| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Ecology, Evolution and Marine Biology, University of California, Santa Barbara, California 93106
b Department of Biology, University of Victoria, Victoria, British Columbia, Canada V8W 2Y2>
ABSTRACT
The involvement of individual molecular forms of GnRH in the regulation of reproductive cyclicity in a viviparous marine teleost, the grass rockfish (Sebastes rastrelliger), was evaluated by relating the brain and pituitary content of the neuropeptide to reproductive status. The presence of sea bream (sb) GnRH, chicken GnRH-II, and salmon GnRH in the brain was confirmed by their elution pattern on HPLC and RIA. In addition, HPLC elution profiles suggest that there may be a fourth form of GnRH. All forms of GnRH were found in male and female brains in all reproductive conditions. However, only sbGnRH could be detected in appreciable amounts in the pituitary. Of the four forms of GnRH found in the rockfish, only sbGnRH fluctuated during the reproductive cycle and large accumulations were detected in the brains and pituitaries of postspawn females and regressed males. The accumulation of sbGnRH at the end of the reproductive cycle is suggested to reflect a decline in GnRH secretion relative to synthesis. The dominance of sbGnRH in the pituitary and its individual fluctuation in relation to seasonal changes in reproductive status suggests that sbGnRH is an important regulator of gonadotropin-mediated reproductive activity in rockfish.
GnRH, hypothalamus, pituitary, seasonal reproduction
INTRODUCTION
Rockfish, members of the genus Sebastes, are viviparous scorpaenid teleosts that typically exhibit a single annual cycle of reproductive activity [1, 2]. In the majority of Northern Pacific rockfish both males and females exhibit reproductive recrudescence in late winter or spring. Groups of synchronously developing oocytes are ovulated and fertilized in the lumen of paired ovaries to initiate a prolonged period of gestation. Fecundity is comparable to that of the most highly fecund oviparous species, and at parturition, large numbers of larvae at an advanced stage of organogenesis are extruded in a single spawn [3, 4].
In vertebrates, hypothalamic GnRH provides the primary neuroendocrine link in the transduction of environmental cues into the cascade of hormonal signals that regulate changes in levels of reproductive activity. In teleosts, as in other vertebrates, the evaluation of the role of GnRH in relation to reproduction is complicated by the presence of more than one molecular form of the peptide in the brain. Most teleost fish have at least two forms of GnRH, usually salmon GnRH (sGnRH) and chicken GnRH-II (cGnRH-II). A third form, sea bream GnRH (sbGnRH) that was first identified in the gilthead sea bream (Sparus aurata) [5] is found together with sGnRH and cGnRH-II in several perciform fish [6, 7]. We have recently identified the same three forms of GnRH in the grass rockfish (Sebastes rastrelliger), thereby extending the phylogenetic origin of sbGnRH to the emergence of the Scorpaeniformes [8].
The role of individual forms of GnRH in the regulation of seasonal changes in reproductive status has received little attention. While the various forms of GnRH identified in the brains of teleosts have the potential to release gonadotropin hormones [5, 9, 10], they are not necessarily all involved in the regulation of pituitary function. Teleost fish lack a hypophysial-portal system [11], and the population of GnRH neurons that regulates pituitary function has axons that extend from the brain to the infundibulum to directly stimulate gonadotroph cells [12]. Accordingly, the determination of levels of individual forms of GnRH in the pituitary gives an indication of the variant responsible for gonadotropic hormone release. Analysis of the distribution of various forms of GnRH between the brain and pituitary in several species of teleosts reveals differences in the degree to which individual forms are transferred to the pituitary [6], indicating differences in regulatory function among the different molecular forms.
Our previous analysis of GnRH forms in the grass rockfish was performed on tissue pooled from male and female fish that were in an immature or regressed condition [8]. Accordingly, it was not possible to relate GnRH content to a particular sex or reproductive condition. The present work was conducted in order to clarify the neuroendocrine mechanisms regulating seasonal aspects of reproduction in rockfish by comparing levels of characterized molecular forms of GnRH in brains and pituitaries at component stages of the annual reproductive cycle of male and female grass rockfish. Grass rockfish (S. rastrelliger) are nearshore representatives of the genus Sebastes [13] and were chosen as our model because of their importance to the fishery of the eastern Pacific Rim.
MATERIALS AND METHODS
Sampling
Grass rockfish (59 females, 43 males) were caught by commercial fishers in the coastal waters of southern and central California. Animal studies were conducted in accordance with the Guide and Use of Laboratory Animals (copyright 1996, National Acadamy of Science) as approved by the University California, Santa Barbara Animal Care Council. Fish were anesthetized in 3-aminobenzoic acid ethyl ester (MS-222, 0.6% in seawater), weighed, and measured for length. The gonads were removed, weighed, and prepared for histological analysis as described previously [4]. This histological material was used to evaluate the stage of germ cell development. For the purpose of the present analysis, females were assigned to previtellogenic, vitellogenic, pregnant, and postspawn groups, and males were classified as either regressed or spermatogenically active. Brains and pituitaries (brains only in previtellogenic females) were removed and stored at -80°C. A sample of larvae was removed from the ovaries of a near-term pregnant female and stored at -80°C.
Peptide Extraction
Brains and pituitaries were pooled according to reproductive condition. Weights of pooled brains and pituitaries were as follows: spermatogenic male brains 11.55 g, pituitaries 0.16 g; regressed male brains 9.38 g, pituitaries 0.18 g; female previtellogenic brains 11.24 g; vitellogenic brains 6.22 g, pituitaries 0.09 g; pregnant brains 3.10 g, pituitaries 0.07 g; postspawned brains 7.82 g, pituitaries 0.23 g; larvae (total weight 9.4 g; approximately 10 000 larvae). Frozen tissues were powdered in liquid nitrogen with a mortar and pestle precooled on dry ice. Powdered material was extracted in 10 ml of a mixture of cold acetone and 0.1 N HCl (100:3 volume). The mixture was stirred for 3 h, then vacuum filtered through Whatman no.1 filter paper. Remaining solids were reextracted in 4–5 ml acetone:0.01 N HCl (4:1 volume), stirred for 3 min, and filtered again. The two filtrates were combined and lipids were removed by five successive petroleum ether treatments (4:1 volume, filtrate:ether). Between each petroleum ether treatment, the top layer was discarded. The bottom layer was recovered and subjected to the next treatment. The remaining extract was concentrated to 1–3 ml on a speed vacuum concentrator (Savant, Farmingdale, NY) and filtered through a 0.45-µm, nonpyrogenic, low protein binding filter (Costar, Cambridge, MA).
High-Performance Liquid Chromatography
Extracts were fractionated on a Supelco (Supelcosil) LC-18 HPLC column (4.6-mm x 25-cm x 5-µm particle diameter) with guard columns. The column was linked to a Beckman solvent module 166 and detector module 125. The column was prewashed initially with increasing and decreasing gradients of 0.25 M triethylammonium formate, pH 6.5 (solvent A) and acetonitrile (solvent B). The column was washed between each extract with increasing and decreasing gradients of solvent B and milli Q water. Extracts in aliquots of 700–900 µl were injected separately onto the column in a solvent environment of 83% A and 17% B, in 2-min intervals. Ten minutes following the first injection, the solvent mixture was changed to 76% A and 24% B over a gradient of 7 min. This mixture was held for 43 min. One fraction was collected each minute for a total of 60 fractions. Flow rate was 1 ml/min. Aliquots of 100 µl were taken from each fraction, vacuum dried, and stored at 4°C for RIA. Prior to the injection of each extract, a blank sample of 800 µl of milli-Q water was injected onto the column and eluted with the same solvent program as the extracts. Aliquots of 100 µl were sampled and assayed in the same manner as the extracts.
Standards
Following HPLC fractionation of all the extracts, the column was washed with solvent A and B as above. Synthetic sea bream, chicken-II, and salmon GnRH standards were mixed and injected onto the column at concentrations of 160 ng (sbGnRH and sGnRH) and 320 ng (cGnRH-II) and eluted with the same program as the extracts. Aliquots of 10 µl were removed and dried for RIA.
Radioimmunoassay
A competitive RIA as described in Sherwood et al. [14] was employed for detecting GnRH immunoreactivity in fractionated extracts. Polyclonal antisera raised in two different rabbits were used to screen extracts for GnRH immunoreactivity. These were antisera GF-6 and 7CR-10 (antisera letter designations refer to the rabbit, tailing number indicates bleed number). Mammalian GnRH125I trace and standard were used with GF-6 at a final dilution of 1:5000. Chicken GnRH-II125I trace and standard were used with 7CR-10 at a final dilution of 1:7500. Antiserum 7CR-10 was raised against dogfish (df) GnRH and has a high cross-reactivity for dfGnRH, cGnRH-II, and sGnRH. Antiserum 7CR-10 has a cross-reactivity of 85% for synthetic sGnRH and a low cross-reactivity of <0.03% for synthetic sbGnRH when referenced against synthetic cGnRH-II standard (100%) and used with lamprey GnRH-I125I trace [15]. In the present study 7CR-10 had an average binding of 22.5% and detection limit (B/B0 = 80%) of 14.1 pg when used with the cGnRH125I trace and cGnRH-II standard. Antiserum GF-6 was raised against synthetic sGnRH and cross-reacts with most known forms of GnRH. Antiserum GF-6 has the same cross-reactivity characteristics as antiserum GF-4, an earlier bleed. GF-4 has a cross-reactivity of 41% for synthetic sbGnRH when synthetic mGnRH is used as the standard [15]. GF-6 had an average binding activity of 10.4% and detection limit of 3.9 pg. Sea bream GnRH immunoreactivity is reported as detected by antiserum GF-6 because of high cross-reactivity. Chicken GnRH-II, sGnRH, and the unknown GnRH immunoreactivities are reported as detected by antiserum 7CR-10, again because of high cross-reactivity. In samples where tracer binding dropped to 20% and below, fractions were reassayed in serial dilutions of 1:2. GnRH content in adults is expressed as total GnRH detected per organ (brain or pituitary), in fractions in the same and adjacent positions in which the synthetic standards eluted. GnRH content in larvae is expressed as the total GnRH (in picograms) per gram of larvae. The GnRH immunoreactivity concentration closest to 50% is reported.
RESULTS
Gonadal Condition
All fish used in this study exceeded the size and age of first maturity for this species [4, 16]. In females, reproductive status was determined in accordance with histological and morphological criteria established for rockfish [4, 17]. In the ovaries of previtellogenic fish (n = 25), the most advanced oocytes ranged in size from 80 to 140 µm and showed no histological evidence of vitellogenesis (Fig. 1A). The ovaries of vitellogenic females (n = 13) contained oocytes ranging in size from 150 to 1000 µm that contained numerous spherical yolk globules (Fig. 1B). In pregnant fish (n = 6) the lumina of the ovaries contained either unhatched embryos or hatched larvae (Fig. 1C). Postspawn fish (n = 15) were readily identified by their gross ovarian morphology. The spent ovaries were flaccid, highly vascularized, and occasionally contained small numbers of residual larvae (Fig. 1D). Vitellogenic females were seen between September and January, pregnant individuals from December to March, and postspawn fish from January to April (Fig. 2). The gonadosomatic indices (GSI; mean ± SEM) were 3.01 ± 0.03 for previtellogenic fish, 2.22 ± 0.51 for vitellogenic fish, 20.09 ± 2.54 for pregnant fish, and 0.68 ± 0.11 for postspawn fish.
|
|
Testes of regressed fish (n = 19) were composed of cysts formed by the Sertoli cells that contained isolated or small groups of spermatogonia (Fig. 1E). The spermatogenically active (n = 24) group included individuals at various stages of testicular recrudescence with gonads containing germ cells at various stages of spermatogenesis (Fig. 1F). Spermatogenically active males were observed between September and March, with peak gonadal activity occurring in November (Fig. 2). The mean GSI for males grouped in the regressed category was 0.042 ± 0.005 while that of spermatogenically active males was 0.166 ± 0.022.
Gonadotropin-Releasing Hormone
Four forms of GnRH were detected in the brains and pituitaries of male and female grass rockfish and in the larvae (Table 1). Each form was distinguished by its elution position on HPLC and cross-reactivity with antisera GF-6 and 7CR-10. The elution position of three of the forms matched those of the synthetic sbGnRH (fractions 22–23), cGnRH-II (fraction 26), and sGnRH (fractions 45–47) standards. The three known forms of GnRH were detected in almost all extracts of both the brains and pituitaries of males and females and in the larval mass. The only exception was in the pituitary of the regressed male where the cGnRH-II concentration was below the lower limit of detection for the assay (Table 1). The fourth form of GnRH was suggested by the HPLC elution profiles where it was detected in fractions 30–34 for all of the extracts (Fig. 3 and Table 1). The unknown GnRH form showed a greater cross-reactivity with 7CR-10 as compared to GF-6. There was insufficient quantities of this immunoreactive material for purification and sequencing.
|
|
All four forms were found in the brains of both sexes and in all reproductive conditions. Chicken GnRH-II, sGnRH, and the unknown GnRH did not fluctuate in relation to the representative stages of the male and female reproductive cycles. In contrast, sbGnRH showed large increases in both postspawn females (6340 pg/brain) and regressed males (1840 pg/brain). Of the four forms of GnRH present in the brain, only sbGnRH could be detected in appreciable amounts in the pituitary (other forms did not exceed 30 pg/organ). As in the brain, massive accumulations of sbGnRH (Fig. 4 and Table 1) were found in the pituitaries of postspawn females (18 218 pg/pituitary) and regressed males (28 500 pg/pituitary). In the larvae, sbGnRH and the unknown form were present in relatively low concentrations (16 pg/g) as compared to cGnRH-II (415 pg/g) and sGnRH (111 pg/g).
|
DISCUSSION
The present study provides the first evidence for the presence of four forms of GnRH with distinct chromatographic and immunologic properties in the brains of both male and female grass rockfish. In accordance with our initial observations [8], three of these GnRH variants exhibit the same elution pattern on HPLC and cross-reactivity in RIA systems as synthetic sbGnRH, cGnRH-II, or sGnRH. In addition, the present study revealed a fourth form of GnRH that eluted in HPLC fractions 30–34 (Fig. 3) and was recognized by both GF-6 and 7CR-10 antisera. The elution pattern of this unknown form of GnRH is similar to a form recently reported in the brain and pituitary of the pejerrey, Odontesthes bonariensis [18]. This form of GnRH has recently been isolated and sequenced in our laboratory and termed pejerrey (p) GnRH. It is possible that the fourth form of GnRH detected in fractions 30–34 in the present study is pGnRH. However, establishment of the identity of the fourth form would require antisera specific to pGnRH or the sequence of the peptide or the cDNA isolated from the rockfish. Confirmation of the unknown form in grass rockfish as a distinct GnRH variant would provide the first evidence of four forms of GnRH in any teleost species.
The earliest species in which three forms of GnRH has been found by primary structure is the teleost herring, Clupea harrengus palasi, Clupeiformes [19]. In phylogeny, the next species in which three forms were sequenced is the euteleost pacu, Piaractus mesopotamicus, Characiformes [7]. The emergence of herring (hr) GnRH, then pacu GnRH (identical to sbGnRH) resulted in three forms of GnRH in the brain. The first event most likely occurred by gene duplication in an ancestor that evolved before the present-day herring. The second event, the appearance of sbGnRH in pacu probably came about from a nucleotide substitution in the herring GnRH gene, giving rise to a new GnRH form. The presence of four forms in the grass rockfish would indicate an additional gene duplication event occurring prior to the evolution of the rockfish (order: Scorpaeniformes). The detection of four forms of GnRH in the rockfish larvae indicates that the genes for these neuropeptides are expressed early during ontogeny in this teleost species.
Although all endogenous forms of GnRH identified to date release gonadotropins from the pituitary of the species of origin [9, 10], it appears that not all forms of GnRH in the brain are directly involved in the regulation of pituitary activity. Of the four forms of GnRH detected in the grass rockfish brain, only sbGnRH appeared in the pituitary in appreciable amounts where the ratio of sbGnRH to cGnRH-II ranged from 55 to 3644. A selective transfer of GnRH variants from the brain to the pituitary has been recorded in other teleost species. For example, in the sea bream (S. aurata) and the cichlids, Haplochromis burtoni and Oreochromis niloticus, three forms of GnRH can be found in the brain (sbGnRH, cGnRH-II, and sGnRH) while only sbGnRH and cGnRH-II can be detected in the pituitary of the sea bream [5, 6] and sbGnRH predominates in the pituitary of the cichlids [20, 21]. Similarly, in the salmon and trout only sGnRH is present in the pituitary, whereas both sGnRH and cGnRH-II are found in the brain [22, 23].
Of the four forms of GnRH present in the rockfish brain, three forms (cGnRH-II, sGnRH, and the uncharacterized GnRH) were present in comparable amounts in each of the reproductive categories established for male and female rockfish. In contrast, sbGnRH showed marked fluctuations in relation to reproductive status in both the brain and pituitary. In female fish, the highest levels of sbGnRH seen in the brains of postspawn fish represented a 33-fold increase from the lowest levels recorded during pregnancy. The corresponding comparison for female pituitaries revealed a 25-fold increase in sbGnRH. In males, sbGnRH levels in the brains and pituitaries of spermatogenically active fish showed a 6.6- and 3.3-fold increase, respectively, from the levels found in the sexually regressed group.
Previous studies have assessed the potential of neurosecretory cells to deliver GnRH to the pituitary by establishing correlations between the morphological attributes of the GnRH neurons and reproductive condition. A review of several studies of teleosts reveals differences in the size or number of GnRH neurons that can be related to relative gonad size in fish that exhibit an intraspecific divergence in the pattern of male reproductive behavior. The size and number of GnRH neurons may be positively or negatively correlated to male reproductive status dependent upon the species [24]. A positive correlation between the soma size of GnRH neurons and the degree of sexual maturation has been recorded not only in fish [25, 26] but also in birds [27] and mammals [28, 29]. The common correlates in these three vertebrate classes suggest the retention of similar neuroendocrine mechanisms for the initiation of puberty during vertebrate evolution. Evidence, primarily from mammalian studies, suggests that the dynamics of the synthesis and release of neuropeptides from GnRH neurons are different during seasonal fluctuations in reproductive status as compared to the initial onset of reproductive competence. Hamsters exhibiting gonadal regression following exposure to short photoperiods possess larger GnRH neuronal perikarya than their controls living in long photoperiods [29] and show no intrinsic decrease in the hypothalamic content of GnRH [30]. The present direct assessment of GnRH levels during the annual reproductive cycle in grass rockfish revealed an accumulation of the presumptive functional form of GnRH (sbGnRH) in the brains and pituitaries of spermatogenically regressed males and postspawn females. The increase in both organs implies that sbGnRH accumulates in both the perikarya and axons of GnRH neurons of regressed males and postspawn females. Our data indicate that the GnRH neurons retain the ability to store and synthesize large amounts of neuropeptides in the seasonally regressed condition and that the pituitary-mediated seasonal decline in gonadal function is, in accordance with evidence from mammals [29], primarily a response to an inhibition of GnRH secretion rather than to a decline in synthesis. Companion studies in grass rockfish [4] have shown a pronounced increase in the production of a series of 17,20-dihydroxylated steroids during pregnancy, a phase characterized by low sbGnRH levels. Conversely, these steroids fall to low levels at spawning, a phase characterized by elevated sbGnRH levels. The reciprocal nature of the relationship between circulating reproductive steroid levels and sbGnRH content in the brain and pituitary in female rockfish is consistent with the general view that the accumulation of sbGnRH at the end of the reproductive cycle is a reflection of diminished activity along the hypothalamo-pituitary-gonadal axis.
GnRH receptors have not been identified to date in rockfish, but have been reported for catfish [31], goldfish [32], and Japanese eel [33]. Binding affinity was not tested with any of the teleost GnRH receptors and the effect of different forms of GnRH was not tested with the eel. However, cGnRH-II was more potent by several hundredfold than catfish GnRH in activating second messenger systems such as cAMP and inositol triphosphate (IP3) in cells transfected with the receptor [31]. In both goldfish GnRH receptors, the order of potency for release of IP3 was cGnRH-II > sGnRH > mGnRH > sbGnRH [32]. The question arises as to the effectiveness of sbGnRH as the regulator of pituitary gonadotropin secretion in the rockfish. While the lower potency of sbGnRH compared to sGnRH and cGnRH-II is a consideration, the higher abundance of sbGnRH in the pituitary compared to cGnRH-II (55- to 3644-fold) and compared to sGnRH (24- to 7125-fold) implies that sbGnRH is likely to be effective as a gonadotropin regulator. Rather, sGnRH is released from axon terminals in the forebrain and cGnRH-II in other parts of the brain and spinal cord where the peptides bind to GnRH receptors on nearby neurons.
In summary, the dominance of sbGnRH in the pituitary and its individual fluctuation in relation to seasonal changes in reproductive condition establishes the concept that sbGnRH is the functional form of GnRH with respect to gonadotropic hormone release and the regulation of reproductive activity in the grass rockfish.
FOOTNOTES
First decision: 26 September 2000.
1 This work was funded in part by a grant from the Committee on Research, University of California; the National Sea Grant College Program, National Oceanic and Atmospheric Administration, U.S. Department of Commerce, under grant number NA66RG0477 project number R/F-59PD; and the Medical Research Council of Canada. 
2 Correspondence: Peter M. Collins, Department of Ecology, Evolution and Marine Biology, Bldg. 478, University of California Santa Barbara, Santa Barbara, CA 93106. FAX: 805 893 4724; collins{at}lifesci.ucsb.edu
Accepted: February 22, 2001.
Received: August 25, 2000.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  E. H. Leder, R. G. Danzmann, and M. M. Ferguson The Candidate Gene, Clock, Localizes to a Strong Spawning Time Quantitative Trait Locus Region in Rainbow Trout J. Hered., January 1, 2006; 97(1): 74 - 80. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G.M. Somoza, D.W. Lescheid, L.A. Miranda, F.L. Lo Nostro, L. Magliulo-Cepriano, A.D. Montaner, M.P. Schreibman, J.E. Rivier, and N.M. Sherwood Expression of Pejerrey Gonadotropin-Releasing Hormone in Three Orders of Fish Biol Reprod, December 1, 2002; 67(6): 1864 - 1871. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Gray, B. A. Adams, C. M. Warby, K. R. von Schalburg, and N. M. Sherwood Transcription and Translation of the Salmon Gonadotropin-Releasing Hormone Genes in Brain and Gonads of Sexually Maturing Rainbow Trout (Oncorhynchus mykiss) Biol Reprod, November 1, 2002; 67(5): 1621 - 1627. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. White, T. Nguyen, and R. D. Fernald Social regulation of gonadotropin-releasing hormone J. Exp. Biol., September 1, 2002; 205(17): 2567 - 2581. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
