Jedediah Tressler
As our understanding of molecular processes has increased, a debate over what force is most important for evolution has arisen between two ideological camps. On the two sides we have the neutralists, the newer idea postulated by Motoo Kimura (1983 and etc.), balanced against the selectionist, the classic theory formulated initially by Wiliams and Dawkins (Winter 1997). This discrepancy was the focus of much work in molecular evolutionary biology for most of the 1980’s, and fueled a great deal of both research and heated discussion (Hey 1999). The debate eventually died down with little consensus being formed.
Herein I intend to examine a brief definition of the neutralist hypothesis, and an overview of some publications from both sides of the debate. I will be focusing on the overall type and nature of the supporting work from each side, omitting much of the detail from any particular experiment’s results in order to focus on the debate as a whole. Ultimately, I hope to show that it is most likely that each theory is valid when viewed in proper context, and the natural state involves some synthesis of the two.
Motoo Kimura summarizes the neutral theory saying,
The neutral theory claims that the great majority of evolutionary changes at the molecular (DNA) level are caused not by Darwinian selection but by random fixation of selectively neutral or nearly neutral mutations (1986).
It is in this statement that the majority of the debate lays. The basis of the theory is rooted in the observation of how consistent the rate of amino acid substitutions is in amino acids in homologous proteins (Kimura 1969). When Kimura looked at hemoglobin µ substitution rates in humans, mice, horses, rabbits, bovines and carp he found that they were statistically consistent across taxa. He coupled this with the observations that the majority of the genetic code is either non-coding, or redundant in some way, i.e. Genetic wobble (Kimura 1983). Because natural selection works only on phenotype, any change in a DNA region that does not result in a phenotypic change would be hidden from selective forces. Finally, the assertion is made that all mutations are either deleterious (natural selection being a negative force) or neutral. Any mutation that is deleterious will be quickly eliminated from the population by natural selection, whereas a neutral mutations fate is dependent entirely on genetic drift (Gillespie 1987). It follows then that, because so much of the genetic code is hidden from selective forces, and most mutations are neutral due to redundancy in the coding triplets, the importance of natural selection is trumped by that of drift, the only force able to act on the neutral mutations. Kimura, also defined a body of mathematical material, regarding the amount of variation expected in a population, rate of molecular evolution, and population sizes needed to account for observed heterozygosity in a given population for example (Kimura 1983, Gillespie 1987). The details of these models have been omitted here do to there complexity. Evidence supporting neutral theory focuses on the rate of change in biological molecules, and in showing that those changes are unrelated to natural selection. Two examples presented here are a more recent paper by Brookfield on the rate of amino-acid sequence evolution, and a classic support paper bye King and Jukes (2000, 1969).
Brookfield examines two systems in which amino-acid sequences are undergoing high rates of sequence evolution, sometimes considered diagnostic of adaptive change. Sequence evolution is examined by comparing the amount of amino-acid sequence evolution to the silent base substitution in the same gene in Drosophila (Brookfield 2000). After examining two different genes he show’s a high rate of evolution, but sites in both cases that there is no knows adaptive function to either gene. Also, he points out that the second gene investigated, transformer, obeys clocklike timing in base pair substitution, further supporting neutral theory. Transformer has a well-defined role in sex determination, so some adaptive value should be expected.
King and Jukes paper presents several points of support for the neutral theory (1969). They begin by accreting that though genetic drift may be slow, it is effective if fixation of a neutral allele, and that given evolutionary time it is easily possible for a gene to become fixed or lost through drift alone. Second they point out that because of the wobble in translation one-fourth (134) of all possible single base substitutions (549) would result in the same amino acid being placed in a polypeptide. Comparisons of various proteins and DNA sequences are examined, predominantly hemoglobins, the treffers mutator gene, and cytochrome c. Finally, arguments are made involving observed rates of molecular evolution, and degrees of diversity. The crux of the argument being that natural selection is insufficient to explain these observations.
Selectionism
As a base explanation, selectionism is simply Darwinian selection. The selectionist maintain that the primary force driving evolution at the molecular level is natural selection. Support of selectionism was initially reactionary, predominantly devoted to refuting claims made by neutralists. Later, selectionists began providing more support by showing selective pressures acting on supposedly neutral mutations, and even alleles.
The first example of support for selectionism is a paper written by Bryan Clarke as a reaction to the King and Jukes paper mentioned above (1970). He begins by challenging King and Jukes conclusions of randomness in the distribution of amino acid changes. The predominant fault according to Clarke is the method by which the data is fitted to a Poisson distribution. Because invariant cites in the sequences were disregarded in analysis, it creates an artificial distribution. Secondly, Clarke points out that even if the distribution fits the Poisson, this doesn’t necessarily support the neutral theory, as examples are known (land snail coloration) of traits that fit a Poison, and are known to be adaptive. Clark also discusses evidence that there is some selective force to be found in synonymous codons. Depending upon the availability of specific nucleotides in the “environment,” the importance of mRNA secondary structure and the affinity of specific tRNAs could lead to a selective situation. It is therefore inappropriate to assume a priori that synonymous codons are selectively neutral. Other assumptions of neutral mutations are also challenged on the grounds that they could be adaptive, such as the loss in primates of the ability to convert 2-keto-L-gulonolactone into ascorbic acid. Even though this means that scurvy results if the animal is deprived vitamin c, it could be adaptive as a measure to conserve some key molecule used in the metabolic pathway. Finally, Clark points out that there is a lacking in contemporary example of neutral mutations.
Two more modern examples of selectionist support can be seen in the work of Gillespie and Ayala (2000). Gillespie utilizes computer modeling to examine the neutral theory in an infinitely large population. He found that a stochastic substitution at a locus undergoing natural selection, which is linked to a neutral locus, would cause the neutral locus to mimic the effects of drift even though change is being driven by selection (Gillespie 2000).
Ayala challenges the consistent change component of neutral theory by examining the evolution of two proteins in regards to several of the mathematical models used to account for the over dispersion of the molecular clock. He found that no modification that has been proposed to Kimura’s original equation could simultaneously account for the discrepancies seen in the two proteins (Ayala 2000).
Consensus and Conclusion
The neutral theory of molecular evolution is an important, and viable model, but it can’t be applied until after natural selection has been eliminated as a possible explanation, both of the mutation in question, as well as any loci that the mutation may be linked to. Also, the possibility of a selective pressure not previously considered must always be examined while interpreting the application of a neutral model.
Few people seem to suggest that Kimura was incorrect in asserting that if a mutation is selectively neutral, or completely masked from selective force, that chance would be more important than selection in driving its evolution. Also, many selectionists would seem to agree that a large proportion of the genetic code is selectively neutral (non-coding regions in particular). The faults found in the neutral theory revolve around the application of the theory to biologically active molecules, functional proteins, or the genes that code for them. The key factor of debate is how importance is defined, and in what context is the term evolution being used. “Broadly speaking, these conflicting views … tend to focus on different, complementary aspects of the evolutionary process.” (Winter 1997) The selectionist stance is particularly sapient when applied to questions concerning the achievement of adaptation. Conversely, neutralism is most appropriate when looking at the rise of molecular diversity.
As for which theory is more important in terms of evolution, it must first be defined what is meant by evolution. In terms of the literal definition of evolution, change in gene frequencies over time, the neutral theory may well be appropriate more often. For evolution viewed as the change in traits of an individual, however, natural selection is a far more important force. As a final argument of this point, consider that, Kimura included strong selection eliminating deleterious alleles as part of the mathematical modeling of the neutral theory. If natural selection is necessary part of an evolutionary model for neutral traits, it follows that it is extremely important for non-neutral traits.
Work Cited
Ayala, Francisco J. 2000 “Neutralism and Selectionism: The Molecular
clock.” Gene-Amsterdam. 12/30; 261 (1): 27-33.
Brookfield, John, F. Y. and Paul M. Sharp. 1994. “Neutralism and selectionism face up to DNA data.” Trends in Genetics. 10 (4) 109-111
Brookfield, J.F.Y. 2000. “Evolution: What determines the rate of sequence evolution?” Current Biology. 10 (11): R410-R411.
Clarke, Bryan. 1970. “Darwinian Evolution of Proteins.” Science. 168. 3934. May 22. 1009-1011
De Winter, Willem. 1997. “The beanbag genetics controversy: Towards a synthesis of opposing views of natural selection.” Biology and Phylosophy. 12 (2). 149-184
Gillespie, J.H. 1987. “Molecular evolution and the neutral allele theory.” Oxford Surveys in Evolutionary Biology 4, 11-37
Gillespie, John H. 2000. “The neutral theory in an infinite
population.” Gene. 261. 11-18
Hey, Jody. 1999. “The neutralist, the fly and the selectionist.” Trends in Ecology and Evolution. 14 (1): 35-37
Kimura, Motoo. 1969. “The rate of Molecular Evolution Considered from the Standpoint of Population Genetics.” Proceedings of the National Academy of Sciences of the United States of America. 63, 4. Aug. 15. 1181-1188
Kimura, M. 1983. Neutral Theory of Molecular Evolution. Cambridge University Press. New York, New York.
Kimura, M. 1986. “DNA and the neutral theory.” Philosophical Transactionsof the Royal
Society of London B. 312. 343-354
King, Jack Lester, Thomas H. Jukes. 1969. “Non-Darwinian Evolution.” Science. 164, 3881. May 16. 788-798