The decline in fitness with inbreeding: evidence for negative dominance-by-dominance epistasis in Drosophila melanogaster.

Genetic interactions can play an important role in the evolution of reproductive strategies. In particular, negative dominance-by-dominance epistasis for fitness can theoretically favour sex and recombination. This form of epistasis can be detected statistically because it generates nonlinearity in the relationship between fitness and inbreeding coefficient. Measures of fitness in progressively inbred lines tend to show limited evidence for epistasis. However, tests of this kind can be biased against detecting an accelerating decline due to line losses at higher inbreeding levels. We tested for dominance-by-dominance epistasis in Drosophila melanogaster by examining viability at five inbreeding levels that were generated simultaneously, avoiding the bias against detecting nonlinearity that has affected previous studies. We find an accelerating rate of fitness decline with inbreeding, indicating that dominance-by-dominance epistasis is negative on average, which should favour sex and recombination.

Selfing, adaptation and background selection in finite populations.

Classic deterministic genetic models of the evolution of selfing predict species should be either completely outcrossing or completely selfing. However, even species considered high selfers outcross to a small degree (e.g. Arabidopsis thaliana and Caenorhabditis elegans). This discrepancy between theory and data may exist because the classic models ignore the effects of drift interacting with selection, that is, Hill-Robertson effects. High selfing rates make the effective rate of recombination near zero, which is expected to cause the build-up of negative disequilibria in finite populations. Despite the transmission advantage associated with complete selfing, low levels of outcrossing may be favoured because of the benefits of increasing the effective rate of recombination to dissipate negative disequilibria. Using multilocus simulations, we confirm that selfing reduces effective population size through background selection and causes negative disequilibria between selected sites. Consequently, the rate of adaptation is substantially reduced in strong selfers. When selfing rate is allowed to evolve, populations evolve to be either strong outcrossers or strong selfers, depending on the parameter values. Amongst selfers, low, but nonzero, levels of outcrossing can be maintained by selection even when all mutations are deleterious; more outcrossing is maintained with higher rates of deleterious mutation. The addition of beneficial mutations can (i) lead to a quantitative increase in the degree of outcrossing amongst stronger selfers but (ii) may cause outcrossing species to evolve into stronger selfers.

Higher rates of sex evolve under K-selection.

The geographical distribution of sexual and related asexual species has been suggested to correlate with habitat stability; sexual species tend to be in stable habitats (K-selection), whereas related asexual taxa tend to be in unstable habitats (r-selection). We test whether this broad-scale pattern can be re-created at a microevolutionary scale by experimentally evolving populations of facultatively sexual rotifers under different ecological conditions. Consistent with the pattern in nature, we find that the rate of sex evolves to lower levels in the r-selected than in K-selection environments. We consider several different explanations for these results.

The effects of competition on the strength and softness of selection.

Deleterious alleles are constantly introduced into populations due to mutation. In subdivided populations, the impact of these mutations depends on the strength of selection as well as the softness of selection, that is, the extent to which fitness is governed by local rather than global competition. It is widely appreciated that the intensity and type of competition will affect selection on deleterious mutations but most empirical work has focused solely on the effects of competition on selection strength. However, competition has rarely been studied in the context of selection 'softness' even though competition is at the conceptual root of soft selection. All other things being equal, theory predicts that inter- and intraspecific competitions have opposing effects on the softness of selection. Using Drosophila melanogaster, we estimated the strength and softness of selection in a 'baseline' competitive environment as well as two additional competitive environments characterized by either additional intra- or interspecific competitors. We found that competitive environment had little effect on the average strength of selection. While the softness of selection was affected by the type of competition, the direction of change varied across tests of different genes, contrary to expectation. Although the 'hard/soft' selection paradigm implicitly assumes that all individuals are equally sensitive to the local competitive environment, we found this not to be the case. Wild-type individuals were more sensitive to changes in the genetic quality of their local competitors than mutant individuals.

The effect of sex on the mean and variance of fitness in facultatively sexual rotifers.

The evolution of sex is a classic problem in evolutionary biology. While this topic has been the focus of much theoretical work, there is a serious dearth of empirical data. A simple yet fundamental question is how sex affects the mean and variance in fitness. Despite its importance to the theory, this type of data is available for only a handful of taxa. Here, we report two experiments in which we measure the effect of sex on the mean and variance in fitness in the monogonont rotifer, Brachionus calyciflorus. Compared to asexually derived offspring, we find that sexual offspring have lower mean fitness and less genetic variance in fitness. These results indicate that, at least in the laboratory, there are both short- and long-term disadvantages associated with sexual reproduction. We briefly review the other available data and highlight the need for future work.

Male Drosophila melanogaster have higher mating success when adapted to their thermal environment.

Adaptation to new environments is a well-documented phenomenon. Individuals from populations maintained in a particular environment for multiple generations tend to be better able to survive and/or reproduce in that environment than their ancestors or other individuals adapted to alternative environments. A third major component of fitness, mating success, has not been well studied in replicated populations under selection in divergent environments. In this study, we used mating trials to compare the mating success of male Drosophila melanogaster adapted for 10 years to two different temperatures, 18 and 25 degrees C. In competition for female partners, males had significantly higher mating success at their adapted temperature compared with males adapted to a different temperature. These results are consistent with the notion that those mutations favoured by natural selection also tend to be favoured by sexual selection.

Parasites and the evolution of self-fertilization.

Assuming all else is equal, an allele for selfing should spread when rare in an outcrossing population and rapidly reach fixation. Such an allele will not spread, however, if self-fertilization results in inbreeding depression so severe that the fitness of selfed offspring is less that half that of outcrossed offspring. Here we consider an ecological force that may also counter the spread of a selfing allele: coevolution with parasites. Computer simulations were conducted for four different genetic models governing the details of infection. Within each of these models, we varied both the level of selfing in the parasite and the level of male-gamete discounting in the host (i.e., the reduction in outcrossing fitness through male function due to the selfing allele). We then sought the equilibrium level of host selfing under the different conditions. The results show that, over a wide range of conditions, parasites can select for host reproductive strategies in which both selfed and outcrossed progeny are produced (mixed mating). In addition, mixed mating, where it exits, tends to be biased toward selfing.

Parent-offspring coadaptation and the dual genetic control of maternal care.

In many animal species, the amount of care provided by parents is determined through a complex interaction of offspring signals and responses by parents to those signals. As predicted by honest signaling theory, we show that in the burrower bug, Sehirus cinctus, maternal provisioning responds to experimental manipulations of offspring condition. Despite this predicted environmental influence, we find evidence from two cross-foster experiments that variation in maternal care also stems from two distinct genetic sources: variation among offspring in their ability to elicit care and variation among parents in their response to offspring signals. Furthermore, as predicted by maternal-offspring coadaptation theory, offspring signaling is negatively genetically correlated with maternal provisioning.

Sexual selection and the maintenance of sexual reproduction.

The maintenance of sexual reproduction is a problem in evolutionary theory because, all else being equal, asexual populations have a twofold fitness advantage over their sexual counterparts and should rapidly outnumber a sexual population because every individual has the potential to reproduce. The twofold cost of sex exists because of anisogamy or gamete dimorphism-egg-producing females make a larger contribution to the zygote compared with the small contribution made by the sperm of males, but both males and females contribute 50% of the genes. Anisogamy also generates the conditions for sexual selection, a powerful evolutionary force that does not exist in asexual populations. The continued prevalence of sexual reproduction indicates that the 'all else being equal' assumption is incorrect. Here I show that sexual selection can mitigate or even eliminate the cost of sex. If sexual selection causes deleterious mutations to be more deleterious in males than females, then deleterious mutations are maintained at lower equilibrium frequency in sexual populations relative to asexual populations. The fitness of sexual females is higher than asexuals because there is no difference in the fecundity of sexual females and asexuals of the same genotype, but the equilibrium frequency of deleterious mutations is lower in sexual populations. The results are not altered by synergistic epistasis in males.

Recessive mutations and the maintenance of sex in structured populations.

The evolutionary maintenance of sexual reproduction remains a controversial problem. It was recently shown that recessive deleterious mutations create differences in the mutation load of sexual vs. asexual populations. Here we show that low levels of population structure or inbreeding can greatly enhance the importance of recessive deleterious mutations in the context of sexual vs. asexual populations. With population structure, the cost of sex can be substantially reduced or even eliminated for realistic levels of dominance.

Maternal effects and the evolution of aposematic signals.

Aposematic signals that warn predators of the noxious qualities of prey gain their greatest selective advantage when predators have already experienced similar signals. Existing theory explains how such signals can spread through selective advantage after they are present at some critical frequency, but is unclear about how warning signals can be selectively advantageous when the trait is initially rare (i.e., when it first arises through mutation) and predators are naive. When aposematism is controlled by a maternal effect gene, the difficulty of initial rarity may be overcome. Unlike a zygotically expressed gene, a maternally expressed aposematism gene will be hidden from selection because it is not phenotypically expressed in the first individual with the mutation. Furthermore, the first individual carrying the new mutation will produce an entire family of aposematic offspring, thereby providing an immediate fitness advantage to this gene.

Kin recognition and the evolution of altruism.

In 1964, Hamilton formalized the idea of kin selection to explain the evolution of altruistic behaviours. Since then, numerous examples from a diverse array of taxa have shown that seemingly altruistic actions towards close relatives are a common phenomenon. Although many species use kin recognition to direct altruistic behaviours preferentially towards relatives, this important aspect of social biology is less well understood theoretically. I extend Hamilton's classic work by defining the conditions for the evolution of kin-directed altruism when recognizers are permitted to make acceptance (type I) and rejection (type II) errors in the identification of social partners with respect to kinship. The effect of errors in recognition on the evolution of kin-directed altruism depends on whether the population initially consists of unconditional altruists or non-altruists (i.e. alternative forms of non-recognizers). Factors affecting the level of these error rates themselves, their evolution and their long-term stability are discussed.

Developmental interactions and the constituents of quantitative variation.

Development is the process by which genotypes are transformed into phenotypes. Consequently, development determines the relationship between allelic and phenotypic variation in a population and, therefore, the patterns of quantitative genetic variation and covariation of traits. Understanding the developmental basis of quantitative traits may lead to insights into the origin and evolution of quantitative genetic variation, the evolutionary fate of populations, and, more generally, the relationship between development and evolution. Herein, we assume a hierarchical, modular structure of trait development and consider how epigenetic interactions among modules during ontogeny affect patterns of phenotypic and genetic variation. We explore two developmental models, one in which the epigenetic interactions between modules result in additive effects on character expression and a second model in which these epigenetic interactions produce nonadditive effects. Using a phenotype landscape approach, we show how changes in the developmental processes underlying phenotypic expression can alter the magnitude and pattern of quantitative genetic variation. Additive epigenetic effects influence genetic variances and covariances, but allow trait means to evolve independently of the genetic variances and covariances, so that phenotypic evolution can proceed without changing the genetic covariance structure that determines future evolutionary response. Nonadditive epigenetic effects, however, can lead to evolution of genetic variances and covariances as the mean phenotype evolves. Our model suggests that an understanding of multivariate evolution can be considerably enriched by knowledge of the mechanistic basis of character development.

On indirect genetic effects in structured populations.

Indirect genetic effects (IGEs) occur when the phenotype of an individual, and possibly its fitness, depends, at least in part, on the genes of its social partners. The effective result is that environmental sources of phenotypic variance can themselves evolve. Simple models have shown that IGEs can alter the rate and direction of evolution for traits involved in interactions. Here we expand the applicability of the theory of IGEs to evolution in metapopulations by including nonlinear interactions between individuals and population genetic structure. Although population subdivision alone generates some dramatic and nonintuitive evolutionary dynamics for interacting phenotypes, the combination of nonlinear interactions with subdivision reveals an even greater importance of IGEs. The presence of genetic structure links the evolution of interacting phenotypes and the traits that influence their expression ("effector traits") even in the absence of genetic correlations. When nonlinear social effects occur in subdivided populations, evolutionary response is altered and can even oppose the direction expected due to direct selection. Because population genetic structure allows for multilevel selection, we also investigate the role of IGEs in determining the response to individual and group selection. We find that nonlinear social effects can cause interference between levels of selection even when they act in the same direction. In some cases, interference can be so extreme that the actual evolutionary response to multilevel selection is opposite in direction to that predicted by summing selection at each level. This theoretical result confirms empirical data that show higher levels of selection cannot be ignored even when selection acts in the same direction at all levels.

Possible consequences of genes of major effect: transient changes in the G-matrix.

Understanding the process of evolutionary divergence requires knowledge of the strength, form, and targets of selection, as well as the genetic architecture of the divergent traits. Quantitative genetic approaches to understanding multivariate selection and genetic response to selection have proven to be powerful tools in this endeavor, particularly with respect to short-term evolution. However, the application of quantitative genetic theory over periods of substantial phenotypic change is controversial because it requires that the requisite genetic parameters remain constant over the period of time in question. We show herein how attempts to determine the stability of key genetic parameters may be misled by the 'many genes of small effect' type of genetic architecture generally assumed in quantitative genetics. The presence of genes of major effect (GOMEs) can alter the genetic variance-covariance matrix dramatically for brief periods of time, significantly alter the rate and trajectory of multivariate evolution, and thereby mislead attempts to reconstruct or predict long term evolution.