| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Biology (B.A.A., J.A.T., C.W., G.O.M., N.M.S.), University of Victoria, Victoria, British Columbia, Canada V8W 3N5; Clayton Foundation Laboratories for Peptide Biology (J.E., W.V., J.E.R.), The Salk Institute, La Jolla, California 92037; and Ferring Research Institute, Inc. (D.J.H., K.O.A.), San Diego, California 92121
Address all correspondence and requests for reprints to: Nancy M. Sherwood, Department of Biology, P.O. Box 3020 STN CSC, University of Victoria, Victoria, British Columbia, Canada V8W 3N5. E-mail: nsherwoo{at}uvic.ca.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Both tunicate genes are expressed as mRNA early and throughout development, measured at the stages of four-cell, gastrulation, tail release, and tail resorption. In a closely related tunicate species, Ciona savignyi, we used in silico analysis to identify two similar genes encoding six peptides, only one of which is unique compared with C. intestinalis.
Immunohistochemistry showed that at least one GnRH peptide was in the nerve net that surrounds the dorsal strand. Synthetic forms of the seven novel tunicate peptides induced release of gametes in adult tunicates. In contrast, the peptides did not activate the human GnRH-I receptor or cause release of LH in a rat pituitary cell assay. These data provide insight into the structural evolution of the GnRH peptides and their genes and show a functional role for GnRH in tunicate spawning.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A GnRH structure of 10 amino acids was first isolated in mammals (5, 6). At present, 16 GnRH forms of 10 amino acids each (7) plus 1 form of 12 amino acids in octopus (8) have been identified, mainly by protein isolation and sequencing techniques. In vertebrates, each form of GnRH is encoded on a separate gene. Mammals, including humans, usually express two forms of GnRH, mammalian (m)GnRH (also known as GnRH-I) and chicken (c)GnRH-II (also known as GnRH-II), although a few species additionally express guinea pig (gp)GnRH. In evolution, mGnRH is first detected in early derived bony fish, and cGnRH-II is detected even earlier in cartilaginous fish. Other forms of GnRH preceded mGnRH and cGnRH-II in evolution. In the teleosts (later-evolving bony fish), most species studied to date express three different forms of GnRH: cGnRH-II, salmon (s)GnRH, and a third form that can vary in structure among species.
The multiple forms of GnRH are distinguished by their separate anatomical location in the brains of mammals (9) and fish (10). Neuronal cell bodies containing cGnRH-II are mainly in the midbrain region for all species, whereas the other GnRH forms are in neurons in the anterior brain. In species with only two forms of GnRH (e.g. most tetrapods and some fish), cGnRH-II is mainly in the midbrain, and the second form is in both the olfactory region/terminal nerve (OLF/TN) and ventral telencephalic/preoptic (VT/PO) region. In species with three GnRH forms, cGnRH-II is in the midbrain, but the two GnRH forms in the anterior brain may be in separate anterior locations or overlap within the same region (11).
Our current understanding of GnRH in invertebrates is derived primarily from three GnRH structures. We identified two GnRH peptides by primary structure as tunicate GnRH-1 and -2 (tGnRH-1 and tGnRH-2) from the protochordate, Chelyosoma productum (12). More recently, a GnRH of 12 amino acids has been identified by protein sequence and cDNA isolation in octopus (8). These studies establish that GnRH is present in protochordates and in animals that evolved earlier, but all of the GnRH forms may not be identified because of the lack of specific antisera and bioassays.
Identification of an ancestral GnRH gene with subsequent duplications or mutations has been approached in several ways. First, the distribution of different GnRH forms with known peptide structures was mapped among animals with an established place in evolution based on the fossil record and morphological comparisons. Second, phylogenetic analysis was used based on the cDNAs for the coding part of the precursors. Third, a linkage method was used to map each GnRH gene and the nearest upstream gene to determine GnRH orthologs in humans and medaka (13). GnRH-coding genes with the same upstream gene were considered to be orthologs. The nearest upstream gene was identified for each of the three GnRH forms in medaka and two forms in human. A highly conserved protein (FLJ20038) preceded the medaka form of GnRH (mdGnRH) and human mGnRH; protein tyrosine phosphatase 
(PTP) was the upstream neighbor for cGnRH-II in both species; and PTP
preceded sGnRH in medaka. This type of linkage analysis for GnRH has not been used to date in tunicates.
In the present study, we began by searching the complete genome for GnRH genes in two tunicate species, Ciona intestinalis and Ciona savignyi. We then used molecular biological techniques to isolate and sequence the gene and cDNA structures from C. intestinalis. In addition, the alternative splicing and expression pattern of mRNA for both genes in C. intestinalis were examined during early development and in adults. tGnRH was localized using immunohistochemical techniques. All six novel GnRH peptides from C. intestinalis and a seventh from C. savignyi were synthesized and tested for biological activity. To examine these novel tGnRH peptides for therapeutic potential and to better understand the nature of the interaction between ligands and mGnRH receptors, we tested activity of the tGnRHs in stimulating LH release in a rat pituitary cell assay and activation of the human GnRH-I receptor using a gene reporter assay. We also examined whether the nearest upstream gene was the same as in medaka fish and human GnRH genes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Identification of the C. savignyi GnRH genes used sequence data from the Whitehead Institute C. savignyi sequencing project (http://www-genome.wi.mit.edu/annotation/ciona/). This database was limited to nucleotide similarity searches using MEGABLAST. To generate an amino acid-searchable database, the genome-read sequences were downloaded and compiled into a local blast database. These sequences were then compared using TBLASTN analysis with the same parameters described above. Each of the C. intestinalis GnRH forms was used to search and resulted in similar blast matches allowing for the characterization of two C. savignyi gene arrangements.
To determine the transcription start site for each gene, a sequence of 1000 bp that was 5' to the GnRH peptide coding region was entered into the neural network promoter prediction site (http://www.fruitfly.org/seq_tools/promoter/html) using the default minimum promoter score of 0.80.
Animals
Adult C. intestinalis (subphylum Tunicata, class Ascidiacea) were obtained from Woods Hole Biological Station (Woods Hole, MA) and treated under the guidelines of the Animal Care Committee at the University of Victoria. The neural (central) ganglion and neural gland, as well as gonads and intestine, were dissected and frozen in liquid nitrogen. Ciona are hermaphrodites, but eggs and sperm were collected from the terminal ends of gonoducts from different animals and mixed for fertilization. C. intestinalis embryos were collected at four stages (four-cell, gastrulation, tail release, and tail resorption) and frozen.
Isolation of mRNA and synthesis of cDNA
mRNA was isolated from tissues and embryos using a Micro Poly(A) Pure mRNA isolation kit (Ambion, Inc., Austin, TX). mRNA was reverse-transcribed in a 50-µl reaction that contained mRNA, 2 mM oligo dT, 2 mM deoxynucleoside triphosphates, 1x first strand reaction buffer, 0.01 M dithiothreitol, 5 U RNase inhibitor, and 100 U Superscript II reverse transcriptase (Invitrogen, San Diego, CA). The reaction was incubated at 42 C for 90 min, and the enzyme was heat-inactivated at 90 C for 10 min.
For rapid amplification of cDNA ends (RACE)-PCR, approximately 200 ng of mRNA was used to prepare RACE-ready cDNA using the RLM-RACE kit (Ambion, Inc.) according to the manufacturer’s instructions.
Isolation of genomic DNA
Genomic DNA was isolated using TRIzol reagent (Invitrogen). DNA was precipitated from the nonaqueous phase of the first phenol-chloroform phase separation following the manufacturer’s instructions, except that the DNA was redissolved in water.
PCR of cDNA and genomic DNA
Oligonucleotides were designed against regions encoding candidate GnRH prepropeptides based on the compiled genomic sequences for C. intestinalis GnRH genes 1 and 2 (Table 1
). Each 50-µl reaction contained 2.5 U Taq polymerase, 1x Taq buffer, 2.5 mM MgCl2, 0.2 mM deoxynucleoside triphosphates (Invitrogen), and 20 pmol of each 5' and 3' primer. PCRs were performed under the following conditions: denaturation at 94 C for 30 sec, annealing at 55 C for 30 sec, extension at 72 C for 30 sec for 35 cycles and a 7-min extension. The PCR products were separated by electrophoresis on a 1.3% agarose gel and visualized with ethidium bromide staining using an Eagle Eye II still video system (Stratagene, La Jolla, CA). Bands were selected, isolated (QIAGEN, Valencia, CA), and cloned or cloned directly as PCR products into pGEM Vector-T (Promega Corp., Madison, WI) and sequenced.
|
Preparative reverse phase-HPLC was used with two different solvent systems, triethylammonium phosphate (TEAP) 2.25 and trifluoroacetic acid (15, 16). A cartridge was packed in the laboratory with reversed-phase Vydac C18 silica (5 x 30 cm, 15- to 20-µm particle size, 300-Å pore size; Grace Vydac, Hesperia, CA) using a Waters Prep LC/System 500A (Millipore Corp., Milford, MA) for the purification of the peptides. The peptides eluted with a flow rate of 100 ml/min using a linear gradient of 1% eluent B per 3 min increase from the baseline % B [eluent A, 0.25 N TEAP (pH 2.25); eluent B, 60% acetonitrile, 40% A]. All peptides were subjected to a second purification step performed with eluents A (0.1% trifluoroacetic acid in water) and B (60% acetonitrile/40% A) on the same cartridge using a linear gradient of 1% B/min increase from the baseline % B. Analytical HPLC screening of the purification was performed on a Vydac C18 column (0.46 x 25 cm, 5-µm particle size, 300-Å pore size) connected to a Rheodyne (Western Analytical Products Inc., Murietta, CA) injector, two Beckman 100A pumps (Beckman Coulter, Fullerton, CA), model 420 System Controller Programer, Kratos 750 UV detector, and a Houston Instruments D-5000 strip chart recorder. The fractions containing the product were pooled and lyophilized. The purity of the final peptides was determined by analytical reverse phase-HPLC performed with a linear gradient using 0.1 M TEAP (pH 2.5) as eluent A and 60% acetonitrile/40% A as eluent B on a Hewlett-Packard Series II 1090 Liquid Chromatograph (Agilent Technologies, Wilmington, DE) connected to a Vydac C18 column (0.21 x 15 cm, 5-µm particle size, 300-Å pore size; Grace Vydac, Hesperia, CA), Controller Model 362. The purity of the peptides was also characterized by capillary zone electrophoresis (CZE) performed on a Beckman P/ACE System 2050 controlled by an IBM Personal System/2 Model 50Z (IBM, White Plains, NY) connected to a ChromJet integrator. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-MS) of the peptides was measured on an ABI-Perseptive DE-STR instrument (PE Applied Biosystems, Foster City, CA).
Screening tunicate peptides with available GnRH antisera
Sixteen different antisera that were raised in rabbits against different forms of GnRH were initially screened for binding to tunicate peptides. The antisera included: Jas-2 through -11 (anti-tGnRH-1); Bla-5, Her-4, Jul-5 (antilamprey GnRH-I); Emily and Gertie (antilamprey GnRH-III); 7CR-10 (antidogfish GnRH); GF-6, FP-5, PBL-45, PBL-49, and Aida (anti-sGnRH); 8CR-6 and -10, 9CR-6 (anticatfish GnRH); Adams-100 (anti-cGnRH-II); and B-7 (anti-mGnRH). Ten antisera were prepared in the Sherwood laboratory; the others included Emily and Gertie (a gift of Dr. Stacia Sower), PBL-45 and -49 (a gift of Dr. Wylie Vale), Aida (a gift of Dr. Katsumi Aida), and Adams-100 (a gift of Dr. Tom Adams). Each antiserum was prepared at a dilution of 1:1,000, except for B-7 (1:2,000), GF-6 and Aida (1:5,000), 7CR-10 (1:7,500), Gertie (1:8,000), Adams-100 (1:10,000), PBL-49 (1:30,000), and PBL-45 (1:50,000). The tGnRH-3, -5, -6, and -7 peptides were iodinated with I125. The percentage binding (maximum binding) of the four iodinated tGnRH peptides with each antiserum was determined, compared with the total counts.
Only six of the antisera (Jas-2 through -6, Bla-5, FP-5, 8CR-6, PBL-45, and PBL-49) had binding of greater than 5%; most antisera had binding of less than 1%. Two antisera were selected for further studies. Bla-5 had 30–32% binding with tGnRH-3 and -5; FP-5 had 7–13% binding with tGnRH-5 and -6. Jas-2 had 38% binding with the tGnRH-3 trace but did not show an immunocytochemical reaction, so it was not tested further. Each of the nine tunicate peptides, plus mGnRH and cGnRH-II were tested at 10, 100, 500, 1,000, 10,000, and 50,000 pg in four assays: with antibody Bla-5 and trace tGnRH-3 or tGnRH-5 or with antibody FP-5 and trace tGnRH-5 or tGnRH-6. The percentage cross-reactivity was calculated as the reference peptide in picomoles at 50% B/Bo divided by the test peptide in picomoles at 50% B/Bo times 100. Thus, the reference peptide has 100% cross-reactivity.
Immunolocalization
Specimens of C. intestinalis were relaxed in 0.01% MS222 for 30 min. Portions of the dorsal fold were pinned out in Petri dishes lined with Sylgard 184 (Dow Corning Corp., Midland, MI) using cactus spines. The dorsal blood sinus was opened to allow access of reagents to the dorsal strand and nerves lying within the sinus. The dorsal strand was fixed for 1 h in 4% paraformaldehyde in 0.1 M PBS (pH 7.3) at room temperature, followed by washing and storage in 0.1 M PBS containing 0.35% Triton X-100 and 0.03% sodium azide (PTA).
Preparations were treated with one of three primary antibodies, Jas-2, Bla-5, or FP-5, in 1:100, 1:1000, or 1:2000 (diluted with PTA), whereas controls were incubated in solutions omitting the antibody. All preparations had 1.5% goat serum added and were incubated for 12–24 h. After a PTA wash, preparations were incubated in fluorescein-isothiocyanate-goat antirabbit 
globulin (F-0382, Sigma, St. Louis, MO) for 12–24 h. After a PTA wash, preparations were mounted in 50% glycerol containing 1.5% N-propyl pyrogallate and examined by laser-scanning confocal microscopy using a Zeiss LSM 410 (Carl Zeiss, Toronto, Ontario, Canada).
Assay for gamete release induced by tGnRH peptides
Seven novel tGnRH peptides (tGnRH-3 through -9) and tGnRH-2 were tested to determine whether they induce gamete release in mature adults. Adult mature C. intestinalis were selected by the presence of a white sperm duct and/or a pink oviduct. GnRH peptides were dissolved in a saline solution (3 g NaCl/100 ml) and diluted at a final injection dose of 10 ng/g tunicate. The injection volume was based on the body mass of representative individuals (6.0–8.0 g). Peptides were injected with a 25-gauge needle beside the gonoducts. Then, each individual was placed in a 300-ml glass stacking dish filled with fresh, filtered seawater and monitored for response. A positive response was the release of a visible plume of either eggs or sperm. A preliminary test injection of only saline solution was conducted on 10 tunicates, none of which released gametes during a 30-min period.
Luciferase reporter gene assay for human GnRH-I receptor
Human embryonic kidney (HEK293) cells, genetically modified to express a cloned human GnRH-I receptor (Dr. Larry Jameson, Northwestern University, Chicago, IL), and a luciferase reporter gene under the control of LH -subunit promoter were cultured in DMEM containing 10% (vol/vol) fetal bovine serum, G418 (0.250 mg/ml), penicillin/streptomycin solution (100 U penicillin and 100 µg streptomycin/ml medium) and L-glutamine (2 mM). Cells were seeded at a density of 4 x 104 cells per well (90 µl volume per well) in phenol red free DMEM containing 10% (vol/vol) fetal bovine serum, penicillin/streptomycin solution (100 U penicillin and 100 µg streptomycin/ml medium), and L-glutamine (2 mM) in white 96-well culture plates (clear bottom with lid, tissue culture treated, polystyrene, sterile; catalog no. 3610, Corning, Inc., Corning, NY). The cells were incubated at 37 C under 5% CO2 for 24 h further before assay. Each compound was assayed in duplicate at 12 different concentrations (final concentration, 10-6 M to 10-12 M). Compounds diluted in 1% dimethyl sulfoxide (10 µl) were added to the HEK293 cells. At the same time, a GnRH standard curve was tested in duplicate ranging from 10-6 M to 10-12 M final concentrations. Plates were incubated for 4 h at 37 C under 5% CO2, after which 100 µl of LucLite was added to each well. The plates were sealed with Packard Topseal (Packard Bioscience Co., Meriden, CT) film and backing tape, gently mixed, and preincubated for 10 min at room temperature in the dark. Luminescence (counts per second) was measured in a Wallac Microbeta Trilux (Wallac, Inc., Gaithersburg, MD). An IC50 value was derived by nonlinear regression to a four-parameter logistic equation [sigmoidal dose-response (variable slope)], using the GraphPad Prism (version 3.0, GraphPad Software, Inc., San Diego, CA) curve fitting software package.
Rat pituitary cell assay
tGnRH peptides 3–9 were tested on dispersed rat pituitary cells for release of LH. Pituitary cells were prepared as described previously (17). Briefly, pituitary glands were removed from 150- to 200-g male rats after decapitation. The glands were quartered and rinsed several times with a HEPES buffer and stirred while digested with 3% BSA-supplemented HEPES buffer and 0.1% hyaluronidase and 0.35% collagenase for 45 min at 37 C. Cells were collected and centrifuged at 475 x g for 2 min, the supernatant was discarded, and the cell pellet was resuspended in 10 ml HEPES buffer containing 0.25% viokase (Life Technologies, Inc., Gaithersburg, MD) and stirred for an additional 15–30 min. The dispersed cells were collected and centrifuged a final time, and the pellet was resuspended in 3 ml sterile nutrient-supplemented DMEM, followed by six washings in medium. Then 1–5 x 105 dispersed cells were added to each dish with 3 ml DMEM and incubated at 37 C under a water-saturated atmosphere of 10% CO2, 90% air. Each tGnRH peptide 3–9 and mGnRH was diluted to a final concentration of 10-7 M. After 2.5 h, media was removed and tested for LH using a RIA specific for rat LH.
In silico analysis of promoter
Upstream promoter regions of 1000 nucleotides preceding the predicted transcription start site for each of the two GnRH genes in C. intestinalis and C. savignyi were generated with BLASTN similarity searches. To build the upstream promoter region for C. intestinalis, we used http://ghost.zool.kyoto-u.ac.jp/indexr1.html. Matching fragments were aligned using Bioedit Software (http://www.mbio.ncsu.edu/ BioEdit/bioedit.html). A technique to walk 1000 bp upstream from each predicted transcription start site by overlapping matching fragments was used to generate a consensus sequence with a minimum of four agreeing fragments covering any given region. The promoter sequences were confirmed using the Department of Energy JGI C. intestinalis v1.0 (http://genome.jgi-psf.org/ciona4/ciona4.home.html) on scaffolds 1051 (Ci-gnrh1) and 410 (Ci-gnrh2).
To predict the transcription factor binding sites on the various promoter regions, each 1000-bp upstream region was entered into the MatInspector input form using Matrix Family Library Version 2.4 (http://www.genomatix.de/cgi-bin/mat_fam.pl) with default settings (cutoff of 0.75 and offset of -1000). This bioinformatics tool recognized transcription factor binding site matrix information and resolved a greater number of GnRH-specific transcription binding sites than other methods such as International Union of Pure and Applied Chemistry consensus (TESS filtered) and context (Alibaba 2.1).
In silico and experimental analysis of nearest gene
To determine whether any tGnRH is an ortholog to a vertebrate GnRH, we identified the nearest upstream gene to GnRH. We searched the C. intestinalis databases using TBLASTN with the last exon in medaka FLJ20038 prepropolypeptide (medaka, Oryzias latipes; gi: 21955956). This gene has been identified as the closest upstream gene to mdGnRH in medaka and to mGnRH gene in human. This search resulted in a very high match (70% amino acid similarity) to a C. intestinalis FLJ20038 gene. A similar procedure of using overlapping fragments as described above was used to determine whether a gene similar to FLJ20038 was located upstream of the GnRH genes. We also used the Department of Energy JGI C. intestinalis v1.0 to identify the scaffolds containing the two Ciona GnRH genes as well as the Ciona FLJ20038 and PTP genes. Primers F1 and F2 (Table 1
) were designed for the region coding for FLJ20038 and used in a PCR with reverse primers G9 for Ci-gnrh1 and G10 for Ci-gnrh2 to amplify possible products. PCR was performed using genomic DNA (see below), High Fidelity platinum Taq DNA polymerase (Invitrogen), and the primers F1 or F2 and G9 or G10 under the following conditions: 94 C for 3 min, followed by 94 C for 30 sec, 55 C for 30 sec, and 68 C for 12 min for 30 cycles, followed by a 7-min extension at 68 C.
A similar procedure, both in silico and PCR using primers P1 or P2 and G9 or G10, was used to determine whether a PTP gene is upstream of the two Ciona GnRH genes. The final exon of the medaka PTP peptide (gi: 21955958) was found to have highly matching fragments within the C. intestinalis database. These matching nucleotide fragments were then used to construct an 1800-bp downstream region.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) as to whether they are present in C. productum only (tGnRH-1 and -2), in C. intestinalis only (tGnRH-3 and -4), in both Ciona species (tGnRH-5 to -8), or in C. savignyi only (tGnRH-9). Analysis of the genomic regions for these candidates showed that three of the peptides, tGnRH-3, tGnRH-5, and tGnRH-6, were found in close proximity as a triplet on one gene (Ci-gnrh1). These peptides are each bordered by basic amino acids (R or KR before the N terminus, and GRR or GKR after the C terminus) that are potential peptide cleavage sites separating the peptides. The GnRHs are separated by intervening peptides of 13 or 8 amino acids (Fig. 2A).
|
|
). The peptides in Ci-gnrh2 are each bordered by basic amino acids (RR or GKR) but lack intervening peptides.
PCR of mRNA transcripts from both GnRH genes in adult C. intestinalis
Using gene-specific primers, we amplified a single product from cDNA prepared from adult C. intestinalis brain tissue for each GnRH gene 1 and 2. Initial Ci-gnrh1 products were amplified with primers G1 and G5, resulting in a 327-bp product, and G1 and G6 resulted in a 343-bp product. 5'RACE-PCR resulted in a number of products for the 5' end of Ci-gnrh1, but only the longest 346-bp product made with G5 and the adapter inner primer matched the expected sequence. This product was used to compare to products in public expressed sequence tag (EST) and genomic databases. 3'RACE-PCR using G1 and the appropriate adapter primer resulted in a 708-bp cDNA fragment. This product was used to overlap and construct a full-length profile cDNA (Fig. 3) for comparison to public EST and genomic databases.
|
) for comparison with public EST and genomic databases. We were not able to isolate products for the 5' end of Ci-gnrh2 cDNA by RACE-PCR. However, we were able to amplify genomic DNA that codes for this region.
|
Triplet peptides on two genes in second tunicate species
Search results of the C. savignyi genome resulted in identification of fragments that group into similar gene structures as those in C. intestinalis. In C. savignyi, Cs-gnrh1 encodes two copies of tGnRH-5 and one copy of tGnRH-6 (Fig. 5A), whereas Cs-gnrh2 encodes tGnRH-7, tGnRH-8, and tGnRH-9 with the appropriate cleavage sites (Fig. 6A). Hence, the two Ciona species share the same general gene structures for gnrh1 and gnrh2 (Figs. 5B and 6B).
|
|
Early developmental expression of mRNA of both genes
Throughout development, both Ci-gnrh1 and Ci-gnrh2 are expressed by C. intestinalis. A single 327-bp product, amplified by PCR for Ci-gnrh1, is expressed at each stage: four-cell, gastrulation, tail release, and tail resorption (Fig. 7). This is the same transcript that was amplified from adult Ciona tissue. In contrast, two transcripts, 310 bp and 602 bp, were amplified for Ci-gnrh2 at the four-cell stage and gastrulation, although only the shorter transcript was detected at the tail release and tail resorption stages. Sequencing of these products revealed that one transcript has introns removed, whereas the second transcript retains an intron. Adult tissue in the same study expressed one transcript for gene 1, but only the longer transcript for gene 2.
|
). Two of nine tunicate peptides (tGnRH-3 and -5) showed the highest cross-reactivity with antiserum Bla-5, which was raised against lamprey GnRH-I (Table 3). Both peptides had between 77 and 117% cross-reactivity. The other tunicate peptides had less than 2.5% cross-reactivity; mGnRH had 3.5% or less cross-reactivity; and cGnRH-II did not cross-react (<0.1%). All of the peptides were tested with antiserum FP-5 and iodinated tGnRH-5 or tGnRH-6, but the cross-reactivity was so weak that only cGnRH-II and tGnRH-5 reached 50% B/Bo binding at a concentration below 50 nM.
|
|
), as previously reported for C. intestinalis using antisera raised against cGnRH-II (18), lamprey GnRH (19), and sGnRH (20). These antisera cross-react with several forms of GnRH. There was no labeling in the preparations incubated with the Jas-2 antibody or in controls.
|
). Administration of the peptide initially caused an increase in the water flow by bodily contraction. At least two tunicates from each group injected with GnRH peptide released gametes. The most effective peptides in this study were tGnRH-5, which caused 63% of tunicates to spawn, and tGnRH-3, in which 50% of tunicates spawned. The time for the tunicates to spawn ranged from 1 min 33 sec to 36 min. There was also a great range in the intensity and number of eggs or sperm released, although this was not associated with a particular peptide. Three animals injected with GnRH peptide underwent increased water flow and large bodily contractions, but they did not release eggs or sperm. This observation suggests that the GnRH may have been effective, but the gametes were not ripe for release at the time of our experiment.
|
-subunit promoter were exposed to peptides for 4 h, at which time luciferase activity was determined. In this assay, mGnRH activates the human GnRH-I receptor with an average EC50 of 0.34 nM. Of the tunicate peptides tested, six had no detectable activation of the human GnRH-I receptor in this assay, and three peptides activated the receptor slightly at the highest dose tested, 1 µM. At 1 µM, the peptides tGnRH-3, tGnRH-5, and tGnRH-7 activated the human GnRH-I receptor an average of 15%, 10%, and 9% of the maximal activation by mGnRH in this assay. The ability of the novel members of the GnRH family to release LH from rat anterior pituitary cells in culture was determined. Each of the tGnRH peptides 3–9 was inactive up to the maximal dose tested, 100 nM. mGnRH was maximally active at 1–2 nM concentration.
Promoter consensus sites for transcription factors
We identified a number of potential binding sites for transcription factors in the Ciona GnRH genes (Fig. 9); these DNA sequences are involved in the regulation of GnRH gene expression in promoter studies of other species. The relevant binding sites in the Ciona promoter regions are predicted to have binding affinity to POU factors (Brn-2, Oct-1, Pit-1, and Tst-1), GATA motif binding factor (GATA), androgen receptors, glucocorticoid receptors, progesterone receptors, cAMP response element (CRE) binding protein (CREB), and its variants, CREB-1 and CREB-1/c-Jun heterodimer, and activator protein (AP)-1. No obvious pattern was seen with respect to the location of the binding sites of the above transcription factors, but the frequencies of many transcription-binding sites are similar between the two Ciona species (Fig. 9).
|
Nearest upstream gene for GnRHs
DNA fragments were identified in C. intestinalis EST databases that had 73% amino acid identity to the last exon of human FLJ20038 and 75% identity to the last exon of human PTP. However, these two genes were not detected upstream of Ci-gnrh1 or Ci-gnrh2. To prove this, a region of approximately 4000 bp 3' to the FLJ20038 coding regions in C. intestinalis was compiled, but neither Ci-gnrh1 nor Ci-gnrh2 was within that distance. We also walked 4000 bp upstream of Ci-gnrh1 and 2500 bp upstream of Ci-gnrh2 in an attempt to locate FLJ20038, but none of these fragments matched the fragments used in constructing the FLJ20038 downstream region. Areas amplified by PCR did not contain the expected products. Also, in silico analysis showed that the two genes were neither within this downstream region of the PTP peptide nor matched with any fragments used to construct the upstream regions of either Ci-gnrh1 or Ci-gnrh2. No products were amplified by PCR using the PTP forward primer and the Ci-gnrh1 or Ci-gnrh2 reverse primer. Furthermore, using the Department of Energy JGI database, we identified our four genes of interest on different scaffolds: Ci-gnrh1 on scaffold 1051, Ci-gnrh2 on scaffold 410, the FLJ20038 gene on scaffold 91, and the PTP gene on scaffold 104. The distance between the genes was even greater than stated above.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Organization of GnRH mRNA is distinct in tunicates and vertebrates
The general gene structure of the Ciona genes is different from vertebrate GnRH genes. In vertebrates, the first exon is noncoding and contains the 5'-untranslated region (UTR); the second exon encodes the signal peptide, the GnRH peptide, and the first portion of the GnRH-associated peptide (GAP). The third exon encodes exclusively the bulk of GAP, and the fourth and final exon encodes the last few amino acids of GAP and the 3'-UTR.
In our study, the entire gene Ci-gnrh1 contains only one exon that encodes a 5'UTR, a putative signal peptide, then three GnRH peptides separated by intervening peptides of 13 and 8 amino acids. These are followed by a candidate GAP of 69 amino acids and finally a 3'UTR. In contrast, Ci-gnrh2 contains three exons, with a large first exon containing the 5'UTR as well as most of the coding region for the three GnRH peptides and their cut sites (RR or GKR) but no intervening peptides, as well as most of a candidate GAP. Exon 2 is 276 bp downstream and has no predictable function. The third exon is 141 bp further downstream and contains a stop codon followed by the 3'UTR.
Regulation of GnRH triplet peptides may enhance tunicate peptide output
In vertebrates, each GnRH peptide is coded on its own gene, allowing for separate regulation of GnRH production for each peptide. However, in Ciona species, three GnRH peptides are encoded on each of two genes, suggesting that the regulation of all three of the peptides coded on one gene is the result of common gene regulation. The multiplication of exons encoding GnRH may simply increase the output of peptides. The large number and type of promoter binding sites (identified in silico) that are shared with the human GNRH1 gene suggest that some aspects of regulation have been conserved.
Analysis of Ciona genome for other forms of GnRH
We have previously identified two novel forms of GnRH, tGnRH-1 and tGnRH-2, in the protochordate tunicate C. productum (12). We did not find any evidence for either of these two peptides in C. intestinalis. Thus, the peptides appear to be genus-specific.
Two GnRH peptides, cGnRH-I and mGnRH, were reported previously to be in the gonads of C. intestinalis using HPLC, RIA, and mass spectrometry (21). We have not found evidence in either the genome or EST databases or using molecular techniques for these two peptides in C. intestinalis.
tGnRH genes are expressed early in development
We found that the two GnRH genes in C. intestinalis are expressed as early as the four-cell stage in development. Both genes were expressed, but Ci-gnrh2 has one transcript that retains an intron. It is not clear whether this is a functional mRNA, a stored mRNA, or a transcript that was not completely processed at the time the tissue was collected for PCR analysis. However, we amplified the transcript with the intron retained in adult tunicate tissue as well, suggesting that this is a common phenomenon. Intron retention in the sGnRH cDNA occurs in adult rainbow trout (22, 23) and in mGnRH cDNA from human reproductive tissues (24). However, the impact of this, if any, on regulation is not clear. GnRH is expressed early in development in fish (25) and in the human placenta (26) and mouse embryo (27), suggesting that a function for GnRH in early development is conserved in tunicates and vertebrates.
Significance of two genes for GnRH in protochordates
The tunicates are invertebrate chordates belonging to the subphylum Urochordata or Tunicata. Tunicates represent an early branch of the phylum before the emergence of Cephalochordates (including Amphioxus) and vertebrates. Tunicates may represent a body plan that is based on a minimum set of mostly single-copy genes that are needed for chordate development. We do not think our finding of two genes for GnRH refutes the concept that an ancestral tunicate had mainly single-copy genes. Instead, we propose that there was one GnRH gene in ancestral tunicates and that a second GnRH gene resulted from duplication after expansion of the peptide-coding region. The two genes we have identified are different from each other in many regards, including the encoded peptides and presence or lack of introns. However, even with these differences, the similarity in number and type of promoter elements suggests conserved regulatory strategies between these two genes and between these genes and human GNRH1 or GNRH2.
tGnRH promoters have binding site sequences that are conserved in humans
A comparison of the tunicate and human GnRH gene promoters includes 1000 bp for each of the genes. Within this 1-kb region, the human GNRH1 gene has both a downstream transcription start site at +1 and an upstream start site at -579 (24). Also, the proximal 1 kb of promoter is the region that restricts expression of GnRH to cells mainly in the hypothalamus; the proof is that transgenic mice with a human GNRH1 construct require the promoter region from -992 to -763 bp to restrict the expression of a luciferase reporter gene (28). The rat Gnrhl promoter is not used for upstream comparison, because a neuron-specific enhancer is located further upstream at -1571 to -1863 (29) and the human and rat promoters have marked differences in the structural organization of the promoters, except in the proximal region at -343 to -1 (30).
Each of the tunicate or human promoter regions had an abundance of POU-family binding sites, except human GNRH2. In the human GNRH1 promoter, Brn-2 (a POU factor) binds to a site within the region essential for cell-specific expression (28). The tunicate promoters each contain one to three Brn-2 binding sites, but it is not known whether the sites are functional. Oct-1, another POU factor, binds within or near to the same cell-specific region, but did not stimulate human GNRH1 transcription of the reporter gene (28, 31). For the Tst-1 transcription factor, one to three binding sites were identified for each of the four tunicate promoters and one Tst-1 site for the human GNRH1 promoter. In the rat Gnrh-1 promoter, Tst-1 (referred to as SCIP/Oct-6/Tst-1) bound three sites within the proximal 333 nucleotides from the transcription start site and repressed GnRH transcription (32). These data may be relevant to human GNRH1 because the proximal promoter is similar to that of the rat.
There is controversy about steroid receptor binding to the GnRH promoter for human or rat. In the tGnRH genes, there are consensus sites for steroid receptors, but the C. intestinalis genome appears not to have genes that encode steroid receptors or genes for enzymes that synthesize steroids (4).
The Ciona genes Ci-gnrh1 and Cs-gnrh1 both contain a single CRE site, whereas none was detected in Ci-gnrh2 and Cs-gnrh2 promoters. CRE is interesting because this is one of only a few binding sites identified for the human GNRH2 promoter; the human GNRH1 promoter does not have sites for binding CREB, although there are CREB-1 and CREB-1/cJun sites. The CRE site in the human GNRH2 promoter appears to be functional as a cAMP analog up-regulated expression of GNRH2 mRNA and GnRH-II peptide (33). Mutation within the CRE site at -67 to -60 reduced both the basal activity and the cAMP analog response of the GNRH2 promoter.
Biological activity of GnRH in tunicates
GnRH has direct effects on gonads in some invertebrates. Five GnRH forms (mGnRH, cGnRH-I, cGnRH-II, sGnRH, and lamprey GnRH-I) increased mitogenic activity in gonial cells of oysters, Crassostrea gigas (34). Injection of tGnRH-1 and -2 induced spawning in C. intestinalis generally within minutes, although tGnRH-2 was less effective than tGnRH-1 (35). Each of the tGnRH peptides we tested induced gamete release, although some peptides appeared to be more effective than others, because the percentage effectiveness ranged from 20–63%. Each gene produces peptides that can induce the release of eggs and/or sperm. It is possible that the time of year of our experiment did not coincide with all of the animals being fully mature. Animals were selected on the basis of visual assessment of maturity, i.e. a white sperm duct and/or a pink oviduct. GnRH may act in concert with other biological or environmental factors to induce spawning but may not be sufficient on its own. Only one of 19 saline-injected C. intestinalis released eggs. The bioactivity of the peptides, including tGnRH-2 found in C. productum and tGnRH-9 that we have identified in the C. savignyi genome, suggests that the receptor(s) in C. intestinalis are able to bind many forms of tGnRH. Likewise, tGnRH-2 induced spawning in a mollusc (36). The GnRH receptor may initiate a direct or indirect effect on the gonoducts, but in either case, the isolation and localization of GnRH receptors are needed to demonstrate the target organs of GnRH in tunicates.
tGnRH peptides are inactive at rat and human GnRH-I receptors
The novel tGnRH-3, -4, -5, -6, -7, -8, and -9 are inactive at the rat and human GnRH-I receptors. Although we have no evidence that the Gln N terminus is processed to the expected pyroglutaminyl residue, it is highly probable. Therefore, all but tGnRH-2 and -7 (with 4 and 2 substitutions, respectively) have three substitutions when compared with mGnRH at positions 5–8. From known studies on structure-activity relationships in mammalian systems, an L-amino acid substitution for Gly at position 6 is conformationally unfavorable. Of the nine tGnRHs, tGnRH-6 was the most likely to have affinity for the mammalian GnRH-I receptor because of an achiral Gly residue is present in position 6. Whether it is the lack of a basic residue at position 8 (Ser) or the presence of a basic residue at position 5 (Lys) that determines the loss of potency at the human GnRH-I receptor is uncertain. The fact that chiral amino acids are found at position 6 of all but one tGnRH suggests that epimerization could occur posttranslationally. Although this could be tested synthetically, the fact that all tGnRHs are active to a different extent in the tunicate suggests otherwise. Because cGnRH-II is biologically active in mammalian systems although it has substitutions at positions 5 (His for Tyr), 7 (Trp for Leu), and 8 (Tyr for Arg) compared with mGnRH, it cannot be excluded that tGnRH analogs with a D-residue at position 6 would have some binding affinity for the mammalian GnRH-I receptor.
In conclusion, tunicates of the genus Ciona are favorable models for the study of the function and regulation of genes important in development (37). The same characteristics make Ciona excellent for hormone studies. Our in silico identification of genes encoding the hormone GnRH was confirmed by our sequencing of genomic DNA. We have shown that both of the GnRH genes are expressed as mRNA early in development and in adult Ciona tissue. The approach of identifying GnRH orthologs in protochordates by the nearest upstream gene does not appear to be suitable because the marker genes that are present in medaka and human are not upstream of C. intestinalis GnRH genes, although the genes are present in tunicate. The novel peptides do not represent potential analogs for rat or human GnRH studies because the peptides do not activate the human GnRH-I receptor or cause LH release in rat pituitary cell cultures. This is most likely due to the presence of an L-amino acid (in place of glycine) at position 6 in the tunicate peptides. It is accepted from structure-activity relationship studies in mammalian systems that nonglycine residues (except possibly for proline) are detrimental to biological activity (in vitro and in vivo; Ref.38). However, this explanation of structure activity relationships from the structural perspective cannot ignore the fact that each residue at positions 5–8 (Tyr-Gly-Leu-Arg) in mGnRH has been selected for optimal interaction with the mammalian receptor type 1 and that any deviation from that sequence results in significant loss of potency.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: AP, Activator protein; cGnRH-II, chicken GnRH-II or GnRH-II; CRE, cAMP response element; CREB, CRE binding protein; CZE, capillary zone electrophoresis; EST, expressed sequence tag; GAP, GnRH-associated peptide; mdGnRH, medaka GnRH; mGnRH, mammalian GnRH or GnRH-I; PTA, PBS containing Triton X-100 and sodium azide; PTP, protein tyrosine phosphatase ; RACE, rapid amplification of cDNA end; sGnRH, salmon GnRH; TEAP, triethylammonium phosphate; TFA, trifluoroacetic acid; tGnRH, tunicate GnRH; UTR, untranslated region.
Received December 31, 2002.
Accepted for publication February 3, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Wu, G. J Roch, L. A Cervini, J. E Rivier, and N. M Sherwood Newly-identified receptors for peptide histidine-isoleucine and GHRH-like peptide in zebrafish help to elucidate the mammalian secretin superfamily J. Mol. Endocrinol., November 1, 2008; 41(5): 343 - 366. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 |
, Androgen response element; 
, AP-1 response element; 
, Brn-2 binding site; 
, CRE for CREB; 
, CRE for CREB-1;
, CRE for CREB-1/cJun heterodimer;
, GATA response element;
, GRE, glucocorticoid response element;
, Oct-1 binding site; 
, Pit-1 binding site; PRE, progesterone response element;
, Tst-1 binding site.