The Social Status of the Male Nile Tilapia (Oreochromis Niloticus) Inﬂuences Testis Structure and Gene Expression

REPRODUCTIONRESEARCH

The social status of the male Nile tilapia (Oreochromis niloticus) inﬂuences testis structure and gene expression

Frank Pfennig, Thomas Kurth1, Stefan Meißner, Andrea Standke, Markus Hoppe, Freia Zieschang, Christine Reitmayer, Andy Go¨bel, Georg Kretzschmar and Herwig O Gutzeit Institute of Zoology, Technische Universita¨t Dresden, D-01062 Dresden, Germany and 1Center for Regenerative Therapies, Technische Universita¨t Dresden, D-01307 Dresden, Germany Correspondence should be addressed to F Pfennig; Email: [email protected]

Abstract

Dominant and territorial behaviour are known social phenomena in cichlids and social stress influences reproduction and growth. The gonadotropic hormones trigger spermatogenesis and subordinate males have typically lower levels of gonadotropins than dominant males. In this study, we compared testis morphology and gene expression of dominant and subordinate Nile tilapia males (d- and s-males) in socially stable communities. The d-males had the highest gonadosomatic index but they were not the largest animals in the majority of studied cases. Long-term d-males showed large groups of Leydig cells and hyperplasia of the tunica albuginea due to numerous cytochrome-P450-11b-hydroxylase (Cyp11b) expressing myoid cells. Increased Cyp11b expression in d-males was reflected by elevated 11-ketotestosterone plasma values. However, immunofluorescence microscopy and expression analysis of selected genes revealed that most s-males conserved their capability for spermatogenesis and are, therefore, ready for reproduction when the social environment changes. Moreover, in s-males gene expression analysis by quantitative RT-PCR showed increased transcript levels for germ line-specific genes (vasa, sox2 and dmc1) and Sertoli-specific genes (amh, amhrII and dmrt1) whereas gene expression of key factors for steroid production (sf1 and cyp11b) were reduced. The Nile tilapia is a promising model to study social cues and gonadotropic signals on testis development in vertebrates. Reproduction (2012) 143 71–84

Introduction germ line cells. Within a single cyst all germ line cells are derived from a single spermatogonium by synchronous The Nile tilapia, Oreochromis niloticus L., is a robust food mitotic divisions. After several rounds of mitotic fish of which several million tons are produced annually divisions the germ line cells also enter meiosis worldwide (FAO 2008). This African cichlid fish came simultaneously. The germ line cells in a developing into the focus of research because knowledge about cyst are in close contact with somatic Sertoli cells that reproduction and physiology is necessary for effective growth in aquaculture (Lim & Webster 2006). Meanwhile, also form the cyst wall. The spermatogenic tubule is O. niloticus has become a valuable model for gonad surrounded by a thin layer composed of basal lamina and development and gene expression studies (Matta et al. peritubular myoid cells. The space between the tubules 2002, Davis et al. 2008, de Alvarenga & de Franca is referred to as intertubular compartment or interstitium 2009, Kobayashi & Nagahama 2009, Lacerda et al. 2010, and is filled with steroidogenic Leydig cells (LC), blood Soler et al.2010). Genetic factors together with environ- vessels, macrophages, mast cells, neuronal cells and mental cues determine the phenotypic gender of this connective tissue cells (reviewed in Schulz et al. (2010)). gonochoristic species (Baroiller et al.2009a,b). The testis is enclosed by an epithelium with a thin Reproduction in nature requires the coordination of subjacent layer of collagenous connective tissue termed gonad development and social behaviour. Seasonal tunica albuginea (TA) by most authors (discussed in changes in gonad structure, onset of puberty or social Koulish et al. (2002)). Little is known about the effects on fertility, are controlled by the endocrine system ultrastructure of the TA in teleost testis. A study on along the brain–pituitary–gonad axis. FSH and LH protogynous fish (Thalassoma bifasciatum) reported mediate neuronal signals to the gonads and stimulate continuity between the subjacent layers close to the spermatogenesis (Weltzien et al. 2004, Levavi-Sivan outer epithelium and the interstitium (Koulish et al. et al. 2010). Spermatogenesis in fish occurs in semi- 2002). In O. niloticus mostbutnotallofthe niferous tubules that consist of numerous cysts filled with undifferentiated spermatogonia are located close to the

q 2012 Society for Reproduction and Fertility DOI: 10.1530/REP-11-0292 ISSN 1470–1626 (paper) 1741–7899 (online) Online version via www.reproduction-online.org Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access 72 F Pfennig and others

TA (Vilela et al. 2003). This organisation may be an Amh receptor-deficient Medaka mutant (Morinaga et interpreted as an intermediate form between the al. 2007) whose phenotype led the authors to conclude restricted and unrestricted spermatogonial distribution. that Amh inhibits gonial proliferation in adult testis and In the first type undifferentiated spermatogonia are initiates premature meiosis in males. restricted to the vicinity of the TA, whereas in the Dominant and territorial behaviour are well known in unrestricted type they are located along the entire tubule cichlids (Turner & Robinson 2000). Such behaviour gives (Grier et al. 1980, Grier 1981, Parenti & Grier 2004, an advantage for the dominant/territorial (d)-male for Schulz et al. 2010). mating with a female (Kelly 2008). In addition, stable Only few data exist about the regulative network social hierarchies save energy and reduce the risk of of endocrine and paracrine factors and about injury (Kaufmann 1983). Dominant behaviour is accom- the interaction between the different testis cell types. panied by an elevation of GnRH1 levels in the The steroid hormone 11-ketotestosterone (11-KT) is the hypothalamus (White et al. 2002, Burmeister et al. major androgen in fish and promotes spermatogenesis. 2005). Experiments with the cichlid Astatotilapia burtoni Cytochrome-P450-11b-hydroxylase (Cyp11b) catalyses showed increased levels of GnRH1, FSH, LH and plasma the conversion from testosterone to 11b-hydroxytestos- 11-KT, as well as an increased size of the gonad in terone, the immediate precursor of 11-KT. In the male socially ascending males while subordinated (s)-males gonad this enzyme was found to be present mainly in LC show the opposite effect (Francis et al. 1993, White et al. (Devlin & Nagahama 2002, Fernandes et al. 2007, 2002, Renn et al. 2008, Maruska & Fernald 2011, Zhang et al. 2010). 11-KT synthesis is stimulated by Maruska et al. 2011). The social ascent to become a gonadotropins (Billard et al. 1982) which act on Sertoli d-male proceeds within only few hours and is promoted cells, which, in turn, respond by producing further by the immediate-early gene (egr-1) in the brain of factors like activin B and anti-Mu¨llerian hormone (Amh) A. burtoni (Burmeister et al. 2005). Once a stable social to regulate proliferation of spermatogonia and the onset status is reached, the expression of egr-1 remains of meiosis (Miura et al. 1991, Nagahama 1994, Ikeuchi unchanged, but the expression of that gene is apparently et al. 2001). In the Japanese eel, activin B is stimulated only required to initiate the social status change. The by 11-KT while the Amh level is reduced and this is stability of the social structure in cichlids (Dijkstra et al. thought to trigger spermatogenesis (Miura et al. 1995, 2007, Maruska et al. 2011) is under endocrine control. 2002). These data are supported by the analysis of 11-KT feeds back from testis to GnRH-containing

A BC Figure 1 Morphological and cytological differences between dominant (d) and subordinate (s) Nile tilapia males. (A) Photograph of a stable community (18 months old) with dark coloured s-males and females near the water surface and a pale d-male that controls the area around the bottom of the tank. (B and C) Body-colouration and eye- darkening pattern for a d-male (B) and an s-male (C). Iris and sclera from the d-male are completely pale. Insets in B and C show the whole animal D 2 mm (yellow barZ30 cm). (D–F) Light microscopic comparison of paraffin-embedded testis cross- sections (4 mm) from a d- and two s-males (sample ID58, 59 and 64, respectively, see Supplementary * * Table 1, see section on supplementary data given at * the end of this article). Immunolocalisation of Vasa protein (brown stain) counterstained with methyl- d-male s-male s-male# green. s-male# in D shows an example for a fully degenerated testis. Note that the degenerated testis EFLC LC completely lacks Vasa protein. Asterisks indicate large groups of Leydig cells (LC) – a higher magnification of this area is shown in Supple- mentary Figure 3, see section on supplementary data given at the end of this article. Cross-sections are orientated with the efferent ducts located towards the top of the picture. (E and F) Details of light micrographs from D. E shows the extended tunica albuginea (TA, arrow) and LC between spermatogenic tubules in a d-male. (F) TA (arrow) 100 µm and spermatogenic tubules of an s-male in comparison to E. E and F same magnification.

Reproduction (2012) 143 71–84 www.reproduction-online.org

Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access Social status and testis structure in tilapia 73 neurons in the hypothalamus (Soma et al. 1996, several subordinates from the same tank and all animals Weltzien et al. 2004) and, furthermore, it is thought were killed at the same time. to trigger aggressive behaviour of A. burtoni d-males The calculated gonadosomatic index (GSI) showed (Francis et al. 1992, Parikh et al. 2006). Such social higher values for the d-male in comparison with the s-males stress suppresses growth and reproduction of Nile in all of the 20 communities (Fig. 2, Supplementary tilapia s-males (Fernandes & Volpato 1993, Correâ Table 1 and Supplementary Figure 1, see section on et al. 2003, Vera Cruz & Brown 2007). Visual and supplementary data given at the end of this article). The chemical communication is essential to establish social plot of the body weight against the quotient between GSI hierarchy in Nile tilapia and d-males can be identified of any male from one community and the corresponding easily by white body colour and a paler coloured iris and d-male’s GSI revealed that the d-male was not necess- sclera (Giaquinto & Volpato 1997, Volpato et al. 2003, arily the largest animal of a community (Fig. 2). Machnes et al. 2008). Interestingly, the GSI of the subordinates varied within Comparative studies between dominant and subor- a broad range from values near the GSI of dominants to dinate Nile tilapia males with detailed cytological very small values (Supplementary Figure 1, see section description of gonads are not available to date. To fill on supplementary data given at the end of this article). this gap and to link social status to morphological and The increased testis weight from d-males was the result functional features of the gonads we compared testes of a larger width but not length of the organ (Fig. 1D, from long-term stable d- and s-males by histological Supplementary Table 1, see section on supplementary criteria and by comparing the expression of key genes. data given at the end of this article). Both, d- and s-males, exhibited numerous Vasa- positive germ line cells in cysts. Also, the large single Results germ line cells were specifically stained. Most of them were located close to the TA (Fig. 1E and F). The testis in Dominant and subordinate males in communities with d-males showed large groups of interstitial cells and the stable social hierarchy lumen of the spermatogenic tubules appeared wider We have evaluated 20 Tilapia communities and observed the behaviour and measured body and testis weight from a total of 20 dominant and 72 subordinate 1.0 37 58 * males. The observation of a community started with * 0.8 * about 15 pubescent, 4–5 months old fish in one tank. The * * percentage of females in the communities varied 59 between 30 and 70%. Within the next 4 weeks the 0.6 social hierarchy was established and aggressive 39 behaviour of individuals was frequently observed. 0.4 During this time, males with paler body colour appeared which is a typical attribute of dominance. At the end of 0.2 41 # # # the first week only one white d-male was present and # # 0.0 defended a territory around the bottom of the tank. All Quotient (GSI s-male)/(GSI d-male) other members of a group stayed close to the water surface. In every tank we always observed the formation 0 500 1000 1500 2000 2500 of stable social structures (Fig. 1A). The d-male Body weight (g) developed the typical eye colour pattern with pale iris Figure 2 Gonadosomatic indices (GSIs) and body weight data of all and sclera (Fig. 1B and C) and randomly attacked all examined animals shown as the quotient of (GSI)/(GSI d-male of other individuals. Only d-males established a spawning the same community) plotted against the body weight. Therefore, the site. A tube at the bottom of the tank was preferentially value for the d-male in each community is 1. Filled symbols represent used for mating (Fig. 1A). Spawning and successful d-males, open symbols represent s-males. Green squares indicate the community that was used for plasma 11-KT measurement (Table 2). reproduction occurred throughout the year. We Triangles in blue and red indicate the two communities that were used observed that the dominance of a particular male can for qRT-PCR and histology in parallel. Symbols highlighted with ID persist for more than 2 years. However, changes in numbers are the samples analysed by qRT-PCR in this study (Fig. 5). dominance also occurred occasionally followed by the The numbered blue triangles correspond to specimens investigated establishment of a new dominant–subordinate relation- in parallel by IF and TEM. Symbols highlighted with asterisks are ship. The replacement was accompanied by a slow ascending males with GSI values near to the d-male. Males with change in the eye colour pattern over several days. The completely degenerated testes (#) show low quotients. A further diagram (GSIs plotted against age) is shown as Supplementary Figure 1, animals chosen for sampling were between 0.5 and 3 see section on supplementary data given at the end of this article. years old and remained at least for 5 months in the status All GSI values and ID numbers are presented in Supplementary Table 1, of a d-male. The d-male was compared with one or see section on supplementary data given at the end of this article. www.reproduction-online.org Reproduction (2012) 143 71–84

Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access 74 F Pfennig and others

d-male s-male The quotient between tubules and cysts was identical A D for d- and s-male. Interestingly, the percentage of the area with interstitial cells, most likely LC, comprised Sz 44% in d-male and only 4% in the s-male. In our samples e Sc A St from subordinates we also noticed different degrees of Sc St testis degeneration. This ranged from some liquid-ﬁlled Ae Vasa cysts on the testicular surface to a complete germ cell- Sz B depleted testis in ﬁve cases (Fig. 1D and Supplementary Cy B Cy Figure 2, see section on supplementary data given at the A A end of this article).

BELC Gene expression analysis of testes from dominant and * subordinate males by qRT-PCR LC We compared the relative transcript levels between d- * and s-males of two different communities by qRT-PCR

Brightfield/DNA Lu (Fig. 5). Selected marker genes (Table 1) of the germ line cy cy (vasa, sox2 and dmc1), of somatic cells (amh, amhrII

CFLC d-male s-male

* A C ery LC ery * LC ery

Vasa/PCNA/DNA Lu * *

TA TA Figure 3 Immunohistological analysis of Vasa- and PCNA protein expression in testis cross-sections (4 mm) from Nile tilapia d- (ID58; B D A–C) and s-males (ID59; D–F). The orientation of the cross-sections is LC LC as shown in Fig. 1D with the efferent ducts located on the dorsal side (top). (A and D) Vasa protein (red) is present in cysts and large, single cy * germ line cells. The fluorescence intensity is decreased from * cy spermatogonia (A and B) to spermatocytes (Sc) and is lost in spermatids (St). The subcellular localisation of Vasa is cytoplasmic. (B and E) ery Brightfield picture overlaid with DNA staining (DAPI, converted to black) to illustrate the testis structure consisting of tubules (dashed line ery in A and C) and interstitium (* in B–F) including Leydig cells (LC). myo Lumen (Lu) of a spermatogenic tubule. (C and F) Identification of proliferating cells by PCNA staining (green); Vasa is shown in red and ery DNA in blue (DAPI). PCNA is located in the nuclei and appears from green to light blue due to the overlay with DNA staining in blue with TA TA different fluorescence intensities. Most spermatogonia are positive for PCNA. PCNA-labelled Sertoli cells are marked with a white arrow. Figure 4 Immunolocalisation of Cyp11b protein (green fluorescence) in Nuclei of early spermatogonia (Ae) stained very faintly (DAPI). testes cross-sections (4 mm) from d- and s-males showing the region Erythrocytes (ery) appear in a bright green due to endogenous close to the tunica albuginea (TA). DAPI staining of DNA (blue) is fluorescence (C and F). Scale bar in F: 20 mm (all pictures). merged with the green channel. (A and B) Testis from a d-male (ID58) at different magnifications. Myoid cells in different orientations (myo) and interstitial cells show strong Cyp11b signals. (C and D) Testis from an s-male (ID59) in comparison. Only few myoid cells were stained in than in s-males (Figs 1D, 3, 4 and Supplementary s-males. Bright green dots in C are autofluorescing erythrocytes (ery). Figure 3, see section on supplementary data given at Erythrocytes in B and D appear in orange because of the additional the end of this article). Furthermore, the TA was overlay with the red channel to show autofluorescence. In part D, a expanded to a multi-layered structure in all d-males single Cyp11b-positive cell (myoid boundary cell) from s-male can be (Fig. 1E) that were dominant for at least 5 months. seen (white arrow). Leydig cells (LC), spermatogenic cysts (cy), interstitium (*), border of tubules (dashed line) and myoid boundary We counted the absolute number of tubules and cells (white arrow). Lumen (Lu) of a spermatogenic tubule. The cysts in representative cross-sections shown in Fig. 1D. dimension of the TA in B and D is highlighted (dotted line). Bars: 50 mm In d-males about 30% more tubules were found. (A, also valid for C) and 10 mm (D, also valid for B).

Reproduction (2012) 143 71–84 www.reproduction-online.org

Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access Social status and testis structure in tilapia 75

Soma Germ line amh amhr2 dmrt1 sf1 cyp11b cyp19a1 vasa sox2 dmc1 S1 S2 S3 S1 S2 S3 S1 S2 S3 S1 S2 S3 S1 S2 S3 S1 S2 S3 S1 S2 S3 S1 S2 S3 S1 S2 S3 16 8 4 2 1 d-males =1 0.5 Fold change in s-males 0.25 0.125

Figure 5 Comparison of testis mRNA levels from Nile tilapia d- and s-males by quantitative real-time PCR (qRT-PCR) with selected marker genes (Table 1). Two d-males and three s-males were analysed. Transcript levels of s-males (s1, s2 and s3) are calculated with reference to the corresponding d-male. d-male ID37 was compared with s-male ID39 (s1) and s-male ID41 (s2) and d-male ID58 was compared with s-male ID59 (s3). GSIs of analysed specimens are marked in Fig. 2 and Supplementary Figure 1, see section on supplementary data given at the end of this article. Error bars represent S.D. and dmrt1) and typical for steroidogenic cells (sf1, cyp11b Zhang et al. 2010). In s-males sf1 and cyp11b expression and cyp19a1) were generally expressed in all samples. was low in contrast to Sertoli cell-specific genes. We also The genes vasa and sox2 are expressed in spermato- estimated aromatase expression (cyp19a1) since the gonia and spermatocytes of Nile tilapia (Kobayashi et al. enzyme also plays a role in testis development (reviewed 2002, Supplementary Figure 4, see section on supple- in Schulz et al. (2010)). Transcript levels of cyp19a1 were mentary data given at the end of this article) while dmc1 increased in s-males. In summary, the gene expression is expressed only in meiotic cells (Kajiura-Kobayashi profile indicated normal spermatogenesis in d- and et al. 2005). In s-males the relative abundance of the s-males, respectively, while cyp11b necessary for 11-KT analysed germ line-specific genes was higher than in production was elevated in d-males. d-males (Fig. 5). We also investigated one of the s-males with a completely degenerated testis (Fig. 1Dand Spermatogenesis in dominant and subordinate males Supplementary Figure 2D, see section on supplementary data given at the end of this article). There were no vasa To evaluate the spermatogenic activity in d- and s-males or dmc1 transcripts detectable and sox2 expression was and to identify a possible arrest of spermatogenesis in extremely weak (data not shown) indicating the absence s-males we analysed cross-sections of 8 d-males and of germ line cells. 18 s-males histologically and by immunofluorescence Amh, its type II receptor AmhrII and Dmrt1 are microscopy (IF). In Fig. 3, we show sections of typically expressed in Sertoli cells of teleosts (Klu¨ver representative d- and s-males that were also analysed et al. 2007, Supplementary Figure 6, see section by qRT-PCR and EM in this study (ID58 and ID59). We on supplementary data given at the end of this used antibodies specific for pre-meiotic germ line cells article, Kobayashi et al.2008). Transcript levels of (Vasa) and for mitotic cells (PCNA, p-Histone H3). The all three genes were reduced in testes from d-males. estimation of the nuclear diameter after DNA staining Steroid biosynthesis is a typical LC function. We allows the identification of different germ line cell types. analysed sf1 and cyp11b transcript levels to study the But secondary spermatogonia and spermatocytes from androgen-producing potential (Yoshiura et al. 2003, leptotene to diplotene show similar nuclear diameters

Table 1 Primer pairs (50–30) used for quantitative real-time RT-PCR analysis.

Gene Database Forward primer Reverse primer Size (bp) amh EF512167 AACGTACGACGCGCAAAGa TCTCGATGTGGGAGTTGAGCa 215 amhrIIb contig32c ACCTGAACACTGGGAACTGCa CATCCTGAAAAGACAAGAGACAGAa 173 dmrt1 AF203489 ACTACCCAGCAGCCACCTACa AACGTGAAGACAGCGTGAAGa 395 sf1 AB060814 TACGATCCAACGGTTTCAAGa ACTGGTACTGGGTGGTGAGGa 231 cyp11b FJ713105 CAAAGAAGTCCTCAGGTTGTACCCAd GGACCAAAGTTCCAGCAGGTATGTd 104 cyp19a1 AF472620 TAGACGCTGTTGTGGGTGAGa TCTACAGGCTGCTGGGAAAGa 442 vasa AB032467 CGATGAGATCTTGGTGGATGa CATGAGATCCCTGCCAGCAGAe 175 sox2 EF431920 TCCTACTCCCAGCAAACCACf GCACGGTGCTCTGGTAGTGf 247 dmc1 AB182646 GAAGATGACTCTAGTTTCCAGGATGATGa GCAGAGATGGCAAACGTAGCCa 974 18SrRNA AF497908 TGAAATTCTTGGACCGGCGCa CCAGACAAATCGCTCCACCAACa 407 aThis work. bFurther information was given in Supplementary Figure 6, see section on supplementary data given at the end of this article. cRBEST http://reprobio.nibb.ac.jp. dIjiri et al. (2008) cyp11b, long form. eKobayashi et al. (2002). fCnaani et al. (2007). www.reproduction-online.org Reproduction (2012) 143 71–84

Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access 76 F Pfennig and others and for this reason we used the higher Vasa concen- reduced number of LC and spermatogenic tubules we tration in spermatogonia as an additional criterion expect a reduced potential for spermatogenesis in for identification. Double IF staining with anti-Vasa/ s-males compared with d-males. anti-PCNA and anti-PCNA/anti-p-Histone H3 antibodies revealed mitotically active germ line cells and Sertoli cells in both d- and s-males. PCNA was detected Dominant males exhibit large groups of LC and a thickened TA generally in cyst-forming spermatogonia, but also in some Sertoli cells mostly located around primary LC and myoid cell were identified based on their spermatogonia or early secondary spermatogonia characteristic ultrastructural features. In d-males LC (Fig. 3). Spermatids were most frequently positive for form large clusters outside the seminiferous tubules p-Histone H3 (Supplementary Figure 5, see section (Fig. 6A and B) and show characteristic smooth on supplementary data given at the end of this article). endoplasmic reticulum and mitochondria with tubular In summary, we have identified mitotic spermatogonia, cristae (Fig. 6C and D) indicating active steroid meiotic spermatocytes and haploid spermatids and production (Nakamura & Nagahama 1989). LC are spermatozoa in the testes of d- and also in s-males located in close proximity only to interstitial myoid with GSI O0.07. We have not seen obvious differences cells or to other LC (Fig. 6E) as has been observed in in the relative abundance of specific germ cell stages to Medaka testes (Gresik et al. 1973). each other. Only s-males with germ cell-depleted testis The thickened TA of the testis from d-males contained and the lowest GSIs had apparently abolished sperm- several cell layers of elongated myoid cells (Fig. 6F). atogenesis. Our histological analysis revealed that most These cells are typical smooth muscle cells as shown by s-males showed all stages of spermatogenesis and the presence of electron-dense myofilaments, dense proliferating cells, but the dynamics of spermatogenesis bodies, dense plaques and caveolae at their cell cannot be assessed by the available morphological data. periphery (Fig. 6G and H). The myoid network from Considering the lower GSI in s-males caused by the the TA appears to be connected with myoid cells from

A B cap LC cy LC LC cap cy LC cy LC

C D E Figure 6 Histology and ultrastructure of a testis from a dominant Nile tilapia male (ID58). (A and B) Histology, 1 mm epon sections stained with toluidine LC sER LC blue. (A) Overview: the thickened periphery myo (arrows), cysts (cy) and groups of Leydig cells (LC) are shown. (B) A group of LC at higher magni- nuc m fication; cap, capillary. (C–H) TEM images. (C and LC D) Ultrastructure of LC. The marked region in C is LC shown at higher magnification in D. LC have extensive smooth endoplasmic reticulum (sER), mitochondria (m) with tubular cristae; nucleus F G col (nuc). (E) Group of LC are often found in close myo dp myf proximity to myoid cells (myo). (F) Myoid cells in the dp periphery of the testis. Two cysts are indicated (cy). TA Some longitudinally sectioned (yellow) and some cross-sectioned (red) myoid cells are highlighted. nuc cy myf Dimension of tunica albuginea (TA) is marked by a dashed line. (G) Myoid cells at higher magnification display myofilaments (myf), dense bodies (db) and H m myf dense plaques (dp), hallmarks of smooth muscle ecm cells. The cells are surrounded by ECM rich in cy collagen (col). (H) Caveolae (arrowheads) at the periphery of myoid cells. Bars: 100 mm (A), 10 mm myf (B and F), 5 mm (C and E) and 500 nm (D, G and H).

Reproduction (2012) 143 71–84 www.reproduction-online.org

Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access Social status and testis structure in tilapia 77

* * cy * cy cy Figure 7 Histology and ultrastructure of a testis from cy a subordinate Nile tilapia male (ID59). (A and B) cy 1 mm epon sections stained with toluidine blue. C cy D (A) Cysts are embedded in a loosely organised ery fibrous interstitium with many gaps (*); no Leydig * cells (LC) can be seen in that area. The square myo indicates the region displayed in (B) at higher ery magnification. (B) Tubules with cysts and the * myo myo * * cy interstitium between them; granule rich cells (possibly mast cells) are abundant (arrows). (C and cy myo * D) TEM micrographs of interstitial tissue with myoid col cells (myo), red blood cells (ery) and numerous gaps * (*), possibly indicating the loss of LC. Dense bodies E H are not recognisable in the myoid cells. (E and F) F mp G Phagocytotic cells (probably macrophages, mp) in mp * ery the interstitium. (E) light micrograph, the box * * indicates the region displayed in F. (F) TEM * micrograph; capillary (cap). (G and H) Mast cells cap * * mp ery (arrows) in the interstitium. (G) Light micrograph. * * (H) TEM micrograph. Bars: 100 mm (A), 50 mm (B), 10 mm (C, E and G), 500 nm (D) and 5 mm (F and H). the interstitium (Fig. 6F). LC are not part of the TA. In used a specific antibody to determine the localisation of contrast, s-males exhibit always a thin TA consisting of Cyp11b expressing cells by IF. Furthermore, we used one an epithelium and a thin subjacent layer that also d- and one s-male, respectively, to quantify the number includes some myoid cells. of Cyp11b-positive cells by flow cytometry and the The testis of s-males is further characterised by a amount of Cyp11b protein by immunoblotting. reduced number of LC (Fig. 7A and B). In s-males large In paraffin sections a strong cytoplasmic Cyp11b parts of the interstitium were loosely organised (Fig. 7D). signal in the vicinity of the nucleus was apparent in the Between myoid cells covered by fibrous material cells of the thickened TA and around the tubules in the numerous gaps appeared, possibly as the result of LC testis of d-males (Fig. 4A). Our ultrastructural analysis loss (Fig. 7C and D). Macrophages and mast cell-like revealed myoid cells as the most prominent cell type of cells were present in the interstitium that was not seen the thickened TA (Fig. 6F). The Cyp11b-positive cells of in d-males (Fig. 7E–H). That observation indicates the TA show the typical elongated myoid cell shape inflammatory reactions and/or tissue remodelling in the (Fig. 4B) while in transversal sections they have a regular testis of s-males. spherical shape (Figs 4B and 6F). Therefore, we conclude that the Cyp11b-positive cells of the TA represent myoid cells. In testes from s-males the thickened periphery was d-males show Cyp11b expressing myoid cells and not observed and Cyp11b-positive myoid cells were rare elevated amounts of Cyp11b protein in the entire testis (Fig. 4C and D). Controls devoid of the To analyse the 11-KT producing potential of testes from primary antibody were without fluorescence (not d- (three individuals) and s-males (six individuals) we shown). Cyp11b was also detected in cells of the

AB 1200 Msddsd s 960 Figure 8 Flow cytometry of Cyp11b expressing cells kDa (A) and immunoblotting (B). (A) Histogram of primary 720 testis cells shows intensity of immunostaining with anti- 55 Cyp11b antibody and number of cells with equivalent 480 Counts value. d-male data (green, GSI: 0.27, ID69) show more 40 240 cells inthe area of high ﬂuorescence incomparison to the SDS–PAGE γ-Tub. Cyp11b s-male (purple, GSI: 0.15, ID70). (B) Immunoblotting of 0 testis extracts from a d- and s-male (ID69 and ID70). 1 10 100 1000 A Cyp11b speciﬁc band (60 kDa) appears in both Cyp11b samples but with higher intensity for the d-male. www.reproduction-online.org Reproduction (2012) 143 71–84

Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access 78 F Pfennig and others

Table 2 Plasma 11-ketotestosterone (11-KT) levels in a socially stable sp. in nature. Typically, d-males occupy a territory and Nile tilapia community. defend it for several weeks (Turner & Robinson 2000). We compared the GSIs from O. niloticus d- and Plasma 11-KT (ng/ml) GSI s-males and in general, the d-males showed twice the d-male 15.4 0.41 values as in s-males and only ascending candidates s1 1.5 0.132 s2 2.0 0.149 reached values close to the d-male of a community. s3 3.6 0.168 Interestingly, the d-male of a community with the highest C s4a(t 3h) 23.3 0.312 GSI was not necessarily the male with the largest body Female 1.7 ND weight. Lower social status is associated with suppressed s1–s3: s-males; s4a: ascending male (3 h after removal of the d-male). growth, probably caused by decreased hepatic IGF1 values and reduced food intake (Vera Cruz & Brown connective tissue layer around epithelial cells of efferent 2007). On the other hand, a d-male needs much more ducts but there was no apparent difference between energy to establish or to defend a territory than the d- and s-males (not shown). others. Obviously, several phenomena seem to interfere To quantify Cyp11b-positive cells we used primary with each other (Vera Cruz & Brown 2007). We can testis cells for flow cytometry. Specific binding of anti- exclude food limitations for s-males in our long-term Cyp11b antibody was detected with an EGFP labelled communities because the animals were fed ad libitum. secondary antibody. Amongst the testis cells of a d-male For A. burtoni it was reported that non-territorial s-males 58% of the cells showed a positive Cyp11b reaction in schools together with females and in absence of while the number for an s-male was 21% (Fig. 8A). In d-males show faster growth than territorial d-males compliance with these data immunoblotting of testis (Hofmann et al. 1999). The body weight is apparently extracts revealed higher levels of Cyp11b protein in not a prerequisite to become a d-male, but early d-males compared with s-males (Fig. 8B). experiences as dominant individual with territoriality and with courtship are more essential (Fraley & Fernald Measurement of plasma 11-KT levels in d- and s-males 1982, Huntingford et al. 1990). In our long-term communities we have observed that To prove an elevated 11-KT biosynthesis in d-males we social stress was sufficient to cause fin disorders, measured the plasma 11-KT concentration in several increased skin mucus and occasionally gall bladder s-males and the corresponding single d-male (Table 2). In hypertrophy in s-males with low GSI (F Pfennig 2010, this d-male the 11-KT value was 15.4 ng/ml while in unpublished observations). Pathological reactions such s-males the 11-KT values were much lower (1.5, 2.0 and as bile retention and immunosuppression were also 3.6 ng/ml) and close to the level of a female control reported for cichlid s-males (Faisal et al. 1989, Earley et (1.7 ng/ml). These data corroborate the observed higher al. 2004). Permanent stress affects testis development Cyp11b expression and the increased number of negatively (Wendelaar Bonga 1997, Celino et al. 2009, Cyp11b-positive cells in d-males. Schreck 2010, Trubiroha et al. 2011) and this conclusion We observed in all communities that apart from the is also supported by the observation that temperature white coloured d-male one or more males existed that decrease leads to spermatogenic arrest at late pachytene were lighter coloured than others. Such s-males were spermatocytes in sexually mature Nile tilapia males also characterised by comparatively high GSI (Fig. 2 and (Vilela et al. 2003). Supplementary Figure 1, see section on supplementary Our experiments revealed ongoing spermatogenesis data given at the end of this article) and might become an and proliferating testis cells also in s-males and no cell ascending male after removal of the d-male. We were cycle arrest was apparent. The maintenance of sperm- able to identify the new d-male in the population after atogenesis in s-males may enable them to reproduce by a about 30 min because this male assumed a greyish-white ‘sneaking’ strategy. The fact that the s-males in a colour in an otherwise black–coloured community and showed the typical aggressive behaviour of d-males. In a community are characterised by testes in a wide range single case we determined the 11-KT concentration 3 h of physiological states allows to quickly replace d-males after ascent in a new d-male and obtained the high value if the necessity arises (Goncalves-de-Freitas & Nishida of 23.3 ng/ml. 1998). Complete testis degeneration in s-males and germ cell depletion was seen in 5 out of 72 cases. We speculate that those s-males were highly suppressed over Discussion a long time and such individuals should have low FSH plasma levels (Weltzien et al. 2004, Levavi-Sivan et al. Social hierarchy and spermatogenesis in d- and s-males 2010, Schreck 2010, Maruska et al. 2011). FSH seems to In our study, we observed a stable social hierarchy be necessary for gonial proliferation and has anti- among Nile tilapia reared in 800-l aquaria, and such a apoptotic activity on germ line cells in fish (Ohta et al. social structure has also been observed for Oreochromis 2007, Iwasaki et al. 2009, Garcia-Lopez et al. 2010).

Reproduction (2012) 143 71–84 www.reproduction-online.org

Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access Social status and testis structure in tilapia 79

Histological and gene expression studies give evidence germ cells in s-males do not reflect an up-regulation in for different steroidogenic potential of d- and s-males the respective cell types. Since the number of germ line cells depends on the number of Sertoli cells (Schulz et al. The number of steroid-producing LC is posi- 2005) the marker genes for both cell types show the tively correlated with spermatogenesis in teleost fish same trend in the RT-PCR analysis. The reduced (Kanamori et al. 1985, Loir et al. 1989, Nobrega & expression of cyp11b in s-males is in accordance with Quagio-Grassiotto 2007). We demonstrated that stable the histological analysis and confirms the lower d-males had testes with large groups of LC. Our steroidogenic potential of s-males. Further studies by estimation of the areas occupied by different testis cell other techniques such as in situ hybridisation, laser types revealed a very high proportion (44%) of LC in dissection or histomorphometry are necessary to quan- d-males. A similarly high value was reported only for tify the cell type-specific expression of genes of interest. Myleus ternetzi at mature stage (Loir et al. 1989). In s-males, this value dropped to 4%, probably due to a loss of LC. Inflammatory reactions and tissue remodelling Role of 11-KT in stable d-males and subsequent regression of the fish testis are known We determined the plasma 11-KT values and measured from seasonal changes in other teleosts (Loir et al. 1989, tenfold higher level in d-males than in s-males. This Koulish et al. 2002, Lo Nostro et al. 2004, Chaves-Pozo observation is in compliance with data from the cichlids et al. 2005). Macrophages, which typically accompany A. burtoni and Pundamilia nyererei (Parikh et al. 2006, these processes, were particularly abundant in s-males. Dijkstra et al. 2007). We presume that d-males in stable The second striking difference in testis structure Nile tilapia communities have elevated plasma 11-KT between d- and s-males was the thickened TA in due to the social status and the permanent presence of d-males. Most likely the myoid cells of the TA formed a potential challengers. That is in agreement with the contractile network with peritubular myoid cells, a ‘challenge hypothesis’ that predicts higher androgen structural arrangement referred to as ‘Y’ zone by Koulish levels during the course of territory establishment and et al. (2002). This structure was also observed in our defence amongst others (discussed by Oliveira et al. histological analysis. A contractile testis wall presumably (2002) and by Maruska & Fernald (2010)). squeezes out the mature sperm during mating. Contrac- In our study, we investigated the long-term effect of tile properties of the TA are also known from mammals social dominance in Nile tilapia males. This system (Banks et al. 2006). We can only speculate about the allows the research of the naturally originated hierarchy. factors that trigger the strong development of myoid cells This situation is in contrast to the artificial pairing studies in d-males, but androgens are likely candidates because with A. burtoni males (Maruska & Fernald 2010). We of their known stimulating effect on muscle development were also able to identify ascending males that will and growth in vertebrates (Schulz et al. 2010). become the new d-male after removal of an actual Remarkably, the myoid cells of the TA in d-males d-male. In undisturbed communities the replacement of expressed Cyp11b, necessary for 11-KT biosynthesis. a d-male by another male is rarely seen. In these rare Therefore, the myoid cells could be an additional source cases the change in dominance takes several days for 11-KT apart from the numerous LC. This cytological whereas the experimentally induced change proceeded observation is corroborated by the elevated cyp11b within 30 min. We did not determine the 11-KT transcript levels, a twofold increase in the number of concentration before removing the d-male, but 3 h Cyp11b expressing cells estimated by flow cytometry, after removal a high 11-KT level was found in the and the detection of higher amounts of Cyp11b protein ascending male. We assume that two effects contributed estimated by immunoblotting. Finally, a tenfold increased to this high 11-KT level: first, the ascending male is plasma 11-KT level in a representative d-male confirmed recruited from males with relatively high GSI close to the the higher androgen production in d-males. d-male and hence the number of LC and steroidogenic We showed that in s-males transcript levels for Sertoli capacity is already elevated. Secondly, a quick ascent in cell genes such as amh, amhrII and dmrt1 and for germ social status is accompanied by a rapid elevation of line-specific genes were higher than in testes of d-males plasma 11-KT values in A. burtoni (Maruska & Fernald although the histological analysis revealed a reduced 2010) and this might also hold for Nile tilapia. number of spermatogenetic tubules in s-males. The Immature Sertoli cells around the spermatogonia are reason for this discrepancy presumably is the strongly the source for Amh in fish testis (Miura et al. 2002, different representation of specific cell types in s- and Yoshinaga et al. 2004, Vizziano et al. 2008). One effect d-males and the quantification of the transcript level of 11-KT (or testosterone in mammals) in a complex with reference to housekeeping genes. Due to the large regulative network is the down-regulation of Amh number of LC and myoid cells in d-males the number of (Al-Attar et al. 1997, Miura et al. 2002, Rey et al. Sertoli cells represents a smaller fraction of the total 2003). Amh, in turn, inhibits LC development and number of testis cells in d-males compared with s-males. androgen synthesis in mammals (Josso et al. 1998, Hence, the high transcript concentrations for Sertoli- and Fynn-Thompson et al. 2003, Salva et al. 2004). This www.reproduction-online.org Reproduction (2012) 143 71–84

Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access 80 F Pfennig and others inhibition might also hold for the fish testis but it seems and dilutions were used for IF (and IHC): anti-zfVasa purified likely that Amh acts locally since at the distant spermatic antiserum (rabbit, Knaut et al. 2000) 1:100 (1:1000 for IHC); ducts we observed numerous LC as a result of active anti-PCNA MAB (mouse, M0879; DAKO, Glostrup, Denmark) proliferation (Schulz et al. 2005). 1:100; anti-phospho-Histone H3 – protein A purified IgG- This study provides information about the structural antiserum (rabbit, #06-570; Abcam, Cambridge, UK) 1:100; differences in testis development between d- and s-males anti-Sox2 affinity purified antiserum (rabbit, NB110-79875; of an economically important fish species. Differently Acris, Herford, Germany) 1:500 and anti-(Medaka)-Cyp11b1 developed gonads as a function of social cues make Nile antiserum (rabbit, ab71561, Abcam) 1:50. Slides were washed tilapia males a promising model for studying the function with PBST and the primary antibodies were detected for IF with of gonadal hormones and their gonadotropic control. appropriate secondary antibodies conjugated with Alexa 488, DyLight 488 or Rhodamine Red-X (Invitrogen or Jackson ImmunoResearch, West Grove, PA, USA). After washing with Materials and Methods PBST, DNA was stained with DAPI (1:20 000; stock solution Fish maintenance and sampling 1 mg/ml in DMSO) and mounted with Vectashield (Vector Laboratories, Burlingame, CA, USA). Anti-Vasa and anti-Sox2 Nile tilapia (O. niloticus L.) were kept in circulating 800-l antibodies were also detected immunohistochemically using a freshwater glass tanks (AQUA SCHWARZ GmbH, Go¨ttingen, biotinylated anti-rabbit antibody (#BA-1000, Vector) 1:1000, Germany) at 26 8C and under constant photoperiod cycle (14 h followed by binding of streptavidin-conjugated HRP (1 mg/ml, light:10 h darkness). Five per cent of the water was exchanged Jackson ImmunoResearch) 1:5000 and terminal staining with daily and the conductivity of the water was adjusted to 450 ms the substrate NovaRed (Vector Laboratories) according to the by addition of Coral Pro Salt (Red Sea Europe, Verneuil s/Avre, manufacturer’s instructions. Methylgreen (1:100; stock solution France) to deionised water. Fish were fed ad libitum (around 1% v/v in 20% v/v EtOH) was used for counterstaining. Sections 1% of biomass) once daily with commercial trout pellets were analysed in a fluorescence microscope (Axiolab, Zeiss) (Trouvit F3P Select, Skretting; Trouw Nutrition Deutschland, with Neofluar objectives (Zeiss). Documentation was done with Burgheim, Germany). One community consisted always of Spot cam model 1.40 and the software SpotAdvanced 4.0.9 siblings from one set of parents. There was no selection for size (Diagnostic Instruments, Heights Sterling, MI, USA). Overview or sex before the experiment. The killed d-males were at least 5 pictures of sections were taken on a BZ-8100E microscope months continuously dominant. The behaviour and social (Keyence) with Plan Apo objectives (Nikon) and assembled with status were controlled daily. We identified each animal by its the BZ analyser software (Keyence). Software Image J 1.41o with individual fin shape. Animal culture was performed according plugin MBF ‘Image J for microscopy’ (free software) was used for to the German national regulations and animal welfare. quantification of areas with defined cell types on testis sections Fish were stunned mechanically before they were killed by (Vasa IHC). decapitation. All animals of a community used for experimental analysis were killed at the same time. The weight of body and testis was estimated and gonadosomatic indices were Epon histology and electron microscopy calculated (GSIZgonad weight/body weight!100; plots were After dissection the testes were fixed in modified Karnovsky’s created with Origin 8.5; OriginLab, Northampton, MA, USA). C Testes were removed immediately and sliced into fragments of fixative (2% v/v glutaraldehyde 2% v/v paraformaldehyde in 8 5 mm thickness from the central part of the organ. Specimens 50 mM HEPES; Karnovsky 1965, Kurth et al. 2010)at4C ! ! were frozen in liquid nitrogen and stored at K80 8C until RNA overnight, and washed 2 in 100 mM HEPES and 2 in PBS. or protein extraction. Adjacent pieces of the same size were The specimens were postfixed with 1% w/v OsO4/PBS for 2 h prepared for IF, immunohistochemistry (IHC) and epon histology on ice, washed with PBS and water and en bloc contrasted with (see below). For flow cytometry, a whole testis was used (see 1% w/v uranyl acetate in water. The samples were then washed below). Blood samples were taken directly from the heart. several times in water, dehydrated in a graded series of ethanol, infiltrated in epon 812 (epon/ethanol mixtures: 1:3, 1:1, 3:1 1.5 h each, pure epon overnight, pure epon 3 h) and embedded IF and IHC in flat embedding moulds. For light microscopy, semithin Testis tissue was fixed in 4% w/v paraformaldehyde in PBS sections were mounted on glass slides and stained with 1% w/v overnight at 4 8C. After dehydration through a series of toluidine blue and 0.5% w/v borax. For electron microscopy, increasing alcohol concentrations the specimens were ultrathin sections were collected on formvar-coated slot grids, embedded in paraffin, sectioned at 4 mm and transferred to stained with lead citrate and uranyl acetate according to L-lysine coated slides. Paraffin cross-sections were deparaffi- Venable & Coggeshall (1965) and analysed on a FEI Morgagni nised and successively rehydrated through xylol/ethanol series. 268 at 80 kV. Antigen unmasking was performed in 10 mM citrate buffer (18 ml of 100 mM citric acid, 82 ml of 100 mM sodium citrate Immunoblotting and 900 ml deionised water, pH 6) in a pressure cooker for 7 min, followed by three washing steps with PBS. Unspecific RIPA buffer (Boehringer Mannheim) including protease binding sites were blocked with 3% w/v BSA in PBST (PBS with inhibitor mix (Roche, cOmplete ULTRA Tablets, Mini, EDTA- 0.1% v/v Tween 20). The samples were incubated overnight at free, EASYpack) was added to the samples (35 mg tissue per 4 8C with primary antibodies. The following primary antibodies 1 ml buffer). After homogenisation with a pestle the samples

Reproduction (2012) 143 71–84 www.reproduction-online.org

Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access Social status and testis structure in tilapia 81 were sonicated twice in an ice bath. After centrifugation for RNA concentrations and purity were measured by spectro- 10 min at 14 000 g and 4 8C, the samples were partitioned in photometry (Nanodrop 1000, Peqlab). Residual genomic DNA aliquots, frozen in liquid nitrogen and stored at K80 8C until was enzymatically eliminated with Desoxyribonuclease I use. Sample aliquots were separated by SDS–PAGE under (Ambion) and completeness of removal was examined by reducing conditions and blotted onto nitrocellulose membrane PCR. MMLV reverse transcriptase (Life Technologies) and according to standard protocols (Rosenberg 2005). Unspecific random hexamers were used for the first-strand cDNA binding sites were blocked with 5% instant non-fat dry milk synthesis. qRT-PCR analysis was performed in three indepen- solution in TBST (TBS with 0.1% v/v Tween 20). Blots were dent runs as described elsewhere using 18SrRNA as an incubated with rabbit polyclonal Cyp11b1 antiserum (see endogenous reference and SybrGreen1 to detect PCR products above, dilution 1:500) or for controls with mouse monoclonal (Kretzschmar et al. 2005). The amount of cDNA per PCR anti-g-tubulin-antibody (T5326, Sigma–Aldrich, 1:5000) over- reaction corresponded to 65 ng of total RNA. The primers that night at 4 8C. After washing specific binding was detected using have been used are listed in Table 1. Properties of designed a HRP-conjugated anti-rabbit or anti-mouse IgG, respectively, primers were examined with AnnHyb version 4.936 (free and visualised by enhanced chemiluminescent detection (ECL software). Results were expressed as relative amounts of KDD Detection Kit, Amersham). mRNA compared with the d-male using the 2 Ct formula (Winer et al. 1999). Flow cytometry Whole testes were minced mechanically with a scalpel in Supplementary data prewarmed Hank’s Balanced Salt Solution (Biochrom, Berlin, Germany) and resulting tissue fragments were treated with This is linked to the online version of the paper at http://dx.doi. 0.34 mg/ml collagenase XI (Sigma–Aldrich) in DMEM (Biochrom) org/10.1530/REP-11-0292. for 2 h at 27.5 8C. After incubation, cells were centrifuged and washed with prewarmed Hank’s Balanced Salt Solution (Biochrom). Aggregates and debris were removed by centri- Declaration of interest fugation for 10 min at 125 g and by filtration steps with different The authors declare that there is no conflict of interest that pore sizes (100, 50 and 20 mm). After washing, the cells were could be perceived as prejudicing the impartiality of the harvested by centrifugation. For immunostaining testis cells research reported. were resuspended in PBS and fixed with 2% w/v paraformaldehyde for 45 min and permeabilised with 0.2% v/v Tween 20 added for the last 15 min of fixation. Cells were centrifuged Funding (10 min at 220 g) and washed two times with PBST. Cells were resuspended in PBST including 3% w/v BSA and incubated for This work was supported by the Deutsche Forschungs- 45 min, then harvested by centrifugation and washed once. gemeinschaft (DFG Gu208/12-3). After resuspending in PBST including anti-Cyp11b1 antiserum (rabbit, ab71561, Abcam, diluted 1:500) and incubating overnight at 4 8C, cells were harvested by centrifugation and Acknowledgements washed twice with PBST. Alexa488-conjugated anti-rabbit Animals as source for our broodstock were kindly provided by antibody (Invitrogen, diluted 1:500) was added and incubated B Rennert and W Kloas from the IGB Berlin (Germany). We for 2 h in the dark. After resuspension in Hank’s Balanced Salt thank H Knaut and H Schwarz (Max Planck Institut fu¨r Solution the cells were measured immediately on CyFlow Entwicklungsbiologie, Tu¨bingen, Germany) for the generous space (Partec, Mu¨nster, Germany). Data were acquired and gift of zfVasa antibody and Ru¨diger W Schulz (Utrecht analysed using FloMax 6.0 (Partec). With reference to the University, the Netherlands) for performing the 11-KT RIA. negative control, a threshold for specific fluorescence was We acknowledge further S Richter for computation of GSI defined and data of measured cells were divided into Cyp11b- plots, Sabine Basche for help with antibody incubations and N positive and Cyp11b-negative. Zimmermann, A Schneider and Y Henker for maintenance of the fish stocks. 11-KT RIA For sampling heparinised canulae and syringes were used. Ten microlitre heparin (5.000 IE/ml) was prepipetted per 1 ml blood References and mixed immediately by inversion. After centrifugation Al-Attar L, Noel K, Dutertre M, Belville C, Forest MG, Burgoyne PS, (5 min at 1500 g and 4 8C), the supernatants were quantified Josso N & Rey R 1997 Hormonal and cellular regulation of Sertoli cell by 11-KT RIA as described by Schulz et al. (1993). anti-Mullerian hormone production in the postnatal mouse. Journal of Clinical Investigation 100 1335–1343. (doi:10.1172/JCI119653) de Alvarenga ER & de Franca LR 2009 Effects of different temperatures on RNA preparation and quantitative real-time PCR testis structure and function, with emphasis on somatic cells, in sexually (qRT-PCR) mature Nile tilapias (Oreochromis niloticus). Biology of Reproduction 80 537–544. (doi:10.1095/biolreprod.108.072827) Total RNA was prepared from testis fragments using Banks FC, Knight GE, Calvert RC, Turmaine M, Thompson CS, Trifast (Peqlab) according to the manufacturer’s protocol. Mikhailidis DP, Morgan RJ & Burnstock G 2006 Smooth muscle and www.reproduction-online.org Reproduction (2012) 143 71–84

Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access 82 F Pfennig and others

purinergic contraction of the human, rabbit, rat, and mouse testicular Francis RC, Soma K & Fernald RD 1993 Social regulation of the brain– capsule. Biology of Reproduction 74 473–480. (doi:10.1095/biolreprod. pituitary–gonadal axis. PNAS 90 7794–7798. (doi:10.1073/pnas.90.16. 105.044602) 7794) Baroiller JF, D’Cotta H, Bezault E, Wessels S & Hoerstgen-Schwark G Fynn-Thompson E, Cheng H & Teixeira J 2003 Inhibition of steroidogenesis 2009a Tilapia sex determination: where temperature and genetics meet. in Leydig cells by Mullerian-inhibiting substance. Molecular and Cellular Comparative Biochemistry and Physiology. Part A: Molecular & Endocrinology 211 99–104. (doi:10.1016/j.mce.2003.09.015) Integrative Physiology 153 30–38. (doi:10.1016/j.cbpa.2008.11.018) Garcia-Lopez A, de Jonge H, Nobrega RH, de Waal PP, van Dijk W, Baroiller JF, D’Cotta H & Saillant E 2009b Environmental effects on fish sex Hemrika W, Taranger GL, Bogerd J & Schulz RW 2010 Studies in determination and differentiation. Sexual Development 3 118–135. zebrafish reveal unusual cellular expression patterns of gonadotropin (doi:10.1159/000223077) receptor messenger ribonucleic acids in the testis and unexpected Billard R, Fostier A, Weil C & Breton B 1982 Endocrine control of functional differentiation of the gonadotropins. Endocrinology 151 spermatogenesis in teleost fish. Canadian Journal of Fisheries and 2349–2360. (doi:10.1210/en.2009-1227) Aquatic Sciences 39 65–79. (doi:10.1139/f82-009) Giaquinto PC & Volpato GL 1997 Chemical communication, aggression, Burmeister SS, Jarvis ED & Fernald RD 2005 Rapid behavioral and genomic and conspecific recognition in the fish Nile tilapia. Physiology & responses to social opportunity. PLoS Biology 3 e363. (doi:10.1371/ Behavior 62 1333–1338. (doi:10.1016/S0031-9384(97)00347-8) journal.pbio.0030363) Goncalves-de-Freitas E & Nishida SM 1998 Sneaking behaviour of the Nile Celino FT, Yamaguchi S, Miura C & Miura T 2009 Arsenic inhibits in vitro tilapia. Boletim Tecnico do CEPTA 11 71–79. (doi:10.1093/icb21.2.345) spermatogenesis and induces germ cell apoptosis in Japanese eel Gresik EW, Quirk JG & Hamilton JB 1973 A fine structural and (Anguilla japonica). Reproduction 138 279–287. (doi:10.1530/REP-09- histochemical study of the Leydig cell in the testis of the teleost, Oryzias 0167) latipes (Cyprinidontiformes). General and Comparative Endocrinology Chaves-Pozo E, Mulero V, Meseguer J & Garcia AA 2005 An overview 20 86–98. (doi:10.1016/0016-6480(73)90133-0) of cell renewal in the testis throughout the reproductive cycle of a Grier HJ 1981 Cellular organization of the testis and spermatogenesis in seasonal breeding teleost, the gilthead seabream (Sparus aurata L.). fishes. American Zoologist 21 345–357. (doi:10.1093/icb/21.2.345) Biology of Reproduction 72 593–601. (doi:10.1095/biolreprod.104. Grier HJ, Linton JR, Leatherland JF & de Vlaming V 1980 Structural 036103) evidence for two different testicular types in teleost fishes. American Cnaani A, Lee BY, Ozouf-Costaz C, Bonillo C, Baroiller JF, D’Cotta H & Journal of Anatomy 159 331–345. (doi:10.1002/aja.1001590307) Kocher T 2007 Mapping of sox2 and sox14 in tilapia (Oreochromis spp.). Hofmann HA, Benson ME & Fernald RD 1999 Social status regulates growth Sexual Development 1 207–210. (doi:10.1159/000102109) rate: consequences for life-history strategies. PNAS 96 14171–14176. Correâ SA, Fernandes MO, Iseki KK & Negrao JA 2003 Effect of the (doi:10.1073/pnas.96.24.14171) establishment of dominance relationships on cortisol and other Huntingford FA, Metcalfe NB, Thorpe JE, Graham WD & Adams CE 1990 metabolic parameters in Nile tilapia (Oreochromis niloticus). Brazilian Social dominance and body size in Atlantic salmon parr, Salmo solar L. Journal of Medical and Biological Research 36 1725–1731. (doi:10. Journal of Fish Biology 36 877–881. (doi:10.1111/j.1095-8649.1990. 1590/S0100-879X2003001200015) tb05635.x) Davis LK, Pierce AL, Hiramatsu N, Sullivan CV, Hirano T & Grau EG 2008 Ijiri S, Kaneko H, Kobayashi T, Wang DS, Sakai F, Paul-Prasanth B, Gender-specific expression of multiple estrogen receptors, growth Nakamura M & Nagahama Y 2008 Sexual dimorphic expression of genes hormone receptors, insulin-like growth factors and vitellogenins, and in gonads during early differentiation of a teleost fish, the Nile tilapia effects of 17 beta-estradiol in the male tilapia (Oreochromis mossambi- Oreochromis niloticus. Biology of Reproduction 78 333–341. (doi:10. cus). General and Comparative Endocrinology 156 544–551. (doi:10. 1095/biolreprod.107.064246) 1016/j.ygcen.2008.03.002) Ikeuchi T, Todo T, Kobayashi T & Nagahama Y 2001 Two subtypes of Devlin RH & Nagahama Y 2002 Sex determination and sex differentiation in androgen and progestogen receptors in fish testes. Comparative fish: an overview of genetic, physiological, and environmental influences. Biochemistry and Physiology. Part B: Biochemistry & Molecular Biology Aquaculture 208 191–364. (doi:10.1016/S0044-8486(02)00057-1) 129 449–455. (doi:10.1016/S1096-4959(01)00375-X) Dijkstra P, Hekman R, Schulz RW & Groothuis T 2007 Social stimulation, Iwasaki Y, Ohkawa K, Sadakata H, Kashiwadate A, Takayama-Watanabe E, nuptial colouration, androgens and immunocompetence in a sexual Onitake K & Watanabe A 2009 Two states of active spermatogenesis dimorphic cichlid fish. Behavioral Ecology and Sociobiology 61 switch between reproductive and non-reproductive seasons in the testes 599–609. (doi:10.1007/s00265-006-0289-7) of the medaka, Oryzias latipes. Development, Growth & Differentiation Earley RL, Blumer LS & Grober MS 2004 The gall of subordination: changes 51 521–532. (doi:10.1111/j.1440-169X.2009.01114.x) in gall bladder function associated with social stress. Proceedings of the Josso N, Racine C, di Clemente N, Rey R & Xavier F 1998 The role of anti- Royal Society B: Biological Sciences 271 7–13. (doi:10.1098/rspb.2003. Mullerian hormone in gonadal development. Molecular and Cellular 2558) Endocrinology 145 3–7. (doi:10.1016/S0303-7207(98)00186-5) Faisal M, Chiappelli F, Ahmed II, Cooper EL & Weiner H 1989 Social Kajiura-Kobayashi H, Kobayashi T & Nagahama Y 2005 Cloning of cDNAs confrontation "stress" in aggressive fish is associated with an endogenous and the differential expression of A-type cyclins and Dmc1 during opioid-mediated suppression of proliferative response to mitogens and spermatogenesis in the Japanese eel, a teleost fish. Developmental nonspecific cytotoxicity. Brain, Behavior, and Immunity 3 223–233. Dynamics 232 1115–1123. (doi:10.1002/dvdy.20289) (doi:10.1016/0889-1591(89)90038-X) Kanamori A, Nagahama Y & Egami N 1985 Development of the tissue FAO 2008 Fishery and aquaculture statistics – aquaculture production. In architecture in the gonads of the medaka Oryzias latipes. Zoological FAO Yearbook 2008, part B-12, p 59. Rome: Food and Agriculture Science 2 707–712. Organization of the United Nations. Karnovsky MJ 1965 A formaldehyde–glutaraldehyde fixative of high Fernandes MO & Volpato GL 1993 Heterogeneous growth in the Nile osmolarity for use in electron microscopy. Journal of Cell Biology tilapia: social stress and carbohydrate metabolism. Physiology & 27 137A. Behavior 54 319–323. (doi:10.1016/0031-9384(93)90117-X) Kaufmann JH 1983 On the definitions and functions of dominance and Fernandes D, Bebianno MJ & Porte C 2007 Mitochondrial metabolism of territoriality. Biological Reviews 58 1–20. (doi:10.1111/j.1469-185X. 17alpha-hydroxyprogesterone in male sea bass (Dicentrarchus labrax): a 1983.tb00379.x) potential target for endocrine disruptors. Aquatic Toxicology 85 Kelly C 2008 The interrelationships between resource-holding potential, 258–266. (doi:10.1016/j.aquatox.2007.09.010) resource-value and reproductive success in territorial males: how much Fraley NB & Fernald RD 1982 Social control of developmental rate in variation can we explain? Behavioral Ecology and Sociobiology 62 the African cichlid, Haplochromis burtoni. Zeitschrift fu¨r Tierpsychologie 855–871. (doi:10.1007/s00265-007-0518-8) 60 66–82. Klu¨ver N, Pfennig F, Pala I, Storch K, Schlieder M, Froschauer A, Francis RC, Jacobson B, Wingfield JC & Fernald RD 1992 Castration lowers Gutzeit HO & Schartl M 2007 Differential expression of anti-Mullerian aggression but not social dominance in male Haplochromis burtoni hormone (amh) and anti-Mullerian hormone receptor type II (amhrII)in (Cichlidae). Ethology 90 247–255. (doi:10.1111/j.1439-0310.1992. the teleost medaka. Developmental Dynamics 236 271–281. (doi:10. tb00836.x) 1002/dvdy.20997)

Reproduction (2012) 143 71–84 www.reproduction-online.org

Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access Social status and testis structure in tilapia 83

Knaut H, Pelegri F, Bohmann K, Schwarz H & Nu¨sslein-Volhard C 2000 Miura T, Miura C, Yamauchi K & Nagahama Y 1995 Human recombinant Zebrafish vasa RNA but not its protein is a component of the germ plasm activin induces proliferation of spermatogonia in vitro in the Japanese and segregates asymmetrically before germline specification. Journal of eel (Anguilla japonica). Fisheries Science 61 434–437. Cell Biology 149 875–888. (doi:10.1083/jcb.149.4.875) Miura T, Miura C, Konda Y & Yamauchi K 2002 Spermatogenesis- Kobayashi T & Nagahama Y 2009 Molecular aspects of gonadal preventing substance in Japanese eel. Development 129 2689–2697. differentiation in a teleost fish, the Nile tilapia. Sexual Development 3 Morinaga C, Saito D, Nakamura S, Sasaki T, Asakawa S, Shimizu N, 108–117. (doi:10.1159/000223076) Mitani H, Furutani-Seiki M, Tanaka M & Kondoh H 2007 The hotei Kobayashi T, Kajiura-Kobayashi H & Nagahama Y 2002 Two isoforms of mutation of medaka in the anti-Mullerian hormone receptor causes the vasa homologs in a teleost fish: their differential expression during germ dysregulation of germ cell and sexual development. PNAS 104 cell differentiation. Mechanisms of Development 111 167–171. (doi:10. 9691–9696. (doi:10.1073/pnas.0611379104) 1016/S0925-4773(01)00613-X) Nagahama Y 1994 Endocrine regulation of gametogenesis in fish. Kobayashi T, Kajiura-Kobayashi H, Guan G & Nagahama Y 2008 Sexual International Journal of Developmental Biology 38 217–229. dimorphic expression of DMRT1 and Sox9a during gonadal differen- Nakamura M & Nagahama Y 1989 Differentiation and development of tiation and hormone-induced sex reversal in the teleost fish Nile tilapia Leydig cells, and changes of testosterone levels during testicular (Oreochromis niloticus). Developmental Dynamics 237 297–306. differentiation in tilapia Oreochromis niloticus. Fish Physiology and (doi:10.1002/dvdy.21409) Biochemistry 7 211–219. (doi:10.1007/BF00004709) Koulish S, Kramer CR & Grier HJ 2002 Organization of the male gonad Nobrega RH & Quagio-Grassiotto I 2007 Morphofunctional changes in in a protogynous fish, Thalassoma bifasciatum (Teleostei: Labridae). Leydig cells throughout the continuous spermatogenesis of the Journal of Morphology 254 292–311. (doi:10.1002/jmor.10037) freshwater teleost fish, Serrasalmus spilopleura (Characiformes, Char- Kretzschmar G, Nisslein T, Zierau O & Vollmer G 2005 No estrogen-like acidae): an ultrastructural and enzyme study. Cell & Tissue Research 329 effects of an isopropanolic extract of Rhizoma Cimicifugae racemosae on 339–349. (doi:10.1007/s00441-006-0377-z) uterus and vena cava of rats after 17 day treatment. Journal of Steroid Ohta T, Miyake H, Miura C, Kamei H, Aida K & Miura T 2007 Follicle- Biochemistry & Molecular Biology 97 271–277. (doi:10.1016/j.jsbmb. stimulating hormone induces spermatogenesis mediated by androgen 2005.05.008) production in Japanese eel, Anguilla japonica. Biology of Reproduction Kurth T, Berger J, Wilsch-Brauninger M, Kretschmar S, Cerny R, Schwarz H, 77 970–977. (doi:10.1095/biolreprod.107.062299) Lofberg J, Piendl T & Epperlein HH 2010 Electron microscopy of the Oliveira RF, Hirschenhauser K, Carneiro LA & Canario AV 2002 Social amphibian model systems Xenopus laevis and Ambystoma mexicanum. modulation of androgen levels in male teleost fish. Comparative Methods in Cell Biology 96 395–423. (doi:10.1016/S0091-679X(10)96017-2) Biochemistry and Physiology. Part B: Biochemistry & Molecular Biology Lacerda SM, Batlouni SR, Costa GM, Segatelli TM, Quirino BR, 132 203–215. (doi:10.1016/S1096-4959(01)00523-1) Queiroz BM, Kalapothakis E & Franca LR 2010 A new and fast Parenti LR & Grier HJ 2004 Evolution and phylogeny of gonad morphology in bony fishes. Integrative and Comparative Biology 44 333–348. technique to generate offspring after germ cells transplantation in adult (doi:10.1093/icb/44.5.333) fish: the Nile tilapia (Oreochromis niloticus) model. PLoS ONE 5 Parikh VN, Clement TS & Fernald RD 2006 Androgen level and male social e10740. (doi:10.1371/journal.pone.0010740) status in the African cichlid, Astatotilapia burtoni. Behavioural Brain Levavi-Sivan B, Bogerd J, Mananos EL, Gomez A & Lareyre JJ 2010 Research 166 291–295. (doi:10.1016/j.bbr.2005.07.011) Perspectives on fish gonadotropins and their receptors. General and Renn SC, Aubin-Horth N & Hofmann HA 2008 Fish and chips: functional Comparative Endocrinology 165 412–437. (doi:10.1016/j.ygcen.2009. genomics of social plasticity in an African cichlid fish. Journal of 07.019) Experimental Biology 211 3041–3056. (doi:10.1242/jeb.018242) Lim C & Webster CD 2006 Tilapia Biology, Culture, and Nutrition. Eds C Rey R, Lukas-Croisier C, Lasala C & Bedecarras P 2003 AMH/MIS: what we Lim & CD Webster. New York: Food Products Pressw. An Imprint of The know already about the gene, the protein and its regulation. Molecular Harworth Press, Inc. and Cellular Endocrinology 211 21–31. (doi:10.1016/j.mce.2003.09. Loir M, Margeridon A & Cauty C 1989 Leydig cells in Myleus 007) ternetzi testes. Aquatic Living Resources 2 57–61. (doi:10.1051/alr: Rosenberg IM 2005 Protein Analysis and Purification: Benchtop 1989006) Techniques, 2nd edn. Boston: Birkhaüser. Lo Nostro FL, Antoneli FN, Quagio-Grassiotto I & Guerrero GA 2004 Salva A, Hardy MP, Wu XF, Sottas CM, MacLaughlin DT, Donahoe PK Testicular interstitial cells, and steroidogenic detection in the proto- & Lee MM 2004 Mullerian-inhibiting substance inhibits rat Leydig gynous fish, Synbranchus marmoratus (Teleostei, Synbranchidae). Tissue cell regeneration after ethylene dimethanesulphonate ablation. & Cell 36 221–231. (doi:10.1016/j.tice.2004.03.001) Biology of Reproduction 70 600–607. (doi:10.1095/biolreprod.103. Machnes Z, Avtalion R, Shirak A, Trombka D, Wides R, Fellous M & Don J 021550) 2008 Male-specific protein (MSP): a new gene linked to sexual behavior Schreck CB 2010 Stress and fish reproduction: the roles of allostasis and and aggressiveness of tilapia males. Hormones and Behavior 54 hormesis. General and Comparative Endocrinology 165 549–556. 442–449. (doi:10.1016/j.yhbeh.2008.03.014) (doi:10.1016/j.ygcen.2009.07.004) Maruska KP & Fernald RD 2010 Behavioral and physiological plasticity: Schulz RW, Paczoska-Eliasiewicz H, Satijn D & Goos H 1993 The feedback rapid changes during social ascent in an African cichlid fish. Hormones regulation of pituitary GTH-II secretion in male African catfish (Clarias and Behavior 58 230–240. (doi:10.1016/j.yhbeh.2010.03.011) gariepinus): participation of 11-ketotestosterone. Fish Physiology and Maruska KP & Fernald RD 2011 Plasticity of the reproductive axis caused by Biochemistry 11 107–115. (doi:10.1007/BF00004556) social status change in an African cichlid fish: II. testicular gene Schulz RW, Menting S, Bogerd J, Franca LR, Vilela DA & Godinho HP 2005 expression and spermatogenesis. Endocrinology 152 291–302. (doi:10. Sertoli cell proliferation in the adult testis – evidence from two fish 1210/en.2010-0876) species belonging to different orders. Biology of Reproduction 73 Maruska KP, Levavi-Sivan B, Biran J & Fernald RD 2011 Plasticity of the 891–898. (doi:10.1095/biolreprod.105.039891) reproductive axis caused by social status change in an African cichlid Schulz RW, de Franca LR, Lareyre JJ, Le GF, Chiarini-Garcia H, fish: I. Pituitary gonadotropins. Endocrinology 152 281–290. (doi:10. Nobrega RH & Miura T 2010 Spermatogenesis in fish. General and 1210/en.2010-0875) Comparative Endocrinology 165 390–411. (doi:10.1016/j.ygcen.2009. Matta SL, Vilela DA, Godinho HP & Franca LR 2002 The goitrogen 6-n- 02.013) propyl-2-thiouracil (PTU) given during testis development increases Soler L, Conte MA, Katagiri T, Howe AE, Lee BY, Amemiya C, Stuart A, Sertoli and germ cell numbers per cyst in fish: the tilapia (Oreochromis Dossat C, Poulain J, Johnson J et al. 2010 Comparative physical maps niloticus) model. Endocrinology 143 970–978. (doi:10.1210/en.143.3. derived from BAC end sequences of tilapia (Oreochromis niloticus). BMC 970) Genomics 11 636. (doi:10.1186/1471-2164-11-636) Miura T, Yamauchi K, Takahashi H & Nagahama Y 1991 Hormonal Soma KK, Francis RC, Wingfield JC & Fernald RD 1996 Androgen induction of all stages of spermatogenesis in vitro in the male Japanese regulation of hypothalamic neurons containing gonadotropin-releasing eel (Anguilla japonica). PNAS 88 5774–5778. (doi:10.1073/pnas.88.13. hormone in a cichlid fish: integration with social cues. Hormones and 5774) Behavior 30 216–226. (doi:10.1006/hbeh.1996.0026) www.reproduction-online.org Reproduction (2012) 143 71–84

Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access 84 F Pfennig and others

Trubiroha A, Kroupova H, Frank SN, Sures B & Kloas W 2011 Inhibition of Wendelaar Bonga SE 1997 The stress response in fish. Physiological gametogenesis by the cestode Ligula intestinalis in roach (Rutilus rutilus) Reviews 77 591–625. is attenuated under laboratory conditions. Parasitology 138 648–659. White SA, Nguyen T & Fernald RD 2002 Social regulation of gonadotropin- (doi:10.1017/S0031182010001514) releasing hormone. Journal of Experimental Biology 205 2567–2581. Turner GF & Robinson RL 2000 Reproductive biology, mating systems and Winer J, Jung CK, Shackel I & Williams PM 1999 Development and parental care. In Tilapias: Biology and Exploitation, pp 33–58. Eds MCM validation of real-time quantitative reverse transcriptase-polymerase Beveridge & BJ McAndrew. London: Kluwer Academic Publishers. chain reaction for monitoring gene expression in cardiac myocytes in Venable JH & Coggeshall R 1965 A simplified lead citrate stain for use in vitro. Analytical Biochemistry 270 41–49. (doi:10.1006/abio.1999. electron microscopy. Journal of Cell Biology 25 407–408. (doi:10.1083/ 4085) jcb.25.2.407) Yoshinaga N, Shiraishi E, Yamamoto T, Iguchi T, Abe S & Kitano T 2004 Vera Cruz EM & Brown CL 2007 The influence of social status on the rate of Sexually dimorphic expression of a teleost homologue of Mullerian growth, eye color pattern and insulin-like growth factor-I gene expression inhibiting substance during gonadal sex differentiation in Japanese in Nile tilapia, Oreochromis niloticus. Hormones and Behavior 51 flounder, Paralichthys olivaceus. Biochemical and Biophysical Research 611–619. (doi:10.1016/j.yhbeh.2007.02.010) Communications 322 508–513. (doi:10.1016/j.bbrc.2004.07.162) Vilela DAR, Silva SGB, Peixoto MTD, Godinho HP & Franca LR 2003 Yoshiura Y, Senthilkumaran B, Watanabe M, Oba Y, Kobayashi T & Spermatogenesis in teleost: insights from the Nile tilapia (Oreochromis Nagahama Y 2003 Synergistic expression of Ad4BP/SF-1 and cyto- niloticus) model. Fish Physiology and Biochemistry 28 187–190. (doi:10. chrome P-450 aromatase (ovarian type) in the ovary of Nile tilapia, 1023/B:FISH.0000030523.16010.62) Oreochromis niloticus, during vitellogenesis suggests transcriptional Vizziano D, Baron D, Randuineau G, Mahe S, Cauty C & Guiguen Y 2008 interaction. Biology of Reproduction 68 1545–1553. (doi:10.1095/ Rainbow trout gonadal masculinization induced by inhibition of biolreprod.102.010843) estrogen synthesis is more physiological than masculinization induced by androgen supplementation. Biology of Reproduction 78 939–946. Zhang WL, Zhou LY, Senthilkumaran B, Huang BF, Sudhakumari CC, (doi:10.1095/biolreprod.107.065961) Kobayashi T, Nagahama Y & Wang DS 2010 Molecular cloning of two Volpato GL, Luchiari AC, Duarte CR, Barreto RE & Ramanzini GC 2003 Eye isoforms of 11beta-hydroxylase and their expressions in the Nile tilapia, color as an indicator of social rank in the fish Nile tilapia. Brazilian Oreochromis niloticus. General and Comparative Endocrinology 165 Journal of Medical and Biological Research 36 1659–1663. (doi:10. 34–41. (doi:10.1016/j.ygcen.2009.05.018) 1590/S0100-879X2003001200007) Weltzien FA, Andersson E, Andersen O, Shalchian-Tabrizi K & Norberg B 2004 The brain–pituitary–gonad axis in male teleosts, with special Received 11 August 2011 emphasis on flatfish (Pleuronectiformes). Comparative Biochemistry and Physiology. Part A: Molecular & Integrative Physiology 137 447–477. First decision 26 September 2011 (doi:10.1016/j.cbpb.2003.11.007) Accepted 26 October 2011

Reproduction (2012) 143 71–84 www.reproduction-online.org

Downloaded from Bioscientifica.com at 10/02/2021 07:41:17AM via free access