Biologists have long known the advantages of sexual reproduction to the evolution and survival of species. With a little sex, a fledgling creature is more likely to pass on the good mutations it may have, and more able to deal with the sort of environmental adversity that would send its asexual neighbors floundering into the shallow end of the gene pool.
The only problem is that it's hard to figure out how sex got started in the first place. Not only do many primitive single-celled organisms do just fine with asexual reproduction, but mathematical models show that a sexual mutant in an asexual population is most likely not favored to compete successfully and pass on its genes.
Now, researchers from the California Institute of Technology and the Jet Propulsion Laboratory, using "digital organisms" and RNA, have concluded that established asexual bacteria could be nudged to evolve into sexual reproduction if there are certain forms of stress on the environment, such as radiation or catastrophic meteor or comet impacts that give rise to a high rate of mutations.
In an article that has significant implications for understanding the origin of sexual reproduction in the early world, Claus Wilke of Caltech and Chris Adami, who holds joint appointments at Caltech and JPL, report that a change in conditions causing higher rates of mutations can lead an asexual population to an adaptation that may be sufficient to give mutant individuals a greater advantage if those mutants reproduce sexually.
The paper, published in the July 22 issue of the Royal Society journal Proceedings: Biological Sciences B, builds on earlier work by Adami and his collaborators, showing that digital organisms—that is, self-replicating computer programs designed to closely resemble the life cycles of living bacteria—can actually adapt to become more robust.
"What we showed in the other paper," says Adami, "is that if you transfer a fragile organism that evolved with a small mutation rate into a high-mutation-rate environment, it will adapt to this environment by becoming more robust."
One of the reasons the origin of sexual reproduction has been a mystery is because of an effect known as "mutation accumulation." Organisms tend to adapt so as to decrease the effects of mutations in order to become less vulnerable.
But this kind of robustness is poisonous, because with sexual recombination, deleterious mutations would simply accumulate in the organism and thus lead to a gradual loss of genes. This handicap of sexual creatures would be enough to guarantee their extinction when competing against asexual ones.
This can be avoided if the effects of mutations are compounding—that is, if the effect of two or more simultaneous deleterious mutations is worse than the combined effect of each of the mutations. In this manner, an organism may be robust to a few mutations, but incapable of surviving a large number of mutations, so that mutations cannot accumulate.
The new revelation by Wilke and Adami is that there is a conservation law at work in the relationship between the compounding of mutations and the fitness decay due to single mutations. This law says that robustness to a few mutations implies vulnerability to a large number, while robustness to many mutations must go hand in hand with vulnerability to single mutations.
Thus, increasing robustness to single mutations automatically makes multiple mutations intolerable, which removes organisms with multiple deleterious mutations from the population and allows sexual recombination to reap the rewards from sharing beneficial mutations.
Because stressful environments with high mutation rates push organisms to become robust to single mutations, the conservation law guarantees that this evolutionary pressure also pushes asexual organisms on to the road toward sexual recombination.
The researchers studied the evolution of digital organisms and RNA secondary structure, because accurate data on the decay of fitness and the effect of multiple mutations (whether they are compounding or mitigating) for living organisms is quite rare. For the RNA study, the researchers used known sequences with well-understood folds and then tried various mutations to see which mutations mattered and which didn't, in a system that computationally predicts RNA secondary structure. The results supported the conservation law.
Though the study did not involve actual living organisms, Adami has collaborated in the past with experts on bacteria to demonstrate that the digital organisms are indeed realistic. In an earlier 1999 study, for example, Adami's collaborator was a leading expert on the evolution of the E. coli bacteria.
The digital organisms have the advantage that many generations can be studied in a brief period of time, but Adami thinks a colony of asexual bacteria subjected to the stress imposed on the digital organisms in the experiment would probably face similar consequences.
"If you took a population of E. coli and subjected it to high mutation rates for many years—for example by irradiation or introducing mutagenic factors—at some point you might observe that exchange of genetic material, a precursor to sexual recombination, would become favorable to the organisms and thus selected for, if at the same time the environment changes fast enough that enough mutations are beneficial," he says.
"But that's a very difficult experiment with living organisms because of the time involved, and because it is difficult to construct constantly changing environments in a petri dish. This is easier with digital organisms, and will probably be first observed there.
"The reason the origin of sexual reproduction has been such a big mystery is that we look at the world as it is now," Adami says. "But the early world was a much more stressful place, sometimes changing very rapidly.
"We can't say how or when sexual reproduction came to take a hold in nature, but we can now say that high mutation rates can, under the right conditions, force an asexual organism to become sexual."
Adami earned his doctorate in theoretical physics at SUNY Stony Brook. He is a faculty associate in the computation and neural systems department at Caltech, and a research scientist at JPL. He is the author of the 1998 book Introduction to Artificial Life. Wilke, also a physicist, is a postdoctoral fellow in Adami's Digital Life Laboratory.
The article appears in Proceedings: Biological Sciences B, volume 268, number 1475, page 1469. The cover date is 22 July 2001, but the article is available on-line at http://www.pubs.royalsoc.ac.uk/proc_bio/proc_bio.html
