Читать книгу Quantum Evolution: Life in the Multiverse - Johnjoe McFadden - Страница 31

By and large, gene sequence data supports the neoDarwinian notion that gene evolution has involved a series of gradual modifications of existing genes through mutations. Nevertheless, problem areas remain. The first (already mentioned): how to account for apparent big jumps. A related problem, apparent in the DNA record, is the relationship between the major protein families. Examination of genes from diverse organisms has established that all modern proteins fall into about a thousand distinct protein families’. Although evolution within protein families, such as the globin (the protein in haemoglobin) gene family, can generally be traced through a number of antecedent proteins present in living creatures, finding the links between the protein families is much more difficult. Animal globins bear some relation to oxygen storage proteins found in bacteria but there is little or no identifiable relationship between these globin-related proteins and any of the nine hundred and ninety-nine or so, other protein families. The same is true for all the other protein families – there is much evidence for Darwinian evolution within the family, but no obvious close relative from which the family could have evolved. Each protein family is like a separate galaxy (of related proteins) in a vast outer space of protein sequences. New protein families must have arisen from existing proteins by some kind of mutational process but how their sequence traversed this vast empty sequence space devoid of Darwinian intermediates, is a mystery. It seems molecular evolution often proceeds though a series of small steps but that sometimes it takes big leaps – rather like the punctuated evolution envisaged by Gould and Eldridge. Big leaps are big problems for neoDarwinian evolution because the chances of a big jump landing anywhere useful are generally thought exceedingly small. As Richard Dawkins states, ‘However many ways there are of being alive, it is certain that there are vastly more ways of being dead …’³

Another problem for the neoDarwinian process is the evolution of metabolic pathways. This is a kind of molecular version of the eye problem – how to evolve complex structures – but its solution is not as apparent as the evolutionary pathway that led to the eye. The basic problem is that the complexity of biochemical pathways (unlike the eye) do not appear reducible. For instance, one of the cell’s essential biochemicals is AMP (adenosine monophosphate) the precursor of ATP (the energy-carrying molecule) which also finds its way into DNA, RNA and many other cellular components. AMP is made from ribose-5-phosphate, but the transformation involves thirteen independent steps involving twelve different enzymes (which we will represent as: A→B→C→D→E→F→G→H→I→J→K→L→M where A is ribose 5-phosphate and M is AMP). Each of the twelve enzymes involved in this pathway is absolutely essential for the biosynthesis of AMP. Darwinian evolution would require this complex system to have evolved from something simpler. But, unlike the eye, we cannot find the relics of simpler systems in any living creatures. As far as we know, nothing simpler works. Half or a quarter or a twelfth of the pathway does not generate any AMP or indeed anything else of value to the cell. It appears that the entire sequence of enzymes is needed to make any AMP. But without viable stepping stones, how can the entire complex system have evolved through Darwinian natural selection?

One explanation often cited is that complex biochemical pathways have evolved backwards. The story goes that the primitive cell initially utilized the final biochemical in the pathway (M or AMP) directly, as it was already available in the primordial soup. However, as primitive cells used up the supplies of M, any cell that evolved the capability of making M from another available biochemical would have had a selective advantage. One of those biochemicals was L, and a cellular innovator soon evolved an enzyme which could perform the transformation of L → M. Eventually supplies of L were, in their turn, depleted, creating selective pressure for a second evolutionary step to acquire the enzymatic capacity to make more L from one more readily available biochemical, K. Eventually, the entire pathway: A → B → C → D → E → F → G → H → I → J → K → L → M, evolved through this series of backward steps.

The problem with this explanation is that it requires all of the intermediate biochemicals (B,C,D,E,F,G,H,I,J,K,L) to have been sloshing about in the environment of the primitive cell. Yet AMP is a ribose sugar, that goes into making RNA. We will soon be exploring the enormous difficulties in making ribose sugars by the kind of inorganic chemical processes going on within the early Earth. As you will see, it is unlikely that even one, never mind all eleven, of the biochemicals in the pathway from ribose 5-phosphate to AMP was present in any significant quantity in the environment of the primitive cell.

In his book Darwin’s Black Box, the biochemist Michael J Bethe of Lehigh University considered the evolution of the AMP pathway and several other complex biochemical systems such as the mechanism of the bacterial flagella. Bethe contends that these complex biochemical systems cannot be broken down into a series of (forward or backward) steps subject to Darwinian evolution. He maintains that their evolution is totally inexplicable in terms of standard evolutionary theory or indeed any scientific theory. Bethe’s solution is either radical or hopelessly archaic depending on your point of view; he proposes that the existence of complex biochemical systems is due to God’s intervention in the evolution of life on Earth.

A third problem for the standard evolutionary theory is the apparent existence of a heretical type of mutation. Standard neoDarwinian evolutionary theory predicts that mutations occur randomly with no respect to the direction of evolutionary change. Natural selection provides the direction of evolution by selecting hosts with beneficial mutations; but those mutations are generated randomly. This does not mean that all sites in DNA have the same mutation rate. In fact we know that there are regions in most chromosomes that are highly mutable (probably due to their local chemical environment, or because they bind mutation-promoting proteins). However, it does mean that the mechanisms that introduce mutations into DNA are presumed not to know which bases are likely to generate advantageous mutations if they mutate.

Nonetheless, when John Cairns of the Harvard School of Public Health in Boston set out to test this reasonable premise he found things were not quite so simple. Cairns incubated cells of the common gut bacterium E. coli in conditions where a single mutation could rescue them from starvation. He used an E. coli lac^- strain deficient in an enzyme called β-galactosidase (β-gal) needed for the cells to eat lactose (milk sugar). He then fed the cells on a diet of only lactose. Without β-gal he expected all the cells to starve. In fact it takes a lot to kill E. coli by starvation, mostly the cells just shut up shop and go into what is called a stationary phase, where either they don’t replicate or only very slowly. E. coli cells can survive for many weeks in this stationary phase.

Cairns fed a parallel culture of E. coli cells on yeast extract, which the cells could eat without need for the β-gal enzyme. The standard neoDarwinian theory would predict that the mutation lac^- → lac⁺, to generate a fully functional β-gal enzyme, should occur at the same rate for the cells fed on yeast extract, compared with cells on the starvation diet of lactose. The only difference should be that, for the cells fed only on lactose, the mutation would rescue them from starvation; whereas the mutation would be irrelevant for the cells happily feeding on the yeast extract. What Cairns actually found was a much higher rate of lac^- → lac⁺ mutation when the cells had only lactose to eat. Cairns examined other genes but their rate of mutation was unchanged by starvation, indicating that the phenomenon was not caused by a general increase in mutation rate.

These adaptive mutations suggested that a starving cell could sense that it was starving and somehow choose the gene it needed to mutate to save itself from starvation. Cairns’ paper describing adaptive mutations was published in Nature in 1988⁴, unleashing a storm of controversy. The difficulty was that there was no known mechanism which could allow the environment of a living cell to influence the targeting of DNA mutations. The direction of information flow in the cell is from DNA through RNA to protein and outwards to the environment. There is currently no known mechanisms by which information can flow backwards from the environment to DNA to account for these mutations.

Since 1988, hundreds of publications have appeared that have either supported or denied the phenomenon of adaptive mutations. Adaptive mutations have been proposed to occur in many types of bacteria as well as more complex yeast and animal cells and have even been implicated in cancer. One of the most impressive demonstrations of the phenomenon was by Barry Hall of Rochester University who demonstrated that two sites just a few bases apart on the same DNA molecule could be subject to widely different mutation rates, dependent on whether or not those mutations were adaptive⁵. Whatever their mechanisms, adaptive mutations appear to be able to bias the mutational process to favour certain genetic changes.

I must emphasize that though there may be some doubt concerning the mechanisms involved in evolutionary change (and there are likely to be many), this should not be confused with any doubt concerning the process of evolution itself. There is overwhelming molecular evidence that all modern species have evolved from earlier species. Indeed, there is considerable evidence that we have all evolved from a single ancestral cell. Let us next examine the probable nature of that common ancestor.