| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Medical Research Council Human Reproductive Sciences Unit, The Queens Medical Research Institute, Edinburgh, Scotland EH16 4TJ, United Kingdom
Address all correspondence and requests for reprints to: Kevin Morgan, Medical Research Council, Human Reproductive Sciences Unit, The Queens Medical Research Institute, Little France Crescent, Old Dalkeith Road, Edinburgh, Scotland EH16 4TJ, United Kingdom. E-mail: k.morgan{at}hrsu.mrc.ac.uk.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Furthermore, the possible occurrence of different GnRH peptides in mammals, in addition to classically described GnRH-I, has been a focus of investigation since the discoveries of GnRH-II in chicken and GnRH-III in salmon (8, 9, 10, 11). The gene encoding GnRH-II decapeptide has subsequently been studied in certain mammals but not in cattle. Coding sequences for GnRH-III have not been identified in mammals.
GnRH-II decapeptide differs from mammalian GnRH-I at three positions: codons specify QHWSHGWYPG vs. QHWSYGLRPG in the two genes, respectively. Both mature peptides contain N-terminal pyroglutamate and C-terminal glycinamide, and they possess different conformations in solution (12). They bind with different affinities to GnRH receptor isoforms (types I and II GnRH receptors in mammals) (12). The type I GnRH receptor binds GnRH-I with higher affinity than GnRH-II, and the type II GnRH receptor has higher affinity for GnRH-II.
The progress of genome sequencing projects and the targeted cloning of GnRH precursor genes and GnRH receptors revealed retention of different GnRH system gene complements in certain animal species (13). However, the available genetic data remain incomplete for many species, and therefore, the scope for functional interactions between different GnRH peptides and GnRH receptor isoforms in vivo is not fully understood. Discrepancies in GnRH ligand and GnRH receptor nomenclature (between mammalian and nonmammalian vertebrates) (13) need to be reassessed (and hopefully rectified) as the GnRH systems become characterized in more detail in different vertebrates. We used the mammalian nomenclature described elsewhere (12).
In mammals, the gene for prepro-GnRH-II has been characterized in humans, monkeys, and shrews. However, the prepro-GnRH-II gene has been deleted from the genome of laboratory rats and mice (13), prompting reevaluation of contradictory immunostaining data (14). In animals possessing two GnRH genes, the prepro-GnRH-II gene is expressed, in part, in anatomically distinct regions of the brain, compared with prepro-GnRH-I and special roles for GnRH-II in regulation of puberty, reproductive behavior, feeding, and energy balance have been proposed (15, 16, 17, 18, 19, 20).
In vitro, both types of mammalian GnRH receptor can bind GnRH-I or GnRH-II, and the peptides can elicit intracellular signaling at either receptor (12). In vivo, histologically separate and histologically overlapping hormone receptor networks, possibly linked by currently uncharacterized neurons, may exist in those species possessing more than one type of ligand and receptor (19, 20).
Surprisingly, the effects of GnRH-II on bovine physiology have not been reported in the literature, and recently the presumed existence of a conserved and functional GnRH-II system in cattle has become uncertain because partial DNA sequence information indicated a premature stop codon in the bovine type II GnRH receptor gene homolog (21), its presumed cognate receptor.
Although it is conceivable that GnRH-II may act through the type I GnRH receptor in the absence of a conventional type II GnRH receptor (22), the characterization of the bovine GnRH-II precursor gene is a necessary step to begin to understand the organization of the GnRH system in cattle. Therefore, we analyzed new information from the bovine genome project for GnRH ligand precursor and GnRH receptor DNA sequences. Two ligand precursor genes (one encoding GnRH-I and one encoding a GnRH-II-like decapeptide) and two receptor genes (types I and II) were identified and characterized. A synthetic bovine GnRH-II-like decapeptide (bovine GnRH-II) was prepared and its receptor binding and activation properties were investigated.
| Materials and Methods |
|---|
|
|
|---|
Ovine GnRH-I precursor (GenBank U02517) and type I GnRH receptor exon 3 (GenBank L43841) DNA sequences were used to identify matching sequences in the Ensembl Bos taurus bovine genome database using basic local alignment search tool software (BLASTn).
A bovine GnRH-II precursor gene was identified using human GnRH-II precursor peptide sequence (GenBank NP_001492) to search for short, nearly exact matches encoded in the Bos taurus genome using BLASTx software to translate bovine DNA sequences and select relevant matching peptide sequences.
The bovine type II GnRH receptor gene locus was identified using a bovine expressed sequence tag (GenBank AV604131) to search the Ensembl database for sequence matches. The identity and relative orientation of flanking genes was recorded for each GnRH ligand and GnRH receptor locus using Ensembl contigview. Cross-species DNA sequence alignments were made using National Center for Biotechnology Information BLAST bl2 software. Polypeptide sequences were elucidated using DNA sequence-translation software (ExPASy translate, Swiss Institute of Bioinformatics, Geneva, Switzerland) and multiple sequence alignments were prepared using Clustalw (European Bioinformatics Institute, Hinxton, Cambridge, UK).
Genomic DNA, PCR amplification, cloning, and DNA sequencing
Bovine genomic DNA was purchased from Novagen (Madison, WI). Ovine genomic DNA was prepared from Soay sheep testis tissue provided by Dr. G. Lincoln (Medical Research Council Human Reproductive Sciences Unit, Edinburgh, UK) using standard proteinase k (Sigma, Poole, UK) digestion and phenol/chloroform extraction procedures. Primers for PCR amplification of bovine GnRH-II coding exon 1 were synthesized by MWG Biotech (London, UK): 5'-ATG GGA GCA GCC CTG CTA TG-3' and 5'-GGG GGC CTA GGA GGA TGC TGA GG-3'. PCR was performed using Easy-A DNA polymerase (Stratagene, La Jolla, CA) using standard buffer, 94 C denaturing incubation, 55 C anneal, and 1 min extension at 72 C. PCR products were purified from 1% agarose-Tris-acetate-EDTA gel slices by centrifugal elution in Ultrafree-MC filter devices (Millipore, Bedford, MA) after electrophoresis. DNA size markers were a 100-bp ladder (Promega, Southampton, UK). The DNA was ligated into pCR4 sequencing vector (Invitrogen, Paisley, UK) and transformed into Escherichia coli Top10 (Invitrogen). Individual clones were picked, and plasmid DNA was subjected to automated sequencing using T3 and T7 sequencing primers.
RT-PCR, cDNA cloning, and expression construct preparation
RNA was purified from whole bovine pituitary tissue (also provided by Dr. G. Lincoln, Edinburgh University, Edinburgh, UK) after homogenization in Tri-Reagent (Sigma). Single-stranded cDNA was generated using random oligonucleotide primers (Promega) and superscript reverse transcriptase following the manufacturers instructions (Invitrogen). DNA spanning the full-length reading frame encoding bovine type I GnRH receptor was amplified using specific primers tagged with restriction endonuclease sites (5'-GAT ATC GGT ACC GCC ACC ATG GCA AAC AGT GAC TCT CCT GAA-3' and 5'-AGG CCT TCT AGA TTA TAG AGA GAA ATA TCC ATA TAT AAG-3') and Easy-A DNA polymerase (Stratagene). The purified PCR product was cloned into pCR4. The DNA sequence was confirmed using automated sequencing, and the cDNA was subcloned into the eukaryotic expression vector pcDNA3.1 (Invitrogen).
Qualitative RT-PCR for analysis of bovine type II GnRH receptor gene expression used forward primer 5'-ATG TCT GCA GGC AAC GTC ACC CCT TGG-3' or 5'-GAG ATC ATC TAC AAC CTC TTC ATC TTC-3' and reverse primer 5'-AGA ACC TCC TTC CCT GGA GGA GTC-3'.
Peptides
GnRH-I, GnRH-II, bovine GnRH-II, His-5, and D-Tyr-6-GnRH were made using solid-phase synthesis and were purified using reverse-phase HPLC to more than 98% purity.
Cell culture
COS-7 cells were prepared and transfected with GnRH receptor cDNA-plasmid expression constructs as described elsewhere (23). Mouse LßT2 cells were cultured on matrigel-coated plasticware (BD Biosciences, Franklin Lakes, NJ) in DMEM (Sigma) containing 10% fetal calf serum (PAA Laboratories GmbH, Cölbe, Germany).
GnRH receptor ligand binding assay
Whole-cell binding assays were performed using 125I-iodinated His-5, D-Tyr-6-GnRH, a tracer enabling sensitive binding assays (24). Ten micrograms of peptide were labeled with 1 mCi Na125I using the chloramine T method (Sigma) and purified using Sephadex G25 chromatography (Sigma). Binding analyses were performed in triplicate.
Inositol phosphate assay
Production of inositol phosphate was assayed as described previously (25), using myo-D-[3H] inositol (7590 Ci/mmol, Amersham, Aylesbury, UK). Assay measurements were performed in triplicate using peptide concentrations in the subnanomolar to micromolar range.
Binding and inositol phosphate assay data were analyzed using Sigmaplot 7.0 software (Systat Software Inc., Point Richmond, CA) or Prism software (GraphPad, San Diego, CA).
| Results |
|---|
|
|
|---|
The bovine GnRH-I precursor gene is physically linked to a gene encoding a potassium channel tetramerization domain protein (KCTD-9) and to the gene encoding a protein involved in cytokinesis (DOC-5) (13) (see Fig. 1A
). The coding exons exhibit highest homology to the partial sequence for sheep prepro-GnRH-I (GenBank GON1_SHEEP, AAA03433, U02517), and the bovine data contribute extra information required to construct models for the gene structure in these two species. This includes details of the 5' flanking region and the 23 codons preceding the decapeptide coding sequence (Fig. 1B
).
|
Bovine GnRH-II precursor gene
A gene encoding a variant of the GnRH-II decapeptide precursor was identified on a 3.98-kb genomic DNA fragment (GenBank AAFC01572942). An Ensembl contiguous DNA sequence (ENSBTAG 000 000 00825) shows that this gene is physically linked to genes for 28S mitochondrial ribosomal protein S-26 (MRPS-26) and receptor-type protein tyrosine phosphatase-
, as found with the GnRH-II precursor gene in other mammalian species (13) (Fig. 2A
). The MRPS-26 translation start codon occurs only 314 bp downstream from the GnRH-II precursor gene stop codon, suggesting that the genes overlap each other.
|
The three coding exons for this gene are spread over almost 1 kb genomic sequence (Fig. 2A
), and splice donor and acceptor sequences are well conserved, compared with sequences present in the human GnRH-II precursor gene. Further comparison of the bovine gene with the human homolog (Ensembl ENSG 000 001 25787) indicates sequence similarities between the first two coding exons but divergence in the third coding exon.
An early Ensembl gene structure model (Ensembl gene identity no. ENSBTAG 000 000 00826) was generated ab initio with GENSCAN software (25). It suggested the bovine sequence equivalent to human prepro-GnRH-II coding exon 3 may be divided into three smaller exons separated by two very small introns of 4 and 1 bp, respectively. This is an unlikely model. A more convincing model is that the reading frame in bovine exon 3 terminates earlier than in the human homolog, resulting in a slightly shorter GnRH-associated peptide (see Fig. 3
).
|
The DNA sequence of bovine prepro-GnRH-II coding exon 1 was confirmed by cloning and sequencing multiple isolates of PCR-amplified bovine DNA generated with a high-fidelity proofreading thermostable DNA polymerase (Easy-A; Stratagene; see Fig. 3A
). All clones encoded QHRSHGWYPG (GenBank DQ359716). The same oligonucleotide primers (as detailed above and indicated in Fig. 2B
) were used to clone multiple isolates of sheep prepro-GnRH-II coding exon 1. DNA sequences derived from both strands of the plasmid clones were translated, and amino acid sequences were aligned with other mammalian prepro-GnRH-II sequences. This revealed that the tryptophan codon at position seven of GnRH-II is replaced by a stop codon in sheep DNA, QHWSHG*YPG (Fig. 3B
and GenBank DQ359717).
Bovine type I GnRH receptor gene
The type I GnRH receptor was identified in a genomic DNA region spanning approximately 17.5 kb on bovine chromosome 6 (Ensembl identification no. ENSBTAG 000 000 27542). The gene is flanked by genes encoding airway trypsin-like protease and a ubiquitin-activating enzyme, a linkage group observed in other mammals (Fig. 4A![]()
) (13). Details are contained in Ensembl ENSBTAG 000 000 00438. Coding exon 1 encompasses GnRH receptor transmembrane domains 14, over a total of 175 codons. The following intron extends for 12.5 kb. Coding exon 2 comprises 73 codons from the end of transmembrane domain 4 to the middle of intracellular loop 3. The next intron is 3.5 kb and coding exon 3 contains 80 codons.
|
|
Bovine type II GnRH receptor gene
A gene encoding a type II GnRH receptor is located on Ensembl GeneScaffold_5442 (Ensembl gene identity no. ENSBTAG 000 000 08370) on bovine chromosome 3. It spans 5 kb and is flanked by genes encoding peroxisomal membrane protein 11ß and RNA-binding motif protein 8A (Fig. 4A![]()
), as described in other mammalian species (13, 21, 26). The Ensembl receptor gene model generated ab initio by GENSCAN suggested 14 coding exons, but this can be simplified in view of the number of exons found in the homologous gene in other species (i.e. three) and by considering the existence of three coding exons in type I GnRH receptor genes. Thus, there are probably only three coding exons in the bovine type II GnRH receptor gene homolog (see Fig. 4![]()
). The first coding exon is disrupted by a deletion of 2 bp, which causes a frame shift in the region for receptor transmembrane domain 3 (Fig. 4C![]()
). This is a unique feature, although most of the DNA sequence is highly conserved relative to the sheep type II GnRH receptor gene homolog (94% sequence identity in this exon), including the splice donor sequence (GTGAGT).
Further comparison with the sheep gene shows that 21 of the 170 codons in the first exon are altered, with five silent changes, 14 amino acid alterations, and two amino acid deletions (one deletion in the bovine sequence and one in the sheep sequence, see Fig. 4C![]()
). The DHQ motif encoded in intracellular loop 2, also present in the sheep sequence (26), reflects divergence from the DRQ motif in primate type II GnRH receptor sequences.
The intron sequence is highly conserved relative to the sheep sequence (91% identity) and is of a similar length. The splice acceptor sequence is completely conserved (ATCTAG).
There is an apparent premature stop codon in coding exon 2, as described previously (21), followed by a frame shift 32 codons downstream potentially caused by deletion of 1 bp in the region encoding transmembrane domain 5. Further analysis in this region showed that there is in fact a deletion of 13 bp relative to the sheep sequence (TCTTTCTGAGGCC in sheep exon 2). Correction of the frame shift by insertion of 1 bp would result in a 5-amino acid residue deletion in transmembrane domain 5 (Fig. 4C![]()
).
The sequence of the following intron is well conserved (92% identity to the sheep sequence), but the splice donor sequences are divergent, whereas the splice acceptor sequences are identical with the sheep (TCCCAG).
Two premature stop codons occur in the third coding exon, in the region encoding the cytoplasmic tail domain. Interestingly, the first stop codon (TGA) is adjacent to the end of transmembrane domain 7 and may have arisen from a single base change in a codon for arginine (CGA). The second premature stop codon in this exon occurs 72 bp downstream and may have arisen from mutation of a glutamine codon (CAA to TAA). A frame shift occurs 12 bp downstream from the second premature stop codon. This results in deletion of two amino acid residues (Fig. 4C![]()
) and may have arisen from the loss of 4 bp. Translation of the alternative reading frames indicates that coding sequences for the C-terminal residues (ISITSI) have been lost, although the 3' flanking region linked to the RNA-binding motif protein 8A gene is similar in length to the sheep sequence.
When translated, the bovine type II GnRH receptor protein sequence exhibits 80% similarity and 75% identity to the marmoset type II GnRH receptor, indicative of poorer conservation relative to the level of sequence conservation in the type I GnRH receptors.
RT-PCR analysis of bovine pituitary RNA demonstrated expression of the type I GnRH receptor gene, but spliced transcripts from the type II GnRH receptor gene were not detectable using primers targeting the coding exons (Fig. 4B![]()
).
Binding assays for bovine GnRH-II peptide
COS-7 cells were transfected with expression constructs for bovine or human type I GnRH receptor or marmoset type II GnRH receptor, and binding assays were performed using iodinated His-5,D-Tyr-6 GnRH. Unlabeled GnRH-I or GnRH-II competed with binding of the radiolabeled ligand most effectively at their respective cognate receptors, with dissociation constants of 9.7, 1.8, and 0.9 nM, respectively (Fig. 5A
, D, and E). However, the bovine GnRH-II peptide could not effectively displace radiolabeled ligand binding from either the bovine or human type I or marmoset type II GnRH receptors (Fig. 5
, C, F, and G), even when added at micromolar concentrations. In addition, micromolar bovine GnRH-II did not displace tracer binding at the mouse type I GnRH receptor endogenously expressed at high level in the gonadotrope cell line LßT2 (Fig. 6
).
|
|
|
| Discussion |
|---|
|
|
|---|
Bovine prepro-GnRH-I and type I GnRH receptor genes are intact. However, analysis of the information contained within the DNA sequences for the bovine prepro-GnRH-II and the type II GnRH receptor genes raises questions concerning their functionality. The bovine prepro-GnRH-II gene encodes an unusual decapeptide containing arginine (Arg) in position 3 of the mature peptide, and the type II GnRH receptor gene is disrupted by frame shifts and premature stop codons. These new data extend what is understood about species-specific differences in the organization of the GnRH system.
We confirmed the DNA sequence for coding exon 1 of the bovine prepro-GnRH-II gene. The substitution of tryptophan at position 3 (Trp3) in bovine GnRH-II by arginine is a unique occurrence; it has not been described in any natural GnRHs (12). Conservative replacement of Trp3 with a polar aromatic tyrosine residue occurs naturally in guinea pig GnRH-I (12, 27) and lamprey GnRH-I (12, 28). Position 3 is important for GnRH receptor activation, because D-Trp3 GnRH analogs behave as antagonists and replacement of Trp3 with nonpolar phenylalanine produces GnRH analogs with low biological activity (29). Arg3 has not been described in the large number of GnRH analogs synthesized, although replacement of position 3 with a variety of residues is associated with loss of activity (29).
Because substitutions of Trp3 are used in GnRH antagonist design, we contemplated that bovine GnRH-II might be a naturally occurring antagonist.
However, our data indicate that substitution of position 3 with Arg prevents GnRH-II binding to GnRH receptors. We showed that bovine GnRH-II has no binding or inositol phosphate stimulating activity at mammalian type II or type I GnRH receptors. Bovine GnRH-II did not bind to GnRH receptors artificially expressed in COS-7 cells or GnRH receptor endogenously expressed in specialized GnRH-responsive pituitary gonadotrope cells (mouse LßT2).
In the absence of binding, we would not expect activation of intracellular signaling. The bovine GnRH-II peptide had no ability to inhibit the stimulation of inositol phosphate production by GnRH-I or GnRH-II at the type II or type I receptors (data not shown). Considering these results it seems unlikely that bovine GnRH-II elicits or modifies alternative signaling events via conventional GnRH receptors.
Further considerations suggest functional inactivation of bovine GnRH-II. For example, assuming the gene is expressed in bovine tissues, the position of the basic Arg3 may be expected to affect correct posttranslational cleavage of the propeptide such that a decapeptide is not generated in vivo. In addition, oligonucleotide primers based on the bovine prepro-GnRH-II gene were used to determine the DNA sequence of part of the homologous sheep gene. Significantly there is a premature stop codon (TGA) at position seven of the region encoding the mature GnRH-II peptide in sheep (i.e. QHWSHG*YPG). An identical truncated peptide sequence is encoded in chimpanzee DNA (13) (see Ensembl ENSPTRG 000 000 13187). Thus, in the absence of an intact reading frame, it is unlikely that GnRH-II is biosynthesized in these species. Therefore, considering their relatively close evolutionary relationship, the data from sheep are consistent with the likely functional inactivation of GnRH-II in cattle.
Loss of functional ligand through mutation is also consistent with coevolutionary loss of cognate receptor function. Using sequence alignment, we confirmed that the bovine type II GnRH receptor gene is disrupted due to multiple alterations relative to functional GnRH receptor genes. Also, spliced RNA transcripts derived from the bovine type II GnRH receptor gene were not detectable in pituitary cDNA, suggesting receptor silencing in this tissue, in contrast to the situation in marmoset pituitary, in which the type II GnRH receptor gene is expressed (30).
A major question is why a loss of functional GnRH-II and/or its cognate receptor has occurred in certain mammalian species, such as cattle, sheep, laboratory rodents, chimpanzee, and man but not in others, such as the marmoset (see Table 1
). It is possible that this loss represents a null mutation in certain species or that other neurohormonal systems have replaced the ancestral role of GnRH-II or its cognate receptor. Perhaps GnRH-I performs these functions in certain species. Interestingly, the prepro-GnRH-II gene is intact in jawed teleosts and amphibians (13), and receptors phylogenetically similar to the mammalian type II GnRH receptor are functional in amphibians (13, 31). However, whether one particular receptor isoform is functionally dedicated to recognition of GnRH-II in these animals has not been resolved (31, 32). Analyses of the type II GnRH receptor family, their evolutionary history, and functional adaptations are ongoing, including jawless teleosts in which the status of the prepro-GnRH-II gene homolog is unknown (28, 33, 34, 35).
|
In mammals, the biological effects of GnRH-II within the brain have been examined in female marmoset monkeys (36), female shrews (15, 16, 17), and female mice (37). Intracerebroventricular injection of GnRH-II stimulates female reproductive behavior in each of these animals and is also associated with decreased food intake in shrews and mice. The experimental data implicate a cognate receptor for GnRH-II (i.e. a type II GnRH receptor) as the site of action in each of these species because GnRH-I does not elicit the same response and type I GnRH receptor antagonists do not block the effects of GnRH-II. However, this conclusion with respect to the mouse (37) must be questioned in the absence of genetic evidence for the existence of a GnRH-II system in this species (13).
Perhaps this paradox may be resolved when the neurotransmitter phenotype and the extent of synaptic connections between neurons secreting GnRH peptide and those possessing GnRH receptors are better understood. In this respect, advances in genome sequencing projects should enable selection of animal models based on detailed knowledge of the genetic variations in the organization of the GnRH system (see Table 1
for comparisons). For example, preliminary findings suggest that prepro-GnRH-II and type II GnRH receptor genes may also be disrupted by reading frame mutations in the rabbit and dog genomes (see Ensembl database). Nevertheless, the data derived from our studies of cattle and sheep should contribute to the further elaboration of models addressing the neuronal regulation of reproductive behavior (38).
| Footnotes |
|---|
First Published Online August 17, 2006
Abbreviations: MRPS-26, Mitochondrial ribosomal protein S-26; Trp3, tryptophan at position 3.
Received February 21, 2006.
Accepted for publication August 4, 2006.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. S. Schneider and E. F. Rissman Gonadotropin-releasing hormone II: a multi-purpose neuropeptide Integr. Comp. Biol., November 1, 2008; 48(5): 588 - 595. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. A. Tello, S. Wu, J. E. Rivier, and N. M. Sherwood Four functional GnRH receptors in zebrafish: analysis of structure, signaling, synteny and phylogeny Integr. Comp. Biol., November 1, 2008; 48(5): 570 - 587. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Faurholm, S. Cochrane, R. R Millar, and A. A Katz Gene structure and promoter functional analysis of the marmoset type II GnRH receptor J. Mol. Endocrinol., August 1, 2007; 39(2): 91 - 104. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z.-L. Lu, M. Coetsee, C. D. White, and R. P. Millar Structural Determinants for Ligand-Receptor Conformational Selection in a Peptide G Protein-coupled Receptor J. Biol. Chem., June 15, 2007; 282(24): 17921 - 17929. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |