Inbreeding load and inbreeding depression estimated from lifetime reproductive success in a small, dispersal-limited population

Abstract

The fitness consequences of inbreeding and the individual behaviors that prevent its detrimental effects can be challenging to document in wild populations. Here, we use field and molecular data from a 17-year study of banner-tailed kangaroo rats (Dipodomys spectabilis) to quantify the relationship between inbreeding, mate kinship, and lifetime reproductive success. Using a pedigree that was reconstructed using genetic and field data within a Bayesian framework (median probability of parental assignment = 0.92, mean pedigree depth = 6 generations), we estimated both inbreeding coefficients and kinship between individuals that produced offspring (mean inbreeding coefficient = 0.07, mean mate kinship = 0.08). We also used the pedigree, in combination with census data, to generate a series of fitness estimates, ranging from survival to reproductive maturity to lifetime reproductive success. We found that the population's inbreeding load was low to moderate (0.98-4.66 haploid lethal equivalents) and increased with the time frame over which fitness was estimated (lowest for survival to maturity, highest for adult-to-adult reproductive success). Fitness decreased with increasing inbreeding coefficients. For example, lifetime reproductive success was reduced by 24% for individuals with inbreeding coefficients greater than twice the population mean. Within full sibling pairs, the sibling with less-related mates produced an average of 30% more offspring over its lifetime. These data further illustrate that inbreeding can have a negative effect on lifetime reproductive success.

Fig. 1

Location of study site in southeast Arizona. Points depict the locations of all occupied mounds across the 17-year-study period. Mounds occurred around the perimeter of a volcanic cinder cone. Top right inset show site location in North America. Bottom right inset illustrates the study species, banner-tailed kangaroo rat (Dipodomys spectabilis)

Fig. 2

Hypothetical pedigree displaying pedigree calculations. In this example, individuals N (female) and P (male) have different histories of inbreeding in their family lines, leading to different inbreeding coefficients (F). The kinship (k) between the two individuals is smaller than either of their inbreeding coefficients because they have only one known ancestor in common (individual c). All letters represent distinct individuals

Fig. 3

Relationship between pedigree-estimated inbreeding coefficient and lifespan. The line depicts the generalized linear regression line (birth year treated as random variable; Poisson errors; variance explained by fixed effects only). On average, individuals that were less inbred survived longer than individuals with higher inbreeding coefficients. (See Table 2 for coefficient estimates and other model results)

Fig. 4

Pedigree estimates of fitness as a function of individual inbreeding coefficient a, c or mean kinship between an individual and all of its mates b, d. In a, b the y-axis depicts the number of juvenile offspring, whereas in c, d it represents lifetime reproductive success. In all plots, inbreeding and fitness estimates were compared using a generalized linear model with Poisson errors, treating parental birth year as a random variable. In each panel, the resulting regression line is plotted over the individual data points. (See Table 2 for coefficient estimates and other model results)

Fig. 5

Difference in relative reproductive success between siblings within a full-sibling pair as a function of difference in mean mate kinship between the same siblings. In a, the y-axis shows the difference between siblings in number of juvenile offspring. In b, it denotes the difference in lifetime reproductive success. In both panels, reproductive success values are relative to the sibling with the larger mean mate kinship such that values <1 indicate that the sibling with the more closely related mates had fewer offspring. The dashed horizontal line indicates the expectation if kinship has no effect on reproductive success. The blue line illustrates the results of a generalized mixed-effect linear model, where birth year was modeled as a random variable. (See Table 2 for coefficient estimates and other model results)