Title: Transient sex change in the immature Malabar grouper, Epinephelus malabaricus, by androgen treatment Running title: Transient sex change in the grouper Summary sentence: Artificial androgen having little effect on the steroidogenic endocrine pathway causes sex change-recovery after treatment

To clarify the cause of sex change recovery after withdrawal of androgen treatment, immature female Malabar grouper were fed a diet containing 17alpha-methyltestosterone (MT) at 50 g/g for 7 mo and then a normal diet for 6 mo. MT brought about precocious sex change from immature ovaries to mature testes with active spermatogenesis, including development of spermatozoa, and sex change reversed soon after MT treatment withdrawal. This result indicates that precocious sex change in immature Malabar grouper with oral MT treatment is impermanent. The expression of three steroidogenic enzymes (Cyp11a, Cyp19a1a, and Cyp11b) in the gonads of the Malabar grouper were analyzed immunohistochemically at the end of the 7-mo treatment. No apparent differences were seen in the expression pattern of these enzymes between the mature testes of MT-treated fish and the immature ovaries of control fish. In addition, serum estradiol-17beta and 11-ketotestosterone levels in treated fish were same as those in control fish. These results indicate that in the case of immature Malabar grouper, MT might have little effect on endogenous steroidogenesis during precocious sex change even though it induced active spermatogenesis in the gonads of treated fish. From these results, we concluded that this fact; MT might have little effect on the steroidogenic endocrine pathway, is one of a cause of sex change recovery after treatment withdrawal. INTRODUCTION Groupers are protogynous hermaphrodites; their sex changes from female to male when they reach a large size [1-7]. Mature males is an important prerequisite in artificial seed production of grouper; however, their availability is limited owing to their reproductive characteristics. Artificial sex change has been successfully performed in some grouper species to overcome the shortage of mature males, and BOR Papers in Press. Published on May 14, 2014 as DOI:10.1095/biolreprod.113.115378 Copyright 2014 by The Society for the Study of Reproduction. these studies have shown successful artificial permanent sex change from female to male induced via downregulation of endogenous estrogen synthesis using exogenous androgen or aromatase inhibitors [817]. Exceptionally only in the young (1~4 years old) dusky grouper, Epinephelus marginatus,, it was seen or mentioned that sex inversion after orally administered or injected MT may not maintained and their sex changes may back to females during the next reproductive period. However, the sexual recovery after MT treatment was not demonstrated, and the cause of this phenomenon remains unclear [18, 19]. To the best of our knowledge, our recent study demonstrated for the first time that precocious sex change from immature ovaries to mature testes can be induced in underyearling Malabar grouper, Epinephelus malabaricus, with 17-methyltestosterone (MT) treatment after ovarian differentiation [20]. This precocious sex change is variable for the aquaculture of grouper; however, testes with active spermatogenic germ cells that have changed from immature ovaries are highly likely to change back to ovaries after treatment withdrawal as the previous report in the dusky grouper [18, 19]. Generally in gonochoristic teleost fish, endogenous sex steroid hormones play a role in natural gonadal sex differentiation and the most effective period to induce artificial sex change via exogenous sex hormones is thought to be during gonadal sex differentiation; it is almost impossible to induce sex reversal after sex differentiation [21]. Exogenous androgen has also been demonstrated to affect the expression of key steroidogenic enzymes, including Cyp19a1a, which is essential for estrogen synthesis, to induce permanent sex reversal in the gonochoristic Nile tilapia, Oreochromis niloticus [22]. In some grouper species, expression patterns of steroidogenic enzymes in the gonads and serum sex steroid hormone levels change during natural or artificial sex change [7, 23-27]. Thus, we believe that completion of permanent sex change requires a change not only in germ cell differentiation from oogenesis to spermatogenesis but also in the expression patterns of steroidogenic enzymes during sex change from ovary to testis. First in this study, we investigated whether the testis which is reversed from immature ovary using MT would recovery after withdrawal of MT treatment by histological analysis. Next, to clarify the cause of recovery after sex change, we analyzed the expression patterns of steroidogenic enzymes in the gonads of MT-treated and control fish during precocious sex change; serum sex hormone levels were also examined. MATERIALS AND METHODS Animals Specimens of E. malabaricus were raised in the Okinawa Prefectural Fisheries and Ocean Research Center in Ishigaki, Okinawa, Japan. The fish were maintained as described previously [28]. We believe that they reach sexual maturity as females approximately 5 years after hatching and change their sex from female to male more than 10 years after hatching. Hormone treatment and sampling MT was dissolved in 96% ethanol and added to commercial fish feed (Pure-gold; Marubeni Nisshin Feed Co., Ltd., Japan) at a concentration of 50 g/g as described previously [21]. Control feed was mixed with 96% ethanol only. After air drying overnight, the feed was stored at room temperature until used for feeding. All fish were anesthetized with 0.05% 2-phenoxyethanol (Wako Chemicals, Osaka, Japan) before sampling. After measurements of total length and body weight, the fish were euthanized by decapitation. All animal handling and experimental procedures were conducted in accordance with our Guide for Care and Use of Laboratory Animals (animal-jikken-kisoku 19.6.26) and were approved by the University of the Ryukyus. For initial controls, seven fish were sacrificed, and their gonads were collected to determine the gonadal status before the start of the treatment. The control group fish (n = 50) received feed without MT, whereas the MT-treated group (n = 52) received a diet containing 50 g/g of MT. Fish were fed once daily at the rate of 0.75% of their body weight. One, 2, 3, 5, and 7 months after the start of treatment, 6 to 10 fish from each group were sacrificed, and their gonads and blood were collected to examine the effects of MT on the gonads and steroid hormone levels. Then, 1, 3, and 6 months after the withdrawal of treatment, four to six fish from each group were sacrificed, and their gonads and blood were collected. An overview of the experimental design is shown in Fig. 1. Control and MT-treated fish were kept in actively aerated seawater in a 1.0 ton outdoor tank respectively before the withdrawal of MT treatment (34 fish / 1.0 ton tank at the end of treatment), then micro tip (Pit tag BIO12A BIO MARK) were injected to each fish to detect which fish is MT-treated or control, and they were kept in same 1.0 ton outdoor tank with some fish which was used in another experiment (over 60 fish / 1.0 ton tank). Histological observation The gonads were fixed in Bouin’s solution, embedded in paraffin, cross-sectioned, and stained with Delafield’s hematoxylin using standard methods for light microscopy. Immunohistochemistry The expression of three steroidogenic enzymes (Cyp11a, Cyp19a1a, and Cyp11b) in the gonads was examined as described by Murata et al. 2011 [29]. Measurement of steroid hormones Serum estradiol-17β (E2) and 11-ketotestosterone (11KT) were quantified by using enzymelinked immunosorbent assay (ELISA) as described previously Asahina et al. 1985 [30]. Statistical analysis All data are expressed as mean ± standard error of the mean. One-way analysis of variance (ANOVA) and the Tukey-Kramer comparison test were used for statistical analysis of E2 and 11KT levels. RESULTS Effects of MT treatment on gonads The size, age, growth performance, and gonadal status of fish during the experimental period are summarized in Table 1. All individuals in the initial control had immature ovaries consisting of the ovarian cavity and a few oogonia (Fig. 2A). As the experiment progressed, the immature ovaries of all control fish developed slowly, as described in our previous report [28], and perinucleolar stage oocytes were detected in the ovary at the end of the experiment (Fig. 2B, D, F, and H). Conversely, 2 months after the start of the experiment, the slit (efferent duct–like structure) appeared between the ovarian cavity and blood vessel on the dorsal side in the gonads of all MT-treated individuals, and spermatocytes were first detected in the gonads of some MT-treated fish, indicating the onset of spermatogenesis (Fig. 2C). In MTtreated fish, spermatogenesis progressed until 7 months after the start of the experiment, and 80% of MTtreated fish developed mature testes with active spermatogenesis and contained spermatozoa in the efferent duct-like structure (Fig. 2E). Spermatozoa in the gonads of almost all MT-treated fish disappeared within 1 month after the withdrawal of MT treatment (Fig. 2G). Three and 6 months after the withdrawal of MT treatment, neither sperms nor spermatogenesis were detected in the gonads of any MTtreated fish, and perinucleolar oocytes were seen in the gonads of some MT-treated fish as in control fish (Fig. 2I). Sterile gonads consisted of somatic cells only; no germ cells could be observed in these sterile, MT-treated fish during the last five months of the experiment. (data are not shown). Expression of steroidogenic enzymes and serum sex steroid hormone level To clarify the effects of MT on the expression of steroidogenic enzymes, we used immunohistochemistry to investigate the expression pattern of three steroidogenic enzymes—Cyp11a, the key steroidogenic enzyme; Cyp19a1a, which is essential for E2 production; and Cyp11b, which is important for fish-specific androgen 11KT production—in the gonads of Malabar grouper 7 months after the start of experiment. In the control immature ovary, Cyp11a-positive cells appeared in th