THE LIPID WORLD

DANIEL SEGRÉ1 , DAFNA BEN-ELI1 , DAVID W. DEAMER2 and DORON

LANCET1∗
1 Dept. of Molecular Genetics and The Crown Human Genome Center, The Weizmann Institute of
Science, Rehovot 76100, Israel
2 Dept. of Chemistry and Biochemistry, University of California, Santa Cruz, California 95064,
U.S.A.
(∗ author for correspondence, e-mail: [Link]@[Link])

(Received 12 November, 1999)

Abstract. The continuity of abiotically formed bilayer membranes with similar structures in contem-
porary cellular life, and the requirement for microenvironments in which large and small molecules
could be compartmentalized, support the idea that amphiphilic boundary structures contributed to
the emergence of life. As an extension of this notion, we propose here a ‘Lipid World’ scenario as an
early evolutionary step in the emergence of cellular life on Earth. This concept combines the poten-
tial chemical activities of lipids and other amphiphiles, with their capacity to undergo spontaneous
self-organization into supramolecular structures such as micelles and bilayers. In particular, the docu-
mented chemical rate enhancements within lipid assemblies suggest that energy-dependent synthetic
reactions could lead to the growth and increased abundance of certain amphiphilic assemblies. We
further propose that selective processes might act on such assemblies, as suggested by our computer
simulations of mutual catalysis among amphiphiles. As demonstrated also by other researchers, such
mutual catalysis within random molecular assemblies could have led to a primordial homeostatic sys-
tem displaying rudimentary life-like properties. Taken together, these concepts provide a theoretical
framework, and suggest experimental tests for a Lipid World model for the origin of life.

Keywords: compositional information, GARD, lipozyme, membrane mimetic chemistry, micellar

catalysis, mutual catalysis, origin of life, prebiotic evolution, self-reproduction

1. Introduction: Did Life Processes Necessarily Begin with Biopolymers?

The molecular systems from which life emerged were likely subject to the same
physical and chemical laws that guide self-assembly processes in contemporary liv-
ing systems (Tanford, 1978). While primitive prebiotic systems were not endowed
with the highly evolved information-coding and replication mechanism found in
contemporary cellular life, it is imperative that some forms of information storage
and transfer have been at work. Since the transition from microscopic chemical
mechanisms to the macroscopically detectable emergent properties that charac-
terize life remains unresolved, there is little knowledge of what such primitive
information-prone mechanism must have been. This is despite the fact that the
principles underlying present-day biochemistry and molecular biology are well
understood.

Origins of Life and Evolution of the Biosphere 31: 119–145, 2001.

© 2001 Kluwer Academic Publishers. Printed in the Netherlands.
120 D. SEGRÉ ET AL.

The first attempt to bridge this gap were Oparin’s experiments with coacervate
droplets (Oparin, 1957; Oparin et al., 1976). Although these are considered an
important historical step (Walde et al., 1994) the attention of investigators today
has shifted from colloid phenomena and protein chemistry to nucleic acids. Many
researchers maintain that the first living cells had predecessors in an RNA world
(Gilbert, 1986; Joyce et al., 1987; Gesteland et al., 1999). This scenario for life’s
origin is supported by striking experimental evidence that RNA can display cata-
lytic activity (Cech, 1993), and under defined conditions can be made to evolve new
catalytic activities through molecular Darwinian selection (Eigen and Schuster,
1982; Beaudry and Joyce, 1992; Wilson and Szostak, 1994). Still, this does not
imply that early evolutionary processes followed the same path. Indeed, RNA it-
self seems unlikely to have emerged spontaneously in a primordial environment
(Shapiro, 1984). Scenarios in which replication is a collective property of a loose
molecular assembly could be as likely to reflect the earliest stages of biogenesis
(Dyson, 1985; Farmer et al., 1986; Morowitz, 1992; Kauffman, 1993; Segré et
al., 1998a; Dyson, 1999; Segré et al., 2000a). The notion that lipids and other
amphiphiles could serve as intermediates in prebiotic evolution has also been spe-
cifically elaborated (Ourisson and Nakatani, 1994; Norris and Raine, 1998; Luisi
et al., 1999). In particular, it has been suggested that lipid membranes may have a
hereditary potential, as most membranes are generated from other membranes but
not created de novo (Cavalier-Smith, 1995; Szathmáry, 1999). These alternative
approaches have received less attention, probably because of the relative paucity
of experimental support, and because of the lack of a scenario explicitly involving
information storage and evolution therein.
Here we critically analyze some known physical and chemical properties of
intermediate-size organic molecules capable of forming non-covalent assemblies.
From this analysis, we propose an alternative view in which an original, high-
probability ‘Lipid World’ later gave rise to a world populated by the complex,
relatively improbable biopolymers that are ubiquitous in all life today. The advent
of the RNA world concept was initiated by the realization that nucleic acids have
catalytic capacities beyond their specific role as information carriers. We review
here the evidence that catalysis is not restricted to proteins and RNA, and that
lipids and other simple amphiphiles, normally considered to be chemically inert
and to have primarily a structural role, have substantial catalytic capacities. We
will also explore the possibility that non-covalent assemblies of amphiphilic mo-
lecules could be endowed with a capacity to store and propagate information, and
to undergo selection and evolution.

2. Lipid-like Amphiphiles Are Highly Diverse

In living cells today, biological membranes have numerous roles, including com-
partmentalization, energy transduction (photosynthesis and oxidative respiration)
 THE LIPID WORLD 121

nutrient and ion transport, signal transduction, and enzyme-catalyzed metabolic

reactions. The latter include the synthesis of the membrane lipids themselves to
allow growth of the bilayers in different cellular compartments. Yet membrane
lipids are still commonly regarded primarily as providing physical partitions, while
the embedded proteins allow them to subserve diverse functions. We argue here
that the active role of membranes is not a late stratagem of advanced cellular life,
but instead has an evolutionary continuity with the earliest life forms, in which
lipid-like molecules may have had a more active metabolic role.
A present-day eukaryotic cell incorporates three primary classes of lipid: phos-
pholipids, sphingolipids and sterols. When these are further differentiated with
respect to head groups, hydrocarbon chains and linking bonds, hundreds of dif-
ferent membrane lipids can be defined. The number grows even larger if we take
into account the variant lipids of prokaryotic cells. A similar diversity of lipophilic
organic compounds may be generated along other chemical pathways, such as
terpenoids (Ourisson and Nakatani, 1994). Although not all of these molecular spe-
cies are necessarily functionally distinct, it is significant that the combinatorics of
molecular structures can easily generate a large diversity (Figure 1A). The diversity
of primitive amphiphiles could be responsible for the rise of a multiplicity of func-
tions in a lipid monolayer or bilayer, through a form of combinatorial chemistry,
even in the absence of classical biopolymer combinatorics (Figure 1B) (cf. also
(Lehn, 1995)). It should be pointed out that present day organisms carry in their
circulation micellar particles with defined but highly heterogeneous compositions,
in the form of lipoproteins (Esterbauer, 1995). These could serve as models for the
proposed heterogeneous amphiphile assemblies discussed in this paper.

3. Amphiphiles Undergo Self-Assembly

Among the large number of molecular species expected to be found on prebiotic

Earth, lipid-like molecules have a distinct property: an ability to undergo spon-
taneous aggregation to form droplets, micelles, bilayers and vesicles within an
aqueous phase, through entropy-driven hydrophobic interactions (Tanford, 1978;
Safran, 1994). While such property seems remote from complex biological func-
tions such as self-replication or information coding, it may underlie some of the
earliest processes that led to the emergence of biological complexity.
The concentration of organic compounds in the water bodies of prebiotic Earth
has been estimated to be approximately 1 micromolar (Lasaga et al., 1971), poten-
tially too low for typical covalent chemical reactions. Yet, because of their spontan-
eous tendency to aggregate, some hydrophobic and amphiphilic molecules would
constitute local phases of highly concentrated organic compounds within a dilute
solution of organic species, and therefore could enhance concentration-dependent
chemical interactions. It should be stressed again that this self-concentration ad-
vantage does not require complex evolution: it is a molecular property that depends
122 D. SEGRÉ ET AL.

Figure 1. Diversity of lipids and of their assemblies. (a) Diversity at the level of single molecules,
examplified by a glycerolipid. The two acyl chains (R1 and R2 ) can vary in terms of length, un-
saturation and chain branching (as, for example, in prokaryotic lipids). The lipid head groups (RH )
are also composed of a variety of molecular species, so that literally thousands of different lipid
species can be produced by combinations of the hydrophobic tails and hydrophilic head groups. (b)
Diversity of non-covalent assemblies compositions. Given a number NG of monomer species, the
number of non-branched polymers of length N that could virtually form is (NG )N (N=60, dashed
line). A similar evaluation of combinatorial diversity can be performed for an assembly composed
of N=60 amphiphiles chosen out of NG possible types. The number of different compositions is
(NG +N-1)!/(N! (NG -1)!)) (continuous line), a remarkably large quantity, although smaller than the
correspondent polymers diversity.
 THE LIPID WORLD 123

solely on physical attributes of the organic molecules.

Vesicles are composed of thousands to millions of individual molecules, held
together by weak non-covalent interactions driven by the hydrophobic forces. The
combined energy of these interactions constitutes an appreciable free energy barrier
that opposes the disruption of the entire assembly. Therefore, the lifetime of a given
amphiphilic assembly may be long, of the order of hours, days or even months
(Kaler et al., 1989). Yet, individual molecular components can enter and leave the
assembly with relative ease, in contrast to the situation of a covalent polymer, in
which monomer exchange is subject to a very high energy of activation. It should
be pointed out, however, that molecular exchanges within amphiphilic assemblies
still have a finite free energy barrier, and may be subject to rate enhancement by
catalytic effects.
The ‘soft-matter’ properties of lipid aggregates, which make them suitable for
rapid material exchange, also render them quite different from crystals, which are
much more rigid in structure, and have defined atomic and molecular coordin-
ates. While bearing considerable similarity to inorganic clay minerals in term of a
potential capacity to underlie the rudiments of life’s origin (Cairns-Smith, 1982)
organic lipid assemblies, similar to other phase-separated organic systems such as
proteinoid microspheres (Fox, 1976), may be much more diverse and chemically
labile, and they also do not come into conflict with the continuity principle (Fry,
1995; Lahav and Nir, 1997)

4. Lipid-Like Amphiphilic Molecules Predated the Origin of Life on the

Earth

While it is reasonable to assume that the first cellular life forms used amphiphilic
molecules for boundary membranes, as well as for other functions, the origin and
diversity of amphiphiles on the early Earth remains to be elucidated. Theoretical
and experimental estimates of prebiotic diversity of organic molecules are avail-
able (Chyba and Sagan, 1992; Morowitz, 1992; Schwartz, 1996), but these do not
directly address the abundance and variety of prebiotic amphiphiles. One of the
reasons is that many of the experiments on prebiotic organic synthesis concentrated
on specific compounds occurring in living systems today, rather than relating to the
entire spectrum of possible primordial molecules.
Several potential prebiotic reactions that could produce simple organic com-
pounds on early Earth have been proposed (Miller and Urey, 1959; Schlesinger
and Miller, 1983; Wächtershäuser, 1988; de Graaf et al., 1995; Huber and Wächter-
shäuser, 1997). Energy sources available to drive such reactions range from volca-
noes and hydrothermal vents to solar photochemistry and pyrite-dependent reduc-
tion. Some of these prebiotic syntheses have been shown to include the formation
of lipid-like amphiphilic molecules, long-chain hydrocarbons and their derivat-
ives (Hargreaves et al., 1977; Leach et al., 1978; Rao et al., 1982). For example,
124 D. SEGRÉ ET AL.

Fischer-Tropsch synthesis which produces organic compounds from simple gasses

such as CO, H2 and CO2 (McCollom et al., 1999) may lead to long chain fatty
acids and fatty alcohols. It should also be stressed that in many cases, in addition to
water-soluble organic compounds that are easily analyzed, insoluble residue (e.g.
‘tar’ or ‘tholins’ (McDonald et al., 1996; Bernstein et al., 1999)) are reported in
prebiotic synthesis simulations (Miller, 1953; Schlesinger and Miller, 1983). These
are likely to contain sparingly soluble hydrophobic and amphiphilic compounds.
A second possible source for prebiotic organic material is delivery to the early
Earth by meteoritic and cometary infall. Anders (Anders, 1989) and Chyba and
Sagan (Chyba and Sagan, 1992) concluded that interplanetary dust particles (IDP)
were the most abundant source of extraterrestrial organic carbon in the late ac-
cretion phase of the early Earth. Comets would contribute smaller amounts, while
carbonaceous meteorites would be a negligible source. The total amount of deliv-
ery over a period of 100 million years of the late Hadean – early Archean eras is
estimated to be on the order of 1016 – 1018 kg (Chyba and Sagan, 1992). For com-
parison, the total organic carbon in the present biosphere is 6 × 1014 kg. It follows
that organic material delivered as extraterrestrial infall was likely to have been a
significant contribution to the organic inventory of the early Earth environment.
Because the parent bodies of carbonaceous meteorites accreted from the same
molecular cloud as comets and IDP (Bernstein et al., 1999) it is possible that
similar organic compounds would be found in all three forms of extraterrestrial
material. One can therefore use the organic components of carbonaceous meteor-
ites as a guide to the organic inventory plausibly available for chemical evolution
on the primitive Earth. Fischer-Tropsch syntheses have been proposed also as one
of the sources of the hydrocarbons found in carbonaceous meteorites (Studier et
al., 1972).
Using the Murchison meteorite as a guide, the most abundant organic material
delivered to the early Earth would be a complex aromatic hydrocarbon polymer
(∼90% of total organic content) followed by a variety of soluble organic acids,
aliphatic and aromatic hydrocarbons, amino acids, ureas, ketones, alcohols, alde-
hydes and purines (Cronin et al., 1988). Of these, both the longer chain mono-
carboxylic acids and certain polycyclic aromatic hydrocarbons (PAHs) derivat-
ives have amphiphilic properties, i.e. solubility in organic solvents and a capa-
city to form aqueous micelles as well as monolayers at the air-water interface.
This suggests the possibility that amphiphilic compounds present in carbonaceous
meteorites could self-assemble into membranous vesicles under appropriate con-
ditions, and in fact such structures have been experimentally demonstrated to be
produced from extracted compounds (Deamer, 1985; Deamer and Pashley, 1989)
(see (Deamer, 1997) for review).
Assuming that a significant amount of the organic substances in cometary and
meteoritic infall survived atmospheric entry, most of the material would presum-
ably enter the oceans and be released over a period of many years. One mechanism
for the release of organic components from extraterrestrial infall is thermal ex-
 THE LIPID WORLD 125

traction (Mautner et al., 1995): over half of the organic compounds present in a
Murchison meteorite sample were shown be released by hydropyrolysis at elevated
temperatures and pressures. For our purposes, we will assume that the primitive
earth, at some point, contained a diverse array of organic chemicals with varying
complexity, including hydrocarbon derivatives with lipid-like properties. While
on a global scale, organic compounds concentrations were probably rather low
(Stribling and Miller, 1987), their self-aggregation properties could lead to local
high concentrations within the assemblies. Given that amphiphilic molecules were
present, it is possible to discuss their catalytic and self-assembly processes, likely
to have played a role in the emergence of primitive precursors of life.

5. Prebiotic Synthesis and Assembly of Membranes

One aspect of early cellular life that is often disregarded is that primordial mem-
branes would need to continuously add amphiphilic components in order to ac-
commodate the growth and replication of the encapsulated macromolecular system
or of the lipid aggregate itself. To understand this process, it is useful to study the
properties of abiotic amphiphilic structures. Bangham et al. (Bangham et al., 1965)
first reported that phospholipids have the capacity to self-assemble into vesicular
structures now called liposomes, in which lipid bilayers act as permeability barriers
to the free diffusion of polar and ionic solutes. Hargreaves et al. (Hargreaves et al.,
1977; Hargreaves and Deamer, 1978) extended these observations to the prebi-
otic environment, asking what minimal properties are required for amphiphiles to
form membranes. It was found that a variety of single-chain amphiphiles can in
fact self-assemble into bilayers under certain conditions. Examples of single chain
amphiphiles include medium- and long-chain monocarboxylic acids (fatty acids),
alcohols, amines, alkyl phosphates, and alkyl sulfates. The minimal chain length
for the microscopic appearance of closed vesicles at physiological temperatures
was 10 carbons. An appropriate balance between charge and hydrophobicity was
also required. For instance, sodium dodecyl sulfate, a 12-carbon alkyl sulfate, is
micellar, but equimolar additions of 1-dodecanol permitted the mixed system to
form very robust membranes.
Hydrocarbon-based amphiphiles would be expected to concentrate at air-water
interfaces. Because sunlight is a primary energy source in the contemporary bio-
sphere, and presumably was equally abundant on the early Earth, it seems reason-
able to ask whether photochemical synthesis of amphiphiles could occur under
simulated prebiotic conditions. Klein and Pilpel (1973) first demonstrated that
amphiphiles can be synthesized by a light-dependent reaction using PAHs as pho-
tosensitizers. This reaction can easily be demonstrated with mixtures of normal
hydrocarbons and PAHs either in the form of aqueous dispersions or as films at
aqueous interfaces (Deamer, 1992; Volkov et al., 1995). In this work, hexadecane
was used as a model aliphatic hydrocarbon, and pyrene, fluoranthene, and anthra-
126 D. SEGRÉ ET AL.

Figure 2. Photocarboxylation of phenanthrene, a model for primitive photosynthetic reactions in-

volving PAH as pigments. Similar reactions occur with other PAH such as naphthalene, anthracene
andpyrene (Tazuke and Ozawa, 1975). Carbon fixation reaction, where COOH derivatives of the PAH
are synthesized.

cene derivatives were used as model polycyclic aromatic hydrocarbons. All of

the latter are present in the Murchison meteorite as free hydrocarbons (Basile et
al., 1978; Cronin et al., 1988). Upon illumination of such mixtures, a variety of
surface-active compounds were rapidly produced, both in air and under anaerobic
conditions. Analysis by gas chromatography – mass spectrometry (GC/MS) indic-
ated that 1-hexadecanol and 2-hexadecanone were major products of the reactions,
together with oxidized derivatives of the PAH. Significantly, if CO2 was present in
the absence of oxygen, a form of carbon fixation reaction can occur (Tazuke and
Ozawa, 1975), in which carboxylic acid derivatives of the PAH are synthesized
(Figure 2).
These results clearly show that UV illumination can drive oxidation of alkanes
to more polar amphiphilic compounds, particularly long chain alcohols, which are
surface-active. Pyrene, fluoranthene, and 2-ethyl anthracene, models for polycyc-
lic aromatic hydrocarbons present in carbonaceous meteorites, are able to act as
photosensitizers. The rate-limiting step does not depend on oxygen concentration,
because carrying out the reaction under aerobic conditions did not markedly in-
crease the rate or yield. It follows that polycyclic aromatic hydrocarbons derived
from meteoritic sources could have served as primitive pigment systems to capture
light energy and drive synthesis of amphiphilic compounds.
The reactions described above do not require a catalyst, but instead are driven
by photochemical activation. In the Lipid World scenario, the lipid assemblies
themselves could provide a weak catalytic activity to enhance the rates of specific
synthesis reactions. One such generalized catalytic activity has been described
by Luisi and co-workers (Blocher et al., 1999; Luisi et al., 1999). One caveat is
that photochemical oxidation of hydrocarbons so far has been shown to produce
only long-chain alcohols and ketones. Neither of these molecules by itself can
form stable membranes, although both are able to contribute to membranes as a
structural component, just as cholesterol does in contemporary cell membranes.
 THE LIPID WORLD 127

There are two challenges for future research in this area. The first is to find
a plausible synthetic pathway for hydrocarbons with 10 or more carbons in their
chains. Such chains must also have modifications, e.g. chain branching, that will
allow them to be fluid at the permissive environmental temperature. Second, reac-
tions must be established by which both polar and ionic character can be added to
the hydrocarbon chains. This might include oxidation of the chains to long-chain
carboxylic acids. Alternatively, chains oxidized to alcohols could conceivably es-
terify with phosphate to produce the necessary ionic characteristics.
For instance, Ourisson and Nakatani (Ourisson and Nakatani, 1994) sugges-
ted that isoprenoids such as isopentenol can condense to form acyclic polyprenol
chains that might serve as hydrophobic moieties in primitive lipids, and showed
that in fact dipolyprenol phosphates are able to self-assemble into bilayer mem-
brane structures. They also note that terpenoid and hopanoid derivatives are com-
monly present as molecular fossils, clearly suggesting that microorganisms utilized
such compounds. Another possible source of longer chain amphiphilic molecules
is the Fischer-Tropsch type synthesis recently reported (McCollom et al., 1999) in
which fatty acids and fatty alcohols are produced. Other investigators have demon-
strated that such compounds, either individually or as mixtures, are able to produce
stable bilayer vesicles (Hargreaves and Deamer, 1978; Walde et al., 1994). This is
exemplified by the stable vesicles produced from a mixture of a carboxylic acid, an
alcohol, and pyrene, a PAH derivative (Figure 3).

6. Lipozymes: Non-covalent Amphiphilic Aggregates with Catalytic

Properties

Several authors have reported cases of chemical reactions whose rates can be en-
hanced by the presence of certain lipid micelles or vesicles (Fendler and Fendler,
1975; Cuccovia et al., 1982; Kust and Rathman, 1995). For example, Cuccovia
et al. have reported a 106 -fold rate enhancement for ester thiolysis catalyzed by
n-heptyl mercaptan in dimethyl di-(n-alkyl) ammonium chloride (Cuccovia et al.,
1982) (Figure 4). A particularly rich compendium of lipid-mediated rate enhance-
ments has been published by Fendler (Fendler and Fendler, 1975) (Figure 6b).
Although most reported cases of micellar catalysis involve hydrolytic reactions,
rate enhancement of synthetic reactions has been demonstrated as well, for example
in non biological surfactants synthesis (Kust and Rathman, 1995) and in liposome-
catalyzed oligomerization of amino-acids (Luisi et al., 1999). In a different realm
of investigation, Luisi has extensively demonstrated the capacity of amphiphile
assemblies to enhance the rate of hydrolytic reactions, in ways that lead to the
autocatalytic growth of such assemblies (Bachmann et al., 1992)
An analogous system is one in which enzyme-catalyzed synthesis of a phos-
pholipid, allows self-assembly into vesicular membranes. Acyltransferase, a com-
mon enzyme in most biological membranes of eukaryotic cells, transfers acyl groups
128 D. SEGRÉ ET AL.

Figure 3. Bilayer vesicles produced by the mixture of amphiphilic compounds isolated from the
Murchison carbonaceous meteorite (A) and by nonanoic acid, a component of the mixture (B) (see
Deamer, 1997).
 THE LIPID WORLD 129

Figure 4. Thiolysis is markedly catalyzed by lipid vesicles. In the presence of DODAC vesicles
(dioctadecyldimethylammonium chloride) thiolysis rates are increased by 106 fold (Cuccovia et al.,
1982).

from acyl-CoA derivatives to lysophospholipids. Since both acyl CoA and lyso-
phosphatidylcholine form micellar dispersions rather than bilayers, the enzyme-
catalyzed reaction reduces the concentration of the detergent-like substrates while
simultaneously producing a double-chained amphiphile. Membranes appear when
the micellar substrates fall below the critical micelle concentration of the mixed
system, with the result that the bilayer-forming double-chain product predominates
(Deamer and Boatman, 1980; Gavino and Deamer, 1983) (Figure 5). The les-
son gleaned from this experiment is that vesicular membranes can assemble from
non-membranous components in a relatively simple one-step enzyme-catalyzed
reaction, which might, under some conditions, be mimicked by the amphiphilic
structures themselves.
Membrane mimetic chemistry (Fendler, 1982) does not necessarily display the
same features classically manifested in enzyme catalysis. Concentration of react-
ants in a dimensionally restricted environment has been held responsible for rate
enhancement in some cases (Cuccovia et al., 1982), while the active role of chem-
ical groups bound to a membrane was considered essential in others (Dugas and
Penney, 1981). In general, both aspects can combine to render lipid aggregates
with an effective rate enhancement capacity. Similar kinetic effects can also be
envisaged for non covalent reactions, e.g. for joining of a free molecule to a previ-
ously formed aggregate or for the transition between two different layers within it
(Zachowsky et al., 1989).
The above reactions may also not strictly conform to the standard attributes of
enzymatic catalysis, such as the degree of turnover, the definition of the transition
state and the stereospecificity of bond formation. Still, in a broader sense, they
could be regarded as manifesting a form of lipid-mediated catalysis.
Can these observations have any implications to the study of the emergence of
life? The earlier studies of living systems have led to the notion that biological
catalysis is mainly attributed to protein enzymes. Subsequently, the discovery that
RNA molecules can act as biological catalysts eliminated this constraint, and ri-
bozymes (Been and Cech, 1988) are now seriously considered as candidate key
components in a primordial ‘RNA world’ (Gesteland et al., 1999). With further
130 D. SEGRÉ ET AL.

Figure 5. Bilayer synthesis from soluble reactants. Acyltransferase catalyzes the transfer of a fatty
acid from acyl CoA derivatives to lysophosphatides, in this case, lysophosphaticylcholine. The
resulting phosphatidylcholine self-assembles into lipid bilayers. (A) Negative stain of the original
dispersion showing mixed micelles of the two detergent-like substrate molecules. (B) Negative
stain of the mixture after 1 hour incubation, showing large numbers of membranous vesicles. (C)
Freeze-fracture image of the lipid vesicles in B. Original magnification 40 000 X.
 THE LIPID WORLD 131

Figure 6a. The different molecules in a GARD system may have varying mutual catalytic enhance-
ment values, shown by the thickness of arrows in the schematic depiction on left and by a gray level
scale in the β matrix representation on the right. The propeties of the mutually catalytic network
depend on the values of the β matrix components, derived from a statistical distribution as shown in
Figure 6b.

extrapolation, it is possible to conceive amphiphilic assemblies as harboring poten-

tial prebiotic catalysis. Based on the above mentioned rate enhancement properties
of amphiphiles as well as on the previously explored concepts of heterogeneous
micelles as structures that bear tangible similarity to globular proteins (Lipowsky,
1995), we introduce here the term ‘lipozymes’ to indicate lipid aggregates that
have a rate enhancement capacity for chemical reactions, either endogenous or
exogenous.

7. Autocatalytic Lipozymes May Self-reproduce

The capacity of a system to transform part of its surrounding environment into its
own similes is considered as one of the fundamental properties of living organisms.
This may be the property of individual molecules, that can replicate autocatalyt-
ically (Ballester and Rebek, 1990; Orgel, 1992; Li and Nicolaou, 1994; Rebek,
1994; Lee et al., 1996; Lifson, 1996). However, the phenomenon of reproduction
is not necessarily confined to single molecules. Assemblies of molecules might
132 D. SEGRÉ ET AL.

Figure 6b. Distribution of catalytic enhancement factors (β), from Fendler’s list of lipid-catalyzed
reactions (Fendler, 1982). The 260 values were taken from tables 12.1, 12.2, 12.3, 12.4, 12.5 of
Fendler’s book, and they correspond to ratios between the catalyzed and the uncatalyzed rates of
different reactions (mostly hydrolytic) that take place in aquaeous micelles. The relative frequency
of these rate enhancement factors (bars) is superimposed here to a theoretical estimate (line) based
on the Receptor Affinity Distribution (RAD) model (Lancet et al., 1993; Lancet et al., 1994; Lancet
et al., 1994) for the statistics of receptor-ligand recognition. The RAD probability function is defined
by three parameters: α, the free energy contribution per elementary subsite interaction; B, the number
of subsites, and S, the number of differnt types of subsite configurations. In the curve plotted here
the parameters used are: B=18, S=13, α/RT=1.

also produce new assemblies by a growth and division process, even if none of
the molecular components is capable of reproduction (Oparin, 1957; Dyson, 1985;
Morowitz, 1992; Dyson, 1999). Such collective self-replication of mutually cata-
lytic species has been demonstrated experimentally (Sievers and Von-Kiedrowski,
1994; Lee et al., 1997), and investigated theoretically (Farmer et al., 1986; Kauff-
man, 1986; Bagley and Farmer, 1991; Jain and Krishna, 1998; Segré and Lancet,
1998; Segré et al., 1998a; Segré et al., 1998b).
Assemblies composed of a single type of amphiphile have been shown to un-
dergo replication-like behavior (Bachmann et al., 1992; Kust and Rathman, 1995),
due to rate-enhancement effects between the lipid micelles and the precursors of
their constituents. These assemblies may be considered as special cases of autocata-
lytic lipozymes. These and similar autopoietic systems (Luisi and Varela, 1989;
Bachmann et al., 1992) have been suggested to represent paradigms of early life,
 THE LIPID WORLD 133

Figure 7a–b. A multicomponent mutually catalytic micelle (a) A schematic view of the proposed
mechanism of amphiphile Graded Autocatalysis Replication Domain (A-GARD). Micelles are com-
posed of amphiphiles with polar head groups (geometric shapes) and hydrophobic tails (sticks). A
chemical reaction that generates an amphiphile Li is catalyzed by the presence of an amphiphile Lj
within the assembly. Numerous such reactions lead to the growth of the micelle. The rate of growth
depends on the summated action of numerous catalytic events, governed by the matrix β, whose
element β ij , when multiplied by the basal rate constants ki and k−i , produces the catalyzed rates.
Upon reaching a critical size, the micelle may split, giving rise to progeny. (b) A starting composition
(i) has most compounds represented scantly and more or less equally. The gradual evolution towards
the establishment of a mutually catalytic network in one of many possible micellar compositions may
result in a highly biased composition (ii), whereby a few species are highly represented, and the rest
are rather low in abundance.

i.e. enclosure and self-reproduction. But they have been argued not to embody
some of the properties essential for initiating an evolutionary process, since they
lack information carriers such as nucleic acids or peptides. Coupling of lipid en-
closure with nucleic acids replication (Chakrabarti et al., 1994; Luisi et al., 1994)
or with peptide oligomerization (Blocher et al., 1999; Luisi et al., 1999) has there-
fore been investigated as a more advanced alternative to the simplest autopoietic
unit.
Here we delineate an alternative scenario, whereby amphiphiles, besides dis-
playing self-assembly and self-reproduction, may also embody the diversity and
134 D. SEGRÉ ET AL.

Figure 7c. Schematic depiction of a scenario merging the results of the chemical kinetics of the
A-GARD model (Segré et al., 1998a; Segré et al., 2000b) with the experimental outcome for
autocatalytic growth of amphiphilic micelles and vesicles (Bachmann et al., 1992; Walde et al.,
1994). It is surmised that if a complex mixture of starting compounds would be used instead of one
amphiphile-generating species, a kinetic behavior as schematically shown in curve (ii) (cf. (Bach-
mann et al., 1992; Walde et al., 1994)) would obtain. This is because a GARD mutually catalytic
network has been demonstrated to behave in a similar manner to a single autocatalyst (Segré et
al., 1998a). Assemblies with compositions deviating considerably from the optimal mutual catalytic
network, including uniform compositions of single species with low autocatalytic capacity, would
display a kinetic behavior as shown in curve (i).

information content needed for the emergence of life. We propose that crucial steps
in the origin of life might have been carried out by lipid-like molecules alone,
potentially prior to the emergence of polynucleic acids and polypeptides. We sug-
gest that heterogeneous autocatalytic lipozymes with defined internal compositions
might have been gradually selected out of an initial highly complex repertoire of
micelles and vesicles formed spontaneously by abiotic processes. This would entail
information content as well as a capacity to undergo natural selection, as delin-
eated below (Figure 7). In the classification proposed by Szathmáry (Szathmáry,
1999), autocatalytic lipozymes would belong to the class of phenotypic replicators,
because of their functional rather than digital inheritance.

8. Statistical Properties of Lipozyme Catalysis: the Importance of Diversity

When the values for the catalytic enhancement factors (kcat /kuncat ) for many dif-
 THE LIPID WORLD 135

ferent lipid-related reactions (Fendler, 1982) are plotted as a frequency histogram

(Figure 6B), certain features are revealed which are often observed in the realm of
random or combinatorial chemistry (Hassan et al., 1996; Borman, 1997; Collins,
1997). Catalysis is in principle a gradative phenomenon, and assays measuring the
catalytic potency of different compounds may display a broad spectrum of values.
General properties of the resulting probability distribution may be inferred both
from experimental data and from theoretical considerations. Along these lines, we
have previously proposed and investigated a model for the distribution of binding
affinities in multireceptor repertoires (Lancet et al., 1993; Lancet et al., 1994; Ros-
enwald, 1998). We have subsequently argued that similar statistical considerations
can be extended to catalysis, and are a viable way for looking at prebiotic scenarios
without questioning the details of each single chemical species (Lancet et al., 1994;
Segré et al., 1998a; Segré and Lancet, 1999). Thus, we demonstrated that the cata-
lytic efficiency values conform to a distributions as manifested in the elements β ij
of the β-matrix (Figure 6A). These may be used as a universal tool for describing
and computationally simulating the kinetic behavior of large molecular repertoires,
as exemplified by the Graded Autocatalysis Replication Domain (GARD) model
(Segré et al., 1998a). In this context, it is significant that the distribution of kinetic
values in Figure 6B indeed generally conforms to the proposed statistical model.
In reality, compounds do not always exert mutual rate enhancements, and some
may manifest inhibitory effects upon the recruitment and/or synthesis of others.
In the model described herein, inhibition is not included, i.e. only positive βij
values are used. Embodiments of the GARD model, in which kinetic inhibition is
addressed, are currently being explored, and it is believed that this will not modify
the central conclusions in a major way.

9. Mutually Catalytic Networks Within Complex Lipozymes

Mutually catalytic networks have been described as primordial metabolism-based

entities which may have preceded the appearance of informational biopolymers
(Dyson, 1982; Kauffman, 1986; Morowitz, 1999). It was further shown that, if a
network of chemical reactions manifests ‘catalytic closure’, it will acquire proper-
ties akin to self-replication (Farmer et al., 1986; Kauffman, 1986). We present here
a case for lipid assemblies as realistic chemical embodiments of such mutually
catalytic networks. Simple lipozymes which behave autocatalytically (Bachmann
et al., 1992; Walde et al., 1994; Luisi et al., 1999) provide an important demon-
stration of the principle: when an amphiphile is formed through an endogenously
catalyzed reaction, vesicle growth becomes a consequence of the catalytic action.
Such principle may be extended to a multi-component scenario. Instead of a single
amphiphile, we consider a large number (NG ) of different components, for example
precursors of vesicle-forming compounds. Chemical conversions could initially
occur spontaneously, giving rise to mixed micellar structures. Subsequently, such
136 D. SEGRÉ ET AL.

micelles might catalyze further reactions in a way that could be described by

a mutually catalytic network formalism. It is conceivable that specific micellar
compositions would display faster growth as compared to others. If the condi-
tions are such that the system is kept far from equilibrium (for example by a
continuous availability of fresh precursors, and the random elimination of some
of the micelles), transients of specific compositions could be kept in a homeostatic
fashion. The compositional bias that ensues would constitute a rudimentary form
of transferable chemical information, as described herein.
It should be borne in mind that in some catalytic networks, certain chemical
species may be detrimental to the system, as they catalyze side reactions that might
disrupt the homeostatic characteristics of the network. It may however be envisaged
that as chemical evolution proceeds, such elements will be selected against, and
‘good’ networks, even if initially rare, will prevail. Alternatively, inhibitory effects
might serve as regulatory mechanisms through negative feedback loops in a net-
work, and could therefore be incorporated fully in the dynamics of a homeostatic
system. In future extensions of the current models these questions could be ad-
dressed quantitatively, and the effects of inhibition on mutually catalytic networks
could be better understood.
Several embodiments of this scenario may be envisaged, in terms of the types
of chemical reactions and the source of free energy. The latter could include pho-
tochemically driven reactions, that might help keeping the system away from equi-
librium. It is important to stress the contrast between two scenarios: (a) an ‘engin-
eered’ chemical systems, in which a specifically designed chemical architecture
underlies the emergence of molecular self-replication, and (b) a naturally occur-
ring mixture of chemicals, where random interactions within large repertoires of
arbitrary organic molecules lead to the appearance of self-reproducing assemblies.
Future experiments could be initiated with a mixture resembling a possible prebi-
otic amphiphile repertoire, and while keeping the system far from equilibrium, the
composition of individual assemblies would be monitored. Such experiments are
still very challenging and could be characterized by utterly slow rate constants of
months and years. For this reason we have resorted to a mathematical modeling ap-
proach, based on computer simulations of the kinetics of self-assembly in complex
molecular mixtures (Segré and Lancet, 1998; Segré and Lancet, 1999; Segré et al.,
2000b) (cf. also (Mayer and Rasmussen, 1998)).
The fluid nature of micellar structures may harbor important advantages in a
prebiotic milieu. While for an evolved polymer structure, such as a folded pro-
tein, the emergence of a new function would require changes in covalent bonds,
a lipozyme might acquire different functions as a consequence of facile joining of
new amphiphiles, or rearrangements of its constituent molecules (Moss et al., 1976;
Sackmann and Feder, 1995). Therefore, the same lipozyme could serve as a poly-
functional endogenously catalytic system and entail the capacity of combinatorially
giving rise to different rate enhancement effects. Consider for example a catalytic
triad composed of the active head groups A, B and C diffusing as amphiphiles
 THE LIPID WORLD 137

within a micelle. While an encounter of all three groups to form the ABC triad
would be much less likely than in a folded protein enzyme, it would still be much
more probable compared to the situation in a dispersed solution. Also, diffusion
encounters that form pairs such as AB or AC could still harbor some catalysis.
This flexibility in the modulation of functions could allow spontaneous ‘screening’
and ‘tuning’ of different catalytic networks with no need for complex evolutionary
processes such as sequence mutations within biopolymers, and a complex coding
apparatus.

10. Compositional Information in Lipozymes

Legitimate concerns about the likelihood of a prebiotic Lipid World scenario could
derive from the seeming discontinuity between catalytic lipid aggregates and present-
day biopolymer-based cellular life. Biochemical information transfer is tradition-
ally identified with sequences of the four letters alphabet of nucleic acid polymers.
Could any information transfer system be envisaged for catalytic lipid aggregates?
We have recently formally analyzed the idea that information may be transferred in
the form of specific molecular compositions (Segré et al., 1997; Segré et al., 1998a;
Segré and Lancet, 1999), an idea initially hinted upon in earlier studies (Oparin,
1957; Morowitz, 1967; Morowitz, 1999). We have now extended this concept of
compositional information to amphiphile assemblies and have provided examples
of how it may be utilized in computer simulations of prebiotic evolution (Segré and
Lancet, 1998; Segré et al., 2000b).
Imagine a population of idiosyncratic assemblies formed randomly out of a
large number of available amphiphiles. If the typical count of molecules N within
a typical assembly is considerably smaller than the number of molecular species
types NG , then practically every assembly will be different. Some of these assem-
blies may harbor superior lipozyme characteristics, e.g. a capacity to catalyze the
recruitment or the synthesis of further amphiphiles. These lipozymes would tend to
increase in size faster and eventually divide through physical forces (Koch, 1985;
Sackmann and Feder, 1995), giving rise to ‘daughter’ assemblies. These might
display varying degrees of similarity to the original composition, depending on
the organization of the catalytic network and on the fate of each molecule upon
splitting. The process described above implies the transfer of a compositional in-
formation from one assembly to its progeny. A capacity for high fidelity transfer
of compositional information could be acquired gradually, after many growth and
division cycles, leading to a process akin to self-reproduction (Segré and Lancet,
1998; Segré et al., 2000b).
It is important to point out that when present-day cells divide, they too transmit
considerable elements of compositional information (including specific gamuts of
lipids, proteins and RNA) which are ‘inherited’ from the mother cell. Higher level
structures may also be inherited epigenetically, as exemplified by the concept of
138 D. SEGRÉ ET AL.

genetic membranes (Cavalier-Smith, 1995; Szathmáry, 1999). This is in addition,

and in parallel to the classical self-replication of DNA sequence information.

11. Evolution of Autocatalytic Lipozymes

The number of compositionally different assemblies that can be formed from a

set of possible amphiphiles can reach very high values (Figure 1B). In a primor-
dial environment, where millions of different chemical species might have been
present, a huge variety of random compositions should be expected, provided that
the assemblies are smaller than a critical value (Segré et al., 1998b; Segré and
Lancet, 1999). In this respect, autocatalytic lipozymes might be viewed as rep-
licators with unlimited hereditary potential (Szathmáry, 1999). The capacity of
‘trying out’ many different combinations, and of searching the fitness landscape for
 THE LIPID WORLD 139

by statistical considerations, that most amphiphiles will provide the highest rate
enhancement to their own kind.
In a further elaboration (Segré, Ben-Eli et al., 1999; Segré et al., 2000b), it
is predicted that within such mutually catalytic assemblies of monomeric am-
phiphiles, dimers and higher oligomers could gradually form. These could replace
some of the monomers, assuming their catalytic roles in the network. We show
that rearrangements of monomers within the oligomers would make the new as-
semblies more successful in propagating their compositions. In addition to this
selective advantage, oligomers could also exhibit higher mutually catalytic poten-
cies because of a ‘combinatorial library’ effect. Our computer model utilizes a
fitness function that balances the thermodynamic price of polymerization with the
advantages of oligomer formation. In preliminary simulations, a defined size of
monomer alphabet was reached and a hierarchy of oligomer ‘words’ crystallized
(Segré et al., 2000b). The large diversity of the molecular components thus gen-
erated would enhance the capacity of a GARD assembly to embody an unlimited
hereditary potential (Szathmáry, 1999), which could eventually lead to the combin-
atorial nucleic acid ‘takeover’, and to the emergence of a genetic code (Szathmáry
and Maynard-Smith, 1995).

12. Summary

The possible role of RNA as the initial carrier of catalytic capacity and genetic
information faces several difficult problems. The monomers of RNA are not readily
synthesized under prebiotic conditions; it is difficult to imagine ways in which they
could be assembled spontaneously into polymeric structures of sufficient complex-
ity; and RNA has no ability to capture energy from the environment and cannot
readily contribute to organized supramolecular structures.
In contrast, amphiphilic molecules were likely to have been relatively abundant
in the prebiotic environment, given that virtually any molecule that contains a long
enough hydrocarbon moiety and a polar group is an amphiphile. Furthermore,
self-assembly of amphiphilic molecules into complex supramolecular structures
is spontaneous. The plausibility that such structures were present in the prebiotic
environment is supported by the occurrence of amphiphilic molecules in carbon-
aceous meteorites and the demonstration that they can assemble into membrane
vesicles. Vesicle structures have the capacity to capture light energy by incor-
poration of pigment molecules that partition into the bilayer, or redox energy by
mediating electron transport reactions across the membrane so that electrochem-
ical potentials are produced. Assemblies of amphiphilic molecules also have the
capacity to act as catalysts for a variety of reactions pertinent to the living state,
including synthetic reactions leading to growth of the bilayer structure from pre-
cursors. Finally, as demonstrated by the GARD model in computer simulations,
140 D. SEGRÉ ET AL.

assemblies of amphiphilic molecules have the capacity to contain compositional

information and to undergo an evolution-like process.
We conclude that a high-probability Lipid World may have preceded a low prob-
ability RNA world. Nonetheless, at some point in prebiotic evolution assemblies
of lipid-like molecules likely began to incorporate additional chemical moieties.
Head groups could diversify, to include monomers of present-day life, such as nuc-
leotides and amino acids. These could oligomerize in the plane of the monolayer
or the bilayer, thus giving rise to entities more akin to contemporary biopolymers,
which would enhance the catalysis and templating capacities within the assemblies.
Some molecules could detach and become soluble, perhaps enclosed in a vesicular
lumen. It is at this stage that a scenario akin to the RNA world could be initiated,
although this does not imply by any mean that RNA chemistry was exclusively
present.
An important goal for future research will be to provide an additional exper-
imental basis for the Lipid World scenario. One could explore ways in which
amphiphilic micelles and vesicles could constitute a suitable microenvironment
in which diverse chemical reactions could occur. This would include rudiment-
ary photosynthesis, as well as the generation of RNA and protein monomers, that
could be synthesized in a chemically active form, followed by oligomerization into
highly catalytic forms. An energy-transducing vesicular system containing tem-
plating molecules as well as catalysts for reactions leading to amphiphile synthesis
would be a clearly recognizable intermediate in the pathway leading to the origin
of life.

Acknowledgments

We are grateful to Helmut Zepik and Charles Apel for critically reading the ma-
nuscript, and to Ora Kedem, Shneior Lifson and Pier Luigi Luisi for inspiring
discussions. Doron Lancet holds the Ralph and Lois Chair in Human Genomics.
Supported by the Crown Human Genome Center, the Krupp foundation, and the
Weizmann Institute Glasberg, Levy, Nathan Brunschwig and Levine funds.

References

Anders, E.: 1989, Pre-Biotic Organic Matter from Comets and Asteroids, Nature 342, 255–257.
Bachmann, P. A., Luisi, P. L. and Lang, J.: 1992, Autocatalytic Self-Replicating Micelles as Models
for Prebiotic Structures, Nature 357, 57–59.
Bagley, R. J. and Farmer, J. D.: 1991, ‘Spontaneous emergence of a metabolism’, in Langton, C. G.,
Taylor, C., Farmer, J. D. and Rasmussen, S. (eds.), Artificial Life II, SFI Studies in the Sciences
of Complexity, Addison-Wesley, X: 93–140.
Ballester, P. and Rebek, J.: 1990, A Self-Replicating System, J. Am. Chem. Soc. 112, 1249–1250.
Bangham, A. D., Standish, M. M. and Watkins, J. C.: 1965, Diffusion of Univalent Ions Across the
Lamellae of Swollen Phospholipids, J. Mol. Biol. 13, 238.
 THE LIPID WORLD 141

Basile, B. P., Middleditch, B. S. and Oró, J.: 1978, Polycyclic Aromatic Hydrocarbons in the
Murchison Meteorite, Org. Geochem. 5, 211–216.
Beaudry, A. A. and Joyce, G. F.: 1992, Directed Evolution of an RNA Enzyme, Science 342, 255–
257.
Been, M. D. and Cech, T. R.: 1988, Science 239, 1412–1414.
Ben-Eli, D.: submitted, ‘From Random Chemistry to Compositional Stationary States in Prebiotic
Non-Covalent Assemblies’„ Rehovot, Weizmann Institute, [Link] thesis.
Bernstein, M. P., Sandford, S. A., Allamandola, L. J., Gillette, J. S., Clemett, S. J. and Zare, R.
N.: 1999, UV Irradiation of Polycyclic Aromatic Hydrocarbons in Ices: Production of Alcohols,
Quinones, and Ethers, Science 283, 1135–1138.
Blocher, M., Liu, D., Walde, P. and Luisi, P. L.: 1999, Liposome-Assisted Selective Polycondensation
of A-Amino Acids and Peptides, ISSOL’99, San Diego, CA, USA, Abstract c1.7.
Borman, S.: 1997, Combinatorial Chemistry, Chemical & Engineering News 24 (February): 43–62.
Cairns-Smith, G.: 1982, Genetic Takeover and the Mineral Origins of Life, Cambridge, UK,
Cambridge University Press.
Cavalier-Smith, T.: 1995, Biodiversity and Evolution, M. Kato and Y. Doi. Tokyo, National Science
Museum Foundation.
Cech, T. R.: 1993, The Efficiency and Versatility of Catalytic RNA: Implications for an RNA world,
Gene 135(1–2), 33–36.
Chakrabarti, A., Breaker, R. R., Joyce, G. F. and Deamer, D. W.: 1994, Production of RNA by a
Polymerase Protein Encapsulated within Phospholipid Vesicles, J. Mol. Evol. 39, 555–559.
Chyba, C. F. and Sagan, C.: 1992, Endogenous Production, Exogenous Delivery and Impact-Shock
Synthesis of Organic Molecules: An Inventory for the Origin of Life, Nature 355, 125–132.
Collins, J.: 1997, ‘Phage display’, in W. H. Moos, M. R. Pavia, A. D. Ellington and B. K. Kay (eds.),
Annual reports in Combinatorial Chemistry and Molecular Diversity, Leiden, 1, 210–262.
Cronin, J. R., Pizzarello, S. and Cruickshank, D. P.: 1988, ‘Organic Matter in Carbonaceous Chon-
drites, Planetary Satellites, Asteroids and Comets’, in J. F. Kerridge and M. S. Matthews (eds.),
Meteorites and the Early Solar System, Tucson AZ, University of Arizona Press, 819–857.
Cuccovia, I. M., Quina, F. H. and Chaimovich, H.: 1982, A Remarkable Enhancement of the Rate of
Ester Thiolysis by Synthetic Amphiphile Vesicles, Tetrahedron 38(7), 917–920.
de Graaf, R. M., Visscher, J. and Schwarz, A. W.: 1995, A Plausibly Prebiotic Synthesis of
Phosphonic Acids, Nature 378, 474–477.
Deamer, D. W.: 1985, Boundary Structures are Formed by Organic Components of the Murchison
Carbonaceous Chondrite, Nature 317, 792–794.
Deamer, D. W.: 1992, Polycyclic Hydrocarbons: Primitive Pigment Systems in the Prebiotic
Environment, Adv. Space Res. 12, 183–189.
Deamer, D. W.: 1997, The First Living Systems: A Bioenergetic Perspective, Microbiol. Molecular
Biology Reviews 61, 239–261.
Deamer, D. W. and Boatman, D. E.: 1980, An Enzymatically Driven Membrane Reconstitution from
Solubilized Components, J. Cell Biol. 84, 461–467.
Deamer, D. W. and Pashley, R. M.: 1989, Amphiphilic Components of Carbonaceous Meteorites,
Origins Life Evol. Biosphere 19, 21–33.
Dugas, H. and Penney, C.: 1981, Bioorganic Chemistry: A Chemical Approach to Enzyme Action,
New York, Springer-Verlag.
Dyson, F.: 1985, Origins of Life, Cambridge, Cambridge University Press.
Dyson, F. J.: 1982, A Model for the Origin of Life, J. Mol. Evol. 18, 344–350.
Dyson, F. J.: 1999, Origins of Life, Cambridge, Cambridge University.
Eigen, M. and Schuster, P.: 1982, Stages of Emerging Life-Five Principles of Early Organization, J.
Mol. Evol. 19(1), 47–61.
142 D. SEGRÉ ET AL.

Esterbauer, H.: 1995, ‘The Chemistry of Oxidation of Lipoproteins’, in C. Rice-Evans and K. R.

Bruckdorfer (eds.), Oxidative Stress, Lipoproteins and Cardiovascular Disfunctions, Portland
Press Research Monograph, No. 6, Portland Press, pp. 55–79.
Farmer, J. D., Kauffman, S. A. and Packard, N. H.: 1986, Autocatalytic Replication of Polymers,
Physica 22D, 50–67.
Fendler, H. J. and Fendler, E. J.: 1975, Catalysis in Micellar and Macromolecular Systems, New
York, Academic Press.
Fendler, J. H.: 1982, Membrane Mimetic Chemistry, New York, Wiley.
Fox, S. W.: 1976, The Evolutionary Significance of Phase-Separated Microsystems, Origin Life Evol.
Biosphere 7, 49–68.
Fry, I.: 1995, Are the Different Hypotheses on the Emergence of Life as Different as they Seem?’,
Biology and Philosophy 10, 389–417.
Gavino, V. C. and Deamer, D. W.: 1983, Purification of Acyl CoA:1-acyl-sn-glycero-3-phos-
phorylcholine Acyltransferase, J. Bioenerg. Biomembr. 14, 513–526.
Gesteland, R. F., Cech, T. R. and Atkins, J. F. (eds.): 1999, The RNA world, Second Edition, Cold
Spring Harbor, New York, Cold Spring Harbor Laboratory Press.
Gilbert, W.: 1986, The RNA world, Nature 319, 618.
Hargreaves, W. R. and Deamer, D. W.: 1978, Liposomes from Ionic, Single-Chain Amphiphiles,
Biochemistry 17, 3759–3768.
Hargreaves, W. R., Mulvihill, S. and Deamer, D. W.: 1977, Synthesis of Phospholipids and
Membranes in Prebiotic Conditions, Nature 266, 78–80.
Hassan, M., Bielawski, J P., Hempel, J. C. and Waldman, H.: 1996, Optimization and Visualization
of Molecular Diversity of Combinatorial Libraries, Mol. Divers. 2(1–2), 64–74.
Huber, C. and Wächtershäuser, G.: 1997, Activated Acetic Acid by Carbon Fixation on (Fe, Ni)S
under Primordial Conditions, Science 276, 245.
Jain, S. and Krishna, S.: 1998, Autocatalytic Sets and the Growth of Complexity in an Evolutionary
Model, Phys. Rev. Lett. 81: 5684–5687.
Joyce, G. F., Schwartz, A. W., Miller, S. L. and Orgel, L. E.: 1987, The Case for an Ancestral Genetic
System Involving Simple Analogues of the Nucleotides, Proc. Natl. Acad. Sci. USA 84, 4398–
4402.
Kaler, W. K., Murthy, A. K., Rodriguez, B. E. and Zasadzinski, J. A. N.: 1989, Sontaneous Vesicle
Formation in Aqueous Mixtures of Single-Tailed Surfactants, Science 245, 1371–1374.
Kauffman, S. A.: 1986, Autocatalytic Sets of Proteins, J. Theor. Biol. 119, 1–24.
Kauffman, S. A.: 1993, The Origins of Order – Self-Organization and Selection in Evolution, Oxford,
Oxford University Press.
Klein, A. E. and Pilpel, N.: 1973, J. Chem. Soc., Faraday I 69, 1729–1736.
Koch, A. L.: 1985, Primeval Cells: Possible Energy-Generating and Cell-Division Mechanisms,
Journal of Molecular Evolution 21, 270–277.
Küppers, B.: 1983, Molecular Theory of Evolution, Berlin-Heidelberg, Springer-Verlag.
Kust, P. R. and Rathman, J. F.: 1995, Synthesis of Surfactants by Micellar Autocatalysis: N,N-
dimethyldodecylamine N-oxide, Langmuir 11, 3007–3012.
Lahav, N. and Nir, S.: 1997, Emergence of Template-and-Sequence-Directed (TSD) Syntheses: I. A
Bio-Geochemical Model, Origins Life Evol. Biosphere 27, 377–395.
Lancet, D., Horovitz, A. and Katchalski-Katzir, E.: 1994, ‘Molecular Recognition in Biology: Models
for Analysis of Protein/Ligand Interactions’, in J.-P. Behr (ed.), Perspectives in Supramolecular
Chemistry, J. Wiley New-York, pp. 25–71.
Lancet, D., Kedem, O. and Pilpel, Y.: 1994, Emergence of Order in Small Autocatalytic Sets Main-
tained far from Equilibrium: Application of Receptor Affinity Distribution (RAD Model), Ber
Bunsenges. Phys. Chem. 98(9), 1166–1169.
 THE LIPID WORLD 143

Lancet, D., Sadovsky, E. and Seidemann, E.: 1993, Probability Model for Molecular Recognition in
Biological Receptor Repertoires: Significance to the Olfactory System, Proc. Natl. Acad. USA.
90, 3715–3719.
Lasaga, A. C., Holland, H. D. and Dwyer, M. J.: 1971, Primordial Oil Slick, Science 174, 53–55.
Leach, W. W., Nooner, D. W. and Oró, J.: 1978, Abiotic Synthesis of Fatty Acids, Origin of Life xx,
113–122.
Lee, D. H., Granja, J. R., Martinez, J. A., Severin, K. and Ghadiri, M. R.: 1996, A Self-Replicating
Peptide, Nature 382, 525–528.
Lee, D. H., Severin, K., Yokobayashi, Y. and Ghadiri, M. R.: 1997, Emergence of Symbiosis in
Peptide Self-Replication Through a Hypercyclic Network, Nature 390, 591–594.
Lehn, J. M.: 1995, Supramolecular Chemistry. Weinheim, VCH.
Li, T. and Nicolaou, K.C.: 1994, Chemical Self-Replication of Palindromic Duplex DNA, Nature
369, 218–221.
Lifson, S.: 1996, On the Crucial Stages in the Origin of Animate Matter, J. Mol. Evol. 44, 1–8.
Lifson, S. and Lifson, H.: 1999, A Model of Prebiotic Replication: Survival of the Fittest Versus
Extinction of the Unfittest, J. Theor. Biol. 199, 425–433.
Lipowsky, R.: 1995, The Morphology of Lipid Membranes, Curr. Opin. Struct. Biol. 5, 531–541.
Luisi, P. L. and Varela, F. J.: 1989, Self-Replicating Micelles – A Chemical Version of a Minimal
Autopoietic System, Origins Life Evol. Biosphere 19, 633–643.
Luisi, P. L., Walde, P. and Oberholzer, T.: 1994, ‘Enzymatic RNA Synthesis in Self-Reproducing
Vesicles: An Approach to the Construction of a Minimal Synthetic Cell’, Ber. Bunsenges. Phys.
Chem. 98, 1160–1165.
Luisi, P. L., Walde, P. and Oberholzer, T.: 1999, Lipid Vesicles as Possible Intermediates in the Origin
of Life, Current Opinions in Colloid & Interface Science 4, 33–39.
Mautner, M. N., Leonard, R. L. and Deamer, D. W.: 1995, Meteorite Organics in Planetary Environ-
ments: Hydrothermal Release, Surface Activity and Microbial Utilization, Planet. Space Sci. 43,
139–147.
Mayer, B. and Rasmussen, S.: 1998, The Lattice Molecular Automaton (LMA): A Simulation System
for Constructive Molecular Dynamics, Internat. J. of Modern Physics C 9, 157–177.
McCollom, T. W., Ritter, G. and Simoneit, B. R. T.: 1999, Lipid Synthesis Under Hydrothermal
Conditions by Fisher-Tropsch-Type Reactions, Orig. Life Evol. Biosphere 29, 153–166.
McDonald, G. D., Whited, L. J., De Ruiter, C., Khare, B. N., Patnaik, A. and Sagan, C.: 1996,
Production and Chemical Analysis of Cometary Ice Tholins, Icarus 122, 107–117.
Miller, S. L.: 1953, A Production of Amino Acids under Possible Primitive Earth Conditions, Science
117, 528–529.
Miller, S. L. and Urey, H. C.: 1959, Organic Compound Synthesis on the Primitive Earth, Science
130, 245–251.
Morowitz, H. J.: 1967, ‘Biological Self-Replicating Systems’, in F. M. Snell (ed.), Progress in
Theoretical Biology, Academic Press. 1, 35–58.
Morowitz, H. J.: 1992, Beginnings of Cellular Life, London, Yale University Press.
Morowitz, H. J.: 1999, A Theory of Biochemical Organization, Metabolic Pathways and Evolution,
Complexity 4, 39–53.
Moss, R. A., Nahas, R. C. and Ramaswami, S.: 1976, ‘Bifunctional Micellar Catalysis’, in K. L.
Mittal (ed.), Micellization, Solubilization, and Microemulsions, New York, Plenum Press, 2, pp.
603–615.
Norris, V. and Raine, D. J.: 1998, A Fission-Fusion Origin for Life, Origins Life Evol. Biosphere 28,
523–537.
Oparin, A. I.: 1957, The Origin of Life on the Earth, London, Oliver and Boyd.
Oparin, A. I., Orlovskii, A. F., Bukhlaeva, V. Y. and Gladilin, K. L.: 1976, Influence of the Enzymatic
Synthesis of Polyadenylic Acid on a Coacervate System, Dokl. Akad. Nauk SSSR 226, 972–974.
Orgel, L. E.: 1992, Molecular Replication, Nature 358, 203–209.
144 D. SEGRÉ ET AL.

Ourisson, G. and Nakatani, Y.: 1994, The Terpenoid Theory of the Origin of Cellular Life: The
Evolution of Terpenoids to Choloesterol, Chemistry & Biology 1, 11–23.
Rao, M., Eichenberg, J. and Oró, J.: 1982, Synthesis of Phosphatidylcholine under Possible Primitive
Earth Conditions, J. Mol. Evol. 18, 196–202.
Rebek, J.: 1994, ‘Synthetic Self-Replicating Molecules’, Scientific American 271, 34.
Rosenwald, S.: 1998, Experimental Testing of the Receptor Affinity Distribution (RAD) model.
Rehovot, Israel, Weizmann Institute, [Link]. thesis.
Sackmann, E. and Feder, T.: 1995, Budding, Fission and Domain Formation in Mixed Lipid Vesicles
Induced by Lateral Phase-Separation and Macromolecular Condensation, Mol. Membr. Biol. 12,
21–28.
Safran, S. A.: 1994, Statistical Thermodynamics of Surfaces, Interfaces, and Membranes, Reading,
MA, Addison-Wesley.
Schlesinger, G. and Miller, S. L.: 1983, Prebiotic Synthesis in Atmospheres Containing CH4 , CO,
and CO2 . I. Amino Acids, J. Mol. Evol. 19, 376–382.
Schwartz, A. W.: 1996, Did Minerals Perform Prebiotic Combinatorial Chemistry?, Chemistry and
Biology 3, 515–518.
Segré, D., Ben-Eli, D. and Lancet, D.: 2000a, Proc. Natl. Acad. Sci. 97, 4112.
Segré, D., Ben-Eli, D. and Lancet, D.: 2000b, in G. Lemarchand and K. Meech (eds.), Bioastronomy
’99 – A New Era in Bioastronomy, ASP Conference Series 213, Kohala Coast, Hawaii, in press.
Segré, D., Ben-Eli, D. and Lancet, D.: 1999, The Prebiotic Transition from Compositional to
Sequence-Based Information, ISSOL ’99, Abstract Book, San Diego, CA, USA.
Segré, D. and Lancet, D.: 1998, ‘Mutually Catalytic Amphiphiles: Simulated Chemical Evolution
and Implications to Exobiology’, in J. Chela-Flores and F. Raulin, Exobiology: Matter, Energy
and Information in the Origin and Evolution of Life in the Universe, Trieste, Italy, Kluwer, pp.
123–131.
Segré, D. and Lancet, D.: 1999, A Statistical Chemistry Approach to the Origin of Life, Chemtracts
– Biochemistry and Molecular Biology 12, 382–397.
Segré, D., Lancet, D., Kedem, O. and Pilpel, Y.: 1998a, Graded Autocatalysis Replication Domain
(GARD): Kinetic Analysis of Self-Replication in Mutually Catalytic Sets, Origins Life Evol.
Biosphere 28, 501–514.
Segré, D., Pilpel, Y., Glusman, G. and Lancet, D.: 1997, ‘Self-Replication and Evolution in Prim-
ordial Mutually Catalytic Sets’, in C. B. Cosmovici, S. Bowyer and D. Werthimer (eds.),
Astronomical and Biochemical Origins and the Search for Life in the Universe, Proceedings
of the 5th International Conference on Bioastronomy, IAU Colloquium N.161, Bologna, Editrice
Compositori, pp. 469–476.
Segré, D., Pilpel, Y. and Lancet, D.: 1998b, Mutual Catalysis in Sets of Prebiotic Organic Molecules:
Evolution Through Computer Simulated Chemical Kinetics, Physica A 249, 558–564.
Shapiro, R.: 1984, The Improbability of Prebiotic Nucleic Acid Synthesis, Origins Life Evol.
Biosphere 14, 565–570.
Sievers, D. and Von-Kiedrowski, G.: 1994, Self-Replication of Complementary Nucleotide-Based
Oligomers, Nature 369, 221–224.
Smith, T. F. and Morowitz, H. J.: 1982, Between History and Physics, J. Mol. Evol. 18, 265–282.
Stribling, R. and Miller, S. L.: 1987, Energy Yields for Hydrogen Cyanide and Formaldehyde: The
HCN and Amino Acid Concentration in the Primitive Ocean, Orig. Life Evol. Biosphere 17,
261–273.
Studier, M. H., Hayatsu, R. and Anders, E.: 1972, Origin of Organic Matter in Early Solar System
– V. Further Studies of Meteoritic Hydrocarbons and a Discussion of Their Origin, Geochim.
Cosmochim. Acta 36, 189–215.
Szathmáry, E.: 1999, Chemes, Genes, Memes: A Revised Classification of Replicators, Lectures on
Mathematics in the Life Sciences, 26, 1–10.
 THE LIPID WORLD 145

Szathmáry, E. and Maynard Smith, J.: 1995, The Major Evolutionary Transitions, Nature 374, 227–
232,
Tanford, C.: 1978, The Hydrophobic Effect and the Organization of Living Matter, Science 200,
1012–1018.
Tazuke, S. and Ozawa, H.: 1975, Photofixation of Carbon Dioxide: Formation of 9,10-
Dihydrophenanthrene 9-Carboxylic Acid from Phenanthrene-amine-carbon Dioxide Systems, J.
Chem. Soc. Chem. Commun. 7, 237–238.
Volkov, A. G., Gugashashdvili, M. I. and Deamer, D. W.: 1995, Energy Conversion at Liquid-Liquid
Interfaces, Electrochimca acta 40, 2849–2868.
Wächtershäuser, G.: 1988, Pyrite Formation, the First Energy Source for Life: A Hypothesis, Syst.
Appl. Microbiol. 10, 207–210.
Walde, P., Goto, A., Monnard, P. A., Wessicken, M. and Luisi, P. L.: 1994, Oparin’s Reactions Re-
visited: Enzymatic Synthesis of Poly(adenylic acid) in Micelles and Self Reproducing Vesicles,
J. Am. Chem. Soc. 116, 7541–7547.
Walde, P., Wick, R., Fresta, M., Mangone, A. and Luisi, P. L.: 1994, Autopoietic Self-Reproduction
of Fatty Acid Vesicles, J. Am. Chem. Soc. 116, 11649–11654.
Wilson, C. and Szostak, J. W.: 1994, In vitro Evolution of a Self-Akylating Ribozyme, Nature 374,
777–782.
Zachowsky, A., Henry, J. P. and Devaux, P. F.: 1989, Control of Transmembrane Lipid Asymmetry
in Chromaffin Granules by an ATP-Dependent Protein, Nature 340, 75–76.