H.Y., Wang, N.N., Chang, D.S., Dunphy, N., Brown,
P.O., 2003. Genomewide view of gene silencing by small interfering
RNAs. Proc. Natl. Acad. Sci. U.S.A. 100, 6343–6346.
Cogoni, C., Macino, G., 2000. Post-transcriptional gene silencing across
kingdoms. Curr. Opin. Genet. Dev. 10, 638–643.
Cristina, J., Yunus, A.S., Rockemann, D.D., Samal, S.K., 2001. Bovine
respiratory syncytial virus can induce apoptosis in MDBK cultured
cells. Vet. Microbiol. 83, 317–320.
Crowe Jr., J.E., Bui, P.T., London, W.T., Davis, A.R., Hung, P.P., Chanock,
R.M., Murphy, B.R., 1994. Satisfactorily attenuated and protective
mutants derived from a partially attenuated cold-passaged respiratory
syncytial virus mutant by introduction of additional attenuating mu-
tations during chemical mutagenesis. Vaccine 12, 691–699.
Crowe Jr., J.E., Williams, J.V., 2003. Immunology of viral respiratory
tract infection in infancy. Paediatr. Respir. Rev. 4, 112–119.
Delarue, M., Poch, O., Tordo, N., Moras, D., Argos, P., 1990. An attempt
to unify the structure of polymerases. Protein Eng. 3, 461–467.
de Magalhaes, J.C., Andrade, A.A., Silva, P.N., Sousa, L.P., Ropert, C.,
Ferreira, P.C., Kroon, E.G., Gazzinelli, R.T., Bonjardim, C.A., 2001. A
mitogenic signal triggered at an early stage of vaccinia virus infection:
implication of MEK/ERK and protein kinase A in virus multiplication.
J. Biol. Chem. 276, 38353–38360.
Domingo, E., Holland, J.J., 1997. RNA virus mutations and fitness for
survival. Annu. Rev. Microbiol. 51, 151–178.
Dutch, R.E., Jardetzky, T.S., Lamb, R.A., 2000. Virus membrane fusion
proteins: biological machines that undergo a metamorphosis. Biosci.
Rep. 20, 597–612.
Elbashir, S.M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., Tuschl,
T., 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference
in cultured mammalian cells. Nature 411, 494–498.
Esolen, L.M., Park, S.W., Hardwick, J.M., Griffin, D.E., 1995. Apoptosis
as a cause of death in measles virus-infected cells. J. Virol. 69, 3955–
3958.
Hamada, M., Ohtsuka, T., Kawaida, R., Koizumi, M., Morita, K., Fu-
rukawa, H., Imanishi, T., Miyagishi, M., Taira, K., 2002. Effects on
RNA interference in gene expression (RNAi) in cultured mammalian
cells of mismatches and the introduction of chemical modifications at
the 3 -ends of siRNAs. Antisense Nucleic Acid Drug Dev. 12, 301–
309.
Hannon, G.J., 2002. RNA interference. Nature 418, 244–251.
Harborth, J., Elbashir, S.M., Bechert, K., Tuschl, T., Weber, K., 2001.
Identification of essential genes in cultured mammalian cells using
small interfering RNAs. J. Cell. Sci. 114, 4557–4565.
Helin, E., Vainionpaa, R., Hyypia, T., Julkunen, I., Matikainen, S., 2001.
Measles virus activates NF-kappa B and STAT transcription factors and
production of IFN-alpha/beta and IL-6 in the human lung epithelial
cell line A549. Virology 290, 1–10.
Heminway, B.R., Yu, Y., Galinski, M.S., 1994a. Paramyxovirus me-
diated cell fusion requires co-expression of both the fusion and
hemagglutinin-neuraminidase glycoproteins. Virus Res. 31, 1–16.
Heminway, B.R., Yu, Y., Tanaka, Y., Perrine, K.G., Gustafson, E., Bern-
stein, J.M., Galinski, M.S., 1994b. Analysis of respiratory syncytial
virus F, G, and SH proteins in cell fusion. Virology 200, 801–805.
Hoffman, L.M., Nix, D.A., Benson, B., Boot-Hanford, R., Gustafsson,
E., Jamora, C., Menzies, A.S., Goh, K.L., Jensen, C.C., Gertler, F.B.,
Fuchs, E., Fassler, R., Beckerle, M.C., 2003. Targeted disruption of
the murine zyxin gene. Mol. Cell. Biol. 23, 70–79.
Holen, T., Amarzguioui, M., Wiiger, M.T., Babaie, E., Prydz, H., 2002.
Positional effects of short interfering RNAs targeting the human co-
agulation trigger Tissue Factor. Nucleic Acids Res. 30, 1757–1766.
Huntley, C.C., Weiss, W.J., Gazumyan, A., Buklan, A., Feld, B., Hu,
W., Jones, T.R., Murphy, T., Nikitenko, A.A., O’Hara, B., Prince,
G., Quartuccio, S., Raifeld, Y.E., Wyde, P., O’Connell, J.F., 2002.
RFI-641, a potent respiratory syncytial virus inhibitor. Antimicrob.
Agents Chemother. 46, 841–847.
Jackson, A.L., Bartz, S.R., Schelter, J., Kobayashi, S.V., Burchard, J., Mao,
M., Li, B., Cavet, G., Linsley, P.S., 2003. Expression profiling reveals
off-target gene regulation by RNAi. Nat. Biotechnol. 21, 635–637.
Jayakar, H.R., Whitt, M.A., 2002. Identification of two additional trans-
lation products from the matrix (M) gene that contribute to vesicular
stomatitis virus cytopathology. J. Virol. 76, 8011–8018.
Kamath, R.S., Martinez-Campos, M., Zipperlen, P., Fraser, A.G., Ahringer,
J., 2001. Effectiveness of specific RNA-mediated interference through
ingested double-stranded RNA in Caenorhabditis elegans. Genome
Biol. 2, RESEARCH0002.
Kawasaki, H., Taira, K., 2003. Short hairpin type of dsRNAs that are
controlled by tRNA(Val) promoter significantly induce RNAi-mediated
 S. Barik / Virus Research 102 (2004) 27–35
gene silencing in the cytoplasm of human cells. Nucleic Acids Res.
31, 700–707.
Kopecky, S.A., Lyles, D.S., 2003. The cell-rounding activity of the vesic-
ular stomatitis virus matrix protein is due to the induction of cell
death. J. Virol. 77, 5524–5528.
Lawson, N., Stillman, E., Whitt, M., Rose, J., 1995. Recombinant vesicular
stomatitis viruses from DNA. Proc. Natl. Acad. Sci. U.S.A. 92, 4477–
4481.
Marriott, A.C., Easton, A.J., 1999. Reverse genetics of the Paramyxoviri-
dae. Adv. Virus Res. 53, 321–340.
Martinez, J., Patkaniowska, A., Urlaub, H., Luhrmann, R., Tuschl, T.,
2002. Single-stranded antisense siRNAs guide target RNA cleavage
in RNAi. Cell 110, 563–574.
Monick, M., Staber, J., Thomas, K., Hunninghake, G., 2001. Respiratory
syncytial virus infection results in activation of multiple protein kinase
C isoforms leading to activation of mitogen-activated protein kinase.
J. Immunol. 166, 2681–2687.
Moyer, S.A., Baker, S.C., Lessard, J.L., 1986. Tubulin: a factor necessary
for the synthesis of both Sendai virus and vesicular stomatitis virus
RNAs. Proc. Natl. Acad. Sci. U.S.A. 83, 5405–5409.
O’Donnell, D.R., Milligan, L., Stark, J.M., 1999. Induction of CD95
(Fas) and apoptosis in respiratory epithelial cell cultures following
respiratory syncytial virus infection. Virology 257, 198–207.
O’Reilly, E.K., Kao, C.C., 1998. Analysis of RNA-dependent RNA poly-
merase structure and function as guided by known polymerase struc-
tures and computer predictions of secondary structure. Virology 252,
287–303.
Palese, P., Zheng, H., Engelhardt, O.G., Pleschka, S., Garcia-Sastre, A.,
1996. Negative-strand RNA viruses: genetic engineering and applica-
tions. Proc. Natl. Acad. Sci. U.S.A. 93, 11354–11358.
35
Pazdrak, K., Olszewska-Pazdrak, B., Liu, T., Takizawa, R., Brasier, A.R.,
Garofalo, R.P., Casola, A., 2002. MAPK activation is involved in post-
transcriptional regulation of RSV-induced RANTES gene expression.
Am. J. Physiol. Lung Cell Mol. Physiol. 283, L364–L372.
Pekosz, A., He, B., Lamb, R.A., 1999. Reverse genetics of negative-strand
RNA viruses: closing the circle. Proc. Natl. Acad. Sci. U.S.A. 96,
8804–8806.
Pickford, A.S., Cogoni, C., 2003. RNA-mediated gene silencing. Cell.
Mol. Life Sci. 60, 871–882.
Porotto, M., Murrell, M., Greengard, O., Moscona, A., 2003. Trig-
gering of human parainfluenza virus 3 fusion protein (F) by the
hemagglutinin-neuraminidase (HN) protein: an HN mutation dimin-
ishes the rate of F activation and fusion. J. Virol. 77, 3647–3654.
Semizarov, D., Frost, L., Sarthy, A., Kroeger, P., Halbert, D.N., Fesik,
S.W., 2003. Specificity of short interfering RNA determined through
gene expression signatures. Proc. Natl. Acad. Sci. U.S.A. 100, 6347–
6352.
Tian, B., Zhang, Y., Luxon, B.A., Garofalo, R.P., Casola, A., Sinha,
M., Brasier, A.R., 2002. Identification of NF-kappaB-dependent gene
networks in respiratory syncytial virus-infected cells. J. Virol. 76,
6800–6814.
Tijsterman, M., Ketting, R.F., Okihara, K.L., Sijen, T., Plasterk, R.H.,
2002. RNA helicase MUT-14-dependent gene silencing triggered in
C. elegans by short antisense RNAs. Science 295, 694–697.
Whelan, S.P., Ball, L.A., Barr, J.N., Wertz, G.T., 1995. Efficient recovery
of infectious vesicular stomatitis virus entirely from cDNA clones.
Proc. Natl. Acad. Sci. U.S.A. 92, 8388–8392.
Yu, J.Y., DeRuiter, S.L., Turner, D.L., 2002. RNA interference by ex-
pression of short-interfering RNAs and hairpin RNAs in mammalian
cells. Proc. Natl. Acad. Sci. U.S.A. 99, 6047–6052.
 Virus Research 102 (2004) 37–42

Use of siRNAs to prevent and treat influenza virus infection

Qing Ge, Herman N. Eisen, Jianzhu Chen∗
Department of Biology, Center for Cancer Research, Massachusetts Institute of Technology, Cambridge, MA 02139, USA

Abstract

Influenza virus causes one of the most prevalent infections in humans. In a typical year, 10–20% of the population in the United States
are infected by influenza virus, resulting in up to 40,000 deaths. Current vaccines can prevent illness in approximately 70–80% of healthy
individuals under age 65, but the protection rate is much lower in those most susceptible to infection, namely infants, the elderly, and
immunocompromised individuals. Although four antiviral drugs have been approved in the United States for treatment and/or prophylaxis of
influenza, their use is limited because of concerns about side effects, compliance, and the possible emergence of resistant virus. We found
that short interfering RNAs (siRNAs) specific for conserved regions of the influenza virus genome are potent inhibitors of influenza virus
replication in both cell lines and embryonated chicken eggs. In this review, we discuss the potential value of siRNAs for preventing and
treating influenza virus infections in humans and the challenges that have to be overcome to realize their potential.
© 2004 Elsevier B.V. All rights reserved.

Keywords: Influenza virus; siRNA; Prophylaxis; Therapy; Delivery

1. Influenza virus infection is a significant
public health problem
Influenza viruses are enveloped, single-stranded RNA
viruses of the Orthomyxoviridae family (Lamb and Krug,
2001). They are classified as influenza virus types A, B, and
C based on differences in their nucleoproteins and matrix
proteins. Among the three types, influenza A virus is most
pathogenic. Influenza A virus contains eight RNA seg-
ments of negative polarity in its genome (Lamb and Krug,
2001). Three of the segments encode the three compo-
nents of RNA-dependent RNA transcriptase complex (PA,
PB1, and PB2). Three additional RNA segments code for
(i) hemagglutinin (HA), a major surface glycoprotein, (ii)
neuraminidase (NA), another major surface glycoprotein,
and (iii) nucleocapsid protein (NP), a scaffolding and RNA
binding protein. Each of the remaining two RNA segments
encodes two proteins, either M1 and M2 or NS1 and NS2,
which function either as viral structural proteins or as non-
structural proteins in the viral life cycle. The RNA segments
present in the fully assembled virus particle are referred
to as virion RNA (vRNA). Transcription of vRNA by the
virus-encoded transcriptase produces mRNA that serves as
∗ Corresponding author. Tel.: +1-617-258-6173;
fax: +1-617-258-6172.
E-mail address: jchen@mit.edu (J. Chen).
a template for synthesis of viral proteins. Transcription of
vRNA also produces complementary RNA (cRNA), which
differs from mRNA by lacking the 5 cap and the 3 poly A
tail and serves as a template for synthesizing more vRNA
for new virion production.
In humans, influenza A virus infection occurs in the upper
respiratory tract and in the lungs (Lamb and Krug, 2001).
Its threat as a pathogen depends on some of the virus’s
unique properties. First, the virus is easily spread by aerosol.
Second, small changes (antigenic drift) in the virus’ prin-
cipal antigens HA and NA are frequent, enabling the virus
to readily escape protective immunity induced by previous
exposure to a different variant of the virus (Webster et al.,
1992). Third, new strains of virus, markedly different anti-
genically, can be generated by reassortment (antigenic shift)
(Webster et al., 1982). When influenza viruses from two
species, such as human and swine, infect the same cell, hy-
brid (reassortant) viruses containing RNA segments from
the two species can emerge (Webby and Webster, 2001).
The hybrid virus, with the resulting major antigenic shift,
can be especially virulent, because it is not restrained by
host immunity to previously prevalent strains of the virus.
The 1918 pandemic, during which over 20 million people
were estimated to have died worldwide was thought to be
caused by a hybrid virus arising from reassortment of swine
and human viruses (Patterson and Pyle, 1991; Taubenberger
et al., 2001). The potential for the emergence of a highly

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.013
38 Q. Ge et al. / Virus Research 102 (2004) 37–42
virulent strain of influenza virus, either naturally occurring
or deliberately constructed, was clearly demonstrated by the
sudden appearance of a lethal form of the virus in Hong
Kong in 1997, which killed 6 out of 18 infected individuals
(Claas et al., 1998; Yuen et al., 1998). Even infection with
comparatively less virulent influenza strains is estimated
to cause up to 40,000 deaths in the US in a typical year
(Thompson et al., 2003). The mortality due to influenza virus
infection generally is found in infants, adults over 65 years
of age, and immunocompromised individuals (Kemble and
Greenberg, 2003; Thompson et al., 2003). As the population
ages in industrialized countries, influenza virus infection and
its associated complications and mortality will likely be-
come an even larger public health problem. Currently, the
economic costs of influenza virus infection are estimated to
be US$ 5 billion per annum in the US alone.
2. More effective influenza therapeutics are needed
The only effective protection from influenza virus infec-
tion now available is vaccination (Palese and Garcia-Sastre,
2002). In the US, inactivated trivalent influenza virus
is the commonly used form of the vaccine. Recently, a
cold-adapted influenza virus vaccine was approved and is
expected to be available in the coming flu season. The
cold-adapted virus is capable of undergoing limited repli-
cation in the upper respiratory tract, where the tempera-
ture is lower than 37 ◦ C, but not in the lung, where the
temperature is higher (Maassab and Bryant, 1999). Thus,
the cold-adapted virus is administered intranasally as a
live-attenuated virus vaccine. Although this is expected to
induce local and perhaps longer-lasting protective immune
responses (Palese and Garcia-Sastre, 2002), its efficacy at
the population level has yet to be determined. Currently,
influenza vaccines are recommended for all individuals
with chronic underlying diseases, those aged 65 or older,
and those who can transmit influenza virus to high-risk
individuals.
Protection provided by vaccines, however, is limited.
First, the vaccinees’ age and immune status significantly af-
fect the vaccines’ effectiveness. Current vaccines, consisting
of either killed virus or recombinant surface glycoproteins,
can prevent illness in approximately 70–80% of healthy in-
dividuals under the age of 65, but only 30–40% in the elderly
(Patriarca et al., 1985). The protection rate is even lower in
infants, immunocompromised individuals, and elderly per-
sons with chronic obstructive pulmonary disease, asthma,
chronic heart disease, diabetes, chronic renal or hepatic dis-
ease, neoplastic disease, or chronic connective tissue disease
(Glezen et al., 1987; Nichol et al., 1998). Thus, the vaccines
are only moderately effective in those who most need pro-
tection. Second, existing vaccines have to be reformulated
each year because the viral antigens (HA and NA) that elicit
the neutralizing antibodies usually undergo changes, ren-
dering the previous year’s vaccine ineffective against new
subtypes. Each year, a decision must be reached before the
flu season as to which virus strains to include in the vaccine;
it then takes 6 months to produce enough vaccine. In the
face of a rapidly approaching epidemic, 6 months is proba-
bly too long to be acceptable. Third, the expense and poten-
tially serious side effects, such as Guillain-Barré syndrome,
associated with annual reformulation of vaccine production
and administration make this approach less than optimal.
Four drugs have been approved for treating and/or pre-
venting influenza virus infection in the US (Cheung and
Lieberman, 2002). Amantadine and rimantadine are in-
hibitors of the M2 ion-channel and can prevent viral un-
coating (Luscher-Mattli, 2000). Zanamivir and oseltamivir
inhibit NA activity and prevent the release of influenza viri-
ons from the cytoplasmic membrane. These drugs usually
reduce the incidence of influenza infection when used pro-
phylactically and reduce the duration of symptoms by 1–2
days when given within 1 or 2 days of infection (Stiver,
2003). However, the widespread use of these drugs is lim-
ited by concerns about side effects, patients’ compliance,
and the possible emergence of drug-resistant variants. For
example, in a family-based clinical study, 33% of recipients
of amantadine began shedding amantadine-resistant virus
by day 5 of treatment (Hayden et al., 1991). For these rea-
sons, these drugs are prescribed only to highly susceptible
individuals.
3. siRNA-based therapy is ideal for inhibiting influenza
virus infection
RNA interference (RNAi) is a process by which
double-stranded RNA (dsRNA) directs sequence-specific
degradation of messenger RNA (mRNA) (Brantl, 2002;
Sharp, 2001; Vaucheret et al., 2001). This phenomenon was
initially observed in plants (Baulcombe, 2002; Vaucheret
et al., 2001), in Caenorhabditis elegans (Fire et al., 1998),
and, recently, in mammalian cells (Elbashir et al., 2001). In
plants, it is an evolutionarily conserved response to virus
infection. Naturally occurring RNAi is initiated by the
dsRNA-specific endonuclease, called Dicer, which proces-
sively cleaves long dsRNA into double-stranded fragments
between 21 and 25 nucleotides long, termed short inter-
fering RNA (siRNA) (Elbashir et al., 2001). siRNAs are
then incorporated into a protein complex that recognizes
and cleaves target mRNAs. RNAi can be triggered in mam-
malian cells by introducing synthetic 21-nucleotide siRNA
duplexes (Elbashir et al., 2001; Fire et al., 1998), bypass-
ing the requirement for Dicer-mediated processing of long
dsRNA. Because 21-nucleotide siRNAs are too short to
induce an interferon response in mammalian cells (Chi
et al., 2003; Kumar and Carmichael, 1998; Semizarov et al.,
2003), yet still able to confer transient interference of gene
expression in a sequence-specific manner, they represent a
new class of molecules that are likely to have significant
medical applications.
 Q. Ge et al. / Virus Research 102 (2004) 37–42
RNAi appears to be ideal for inhibiting influenza virus in-
fection. First, influenza virus is an RNA virus, without any
DNA intermediates during its entire life cycle (Lamb and
Krug, 2001). Besides mRNA, both vRNA and cRNA could
also be potential targets for siRNA-mediated degradation.
Second, the influenza virus genome consists of eight seg-
mented RNAs, encoding a total of 10 proteins. Each pro-
tein is either an integral component of the viral structure or
plays a critical role during the virus life cycle. Interfering
with the production of any one of them is likely to have se-
vere consequences on viral replication and production. Thus,
there are multiple siRNA targets, and combinations of siR-
NAs to different targets may be used simultaneously. The
use of two or more siRNAs simultaneously may be required
to prevent the emergence of resistant virus, analogous to the
use of drug “cocktails” for treating other infectious diseases
(caused by Mycobacteria, HIV, etc.). Third, influenza virus
naturally infects epithelial cells in the upper respiratory tract
and the lungs in humans. Thus, siRNAs can be administered
by inhalation, which would not only be convenient but may
also result in much higher local siRNA concentrations than
could be achieved by parenteral injection. Considering that
the number of virions is probably small at the onset of a
natural infection, sufficient amounts of siRNA may be de-
livered to epithelial cells in the upper airways and the lungs
to inhibit virus replication or production, thus potentially
achieving preventive and/or therapeutic effects. Finally, un-
like vaccines that require the recipients to have a relatively
normal immune system, siRNA-based treatment does not
depend on a functional immune system and should be as ef-
fective in the elderly or immuno-compromised individuals
as in immunocompetent individuals.
4. siRNAs specific for nucleocapsid and transcriptase
are particularly potent in inhibiting influenza virus
production
Among influenza A viruses, 15 HA subtypes and 9 NA
subtypes are known. There are also extensive differences in
nucleotide sequences of other genes among influenza virus
isolates from different species. To design siRNAs that re-
main effective despite antigenic drifts and shifts, we have
focused on regions of the viral genome that are conserved
among different subtypes and strains of virus from human,
chicken, duck, equine, and swine (the influenza sequence
database, www.flu.lanl.gov). We designed a total of 20 siR-
NAs specific for NP, PA, PB1, PB2, M, and NS genes, but not
for HA and NA because they contain no stretch of 21 con-
served nucleotides, a result of extensive variations in these
genes among different virus isolates from humans and other
species.
The ability of synthetic siRNAs to inhibit influenza virus
production was determined in cultured cells by using graded
amounts of siRNAs, introducing siRNAs into the cells ei-
ther before or after virus infection, and assaying virus titers
39
PB2
5’ 3’
UTR
PB1
Coding sequence
PA
NP siRNA
M
No inhibition
Partial inhibition
NS Strong inhibition
Fig. 1. Comparison of relative potency of siRNAs specific for different
influenza viral genes in inhibiting influenza virus production. Locations
of siRNAs relative to their respective viral genes (PB2, PB1, etc.) are
depicted. The relative potency of siRNAs in inhibiting influenza virus
production is based on results in MDCK cells and chicken embryos. UTR:
untranslated regions.
in the culture supernatants at different times after infection
(Ge et al., 2003). The ability of siRNAs to inhibit influenza
virus production was also verified in embryonated chicken
eggs. We found that: (i) some siRNAs can potently inhibit
influenza virus production in cultured cells; (ii) influenza
virus production can be inhibited by siRNAs specific for
different viral genes, especially those encoding NP, PA, and
PB1; (iii) siRNA inhibition can still occur in cells that were
infected with virus prior to siRNA introduction; (iv) the
amount of siRNA required for significant inhibition is small
(sub-nanomoles); and (v) siRNAs that potently inhibited in-
fluenza virus production in cultured cells also inhibited virus
production in chicken embryos (Fig. 1).
Because siRNAs are double-stranded, the antisense strand
is complementary to mRNA and cRNA, and the sense strand
is complementary to vRNA. Thus, besides mRNA, siRNA
might also interfere with vRNA and cRNA. To investigate
these possibilities, we used two different approaches: First,
we used NP-specific siRNA in which either the sense (+) or
antisense (−) strand was modified. The modification, which
substituted the 2 -hydroxyl group with a 2 -O-methyl group
at every nucleotide residue, does not affect base-pairing for
duplex formation, but the modified RNA strand no longer
supports RNA interference. We found that inhibition of in-
fluenza virus production required a wildtype antisense (−)
strand of siRNA duplex, indicating that the target of RNA
interference is either mRNA (+) or cRNA (+) or both (Ge
et al., 2003).
Second, we examined the effect of siRNA on the accu-
mulation of the corresponding mRNA, cRNA, and vRNA
using reverse transcription followed by real time PCR (Ge
et al., 2003). We found that M-specific siRNA specifically
inhibited the accumulation of M mRNA. However, siRNA
specific for NP or PA abolished the accumulation of not
only the corresponding mRNA but also vRNA and cRNA
(Fig. 2). These siRNAs also broadly inhibited the accu-
mulation of other viral RNAs. This broad inhibition was
virus-specific as it did not significantly affect RNAs tran-
scribed from cellular genes. The virus-specific inhibition
was also observed in Vero cells, which lack the interferon
␣, ␤, and ␻ genes, indicating that the broad inhibition was
not a result of an interferon response that shuts off all
transcription in the infected cells. In addition, the broad
inhibition was not a result of a general degradation of
40 Q. Ge et al. / Virus Research 102 (2004) 37–42

(A) NP cRNA (B) NP cRNA 3

Virion NP vRNA New virion Virion NP vRNA 3 New virion 4

NP mRNA NP mRNA Other viral RNAs 3

NP siRNA
1
NP protein Degradation NP protein 2

Fig. 2. NP-specific siRNAs inhibit influenza virus production through both direct and indirect mechanisms. (A) Following influenza virus infection,
vRNA is transcribed into mRNA, which serves as a template for synthesis of viral proteins. Transcription of vRNA also produces cRNA, which in turn
serves as a template for synthesis of more vRNA for new virion production. Transcription of NP RNAs and synthesis of NP protein are depicted as an
example. (B) In the presence of NP-specific siRNA, NP mRNA is targeted for degradation (step 1). The resulting decrease in NP mRNA level results in
a decrease in NP protein synthesis (step 2). As NP protein normally binds to vRNA and cRNA and is required for their transcription and replication,
the decrease in NP protein level then indirectly results in a decrease in NP vRNA and cRNA synthesis as well as other viral RNA synthesis (step 3).
The broad shutoff of viral RNA and protein synthesis results in a severe reduction of new virion production (step 4). siRNAs specific for viral RNA
transcriptase also directly target specific mRNA for degradation and indirectly interfere with all viral RNA accumulation.

virus-specific RNAs, because activation of protein kinase R
was not affected by the presence of siRNA in infected cells.
Instead, these findings reveal a critical role of newly
synthesized NP and PA proteins in viral transcription
and replication. NP protein is required for elongation and
antitermination of nascent cRNA and vRNA transcripts
(Beaton and Krug, 1986; Shapiro and Krug, 1988). Without
newly synthesized NP, further viral transcription and repli-
cation are blocked, as is new virion production. Similarly,
without newly synthesized RNA transcriptase, further viral
transcription and replication are evidently inhibited. These
findings are consistent with studies showing that siRNAs
target the degradation of mRNAs but not vRNA of respira-
tory syncytial virus (RSV) (Bitko and Barik, 2001). siRNA
specific for RSV P gene, which encodes a subunit of the
RNA-dependent RNA polymerase, inhibits not only virus
P protein expression but also all other viral protein expres-
sion (Bitko and Barik, 2001). Both the targeted mRNA
degradation and the resulting global inhibition of other viral
RNA transcription make the NP- and PA-specific siRNAs
especially potent inhibitors of influenza virus infection.
and to have a long-lasting effect. These requirements could
be met by intranasal administration of siRNA, which might
be able to achieve a high enough local siRNA concentration
to result in sufficient uptake of siRNA by epithelial cells. In
rapidly dividing cells in cultures, sufficient amount of siRNA
can be introduced to mediate effective gene silencing that
persisted for five cell divisions (McManus et al., 2002). In
humans, epithelial cells in the respiratory tract do not divide
significantly. Thus, it might be possible to achieve preven-
tion of influenza infection by inhalation of siRNA once per
week or even less often.
In infants, the elderly, and in immunocompromised in-
dividuals, influenza virus infection can result in secondary
infections, severe complications, and death. For these seri-
ously ill patients, currently there is no effective treatment.
siRNA-based intervention may be able to fill this spe-
cific gap. Besides intranasal administration, other treatment
routes, such as intravenous injection, may be reasonable
in these life-threatening situations. Besides synthetic siR-
NAs, siRNA transcribed from viral or DNA vectors might
also prove to be useful therapeutically and justifiable under
some circumstances.

5. Potential siRNA-based prophylaxis and therapy of

influenza infection in humans 6. Delivery: the challenge of siRNA-based intervention

Influenza virus normally infects epithelial cells in the up-
per airway and the lungs. In healthy individuals, virus is
cleared about 7 days after infection, and tissue damage is re-
paired about 10–12 days after infection. Studies have shown
that antiviral drugs are effective only when they are admin-
istered prior to virus infection or within 1 or 2 days of in-
fection (Hayden et al., 1991; Stiver, 2003). It is possible that
siRNA may also have to be administered prior to infection
or during the early phase of infection in order to achieve pro-
phylactic or therapeutic effect. For prevention or treatment
of influenza infection on a population-wide scale, siRNA
would have to be convenient to use, relatively inexpensive,
siRNAs are negatively charged and do not readily cross
cell membrane. An effective siRNA-mediated prevention
and treatment of influenza virus infection requires efficient
non-toxic means to deliver siRNAs into epithelial cells of
the respiratory tract or other cell types of interest for other
diseases. Because of the structural and chemical similarities
between siRNA and DNA, methodologies that have been, or
are being, developed for DNA delivery (gene therapy) will
likely be useful for siRNA delivery. Compared to plasmid
DNAs, siRNAs are much smaller and therefore probably
easier to introduce into cells than DNA molecules. In addi-
tion, siRNAs can function in the cytosol whereas DNA has
 Q. Ge et al. / Virus Research 102 (2004) 37–42
to enter the nucleus to initiate transcription (Simeoni et al.,
2003). Thus, it is possible that effective siRNA delivery will
be more easily achievable than DNA delivery.
Currently, there are two classes of DNA delivery vehicles:
genetically engineered viruses and synthetic carriers (non-
viral carriers). Adenovirus, lentivirus, and retrovirus have
been widely investigated for gene delivery because of their
high infection efficiency (Thomas et al., 2003). These virus
vectors have also been successfully used to encode short
hairpin RNAs (shRNAs) for expressing siRNAs in cultured
cells and in experimental animals (Hemann et al., 2003;
Rubinson et al., 2003; Shen et al., 2003; Xia et al., 2002). Al-
though these viral vectors are efficient research tools, there
will likely be serious concerns about their medical use in
humans. For example, the viruses tend to infect many cell
types and may induce strong immune responses against in-
fected cells, resulting in the elimination of the infected cells.
Some viruses can also integrate into the host genome and
might result in tumorigenesis. In particular, the idea of us-
ing one virus to prevent or treat infection caused by another
virus may be difficult to gain regulatory approval, perhaps
under some circumstances.
Synthetic carriers, such as peptides, cationic lipids, and
cationic polymers, have also been investigated for DNA
delivery. The most widely investigated non-viral carriers
for DNA transfection are cationic polymers (Han et al.,
2000). Cationic polymers bind electrostatically to DNA
and condense large plasmid DNA molecules into smaller
DNA/polymer complexes for more efficient endocytosis.
The DNA/cationic polymer complexes also act as bioad-
hesives because of their electrostatic interaction with nega-
tively charged sialic acid residues of cell surface glycopro-
teins (Soane et al., 1999). The ability of cationic polymers
to bind to negatively charged siRNA and still interact with
the negatively charged cell surface may facilitate endo-
cytic uptake of siRNAs. In addition, some polymers, such
as imidazole group-modified polylysine, can apparently
disrupt the acidic endosomes and, therefore, facilitate re-
lease of DNA or RNA into the cytosol (Putnam et al.,
2001).
The advantages of nonviral carriers include greater con-
trol of the molecular composition of the final products, flex-
ibility in the amount of siRNAs to be delivered, and reduced
safety concerns. In addition, some polymers are also low
in immunogenicity. However, the delivery efficiency of the
nonviral carriers is generally low compared to viral vectors.
Relatively large doses of DNA, and perhaps siRNA, are usu-
ally required for animal experiments as well as clinical trials
in humans (Donnelly et al., 2003). It is reported that approx-
imately 106 copies of plasmid DNA molecules are needed
to transfect a single cell, with ∼102 –104 copies having to
make it into the nucleus for transcription (Tachibana et al.,
2002). Nevertheless, rapid development of novel polymers
or modification of existing polymers or cationic lipids will
likely yield carriers that can deliver DNA and siRNA into
cells efficiently (Anderson et al., 2003).
41
Recently, synthetic siRNAs and shRNAs transcribed from
DNA templates were shown to potently inhibit gene ex-
pression in mice (McCaffrey et al., 2002, 2003). These re-
sults suggest that siRNA can be introduced into the animals
and mediate effective gene silencing. However, the hydrody-
namic intravascular infusion that was used to deliver siRNA
or DNA into mouse liver can result in transient right-sided
heart failure and liver injury (Zhang et al., 2000), making it
unlikely to be applicable to humans. Development of tech-
nologies for safe, efficient, and target cell-specific delivery
of siRNA will be the key to successful utilization of siRNA
in medical applications.
7. Conclusions
Influenza virus infection is a severe public health prob-
lem with significant personal, social, and economical conse-
quences. More effective approaches are urgently needed to
treat influenza virus infections, especially in infants, the el-
derly, and immunocompromised individuals. siRNA-based
intervention seems ideal for this purpose and may meet this
specific need. siRNAs specific for nucleocapsid and tran-
scriptase have been shown to be particularly potent in in-
hibiting influenza virus production in cell lines and chicken
embryos. The challenge of successful siRNA-based therapy
in the near future is to develop efficient and safe means to
deliver siRNA into cells of interest.
Acknowledgements
We thank Drs. Peter Palese and Adolfo Garcia-Sastre for
kindly allowing Q.G. to visit their laboratories, Drs. Christo-
pher Basler, Astrid Flandorfer, Luis Martinez-Sobrido, and
Yurie Nakaya for teaching Q.G. various techniques for han-
dling influenza virus; and members of the Chen and Eisen
laboratories for helpful discussions. This work was sup-
ported in part by grants from the National Institutes of Health
AI40146, AI44478, and AI50631 (to J.C.), AI44477 and
CA60686 (to H.N.E.).
References
Anderson, D.G., Lynn, D.M., Langer, R., 2003. Semi-automated synthesis
and screening of a large library of degradable cationic polymers for
gene delivery. Angew Chem. Int. Ed. Engl. 42, 3153–3158.
Baulcombe, D., 2002. RNA silencing. Curr. Biol. 12, R82–R84.
Beaton, A.R., Krug, R.M., 1986. Transcription antitermination during in-
fluenza viral template RNA synthesis requires the nucleocapsid pro-
tein and the absence of a 5 capped end. Proc. Natl. Acad. Sci. U.S.A.
83, 6282–6286.
Bitko, V., Barik, S., 2001. Phenotypic silencing of cytoplasmic genes
using sequence-specific double-stranded short interfering RNA and its
application in the reverse genetics of wild type negative-strand RNA
viruses. BMC Microbiol. 1, 34.
42 Q. Ge et al. / Virus Research 102 (2004) 37–42
Brantl, S., 2002. Antisense-RNA regulation and RNA interference.
Biochim. Biophys. Acta 1575, 15–25.
Cheung, M., Lieberman, J.M., 2002. Influenza: update on strategies for
management. Contemp. Pediatr. 19, 82–114.
Chi, J.T., Chang, H.Y., Wang, N.N., Chang, D.S., Dunphy, N., Brown,
P.O., 2003. Genomewide view of gene silencing by small interfering
RNAs. Proc. Natl. Acad. Sci. U.S.A. 100, 6343–6346.
Claas, E.C., A.D., O., van Beek, R., De Jong, J.C., Rimmelzwaan, G.F.,
Senne, D.A., Krauss, S., Shortridge, K.F., Webster, R.G., 1998. Human
influenza A H5N1 virus related to a highly pathogenic avian influenza
virus. Lancet 351, 472–477.
Donnelly, J., Berry, K., Ulmer, J.B., 2003. Technical and regulatory hurdles
for DNA vaccines. Int. J. Parasitol. 33, 457–467.
Elbashir, S.M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., Tuschl,
T., 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference
in cultured mammalian cells. Nature 411, 494–498.
Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver, S.E., Mello,
C.C., 1998. Potent and specific genetic interference by double-stranded
RNA in Caenorhabditis elegans. Nature 391, 806–811.
Ge, Q., McManus, M., Nguyen, T., Shen, C.-H., Sharp, P.A., Eisen, H.N.,
Chen, J., 2003. RNA interference of influenza virus production by
directly targeting mRNA for degradation and indirectly inhibiting all
viral RNA transcription. Proc. Natl. Acad. Sci. U.S.A. 100, 2718–
2723.
Glezen, W.P., Decker, M., Perrotta, D.M., 1987. Survey of underlying
conditions of persons hospitalized with acute respiratory disease during
influenza epidemics in Houston, 1978–1981. Am. Rev. Respir. Dis.
136, 550–555.
Han, S.-O., Mahato, R.I., Sung, Y.K., Kim, S.W., 2000. Development of
biomaterials for gene therapy. Mol. Ther. 2, 302–317.
Hayden, F.G., Sperber, S.J., Belshe, R.B., Clover, R.D., Hay, A.J., Pyke, S.,
1991. Recovery of drug-resistant influenza A virus during therapeutic
use of rimantadine. Antimicrob. Agents. Chemother. 35, 1741–1747.
Hemann, M.T., Fridman, J.S., Zilfou, J.T., Hernando, E., Paddison, P.J.,
Cordon-Cardo, C., Hannon, G.J., Lowe, S.W., 2003. An epi-allelic
series of p53 hypomorphs created by stable RNAi produces distinct
tumor phenotypes in vivo. Nat. Genet. 33, 396–400.
Kemble, G., Greenberg, H., 2003. Novel generations of influenza vaccines.
Vaccine 21, 1789–1795.
Kumar, M., Carmichael, G.G., 1998. Antisense RNA: function and fate
of duplex RNA in cells of higher eukaryotes. Microbiol. Mol. Biol.
Rev. 62, 1415–1434.
Lamb, R.A., Krug, R.M., 2001. Orthomyxoviridae: the viruses and their
replication. In: Knipe, D.M., Howley, P.M. (Eds.), Fundamental Vi-
rology, pp. 725–770.
Luscher-Mattli, M., 2000. Influenza chemotherapy: a review of the present
state of art and of new drugs in development. Arch. Virol. 145, 2233–
2248.
Maassab, H.F., Bryant, M.L., 1999. The development of live attenuated
cold-adapted influenza virus vaccine for humans. Rev. Med. Virol. 9,
237–244.
McCaffrey, A.P., Meuse, L., Pham, T.T., Conklin, D.S., Hannon, G.J.,
Kay, M.A., 2002. RNA interference in adult mice. Nature 418, 38–39.
McCaffrey, A.P., Nakai, H., Pandey, K., Huang, Z., Salazar, F.H., Xu, H.,
Wieland, S.F., Marion, P.L., Kay, M.A., 2003. Inhibition of hepatitis
B virus in mice by RNA interference. Nat. Biotechnol. 21, 639–644.
McManus, M.T., Haines, B.B., Dillon, C.P., Whitehurst, C.E., van Parijs,
L., Chen, J., Sharp, P.A., 2002. Small interfering RNA-mediated gene
silencing in T lymphocytes. J. Immunol. 169, 5754–5760.
Nichol, K.L., Wuorenma, J., von Sternberg, T., 1998. Benefits of influenza
vaccination for low-, intermediate-, and high-risk senior citizens. Arch.
Intern. Med. 158, 1769–1776.
Palese, P., Garcia-Sastre, A., 2002. Influenza vaccines: present and future.
J. Clin. Invest. 110, 9–13.
Patriarca, P.A., Weber, J.A., Parker, R.A., Hall, W.N., Kendal, A.P., Breg-
man, D.J., Schonberger, L.B., 1985. Efficacy of influenza vaccine in
nursing homes. Reduction in illness and complications during an in-
fluenza A (H3N2) epidemic. JAMA 253, 1136–1139.
Patterson, K.D., Pyle, G.F., 1991. The geography and mortality of the
1918 influenza pandemic. Bull. Hist. Med. 65, 4–21.
Putnam, D., Gentry, C.A., Pack, D.W., Langer, R., 2001. Polymer-based
gene delivery with low cytotoxicity by a unique balance of side-chain
termini. Proc. Natl. Acad. Sci. U.S.A. 98, 1200–1205.
Rubinson, D.A., Dillon, C.P., Kwiatkowski, A.V., Sievers, C., Yang, L.,
Kopinja, J., Rooney, D.L., Ihrig, M.M., McManus, M.T., Gertler,
F.B., Scott, M.L., Van Parijs, L., 2003. A lentivirus-based system to
functionally silence genes in primary mammalian cells, stem cells
and transgenic mice by RNA interference. Nat. Genet. 33, 401–
406.
Semizarov, D., Frost, L., Sarthy, A., Kroeger, P., Halbert, D.N., Fesik,
S.W., 2003. Specificity of short interfering RNA determined through
gene expression signatures. Proc. Natl. Acad. Sci. U.S.A. 100, 6347–
6352.
Shapiro, G.I., Krug, R.M., 1988. Influenza virus RNA replication in vitro:
synthesis of viral template RNAs and virion RNAs in the absence of
an added primer. J. Virol. 62, 2285–2290.
Sharp, P.A., 2001. RNA interference—2001. Genes Dev. 15, 485–
490.
Shen, C., Buck, A.K., Liu, X., Winkler, M., Reske, S.N., 2003. Gene
silencing by adenovirus-delivered siRNA. FEBS Lett. 539, 111–
114.
Simeoni, F., Morris, M.C., Heitz, F., Divita, G., 2003. Insight into the
mechanism of the peptide-based gene delivery system MPG: impli-
cations for delivery of siRNA into mammalian cells. Nucleic Acids
Res. 31, 2717–2724.
Soane, R.J., Frier, M., Perkins, A.C., Jones, N.S., Davis, S.S., Illum,
L., 1999. Evaluation of the clearance characteristics of bioadhesive
systems in humans. Int. J. Pharm. 178, 55–65.
Stiver, G., 2003. The treatment of influenza with antiviral drugs. CMAJ
168, 49–56.
Tachibana, R., Harashima, H., Ide, N., Ukitsu, S., Ohta, Y., Suzuki, N.,
Kikuchi, H., Shinohara, Y., Kiwada, H., 2002. Quantitative analysis of
correlation between number of nuclear plasmids and gene expression
activity after transfection with cationic liposomes. Pharm. Res. 19,
377–381.
Taubenberger, J.K., Reid, A.H., Janczewski, T.A., Fanning, T.G., 2001.
Integrating historical. Phil. Trans. R. Soc. Lond. B: Biol. Sci. 356,
1829–1839.
Thomas, C.E., Ehrhardt, A., Kay, M.A., 2003. Progress and problems with
the use of viral vectors for gene therapy. Nat. Rev. Genet. 4, 346–358.
Thompson, W.W., Shay, D.K., Weintraub, E., Brammer, L., Cox, N.,
Anderson, L.J., Fukuda, K., 2003. Mortality associated with influenza
and respiratory syncytial virus in the United States. JAMA 289, 179–
186.
Vaucheret, H., Beclin, C., Fagard, M., 2001. Post-transcriptional gene
silencing in plants. J. Cell Sci. 114, 3083–3091.
Webby, R.J., Webster, R.G., 2001. Emergence of influenza A viruses.
Phil. Trans. R. Soc. Lond. B: Biol. Sci. 356, 1817–1828.
Webster, R.G., Bean, W.J., Gorman, O.T., Chambers, T.M., Kawaoka, Y.,
1992. Evolution and ecology of influenza A viruses. Microbiol. Rev.
56, 152–179.
Webster, R.G., Laver, W.G., Air, G.M., Schild, G.C., 1982. Molecular
mechanisms of variation in influenza viruses. Nature 296, 115–121.
Xia, H., Mao, Q., Paulson, H.L., Davidson, B.L., 2002. siRNA-mediated
gene silencing in vitro and in vivo. Nat. Biotechnol. 20, 1006–1010.
Yuen, K.Y., Chan, P.K., Peiris, M., Tsang, D.N., Que, T.L., Shortridge,
K.F., Cheung, P.T., To, W.K., Ho, E.T., Sung, R., Cheng, A.F., 1998.
Clinical features and rapid viral diagnosis of human disease associated
with avian influenza A H5N1 virus. Lancet 351, 467–471.
Zhang, G., Song, Y.K., Liu, D., 2000. Long-term expression of human
alpha1-antitrypsin gene in mouse liver achieved by intravenous ad-
ministration of plasmid DNA using a hydrodynamics-based procedure.
Gene Ther. 7, 1344–1349.
 Virus Research 102 (2004) 43–51

RNA silencing of rotavirus gene expression

Carlos F. Arias, Miguel A. Dector, Lorenzo Segovia, Tomás López, Minerva Camacho,
Pavel Isa, Rafaela Espinosa, Susana López∗
Departamento de Génetica del Desarrollo y Fisiologı́a Molecular, Instituto de Biotecnologı́a, Universidad Nacional Autónoma de México,
Avenida Universidad 2001, Col. Chamilpa, Cuernavaca, Morelos 62210, Mexico

Abstract

RNA interference (RNAi) is a double-stranded RNA (dsRNA)-triggered mechanism for suppressing gene expression, which is conserved
in evolution and has emerged as a powerful tool to study gene function. Rotaviruses, the leading cause of severe diarrhea in young children,
are formed by three concentric layers of protein, and a genome composed of 11 segments of dsRNA. Here, we show that the RNAi machinery
can be triggered to silence rotavirus gene expression by sequence-specific short interfering RNAs (siRNAs). RNAi is also useful for the study
of the virus-cell interactions, through the silencing of cellular genes that are potentially important for the replication of the virus. Interestingly,
while the translation of mRNAs is readily stopped by the RNAi machinery, the viral transcripts involved in virus genome replication do not
seem to be susceptible to RNAi. Since gene silencing by RNAi is very efficient and specific, this system could become a novel therapeutic
approach for rotavirus and other virus infections, once efficient methods for in vivo delivery of siRNAs are developed. Although the use of
RNAi as an antiviral therapeutic tool remains to be demonstrated, there is no doubt that this technology will influence drastically the way
postgenomic virus research is conducted.
© 2004 Elsevier B.V. All rights reserved.

Keywords: Rotavirus; RNA interference; siRNAs; Antiviral agents; Reoviridae

1. Introduction trol the spread of rotavirus infections, there is an urgent need

Acute, infectious diarrhea is the most common cause of
morbidity and mortality among young children living in de-
veloping countries, accounting for as many as one billion
illnesses and between 2.5 and 3.2 million deaths annually
(Parashar and Glass, 2003). Rotaviruses are the leading eti-
ologic agent of severe diarrheal disease in infants and young
children worldwide. In developed countries, rotaviruses have
been detected in 30–50% of infants hospitalized with acute
diarrhea. In less developed countries, rotaviruses are also
the most frequently detected pathogen in children with se-
vere gastroenteritis. While the mortality from rotavirus dis-
ease in developed countries is very low, rotavirus causes an
estimated 500,000–600,000 deaths each year (Parashar and
Glass, 2003). The frequency of rotavirus infection is remark-
ably similar in both settings. Since rotaviruses play such an
important role in severe dehydrating gastroenteritis, and be-
cause even advanced levels of hygiene seem unable to con-
∗ Corresponding author. Tel.: +52-777-3291615;
fax: +52-777-3172388.
E-mail address: susana@ibt.unam.mx (S. López).
to develop effective vaccination and therapeutic strategies.
Fundamental to these developments is a basic understanding
of the molecular mechanisms by which rotaviruses interact
with their host cell.
Rotaviruses, a genus of the Reoviridae family, are formed
by three concentric layers of protein that enclose a genome
composed of 11 segments of double stranded RNA (dsRNA),
ranging in size from approximately 660–3300 base pairs
(bp). The total genomic size is about 18,500 bp. The inner-
most layer of the virion is formed by 120 molecules of pro-
tein VP2, which surrounds the viral genome. Twelve copies
each of VP1, the RNA polymerase, and VP3, a guanylyl-
transferase and methylase, constitute the core of the virus.
The addition of 260 trimers of VP6 on top of the VP2
layer produces double-layered particles (DLPs). The outer-
most layer, characteristic of triple-layered particles (TLPs),
is composed of two proteins, VP4 and VP7. The smooth
external surface of the virus is made up of 780 copies of
glycoprotein VP7, organized as trimers, while 60 spike-like
structures, formed by dimers of VP4, extend about 12 nm
from the VP7 surface. The mature virus particle is ap-
proximately 100 nm in diameter and contains 132 porous

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.014
44 C.F. Arias et al. / Virus Research 102 (2004) 43–51

channels, which allow the influx of compounds in aqueous
solution to the inside of the capsid and the efflux of newly
formed mRNAs (Pesavento et al., 2003).
VP4 has essential functions in the virus life cycle, includ-
ing receptor binding and cell penetration. The role of VP7
during the initial interactions of the virus with the cell is less
clear, although it has been recently shown that VP7 interacts
with cell surface molecules at a step subsequent to the initial
attachment of the virus through the spike protein (Graham
et al., 2003; Zárate et al., 2002). After the virus attaches to
the cell surface, it has to penetrate the plasma membrane to
productively infect the cell. This penetration is increased by,
and most probably dependent on, trypsin treatment of the
virus that results in the specific cleavage of VP4 to polypep-
tides VP8 and VP5. The cleavage of VP4 does not affect
cell binding and is rather associated with the entry of the
virus into the cell (Arias et al., 2001; Estes, 2001).
During, or shortly after cell entry, the infecting TLP is
uncoated, loosing the two proteins of the outer layer, and
yielding a DLP, which is transcriptionally active. VP1 syn-
thesizes the primary viral transcripts, which are extruded into
the cell’s cytoplasm through the class I channels located at
the icosahedral five-fold vertices of the particle (Pesavento
et al., 2003).
The RNA transcripts direct the synthesis of six structural
and six non-structural viral proteins (i.e., function as mR-
NAs) and also serve as RNA templates (RNA(+)) for the
synthesis of the RNA negative strands (RNA(−)), to form
the dsRNA genome segments. Once a critical mass of viral
proteins is accumulated into structures known as viroplasms,
core replication intermediate (RI) particles assemble. The
synthesis of RNA(−) has been proposed to occur concur-
rently with the packaging of RNA(+) into the core RIs, and
is highly coordinated in such a way that packaging and repli-
cation lead to the formation of cores containing one copy
of each of the 11 genome segments of dsRNA (Patton and
Spencer, 2000). In addition to the virus RNA polymerase,
the non-structural proteins NSP2 and NSP5 are thought to
be essential for the early steps of morphogenesis. All these
processes lead to the production of transcriptionally ac-
tive, dsRNA-containing double-layered RI particles. These
particles are responsible for an enhanced second round of

Fig. 1. Replication cycle of rotaviruses. The different steps in the replication cycle of the virus are indicated by numbers—1: attachment of the virion to
the cell surface; 2: penetration and uncoating of the virus particle to yield DLPs; 3: primary transcription of the genomic dsRNA; 4: synthesis of viral
proteins; 5: assembly of core RIs and negative strand RNA synthesis; 6: assembly of double-layered RIs; 7: secondary transcription from double-layered
RIs; 8: secondary, enhanced synthesis of viral proteins; 9: secondary, increased assembly of core RIs and negative strand RNA synthesis; 10: secondary,
increased assembly of double-layered RIs; 11: budding of double-layered RIs through the membrane of the endoplasmic reticulum (ER), and acquisition
of a membrane envelope; 12: loss of the membrane envelope and generation of mature triple-layered virions.
 C.F. Arias et al. / Virus Research 102 (2004) 43–51
transcription, which results in a second wave of assembly
of double-layered RI particles, which then bud through the
membrane of the ER. During this budding, which is medi-
ated by the interaction of the double-layered RIs with the ER
membrane-associated rotavirus protein NSP4, the particles
acquire a transient membrane envelope (Estes, 2001). The
transiently enveloped particles contain, in addition to NSP4,
the virus surface proteins VP4 and VP7, as well as minor
amounts of other non-structural proteins (Poruchynsky and
Atkinson, 1991). The lipid envelope is then removed by a
largely unknown mechanism, to yield mature TLPs (Fig. 1).
Apart from glycoprotein NSP4, the other five nonstruc-
tural proteins (NSP1 to NSP3, NSP5, and NSP6) have the
ability to bind RNA (Patton, 1995) and are thought to be in-
volved in the replication of the viral genome, although their
precise function is not known. Despite the efforts of many
laboratories, at present there is no reverse genetics system
available for rotaviruses. Thus, the in vivo characterization of
rotavirus gene function has been limited so far to the charac-
terization of antibody escape variants, temperature-sensitive
mutants, and reassortants.
2. RNA interference and rotaviruses
RNA interference (RNAi) has become a powerful and
widely used tool for the analysis of gene functions.
This evolutionarily conserved mechanism, triggered by
double-stranded RNA (dsRNA), specifically suppresses
gene expression by selectively degrading mRNAs matching
the sequence of the dsRNA that triggered the response,
without affecting the expression of other genes. During
RNAi, long dsRNA molecules are processed into 21–25 bp
RNAs known as short-interfering RNAs (siRNAs) that serve
as guides for enzymatic cleavage of complementary RNAs,
thus offering an exquisite mode to specifically knockdown
the expression of a particular gene.
This phenomenon has been observed in invertebrates and
plants (Hannon, 2002; Sharp, 2001; Zamore, 2001); how-
ever, the demonstration of an RNAi-like response in somatic
mammalian cells had been hampered by dsRNA-induced
mechanisms, which non-specifically inhibit gene expres-
sion. In mammalian cells, dsRNA fragments longer than
30 bp induce components of the interferon response, in-
cluding the dsRNA-dependent protein kinase (PKR), the
2 ,5 -oligoadenylate synthetase (OAS) and the non-specific
RNAse L (Katze et al., 2002). These factors mediate a gener-
alized inhibition of gene expression through the non-specific
shutdown of protein synthesis and mRNA degradation, what
eventually leads to cell apoptosis. The recent demonstration
that synthetic 21-nt siRNA duplexes effectively inhibit mam-
malian endogenous gene expression in a sequence-specific
manner without activating the nonspecific dsRNA responses
(Caplen, 2002; Elbashir et al., 2001) offers great opportuni-
ties to use this pathway of gene silencing to study the func-
tion of mammalian virus genes.
45
Rotaviruses have a segmented dsRNA genome, which is
replicated in the cytoplasm of cells. The dsRNA genome,
however, is masked at all times during the virus replication
cycle, being found always associated with subviral particles
and not free in the cellular cytosol. This probably prevents
the rotavirus genome to be detected and targeted by gen-
eral defense mechanisms of the cell. In addition, the virus
replication cycle is highly lytic and rapid, being completed
in about 12 h. The features of rotavirus replication repre-
sented a challenge for the RNAi machinery to silence the
virus gene expression and to block the replication of the
virus.
We showed that a siRNA with a sequence homologous to
the gene encoding the virus spike protein VP4 (siRNAVP4 )
efficiently inhibited the synthesis of VP4 to a barely de-
tectable level in cells transfected with the siRNA, whereas
all other structural proteins remained unaltered (Dector
et al., 2002). Similarly, the yield of progeny virus that
resulted from an infection carried out in the presence of
siRNAVP4 was reduced to 15–25% that of a control in-
fection performed in the presence of an unrelated siRNA
(homologous to the nuclear proteins laminA/C). The resid-
ual infectious virus produced in the presence of siRNAVP4
was mostly the result of virus replication in those cells that
were not transfected with the siRNA, the transfection effi-
ciency achieved being about 75%. As result of the inhibition
of VP4 synthesis, non-infectious TLPs, lacking the VP4
protein (appearing as “spike-less” TLPs), were obtained
(Dector et al., 2002).
We have also addressed the question of the duration of
an effective RNAi response in a rotavirus infection. For
this, we transfected MA104 cells with either a siRNA to
VP4 or to VP7, and then infected with rotavirus at differ-
ent times post-transfection. The inhibition of the synthe-
sis of both viral proteins could be achieved even when the
cells were infected 7 days after transfection (shown up to
day 5 in Fig. 2A), with the consequence of a significant
decrease on the yield of viral progeny (Fig. 2B). The in-
hibitory effect could probably last even longer, however,
after the fifth day cells in culture were notoriously dam-
aged, even in the untransfected controls, so that it was dif-
ficult to have reproducible results at later times. Of inter-
est, we consistently found that the highest inhibition was
achieved when the cells were infected 72 h after transfec-
tion with the siRNA, suggesting that some of the elements
of the RNAi response could be induced or activated by
the presence of the siRNA, increasing their effective con-
centration inside the cell. Our observations are in agree-
ment with other studies carried out with different viruses
and different cell lines, where the RNAi effect was re-
ported to last also about 5 days (Flores-Jasso et al., 2004;
Ge et al., 2004; Kapadia et al., 2003). It is not clear, how-
ever, whether this effect is due to the fact that the siRNAs
are stable within the cells, or whether they are slowly re-
leased from the membrane-associated transfection mixture
(Gitlin and Andino, 2003).
46 C.F. Arias et al. / Virus Research 102 (2004) 43–51
Fig. 2. The RNAi response lasts up to 7 days. MA104 cells in 48-well
plates were transfected with an siRNA directed to VP4 (4), VP7 (7),
or lamin A/C (C), and after 8 h the lipofectamine–siRNA mixture
was removed and replaced by medium. At the indicated times (days
post-transfection, d.p.t.), the cells were infected with rotavirus RRV at an
MOI of 1 and 12 h post-infection (p.i.), the cells were harvested and pro-
cessed for electrophoresis or infectivity assays. (A) Western blot stained
with a rabbit polyclonal antibody to rotavirus TLPs. Arrows indicate the
positions of VP4, VP6, and VP7. (B) The yield of progeny virus ob-
tained after transfection with the siRNA to VP4 was determined by an
immunoperoxidase assay, as described (Lizano et al., 1991). Data are ex-
pressed as the percentage of the infectivity obtained when the cells were
transfected with the control siRNA to lamin A/C.
The results obtained in our work indicate that rotaviruses
are sensitive to the RNAi response of the cell, as has been
described for viruses belonging to other families, such as po-
liovirus, hepatitis B and C viruses, human immunodeficiency
virus, respiratory syncytial virus, vesicular stomatitis virus,
papillomavirus, herpes virus, influenza virus, dengue virus,
and baculovirus, some of which are discussed in accom-
panying articles of this Issue (Adelman et al., 2002; Bitko
and Barik, 2001; Capodici et al., 2002; Coburn and Cullen,
2002; Ge et al., 2003; Gitlin et al., 2002; Hall and
Alexander, 2003; Hamasaki et al., 2003; Hu et al., 2002;
Jacque et al., 2002; Jia and Sun, 2003; Jiang and Milner,
2002; Kapadia et al., 2003; Klein et al., 2003; Martinez
et al., 2002; McCaffrey et al., 2003; Novina et al., 2002;
Randall et al., 2003; Shlomai and Shaul, 2003;Valdes et al.,
2003; Wilson et al., 2003).
3. Is there a suppression of the RNAi response during
rotavirus infection?
It has been found that many plant viruses, and at least
one insect virus (flock house virus) are able to counteract
Fig. 3. Rotavirus gene expression is efficiently inhibited even if RNAi is
induced after viral infection. MA104 cells in 48-well plates were infected
for 1 h with rotavirus RRV, at an MOI of 3. At the indicated times
post-infection (hours post-infection, h.p.i.), the cells were transfected with
a siRNA to VP4 (4), or lam A/C(C) as control, for 8 h. At 12 h.p.i. the
cells were labeled with [35 S]-methionine for 1 h, lysed, and the proteins
separated by gel electrophoresis and analyzed by fluorography (A), or by
Western blot using a monoclonal antibody to VP4 (B). The dots in (A)
indicate the migration of viral proteins other than VP4.
the RNA interference of the cell by producing proteins that
suppress this response at different levels (Ding et al., 2004;
Li et al., 2002; Roth et al., 2004; Vance and Vaucheret, 2001).
Although it is not yet known whether the RNAi has antiviral
functions in mammalian cells, it is tempting to speculate that
one of the rotavirus proteins could act as a suppressor of a
cellular RNAi response triggered by the viral infection.
In a previous publication we described that the synthesis
of a viral protein in MA104 cells lipofected with a siRNA
specific for a viral protein 48–72 h prior to infection, was
silenced in a sequence-specific manner (Dector et al., 2002,
unpublished results), suggesting that rotavirus infection can-
not suppress the interfering process. However, since in those
experiments the siRNA was added to the cells before virus
infection, we reasoned that for a virus-encoded suppressor
of RNAi, there might have not been enough time for its syn-
thesis before the RNAi response (triggered by the siRNA)
inhibited the virus replication. To address this issue, we eval-
uated the capacity of the RNAi response to silence rotavirus
gene expression when the siRNAs were transfected into cells
after infection with the virus. We found that the synthesis
of VP4 was silenced even when the siRNAVP4 was added
at 4 h post-infection (p.i.), a time point at which viral pro-
tein synthesis is already predominant in the infected cells
(Fig. 3). These results suggest that the RNAi machinery is
active and effective even in the presence of previously syn-
thesized rotaviral proteins, and thus that rotaviruses do not
encode an RNAi suppressor. However, in order to formally
 C.F. Arias et al. / Virus Research 102 (2004) 43–51
demonstrate that rotaviruses do not code for suppressors of
RNAi, further experiments need to be done. For example, it
is necessary to determine if rotavirus infection could block
an RNAi response triggered by short hairpin RNAs, which
have been shown to specifically induce the RNAi system.
These short hairpin RNAs are substrates for the cellular en-
zyme Dicer (Brummelkamp et al., 2002), which is respon-
sible for processing long dsRNA molecules into siRNAs.
Also, it could be helpful to assay candidate viral proteins
as RNAi suppressors in a heterologous system, like plants.
With such an approach, it was recently reported that the ␴3
protein of reovirus, which is a dsRNA binding protein, has
a supressor activity that blocks the synthesis of GFP in an
assay for suppressor silencing in tobacco plants (Lichner
et al., 2003; Roth et al., 2004).
4. RNA interference to study the function of rotavirus
proteins
The segmented nature of the genome of the viruses in
the Reoviridae family makes them particularly amenable to
analysis by RNAi. In the case of rotaviruses, each of the
11 segments of dsRNA is transcribed into a single mRNA,
each encoding a single protein (with the exception of one
bicistronic segment). This makes it possible to silence the
expression of individual genes without affecting the expres-
sion of the others. The analysis of the phenotypes generated
should allow the characterization of the function of proteins
encoded by the silenced genes.
This sort of individual gene function analysis is more dif-
ficult to carry out in viruses with a positive strand RNA
genome, since, as it has been shown for poliovirus, hep-
atitis C (Randall and Rice, 2004; Saleh et al., 2004), and
SARS-coronavirus (Zhang et al., 2003), their genome is
targeted by the RNAi machinery, inhibiting the replication
of the virus. Since the genome of these viruses functions
as mRNA to direct the synthesis of precursor polyproteins
which are post-translationally cleaved to yield the individual
viral polypeptides, any siRNA directed against the mRNA
(represented by either the genomic RNA, or the subgenomic
mRNAs in some virus families) would result in a decreased
synthesis of the entire polyprotein, preventing the study
of the role of individual polypeptides. In this regard, it is
of interest that in viruses with a non-segmented, negative
strand RNA genome, like respiratory syncytial virus, vesic-
ular stomatitis virus, and human parainfluenza virus, the ge-
nomic RNA as well as the replication intermediate RNA(+),
are protected from the action of RNAi, while only the mR-
NAs are degraded (Barik, 2004; Bitko and Barik, 2001).
This also seems to be the case for influenza A viruses which
have a genome composed of eight single-stranded RNA seg-
ments of negative polarity (Ge et al., 2003). The functions
of proteins of DNA viruses and retroviruses, whose genome
is in general transcribed into monocistronic mRNAs, is also
amenable to dissection by RNAi, as has been shown for bac-
47
ulovirus, human immunodeficiency virus, and retroviruses
in this issue, and for herpes virus, papillomavirus and the
hepatitis delta agent (Chang and Taylor, 2003; Hall and
Alexander, 2003; Jia and Sun, 2003).
In the case of rotaviruses, one could explore the function
of the viral proteins in three different conceptual areas: (i)
proteins that do not participate in the replication of the virus
genome, but are involved in the late steps of virus morpho-
genesis; (ii) proteins involved directly or indirectly in the
replication of the genomic RNA; and (iii) proteins that, al-
though not directly involved in virus replication, contribute
to make the replication cycle more efficient. Some viral pro-
teins could certainly have activities that could fit into more
than one of the areas described.
Silencing the viral genes that encode proteins involved in
the late steps of morphogenesis, like VP4 and VP7, should
allow the replication cycle to reach the synthesis of the
second wave of double-layered RIs. However, after this
point the morphogenesis of the virus should be impaired.
In agreement with this hypothesis, we have shown that the
silencing of the VP4 gene allowed the synthesis of abun-
dant double-layered RIs, that efficiently bud through the ER
membrane, eventually loosing the transient lipid membrane
and incorporating the outer protein layer formed by VP7,
to form spike-less TLPs. Of interest, the VP7 outer layer
assembled on the spike-less TLPs was found by cryoelec-
tron microscopy to have an structure identical to that of the
outer layer in wild-type TLPs (B.V.V. Prasad, unpublished
results), clearly showing that the assembly of VP7 on DLPs
is a process independent of the assembly of VP4. In addition,
these findings indicated that VP4 is neither required for the
budding process nor for the removal of the lipid membrane
from the enveloped particles, as had been suggested (Estes,
2001). Similarly, silencing the VP7 gene prevented the for-
mation of TLPs, and caused DLPs to accumulate (Patton,
2003; Camacho et al., manuscript in preparation). A priori,
one would expect that inhibition of the synthesis of NSP4,
the ER membrane receptor for DLPs, should also lead to an
accumulation of DLPs. However, since it is known that this
protein alters the calcium homeostasis of cells (Tian et al.,
1994, 1995), its absence might have additional effects on the
virus replication cycle. Silencing the NSP4 and VP7 genes
should allow us to establish directly the role of these pro-
teins in the translocation of DLPs into the lumen of the ER
and in the removal of the intermediary lipid envelope.
RNAi should also be useful to study in vivo the role of
proteins that have been suggested to be involved in the early
stages of the virus replication cycle, like the viral polymerase
VP1, the protein that forms the innermost shell of the virion
VP2, and the nonstructural proteins NSP2 and NSP5, all
of which are thought to be necessary to generate core RIs
containing newly formed genomic dsRNA segments (Patton
and Spencer, 2000). One would expect that silencing the
expression of either of these genes would block the virus
secondary transcription, with the concomitant inhibition of
the synthesis of all viral proteins, as it has been found for
48 C.F. Arias et al. / Virus Research 102 (2004) 43–51
VP1 and NSP5 (Campagna et al., 2003; T. López and M.
Arias, manuscript in preparation). A more detailed anal-
ysis of these data should allow us to define the role of
each protein in the formation of functional cores. Also,
since most viral proteins do not have a random distribu-
tion within the cell, the inhibition of the synthesis of par-
ticular proteins should help to identify the role they might
have in the overall intracellular organization of the viral
polypeptides.
In addition, RNAi could serve to approach the character-
ization of the role of viral proteins that have been proposed
to favor the replication cycle of the virus through, for in-
stance, inhibiting the cellular protein synthesis (NSP3), in-
creasing the translation rate of the viral proteins (Piron et al.,
1998), counteracting cellular activities that might be noxious
to virus replication, such as the interferon response (NSP1)
(Graff et al., 2002), or promoting the cytopathic effect to
facilitate the release of virus particles (NSP4) (Estes, 2003).
It is foreseen that in the near future many new studies
using this methodology will help to unravel the role of all
rotaviral proteins in vivo, both in cell culture and in ani-
mal models. The feasibility of inhibiting virus replication
in complete animals has already been shown for hepatitis C
using the mouse as a model (McCaffrey et al., 2002).
5. RNAi to study rotavirus–host cell interactions
Recently, a number of cellular proteins have been sug-
gested to participate in the virus replication cycle. These
proteins range from virus cellular receptors to chaperones
and transcription factors. It is clear that RNAi will help to
elucidate the role of these cell molecules during the infec-
tious process.
The entry of rotaviruses into cells has been described as
a multistep process where at least four interactions between
the viral surface proteins and cellular surface molecules
occur. The molecules implicated as virus receptors include
gangliosides GM1 and GM3, integrins ␣2␤1, ␣x␤2, and
␣v␤3, and the heat shock cognate protein hsc70 (Arias et al.,
2001). Silencing of individual receptors or combinations
of them should help to establish their role during virus
entry. In fact, preliminary experiments have shown that
inhibition of ␣v␤3 expression decreases the ability of
rotaviruses to infect epithelial cells (P. Isa, unpublished
results).
It has been shown that rotavirus infection induces the
expression of a large set of genes, including those encoding
heat shock proteins hsp90 and hsp70, as well as the glucose
regulated proteins grp78 and grp94 (Cuadras et al., 2002;
Xu et al., 1998; L. Maruri, unpublished). These findings,
together with the fact that cells that are poorly susceptible to
rotavirus infection increase their susceptibility to the virus
up to 100-fold when subjected to a heat shock (T. Lopez,
unpublished results), suggest that stress proteins might be
important for the efficient replication of the virus. The role
of these proteins in the virus replication cycle should be
amenable to analysis by RNAi.
Knockdown of the caveolin-1 gene and other genes en-
coding proteins involved in different types of cell endocyto-
sis, like those encoding clathrin and Eps15, should also help
to study the role of these proteins in virus entry. Using this
approach, together with pharmacological and other genetic
approaches, we have recently described that rotavirus cell
entry occurs through a caveolae- and clathrin-independent
endocytosis, (Sánchez-SanMartı́n et al., submitted).
6. RNAi induces the degradation of only a subset of the
viral transcripts
During rotavirus infection, the DLPs derived from the
infecting viruses synthesize 11 viral transcripts. Since these
transcripts function both as mRNA and as RNA(+), it was
expected that siRNAs directed to a given transcript would
impair the synthesis of both the corresponding protein and
the corresponding dsRNA segment. However, we and others
have found that when silencing rotavirus genes encoding
proteins not involved in the replication of the viral genome,
such as VP4 and VP7, the newly assembled particles, either
spike-less TLPs (in the case of silencing VP4) or DLPs (in
the case of silencing VP7), contain all 11 RNA segments
in equimolar amounts (Dector et al., 2002; Patton, 2003;
Camacho et al., manuscript in preparation), despite the fact
that the synthesis of the target protein is almost completely
ablated, indicating that the mRNA has been degraded.
The previous observations have been taken to suggest
that viral transcripts establish two, functionally different and
physically separated pools (Patton, 2003; Camacho et al.,
manuscript in preparation), one that directs the synthesis of
proteins, and the other that is used as template for repli-
cation of the virus genome. The mRNAs are accessible in
the cytoplasm to be translated (where they can be efficiently
targeted by RNAi), whereas the RNA(+) molecules act-
ing as template for dsRNA replication could be either kept
within the so-called viroplasms (where the synthesis of the
RNA(−) is thought to take place) or could be associated
in the cytoplasm with RNA binding proteins, making them
inaccessible to the RNAi machinery. Even though the ex-
istence of two separated viral transcripts pools would ap-
pear to be logical, the common belief is that viral tran-
scripts represent a single pool of RNA that could be used at
any time, and stochastically, for either of the two functions.
The hypothesis that a mRNA pool could be physically sep-
arated from the RNA(+) (replication) pool might explain
why the attempts to develop a reverse genetic system for ro-
taviruses have so far failed; the introduction of exogenous
transcripts that represent rotavirus mRNAs (obtained by in
vitro synthesis), or the production inside the cell of these
transcripts from recombinant vectors, might not reach the
compartment where such molecules are used as templates for
replication.
 C.F. Arias et al. / Virus Research 102 (2004) 43–51
7. Perspectives for RNAi in rotavirus research
The ability to knock down the expression of individ-
ual rotavirus genes should provide the basis to develop
phenotypic complementation systems in the short term. In
principle, given the high efficiency of RNAi to prevent the
synthesis of a given viral protein, it is conceivable that an al-
most homologous polypeptide (encoded by a gene having a
few nucleotide differences in the siRNA target site) could be
synthesized from a recombinant expression vector, to phe-
notypically rescue the activity of the missing protein. Once
established, such system should allow the expression of
modified versions of the protein to dissect its function. In the
absence of a reverse genetics system, this approach should be
very useful to carry out functional genomics in rotaviruses,
and in viruses of other genera of the Reoviridae. In addition,
it can be forseen that the studies employing RNAi will soon
extend to the analysis of a number of cellular genes whose
protein products influence the replication of the virus in a
positive or negative manner, and more in the mid term, to the
analysis of viral gene function in complete animal models.
One of the limiting steps using RNAi to silence viral
genes is the siRNA transfection efficiency. Thus, the RNAi
studies, including the phenotypic rescue system described
above, will be greatly improved and simplified if all cells
in a culture could harbor the silencing siRNA. This could
be achieved by constructing stable cell lines expressing the
siRNA (either as two complementary strands that will hy-
bridize inside the cell, or as precursor short hairpin RNAs;
Miyagishi et al., 2004), or alternatively, by developing bet-
ter systems to deliver the siRNAs into the cells, for instance,
using lentiviruses or flavivirus replicons as vectors to direct
the intracellular synthesis of the siRNAs. These approaches
should be evaluated carefully to avoid the induction of the
non-specific interferon response by the siRNAs, as has been
recently reported (Bridge et al., 2003; Sledz et al., 2003).
The sequence specificity of RNAi is remarkable; siRNAs
that differ from their target sequence in one or more bases,
depending on the position of the mismatch, do not efficiently
silence the expression of that gene. This fact has been re-
garded at as a drawback for the use of siRNAs as potential
therapeutic antiviral compounds, since variant viruses could
escape the RNAi response, as has been shown for poliovirus
in cell culture (Gitlin et al., 2002). However, based on the
high specificity of RNAi, an interesting potential applica-
tion could be the use of siRNAs directed to a specific region
of a gene, to select viral escape mutants that will contain
nucleotide changes in the specific targeted region. SiRNAs
could then be used as region-specific mutagens to generate
pre-designed mutants.
On the other hand, the inhibition of viral gene expres-
sion by siRNAs offers the potential for a novel therapeu-
tic approach for infections by rotaviruses and other viruses,
once efficient methods for in vivo delivery of siRNAs have
been developed. In addition to the standard management of
rotavirus infection symptoms, directed towards restoration
49
of fluid and electrolyte balance until the infection resolves
(symtomatic treatment), specific interventions aimed at in-
hibiting the viral replication could be designed. The intro-
duction of siRNAs preventing the replication of the virus
(for instance silencing genes involved in the replication of
the viral genome) could shorten the duration of the illness
and possibly help to prevent dehydration, the principal cause
of death of a rotavirus infection. Another attractive target
for antiviral therapy could be NSP4, since this nonstructural
protein, in addition to its important role during virus mor-
phogenesis, has been shown to function as a viral entero-
toxin that contributes to the diarrheal illness (Estes, 2003).
Although the use of RNAi as an antiviral therapeutic prin-
ciple and method still remains to be demonstrated, there is
no doubt that this technology will influence drastically the
way postgenomic viral and cellular research is conducted.
Acknowledgements
We thank Ulrich Desselberger for critical reading the
manuscript. The excellent technical assistance of Pedro
Romero is acknowledged. This work was partially sup-
ported by grants 55003662 and 55000613 from the Howard
Hughes Medical Institute, and by grant G37621N from the
National Council for Science and Technology—Mexico.
References
Adelman, Z.N., Sanchez-Vargas, I., Travanty, E.A., Carlson, J.O., Beaty,
B.J., Blair, C.D., Olson, K.E., 2002. RNA silencing of dengue virus
type 2 replication in transformed C6/36 mosquito cells transcribing
an inverted-repeat RNA derived from the virus genome. J. Virol. 76,
12925–12933.
Arias, C.F., Guerrero, C.A., Mendez, E., Zarate, S., Isa, P., Espinosa,
R., Romero, P., Lopez, S., 2001. Early events of rotavirus infection:
the search for the receptor(s). Novartis Found. Symp. 238, 47–60,
discussion 60–43.
Barik, S., 2004. Control of nonsegmented negative-strand RNA virus
replication by siRNA. Virus Res., this issue.
Bitko, V., Barik, S., 2001. Phenotypic silencing of cytoplasmic genes
using sequence-specific double-stranded short interfering RNA and its
application in the reverse genetics of wild type negative-strand RNA
viruses. BMC Microbiol. 1, 34.
Bridge, A.J., Pebernard, S., Ducraux, A., Nicoulaz, A.L., Iggo, R., 2003.
Induction of an interferon response by RNAi vectors in mammalian
cells. Nat. Genet. 34, 263–264.
Brummelkamp, T.R., Bernards, R., Agami, R., 2002. A system for stable
expression of short interfering RNAs in mammalian cells. Science
296, 550–553.
Campagna, M., Vascotto, F., Eichwald, C., Burrone, O.R., 2003. Interfering
with rotavirus NSP5. In: Proceedings of 8th International Symposium
on Double-Stranded RNA Viruses, Lucca, Italy.
Caplen, N.J., 2002. A new approach to the inhibition of gene expression.
Trends Biotechnol. 20, 49–51.
Capodici, J., Kariko, K., Weissman, D., 2002. Inhibition of HIV-1 infection
by small interfering RNA-mediated RNA interference. J. Immunol.
169, 5196–5201.
Chang, J., Taylor, J.M., 2003. Susceptibility of human hepatitis delta virus
RNAs to small interfering RNA action. J. Virol. 77, 9728–9731.
50 C.F. Arias et al. / Virus Research 102 (2004) 43–51
Coburn, G.A., Cullen, B.R., 2002. Potent and specific inhibition of human
immunodeficiency virus type 1 replication by RNA interference. J.
Virol. 76, 9225–9231.
Cuadras, M.A., Feigelstock, D.A., An, S., Greenberg, H.B., 2002. Gene
expression pattern in Caco-2 cells following rotavirus infection. J.
Virol. 76, 4467–4482.
Dector, M.A., Romero, P., Lopez, S., Arias, C.F., 2002. Rotavirus gene
silencing by small interfering RNAs. EMBO Rep. 3, 1175–1180.
Ding, S.W., Li, H., Lu, R., Li, F., Li, W.X., 2004. RNA silencing: a
conserved antiviral immunity of plants and animals. Virus Res., this
issue.
Elbashir, S.M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., Tuschl,
T., 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference
in cultured mammalian cells. Nature 411, 494–498.
Estes, M.K., 2001. Rotaviruses and their replication. In: Knipe, D.N.,
Howley, P.M. (Eds.), Virology, fourth ed. Lippincott Williams and
Wilkins, Philadelphia, PA, pp. 1747–1785.
Estes, M.K., 2003. The rotavirus NSP4 enterotoxin: current status and
challenges. In: Desselberger, U., Gray, J. (Eds.), Viral Gastroenteritis.
Elsevier Science, Amsterdam, pp. 207–224.
Flores-Jasso, C.F., Valdes, V.J., Sampieri, A., Valadez-Graham, V.,
Recillas-Targa, F., Vaca, L., 2004. Silencing structural and nonstruc-
tural genes in baculovirus by RNA interference. Virus Res., this issue.
Ge, Q., Eisen, H.N., Chen, J., 2004. Use of siRNAs to prevent and treat
influenza virus infection. Virus Res., this issue.
Ge, Q., McManus, M.T., Nguyen, T., Shen, C.H., Sharp, P.A., Eisen,
H.N., Chen, J., 2003. RNA interference of influenza virus production
by directly targeting mRNA for degradation and indirectly inhibiting
all viral RNA transcription. Proc. Natl. Acad. Sci. U.S.A. 100, 2718–
2723.
Gitlin, L., Andino, R., 2003. Nucleic acid-based immune system: the
antiviral potential of mammalian RNA silencing. J. Virol. 77, 7159–
7165.
Gitlin, L., Karelsky, S., Andino, R., 2002. Short interfering RNA confers
intracellular antiviral immunity in human cells. Nature 418, 430–434.
Graff, J.W., Mitzel, D.N., Weisend, C.M., Flenniken, M.L., Hardy, M.E.,
2002. Interferon regulatory factor 3 is a cellular partner of rotavirus
NSP1. J. Virol. 76, 9545–9550.
Graham, K.L., Halasz, P., Tan, Y., Hewish, M.J., Takada, Y., Mackow,
E.R., Robinson, M.K., Coulson, B.S., 2003. Integrin-using rotaviruses
bind a2b1 integrin a2 I domain via VP4 DGE sequence and recognize
aXb2 and aVb3 by using VP7 during cell entry. J. Virol. 77, 9969–
9978.
Hall, A.H., Alexander, K.A., 2003. RNA interference of human papil-
lomavirus type 18 E6 and E7 induces senescence in HeLa cells. J.
Virol. 77, 6066–6069.
Hamasaki, K., Nakao, K., Matsumoto, K., Ichikawa, T., Ishikawa, H.,
Eguchi, K., 2003. Short interfering RNA-directed inhibition of hepatitis
B virus replication. FEBS Lett. 543, 51–54.
Hannon, G.J., 2002. RNA interference. Nature 418, 244–251.
Hu, W.Y., Myers, C.P., Kilzer, J.M., Pfaff, S.L., Bushman, F.D., 2002.
Inhibition of retroviral pathogenesis by RNA interference. Curr. Biol.
12, 1301–1311.
Jacque, J.M., Triques, K., Stevenson, M., 2002. Modulation of HIV-1
replication by RNA interference. Nature 418, 435–438.
Jia, Q., Sun, R., 2003. Inhibition of gammaherpesvirus replication by
RNA interference. J. Virol. 77, 3301–3306.
Jiang, M., Milner, J., 2002. Selective silencing of viral gene expression
in HPV-positive human cervical carcinoma cells treated with siRNA,
a primer of RNA interference. Oncogene 21, 6041–6048.
Kapadia, S.B., Brideau-Andersen, A., Chisari, F.V., 2003. Interference of
hepatitis C virus RNA replication by short interfering RNAs. Proc.
Natl. Acad. Sci. U.S.A. 100, 2014–2018.
Katze, M.G., He, Y., Gale Jr, M., 2002. Viruses and interferon: a fight
for supremacy. Nat. Rev. Immunol. 2, 675–687.
Klein, C., Bock, C.T., Wedemeyer, H., Wustefeld, T., Locarnini, S., Dienes,
H.P., Kubicka, S., Manns, M.P., Trautwein, C., 2003. Inhibition of
hepatitis B virus replication in vivo by nucleoside analogues and
siRNA. Gastroenterology 125, 9–18.
Li, H., Li, W.X., Ding, S.W., 2002. Induction and suppression of RNA
silencing by an animal virus. Science 296, 1319–1321.
Lichner, Z., Silhavy, D., Burgyan, J., 2003. Double-stranded RNA-binding
proteins could suppress RNA interference-mediated antiviral defences.
J. Gen. Virol. 84, 975–980.
Lizano, M., Lopez, S., Arias, C.F., 1991. The amino-terminal half of
rotavirus SA114fM VP4 protein contains a hemagglutination domain
and primes for neutralizing antibodies to the virus. J. Virol. 65, 1383–
1391.
Martinez, M.A., Clotet, B., Este, J.A., 2002. RNA interference of HIV
replication. Trends Immunol. 23, 559–561.
McCaffrey, A.P., Meuse, L., Pham, T.T., Conklin, D.S., Hannon, G.J.,
Kay, M.A., 2002. RNA interference in adult mice. Nature 418, 38–
39.
McCaffrey, A.P., Nakai, H., Pandey, K., Huang, Z., Salazar, F.H., Xu, H.,
Wieland, S.F., Marion, P.L., Kay, M.A., 2003. Inhibition of hepatitis
B virus in mice by RNA interference. Nat. Biotechnol. 21, 639–
644.
Miyagishi, M., Matsumoto, S., Taira, K., 2004. Generation of an RNA
expression library against the human genome. Virus Res., this issue.
Novina, C.D., Murray, M.F., Dykxhoorn, D.M., Beresford, P.J., Riess, J.,
Lee, S.K., Collman, R.G., Lieberman, J., Shankar, P., Sharp, P.A.,
2002. siRNA-directed inhibition of HIV-1 infection. Nat. Med. 8,
681–686.
Parashar, U.D., Glass, R.I., 2003. Viral causes of gastroenteritis. In: Des-
selberger, U., Gray, J. (Eds.), Viral Gastroenteritis. Elsevier Science,
Amsterdam, pp. 9–22.
Patton, J.T., 1995. Structure and function of the rotavirus RNA-binding
proteins. J. Gen. Virol. 76 (Pt 11), 2633–2644.
Patton, J.T., 2003. Rotavirus genome replication: sites and signals. In:
Proceedings of 8th International Symposium on Double-Stranded RNA
Viruses, Lucca, Italy.
Patton, J.T., Spencer, E., 2000. Genome replication and packaging
of segmented double-stranded RNA viruses. Virology 277, 217–
225.
Pesavento, J.B., Estes, M.K., Prasad, B.V.V., 2003. Structural organization
of the genome in rotavirus. In: Desselberger, U., Gray, J. (Eds.),
Viral Gastroenteritis, first ed., vol. 9. Elsevier Science, Amsterdam,
pp. 115–128.
Piron, M., Vende, P., Cohen, J., Poncet, D., 1998. Rotavirus RNA-binding
protein NSP3 interacts with eIF4GI and evicts the poly(A) binding
protein from eIF4F. EMBO J. 17, 5811–5821.
Poruchynsky, M.S., Atkinson, P.H., 1991. Rotavirus protein rearrange-
ments in purified membrane-enveloped intermediate particles. J. Virol.
65, 4720–4727.
Randall, G., Grakoui, A., Rice, C.M., 2003. Clearance of replicating
hepatitis C virus replicon RNAs in cell culture by small interfering
RNAs. Proc. Natl. Acad. Sci. U.S.A. 100, 235–240.
Randall, G., Rice, C.M., 2004. Interfering with hepatitis C virus RNA
replication. Virus Res., this issue.
Roth, B.M., Pruss, G.J., Vance, V., 2004. Plant viral suppressors of RNA
silencing. Virus Res., this issue.
Saleh, M., Van Rij, R.P., Andino, R., 2004. RNA silencing in viral
infections: insights from poliovirus. Virus Res., this issue.
Sharp, P.A., 2001. RNA interference—2001. Genes Dev. 15, 485–490.
Shlomai, A., Shaul, Y., 2003. Inhibition of hepatitis B virus expres-
sion and replication by RNA interference. Hepatology 37, 764–
770.
Sledz, C.A., Holko, M., de Veer, M.J., Silverman, R.H., Williams, B.R.,
2003. Activation of the interferon system by short-interfering RNAs.
Nat. Cell Biol. 9, 834–839.
Tian, P., Estes, M.K., Hu, Y., Ball, J.M., Zeng, C.Q., Schilling, W.P.,
1995. The rotavirus nonstructural glycoprotein NSP4 mobilizes Ca2+
from the endoplasmic reticulum. J. Virol. 69, 5763–5772.
 C.F. Arias et al. / Virus Research 102 (2004) 43–51
Tian, P., Hu, Y., Schilling, W.P., Lindsay, D.A., Eiden, J., Estes, M.K.,
1994. The nonstructural glycoprotein of rotavirus affects intracellular
calcium levels. J. Virol. 68, 251–257.
Valdes, V.J., Sampieri, A., Sepulveda, J., Vaca, L., 2003. Using
double-stranded RNA to prevent in vitro and in vivo viral infections
by recombinant baculovirus. J. Biol. Chem. 278, 19317–19324.
Vance, V., Vaucheret, H., 2001. RNA silencing in plants—defense and
counterdefense. Science 292, 2277–2280.
Wilson, J.A., Jayasena, S., Khvorova, A., Sabatinos, S., Rodrigue-Gervais,
I.G., Arya, S., Sarangi, F., Harris-Brandts, M., Beaulieu, S., Richard-
son, C.D., 2003. RNA interference blocks gene expression and RNA
synthesis from hepatitis C replicons propagated in human liver cells.
Proc. Natl. Acad. Sci. U.S.A. 100, 2783–2788.
Xu, A., Bellamy, A.R., Taylor, J.A., 1998. BiP (GRP78) and endoplasmin
(GRP94) are induced following rotavirus infection and bind transiently
51
to an endoplasmic reticulum-localized virion component. J. Virol. 72,
9865–9872.
Zamore, P., 2001. RNA interference: listening to the sound silence. Nat.
Struct. Biol. 8, 746–750.
Zárate, S., Espinosa, R., Romero, P., Arias, C.F., López, S., 2002. Ro-
tavirus interacts with avb3 through a novel binding site. In: Proceed-
ings of Meeting of the Howard Hughes Medical Institute International
Research Scholars, Palm Cove, Australia.
Zhang, R., Guo, Z., Lu, J., Meng, J., Zhou, C., Zhan, X., Huang, B.,
Yu, X., Huang, M., Pan, X., Ling, W., Chen, X., Wan, Z., Zheng,
H., Yan, X., Wang, Y., Ran, Y., Liu, X., Ma, J., Wang, C., Zhang,
B., 2003. Inhibiting severe acute respiratory syndrome-associated
coronavirus by small interfering RNA. Chin. Med. J. 116, 1262–
1264.
 Virus Research 102 (2004) 53–58

Control of HIV-1 replication by RNA interference

Nan Sook Lee, John J. Rossi∗
Division of Molecular Biology, Beckman Research Institute of the City of Hope, Duarte, CA 91010, USA

Abstract

Small interfering RNAs (siRNAs) have been shown to direct sequence-specific inhibition of gene expression in mammalian cells. siRNAs
are RNA duplexes of 21–23 nucleotides (nts) with ∼2 nt 3 overhangs that can induce degradation of their homologous target mRNAs without
interferon responses in mammalian cells. The degradation of the target occurs at the post-transcriptional level, meaning a post-transcriptional
gene silencing (PTGS) mechanism called as RNA interference (RNAi). RNAi has emerged as an efficient method to inhibit gene expression
in mammalian cells with increasingly successful cases of knockdown of many specific genes. Recent works have shown that the use of
RNAi could inhibit HIV-1 replication by targeting viral or cellular genes. RNAi can be considered as a gene-specific therapeutic option for
controlling HIV-1 replication. However, the control of HIV-1 replication has become complex because of the limited effectiveness of existing
anti-HIV-1 agents and the high speed mutation rate of the HIV-1 genome. Careful assessments are required for the potential of RNAi as a
gene therapy approach for controlling HIV-1 replication. This review will discuss the status of the science using RNAi for controlling HIV-1
replication and will describe possible problems for therapeutic applications of RNAi-mediated technologies for HIV-1 behind this novel
mechanism.
© 2004 Elsevier B.V. All rights reserved.

Keywords: RNAi; siRNA; shRNA; HIV-1; PTGS

1. Introduction
Gene therapy has gained consideration as a possible treat-
ment for acquired immunodeficiency syndrome (AIDS), ei-
ther as an alternative or as an addition to anti-retroviral
chemotherapy. Several types of RNA gene therapies have
been developed and shown to inhibit HIV-1 replication in
mammalian cell cultures; these include antisense RNA, cat-
alytic RNA (ribozyme) and high-affinity RNA ligands (ap-
tamers or decoys). Several of these RNA-based approaches
have been safety tested in phase I clinical trials, although
there are no reports yet of efficacy trials. Nevertheless, in
vitro data suggest that their anti-viral efficiencies are vari-
able and can often be overcome by increasing the multi-
plicity of HIV infectious particles. The limitations have led
several investigators to explore the potential of RNA inter-
ference (RNAi) as an anti-HIV therapy.
RNAi is the process of sequence-specific, post-transcrip-
tional gene silencing (PTGS) in animals and plants triggered
by double-stranded RNA (dsRNA) that is homologous in se-
∗ Corresponding author. Tel.: +1-626-301-8360.
E-mail address: jrossi@coh.org (J.J. Rossi).
quence to the silenced gene (Elbashir et al., 2001; Fire et al.,
1998; Hutvagner and Zamore, 2002; Sharp, 2001; Zamore
et al., 2000). RNAi is initiated by the dsRNA-specific en-
donuclease Dicer that promotes processive cleavage of long
dsRNA into small 21–25 nucleotide (nt) small interfering
RNAs (siRNAs) (Bernstein et al., 2001; Hutvagner et al.,
2001; Ketting et al., 2001; Knight and Bass, 2001). The du-
plex siRNA is unwound, and one of the two strands is in-
corporated into an RNA-induced silencing complex (RISC)
(Bernstein et al., 2001; Hammond et al., 2001) guiding it
to a homologous target mRNA, which in turn is cleaved by
an endonuclease in RISC. This powerful sequence-specific
knockdown mechanism has rapidly gained wide acceptance
as a surrogate genetic tool and possible therapeutic modality
in mammalian cells.
It is known that dsRNA of ≥30 bp in size can trigger in-
terferon responses in mammalian cells that are intrinsically
sequence-nonspecific to the inducing dsRNA (Paddison
et al., 2002b). However, introduction of shorter siRNAs
(<30 bp) into mammalian cells leads to mRNA degrada-
tion with sequence specificity without activating the inter-
feron response (Elbashir et al., 2001). Synthetic duplexes
of 21-nucleotide siRNAs with protruding 3 termini can
mediate RNAi in a sequence-specific manner in cultured

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.015
54 N.S. Lee, J.J. Rossi / Virus Research 102 (2004) 53–58

mammalian cells. This siRNA technology takes advantage
of the fact that RNAi is a natural biological mechanism for
silencing genes in most eukaryotic organisms ranging from
fission yeast through plants and mammals (Agami, 2002;
Cottrell and Doering, 2003; Paddison and Hannon, 2002;
Scherr et al., 2003). RNAi functions as an antiviral mech-
anism in some organisms such as Drosophila and plants,
although the role of RNAi in controlling viral infection in
mammals is unknown. siRNA-mediated PTGS, therefore,
offers a potentially powerful tool for inhibiting HIV repli-
cation. This mechanism can be targeted to both viral and
cellular transcripts.
2. Targeting HIV-1 with RNAi
Introduction of siRNAs specific for HIV-1 into mam-
malian cells could lead to viral RNA degradation and in-
hibition of HIV-1 gene expression and replication during
different stages of the viral life cycle (Fig. 1). HIV-1 has
∼9 kbp genome that contains nine genes encoding 15 pro-
teins (Greene and Peterlin, 2002). The first potential tar-
get for siRNAs is the viral genomic RNA upon viral en-
try and uncoating. At least one report demonstrates that
siRNA-mediated destruction of incoming HIV-1 can take
place (Jacque et al., 2002) although other studies of RNAi

Fig. 1. HIV-1 life cycle and potential intervention by RNAi. HIV-1 utilizes the CD4 receptor and the chemokine receptors CCR5/and or CXCR4 for
entry. Each of these receptors is a potential RNAi target to block viral entry. In lieu of targeting the receptor or co-receptor mRNAs, incoming genomic
viral RNA is a potential target as well, as are all classes of viral transcripts. RISC is the RNAi induced silencing complex that interacts with one of
the siRNA strands and positions the strand for complementary base pairing with the target RNA. A component of RISC endonucleolytically cleaves the
target RNA allowing the RISC complex to recycle.
 N.S. Lee, J.J. Rossi / Virus Research 102 (2004) 53–58 55

inhibition of retroviral infection suggested that incoming ge-
nomic RNA may not be accessible to siRNAs (Hu et al.,
2002b; Waterhouse et al., 2001). Once the viral genomic
DNA has been integrated, the viral mRNA transcripts as
well as the unspliced genomic length RNA are targets for
RNAi. Early transcripts such as HIV-1 rev and tat are good
targets for siRNAs since the Tat and Rev proteins encoded
by these RNAs are essential for subsequent expression of
HIV-1 structural genes (Gag, Pol, and Env) and for the syn-
thesis of full length viral genomic RNA.
In addition to viral targets, the cellular chemokine recep-
tors CCR5 and CXCR4, which function as co-receptors for
HIV-1, have provided new therapeutic targets and a better
understanding of the progression of viral infection. CCR5,
an HIV-1 co-receptor for M-tropic HIV-1 provides an attrac-
tive cellular target for siRNAs since homozygous deletions
of CCR5 effectively confer protection from HIV-1 without
any serious deleterious effects in immune function (Samson
et al., 1996). At least one group has taken advantage of this
target for RNAi mediated gene silencing demonstrating in
vitro knockdown of CCR5 by siRNAs provided marked pro-
tection from HIV-1 infection (Martinez et al., 2002).
3. Exogenous delivery versus endogenous expression of
si/shRNAs
In mammalian systems, the sequence-specific anti-HIV-1
effects have been observed following introduction of syn-
thetic siRNAs (ssiRNAs) via transfection into HIV-1 in-
fectable cells (Coburn and Cullen, 2002; Gitlin et al., 2002;
Jacque et al., 2002; Novina et al., 2002), or via endogenous
expression of 21–23 nt transcripts (psiRNAs) or hairpin
RNAs (pshRNAs) from DNA plasmids (Jacque et al., 2002;
Lee et al., 2002a) (Fig. 2). The expression of siRNAs
from psiRNAs appears to bypass the need for Dicer or
other endonucleases to produce siRNAs (Lee et al., 2002a;
Miyagishi and Taira, 2002). The same is true for ssiR-
NAs, but these must be used in multiple transfections to
maintain silencing. The psiRNA system can potentially be
used in differentiated mammalian cells containing no or
low amounts of Dicer (Kitabwalla and Ruprecht, 2002).
However, shRNAs with different stem-loop structures ex-
pressed from pshRNAs in mammalian cells require pro-
cessing to remove the loop. Depending upon the length
of the duplex, these may utilize Dicer, but perhaps one or
more other endonucleases are involved in this processing
step. (Brummelkamp et al., 2002; Parrish et al., 2000; Paul
et al., 2002; Sui et al., 2002; Yu et al., 2002; Zeng et al.,
2002). The extent of siRNA-directed silencing from both
psiRNAs and pshRNAs may be limited by the concentration
of active siRNAs in the target cell. The stem-loop struc-
tures of shRNAs mimic the naturally occurring stem-loop
structures found in micro-RNA precursors which are pro-
cessed by Dicer to yield small temporal RNAs (stRNA) or
microRNA (miRNAs) (Lee et al., 2002b). Processing of
anti-HIV siRNAs from a miRNA precursor has recently
been demonstrated (Zeng and Cullen, 2002).
Inhibition of HIV-1 infection using ssiRNAs, or siRNAs
produced from transcription of psiDNAs and pshDNAs has
served as a proof-of-principle that siRNA technology can be
a possible therapeutic approach for the inhibition of HIV-1
replication in hematopoietic cells (Coburn and Cullen, 2002;
Hu et al., 2002a; Jacque et al., 2002; Lee et al., 2002a;
Novina et al., 2002). We have demonstrated that a psiRNA
approach can be used to inhibit expression of HIV-1 rev
and/or tat transcript (s) in both transient transfections (Lee
et al., 2002a) or from lentiviral transduced hematopoietic

Fig. 2. Gene expression systems for intracellular transcription of siRNAs or shRNAs. For purposes of simplicity, only the Pol III promoted transcripts
are depicted, but Pol II systems can be used as well provided that a polyA signal is included downstream of the shRNA genes. The Pol III transcripts
terminate within a string of 5–6 uridines providing a defined 3 end for the transcripts. When sense and antisense strands are transcribed from separate
promoters, these strands must anneal to form an siRNA. In contrast, linked shRNA strands readily form a duplex, but the loop joining these must be
processed to generate siRNAs.
56 N.S. Lee, J.J. Rossi / Virus Research 102 (2004) 53–58
progenitor cells (Banerjea et al., 2003). In this approach,
the vectors contain two tandem human U6 snRNA promot-
ers followed by 21-mers encoding sense and antisense siR-
NAs. In the co-transfection experiments, psiRNAs cotrans-
fected with the HIV-1 pNL4-3 proviral DNA led to up to
4 logs of inhibition of HIV-1 p24 antigen expression (Lee
et al., 2002a). The high degree of inhibition was achieved
by simultaneously targeting two essential sites (rev and tat).
It was also demonstrated that synthetic siRNAs targeted to
HIV-1 rev and tat mRNAs inhibit HIV-1 gene expression
and replication in human T-cell lines and primary lympho-
cytes (Coburn and Cullen, 2002). Novina et al. (2002) re-
ported that silencing CD4 expression inhibited viral entry,
syncytia formation and reduced free viral titers, although
CD4 may not be a good therapeutic target because of the
receptor’s critical role in T-cell function.
Gitlin et al. (2002) reported that siRNA against poliovirus
provided intracellular anti-viral immunity to human cells,
promoting viral clearance and cell survival. They suggested
that gene silencing by siRNAs functions as an adaptive,
RNA-based defense mechanism in mammalian cells. Jacque
et al. (2002) reported similar results in their studies with
HIV-1. They utilized intracellular T7 transcription from a
plasmid DNA template to produce siRNAs in the cyto-
plasm of mammalian cells. Utilizing this system to produce
shRNAs targeting HIV-1 vif they were able to confer intra-
cellular immunity to HIV-1 in cell lines as well as primary
lymphcytes, in part by blocking reverse transcription of ge-
nomic RNA into proviral cDNA. Capodici et al. (2002) used
synthetic and in vitro transcribed siRNAs and obtained sim-
ilar suppression of HIV-1 specific proviral DNA formation
and replication in cell lines and primary, activated CD4+ T
lymphocytes. In addition to demonstrating that siRNA can
inhibit viral infection at pre- and post-integration stages in
mammalian cells, they also showed that primary CD4+ T
cells could be targeted by siRNAs using lipofectin, or by
non-lipid complexed fluorine-derivatized siRNAs.
Although transfected siRNAs can effectively inhibit viral
replication, the effects are largely transient (Novina et al.,
2002). Therefore, even if effective in vivo delivery is ob-
tained, the siRNAs will have to be infused on a regular
basis to treat chronic diseases such as AIDS. The use of
engineered DNA vectors capable of long term production
of siRNAs or shRNAs has facilitated the practical applica-
tion of RNAi in human primary cells (Brummelkamp et al.,
2002; Lee et al., 2002a; Miyagishi and Taira, 2002;
Paddison et al., 2002a; Paul et al., 2002; Xia et al., 2002).
The demonstrations that RNAi could be activated by ex-
pressing siRNAs or shRNAs from eukaryotic promoters has
prompted the inclusion of these expression systems into vi-
ral vectors that can be used to transduce cells with siRNA or
shRNA genes. For treatment of HIV-1 infection, lentiviral
vector-based transduction of siRNA and shRNA encoding
genes into hematopoietic cells is particularly attractive from
the experimental point of view. Lentiviruses are effective
for delivery of siRNAs into mammalian cells because they
can effectively transduce both dividing and non-dividing
cells and provide long-term gene expression (Abbas-Terki
et al., 2002; Matta et al., 2003; Qin et al., 2003; Rubinson
et al., 2003; Tiscornia et al., 2003). Anti-CCR5-shRNA
genes inserted in a lentiviral vector and introduced into hu-
man peripheral blood T lymphocytes (PBLs) resulted in a
10-fold reduction of CCR5 in PBLs and a five-fold reduc-
tion of HIV-1 replication. This protection was long term
and sequence-specific (Qin et al., 2003). It has also been
demonstrated that anti-Rev siRNA genes expressed from
lentiviral vector-transduced hematopoietic precursor cells
provided in vitro derived macrophages and SCID-hu mouse
differentiated T-lymphocytes with long term protection
from HIV-1 infection and replication (Banerjea et al., 2003).
It should be pointed out that long term RNAi effected by
ssiRNAs may also be possible in certain cell types. For in-
stance non-dividing macrophages were shown to be resistant
to HIV-1 challenge for over two weeks following treatment
with ssiRNAs targeting CCR5(Song et al., 2003). Interest-
ingly, these same investigators demonstrated that the RNAi
effect is more long-lived if there is a target RNA present
since RNAi was lost in cells that were not continually chal-
lenged with viral gene expression.
4. Possible problems for therapeutic applications
While the results obtained to date should be considered
preliminary in terms of application to humans, they do pro-
vide strong justification for further investigating the use of
RNAi for treatment of HIV-1 in a clinical setting. It has
largely been assumed that siRNAs appear to be under the
radar of the host interference response mechanism. It has
recently been demonstrated that some expressed shRNAs
can activate at least one of the arms of the human inter-
feron response mechanism (Bridge et al., 2003). Since many
commonly used tumor cells cannot respond to interferon be-
cause of specific mutations of genes in the interferon path-
way (Stojdl et al., 2000), careful studies with primary cells
need to be carried out to ensure that potentially serious side
effects due to interferon induction are not occurring.
An additional concern is that the success of RNAi against
a viral pathogen is not automatically assured. One impor-
tant factor is that not all viral mRNA sequences are equally
accessible to siRNAs. Upon HIV-1 infection, genomic viral
RNA is introduced into the host cell cytoplasm in the form
of a nucleoprotein complex that can obscure the recogni-
tion by siRNAs. Some viral RNA sequences might be buried
within secondary structures or within highly folded regions.
It may often be necessary to test several different target sites
before a potent RNAi/target combination can be identified.
For HIV-1 there is the compounding problem of rapid muta-
tion, rendering the virus resistant to therapeutic agents. Us-
ing RNAi agents directed against multiple conserved RNA
target sequences in combination with targeting host factors
may overcome this problem. Another concern is maintaining
 N.S. Lee, J.J. Rossi / Virus Research 102 (2004) 53–58
siRNA expression long term in stably transduced primary
cells. It is currently unknown whether or not silencing of
Pol III-promoted siRNA and shRNA transcription will be a
problem. Finally, since RNAi is a native cellular mechanism
that most likely has an important, if not essential function
in most cell types, there is the concern that usurping the
RNAi machinery with foreign siRNAs could have adverse
developmental and functional consequences. Only by car-
rying out the appropriate experiments in relevant cells and
animal models can the potential problems be addressed.
Acknowledgements
This work was supported by NIH grants AI 29329,
AI42552, and HL074704 to JJR and a GlaxoSmith Kline
(GSK) grant to NSL.
References
Abbas-Terki, T., Blanco-Bose, W., Deglon, N., Pralong, W., Aebischer,
P., 2002. Lentiviral-mediated RNA interference. Hum. Gene. Ther. 13,
2197–2201.
Agami, R., 2002. RNAi and related mechanisms and their potential use
for therapy. Curr. Opin. Chem. Biol. 6, 829–834.
Banerjea, A., Li, M.J., Bauer, G., Remling, L., Lee, N.S., Rossi, J., Akkina,
R., 2003. Inhibition of HIV-1 by lentiviral vector-transduced siRNAs in
T lymphocytes differentiated in SCID-hu mice and CD34+ progenitor
cell-derived macrophages. Mol. Ther. 8, 62–71.
Bernstein, E., Caudy, A.A., Hammond, S.M., Hannon, G.J., 2001. Role
for a bidentate ribonuclease in the initiation step of RNA interference.
Nature 409, 363–366.
Bridge, A.J., Pebernard, S., Ducraux, A., Nicoulaz, A.L., Iggo, R., 2003.
Induction of an interferon response by RNAi vectors in mammalian
cells. Nat. Genet. 34, 263–264.
Brummelkamp, T.R., Bernards, R., Agami, R., 2002. A system for stable
expression of short interfering RNAs in mammalian cells. Science
296, 550–553.
Capodici, J., Kariko, K., Weissman, D., 2002. Inhibition of HIV-1 infection
by small interfering RNA-mediated RNA interference. J. Immunol.
169, 5196–5201.
Coburn, G.A., Cullen, B.R., 2002. Potent and specific inhibition of human
immunodeficiency virus type 1 replication by RNA interference. J.
Virol. 76, 9225–9231.
Cottrell, T.R., Doering, T.L., 2003. Silence of the strands: RNA interfer-
ence in eukaryotic pathogens. Trends Microbiol. 11, 37–43.
Elbashir, S.M., Lendeckel, W., Tuschl, T., 2001. RNA interference is
mediated by 21- and 22-nucleotide RNAs. Genes Dev. 15, 188–200.
Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver, S.E., Mello,
C.C., 1998. Potent and specific genetic interference by double-stranded
RNA in Caenorhabditis elegans. Nature 391, 806–811.
Gitlin, L., Karelsky, S., Andino, R., 2002. Short interfering RNA confers
intracellular antiviral immunity in human cells. Nature 418, 430–434.
Greene, W.C., Peterlin, B.M., 2002. Charting HIV’s remarkable voyage
through the cell: Basic science as a passport to future therapy. Nat.
Med. 8, 673–680.
Hammond, S.M., Boettcher, S., Caudy, A.A., Kobayashi, R., Hannon, G.J.,
2001. Argonaute2, a link between genetic and biochemical analyses
of RNAi. Science 293, 1146–1150.
Hu, W., Myers, C., Kilzer, J., Pfaff, S., Bushman, F., 2002a. Inhibition
of retroviral pathogenesis by RNA interference. Curr. Biol. 12, 1301.
57
Hu, W.Y., Myers, C.P., Kilzer, J.M., Pfaff, S.L., Bushman, F.D., 2002b.
Inhibition of retroviral pathogenesis by RNA interference. Curr. Biol.
12, 1301–1311.
Hutvagner, G., McLachlan, J., Pasquinelli, A.E., Balint, E., Tuschl, T.,
Zamore, P.D., 2001. A cellular function for the RNA-interference
enzyme Dicer in the maturation of the let-7 small temporal RNA.
Science 293, 834–838.
Hutvagner, G., Zamore, P.D., 2002. RNAi: nature abhors a double-strand.
Curr. Opin. Genet. Dev. 12, 225–232.
Jacque, J.M., Triques, K., Stevenson, M., 2002. Modulation of HIV-1
replication by RNA interference. Nature 418, 435–438.
Ketting, R.F., Fischer, S.E., Bernstein, E., Sijen, T., Hannon, G.J., Plasterk,
R.H., 2001. Dicer functions in RNA interference and in synthesis of
small RNA involved in developmental timing in C. elegans. Genes
Dev. 15, 2654–2659.
Kitabwalla, M., Ruprecht, R.M., 2002. RNA interference—a new weapon
against HIV and beyond. N. Engl. J. Med. 347, 1364–1367.
Knight, S.W., Bass, B.L., 2001. A role for the RNase III enzyme DCR-1
in RNA interference and germ line development in Caenorhabditis
elegans. Science 293, 2269–2271.
Lee, N.S., Dohjima, T., Bauer, G., Li, H., Li, M.J., Ehsani, A., Salvaterra,
P., Rossi, J., 2002a. Expression of small interfering RNAs targeted
against HIV-1 rev transcripts in human cells. Nat. Biotechnol. 20,
500–505.
Lee, Y., Jeon, K., Lee, J.T., Kim, S., Kim, V.N., 2002b. MicroRNA
maturation: stepwise processing and subcellular localization. EMBO
J. 21, 4663–4670.
Martinez, M.A., Gutierrez, A., Armand-Ugon, M., Blanco, J., Parera, M.,
Gomez, J., Clotet, B., Este, J.A., 2002. Suppression of chemokine
receptor expression by RNA interference allows for inhibition of
HIV-1 replication. AIDS 16, 2385–2390.
Matta, H., Hozayev, B., Tomar, R., Chugh, P., Chaudhary, P.M., 2003.
Use of lentiviral vectors for delivery of small interfering RNA. Cancer
Biol. Ther. 2, 206–210.
Miyagishi, M., Taira, K., 2002. U6 promoter-driven siRNAs with four
uridine 3 overhangs efficiently suppress targeted gene expression in
mammalian cells. Nat. Biotechnol. 20, 497–500.
Novina, C.D., Murray, M.F., Dykxhoorn, D.M., Beresford, P.J., Riess, J.,
Lee, S.K., Collman, R.G., Lieberman, J., Shankar, P., Sharp, P.A.,
2002. siRNA-directed inhibition of HIV-1 infection. Nat. Med. 8,
681–686.
Paddison, P.J., Caudy, A.A., Bernstein, E., Hannon, G.J., Conklin, D.S.,
2002a. Short hairpin RNAs (shRNAs) induce sequence-specific silenc-
ing in mammalian cells. Genes Dev. 16, 948–958.
Paddison, P.J., Caudy, A.A., Hannon, G.J., 2002b. Stable suppression of
gene expression by RNAi in mammalian cells. Proc. Natl. Acad. Sci.
U.S.A. 99, 1443–1448.
Paddison, P.J., Hannon, G.J., 2002. RNA interference: the new somatic
cell genetics. Cancer Cell 2, 17–23.
Parrish, S., Fleenor, J., Xu, S., Mello, C., Fire, A., 2000. Functional
anatomy of a dsRNA trigger: differential requirement for the two
trigger strands in RNA interference. Mol. Cell 6, 1077–1087.
Paul, C.P., Good, P.D., Winer, I., Engelke, D.R., 2002. Effective expression
of small interfering RNA in human cells. Nat. Biotechnol. 20, 505–
508.
Qin, X.F., An, D.S., Chen, I.S., Baltimore, D., 2003. Inhibiting HIV-1
infection in human T cells by lentiviral-mediated delivery of small
interfering RNA against CCR5. Proc. Natl. Acad. Sci. U.S.A. 100,
183–188.
Rubinson, D.A., Dillon, C.P., Kwiatkowski, A.V., Sievers, C., Yang, L.,
Kopinja, J., Rooney, D.L., Ihrig, M.M., McManus, M.T., Gertler,
F.B., Scott, M.L., Van Parijs, L., 2003. A lentivirus-based system to
functionally silence genes in primary mammalian cells, stem cells and
transgenic mice by RNA interference. Nat. Genet. 33, 401–406.
Samson, M., Libert, F., Doranz, B.J., Rucker, J., Liesnard, C., Farber,
C.M., Saragosti, S., Lapoumeroulie, C., Cognaux, J., Forceille, C.,
Muyldermans, G., Verhofstede, C., Burtonboy, G., Georges, M., Imai,
58 N.S. Lee, J.J. Rossi / Virus Research 102 (2004) 53–58
T., Rana, S., Yi, Y., Smyth, R.J., Collman, R.G., Doms, R.W., Vassart,
G., Parmentier, M., 1996. Resistance to HIV-1 infection in caucasian
individuals bearing mutant alleles of the CCR-5 chemokine receptor
gene. Nature 382, 722–725.
Scherr, M., Morgan, M.A., Eder, M., 2003. Gene silencing mediated by
small interfering RNAs in mammalian cells. Curr. Med. Chem. 10,
245–256.
Sharp, P.A., 2001. RNA interference—2001. Genes Dev. 15, 485–
490.
Song, E., Lee, S.K., Dykxhoorn, D.M., Novina, C., Zhang, D., Crawford,
K., Cerny, J., Sharp, P.A., Lieberman, J., Manjunath, N., Shankar, P.,
2003. Sustained small interfering RNA-mediated human immunode-
ficiency virus type 1 inhibition in primary macrophages. J. Virol. 77,
7174–7181.
Stojdl, D.F., Lichty, B., Knowles, S., Marius, R., Atkins, H., Sonenberg,
N., Bell, J.C., 2000. Exploiting tumor-specific defects in the interferon
pathway with a previously unknown oncolytic virus. Nat. Med. 6,
821–825.
Sui, G., Soohoo, C., Affar el, B., Gay, F., Shi, Y., Forrester, W.C.,
2002. A DNA vector-based RNAi technology to suppress gene ex-
pression in mammalian cells. Proc. Natl. Acad. Sci. U.S.A. 99, 5515–
5520.
Tiscornia, G., Singer, O., Ikawa, M., Verma, I.M., 2003. A general method
for gene knockdown in mice by using lentiviral vectors expressing
small interfering RNA. Proc. Natl. Acad. Sci. U.S.A. 100, 1844–1848.
Waterhouse, P.M., Wang, M.B., Lough, T., 2001. Gene silencing as an
adaptive defence against viruses. Nature 411, 834–842.
Xia, H., Mao, Q., Paulson, H.L., Davidson, B.L., 2002. siRNA-mediated
gene silencing in vitro and in vivo. Nat. Biotechnol. 20, 1006–1010.
Yu, J.Y., DeRuiter, S.L., Turner, D.L., 2002. RNA interference by ex-
pression of short-interfering RNAs and hairpin RNAs in mammalian
cells. Proc. Natl. Acad. Sci. U.S.A. 99, 6047–6052.
Zamore, P.D., Tuschl, T., Sharp, P.A., Bartel, D.P., 2000. RNAi:
double-stranded RNA directs the ATP-dependent cleavage of mRNA
at 21 to 23 nucleotide intervals. Cell 101, 25–33.
Zeng, Y., Cullen, B.R., 2002. RNA interference in human cells is restricted
to the cytoplasm. RNA 8, 855–860.
Zeng, Y., Wagner, E.J., Cullen, B.R., 2002. Both natural and designed
micro RNAs can inhibit the expression of cognate mRNAs when
expressed in human cells. Mol. Cell 9, 1327–1333.
 Virus Research 102 (2004) 59–64

RNA interference against retroviruses

Wen-Yuan Hu, Frederic D. Bushman∗ , Amara C. Siva
Infectious Disease Laboratory, The Salk Institute, 10010 North Torrey Pines Rd., La Jolla, CA 92037, USA

Abstract

Bang and Ellerman, and later Peyton Rous, reported the first identification of transmissible cancer-causing agents, which later turned out to
be avian retroviruses. Today avian retroviruses are important models for study of retrovirus replication and pathogenesis, and also important
pathogens of domestic fowl. Here we describe the use of RNA interference (RNAi) in live chick embryos to block replication of an avian
retrovirus. We also describe inhibition of ASLV and HIV replication in cell culture with RNAi.
© 2004 Elsevier B.V. All rights reserved.

Keywords: Retrovirus; RNA interference; siRNA; ASLV; HIV

1. Introduction natively, RNAi can also be induced by direct transfection of

1.1. RNAi background
RNA interference (RNAi) is a process in which
double-stranded RNA (dsRNA) triggers the silencing of
gene expression in a sequence-specific manner. RNAi has
been observed in organisms as diverse as plants, protozoa,
nematodes, fungi, insects, and mammals (Caplen et al.,
2000, 2001; Cogoni and Macino, 1999; Elbashir et al.,
2001b; Fire et al., 1998; Hamilton and Baulcombe, 1999;
Hammond et al., 2000; Jorgensen, 1990; Ngo et al., 1998;
Romano and Macino, 1992). Some of the components of
the RNAi machinery are also involved in the processing
and function of microRNAs (miRNAs; also called small
temporal RNAs), a class of small endogenous noncoding
RNA molecules that carry out developmental gene control
via translational repression (Doench et al., 2003; Palatnik
et al., 2003; Zeng et al., 2000, 2003).
RNAi can be induced by introduction of dsRNA into
cells in a variety of ways. These include endogenous syn-
thesis of complementary RNA strands, by infection with
viruses that produce dsRNA, and injection or transfection
of dsRNAs. Long dsRNA are processed to small interfering
RNAs (siRNAs) of about 21–23 nt by the RNase III en-
zyme Dicer (for recent reviews of the RNAi mechanism, see
Dykxhoorn et al., 2003; Pickford and Cogoni, 2003). Alter-
∗ Corresponding author.
E-mail address: bushman@salk.edu (F.D. Bushman).
siRNAs, short RNAs mimicking the product of Dicer pro-
cessing.
The siRNAs are incorporated into a multicomponent RNA
induced silencing complex (RISC), and act as guides for the
targeting of mRNAs for cleavage and degradation (Martinez
et al., 2002a; Schwarz et al., 2002). The identities of all
RISC components have not been characterized to date. At
least one of the members of the Argonaute gene family is a
component of this complex (for review, see Carmell et al.,
2002), and a micrococcal nuclease family member has been
also been identified that may be the enzyme responsible for
mRNA degradation (Caudy et al., 2003).
In invertebrates, the RNA silencing machinery appears to
be able to amplify the degradation signal. A strand from an
siRNA bound to the target mRNA can apparently function
as a primer that extends a complementary strand along the
targeted mRNA by an RNA-dependent RNA polymerase
(RdRP). Further processing of the dsRNA generated results
in generation of new siRNAs (Lipardi et al., 2001; Plasterk,
2002; Sijen et al., 2001). However, such amplification has
not been observed in mammalian cells to date.
1.2. RNAi in vertebrate cells
Prior to 2001, there appeared to be little chance that
RNAi could be employed to inhibit the replication of ver-
tebrate viruses. Vertebrate cells were known to respond to
double-stranded RNA by inducing the interferon response.
Cells respond to interferon combined with a double-stranded
RNA inducer by shutting down translation and other

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.016
60 W.-Y. Hu et al. / Virus Research 102 (2004) 59–64

responses designed to block viral replication and so block
spread in the organism. However, key findings in the labo-
ratories of Tuschl and Fire revealed that the length of the
double-stranded RNA determined the nature of the response
(Caplen et al., 2001; Elbashir et al., 2001a). Long dsRNA
induced the interferon response, while short dsRNAs in-
duced RNAi. This opened the door to studies of RNAi
against vertebrate viruses.
RNAi has been found to act as an antiviral response in
plants. Plants inhibited for the RNAi response show exac-
erbated symptoms upon infection with some viruses. Fur-
thermore, several plant viruses have been shown to encode
proteins that antagonize the RNAi system, providing strong
evidence that RNAi is part of the normal plant response to
infection (Al-Kaff et al., 1998; Marathe et al., 2000; Ratcliff
et al., 1999). Recently a protein inhibiting RNAi has been
detected in Flock House Virus (FHV), a nodavirus that in-
fects insects, implicating RNAi in defense against viruses in
insects as well (Li et al., 2002). The FHV B2 protein, the
inhibitor of RNAi, has been found to block RNA silencing
in both plant and invertebrate cells, and can functionally re-
place the suppressor of RNA silencing 2b protein of cucum-
ber mosaic virus (Li et al., 2002).
Is RNAi a component of the normal vertebrate response
to viral infection? This question is still unanswered. We and
others have investigated this question by modeling inhibition
of viral replication by RNAi using retroviral models. This
review focuses on studies of avian retroviruses, for which we
have an in vivo model. Many publications have also reported
studies of inhibition of HIV. Below we first summarize the
HIV work (see also Bushman, 2003, and the article by Lee
and Rossi in this volume), then studies of RNAi against
avian retroviruses. These studies show that RNAi can inhibit
viral replication in vivo and in cell culture, but leave the
question of its normal role unanswered. Regardless of its
normal function, it seems likely that RNAi can be adapted
to mitigate viral pathogenesis in vertebrates.
2. RNA interference against HIV in cell culture
Inhibition of HIV-1 replication in cell culture was first re-
ported by Garrus et al. (2001). In this study, Tsg101, a pro-
tein involved in vacuolar protein sorting, was shown to be
required for HIV-1 budding through the targeting of Tsg101
mRNA by RNAi (Garrus et al., 2001). This beautiful in-
vestigation launched an explosion of studies using RNAi to
inhibit HIV replication (summarized in Table 1).
Many positions in the HIV genomic RNA have been tar-
geted by siRNAs, including the gag and pol regions (encod-
ing viral structural proteins), and the genes for the regula-
tory proteins tat and rev. Cellular factors required for viral
replication in addition to the Tsg101 coding region have also
been targeted, specifically the HIV receptor CD4 and the
coreceptors CXCR4 and CCR5 (references in Table 1).
Several methods were used to initiate RNAi in these stud-
ies. Some of these techniques in fact were first used in the
HIV model system, and each suggests a possible approach
for using RNAi to treat HIV infection. Transfection of cells
with chemically synthesized siRNA was first used to de-
liver RNAi, but disadvantages of this method include that
the RNAi effect is transient, and many primary cells are not

Table 1
Studies reporting inhibition of retroviral replication by RNAi
Target gene Means of inducing RNAi Cell line or tissue Reference

Viral gene
HIV-1 Tat, Rev siRNA transfection, siRNA electroporation 293T, Jurkat, PBMC Coburn and Cullen (2002)
HIV-1 Vif siRNA transfection Magi cells Jacque et al. (2002)
Nef, LTR TAR T7-shRNA (vector) transfection PBL
HIV-1 Gag, Pol siRNA transfection HOS Hu et al. (2002)
HIV-1 Gag, LTR siRNA (synthetic and transcribed from 293T, U87, PBMC Capodici et al. (2002)
T7 in vitro) transfection
HIV-1 Rev, Tat siRNA transfection, tandem U6-siRNA 293/EcR Lee et al. (2002)
(vector) transfection
HIV-1 Rev Tandem U6-siRNA (vector) transfection CD34+ hematopoetic Banerjea et al. (2003)
progenitor cells
HIV-1 Gag, Env Long dsRNA transfection COS, HeLa, PBMC Park et al. (2002)
HIV-1 Tat, RT, human siRNA transfection Magi cells, Jurkat cells Surabhi and Gaynor (2002)
NF-␬B p65 subunit
HIV-1 Gag, human CD4 siRNA transfection Magi cells Novina et al. (2002)
ASLV Gag siRNA transfection, siRNA in ovo DF1, chicken embryo Hu et al. (2002)
electroporation
Host genes
Human TSG101 siRNA transfection 293T Garrus et al. (2001)
Human CCR5 U6-shRNA on lentiviral vector transduction PBMC Li et al. (2003)
Human CCR5 U6-shRNA on lentiviral vector transduction Magi cells, PBL Qin et al. (2003)
Human CXCR4, CCR5 siRNA transfection U87, HeLa Martinez et al. (2002b)
 W.-Y. Hu et al. / Virus Research 102 (2004) 59–64
readily transfected. RNAi can also be delivered via a DNA
vector encoding a short hairpin RNA expressed under the
control of an RNA Pol III promoter (e.g. H1 and U6) or RNA
Pol II promoter (e.g. CMV). Use of viral vectors has also
been reported, including vectors based on adeno-associated
virus, retroviruses, and lentiviruses. These three viruses have
the advantage of integrating the RNAi-initiating construct
in the host genome, thereby allowing inhibition to persist
during cell division.
3. RNA interference against avian retroviruses in cell
culture
3.1. ASLV history
Ellerman and Bang (1908) demonstrated that a filterable
agent was responsible for causing chicken leukosis, a type
of leukemia/lymphoma. The cell-free transmission of solid
tumors in chickens was reported by Rous (1911) at the Rock-
efeller Institute in New York. It was subsequently observed
that inoculation of RSV onto the chorioallantonic membrane
of the chick embryo resulted in the appearance of individual
small tumors whose numbers could be correlated with vi-
ral concentrations (Keogh, 1938). Temin and Rubin (1958)
demonstrated the oncogenic transformation of chick embryo
fibroblasts in culture by Rous sarcoma virus (RSV), and that
single viral particles induced discrete foci of transformed
cells. Temin later used an avian retrovirus in his demon-
stration that retroviral particles contain reverse transcriptase
activity (Temin and Mizutani, 1970). Studies of the RSV
also led to the identification of cellular homologs of the
oncogenes transduced by the acute transforming retroviruses
(Bishop, 1991). Today avian retroviruses are important mod-
els for retroviral replication and significant pathogens affect-
ing the poultry industry.
3.2. RNAi against ASLV replication
We have used avian retroviruses as a model to study
RNAi against retroviral replication. In these studies, siRNAs
matching two sequences in the gag gene of Avian Sarcoma
Leucosis Virus (ASLV) were introduced into cells by trans-
fection, one matching the p19 matrix (MA) coding region
and the other matching the p12 nucleocapsid (NC) coding
region. As a model, cultured chicken DF-1 cells were stud-
ied, and the RCAS version of ASLV was used. RCAS is
a replication-competent retroviral vector derived from RSV
by removal of the src oncogene so as to allow introduc-
tion of new genetic information in its place (Federspiel and
Hughes, 1997; Hughes et al., 1987).
RCAS virus was introduced either by infecting DF-1
cells or by transfecting them with RCAS proviral DNA.
The siRNAs were introduced by cotransfection concur-
rently. In the presence of specific siRNA, the production of
the RCAS virus in culture supernatants was dramatically
61
reduced (90%) 2 days after transfection as monitored by
accumulation of the p19 matrix protein (Hu et al., 2002).
Analysis of provirus formation in a spreading infection
also revealed that the two siGAG RNAs inhibited RCAS
replication. As controls, cultures were treated with siRNAs
against irrelevant proteins or mock transfected, and these
showed no significant inhibition. These data document that
RNAi can efficiently inhibit RCAS replication.
3.3. Inhibition early or late?
There are two parts of the retroviral life cycle that could
potentially be targeted by RNAi. The incoming viral RNA
might be degraded by RNAi prior to completing reverse
transcription, or the late viral mRNA and genomic RNA
transcribed from the integrated proviral DNA might also be
cleaved. Several groups have investigated the viral RNAs
attacked by RNAi, with somewhat different pictures emerg-
ing.
We have analyzed this issue for both ASLV and HIV. To
model late expression from integrated proviruses, we trans-
fected DNA copies of the ASLV or HIV genomes along with
siRNAs against the retroviral genome or with control siR-
NAs. Inhibition was found to be efficient for both viruses.
Thus RNAi worked efficiently against the viral mRNAs and
genomic RNAs expressed from a DNA copy, modeling ac-
tivity against the late viral RNAs produced by transcription.
What about RNAi against the viral RNA genome intro-
duced into cells in the early steps of infection, i.e. initial fu-
sion of the viral and cellular membranes? To study this step,
cells were infected with ASLV or HIV and the production
of early reverse transcription products measured. Because
the viral RNA genome serves as the template for DNA syn-
thesis, we reasoned that degradation of the incoming viral
RNA would result in reduced accumulation of viral DNA.
However, no significant difference was seen in either ASLV
or HIV. Thus we concluded that the incoming genome was
not a substrate for RNAi.
This conclusion has not been reached by all investiga-
tors that examined this issue. Cullen and coworkers found
that the incoming genome could be a substrate for RNAi,
though not utilized as efficiently as the newly synthe-
sized RNAs (Coburn and Cullen, 2002). Stevenson and
coworkers did not detect much difference in sensitivity to
RNAi in the early and late RNAs (Jacque et al., 2002).
The reasons for these differences remain to be determined.
Possibly different segments of the early RNAs are differ-
entially exposed in the viral core, so that the choice of
RNA sequence to target determines whether or not they
are sensitive. Perhaps the detailed differences in exper-
imental protocols somehow dictate the outcome. It will
be interesting to work out these differences with further
experimentation.
Why would the incoming viral RNA genome be less sen-
sitive to RNAi? One possibility is that the incoming genomic
RNA may be protected by viral proteins, thereby blocking
62 W.-Y. Hu et al. / Virus Research 102 (2004) 59–64
access of the RNAi machinery. Another possibility is that
incoming genomic RNA does not traffic through a cellular
compartment that harbors the RNAi machinery. For exam-
ple, if the RNAi machinery resides at the ribosome or the
nuclear pore, the incoming genomic RNA may not encounter
RNAi factors. In support of this, the RNAi machinery is re-
ported to be associated with the translational factor eIF2c,
suggesting ribosomal association (Doi et al., 2003; Sasaki
et al., 2003; Tabara et al., 1999).
4. RNA interference against chicken retroviruses in an
embryo electroporation model
4.1. RNAi in chicken embryos
In ovo electroporation of chick embryos provides a sim-
ple and convenient system for studying gene activity during
development (Nakamura and Funahashi, 2001). In ovo elec-
troporation involves the injection of nucleic acids into the
central canal of the spinal cord, followed by electroporation
with five short (25-ms) square wave pulses at low voltages
(25 V). Nucleic acid enters the side of the embryo near the
positive electrode. The other half of the embryo is not elec-
troporated and serves as a control. In previous studies, elec-
troporation of dominant-negative gene constructs into chick
embryos has allowed analysis of vertebrate gene function
in the developing spinal cord (Thaler et al., 2002). We have
found that RNA interference can be readily induced in the
chick embryo spinal cord by in ovo electroporation, which
allowed the demonstration in vivo of inhibition of retroviral
replication by RNAi. Pekarik et al. (2003) have also demon-
strated the use of RNAi by in ovo electroporation.
Initially, to investigate whether RNAi took place in the
chick embryo, we used GFP as a target. A gfp encoding plas-
mid was electroporated into the chicken neural tube together
with either specific (siGFP) or nonspecific (siLUC against
firefly luciferase) siRNAs. Two days after electroporation,
in the absence of siRNA or in the presence of the control
siRNA bright green cells were detected in the half of the
spinal cord proximal to the positive electrode. In embryos
electroporated with the gfp plasmid and siGFP, a greater
than 90% reduction in GFP expression was observed. Elec-
troporation of siRNAs up to 3 ␮g/␮l showed no interruption
of embryo development (Hu et al., 2002). These studies and
those of Pekarik’s group establish that RNAi can operate in
the developing chick spinal cord.
4.2. Inhibiting ALV infection by in ovo electroporation
To investigate whether RNAi could inhibit replication of
RCAS in embryos, the two siRNAs against gag that had been
shown above to be active in cell culture were used. siLUC
was used as a nonspecific control. Two days after fertiliza-
tion, embryos were electroporated with a plasmid encod-
ing the RCAS genome together with siRNAs. The embryos
were analyzed at day 4 by sectioning and staining with an
antibody against ASLV Gag protein p19.
Electroporation of RCAS DNA with no siRNA or control
siRNA resulted in virus spread from the electroporated half
of the spinal cord into the surrounding mesenchyme (due to
secondary infection). At 2 days post-electroporation, ASLV
replication was inhibited efficiently (>90%) by either of the
siRNAs targeting gag: infection was evident in only a few
cells by anti-Gag staining. These results indicate that RNAi
can inhibit ASLV replication in chick embryos.
4.3. Inhibiting pathogenesis by RSV
We next asked whether RNAi could inhibit transforma-
tion by RSV in the chick embryo model. RSV encodes the
oncogenic src gene, which disrupts signal transduction and
causes sarcomas in chickens. We electroporated embryos
with DNA encoding RSV and different siRNAs. Three days
after electroporation, none of chick embryos treated with
RSV plus control siRNA survived. In contrast, 7 of 12 em-
bryos treated with the siGAG against p12 survived.
In order to examine the effect of RNAi on RSV patho-
genesis, chick embryo sections were stained with different
antibodies at 36 h post-electroporation. The mpm2 mono-
clonal antibody detected M-phase phosphoproteins, thereby
identifying mitotic cells, and kip1 (cyclin-dependent ki-
nase inhibitor p27) staining detected inhibition of cell
cycle progression, showing cells arrested in G1. The Gag
(p19) antibody was used to monitor the spread of RSV.
In RSV-infected chick embryos, the neural tube was dis-
organized and normal cell cycle progression disrupted in
the electroporated half. In contrast, normal development
and properly controlled cell cycle progression were seen
when RSV replication was inhibited by RNAi, thereby
documenting inhibition of pathogenesis.
5. Applications of RNAi against animal retroviruses
Today, avian retroviruses are a serious problem in the
poultry industry. Crowded conditions in poultry farms are
ideal for spread of epidemics of infectious disease, and retro-
viruses remain a prominent threat. Vaccines are largely in-
effective, so new measures against disease caused by avian
retroviruses are needed.
The three main disease-causing avian retrovirus of agri-
cultural significance are the ASLV, reticuloendotheliosis
virus (REV), and lymphoproliferative disease virus (LPDV)
of turkeys (for review, see Payne, 1998). The ASLVs are
categorized into six subgroups based on their receptor
tropism, A–E and J. Of these, A, B, and J occur com-
monly. Although genetic resistance (due to recessive re-
sistance genes) to subgroups A–E have been observed in
some chickens, all chicken strains are susceptible to sub-
group J (ALV-J). ALV-J induces myelocytic leukosis, a
tumor-inducing infection of white blood cells from the bone
 W.-Y. Hu et al. / Virus Research 102 (2004) 59–64
marrow. Control of ALV-J has become a major concern to
the poultry industry as losses of breeder birds to ALV-J
can be as high as 20–40% in severe outbreaks. The REV
family of retroviruses causes a variety of diseases including
chronic lymphatomas in chickens, ducks, turkeys, geese,
pheasants, and quail. In turkeys, LPDV infection causes a
lymphoproliferative disease.
Furthermore, some poxviruses responsible for fowlpox
outbreaks in the United States have been found to harbor an
intact REV provirus in their genomes (Kim and Tripathy,
2001; Singh et al., 2003). Evidently the REV virus used
the poxvirus genome as an integration target rather than the
host cell chromosome, thereby incorporating the retroviral
genome in the poxvirus genome. The addition of REV to the
fowlpox genome has been implicated in increased poxvirus
replication in host chickens, and may facilitate infection of
previously vaccinated poultry.
An attractive potential technology to suppress retroviral
diseases of poultry would be germ line insertion of RNAi
constructs targeting either retroviral sequences or dispens-
able host proteins involved in the infectious process. For
example, DNA hairpins initiating RNAi against the gag
targets described above could be installed in the chicken
genome, blocking replication of pathogenic ASLVs. Such an
approach, if successful, might be applied to inhibiting viral
infection and pathogenesis in a wide range of domestic ani-
mals. RNAi is also being investigated as a means of treating
HIV infection in humans, but development of RNAi as a
drug seems much more difficult—germ line modification of
farm animals, in contrast, seems potentially feasible in the
near term.
In summary, RNAi appears to be a promising new tech-
nology for preventing and treating retroviral diseases, well
warranting a review that begins and ends with a bang!
References
Al-Kaff, N.S., Covey, S.N., Kreike, M.M., Page, A.M., Pinder, R., Dale,
P.J., 1998. Transcriptional and posttranscriptional plant gene silencing
in response to a pathogen. Science 279, 2113–2115.
Banerjea, A., Li, M.J., Bauer, G., Remling, L., Lee, N.S., Rossi, J., Akkina,
R., 2003. Inhibition of HIV-1 by lentiviral vector-transduced siRNAs in
T lymphocytes differentiated in SCID-hu mice and CD34+ progenitor
cell-derived macrophages. Mol. Ther. 8, 62–71.
Bishop, J.M., 1991. Molecular themes in oncogenesis. Cell 64, 235–248.
Bushman, F.D., 2003. RNA interference: applications in vertebrates. Mol.
Ther. 7, 9–10.
Caplen, N.J., Fleenor, J., Fire, A., Morgan, R.A., 2000. dsRNA-mediated
gene silencing in cultured Drosophila cells: a tissue culture model for
the analysis of RNA interference. Gene 252, 95–105.
Caplen, N.J., Parrish, S., Imani, F., Fire, A., Morgan, R.A., 2001. Spe-
cific inhibition of gene expression by small double-stranded RNAs in
invertebrate and vertebrate systems. Proc. Natl. Acad. Sci. U.S.A. 98,
9742–9747.
Capodici, J., Kariko, K., Weissman, D., 2002. Inhibition of HIV-1 infection
by small interfering RNA-mediated RNA interference. J. Immunol.
169, 5196–5201.
63
Carmell, M.A., Xuan, Z., Zhang, M.Q., Hannon, G.J., 2002. The Arg-
onaute family: tentacles that reach into RNAi, developmental control,
stem cell maintenance, and tumorigenesis. Genes Dev. 16, 2733–2742.
Caudy, A.A., Ketting, R.F., Hammond, S.M., Denli, A.M., Bathoorn,
A.M., Tops, B.B.J., Silva, J.M., Myers, M.M., Hannon, G.J., Plasterk,
R.H.A., 2003. A micrococcal nuclease homologue in RNAi effector
complexes. Nature 425, 411–415.
Coburn, G.A., Cullen, B.R., 2002. Potent and specific inhibition of human
immunodeficiency virus type 1 replication by RNA interference. J.
Virol. 76, 9225–9231.
Cogoni, C., Macino, G., 1999. Gene silencing in Neurospora crassa
requires a protein homologous to RNA-dependent RNA polymerase.
Nature 399, 166–169.
Doench, J.G., Petersen, C.P., Sharp, P.A., 2003. siRNAs can function as
miRNAs. Genes Dev. 17, 438–442.
Doi, N., Zenno, S., Ueda, R., Ohki-Hamazaki, H., Ui-Tei, K., Saigo, K.,
2003. Short-interfering-RNA-mediated gene silencing in mammalian
cells requires Dicer and eIF2C translation initiation factors. Curr. Biol.
13, 41–46.
Dykxhoorn, D.M., Novina, C.D., Sharp, P.A., 2003. Killing the messenger:
short RNAs that silence gene expression. Nat. Rev. Mol. Cell. Biol.
4, 457–467.
Elbashir, S.M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., Tuschl,
T., 2001a. Duplexes of 21-nucleotide RNAs mediate RNA interference
in cultured mammalian cells. Nature 411, 494–498.
Elbashir, S.M., Lendeckel, W., Tuschl, T., 2001b. RNA interference is
mediated by 21 and 22-nucleotide RNAs. Genes Dev. 15, 188–200.
Ellerman, V., Bang, O., 1908. Experimentelle Leukamie bei Huhnern.
Zentralbl. Bakteriol. Parasitenkd. Infectionskr. Hyg. Abt. Orig. 46,
595–609.
Federspiel, M.J., Hughes, S.H., 1997. Retroviral gene delivery. Methods
Cell. Biol. 52, 179–214.
Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver, S.E., Mello,
C.C., 1998. Potent and specific genetic interference by double-stranded
RNA in Caenorhabditis elegans. Nature 391, 806–811.
Garrus, J.E., von Schwedler, U.K., Pornillos, O.W., Morham, S.G., Zavitz,
K.H., Wang, H.E., Wettstein, D.A., Stray, K.M., Cote, M., Rich, R.L.,
Myszka, D.G., Sundquist, W.I., 2001. Tsg101 and the vacuolar protein
sorting pathway are essential for hiv-1 budding. Cell 107, 55–65.
Hamilton, A.J., Baulcombe, D.C., 1999. A species of small antisense RNA
in posttranscriptional gene silencing in plants. Science 286, 950–952.
Hammond, S.M., Bernstein, E., Beach, D., Hannon, G.J., 2000. An
RNA-directed nuclease mediates post-transcriptional gene silencing in
Drosophila cells. Nature 404, 293–296.
Hu, W.Y., Meyers, C.P., Kilzer, J.M., Pfaff, S.L., Bushman, F.D., 2002.
Inhibition of retroviral pathogenesis by RNA interference. Curr. Biol.
12, 1301–1311.
Hughes, S.H., Greenhouse, J.J., Petropoulos, C.J., Sutrave, P., 1987.
Adaptor plasmids simplify the insertion of foreign DNA into
helper-independent retroviral vectors. J. Virol. 61, 3004–3012.
Jacque, J.M., Triques, K., Stevenson, M., 2002. Modulation of HIV-1
replication by RNA interference. Nature 418, 435–438.
Jorgensen, R., 1990. Altered gene expression in plants due to trans
interactions between homologous genes. Trends Biotechnol. 8, 340–
344.
Keogh, E.V., 1938. Ectodermal lesions produced by the virus of Rous
sarcoma. Br. J. Exp. Pathol. 19, 1–9.
Kim, T.J., Tripathy, D.N., 2001. Reticuloendotheliosis virus integration in
the fowl poxvirus genome: not a recent event. Avian Dis. 45, 639–663.
Lee, N.S., Dohjima, T., Bauer, G., Li, H., Li, M.J., Ehsani, A., Salvaterra,
P., Rossi, J., 2002. Expression of small interfering RNAs targeted
against HIV-1 rev transcripts in human cells. Nat. Biotechnol. 20,
500–505.
Li, H., Li, W.X., Ding, S.W., 2002. Induction and suppression of RNA
silencing by an animal virus. Science 296, 1319–1321.
Li, M.J., Bauer, G., Michienzi, A., Yee, J.K., Lee, N.S., Kim, J., Li,
S., Castanotto, D., Zaia, J., Rossi, J.J., 2003. Inhibition of HIV-1
64 W.-Y. Hu et al. / Virus Research 102 (2004) 59–64
infection by lentiviral vectors expressing Pol III-promoted anti-HIV
RNAs. Mol. Ther. 8, 196–206.
Lipardi, C., Wei, Q., Paterson, B.M., 2001. RNAi as random degradative
PCR: siRNA primers convert mRNA into dsRNAs that are degraded
to generate new siRNAs. Cell 107, 297–307.
Marathe, R., Anandalakshmi, R., Smith, T.H., Pruss, G.J., Vance, V.B.,
2000. RNA viruses as inducers. Plant Mol. Biol. 43, 295–306.
Martinez, J., Patkaniowska, A., Urlaub, H., Luhrmann, R., Tuschl, T.,
2002a. Single-Stranded Antisense siRNAs Guide Target RNA Cleavage
in RNAi. Cell 110, 563–574.
Martinez, M.A., Gutierrez, A., Armand-Ugon, M., Blanco, J., Parera, M.,
Gomez, J., Clotet, B., Este, J.A., 2002b. Suppression of chemokine
receptor expression by RNA interference allows for inhibition of
HIV-1 replication. AIDS 16, 2385–2390.
Nakamura, H., Funahashi, J., 2001. Introduction of DNA into chick
embryos by in ovo electroporation. Methods 24, 43–48.
Ngo, H., Tschudi, C., Gull, K., Ullu, E., 1998. Double-stranded RNA
induces mRNA degradation in Trypanosoma brucei. Proc. Natl. Acad.
Sci. U.S.A. 95, 14687–14692.
Novina, C.D., Murray, M.F., Dykxhoorn, D.M., Beresford, P.J., Riess, J.,
Lee, S.K., Collman, R.G., Lieberman, J., Shankar, P., Sharp, P.A.,
2002. siRNA-directed inhibition of HIV-1 infection. Nat. Med. 8,
681–686.
Palatnik, J.F., Allen, E., Wu, X., Schommer, C., Schwab, R., Carrington,
J.C., Weigel, D., 2003. Control of leaf morphogenesis by microRNAs.
Nature 425, 257–263.
Park, W.S., Miyano-Kurosaki, N., Hayafune, M., Nakajima, E., Matsuzaki,
T., Shimada, F., Takaku, H., 2002. Prevention of HIV-1 infection in
human peripheral blood mononuclear cells by specific RNA interfer-
ence. Nucleic Acids Res. 30, 4830–4835.
Payne, L.N., 1998. Retrovirus-induced disease in poultry. Poult. Sci. 77,
1204–1212.
Pekarik, V., Bourikas, D., Miglino, N., Joset, P., Preiswerk, S., Stoeckli,
E.T., 2003. Screening for gene function in chicken embryo using
RNAi and electroporation. Nat. Biotechnol. 21, 93–96.
Pickford, A.S., Cogoni, C., 2003. RNA-mediated gene silencing. Cell
Mol. Life Sci. 60, 871–882.
Plasterk, R.H., 2002. RNA silencing: the genome’s immune system. Sci-
ence 296, 1263–1265.
Qin, X.F., An, D.S., Chen, I.S., Baltimore, D., 2003. Inhibiting HIV-1
infection in human T cells by lentiviral-mediated delivery of small
interfering RNA against CCR5. Proc. Natl. Acad. Sci. U.S.A. 100,
183–188.
Ratcliff, F.G., MacFarlane, S.A., Baulcombe, D.C., 1999. Gene silencing
without DNA. RNA-mediated cross-protection between viruses. Plant
Cell 11, 1207–1216.
Romano, N., Macino, G., 1992. Quelling: transient inactivation of gene
expression in Neurospora crassa by transformation with homologous
sequences. Mol. Microbiol. 6, 3343–3353.
Rous, P., 1911. A sarcoma of the fowl transmissible by an agent separable
from the tumor cells. J. Exp. Med. 13, 397–411.
Sasaki, T., Shiohama, A., Minoshima, S., Shimizu, N., 2003. Identification
of eight members of the Argonaute family in the human genome small
star, filled. Genomics 82, 323–330.
Schwarz, D.S., Hutvagner, G., Haley, B., Zamore, P.D., 2002. Evidence
that siRNAs function as guides, not primers, in the Drosophila and
human RNAi pathways. Mol. Cell. 10, 537–548.
Sijen, T., Fleenor, J., Simmer, F., Thijssen, K.L., Parrish, S., Tim-
mons, L., Plasterk, R., Fire, A., 2001. On the role of RNA
amplification in dsRNA-triggered gene silencing. Cell 107, 465–
476.
Singh, P., Schnitzlein, W.M., Tripathy, D.N., 2003. Reticuloendotheliosis
virus sequences within the genomes of field strains of fowlpox virus
display variability. J. Virol. 77, 5855–5862.
Surabhi, R.M., Gaynor, R.B., 2002. RNA interference directed against
viral and cellular targets inhibits human immunodeficiency virus type
1 replication. J. Virol. 76, 12963–12973.
Tabara, H., Sarkissian, M., Kelly, W.G., Fleenor, J., Grishok, A., Tim-
mons, L., Fire, A., Mello, C.C., 1999. The rde-1 gene, RNA in-
terference, and transposon silencing in C. elegans. Cell 99, 123–
132.
Temin, H., Mizutani, S., 1970. RNA-dependent DNA polymerase in viri-
ons of Rous sarcoma virus. Nature 226, 1211–1213.
Temin, H.M., Rubin, H., 1958. Characteristics of an assay for Rous
sarcoma virus and Rous sarcoma cells in tissue culture. Virology 6,
669–688.
Thaler, J.P., Lee, S.K., Jurata, L.W., Gill, G.N., Pfaff, S.L., 2002. LIM
factor Lhx3 contributes to the specification of motor neuron and in-
terneuron identity through cell-type-specific protein-protein interac-
tions. Cell 110, 237–249.
Zeng, Y., Wagner, E.J., Cullen, B.R., 2000. Both natural and designed
micro RNAs can inhibit the expression of cognate mRNAs when
expressed in human cells. Mol. Cell. 9, 1327–1333.
Zeng, Y., Yi, R., Cullen, B.R., 2003. MicroRNAs and small interfering
RNAs can inhibit mRNA expression by similar mechanisms. Proc.
Natl. Acad. Sci. U.S.A. 100, 9779–9784.
 Virus Research 102 (2004) 65–74

RNA interference, arthropod-borne viruses, and mosquitoes

Irma Sanchez-Vargas, Emily A. Travanty, Kimberly M. Keene, Alexander W.E. Franz,
Barry J. Beaty, Carol D. Blair, Ken E. Olson∗
Arthropod-Borne and Infectious Diseases Laboratory, Department of Microbiology, Immunology, and Pathology,
Colorado State University, Fort Collins, CO 80523, USA

Abstract

RNA interference (RNAi) probably functions as an antiviral mechanism in most eukaryotic organisms. Variations in the activity of this
antiviral pathway in mosquitoes could explain, in part, why some mosquitoes are competent vectors of medically important, arthropod-borne
viruses (arboviruses) and others are not. There are three lines of evidence that show the RNAi pathway exists in Aedes species that transmit
arboviruses. The first is that recombinant Sindbis viruses expressing a RNA fragment from a genetically unrelated dengue-2 virus (DENV-2)
interfere with DENV-2 replication in Aedes aegypti mosquitoes by a mechanism similar to virus-induced gene silencing described in plants.
The second is that transfection of C6/36 (Aedes albopictus) cells with either double-stranded RNA or synthetic small interfering RNAs derived
from an arbovirus genome interferes with replication of the homologous virus. The third is that a hairpin DENV-2-specific RNA transcribed
from a plasmid can generate virus-resistant C6/36 cells. We hypothesize that genetically modified mosquitoes can be generated that transcribe
a flavivirus-specific dsRNA, triggering the RNAi response soon after ingestion of a blood meal. This could induce the RNAi pathway in the
midgut prior to establishment of virus infection and profoundly change vector competence. Towards this goal, we are developing transgenic
A. aegypti lines that are refractory to DENV by exploiting the RNAi pathway.
© 2004 Elsevier B.V. All rights reserved.

Keywords: Mosquitoes; RNAi; dsRNA; siRNA; Dengue viruses; Arthropod-borne viruses

1. Introduction competent mosquito populations that support virus infec-

Mosquitoes transmit a number of medically important
arthropod-borne viral pathogens to humans and other an-
imals. Dengue viruses (DENV: family Flaviviridae, genus
Flavivirus, species Dengue virus) cause an estimated
50–100 million human cases of dengue fever (DF) every
year (Monath, 1994). The rapid spread of West Nile virus
(family Flaviviridae, genus Flavivirus, species West Nile
virus) in North America and outbreaks of Venezuelan equine
encephalitis virus (family Togaviridae, genus Alphavirus,
species Venezuelan equine encephalitis virus) in South
America and Rift Valley fever virus (RVFV, family Bun-
yaviridae, genus Phlebovirus, species Rift Valley fever virus)
in the Middle East during the last decade further demon-
strate the enormous public health problems associated with
arbovirus infections. The maintenance and transmission of
arboviruses in nature is, in part, dependent on genetically
∗ Corresponding author. Tel.: +1-970-491-4383;
fax: +1-970-491-8323.
E-mail address: kolson@lamar.colostate.edu (K.E. Olson).
tion. Traditional mosquito control to lower the population
density of vectors and prevent arboviral disease has been
unsuccessful due to cost, expanding urbanization, increases
in human populations, pesticide resistance, and spreading
of vector species due to globalization of trade and air travel.
New strategies are needed to control these diseases.
Dengue viruses arguably cause the most important
arthropod-borne viral infections. DENV serotypes 1–4 are
transmitted to humans by Aedes (A.) aegypti and A. albopic-
tus. Severe forms of DENV disease, DEN hemorrhagic fever
(DHF) and DEN shock syndrome (DSS) are probably caused
by the global trafficking of virulent genotypes of DENV
(Gubler, 1998) and sequential infection of individuals with
heterotypic virus serotypes leading to antibody-mediated
disease enhancement (Halstead, 1989). DHF/DSS are in-
creasing at an alarming rate due to the hyperendemnicity
of DENV in many parts of the world and the trafficking of
DENV serotypes to new geographic regions (Gubler, 1998).
Dengue viruses, like other flaviviruses, are small, enve-
loped particles (50 nm in diameter) containing a single-
stranded, positive-sense RNA of approximately 11 Kb. The

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.017
66 I. Sanchez-Vargas et al. / Virus Research 102 (2004) 65–74
genomic RNA is capped at the 5 end but not polyadeny-
lated at the 3 end. The virion RNA contains a single long
open reading frame encoding three structural proteins, cap-
sid (C), premembrane (prM), and envelope (E), and seven
non-structural proteins, NS1, NS2A, NS2B, NS3, NS4A,
NS4B, and NS5. The genomic RNA is translated into a
polyprotein precursor that is co- and post-translationally
processed by cellular and viral proteases into individual
proteins (Chambers et al., 1990).
Dengue viruses enter the mosquito when the adult fe-
male takes an infectious blood meal. In competent mosquito
vectors, viral amplification begins with the initial, unidirec-
tional, productive infection of the midgut epithelium cells.
Following replication of virus in the epithelial cells, virus
escapes the midgut and disseminates to secondary target or-
gans such as the salivary glands. Further replication occurs
in the salivary glands, and virus is eventually shed into the
salivary gland ducts. Transmission occurs by way of infected
saliva in a subsequent bite. The time between initial infection
of the midgut and transmission of virus in saliva is termed
the extrinsic incubation period (EIP). The EIP for DENV in
competent A. aegypti is usually 7–14 days (Chandler et al.,
1998; Scott and Burrage, 1984).
Transmission of virus to the vertebrate host is prevented
if one or more barriers to infection and dissemination are
present in the mosquito (Hardy and Strauss, 1988; Woodring
et al., 1996). Little is known about the molecular basis of
these barriers. However, it is possible to speculate that a
midgut infection barrier (MIB) could occur if the arbovirus
fails to recognize critical receptors for entry into the midgut
epithelia, the virus is not processed correctly by proteolytic
enzymes in the midgut lumen, or the virus fails to uncoat
during initial steps of infection. MIBs have been observed
in certain A. aegypti populations after oral infection with a
strain of DENV-2 that is infectious for other populations
(Bennett et al., 2002; Black et al., 2002). A. aegypti can
also exhibit a midgut escape barrier (MEB). Here, virus in-
fects the midgut epithelium but is unable to disseminate
to other organs and remains restricted and sequestered in
the midgut cells of some populations of mosquitoes (Black
et al., 2002). Although the molecular basis for MEBs is cur-
rently unknown, one can hypothesize that virus replication
in the midgut or secondary tissues is somehow modulated.
A. aegypti with MEBs to DENV infections have also been
described (Bennett et al., 2002; Black et al., 2002).
The fields of molecular biology, vector biology, and
bioinformatics are now converging to define the nature of
these barriers to infection. For example, the genomes of
two insects in the order Diptera, the fruit fly Drosophila
(D.) melanogaster and the medically important mosquito
vector of malaria, Anopheles (A.) gambiae, have now been
sequenced and annotated (Adams et al., 2000; Holt et al.,
2002). The genome sequence determination of a third
dipteran, A. aegypti, is currently underway and should be
available in the next several years. We can now use bioin-
formatic and functional genomic analyses to develop a
more complete understanding of arbovirus-mosquito vec-
tor interactions. Identifying molecules that modulate ar-
bovirus replication in mosquitoes and the characterization
of mosquito genes that encode these molecules may lead
to powerful new approaches to assess risk and attempt
new strategies to control arbovirus transmission (Beerntsen
et al., 2000; Blair et al., 2000).
2. Arboviruses, mosquitoes and RNA interference
What could be the nature of arbovirus modulation in
mosquito cells? A clue may lie in the replication strat-
egy used by positive-sense RNA viruses like DENV. Fla-
vivirus RNA replication occurs in the cytoplasm of infected
cells and is asymmetric and semi-conservative (Chu and
Westaway, 1985). RNA replicative intermediates (RI) are
partially double-stranded RNAs (dsRNA) generated within
infected cells (Westaway et al., 1997). Recently, Uchil
and Satchidanandam (2003) provided biochemical evi-
dence that flaviviruses rearrange the architecture of cell
membranes associated with virus replication to sequester
the dsRNAs in a double membrane barrier enclosing the
viral replication complex (RC). The intricate mechanism
adopted by flaviviruses to hide the RI suggests that this
adaptation may have evolved to limit exposure of the RC
to dsRNA-mediated host defenses such as protein kinase
R (PKR), RNase L, ␣/␤ interferon, and RNA interference
(RNAi). Simply put, the host cell and flaviviruses appear
to have co-evolved defense-counter defense strategies.
Mosquito cells do not appear to have PKR, RNase L, or ␣/␤
interferon pathways comparable to those found in mam-
malian cells, but mosquito cells most likely have an RNAi
pathway. Therefore arbovirus interactions with mosquito
cells should be an excellent model for understanding how
animal viruses interact with RNAi pathway. Can RNAi act
as an antiviral pathway in mosquitoes, and if so, how do
arboviruses overcome this pathway in competent mosquito
vectors?
RNAi is an evolutionarily conserved process through
which dsRNA induces the silencing of cognate gene expres-
sion (Bernstein et al., 2001; Carthew, 2001). RNAi results
in the degradation of dsRNA and of any mRNA present
in the organism with high sequence homology with the
dsRNA trigger (Bernstein et al., 2001). Synthetic dsRNA
can be introduced by transfection or can be produced intra-
cellularly by the introduction of viruses or transposons that
generate dsRNA during their replication cycle (Hammond
et al., 2001b). In addition, transcripts from short hairpin
sequences encoded within the genome of many organisms
appear to enter the RNAi pathway and function to reg-
ulate the expression of endogenous protein-coding genes
(Grishok et al., 2001; Hutvagner et al., 2001; Hutvagner
and Zamore, 2002; Ketting et al., 2001; Knight and Bass,
2001). DsRNA is initially cleaved into small interfering
RNA (siRNA), 21–25 nucleotides in length, by a Dicer-1
 I. Sanchez-Vargas et al. / Virus Research 102 (2004) 65–74 67

Fig. 1. Proposed model for RNA-silencing pathway in insects (Drosophila). dsRNA is recognized by Dicer (Step 1) and cleaved into 21–25 nt dsRNA
fragments (siRNAs) forming siRNA-protein complexes (Step 2). Double-stranded siRNA molecules are unwound by the Dicer-associated helicase and
incorporated into RISC (Step 3). RISC is guided by the incorporated siRNAs to target RNA molecules of complementary sequence (Step 4). RISC then
cleaves target RNA (Step 5). Dicer: RNase III-like nuclease, helicase activity, ATP dependent. RISC: RNA-induced silencing complex (endonuclease).

protein (Fig. 1). The siRNAs are then incorporated into
the RNA-induced silencing complex (RISC), which con-
tains proteins of the Argonaute family (rde-4, rde-1 and
drh-1/2 in Caenorhabditis (C.) elegans or Argonaute 2
in D. melanogaster) (Tabara et al., 2002; Williams and
Rubin, 2002; Zamore et al., 2000). siRNAs are unwound
by an RNA helicase to act as guide sequences that direct
the RNA degradation machinery to the target mRNAs. The
RISC monitors the sequences of cytoplasmic RNA and is
able to cleave the cognate target RNA in a sequence spe-
cific, siRNA-dependent manner (Martinez et al., 2002). It
has been proposed for plants and lower eukaryotes such as
C. elegans, that siRNAs also function as primers that are
extended on the targeted RNA by an RNA-dependent RNA
polymerase (RdRp) to amplify the dsRNA trigger (Lipardi
et al., 2001; Sijen et al., 2001). Although RdRp-like activ-
ity has been detected in D. melanogaster embryos, it does
not appear to play a significant role in RNAi within D.
melanogaster and A. gambiae (Hoa et al., 2003; Roignant
et al., 2003).
What is the evidence that RNAi is an important antiviral
defense strategy used by the host to modulate virus infec-
tions? First, there is a wealth of evidence that plant RNA
viruses act as both inducers and targets of RNAi (Ruiz
et al., 1998; Vance and Vaucheret, 2001). In addition, many
plant RNA viruses encode suppressors of RNAi, supporting
the hypothesis that RNAi is a viral defense system. For
example, the tomato bushy stunt virus (family Tombusviri-
dae, genus Tombusvirus, species Tomato bushy stunt virus)
p19 protein suppresses post-transcriptional gene silenc-
ing (PTGS) in plants by binding siRNAs generated after
virus infection (Silhavy et al., 2002). Furthermore, the in-
sect virus, Flock house virus (family Nodaviridae, genus
Alphanodavirus, species Flock house virus) B2 protein
suppresses RNAi activity in both plants and Drosophila S2
cells, emphasizing that an evolutionarily conserved RNAi
pathway has a natural antiviral role (Li et al., 2002b).
The sequencing of the A. gambiae genome now makes
it possible to identify genes that may be involved in the
RNAi pathway and what role, if any, these genes have in ar-
bovirus infection. Recent studies have demonstrated that sev-
eral members of the Dicer and Argonaute protein families,
homologous to D. melanogaster proteins, are encoded in the
A. gambiae genome (Hoa et al., 2003). To identify the role of
these proteins in RNAi, Hoa and colleagues initially showed
that transfection of A. gambiae cells with dsRNA derived
from a 500 bp region of the firefly luciferase gene could effi-
ciently silence transient expression of a luciferase transgene.
A. gambiae Sua1B cells were pretreated with dsRNA derived
from A. gambiae dcr1, dcr-2, or ago1-5 prior to silencing lu-
ciferase expression. Pretreatment with dsRNA derived from
A. gambiae dcr2, ago2, or ago3 consistently led to recovery
of luciferase activity, by knocking down expression of com-
ponents of the RNAi pathway. That ago2 gene expression is
68 I. Sanchez-Vargas et al. / Virus Research 102 (2004) 65–74
important in RNAi is not surprising since it is thought to be
an important component of RISC (Hammond et al., 2001a).
However, significant recovery of luciferase activity was not
found in silenced Sua1B cells following pretreatment of cells
with dsRNA from A. gambiae dcr1, ago1, ago4, ago5, or
β-galactosidase genes. The gene dcr1 encodes the Dicer-1
protein that is essential for cleaving dsRNAs into siRNAs
in a number of organisms. Why down-regulation of dcr1
gene expression did not affect RNAi in A. gambiae cells
is not easily explained, but may be due to the cell culture
system used in these experiments. We have recently used a
different strategy in A. gambiae mosquitoes to assign func-
tion to these genes by co-injecting 103 plaque forming units
of O’nyong-nyong virus (ONNV; family Togaviridae, genus
Alphavirus, SFV complex) with ∼200 ng of dsRNA de-
rived from dcr1, dcr2, ago2 or ago3 genes. ONNV is one
of the few arboviruses naturally transmitted by A. gambiae.
We have observed that virus replication is significantly in-
creased in mosquitoes co-injected with ONNV and dsRNA
from dcr1, dcr2, ago2, or ago3 and not with ␤-gal dsRNA,
suggesting that these genes play an important role in RNAi
in the mosquito and that RNAi indeed functions as an antivi-
ral pathway in A. gambiae (K.M. Keene, unpublished data).
Also, we have shown that co-injection of dsRNA from the
nsp1 and nsp3 genome regions from ONNV significantly
decreases the amount of virus found in the mosquito (K.M.
Keene and K.E. Olson, unpublished data).
It is very likely that an RNA silencing mechanism also
exists in A. aegypti and A. albopictus mosquitoes. Aedes
species of mosquitoes transmit a number of medically impor-
tant arboviruses, such as DENV, yellow fever virus (family
Flaviviridae, genus Flavivirus, species yellow fever virus),
Chikungunya virus (family Togaviridae, genus Alphavirus,
SFV complex), and RVFV. If RNAi is present, it might be a
likely mechanism for modulating arbovirus replication in the
natural host and may explain, in part, why some mosquitoes
are refractory for transmission of arboviruses such as DENV.
3. Evidence for RNA-mediated silencing of DENV
replication in mosquito cells
The first evidence that a functional RNAi pathway ex-
ists in Aedes mosquitoes came from studies to generate A.
albopictus C6/36 cells that were resistant to DENV-2 infec-
tion. We observed that transduction by recombinant Sindbis
viruses (family Togaviridae, genus Alphavirus, WEEV com-
plex) carrying the prM coding region of DENV-2 in their
genome in either sense or antisense orientation made the
cells resistant to DENV-2, but not DENV-3 (Gaines et al.,
1996; Olson et al., 2002). We then showed that this resis-
tance phenomenon could be reproduced in adult, female A.
aegypti by co-injecting 103 pfu/ml of recombinant Sindbis
virus and DENV-2 (Olson et al., 1996). Eleven days after
injection, DENV-2 was absent only in mosquitoes that were
co-infected with Sindbis viruses carrying the prM fragment.
This is reminiscent of the use of RNA virus expression vec-
tors to silence host genes in plants. Nicotiana benthamiana
inoculated with modified tobacco mosaic virus (family To-
bamoviridae, genus Tobamovirus, species tobacco mosaic
virus) or potato virus X (genus Potexvirus, species potato
virus X) carrying host-related inserts suppressed genes ho-
mologous to the inserts (Kumagai et al., 1995; Ruiz et al.,
1998). Some plant viruses also were able to induce silenc-
ing of either transgenes or genes carried by genetically
unrelated viruses having significant sequence identity with
the inoculated virus (Lindbo et al., 1993; Ratcliff et al.,
1999). This phenomenon was termed virus-induced gene
silencing (Angell and Baulcombe, 1997; Ruiz et al., 1998),
and the mechanism of action for VIGS was similar to
post-transcriptional gene silencing or RNAi.
We then conducted experiments to observe whether the
mechanism by which recombinant Sindbis viruses induced
resistance in mosquitoes to DENV-2 was similar to RNAi.
We engineered a Sindbis virus that contained a 290 base
region of DENV-2 prM in either sense or antisense ori-
entation and infected C6/36 cells with the recombinant
virus. Infection with Sindbis virus with both orientations
of DENV-2 prM RNA in the genome ablated the ability of
DENV-2 to infect cells, but not other serotypes of DENV.
We observed by Northern blot analysis that intracellular
concentrations of Sindbis genomic and subgenomic RNAs
carrying the DENV-2 prM RNA significantly declined by
96 h post-infection, but cells retained profound resistance
to DENV-2 challenge (Adelman et al., 2001). We reasoned
that if an RNAi phenomenon were responsible for the ob-
served resistance in C6/36 cells, then DENV-2 should be
unable to replicate in the cells even though the effector prM
sequences were undetectable. Three important features of
the resistance phenomenon suggested to us that recombi-
nant Sindbis viruses were triggering an RNAi-like response
in mosquito cells. (1) The interference observed was not re-
lated to the polarity of the inserted DENV sequence, because
Sindbis viruses with either sense or antisense DENV RNA
inserts conferred similar levels of resistance in mosquitoes
when challenged with the homologous DENV serotype. (2)
DENV-2 RNA failed to accumulate after challenge only
in cells infected with the recombinant Sindbis virus carry-
ing the DENV-2 sequence. (3) The resistance phenotype
was not dependent on high concentrations of effector RNA
transcribed by Sindbis virus as one might expect with an
antisense RNA mechanism. Finally, we have now shown
that the Sindbis virus used to transcribe the DENV-2 prM
effector RNA triggers formation of RNA species that are
21–23 nucleotides in size and consistent with siRNAs, the
hallmark of RNAi (Olson et al., 2002). These small RNAs
hybridized with sense and antisense RNA probes derived
from either Sindbis or DENV insert RNA sequences.
To test whether Sindbis viruses can silence endogenous
genes in A. aegypti, we engineered the Sindbis virus genome
to contain the coding region for early trypsin (TrypEarl)
protein, an important regulator of the proteolytic cascade
 I. Sanchez-Vargas et al. / Virus Research 102 (2004) 65–74
Fig. 2. Detection of siRNA in midgut of mosquitoes infected with re-
combinant dsSIN viruses. Fourteen days after infection the midguts of 70
mosquitoes were obtained and small RNAs were purified following the
protocol described by Hamilton and Baulcombe (1999) and hybridized
with 32 P-labeled probe complementary to TrypEarl gene. Lane 1, 22 nt
RNA marker; lane 2, uninfected mosquitoes; lane 3, mosquitoes infected
with dsSIN/GFP (MRE3 2J-GFP, negative control); lane 4, mosquitoes
infected with dsSIN/ETas (MRE3 2J-ETas-500 bp); lane 5, mosquitoes in-
fected with dsSIN/FETas (MRE3 2J-FETas-800 bp); lane 6, mosquitoes
infected with dsSIN/FETs (MRE3 2J-FETs-800 bp). FETas: full length
TrypEarl gene antisense; FETs: full length TrypEarl gene sense; ETas:
fragment of TrypEarl gene antisense derived from the 3 500 bp of the
gene.
in the midguts of blood fed mosquitoes. A recombinant
Sindbis virus was made that transcribed a 500 bp region
of TrypEarl gene. Expression of TrypEarl RNA sequence
from recombinant Sindbis viruses led to silencing of early
trypsin expression and the degradation of TrypEarl mRNA
in mosquito midguts. The phenomenon appeared to be
post-transcriptional, because we were able to detect siRNAs
(21–25 nt) using a TrypEarl RNA probe in the midguts of
these mosquitoes (Fig. 2). Inhibition of the expression of
TrypEarl also affected dissemination of DENV-2 to head
tissue. The experiments described above suggested that
virus-induced gene silencing (VIGS) occurs in mosquitoes
and that Sindbis virus expression vectors will be important
tools for understanding the role of mosquito genes in de-
termining vector competence and virus transmission. The
Sindbis expression system has already been used to knock-
down expression of reporter genes in transgenic A. aegypti
mosquitoes and determine gene function of phenoloxidase
in Armigeres subalbatus mosquitoes (Johnson et al., 1999;
Shiao et al., 2001).
Others have presented evidence that a RNAi-like defense
against DENV can be efficiently triggered in C6/36 mosquito
cells by dsRNA corresponding to a portion of the DENV-1
or DENV-2 genome. The study by Caplen et al. (2002)
69
demonstrated silencing of DENV-1 replication after trans-
fection of C6/36 cells with DENV-1 specific dsRNA. They
observed that different dsRNAs targeted to DENV-1 RNA
inhibited replication at similar levels, suggesting that degra-
dation of the whole genome RNA can be induced by cleav-
age within a region that corresponds to less that 1% of the
genome. However, the degree of inhibition reported using
Sindbis-induced silencing of DENV replication was greater
than reported for transfected dsRNA (Caplen et al., 2002).
Adelman et al. (2002) have shown that virus resistance
also can be generated in cultured C6/36 cells by transcription
of DENV-specific RNA that was designed to form a hair-
pin structure from a plasmid. The hairpin was transcribed
from the constitutive IE-1 promoter of Autographa califor-
nica baculovirus and consisted of the 567 base positive-sense
DENV-2 sequences encoding the prM protein followed by
a 290 base antisense RNA complementary to the first 290
bases of the prM RNA. After transformation with the plas-
mid, clonal cell lines were selected using an antibiotic resis-
tance marker and challenged with DENV-2. The cell lines
were classified as resistant to virus replication if fewer than
5% of cells in a culture expressed DENV-2 envelope (E)
antigen 7 days after challenge. Cells in the line designated
FB9.1 that was highly resistant to DENV-2 failed to accumu-
late DENV-2 genome RNA after challenge, indicating that
viral RNA can be targeted by expression of a dsRNA-trigger
(Fig. 3B). Most importantly, DENV-2 prM-specific, small
RNAs 21–25-nt in length were detected in resistant FB9.1
cells using protocols described by Hamilton and Baulcombe
(1999), consistent with RNA silencing as the mechanism
of resistance (Fig. 3A). These small RNA species were not
found in control cells. Surprisingly, in the FB9.1 cell line,
the rate of DENV-3 RNA and envelope protein accumulation
also was greatly reduced when compared to infection of con-
trol cells (Fig. 3B). The viral interference appeared to be se-
quence specific because the replication of the more distantly
related flavivirus, West Nile virus, was not affected (unpub-
lished data). The cross-resistance to DENV-3 seen in the
FB9.1 cells is consistent with RNAi. When the prM coding
regions from DENV-2 and DENV-3 are aligned, a number
of short homologous regions are found; one of these regions
is 23 nt in length with 2 mismatches and if it is produced as
siRNA, it might direct the silencing of both DENV-2 and
DENV-3. According to Elbashir et al. (2001) mismatches
are not tolerated at the middle of the siRNA as this is
the region that determines the cleavage site in the target
mRNA. The proposed DENV-2/DENV-3 siRNA does not
contain any mismatches in the vital cleavage area. Sequence
comparisons between DENV-2 and DENV-1, and DENV-2
and DENV-4 genomes reveal similar potential siRNA re-
gions. Experiments are currently underway to determine if
the cross-protection phenomenon extends to all 4 DENV
serotypes and if the cross-protection is indeed due to siR-
NAs resulting from the near-homologous regions.
Recently, we have directly demonstrated the role of
DENV-2 specific siRNAs in inducing RNAi to DENV in
70 I. Sanchez-Vargas et al. / Virus Research 102 (2004) 65–74

Fig. 3. Detection of siRNA in FB9.1 cells and DENV-2 and DENV-3 RNA accumulation in FB9.1 and H9.1 cells after DENV challenge. (A) Northern
blot hybridization to detect siRNA in FB9.1 cells. Small RNAs were obtained following the protocol described by Hamilton and Baulcombe (1999). Low
molecular weight RNAs were fractionated by 12% urea-PAGE, blotted and hybridized with 32 P-labeled sense and antisense RNA probes complementary
to prM gene region of DENV-2 RNA. (a) Untransformed C6/36 cells; (b and d) H9.1 control cells; (c and e). FB9.1 cells. The arrow shows DENV-2
prM-specific, small RNAs 21–25-nt in length. (B) Slot blot analysis comparing DENV-2 and DENV-3 RNA accumulation in FB9.1 and H9.1 cells after
DENV challenge. Total cellular RNA from each cell line on days 1–14 following DENV challenge was blotted and probed for DENV RNA. 32 P-labeled
probe complementary to the NS2b-NS3 gene region of DENV-2 RNA was used for detecting DENV-2 and a 32 P-labeled probe complementary to the C
gene region was used for detecting DENV-3. FB9.1: C6/36 cell line transformed with DENV-2 prM hairpin-expression plasmid. H9.1: C6/36 cell line
transformed with vector plasmid (control).

mosquito cells. We observed that transfection of C6/36
cells with synthetic 21 bp siRNA designed from a region of
DENV-2/DENV-3 prM homology (Dharmacon Research
Inc., Lafayette, CO) resulted in interference with DENV-2
replication (Fig. 4A). Experiments are underway to see if
this interference will extend to DENV-3. We also gener-
ated siRNAs in vitro by digesting a dsRNA derived from a
DENV-2 prM sequence with recombinant human Dicer-1
protein (Gene Therapy Systems). After transfection of the
siRNAs into C6/36 cells, the cells were challenged with

Fig. 4. Effect of siRNA or d-siRNA on DENV-2 mRNA stability. (A) Effect of siRNA on DENV-2 mRNA stability. C6/36 cells were transfected with
TransIT-TKO (Mirus, Corp. Madison, WI) as a negative control or in conjunction with either luciferase GL2 siRNA duplex or DENV siRNA duplex
(12 mM). After 48 h post-transfection, cells were infected with DENV-2. Total cellular RNA was isolated from transfected cells on day 6 following DENV
challenge, and was then blotted and probed for DENV RNA using a 32 P-labeled probe complementary to the NS2b-NS3 gene region of DENV-2. Lane
1, DENV-2 alone, 6 dpi; lane 2, control luciferase GL2 siRNA and DENV-2, 6 dpi; lane 3, DENV siRNA and DENV-2, 6 dpi. (B). Effect of a pool of
22 bp siRNA (d-siRNA) generated by cleavage of in vitro transcribed dsRNA corresponding to DENV-2 C-prM-M (485bp) using recombinant human
Dicer enzyme (Gene Therapy Systems San Diego, CA). C6/36 cells were transfected with Gene Silencer (Gene Therapy Systems) as a negative control
either in conjunction with d-siRNA/GFP or d-siRNA/DENV-2-M. After 48 h post-transfection, cells were infected with DENV-2. Total cellular RNA
was isolated from transfected cells on day 6 following DENV challenge, and was then blotted and probed for DENV RNA using a 32 P-labeled probe
complementary to the NS2b-NS3 gene region of DENV-2. Lane 1, DENV-2 alone; lane 2, d-siRNA/GFP and DEN2 6 dpi; lane 3, d-siRNA/ DENV-2-M
and DEN-2 6 dpi. The bottom panel shows the ethidium bromide stained rRNA.
 I. Sanchez-Vargas et al. / Virus Research 102 (2004) 65–74 71

DENV-2 and failed to accumulate viral RNA. Cells trans- as a disease control method, but as we learn more about
fected with siRNAs derived from GFP mRNA accumulated arbovirus-vector interactions we will learn more about the
normal concentrations of DENV-2 RNA after challenge feasibility of this type of approach.
(Fig. 4B). These experiments demonstrate that direct ap-
plication of siRNAs targeted to DENV RNA can cause
degradation of DENV genomic RNA in mosquito cells and 5. Conclusions
suggest that siRNAs may be an important tool for antiviral
therapy of patients infected with flaviviruses (Song et al., An important opportunity for investigation is to determine
2003). how arboviruses overcome the RNAi pathway. Arboviruses
could replicate faster than the ability of RNAi to overcome
the virus. Alphaviruses such as Sindbis may use this strat-
4. RNAi and new strategies for arbovirus disease control egy. We know that Sindbis viruses replicate and generate
virus in C6/36 cells within 12 h post infection. We also know
How can the information presented so far be used to de- that Sindbis viruses can induce an RNAi-like response in
velop new tools for predicting DENV disease outbreaks or mosquito cells at about 48 h post infection, at which time
controlling the spread of DENV disease in the future? Once we first detect the appearance of siRNAs. The appearance
we identify genes in the RNAi pathway that are involved in of siRNAs roughly correlates with decline in virus mRNAs,
the antiviral response we can, in theory, look for variations and detection of maximum concentration of siRNA occurs
in these genes in populations of vectors. If, as an example, at 120 h post-infection, at which time little viral RNA is de-
we can associate genetic variation in ago2 genes with vec- tected by Northern blot analysis (Fig. 5). In natural mosquito
tor competence for DENV-2 and incidence of transmission, hosts, the EIP for some alphaviruses is as short as 4 days
it may be possible to rapidly screen for and identify vector
populations with a high risk for transmitting the arbovirus
to humans. These vector populations can be targeted for
control or elimination.
We can also think about new control strategies that in-
duce a more robust antiviral RNAi pathway in mosquitoes.
We have hypothesized that genetically modified mosquitoes
could be generated transcribing a flavivirus-specific dsRNA
that triggers the RNAi response soon after ingestion of
a blood meal. This could induce the RNAi pathway in
midguts prior to (or concurrent with) the establishment of
virus infection and change a highly competent mosquito
vector for a flavivirus to a poorly competent vector or a
vector that completely eliminates the virus infection. Re-
ducing the number of infected vectors will in turn reduce
the incidence of human disease. Couple this strategy with a
genetic drive mechanism such as transposable elements, and
it may be possible to drive the resistance gene into vector
populations (Beerntsen et al., 2000; Ribeiro and Kidwell,
1994). Towards this goal, we are developing transgenic A.
aegypti lines that are refractory to DENV by exploiting the
RNAi pathway of mosquitoes (Adelman et al., 2002; Olson
et al., 2002). We are currently generating genetically mod-
ified A. aegypti that transcribe a dsRNA derived from the
DENV-2 prM sequence. The dsRNA is transcribed from a Fig. 5. Detection of Sindbis virus-specific siRNA in C6/36 cells. (A)
midgut-specific (e.g. carboxypeptidase) promoter to initiate Northern blot analysis of dsSIN virus transcripts. Total RNA from C6/36
resistance in the organ system first encountering virus fol- cells infected with TE3 2J virus were isolated at 1, 2, 3, 4, 5, 6 or 7
days post-infection, 32 P-labeled sense or antisense probes to E1 gene re-
lowing ingestion of an infected blood meal (Moreira et al., gion of Sindbis RNA were used for detection of Sindbis mRNA. The
2000). The DENV-2-derived prM cDNA fragments are top panel shows ethidium bromide stained rRNA. (B) RNA blot analysis
inserted as inverted-repeat sequences into a Mariner MosI of Sindbis-specific siRNA in uninfected C6/36 cells (mock) or in C6/36
transposable element used for germ line transformation. cells infected with TE3 2J virus. Small RNAs were obtained following
The inverted repeat sequences contain a 291 nt fragment of the protocol described by Hamilton and Baulcombe (1999). Low molec-
ular weight RNAs were fractionated by 12% urea-PAGE, blotted and hy-
prM region in sense and antisense orientations separated by bridized with 32 P-labeled sense or antisense RNA probe complementary
a major intron of the A. aegypti Sialokinin I gene. Obvi- to E1 gene region of Sindbis RNA. The positions of the 10, 20 and 30
ously, much more work needs to be done to use this strategy nucleotide markers are shown at the left.
72 I. Sanchez-Vargas et al. / Virus Research 102 (2004) 65–74
(Myles et al., 2003; Scott et al., 1984), suggesting that the
virus may replicate in and escape the midgut before the
RNAi response is underway. Arboviruses also could limit the
number of dsRNA replication forms in an infected mosquito
cell. Perhaps a threshold level of dsRNA is needed to trigger
RNAi. A corollary to this is that some arboviruses appear
to sequester their dsRNA replicative forms from the cellular
RNAi machinery (Uchil and Satchidanandam, 2003) sug-
gesting a long evolutionary history of interaction between
arboviruses and host dsRNA-mediated defense pathways.
Viruses also may encode RNAi suppressors to over-
come the host defense. Examples of plant RNA and DNA
viruses encoding suppressors to down-regulate RNAi have
now been described that presumably give the viruses an
advantage during their replication and spread in the plant
host (Llave et al., 2000). As noted earlier, at least one ex-
ample has been described of an insect virus (FHV) that
suppresses RNAi in Drosophila S2 cells (Li et al., 2002a).
Many virus proteins are multifunctional and some arboviral
proteins may have RNAi suppressor activity. This will be
an important area of investigation in the future. Finally,
plant geneticists have noted that developmental mutants of
plants with defined mutations in the Argonaute 1 protein
are highly susceptible to virus infection, perhaps because
the PTGS or RNAi pathway is defective (Williams and
Rubin, 2002). This suggests that a functional RNAi pathway
is essential in many organisms to counteract RNA virus in-
fection, and malfunctioning genes in the pathway could lead
to increased virus replication and pathology. It will be in-
teresting to see whether vector competence for arboviruses
can be associated with mutant gene products in the RNAi
pathway. We believe that further investigations of RNAi and
other RNA-mediated silencing pathways in mosquitoes will
be critical to the understanding of virus-vector interactions,
vector competence, translational control of gene expression,
and vector developmental biology.
Acknowledgements
This worked was supported by the National Institute of
Allergy and Infectious Diseases, grants AI34014, AI48740,
TW1130, and the John T. and Catherine D. MacArthur
Foundation.
References
Adams, M.D., Celniker, S.E., Holt, R.A., Evans, C.A., Gocayne, J.D.,
Amanatides, P.G., Scherer, S.E., Li, P.W., Hoskins, R.A., Galle, R.F.,
George, R.A., Lewis, S.E., Richards, S., Ashburner, M., Henderson,
S.N., Sutton, G.G., Wortman, J.R., Yandell, M.D., Zhang, Q., Chen,
L.X., Brandon, R.C., Rogers, Y.H., Blazej, R.G., Champe, M., Pfeif-
fer, B.D., Wan, K.H., Doyle, C., Baxter, E.G., Helt, G., Nelson, C.R.,
Gabor, G.L., Abril, J.F., Agbayani, A., An, H.J., Andrews-Pfannkoch,
C., Baldwin, D., Ballew, R.M., Basu, A., Baxendale, J., Bayraktaroglu,
L., Beasley, E.M., Beeson, K.Y., Benos, P.V., Berman, B.P., Bhandari,
D., Bolshakov, S., Borkova, D., Botchan, M.R., Bouck, J., Brokstein,
P., Brottier, P., Burtis, K.C., Busam, D.A., Butler, H., Cadieu, E.,
Center, A., Chandra, I., Cherry, J.M., Cawley, S., Dahlke, C., Dav-
enport, L.B., Davies, P., de Pablos, B., Delcher, A., Deng, Z., Mays,
A.D., Dew, I., Dietz, S.M., Dodson, K., Doup, L.E., Downes, M.,
Dugan-Rocha, S., Dunkov, B.C., Dunn, P., Durbin, K.J., Evangelista,
C.C., Ferraz, C., Ferriera, S., Fleischmann, W., Fosler, C., Gabrielian,
A.E., Garg, N.S., Gelbart, W.M., Glasser, K., Glodek, A., Gong, F.,
Gorrell, J.H., Gu, Z., Guan, P., Harris, M., Harris, N.L., Harvey, D.,
Heiman, T.J., Hernandez, J.R., Houck, J., Hostin, D., Houston, K.A.,
Howland, T.J., Wei, M.H., Ibegwam, C., et al., 2000. The genome
sequence of Drosophila melanogaster. Science 287, 2185–2195.
Adelman, Z.N., Blair, C.D., Carlson, J.O., Beaty, B.J., Olson, K.E., 2001.
Sindbis virus-induced silencing of dengue viruses in mosquitoes. In-
sect. Mol. Biol. 10, 265–273.
Adelman, Z.N., Sanchez-Vargas, I., Travanty, E.A., Carlson, J.O., Beaty,
B.J., Blair, C.D., Olson, K.E., 2002. RNA silencing of dengue virus
type 2 replication in transformed C6/36 mosquito cells transcribing
an inverted-repeat RNA derived from the virus genome. J. Virol. 76,
12925–12933.
Angell, S.M., Baulcombe, D.C., 1997. Consistent gene silencing in trans-
genic plants expressing a replicating potato virus X RNA. EMBO J.
16, 3675–3684.
Beerntsen, B.T., James, A.A., Christensen, B.M., 2000. Genetics of
mosquito vector competence. Microbiol. Mol. Biol. Rev. 64, 115–137.
Bennett, K.E., Olson, K.E., Munoz Mde, L., Fernandez-Salas, I.,
Farfan-Ale, J.A., Higgs, S., Black 4th, W.C., Beaty, B.J., 2002. Varia-
tion in vector competence for dengue 2 virus among 24 collections of
Aedes aegypti from Mexico and the United States. Am. J. Trop. Med.
Hyg. 67, 85–92.
Bernstein, E., Denli, A.M., Hannon, G.J., 2001. The rest is silence. RNA
7, 1509–1521.
Black 4th, W.C., Bennett, K.E., Gorrochotegui-Escalante, N.,
Barillas-Mury, C.V., Fernandez-Salas, I., de Lourdes Munoz, M.,
Farfan-Ale, J.A., Olson, K.E., Beaty, B.J., 2002. Flavivirus suscepti-
bility in Aedes aegypti. Arch. Med. Res. 33, 379–388.
Blair, C.D., Adelman, Z.N., Olson, K.E., 2000. Molecular strategies for
interrupting arthropod-borne virus transmission by mosquitoes. Clin.
Microbiol. Rev. 13, 651–661.
Caplen, N.J., Zheng, Z., Falgout, B., Morgan, R.A., 2002. Inhibi-
tion of viral gene expression and replication in mosquito cells by
dsRNA-triggered RNA interference. Mol. Ther. 6, 243–251.
Carthew, R.W., 2001. Gene silencing by double-stranded RNA. Curr.
Opin. Cell Biol. 13, 244–248.
Chambers, T.J., Hahn, C.S., Galler, R., Rice, C.M., 1990. Flavivirus
genome organization, expression, and replication. Annu. Rev. Micro-
biol. 44, 649–688.
Chandler, L.J., Blair, C.D., Beaty, B.J., 1998. La Crosse virus infection
of Aedes triseriatus (Diptera: Culicidae) ovaries before dissemination
of virus from the midgut. J. Med. Entomol. 35, 567–572.
Chu, P.W., Westaway, E.G., 1985. Replication strategy of Kunjin virus:
evidence for recycling role of replicative form RNA as template in
semiconservative and asymmetric replication. Virology 140, 68–79.
Elbashir, S.M., Martinez, J., Patkaniowska, A., Lendeckel, W., Tuschl, T.,
2001. Functional anatomy of siRNAs for mediating efficient RNAi in
Drosophila melanogaster embryo lysate. EMBO J. 20, 6877–6888.
Gaines, P.J., Olson, K.E., Higgs, S., Powers, A.M., Beaty, B.J., Blair,
C.D., 1996. Pathogen-derived resistance to dengue type 2 virus in
mosquito cells by expression of the premembrane coding region of
the viral genome. J. Virol. 70, 2132–2137.
Grishok, A., Pasquinelli, A.E., Conte, D., Li, N., Parrish, S., Ha, I., Baillie,
D.L., Fire, A., Ruvkun, G., Mello, C.C., 2001. Genes and mechanisms
related to RNA interference regulate expression of the small temporal
RNAs that control C. elegans developmental timing. Cell 106, 23–34.
Gubler, D.J., 1998. Dengue and dengue hemorrhagic fever. Clin. Micro-
biol. Rev. 11, 480–496.
 I. Sanchez-Vargas et al. / Virus Research 102 (2004) 65–74
Halstead, S.B., 1989. Antibody, macrophages, dengue virus infection,
shock, and hemorrhage: a pathogenic cascade. Rev. Infect. Dis. 11,
S830–839.
Hamilton, A.J., Baulcombe, D.C., 1999. A species of small antisense RNA
in posttranscriptional gene silencing in plants. Science 286, 950–952.
Hammond, S.M., Boettcher, S., Caudy, A.A., Kobayashi, R., Hannon,
G.J., 2001a. Argonaute 2, a link between genetic and biochemical
analyses of RNAi. Science 293, 1146–1150.
Hammond, S.M., Caudy, A.A., Hannon, G.J., 2001b. Post-transcriptional
gene silencing by double-stranded RNA. Nat. Rev. Genet. 2, 110–119.
Hardy, W.R., Strauss, J.H., 1988. Processing the nonstructural polyproteins
of Sindbis virus: study of the kinetics in vivo by using monospecific
antibodies. J. Virol. 62, 998–1007.
Hoa, N.T., Keene, K.M., Olson, K.E., Zheng, L., 2003. Characterization of
RNA interference in an Anopheles gambiae cell line. Insect. Biochem.
Mol. Biol. 33, 949–957.
Holt, R.A., Subramanian, G.M., Halpern, A., Sutton, G.G., Charlab, R.,
Nusskern, D.R., Wincker, P., Clark, A.G., Ribeiro, J.M., Wides, R.,
Salzberg, S.L., Loftus, B., Yandell, M., Majoros, W.H., Rusch, D.B.,
Lai, Z., Kraft, C.L., Abril, J.F., Anthouard, V., Arensburger, P., Atkin-
son, P.W., Baden, H., de Berardinis, V., Baldwin, D., Benes, V.,
Biedler, J., Blass, C., Bolanos, R., Boscus, D., Barnstead, M., Cai,
S., Center, A., Chaturverdi, K., Christophides, G.K., Chrystal, M.A.,
Clamp, M., Cravchik, A., Curwen, V., Dana, A., Delcher, A., Dew, I.,
Evans, C.A., Flanigan, M., Grundschober-Freimoser, A., Friedli, L.,
Gu, Z., Guan, P., Guigo, R., Hillenmeyer, M.E., Hladun, S.L., Hogan,
J.R., Hong, Y.S., Hoover, J., Jaillon, O., Ke, Z., Kodira, C., Kokoza,
E., Koutsos, A., Letunic, I., Levitsky, A., Liang, Y., Lin, J.J., Lobo,
N.F., Lopez, J.R., Malek, J.A., McIntosh, T.C., Meister, S., Miller, J.,
Mobarry, C., Mongin, E., Murphy, S.D., O’Brochta, D.A., Pfannkoch,
C., Qi, R., Regier, M.A., Remington, K., Shao, H., Sharakhova, M.V.,
Sitter, C.D., Shetty, J., Smith, T.J., Strong, R., Sun, J., Thomasova,
D., Ton, L.Q., Topalis, P., Tu, Z., Unger, M.F., Walenz, B., Wang,
A., Wang, J., Wang, M., Wang, X., Woodford, K.J., Wortman, J.R.,
Wu, M., Yao, A., Zdobnov, E.M., Zhang, H., Zhao, Q., et al., 2002.
The genome sequence of the malaria mosquito Anopheles gambiae.
Science 298, 129–149.
Hutvagner, G., McLachlan, J., Pasquinelli, A.E., Balint, E., Tuschl, T.,
Zamore, P.D., 2001. A cellular function for the RNA-interference
enzyme Dicer in the maturation of the let-7 small temporal RNA.
Science 293, 834–838.
Hutvagner, G., Zamore, P.D., 2002. A microRNA in a multiple-turnover
RNAi enzyme complex. Science 297, 2056–2060.
Johnson, B.W., Olson, K.E., Allen-Miura, T., Rayms-Keller, A., Carlson,
J.O., Coates, C.J., Jasinskiene, N., James, A.A., Beaty, B.J., Higgs, S.,
1999. Inhibition of luciferase expression in transgenic Aedes aegypti
mosquitoes by Sindbis virus expression of antisense luciferase RNA.
Proc. Natl. Acad. Sci. USA 96, 13399–13403.
Ketting, R.F., Fischer, S.E., Bernstein, E., Sijen, T., Hannon, G.J., Plasterk,
R.H., 2001. Dicer functions in RNA interference and in synthesis of
small RNA involved in developmental timing in C. elegans. Genes
Dev. 15, 2654–2659.
Knight, S.W., Bass, B.L., 2001. A role for the RNase III enzyme DCR-1
in RNA interference and germ line development in Caenorhabditis
elegans. Science 293, 2269–2271.
Kumagai, M.H., Donson, J., della-Cioppa, G., Harvey, D., Hanley, K.,
Grill, L.K., 1995. Cytoplasmic inhibition of carotenoid biosynthesis
with virus-derived RNA. Proc. Natl. Acad. Sci. USA 92, 1679–1683.
Li, H., Li, W.X., Ding, S.W., 2002a. Induction and suppression of RNA
silencing by an animal virus. Science 296, 1319–1321.
Li, X., Scuderi, A., Letsou, A., Virshup, D.M., 2002b. B56-associated
protein phosphatase 2A is required for survival and protects from
apoptosis in Drosophila melanogaster. Mol. Cell Biol. 22, 3674–3684.
Lindbo, J.A., Silva-Rosales, L., Proebsting, W.M., Dougherty, W.G., 1993.
Induction of a highly specific antiviral state in transgenic plants:
implications for regulation of gene expression and virus resistance.
Plant. Cell 5, 1749–1759.
73
Lipardi, C., Wei, Q., Paterson, B.M., 2001. RNAi as random degradative
PCR: siRNA primers convert mRNA into dsRNAs that are degraded
to generate new siRNAs. Cell 107, 297–307.
Llave, C., Kasschau, K.D., Carrington, J.C., 2000. Virus-encoded suppres-
sor of posttranscriptional gene silencing targets a maintenance step in
the silencing pathway. Proc. Natl. Acad. Sci. USA 97, 13401–13406.
Martinez, J., Patkaniowska, A., Urlaub, H., Luhrmann, R., Tuschl, T.,
2002. Single-stranded antisense siRNAs guide target RNA cleavage
in RNAi. Cell 110, 563–574.
Monath, T.P., 1994. Dengue: the risk to developed and developing coun-
tries. Proc. Natl. Acad. Sci. USA 91, 2395–2400.
Moreira, L.A., Edwards, M.J., Adhami, F., Jasinskiene, N., James, A.A.,
Jacobs-Lorena, M., 2000. Robust gut-specific gene expression in trans-
genic Aedes aegypti mosquitoes. Proc. Natl. Acad. Sci. USA 97,
10895–10898.
Myles, K.M., Pierro, D.J., Olson, K.E., 2003. A comparison of the trans-
mission potential of two genetically distinct Sindbis viruses following
per os infetion of Aedes aegypti (Diptera: Culicidae). J. Med. Ento-
mol. 41, 95–106.
Olson, K.E., Adelman, Z.N., Travanty, E.A., Sanchez-Vargas, I., Beaty,
B.J., Blair, C.D., 2002. Developing arbovirus resistance in mosquitoes.
Insect Biochem. Mol. Biol. 32, 1333–1343.
Olson, K.E., Higgs, S., Gaines, P.J., Powers, A.M., Davis, B.S., Kam-
rud, K.I., Carlson, J.O., Blair, C.D., Beaty, B.J., 1996. Genetically
engineered resistance to dengue-2 virus transmission in mosquitoes.
Science 272, 884–886.
Ratcliff, F.G., MacFarlane, S.A., Baulcombe, D.C., 1999. Gene silencing
without DNA. RNA-mediated cross-protection between viruses. Plant
Cell 11, 1207–1216.
Ribeiro, J.M., Kidwell, M.G., 1994. Transposable elements as population
drive mechanisms: specification of critical parameter values. J. Med.
Entomol. 31, 10–16.
Roignant, J.Y., Carre, C., Mugat, B., Szymczak, D., Lepesant, J.A., An-
toniewski, C., 2003. Absence of transitive and systemic pathways al-
lows cell-specific and isoform-specific RNAi in Drosophila. RNA 9,
299–308.
Ruiz, M.T., Voinnet, O., Baulcombe, D.C., 1998. Initiation and mainte-
nance of virus-induced gene silencing. Plant Cell 10, 937–946.
Scott, T.W., Burrage, T.G., 1984. Rapid infection of salivary glands in
Culiseta melanura with eastern equine encephalitis virus: an electron
microscopic study. Am. J. Trop. Med. Hyg. 33, 961–964.
Scott, T.W., Hildreth, S.W., Beaty, B.J., 1984. The distribution and devel-
opment of eastern equine encephalitis virus in its enzootic mosquito
vector, Culiseta melanura. Am. J. Trop. Med. Hyg. 33, 300–310.
Shiao, S.H., Higgs, S., Adelman, Z., Christensen, B.M., Liu, S.H., Chen,
C.C., 2001. Effect of prophenoloxidase expression knockout on the
melanization of microfilariae in the mosquito Armigeres subalbatus.
Insect Mol. Biol. 10, 315–321.
Sijen, T., Fleenor, J., Simmer, F., Thijssen, K.L., Parrish, S., Timmons,
L., Plasterk, R.H., Fire, A., 2001. On the role of RNA amplification
in dsRNA-triggered gene silencing. Cell 107, 465–476.
Silhavy, D., Molnar, A., Lucioli, A., Szittya, G., Hornyik, C., Tavazza,
M., Burgyan, J., 2002. A viral protein suppresses RNA silencing
and binds silencing-generated, 21- to 25-nucleotide double-stranded
RNAs. EMBO J. 21, 3070–3080.
Song, E., Lee, S.K., Wang, J., Ince, N., Ouyang, N., Min, J., Chen,
J., Shankar, P., Lieberman, J., 2003. RNA interference targeting Fas
protects mice from fulminant hepatitis. Nat. Med. 9, 347–351.
Tabara, H., Yigit, E., Siomi, H., Mello, C.C., 2002. The dsRNA bind-
ing protein RDE-4 interacts with RDE-1, DCR-1, and a DExH-box
helicase to direct RNAi in C. elegans. Cell 109, 861–871.
Uchil, P.D., Satchidanandam, V., 2003. Characterization of RNA synthesis,
replication mechanism, and in vitro RNA-dependent RNA polymerase
activity of Japanese encephalitis virus. Virology 307, 358–371.
Vance, V., Vaucheret, H., 2001. RNA silencing in plants–defense and
counterdefense. Science 292, 2277–2280.
74 I. Sanchez-Vargas et al. / Virus Research 102 (2004) 65–74

Westaway, E.G., Mackenzie, J.M., Kenney, M.T., Jones, M.K., Khromykh,
A.A., 1997. Ultrastructure of Kunjin virus-infected cells: colocalization
of NS1 and NS3 with double-stranded RNA, and of NS2B with NS3,
in virus-induced membrane structures. J. Virol. 71, 6650–6661.
Williams, R.W., Rubin, G.M., 2002. ARGONAUTE1 is required for ef-
ficient RNA interference in Drosophila embryos. Proc. Natl. Acad.
Sci. USA 99, 6889–6894.
Woodring, J.I., Higgs, S., Beaty, B.J., 1996. Natural cycles of vector-borne
pathogens. In: Marquardt, W.C. (Ed.), The Biology of Disease Vectors.
Universitiy of Colorado Press, Niwot, CO, pp. 51–72.
Zamore, P.D., Tuschl, T., Sharp, P.A., Bartel, D.P., 2000. RNAi:
double-stranded RNA directs the ATP-dependent cleavage of mRNA
at 21–23 nucleotide intervals. Cell 101, 25–33.
 Virus Research 102 (2004) 75–84

Silencing structural and nonstructural genes in

baculovirus by RNA interference
C. Fabian Flores-Jasso a , Victor Julian Valdes a , Alicia Sampieri a ,
Viviana Valadez-Graham b , Felix Recillas-Targa b , Luis Vaca a,∗
a Departamento de Biologı́a Celular, Instituto de Fisiologı́a Celular, UNAM, Ciudad Universitaria, Mexico, D.F. 04510, Mexico
b Departamento de Genética Molecular, UNAM, Mexico, D.F. 04510, Mexico

Abstract

We review several aspects of RNAi and gene silencing with baculovirus. We show that the potency of RNAi in Spodoptera frugiperda
(Sf21) insect cells correlates well with the efficiency of transfection of the siRNA. Using a fluorescein-labeled siRNA we found that the siRNA
localized in areas surrounding the endoplasmic reticulum (ER). Both long (700 nucleotides long) and small (∼25 nucleotides long) interfering
RNAs were equally effective in initiating RNA interference (RNAi), and the duration of the interfering effect was indistinguishable. Even
though RNAi in Sf21 cells is very effective, in vitro experiments show that these cells fragment the long dsRNA into siRNA poorly, when
compared to HEK cells.
Finally, we show that in vivo inhibition of baculovirus infection with dsRNA homologous to genes that are essential for baculovirus infectivity
depends strongly on the amount of dsRNA used in the assays. Five hundred nanogram of dsRNA directly injected into the haemolymph of
insects prevent animal death to over 95%. In control experiments, over 96% of insects not injected with dsRNA or injected with an irrelevant
dsRNA died within a week. These results demonstrate the efficiency of dsRNA for in vivo prevention of a viral infection by virus that is very
cytotoxic and lytic in animals.
© 2004 Elsevier B.V. All rights reserved.

Keywords: Systemic RNAi; Double stranded RNA; siRNA; Gene silencing; Baculovirus; Viral infection; In vivo inhibition

1. Introduction is a reality today. With RNAi one can knockdown as many

Introduction of double stranded RNA (dsRNA) into a wide
variety of cells and organisms results in depletion of the
homologue endogenous messenger RNA (mRNA) (Caplen
et al., 2000; Chuang and Meyerowitz, 2000; Fire et al., 1998;
Romano and Macino, 1992; Wianny and Zernicka-Goetz,
2000).
This phenomenon now known as RNA interference or
RNAi in animals is related to the post transcriptional gene
silencing (PTGS) phenomena first discovered in plants as
co-suppression (Napoli et al., 1990; Tabara, 2001; Vaucheret
et al., 2001).
RNAi is rapidly becoming a powerful tool for gene silenc-
ing (Voorhoeve and Agami, 2003). What only a few years
ago was the dream of many cell biologists, namely to have a
way to explore the function of several genes simultaneously,
∗ Corresponding author. Tel.: +52-5-622-5654; fax: +52-5-622-5611.
E-mail address: lvaca@ifc.unam.mx (L. Vaca).
genes as one desires efficiently and selectively in cell cul-
tures and even in living organisms. The most dramatic exam-
ple was provided with the nematode Caenorhabditis elegans,
where the term “genome-wide RNAi” has taken gene si-
lencing technology to a new level (Maeda et al., 2001).
Even though the mechanism by which RNAi operates is
not yet completely understood, many researchers are taking
advantage of this phenomenon in different ways. The rapid
production of knockdown animals where genes of interest
have been selectively silenced is just one of many possible
uses for RNAi (Voorhoeve and Agami, 2003).
Another obvious use for RNAi is the selective silencing of
viral genes essential for virulence (Lawrence, 2002; Shuey
et al., 2002). This objective is particularly important given
the lack of good antiviral agents in the face of a large num-
ber of viral diseases afflicting humans. RNAi cannot only
be used as a possible antiviral therapy, but also as a pow-
erful tool to explore and better understand the basic aspects
of viral entry into host cells, disassembly, replication, par-
ticle assembly and export out of the infected cell. This tool

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.018
76 C.F. Flores-Jasso et al. / Virus Research 102 (2004) 75–84
allows exploring simultaneously the role of several genes in
viral function, especially in viruses with large and complex
genomes containing many putative proteins with unknown
function.
In the present review we will describe some of the re-
cent advances in RNAi using the baculovirus as a model
system. We will also speculate about possible uses of RNAi
combined with baculovirus technologies to produce power-
ful tools for gene function exploration.
2. The Autographa californica nucleopolyhedrosis virus
Baculoviruses comprise a large family of viruses which
naturally infect many different insect species (Jackson,
1991). This type of virus has been the cause of severe eco-
nomic loses in the silk industry over many years. Literature
published in the 16th century contains detailed descriptions
of the “wilting” disease in silkworms. A direct correlation
between wilting disease and baculovirus infection was not
found until the beginning of the 19th century with the
discovery of small crystalline bodies known as polyhedra,
obtained from infected silkworms, and the identification of
rod-shaped virions contained inside this structure.
By far, the best studied member of this family is the A.
californica nucleopolyhedrovirus (AcNPV). This virus has
a large double stranded DNA genome with over 135 open
reading frames (Ayres et al., 1994). The functions of most of
the protein products encoded by this genome are unknown.
Virus morphogenesis is a rather complex process involv-
ing early, late and very late events (Yamagishi et al., 2003).
Some of the virions will bud out of the cell to infect adja-
cent cells, while others will be occluded into a protein ma-
trix named polyhedra (Summers and Volkman, 1976). The
polyhedra are formed of the aggregates of polyhedrin, a
highly expressed protein produced during the late phase of
baculovirus replication. Polyhedrin production is driven by
the late and very strong polyhedrin promoter. This promoter
interacts selectively with baculovirus ␣-amanitin-resistant
RNA polymerase. Occluded virions (contained inside the
polyhedra) will leave the cell after cell lysis. In the occluded
form, baculovirus is extremely stable and may maintain in-
fectivity for several years at room temperature. There are
many unanswered questions regarding baculovirus morpho-
genesis, for instance, the mechanism by which viruses are
selected to be occluded or budded out of the infected cell.
The sorting system that operates to determine the fate of viri-
ons remains unclear, and the use of RNAi may help to elu-
cidate this and other intriguing questions about baculovirus
morphogenesis in the future.
One of the most common systems used to study Ac-
NPV are cells originally isolated from larvae of the insect
Spodoptera frugiperda (Sf21 and Sf9 cells). These cell lines
are extremely susceptible to AcNPV infection and in a few
days become highly lysed by the cytolytic viral effect. Many
groups have taken advantage of this kind of susceptibil-
ity and have produced recombinant baculovirus expressing
exogenous proteins from a wide variety of organisms. As
mentioned above, the polyhedrin gene is driven by its po-
tent promoter and thus, one can replace polyhedrin by the
gene of interest, resulting in the production of large amounts
of protein (Luckow and Summers, 1988). This protein pro-
duction system facilitates protein purification for functional
studies. Many proteins of different origins, including solu-
ble and plasma membrane proteins have been produced with
this system. The proteins produced in this form are func-
tional in the insect cell, an advantage that has positioned this
system as an excellent option to express and study the func-
tion of many G protein-coupled receptors and ionic channels
(Massotte, 2003).
Recently, new applications have been described for re-
combinant or modified baculovirus. Special interest has been
directed to the use of recombinant baculovirus as gene ther-
apy vectors (Pieroni and La Monica, 2001). It has been
shown that modified baculovirus containing suitable mam-
malian promoters can efficiently transfer genetic material
into mammalian cells from different tissues (Hu et al., 2003).
Another recent use for baculovirus is in the display of
peptides and proteins of interest. Because the modification
of the most abundant glycoprotein in the baculovirus nucle-
ocapsid (gp64) is relatively easy, several groups have used
recombinant baculovirus to display proteins and peptides
fused to GP64 (Grabherr et al., 2001; Ojala et al., 2001).
Thus, baculovirus display has recently emerged as a new
powerful technique to identify ligands and in the presenta-
tion of antigens (Ojala et al., 2001).
Our group has explored recombinant baculovirus in com-
bination with RNAi technology to test two important con-
cepts: (a) that RNAi is a feasible alternative to classic reverse
genetics with baculovirus and (2) that interfering with genes
essential for virulence prevents viral infections in vivo.
Recently we have shown for the first time that injecting an-
imals with dsRNA corresponding to genes essential for bac-
ulovirus infection prevents insect death by over 95%.(Valdes
et al., 2003). The surviving insects do not show activity of
the reporter genes carried by the recombinant baculovirus,
indicating that virus replication was inhibited by the dsRNA
treatment (Valdes et al., 2003). We have explored the effect
of dsRNA on two genes essential for baculovirus infectiv-
ity. The first gene is gp64, the major protein found in bac-
ulovirus nucleocapsid, and the second gene is ie1, an early
transcriptional activator essential for baculovirus replication.
We show that RNAi is very efficient and selective in si-
lencing structural (gp64) and nonstructural (ie1) baculovirus
genes.
3. RNAi is very potent and selective in insect cells
Introduction of dsRNA containing a sequence from ie1
into Sf21 cells and living insects protects against baculovirus
infection. Surprisingly, the protective effect of dsRNA was
 C.F. Flores-Jasso et al. / Virus Research 102 (2004) 75–84 77

very potent in both, cell cultures and whole organisms. Over
95% of Sf21 cells transfected with 5 ␮g of dsRNA corre-
sponding to the ie1 gene were not infected with recombinant
baculovirus, even at multiplicity of infections (MOI) of five
and higher (Valdes et al., 2003). Furthermore, larvae from
the insect Tenebrio mollitor were systemically protected with
as few as 0.5 ␮g of dsRNA that was directly injected in the
haemolymph (Valdes et al., 2003). This is the first report
showing systemic protection against a viral infection in an
animal by the injection of nude dsRNA.
The excellent protection conferred by the dsRNA in cell
culture was most likely the result of the majority of the Sf21
insect cells receiving dsRNA. To determine the transfection
efficiency in insect cells, we used a siRNA (21 nucleotides
long) conjugated with fluorescein to assess cell fluorescence
after transfection. Fig. 1A illustrates that the majority of

Fig. 1. Distribution of siRNA inside Sf21 insect cells. Sf21 insect cells were transfected with 0.5 ␮M of siRNA-fluorescein (QIAGEN, Hilden, Germany)
using cellfectin (INVITROGEN, Carlsbad, CA) and maintained in culture for several days as previously described (Valdes et al., 2003). Cells were
mounted on glass coverslips and observed with a confocal microscope (BIORAD, Hercules, CA). (A) Cells observed under low magnification to illustrate
that most of cells show red (brefeldin A) and green (siRNA) labeling. White arrow heads point to cells showing co-localization of brefeldin A and siRNA
(indicated by yellow color). (B) Magnification of an area in (A) to illustrate co-localization of brefeldin A and siRNA label. (C) Three-dimensional
reconstruction of two cells obtained a week after siRNA transfection showing siRNA distribution. Notice that the nuclei (Ni) of both cells do not contain
siRNA. (D) 90◦ rotation of the image in (C) to illustrate the siRNA distribution inside the cell. White arrows in (D) show cell limits. The scale bar
represents 50 ␮s for panel (A), 20 ␮s for (B and C) and 10 ␮s for (D). In all cases, cells were incubated for 5 min with 2 ␮M brefeldin A BODIPY
558/568 (MOLECULAR PROBES, Eugene, OR).
78 C.F. Flores-Jasso et al. / Virus Research 102 (2004) 75–84

insect cells transfected with the fluorescein-labeled siRNA
fluoresce (shown in green). The transfected siRNA was ob-
served in well defined areas surrounding the endoplasmic
reticulum as illustrated in Fig. 1B. In this series of exper-
iments we used two selective markers of the endoplasmic
reticulum, the first one was brefeldin A (Nebenfuhr et al.,
2002) conjugated with BODIPY (illustrated in the figure
in red), and the second selective marker was thapsigargin
(Inesi and Sagara, 1992), a specific inhibitor of the microso-
mal calcium ATPase (data not shown). These results showed
that the siRNA fluorescence was absent from the cell nuclei
(Fig. 1C) and was not found dispersed throughout the cy-
tosol (Fig. 1D), but only in areas surrounding ER. In several
cells we observed co localization of ER and siRNA fluo-
rescence (Fig. 1A, arrow heads). These results indicate that
siRNA is concentrated near the ER and does not cross the
nuclear membrane in Sf21 cells.
A recent study has shown that Dicer, a type III RNase re-
sponsible for fragmenting long dsRNA into small interfering
RNA (siRNA), is only present in the cytosol in an area that

Fig. 2. Efficient, long lasting RNAi in Sf21 insect cells. (A) GFP-positive cells measured with a cell sorter as previously described (Valdes et al.,
2003). No virus (NV) indicate uninfected cells (cell auto fluorescence), and cells transfected with dsRNA–gfp and infected with recombinant baculovirus
carrying GFP gene on days 2, 4, 6, and 9 after transfection. In all cases the GFP fluorescence was measured 48 h after infection with baculovirus, and
only transfection times varied. Cells infected with recombinant baculovirus and not exposed to dsRNA–gfp illustrate the maximum GFP fluorescence
obtained (no dsRNA). (B) Percentage of GFP-positive cells (mean ± standard deviation) obtained from 10,000 events measured with the cell sorter
(FACScalibur; Becton Dickinson). The percentage was calculated from the gray and white rectangles shown in (A) which separate GFP-positive from
GFP-negative cells (auto-fluorescence from uninfected cells). (C) Representative examples comparing the RNAi potency obtained with dsRNA–gfp (700
nucleotides) and siRNA–gfp (∼25 nucleotides) on the GFP production by recombinant baculovirus. Cells were infected with baculovirus 5 days after
transfection of the RNA. Infection was carried out in all cases with MOI of 1. Notice that both 700 nucleotide long dsRNA–gfp and the 25 nucleotide
long siRNA–gfp were equally efficient, inhibiting approximately 60% of GFP fluorescence. The mean of two independent observations is shown. All
cytometer measurements were performed as previously described (Valdes et al., 2003).
 C.F. Flores-Jasso et al. / Virus Research 102 (2004) 75–84
correlates well with markers of the ER (Provost et al., 2002).
It remains to be explored whether the well defined spots
observed in Sf21 cells transfected with fluorescein-labeled
siRNA co-localize with Dicer.
Fluorescence had decayed only moderately a week after
siRNA transfection into Sf21 cells. We could not determine
whether the fluorescence decay observed after a week of
transfection reflected siRNA degradation or simply fluores-
cein bleaching. Since in these experiments we used siRNA
formed of 21 nucleotides, the data suggest that small inter-
fering RNA (siRNA) may be stable inside insect cells for
over a week. Similar results have been observed in mam-
malian cells, where siRNA protected against viral infection
for several days with poliovirus (Gitlin et al., 2002), hep-
atitis C virus (Kapadia et al., 2003) and influenza virus (Ge
et al., 2003). The stability of the siRNA, combined with the
efficient transfection observed in the experiments reported
here, may explain the potent RNAi we have previously pub-
lished with insects (Valdes et al., 2003).
Although the fluorescence experiments suggested that
siRNAs are stable for over a week inside insect cells, it was
necessary to determine whether the RNAi effects driven by
siRNA could also be observed one week after transfection.
To determine if the siRNA could silence genes a week af-
ter transfection, insect cells were transfected with either long
dsRNA–gfp (700 nucleotides long) or the same dsRNA di-
gested in vitro by recombinant human Dicer to obtain siRNA
(21 nucleotides long). As illustrated in Fig. 2, interference
of GFP production by recombinant baculovirus carrying
the GFP gene was observed even 9 days after transfecting
the cells with dsRNA–GFP. Interference efficiency decayed
slowly in correlation with the number of days after transfec-
tion (Fig. 2B). Fifty percent GFP inhibition was obtained be-
tween days 6 and 9 post transfection. At day 9 only 30% GFP
inhibition was obtained. This apparently small inhibition is,
however, rather significant since we are not interfering with
baculovirus replication, but only with GFP production.
The results presented here indicate that the protective ef-
fect of dsRNA–gfp remained active after 6–9 days after
transfection. Similar results were obtained using small inter-
fering RNAs (siRNA–gfp). Both dsRNA (700 nt) and siRNA
(∼25 nt) were equally effective in preventing GFP produc-
tion in cells infected with recombinant baculovirus, as il-
lustrated in Fig. 2C. At day 5 post transfection, both long
dsRNA and siRNA prevented 60% of GFP fluorescence.
The long lasting effect of RNAi has not been extensively
explored in different cells. In the nematode C. elegans RNAi
can last for a couple of generations (Fire et al., 1998). This
lasting effect of RNAi is essentially explained by the fact that
this nematode has RNA-directed RNA polymerases (RdRP),
which produce amplification of the dsRNA originally intro-
duced (Nishikura, 2001). In plants the RdRP gene (SDE-1)
is required only for transgene-induced PTGS, but not for
virus-induced PTGS (Dalmay et al., 2000). EGO-1, the or-
thologue in C. elegans of SDE-1 is essential for RNAi, but
not in the germline (Smardon et al., 2000). RdRP activity has
79
been reported in Drosophila embryos (Lipardi et al., 2001);
on the other hand, have been reports that question the exis-
tence of a RdRP in Drosophila and mammals (Schwarz et al.,
2002; Stein et al., 2003). In fact, a single siRNA would be
sufficient for priming the entire sequence of the gene of in-
terest, thus producing full-length dsRNAs, which in turn can
enter a new dicing cycle to produce more siRNAs (Lipardi
et al., 2001). Whether Sf21 cells possess RdRP activity is
still an open question.
The duration of the fluorescence in cells transfected with
fluorescein-labeled siRNA correlates well with the persever-
ance of the RNAi effects on the inhibition of GFP production
by recombinant baculovirus. This observation alone might
be sufficient to explain the long lasting duration of RNAi in
insect cells, but one cannot discard the involvement of an
RdRP complex equivalent to the one reported in C. elegans.
Future experiments may help to identify RdRP activity in
Sf21 cells.
4. Dicer activity in insect and human cells
Dicer is a type III RNase essential for RNAi. RNases
III are endonucleases with high specificity for dsRNA
(Robertson et al., 1968). Dicer homologues have been iden-
tified in bacteria, yeast, insects, nematodes and mammals
(Blaszczyk et al., 2001). This protein is responsible for
cleaving (dicing) long dsRNA into small interfering dsRNA
(siRNA) fragments of ∼21–25 nucleotides long.
A large body of experimental evidence strongly sug-
gests that siRNA is the functional unit in charge of direct-
ing the degradation of the messenger RNA by the RISC
(RNA-induced silencing complex) (Ramaswamy and Slack,
2002). Therefore, the ability of different cells to produce
siRNA by fragmenting long dsRNA may correlate with
efficient RNAi.
To determine the Dicer activity in Sf21 insect cells, we
performed experiments using long dsRNAs (700 nucleotides
long) that were radiolabeled. This labeled dsRNA was used
with cell extracts obtained from Sf21 insect and human em-
bryonic kidney (HEK) cells in dicing assays. As illustrated
in Fig. 3, human cell extracts efficiently cleaved the long
dsRNA (700 nt long) into ∼21 nucleotide siRNA fragments
within 1 h. In comparison, Sf21 cell extracts showed re-
duced Dicer activity when compared to human cells, since
after 90 min only a small portion of the dsRNA had been
processed into siRNA fragments (Fig. 3). The Dicing ac-
tivity from cell extracts was compared with that of purified
human Dicer. As shown in Fig. 3, overnight incubation of
the dsRNA with purified human Dicer resulted in fragmen-
tation of the dsRNA into siRNA. These experiments were
performed to compare the molecular weight of the products
resulting from cleaving with purified Dicer and cell extracts.
As illustrated in the figure, HEK and Sf21 cell extracts
produced siRNA fragments of molecular weight equivalent
to the fragments produced by the purified enzyme. These
80 C.F. Flores-Jasso et al. / Virus Research 102 (2004) 75–84
Fig. 3. In vitro Dicer activity of Sf21 insect and human cells. In vitro RNA
cleavage experiments using extracts from Sf21 insect and HEK human
cells. Dicer reactions were stopped at 10, 30, 60, and 90 min as indicated
below. Recombinant human Dicer was used in overnight experiments.
Control (C) indicates reaction being carried out in the absence of cell
extracts or recombinant Dicer (Di) (INVITROGEN, Carlsbad, CA). MW
indicates molecular weight markers. The horizontal thick arrow points to
the 20 nucleotide molecular weight marker. Methods were in brief: Either
insect or human cells (∼3 × 107 ) were lysed with RIPA buffer (30 mM,
Tris–HCl; 50 mM, NaCl; 0.1%, NP40; 20%, Glycerol; 3 mM, MgCl2 ; pH
6.8 final) and a cocktail of protease inhibitors. Assays were done at a final
concentration of 3 ␮g/␮l protein and contained 2 ␮g of dsRNA (700 nt
long) internally labeled with [32 P]-UTP (∼2 × 105 cpm/␮g). The final
reaction volume was 20 ␮l. During experiments, Sf21 cells were incubated
at 27 ◦ C and HEK cells were incubated at 37 ◦ C. Phenol/chloroform (5 ␮l)
were added to stop reactions. Assays were separated on 10% PAGE-8M
urea. Gels were dried, exposed to an amplifying screen and analyzed with
a PhosphorImager (AMERSHAM BIOSCIENCES, Sunnyvale, CA).
observations indicate that the fragmentation of the long
dsRNA using the cells extracts result from the activity of
a type III RNase. Increasing the amount of purified Dicer
used in the assay or reducing the amount of dsRNA both re-
sulted in complete fragmentation of the substrates (data not
shown).
These results suggest that Sf21 cells are less efficient in
cleaving long dsRNA than HEK cells. There are several ex-
planations for this observation. For instance, it is possible
that the reduced Dicer activity of insect cell extracts may
result of inadequate conditions for the assay. Although we
explored different buffer solutions in which we changed pH,
magnesium, temperature and ATP contents, all yielded sim-
ilar results. However, we cannot discard the possibility that
the Dicer from insect cells may not be fully functional in the
buffers used for these experiments. Another possibility may
be that insect cells have less Dicer protein than HEK cells.
Ongoing experiments may help to elucidate this point.
As shown in the previous section, both long dsRNA and
siRNA were effective after 7 days or more post transfection,
suggesting that double stranded RNA may be relatively re-
sistant to degradation in insect cells. The mechanisms by
which siRNAs degrade are still poorly explored. It is possi-
ble that degradation may take place after the double strand
is disrupted by a helicase or other yet unidentified mecha-
nism. We have maintained successfully dsRNA for several
months at −20 ◦ C. Not surprisingly, single strand RNA is
much more labile under these storage conditions and de-
grades relatively fast. Another explanation for the dsRNA
stability in Sf21 cells is that insects lack the interferon re-
sponse characteristic of higher eukaryotes (Clemens, 1997;
de Haro et al., 1996; Katze, 1995; Tan and Katze, 1999).
Thus, the stability of dsRNA could be enhanced because no
apoptosis will be induced by the presence of long dsRNA
in the cytosol of Sf21 cells.
5. Protecting living animals from a highly cytotoxic
baculovirus with dsRNA
We have observed systemic RNAi effects in larvae from
the insect T. mollitor. Injecting as few as 0.5 ␮g of dsRNA
from ie1 directly into the haemolymph followed 24 h later
by the injection of recombinant baculovirus prevents insect
death in over 95% of the animals, as previously reported
(Valdes et al., 2003). Surviving insects do not show beta
galactosidase activity or GFP fluorescence, the two reporter
genes carried by the recombinant baculovirus.
We explored insect survival using different dsRNA con-
centrations and followed survival for up to 7 days. Fig. 4 il-
lustrates the results obtained with larvae injected with 0.05,
0.1, and 0.5 ␮g dsRNA–ie1 24 h prior to the challenge with
recombinant baculovirus. The dsRNA–ie1 was synthesized
from nucleotides 19-470 from the ie1 gene as previously
described (Valdes et al., 2003).
As illustrated in the figure, insect survival declined
when the animals were injected with reducing amounts
of dsRNA–ie1. Dead insects previously injected with
0.1 ␮g dsRNA–ie1 showed a lighter blue color and reduced
GFP fluorescence when compared to insects injected with
0.05 ␮g dsRNA–ie1. These results show that the expression
in infected animals of the two reporter genes carried by
the recombinant baculovirus (beta galactosidase and GFP)
correlates well with the survival rate. This observation is
more clearly illustrated in Fig. 4B, where the GFP fluo-
rescence intensity obtained from larvae injected with 0.05,
0.1, and 0.5 ␮g dsRNA–ie1 prior to baculovirus infection
is plotted. Notice the correlation between dead insects col-
lected at day 7 after baculovirus infection and increased
GFP fluorescence.
 C.F. Flores-Jasso et al. / Virus Research 102 (2004) 75–84 81

Fig. 4. In vivo protection of insect larvae from baculovirus infection by the use of dsRNA. Panel (A) larvae from the insect T. mollitor (mealworm
beetle) injected in the haemolymph with the 0.05, 0.1, and 0.5 ␮g of dsRNA–ie1 or 0.5 ␮g dsRNA–gfp 24 h prior to the injection of 2 ␮l of recombinant
baculovirus AcNPV–gfp (viral stock with a titer of 1 × 10−7 pfu). To the left representative examples of larvae injected with 2 ␮l of 0.05% X-gal in
dimethyl sulfoxide on day 7 after baculovirus injection are shown. Notice the blue color in larvae infected with recombinant baculovirus and not protected
by the dsRNA–ie1. Notice also the lighter blue color obtained from the larva injected with 0.1 ␮g dsRNA–ie1. The middle panel shows confocal images
illustrating the GFP fluorescence from the head of larvae under the different experimental conditions mentioned above. Notice the lack of blue color
(beta galactosidase activity) and GFP fluorescence in larvae injected with 0.5 ␮g dsRNA–ie1. The plots to the right indicate the percentage of dead
insects at different days post infection with baculovirus. The numbers in the inserts indicate the number of dead larvae and the total larvae injected
for each condition (e.g. 57 dead larvae from 60 injected with 0.05 ␮g dsRNA–ie1). Panel (B), the plot to the left illustrates the number of dead larvae
obtained at day 7 post infection for the three different dsRNA–ie1 concentrations assayed. The plot to the right shows the GFP fluorescence intensity (in
arbitrary units, AU) obtained at day 7 post infection from the head of randomly selected larvae, injected with 0.5, 0.1, and 0.05 ␮g dsRNA–ie1 (n = 10
for each condition). Solid circles indicate dead larvae and open circles living larvae. Autofluorescence was subtracted from all experiments. For details
on experimental conditions see Valdes et al. (2003).
82 C.F. Flores-Jasso et al. / Virus Research 102 (2004) 75–84
These results demonstrate that the protective effect of
dsRNA–ie1 is dose-dependent. When 0.05 ␮g dsRNA–ie1
is used, only 3 out of 60 (5%) larvae injected survived bac-
ulovirus infection (Fig. 4A, upper panel). All insects showed
beta galactosidase activity and high GFP fluorescence. The
amount of GFP fluorescence obtained from these animals
was indistinguishable from that obtained with larvae not in-
jected with dsRNA–ie1 (data not shown), suggesting that this
dosage of dsRNA is insufficient to protect insects from bac-
ulovirus infection. Injecting larvae with 0.5 ␮g dsRNA–gfp
prevented GFP fluorescence in infected animals but did not
alter insect survival to the baculovirus infection. Insects in-
jected with dsRNA–gfp turned blue upon X-gal injection, in-
dicating the presence of beta galactosidase in the baculovirus
infected animals (Fig. 4A). These results demonstrate that
the in vivo RNAi effects are selective for the dsRNA se-
quence used in the experiments, since we could interfere
with GFP but not beta galactosidase. The high mortality rate
observed in animals injected with dsRNA–gfp demonstrates
that baculovirus replication and infectivity were not affected
by irrelevant dsRNAs.
Since the dsRNA injection was into the haemolymph and
baculovirus infection in insects is systemic, the protective
effects of dsRNA–ie1 strongly suggest that the effects of
RNAi in these larvae are systemic. Systemic RNAi has been
observed in plants (Palauqui et al., 1997) and several animals
including C. elegans (Fire et al., 1998; Winston et al., 2002)
and mice (Sorensen et al., 2003). More recently it has been
shown systemic and inheritable RNAi in the beetle Tribolium
castaneum (Bucher et al., 2002).
6. Conclusions
We have used the baculovirus AcNPV to explore two
hypotheses: (a) that interfering with the synthesis of proteins
essential for baculovirus infectivity prevents in vitro and in
vivo viral infections and (b) that RNAi is a powerful tool to
explore the morphogenesis of a complex virus.
We have shown, for the first time, the use of RNAi to
selectively inhibit viral genes essential for baculovirus vir-
ulence. Furthermore, systemic application of dsRNA into a
living animal protected insects from infection by this very
infective and cytolytic virus. These results demonstrate the
potential of using dsRNA as a therapeutic drug in the treat-
ment or prevention of viral infections in animals.
6.1. Exploring the morphogenesis of a complex virus with
RNAi
Understanding the functional role of several genes from
a complex genome the size of the AcNPV genome is not an
easy task. The production of knockout virus is a lengthy pro-
cess. On many occasions, the gene deletions result in virus
that cannot be replicated and therefore cannot be studied.
The advantage of using RNAi is that one can have the wild
type virus or a recombinant virus, which are easy to obtain
and replicate, and use RNAi with insect cells to silence the
genes of interest. We have successfully silenced two genes
essential for infectivity in the AcNPV baculovirus, gp64 and
ie1.
There are several important lessons we have learned from
these experiments. A large body of evidence shows that
GP64 is a protein essential for virus entry into insect cells
(Robertson et al., 1974). Deletion of this protein resulted
in baculovirus that cannot infect cells and therefore cannot
be replicated (Monsma et al., 1996). Thus studying the role
of GP64 in baculovirus infectivity cannot be accomplished
with traditional knockout technologies. We have used RNAi
to produce viruses lacking GP64, by transfecting cells with
dsRNA formed of 300 nucleotides of gp64 (dsRNA–gp64)
following infection with recombinant baculovirus carrying a
wild type copy of gp64. Virus produced by cells transfected
with dsRNA–gp64 lacked this glycoprotein, as we have pre-
viously shown (Valdes et al., 2003). The virus produced by
these cells could not bud out of the insect cells. Virions were
identified inside the cells nuclei by electron microscopy, but
they could not be collected from the supernatant of infected
cells even at 96 h post infection. These results strongly sug-
gest that GP64 glycoprotein is not only needed for virus en-
try (as previously shown by Monsma et al., 1996), but also
for budding out of infected cells.
RNAi can also be used to identify novel proteins in the
genome of baculovirus. Recently, RNAi has been used
to identify a novel antiapoptotic gene (Op-iap3) from the
baculovirus Orgyia pseudotsugata M nucleopolyhedrovirus
(Means et al., 2003). Interfering with Op-iap3 resulted in
apoptosis in insect cells infected with this baculovirus, and
virus production was markedly reduced due to early cell
death (Means et al., 2003). Thus RNAi provides a rapid
and efficient method for the screening of protein function
in different members from the baculovirus family.
6.2. Baculovirus as RNAi vectors
Major efforts are directed worldwide to find a suitable
vector to induce RNAi in living organisms. Researchers
have injected nude siRNA or combined with lipids in mice
(Sorensen et al., 2003). However, it has been difficult to
obtain systemic distribution of the siRNA. More recently,
several groups have developed viral vectors based on aden-
ovirus (Shen et al., 2003) and lentivirus (Stewart et al., 2003)
carrying a polymerase III promoter (such as H1 or U6) and
a hairpin sequence to produce siRNAs. Although both aden-
ovirus and lentivirus are excellent vectors to transfer genetic
material in mammalian cells, the production of these vec-
tors is time consuming, making the exploration of several
siRNA sequences difficult. On the other hand, it is well es-
tablished that baculovirus can transfer genetic material into
mammalian cells (Hu et al., 2003), yet this virus is relatively
easy to prepare, and production of high titer viral stocks is
achieved in a short period of time. Thus, the development
 C.F. Flores-Jasso et al. / Virus Research 102 (2004) 75–84
of a technology for RNAi based on baculovirus to trans-
fer genetic material in mammalian cells is feasible. This is
one of many possible uses for baculovirus technology in the
RNAi field. Baculovirus can also be used as RNAi vectors
in insects. One of the advantages of baculovirus is that they
are highly infective and capable of producing large amounts
of dsRNA. The major disadvantage is that baculovirus are
also highly cytolytic in insect cells, which complicates the
interpretation of the gene interference results.
Acknowledgements
The excellent services from the Molecular Biology and
Microscopy units and the library at the Instituto de Fisi-
ologia Celular are greatly appreciated. Part of this work
was produced with equipment donated by the Alexander
von Humboldt-Stiftung and a grant from Fundación Miguel
Alemán to LV.
References
Ayres, M.D., Howard, S.C., Kuzio, J., Lopez-Ferber, M., Possee, R.D.,
1994. The complete DNA sequence of Autographa californica nuclear
polyhedrosis virus. Virology 202, 586–605.
Blaszczyk, J., Tropea, J.E., Bubunenko, M., Routzahn, K.M., Waugh,
D.S., Court, D.L., Ji, X., 2001. Crystallographic and modeling studies
of RNase III suggest a mechanism for double-stranded RNA cleavage.
Structure (Camb.) 9, 1225–1236.
Bucher, G., Scholten, J., Klingler, M., 2002. Parental RNAi in Tribolium
(Coleoptera). Curr. Biol. 12, R85–R86.
Caplen, N.J., Fleenor, J., Fire, A., Morgan, R.A., 2000. dsRNA-mediated
gene silencing in cultured Drosophila cells: a tissue culture model for
the analysis of RNA interference. Gene 252, 95–105.
Chuang, C.F., Meyerowitz, E.M., 2000. Specific and heritable genetic
interference by double-stranded RNA in Arabidopsis thaliana. Proc.
Natl. Acad. Sci. U. S. A. 97, 4985–4990.
Clemens, M.J., 1997. PKR—a protein kinase regulated by double-stranded
RNA. Int. J. Biochem. Cell Biol. 29, 945–949.
Dalmay, T., Hamilton, A., Rudd, S., Angell, S., Baulcombe, D.C., 2000.
An RNA-dependent RNA polymerase gene in Arabidopsis is required
for posttranscriptional gene silencing mediated by a transgene but not
by a virus. Cell 101, 543–553.
de Haro, C., Mendez, R., Santoyo, J., 1996. The eIF-2alpha kinases and
the control of protein synthesis. FASEB J. 10, 1378–1387.
Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver. S.E., Mello,
C.C., 1998. Potent and specific genetic interference by double-stranded
RNA in Caenorhabditis elegans. Nature 391, 806–811.
Ge, Q., McManus, M.T., Nguyen, T., Shen, C.H., Sharp, P.A., Eisen,
H.N., Chen, J., 2003. RNA interference of influenza virus production
by directly targeting mRNA for degradation and indirectly inhibiting
all viral RNA transcription. Proc. Natl. Acad. Sci. U. S. A. 100,
2718–2723.
Gitlin, L., Karelsky, S., Andino, R., 2002. Short interfering RNA confers
intracellular antiviral immunity in human cells. Nature 418, 430–434.
Grabherr, R., Ernst, W., Oker-Blom, C., Jones, I., 2001. Developments in
the use of baculoviruses for the surface display of complex eukaryotic
proteins. Trends Biotechnol. 19, 231–236
Hu, Y.C., Tsai, C.T., Chang, Y.J., Huang, J.H., 2003. Enhancement and
prolongation of baculovirus-mediated expression in mammalian cells:
83
focuses on strategic infection and feeding. Biotechnol. Prog. 19, 373–
379.
Inesi, G., Sagara, Y., 1992. Thapsigargin, a high affinity and global
inhibitor of intracellular Ca2+ transport ATPases. Arch. Biochem.
Biophys. 298, 313–317.
Jackson, J.A., 1991. Baculovirus expression systems and insect cells.
Bioprocess Technol. 13, 402–413.
Kapadia, S.B., Brideau-Andersen, A., Chisari, F.V., 2003. Interference of
hepatitis C virus RNA replication by short interfering RNAs. Proc.
Natl. Acad. Sci. U. S. A. 100, 2014–2018.
Katze, M.G., 1995. Regulation of the interferon-induced PKR: can viruses
cope? Trends Microbiol. 3, 75–78.
Lawrence, D., 2002. RNAi could hold promise in the treatment of HIV.
Lancet 359, 2007.
Lipardi, C., Wei, Q., Paterson, B.M., 2001. RNAi as random degradative
PCR: siRNA primers convert mRNA into dsRNAs that are degraded
to generate new siRNAs. Cell 107, 297–307.
Luckow, V.A., Summers, M.D., 1988. Signals important for high-level
expression of foreign genes in Autographa californica nuclear poly-
hedrosis virus expression vectors. Virology 167, 56–71.
Maeda, I., Kohara, Y., Yamamoto, M., Sugimoto, A., 2001. Large-
scale analysis of gene function in Caenorhabditis elegans by
high-throughput RNAi. Curr. Biol. 11, 171–176.
Massotte, D., 2003. G protein-coupled receptor overexpression with the
baculovirus-insect cell system: a tool for structural and functional
studies. Biochim. Biophys. Acta 1610, 77–89.
Means, J.C., Muro, I., Clem, R.J., 2003. Silencing of the baculovirus
Op-iap3 gene by RNA interference reveals that it is required for pre-
vention of apoptosis during Orgyia pseudotsugata M nucleopolyhe-
drovirus infection of Ld652Y cells. J. Virol. 77, 4481–4488.
Monsma, S.A., Oomens, A.G., Blissard, G.W., 1996. The GP64 enve-
lope fusion protein is an essential baculovirus protein required for
cell-to-cell transmission of infection. J. Virol. 70, 4607–4616.
Napoli, C., Lemieux, C., Jorgensen, R., 1990. Introduction of a chimeric
chalcone synthase gene into petunia results in reversible co-suppression
of homologous genes in trans. Plant Cell 2, 279–289.
Nebenfuhr, A., Ritzenthaler, C., Robinson, D.G., 2002. Brefeldin A: deci-
phering an enigmatic inhibitor of secretion. Plant Physiol. 130, 1102–
1108.
Nishikura, K., 2001. A short primer on RNAi: RNA-directed RNA poly-
merase acts as a key catalyst. Cell 107, 415–418.
Ojala, K., Mottershead, D.G., Suokko, A., Oker-Blom, C., 2001. Specific
binding of baculoviruses displaying gp64 fusion proteins to mam-
malian cells. Biochem. Biophys. Res. Commun. 284, 777–784.
Palauqui, J.C., Elmayan, T., Pollien, J.M., Vaucheret, H., 1997. Systemic
acquired silencing: transgene-specific post-transcriptional silencing is
transmitted by grafting from silenced stocks to non-silenced scions.
EMBO J. 16, 4738–4745.
Pieroni, L., La Monica, N., 2001. Towards the use of baculovirus as a
gene therapy vector. Curr. Opin. Mol. Ther. 3, 464–467.
Provost, P., Dishart, D., Doucet, J., Frendewey, D., Samuelsson, B., Rad-
mark, O., 2002. Ribonuclease activity and RNA binding of recombi-
nant human Dicer. EMBO J. 21, 5864–5874.
Ramaswamy, G., Slack, F.J., 2002. siRNA. A guide for RNA silencing.
Chem. Biol. 9, 1053–1055.
Robertson, H.D., Webster, R.E., Zinder, N.D., 1968. Purification and
properties of ribonuclease III from Escherichia coli. J. Biol. Chem.
243, 82–91.
Robertson, J.S., Harrap, K.A., Longworth, J.F., 1974. Baculovirus mor-
phogenesis: the acquisition of the virus envelope. J. Invertebr. Pathol.
23, 248–251.
Romano, N., Macino, G., 1992. Quelling: transient inactivation of gene
expression in Neurospora crassa by transformation with homologous
sequences. Mol. Microbiol. 6, 3343–3353.
Schwarz, D.S., Hutvagner, G., Haley, B., Zamore, P.D., 2002. Evidence
that siRNAs function as guides, not primers, in the drosophila and
human RNAi pathways. Mol. Cell 10, 537–548.
84 C.F. Flores-Jasso et al. / Virus Research 102 (2004) 75–84
Shen, C., Buck, A.K., Liu, X., Winkler, M., Reske, S.N., 2003. Gene
silencing by adenovirus-delivered siRNA. FEBS Lett. 539, 111–114.
Shuey, D.J., McCallus, D.E., Giordano, T., 2002. RNAi: gene-silencing
in therapeutic intervention. Drug Discov. Today 7, 1040–1046.
Smardon, A., Spoerke, J.M., Stacey, S.C., Klein, M.E., Mackin, N., Maine,
E.M., 2000. EGO-1 is related to RNA-directed RNA polymerase
and functions in germ-line development and RNA interference in C.
elegans. Curr. Biol. 10, 169–178.
Sorensen, D.R., Leirdal, M., Sioud, M., 2003. Gene silencing by systemic
delivery of synthetic siRNAs in adult mice. J. Mol. Biol. 327, 761–
766.
Stein, P., Svoboda, P., Anger, M., Schultz, R.M., 2003. RNAi: mammalian
oocytes do it without RNA-dependent RNA polymerase. RNA 9, 187–
192.
Stewart, S.A., Dykxhoorn, D.M., Palliser, D., Mizuno, H., Yu, E.Y., An,
D.S., Sabatini, D.M., Chen, I.S., Hahn, W.C., Sharp, P.A., Weinberg,
R.A., Novina, C.D., 2003. Lentivirus-delivered stable gene silencing
by RNAi in primary cells. RNA 9, 493–501.
Summers, M.D., Volkman, L.E., 1976. Comparison of biophysical and
morphological properties of occluded and extracellular nonoccluded
baculovirus from in vivo and in vitro host systems. J. Virol. 17, 962–
972.
Tabara, H., 2001. RNAi and PTGS in eukaryotes. Tanpakushitsu Kakusan
Koso 46, 2017–2024.
Tan, S.L., Katze, M.G., 1999. The emerging role of the interferon-induced
PKR protein kinase as an apoptotic effector: a new face of death? J.
Interferon Cytokine Res. 19, 543–554.
Valdes, V.J., Sampieri, A., Sepulveda, J., Vaca, L., 2003. Using
double-stranded RNA to prevent in vitro and in vivo viral infections
by recombinant baculovirus. J. Biol. Chem. 278, 19317–19324.
Vaucheret, H., Beclin, C., Fagard, M., 2001. Post-transcriptional gene
silencing in plants. J. Cell Sci. 114, 3083–3091.
Voorhoeve, P.M., Agami, R., 2003. Knockdown stands up. Trends Biotech-
nol. 21, 2–4.
Wianny, F., Zernicka-Goetz, M., 2000. Specific interference with gene
function by double-stranded RNA in early mouse development. Nat.
Cell Biol. 2, 70–75.
Winston, W.M., Molodowitch, C., Hunter, C.P., 2002. Systemic RNAi in
C. elegans requires the putative transmembrane protein SID-1. Science
295, 2456–2459.
Yamagishi, J., Isobe, R., Takebuchi, T., Bando, H., 2003. DNA microarrays
of baculovirus genomes: differential expression of viral genes in two
susceptible insect cell lines. Arch. Virol. 148, 587–597.
 Virus Research 102 (2004) 85–96

RNA interference as a new biotechnological tool for the control of virus

diseases in plants
Francisco Tenllado, César Llave, José Ramón Dı́az-Ruı́z∗
Departamento de Biologı́a de Plantas, Centro de Investigaciones Biológicas, CSIC, Ramiro de Maeztu 9, Madrid 28040, Spain

Abstract

RNA silencing occurs in a wide variety of organisms, including protozoa, fungi, plants and animals and involves recognition of a target
RNA and initiation of a sequence-specific RNA degradation pathway in the cytoplasm. In the last few years, there have been considerable
advances in our understanding of post-transcriptional gene silencing (PTGS). This mechanism is conceived as a natural antiviral defense
system in plants that is activated as a response to double-stranded RNA (dsRNA) formed during virus replication. To develop new approaches
for plant protection against virus diseases based on PTGS we have expanded previous findings on RNA interference (RNAi) in animals by
using dsRNA to specifically interfere with virus infection in plants. This approach differs from strategies based on transgenic expression of
RNAs but still relies on PTGS as a means to achieve pathogen-derived resistance (PDR). Our findings suggest that exogenously supplied
dsRNA could form the basis for the development of an environmentally safe, new biotechnological tool aimed at protecting crops against
virus diseases, provided that some limitations of the current status of the approach could be overcome.
© 2004 Elsevier B.V. All rights reserved.

Keywords: RNA interference; Post-transcriptional gene silencing; Double-stranded RNA; Virus diseases; Plant protection

1. Introduction specific step during virus multiplication in the plant. Many

A number of homology-dependent gene silencing phe-
nomena have been reported in different organisms. They de-
scribe specific variants of the same basic mechanism which
acts mainly at the post-transcriptional level of gene ex-
pression (Hammond et al., 2001). Post-transcriptional gene
silencing (PTGS) was first reported in plants 13 years ago as
a coordinated and reciprocal inactivation of host genes and
transgenes encoding the same RNA (Napoli et al., 1990).
Shortly after, a similar phenomenon was observed in the
filamentous fungus Neurospora crassa and termed quelling
(Romano and Macino, 1992). Historically, the principle
of pathogen-derived resistance (PDR) as a way to confer
protection against viral infection has been strongly linked
to genetic manipulation of plants in order to obtain trans-
genic resistance (Sanford and Johnston, 1985). This well-
established principle involves the use of virus-derived genes
or genome fragments with the purpose to interfere with a
∗ Corresponding author. Present address: Centro de Investigaciones
Biológicas, Ramiro de Maeztu 9, 28040 Madrid, Spain.
Tel.: +34-1-5611800; fax: +34-1-5627518.
E-mail address: jrdiazruiz@cib.csic.es (J.R. Dı́az-Ruı́z).
papers have been devoted to RNA-mediated virus resistance,
nearly all of which can be classified as a specific manifesta-
tion of PTGS in plants (reviewed by Vazquez Rovere et al.,
2002; Goldbach et al., 2003). In this particular case of PTGS,
both the viral RNA and the virus-derived transgene mRNA
are targeted for degradation rendering plants that display
protection against virus infection (Baulcombe, 1999).
A major breakthrough in the silencing history was the
discovery of a gene silencing response in the nematode
Caenorhabditis elegans (Fire et al., 1998). This phe-
nomenon, in which experimentally introduced double-
stranded RNA (dsRNA) leads to loss of expression of the
corresponding cellular gene by sequence-specific RNA
degradation, was called RNA interference (RNAi). RNAi
could be induced by injection of dsRNA into any site of
the animal (Fire et al., 1998), by feeding the worms on
engineered Escherichia coli producing dsRNA against the
target gene (Timmons and Fire, 1998), by in vivo transcrip-
tion of dsRNA from transgene promoters (Tavernarakis
et al., 2000), or simply by soaking the worms in a solution
containing the dsRNA (Tabara et al., 1998). Since this dis-
covery emerged in C. elegans, RNAi has been documented
in a wide variety of animals, ranging from protozoans, such

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.019
86 F. Tenllado et al. / Virus Research 102 (2004) 85–96
as Trypanosome, to metazoans, such as Drosophila and
mammals (Tijsterman et al., 2002).
A growing body of biochemical and genetic data has
further demonstrated that plants, animals and yeasts share
related mechanisms of specific degradation of RNAs in
which double-stranded forms of RNA are initiator molecules
(Dalmay et al., 2000; Fagard et al., 2000; Bernstein et al.,
2001). In plants, dsRNA or self-complementary hairpin
RNA that are transcribed from engineered inverted repeats
were shown to be potent inducers of a gene silencing re-
sponse when directed against trangenes (Hamilton et al.,
1998; Waterhouse et al., 1998; Johansen and Carrington,
2001). Furthermore, plants transformed with constructs that
produce RNAs capable of duplex formation induce virus im-
munity with almost 100% efficiency when targeted against
viruses (Smith et al., 2000; Wang et al., 2000; Kalantidis
et al., 2002).
As early as 1978, virus-specific dsRNA forms were iso-
lated from infected plant tissues (Dı́az-Ruı́z and Kaper,
1978). More recently, several lines of evidence indicate that
virus-induced gene silencing (VIGS), a particular manifes-
tation of PTGS in which viruses are both triggers and targets
of silencing, has evolved into a viral surveillance system
triggered by dsRNA produced during the intermediate steps
of virus replication (Ratcliff et al., 1999; Waterhouse et al.,
2001). Since dsRNA does not naturally occur in the cyto-
plasm of plant cells, recognition of nucleic acid structures
that are present only in infected cells seems to induce a
natural line of defense against viral infections (Voinnet,
2001). Moreover, it has been speculated that silencing pro-
cesses may also serve as an antiviral system in vertebrates
(Li et al., 2002). In the PTGS-associated antiviral response,
viral dsRNA is first processed by an RNase III-like nucle-
ase, which was named Dicer in Drosophila melanogaster
and for which several homologs in Arabidopsis have been
found, into 21–25 nt dsRNAs (small interfering RNAs, siR-
NAs) that guide another nuclease complex (RNA-induced
silencing complex, RISC) to cleave the RNA genome of
the invading virus (Hamilton and Baulcombe, 1999; Tang
et al., 2003). As a counterdefensive strategy, many viruses
encode proteins that suppress silencing at different points
of the PTGS pathway (Mlotshwa et al., 2002). Thus, the
discovery that viruses are inducers, and targets, and may
carry suppressors of PTGS adds further support to the in-
volvement of dsRNA-induced silencing processes as an
innate defense mechanism against virus infection.
In the last few years, there have been considerable ad-
vances in our understanding of the molecular mechanism of
PTGS. We now know that both PTGS and RNAi are man-
ifestations of a broader set of gene silencing phenomena
common to most, if not all, eukaryotes. For instance, one
emerging theme is that besides an important biological role
in protecting the organism against exogenous nucleic acids,
PTGS-related processes are implicated in maintenance of
genome stability (Ketting et al., 1999; Djikeng et al., 2001;
Hamilton et al., 2002) and in the formation of heterochro-
matin (Volpe et al., 2002) in different organisms. More-
over, recent findings have pointed to an even more central
role for RNA silencing in an RNA-based regulatory network
aimed at regulating gene expression in plants and animals
(Hutvágner and Zamore, 2002; Llave et al., 2002). Details
on molecular mechanisms of PTGS/RNAi and its natural
role for endogenous gene regulation have been considered in
previous reviews (Hannon, 2002; Pasquinelli and Ruvkun,
2002; Denli and Hannon, 2003).
The concept of obtaining resistance to viruses by transfor-
mation with genes derived from the genome of the pathogen
is a well-exploited and highly effective procedure to fight
viruses as causal agents of diseases in plants. In the past,
we have genetically engineered Nicotiana plants using viral
gene sequences with a view of providing the basis for a po-
tential control of virus diseases in crops (Peña et al., 1994;
Tenllado et al., 1995, 1996). In particular, we were interested
in tobamovirus strains which are able to infect commer-
cial pepper cultivars with genetic resistances introduced by
classical breeding against common strains of tobamoviruses.
Like most cases in PDR, transgenic resistance against Pepper
mild mottle virus (PMMoV) using replicase gene sequences
was based on PTGS (Tenllado and Dı́az-Ruı́z, 1999). How-
ever, because most pepper varieties are recalcitrant to ge-
netic transformation, control of diseases caused by pepper
strains of tobamoviruses using this strategy await future
progress. One important aim of our current research is to
develop new biotechnological approaches for plant protec-
tion against virus diseases based on PTGS. In this sense,
we have expanded previous findings on RNAi in animals
by using dsRNA to specifically interfere with virus infec-
tion in plants. This approach differs from strategies based
on transgenic expression of RNAs but still relies on PTGS
as a means to achieve PDR. Some of the results obtained in
our laboratory are presented here.
2. RNAi in plants
The essence of RNAi technology is the delivery of dsRNA
as a potent activator of RNA silencing into an organism,
or cell, with the purpose of triggering sequence-specific
degradation of homologous target RNAs. As we mentioned
above, dsRNA can be delivered by stably transforming
plants with transgenes that express a self-complementary
RNA. The resulting transcript hybridizes with itself to
form a hairpin structure that comprises a single-stranded
loop region (usually corresponding to an intron) and a
base-paired stem, which mimics the dsRNA structure that
induces RNAi. It has been shown that transgene constructs
encoding intron-spliced RNA with hairpin structure give
stable silencing with almost 100% efficiency when directed
against viruses (Smith et al., 2000; Kalantidis et al., 2002).
The major technical limitation for this technology is that
many important plant crop species are difficult or impossi-
ble to transform, precluding the constitutive expression of
 F. Tenllado et al. / Virus Research 102 (2004) 85–96
constructs directing production of dsRNA. Moreover,
questions concerning the potential ecological impact of
virus-resistant transgenic plants have so far significantly
limited their use (Tepfer, 2002).
When the RNAi technology was first applied in mam-
malian cells, investigators had to tackle the fact that dsRNA
molecules longer than 30 base pairs trigger a non-specific
interferon-mediated response that leads to general suppres-
sion of protein synthesis and ultimately cell death (Gil
and Esteban, 2000). Afterwards, an outstanding discovery
demonstrated that RNAi can also be initiated in different cul-
tured mammalian cell lines by transfecting synthetic siRNA
duplexes (Elbashir et al., 2001). Indeed, sequence-specific
silencing at the post-transcriptional level can be achieved
in Drosophila embryos, C. elegans or cultured mammalian
cells by injection of synthetic dsRNA molecules. From
this discovery emerged the notion that in vitro transcribed
dsRNA might also act as an activator of RNAi in plants.
Moreover, plant cells seem to lack the mechanisms that
trigger non-specific inhibition of gene expression by long
dsRNA, such as the activation of a protein kinase–mediated
antiviral defense presents in mammalian cells (Cayley et al.,
1982). Thus, the use of synthetic siRNAs to induce RNAi
is not obligatory in plants as it is in mammalian systems.
In our laboratory we have made intensive research efforts
to assess the potential of dsRNA-mediated interference in
plant virus infections. As an initial approach, we investi-
gated whether direct delivery by mechanical inoculation of
dsRNA together with either virion particles or viral tran-
scripts could interfere with infection in a sequence-specific
manner (Tenllado and Dı́az-Ruı́z, 2001). We made use of
three viruses belonging to the tobamovirus (PMMoV), po-
tyvirus (Tobacco etch virus, TEV) and alfamovirus (Alfalfa
mosaic virus, AMV) groups, as representative examples
in the diversity of positive-strand RNA viruses in plants.
Sense single-stranded RNA (ssRNA) and antisense ssRNA
corresponding to different parts of the viral genomes were
transcribed in vitro from partial cDNA clones and even-
tually annealed to each other to produce dsRNAs. In our
initial experiments, the ability of sense ssRNA, antisense
Fig. 1. Specific interference with PMMoV infection by dsRNA in a hypersensitive host. Response of N. tabacum cv. ‘Xanthi nc’ to PMMoV alone (left
halves of leaves) or to a combination of PMMoV plus either 54-kDa antisense (as) RNA (A), 54-kDa dsRNA (B) or TEV-HC dsRNA (C) (right halves
of the leaves). Similar number of local lesions was observed in both halves of the leaves in panels (A) and (C). No visible local response was observed
in the half-leaf inoculated with PMMoV plus 54-kDa dsRNA (B).
87
ssRNA or dsRNA derived from the replicase gene of PM-
MoV (54-kDa-protein region) to inhibit local lesion forma-
tion elicited by PMMoV was evaluated using two different
hypersensitive hosts, Nicotiana tabacum cv. Xanthi nc and
Capsicum chinense. We found that mechanical inoculation
of neither sense ssRNA, antisense ssRNA nor cDNA corre-
sponding to the 54-kDa-protein region of the virus caused
reduction in the number of local lesions elicited by PMMoV
in plants. However, virus infectivity was completely blocked
by coinoculation of PMMoV particles with a 977-bp dsRNA
derived from the 54-kDa-coding region (Fig. 1). Further-
more, we found interference with virus infection only when
dsRNA molecules share sequence identity with the virus.
Coinoculation of PMMoV together with a 1483-bp dsRNA
corresponding to most of the Helper component-Proteinase
(HC-Pro) gene of TEV (TEV-HC dsRNA), did not have any
effect on PMMoV infectivity. Thus, these results clearly
demonstrate that RNAi can be successfully triggered against
plant viruses by mechanically inoculating in vitro tran-
scribed dsRNA to leaf cells together with the target virus.
Moreover, we also found that homologous dsRNA was
competent to interfere with virus infection in a systemic host
infection. Nicotiana benthamiana plants that had been in-
oculated with mixtures of PMMoV particles and different
PMMoV-derived dsRNAs were resistant to systemic virus
infection. We further confirmed that dsRNA molecules as
short as 596 bp from either the 54- or 30-kDa coding genes
were effective in protecting plants against PMMoV infec-
tion. Not only did most plants not display disease symp-
toms during the completion of their life cycles, but more im-
portantly, neither viral RNA nor viral particles accumulated
to detectable levels in both inoculated and systemically in-
fected N. benthamiana leaves. We found that plants coinocu-
lated with PMMoV particles and a homologous 315 bp long
dsRNA displayed viral symptoms with a delay of 1–2 days
compared to plants inoculated with PMMoV alone, with vi-
ral RNA accumulating at levels similar to the control. This
may indicate that the ability of dsRNA to specifically in-
terfere with virus infection seems to require a minimum
length of dsRNA. In fact, RNAi has also been reported to
88 F. Tenllado et al. / Virus Research 102 (2004) 85–96
depend on the length of dsRNA in plants, C. elegans and
Drosophila (Fire et al., 1998; Hammond et al., 2000). Re-
combinant viruses carrying fragments of usually 300–800
nucleotides derived from the target gene efficiently trigger
RNAi of a transgene in plants, although fragments as short
as 23–60 nucleotides may also be effective (Thomas et al.,
2001). In these experiments, the source of siRNAs would
be the sum of the sequence-specific RNA degradation prod-
ucts from the recombinant virus and the transgene mRNA,
conceivably a higher dose than produced in our experimen-
tal system. Neither sense ssRNA nor antisense ssRNA de-
rived from PMMoV or nonhomologous dsRNA (TEV-HC
dsRNA) protected N. benthamiana plants against systemic
infection.
Our experiments have also verified that the use of
dsRNA molecules is a general strategy to promote inter-
ference with the infection of plants by a variety of plant
viruses. N. tabacum plants inoculated with a mixture con-
taining infectious TEV genomic RNA transcripts and TEV
HC-Pro-derived dsRNA were protected against disease
symptoms, and viral RNA did not accumulate in neither
inoculated nor upper leaves of these plants. Likewise, none
of the N. benthamiana plants inoculated with a mixture of
in vitro transcripts of AMV RNAs 1, 2 and 3 plus a dsRNA
covering a 1124-nt region of AMV RNA 3 became infected,
and AMV RNAs were not detectable in these plants.
Because entrance of mechanically inoculated-dsRNA into
the cell occurs upon abrasion of the epidermal tissue on the
leaf surface, accurate quantitation of dsRNA molecules en-
tering the cell is not possible, and consequently optimal con-
ditions for induction of an RNAi response against viruses
are difficult to establish. Using PMMoV, we could deter-
mine that a 500-fold excess of dsRNA over viral RNA is
the threshold required for interference with virus multipli-
cation under the experimental conditions used. This appar-
ently argues against an amplification step in the interference
process similar to that described in C. elegans, where a few
molecules of dsRNA are sufficient to trigger RNAi (Fire
et al., 1998). Amplification of RNA silencing in cases in-
volving transgenes likely reflects a transitive phenomenon
for which the initial pool of siRNAs generated from dsRNA
inducer molecules is increased through the formation of sec-
ondary siRNAs. It is suggested that primary siRNAs anneal
with target RNA to prime a PCR-like reaction catalyzed by
a host encoded-RNA dependent RNA polymerase (RdRp)
to form dsRNA corresponding to sequences located out-
side the primary targeted region of the RNA. In contrast,
RdRp is dispensable in cases involving VIGS since the virus
encoded-RNA polymerase would produce dsRNA (Vaistij
et al., 2002). Why amplification of RNA silencing does not
occur in our system remains unclear although it may be due
to the inability of the in vitro synthetized dsRNA to prime the
production of new dsRNA from the target viral RNA. Alter-
natively, as mechanical inoculation of plant viruses is based
on limited damage to epidermal cells by abrasion, the dose
effect described above probably reflects a minimum amount
of dsRNA required to ensure that most viral RNA pene-
trates into the damaged cells in combination with appropri-
ate quantities of dsRNA. From a mechanistic perspective, all
evidence favors the hypothesis that dsRNA interferes with
virus infection at the cell level in the inoculated leaves: (i)
dsRNA interferes with PMMoV infection on hypersensitive
hosts and (ii) in cases where the virus is able to overcome
the protection conferred by dsRNA in a systemic host, there
is a delay in symptom expression of several weeks, suggest-
ing a reduced number of competent infection foci in the in-
oculated leaves. Furthermore, delivery of virus and dsRNA
into the same cells seems necessary to trigger interference,
because only a short interval (24 h) between sequential in-
oculation of dsRNA and virus resulted in a complete sus-
ceptibility of the plants to virus infection. This phenomenon
may be explained if we assume that during secondary inoc-
ulation the challenging virus enters the plant tissue through
a number of cells that were not originally damaged during
the first inoculation round and that therefore do not contain
dsRNA molecules. This finding gives rise to one remark-
able feature of RNAi in our system, i.e. the failure of locally
introduced dsRNA to trigger a systemic antiviral response
in plants. When dsRNA-treated N. benthamiana plants are
inoculated with virus in upper, non-treated leaves, they are
found to be susceptible to virus infection.
Experiments performed by others have revealed that
microprojectile bombardment into leaf epidermal cells
with beads coated with dsRNA molecules homologous to
endogenous genes triggered local RNAi, although its ef-
ficiency appeared to decrease along with time (Schweizer
et al., 2000). In plants, analysis of PTGS revealed that there
are three separate components in the dynamics of RNA
silencing: initiation, propagation of a systemic silencing
signal, and maintenance (Ruiz et al., 1998). Furthermore,
it has been reported that a nuclear component, transgene or
endogenous gene, homologous to the target RNA is required
for propagation and amplification of the systemic silencing
signal (Palauqui and Balzergue, 1999). In our scenario,
dsRNA delivered by inoculation or expressed directly inside
plant cells by means of Agrobacterium tumefaciens (see
below) would trigger the initiation step of PTGS targeting
viral RNA for degradation. The maintenance step of PTGS
would require a nuclear component, transgene or DNA se-
quences, homologous to the target viral RNA. The absence
of such a component in our nontransgenic system together
with the failure of input dsRNA to move between cells at
detectable levels (unpublished data), presumably precludes
spreading of the sequence-specific signal involved in PTGS
to other parts of the plant.
3. Interference with plant virus by transiently
expressed hairpin RNA
Although the effects mediated by dsRNA in plant virus
infection recapitulate many aspects of RNAi reported in
 F. Tenllado et al. / Virus Research 102 (2004) 85–96
animals, a direct link between inhibition of plant virus in-
fection by exogenously supplied dsRNA and PTGS-related
processes was unequivocally provided by using A. tumefa-
ciens as a transient expression system. Transient expression
of genes through infiltration of A. tumefaciens cultures into
leaf tissue (agroinfiltration) has been widely used as a means
to deliver inducers, targets and suppressors of PTGS into
plants (Llave et al., 2000; Johansen and Carrington, 2001).
For instance, expression of a given transgene in a plant can
be suppressed by infiltrating the plant with a recombinant bi-
nary vector carrying the corresponding transgene sequence.
In our laboratory, we have investigated the silencing poten-
tial of a construct designed to generate a hairpin RNA after
agroinfiltration in the plant tissue to interfere with a rapidly
replicating plant virus (Tenllado et al., 2003a).
As mentioned in the introduction, virus resistance in
plants has been reported to be induced at high efficiency
by stably integrated transgenes expressing inverted-repeat
structures (Smith et al., 2000). We have demonstrated that
transient expression of a hairpin RNA can also effectively
block multiplication and spread of a homologous highly
replicating plant virus in nontransgenic plants. As a result,
hairpin-dependent interference with virus infection by tran-
siently expressed constructs provides a potential tool to
study the onset of PTGS in the context of a viral infection,
in contrast to dsRNA-transgenic plants which primarily al-
low the study of PTGS maintenance. N. benthamiana plants
infiltrated with cultures of Agrobacterium carrying a hair-
pin RNA derived from the 54-kDa region of PMMoV were
resistant to subsequent infection by PMMoV. In contrast,
Agrobacterium-mediated expression of constructs contain-
ing duplicated (head-to-tail) PMMoV 54-kDa sequences
in the sense or antisense orientations were ineffective as
initiators of RNAi, and disease symptoms were displayed
in upper leaves. At an interval of 3 or more days between
agroinfiltration with PMMoV hairpin RNA and virus in-
oculation plants were protected against virus infection, as
indicated by the absence of PMMoV RNA in both the
inoculated and the upper leaf tissue.
The hairpin RNA-dependent interference with virus in-
fection observed with transiently expressed hairpin RNA
resembled interference with viral infection by direct deliv-
ery of dsRNA to leaf cells by mechanical inoculation. We
found several features associated with transient expression
of hairpin RNA that are hallmarks of PTGS. First, the in-
terference with virus infection exhibited by hairpin RNA
was dependent on a high level of sequence identity between
the hairpin RNA and the target RNA. We observed that
plants transiently expressing PMMoV-derived hairpin RNA
exhibited systemic infection when inoculated with an unre-
lated virus such as AMV or Tobacco mosaic virus (TMV)
(73% identity). By contrast, agro-infiltrated plants remain
uninfected when inoculated with a related strain of PMMoV
(PMMoV-I), which shares 93% sequence identity with the
hairpin RNA. Further evidence for a homology-dependent
type of resistance came from the observation that plants
89
infiltrated with hairpin RNA were resistant to a Potato virus
X (PVX)-based vector carrying sequences homologous to
the PMMoV hairpin RNA. Second, siRNAs of two size
classes were found in extracts from hairpin RNA-infiltrated
tissue. These siRNAs are consistently associated with an
activated PTGS status (Hamilton and Baulcombe, 1999)
and result from specific cleavage of dsRNA by an RNase
III-like enzyme. Third, the interference with PMMoV infec-
tion triggered by a hairpin RNA was partially overcome by
transiently expressing a construct encoding the HC-Pro of
Plum pox virus (PPV). In the last years, it has been demon-
strated that HC-Pro of some potyviruses act as suppressors
of PTGS (Brigneti et al., 1998). Although not formally
demonstrated, we assumed that the HC-Pro of PPV could
behave as a suppressor of PTGS, since it has been previ-
ously shown to be involved in viral synergism with PVX in
N. benthamiana (Sáenz et al., 2002; P. González-Jara et al.,
unpublished data). This phenomenon is closely related to the
silencing suppressor activity of HC-Pro in other potyviruses,
which supports its capability to suppress PTGS induced by
transiently expressed hairpin RNA (Pruss et al., 1997).
Although hairpin-dependent interference with viral in-
fection has been described for PMMoV, it is reasonable
to expect that many plant RNA viruses will be inhibited
by Agrobacterium-mediated transient expression of homol-
ogous hairpin RNAs. In addition, this approach can give
valuable information about the step in the PTGS pathway
where viral proteins suppress RNA silencing. In this sense,
available evidence indicates that transgene expression ob-
tained by infiltrating plant tissues with Agrobacterium is pri-
marily due to transcription of episomal T-DNA, rather than
the generation of stably integrated transgenes that are in-
volved in PTGS maintenance (Sonti et al., 1995). In previous
studies, HC-Pro was shown to interfere with a maintenance
step of the silenced state (Llave et al., 2000). Our coinfil-
tration assays, in which expression of HC-Pro occurs at the
same time as initiation of silencing by hairpin RNA, suggest
that HC-Pro of potyviruses also interferes with the initia-
tion of PTGS. Supporting this, it has been reported that the
HC-Pro of TEV partially inhibits PTGS induced by a tran-
siently expressed dsRNA from a jellyfish green fluorescent
protein at early stages of PTGS (Johansen and Carrington,
2001).
Some limiting aspects of dsRNA-mediated interference
with virus infection by mechanical inoculation of dsRNA
to leaf cells were also described in Agrobacterium-mediated
interference, i.e. the lack of a systemic antiviral response in
upper parts of the plant triggered by locally induced PTGS.
Delivery of virus and hairpin RNA into the same tissues
seemed necessary to achieve protection against virus in-
fection, because Agrobacterium-infiltrated plants challenged
with PMMoV in upper, noninfiltrated leaves became in-
fected. Similar arguments to those expressed for direct de-
livery of dsRNA by mechanical inoculation, together with
the failure of Agrobacterium to efficiently move for long
distances, can be made to explain this failure (see above).
90 F. Tenllado et al. / Virus Research 102 (2004) 85–96

RNA silencing represents an inherent defense mechanism
whereby a sequence-specific response limits the extent of
virus infection (Voinnet, 2001). Unequivocal evidence exists
that RNA silencing can be transmitted throughout plants by
a mobile systemic signal (Mlotshwa et al., 2002). Viruses
have moved forward with the development of efficient strate-
gies to counteract or to escape RNA silencing. For instance,
several plant viral silencing suppressors are known to af-
fect systemic signaling of silencing, in agreement with the
notion that systemic infection by a virus requires dimin-
ishing the antiviral silencing defense response in the plant
(Voinnet et al., 2000). In this sense, the suppression of PTGS
by HC-Pro of potyviruses could be potentially mediated, at
least in part, by the activation of a pathway that transmits
the inhibition of silencing to other parts of the plant by a
mechanism similar to the trafficking of endogenous proteins
and their transcripts between cells and through the phloem
(Jorgensen et al., 1998). The combination of both an in-
ducer (hairpin RNA) and a suppressor (HC-Pro) of PTGS
transiently expressed by Agrobacterium allowed us to de-
vise a set of experiments involving the sequential infiltra-
tion of plants in different leaves with single Agrobacterium
cultures carrying either the PPV HC-Pro or the PMMoV
hairpin RNA. HC-Pro transiently expressed in lower leaves
failed to suppress the inhibitory effect on PMMoV replica-
tion triggered by infiltration of hairpin RNA in upper leaves
(Fig. 2). Thus, no evidence for a mobile silencing suppres-
sion signal induced by HC-Pro of PPV was observed in our
system. However, this result does not completely rule out
the involvement of a mobile silencing suppression signal in
natural potyvirus infection. For instance, PTGS suppression
signaling to distal parts of the plant may require the accumu-
lation of a threshold level of HC-Pro on the onset of a virus
infection, or perhaps, any other potyviral proteins. The P1
proteinase of TEV has been shown to enhance suppression
of VIGS by HC-Pro (Anandalakshmi et al., 1998).
4. The potential of RNAi in plant virus disease control
The findings presented above suggest that exogenously
supplied dsRNA could form the basis for the development
of a new biotechnological tool aimed at protecting crops
against virus diseases, provided that some limitations of
the current status of the approach could be overcome. Ob-
viously, interference with virus infection mediated by in
vitro-transcription products or via Agrobacterium-mediated
transient expression of dsRNA is limited to fundamental
studies. Thus, effective means of production and delivery of
adequate interference products onto plant surfaces should
be developed before a practical use of RNAi in plant pro-
tection could be envisioned. As an attempt to achieve this,
we established a simple, fast, safe and inexpensive proce-
dure to produce large amounts of dsRNA derived from viral
sequences using an inducible expression system based on a
genetically modified E. coli strain (Tenllado et al., 2003b).
In C. elegans, the E. coli strain HT115(DE3) has been
used to produce large amounts of dsRNA with the purpose
of triggering RNAi of endogenous genes in the nematode
(Timmons and Fire, 1998; Timmons et al., 2001). This strain
is deficient for RNase III, an enzyme that normally de-
grades the majority of dsRNAs in the bacterial cell, thereby
improving the accumulation of extended dsRNA duplexes
compared to common strains of E. coli. We adopted this
promising strain, which had also been modified to express
T7 RNA polymerase from an IPTG-inducible promoter, for
the preparation of massive amounts of viral dsRNA in vivo.
Upon induction, HT115(DE3) transformed with a construct
encoding a hairpin RNA derived from the 54-kDa region of
PMMoV accumulated substantial levels of PMMoV-derived
dsRNA. More importantly, the 54-kDa dsRNA produced
in bacteria was capable of interfering with PMMoV infec-
tion. In fact, N. benthamiana plants inoculated with mix-
tures of PMMoV and phenol-extracted nucleic acid extracts

Fig. 2. PPV HC-Pro expressed in lower leaves does not suppress hairpin-mediated interference with PMMoV infection in upper leaves. (A) Schematic
representation of the experimental procedure. Two lower, opposite leaves of N. benthamiana plants were first infiltrated with Agrobacterium containing
either vector alone or HC-Pro. After 7 days, upper leaves of these plants were infiltrated in all the combinations with Agrobacterium containing either
vector alone or hairpin RNA. Four days later, PMMoV was inoculated on the uppermost, infiltrated leaves. (B) Northern blot analysis of total RNA
extracted from the inoculated leaves of plants inoculated with PMMoV at 15 days post-inoculation.
 F. Tenllado et al. / Virus Research 102 (2004) 85–96 91

prepared from HT115(DE3) expressing dsRNA were free of
symptoms during the completion of their life cycles. Fur-
thermore, viral RNA levels were below the limit of detec-
tion in both the inoculated and the upper leaf tissue. In
contrast, all the plants that were inoculated with PMMoV
plus extracts derived from HT115(DE3) harboring the empty
cloning vector displayed high levels of viral RNA. In agree-
ment with previous results that confirmed the remarkable
sequence-specificity of the RNA silencing process, coinoc-
ulation of nucleic acid extracts containing PMMoV dsRNA
together with a nonhomologous virus, TMV, did not affect
virus infection in plants. Thus, interference with virus in-
fection exhibited by nucleic acid extracts derived from bac-
teria expressing dsRNA was dependent on a high level of
sequence identity with the target viral RNA.
Large-scale production of dsRNA required for the imple-
mentation of RNAi in plant protection depends on a sim-
ple and quick procedure to obtain dsRNA-related interfering
products from bacteria, avoiding the use of time-consuming
protocols to purify nucleic acids. This prompted us to an-
alyze the interfering activity on virus infection of bacterial
crude preparations obtained by lysing cell pellets with a
French Press. We found that viral accumulation was com-
pletely inhibited in N. benthamiana leaves that had been
inoculated with a mixture of a bacterial lysate expressing
PMMoV-derived dsRNA and PMMoV particles. In con-
trast, plants coinoculated with virus and bacterial lysate
preparations expressing the empty vector or a nonhomol-
ogous dsRNA (see below) accumulated PMMoV RNA at
high levels (Fig. 3). This precluded any effect concerning
non-specific toxicity of the bacterial extract on the inter-
ference with virus infection observed by dsRNA-expressing
preparations.
Further experiments in our laboratory have proved RNAi
technology to be suitable to control infection by a variety of
plant viruses. PPV is the causal agent of sharka or plum pox,
the most serious disease on woody trees of the genus Prunus.
Most cultivated Prunus species are highly susceptible, and
conventional breeding has not produced highly resistant
and commercially acceptable varieties (Ravelonandro et al.,
2000). This situation encouraged us to apply RNAi against
PPV in an experimental host, N. benthamiana, with the per-
spective of providing the basis for the potential control of
PPV in its natural host. As predicted, a high proportion of
plants inoculated with combinations of PPV plus homolo-
gous dsRNA-expressing preparations showed no symptoms
of virus disease. Furthermore, no traces of PPV replication
were detected by several diagnostic tests (ELISA, RT-PCR)
in the upper leaves of these plants. Thus, our work has
demonstrated that easy-to-obtain, crude extracts of bacte-
rially expressed dsRNAs are highly effective in protecting
plants against infection by PMMoV and PPV, pathogens
that represent two of the main widespread groups of plant
viruses in nature.
From a practical standpoint, we developed a simple
spray technique for the delivery of dsRNA-containing
crude preparations onto the surface of plant leaves. We
reported that virus infectivity was specifically abolished
when plants were sprayed with bacterial lysates several
days before mechanical inoculation with either PMMoV
or PPV in the same leaves. Analysis of the upper leaves
from dsRNA-treated plants confirmed the absence of either
PMMoV or PPV accumulation. The rationale of this ob-
servation is that dsRNA molecules, which remain on the
leaf surface after being sprayed, enter the cell along with
viral particles due to mechanical damage in the epidermal

Fig. 3. Interference with PMMoV infection by bacterial crude preparations expressing PMMoV dsRNA. (A) Northern blot analysis of total RNA
extracted from inoculated leaves of N. benthamiana at 7 days post-inoculation. Plants were inoculated with mixtures of PMMoV plus either a series of
dilutions (1/2–1/20) of the PMMoV dsRNA preparation, or the PPV dsRNA preparation diluted 1/2, as indicated. (B) Response of N. benthamiana to a
combination of PMMoV plus French Press preparations derived from either PMMoV dsRNA (left), PPV dsRNA (middle) or empty vector (right). Plants
displaying disease symptoms (one-half-diluted, PPV dsRNA and vector preparations) or showing protection to virus infection (1/10-diluted PMMoV
dsRNA preparation) were photographed at 30 days post-inoculation.
92 F. Tenllado et al. / Virus Research 102 (2004) 85–96
cells. As a result, the dsRNA is converted into siRNAs that
guide sequence-specific cleavage of the homologous viral
RNA. The presence of interfering products on leaf surfaces
for several days is not surprising as a previous analysis of
the stability of in vitro-synthesized dsRNA molecules after
delivery into leaves, had shown most of the input dsRNA
to be relatively stable and to persist in the leaves for sev-
eral days after inoculation, even after exuberant washing
(Tenllado and Dı́az-Ruı́z, 2001). These encouraging re-
sults lead to the concept of spraying crude preparations of
dsRNA-expressing E. coli HT115(DE3) as a highly con-
venient method for delivery of RNAi inducers in plants
and further contribute to making RNAi technology widely
available and applicable for deployment in the field.
5. Perspectives of the use of RNAi for crop protection
Until recently, the only available direct strategy, beyond
classical plant breeding, for the potential control of virus
diseases in crops was the use of virus-resistant transgenic
plants. The demonstration in 1986 that the expression of
a viral coat protein gene in a transgenic plant could con-
fer resistance to the virus represented a major breakthrough
(Powell-Abel et al., 1986). It opened the prospect of a virtu-
ally unlimited source of virus resistance genes that could be
used in crops in which sources of natural resistance genes
were unavailable or inadequate. However, over the past 10
years, questions concerning the potential ecological impact
of transgenic plants have been raised (Tepfer, 2002). The
areas of potential risk include complementation, heterolo-
gous encapsidation, synergy, and genetic flow between or-
ganisms. In view of these concerns, alternative strategies
aimed at protecting crops against virus diseases reducing or
eliminating certain risks possibly associated with transgenic
expression of viral sequences should be evaluated.
In just 5 years, molecular biologists have uncovered a pre-
viously unsuspected biological mechanism which is used by
nature to silence genes in many different settings. Besides
having become a prominent tool in basic research, RNAi
technology provides a significant new strategy to combat vi-
ral diseases in many organisms, from plants to mammals.
It offers a combination of extreme specificity and the ex-
ploitation of an underlying cellular pathway that has seem-
ingly evolved naturally for this purpose. A common feature
of silencing-related processes in different organisms is the
involvement of dsRNA as an initiator molecule. We have
seen that during cytoplasmic replication of RNA viruses, the
replicative forms and replicative intermediates may repre-
sent the pool of dsRNAs that trigger induction of the PTGS
antiviral defense mechanism in plants, at least under certain
circumstances (Dalmay et al., 2000; Voinnet et al., 2000).
At present, it is not clear whether RNAi is part of a vi-
ral surveillance system in animal cells. However, antiviral
RNAi technology has been recently applied for the con-
trol of human immunodeficiency virus type 1, hepatitis C
virus, poliovirus, rotavirus, and respiratory syncytial virus
in human cell lines using siRNAs (Bitko and Barik, 2001;
Gitlin et al., 2002; Jacque et al., 2002; Novina et al., 2002).
Our results summarized here contribute to establish the ba-
sis of a new application of RNAi for the control of diseases
caused by plant viruses. This approach differs from strate-
gies based on transgenic plants capable of expressing dsRNA
that are immune to viruses (Smith et al., 2000; Kalantidis
et al., 2002), but still relies on PTGS as a means to achieve
PDR. We established a simple and inexpensive procedure
to produce large amounts of dsRNA derived from viral se-
quences in bacteria, with the view of providing the starting
point for the industrial production of these molecules. When
applied on plants, the dsRNAs caused specific degradation
of the homologous viral RNA with the resultant protection
against virus infections. The infectivity of virtually any RNA
virus in plants can now be inhibited with the correspond-
ing dsRNA, without detrimental effect on plant surfaces. We
propose that exogenously supplied dsRNA would mimic the
dsRNA intermediate involved in viral replication, triggering
the initiation step of PTGS that leads to the production of
siRNAs which are incorporated into the nuclease complex
(RISC) responsible for the degradation of the target RNA
(Hammond et al., 2000). Thus, the invading virus contain-
ing sequences homologous to the dsRNA is recognized and
degraded by this defense mechanism.
As mentioned above, numerous studies have established
that synthetic siRNAs used to target viral genes can con-
fer resistance to several cytoplasmically replicating animal
viruses. To date, siRNAs have only been occasionally used
as inducers of RNAi in plants (Klahre et al., 2002), proba-
bly due to the likelihood of triggering silencing using long
dsRNA molecules. Moreover, delivery of a unique siRNA
species into a cell may compromise the efficiency of RNAi,
since it may not be sufficient to activate a potent silenc-
ing response (Paddison et al., 2002). At present, little is
known about the structural and sequence requirements that
govern target recognition and subsequent processing by the
siRNA-containing RISC complex. For instance, it is not pos-
sible to anticipate the most optimal viral RNA region for
siRNA targeting. Also, the viral genome might be inacces-
sible to some siRNAs, perhaps due to being enclosed in a
compacted protein-RNA complex or to possessing a partic-
ular RNA structure (Bitko and Barik, 2001). Because plant
cells lack non-specific inhibitory effects activated by dsRNA
in mammals, limitations caused by siRNAs can be circum-
vented by the use of long dsRNA. Processing of long dsR-
NAs by the PTGS machinery present in plant cells should
generate a mixture of siRNAs corresponding to different re-
gions of the precursor dsRNA. Sequence-specific degrada-
tion of viral RNA would occur as a result of multiple siRNAs
functioning cooperatively at different positions of the target
RNA. This most likely enhances the robustness of antiviral
RNAi response in plants.
We anticipate that inducible expression of dsRNA in bac-
teria could easily be scaled-up as a rapid and cost-effective
 F. Tenllado et al. / Virus Research 102 (2004) 85–96
expression system for a large variety of viral dsRNAs. In
our experience, average yield of dsRNA produced in bacte-
rial cells was 4 ␮g/ml of culture. Large fermenters of up to
100,000 l, such as those used for the commercial production
of entomopathogenic nematodes for the control of insect
pests (Georgis and Manwelier, 1994), would potentially
meet the enormous demand for the use of these interfering
products in agriculture. Moreover, bacterial cell cultures
provide a practicable and reproducible approach for mass
production of dsRNA. They offer the possibility to be easily
pelleted, stored and distributed, without losing their inter-
fering properties with plant virus infection. Large-scale pro-
duction of dsRNA for its use in agriculture can be achieved
only through simple and inexpensive downstream process-
ing steps. Lysing cells with a French Press or any other
mechanical means is a rapid and efficient procedure that
avoids the use of chemicals and time-consuming protocols
to purify nucleic acids from bacteria. The demonstration
under experimental conditions that dsRNA also works in
interfering with virus infection when bacterial crude prepa-
rations are sprayed onto plant surfaces, further contributes
to making this technology widely available and applicable
for deployment in the field.
Although this approach is very exciting, important prob-
lems must be solved before this RNAi-based technology can
be used as a tool for virus control in the field on a commercial
scale. First, plants that are protected against mechanically in-
oculated target viruses may not be protected when the virus
is inoculated by natural vectors. This is especially problem-
atic since the vast majority of plant viruses are transmitted
from plant to plant in nature by invertebrate vectors which,
in many cases, release the virus particles directly inside the
plant cell (López-Moya, 2002). This probably precludes
the possibility that the interfering dsRNA sprayed onto leaf
surfaces enters the cell together with the virus. Since most
ssRNA plant viruses, including those in the economically
important group of potyviruses, are transmitted by aphids
in nature, we are especially interested in using RNAi tech-
nology to extend the resistance against viral inoculation by
aphids. Studies have been initiated trying to facilitate the
penetration of bacterially expressed dsRNA into plant tis-
sues other than through epidermal cells. Finding appropriate
coadyuvants would provide a starting line for the design
of new formulations based on dsRNA aimed at protecting
plants against this natural means of infection. Successful
interference with virus infection by RNAi would reduce the
use of pesticides in agriculture thus minimizing potential
ecological and environmental risks. At present, the most
widely used strategy to control virus diseases is by chemical
control of viral vectors. Alternatives to chemical pesticides
are desirable for several reasons, such as an increased aware-
ness of the potentially harmful effects of chemical residues
in food and the environment, and the increasing resistance of
insects to insecticides. Second, dsRNA induces RNA silenc-
ing in the inoculated tissue but not in the upper parts of the
plants, i.e. intercellular signaling to trigger RNA silencing
93
in distal tissues does not occur. As a result, protection would
be feasible only if adequate coverage of the whole plant
with interfering dsRNA is guaranteed. Alternatively, the use
of coadyuvants in the dsRNA-based formulations could fa-
cilitate the spread of interfering products to distant parts of
the plant triggering a systemic protective effect against virus
infection. Third, dsRNA retains its activity as an inducer of
RNAi against viruses only for a transitory period of time.
This fact creates a number of inconveniences since frequent
protective treatment may be impracticable or not be appli-
cable at the time required in open-field crops. Alternatively,
application of formulations based on viral dsRNA made
at appropriate times during the growing season, coincident
with vector epidemics, should have the potential of activat-
ing PTGS mechanisms of virus resistance in crops grown
under plastic or glass. Fourth, RNAi mediated by direct
delivery of dsRNA onto plants is intended to be a practical
tool for virus control on a commercial scale. Unfortunately,
at present, large-scale applications have not been accom-
plished yet so it remains unknown whether immunization
would be successful. In addition, the protective effect has so
far been tested in model plants but not in crop species. Cou-
pled with these current problems, costs regarding long-term
applications in crops should be evaluated before directly de-
livering dsRNA as a means to initiate RNAi is implemented
in a therapeutic setting. Another determining consideration
to keep in mind is that spraying dsRNA-based formula-
tions could be more competitive with the costs of chemical
insecticide programmes currently used to control insect
vectors.
At present, provided that an efficient transformation tech-
nique is available, RNAi based on stable transformation of
host plants with viral sequences in the form of constructs
directing dsRNA formation, is likely the most efficient
method to confer resistance against pathogenic viruses in
crops. Biotechnological utilization of RNAi-based engi-
neered resistance is appealing for biosafety reasons as well.
RNA silencing-based resistance does not involve the trans-
genic production of functional viral genes or proteins, nor
does it lead to the presence of transgene RNA thus mini-
mizing potential ecological and environmental risks (Tepfer,
2002). Since little or no transgene mRNA is accumulated in
plant cells, there is essentially no template for events such
as complementation, heterologous encapsidation, synergy,
and recombination. However, despite the enormous benefit
for agriculture, questions concerning the potential ecologi-
cal impact of virus-resistant transgenic plants have signifi-
cantly limited their use. For instance, reversal of PTGS by
virus-encoded suppressors would severely limit the biotech-
nological applications of RNA silencing-based transgenic
resistance. Infection of a PTGS-based virus-resistant trans-
genic plant by a nonhomologous virus with silencing sup-
pression capabilities, may reverse the silencing status of the
transgene resulting in high expression of the transgene and
loss of transgene-homologous virus resistance (Savenkov
and Valkonen, 2002). This creates a situation in which
94 F. Tenllado et al. / Virus Research 102 (2004) 85–96
recombination between transgene mRNA and the RNA of
the infecting virus could potentially occur.
RNAi technology based on exogenously supplied dsRNA
has several apparent advantages compared to the require-
ments for regenerating transgenic plants capable to interfere
with plant virus infections. The experimental procedure
does not require sophisticated equipment and is inexpen-
sive, compared to labor- and cost-intensive production of
transgenic plants encoding virus genes. A further advantage
of the RNAi technology described here is the facility to de-
liver several dsRNAs targeting different viruses in a single
formulation, so that multiple resistance traits can be gener-
ated. In stable transgenic plants such manipulation would
require sequential transformations or crosses between trans-
genic plants that could take prolonged periods of time.
Although PTGS triggered by exogenously supplied dsRNA
could also be broken down by suppressor viruses, dsRNA
molecules can be designed to lack transcriptional regulatory
regions, such as the CaMV 35S promoter commonly used
in transgenic plants. This would probably impair overaccu-
mulation of interfering products on dsRNA-treated plants,
reducing the chance of RNA recombination.
6. Conclusions
Efficient control of plant viruses affecting economi-
cally important crop species represents one of the major
challenges in the development of effective and sustainable
measures of plant health management. As a necessary com-
plement to traditional methods of virus control, emerging
technologies should be based on a better understanding of
the natural defense mechanisms of the host plants and the
counteracting strategies displayed by the viruses. RNA si-
lencing and other related pathways play an important role
in the regulation of basic cellular processes such as gene ex-
pression and heterochromatin formation. In addition, RNA
silencing represents an adaptive antiviral defense mecha-
nism in plants. RNA silencing is activated in the presence of
either dsRNA molecules or siRNAs leading to degradation
of homologous RNAs in a sequence-specific manner. The
discovery that inducer molecules can trigger inactivation of
target RNAs when directed against virus genomes opens
the way to investigate the development of methodologies
to deliver RNA silencing inducers into the plant tissue with
the purpose of interfering with a specific step of virus repli-
cation and/or movement. In this article, we have discussed
the concept of RNA silencing as an elegant natural antiviral
mechanism in plants and dissected the prospective of RNAi
technology as a tool for control of virus diseases in crop
plants.
Our long-term goal is the development of RNAi as a
biotechnological approach to control plant virus diseases
in crops in an effective and environmentally safety manner.
Although much progress has been made recently in our
understanding of PTGS, much still remains to be discov-
ered. The successful exploitation of bacterially expressed,
exogenously supplied dsRNA for the control of plant virus
diseases by RNAi seems to be promising.
Acknowledgements
We wish to thank Natalia Wright for critical reading of
the manuscript. F.T has been the recipient of a “Ramón y
Cajal” contract from MCyT (Spain) and he is currently Se-
nior Investigator of CSIC. C.L. is the recipient of an “I3P”
contract from CSIC (Spain). J.R.D.-R. is Research Profes-
sor Staff Scientist of CSIC. The research done in the lab-
oratory has been founded with grants BIO2000-0914 and
BIO2003-03428 from CICYT-MCyT and 07M/0123/2000
and 07G/0043/2003 from Comunidad de Madrid (Spain).
References
Anandalakshmi, R., Pruss, G.J., Ge, X., Marathe, R., Mallory, A.C.,
Smith, T.H., Vance, V.B., 1998. A viral suppressor of gene silencing
in plants. Proc. Natl. Acad. Sci. U.S.A. 95, 13079–13084.
Baulcombe, D.C., 1999. Viruses and gene silencing in plants. Arch. Virol.
Suppl. 15, 189–201.
Bernstein, E., Caudy, A.A., Hammond, S.M., Hannon, G.J., 2001. Role
for a bidentate ribonuclease in the initiation step of RNA interference.
Nature 409, 363–366.
Bitko, V., Barik, S., 2001. Phenotypic silencing of cytoplasmic genes
using sequence-specific double-stranded short interfering RNA and its
application in the reverse genetics of wild type negative-strand RNA
viruses. BMC Biotechnol. 1, 34–46.
Brigneti, G., Voinnet, O., Li, W.X., Ji, L.H., Ding, S.W., Baulcombe, D.C.,
1998. Viral pathogenicity determinants are suppressors of transgene
silencing in Nicotiana benthamiana. EMBO J. 17, 6739–6746.
Cayley, P.J., White, R.F., Antoniw, J.F., Walesby, N.J., Kerr, I.M., 1982.
Distribution of the ppp(A2’p)nA-binding protein and interferon-related
enzymes in animals, plants, and lower organisms. Biochem. Biophys.
Res. Commun. 108, 1243–1250.
Dalmay, T., Hamilton, A., Rudd, S., Angell, S., Baulcombe, D.C., 2000.
An RNA-dependent RNA polymerase gene in Arabidopsis is required
for posttranscriptional gene silencing mediated by a transgene but not
by a virus. Cell 101, 543–553.
Denli, A.M., Hannon, G.J., 2003. RNAi: an ever-growing puzzle. Trends
Biochem. Sci. 28, 196–201.
Dı́az-Ruı́z, J.R., Kaper, J.M., 1978. Isolation of viral double-stranded
RNAs using a LiCl fractionation procedure. Prep. Biochem. 8, 1–17.
Djikeng, A., Shi, H., Tschudl, C., Ullu, E., 2001. RNA interference in
Trypanosoma brucei: cloning of small interfering RNAs provides evi-
dence for retroposon-derived 24–26-nucleotide RNAs. RNA 7, 1522–
1530.
Elbashir, S.M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., Tuschl,
T., 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference
in cultured mammalian cells. Nature 411, 494–498.
Fagard, M., Boutet, S., Morel, J.-B., Bellini, C., Vaucheret, H., 2000.
AGO-1, QDE-2, and RDE-1 are related proteins required for
post-transcriptional gene silencing in plants, quelling in fungi, and
RNA interference in animals. Proc. Natl. Acad. Sci. U.S.A. 97, 11650–
11654.
Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver, S.E., Mello,
C.C., 1998. Potent and specific genetic interference by double-stranded
RNA in Caenorhabditis elegans. Nature 391, 806–811.
Georgis, R., Manwelier, S.A., 1994. Entomopathogenic nematodes: a
developing biological control technology. Agric. Zool. Rev. 6, 63–94.
 F. Tenllado et al. / Virus Research 102 (2004) 85–96
Gil, J., Esteban, M., 2000. Induction of apoptosis by the dsRNA-dependent
protein kinase (PKR): mechanism of action. Apoptosis 5, 107–114.
Gitlin, L., Karelsky, S., Andino, R., 2002. Short interfering RNA confers
intracellular antiviral immunity in human cells. Nature 418, 430–434.
Goldbach, R., Bucher, E., Prins, M., 2003. Resistance mechanisms to
plant viruses: an overview. Virus Res. 92, 207–212.
Hamilton, A., Voinnet, O., Chappell, L., Baulcombe, D.C., 2002. Two
classes of short interfering RNA in RNA silencing. EMBO J. 21,
4671–4679.
Hamilton, A.J., Baulcombe, D.C., 1999. A species of small antisense RNA
in posttranscriptional gene silencing in plants. Science 286, 950–952.
Hamilton, A.J., Brown, S., Yuanhai, H., Ishizuka, M., Lowe, A., Solis,
A.G.A., Grierson, D., 1998. A transgene with repeat DNA causes
high frequency, post-transcriptional suppression of ACC-oxidase gene
expression in tomato. Plant J. 15, 737–746.
Hammond, S., Bernstein, E., Beach, D., Hannon, G., 2000. An
RNA-directed nuclease mediates post-transcriptional gene silencing in
Drosophila cells. Nature 404, 293–296.
Hammond, S.M., Caudy, A.A., Hannon, G.J., 2001. Post-transcriptional
gene silencing by double-stranded RNA. Nat. Rev. Genet. 2, 110–119.
Hannon, G.J., 2002. RNA interference. Nature 418, 244–251.
Hutvágner, G., Zamore, P.D., 2002. A microRNA in a multiple-turnover
RNAi enzyme comple. Science 297, 2056–2060.
Jacque, J.M., Triques, K., Stevenson, M., 2002. Modulation of HIV-1
replication by RNA interference. Nature 418, 435–438.
Johansen, L.K., Carrington, J.C., 2001. Silencing on the spot. Induction
and suppression of RNA silencing in the Agrobacterium-mediated
transient expression system. Plant Physiol. 126, 930–938.
Jorgensen, R.A., Atkinson, R.G., Forster, R.L., Lucas, W.J., 1998. An
RNA-based information superhighway in plants. Science 279, 1486–
1487.
Kalantidis, K., Psaradakis, S., Tabler, M., Tsagris, M., 2002. The occur-
rence of CMV-specific short RNAs in transgenic tobacco expressing
virus-derived double-stranded RNA is indicative of resistance to the
virus. Mol. Plant-Microbe Interact. 15, 826–833.
Ketting, R.F., Haverkamp, T.H., van Luenen, H.G., Plasterk, R.H., 1999.
Mut-9 of C. elegans, required for transposon silencing and RNA
interference, is a homolog of Werner syndrome helicase and RNaseD.
Cell 99, 133–141.
Klahre, U., Crete, P., Leuenberger, S.A., Iglesias, V.A., Meins, F., 2002.
High molecular weight RNAs and small interfering RNAs induce
systemic posttranscriptional gene silencing in plants. Proc. Natl. Acad.
Sci. U.S.A. 99, 11981–11986.
Li, H., Li, W.X., Ding, S.W., 2002. Induction of suppression of RNA
silencing by an animal virus. Science 296, 1319–1321.
López-Moya, J.J., 2002. Genes involved in insect-mediated transmission
of plant viruses. In: Khan, J.A., Dijkstra, J. (Eds.), Plant Viruses as
Molecular Pathogens. The Haworth Press Inc., London, pp. 31–61.
Llave, C., Kasschau, K.D., Carrington, J.C., 2000. Virus-encoded suppres-
sor of posttranscriptional gene silencing targets a maintenance step in
the silencing pathway. Proc. Natl. Acad. Sci. U.S.A. 97, 13401–13406.
Llave, C., Xie, Z., Kasschau, K.D., Carrington, J.C., 2002. Cleavage
of Scarecrow-like mRNA targets directed by a class of Arabidopsis
miRNA. Science 297, 2053–2056.
Mlotshwa, S., Voinnet, O., Mette, M.F., Matzke, M., Vaucheret, H., Ding,
S.W., Pruss, G., Vance, V.B., 2002. RNA silencing and the mobile
silencing signal. Plant Cell 14 (Suppl.), S289–S301.
Napoli, C., Lemieux, C., Jorgensen, R., 1990. Introduction of a
chimeric chalcone synthase gene into Petunia results in reversible
co-suppression of homologous genes in trans. Plant Cell 2, 279–289.
Novina, C.D., Murray, M.F., Dykxhoorn, D.M., Beresford, P.J., Riess, J.,
Lee, S.K., Collman, R.G., Lieberman, J., Shankar, P., Sharp, P.A.,
2002. siRNA-directed inhibition of HIV-1 infection. Nat. Med. 8,
681–686.
Paddison, P.J., Caudy, A.A., Bernstein, E., Hannon, G.J., Conklin, D.S.,
2002. Short hairpin RNAs (shRNAs) induce sequence-specific silenc-
ing in mammalian cells. Genes Dev. 16, 948–958.
95
Palauqui, J.C., Balzergue, S., 1999. Activation of systemic acquired si-
lencing by localised introduction of DNA. Curr. Biol. 9, 59–66.
Pasquinelli, A.E., Ruvkun, G., 2002. Control of developmental timing by
microRNAs and their targets. Annu. Rev. Cell Dev. Biol. 18, 495–513.
Peña, L., Trad, J., Dı́az-Ruı́z, J.R., McGarvey, P.B., Kaper, J.M., 1994.
Cucumber mosaic virus protection in transgenic tobacco plants ex-
pressing monomeric, dimeric or partial sequences of a bening satellite
RNA. Plant Sci. 100, 71–81.
Powell-Abel, P., Nelson, R.S., De, B., Hoffmann, N., Rogers, S.G., Fraley,
R.T., Beachy, R.N., 1986. Delay of disease development in transgenic
plants that express the tobacco mosaic virus coat protein gene. Science
232, 738–743.
Pruss, G., Ge, X., Shi, X.M., Carrington, G.C., Vance, V.B., 1997.
Plant viral synergism: the potyviral genome encodes a broad-range
pathogenicity enhancer that transactivates replication of heterologous
viruses. Plant Cell 9, 1–11.
Ratcliff, F., MacFarlane, S., Baulcombe, D.C., 1999. Gene silencing with-
out DNA: RNA-mediated cross protection between viruses. Plant Cell
11, 1207–1215.
Ravelonandro, M., Scorza, R., Callahan, A., Levy, L., Jacquet, C., Mon-
sion, M., Damsteegt, V., 2000. The use of transgenic fruit trees as a
resistance strategy for virus epidemics: the plum pox (sharka) model.
Virus Res. 71, 63–69.
Romano, N., Macino, G., 1992. Quelling: transient inactivation of gene
expression in Neurospora crassa by transformation with homologous
sequences. Mol. Microbiol. 22, 3343–3353.
Ruiz, M.T., Voinnet, O., Baulcombe, D.C., 1998. Initiation and mainte-
nance of virus-induced gene silencing. Plant Cell 10, 937–946.
Sáenz, P., Salvador, B., Simón-Mateo, C., Kasschau, K.D., Carrington,
J.C., Garcı́a, J.A., 2002. Host-specific involvement of the HC protein
in the long-distance movement of potyviruses. J. Virol. 76, 1922–1930.
Sanford, J.C., Johnston, S.A., 1985. The concept of parasite-derived re-
sistance. Deriving resistance genes from the parasite’s own genome.
J. Theor. Biol. 113, 395–405.
Savenkov, E.I., Valkonen, J.P., 2002. Silencing of a viral RNA silencing
suppressor in transgenic plants. J. Gen. Virol. 83, 2325–2335.
Schweizer, P., Pokorny, J., Schulze-Lefert, P., Dudler, R., 2000.
Double-stranded RNA interferes with gene function at the single-cell
level in cereals. Plant J. 24, 895–903.
Smith, N., Singh, S., Wang, M.B., Stoutjesdijk, P., Green, A., Waterhouse,
P.M., 2000. Total silencing by intron-spliced hairpin RNAs. Nature
407, 319–320.
Sonti, R.V., Chiurazzi, M., Wong, D., Davies, C.S., Harlow, G.R., Mount,
D.W., Signer, E.R., 1995. Arabidopsis mutants deficient in T-DNA
integration. Proc. Natl. Acad. Sci. U.S.A. 92, 11786–11790.
Tabara, H., Grishok, A., Mello, C.C., 1998. RNAi in C. elegans: soaking
in the genome sequence. Science 282, 430–431.
Tang, G., Reinhart, B.J., Bartel, D.P., Zamore, P.D., 2003. A biochemical
framework for RNA silencing in plants. Genes Dev. 17, 49–63.
Tavernarakis, N., Wang, S.L., Dorovkov, M., Ryazanov, A., Driscoll, M.,
2000. Heritable and inducible genetic interference by double-stranded
RNA encoded by transgenes. Nat. Genet. 24, 180–183.
Tenllado, F., Barajas, D., Vargas, M., Atencio, F.A., González-Jara, P.,
Dı́az-Ruı́z, J.R., 2003a. Transient expression of homologous hairpin
RNA causes interference with plant virus infection and is overcome
by a virus encoded suppressor of gene silencing. Mol. Plant-Microbe
Interact. 16, 149–158.
Tenllado, F., Dı́az-Ruı́z, J.R., 1999. Complete resistance to pepper mild
mottle tobamovirus mediated by viral replicase sequences partially
depends on transgene homozygosity and is based on a gene silencing
mechanism. Transgenic Res. 8, 83–93.
Tenllado, F., Dı́az-Ruı́z, J.R., 2001. Double-stranded RNA-mediated in-
terference with plant virus infection. J. Virol. 75, 12288–12297.
Tenllado, F., Garcı́a-Luque, I., Serra, M.T., Dı́az-Ruı́z, J.R., 1995. Nico-
tiana benthamiana plants transformed with the 54-kDa region of the
pepper mild mottle tobamovirus replicase gene exhibit two types of
resistance responses against viral infection. Virology 211, 170–183.
96 F. Tenllado et al. / Virus Research 102 (2004) 85–96
Tenllado, F., Garcı́a-Luque, I., Serra, M.T., Dı́az-Ruı́z, J.R., 1996. Re-
sistance to pepper mild mottle tobamovirus conferred by the 54-kDa
gene sequence in transgenic plants does not require expression of the
wild-type 54-kDa protein. Virology 219, 330–335.
Tenllado, F., Martinez-Garcia, B., Vargas, M., Dı́az-Ruı́z, J.R., 2003b.
Crude extracts of bacterially expressed dsRNA can be used to protect
plants against virus infections. BMC Biotechnol. 3, 3.
Tepfer, M., 2002. Risk assessment of virus-resistant transgenic plants.
Annu. Rev. Phytopathol. 40, 467–491.
Thomas, C.L., Jones, L., Baulcombe, D.C., Maule, A.J., 2001. Size
constraints for targeting post-transcriptional gene silencing and for
RNA-directed methylation in Nicotiana benthamiana using a potato
virus X vector. Plant J. 25, 417–425.
Tijsterman, M., Ketting, R.F., Plasterk, R.H., 2002. The genetic of RNA
silencing. Annu. Rev. Genet. 36, 489–519.
Timmons, L., Court, D.L., Fire, A., 2001. Ingestion of bacterially ex-
pressed dsRNAs can produce specific and potent genetic interference
in Caenorhabditis elegans. Gene 263, 103–112.
Timmons, L., Fire, A., 1998. Specific interference by ingested dsRNA.
Nature 395, 854.
Vaistij, F.E., Jones, L., Baulcombe, D.C., 2002. Spreading of RNA tar-
geting and DNA methylation in RNA silencing requires transcription
of the target gene and a putative RNA-dependent RNA polymerase.
Plant Cell 14, 857–867.
Vazquez Rovere, C., del Vas, M., Hopp, H.E., 2002. RNA-mediated virus
resistance. Curr. Opin. Biotechnol. 13, 167–172.
Voinnet, O., 2001. RNA silencing as a plant immune system against
viruses. Trends Genet. 17, 449–459.
Voinnet, O., Lederer, C., Baulcombe, D.C., 2000. A viral movement
protein prevents spread of the gene silencing signal in Nicotiana
benthamiana. Cell 103, 157–167.
Volpe, T.A., Kidner, C., Hall, I.M., Teng, G., Grewal, S.L., Martienssen,
R.A., 2002. Regulation of heterochromatic silencing and histone H3
lysine-9 methylation by RNAi. Science 297, 1833–1837.
Wang, M.B., Abbott, D.C., Waterhouse, P.M., 2000. A single copy
of a virus-derived transgene encoding hairpin RNA gives immu-
nity to barley yellow dwarf virus. Mol. Plant Pathol. 1, 347–
356.
Waterhouse, P.M., Gramham, M., Wang, M.B., 1998. Virus resistance and
gene silencing in plants can be induced by simultaneous expression of
sense and antisense RNA. Proc. Natl. Acad. Sci. U.S.A. 95, 13959–
13964.
Waterhouse, P.M., Wang, M.B., Lough, T., 2001. Gene silencing as an
adaptive defense against viruses. Nature 411, 834–842.
 Virus Research 102 (2004) 97–108

Plant viral suppressors of RNA silencing

Braden M. Roth, Gail J. Pruss, Vicki B. Vance∗
Department of Biological Sciences, University of South Carolina, Columbia, SC 29208, USA

Abstract

RNA silencing is an ancient eukaryotic pathway in which double stranded RNA (dsRNA) triggers destruction of related RNAs in the cell.
Early studies in plants pointed to a role for this pathway as a defense against viruses. Most known plant viruses have RNA genomes and
replicate via dsRNA intermediates, thereby serving as potent inducers of RNA silencing early in replication and as silencing targets later in
infection. Because RNA silencing is an antiviral mechanism, it is not surprising that many plant viruses encode suppressors of RNA silencing.
This review focuses on the currently known plant virus encoded suppressors of silencing and the functional assays used to identify these
proteins. Because they interfere with silencing at different points in the pathway, these viral suppressors are powerful tools to help unravel
the mechanism of RNA silencing in plants.
© 2004 Elsevier B.V. All rights reserved.

Keywords: RNA silencing; RNAi; Viral suppressor protein; HC-Pro; CMV 2b

1. Introduction Vance et al., 1995). Expression of HC-Pro in transgenic

RNA silencing is an adaptive defense mechanism that is
triggered by double stranded RNA (dsRNA). Three initially
unrelated lines of research led to the recognition of RNA si-
lencing as an important means of defense against viruses and
other nucleic acid invaders. The discovery that plant viruses
encode suppressors of RNA silencing is a key element in
the story. The first line of research led to the discovery of
transgene-induced RNA silencing. Attempts to over-express
endogenous genes by introducing additional copies resulted
instead in turning off the endogenous gene as well as the
transgene (Napoli et al., 1990; Smith et al., 1990; van der
Krol et al., 1990). The next piece of the puzzle came from
studies of pathogen-derived resistance in which RNA silenc-
ing directed against a viral transgene provided resistance to
any virus carrying the targeted sequence (Baulcombe, 1996;
Dougherty and Parks, 1995; English et al., 1996; Goodwin
et al., 1996; Lindbo et al., 1993; Smith et al., 1994). Thus,
viruses could be targets of RNA silencing. The third clue
came from studies of synergistic viral diseases caused by
certain pairs of co-infecting viruses. A viral protein called
helper component proteinase (HC-Pro) was shown to medi-
ate one class of viral synergistic disease (Shi et al., 1997;
∗ Corresponding author.
E-mail address: vance@biol.sc.edu (V.B. Vance).
plants allowed a broad range of heterologous viruses to ac-
cumulate beyond the normal level, suggesting that HC-Pro
blocked a general plant defense mechanism (Pruss et al.,
1997). Remarkably, the mechanism blocked by HC-Pro was
found to be RNA silencing (Anandalakshmi et al., 1998;
Brigneti et al., 1998; Kasschau and Carrington, 1998). Since
the initial demonstration that HC-Pro blocks RNA silencing,
many other plant viral suppressors of silencing have been
identified (see Table 1).
2. RNA silencing
The term RNA silencing refers to a set of related pathways
found in a broad range of eukaryotic organisms. Genetic and
biochemical experiments have established a general mecha-
nistic model for these related pathways and identified factors
that are required for RNA silencing in a variety of organ-
isms (Fig. 1). The process is initially triggered by dsRNA,
which can be introduced experimentally or arise from en-
dogenous transposons, replicating RNA viruses, or the tran-
scription of transgenes. The dsRNA trigger is cleaved by a
ribonuclease III (RNAse III)-like enzyme termed Dicer into
21–24 nucleotide duplexes termed short-interfering RNAs
(siRNAs) (Bernstein et al., 2001; Hamilton and Baulcombe,
1999; Zamore et al., 2000). The production of siRNAs by
Dicer is an ATP-dependent step (Bernstein et al., 2001;

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.020
 98
Table 1
Plant viral suppressors of RNA silencing
Genus Virus Suppressor Evidence Reference

Carmovirus Turnip crinkle virus (TCV) CP TCV infection does not reverse silencing. In agro-coinfiltration Qu et al. (2003) and Thomas et al. (2003)
assay, CP blocks sense and antisense induced local silencing and
prevents systemic silencing.
Closterovirus Beet yellows virus (BYV) p21 Suppresses inverted repeat (IR) induced local silencing in Reed et al. (2003)
Beet yellow stunt virus (BYSV) p22 agro-coinfiltration assay. BYV p21 corresponds to BYSV p22.
Cucumovirus Cucumber mosaic virus (CMV) 2b Infection with CMV or with PVX-2b vector blocks silencing. Li et al. (2002), see text
Tomato aspermy virus (TAV) Interferes with systemic signal (see text).
Furovirus Beet necrotic yellow vein virus (BNYVV) P14 Agro-coinfiltration assay with sense induced silencing. BNYVV Dunoyer et al. (2002)

B.M. Roth et al. / Virus Research 102 (2004) 97–108

P14 corresponds to PCV P15.
Geminivirus African cassava mosaic virus (ACMV) AC2 Infection with ACMV, PVX-AC2, or PVX-C2 reverses silencing. Dong et al. (2003), Voinnet et al. (1999)
Tomato yellow leaf curl virus-China (TYLCV-C) C2 Blocks sense induced silencing in agro-coinfiltration assay. AC2 and van Wezel et al. (2002)
and C2 are homologs.
Hordeivirus Barley stripe mosaic virus (BSMV) ␥b RNA mediated cross protection between PVX-GFP and TMV-GFP Yelina et al. (2002)
Poa semilatent virus (PSLV) vectors is eliminated when ␥b is expressed from the PVX vector.
Pecluvirus Peanut clump virus (PCV) P15 PCV infection blocks silencing. p15 blocks local and delays Dunoyer et al. (2002)
systemic sense-induced silencing in agro-coinfiltration assay.
Polerovirus Beet western yellows virus (BWYV) PO BWYV PO suppresses local but not systemic sense-induced Pfeffer et al. (2002)
Cucurbit aphid-borne yellows virus (CABYV) silencing in agro-coinfiltration assay. CABYV PO tested only on
local silencing.
Potexvirus Potato virus X (PVX) p25 PVX infection does not suppress silencing. In agro-coinfiltration, See text
p25 blocks systemic but not always local silencing (see text).
Potyvirus Potato virus Y (PVY) HC-Pro Evidence from multiple types of assay. Does not block systemic See text
Tobacco etch virus (TEV) silencing in stable expression grafting assay, but does in
agro-coinfiltration assay (see text).
Sobemovirus Rice yellow mottle virus (RYMV) P1 Infection with PVX-P1 viral vector reverses silencing. Voinnet et al. (1999)
Tenuivirusa Rice hoja blanca virus (RHBV) NS3 Agro-coinfiltration assay of sense induced local silencing. Bucher et al. (2003)
Tombusvirus Tomato bushy stunt virus (TBSV) P19 Limited activity in reversal of silencing; strong activity in Voinnet et al. (2003), Qu and Morris
Cymbidium ringspot virus (CymRSV) agro-coinfiltration (see text). AMCV (artichoke mottled crinkle (2002) and Takeda et al. (2002), see text
virus) P19 also works as a suppressor.

Tospovirusa Tomato spotted wilt virus (TSWV) NSs TSWV infection reverses silencing. In agro-coinfiltration, NSs Bucher et al. (2003) and Takeda et al. (2002)
suppressed sense, but not IR, induced local and systemic silencing.
a Tospoviruses and tenuiviruses replicate in their insect vectors and in plants.
 B.M. Roth et al. / Virus Research 102 (2004) 97–108
Fig. 1. RNA silencing pathway.
Zamore et al., 2000) and probably involves interactions
with other proteins, including an argonaute-like protein,
a dsRNA binding protein, and an RNA helicase (Tabara
et al., 2002). There are four Dicer-like (DCL) homologs
in Arabidopsis; however, the specific enzyme responsible
for siRNA production in plants has not yet been identified.
The siRNAs produced from a fully double-stranded RNA
substrate by Dicer have distinctive characteristics: they
represent both polarities and have two nucleotide 3 over-
hangs with 5 phosphate and 3 hydroxyl groups (Elbashir
et al., 2001a,b). In another ATP-dependent step (Nykanen
et al., 2001), the siRNAs are denatured and incorporated
into a multi-subunit endonuclease silencing complex called
RNA-induced silencing complex (RISC; Hammond et al.,
2000). Within the activated RISC, single-stranded siRNAs
act as guides to bring the complex into contact with com-
plementary mRNAs and thereby cause their degradation
(Bernstein et al., 2001; Elbashir et al., 2001a; Hammond
et al., 2000, 2001; Zamore et al., 2000).
99
One fascinating aspect of RNA silencing is that it is
non-cell-autonomous, and this feature may reflect the antivi-
ral nature of the process (for a recent review, see Mlotshwa
et al., 2002b). Virus infection usually starts with entry via
a small wound. The virus replicates in the initially infected
cell and then moves into adjacent cells, spreading from cell
to cell until it enters the vascular system, which allows rapid
movement to distant parts of the plant. In response, the host
plant initiates RNA silencing against the viral RNA and pro-
duces the mobile silencing signal. The mobile signal moves
along the same route the virus takes. Thus, the plant and the
virus enter a race. If the virus moves ahead of the signal, it
can establish infection when it enters the distant cells. How-
ever, if the mobile silencing signal gets there first, the virus
will enter the distant cell only to find itself targeted by RNA
silencing, and the infection will fail to become systemic.
One major class of viral suppressors comprises proteins that
block systemic silencing, suggesting the co-evolution of de-
fense and counter-defense between the host plant and the
invading virus at the level of systemic spread.
3. Functional assays used to identify suppressors of
RNA silencing
Three major approaches have been widely used to iden-
tify plant viral suppressors of RNA silencing: (1) transient
expression assays, (2) the reversal of silencing assay, and (3)
stable expression assays. These assays are described below
and shown in cartoon form in Figs. 2 through 4.
3.1. Transient expression assays—Agrobacterium
co-infiltration
This approach provides a rapid and easy test of sup-
pressor activity and is currently the technique most com-
monly used to identify viral suppressors (Llave et al., 2000;
Voinnet et al., 2000). The method makes use of a commonly
used bacterial pathogen of plants, Agrobacterium tumefa-
ciens. The Agrobacterium serves two purposes: one strain
is used to induce RNA silencing of a reporter gene (usually
green fluorescent protein (GFP)), and another strain is used
to express the candidate suppressor. The overall strategy is
to co-infiltrate mixtures of the two bacterial strains (one act-
ing to induce silencing, the other to suppress it) into a plant
leaf and then examine the infiltrated patch over time for si-
lencing of the reporter (Fig. 2A). Nicotiana benthamiana is
well suited to these assays because the leaves are easily in-
filtrated and produce high quantities of protein in response
to agro-infiltration. Non-transgenic N. benthamiana as well
as N. benthamiana expressing the reporter gene can be used.
In a typical experiment with Agrobacterium expressing GFP
as the inducer, the infiltrated patch initially expresses high
levels of GFP and glows bright green under ultraviolet (UV)
light (Fig. 3A, first panel). However, in three to five days,
the procedure triggers local RNA silencing, and the patch
100 B.M. Roth et al. / Virus Research 102 (2004) 97–108

Fig. 2. Cartoon guide to transient expression assays. (A) Assay for suppressors of local silencing. (B) Assay for suppressors of systemic silencing.

Fig. 3. Agrobacterium-induced systemic silencing (A) and cartoon guide to reversal of silencing assay (B).
 B.M. Roth et al. / Virus Research 102 (2004) 97–108
becomes a dirty red color under UV light due to a mixture
of green from residual GFP and red from chlorophyll in the
leaves (Fig. 3A, second panel). If the candidate suppressor
expressed from the co-infiltrated Agrobacterium interferes
with RNA silencing, the patch will remain bright green: if it
does not, the patch will turn red (Fig. 2A). Variations of this
technique include using Agrobacterium expressing inverted
repeat (IR) or viral cDNA constructs to induce silencing
in combination with or in place of Agrobacterium express-
ing a sense gene construct (Johansen and Carrington, 2001;
Voinnet et al., 2000).
Another variation of the transient expression approach has
been widely used to investigate the effect of silencing sup-
pressors on the mobile silencing signal (Fig. 2B). For this
purpose, infiltration is performed on a transgenic N. ben-
thamiana line that expresses GFP (line 16C, Ruiz et al.,
1998). Early experiments using an Agrobacterium strain ex-
pressing a sense GFP construct as the inducer of silencing
demonstrated that local induction of GFP silencing produced
a mobile silencing signal that moved out of the infiltrated
spot, along veins, and into surrounding tissues, leaving a trail
of red marking its path (Fig. 3A, third panel) and eventually
silencing GFP throughout the whole plant (Fig. 3A, fourth
panel) (Voinnet and Baulcombe, 1997). Thus, the ability
of suppressors to interfere with systemic silencing can be
tested in this system by co-infiltrating the inducer together
with a putative suppressor of silencing and observing long
enough to determine whether the plant becomes systemi-
cally silenced (Fig. 2B).
3.2. Reversal of silencing assay
This versatile assay can be used to first identify candi-
date viruses that may suppress silencing as a major line of
counter-defense (Fig. 3B; Brigneti et al., 1998). A modifi-
cation of the same technique can then be used to identify
the specific viral gene product that suppresses silencing. The
overall strategy is to infect a silenced plant with the candi-
date virus and determine whether the silenced phenotype is
reversed (Fig. 3B). The most common version of this tech-
nique uses the GFP-expressing transgenic N. benthamiana
line 16C described above. Soon after germination (at about
the four-leaf stage), the plant is infiltrated with Agrobac-
terium expressing GFP, which triggers local silencing and
then systemic silencing as described above. Ultimately, the
plant becomes completely silenced for GFP (red in UV light)
(Fig. 3A, fourth panel). At this point, the plant is inocu-
lated with the virus being tested for suppressor activity. If
virus infection allows the plant to express GFP, the infect-
ing virus likely encodes a suppressor of silencing (Voinnet
et al., 1999). Individual genes can be assayed for suppres-
sor activity using a modification of the reversal of silencing
technique that employs potato virus X (PVX). PVX is an ef-
ficient vector for systemic expression of heterologous genes
in N. benthamiana and does not, itself, encode a suppres-
sor of silencing that works in the reversal of silencing assay.
101
Thus, candidate suppressors from heterologous viruses can
be tested in the reversal of silencing assay by expressing
them from a PVX vector (Fig. 3B). Many viral suppressors
have been identified by expression from PVX in this manner
(Table 1; Voinnet et al., 1999).
3.3. Stable expression assays
In this approach, a stable transgenic line expressing a can-
didate suppressor of silencing (usually initially identified in
one of the previous two assays) is crossed to a series of
well-characterized transgenic lines silenced for a reporter
gene (Fig. 4A; Anandalakshmi et al., 1998; Kasschau and
Carrington, 1998). An advantage of this approach is that
it offers the opportunity to examine the effect of the sup-
pressor on different well-defined types of transgene-induced
RNA silencing, thus, providing information about the mech-
anism of suppression. The stable expression assays are also
well suited to investigate the role of suppressors in sys-
temic silencing using grafting (Fig. 4B; Guo and Ding, 2002;
Mallory et al., 2001, 2003). In these assays, the ability of a
plant to send a mobile silencing signal is assayed by graft-
ing an expressing line onto the top of it (the bottom plant
is called the rootstock and the expressing plant grafted onto
the top is called the scion) (Palauqui et al., 1997). If the sup-
pressor of silencing in the rootstock blocks either the pro-
duction or movement of the systemic silencing signal, then
the transgene in the scion will continue to be expressed. If
a suppressor does not block the systemic silencing signal
in either of these ways, then the transgene in the scion will
become silenced (Fig. 4B).
4. Mechanism of suppression
The assays discussed above have proven extremely use-
ful as rapid and sensitive methods to identify suppressors
of silencing. They have also enabled rudimentary charac-
terization of suppressor mechanism, but conflicting results
from different assays have made it difficult to draw firm con-
clusions for many suppressors. Interestingly, the currently
known suppressors share no obvious similarities at either
the nucleic acid or the protein level, perhaps reflecting dif-
ferences at the mechanistic level as well. At present, two
major classes of suppressor action have been identified.
4.1. Suppressors that affect small RNA metabolism
Many suppressors reduce the accumulation of siRNAs,
raising the possibility that silencing is blocked at the step
at which Dicer processes the dsRNA that triggers silencing.
Thus, it may be that many suppressors prevent silencing by
blocking production of the siRNAs that provide the sequence
specificity of the process. A second mechanism involving
siRNAs is exemplified by the suppressor P19, which has
been shown to bind siRNAs, perhaps sequestering them and
102 B.M. Roth et al. / Virus Research 102 (2004) 97–108
Fig. 4. Cartoon guide to stable expression assays. (A) Genetic crosses with silenced transgenic lines. (B) Grafting assay for suppression of systemic
silencing.
thereby blocking their function (Silhavy et al., 2002). Inter-
estingly, the effect of suppressors on small RNA metabolism
may extend to other types of small regulatory RNAs. For
example the silencing suppressor HC-Pro affects the accu-
mulation not only of the siRNAs that mediate silencing but
also of endogenous microRNAs (miRNAs), which have been
implicated in development (Kasschau et al., 2003; Mallory
et al., 2002b). Surprisingly, although HC-Pro blocks the ac-
cumulation of siRNAs, it enhances the accumulation of miR-
NAs (Mallory et al., 2002b). Even more interesting, there
is evidence that HC-Pro might block the function of miR-
NAs (Kasschau et al., 2003). It remains to be seen if other
viral suppressors also affect the miRNA pathway, perhaps
using that pathway to turn off expression of genes required
for silencing.
4.2. Suppressors that affect systemic silencing
Systemic silencing can be assayed in either transient ex-
pression experiments or in experiments with stably trans-
formed transgenic lines (Figs. 2B and 4B). Many suppressors
have been demonstrated to block systemic silencing in at
least one of these assays (Table 1). Because of the conflicting
results sometimes obtained with different assays, the most
successful attempts to determine whether a suppressor pri-
marily affects systemic silencing have used a multi-pronged
approach rather than relying on just one type of assay.
For example, although HC-Pro did not block systemic si-
lencing in grafting experiments using stable transgenic lines
(Mallory et al., 2001, 2003), it interfered with systemic si-
lencing in the transient expression assay (Hamilton et al.,
2002). Because HC-Pro is a powerful suppressor of local
silencing in all assays, these results suggest that HC-Pro pri-
marily affects local silencing, but also has a smaller effect
on systemic silencing. In contrast, CMV 2b primarily targets
systemic silencing because it blocks movement of the sig-
nal in many assays, but has a lesser effect on local silencing
(Brigneti et al., 1998; Bucher et al., 2003; Guo and Ding,
2002). It may not be surprising that a primary effect of a
particular suppressor on one aspect of silencing could lead,
perhaps through feedback mechanisms, to secondary effects
on other parts of the pathway, thereby making it appear that
the suppressor works at multiple points.
4.3. Conflicting results using different assays
Why do different assays give different results when look-
ing at effects of viral suppressors on systemic silencing?
The major conflicts are seen with stable expression assays
versus transient ones and probably reflect a number of in-
trinsic differences between the systems. Transient expres-
sion assays and the reversal of silencing assay use Agrobac-
terium to induce silencing. Because Agrobacterium is a plant
pathogen, infiltration likely induces plant defensive and bac-
terial counter-defensive interactions, which might modify
the activity of some viral suppressors. Thus, attempts to
compare suppressor activities or to understand the mecha-
nism of action of the different suppressors are complicated
 B.M. Roth et al. / Virus Research 102 (2004) 97–108
by the largely unknown effects of Agrobacterium on the
system. Similarly, the reversal of silencing assay involves
virus infection and likely induces accompanying defense
and counter-defensive responses that complicate the inter-
pretation of results. Further sources of variation include dif-
ferences in the level and time course of expression of the
silencing suppressor and silencing inducer in the different
systems.
Of the three widely used types of assay (Section 3), stable
expression assays are the ones most free of complications
due to extraneous pathogens. In addition, use of the same
well-characterized silenced transgenic line to test the effects
of different viral suppressors affords a degree of standard-
ization from lab to lab not yet found with transient expres-
sion assays. It is important to use well-characterized lines
to compare suppressor activities because different types of
transgenes trigger silencing in different ways (Vance and
Vaucheret, 2001).
Understanding the basis of the conflicting results given
by the currently widely used assays will likely offer con-
siderable insight into the mechanisms of suppressor action.
Alternative approaches such as yeast two-hybrid, biochemi-
cal, and localization studies (see Section 5) are increasingly
being used to investigate the mechanisms of action of the
different viral suppressors of silencing and will undoubtedly
help clear up the confusion.
5. Better studied suppressors of silencing
5.1. Cucumber mosaic virus (CMV) 2b
The CMV 2b protein was one of the first identified sup-
pressors of RNA silencing and also one of the best studied
from a mechanistic standpoint. The initial indication that
CMV 2b suppressed silencing came from the reversal of
silencing assay in which 2b expressed from PVX could
prevent the initiation of silencing but could not reverse si-
lencing that was already established (Brigneti et al., 1998).
That early result raised the possibility that 2b might block
systemic silencing. Subsequently, stable expression assays
and grafting experiments provided an elegant demonstration
that 2b blocks the movement of the systemic silencing signal
(Guo and Ding, 2002). The stable expression assays made
use of a well-characterized silenced tobacco line called
6b5 (Elmayan and Vaucheret, 1996), which is silenced for
the reporter gene GUS and is a model for sense-transgene
induced RNA silencing. The 6b5-GUS locus triggers si-
lencing even when it is hemizygous, produces high levels
of GUS siRNAs, and is highly effective in the production
and transmission of a systemic silencing signal as demon-
strated by grafting experiments. A genetic cross between
line 6b5 and a stable transgenic tobacco line expressing
CMV 2b established that 2b could partially suppress local
sense-transgene induced silencing: offspring of the cross
accumulated GUS mRNA, but GUS siRNA accumulation,
103
although reduced, was not eliminated. However, grafting
experiments definitively demonstrated that CMV 2b blocks
the movement of the systemic silencing signal. In the first
protocol, 6b5 rootstocks suppressed for silencing by CMV
2b did not silence grafted GUS-expressing scions, showing
that expression of CMV 2b in the rootstocks prevented
production or transmission of the systemic silencing signal
(Fig. 5A, first panel). In the second protocol, a small spacer
of transgenic tobacco expressing CMV 2b, grafted between
a GUS-silenced 6b5 rootstock and a GUS-expressing scion,
blocked systemic silencing (Fig. 5A, second panel). Thus,
CMV 2b prevents transmission of the systemic silencing
signal in a stable expression assay.
The experiments described above provide compelling ev-
idence that the mechanism of action of CMV 2b, at least in
part, is to block systemic silencing. Guo and Ding (2002)
also report similar conclusions based on Agrobacterium
co-infiltration experiments. Paradoxically, the same type of
co-infiltration experiments in another laboratory produced a
different result: CMV 2b delayed but did not block systemic
silencing (Hamilton et al., 2002). A possible cause of this
discrepancy is that the two labs were using different ratios
of silencing inducer to suppressor in the co-infiltrations
(S.W. Ding, personal communication), suggesting that a
standardized methodology for co-infiltration assays would
help resolve some of the current inconsistencies.
How does CMV 2b protein block the movement of the
mobile silencing signal? The ability of 2b protein to pre-
vent the transmission of the signal suggests that it either se-
questers or inactivates the signal in the phloem stream. One
possibility is that 2b acts directly by binding to the signal.
However, the finding that 2b localizes to the nucleus (Lucy
et al., 2000) suggests that the suppressor acts indirectly, per-
haps by activating one or more processes that subsequently
affect the signal.
5.2. Potyviral helper-component protease (HC-Pro)
HC-Pro was the first identified suppressor of RNA silenc-
ing. The original reports demonstrated that it suppresses both
transgene- and virus-induced silencing (Anandalakshmi
et al., 1998; Kasschau and Carrington, 1998). In contrast to
CMV 2b protein, it is able to reverse established silencing
in the reversal of silencing assay (Brigneti et al., 1998), sug-
gesting that the two suppressors work at different steps in the
silencing pathway. Although HC-Pro alone has suppressor
activity (Anandalakshmi et al., 1998; Brigneti et al., 1998),
it is frequently expressed as the proteinase 1 (P1)/HC-Pro
polyprotein in suppression of silencing studies. The
P1/HC-Pro construct is the N-terminal portion of the natural
viral polyprotein and allows HC-Pro to be proteolytically
processed just as when expressed from the viral genome.
Some evidence suggests that P1 might enhance HC-Pro
activity as a suppressor of silencing (Pruss et al., 1997).
A variety of approaches, including both transient and sta-
ble expression assays, have been used to investigate how
104 B.M. Roth et al. / Virus Research 102 (2004) 97–108
Fig. 5. Cartoon guide to (A) CMV 2b and (B) HC-Pro grafting experiments.
HC-Pro suppresses RNA silencing. Despite some conflicting
results, at least one common finding has emerged: HC-Pro
affects small RNA metabolism. Exactly how HC-Pro exerts
its effect on small RNAs and how this leads to suppression
of silencing are questions that remain to be resolved. The
interaction(s) of HC-Pro and its cellular effector(s) probably
occur in the cytoplasm because HC-Pro is found primarily
in cytoplasm (Mlotshwa et al., 2002a).
5.2.1. HC-Pro and small RNAs
HC-Pro has been reported to alter the accumulation of
several classes of small RNAs: the siRNAs that direct RNA
degradation during silencing, a novel class of slightly-larger
small RNAs of unknown function, and the endogenous mi-
croRNAs (miRNAs) that have been implicated in regulation
of development. In stable expression assays in tobacco,
HC-Pro has been reported to suppress three classes of
transgene-induced RNA silencing, in each case interfering
with the accumulation of siRNAs (Mallory et al., 2001,
2002b). Similarly, siRNA accumulation is dramatically
reduced during HC-Pro suppression of silencing in tran-
sient expression assays (Johansen and Carrington, 2001;
Llave et al., 2000). In stable expression assays in tobacco,
HC-Pro suppression of IR transgene-induced silencing
or amplicon-transgene-induced silencing—but not sense
transgene-induced silencing—resulted in the accumulation
of a novel class of slightly-larger small RNAs (Mallory
et al., 2002b). Similar slightly-larger small RNAs have also
been observed in transient expression assays (Hamilton
et al., 2002). The function of this class of small RNAs is
not clear; however, they do not appear to act as siRNAs
because their presence is not correlated with RNA degra-
dation (Hamilton et al., 2002; Mallory et al., 2002b). They
correlate with systemic silencing in transient expression
assays (Hamilton et al., 2002) but not in stable expression
assays (Mallory et al., 2002b). Finally and surprisingly,
the accumulation of endogenous miRNAs is increased in
both tobacco and Arabidopsis transgenic lines that express
HC-Pro (Kasschau et al., 2003; Mallory et al., 2002b),
suggesting a more general role in the biogenesis of small
regulatory RNAs. A recent report that HC-Pro enhances
the stability of several miRNA target messages raises the
possibility that HC-Pro affects the function as well as the
biogenesis of small RNAs (Kasschau et al., 2003).
5.2.2. HC-Pro and systemic silencing
In stable expression experiments utilizing grafting,
HC-Pro failed to block systemic silencing (Fig. 5B, first
panel; Mallory et al., 2001, 2003). Three different trans-
genic tobacco lines were used as rootstocks, representing
three different means of inducing silencing: sense trans-
gene, inverted repeat transgene, and amplicon transgene.
In the case of the amplicon-transgene induced silencing,
in which the transgene is the cDNA of a replication com-
petent virus, HC-Pro actually promoted systemic silencing
(Mallory et al., 2003). These results suggest that production
and transmission of the systemic silencing signal are largely
unaffected by HC-Pro, implying that HC-Pro suppression
of silencing occurs downstream of the signal. To test that
hypothesis, the effect of HC-Pro on perception of and/or
response to the silencing signal was tested in additional
grafting experiments (Fig. 5B, second panel; Mallory et al.,
2001). A silenced GUS line previously shown to silence
GUS systemically across a graft junction was used as root-
stock, and the scion was a line that expressed both GUS
and HC-Pro. The scion failed to become silenced for GUS,
 B.M. Roth et al. / Virus Research 102 (2004) 97–108
showing definitively that HC-Pro works downstream of the
systemic silencing signal: The rootstock is known to send
a signal; therefore, in the presence of HC-Pro, the scion
either fails to perceive the signal or fails to respond to it.
Although the grafting experiments described above make
a convincing case that HC-Pro does not block the systemic
silencing signal, results from Agrobacterium co-infiltration
assays suggest that this suppressor does interfere with sys-
temic silencing (Hamilton et al., 2002). There are a num-
ber of possible reasons for obtaining different results with
the two assays. First, the stable transgenic lines express the
P1/HC-Pro polyprotein, whereas the transient assay exper-
iments used a construct that encodes only HC-Pro and has
a methionine substituted for the natural N-terminal amino
acid of the protein. The protein produced in the transient as-
say, therefore, is not identical to HC-Pro produced from the
virus. Thus, it is possible that the difference between the two
assays reflects this small difference in the proteins. Indeed,
earlier experiments expressing either P1/HC-Pro or HC-Pro
(with an N-terminal methionine) from a PVX vector reported
a dramatic difference in the effect on PVX replication (Pruss
et al., 1997). A second possibility is that the result in the
transient assay depends on the ratio of inducer to suppres-
sor, as suggested for CMV 2b protein (see Section 5.1), and
this possibility could be tested by using a range of ratios.
A more exciting possibility is that the discrepancy reflects
an intrinsic difference between the two assays, as discussed
in Section 4.3. Thus, HC-Pro might block systemic silenc-
ing in some circumstances, but not in others, and identify-
ing the important variables will help in our understanding
of HC-Pro suppression of silencing.
5.2.3. HC-Pro interacting proteins
If HC-Pro works by interacting with components or regu-
lators of the silencing machinery, identification of plant pro-
teins that interact with HC-Pro should provide clues about
its mechanism of action. Using the yeast two-hybrid sys-
tem, an HC-Pro-interacting protein called regulator of gene
silencing calmodulin-like protein (rgsCaM) was identified
(Anandalakshmi et al., 2000). Like HC-Pro, rgsCaM could
reverse silencing when expressed from PVX in the rever-
sal of silencing assay. In addition, when over-expressed in
stable expression experiments, rgsCaM caused a develop-
mental phenotype typical of HC-Pro-expressing transgenic
lines and interfered with virus induced gene silencing. Be-
cause calmodulins are classic signaling molecules, these re-
sults raise the possibility that HC-Pro suppresses silencing
indirectly via a signal cascade involving rgsCaM.
5.3. Tombusvirus P19
This protein has been an exciting addition to the reper-
toire of plant viral suppressors. Initially described in the re-
versal of silencing assay (Voinnet et al., 1999), it appeared
to be a weak suppressor, only reversing silencing in the re-
gion of veins. Similarly, P19 delayed but did not prevent
105
virus-induced gene silencing and did not suppress silencing
induced by a defective interfering RNA that accumulates to
high levels (Qiu et al., 2002). In transient expression as-
says, however, P19 is a star, blocking both local and sys-
temic silencing and apparently eliminating all small RNAs
(Hamilton et al., 2002; Silhavy et al., 2002; Takeda et al.,
2002; Voinnet et al., 2003). Interestingly, biochemical stud-
ies have shown that P19 binds siRNAs and that binding de-
pends on characteristics of RNase III products (dsRNAs with
two nucleotide 3 overhangs) (Silhavy et al., 2002). This re-
sult raises the possibility that P19 suppresses silencing by
sequestering siRNAs, thereby preventing their incorporation
into the RISC complex to serve as guides. This is a novel
mechanism among suppressors, and because it theoretically
stems from an intrinsic property of the protein to bind func-
tional siRNAs, it is possible that P19 could interfere with
silencing in a broad range of different plants (and even other
organisms).
5.4. Potato virus X (PVX) p25
Three of five potexviruses tested in the reversal of silenc-
ing assay suppressed RNA silencing, although PVX was
one that did not show suppressor activity (Voinnet et al.,
1999). Paradoxically, however, PVX is the only potexvirus
for which a suppressor of RNA silencing has subsequently
been characterized. In grafting experiments designed to sep-
arate movement of the systemic silencing signal from that
of the virus in virus induced silencing (VIGS), Voinnet et al.
(2000) unexpectedly found that PVX infection failed to pro-
duce systemic silencing independent of the presence of virus,
suggesting that a PVX-encoded protein interferes with pro-
duction or transmission of the systemic signal. Protein p25,
which is required for cell-to-cell movement of potexviruses,
was identified as the culprit in a set of experiments in which
deletion derivatives of replication competent PVX-GFP viral
vector constructs were agro-infiltrated into plants express-
ing GFP. The PVX-GFP constructs were restricted to ini-
tially infiltrated cells because they all lacked coat protein,
which is also required for potexviral movement. Induction
of systemic silencing in these experiments, therefore, de-
pends entirely on the systemic signal. Only viral constructs
from which p25 expression had been eliminated induced
widespread systemic silencing in all plants.
Agro-coinfiltration experiments showed that p25, with-
out any other PVX protein, was sufficient to block sys-
temic silencing. Moreover, p25 blocked systemic silencing
whether silencing was induced by a simple transgene con-
struct (35S-GFP) or by a replication competent PVX-GFP
construct. The effect of p25 on local silencing in these ex-
periments, however, depended on the nature of the construct
used to induce silencing. Although p25 suppressed local si-
lencing in agro-coinfiltration assays with 35S-GFP, it did
not suppress local silencing induced by infiltration of repli-
cation competent PVX-GFP constructs. p25 suppression of
systemic silencing in agro-coinfiltration assays is correlated
106 B.M. Roth et al. / Virus Research 102 (2004) 97–108
with the absence of a slightly-larger class of small RNAs,
possibly indicating the step in the pathway affected by p25
(Hamilton et al., 2002).
Although these transient expression experiments make a
convincing case that PVX p25 blocks systemic silencing,
a different result was obtained in experiments using trans-
genic plants constitutively expressing white clover mosaic
potexvirus (WClMV) p25. Agro-infiltration of a 35S-IR con-
struct systemically silenced these plants (Foster et al., 2002).
Moreover, the systemic silencing reverted a severe develop-
mental phenotype, indicating that the systemic signal was
able to enter the shoot apical meristem. Thus, WClMV p25
did not prevent systemic silencing in this experimental sys-
tem. Whether the difference in the results obtained with the
PVX and WClMV p25 proteins reflects differences in the
assays or in the activities of the proteins has not yet been
determined.
6. Do all plant viruses have suppressors of silencing?
The discovery that plants have a generalized antiviral de-
fense mechanism triggered by dsRNA has revolutionized
our thinking about plant–virus interactions. In the euphoric
aftermath of the initial identification of plant viral suppres-
sors of silencing, a popular expectation was that most or all
plant viruses would encode a suppressor of RNA silencing.
Although many different viral suppressors have been iden-
tified, the fact that HC-Pro helps so many different viruses
suggests that a lot of viruses do not effectively suppress si-
lencing. Such viruses may have evolved other ways to try
to avoid silencing, such as by replicating within spherules
in the ER (Schwartz et al., 2002), where the dsRNA is hid-
den, or by replicating and moving rapidly enough to outrun
the mobile silencing signal. Furthermore, plants have other
defense mechanisms, and silencing might not be the major
threat for all viruses. Some viruses, therefore, may well have
suppressors of other defense pathways.
7. Suppressors of silencing as tools
The finding that certain viral proteins suppress RNA si-
lencing has provided a new tool for technologies utilizing
genetically modified plants and is, therefore, of practical sig-
nificance. Many biotechnological applications are impaired
by RNA silencing, and suppressors of silencing can be used
to attain consistent, high-level expression of transgenes in
plants (Mallory et al., 2002a; Voinnet et al., 2003). With si-
lencing under control, transgenic plants can be engineered to
produce a range of transgene expression: moderate levels of
expression to produce desired plant traits or very high-level
expression to use the plant as a factory producing pharma-
ceuticals, vaccines or other high-value gene products.
Perhaps more importantly, viral suppressors of silencing
also provide unique tools to understand the mechanism of
RNA silencing. Much of what is currently known about
the RNA silencing pathway comes from elegant in vitro
and genetic studies in organisms other than plants (for a
recent review see Tijsterman et al., 2002). In plants, tra-
ditional genetic approaches have led to the identification
of a number of cellular genes required for RNA silencing
(Dalmay et al., 2000, 2001; Fagard et al., 2000; Mourrain
et al., 2000). Surprisingly, however, all of these genes are
required for sense-, but not IR-transgene induced silencing
(Boutet et al., 2003). The plant viral suppressors, many of
which appear to work downstream of dsRNA, provide a
novel means of entry into parts of the silencing pathway
that are not easily accessible by genetic means. The cur-
rently known suppressors appear to work at different steps
in silencing, thereby providing access to a number of points
in the pathway where silencing can be controlled.
Identifying host proteins that interact with a viral sup-
pressor of RNA silencing is one very promising approach
that is being used to take advantage of viral suppressors
to elucidate the silencing pathway. The yeast two-hybrid
system has been used to find tobacco proteins that in-
teract with HC-Pro, identifying a calmodulin-related pro-
tein called rgsCaM that suppresses RNA silencing when
over-expressed (Anandalakshmi et al., 2000). This result
suggests that a calcium controlled signal transduction path-
way involving rgsCaM is one of the mechanisms regulating
RNA silencing. Intriguingly, in the case of geminiviruses,
yeast two-hybrid studies have identified SNF1 kinase as a
cellular interactor of the tomato golden mosaic virus AL2
protein (Hao et al., 2003). AL2 is a homologue of the
ACMV and TYLCV-C suppressors of silencing. Whether
the interaction of AL2 with SNF1, which is a regulator of
metabolism in response to stress, plays any role in suppres-
sion of silencing is unknown as yet.
The potential of using viral suppressors to help understand
the mechanism of RNA silencing in plants is largely un-
tapped, and these studies promise to be an exciting and fer-
tile area of research. The recent identification of a viral sup-
pressor that works in animal cells (Li et al., 2002) offers the
possibility that such proteins may provide a similar tool to
understand the silencing pathway in other organisms as well.
Note added in proof
The tomato mosaic virus replication protein has recently
been reported to suppress RNA silencing (Kubota, K., Tsuda,
S., Tamai, A., Meshi, T., 2003. Tomato mosaic virus repli-
cation protein suppresses virus-targeted posttranscriptional
gene silencing. J. Virol. 77, 11016–11026).
Acknowledgements
VBV gratefully acknowledges support from the USDA
Competitive Grants Program, NIH, and Dow AgroSciences
LLC.
 B.M. Roth et al. / Virus Research 102 (2004) 97–108
References
Anandalakshmi, R., Marathe, R., Ge, X., Herr Jr., J.M., Mau, C., Mallory,
A., Pruss, G., Bowman, L., Vance, V.B., 2000. A calmodulin-related
protein that suppresses posttranscriptional gene silencing in plants.
Science 290, 142–144.
Anandalakshmi, R., Pruss, G.J., Ge, X., Marathe, R., Mallory, A.C.,
Smith, T.H., Vance, V.B., 1998. A viral suppressor of gene silencing
in plants. Proc. Natl. Acad. Sci. USA 95, 13079–13084.
Baulcombe, D.C., 1996. RNA as a target and an initiator of
post-transcriptional gene silencing in transgenic plants. Plant Mol.
Biol. 32, 79–88.
Bernstein, E., Caudy, A.A., Hammond, S.M., Hannon, G.J., 2001. Role
for a bidentate ribonuclease in the initiation step of RNA interference.
Nature 409, 363–366.
Boutet, S., Vazquez, F., Liu, J., Beclin, C., Fagard, M., Gratias, A., Morel,
J.B., Crete, P., Chen, X., Vaucheret, H., 2003. Arabidopsis HEN1.
A genetic link between endogenous miRNA controlling development
and siRNA controlling transgene silencing and virus resistance. Curr.
Biol. 13, 843–848.
Brigneti, G., Voinnet, O., Li, W.X., Ji, L.H., Ding, S.W., Baulcombe, D.C.,
1998. Viral pathogenicity determinants are suppressors of transgene
silencing in Nicotiana benthamiana. EMBO J. 17, 6739–6746.
Bucher, E., Sijen, T., De Haan, P., Goldbach, R., Prins, M., 2003.
Negative-strand tospoviruses and tenuiviruses carry a gene for a sup-
pressor of gene silencing at analogous genomic positions. J. Virol.
77, 1329–1336.
Dalmay, T., Hamilton, A., Rudd, S., Angell, S., Baulcombe, D.C., 2000.
An RNA-dependent RNA polymerase gene in Arabidopsis is required
for posttranscriptional gene silencing mediated by a transgene but not
by a virus. Cell 101, 543–553.
Dalmay, T., Horsefield, R., Braunstein, T.H., Baulcombe, D.C., 2001.
SDE3 encodes an RNA helicase required for post-transcriptional gene
silencing in Arabidopsis. EMBO J. 20, 2069–2078.
Dong, X., van Wezel, R., Stanley, J., Hong, Y., 2003. Functional charac-
terization of the nuclear localization signal for a suppressor of post-
transcriptional gene silencing. J. Virol. 77, 7026–7033.
Dougherty, W.G., Parks, T.D., 1995. Transgenes and gene suppression:
telling us something new? Curr. Opin. Cell Biol. 7, 399–405.
Dunoyer, P., Pfeffer, S., Fritsch, C., Hemmer, O., Voinnet, O., Richards,
K.E., 2002. Identification, subcellular localization and some properties
of a cysteine-rich suppressor of gene silencing encoded by peanut
clump virus. Plant J. 29, 555–567.
Elbashir, S.M., Lendeckel, W., Tuschl, T., 2001a. RNA interference is
mediated by 21- and 22-nucleotide RNAs. Genes Dev. 15, 188–200.
Elbashir, S.M., Martinez, J., Patkaniowska, A., Lendeckel, W., Tuschl, T.,
2001b. Functional anatomy of siRNAs for mediating efficient RNAi
in Drosophila melanogaster embryo lysate. EMBO J. 20, 6877–6888.
Elmayan, T., Vaucheret, H., 1996. Expression of single copies of a strongly
expressed 35S transgene can be silenced post-transcriptionally. Plant
J. 9, 787–797.
English, J.J., Mueller, E., Baulcombe, D.C., 1996. Suppression of virus
accumulation in transgenic plants exhibiting silencing of nuclear genes.
Plant Cell 8, 179–188.
Fagard, M., Boutet, S., Morel, J.B., Bellini, C., Vaucheret, H., 2000.
AGO1, QDE-2, and RDE-1 are related proteins required for
post-transcriptional gene silencing in plants, quelling in fungi, and
RNA interference in animals. Proc. Natl. Acad. Sci. USA 97, 11650–
11654.
Foster, T.M., Lough, T.J., Emerson, S.J., Lee, R.H., Bowman, J.L., Forster,
R.L., Lucas, W.J., 2002. A surveillance system regulates selective
entry of RNA into the shoot apex. Plant Cell 14, 1497–1508.
Goodwin, J., Chapman, K., Swaney, S., Parks, T.D., Wernsman, E.A.,
Dougherty, W.G., 1996. Genetic and biochemical dissection of trans-
genic RNA-mediated virus resistance. Plant Cell 8, 95–105.
107
Guo, H.S., Ding, S.W., 2002. A viral protein inhibits the long range
signaling activity of the gene silencing signal. EMBO J. 21, 398–407.
Hamilton, A., Voinnet, O., Chappell, L., Baulcombe, D., 2002. Two classes
of short interfering RNA in RNA silencing. EMBO J. 21, 4671–4679.
Hamilton, A.J., Baulcombe, D.C., 1999. A species of small antisense RNA
in posttranscriptional gene silencing in plants. Science 286, 950–952.
Hammond, S.M., Bernstein, E., Beach, D., Hannon, G.J., 2000. An
RNA-directed nuclease mediates post-transcriptional gene silencing in
Drosophila cells. Nature 404, 293–296.
Hammond, S.M., Caudy, A.A., Hannon, G.J., 2001. Post-transcriptional
gene silencing by double-stranded RNA. Nat. Rev. Genet. 2, 110–119.
Hao, L., Wang, H., Sunter, G., Bisaro, D.M., 2003. Geminivirus AL2 and
L2 proteins interact with and inactivate SNF1 kinase. Plant Cell 15,
1034–1048.
Johansen, L.K., Carrington, J.C., 2001. Silencing on the spot. Induction
and suppression of RNA silencing in the Agrobacterium-mediated
transient expression system. Plant Physiol. 126, 930–938.
Kasschau, K.D., Carrington, J.C., 1998. A counterdefensive strategy of
plant viruses: suppression of posttranscriptional gene silencing. Cell
95, 461–470.
Kasschau, K.D., Xie, Z., Allen, E., Llave, C., Chapman, E.J., Krizan,
K.A., Carrington, J.C., 2003. P1/HC-Pro, a viral suppressor of RNA
silencing, interferes with Arabidopsis development and miRNA unc-
tion. Dev. Cell 4, 205–217.
Li, H., Li, W.X., Ding, S.W., 2002. Induction and suppression of RNA
silencing by an animal virus. Science 296, 1319–1321.
Lindbo, J.A., Silva-Rosales, L., Proebsting, W.M., Dougherty, W.G., 1993.
Induction of a highly specific antiviral state in transgenic plants:
implications for regulation of gene expression and virus resistance.
Plant Cell 5, 1749–1759.
Llave, C., Kasschau, K.D., Carrington, J.C., 2000. Virus-encoded suppres-
sor of posttranscriptional gene silencing targets a maintenance step in
the silencing pathway. Proc. Natl. Acad. Sci. USA 97, 13401–13406.
Lucy, A.P., Guo, H.S., Li, W.X., Ding, S.W., 2000. Suppression of
post-transcriptional gene silencing by a plant viral protein localized
in the nucleus. EMBO J. 19, 1672–1680.
Mallory, A.C., Ely, L., Smith, T.H., Marathe, R., Anandalakshmi, R.,
Fagard, M., Vaucheret, H., Pruss, G., Bowman, L., Vance, V.B., 2001.
HC-Pro suppression of transgene silencing eliminates the small RNAs
but not transgene methylation or the mobile signal. Plant Cell 13,
571–583.
Mallory, A.C., Mlotshwa, S., Bowman, L.H., Vance, V.B., 2003. The
capacity of transgenic tobacco to send a systemic RNA silencing
signal depends on the nature of the inducing transgene locus. Plant J.
35, 82–92.
Mallory, A.C., Parks, G., Endres, M.W., Baulcombe, D., Bowman,
L.H., Pruss, G.J., Vance, V.B., 2002a. The amplicon-plus system for
high-level expression of transgenes in plants. Nat. Biotechnol. 20,
622–625.
Mallory, A.C., Reinhart, B.J., Bartel, D., Vance, V.B., Bowman, L.H.,
2002b. From the cover: a viral suppressor of RNA silencing dif-
ferentially regulates the accumulation of short interfering RNAs and
micro-RNAs in tobacco. Proc. Natl. Acad. Sci. USA 99, 15228–
15233.
Mlotshwa, S., Verver, J., Sithole-Niang, I., Prins, M., Van Kammen,
A.B., Wellink, J., 2002a. Transgenic plants expressing HC-Pro show
enhanced virus sensitivity while silencing of the transgene results in
resistance. Virus Genes 25, 45–57.
Mlotshwa, S., Voinnet, O., Mette, M.F., Matzke, M., Vaucheret, H., Ding,
S.W., Pruss, G., Vance, V.B., 2002b. RNA silencing and the mobile
silencing signal. Plant Cell 14, S289–301.
Mourrain, P., Beclin, C., Elmayan, T., Feuerbach, F., Godon, C., Morel,
J.B., Jouette, D., Lacombe, A.M., Nikic, S., Picault, N., Remoue, K.,
Sanial, M., Vo, T.A., Vaucheret, H., 2000. Arabidopsis SGS2 and
SGS3 genes are required for posttranscriptional gene silencing and
natural virus resistance. Cell 101, 533–542.
108 B.M. Roth et al. / Virus Research 102 (2004) 97–108
Napoli, C., Lemieux, C., Jorgensen, R., 1990. Introduction of a chimeric
chalcone sythase gene into Petunia results in reversible co-suppression
of homologous gene in trans. Plant Cell 2, 279–289.
Nykanen, A., Haley, B., Zamore, P.D., 2001. ATP requirements and small
interfering RNA structure in the RNA interference pathway. Cell 107,
309–321.
Palauqui, J.C., Elmayan, T., Pollien, J.M., Vaucheret, H., 1997. Systemic
acquired silencing: transgene-specific post-transcriptional silencing is
transmitted by grafting from silenced stocks to non-silenced scions.
EMBO J. 16, 4738–4745.
Pfeffer, S., Dunoyer, P., Heim, F., Richards, K.E., Jonard, G., Ziegler-Graff,
V., 2002. P0 of beet Western yellows virus is a suppressor of post-
transcriptional gene silencing. J. Virol. 76, 6815–6824.
Pruss, G., Ge, X., Shi, X.M., Carrington, J.C., Bowman Vance, V., 1997.
Plant viral synergism: the potyviral genome encodes a broad-range
pathogenicity enhancer that transactivates replication of heterologous
viruses. Plant Cell 9, 859–868.
Qiu, W., Park, J.W., Scholthof, H.B., 2002. Tombusvirus P19-mediated
suppression of virus-induced gene silencing is controlled by genetic
and dosage features that influence pathogenicity. Mol. Plant Microbe
Interact. 15, 269–280.
Qu, F., Morris, T.J., 2002. Efficient infection of Nicotiana benthamiana
by Tomato bushy stunt virus is facilitated by the coat protein and
maintained by p19 through suppression of gene silencing. Mol. Plant
Microbe Interact. 15, 193–202.
Qu, F., Ren, T., Morris, T.J., 2003. The coat protein of turnip crinkle virus
suppresses posttranscriptional gene silencing at an early initiation step.
J. Virol. 77, 511–522.
Reed, J.C., Kasschau, K.D., Prokhnevsky, A.I., Gopinath, K., Pogue, G.P.,
Carrington, J.C., Dolja, V.V., 2003. Suppressor of RNA silencing
encoded by Beet yellows virus. Virology 306, 203–209.
Ruiz, M.T., Voinnet, O., Baulcombe, D.C., 1998. Initiation and main-
tenance of virus-induced gene silencing. Plant Cell 10, 937–
946.
Schwartz, M., Chen, J., Janda, M., Sullivan, M., den Boon, J., Ahlquist,
P., 2002. A positive-strand RNA virus replication complex paral-
lels form and function of retrovirus capsids. Mol. Cell 9, 505–
514.
Shi, X.M., Miller, H., Verchot, J., Carrington, J.C., Vance, V.B., 1997.
Mutations in the region encoding the central domain of helper
component-proteinase (HC-Pro) eliminate potato virus X/potyviral
synergism. Virology 231, 35–42.
Silhavy, D., Molnar, A., Lucioli, A., Szittya, G., Hornyik, C., Tavazza,
M., Burgyan, J., 2002. A viral protein suppresses RNA silencing
and binds silencing-generated, 21- to 25-nucleotide double-stranded
RNAs. EMBO J. 21, 3070–3080.
Smith, C.J., Watson, C.F., Bird, C.R., Ray, J., Schuch, W., Grierson,
D., 1990. Expression of a truncated tomato polygalacturonase gene
inhibits expression of the endogenous gene in transgenic plants. Mol.
Gen. Genet. 224, 477–481.
Smith, H.A., Swaney, S.L., Parks, T.D., Wernsman, E.A., Dougherty, W.G.,
1994. Transgenic plant virus resistance mediated by untranslatable
sense RNAs: expression, regulation, and fate of nonessential RNAs.
Plant Cell 6, 1441–1453.
Tabara, H., Yigit, E., Siomi, H., Mello, C.C., 2002. The dsRNA bind-
ing protein RDE-4 interacts with RDE-1, DCR-1, and a DExH-box
helicase to direct RNAi in C. elegans. Cell 109, 861–871.
Takeda, A., Sugiyama, K., Nagano, H., Mori, M., Kaido, M., Mise, K.,
Tsuda, S., Okuno, T., 2002. Identification of a novel RNA silencing
suppressor, NSs protein of Tomato spotted wilt virus. FEBS Lett. 532,
75–79.
Thomas, C.L., Leh, V., Lederer, C., Maule, A.J., 2003. Turnip crinkle
virus coat protein mediates suppression of RNA silencing in Nicotiana
benthamiana. Virology 306, 33–41.
Tijsterman, M., Ketting, R.F., Plasterk, R.H., 2002. The genetics of RNA
silencing. Annu. Rev. Genet. 36, 489–519.
van der Krol, A.R., Mur, L.A., Beld, M., Mol, J.N., Stuitje, A.R., 1990.
Flavonoid genes in petunia: addition of a limited number of gene
copies may lead to a suppression of gene expression. Plant Cell 2,
291–299.
van Wezel, R., Dong, X., Liu, H., Tien, P., Stanley, J., Hong, Y.,
2002. Mutation of three cysteine residues in Tomato yellow leaf curl
virus—China C2 protein causes dysfunction in pathogenesis and post-
transcriptional gene silencing suppression. Mol. Plant Microbe Inter-
act. 15, 203–208.
Vance, V., Vaucheret, H., 2001. RNA silencing in plants—defense and
counterdefense. Science 292, 2277–2280.
Vance, V.B., Berger, P.H., Carrington, J.C., Hunt, A.G., Shi, X.M., 1995.
5 proximal potyviral sequences mediate potato virus X/potyviral syn-
ergistic disease in transgenic tobacco. Virology 206, 583–590.
Voinnet, O., Baulcombe, D.C., 1997. Systemic signalling in gene silencing.
Nature 389, 553.
Voinnet, O., Lederer, C., Baulcombe, D.C., 2000. A viral movement
protein prevents spread of the gene silencing signal in Nicotiana
benthamiana. Cell 103, 157–167.
Voinnet, O., Pinto, Y.M., Baulcombe, D.C., 1999. Suppression of gene
silencing: a general strategy used by diverse DNA and RNA viruses
of plants. Proc. Natl. Acad. Sci. USA 96, 14147–14152.
Voinnet, O., Rivas, S., Mestre, P., Baulcombe, D., 2003. An enhanced
transient expression system in plants based on suppression of gene
silencing by the p19 protein of tomato bushy stunt virus. Plant J. 33,
949–956.
Yelina, N.E., Savenkov, E.I., Solovyev, A.G., Morozov, S.Y., Valkonen,
J.P., 2002. Long-distance movement, virulence, and RNA silencing
suppression controlled by a single protein in hordei- and potyviruses:
complementary functions between virus families. J. Virol. 76, 12981–
12991.
Zamore, P.D., Tuschl, T., Sharp, P.A., Bartel, D.P., 2000. RNAi:
double-stranded RNA directs the ATP-dependent cleavage of mRNA
at 21 to 23 nucleotide intervals. Cell 101, 25–33.
 Virus Research 102 (2004) 109–115

RNA silencing: a conserved antiviral immunity of plants and animals

Shou-Wei Ding a,b,∗ , Hongwei Li a , Rui Lu a , Feng Li b , Wan-Xiang Li a
a Center for Plant Cell Biology, Department of Plant Pathology, Riverside, CA 92521, USA
b Microbiology Program, University of California, Riverside, CA 92521, USA

Abstract

RNA silencing is a novel RNA-guided gene regulatory mechanism operational in a wide range of eukaryotic organisms from fission yeast,
plants, to mammals. This article reviews the recent progress on aspects of RNA silencing that are related to its biological function as a conserved
antiviral immunity of plants and animals, and highlights features of this novel antiviral response in invertebrate animals as compared to the
known innate and adaptive immunities. Finally, we discuss evidence that suggests a natural antiviral role for RNA silencing in vertebrates as
well as experimental approaches that may facilitate the identification of first mammalian viral suppressors of RNA silencing.
© 2004 Elsevier B.V. All rights reserved.

Keywords: RNA silencing; RNAi; Antiviral immunity; Drosophila

1. Introduction Martinez et al., 2002), siRNAs act as guides to select mRNA

RNA silencing is a generic term describing related RNA-
guided gene regulatory mechanisms that include post-
transcriptional gene silencing (PTGS) in plants, quelling
in fungi, and RNA interference (RNAi) in animals
(Baulcombe, 1999; Ding, 2000; Vance and Vaucheret,
2001). The first examples of RNA silencing were reported
in transgenic plants carrying artificially introduced trans-
genes (Matzke et al., 1989; Napoli et al., 1990; Van der
Krol et al., 1990). Similar RNA silencing phenomena have
subsequently been described in a wide range of eukaryotic
organisms such as fungi, worms, flies, and mammals (Denli
and Hannon, 2003; Fire et al., 1998).
Small interfering RNAs (siRNAs) are considered as the
central component of RNA silencing in all of the silencing
systems characterized to date (Denli and Hannon, 2003;
Hamilton and Baulcombe, 1999). siRNAs are processed
from double-stranded RNA (dsRNA) precursors by the
type III endoribonuclease Dicer (Hammond et al., 2000;
Zamore et al., 2000), which also recognizes fold-back,
imperfectly base-paired single-stranded RNA (ssRNA) sub-
strates (Hutvagner et al., 2001). After incorporation into the
RNA-induced silencing complex (RISC), which contains at
least one Argonaute (AGO) protein (Hammond et al., 2001;
∗ Corresponding author.
E-mail address: shou-wei.ding@ucr.edu (S.-W. Ding).
targets for degradation by siRNA/mRNA base-pairing. A
fascinating feature of RNA silencing is its ability to spread
both cell-to-cell and long distantly to cause systemic RNA
silencing in whole organisms via a sequence-specific si-
lencing signal produced following the induction of RNA
silencing in single cells (Mlotshwa et al., 2002; Palauqui
et al., 1997; Voinnet and Baulcombe, 1997).
RNA silencing and related pathways are involved not only
in the response to viruses and transposable elements, but also
in gene regulation and heterochromatin formation (Denli and
Hannon, 2003). This short review covers recent progress on
RNA silencing as a natural antiviral immunity in plants, in
invertebrate and vertebrate animals.
2. RNA silencing is a natural antiviral response
in plants
Three main lines of evidence support a naturally antiviral
role for RNA silencing in plants. Firstly, virus infection trig-
gers RNA silencing in infected plants that specifically targets
the viral and homologous RNAs for degradation. This is best
illustrated by the detection of virus-specific siRNAs of both
sense and antisense polarities in wild type plants infected
with plus-strand RNA viruses (Hamilton and Baulcombe,
1999). Presumably, silencing of viral RNAs is initiated by
Dicer cleavage of the viral dsRNA replicative intermediates,
which predicts that the host has the potential to recognize

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.virusres.2004.01.021
110 S.-W. Ding et al. / Virus Research 102 (2004) 109–115
and target the entire length of a replicating viral RNA. How-
ever, the viral siRNA population may also contain siRNAs
derived from Dicer cleavage of either highly structured ss-
RNAs (Szittya et al., 2002) or dsRNAs generated by the ac-
tion of a cellular RNA-dependent RNA polymerase (RdRP)
(Sijen et al., 2001; Vaistij et al., 2002). The specific cleav-
age of ssRNA targets guided by viral siRNAs is consistent
with a number of key observations that provided the first
clues for the antiviral role of RNA silencing. For example,
the observed induction of RNA silencing of a tobacco etch
virus (TEV) transgene in tobacco plants by TEV infection
(Lindbo et al., 1993) could be initiated by antisense TEV
siRNAs produced in the infected plants in response to the
replicating TEV RNA. Similarly, this would also explain
why transgenes encoding a replication-competent viral RNA
genome are the more consistent inducers of RNA silenc-
ing than simple sense transgenes (Angell and Baulcombe,
1997) and why non-virus-derived transgenes and endoge-
nous genes can be specifically silenced following infection
with a recombinant RNA or DNA virus carrying a cDNA
fragment of the corresponding gene (Kjemtrup et al., 1998;
Kumagai et al., 1995; Ruiz et al., 1998). The latter tech-
nique, known as virus-induced gene silencing (VIGS), has
been developed into a high throughput platform technology
for functional genomics studies in several host species (Lu
et al., 2003).
The second and perhaps the strongest support for a
naturally antiviral role of RNA silencing in plants is the
demonstration that plant viruses encode proteins capable of
suppressing RNA silencing (Li and Ding, 2001; Roth et al.,
this issue). The first viral suppressors of RNA silencing
reported were HC-Pro and the 2b protein encoded by po-
tyviruses and cucumoviruses, respectively (Anandalakshmi
et al., 1998; Brigneti et al., 1998; Kasschau and Carrington,
1998; Li et al., 1999), both of which were previously identi-
fied as determinants of synergistic viral diseases (Ding et al.,
1996; Pruss et al., 1997). Many plant viruses have since
been found to encode silencing suppression activity, includ-
ing a DNA virus and potato virus X (PVX) that did not
exhibit any suppressor activity in a ‘reversal of silencing’
assay, even though suppressors encoded by viruses from
distinct taxa may share little sequence homology (Li et al.,
1999; Li and Ding, 2001; Voinnet et al., 1999, 2000). It
would not be unexpected if all plant viruses are found to
encode at least one suppressor of RNA silencing, provided
each of the viral proteins is assayed for possible suppression
of both intra- and inter-cellular RNA silencing. Silencing
suppression clearly plays a role in establishing virus infec-
tions as most of these suppressors were previously shown
essential for systemic virus spread in their hosts.
In contrast to the specificity involved in the selection
of silencing targets, viral suppression of RNA silencing is
broad-spectrum, suggesting that they target a core RNA si-
lencing pathway. It is important to note that silencing sup-
pression by all known viral proteins is partial or incomplete.
For example, the cucumber mosaic virus (CMV) 2b protein
prevents the spread of RNA silencing into tissues that do not
contain the initial silencing trigger, but it does not inhibit the
initiation of intracellular silencing (Guo and Ding, 2002).
Thus, while a portion of CMV RNAs is constantly destroyed
in infected cells, uninfected tissues will remain suscepti-
ble due to lack of import of the antiviral silencing signal.
In contrast, the potyviral HC-Pro is a potent suppressor of
intracellular silencing, but silencing signal is continuously
being produced and exported out of the HC-Pro-expressing
cells (Mallory et al., 2001) and as a result, may reduce the
rate of virus spread in infected plants. Thus, mixed viral in-
fections will induce synergistic diseases when suppressors
carried by unrelated viruses target distinct steps in the si-
lencing pathway or when a related virus provides a more
efficient suppression of RNA silencing at the same step. No-
tably, most of the plant viruses contain a compact genome
shorter than 10 kilobases encoding on the average four–five
proteins. Therefore, the fact that silencing suppression re-
mains as a conserved genomic function of plant viruses is
also indicative of RNA silencing as being a major antiviral
response of plants.
The third support came from the observation that host
plants compromised in RNA silencing exhibit enhanced
susceptibility to virus infection. In this respect, SGS2/SDE1,
SGS3, SDE3 and AGO1 have been identified in Arabidop-
sis thaliana and are essential for transgene-induced RNA
silencing, possibly by contributing to production of dsRNA
(Vance and Vaucheret, 2001). Intriguingly, A. thaliana mu-
tants defective in either of these genes were all hypersensi-
tive to CMV, but were as susceptible as the wild type plants
to the infection of at least five other viruses tested (Boutet
et al., 2003; Dalmay et al., 2000, 2001; Morel et al., 2002;
Mourrain et al., 2000). In contrast, tobacco and A. thaliana
plants deficient in a gene related to SGS2/SDE1, which,
unlike SGS2/SDE1, is induced by salicylic acid and virus
infection, were more susceptible to both tobacco mosaic
virus and PVX (Xie et al., 2001; Yu et al., 2003). It will
be informative to determine if infection with virus mutants
that lack a functional suppressor of RNA silencing could
be complemented specifically in any of those host mutants
compromised in RNA silencing.
3. Animal viral proteins that suppress RNA
silencing in plants
Detection of silencing suppressor activity for the flock
house virus (FHV) B2 protein, using a silencing assay es-
tablished in plants, provided the first indication that RNA
silencing might be antiviral in animals (Li et al., 2002). The
animal nodaviral B2 gene shares several features with the
plant cucumoviral 2b gene (Ding et al., 1995), known to en-
code a suppressor of RNA silencing (Brigneti et al., 1998;
Li et al., 1999). Both genes overlap the C-terminus, and
occupy the +1 reading frame, of the viral RdRP gene and
are translated in vivo from a subgenomic RNA, though their
 S.-W. Ding et al. / Virus Research 102 (2004) 109–115
Fig. 1. Open reading frames (ORFs) encoded by CMV RNA 2 and FHV RNA 1. The maps show the location of both start (arrow) and stop (triangle)
codons in each reading frame represented by, respectively, short and long vertical lines. The ORFs encoding the viral RdRP are called 2a in CMV and
A in FHV, respectively. The start and stop codons for proteins B2 and 2b are indicated. The location of the conserved GDD codons of the RdRP of
CMV (nt 1908–1916) and FHV (nt 2110–2118) is also shown.
encoded proteins do not exhibit any detectable sequence
similarities (Fig. 1). Knockout of either B2 in FHV or 2b in
CMV resulted in a defect in virus spreading, but neither is
required for viral RNA replication (Ball, 1995; Ding et al.,
1995). In Nicotiana benthamiana plants, B2 suppression of
transgene-induced RNA silencing was as potent as the 2b
protein encoded by tomato aspermy cucumovirus, which is
among the strongest plant viral suppressors known (Li et al.,
2002). By comparison, the 2b of CMV was a much weaker
suppressor that delayed, but did not prevent the initiation
of RNA silencing (Guo and Ding, 2002). Significantly, B2
was able to functionally substitute for 2b of CMV in whole
plant infections (Li et al., 2002).
A recent work found that the mammalian reovirus ␴3 pro-
tein also suppresses RNA silencing in the same Agrobac-
terium co-infiltration assay established in N. benthamiana
plants carrying a highly expressed green fluorescent protein
(GFP) transgene (Lichner et al., 2003). ␴3 is a virion outer
shell protein that binds dsRNA (Miller and Samuel, 1992),
suggesting a possible link between dsRNA binding and si-
lencing suppression (Lichner et al., 2003).
4. RNA silencing as an adaptive antiviral
immunity in animals
Direct experimental evidence supporting an antiviral role
of RNA silencing in the animal kingdom has recently been
documented in Drosophila (Li et al., 2002). In cultured
Drosophila S2 cells, infection with FHV led to the accu-
mulation of the 22-nt, FHV-specific siRNAs in both the
sense and antisense polarities. Virus-specific siRNAs were
also detected in the silkmoth (Bombyx mori) larvae infected
with Sindbis virus (Uhlirova et al., 2003). FHV RNAs ac-
cumulated to higher levels in the fly cells after the RNAi
111
pathway was made partially defective by depleting AGO2,
an essential component of RISC (Hammond et al., 2001).
These findings indicate that recognition and targeting of the
replicating FHV RNAs for degradation by the RNAi path-
way represents a natural response of the wild type fly cells
to virus infection. Further mutagenesis analysis showed an
essential role of B2 in FHV infection since a B2-deletion
mutant of FHV did not accumulate to a detectable level in
wild type fly cells. However, an abundant accumulation of
the B2-knockout mutant was detected in the fly cells that
either carried a B2-expressing plasmid or were depleted for
AGO2. Thus, the observed essential role of B2 is to sup-
press the RNA silencing antiviral response, rather than being
required for viral replication, which is in agreement with a
previous work carried out in mammalian cells (Ball, 1995).
Recent work showed that B2 is also a suppressor of RNAi
since specific degradation of a GFP mRNA targeted by an
introduced GFP dsRNA was inhibited in the B2-expressing
fly cells (Li et al., 2004).
Similar to fruit flies, it was found recently that infection of
mosquito cells with a HIV-related nodavirus also triggered
an AGO2-dependent antiviral immunity that was sensitive
to suppression of either the homologous or heterologous
nodaviral B2 (Li et al., 2004).
Therefore, nodavirus infection of invertebrate animal
cells triggers RNA silencing that specifically targets the
invading viral RNA, and nodaviruses encode a B2 protein
that is able to suppress PTGS in plants and RNAi in fruit
flies. Significantly, B2 is essential for nodavirus infection of
wild type animal cells, but is dispensable in cells depleted
for a RISC component, AGO2, indicating that nodavirus
RNAs are targeted for degradation by the RISC-mediated
RNA silencing antiviral response. These data establish RNA
silencing as a natural antiviral mechanism in invertebrate
animals.
112 S.-W. Ding et al. / Virus Research 102 (2004) 109–115
5. Features of the RNA silencing antiviral response in
invertebrate animal cells
The RNA silencing antiviral response of invertebrates
is adaptive, potent and rapid. In contrast to the broad-
spectrum innate immunity in Drosophila and mosquitoes
(Christophides et al., 2002), the RNA silencing immunity
has features similar to the peptide-based adaptive immunity
(Whitton and Oldstone, 2001). The specificity determinants
of the RNA silencing response are siRNAs, which are
derived and processed from the invading virus. After up-
loaded into RISC, these virus-specific siRNAs selectively
recruit target ssRNAs by base-pairing for RISC-mediated
cleavage. At whole organism level, RNA silencing may
also have the capacity to provide a long-term memory,
as has been detected in plants, which, after being re-
covered from a virulent primary infection, maintain an
RNA silencing-mediated resistance to secondary infections
(Covey et al., 1997; Ratcliff et al., 1997, 1999; Xin and
Ding, 2003). However, while the peptide-based immunity
in vertebrates usually takes more than a week to respond
(Whitton and Oldstone, 2001), the RNA silencing response
in fruit flies and mosquitoes is rapid and effective in primar-
ily infected cells, which is similar to the innate immunity.
In fact, the genome of the B2-knockout mutant of FHV
never accumulated to detectable levels after transfection,
indicating that in the absence of viral suppression, RNA
silencing is sufficient to rapidly clear viral RNAs within
primarily infected cells (Li et al., 2002). In this regard, it is
possible that RNA silencing may contribute to, or be part
of, the robust innate immunity occurred in early stages of
viral infections in vertebrates (Parham, 2003).
It is known that common components, such as siRNA,
Dicer and RISC, are involved in PTGS in plants and RNAi
in animals (Tang et al., 2003; Vance and Vaucheret, 2001).
Suppression of the two processes in plants and Drosophila
by both the animal viral B2 protein and the plant viral p19
provides direct experimental evidence for a conserved RNA
silencing pathway in the plant and animal kingdoms (Voinnet
et al., 1999; Li et al., 2002; Lindenbach and Rice, 2002; Li
et al., 2004). However, the cellular RdRP is found only in
plants, fungi and worms, but not in fruit flies, mouse and
human. Furthermore, it appears that at the single cell level,
the RNA silencing response targeting a replicating virus in
plants is not as potent as in fruit flies and mosquitoes. For
example, plant virus mutants that lack a functional suppres-
sor have defects in spreading cell-to-cell or systemically in
whole plant infections, but they exhibit no obvious defect
in isolated single cells (protoplasts) (Li and Ding, 2001;
Silhavy et al., 2002). This is in contrast to an essential role
for B2 suppression in FHV infection of cultured fly cells (Li
et al., 2002), suggesting that RISC as a multiple turnover
enzymatic complex may be more efficient in organisms that
do not carry a cellular RdRP. These results also suggest that
plant viral suppressors may mainly interfere with the spread
of the RNA silencing antiviral response away from the pri-
marily infected cells, whereas it may be essential for animal
viral suppressors to inhibit the intracellular silencing.
6. Is RNA silencing naturally antiviral in vertebrates?
The RNA silencing pathway is also operational in mam-
malian cells since specific mRNA degradation is achieved
by introducing exogenous synthetic siRNAs (Caplen et al.,
2001; Elbashir et al., 2001). Remarkably, the siRNAs intro-
duced into mammalian cells are found in a silencing com-
plex essentially identical to the RISC initially characterized
in Drosophila (Hammond et al., 2001; Martinez et al., 2002).
As in fruit flies, the cellular RdRP involved in transitive RNA
silencing in A. thaliana and C. elegans is not found in either
mouse or human (Denli and Hannon, 2003; Schwarz et al.,
2002). Furthermore, the human genome encodes a Dicer that
recognizes both dsRNA and hairpin ssRNA substrates as the
Drosophila Dicer does (Bernstein et al., 2001; Hutvagner
et al., 2001; Provost et al., 2002; Zhang et al., 2002).
Thus, it is conceivable that the mammalian Dicer could
also detect viral dsRNA replicative intermediates and/or
stem-loop structures to initiate RNA silencing in response
to virus infection of mammalian cells, as has been demon-
strated in fruit flies (Li et al., 2002). Recent reports have
elegantly demonstrated potent inhibition of production of
several types of mammalian viruses by synthetic siRNAs
(Gitlin and Andino, 2003), indicating that viral RNAs are
accessible to an activated RISC in mammalian cells. The
mammalian Dicer could act in an antagonist pathway to the
dsRNA-specific protein kinase (PKR). Alternatively, initi-
ation of the RNAi antiviral immunity by the mammalian
Dicer could occur in specific cell types or at certain de-
velopmental stages in which the non-specific translational
shutdown response mediated by PKR is inactive (Wianny
and Zernicka-Goetz, 2000; Svoboda et al., 2000; Billy et al.,
2001).
There is evidence suggesting that the RNAi machinery
in mammalian cells is capable of detecting the replicating
viral RNAs. For example, FHV also replicates and accu-
mulates in mammalian cells. Although B2 was not essen-
tial for FHV replication, B2-deficient FHV production in
mammalian cells impaired infectivity of viral progeny (Ball,
1995), suggesting that the newly identified activity of B2 in
silencing suppression may play a role in FHV infection of
mammalian cells. Similarly, in reovirus-infected mammalian
cells, the expression of ␴3 might lead to inactivation of the
RNA silencing antiviral response in addition to inhibition of
PKR-mediated responses (Lichner et al., 2003; Miller and
Samuel, 1992).
Future work will be necessary to determine if virus
infection indeed elicits the RNA silencing antiviral re-
sponse in mammalian cells. This may include detection of
virus-specific siRNAs, a homology-dependent RNA degra-
dation, or viral suppression of RNA silencing in infected
mammalian cells. The in planta silencing assay will be
 S.-W. Ding et al. / Virus Research 102 (2004) 109–115
useful in identifying novel animal viral suppressors of
RNA silencing. However, it will be limited to those sup-
pressors that target the elements conserved between the
RdRP-dependent silencing pathway in plants and the
RdRP-independent silencing pathway in animals. As a re-
sult, a suppressor of RNA silencing identified in plant assays
may not necessarily suppress RNA silencing in animal cells
and vice versa. Therefore, it will be important to establish
an RNA silencing system in mammalian cells suitable for
assaying viral suppression of RNA silencing. It is of interest
to consider that most, if not all, of the characterized plant
viral suppressors target steps upstream of siRNAs or RISC
(Li and Ding, 2001; Roth et al., this issue). Identification of
similar animal viral suppressors will likely require a silenc-
ing system that initiates early in the silencing pathway and
is not initiated by siRNAs. Thus, clearance of virus RNAs
in infected cells by introduced siRNAs (Dector et al., 2002;
Gitlin et al., 2002; Randall et al., 2003) may not necessarily
argue against the possibility that these mammalian viruses
carry a suppressor of the RNA silencing antiviral immunity.
Given the closer similarity of the RNA silencing pathway
between Drosophila and humans, the nodavirus RNA si-
lencing system established in cultured fruit fly cells (Li
et al., 2002) may provide a valuable assay for the identifica-
tion of mammalian viral suppressors. Indeed, NS1 encoded
by influenza A/B/C viruses and E3L by vaccinia virus were
recently identified as suppressors of the nodavirus-induced
RNA silencing in Drosophila cells, implicating RNA si-
lencing as an RNA-based antiviral immunity in mammalian
cells (Li et al., 2004).
Acknowledgements
The research projects on the RNA silencing antiviral re-
sponse in authors’ lab are supported by grants from NIH,
USDA, CRB, CDFA and UC-BIOSTAR. The authors also
wish to thank Karla Kirkegaard for discussion.
References
Anandalakshmi, R., Pruss, G.J., Ge, X., Marathe, R., Mallory, A.C.,
Smith, T.H., Vance, V.B., 1998. A viral suppressor of gene silencing
in plants. Proc. Natl. Acad. Sci. U.S.A. 95, 13079–13084.
Angell, S.M., Baulcombe, D.C., 1997. Consistent gene silencing in trans-
genic plants expressing a replicating potato virus X RNA. EMBO J.
16, 3675–3684.
Ball, L.A., 1995. Requirements for the self-directed replication of flock
house virus RNA 1. J. Virol. 69, 2722.
Baulcombe, D.C., 1999. Gene silencing: RNA makes RNA makes no
protein. Curr. Biol. 9, R599–R601.
Bernstein, E., Caudy, A.A., Hammond, S.M., Hannon, G.J., 2001. Role
for a bidentate ribonuclease in the initiation step of RNA interference.
Nature 409, 363–366.
Billy, E., Brondani, V., Zhang, H., Muller, U., Filipowicz, W.,
2001. Specific interference with gene expression induced by long,
double-stranded RNA in mouse embryonal teratocarcinoma cell lines.
Proc. Natl. Acad. Sci. U.S.A. 98, 14428–14433.
113
Boutet, S., Vazquez, F., Liu, J., Beclin, C., Fagard, M., Gratias, A., Morel,
J.B., Crete, P., Chen, X., Vaucheret, H., 2003. Arabidopsis HEN1:
a genetic link between endogenous miRNA controlling development
and siRNA controlling transgene silencing and virus resistance. Curr.
Biol. 13, 843–848.
Brigneti, G., Voinnet, O., Li, W.X., Ji, L.H., Ding, S.W., Baulcombe, D.C.,
1998. Viral pathogenicity determinants are suppressors of transgene
silencing in Nicotiana benthamiana. EMBO J. 17, 6739–6746.
Caplen, N.J., Parrish, S., Imani, F., Fire, A., Morgan, R.A., 2001. Spe-
cific inhibition of gene expression by small double-stranded RNAs in
invertebrate and vertebrate systems. Proc. Natl. Acad. Sci. U.S.A. 98,
9742–9747.
Christophides, G.K., Zdobnov, E., Barillas-Mury, C., Birney, E., Blandin,
S., Blass, C., Brey, P.T., Collins, F.H., Danielli, A., Dimopoulos, G.,
Hetru, C., Hoa, N.T., Hoffmann, J.A., Kanzok, S.M., Letunic, I.,
Levashina, E.A., Loukeris, T.G., Lycett, G., Meister, S., Michel, K.,
Moita, L.F., Muller, H.M., Osta, M.A., Paskewitz, S.M., Reichhart,
J.M., Rzhetsky, A., Troxler, L., Vernick, K.D., Vlachou, D., Volz,
J., von Mering, C., Xu, J., Zheng, L., Bork, P., Kafatos, F.C., 2002.
Immunity-related genes and gene families in Anopheles gambiae.
Science 298, 159–165.
Covey, S.N.A.-K., N.S., Langara, A., Turner, D.S., 1997. Plants combat
infection by gene silencing. Nature 385, 781–782.
Dalmay, T., Hamilton, A., Rudd, S., Angell, S., Baulcombe, D.C., 2000.
An RNA-dependent RNA polymerase gene in Arabidopsis is required
for posttranscriptional gene silencing mediated by a transgene but not
by a virus. Cell 101, 543–553.
Dalmay, T., Horsefield, R., Braunstein, T.H., Baulcombe, D.C., 2001.
SDE3 encodes an RNA helicase required for post-transcriptional gene
silencing in Arabidopsis. EMBO J. 20, 2069–2078.
Dector, M.A., Romero, P., Lopez, S., Arias, C.F., 2002. Rotavirus gene
silencing by small interfering RNAs. EMBO Rep. 3, 1175–1180.
Denli, A.M., Hannon, G.J., 2003. RNAi: an ever-growing puzzle. Trends
Biochem. Sci. 28, 196–201.
Ding, S.W., 2000. RNA silencing. Curr. Opin. Biotechnol. 11, 152–156.
Ding, S.W., Li, W.X., Symons, R.H., 1995. A novel naturally-occurring
hybrid gene encoded by a plant RNA virus facilitates long-distance
virus movement. EMBO J. 14, 5762–5772.
Ding, S.W., Shi, B.J., Li, W.X., Symons, R.H., 1996. An interspecies
hybrid RNA virus is significantly more virulent than either parental
virus. Proc. Natl. Acad. Sci. U.S.A. 93, 7470–7474.
Elbashir, S.M., Lendeckel, W., Tuschl, T., 2001. RNA interference is
mediated by 21-and 22-nucleotide RNAs. Genes. Dev. 15, 188–
200.
Fire, A., Xu, S.Q., Montgomery, M.K., Kostas, S.A., Driver, S.E., Mello,
C.C., 1998. Potent and specific genetic interference by double-stranded
RNA in Caenorhabditis elegans. Nature 391, 806–811.
Gitlin, L., Andino, R., 2003. Nucleic acid-based immune system: the
antiviral potential of mammalian RNA silencing. J. Virol. 77, 7159–
7165.
Gitlin, L., Karelsky, S., Andino, R., 2002. Short interfering RNA confers
intracellular antiviral immunity in human cells. Nature 418, 430–
434.
Guo, H.S., Ding, S.W., 2002. A viral protein inhibits the long range
signaling activity of the gene silencing signal. EMBO J. 21, 398–407.
Hamilton, A.J., Baulcombe, D.C., 1999. A species of small antisense RNA
in posttranscriptional gene silencing in plants. Science 286, 950–952.
Hammond, S.M., Bernstein, E., Beach, D., Hannon, G.J., 2000. An
RNA-directed nuclease mediates post-transcriptional gene silencing in
Drosophila cells. Nature 404, 293–296.
Hammond, S.M., Boettcher, S., Caudy, A.A., Kobayashi, R., Hannon, G.J.,
2001. Argonaute 2, a link between genetic and biochemical analyses
of RNAi. Science 293, 1146–1150.
Hutvagner, G., McLachlan, J., Pasquinelli, A.E., Balint, E., Tuschl, T.,
Zamore, P.D., 2001. A cellular function for the RNA-interference
enzyme Dicer in the maturation of the let-7 small temporal RNA.
Science 293, 834–838.
114 S.-W. Ding et al. / Virus Research 102 (2004) 109–115
Kasschau, K.D., Carrington, J.C., 1998. A counterdefensive strategy of
plant viruses: Suppression of posttranscriptional gene silencing. Cell
95, 461–470.
Kjemtrup, S., Sampson, K.S., Peele, C.G., Nguyen, L.V., Conkling, M.A.,
Thompsosn, W.F., Robertson, D., 1998. Gene silencing from plant
DNA carried by a geminivirus. Plant J. 14, 91–100.
Kumagai, M.H., Donson, J., della-Cioppa, G., Harvey, D., Hanley, K.,
Grill, L.K., 1995. Cytoplasmic inhibition of carotenoid biosynthesis
with virus-derived RNA. Proc. Natl. Acad. Sci. U.S.A. 92, 1679–1683.
Li, H.W., Li, W.X., Ding, S.W., 2002. Induction and suppression of RNA
silencing by an animal virus. Science 296, 1319–1321.
Li, H.W., Lucy, A.P., Guo, H.S., Li, W.X., Ji, L.H., Wong, S.M., Ding,
S.W., 1999. Strong host resistance targeted against a viral suppressor
of the plant gene silencing defence mechanism. EMBO J. 18, 2683–
2691.
Li, W.X., Ding, S.W., 2001. Viral suppressors of RNA silencing. Curr.
Opin. Biotechnol. 12, 150–154.
Li, W.X., Li, H.W., Lu, R., Li, F., Dus, M., Atkinson, P.W., Brydon, E.,
Johnson, K.L., Garcia-Sastre, A., Ball, L.A., Palese, P., Ding, S.W.,
2004. Interferon antagonist proteins of influenza and vaccinia viruses
are suppressors of RNA silencing. Proc. Natl. Acad. Sci. USA 101,
1350–1355.
Lichner, Z., Silhavy, D., Burgyan, J., 2003. Double-stranded RNA-binding
proteins could suppress RNA interference-mediated antiviral defences.
J. Gen. Virol. 84, 975–980.
Lindbo, J.A., Silva-Rosales, L., Proebsting, W.M., Dougherty, W.G., 1993.
Induction of a highly specific antiviral state in transgenic plants:
implications for regulation of gene expression and virus resistance.
Plant Cell 5, 1749–1759.
Lindenbach, B.D., Rice, C.M., 2002. RNAi targeting an animal virus:
news from the front. Mol. Cell 9, 925–927.
Lu, R., Martin-Hernandez, A.M., Peart, J.R., Malcuit, I., Baulcombe, D.C.,
2003. Virus-induced gene silencing in plants. Methods 30, 296–303.
Mallory, A.C., Ely, L., Smith, T.H., Marathe, R., Anandalakshmi, R.,
Fagard, M., Vaucheret, H., Pruss, G., Bowman, L., Vance, V.B., 2001.
HC-Pro suppression of transgene silencing eliminates the small RNAs
but not transgene methylation or the mobile signal. Plant Cell 13,
571–583.
Martinez, J., Patkaniowska, A., Urlaub, H., Luhrmann, R., Tuschl, T.,
2002. Single-stranded antisense siRNAs guide target RNA cleavage
in RNAi. Cell 110, 563–574.
Matzke, M.A., Primig, M., Trnovsky, J., Matzke, A.J.M., 1989. Reversible
methylation and inactivation of marker genes in sequentially trans-
formed tobacco plants. EMBO J. 8, 643–649.
Miller, J.E., Samuel, T., 1992. Proteolytic cleavage of the reovirus sigma
3 protein results in enhanced double-stranded RNA-binding activ-
ity: identification of a repeated basic amino acid motif within the
C-terminal binding region. J. Virol. 66, 5347–5356.
Mlotshwa, S., Voinnet, O., Mette, M.F., Matzke, M., Vaucheret, H., Ding,
S.W., Pruss, G., Vance, V.B., 2002. RNA silencing and the mobile
silencing signal. Plant Cell 14, S289–S301.
Morel, J.B., Godon, C., Mourrain, P., Beclin, C., Boutet, S., Feuerbach, F.,
Proux, F., Vaucheret, H., 2002. Fertile hypomorphic ARGONAUTE
(ago1) mutants impaired in post-transcriptional gene silencing and
virus resistance. Plant Cell 14, 629–639.
Mourrain, P., Beclin, C., Elmayan, T., Feuerbach, F., Godon, C., Morel,
J.B., Jouette, D., Lacombe, A.M., Nikic, S., Picault, N., Remoue,
K., Sanial, M., Vo, T.A., Vaucheret, H., 2000. Arabidopsis SGS2 and
SGS3 genes are required for posttranscriptional gene silencing and
natural virus resistance. Cell 101, 533–542.
Napoli, C., Lemieux, C., Jorgensen, R., 1990. Introduction of a
chimeric chalcone synthase gene into Petunia results in reversible
co-suppression of homologous genes in trans. Plant Cell 2, 279–289.
Palauqui, J.C., Elmayan, T., Pollien, J.M., Vaucheret, H., 1997. Systemic
acquired silencing: transgene-specific post-transcriptional silencing is
transmitted by grafting from silenced stocks to non-silenced scions.
EMBO J. 16, 4738–4745.
Parham, P., 2003. Innate immunity: the unsung heroes. Nature 423, 20.
Provost, P., Dishart, D., Doucet, J., Frendewey, D., Samuelsson, B., Rad-
mark, O., 2002. Ribonuclease activity and RNA binding of recombi-
nant human Dicer. EMBO J. 21, 5864–5874.
Pruss, G., Ge, X., Shi, X.M., Carrington, J.C., Vance, V.B., 1997. Plant vi-
ral synergism: the potyviral genome encodes a broad-range pathogenic-
ity enhancer that transactivates replication of heterologous viruses.
Plant Cell 9, 859–868.
Randall, G., Grakoui, A., Rice, C.M., 2003. Clearance of replicating
hepatitis C virus replicon RNAs in cell culture by small interfering
RNAs. Proc. Natl. Acad. Sci. U.S.A. 100, 235–240.
Ratcliff, F., Harrison, B.D., Baulcombe, D.C., 1997. A similarity between
viral defense and gene silencing in plants. Science 276, 1558–1560.
Ratcliff, F.G., MacFarlane, S.A., Baulcombe, D.C., 1999. Gene silencing
without DNA: RNA-mediated cross-protection between viruses. Plant
Cell 11, 1207–1215.
Roth, B.M., Pruss, G.J., Vance, V.B., 2004. Plant viral suppressors of
RNA silencing. Virus Res., this issue.
Ruiz, M.T., Voinnet, O., Baulcombe, D.C., 1998. Initiation and mainte-
nance of virus-induced gene silencing. Plant Cell 10, 937–946.
Schwarz, D.S., Hutvagner, G., Haley, B., Zamore, P.D., 2002. Evidence
that siRNAs function as guides, not primers, in the Drosophila and
human RNAi pathways. Mol. Cell 10, 537–548.
Sijen, T., Fleenor, J., Simmer, F., Thijssen, K.L., Parrish, S., Timmons,
L., Plasterk, R.H.A., Fire, A., 2001. On the role of RNA amplification
in dsRNA-triggered gene silencing. Cell 107, 465–476.
Silhavy, D., Molnar, A., Lucioli, A., Szittya, G., Hornyik, C., Tavazza,
M., Burgyan, J., 2002. A viral protein suppresses RNA silencing
and binds silencing-generated, 21- to 25-nucleotide double-stranded
RNAs. EMBO J. 21, 3070–3080.
Svoboda, P., Stein, P., Hayashi, H., Schultz, R.M., 2000. Selective reduc-
tion of dormant maternal mRNAs in mouse oocytes by RNA interfer-
ence. Development 127, 4147–4156.
Szittya, G., Molnar, A., Silhavy, D., Hornyik, C., Burgyan, J., 2002.
Short defective interfering RNAs of tombusviruses are not targeted but
trigger post-transcriptional gene silencing against their helper virus.
Plant Cell 14, 359–372.
Tang, G., Reinhart, B.J., Bartel, D.P., Zamore, P.D., 2003. A biochemical
framework for RNA silencing in plants. Genes Dev. 17, 49–63.
Uhlirova, M., Foy, B.D., Beaty, B.J., Olson, K.E., Riddiford, L.M., Jindra,
M., 2003. Use of Sindbis virus-mediated RNA interference to demon-
strate a conserved role of Broad-Complex in insect metamorphosis.
Proc. Natl. Acad. Sci. USA 100, 15607–15612.
Vaistij, F.E., Jones, L., Baulcombe, D.C., 2002. Spreading of RNA tar-
geting and DNA methylation in RNA silencing requires transcription
of the target gene and a putative RNA-dependent RNA polymerase.
Plant Cell 14, 857–867.
Van der Krol, A.R., Mur, L.A., Beld, M., Mol, J.N., Stuitje, A.R., 1990.
Flavonoid genes in petunia: addition of a limited number of gene
copies may lead to a suppression of gene expression. Plant Cell 2,
291–299.
Vance, V., Vaucheret, H., 2001. RNA silencing in plants—defense and
counterdefense. Science 292, 2277–2280.
Voinnet, O., Baulcombe, D.C., 1997. Systemic signalling in gene silencing.
Nature 389, 553.
Voinnet, O., Lederer, C., Baulcombe, D.C., 2000. A viral movement
protein prevents spread of the gene silencing signal in Nicotiana
benthamiana. Cell 103, 157–167.
Voinnet, O., Pinto, Y.M., Baulcombe, D.C., 1999. Suppression of gene
silencing: a general strategy used by diverse DNA and RNA viruses
of plants. Proc. Natl. Acad. Sci. U.S.A. 96, 14147–14152.
Whitton, J.L., Oldstone, B.A., 2001. The immune responses to viruses. In:
Knipe, D.M., Howley, P.M. (Eds.), Fields Virology, 4th ed. Lippincott
Williams & Wilkins, Philadephia, pp. 285–320.
Wianny, F., Zernicka-Goetz, M., 2000. Specific interference with gene
function by double-stranded RNA in early mouse development. Nat.
Cell Biol. 2, 70–75.
 S.-W. Ding et al. / Virus Research 102 (2004) 109–115 115

Xie, Z., Fan, B., Chen, C., Chen, Z., 2001. An important role of an
inducible RNA-dependent RNA polymerase in plant antiviral defense.
Proc. Natl. Acad. Sci. U.S.A. 98, 6516–6521.
Xin, H.W., Ding, S.W., 2003. Identification and molecular characterization
of a naturally occurring RNA virus mutant defective in the initiation
of host recovery. Virology 317, 253–262.
Yu, D., Fan, B., MacFarlane, S.A., Chen, Z., 2003. Analysis of the in-
volvement of an inducible Arabidopsis RNA-dependent RNA poly-
merase in antiviral defense. Mol. Plant Microb. Interact. 16, 206–
216.
Zamore, P.D., Tuschl, T., Sharp, P.A., Bartel, D.P., 2000. RNAi:
Double-stranded RNA directs the ATP-dependent cleavage of mRNA
at 21–23 nucleotide intervals. Cell 101, 25–33.
Zhang, H., Kolb, F.A., Brondani, V., Billy, E., Filipowicz, W., 2002.
Human Dicer preferentially cleaves dsRNAs at their termini without
a requirement for ATP. EMBO J. 21, 5875–5885.
 Virus Research 102 (2004) 117–124

Generation of an shRNAi expression library against

the whole human transcripts
Makoto Miyagishi a,b , Sahohime Matsumoto a,b,c , Kazunari Taira a,b,∗
aDepartment of Chemistry and Biotechnology, School of Engineering, The University of Tokyo, Hongo, Tokyo 113-8656, Japan
b Gene Function Research Center, National Institute of Advanced Industrial Science and Technology (AIST),

Central 4, 1-1-1 Higashi, Tsukuba Science City 305-8562, Japan

c Department of Cardiovascular Medicine, Graduate School of Medicine, The University of Tokyo, Hongo, Tokyo 113-0033, Japan

Abstract

RNA interference is a phenomenon in which expression of an individual gene can be specifically silenced by introducing a double-stranded
RNA, homologous to the gene, and is receiving attention as a powerful tool for reverse genetics in the post-genome era. Throughout our
current research to generate an siRNA expression library for the whole human genome, we face many technical difficulties. We present here the
strategies for overcoming some of the difficulties, including the development of genetically stable and highly active siRNA expression vectors,
the selection procedure of the favorable target sites, and the efficient and low cost procedure for constructing an siRNA expression library.
Furthermore, we demonstrate that the screening using the constructed siRNA-expression library indeed works, by evaluating siRNA-expression
library against apoptosis-related genes.
© 2004 Published by Elsevier B.V.

Keywords: RNA interference; U6 promoter; Algorithm; RNAi library

1. Introduction co-workers demonstrated that 21- or 22-nt RNAs with two

Recent advances in the Human Genome Project and avail-
ability of the complete genome sequence have clarified the
existence of numerous genes whose functions are unknown.
Together with the hybrid ribozyme technology (Warashina
et al., 2001; Kawasaki and Taira, 2002; Kawasaki et al.,
2002; Suyama et al., 2003a,b), RNA interference (RNAi) is
receiving attention as the technology that could unveil their
functions (McManus and Sharp, 2002).
RNAi is an evolutionarily conserved process in plants and
animals by which double-stranded RNA (dsRNA) induces
sequence specific degradation of homologous RNA (Fire
et al., 1998; reviewed in Zamore, 2001). Although RNAi
works well in various organisms, the silencing of specific
genes by RNAi has proven difficult to detect in mammalian
systems, because of the dsRNA-dependent nonspecific in-
hibition of protein synthesis by the PKR pathway and the
nonspecific RNA degradation by activation of RNase L
(Elbashir et al., 2001). However, recent works by Tuschl and
∗ Corresponding author. Tel.: +81-3-5841-8828/298-61-3015;
fax: +81-3-5841-7340/8828.
E-mail address: taira@chembio.t.u-tokyo.ac.jp (K. Taira).
nt 3 -overhangs (siRNA) can induce gene silencing without
the nonspecific inhibition of the gene expression in cultured
mammalian cells (Caplen et al., 2001; Elbashir et al., 2001).
Furthermore, in the past year, various groups, including our
own, have developed systems for vector-mediated specific
RNAi in mammalian cells (Brummelkamp et al., 2002;
Kawasaki and Taira, 2003; Lee et al., 2002; Miyagishi and
Taira, 2002; Paddison et al., 2002; Sui et al., 2002; Paul
et al., 2002; Yu et al., 2002; reviewed in McManus and
Sharp, 2002; Tuschl, 2002). These vector systems use a pol
III promoter, such as the U6, H1, and tRNA promoters, and
the systems have been classified into two groups depend-
ing on whether the expressed RNAs are tandem-type or
hairpin-type.
Since gene silencing by siRNA is highly specific and has
extremely high activity, genome-wide and comprehensive
gene analysis in various organism should be expected. Re-
cently, in Caenorhabditis elegans, a systematic functional
analysis of a number of genes by RNAi has been per-
formed, and many of these genes were identified as genes
that show mutant phenotype (Fraser et al., 2000; Gonczy
et al., 2000). In the near future, it is predicted that a similar
genome-wide RNAi library approach could be extended

0168-1702/$ – see front matter © 2004 Published by Elsevier B.V.

doi:10.1016/j.virusres.2004.01.022
118 M. Miyagishi et al. / Virus Research 102 (2004) 117–124

to mammalian cells by an siRNA oligonucleotide library Tandem type

or an siRNA expression library, with the success that ran-
+1 +1
domized ribozyme libraries have had in the identification U6 or H1 promoter Sense TTTTT U6 or H1 promoter
Antisense
TTTTT
of novel genes (Beger et al., 2001; Kawasaki and Taira,
2002; Kawasaki et al., 2002; Kruger et al., 2000; Li et al.,
2000; Onuki et al., 2002; Rhoades and Wong-Staal, 2003; Transcription,
hybridization
Suyama et al., 2003a,b; Welch et al., 2000). In fact, several
Sense
groups, including ours, have already begun the generation
of an siRNA library against the whole human genome. In Antisense

this report, we present the practical problems and meth- siRNA duplex
ods for solving some of the difficulties that were faced in
generating the siRNA expression library. Hairpin type
+1 Loop
Antisense
U6 or H1 promoter TTTTT
2. Materials and methods

2.1. Constructs used in transient transfections

Transcription

We prepared tandem-type and hairpin-type siRNA expres-

Sense
sion vectors using the piGENE hU6 vector (Miyagishi and
Taira, 2003; iGENE Therapeutics, Inc.; http//www.iGENE- Antisense

therapeutics.co.jp), which contains a human U6 promoter Hairpin RNA

and two BspMI sites. For the construction of tandem-type
Processing
siRNA expression plasmids, we amplified DNA fragments
that included the sequences of the sense and antisense
Sense
regions and the U6 promoter by PCR with the piGENE
hU6 vector as a template, and primers that included the Antisense

sequences of the sense or antisense sequence plus the ter-
minator (Fig. 1). After digestion of products of PCR with
BspMI, each fragment was ligated into the BspMI sites of
the piGENE hU6 vector to yield a series of siRNA expres-
sion vectors. For the construction of hairpin-type siRNA
expression vectors, we synthesized oligonucleotides with
the hairpin sequence, the terminator sequence, and over-
hanging sequences. Then we annealed the fragments and
inserted them into the BspMI sites of the piGENE hU6
vector. H1 promoter-driven siRNA expression vectors were
generated as described by Brummelkamp et al. (2002).
Sequences inserted immediately downstream of the U6 pro-
moters or the H1 promoters were as follows (only the sense
sequences are shown): in pU6tandem19, pU6hairpin19,
and pH1hairpin19, GTG CGC TGC TGG TGC CAA C; in
pU6tandem26, GTG CGC TGC TGG TGC CAA CCC TAT
TC; in pU6hairpin21, GTG CGC TGC TGG TGC CAA
CCC; in pU6hairpin23, GTG CGC TGC TGG TGC CAA
CCC TA; in pU6hairpin25, GTG CGC TGC TGG TGC
CAA CCC TAT T. The sequences of the various mutated
constructs are indicated in Fig. 2.
2.2. Culture and transfection of cells and assays of the
expression of reporter genes
HeLa S3 cells were cultured in Dulbecco’s modified
Eagle’s medium supplemented with 10% fetal bovine serum.
Transfections were performed with LipofectamineTM 2000
reagent (Life Technologies, Rockville, MD, USA) using
30 ng of siRNA expression plasmid, 30 ng of firefly lu-
siRNA duplex
Fig. 1. Schematic representation of tandem-type and hairpin-type
siRNA-expression vectors.
ciferase expression plasmid (pGL3; Promega, Madison,
WI, USA), and 10 ng of Renilla luciferase expression plas-
mid (pRL-RSV; Miyagishi et al., 2000), as described by
Promega. Luciferase activities were analyzed after 24 h
with a Dual Luciferase System (Promega). To ensure equal
DNA amounts, empty plasmids were added at appropriate
levels in each transfection.
2.3. Algorithm for searching favorable target sites
Prediction modeling of the RNAi effect was performed
by partial least squares (PLS) method. One hundred and
fourteen data sets of siRNA sequences and their suppression
values, obtained from published results and our own data,
were analyzed and significantly correlated factors, including
GC contents, were extracted. A PLS calibration model was
made with the target site sequences, their suppression values
and the extracted factors using the PLS1 algorithm of our
own making (Lindberg et al., 1983). The optimum number
of PLS factors was determined by cross-validation, that is,
the best calibration as judged by the lowest standard error
of cross-validation. After equations were established, the
suppression values at the various sites of the firefly luciferase
were predicted using their target sequences.
 M. Miyagishi et al. / Virus Research 102 (2004) 117–124 119

Fig. 2. (A) Comparative analysis of the effects of the U6 promoter-driven 21-nt tandem-type and 21-nt hairpin-type siRNA expression vectors, the H1
promoter-driven 21-nt hairpin-type siRNA-expression vector and the U6-promoter-driven 21-nt hairpin-type with various mutations in hairpin region.
HeLa S3 cells were cotransfected with 10 ng of pRL-RSV, 30 ng of pGL3, and 30 ng of each siRNA expression vector targeted to the firefly gene
for luciferase. The activities of firefly luciferase were normalized by reference to those of Renilla luciferase. Each bar indicates an average value and
horizontal bars indicate standard errors of triplicate assays. (B) Schematic representation of the proposed siRNA expression system.

3. Results and discussion
3.1. Optimization of siRNA expression system: tandem-type
or hairpin-type?
To construct the siRNA expression library, an optimiza-
tion of the siRNA expression system, that has high suppres-
sive activity and high genetic stability, should be initially
addressed.
The major class of siRNA expression systems uses pol
III promoter, such as U6 or H1 promoter. The systems have
been classified into two groups, tandem-type or hairpin-type,
depending on their expressing RNA (Fig. 1). In the case
of tandem-type siRNA expression vector, both sense and
antisense strands are driven separately by a respective pol
III promoter, anneal in the cells and form siRNA duplexes
with about four-nt overhang at each 3 -end (Lee et al., 2002;
Miyagishi and Taira, 2002). In the hairpin-type of siRNA ex-
pression, sense and antisense nucleotides are connected by a
loop and expressed as a single chain (Brummelkamp et al.,
2002; Kawasaki and Taira, 2003; Paddison et al., 2002; Paul
et al., 2002; Sui et al., 2002; Taira and Miyagishi, 2001; Yu
et al., 2002). The transcribed RNA will rapidly form a hair-
pin structure with a stem and loop and seems to be processed
to siRNA by Dicer, like micro RNAs (miRNAs) (Ketting
et al., 2001). We tried to examine which vector system had
higher suppressive activity. We constructed both types of
siRNA expression vectors, targeted against the same site of
firefly luciferase gene, and compared their suppressive ac-
tivities. Although both had high levels of suppressive activ-
ities at a high concentration of siRNA expression vectors,
it was found that the hairpin-type siRNA expression vec-
tor (21-nt stem length) had significantly higher suppressive
activity than the tandem-type siRNA expression vector at a
low concentration of the siRNA expression vector (Fig. 2A).
3.2. Genetically stable hairpin-type siRNA expression
vector
Although the hairpin-type siRNA expression vector has
higher suppressive activity than the tandem-type siRNA ex-
pression vector, the palindromic sequence of the hairpin-type
siRNA expression vector frequently causes two serious tech-
nical problems. Firstly, it is difficult to sequence constructs
that contained a hairpin region, probably because of the tight
palindromic structure. Secondly, approximately 20–40% of
constructs are mutated within the hairpin region of the con-
structs upon introduction into Escherichia coli, which is the
most serious problem for the library construction. We were
aware that constructs with two or more point mutations or
insertions/deletions in the sense or antisense region could be
sequenced without any problems and some of them retained
the suppressive activity. Therefore, we examined in detail
the activities of constructs with various mutations and/or
120
insertions/deletions in the sense or antisense region (detailed
analysis will be published elsewhere). As shown in Fig. 2A,
as many as five-point mutations from C to T, which gen-
erated G:U base-pairing, in the sense strand did not affect
the silencing effect (pU6hairpin21C > T5 in Fig. 2A). In
contrast, insertion of a single C–T mutation in the antisense
strand (pU6hairpinG > A1(AS) in Fig. 2A), which recog-
nizes the target transcript, resulted in reduced suppression.
Therefore, it appears that, in practice, it is not possible to
introduce mutations into the antisense strand and retain ac-
tivity.
While generating these constructs with mutated hairpin
structures, we found that the rates of mutation in E. coli of
the hairpin region of the constructs were markedly reduced
when we inserted more than three mutations in the sense or
antisense strand (Taira and Miyagishi, 2001). This property
of such constructs is obviously important for the mainte-
nance of corresponding plasmids in their prokaryotic host.
Collectively, our data suggest that the way to generate a use-
ful hairpin-type siRNA expression vector that enables us to
construct an siRNA expression library is to introduce multi-
ple C–T (or A–G) mutations only into the sense strand part
of the hairpin (Fig. 2B).
3.3. Comparative analysis of various pol III promoters for
silencing activity
We also compared the RNAi activity of the U6 promoter
and the H1 promoter, both of which belong to the type II
Polymerase III promoters, and contain the same consensus
motifs, distal sequence element (DSE), proximal sequence
element (PSE) and TATA box. As shown in Fig. 2A, H1
1.0
Observed suppressive activity (relative)
0.8
0.6
0.4
0.2
0.0
0.0
Fig. 3. Relationship between predicted and actual values of suppression activity. The actual values were measured by co-transfection experiments with
10 ng of pRL-RSV, 30 ng of pGL3, and 10 nM of each siRNA oligonucleotides targeted to the firefly gene for luciferase into HeLa cells. The activities
of firefly luciferase were normalized by reference to those of Renilla luciferase and the suppression values were calculated as the ratio to the negative
control. The predicted values were calculated from the target sequence of the siRNA with the PLS algorithm.
M. Miyagishi et al. / Virus Research 102 (2004) 117–124
promoter-driven siRNA construct targeted against firefly
luciferase has an activity comparable, but slightly lower,
than that of the U6-driven siRNA construct (comparing
pH1hairpin21C > T5 with pU6hairpin21C > T5 in Fig 2A,
and other unpublished results). Therefore, we selected
U6-driven siRNA expression vector as the promoter for
the siRNA expression library. The tRNA-linked siRNA ex-
pression vector is a recently developed expression system,
which would enable us to forcibly transport the hairpin RNA
to the cytoplasm (Kawasaki and Taira, 2003). However, it
requires a longer stem length (30-nt) and it would be more
costly; hence we adopted the U6-driven siRNA expression
vector for the construction of the siRNA expression library.
3.4. Optimization of the loop sequence
Different loop sequences of a hairpin-type siRNA ex-
pression were used according to respective researchers.
For example, Agami’s group uses a 9-nt loop sequence,
5 -UUCAAGAGA-3 , (Brummelkamp et al., 2002) and
Paul et al. (2002) use a 4-nt, 5 -UUCG-3 , tetra nucleotide
sequence. Our experiments using hairpin-type siRNA ex-
pression vector with randomized 6-nt loop sequences re-
vealed that the hairpin RNAs with different loop sequences
had different suppression activities (Kawasaki and Taira,
2003; Miyagishi et al., 2004), therefore, the loop sequence
appears to somewhat influence the RNAi effect.
We wondered whether the natural loop sequence of miR-
NAs might be a preferable sequence for the loop of the
hairpin RNA produced from hairpin-type siRNA expres-
sion vectors. Examination of several miRNA-derived loop
sequences revealed that loop 7, which is 11-nt in length and
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Predicted suppressive activity (relative)
 M. Miyagishi et al. / Virus Research 102 (2004) 117–124 121

derived from the human miRNA miR-26b (the sequence is
5 -GUG UGC UGU CC-3 ), is highly effective (the loop
is shown in Fig. 2B). Therefore, we adopted this loop as
the optimized loop sequence of the hairpin-type siRNA
expression vector for the siRNA expression library.
3.5. Method for selection of favorable target site
One of the most important and critical problems of the
siRNA expression library would be selection of the tar-
get sites. It is well-known that the effectiveness of siRNA
is very dependent on the target site of a message and the
possibility that siRNAs at an arbitrarily selected target site
have high suppressive activity, seems to be only at about
20–40% or less. To generate an siRNA expression library of
high-quality, an algorithm that can accurately predict an ef-
fective site must be developed. We analyzed siRNA data sets
of several hundreds of target sites, including published work
and our own experimental data, extracted several correlated
factors, and generated an algorithm by nonlinear regression
methods.
Fig. 3 shows predicted versus measured results for siR-
NAs targeted against firefly luciferase, which were not used
in the generation of the algorithm. The correlation coeffi-
cient was not high (0.58) but, notably, all of the siRNAs that
were calculated as scoring more than 0.73 showed high sup-
pressive activities (more than 90%), therefore, especially at
high prediction values, the predicted scores seem to be ac-
curate. We use target sites that have a predicted suppressive
activity of more than 0.75 for the construction of the siRNA
expression library.
3.6. Confirmation of specificity of the target sequence
When one selects a certain target site in a specific gene,
there is a need to check the specificity of the sequence of the

(A) piGENE hU6

EcoRI BspMI BspMI
TTTTT
U6 promoter

Transcription start site

U6 promoter BspMI
+1
linker
CTTGTGGAAAGGACGAAA CACCGTGAGCAGGT---------------ACCTGCAGGC ATGCAAGCTT
GAACACCTTTCCTGCTTTGTGG CACTCGTCCA---------------TGGACGTCCGTACG TTCGAA
Cleavage site Cleavage site
BspMI

BspMI Loop BspMI

5'-CACC sense gtgtgctgtcc antisense ttttt-3
3'- sense cacacgacacc antisense aaaaaTACG-5'

(B)

Annealing individual
oligos (96 oligos) Ligated into
the U6 plasmid

Pick up 384 colonies

Sequencing
About 90% clones could be recovered

Fig. 4. (A) Cloning site and the insert sequence of hairpin-type siRNA-expression vector. (B) Our strategy for the construction of siRNA-expression library.
 122
M. Miyagishi et al. / Virus Research 102 (2004) 117–124
Fig. 5. Identification of genes involved in apoptosis by using shRNAi library. (A) The siRNA-expression vectors targeted against PKR can block dsRNA-induced apoptosis. HeLaS3 cells were transfected
with siRNA-expression vectors targeted against two sites of PKR transcript (Site A; 5’TAA TGA ATC AAT CAA TTC ATA TC-3’ and Site B; 5’AAG ACT AAC TGT AAA TTA TGA AC-3’). After
the selection with puromycin, the cells were induced to undergo apoptosis by dsRNA transfection. The survived cells were fixed, and stained with crystal violet (0.2%). To check the reproducibility, each
independent plasmid was transfected into cells in three different wells (triplicate experiments). NC: negative control (unrelated siRNA-expression vector). (B) Light microscopic images of the cells in the
experiment of (B). (C) An example of the screening using siRNA-expression library. PC: positive control; NC1, NC2: negative controls (unrelated siRNA-expression vectors); C1–C11: siRNA-expression
vectors targeted against specific genes, for example kinases, transcription factors, or apoptosis-related genes. The shRNA expression vector in C7 inhibits the dsRNA-dependent apoptosis.
 M. Miyagishi et al. / Virus Research 102 (2004) 117–124
target site, that is, to examine that there are no highly ho-
mologous sequences in other genes. Under what conditions
should one check the specificity? If very stringent condi-
tions are applied, for example, more than three base mis-
matches are needed to determine the discrimination between
the correct and homologous targets, significant portions of
favorable target sites that were predicted as highly active
sites could be eliminated by the BLAST search program. On
the other hand, if less stringent conditions are applied, for
example, only one base mismatch is allowed for the dis-
crimination, the designed siRNA might have the potential to
disrupt not only the correct target but also other homologous
genes as well, although the cleavage efficiency is reduced
for the latter. Under these circumstances, we have decided
to use the relatively low stringent condition for examining
target site specificity, because multiple positive candidates
can be further analyzed by creating new siRNAs targeted at
different sites within the candidate gene.
Alternative splicing and highly homologous gene fami-
lies are also major problems for target site selection. Re-
cent genome-wide analysis of transcripts reveals that many
genes have alternative splicing forms. For example, in case
where one tries to disrupt a specific alternative variant, an
siRNA expression vector targeted against the specific vari-
ant can be created, without damaging other alternative vari-
ants originating from the same pre-spliced form. However,
for the generation of the first draft library, in the targeting
of genes which have multiple alternative splicing forms, the
common sequence of all alternative transcripts (similarly,
the common sequence for the homologous gene family) can
be aimed as a target site. This is more economical for the
creation of the first library and additional siRNAs can eas-
ily be made once the targeted gene is identified as being an
attractive candidate.
3.7. Improved method for library construction
In constructing an siRNA expression library against all
genes in the genome, one would become focused on practical
problems, such as how to construct the library efficiently,
rapidly, and at a reduced price.
We have established a large-scale method for construct-
ing a library, as shown in Fig. 4. Specifically, 96 sets of
oligonucleotides that include hairpin sequences correspond-
ing to 96 target sequences were synthesized and annealed
separately. Then the annealed oligonucleotides were mixed,
and were ligated into BspMI site of the U6-driven siRNA
expression vector. After transformation into E. coli, 384
clones were picked up and were sequenced to identify each
clone. In our data, about 88.5% of the clones were recov-
ered in a single procedure. The unrecovered clones were
followed by the next round of the construction. This bulk
procedure allowed the production of an siRNA expression
vector dramatically cheaper and faster than the traditional
procedure in which each clone is made separately. Us-
ing this procedure, we can practically make 3000–5000
123
siRNA expression vectors per month (Taira and Miyagishi,
2001).
3.8. Advantages and disadvantages of siRNA expression
library compared to siRNA oligonucleotide library
For the plasmid-based siRNA expression library and the
alternative oligonucleotide-based siRNA library, there are
advantages and disadvantages.
Transfection efficiency of siRNA into cells using lipofec-
tion depends on cell types and the RNAi effect only seems
to be sustained for a period of time. The advantage of the
plasmid-based siRNA is the capability of removing those
cells that were not transfected with the plasmids by se-
lecting the transfected cells with antibiotic resistant genes.
Moreover, the RNAi effect lasts for much longer periods
when plasmid-based siRNAs are used. Additionally, virus
vectors enable us to deliver siRNA expression cassettes into
cells with higher transfection efficiency, and in the case of
lentivirus and retrovirus, it is easy to make stable knock-
down cells by the integration of the virus vector into the
genome. If one wants to use siRNA expression vectors as
a bulk library, these virus vectors, which can generate sta-
ble knockdown cells, would be most suitable for the bulk
library screen.
The advantage of oligonucleotide-based siRNA is that, in
some favorable cells, transfection efficiency is greater than
90%, and thus higher than that of the plasmid. However, for
the library, once plasmid-based siRNAs are generated, they
can be amplified unlimitedly, especially with our construct
(Fig. 2B) which does not undergo unwanted mutation (un-
like conventional ones) during amplification in E. coli cells,
resulting in a more economical construction of a library
set.
4. Conclusions
During the present work, we have developed a genet-
ically stable and highly active siRNA expression vector,
and have established a system for constructing high-quality
and genome-wide scale siRNA expression libraries very
efficiently. Recently, we have started to screen functional
genes by using the constructed siRNA-expression library.
Fig. 5 shows an example of screening of functional genes in
the apoptosis pathway. Positive controls (siRNA-expression
vectors targeted against PKR) could block dsRNA-induced
apoptosis (Fig. 5A), and in the experiments using siRNA ex-
pression vectors targeted against unknown genes (Fig. 5B;
C1–C11), one positive clone could be identified fron 11 can-
didates (Fig. 5B; C7). Within the next few years, several
groups, including our own would expect to identify a num-
ber of genes through the use of various siRNA libraries.
In the past, we were able to identify miRNA sequences by
our ribozyme library (Kuwabara et al., 2004). Since miRNA
plays an important role in development, and possibly in other
124 M. Miyagishi et al. / Virus Research 102 (2004) 117–124
biological phenomena of mammalian cells (Kawasaki et al.,
2004), the siRNA library should also cover miRNAs and un-
defined noncoding RNAs as targets. Scientists are making
the first great step in functional genome analysis, and human
beings will soon enjoy deeper knowledge of the unraveled
life of cells.
References
Beger, C., Pierce, L.N., Kruger, M., Marcusson, E.G., Robbins, J.M.,
Welcsh, P., Welch, P.J., Welte, K., King, M.C., Barber, J.R.,
Wong-Staal, F., 2001. Identification of Id4 as a regulator of BRCA1
expression by using a ribozyme-library-based inverse genomics ap-
proach. Proc. Natl. Acad. Sci. U.S.A. 98, 130–135.
Brummelkamp, T.R., Bernards, R., Agami, R., 2002. A system for stable
expression of short interfering RNAs in mammalian cells. Science
296, 550–553.
Caplen, N.J., Parrish, S., Imani, F., Fire, A., Morgan, R.A., 2001. Spe-
cific inhibition of gene expression by small double-stranded RNAs in
invertebrate and vertebrate systems. Proc. Natl. Acad. Sci. U.S.A. 98,
9742–9747.
Elbashir, S.M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., Tuschl,
T., 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference
in cultured mammalian cells. Nature 411, 494–498.
Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver, S.E., Mello,
C.C., 1998. Potent and specific genetic interference by double-stranded
RNA in Caenorhabditis elegans. Nature 391, 806–811.
Fraser, A.G., Kamath, R.S., Zipperlen, P., Martinez-Campos, M.,
Sohrmann, M., Ahringer, J., 2000. Functional genomic analysis of C.
elegans chromosome I by systematic RNA interference. Nature 408,
325–330.
Gonczy, P., Echeverri, C., Oegema, K., Coulson, A., Jones, S.J., Copley,
R.R., Duperon, J., Oegema, J., Brehm, M., Cassin, E., Hannak, E.,
Kirkham, M., Pichler, S., Flohrs, K., Goessen, A., Leidel, S., Alleaume,
A.M., Martin, C., Ozlu, N., Bork, P., Hyman, A.A., 2000. Functional
genomic analysis of cell division in C. elegans using RNAi of genes
on chromosome III. Nature 408, 331–336.
Kawasaki, H., Taira, K., 2002. Identification of genes by hybrid ribozymes
that couple cleavage activity with the unwinding activity of an en-
dogenous RNA helicase. EMBO Rep. 3, 443–450.
Kawasaki, H., Taira, K., 2003. Short hairpin type of dsRNAs that are
controlled by tRNAVal promoter significantly induce RNAi-mediated
gene silencing in the cytoplasm of human cells. Nucleic Acids Res.
31, 700–707.
Kawasaki, H., Onuki, R., Suyama, E., Taira, K., 2002. Identification of
genes that function in the TNF-alpha-mediated apoptotic pathway
using randomized hybrid ribozyme libraries. Nat. Biotechnol. 20, 376–
380.
Kawasaki, H., Wadhwa, R., Taira, K., 2004. World of small RNAs: from
ribozymes to siRNA and miRNA. Differentiation, in press.
Ketting, R.F., Fischer, S.E., Bernstein, E., Sijen, T., Hannon, G.J., Plasterk,
R.H., 2001. Dicer functions in RNA interference and in synthesis
of small RNA involved in developmental timing in Caenorhabditis
elegans. Genes Dev. 15, 2654–2659.
Kruger, M., Beger, C., Li, Q.X., Welch, P.J., Tritz, R., Leavitt, M.,
Barber, J.R., Wong-Staal, F., 2000. Identification of eIF2Bgamma and
eIF2gamma as cofactors of hepatitis C virus internal ribosome entry
site-mediated translation using a functional genomics approach. Proc.
Natl. Acad. Sci. U.S.A. 97, 8566–8571.
Kuwabara, T., Hsieh, J., Nakashima, K., Taira, K., Gage, F.H., 2004.
Small Modulatory dsRNA specifies the Fate of Adult Neural Stem
Cells. Cell, in press.
Lee, N.S., Dohjima, T., Bauer, G., Li, H., Li, M.J., Ehsani, A., Salvaterra,
P., Rossi, J., 2002. Expression of small interfering RNAs targeted
against HIV-1 rev transcripts in human cells. Nat. Biotechnol. 20,
500–505.
Li, Q.X., Robbins, J.M., Welch, P.J., Wong-Staal, F., Barber, J.R., 2000. A
novel functional genomics approach identifies mTERT as a suppressor
of fibroblast transformation. Nucleic Acids Res. 28, 2605–2612.
Lindberg, W., Persson, J.A., Wold, S., 1983. Partial least-squares method
for spectrofluorimetric analysis of mixtures of humic acid and lignin
sulfonate. Anal. Chem. 55, 643–648.
McManus, M.T., Sharp, P.A., 2002. Gene silencing in mammals by small
interfering RNAs. Nat. Rev. Genet. 3, 737–747.
Miyagishi, M., Taira, K., 2002. U6 promoter-driven siRNAs with four
uridine 3 overhangs efficiently suppress targeted gene expression in
mammalian cells. Nat. Biotechnol. 20, 497–500.
Miyagishi, M., Taira, K., 2003. Expression of siRNA from a pol III
promoter in mammalian cells. In: Appasani, K. (Ed.), Perspectives
in Gene Expression, The Eaton Publishers, Westboro, MA. pp. 361–
376.
Miyagishi, M., Fujii, R., Hatta, M., Yoshida, E., Araya, N., Nagafuchi,
A., Ishihara, S., Nakajima, T., Fukamizu, A., 2000. Regulation of
Lef-mediated transcription and p53-dependent pathway by associating
␤-catenin with CBP/p300. J. Biol. Chem. 275, 35170–35175.
Miyagishi, M., Suminoto, H., Miyoshi, H., Kawakami, Y., Taira, K., 2004.
Optimization of an siRNA-expression system with a mutated hair-
pin and its significant suppressive effects upon HIV vector-mediated
transfer into mammalian cells. J. Gene Med., in press.
Onuki, R., Nagasaki, A., Kawasaki, H., Baba, T., Uyeda, T.Q., Taira, K.,
2002. Confirmation by FRET in individual living cells of the absence
of significant amyloid beta-mediated caspase 8 activation. Proc. Natl.
Acad. Sci. U.S.A. 99, 14716–14721.
Paddison, P.J., Caudy, A.A., Bernstein, E., Hannon, G.J., Conklin, D.S.,
2002. Short hairpin RNAs (shRNAs) induce sequence-specific silenc-
ing in mammalian cells. Genes Dev. 16, 948–958.
Paul, C.P., Good, P.D., Winer, I., Engelke, D.R., 2002. Effective expression
of small interfering RNA in human cells. Nat. Biotechnol. 20, 505–
508.
Rhoades, K., Wong-Staal, F., 2003. Inverse Genomics as a powerful tool to
identify novel targets for the treatment of neurodegenerative diseases.
Mech. Ageing Dev. 124, 125–132.
Sui, G., Soohoo, C., Affarel, B., Gay, F., Shi, Y., Forrester, W.C., Shi,
Y., 2002. A DNA vector-based RNAi technology to suppress gene
expression in mammalian cells. Proc. Natl. Acad. Sci. U.S.A. 99,
5515–5520.
Suyama, E., Kawasaki, H., Kasaoka, T., Taira, K., 2003a. Identification
of genes responsible for cell migration by a library of randomized
ribozymes. Cancer Res. 63, 119–124.
Suyama, E., Kawasaki, H., Nakajima, M., Taira, K., 2003b. Identification
of genes involved in cell invasion by using a library of randomized
hybrid ribozymes. Proc. Natl. Acad. Sci. U.S.A. 100, 5616–5621.
Taira, K., Miyagishi, M., 2001. siRNA Expression System and Method
for Producing Functional Gene Knock-down Cell Using the System.
Japanese Patent Application, Heiser–13–363385 (PCT/JP02/12447).
Tuschl, T., 2002. Expanding small RNA interference. Nat. Biotechnol.
20, 446–448.
Warashina, M., Kuwabara, T., Kato, Y., Sano, M., Taira, K., 2001.
RNA-protein hybrid ribozymes that efficiently cleave any mRNA in-
dependently of the structure of the target RNA. Proc. Natl. Acad. Sci.
U.S.A. 98, 5572–5577.
Welch, P.J., Marcusson, E.G., Li, Q.X., Beger, C., Kruger, M., Zhou,
C., Leavitt, M., Wong-Staal, F., Barber, J.R., 2000. Identification and
validation of a gene involved in anchorage-independent cell growth
control using a library of randomized hairpin ribozymes. Genomics
66, 74–83.
Yu, J.Y., DeRuiter, S.L., Turner, D.L., 2002. RNA interference by ex-
pression of short-interfering RNAs and hairpin RNAs in mammalian
cells. Proc. Natl. Acad. Sci. U.S.A. 99, 6047–6052.
Zamore, P.D., 2001. RNA interference: listening to the sound of silence.
Nat. Struct. Biol. 8, 746–750.
 | |

An International Journal of Molecular and Cellular Virology

Instructions to Authors
EDITOR-IN-CHIEF
These can be found in the prelims of the first issue of each volume, and accessed on the
Dr. B.W.J. Mahy, Virus Research, Mailstop C12, National Center for Infectious Diseases, Centers for Disease Control and Prevention, World Wide Web under http://www.elsevier.com
Atlanta, GA 30333, USA. Tel. No.: þ1 404 728-0564; Fax No.: þ1 404 728-0032. E-mail: virology@bellsouth.net
This Journal is edited by Brian WJ Mahy in his private capacity. No official support or endorsement by CDC is intended or should be
inferred.

EDITORS Author enquiries

Dr. L. Enjuanes (European Editor), Virus Research, Department of Molecular and Cell Biology, National Center of Biotechnology For enquiries relating to the submission of articles (including electronic submission
(CNB), CSIC, Campus Universidad Autonoma, 28049 Madrid, Spain, Tel. No.: þ34 91 585 4555; Fax No.: þ34 91 585 4915. where available) please visit Elsevier’s Author Gateway at http://authors.elsevier.com.
E-mail: L.Enjuanes@cnb.uam.es The Author Gateway also provides the facility to track accepted articles and set up
Dr. Y. Kawaoka (Address for manuscript submission) Virus Research, Department of Pathobiological Sciences, School of Veterinary
Medicine, University of Wisconsin-Madison, 2015 Linden Drive, Madison, WI 53706, USA, Tel. No.: þ1 608 265 4925; Fax No.:
e-mail alerts to inform you of when an article’s status has changed, as well as detailed
þ1 608 265 5622. E-mail: VirusResearch@svm.vetmed.wisc.edu artwork guidelines, copyright information, frequently asked questions and more.
Division of Virology, Department of Microbiology and Immunology, Institute of Medical Science, University of Tokyo, 4-6-1,
Shirokanedai, Minato-Ku, Tokyo 108-8639, Japan Contact details for questions arising after acceptance of an article, especially those
relating to proofs, are provided after registration of an article for publication.
EDITORIAL BOARD
A. Barrett, Galveston, TX, USA R. Goldbach, Wageningen, The S. Nichol, Atlanta, GA, USA
T. Barrett, Pirbright, UK Netherlands A. Nomoto, Tokyo, Japan
Elsevier will do everything possible to get your article corrected and published as
W.J. Bellini, Atlanta, GA, USA F. Gonzales-Scarano, Philadelphia, F. Novembre, Atlanta, GA, USA quickly and accurately as possible. Therefore, it is important to ensure that all of your
W. Britt, Birmingham, AL, USA PA, USA D. O’Callaghan, Shreveport, LA, USA corrections are sent back to us in one communication. Subsequent corrections will not
M.S. Campo, Glasgow, UK D. Griffin, Baltimore, MD, USA N. Raab-Traub, Chapel Hill, NC, USA be possible, so please ensure that your first sending is complete.
D. Cavanagh, Compton, UK L. Guarino, College Station, TX, USA R. Ray, St Louis, MO, USA
P. Collins, Bethesda, MD, USA I. Jones, Reading, UK A.B. Rickinson, Birmingham, UK
R.W. Compans, Atlanta, GA, USA J. Katz, Atlanta, GA, USA J. Ridpath, Ames, IA, USA
K. Conzelmann, Muenchen, Germany H.-D. Klenk, Marburg, Germany B. Robertson, Atlanta, GA, USA
T. Dermody, Nashville, TN, USA M. Koopmans, Bilthoven, L. Roux, Geneva, Switzerland
V. Deubel, Lyon, France The Netherlands D. Rowlands, Leeds, UK
W. Doerfler, Cologne, Germany M. Lai, Los Angeles, CA, USA P. Roy, London, UK
E. Domingo, Madrid, Spain I. Lipkin, New York, NY, USA C. Schmaljohn, Frederick, MD, USA USA mailing notice: Virus Research (ISSN 0168-1702) is published monthly January to
R. Elliott, Glasgow, UK R.B. Luftig, New Orleans, LA, USA M.J. Studdert, Parkville, Victoria, December by Elsevier B.V. (P.O. Box 211, 1000 AE Amsterdam, The Netherlands).
H. Feldmann, Winnipeg, Canada W.S. Mason, Philadelphia, PA, USA Australia
B. Fleckenstein, Erlangen, Germany J.W. McCauley, Compton, UK G.T.W. Wertz, Birmingham, AL, USA Annual subscription price in the USA US$3245 (valid in North, Central and South
T. Folks, Atlanta, GA, USA P. Moore, Pittsburgh, PA, USA America), including air speed delivery. Periodical postage rate paid at Jamaica, NY
A. Garcia-Sastre, New York, NY, USA M. Mulligan, Birmingham, AL, USA 11431.
USA POSTMASTER: Send address changes to Virus Research, Publications Expedit-
AIMS AND SCOPE
ing Inc., 200 Meacham Avenue, Elmont, NY 11003.
Virus Research provides a means of rapid publication for original papers on fundamental research concerning virus structure, replication, AIRFREIGHT AND MAILING in the USA by Publications Expediting Inc., 200
and pathogenesis. These include reports describing virus morphology, the function and antigenic analysis of virus structural components,
virus genome structure and expression, analysis of virus replication processes, effects of viruses on their host cells, including oncogene Meacham Avenue, Elmont, NY 11003.
activation and transduction, neoplastic transformation, and the pathogenesis of virus infections including tumour induction. The journal
also publishes review articles on topics of current interest and occasional book reviews and meeting reports.

Publication information: Virus Research (ISSN 0168-1702). For 2004, volumes 99–106 are scheduled for publication.
Subscription prices are available upon request from the Publisher or from the Regional Sales Office nearest you or from Advertising information: Advertising orders and enquiries can be sent to: USA,
this journal’s website (http://www.elsevier.com/locate/virusres). Further information is available on this journal and other Canada and South America: Mr Tino DeCarlo, The Advertising Department,
Elsevier products through Elsevier’s website: (http://www.elsevier.com). Elsevier Inc., 360 Park Avenue South, New York, NY 10010-1710, USA; phone:
Subscriptions are accepted on a prepaid basis only and are entered on a calendar year basis. Issues are sent by standard
mail (surface within Europe, air delivery outside Europe). Priority rates are available upon request. Claims for missing (þ1) (212) 633 3815; fax: (þ1) (212) 633 3820; e-mail: t.decarlo@elsevier.com.
issues should be made within six months of the date of dispatch. Japan: The Advertising Department, Elsevier K.K., 4F Higashi-Azabu, 1-Chome
Bldg, 1-9-15 Higashi-Azabu, Minato-ku, Tokyo 106-0044, Japan; phone: (þ81) (3)
Orders, claims, and journal enquiries: please contact the Customer Service Department at the Regional Sales Office nearest you:
Orlando: Elsevier, Customer Service Department, 6277 Sea Harbor Drive, Orlando, FL 32887-4800, USA; phone: (þ1) (877) 8397126
5561 5037; fax: (þ81) (3) 5561 5047; e-mail: jp.info@elsevier.com. Europe and
[toll free number for US customers], or (þ1) (407) 3454020 [customers outside US]; fax: (þ1) (407) 3631354; e-mail: usjcs@ ROW: Katrina Barton, Print Operations Co-ordinator, Elsevier, 84 Theobald’s Road,
elsevier.com London WC1X 8RR, UK. Tel.: þ44 (0) 20 7611 4117; Fax: þ44 (0) 20 7611 4463;
Amsterdam: Elsevier, Customer Service Department, PO Box 211, 1000 AE Amsterdam, The Netherlands; phone: (þ31) (20) 4853757;
fax: (þ31) (20) 4853432; e-mail: nlinfo-f@elsevier.com
E-mail: k.barton@elsevier.com
Tokyo: Elsevier, Customer Service Department, 4F Higashi-Azabu, 1-Chome Bldg, 1-9-15 Higashi-Azabu, Minato-ku, Tokyo 106-0044,
Japan; phone: (þ81) (3) 5561 5037; fax: (þ81) (3) 5561 5047; e-mail: jp.info@elsevier.com
Singapore: Elsevier, Customer Service Department, 3 Killiney Road, #08-01 Winsland House I, Singapore 239519; phone: (þ65)
63490222; fax: (þ65) 67331510; e-mail: asiainfo@elsevier.com

doi:10.1016/S0168-1702(04)00146-7

| |
 Viral Gene Silencing by RNA
Interference

Guest Editors
Susana Lopez
&
Carlos F. Arias
(Department of Developmental Genetics and Molecular Physiology,
Instituto de Biotechnologia, Universidad Nacional Autonoma de Mexico,
Avenida Universidad 2001,Colonia Chamilpa,
Cuernavaca, Morelos, 62210, Mexico)

Cover Image: Cryoelectron micrograph of spikeless rotavirus particles obtained by silencing

the spike protein VP4. The red asterisks indicate where the spikes should have
been.
[Courtesy of Dr. B.V.V. Prasad. Baylor College of Medicine, Houston, Texas].

doi:10.1016/S0168-1702(04)00155-8