Postcopulatory sexual selection is associated with sperm aggregate quality in Peromyscus mice

Abstract

In some species, sperm form coordinated groups that are hypothesized to improve their swimming performance in competitive contexts or to navigate through the viscous fluids of the female reproductive tract. Here we investigate sperm aggregation across closely related species of Peromyscus mice that naturally vary by mating system to test the predictions that sperm aggregates 1) are faster than solitary sperm in species that females mate multiply to aid cells in sperm competition, and 2) outperform solitary sperm cells in viscous environments. We find significant variation in the size of sperm aggregates, which negatively associates with relative testis mass, a proxy for sperm competition risk, suggesting that postcopulatory sexual selection has a stabilizing effect on sperm group size. Moreover, our results show that sperm aggregates are faster than solitary sperm in some, but not all, species, and this can vary by fluid viscosity. Of the two species that produce the largest and most frequent groups, we find that sperm aggregates from the promiscuous P. maniculatus are faster than solitary sperm in every experimentally viscous environment but aggregation provides no such kinematic advantage under these same conditions for the monogamous P. polionotus. The reduced performance of P. polionotus aggregates is associated with less efficient aggregate geometry and the inclusion of immotile or morphological abnormal sperm. Our cross-species comparison yields insight into the evolution of sperm social behaviors, provides evidence of extensive variation in the Peromyscus lineage, and reveals that differences in sperm aggregate quality associate with postcopulatory sexual selection.

Keywords: mating systems; sexual selection; sperm competition; sperm conjugation; sperm motility.

Figure 1

Sperm aggregate size differs within and among Peromyscus mice, and variability in this trait negatively associates with relative testis size. (A) Observed variation in the number of sperm cells per aggregate produced by males under controlled, in vitro conditions. Box-plots represent median and interquartile ranges with mean values per male overlaid as black dots. Species relationships are indicated within the phylogeny (adapted from Bradley et al. 2007), and open circles represent the evolutionary transitions to monogamy (Bedford & Hoekstra 2015) such that P. californicus, P. eremicus, and P. polionotus are monogamous, while the other species are promiscuous. (B) When controlling for phylogenetic relationships across these species using a PGLS model, the coefficient of variation (CV) for the number of sperm cells in aggregate negatively correlates with relative testes mass. Note truncated y-axes.

Figure 2

The frequency of sperm aggregates varies across focal Peromyscus species and can affect sperm swimming performance. (A) Sperm aggregates (solid bars) and solitary sperm (open bars) are simultaneously produced by males of all focal species, allowing for a direct comparison of their speeds while controlling for inter-male variation to test the hypothesis that sperm aggregates provide a competitive advantage. The summed number of sperm aggregates observed within each species is indicated in brackets. The curvilinear velocity of aggregated (closed circles) and solitary sperm (open circles) were observed in either low (B) or high (C) viscosity media to test the hypothesis that sperm aggregates are faster in viscous environments. In both environments, sperm aggregates are faster in P. maniculatus. Aggregates are also faster in P. californicus in the high-viscosity environment, but aggregates are slower in P. polionotus and P. gossypinus in the low-viscosity environment. All other species produced aggregates that did not differ in speed from solitary sperm. Circles represent mean values per male, and black dashes represent species mean per sperm type. P. californicus, P. eremicus, and P. polionotus are monogamous, while the other species are promiscuous. * P < 0.05, ** P < 0.01, *** P < 0.001. Note the truncated x-axes.

Figure 3

The alignment of sperm within aggregates varies across focal Peromyscus species and impacts aggregate speed. (A) Sperm aggregates and composition were scored as “abnormal” if they featured one or more cells that were immotile, misaligned, morphologically abnormal, or had a combination of these issues. (B) Abnormal aggregates (triangles) are slower than normal aggregates (circles) in all but two species (P. leucopus and P. gossypinus). P. californicus, P. eremicus, and P. polionotus are monogamous, while the other species are significantly promiscuous. Symbols represent mean values per male, and black dashes represent species means per aggregate type.