| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Mechanisms of Hormone Action |
Junior Research Group of Molecular Animal Cell Toxicology,2 UFZ Centre for Environmental Research, D-04318 Leipzig, Germany
Institute of Zoology,3 University of Dresden, D-01062 Dresden, Germany
Department of Physiological Chemistry,4 University of Würzburg, D-97074 Würzburg, Germany
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
), was analyzed in monosex populations. Dmrt1b(Y) expression correlated strictly with the genotype but not the sexual phenotype. When XY juvenile fish were exposed to 17-ethynylestradiol at concentrations that induce sex reversal, Dmrt1b(Y) expression was not repressed. However, Dmrt1b(Y) was expressed in XY or YY gonads regardless of the sex and could not be detected in XX individuals. In contrast, the expression of Fig1a correlated with the phenotypic sex: Fig1a was expressed in male juvenile fish exposed to 17-ethynylestradiol and repressed in fish exposed to 17-methyltestosterone. The Dmrt1b(Y) expression appears to reflect an early and important event in sex determination and lends support to the suggested key regulatory role of the Dmrt1b(Y) gene in sex determination. This process is apparently hormone insensitive, and the expression of further downstream acting genes can be regulated (directly or indirectly) by sex steroids.
early development, embryo, estradiol, gene regulation, testis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Recently a different sex-determining gene of the medaka (Oryzias latipes) has been identified. The gene Dmrt1b(Y) (for Doublesex- and Mab-3-related transcription factors), a duplicated copy of the autosomal Dmrt1, is the only functional gene in the Y specific region and is expressed in pre-Sertoli and Sertoli cells (the gene was originally called DmY or Dmrt1Y; according to the nomenclature proposed by Volff et al. [8], Dmrt1b(Y) will be used throughout the text). The expression pattern makes it very likely that Dmrt1b(Y) is the male sex-determining gene, but the biochemical function of Dmrt1b(Y) is not known. The DM-domain, a DNA-binding motif, is also found in Drosophila doublesex (Dsx) and Chaenorabditis Mab-3 genes, which are both expressed downstream in the sex-determination cascade [9, 10].
In contrast to mammals, sex determination in many fish including the medaka is labile, and the genetic sex can be reversed functionally by exposure to androgens or estrogens during a certain period of early juvenile development [11–14]. Monosex populations, offspring with only one sex, can be produced by crossing experimentally produced YY or XX males with normal females. This approach has been widely used in aquaculture because one sex is often favored for fish production because of sex-specific growth differences [15]. For scientific purpose, monosex populations allow the study of sex-specific genes during early development when phenotypic sex differences are not yet discernible. In the present study, we report the application of an improved protocol for the establishment of monosex populations using the d-rR.YHNI strain of the model teleost medaka. This strain is characterized by several sex markers: 1) The phenotypic sex can be determined by sex-specific fin characters (shape and size of the anal fin [12]); 2) the genetic sex is visible by the orange body color because the R-locus coding for a gene involved in the formation of orange chromatophores is active only on the Y chromosome [12]; and 3) a deletion in the SL1-locus on the Y chromosome allows identification of the genetic sex in embryos. Amplification with primers flanking this deletion produces polymerase chain reaction (PCR) products of different length for X and Y chromosomes [16] and allows the XY and YY genotypes to be distinguished. Both the r- and SL1-loci are sex-linked markers in the pseudoautosomal region and are not sex determining. Their distance to the sex-determining locus is 1.6–2.1 cM for SL1 [9, 10] and 0.36 cM for the r-locus [9].
The established monosex populations were used to study the effect of sex hormone exposure on two early expressed sex-specific genes: Dmrt1b(Y), the sex-determining gene of medaka [9, 10] and Fig1a (factor in the germline ), an oocyte-specific gene expressed early in development [17].
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The medaka d-rR and Hd-rR.YHNI strains were kindly provided by Dr. Y. Wakamatsu (laboratory of fresh water fish stocks, Bioscience Center, Nagoya University, Nagoya, Japan) and Prof. M. Sakaizumi (Department of Environmental Science, Niigata University, Niigata, Japan). The animals were kept under a constant cycle of 16L:8D at 26°C (±1°C). Fish were fed ad libitum three times daily, once with Artemia and twice with commercial flake food (TetraMin, Tetra GmbH, Melle, Germany). Breeding was performed in 10-L tanks containing one male and one to three females.
Exposure of Fish
Exposure concentrations and conditions that are known to induce complete functional sex reversal were chosen [13, 18]. All exposure schemes started at the time of hatching. Genetic males were exposed to 17-ethynylestradiol (EE2, purity >97%, Merck, Darmstadt, Germany) added to the water. Because addition of 17-methyltestosterone (MT, purity >98%, Sigma, Deisenhofen, Germany) to the water caused sterility in all sex-reversed XX males (unpublished data), fish were fed with MT-containing flake food. Additional feeding of fish with Artemia may lead to a reduced uptake of contaminated flake food. Thus, to ensure sufficient dosage of MT for sex reversal and according to the original protocol of Yamamoto [13], fish exposed to MT (and appropriate control fish) were not fed with artemia.
EE2 was dissolved in dimethylsulfoxide (DMSO, purity 99%, Sigma) and added to the aquarium water to reach a final concentration of 100 ng/L EE2 and 0.005% (v/v) DMSO. Control fish were exposed to the same DMSO concentration. To avoid metabolic and microbial breakdown of EE2, 90% of the water was removed and replaced every 2 days with fresh EE2-contaminated water. Every 7 days all groups of experimental animals were placed into new aquaria with completely exchanged tank water. Sex-reversed fish were produced by exposing them for 2 months to EE2. Following the exposure the animals were kept in clean water until they had reached sexual maturity.
To prepare MT-containing food, 25 µg MT was dissolved in 5 ml ethanol, mixed with 1 g flake food (Tetramin, Tetra), and the ethanol was evaporated overnight. Fertile sex-reversed XX males were produced by exposure for 3 months. Following the exposure the animals were fed with normal food until sexual maturity.
Preparation of the d-rR.YHNI Strain
Males of the Hd-rR.YHNI strain were crossed with females of the d-rR strain. Males from the offspring of this strain were crossed once again with females of the d-rR strain. The resulting strain was called d-rR.YHNI and maintained as a stock of approximately 100 individuals per generation. Animals used for the experiments were from the fourth-generation offspring.
Production of Monosex Populations
Monosex populations were produced according to Yamamoto [12, 13, 19] with some modifications. Offspring of the d-rR.YHNI strain were exposed to EE2 to produce XY females. XY females were identified by female fin characters associated with the Y-linked orange body color and crossed with normal XY males. Twenty-five percent of the offspring could be expected to be YY males, which could not be distinguished from XY males by fin morphology or body color. By analyzing the sex chromosome-specific DNA sequences, YY males could be identified and were crossed with XY females. The offspring of this cross was exposed to EE2 and YY females were identified by analysis of sex chromosome specific sequences. To maintain a stock of YY medaka, YY males and YY females were mated and some of their offspring were exposed to EE2. XX males were obtained by crossing normal d-rR.YHNI males and females and treating the offspring with MT. XX males could be identified by their body color. To maintain a stock of XX medaka, XX males were crossed with XX females and some of their offspring were exposed to MT.
PCR of Sex Chromosome-Specific Sequences
DNA was isolated from either a 2 x 3 mm2 piece of the anal fin or whole 6-day-old embryos after removing the chorion, according to Shinomiya et al. [20]. Briefly, fins or embryos were incubated for 60 min in 100 µl lysis buffer (0.1 mg/ml proteinase K, 100 mM EDTA, 50 mM Tris pH 8.0, 100 mM NaCl, 1% SDS). Then 100 µl of chloroform:phenol (1:1, buffer saturated) was added and mixed by vigorous shaking for 1 min. After centrifugation for 5 min at 12 000 g, DNA was precipitated from the upper phase by adding one tenth of the volume sodium acetate (3 M, pH 5.2) and an equal amount of isopropyl alcohol. After cooling for 20 min at -20°C, the samples were centrifuged for 10 min at 12 000 x g, washed with 70% ethanol, and air dried. The DNA pellet was solubilized in 50 µl of sterile water. PCR was performed using the primers pHO5.5F (5'-CCT GCA ATG GGA AAT TAT TCT GCT C-3') and pHO5.5RV (5'-CTT TTG TGT CTT TGG TTA TGA AAC GAT G-3') at 55°C annealing temperature and 30 cycles.
Reverse Transcription-Polymerase Chain Reaction
RNA was prepared from medaka embryos or juvenile fish with Trifast (Peqlab, Erlangen, Germany) according to the manufacturer's instructions. If juvenile fish were used, head and tail were removed prior to RNA analysis. Because of the different exposure conditions, growth of juvenile fish exposed to MT was delayed when compared with fish exposed to EE2. To analyze gene expression in juvenile fish of the same size and stage of development, samples for RNA isolation were taken at 23 days for fish exposed to EE2 and 40 days for fish exposed to MT. The cDNA was synthesized from DNAse (Roche Diagnostics, Mannheim, Germany)-treated RNA using oligo dT primer and Muloney murine leukemia virus reverse transcriptase (Invitrogen, Karlsruhe, Germany). PCR was performed at 60°C annealing temperature for 30 (Actin, Fig1a) or 40 cycles (Dmrt1b(Y)) using Taq-polymerase (Peqlab) with 0.5 µl cDNA per 15 µl reaction volume. Actin transcripts were amplified using the primers 5'-AGG GAG AAG ATG ACC CAG ATC-3' and 5'-CGC AGG ACG CCA TAC CAA-3'. Dmrt1b(Y) was amplified using 5'-CCG GGT GCC CAA GTG CTC CCG CTG-3' as forward and 5'-GAT CGT CCC TCC ACA GAG AAG AGA-3' as reverse primer. Fig1a was amplified using 5'-TAC TGC TGC ATC GAG AAG TAC AAG-3' as forward and 5'-AGA GTA CAG CTG AAA GCA AAA TGA G-3' as reverse primer. The primers for Actin and Dmrt1b(Y) were cDNA specific (i.e., because of introns in the sequence flanked by primers, amplification of genomic DNA resulted in larger fragments.)
Alternative primers were used for the amplification of Dmrt1b(Y) and actin in embryos: For the detection of Dmrt1b(Y), the primers 5'-GGC CGG GTC CCC GGG TG-3' and 5'-TGC GGC AGA CAG AGG ATT GG-3' were used. Actin was amplified with 5'-TTA AAC AGC CCT GCC ATG TA-3' as forward and 5'-GCA GCT CAT AGC TCT TCT CCA GGG AG-3' as reverse primers.
Statistical Analysis
Statistically significant differences of the proportions of male offspring (P < 0.05) were analyzed by the Fisher's exact test using the program GraphPad InStat, version 3.00 (GraphPad Software, San Diego CA, www.graphpad.com).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Matsuda et al. [16] have established the medaka inbred strain Hd-rR.YHNI with a sex chromosome-specific DNA sequence. This strain showed reduced fecundity in our lab (i.e., the first spawning of females was delayed 1–2 months and fewer eggs were produced when compared with the d-rR strain). The d-rR strain [12] has the same Y-linked orange color marker as the Hdr-R.YHNI strain but lacks the sex chromosome-specific DNA sequences. By crossing of d-rR females with Hd-rR.YHNI males, the sex-specific chromosome sequence was introduced into the offspring of these parents. The resulting strain was called d-rR.YHNI, and no difference in fecundity could be observed, compared with the d-rR strain.
By means of a series of hormone exposures and crossings described by Yamamoto [12, 13, 19], the d-rR.YHNI strain was used to produce XY females, YY males, YY females, and XX males. Because YY medaka cannot be distinguished from XY medaka by morphological criteria, Yamamoto performed backcrossings to identify the YY genotype. In contrast to the d-rR strain used by Yamamoto, the d-rR.YHNI strain allowed YY males and females to be identified by analysis of sex chromosome-specific DNA sequences. PCR of DNA from XY fish resulted in two distinct bands, but XX and YY fish showed only one specific band corresponding to either the X or Y chromosome (Fig. 1). By isolating DNA from the anal fin, YY animals were identified by the lack of the X-specific fragment after amplification of sex chromosome-specific DNA. The same fish from which the DNA samples of fin tissue had been taken were used in subsequent mating experiments.
|
The genetic sex of the offspring was analyzed by isolating DNA from individual 6-day-old embryos. The determined genotypes of the offspring were consistent with the known genotypes of the parents: 1) a cross of normal XY males and XX females yielded embryos of either the genotype XY or XX; 2) offspring of the cross of YY males and XX females all had the male genotype; 3) by crossing sex-reversed XX males with XX females, 100% of the offspring were of the XX genotype; and 4) the offspring of YY males mated with sex-reversed YY females were all of the YY genotype.
To test whether the phenotype matched the genotype in all crosses, the offspring were reared until sexual maturity and the phenotypic sex was determined by observation of the anal fins and dissection of the gonads (Table 1). In keeping with the Y-linked orange body color, all orange fish were phenotypic males. However, some males (2%) with white body color developed from the offspring of XY males mated with XX females. Among the offspring of sex-reversed XX males with XX females, 17% developed as males. Analysis of the sex chromosome-specific DNA confirmed the XX genotype of these males. Furthermore, the lack of Dmrt1b(Y) expression indicated that male differentiation was not caused by Dmrt1b(Y) transferred to the X chromosome by a historical cross-over event (Fig. 2). These result confirmed the findings on spontaneous sex-reversed XX males obtained for other strains of the medaka [21].
|
|
The reason for the occurrence of male differentiation in XX individuals is unclear. To test whether we could select for (or against) spontaneous male differentiation in this genotype, we analyzed offspring of 10 different XX mating pairs. Indeed, pairs with no male offspring and pairs with male offspring varying between 3.6% and 18.5% were observed (Fig. 3a). Statistical significance was reached only when the mating pair with the highest proportion of males was compared with pairs with no male offspring or compared with all other pairs. By testing the presumption that pairs with a certain percentage of male offspring belong to similar groups of a certain genetic predisposition, a statistically significant difference could be found for pairs with a high proportion of male offspring when compared with pairs with a low proportion of male offspring (Fig. 3b).
|
If the difference in the percentage of XX males between individual mating pairs was based on a genetic difference, mating of spontaneous XX males (i.e., males that have occurred without exposure to hormones) to normal females might result in an increased percentage of male offspring in the following generation. However, this increase was not observed (Table 1, experiment 5).
Expression of Sex-Specific Genes in Monosex Populations
The XX mating pairs that previously did not show any spontaneous development of XX males in their offspring were chosen to analyze gene expression and its relationship to phenotypic and genotypic sex.
Using the previously established monosex populations, Dmrt1b(Y) expression was detected specifically by RT-PCR in RNA pools of XY embryos. However, Dmrt1b(Y) was not expressed in XX embryo populations (Fig. 4).
|
To analyze the effect of hormonal sex reversal on the expression of Dmrt1b(Y) and Fig1a, fish were exposed to EE2 or MT. Because Fig1a could not be detected in female (XX) fish prior to 23 days after hatching, the reverse transcription (RT)-PCR analysis of gene expression after sex reversal was carried out at this stage of development. Fig1a was expressed in normal juvenile XX females and experimentally produced sex-reversed XY females, but XY males did not express this gene. Exposure of XX fish to MT repressed expression of Fig1a (Fig. 5). These experiments show that the expression of Fig1a was linked to the phenotypic sex. The lack of expression of Fig1a in one sex-reversed XY female may be attributed to a delay in growth, which has been observed occasionally for individual fish. In contrast to Fig1a, expression of the master regulating gene Dmrt1b(Y) appears to be linked to the genetic sex: Dmrt1b(Y) expression was not repressed in sex-reversed XY females. Because Dmrt1b(Y) is located only on the Y chromosome, no expression was observed in XX medaka (Fig. 5). Some very weak bands in RT-PCR samples of XX females in the size of the Dmrt1b(Y) band were unspecific artifacts because Dmrt1b(Y) principally cannot be detected in genomic DNA preparations of XX medaka (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The production of monosex populations is an alternative approach circumventing the sexing of individual fish. The present study has demonstrated that monosex population can be established and controlled using a molecular marker. YY and XX males were mated with normal females, and the unique genetic sex of the offspring was detected by the analysis of sex chromosome-specific DNA sequences. For male monosex populations, genetic sex clearly matched the phenotypic sex. However, XX monosex populations consisting of only female genotypes had up to 18.5% phenotypic males. Because these males were lacking the expression of the sex-determining gene Dmrt1b(Y), it could not be excluded that spontaneous development of XX males without exposure to androgens is caused by a combination of unknown genetic factors. Different frequencies of XX males in medaka, depending on the strain, have also been observed: Some strains did not show any occurrence of XX males, suggesting that the development of XX males might be caused by unknown genetic factors in the d-rR.YHNI strain [21]. Crossing the offspring of those mating pairs with a high percentage of XX males (this study) did not result in a further increase in the percentage of XX males or at least a reproducibly high percentage. The genetic (or epigenetic) basis for the observed instability remains to be elucidated. To establish XX female monosex populations with virtually no male offspring, selection of appropriate XX breeding pairs will be continued for several generations and may support or disprove a genetic basis for the spontaneous development of XX males.
Monosex populations were used to analyze Dmrt1b(Y) and Fig1a, two sex-specific gonadal genes expressed during early development. Dmrt1b(Y) could be detected in 7-day-old embryos of male monosex populations but was not detectable in XX embryos. Dmrt1b(Y) expression was hormone insensitive and appeared to be linked to the genetic sex: If juvenile XY fish were exposed to EE2 at sex-reversing concentrations, expression of Dmrt1b(Y) was not repressed. In contrast, Fig1a was hormone sensitive, could be induced in genetic males by exposure to EE2, and was repressed by MT. The difference in the expression of both genes is in keeping with their suggested function. Although not structurally related, Dmrt1b(Y) resembles the mammalian Sry master regulating function for sex determination [23]. It is probably the first gene in the male differentiation pathway of medaka, and it may not have any function other than initiating a gene cascade leading to male development [9, 10]. The apparent hormone insensitivity of Dmrt1b(Y) supports this key regulatory function in sex determination. Unlike Dmrt1b(Y), Fig1a is not a master regulating gene but, with respect to the sex-determining cascade, is probably a further downstream acting gene involved in the phenotypic differentiation of the ovary [24]. The gene is expressed during early oocyte development and has been identified by subtractive cloning in fish larvae just 1 day after hatching [17]. The mouse homolog is a germ cell-specific transcription factor involved in the co-ordinate expression of the zona pellucida genes [24]. Its linkage to an important structural component of the oocyte explains its expression in genetic males exposed to sex-reversing estrogen concentrations.
Analysis of the sex specifically expressed genes Dmrt1b(Y) and Fig1a in medaka and the effect of hormone exposure have clearly demonstrated the usefulness of monosex populations. The relatively easy experimental manipulation of sexual differentiation in fish and the analysis of key genes present in all vertebrates offers an attractive model system to unravel the molecular mechanisms of sex determination and differentiation.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received: 7 March 2003.
First decision: 26 March 2003.
Accepted: 10 April 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. F Orlando, Y. Katsu, S. Miyagawa, and T. Iguchi Cloning and differential expression of estrogen receptor and aromatase genes in the self-fertilizing hermaphrodite and male mangrove rivulus, Kryptolebias marmoratus. J. Mol. Endocrinol., October 1, 2006; 37(2): 353 - 365. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
