Dangerous Experiments is the LabSpaces spot for guest bloggers. The purpose of the blog is to give new and old bloggers a space to experiment with blogging. If you'd like to contribute to this experiment, send us an e-mail or contact us on twitter at either @LSBlogs or @LabSpaces.
My posts are presented as opinion and commentary and do not represent the views of LabSpaces Productions, LLC, my employer, or my educational institution.
Please wait while my tweets load
This week's guest blogger is Rachana Bhatawdekar. She's a budding astrophysicist currently traveling Europe and Asia. You can find her and a bunch of sciency tweets on twitter as @astrogeek03.
When astronomers look in to deep space they can ‘look-back’ in to time, billions of years. Now I can understand it can take light a while to get here, but how did WE get here, considering we originated in the big bang just like all the stuff created whose light reaches us many years later?
Was the emergence of life unavoidable? Is it the result of a process that would have had to occur sooner or later? Or else is it the outcome of coincidences so improbable that time spans much longer than the age of the Universe would be insufficient to explain it by a random process?
Of course, when one has plenty of time, even the improbable becomes possible. When one plays dice for a very long time, one always ends up by throwing a double six three times in a row.
The theory of evolution of species initially gave rise to much disbelief. In the 19th century, this was in part because geological periods were believed to be much shorter. Today, it is probably because the disbelievers cannot imagine the immensity of the time lapse corresponding to four billion years. However, the recent phylogenetic studies have convinced all serious scientists that evolution is no longer in doubt. The most important unsolved problem is the time span needed to produce the first bacteria. Between the decline in the number of large cometary impacts and the existence of the first fossil bacteria there cannot be much more than 200 million years. In our ignorance of the mechanisms needed to assemble the first self-reproducing cell, can we assume that a sufficient time has elapsed? Most biologists answer in the affirmative. But in 1970 the French biochemist Jacques Monod wrote a widely read book entitled Chance and Necessity wherein he expressed the opinion that the origin of life was the result of a very unlikely chance that would not repeat itself elsewhere. However, Monod was unaware of all the more recent evidence that suggests otherwise.
Other scientists have addressed the residual problem of the time span needed to put together the first bacteria. In particular, the well known British astrophysicist Fred Hoyle and his Sri-Lankan collaborator N. Chandra Wickramasinghe have endeavored to demostrate that it is utterly impossible to build a bacterium in so little time. This is a useful effort, because it compels us to reexamine the situation. They started from the three-dimensional form of an enzyme, which gives it its specificity (for instance, its ability to fit the shape of another molecule in the way a key fits a lock). They computed that there is a 1 in 1015 chance of assembling amino acids by chance to build the required geometric solid. Then there is at most 1 chance in 105 of positioning the active site of the catalyst at the best solution. Thus, there is only one chance in 1015+5 = 1020 of obtaining the required enzyme that is able to function. Trying to assemble it by chance at the rate of 1000 combinations per second would require 3 billion years !
This is not the problem, because it could be tried in a billion different locations in the early seas. Only three years would then be required for the enzyme to form by chance in at least one location. The problem lies in the fact that, in order to make a basic bacterium, about 2000 enzymes are required, each having a specific shape and a different catalytic action. In order to make them all by chance, at least 1020*2000 = 1040,000 trials are needed.
The huge improbability of assembling a bacterium by chance within the 100 or 200 million years available on the primitive Earth is used by Hoyle and Wickramasinghe as the starting point for their hypothesis that bacteria preexisted in comets.
We now return to their probability assessment to show where it is misleading. As a matter of fact, their computation is based on pure chance. They attach equal probabilities to all possible cases, which comes down to constructing a bacterium all at once from nothing. Despite its small size, it is true that a bacterium would be nearly as complicated as a Boeing 747, but there is no reason to assemble either a Boeing or a bacterium in a single operation. Of course, the difficulty lies in the nature of the evolutionary process, when all evolving proteins must help one another and act in concert for the sake of the whole organism.
Research has begun to elucidate this question; it seems to imply that the emergence of life comes from the collective properties of the polymers that show catalytic features. This explains the recent interest in the fact that RNA is autocatalytic. Eigen has recently helped to clarify the problem. He has studied the way in which what he calls chemical ‘hypercycles’ can evolve. The main point of a hypercycle is that it involves several chemical reactions coupled by several feedback loops acting upon one another. The simplest feedback loop (e.g. in a central heating system) gives a false sense of finality. When many feedback loops are coupled, this feeling is even more marked, because the system seems to ‘know’ how to influence the future. Chemical hypercycles have now been studied in the laboratory, and their spontanenous evolution in the course of time shows natural selection in action, as if it ‘knew’ the way to build more stable and more complex hypercycles.
Lets say there are three reaction products A, B and C that act indirectly upon each other. A catalyses the reaction leading to B, B does the same for C, and C for A, closing the feedback loop. If some of the reaction products A’, A’’,A’’’ etc. are less fit to survive, they trigger the disappearance of their whole cycle. The same happens for B’, B’’, B’’’ and for C’, C’’, C’’’ etc. If A survives, this is because it has selected the right choices for B and C, and vice versa. The hypercycle is therefore a machine for selecting and encoding the proper information for survival at least expense, and beginning with zero information. In a word, it is the Darwinian mechanism of evolution of the species and of survival of the fittest that has just been moved back to the level of a purely chemical process.
Laboratory experiments have now shown that the short period of time needed to begin early life on Earth is not a problem. With a simplified enzyme including only few amino acid residues and maybe some fragments of RNA, a hypercycle could have produced the first ‘protobionts’ in much less than 100 million years. Thus the way would have been open for the first bacteria to appear.
All this can be summarized by emphasizing that Prigogine’s ‘dissipative structures’ that are needed for life can be developed from scratch, with no previous information needed, by using Eigen’s hypercyles to set in motion the evolution of species and the survival of the fittest, at an extraordinarily simple chemical level.
Thus, genetic information is coded from the beginning by the survival of the most suitable chemical process, and it accumulates in small stages which are the most probable at each time point. The questions raised by those who feel intuitively that life is a phenomenon too complex to emerge simply by chance no longer stand up, and the quantitative statement of this problem, as expressed by Hoyle and Wickramasinghe, has proved to be totally misleading. We can in fact readily accept that ‘life’ is a very probable physico-chemical phenomenon that will appear soon after the prerequisite conditions are met. On the Earth, it could easily have emerged in the time available after the biosphere emerged as a result of the cometary bombardment.
This post has been viewed: 1379 time(s)
A very interesting article! However, your focus seems to be on the emergence of bacteria as the start of life, If we are to consider the formation of autocatalytic RNA as part of the process towards life, what about the place of viruses in the evolution of life? RNA is a fundamental part of viruses, and it is possible that in one of the billions of pools across ancient Earth the RNA could have reacted with enzymes to form primitive strands of DNA.
Also consider the possibility that these early pools of enzymes and autocatalytic RNA could have imitated some of the functions we've come to expect from cells until outside reactants, such as changes in pH or H2O that could have come from comet impacts, forced changes in the interaction of "native pool" reactants. Through millions to billions of years such changes would amount to the tried and true stimulus - response, an environmental change forcing a corresponding change in the reactions and formations of RNA /DNA /enzymes. It could be said that the formation of life was an effort by these primitive enzymatic "pools" to escape extinction, and that subsequent mutations seen in the formation of various life forms as an effort by the RNA /DNA /enzymes that make up all life to duplicate the effect of the billions of pools on ancient Earth - to give the molecules that make up our cell nuclei the broadest and best possible chances of surviving all that the universe is throwing at it! And those changes has thus far culminated with humans - what better survival adaptation for the RNA /DNA /enzymes than to evolve a being that can escape Earth spanning extinction events?
Speaking of the universe throwing stuff at the pre-life enzymatic pools, and specifically comets, I dismiss the idea that Earth was seeded with bacteria from comets since it still does not answer the question of how life came to be, and is a cop-out on trying to find the real answer to life's emergence.
I'm a science fiction writer with one of my three degrees being an associates in biology, and it's always a pleasure to discuss the possibilities of how we, and life, came to be. Thanks to Labspaces for tweeting a path to this article. :)
R. C. Qualls
The events leading to the first bacterial cells could have started whenever chance processes produced somethng that natural selecion could act on (something with replication and heritable variation). Eigen's hypercycles are one way to do this.
Life's "dangerous experiment"; much appreciated!
- Whenever I hear of Monod (which I haven't read), I get the feeling that he isn't considering an actual statistical process! It sounds more like a dynamic process, where in a vast phase space that is frequently revisited he puts a small volume where there is 'life'.
It is trivial that when you look at stochastic processes, a simplest possible Poisson process model for abiogenetic attempts is both suitable and informative. Life within ~ 1 Gy out of ~ 5 Gy history makes a normalized delay of ~ 0.2, and since exponential distributions stacks their probability mass so early this actually looks to have 3 sigma testability!?
In a deterministic interpretation, abiogenesis attempts are frequent and/or they are easy. Conversely from P.m. ~ 30 % of ~ 5 Gy old habitable worlds could have life, which rapidly rises to ~ 100 % over some delays (< 1 Gy) time. So informative, while bad stats (1 data-point).
- The death under the impact tail has been vastly exaggerated. ["Microbial habitability of the Hadean Earth during the late heavy bombardment", Oleg Abramov et al, Nature 2009.] Cells multiply and spread faster than sterilization ("life is a plague"). In later papers there is a "goldilocks" crustal habitable zone ~ 1 km down that fails crust buster sterilization.
Life could have started out right away. Mantle diamond inclusions @ ~ 4.3 Ga tells of a vast reservoir of light 13C with fractionation that only the acetyl-CoA pathway reach @ ~ - 50 o/oo. (Fischer-Tropsch abiotic reactions reach a steady ~ - 30 o/oo 13C fractionation; see McCollom et al IIRC.)
- Some biologists with what looks like chemical expertise like Orgel has criticized the viability of autocatalytic cycles. They "leak" nonfunctional side product and IIRC extinguish. (Don't have the ref handy.) Has this changed, which lab experiments are we discussing?
Just for kicks (and since I can quickly adapt a recent comment of mine), here is my take on the necessity of chemical evolution:
DNA-protein cell machinery, RNA or ATP biosynthesis before the first membranes, the first enzymes are examples of (not fully exclusive) common evolutionary chicken-and-egg problems. Luckily such problems conveniently bottleneck possible pathways to a smaller set.
Bottom up, chemical network enzymes are a natural outcome in newer scenarios. High-temperature reactions seems to be much faster than orthodox theory believed from scant data. This temperature dependence gives a self-selection for enthalpic pre-proteinous enzymes. ["Impact of temperature on the time required for the establishment of primordial biochemistry, and for the evolution of enzymes", Stockbridge et al, PNAS, 2010.]
Now looking top down, we see that pathways meet. The first modern metabolic networks originated with purine metabolism, and specifically with the gene family of the P-loop-containing ATP hydrolase fold. ["The origin of modern metabolic networks inferred from phylogenomic analysis of protein architecture", Caetano-Anollés et al. PNAS, 2007; "Rapid evolutionary innovation during an Archaean genetic expansion", David et al, Nature, 2010.]
That is, ATP sits at the intersection between a cooling and/or hydrothermal vent active Earth prometabolism and the nucleotide protometabolism. (That later seems to have been exaptated by modern proteinous metabolic genes as coenzyme/energy currency.) Minimum change of traits picks ATP use before RNA evolution.*
* Note that this is an (informal) test of a phylogenetic pathway. Abiogenesis is actually slightly testable today as far as I can see; and see my previous comment.
I think Shostak's spontaneous assembling protocells may be testable as well; they are necessary both for RNA polymerases self-replication (since their promiscous nature are prone to self-extinction otherwise) or mature ATP biosynthesis (cellular metabolism). They too straddles the RNA world gap.
You might find this interesting:
Life might be rare despite its early emergence on Earth: a Bayesian analysis of the probability of abiogenesis -
[ - Is this thing still on? Thumps thread. An answering comment makes him jump. - Why, so it is!]
Thanks, Rachana. Yes, I was actually currently looking for such models, it is again much appreciated!
I will look it over. unfortunately I don't know much about bayesian modeling as regards stochastic processes, I have only the basic stuff under my vest. I do know bayesian modeling is emulated by hidden markov models, so there should be some connection.
I don't want to make this a comment dump, but I added something on a whim last time that doesn't hold up. The claim that self-assembled membranes is a chicken-and-egg problem for cellular ATP production specifically is wrong.
I was thinking of modern metabolism, where chemiosmosis over membranes stand for the majority of ATP production. However, ATP is also a net product out of glycolysis in the cell cytosol, not dependent on membranes.
On the contrary one can entertain the notion that glycolysis is another early trait. As far as I can see it is claimed to be extant in all cells, and incorporates key steps for both anabolic and catabolic pathways.
It is also notable in this context of abiogenesis and early ATP that the steps that lies around the start of the pay-off phase (later stage of glycolysis) are populated by DHAP and GADP. <a href="http://books.google.se/books?id=U_QqAd1QlIgC&pg=PA18&lpg=PA18&dq=dhap+abiotically&source=bl&ots=VVz16iNZEj&sig=WoRRyOtM3GI-3c1GcjBHzoCNmig&hl=sv&ei=-0cnTurOPM-QswbKoqW3CQ&sa=X&oi=book_result&ct=result&resnum=4&ved=0CD0Q6AEwAw#v=onepage&q&f=false">DHAP and GADP have been claimed to be abiotic products of formaldehyde in alkaline fluids with FeS</a> (say, in hydrothermal vents).