In my lab, we perform a version of combinatorial chemistry when we prepare liposomes by adding water to a few milligrams of dry lipid in a flask. A milky suspension is produced that contains, not thousands, but trillions of individual microscopic vesicles in the size range of small bacteria–half a micrometer in diameter. If the vesicles are prepared in a solution containing small peptides and short nucleic acids such as RNA, each of the vesicles will contain a different set of components, so each represents a microscopic experiment. Now let’s think about the early Earth. Instead of milligrams of lipid in a flask, it would have had billions of tons of organic material assembling into enormous numbers of microscopic structures, and half a billion years to do the experiment.
The origin of life can be understood metaphorically as combinatorial chemistry on a global scale. A few of the microscopic experiments must have been successful, resulting in primitive cells capable of capturing energy and nutrients in order to grow by means of polymerization reactions. Evolution began when the cells filled a limited niche and competed for resources. At that point, natural selection took over, placing a premium on how efficiently a given cell could capture nutrients in order to grow. I imagine that once robust cellular life got under way, it expanded exponentially. Earth, seen from space, may even have blushed red or turned green for a while when photosynthetic bacteria filled the oceans.
Will we ever discover the combination of ingredients that gave rise to life? Again, I am optimistic. We need to apply what we know about the chemistry and physics of living systems to narrow down the possibilities, then be brave enough to actually do some experiments. But what experiments should we try? This is where theory can guide us. Freeman Dyson, one of the great theoretical physicists of our time, has also taken an interest in the origin of life. In his book Origins of Life, Dyson succinctly summarizes what I have told you:
Life began with little bags, the precursors of cells, enclosing small volumes of dirty water containing miscellaneous garbage. A random collection of molecules in a bag may occasionally contain catalysts that cause synthesis of other molecules that act as catalysts to synthesize other molecules, and so on. Very rarely a collection of molecules may arise that contains enough catalysts to reproduce the whole population as time goes on. The reproduction does not need to be precise. It is enough if the catalysts are maintained in a rough statistical fashion. The population of molecules in the bag is reproducing itself without any exact replication. While this is happening, the bag may be growing by accretion of fresh garbage from the outside, and the bag may occasionally be broken into two bags when it is thrown around by turbulent motions. The critical question is then, what is the probability that a daughter bag produced from the splitting of a bag with a self-reproducing population of molecules will itself contain a self-reproducing population? When this probability is greater than one half, a parent produces on the average more than one functional daughter, a divergent chain reaction can occur, the bags containing self-reproducing populations will multiply, and life of a sort has begun.
The life that begins in this way is the garbage-bag world. It is a world of little protocells that only metabolize and reproduce themselves statistically. The molecules that they contain do not replicate themselves exactly. Statistical reproduction is a good enough basis for natural selection. As soon as the garbage-bag world begins with crudely reproducing protocells, natural selection will operate to improve the quality of the catalysts and the accuracy of the reproduction. It would not be surprising if a million years of selection would produce protocells with many of the chemical refinements that we see in modern cells.