Induction of vitellogenesis, methyl farnesoate synthesis and ecdysteroidogenesis in two edible crabs by arachidonic acid and prostaglandins

In decapods, reproduction is mainly controlled by two antagonistically acting peptide hormones, i.e. gonad-inhibiting hormone (GIH), synthesized and secreted from the X-organ–sinus gland complex of eyestalks, which inhibits gonad development (Nagabhushnam et al., 1980; Rotllant et al., 1993), and gonad-stimulating hormone (GSH), secreted from the brain and thoracic ganglia, which stimulates gonad development (Otsu, 1963; Eastman-Reks and Fingerman, 1984). Later studies suggest that ecdysteroids secreted from Y-organs (YO) trigger reproduction in crustaceans (Chang, 1997; Laufer et al., 1998; Sumiya et al., 2014; Subramoniam, 2000, 2017; Gong et al., 2015, 2016; Medesani et al., 2015). In addition, methyl farnesoate (MF), a secretory product of the mandibular organ (MO), has also been implicated in inducing ovarian maturation in crustaceans (Reddy et al., 2004; Nagaraju, 2007; Kakaley et al., 2017; Li et al., 2019; Xie et al., 2018; Neelima and Reddy, 2019). The ovarian enhancement effect of MF may be indirect by stimulating the YO to synthesize and secrete ecdysteroids (Tamone and Chang, 1993).

Prostaglandins are oxygenated metabolites of arachidonic acid (AA) and two other polyunsaturated fatty acids, eicosatrienoic and eicosapentaenoic acids. Similar to vertebrates, prostaglandins play an important role in regulating several physiological processes in invertebrates including crustaceans (Stanley and Howard, 1998; Rowley et al., 2005; Di Costanzo et al., 2019). Prostaglandins such as PGE₂ and PGF_2α were discovered in several crustaceans and their involvement in the regulation of reproduction was also documented (Spaziani et al., 1993, 1995; Muriana et al., 1995; Raviv et al., 1999; Tahara and Yano, 2004; Meunpol et al., 2005; Huang et al., 2008; González-Félix et al., 2009; Preechaphol et al., 2010; Chansela et al., 2012; Wimuttisuk et al., 2013; Waiho et al., 2017; Di Costanzo et al., 2019). Supplementation of AA, through diet, promotes egg development and spawning in Penaeus monodon (Coman et al., 2011). Administration of PGE₂ stimulated ovarian maturation, oocyte proliferation and vitellogenin synthesis in the giant freshwater prawn, Macrobrachium rosenbergii (Sumpownon et al., 2015). Although we have also demonstrated that injection of PGE₂ and PGF_2α resulted in significant increase in ovarian index and oocyte diameter in a dose-dependent manner in the freshwater crab Oziotelphusa senex senex (Reddy et al., 2004), the interaction of prostaglandins with other crustacean reproductive hormones is not known. The aim of the present study was to determine the effect of prostaglandins in inducing vitellogenesis, ecdysteroidogenesis and MF synthesis in selected edible crustaceans. For the present study, we selected the freshwater rice field crab, O. senex senex and the giant mud crab, Scylla serrata, as test animals. Oziotelphusa senex senex is well known as ‘poor man's protein’ and is one of the important species for freshwater aquaculture; S. serrata is well known for its commercial importance, with the commencement, since the early 1980s, of live crab export from India.

Intact adult crabs, Oziotelphusa senex senex (Fabricius 1798) (body mass of 30±3 g), were collected from paddy fields around Tirupati, India, and maintained at room temperature under a 12 h:12 h light:dark cycle. Crabs were housed 6–8 per glass aquaria (length:width:height=60:30:30 cm) with sufficient ambient medium (salinity: 0.5 ppt) and transferred to fresh medium every day. Scylla serrata (Forskål 1775) (150–170 g body mass) were collected from the Dugarajapatnam coast, PSR Nellore district, Andhra Pradesh, India, and maintained in filtered seawater (salinity 33 ppt) at room temperature under a 12 h:12 h light:dark cycle. The crabs were fed with sheep meat ad libitum while in the laboratory and the ambient medium was changed daily. Female crabs in intermolt stage (stage C₄, according to the criteria described by Reddy, 1991; detailed description is given in Table S1) were selected for the present study.

The crabs were divided into six groups of 20 animals each. Animals in the first group served as initial controls and did not receive any treatment and were killed on the first day of the experiment. Animals in the second group served as concurrent controls and were injected with 10 µl of 5% ethanol in saline, and animals in groups 3, 4, 5 and 6 were injected with 1.0 µg AA, PGF_2α, PGE₂ and PGD₂ per gram body mass, respectively, in 10 µl volume. Injections into groups 2–6 were administered on the 1st, 7th, 14th and 21st day and the crabs were killed on the 28th day. The injections were administered at the base of the coxa of the third walking leg. AA, PGF_2α, PGE₂ and PGD₂ were purchased from Sigma-Aldrich, Mumbai, India, and test solutions were prepared fresh by dissolving AA, PGF_2α, PGE₂ and PGD₂ in 5% ethanol in crustacean saline. Test doses were selected based on our dose–response studies in O. senex senex (Reddy et al., 2004). Ten crabs from each group were weighed, and the hemolymph and ovaries were isolated and used for further analysis. The remaining 10 crabs from each group were used for isolation of the MOs and YOs for determination of MF secretion and ecdysteroid secretion, respectively. No significant change in the body mass of crabs either within or between the six groups was observed during the experimental duration.

where M_ovary is the wet mass of the ovary (g) and M_crab is the mass of the crab (g). For oocyte diameter determination, the ovaries were placed in Bouin's fixative (picric acid:formaldehyde:acetic acid, 75:25:5) immediately after their isolation. After 24 h, they were washed thoroughly and dehydrated with an ascending alcohol series. Dehydrated ovaries were embedded in paraffin wax (melting point 56–58°C) after clearing in xylene, sectioned transversely at 5 μm and stained with hematoxylin and counter-stained with eosin. The maximum and minimum diameters of up to 50 randomly sectioned oocytes of each sectioned ovary were measured using an ocular micrometer under a compound microscope (Olympus, Model-BX41TF HB, Japan). The measurements were made on the longest and shortest axes of each oocyte, both dimensions were added, and the mean was taken as mean oocyte diameter.

Vitellogenin (VtG) was isolated from the ovaries (100 mg) of animals using the protocol described by (Tsukimura et al., 2000). The ovarian VtG content was estimated by enzyme linked immunosorbent assay (ELISA). The robustness of the ELISA was assessed with the measurement of inter- and intra-assay variation, where the variation of intra-assay was measured by adding the same concentration of standard to one microtiter plate 12 times, and the inter-assay variation was measured by comparing various dilutions of the standard added to 12 different microtiter plates. The detection limit of the ELISA was calculated as the lowest standard with an absorbance at least three times higher than the standard deviation of the first value that was significantly different from the ELISA control wells. The data were analyzed with Kyplot (version 2.0). Sensitivity of the ELISA was as little as 5.0 ng of VtG; intra-assay variation was 4% to 6% (n=12) and the inter-assay variation was 9% to 10.6% (n=12). A typical standard curve using purified O. senex senex VtG was presented in our earlier paper (Girish et al., 2015).

Hemolymph was collected from the sinuses at the base of the third walking leg with a syringe and mixed (1:2) with anticoagulant (containing 1.4% Na₂HPO₄, 1.3% KH₂PO₄, 3.2% EDTA, 2% dextrose and 0.25% sodium citrate). The hemolymph samples were stored at −40°C, after centrifugation for 10 min at 4000 g at 4°C.

MOs and YOs were isolated from control and experimental crabs and incubated in 2.0 ml culture medium at 18°C on a shaker (50 rpm) in a CO₂ incubator. The levels of MF and ecdysteroids were determined from culture medium after 48 h. The culture medium for O. senex senex endocrine glands was prepared by dissolving 10.5 g of dried 199 medium (containing Earle's salts, l-glutamine without sodium bicarbonate) in 1 liter of distilled water, and finally, sodium bicarbonate (2.2 g) was added to the medium after sterilization to obtain a pH of 7.0. Further, the medium was supplemented with fetal bovine serum (10%), penicillin-G (12.5 µg ml⁻¹) and streptomycin (12.5 µg ml⁻¹). The osmolarity of the medium used to culture the glands isolated from O. senex senex was standardized earlier (Nagaraju et al., 2005, 2006).

Both MOs and YOs were dissected from ice-anesthetized intermolt (stage C₄) crabs in avitellogenic stage and washed in ice-cold culture medium for 10 min. The incubated glands were observed under a microscope and no significant changes were detected in the size and shape of the cells. The cells were apparently normal even after 72 h incubation in the culture medium. Incubation of MOs/YOs in culture medium resulted in MF/ecdysteroid secretion in a time-dependent manner up to 48 h, whereas incubation of glands up to 72 h resulted in no further increase in secretion (Nagaraju et al., 2006; Reddy and Reddy, 2006). Hence, we restricted our in vitro studies to 48 h.

The culture medium for S. serrata glands was prepared by adding Leibovitz's medium+crab saline (1:1), in which survivability and proliferation of testicular cells of S. serrata was at a maximum (Shashikumar and Desai, 2011). Leibovitz L-15 medium with l-glutamine powder purchased from Thermo Fisher Scientific was used. The medium as powder was reconstituted with sterile filtered (0.22 μm porosity filters) artificial seawater (salinity 29%). Crab (S. serrata) saline was constituted with NaCl (440 mmol l⁻¹), KCl (11.3 mmol l⁻¹), CaCl₂ (13.3 mmol l⁻¹), MgCl₂ (26 mmol l⁻¹), Na₂SO₄ (23 mmol l⁻¹) and Hepes (10 mmol l⁻¹). All media were filtered through 0.22 μm pore size cellulose membrane prior to use. Antibiotics were routinely used and included penicillin-G (12.5 µg ml⁻¹) and streptomycin (12.5 µg ml⁻¹). The osmolarity was adjusted to 1050 mOsm kg⁻¹ with crab saline having poly vinyl pyrrolidone (15 mmol l⁻¹). The pH was adjusted to 7.4.

Individual MOs and YOs were subsequently cultured in 24-well tissue culture plates in 2 ml of the culture medium and either 10 µl of test chemical (30 ng) or 10 µl of 5% ethanol for up to 48 h at 18°C on shaker (50 rpm) in a CO₂ incubator. The secretory products of MO (MF) and YO (ecdysteroids) were determined as specified below at different time points of incubation (6, 12, 24 and 48 h).

MF was extracted from the medium (50 µl) by adding 2.5 ml of acetonitrile and 0.5 ml of 4% sodium chloride (w/v), centrifuged at 1000 g for 10 min at 4°C, and partitioned against 1 ml of hexane. The dried hexane extract was used for the determination of MF using HPLC. To determine circulatory MF levels, freshly drawn hemolymph samples (1.0 ml) were mixed with 2.5 ml of acetonitrile and 1.0 ml of 1% sodium chloride, and then extracted with hexane. In brief, samples containing MF were separated using a Bondapack C18 column (Waters, Milliford, MA, USA) with an isocratic elution of 70% acetonitrile in water (1.0 ml min⁻¹). Peaks were detected at 254 nm by UV absorption. Retention times of the peaks were compared with those of standard MF. The MF was also confirmed by co-chromatography with authentic all-trans-methyl farnesoate (a gift from Dr P. Ramachandra Reddy, Department of Biochemistry, Yogi Vemana University, Y.S.R. Kadapa District, Andhra Pradesh, India). MF levels were expressed as ng ml⁻¹ and ng per MO.

To determine ecdysteroid secretion from YOs, 50 µl of the medium was removed from each well and extracted in 75% methanol. The dried methanolic extract was used for the determination of ecdysteroid level by radioimmunoassay. To determine circulating ecdysteroid levels, freshly drawn hemolymph samples (1.0 ml) were mixed with 2.5 ml of methanol, and the methanolic extract was used for the estimation of ecdysteroids. The radioimmunoassay was conducted as previously described (Chang and O'Connor, 1979) using the ecdysone antiserum (a generous gift from Dr W. E. Bollenbacher, University of North Carolina, Chapel Hill, NC, USA). The sensitivity of the assay was calculated as 0.3 ng and the intra-assay variation was found to be 6.2%. All of the samples were run at the same time to avoid inter-assay variation. Ecdysteroid concentrations were expressed as ng ml⁻¹ and ng per YO.

Statistical analyses were performed by one-way ANOVA followed by Tukey's post hoc HSD test using SPSS.

Following the 28 day treatment regimen, the mean ovarian index, oocyte diameter and ovarian vitellogenin levels in the crabs administered with AA, PGF_2α and PGE₂ were significantly (P<0.05) higher than in the control crabs, whereas mean ovarian index, oocyte diameter and ovarian vitellogenin levels in PGD₂ administered crabs were comparable with those in control crabs (Tables 1 and 2). The mean ovarian index, oocyte diameter in both control and concurrent control crabs were not significantly different following the 28 day experimental period.

The MF levels in the hemolymph of control O. senex senex (Fig. 1A) and S. serrata (Fig. 1B) were 0.25±0.04 and 0.32±0.05 ng ml⁻¹, respectively, and in control crabs following the 28 day experimental period the levels were 0.27±0.04 and 0.35±0.04 ng ml⁻¹. Administration of AA, PGF_2α and PGE₂ resulted in a significant increase (P<0.05) in the hemolymph MF levels in the experimental crabs when compared with control crabs. Conversely, administration of PGD₂ resulted in no significant changes in MF levels when compared with control crabs (Fig. 1).

The ecdysteroid levels in the hemolymph of control O. senex senex and S. serrata were 20.78±3.87 and 21.09±2.35 ng ml⁻¹, respectively (Fig. 2). The mean ecdysteroid levels in hemolymph of concurrent control crabs were comparable with control crabs following the 28 day experimental period. A significant increase (P<0.05) in ecdysteroid levels was observed in the hemolymph of AA, PGF_2α and PGE₂ injected animals when compared with controls, whereas no significant changes were observed in the ecdysteroid levels in PGD₂ administered crabs (Fig. 2).

Injection of AA, PGF_2α and PGE₂ resulted in a significant (P<0.05) increase in MF secretion by MOs when compared with their respective controls (Fig. 3). Similarly, a significant (P<0.05) increase was observed in ecdysteroid secretion by YOs isolated from crabs injected with AA, PGF_2α and PGE₂ when compared with controls (Fig. 4). The MF secretion by MOs and ecdysteroid secretion by YOs isolated from PGD₂ administered crabs was comparable with that of control groups (Figs 3 and 4).

Incubation of MOs isolated from avitellogenic crabs with AA, PGF_2α and PGE₂ resulted in a significant (P<0.05) increase in the MF secretions in a time-dependent manner when compared with their respective controls, whereas the secretion of MF from MO incubated with PGD₂ was comparable with controls (Tables 3 and 4). No significant change in ecdysteroid secretion was observed when YOs isolated from avitellogenic crabs were incubated with AA, PGF_2α, PGE₂ and PGD₂ (Tables 3 and 4).

In crustaceans, maturation of the ovary can be identified based upon change in color, increase in the size of the ovary and increase in oocyte diameter (Nagaraju, 2011; Swetha et al., 2015). The changes in ovarian VtG levels were also used as a sensitive biochemical marker to determine the maturation of the ovary. The ovarian index at the immature stage was 0.367±0.035 (g %) and increased to 0.859±0.039, 1.650±0.060 and 3.238±0.082 (g %) during vitellogenic stages I, II and III, respectively, during the natural reproductive cycle of O. senex senex. Oocyte diameter also significantly increased (P<0.0001) from the immature stage to vitellogenic stage III (20.71±0.528 to 66.31±2.477 μm) during the natural reproductive cycle (Swetha et al., 2015). The ovarian VtG levels at pre-vitellogenic stage were 70±4.7 μg g⁻¹ tissue, and the ovarian VtG levels gradually increased from vitellogenic stage I to vitellogenic stage III (240±16.2 to 690±21.3 μg g⁻¹ tissue) (Girish et al., 2015). In the present study, injection of AA, PGF_2α and PGE₂ into avitellogenic female crabs resulted in significant increases over a 28 day treatment period in ovarian index, oocyte diameter and VtG levels in the freshwater rice field crab, O. senex senex, and in the mud crab, S. serrata. These results are in agreement with earlier results (Wouter et al., 2001; Glencross, 2009; Coman et al., 2011; Maheswarudu and Vineetha, 2013; Kangpanich et al., 2016; Prameswari et al., 2017; Xu et al., 2017). The results indicate that prostaglandins induce ovarian development, which could be (1) a direct action on the ovary or (2) an indirect action by blocking the eyestalk hormones or increased production of reproductive enhancing hormones (MF and ecdysteroids) and/or a combination of internal ovarian and external hormonal factors.

In the present study, administration of AA, PGF_2α and PGE₂ into female animals following a 28 day injection cycle resulted in a significant increase in MF and ecdysteroid levels in the hemolymph when compared with controls. MF and ecdysteroids are synthesized and secreted from the MO and YO, respectively. Circulatory MF levels in O. senex senex increased gradually from the pre-vitellogenic stage to vitellogenic stage III (16.3±3.1 to 56.7±6.2) during natural reproductive stages (Nagaraju et al., 2006). The data clearly indicate that MOs and YOs isolated from AA, PGF_2α and PGE₂ injected crabs showed a marked increase in secretory rates when compared with MOs and YOs isolated from control crabs. This indicates that prostaglandin-induced vitellogenesis might be due to enhanced ecdysteroidogenesis and MF synthesis from YOs and MOs, respectively. It is well established that the synthesis and secretion of MF and ecdysteroids are negatively regulated by mandibular organ inhibiting hormone (MOIH) and molt inhibiting hormone (MIH), respectively, produced in the X-organ–sinus gland complex of eyestalks. The elevated circulatory levels of MF and ecdysteroids and secretory rates of MOs and YOs in O. senex senex after administration of AA, PGF_2α and PGE₂ may be due to inhibition of release of MOIH and MIH from eyestalks or direct action of AA or PGs on MO and YO, or both.

Increased secretory rates of MOs and YOs isolated from AA, PGF_2α and PGE₂ injected crabs following a 28 day injection cycle motivated us to study the secretory rates of endocrine glands isolated from avitellogenic crabs incubated with AA, PGF_2α and PGE₂ in vitro. In the present study, incubation of MOs isolated from avitellogenic crabs with AA, PGF_2α and PGE₂ resulted in a significant increase in MF production in a time-dependent manner indicating the direct action of AA, PGF_2α and PGE₂ on MOs and induction of MF synthesis. Conversely, no significant change was observed in ecdysteroid secretion from YOs isolated from avitellogenic crabs and incubated with AA, PGF_2α and PGE₂. Tamone and Chang (1993) demonstrated that co-incubation of MOs with YOs isolated from Cancer magister stimulated ecdysteroid secretion from YOs in vitro. Elevated circulatory levels of ecdysteroids and increased secretions of ecdysteroids from YOs isolated from AA, PGF_2α and PGE₂ injected crabs in the present study may be due to an ecdysiotropic effect of MF, as suggested by Tamone and Chang (1993). It is well established that the circulatory levels of MF and ecdysteroids are at low levels prior to oocyte maturation and increase during vitellogenesis (Subramoniam, 2000; Gunamalai et al., 2004; Reddy et al., 2004). The induced vitellogenesis in crabs injected with AA, PGF_2α and PGE₂ is in accordance with the elevated circulatory levels of MF and ecdysteroids after AA and PGF_2α and PGE₂ administration. Conversely, PGD₂ has no effect on reproduction in O. senex senex. We have previously observed levels of PGF_2α and PGE₂ that increase in concert with oocyte maturation in O. senex senex, while levels of PGD₂ remain relatively low and constant (Reddy et al., 2004). The present data also reveal that PGD₂ is not involved in regulation of synthesis of MF and/or ecdysteroids. Piecing this evidence together, one may conclude that PGF_2α and PGE₂ are indeed involved in the induction of vitellogenesis, MF synthesis and ecdysteroidogenesis, whereas the involvement of PGD₂ in the regulation of reproduction is indefinable.

The present data suggest that PGD₂ is not involved in the regulation of MF or ecdysteroid synthesis, whereas PGF_2α and PGE₂ may mediate the induction of vitellogenesis by stimulating MF synthesis, and consequent ecdysteroid production. These are the first data showing an association between the prostaglandins and MF synthesis and ecdysteroidogenesis, and they provide an alternative molecular intervention mechanism to the traditional eyestalk ablation methodology to induce fecundity in aquaculturally important crustaceans.