A sequence of progesterone receptor homologue in freshwater crocodile (Crocodylus siamensis)

Crocodile farming has been established in Thailand for over 60 years ( Ratanakorn et al . 1993) and is currently performed on a large scale. Today over 1000 established crocodile farms in Thailand are housing over 80 000 crocodiles. Skin of 3–4 year old crocodiles is the main product from crocodile farming. Two different species are currently being farmed in Thailand, the freshwater crocodile ( Crocodylus siamensis ) and the saltwater crocodile ( Crocodylus porosus ). The saltwater crocodile is very difficult to farm. It is a fierce animal, very territorial and prone to fighting when penned in high densities. In nature less than 1% of all hatchlings will reach maturity because of predators, lack of food and social pressures (territorial males will kill and eat juveniles). However due to the lack of ventral osteoderms, this species provides the most valuable Crocodilian hide. The freshwater crocodiles are smaller than the saltwater crocodiles. The males are always bigger and grow faster than the females. Freshwater animals reach maturity earlier than saltwater crocodiles at around 10 years. The species are bred extensively in captivity in Thailand and knowledge that is based on long‐term farming operations exists. The hybrid crosses between freshwater and saltwater crocodiles are used for the skin trade because of their superior growth rates and increased yield. The skins from these crosses are valuable for commercial use. The species is considered to be relatively inoffensive and only a moderate threat to humans compared with the saltwater crocodile. Much research has been carried out on temperature‐dependent sex determination (TSD) in crocodiles, which is important for breeding programmes to ensure the desired sex ratio, or to produce faster growing males for farming purposes (Lang & Andrews 1994). The steroid hormone, progesterone, is known to be important in the development and regulation of normal reproductive tissues in all vertebrates ( Evans 1988). The effect of this hormone is mediated by its binding to the progesterone receptor (PR) found in the target tissues. Recently, a mouse model carrying a null mutation of the PR gene was found to exhibit pleiotropic reproductive abnormalities (Lydon et al . 1995). The results strongly support the belief that progesterone’s role has developed as an essential and pleiotropic coordinator of diverse physiological events during the evolution of the reproductive system. The PR is composed of two naturally occurring ligand‐binding forms, termed PR‐A and PR‐B ( Schrader & O’Malley 1972), which derive from the same gene (Conneely et al . 1987; Kastner et al . 1990). Complete amino acid and nucleotide sequences from chicken (Conneely et al . 1987; Gronemeyer et al . 1987), human (Misrahi et al . 1987) mouse (Schott et al . 1991), and rabbit (Loosfelt et al . 1986) have been established and indicate a high level of similarity. The molecular cloning of the avian PR demonstrated that this receptor is a member of the steroid/nuclear receptors superfamily that includes receptors for steroids, thyroids, retinoids, and vitamin D (Conneely et al . 1986; Tsai & O’Malley 1994). The PR, as all members of this superfamily, is defined by a common structural motif that is organized into defined domains in terms of structure and function (Carson‐Jurica et al . 1990b). The single‐copy transcription unit of the PR consists of eight exons. The conserved functional domains of the receptor protein are encoded by distinct exons. The most conserved region, the DNA‐binding domain, is located centrally in the molecule and is encoded by distinct small exons 2 and 3 (Huckaby et al . 1987). This domain is determined by two type II zinc fingers which facilitate DNA binding to response elements (Umesono & Evans 1989). The moderately conserved ligand binding domain is located in the carboxy‐terminal part of the receptor and is encoded by exons 5 to 8 (Huckaby et al . 1987). It contains sequences required for receptor dimerization and target gene trans ‐activation (Gronemeyer 1991). The short variable hinge region, encoded by exon 4 (Huckaby et al . 1987), is located between the DNA‐ and the steroid‐binding domains and contains a nuclear localization signal (Carson‐Jurica et al . 1990a). The amino‐terminal region of the PR, encoded by exon 1 (Huckaby et al . 1987), is the most hypervariable region of both size and sequence among members of this superfamily. In crocodiles, the incubation temperature of the eggs determines the sex. This phenomenon is either termed temperature dependent sex determination (TSD) or environmental sex determination (ESD). Very little is known about the function of sex steroid hormones in these animals. As a contribution to understanding the molecular mechanisms of TSD/ESD in animals, a homologue of PR from the crocodile oviduct was amplified and sequenced using polymerase chain reaction (PCR) methods. Materials and methods Amplification of crocodile progesterone receptor cDNA using reverse transcription‐polymerase chain reaction Male and female laryngeal muscles and oviduct tissue were collected from freshwater crocodiles ( C. siamensis ) at the age of 3–4 years and kept in liquid nitrogen during transportation. Approximately 300 mg of each sample was homogenized in 5 ml of denaturation solution (4  m guanidium thiocyanate, 25 m m Na‐citrate pH 7.0, 0.5% sarcosyl and 0.1  m 2‐mercaptoethanol) by homogenizer Polytron ® System (Zurich, Switzerland). Approximately two‐thirds of the homogenate and one‐third of CsCl, 5.7 m , 25 m m NaAc pH 5.2 was centrifuged at 23°C, 90 000 g for 18–20 h in order to extract total RNA. The RNA pellet was then dissolved in DepC H 2 O, phenol (H 2 O saturated) and chloroform (25 : 1, chloroform : isoamylalcohol). The mixture was centrifuged at 4°C, 13 000 rpm for 5 min. RNA in the upper part of solution was precipitated with 1/20 volume of 5  m NaCl and 2 volumes of absolute ethanol and kept overnight at −20°C. The mixture was then centrifuged at 4°C, 13 000 rpm for 30 min. the pellet was carefully washed with 75% ethanol and centrifuged at 4°C, 20 000 g for 5 min. After air drying the pellet was resuspended in DepC H 2 O and incubated at 45°C for 5–10 min. The total RNA was quantified using a spectrophotometer (Perkin‐Elmer, Norwalk, CT, USA). Quality of the RNA was assessed by visual inspection on formaldehyde agarose gel. Reverse transcriptase (RT)‐PCR was performed with 2.5 μg of total RNA extracted from oviduct tissue, using 20 μ m of cross‐species primer (primer No. 6, below) homologous to the sequence previously described (Conneely et al . 1987) and AMV reverse transcriptase (Promega, Wallisellen, Switzerland). Conditions were at 70°C for 5 min, ramp to room temperature in 10 min, 42–55°C for 75 min and finally at 70°C for 15 min. The primer sequences used for each PCR amplification were either specific (primer No. 3–5) or cross‐species primers [primer No. 1–2 (from frog androgen receptor) and No. 6 (from chicken PR)]. The cross‐species primers were designed based on the highly conserved regions of published nucleotide sequences for androgen and progesterone receptors from different species. Figure 1 illustrates the position of the primers. The primer sequences are as follows: 1Nucleotide sequence and predicted amino acid sequence of the crocodile progesterone receptor. The underlined sequences indicate the position of each primer Primer 1 5′ GTC TGA TCT GTG GCG ATG AGG C 3′ Primer 2 5′ CTT GTC AAT TGT GCA GTC ATT TCT 3′ Primer 3 5′ GGG CAG CAT AAC TAT TTA TGT GC 3′ Primer 4 5′ GCA CCT GAT TTG ATC CTA AAT GA 3′ Primer 5 5′ GTG CAA ATT CCG GAA TCC 3′ Primer 6 5′ GAA GAG GTT TCA CCA TCC CTG C 3′ Complementary DNA was amplified in 25 μl of reaction mixture containing: 5 μl of RT product, 1.25 m m dNTP’s, 20 μ m of each primer, Taq high fidelity polymerase and 10× buffer with 15 m m MgCl 2 (Boehringer Mannheim, Rotkreuz, Switzerland). Conditions for PCR were 95°C, 30 s (denaturation), 48–56°C, 30 sec (annealing) and 72°C, 1–2 min (extention) for 35–40 cycles. PCR products were analysed on gels. Approximately 100 mg of gel containing a fragment in the range of the expected size were extracted by Gel Extraction Kit (QIAEX II, Qiagen, Basel, Switzerland). All products from the gel extraction were sequenced using the ABI PRISM TM Dye Terminator cycle sequencing ready reaction kit (Perkin‐Elmer) and analysed by an automated method with the 373 DNA sequencing system (Applied Biosystems, Perkin‐Elmer). Northern blot hybridization According to standard procedures, approximately 15 μg of total RNA, isolated from each tissue sample, was electrophoretically separated on a 1.25% formaldehyde agarose gel and blotted onto Duralose‐UV TM Membranes (Stratagene, Basel, Switzerland). Using the primers No. 1 and No. 2, 25 ng of the crocodile cDNA fragment were labelled with [ 32 P]‐dATP using a random primer labelling kit (Prime It II, Stratagene). Hybridization was carried out overnight at 42–45°C and then washed under high stringency conditions several times using 2× standard saline citrate (SSC), 0.1% sodium dodecyl sulfate (SDS) at room temperature for 5–10 min and 0.1× SSC, 0.1% SDS at 50–65°C for 5–10 min. The density of the signals was observed by a Kaiker counter until an empirical optimal density was reached. The membranes were then exposed to X‐ray films for 1, 3 and 5 days depending on appearance of the signals. Fluoresence in situ hybridization Chromosome preparations were obtained from fibroblast cultures. The Q‐banding was performed using 0.002% quinacrine mustard for 30–60 s, prior to FISH. Well‐spread metaphases were photographed using a Leitz Laborlux K fluorescence microscope coupled to a CCD camera (Photometrics, Tucson, AZ, USA), controlled by a Power Macintosh 7100. Image acquisition was performed using the software package IPLab Spectrum 2.5.7 (Signal Analytics, Fairfax, VA, USA). Approximately 1 kb of crocodile cDNA probes were labelled with biotin‐16‐dUTP using a random primer labelling kit (Prime It ® FLUOR, Stratagene) followed by a dot blot assay (BluGene Kit BRL, Basel, Switzerland) as a control. Fluorescence in situ hybridization was performed according to standard procedures in livestock species (Solinas et al . 1995). Results Amplification of the crocodile progesterone receptor cDNA The fragments of crocodile cDNA (sequence data described is available at GenBank under Accession No. AF030321 ) were consistently found to encode 1039 nucleotides, containing 346 deduced amino acids ( Fig. 1). The amino acid and cDNA nucleotide sequences of the crocodile PR fragment were compared with sequences from chicken (Conneely et al . 1987), lizard (Young et al . 1995), human (Misrahi et al . 1987), mouse (Schot et al . 1991), and rabbit (Loosfelt et al . 1986) based on the GCG package analysis (Devereux et al . 1984). The amino acid sequence with greatest similarity was that of the chicken (91%) but the human, lizard and mouse also showed a high degree of similarity (87%). The sheep and rabbit sequences revealed approximately 85% amino acid similarity. At nucleotide level, the fragment showed 87.7% overall identity to chicken PR, and identities with the other species were in the range of 79.0–81.1% due to a few inserted or deleted nucleotides. Compared with the complete sequence of chicken PR mRNA (containing eight exons), that of the crocodile fragment contained seven exons, beginning from terminal‐half of exon 2 to the beginning of exon 8. Each part of the crocodile sequence was compared with corresponding exon regions (2–8) of PR in other vetebrates ( Fig. 2). The crocodile homologous to the DNA‐ and steroid‐binding domains, separated by the hinge region, are also shown. The DNA‐binding domain (exons 2 and 3) showed nucleotide identities from 81.8 to 97.6% while those of steroid‐binding domain (exons 5–8) ranged between 71.2 and 89.3%. The identity of the hinge region (exon 4) was 84.4%. The overall homology of crocodile PR when compared with frog androgen receptor (AR) (Fischer et al . 1995), human AR (Chang et al . 1988), human glucocorticoid receptor (GR) (Hollenberg et al . 1985) and human mineralocorticoid receptor (MR) (Arriza et al . 1987), ranged from 60.4 to 62.0% with variation in size. 2The illustration of the crocodile cDNA fragment compared with the complete genomic picture of chicken PR. The coding sequence is indicated by dark and shaded boxes, the untranslated regions by white boxes, and the intron sequence by single lines. Dot lines indicate parts of the crocodile nucleotide sequence which are homolog to the chicken PR exons. The data in percentage represents sequence identity for each part of the crocodile PR homolog with different species. The data in parenthesis shows overall sequence identity with the crocodile PR homolog.Northern blot hybridization and fluorescence in situ hybridization Using Northern blot hybridization, a single transcript of approximately 3–4 kb was detected in the oviduct and laryngeal muscles of young crocodiles with a lower intensity of the signal on male laryngeal tissue ( Fig. 3). Using fluorescence in‐situ hybridization (FISH) technique, crocodile PR cDNA probes were mapped to a subtelomeric region of the p‐arm of the longest chromosome putative chromosome 1 ( Fig. 4). 3Northern blot analysis demonstrates that only a single transcript of approximately 3–4 kb is expressed4Mapping of the crocodile progesterone receptor homolog by FISH. (A) metaphase spread with Q‐stained chromosomes. (B) The same metaphase spread with the hybridization signals on a subtelomeric region of the p‐arm of the longest chromosome [putative chromosome; Chr.1]Discussion Comparison of the crocodile sequence with published steroid receptors in different species by GenBank Database indicates that the sequence described indeed represents a fragment of the gene encoding the progesterone receptor. It is not surprising that the most conserved region is the DNA‐binding domain but the region of greatest diversity is the so called ‘hinge region’ which has been also found in other species ( Conneely et al . 1987; Misrahi et al . 1987; Loosfelt et al . 1986). High homology of conserved DNA‐ and steroid‐binding regions were found among the steroid receptor superfamily, especially in the same subgroup. In addition, PR, AR, GR, and MR are the steroid receptors carrying the GS‐V motif in the P‐box and were therefore considered as a group (Luisi et al . 1994). The low degree of homology between PR and AR, GR, and MR could obviously be recognized in this case. Furthermore, the result from Northern blot corresponded to previous observations on PR transcript (3.7–5.2 kb) in other animals ( Conneely et al . 1987; Gronemeyer et al . 1987; Misrahi et al . 1987). While the expressions of AR (Luisi et al . 1994) and GR (Yang et al . 1992) were found at the positions 8–9 kb and 5–7 kb, respectively, no information of blot analysis on MR has been reported yet. The stronger signal found in female laryngeal tissue by Northern blot indicated a different degree of gene expression in male and female crocodiles. the location of PR gene, demonstrated by FISH in other species, was found on autosomes as chromosome 1 in chicken (Dominguez‐Steglich et al . 1992), chromosome 9 in mouse (Naylor et al . 1989), and chrosome 11 in human (Rousseau‐Merck et al . 1987). Sex steroid receptor genes in the crocodile species have never been reported previously. The high degree of similarity in this study confirmed the strict conservation of sex steroid hormones from ancient animals to modern ones. Further studies will try to produce a specific antibody to the progesterone receptor using the crocodile PR sequence to establish a sex specific test. These data indicate that a characterization of genes functioning early in the cascade of sex determination would be an appropriate model to understand the genetic mechanisms of temperature dependent and environmental sex determination in animals, respectively. Acknowledgements We are grateful to Dr P anya Y oungprapakorn and staff (the Samuthprakarn crocodile farm and zoo in Thailand) to provide us with the specimens. We thank Dr M eena S arikaputi and Dr W eerapong K oykul (Chulalongkorn University, Thailand) for all assistance. This study was supported by the Department of Animal Science, Federal Institute of Technology, Zurich, Switzerland and The Federal Swiss Scholarship Programmes. A homologue to the progesterone receptor mRNA in freshwater crocodiles ( Crocodylus siamensis ), encoding 1039 nucleotides, was established by reverse transcriptase‐polymerase chain reaction. The cDNA fragment, amplified and sequenced from oviduct tissue, was found to contain seven exons which encoded a deduced polypeptide of 346 amino acids in length with a high nucleotide and amino acid identity to known progesterone receptor sequences in other animals. Comparison of the crocodile cDNA fragment and the chicken progesterone receptor gene indicated the presence of DNA‐ and steroid‐binding domains, including the hinge region. In addition, young freshwater crocodiles showed a sex‐specific expression of this gene in laryngeal muscles recognized by Northern blot analysis. The cDNA probes were biotin‐labelled and mapped by FISH to a subtelomeric region of the p‐arm of the longest chromosome (putative chromosome 1). This is a contribution to a new approach towards understanding the molecular mechanisms of temperature‐dependent sex determination.