J Mol Evol (1998) 47:517530
Springer-Verlag New York Inc. 1998
Symbiosis Between Methanogenic Archaea and -Proteobacteria as the
Origin of Eukaryotes: The Syntrophic Hypothesis
David Moreira,1 Purificacion Lopez-Garca2
1
Laboratoire de Biologie Cellulaire (BC4), Batiment 444, URA CNRS 2227, Universite Paris-Sud, 91405 Orsay Cedex, France
2
Institut de Genetique et Microbiologie, Batiment 409, Universite Paris-Sud, 91405 Orsay Cedex, France
Received: 5 January 1998 / Accepted: 18 March 1998
Abstract. We present a novel hypothesis for the origin
of the eukaryotic cell, or eukaryogenesis, based on a
metabolic symbiosis (syntrophy) between a methano-
genic archaeon (methanobacterial-like) and a -
proteobacterium (an ancestral sulfate-reducing myxobac-
terium). This syntrophic symbiosis was originally
mediated by interspecies H2 transfer in anaerobic, pos-
sibly moderately thermophilic, environments. During eu-
karyogenesis, progressive cellular and genomic cointe-
gration of both types of prokaryotic partners occurred.
Initially, the establishment of permanent consortia, ac-
companied by extensive membrane development and
close cellcell interactions, led to a highly evolved sym-
biotic structure already endowed with some primitive
eukaryotic features, such as a complex membrane system
defining a protonuclear space (corresponding to the ar-
chaeal cytoplasm), and a protoplasmic region (derived
from fusion of the surrounding bacterial cells). Simulta-
neously, bacterial-to-archaeal preferential gene transfer
and eventual replacement took place. Bacterial genome
extinction was thus accomplished by gradual transfer to
the archaeal host, where genes adapted to a new genetic
environment. Emerging eukaryotes would have inherited
archaeal genome organization and dynamics and, conse-
quently, most DNA-processing information systems.
Conversely, primordial genes for social and developmen-
tal behavior would have been provided by the ancient
myxobacterial symbiont. Metabolism would have been
Correspondence to: P. Lopez-Garca; e-mail: puri@igmors.u-psud.fr
issued mainly from the versatile bacterial organotrophy,
and progressively, methanogenesis was lost.
Key words: Eukaryogenesis Syntrophy Sym-
biosis Eukaryotic evolution Interspecies H2 trans-
fer Methanogenic archaea Histones -
Proteobacteria Myxobacteria Anaerobic consor-
tium
Introduction
Evolutionary biology has entered an era characterized by
the widespread use of phylogenetic reconstruction tech-
niques based on protein and nucleic acid sequence com-
parison. The most striking breakthrough was the so-
called Woesian revolution, whose origin dates back to
1977 when, after comparison of ribosomal RNA (rRNA)
oligonucleotide catalogs, a tripartite division of the living
world into primary kingdoms, Eubacteria, Archaebacte-
ria, and Eukaryotes, was claimed (Woese and Fox 1977).
These lineages were later reclassified as the domains
Bacteria, Archaea, and Eucarya, respectively (Woese et
al. 1990). Additional work on protein sequence compari-
son further supported the phylogenetic distinctness of the
three groups. Subsequently, sequence comparison of an-
cestral paralogous genes, i.e., those derived from genes
duplicated before the separation of the three domains,
such as the elongation factors EF-Tu (1) and EF-G (2)
and ATPase V and F genes, was used by some authors to
root the tree of life in the bacterial branch (Gogarten et
518
al. 1989; Iwabe et al. 1989). Such classification and root-
ing have become widely accepted and represent the
prevalent viewpoint on biological evolution, although its
validity remains controversial (Forterre et al. 1993;
Doolittle and Brown 1994).
Those ideas have essential implications regarding one
of the major unsolved questions in contemporary biol-
ogy: the origin of the eukaryotic cell or eukaryogenesis.
Notwithstanding Woeses original speculation of a very
primitive last common ancestor, a progenote, from
which the three domains would have been directly de-
rived (Woese 1987), the Gogarten and Iwabe rooting
implies two things: first, that the last common ancestor
(or cenancestor) was prokaryotic and, second, that eu-
karyotes were direct descendants of a common archaeal
eukaryal ancestor (Gogarten et al. 1989; Iwabe et al.
1989; Woese 1990). Despite the scientific success of this
so-called archaeal model, alternative hypotheses ad- quence comparison. Some phylogenies, especially those
dressing the question of eukaryotic origins have been
proposed. Detailed discussion of these is beyond the
scope of this work; however, we briefly summarize them
in order to provide a general framework as a base to
formulate a different model for eukaryogenesis.
Hypotheses for the Origin of Eukaryotes: State of
the Art and Criticisms
A completely independent hypothesis was originally
stated by Reanney (1974) and Darnell (1978), who pro-
posed that the cenancestor was a eukaryote itself. Nev-
ertheless, these authors did not explain how the transition
from this ancestor to prokaryotic cells proceeded. The
idea was recently revived by Forterre, who argued that
the results shown by Woese and others are artifacts de-
rived from the inadequacy of the phylogenetic methods
employed (Forterre 1996). He hypothesized that prokary-
otes issued from a eukaryote-like ancestor by a reductive
process as a consequence of adaptation to hyperthermo-
philic environments (Forterre 1995). However, these
ideas fail to explain the evolution of the extremely com-
plex eukaryotic cellular organization directly from pre-
cellular systems. If eukaryotes evolved directly from pre-
cellular systems, they must have relied on a continuous
supply of prebiotic organic substrates over a long period
of time, since bacterial-derived chloroplasts appeared
later in evolution. It is highly unlikely that any prebiotic
soup provided a regular source of organic nutrients up potheses appealing to an archaealbacterial partnership
to a eukaryote-like stage of organismal evolution. In ad-
dition, eukaryotic life seems incompatible with the high
temperature regimes of the archaean Earth, for which
increasing evidence exists (Knauth 1992; Schwartzman
and Shore 1996). The antagonism with high-temperature
regimes could be circumvented by alluding to a rela-
tively moderate thermophilic environment compatible
with higher temperature limits for eukaryotes (Brock
1986). However, the essential advantage of a hot origin
hypothesis is the implication that the first living beings
were autotrophic, relying only on chemolithotrophic re-
sources. Finally, a eukaryotic ancestor is at odds with the
Precambrian fossil record, which suggests that the most
ancient fossils (3500 million years ago) are prokaryotic
(Schopf 1993), whereas the earliest eukaryotic fossils
date at 1800 to 2100 million years ago (Han and Run-
negar 1992).
The remaining hypotheses are based on the currently
accepted idea of a prokaryotic cenancestor from which
the prokaryotic domains, i.e., Archaea and Bacteria, were
the primary offshoots. The above-mentioned archaeal
model, which entails the direct evolution of the eukary-
otic nuclear genome from a common archaealeukaryal
ancestor, could be classified as autogenous. This
model has not always been supported by protein se-
based on elongation factors EF-1 and EF-2 (Rivera and
Lake 1992; Baldauf et al. 1996), but also those based on
rRNA sequence parsimony analysis (Lake 1988), argued
in favor of a specific sisterhood between eukaryotes and
crenarchaea (kingdom Crenarchaeota), or eocytes, ac-
cording to the nomenclature of Rivera and Lake (1992).
The other archaeal lineage (kingdom Euryarchaeota)
would be closer to Bacteria (especially Gram-positive
bacteria), and the paraphyly of Archaea would be in-
ferred (Lake 1988; Gupta and Singh 1992; Rivera and
Lake 1992; Gupta and Golding 1993; Doolittle and
Brown 1994). However, some of these results have been
interpreted as artifacts derived from undetected hidden
paralogies and presumptive horizontal gene transfers
(Hilario and Gogarten 1993; Brown et al. 1994; Roger
and Brown 1996). Bias in species sampling could also
account for some molecular characteristics claimed to be
shared only by eukaryotes and crenarchaea; for instance,
several of these have recently been detected in some
methanogenic euryarchaea, including Methanopyrus
kandleri (Lake and Rivera 1996; Feng et al. 1997). In
addition, limitations in current phylogeny reconstruction
methods may constitute a major pitfall in this kind of
analysis, since very often the approximations and as-
sumptions utilized are quite unrealistic. These methods
are especially sensitive to different evolutionary rates
among the organisms studied, a problem known as un-
equal rate effect (Felsenstein 1978; Lake 1991).
As a result of all these apparent contradictions, hy-
at the origin of eukaryotes have been on the rise. These
could be considered chimeric or heterogenous,
and may be classified into three subgroups according to
the kind of relationship that putatively occurred between
the two partners: fusion, phagotrophy, or symbiosis. The
fusion model postulates that the eukaryotic nuclear ge-
nome is a chimera resulting from the amalgamation of an
archaeon and a bacterium (Zillig et al. 1989; Zillig

1991). Among other mechanistic difficulties, this hy-
pothesis fails to explain the origin of the nuclear mem-
brane. A previous model attempting to overcome this
problem had already been proposed by Cavalier-Smith,
suggesting that the origin of this structure was the fusion
of cytoplasmic cisternae (Cavalier-Smith 1980) in a tran-
sient intermediate between an archaeon and a Gram-
positive bacterium (Cavalier-Smith 1987). In addition,
fusion models are based on events never observed in
nature, i.e., cell fusion between lineages as phylogeneti-
cally distant as archaea and bacteria. Phagotrophic mod-
els propose the engulfment of one of the partners by the
other. Thus, Lake and Rivera (1994) suggested that a
bacterium engulfed a crenarchaeon (eocyte), which
would have originated the nucleus (eocyte hypoth-
esis). Similarly, Gupta and Golding (1996) postulated
the engulfment of an eocyte by a Gram-negative bacte-
rium lacking a cell wall. In these models, the nuclear
envelope and the endoplasmic reticulum would be rem-
nants of the eocytic plasma membrane and the bacterial
phagotrophic vesicle membrane, respectively. In a re-
lated but quite different model, Sogin (1991) proposed
the engulfment of an archaeon by a protoeukaryote still
endowed with RNA-based metabolism. However, it is
difficult to conceive the persistence of such ancient pro-
toeukaryotes having a primitive and inefficient RNA-
based metabolism in a world already inhabited by eco-
logically competitive DNA-based prokaryotes. However,
this hypothesis overcomes the main objection for the rest
of the phagotrophic models, which is the inability for
bacteria to accomplish phagocytosis.
Considering these major drawbacks, the most realistic
models could be those based on symbiotic relationships
between archaea and bacteria, i.e., the symbiotic models.
Symbiosis does not depend on single, perhaps unlikely,
events, such as cell fusion or phagocytosis, but offers the
possibility of coevolution (eventually cointegration) over
long periods of time. To date (see also Note Added in
Proof), the only symbiotic model described is the serial
endosymbiotic theory (SET) proposed by Margulis
(1970, 1993), which recapitulates some of the pioneering
ideas of Mereschkowsky (1905). This hypothesis postu-
lates that eukaryotes evolved from the symbiosis be-
tween an archaeon (a thermoacidophilic Thermoplasma-
like organism) and a bacterium (a spirochete). In this
way, the archaeon acquired swimming motility, while
the spirochete obtained useful metabolites (Margulis
1996). Subsequent endosymbiotic incorporations of -
proteobacteria and cyanobacteria originated mitochon-
dria and plastids, respectively. This hypothesis does not
favor the endosymbiotic origin of the nucleus. Although
this kind of motile symbiosis is known to occur between
eukaryotic protista and spirochetes (Cleveland and Grim-
stone 1964; Bloodgood and Fitzharris 1976), it has never
been observed between archaea and spirochetes, which,
in addition, are not known to thrive in extreme acido-
519
philic environments. T. acidophilum lacks a cell wall and
was the first archaeal species where putative histone
proteins were characterized (Searcy and DeLange 1980).
Those were regarded as crucial eukaryotic-like traits, and
some others followed (Searcy 1987), all of which made
this archaeon a candidate of choice for that proposal. But
in the end, its histone-like proteins turned out to be much
more closely related to bacterial HU proteins rather than
to eukaryotic histones (DeLange et al. 1981; Drlica and
Rouvie`re-Yaniv 1987; Grayling et al. 1994).
If we dismiss this (quite unlikely) Thermoplasma
spirochete symbiosis as the origin of eukaryotes, does
any kind of primary symbiosis exist that could have
paved the way to eukaryotic evolution? A look at the
present-day archaealbacterial symbioses may help an-
swer this question.
Contemporary Symbioses Between Archaea
and Bacteria
The existence of symbiosis has not been extensively ex-
plored within the domain Archaea. Most characterized
symbioses between archaea and bacteria are related to
the process of interspecies H2 transfer, which is meta-
bolically advantageous for both partners. The bacterial
partner ferments organic compounds producing more
oxidized organic molecules and H2 as end products. This
H2, together with environmental CO2, is assimilated by
the archaeal partner, a methanogen, to form methane in
an energy-conserving reaction. The methanogen acts as a
sink of H2, thus increasing the velocity of the otherwise
less efficient bacterial fermentative reactions. This kind
of mutualistic symbiosis is known as syntrophy and it is
frequently found in anaerobic natural communities (for a
review, see Fenchel and Finlay 1995). Generally, a close
proximity between partners is required to avoid loss of
the volatile H2. This requirement leads to the establish-
ment of well-integrated consortia. A most striking ex-
ample is the case of the so-called Methanobacillus om-
elianskii, which, 50 years after its discovery, was
shown to be not a single organism but a pair of dissimilar
species in close contact: a chemotrophic bacterium
known as S-organism and the methanogenic archaeon
Methanobacterium bryantii (Bryant et al. 1967). When
isolated, each organism grows poorly on its own. For this
reason they were classified as obligate syntrophs.
Many more examples of this kind of syntrophic sym-
biosis between methanogens and bacteria are presently
known. They involve a variety of methanogenic archaea
and a more restricted diversity of bacteria, usually mem-
bers of the subdivision of the Proteobacteria, which
groups the vast majority of the sulfate-reducing bacteria
(Woese 1987). Some of these bacteria, e.g., species of
the genera Desulfovibrio and Syntrophobacter, can de-
velop syntrophic relationships with methanogens (Tatton
520
et al. 1989). Also, some nonstrict sulfate-reducing -
proteobacterial genera are specialized in obligate syntro-
phic symbioses with methanogens, such as Pelobacter
and Syntrophus (Zinder 1993).
The existence of such intimate relationships raises an
intriguing and exciting question: Might a methanogenic
archaeon/-proteobacterium syntrophic symbiosis have
triggered eukaryogenesis?
A Syntrophic Symbiosis at the Origin of Eukaryotes
We hypothesize that eukaryotes emerged from a syntro-
phic symbiotic event involving a methanogenic archaeon
and a -proteobacterium in an anaerobic, likely moder-
ately thermophilic, context. The former would have pro-
vided the basic genome and nucleic acid metabolism, and
the latter, most metabolic capabilities. Presumably, the
eukaryotic endoplasmic reticulum and nuclear mem-
brane arose, at least in part, from archaeal membranes or
membrane components, while the plasmic membrane de-
rived mostly from a bacterial membrane.
We try now to justify this proposal briefly by first
analyzing some key eukaryotic-like features of the re-
spective candidate partners.
The Archaeal Partner
Since the recognition of Archaea as a distinct lineage,
archaeal research has been prompted largely by the dis-
covery of many essential features in common with eu-
karyotes but absent from bacteria. Transcriptional ma-
chinery similarities were among the first to be
documented (Zillig et al. 1978, 1979). Subsequent work,
including completion of the first archaeal genome se-
quences, allowed the identification of homologies in rep-
lication and translation as well (Bult et al. 1996; Klenk et
al. 1997; Olsen and Woese 1997; Smith et al. 1997).
Most authors interprete these as evidence of archaeal
eukaryal sisterhood rather than specific bacterial loss
from a complex ancestor. Many of these shared features
would equally support either an autogenous or any bac-
terialarchaeal chimeric (heterogenous) hypothesis. In
this section, we do not recapitulate all these characteris-
tics, extensively revised elsewhere (Klenk and Doolittle
1994; Langer et al. 1995; Thomm 1996; Belfort and
Weiner 1997; Dennis 1997; Brown and Doolittle 1997;
Edgell and Doolittle 1997; Reeve et al. 1997). Alterna-
tively, we focus on certain essential traits shared specifi-
cally by eukaryotes and archaea belonging to the king-
dom Euryarchaeota (particularly a subgroup of
euryarchaeal methanogens), which might reveal an inti-
mate phyletic relationship between both (a common his-
tory). These are the presence of histones and nucleo-
somes, DNA topoisomerases, and lipids.
A classical eukaryotic hallmark is the organization of
chromatin in nucleosome arrays. Eukaryotic nucleo-
somes are formed by a histone core, consisting of an
octamer built upon an (H3H4)2 tetramer plus two H2A
H2B dimers, around which DNA wraps (Kornberg 1977;
Luger et al. 1997). The tetramer interacts first with DNA,
and it is critical to determine nucleosome positioning,
whereas H2AH2B dimers assemble later and may have
a role in transcriptional regulation. Likely, H2A and H2B
evolved after H3H4, allowing further DNA condensa-
tion (Ramakrisnan 1995) and increasing the regulatory
repertoire (Luger et al. 1997). Histone homologues spe-
cifically related, by sequence and structural data, to the
highly conserved H3H4 have been discovered in Ar-
chaea, but exclusively in the euryarchaeal branch (Gray-
ling et al. 1994; Starich et al. 1996; Pereira et al. 1997;
Reeve et al. 1997; Zlatanova 1997). No core histone has
been discovered so far in Bacteria or in the crenarchaea,
in spite of intense search and of the fact that these pro-
teins are very abundant in the cell. They have been found
in the hyperthermophilic Thermococcales (Sandman et
al. 1994; Ronimus and Musgrave 1996), and in several
hyperthermophilic and mesophilic methanogens from the
orders Methanococcales and Methanobacteriales (Gray-
ling et al. 1994; Darcy et al. 1995; Pereira et al. 1997)
(Fig. 1). However, archaeal histones are not present in
the related Thermoplasma acidophilum, which contains
instead a bacterial HU-like protein, HTa (DeLange et al.
1981; Grayling et al. 1994). The presence of histones in
these archaea and in eukaryotes establishes an intriguing
link between both groups of organisms. Plausible expla-
nations of the occurrence of this essential eukaryotic
character in some archaea are (i) independent loss of
histones inherited from a histone-endowed common ar-
chaealeukaryal ancestor (or even cenancestor), but re-
tention in eukaryotes and some archaea for unknown
reasons, and (ii) horizontal transfer. However, the most
parsimonious possibility may be that histones evolved in
the euryarchaeal branch and eukaryotes simply inherited
them. From this viewpoint, eukaryotes would have
implemented and optimized histone function, may be
more related to transcriptional regulation at the begin-
ning. Indeed, a parallel origin of the nucleosome core
and eukaryotic transcription from archaea has already
been formulated based on the similarities between the
protein families H3, H4, archaeal histones, and the cen-
tral domain of subunits A and B of the CCAAT-binding
factor (CBF), a transcription factor associated with eu-
karyotic promoters (Ouzounis and Kyrpides 1996).
A second euryarchaealeukaryotic connection con-
cerns DNA topoisomerase distribution. These universal
proteins are required to disentangle DNA strands or du-
plexes and are, therefore, indispensable for cellular pro-
cesses such as replication, transcription, recombination,
and regulation of DNA supercoiling (Drlica 1992; Wang
1996). Depending on their ability to cut one or both DNA
strands, they are classified into two mechanistic types, I

Fig. 1. Distribution of histones, reverse gyrase, and gyrase in an rRNA-based archaeal tree and possible phylogenetic position of the archaeal
syntrophic partner. The tree (adapted from Woese et al. 1990; Stetter 1996) is idealized, so that branch lengths are not necessarily proportional to
phylogenetic distances. Thick lines indicate thermophilic lineages; asterisks, moderately thermophilic species.
and II, each one of which is subdivided into two phylo-
genetic families (Wang 1996; Bergerat et al. 1997). The
presence of at least one type I and one type II enzyme
member seems to be necessary for any organism. In this
regard, and strictly speaking, the only archaeal trace pre-
sent in eukaryotes is a protein involved in meiotic re-
combination that is homologous to one of the two sub-
units of topoisomerase VI (Topo VI) (Bergerat et al.
1997). Topo VI is the prototype of a recently recognized
type II phylogenetic family, which is widespread among
the archaea but not found in bacteria. Nevertheless, an-
other interesting observation comes from the distribution
in archaea of the two supercoil-introducing activities
known to date, gyrase and reverse gyrase. Reverse gy-
rase has attracted much attention because of its ability to
introduce positive supercoils in DNA and because it ap-
pears to be a marker of hyperthermophily. It was first
thought to be only archaeal. However, it was found not
only in all hyperthermophilic archaea, but also in hyper-
thermophilic bacteria (for reviews, see Duguet 1995;
Forterre et al. 1996). Thus, it is present in hyperthermo-
philic methanogens, but not in mesophilic or thermo-
philic (such as Methanobacterium thermoautotrophicum,
growing optimally at 65C) species (Fig. 1). On the other
hand, gyrase, a type II DNA topoisomerase specifically
introducing negative supercoils, is present in all bacteria.
It may be an original bacterial feature, since it is the only
type II enzyme in the deep-branching Thermotoga mar-
itima (Guipaud et al. 1997) and likely in many other
bacteria (Huang 1996). It also appears in some euryar-
521
chaea (Archaeoglobus and haloarchaea; see Fig. 1), pre-
sumably by horizontal import from bacteria (Forterre et
al. 1994; Klenk et al. 1997; Lopez-Garca 1998). Neither
gyrase nor reverse gyrase is present in eukaryotes, which
possess only topoisomerases with relaxing activities. If
we consider, for instance, the rRNA phylogenetic tree, a
sort of living temperature gradient may be established
along the euryarchaeal branch, from the deep-branching
hyperthermophiles to the mesophilic halophiles (Fig. 1).
Roughly, reverse gyrase is present at the base of the
branch, coincident with hyperthermophily, whereas gy-
rase appears toward the tip. Interestingly, M. thermoau-
totrophicum, located in the middle of this gradient, is
devoid of the genes coding for both supercoil-
introducing activities (Smith et al. 1997). Although this
is not a conclusive piece of evidence, it is interesting to
note that this organism is endowed with only relaxing
topoisomerases, which, in combination with histones,
highly recalls the situation found in eukaryotic cells.
One major argument against eukaryotic membranes
being derived from archaeal ones concerns lipids. Ar-
chaeal lipids are mainly isoprenoid glycerol ethers, in
contrast to the bacterial and eukaryal ester-based link-
ages. However, the presence of ether lipids is not re-
stricted to archaea, being present also in hyperthermo-
philic bacteria of the genera Thermotoga and Aquifex
(De Rosa et al. 1989; Huber et al. 1992). Only the si-
multaneous occurrence of ether linkages, the isopranic
nature of the aliphatic chains, and the chirality of glyc-
erol (2,3-sn-glycerol instead of 1,2-sn stereochemistry)
522

Table 1. General lipid occurrence and distribution in Archaeaa

Extreme thermophiles Methanogens Halophiles

Core lipids Caldarchaeol-based Archaeol-based Archaeol-based only

Polar lipids (headgroups)
Nonitol caldarchaeol
(with or without rings)
Phosphoinositolb
Sugars (glucose, galactose,
usually unsulfated)
Caldarchaeol (no rings)-basedb
Phosphoamino polar headgroups
(serine and ethanolamine, mostly)b
Unsulfated sugar groups
(glucose, galactose)
Phosphoglycerol-derived
Sulfated/unsulfated glycosyl groups
(glucose, mannose, galactose)

a
Archaeol, diphytanylglycerol diether; caldarchaeol, dibiphytanyldiglycerol tetraether. Based on Kates (1993).
b
Important traits (molecules or synthetic pathways) found also in eukaryotes (see text).

can be considered of phylogenetic relevance (Gamba- The Bacterial Partner

corta et al. 1994). Archaeal lipids are diverse and com-
plex (Table 1). This turns out to be especially true for
lipids in methanogenic species, whose variability is de-
rived mainly from the phosphodiester-bonded, water-
soluble alcoholic residues, in contrast to the lipid struc-
ture in extreme halophiles or sulfur-dependent archaea,
whose diversity is due to the sugar residues of glycolip-
ids (Koga et al. 1993).
Glucosaminyl archaetidylinositol was the first polar
lipid reported to share a characteristic with eukaryal-
specific lipids. It is the major either polar lipid in Metha-
nosarcina barkeri, and it consists of GlcNp(1-6)-myo-
inositol 1-phosphate as the polar head group, linked via
a phosphodiester bond to an archaeol (2,3-di-O-
phytanyl-sn-glycerol). Interestingly, this polar group is
identical to the conserved core structure of glycosylated
phosphatidylinositol, which serves as a membrane pro-
tein anchor in eukaryotes (Nishihara et al. 1992).
Another differential feature of archaeal lipids is the
occurrence of tetraether lipids in some archaeal mem-
bers, including methanogens and sulfur-dependent ar-
chaea (but not Methanopyrus or Thermococcus species)
(Table 1). It has been demonstrated recently that the
synthesis of tetraether from diether lipids in T. acidophi-
lum, a close relative of methanogens (Olsen et al. 1994),
is inhibited by a potent inhibitor of eukaryotic steroid
biosynthesis acting on squalene epoxidase. Both biosyn-
thetic enzymes appear to have structural similarity. Since
squalene epoxidase is located in the endoplasmic reticu-
lum, it has already been suggested that this structure
originated from cytoplasmic protoarchaeal membranes
(Yamagishi et al. 1996).
Although inositol and tetraether lipids are present in
some sulfur-dependent archaea and in methanogens, they
have not been detected in extreme halophiles. Glycerol
and amino compounds have been found as polar head
groups only in methanogens (Koga et al. 1993) (Table 1).
Taken together, these data suggest that there is a re-
stricted range of methanogenic archaea (centered around
Methanobacteriales) as putative candidates to have set up
a syntrophic eukaryogenetic symbiosis.
The subdivision of the Proteobacteria clusters three
groups that are phenotypically different, the dissimila-
tory sulfate-reducing bacteria, the myxobacteria, and the
bdellovibrios (Oyaizu and Woese 1985). All these
groups are able to perform a broad variety of energetic
metabolisms, and they may have evolved their respira-
tory chains from that of an anaerobic sulfate-reducing
ancestor (Woese 1987). Some genera within this subdi-
vision exhibit gliding motility, and the myxobacteria and
the bdellovibrios also exhibit complex developmental
and cell division cycles (Reichenbach and Dworkin
1992; Shimkets and Woese 1992).
Myxobacteria are of special interest with regard to the
problem of eukaryogenesis since, indeed, they display a
surprising number of similarities with eukaryotes at the
molecular level. The most remarkable are (a) the pres-
ence of serinethreonine kinases (previously thought to
be unique to Eukarya) (Munoz-Dorado et al. 1991); (b) a
protein involved in spore morphogenesis that shares
some structural properties with calmodulin and with ver-
tebrate and crystallins (Wistow et al. 1985; Inouye et
al. 1993); (c) an intercellular signaling factor (factor C)
homologous to vertebrate 17 -hydroxysteroid dehydro-
genase (Baker 1994), which is related to the existence of
steroids in this group (Kohl et al. 1983); (d) the presence
of a phosphatidyl-inositol cycle with the possible partici-
pation of a G protein in the process (Benassa et al. 1994)
(inositol phosphates are second messengers in eukaryotic
signal transduction); (e) the existence of reverse tran-
scriptase and retron elements (Rice et al. 1993); (f) the
presence of a GTPase involved in motility homologous
to the Ras/Rab/Rho superfamily of small eukaryotic
GTPases (Hartzell 1997); and (g) the existence of a tran-
scriptional factor homologous to the eukaryotic high-
mobility group (Y) proteins (Murillo 1997).
In addition, myxobacteria display interesting similari-
ties to some eukaryotic groups at the level of primary and
secondary metabolism. They secrete a great variety of
hydrolytic enzymes and antibiotics, likely in relation to
their ecological role as predators of other microorgan-

Fig. 2. Schematic representation of eukaryogenesis from syntrophic
symbiosis between methanogenic archaea and ancestral sulfate-
respiring -proteobacteria. Initially, a syntrophic consortium is estab-
lished (A) where each methanogenic cell (black) is surrounded by
several bacterial partners endowed with double membranes. Extensive
membrane development for efficient interspecies H2 transfer and in-
isms or scavengers of organic matter (Dworkin 1996).
This ability, together with pigment production, like
melanin (Burchard and Dworkin 1966), and their com-
plex developmental cycle, explains why botanists and
microbiologists classified them as fungi for almost a cen-
tury (reviewed by Reichenbach and Dworkin 1992).
Taken together, all these data allowed some authors to
conclude that there is little doubt that the myxobacteria
have played some kind of role in the evolution of eu-
karyotic multicellularity (Dworkin 1996). We think
that this kind of role may have been indeed played by
ancestral sulfate-reducing fermentative myxobacteria at
the very origin of eukaryotes.
The Eukaryogenetic Pathway
Formation of Stable Consortia and Close
CellCell Interactions
As depicted schematically in Fig. 2, the first eukaryoge-
netic step would consist of the consolidation of a perma-
nent syntrophic consortium integrated by a core metha-
nogenic archaeon entirely wrapped by -proteobacterial
cells. This situation can actually be found in present-day
microbial flocs, such as those formed by Methanobacte-
rium and Syntrophobacter species (Dubourgier et al.
1988; Thiele et al. 1988; Fenchel and Finlay 1995). Flocs
are generally composed of large numbers of both cell
types, sustaining a maximum efficiency by the adoption
of mosaic arrangements allowing extensive mixing of
523
creased cellcell interactions progressively occur (B), leading eventu-
ally to bacterial cytoplasmic fusion (C). Ultimately, redundant mem-
branes are lost (D), an endoplasmic reticulum develops, and nuclear
pores appear (E). The process is accompanied by gene transfer, gene
replacement, and bacterial genome extinction. Mitochondria may have
been acquired at the latest stages. See text.
both partners (Schink 1992). The production of mucus,
slime, and other extracellular materials by some syntro-
phic species greatly enhances the cohesion of these struc-
tures (Fenchel and Finlay 1995).
In this way, the physical juxtaposition of the syntro-
phic partners facilitates the interspecies H2 transfer,
which would be strongly improved by subsequent devel-
opment of coupled membrane invaginations and extru-
sions. Interestingly, methanobacteria are able to develop
extensive membrane invaginations forming internal
membrane systems. Indeed, they have been referred to as
methanochondria, due to their mitochondrial-like ap-
pearance under the microscope (Zeikus and Bowen
1975). They are also able to form membrane extrusions,
and even the production of branched cellular extensions
by an apparent cell wall-less species, Methanoplasma
elizabethii, was reported (Rose and Pirt 1981). These
kinds of membrane invaginations and extrusions are also
widespread among the proteobacteria, improving consid-
erably diverse metabolic capacities with the increase in
cell surface (Leive 1973).
Once this putative consortium of methanoarchaea and
anaerobic ancestral sulfate-reducing myxobacteria was
established, a further step would involve bacterial mem-
brane fusion around the methanogenic core (Fig. 2). This
would have been facilitated by the intimate relationships
already existing between the bacterial cells surrounding
the methanogen, in an analogous way to contemporary
myxobacterial interactions. The result would be a meth-
anogen embedded within a continuous layer of bacterial
cells with a complex membrane interphase. This arrange-
ment may be regarded as similar to the final situation of
524
a putative phagocytosis of an archaeon by a bacterium,
although not requiring the necessity of such a very un-
likely event.
Membrane Development and Membrane Loss
A later eukaryogenetic state would imply loss of one of
the plasma membranes (Fig. 2). Membrane loss occurs in
nature; for instance, phagocytotic membranes engulfing
some cellular endosymbionts, including mitochondria
and plastids (Gray 1992). Even the outer membrane of
Gram-positive bacteria could have been lost if, as sug-
gested, the presence of two cell membranes is an ancient
character in bacteria (Blobel 1980; Rachel et al. 1990).
Two possibilities can be considered. One involves the
loss of the external membrane of bacterial cells, i.e., the
bacterial membrane in closest contact with the methano-
gen. Hence, a single eukaryotic plasma membrane could
be generated. The preservation of the archaeal mem-
brane, currently used as an argument against a putative
archaeal origin of the eukaryotic nucleus, could in this
way explain some archaeal-like features of nuclear and
endoplasmic reticulum membranes (see above). If this
turns out to be the case, substitution of ether-linked lipids
by bacterial ester-linked ones would have been required,
while some polar lipids and metabolic pathways would
have been conserved. However, these similarities do not
argue specifically in favor of this possibility, since a
similar situation could result if the archaeal membrane
were lost, but some of the genes involved were con-
served, and their products began to interact with bacterial
membranes. This suggestive possibility would entail
elimination of the unique membrane of the methanogenic
archaeon, while maintaining useful membrane compo-
nents and their associated synthetic pathways. In such a
scenario, it might be speculated that the archaeal mem-
brane, completely encircled by a continuous layer of bac-
terial ones, became redundant and was lost, as well as the
outer bacterial membranes exposed to the environment.
In any case, the final situation would have corre-
sponded to a methanogenic core wrapped by a double
membrane with a complex system of invaginations and
extrusions. This eukaryotic-like membrane arrangement
could be considered preadaptative, having evolved ini-
tially with a function (increase in contact surface be-
tween syntrophic partners) quite different from that of
present-day eukaryotic membranes. The central compart-
ment surrounded by a double membrane could be con-
sidered as a protonucleus plus a protoendoplasmic re-
ticulum.
Gene Flow, Gene Replacement, and Eukaryotic
Genome Emergence
We suggest that, during this cointegrative process, a bi-
directional flow of genetic material between both part-
ners took place. Since single archaeal cells were sur-
rounded by several bacterial ones, horizontal transfer
from bacterial to archaeal genomes would have been
more important. Two critical consequences must have
followed. First, transfer of some housekeeping genes
from one species to the other led to an irreversible sta-
bilization of the symbiotic consortium. Second, the trans-
ferred bacterial genes met a new genetic environment in
the archaeal genome, and therefore, they had to adapt to
it by acquisition of some (eukaryotic-like) archaeal fea-
tures such as TATA boxes or histone-mediated regula-
tion. This symbiotic gene transfer could be compared to
that undergone by mitochondrial and plastid genomes
towards the nuclear genome (Gray 1992). In some cases,
the complete genome of a symbiont can be transferred in
a so-called genome extinction process (Palmer 1997).
Interesting examples are the hydrogenosomes (Muller
1993) and some secondary plastid endosymbionts
(Palmer and Delwiche 1996). Obviously, each trans-
ferred gene had to acquire a targeting signal ensuring
efficient posttranslational transport of the encoded pro-
tein to the donor symbiont. Such targeting signals gen-
erally consist of short sequences, which seem to be easily
incorporated, as testified by all mitochondrial and plas-
tid-derived nuclear genes. This may happen in a short
period of time, as has been documented for coxII. This
gene is normally transcribed from mitochondrial DNA,
but in some legumes it was recently transferred to the
nucleus, where it acquired a classical sequence to target
the protein back into the mitochondrion (Nugent and
Palmer 1991).
Logically, bacterial genes having homologues within
the host archaeal genome should also have been trans-
ferred, and these genes may have often replaced the host
copies. This is a well-known process for some eukaryotic
genes of mitochondrial or plastidial origin, and is re-
ferred to as symbiotic gene replacement (Martin et al.
1993; Martin and Schnarrenberger 1997). Obviously, the
probability of a successful gene replacement was higher
when the transferred gene encoded a protein that was not
involved in a large number of interactions with other
proteins, as could be the case for many enzymes of dif-
ferent metabolic pathways. In contrast, host genes en-
coding proteins interacting with many others were un-
likely to be replaced by the transferred homologues. This
would also be in agreement with the finding that eukary-
otic replication, transcription, and translation are ar-
chaeal-like, whereas some eukaryotic metabolic genes
are more akin to those of bacteria (Feng et al. 1997;
Brown and Doolittle 1997).
Such a putative gene flow, and eventually gene re-
placement, toward the archaeal symbiont would explain
eukaryotic genomes birth more easily than fusion mod-
els of eukaryogenesis do (Zillig et al. 1989). Whereas, as
stated by Roger and Brown (1996), it is very difficult to
explain how two very different genetic systems (archaeal

and bacterial) could be successfully integrated by such a
radical event as cell fusion, a progressive integration by
transfer of discrete amounts of genetic information, as
occurred in the case of mitochondria and plastids, seems
much more feasible. One simple and efficient genetic
transfer mechanism to be considered, although not ex-
clusive, would be conjugation, which, indeed, can effi-
ciently mobilize DNA between distantly related prokary-
otes (Mazodier and Davies 1991) and even between
individuals from different domains, including eukaryotes
(Buchanen-Wollaston et al. 1987; Heineman and Spra-
gue 1989). Notably, conjugative plasmids and conjuga-
tion are widespread in Archaea (Rosenshine et al. 1989;
Schleper et al. 1995; Zillig et al. 1996) and also in Pro-
teobacteria, including the subdivision (Argyle et al.
1992; Wall et al. 1993).
All these processes of gene uptake would have been
accompanied by an increase in genome size and, ulti-
mately, genome partition and linearization. These phe-
nomena occur in present-day prokaryotes. Indeed, some
(relatively large) prokaryotic genomes are integrated by
more than one large replicon (chromosomes and/or giant
plasmids) (Allardet-Servent et al. 1993; Michaux et al.
1993; Choudhary et al. 1994; Netolitzky 1995), and
many bacteria have linear chromosomes and plasmids
(Hinnebusch and Tilly 1993; Casjens et al. 1995). In
halophilic euryarchaea, some of which carry the largest
known genomes among the Archaea, large extrachromo-
somal elements (up to approximately 700 kb) are wide-
spread too (Lopez-Garca et al. 1996). Remarkably,
myxobacterial genomes are the largest found in prokary-
otes (around 910 mbp) (Chen et al. 1990; Neumann et
al. 1992), which might be required for their complex
social and developmental biology. Such values are com-
parable to those of free-living unicellular eukaryotes
such as Saccharomyces cerevisiae (13 mbp) and the mi-
croalga Galdieria sulphuraria (10 mbp) (Coetzee et al.
1987; Moreira et al. 1994).
Thus, finally, a developed integrated consortium al-
ready endowed with a complex large genome would
have emerged. Metabolic, social, and developmental
functions would have had a bacterial origin, whereas
genetic information-processing activities, less suscep-
tible to the interspecific transfer, would have been ar-
chaeal-like. This is the overall case in contemporary eu-
karyotic genomes.
Metabolism
Initially, the metabolic activities of the eukaryogenetic
syntrophic consortium would have been greatly depen-
dent on interspecific H2 transfer via intimate cell-to-cell
interactions. General metabolism would have mostly re-
lied on the bacterial H2-releasing fermentation of organic
substrates, followed by archaeal H2-dependent methano-
genesis. The end products were diverse oxidized organic
525
molecules and methane. Likely, the consortium dis-
played broad metabolic abilities, contributed mainly by
the bacterial partner. In fact, most -proteobacteria (both
sulfate-reducing and myxobacteria) are able to decom-
pose a great diversity of organic compounds (Reichen-
bach and Dworkin 1992; Widdel and Hansen 1992). In-
terestingly, myxobacteria produce extracellular enzymes
hydrolyzing several macromolecules (Reichenbach and
Dworkin 1992), which might have already evolved in the
putative ancestral sulfate-reducing myxobacterial part-
ner. Consequently with these capabilities, these consortia
would have thrived in anaerobic (maybe microaerobic)
environments, gaining energy from the fermentation of a
variety of organic compounds and, possibly, playing the
ecological role of organotrophic scavengers, as some eu-
karyotes do today.
During the eukaryogenetic process, a dramatic meta-
bolic change must have occurred, leading to the complete
loss of methanogenesis, absent in contemporary eukary-
otes. This change can be explained in several ways. On
the one hand, adaptation to an efficient saprophytism
(even phagotrophism after cytoskeleton development,
likely favored by size increase) made methanogenesis no
more selectively advantageous, and it was finally lost.
The methanogenic partner was, however, already essen-
tial to the consortium as recipient of the bacterial ge-
nome. On the other hand, the acquisition of the mito-
chondrial endosymbiont undoubtedly caused substantial
changes in the protoeukaryotic metabolism. All known
eukaryotes appear to have harbored mitochondria once
(Germot et al. 1996; Roger et al. 1996, 1998). Therefore,
mitochondrial symbiosis must have been a very early
event in the history of eukayotes, perhaps as ancient as
they are. At this point, it is tempting to speculate that the
mitochondrial symbiosis driving force might have been
methane production by the syntrophic consortium itself.
Most methanotrophic bacteria, which derive energy from
methane oxidation under aerobic or anaerobic condi-
tions, belong to the Proteobacteria, and some genera
(such as Methylocystis, Methylosinus, and Methylobac-
terium) to the subdivision (Lidstrom 1991; Hanson and
Hanson 1996). Methanotrophs are ubiquitous; some spe-
cies associate with methanogens to form consortia under
anoxic conditions (Hoehler et al. 1995), and some are
intracellular symbionts of diverse invertebrates (Cavan-
augh 1985, 1993). Furthermore, methanotrophs develop
internal membrane systems strikingly similar to those of
mitochondria (Davies and Whittemburg 1970; Takeda
and Tanaka 1980). If ancestral eukaryotes were methane
producers, the first mitochondrial endosymbionts could
have been anaerobic or microaerophilic methanotrophs
of the -Proteobacteria. Taking into account the fact that
mitochondrial acquisition is a very ancient event, per-
haps even prior to the massive accumulation of atmo-
spheric O2, the original endosymbionts metabolism may
have been different from the present-day aerobic one.
526
The common evolutionary origin of aerobic mitochon-
dria and anaerobic hydrogenosomes may actually sug-
gest this (Bui et al. 1996). A mitochondrial transition to
aerobic respiration would have been subsequent, as an
adaptative consequence to increasing environmental O2
concentrations. Methanogenesis was, thus, ultimately
lost in favor of a much more productive aerobic metabo-
lism, and old-fashioned methanogenic (proto)eukaryotes
might have been completely outcompeted by the novel
efficient organotrophs.
Discussion and Perspectives
We have reported here an alternative hypothesis for the
origin of eukaryotes in an attempt to overcome some of
the unsolved questions in current models, many of which
were artificially born to explain particular contradictions
in protein phylogenies. We believe that the idea that
eukaryotes derived from syntrophic symbiosis between
certain methanogenic archaea and ancestral anaerobic
sulfate-respiring -proteobacteria provides a coherent
molecular and cellular evolutionary scenario compatible
with actual phenomena in nature.
This evolutionary syntrophic symbiotic process pre-
sumably took place under anaerobic (or microaerobic),
and possibly moderately thermophilic (Schwartzman and
Shore 1996), conditions, as can be found in many pre-
sent-day methanogenic consortia (Zinder and Koch
1984). The inferred primeval eukaryotic anaerobic nature
could be in agreement with the reported existence of
eukaryotic fossils predating the massive oxygen release
to the atmosphere (Han and Runnegar 1992; Canfield
and Teske 1996).
In addition to the compatibility with observations
from microbial ecology, this eukaryogenetic pathway ac-
commodates a continuous evolution, allowing a parsimo-
nious explanation of the existence of many eukaryotic
characters in common with archaea or bacteria, in many
cases with specific groups within them. This is seen in
the case of histone- and nucleosome-based genome or-
ganization shared with the euryarchaea and of many mo-
lecular features of the complex social behavior and de-
velopmental cycles of some -proteobacteria.
Curiously, as mentioned in a previous section, certain
molecular phylogenies may favor a closer proximity of
eukaryotes and crenarchaea (Rivera and Lake 1992;
Baldauf et al. 1996). In addition to possible unrecognized
paralogies, horizontal transfers, or biased sampling, un-
equal rate effects may mask their real evolutionary rela-
tionships, especially since very different evolutionary
rates often occur among eukaryotes and archaea. Certain
insertion/deletion sequences are more difficult to ex-
plain. The example often cited corresponds to an oligo-
peptide stretch of 11 amino acids shared by EF-1 se-
quences from crenarchaea and eukaryotes, which is used
as evidence for their specific sisterhood (Rivera and
Lake 1992). However, it has been noted that this inser-
tion/deletion is not a stable character. In fact, among the
Euryarchaeota, this particular region has changed at least
three times (Creti et al. 1994). Therefore, an alternative
explanation would be that this oligopeptide stretch rep-
resents an ancestral archaeal character (perhaps present
in the euryarchaeon that we propose to be involved in the
eukaryogenetic pathway) that has been lost several inde-
pendent times in the euryarchaeal branch. Additionally,
although the EF-1 and EF-2 analysis indicates that the
eukaryotecrenarchaea relationship seems the best sup-
ported, other possibilities cannot be statistically rejected
(Baldauf et al. 1996). In any case, an essential conse-
quence of the proposed syntrophic hypothesis is that the
resulting eukaryotic genome would have had, at least
partly, a high evolutionary rate. Acceleration of evolu-
tionary rates appears to be a common phenomenon
among symbiotic partners (Lutzoni and Pagel 1997; Ma-
tic et al. 1997), and indeed, some eukaryotic genes, such
as rRNA genes, have evolved rapidly (De Rijk et al.
1995; Baldauf et al. 1996). Besides, the generation of
innovative properties is an intrinsic, even conclusive,
characteristic of symbiosis (Smith 1989; Margulis and
Fester 1993). Although mosaic evolution and differential
gene loss can complicate our understanding of lifes evo-
lutionary history, the differential presence of a signifi-
cant number of relevant characters, such as the common
occurrence of histones and associated pathways of geno-
mic regulation in euryarchaea and eukaryotes, may con-
stitute a valuable phylogenetic tool.
This syntrophic eukaryogenetic hypothesis is capable
of being tested. First, some nuclear eukaryotic genes are
thought to have been acquired via endosymbiotic gene
replacement by mitochondrial genes (Martin and Schnar-
renberger 1997). Perhaps a closer investigation of some
of them could reveal more similarity to their -proteo-
bacterial than to their -proteobacterial counterparts.
Even if many phylogenetic signals permitting discrimi-
nation between both possibilities could have been extin-
guished after such a long evolutionary period, having
-proteobacterial genomes sequenced may be very use-
ful. Unfortunately, few -proteobacterial gene sequences
are presently available to perform a meaningful analysis.
Second, evidence may come from the analysis of pre-
sent-day consortia. It would be most interesting to ana-
lyze in more detail the molecular biology of obligatory
symbiosis and try to detect, for instance, recent gene
horizontal transfers or how, if detected, they may be
taking place. Finally, a different look at the fossil record
(especially of prokaryotic consortia) might reveal impor-
tant information concerning one of the most exciting
intelectual challenges in evolutionary biology, the
enigma of the origin of eukaryotes.
Acknowledgments. We wish to thank David Musgrave and Henner
Brinkmann for critical reading of the manuscript. D.M. is a postdoc-

toral fellow of the spanish Ministerio de Educacion y Ciencia, and
P.L.G. is a postdoctoral fellow of the European Community (Biotech-
nology Program).
Note Added in Proof. During the reviewing process of the manu-
script, a new symbiotic hypothesis for the origin of eukayotes was
proposed founded on metabolic grounds: the hydrogen hypothesis
(Martin and Muller 1998). In this case, an anaerobic H2-dependent
autotrophic archaeon (possibly a methanogen) would have established
a syntrophic relationship with a bacterium (-proteobacterium-like)
generating H2 as a waste product of heterotrophic metabolism. The
archaeon would have finished by engulfing the bacterial symbiont to
increase contact surface. The bacterium would have subsequently
evolved to originate either hydrogenosomes or mitochondria in differ-
ent present-day eukaryotes. Although both hydrogen and syntro-
phic hypotheses are complementary in several aspects (especially
some metabolic details of the symbiosis and archaeal molecular fea-
tures), they differ substantially in the bacterial partnership. In our pro-
posal, mitochondria would derive from an independent, maybe simul-
taneous, symbiotic event. Interestingly, both hypotheses concur in the
suggestion of an anaerobic metabolism for the origin of protomitochon-
drial symbiosis.
References
Allardet-Servent A, Michaux-Charachon S, Jumas-Bilak E, Karayan L,
Ramuz M (1993) Presence of one linear and one circular chromo-
some in the Agrobacterium tumefaciens C58 genome. J Bacteriol
175:78697874
Argyle JL, Rapp-Giles BJ, Wall JD (1992) Plasmid transfer by conju-
gation in Desulfovibrio desulfuricans. FEMS Microbiol Lett 73:
255262
Baldauf SL, Palmer JD, Doolittle WF (1996) The root of the universal
tree and the origin of eukaryotes based on elongation factor phy-
logeny. Proc Natl Acad Sci USA 93:77497754
Baker ME (1994) Myxococcus xanthus C-factor, a morphogenetic para-
crine signal similar to Escherichia coli 3-oxoacyl-[acyl carrier pro-
tein] reductase and human 17b-hydroxysteroid dehydrogenase.
Biochem J 301:311312
Belfort M, Weiner A (1997) Another bridge between kingdoms: rRNA
splicing in archaea and eukaryotes. Cell 89:10031006
Benassa M, Vieyres-Lubochinsky J, Odeide R, Lubochinsky B (1994)
Stimulation of inositide degradation in Stigmatella aurantiaca. J
Bacteriol 176:13901393
Bergerat A, de Massy B, Gadelle D, Varoutas PC, Nicolas A, Forterre
P (1997) An atypical topoisomerase II from archaea with implica-
tions for meiotic recombination. Nature 386:414417
Blobel G (1980) Intracellular protein topogenesis. Proc Natl Acad Sci
USA 77:14961500
Bloodgood RA, Fitzharris TP (1976) Specific associations of prokary-
otes with symbiotic flagellate protozoa from the hindgut of the
termite Reticulitermes and the wood-eating roack Cryptocercus.
Cytobios 17:103122
Brock TD (1986) Introduction: an overview of the thermophiles. In:
Brock TD (ed) Thermophiles. Wiley-Interscience, New York, p 1
Brown JR, Doolittle WF (1997) Archaea and the prokaryote-to-
eukaryote transition. Microbiol Mol Biol Rev 61:456502
Brown JR, Masuchi Y, Robb FT, Doolittle WF (1994) Evolutionary
relationships of bacterial and archaeal glutamine synthetase genes.
J Mol Evol 38:566576
Bryant MP, Wolin EA, Wolin MJ, Wolfe RS (1967) Methanobacillus
omelianskii, a symbiotic association of two species of bacteria.
Arch Microbiol 59:2031
Buchanen-Wollaston V, Passiatore JE, Canon F (1987) The mob and
oriT mobilization functions of a bacterial plasmid promote its trans-
fer to plants. Nature 328:172175
527
Bui ETN, Bradley PJ, Johnson PJ (1996) A common evolutionary
origin for mitochondria and hydrogenosomes. Proc Natl Acad Sci
USA 93:96519656
Bult CJ, White O, Olsen GJ, Zhou L, Fleischmann RD, Sutton GG,
Blake JA, FitzGerald LM, Clayton RA, Gocayne JD, Kerlavage
AR, Dougherty BA, Tomb JF, Adams MD, Reich CI, Overbeek R,
Kirkness EF, Weinstock KG, Merrick JM, Glodek A, Scott JL,
Geoghagen NSM, Weidman JF, Fuhrmann JL, Nguyen D, Utter-
back TR, Kelley KM, Peterson JD, Sadow PW, Hanna MC, Cotton
MD, Roberts KM, Hurst MA, Kaine BP, Borodovsky M, Klenk HP,
Fraser CM, Smith HO, Woese CR, Venter JC (1996) Complete
genome sequence of the methanogenic archaeon, Methanococcus
jannaschii. Science 273:10581073
Burchard RP, Dworkin M (1966) Light-induced lysis and carotenogen-
esis in Myxococcus xanthus. J Bacteriol 91:535545
Canfield DE, Teske A (1996) Late proterozoic rise in atmospheric
oxygen concentration inferred from phylogenetic and sulphur-
isotope studies. Nature 382:127132
Casjens S, Delange M, Ley III HL, Rosa P, Huang WM (1995) Linear
chromosomes of Lyme disease agent spirochetes: genetic diversity
and conservation of gene order. J Bacteriol 177:27692780
Cavalier-Smith T (1980) Cell compartmentation and the origin of eu-
karyote membrane organelles. In: Schwemmler W, Schenk HEA
(eds) Endocytobiology, endosymbiosis and cell biology. de Gruy-
ter, Berlin, p 831
Cavalier-Smith T (1987) The origin of eukaryotic and archaebacterial
cells. Ann NY Acad Sci 503:1754
Cavanaugh CM (1985) Symbiosis of chemoautotrophic bacteria and
marine invertebrates from hydrothermal vents and reducing sedi-
ments. Bull Biol Soc Wash, p 373
Cavanaugh CM (1993) Methanotroph-invertebrate symbioses in the
marine environment: ultrastructural, biochemical and molecular
studies. In: Murell JC, Kelly DP (eds) Microbial growth on C1
compounds. Intercept Press, Andover, UK, p 315
Chen H, Keseler IM, Shimkets LJ (1990) Genome size of Myxococcus
xanthus determined by pulsed-field gel electrophoresis. J Bacteriol
172:42064213
Choudhary M, Mackenzie C, Nereng KS, Sodergren E, Weinstock GM,
Kaplan S (1994) Multiple chromosomes in bacteria: structure and
function of chromosome II of Rhodobacter sphaeroides 2.4.1T. J
Bacteriol 176:76947702
Cleveland LR, Grimstone AV (1964) The fine structure of the flagellate
Mixotricha paradoxa and its associated micro-organisms. Proc R
Soc Lond B Biol Sci 159:668686
Coetzee DJ, Kock JLF, Pretorius GHJ (1987) The value of orthogonal-
field-alternation gel electrophoresis as a rapid identification process
for some species representing the genus Saccharomyces. J Micro-
biol Methods 7:219225
Creti R, Ceccarelli E, Bocchetta M, Sanangelantoni AM, Tiboni O,
Palm P, Cammarano P (1994) Evolution of translation elongation
factor (EF) sequences: reliability of global phylogenies inferred
from EF-1(Tu) and EF-2(G) proteins. Proc Natl Acad Sci USA
91:32553259
Darcy TJ, Sandman K, Reeve JN (1995) Methanobacterium formici-
cum, a mesophilic methanogen, contains three HFo histones. J Bac-
teriol 177:858860
Darnell JE (1978) Implication of RNA-RNA splicing in evolution of
eukaryotic cells. Science 202:12571260
Davies SL, Whittenbury R (1970) Fine structure of methane and other
hydrocarbon oxidizing bacteria. J Gen Microbiol 61:227232
DeLange RJ, Williams LC, Searcy D (1981) A histone-like protein
(HTa) from Thermoplasma acidophilum. II. Complete amino acid
sequence. J Biol Chem 256:905911
De Rijk P, van de Peer Y, van den Broeck I, De Wachter D (1995)
Evolution according to large ribosomal subunit RNA. J Mol Evol
41:366375
Dennis P (1997) Ancient ciphers: translation in Archaea. Cell 89:1007
1010
528
De Rosa M, Gambacorta A, Huber R, Lanzotti V, Nicolaus B, Stetter
KO, Trincone A (1989) Lipid structures in Thermotoga maritima.
In: da Costa MS, Duarte JC, Williams RAD (eds) Microbiology of
extreme environments and its potential for biotechnology. Elsevier
Applied Science, London, p 167
Doolittle WF, Brown JR (1994) Tempo, mode, the progenote, and the
universal root. Proc Natl Acad Sci USA 91:67216728
Drlica K (1992) Control of bacterial DNA supercoiling. Mol Microbiol
6:425433
Drlica K, Rouvie`re-Yaniv J (1987) Histone-like proteins of bacteria.
Microbiol Rev 51:301319
Dubourgier HC, Archer DB, Algagnac G, Prensier G (1988) Structure
and metabolism of methanogenic microbial conglomerates. In: Hall
ER, Hobson PN (eds) Anaerobic digestion. Pergamon Press, Ox-
ford, p 13
Duguet M (1995) Reverse gyrase. In: Eckstein F, Lilley DMJ (eds)
Nucleic acids and molecular biology, Vol 9. Springer-Verlag, Ber-
linHeidelberg, p 84
Dworkin M (1996) Recent advances in the social and developmental
biology of the myxobacteria. Microbiol Rev 60:70102
Edgell DR, Doolittle WF (1997) Archaea and the origin(s) of DNA
replication proteins. Cell 89:995998
Felsenstein J (1978) Cases in which parsimony and compatibility meth-
ods will be positively misleading. Syst Zool 27:401410
Fenchel T, Finlay BJ (1995) Anaerobic prokaryotes: competition and
syntrophy. In: May RM, Harvey PH (eds) Ecology and evolution in
anoxic worlds. Oxford University Press, Oxford, p 85
Feng DF, Cho G, Doolittle RF (1997) Determining divergence times
with a protein clock: update and reevaluation. Proc Natl Acad Sci
USA 94:1302813033
Forterre P (1995) Thermoreduction, a hypothesis for the origin of pro-
karyotes. CR Acad Sci Paris 318:415422
Forterre P (1996) Protein versus rRNA: problems in rooting the uni-
versal tree of life. ASM News 63:8995
Forterre P, Benachenou-Lahfa N, Confalonieri F, Duguet M, Elie C,
Labedan B (1993) The nature of the last universal ancestor and the
root of the tree of life, still open questions. BioSystems 28:1532
Forterre P, Bergerat A, Gadelle D, Elie C, Lottspeich F, Confalonieri F,
Duguet M, Holmes M, Dyall-Smith M (1994) Evolution of DNA
topoisomerases and DNA polymerases: a perspective from Ar-
chaea. Syst Appl Microbiol 16:746758
Forterre P, Bergerat A, Lopez-Garca P (1996) The unique DNA to-
pology and DNA topoisomerases of hyperthermophilic archaea.
FEMS Microbiol Rev 18:237248
Gambacorta A, Trincone A, Nicolaus B, Lama L, De Rosa M (1994)
Unique features of lipids of Archaea. Syst Appl Microbiol 16:518
527
Germot A, Philippe H, Le Guyader (1996) Presence of a mitochondrial-
type 70-KDa heat shock protein in Trichomonas vaginalis suggests
a very early mitochondrial endosymbiosis in eukaryotes. Proc Natl
Acad Sci USA 93:1461414617
Gogarten JP, Kibak H, Dittrich P, Taiz L, Bowman EJ, Bowman BJ,
Manolson MF, Poole RJ, Date T, Oshima T, Konishi J, Denda K,
Yoshida M (1989) Evolution of the vacuolar H+-ATPase: implica-
tions for the origin of eukaryotes. Proc Natl Acad Sci USA 86:
66616665
Gray MW (1992) The endosymbiont hypothesis revisited. In: Wolsten-
home DR, Jeon KW (eds) Mitochondrial genomes. Int Rev Cytol
141, Academic Press, New York, p 233
Grayling RA, Sandman K, Reeve JN (1994) Archaeal DNA binding
proteins and chromosome structure. Syst Appl Microbiol 16:582
590
Guipaud O, Marguet E, Knoll KM, Bouthier de la Tour C, Forterre P
(1997) Both DNA gyrase and reverse gyrase are present in the
hyperthermophilic bacterium Thermotoga maritima. Proc Natl
Acad Sci USA 94:1060610611
Gupta R, Golding GB (1993) Evolution of hsp70 gene and its impli-
cations regarding relationships between Archaebacteria, Eubacteria
and Eukaryotes. J Mol Evol 37:573582
Gupta R, Golding GB (1996) The origin of the eukaryotic cell. Trends
Biochem Sci 21:166171
Gupta R, Singh (1992) Cloning of the HSP70 gene from Halobacte-
rium marismortui: relatedness of archaebacterial HSP70 to its eu-
bacterial homologs and a model for the evolution of the HSP70
gene. J Bacteriol 174:45944605
Han TM, Runnegar B (1992) Megascopic eukaryotic algae from the
2.1-billion-year-old nagaunee iron-formation, Michigan. Science
257:232235
Hanson RS, Hanson TE (1996) Methanotrophic bacteria. Microbiol
Rev 60:439471
Hartzell PL (1997) Complementation of sporulation and motility de-
fects in a prokaryote by a eukaryotic GTPase. Proc Natl Acad Sci
USA 94:98819886
Heinemann JA, Sprague GF (1989) Bacterial conjugative plasmids
mobilize DNA transfer between bacteria and yeasts. Nature 340:
205209
Hilario E, Gogarten JP (1993) Horizontal transfer of ATPase genes
the tree of life becomes a net of life. Biosystems 31:111119
Hinnebusch J, Tilly K (1993) Linear plasmids and chromosomes in
bacteria. Mol Microbiol 10:917922
Hoehler TM, Alperin MJ, Albert DB, Martens CS (1995) Field and
laboratory studies of methane oxidation in an anoxic sediment:
evidence for a methanogen-sulfate reducer consortium. Global Bio-
geochem Cycles 8:451463
Huang WM (1996) Bacterial diversity based on type II DNA topo-
isomerase genes. Annu Rev Genet 30:79107
Huber R, Wilharm T, Huber D, Trincone A, Burggraf S, Konig H,
Rachel R, Rockinger I, Frike H, Stetter KO (1992) Aquifex pirophi-
lus gen. nov. sp. nov., represents a novel group of marine hyper-
thermophilic hydrogen-oxidizing bacteria. Syst Appl Microbiol 15:
340351
Inouye S, Franceschini T, Inouye M (1993) Structural similarities be-
tween the developmental-specific protein from a Gram-negative
bacterium, Myxococcus xanthus, and calmodulin. Proc Natl Acad
Sci USA 80:68296833
Iwabe N, Kuma K, Hasegawa M, Osawa S, Miyata T (1989) Evolu-
tionary relationship of archaebacteria, eubacteria and eukaryotes
inferred from phylogenetic trees of duplicated genes. Proc Natl
Acad Sci USA 86:93559359
Kates M (1993) Membrane lipids of archaea. In: Kates M, Kushner DJ,
Matheson AT (eds) The biochemistry of Archaea (Archaebacteria).
Elsevier Science, Amsterdam, p 261
Klenk HP, Doolittle WF (1994) Archaea and eukaryotes versus bacte-
ria. Curr Biol 4:920922
Klenk HP, Clayton RA, Tomb JF, White O, Nelson KE, Ketchum KA,
Dodson RJ, Gwin M, Hickey EK, Peterson JD, Richardson DL,
Kerlavage AR, Graham DE, Kyrpides NC, Fleischmann RD,
Quackenbush J, Lee NH, Sutton GG, Gill S, Kirkness EF, Dough-
erty BA, McKenney K, Adams MD, Loftus B, Peterson S, Reich
CI, McNeil LK, Badger JH, Glodeck A, Zhou L, Overbeek R,
Gocayne JD, Weidman JF, McDonald L, Utterback T, Cotton MD,
Spriggs T, Artiach P, Kaine BP, Sykes SM, Sadow PW, DAndrea
KP, Bowman C, Fujii C, Garland SA, Mason TM, Olsen GJ, Fraser
CM, Smith HO, Woese CR, Venter JC (1997) The complete ge-
nome sequence of the hyperthermophilic, sulphate-reducing ar-
chaeon Archaeoglobus fulgidus. Nature 390:364370
Knauth LP (1992) Origin and diagenesis of cherts: an isotopic perspec-
tive. In: Clauer N, Chaudhuri S (eds) Isotopic signatures and sedi-
mentary records, Lecture Notes in Earth Sciences No 43. Springer
Verlag, Berlin, p 123
Koga Y, Nishihara M, Morii H, Akagawa-Matsushita M (1993) Ether
polar lipids of methanogenic bacteria: structures, comparative as-
pects, and biosynthesis. Microbiol Rev 57:164182
Kohl W, Gloe A, Reichenbach H (1983) Steroids from the myxobac-
terium Nannocystis exedens. J Gen Microbiol 129:16291635

Kornberg RD (1977) Structure of chromatin. Annu Rev Biochem 46:
931954
Lake JA (1988) Origin of the eukaryotic nucleus determined by rate-
invariant analysis of rRNA sequences. Nature 331:184186
Lake JA (1991) Tracing origins with molecular sequences: metazoan
and eukaryote beginnings. Trends Biochem Sci 16:4650
Lake JA, Rivera MC (1994) Was the nucleus the first endosymbiont?
Proc Natl Acad Sci USA 91:28802881
Lake JA, Rivera MC (1996) The prokaryotic ancestry of eukaryotes. In:
Roberts DMcL, Sharp P, Alderson G, Collins M (eds) Evolution of
microbial life. Cambridge University Press, Cambridge, p 87
Langer D, Hain J, Thuriaux P, Zillig W (1995) Transcription in Ar-
chaea: similarity to that in Eucarya. Proc Natl Acad Sci USA 92:
57685772
Leive L (1973) Bacterial membranes and walls. Marcel Dekker, New
York
Lidstrom ME (1991) The aerobic methylotrophic bacteria. In: Balows
A, Truper HG, Dworkin M, Harder W, Schleifer KH (eds) The
prokaryotes, 2nd ed. Springer-Verlag, New York, p 431
Lopez-Garca P (1998) DNA topoisomerases, temperature adaptation,
and early diversification of life. In: Wiegel J, Adams MWW (eds)
Thermophiles: the keys to molecular evolution and the origin of
life? Taylor & Francis, London, pp 201216
Lopez-Garca P, Amils R, Anton P (1996) Sizing chromosomes and
megaplasmids in haloarchaea. Microbiology 142:14231428
Luger K, Mader AW, Richmond RK, Sargent DF, Richmond TJ (1997)
Crystal structure of the nucleosome core particle at 2.8 resolu-
tion. Nature 389:251260
Lutzoni F, Pagel M (1997) Accelerated evolution as a consequence of
transitions to mutualism. Proc Natl Acad Sci USA 94:1142211427
Margulis L (1970) Origin of eukaryotic cells. Yale University Press,
New Haven, CT
Margulis L (1996) Archaeal-eubacterial mergers in the origin of Eu-
karya: phylogenetic classification of life. Proc Natl Acad Sci USA
93:10711076
Margulis L, Fester R (1993). Symbiosis as a source of evolutionary
innovation. MIT Press, Cambridge, MA
Martin W, Muller M (1998) The hydrogen hypothesis for the first
eukaryote. Nature 392:3741
Martin W, Schnarrenberger C (1997) The evolution of the Calvin cycle
from prokaryotic to eukaryotic chromosomes: a case study of func-
tional redundancy in ancient pathways through endosymbiosis.
Curr Genet 32:118
Martin W, Brinkmann H, Savonna C, Cerff R (1993) Evidence for a
chimeric nature of nuclear genomes: eubacterial origin of eukary-
otic glyceraldehyde-3-phosphate dehydrogenase genes. Proc Natl
Acad Sci USA 90:86928696
Matic I, Radman M, Taddei F, Picard B, Doit C, Bingen E, Denamur
E, Elion J (1997) Highly variable mutation rates in commensal and
pathogenic Escherichia coli. Science 277:18331834
Mazodier P, Davies J (1991) Gene transfer between distantly related
bacteria. Annu Rev Genet 25:147171
Mereschkowsky C (1905) Uber natur und ursprung der chromatophores
in pflanzenteilen. Biol Zentrabl 25:593635
Michaux S, Paillison J, Carles-Nurit MJ, Bourg G, Allardet-Servent A,
Ramuz, M (1993) Presence of two independent chromosomes in the
Brucella melitensis 16M genome. J Bacteriol 175:701705
Moreira D, Lopez-Archilla AI, Amils R, Marin I (1994) Characteriza-
tion of two new thermoacidophilic microalgae: genome organiza-
tion and comparison with Galdieria sulphuraria. FEMS Microbiol
Lett 122:109114
Muller M (1993) The hydrogenosome. J Gen Microbiol 139:2879
2889
Munoz-Dorado J, Inouye S, Inouye M (1991) A gene encoding a pro-
tein serine/threonine kinase is required for normal development of
Myxococcus xanthus, a Gram-negative bacterium. Cell 67:995
1006
Murillo FJ (1997) Un factor de tipo eucariotico en el proceso de de-
529
sarrollo multicelular de la bacteria Myxococcus xanthus. Proceed-
ings of the XVI Congress of the Spanish Microbiological Society,
p 54
Netolitzky DJ, Wu X, Jensen SE, Roy KL (1995) Giant linear plasmids
of -lactam antibiotic producing Streptomyces. FEMS Microbiol
Lett 131:2734
Neumann B, Pospiech A, Schairer HU (1992) Size and stability of the
genomes of the myxobacteria Stigmatella aurantiaca and Stigma-
tella erecta. J Bacteriol 174:63076310
Nishihara M, Utagawa M, Akutsu H, Koga Y (1992) Archaea contain
a novel diether phosphoglycolipid with a polar head group identical
to the conserved core of eucaryal glycosyl phosphatidylinositol. J
Biol Chem 267:1243212435
Nugent JM, Palmer JD (1991) RNA-mediated transfer of the gene coxII
from the mitochondrion to the nucleus during flowering plant evo-
lution. Cell 66:473481
Olsen GJ, Woese CR (1997) Archaeal genomics: an overview. Cell
89:991994
Olsen GJ, Woese CR, Overbeek R (1994) The winds of (evolutionary)
change: breathing new life new life into microbiology. J Bacteriol
176:16
Ouzounis CA, Kyrpides NC (1996) Parallel origins of the nucleosome
core and eukaryotic transcription from Archaea. J Mol Evol 42:
234239
Oyaizu H, Woese CR (1985) Phylogenetic relationships among the
sulfate respiring bacteria, myxobacteria and purple bacteria. Syst
Appl Microbiol 6:257263
Palmer JD (1997) Organelle genomes: going, going, gone! Science
275:790791
Palmer JD, Delwiche CF (1996) Second-hand chloroplasts and the case
of the dissapearing nucleus. Proc Natl Acad Sci USA 93:7432
7435
Pereira SL, Grayling RA, Lurz R, Reeve, J (1997) Archaeal nucleo-
somes. Proc Natl Acad Sci USA 94:1263312637
Rachel R, Engel AM, Huber R, Stetter KO, Baumeister W (1990) A
porin-type protein is the main constituent of the cell envelope of the
ancestral eubacterium Thermotoga maritima. FEBS Lett 262:6468
Ramakrisnan V (1995) The histone fold: evolutionary questions. Proc
Natl Acad Sci USA 92:1132811330
Reanney DC (1974) On the origin of prokaryotes. J Theor Biol 48:
243351
Reeve JN, Sandman K, Daniels C (1997) Archaeal histones, nucleo-
somes, and transcription. Cell 89:9991002
Reichenbach H, Dworkin M (1992) The myxobacteria. In: Balows A,
Truper HG, Dworkin M, Harder W, Schleifer KH (eds) The pro-
karyotes, 2nd ed. Springer-Verlag, New York, p 3416
Rice S, Bieber J, Chun JY, Stacey G, Lampson BC (1993) Diversity of
retron elements among a population of rhizobia and other Gram-
negative bacteria. J Bacteriol 175:42504254
Rivera MC, Lake JA (1992) Evidence that eukaryotes and eocyte pro-
karyotes are immediate relatives. Science 257:7476
Roger AJ, Brown JR (1996) A chimeric origin for eukaryotes re-
examined. Trends Biochem Sci 21:370371
Roger AJ, Clark CG, Doolittle WF (1996) A possible mitochondrial
gene in the early-branching amitochondriate protist Trichomonas
vaginalis. Proc Natl Acad Sci USA 93:1461814622
Roger AJ, Svard SG, Tovar J, Clark CG, Smith MW, Gillin FD, Sogin
ML (1998) A mitochondrial-like chaperonin 60 gene in Giardia
lamblia: evidence that diplomonads once harbored an endosymbi-
ont related to the progenito of mitochondria. Proc Natl Acad Sci
USA 95:229234
Ronimus RS, Musgrave DR (1996) A gene, han1A, encoding an ar-
chaeal histone-like protein from the Thermococcus species AN1:
homology with eukaryal histone consensus sequences and the im-
plications for delineation of the histone fold. Biochim Biophys Acta
1307:17
Rose CS, Pirt SJ (1981) Conversion of glucose to fatty acids and
530
methane: roles of two mycoplasmal agents. J Bacteriol 147:248
254
Rosenshine I, Tchelet R, Mevarech M (1989) The mechanism of DNA
transfer in the mating system of an archaebacterium. Science 245:
13871389
Sandman K, Perler F, Reeve JN (1994) Histone-encoding genes from
Pyrococcus: evidence for members of the HMf family of archaeal
histones in a non-methanogenic Archaeon Gene 150:207208
Schink B (1992) Syntrophism among prokaryotes. In: Balows A,
Truper HG, Dworkin M, Harder W, Schleifer KH (eds) The pro-
karyotes, 2nd ed. Springer-Verlag, New York, p 276
Schleper C, Holz I, Janekovic D, Murphy J, Zillig W (1995) A multi-
copy plasmid of the extremely thermophilic archaeon Sulfolobus
effects its transfer to recipients by mating. J Bacteriol 177:4417
4426
Schopf JW (1993) Microfossils of the early Archaean apex chert: new
evidence of the antiquity of life. Science 260:640646
Schwartzman DW, Shore SN (1996) Biotically mediated surface cool-
ing and habitability for complex life. In: Doyle LR (ed) Circum-
stellar habitable zones. Trevis House, Menlo Park, CA, p 421
Searcy D (1987) Phylogenetic and phenotypic relationships between
the eukaryotic nucleocytoplasm and thermophilic Archaebacteria.
Endocytobiology III 503:168179
Searcy D, DeLange RJ (1980) Thermoplasma acidophilum histone-like
protein. Partial amino acid sequence suggestive of homology to
eukaryotic histones. Biochim Biophys Acta 609:197200
Shimkets L, Woese CR (1992) A phylogenetic analysis of the myxo-
bacteria: basis for their classification. Proc Natl Acad Sci USA
89:94599463
Smith M (1989) Generating novelty by symbiosis. Nature 341:284285
Smith DR, Doucette-Stamm LA, Deloughery C, Lee H, Dubois J,
Aldredge T, Bashirzadeh R, Blakely D, Cook R, Gilbert K, Harri-
son D, Hoang L, Keagle P, Lumm W, Pothier B, Qiu D, Spadafora
R, Vicaire R, Wang Y, Wierzbowski J, Gibson, Jiwani N, Caruso
A, Bush D, Safer H, Patwell D, Prabhakar S, McDougall S, Shimer
G, Goyal A, Pietrokovski S, Church GM, Daniels CJ, Mao JI, Rice
P, Nolling J, Reeve JN (1997) Complete genome sequence of
Methanobacterium thermoautrophicum H: functional analysis
and comparative genomics. J Bacteriol 179:71357155
Sogin ML (1991) Early evolution and the origin of the eukaryotes. Curr
Opin Genet Dev 1:457463
Starich MR, Sandman K, Reeve JN, Summers MF (1996) NMR struc-
ture of HMf from the hyperthermophile Methanothermus fervidus
confirms that this archaeal protein is a histone. J Mol Biol 255:
187203
Stetter KO (1996) Hyperthermophilic procaryotes. FEMS Microbiol
Rev 18:149158
Takeda K, Tanaka K (1980) Ultrastructure of intracytoplasmic mem-
branes of Methylomonas margaritae cells grown under different
conditions. Antonie van Leeuwenhoek J Microbiol Serol 46:1525
Tatton MJ, Archer DB, Powell GE, Parker ML (1989) Methanogenesis
from ethanol by defined mixed continuous cultures. Appl Environ
Microbiol 55:440445
Thiele JH, Chartrain M, Zeikus JG (1988) Control of interspecies elec-
tron flow during anaerobic digestion: role of floc formation in
syntrophic methanogenesis. Appl Environ Microbiol 54:1019
Thomm M (1996) Archaeal transcription factors and their role in tran-
scription initiation. FEMS Microbiol Rev 18:159171
Wall JD, Rapp-Giles BJ, Rousset M (1993) Characterization of a small
plasmid from Desulfovibrio desulfuricans and its use for shuttle
vector construction. J Bacteriol 175:41214128
Wang JC (1996) DNA topoisomerases. Annu Rev Biochem 65:635
692
Widdel F, Hansen TA (1992) The dissimilatory sulfate- and sulfur-
reducing bacteria. In: Balows A, Truper HG, Dworkin M, Harder
W, Schleifer KH (eds) The prokaryotes, 2nd ed. Springer-Verlag,
New York, p 583
Wistow G, Summers L, Blundell T (1985) Myxococcus xanthus spore
coat protein S may have a similar structure to vertebrate lens ,-
crystallins. Nature 315:771773
Woese CR (1987) Bacterial evolution. Microbiol Rev 51:221271
Woese CR, Fox GE (1977) Phylogenetic structure of the prokaryotic
domain: the primary kingdoms. Proc Natl Acad Sci USA 74:5088
5090
Woese CR, Kandler O, Wheelis ML (1990) Towards a natural system
of organisms. Proposal for the domains Archaea, Bacteria and Eu-
carya. Proc Natl Acad Sci USA 87:45764579
Wu SS, Kaiser D (1995) Genetic and functional evidence that type IV
pili are required for social gliding motility in Myxococcus xanthus.
Mol Microbiol 18:547558
Yamagishi A, Kon T, Takahashi G, Oshima T (1996) Cytoplasmic
membrane of Themoplasma may be related to endoplasmic reticu-
lum of eukaryotic cells. Thermophiles 96, Athens, GA
Zeikus JG, Bowen VG (1975) Fine structure of Methanospirillum hun-
gatei. J Bacteriol 121:373580
Zillig W (1991) Comparative biochemistry of Archaea and Bacteria.
Curr Opin Genet Dev 1:544551
Zillig W, Stetter KO, Tobien M (1978) DNA-dependent RNA poly-
merase from Halobacterium halobium. Eur J Biochem 91:193199
Zillig W, Stetter KO, Janekovic D (1979) DNA-dependent RNA poly-
merase from the archaebacterium Sulfolobus acidocaldarius. Eur J
Biochem 96:597604
Zillig W, Klenk HP, Palm P, Leffers H, Puhler G, Gropp F, Garret RA
(1989) Did eukaryotes originate by a fusion event? Endocyt Cell
Res 6:125
Zillig W, Prangishvilli D, Schleper C, Elferink M, Holz I, Albers S,
Janekovic D, Gotz D (1996) Viruses, plasmids and other genetic
elements of thermophilic and hyperthermophilic Archaea. FEMS
Microbiol Rev 18:225236
Zinder SH (1993) Physiological ecology of methanogens. In: Ferry JG
(ed) Methanogenesis: ecology, physiology, biochemistry and ge-
netics. Chapman & Hall, New York, p 128
Zinder SH, Koch M (1984) Non-aceticlastic methanogenesis from ac-
etate: Acetate oxidation by a thermophilic syntrophic coculture.
Arch Microbiol 138:263272
Zlatanova J (1997) Archaeal chromatin: Virtual or real? Proc Natl Acad
Sci USA 94:1225112254