Frequency of Multiple Paternity in Tope (Galeorhinus galeus) from California and Australia
Multiple paternity (MP) has been widely documented in elasmobranch fishes (sharks, rays, and skates); however, the frequency of multiple paternity (FMP; the percentage of sampled litters having multiple sires) varies greatly between and within species. Understanding the reasons for this inter- and intraspecific variation in FMP may be key to understanding why females mate with multiple males. For the Tope (Galeorhinus galeus), one previous study detected multiple paternity in two out of five litters sampled from New Zealand (FMP = 40.0%), an unexpectedly low FMP given this species’ reproductive biology and ecology. To further test the hypothesis of low FMP in G. galeus, we expanded on previous work by genotyping an additional ten litters sampled from central South Australia, Australia (mean litter size = 32.6 ± 7.8 pups) as well as ten litters sampled from southern California, USA (mean litter size = 30.9 ± 8.5 pups). Multiple paternity was detected in four of the ten litters sampled from Australia (FMP = 40.0%) and two of the ten litters sampled from California (FMP = 20.0%). Our results support the hypothesis that MP occurs at a relatively low frequency in G. galeus, as compared to other elasmobranch species.
ELASMOBRANCH fishes (sharks, rays, and skates) generally exhibit a polyandrous mating system, where females mate with multiple males in a single breeding season (Bester-van der Merwe et al., 2022). This behavior, combined with sperm storage in the reproductive tracts of females (Pratt, 1993), can result in multiple paternity (MP), that is, litters sired by multiple males (reviewed by Lyons et al., 2021). The reasons why female elasmobranchs mate with multiple males remain elusive given the potential physical costs of mating to females and that the sperm from just one male is likely more than enough to fertilize all of a female’s eggs. One possible explanation for the presence of MP in elasmobranchs is convenience polyandry where mating with multiple males is simply the result of sexual conflict and females concede to superfluous matings to avoid excessive harassment by males (e.g., Feldheim et al., 2004; DiBattista et al., 2008; Daly-Engel et al., 2010). While convenience polyandry may explain this behavior for some species, it has not been systematically tested and ignores possible benefits to females (Bester-van der Merwe et al., 2022). As in other taxa, polyandry may facilitate genetic bet-hedging (increasing genetic diversity among offspring), trading up (accepting mating attempts from males of higher genetic quality than those previously encountered), cryptic female choice (postcopulatory mechanisms that favor fertilization by sperm of some males over others), and sperm competition (male offspring may have greater reproductive success if competitive sperm is a heritable trait), that could benefit the female indirectly by producing higher quality offspring with better survival and mating success (Lyons et al., 2021). If this is the case and females do benefit from mating with multiple males, we might expect the frequency of multiple paternity (FMP; the percentage of sampled litters sired by multiple males) to be high for most elasmobranch species.
Although MP has been documented in most elasmobranch species where it has been studied, the FMP varies greatly between species and sometimes between populations of the same species (reviewed by Lyons et al., 2021). For example, the Small-spotted Catshark Scyliorhinus canicula has a high FMP (92.3%; Griffiths et al., 2012), the Shortspine Spurdog Squalus mitsukurii a low FMP (11.1%; Daly-Engel et al., 2010), and the Blacktip Shark Carcharhinus limbatus an intermediate FMP (50.0%; Bester-van der Merwe et al., 2019). Among species with intraspecific variation in FMP, the Brown Smoothhound Mustelus henlei is particularly notable with an FMP of 92.9% off the Pacific coast of Baja California Sur, Mexico (Byrne and Avise, 2011), but only 22.2% off Santa Catalina Island, California, USA (Chabot and Haggin, 2014). Understanding the reasons for this inter- and intraspecific variation in FMP may be the key to understanding the reasons why female elasmobranchs mate with multiple males.
For Tope, Galeorhinus galeus, also known as Soupfin Sharks and School Sharks, only Hernández et al. (2014) have investigated MP, detecting multiple sires in two out of five litters (FMP = 40.0%) sampled from New Zealand waters. This FMP was unexpectedly low, given this species’ reproductive biology and ecology (see below), and was based on a small sample size of only five litters. We sought to further test the hypothesis of low FMP by genotyping an additional ten litters of G. galeus from Australia (G. galeus from Australia and New Zealand comprise a panmictic population; Hernández et al., 2015; Devloo-Delva et al., 2019), as well as ten litters from California, part of a population that is genetically distinct from the Australia–New Zealand population (Chabot and Allen, 2009; Chabot, 2015).
We hypothesized a higher FMP for both populations of G. galeus than that reported by Hernández et al. (2014), given that this species exhibits sperm storage and has a protracted mating season (that may exceed five months; Peres and Vooren, 1991). This species also has a propensity to aggregate, which may increase the amount of contact between the sexes and therefore the probability of multiple mating leading to multiple paternity. Although Ripley (1946) reported latitudinal segregation of the sexes between northern California (97.5% of landings were males) and southern California (98.7% of landings were females), an approximately 1:1 sex ratio was reported for G. galeus captured off the Central Coast of California, between San Francisco and the Northern Channel Islands. A long-term acoustic telemetry study conducted by Nosal et al. (2021) found that mature females spend two years of their triennial reproductive cycle along the Central Coast of California, where they were expected to give birth and mate, before migrating to southern California for their 12-month gestation period. Similar latitudinal sexual segregation has been reported for Australia, with males dominating higher latitudes off Tasmania while females frequent the warmer waters of the Great Australian Bight (Olsen, 1984; Walker et al., 2008). In May and June (Austral winter), breeding aggregations form on the edge of the continental shelf where the sexes are well mixed (Olsen, 1954). Gestating females aggregate in the Great Australian Bight before undertaking a partial migration in which some gravid females migrate to Tasmania and New Zealand while others remain in the Great Australian Bight to give birth (McMillan et al., 2019). In short, given the mixing of sexes reported for both populations, we expected a relatively high FMP for G. galeus off Australia and California.
MATERIALS AND METHODS
Tissue sample collection and DNA extraction.—
Pregnant G. galeus were captured opportunistically off southern California, USA, and central South Australia, Australia between 2010 and 2020, with most females captured between 2018 and 2020 (Table 1). Total length was recorded for each mother and, when possible, each pup, along with their sex. Tissue samples (fin clips) were taken from each mother and pup and preserved in 95% ethanol until DNA extraction. Fin clips were also taken from non-pregnant G. galeus captured from each population (to calculate population-specific allele frequencies) and preserved in 95% ethanol. These included 45 G. galeus captured off San Diego County, California, 35 around the Northern Channel Islands, California, 23 off Central South Australia, Australia, and 17 from Western Bass Strait, Australia. DNA was extracted from each fin clip using a PureLink Genomic DNA Mini Kit (Invitrogen) following the manufacturer’s instructions, and the eluted DNA was stored at –20°C until polymerase chain reaction (PCR) amplification. Samples from California were collected under University of California San Diego IACUC Protocol S00080 and California Department of Fish and Wildlife Scientific Collecting Permit 183020007. Australian samples were collected by commercial vessels operating under scientific permits from the Australian Fisheries Management Authority (AFMA) for project 2019/0841.
Microsatellite amplification and genotyping.—
A total of six microsatellites were amplified by PCR from each tissue sample. These included five species-specific microsatellites developed for G. galeus by Chabot and Nigenda (2011; Gg15, Gg18, Gg22, Gg23) and Hernández et al. (2015; Ggal15), and another microsatellite originally developed for Mustelus canis by Giresi et al. (2012; Mca37) but known to cross-amplify with G. galeus (Maduna et al., 2014). Each PCR contained three oligonucleotide primers (Boutin-Ganache et al., 2001): a forward primer with a 5′ M13 tail (5′-GTAAAACGACGGCCAG-3′), a reverse primer with a 5′ pigtail (5′-GTTTCT-3′), and a 5′ fluorescently labeled (6FAM or HEX) M13 primer (5′-GTAAAACGACGGCCAG-3′). The complete sequences for all primers are given in Supplemental Table 1 (see Data Accessibility). Each 10 μl PCR contained 2.5 μl BSA solution (1.6 mg/ml), 1.0 μl primer mix (0.1 μM forward primer with M13 tail, 2 μM reverse primer with pigtail, 0.1 μM fluorescently labeled M13 primer), 5.0 μl Multiplex PCR Master Mix (Qiagen), and 1.5 μl template DNA. Amplifications were performed on a VeritiPro Thermal Cycler (Applied Biosystems) with an initial denaturation step at 95°C for 15 min, followed by 25 cycles of denaturation at 95°C for 30 s, annealing at 59°C for 90 s, and elongation at 72°C for 60 s, followed by 20 cycles of 95°C for 30 s, 53°C for 90 s, and 72°C for 60 s, followed by a final elongation step at 60°C for 30 min. PCR products were diluted with ultrapure water by a factor of 10. Diluted PCR product (1 μl) was added to 10 μl of Hi-Di Formamide (Applied Biosystems) pre-mixed with GeneScan 500 ROX size standard (Applied Biosystems) in a 100:1 ratio (1000 μl formamide mixed with 10 μl ROX size standard), denatured for 3 min at 95°C, and separated by capillary electrophoresis on an ABI 3730 Genetic Analyzer (Applied Biosystems). Allele sizes were visualized and scored using Peak Scanner Software v1.0 (Applied Biosystems).
Data analyses.—
To ensure genetic homogeneity of the California samples, we employed Structure v.2.3.2 (Pritchard et al., 2000; Falush et al., 2003; Hubisz et al., 2009) using the following parameters: admixture with independent allele frequencies between populations with length of burnin period and number of MCMC reps after burnin both set to 1,000,000. This analysis was repeated with sampling locations as priors (Hubisz et al., 2009). We tested K = 1–5 with ten iterations for each K-value. To assess the most likely number of populations from these analyses, we employed STRUCTURESELECTOR (Li and Liu, 2018).
Prior to analyses, pup genotypes were checked to ensure they shared at least one allele per microsatellite locus with their respective mother. For each microsatellite locus in each population (Australia and California), Genepop Web Service v4.7.5 (Raymond and Rousset, 1995) was used to determine allele frequencies, as well as the observed and expected heterozygosity, and to test for linkage disequilibrium and conformance to Hardy-Weinberg equilibrium. PrDM v1.0 (Neff and Pitcher, 2002) was used post hoc to calculate the probability of detecting MP for each litter, for two fathers (the maximum we observed for any litter) with each of six hypothetical paternal skews: 50:50, 60:40, 70:30, 80:20, 90:10, and 95:5. GERUD v2.0 (Jones, 2005) was used to calculate exclusion probabilities for each population and to estimate the minimum number of sires and paternal skews for each litter. COLONY v2.0 (Jones and Wang, 2010) was also used to estimate the number of sires per litter using a maximum likelihood approach. For analyses in COLONY v2.0, the allelic dropout rate was set to 0 and the error rate to 0.02. The input parameters for COLONY analyses were Mating System I: ‘Female Polygamy’ and ‘Male Monogamy’; Mating System II: ‘Without Inbreeding’ and ‘Without Clone’; Species: ‘Dioecious’ and ‘Diploid’; Length of Run: ‘Medium’; Analysis Method: ‘Full-Likelihood’; Run Specifications: ‘No Update Allele Frequency’, ‘Yes Sibship Scaling’, ‘1 Number of Run’, and ‘1234 Random Number Seed’; and Sibship Prior: ‘No Prior’. Population allele frequencies were incorporated into all PrDM, GERUD, and COLONY analyses. Maternal genotypes were also incorporated into all these analyses, except for litter AU-8 (see below).
RESULTS
Mean total length (TL) ± SD of G. galeus collected off California (n = 10) was 175.8 ± 9.8 cm (range: 162–190 cm), and mean litter size ± SD was 30.9 ± 8.5 pups (range: 15–44 pups). Mean TL ± SD of G. galeus collected off Australia (n = 10) was 157.7 ± 8.2 cm (range: 141–168 cm), and mean litter size ± SD was 32.6 ± 7.8 pups (range: 24–51 pups). Maternal TL was significantly higher for California than for Australia (t-test; P < 0.001); however, there was no significant difference in litter size between the populations (t-test; P = 0.646). There was no significant relationship between maternal TL and litter size for California (R2 = 0.19, F1,8 = 1.9, ß = 0.38, P = 0.205) or Australia (R2 = 0.35, F1,8 = 4.4, ß = 0.56, P = 0.069; Table 1).
For the Structure analyses, every estimator of K yielded two populations, supporting the pooling of the California adults (including the mothers) into a single group (n = 90) and the Australia adults (including the mothers) into a single group (initially n = 50 but later revised to n = 49; see below), for the purpose of calculating allele frequencies. Mean allelic richness ± SD for the six microsatellite loci was 5.3 ± 2.5 (range: 3–10 alleles) for California and 6.5 ± 4.0 (range: 3–13 alleles) for Australia. Mean observed heterozygosity ± SD was 0.546 ± 0.154 (range: 0.378–0.800) for California and 0.534 ± 0.255 (range: 0.204–0.857) for Australia. Mean expected heterozygosity ± SD was 0.565 ± 0.156 (range: 0.425–0.843) for California and 0.613 ± 0.231 (range: 0.206–0.857) for Australia. All loci were found to be in Hardy-Weinberg equilibrium, except Gg18 in California (P = 0.006) and Gg23 in Australia (P = 0.001). However, these loci were retained because they were highly informative and the heterozygote deficiency was small (California: 42 observed vs. 42.5 expected heterozygotes for locus Gg18; Australia: 72 observed vs. 75.9 expected heterozygotes for locus Gg23). These data are summarized in Table 2.
For each litter, every pup was found to share at least one allele per locus with their mother, except for litter 8 from Australia (AU-8), for which the genotype of the mother could not be confidently established. Therefore, the putative mother of litter AU-8 was omitted from allele frequency calculations, reducing the number of unrelated adults sampled from the Australia population from 50 to 49. However, the pups from litter AU-8 were retained for estimating FMP because GERUD and COLONY have an option for estimating the number of sires even when the maternal genotype is not known.
The probability of detecting multiple paternity (PrDM) with two sires was high across all litters (except litter AU-8, because post hoc analysis of PrDM relies on knowing the maternal genotype) for low to moderate hypothetical paternal skews (Table 3): mean PrDM ± SD was 0.963 ± 0.020 for a skew of 50:50, 0.962 ± 0.021 for a skew of 60:40, 0.959 ± 0.024 for a skew of 70:30, and 0.945 ± 0.036 for a skew of 80:20. However, PrDM was lower for the highest hypothetical paternal skews: 0.862 ± 0.074 for a skew of 90:10 and 0.672 ± 0.010 for a skew of 95:5.
GERUD indicated that two of the ten California litters (CA-1 and CA-2) and four of the ten Australia litters (AU-5, AU-6, AU-7, and AU-8) exhibited MP (Table 1), with no more than two sires detected for each of these litters. Thus, the frequency of multiple paternity (FMP) was 20.0% for California and 40.0% for Australia, with no significant difference between the two populations (Fisher’s Exact Test, two-tailed P = 0.629). Two of these litters had paternal skews higher than 80:20, with the dominant male for litter AU-6 siring 94.6% of the pups and the dominant male for litter AU-7 siring 82.1% of the pups. There was no statistically significant difference between singly and multiply sired litters in maternal TL (two-sample t-test, two-tailed P = 0.476) or litter size (two-sample t-test, two-tailed P = 0.057). COLONY indicated that all 20 litters exhibited MP, with a mean number of sires ± SD of 5.2 ± 1.9 sires (range: 2–9 sires; see Supplemental Text and Supplemental Table 2; see Data Accessibility); however, given the heterozygosity of the microsatellite loci in each population, these were likely gross overestimates (see Discussion).
DISCUSSION
Our study investigated the frequency of multiple paternity (FMP) in G. galeus and found an FMP of 40.0% (4 of 10 litters had multiple sires) in the Australian population and 20.0% (2 of 10 litters had multiple sires) in the Californian population. This is one of the few studies of multiple paternity (MP) to compare FMP in different populations of the same species (see below). Our values of FMP are consistent with previous findings by Hernández et al. (2014), who reported a similar FMP (40.0%), but in a smaller sample size (5 litters) from New Zealand. By using a larger sample size (20 litters) across two populations, our study corroborates the notion that G. galeus exhibits a low to intermediate FMP compared to other elasmobranch species. The probability of detecting multiple paternity (PrDM; Table 3) in our study was high (>0.95) for paternal skews up to 80:20, but lower for the highest paternal skews simulated (90:10 and 95:5). One litter in which MP was detected (AU-6) demonstrated a high paternal skew (35:2; Table 1), which suggests one or more litters we identified as having a single sire might actually have multiple sires, meaning the true FMPs could be slightly higher than reported here. In short, our data support the hypothesis of low to intermediate FMP in G. galeus.
The FMPs we report are based on the sire numbers estimated by GERUD v2.0 and not COLONY v2.0. Compared to GERUD, COLONY suggested all 20 litters had MP, with 2–9 sires per litter (see Supplemental Text and Supplemental Table 2; see Data Accessibility). However, COLONY is known to overestimate the number of sires, and therefore FMP, especially when the heterozygosity of the markers used for genotyping is low (Sefc and Koblmüller, 2009). Thus, our FMPs are only based on sire estimates from GERUD, and, for accurate comparison to other studies, the FMPs discussed below were also based on GERUD. Additionally, FMP is only considered for comparison if it was estimated from at least five litters.
Multiple paternity has been widely documented in elasmobranch fishes, with high variation in FMP (Lyons et al., 2021). Relatively high FMPs have been reported for some species, such as S. canicula (92.3%; Griffiths et al., 2012) and Round Stingrays Urobatis halleri (90%; Lyons et al., 2017). Relatively low FMPs have been reported for S. mitsukurii (11.1%; Daly-Engel et al., 2010), Spotted Estuary Smoothhounds Mustelus lenticulatus (12.5%; Boomer et al., 2013), Spiny Dogfish Squalus acanthias (17.2%; Veríssimo et al., 2011), Bonnethead Sphyrna tiburo (18.2%; Chapman et al., 2004), and Gummy Shark Mustelus antarcticus (24.1%; Boomer et al., 2013). Lastly, intermediate FMPs have been reported for the Dusky Shark Carcharhinus obscurus (35.7%; Rossouw et al., 2016), Leopard Shark Triakis semifasciata (36.4%; Nosal et al., 2013), Common Smoothhound Mustelus mustelus (47.4%; Marino et al., 2015), Blacktip Shark Carcharhinus limbatus (50.0%; Bester-van der Merwe et al., 2019), Black Spotted Smoothhound Mustelus punctulatus (53.8%; Marino et al., 2015), and Starry Smoothhound Mustelus asterias (58.3%; Farrell et al., 2013).
Galeorhinus galeus falls in the low to intermediate range of FMP, given the values of 40.0% and 20.0% we report for litters sampled from Australia and California, respectively, and the value of 40.0% previously reported by Hernández et al. (2014) from New Zealand. We detected no significant difference in FMP between the two populations; however, such population differences in FMP have been documented for other species. For example, Portnoy et al. (2007) found an FMP of 85.0% for Sandbar Sharks Carcharhinus plumbeus in the Western North Atlantic, while Daly-Engel et al. (2006) found an FMP of only 40.0% for this species in the Central Pacific. Similarly, Scalloped Hammerhead Sphyrna lewini have an FMP of 100% off Papua New Guinea (Green et al., 2017) but an FMP of only 46.2% off South Africa (Rossouw et al., 2016). Perhaps the most striking difference between populations comes from M. henlei, which has an FMP of 92.9% along the Pacific coast of Baja California Sur, Mexico (Byrne and Avise, 2011) and 86.7% in the northern Gulf of California (Rendón-Herrera et al., 2022), but an FMP of only 22.2% around Santa Catalina Island (Chabot and Haggin, 2014).
Our hypothesis of a high FMP, particularly for G. galeus off California, was not supported. Nosal et al. (2021) found that mature females spend two years of their triennial reproductive cycle along the Central Coast of California, where they were expected to give birth and mate, before migrating to southern California for gestation. This was consistent with fishery data analyzed by Ripley (1946), who reported an approximately 1:1 sex ratio (46.5% female) for G. galeus off central California, compared to 98.7% females off southern California and 2.5% females off northern California. Given the apparent lack of sexual segregation where G. galeus is believed to mate, protracted mating season, and this species’ propensity to aggregate, we expected high levels of contact between the sexes and thus a high probability of multiple mating leading to MP. This would be especially true if convenience polyandry explained the prevalence of MP in elasmobranch fishes (but see Lyons et al., 2021). However, we found a relatively low FMP of 20.0% for California. One explanation for this is that, although there may be little geographical (horizontal) segregation of the sexes off central California, there may be substantial depth (vertical) segregation. At least for G. galeus captured around the Northern Channel Islands, Ripley (1946) reported that 80.8% of females were captured in depths less than 30 fathoms (54.9 m), compared to only 27.4% of males. Meanwhile, 49.9% of males were captured at depths of 41–70 fathoms (75.0–128.0 m), compared to only 11.0% of females, and 14.7% of males were captured at depths of 111–120 fathoms (203.0–219.5 m), compared to only 0.4% of females. Similar sexual segregation by depth is observed in Australia, with males generally preferring deeper areas than females except during the winter breeding season when mixing occurs on the edge of the continental shelf (Olsen, 1954). Thus, depth segregation may limit the amount of contact between the sexes and result in lower frequency of multiple mating, and therefore MP.
Lyons et al. (2021) noted that convenience polyandry is not the only possible explanation for MP in elasmobranch fishes. Alternative explanations, such as cryptic female choice, genetic bet-hedging, sperm competition, and trading up, have potential indirect benefits to females mating with multiple males. By extension, multiple mating does not necessarily lead to MP. It is possible that female G. galeus mate with multiple males, but subsequent sperm competition or cryptic female choice limit the fertilization success of some males’ sperm. The occasional high paternal skew in litters with multiple sires would also be consistent with sperm competition or cryptic female choice. Convenience polyandry and indirect female benefits are not mutually exclusive explanations. Interestingly, mean litter size ± SD was 34.0 ± 7.5 pups for singly sired litters and 26.5 ± 7.0 pups for multiply sired litters. Although this difference was not statistically significant (P = 0.057), it is worth noting that if MP somehow results in fewer pups per litter, any indirect benefits to the female, if they exist, would have to be substantial to outweigh the potential cost to the female’s fecundity.
Frequency of multiple paternity is highly variable among species (Lyons et al., 2021), and there are often too many differences in their biology and ecology to determine why some species have higher FMPs than others. However, comparing different populations of the same species reduces the number of confounding variables and may offer clues as to the factors influencing FMP. For example, the regional differences in FMP for M. henlei were largely attributed to differences in sex ratios during breeding season, with female-biased sex ratios correlated with relatively low FMP, lending support for convenience polyandry as an overall explanation for multiple mating in female elasmobranchs (Byrne and Avise, 2011; Chabot and Haggin, 2014; Rendón-Herrera et al., 2022). The importance of comparing multiple populations of a species is why we determined FMP for both Australia and California in G. galeus, and while the difference in FMP we report (40.0% for Australia and 20.0% for California) is not statistically significant, this could be due to the relatively small sample size of ten litters sampled from each population. Future work might attempt to sample additional litters from both populations, and from other genetically distinct populations, as well as from other species with multiple populations. To unravel the genetic drivers and heritability of MP and litter size, future studies could target nursery areas to sample adult females and neonates non-destructively across cohorts (e.g., Reid et al., 2011), which could theoretically be achieved in a close-kin mark–recapture study (see Thomson et al., 2020).
In conclusion, our study provides new insights into MP in G. galeus, corroborating previous findings of a low to intermediate FMP in this species. Additionally, we found that the FMP for the California population (20.0%) was half that of the Australia population (40.0%). These results underscore the importance of studying multiple populations within species to better understand the ecological and behavioral factors influencing MP.
DATA ACCESSIBILITY
The data generated and analyzed in this study are available by request of the authors. Supplemental material is available at https://www.ichthyologyandherpetology.org/i2025002. Unless an alternative copyright or statement noting that a figure is reprinted from a previous source is noted in a figure caption, the published images and illustrations in this article are licensed by the American Society of Ichthyologists and Herpetologists for use if the use includes a citation to the original source (American Society of Ichthyologists and Herpetologists, the DOI of the Ichthyology & Herpetology article, and any individual image credits listed in the figure caption) in accordance with the Creative Commons Attribution CC BY License.
AI STATEMENT
The authors declare that ChatGPT was used to refine and distill some paragraphs to improve flow and clarity, and to reduce word count.
Contributor Notes
Associate Editor: T. Grande.