|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Neuroendocrinology |
a Instituto Tecnológico de Chascomús (IIB-INTECH, CONICET-Universidad de San Martín), Camino de Circunvalación Laguna, Km 6. CC 164 (B7130IWA) Chascomús, Provincia de Buenos Aires, Argentina
b Department of Biology, University of Victoria, Victoria, British Columbia, Canada V8W 2Y2
c Laboratorio de Embriología Animal, Departamento de Ciencias Biológicas, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires (C1428EHA), Buenos Aires, Argentina
d Aquatic Research and Environmental Assessment Center (AREAC), Brooklyn College of the City University> of New York, New York, New York 11210
e Instituto de Investigaciones Biomédicas, Fundación Pablo Cassará, Saladillo 2452 (C1440FFX), Buenos Aires, Argentina
f The Clayton Foundation Laboratories for Peptide Biology, The Salk Institute, La Jolla, California 92037
| ABSTRACT |
|---|
|
|
|---|
anterior pituitary, central nervous system, gonadotropin-releasing hormone, hypothalamic hormones, neuropeptides
| INTRODUCTION |
|---|
|
|
|---|
Studies regarding the distribution of GnRH within the brain and among species show that chicken GnRH-II (cGnRH-II) is always found in the same brain region, the mesencephalic tegmentum [1]. A second GnRH form in bony fish can vary depending on the species; to date, the second form has been identified as mammalian GnRH (mGnRH), salmon GnRH (sGnRH), or catfish GnRH (cfGnRH). The mammalian form was found in lungfish [10, 11], sturgeon, and other bony fish that evolved early [12, 13], such as Pantodon [14] and eels [15]. Salmon GnRH was detected in later-evolving bony fish [5, 14, 1619], except for the catfish, which has cfGnRH [20, 21]. Furthermore, the occurrence within a single species of a third GnRH in addition to cGnRH-II and sGnRH has also been reported. The third form varies among species representing different taxa: herring GnRH (hrGnRH) [5], seabream GnRH (sbGnRH) [22], or pejerrey GnRH (pjGnRH) [6, 7].
To date, hrGnRH has been found in only one species [5], whereas sbGnRH has been sequenced and/or detected in the brain and pituitary gland among teleosts from a number of different orders: Characiformes [23, 24], Perciformes [22, 2528], Pleuronectiformes [29], and Scorpaeniformes [30].
Recently, two different groups have demonstrated the presence of a new GnRH family member in two different teleosts: the medaka (Oryzias latipes), by sequencing the GnRH cDNA [6], and the pejerrey (Odontesthes bonariensis), by isolating and sequencing the mature peptide [7]. This form, [Phe5-Ser8]GnRH, has been given two names: medaka GnRH (mdGnRH) [6] and pjGnRH [7]. In the present study, we refer to the peptide as pjGnRH. After we reported the structure of pjGnRH, we noticed in earlier studies that small immunoreactive GnRH (ir-GnRH) peaks of unknown identity eluted in the same position as pjGnRH. The earlier reports included the molly (Poecilia latipinna [Cyprinidontiformes]) [31] and the grass rockfish (Sebastes rastrelliger [Scorpaeniformes]) [30]. In the pejerrey fish, the pjGnRH form has been localized within the preoptic area [7, 32] and has been the only form detected at the pituitary level [33], suggesting a hypophysiotropic function. However, a variable number of GnRH forms have been reported to be present in the pituitary gland of different teleost species [19, 2426, 28, 30], suggesting that further studies are needed to clarify this subject.
In previous reports, it was hypothesized that pjGnRH and sbGnRH forms are orthologous, in which pjGnRH was derived from the ancestral and widespread sbGnRH by nucleotide substitutions [6, 7]. The phylogenetic position of pejerrey fish in the order Atheriniformes is between two orders of fish (Characiformes and Perciformes) that express sbGnRH [7].
The main objective of the present study was to determine if pjGnRH has a more widespread presence among different fish species than is currently accepted. To address this question, we analyzed GnRH forms in several species belonging to three different orders grouped in the taxon Smegmamorpha: Synbranchiformes (swamp eel [Synbranchus marmoratus]), Cyprinidontiformes (platyfish [Xiphophorus maculatus] and green swordtail [X. helleri]), and Atheriniformes (Patagonia pejerrey [Odontesthes hatchery]). Moreover, to determine whether pjGnRH was the only GnRH form present in the pituitary, a careful screening was performed using our model species, the pejerrey (O. bonariensis [Atheriniformes]).
| MATERIALS AND METHODS |
|---|
|
|
|---|
All animal studies were performed in accordance with the Guide for the Care and Use of Laboratory Animals (National Academy of Science, 1996) or were approved by the University of Victoria Animal Care Committee.
Swamp eel A total of 11.8 g of whole brains with pituitaries attached (n = 200) was collected during the spawning season from adult swamp eels in freshwater lagoons of the Province of Chaco, Argentina. The animals were transported to the laboratory. The tissues were removed from freshly killed fish, placed on dry ice, and stored at -80°C.
Platyfish Male and female brains with pituitaries (n = 340, 0.35 g) were collected from known genetic stock that were bred at the Aquatic Research and Environmental Assessment Center at Brooklyn College. Fish were decapitated. The tissues were immediately removed, frozen in liquid nitrogen, and shipped on dry ice to Victoria, BC, Canada, for storage at -80°C.
Swordtail Male and female green swordtail fish (n = 1000) were grown at the Tropical Gardens Fish Farm, Florida, and shipped to Victoria. The original fish came from a pure wild stock collected in Princess Margaret Creek, Costa Rica. After anesthesia, the fish were decapitated, and the brains with pituitaries (12.0 g) were removed, quickly frozen on dry ice, and stored at -80°C.
Patagonia pejerrey A total of 21 adult male and female Patagonian pejerrey were captured at the Embalse Alicurá, Province of Neuquén, Argentina. The fish were transported to the Center of Applied Ecology of Neuquén (CAEN), Argentina, where they were killed. The brain and pituitary tissues (5.3 g) were quickly removed, frozen on dry ice, transported to Buenos Aires, and stored at -80°C.
Pejerrey A total of 300 male and female pejerrey were captured at Chascomús Lagoon, Province of Buenos Aires, Argentina. The fish were transported to the laboratory, where they were killed. The pituitaries (0.6 g) were removed and frozen on dry ice, then stored at -80°C. The brains from these fish were used previously for the identification and sequencing of pjGnRH [7].
Peptide Extraction
Frozen tissues were powdered in a cold mortar and pestle with liquid nitrogen and added to a mixture of cold 1 N HCl and acetone (3:100, v/v). For every gram of tissue extracted, 5 ml of the acid-acetone mixture was used. The extraction mixture was stirred on dry ice for 3 h and then filtered (Whatman No. 1; Whatman, Maidstone, England). The insoluble material was re-extracted for a further 5 min in a mixture of cold 0.01 N HCl and acetone (20:80, v/v) and refiltered. The combined filtrates were treated with petroleum ether (boiling point 3060°C) for five successive times [20]. The final aqueous phases were then concentrated to less than 1 ml using a vacuum concentrator.
Partial Reverse-Phase, High-Performance Liquid Chromatographic Purification of GnRH Forms
Concentrated brain extracts were filtered through a low protein absorbing filter (45 µm; µCoStar LBTM, Keenebunk, ME) and injected onto a Supelcosil LC-18 analytical column (Supelco, St. Louis, MO) using a Beckman model 166 liquid chromatograph (Irvine, CA). The sample was applied at the beginning of a 10-min isocratic period of 17% acetonitrile (ACN) in 0.25 M triethylammoniumformate (pH 6.5); ACN was then increased to 24% over a 7-min period and held isocratically for 43 more minutes. The flow rate was kept at 1 ml/min, and 1-ml fractions were collected [20]. The fractions were vacuum-dried, reconstituted in RIA buffer, and assayed for ir-GnRH using RIA. Each injection of the tissue extract was preceded by a blank run in which the mobile phase was injected. The fractions were radioimmunoassayed using the same conditions as the samples to ensure that the column was not contaminated. Fresh standards were chromatographed onto a different column to compare the elution position of each ir-GnRH peak.
For green swordtail brain extract, an additional purification step was done. The brain extract was filtered and applied (step 1) to a column that consisted of a series of 10 SepPak cartridges (Waters, Milford, MA) connected together. The applied material was eluted with mobile phases A (0.5% [v/v] trifluoroacetic acid [TFA]) and B (0.5% TFA and 80% [v/v] ACN). Sixty consecutive fractions of 1 ml each were collected, and an aliquot of 50 µl from each fraction was removed, vacuum-dried, and radioimmunoassayed for ir-GnRH using GF-4 antiserum (data not shown). Immunoreactive fractions were combined and vacuum concentrated before the next purification step.
Synthetic GnRH Standards
Fifteen forms of GnRH standards have been run many times in our laboratories on the same type of column and reverse-phase, high-performance liquid chromatographic (RP-HPLC) system. In the present study, we applied five standards to cover a broad range of elution times. Seabream GnRH, pjGnRH, and dogfish GnRH (dfGnRH) were synthesized using a solid-phase method on a methylbenzhydrylamine resin as described elsewhere [34]. Peninsula Laboratories, Inc. (Belmont, CA), supplied the cGnRH-II and sGnRH.
RIA Measurements
Radioimmunoassays were performed as previously described [7], and standards were radiolabeled according to established protocols [24]. Cross-reactivity for the antisera listed below and assay sensitivity are shown in Table 1. The RIA systems were as follows: The GF-4 antiserum was raised by NMS in rabbits against sGnRH and used at 1:25 000 (v/v) final dilution with synthetic mGnRH as tracer and standard [26, 35]. The Bla-5 was raised against lamprey GnRH-I by Dr. N.M. Sherwood in rabbits and used at a final dilution of 1:5000 (v/v) with mGnRH as tracer and standard [35]. The PBL 49, a donation of Dr. W. Vale (The Salk Institute), was used at a final dilution of 1:150 000 (v/v) with sGnRH as tracer and standard [19, 26]. The PBL 45, also a donation of Dr. W. Vale, was used at a final dilution of 1:250 000 with sGnRH as tracer and standard [19, 26]. The AS-691 antiserum, a donation of Dr. K. Okuzawa (National Research Institute of Aquaculture, Tamaki, Japan), was used at a final dilution of 1:40 000 with sbGnRH as tracer and standard. The sGnRH 2 antiserum was a donation of Dr. K. Aida (University of Tokyo, Japan). It was used at a final dilution of 1:300 000 with sGnRH as tracer and standard [33].
|
| RESULTS |
|---|
|
|
|---|
Swamp Eel Brain Extract Elution from RP-HPLC
Three ir-GnRH peaks were detected by PBL 45 antiserum: peak I, fractions 2627; peak II, fractions 3032; and peak III, fractions 4653 (Fig. 1). The same three peaks were identified using antiserum PBL 49 (data not shown). Also, the fractions were analyzed using an antiserum (AS-691) raised against sbGnRH. All three antisera cross-react with pjGnRH (Table 1). A major ir-GnRH peak was detected in fractions 3032 with AS-691 (Fig. 2). The elution positions of the three peaks matched the GnRH standards of cGnRH-II, pjGnRH, and sGnRH.
|
|
Platyfish Brain Extract Elution from RP-HPLC
Platyfish brain extract resulted in three ir-GnRH peaks that were detected by antiserum GF-4 after elution from the RP-HPLC column: peak I, fractions 2527; peak II, fractions 3032; and peak III, fractions 4548 (Fig. 3). These three peaks matched the synthetic standards of cGnRH-II, pjGnRH, and sGnRH. The GF-4 cross-reacts with these three GnRHs (Table 1), although the cross-reactivity is low with cGnRH-II.
|
Green Swordtail Brain Extract Elution from RP-HPLC
In swordtail brain extract, the first step of purification revealed a broad ir-GnRH peak in fractions 3044 using GF-4 antiserum (data not shown). A further purification of these fractions was performed with RP-HPLC, and four ir-GnRH peaks were revealed with the aid of Bla-5 antiserum: peak I, fraction 26; peak II, fractions 3134; peak III, fractions 4344; and peak IV, fractions 4851 (Fig. 4). Three of these peaks matched the same standards as for platyfish, and the fourth peak was distinct from the five standards. Both GF-4 and Bla-5 have high cross-reactivity with pjGnRH and sGnRH and low reactivity with cGnRH-II (Table 1).
|
Patagonia Pejerrey Brain Extract Elution from RP-HPLC
Three ir-GnRH peaks were detected by PBL 49 antiserum: peak I, fractions 2628; peak II, fractions 3034; and peak III, fractions 5053 (Fig. 5). Also, the chromatographic fractions were tested with AS-691 antiserum, and only one ir-GnRH peak was determined on fractions 3033 (Fig. 6). The three peaks eluted with cGnRH-II, pjGnRH, and sGnRH standards; peak II was further identified as pjGnRH by cross-reactivity with AS-691, PBL 49, and PBL 45.
|
|
Pejerrey Pituitary Extract Elution from RP-HPLC
From the pejerrey pituitary extract, one peak eluted (fractions 3132) and was identified as pjGnRH with four different well-characterized antisera: PBL 45 (Fig. 7), PBL 49 (Fig. 8), GF-4 (Fig. 9), and Bla-5 (Fig. 10). Furthermore, ir-GnRH fractions were not revealed using a highly specific sGnRH antiserum (sGnRH 2, data not shown). Also, antisera GF-4, Bla-5, PBL 45, and PBL 49 have high cross-reactivity with sGnRH (Table 1) if it is present.
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
Evidence Supports Peak I as cGnRH-II
The elution position of the ir-GnRH peak I in brain extracts of these four species is the same as that of synthetic cGnRH-II standard under identical RP-HPLC conditions (Figs. 1 and 35). These results suggest that cGnRH-II is present, because five of the antisera used for the present study have cross-reactivity against this neuropeptide. This idea is strengthened by the demonstrated presence of cGnRH-II in two closely related species: the pejerrey (Odontesthes bonariensis) by isolating and sequencing the mature peptide [7] and the medaka (Oryzias latipes) by sequencing the cDNA encoding the cGnRH-II precursor [6]. Also, previous reports have shown the presence of an ir-GnRH peak in the position of cGnRH-II in the Japanese medaka [36], green molly (Poecilia latipinna) [31], and rockfish (Sebastes rastrelliger) [30]. In addition, the brain of platyfish has been shown previously to contain cGnRH-II neurons, as revealed with a cGnRH-II-specific antiserum [37].
The stringent conservation of the cGnRH-II form throughout millions of years of vertebrate evolution suggests that its presence in the brain is essential for the reproductive success of a species. However, the function of cGnRH-II has proved elusive. For example, even though cGnRH-II clearly has gonadotropin-releasing activity in mammals [35], birds [38], and amphibians [39], it is not evident that axonal fibers from neurons that express cGnRH-II extend to the hypothalamo-hypophyseal portal system. Recent studies in primates have shown that cGnRH-II neurons are present in the mesencephalon and posterior hypothalamus [35] but are reported also for the suprachiasmatic and paraventricular nuclei and the mediobasal hypothalamus of rhesus monkeys [40], although whether cGnRH-II axons project to the median eminence to secrete the peptide into the portal system is currently unknown.
In fish, cGnRH-II not only has gonadotropin-releasing activity [7, 41], the peptide is present in the pituitary gland of elasmobranchs [42] and some teleost species, such as goldfish [19], European eel [15], African catfish [21], tilapia [26], striped bass [43], herring [5], and European sea bass [44]. Nonetheless, cGnRH-II levels in the brain remain constant during the sexual cycle in different bony fish, suggesting that its involvement in the regulation of gonadotropins or in the reproduction cycle is limited [15, 18, 28, 30, 45, 46].
In most vertebrates, cGnRH-II cells are found in the midbrain, with fibers extending to the forebrain, cerebellum, medulla oblongata, and spinal cord [1]. Although the role of this midbrain system is not known in fish, its involvement in sexual behavior has been suggested in a mammal, the musk shrew (Suncus murinus) [47]. Future studies regarding the biological function of the midbrain GnRH system in bony fish need to be done.
Evidence Supports Peak III as sGnRH
The presence of a late-eluting ir-GnRH peak (around fraction 50, peak III) in brain extracts of the studied species suggests the presence of sGnRH. In the present study, the four antisera known to cross-react with sGnRH (PBL 45, PBL 49, GF-4, and Bla-5) recognized sGnRH in the brain extracts of the four species (Figs. 1 and 35). Over the years, we have found that sGnRH, the most hydrophobic of the GnRH forms, elutes between 42 and 55 min from the same type of column and RP-HPLC system. To our knowledge, no other GnRH forms elute so late under these RP-HPLC conditions. Also, the ir-GnRH fractions 4344 of green swordtail brain extract (Fig. 4) most likely represent sGnRH. The identification of a small peak that elutes a few fractions before sGnRH in some brain extracts was previously demonstrated to be sGnRH by sequence and mass analysis of the purified peak [25].
Salmon GnRH was the first GnRH to be characterized by primary structure from the brain of a teleost (late-evolving bony fish) [17], and subsequently, it was demonstrated in the brain of most teleost species, except for catfish [20, 21]. The sGnRH form is not present in lobe-finned fish [10, 11], the ancient ray-finned bony fish that evolved before the teleosts [12, 13], and in teleosts that evolved early in evolution, such as Pantodon [14] and eels [15]. Therefore, sGnRH has been retained during evolution in most teleost species.
In contrast to tetrapods, teleostean fish lack a hypophysial portal blood system [2]. Instead, the axons containing GnRH grow directly into the pituitary. Salmon GnRH is present in nerve terminals that are distributed to the pituitary gland of many teleostean fish [5, 7, 18, 19, 24, 28, 45], and this form has been shown to have gonadotropin-releasing activity [2].
Within the brain, sGnRH, as shown by in situ hybridization, is always present in the anterior region of the ventromedial olfactory bulbs and/or terminal nerve [27, 48, 49]; in addition, some species, such as salmonids, have sGnRH in the preoptic area. In the terminal nerve region, the sGnRH-expressing neurons have intrinsic rhythmicity and project widely in the brain, but not to the pituitary gland [50]. In support of the lack of distribution of sGnRH axons to the median eminence, the adult goldfish terminal nerve ganglia were shown not to be essential for ovarian development and ovulation [51]. Instead, the primary role of sGnRH in the terminal nerve ganglia may be to coordinate olfactory and visual input for reproduction, as shown by an increase in sGnRH activity in the olfactory bulbs of migratory salmon [52, 53]. Some species, like salmonids [18] and cyprinids [19], have only two forms of GnRH instead of three. In these fish, sGnRH neurons are in both the terminal nerve and in the preoptic-ventral telencephalon region. The sGnRH in the preoptic region is related to control of the pituitary and gonadal maturation [54, 55]. Taken together, these studies indicate that sGnRH has multiple physiological functions according to the location of the neurons in the brain. In the present study, in which the four species have three forms of GnRH, it can be predicted that sGnRH is located in the anterior terminal nerve and not in the preoptic area.
Evidence Supports Peak II as pjGnRH
The ir-GnRH material detected in peak II in swamp eel (Figs. 1 and 2), platyfish (Fig. 3), green swordtail (Fig. 4), and Patagonia pejerrey (Fig. 5) represents the expression of pjGnRH. The structure of this form was recently demonstrated not only by sequencing the mature peptide in the pejerrey [7] but also by characterizing the cDNA encoding its preprohormone in medaka [6].
Pattern of Distribution of pjGnRH in Teleostean Fish
Evidence in the present study suggests that the novel pjGnRH is common to platyfish and green swordtail (Cyprinidontiformes). Likewise, the brain of another cyprinidontiform, the molly, was previously shown to contain ir-GnRH in the elution position of pjGnRH [31].
The presence of pjGnRH in brain extract of Patagonia pejerrey, together with its demonstration in pejerrey, suggests a common pattern of expression in Atheriniformes. These results, together with the presence of pjGnRH in the Japanese medaka [6], indicate that the expression pjGnRH, along with sGnRH and cGnRH-II, is a common pattern in Atherinomorpha, which groups together Atheriniformes, Beloniformes, and Cyprinidontiformes [56]. Furthermore, we present strong evidences for the presence of pjGnRH in the brain of swamp eel (Synbranchiformes). Previous data showed the presence of sGnRH and cGnRH-II in the brain of this species [57], but in that study, it was not possible to demonstrate the presence of any GnRH form in pituitary gland extracts. The main difference between those and the present data is the different sexual stage of the fish used. The present data suggest that pjGnRH is the predominant, if not the only, GnRH isoform in the pituitary gland of swamp eels, reinforcing the idea of the hypophysiotropic action of pjGnRH.
These data, together with previous reports, show that pjGnRH is expressed in Atheriniformes, Beloniformes, Cyprinidontiformes, and Synbranchiformes. These orders have been grouped together with Mugilomorpha (several orders), Gasterosteiformes (a single order), and Elassomatidae (a family) in the taxon Smegmamorpha. This group is supported as monophyletic by a single morphological synapomorphy [58]. However, the monophyly has not been corroborated by DNA sequence data derived from the mitochondrial genome [59, 60] and nuclear genes [61]. An even larger grouping for fish that have pjGnRH would have to include the order Scorpaeniformes (rockfish). Although this group evolved later than the fish mentioned above, it shares pjGnRH, sGnRH, and cGnRH-II. In contrast, the rockfish has sbGnRH in addition to the other three forms.
In conclusion, the present study has characterized the presence of cGnRH-II, pjGnRH, and sGnRH in different orders of bony fish. This evidence supports the previous hypothesis [6, 7] that pjGnRH and sbGnRH are orthologous forms and adds data regarding the taxonomic distribution of pjGnRH.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Correspondence. FAX: 54 2241 424048; somoza{at}intech.gov.ar ![]()
Received: 14 February 2002.
First decision: 19 March 2002.
Accepted: 1 July 2002.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
B. Levavi-Sivan, H. Safarian, H. Rosenfeld, A. Elizur, and A. Avitan Regulation of Gonadotropin-Releasing Hormone (GnRH)-Receptor Gene Expression in Tilapia: Effect of GnRH and Dopamine Biol Reprod, June 1, 2004; 70(6): 1545 - 1551. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |