Published On: Wed, Mar 27th, 2019

Sperm-duct gland content increases sperm velocity in the sand goby [RESEARCH ARTICLE]


No difference was found in the percent of motile sperm [generalized linear mixed effects model, χ2 (1)=0.1374, P=0.7109], detection threshold [generalized linear mixed effects model, χ2 (1)=0.8671, P=0.3518], or number of tracked sperm [generalized linear mixed effects model, χ2 (1)=2.5473, P=0.1105], which supports the sampling methodology. To control for a potential effect of sperm numbers on velocity, a general linear model analysis of covariance was performed with VCL as dependent variable, treatment as factor and percent motile sperm as covariate. Non-significant interactions (P0.05) were deleted from the model. Despite heterogeneous variances, the viability data were analysed with a repeated measures general linear model with both time and treatment as repeated measures for each individual replicate.

We found that the treatment with content from the sperm-duct glands significantly increased the velocity of sperm in the sand goby [tested with SDG content mean±s.e.m. VCL: 75.38±2.07 μm s−1; tested without SDG content mean±s.e.m. VCL: 70.26±1.70 μm s−1; generalized linear mixed effects model, χ2 (1)=5.475, P=0.019, Fig. 1A]. Sperm tested with SDG content in the seawater showed an average increase in velocity by 5.12 μm s−1 (7.3%) compared to sperm tested without SDG content. Sperm velocity was still significantly different between treatments when controlling for the percentage of motile sperm [general linear model, treatment (factor): F1,106=5.125, P=0.026; percentage of motile sperm (covariate): F1,106=40.866, P0.001] (Fig. 1B). The percentage of motile sperm was unaffected by treatment [generalized linear mixed effects model, χ2(1)=0.1374, P=0.7109].

Fig. 1.

Fig. 1.

Effects from SDG treatment on sperm traits. Symbols show sperm from 10 males tested without (sperm only, blue dots) or with SDG content (sperm with SDG content, green triangles Δ) in seawater (see Materials and Methods for further details). Different letters show statistically significant differences (see Results and Discussion for all test values). Horizontal bars show means. (A) Sperm velocity increases in SDG content [generalized linear mixed effects model, χ2 (1)=5.475, P=0.019] shown as curvilinear velocity (VCL μm s−1). Each data point is the average of six technical replicates (n=10). (B) The percentage of motile sperm affects sperm velocity [general linear model, percentage of motile sperm (covariate): F1,106=40.866, P0.001]. However, sperm velocity (VCL μm s−1) was still significantly higher for sperm with SDG content when controlling for percentage of motile sperm [general linear model, treatment (factor): F1,106=5.125, P=0.026], including all technical replicates (n=60). (C) The proportion of live sperm dropped slightly after 24 h compared to right after sampling (10 min), but the drop was similar in the two treatments [general linear model, time (repeated): F1,9=8.92, P=0.015] (n=10).

Comparing the proportion of sperm that were alive after 10 min and 24 h, time had a significant effect on the viability of the sperm, but no statistically significant effect of treatment or interaction between time and treatment was found [general linear model, time (repeated): F1,9=8.92, P=0.015; treatment (repeated): F1,9=0.10, P=0.76; time*treatment: F1,9=0.03, P=0.87] (Fig. 1C). In both treatments, sand goby sperm showed a viability of over 86% of the sperm still alive after 24 h exposure to seawater.

Sperm velocity is directly linked to increased fertilization success in many taxa (Gage et al., 2004; Snook, 2005; Gasparini et al., 2010), and an increase in velocity of 5% can increase the relative fertilization success of a male by 5–6% (5% for an external fertilizer: from fig. 4C in Gallego et al., 2013; 6% for an internal fertilizer: from fig. 1 in Gasparini et al., 2010, value obtained through manually scaled measurements using GraphClick from Arizona Software). Such improved fertilization success is likely to have an important effect on male fitness, whether reproduction occurs under sperm competition or sperm limitation, both of which are common among externally fertilizing fish (including gobies) (Levitan, 1998; Petersen and Warner, 1998). Therefore, the increase in sperm velocity by 7.3% found in our study is expected to have a fitness effect through the males’ fertilization success.

Viability is typically considered less important than velocity since fertilization during spawning in most species takes place within seconds of ejaculation (Cosson et al., 2008). Our results show that sand gobies are able to ‘boost’ sperm swimming speed through accessory substances without any measurable negative effect on viability. Levitan (1998, 2000) has suggested that long sperm viability would evolve as an adaptation to sperm limitation, whereas fast but short-lived sperm would evolve under sperm competition. Since sperm competition is well documented in sand gobies, the current result does not fit this picture. Instead, we suggest that the long lifespan of sand goby sperm may be an adaptation to ensure continued sperm function when egg deposition lasts one to several hours (Svensson and Kvarnemo, 2005). Considering that two to six females typically spawn sequentially in one nest (Jones et al., 2001a), the whole session can last even longer. Though the time window for fertilization of sand goby eggs has not been tested experimentally, other species’ eggs remain fertilizable for several hours, enabling competition from parasitically spawning males, as found in grass goby (Zosterisessor ophiocephalus, Pallas 1814) and their close relative, the black goby (Gobius niger, Linnaeus 1758) (Mazzoldi, 1999; Scaggiante et al., 1999; Pilastro et al., 2002; Locatello et al., 2007).

The otherwise short lifespan of fish sperm in general has been attributed to an inability to handle osmotic change when ejaculated into an environment of different osmolality (Alavi and Cosson, 2006). However, sand goby sperm tolerate a wide range of osmolality and within the geographic range of the species, different populations are able to spawn successfully in salinities ranging from 35–3 practical salinity units (PSU) (Fonds and Van Buurt, 1974; Svensson et al., 2017). Hence, a general tolerance of an osmotically challenging environment may potentially explain the unusually long lifespan of the sperm.

Alternative reproductive tactics were not a focus of this study; yet, their role for SDG evolution makes them relevant to discuss in this context. As expected from theory (Parker, 1990; Parker et al., 1997), sneaker males of gobiid species typically have very large testes compared to nest-holding males, but they also have small SDGs (Kvarnemo et al., 2010; Locatello et al., 2013). In the sand goby, a distinct ‘sneaker morph’ is present. These males lack breeding colour and have testes three to four times the size of breeding coloured males, but smaller SDGs (approximately one fourth the size of breeding coloured males; Kvarnemo et al., 2010). Similar patterns are found in e.g. black goby (Rasotto and Mazzoldi, 2002) and grass goby (Scaggiante et al., 1999). In the black goby, sperm of sneaker males survive better over time, swim faster and have more adenosine triphosphate content than the sperm of nest-holding males, as tested without the aid of SDG content (Locatello et al., 2007; Poli et al., 2018). In the grass goby, while SDG function of nest-holders is mainly to produce mucus, in sneaker males it is primarily a sperm storage organ (Scaggiante et al., 1999). Nevertheless, both nest-holding and sneaker males are able to produce sperm trails (Mazzoldi et al., 2000). Sneaker males produce white sperm trails with many times higher sperm concentration than the opaque trails of nest-holding males (Mazzoldi et al., 2000). Furthermore, sperm velocity and fertilization success increase when the sneaker male sperm are exposed to seminal fluid from nest-holders (here: stripped fluid from testes and SDGs, with sperm removed), whereas for nest-holders the opposite is true (Locatello et al., 2013). In black gobies, however, seminal fluid increases sperm velocity of nest-holders, but not of sneaker males (Poli et al., 2018). In some populations of the sand goby, sneaker-morph males represent 10% of all males (Kvarnemo et al., 2010). However, parasitic spawnings also occur by other nest-holding males (Singer et al., 2006). Studies have shown close to 50% of broods to be partly fertilized by a male other than the nest-holder, independent of nest site availability (Jones et al., 2001b). Consequently, sperm competition is common in this species. Sneakers can change into nest-holders and develop breeding colouration (Takegaki et al., 2012). During this change, the SDG size increases, while testes size does not change, pointing to the importance of SDGs for the nest-holding reproductive tactic (Takegaki et al., 2012). Similar results from black goby indicate that plasticity in alternative reproductive tactics could be common among gobies with similar reproductive systems (Immler et al., 2004).

Our results are thus mirrored in grass and black goby, where SDG content also has a positive effect on sperm performance (Locatello et al., 2013; Poli et al., 2018). Since the genera Zosterisessor and Gobius, which are closely related, and Pomatoschistus belong to two distinct lineages (Agorreta et al., 2013), our results show support of a preserved effect of SDG content on sperm velocity in Gobiidae (Fig. S1). The SDG adaptations in Gobiidae possibly have an even older origin, as the sister families Butidae, Eleotridae and Odontobutidae also have SDG structures (Fishelson, 1991). Gobiidae is the second most species-rich vertebrate family known, and the most species-rich marine vertebrate family, with around 2000 described species, and 10 new species or more reported close to every year (Patzner et al., 2011). Their successful diversification and adaptation to spawning in fresh, brackish and marine water, in burrows and anadromously (Adrian-Kalchhauser et al., 2017), together with their potential as invasive species (Wonham et al., 2000) points to their ability to adapt into reproducing in novel environments. With SDGs being a conserved organ in gobies (Miller, 1984; Fishelson, 1991; Fig. S1), the ability to influence sperm function in the fertilization micro-environment is of interest for future research, in particular as this factor may contribute to their ability to spread into a range of environments.

Another example of a fish that modulates the direct environment of its spermatozoa is the three-spined stickleback (Gasterosteus aculeatus, Linnaeus 1758), which has ovarian fluid that enables its sperm to function in a range of salinities (Elofsson et al., 2003, 2006). Presumably, this function of the ovarian fluid has helped this species of stickleback to repeatedly colonize freshwater (Elofsson et al., 2003). Ovarian fluid has been shown to affect sperm function in several fish families (summarized in Elofsson et al., 2006), but to our knowledge, this is still uninvestigated in gobies. In gobies, eggs are attached to the substrate one-by-one (by an attachment network formed by a layer of filaments; Miller, 1984: Kramer and Patzner 2008), and the eggs appear ‘clean’. At this point, it is unknown if ovarian fluid might influence fertilization in gobies, alone or in combination with the SDG content studied here.

Gobies and their reproduction are studied as model organisms of sexual selection and evolutionary ecology (Locatello et al., 2007, 2013; Patzner et al., 2011; Svensson et al., 2017), and our results contribute to this growing body of literature. In conclusion, our study demonstrated that SDG content positively influences sperm velocity in the sand goby without affecting sperm viability. Whether or not the adaptation to alter the micro-environment of the sperm is widespread in the Gobiidae family, and how the trait is linked to their reproductive success, is still in need of investigation.

About the Author

Leave a comment

XHTML: You can use these html tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>