Inbreeding depression is a major evolutionary and ecological force influencing population dynamics and the evolution of inbreeding-avoidance traits such as mating systems and dispersal. Mating systems and dispersal are fundamental determinants of population genetic structure. Resolving the relationships among genetic structure, seasonal breeding-related mating systems and dispersal will facilitate our understanding of the evolution of inbreeding avoidance. The goals of this study were as follows: (i) to determine whether females actively avoided mating with relatives in a group-living rodent species, Brandt’s voles (Lasiopodomys brandtii), by combined analysis of their mating system, dispersal and genetic structure; and (ii) to analyze the relationships among the variation in fine-genetic structure, inbreeding avoidance, season-dependent mating strategies and individual dispersal. Using both individual- and population-level analyses, we found that the majority of Brandt’s vole groups consisted of close relatives. However, both group-specific FISs, an inbreeding coefficient that expresses the expected percentage rate of homozygosity arising from a given breeding system, and relatedness of mates showed no sign of inbreeding. Using group pedigrees and paternity analysis, we show that the mating system of Brandt’s voles consists of a type of polygyny for males and extra-group polyandry for females, which may decrease inbreeding by increasing the frequency of mating among distantly-related individuals. The consistent variation in within-group relatedness, among-group relatedness and fine-scale genetic structures was mostly due to dispersal, which primarily occurred during the breeding season. Biologically relevant variation in the fine-scale genetic structure suggests that dispersal during the mating season may be a strategy to avoid inbreeding and drive the polygynous and extra-group polyandrous mating system of this species.
Citation: Liu XH, Yue LF, Wang DW, Li N, Cong L (2013) Inbreeding Avoidance Drives Consistent Variation of Fine-Scale Genetic Structure Caused by Dispersal in the Seasonal Mating System of Brandt’s Voles. PLoS ONE 8(3): e58101. https://doi.org/10.1371/journal.pone.0058101
Editor: Trine Bilde, Aarhus University, Denmark
Received: April 23, 2012; Accepted: February 3, 2013; Published: March 14, 2013
Copyright: © 2013 Liu et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This research was funded by the National Basic Research Program of China (project no. 2007CB109104). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Inbreeding depression is a major evolutionary and ecological force influencing population dynamics and the evolution of inbreeding-avoidance traits such as mating systems and dispersal . Population genetic structure is fundamentally determined by mating systems and dispersal, which have long been recognized as be the primary factors influencing the rate and outcome of evolution . Resolving the relationship among genetic structure, mating systems and dispersal will facilitate our understanding of the evolution of inbreeding avoidance.
Inbreeding depression is usually substantial enough to affect both individual and population performance ,  by increasing the chances of offspring being affected by recessive or deleterious traits . Pusey and Wolf (1996) proposed that inbreeding depression is often sufficiently severe to lead to the evolution of inbreeding avoidance mechanisms. In many species, individuals optimize genetic compatibility and avoid costly inbreeding depression by choosing mates based on their genetic relatedness –. The almost ubiquitous phenomenon of females mating with more than one male, polyandry, is proposed as a mechanism to avoid reproducing with genetically incompatible mates – and a means of inbreeding avoidance . The idea that females engage in polyandry for genetic benefits is supported by a great deal of increasingly rigorous empirical evidence –, –.
The evolutionary causes of dispersal have been the focus of much theoretical work. Kin competition, inbreeding, resource competition and environmental stochasticity have been identified as potential driving forces, but inbreeding avoidance is recognized as one of the primary causes of dispersal . The observations that dispersal occurs more often in one sex or the other  and that the relatedness between mates decreases with dispersal distance suggests that dispersal might be an efficient means to avoid inbreeding . Other means of avoiding inbreeding, such as discrimination against kin when choosing a mate, have also evolved . Dispersal and kin discrimination behaviors are inherently connected and should be studied in parallel to better understand their relative effect on the patterns of inbreeding in natural populations –.
Mating systems and dispersal are fundamental determinants of population genetic structure . Dispersal can lead to the variations in allele frequencies of local populations, indicated by variation in Fst (Fixation index), a measure of the diversity of randomly chosen alleles within the same sub-population relative to that found in the entire population. The relative fitness of individuals and/or groups may be indirectly indicated by genetic structure, and genetic structure may also be used to predict current and future reproductive success . Fine-scale genetic structure during the breeding season may represent a complex interactive consequence of multiple biological processes . For example, the genetic structure and demography of local populations are tightly linked to the rate and scale of dispersal in the common vole (Microtus arvalis) . Patterns of spatial genetic structure may vary during the year because juveniles appear in the population and dispersal occurs . Fine-scale genetic structure in Malurus cyaneus is attributed to a high rate of extra-pair paternity and skewed reproductive success among individuals . Genetic assignment methods permit the identification of dispersers in populations , and the analysis of genetic structure among populations over time can provide estimates for the effectiveness of dispersal . Thus, resolving the relationship of genetic structure, mating systems and dispersal will allow us to better understand the evolution of inbreeding avoidance.
Brandt’s voles (Lasiopodomys brandtii) are non-hibernating herbivorous rodent species that live in complex social groups. These animals are primarily distributed in typical steppe regions in the middle-eastern Inner Mongolia of China, the Republic of Mongolia and the Baikal Lake region of Russia. Continual outbreaks of Brandt’s vole populations have accelerated the erosion and desertification of grasslands, resulting in a heightened focus on this key rodent pest over the last 30 years. Accordingly, the social behavior and mating system of Brandt’s voles have been carefully investigated, including their mate choice patterns, parental care, olfactory communication, spatial distribution and home range –. The results of these studies indicated that the social behaviors of Brandt’s voles are influenced by many factors, including sex, social hierarchy, reproductive status and seasonal environmental factors. However, the mating system of Brandt’s voles has remained controversial, with previous studies postulating systems ranging from promiscuity and polygyny to monogamy. Molecular detection of multiple paternity revealed that Brandt’s vole females are polyandrous in natural populations . Varied individual exchange ratios among groups during the breeding and non-breeding seasons  suggest varied dispersal patterns in different seasons.
Brandt’s voles live in groups, each of which occupies a burrow system that is conspicuously visible due to connected burrows and runways . In the wild, the earliest births occurs in March , and no over-winter individuals survive through the next August. Thus, in this vole species, no individuals survive for 18 months under the natural environment. Brandt’s voles undergo seasonal reproduction, and the group structure of natural Brandt’s vole populations shows strong periodic seasonal changes. In the non-breeding season, groups are composed of a completely different set of individuals relative to the non-breeding season the year before. At the beginning of the reproductive season, which occurs from approximately March to April in Inner Mongolia, the number of individuals in a group gradually decreases to its lowest level. From about the last 10 days of April to the first 10 days of May, groups are frequently composed of only a male and a female; other group members die or emigrate to a newly recruited breeding group because of mating competition. During the middle and later stages of the reproductive season, which lasts from approximately June to August in Inner Mongolia, the number of individual per group increases to its highest level. In June, groups are frequently composed of a pair of over-winter mates and their progeny, and dispersal becomes common following the sexual maturation of the first generation of the year. Wan reported that males born in April and May can reach sexual maturation at approximately 1.5 months of age, whereas males born in June, July, and August do not reach sexual maturation in the same year. Similarly, females born in April, May, and June can reach sexual maturation at approximately 1 month old and firstly breed at approximately 2 months, whereas females born in July and August do not breed in the same year . The number of litters that a female can produce per year is related to its age: over-winter voles can produce 3–4 litters, and those born in April, May, and June produce 2–3, 1–2, and 0–1 litter in the year of their birth, respectively . At this stage, over-winter individuals begin to die, and the group structure becomes more complex due to dispersal and individual exchanges between groups. Mating behaviors stop around the last 10 days of July and the last progeny are born during the first 10 days of August. September to October is the food-storing stage, a cooperative stage for over-wintering. Exchange of individuals between groups has already been completed, and mating competition has likewise ended . After a peaceful over-winter stage, the new annual reproductive cycle will begin in the spring.
To date, all conclusions about the mating system of Brandt’s voles have been based on observational studies and were primarily conducted under laboratory conditions. No information about the degree of inbreeding or its relationship with the mating system and dispersal of Brandt’s voles in natural populations has been reported. A set of microsatellites recently developed for Brandt’s voles ,  provides a powerful tool to study the group genetic structure and the mating system of this species in depth. The goals of this study were as follows: (i) to determine whether females actively avoided mating with relatives in a group-living rodent species, Brandt’s voles (Lasiopodomys brandtii), by combined analysis of mating systems, dispersal and genetic structure; and (ii) to analyze the relationships among variation in fine-genetic structure, inbreeding avoidance, season-dependent mating strategies and individual dispersal.
Characteristics of the vole microsatellite markers are summarized in Table S1. Typing errors and small allele dominance were not detected by MICRO-CHECKER at any of the 14 loci. The number of alleles at the 14 loci ranged from 2 to 21 with an average of 10.64 alleles per locus. The mean polymorphism information content was 0.690, with a range of 0.363 to 0.880. The average values of the observed and the expected heterozygosities were 0.713 (0.480–0.874) and 0.727 (0.477–0.891), respectively. No loci were detected with null alleles for which the frequencies were greater than 0.05. All 14 loci deviated from Hardy–Weinberg equilibrium (HWE) after the Bonferroni correction among some of the 29 sampled groups (ranging from 2 to 8 groups). All but 6 of the 29 groups had at least 1 locus that deviated from HWE (Supporting information, Table S2). We suggest that non-random mating among groups may explain these departures from HWE.
The Social Composition, Inbreeding Status and Mating System
During the breeding season, only 7 of the 15 groups contained male breeders, and the average across all 15 groups was 0.67 (±0.90) per group (Table 1). Of these, 5 groups contained only 1 male breeder, 1 groups contained 2 male breeders, and 1 groups contained 3 male breeders. In addition, 10 of 15 groups contained female breeders, and the average across all 15 groups was 2.00 (±2.33) per group, ranging from 1 to 8 (Table 1). A paired-samples T-test for the average breeder coefficient [the number of reproductive females (or males)/the number of adult females (or males) in a group] indicated that the proportion of female adults taking part in reproduction was significantly larger than that of males (t = 3.152, df = 13, P = 0.008) (Table 1).
Group pedigree analysis showed that 51.72% of the groups had no identifiable adult female-male pairs with a close-kin relationship (full sibs or half sibs). This included 7 of 15 groups during the breeding season and 8 of 14 groups during the non-breeding season. This result indicates that approximately half of the sampled groups had no risk of inbreeding. The number of identified adult female-male pairs with a close-kin relationship (full sibs or half sibs) was 65 for the other 8 groups during the breeding season. This was an average of 8.13 (±3.56) pairs per group. During the non-breeding season, there were 51 close-kin pairs in the other 6 groups, with an average of 8.50 (±7.74) pairs per group. If mating had occurred randomly, the potential proportion of inbreeding mates would be 6.38% (the number of close-kin pairs/the sum of possible mates) among the whole population, including all samples collected during the two seasons. Overall, 23 pairs of breeding males and females were identified as mates, and no actual pairing was observed between close relatives, which indicates non-random mating in this species. FIS is an inbreeding coefficient introduced by Wright  that expresses the expected rate of homozygosity arising from a given system of breeding, Group specific FIS values showed that none of the 10 groups collected during the breeding season exhibited significant inbreeding (Table 2). Thus, both the results of the group specific FIS analysis and the relationships between identified mates revealed no close-kin mating among these natural Brandt’s vole populations.
Overall, 109 embryos from 13 litters (average litter size of 8.38 (±1.66), ranging from 5–12) were used for paternity analysis (Table 3). Paternity analysis indicated multiple paternity in 92% (12/13) of the litters examined. The average number of fathers per litter, including captured fathers and inferred fathers, was 2.38±0.77 and ranged from 1 to 4. Nine litters contained embryos whose fathers were identified among the captured adult males from the group to which the pregnant female belonged. Of these 9 litters, 8 were sired by only 1 captured father, and 1 was sired by 2 captured fathers. Of 10 identified fathers, 6 were the heaviest individuals and 2 were the second heaviest individuals in the group (Table 1). In addition, a subset of embryos from all of these 9 litters were also sired by at least 1 father that escaped capture but was inferred by the genotype of the pregnant females and her embryos. This indicates that these pregnant females mated with at least 1 male not belonging to her group and supports the theory of extra-group polyandry. In total, 22 fathers, including captured fathers and inferred fathers, were identified in these 9 litters through the genotype of the mother and her progeny. Of the 22 fathers that sired these 9 litters, 1 captured father always inseminated the greatest number of embryos in 8 of the 9 litters (Table 3). These results indicate that females preferentially mate with the heaviest male in their own group, but also frequently mate with males outside of their social group. The results of the pedigree analysis showed that the breeding male copulated with all breeding females in a group. For example, all 6 breeding females copulated with only 1 of 7 adult males in group JUN12. This indicates not only polygyny in this species, but also possible hierarchies in the male society of this species.
Preferential Dispersal of Males Inferred by Variation in Relatedness
A. Within-group relatedness.
During the breeding season, mean pairwise relatedness within groups was significantly greater than zero in 85.7% (12/14) of groups. Mean pairwise relatedness for females within groups was significantly greater than zero in 75.0% (9/12) of groups. Mean pairwise relatedness for males within groups was significantly greater than zero in 41.7% (5/12) of groups (Fig. 1a –1c). During the non-breeding season, mean pairwise relatedness for females within groups was significantly greater than zero in 84.6% (11/13) of groups, and mean pairwise relatedness for males within groups was significantly greater than zero in 84.6% (11/13) of groups (Fig. 1d –1f). These results indicate that there were more groups comprised of close male relatives during the non-breeding season.
Grey lines represent permuted 95% confidence intervals around the null hypothesis of zero relatedness and error bars represent bootstrapped confidence intervals around the mean. a, for the entire group during breeding season; b, for the female members of the group during breeding season; c, for the male members of the group during breeding season; d, for the entire group during non-breeding season; e, for the female members of the group during non-breeding season; f, for the male members of the group during non-breeding season.
During the breeding season, an independent T-test revealed that within-group pairwise relatedness estimates for females were significantly higher than those calculated for males [t = 2.638, df = 22, P (two tailed) = 0.015], though within-group pairwise relatedness estimates for females and for males were not significantly different from within-group pairwise relatedness estimates for the entire group (P>0.05) (Fig. 2). During the non-breeding season, an independent T-test revealed that there were no significant differences among within-group pairwise relatedness estimates for females, males or the entire group (P>0.05) (Fig. 2). These results indicated that within-group close relatives were more commonly female rather than male during the breeding season and that mean pairwise relatedness of males within groups was increased during the non-breeding season. An independent T-test revealed that within-group pairwise relatedness estimates for males were significantly higher during the non-breeding season relative to the breeding season [t = 2.738, df = 23, P (two tailed) = 0.012]; however, there were no significant differences for females or for the entire group (P>0.05) (Fig. 2). These results indicate that there were more close male relatives within a group during the non-breeding season.
B, the entire group during breeding season; NB, the entire group during non-breeding season; BF, the female members of the group during breeding season; NBF, the female members of the group during non-breeding season; BM, the male members of the group during breeding season; NBM, the male members of the group during non-breeding season.
B. Among-group relatedness.
The number of individual pairs identified as close relatives (r ≥0.25) between groups was compared to analyze dispersal between groups.
For the breeding season, a paired-sample T test revealed that the number of female-male pairs identified as close relatives was significantly greater than that of female-female pairs [paired-sample correlation = 0.776, P<0.0001; t = 5.413, df = 90, P (two tailed) <0.0001] and male-male pairs [paired-sample correlation = 0.560, P<0.0001; t = 4.668, df = 92, P (two tailed) <0.001], with no significant difference between the number of female-female pairs and male-male pairs (P>0.05) (Fig. 3).
FF, female-female pairs; MF, female-male pairs; MM, male-male pairs.
For the non-breeding season, a paired-sample T test revealed that the number of female-male pairs identified as close relatives was significantly higher than that of male-male pairs [paired-sample correlation = 0.772, P<0.0001; t = 5.031, df = 90, t P (two tailed) <0.001], and that the number of male-male pairs identified as close relatives was significantly greater than that of female-female pairs [paired-sample correlation = 0.772, P<0.0001; t = 5.14, df = 90, P (two tailed) <0.001] (Fig. 3).
In general, regardless of season, the number of between-group female-male pairs identified as close relatives was the highest and the number of closely related female-female pairs was the lowest.
Together with the variation in the within-group relatedness, these results indicate male-biased dispersal. However, during the breeding season, within-group pairwise relatedness estimates for females were significantly higher than those calculated for males (P<0.01). This result indicates that male-biased dispersal is season-dependent.
Seasonal Variation in Fine-scale Genetic Structure
When individuals were grouped according to the season in which they were collected, an analysis of molecular variance revealed slight but significant differences in genetic structure between the two groups (Pairwise FST = 0.017). Although there was no significant difference between females and males (Pairwise FST <0.001) during the non-breeding season, small but significant differences were observed between females and males during the breeding season, (Pairwise FST = 0.003). Consistent with genetic differentiation between the two seasonal groups, there was significant differentiation between females (Pairwise FST = 0.017) and males (Pairwise FST = 0.018) from the two seasons.
The small difference in pairwise comparison of females and males may result from dispersal during the breeding season. The fact that there was no significant difference between females and males in pairwise comparisons during the non-breeding season was most likely due to the low rate of dispersal and individual exchange during this season. The higher genetic variance observed between the two seasonal populations might be attributed to dispersal and individual exchange during the three months of sampling because all samples were collected from the same region.
At the beginning of the breeding season, Brandt’s vole groups are frequently composed of a single male and a female. If dispersal and individual exchanges do not occur, groups will consist of close relatives after the birth of offspring. Estimates of mean group relatedness revealed that the majority of sampled Brandt’s vole groups were comprised of closely related individuals. Mean pairwise relatedness within groups was significantly greater than zero in 85.7% (12/14) of groups sampled during the breeding season and in 92.9% (13/14) during the non-breeding season. However, only 6.38% of adult female-male pairs were identified as close relatives (including half sibs and full sibs). If random mating occurred, the expected ratio of inbreeding (the number of close-kin pairs/the sum of all possible mates) was only 6.38%. However, no actual pairings between close relatives were observed. This supports the theory that non-random mating occurs in these groups. Furthermore, the group specific FIS values showed that there was no inbreeding in this natural Brandt’s vole population, suggesting that mating occurred in ways that precluded inbreeding.
A pedigree analysis performed on our natural Brandt’s vole population confirmed the polygyny found in previous studies –. Our results showed that the majority of breeding males were the heaviest or second heaviest individual within the groups, and that breeding females mated only with this male within groups. These results indicate that mating within a group is not random, and it may favor inbreeding avoidance by decreasing the probability of females mating with other within-group males that might be close relative. This result also provides an explanation for the observed deviation from HWE. Further parentage analysis showed that Brandt’s vole females were extra-group polyandrous, and that the majority of pregnant females had mated with one or more males outside the group, along with one within-group male. We suggest that extra-group polyandry is a reasonable explanation for the low incidence of inbreeding, because it may increase the frequency of mating among distantly-related individuals.
The mating system of this species implies a possible social hierarchy among male Brandt’s voles. The majority of reproductive males were the heaviest or second heaviest individuals in the group (Table 1). While this strategy prevents random mating within a group, the majority of males have no opportunity to copulate. Mating competition drives low-rank males to emigrate to a new breeding group and increases individual exchange between groups. While females copulate with only one male within the group, the frequent male dispersal facilitates female copulation with more distantly related out-group males.
Increased mean within-group relatedness of males and decreased genetic difference between females and males during the non-breeding season indicate that dispersal occurs mainly during the breeding season. Seasonal variation in female fine-scale genetic structures also supports the dispersal of females, though at a lower frequency than that of males. There were a greater number of between-group female-male pairs identified as close relatives than female-female pairs or male-male pairs, regardless of season. This pattern highlights the preferential separation of closely-related opposite-sex individuals, facilitated by male-biased dispersal. Thus, it appears that inbreeding avoidance during the breeding season drives male-biased dispersal in this species.
Genetic structure refers to any pattern in the genetic makeup of individuals within a population. Due to physical barriers to migration, along with a limited tendency for individuals to move or spread (vagility), and a tendency to remain or come back to their natal place (philopatry), natural populations rarely all interbreed as assumed in theoretical random models (panmixy) . The genetic structure and demography of local populations are tightly linked to the rate and scale of dispersal . The fact that the temporal variance in genetic structure was consistent with the season-dependent dispersal in this species supports this hypothesis. The season-dependent male-biased dispersal is not only the basis of the polygynous and extra-group polyandrous mating strategy in this species but also caused the fine variance in genetic structure between females and males as well as the temporal variance of genetic structure between populations from different seasons.
Many animals have evolved mechanisms that prevent them from breeding with close relatives. Inbreeding avoidance has commonly been used to interpret social behaviors such as reproduction skew, individual dispersal, and construction of the mating system. However, ecological constraints are ultimately believed to provide the selective pressure for these reproductive and behavioral differences, and the levels of inbreeding depression vary across taxa, populations and environments , . In stringent and severe environments, inbreeding can become necessary for the survival and persistence of a group. Cheptou and Donohue (2011) indicated that inbreeding depression is environment-dependent, which often has important ecological and evolutionary consequences. Inbreeding depression in some environments may even contribute to adaptation in others . Inbreeding has been described as a derived trait in the naked mole-rat, Heterocephalus glaber, and might have evolved as an adaptive response to the high costs of dispersal . Inbreeding is tolerable and common in Brandt’s voles under laboratory conditions, and some ecologists have suggested that the mating system changes with habitat. Wang reported evidence of inbreeding in this species , though 3 of 7 microsatellite markers used in their analysis had a high frequency of null alleles, possibly indicating a problem in their analysis. However, the discrepancy with their results might be caused by the different sampling conditions. Additional studies should therefore be conducted to obtain more detailed information regarding variations among groups and social structures throughout a complete annual cycle under different ecological habitats. Such studies will help us to better understand the mechanism of inbreeding avoidance in this species.
In conclusion, the levels of genetic diversity within our natural Brandt’s vole population were affected by seasonal changes in group structure and seasonal reproduction. We conclude that inbreeding avoidance during the breeding season drives the polygynous and extra-group polyandrous mating system that determines the fine-scale genetic structure.
Materials and Methods
Sampling Design and Animal Capture
To assess the genetic and social structure of Brandt’s vole populations over the breeding and non-breeding seasons, a total of 249 individuals were captured from 15 groups during the breeding season, and a total of 200 individuals were captured from 14 groups during the non-breeding season. A summary of the group structures is provided in Table 1. Trapping plots were established in a typical steppe near the town of Aershanbaolige in Xinlingol (N44°40′, E115°40′), Inner Mongolia, China (Fig. S1), which is a typical habitat for Brandt’s voles. Voles were captured on June 18, 2007, for the breeding season analysis. This season was predicted to show the most complex group structure because of sexual maturity, dispersal, and individual exchange. For the non-breeding season analysis, voles were captured on September 28, 2007 This season was predicted to exhibit the most stable group structure because dispersal and individual exchange were complete and mating competition had ended , . All of the individuals in every group were captured by snap-trapping using peanuts as bait. To ensure that all of the individuals were captured, snap-traps were set at all of the group’s burrow entrances, and trapping continued until individuals had stopped merging. The burrow entrances were then blocked with soil and checked the following day to determine whether any individuals entered or exited the group. Each animal captured was weighed and sexed, after which tissue samples were collected from the foot and stored in 75% ethanol until analysis. Embryos of pregnant females large enough to rule out contamination with maternal tissue were carefully separated and stored in 75% ethanol for paternity analysis. Our trapping and handling of Brandt’s voles in the field was approved by the Institutional Animal Use and Care Committee of the Institute of Plant Protection, Chinese Academy of Agricultural Sciences.
Genomic DNA was extracted from tissue samples using the standard proteinase K, phenol, and chloroform protocol . Genetic variation was examined at 14 microsatellite loci that had previously been identified in Brandt’s voles , . Characteristics of the microsatellite markers are summarized in Table S1. One primer of each of the 14 pairs was end-labeled with either FAM or HEX fluorescent dyes (Sangon). The PCRs were performed in 15 µL reaction volumes containing 200 µM of each dNTP, 8 mM Tris-HCl (pH 8.3), 40 mM KCl, 1.2 mM MgCl2, 0.4 µM of each primer, 80 ng of genomic DNA template and 0.25 U Golden DNA Polymerase (TIANGEN). The amplification profiles consisted of initial denaturation at 94°C for 5 min, 35 cycles of 94°C for 30 s, 45 s at the annealing temperature (see Ta in Table 1), and 72°C for 80 s, and a final extension at 72°C for 5 min. The PCR products were then diluted to 50 ng/µL, after which 1-µL of each diluted PCR products was mixed with 1.2 µL deionized formamide and 0.2 µL GeneScan350 ROXTM or GeneScan-500 ROXTM (ABI) internal standard and run on a 377 genetic analyzer (ABI). The fragment sizes of the PCR products were analyzed using GeneScan 3.7 and GeneMarker 1.75. PCR mixtures with known allele sizes were also added to each acrylamide gel as another standard. The program MICRO-CHECKER 2.0  was used to test for the presence of null alleles, short dominance, and typing errors. The Cervus 3.0 software  was used to calculate the number of alleles, the frequency of null alleles, observed heterozygosities, expected heterozygosities, and polymorphic information content (PIC). Genalex 6.1 (Peakall and Smouse 2006) was used to test Hardy–Weinberg equilibrium (HWE).
Group Pedigree Reconstruction and Mating System Inference
The relatedness of individuals within groups was inferred using COLONY version 2.0, which is based on the maximum-likelihood (ML) method. For this analysis, we set the typing error rate at the suggested 0.025 , . The male mating system and female mating system were both specified as polygamous, and all other parameters were set as default. Individuals whi weighed more than 10 g after the removal of viscera were considered physiologically mature  and were used for data analysis. Only the inferred fathers, mothers, full sib pairs and half sib pairs that had a probability of ≥0.95 (as determined by the software) were included in the statistical analysis. During the breeding season, all of the non-adults, including all of the embryos that were developed enough to be separated clearly from the uterus, were used for group pedigree reconstruction. However, only adults were used for comparison of group structures during the breeding season. Paternity of the embryos was inferred using COLONY version 2.0.
Unless stated otherwise, all analyses were performed using Genalex 6.1  and SPSS 15.0 for Windows (SPSS Inc). The means, plus or minus one standard deviation, are reported, unless otherwise noted.
Group specific FIS values were estimated using the Arlequin 3.11 software  to detect signs of inbreeding. Statistical significance was tested by1000 random permutations. To avoid the possible influences of emigrated and immigrated group members, only embryos and juveniles were used to estimate the group-specific FIS values.
Individual dispersal and group member exchange inference.
Individual dispersal and recombination were inferred by analyzing the relatedness of individuals from different groups. Pairwise relatedness estimates devised by Queller and Goodnight  were calculated for each pair of individuals, and mean within-group pairwise relatedness for each group was estimated at the 95% confidence interval via bootstrapping. Random permutation of the dataset was used to generate a distribution for the null hypothesis of no relatedness among individuals within groups and to provide a test for significance. All bootstrapping and permutational tests were performed 1000 times. Mean within-group pairwise relatedness for each sex in each group was estimated and compared with the total within-group pairwise relatedness of each group.
Close relative pairs (individual pairs with r ≥0.25) between every two groups, including female-female pairs, female-male pairs and male-male pairs, were calculated separately and compared to each other using one-way analysis of variance (ANOVA) in order to detect the evidence of sex-biased dispersal.
Analyses and comparison of genetic structure.
An analysis of molecular variance (AMOVA) was used to estimate the group pairwise FSTs. These group pairwise FSTs were estimated separately for each sex in order to detect evidence of sex-biased dispersal. Statistical significance was tested by 1000 random permutations. Individuals collected during the same season were treated as a population, and AMOVA was used to estimate population pairwise FSTs and the partitioning of genetic variation within and between seasonal populations.
Geographical positions of sampled Brandt’s vole groups by GPS in Inner Mongolia, China. Groups with blue flags have been sampled on June 18th, 2007 for breeding season, and groups with red flags have been sampled on September 28th, 2007 for non-breeding season.
Summary statistics for 19 microsatellite loci of Lasiopodomys brandtii .
Conceived and designed the experiments: XHL. Performed the experiments: LFY DWW NL LC. Analyzed the data: XHL LFY. Wrote the paper: LXH YLF.
- 1. Cheptou PO, Donohue K (2011) Environment-dependent inbreeding depression: its ecological and evolutionary significance. New Phytol 189: 395–407.
- 2. Wright S (1931) Evolution in Mendelian Populations. Genetics 16: 97–159.
- 3. Keller LF, Waller DM (2002) Inbreeding effects in wild populations. Trends in Ecology & Evolution 17: 230–241.
- 4. Charlesworth B, Charlesworth D (1999) The genetic basis of inbreeding depression. Genet Res 74: 329–340.
- 5. Charlesworth D, Willis JH (2009) The genetics of inbreeding depression. Nat Rev Genet 10: 783–796.
- 6. Lihoreau M, Zimmer C, Rivault C (2008) Mutual mate choice: when it pays both sexes to avoid inbreeding. PLoS One 3: e3365.
- 7. Archie EA, Hollister-Smith JA, Poole JH, Lee PC, Moss CJ, et al. (2007) Behavioural inbreeding avoidance in wild African elephants. Mol Ecol 16: 4138–4148.
- 8. Moore J (2007) Phenotype matching and inbreeding avoidance in African elephants. Mol Ecol 16: 4421–4423.
- 9. Tregenza T, Wedell N (2000) Genetic compatibility, mate choice and patterns of parentage: invited review. Mol Ecol 9: 1013–1027.
- 10. Pusey A, Wolf M (1996) Inbreeding avoidance in animals. Trends in Ecology & Evolution 11: 201–206.
- 11. Zeh JA, Zeh DW (1996) The Evolution of Polyandry I: Intragenomic Conflict and Genetic Incompatibility. Proceedings of the Royal Society of London. Series B: Biological Sciences 263: 1711–1717.
- 12. Zeh JA, Zeh DW (1997) The evolution of polyandry II: post–copulatory defenses against genetic incompatibility. Proceedings of the Royal Society of London. Series B: Biological Sciences 264: 69–75.
- 13. Zeh J, Zeh D (2008) Maternal inheritance, epigenetics and the evolution of polyandry. Genetica 134: 45–54.
- 14. Cornell SJ, Tregenza T (2007) A new theory for the evolution of polyandry as a means of inbreeding avoidance. Proc Biol Sci 274: 2873–2879.
- 15. Bretman A, Newcombe D, Tregenza T (2009) Promiscuous females avoid inbreeding by controlling sperm storage. Mol Ecol 18: 3340–3345.
- 16. Firman RC, Simmons LW (2008) Polyandry facilitates postcopulatory inbreeding avoidance in house mice. Evolution 62: 603–611.
- 17. Simmons LW, Beveridge M, Wedell N, Tregenza T (2006) Postcopulatory inbreeding avoidance by female crickets only revealed by molecular markers. Mol Ecol 15: 3817–3824.
- 18. Fisher DO, Double MC, Blomberg SP, Jennions MD, Cockburn A (2006) Post-mating sexual selection increases lifetime fitness of polyandrous females in the wild. Nature 444: 89–92.
- 19. Thuman KA, Griffith SC (2005) Genetic similarity and the nonrandom distribution of paternity in a genetically highly polyandrous shorebird. Animal Behaviour 69: 765–770.
- 20. Ivy TM, Sakaluk SK (2005) Polyandry promotes enhanced offspring survival in decorated crickets. Evolution 59: 152–159.
- 21. Evans JP, Marshall DJ (2005) Male-by-female interactions influence fertilization success and mediate the benefits of polyandry in the sea urchin Heliocidaris erythrogramma. Evolution 59: 106–112.
- 22. Tregenza T, Wedell N (2002) Polyandrous females avoid costs of inbreeding. Nature 415.
- 23. Newcomer SD, Zeh JA, Zeh DW (1999) Genetic benefits enhance the reproductive success of polyandrous females. Proc Natl Acad Sci U S A 96: 10236–10241.
- 24. Olsson M, Shine R, Madsen T, Gullberg A, Aring H, et al.. (1996) Sperm selection by females. Nature 383.
- 25. Bowler DE, Benton TG (2005) Causes and consequences of animal dispersal strategies: relating individual behaviour to spatial dynamics. Biol Rev Camb Philos Soc 80: 205–225.
- 26. Greenwood PJ (1980) Mating systems, philopatry and dispersal in birds and mammals. Animal Behaviour 28: 1140–1162.
- 27. Lebigre C, Alatalo RV, Siitari H (2010) Female-biased dispersal alone can reduce the occurrence of inbreeding in black grouse (Tetrao tetrix). Mol Ecol 19: 1929–1939.
- 28. Lehmann L, Perrin N (2003) Inbreeding avoidance through kin recognition: choosy females boost male dispersal. Am Nat 162: 638–652.
- 29. Höner OP, Wachter B, East ML, Streich WJ, Wilhelm K, et al.. (2007) Female mate-choice drives the evolution of male-biased dispersal in a social mammal. Nature 448.
- 30. van de Casteele T, Matthysen E (2006) Natal dispersal and parental escorting predict relatedness between mates in a passerine bird. Mol Ecol 15: 2557–2565.
- 31. Lee JW, Jang BS, Dawson DA, Burke T, Hatchwell BJ (2009) Fine-scale genetic structure and its consequence in breeding aggregations of a passerine bird. Mol Ecol 18: 2728–2739.
- 32. Bretman A, Rodriguez-Munoz R, Walling C, Slate J, Tregenza T (2011) Fine-scale population structure, inbreeding risk and avoidance in a wild insect population. Mol Ecol 20: 3045–3055.
- 33. Scribner KT, Chesser RK (1993) Environmental and demographic correlates of spatial and seasonal genetic structure in the eastern cottontail (Sylvilagus floridanus). Journal of mammalogy: 1026–1044.
- 34. Double MC, Peakall R, Beck NR, Cockburn A (2005) Dispersal, philopatry, and infidelity: dissecting local genetic structure in superb fairy-wrens (Malurus cyaneus). Evolution 59: 625–635.
- 35. Manel S, Gaggiotti OE, Waples RS (2005) Assignment methods: matching biological questions with appropriate techniques. Trends Ecol Evol 20: 136–142.
- 36. Schweizer M, Excoffier L, Heckel G (2007) Fine-scale genetic structure and dispersal in the common vole (Microtus arvalis). Mol Ecol 16: 2463–2473.
- 37. Chen G, Shi D (2003) Study on Reproductive Behavior of Different Social Hierarchy of Brandt’s Vole. Acta Theriologica Sinica 23: 220–224.
- 38. Shi D, Zheng S (1998) Study on the Affinity Relation in Colonies of Brandt’s Vole. ACTA AGRESTIA SINICA 6: 53–58.
- 39. Wan X, Liu W, Wang G, Wang M, Zhong W (2006) Seasonal changes of the activity patterns of Brandt’s vole (Lasiopodomy brandtii) in the typical steppe in Inner Mongolia. Acta Theriologica Sinica 26: 226–234.
- 40. Wan X, Zhang X, Liu W, Wang G, Wang M, et al. (2007) Social hierarchy and its seasonal changes of marked Lasiopodomys brandtii population. Chinese Journal of Ecolog 26: 336–359.
- 41. Xie X, Sun R (1994) The mating system and reproduction of Brandt’s voles. Acta Zoologica Sinica 40: 262–265.
- 42. Yin F, Fang J (1998) Mate choice in Brandt’s voles (Microtus brandtii). Acta Zoologica Sinica 44: 162–169.
- 43. Yin F, Fang J (1998) Comparison of parental behavior in Brandt’s voles. Acta Theriologica Sinica 18: 277–281.
- 44. Yu X, Sun R, Fang J (2004) Effect of kinship on social behaviors in Brandt’s voles (Microtus brandtii). Journal of Ethology 22: 17–22.
- 45. Zhang J, Shi D (2007) Influence of male hierarchy on female choice of Lasiopodomys brandtii. Chinese Journal of Ecology 26: 219–222.
- 46. Zhang J, Shi D (2006) Selection of group odor and its relationship with dominance of Brandt’s vole Lasiopodomys brandtii. Acta Zoologica Sinica 52: 224–228.
- 47. Zhang J, Shi D (2005) Social Dominance of Male Brandt’s Vole. Chinese Journal of Zoology 40: 19–24.
- 48. Huo Y, Wan X, Wolff JO, Wang G, Thomas S, et al. (2010) Multiple paternities increase genetic diversity of offspring in Brandt’s voles. Behavioural Processes 84: 745–749.
- 49. Wan X, Zhong W, Wang M (2001) Estimating the group reformative rate for social rodent species. Acta Theriologica Sinica 21: 67–72.
- 50. Zhong W, Wang G, Zhou Q, Wang G (2007) Communal food caches and social groups of Brandt’s voles in the typical steppes of Inner Mongolia, China. Journal of Arid Environments 68: 398–407.
- 51. Zhang J, Zhong W (1979) Investigations of reproduction in populations of Brandt’s voles. Acta Zoologica Sinica 25: 250–259.
- 52. Yue L, Wang D, Huang B, Liu X (2009) Characterization of nine novel microsatellite markers from Brandt’s vole (Lasiopodomys brandtii). Molecular Ecology Resources 9: 1194–1196.
- 53. Wan X, Wang M, Wang G, Liu W, Zhong W (2002) The reproductive parameters in the marked populations of Brandt’s vole. Acta Theriologica Sinica 22: 116–122.
- 54. Wang D, Shi D (2007) Isolation and characterization of polymorphic microsatellite loci from Brandt’s voles (Lasiopodomys brandtii). Molecular Ecology Notes 7: 671–673.
- 55. Faulkes CG, Bennett NC (2001) Family values: group dynamics and social control of reproduction in African mole-rats. Trends in Ecology & Evolution 16: 184–190.
- 56. Wright S (1921) Systems of Mating. I. the Biometric Relations between Parent and Offspring. Genetics 6: 111–123.
- 57. Duffie CV, Glenn TC, Vargas FH, Parker PG (2009) Genetic structure within and between island populations of the flightless cormorant (Phalacrocorax harrisi). Mol Ecol 18: 2103–2111.
- 58. Faulkes CG, Bennett NC (2001) Family values: group dynamics and social control of reproduction in African mole-rats. Trends in Ecology & Evolution 16: 184–190.
- 59. Wang D, Guo Y, Shi D (2011) Genetic structure of Brandt’s vole (Lasiopodomys brandtii) populations in Inner Mongolia, China, based on microsatellite analysis. Conservation Genetics 12: 659.
- 60. Russell DW, Sambrook J (2001) Molecular cloning: a laboratory manual. N.Y.: Cold Spring Harbor Laboratory Press.
- 61. Van Oosterhout C, Hutchinson WF, Wills DPM, Shipley P (2004) micro-checker: software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes 4: 535–538.
- 62. Kalinowski ST, Taper ML, Marshall TC (2007) Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment. Mol Ecol 16: 1099–1106.
- 63. Wang J, Santure AW (2009) Parentage and sibship inference from multilocus genotype data under polygamy. Genetics 181: 1579–1594.
- 64. Wang J (2004) Sibship reconstruction from genetic data with typing errors. Genetics 166: 1963–1979.
- 65. Wang D, Cong L, Liu X (2010) Seasonal variation of population parameters and physiological characteristics of Brandt’s vole (Lasiopodomys brandtii). ACTA ECOLOGICA SINICA 30: 3562–3568.
- 66. Peakall R, Smouse PE (2006) genalex 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes 6: 288–295.
- 67. Excoffier LGL, Schneider S (2005) Arlequin ver. 3.0: An integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online 1: 47–50.
- 68. Queller D, Goodnight K (1989) Estimating relatedness using genetic markers. Evolution 43: 258–275.