It is my opinion that abiogenesis occurred at a fairly localized site and that it happened in a geological flash. My opinions are only speculation, of course. I will dismiss out of hand the "Thin dilute soup" in which the prebiotic ocean becomes an organic broth. The chemicals of life have to be far more concentrated than they would be in large bodies of water. So our first clue is that we need to think about how they would be concentrated. There are some basic elements:
The region of interest must be enclosed, i.e., it must be a relatively small well defined volume of water which is not connected to a larger volume. This is essential; otherwise the concentration of organics will diffuse away and the requisite concentration will be lost.
It takes energy to build big molecules out of little ones. To build up a concentration of complex molecules there would need to be a continuing process of building these molecules. In pre-biotic Earth there would have been no shortage of energy sources. As a matter of speculation, I opt for a local concentration of U-235 (which was much more common 4000 Myr ago than it is now) rather than volcanic smokers.
One of the things that we have learned from Miller et. al. is that these natural processes which make big ones out of little ones produce a lot of insoluble organics, mostly tars, that precipitate out. They have to go somewhere.
We need a source of the sundry atoms that will go into the making of the organic molecules. In the Miller experiments, CHON comes from dissolved gases in water. Although these would be important I opine that the principal source would be the leaching of weathered rock surfaces.
The enclosed region cannot have been enclosed indefinitely; our new little bacteria have to get out into the big wide world.
Putting these elements together we have something like a deep pit in a cave near the edge of an ocean. The pit is filled with water; there are clay shelf surfaces which the water laps onto. Within the water there is a strip of concentrated radioactive rock which is near critical; the water is hot, around 70 degrees centigrade. There is a continued leaching of the walls of the pit and a continued precipitation of chemicals. What we have is a natural cooker.
An essential part of the process is the presence of chemicals that serve as energy currency, the role that is filled by ATP today. These were probably thio-esters. Another essential part of the process is the formation of lipids -- these are the molecules that form cell walls.
The production of organics in the cooker would not suffice to produce the requisite concentration of organics; for this we need further enclosure. The means is at hand. Lipids very naturally form microspheres; this is important later on. They also can form films over surfaces and that's where the magic happens.
One of the really important things about lipid membranes is that they are permeable to small molecules but not to large ones. Remember those clay shelves that straddle the surface? Those clay shelves that are alternately exposed to air and under water with the tides? Those clay shelves which have films on them?
Small organic molecules, including energy rich ones, constantly diffuse in through the film surface. The clay surface acts both a source and as a catalyst surface. Complex molecules build up rapidly within the region enclosed by a film.
The clay surface is not the only catalyst within the film. The organic molecules themselves are catalysts. What we have here is the makings of Kaufman's autocatalytic set. What are the features of this set? I don't think it was the RNA world at least not as it is commonly pictured. The proteins were always there along with RNA; however they were manufactured by cumbersome pre-cursor reactions rather than by the current RNA synthesis.
What I suggest happened is that there were two critical events. The first is the autocatalytic set phase transition within a film or films. This would be followed by rapid chemical evolution within the film including the evolution of a primitive version of RNA mediated protein synthesis.
We don't have life yet but we are very close. What are we missing? The first thing is heredity. The autocatalytic set maintains its integrity by a network of reactions; there is no single point serving as the keeper of the plans. However within that network of reactions a sub-network that constituted the elements of heredity would naturally develop. In fact, I suggest that DNA appeared fairly early in the set, albeit not with the full functionality that it has today. The second thing missing is autonomy. The autocatalytic set of reactions within the film is dependent on, interlocked with the catalysis supplied by the clay layer.
What is the second critical event? The films would bud off microspheres. These would not, in the beginning, be viable. A living cell is under a lot of functional constraints, ingestion, excretion, the utililization of incoming molecules, and successful division. For a while, the microspheres budding off from one particular film [films came and went] launched off into the pit, "lived" a little while, and then died. But one fine day, 3850 Myr ago, a microsphere budded off and it had its act together; all the pieces were there. And that first prokaryote went out and conquered the pit. In the course of doing so it evolved a lot of tricks; the DNA heredity trick was refined; the protein sythesis trick was refined. And its descendents went out and spread life throughout the world.
How long did it take? Not very long at all. Chemical reactions happen very rapidly. The autocatalytic set phase transition was probably on the order of minutes. The bulk of the time was probably spent in the evolution from proto-prokaryote to prokaryote and that only because the increasing stability induced by reliable heredity would slow down chemical evolution. The total time -- a few hundred years at most. A century is a 100 million prokaryote generations; it's a very long time.
And that's what I think happened.
GCU seems to form more often than predicted, as to why, who knows. (Trifonov EN, and Bettecken T, Sequence fossils, triplet expansion, and reconstruction of earliest codons. Gene , 205: 1-6, 1997 Dec 31) Not that you really need it, one in every 10^17 200 unit random nucleotide sequences is a high efficency ligase anyway (Ekland EH, Szostak JW, and Bartel DP, Structurally complex and highly active RNA ligases derived from random RNA sequences. Science , 269: 364-70, 1995 Jul 21) Also of all possible 1 x 10^130 100 unit proteins, 3.8 x 10^61 represent cytochrome C alone!! (Yockey HP, On the information content of cytochrome c. J Theor Biol , 67: 345-76, 1977)
Experimental Testing of Theories of an Early RNA World
THE ORIGINS AND EARLY EVOLUTION OF LIFE
From Primordial Soup to the Prebiotic Beach
This page was last updated june 3, 1998.