Introduction
Since sexual selection in both males and females is influenced by the number of mating partners, extra-pair paternities (EPP) play an important role in the evolution of mating systems 1,2. EPP are common in pair-living, or socially monogamous birds and mammals (see Table 1 for definitions used in this study), including humans, while genetic monogamy is a very rare phenomenon 1,3,4. Among pair-living mammals — which constitute up to 9% of mammal species, depending on the classification method 5,6 — strict genetic monogamy (no cases of EPP) has been reported for only seven species so far (Table 2). Four other species can be considered as “mostly” genetically monogamous, with the rate of EPP <10%. However, for most pair-living mammal species, genetic paternity data simply does not exist yet, and therefore our understanding of the frequency of genetic monogamy is very incomplete.
Rates of EPP vary substantially between species and populations and have been shown to be affected by various factors, such as, for example, intensity of male care, pair-bond strength and population density 3,4,7,8. The intriguing question is why some individuals engage in mating with multiple partners while others do not. The advantages to males of engaging in extra-pair copulations (EPC) are well recognized, as males are expected to increase their fitness by increasing the number of mating partners as the result of their higher potential reproductive rate 2,9. However, in pair-living species with biparental care, potential reproductive rates and, consequently, levels of intra-sexual competition will be more similar for males and females 2. As a result, both sexes might be expected to gain benefits from engaging in EPC 10.
One potential advantage of EPC to females could to be related to limitations in mate choice. In pair-living species with biparental care, especially in those with low mobility and low breeding density, mate choice can be highly constrained. Not only do mates become unavailable once paired, but also individuals may face a conflict between choice for direct benefits (territory, paternal care) and indirect genetic benefits. As a result, individuals may end up paired to a genetically incompatible or to a closely related partner. To escape these constraints and minimize inbreeding, animals might seek EPC that would allow them to gain indirect benefits while still taking advantage of direct benefits provided by the social partner 10. This strategy has been demonstrated in various bird species 7,11. In mammals, the evidence is much more limited. In Alpine marmots, Marmota marmota, and meerkats, Suricata suricatta EPP rates were found to be higher in pairs where partners were more closely related 12,13. But, to our knowledge, the only pair-living mammal species for which this effect has been demonstrated is the fat-tailed dwarf lemur, Cheirogaleus medius, where females sharing more major histocompatibility complex (MHC)-supertypes with their social partner engaged in more EPC 14.
Whether an individual chooses to restrict matings to its social partner or to seek EPC, it is expected to do so to maximize not only its direct fitness benefits, but also the indirect (genetic) benefits, expressed as increased genetic quality of offspring. The closely related hypotheses of genetic compatibility and heterozygosity posit that individuals benefit from choosing a mate that will maximize offspring heterozygosity 15–17. Thus, animals are expected to choose mates that are genetically unrelated or dissimilar at some fitness-related genes (e.g., MHC genes). An increase in offspring heterozygosity resulting from this disassortative mating is expected increase offspring fitness, as indicated by links between individual heterozygosity and various fitness proxies, such as survival, reproductive success and parasite resistance (e.g., Coltman et al., 1999; Foerster et al., 2003; Ortego et al., 2007) reviewed in Kempenaers (2007). In addition, irrespective of genetic compatibility, individuals might also benefit from choosing heterozygous mates, because heterozygous partners are expected to have higher fitness and should be more likely to provide direct benefits such as increased parental care, fertility or good quality territory 15,20.
Mate choice based on heterozygosity was demonstrated in various species of birds and mammals. For example, in blue tits, Cyanistes caeruleus, heterozygosity was positively correlated between social mates, indicating that mating preferences were based on partner’s heterozygosity 21. In Antarctic fur seals, Arctocephalus gazella, where females exert choice by moving across a breeding colony to visit largely stationary males, females were shown to move further to optimize between high heterozygosity and low relatedness 20. In many species, mate choice was shown to be based on MHC loci dissimilarity (e.g., fat-tailed dwarf lemur, Cheirogaleus medius: 14 or, conversely, similarity (probably an adaptation to local pathogens, shown in, e.g., Malagasy giant jumping rat, Hypogeomys antimena, and European badgers, Meles meles 22,23. Finally, relatedness-based mate choice, while demonstrated in some species, such as Antarctic fur seals, was not found in many other studied species, such as fat-tailed dwarf lemurs, blue titis and great reed warblers, Acrocephalus arundinaceus (García-Navas et al., 2009; Hansson et al., 2007; Schwensow et al., 2008; Sommer, 2005).
One of the biggest problems arising from the constraints of mate choice in pair-living animals is the risk of inbreeding. In the absence of other options, or as a result of a trade-off between choosing a good territory and unrelated/compatible partner, individuals might pair with too closely related mates. This problem can be solved “actively” by either avoiding matings with closely related individuals (through kin recognition) or engaging in EPC with less related individuals, as discussed above 10,25. Alternatively, “passive” inbreeding avoidance can be ensured by natal dispersal that disrupts opposite-sex kin associations and thus allows to avoid matings between them 26. Dispersal was shown to be sufficient to avoid inbreeding or reach a certain level of genetic dissimilarity in many situations 21,24,27. However, it remains unclear if dispersal has to be sex-biased to generate enough local genetic dissimilarity between breeding females and males to avoid inbreeding. In most mammals, males are the dispersing sex, because in polygynous mating systems, which are prevailing in mammals, males experience stronger intra-sexual competition for mates than females 26,28. Following the same logic, mammals that mate monogamously or cooperatively with high levels of reproductive monopolization by a dominant pair are expected to have little or no sex bias in dispersal. This was found to be true in some mammals, such as the genetically monogamous Azara’s owl monkey, Aotus azarae, where both sexes disperse, or cooperatively breeding meerkats, where dispersal is only slightly male-biased 29,30. However, in other mammals, e.g., genetically monogamous California mice, Peromyscus californicus, or socially monogamous greater white-toothed shrew, Crocidura russula, dispersal was found to be female-biased 31,32.
Here, we present a comprehensive study of the genetic mating system, mate choice and dispersal in a wild population of coppery titi monkeys, Plecturocebus cupreus. Titi monkeys (genera Callicebus, Plecturocebus, and Cheracebus) exhibit almost all the elements of the “monogamy package”, such as pair living, strong long-term pair bonds, an exceptionally high level of male care (the infant is carried almost exclusively by the social father), territoriality, and sexual monomorphism 33–36. The only missing component which has yet to be characterized is the genetic mating system. Titis are one of the very few mammalian taxa that exhibit both high level of male care and strong pair bonds, two characteristics shown to affect the rates of EPP in mammals 3. The examination of their mating system and the proximate mechanisms of its maintenance may, therefore, shed light on the evolution of social and genetic monogamy in mammals. In this study, we first examined the mating system of coppery titis using a set of 27 newly developed microsatellite loci that can be universally applied to New World monkeys. Second, we tested for evidence of relatedness- and/or heterozygosity-based mate choice. Finally, to see if dispersal is sex-biased, we compared genetic relatedness and diversity patterns in adult females and males and performed spatial genetic analysis. Given consistent pair living, strong pair bonds and high levels of male care in coppery titis, we predicted them to be genetically monogamous or have a very low rate of EPP. Since the risk of inbreeding is expected to be especially high for long-lived pair-living species such as titis, we expected to find evidence for active inbreeding avoidance via mate choice and/or for heterozygosity-based mate choice. We predicted both sexes to disperse, as expected from a pair-living territorial mammal with biparental care.