| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Experimental Zoology, Utrecht University, NL-3584 CH Utrecht, The Netherlands
ABSTRACT
A putative FSH receptor (FSH-R) cDNA was cloned from African catfish testis. Alignment of the deduced amino acid sequence with other (putative) glycoprotein hormone receptors and analysis of the African catfish gene indicated that the cloned receptor belonged to the FSH receptor subfamily. Catfish FSH-R (cfFSH-R) mRNA expression was observed in testis and ovary; abundant mRNA expression was also detected in seminal vesicles. The isolated cDNA encoded a functional receptor since its transient expression in human embryonic kidney (HEK-T) 293 cells resulted in ligand-dependent cAMP production. Remarkably, African catfish LH (cfLH; the catfish FSH-like gonadotropin has not been purified yet) had the highest potency in this system. From the other ligands tested, only human recombinant FSH (hrFSH) was active, showing a fourfold lower potency than cfLH, while hCG and human TSH (hTSH) were inactive. Human CG (as well as cfLH, hrFSH, eCG, but not hTSH) stimulated testicular androgen secretion in vitro but seemed to be unable to bind to the cfFSH-R. However, it was known that hCG is biologically active in African catfish (e.g., induction of ovulation). This indicated that an LH receptor is also expressed in African catfish testis. We conclude that we have cloned a cDNA encoding a functional FSH-R from African catfish testis. The cfFSH-R appears to be less discriminatory for its species-specific LH than its avian and mammalian counterparts.
FSH, FSH receptor, LH, male sexual function, testes
INTRODUCTION
The gonadotropins regulate the activity of their gonadal target cells via specific, membrane-associated receptors [1]. The gonadotropins, as well as thyroid-stimulating hormone (TSH), are heterodimeric molecules composed of a common glycoprotein hormone 
-subunit and a hormone-specific ß-subunit that determines the biological activity. The structural resemblance among the ß-subunits of these hormones and also among the - and ß-subunits suggests that the subunits of glycoprotein hormones have evolved from a single ancestor molecule [2].
In several teleost fishes two distinct gonadotropins (GTH I and GTH II) have been characterized [3–10]. While teleost GTH II is believed to be homologous to tetrapod LH, the homology of teleost GTH I with tetrapod FSH has been debated. It has recently been suggested [11], however, that fish GTH I and GTH II are structurally and functionally equivalent to tetrapod FSH and LH, respectively. Therefore, we will refer to these hormones as FSH and LH in this paper, as has been proposed previously by others as well [12, 13]. In some fishes, it has been shown that two distinct types of gonadotroph cells produce these two hormones (for salmonids [14–16]; for tuna [17]). Despite repeated attempts, in the African catfish only the LH-like gonadotropin (African catfish LH; cfLH) has been purified to date [18, 19], and only one type of pituitary gonadotroph cell has been identified [20, 21]. Recently, however, we have isolated a cDNA from a catfish pituitary cDNA library that shows significant sequence homology with other teleost FSH ß-subunit cDNAs (unpublished results); it cannot be excluded that the FSH in catfish is produced by the same cell type that produces cfLH, a situation well known from higher vertebrates.
The cloning of the cDNAs encoding FSH receptors (FSH-Rs), LH/CG receptors (LH-Rs), and TSH receptors (TSH-Rs) from mammalian and avian species revealed that these receptors share several structural similarities. These glycoprotein hormone receptors form a subfamily among the G protein-coupled receptors and are characterized by large extracellular N-terminal domains [22] involved in high-affinity binding of their respective ligands. The N-terminal domains are followed by the typical seven-transmembrane helical motifs ending with intracellular C-terminal domains. The amino acid sequences of the glycoprotein hormone receptors are highly homologous, especially in the transmembrane regions, suggesting that these genes, similar to the genes encoding their ligands, may have evolved from a common ancestor.
In order to be able to study the spatiotemporal expression pattern of the catfish gonadotropin receptor(s), we initiated the identification of such receptors by cloning the gonadotropin receptor cDNAs from African catfish testis. Here, we report on the cloning and functional characterization of such a receptor. The present data regarding the structural characteristics suggest that the cloned receptor (catfish FSH-R; cfFSH-R) cDNA is of the FSH-R type. Functional studies using human embryonic kidney (HEK-T) 293 cells expressing the cfFSH-R, however, showed that this receptor is most potently stimulated, of the ligands tested, by cfLH. Moreover, we also provide circumstantial evidence showing that a catfish LH receptor (cfLH-R) is expressed in testis.
MATERIALS AND METHODS
Experimental Animals
African catfish were bred and raised in the laboratory as described previously [23], except that catfish pituitary extracts instead of hCG were used to induce ovulation. Animal culture and experimentation were consistent with the Dutch national regulations; experimental protocols were submitted to and approved by the respective University committee. In our hands, sexual maturation starts at 
3 mo of age when the first indications are found in which spermatogonial proliferation is initiated. At 6 mo of age spermatozoa are found in the testes of most fish. The fully mature males used for breeding purposes are 10–12 mo old. All tissues and sperm samples used in the present study for RNA isolation, genomic DNA isolation, or in vitro experiments on androgen secretion were collected from 10- to 12-mo-old, fully mature catfish.
Primers
Primers were obtained from Pharmacia (Roosendaal, The Netherlands) or Life Technologies (Breda, The Netherlands): TM I-1, 5'-GATGGATCCAAYYCNTGYGARGAYHTNATG-3'; TM I-2, 5'-TTCACTTTCCTGCGTGTCCTCATC-3'; TM III-1, 5'-GATGAATTCCANCKYTCNARNGTDAT-3'; TM III-2, 5'-CCTGAATTCRAAIRYISWIAMRAAICCI GC-3'; TM III-3, 5'-ACCCGGGACCCGTTTGCCATTC-3'; ex-9 (corresponding to nucleotides 1121–1147 in Fig. 1A), 5'-GGCACCTCAGTCTACTCTCTGAGATGG-3'; ex-10/11 (corresponding to nucleotides 1346–1371 in Fig. 1A), 5'-TCCTCGCACGGGTTGAAGGCATCTGG-3'; in which Y = C or T, S = G or C, R = G or A, K = G or T, W = A or T, M = A or C, H = A or T or C, D = G or A or T, N = any deoxynucleotide, and I = deoxyinosine. Some primers contained restriction enzyme recognition sites (underlined) at their 5' ends to facilitate subcloning.
|
RNA Isolation and Genomic DNA Isolation
Total RNA was isolated from different tissues including mature testis by the method of Chirgwin et al. [24]. Poly(A)-rich RNA was prepared using Dynabeads-oligo dT25 (Dynal A.S., Oslo, Norway) according to the manufacturer's instructions. Genomic DNA was isolated from sperm of a mature male according to Ausubel et al. [25].
Reverse Transcription-Polymerase Chain Reaction, Polymerase Chain Reaction, and DNA Sequence Analysis
African catfish poly(A)-rich RNA was reverse transcribed with random hexamers using a first-strand cDNA synthesis kit according to the instructions of the manufacturer (Amersham, Roosendaal, The Netherlands). An aliquot of the cDNA preparation was used in a polymerase chain reaction (PCR) together with two degenerate oligonucleotide primers (TM I-1 and TM III-1). Next, the PCR was diluted 20-fold in water, and a seminested PCR was performed using primers TM I-1 and TM III-2. The primers used correspond to DNA sequences encoding conserved amino acid sequences within the mammalian glycoprotein hormone receptor family. The PCR cycling conditions (both PCRs: 40 cycles) were at 94°C for 45 sec, at 37°C for 30 sec, and at 72°C for 1 min. After the second PCR, a PCR product of 
310 base pairs (bp) was generated and cloned into pBluescript KS+ (Stratagene, La Jolla, CA) for sequence analysis.
One microgram of genomic catfish DNA was used as template for PCR amplification of the region between the ex-9 and ex-10/11 primers in 100-µl volumes containing 50 mM KCl, 10 mM Tris/HCl (pH 8.3), 1.5 mM MgCl2, 0.01% gelatin, 200 µM each dNTP, and 20–100 pmol primers in a Perkin-Elmer Cetus cycler (PE Biosystems, Foster City, CA) using 0.5 U SuperTaq (HT Biotechnology Ltd., Cambridge, U.K.). The PCR product obtained (2.2 kilobases [kb]) was cloned in pGEM-T (Promega, Madison, WI).
DNA sequences were determined from both strands using the dideoxy chain termination method [26] with denatured, double-stranded DNA as template in combination with a T7 DNA polymerase-based sequencing kit (Pharmacia) or in combination with a Dye-Terminator cycle-sequencing kit (PE Biosystems). DNA sequence analysis and multiple sequence alignment analysis were carried out using GeneWorks 2.3 software (IntelliGenetics Inc., Mountain View, CA) and Lasergene software (DNASTAR Inc., Madison, WI).
Isolation of cDNA Clones
A unidirectional, randomly primed African catfish testis cDNA library (in total 1 x 106 independent clones, amplified in 32 aliquots of 3 x 104 original clones each) was constructed in ZAP Express (Stratagene) from African catfish testis poly(A)-rich RNA using a commercial cDNA synthesis and cloning system, according to the instructions of the manufacturer (Stratagene).
A PCR-based screening method of the amplified African catfish testis cDNA sublibraries, using oligonucleotides (TM I-2 and TM III-3) corresponding to the characterized 310-bp PCR product, revealed that several sublibraries contained the cDNA insert of interest. Approximately 5 x 104 clones of these sublibraries were absorbed to two replica Hybond-N filters (Amersham) and hybridized with the radiolabeled [TM I-2–TM III-3] PCR product. From each of sublibraries 15 and 20, a single 
clone was isolated and excised in vivo as pBK-CMV phagemids (designated pBK-7 and pBK-13, respectively).
Real-Time Quantitative PCR
Primers and fluorogenic probes (Table 1), specific for the cfFSH-R mRNA and specific for the endogenous control (catfish 28S rRNA [cf28S]), were designed with Primer Express software (PE Biosystems), according to the manufacturer's guidelines as described previously [27] and were purchased from PE Biosystems. Briefly, primers and probes had 30%–80% GC content avoiding runs of more than three consecutive Gs, and had Tm's of 58–60 and 68–70°C, respectively. Fluorogenic probes were synthesized with the fluorescent reporter dye FAM (cfFSH-R-Pr) or VIC (cf28S-Pr) attached to the 5'-end and a quencher dye TAMRA to the 3'-end and were selected from the strand with more Cs than Gs (as recommended by the manufacturer). In addition, probes with a G at the 5'-end were avoided as this has been shown by the manufacturer to exert a quenching effect on the reporter dye [27]. Forward and reverse primers were positioned as close as possible to each other without overlapping the probe, and each had less than three Gs or Cs in the five most 3' nucleotides. By having these strict primer design criteria we were able to use a TaqMan Universal PCR Master Mix (Mg2+ concentration 5.5 mM; PE Biosystems) for all reactions that circumvented the necessity for rigorous optimization.
|
To analyze the relative cfFSH-R mRNA expression levels (i.e., the cfFSH-R mRNA expression after normalization to cf28S rRNA expression), 2 µg of total RNA from different tissues was reverse transcribed in a final volume of 100 µl containing 1x first strand buffer, 0.01 M dithiothreitol, 750 ng random hexamers, 250 U of Supercript II RNase H- reverse transcriptase (Life Technologies), and 500 nM dNTP (Pharmacia), and incubated at 42°C for 50 min. Next, the reverse transcriptase was inactivated by heating at 70°C for 10 min, and the volume was increased to 200 µl. TaqMan PCR assays were performed in triplicate, using 96-well optical plates on an ABI Prism 7700 Sequence Detection System (PE Biosystems) using default settings. For each 25-µl PCR reaction, 5 µl cDNA was mixed with 100 nM fluorogenic probe, 300 nM sense primer, and 300 nM antisense primer in 1x TaqMan Universal PCR Master Mix (PE Biosystems). All TaqMan PCR data were captured using Sequence Detection Software (SDS version 1.7; PE Biosystems). For every sample, an amplification plot was generated, showing the increase in the reporter dye fluorescence with each cycle of PCR. From each amplification plot, a threshold cycle (Ct) value was determined. The Ct values were then exported into Microsoft Excel worksheets for further analysis.
The concept of the threshold cycle (Ct), which is at the heart of accurate and reproducible quantification using fluorescence-based, real-time quantitative PCR, has recently been decribed in detail [28]. Briefly, from each amplification plot, a threshold cycle (Ct) value is calculated, representing the PCR cycle number, in the exponential phase of the PCR, at which the fluorescence is detectable above an arbitrary threshold, based on the variability of baseline data in cycles 3–15 (or 3–7 for cf28S rRNA). The reporter signal is normalized to the fluorescence of an internal reference dye, to allow for corrections in fluorescent fluctuations caused by changes in concentrations or volume, and a Ct value is reported for each sample. The Ct values can be translated into a quantitative result by constructing a standard curve or by using an arithmetic comparative method (
Ct method), after normalization to the internal reference Ct values, according to the manufacturer's instructions.
The Ct value is inversely proportional to the log of the initial mRNA copy number. Therefore, standard curves for each primer/probe set (cf. Fig. 3, A and B, and Table 1) were generated by plotting the Ct values (y-axis), with 95% confidence intervals, against the logarithm of input testis cDNA (x-axis). The slope (m) of the standard curve describes the PCR efficiency and is defined from the equation Ct = m(log Q) + c, where Ct is the threshold cycle, Q is the initial copy number, and c is the intercept on the y-axis. If the PCR is exponential, resulting in the doubling of product in every cycle, the slope will be -3.32, as 3.32 cycles are required to generate a 10-fold increase in product. This allows for quantification of target genes without concerns over differences in amplification efficiency [29]. If the PCR amplification is less efficient the gradient will be steeper.
|
The relative amounts of cfFSH-R message in the cDNA samples from different tissues (expressed as percentages of the tissue with the highest expression) were calculated using the arithmetic comparative method (Ct method). For this method to be valid, the efficiency of the target (cfFSH-R mRNA) amplification and the efficiency of the internal reference (cf28S rRNA) must be approximately equal. To this end a validation experiment, where the Ct variance is tested in fourfold serial dilutions of testis cDNA was performed. For both primer/probe combinations, the absolute value of the slope of the Ct versus log input amount of testis cDNA should be <0.1 (cf. Fig. 3C).
Tissues yielding a Ct value of 40 were considered not to express the cfFSH-R mRNA (cf. Fig. 4). In addition, tissues yielding Ct values 6.64 higher than in the gonadal tissue with the highest Ct value, were considered to not express physiologically significant amounts of cfFSH-R mRNA (cf. Fig. 4); after all, 100-fold lower cfFSH-R mRNA levels would require 6.64 cycles more to generate a similar detectable fluorescence signal.
|
Expression of the cfFSH-R, Cell Culture Techniques, and Transient Transfection
The HEK-T 293 cells [30] were maintained under 5% CO2 in culture medium (Dulbecco modified Eagle medium [DMEM] containing 2 mM glutamine, 10% fetal bovine serum, and 1x antibiotic/antimyotic solution; all from Life Technologies). The HEK-T 293 cells were transiently transfected with pBK-7 and a plasmid (pCRE/ß-gal) containing a ß-galactosidase gene under the control of the human vasoactive intestinal peptide promoter containing five cAMP response elements (CREs) [31] using the modified bovine serum transfection method (MBS; Stratagene; 1 µg pBK-7 and 10 µg pCRE/ß-gal/10-cm dish containing approximately 5 x 106 cells). After 16–20 h, the cells were split into two 96-well plates. One day later, the cells were stimulated for 6 h with various concentrations of cfLH [18], other ligands for gonadotropin receptors (human recombinant LH [hrLH], human recombinant FSH [hrFSH], eCG, hCG; all were a kind gift of Dr. W.G.E.J. Schoonen, Organon-Oss, The Netherlands), or human TSH (hTSH; Sigma, St. Louis, MO) in Hepes-modified DMEM (Sigma) containing 0.1% BSA (final volume of 50 µl).
Colorimetric Detection of Ligand-Induced Activation of Transiently Expressed cfFSH-R
The ß-galactosidase activity was measured according to Chen et al. [31] with minor modifications as described previously [32]. The absorbance at 405 nm was measured in a 96-well microplate reader (Bio-Rad, Hercules, CA). To take into account interassay variations based on the reporter gene's transfection/expression efficiencies, we measured the absorbance at 405 nm after 10 µM forskolin stimulation. Ligand-induced changes in the absorbance were then related to the forskolin-induced changes in the same transfection series. Hence, the results are expressed as arbitrary assay units, related to the forskolin-induced cAMP-mediated reporter gene activation. The ligand concentrations inducing a half-maximal stimulation (EC50) were calculated using the GraphPad PRISM2 software package (San Diego, CA). The EC50 values from three independent experiments were used for statistical comparison of cfLH and hrFSH by one-way ANOVA, followed by Fisher's probable least-squares difference test. A P < 0.05 was considered significant.
Inositol Phosphate Production
Total inositol phosphates (IPs) were extracted and separated as described previously [33]. Briefly, 24 h after transfection with either pBK-7 or pCGR III (i.e., the eukaryotic expression vector pcDNA3 [Invitrogen, Groningen, The Netherlands], containing the African catfish GnRH receptor [cfGnRH-R] cDNA insert) [32], HEK-T 293 cells were transferred to 48-well plates (2.5 x 105 cells/well in 0.5 ml inositol-free DMEM containing 10% dialyzed fetal calf serum) and incubated for 24 h with [3H]inositol (1 µCi/ml; Amersham). Medium was removed, and cells were washed and preincubated for 10 min with assay medium (Hepes-modified DMEM containing 10 mM LiCl). After removing the assay medium, cells tranfected with the cfFSH-R construct (pBK-7) were incubated with 200 µl assay medium (negative control) or with 1 µg/ml of the various ligands for gonadotropin receptors or hTSH in 200 µl assay medium at 37°C for 45 min, then the assay medium was aspirated. As a control for the procedure, various concentrations of chicken GnRH-II were added to the cells transfected with the cfGnRH-R construct (pCGR III). After extraction with 10 mM formic acid at 4°C for at least 90 min, extracts were transferred to columns containing Dowex (AG 1x8) anion-exchange resin (Sigma). Total IPs were then eluted, and the amount of radioactivity was counted. Assays were performed in duplicate in three separate experiments.
Androgen Secretion by African Catfish Testis In Vitro
Testicular tissue was prepared for in vitro incubations as described previously [34]. In short, per experiment testis tissue from three males was placed in a Petri dish containing Earle balanced salt solution, pH 7.18 (M199 EBSS), supplemented with Hepes (0.02 M), penicillin G, and streptomycin (100 U/ml each; antibiotics and medium from Life Technologies). Being submersed in medium, the tissue was cut into fragments of 2 mm3, after which the fragments were filtered over two layers of cheesecloth to remove small tissue pieces and suspended sperm. Four randomly selected fragments (wet weight 25–40 mg) were placed in wells of 24-well culture plates in 1 ml of medium. For the dose-response experiments with cfLH, hrFSH, and hCG, six replicates were incubated per ligand concentration that ranged from 3 ng to 30 µg/ml medium (Fig. 5B), while 1 µg/ml was chosen as ligand concentration for single dose experiments with cfLH, hrFSH, hCG, and eCG (Fig. 5C). After 18 h of incubation at 25°C, the incubation media were collected, heated for 10 min at 80°C, centrifuged at 15 000 x g for 15 min at 4°C, and the supernatants were stored at -25°C until RIA quantification of 11ß-hydroxyandrostenedione (OHA; 4-androsten-11ß-ol-3,17-dione), the quantitatively dominating androgen produced by African catfish testis [34, 35]. The results are expressed as nanogram OHA secreted per milligram of testis tissue incubated. The gonadotropin concentrations inducing the half-maximal stimulation of OHA secretion (EC50) were estimated by nonlinear regression fits as described above.
|
RESULTS
Degenerate oligonucleotide primers, derived from conserved amino acid sequences flanking the transmembrane domains I and III (TM I and III) of mammalian glycoprotein hormone receptors, were used in two consecutive PCRs with African catfish testis cDNA as template. A PCR product of the expected length was cloned and sequenced. Two new primers (TM I-2 and TM III-3) were designed based on the DNA sequence of the cloned PCR product, in order to obtain a plasmid harboring the corresponding full-length cDNA sequence for expression in eukaryotic cells. To this end, the TM I-2 and TM III-3 primers were used for a PCR-based evaluation to identify the African catfish testis cDNA sublibraries containing such clones. Plaque lifts of the positive sublibraries (each 5 x 104 plaques/plate) were performed, and the corresponding filters were hybridized with the radiolabeled [TM I-2–TM III-3] PCR fragment. Two individual clones, designated pBK-7 and pBK-13 (derived from sublibraries 15 and 20, respectively), were selected. After in vivo excision, sequence analysis showed that both pBK-7 and pBK-13 carried cDNA inserts of 2.1 kb. The cDNA insert of pBK-7 contained an open reading frame of 1986 bp that was flanked by a leader sequence of 39 nucleotides and a trailer sequence of 68 nucleotides (Fig. 1A). The second clone, pBK-13, was a partial clone, although the sequence was in complete agreement with the pBK-7 sequence between nucleotides 335 and 2163. However, pBK-13 contained an additional 334 nucleotides at the 5'-untranslated end (Fig. 1A). Sequence analysis of clones, derived from sublibraries other than 15 and 20, revealed that some of them contained a retained intron between nucleotides 753 and 754 in Figure 1A (for the DNA sequence: see Fig. 1B), which contains an in-frame stop codon (bold in Fig. 1B) immediately after the 5' splice site.
Genomic DNA of the African catfish was used as template for PCR amplification between primer ex-9, corresponding to part of exon 9 (in LH-R, FSH-R, and TSH-R genes), and primer ex-10/11, corresponding to part of exon 10 (in FSH-R and TSH-R genes) and exon 11 (in LH-R genes). Sequence analysis revealed that the cloned PCR product contained only a single intron, present on the position homologous to the one in mammalian FSH-R and TSH-R genes (Fig. 1C).
Two putative translation initiation codons, starting at positions 362 and 374 (both underlined in Fig. 1A), respectively, could be identified. The second ATG starting at position 374 is considered to be the translation initiation codon because it has a more favorable sequence context for translation initiation (CC[A/G]CCATG[G]) [36] and taking into account the multiple sequence alignment analysis of the deduced amino acid sequence of pBK-7 and various other glycoprotein hormone receptor amino acid sequences (in particular the FSH-Rs; see below, and Fig. 2).
|
Conceptual translation of the open reading frame of pBK-7 following the proposed initiation codon predicted a protein of 662 amino acids, the first 22 amino acids of which represent the proposed signal peptide [37]. Accordingly, the mature protein (designated cfFSH-R; see below) consists of 640 amino acids and shows typical characteristics of the glycoprotein hormone receptor subfamily of G protein-coupled receptors (Fig. 1A). The extracellular N-terminal domain region consists of 314 amino acids and has five potential N-linked glycosylation sites and a total of eight cysteine residues (an N-terminal cluster of two cysteines, and a C-terminal cluster of six cysteines). The seven transmembrane-spanning segments together consist of 266 amino acids (Fig. 1A). The extracellular loops between TM II–III and TM IV–V each contain a cysteine residue at conserved positions (Cys417 and Cys492) that are thought to link these extracellular loops via a disulfide bridge (Fig. 1A) [38]. The intracellular loops between TM I–II, TM III–IV, and TM V–VI, as well as the intracellular C-terminal domain, contain multiple tyrosine, serine, and threonine residues that may represent potential phosphorylation sites (Fig. 1A). Two sites within the TM domain (Thr530 and Ser536) represent consensus sites for protein kinase C and casein kinase II phosphorylation, respectively [39, 40]. The intracellular C-terminal domain consists of 60 amino acids, including two potential sites for phosphorylation by protein kinase C (Ser607 and Thr616) [39].
Computer-assisted searches of the GenBank database [41] and multiple sequence alignment (Fig. 2) showed that the cfFSH-R has the highest similarity to mammalian and avian FSH-Rs (43%–52% overall, 65%–69% in transmembrane regions), followed by mammalian and avian LH-Rs (43%–49% overall, 67%–70% in transmembrane regions) and mammalian TSH-Rs (34%–44 % overall, 68%–69% in transmembrane regions).
In order to be able to determine the relative cfFSH-R mRNA expression levels in different tissues using real-time quantitative PCR, we first determined the PCR efficiency and whether the relationship between Ct and log starting copy number was linear for both primer/probe sets, using defined amounts of testis cDNA. As exemplified in Figure 3, A and B, a linear relationship was detected over four orders of magnitude for both the cfFSH-R mRNA and the cf28S rRNA primer/probe sets, respectively. The correlation coefficients were close to unity and the slope of the standard curves were close to -3.32 (Table 1), indicating maximal PCR amplification. Next, the absolute value of the slope of the Ct versus the log input testis cDNA was <0.1 for both primer/probe combinations (Fig. 3C), allowing the quantification of the relative amounts of cfFSH-R message in the cDNA samples from different tissues using the arithmetic comparative method (the Ct method).
The localization of cfFSH-R mRNA normalized to cf28S rRNA is shown in Figure 4. Catfish FSH-R mRNA expression was greatly enriched in seminal vesicles, testis, and ovary. A borderline signal was also obtained from the cerebellum (cf. Fig. 4). Expression of cfFSH-R mRNA was not observed in the other tissues tested.
In order to prove that pBK-7 contained an insert encoding a functional glycoprotein hormone receptor, we transiently transfected HEK-T 293 cells. Next, various glycoprotein hormones were tested for their ability to increase adenylyl cyclase activity in these transfected cells, using a colorimetric reporter gene assay [31]. As shown in Figure 5A, cfLH (1–5000 ng/ml) and hrFSH (1–10 000 ng/ml) were able to increase the intracellular cAMP levels in a dose-dependent manner, with significantly different (P < 0.02) EC50 concentrations of 30 ng/ml and 130 ng/ml, respectively. The maximal stimulations induced by cfLH and hrFSH each were about eightfold over the basal level (Fig. 5A). Both cfLH and hrFSH had no effect in either mock-transfected (pBK-CMV) or untransfected cells (data not shown). All other ligands tested (hrLH [0.1–50 000 ng/ml], hCG [0.1–100 000 ng/ml], eCG [0.1–100 000 ng/ml] and hTSH [1–10 000 ng/ml]) had no measurable effect on reporter gene expression.
We then tested the possibility that hCG, which evoked a clear steroidogenic response on testis tissue (see below; Fig. 5B), may bind to the receptor but that coupling to G proteins does not occur. To this end, HEK-T 293 cells transiently transfected with pBK-7 were incubated with the EC50 concentration of cfLH (30 ng/ml) to test if increasing concentrations of hCG (1 ng/ml to 10 µg/ml) attenuated the cfLH-induced, cAMP-mediated response on reporter gene expression. None of the hCG concentrations tested led to a significant change in the activity of cfLH on the reporter gene expression (data not shown), suggesting that hCG does not bind to the cfFSH-R.
Finally, we tested all ligands (including cfLH and hCG) for their capacity to increase intracellular inositol phosphate production in HEK-T 293 cells transiently transfected with pBK-7. Significant differences from basal inositol phosphate levels were not observed (data not shown), while HEK-T 293 cells transiently transfected with the catfish GnRH receptor showed a clear inositol phosphate response to chicken GnRH-II, as shown previously [32].
Incubation of testis tissue fragments from mature male African catfish with increasing concentrations of cfLH resulted in a dose-dependent stimulation of androgen secretion (Fig. 5B); half-maximal stimulation was achieved with 23 ng cfLH/ml. Androgen secretion was stimulated in a similar manner by hrFSH which, however, was 10-fold less effective than cfLH (EC50 for hrFSH: 210 ng/ml). Although completely ineffective on the transfected cells using the reporter gene assay, hCG also stimulated androgen secretion dose dependently, showing an EC50 concentration of 869 ng/ml. Equine CG also stimulated androgen secretion (Fig. 5C) but, like hCG, did not stimulate the receptor expressed in HEK-T 293 cells. Human TSH was ineffective in the androgen secretion assays as well.
DISCUSSION
In this study we report on the cloning, tissue expression pattern, structural characterization, and functional expression of an African cfFSH-R, a new member of the glycoprotein hormone receptor subfamily. Comparison of its amino acid sequence with those of other (putative) glycoprotein hormone receptors shows the closest resemblance to the FSH-R subfamily.
Additional evidence indicating that we cloned a gonadotropin receptor, displaying structural characteristics of an FSH-R, and not of an LH-R or a TSH-R, was obtained by partial analysis of the genomic organization of the cfFSH-R gene. While in mammalian LH-R genes two introns are encountered 3' of exon 9, only a single intron was found 3' of the homologous exon 9 in the cfFSH-R genomic PCR product (cf. Fig. 1C), resembling the situation in mammalian FSH-R and TSH-R genes [42, 43]. Moreover, no sequence homologous to the LH-R exon 10 was detected in the cfFSH-R intervening sequence, ruling out the possibility of skipping exon 10 by (alternative) splicing as was found for the marmoset monkey LH-R cDNA [44]. The 5' end of the presumed exon 10 of the cfFSH-R gene as well as the cDNA sequence lack the coding region for 50 additional amino acids found in TSH-R genes [43]. A peculiarity observed in the cfFSH-R, i.e., the absence of 30 amino acids at the junctional region between the N-terminal extracellular domain and the transmembrane domain, is caused by a reduction of 90 nucleotides at the 5' end of the presumed last exon of the cfFSH-R, compared to mammalian LH-R and FSH-R genes. Finally, we recently obtained two types of PCR products, of which the sequences are more homologous to LH-Rs and TSH-Rs, respectively (unpublished results) than to FSH-Rs, further substantiating that pBK-7 encodes a cfFSH-R.
Catfish FSH-R mRNA expression has been detected in seminal vesicles, testis, and ovary. While cfFSH-R mRNA expression in testis and ovary were expected, the detection of cfFSH-R mRNA expression in seminal vesicles may seem surprising. However, the epithelial cells of the catfish seminal vesicles (a structure not homologous to mammalian seminal vesicles) are thought to be homologous to Sertoli cells [45]. This assumption appears to be supported by the present detection of cfFSH-R mRNA in catfish seminal vesicle tissue. The relatively low cfFSH-R mRNA expression in ovary (relative to the expression in testis) may be related to the fact that the ovarian tissue was collected from a postvitellogenic, preovulatory fish. At this stage, FSH binding to granulosa cells was reported to be lower than during active vitellogenesis in the salmon ovary [46].
The amino acid sequences and sizes of each exon in mammalian LH-Rs, FSH-Rs, and TSH-Rs reveal remarkable similarities with the exception of exon 1 in all receptors and exon 10 in LH-Rs, and the introns (all in phase 2) of these genes also share important similarities in spite of different sizes [47]. In addition to the three glycoprotein hormone receptors (i.e., the LH-Rs, the FSH-Rs, and TSH-Rs from different species), several other leucine-rich repeat-containing, G protein-coupled receptors have been identified from several species (see Fig. 2). The overall architecture of these receptors is highly conserved, suggesting their common evolutionary origin. Comparison of the amino acid sequences of these receptors suggests that a common ancestral gene for these receptors may have evolved from a fusion of a gene consisting of leucine-rich repeat-containing exons (encoding parts of the N-terminal extracellular domain) to an ancestral (probably intronless) G protein-coupled receptor gene (encoding the C-terminal half with seven TM domains and the cytoplasmic tail), accompanied by gene duplication events. Evolution of such an ancestral gene to the above mentioned receptor genes might explain the notable differences observed in the N-terminal extracellular domain, and at the junction of the N-terminal extracellular domain and the TM domain.
The entire transmembrane domain of cfFSH-R shares higher than 65% identity to mammalian and avian FSH-R, LH-R, and TSH-R protein sequences, and in general, there is a high degree of homology within the seven transmembrane spanning regions of the G protein-coupled receptor family. A characteristic feature of the glycoprotein hormone receptors within the G protein-coupled receptor family is the presence of a relatively large N-terminal extracellular domain. Inspection of the deduced cfFSH-R amino acid sequence also revealed such a domain containing, however, some features that deviate from mammalian and chicken FSH-Rs.
First, the N-terminal domain contains five potential sites for N-linked glycosylation (Fig. 1A). The first potential site (NTT; amino acids 49–51) is also present in the chicken FSH-R, the second site (NLT; amino acids 95–97) is found only in TSH-Rs, the third site (NGT; amino acids 195–197) is found in all mammalian and avian glycoprotein hormone receptors, the fourth site (NLT; amino acids 272–274) is present only in horse and chicken FSH-Rs, and the fifth site (NVS; amino acids 297–299) is found in all mammalian FSH-Rs but not in the chicken FSH-R. In addition, the potential glycosylation site found in all mammalian and the chicken FSH-Rs is absent in the cfFSH-R (corresponding with amino acids 202–204).
Second, the extracellular domain of cfFSH-R is characterized by imperfectly repeating leucine-rich units of approximately 25 residues in length, first identified in the LH-R [48] and corresponding to one repeat per exon [49]. In a large number of proteins (mostly adhesive proteins and receptors) containing similar motifs known as leucine-rich repeats (LRRs), the N and C termini are flanked by cysteine clusters. The cfFSH-R deviates from the other glycoprotein hormone receptors that are characterized by four similarly spaced cysteines in a stretch of about 20 residues in the amino-flanking region of the LRRs, in containing only two cysteines in the homologous region. The C-terminal cysteine cluster may add a further two strands to the LRR domain, acting as a molecular spacer allowing for the correct location of the extracellular domain in relation to the transmembrane domain, and is involved in ligand-mediated transmembrane signaling [49]. In the cfFSH-R, the 16 residues spacing between the third and fourth cysteine residue in the carboxy-flanking region of the LRRs is much shorter than in other glycoprotein hormone receptors (44–45 residues in FSH-Rs [37 residues in donkey FSH-R], 31–32 residues in LH-Rs [61 residues in the chicken LH-R; the marmoset monkey and common turkey LH-Rs lack the fourth cysteine residue in this region], 82–83 residues in TSH-Rs).
In common with other glycoprotein hormone receptors a highly conserved amino acid sequence (YPSHCCAF), proposed to form a pocket for specific glycoprotein hormone binding [50] and located at the carboxy terminal of exon 9 from the extracellular domain of mammalian LH-Rs, FSH-Rs, and TSH-Rs, is entirely conserved within the cfFSH-R sequence (corresponding to amino acids 275–282 in Fig. 1A).
Both cfLH and hrFSH were able to stimulate androgen secretion by African catfish testis tissue in vitro (Fig. 5B), with EC50 concentrations similar to those found when using cfFSH-R-expressing HEK-T 293 cells (Fig. 5A). However, hCG was unable to elevate cAMP levels in these cells and also did not interfere with cfLH-induced signaling in cfFSH-R-expressing HEK-T 293 cells. Because cfFSH-R expressing HEK-T 293 cells, upon stimulation with all ligands tested (including hCG and cfLH), do not elevate their intracellular levels of inositol phosphates, it is not likely that the inositol phosphate pathway is involved to a significant degree in the cfFSH-R signaling. This suggests that hCG is not bound by the cfFSH-R. Nevertheless, hCG (and other mammalian hormones known to bind to the LH-R in mammals) did stimulate African catfish testicular androgen secretion in vitro. The biological activity of hCG in African catfish (induction of ovulation [51]) and other fish species (spermatogenesis [52]; testicular androgen production [53]) has been known for many years. Apparently the cfFSH-R, although responding well to cfLH, is able to discriminate between hrFSH and hCG. A possibility to explain the steroidogenic response to hCG is to assume that this response is mediated by an additional gonadotropin receptor.
The biological activities of FSH and LH in the mammalian testis are directed to two cell types by the expression of FSH-Rs and LH-Rs being restricted to Sertoli and Leydig cells, respectively [54]; there is little (0.001%–0.1%) cross-activation of the FSH-R by LH or of the LH-R by FSH [55, 56]. On the contrary, the biological activities of the two salmonid gonadotropins appear to be less well separated. A two-receptor model has been proposed for the salmonid testis based on receptor binding studies [46, 57, 58]. This model assumes the presence of a receptor specific for LH (designated type II) that was restricted to Leydig cells and was detected only during the short spawning period. It is possible that the receptor we assume to mediate hCG action in catfish corresponds to this LH-specific receptor. The salmonid gonadotropin receptor type I, however, is unusual when compared to its mammalian and avian counterparts, as it binds both FSH and LH, with merely a slight preference for FSH. However, elevated plasma LH levels are reported for only the short spawning period in the annual reproductive cycle of salmonids while higher FSH levels are associated with the gonadal growth period [12, 59]. Hence, although the salmonid receptor type I has the potential to bind LH, this ligand is unavailable during most of the year. This type I receptor has been detected on Sertoli cells [46]. The fact that no clear signal was obtained for Leydig cells is puzzling because salmon FSH clearly stimulates testicular steroid production, also in prespawning testis where the LH-specific receptor type II cannot yet be detected [60]. Future work will show if techniques more sensitive and discriminating than autoradiographic ligand localization will detect receptors binding LH on Leydig cells of prespawning males, or if Sertoli cell-associated type I receptors stimulate Leydig cell steroidogenesis in a paracrine fashion such as described for FSH in rat [61]. Recent molecular work supports the two-receptor model for salmonid gonads, where one receptor type responds to both gonadotropins [62, 63]. Considering that cfLH stimulates the cfFSH-R with a rather low EC50 concentration of 25–30 ng/ml, the cloned catfish FSH-R may be similar to the salmonid receptor type showing a reduced ligand-binding specifity. This suggests that the situation of a gonadotropin receptor discriminating poorly between LH and FSH may not be a peculiarity in salmonid fish but is possibly a general characteristic of one type of fish gonadotropin receptors. Unfortunately, African cfFSH is not available for experimentation yet.
In summary, the cloning and functional characterization of an FSH receptor from the African catfish provides additional information regarding the conservation of G protein-coupled receptors among vertebrate classes (fishes versus birds and mammals). This receptor will be an important tool in further investigations on the physiology of the regulation of testicular functions in African catfish. One interesting aspect is the possibility that receptors in fish (at least in salmon and African catfish) similar to FSH-Rs of higher vertebrates may not yet have evolved to discriminate clearly between FSH- and LH-type fish gonadotropins. The clear distinction of the cfFSH-R between hrFSH and hCG, on the other hand, may be indicative of a coevolution of the ligands and it might be interesting to study the behavior of cfLH toward the human gonadotropin receptors.
ACKNOWLEDGMENTS
We are grateful to Dr. W.G.E.J. Schoonen (Organon-Oss, The Netherlands) for kindly providing the recombinant and purified human gonadotropins.
FOOTNOTES
1 This work was financially supported by the Swedish Council for Forestry and Agricultural Research (E.A.). 
2 Correspondence: R.W. Schulz, Department of Experimental Zoology, Utrecht University, Padualaan 8, NL-3584 CH Utrecht, The Netherlands. FAX: 31 30 253 2837; r.w.schulz{at}bio.uu.nl
Accepted: January 16, 2001.
Received: April 28, 1999.
REFERENCES
- and gonadotropin II ß-subunit mRNA levels in the pituitary gland of mature male African catfish, Clarias gariepinus. Fish Physiol Biochem 1997; 17:99–108[CrossRef]This article has been cited by other articles:
|
|
 |
|
  E. R. d. Alvarenga and L. R. d. Franca Effects of Different Temperatures on Testis Structure and Function, with Emphasis on Somatic Cells, in Sexually Mature Nile Tilapias (Oreochromis niloticus) Biol Reprod, March 1, 2009; 80(3): 537 - 544. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Garcia-Lopez, J. Bogerd, J. C. M. Granneman, W. van Dijk, J. M. Trant, G. L. Taranger, and R. W. Schulz Leydig Cells Express Follicle-Stimulating Hormone Receptors in African Catfish Endocrinology, January 1, 2009; 150(1): 357 - 365. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Sambroni, F. Le Gac, B. Breton, and J.-J. Lareyre Functional specificity of the rainbow trout (Oncorhynchus mykiss) gonadotropin receptors as assayed in a mammalian cell line J. Endocrinol., November 1, 2007; 195(2): 213 - 228. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Zmora, Y. Kazeto, R S. Kumar, R. W Schulz, and J. M Trant Production of recombinant channel catfish (Ictalurus punctatus) FSH and LH in S2 Drosophila cell line and an indication of their different actions J. Endocrinol., August 1, 2007; 194(2): 407 - 416. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Ko, W. Park, D.-J. Kim, M. Kobayashi, and Y. C. Sohn Biological activities of recombinant Manchurian trout FSH and LH: their receptor specificity, steroidogenic and vitellogenic potencies J. Mol. Endocrinol., January 1, 2007; 38(1): 99 - 111. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Freamat, H. Kawauchi, M. Nozaki, and S. A Sower Identification and cloning of a glycoprotein hormone receptor from sea lamprey, Petromyzon marinus. J. Mol. Endocrinol., August 1, 2006; 37(1): 135 - 146. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W.-K. So, H.-F. Kwok, and W. Ge Zebrafish Gonadotropins and Their Receptors: II. Cloning and Characterization of Zebrafish Follicle-Stimulating Hormone and Luteinizing Hormone Subunits--Their Spatial-Temporal Expression Patterns and Receptor Specificity Biol Reprod, June 1, 2005; 72(6): 1382 - 1396. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-F. Kwok, W.-K. So, Y. Wang, and W. Ge Zebrafish Gonadotropins and Their Receptors: I. Cloning and Characterization of Zebrafish Follicle-Stimulating Hormone and Luteinizing Hormone Receptors-- Evidence for Their Distinct Functions in Follicle Development Biol Reprod, June 1, 2005; 72(6): 1370 - 1381. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.F. Vischer, A.C.C. Teves, J.C.M. Ackermans, W. van Dijk, R.W. Schulz, and J. Bogerd Cloning and Spatiotemporal Expression of the Follicle-Stimulating Hormone {beta} Subunit Complementary DNA in the African Catfish (Clarias gariepinus) Biol Reprod, April 1, 2003; 68(4): 1324 - 1332. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.F. Vischer and J. Bogerd Cloning and Functional Characterization of a Gonadal Luteinizing Hormone Receptor Complementary DNA from the African Catfish (Clarias gariepinus) Biol Reprod, January 1, 2003; 68(1): 262 - 271. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bogerd, W. B. Diepenbroek, E. Hund, F. van Oosterhout, A. C. C. Teves, R. Leurs, and M. Blomenrohr Two Gonadotropin-Releasing Hormone Receptors in the African Catfish: No Differences in Ligand Selectivity, but Differences in Tissue Distribution Endocrinology, December 1, 2002; 143(12): 4673 - 4682. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. B. Cavaco, J. Bogerd, H. Goos, and R. W. Schulz Testosterone Inhibits 11-Ketotestosterone-Induced Spermatogenesis in African Catfish (Clarias gariepinus) Biol Reprod, December 1, 2001; 65(6): 1807 - 1812. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
