Friday, May 15, 2009

Metabolism First and the Origin of Life

There are several competing hypotheses about the origin of life. Most people know about the Primordial Soup scenario; that's the one where complex organic molecules are created by spontaneous chemical reactions. Over time these complex molecules, such as amino acids and nucleotides, accumulate in a warm little pond and eventually they come together to form proteins and nucleic acids.

The RNA World scenario is similar except that nucleic acids (RNA) are thought to form before proteins. For a while, RNA molecules are the main catalysts in the primordial soup. Later on, proteins take over some of the catalytic roles. One of the problems with the RNA world hypothesis is that you have to have a reasonable concentration of nucleotides before the process can begin.

The third hypothesis is called Metabolism First. In this scheme, the first reactions involve spontaneous formation of simple molecules such as acetate, a two-carbon compound formed from carbon dioxide and water. Pathways leading to the synthesis of simple organic molecules might be promoted by natural catalysts such as minerals and porous surfaces in rocks. The point is that the origin of life is triggered by the accumulation of very simple organic molecules in thermodynamically favorable circumstances.

Simple organic molecules can then be combined in various ways that result in simple amino acids, lipids, etc. These, in turn, could act as catalysts for the formation of more organic molecules. This is the beginning of metabolism.

Eventually simple peptides will be formed and this could lead to better catalysts. Nucleic acids and complex amino acids will only form near the end of this process.

One of the advantages of the metabolism first scenario is that it offers a simple "solution" to the chirality/racemization problem by explaining why all naturally occurring amino acids are left-handed [see Can watery asteroids explain why life is 'left-handed'?]. Another advantage is that it doesn't require spontaneous formation of nucleotides—a major limitation of the RNA world scenario since spontaneous formation of such molecules is very improbable.1

James Trefil, Harold Morowitz, and Eric Smith have written up a very nice summary of the Metabolism First hypothesis for American Scientist: The Origin of Life. The subtitle, "A case is made for the descent of electrons," is a clever play on words. It illustrates the point that synthesis of simple organic molecules such as acetate are thermodynamically favorable. This is science writing at its best.2

The authors have reconstructed the simplest, most fundamental, biochemical pathways concluding that a reductive citric acid cycle is probably the best example of the first metabolic pathway. In this pathway, the two-carbon acetate molecule is made from carbon dioxide and water in the reverse of the common citric acid pathway found in eukaryotes.

In fact, the reductive pathway occurs in many bacteria. They can still use it to fix carbon. The authors use the figure on the left to illustrate the basic pathway.

Almost all of the common molecules of life are synthesized from acetate or the molecules of the citric acid cycle. The simple amino acids, for example, are formed in one step. More complex amino acids are derived from the simple amino acids, etc. Similarly, simple fatty acids can be formed from acetate and more complex ones come later; once the simple ones accumulate.

The central role of citric acid cycle metabolism in biochemistry has been known for decades. It's involvement in biosynthesis pathways is often ignored in introductory biochemistry courses because they are heavily focused on fuel metabolism in mammals and biosynthetic pathways get short shrift in such courses.

The essence of Metabolism First is that the various complex molecules of life came after the spontaneous formation of very simple molecules. Pathways leading to the complex molecules evolved and their evolution was assisted by the evolution of various catalysts, some of which were biological in nature.

1. In spite of the claims surrounding a recent paper in Nature: RNA world easier to make.

2. Probably good science editing as well. My friend Morgan Ryan is managing editor and he is very good.

[Photo Credit: American Scientist, courtesy of Scripps Institution of Oceanography, University of California, San Diego.]


  1. Bayesian Bouffant, FCDFriday, May 15, 2009 1:03:00 PM

    I don't see a conflict between RNA World and Metabolism First. To me, RNA World says that RNA came before DNA, specifically sequenced proteins and any other complex biopolymers.

    But of course those polymers developed in a suitable environment of precursors and favorable energetics. But I don't think of those as life, and most certainly not life as we know it. That would be like calling the phosphorus and nitrogen fertilizer I spread on my lawn life. They are the background, the environment.

  2. The NYTimes article on this:
    Chemist Shows How RNA Can Be the Starting Point for Life
    Dr. Sutherland’s proposal has not convinced everyone. Dr. Robert Shapiro, a chemist at New York University, said the recipe “definitely does not meet my criteria for a plausible pathway to the RNA world.” He said that cyano-acetylene, one of Dr. Sutherland’s assumed starting materials, is quickly destroyed by other chemicals and its appearance in pure form on the early earth “could be considered a fantasy.”
    Dr. Sutherland replied that the chemical is consumed fastest in the reaction he proposes, and that since it has been detected on Titan there is no reason it should not have been present on the early earth.

    I think Shapiro has simply become too accustomed to saying no.

    1. Shapiro has a nickname Dr No he is very proud of

  3. I don't see a conflict between RNA World and Metabolism First.

    Me neither. I thought the recent paper in Nature on how nucleotides may form spontaneously is great and, in essence, illustrates a possible pathway of transition from one to another.

  4. V. interesting; This supplements what I have already heard about "Metabolism first".

    Tell me Larry; when you delve into these rather unfossilferous speculative areas on the origin of life, do you ever get doubts? No, of course not: I get doubts, but you just react with a "bah humbug!"

  5. Timothy V Reeves writes: "...unfossiliferous speculative areas on the origin of life...."

    As I read it, the post concerns simple(r) chemistry being a more probable starting point for life than more complex chemistry. Speculative in the sense that we don't have a record of events as they unfolded, yes; but not so in the sense that the chemistry involved is familiar and well-understood.

  6. Late to this, but love the subject and wonder if you are familiar with the Morris/Russell treatment of the subject? here and here? I loved the American Science article focussing on the movement of electrons through the metabolic cycles, while Martin and Russell focus on the pumping of protons through the same cycles. An interesting suggestion is that since archaebacteria and eubacteria share DNA/RNA and various metabolic pathways, but have utterly different ways to form cell membranes, their shared history includes a period of development within naturally occurring "cell-type" spaces, but diverges prior to the independent membrane-bounded emergence of each to the external environment.

  7. There is a third possibility, which I am afraid, you have overlooked. Life has three fundamental characteristics. 1. replication, 2. metabolism and 3. consciousness (or irritability). I think the last one is the most essential feature for life to form. So I propose "Consciousness First Theory". Consciousness has originated from primordial cell membrane in the form of membrane potentials. I have proposed that primordial membranes have formed from "Hydrocarbon Mass" in the Earth's crust 3.7 billion years ago. Organic molecules have formed due to erosion of hydrocarbon chains by the electrical (membrane) potentials.

    I have written a book on this subject "The Role of Cell Membrane in the Origin of Life and in Cell Biology", which explains this theory in detail.

    Any takers of this theory?


  9. There is only 2 articles on the origins of life? I wonder why?

  10. In this question I don´t think you have got the right answer. I see no possibility of life without replication, and the most primitive replicator we know of is RNA. As RNA can also catalyze reactions, I see it as evident that there was an RNA world. There is still a problem to be solved: how the first RNA molecules were created. But I am sure this question will be solved.

    1. "But I am sure this question will be solved."

      It is solved by postulating that basic metabolism arose before there was RNA. The beginnings of life required ways to make sugars and nucleotides and confine them to small volumes. The catalysts for those reactions were inorganic molecules (e.g. metals) and small peptides.

      Of course you need replicators and of course RNA is the most logical choice. But you just can't get to RNA from carbon dioxide without metabolism first.

    2. Larry,

      While metabolism first is an idea, there had to be more to it than that I think. There had to be some kind of a shield from harsh environment for life to continue. A protective membrane of some kind before metabolism.

      I'm also puzzled as to how even the simplest lifeforms developed without information for their development. They need it now how come they didn't need it early on?

    3. "But you just can't get to RNA from carbon dioxide without metabolism first."


      You are right that synthesis of nucleotides is not a simple process. And there were other sources than carbon dioxide, which probably was the main component of the atmosphere. Atmosphere may at an early stage also have contained ammoniac, methane and even some free hydrogen. There may also have been some free chlorine, as there may not have been enough sodium to bind all of it. This chlorine may have been crucial for many reactions by activating molecules.

      If the only possibility was creation of life in the oceans, then I would choose an active site like a hydrothermal vents. But I do not think there is a choice only between "soupers" and "smokers", even if these two theories are the most popular today. I have seen no proof that the early Earth was totally covered with water. On the contrary, the processes that created life must have started once there was liquid water, and in the beginning most of the surface would have been land. Although there would have been dramatic greenhouse effect, there would have been a clear difference between daytime and nighttime temperatures. And I am not convinced that the lack of an ozone layer would have let through so much UV light that primitive life was impossible, at least not in areas shadowed by local stones or by clouds.

      I don’t think oceans ever have been "sweet". There would simply not have been enough carbon for that. But there could locally have been sweet spots. Areas around hydrothermal vents would have had high concentrations of minerals, but not organic molecules, and probably not either much phosphate.

      If we look at the basics of life, nucleotides, proteins and fatty acids, then we see that all the elements that they consist of are gaseous, except phosphor and sulphur. We can easily imagine that the first proteins did not contain methionine or cysteine, but nucleotides without phosphate is not possible unless there was initially another backbone. Leslie Orgel was convinced that there was a change of backbone, but I find such a change difficult. I have instead speculated how phosphate could have been incorporated directly.

      As Miller-Urey showed with their famous experiment, molecules like amino acids are synthesized from gaseous molecules if there is something triggering reactions and there is good contact between an aqueous phase and the gaseous phase. I imagine that good contact will be present on a changing surface like the one that is formed when dew drops form on a solid surface, i.e. on a mineral. An extra effect of the mineral could be to catalyze reactions, and substances in the surface could even take place in the reactions.

      If the surface was apatite, then it could work as a source of phosphate. The ribose may have been synthesized on another mineral. Different surfaces could be connected e.g. through rain drops, and various substances could be purified during the day, when heating dried up the dew.

      Slower drying processes combined with drops e.g. in caves could purify sugars and other easily crystallizable substances. Different minerals would sort out specific molecules due to their catalytic effect.

      With the right positioning of different mineral surfaces and maybe also surfaces that created shadow at the right periods during the day, there could be factories suited especially for connecting ribose or other sugars to phosphate. The source of sugars could be a cave that e.g. during the nights had extra supply of water and therefore washed out some crystalized sugar. Further there could also have been purine/pyrimidine factories that could also supply the RNA bases. In apatite, the phosphate groups are perfectly aligned in parallel lines with a distance that gives the required space for sugar and bases.

    4. Ad RNA world starting on mineral surfaces: Dew drops may have been the first encapsulations. Large fatty molecules that did not dissolve in water would have migrated to the drop boundary, and eventually, a beehive structure of a tar-like substance would be built between the drops. Thereby drops would form at the same places every day. With higher humidity the drops would have fused across this boundary, and larger drops would have formed. It is therefore likely that beehive structures would have formed in a hierarchical system. This would have given room for fusion during moisture increase and fission during drying up. This could have been the first reproduction.

      Fatty acids would migrate to the surface of the drops, and before they dried up completely, they could sometimes be completely covered by a single layer membrane. We can also imagine that an extra layer formed if these drops crossed a fluid surface that was covered by a fatty acid surface. Fusion and fission processes between these cells could have been the first cellular reproduction.

      I am not saying that there were no proteins in the RNA world, but they had no simple coding, and therefore they had to be built in quite complex ways. I am not either saying that there were no DNA in the RNA world. DNA was probably soon invented as a substitute for RNA for long term storage of information.

      I have not performed any experiment to support this theory, and I do not know of any others that have similar theories. I would however be very pleased if somebody knows about any similar. I would especially appreciate if somebody could design and maybe also perform experiments that could support origin of life on mineral surfaces.

    5. Phosphate has been crucial not only for replication, i.e. in the synthesis of RNA, but also for metabolism. The general energy carrier, ATP, is just a nucleotide that has been extra phosphorylated two times. Another variant, based on the G base instead of A is also in use. The energy lies in the phosphate group. But also other molecules take up one or two phosphate groups, probably from a mineral source. These phosphate groups can be transferred to ADP, thereby producing ATP. Two of the phosphorylated molecules are 1,3-bisphosphoglycerate and phosphoenolpyruvate. They are part of the most important reaction chains of life, glycolysis and gluconeogenesis. These reaction chains involve electron transport, which is also performed by an RNA-like molecule: NAD+/NADH.

      Different reaction chains established to take care of the released electron pair. One of them produced succinate as end product via fumarate. But also another reaction chain with succinate as end product established, this one via citrate. This chain was used more for anabolic purposes, but it could only be used in limited amount because it produced free electrons. A bridge established between the start point of these two chains, and the anabolic chain could run the reaction in the opposite direction. Thereby a cycle was established, which we can call the reverse Krebs cycle. It could effectively consume free electrons when there was enough carbon dioxide available. It was binding carbon dioxide, and it could serve the anabolic purpose when there were enough free electrons.

      At a later stage, membranous release of electrons also established, first by something quite similar to hydrogenosomes. As earth was oxidized, more and more efficient electron acceptors became available: sulfur, iron ions, and eventually oxygen. An electron transport chain established in the membrane of organelles, that eventually resulted in the mitochondrion. Membrane processes could help the Krebs cycle run in the succinate - fumarate direction, and thereby this cycle could be used as a supply of electrons instead of consumption. Instead of using carbon dioxide as an energy source (by reacting with hydrogen, producing methane), it was now an end product, and Krebs cycle was now mainly used for breaking down the end product of glycolysis. Thereby it was producing electrons to drive the electron transport chain in the membrane.

    6. James Trefil, Harold Morowitz, and Eric Smith came to the conclusion that the reverse Krebs cycle established before the version that we see today. I have come to the same conclusion, but I have seen it as an extension of the glycolysis pathway. Both these authors and later
      Bill Martin and Nick Lane saw the consumption of electrons in this cycle as primary to life. But the source of electrons was not the glycolysis pathway, as in my scenario, but membrane processes. My theory is based on RNA first, and I see that RNA components are used in electron carriers and in energy carriers like NAD+/NADH. They assume that some metabolism established before RNA was in place, and the problem that this should solve in this way is the creation of nucleotides. But can anybody tell me how they explain the existence of nucleotides in the electron and energy carriers?

      Or do they see it as Bayesian Bouffant does: "I don't see a conflict between RNA World and Metabolism First. To me, RNA World says that RNA came before DNA, specifically sequenced proteins and any other complex biopolymers." He then argues that everything based on RNA, without reproductive proteins, were not life, because there was no metabolism. Thereby he can combine the two theories. Even though metabolism was the last component to establish, according to common definitions of life the occurrence of metabolism was tantamount to occurrence of life. But I am sure there was metabolism also in the RNA world, even though none of the ribozymes we have today are engaged in metabolism.

      There are different definitions of "RNA world". Some see it as the world that existed before DNA. But I see no need for a separate "DNA world". The transition from RNA to DNA is quite simple, and I assume DNA was used quite early as a more reliable long term storage of genetic information. But RNA was still the building block of catalysts. Translation is a complex process, and it was probably not invented before a lot of metabolic pathways built on ribozymes were in place. That does not mean proteins were not in use. They were probably parts of many ribozymes. We see that even today in some of the remaining ribozymes, e.g. the ribosomes. They may even have played more active roles than what we see in the ribosomes. But as there was no translation, the synthesis of proteins must have been based on a lot of special mechanisms, in much the same way as sugar chains are produced even today.

    7. Velhovsky writes: "There had to be some kind of a shield from harsh environment for life to continue. A protective membrane of some kind before metabolism."

      A membrane, yes, but not to protect the organism from the environment. The membrane *is* the organism, and the role of metabolism is to make the membrane grow, and ultimately reproduce by fission. Only later in life's history did the membrane assume the role of a fence, separating the soluble molecules 'owned' by the organism from those belonging to the environment, and preventing the former from diffusing away.

      Velhovsky: "I'm also puzzled as to how even the simplest lifeforms developed without information for their development. They need it now how come they didn't need it early on?"

      Oh, they needed information. The information just took a different physical form. For example, one gene, or bit of information, might have been "This organism contains much more D-Glyceraldehyde than L-Glyceraldehyde." Another might be "This organism contains acetyl CoA". A third: "This organism does not contain pyruval CoA".

      It should be obvious that this kind of 'gene' is automatically inherited when the organism reproduces by fission. The gene is passed from parent to child. The tricky thing is to establish that it is preserved by growth - the gene remains unchanged as the child becomes the parent of the next generation.

  11. The conceptual gulf that separates the 'metabolism first' and 'replication first' mechanisms for the emergence of life continues to cloud the origin of life debate. In the present paper we analyze this aspect of the origin of life problem and offer arguments in favor of the 'replication first' school. Utilizing Wicken's two-tier approach to causation we argue that a causal connection between replication and metabolism can only be demonstrated if replication would have preceded metabolism. Source 1 There are also several metabolism foods which helps to burn fat.

  12. Jim Menegay writes: "The membrane *is* the organism, and the role of metabolism is to make the membrane grow, and ultimately reproduce by fission."

    Membrane metabolism is among the most complex of life. The heart of it is membraneous ATPase. Do you anticipate that such systems have occurred from scratch? And if so, how could they have been produced repeatably? I suppose you know about Eigen’s theories about self-regulatory systems consisting of a few molecule types. We could compare such systems with "non-intelligent" electronic regulatory systems. It is obvious that such systems are much more limited than computer controlled systems. It is quite evident that the intelligent electronic systems do not have the same limitations. And with nature’s systems we are talking about systems that are immensely more complex than the most complex systems that humans have built.

    1. The original membrane anabolism could have been quite simple, consuming CO or HCN and yielding (disproportionating to) CO2 and straight chain hydrocarbons with a hydrophilic end group. Think Fischer-Tropsch. No additional energy source required.

      Later, a switch could be made to a pathway more like modern biological ones, like reductive TCA or the Wood-Ljungdahl.

      As for the ATPase, that arose much later in evolution. However, if you look at the mechanism of, say, joining two phosphorylated lipid heads into a high-energy anhydride, you will see that bringing in a proton or two from across the membrane can make it easy to extract a hydroxyl from one phosphate and neutralize the negative charge on the other.

      Modern metabolism is complicated. Primitive metabolism need not have been. In fact, the metabolism-first idea insists that almost all biological complexity arose gradually over time under the direction of natural selection.

    2. I wrote: "... almost all biological complexity arose gradually over time under the direction of natural selection." Hmmm. Since Larry has recently been promoting the work of Michael Lynch on the origin of genome complexity, I suppose I should back away from this a little.

      As Lynch points out, complexity in the genome often arises from an accumulation of almost-neutral changes. Natural selection's role in this is nothing like that of the director of a stage-play. Instead it is the looser kind of guidance provided by a referee in a sporting match.

      As is the case with the complexity of the modern genome, many aspects of the complexity of modern metabolism are best understood as the result of a co-evolutionary modus vivendi, rather than as a Panglossian optimization of a unified objective function.

  13. This paper from my lab entitled "Metals Promote Sequences of the Reverse Krebs Cycle" might add to the discussion.


    Also this blog entry:

  14. Thanks a lot for your posting!
    I just wonder why nothing of the pioneering work from Marcello Guzman (or Scott Martin's group before) has been mentioned in your remarkable new paper in this regard.

    1. Thanks! We did in fact cite the pioneering work of Scot Martin in ref. 32 of our paper. I also cited Guzman's independent work (he was a postdoc in Martin's group), which is essentially studying the same reactions as Martin in further detail, in earlier drafts of the manuscript but it had to be cut due to a 50 reference limit for that journal. Glad you enjoyed the paper!

  15. To those of you who still are concerned about any possibilities for precursory forms of proto-life without replication, here is some pertinent input from the Loren Williams group in pointing out the potentially deceptive idolizing of tentative “privileged functions” in various conventional models in Origins of Life (OoL) research “The Origin of Life: Models and Data”:
    “A privileged function is an extant biological function that is excised from its biological context, elevated in importance over other functions, and transported back in time to a primitive chemical or geological environment [… but] the simplicity of these models is seen to be an illusion on the realization that the models require fluidity in principles of evolution”.
    Most notably, ‘replication’ is one of these illusively simple ‘privileged functions’.

    Contrary to the opinion expressed in that paper, however, I do not consider ‘metabolism’ to rank at the very same level in this regard. Metabolism is the “sum total of the chemical processes that occur in living organisms”. By this token, metabolism is more inclusive than the faithful replication of certain molecules, such as RNA or DNA. As far as science can tell and evaluate, life (the living state) is strictly coupled to the physical coherence, chemical functionality and unbroken historical continuity of certain material entities. Strictly following the arrow of time, it is the sum total of chemical reactions that is evolving together with the physical cohesion as material entities, which thus qualify as coherent and historically connected “systems”. To my views (as well as of others), therefore, molecular replication is a subsystem of metabolism but not the other way round.

    Shortly after I first read the Lanier & Williams article, I also hit upon another one from the same group, “Frozen in Time: The History of Proteins”:
    Yet, the following passage in the abstract appeared strangely provocative to me:
    “Coded proteins originated as oligomers and polymers created by the ribosome, on the ribosome and for the ribosome. […] Protein catalysis appears to be a late byproduct of selection for sophisticated and finely controlled assembly”. These statements, in fact, happen to isolate ‘ribosomal protein synthesis’ as yet another ‘privileged function’, deceptively “transported back in time” before the posited later beginnings of metabolism as a protein-catalyzed phenomenon.

    To my views, these statements postulate a partisan selfishness for emerging ribosomes that seems strangely detached from their pivotal system-supporting role at later stages, when the vast majority of proteins made by ribosomes are NOT incorporated in the ribosomal particles themselves. This caveat, in fact, is calling for alternative interpretations about the transition from uncoded to coded protein synthesis when the stochastic variation of peptide chain elongation was strongly reduced. First of all, ribosomes are not acting alone in making proteins; they heavily depend on the presence of aminoacylated tRNAs. Arguably, tRNAs are substantially older than ribosomes and may originally have functioned as genuine peptidyl transferases for uncoded peptide/protein synthesis. See:
    From a general systems continuity perspective, therefore, a significant share of early uncoded peptides supposedly began to participate in similar functions to those of coded proteins later on, and main progress resulting from the ribosomal assistance of tRNAs was a matter of vastly increased overall efficiency rather than a qualitative change in the most basic systems properties.

    1. Richard, I can't help but ask what you (or anyone else) would add to the water and if necessary gasses to form RNA from common household items, in this home/classroom experiment of mine:

    2. Gary, first of all I'm a biologist and not a chemist, so I shouln't dare making practical suggestions for your kitchen-type experiments. This said, however, it is (almost) basic knowledge that (terrestrial) geo/hydrothermic vents or field exhale a lot of noxious (chemically active) stuff, which under unoxic conditions triggers all kinds of fancy reactions.
      Perhaps you might want to read what Matthew W. Powner(et al.) has accomplished in somewhat 'dirty' "One-Pot" concoctions.
      Powner, M. W., Sutherland, J. D., & Szostak, J. W. (2010). Chemoselective multicomponent one-pot assembly of purine precursors in water. Journal of the American Chemical Society, 132(46), 16677-16688.
      Powner, M. W., Gerland, B., & Sutherland, J. D. (2009). Synthesis of activated pyrimidine ribonucleotides in prebiotically plausible conditions. Nature, 459(7244), 239-242.
      Powner, M. W., & Sutherland, J. D. (2011). Prebiotic chemistry: a new modus operandi. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 366(1580), 2870-2877.