Enhancing spawning in the grey mullet (Mugil ... - Semantic Scholar

General and Comparative Endocrinology 142 (2005) 212–221 www.elsevier.com/locate/ygcen

Enhancing spawning in the grey mullet (Mugil cephalus) by removal of dopaminergic inhibition Joseph Aizena,b, Iris Meiria, Itai Tzchoria, Berta Levavi-Sivanb, Hanna Rosenfelda,¤ a

Israel Oceanographic and Limnological Research, National Center for Mariculture, Eilat 88112, Israel b Department of Animal Sciences, Faculty of Agricultural, Food and Environmental Quality Sciences, The Hebrew University of Jerusalem, Rehovot 76100, Israel Received 19 September 2004; revised 4 January 2005; accepted 6 January 2005 Available online 3 February 2005

Abstract A dot-blot immunoassay for the detection of vitellogenin (Vtg) in plasma of adult grey mullet (Mugil cephalus) was developed. The assay identiWed the sex of the tested Wsh prior to detectable gonadal development, enabling the establishment of broodstock at the desired ratio of 7:4 females to males. This broodstock was maintained under natural photoperiod, and used to study the relative eVect of gonadotropin-releasing hormone (GnRH) and dopamine antagonists on oocyte maturation and ovulation, as well as the eVect of 17-methyltestosterone (MT) on spermiation. Three groups of females were treated with: (i) a single injection of dopamine antagonist, domperidone (Dom), (ii) GnRH analog (GnRHa) administered via ethylene-vinyl acetate copolymer (EVAc) slowrelease implants or (iii) a combination of both Dom and GnRHa. Males were treated with MT, administered via EVAc slow-release implants. An additional group of untreated Wsh was used as a control. The Dom treatment proved to be more potent than the GnRHa treatment, and did not diVer signiWcantly from the combined treatment. The Dom and Dom + GnRHa treatments accelerated oocyte development and increased plasma estradiol levels equally, whereas the GnRHa treatment did not vary signiWcantly from the control. MT was found to be a potent spermiating agent, which enhanced steady milt production in all treated males. In contrast, no spontaneous spermiation occurred in untreated males. Plasma 11-ketotestosterone (11-KT) levels were signiWcantly higher in MT-treated males than in the controls. Interestingly, MT-treated males held with the GnRHa + Dom-treated females showed higher levels of plasma 11-KT than those held with GnRHa-treated females, indicating an additive eVect which is probably attributable to female pheromones. Fully mature females were induced to spawn by injecting GnRHa alone or coupled with metaclopramide (a dopamine D2 receptors antagonist). The combined treatment, which included a dopamine antagonist, was found to be more potent in inducing ovulation and spawning as compared to GnRHa alone. In conclusion, our data suggest that dopaminergic inhibition is a major barrier along the reproductive axis that arrests spontaneous spawning in captive mullets.  2005 Elsevier Inc. All rights reserved. Keywords: Dopamine; Gonadotropin-releasing hormone; Methyltestosterone; Mugil cephalus; Pheromones; Spawning induction; Steroids; Vitellogenin

1. Introduction The grey mullet (Mugil cephalus) is a euryhaline marine teleost widely cultured in brackish and fresh water semi-intensive Wshponds. Due to its benthic ¤

Corresponding author. Fax: +972 8 6375761. E-mail address: hanna@ocean.org.il (H. Rosenfeld).

0016-6480/$ - see front matter  2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ygcen.2005.01.002

feeding behavior it is considered an eYcient bioremediator in aquaculture (Lupatsch et al., 2003), and it is often used in polyculture systems to improve sediment quality (Sarig, 1981). In Asian and Mediterranean markets, mullet roe is a high-priced product, supplementing the Wsh Xesh. These commercial and environmental attributes make mullet an important candidate for domestication. However, Wngerling supplies for

J. Aizen et al. / General and Comparative Endocrinology 142 (2005) 212–221

aquaculture have come almost exclusively from the wild. In nature, the grey mullet only spawns in seawater; but, then can grow and thrive in lower-salinity water. The reproductive season of the Mediterranean population ranges from July to December (Abraham et al., 1966). However, under captive conditions, mullet brooders do not spawn spontaneously, with both genders displaying reproductive dysfunctions, even in full strength seawater. In Hawaii, cultured mullet males have been found to mature and regress earlier than females (Weber and Lee, 1985), and even then, in most cases, the produced milt is highly viscous and fails to fertilize eggs (Shehadeh et al., 1973). Female striped mullet, usually completes vitellogenesis when kept in captivity, but does not undergo Wnal maturation and spawning (Kuo et al., 1974a; Liao, 1981; Nash and Koningsberger, 1981). A diVerent situation has been found in the Mediterranean grey mullet brooders, where males with running milt are rarely observed (De Monbrison et al., 1997; Yashouv, 1969), and female dysfunctions were conWned to two critical phases, i.e., the early stages of vitellogenesis, and Wnal oocyte maturation and ovulation (De Monbrison et al., 1997). The assumption was that lower concentrations or the absence of pituitary gonadotropins (mostly LH) underlie mullet reproductive failures, as has been found in other Wn-Wsh species (Zohar, 1989; Zohar and Mylonas, 2001). Accordingly, co-treatments based on carp pituitary homogenate (CPH) and gonadotropinreleasing hormone (GnRH) were adopted to induce spawning (Lee and Tamaru, 1988). These treatments were eVective, but required extremely high hormone doses relative to the amounts used for spawning induction in other species, suggesting the involvement of an inhibitory factor(s). Such inhibition has been well characterized in cyprinids (reviewed by Yaron, 1995), salmonids, and other teleosts (reviewed by Saligaut et al., 1999). In these Wsh, a strong endogenous inhibitory impact of dopamine compromises the ability of externally applied GnRH to increase LH release, and consequently the Wsh fail to ovulate (Yaron, 1995). The widely adopted method of spawning induction in Wsh with strong dopaminergic inhibition relies on the simultaneous application of a potent dopamine receptor antagonist and GnRH superactive analog (GnRHa) (Peter et al., 1988; Yaron, 1995). Using the same approach, Glubokov et al. (1994) demonstrated that the combination of GnRHa with dopamine antagonists is most eVective at inducing spawning in the PaciWc mullet (Mugil soiuy Bas). In the current study, with the aim of inducing successful breeding in captive Mediterranean grey mullets, a practical technique for sexing mullets at the onset of their reproductive season was developed. As mullets do not display sexual dimorphisms, our method helps establishing brooder groups at desired sex ratios. Based on

213

preliminary study with the Mediterranean grey mullets (Tzchori et al., 1997), hormonal treatments were designed to address the two blockages identiWed during the reproductive cycle of this Wsh. Particular attention was devoted to the relative eVects of GnRHa and antagonists of dopamine D2 receptors on ovarian recrudescence.

2. Materials and methods 2.1. Broodstock maintenance Grey mullet brooders, originally obtained from the Israeli Mediterranean coast, consisted of 5- to 8-year-old females (body weight [BW] 2.3 § 0.08 kg), and 3- to 5year-old males (BW 1.1 § 0.07 kg). Fish were maintained outdoors in 5-m3 covered Wberglass tanks supplied with seawater at 40-ppt salinity from the Gulf of Eilat, and subjected to natural conditions of light and water temperature. The experimental tanks were stocked equally with respect to density (4.0 § 0.174 kg/m3) and sex ratio (7:4 females to males). Fish were fed daily at the rate of 1.5% of their body weight using a 35% crude protein and 7.2% lipid feed formulation prepared at the National Center for Mariculture (Eilat, Israel). 2.2. Hormonal acceleration of gonadal development Mullet females were injected intramuscularly with either (i) GnRH agonist (GnRHa) (des Gly10 [D-Ala]6; Sigma, St. Louis, MO, USA) loaded on ethylene-vinyl acetate copolymer (EVAc) slow-release implants, (ii) dopamine antagonist, domperidone (Dom; Sigma) dissolved in a vehicle (0.7% NaCl, 0.1 M sodium metabisulWte) or (iii) a combination of the two. Mullet males were injected with 17-methyltestosterone (MT) (Sigma) loaded on EVAc slow-release implants. The control Wsh were injected with saline. EVAc implants containing GnRHa or MT were prepared at the Hellenic Center for Marine Research, Institute of Aquaculture (Iraklion, Crete, Greece), according to Zohar et al. (1990, 1995). Prior to all handling procedures, Wsh were anesthetized in 0.07% clove oil (Frutarom, Haifa, Israel). Experiments were run from July to October, the natural spawning season of the Mediterranean grey mullet, for two consecutive years, 2002 and 2003. The Wrst experiment investigated the eVects of GnRHa-EVAc (10 g/kg BW), alone or in combination with Dom (5 mg/kg BW) on ovarian development; the second experiment investigated the eVects of Dom (5 mg/kg BW) alone. In parallel, the eVects of MT-EVAc (4 mg/kg BW) on testicular development were studied. Control Wsh were injected with saline. Each experimental group consisted of 14 females and eight males split in two tank replicates.

214


2.3. Spawning induction

2.6. SDS–PAGE, Western blot, and dot-blot analyses

A reproductively mature female (as determined by gonadal biopsy) was stocked with two spermiating males in a 1-m3 tank supplied with seawater at 24–28 °C. The selected Wsh were treated with either GnRHa alone or in combination with a dopamine antagonist, metaclopramide (MET) (Sigma). Each treatment consisted of priming (GnRHa 10g/kg; Met 15 mg/kg) and resolving (GnRHa 20 g/kg; Met 15 mg/kg) injections given 22.5 h apart. Females selected for spawning-induction trials (2002, n D 14; 2003, n D 7) included those previously treated with GnRHa (2002, n D 3), GnRH + Dom (2002, n D 10; 2003, n D 2), or Dom (2003, n D 5) to enhance ovarian maturation, as well as one untreated female (2002). Once spawned, the fertilized eggs were collected into a 500-m mesh net and placed in a 160-L tank with running seawater. The average diameter of the spawned eggs and total egg volume and weight were recorded for each spawn. The number of eggs per liter (N) was calculated according to the equation: N D 1,222,000 £ diameter¡2.71 (De Monbrison et al., 1997).

Plasma samples were subjected to polyacrylamide gel electrophoresis (PAGE) using the SE 250 Mighty Small apparatus (Hoefer, San Francisco, California, USA).

2.4. Assessment of gonadal development The state of ovarian development, as determined by oocyte growth, was examined bi-weekly from biopsies using polyethylene cannulae. Ovarian maturity was determined according to Kuo et al. (1974b); except that in the Mediterranean strain, female mullets are considered fully mature once their oocytes reach an average diameter greater than 500 m (De Monbrison et al., 1997). In parallel, male mullets were checked for the presence of milt by applying gentle abdominal pressure. 2.5. Steroid hormone determination Blood samples of mullet broodstock were collected from July through October using heparinized syringes and transferred into tubes containing 0.1 mM Complete (Roche, Mannheim, Germany) to prevent proteolysis. The blood was centrifuged (1500g for 30 min at 4 °C) and plasma was recovered. Steroids were extracted with ethyl ether. The 11-ketotestosterone (11-KT) in males and 17estradiol (E2) in females were determined by speciWc enzyme-linked immunosorbent assay (ELISA), according to Cuisset et al. (1994) and Nash et al. (2000), using acetylcholinesterase as a label. The anti-E2 was previously described in Levavi-Zermonsky and Yaron (1986). The anti-11-KT was donated by Dr. D.E. Kime (SheYeld, UK) and was previously described in Cuisset et al. (1994). All samples were analyzed in duplicate, and for each ELISA plate, a separate standard curve was run. The lower limits of detection were 0.93 and 0.5 pg/ml for 11KT and E2, respectively. The intra- and interassay coeYcients of variance were less than 7 and 11%, respectively.

2.6.1. Western-blot analyses Plasma samples of adult grey mullets, including two females, one male, and six Wsh of unknown sex, were electro-transferred from the polyacrylamide gel to a nitrocellulose membrane using the TE 22 Mighty Small Transphor Unit (Hoefer). Non-speciWc binding sites were blocked with 5% skim milk in PBS (137 mM NaCl, 8 mM Na2PO4, 1.4 mM KH2PO4, and 6.6 mM KCl, pH 7.5). For detection of vitellogenin (Vtg), the membrane was incubated for 1 h at room temperature with polyclonal antibodies developed against the 180-kDa subunit of seabream Vtg (anti-sb-Vtg; donated by Dr. E. Lubzens, Haifa, Israel), diluted 1:5000, washed Wve times with PBS-T (PBS plus 0.05% Tween 20), and then incubated with horseradish peroxidase (HRP) conjugated to goat-anti-rabbit-IgG. After 1 h of incubation, the membrane was washed as described above and the signal was visualized by an ECL detection system (Amersham– Pharmacia-Biotech, Uppsala, Sweden). 2.6.2. Dot-blot analyses Plasma samples of 3-year-old mullets of unknown sex were placed directly on a nitrocellulose membrane using the Bio-Dot apparatus (Bio-Rad, Rishon Le Zion, Israel). The following consecutive steps (i.e., blocking, exposure to the anti-sb-Vtg, washes, exposure to the HRP-labeled goat-anti-rabbit-IgG, and ECL visualization) were performed as described above. However, to suppress nonspeciWc binding of the antiserum, it was Wrst absorbed on a nitrocellulose membrane loaded with the equivalent amount of plasma taken from mature mullet males as described by Heppell and Sullivan (1999). 2.7. Statistical analysis The results for each treatment are expressed as means § SEM. Statistical analyses were performed using Prism 4 (Graph-Pad Software, San Diego, CA). Homogeneity of variance was assessed by Bartlett’s test and data were compared by ANOVA followed by Student– Newman–Keuls multiple-range test.

3. Results 3.1. Vtg immunoassay for assessing mullet gender Plasma samples of adult grey mullet were collected in mid-April 2002, when the gonads were not yet accessible for biopsy, and immuno-analyzed with heterolo-


215

Fig. 1. Detection of mullet Vtg by heterologous antibody raised against seabream vitellogenin (anti-sb-Vtg). (A) Western-blot analysis. Mullet plasma proteins (lanes 1–9) were separated by SDS–PAGE (7.5%) and analyzed by Western blot using anti-sb-Vtg polyclonal antibody. The position of Vtg (209 kDa) was ascertained by standard molecular-weight markers, and is depicted with an arrow. (B) Dot-blot immunoassay. Mullet plasma samples (dots 1–13) were applied to a nitrocellulose membrane and incubated with anti-sb-Vtg, which was pre-adsorbed onto an equivalent membrane loaded with mature male mullet plasma.

gous antisera against Vtg. Western-blot analysis revealed antibody speciWcity to a high-molecularweight protein (approximately 209 kDa) in the plasma of the females (Fig. 1, lanes 1 and 9). A faint band at the equivalent size (»209 kDa) was also detected in plasma samples of three Wsh of unknown sex (lanes 2, 6, and 7), indicating their female sex. Nevertheless, one additional low-molecular-weight band (approximately 80 kDa) was detected in all plasma samples, including that of a known male (lane 8). Therefore, for sexing a new batch of 3-year-old mullets by dot-blot immunoassay, the antisera were pre-adsorbed to male plasma. As shown in Fig. 1B, putative vitellogenic females could be distinguished by a strong positive signal (dots 4, 6, 7, 10, 11, 12, and 13), as compared to putative males (dots 3, 5, 8, and 9), for which no signal was detected. These results were conWrmed by biopsy at the more advanced developmental stages of gonadal growth (mid-July). 3.2. EVects of GnRHa and domperidone on mullet ovarian performance During the Wrst experimental period (July–October, 2002), mullet females were treated with GnRHa, administered via EVAc implants, alone or in combination with Dom (GnRHa + Dom), and were compared to salineinjected Wsh (control). Within 66 days post-treatment (dpt), a signiWcantly (P < 0.05) higher oocyte growth rate was recorded in the GnRHa + Dom group as compared to the GnRHa or control groups, which did not vary sig-

niWcantly from each other (Fig. 2A). The only signiWcant (P < 0.01) diVerence between the GnRHa and control groups was detected at the last sampling point (99 dpt). At that stage, most (>85%) GnRHa + Dom-treated females exhibited fully mature oocytes ranging from 500 to 650 m, while a lower percentage of equivalent females was recorded among the GnRHa (85%) of fully mature females in this group. These Wndings suggest an additive eVect attributable to female mullet pheromones, and expand our inferred notion regarding the importance of female–male hormonal relationships in this species. In that regard, extensive studies with salmonid Wsh suggest that urinary pheromones released by ovulatory females trigger endocrine changes in spawning males (Ako et al., 1994; Cardwell et al., 1996; Liley et al., 1993; Olsén et al., 2001; Olsén and Liley, 1992; Scott et al., 1994; Scott and Sorensen, 1994). In the course of this study, we also deWned the relative eVects of GnRHa and Dom, a well-known antagonist of


dopamine D2 receptors, on mullet ovarian development. The Dom treatment was more eVective than the GnRHa treatment, and was not signiWcantly diVerent from the combined treatment. The Dom and Dom + GnRHa treatments accelerated oocyte development equally, leading to high percentages (>85%) of fully mature females within these groups. In comparison, signiWcantly lower rates of fully mature females were recorded among the GnRHa-treated (