BIOLOGY OF REPRODUCTION 62, 1329–1334 (2000)
Differential Splicing of Three Gonadotropin-Releasing Hormone Transcripts in the
Ovary of Seabream (Sparus aurata)1
Massimo Nabissi,3 Laura Soverchia,3 Alberta M. Polzonetti-Magni,2,3 and Hamid R. Habibi4
Dipartimento di Scienze Morfologiche e Biochimiche Comparate,3 University of Camerino, 62032 Camerino (MC), Italy
Department of Biological Sciences,4 University of Calgary, Calgary, Alberta T2N 1N4, Canada
ABSTRACT
Previous studies demonstrated the presence of high-affinity
GnRH binding sites and compounds with GnRH-like activity in
the ovary of seabream, Sparus aurata, providing evidence for the
role of GnRH as a paracrine/autocrine regulator of ovarian func-
tion in this species. In the present study, the expression of three
forms of GnRH (salmon, chicken-II, and seabream) genes in this
marine teleost species was demonstrated for the first time.
Moreover, there is evidence for differential splicing and intronic
expression of cGnRH-II and sbGnRH. Treatment of seabream
follicle-enclosed oocytes with salmon GnRH stimulated reinitia-
tion of oocyte meiosis, whereas chicken GnRH-II treatment was
without effect. Novel information was also provided about or-
ganization of cGnRH-II and seabream GnRH transcripts, con-
firming that GnRH gene organization is maintained through evo-
lution, despite changes in the size and sequence of exons and
introns.
INTRODUCTION
GnRH plays a pivotal role in reproduction by regulating
pituitary gonadotropin synthesis and release, which in turn
modulate gonadal activity [1, 2], including that of seabream
[3]. To date, the primary structures of eleven GnRH vari-
ants have been elucidated, and two or more molecular
forms have been found to be expressed in species from all
vertebrate classes [1–4], including placental mammals [5,
6]. There has also been parallel evolution of the GnRH
receptors, resulting in diverse regulatory functions of
GnRH peptides in the brain, pituitary, and other peripheral
tissues. In particular, GnRH molecules have been shown to
function as neurotransmitters, as well as a local signal in
the gonads [7]. There is also evidence for the presence of
GnRH in the ovary of a number of mammalian and non-
mammalian species [8, 9] including seabream [10]. The
cDNA sequences encoding the precursor of mammalian
GnRH (mGnRH), salmon GnRH (sGnRH), chicken GnRH-
I, chicken GnRH-II (cGnRH-II), catfish GnRH, and sea-
bream GnRH (sbGnRH) have been isolated and character-
ized from different species [4, 11–13], and GnRH gene ex-
pression has been demonstrated in the ovary of a number
of mammalian species [14–18] and other vertebrates in-
cluding fish [19, 20].
Primary structure of GnRH is largely conserved, and
1
This study was supported by a grant to Prof. Alberta M. Polzonetti-Magni
from the MURST (PRIN) Italy.
2
Correspondence: Alberta Maria Polzonetti-Magni, Dipartimento di Scien-
ze Morfologiche e Biochimiche Comparate, University of Camerino, Via
Camerini no 3, 62032 Camerino (MC), Italy. FAX: 39 0737 403217;
e-mail: alberta@camserv.unicam.it
Received: 10 March 1999.
First decision: 6 April 1999.
Accepted: 20 December 1999.
Q 2000 by the Society for the Study of Reproduction, Inc.
ISSN: 0006-3363. http://www.biolreprod.org
1329
GnRH precursors in all species so far studied show the
same general organization, containing regions for the signal
peptide, GnRH, amide-donating glycine, processing site,
and GnRH-associated peptide (GAP). Whereas the segment
coding for the mature GnRH has a high degree of similarity
among different species, there is less than 50% similarity
in the signal peptide and GAP regions. Previous studies in
seabream demonstrated the presence of high-affinity
GnRH-binding sites and compounds with GnRH-like activ-
ity in the ovary, providing evidence for the role of GnRH
as a paracrine regulator of ovarian function in this species
[10]. The present study extends this information and dem-
onstrates the expression of GnRH genes and action of
GnRH peptides on reinitiation of oocyte meiosis in the sea-
bream ovary.
MATERIALS AND METHODS
Animals
Adult female seabream (Sparus aurata, 0.7–1.5 kg BW)
were purchased from a commercial fish farm (La Rosa, Or-
betello, west coast of Italy; lat 428289N, long 118129E). Fish
were maintained in aquaria at 22–238C under natural pho-
toperiod (10L:14D) and 35 parts per thousand salinity, and
fed with commercial fish food. All fish were netted, anes-
thetized in MS-222 (100 mg/L; Sigma, Milan, Italy), and
decapitated before removal of ovaries. Ovaries were im-
mediately frozen in liquid nitrogen and stored at 2708C
until use in molecular biology studies.
Hormones
sGnRH-A (D-Arg6,Trp7,Leu8,Pro9]-N[Et]-GnRH), SGnRH
([Trp 7 ,Leu 8 ]-GnRH), and cGnRH-II ([His 5 ,Trp 7 ,Tyr 8 ]-
GnRH) were purchased from Peninsula Laboratories Ltd.,
Belmont, NY. Peptides were solubilized in 0.1 M acetic
acid (10 mg/20 ml), stored at 2208C, and used within 4 wk.
Appropriate concentrations of the peptide were prepared by
diluting the stock solutions immediately before use in an
experiment.
Follicle Incubation
Ovaries obtained from five different animals in the re-
crudescence period were dissected out and washed in min-
imum essential medium (MEM; Gibco/BRL, Gaithersburg,
MD) in a sterile Petri dish. Pooled follicles were manually
separated under a stereomicroscope, and only 0.6- to 0.7-
mm follicles were used for in vitro incubation. Follicles
were incubated in MEM (with 0.05% streptomycin) in ster-
ile multiwell plates; each well contained 20 follicle-en-
closed oocytes in 1 ml of MEM plus the appropriate com-
pounds. Incubations were performed for 24, 36, 48, 60, 72,
84, and 96 h at 188C in the dark. After each period, the
incubation was stopped by addition of acetic acid (50%
final concentration) to each well, and the incubate was al-
1330 NABISSI ET AL.
FIG. 1. Effect of sGnRH and cGnRH-II on reinitiation of meiosis deter-
mined by means of GVBD in cultured follicle-enclosed seabream oocytes.
GVBD values were determined after 96 h of incubation, and nonstatisti-
cally significant incubation periods are not shown. Each value represents
a percentage of GVBD determined using 60 oocytes (20 oocytes per well).
Values are means 6 SE. *Significantly different from control (P , 0.05).
sGnRH-A was used as positive control, MEM (culture medium).
lowed to stand for 10 min at 188C to clear the oocyte cy-
toplasm. The cleared oocytes were observed under the ste-
reomicroscope, and absence of the germinal vesicle (oocyte
nucleus) indicated breakdown of the nuclear envelope (ger-
minal vesicle breakdown [GVBD]) and reinitiation of mei-
osis, as described previously for goldfish [21].
Total RNA Extraction
Total RNA was extracted, for each reproductive period,
from 5 g of an ovarian pool from five fish, using Trizol
RNA isolation reagent (Gibco/BRL) based on the acid
guanidinium thiocyanate-phenol-chloroform extraction
method [22]. Final RNA concentration was determined by
optical density (OD) reading at 260 l, and integrity was
verified by ethidium bromide staining of 28S and 18S RNA
bands on a denaturing agarose gel.
Oligonucleotides
GnRH primers were designed on the basis of the se-
quence of the salmon, chicken-II, and seabream GnRH
cDNAs of seabream (U30311; U30325; U30320) [12], and
were purchased from Gibco/BRL. The 19-mer sGnRH
sense primer (59 ATGGAGGCGAGCAGCAGAG 39) cor-
responded to nucleotide regions 233–251, and the 20-mer
sGnRH antisense primer was complementary to nucleotide
region 477–496 of sGnRH cDNA. The 19-mer cGnRH-II
sense primer (59 ATGTGTGTATCTCGGCTGG 39) corre-
sponded to nucleotide region 118–137, and the 19-mer
cGnRH-II antisense primer (59 CTTCCTCTTCTGGAGC-
TCT 39) was complementary to nucleotide region 355–373
of cGnRH-II cDNA. The 19-mer sbGnRH sense primer (59
CTCAAACCTCTGGATCCTG 39) corresponded to nucle-
otide region 52–70, and the 19-mer sbGnRH antisense
FIG. 2. PCR results from analyses in different reproductive periods. NR,
Nonreproductive; R, reproductive; PS, postspawning; T, PCR from total
RNA; N, PCR without template. s, c-II, sb: RT-PCR products using sGnRH,
cGnRH-II, sbGnRH sets of primers, respectively. M, Amplisize agarose
markers (Bio-Rad, Milan, Italy). The data were considered useful only if
no bands were observed in the negative controls.
primer (59 TTCTGTGTCCGATGTCCCG 39) was comple-
mentary to nucleotide region 303–311 of sbGnRH cDNA.
Reverse Transcription (RT)-Polymerase Chain Reaction
(PCR)
Ten micrograms of total RNA from seabream ovary were
reverse-transcribed into cDNA in a total volume of 25 ml
containing 1 mg of oligo(dT) 12–18 primers (Gibco/BRL)
and 100 IU of Moloney murine leukemia virus (MMLV)
reverse transcriptase (Gibco/BRL). The reaction was car-
ried out for 15 min at 258C, and then for 90 min at 378C.
After deoxyribonuclease (DNase) treatment of the RT mix-
ture, an aliquot (1 ml) of the resulting cDNA products was
subsequently amplified with 2.5 IU Taq DNA polymerase
(Gibco/BRL) in 50 ml of master mix containing single-
strength PCR buffer (20 mM Tris-HCl pH 8.4, 50 mM
KCl), 1.5 mM MgCl2, 0.5 mM dNTP, and GnRH primers
(50 pmol each). The PCR amplification, for all sets of prim-
ers, was carried out for 35 cycles in an automated ther-
mocycler (MJ Research, Watertown, MA) with thermocycle
profile (denaturation at 948C for 40 sec, primer annealing
at 588C for 40 sec, primer extension at 728C for 1 min)
followed by a post-PCR incubation at 728C for 7 min. PCR
products were then extracted with phenol-chloroform-iso-
amyl alcohol, and an aliquot of each sample was electro-
phoresed on 2% agarose gel in 0.5-strength TAE buffer
(Tris-HCl 20 mM pH 7.5, acetic acid 20 mM, EDTA 0.5
mM, pH 8.0). Different negative controls were prepared.
First, the same amount of total RNA as that used for the
RT-PCR assay was added to the RT reaction mixture with-
out reverse transcriptase and was subsequently amplified,
to confirm the absence of genomic contamination. Second,
all the components of the RT reaction were prepared with-
out RNA and subsequently amplified to confirm the ab-
sence of contamination in the reagents used. The data were
considered useful only if no bands were observed in the
negative controls.
Cloning and Sequencing
Twenty nanograms of PCR products from sGnRH and
28 ng of PCR products from cGnRH-II and sbGnRH primer
amplifications were cloned using pGEM-T vector system
(Promega, Madison, WI). Ten positive clones for each sam-
 OVARIAN GnRH GENE EXPRESSION
ple were screened to confirm the presence of an insert and
subsequently sequenced. DNA sequence analysis was per-
formed by dye-labeled terminators using a DNA sequenc-
ing kit (Perkin Elmer, Irvine, CA) and M13 universal prim-
ers that flank the inserted DNA. The nucleotide sequences
were read in both directions.
Statistical Analysis
Follicle incubation results were analyzed by one-way
ANOVA with Stat View 5121 (Brain Power, Calabasas,
CA). P , 0.05 was taken to indicate a statistically signif-
icant difference between means. Results are expressed as
means 6 SE.
1331
FIG. 3. A) The cDNA nucleotide se-
quence encoding the partial cGnRH-II pre-
cursor and the deduced amino acid se-
quence. The 59 and 39 intron borders are
in boldface; the 100% identity regions
with cGnRH-II cDNA from seabream pitu-
itary are underlined. *Stop codon. The
length of each region is numbered in the
schematic representation. B) The cDNA
nucleotide sequence encoding the partial
sbGnRH precursor and the deduced ami-
no acid sequence. The 59 and 39 intron
borders are in boldface; the 98% identity
regions with sbGnRH cDNA from sea-
bream pituitary are underlined. *Stop co-
don. The length of each region is num-
bered in the schematic representation.
RESULTS
In Vitro Experiments on Oocyte Maturation
Previous studies provided characterization of three forms
of GnRH (sGnRH, cGnRH-II, and sbGnRH) in the sea-
bream brain [23], as well as evidence for the presence of
high-affinity GnRH binding sites and compounds with
GnRH-like activity in the seabream ovary [10]. We used
isolated seabream follicle-enclosed oocytes, in vitro, to in-
vestigate direct action of sGnRH and cGnRH-II on the rein-
itiation of oocyte meiosis as indicated by GVBD. Admin-
istration of sGnRH at 1026 M concentration significantly
increased the GVBD response, while GVBD responses at
lower (1027 M, 1028 M) concentrations of sGnRH were not
1332 NABISSI ET AL.

significantly different from that of MEM (Fig. 1). Treatment

with cGnRH-II did not affect GVBD response in the sea-
bream ovary. A significantly high GVBD response was ob-
served after treatment with 1027 M of sGnRH-A (used as
positive control) compared to sGnRH (Fig. 1).

GnRH Gene Expression and Characterization

Three sets of primers were designed on the basis of
sGnRH, cGnRH-II, and sbGnRH cDNA sequences char-
acterized previously in the seabream brain [12, 13]. The
primers amplified regions between the signal peptide and
GAP for all three GnRH forms. For each set of primers,
the predicted PCR products were 264 base pairs (bp) for
sGnRH, 255 bp for cGnRH-II, and 259 bp for sbGnRH.
Using sGnRH primers, amplification of cDNA from sea-
bream ovarian total RNA at the recrudescence phase re-
sulted in a PCR product of 264 bp, while the use of c-
GnRH-II and sbGnRH primer sets resulted in larger than
expected products of 549 bp and of 551 bp, respectively
(Fig. 2). Two independently amplified 264-bp fragments
were cloned, and nucleotide sequences were found to have
100% identity with the appropriate region of sGnRH cDNA
characterized from seabream brain. Cloning of the 549-bp
fragments (from cGnRH-II primer) and nucleotide sequence
analysis of two different clones revealed in each case a
number of matching regions with 100% similarity to c-
GnRH-II cDNA characterized in seabream brain. The re-
gions of similarity in the 549-bp fragment were 1–135,
272–355, and 514–549, corresponding to cGnRH-II cDNA
regions of 119–253, 254–337, and 338–373, respectively
(Fig. 3A). The genetic information contained in the non-
matching regions of the 549-bp PCR product corresponded
to an intervening sequence (introns) in-frame with exons,
due to the presence of invariant GT and AG dinucleotides
found at the 59 and 39 intron borders, which are part of the
largest consensus sequence characteristic of nuclear pre-
mRNA introns. These regions had higher A1T contents
than their neighboring exons, and the sequence upstream of
39 ends was preceded by a stretch of pyrimidine and T-run,
T, and A homopolymers. The branch point sequences were
also determined, and the presence of inverted repeats that
could interfere with splicing was revealed (Fig. 4). Cloning
and sequencing of the 551-bp fragments (from sbGnRH
primer) also revealed a number of matching regions with
98% similarity to sbGnRH cDNA characterized in the sea-
bream brain. The regions of similarity in the 551-bp frag-
ment were 1–133, 314–412, and 525–551, corresponding
to sbGnRH cDNA regions 52–184, 185–283, and 284–311,
respectively (Fig. 3B). Similarly, the nonmatching regions
of the 551-bp PCR product contained genetic information
corresponding to the intervening sequence (introns) be-
cause of the presence of invariant GT and AG dinucleotides
found at the 59 and 39 intron borders (Fig. 4).
The predicted amino acid sequence derived from the
ovarian cGnRH-II and sbGnRH partial transcript nucleotide
sequence was compared to the amino acid sequence of
brain cGnRH-II and sbGnRH [12, 13]. In the case of c-
GnRH-II, the presence of a stop codon (TAA) in nucleotide
position 286–288 may give rise to the translation of a short-
er GAP peptide than the pre-cGnRH-II peptide produced in
the seabream brain. Similarly, a stop codon (TAA) in nu-
cleotide position 212–214 of ovarian sbGnRH may result
in a shorter GAP peptide compared to the pre-sbGnRH pep-
tide produced in the seabream brain. From the sequence
data, we can also indicate, in terms of exon/intron, the or-
FIG. 4. Analysis of intron regions of cGnRH-II and sbGnRH partial tran-
script sequence. Top) Sequence and nucleotide position of intron junc-
tions. Invariant GT dinucleotides (at 59 intron border) and terminal AG
dinucleotide (at 39 end of the pre-mRNA introns) are in boldface. The
characteristic sequence of introns is represented as nucleotide positions
in the cGnRH-II and sbGnRH partial transcript sequences. Bottom) Pyrim-
idine stretch (pyrimidine stretch that preceded 39 end of introns); T/A hom-
op. (polymers composed only of A or T nucleotides); Inv.repeat (inverted
repeat sequences between branch points and 39 splice sites); % ex/int TA
(percentage of T and A nucleotides in exons and introns).
ganization of the cGnRH-II and sbGnRH genes in seabream
(Fig. 3). Analysis of size distribution of exons and introns
suggests that the cGnRH-II and sbGnRH genes in seabream
ovary maintain the same exon/intron distribution as in the
other GnRH genes characterized (in human, rat, and salm-
on). In particular, from the sequences of cGnRH-II and
sbGnRH partial transcripts, we may predict the gene or-
ganization in those regions. As shown in Figure 3, the cod-
ing region for signal peptide, GnRH, and processing sites
is separated from the coding regions for GAP peptide by
two introns, both in cGnRH-II and in sbGnRH partial tran-
scripts.
To investigate possible seasonal variation in splicing of
these three GnRH genes in seabream, we performed RT-
PCR from total RNA obtained during the nonreproductive
and postspawning periods. The results indicated no season-
al variations in the processing of these pre-mRNA in the
seabream ovary (Fig. 2). In these studies, we were not able
to demonstrate ovarian expression of GnRH by Northern
hybridization, presumably because of low levels of expres-
sion at the time of the experiments.
DISCUSSION
The expression of the GnRH gene was previously re-
ported in human and rat ovary [11, 14, 15]. The study in
the rat suggests that a proportion of the ovarian GnRH tran-
scripts contain intronic sequences and utilize a different
transcription start, compared to hypothalamic GnRH [14].
In the light of previous studies demonstrating expression of
 OVARIAN GnRH GENE EXPRESSION
GnRH genes in the ovary of various vertebrates [14–20],
and the existence of sGnRH, cGnRH-II, and sbGnRH in
the seabream brain [23], we investigated the expression of
these three different GnRH forms in the seabream ovary by
RT-PCR, demonstrating that cGnRH-II and sbGnRH tran-
scripts undergo differential splicing compared to the same
GnRH mRNA forms expressed in the seabream brain. Most
nuclear messenger RNA precursors (pre-mRNA or hetero-
nuclear (hn) RNA) in higher eukaryotes contain multiple
introns, which must be precisely excised by RNA splicing
[24]. Alternative RNA splicing is the process that allows
the selection of different combinations of splice sites within
an hnRNA, and is part of the expression program of a large
number of genes implicated in cell growth and differenti-
ation [24, 25]. Patterns of alternative splicing can be very
complex and can involve alternative introns and exons as
well as variations in the position of individual splice junc-
tions [24]. Regulated alternative splicing can lead to the
production of different proteins from a single hnRNA or
can function as an on/off switch during development [24,
25]. In this context, the presence of sGnRH, cGnRH-II, and
sbGnRH transcripts has been demonstrated previously in
the seabream brain [10], and our findings demonstrate that
cGnRH-II and sbGnRH are alternatively spliced in different
tissues, presumably because of different splicing mecha-
nisms. It has also been demonstrated that the insertion of
inverted repeats between the branch point and 39 splice sites
of yeast and mammals interferes with splicing [26], which
might also be responsible for differential splicing of c-
GnRH-II and sbGnRH in seabream ovary. The differential
splicing observed in seabream ovary was not influenced by
reproductive stages, suggesting a major role of the intron/
exon element in controlling the differential splicing rather
than a variation in the abundance of a specific factor reg-
ulating alternative splicing [27]. It should be noted, how-
ever, that the present results are based on studies carried
out during a rather narrow period of seabream life history,
and we cannot rule out the possibility of seasonal variations
encompassing different stages of sexual development in this
species. The involvement of macro (pituitary gonadotro-
pins) and micro (ovarian peptides) regulatory systems is
particularly important in species such as seabream, which
contain asynchronous ovary continuously undergoing
change and development. While the relationship between
GAP and GnRH may not be clear, there is evidence that
GAP and GnRH may have related functions [28] and that
other cryptic peptides associated with neurohormones may
be involved in the correct processing and packaging of the
hormone [29]. In this respect, the presence of a stop codon
in the introns of the cGnRH-II and sbGnRH partial tran-
scripts may be related to the biological activity of GnRH
molecular forms. Study of paracrine ovarian function in
seabream is very interesting, since this species starts de-
velopment as a hermaphrodite and undergoes significant
changes, from being functional males in puberty to becom-
ing functional females at later stages.
The present results also provide information about or-
ganization of the cGnRH-II and sbGnRH genes. In this con-
text, the exon/intron organization of cGnRH-II and sb-
GnRH partial transcripts was found to be similar to that of
the other GnRH genes characterized previously, indicating
that GnRH gene organization, in terms of exon/intron lo-
cation, is maintained through evolution, in spite of changes
in the size and sequence of exons and introns [30].
In conclusion, the present findings strongly support the
hypothesis that sGnRH, and possibly other forms of GnRH
1333
variants, are involved in paracrine/autocrine regulation of
seabream ovarian function.
REFERENCES
1. Peter RE, Yu KL. Neuroendocrine regulation of ovulation in fishes:
basic and applied aspects. Rev Fish Biol Fish 1997; 7:173–197.
2. Nagahama Y, Yoshikuni M, Yamashita M, Tokumoto T, Katsu Y. Reg-
ulation of oocyte growth and maturation in fish. Curr Top Dev Biol
1995; 30:103–145.
3. Zohar Y, Elizur A, Sherwood NM, Powell JFF, Rivier JE, Zmora
N.Gonadotropin-releasing activities of three native forms of gonado-
tropin-releasing hormones present in the brain of gilthead seabream,
Sparus aurata. Gen Comp Endocrinol 1995; 97:289–299.
4. Millar RP, Troskie B, Sun YM, Ott T, Wakefield I, Myburch D, Paw-
son A, Davidson JS, Flanagan C, Katz A, Hapgood J, Sealfon SC,
Peter RE, Terasawa E, King JA. Plasticity in the structural and func-
tional evolution of GnRH: a peptide for all seasons. In: Kawashima
S, Kikuyama S (eds.), Proceedings of XIII Congress of Comparative
Endocrinology. Yokohama, Japan: Monduzzi Editore, Bologna (Italy);
1997: 15–28.
5. Kasten TL, White SA, Norton TT, Bond CT, Adelman JP, Fernald RD.
Characterization of two new preproGnRH mRNAs in the tree shrew:
first direct evidence for mesencephalic GnRH gene expression in a
placental mammal. Gen Comp Endocrinol 1996; 104:7–19.
6. Lescheid DW, Terasawa E, Quanbeck C, Urbanski HF, Warby CM,
Sherwood NM. A second form of gonadotropin-releasing hormone
(GnRH) in primate brain. In: Program of the 10th International Con-
gress of Endocrinology; 1996; San Francisco, CA. OR63 (Abstract).
7. Habibi HR, Pati D. Extrapituitary gonadotropin-releasing hormone re-
ceptors in teleosts. Fish Physiol Biochem 1993; 11:43–49.
8. Pati D, Habibi HR. Presence of salmon GnRH and compounds with
GnRH like activity in the ovary of goldfish. Endocrinology 1998; 139:
2015–2024.
9. Habibi HR, Pati D. Endocrine and paracrine control of ovarian func-
tion: role of compounds with GnRH-like activity in goldfish. In: Fac-
chinetti F, Henderson IW, Pierantoni R, Polzonetti-Magni AM (eds.),
Cellular Communication in Reproduction. Bristol, UK: J Endocrinol;
1993: 59–70.
10. Nabissi M, Pati D, Polzonett-Magni AM, Habibi HR. Presence and
activity of compounds with GnRH-like activity in the ovary of sea-
bream, Sparus aurata. Am J Physiol 1997; 272:R111-R117.
11. Sherwood NM, Parker DB, McRory JE, Lescheid DW. Molecular
evolution of GnRH and GnRH. In: Sherwood NM, Hew CL (eds.),
Molecular Endocrinology of Fish; Farrel AP, Randall DJ (eds.), Fish
Physiology. New York: Academic Press; 1994: 13:29–66.
12. Gothilf Y, Chrow MM, Elizur A, Chen TT, Zohar Y. Molecular clon-
ing and characterization of a novel gonadotropin-releasing hormone
from the gilthead seabream (Sparus aurata). Mol Marine Biol Biotech
1995; 4:27–35.
13. Gothilf Y, Munoz-Cueto JA, Sagrillo CA, Selmanoff M, Chen TT,
Kah O, Elizur A, Zohar Y. Three forms of gonadotropin-releasing
hormone in a perciform fish (Sparus aurata): complementary deoxy-
ribonucleic acid characterization and brain localization. Biol Reprod
1996; 55:636–645.
14. Goubau S, Bond CT, Adelman JP, Misra V, Hynes MF, Murphy BD.
Partial characterization of the gonadotropin-releasing hormone
(GnRH) gene transcript in the rat ovary. Endocrinology 1992; 130:
3098–4000.
15. Oikawa M, Dargan C, Ny T, Hsueh AJW. Expression of the gonad-
otropin-releasing hormone and prothymosin—a messenger ribonucleic
acid in the ovary. Endocrinology 1990; 127:2350–2356.
16. Clayton RN, Eccleston L, Gossard F, Thalbard JC, Morel G. Rat gran-
ulosa cells express the gonadotrophin-releasing hormone gene: evi-
dence from in-situ hybridization histochemistry. J Mol Endocrinol
1992; 9:189–195.
17. Peng C, Fan NC, Ligier M, Vaananen J, Leung PCK. Expression and
regulation of gonadotropin-releasing hormone (GnRH) and GnRH re-
ceptor messenger ribonucleic acid in human granulosa-luteal cells.
Endocrinology 1994; 135:1740–1746.
18. Dong KW, Yu KL, Roberts JL. Identification of a major upstream
transcription start site for the human progonadotropin-releasing hor-
mone gene used in reproductive tissue and cell-lines. Mol Endocrinol
1993; 7:1654–1666.
19. Lin XW, Peter RE. Expression of salmon gonadotropin-releasing hor-
mone (GnRH) and chicken GnRH-II precursor messenger ribonucleic
1334 NABISSI ET AL.
acids in the brain and ovary of goldfish. Gen Comp Endocrinol 1996;
101:282–296.
20. Grober MS, Myers TR, Marchaterre MA, Bass AH, Myers DA. Struc-
ture, localization and molecular phylogeny of a GnRH cDNA from a
paracanthoterygian fish, the plainfin midshipman, Porichthys notatus.
Gen Comp Endocrinol 1995; 99:85–99.
21. Habibi HR, Van Der Kraak G, Bulanski E, Peter RE. Effects of teleost
GnRH on reinitiation of oocytes meiosis in goldfish in vitro. Am J
Physiol 1998; 255:R268–273.
22. Chomczynski P, Sacchi N. Single-step method of RNA isolation by
guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem
1987; 162:156–159.
23. Powell JFF, Zohar Y, Elizur A, Park M, Fischer WH, Graig AG, Rivier
JE, Lovejoy DA, Sherwood NM. Three forms of gonadotropin-re-
leasing hormone characterized from brains of one species. Proc Natl
Acad Sci U S A 1994; 91:12081–12085.
24. Maniatis T. Mechanism of alternative pre-mRNA splicing. Science
1991; 251:33–34.
25. Amara SG, Rosenfeld JMG. Alternative RNA processing in calcitonin
gene expression generates mRNAs encoding different polypeptide
products. Nature 1982; 298:240–244.
26. Csank C, Taylor FM, Martindale DW. Nuclear pre-mRNA introns:
analysis and comparison of intron sequences from Tetrahymena ther-
mophila and other eukaryotes. Nucleic Acid Res 1990; 18:5133–5141.
27. Chabot B. Directing alternative splicing: cast and scenarious. TIG
1996; 12:472–478.
28. Millar RP, Wormald PJ, Milton RC de L. Stimulation of gonadotropin
release by a non-GnRH peptide sequence of the GnRH precursor.
Science 1986; 232:68–70.
29. Heierhorst J, Mahlmann S, Morley SD, Coe IR, Sherwood NM. Mo-
lecular cloning of two distinct cDNAs from chum salmon (Onchoryn-
chus keta) suggests an ancient gene duplication. FEBS Lett 1990; 260:
301–304.
30. Sherwood NM, Lovejoy DA, Coe IR. Origin of mammalian gonad-
otropin-releasing hormones. Endocr Rev 1993; 14:241–253.