Genetic Aspects of Sexual Selection Research Paper

1. Sexual Selection And The Sex Roles

Males and females in most animal species differ to a large extent in their morphology and their behavior. Female birds are, for example, often drab and coy compared to their colorful and sexually active mates. In many species, males compete with other males for access to females and often use specialized horns, teeth or other weapons, whereas their mates often do not fight with other females. The basic reason behind these sex differences is gamete size that often differs by several orders of magnitude between the sexes. Females produce relatively large gametes, the eggs, and accordingly can only produce relatively few of them. Males, on the other hand, produce large numbers of tiny sperm cells. Due to this difference in the number of gametes available, female reproductive success usually is limited by the number of eggs produced, and multiple matings by females have at most only a small effect on female reproductive success. Male reproductive success, however, is mainly limited by the number of eggs fertilized and thus by the number of mates obtained and not so much by sperm number (Clutton-Brock and Vincent 1991). Males therefore can benefit from attractiveness and from competition for mating partners whereas females probably benefit more by choosing the best ones among the many available mates. In accordance with this expectation, females often seem to carefully examine the available males and reject mating attempts by nonpreferred males (Andersson 1994). In peacocks, for example, hens prefer males with large colorful feathers that are distributed symmetrically, and in some other birds, females prefer males with large song repertoires. The few species where males heavily invest in offspring, e.g., some katydids with large parental investment, as well as seahorses and wading birds with exclusive paternal care, confirm the rule. In these species females usually compete for mates and males choose among females because males are limited by the number of offspring they can care for and females are limited by the number of males they can obtain to care for their offspring. These studies on sex role reversed species show that the sex difference in the benefit of additional matings after the first one is the driving force behind sexual selection. With standard sex roles, male traits that impair survival but render males especially attractive to females can evolve since male traits that are preferred by females will cause an increased mating frequency. Due to the effect of mating frequency on male reproductive success, these traits can lead to an increased lifetime mating success even when male survival decreases. The splendid feathers of male peacocks for example probably decrease male survival because this shiny ornament will also attract predators and make escape less efficient. Less adorned males may live longer. However, when they leave, on average, fewer offspring than attractive males, genes for being less adorned will go extinct. Compared to the evolution of male traits under the influence of female choice, the evolution of female preference is less easy to explain. In the next section the possible routes for the evolution of female preference are described.

2. Theoretical Models For The Evolution Of Preferences

Several hypothesis have been presented to explain the evolution of female preference that causes females to choose their mating partners among the available males (see Kirkpatrick and Ryan 1991, Maynard Smith 1991, for reviews). The simplest possibility would be that males differ in the amount of resources provided or in their fertilization rate. According to this hypothesis, choosy females increase the number of offspring produced and they thus benefit directly from their choice. In some birds, for example, those males are preferred as mates that provide superior territories and outstanding paternal care. Another possible explanation for the evolution of female preferences is that choice has an influence on the genetic quality of the offspring. According to this hypothesis females benefit indirectly when offspring of preferred males have superior viability or increased mating success compared to the offspring of nonpreferred males. And finally, female preference might have evolved in another context than sexual selection, that is, female birds might prefer males with blue feathers because it is adaptive to react positively to blue when blue berries constitute a valuable resource so that female sensory physiology became tuned to this color. This model has accordingly been termed sensory exploitation hypothesis, meaning that males exploit female sensory physiology to attain attractiveness (Ryan and KeddyHector 1992).

For all of these models, genetic variation in preference is essential since evolution of all traits rests on the existence of genetic variance. Despite the importance of such data, not very many species have been examined for genetic variance in female preference. In most of the examined species, ranging from insects to mice, within population variation in female preference seems generally to be influenced by additive genetic variance (Bakker and Pomiankowski 1995). Such genetic variance in preference means that this trait can easily evolve whenever there is a selective advantage to a specific preference. In the following paragraph the different hypotheses for the evolution of female preferences will be discussed in more detail and the genetic aspects of these models will be concentrated on.

When females benefit directly from their choice by increased lifetime fecundity, it is easy to see that any new choice will evolve as long as the extra benefit gained is larger than the extra cost paid for female choice. For example, females might prefer healthy mates because this might reduce the risk of infection during mating. Choosing males with intact and shiny feathers might thus reduce the risk of ectoparasite transfer during copulation. For models that rest on indirect benefits of female preference, females only gain from choice regarding the genetic quality of their offspring. For these indirect benefit models of female preference, the male trait also needs to have genetic variance. One of these hypotheses, Fisher’s arbitrary sexual selection model, predicts that female choice should evolve in response to the evolution of male traits. If females prefer males with a specific trait size, a linkage disequilibrium (a nonrandom association of genes) will build up, since the genes for the preference and the genes for the male trait will co-occur together more frequently than by chance. The existence of choosy females will, in addition, cause increased reproductive success of these preferred males. Since these males also carry a disproportional share of the preference allele, due to the linkage disequilibrium that is caused by the preference, the preference evolves in response to the evolution of the male trait (Kirkpatrick and Ryan 1991). If sexual selection is strong and if preference and male trait are heritable, both the male trait and the female preference are predicted to evolve by a positive feedback that eventually might lead to runaway selection. The distinctive and sufficient genetic condition for Fisher’s arbitrary sexual selection model thus is a genetic correlation between female preference and male traits. According to the good genes hypothesis, another hypothesis that suggests indirect benefits of female preferences, choosy females benefit not only by producing preferred sons but also by producing offspring with superior viability. For such a process to work, male attractiveness needs to indicate male viability. Speaking in genetic terminology, there has to be a genetic correlation between male signaling traits and viability. Since a genetic correlation between female preference and male signaling traits will also build up as a result under the good genes model, such a correlation thus cannot be taken as evidence for the arbitrary sexual selection model but it means that one of the indirect benefit models helps to maintain the preference. To provide evidence that the sensory exploitation model helps to explain the evolution of female preference, one has to show that female preference evolved before the male trait and thus independent of mate choice. The usual way is to show that female preference is more ancestral in phylogeny than the male trait. One of the best studied examples of this hypothesis is the tungara frog, Physalaemus pustulosus, where females have a preference for a male acoustic signal that is not produced by the males, but males of a closely related species do produce it. The suggested explanation for this pattern is that female choice evolved first in the ancestor of both species and the male trait evolved later in only one of those species. However, the loss of the attractive male signal during phylogeny cannot be excluded as an alternative explanation.

3. Testing Genetic Aspects Of Sexual Selection Models

The good genes hypothesis predicts that females benefit from choosing attractive males because these males will produce offspring with superior viability. The critical problem for this hypothesis is that attractive males need to have higher viability. Why should such a correlation between attractiveness and viability exist? The most prominent explanation is the following: Attractive signals are costly and only males with superior viability are able to afford these costs so that only those males can attain high attractiveness. This view of condition-dependent signaling is supported by empirical data showing that in various insects, spiders, and frogs the production of attractive courtship signals consumes more energy than the production of less attractive signals. Also, males that suffer from disease or are starving are usually not as attractive as healthy competitors. Such a process of condition-dependent signaling leads to a genetic correlation between male signaling traits and male viability when there is genetic variation for both traits. If females prefer males with signaling traits that are correlated with viability, female choice can evolve since it will become associated with high viability and a rare female choice allele can thus be expected to increase in frequency when the costs of female choice are low. If females choose among males on the basis of a male trait, a genetic correlation will build up between male trait and female preference at the same time. Since this prediction is identical to the critical and sufficient condition for the arbitrary sexual selection model, experimental separation of these two nonexclusive hypotheses is difficult.

The most frequent method to examine whether the good genes hypothesis contributes to maintain female preference is to compare the viability of the offspring of preferred males with the viability of average or nonpreferred males. In some studies using this method, female choice seems to have an astonishingly large effect on offspring viability, in other studies, no significant effect was observed (see Møller and Alatalo 1999 for a review).

Theoretical models show that these indirect sexual selection benefits are unlikely to increase offspring fitness by more than a few percent (Kirkpatrick 1996). Since direct benefits are suggested to incur larger benefits, evolution of female preferences is predicted to be dominated by direct benefits if these can be obtained by female choice. However, even a small indirect benefit to sexual preferences may be large enough for quick evolution when choice is not very costly. With an exhaustive literature review Møller and Alatalo (1999) showed that a significant positive correlation coefficient between male viability and attractiveness exists on average and they therefore proposed that females can gain a few percent regarding the viability of their offspring when they carefully choose among the potential mates.

In general, the evolution and maintenance of sexual preference seems to be due to various factors. Empirical evidence exists in favor of each of the most prominent hypotheses—direct benefits, sensory exploitation, Fisher’s arbitrary sexual selection and good gene models. These hypotheses are not mutually exclusive and some of them might work synergistically. In the future, more studies examining the quantitative genetics of sexually selected traits are necessary to evaluate the importance of the different hypotheses (Bakker 1999).

4. The Maintenance Of Genetic Variation In Male Traits

4.1 Theoretical Expectation

Both indirect sexual selection models depend on the existence of genetic variance in male traits and will in turn also cause strong selection on these traits. For theoretical reasons it has often been argued that genetic variance of traits under strong selection will usually be smaller than the genetic variance of traits that are less strongly selected (Fisher 1930). This difference occurs because under strong selection, all but the most beneficial alleles will disappear quickly from a population. In line with these arguments, life history traits that are closely connected to fitness have a lower additive genetic variance than morphological traits that are believed to be under weaker selection. If strong selection would deplete genetic variance in male attractiveness or viability, the benefit of female preference would decrease and if there are costs to female choice, females would accordingly benefit from refraining from choice and saving these costs. However, females seem to choose strongly among males in most animal species (Andersson 1994). Furthermore, female choice and the effects of sexual selection on male traits are most obvious in lekking species, where males court females at courtship territories and do not provide any resources for the females. In these cases, direct benefits are unlikely and it is not easy to see why sensory exploitation should be more frequent in these species. The indirect benefit models are thus the only ones that are likely to explain why females strongly select among the available males in lekking species and elsewhere when males do not provide resources. However, when females exert sexual selection on male viability or attractiveness, the genetic variance in male quality should decrease and the benefit of the preference will also decrease in response. The important question thus is whether male genetic variance is large enough to counterbalance the costs of being choosy.

4.2 Empirical Evidence For Genetic Variance Among Males

There is ample evidence for genetic variance in male signaling traits and in attractiveness: crosses between populations that differ in the male signaling trait show that the difference is inherited; heritability estimates from father-son comparisons show that genetic variation exists even within populations and artificial selection experiments have proven that sexually selected male traits can evolve quickly. Despite strong sexual selection on male signaling traits, sexually selected traits generally seem to have heritabilities that are as large as the values for traits that are assumed to be only under weak selection (Pomiankowski and Møller 1995). The deviation from the theoretical expectation is even more impressive when one compares the additive genetic variance (another measure for genetic variance that does not depend on the extent of environmental influence on the trait under consideration) between sexually selected traits and traits that are assumed to be not under sexual selection. Sexually selected traits have significantly larger additive genetic variance than other traits, showing that the existing genetic variation is sufficient for both indirect sexual selection models.

4.3 Possible Reasons For The Maintenance Of Genetic Variance

The extent of genetic variation present in natural populations thus seems to contradict the theoretical expectations. It is, therefore, important to understand how extensive genetic variance can be maintained in the face of strong sexual selection. Among the hypotheses put forward to explain these data that seem to contradict the theory, the following three causes for the maintenance of genetic variance shall be discussed in some detail because they have received considerable credit: capture of genetic variance in condition, hostparasite coevolution, and meiotic drive.

Male signals are only likely to honestly indicate male viability, when attractive signals are costly because otherwise males with low viability will also be able to produce these attractive signals. When signals are costly, only males in good condition will be able to produce attractive signals (Zahavi 1975). Due to this process, male signal quality will indicate a male’s condition. Since male condition is assumed to possess large genetic variance because many traits will influence condition, male signals will capture genetic variance in condition and can in this way maintain the genetic variance in male viability necessary for the good genes model (Rowe and Houle 1996).

Another hypothesis suggests that host-parasite coevolution is important in maintaining genetic variance in male signaling traits (Hamilton and Zuk 1982, Westneat and Birkhead 1998). Let us assume there are male traits that indicate that a male is free of parasites or resistant to parasites. If such resistance is heritable, females would clearly benefit from choosing those males and female preference could accordingly evolve. With increasing resistance, selection on the parasite would lead to new types of parasites that in turn will cause the selection of new resistance. This arms race can in principle lead to a persistent and large advantage to female preference for resistant males because genetic variance in males is maintained since superior male genotypes change with time. There is some indication that such a process also works in humans, where scents from potential partners with dissimilar MHC-genes (an immunologically important group of genes) are preferred ( Wedekind and Furi 1997).

Preference of females for males resistant to the action of meiotic drive has also recently been proposed as a scenario that allows persistent benefits of female choice. In stalk-eyed flies, females prefer males with longer eyestalks, and lines selected for longer eyestalks show increased resistance to meiotic drive (Wilkinson et al. 1998). It was, therefore, suggested that females benefit from choosing males with large eyestalks because resistance to meiotic drive is more frequent in these males. Theoretical simulations, however, revealed that the predicted process cannot occur, but they showed that the avoidance of males possessing meiotic drive does allow persistent benefits for female preference (Reinhold et al. 1999).

5. Genetics Of Sexually Selected Traits: X-Chromosomal Bias

Based on reciprocal crosses between two Drosophila species, Ewing (1969) long ago proposed that a disproportional part of the genes that influence traits important for mate recognition reside on the X-chromosome. Recently, two reviews revealed that X-chromosomal genes actually have a disproportionate effect on sex and reproduction related traits in humans and on sexually selected traits in animals. Using large molecular databases, Saifi and Chandra (1999) compared the linkage of mutations influencing traits related to sex and reproduction with all other traits. Their analysis shows that genes influencing traits related to sex and reproduction are several times more likely to be linked to the X-chromosome than other traits. With a different method using literature data on reciprocal crosses, Reinhold (1998) found that the influence of X-chromosomal genes is much stronger for traits that are likely to be under sexual selection than for other genes. On average, about one third of the difference between the two parental lines used for the reciprocal crosses was caused by X-chromosomal genes when sexually selected traits were considered. For those traits that were classified to be not under sexual selection, this value was much smaller—on average two percent of the difference was due to X-linked genes—and was not significantly different from zero.

Such a bias towards the X-chromosome can be expected for traits that are influenced by sexually antagonistic selection (Rice 1984) and for sex-limited traits that are under fluctuating selection (Reinhold 1999). Antagonistic selection occurs if the optimal phenotype differs for males and females and if a genetic correlation between the sexes prevents the phenotypes to reach their evolutionary stable optimum. Under sexually antagonistic selection, sex-linked traits can be expected to evolve faster than other traits because sex-linked traits almost always differ in their expression in the two sexes. Rare recessive X-chromosomal genes, for example, will always be expressed when they occur in the heterogametic sex (the sex that has two different sex chromosomes; in humans the males, they posses an X as well as a Y-chromosome) and will be expressed almost not at all in the homogametic sex. This difference in expression then provides the raw material selection can work on, so that preferential X-linkage can be expected for traits under sexually antagonistic selection. Under fluctuating selection the fate of an allele is influenced by its geometric mean fitness a fitness measure that is equivalent to the case of effective interest rates when the interest rate on a financial investment fluctuates in time. This fitness measure is influenced by the extent of expression of a trait. Between autosomal and X-chromosomal genes there is such a difference in the extent of expression: (a) X-chromosomal genes coding for sex limited male traits in heterogametic males are only expressed to one third because the other two thirds of all X-chromosomes are present in females that do not express the genes under consideration; (b) Autosomal sex-limited genes, i.e. genes that are expressed in only one sex and do not reside on the sex chromosomes but lie on any of the other chromosomes, are, in contrast, expressed to one half provided they are not totally recessive. Due to this difference in expression, autosomal genes coding for the same phenotype as X-chromosomal genes have a disadvantage compared to X-linked genes. As a consequence, X-chromosomal genes coding for sex limited traits are expected to evolve more easily than autosomal genes (Reinhold 1999). The observed X-bias for sexually selected traits can accordingly be explained by the effect of fluctuating selection on these sex-limited traits.

Bibliography:

1. Andersson M 1994 Sexual Selection. Princeton University Press, Princeton, NJ
2. Bakker T C M 1999 The study of intersexual selection using quantitative genetics. Behaviour 136: 1237–65
3. Bakker T C M, Pomiankowksi A 1995 The genetic basis of female mate preferences. Journal of Evolutionary Biology 8: 129–71
4. Clutton-Brock T H, Vincent A C J 1991 Sexual selection and the potential reproductive rates of males and females. Nature 351: 58–60
5. Ewing A W 1969 The genetic basis of sound production in Drosophila pseudobscura and D. persimilis. Animal Behaviour 17: 555–60
6. Fisher R A 1930 The Genetical Theory of Natural Selection. Clarendon Press, Oxford, UK
7. Hamilton W D, Zuk M 1982 Heritable true fitness and bright birds: A role for parasites. Science 218: 384–7
8. Kirkpatrick M 1996 Good genes and direct selection in the evolution of mating preferences. Evolution 50: 2125–40
9. Kirkpatrick M, Ryan M J 1991 The evolution of mating preferences and the paradox of the lek. Nature 350: 33–8
10. Møller A P, Alatalo R V 1999 Good-genes effects in sexual selection. Proceedings of the Royal Society of London Series B 266: 85–91
11. Pomiankowski A, Møller A P 1995 A resolution of the lek paradox. Proceedings of the Royal Society of London Series B 260: 21–9
12. Reinhold K 1998 Sex linkage among genes controlling sexually selected traits. Behavioural Ecology and Sociobiology 44: 1–7
13. Reinhold K 1999 Evolutionary genetics of sex-limited traits under fluctuating selection. Journal of Evolutionary Biology 12: 897–902
14. Reinhold K, Engqvist L, Misof B, Kurtz J 1999 Meiotic drive and evolution of female choice. Proceedings of the Royal Society of London Series B 266: 1341–5
15. Rice W R 1984 Sex chromosomes and the evolution of sexual dimorphism. Evolution 38: 735–42
16. Rowe L, Houle D 1996 The lek paradox and the capture of genetic variance by condition dependent traits. Proceedings of the Royal Society of London Series B 263: 1415–21
17. Ryan M J, Keddy-Hector A 1992 Directional patterns of female mate choice and the role of sensory biases. The American Naturalist 139: S4–S35
18. Saifi G M, Chandra H S 1999 An apparent excess of sex-and reproduction-related genes on the human X chromosome. Proceedings of the Royal Society of London Series B 266: 203–9
19. Smith J M 1991 Theories of sexual selection. Trends in Ecology and Evolution 6: 146–51
20. Wedekind C, Furi S 1997 Body odour preferences in men and women: Do they aim for specific MHC combinations or simply heterozygosity? Proceedings of the Royal Society of London Series B 264: 1471–9
21. Westneat D F, Birkhead T R 1998 Alternative hypotheses linking the immune system and mate choice for good genes. Proceedings of the Royal Society of London B 265: 1065–73
22. Wilkinson G S, Presgraves D C, Crymes L 1998 Male eye span in stalk eyed flies indicate genetic quality by meiotic drive suppression. Nature 391: 276–79
23. Zahavi A 1975 Mate selection—a selection for a handicap. Journal of Theoretical Biology 53: 205–14