The Equine Luteinizing Hormone -Subunit Promoter Contains Two ...

8 downloads 1095 Views 4MB Size Report
University of Kansas Medical Center ... genesis in males, and directs gonadal steroidogenesis ...... Wolfe MW, Call GB 1999 Early growth response protein.
The Equine Luteinizing Hormone b-Subunit Promoter Contains Two Functional Steroidogenic Factor-1 Response Elements

Michael W. Wolfe Department of Molecular and Integrative Physiology University of Kansas Medical Center Kansas City, Kansas 66160-7401

The requirements for basal expression of the LH b-subunit promoter in pituitary gonadotropes are largely unknown. We have used the equine (e) LHb subunit promoter as a model to unravel the combinatorial code required for gonadotrope expression. Through the use of 5*-deletion mutagenesis, a region between 2185 and 2100 of the eLHb promoter was shown to play a critical role in maintaining basal promoter activity in aT3–1 and LbT2 cells. This region encompasses the steroidogenic factor-1 (SF-1) binding site that has been reported to have a functional role in expression of the LHb promoter in other species. We have also identified an additional SF-1 site at 255 to 248. Binding of SF-1 to both sites was confirmed by electrophoretic mobility shift assays. Mutations within these sites, either individually or in combination, did not attenuate basal activity of the eLHb promoter in aT3–1 cells, but did diminish promoter activity in LbT2 cells. Interestingly, cotransfection with an expression vector encoding SF-1 induced eLHb promoter activity, and this induction was abrogated by mutations within the SF-1 sites in aT3–1 cells. Block replacement mutagenesis was performed on the 2185/2100 region of the eLHb promoter to identify DNA response elements responsible for maintaining basal promoter activity. From this analysis, two regions emerged as being important: a distal 31-bp segment (2181 to 2150) and an element located immediately 3* to the distal SF-1 site (2119 to 2106). It is hypothesized that these two regions as well as the SF-1 sites represent regulatory elements that contribute to a combinatorial code involved in targeting expression of the eLHb promoter to gonadotropes. (Molecular Endocrinology 13: 1497–1510, 1999)

0888-8809/99/$3.00/0 Molecular Endocrinology Copyright © 1999 by The Endocrine Society

INTRODUCTION LH and CG are heterodimeric glycoprotein hormones composed of a common a-subunit noncovalently linked to a unique b-subunit. LH is synthesized in the pituitary of all mammals (1). This glycoprotein hormone stimulates ovulation in females, promotes spermatogenesis in males, and directs gonadal steroidogenesis in both sexes. In contrast, synthesis of CG occurs only in placenta of primates and equids (2). In primates, and presumably equids, CG maintains function of the corpus luteum during the early stages of pregnancy, which in turn sustains pregnancy. A single gene encodes the a-subunit in all mammals studied to date (1). Thus, synthesis of LH and CG requires that expression of the a-subunit gene occurs in two locations: gonadotropes of the pituitary and trophoblasts of the placenta (1, 3–5). Placenta-specific expression is achieved through a compact and interactive array of regulatory elements located within the proximal 200 bp of the human a-subunit 59-flanking region (5, 6). Some of the transcription factors that interact with these elements have been identified and consist of ubiquitous proteins such as members of the cAMP-response element binding protein (CREB)/ activating transcription factor (ATF) family (4, 7, 8), the GATA family of DNA binding proteins (9), and a protein that appears to be unique to the placenta (trophoblastspecific element binding protein or TSEB; Refs. 9 and 10). In contrast, a different but overlapping set of regulatory elements appears to be required for pituitaryspecific expression of the a-subunit gene. For example, at least one of the cis-acting elements involved in placenta-specific expression (the GATA binding site) may also contribute to expression in pituitary gonadotropes (11, 12). Additional elements upstream of the 200-bp promoter-proximal region have also been identified. These include an element that binds a member of the orphan nuclear receptor family, steroidogenic factor-1 (SF-1; Ref. 13) and a site that binds a member of the LIM-homeodomain family of DNAbinding proteins (LH-2; Ref. 14). 1497

MOL ENDO · 1999 1498

Synthesis of LH and CG also requires expression of the hormone-defining b- subunit gene. Most mammals have a single LHb gene with transcription occurring only in gonadotropes of the pituitary. In primates, the single-copy LHb gene has undergone a series of gene duplications resulting in the formation of a linked array of multiple CGb genes (2). Although LHb and CGb genes maintain a homology of greater than 90%, transcription of CGb genes occurs only in placenta and initiates at a site 366 bp upstream from that used for transcription of the LHb gene (15, 16). Thus, different promoters and regulatory elements are responsible for the pituitary- and placenta-specific expression of primate LHb and CGb genes. As indicated from several transfection studies (17– 19), the human CGb promoter contains one element located between nucleotides 2305 and 2279 that appears important for both basal transcription and responsiveness to cAMP. Although less well defined, full responsiveness to cAMP requires at least one other element located between 2248/2210 (18). Interestingly, both of these elements can bind TSEB, the aforementioned placenta-specific protein that forms part of the regulatory code required for targeting expression of the a-subunit gene to the placenta (19). While the binding of TSEB may provide a mechanism for coordinating placenta-specific expression of the aand CGb-subunit genes, functional studies that test this possibility are lacking. Resolution of regulatory elements required for pituitary-specific expression of mammalian LHb genes has been hampered by the lack of cell lines that actively express either the endogenous LHb gene or transfected LHb promoter-reporter genes. For reasons that still remain unresolved, transfection studies that employ primary cultures of pituitary cells have also been relatively uninformative (20). Due to these limitations, transgenic mice have become the model of choice for studying the LHb promoter. Data from transgenic studies suggest that elements required for gonadotrope-specific expression and responsiveness to GnRH and sex steroids reside within the first 800 bp of the LHb-promoter proximal region (21–23). It has been demonstrated both in vitro and in vivo that SF-1 can bind to and transactivate the rat and bovine LHb promoters (24, 25). In fact, mutation of the SF-1 site severely attenuated activity of the bovine promoter in transgenic mice (25). Recent reports indicate that SF-1 interacts with an immediate-early response gene product, early growth response protein 1 (Egr-1), and that these two transcription factors mediate GnRH regulation of the LHb gene (26–28). Thus, SF-1 appears to play an important role in regulating gonadotrope expression of both the a- and LHb-subunit genes. Characterization of the equine LH and CG b-subunit gene (29) has shown that, in contrast to primates, the b-subunits of eLH and eCG are encoded by the same single-copy gene. The single eLH/CGb transcript gives rise to proteins having identical amino acid se-

Vol 13 No. 9

quences (30). This protein is more like the human and primate CGb than LHb since it contains a carboxylterminal peptide unique to primate CGb genes. In contrast, initiation of transcription of the eLH/CGb gene occurs at the same nucleotide position in placenta and pituitary (29). The TATA-containing promoter responsible for this event is comparable to the primate LHb promoter. Therefore, the eLH/CGb gene shares features that are common to the primate and equid a-subunit gene in that they are single-copy genes, transcription is initiated through the use of a TATA-containing promoter, and both subunits are expressed in pituitary and placenta. Given the unique structural configuration of the eLH/ CGb gene, and the paucity of information regarding mechanisms regulating pituitary expression of mammalian LHb genes, activity of the eLHb promoter was evaluated in the aT3–1 (31) and LbT2 (32, 33) gonadotrope cell lines. Data reported herein identify two functional SF-1 sites and two additional proximal activating elements/regions. It is hypothesized that these DNA response elements contribute to a combinatorial code that directs LHb expression in pituitary gonadotropes.

RESULTS Cloning of Additional eLHb Sequence Expression of the LHb promoter in the pituitaries of transgenic mice has been reported in three studies that used 1.9 kb of ovine (21), 1.5 kb of rat (23), or 776 bp of bovine (22) 59-flanking sequence. The bovine construct was expressed exclusively in the pituitary and was regulated appropriately by GnRH, estrogen, and androgen (22). These data suggested that the cis-elements required for pituitary-specific expression of LHb are contained within 800 bp of the proximal 59-flanking region. Our original eLHb promoter clone extended only to 2448 and, hence, may not have contained important cis-acting elements necessary for gonadotrope expression. To generate additional 59flanking sequences, we screened an equine genomic

Fig. 1. Schematic Representation of eLHb Genomic Clones The l-clones were isolated as described in Materials and Methods. Shown are the three exons of the eLHb gene (black boxes) and HindIII restriction sites (H). Overall size of the clones is indicated on the lower line with 0 denoting the start site of transcription.

Basal Expression of Equine LHb

library and isolated two overlapping l-clones (Fig. 1). The promoter-proximal region of one the clones (eb5l) was sequenced to confirm that it contained an authentic eLHb 59-flanking region. This clone contains approximately 3 kb of 59-flanking sequence, the entire coding region, and approximately 12 kb of 39-flanking sequence. The eLHb Promoter Is Active in the aT3–1 Gonadotrope Cell Line Previous studies have demonstrated that the rat and bovine LHb promoters are inactive in the aT3–1 cell line (22–24). We examined transcriptional activity of 59-deletion mutants of the eLHb promoter to determine whether it was also inactive in aT3–1 cells. Although these cells lack endogenous expression of LHb, they have many of the characteristics exhibited by mature gonadotropes (31). Unlike previous reports for the rat and bovine LHb promoters, activity of all of the eLHb promoter constructs was greater than that of the promoterless control (pGL2 basic; Fig. 2A). Transcriptional activity of the 2185/160 promoter was 17-fold greater than the promoterless control. In contrast, these same constructs were inactive in the BeWo human choriocarcinoma cell line (1- to 3-fold over pGL2 basic; Fig. 2B). All of the eLHb promoter constructs had similar levels of basal activity except for the shortest construct (2100/160). Truncation of the promoter from 2185 to 2100 resulted in a 65% decrease in basal activity. Due to the elevated basal activity of the eLHb promoter in aT3–1 cells, we evaluated whether this was unique to the equine promoter. Similar regions of the bovine and mouse LHb promoters (based on sequence homology and alignment) were cloned by PCR

1499

and evaluated for basal activity in aT3–1 cells. Both the bovine and mouse LH promoters were inactive in the aT3–1 cells, corroborating previous data (Fig. 3 and Ref. 22). Furthermore, we also observed variation between species in regard to activity of the a-subunit promoter. Some of the differences in activity of the a-subunit promoters could be attributed to variation in promoter length, but not all. Thus, species differences appear to exist in the requirements for expression of both the a and LHb-subunit promoters in aT3–1 cells. The Equine LHb Promoter Contains Two SF-1 Binding Sites Since SF-1 has been shown to be an important regulator of LHb expression (24–28), it was important to determine whether a SF-1 site located within the 2185 to 2100 region of the eLHb promoter contributed to basal promoter activity. This site is homologous to the SF-1 response element identified in the LHb promoter of other species (Fig. 4, dSF-1). Furthermore, an additional more proximal SF-1-like sequence is also present in the LHb promoters of other species including the horse (Fig. 4, pSF-1). Electrophoretic mobility shift assays (EMSAs) were performed to determine whether these eLHb SF-1 sites were authentic SF-1 binding sites. Nuclei from aT3–1 cells were used as the source for SF-1 (34, 35). A single, predominant complex was detected with both the distal and proximal eLHb SF-1 probes (Fig. 5A, lanes 1 and 6). An antibody that recognizes an epitope within the DNAbinding domain of SF-1 was used to determine whether SF-1 was a component of this complex. Inclusion of the SF-1-specific antibody blocked the formation of the complex on both probes (Fig. 5A, lanes 2 and 7), indicating that SF-1 can bind to both sites. In

Fig. 2. The eLHb Promoter Is Active in the aT3–1 Cell Line, but Inactive in BeWo Cells The plasmid constructs indicated along the left side of the figure were transfected into aT3–1 (A) and BeWo (B) cells and evaluated for luciferase reporter gene activity as described in Materials and Methods. Luciferase activity was normalized to b-galactosidase levels. Values shown are the mean 6 SEM for triplicate wells from a minimum of three independent transfections.

MOL ENDO · 1999 1500

Vol 13 No. 9

(lanes 2 vs. 7). A similar pattern of competition was observed when the proximal SF-1 site was used as the labeled probe (data not shown). These data suggest that the distal site binds SF-1 with an affinity that is approximately 10-fold higher than that of the proximal site in the eLHb promoter. The Equine LHb SF-1 Sites Are Not Required for Basal Activity in aT3–1 Cells Fig. 3. Basal Activity of a and LHb Promoters Varies across Species in aT3–1 Cells Basal activity of the human (ha 2485/148), mouse (ma 2507142), and bovine (ba 2315/145) a-subunit promoters and equine (eLHb 2185/160), bovine (bLHb 2185/110) and mouse (mLHb 2196/18) LHb-subunit promoters were elevated in aT3–1 cells. Data are expressed as fold induction over a promoterless control luciferase vector (pGL2 basic). Values shown are the mean 6 SEM for triplicate wells from two independent transfections.

contrast, inclusion of nonimmune rabbit serum did not disrupt binding of the complex to the distal and proximal SF-1 sites (Fig. 5A, lanes 3 and 8). A similar scenario was observed when the SF-1 site within the human a-subunit promoter (haGSE) was used as labeled probe (Fig. 5A, lanes 11–13). Thus, SF-1 can bind to both the proximal and distal SF-1 sites. Additional confirmation of these data was provided by competitions with unlabeled SF-1 sites. Lanes 4 and 9 (Fig. 5A) represent competition of protein binding by 100-fold molar excess of homologous DNA, while lanes 5 and 10 depict competition with the human a-subunit promoter SF-1 site. The haGSE oligo competed for protein binding as well as did the homologous competitors. We also evaluated the ability of the eLHb SF-1 sites to compete for binding of SF-1 to the radiolabeled haGSE site (Fig. 5B). Both eLHb SF-1 sites competed for SF-1 binding to the haGSE oligo (lanes 15 and 16). The pSF-1 site was less effective at competing for SF-1 binding to the haGSE oligo as compared with the dSF-1 and haGSE oligos. These data suggested that the pSF-1 response element had a weaker affinity for SF-1 as compared with the dSF-1 and haGSE response elements and was supported by the binding data shown in Fig. 5A (decreased SF-1 complex formed with the labeled eLHb pSF-1 oligo when compared with that observed when the dSF-1 or haGSE oligos were used as probes; lanes 6 vs. 1 and 11). To further document the affinity differences between the distal and proximal SF-1 sites, a more thorough competition analysis was performed (Fig. 5C). Binding of SF-1 to the dSF-1 site was effectively blocked using as little as a 10-fold molar excess of homologous competitor (eb dSF1, lane 2) or the ha GSE site as competitor (lane 8), while a 10-fold molar excess of eb pSF1 was ineffective (lane 5) at competing for SF-1 binding. It required 10 times more eb pSF1 than eb dSF1 to achieve equivalent levels of competition

To address the functional significance of the eLHb SF-1 sites, transient transfections were performed in aT3–1 cells with the wild-type 2448/160 eLHb promoter linked to luciferase or constructs that harbored mutations in the SF-1 sites (single or double mutations). The distal and proximal SF-1 sites (TGACCTTG and TGGCCTTG, respectively) in the eLHb promoter were mutated to aGatCTTG. These mutations completely abolished SF-1 binding in an EMSA (data not shown). Mutation of the distal or proximal SF-1 sites, individually or in combination, had little impact on basal promoter activity (Fig. 6A). These data suggest that although SF-1 can bind to the eLHb promoter, SF-1 does not contribute to basal expression in aT3–1 cells. These data conflict with previous findings regarding SF-1 regulation of the bovine and rat LHb promoters (24, 25). The previous studies used an overexpression model to evaluate promoter regulation by SF-1. Therefore, we evaluated the ability of SF-1 to transactivate the eLHb promoter in an overexpression experiment. The constructs described above were cotransfected into aT3–1 cells along with an expression vector encoding the SF-1 cDNA. Overexpression of SF-1 increased activity of the 2448/160 eLHb promoter by 2.2-fold (Fig. 6B), while it had no effect on the Rous sarcoma virus (RSV) promoter. We also have observed a similar 2- to 3-fold induction by SF-1 of the 2185/ 110 bovine LHb promoter (data not shown). Mutation of both SF-1 sites in the eLHb promoter completely abrogated the ability of SF-1 to transactivate the promoter. Furthermore, the individual mutations reduced, but did not completely abrogate, SF-1 induction of eLHb promoter activity (1.4- and 1.5-fold for the distal and proximal mutants, respectively, vs. 2.2-fold for the wild-type construct). These data indicate that SF-1 can transactivate the eLHb promoter when levels of SF-1 are elevated in aT3–1 cells and that full transactivation requires both SF-1 sites. Two Regions Flanking the SF-1 Site Are Responsible for Basal Activity of the eLHb Promoter in aT3–1 Cells As shown earlier, sequences between 2185 and 2100 are needed for full promoter activity (Fig. 2A). Therefore, to further delineate the sequences required for basal activity of the eLHb promoter in aT3–1 cells, five block replacement mutations (A–E) were generated that scanned through the 2185/2100 region (Fig. 7A).

Basal Expression of Equine LHb

1501

Fig. 4. Species Conservation of SF-1 Sites within the LHb Promoter The proximal promoter region of the LHb gene from various species was aligned to identify homologous regions. A distal SF-1 site was identified in the equine (E;29), human (H;40), bovine (B;37), ovine (O;21), porcine (P;38), mouse (M;41), and rat (R;39) LHb promoters. This site has been previously characterized in the bovine (25) and rat (24) LHb promoters. A similar DNA sequence was identified more proximal (pSF-1) in all of these species except bovine and ovine. The numbers indicate the position of the sequence relative to the transcription start site (11) for the eLHb promoter.

These mutations were made within the context of the 2448/160 promoter. As an additional test of the importance of this 85-bp segment, we evaluated activity of a promoter that had the bases between 2185 and 2100 deleted (eb D85). Mutation of regions A, B, and E resulted in a decrease in basal promoter activity of 46, 50, and 85%, respectively (Fig. 7B). An additional E mutant was generated (changed E to a different mutant sequence; mE1.2) that also severely attenuated promoter activity (data not shown). The impact of these E mutations was essentially indistinguishable from the effect of the 85-bp internal deletion (eb D85). In contrast, the C and D mutations had little to no effect on promoter activity. Interestingly, the D mutation disrupts two cytosines that have been shown to be critical for SF-1 binding (13, 35). These data from the D mutation further support the data described above for the dSF-1 mutant. Collectively, the data reveal that the regions defined by A, B, and E are essential for full promoter activity, while the greatest impact was seen from mutation of E. Attempts were subsequently made at characterizing the protein(s) that interacted with the E region of the promoter. EMSAs were performed using 21- and 51-bp oligodeoxynucleotide probes encompassing the E region. The shorter probe represented sequence within the E region and included additional 39-sequence. This probe should not interact with SF-1. The

longer probe extended both 59 and 39 to region E and encompassed the SF-1 site. A representative EMSA using these probes is shown in Fig. 8A. We were unable to detect protein binding to ebE (lane 1, Fig. 8A), while ebDEF interacted with SF-1 (lane 2). Upon overexposure, an extremely weak band could be detected with the ebDEF probe that migrated more slowly than SF-1. The EMSA conditions that were used did not appear to be problematic in that they allowed binding of SF-1 to the probes containing an SF-1 response element (Figs. 5 and 8A) and also binding of proteins to consensus AP1 and Sp1 probes (Fig. 8B). Due to the inability to conclusively identify protein binding to the E region with the eb E and DEF probes, protein binding was evaluated using a larger fragment of the eLHb promoter. It was reasoned that protein binding to the E region might be unstable or transient. Use of a larger probe may allow for protein-protein interactions that might stabilize such interactions and allow complex formation and visualization in an EMSA. A restriction fragment encompassing the 2185/278 region of the promoter was isolated, radiolabeled, and used as an EMSA probe. Comparable restriction fragments were also isolated from promoters containing mutations within region E (mE1.1 and mE1.2) and the distal SF-1 site (mdSF1), radiolabeled, and used as probes. Inclusion of the mutant probes permitted a

MOL ENDO · 1999 1502

Vol 13 No. 9

Fig. 5. SF-1 Can Bind to Both the Distal and Proximal SF-1 Sites in the eLHb Promoter EMSAs were performed with nuclei from aT3–1 cells and labeled oligodeoxynucleotide probes representing the eLHb distal SF-1 site (dSF1, 2135/2104), the eLHb proximal SF-1 site (pSF1, 264/239), and the human a-subunit GSE/SF-1 (GSE, 2228/2198 of the human a-subunit promoter; Ref. 30). A, Included in the reactions were antisera to SF-1 (directed against the DNA binding domain of SF-1), normal rabbit sera (NRS), or the indicated unlabeled, competitor DNA (100-fold molar excess). B, Binding of SF-1 to the radiolabeled ha GSE probe was competed for by using 100-fold molar excess of unlabeled eb dSF-1 (lane 15), eb pSF-1 (lane 16), or ha GSE (lane 17). C, Competition for SF-1 binding to radiolabeled eb dSF-1 by no competitor (lane 1), 10-fold (lanes 2, 5, and 8), 50-fold (lanes 3, 6, and 9) or 100-fold (lanes 4, 7, and 10) molar excess of dSF-1 or pSF-1 eLHb oligos or the ha GSE. The protein complex corresponding to SF-1 is indicated by the arrow.

Basal Expression of Equine LHb

Fig. 6. The Distal and Proximal SF-1 Sites in the eLHb Promoter Are not Essential for Maintaining Basal Activity in aT3–1 Cells The plasmid constructs indicated along the left side of the figures were transfected into aT3–1 cells and evaluated for basal transcriptional activity (A). Luciferase activity was normalized to b-galactosidase levels. The ability of SF-1 to transactivate the eLHb promoter was also evaluated (B). aT3–1 cells were cotransfected with various eLHb promoter constructs or a RSV Luc construct along with a RSV expression vector encoding SF-1. Data (B) are expressed as fold induction over cotransfection with a RSV globin expression vector. Values shown are the mean 6 SEM for triplicate wells from a minimum of three independent transfections.

correlation between protein binding and functional activity since regions covered by the mutations were the same as those tested in the transfection assays. Multiple protein complexes were observed when the wildtype region of the promoter was used as probe (Fig. 8B, lane 5). One of the complexes was absent when the regions containing the E mutations were used as probes (lanes 6 and 7). This complex was also absent when the probe containing the mutant SF1 site was used (lane 8) and suggests that this complex represents SF-1. Furthermore, since the mdSF1 construct was fully functional (Fig. 5A), this complex is presumably not critical for basal activity of the eLHb promoter. None of the protein complexes shown in Fig. 8B could be definitively assigned to the E region. However, it is interesting to note that the intensity of the predominant complex (as well as an additional slower migrating complex; bracketed in Fig. 8B) was slightly diminished with the mE1.1 and mE1.2 probes. Activity of the eLHb Promoter in LbT2 Cells To affirm the results obtained in aT3–1 cells, the LbT2 gonadotrope cell line was obtained, and the previous

1503

promoter constructs were reevaluated. LbT2 cells are a newly derived gonadotrope cell line and, unlike the aT3–1 cells, they express their endogenous LHb-subunit gene and it is responsive to GnRH (32). As was the case for aT3–1 cells, the equine promoter was active in LbT2 cells, and the region between 2185 and 2100 retained its importance in enhancing basal promoter activity (Fig. 9). Unlike aT3–1 cells, the bovine and mouse LHb promoters exhibited some basal promoter activity (Fig. 10A); however, the equine promoter had considerably more activity. The importance of the SF-1 sites as well as the A, B, and E regions were subsequently evaluated in the LbT2 cells. Mutation of the distal, proximal, or both SF-1 sites resulted in a 34, 14, and 24% decrease, respectively, in basal promoter activity of the eLHb promoter (Fig. 10B). Thus, the SF-1 sites appear to have a more critical role in regulating promoter activity in this cell line. Analysis of the A–E mutations confirmed the results from aT3–1 cells with one exception (Fig. 10C). Mutation of region D had a more severe affect on promoter activity and presumably reflects the increased importance placed on the distal SF-1 site in this cell line. Furthermore, these data confirm the importance of regions B and E in maintaining basal activity of the eLHb promoter.

DISCUSSION Much has been learned about regulation of LH secretion during various physiological states and the roles played by GnRH, steroids, and other factors (3). Information is also available as to the associated changes that occur in gonadotropin mRNA levels (3). However, a paucity of information exists as to the molecular events that are required to produce these changes, and in particular, the requirements for transcription of the LHb-subunit gene. This has been an area of intense investigation, but one in which little progress has been made. The major obstacle impeding rapid progress has been the lack of an in vitro model for dissecting LHb promoter regulation. We have used the aT3–1 and LbT2 cell lines to study expression of the eLHb-subunit promoter. Unlike the bovine, mouse, and rat (23, 24) LHb promoters, the eLHb promoter is active in the aT3–1 cell line. Furthermore, the equine promoter is approximately 7 times more active in LbT2 cells than are the bovine and mouse LHb promoters. This has allowed us to begin to identify DNA response elements and trans-acting factors that are involved in regulating basal expression of the eLHb promoter. Analysis of 59-deletion mutants of the eLHb promoter indicated that all of the constructs had basal activities greater than that of the promoterless control in aT3–1 (ranged from 7- to 18-fold over pGL2 basic) and LbT2 cells (16- to 112-fold over pGL2 basic). These same constructs were inactive in a human choriocarcinoma cell line. Thus, the transcription factors

MOL ENDO · 1999 1504

Vol 13 No. 9

Fig. 7. Multiple Regions between 2185 and 2100 Regulate Basal Activity of the eLHb Promoter Shown in uppercase letters is the nucleotide sequence of the eLHb promoter between 2181 and 2106 (A). The A–E regions are indicated above the sequence. In lowercase below the native sequence are the mutations that were made in each of these regions (mA–mE). B, The mutations shown in panel A were placed individually into the eb 2448/160 Luc vector and tested for functional activity in aT3–1 cells as described in Materials and Methods. The ebD85 construct had the bases between 2185 and 2100 deleted. Schematic representations of these constructs are shown along the left side of the figure and functional activity is shown within the graph. Luciferase activity was normalized to b-galactosidase levels and is expressed as a percentage of the wild-type promoter (eb 2448/160). Values shown are the mean 6 SEM for triplicate wells from a minimum of three independent transfections.

involved in expression of the eLHb promoter in aT3–1 and LbT2 cells appear to be absent in BeWo cells. Of interest is the fact that we and others (22) have not observed any significant activity of the bLHb promoter in aT3–1 or BeWo cells. Similarly, the rat LHb promoter has been reported to be inactive in aT3–1 cells (24), as is the mouse LHb promoter based on our own data. Together these findings, as well as those from LbT2 cells (Fig. 10A), suggest that differences exist between species in the requirements for gonadotrope (aT3–1, LbT2) expression of the LHb gene. The unique features that result in elevated activity of the eLHb promoter appear to reside within the 2185/160 region. Four primary areas of sequence divergence exist in this region between the equine and bovine promoters: 2185 to 2130, 2115 to 2113, 275 to 257, and 216 to 145. The two most distal regions are located within the block replacements shown in Fig. 7 and are involved in maintaining basal promoter activity. It is interesting to note that region E was extremely important for basal activity of the promoter, and this sequence differs from bovine by only two nucleotides. Minor sequence variation in these regions may account for the elevated basal activity of the equine promoter as compared with bovine. Basal activity of the eLHb promoter is primarily driven by sequences located between 2185 and 2100. This is in contrast to data obtained from a study

of the rat LHb promoter in primary cultures of rat pituitary cells (20). A gradual decline in basal promoter activity was observed when serial 59-deletions were performed on a 1.7-kb promoter. The shortest rat LHb promoter tested (75 bp) maintained 37% of the activity of the 1.7-kb promoter. It is not clear as to whether these discrepancies are due to species differences, use of a different model, or a combination of these two. It was initially hypothesized that a SF-1 site located between 2185 and 2100 was responsible for regulating basal activity of the eLHb promoter. This hypothesis was formulated based on 1) data shown in Fig. 2A; 2) the striking conservation of this sequence across species (Fig. 4, dSF-1); 3) previous reports regarding SF-1 regulation of the bovine and rat LHb promoters (24, 25); and 4) the loss of LHb expression in mice harboring a homozygous disruption of the Ftz-F1 gene, which encodes SF-1 (29). Furthermore, as we began to analyze the eLHb promoter, we identified a second, putative SF-1 binding site (Fig. 4, pSF-1). This site was similar to the distal SF-1 site and was conserved in the human, porcine, rat, and mouse LHb promoters. This proximal SF-1 site has recently been determined to be functional in the rat LHb promoter (26). Due to the conservation and utilization of SF-1 response elements in the GnRH receptor (36), LHb(21, 37–41), and a-subunit promoters (13, 35), it sug-

Basal Expression of Equine LHb

Fig. 8. Multiple Proteins Bind to the 2185 to 278 Region of the eLHb Promoter A, EMSAs were performed with nuclei from aT3–1 cells and labeled oligodeoxynucleotide probes representing the eLHb E region (E, lane 1; 2119 to 299), the eLHb DEF region (DEF, lane 2; 2128 to 279), as well as with consensus AP1 (lane 3) and Sp1 (lane 4) probes. A protein complex representing SF-1 is indicated by the arrow. B, Restriction fragments encompassing the 2185 to 278 region of the eLHb promoter were isolated from the wild type (lane 5), mE (lanes 6 and 7), and mdSF1 (lane 8) clones, radiolabeled, and used as EMSA probes. Multiple protein complexes interacted with these probes. A complex representing SF-1 is indicated as well as additional complexes of proteins (bracketed).

gests that SF-1 may be serving a role in the gonadotrope analogous to the trophoblast-specific element (TSE) and TSEB in regulating placental expression of the human a- and CGb-subunits (19). Data from the current study indicate that SF-1 can indeed bind to both the distal and proximal SF-1 sites and activate the eLHb promoter (Figs. 5 and 10B). Mutation of the SF-1 sites individually or in combination did not alter basal activity of the eLHb promoter in aT3–1 cells (Fig. 6A) but did so in LbT2 cells. Fold activation of the equine promoter by SF-1 was similar to that reported for the rat LHb promoter in LbT2 cells (26). Data from the EMSAs suggested that SF-1 was not limiting in these cells. This contention was supported by a report indicating that mutation of the SF-1 site within the human a-subunit promoter decreased basal promoter activity (35). Thus, SF-1 is not limiting for transactivation of the a-subunit promoter. Based on the data presented in Fig. 5C, it appears that the human a-subunit SF-1 site has a similar affinity toward SF-1 as does the dSF-1 in the eLHb promoter. It is also interesting to note that SF-1 was unable to induce activity of a 282/15 rat LHb promoter containing the pSF-1 site that we have identified (24). However, in a

1505

Fig. 9. The eLHb Promoter Is Active in LbT2 Cells The plasmid constructs indicated along the left side of the figure were transfected into LbT2 cells and evaluated for luciferase reporter gene activity as described in Materials and Methods. Luciferase activity was normalized to b-galactosidase levels. Values shown are the mean 6 SEM for triplicate wells from a minimum of three independent transfections.

subsequent study this proximal site was shown to be functional (26). Several potential explanations exist as for why mutations of the SF-1 sites had no detrimental effect on activity of the equine LHb promoter, but does affect activity of the human a-subunit promoter. SF-1 may be interacting with other transcription factors, and these factors may differ between the a- and LHbsubunits. This is supported by data indicating that an immediate early response gene (Egr1/NGFIA) can interact with SF-1 and regulate expression of the rat LHb promoter (42). Effective interaction with a second transcription factor such as Egr1 may require higher cellular concentrations of SF-1, hence, the lack of SF-1 regulation of the eLHb promoter under basal conditions in aT3–1 cells. Alternatively, the lack of a response in aT3 cells may be due to the fact that these cells also express Dax-1 (43), which can block or repress SF-1-mediated transcription (M. Wolfe, unpublished data). Overexpression of SF-1 may have overcome this block and revealed the functional importance of the SF-1 response elements. In light of the recent SF-1 transgenic mouse data (25, 34), further experimentation is warranted to determine the in vivo significance (i.e. transgenic mice) of the eLHb SF-1 sites and what role, if any, they play in the spatiotemporal expression pattern and hormonal regulation of the eLHb-subunit gene. It is interesting to note that exogenous administration of GnRH to the SF-1-deficient mice activated expression of LHb (44), suggesting that SF-1 is not essential for expression of LHb in vivo. However, it has also been shown that GnRH can increase the expression of SF-1 in gonadotropes (45). We and others have recently shown that SF-1 and Egr1 interact and that GnRH regulation of the LHb promoter occurs through the SF-1 and Egr response elements (27, 28). Thus, GnRH appears to be

MOL ENDO · 1999 1506

Fig. 10. The SF-1 Sites and Regions B, D, and E Are Required for Basal Activity of eLHb Promoter in LbT2 Cells A, Basal activity of the human (ha 2485/148), mouse (ma 2507142), and bovine (ba 2315/145) a-subunit promoters and equine (eLHb 2185/160), bovine (bLHb 2185/ 110), and mouse (mLHb 2196/18) LHb-subunit promoters were elevated in LbT2 cells. Data are expressed as fold induction over a promoterless control luciferase vector (pGL2 basic). B, The plasmid constructs described in Fig. 6 and indicated along the left side of the figures were transfected into LbT2 cells and evaluated for basal transcriptional activity. Luciferase activity was normalized to b-galactosidase levels. C, eLHb promoters containing the A–E mutations and D 85 deletion (described in Fig. 7) were tested for functional activity in LbT2 cells as described in Materials and Methods. Luciferase activity was normalized to b-galactosidase levels and is expressed as a percentage of the wild-type promoter (eb 2448/160). Values shown are the mean 6 SEM for triplicate wells from a minimum of three independent transfections.

able to regulate LHb through SF-1-dependent and -independent pathways. The most striking outcome of this study was the discovery that an 85-bp fragment of the eLHb pro-

Vol 13 No. 9

moter was required for basal activity in aT3–1 and LbT2 cells. Deletion of the bases between 2185 and 2100, within the context of the 2448/160 construct, severely attenuated promoter activity. Furthermore, attenuated promoter activity could be recapitulated by mutating the bases between 2119 and 2105 (region E). The block replacement mutagenesis uncovered an additional segment of DNA within the 85-bp region (regions A and B) that also plays a role in promoter activation. Mutation of these bases attenuated promoter activity, but not as effectively as mutations within region E. Attempts at further defining the DNA response element in the E region using smaller mutations have been unsuccessful. We are currently focusing on this region, as well as B, to identify the transcription factors responsible for maintaining elevated basal activity of the eLHb promoter. Both regions are G/C rich, and preliminary data suggest that they may represent weak Sp1 binding sites (M. Wolfe, unpublished). Additional evidence suggests an Egr site lies immediately 39 to region E. Therefore, the E mutation would disrupt Egr binding. We have been unable to detect expression of Egr1 or binding to this site using nuclear proteins from unstimulated aT3–1 cells (Fig. 8 and Ref. 28). Unlike the E mutant, mutation of this Egr site has no detrimental effect on eLHb promoter activity in aT3–1 cells (28). These data suggest that some other transcription factor may bind to region E. In contrast, basal promoter activity is attenuated in LbT2 cells when the Egr site is mutated. Furthermore, Egr proteins are expressed basally in LbT2 cells (46). This could be one explanation as to why the E mutation leads to attenuated promoter activity in LbT2 cells. In summary, the aT3–1 and LbT2 cell lines are useful models for studying expression of the eLHb gene. Two SF-1 response elements were identified within the eLHb promoter that have been conserved across other species. These sites are functional and contribute to basal activity of the eLHb promoter in LbT2, but not aT3–1, cells. Two other regions of the promoter that lie between 2185 and 2100 were identified as being required for basal activity. The most important of these lies immediately downstream of the distal SF-1 site. At present, it is unclear as to what protein(s) binds to either of these regions, although they have some homology to Sp1 sites. These data are some of the first to identify functional cis-acting elements that play critical roles in regulating basal activity of a LHb gene. Transcriptional regulation of the eLHb gene appears to involve multiple elements as does expression of the glycoprotein hormone a-subunit gene. Further experimentation will be required to determine whether other similarities exist in regulation of gonadotrope expression of the a- and LHb- subunit genes.

Basal Expression of Equine LHb

MATERIALS AND METHODS Materials Restriction enzymes and other enzymes were obtained from the Promega Corp. (Madison, WI) and Life Technologies, Inc., Inc. (Gaithersburg, MD). Oligodeoxynucleotides were obtained from Midland Certified Reagent Co. (Midland, TX) or Life Technologies, Inc.. Bluescript plasmid vector (pBSK-) used for DNA cloning was obtained from Stratagene (La Jolla, CA), and the luciferase reporter vectors (pGL2 basic and control) were obtained from Promega Corp.. All radionuclides were purchased from New England Nuclear Life Science Products, Inc. (Boston, MA). A random prime labeling kit was purchased from Roche Molecular Biochemicals (Indianapolis, IN). DNA sequencing was conducted using Sequenase purchased from United States Biochemical Corp. (Cleveland, OH) or through cycling sequencing using reagents purchased from PE Applied Biosystems and subsequently run on the ABI 310 sequencer (Perkin Elmer Corp./ABD, Norwalk, CT). PCR amplification of DNA was performed using Taq polymerase (Life Technologies, Inc.) or Deep Vent DNA polymerase (New England Biolabs, Inc., Beverly, MA). All other chemicals and reagents were obtained from Pharmacia Biotech (Piscataway, NJ), Fisher Scientific (Pittsburgh, PA), Sigma Chemical Co. (St. Louis, MO), and Life Technologies, Inc.. Equine Genomic Clone To isolate additional 59-flanking sequence for the eLH/CGb gene, a Lambda FIX II equine genomic library (Stratagene; La Jolla, CA) was screened using standard procedures (40) and the 2448/160 equine promoter fragment as radiolabeled probe (29). Approximately 1.1 3 106 plaque-forming units were screened, from which three clones were isolated, each containing a 16-kb insert. Restriction analysis revealed that two of the clones were identical. Further characterization of the clones was determined by performing Southern analyses using either the eLH/CGb promoter fragment or cDNA as probes (29) and by sequencing a portion of the clone (48). Plasmid Constructs The pGL2 basic plasmid was used as a reporter vector for all of the promoter constructs used in this study. The original promoter clone isolated by Sherman et al. (29) was used to generate all of the constructs containing 448 bp or less of 59-flanking sequence. The original equine clone was generated by PCR using an upstream bovine consensus oligodeoxynucleotide (2440/2420; Ref. 29) that contained a 59 HindIII site and an oligodeoxynucleotide specific to the 59 untranslated region of the equine LH/CGb gene (141/160). PstI linkers were added to this PCR product and subsequently digested with PstI and partially with HindIII. The PstI and HindIII fragment was subcloned into the HindIII and PstI sites of BSK. The longer eLHb promoter constructs were generated as follows. The eb5l clone was digested with SacI and HindIII, gel isolated (2.5-kb fragment), partially digested with SalI, and reisolated as a 2.5-kb fragment containing the promoter. This represented a fragment cut at a SalI site in the lambda multiple cloning site and 39 at a HindIII site located at 2387 in the promoter. The cloning vector was made by cutting the original promoter clone (29) with XhoI (cuts in multiple cloning site and is compatible with SalI) and HindIII (cuts at 2387 in promoter) and isolating the vector. The SalI/HindIII promoter fragment was ligated into the XhoI/HindIII sites of the eLHb BSK vector, resulting in a 3-kb promoter construct. The 2448/160, 2185/160, and 2100/160 constructs were made by digesting the parent vector with ClaI and SauI (2448), NarI (2185), or SmaI (2100), respectively, filling in the

1507

ends and religating. This removed the upstream bovine oligodeoxynucleotide and additional 59-flanking sequence in the case of the shorter constructs. The 2387/160 promoter was constructed by digesting the parent promoter with HindIII and religating. These promoter constructs were subsequently subcloned into the pGL2 basic reporter vector. Promoter-containing plasmids were digested with PstI, blunted, digested with XhoI, and gel isolated. These fragments were ligated into the XhoI and blunted BglII sites of pGL2 basic. The eb 22200/160 and 21400/160 constructs were generated from the eLHb 23000/160 BSK vector by partial digestion with XhoI (sites at 22200 and 21400) and complete digestion with SstI. Promoter fragments were isolated and subsequently ligated into the eb 2448/160 luc vector that had been cut with XhoI and SstI. All mutant promoter constructs were made in the context of the 2448/160 promoter. Mutation of the dSF-1 site was accomplished by PCR using an upstream oligodeoxynucleotide located in the pGL2 vector (GL1), a downstream mdSF-1 oligodeoxynucleotide encompassing bases 2135 to 273, and eb2448/160 Luc as template. The dSF-1 site was mutated from TGACCTTG to aGAtCTTG. It has previously been shown that the CC pair within the GSE was critical for protein binding (13). The PCR product was digested with SstI and BstEII, and this fragment (2343 to 278) was gel isolated. The eb2448/160 Luc vector was digested with SstI and BglII and both fragments were isolated. The smaller fragment (2343 to the BglII site 59 to the luciferase gene) was digested with BstEII, and the BstEII/BglII fragment was isolated. The SstI/BglII digested eb 2448/160 Luc vector, the BstEII/BglII fragment, and the SstI/BstEII PCR fragment were subsequently ligated to generate eb 2448/160 mdSF1 Luc. The proximal SF-1 site was mutated by a similar PCR strategy. Two PCR reactions were performed using eb 2448/ 160 Luc as template. The first used GL2 (downstream oligodeoxynucleotide located in luciferase) and an oligodeoxynucleotide encompassing bases 264 to 227 (sense strand), while the second reaction used GL1 and an antisense oligodeoxynucleotide encompassing bases 264 to 227. The oligodeoxynucleotides encompassing 264 to 227 mutated the pSF-1 from TGGCCTTG to aGatCTTG and generated a BglII site. The GL1/mpSF-1 PCR product was digested with SstI and BglII, while the mpSF-1/GL2 PCR product was digested with BglII alone. These fragments were subsequently ligated into the eb 2448/160 Luc vector digested with SstI and BglII. Positive clones were evaluated for the correct orientation. Both SF-1 mutant clones were sequenced to confirm that the appropriate mutations had been made. A similar paradigm was used for the block replacement mutants. Oligodeoxynucleotides were synthesized containing the mutations shown in Fig. 7A. Additional 59 and 39 sequence was incorporated onto these oligodeoxynucleotides to allow for annealing to the eb 2448/160 Luc template (mA 2191/2150; mB 2191/2136; mD 2149/294; mE 2135/ 294; the latter two oligos were in the reverse orientation). The second oligo that was used in PCR corresponded to bases 2224/2205 (for mutants D and E) or 141/160 (reverse orientation; for mutants A and B). These PCR products were gel isolated, digested with NarI and SmaI (or AvaI), and ligated into eb 2448/160 BSK that had been digested with NarI and SmaI (or AvaI). Generation of the C mutant required a twostep process with two mutant oligos. Two PCR reactions were performed: the first with the 2224/2205 oligo and a reverse orientation mC (2164/2136) and the second with a positive orientation mC (2149/2120) oligo and the 141/160 reverse orientation oligo. These PCR products were gel isolated and digested with NarI and BglII or BglII and SmaI (or AvaI), respectively (the BglII site is within the mutation), and gel isolated again. The two DNA fragments were ligated, digested with NarI and SmaI (or AvaI), and subsequently ligated into the NarI/SmaI (or AvaI) digested vector. The ebD85 Luc construct was made by blunting the NarI/SmaI digested eb 2448/160 BSK followed by religation. These

MOL ENDO · 1999 1508

mutant promoters were subcloned into pGL2 basic as described above and were sequenced to confirm that the appropriate mutation had been made. The second E mutation (E1.2) was generated using a strategy similar to that used to make the mdSF1 clone. The mutant promoter fragment was amplified using GL1 and an oligo encompassing the bases between 2128 and 273. This mutated the bases between 2119 and 2105 from TTGTCCGCCTCTCGC to ggtaaCtagTacgta and differs from the original mutation (Fig. 7A). The PCR product was digested with SstI and BstEII and ligated into the eb 2448/160 Luc vector previously cut with SstI and BstEII as described above. Positive clones were sequenced to confirm that the appropriate mutation had been made. The bovine and mouse LHb promoter fragments were generated by PCR using the following oligos (59–39): 59-bLHb oligo AATCTCGAGTACGGGAGCCACTCAGG (2185/1168), 39-bLHb oligo GTTAAGCTTCTTGGTGCCTCCCCTGC (27/ 110), 59-mLHb oligo AGGGCTAGCTCGAGCCCTGACACCTGGGC (2196/1181), 39-mLHb oligo AGGAAGCTTAGATCTTTGATACCCTTCCCTAC (212/18). Bovine or mouse genomic DNA served as template. Products from the PCR reactions were digested with XhoI and HindIII (engineered into the oligos) and ligated into pGL2 basic previously cut with XhoI and HindIII. Positive clones were isolated and sequenced to confirm the accuracy of the PCR reaction. The pGL2 basic vector served as a promoterless control, while the the pGL2 control (contains the SV40 promoter and enhancer) vector served as a positive control in the transient transfection experiments. An additional viral promoter linked to luciferase (pGL2) that was used as a positive control was the RSV long-terminal repeat. A 2.1-kb mouse SF-1 cDNA was obtained for experiments involving overexpression of SF-1 (49). This cDNA was subcloned into a RSV-driven expression vector (50). A similar construct containing a globin cDNA was used as a control. Cell Culture and Transient Transfections Cultures of aT3–1 cells (31) were plated in DMEM with 5% FBS, 5% horse serum, and antibiotics. On the day before transfection, aT3–1 cells were plated at a density of 1.8 3 105 cells per well in six-well plates. Cells were transfected with up to 1.5 mg of plasmid DNA, 400 ng of RSVbGal (internal control of transfection efficiency), and 7 ml of LipofectAmine (Life Technologies, Inc.) according to the manufacturers recommendations. Briefly, DNA and LipofectAmine were diluted separately in OptiMEM, combined, and incubated at room temperature for approximately 30 min. Media were aspirated from the cells and replaced with the DNA/LipofectAmine mix. The plates were then returned to the CO2 incubator. After an overnight incubation, the DNA/LipofectAmine mix was removed, fresh media were added, and the plates were returned to the incubator. Cells were harvested 2 days posttransfection. Plasmid constructs were evaluated in triplicate within each transfection, and transfections were performed a minimum of three times unless noted otherwise. Cultures of LbT2 cells (32, 33) were plated on ECM (Sigma Chemical Co.)-coated plates in DMEM containing 10% FBS and antibiotics. On the day before transfection, cells were plated at 1.5 3 105 cells per well in 12-well plates. Cells were transfected with 1 mg of plasmid DNA, 400 ng of RSVbGal, 3 l of LipofectAmine Plus, and 2 l of LipofectAmine following the manufacturers recommendations. Cells were incubated with the DNA/Lipid mix for 4–5 h. Media containing 20% FBS were then added and the incubation was continued overnight. On the following morning, the transfection media were aspirated and replaced with fresh media. Cells were harvested 2 days posttransfection. Plasmid constructs were evaluated in triplicate within each transfection, and transfections were performed a minimum of three times unless noted otherwise.

Vol 13 No. 9

Reporter Assays Luciferase assays were performed following the protocol for the Promega Luciferase assay system (Promega Corp.). Briefly, cells were washed twice in PBS and harvested in 150 ml of reporter lysis buffer (Promega Corp.). After the addition of luciferase assay buffer (100 ml), relative luciferase activity (20 ml of lysate) was measured for 10 sec in a Berthold Lumat LB 9501 luminometer (Wallac, Inc. Gaithersburg, MD). The Galacto-Light b-galactosidase reporter gene assay system (Tropix, Bedford, MA) was used to anaylze b-galactosidase activity. Reaction buffer (100 ml) was added to cell lysates (20 ml) and incubated at room temperature for 1 h. Light emission accelerator (150 ml) was added, and light emission was measured for 5 sec in a luminometer. Luciferase activity (relative light units) was normalized to the activity of b-galactosidase. Nuclear Preparation and EMSAs Nuclei were prepared from aT3–1 cells according to the methods of Hagenbuchle and Wellaur (51). Nuclei were diluted to a concentration of 2–4 3 105 nuclei/ml and stored at 280 C. Oligodeoxynucleotides were end labeled with [32P]ATP and T4 kinase while restriction fragments were labeled with [32P]dCTP and [32P]dATP using Klenow polymerase. These labeled DNAs were used as probes in EMSAs. Nuclei (1–2 ml) were incubated for 15 min at room temperature in binding buffer (10 mM HEPES, pH 7.9, 100 mM KCL, 5 mM MgCl2, 10 mM ZnCl2, 1 mM EDTA, 10% glycerol) containing 0.5 mg poly (dA-dT), 0.25 mg poly (dI-dC), and 0.1 mg salmon sperm DNA. Labeled probe (50 fmol) and competitor were then added and incubated for an additional 15 min at room temperature (total reaction volume of 20 ml). The DNAprotein complexes were resolved on a 4% native polyacrylamide gel (prerun for ;30 min) in 0.53 Tris-borate-EDTA buffer. In experiments where the antibody against the DNAbinding domain of SF-1 or normal rabbit sera were used, sera (2 ml) were added to the reaction 30 min before addition of labeled probe. The reaction was allowed to incubate for an additional 15 min after inclusion of labeled probe.

Acknowledgments We wish to thank Weiwei Zhao, Gerald Call, Dr. Leslie L. Heckert, and Dr. Michael J. Soares for their technical advice and assistance. Sincere appreciation and gratitude is expressed to Dr. John H. Nilson for the bovine a-subunit promoter and for his generous advice and support of this work. We would also like to thank Dr. Pamela Mellon for the aT3–1 and LbT2 cell lines, Dr. Keith Parker for the SF-1 cDNA, Dr. Ulf Rapp for the RSV expression vector, Dr. Leslie L. Heckert for the human a-subunit promoter, and Dr. Mark S. Roberson for the mouse a-subunit promoter.

Received April 28, 1998. Revision received April 22, 1999. Accepted June 2, 1999. Address requests for reprints to: Michael W. Wolfe, Ph.D., Department of Molecular and Integrative Physiology, University of Kansas Medical Center, 3901 Rainbow Boulevard, Kansas City, Kansas 66160-7401. E-mail: mwolfe2@kumc. edu. This work was supported by NIH Grant DK-50668 (M.W.W.) and was performed with the assistance of the Imaging/Photography and Cell Culture Cores of the NIHsupported Center of Reproductive Sciences (Grant HD-33994).

Basal Expression of Equine LHb

1509

REFERENCES 1. Pierce J, Parsons TF 1981 Glycoprotein hormones: structure, function. Annu Rev Biochem 50:465–495 2. Fiddes JC, Talmadge K 1984 Structure, expression, and evolution of the genes encoding the human glycoprotein hormones. Recent Prog Horm Res 40:43–78 3. Gharib SD, Wierman ME, Shupnik MA, Chin WW 1990 Molecular biology of the pituitary gonadotropins. Endocr Rev 11:177–199 4. Jameson JL, Hollenberg AN 1993 Regulation of chorionic gonadotropin gene expression. Endocr Rev 14:203–221 5. Bousfield GR, Perry WM, Ward DN 1994 Gonadotropins, chemistry and biosynthesis. In: Knobil E, Neill JD (eds) The Physiology of Reproduction, ed. 2. Raven Press, New York, vol. 1:1749–1785 6. Pittman RH, Clay CM, Farmerie TA, Nilson JH 1994 Functional analysis of the placenta-specific enhancer of the human glycoprotein hormone a subunit gene: emergence of a new element. J Biol Chem 269:19360–19368 7. Silver BJ, Bokar J, Virgin JB, Valen E, Milsted A, Nilson JH 1987 Cyclic AMP regulation of the human glycoprotein hormone a-gene is mediated by an 18 base pair element. Proc Natl Acad Sci USA 84:2198–2202 8. Meyer TE, Habener JF 1993 Cyclic adenosine 39,59monophosphate response element binding protein (CREB) and related transcription-activating deoxyribonucleic acid-binding proteins. Endocr Rev 14:269–290 9. Steger DJ, Altschmied J, Buscher M, Mellon PL 1991 Evolution of placenta-specific gene expression: comparison of the equine and human gonadotropin a-subunit genes. Mol Endocrinol 5:243–255 10. Delegeane AM, Ferland LH, Mellon PL 1987 Tissue-specific enhancer of the human glycoprotein hormone a-subunit gene: dependence on cyclic AMP-inducible elements. Mol Cell Biol 7:3994–4002 11. Schoderbek WE, Kim KW, Ridgway EC, Mellon PL, Maurer RA 1992 Analysis of DNA sequences required for pituitary-specific expression of the glycoprotein hormone a-subunit gene. Mol Endocrinol 6:893–903 12. Steger DJ, Hecht JH, Mellon PL 1994 GATA-binding proteins regulate the human gonadotropin a-subunit gene in the placenta and pituitary gland. Mol Cell Biol 14:5592–5602 13. Horn F, Windle JJ, Barnhart KM, Mellon PL 1992 Tissuespecific gene expression in the pituitary: the glycoprotein hormone a-subunit gene is regulated by a gonadotropespecific protein. Mol Cell Biol 12:2143–2153 14. Roberson MS, Schoderbek WE, Tremml G, Maurer RA 1994 Activation of the glycoprotein hormone a-subunit promoter by a LIM-homeodomain transcription factor. Mol Cell Biol 14:2985–2993 15. Talmadge K, Vamvakopoulos NC, Fiddes JC 1984 Evolution of the genes for the b subunits of human chorionic gonadotropin and luteinizing hormone. Nature 307: 37–40 16. Jameson JL, Lindell CM, Habener JF 1986 Evolution of different transcriptional start sites in the human luteinizing hormone and chorionic gonadotropin b-subunit genes. DNA 5:227–234 17. Fenstermaker RA, Milsted A, Virgin JB, Miller WL, Nilson JH 1989 The transcriptional response of the human chorionic gonadotropin b-subunit gene to cAMP is cycloheximide sensitive and is mediated by cis-acting sequences different from that found in the a-subunit gene. Mol Endocrinol 3:1070–1076 18. Albanese C, Kay TWH, Troccoli NM, Jameson JL 1991 Novel cyclic adenosine 39,59-monophosphate response element in the human chorionic gonadotropin b-subunit gene. Mol Endocrinol 5:693–702 19. Steger DJ, Buscher M, Hecht JH, Mellon PL 1993 Coordinate control of the a- and b-subunit genes of human

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30. 31. 32.

33.

34.

35.

36.

chorionic gonadotropin by trophoblast-specific elementbinding protein. Mol Endocrinol 7:1579–1588 Kim KE, Day KH, Howard P, Salton SRJ, Roberts JL, Maurer RA 1990 DNA sequences required for expression of the LHb promoter in primary cultures of rat pituitary cells. Mol Cell Endocrinol 74:101–107 Brown P, McNeilly JR, Wallace RM, McNeilly AS, Clark AJ 1993 Characterization of the ovine LH b-subunit gene: the promoter directs gonadotrope-specific expression in transgenic mice. Mol Cell Endocrinol 93:157–165 Keri RA, Wolfe MW, Saunders TL, Anderson I, Kendall SK, Wagner T, Yun J, Gorski J, Nett TM, Camper SA, Nilson JH 1994 The proximal promoter of the bovine luteinizing hormone b-subunit gene confers gonadotrope-specific expression and regulation by gonadotropin-releasing hormone, testosterone, and 17b-estradiol in transgenic mice. Mol Endocrinol 8:1807–1816 Fallest PC, Trader GL, Darrow JM, Shupnik MA 1995 Regulation of rat luteinizing hormone b gene expression in transgenic mice by steroids and a gonadotropin-releasing hormone antagonist. Biol Reprod 53:103–109 Halvorson LM, Kaiser UB, Chin WW 1996 Stimulation of luteinizing hormone b gene promoter activity by the orphan nuclear receptor, steroidogenic factor-1. J Biol Chem 271:6645–6650 Keri RA, Nilson JH 1996 A steroidogenic factor-1 binding site is required for activity of the luteinizing hormone b subunit promoter in gonadotropes of transgenic mice. J Biol Chem 271:10782–10785 Halvorson LM, Ito M, Jameson JL, Chin WW 1998 Steroidogenic factor-1 and early growth response protein 1 act through two composite DNA binding sites to regulate luteinizing hormone b-subunit gene expression. J Biol Chem 273:14712–14720 Halvorson LM, Kaiser UB, Chin WW 1999 The protein kinase C system acts through the early growth response protein 1 to increase LHb gene expression in synergy with steroidogenic factor-1. Mol Endocrinol 13:106–110 Wolfe MW, Call GB 1999 Early growth response protein 1 binds to the luteinizing hormone-b promoter and mediates gonadotropin-releasing hormone-stimulated gene expression. Mol Endocrinol 13:752–763 Sherman GB, Wolfe MW, Farmerie TA, Clay CM, Threadgill DS, Sharp DC, Nilson JH 1992 A single gene encodes the b-subunits of equine luteinizing hormone and chorionic gonadotropin. Mol Endocrinol 6:951–959 Sugino H, Bousfield GR, Moore Jr WT, Ward DN 1987 Structural studies on equine glycoprotein hormones. J Biol Chem 262:8603–8609 Windle JJ, Weiner RI, Mellon PL 1990 Cell lines of the pituitary gonadotrope lineage derived by targeted oncogenesis in transgenic mice. Mol Endocrinol 4:597–603 Turgeon JL, Kimura Y, Waring DW, Mellon PL 1996 Steroid and pulsatile gonadotropin-releasing hormone (GnRH) regulation of luteinizing hormone and GnRH receptor in a novel gonadotrope cell line. Mol Endocrinol 10:439–450 Thomas P, Mellon PL, Turgeon JL, Waring DW 1996 The LbT2 clonal gonadotrope: a model for single cell studies of endocrine cell secretion. Endocrinology 137: 2979–2989 Ingraham HA, Lala DS, Ikeda Y, Luo X, Shen W, Nachtigal MW, Abbud R, Nilson JH, Parker KL 1994 The nuclear receptor steroidogenic factor 1 acts at multiple levels of the reproductive axis. Genes Dev 8:2302–2312 Barnhart KM, Mellon PL 1994 The orphan nuclear receptor, steroidogenic factor-1, regulates the glycoprotein hormone a-subunit gene in pituitary gonadotropes. Mol Endocrinol 8:878–885 Duval DL, Nelson SE, Clay CM 1997 A binding site for steroidogenic factor-1 is part of a complex enhancer that mediates expression of the murine gonadotropin-releasing hormone receptor gene. Biol Reprod 56:160–168

MOL ENDO · 1999 1510

37. Virgin JB, Silver BJ, Thomason AR, Nilson JH 1985 The gene for the b subunit of bovine luteinizing hormone encodes a gonadotropin mRNA with an unusually short 59-untranslated region. J Biol Chem 260:7072–7077 38. Ezashi T, Hirai T, Kato T, Wakabayashi K, Kato Y 1990 The gene for the b subunit of porcine LH: clusters of GC boxes and CACCC elements. J Mol Endocrinol 5: 137–146 39. Jameson JL, Chin WW, Hollenberg AN, Chong AS, Habener JF 1984 The gene encoding the b-subunit of rat luteinizing hormone. J Biol Chem 259:15474–15480 40. Talmadge K, Vamvakopoulos NC, Fiddes JC 1984 Evolution of the genes for the b subunits of chorionic gonadotropin and luteinizing hormone. Nature 307:37–40 41. Kumar TR, Matzuk MM 1995 Cloning of the mouse gonadotropin b-subunit-encoding genes, II. Structure of the luteinizing hormone b-subunit-encoding genes. Gene 166:335–336 42. Lee SL, Sadovsky Y, Swirnoff AH, Polish JA, Goda P, Gavrilina G, Milbrandt J 1996 Luteinizing hormone deficiency and female infertility in mice lacking the transcription factor NGFI-A (Egr-1). Science 273:1219–1221 43. Ikeda Y, Swain A, Weber TJ, Hentges KE, Zanaria E, Lalli E, Tamai KT, Sassone-Corsi P, Lovell-Badge R, Camerino G, Parker KL 1996 Steroidogenic factor 1 and Dax-1 colocalize in multiple cell lineages: potential links in endocrine development. Mol Endocrinol 10: 1261–1272 44. Ikeda Y, Luo X, Abbud R, Nilson JH, Parker KL 1995 The nuclear receptor steroidogenic factor 1 is essential for

Vol 13 No. 9

45.

46.

47.

48.

49.

50.

51.

the formation of the ventromedial hypothalamic nucleus. Mol Endocrinol 9:478–486 Haisenleder DJ, Yasin M, Dalkin AC, Gilrain J, Marshall JC 1996 GnRH regulates steroidogenic factor-1 (SF-1) gene expression in the rat pituitary. Endocrinology 137:5719–5722 Dorn C, Ou Q, Svaren J, Crawford PA, Sadovsky Y 1999 Activation of luteinizing hormone-b gene by gonadotropin-releasing hormone requires the synergy of early growth response-1 and steroidogenic factor-1. J Biol Chem 274:13870–13876 Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K (eds) 1994 Current Protocols in Molecular Biology. John Wiley & Sons, Inc., New York Sanger F, Nicklen S, Coulson AR 1977 DNA sequencing with chain-termination inhibitors. Proc Natl Acad Sci USA 74:5463–5467 Lala DS, Rice DA, Parker KL 1992 Steroidogenic factor I, a key regulator of steroidogenic enzyme expression, is the mouse homolog of fushi tarazu-factor I. Mol Endocrinol 6:1249–1258 Dorn P, DaSilva L, Martarano L, Derse D 1990 Equine infectious anemia virus tat: insights into the structure, function, and evolution of lentivirus trans-activator proteins. J Virol 64:1616–1624 Hagenbuchle O, Wellaur PK 1992 A rapid method for the isolation of DNA-binding proteins from purified nuclei of tissues and cells in culture. Nucleic Acids Res 20: 3555–3559