More Recent Comments

Thursday, January 27, 2011

More Prebiotic Soup Nonsense

There's another round of nonsense under way, fueled by the discovery that chemical trickery can lead to a slight excess of L-amino acids over D-amino acids. There's a BBC News story about it at: 'Life chemicals' may have formed around far-flung star. The story is reproduced without comment on RichardDawkins.net.

The press reports refer to an article published in The Astrophysical Journal Letters [NON-RACEMIC AMINO ACID PRODUCTION BY ULTRAVIOLET IRRADIATION OF ACHIRAL INTERSTELLAR ICE ANALOGS WITH CIRCULARLY POLARIZED LIGHT]. Here's the abstract.
The delivery of organic matter to the primitive Earth via comets and meteorites has long been hypothesized to be an important source for prebiotic compounds such as amino acids or their chemical precursors that contributed to the development of prebiotic chemistry leading, on Earth, to the emergence of life. Photochemistry of inter/circumstellar ices around protostellar objects is a potential process leading to complex organic species, although difficult to establish from limited infrared observations only. Here we report the first abiotic cosmic ice simulation experiments that produce species with enantiomeric excesses (e.e.'s). Circularly polarized ultraviolet light (UV-CPL) from a synchrotron source induces asymmetric photochemistry on initially achiral inter/circumstellar ice analogs. Enantioselective multidimensional gas chromatography measurements show significant e.e.'s of up to 1.34% for (13C)-alanine, for which the signs and absolute values are related to the helicity and number of CPL photons per deposited molecule. This result, directly comparable with some L excesses measured in meteorites, supports a scenario in which exogenous delivery of organics displaying a slight L excess, produced in an extraterrestrial environment by an asymmetric astrophysical process, is at the origin of biomolecular asymmetry on Earth. As a consequence, a fraction of the meteoritic organic material consisting of non-racemic compounds may well have been formed outside the solar system. Finally, following this hypothesis, we support the idea that the protosolar nebula has indeed been formed in a region of massive star formation, regions where UV-CPL of the same helicity is actually observed over large spatial areas.
The authors assume that the primodial soup speculation about the origin of life is the most reasonable explanation. According to this widely believed scenario, life originated in a soup of organic molecules that supplied most of the molecules of metabolism such as glucose and amino acids (and nucleotides?). Presumably once life got underway these molecules were used up and only then did metabolic pathways evolve to synthesize these molecules.

The competing hypothesis is Metabolism First [Metabolism First and the Origin of Life]. In this scenario, the first steps involved the establishment of simple oxidation-reduction reactions across a "membrane" using inorganic molecules. Once this supply of energy was in place the first pathways led to synthesis of simple organic molecules like acetate and glycine.

What's wrong with the Primordial Soup model? Well, for one thing, it's awfully hard to imagine how incoming asteroids could supply enough material to make a difference. The maximum concentration of all amino acids in the ocean, for example, could never have been more than 10-100 pM and that's optimistic [Can watery asteroids explain why life is 'left-handed'?].1

Instead of trying to prove that asteroids could carry a slight excess of L-amino acids, I wish these workers would apply a bit of healthy skepticism to the subsequent steps of the scenario. It's not reasonable to assume that minute quantities of amino acids could ever fuel the origin of life. Incidentally, the Primordial Soup Hypothesis also imagines that early cells used exogenous glucose as a fuel. This implies that the glycolysis pathway is more primitive that the gluconeogenesis pathway for synthesis of glucose. Unfortunately the data disproves this prediction. Gluconeogenesis is more ancient and glycolysis evolved later [Aldolase in Gluconeogenesis & Glycolysis]. A nasty little fact.

The real problem is not that metabolism firstists such as Bill Martin are right and soupists are wrong—although that's a very real possibility. The problem is that most scientists are not thinking critically about the origin of life. There are several possibilities and none of them are particularly convincing. However, the Primordial Soup Hypothesis has a number of glaring weaknesses that need to be addressed honestly and it doesn't do anyone any good if scientists sweep these weaknesses under the rug.


1. We're talking about a primordial soup where the concentration of L-alanine might be 0.50 pM and the concentration of D-alanine might be 0.49 pM. That's supposed to be enough for life based on amino acids to evolve and to lead to the subsequent preference for synthesizing exclusively L-amino acids. How, exactly, does that work?

21 comments :

Anonymous said...

Small enantiomeric excesses can be amplified by asymmetric autocatalytic reactions like the Soai reaction. Note: I'm not saying the Soai reaction was involved in the origins of life, just that processes like the Soai reaction can significantly shift the enatiomeric ratio of scalemic mixtures.

RaulFelix said...

Hi, Larry,

Can't you have a look at my hypothesis, at
http://www.ecopoese.bio.br/THE_ECOPOESIS_MODEL.pdf ?

Thanks,

Raul A. Félix de Sousa

Anonymous said...

Surely the fixation by life on L acids (and D sugars) is most likely simply a 'frozen accident'? If you have a set of enzymes (or ribozymes) that work with non-isomeric glycine, then we may develop one whose 'pocket' is reshaped to allow replacement of one or the other hydrogen by a small side chain. Far simpler to use this asymmetric enzyme as the basis for further retooling to deal with other side groups than to attempt to evolve a mirror version, whose products would kink proteins, even if an AA-tRNA synthetase could handle them.

Unknown said...

So based on the frozen accident hypothesis, it is slightly more likely that this would happen if there was 51 molecules of L and 49 of D, but only very slightly. And if the frozen accident were true, then why would that be true for all amino acids. One would think that all frozen accidents might not be dependent on the previous frozen accidents

Larry Moran said...

I don't think you understand the concept of frozen accident. It's an "accident" because the D-isomer could just as easily been the "winner" in the lottery.

It's "frozen" because once you start selecting for one or other of the isomers you can't go back.

In the Metabolism First scenario the first chiral amino acid might have been L-alanine (by accident) but then it would have been advantageous to evolve a pathway for synthesis of L-serine instead of D-serine. Eventually you bootstrap your way to 20 common amino acids that are all L isomers.

Anonymous said...

@Dave Bridges So based on the frozen accident hypothesis, it is slightly more likely that this would happen if there was 51 molecules of L and 49 of D, but only very slightly. And if the frozen accident were true, then why would that be true for all amino acids. One would think that all frozen accidents might not be dependent on the previous frozen accidents

The ratio of L and D acids in the soup is irrelevant to the argument. I am suggesting that the nature of biological catalysis and catalyst evolution are the constraints that cause settlement on one form or the other; I think the chances are pretty much 50/50. The scenario I painted derives all subsequent orientation from that of the first biologically useful chiral form. Having a ribozyme that works with one chiral form gives a template for gene duplication and modification to work with variants on that theme. Whichever side is 'chosen', it is more likely that further side-chains are catered for by modification of this expandable side-chain pocket than that a mixture of L-specific and D-specific enzymes would evolve from independent sources, with L somehow winning out because of a fractional superiority of numbers

So (on the hypothesis) it is true for all amino acids because the enzymes that deal with the many are descendants of those that dealt with the few. It only needs to be frozen once - like driving on the left (or right).

There is good reason to think that the current 20 is an expanded set derived from a more ancient 10 in a code with only doublet specificity (in a triplet frame). So at least 10 of the set are constrained. I'm suggesting that they all are, except for the one from which the rest derive.

Another factor is the architecture of the ribosome, once you start to use acids for polypeptide formation - wherever they come from. A 'mirror' ribosome could condense D-acids instead, but you would be unlikely to get one that could handle both - if it could, it would be somewhat susceptible to hugely disruptive changes of backbone direction by mutation.

Similar constraints could drive the evolution of a family of sugar-specific enzymes, as life takes on the task of biosynthesis as well as metabolism. The structural centrality of ribose in the molecules RNA, DNA, ATP and NADH may be another 'freezing point'.

Once an orientation is favoured, which can be entirely at random, it becomes the stock template.

Anonymous said...

In the Metabolism First scenario the first chiral amino acid might have been L-alanine (by accident) but then it would have been advantageous to evolve a pathway for synthesis of L-serine instead of D-serine. Eventually you bootstrap your way to 20 common amino acids that are all L isomers.

This is where I may part company with you. I'm not a fully convinced 'souper', but I don't see how a non-replicating system can evolve, in the biological sense. Still, as far as amino acids are concerned, we can actually sweep 'em under the carpet until we do have replicators, with the ability to produce template-based polypeptides from them. Till then, it seems that amino acids aren't of much use to a 'raw' metabolism?

RaulFelix said...

The "frozen accident" hypothesis is a very compelling possibility to explain the fixation of D-sugars since once the first enzymatic process from achiral phosphoenolpyruvate to chiral 2-phosphoglycerate evolved, all the subsequent steps in the glycosynthetic pathways would inherit that stereochemical conformation. For amino acids, however, things are not that simple, because there is not one single step for the generation of chirality. The chirogenic step, which in general the reductive amination of an alpha-keto acid, has to be performed every time each chiral amino acid is generated. Furthermore, much of the initial metabolism, including the reductive tricarboxilic acid cycle and the amino acids immediately related to it, would have to exist before there were any chiral catalysts around.

Anonymous said...

@RaulFelix As mentioned in another post, polypeptide condensation in the ribosome could be the 'filter' that forces a specific isomeric orientation upon amino acid pathways feeding into it.

The chirogenic step, which in general the reductive amination of an alpha-keto acid, has to be performed every time each chiral amino acid is generated.

If, however, the (en/ribo)zymes dealing with various side-chains are paralogous, chirality may be inherited as a 'family' trait, by modifications of the side-chain binding site.

Furthermore, much of the initial metabolism, including the reductive tricarboxilic acid cycle and the amino acids immediately related to it, would have to exist before there were any chiral catalysts around.

True. I think a distinction could be drawn between amino acids as a source of 2-carbon units or nitrogen (which can be blind to side-chain nature and site of attachment, particularly if catalysis is 'non-bio') and biosynthesised amino acids. In the primitive metabolism, chirality need not be an issue - if you get your building blocks from the outside world, and that supplies a racemic mixture, then there is advantage in being blind to handedness. But once you start to biosynthesise, using these new-fangled biological catalysts, you have to pick an orientation. If you are a consumer, you find that the 'outside world' has started supplying an excess of L.

Larry Moran said...

RaulFelix says,

The chirogenic step, which in general the reductive amination of an alpha-keto acid, has to be performed every time each chiral amino acid is generated.

This is clearly not correct since many amino acids are derived from other amino acids. In addition I suspect that many of the original reactions were performed by the same enzymatic activity.

For example, the same primitive stereospecific animotransferse might have recognized pyruvate, alpha-ketobutyrate, and oxaloacetate.

Larry Moran said...

Alan Miller says,

This is where I may part company with you. I'm not a fully convinced 'souper', but I don't see how a non-replicating system can evolve, in the biological sense.

I shouldn't have used the word "evolve." It would be better to think of early chemical mixtures as just changing over time.

I imagine that these primitive chemical mixtures were around for thousands of years before there was any hint of a self-replicating system.

Still, as far as amino acids are concerned, we can actually sweep 'em under the carpet until we do have replicators, with the ability to produce template-based polypeptides from them. Till then, it seems that amino acids aren't of much use to a 'raw' metabolism?

Mixtures of amino acids can spontaneously form peptides and all kinds of catalytic activities have been found in these mixtures. No matter how you imagine the origin of life there has to be a long period of chemical activity before you ever get nucleic acid replicators and the genetic code.

All that comes much, much, later.

Anonymous said...

I imagine that these primitive chemical mixtures were around for thousands of years before there was any hint of a self-replicating system.

The other missing ingredient is energy. Replicator and constructive metabolism both require some means of pushing reactions against the thermodynamic gradient, and for similar reasons - replication is simply a particular kind of constructive metabolism. The minimum requirement for a nucleic acid replicator is a means to create its subunits: nucleotide phosphates. But first you have to get their subunits - - a pentose sugar, a nucleotide and 1-3 phosphate groups.

So to that extent, yes: metabolism first (including harnessing energy in phosphate bonds). Something I find remarkable is the way the phosphate-pentose-nucleotide group is ingrained so deeply within the fabric of life. ATP and GTP, obviously, but also NAD and FAD ... yet they also form the 'informational' subunits of a replicator, complete with the energy required for their own polymerisation. No coincidence, I feel - but whether the energetic role grew out of the replicator one, or vice versa, who knows.

Mixtures of amino acids can spontaneously form peptides and all kinds of catalytic activities have been found in these mixtures. No matter how you imagine the origin of life there has to be a long period of chemical activity before you ever get nucleic acid replicators and the genetic code.

All that comes much, much, later.


Yes. The genetic code certainly seems like a much later invention than replication.

It would be surprising if there weren't catalytic activity in random polypeptides. But it's not clear how uncoordinated combination can achieve anything much - or what the coordinating principle is in a metabolism-only world, other than the differential energetic likelihood of some molecules over others, which soon reaches a limit on novelty. Replicators give fundamental organising principles on two levels: they allow the same chemical 'trick' to be performed repeatedly, and once they start to compete for resources, they create a system where better tricks, or better performance, carries a premium.

RaulFelix said...

What I meant is that many of the most abundant amino acids (which are also among those that are thought to be more primitive) are only one or two synthetic steps away from important (achiral) metabolic intermediates. If we think about how dynamic this part of small-molecule metabolism is, it is obvious that, in the absence of very specific chiral catalysis, racemisation would be the natural outcome for those amino acids.
A primitive stereospecific aminotransferase might certainly have been able to recognise every keto acid available, but it shouldn't be too primitive if it were to yield a significant enantiomeric excess. Besides, its (equally primitive) enantiomer would be just as easily available.
It is therefore completely reasonable to suppose that, instead of an early "frozen accident" (which has never been clearly specified), primitive life was involved in a prolonged chirality
war.
If that battle lasted beyond what Woese called the "darwinian threshold" we might have had a period in the story of life when the Earth was inhabited by two types of "mirror image" organisms.

Monado, FCD said...

Even thirty -- forty! -- years ago scientists noted that it was easier for the molecules to form, or to gather and interact, on the surface of clay particles, raising the odds that the molecules would form and modify themselves into new molecules.

Anonymous said...

@RaulFelix

It is therefore completely reasonable to suppose that, instead of an early "frozen accident" (which has never been clearly specified), primitive life was involved in a prolonged chirality war.
If that battle lasted beyond what Woese called the "darwinian threshold" we might have had a period in the story of life when the Earth was inhabited by two types of "mirror image" organisms.


Ummm... I don't think so. This picture is no more clearly specified than the frozen accident. Stage 1 - the putative RNA world - has at its heart a pentose sugar in the D orientation. The primary constraint here is that the stacking of such units in an RNA (then a DNA) polymer selects for isomeric consistency. You are suggesting that the 'spark of life', moving to replicator polymerisation, must have happened twice, once for D pentose and once for L pentose. You may be right - but whichever survives is legitimately termed a 'frozen accident'. There needs to be some demonstration of a selective advantage for any 'chirality war' scenario, rather than the likeliest "null hypothesis", good old genetic drift.

At a later date, we postulate another branching point for amino acids - again, polymerisation is a potential constraint. The D-pentose survivors of the first 'war' hit upon template-based polypeptide catalysis. Do we have any reason at all to suppose that two kinds of ribosome arose, and two sets of amino-acid metabolising catalysts, with their descendants again left to fight it out, chirality being the only fundamental difference between them?

If there ever were 'mirror-image' organisms, any selective differential is far more likely to reside in some accidental inherited fitness difference in the genome than in any inherent superiority of one chiral form over another. If we could look around the universe at planets whose life-forms have discovered RNA and protein, I would expect a 50/50 split between fixation on specific enantiomers. You seem to be suggesting that we would find an excess of one or the other - presumably, 51% of planets (or more?) would have L acids.

I would allow the possibility of coupling - it may conceivably be the case that D sugars select for L amino acids, and vice versa, due to constraints on the ribosome (which is not the same as an inherent superiority for given enantiomers).

RaulFelix said...

@Allan Miller

I completely agree that "if there ever were 'mirror-image' organisms, any selective differential is far more likely to reside in some accidental inherited fitness difference in the genome than in any inherent superiority of one chiral form over another." Nowhere did I intend to portray the "chirality war" as based on any intrinsic merit of each one of the stereochemical configurations. My contention is simply that early life would probably have lacked control mechanisms capable to enforce the predominance of anyone of them, and that there might have been quite a long time before it did.

OgreMkV said...

Here's a recent research article on chirality and how UV could cause it:

http://www.webmedcentral.com/article_view/924

Anonymous said...

I have an layperson question to ask. What is it about L acids as a group that distinguishes them from D acids as a group?

I ask because if I can pick out a glove for a left hand and a glove for a left foot because being human I know what left hands and left feet look like. If I were an intelligent alien octopus I wouldn't be able to do this without knowing human anatomy. What is it about different amino acids that makes them L as a group or D as a group and is there some reason why proteins should be made up only of L acids?

Apologies if this question is totally incoherent and turns out not to make any sense. I'm not a biochemist (obviously)

Anonymous said...

LOL!!!!

Torbjörn Larsson said...

I am currently backtracking some of these "smokers" vs "soupers" posts, and I'll note that the gluconeogenesis/glycolysis analysis isn't entirely convincing today. In what was a first (I think) demonstration of non-enzymatic metabolic like pathway, Keller et al showed that a Hadean gepohysical setting drove glycolysis, and by product separation gluconeogenesis. [ http://msb.embopress.org/content/10/4/725 ; and see the accompanying editorial for gluconeogenesis.] The setting is, you guessed it, a vent dominated anoxic FeII solute ocean with temperature differentials from 70+ degC to 70- degC, so it is more a "smoker" than a "souper" pathway.

There is no reason to believe, if these reaction pathways were environmentally driven at first, the order of later enzymatic co-option would have to promote a specific one of them. But since the environment as well as early cells were leaky, glucose loss, evolving gluconeogenesis co-option first may well have netted a fitness increase.

Larry Moran said...

In order to derive energy from the breakdown of glucose you need to start with an environment that contains a significant concentration of glucose. There is no evidence to support the idea that ancient oceans were sweet enough and no reasonable scenario that could accomplish the goal.