Phil. Trans. R. Soc. B (2011) 366 , 2949–2958 doi:10.1098/rstb.2011.0135
*
Institut des Biomole´cules Max Mousseron ( IBMM ) , UMR 5247, Universite´ Montpellier 1 and 2 – CNRS,
CC DSBC 1706 – Universite´ Montpellier 2, Place Euge`ne Bataillon, 34095 Montpellier Cedex 5, France
Thermodynamics provides an essential approach to understanding how living organisms survive in an organized state despite the second law. Exchanges with the environment constantly produce large amounts of entropy compensating for their own organized state. In addition to this constraint on self-organization, the free energy delivered to the system, in terms of potential, is essential to understand how a complex chemistry based on carbon has emerged. Accordingly, the amount of free energy brought about through discrete events must reach the strength needed to induce chemical changes in which covalent bonds are reorganized. The consequence of this constraint was scrutinized in relation to both the development of a carbon metabolism and that of translation. Amino acyl adenylates involved as aminoacylation intermediates of the latter process reach one of the higher free energy levels found in biochemistry, which may be informative on the range in which energy was exchanged in essential early biochemical processes. The consistency of this range with the amount of energy needed to weaken covalent bonds involving carbon may not be accidental but the consequence of the abovementioned thermodynamic constraints. This could be useful in building scenarios for the emergence and early development of translation.
Keywords: carbon metabolism; aminoacyl adenylates; aminoacylation; chemiosmosis; methanogenesis; heterotrophic hypothesis
1. INTRODUCTION
Cells are characterized by the presence of genetic information, of a metabolism and of compartments; there has been an ongoing debate on the features
that came first [ 1
–
3
]. This debate has also been com-
plicated by an excessive simplification of positions [ 4 ].
But the simultaneous requirement of two or all of the sub-systems can be considered a likely possibility as
well [ 5 ]. Independent of that choice, a metabolic con-
tribution cannot be precluded as the presence of genetic material or that of membrane components requires synthetic pathways supporting, for example, a preparatory metabolism variant of the genetic polymer
first option [ 4 ]. Therefore, the chemical free energy
released or used in these pathways represents an essential component in the majority of the hypotheses on the origin of life. This is obviously the case for the chemoautotrophic hypothesis
[ 6 , 7
], in which the formation of organic matter relies on mineral sources of energy. In the earlier heterotrophic hypothesis for
the origin of life [ 8 , 9
], the energy brought about by organic molecules has been considered as an essential factor in addition to their role as building blocks. It is likely that some of these molecules have constituted the starting material yielding some of the high-energy intermediates (thioesters, acyl phosphates, acyl adenylates, phosphoenol pyruvate, aminoacyl adenylates) that are nowadays involved in the main biochemical
* Author for correspondence ( rpascal@univ-montp2.fr
).
One contribution of 17 to a Discussion Meeting Issue ‘The chemical origins of life and its early evolution’.
2949 pathways. Building an inventory of processes capable of providing energy to early living organisms is then a major goal in the origin of life studies. Considering thermodynamic capabilities of inorganic reactions has commonly been used in the analysis of hydrothermal pathways, mostly performed by geochemists, but it is less spontaneous for organic chemists who are used to the fact that thermodynamic constraints are often of little help in the analysis of organic reactions.
This overlook of thermodynamic constraints has chemical grounds. It is related to the tendency of carbon to form covalent bonds with other elements and then to the fact that organic reactions usually require these bonds to be cleaved to some extent at the transition state, which raises high kinetic barriers that partly reflect the strength of the full bond and allow reaction products to be determined in many instances by kinetics rather than thermodynamics. It is worth noticing that this mere importance of kinetic barriers in determining organic reactivity has also been proposed as a key factor to induce self-organization as it is a prerequisite for efficient catalytic or autocatalytic pathways [
10
,
11
]. It may eventually be helpful in understanding why life, defined as a state of matter in which stability is governed by dynamics
[ 12 ], is based on carbon. However, even if many
organic reactions are under kinetic control, the fact remains that thermodynamic constraints apply and can be crucial in identifying prebiotic pathways. Moreover, life presents very specific thermodynamic features useful in understanding the specificity of
living organisms [ 13 ]. Lokta [ 14 ] initially pointed
out this peculiarity and considered that
This journal is q
2011 The Royal Society

2950 R. Pascal & L. Boiteau Review. Energy flows thermodynamics is not sufficient to drive evolution, and proposed to set natural selection as an additional physical principle [
15
] in a way similar to the description proposed by Pross
[ 12 ].
Anyway, the developments of thermodynamics have shown that complex dissipative structures can develop in systems
maintained away from equilibrium [ 16 ] and flows of
energy that are essential to living organisms may also have played a role in the emergence of life [
17
].
An inventory of processes leading to the presence of organic matter on the primitive Earth has been made in the past, including synthesis driven by ultraviolet light or electric discharges in a reducing atmosphere
and the delivery by impacts [ 18 ]. Since, at that time,
a non-reducing atmosphere was considered as likely, an emphasis was made on the latter process. But, new pieces of information have been published that suggest an active atmospheric synthesis owing to a more reducing character than previously believed
[ 19 ]. Moreover, even neutral atmospheres have been
reconsidered and have been shown to produce an unexpected yield of amino acids when submitted to
electric discharges [ 20
]. Other sources of organic matter have been proposed as the result of redox gradients that are present at the interface of the mantle with oceans in hydrothermal systems. It is obvious that the kind of chemical intermediates and products that can be obtained would depend on the process involved. It is then appealing to compare the energetic properties of the most ancient biochemical intermediates to the possibilities of different prebiotic pathways.
The existence of a driving force for the transition from non-living to living is suggested by thermodynamic considerations. Without underestimating the role of natural selection as a trigger for life’s origin, it is likely that energy sources could be an important factor in selecting early biochemical pathways and, on the other hand, by analysing the potential dynamics and the thermodynamics of early biochemistry, we may infer some conclusions on both the nature of free energy sources capable of driving the emergence of life as we know it and how subsequent processes were able to guide evolution. Therefore, this work is aimed at determining the kind of chemistry that could arise from these energy sources and their respective limitations. We report here the progress made in these investigations, which led us to several conclusions supporting the heterotrophic nature of early living organisms and provided new insight into the chemistry that led to the emergence of the translation apparatus and the emergence of nucleoside triphosphates as energy carriers. The suggested picture is in agreement with the principle of evolutionary continuity in which random changes were selected as a result of their ability to give a selective advantage to the corresponding organisms at every stage of evolution from the very beginning.
2. THE IMPORTANCE OF ENERGY IN LIFE
EMERGENCE
The idea that physical laws provide a driving force for the emergence of dissipative structures , which near bifurcation points depend on the behaviour of
Phil. Trans. R. Soc. B (2011) energy chemical carriers ( Δ G c photons…
)
Δ S <0 high-energy metabolites
Δ G m inactivated products/heat
Δ
S 0 living organism/ chemical system
Scheme 1. Self-organization in chemical systems induced by a flow of energy and a local decrease in entropy compensated by an overall increase. The formation of high-energy metabolites/intermediates is dependent upon the supply of energy (as chemical carriers or under physical forms) into the system and on its degradation into high entropy forms.
For closed systems, at the steady state, the same amounts of energy are entering and leaving the system but energy is released in a degenerate form characterized by high entropy.
Self-organization in this kind of system can depend upon the occurrence of autocatalytic pathways, self-replication processes or metabolic cycles increasing the rate of decay of high-energy reactants and constituting dissipative structures.
Except unstable intermediates with very short lifetimes, the free energy potential of the metabolites accumulating in significant concentration is usually limited by the energy of chemical carriers (reactants): 2 D G m
, – D G c
, demonstrating that the incoming energy source determines the energy of proto-metabolic intermediates. Generating metabolites with higher energy potentials would require the presence of a thermodynamic machine (such as a heat engine capable of extracting work or heat from high-entropy energy forms using an entropy sink), the complexity of which is highly unlikely for primitive living forms. Therefore, the energy flow constitutes the driving force and the nature of the energy source is likely to have decisive consequences on the kind of metabolism accessible from the self-organization process.
fluctuations involving a limited number of events occurring at the molecular scale rather than on macro-
scopic properties [ 16 ], has emerged in the second half
of last century. At a sufficient distance from equilibrium, the evolution of a system may not obey linear laws but follow nonlinear dynamics characterized by the formation of these dissipative structures. Most chemical and biochemical reactions (except those having very low barriers or when reactants are at concentrations very close to equilibrium) satisfy the criteria of distance from equilibrium [
16
] so that, in chemistry, the occurrence of dissipative structures is mainly governed by the availability of feedback mechanisms able to amplify fluctuations, which are usually the result of autocatalysis or self-reproduction of molecules. As life is rooted in the properties of molecules, this view means that amplification processes are needed to reach the micro- or macroscopic size of living organisms or cells starting from singularities occurring at scales smaller by orders of magnitude.
These processes need the system to be maintained
away from equilibrium by an energy flow ( scheme 1
).
In addition to these requirements on the dynamics of the processes, it must also be considered that the nature of the system involved, composed of organic molecules, has decisive consequences. Indeed, even in the absence of oxygen, the large majority of

metabolites and biopolymers are generally not thermo-
dynamically stable [ 21 ] and, depending on the
conditions, can be converted into inorganic carbon,
CH
4 or CO
2
. As a result, driving the system towards equilibrium (for instance, by heating and then increasing the rates of irreversible processes) will not ultimately give a complex mixture of organic molecules as products. Then, for both thermodynamic and dynamic reasons, self-organization in biomolecular systems requires a flow of energy [
13
,
14
]. In real irreversible systems, energy flows from sub-systems having high potential towards those of low potential and is usually released as heat and radiated at low frequencies. As the life form that we know on the Earth is mostly based on carbon chemistry and covalent bonds, the amount of free energy needed to induce self-organization in chemical systems very likely approaches the order of magnitude needed to destabilize covalent bonds (representing at least a significant fraction of the corresponding bonding free energy, e.g.
350 kJ mol
2 1 for a C – C bond). This value is consistent with the free energy potential of many high-energy biochemical metabolites, 50 – 70 kJ mol
2 1
[
11 ], meaning
that energy must have been brought about at a molecular scale by carriers capable of delivering quanta of energy in the corresponding range to trigger the emergence of life (
scheme 1
).
1
This free energy requirement is fulfilled by electromagnetic radiations with wavelengths close to the range of visible light emitted by the Sun, a blackbody heated at ca 6000 K. But, as previously mentioned by others [
17
], a repeated absorption of photons by early biological systems seems impracticable, though more or less direct mechanisms can be responsible for the photochemical generation of high-energy molecules. As a matter of fact, ions, atoms or radicals can be generated by photolysis, and recombination may lead to activated molecules with a sufficient lifetime
(isolated in a free energy well by high enough kinetic barriers) so that they can reach the ground or a location in which a more complex chemistry can take place [
11
]. Considering a range of lifetimes from 1 s to 1 year leads to 85 – 120 kJ mol
2 1 height for the kinetic barrier, which also matches the above-deduced amount corresponding to a fraction of the free energy of a covalent bond.
( a ) The energy quality requirements for life
In any discussion of metabolisms available for the origin of life, it is essential to take into account that a consequence of the second law is that free energy requirements cannot be fully understood at a macroscopic—extensive—scale, but are associated with discrete chemical events occurring at a molecular scale. As a matter of fact, providing a determined amount of energy either as a single energy carrier or as an identical quantity shared by a huge number of molecules is not equivalent. Moreover, this non-equivalence corresponds actually to an entropy production that is precisely what is needed for compensating the loss of entropy associated with self-organization. It has then been suggested that a minimal amount of free energy per chemical event, called the
Phil. Trans. R. Soc. B (2011) minimum
Review. Energy flows R. Pascal & L. Boiteau 2951 biological energy quantum , is needed to support life
[ 23 ]. The relevance of a minimal quantum of free
energy in biology has been debated and it has been proposed that in certain environments, this need can be overcome by the cooperation of different forms of
life based on complementary metabolisms [ 24 ]. But
complex associations of different species based on different metabolisms are unlikely for emerging living forms so that the existence of a minimum early biochemical energy quantum remains relevant in this
context [ 11 ]. In fact, the formation of one adenosine
triphosphate (ATP) molecule from a proton gradient is in principle capable of taking advantage of a minimum amount of energy corresponding to ca onethird of the free energy content of ATP (i.e.
ca
20 kJ mol
2 1 of energy
[ 23
as more than three proton translocations are needed to build one ATP from adenosine diphosphate (ADP) and inorganic phosphate). This amount has been called a biological minimal quantum
]; it is related to the requirement for a dis-
crete minimum value in energy delivered by unit event below which function is not possible. As far as early biochemical pathways are concerned, it is then reasonable to consider that the minimum amount of energy possible to integrate in the metabolism was much higher for less evolved organisms.
( b ) Energy and carbon metabolism
The indication that aldehydes are essential components for the origin of life, unambiguous from their role in the formose reaction [
25
] or in the formation of amino acids [
26
], can also be appreciated by an analysis of potential carbon metabolisms
[ 27 , 28 ]. Actually, the first attempt to introduce ther-
modynamic data in the discussion of possible early carbon metabolisms has been made by Urey [
29
]; the corresponding data are quoted in the top left part of
table 1 . The first conclusion inferred from
those data is the instability of organics in the presence of oxygen (second column), which is the driving force for the respiratory metabolism of contemporary living organisms. But, under anoxic condition, organic molecules become much less unstable. This is consistent with the formation of organics in the reduced environments found in the interstellar medium, on other bodies of the solar system (Titan) or even in hydrothermal systems present in mid-ocean ridges.
Considering only one-carbon derivatives (CO
2
, formic acid, formaldehyde, methanol and methane), as in the original work of Urey, leads to the conclusion that methanogenesis is the easiest pathway for the autotrophic generation of energy from inorganic sources. Though it requires (directly or indirectly) the reduction of carbon dioxide in four steps, only the last two are strongly exergonic, leaving unresolved the question of the formation of significant amounts of formaldehyde (and sugars). But expanding Urey’s work by taking into account a more complex carbon chemistry, including carbohydrate ( – CHOH – ), aliphatic ( – CH
2
– ) carbons, as well as graphite (inor-
ganic carbon) ( table 1 ), it appears that several other
processes are capable of providing energy to the metabolism. A chemistry based on exergonic reactions

2952 R. Pascal & L. Boiteau Review. Energy flows
Table 1. Standard free energy of reactions ( D r
G 8 in kJ mol
2 1
) of simple carbon derivatives susceptible to giving rise to early
metabolisms. Values in parentheses are published by Urey in 1952 [ 29 ].
compound oxidation with
O
2 to CO
2 and H
2
O reduction with
H
2 to CH
4 and H
2
O disproportion to
CH
4
, CO
2
H
2
O and reduction with
H
2 to -CH
2
a and H
2
O conversion to
-CHOHb,c
CO
2
(g)
HCOOH (l)
CH
2
O (g)
CH
3
OH (l)
CH
4
(g)
– CHOH – (l) b
– CH
2
– (l) a
C (graphite)
0 2 130 ( 2 131) 0
2 270 ( 2 296) 2 163 ( 2 189) 2 66 ( 2 91)
2 529 ( 2 522) 2 185 ( 2 178) 2 120 ( 2 113)
2 702 ( 2 707) 2 121 ( 2 126) 2 88 ( 2 93)
2 818 ( 2 818) 0
2
2
2
480
630
394
2
2
2
136
49
50
0
2 71
þ 4 d
þ 15 d
2 74
2 107
2 129
2 65 n.a.
2 80
0
þ 6
þ 6
2 27
2 49
þ 15
þ 136
0
þ 80
þ 86 a b
As – CHOH – group in glycerol compared with ethylene glycol, D f
G 8 ( – CHOH– ) ¼ 2 151 kJ mol c
As a methylene group in pentanoic acid compared with butanoic acid, D f
G 8 ( – CH
2 and/or H
2
– ) ¼ 2 1 kJ mol
2 1
2 1
.
.
O as reagents or products as necessary.
d
By hydration, hydrogenation or dehydrogenation involving H
2
Water is consumed (instead of being produced).
carbonization to graphite, H
2
O and H
2 n.a.
n.a.
2 135
2 70
þ 50
2 86
þ 1
0 can be built from carbohydrates ( – CHOH– ) n
, whereas inorganic carbon and hydrocarbons appear to constitute thermodynamic ends (no exergonic reaction possible except hydrogenation under strongly reducing conditions). Although they behave as energetic biomolecules through respiratory metabolism, lipids made of aliphatic hydrocarbons ( – CH
2
– ) n are non-reactive end-products in anoxic environments, which is consistent with the long-term stability of oil as well as coal in sediments.
In addition to this nature of waste in an energy producing pathway, fatty acids present the strong advantage of selfaggregating into lipid bilayers and constitute boundaries
for early living entities [ 30 , 31 ], whereas carbonization or
methanogenesis leads to products that are useless for living organisms. The potency of carbohydrate transformations has already been emphasized in the
literature [ 24 , 27 , 28
,
32
] and is clearly shown in
table 1
, as well as the difficulty in their synthesis, strongly favourable starting from formaldehyde only. The need of abundant abiotic sources of formaldehyde, or an equivalent capable of generating carbohydrates, may then be considered as an energetic constraint on the origin of life, which is consistent with an atmospheric or extraterrestrial delivery to the early Earth [
33
]. The fact that carbonyl carbon is reduced more easily than it
is formed from oxidized precursors ( table 1
) confirms that the formation of substantial amounts of formaldehyde and formose products by reduction of CO
2 requires very selective pathways that may be accessible to living organisms having developed very specific catalysts, but may be difficult for early life in the environment of hydrothermal systems. This inadequacy is also consistent with the well-known high kinetic reactivity of the carbonyl group [
34
], which is the basis of many organic synthesis routes but which also increases the susceptibility of this group to reducing agents including hydride.
( c ) Heterotrophy
The line of reasoning developed above suggests that the most important organic constituents of the first living entities were not formed from CO
2 by a direct reduction that could hardly provide biochemical
Phil. Trans. R. Soc. B (2011) metabolites and free energy simultaneously. Before examining the possibility that the anabolism of autotrophic early living organisms may also have coupled the thermodynamically unfavourable formation of important biochemical constituents with the consumption of chemical energy sources, we can emphasize that this mere observation is consistent
with the heterotrophic hypothesis [ 8 , 9
]. The metabolism of early living entities could then have been dependent on the transformation of high-energy organic molecules formed in the environment that could be converted into metabolic end-products, through processes able, at the same time, to provide the amount of energy needed for the metabolism.
The class of molecules formed in this way includes low-molecular weight reactants with double or triple bonds (formaldehyde, hydrogen cyanide, cyanic acid, urea, cyanamide, cyanoacetylene, etc.) that have
been observed in the interstellar medium [ 35 ]. For
instance, it has been demonstrated for years that molecules of this kind can sustain a very rich prebiotic chemistry leading to the formation of building blocks
such as sugars [ 25 ] and amino acids [ 26
]. But a much more sophisticated chemistry has been observed from these intermediates: they are additionally able to
activate phosphate [ 36 ], to phosphorylate nucleosides
and activate nucleotides [
37
], or even to directly drive
the formation of activated nucleotides [ 38 ]. Although,
they are found in interstellar media, it is unlikely that high-energy intermediates survived for long periods in the parent bodies of meteorites, so that their source has to be found close to the location where
life emerged. Photolysis, lightning in reducing [ 39 ]
or neutral atmospheres [
20
], or impacts [
18
] have been proposed as likely processes for the formation of high-energy intermediates. Activated species built in this way result from the recombination of atoms, ions and radicals transiently formed during events occurring during short timescales followed by associat-
ive processes leading to metastable molecules [ 40 ]
with lifetimes sufficient to reach locations where self-
organization can take place [ 11 ]. Ultraviolet light is
in principle sufficient to break most components of the atmosphere although many reactions require

short wavelengths. Other processes such as impact and lightning (leading to temperatures locally exceeding 10 000 K) are also able to break most atmospheric molecules through transient heating and lead subsequently to efficient recombination processes occurring at lower temperatures. Molecules formed in this way are, in principle, able to supply energy, though their role has been challenged because they could hardly accumulate in favourable locations. But it seems difficult to find alternative pathways that could deliver organic species capable of performing this task. This conclusion does not mean that hydrothermal formation of organic molecules was devoid of any utility for the origin of life but that other sources leading to high-energy intermediates were also needed.
Review. Energy flows R. Pascal & L. Boiteau 2953
E
Δ
G
ATP
[H + ]
ADP
[H
+
] e – e
–
[H + ] e –
[H + ]
[H
+
]
[H
+
]
[H + ]
[H + ]
( d ) Adenosine triphosphate and chemiosmosis
Owing to its prominence in the metabolism of extant living organisms, chemiosmosis is a potential pathway for building the ATP needed for the survival of early living organisms. The synthesis of ATP (or any other alternative exchangeable energy carrier) through a pathway independent of the formation of organic carbon would be advantageous because of its subsequent use in the development of an organic chemistry starting from inactivated carbon sources and reducing agents or sources of inactivated organic matter (brought to the Earth by external delivery or formed in hydrothermal vents). This kind of metabolism, constituting an anabolism, might have been based on a redox disequilibrium present in hydrothermal environments. It is often considered that the reducing power of rocks reaching the surface in ocean ridges may have generated a redox gradient at the contact with the ocean, which has a more oxidized
state (for recent examples, see [ 41 , 42 ]). Chemiosmo-
sis may then constitute a unique pathway to collect the energy available from the redox processes that are available from the hydrothermal environments, and
ATP produced in this way could have provided energy
to anabolic processes. This sophisticated process [ 43 ]
proceeds in two well-separated stages with a transient storage of energy under the physical form of a proton concentration gradient (
scheme 2
).
In the first stage, the proton concentration gradient across the membrane is transiently generated through an electron transfer cascade coupling the translocation of as many protons as possible across the membrane to the transfer of electrons between redox centres of the electron transfer chain. In a second stage, ATP is synthesized at the expense of the pH gradient using membrane ATP-synthase, which couples proton translocation back to the cytoplasmic compartment to ATP synthesis. Peter Mitchell proposed the chemiosmotic theory [
43
] to explain how ATP is formed by respiration, through oxidative phosphorylation, which is now universally accepted. Photophosphorylation, the process by which ATP is synthesized in photosynthetic organisms, proceeds in a very similar way as well as other redox metabolisms developed in bacteria and
archaea [ 44 ] that involve final electron acceptors
other than oxygen and donors other than the redox cofactors generated by glycolysis (NADH and
Phil. Trans. R. Soc. B (2011)
Scheme 2. The principle of generation of ATP through a chemiosmosis pathway. The electron transfer cascade associated with redox reactions is coupled to the translocation of protons towards the exterior of the cell, which generates a proton concentration gradient. This gradient is used in a second stage by membrane ATP-synthase (the proton pump) to synthesize ATP by coupling phosphorylation of
ADP with the re-entry of H
þ in the cytoplasmic compartment.
FADH2) [
44
]. All these processes have very acute requirements to be efficient: firstly, a membrane that is impermeable to protons, secondly, a series of membrane proteins capable of coupling electron transfers to proton translocations, and finally, membrane ATPsynthase that is by itself a very complex cellular machine
[ 45
]. Performing only one of these tasks would hardly be achievable by an emerging living organism, which raises a series of unanswered questions:
— How could an impermeable membrane be formed whereas fatty acid-made membranes are notoriously
leaky [ 46
], except in the absence of alkali-metal
cations or other permeable cations [ 47 ]?
— How could the free energy from several events, each involving the translocation of a proton through the membrane, be harvested to build an
ATP molecule from ADP and inorganic phosphate without a highly evolved molecular machine?
Otherwise, a pH gradient value exceeding the possibilities of simple vesicles would be needed to build ATP in a single step and maintain its concentration to a significant level in a primitive cell. The fact that membrane ATP-synthases are complex proteins made of several subunits renders the problem even more complex as, in this hypothesis,
ATP synthesis should have evolved before translation, which requires this task to be performed by a ribozyme or an unknown catalyst.
It is then likely that chemiosmosis emerged later than translation, rendering highly speculative the hypotheses supporting that early living organisms exploited inorganic electron donors for their metabolism [
24
]. The reasons why evolution led cells to use this complex mechanism are far from being fully understood.
However, compared with a direct conversion, the biochemical solution presents decisive advantages. The first one may be the high rates of proton transfers,

2954 R. Pascal & L. Boiteau Review. Energy flows
O
N
H
R 1
O
O – H
3
N
R 2
O
N
H
K pep
±H
2
O
O
N
H
R 1
O
O
H
N
R 2
N
H
Scheme 3. Formation of a peptide bond from two peptide segments.
which can reach the diffusion limit in aqueous solution
[ 48 ] and are potentially capable of being concerted [ 49 ]
with electron transfers, which may be used to get energy from transient conformational states of membrane proteins involved in the electron transfer cascade. A second one may be related to the possibility of adaptation of a wide range of redox potentials to the generation of a single universal energy currency, namely ATP, which can then be used to bring about energy in a diversity of endergonic metabolic transformations. Both oxidative phosphorylation and photophosphorylation can then be considered as unlikely systems to get energy for primitive cells.
3. FREE ENERGY REQUIREMENTS FOR THE
EMERGENCE OF TRANSLATION
(
The formation of an amide bond from a peptide segment with a free C-terminal carboxylic acid and the
N-terminal amine of a second one (
0.1 M
( a b
)
)
Bioenergetics of peptide formation
2 1
) at moderate pH [
A driving force for the evolution of translation?
Harry Noller [ 54
] pointed out that finding a driving force supporting a scenario for the development of translation in an RNA world is not easy. The emergence of enzymes that stabilize the intermediates of the translation process is especially puzzling in this discussion as translation could not develop from pre-existing folded proteins. For instance, understanding why adenylates have been selected as intermediates cannot be easily understood as the equilibrium for their formation from the ATP in solution is highly unfavourable ([aa-
AMP][PP i
]/[ATP][AA] ¼ 3.5
10
2 7
[ 55
]), and they would not be formed in significant equilibrium
Phil. Trans. R. Soc. B (2011)
50 ].
scheme 3
ditions, moderately activated esters such as aminoacyltransfer RNAs (aa-tRNA, D G 8
0 ¼ 2 35 kJ mol
2 1
[
51
]) are fully adapted to peptide biosynthesis provided that the aminolysis step (peptide bond formation) can take advantage of a catalyst capable of approximating reactants, a role of entropy trap fulfilled by the ribosome
[ 52
]. But, aa-tRNAs are not the only intermediates of protein biosynthesis as amino acids are universally activated as adenylates (aa-AMP) before aminoacylation of tRNA. The development of aminoacyl-tRNA synthetases (aaRS) was one of the first processes needed to establish an RNA– protein world and the corresponding genes are among the most conserved ones [
53
].
) is not very far from thermodynamic equilibrium ( K pep
Therefore, in water and under physiological conconcentration from ATP unless a specific stabilization is present as in the active site of aaRS. But the development of the translation machinery required the availability of a source of energy capable of promoting the formation of aa-tRNA (or any of its precursors as
for example aminoacylated RNA mini-helices [ 56 ]).
The fact that the free energy content of aa-AMP is far beyond that of ATP by as much as ca 37 kJ mol
2 1
[ 55 ]
suggests that ATP was not the early activating agent for amino acids [
50
]. A first explanation may be that
ATP was settled as a universal currency, whereas tRNA amino-acid esters formed in a completely independent way and had already been selected as intermediates. A pathway leading from ATP to aatRNA via aa-AMP may have afterwards improved this system as a result of an unexpected event in evolution
(exaptation), for example through catalytic promiscuity.
Otherwise, the aaRS function may have emerged as ribozymes in an RNA world with both an adenylation domain capable of catalysing the formation of aa-
AMP (the thermodynamically favourable reverse reaction of pyrophosphate with adenylate to give ATP is not spontaneous in water) while stabilizing this intermediate and abilities in the aminoacylation reaction.
Additionally, these catalysts should have to hinder the spontaneous reaction with carbon dioxide known to consume aminoacyl adenylates and related mixed
anhydrides [ 57 ]. The alternative is to consider that
adenylates (or similar mixed anhydrides formed from nucleotides) were formed abiotically and available for early living organisms; in this view, the role of ATP in
adenylate formation must have emerged later [ 57
].
This view is supported by the fact that amino acid Ncarboxy anhydrides (NCA,
scheme 4
) can be abiotically formed [
50 , 58 , 59 ] and have been shown to spontaneou-
sly react to give mixed anhydride with phosphate
and nucleotides [ 60 , 61
] with no need of enzyme at neutral pH.
It is important to notice that as soon as NCA are considered as prebiotically relevant, peptides must have been formed spontaneously as NCA polymerize easily.
Living organisms depending only on RNA for genetic continuity and lacking encoded proteins (in agreement with the RNA world hypothesis) would only have to improve this process to take advantage of the formation of non-coded random peptides, without the need of specifically built machinery. It must be emphasized that, like aa-AMP, the large majority of carboxy-activated amino acids, including thioesters, would have
been converted into NCA ( scheme 4
) [
62
] because of the reaction of the amino group with CO
2 of the abundance of CO
2 as a result and bicarbonate on the early

R 1
O
OH
NH
2
R 1
O
X
NH
2 activating agent
CO
2
O
CO
2
R 1
N
H
O
O
Review. Energy flows R. Pascal & L. Boiteau 2955 fast
N
H
R
1
H
N
O
O
OH
R 1
CO
2 fast slow
HN X
O
R
1
O
O
O P O
OH
NH
2
R 2
Scheme 4. NCA can be formed starting from sufficiently activated amino acid derivatives including mixed anhydrides with phosphate and nucleotides. Peptide formation from a carboxy activated amino acid usually takes place through the intermediacy of NCA.
O
O
B
O
O
O
O
P
OH
O
B
OH
NCA
O
O
B
H
2
N
O
O
P
O
O
OH
O
R
1
O
HO
P
O
O
O
O
O
O
R
B
NH
1
2
O
P
O O
O
Scheme 5. Intramolecular aminoacyl transfer from nucleotide phosphate mixed anhydrides occurring from mixed anhydrides of 3
0
-phosphorylated nucleosides. Spontaneous aminoacylation of nucleotides by NCA is taking place as well as cyclization.
Earth [
63 ]. Moreover, additional experiments [ 57 ] have
shown that aa-AMP like other amino acid phosphatemixed anhydrides do not give rise to efficient peptide chain formation except in the presence of carbon dioxide demonstrating that peptides are mainly formed through the intermediacy of NCA, which removes any evolutionary advantage of adenylates as peptide precursors. Actually, living organisms have probably developed mechanisms to avoid the formation of NCA in the cell in order to prevent uncontrolled aminoacylation of proteins so that NCA have presently no identified role in cells. Finally, NCA have also demonstrated abilities in aminoacylating nucleotides by reaction with a 3 0
-phosphorylated end [ 60
] (
scheme 5
).
4. A SCENARIO FOR THE EMERGENCE
OF BOTH AMINOACYLATION AND ADENOSINE
TRIPHOSPHATE
An appealing scenario to explain the selection of adenylates by evolution in spite of their ineffectiveness as
Phil. Trans. R. Soc. B (2011) peptide precursors is to consider that their initial role lay in the conversion of the free energy available from amino acid chemistry as NCA into a form useful for nucleic acid chemistry. The spontaneous reactions of NCA in aqueous solution leading both to phosphate-mixed anhydride and aminoacylated esters (
scheme 5
) constitute a chemical substratum from which an early translation apparatus may have evolved through selective recognition of amino acid side-chains by RNA and ribosomal activity may have developed. Independent of this process, any activated form of nucleotides (mixed anhydrides, pyrophosphates, or triphosphates) may have been useful for
RNA replication and polymerization, providing a selective advantage to the NCA-driven pathway in an RNA world. Afterwards, the selective advantage of translated proteins may have driven evolution, and the development of protein enzymes capable of sequestrating and stabilizing adenylates would have been a key improvement as well as their interconversion into more stable phosphoanhydrides having an increased lifetime

2956 R. Pascal & L. Boiteau Review. Energy flows initial stage
RNA 2 -RNA 1
RNA 2
PP i
5 ¢ p-RNA 1 aa-p-RNA 1 ppp-RNA 1
NCA RNA-bound random peptides
RNA
HO
3
2
¢
,3
¢
RNA peptide-O
3
2 ¢ ,3 ¢
RNA world: ribozyme-mediated mixed anhydride stabilization.
Catalysis of phosphoanhydride formation.
Selectivity in peptide formation.
intermediate stage improvements in
RNAsynthesis ppp-RNA 1
PP i
5
¢ p-RNA
1
NCA
RNA
HO
3
2 ¢ ,3 ¢ aa-p-RNA ribozyme
1
RNA
Peptide-O
3
2
¢
,3
¢
RNA aa-O
3
2 ¢ ,3 ¢ primitive ribosomal activity
RNA–protein world: availability of independent pathways for ATP synthesis.
aaRS protein family replaces ribozymes and improves amino acid specificity.
late stage
ATP
PP i aa-AMP aaRS
NCA
RNA
HO
3
2 ¢ ,3 ¢ loss of the useless
NCA pathways
RNA aa-O
3
2 ¢ ,3 ¢ reversion of reactant flow due to the availability of ATP proteins
Scheme 6. A scenario for the evolution of amino acid activation pathway starting from an NCA-driven nucleotide activation towards an ATP-driven protein formation.
inducing the possibility of free energy exchanges
between different metabolic pathways ( scheme 6
).
5. CONCLUSIONS
Considering thermodynamics constraints leads to a better understanding of the driving forces by which life may have arisen or evolved during its early steps.
The most important one is the need for an energy flow continuously compensating for the loss of entropy associated with self-organization, which requires the input of low-entropy energy sources (light for instance) and high-entropy energy (heat or radiation at long wavelengths) flowing out of the system. Natural selection may arise in those systems as soon as structures that are capable of self-reproduction emerge, so that the ones that replicate faster from precursors and energy tend to predominate owing to their kinetic behaviour, and as soon as their variability is not limited.
The consequences of the need for an energy flow have been discussed with respect to carbon metabolism and translation. Both processes are consistent with a heterotrophic origin in which early life depended on the availability of energy-rich building blocks. A heterotrophic supply of formaldehyde may have been useful for conversion into carbohydrates and aliphatic chains through processes releasing energy and leading to lipidic end-products capable of constituting a boundary.
The analysis of amino acid activation shows that
Phil. Trans. R. Soc. B (2011) high-energy mixed anhydrides were involved early, which suggests that the driving force for the emergence of translation was the delivery of energy from amino acid chemistry to nucleotide chemistry. This role of amino acid derivatives cannot be deduced from present biochemical pathways but provides an explanation both of the emergence of ATP as a universal biochemical currency and of translation.
This work was supported by EPOV interdisciplinary programme of the CNRS and COST action CM0703.
NOTE
1
The occurrence of very specific enzyme catalysts capable of enhancing reaction rates with proficiencies that can exceed factors of 10
19
[
22
] by stabilizing transition states and unstable intermediates conceals the need of such amounts of free energy in contemporary biochemistry, which was not the case in prebiotic chemical systems as catalytic efficiencies were probably limited.
REFERENCES
1 Pross, A. 2004 Causation and the origin of life. Metabolism or replication first?
Orig. Life Evol. Biosph.
34 ,
307 – 321. ( doi:10.1023/B:ORIG.0000016446.51012.bc
)
2 Orgel, L. E. 2008 The implausibility of metabolic cycles on the prebiotic earth.
PLoS Biol .
6 , 5 – 13. ( doi:10.1371/ journal.pbio.0060018
)
3 Shapiro, R. 2006 Small molecule interactions were central to the origin of life.
Q. Rev. Biol.
81 , 106 – 125.

4 Fry, I. 2011 The role of natural selection in the origin of life.
Orig. Life Evol. Biosph.
41 , 3 – 16. ( doi:10.1007/ s11084-010-9214-1 )
5 Ga´nti, T. 2003 The principles of life . Oxford, UK: Oxford
University Press.
6 Wa¨chtershauser, G. 1988 Before enzymes and templates: theory of surface metabolism.
Microbiol. Rev.
52 , 452–484.
7 Martin, W. & Russell, M. J. 2003 On the origins of cells: a hypothesis for the evolutionary transitions from abiotic geochemistry to chemoautotrophic prokaryotes, and from prokaryotes to nucleated cells.
Phil. Trans. R. Soc.
Lond. B 358 , 59 – 85. ( doi:10.1098/rstb.2002.1183
)
8 Oparin, A. I. 1924 The origin of life. Proiskhodenie Zhini , pp. 199 – 234. [English transl.: Bernal, J. D. 1967 The origin of life.
The Weidenfeld and Nicolson Natural History (ed. R. Carrington). London, UK: Readers Union].
9 Haldane, J. B. S. 1929 The origin of life. The rationalist annual , pp. 242 – 249. [Reprinted in: Bernal, J. D. 1967
The origin of life.
The Weidenfeld and Nicolson Natural
History (ed. R. Carrington). London, UK: Readers
Union.].
10 Eschenmoser, A. 2007 Question 1: commentary referring to the statement ‘The Origin of Life can be
Traced Back to the Origin of Kinetic Control’.
Orig.
Life Evol. Biosph.
37 , 309 – 314. ( doi:10.1007/s11084-
007-9102-5 )
11 Boiteau, L. & Pascal, R. 2011 Energy sources, selforganization, and the origin of life.
Orig. Life Evol.
Biosph.
40 , 23 – 33. ( doi:10.1007/s11084-010-9209-y )
12 Pross, A. 2009 Seeking the chemical roots of Darwinism: bridging between chemistry and biology.
Chem. Eur. J.
15 , 8374 – 8381. ( doi:10.1002/chem.200900805
)
13 Schro What’s life.
New York, NY:
McMillan.
14 Lotka, A. J. 1922 Contribution to the energetics of evolution.
Proc. Natl Acad. Sci. USA 8 , 147 – 151. ( doi:10.
1073/pnas.8.6.147
)
15 Lotka, A. J. 1922 Natural selection as a physical principle.
Proc. Natl Acad. Sci. USA 8 , 151 – 154. ( doi:10.
1073/pnas.8.6.151
)
16 Nicolis, G. & Prigogine, I. 1977 Self-organization in nonequilibrium systems.
New York, NY: Wiley.
17 Morowitz, H. & Smith, E. 2007 Energy flow and the organization of life.
Complexity 13 , 51 – 59. ( doi:10.
1002/cplx.20191
)
18 Chyba, C. & Sagan, C. 1992 Endogenous production, exogenous delivery and impact-shock synthesis of organic molecules: an inventory for the origins of life.
Nature 355 , 125 – 132. ( doi:10.1038/355125a0 )
19 Tian, F., Toon, O. B., Pavlov, A. A. & De Sterck, H.
2005 A hydrogen-rich early earth atmosphere.
Science
308 , 1014 – 1017. ( doi:10.1126/science.1106983
)
20 Cleaves, H. J., Chalmers, J. H., Lazcano, A., Miller, S. L. &
Bada, J. L. 2008 A reassessment of prebiotic organic synthesis in neutral planetary atmospheres.
Orig. Life Evol.
Biosph.
38 , 105 –115. ( doi:10.1007/s11084-007-9120-3 )
21 Deamer, D. & Weber, A. L. 2010 Bioenergetics and life’s origins.
Cold Spring Harb. Perspect. Biol.
2 , a00492.
( doi:10.1101/cshperspect.a004929
)
22 Wolfenden, R. 2006 Degrees of difficulty of waterconsuming reactions in the absence of enzymes.
Chem.
Rev.
106 , 3379 – 3396. ( doi:10.1021/cr050311y )
23 Hoehler, T. M. 2007 An energy balance concept for habitability.
Astrobiology 7 , 824 – 838. ( doi:10.1089/ast.2006.
0095 )
24 McCollom, T. M. & Amend, J. P. 2005 A thermodynamic assessment of energy requirements for biomass synthesis by chemolithoautotrophic micro-organisms in oxic and anoxic environments.
Geobiology 3 , 135 – 144.
( doi:10.1111/j.1472-4669.2005.00045.x
)
Phil. Trans. R. Soc. B (2011)
Review. Energy flows R. Pascal & L. Boiteau 2957
25 Butlerov, A. M. 1861 Formation synthe´tique d’une substance sucre´e.
Comptes Rendus Acad. Sci.
53 , 145 – 147.
26 Miller, S. L. 1953 A production of amino acids under possible primitive earth conditions.
Science 117 , 528 –
529. ( doi:10.1126/science.117.3046.528
)
27 Weber, A. L. 2000 Sugars as the optimal biosynthetic carbon substrate of aqueous life throughout the universe.
Orig. Life Evol. Biosph.
30 , 33 – 43. ( doi:10.1023/
A:1006627406047 )
28 Weber, A. L. 2002 Chemical constraints governing the origin of metabolism: the thermodynamic landscape of carbon group transformations under mild aqueous conditions.
Orig. Life Evol. Biosph.
32 , 333 – 357. ( doi:10.
1023/A:1020588925703 )
29 Urey, H. C. 1952 On the early chemical history of the earth and the origin of life.
Proc. Natl Acad. Sci. USA
38 , 351 – 363. ( doi:10.1073/pnas.38.4.351
)
30 Deamer, D. W. 1997 The first living systems: a bioenergetic perspective.
Microbiol. Mol. Biol. Rev.
61 , 239 – 261.
31 Hanczyc, M. M. & Szostak, J. W. 2004 Replicating vesicles as models of primitive cell growth and division.
Curr. Opin. Chem. Biol.
8 , 660 – 664. ( doi:10.1016/j.
cbpa.2004.10.002
)
32 Benner, S. A., Kim, H.-J., Kim, M.-J. & Ricardo, A.
2010 Planetary organic chemistry and the origins of biomolecules.
Cold Spring Harb. Perspect. Biol.
2 , a003467.
( doi:10.1101/cshperspect.a003467
)
33 Cleaves, H. J. 2008 The prebiotic geochemistry of formaldehyde.
Precamb. Res.
164 , 111 – 118. ( doi:10.1016/ j.precamres.2008.04.002
)
34 Jencks, W. P. 1969 Catalysis in chemistry and enzymology.
New York, NY: McGraw-Hill.
35 Thaddeus, P. 2006 The prebiotic molecules observed in the interstellar gas.
Phil. Trans. R. Soc. B 361 , 1681 –
1687. ( doi:10.1098/rstb.2006.1897
)
36 Miller, S. L. & Parris, M. 1964 Synthesis of pyrophosphate under primitive Earth conditions.
Nature 204 ,
1248 – 1250. ( doi:10.1038/2041248a0 )
37 Orgel, L. E. & Lohrmann, R. 1974 Prebiotic chemistry and nucleic acid replication.
Acc. Chem. Res.
7 , 368 –
377. ( doi:10.1021/ar50083a002 )
38 Powner, M. W., Gerland, B. & Sutherland, J. D. 2009
Synthesis of activated pyrimidine ribonucleotides in prebiotically plausible conditions.
Nature 459 , 239 – 242.
( doi:10.1038/nature08013 )
39 Miller, S. L. 1998 The endogenous synthesis of organic compounds. In The molecular origins of life. Assembling pieces of the puzzle (ed. A. Brack), pp. 59 – 85. Cambridge,
UK: Cambridge University Press.
40 Kasting, J. F. & Brown, L. L. 1998 The early atmosphere as a source of biogenic compounds. In The molecular origins of life. Assembling pieces of the puzzle (ed. A.
Brack), pp. 35 – 56. Cambridge, UK: Cambridge University Press.
41 Russell, M. J., Hall, A. J. & Martin, W. 2010 Serpentinization as a source of energy at the origin of life.
Geobiology 8 , 355 – 371. ( doi:10.1111/j.1472-4669.2010.
00249.x
)
42 Lane, N., Allen, J. F. & Martin, W. 2010 How did LUCA make a living? Chemiosmosis in the origin of life.
BioEssays 32 , 271 – 280. ( doi:10.1002/bies.200900131
)
43 Mitchell, P. 1961 Coupling of phosphorylation to electron and hydrogen transfer by a chemi-osmotic type of mechanism.
Nature 191 , 144 – 148.
( doi:10.1038/
191144a0 )
44 Madigan, M. T., Martinko, J. M. & Parker, J. 2003 Brock biology of microorganisms . Upper Saddle River, NJ:
Prentice Hall.
45 Voet, D., Voet, J. G. & Pratt, C. W. 2006 Fundamentals of biochemistry , 2nd edn. New York, NY: Wiley.

2958 R. Pascal & L. Boiteau Review. Energy flows
46 Mansy, S. S. 2010 Membrane transport in primitive cells.
Cold Spring Harb. Perspect. Biol.
2 , a002188.
( doi:10.1101/cshperspect.a002188
)
47 Chen, I. A. & Szostak, J. W. 2004 Membrane growth can generate a transmembrane pH gradient in fatty acid vesicles.
Proc. Natl Acad. Sci. USA 101 , 7965 – 7970.
( doi:10.1073/pnas.0308045101
)
48 Eigen, M. 1964 Proton transfer, acid-base catalysis, and enzymatic hydrolysis.
Angew. Chem. Int. Ed. Engl.
3 , 1 – 19. ( doi:10.1002/anie.196400011
)
49 Reece, S. Y., Hodgkiss, J. M., Stubbe, J. & Nocera, D. G.
2006 Proton-coupled electron transfer: the mechanistic underpinning for radical transport and catalysis in biology.
Phil. Trans. R. Soc. B 361 , 1351 – 1364.
( doi:10.1098/rstb.2006.1874
)
50 Pascal, R., Boiteau, L. & Commeyras, A. 2005 From the prebiotic synthesis of a -amino acids towards a primitive translation apparatus for the synthesis of peptides.
Top.
Curr. Chem.
259 , 69 – 122. ( doi:10.1007/b136707 )
51 Jencks, W. P. 1976 Free energies of hydrolysis and decarboxylation. In Handbook of biochemistry and molecular biology , vol. 1 (ed. G. D. Fasman), pp. 296 – 304.
3rd edn, Cleveland, OH: CRC Press.
52 Sievers, A., Beringer, M., Rodnina, M. V. & Wolfenden,
R. 2004 The ribosome as an entropy trap.
Proc. Natl
Acad. Sci. USA 101 , 7897 – 7901. ( doi:10.1073/pnas.
0402488101 )
53 Pascal, R.
et al.
2006 Prebiotic chemistry—biochemistry—emergence of life (4.4 – 2 Ga).
Earth Moon Planets
98 , 153 – 203. ( doi:10.1007/s11038-006-9089-3 )
54 Noller, H. F. 2004 The driving force for molecular evolution of translation.
RNA 10 , 1833 – 1837. ( doi:10.1261/ rna.7142404
)
55 Wells, T. N. C., Ho, C. K. & Fersht, A. R. 1986 Free energy of hydrolysis of tyrosyl adenylate and its binding to wild-type and engineered mutant tyrosyl-tRNA synthetases.
Biochemistry 25 , 6603 – 6608. ( doi:10.1021/ bi00369a040 )
56 Schimmel, P. & Henderson, B. 1994 Possible role of aminoacyl-RNA complexes in noncoded peptide synthesis and origin of coded synthesis.
Proc. Natl Acad. Sci. USA
91 , 11 283 – 11 286. ( doi:10.1073/pnas.91.24.11283
)
57 Pascal, R. & Boiteau, L. 2010 Energetic constraints on prebiotic pathways: application to the emergence of translation. In Origins and evolution of life: an astrobiological perspective (eds M. Gargaud, P. Lo´pez-Garcı`a &
H. Martin), pp. 247 – 258. Cambridge, UK: Cambridge
University Press.
58 Leman, L., Orgel, L. & Ghadiri, M. R. 2004 Carbonyl sulfide-mediated prebiotic formation of peptides.
Science
306 , 283 – 286. ( doi:10.1126/science.1102722
)
59 Danger, G., Boiteau, L., Cottet, H. & Pascal, R. 2006
The peptide formation mediated by cyanate revisited.
N -carboxyanhydrides as accessible intermediates in the decomposition of N -carbamoylaminoacids.
J.
Am.
Chem. Soc.
128 , 7412 – 7413. ( doi:10.1021/ja061339+ )
60 Biron, J.-P., Parkes, A. L., Pascal, R. & Sutherland, J. D.
2005 Expeditious, potentially primordial, aminoacylation of nucleotides.
Angew. Chem. Int. Ed.
44 , 6731 –
6734. ( doi:10.1002/anie.200501591
)
61 Leman, L. J., Orgel, L. E. & Ghadiri, M. R. 2006 Amino acid dependent formation of phosphate anhydrides in water mediated by carbonyl sulfide.
J. Am. Chem. Soc.
128 , 20 – 21. ( doi:10.1021/ja056036e )
62 Brack, A. 1987 Selective emergence and survival of early polypeptides in water.
Origins Life 17 , 367 – 379. ( doi:10.
1007/BF02386475 )
63 Zahnle, K., Schaefer, L. & Fegley, B. 2010 Earth’s earliest atmospheres.
Cold Spring Harb. Perspect. Biol.
2 , a004895.
( doi:10.1101/cshperspect.a004895
)
Phil. Trans. R. Soc. B (2011)