Pross, Addy. 2012. What is Life? How chemistry becomes biology. Oxford University Press. 200 pp.
Chapter 5: Origin of life
Pross gives a summary of research on this question that seems fairly reasonable, although he clearly doesn’t think much of historical approaches. I wonder whether he is not giving enough credit to geochemical analysis of rocks from the period before we find microfossils, that is to possible evidence of biogeochemistry back before the oldest fossil organisms. Also, he has not mentioned cosmochemistry – what was available in the part of the solar nebula that became the earth? None of that evidence in itself would answer the question, but he earlier talked about how historical studies could supply useful constraints on the free flow of speculative ahistorical studies of prebiotic chemistry.
He says sequence analysis fails on the origin problem because of horizontal gene transfer. If you start to see networks instead of trees, he claims that you can’t tell anything from the results. Is that so, or is that just a further challenge for clever analysts to overcome? After all, trees took a while to be generally useful. There still are lots of difficulties, but horizontal gene transfer isn’t just chaos. The process must have some logic, ultimately controlled by natural selection, like “normal” vertical gene transfer. I think he might be giving these approaches short shrift, because he has his own agenda.
He also assesses RNA world as unlikely, given the failure to create really complex self-catalyzing molecules in decades of lab studies. This despite his earlier claim that negative findings could not be used to rule out this very scenario. Well, if there was an RNA world, we haven’t been able to create a similar thing in vitro.
The other current scenario has a closed metabolic cycle evolving before self-replication kicked in. He calls such a cycle, “holistic autocatalysis.” So far, attempts to develop such systems by evolution in vitro have also not gone very far, according to Pross.
Biology’s Crisis of Identity
Pross asks three questions: what is life? How did it originate? And how would one make it? He says biology has reached a point, with the completion of the human DNA sequence project, that physics had reached in the late 19th century, prior to relativity, quantum mechanics and subatomic particles. How you can judge the state of mind of a body of scientists, I don’t know, but such an assessment feeds into his attempt to portray himself as breaking through confusion and complacency. To him, the problem is complexity. Is complexity a substance? Can there be a theory of complexity, as opposed to a complex theory?
Does all complexity go back to symmetry breaking, like quantum theory says, if I understand correctly? Life’s complexity clearly arises from the pure combinatorial possibilities of sets of fairly simple elements – four nucleotides, twenty amino acids, thousands of enzymes and similar numbers of intermediary products to create all those metabolic cycles. But they wouldn’t be of much use in a totally homogeneous environment. That’s the competitive exclusion principle. Life is complex because it exists in a large and complex environment, whose complexity is the result of irregularities in composition and past impacts, etc. leading to plate tectonics, and the uneven heating of a rotating almost sphere by the sun, leading to circulation of atmosphere and hydrosphere.
Pross says, “It is the organization of life, rather than the stuff of life, that makes life the unique phenomenon that it is.” Well, duh. He says “systems biology,” which tries to explain cells functions using mathematical ideas like “network topology, ” has not produced much in the way of insight. He also says that a holistic approach can be reductionism “dressed up.”
Another favorite of complexity mavens: non-equilibrium thermodynamics. Life, Pross says, can be said to be a dissipative structure, but what further insight comes from that? None, he thinks.
He then turns to John Conway’s Game of Life, the cellular automaton computer program, beloved of Gaia worshipers. These programs illustrate how simple deterministic games can generate complex patterns, but like the physical insights into complexity, there mathematical discoveries don’t seem to throw light on what Pross claims is the tough question about life: how does teleonomy arise within non-teleonomic worlds? I wonder if there is a fallacy in looking for the origin of “apparent purposiveness” when things apparent are clearly in the mind of the beholder. Can science find any sort of purposiveness at all? That’s a philosophic problem, as Socrates pointed out long ago. And as to “apparent purposiveness” is that anything at all? It’s not hard to explain how natural selection acts to give things apparent purposiveness: purposelessness is clearly maladaptive, it is not bothering to try. Is this his great insight?
Biology is Chemistry
The answer, he says, lies in systems chemistry. What defines it is that it deals with simple chemical systems that have life-like properties of self-replication. After dismissing all the previous attempts involving RNA or metabolic cycles, what is he offering that is different? He starts by justifying all over again the utility of simple systems, with the argument that since we think life started from simple stuff it will be informative to experiment with simple systems. This, however, is unproved: what if comets bombarded the proto-earth with really complex stuff, like Buckyballs and other cosmic macromolecules? Also, this comes after he says that we have no idea what sort of simple stuff life came from. I wonder if he’s headed for another case like those he dismisses.
He claims that systems chemistry is like looking at the Wright brother’s flyer to understand flight, as opposed to a 747. That is, if we can strip down to the simplest possible replicating system, we can get somewhere. But he just said that’s not possible because we don’t have any idea what the earliest living organisms were like. As if we did not know anything about airplanes prior to say, WWII, and we’re trying to imagine the ones from1903, could we do it? He seems to be saying both yes and no.
So here comes his “bombshell,” Darwin applies to replicating chemical systems, thus removing the distinction between chemistry and biology. Fine. But if this is really a momentous original discovery, a lot of folks must not have been thinking very clearly. Anyhow, we know Darwinian theory can apply to designing electrical circuits, why not replicating molecules? But can you actually use that to account for life on earth, more than just in principle? Now he brings in competitive exclusion, and we are off to the races. How well can you demonstrate this principle in a purely chemical system? He says replicating RNA molecules competing for different substrates, evolved to optimize their use of two different substrates, thus precisely mimicking the evolution of Darwin’s finches. Well, precisely is putting it a bit strongly. He claims totally without conclusive evidence that the finches are only doing what molecules were doing five billion years ago. He says that somehow replicating molecules transformed into living cells. I agree, but this is no profound insight, just an attempt to dress up a few clever experiments as a major breakthrough. And maybe the fact that a chemist can learn something from paying attention to ecology and evolution.
The earlier chemists, whose work he seems to dismiss, we’re studying the same things as he is, and he still has no idea what molecules to study. It seems exactly like non-equilibrium thermodynamics or systems biology or Game of life: some clever demonstrations, but no meaningful answers. On pages 132-134, he cites experiments that laboriously mimic the process that was already obvious, that evolving systems become more complex over time, but actually the experiment only shows that two interacting molecular species replicate more efficiently than a single species. Cross catalysis, in this case, speeds things up. So is all life one giant cross catalytic system? Of course it is. Herclitus’s ONE:EVERYTHING::EVERYTHING:ONE holds. Yes, it is chemical; life is an interacting system of macromolecules in an aqueous medium, but it is more. For one, it is largely cellular. Why? Can Pross explain that transition from chemistry to biology with more than a somehow?
Pross wants to add complexification into the sequence replication, mutation, selection, evolution. He puts it after mutation, but that makes no sense, and in his experiment it was the experimenter who in effect introduced it. Even the bare sequence is not right. Evolution doesn’t belong. It is not inevitable, it only happens if the frequencies of the interacting elements change, and that requires an outside physical/chemical/biological cause, a selective force. The system only evolves because of some constraint. Complexification is not a force, no more than evolution; it is the outcome of selection operating under varying conditions. It isn’t a cause. Evolution is change. Complexity is variability, they are not causes, they are results. True, it seems as if complexity is somehow auto catalytic, generating more and more complexity, but there is no law that says that has to be. Diversity does not necessarily result in stability or increasing diversity. Those outside constraints ultimately set the limits. Pross knows a little ecology and evolution, but not enough.
Pross says chemistry and biology are connected by a complexity continuum. What does that mean? Just that he’s repeating his claim in a different way? Wouldn’t discontinuity be more complex? His holistic claims seem more like good old reductionism dressed up. Is his bridge between the two more than just analogical? Physically, of course, it is the same stuff, but until you can actually make molecules evolve into living cells, what have you added to our understanding?
Is the first gene or the first enzyme buried somewhere in our cells, still doing a job, albeit not necessarily what it did billions of years ago? Or did it go the way of the protobiont and so many other species that are now extinct? If we could reverse engineer a simple bacterium into an even more minimal creature, would we be replicating our now vanished ancestors, or just making test tube freaks that never could have competed in the biosphere? Pross says the bacteria have remained simple, but how does he know? Is the bacterial component of the biosphere becoming ever more complex, just in a different way, than the higher plants and animals?
Assume he’s right, and some bit of RNA started the whole thing. Did it manage to do this in some primordial soup competing with uncounted numbers of other molecules, or was it in some incredibly sheltered, simplified environment, like those laboratory test tubes? One thing you don’t have to worry about is sufficient numbers to let mutation and selection act on. Enough might be produced in seconds, if you hit on the right mix. Even if it was much less rapid, as Pross notes, there was certainly plenty of time back then.
Natural selection is kinetic selection
Are competing organisms much like competing molecules? That’s a very loose analogy. Organisms don’t just compete for substrate. He claims we have to explain biology in the language of chemistry, but he uses all language very loosely. He really makes an unwarranted jump in equating chemical kinetics with biological reproduction. If you say that one species winning out over another is just chemical kinetics, I think you will get demurrals from most biologists. He’s back to hiding crude reductionism under his holistic claims. What he says about chemical systems being more amenable to mathematical analysis is just wrong, too (p. 139-140).
Fitness equals dynamic kinetic stability
He’s already in trouble by claiming fitness is a population phenomenon, not an individual one. Even chemically, I’d say that’s dubious, although there may be a population aspect. He is shoehorning a biological idea into a much simpler chemical concept. He claims you can focus on the population aspect, evidently without considering the individuals. But that is just wrong. The only real aspect of fitness is which individuals are the parents of future generations. Who is going to have descendants? Perhaps highly predictable with molecules that replicate. Not so easy with organisms. Even in general it isn’t easy. Who would have picked out the ancestors of angiosperms and placental mammals in the Jurassic? Connecting fitness to stability seems hugely wrong. On the level of the persistence of simple forms, maybe. Lots of genes seem not to have changed all that much.
His attempt to explain fitness landscapes and to make an analogy to a flock of birds seeking higher peaks is not particularly helpful, and didn’t that come from Richard Levin’s work in ecology? Actually the Eigen-Schuster Quasispecies concept is a neat mathematical formulation, but it is not clear what it applies to. Maybe viruses, maybe the origins of DNA RNA transcription/translation! maybe sex (see Wikipedia on quasispecies model) Certainly nothing like all evolving species. This is another analogy that seems to break down on close inspection. He’s trying to bridge the gap by forcing these analogies to do more than they are suited to do. After all, the real unification would mean that you can reduce equations of population genetics to chemical equations, doesn’t it?
He ends up not making a clear connection to the quasispecies concept and goes on to talk about his dynamic kinetic stability, which he admits can’t be measured absolutely, just like fitness, which also depends on the environment in which it is measured. Given how vague DKS seems, it does share the character of “fitness,” in as much as both can be what you want them to be. He suggests (p. 146-147) two measures: abundance and persistence, that are like part of Wilson’s definition of ecological success.
Incidentally, why does he not discuss the Eigen-Schuster hypercycle idea, which seems like a real theory of evolution of simple replicating molecules into linked pathways?
He now says that the cause of evolution is the drive toward greater DKS. But isn’t the cause self copying, with imperfections in a variable, limited environment? It’s differential reproduction, not any drive to achieve stability in any sense of stability I understand. A driving force towards something that he admits can’t be quantified and a mechanism that is a process of becoming a mechanism that is made up of more diversely interacting components (complexification) Seems pretty incoherent to me. He can’t put this into an equation, can he?
In arguing for the idea that life has undergone complexification he points to the fragility of self replicating molecules in the lab. I don’t see that that self-evidently applies to the first replicators in nature. Maybe we are all descended from a horrendously tough little replicator that just happened along out of the seemingly infinite possibilities. Maybe there are theoretical limits set by the problem of mutation in a small set of elements, something seemingly discussed by Eigen and Schuster. Small sets are inherently unstable, so it’s hard to conserve the replication when the replicates are too unlike the original. If a sequence is going to assume the role of a template, or even just determine catalytic properties, it can’t vary too much. Isn’t that just a trivial result, though? It sounds more profound if you introduce the term information into the discussion, but is that really necessary? Jacob Klein always denied that what geneticists talked about was information. I’ll stop at this point, because I think I have about reached my limit in thinking about where life comes from. Pross has made an interesting attempt to define a new agenda for research in this area. I don’t think he’s got anything really significant, though. Perhaps if we can ever find another biosphere to examine, we will see just how narrow or how loose the constraints are.