Integrating field and laboratory evidence for environmental sex determination in the amphipod, Echinogammarus marinus
- First Online:
- Cite this article as:
- Guler, Y., Short, S., Kile, P. et al. Mar Biol (2012) 159: 2885. doi:10.1007/s00227-012-2042-2
- 537 Views
The causes of sex determination in amphipods are believed to be multi-factorial. Sex determination in Echinogammarus marinus (Leach, 1815) has been reported to be linked with feminising parasites. To date, however, no such studies have linked this species with environmental sex determination (ESD). A field study and laboratory breeding experiments were conducted to determine the influence of photoperiod on sex determination. Over the two-year field study, males dominated during August to November, whilst female-biased populations were observed during April to July. A significant linear relationship was observed between photoperiods and sex ratios from the field data. Under laboratory conditions, photoperiod was also shown to be an influential factor in sex determination, with a male bias over a long-day photo regime (61.5 % male broods) and a female biased over a short-day photoperiod regime (43.5 % male broods). Findings suggest that there is some level of ESD present within the species, suggesting considerable plasticity in the sex differentiation pathway.
There are multiple factors that are known to trigger or influence sex determination in Crustacea. Sex can be determined by environmental (Bulnheim 1978), parasitic (Mautner et al. 2007), as well as genetic factors (Legrand et al. 1987). Kato et al. (2011) recently revealed the role played by the highly conserved Doublesex gene in parthenogenic crustacean, Daphnia magna; however, molecular mechanisms for sex determination in crustaceans are still largely unknown with the only well-characterised sex determination pathway being that of the highly divergent Drosophila (Sanchez 2008). Sex-determining mechanisms drive a population’s sex ratio, which in turn affects the size of the reproducing population. Fisher’s principle of equal investment states that natural selection favours equal frequency of males and females (Fisher 1958). This evolutionary theory of a stable 1:1 sex ratio model has been generally favoured when producing males or females has similar costs (Fisher 1958; MacArthur 1965; May 1983). In crustaceans, however, it is rare to find a species with an unbiased sex ratio (Lasker et al. 1970; Maly 1970; Litulo 2005; Castiglioni and Buckup 2008; Doi et al. 2008; Ford and Glazier 2008; Prato et al. 2009; Saher and Qureshi 2011). This can be a consequence of a gender bias in the production of offspring, or mortality rates could be sexually differentiated, such as cases of sex-biased predation (Appadoo and Myers 2004).
Environmental sex determination (ESD) is defined as the determination of gender by environmental cues or stimuli during development (Korpelainen 1990). Environmental sex determination has been found in diverse groups of organisms including Echiura, reptiles, fish, nematodes and crustaceans (Petersen 1972; Bull 1980; Conover and Kynard 1981; Adams et al. 1987; Korpelainen 1990; Janzen and Paukstis 1991). Not all populations within a species necessarily possess ESD, and it is more prevalent in populations that have a limited breeding season (Watt and Adams 1994). This variation in ESD indicates the adaptive response of reproductive strategies under varying environmental conditions, allowing an individual to develop into the gender that provides the best ecological fitness at the time, given the environment they encounter (Naylor et al. 1988b; Watt and Adams 1994).
The adaptive benefit for ESD is usually size related, as size is an important factor in reproductive success in males and females (Naylor and Adams 1987). In amphipods, increased female fecundity correlates with increase in size. However, as males guard the females in pre-copulatory behaviour, they need to be larger than the females. To ensure a larger size, a longer period of growth is needed and employing an ESD system, where males are produced earlier in the year, giving a longer period for growth, maximises reproductive success (Naylor et al. 1988a; Watt and Adams 1994). This has been shown in a brackish water amphipod, Gammarus duebeni, a species exhibiting seasonal sex-biased ratios, favouring males in the summer and that shift to female bias in the autumn (Naylor et al. 1988b). Sex determination can be highly influenced by photoperiod; however, the ESD response by G. duebeni in the laboratory did not correlate with what was occurring in the natural populations (Watt and Adams 1994). Temperature was then examined by Dunn et al. (2005) as a secondary cue comparing four geographically different populations. Dunn et al. (2005) suggested there may be an interaction of temperature and day length as cues for ESD as well as variation of degrees of ESD between the different populations, thus demonstrating adaptive variation within the species.
Echinogammarus marinus (Leach, 1815) is an intertidal amphipod, which is highly abundant and has a wide distribution range from polar regions down to southern Portugal (Lincoln 1979). The species has an important role in ecosystem dynamics, mainly as a food resource for upper trophic levels, which can in turn influence the structure and functionality of aquatic communities (Múrias et al. 1996; Duffy and Hay 2000). It is well-documented that E. marinus presents a range of sexual phenotypes (Ford et al. 2005). Intersexuality and female bias have been linked with vertically transmitting parasites that are believed to feminise male embryo hosts as a reproductive strategy to facilitate parasite transmission to the next generation (Kelly et al. 2004; Yang et al. 2011; Short et al. 2012). However, other factors known to influence crustacean sex determination pathways, such as ESD, have yet to be explored in E. marinus. Past studies have reported E. marinus populations, showing temporal and geographical variability in their sex ratios (Vlasblom 1969; Martins et al. 2009; Yang et al. 2011); however, no current published data have shown whether ESD is present within this species. To address this issue, this study investigated the effect of photoperiod on the sex ratio of broods in E. marinus. In addition, these results were compared with a two-year field study from a population of E. marinus without known feminising parasites (Yang et al. 2011).
Animals were collected from Langstone Harbour, situated in Portsmouth, southern England, UK (50°47′23.13N 1°02′37.25W), and a total of 60 pre-copular pairs were placed in 150-ml pots that contained filtered seawater with fucoid seaweed (Ascophyllum nodosum) for food. Both food and seawater were changed approximately every 4–5 days. In other gammarid species, ESD is influenced 3–4 weeks after release from the mother’s brood pouch (Bulnheim 1978; Naylor et al. 1988b). However, no assumption was made at what point environmental cues might influence sex determination (either as a zygote or juvenile stage). Therefore, pre-copular pairs were assigned at random to one of the photoperiod regimes: 16-h light and 8-h dark or 8-h light and 16-h dark (30 broods per light regime at 15°C) to mimic the extremes of long-day and short-day conditions in the field. This ensured that the broods were exposed to the chosen photoperiod at all developmental stages. Once juveniles left the female brood pouch (approximately 30 days after egg release), all adults were removed to avoid cannibalism. Despite the microsporidia infecting this population, Dictyocoela berillonum, being a non-feminiser, all brooding females were screened for microsporidian parasites using PCR (Yang et al. 2011) and broods were eliminated if the mother tested positive (n = 11/60). Echinogammarus marinus sex ratios were determined when sex could be distinguished morphologically after approximately five months.
To assess sex ratios in the field, E. marinus were collected over a two-year period between December 2009 and December 2011 from Langstone Harbour (Portsmouth, UK). Hours of daylight at this latitude vary between approximately 8 and 16 h throughout the year. Samples were taken by selecting five 1-m2 quadrats (total area = 5 m2) in the intertidal zone during low tide. All algae and surface sediment (approximately 2 cm in depth) were retrieved and stored in polythene bags. In the laboratory, samples were washed and decanted through a 0.7-mm sieve, and all algae were scraped to ensure no individuals were left. All amphipods were collected and stored in 70 % ethanol where E. marinus specimens were separated into males, females and juveniles. Generally, sex could be determined within individuals that were approximately over 10 mm in length. E. marinus males were distinguished by the presence of enlarged gnathopods and genital papillae, whereas females were distinguished by much smaller gnathopods and oostegites (brood plates). Individuals not presenting any of these features were grouped as juveniles.
A significant difference in the sex ratios between the two light regimes was observed (Mann–Whitney test; p < 0.001) in the laboratory breeding experiments with a male bias recorded over a long-day photo regime (mean = 61.5 ± 0.84 % male broods; n = 16) and a female biased over a short-day photoperiod regime (mean = 43.5 ± 0.94 % male broods; n = 12). No significant difference in the mean number of eggs per female was observed for short- and long-day regimes (27.5 ± 0.9 and 26.6 ± 0.8, respectively), when normalised to size (ANCOVA; p > 0.05). Furthermore, there was no significant difference (chi-square p > 0.05) in the brood survival between the long-day (67 %) and short-day regimes (69 %).
Sex determination and other reproductive processes still remain largely unknown in many aquatic invertebrate species. Amphipods are extensively used in ecotoxicology studies with an increasing emphasis on reproductive endpoints (Hyne 2011). However, to truly understand whether anthropogenic influence is currently an issue, it is critical that all the mechanisms governing reproductive processes are fully evaluated. Ecologists have criticised the lack of basic knowledge in the biology of well-studied fauna, as well as highlighted an extreme bias in our knowledge towards vertebrates (Tyler et al. 2012). Environmental conditions can be influential factors in reproduction and development, and a better understanding of how these altered conditions affect reproductive biology in organisms is required. Such an understanding will be crucial in our ability to model and predict population levels in a changeable environment (Visser et al. 2004; Martins et al. 2009).
The aim of this study was to demonstrate whether E. marinus displays ESD under laboratory conditions and whether this correlates with sex ratios detected in the field. Photoperiod was shown to be an influential factor in sex determination, and a significant correlation was observed between sex ratios detected in the laboratory and field. Despite this, the regression model only accounted for 56 % of the variation in the data set, which suggests that other environmental factors (e.g. temperature) and interindividual variation may also be involved. The range of sex ratios from the field study displayed large variation, with extremes of 36 % males (Mar 2011) and 71 % males (October 2010), strongly suggesting adaptive sex ratio variation over the 1:1 Mendelian sex ratio. These swings in sex ratio were mirrored over the two-year study, with the October months possessing the highest proportions of males and January through to March having the lowest. Interestingly, when comparing the total number of males and females over the two-year study, the Portsmouth population displayed an overall 1:1 sex ratio (50.3 % males). The two-year field study does highlight a seasonal gender bias; however, overall the population does not produce more of one gender over the course of two years. The laboratory experiments resulted in a male bias over long-day and female bias over the short-day photoperiod. These findings are consistent with those of an estuarine temperate species, Gammarus duebeni, that showed females (taken from in southern sites within the UK) produced male-biased broods under long-day laboratory conditions (Dunn et al. 2005). Although laboratory and field data correlate well and support photoperiod as an ESD cue, a second cue (e.g. temperature) for ESD has been inferred in other aquatic species (Baron et al. 2002; Dunn et al. 2005) and should also be considered for E. marinus.
The E. marinus population used in this study has a continuous reproductive output and breeds throughout the year, producing male and female bias seasonally. This is despite the prediction that if a breeding season is unrestricted and there is a generations overlap, ESD is no longer advantageous and will revert to a genetic system where males and females are produced simultaneously (Naylor et al. 1988b). Given the apparent costs of using ESD, such as intersexuality and inconsistent environmental conditions, the benefits for this population are not obvious. This could suggest that ESD in this population is ancestral and, whether advantageous or not, has been retained.
The E. marinus population used in this study enabled the reliable detection of ESD because it is uninfluenced by known feminising parasites (Yang et al. 2011). However, the variable presence of such parasites makes direct comparisons of sex ratios between populations difficult. Amongst E. marinus populations so far studied, female bias is common. Vlasbloom (1969) observed female bias in E. marinus populations from the Netherlands, with an approximate average of 40 % males. In addition, although a female bias was mainly present, sex ratio fluctuations similar to those seen in this study were observed but with male increase occurring two months earlier. In comparison, southern latitudinal populations (Mondego estuary, Portugal) have shown female bias in autumn and winter, with a reversal in the spring and summer months. However, further south in the estuary, male bias was observed during winter, with female bias occurring during part of the summer and autumn (Maranhao et al. 2001), similar to that found in this study. The cause of this variation is unknown but could be due to environmental conditions specific to the latitudinal positions, as well as other sex-determining factors, such as the already mentioned feminising parasites. Within the UK, well-documented E. marinus populations have high female bias that is clearly correlated with parasite infection (Ford et al. 2006; Short et al. 2012), and although these populations also present fluctuating sex ratios, it is problematic to separate the influence of photoperiod and parasites. Although it is clear that E. marinus sex ratios can vary considerably, reasonable comparisons of environmentally induced sex ratio fluctuations in E. marinus will require detailed surveys of other populations not influenced by feminising parasites.
In addition to being linked to parasite infection, amphipod intersexuality has been associated with populations that possess ESD (Dunn et al. 1996). Echinogammarus marinus populations consistently present a fraction of intersex individuals that are not infected with feminising parasites (Yang et al. 2011; Short et al. 2012). Indeed, the population used for this study has no known parasitic feminiser, yet presents notable levels of intersex (Yang et al. 2011). The non-parasite-induced intersex phenotypes could be the result of an underlying ESD mechanism that has been disrupted due to intermediate environmental signals (Dunn et al. 1993). The mechanism of ESD could also increase the susceptibility of the host to parasite-induced feminisation. Feminising parasites have been shown to have higher prevalence in populations with high levels of ESD. It has been suggested that the delay in sex determination resulting from the ESD pathway causes host vulnerability to the manipulation by feminising parasites, as the parasites are more easily able to override an ESD pathway than they are a genetic-based system (Dunn et al. 1996).
Echinogammarus marinus has a large geographical range from approximately 39oN, where average daylight hours can range from 9 to 15 h throughout the year, to 65oN, where 24 h of daylight occurs at certain times of the year. This is important when considering the potential implications of climate change. There is evidence suggesting that species distribution shifts occur as a result of a changing climate (Parmesan and Yohe 2003). Species that have a sex determination pathway influenced by photoperiod may well be latitudinally constrained. As a result, E. marinus populations may be forced to adapt to increased temperatures or altered photoperiods. Given the ecological importance of the species, a better knowledge of sex-determining factors, in particular ESD, will be required if we are to fully understand the impact of a changing environment.
YG and SS are supported by the Natural Environment Research Council (UK) grant (NE/G004587/1) awarded to PK and ATF. We are appreciative to Amaia Etxabe Green and Joanna Murry for their assistance with field work and figure preparation, respectively.