Our results indicated that male breeders tended to have higher genome-wide heterozygosity, whereas female breeders have a lower level of MHC-heterozygosity

Download (0)




To our best knowledge, this study is the first attempt to demonstrate both genetic

heterozygote advantage and genetic compatibility effects at neutral as well as functional loci in a single avian population. Our results indicated that male breeders tended to have higher genome-wide heterozygosity, whereas female breeders have a lower level of MHC-heterozygosity. We also found that females preferred to mate disassortatively at genomic level, whereas they engaged in EPF when having mated with individuals with high MHC-dissimilarity.

Heterozygosity advantage

The observation that male breeders had higher genome-wide heterozygosity as revealed by microsatellite loci (unweighted: p = 0.06; weighted: p = 0.20) is consistent with the expectation that only advantageous males could compete for territory and opportunity to mate. If males with high heterozygosity had more vigor, they should have better chances to possess breeding territories and gain breeding opportunities. The absence of

association between female heterozygosity and their breeding status in the green-backed is also consistent with a conventional perception that females don’t have to go through fierce competition to acquire breeding status. However, the link between genome-wide heterozygosity between females and their breeding status might be concealed by other factors such as high cost of autoimmunity due to high MHC heterozygosity, as discussed below.

Although accumulating evidences suggested the beneficial effect of high individual MHC heterozygosity with regard to parasite resistance (e.g. Doherty &


Zinkernagel, 1975; Paterson et al., 1998; Lohm et al., 2002), elevated level of MHC heterozygosity may also result in an inability to detect T cells that react with

self-peptide-MHC recombination and lead to autoimmunity (Nowak et al., 1992).

Selective pressure of autoimmunity may favor an optimal, rather than maximum, level of MHC-heterozygosity. Studies on both rodents and humans (reviewed by Whitacre, 2001) had demonstrated that autoimmune responses are more prevalent in females than males, especially during pregnancy, when the concentrations of estrogen and progesterone increases greatly, thereby enhancing immune response and susceptibility to autoimmune diseases. This may explain why the female breeders had fewer number of MHC alleles (unweighted: p = 0.04; weighted: p = 0.04): the hormone concentration change before onset of breeding in birds might increase females’ cost of autoimmune responses and reduce their opportunity to breed. The optimal number of MHC alleles in female breeders was further supported by the smaller variance of number of MHC alleles than female non-breedes (citations). If the cost of autoimmunity is sufficient highly, it might compromise the fitness of females, even when they carried high level of genome-wide heterozygosity.

However, either MHC-heterozygosity advantage, nor the optimal number of MHC alleles (small variance of allele number) was evidenced in the male breeders.

It is consistent with results of studies that failed to detect MHC-related mating advantage (Ekblom et al., 2004; Westerdahl, 2004; Freeman-Gallant et al., 2003), which may be attributed to several factors. First of all, female birds might not be able to utilize males’

general conditions to evaluate the degree of MHC heterozygosity, rather, females may favor a particular MHC allele conferring resistance to a specific pathogen (e.g. Ekblom et al., 2004); there may also be trade-off between health and reproduction: rising of


testosterone during breeding season would compromise immune response but strengthen males’ attractiveness (e.g. Peters et al., 2004), thereby masking the effect of MHC genotypes. Alternatively, heterozygosity advantage of MHC might be

condition-dependent. Without strong selection pressure of parasite in current population, males with high MHC heterozygosity might not be able to outperform the others to gain a breeding territory. Finally, the failure to detect MHC-based mating advantage may be partially caused by the undetected MHC polymorphism: the CE-SSCP method employed in current study only permits crude detection of MHC alleles (Bryja et al., 2005). It left MHC alleles with little sequence differences unresolved and concealed the MHC-based mating pattern.

Despite a growing number of studies corroborating the relationship between level of neutral genetic diversity and phenotypic traits (Rupert et al., 2003; Seddon et al., 2004;

Roberts et al., 2005; Amos et al., 2001; Hoffman et al., 2004), the underlying

mechanisms for such relationship remain debated. One of the possible mechanisms may be that the cause of a seeming heterozygosity-fitness correlations is in fact an artifact of inbreeding depression, in which the fitness of homozygotes is pronouncedly reduced (Hansson & Westerberg, 2002) (the general effect hypothesis). However, calculation of overall relatedness of all putative dyads among breeders showed that the possibility of inbreeding within this population was relatively low (average relatedness = -0.04). Also, only one dyad in our population was consisted of putative relatives. The low inbreeding possibility makes the general effect hypothesis an unlikely explanation for the correlation between genome-wide heterozygosity and breeding status of male green-backed tits.

Alternatively, such heterozygote advantage might be due to the local effect (reviewed by(Hansson & Westerberg, 2002), i.e. heterosis resulting from markers cosegregating


with genes that are under selection. Among functional genes, those maintained by balancing selection, such as the MHC, are most likely to confer heterozygote advantage (Paterson et al., 1998; Foerster et al., 2003). However, individual microsatellite

heterozygosity was not correlated with the number of MHC per individual (p = 0.99) in the green-backed tit. If genome-wide heterozygosity advantage is attributed to local effect, then fitness-related loci other than the MHC may be at genetic disequilibrium with the microsatellites we studied

Genetic compatibility-based mating

Although genetic compatibility hypothesis predicted that females should seek to mate with genetically dissimilar males to maximize heterozygosity of their offspring in general, ironically, only its collateral, that females increase the heterozygosity of offspring

through EPFs when they socially mated with genetically similar males, was the focus of recent studies (Blomqvist et al., 2002; Foerster et al., 2003). Genetic disassortative mating between social pairs, the dominant type of mating, had been somehow overlooked and underreported. Indeed, that average microsatellite dissimilarity of withinpairs was higher than that expected from random breeder dyads (unweighted: p = 0.07; weighted: p

= 0.06) implied that the green-backed tits might tend to mate with genetically dissimilar mates. Genetic similarity of cuckolded social pairs were slightly higher than that of the withinpairs and not significantly different from that expected from dyads among random breeders, which are also consistent with the genetic compatibility hypotheses. However, in the blue tits (Parus caeruleus), females chose genetically dissimilar males only when engaged in EPF, even males’ heterozygosity was positively correlated with the

expression of traits preferred by females (Foerster et al., 2003), whereas in black-throated


blue warbler (Dendroica caerulescens), neither heterozygosity advantage nor genetic compatibility preference was found (Smith et al., 2005). Failures to detect genetic compatibility-based mating pattern might have partly resulted from using limited number of microsatellite loci, thus not fully estimating genetic similarity between mates in most studies (Smith et al., 2005).

In our study, females were likely to engage in EPF when sharing fewer MHC alleles with their social mates (unweighted: p = 0.12; weighted: p = 0.04). Similar evidence was found in the house sparrow (Passer domesticus), where males with high MHC dissimilarity to females were excluded from reproductive events, while variation at neutral microsatellite markers had no effect, suggesting that mate choice was driven by MHC loci per se (Bonneaud et al., 2006). Nevertheless, in green-backed tits, neither did the level of MHC allele-sharing, nor the MHC allele-sharing coefficient variance, of whitinpairs differred from that expected from dyads among random breeders (for allele-sharing: unweighted: p = 0.46; weighted: p = 0.47; for variance of allele-sharing:

unweighted: p = 0.91; weighted: p = 0.83). Therefore, our results did not support that female tits nonrandomly mated with males with higher MHC dissimilarity nor shared particular level of MHC alleles with themselves.

Our results on genetic similarity of microsatellites and MHC suggested that mechanisms for genetically dissasortative mating, a self-reference mating pattern, could be more complicated than previously imagined. The fundamental assumption of

self-reference mating pattern is that genetic dissimilarity could be assessed by phenotypic traits, and MHC genotype is considered to be one of the most possible cues (Penn & Potts, 1998a; Penn & Potts, 1998b; Haberli & Aeschlimann, 2004) manifested through

olfaction. Yet the microsatellite and MHC-based dissasortative mating patterns were not


congruent in our study, and the relatedness at microsatellites was not correlated with the level MHC allele-sharing (p = 0.45), suggesting that green-backed tits might not use MHC as a cue to assess genetic similarity between themselves and candidate males for mate choice, but through other pathways and awaits further exploration.

Compared to species in which MHC-based disassortative mating was found (reviewed by Bernatchez & Landry, 2003), the green-backed tit seems to have higher number of MHC loci (at least ten loci). Compared with human (three loci) and mice (two loci), relatively high copy number of MHC genes (at least six) was also found in house sparrows (Bonneaud et al., 2006) and sticklebacks (Gasterosteus aculeatus) (Wegner et al., 2004). Interestingly, MHC-based disassortative mating was not evidenced in these organisms (Wegner et al., 2004; Bonneaud et al., 2006). Instead, females tended to mate with individuals with intermediate (optimal) level of MHC similarity. Such pattern implied that organisms with different number of MHC genes might adopt different mating strategy. For species having fewer MHC loci, selection pressure of pathogens might outweigh the cost of autoimmunity, hence a pronounced heterozygote advantage is expected and selection should favor disassortative mating based on MHC genotypes. On the other hand, for species with numerous MHC loci, chances for autoimmunity may be higher. Consequently selection may favor females to optimize, instead of maximize, offspirng’s MHC heterozygosity.

Extrapair mating

In this study I found that female green-backed tits tended to engage in EPFs when having social mates with relatively lower genome-wide dissimilarity or high MHC dissimilarity.

However, contradicting the expectations of genetic compatibility hypothesis, their EPF


partners were not more genetically compatible with respect to either genome-wide or MHC dissimilarity. One explanation might be that only four (five when weight by times) extrapair males (33.3% of EPF broods) were identified and such a small sample size lacked sufficient statistical power to reveal the association between EPFs and

genome-wide and MHC heterozygosities. Alternatively, it may be similar to that found in the blue tits (Foerster et al., 2003), in which extrapair young sired by close neighbors who were more heterozygous as well as more attractive, were not more heterozygous

compared to extrapair young of non-neighbors. It suggested that females might involve in EPFs to acquire direct benefits from their close neighbors. Resighting records during the breeding season for EPF fathers in the nearby regions led me to infer that these males were likely to be neighboring breeders. As in the case of blue tits, I suspected that female green-backed tits might accept these males as EPF partners for nongenetic benefits.

However, two thirds of EPF fathers (the unidentified ones) were presumably

non-neighbors in this study, without further genetic data, it is premature to reject the possibility that female green-backed tits pursue EPFs to maximize the heterozygosity of offspring . Nerveless, it is also possible that EPF was forced upon females.

Correlations between genetic dissimilarity and heterozygosity

Although females might seek to mate with males with higher heterozygosity or genetic dissimilarity, however, mate choice based on heterozygosity or compatibility may not be all that exclusive. In the present study, I found that individual heterozygosity and average geneome-wide similarity with individuals of opposite sex were highly negatively

correlated in the green-backed tit. It suggested that female tits might be able to find males with higher heterozygosity and higher genome-wide genetic dissimilarity at the same


time. Negative correlation between genome-wide similarity and individual

heterozygostiy revealed here might imply that as long as a females could assess a male’s heterozygosity through expression of phenotypic traits, it would not be essential for her to evaluate the genetic similarity.

A positive correlation was also found between MHC-heterozygosity and average allele-sharing in the green-back tits, as in human and peafowl (Roberts et al., 2006). It implies that a tits with high MHC diversity have higher probability to share more MHC alleles with individuals in the opposite sex. Consequently, individuals with high MHC heterozygosity might suffer the risk of autoimmune responses and are less likely to find compatible MHC genotypes because of an increased proportion of alleles shared with the average opposite-sex individual. One the other hand, those individuals with optimal MHC-heterozygosity is favored by natural selection for disease resistance and sexual selection. Similarly, by assessing MHC diversity of an individual, a female would not have to evaluating MHC allele sharing to produce offspring with optimal level of MHC diversity. These correlations might shed new lights on how different genetic aspects need could be integrated during mate choice.




相關主題 :