The heat-shock protein Hsp90 supports diverse but specific signal transducers and lies at the interface of several developmental pathways. We report here that when Drosophila Hsp90 is mutant or pharmacologically impaired, phenotypic variation affecting nearly any adult structure is produced, with specific variants depending on the genetic background and occurring both in laboratory strains and in wild populations. Multiple, previously silent, genetic determinants produced these variants and, when enriched by selection, they rapidly became independent of the Hsp90 mutation. Therefore, widespread variation affecting morphogenic pathways exists in nature, but is usually silent; Hsp90 buffers this variation, allowing it to accumulate under neutral conditions. When Hsp90 buffering is compromised, for example by temperature, cryptic variants are expressed and selection can lead to the continued expression of these traits, even when Hsp90 function is restored. This provides a plausible mechanism for promoting evolutionary change in otherwise entrenched developmental processes.

Hsp90 as a capacitor for morphological evolution

In the late sixties the neo-Darwinian theory of evolution, hitherto generally accepted by biologists, was confronted with a new, revolutionary view: the theory of neutral or non-Darwinian evolution. In the neutralist view amino acid and nucleotide changes in the course of evolution are mainly due to random fixation of selectively neutral mutants (Kimura, 1968; King and Jukes, 1969). The approximately constant rate of evolution in terms of amino acid substitutions per site per year for various lineages, as claimed by neutralists, forms one of the main arguments for the neutralist theory. Enzyme polymorphisms in present-day populations are considered as a phase in molecular evolution (Kimura and Ohta, 1971; Kimura, 1977). The first estimates of the extent of enzyme polymorphisms in populations of Drosophila pseudoobscura (Lewontin and Hubby, 1966) and humans (Harris, 1966) based on electrophoresis of proteins were soon followed by many others, covering a wide range of animal and plant species. These surveys showed that most species are highly polymorphic [see reviews by Powell (1975), Nevo (1978), A. H. D. Brown (1979), and Hamrick et al. (1979)]. In the neutralist versus selectionist controversy the nature of these allozyme polymorphisms is disputed. Thw neutralist hypothesis states that the observed variation is mainly a product of mutation and drift of selectively neutral genes. The selectionist hypothesis claims that some from of balancing selection is resposible for the maintenance of allozyme polymorphisms.

The alcohol-dehydrogenase polymorphism in drosophila-melanogaster - selection at an enzyme locus

The advantages of a broad ecological niche are evident (Futuyma nd Moreno 1988; Jaenike 1990). The costs of being a generalist are much less obvious. Yet some costs must exist or there would be no specialist species. Theories of the evolution of ecological specialization routinely assume that adaptations to different habitats (i.e., to different conditions including food and other resources) are antagonistic (Levins 1968; Taper and Case 1985; Lively 1986; Lynch and Gabriel 1987; Futuyma nd Moreno 1988; Van Tienderen 1991). Hence, trade-offs between performance in different habitats should be ubiquitous. Such trade-offs should be reflected in negative cross-habitat genetic orrelations: genotypes doing better in one habitat would be inferior in other habitats (Via 1990). However, in experimental studies, mostly of herbivorous insects, more positive than negative correlations have been found, and in most cases they have been close to zero (Futuyma nd Moreno 1988; Jaenike 1990; Via 1990). For several reasons, the predicted correlations may not have been found although the tradeoffs did exist (Rausher 1988; Charlesworth 1990). Nonetheless, the high number of correlations close to zero suggests that fitness components expressed in different habitats are to some degree genetically independent. This independence can result from allele substitutions at universally expressed loci that have fitness consequences only under some conditions or from allelic variation at loci expressed only in some habitats (Via and Lande 1985). Examples of the latter include mutations at loci involved in utilization of resources available only in some habitats or in processing secondary compounds pecific to a certain host plant of the range used by a herbivorous insect. It has been suggested (Rausher 1988) that reduction of the overlap between genes affecting performance indifferent habitats could be favored by selection in a generalist species. In this note I suggest hat, although not involved in the trade-offs, deleterious alleles affecting performance only in some habitats can produce another, complementary cost of being a generalist. Such alleles are exposed to natural selection only in a part of a generalist species' niche, and thus only in some individuals, but the mutation rate to such alleles should be independent of the habitat. Hence, a generalist species is more prone to accumulate them than a specialist species confined to that habitat. This has been proposed as a mechanism maintaining

Accumulation of Deleterious Mutations and the Evolutionary Cost of Being a Generalist

We have used gluconate-limited chemostats to study selective differences between isogenic strains of Escherichia coli K12 into which four naturally occurring alleles coding for allozymes of 6-phosphogluconate dehydrogenase (6PGD) had been transferred. The limit of detectability of selection with our procedures is a selection coefficient of 0.5%. In the normal E. coli K12 genetic background, all alleles are selectively neutral or nearly neutral. The absence of detectable selection does, however, depend on genetic background and on such environmental factors as cell density. In a genetic background containing a mutation that cuts off the alternative metabolic route for 6-phosphogluconate, selection between allozymes can be detected, and the selection is in the direction expected from the measured apparent K m values of the allozymes. Even when the alternative metabolic route is not blocked by mutation, one of the 6PGD allozymes has a detrimental, but density-dependent, interaction with a mutation conferring resistance to bacteriophage T5. In all cases, the observed selection is due to the allozymes themselves (or to associated regulatory elements), as the selection disappears when the chemostats are limited by a different carbon source (ribose plus succinate). Nevertheless, the four alleles do seem to be selectively neutral or nearly neutral in the normal E. coli K12 genetic background. Moreover, the distribution of allele frequencies in natural populations of E. coli is in accord with the expectations of selective neutrality. I am inclined to suspect that we see, at least in some [cases], variations which are of no service to the species, and which consequently have not been seized on and rendered definite by natural selection…. Variations neither useful nor injurious would not be affected by natural selection, and would be left either a fluctuating element, as perhaps we see in certain polymorphic species, or would ultimately become fixed…. We may easily err in attributing importance to characters, and in believing that they have been developed through natural selection;… many structures are now of no direct use to their possessors, and may never have been of any use to their progenitors…. [On the other hand,] we are much too ignorant in regard to the whole economy of any organic being to say what slight modifications would be of importance or not.

https://www.genetics.org/content/genetics/96/4/801.full.pdf

Selective Neutrality of 6pgd Allozymes in E. COLI and the Effects of Genetic Background

Theoretical studies have suggested that the evolution of habitat (host) races, regarded as a prelude to sympatric speciation, requires strong trade-offs in adaptation to different habitats: alleles that improve fitness in some habitats and have deleterious effects of similar magnitude in other habitats must be segregating in the population. I argue that such trade-offs are not necessary; the evolution of habitat races can also be driven by genetic variation due to loci that affect fitness in one habitat and are neutral or nearly so in others, that is, when performance in different habitats is genetically independent. One source of such genetic variation are deleterious mutations wit:? habitat-specific fitness effects. I use deterministic two-locus and multilocus models to show that the presence of such mutations in the gene pool results in indirect selection favoring habitat fidelity or habitat preference over acceptance of both suitable habitats. This leads to the evolution of largely genetically isolated populations that use different habitats, from a single panmictic population of individuals accepting both habitats. This study suggests that the conditions favoring habitat race formation, and thus possibly sympatric speciation, are much less stringent than previously thought.

https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1558-5646.1997.tb05099.x

Sympatric speciation via habitat specialization driven by deleterious mutations.

Strains homozygous at the amylase locus were derived from a polymorphic laboratory population of Drosophila melanogaster . The Amy 4,6 strain has higher enzyme activity than the Amy 1 strain.—Maltose has the same nutritional value as starch.—The effect of starch in pure culture depends on the yeast level. At low yeast level increasing starch increases survival, at high yeast level increasing starch increases mean dry weight. The strains do not differ in survival or mean dry weight in pure culture.—In mixed cultures at 50% input of Amy 4,6 and Amy 1 as larvae the percentage Amy 4,6 in adults increases with increasing starch at low yeast levels, but equals input frequency at high yeast levels. No increase in percentage Amy 4,6 in adults is present with increasing maltose at low yeast levels in mixed culture. The increase in percentage Amy 4,6 with increasing starch must be due to selection on the amylase locus working by competition for food in the larval stage. The single locus selection coefficient is determined by the environment and can reach quite high values.—Viability selection in the presence of starch is in the direction indicated by the enzyme activities.

https://www.genetics.org/content/genetics/84/1/77.full.pdf

W. Scharloo

Papers

Environmental determination of selective significance or neutrality of amylase variants in drosophila melanogaster