Review

Special Issue: Translational Cell Biology

HIV trafficking in host cells: motors

wanted!
Raphael Gaudin1,2*, Bruna Cunha de Alencar1,2,
Nathalie Arhel3, and Philippe Benaroch1,2
1

Institut Curie, Centre de Recherche, 26 rue dUlm, 75248 Paris Cedex 05, France
INSERM, U932, 26 rue dUlm, 75248 Paris Cedex 05, France
3
INSERM, U941, Hopital Saint Louis, 1 avenue Claude Vellefaux, 75010 Paris, France
2

Throughout the viral replication cycle, viral proteins,

complexes, and particles need to be transported within
host cells. These transport events are dependent on the
host cell cytoskeleton and molecular motors. However,
the mechanisms by which virus is trafficked along cytoskeleton filaments and how molecular motors are
recruited and regulated to guarantee successful integration of the viral genome and production of new viruses
has only recently begun to be understood. Recent studies on HIV have identified specific molecular motors
involved in the trafficking of these viral particles. Here
we review recent literature on the transport of HIV
components in the cell, provide evidence for the identity
and role of molecular motors in this process, and highlight how these trafficking events may be related to
those occurring with other viruses.
Viral hijacking of cytoskeletal transport systems
The cytoplasm of eukaryotic cells is a complex and molecularly crowded environment [1] in which proteins move
primarily by diffusion. In the case of larger macromolecules and organelles, or on the need to translocate from one
compartment to another, proteins can attach to the cytoskeletal network and undergo active directed movement
along filaments. Viruses fit both requirements for active
cytoplasmic transport because their components are often
too large to diffuse freely within the cytosol and they need
to reach their site of replication rapidly. As such, viruses
operate as molecular imposters that use well-established
cellular processes [28]. In particular, viruses that replicate in the nucleus, such as adenoviruses [9,10] parvoviruses [11], retroviruses [12,13], and herpes viruses
[14,15], use the microtubule network and associated
motors to traffic within the cytoplasm either from the cell
Corresponding authors: Arhel, N. (nathalie.arhel@inserm.fr);
Benaroch, P. (benaroch@curie.fr).
Keywords: cytoskeleton; microtubule; retrovirus; actin; kinesin; dynein; myosin.
*
Current address: Department of Cell Biology, Harvard Medical School and Program in Cellular and Molecular Medicine at Childrens Hospital, 200 Longwood
Avenue, Boston, MA 02215, USA.
0962-8924/$ see front matter
2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tcb.2013.09.004

652

Trends in Cell Biology, December 2013, Vol. 23, No. 12

periphery to the center of the cell (dynein) or in the reverse

direction (kinesins) (Box 1). Motility on actin filaments
with the involvement of actin-driven motors (myosins)
has also been described for retroviruses [13,16] and other
viruses [17,18]. To date, much of the data surrounding the
role of molecular motors during retroviral trafficking has
been understood through studies of the lentivirus HIV.
Both the early and late phases of HIV viral replication
are highly regulated events likely to require different
molecular motors and numerous actin- and tubulin-associated proteins [19] (Table 1) at various stages (Figure 1)
[20]. During the early phase, incoming virions (Box 2) fuse
with the host cell membrane, allowing the viral complex
referred to as reverse transcription complex (RTC) to be
transported through the cytoplasm toward the nucleus. On
successful conversion of the viral RNA into DNA, the
complex, then referred to as a pre-integration complex
(PIC) and devoid of capsid, enters the nucleus where it
integrates its genome into host-cell DNA. During the late
phase, the transcribed viral RNA is exported to the cytoplasm where it drives the synthesis of new viral proteins.
These viral components are then transported in a coordinated fashion to the site of virus assembly leading to the
release of newly formed virions.
With new technologies including gene editing, improved
microscopy techniques, and high-throughput screening,
the identification of molecular motors involved in the
retroviral cycle and their characterization is achievable.
This review discusses our current understanding of retroviral cytoplasmic trafficking during host cell infection, with
an emphasis on HIV trafficking, and the questions remaining to be addressed. We also discuss HIV interaction with
the cytoskeleton and associated motors in both the early
and late phases of the replication cycle and compare this
with other retroviruses or other virus families for which
data are available.
HIV translocation: from plasma membrane to nuclear
pore
The hallmark of all retroviral infections is integration of
the viral genome into host-cell chromatin. For most retroviruses, such as murine leukemia virus (MLV) or human
foamy virus (HFV), this requires nuclear envelope breakdown during prometaphase. By contrast, lentiviruses
are actively imported to the nucleus via nuclear pore

Review
Box 1. Cytoskeleton and motors
The cytoskeleton is a dynamic assembly of structures that regulates
cell shape and movement, chromosomal segregation, cell division,
and intracellular trafficking of molecular complexes and organelles. It
comprises three types of network: microtubules, intermediary
filaments, and actin filaments, also called microfilaments. Microtubules are hollow tubes measuring 25 nm in diameter and several
microns in length, comprising 1315 protofilaments of a/b tubulin
dimers arranged in parallel [93]. They undergo continuous assembly
and disassembly, a process known as treadmilling, with the growing
plus end oriented toward the cell periphery and the minus end toward
the MTOC. Microfilaments are thinner filaments (7 nm diameter)
comprising two parallel strands of actin monomers [94]. They are
found throughout the cell but are particularly abundant beneath the
plasma membrane, where they form an actin cortex. Similarly to
microtubules, microfilaments are polar dynamic filaments that can
ensure directed transport. By contrast, intermediate filaments are
essentially involved in cell shape and movement and are generally
not involved in long-range movement of cargos.
Myosins belong to a superfamily of mechanoenzymes sharing a
conserved motor domain that is responsible for actin binding and
for power force production through ATP hydrolysis and conformational change [95]. The myosin tail domain mediates interaction
with cargos and, in certain cases, other myosin motors. Myosins are
involved in the regulation of the actin cytoskeleton dynamics, in the
control of cell shape and movement, and in the transport of many
different cargos including organelles.
Kinesins belong to a superfamily that shares a motor domain able
to bind microtubules in an ATP-dependent manner [95,96]. Typically,
active kinesins also possess a coiled-coil domain that mediates homo/
heterodimerization and enables their association with other regulatory subunits. Most kinesins move towards the plus end of microtubules, although a few kinesins move in the opposite direction. They
possess divergent tail domains with various cargo-binding specificities. It allows them to transport particular cargos such as proteins,
protein complexes, vesicles, or organelles.
Cytoplasmic dynein is a very complex molecular motor comprising two heavy chains, two intermediate chains, and several light
and light intermediate chains [95,96]. This motor typically functions in conjunction with the dynactin complex, which comprises at
least ten additional proteins and participates in motor processivity
and cargo binding. Cytoplasmic dynein is responsible for the
transport of many cellular cargos towards the minus end of
microtubules.
Many cellular cargos exhibit bidirectional movement on microtubules due to their simultaneous attachment to dyneins and
kinesins. Thus, opposite-polarity motors can be engaged in a tugof-war (Figure 1, white arrows) [28]. These opposite-polarity motor
activities are coordinated, resulting in proper spatial and temporal
sorting and in the establishment and maintenance of the subcellular
organization.

complexes (NPCs), which allows them to infect both dividing and non-dividing cells [21]. Because the nuclear entry
site differs between lentiviruses such as HIV and most
other retroviruses, it is likely that they also employ different trafficking pathways toward this site. Therefore, although our understanding of HIV trafficking can be
strengthened by studies of other retroviruses, similar pathways and mechanisms cannot necessarily be inferred.
After cell entry, evidence suggests that microtubuledirected trafficking propels HIV RTCs toward the nucleus;
single-particle tracking of incoming HIV revealed rapid
and directed movement over several microns with burst
velocities of the order of 1 mm/s [12,13]. In addition, treatment with nocodazole, which induces the disruption of the
microtubule network, leads to the accumulation of viral
complexes at the cell periphery and hinders infection
[13,22]. With regard to MLV and HFV, incoming viruses

Trends in Cell Biology December 2013, Vol. 23, No. 12

traffic to and accumulate at the Microtubule-Organizing

Center (MTOC) until nuclear envelope breakdown at the
onset of mitosis [2325]. Because chromosomes interact
with the mitotic spindle during chromatid separation,
accumulation at the MTOC may be an efficient mechanism
allowing viral genomes to quickly reach chromosomes on
nuclear envelope breakdown. As reported for HFV, retroviral Gag proteins may further assist both tethering of the
viral DNA to the host cell chromatin and integration [26]. By
contrast, lentiviruses access the nucleus via NPCs and,
although RTC accumulation at the MTOC has been reported
[12,13,27], the physiological relevance of this accumulation
is unclear because it may prevent or delay nuclear entry by
localizing viruses away from NPCs [13,27].
To properly engage and traffic along microtubules, some
retroviruses probably rely on molecular motors of both the
dynein and kinesin families. Single-particle tracking in
living cells exposed to fluorescent viruses revealed bidirectional microtubular displacement, characterized by a series of saltatory retrograde and anterograde movements
with overall directionality toward the nuclear compartment [13]. Retroviruses could therefore bind to molecular
motors of opposite polarity sequentially or simultaneously
as has been proposed for herpes simplex virus type 1 (HSV1) [15], and undergo a tug-of-war phenomenon [28] with
overall movement toward the center of the cell (Box 1 and
Figure 2). The minus-ended motor dynein is thought to
account for HIV and HFV movement toward the nucleus
because treatment with a dominant-negative inhibitor of
the p150 CC1 domain of dynactin (which uncouples dyneinbased transport) or microinjection of antidynein antibodies
led to clustering of viruses in the cell periphery [12,13,23].
Furthermore, a direct interaction between HFV Gag and
LC8, the light chain of dynein, has been demonstrated [23].
However, recent evidence suggests that interaction with
LC8 might be involved in other cellular processes and thus
is not specific for retrograde transport [29]. Although
microfilaments are also thought to be involved in trafficking incoming HIV [13,30], no functional interaction with
myosin motors has been demonstrated. The difficulty in
identifying direct HIVmotor interactions that specifically
mediate retrograde movement could be due to the size and
complexity of the motors, which can comprise several
heavy and light chains and work in conjunction with other
proteins. It may also be due to the functional redundancy of
family members (particularly for kinesins) or to the fragile
nature of the HIV capsid, which renders interaction
approaches complex [31].
The precise composition of the HIV RTC that interacts
with the cytoskeleton during the early phases of infection
has been a source of debate. Other viruses that traffic to the
nucleus interact with the cytoskeleton and/or with molecular motors via their structural shell. For example, the
adenovirus capsid hexon interacts with cytoplasmic dynein
intermediate and light intermediate chains [32], the herpes virus inner tegument protein VP1/2 recruits the dynein/dynactin motor complex [15,33], and HFV Gag
interacts with dynein light chain 8 [23]. In the case of
HIV, however, the field has been divided regarding the
structure and composition of the trafficking complex. Some
support the view that it no longer contains viral capsid
653

Review

Trends in Cell Biology December 2013, Vol. 23, No. 12

Table 1. Cytoskeletal proteins identified by genome-wide RNAi screens potentially involved in the HIV-1 cycle
Abbreviation
Cellular
Microtubule-related proteins
Tubulin
TUBAL3
Tubulin
TUBA8 (TUBAL2)
Tubulin
TUBA1B
Tubulin
TUBA1C (TUBA6)
Dynein,
DNAL1

component

Uniprot ID

Function (adapted from Uniprot and NCBI)

Refs

alpha chain-like 3
alpha-8 chain
alpha-1B chain
alpha-1C chain
axonemal, light chain 1

A6NHL2
Q9NY65
P68363
Q9BQE3
Q4LDG9

Tubulin is the major constituent of microtubules. It binds

two moles of GTP, one at an exchangeable site on the beta
chain and one at a non-exchangeable site on the alpha
chain.

[19]
[55]
[97]
[97]
[19]

KIF3A

Kinesin-like protein KIF3A

Q9Y496

KIF3C

Kinesin family member 3C

O14782

KIF2B

Kinesin-like protein KIF2B

Q8N4N8

KIF17

Kinesin-like protein KIF17

Q9P2E2

MAP4

Microtubule-associated protein
4
Pleckstrin homology domain
containing, family A
(phosphoinositide bindingspecific) member 3
Spastin
Mid1-interacting protein 1

P27816

PLEKHA3 (FAPP1)

Q9HB20

Acts as one of several non-catalytic accessory components

of the cytoplasmic dynein 1 complex that are thought to be
involved in linking dynein to cargos and to adapter proteins
that regulate dynein function. Cytoplasmic dynein 1 acts as
a motor for the intracellular retrograde motility of vesicles
and organelles along microtubules. DNAL1 is cilium
specific.
Microtubule-based anterograde translocator for
membranous organelles. Plus end-directed microtubulesliding activity in vitro. Plays a role in primary cilium
formation. Heterodimer with KIF3B or KIF3C.
Microtubule-based anterograde movement. Heterodimer of
KIF3A and KIF3C.
Plus end-directed microtubule-dependent motor required
for spindle assembly and chromosome movement. Has
microtubule depolymerization activity.
Plus end-directed microtubule-dependent motor. KIF3Related Motor Protein.
Non-neuronal microtubule-associated protein. Promotes
microtubule assembly.
Involved in Golgi-to-cell surface membrane traffic. Induces
membrane tubulation. Binds preferentially to
phosphatidylinositol 4-phosphate (PtdIns4P).

[98]

[19]
[55]

[55]
[19,98]
[19]

Q9UBP0
Q9NPA3

ATP-dependent microtubule-severing protein.

Involved in stabilization of microtubules.

[19]
[98]

Actin-like protein 6B

O94805

[97]

MYL6
MYO1F

Myosin light chain 6, alkali

Unconventional myosin-1F

P60660
O00160

MYOM1

Myomesin 1 (skelemin)

P52179

MSN

Moesin

P26038

ANKRD30A
SOWAHA (ANKRD43)

Ankyrin repeat domain 30A

Ankyrin repeat domaincontaining protein SOWAHA
Ankyrin repeat domaincontaining protein 6
Ankyrin repeat domaincontaining protein 9
Ankyrin repeat and FYVE
domain-containing protein 1
Ankyrin repeat domain 1
Spectrin, alpha, nonerythrocytic 1 (alpha-fodrin)
Spectrin beta chain, nonerythrocytic 1 (fodrin beta chain)

Q9BXX3
Q2M3V2

Belongs to the chromatin remodeling brain-specific BAF

(bBAF) complex; as such, plays a role in remodeling
mononucleosomes in an ATP-dependent fashion.
Regulatory light chain of myosin.
Myosins are actin-based motor molecules with ATPase
activity. Unconventional myosins serve in intracellular
movements.
Major component of the vertebrate myofibrillar M band.
Binds myosin, titin, and light meromyosin.
Member of the ERM family. ERM proteins function as
crosslinkers between plasma membranes and actin-based
cytoskeletons. Moesin is localized to filopodia and other
membranous protrusions.
Ankyrins link the integral membrane proteins to the
underlying spectrinactin cytoskeleton. They play key roles
in activities such as cell motility, activation, proliferation,
and contact and the maintenance of specialized membrane
domains. Most ankyrins typically comprise three structural
domains: an amino-terminal domain containing multiple
ankyrin repeats; a central region with a highly conserved
spectrin-binding domain; and a carboxy-terminal regulatory
domain that is the least conserved and subject to variation.

SPAST (ADPSP, FSP2)

MID1IP1
Actin-related proteins
ACTL6B

ANKRD6
ANKRD9
ANKFY1
ANKRD1
SPTAN1
SPTBN1

Q9Y2G4
Q96BM1
Q9P2R3
Q05DQ9
Q13813
Q01082

VCL

Vinculin

P18206

TAGLN2

Transgelin-2 (SM22 alpha

homolog)

P37802

654

Spectrin is the major constituent of the cytoskeletal network

underlying the plasma membrane. It associates with actin to
form the cytoskeletal superstructure of the plasma
membrane. Comprises nonhomologous chains, alpha and
beta, that aggregate to form dimers, tetramers, and higher
polymers. Calcium-dependent movement of the
cytoskeleton at the membrane. Interacts with ankyrin.
Actin filament (F-actin)-binding protein involved in cell
matrix and cellcell adhesion.
SM22 alpha is a shape-change sensitive actin crosslinking/
gelling protein.

[97]
[98]

[97]
[97,98]

[19]
[19]
[19]
[19]
[55]
[55]
[19]
[19]

[55]
[98]

Review

Trends in Cell Biology December 2013, Vol. 23, No. 12

Table 1 (Continued )
Abbreviation
TWF1 (PTK9)

Cellular component
Twinfilin-1

Uniprot ID
Q12792

ARPC1A (SOP2L)

Q92747

CDC42EP3 (BORG2)

Actin-related protein 2/3

complex subunit 1A
CDC42 Effector Protein 3

NF2

Merlin (neurofibromin-2)

P35240

FHL3

Four and a half LIM domains 3

Q13643

FLII

Flightless I homolog
(Drosophila)

Kelch domain-containing
protein 2
Kelch-like protein 1
KLHL1
Intermediate filament-related proteins
Keratin, type I cytoskeletal 17
KRT17
KLHDC2

KRT4

Keratin 4

Q9UKI2

Q9Y2U9

Function (adapted from Uniprot and NCBI)

Actin-binding protein involved in motility and
morphological processes. Inhibits actin polymerization,
probably by sequestering G-actin. By capping the barbed
ends of filaments, it also regulates motility.
Component of the Arp2/3 complex that is involved in
regulation of actin polymerization.
Involved in organization of the actin cytoskeleton. May act
downstream of CDC42 to induce actin filament assembly
leading to cell-shape changes. Induces pseudopodium
formation in fibroblasts.
May act as a membrane-stabilizing protein. Regulator of the
Hippo/Sav/Wts/Hpo (SWH) signaling pathway. Strong
staining in ruffling membranes and filopodia.
Actin-binding protein that regulates alpha-actinin-mediated
actin bundling. Localizes to actin stress fibers and enhances
cell spreading and stress fiber disassembly.
Regulation of cytoskeletal rearrangements involved in
cytokinesis and cell migration, by inhibiting Rac1dependent paxillin phosphorylation.
Actin-binding site. May play a role in organizing the actin
cytoskeleton.

Q9NR64
Q04695

Q6PIN2

Refs
[55]

[19]
[98]

[19]

[19]

[19]

[19]
[19]

Type I intermediate filament chain keratin 17, expressed in

nail bed, hair follicle, sebaceous glands, and other
epidermal appendages.
Type II cytokeratins coexpressed during differentiation of
simple and stratified epithelial tissues. This type II
cytokeratin is specifically expressed in differentiated layers
of the mucosal and esophageal epithelia.

[97]

[97]

NCBI, National Center for Biotechnology Information.

protein, others suggest that the capsid remains intact for

the duration of transport to the nuclear pore, whereas
others propose that it undergoes a series of transformations throughout transport [31]. Studies may have been
confounded by the fragile nature of the HIV-1 capsid
[34,35] and the fact that not all intracellular viral complexes lead to productive infection [36]. Furthermore, some
confusion might also be due to the use of physiologically
irrelevant model cell systems. Nevertheless, recent evidence shows that the HIV capsid mediates interaction with
the nuclear pore and its machinery [3739], suggesting
that the viral capsid is maintained during trafficking
toward the nucleus.
Although sufficient evidence supports a role for microtubules directing HIV to the NPC, it is unclear how HIV is
transferred from microtubules to NPCs. The proximity of
the centrosome to the nuclear envelope could be sufficient
for cargos to detach from microtubules and diffuse to the
nearest NPC, which would imply cargo translocation
through NPCs located closest to centrosomes. However,
this hypothesis seems unlikely because analysis by in situ
hybridization electron microscopy revealed homogenous
HIV-1 genome integration throughout euchromatin [40].
Alternatively, after retrograde transport to the MTOC,
lentiviruses could adopt kinesin-dependent anterograde
movement over a short distance to NPCs, as has been
suggested for herpes virus [15]. However, this may also
be unlikely because physical connections between microtubules and the nuclear envelope occur away from NPCs,
probably to avoid mechanical deformation of NPCs [41].

Despite these unlikely hypotheses, there is good evidence to suggest that the transfer of incoming HIV-1 from
microtubules to NPCs may occur via perinuclear actin
filaments, also called microfilaments (Box 1). Single-particle tracking of HIV-1 revealed directed movement close to
the nucleus before docking at the nuclear envelope at a
slow velocity in a range of myosin motors [13]. Moreover,
colocalization of HIV-1 with microfilaments was observed
and depolymerization of actin networks shortly after infection led to accumulation of RTCs close to the nuclear
compartment but not at the nuclear envelope [13]. Furthermore, it appears that, in some CD4+ T cell lines,
trafficking of incoming HIV-1 might occur exclusively on
microfilaments [30]. The transfer from microtubules to
microfilaments may be mediated via microtubule-associated proteins (MAPs) or motor proteins that possess both
actin- and microtubule-binding motifs [42], whereas transfer from microfilaments to NPCs may occur via nuclear
pore cytoplasmic filaments, which interact with microfilaments near the nucleus [43]. To reach the host-cell genome
efficiently, gammaretroviruses and spumaviruses rely on
microtubule-directed trafficking and MTOC accumulation,
whereas lentiviruses may rely initially on microtubuledirected trafficking but reach NPCs via trafficking along
microfilaments in the proximity of the nucleus.
In conclusion, HIV translocation from the plasma membrane to the nuclear membrane involves both microtubules
and microfilaments near the nucleus. The movement properties of HIV and the use of dominant-negative inhibitors
suggest the involvement of a dynein motor, but a direct
655

Review

Trends in Cell Biology December 2013, Vol. 23, No. 12

Kinesin

Myosin

Dynein

ARN
Gag

Tug-of-war

TRENDS in Cell Biology

Figure 1. Schematic view of microtubules, actin filaments, and molecular motors with trafficking HIV components. The various types of motor carrying their cargo are
represented. A kinesin is transporting two molecules of Gag associated with a viral RNA toward the plus end of a microtubule. Two dyneins associated with a viral capsid
are running toward the minus end of the microtubule, while two myosins are moving on a microfilament carrying an incoming viral reverse transcription complex (RTC).
The tug-of-war concept is illustrated with a Gag molecule simultaneously binding a kinesin and a dynein pulling in opposite direction.

interaction remains to be demonstrated. A better understanding of the structure and composition of the trafficking
RTC will no doubt provide a step forward in this endeavor.
HIV synthesis and egress: from nuclear pore to plasma
membrane
On synthesis, different components of the viral particle
reach the virus assembly site in a time- and space-coordinated manner. The site of HIV assembly is defined as a lipid
bilayer specialized platform. Here, newly synthesized Gag,
viral RNAs (vRNAs), and Env are transported, with the help
of cellular machineries, over long intracellular distances (in
the few-micron range) to the viral assembly site. Such a
journey has long been suspected to be cytoskeleton driven,
but discrepancies remain. Below we review the evidence
supporting the involvement of filaments and associated
motors in the late phases of the HIV replication cycle.
Role of the microtubule network and associated motors
The microtubule cytoskeleton regulates many cellular processes including intracellular trafficking of molecular complexes and organelles. It is a useful network for viruses
because it is generally involved in long-range movement of
cargos. Gag, which is the major structural protein of HIV,
is synthesized in the cytosol as a 55-kD precursor and
transported from the cytoplasm to the plasma membrane,
where it orchestrates HIV assembly. Soon after synthesis,
the N terminus of Gag is myristoylated, allowing the Gag
precursor to be anchored to the inner leaflet of the plasma
membrane, where it oligomerizes [44]. Interestingly, Gag
656

expression alone is sufficient for the assembly, budding,

and release of Virus-Like Particles (VLPs) [45], which
constitute a useful tool for studying Gag trafficking. Nevertheless, this minimal system may diverge from full HIV
component trafficking.
It is unclear how newly synthesized Gag is transported
to the site of viral assembly. Movement of Gag along
microtubules is supported by studies showing colocalization of HIV-1 Gag with tubulin and its copurification with
microtubules in COS-1 cells [46]. Moreover, in three of four
genome-wide small interfering RNA (siRNA) screens performed on various HIV-infected HeLa, 293T, or Jurkat
cells, at least one tubulin isoform was found to be important for efficient replication of HIV-1 (Table 1). Studies that
dispute the role of microtubules in Gag trafficking [47,48]
may be accounted for by the fact that Gag localization was
examined in the absence of a full infection or in cell lines
rather than primary target cells.
Regardless of the precise route that HIV Gag takes to
reach the plasma membrane, both MAPs and molecular
motors have been implicated in efficient Gag trafficking to
the plasma membrane, thus substantiating the involvement of the microtubular network. Host proteins such as
SOCS-1 can associate with and allow Gag transport along
microtubules. SOCS-1 expression is induced by HIV infection, colocalizes with HIV-1 Gag along microtubules, and
regulates the intracellular trafficking and stability of Gag
during HIV-1 infection through an unknown mechanism
[46,49,50]. In addition to MAPs, multiple lines of evidence
suggest a role for molecular motors during Gag egress. The

Review
Box 2. Viruses and viral proteins that undergo trafficking
Viral particles, referred to as virions, typically comprise a nucleic
acid (DNA or RNA) protected by a protein shell (a capsid). In some
viruses, the capsid is surrounded by another protein layer (matrix or
tegument), and a lipid membrane called an envelope.
Retroviruses are enveloped positive RNA viruses whose genome
is reverse transcribed into double-stranded DNA by viral reverse
transcriptase during translocation to the nucleus. The hallmark of
retroviruses is the integration of their viral genome into host-cell
chromatin [99]. All retroviruses contain three essential genes Gag,
Pol, and Env which encode polyprotein precursors that are
processed by proteolysis during late phases of the replication cycle
in the Golgi apparatus (Env) or after budding (Gag and GagPol).
Gag encodes structural proteins (matrix, capsid, and nucleocapsid);
Pol encodes viral enzymes (protease, reverse transcriptase, and
integrase); and Env encodes the surface and transmembrane
glycoproteins of the virion. For simple retroviruses such as MLV,
these are the only genes. Complex retroviruses, which include
HTLV, lentiviruses such as HIV and SIV, and spumaviruses such as
HFV, also contain accessory genes that regulate viral gene expression and replication. HIV encodes two regulatory proteins (Tat and
Rev) and four accessory proteins (Vpr, Vpu/Vpx, Vif, Nef). All
retroviruses undergo cytoplasmic trafficking to and from the
nucleus in their target cell.
Herpes viruses are a large family of enveloped DNA viruses of
which HSV-1 has been the most extensively studied. The virus
genome is very large and encodes nearly 100 transcripts. Adenoviruses are non-enveloped DNA viruses whose icosahedral capsid
comprises several hundred hexons and 12 pentons. The core of the
particle comprises two major proteins (polypeptides V and VII).
The viral replication cycle refers to the successive steps from entry
into the target cell, to genomic replication, to formation of new
particles and transmission to new cells. The virion enters the cell by
either fusing with the cell membrane or being engulfed in the
cytoplasm by endocytosis. All retroviruses discussed in this review
traffic to the nucleus, where their genome is imported and
replicated. Viral mRNA is then transported back into the cytoplasm
where it is translated into viral proteins. Ultimately, viral components or virions traffic back to the cell membrane and are released
into the extracellular environment to spread to other cells. For some
retroviruses, the infection spreads to new cells directly via cell-tocell contacts at a focal point named the virological synapse.

chromokinesin KIF4 directly interacts with the Gag precursor of MLV, MasonPfizer monkey virus, simian immunodeficiency virus (SIV), and HIV-1 [51,52] and regulates
intracellular trafficking and stabilization of the HIV-1 Gag
precursor [53]. Indeed, KIF4 silencing or expression of a
dominant-negative form of KIF4 reduces intracellular
levels of Gag. However, as a chromokinesin, KIF4 is mainly
found in the nucleus [54]. Therefore, further analysis is
needed to characterize the precise role of KIF4 in the
different steps of Gag trafficking.
Interestingly, whereas disrupting kinesins decreases
HIV particle release [19,53,5557], silencing expression
of the dynein heavy chain, as well as overexpression of
the dynactin subunit p50/dynamitin, led to a threefold
increase in HIV-1 release [58]. The role of molecular motors
of opposite direction suggests that Gag could undergo a
tug-of-war (Box 1 and Figure 1) by kinesin and dynein
motors [28] during egress and that retrograde trafficking
could counteract efficient HIV-1 release. Unknown specific
stimuli, such as the formation of virological synapse (Box 2)
or cytokine signaling, would be required to tip the scale in
one direction. Further work will be needed to define the
stage and the molecular state at which Gag may interact
with the different cytoskeleton components. Again, the cell

Trends in Cell Biology December 2013, Vol. 23, No. 12

model used to address those questions will be crucial

because it may lead to different conclusions. Along these
lines, the particular case of infected macrophages that
accumulate newly formed virions in intracellular compartments called virus-containing compartments (VCCs) is of
interest. A recent study [57] has identified kinesins required for the proper transport and release of VCCs
(Figure 2 and Box 3).
Whereas Gag trafficking has been well studied, little is
known regarding the trafficking of the other viral components. After transcription, vRNA reaches the site of viral
assembly in association with Gag. This interaction occurs
near the nucleus shortly after Gag synthesis and promotes
the dimerization of Gag [59,60]. According to the current
model, the HIV-1 GagvRNA interaction stabilizes Gag at
the plasma membrane, allowing higher-order Gag multimerization and viral budding [6062]. Whether formation
of cytosolic low-order complexes is required for interactions
with molecular motors and transport remains to be established but represents a likely hypothesis because structural conformational change, phosphorylation, or other
physical modifications could constitute a signal for motor
recruitment. Although other viral components are believed
to associate with molecular motors, none has yet been
associated with the transport of retrovirus envelopes
(Env). Env is a transmembrane glycoprotein synthesized
as a precursor in the endoplasmic reticulum and exported
via the secretory pathway to the plasma membrane. The
precise intracellular route followed by Env proteins
remains to be established and may involve passage
through secretory granules in the case of HIV-1-infected
T cells [63]. Several molecular motors have been associated
with the trafficking of endosomes from the secretory pathway [64], but none has been associated with Env trafficking. One possibility to be explored is that Env, while still in
endosomes, could hijack molecular motors through its long
cytoplasmic tail to traffic to the plasma membrane. Thus,
identification of such molecular motors could lead to the
development of specific inhibitors preventing proper Env
incorporation into virions and thus loss of infectivity of the
released viruses.
Although still incomplete, the current view emerging
from the literature indicates that the microtubule network
and associated motors are involved in the intracellular
transport of viral and cellular components of the future
virion.
Role of the actin network
Beside microtubules, actin filaments represent the other
important network of the cytoskeleton for intracellular
transport of proteins and organelles. Several early studies
have shown that Gag directly interacts with actin [6567]
and that such an interaction, together with the myosin
light chain, is required during HIV-1 late phases. Other
HIV components (Box 2), such as reverse transcriptase
[68], Nef [69], and nucleocapsid [66], also interact with
actin at several time points during the replication cycle. In
addition, actin plays an important role during the export of
vRNA from the nucleus to the cytosol, because chemical
inhibition of actin bundle formation leads to HIV-1 Gag
mRNA accumulation in the nucleus [70]. This retention
657

Review

Trends in Cell Biology December 2013, Vol. 23, No. 12

Acn
Microtubules
Acn
cortex

Plasma
membrane

Tran

slo

Acn

Nucleus

on
ca

Microtubules
Egress

Acn
cortex

Plasma
membrane

Mature virus
Dynein
Envelope glycoprotein
Kinesin

Capsid

Genomic viral RNA

VCC

Myosin

Reverse transcripon

Nuclear pore

Acn

P55 gag

MTOC

Microtubule
TRENDS in Cell Biology

Figure 2. Schematic representation of viral trafficking during the replication cycle of HIV-1. The HIV replication cycle can be divided into early and late phases, which are
represented to the left and right of the schematic cell, respectively. During the early phase, viruses enter target cells by receptor-mediated fusion, cross the actin cortex
underlying the plasma membrane, and release their viral core into the cytosol. HIV reverse transcription complexes (RTCs) are propelled along microtubules toward the nucleus
in a movement that is thought to involve dynein motors. During transport, the viral RNA (vRNA) genome is reverse transcribed into DNA. The literature also reports directed
movement of HIV-1 on microfilaments of actin, which could mediate the transition between microtubules and nuclear pores. On completion of reverse transcription, the viral
pre-integration complex (PIC) comprising viral DNA and associated proteins is imported to the nucleus via the nuclear pore, where the viral genome either is integrated into host
chromosomes or circularizes. During the late phase, vRNA is synthesized in the nucleus and exported into the cytosol, where viral protein synthesis occurs. The viral envelope is
synthesized in the endoplasmic reticulum and transported to the Golgi apparatus where it is cleaved before reaching the plasma membrane. Through interaction with kinesin
KIF4, Gag might be transported along microtubules toward the cell periphery. Gag associates and forms complexes with vRNA present in the cytosol. Particle assembly is driven
by Gag oligomerization at the plasma membrane and within virus-containing compartments (VCCs) in macrophages (below on the cell).

could be linked to the interaction of actin with the nucleocytoplasmic filaments of NPCs, which is critical for CRM1dependant export of HIV-1 Rev protein [71]. This nucleoskeleton comprises over ten different myosins and kinesins
(for a review, see [72]) but has not been studied in detail
during the course of a retroviral infection. Cell-free HIV-1
658

particles contain, in addition to actin, many actin-related

proteins such as ezrin, moesin, cofilin, myosins, profilin-1,
vimentin, filamin A, Arp2/3, talin, cdc42, and ankyrin [73
78], which is indicative of tight regulation of actin organization during retroviral budding. One likely hypothesis is
that cortical actin represents a physical barrier that needs

Review

Trends in Cell Biology December 2013, Vol. 23, No. 12

Box 3. The enigmatic VCC of macrophages

+

HIV-1 has two main cell targets: CD4 T cells and macrophages. HIV1-infected macrophages have been found in many tissues from
seropositive patients. Although HIV buds primarily at the plasma
membrane of CD4+ T cells and most cell lines, early electron
microscopy studies established that infected macrophages accumulate a large number of infectious viral particles within intracellular compartments [100]. Budding profiles observed at the limiting
membrane of these VCCs strongly suggested that VCCs are the site
of viral assembly in macrophages (Figure 2, inset) [101103]. VCCs
were initially considered part of the endocytic pathway because they
share morphological and molecular similarities with multivesicular
bodies (late endosomes) [102,102]. However, further studies established that they lack important characteristics of endosomes
[104,105] including lysosomal markers [104]. Moreover, VCCs
exhibit a near-neutral luminal pH and do not progress toward
fusion with lysosomes [106]. VCCs may originate from the
sequestration of areas of the plasma membrane rich in tetraspanins
and cholesterol [107,108]. Potentially because of their origin, they
appear connected to the extracellular medium through microchannels [107,108] that are generally considered too narrow to allow
passage of viral particles. Recent studies [109111] indicate that they
exist before infection. Whether these compartments are unique to
macrophages remains an open question. Regardless, VCCs may
protect newly formed virions from both exogenous and endogenous attacks and therefore contribute to the role of reservoirs
attributed to infected macrophages.
The mechanism allowing the delivery of these intracellular stocks
of virions to the extracellular milieu remains unclear. Recent results
indicate that transport of the VCC within macrophages relies on the
molecular motor KIF3A [57]. The limiting membrane of the VCC
was observed to be tightly associated with the microtubule
network and the kinesin KIF3A. Moreover, the kinesins KIF3A [57]
and, possibly, KIF3C (unpublished observation) are important for
HIV-1 release by primary macrophages, probably by driving the
motion of the VCCs toward the cell periphery for efficient viral
particle release. Two independent genome-wide screens performed on HIV-1-infected cell lines identified KIF3A [98] and KIF3C
[19] as cellular proteins potentially involved in the HIV-1 cycle
(Table 1), further indicating that kinesin-II plays an important role in
the HIV-1 cycle. Such a role might be cell-type dependent because
KIF3A is not involved in HIV-1 release from Jurkat T cells [57] but is
important for efficient VLP release from HeLa cells [56]. Finally,
several molecular motors probably control the motion of VCCs and
their spatial distribution, which is strictly dependent on the
integrity of the microtubule network.

to be loosened to allow budding to occur. By contrast, actin

polymerization could also represent an important driving
force required under membrane tension for the budding
process in a similar way, but topologically inverted, as seen
during endocytosis when clathrin-coated pits pinch off the
plasma membrane [79].
In addition to actin, several studies also point to a role
for myosins (Box 1) in the HIV-1 life cycle. Indeed, budding
of HIV-1 particles, but not transport of Gag to the plasma
membrane, is strongly affected by chemical inhibition of
the myosin light chain [73]. Moreover, many actin-related
proteins have been found in siRNA genome-wide screens
as important host proteins during the HIV-1 life cycle,
including three actin-related molecular motors (Table 1).
One, myosin-1F, is of particular interest because it is
primarily expressed in lymphoid and myeloid cells, including CD4+ T cells and the monocyte lineage (BioGPS) two
main targets of HIV. However, this myosin has not been
investigated in detail.
In conclusion, although it is clear that the actin cytoskeleton plays important roles at various stages of the

HIV-1 late cycle, assays that are more specific will be

necessary to study individual trafficking events and identify the specific myosins involved.
HIV cell-to-cell transmission
HIV cell-to-cell transmission occurs through specialized
virus-induced contact areas between infected cells and
target cells and is more efficient than cell-free virus infection [80]. Several transmission modes have been reported,
including virological synapses, cellular conduits, nanotubes, and biofilms, although virological synapses seem
to be more frequent [81].
The cytoskeleton is essential for virological synapse
formation, because the MTOC polarizes toward the synapse in HIV-1 or human T-lymphotropic virus (HTLV)infected T cells [8284]. Inhibition of actin and tubulin
polymerization impairs both Gag and Env accumulation at
the contact site and virus transmission to uninfected targets [85]. In addition, exposure of migrating lymphocytes to
a myosin light chain inhibitor impaired polarization of Gag
to uropods and virus transmission [86]. These data strongly suggest the involvement of molecular motors; however,
the exact identity of these specific motors remains elusive.
In addition to virological synapse formation, retrovirusinfected cells have been shown to connect to uninfected
targets via intercellular conduits [8791]. Actin filament
bundles form these intercellular conduits, along which viral
proteins/virus can slide en route to the uninfected cell. The
mechanisms that allow virus to access such conduits remain
unknown but are likely to involve actin-based motors.
Concluding remarks
Despite a large body of work, we remain far from elucidating the complexity of interactions between retroviruses,
the host cell cytoskeleton, and molecular motors.
Genome-wide screens of the HIV cycle have identified
numerous cytoskeleton-related proteins but few motors,
though it is clear that molecular motors are likely to be
involved in many steps of transport during the HIV cycle.
Several points can explain the failure to identify these
motors. First, genome-wide screens based on RNAi generate numerous false-negative results due to problems with
silencing efficiency, toxicity, and protein half-life. Second,
the original assays used for RNAi screening were primarily
based on final measurements of viral production in cells
transfected with siRNAs and infected. Therefore, only a
knock down resulting in blockade of the viral cycle without
affecting cell viability can generate a hit. Third, molecular
motors are numerous and may be essential or redundant;
therefore, single-gene silencing of the motors may not yield
a phenotype. Moreover, in the case of VCCs observed in
macrophages, indirect evidence [57] suggests that different
motors are recruited to ensure VCC motion (Box 3). In
addition, the structure and composition of the viral cargo,
such as the RTC, evolves during transport toward the
nucleus. The same holds true for Gag monomers, dimers,
or complexes with vRNA on their way to the assembly site.
These changes in cargo might be associated with the
recruitment of different motors, as shown for HSV-1 [15].
New genetic screens based on measurement of the
precise steps of viral transport rather than the entire viral
659

Review
cycle might be more fruitful in further identifying cytoskeleton-related proteins and motors. Each trafficking step of
the viral components deserves to be individually deciphered to identify the specific molecular motors involved.
On this road, the switch from model cell lines to primary
cells will represent an important milestone in validating
the screens and their results and searching for motors with
tissue specificity (T cells versus macrophages). From entry
to egress, we first need an inventory of the molecular
motors involved in the viral cycle. In a second step, integration of this information will represent an exciting challenge in understanding how HIV hijacks and regulates
these machineries for its own purpose.
In addition, characterization of the role of potential new
molecular players should benefit from dynamic imaging
techniques and fluorescent probes such as high-resolution
microscopy and high-speed time-lapse microscopy, which is
required to follow motors and their cargo in motion. The
viral cycle may also prove to be very different in vivo, where
the microenvironment may strongly differ from the in vitro
conditions of culture in terms of oxygen, nutrients, and
energy as well as the 3D matrix and cell surroundings.
Development and usage of in vivo models of retroviral
infection such as the recently developed humanized mouse
models (e.g., [92]) should help to define new mechanisms of
viral trafficking.
Acknowledgments
This study was supported by grants from Ensemble contre le SIDA, the
DCbiol labex, and Agence Nationale de Recherche contre le SIDA to P.B.
and a fellowship to B.C.d.A.

References
1 Ellis, R.J. (2001) Macromolecular crowding: an important but
neglected aspect of the intracellular environment. Curr. Opin.
Struct. Biol. 11, 114119
2 Brandenburg, B. and Zhuang, X. (2007) Virus trafficking learning
from single-virus tracking. Nat. Rev. Microbiol. 5, 197208
3 Greber, U.F. and Fassati, A. (2003) Nuclear import of viral DNA
genomes. Traffic 4, 136143
4 Greber, U.F. and Way, M. (2006) A superhighway to virus infection.
Cell 124, 741754
5 Marsh, M. and Helenius, A. (2006) Virus entry: open sesame. Cell 124,
729740
6 Dodding, M.P. and Way, M. (2011) Coupling viruses to dynein and
kinesin-1. EMBO J. 30, 35273539
7 Radtke, K. et al. (2006) Viral interactions with the cytoskeleton: a
hitchhikers guide to the cell. Cell. Microbiol. 8, 387400
8 Cohen, S. et al. (2011) How viruses access the nucleus. Biochim.
Biophys. Acta 1813, 16341645
9 Suomalainen, M. et al. (1999) Microtubule-dependent plus- and minus
end-directed motilities are competing processes for nuclear targeting
of adenovirus. J. Cell Biol. 144, 657672
10 Leopold, P.L. et al. (2000) Dynein- and microtubule-mediated
translocation of adenovirus serotype 5 occurs after endosomal lysis.
Hum. Gene Ther. 11, 151165
11 Xiao, P.J. and Samulski, R.J. (2012) Cytoplasmic trafficking,
endosomal escape, and perinuclear accumulation of adenoassociated virus type 2 particles are facilitated by microtubule
network. J. Virol. 86, 1046210473
12 McDonald, D. et al. (2002) Visualization of the intracellular behavior
of HIV in living cells. J. Cell Biol. 159, 441452
13 Arhel, N. et al. (2006) Quantitative four-dimensional tracking of
cytoplasmic and nuclear HIV-1 complexes. Nat. Methods 3, 817824
14 Wolfstein, A. et al. (2006) The inner tegument promotes herpes
simplex virus capsid motility along microtubules in vitro. Traffic 7,
227237

660

Trends in Cell Biology December 2013, Vol. 23, No. 12

15 Radtke, K. et al. (2010) Plus- and minus-end directed microtubule

motors bind simultaneously to herpes simplex virus capsids using
different inner tegument structures. PLoS Pathog. 6, e1000991
16 Lehmann, M.J. et al. (2005) Actin- and myosin-driven movement of
viruses along filopodia precedes their entry into cells. J. Cell Biol. 170,
317325
17 van Leeuwen, H. et al. (2002) Evidence of a role for nonmuscle myosin
II in herpes simplex virus type 1 egress. J. Virol. 76, 34713481
18 Sun, X. and Whittaker, G.R. (2007) Role of the actin cytoskeleton
during influenza virus internalization into polarized epithelial cells.
Cell. Microbiol. 9, 16721682
19 Brass, A.L. et al. (2008) Identification of host proteins required for HIV
infection through a functional genomic screen. Science 319, 921926
20 Naghavi, M.H. and Goff, S.P. (2007) Retroviral proteins that interact
with the host cell cytoskeleton. Curr. Opin. Immunol. 19, 402407
21 Suzuki, Y. and Craigie, R. (2007) The road to chromatin nuclear
entry of retroviruses. Nat. Rev. Microbiol. 5, 187196
22 Bukrinskaya, A. et al. (1998) Establishment of a functional human
immunodeficiency virus type 1 (HIV-1) reverse transcription complex
involves the cytoskeleton. J. Exp. Med. 188, 21132125
23 Petit, C. et al. (2003) Targeting of incoming retroviral Gag to the
centrosome involves a direct interaction with the dynein light chain 8.
J. Cell Sci. 116, 34333442
24 Saib, A. et al. (1997) Nuclear targeting of incoming human foamy virus
Gag proteins involves a centriolar step. J. Virol. 71, 11551161
25 Afonso, P.V. et al. (2007) Centrosome and retroviruses: the dangerous
liaisons. Retrovirology 4, 27
26 Tobaly-Tapiero, J. et al. (2008) Chromatin tethering of incoming
foamy virus by the structural Gag protein. Traffic 9, 17171727
27 Zamborlini, A. et al. (2007) Centrosomal pre-integration latency of
HIV-1 in quiescent cells. Retrovirology 4, 63
28 Hendricks, A.G. et al. (2010) Motor coordination via a tug-of-war
mechanism drives bidirectional vesicle transport. Curr. Biol. 20,
697702
29 Tan, G.S. et al. (2007) The dynein light chain 8 binding motif of rabies
virus phosphoprotein promotes efficient viral transcription. Proc.
Natl. Acad. Sci. U.S.A. 104, 72297234
30 Yoder, A. et al. (2011) Effects of microtubule modulators on HIV-1
infection of transformed and resting CD4 T cells. J. Virol. 85,
30203024
31 Arhel, N. (2010) Revisiting HIV-1 uncoating. Retrovirology 7, 96
32 Bremner, K.H. et al. (2009) Adenovirus transport via direct
interaction of cytoplasmic dynein with the viral capsid hexon
subunit. Cell Host Microbe 6, 523535
33 Zaichick, S.V. et al. (2013) The herpesvirus VP1/2 protein is an effector
of dynein-mediated capsid transport and neuroinvasion. Cell Host
Microbe 13, 193203
34 Borroto-Esoda, K. and Boone, L.R. (1991) Equine infectious anemia
virus and human immunodeficiency virus DNA synthesis in vitro:
characterization of the endogenous reverse transcriptase reaction. J.
Virol. 65, 19521959
35 Welker, R. et al. (2000) Biochemical and structural analysis of isolated
mature cores of human immunodeficiency virus type 1. J. Virol. 74,
11681177
36 Marechal, V. et al. (1998) Cytosolic Gag p24 as an index of productive
entry of human immunodeficiency virus type 1. J. Virol. 72,
22082212
37 Arhel, N.J. et al. (2007) HIV-1 DNA Flap formation promotes
uncoating of the pre-integration complex at the nuclear pore.
EMBO J. 26, 30253037
38 Krishnan, L. et al. (2010) The requirement for cellular transportin 3
(TNPO3 or TRN-SR2) during infection maps to human
immunodeficiency virus type 1 capsid and not integrase. J. Virol. 84,
397406 http://dx.doi.org/10.1128/JVI.01899-09
39 Di Nunzio, F. et al. (2012) Human nucleoporins promote HIV-1
docking at the nuclear pore, nuclear import and integration. PLoS
ONE 7, e46037
40 Arhel, N.J. et al. (2006) Wild-type and central DNA flap defective HIV1 lentiviral vector genomes: intracellular visualization at
ultrastructural resolution levels. Retrovirology 3, 38
41 Tzur, Y.B. et al. (2006) SUN-domain proteins: Velcro that links
the nucleoskeleton to the cytoskeleton. Nat. Rev. Mol. Cell Biol. 7,
782788

Review
42 Rodriguez, O.C. et al. (2003) Conserved microtubule-actin
interactions in cell movement and morphogenesis. Nat. Cell Biol. 5,
599609
43 Munter, S. et al. (2006) Actin polymerisation at the cytoplasmic face of
eukaryotic nuclei. BMC Cell Biol. 7, 23
44 Balasubramaniam, M. and Freed, E.O. (2011) New insights into HIV
assembly and trafficking. Physiology (Bethesda) 26, 236251
45 Gheysen, D. et al. (1989) Assembly and release of HIV-1 precursor
Pr55gag virus-like particles from recombinant baculovirus-infected
insect cells. Cell 59, 103112
46 Nishi, M. et al. (2009) Requirement for microtubule integrity in the
SOCS1-mediated intracellular dynamics of HIV-1 Gag. FEBS Lett.
583, 12431250
47 Jouvenet, N. et al. (2006) Plasma membrane is the site of productive
HIV-1 particle assembly. PLoS Biol. 4, e435
48 Finzi, A. et al. (2007) Productive human immunodeficiency virus
type 1 assembly takes place at the plasma membrane. J. Virol. 81,
74767490
49 Ryo, A. et al. (2008) SOCS1 is an inducible host factor during HIV-1
infection and regulates the intracellular trafficking and stability of
HIV-1 Gag. Proc. Natl. Acad. Sci. U.S.A. 105, 294299
50 Vuong, B.Q. et al. (2004) SOCS-1 localizes to the microtubule
organizing complex-associated 20S proteasome. Mol. Cell. Biol. 24,
90929101
51 Kim, W. et al. (1998) Binding of murine leukemia virus Gag
polyproteins to KIF4, a microtubule-based motor protein. J. Virol.
72, 68986901
52 Tang, Y. et al. (1999) Cellular motor protein KIF-4 associates with
retroviral Gag. J. Virol. 73, 1050810513
53 Martinez, N.W. et al. (2008) Kinesin KIF4 regulates intracellular
trafficking and stability of the human immunodeficiency virus type 1
Gag polyprotein. J. Virol. 82, 99379950
54 Samejima, K. et al. (2012) Mitotic chromosomes are compacted
laterally by KIF4 and condensin and axially by topoisomerase
IIalpha. J. Cell Biol. 199, 755770
55 Zhou, H. et al. (2008) Genome-scale RNAi screen for host factors
required for HIV replication. Cell Host Microbe 4, 495504
56 Azevedo, C. et al. (2009) Inositol pyrophosphate mediated
pyrophosphorylation of AP3B1 regulates HIV-1 Gag release. Proc.
Natl. Acad. Sci. U.S.A. 106, 2116121166
57 Gaudin, R. et al. (2012) Critical role for the kinesin KIF3A in the
HIV life cycle in primary human macrophages. J. Cell Biol. 199,
467479
58 Lehmann, M. et al. (2009) Intracellular transport of human
immunodeficiency virus type 1 genomic RNA and viral production
are dependent on dynein motor function and late endosome
positioning. J. Biol. Chem. 284, 1457214585
59 Poole, E. et al. (2005) HIV-1 Gag-RNA interaction occurs at a
perinuclear/centrosomal site; analysis by confocal microscopy and
FRET. Traffic 6, 741755
60 Kutluay, S.B. and Bieniasz, P.D. (2011) Analysis of the initiating
events in HIV-1 particle assembly and genome packaging. PLoS
Pathog. 6, e1001200
61 Ott, D. et al. (2009) The nucleocapsid region of human
immunodeficiency virus type 1 Gag assists in the coordination of
assembly and Gag processing: role for RNAGag binding in the
early stages of assembly. J. Virol. 83, 77187727
62 Jouvenet, N. et al. (2009) Imaging the interaction of HIV-1 genomes
and Gag during assembly of individual viral particles. Proc. Natl.
Acad. Sci. U.S.A. 106, 1911419119
63 Miranda, L.R. et al. (2002) Cell surface expression of the HIV-1
envelope glycoproteins is directed from intracellular CTLA-4containing regulated secretory granules. Proc. Natl. Acad. Sci.
U.S.A. 99, 80318036
64 Soldati, T. and Schliwa, M. (2006) Powering membrane traffic in
endocytosis and recycling. Nat. Rev. Mol. Cell Biol. 7, 897908
65 Rey, O. et al. (1996) HIV-1 Gag protein associates with F-actin present
in microfilaments. Virology 220, 530534
66 Liu, B. et al. (1999) Interaction of the human immunodeficiency virus
type 1 nucleocapsid with actin. J. Virol. 73, 29012908
67 Wilk, T. and Fuller, S.D. (1999) Towards the structure of the human
immunodeficiency virus: divide and conquer. Curr. Opin. Struct. Biol.
9, 231243

Trends in Cell Biology December 2013, Vol. 23, No. 12

68 Hottiger, M. et al. (1995) The large subunit of HIV-1 reverse

transcriptase interacts with beta-actin. Nucleic Acids Res. 23,
736741
69 Fackler, O.T. et al. (1997) Association of human immunodeficiency
virus Nef protein with actin is myristoylation dependent and
influences its subcellular localization. Eur. J. Biochem. 247, 843851
70 Kimura, T. et al. (2000) Rev-dependent association of the introncontaining HIV-1 gag mRNA with the nuclear actin bundles and
the inhibition of its nucleocytoplasmic transport by latrunculin-B.
Genes Cells 5, 289307
71 Hofmann, W. et al. (2001) Cofactor requirements for nuclear export of
Rev response element (RRE)- and constitutive transport element
(CTE)-containing retroviral RNAs. An unexpected role for actin. J.
Cell Biol. 152, 895910
72 Simon, D.N. and Wilson, K.L. (2011) The nucleoskeleton as a genomeassociated dynamic network of networks. Nat. Rev. Mol. Cell Biol. 12,
695708
73 Sasaki, H. et al. (1995) Myosinactin interaction plays an important
role in human immunodeficiency virus type 1 release from host cells.
Proc. Natl. Acad. Sci. U.S.A. 92, 20262030
74 Ott, D.E. et al. (1996) Cytoskeletal proteins inside human
immunodeficiency virus type 1 virions. J. Virol. 70, 77347743
75 Wilk, T. et al. (1999) Actin associates with the nucleocapsid domain of
the human immunodeficiency virus Gag polyprotein. J. Virol. 73,
19311940
76 Chertova, E. et al. (2006) Proteomic and biochemical analysis of
purified human immunodeficiency virus type 1 produced from
infected monocyte-derived macrophages. J. Virol. 80, 90399052
77 Cooper, J. et al. (2011) Filamin A protein interacts with human
immunodeficiency virus type 1 Gag protein and contributes to
productive particle assembly. J. Biol. Chem. 286, 2849828510
78 Santos, S. et al. (2012) Virus-producing cells determine the host
protein profiles of HIV-1 virion cores. Retrovirology 9, 65
79 Boulant, S. et al. (2011) Actin dynamics counteract membrane
tension during clathrin-mediated endocytosis. Nat. Cell Biol. 13,
11241131
80 Chen, P. et al. (2007) Predominant mode of human immunodeficiency
virus transfer between T cells is mediated by sustained Envdependent neutralization-resistant virological synapses. J. Virol.
81, 1258212595
81 Rudnicka, D. et al. (2009) Simultaneous cell-to-cell transmission of
human immunodeficiency virus to multiple targets through
polysynapses. J. Virol. 83, 62346246
82 Nejmeddine, M. et al. (2005) Human T-lymphotropic virus, type 1, tax
protein triggers microtubule reorientation in the virological synapse.
J. Biol. Chem. 280, 2965329660
83 Sol-Foulon, N. et al. (2007) ZAP-70 kinase regulates HIV cell-to-cell
spread and virological synapse formation. EMBO J. 26, 516526
84 Jolly, C. et al. (2011) The regulated secretory pathway in CD4+ T cells
contributes to human immunodeficiency virus type-1 cell-to-cell
spread at the virological synapse. PLoS Pathog. 7, e1002226
85 Lehmann, M. et al. (2011) How HIV-1 takes advantage of the
cytoskeleton during replication and cell-to-cell transmission.
Viruses 3, 17571776
86 Llewellyn, G.N. et al. (2010) Nucleocapsid promotes localization of
HIV-1 gag to uropods that participate in virological synapses between
T cells. PLoS Pathog. 6, e1001167
87 Sherer, N.M. et al. (2007) Retroviruses can establish filopodial bridges
for efficient cell-to-cell transmission. Nat. Cell Biol. 9, 310315
88 Sowinski, S. et al. (2008) Membrane nanotubes physically connect T
cells over long distances presenting a novel route for HIV-1
transmission. Nat. Cell Biol. 10, 211219
89 Eugenin, E.A. et al. (2009) Tunneling nanotubes (TNT) are induced by
HIV-infection of macrophages: a potential mechanism for intercellular
HIV trafficking. Cell. Immunol. 254, 142148
90 Van Prooyen, N. et al. (2010) Human T-cell leukemia virus type 1 p8
protein increases cellular conduits and virus transmission. Proc. Natl.
Acad. Sci. U.S.A. 107, 2073820743
91 Xu, W. et al. (2009) HIV-1 evades virus-specific IgG2 and IgA
responses by targeting systemic and intestinal B cells via longrange intercellular conduits. Nat. Immunol. 10, 10081017
92 Murooka, T.T. et al. (2012) HIV-infected T cells are migratory vehicles
for viral dissemination. Nature 490, 283287
661

Review
93 Cooper, G.M. (2000) Microtubules. In The Cell: A Molecular Approach
(2nd edn), Sinauer Associates
94 Cooper, G.M. (2000) Structure and organization of actin filaments. In
The Cell: A Molecular Approach (2nd edn), Sinauer Associates
95 Vale, R.D. (2003) The molecular motor toolbox for intracellular
transport. Cell 112, 467480
96 Caviston, J.P. and Holzbaur, E.L. (2006) Microtubule motors at the
intersection of trafficking and transport. Trends Cell Biol. 16, 530537
97 Yeung, M.L. et al. (2009) A genome-wide short hairpin RNA screening
of Jurkat T-cells for human proteins contributing to productive HIV-1
replication. J. Biol. Chem. 284, 1946319473
98 Konig, R. et al. (2008) Global analysis of host-pathogen interactions
that regulate early-stage HIV-1 replication. Cell 135, 4960
99 Goff, S. (2007) Retroviridae: the retroviruses and their replication. In
Fields Virology (5th edn) (Knipe, D.M. and Howley, P.M., eds), pp.
19992070, Lippincott Williams & Wilkins
100 Sharova, N. et al. (2005) Macrophages archive HIV-1 virions for
dissemination in trans. EMBO J. 24, 24812489
101 Raposo, G. et al. (2002) Human macrophages accumulate HIV-1
particles in MHC II compartments. Traffic 3, 718729
102 Pelchen-Matthews, A. et al. (2003) Infectious HIV-1 assembles in late
endosomes in primary macrophages. J. Cell Biol. 162, 443455

662

Trends in Cell Biology December 2013, Vol. 23, No. 12

103 Kramer, B. et al. (2005) HIV interaction with endosomes in

macrophages and dendritic cells. Blood Cells Mol. Dis. 35, 136142
104 Marsh, M. et al. (2009) HIV assembly and budding in macrophages.
Biochem. Soc. Trans. 37, 185189
105 Benaroch, P. et al. (2010) HIV-1 assembly in macrophages.
Retrovirology 7, 29
106 Jouve, M. et al. (2007) HIV-1 buds and accumulates in nonacidic
endosomes of macrophages. Cell Host Microbe 2, 8595
107 Deneka, M. et al. (2007) In macrophages, HIV-1 assembles into an
intracellular plasma membrane domain containing the tetraspanins
CD81, CD9, and CD53. J. Cell Biol. 177, 329341
108 Welsch, S. et al. (2007) HIV-1 buds predominantly at the plasma
membrane of primary human macrophages. PLoS Pathog. 3, e36
109 Gaudin, R. et al. (2013) Dynamics of HIV-containing compartments in
macrophages reveal sequestration of virions and transient surface
connections. PLoS ONE 8, e69450
110 Mlcochova, P. et al. (2013) Organization and regulation of
intracellular plasma membrane-connected HIV-1 assembly
compartments in macrophages. BMC Biol. 11, 89
111 Berre, S. et al. CD36-specific antibodies block release of HIV-1 from
infected primary macrophages and its transmission to T cells. J. Exp.
Med. (http://dx.doi.org/10.1084/jem.20130566)