Gonad development and spawning of the Vulnerable commercial sea cucumber, Stichopus herrmanni, in the southern Great Barrier Reef

Abstract

Despite the important ecological roles of commercial bêche-de-mer holothuroids in coral reef ecosystems their reproductive biology is poorly studied, including on the Great Barrier Reef (GBR). We investigated reproduction of Stichopus herrmanni, a commercially important species listed as Vulnerable, at One Tree Island, southern GBR. Gonad index, histology and spawning observations indicated an annual reproductive cycle with gamete release in the Austral spring and summer (November–February), as for populations of this species at a similar latitude in New Caledonia. Stichopus herrmanni releases gametes episodically, spawning multiple times during summer. Assimilation of spawning observations from OTI and elsewhere along the GBR and tropical Pacific revealed that gamete release by S. herrmanni is influenced by the lunar cycle, with spawning taking place around the new moon in summer. This species is an aggregative spawner with a behavioural change to attain elevated positions on the reef at dusk prior to spawning. After the spawning season, gametes remaining in the gonads are reabsorbed. Spent gonads completely lacked gametes. There was a quiescence in gonad development in winter with an absence of gonads in some specimens, indicating an aestivation-like period for reproduction. By late-winter (August) recovery stage gonads were distinguished by the initiation of gametogenesis, which coincided with increasing temperature and day length. Our findings contribute to the understanding of the reproductive biology of S. herrmanni, a consideration for future fisheries management in the protection of this Vulnerable species, especially with respect to the increasing global trade in bêche-de-mer.

INTRODUCTION

Aspidochirotid holothuroids are a conspicuous component of the macro-benthos of marine environments. As deposit feeders, they process vast quantities of sediment, and thus play an important role in the mineralization and cycling of nutrients in benthic habitats (Uthicke & Klumpp, 1998; Uthicke, 1999, 2001; Purcell et al., 2016; Lee et al., 2017). The release of nitrogenous waste by aspidochirotids can increase the productivity of benthic microalgae and seagrass systems, a particularly important functional role in oligotrophic coral reef ecosystems (Uthicke & Klumpp, 1998; Uthicke, 2001; Eriksson et al., 2010; Wolkenhauer et al., 2010; Purcell et al., 2016; Wolfe & Byrne, 2017a, b). The digestion and dissolution of carbonate sands by tropical holothuroids can also influence local biogeochemistry by increasing local alkalinity, potentially buffering against ocean acidification and enhancing reef resilience (Hammond, 1981; Schneider et al., 2011, 2013; Wolfe et al., 2018).

Many holothuroid species are harvested for the lucrative dried seafood trade, with the dried body wall product (bêche-de-mer) highly prized in the Asian market (Purcell et al., 2013, 2016; Eriksson & Byrne, 2015; Eriksson & Clarke, 2015). Many commercial species are in a perilous state of conservation with 16 species, largely from tropical regions, recently listed as Threatened with extinction by the International Union for Conservation of Nature (IUCN) (Purcell et al., 2013, 2014; Conand et al., 2014). At least 70% of the world's tropical holothuroid fisheries are considered exploited, over-exploited or depleted (Purcell et al., 2013), with many species now locally extinct (Hasan, 2005; Anderson et al., 2011; Branch et al., 2013; Price et al., 2013; Purcell et al., 2014).

Despite the important ecological roles of commercial holothuroids in coral reef ecosystems and their commercial value, we have a limited understanding of their reproductive cycles and spawning periodicity. These animals are iteroparous and investigation of 11 Indo-Pacific species including some of the most commercially valuable (Holothuria fuscogilva, Actinopyga mauritiana, Thelenota ananas) indicates that they spawn annually during summer (Conand, 1981, 1993a, b; Conand et al., 2002; Ramofafia et al., 2003). Spawning during winter is also documented including for H. whitmaei (Conand, 1981, 1993a, b; Ramofafia et al., 2000, 2001; Shiell & Uthicke, 2006; Asha & Muthiah, 2008). While less common, biannual spawning has been observed for H. atra in New Caledonia (Conand, 1993a, b). The timing of spawning can differ among geographically separated conspecific holothurian populations (Ramofafia et al., 2003). Holothuria scabra spawns year round in equatorial regions (Ramofafia et al., 2003), but spawns annually in higher latitude regions (Conand, 1993a, b). These differences in spawning within a species are largely restricted to populations separated by latitudinal (north-south) oriented differences (Shiell & Uthicke, 2006). Due to the threatened status of commercial sea cucumbers, globally, we need a better understanding of their reproductive biology as a key fishery biology trait.

The curryfish, Stichopus herrmanni, is an abundant Indo-Pacific sea cucumber distributed from East Africa to Australia and Indonesia, which inhabits sandy lagoons, seagrass and coral reef habitats between depths of 0–30 m (Conand, 1993a, b; Eriksson et al., 2013; Wolfe & Byrne, 2017a, b). Stichopus herrmanni, a mid- to low-value bêche-de-mer species, has become a major harvest target as populations of higher value species continue to decline (Purcell et al., 2013, 2014; Eriksson & Byrne, 2015). Between 2007 and 2011, curryfish catches along the Great Barrier Reef (GBR) Marine Park, increased at an average annual rate of 200% (Eriksson & Byrne, 2015). Overall fishing pressure has resulted in a 60–90% population decline over half of the species' global range (Conand et al., 2014). Stichopus herrmanni is listed as Vulnerable to extinction by the IUCN (Purcell et al., 2014; Eriksson & Byrne, 2015).

We investigated the reproductive biology of S. herrmanni at One Tree Island (OTI), southern GBR. Our current understanding of the reproduction of S. herrmanni is based on data obtained from a population in New Caledonia, where spawning over 5 years occurred in summer (Conand, 1993a, b). Thus far, the reproductive cycle of only two bêche-de-mer species has been determined for populations on the GBR; H. whitmaei spawns in winter and H. atra spawns in summer (Shiell & Uthicke, 2006; Lee et al., 2008; Thorne et al., 2012). We determined the timing of gonad development and gamete maturation for S. herrmanni using gonad histology and the gonad index method. We also compiled observations of spawning by S. herrmanni in situ at OTI and elsewhere along the GBR and the Indo-Pacific. We hypothesized that reproduction of this species at OTI would be similar to that observed in New Caledonia (Conand, 1993a, b), as OTI and New Caledonia exist at similar latitudes. Our goal was to identify the spawning period of S. herrmanni at OTI and where possible, elsewhere on the GBR, providing important new information for the curryfish fishery to facilitate development of sustainable fisheries practices for this key bêche-de-mer species.

MATERIALS AND METHODS

Stichopus herrmanni (120–600 mm length) were collected from the lagoon (2–4 m depth) at One Tree Island (OTI) (23°30′S 152°05′E), southern GBR. Due to permit restrictions (GBR Marine Park Authority Permit G13/36027.1.) on the total number of sea cucumber species allowed to be collected annually we used data from collections over many years (Jan 2009, 2016, Feb 2011, 2013, Apr 2012, May 2009, Jul 2009, Aug 2015, Sep 2009, Oct 2013, Nov 2008, Dec 2009). We combined data per collection month across years. To determine the gonad index, S. herrmanni were dissected and the coelomic fluid drained. The gonad index was calculated as the weight % of the gonads relative to the combined drained body wall and viscera weights. Averages were taken per month. The seasonal temperature cycle of the lagoon was determined from a long-term monitoring programme (http://data.aims.gov.au/aiemsrtds/station.html?station=131). Average daily temperatures in OTI lagoon were combined to calculate monthly averages between 2008 and 2017 (N = 66–99 month⁻¹).

A portion of the gonad was fixed in Bouin's fluid and processed for routine wax histology. The gonad sections (7 µm thick) were stained in haematoxylin and eosin, and gonad histological condition was determined by microscopic examination. The gametogenic state of the gonads were scored in four stages: mature, partly spawned, post-spawned/spent and recovering, as in previous studies of sea cucumber gonad histology (Ramofafia et al., 2001, 2003). Digital images of gonad sections were captured using an Olympus DP70 digital camera mounted onto an Olympus BX50 microscope. Maximum egg size in mature ovaries and the thickness of the gonad wall of testes and ovaries were determined using Image-J (NIH, Bethesda, MD, USA).

Observations of spawning of S. herrmanni with information of the date and time of gamete release were assimilated from personal and published observations collected at OTI, elsewhere along the GBR and at other locations in the Pacific (see Table 1). Data were assessed with respect to the lunar cycle. Based on the pattern that emerged (see Results), we undertook a targeted survey for spawning behaviour at OTI around the new moon in January 2017. On three days (28–30 January, 17:30–18:30 h) coinciding with the new moon (NM), NM + 1 day and NM + 2 days, the number of spawning individuals were counted in the first 18–20 individuals randomly encountered along two transects (10–20 m) along the reef edge at Shark Alley (0.5–3 m depth), OTI, on snorkel.

Table 1. Observations of Stichopus herrmanni spawning on the Great Barrier Reef between 1987 and 2016, with indication of lunar cycle (NM = new moon). Data taken from: 1. Uthicke (1994); 2. Conand (1989); 3. Desurmont (2008).

To determine whether the gonad index differed among months, these data were compared by one-way ANOVA, with month as the fixed factor. Percentage data were arcsine transformed before analysis using JMP 501 (Cary, NC, USA). As required for ANOVA, homogeneity of variance and normality was checked and confirmed for all data series (Quinn & Keough, 2003). Post-hoc Tukey's HSD tests were used to determine where significant differences lay.

RESULTS

Ovary histology

Mature ovaries Stichopus herrmanni had late stage oocytes (mean diameter: 94.3 ± 0.9 µm; N = 420) filling the lumen, each located in an individual follicle (Figure 1A, B). During the mature stage the ovary wall was at its minimal thickness (14.3 ± 1.5 µm; N = 120). Partly spawned ovaries were marked by loosely packed unspawned oocytes and the presence of aggregations of phagocytes in the lumen (Figure 1C, D). Overlapping generations of oocytes were present in mature and partly spawned ovaries with a renewal of gametogenesis evident along the germinal epithelium (Figure 1C, D). Fully mature and partly spawned gonads were observed in late spring and summer (Nov–Feb). The largest oocytes had a mean diameter of 101.1 ± 2.8 µm (N = 120) (Figure 1B).

Fig. 1. Ovary histology of Stichopus herrmanni. (A, B) Mature ovary with the lumen filled with late stage oocytes within individual follicles (F). Note the thin gonad wall (arrow); (C, D) partly spawned ovary with oocytes scattered in the lumen (L) and early developing oocytes along the germinal epithelium (arrow); (E) spent ovaries with a few remaining oocytes and a thick wall; (F) recovering ovary with a new cohort of developing oocytes along the germinal epithelium (arrows) (scale bars = 200 µm).

Spent ovaries of S. herrmanni were largely empty and had a thick gonad wall (27.7 ± 2.0 µm; N = 120) (Figure 1E). Variable numbers of relict oocytes and phagocytes were present. It appears that the unspawned eggs were broken down and reabsorbed. Recovery stage ovaries had basophilic previtellogenic oocytes developing along the germinal epithelium (Figure 1F). As the oocytes developed they became increasingly eosinophilic, indicating vitellogenesis.

Testis histology

Mature testes were packed with basophilic spermatozoa (Figure 2A, B). Infolds of the germinal epithelium were absent and the wall of the testis was at its thinnest (7.0 ± 0.8 µm; N = 120). Partly spawned testes had fewer sperm in the lumen and an increase in the thickness of the gonad wall (Figure 2C, D). There was evidence of renewed spermatogenesis in partly spawned testes with developing sperm present along the re-appeared folds of the germinal epithelium. Spent testes were devoid of contents, with the exception of a few unspawned spermatozoa (Figure 2E) and the gonad wall was thick (24.1 ± 1.7 µm; N = 120) and had a wrinkled appearance. In recovery stage testes, developing sperm lined the germinal epithelium. The layer of developing sperm increased as the testis developed with invaginations of the germinal epithelium protruding towards the centre of the tubule, increasing the surface area of the germinal epithelium (Figure 2F).

Fig. 2. Testis histology of Stichopus herrmanni. (A, B) Mature testis with the lumen filled with sperm; (C) partly spawned testis with sperm scattered in the lumen; (D, E) spent testis with an empty lumen. The folds of the germinal epithelium are evident (arrows) as are aggregations of phagocytes (P); (F) recovering testis with a new cohort of developing sperm along the folds of the germinal epithelium (scale bars = 200 µm).

Seasonal trends in histology and the gonad index

Gametogenesis of S. herrmanni showed a seasonal pattern (Figure 3). In both sexes, mature gametes were observed from late-spring (Nov, 25%) through summer (Dec, 80%; Jan, 5.6%; Feb, 64.7%). In summer, most S. herrmanni were mature (Figure 3). Partly spawned individuals were also present in summer, especially in January (77.8%). Almost all S. herrmanni observed in autumn (April–May) and winter (July–Aug) had spent gonads. Recovery stage gonads were evident from August (55%) through to November (75%).

Fig. 3. Histological condition of the gonads of Stichopus herrmanni (sample size in parentheses) across the seasons.

OTI lagoon has a distinct seasonal temperature cycle, with daily averages ranging from 25.2–29.4°C in the summer months (mean: Dec = 26.7°C, Jan = 27.3°C, Feb = 27.2°C), to 19.4–23.9°C in the winter months (mean: Jun = 22.3°C, Jul = 21.2°C, Aug = 21.2°C) (Figure 4). The annual mean daily temperature was 24.4°C (± 0.7), with an average range of 19.4–29.4°C across the year. The gonad index of S. herrmanni differed between months (F _8,59 = 10.16, P < 0.0001), in parallel with water temperature (Figure 4). Tukey's HSD tests revealed that the average gonad index was highest in January (8.26 ± 1.4%) and February (3.65 ± 0.5%) (Figure 4). Some S. herrmanni lacked identifiable gonads in the winter.

Fig. 4. Annual temperatures at One Tree Island lagoon (2008–2017) and gonad index for Stichopus herrmanni (sample size in parentheses).

Spawning observations

Records from in situ observations (N = 13) show that Stichopus herrmanni spawns from late spring (Nov) through summer (Dec–Feb) near sunset (1730–1830) (Figure 5), typically around the new moon at several sites along the GBR from the southern (One Tree Island, Heron Island), central (Little Broadhurst Reef, Davies Reef), northern (Lizard Island) GBR, and also in New Caledonia (Table 1). One spawning observation was made ~1 week before a new moon event in December 2016 (Table 1). As is typical of aspidochirotids, S. herrmanni adopts an erect posture during spawning (Figure 6A).

Fig. 5. Number of observations of spawning of Stichopus herrmanni on the Great Barrier Reef and New Caledonia with respect to lunar cycle (NM = new moon).

Fig. 6. Broadcast spawning behaviour of Stichopus herrmanni on One Tree Island, showing the (A) erect posture typical of spawning holothuroids and (B) sperm release. Spawning individuals attract fishes (arrow) to the released gametes.

Observations of spawning by S. herrmanni were most frequently documented in January with additional observations in November, December and February (Figure 5). Overall, most spawning observations were recorded on the day following the new moon, as well as on the new moon and a day before the new moon. Of the months where spawning was recorded, February had the lowest number of spawning events, with only one event recorded on the day after a new moon. There was a single outlier observation, a report of spawning in late October (Table 1).

In the surveys conducted at OTI on 28–29 January 2017 (new moon and one day following) (Table 1), spawning was observed late afternoon of the NM (5/20 spawning) and NM + 1 (16/21 spawning). No spawning was observed on NM + 2 (Table 1). Males spawned before females. As sunset approached individuals were still in the process of moving to elevated positions (Figure 6A). Small fishes attacked the anterior end of spawning individuals and appeared to be consuming the released gametes (Figure 6B).

DISCUSSION

Stichopus herrmanni has an annual reproductive cycle at OTI, spawning from late spring through summer. This pattern, based on evidence from gonad index, gonad histology and spawning observations, is similar to that reported for this species in New Caledonia in a 5-year study (Conand, 1993a, b), at a similar latitude to the population investigated here. Summer spawning is reported for several tropical aspidochirotids, including other stichopodids (e.g. Thelonata ananas) and holothurids (e.g. Holothuria atra) (Conand, 1981, 1993a, b; Conand et al., 2002; Ramofafia et al., 2003). Although our data for elsewhere on the GBR is limited, the spawning observations suggest that S. herrmanni spawns during the summer in this region.

Stichopus herrmanni at OTI has an annual gametogenic cycle. Oocytes developed from late winter to summer with the fully grown oocytes evident from November, as also reported in New Caledonia (Conand, 1993a, b). The histological condition of the gonads indicated that spawning in S. herrmanni is partial with fully mature and partly spawned gonad tubules present throughout summer. Spawning observations also indicate that S. herrmanni has episodic gamete release (i.e. multiple spawning) through late-spring and summer. By April, the gonads were post-spawned and the remaining gametes appear to be reabsorbed. Resorption of unspawned gametes is a common feature of aspidochirotid and echinoderm gonads and is suggested to be associated with exogenous cues and/or an underlying endogenous rhythm (Eckelbarger & Young, 1992; Morgan, 2000; Ramofafia et al., 2003). After the spawning period, the gonads enter a spent/resting phase with little gametogenic activity for 6 months over winter (March/April to Sept/Oct) as also reported for S. herrmanni in New Caledonia (Conand, 1993a, b).

Temperature, salinity, photoperiod and food availability have been suggested to regulate the process of gametogenesis in sea cucumbers (Cameron & Fankboner, 1986; Conand, 1993a, b; Morgan, 2000; Hamel et al., 2001; Conand et al., 2002; Ramofafia et al., 2003; Shiell & Uthicke, 2006). The gonads of S. herrmanni had minimum development in winter and were mature in summer, indicating gametogenesis is controlled by temperature and day length. In both temperate and tropical species, the early gametogenic phase appears to coincide with the cooler months and when days are shorter than nights, as suggested for Actinopyga mauritiana, H. scabra, S. californicus, Thelenota ananas (Cameron & Fankboner, 1986; Conand, 1993a, b; Ramofafia et al., 2001). A resting phase in the winter, when the gonads are at their minimal size (or absent), is a feature described for temperate stichopodids, and is suggested to represent an aestivation-like cessation of activity due to cold conditions at high latitudes (Cameron & Fankboner, 1986). This phenomenon is also observed for tropical species, indicating that a quiescent gonad phase may be common among aspidochirotids (Morgan, 2000; Hamel et al., 2001; Ramofafia et al., 2003).

At OTI, S. herrmanni experiences distinct seasonal temperature fluctuations (~10°C), and the cold season might initiate an aestivation-like response in reproductive activity. However, S. herrmanni is abundant in equatorial locations where annual temperature flux is minimal. In these regions with minimal seasonal cues, continuous reproduction is predicted for echinoderms (Giese & Pearse, 1974) as occurs in equatorial populations of H. scabra (Ramofafia et al., 2003). An assessment of the reproductive cycle of low latitude populations of S. herrmanni would provide insight into whether a gonadal resting phase is species-specific or varies with latitude (i.e. temperature-driven). In addition to the seasonal temperature cycle, day length and lunar cycle, which are fairly predictable among years, it would also be important to consider the influence of other environmental variables such as food availability and cyclones on reproduction of S. herrmanni.

The onset of gonad growth occurred in late-winter (August), correlating with increasing day length. There was an overlap in spent and recovering stages of gametogenesis in the gonads of S. herrmanni at OTI from August to October. Gonad maturation is reached by November at OTI and New Caledonia (Conand, 1993a, b), as temperature and day length increase. Summer spawning indicates that the small individuals observed in autumn (four months later) on nearby reefs are likely to be new recruits (Wolfe & Byrne, 2017b).

While gonad development and maturation appear to be entrained by a seasonal cycle, coinciding with the warmest temperatures and longest days, spawning and the behavioural change to move to elevated places such as coral bommies appears to be cued by the lunar cycle. This behavioural change brings individuals closer together, an aggregation-like response also observed in H. scabra (Morgan, 2000; Hamel et al., 2001). At OTI, S. herrmanni began moving up coral bommies before dusk and spawning started soon after, between 1645 and 2050 h. It is likely that a more aggregated distribution and spawning continues into the night, as for S. chloronotus on the GBR (Babcock et al., 1992). Observations in the Solomon Islands indicate that H. scabra are normally scattered but progressively form pairs, trios and larger groups, peaking a few days before the new moon. Pheromones released by males have been shown to induce aggregation and spawning in H. arguinensis and sympatric holothuroids (Marquet et al., 2018). Aggregative spawning is suggested to be an evolved strategy to increase fertilization rates and reduce predation of gametes through predator satiation (Babcock et al., 1992). As for S. herrmanni, males spawn prior to females in other aspidochirotids (Babcock et al., 1992; Morgan, 2000; Hamel et al., 2001; Marquet et al., 2018). Males release sperm over long periods in contrast to the short bursts of egg release by females (Babcock et al., 1992).

The timing of spawning can differ among geographically separated conspecific holothurian populations (Ramofafia et al., 2003). The tropical sea cucumber H. scabra spawns all year round in the Solomon Islands near the equator and annually in higher latitude regions, a disparity reflecting latitudinal (north-south) differences (see review, Ramofafia et al., 2003). For H. whitmaei, spawning occurs in winter for western and eastern Australian populations at the same latitude, in the Indian and Pacific Oceans, respectively (Shiell & Uthicke, 2006). As S. herrmanni is such a widely distributed species in the Indo-Pacific, it remains to be determined if its reproductive cycle and spawning periodicity differs across oceans and latitude.

Successful reproduction of marine invertebrates requires individuals to be in close proximity (Pakoa et al., 2014). Low population densities effectively reduce the chances of the successful fertilization, leading to reproductive failure – the Allee effect (Allee, 1938). Unlike several tropical holothuroids that can reproduce asexually (e.g. H. atra, Lee et al., 2008), S. herrmanni only reproduces by sexual means. Weakened mating capacity of stocks due to overfishing leads to population declines and local extinctions with minimal population recovery, as observed for several commercial species (Uthicke et al., 2004; Hasan, 2005; Purcell, 2010; Friedman et al., 2011). Although the population density of S. herrmanni is particularly high on OTI (Eriksson et al., 2013; Wolfe & Byrne, 2017a), which has been an unfished reef for decades, populations on reefs open to fishing are vulnerable to exploitation. There are no pre- or post-harvest data for any of the fished areas on the GBR, so the impact of removal of mature adults on reproductive success and the persistence of populations is not known. This is an urgent knowledge gap to address considering the Vulnerable status of S. herrmanni (Eriksson & Byrne, 2015).

Early research on the primary high-value target bêche-de-mer species on the GBR, H. whitmaei, prompted the closure of its fishery over a decade ago (Roelofs, 2004; Uthicke et al., 2004). As populations of high-value species decline, low- to mid-value species are targeted leading to a sequential pattern of depletion across the Indo-Pacific (Purcell et al., 2014; Eriksson & Byrne, 2015; Eriksson & Clarke, 2015). As a result, the mid-value curryfish, S. herrmanni, is now a major target species on the GBR (Eriksson & Byrne, 2015). It is likely that the commercial exploitation of S. herrmanni will mimic the patterns documented for previous target species if appropriate management strategies are not implemented (Eriksson & Byrne, 2015; Eriksson et al., 2015). Based on our data, spawning closures could be considered. At present there are no constraints on fishing throughout the year for S. herrmanni. Importantly, this species exhibits an ontogenetic migration within their recruitment reef (Eriksson et al., 2013; Wolfe & Byrne, 2017b). This connectivity between juvenile recruitment and adult habitats highlights the vulnerability of many marine species to overfishing (Gillanders et al., 2003; Grüss et al., 2011). However, there are no data available on S. herrmanni, and many other target commercial species, on fished reefs along the GBR.

Current approaches of bêche-de-mer fisheries are extremely exploitive and are not viable for holothuroid resources (Conand et al., 2014; Purcell et al., 2012, 2013, 2014; Eriksson & Byrne, 2015), prompting the suggestion that a paradigm shift in fisheries management is needed. Fisheries-independent information is necessary to understand the true status of the commercial fishery on the GBR, and elsewhere. This is particularly important with regard to the aggregative behaviour exhibited during spawning and the clear implications for negative Allee effects following unsustainable fishing practices.