Fertilization mode differentially impacts the evolution of vertebrate sperm components

Abstract

Environmental change frequently drives morphological diversification, including at the cellular level. Transitions in the environment where fertilization occurs (i.e., fertilization mode) are hypothesized to be a driver of the extreme diversity in sperm morphology observed in animals. Yet how fertilization mode impacts the evolution of sperm components-head, midpiece, and flagellum-each with different functional roles that must act as an integrated unit remains unclear. Here, we test this hypothesis by examining the evolution of sperm component lengths across 1103 species of vertebrates varying in fertilization mode (external vs. internal fertilization). Sperm component length is explained in part by fertilization mode across vertebrates, but how fertilization mode influences sperm evolution varies among sperm components and vertebrate clades. We also identify evolutionary responses not influenced by fertilization mode: midpieces evolve rapidly in both external and internal fertilizers. Fertilization mode thus influences vertebrate sperm evolution through complex component- and clade-specific evolutionary responses.

Fig. 1. Sperm morphology across vertebrates.

a The phylogeny of the species in our dataset where sperm data are available. Sperm head (orange), midpiece (green), and flagellum (purple) length (plotted as natural log-transformed values) varied within and among vertebrate clades. Colors on the tips of the tree indicate whether the species is an external fertilizer (blue) or an internal fertilizer (red). Data were available from six vertebrate clades (i.e., superclass/class), including Chondrichthyes (n = 51), Osteichthyes (n = 134), Amphibia (n = 104), Reptilia (n = 117), Aves (n = 237) and Mammalia (n = 460, Table S1). b Sperm morphology varied among major vertebrate clades. The average percent of each sperm component for each vertebrate clade is presented for (from top to bottom) Chondrichthyes, Osteichthyes, Amphibia, Reptilia, Aves, and Mammalia. For Osteichthyes and Amphibia, average sperm morphologies are presented separately for external (Ext.) and internal (Int.) fertilizing species. As expected, there were clear differences in the lengths of different sperm components across vertebrates: sperm head was shortest, followed by midpiece, and the flagellum was longest on average (Table S1). This general pattern (i.e., average sperm head <midpiece <flagellum length) was also found among Mammalia and Reptilia (Table S1). However, a different pattern was observed in other vertebrate clades. The sperm head was longer on average than the midpiece in Chondrichthyes, Osteichthyes, and Amphibia, while in Aves the sperm midpiece was longer on average than either the sperm head or flagellum (Table S1). The stylized sperm drawings represent common sperm morphologies in each group and are used to illustrate broad similarities and differences among vertebrate clades. However, we acknowledge that there is wide variation in sperm morphologies within these vertebrate clades that is not depicted here. Silhouette illustrations contributed by various authors under public domain license (CC0 1.0 license) from PhyloPic (http://phylopic.org). Source data are provided as a source data file.

Fig. 2. Sperm length for internally and externally fertilizing vertebrates.

Sperm length (on natural log scale) of the sperm head (a), midpiece (b), and flagellum (c) for external (blue) and internal fertilizers (red), plotted with the mean of each group. Plotted are values for all external and internal fertilizers in our dataset, and each of the major vertebrate classes: (from left to right) Osteichthyes, Amphibia, Chondrichthyes, Reptilia, Aves, and Mammalia. Osteichthyes and Amphibia are broken into the external and internal fertilizing species. Significant differences in sperm component lengths between fertilization modes are provided for comparisons between externally fertilizing vertebrates (N = 1103), fish (N = 134), and amphibians (N = 104) which were calculated using phylogenetic linear models (* = P < 0.05, see Table S3 for statistical model outputs for each comparison). Silhouette illustrations contributed by various authors under public domain license (CC0 1.0 license) from PhyloPic (http://phylopic.org). Source data are provided as a Source Data file.

Fig. 3. Rates of evolution of the sperm head, midpiece, and flagellum lengths.

Rates of evolution were estimated using mvMORPH for all vertebrates in our dataset (Full dataset, a), external and internal fertilizers (divided under Vertebrates), and for each vertebrate class (from left to right) including Osteichthyes, Amphibia, Chondrichthyes, Reptilia, Aves, and Mammalia (b). Separate analyses for internal and externally fertilizing Osteichthyes and Amphibia are presented. Rates of evolution are presented for sperm head (orange), midpiece (green) and flagellum (purple) in all plots. Internal fertilizers are on the top row, and external fertilizers are on the bottom row. Mean and standard error estimates for each evolutionary rate were estimated using a boot-strapping approach with 100 bootstrap samples. Differences in the rates of evolution of sperm components were determined using post hoc tests that compared the rates of evolution in a pairwise manner using AICc to determine the model with the best fit (rates were equal, or rates were different, see details in “Methods”). When models with different rates had the lowest AICc value, rates were considered to be significantly difference from one another, which is indicated with letters above the estimated rate values. Silhouette illustrations contributed by various authors under public domain license (CC0 1.0 license) from PhyloPic (http://phylopic.org). Source data are provided as a Source Data file.