This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kaiser, U. B. Articles by Chin, W. W. Search for Related Content PubMed PubMed Citation Articles by Kaiser, U. B. Articles by Chin, W. W.

Studies of Gonadotropin-Releasing Hormone (GnRH) Action Using GnRH Receptor-Expressing Pituitary Cell Lines¹

Division of Genetics (U.B.K., W.W.C.), Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts 02115; Oregon Regional Primate Research Center (P.M.C.), Beaverton, Oregon 97006; and Oregon Health Sciences University, Portland, Oregon 97201-3098

	Abstract

Top Abstract I. Introduction II. GnRHR Structure Analysis III. Studies of GnRH... IV. Studies of GnRH... V. Studies of GnRH... VI. Future Directions References

 

I. Introduction

II. GnRHR Structure Analysis

III. Studies of GnRH Action in T3-1 Cells

A. Derivation of T3-1 cells

B. Characterization of T3-1 cells

C. GnRH binding

D. GnRHR regulation

1. Homologous regulation by GnRH

2. Regulation by gonadal steroid hormones

3. Regulation by gonadal peptides

4. Regulation by second messenger activators

E. Intracellular second messengers

1. G protein coupling

2. Inositol phosphates

3. Intracellular calcium

4. Protein kinase C

5. cAMP

6. Mitogen-activated protein kinases

F. -Subunit gene expression

1. Cell-specific expression

2. GnRH-stimulated expression

G. Desensitization

H. Summary of GnRH action in T3-1 cells

IV. Studies of GnRH Action in GH₃ Cells Transfected with the GnRH Receptor (GGH₃ Cells)

A. Derivation of GH₃ cells

B. Characterization of GH₃ cells

C. Derivation of GH₃ cells transfected with the GnRHR (GGH₃ cells)

D. GnRH binding

E. GnRHR regulation

F. Intracellular second messengers

1. G protein coupling

2. Inositol phosphates

3. cAMP

G. Regulation of secretion

1. PRL

2. LH and FSH (in GH₃ cells transfected with the -, LHß-, and FSHß-subunit genes)

3. Secretogranin-II

H. Regulation of PRL mRNA

I. Regulation of expression of transiently expressed reporter genes

1. PRL vs. -subunit gene

2. -, LHß-, and FSHß-subunit genes

J. Summary of GnRH action in GGH₃ cells

V. Studies of GnRH Action in Other Pituitary Cell Lines

A. RC-4B/C cells

B. LßT2 cells

VI. Future Directions

	I. Introduction

Top Abstract I. Introduction II. GnRHR Structure Analysis III. Studies of GnRH... IV. Studies of GnRH... V. Studies of GnRH... VI. Future Directions References

 
THE regulation of normal mammalian sexual maturation and reproductive function requires the integration and precise orchestration of hormonal regulation at the hypothalamic, pituitary, and gonadal levels. GnRH is a decapeptide synthesized in neurosecretory cells in the preoptic area of the hypothalamus. GnRH is secreted into the hypophysial portal circulation and is transported to the anterior pituitary gland, where it binds to receptors on a specific pituitary cell type, the gonadotrope, to modulate the synthesis and secretion of the gonadotropins, LH and FSH. Gonadotropins, in turn, are secreted into the systemic circulation and act on the gonads to regulate steroidogenesis and gametogenesis. LH stimulates ovulation and corpus luteum formation in females and androgen secretion in males; FSH stimulates the growth and maturation of ovarian follicles in females and spermatogenesis in males. Gonadal steroids and peptides, in turn, are secreted into the systemic circulation and act to modulate hypothalamic and pituitary function in both positive and negative feedback loops (1, 2).

Research into the neuroendocrine control of reproductive function by GnRH has undergone an explosion in the past 25 yr, marked first by the isolation and chemical characterization of GnRH (3, 4, 5). This led to the development of both agonist and antagonist analogs, resulting in rapid advances in our basic understanding as well as clinical applications to the treatment of disorders such as prostate cancer, endometriosis, precocious puberty, and infertility (6, 7). More recently, the molecular cloning of cDNAs encoding receptors for GnRH (GnRHR)² was achieved, first in mouse (8, 9) and subsequently in human, rat, cow, and sheep (10, 11, 12, 13, 14, 15, 16, 17). The availability of the GnRHR cDNA has allowed studies leading to further understanding of the mechanisms of GnRH action.

Primary anterior pituitary cells are comprised of a heterogeneous population of well differentiated, secretory cell types. These include somatotropes, which secrete GH; lactotropes, which secrete PRL; corticotropes, which secrete ACTH as well as other hormones derived from the peptide precursor, POMC, including MSH, lipotropins, endorphins, and enkephalin; thyrotropes, which secrete TSH; and gonadotropes, which secrete LH and/or FSH (18, 19). Several anterior pituitary cell types produce more than one of the anterior pituitary hormones; for example, LH and FSH are often colocalized to the same cell, as are GH and PRL. More recently, there has been evidence of colocalization of GH with LH or FSH (20) .

A major hindrance to progress in our understanding of the mechanisms of neuroendocrine control of reproduction at the hypothalamo-pituitary level is the lack of an ideal cell model for these studies. Historically, such studies have been performed in vivo in a variety of animal models and in vitro in dispersed primary pituitary cell cultures. These studies are limited by the heterogeneity of anterior pituitary cell types; gonadotropes make up only 6–15% of anterior pituitary secretory cells in adult animals (21). In addition, anterior pituitary cells cannot be propagated in culture systems, thus limiting the feasibility of many studies. Recently, a number of immortalized pituitary cell lines have been used as models for studies of the mechanisms of action of GnRH and its receptor.

Several aspects of the GnRHR and its signaling properties have been reviewed previously (22, 23, 24, 25, 26, 27, 28, 29, 30). Past reviews have focused on the molecular mechanisms of action of GnRH and the signaling properties of the GnRHR in primary pituitary cells. In this review, we will focus on studies of GnRH action using GnRHR-expressing pituitary cell lines as model systems. The results of these studies will be compared with what is known about GnRH signaling in primary pituitary cells. In addition, we will focus on the role of the GnRHR pathway in the regulation of gene expression.

	II. GnRHR Structure Analysis

Top Abstract I. Introduction II. GnRHR Structure Analysis III. Studies of GnRH... IV. Studies of GnRH... V. Studies of GnRH... VI. Future Directions References

 
The GnRHR cDNA encodes a 327- to 328-amino acid protein with seven putative membrane-spanning domains, characteristic of the family of G protein-coupled receptors (Fig. 1) (31). Interestingly, the GnRHR lacks the typical intracellular carboxyl terminus, making it one of the smallest receptors with the seven-transmembrane segment motif. The lack of a carboxyl-terminal tail domain is a unique feature of the GnRHR among G protein-coupled receptors.

View larger version (50K): [in this window] [in a new window]   Figure 1. Model of the rat GnRHR. Amino acid residues in black represent nonconserved amino acids between the rat and mouse GnRHR; shaded amino acid residues are nonidentical but conserved between the two species. Asterisks denote potential glycosylation sites. Potential phosphorylation sites are indicated for protein kinase C (arrowheads), casein kinase II (arrow), and protein kinase A (cross). [Reprinted with permission from U. B. Kaiser et al: Biochem Biophys Res Commun 189:1645–1652, 1992 (13) (Fig. 2A).]

Cloning of the mouse and human GnRHR genes reveals the presence of two introns (Fig. 2) (33, 34). The introns in the mouse gene occur in the sequences encoding the fourth transmembrane helix and the third intracellular loop. The human gene has the same structure, with the introns interrupting the coding sequences at the same locations, although the introns appear to vary in size. Both the human and the mouse appear to have only a single GnRHR gene, as determined by Southern blot analysis. Analysis of multiple cDNA clones obtained from T3-1 cells revealed the presence of at least four alternative transcripts, derived largely by alternate splicing (34). It is possible that these alternative transcripts account for some of the additional bands seen on Northern blot analysis. However, these alternative transcripts are less abundant than the original cDNA clone and appear to encode nonfunctional, truncated GnRHRs.

View larger version (13K): [in this window] [in a new window]   Figure 2. Schematic representation of the human GnRHR gene. A, Exon-intron localization. The shaded boxes indicate exons and the intervening lines indicate introns. B, The structure of the human GnRHR cDNA. The open box indicates the protein-coding regions, and hatched boxes are the putative transmembrane domains. [Reprinted from Mol Cell Endocrinol 103:R1-R6, (Fig. 1, C and D), N. C. Fan et al., "The human gonadotropin-releasing hormone (GnRH) receptor gene: cloning, genomic organization and chromosomal assignment" 1994 (33) with kind permission from Elsevier Science Ireland Ltd., Bay 15K, Shannon Industrial Estate, Co. Clare, Ireland.]

The 5'-flanking region of the human GnRHR gene has also been cloned and sequenced (37). Five consensus TATA boxes were identified, distributed within a 700-nucleotide region, and multiple transcriptional start sites were detected associated with these TATA sequences. These transcriptional start sites reside further upstream than the major transcriptional start site identified in the mouse, although the mouse 5'-flanking sequence also reveals several putative TATA boxes. These findings raise the possibility of species-specific or tissue-specific transcription initiation sites. The 3'-end of the human GnRHR gene has also been sequenced, revealing five classical polyadenylation signals (37). The large 3'-untranslated sequence likely accounts for the greatest portion of the major mRNA species observed by Northern blot analysis.

	III. Studies of GnRH Action in T3-1 Cells

Top Abstract I. Introduction II. GnRHR Structure Analysis III. Studies of GnRH... IV. Studies of GnRH... V. Studies of GnRH... VI. Future Directions References

 
A. Derivation of T3-1 cells
A fusion gene containing 1.8 kb of 5'-flanking sequences of the human glycoprotein hormone -subunit gene linked to the protein-coding sequences of the simian virus-40 (SV-40) T antigen oncogene was used to generate transgenic mice. Mice carrying this fusion gene developed tumors of the anterior pituitary. The T3-1 cell line was derived from a pituitary tumor in such a mouse. Cells from this tumor were dispersed and maintained in monolayer culture. Stable cultures were established, and monoclonal cell lines were derived and characterized (38). These cells have provided a continuous cell model system for the study of the GnRHR and GnRH action, as well as for cell-specific expression of the -subunit; indeed, the availability of T3-1 cells was critical for the molecular cloning of cDNAs encoding the GnRHR (8, 9).

B. Characterization of T3-1 cells
T3-1 cells express -subunit mRNA. In addition, -subunit protein is synthesized and secreted by these cells. The cells do not express TSHß, GH, PRL, or POMC genes, the hormones expressed in other, nongonadotrope anterior pituitary cell types. However, neither LHß nor FSHß subunit mRNA, expressed in primary pituitary gonadotropes, is expressed in the T3-1 cells. The cells respond to GnRH with an increase in -subunit mRNA levels, whereas levels remain unchanged after exposure to TRH. The GnRH response is time- and dose-dependent and blocked by a GnRH antagonist, consistent with action through the GnRHR (38). Furthermore, GnRH binding and expression of GnRHR mRNA in T3-1 cells have been shown (39). T3-1 cells also bind activin A and express mRNAs for the activin receptor types I, II, and IIB, as well as for the inhibin ßB-subunit (40, 41). The expression of the gonadotropin -subunit and GnRHR in T3-1 cells is consistent with their derivation from the gonadotrope lineage; however, they fail to express the full complement of gonadotrope-specific proteins, specifically the LHß and FSHß subunits. This suggests that T3-1 cells are derived from precursor cells that were not fully differentiated into gonadotropes. This is supported by observations that -subunit expression occurs early in ontogeny before LHß or FSHß (42, 43). The presence of GnRH responsiveness indicates that these cells likely arose after the expression of GnRHR; GnRH-binding sites have been reported to appear, albeit at very low levels, several days earlier in development than the ß-subunits (44).

C. GnRH binding
Specific, high-affinity binding sites for GnRH have been identified in T3-1 cell membrane preparations (39). A GnRH analog binds to these sites with a dissociation constant of 0.50 nM, similar to that measured in normal mouse (0.51 nM) and rat (0.20 nM) anterior pituitary. The total number of binding sites for GnRH is 1.6 pmol/mg protein, about 5 times higher than in normal mouse (0.33 pmol/mg) and rat (0.31 pmol/mg) anterior pituitary (Table 1) (39). However, one must take into account that T3-1 cells represent a homogeneous cell population, in which all the cells express the GnRHR and bind the GnRH analog, whereas anterior pituitary cells are a heterogeneous cell population, in which only approximately 10% of the cells, the gonadotropes, express the GnRHR. Therefore, the estimated number of GnRH-binding sites on T3-1 cells is approximately 50% of the number on primary gonadotropes.

View this table: [in this window] [in a new window]   Table 1. Comparison of the GnRH receptor on mouse, T3-1, and rat anterior pituitary membrane homogenates

2. Regulation by gonadal steroid hormones. Estradiol has been shown to reduce GnRHR number in T3-1 cells, as determined by GnRH-binding studies, without significantly altering the dissociation constant (K_d) (52). This inhibitory effect of estradiol is dose- and time-dependent. A reduction in GnRHR number was measurable after 24 h of exposure to estradiol and was maximal after 4–5 days. The EC₅₀ of the estradiol effect was approximately 10^-11 M. In primary cultures of rat pituitary cells, estradiol can both increase (chronic exposure) and decrease (short-term exposure) GnRH binding (53, 54). In ovine pituitary cultures, estradiol increased GnRH binding by 10 h, and this increase was maintained up to 48 h (55). Thus, there appear to be some differences in the responses of T3-1 cells and primary gonadotropes to estradiol. These discrepancies may be attributable to differences between physiological cellular responses of T3-1 cells and primary gonadotropes; alternatively, the up-regulation of GnRHR number seen in primary cultures may occur indirectly, involving steroid hormone effects on cells other than gonadotropes.

3. Regulation by gonadal peptides. Activin A increases GnRHR mRNA levels in T3-1 cells in a time- and dose-dependent fashion, with maximal stimulation occurring after 24–48 h of exposure (40). This stimulation of GnRHR mRNA levels by activin A occurs at the transcriptional level, as indicated by nuclear run-off and transient transfection experiments. Furthermore, pretreatment of T3-1 cells with activin A is able to enhance GnRH-induced activation of the gonadotropin -subunit promoter, suggesting that activin A may have a functional role in modulating the responsiveness of the gonadotrope to GnRH by increasing the expression of the GnRHR. Follistatin is able to block the effects of activin on the GnRHR gene, possibly by binding to and inactivating activin. These data are consistent with data in primary pituitary cells, demonstrating stimulation of the synthetic rate of GnRHRs by activin A (56). In contrast, recent data demonstrated that activin A blocked the stimulatory effect of GnRH on -subunit promoter activity in T3-1 cells; whether this was a receptor or postreceptor effect was not determined (57) .

4. Regulation by second messenger activators. In an attempt to identify possible regulators of GnRHR, T3-1 cells were treated with the second messenger activators, phorbol myristic acid (PMA) and forskolin (50). These agents activate the signal transduction pathways of a multitude of potential effectors that might regulate GnRHR. PMA, a phorbol ester that activates protein kinase C (PKC), had no effects on GnRHR mRNA levels in T3-1 cells. However, forskolin, which activates adenylyl cyclase, leading to increases in intracellular cAMP levels and hence activation of protein kinase A (PKA), decreased GnRHR mRNA levels by up to 6-fold. This effect was maximal after 8 h, but was transient, with GnRHR mRNA levels returning to control levels by 24 h after treatment. Correlation with GnRH binding is not yet known. Thus, in T3-1 cells, factors that activate the PKA pathway may decrease GnRHR mRNA levels, whereas activation of the PKC pathway appears to have no effect. In contrast, activation of PKC appears to play a role in mediating up-regulation of the GnRHR by GnRH in primary rat pituitary cells (27, 58, 59).

E. Intracellular second messengers
Studies of signal transduction pathways activated by GnRH in T3-1 cells have included studies of G protein coupling, generation of inositol phosphates, stimulation of increases in intracellular calcium concentration, activation of PKC, generation of cAMP, and activation of mitogen-activated protein kinases. The majority of studies have observed the responses to a single pulse of GnRH or to continuous GnRH; the responses to pulsatile administration of GnRH have not yet been described.

1. G protein coupling. Activation of the GnRHR by GnRH has long been known to result in the activation of heterotrimeric GTP-binding (G) proteins. Therefore, when the GnRHR cDNA was cloned, it was no surprise to find that it encoded a protein predicted to be a member of the family of cell surface, seven-transmembrane domain, G protein-coupled receptors (31). Because GnRH actions are generally not affected by cholera or pertussis toxin, a novel G protein (G_p) was suggested to mediate receptor activation. Using an antibody to the common G_q/G₁₁ carboxy-terminal sequence, it has been shown that GnRH activation of phospholipase C (PLC) in T3-1 cells requires GnRHR coupling to G_q, G₁₁, or both (60). Sustained exposure of T3-1 cells to a GnRH agonist results in the specific down-regulation of cellular levels of both G_q and G₁₁ (Fig. 3) (61, 62, 63). This was attributable to enhanced proteolysis of the activated G proteins; there was no change in G_q or G₁₁ mRNA levels (64). Sustained activation of PKC with the phorbol ester, PMA, was unable to mimic the GnRH agonist-mediated down-regulation of G_q and G₁₁, and inhibition of PKC with the selective inhibitor chelerythrine did not prevent this effect of GnRH, suggesting that the down-regulation of the G protein -subunits is a direct result of activation of the G protein, and does not require activation of a downstream second messenger-activated protein kinase. Interestingly, the rate of decay of G_q/G₁₁ in the presence of GnRH agonist had two components: an initial rapid rate and a slower secondary phase. It is possible that the initial fast decay rate occurring upon receptor occupancy is reduced to a lower rate with desensitization of the receptor response; alternatively, the fast decay rate may be dependent on the fraction of the cellular G protein that becomes activated upon occupancy of the GnRHR, whereas the lower decay rate depends on the residual G protein pool. The down-regulation of G_q and G₁₁ may, in turn, be a component of the desensitization of the cellular response to GnRH upon sustained exposure to GnRH or to an agonist.

View larger version (25K): [in this window] [in a new window]   Figure 3. The turnover of Gq/G11 and Gi2 in control and LHRH-E-treated T3-1 cells. A, Autoradiograph of a pulse-chase experiment with [35S]methionine in T3-1 cells treated or not for various times with LHRH-E. The turnover of Gq/G11 was monitored in T3-1 cells in the presence (+) or absence (-) of LHRH-E (1 µM) as described. Immunoprecipitates using antiserum CQ (Gq/G11) were subjected to SDS-PAGE, and the resulting gel was exposed to a phosphor storage plate for 48 h. The indicated 35S-labeled band was not present in immunoprecipitations done with preimmune serum (data not shown). B, Quantitative analysis of the effect of LHRH-E on the turnover of Gq/G11. Data such as that presented in panel A were quantitated and are displayed as means ± SEM of four individual experiments. , Control; •, LHRH-E treated. C, LHRH-E treatment does not alter the turnover of Gi2. Samples such as those of panel A were immunoprecipitated with the anti-Gi2 antiserum, SG, and exposed to a phosphor storage plate; the images were analyzed as for panel B. Data represent the means ± SEM of three experiments. , Control; , LHRH-E treated. LHRH-E, des-Gly10-[D-Ala6] LH-releasing hormone ethylamide. (The term LHRH used in this figure is synonymous with GnRH used elsewhere.) [Reprinted with permission from B. H. Shah et al: Proc Natl Acad Sci USA 92:1886–1890, 1995 (64) (Fig. 3).].

q

11

View larger version (14K): [in this window] [in a new window]   Figure 4. Time course of total IP production in unstimulated T3-1 cells () or cells stimulated with either GnRH (GnRH; 10 µmol/liter (•) or the GnRH agonist, buserelin (10 nmol/liter; (). Monolayer cultures were incubated for the times indicated and total IPs were measured. Results are the mean ± SD of triplicate determinations in three separate experiments. [Reproduced by permission of The Journal of Endocrinology, Ltd. From L. Anderson et al: J Endocrinol 136:51–58, 1993 (61) (Fig. 1).].

View larger version (19K): [in this window] [in a new window]   Figure 5. The effect of GnRH (10-8 M, t = 118 s, n = 6, upper trace) alone or after pretreatment with a GnRH antagonist (10-6 M for 2 min, lower trace, n = 10) on [Ca2+]i. [Ca2+]i, Ionized intracellular calcium concentration. [Reprinted from Mol Cell Endocrinol 86:167–175, Fig. 1, L. Anderson et al., "Characterization of the gonadotropin-releasing hormone calcium response in single T3-1 pituitary gonadotroph cells" 1992 (66) with kind permission from Elsevier Science Ireland Ltd., Bay 15K, Shannon Industrial Estate, Co. Clare, Ireland.].

View larger version (30K): [in this window] [in a new window]   Figure 6. GnRH-induced oscillations of outward K+ current and [Ca2+]i. The K+ current is measured under voltage clamp conditions at -50 mV, and [Ca2+]i is measured simultaneously with 50 µM indo-1 in the pipette. GnRH (2 nM) is perfused in the bath during the period marked with a bar. The opening of K+ channels is strictly synchronous with [Ca2+]i elevations. I, Current; Ca2+, ionized calcium concentration. [Reprinted with permission from B. L. Hille et al: Recent Prog Horm Res 50:75–95, 1995 (25) (Fig. 3).]

5. cAMP. No significant change in cAMP levels could be detected in T3-1 cells after treatment with a GnRH agonist, even in the presence of a phosphodiesterase inhibitor to prevent the degradation of cAMP (39). This is in contrast to the rise in cAMP levels that has been observed in whole pituitaries (76). This difference may lie in the possible need for the presence of testosterone for this response; the GnRH-induced rise in cAMP levels was observed in intact male rats only (77). Others have not been able to detect significant changes in cAMP levels after GnRH treatment of primary gonadotropes (78).

6. Mitogen-activated protein kinases (MAPKs). MAPKs, also known as extracellular signal-related kinases (ERKs), are a family of serine/threonine protein kinases that are rapidly activated in response to a wide variety of stimuli (Fig. 7) (79, 80, 81, 82, 83). Several members of the MAPK family have been identified, including p42^mapk (ERK2) and p44^mapk (ERK1). Stimuli for their activation include growth factors, many of which have receptors with intrinsic protein tyrosine kinase activity. MAPKs are involved in transmitting extracellular growth and differentiation signals into the cell nucleus, resulting in an array of transcriptional and mitogenic effects. Recent evidence indicates that some G protein-coupled receptors can activate the MAPK family of enzymes and that MAPKs may also be involved in nonproliferative signaling cascades (84, 85, 86, 87). G protein-coupled receptors appear to activate MAPK through Ras-dependent and -independent pathways, and both G- and Gß-subunits appear to be variably involved. These findings have led several investigators to study the ability of the GnRHR to activate MAPK and the role of MAPK in mediating cellular effects of GnRH (88, 89, 90, 91, 92, 93).

View larger version (30K): [in this window] [in a new window]   Figure 7. ERK1/ERK2 MAPK pathway. A schematic illustration of the MAPK pathway. RPTK, receptor protein tyrosine kinase; ERK, extracellular signal-related kinase; MEK, mitogen activated protein kinase (MAPK)/ERK; PKA, protein kinase A; cPLA2, cytosolic phospholipase A2; PP2A, protein phosphatase 2A; G, G protein; PTP, protein tyrosine phosphatase; RSK, ribosomal S6 kinase; TF, TCF, ELK1, transcription factors; MKP1, PAC1, protein phosphatases. [Reprinted with permission from T. Hunter: Cell 80:225–236, 1995 (79) (Fig. 1). © 1995 by Cell Press].

F. -Subunit gene expression
1. Cell-specific expression. T3-1 cells have proven to be a useful cell model for the isolation and characterization of transcription factors that appear to be involved in mediating gonadotrope-specific expression of the -subunit gene (Fig. 8). Some of these factors may be involved in mediating stimulation of -subunit gene expression by GnRH as well. However, because these factors appear to be more important for basal or tissue-specific -subunit gene expression rather than GnRH-stimulated expression, they will be mentioned only briefly here.

View larger version (8K): [in this window] [in a new window]   Figure 8. cis-Acting elements and transcription factors important for cell-specific and regulated expression of the glycoprotein hormone -subunit gene that have been characterized in T3-1 cells. GnRH-RE, GnRH response element; PGBE, pituitary glycoprotein hormone basal element; GSE, gonadotrope-specific element; EB2, E-box; LIM, LIM homeodomain protein; SF-1, steroidogenic factor 1; ßHLH, basic helix-loop-helix protein.

An additional putative basal enhancer, referred to as the pituitary glycoprotein hormone basal element (PGBE), has been identified at -344/-300 of the mouse -subunit gene (104). The PGBE is able to direct expression of the -subunit promoter to cells of both gonadotrope and thyrotrope lineages, but not to placenta. A member of the LIM (lin-11, isl-1, mec-3)-homeodomain family of transcription factors, LH-2, binds to a 14-bp imperfect palindrome within the PGBE domain in vitro (105). This element and factor are discussed further below.

Other elements that have been identified to play a role in expression of the -subunit gene in T3-1 cells include a GATA element, bound by GATA-binding proteins (106), and two E boxes, which bind members of the family of basic-helix-loop-helix-zipper proteins (107). The optimum level of -subunit gene expression in gonadotropes is probably determined by the combined actions of widely expressed, pituitary-restricted, and gonadotrope-specific transcriptional activators that act in combination and synergistically.

2. GnRH-stimulated expression. Although a number of factors that may be necessary for maintenance of basal levels of gonadotrope-specific gene expression have been identified in T3-1 cells, the identification of mechanisms for GnRH-stimulated expression have been less forthcoming. Windle et al. (38) have demonstrated that T3-1 cells respond to GnRH by elevating -subunit gene expression. A similar increase of -subunit mRNA levels was observed in response to PMA, and this increase was not additive with GnRH, suggesting that PKC may play a role in transducing the GnRH signal to the nucleus (39). The calcium ionophore, ionomycin, also stimulates -subunit mRNA levels. In contrast, an inhibitor of cAMP-dependent protein kinase did not affect the ability of GnRH or PMA to stimulate expression of an -subunit promoter/luciferase reporter gene (LUC), indicating that cAMP-dependent protein kinase is not required for transcriptional activation by GnRH (104).

The increase in -subunit mRNA levels in response to GnRH was maximal at 12–24 h and maintained for a further 24 h (Fig. 9) (108). The observed increase in mRNA levels appears to be mediated by both an increase in -subunit gene transcription and mRNA stability. Nuclear run-off assays demonstrated an increase in -subunit gene transcription of 2- to 3-fold within 1 h after exposure to GnRH but returned to baseline by 12 h. GnRH also stimulated the activity of LUC, apparent after 1 h, maximal after 4–6 h, but back to baseline by 24 h of GnRH treatment (Fig. 9). Thus, GnRH appears to stimulate a burst of -subunit gene transcription lasting less than 4–6 h. The persistent elevation of -subunit mRNA levels for at least 48 h suggests that the mRNA has a long half-life and/or that GnRH stabilizes the mRNA in addition to its transcriptional effects. Indeed, pulse-chase experiments showed that the half-life of the -subunit mRNA increased from 1.2 h in the absence of GnRH to 8 h in the presence of GnRH in T3-1 cells. Whether this mechanism also occurs in primary gonadotropes is unclear, as the half-life of -subunit mRNA in primary pituitary cultures is 6.5 h; however, in this case both gonadotropes and thyrotropes contribute to -subunit mRNA levels (109). Interestingly, while the stimulatory effects of GnRH on -subunit gene transcription and mRNA levels were evident very rapidly, within 1 h after exposure to GnRH, GnRH-induced -subunit release was detected only after a lag of 4 h of incubation (110). Thus, there appears to be dissociation between the stimulation of gene expression and exocytosis.

View larger version (28K): [in this window] [in a new window]   Figure 9. Effect of GnRH on LUC expression and -subunit mRNA levels. T3 cells stably transfected with LUC were incubated in the absence or presence of GnRH (10-7 M) for the indicated periods of time. Cells were harvested and assayed for luciferase activity. Luciferase activity (LUC) is expressed in arbitrary light units (ALU) and is the mean ± SEM of triplicate plates of cells. Basal expression of LUC was 415,000 ALU. Background luciferase activity was below 120 ALU. Total RNA (5 µg) from triplicate plates of T3 cells treated in the absence or presence of GnRH was analyzed by Northern blot for -subunit and GAPDH mRNAs. mRNA levels were quantitated using scanning densitometry, and -subunit mRNA levels were corrected for hybridization to GAPDH mRNA. The mean ± SEM of three separate experiments are expressed relative to the basal -subunit mRNA level in the absence of GnRH. [Reprinted with permission from P. J. Chedrese et al: Endocrinology 134:2475–2481, 1994 (108) (Fig. 1). © The Endocrine Society.]

View larger version (27K): [in this window] [in a new window]   Figure 10. Multimers of the -416 to -385 region function as a GnRH-responsive element. A, Synthetic DNA elements were prepared that included the sequences that were shown to be important by mutation analysis. The sequence of the mouse -subunit gene, which was used as a synthetic DNA element, is aligned with the corresponding region of the human and pig -subunit genes. Positions in which the human or pig sequence are identical to the mouse sequence are indicated by uppercase letters. The locations of the -subunit sequences where mutations reduced GnRH and phorbol responses are indicated by overbars. B, To assess the functional properties of these elements, multimers of the synthetic DNA elements were placed upstream of a minimal promoter, which was linked to luciferase, and the reporter genes were transfected into T3-1 cells. Cells were treated with vehicle alone, 10-5 M buserelin (GnRHa), 10-7 M phorbol myristic acid (PMA), or 0.5 mM 8-(4-chlorophenylthio)cAMP (cAMP) 18 h after transfection. Cells were collected 24 h after transfection (6 h after treatment), and luciferase activity was determined. All values are means ± SE from two to four separate experiments; each experiment included three transfections for each DNA construct. The luciferase data were normalized for transfection efficiency between experiments. Responses to different agents are indicated as the ratio of luciferase activity in the treated cells to that in vehicle-treated cells. A schematic representation of the organization of each of the constructs is shown at the left. The -416 to -385 element is indicated by a black arrow; the -344 to -300 element is indicated by a white arrow; the minimal promoter sequences are indicated by gray shading. [Reprinted with permission from W. E. Schoderbek et al: J Biol Chem 268:3903–3910, 1993 (104) (Fig. 7).].

Another candidate factor for a role in mediating -subunit gene expression by binding to the PGBE is mLim-3, a related member of the family of LIM-homeodomain proteins. mLim-3, also known as P-Lim or Lhx3, is a mouse gene expressed in the pituitary throughout development and in the adult, as well as transiently in the spinal cord, pons, and medulla oblongata, but with no detectable expression elsewhere. mLim-3 expression was detected in cell lines of pituitary origin, including cells representative of somato-lactotropes (GH₃, GH₄C1, GC), thyrotropes (TSH), gonadotropes (T3), and corticotropes (AtT-20), but not in cell lines derived from peripheral, other endocrine, or neural tissues (112, 113). mLim-3 is able to bind to the PGBE sequence in vitro and is a strong activator of transcriptional activity of the -subunit promoter, as well as the PRL, TSHß, and Pit-1 promoters (112). Interestingly, it was recently reported that targeted disruption of the mLim-3 gene in mice leads to failure of growth and differentiation of the anterior and intermediate lobes of the pituitary (114). The development of all pituitary cell lineages, except the corticotropes, was affected. This suggests that mLim-3 plays an important role not only in -subunit gene expression, but in differentiation and proliferation of nearly all the pituitary cell lineages.

Further studies of the putative GnRH-RE in the mouse -subunit promoter have identified a core Ets factor (a family of transcription factors that have been implicated in mediating transcriptional responses to MAPK activation) binding site within the GnRH-RE, which appears to be important in mediating GnRH stimulation of -subunit gene expression (Fig. 8) (92). Recent evidence that GnRH activates the MAPK signal transduction pathway, as discussed above, is relevant in terms of the mechanisms of transcriptional stimulation of the -subunit gene by GnRH. Activation of the MAPK cascade by a constitutively active form of Raf kinase in T3-1 cells leads to stimulation of the -subunit promoter. Furthermore, inhibition of MAPK activity by kinase-defective ERK1 or ERK2, or overexpression of MAPK phosphatase 2, which dephosphorylates and inactivates MAPK, leads to the attenuation of GnRH-induced activation of the -subunit promoter. The DNA-binding domain of Ets-2 was able to bind specifically to a site within the GnRH-RE, and a dominant negative Ets-2 expression vector reduced the ability of GnRH to stimulate expression of LUC. These findings suggest that the Ets factor-binding site in the GnRH-RE may contribute to transcriptional stimulation of the -subunit gene by GnRH, via activation of the MAPK pathway. In contrast, however, Sundaresan et al. (91) found that dominant negative mutant forms of Ras, ERK1, and ERK2 reduced basal expression of a human LUC but had no effect on GnRH-stimulated expression. The reasons for the differences between these two studies are not clear, although Roberson et al. (92) used the mouse -subunit promoter, whereas Sundaresan et al. used the human gene.

In addition to the studies characterizing GnRH-responsive DNA sequences in the mouse -subunit gene using T3-1 cells as described above, a GnRH-responsive region in the human gene was identified by transfection analyses in primary rat pituitary cell cultures (115). Deletion analyses suggested that one or more GnRH-responsive sequences reside between -346 and -244 in the human -subunit promoter. This GnRH-responsive region does not include the GnRH-RE defined in the mouse -subunit promoter. In contrast to the findings with the mouse -subunit gene in T3-1 cells, the regions of the human -subunit gene that are important for the GnRH response appear to be distinct from those required for basal activity. Basal expression appeared to be primarily mediated through the proximal promoter and cAMP-responsive regions. These differences may reflect different mechanisms of GnRH stimulation of the human vs. the mouse -subunit gene or differences in the mechanisms of regulation in T3-1 cells vs. primary pituitary gonadotropes.

G. Desensitization
GnRH is secreted from the hypothalamus in a pulsatile fashion, and pulsatile GnRH stimulates LH and FSH biosynthesis and secretion (116). In contrast to the stimulatory effects of pulsatile GnRH, sustained exposure to high concentrations of GnRH reduces the response of gonadotropes to subsequent stimulation with GnRH (homologous desensitization), leading to suppression of gonadotropin secretion (117). This homologous desensitization to GnRH can occur rapidly, within the time frame of endogenous GnRH pulses (118). The mechanism of this desensitization is not known, and both receptor (119) and postreceptor (120, 121) mechanisms have been proposed. For a number of other G protein-coupled receptors, early desensitization events are thought to involve the uncoupling of the receptor from its regulatory G protein, with loss of downstream-signaling events (122). Rapid desensitization appears to involve phosphorylation by specific intracellular kinases of the third intracellular loop or the C-terminal tail (123, 124). However, the GnRHR lacks the C-terminal cytoplasmic tail as well as the third intracellular loop sequences implicated in the desensitization of other receptors (31).

T3-1 cells have been used as a model for the study of mechanisms of desensitization to GnRH. Stimulation of LUC activity in transfected T3-1 cells was maximal 4–6 h after exposure to GnRH but thereafter declined, returning to levels in unstimulated control cells by 12–24 h. LUC activity was also stimulated by a PKC activator, PMA, a calcium channel agonist, BAY K 8644, and an activator of the PKA pathway, 8-bromo-cAMP. Maximal responses to these agents also occurred after 4–6 h of exposure, although the maximal levels of activity were less than those observed in response to GnRH. A decline in LUC activity over time with continuous exposure to these agents was particularly marked for PMA, but was also seen with BAY K 8644, whereas stimulation by 8-bromo-cAMP was maintained for at least 24 h. Pretreatment of T3-1 cells with GnRH blocked subsequent stimulation of LUC activity by either GnRH or PMA. In contrast, both 8-bromo-cAMP and BAY K 8644 were still able to stimulate LUC activity after pretreatment with GnRH. These data suggest that the transcriptional stimulation of the -subunit gene by GnRH is mediated by the PKC pathway, and that this pathway can be desensitized in T3-1 cells by continuous exposure to GnRH. The kinetics of desensitization are difficult to infer from these studies; exposure to GnRH may incite a short burst of transcriptional activity of the -subunit promoter, which then leads to a more gradual accumulation of the luciferase product. However, the addition of the GnRH antagonist, Antide, after treatment of the cells with GnRH resulted in a reduction of luciferase activity compared with exposure to GnRH alone, even when Antide was added up to 6 h after GnRH, indicating that some stimulation of the -subunit promoter by GnRH was still occurring, i.e. the cells were not fully desensitized to GnRH. Continuous exposure of primary pituitary cells to GnRH causes rapid desensitization at the secretory level for free -subunit as well as intact LH and FSH, evident within 15 min (125). The differences in kinetics for transcriptional and secretory desensitization may reflect different cellular mechanisms or differences between the T3-1 cell line and primary gonadotropes.

Regulation of -subunit gene transcription is a relatively downstream endpoint for the study of homologous GnRH desensitization. Measurements of second messengers may lead to insights into early or short-term desensitization events. GnRH treatment led to a linear increase in total IP production in T3-1 cells over 0–15 min (126, 127, 128). Furthermore, GnRH pretreatment for 5 min did not alter subsequent stimulation of IP₃ production by GnRH 15 min later. These data indicate a lack of desensitization of the rapid GnRH-induced IP₃ response in T3-1 cells. Pretreatment with GnRH for 1 h did reduce subsequent cellular IP accumulation in response to GnRH, but this may be attributable to a reduction in GnRHR numbers. GnRH pretreatment of T3-1 cells for short times (5–15 min) had no effect on GnRHR number; however, treatment for 1 h with 10^-7 M GnRH reduced GnRHR number by 48%. The affinity for GnRH was not altered. Desensitization of both the extracellular Ca²⁺-dependent and -independent phases of the Ca²⁺ response to GnRH were observed after pretreatment with 10^-7 M GnRH for 1 h (128). Thus, one mechanism of intermediate desensitization to GnRH may be receptor loss.

However, this does not account for rapid or early desensitization or the degree of desensitization of the Ca²⁺ response. An additional uncoupling event may occur during the pretreatment, which reduces the ability of the agonist-occupied GnRHR to elevate intracellular Ca²⁺. Treatment of T3-1 cells with 5-min pulses of GnRH every 15 min resulted in desensitization of the Ca²⁺ response after the first pulse in a dose-dependent manner, being evident at GnRH concentrations greater than 2 x 10^-9 M (126). The mechanisms underlying this desensitization are not known but could include loss of IP₃ receptors, depletion of intracellular Ca²⁺ stores, and inactivation of Ca²⁺ channels, as has been suggested in studies of primary pituitary cells (120). The dissociation of IP production and calcium stimulation suggests that desensitization of GnRH-induced calcium mobilization is a postreceptor phenomenon occurring distal to PLC activation. The lack of the C-terminal cytoplasmic tail, implicated in the desensitization of other G protein-coupled receptors, in the GnRHR therefore appears to correlate with a lack of receptor desensitization; rather, desensitization to GnRH appears to be primarily a postreceptor phenomenon. Alternatively, T3-1 cells may be lacking a factor(s) necessary for mediating rapid receptor desensitization in primary gonadotropes.

H. Summary of GnRH action in T3-1 cells
The development of the T3-1 gonadotropic cell line has enabled significant advances in our understanding of gonadotrope function and gonadotropin regulation, particularly in the areas of -subunit gene expression and GnRHR structure and function. T3-1 cells were critical for the initial cloning of GnRHR cDNAs as well as for elucidation of the GnRHR gene structure, confirming previous findings in primary pituitary cells which suggested that the GnRHR was a member of the G protein-coupled receptor family. The absence of a carboxy-terminal intracellular tail on the receptor was a surprising finding, which makes questions about the mechanisms of gonadotrope desensitization to GnRH all the more intriguing.

T3-1 cells have been used to elucidate a number of components of the GnRH signal transduction pathway (Fig. 11). The GnRHR in T3-1 cells is coupled to G proteins of the G_q/G₁₁ family, leading to production of IPs and increases in intracellular calcium levels, which, in turn, leads to activation of PKC. While cAMP has, in some studies, been suggested to be activated by GnRH, and has been shown to lead to increases in expression of the -subunit gene, there is no evidence for increases in cAMP levels in response to GnRH in T3-1 cells. Furthermore, there is now evidence that the MAPK pathway is activated by GnRH in T3-1 cells and may be important in the stimulation of -subunit gene expression by GnRH.

View larger version (25K): [in this window] [in a new window]   Figure 11. Summary of known GnRH actions on -subunit gene expression in T3-1 cells. GnRH binds to the seven-transmembrane domain GnRHR, which is coupled to Gq/G11. Activation of Gq/G11 activates phospholipase C, which stimulates the production of inositol triphosphate and an increase in [Ca2+]i, leading to activation of PKC. PKC, in turn, leads to stimulation of -subunit gene expression, either directly, or indirectly by activating the MAPK cascade. GnRHR may also be coupled to Gs, leading to activation of adenylyl cyclase and stimulation of cAMP production, which may also influence -subunit gene expression. Third, activation of the GnRHR may also activate the MAPK cascade via Gi.

	IV. Studies of GnRH Action in GH3 Cells Transfected with the GnRH Receptor (GGH3 Cells)

Top Abstract I. Introduction II. GnRHR Structure Analysis III. Studies of GnRH... IV. Studies of GnRH... V. Studies of GnRH... VI. Future Directions References

 
A. Derivation of GH₃ cells
The GH₃ cell is a well characterized pituitary cell strain established from a GH-producing rat pituitary tumor, MtT/W5, that was propagated as a transplantable rat pituitary tumor. By a method of alternate culture and animal passage, several clonal strains of epithelial cells were established (129, 130).

B. Characterization of GH₃ cells
These cells are somatolactotropic in origin. They secrete large amounts of GH into culture medium and stimulate body weight gain and growth after injection into normal or hypophysectomized rats (129, 130). They express PRL and GH genes and also secrete PRL and GH in a regulated fashion. GH₃ cells express TRH receptors (TRHR) and respond to TRH with an increase in PRL biosynthesis and secretion, and a reduction in GH production (131, 132). GH₃ cells do not express -subunit, TSHß, LHß, FSHß, and POMC genes, hormones expressed in other, nonsomatolactotropic anterior pituitary cell types. However, they are capable of supporting the expression of exogenous - and TSHß-subunit genes, introduced into the cells by transient transfection (133, 134, 135, 136, 137, 138). In addition to TRHR, GH₃ cells al