2.) Exotic five-carbon sugars
Some five carbon sugars are very rare in nature, so very few organisms have the ability to use these exotic compounds in their metabolism. Robert Mortlock determined that the bacteria Klebsiella aerogenes was not immediately able to metabolize D-arabinose and xylitol by growing strains in media containing those compounds and noting the strains that were able to grow only after a lag time. This indicated that the original strain did not have the ability to process the compounds, but was able to evolve such a capability. Mortlock then went on to see how this capability was evolved.
In the case of D-arabinose, Mortlock showed that the arabinose could be utilized if it could be converted to D-ribulose by an enzyme (an isomerase). Unfortunately, K. aerogenes has no such isomerase for the conversion of D-arabinose. However, the isomerase for L-fucose has a low activity for D-arabinose. But, the bad news is that the L-fucose isomerase is normally produced only when the cell is exposed to fucose. Nonetheless, in a few individuals, mutations occurred that allowed the fucose isomerase to be produced at all times - not just when L-fucose is present. This is normally a bad thing and would be selected against because it wastes the cells resources by constantly producing an unneeded enzyme. In this situation though, the mutation is a very good thing, and allows the cell to survive because it can now metabolize arabinose (albeit rather poorly). Although production of the fucose isomerase has been deregulated, the structure of the isomerase itself has not been changed. The next mutation was a change to the isomerase to make it more effective in the conversion of arabitol to ribulose. Finally (although I can't tell from Bell's description if this was actually done in the experiments), the culture could be selected to regain control of the expression of the isomerase - so that it is produced only when arabitol is present.
Xylitol is also not normally metabolized, but Mortlock and his colleagues were able to develop strains (generally through spontaneous mutations, but sometimes with u.v. ray or chemical induced mutations) that could use it because ribitol dehydrogenase (which is usually present in the cells to convert ribitol to D-ribulose) was able to slightly speed up the conversion of xylitol to D-xylulose, for which metabolic pathways already exist. The ability of the strains to utilize xylitol was increased as much as 20 fold when first production of ribitol dehydrogenase was deregulated (the enzyme was produced all the time, not just when ribitol was present), then duplication of the ribitol dehydrogenase genes occurred, then the structure of the enzyme was changed such that its efficiency at working with xylitol was improved, and finally, in at least one case, a line regained control of the modified ribitol dehydrogenase gene so that the enzyme was only produced in the presence of xylitol. Here we have a complete example of a new metabolic pathway being developed through duplication and modification of an existing pathway.
Many papers were published concerning this group of experiments. For a review, see:
Hartley, B.S. (1984), Experimental evolution of ribitol dehydrogenase. In R.P. Mortlock (ed.), "Microorganisms as Model Systems for Studying Evolution" (pp. 23 - 54) Plenum, New York.
Bell goes on to give at least two more examples of the evolution of new metabolic pathways.
! Since such a power by definition should be more complex than nature/humans, it is not a serious option to believe it has ever existed....
