| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Neuroendocrinology; |
Department of Anatomy, University of Bristol, Bristol BS2 8EJ, United Kingdom
ABSTRACT
Previous studies have provided evidence for a paracrine interaction between pituitary gonadotrophs and lactotrophs. Here, we show that GnRH is able to stimulate prolactin (PRL) release in ovine primary pituitary cultures. This effect was observed during the breeding season (BS), but not during the nonbreeding season (NBS), and was abolished by the application of bromocriptine, a specific dopamine agonist. Interestingly, GnRH gained the ability to stimulate PRL release in NBS cultures following treatment with bromocriptine. In contrast, thyrotropin-releasing hormone, a potent secretagogue of PRL, stimulated PRL release during both the BS and NBS and significantly enhanced the PRL response to GnRH during the BS. These results provide evidence for a photoperiodically modulated functional interaction between the GnRH/gonadotropic and prolactin axes in the pituitary gland of a short day breeder. Moreover, the stimulation of PRL release by GnRH was shown not to be mediated by the gonadotropins, since immunocytochemical, Western blotting, and PCR studies failed to detect pituitary LH or FSH receptor protein and mRNA expressions. Similarly, no gonadotropin receptor expression was observed in the pituitary gland of the horse, a long day breeder. In contrast, S100 protein, a marker of folliculostellate cells, which are known to participate in paracrine mechanisms within this tissue, was detected throughout the pituitaries of both these seasonal breeders. Therefore, an alternative gonadotroph secretory product, a direct effect of GnRH on the lactotroph, or another cell type, such as the folliculostellate cell, may be involved in the PRL response to GnRH in these species.
follicle-stimulating hormone receptor, gonadotroph, gonadotropin-releasing hormone, lactotroph, pituitary, prolactin, seasonal reproduction
Paracrine interactions within the pituitary gland are thought to play an important role in the regulation of fertility. Whereas physical associations, in the form of gap junctions, have been documented between gonadotrophs and lactotrophs in the rat [1], the presence of gonadotrophs within clusters of lactotrophs has been observed not only in this species but also in large vertebrates such as the sheep and the horse [2–4]. These cellular associations provide morphological evidence for possible regulation of gonadotroph function by lactotroph cells, or, alternatively, of lactotroph function by gonadotroph cells. Indeed, prolactin (PRL) receptors have been detected in the pituitary gland of several species, and their expression was shown to be specific to gonadotrophs in the sheep, supporting a role of PRL in the control of gonadotropin secretion [3, 5, 6]. Moreover, PRL was reported to reduce both basal and GnRH-stimulated LH release in the rat [7], and a recent study by our group demonstrated the suppressive effects of PRL and dopamine upon the LH response to GnRH in ovine primary pituitary cultures [8].
The possibility that gonadotoph-lactotroph associations could underlie a regulatory role of gonadotroph cells on lactotroph function has also been demonstrated. Treatment of rat pituitary cultures with GnRH leads to a rapid and sustained rise in PRL, an effect not observed in lactotroph-enriched cultures [2]. Notably, in the same study, conditioned media obtained from GnRH-stimulated gonadotroph-enriched aggregates was able to stimulate PRL release from lactotroph-enriched cultures [2]. Furthermore, it was subsequently shown that the addition of a population of 
T3–1 gonadotroph cells, which are known to express the GnRH receptor [9], restored the stimulation of PRL release by GnRH in rat lactotroph-enriched aggregates [10]. These studies, therefore, provide compelling evidence for a paracrine mode of interaction between the gonadotropic and lactotropic axes, rather than a direct effect of GnRH on the lactotroph.
Since LH and FSH are the major secretory products of the gonadotroph, it is plausible that these gonadotropins play a role in the transfer of the GnRH-stimulated gonadotroph signal to the lactotroph in the adult sheep. Indeed, Chabot et al. [11] demonstrated that the subunit of the gonadotropin glycoprotein hormones had a stimulatory effect upon PRL release from 50-day-old ovine fetal pituitary explants.
Interestingly, folliculostellate (FS) cells are also able to elicit the release of PRL from lactotrophs [12]. These agranular, nonendocrine cells have numerous gap junction connections with lactotrophs and can secrete a wide range of substances [13–15]. They are electrically excitable, and a seasonal change in gap junction content has been reported in the mink; it has been hypothesized that FS cells may contribute dynamically to the control of PRL secretion via an effect on cellular synchronization during the reproductive cycle [16, 17]. To our knowledge, FS cell expression has not been described in the sheep or horse pituitary glands. Their characterization is thus essential before a role of the FS cell can be implicated in the temporal regulation of PRL secretion and fertility in these seasonal breeders.
Therefore, the aims of this study were to: 1) determine whether GnRH can readily affect PRL release from ovine pituitary cultures during the breeding (BS) and nonbreeding (NBS) seasons, 2) examine the role of gonadotropins in this response by assessing LH receptor (LHCGR) and FSH receptor (FSHR) mRNA and protein expression within the pituitary of this short day breeder, 3) characterize the expression and morphology of FS cells in the ovine pituitary gland, and 4) investigate the expression of LHCGR and FSHR and the presence of FS cells in the equine pituitary gland to allow comparison with a long day breeder.
Ovine pituitary glands were obtained from ewes during the BS (January–February) and NBS (July–August). Animals were killed for commercial reasons at an abattoir (Baker's, Nailsea, U.K.), and the pituitaries were dissected out immediately after death.
Equine pituitary glands were obtained from gonadal-intact female and gonadectomized male horses during the BS (July–August) and the NBS (January–February). Animals were killed for commercial reasons at an abattoir (Potter Abattoir, Taunton, U.K.), and the pituitaries were dissected out immediately after death. Mares were considered to be sexually active on the basis of a recently formed corpus luteum, together with the presence of a large follicle (>2 cm). In contrast, mares were considered to be anestrus when no corpora lutea but a corpus albicans was observed in the gonad, and follicles present were <2 cm diameter. All animals were sexually mature Thoroughbreds between 3 and 18 yr of age.
Ovine primary pituitary cell cultures were produced following a method previously described [18]. Briefly, 2-mm3 pituitary blocks were incubated in 12.5 ml incomplete M199 containing 0.006 g collagenase and 0.006 g hyaluronidase (Roche Diagnostics, Hertfordshire, U.K.) for 75 min at 37°C in a shaking water bath. Following incubation, the explants were immersed in Ca- and Mg-free sterile PBS (Invitrogen, Paisley, U.K.) containing 0.2 mM EDTA (Sigma-Aldrich, Poole, U.K.) before dispersion in Ca- and Mg-free PBS. The supernatant was then aspirated, and an equal volume of complete M199 containing 10 U/ml penicillin, 10 µg/ml streptomycin, 20 µg/ml gentamicin, 5 µg/ml insulin (all Sigma-Aldrich), and 10% charcoal stripped lamb serum (Invitrogen) was added. The resulting suspension was then centrifuged, the supernatant was removed, and the cells were resuspended in complete M199. The cell number was calculated using a Neubauer hemocytometer, and the cells were plated at 400 000 cells per well onto 24-well plates (Costar, Japan). The cells were incubated for 7 days at 37°C and 5% CO2, and the medium was replaced every other day. Rat pituitary monolayer cultures have previously been reported to retain cellular contacts similar to those observed in vivo [19]. Microscopically, ovine primary pituitary cultures also appear to maintain these contacts [8].
On Day 7 of culture, cells received GnRH (Sigma-Aldrich) at concentrations of 0, 10–10, 10–9, 10–8, or 10–7 M for 90 min; this incubation time has previously been shown to lead to optimal LH and FSH release in an identical culture system [8]. To examine whether the level of secretory activity of lactotroph cells could influence their response to GnRH, all wells were washed twice with M199 prior to application of one of the following treatments in conjunction with GnRH: a) Control (medium alone), b) Bromocriptine (Br), or c) thyrotropin-releasing hormone (TRH). Bromocriptine (10–8 M; Sigma-Aldrich) was administered over the 7-day culture period to chronically suppress endogenous PRL release. To increase the concentration of PRL in the culture environment, 10–7 M TRH (Sigma-Aldrich) was applied immediately before each dose of GnRH for 90 min. Seasonal changes in the combined effects of PRL and dopamine on the LH response to GnRH have previously been shown to persist after 7 days in an identical culture system, and it is known that the mechanisms involved in circannual cell memory show long-term persistence in vivo [8, 20].
The concentration of PRL in the culture wells was measured by radioimmunoassay following a method described by McNeilly and Andrews [21]. NIAMDD-oPRL-I-1 (NIH) was used for iodination and as standard. The limit of detection of the assay was 0.5 ng/ml. The ovine PRL antibody (ASMcN R50; a gift from Professor Alan S. McNeilly, Medical Research Council, Human Reproductive Sciences Unit, Edinburgh, U.K.) was used at a concentration of 1:128 000. All samples were run in duplicate in a single assay, and the mean intra-assay coefficient of variation was 6.7% for low, medium, and high control samples.
Total RNA was extracted from samples of equine pituitary, ovine pituitary, and ovine testis using the RNeasy midi kit (Qiagen, Crawley, U.K.). RNA was treated with DNase (Invitrogen) to remove genomic DNA products. Reverse transcription was performed using a first strand cDNA synthesis kit (Roche). Four sets of primers were used for PCR amplification: 1) primers for the ovine glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA sequence: sense 5'-GAACGGGAAGCTCACTGGCAT-3', antisense 5'-GTCCACCACCCTGTTGCTGTAG-3' predicted to produce a 306-bp product; 2) primers for the ovine PRL receptor (PRLR), common to long and short forms: sense 5'-GACGATGGAGGACTTCCTACCAATTA-3', antisense 5'-GCAGGTCACCATGCTATAGCCCTT-3', predicted to produce a 645-bp product [3]; 3) primers for the ovine FSH receptor (FSHR): sense 5'-CACCATCGACTCTGTCACTGCTC-3', antisense 5'-GAATGACCGGTCCAGAGGCTCC-3', predicted to produce a 613-bp product; and 4) primers for the bovine LH receptor (LHCGR): sense 5'-CTTCTGCATGGGGCTCTA-3', antisense 5'-TCACGTTGAGAATCAGGATGG-3', predicted to produce a 397-bp product [22]. All primers were designed to span an intron to ensure genomic DNA was not amplified. Primers for the constitutively expressed gene GAPDH were used to confirm the integrity of the RNA and efficacy of the PCR reaction. Complementary DNA samples were subjected to PCR amplification consisting of an initial denaturation step at 2 min followed by 35 cycles of an annealing step at 55°C (GAPDH and LHCGR) or 60°C (PRLR and FSHR) for 45 sec, and extension at 72°C for 1 min. A final extension period at 72°C for 10 min completed the amplification. Three control tubes were run in parallel, one in which water replaced the RNA, one in which water replaced the RNA and reverse transcriptase was omitted, and one in which ovine pituitary RNA was used but reverse transcriptase was omitted. After amplification, PCR products were subjected to electrophoresis through a 1.5% agarose gel containing 0.1 µg/ml ethidium bromide and were visualized and photographed under UV light.
Western Blot Analysis: LHCGR and FSHR
Western blot analysis was carried out following the method of Laemmli et al. [23]. Briefly, tissue samples (ovine pituitary, equine pituitary, rat testis, and equine ovary) were homogenized in whole cell lysis buffer (250 mM NaCl, 5 mM Hepes, 2 mM EDTA, 10% glycerol, 0.5% NP40, 0.2 mM PMSF, 0.2 mM sodium orthovanadate and protease inhibitor; Bio-Rad, Hertfordshire, U.K.) before removal of the supernatant for determination of protein concentration. Protein (100 µg) was added to an equal volume of SDS-gel loading buffer (125 nM Tris-HCl [pH 8.0] containing 4% SDS, 2% 2-mercaptoethanol, 20% glycerol, 0.02% bromo-phenol blue), and denatured before electrophoresis through a 10% SDS-polyacrylamide gel and electrotransfer to polyvinylidene difluoride membrane (Roche). Membranes were incubated as follows: 1) 1% Western blocking reagent (Roche) overnight at 4°C; 2) rabbit anti-LHCGR immunoglobulins 1:1000 (a gift from Dr. P. Roche, Mayo Clinic, MN) or rabbit anti-FSHR immunoglobulins 1:400 (Zymed Laboratories Inc., CA) for 2 h at room temperature; 3) donkey anti-rabbit IgG linked to horseradish peroxidase (DAKO, Ely, U.K.) 1:2000 for 1 h at room temperature; and 4) chemiluminescence blotting substrate (Roche) for 1 min followed by exposure to hyperfilm ECL (Amersham, Buckinghamshire, U.K.).
Nonfluorescent Staining: LH, LHCGR, FSHR, and S100
Ovine and equine pituitary glands were prepared for immunocytochemistry as previously described [3, 4]. The sections were incubated for 24 h at 4°C with one of the following specific primary antibodies: (i) mouse monoclonal antibody specific to the bovine LH β subunit (LHB; 518 B7, 1:1000; a gift from Dr. Jan F. Roser, University of California, Davis, CA); (ii) rabbit polyclonal antibody specific to the rat LHCGR (1:50; Dr. P. Roche); (iii) rabbit polyclonal antibody specific to the human FSHR (1:50; Zymed); (iv) mouse monoclonal antibody specific to bovine S100 protein (a specific FS cell marker; 1:200; Sigma-Aldrich); and (v) rabbit polyclonal antibody specific to bovine S100 protein (1:100; DAKO). Slides were washed 4 x 5 min in 0.01 M PBS/BSA (pH 7.4) prior to application of the secondary antibodies. For polyclonal antibodies (LHCGR, FSHR, and S100), a swine anti-rabbit biotinylated antibody was used (1:200; DAKO). For monoclonal antibodies (LHB or S100), a rabbit anti-mouse biotinylated antibody was used (1:200; DAKO). Sections were incubated for 2 h at room temperature and washed as previously before the application of ABC complex (DAKO) for 30 min at room temperature. Slides were developed in diaminobendizine tetrahydrochloride (DAB; DAKO), counter-stained with hematoxylin, and coverslipped using DePeX (Merck Ltd., Hull, U.K.).
Single Immunofluorescent Staining: LH, LHCGR, FSHR, and S100
Immunofluorescent staining was conducted following the method described by Tortonese et al. [3]. For polyclonal antibodies (LHCGR, FSHR, S100p, and PRLR), a fluorescein-conjugated donkey anti-rabbit serum was used (1:20; DAKO). For monoclonal antibodies (LHB, S100m), a rhodamine-conjugated goat anti-mouse serum was used (1:20; DAKO). Slides were cover-slipped with Vectashield (Vector Laboratories, Peterborough, U.K.).
Double Immunofluorescent Staining: LH/S100 and PRLR/S100
For double immunofluorescent staining, the LHB monoclonal antibody was used in combination with the S100 polyclonal antibody (S100p), and the S100 monoclonal antibody (S100m) was used in combination with a PRLR polyclonal antibody to the extracellular domain of the rat PRLR (R120). Sections were blocked in a combination of normal goat (DAKO) and donkey (Sigma-Aldrich) sera. The first primary antibody was applied at twice the final concentration (LHB, 1:500; S100p, 1:50) for 20 min before application of the second primary antibody at twice the final concentration (S100m, 1:25; PRLR, 1:25) for 24 h at 4°C. Following four 5-min washes in 0.01 M PBS/BSA (pH 7.4), slides were incubated for 1 h at room temperature with rhodamine-conjugated goat anti-mouse secondary antibody applied at 1:20 dilution. After repeating the washes as before, the fluorescein-conjugated donkey anti-rabbit secondary antibody was applied at 1:20 for 1 h at room temperature, and the slides were washed as previously and cover-slipped with Vectashield.
For both fluorescent and nonfluorescent staining, control sections in which the primary antibodies were either omitted or replaced by their equivalent concentration of normal serum were included in each staining run.
In both the BS and NBS cultures, three wells were assigned to each dose of GnRH for each treatment group, and the concentrations of PRL were measured in duplicate. Each experiment was carried out three times. Reported values represent the mean ± SEM. The effects of treatment and season on the PRL response to GnRH were examined by ANOVA and Fisher PLSD post hoc test (Statview; SAS, Cary, NC).
PRL Response to GnRH in Ovine Primary Pituitary Cell Cultures During the Breeding Season
Under control conditions, GnRH significantly stimulated PRL secretion (P < 0.05; Fig. 1A, CON). Indeed, at a dose of 10–7 M, GnRH was able to induce a 38% increase in PRL release over basal values (311.13 ± 10.14 vs. 225.78 ± 12.24 ng/ml, for 10–7 M and 0 M GnRH, respectively). Treatment with the dopamine agonist Br for 7 days significantly suppressed basal PRL secretion (98.42 ± 5.12 ng/ml; P < 0.01) and completely abolished the PRL response to GnRH (100.72 ± 12.65 ng/ml, for 10–7 M GnRH; P < 0.01; Fig. 1A, Br). Treatment with TRH significantly increased basal PRL release (407.44 ± 34.6 ng/ml; P < 0.01) and did not prevent but actually enhanced the PRL response to GnRH (592.62 ± 39.55 vs. 407.44 ± 34.6 ng/ml, for 10–7 M and 0 M GnRH, respectively, a 45% increase; P < 0.01; Fig. 1A, TRH).
|
PRL Response to GnRH in Ovine Primary Pituitary Cell Cultures During the Nonbreeding Season
Under control conditions, GnRH was unable to significantly stimulate PRL release (Fig. 1B, CON). Treatment with Br for 7 days effectively suppressed basal PRL release (93.84 ± 0.42 vs. 176.72 ± 8.78 ng/ml, for Br and CON, respectively; Fig. 1B, Br and CON). In contrast to results for the BS, GnRH was able to stimulate PRL release at all doses tested in the presence of Br (P < 0.05; Fig. 1B, Br). The maximal response was observed at 10–7 M GnRH (197.06 ± 17.39 ng/ml), representing a 100% increase over basal values. Treatment with TRH significantly enhanced basal PRL release during the NBS; however, GnRH was unable to modify this response (377.86 ± 17.39 vs. 8.78 ng/ml for TRH and CON, respectively; P < 0.05; Fig. 1B, TRH and CON).
Assessment of Gonadotropin and PRL Receptor Gene Expression
LHCGR mRNA was clearly visible as a distinct band of 397 bp in the positive control tissue (ovine testis; Fig. 2G). In contrast, no band was apparent in either ovine or equine pituitary extracts obtained during the summer (equine BS and ovine NBS; Fig. 2, D and F). A similar result was obtained for FSHR gene expression, with no receptor mRNA detected in the pituitary glands of either species. As reported in previous studies, expression of the PRLR in ovine pituitary tissue was clearly indicated by a 645-bp band (Fig. 2D). Interestingly, a product of identical size was amplified in the equine pituitary (Fig. 2F). GAPDH expression was observed in all samples (except controls; Fig. 2, B, C, and E).
|
Detection of LH and FSH Receptor Protein by Western Blotting
In the positive control tissue, LHCGR and FSHR protein were observed as distinct bands of approximately 60 kDa and 87 kDa, respectively (rat testis and equine ovary; Fig. 3, C and F, respectively). However, equivalent bands were not apparent in either the ovine or equine pituitary extracts obtained during the winter (equine NBS and ovine BS; Fig. 3, A and B for LHCGR; Fig. 3, D and E for FSHR).
|
Gonadotropin Receptor Protein Visualization
Both nonfluorescent and fluorescent immunocytochemistry identified LH-positive gonadotrophs, illustrating their widespread localization throughout the pituitary. These results were used to verify the validity of the techniques and quality of specimens used in each staining run (Fig. 4, A and B).
|
No LHCGR immunostaining was apparent using either fluorescent (data not shown) or nonfluorescent methods in male or female equine pituitaries or in female ovine pituitaries obtained during either season in both species. The nonfluorescently (DAB) stained sections (Fig. 4, D and E) demonstrate this clearly, with a complete absence of immunoreactivity in the equine or ovine pituitaries and no difference with the negative controls (Fig. 4F). Similar to the results obtained for the LHCGR, no positive staining was detected for the FSHR in either equine or ovine pituitaries obtained during either season (Fig. 4, G and H). In contrast, ovine ovary, used as a positive control, exhibited strong immunoreactivity in the follicular granulosa cells, demonstrating the effectiveness of the LHCGR antibody to bind to its specific antigen when the receptor is present (Fig. 4C).
Positive fluorescent staining was clearly apparent for the PRLR (Fig. 4I), demonstrating widespread distribution of these receptors in the pars distalis. S100 protein (an FS cell marker) showed abundant staining in the nonfluorescent sections from both equine and ovine pituitaries (Fig. 4, J and K). Interestingly, positive fluorescent staining for S100 protein was clearly detected in neuronal-like cells of the infundibular stalk of the equine pituitary gland (Fig. 4L).
The results of this study show that GnRH is able to dose-dependently stimulate PRL release from ovine primary pituitary cultures and that this effect is not mediated by the gonadotropins LH and FSH. Interestingly, the PRL response to GnRH was abolished by the concomitant application of Br, a dopamine agonist and powerful suppressor of PRL release, whereas TRH, a potent secretagogue of PRL, enhanced the stimulatory effects of GnRH. Previous studies in the rat had shown that both GnRH and gonadotroph-conditioned medium were able to stimulate PRL release when applied to pituitary cultures [2]. However, an interaction of this kind between the gonadotropic and PRL axes has not been previously reported in a photoperiodic species such as the sheep. Intriguingly, the stimulation of PRL release by GnRH and the effects of Br and TRH observed here displayed a marked seasonal differential response. Stimulation of PRL release by GnRH could only be observed in pituitary cultures generated from animals in the BS. Although Br abolished this response, it restored the ability of GnRH to stimulate PRL release in NBS cultures. Furthermore, TRH stimulated basal PRL release during both the BS and NBS, but significantly enhanced the PRL response to GnRH only during the BS.
The slightly lower PRL concentrations detected under control conditions during the NBS (summer) when compared to the BS (winter) reflect the increased synthesis and secretion of PRL at that time of year in vivo, which result in less PRL being available for secretion in vitro. This phenomenon has previously been shown to account for the decreased in vitro secretion of LH during the BS (winter), when gonadotropin secretion in vivo is highest [8]. GnRH, at a dose of 10–7 M, induced a 38% increase in PRL release, a similar response to that observed in previous studies in vitro using fetal ovine pituitaries [11]. The magnitude of this response does not differ from that detected in vivo, as GnRH was shown to elicit a 30% increase in PRL concentrations in humans [24].
The seasonal effect upon the stimulation of PRL release by GnRH is likely to be mediated by photoperiod and melatonin. This neurohormone is released from the pineal gland during darkness. Therefore, short- and long-duration nocturnal peaks of melatonin occur in the summer and winter, respectively [25, 26]. Melatonin receptors are densely expressed in the ovine pars tuberalis [27], and their reduced activation under long days leads to the stimulation of PRL release [28, 29]. Hence, during the summer, when sheep are sexually inactive (NBS), the lactotroph may be unresponsive to GnRH since, at this time of year, PRL secretion is maximal. However, if lactotroph output was the sole determinant of the seasonal change in responsiveness to GnRH, then a robust stimulation of PRL secretion by TRH, irrespective of season, as well as a seasonal effect of TRH upon the PRL response to GnRH, would not be observed. During the NBS, high PRL concentrations lead to increased dopamine release from hypothalamic dopaminergic neurons via a short loop feedback mechanism [30, 31]. Dopamine binds to its specific D2 receptors on the lactotroph membrane to suppress PRL release, allowing PRL to regulate its own secretion [32]. In the present study, it is the ability of GnRH to overcome the inhibitory effects of Br on PRL release during the NBS (summer) which is most physiologically relevant since the resulting enhancement of PRL levels may allow the subsequent suppression of the gonadotroph response to GnRH [33, 34]. Indeed, we have previously shown that PRL only suppresses the LH response to GnRH when applied concomitantly with dopamine during the ovine NBS [8]. Conversely, the abolition of the PRL response to GnRH by Br during the BS (winter) is likely to relate to the reduced dopaminergic tone in vivo at the time the specimens were obtained. Increased sensitivity to exogenously applied dopamine at this time may lead to an enhancement of its suppressive effects on the lactotroph and the subsequent inability of GnRH to stimulate PRL release.
To investigate whether the gonadotropins themselves play a role in the gonadotroph-lactotroph transfer of the GnRH signal, expression of LH and FSH receptor protein and mRNA in the ovine and equine pituitary glands was characterized. Immunocytochemical techniques failed to detect the presence of gonadotropin receptor protein within the pituitaries of both species during their BS and NBS. To account for either protein degradation or a lack of protein transcription, RT-PCR was used to assess receptor mRNA expression. The results showed that LHCGR and FSHR mRNA was undetectable in ovine or equine pituitaries obtained during the summer (ovine NBS and equine BS). It is unlikely that tissue preparation or season could have had a detrimental effect on receptor expression, as LHCGR and FSHR proteins were also not detected by Western blotting in ovine or equine pituitaries obtained during the winter (ovine BS and equine NBS), the only time of year when GnRH was able to induce PRL release. Similar-size bands to those observed in the positive controls have previously been reported [35, 36]. Since LH and FSH receptors were shown not to be expressed in the pituitaries of the sheep and horse, the stimulatory effects of GnRH on PRL release are unlikely to be mediated by LH or FSH. Therefore, these effects probably result from either a direct action of GnRH on the lactotroph cell, or from an indirect action involving intercellular signals other than gonadotropins.
Interestingly, RT-PCR also detected the expression of PRLR mRNA within both equine and ovine pituitary glands. This is consistent with evidence from previous investigations in the rat [6] and sheep [3], and is a novel demonstration of PRLR gene expression in the equine pituitary, supporting a paracrine role of PRL within this tissue.
The lack of gonadotropin receptor expression in the ovine and equine pituitaries appears to be in contrast to the findings by Chabot et al., who reported that, in fetal lambs, the glycoprotein subunit (-GSU) was able to stimulate PRL secretion [11]. It is possible that fetal lambs transiently express pituitary gonadotropin receptors, with downregulation occurring shortly after birth. This phenomenon has been previously demonstrated in fetal rats where the expression of melatonin receptors was reported in the pars distalis during development, but declined shortly after parturition [37]. In the previous study in fetal sheep, it was noted that -GSU expression was only observed in cells staining positively for LHB protein [11]. LH, FSH, thyrotropin, eCG, and hCG are all composed of a common subunit and a β subunit specific to each hormone; the overlap between the and β subunits is thought to confer the ability to bind to their specific receptors [38, 39]. Although upregulation of -GSU expression by GnRH in gonadotrophs may provide the necessary means for the stimulation of PRL release, it is not clear how -GSU could result in this effect. Indeed, the gonadotropins have a low dissociation constant and so degrade into free and β subunits slowly [40]. Nevertheless, studies in the sheep and the hamster have described the re-expression of -GSU in the pars tuberalis, which coincided with the onset of short day photo-refractoriness and the consequent increase in prolactin concentrations [41, 42].
An alternative mechanism underlying the stimulatory effects of GnRH on PRL release is that a tertiary cell type may be responsible for the mediation of the signaling processes involved. Immunocytochemical staining demonstrated the abundant expression of folliculostellate cells in both the pars distalis and pars tuberalis of the ovine and equine pituitary glands. It has previously been documented that folliculostellate cells, like gonadotrophs, are able to stimulate the release of PRL [12]. Conversely, the same cell has been shown to inhibit the PRL response to TRH and angiotensin II in a paracrine manner [43]. These differential effects are to be expected given the array of second messengers produced by these cells; indeed, inositol phosphates, cyclic nucleotides, calcium, fibroblast growth factor, vascular endothelial growth factor, interleukins, and follistatin have all been shown to be secreted [14, 15, 44, 45]. Interestingly, we have observed a change in morphology of this cell type in the equine pituitary throughout the annual reproductive cycle, and so it is not inconceivable that there is also a seasonal modulation of the paracrine factors released [46]. It is unlikely that the previously reported [17, 19] seasonal modulation of gap junction content plays a role, as despite the similar topographical nature of pituitary monolayer cultures to that seen in vivo, gap junctions do not persist in vitro [17, 19]. Therefore, seasonal changes in the release of paracrine factors by FS cells may represent a mechanism by which this cell type can alter the lactotroph response to the gonadotroph.
In contrast, it is possible that the gonadotroph itself can still mediate the PRL responses to GnRH, especially in nonphotoperiodic species such as the rat, via an alternative secretory product. Immunocytochemical studies have identified a specific subtype of gonadotroph that possesses serotonin (5-HT) granules as well as LH and FSH granules [47, 48]. Since 5-HT is a potent inducer of PRL release from pituitary cultures [49, 50] and has stimulatory effects upon immortalized GnRH neurons depending on the receptor subtype activated [51], a regulatory role of 5-HT in GnRH-induced PRL release cannot be ruled out.
It is unlikely that GnRH can have a direct effect upon the lactotroph to stimulate PRL release, since an enriched population of lactotrophs only responded to GnRH in the presence of gonadotrophs [2, 10]. However, the study was performed using rat pituitary cells, and differences in the distribution of specific receptors have been highlighted between the pituitaries of the sheep and rat [3, 52]. Also, there may be a requirement for the simultaneous occupancy of the GnRH receptor in both the lactotroph and gonadotroph before stimulation of PRL can be achieved. Further hybridization and immunocytochemical studies are required to exclude these possibilities since GnRH receptor distribution in the ovine pituitary has not been characterized.
In conclusion, the results of the present study demonstrate that GnRH is able to elicit the release of PRL in ovine pituitary cultures and that this effect is modulated by season/photoperiod. Moreover, since neither ovine nor equine pituitaries were shown to express either form of the gonadotropin receptor, it is unlikely that LH or FSH play any part in mediating this effect of GnRH, or indeed, in any paracrine mechanisms operating within the pituitary. In contrast, this study has shown that PRLR mRNA is expressed in the equine pituitary gland, corroborating previous findings reporting the presence of PRLR protein in the pituitary of this species. In addition, the abundant expression of S100 protein detected throughout the pituitary glands of these seasonal breeders suggests a possible role of the FS cell in the photoperiodic modulation of the GnRH-stimulated PRL release.
ACKNOWLEDGMENTS
We would like to thank the NIDDK's National Hormone and Peptide Program and Dr. A.F. Parlow for providing biologically active rat PRL, and Potter Abattoir (Cappards Farm, Bishop Sutton, Bristol, U.K.) and Baker's Abattoir (Nailsea, Bristol, U.K.) for providing the sheep and horse specimens used in these studies.
FOOTNOTES
Correspondence: 1Domingo J. Tortonese, Department of Anatomy, University of Bristol, Southwell Street, Bristol BS2 8EJ, United Kingdom. FAX: 44 117 925 4794; e-mail: d.tortonese{at}bristol.ac.uk
2These authors contributed equally to this work. 
Received: 5 July 2007.
First decision: 26 September 2007.
Accepted: 5 November 2007.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
