Background
The hypothalamic decapeptide, gonadotropin-releasing hormone (GnRH), binds to its cognate receptor (GnRHR) on the surface of gonadotrope cells within the anterior pituitary gland, stimulating the synthesis and secretion of the gonadotropins, follicle stimulating hormone (FSH) and luteinizing hormone (LH) [
1,
2]. The gonadotropins subsequently act on the gonads to trigger secretion of steroid hormones, which feedback at the level of the hypothalamus and anterior pituitary gland to regulate GnRH and gonadotropin levels, respectively. Binding of GnRH to its seven transmembrane, G-protein-coupled receptor activates multiple signal transduction cascades, ultimately resulting in up-regulation of the genes that encode the common α- and unique β-subunits of FSH and LH [
3,
4] as well as the GnRHR itself [
5]. Therefore, the interaction between GnRH and GnRHR represents a crucial point for regulation of reproductive function in mammals. Furthermore, the porcine GnRHR gene is located on chromosome 8, in close proximity to a quantitative trait locus for ovulation rate, a primary determinant of litter size [
6]. Consistent with this, a C/G substitution in the 3’ untranslated region was shown to be significantly associated with ovulation rate [
7]. Consequently, the GnRHR gene represents both a physiological and positional candidate for genes influencing prolificacy in pigs.
The GnRHR gene promoter has been extensively studied in the mouse, rat, human and sheep [
8,
9]. In the mouse, gonadotrope-specific expression is conferred by 500 bp of GnRHR gene promoter [
10] comprised of binding sites for steroidogenic factor 1 (SF1), activator protein 1 (AP1) and a GnRHR activating sequence (GRAS) [
11,
12], although pituitary homeobox (Pitx)-1 and a member of the LIM homeodomain family, Lhx3, have also been implicated [
13,
14]. Additionally, protein kinase C (PKC) activation of AP1 is critical for GnRH responsiveness of the murine GnRHR gene promoter [
15], whereas the GRAS binding site acts in conjunction with a downstream activin regulatory element to control activin responsiveness [
16]. Subsequent studies revealed that the GRAS element binds a complex of transcription factors including AP1, Smad3 and 4 and FOXL2, a member of the forkhead family of transcription factors [
17]. Furthermore, an enhancer element, sequence underlying responsiveness to GnRH (SURG)-1 [
18], binds octamer transcription factor-1 (OCT1) and nuclear factor (NF)-Y [
19] for basal and maximal GnRH stimulation of GnRHR gene transcription. A proximal homeodomain (Hbox) binding motif also binds OCT1, indicating the transcription factor acts at multiple TAAT sites to direct basal expression [
20]. Interestingly, the CLOCK and BMAL1 drive activity of the murine GnRHR promoter through their interaction with E-box enhancer sequences [
21]. The importance of protein kinase A (PKA) signaling was also illustrated by the role of a cAMP responsive element (CRE) in activation of the GnRHR gene [
22].
The rat and mouse GnRHR gene promoters appear consistent, both containing SF1, AP1 and GRAS binding sites [
23]. However, the GRAS element in the rat GnRHR promoter harbors an A → G bp alteration, dramatically reducing effectiveness of this binding site [
24]. Furthermore, the rat promoter contains CRE-like and SF1 adjacent protein (SAP) binding sites involved in basal activity [
25] and a GnRHR specific enhancer (GnSE) [
24] that interacts with GATA factors and the LIM-related factors, Isl-1 and Lhx3, to promote maximal basal activity [
25]. Although activity of the human GnRHR promoter in cell lines of non-gonadotrope origin and in response to hormones have been characterized [
8], investigation into gonadotrope-specific activity has only elucidated a SF1 binding site [
26]. Additional studies revealed that AP1 confers down-regulation following GnRH stimulation [
27] and OCT1 serves as a strong constitutive repressor [
28]. Elements conferring gonadotrope-specific expression of the ovine GnRHR gene remain to be elucidated; however, Duval and coworkers [
29] reported a SF1 binding site that mediates basal expression.
Our laboratory has previously shown that gonadotrope-specific activity of the porcine GnRHR gene is partially conferred by a SF1 binding site positioned within a 112-bp upstream enhancer (−1779/−1667), as well as two additional SF1 and one retinoid X receptor (RXR) binding sites located within 315 bp of proximal promoter [
30]. In order to compare transcriptional regulation of the GnRHR gene among pig lines with divergent ovulation rates, we constructed luciferase reporter constructs containing 5118 bp of 5’ flanking sequence from three genetic lines of swine: a Control white-crossbred line; a Nebraska Index line selected for over 14 generations based on an index of ovulation rate and embryonic survival [
31]; and the Chinese Meishan breed, a line with increased prolificacy over white-crossbred lines, largely due to a greater ovulation rate [
32,
33]. Previously, our laboratory has shown that anterior pituitary levels of GnRHR mRNA were highest in Meishan, intermediate in Index and lowest in Control [
34]. Herein, we demonstrate differential activity among these line-specific GnRHR promoters utilizing transient transfections assays in gonadotrope-derived αT3-1 cells. In addition, we identified three bp substitutions at −1690 (T → C), −1235 (C → G) and −845 (G → T) of proximal promoter that allow GATA-4, the p52 and p65 subunits of nuclear factor (NF)-κB as well as a specificity protein (SP)1-like factor, and GATA-4, respectively, to preferentially bind the Meishan compared to Index or Control GnRHR gene promoters.
Materials
The antibody directed against the p65 subunit of NF-κB (catalog no. PC137) was purchased from Calbiochem (La Jolla, CA), the antibodies specific for the p52 subunit of NF-κB (catalog no. 06–413), SP1 (catalog no. 07–645), SP3 (catalog no. 07–107) were from Upstate (Charlottesville, VA), the specific antibodies for the p50 subunit of NF-κB (catalog no. sc-114X), SP1 (catalog no. sc-59X), SP2 (catalog no. sc-643X), SP4 (catalog no. sc-13019X), GATA-1 (catalog no. sc-1234X), GATA-2 (catalog no. sc-9008X), GATA-4 (catalog no. sc-1237) and normal rabbit IgG (catalog no. sc-2027) were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). For experiments using EMSA, competitive oligonucleotides containing consensus binding sites for activator protein (AP)2, NF-κB, SP1 or GATA were synthesized by Integrated DNA Technologies (Coralville, IA; Table
1).
Table 1
Sense strand of EMSA oligosa
AP2 consensusb
| 5’-GATCGAACTGACCGCCCGCGGCCCGT-3’ |
GATA consensusb
| 5’-CACTTGATAACAGAAAGTGATAACTCT-3’ |
GR consensusb
| 5’-AGAGGATCTGTACAGGATGTTCTAGAT-3’ |
NF1 consensusb
| 5’-TTTTGGATTGAAGCCAATATGATA-3’ |
NF-κB consensusb
| 5’-AGTTGAGGGGACTTTCCCAGGC-3’ |
SP1 consensusb
| 5’-AATCGATCGGGGCGGGGCGAG-3’ |
C/I −855/−835c
| 5’-GCATACAAAGGGATATAAACA-3’ |
M −853/−833c
| 5’-GCATACAAAGTGATATAAAC-3’ |
C/I −1245/−1225c
| 5’-AGCTTCCTCACGGCCTGGATG-3’ |
M −1243/−1223c
| 5’-AGCTTCCTCAGGGCCTGGATG-3’ |
C/I −1700/−1680c
| 5’-AACCCCATATTTCCACTGAGA-3’ |
M −1676/−1656c
| 5’-AACCCCATATCTAGGCACTAA-3’ |
Plasmids
Using primers specific for the porcine GnRHR gene promoter originally isolated from the Control line [
30], we sequenced promoters from genomic DNA of the Meishan and Index lines. Full-length GnRHR gene promoters (−5118) from the three genetic pig lines were sub-cloned into the pGL3 basic reporter vector (Promega Corp., Madison, WI). Studies involving constructs containing progressively less 5’ flanking sequence of the GnRHR gene promoter for all three lines of pigs were generated by restriction endonuclease digestion of vectors containing the full-length GnRHR gene promoter for each line and subsequent intramolecular ligation of the remaining vector backbone (
PvuII, SpeI and
BlpI). The promoter “swap” vectors containing full-length GnRHR gene promoter with the region from the −1915 to −1431 bp exchanged between Control and Meishan promoters was constructed from vectors containing 5118 bp of native sequence. Restriction endonuclease digestion of the internal 484 bp and subsequent ligation of the corresponding region for the promoter of the other line of swine was performed. Overlap extension PCR mutagenesis was performed through two rounds of PCR in order to specifically mutate the binding element of interest [
35]. The first round of PCR utilized primers replacing the binding site of interest with a restriction site, and the second round used product from the first round as template to anneal and replicate the mutated element and flanking sequence (Table
2). The mutation of the SF1 binding sites located at −179/−171 in each of the promoters was performed with the same set of primers and generated a
NotI restriction enzyme site. The −MμGATAUEpGL3, −MμNF-κBpGL3, −MμSP1pGL3 and −MμGATA4pGL3 plasmids were composed of 5118 bp of 5’ flanking sequence for the Meishan GnRHR gene with individual elements mutated to contain either a
SpeI (−MμGATAUEpGL3),
NsiI (−MμNF-κBpGL3),
SpeI (−MμSP1pGL3) or
PstI (−MμGATA4pGL3) site. The −Mμ1240pGL3 plasmid contained a double mutation of the NF-κB and SP1 sites discussed above. The −CμNF-κBpGL3 plasmid was made by substituting the NF-κB site for
EcoRI. To verify that the correct mutations had been introduced, vectors were sequenced before use in transient transfection experiments. The vector used as a control for transfection efficiency in all experiments contained the Rous Sarcoma Virus (RSV) promoter fused to the cDNA encoding β-galactosidase (RSV-βgal, Stratagene, La Jolla, CA). A midi plasmid preparation kit (Qiagen, Valencia, CA) was used to isolate transfection quality DNA.
Table 2
Primers used to generate reporter vectors
−5118pGL3 F | 5’-CAGACAATTAGATTCCAGGGC-3’ |
Promoter R | 5’-TCCTTCCCCAACTGATGTAG-3’ |
μSF1pGL3 Fa
| 5’-AAGTACACAAAACAAGTTGCGGCCGGCTCTTTCACATTAAATATA-3’ |
proximal A OF | 5’-GTTATGTGGAAGAGCCGGTG-3’ |
proximal OR | 5’-CTTTATGTTTTTGGCGTCTTCC-3’ |
MμGATAUEpGL3Fa
| 5’-TTGCAGAAACCTAACCCCACTAGTAGGCACTAATCCAGTGTC-3’ |
CμNF-κBpGL3 Fa
| 5’-TTGGCTTGCAGAAACCTAGAATTCTATTTCCACTGAGAGCAA-3’ |
distal OF | 5’-CAGAGAATGCTATTGCTCTC-3’ |
distal OR | 5’-GTGTAAGTGTTGGAACCACATC-3’ |
Mμ1240pGL3 Fa
| 5’-CATAGCACCAAGGAAGCTATGCATACTAGTGGATGATACTGTGTGCAG-3’ |
proximal B OF | 5’-AGGCACTAATCCAGTGTCTGC-3’ |
proximal OR | 5’-CTTTATGTTTTTGGCGTCTTCC-3’ |
MμSP1pGL3 Fa
| 5’-ACCAAGGAAGCTTCCTCAACTAGTGGATGATACTGTGTGCAG-3’ |
MμNF-κBpGL3 Fa
| 5’-CATAGCACCAAGGAAGCTATGCATGGGCCTGGATGATACTGT-3’ |
MμGATA4pGL3 Fa
| 5’-ATTAGATTGCATACAAAGCTGCAGAAACAAATATTCATATTA-3’ |
proximal C OF | 5’-TACTCCTCTTGATTTCTGACTC-3’ |
proximal OR | 5’-CTTTATGTTTTTGGCGTCTTCC-3’ |
Cell culture and transient transfections
Cultures of αT3-1 cells (Dr. Pam Mellon, Salk Institute, La Jolla, CA) were maintained at 37 °C in a humidified 5 % CO
2 in air atmosphere. The αT3-1 cells were cultured in high-glucose DMEM (4.5 g/L; Mediatech, Herndon, VA) supplemented with 5 % fetal bovine serum, 5 % horse serum, 2 mM glutamine, 100 U/ml penicillin and 100 μg/ml streptomycin sulfate (Gibco, Grand Island, NY). Transient transfections were carried out using a liposome-mediated protocol (Fugene6, Roche Diagnostics Corp., Indianapolis, IN) according to manufacturer’s instructions. Briefly, 2 × 10
6 cells were plated in 6-well culture dishes 1 d prior to transfection. Cells were transfected with a 3:1 Fugene6 to DNA ratio. A total of 1 μg of DNA, 0.9 μg of luciferase test vector and 0.1 μg of RSV-βgal control vector were used per well. Approximately 20–24 h post-transfection, cells were washed twice with ice-cold PBS and harvested with 200 μl of lysis buffer [100 mM potassium phosphate (pH 7.8), 0.2 % Triton X-100 and 1 mM dithiothreitol (DTT)]. Lysates were cleared by centrifugation at 14,000 X
g for 2 min at 4 °C. Lysates (20 μl) were immediately analyzed according to manufacturer’s instructions for both luciferase (Promega Corp.) and β-gal (Applied Biosystems, Bedford, MA) activity using a Wallac Victor
2 microplate reader (PerkinElmer Life Sciences, Boston, MA). Luciferase values were divided by β-gal values to adjust for transfection efficiency. The raw data for all transfections utilized in this study have been included (Additional file
1).
EMSA
Nuclear protein extracts were obtained from approximately 2.8 × 108 αT3-1 cells using the NE-PER® Nuclear and Cytoplasmic Extraction Reagents Kit (Pierce Biotechnology, Rockford, IL). The nuclear extracts were treated with protease (catalog no. P8340; Sigma Chemical Co., St. Louis, MO) and phosphatase (catalog no. 524625; Calbiochem, La Jolla, CA) inhibitor cocktail solutions to prevent enzymatic degradation of proteins. The amount of protein present in the extracts was determined using bicinchoninic acid (BCA assay, Pierce Biotechnology). Oligonucleotides were end-labeled with [γ-32P]ATP using T4 polynucleotide kinase (Fermentas Inc., Hanover, MD) and purified using sephadex G-25 spin columns (Amersham Biosciences Corp., Piscataway, NJ). EMSAs were completed through incubation of nuclear extracts (5 μg) in 20 μl reactions containing 4 μl of Dignam D buffer (20 mM HEPES, 20 % glycerol, 0.1 M potassium chloride, 0.2 mM EDTA and 0.5 mM DTT), 1 mM DTT, 2 μg of poly(dI•dC) (Amersham Biosciences) and, where indicated, a rabbit polyclonal antibody directed against the p65 (Calbiochem), p52 (Upstate, Lake Placid, NY), or p50 (Santa Cruz Biotechnology) subunits of NF-κB; GATA-1, -2 and -4 (Santa Cruz Biotechnology); SP1 (Upstate), SP1, SP2, SP3 and SP4 (Santa Cruz Biotechnology) or an equal mass of rabbit IgG (Santa Cruz Biotechnology) at 4 °C for 2 h. Following incubation, radiolabeled probe (100,000 cpm) and 50-fold molar excess of either homologous or heterologous unlabeled competitor was added. Where indicated, 50-fold molar excess of unlabeled oligonucleotides containing consensus binding sequences for AP2, NF-κB, SP1, glucocorticoid receptor (GR), nuclear factor (NF)-1 or GATA-4 were also added. The final reactions were incubated at 25 °C for 20 min before bound probe was separated from free at 30 mA for 1.5 h on a 5 % polyacrylamide gel that had been prerun at 100 V for 1 h in 1X TGE [25 mM Tris (pH 8.3), 190 mM glycine and 1 mM EDTA]. Gels were transferred to blotting paper, dried, and exposed to Biomax MS film (Eastman Kodak Co., Rochester, NY) for 20–24 h at −80 °C before being developed.
Western blot
Nuclear proteins from αT3-1 cells were extracted using the Nuclear Complex Co-IP kit from Active Motif (Carlsbad, CA), quantitated with a BCA protein assay kit (Pierce) and stored at −80 °C. Protein samples (40 μg) were boiled for 5 min in a 2X reducing loading buffer (130 mM Tris pH 6.8, 4 % SDS, 0.02 % Orange G, 20 % glycerol, 100 mM DTT), cooled to room temperature (RT) and loaded onto an SDS polyacrylamide gel (PAGE) with a 5 % stacking and 10 % resolving gel. Gels were run at 40 mA for approximately 90 min and electrophoresed proteins were transferred to polyvinylidene difluoride (PVDF, Immobilon -FL, Millipore, Billerica, MA) membrane with a semi-dry electroblotter (Panther, Owl Separation Systems, Portsmouth, NH). Briefly, PVDF membrane was pre-wetted in 100 % methanol and soaked with the gel in transfer buffer (25 mM Tris pH 8.3, 192 mM glycine, 0.1 % SDS, 20 % methanol) for 15 min. The proteins were transferred at 200 mA for 1 h. Membranes were blocked with StartingBlock™ TBS buffer (Pierce) for 30 min at room temperature with agitation. Incubation of primary antibodies directed against the p50 (Santa Cruz Biotechnology), p52 (Upstate) and p65 (Calbiochem) subunits of NF-κB were performed in StartingBlock™ TBS buffer supplemented with 0.05 % Tween-20. Antibodies were used at 1:500 (p50) or 1:5000 (p52 and p65) dilutions. Blots were incubated with primary antibody overnight at 4 °C with gentle shaking. After incubation, the blots were washed four times with TBS-T (20 mM Tris pH 7.6, 137 mM sodium chloride, 0.1 % Tween-20). Each wash was performed for 5 min with gentle agitation. The secondary antibody, Alexa Fluor 680 goat anti-rabbit IgG (A21076, Invitrogen, Carlsbad, CA) was diluted 1:15,000 in StartingBlock™ TBS buffer (Pierce) supplemented with 0.01 % SDS and 0.05 % Tween-20. The incubation was performed at RT for 1 h with gentle shaking. Blots were washed four times with TBS-T for 5 min with gentle agitation. After a final rinse with TBS, blots were scanned on the 700 channel of the Odyssey Infrared Imaging System (LI-COR, Lincoln, NE) following manufacturer’s instructions.
Statistical analysis
Data were analyzed using the general linear models (GLM) procedure of the Statistical Analysis System (SAS, version 8.2, Cary, NC). To control for transfection efficiency, the arbitrary light value for each replicate was divided by the respective β-gal value. These values were then divided by the mean of the empty vector and reported as fold activity over pGL3. All transfections were performed a minimum of three times, with samples in triplicate using different plasmid preparations for each transfection. Individual values from all the replicates were used to generate the mean ± SEM. Comparisons between pGL3 and test vectors were evaluated with Dunnett’s t-test. Least squares means for luciferase activity were compared among test vectors using least significant differences.
Discussion
Here, we have demonstrated a dramatic variance in activity among reporter vectors containing GnRHR promoters from the Control, Index and Meishan lines of pigs using transient transfections in the gonadotrope-derived αT3-1 cell line. Divergent activity of the Control, Index, and Meishan promoters illustrates a potential alteration in the mechanisms underlying transcriptional regulation of the porcine GnRHR gene among genetic strains. Ultimately, differential regulation of GnRHR gene expression may be correlated with divergent ovulation rates as well as other reproductive traits observed among the Control, Index and Meishan pig lines. This characteristic difference in activity among pig strains was lost after promoter constructs were reduced from −1431 to −1004 bp of 5’ flanking region, inferring that the element(s) responsible for the elevated luciferase activity of the Meishan promoter construct is located within the proximal 1431 bp of GnRHR promoter. Within gonadotrope-derived cell lines, the majority of elements responsible for basal and hormonally-induced expression of the GnRHR gene in other species are located within 1000 bp of proximal promoter [
10‐
19,
22‐
28,
36,
37], although placental-, granulosa/luteal cell- and neuronal-specific promoters have been identified further upstream in the human [
38‐
40]. On the other hand, the spatial arrangement of the porcine GnRHR promoter is somewhat unique because it requires approximately 1800 bp of 5’ flanking sequence for basal activity in αT3-1 cells [
30]. Despite the fact that the elements which confer line-specific activity of the porcine GnRHR promoter lie further upstream than 1000 bp, they still remain within the boundaries of the gonadotrope-specific promoter.
Transcriptional regulation of the GnRHR gene in different species is achieved through a variety of mechanisms and a number of different transcription factors [
8,
9]. One of the most characterized mechanisms for transcriptional regulation of the GnRHR gene involves the orphan nuclear receptor, SF1, known to be vital for gonadotrope-specific expression of the GnRHR gene in the human [
26], mouse [
11], rat [
24], sheep [
29] and pig [
30]. SF1 binding is also known to confer expression of the gonadotropin subunit genes within gonadotrope cells [
41‐
43]. Previously, our laboratory demonstrated that three SF1 binding sites are involved in transcriptional regulation of the porcine GnRHR gene promoter in αT3-1 cells. In reporter assays, mutation of the proximal SF1 binding site at −179/−171 bp resulted in complete ablation of promoter activity, indicating that this site is essential for gene expression [
30]. In this study, however, we have shown that the critical proximal SF1 binding site is not responsible for line-specific regulation of the GnRHR gene in swine (Fig.
2).
Identification of an NF-κB site involved in line-specific expression of the porcine GnRHR gene in gonadotrope cells is unique. While NF-κB has traditionally been associated with genes involved in immune and inflammatory responses [
44], it has been implicated in mediating the apoptotic effects of GnRH in ovarian cancer cells [
45]. As reviewed by Hayden and Ghosh [
46], NF-κB typically exists as a heterodimer composed of p65 (RelA), p50, p52, RelB and/or c-Rel. Although p52/p65 and p52/RelB heterodimers are prevalent, most commonly the heterodimer is comprised of the p65 and p50 subunits. Previously we reported that p52/p65 subunits are involved in the transcriptional regulation of the porcine GnRHR2 gene in the testis [
47]. Regarding gonadotrope cells of the anterior pituitary gland, however, NF-κB remains noticeably absent from the cast of transcription factors known to regulate the GnRHR or gonadotropin subunit genes. In this study, we detected p52 and p65 proteins within αT3-1 nuclear extracts and established the role of p52/p65 heterodimer binding in transcriptional regulation of the GnRHR gene. Consistent with this, NF-κB is an important regulator of Cox-2 promoter activity in the gonadotrope-derived cell line, LβT2, and treatment with GnRH stimulated phosphorylation of the p65 subunit by 22-fold [
48]. In addition, NF-κB binding sites have also been associated with regulation of genes expressed in other pituitary cell types including somatotropes [
49] and corticotropes [
50].
The −1240/−1230 region of the Meishan GnRHR promoter also appears to bind another transcription factor capable of interacting with the SP1 consensus binding site. Although the use of an SP1 consensus oligonucleotide revealed competition for DNA-protein binding (Fig.
4), the inability of SP1-specific antibodies acquired from two separate commercial vendors to bind the DNA-protein complex (Fig.
4) suggests that an alternative transcription factor recognizes the SP1 consensus binding sequence. To further verify the integrity of the antibodies directed against SP1, we also performed EMSAs with αT3-1 nuclear extracts and radiolabeled oligonucleotide containing consensus binding sites for SP1. In this instance, a specific complex was formed and the addition of SP1-specific antibodies resulted in supershifted DNA-protein complexes (data not shown), confirming the presence of SP1 in nuclei of αT3-1 cells and effectiveness of the SP1 antibodies. Another concern was whether the SP1 consensus oligonucleotides actually contained one or more NF-κB elements and therefore, was merely mimicking the NF-κB consensus oligonucleotide. However, sequence analysis of the oligonucleotide containing consensus SP1 binding sites did not reveal any NF-κB elements. Next, we examined other members of the SP1 family of transcription factors. Of primary interest, SP3 and SP4 recognize the same binding sequence with similar affinities [
51]. However, the addition of antibodies directed against SP2, SP3 and SP4 were unable to bind to the specific complex. Thus, we were able to eliminate SP1-4 as potential transcription factors binding to the SP1 element located within −1240/−1230 of the Meishan GnRHR promoter. Currently, nine members of the SP1 family (SP1-9) have been identified [
52,
53], and represent a subgroup of a larger class of transcription factors, the SP1-like/Krüppel-like factor (KLF) family [
54]. These factors share a highly conserved DNA-binding domain containing three Cys
2/His
2 zinc fingers [
51], which is the most abundant transcription factor motif in the human genome [
55]. In fact, over 25 SP1-like/KLF genes have been reported in mammals [
56]. The nine SP transcription factors can further be divided into the SP1-like family (SP1-4) and SP8-like family (SP5-9) [
57]. While SP1 and SP3 are ubiquitously expressed, SP5-9 expression patterns are more specific and temporal. SP5 and SP8 expression has been linked with Wnt activity and are important for stem cell differentiation and early embryonic development [
58,
59]. SP6, also known as epiprofin, is involved in epidermal differentiation [
60] whereas SP7, known as osterix, is expressed in osteoblasts [
61]. Additionally, SP9 is also expressed embryonically and affects limb outgrowth [
62]. Given the developmental roles that SP5-9 frequently play, it is unlikely that any of these factors are binding the GnRHR gene promoter. Thus, more studies are required to determine the identity of the factor(s) binding to the SP1 site within the −1240/−1230 promoter region of the Meishan GnRHR gene.
Mutation of the individual binding sites for NF-κB and the SP1-like factor within the Meishan GnRHR promoter demonstrated a significant loss of promoter activity. In addition, mutation of both the NF-κB and the SP1 recognition sequences (−Mμ1240pGL3) diminished luciferase activity to approximately the same level as either single block mutation. Due to the lack of further reduction of luciferase activity by the double block replacement mutation, it would appear that the transcription factors binding to the two elements work synergistically, but not additively, to stimulate increased GnRHR promoter activity in the Meishan line of swine. Indeed, NF-κB interacts with a variety of other transcription factors including AP1, estrogen receptor α, C/EBP, SF1 and SP1 [
63‐
66]. The Nabel laboratory determined that SP1 interacts with p65 via the DNA binding regions of each factor and that this interaction is necessary for activation of the HIV-1 gene [
67]. Despite the importance of the complex of transcription factors binding at −1240/−1230, none of the mutations lowered Meishan GnRHR promoter activity to that of the native full-length Control promoter suggesting another element(s) within the Meishan GnRHR promoter also confers line-specific expression.
We also identified 2 additional single-bp alterations located at −845 (G → T) and at −1690 (T → C) within the Meishan GnRHR promoter region that allowed binding of GATA-4 to the recognition sites. While GATA-4 has not previously been implicated in transcriptional regulation of the GnRHR gene, it is involved in the expression of other gonadotropic genes. A GATA motif detected within the human α-subunit gene promoter binds GATA-2 and a GATA-4-related protein in αT3-1 cells [
68]. These investigators were unable to confirm GATA-4 binding because specific antibodies directed against GATA-4 were not commercially available at the time. Consistent with these results, our study confirmed the presence of GATA-4 in gonadotrope-derived αT3-1 cells and implicated its importance in regulation of gonadotropic gene expression. Steger and coworkers [
68] also reported that the same GATA element binds GATA-2 and -3, but not GATA-4-related protein, in placental-derived cell lines. Binding of GATA-4 also regulates other genes essential to reproduction [
69]. In the neuronal GT1-7 cell line, GATA-4 binds recognition sites within the promoter for the GnRH gene [
70]. Transcription of the Müllerian inhibiting substance (MIS) gene in Sertoli cells is enhanced by the direct interaction of GATA-4 and SF1, although GATA-4 binding to the DNA is not required for this synergistic effect [
71,
72]. Additionally, adrenal-specific transcription of the human P450c17 gene is regulated by the interaction of GATA-4 or GATA-6 with SP1 [
73]. However, despite its contribution to enhanced promoter activity of the Meishan GnRHR gene, the GATA-4 binding sites, like the NF-κB and SP1 elements, do not fully explain the increased activity of reporter constructs containing the Meishan compared to Control/Index promoters. Therefore, future studies in our laboratory will focus on identifying the remaining elements and corresponding binding factors that contribute to the enhanced activity of the Meishan GnRHR promoter in αT3-1 cells.
Abbreviations
AP1, activator protein 1; CRE, cAMP responsive element; EMSA, electrophoretic mobility shift assay; FOXL2, forkhead box L2; FSH, follicle stimulating hormone; GnRH, gonadotropin-releasing hormone; GnRHR, gonadotropin-releasing hormone receptor; GnSE, GnRHR specific enhancer; GRAS, GnRH receptor activating sequence; GSE, gonadotrope specific element; JNK, c-Jun N-terminal kinase; LH, luteinizing hormone; LHX3, LIM homeobox 3; NF-Y, nuclear factor Y; NF-κB, nuclear factor-kB; OCT1, octamer transcription factor 1; PCR, polymerase chain reaction; Pitx-1, pituitary homeobox 1; PKA, protein kinase A; PKC, protein kinase C; RSV, Rous sarcoma virus; RXR, retinoid X receptor; SAP, SF1 adjacent protein; SF1, steroidogenic factor 1; SP, specificity protein; SURG-1, sequence underlying responsiveness to GnRH
Acknowledgments
The αT3-1 cells were a generous gift from Dr. Pamela Mellon (Salk Institute, La Jolla, CA). We would like to thank Dr. Colin Clay (Colorado State University, Ft Collins, CO) for the –m600pGL3 plasmid. Genomic DNA from the Meishan line of swine was kindly provided by Dr. Matthew B. Wheeler (University of Illinois, Urbana, IL). The authors would like to thank Rachel Friedrich, Chanho Lee and Jocelyn Wiarda for their assistance with completion of this study, as well as Dr. Janos Zempleni, Department of Nutrition and Health Sciences at UNL, for the use of his film developer. A contribution of the University of Nebraska Agricultural Research Division, Lincoln, Nebraska 68583. Journal Series No. 15039.