Artigo Acesso aberto Revisado por pares

Steroidogenic Factor-1 Is an Essential Transcriptional Activator for Gonad-specific Expression of Promoter I of the Rat Prolactin Receptor Gene

1997; Elsevier BV; Volume: 272; Issue: 22 Linguagem: Inglês

10.1074/jbc.272.22.14263

ISSN

1083-351X

Autores

Zhang-Zhi Hu, Li Zhuang, Xin‐Yuan Guan, Jianping Meng, Maria Dufau,

Tópico(s)

Sexual Differentiation and Disorders

Resumo

The expression of the prolactin receptor is under the control of two putative tissue-specific (PI, gonads; PII, liver) and one common (PIII) promoters (Hu, Z. Z., Zhuang, L., and Dufau, M. L. (1996) J. Biol. Chem.271, 10242–10246). The three promoter regions were co-localized to the rat chromosomal locus 2ql6, in the order 5′-PIII-PI-PII-3′. To investigate the mechanisms of gonad-specific utilization of PI, the promoter domain, regulatory cis-elements, and trans-factors were identified in gonadal cells. The promoter domain localized to the 152-base pair 5′ of the transcriptional start site at −549 is highly active in gonadal cells but has minimal activity in hepatoma cells. It contains a steroidogenic factor 1 (SF-1) element (−668) that binds the SF-1 protein of nuclear extracts from gonadal cells and is essential for promoter activation. A CCAAT box (−623) contributes minimally to basal activity in the absence of the SF-1 element, and two adjacent TATA-like sequences act as inhibitory elements. Thus, PI belongs to a class of TATA-less/non-initiator gene promoters. These findings demonstrate an essential role for SF-1 in transcriptional activation of promoter I of the prolactin receptor gene, which may explain the tissue-specific expression of PI in the gonads but not in the liver and the mammary gland. The expression of the prolactin receptor is under the control of two putative tissue-specific (PI, gonads; PII, liver) and one common (PIII) promoters (Hu, Z. Z., Zhuang, L., and Dufau, M. L. (1996) J. Biol. Chem.271, 10242–10246). The three promoter regions were co-localized to the rat chromosomal locus 2ql6, in the order 5′-PIII-PI-PII-3′. To investigate the mechanisms of gonad-specific utilization of PI, the promoter domain, regulatory cis-elements, and trans-factors were identified in gonadal cells. The promoter domain localized to the 152-base pair 5′ of the transcriptional start site at −549 is highly active in gonadal cells but has minimal activity in hepatoma cells. It contains a steroidogenic factor 1 (SF-1) element (−668) that binds the SF-1 protein of nuclear extracts from gonadal cells and is essential for promoter activation. A CCAAT box (−623) contributes minimally to basal activity in the absence of the SF-1 element, and two adjacent TATA-like sequences act as inhibitory elements. Thus, PI belongs to a class of TATA-less/non-initiator gene promoters. These findings demonstrate an essential role for SF-1 in transcriptional activation of promoter I of the prolactin receptor gene, which may explain the tissue-specific expression of PI in the gonads but not in the liver and the mammary gland. The functional diversity of prolactin, involving lactation and reproduction, growth and metabolism, immune regulation, behavior, and homeostasis, has been well documented in the past several decades (2Nicoll C.S. Knobil E. Saywer W.H. Handbook of Physiology.American Physiological Society, Washington, D. C. 1974; 4: 253-292Google Scholar,3Doppler W. Rev. Physiol. Biochem. Pharmacol. 1994; 124: 94-130Google Scholar). However, the mechanisms underlying the regulation of prolactin receptors present in specific target tissues have only recently been investigated. Prolactin receptors are widely expressed, and multiple mRNA transcripts corresponding to the long and short forms of the receptor are present with various proportions in different tissues (4Hu Z.Z. Dufau M.L. Biochem. Biophys. Res. Commun. 1991; 181: 219-225Crossref PubMed Scopus (24) Google Scholar,5Nagano M. Kelly P.A. J. Biol. Chem. 1994; 269: 13337-13345Abstract Full Text PDF PubMed Google Scholar). The diverse actions of prolactin could be manifested by the expression of different receptor forms and signal transduction pathways and also by differential control of gene transcription and mRNA regulation of PRLR 1The abbreviations used are: PRLR, prolactin receptor; SF-1, steroidogenic factor-1; 5′-RACE, rapid amplification of cDNA 5′ ends; RT-PCR, reverse transcriptase-polymerase chain reaction; bp, base pairs; kb, kilobase(s); DTT, dithiothreitol; LUC, luciferase; MLTC, mouse Leydig tumor cells; EMSA, electrophoresis mobility shift assay; C/EBP, CCAAT/enhancer-binding protein; TSS, transcriptional start site; Adv, adenovirus. in target tissues. Recently, the complexity of the mechanism by which the PRLR gene is controlled was revealed by the demonstration of multiple and tissue-specific promoters of the rat prolactin receptor gene (1Hu Z. Zhuang L. Dufau M.L. J. Biol. Chem. 1996; 271: 10242-10246Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar). Three putative promoters PI, PII, and PIII that initiate the transcription of alternative first exons E11, E12, and E13, respectively, were identified in the rat gonads and the liver. The three alternate first exons are alternatively spliced to a common noncoding exon 2 that precedes the third exon containing the translation initiation codon of the prolactin receptor. E11is expressed in ovaries and in Leydig cells, E12 in the liver, and E13 in all three tissues, indicating that PI is gonad-specific, PII is liver-specific, and PIII is a common promoter for the prolactin receptor in these tissues and is the sole promoter utilized in the rat mammary gland (our current study). The various prolactin receptor forms were found not to be dependent on the utilization of individual promoters, which may result from a post-transcriptionally regulated process. However, these accounted for the 5′-untranslated region heterogeneity of the first exons of the PRLR mRNA transcripts (E11, E12, and E13) (1Hu Z. Zhuang L. Dufau M.L. J. Biol. Chem. 1996; 271: 10242-10246Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar). The heterogeneity of the 5′-untranslated region of the PRLR transcripts was also reported recently in the gonads and the liver by others (6Moldrup A. Ormandy C. Nagano M. Murthy K. Banville D. Tronche F. Kelly P.A. Mol. Endocrinol. 1996; 10: 661-671PubMed Google Scholar, 7Rubtsov P.M. Lonina D.A. Mol. Biol. ( Moscow ). 1996; 30: 193-198Google Scholar). Sequence analyses of 5′-flanking regions showed a lack of consensus TATA box sequences positioned within the expected distance from the transcriptional start site (TSS) in all three putative promoter regions (1Hu Z. Zhuang L. Dufau M.L. J. Biol. Chem. 1996; 271: 10242-10246Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar). However, potentially functional non-canonical TATA box sequences are present in these regions, and in PI two adjacent TATA-like sequences reside 10 and 23 bp 5′ from the TSS. In addition, consensus sequences for several transcription factors are also present in these promoters, specifically SF-1, C/EBP, and CAAT box in PI; GATA-1, ERE and AP-1 in PII; and GAGA, GATA-1 and AABS in PIII. Therefore, it is assumed that different transcription factors and other cellular and extracellular regulators may be involved in the control and regulation of the PRLR gene in diverse target tissues. In the present study, we have determined the co-localization and the relative orientation of the three 5′-untranslated region/promoter regions in the rat chromosome, and we have investigated the underlying mechanisms of the tissue-specific promoter control of the PRLR gene in the gonads by the characterization of the gonad-specific promoter domain PI and its relevant cis-elements and trans-factors. These studies have demonstrated that steroidogenic factor-1 (SF-1) (8Luo X. Ikeda Y. Parker K.L. Cell. 1994; 77: 481-490Abstract Full Text PDF PubMed Scopus (1397) Google Scholar), also known as Ad4BP (9Honda S. Morohashi K. Nomura M. Takeya H. Kitajima M. Omura T. J. Biol. Chem. 1993; 268: 7494-7502Abstract Full Text PDF PubMed Google Scholar), is an essential transcriptional activator of PI in testicular and ovarian cells. Such regulation by SF-1 appears to determine the specific gonadal utilization of promoter I. Adult male, prepubertal female, and lactating Harlan Sprague Dawley rats (Charles River, Wilmington, MA) were housed in pathogen-free, temperature- and light-controlled conditions (20 °C; alternating light-dark cycle with 14 h of light and 10 h of darkness; lights on at 0600 h). The animals were given free access to tap water and standard Purina lab chow (Ralston-Purina, St. Louis, MO). All animals were killed between 1000 and 1100 h by asphyxiation with CO2. All animal studies were approved by the NICHD Animal Care and Use Committee (protocols 94-040 and 94-041). 40-day-old adult male rats were used for preparation of Leydig cells. 24-day-old immature female rats were injected subcutaneously daily for 3 days with 1.5 mg/day 17β-estradiol (dissolved in propylene glycol) (Sigma). On the 4th day, the animals were sacrificed for preparation of ovarian granulosa cells. The mammary gland of lactating female rats were resected and freed of surrounding adipose tissue and used for mRNA preparation. Cultures of mouse Leydig tumor cells (MLTC-1), a stable steroidogenic cell line that expresses prolactin and luteinizing hormone receptors (kindly provided by Dr. R. V. Rebois, National Institutes of Health, Bethesda, MD), were maintained in RPMI 1640 (BioWhittaker, Walkersville, MD) supplemented with 10% fetal bovine serum and 1 × antibiotic/antimycotic mixture (Life Technologies, Inc.), and the human hepatoma cell line (HepG2, American Type Culture Collection, Rockville, MD), which expresses prolactin receptors, was maintained in minimal essential medium (Life Technologies, Inc.) supplemented with 10% fetal bovine serum, non-essential amino acids, and sodium pyruvate. Rat ovarian granulosa cells were prepared from estrogen-primed rat ovaries as described previously (10Sirois J. Simmons D.L. Richards J.S. J. Biol. Chem. 1992; 267: 11586-11592Abstract Full Text PDF PubMed Google Scholar) with some modifications. Briefly, the ovaries were removed and cleared of the surrounding fat tissues. These were washed in Medium 199 (M199, BioWhittaker) containing 0.2% bovine serum albumin and 10 mm HEPES, pH 7.2 (Medium A), and punctured 20 times with a 23-gauge needle in Medium B (Medium A with additional 6.8 mm EGTA) followed by incubation in a CO2 incubator at 37 °C for 10 min and centrifugation at 700 rpm for 10 min. The ovaries were then incubated with Medium C (Medium A with additional 1.8 mm EGTA and 0.5 msucrose) for 6 min in CO2 incubator at 37 °C and centrifuged as above. To disperse the granulosa cells, the ovaries were pressed against a 40-mesh metal sieve with a spatula, and the dispersed cells were collected. The cells were treated with trypsin (Sigma) (20 mg/ml) for 1 min at 37 °C followed by addition of soybean trypsin inhibitor (Sigma) (300 μg/ml) and DNase I (Sigma) (100 μg/ml) for 4 min at 37 °C. The enzymes were removed by centrifugation, and the cells were washed twice with Medium A. Cells were suspended in medium Dulbecco's modified Eagle's medium/F12 (Life Technologies, Inc.) supplemented with 1% fetal bovine serum, and with 15 ng/ml ovine follicle stimulating hormone (ovine FSH-20, National Pituitary Program, NIDDK), 10 ng/ml testosterone (Sigma) and plated at a density of 0.5 × 106 cells/cm2 in 24-well plates and cultured at a CO2 incubator for 3–5 days. Rat Leydig cells were prepared from adult male rat testes by collagenase digestion and purified by centrifugal elutriation (11Aquilano D.R. Dufau M.L. Endocrinology. 1984; 114: 499-510Crossref PubMed Scopus (50) Google Scholar). The purified Leydig cells were suspended in M199 containing 0.1% bovine serum albumin and 1 × antibiotic/antimycotic and plated at the density of 0.5 × 106 cells/cm2 and cultured for up to 20 h. Lambda phage DNA with genomic PRLR DNA inserts of 15 and 14.5 kb were used for fluorescence in situ hybridization analysis. The probes were labeled with biotin (lambda phage clone λ11-1, 15 kb, containing PRLR PI/PII regions) or spectrum orange (lambda phage clone λ3a, 14.5 kb, containing PIII region) by nick translation (Life Technologies, Inc.) and hybridized to rat metaphase chromosomes as described previously (12Guan X.-Y. Meltzer P.S. Trent J.M. Genomics. 1994; 22: 101-107Crossref PubMed Scopus (92) Google Scholar). Briefly, about 200 ng of each probe was used in 10 μl of hybridization mixture containing 55% formamide, 2 × SSC, and 1 μg of humanCot I DNA (Life Technologies, Inc.) that was denatured at 75 °C for 5 min. Slides with rat metaphase chromosomal spreads were denatured in 70% formamide, 2 × SSC at 72 °C for 2 min, and hybridized with the specified probes at 37 °C in a moist chamber overnight. Slides were then washed three times in 50% formamide, 2 × SSC at 45 °C for 3 min each. The hybridization signal of the probe was detected by two layers of fluorescein isothiocyanate-conjugated avidin (Vector Laboratories, Inc, Burlingame, CA) and amplified with one layer of anti-avidin antibody (Vector). Slides were counterstained with 0.5 μg/ml 4,6-diamidino-2-phenylindole in an antifade solution (1 mg/mlp-phenylenediamine dihydrochloride, 10% phosphate-buffered saline (v/v), 90% glycerol (v/v), 4.2% sodium carbonate (w/v)) and was examined with a Zeiss Axiophot microscope equipped with a dual bandpass filter. For mapping of the plasmid DNA isolated from a P1 genomic library (Genome Systems, Inc., St. Louis, MO) containing the 5′-flanking regions of the rat PRLR gene, the plasmid was digested withBam HI or in combination with Not I followed by Southern blot hybridization with oligonucleotide probes derived from E11, E12, and E13. DNA fragments of the 5′-flanking region containing promoter I were either generated by restriction enzyme digestion (Kpn I/Xba I, −1566/−124) or by PCR amplification. For generation of plasmid constructs, the 5′-flanking DNA fragments of PI were ligated 5′ to the luciferase gene (LUC) of the linearized plasmid pGL2 (Promega, Madison, WI). The pGL2 constructs were numbered relative to the translation initiation codon (PI(−#/−#)/LUC). The recombinant adenovirus (Adv) luciferase reporter constructs containing the putative promoter region (5′-flanking region of the PRLR) 5′ to the luciferase gene [PI(−#/−#)/LUC/Adv] were prepared as described previously (13Alcorn J.L. Gao E. Chen Q. Smith M.E. Gerard R.D. Mendelson C.R. Mol. Endocrinol. 1993; 7: 1072-1085Crossref PubMed Scopus (77) Google Scholar). Briefly, the 5′-flanking region of PRLR-luciferase (5′PRLR/LUC) minigenes were excised from pGL2 plasmid constructs using Sma I/Bam HI and inserted to the Eco RV/Bam HI site of pAC plasmid, which contains partial sequences of the adenovirus 5 genome. The resulting pAC/PI/LUC constructs were co-transfected with plasmid pJM17, which contains the remainder of the adenovirus genome into human embryonic 293 cells. In vivo homologous recombination of the plasmids yields recombinant viral genome (PI/LUC/Adv) and subsequent generation of infectious viral particles. Viral plaques were isolated and propagated to a titer of 109 ml−1 in the 293 cells and were used for infection of primary cultures of rat gonadal cells. Structure of the fusion genes was verified by PCR amplification of the insert and the subsequent DNA sequence analysis. For RT-PCR analyses of RNA from MLTC, first strand cDNA was synthesized with random primers using Superscript reverse transcriptase (Life Technologies, Inc.) at 42 °C for 30 min. Primers used for PCR amplification of exons E11 and E13 were 5′-GTGGCCAGAGCCATGGACAG-3′ (forward) and 5′-AAACTCTTTCCTCGGAGGTCACTAG-3′ (reverse), 5′-TCTCAGAGACACGCGGCTG-3′ (forward) and 5′-TTCTGCTGGAGAGAAAAGTCTG-3′ (reverse), respectively. PCR fragments were resolved on 1.5% agarose gel. 5′-RACE PCR analyses of RNA from luciferase reporter plasmid PI(−1566/−124)/LUC transfected in MLTC were performed as described previously (1Hu Z. Zhuang L. Dufau M.L. J. Biol. Chem. 1996; 271: 10242-10246Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar). First strand cDNA was synthesized using primer GS1 (5′-AGCGGTTCCATCCTCTAG-3′) (+7 to +24 of the luciferase coding region) and 3′-end-tailed with dCTP using terminal deoxynucleotidyltransferase followed by PCR with primer GS2 (5′-CTTTATGTTTTTGGCGTCTTCCA-3′) (+44 to +66 of the LUC coding region) and dG-adaptor primer (5′-GCGAATTCTCGAGATCTGGGII GGGII GGGII G-3′), where I represents inosine. The PCR products were resolved on 1.5% agarose gel and subjected to Southern blot analyses using nested oligonucleotide probe GS3 (5′-TCTACCTAACCCGCCCACTGGTT-3′) within E11. Primers used for 5′-RACE PCR analyses of rat mammary gland PRLR mRNA were same as described previously (1Hu Z. Zhuang L. Dufau M.L. J. Biol. Chem. 1996; 271: 10242-10246Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar). Poly(A)+ RNAs from the rat ovary, Leydig cells, mammary gland, and MLTC were prepared and analyzed by Northern blot as described previously (1Hu Z. Zhuang L. Dufau M.L. J. Biol. Chem. 1996; 271: 10242-10246Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar, 4Hu Z.Z. Dufau M.L. Biochem. Biophys. Res. Commun. 1991; 181: 219-225Crossref PubMed Scopus (24) Google Scholar). Nuclear proteins from MLTC, rat granulosa cells, and rat Leydig cells were extracted as described (14Tsai-Morris C.H. Xie X. Wang W. Buczko E. Dufau M.L. J. Biol. Chem. 1993; 268: 4447-4452Abstract Full Text PDF PubMed Google Scholar) with some modifications. Briefly, cells were suspended in buffer A containing 10 mm HEPES, pH 7.9, 10 mm KCl, 1.5 mm MgCl, 0.1 mm EDTA, 0.5 mm DTT, 0.4 mm Pefabloc SC (Boehringer Mannheim), 2 μg/ml leupeptin (Sigma), and 2 μg/ml pepstatin A (Sigma) with additional 0.3 m sucrose and 2% Nonidet P-40 and homogenized 20 strokes with type B pestle. The nuclei were recovered by centrifugation of the cell homogenate through a 1.5 m sucrose cushion contained in buffer A and lysed in buffer B (10 mm HEPES, pH 7.9, 420 mm KCl, 1.5 mm MgCl, 0.1 mm EDTA, 10% glycerol, 0.5 mm DTT, 0.4 mm pefabloc SC, 2 μg/ml leupeptin, and 2 μg/ml pepstatin A). The nuclear lysate was dialyzed against buffer D (20 mm HEPES, pH 7.9, 100 mm KCl, 0.1 mm EDTA, 20% glycerol, 0.5 mm DTT, 0.4 mm pefabloc SC, 2 μg/ml leupeptin and pepstatin A) overnight at 4 °C. The concentration of the nuclear proteins was measured by the protein assay kit (Bio-Rad). Double-stranded DNA fragment corresponding to the region of PI at −827 to −440 was used for DNase I footprinting analysis. One strand end-labeled DNA probes were generated by first labeling the 5′ end with [γ-32P]ATP (3000 mCi/mmol, DuPont NEN) of either forward primer (5′-TGTCTGCCTCATGAGAATAC-3′) or reverse primer (5′-TCTACCTAACCCGCCCACTGGTT-3′) followed by PCR amplification of the DNA fragment with one labeled and one unlabeled primer. For each footprinting reaction, 5 × 104 cpm of the probe was added to 10–20 μg of nuclear protein in 50 μl of mixture containing 20 mm HEPES, pH 7.5, 50 mm KCl, 0.5 mm EDTA, 1 mm DTT, 5% glycerol and incubated at room temperature for 15 min. DNase I digestion was performed by adding 1 × DNase I buffer (25 mm NaCl, 10 mm HEPES, 5 mm MgCl, and 1 mmCaCl2) containing 1 unit of DNase I (Promega) and incubation for 1 min at 22 °C. The reaction was terminated by adding 10 μl of stop buffer (200 mm NaCl, 30 mmEDTA, 1% SDS, and 100 μg/ml yeast RNA) and 110 μl of phenol/chloroform. The digested DNA fragments were recovered by ethanol precipitation and resolved on 6% polyacrylamide-urea gel electrophoresis. Synthesized oligonucleotides were used for gel mobility shift assays unless otherwise indicated. The oligomers were annealed and 5′-end-labeled with [γ-32P]ATP (3000 mCi/mmol, DuPont NEN). 1–3 μg of nuclear protein was added to 20 μl of reaction containing 12 mm HEPES, pH 7.6, 60 mm KCl, 4 mmTris-HCl, 5% glycerol, 1 mm EDTA, 1 mm DTT, and 25 μg/ml polydeoxyinosinic deoxycytidylic acid on ice for 15–30 min followed by addition of 5 × 104 cpm of the probe for additional 15 min. For competition assay, unlabeled DNA sequences were added to the reaction 15 min prior to the addition of the probe. For supershift assay, SF-1 antibody 1 (a gift from Dr. Ken-ichirou Morohashi, Kyushu University, Japan) (15Morahashi K. Iida H. Nomura M. Hatano O. Honda S. Tsukiyama T. Niwa O. Hara T. Takakusu A. Shibta Y. Omura T. Mol. Endocrinol. 1994; 8: 643-653PubMed Google Scholar) or antibody 2 (8Luo X. Ikeda Y. Parker K.L. Cell. 1994; 77: 481-490Abstract Full Text PDF PubMed Scopus (1397) Google Scholar) (purchased from Upstate Biotechnology, Inc, Lake Placid, NY) was incubated with nuclear proteins for 30 min prior to the addition of the probe. DNA-protein complexes were resolved on 5% native polyacrylamide gel electrophoresis. MLTC cells were transfected at 50–70% confluency ∼48 h after plating usingN-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium methylsulfate (Boehringer Mannheim) as described previously (16Leventis R. Silvius J.R. Biochim. Biophys. Acta. 1990; 1023: 124-132Crossref PubMed Scopus (439) Google Scholar). Cells were harvested 40–60 h after transfection, and whole cell lysates were assayed for luciferase activity (17Alam J. Cook J.L. Anal. Biochem. 1990; 188: 245-254Crossref PubMed Scopus (407) Google Scholar). Since the primary rat gonadal cells were not amenable for transient expression of PRLR-PI/LUC reporter gene constructs by conventional liposome-mediated transfection in our experiments, the recombinant adenovirus infection method was applied in this study. Adenovirus reporter gene constructs (PI/LUC/Adv) were added at multiplicity of infection of 100 to the granulosa or Leydig cell culture and incubated for 20 h. Adenovirus construct containing the promoterless luciferase gene served as background control and a construct containing the SV40 promoter-luciferase gene as positive control. The granulosa cells were infected after 3 days of culture, and the infection was allowed to continue for 20 h before termination. The Leydig cells were infected 2 h after plating, and human chorionic gonadotropin was added to a final concentration of 20 ng/ml 6 h before termination. The infection was also allowed to continue for 20 h. Whole cell lysates were used for measurement of luciferase activity. The fluorescence in situ hybridization method was used to establish the chromosomal location of the rat PRLR gene and to determine whether the three PRLR gene promoter regions that direct transcription of alternative first exons reside at the same gene locus. The genomic probe (15 kb) containing the 5′-flanking region corresponding to both PI and PII promoter regions has located these regions at the chromosomal locus 2q16 (Fig. 1,above). The same finding was obtained using the genomic probe containing coding exons 4 and 5 of the rat PRLR (not shown). The PIII promoter region was also located at the identical locus on the same metaphase chromosome when probed with a genomic fragment (14.5 kb) containing the PIII region (Fig. 1, above). The order of the three promoters in the gene was determined by mapping overlapping genomic clones (Fig. 1, middle). The size of the DNA insert cloned from the rat P1 genomic library is ∼96 kb as determined by restriction digestion and agarose gel electrophoresis (Fig. 1,middle left). A 55-kb Bam HI fragment was hybridized by two oligomer probes derived from E11 (Fig. 1,middle right, lanes 1-3) but not from E12 (not shown) or E13 (lane 4).Lanes 2 and 3 show a 35-kb positive fragment generated from the 55-kb fragment with Not I digestion, which cleaves at a unique site of the 16-kb vector. Furthermore, aBam HI-digested fragment of 8 kb was specifically hybridized by the oligomer probe derived from E13 (lane 4). However, no hybridization of this genomic DNA was revealed by the oligomer probes derived from E12 (not shown). These results indicated that this 96-kb genomic fragment of the rat PRLR gene contained both regions corresponding to exons E11 and E13 as well as their putative promoter regions PI and PIII but not PII and exon E12. Since previously we have established that PI region is located 10 kb 5′ of PII region (1Hu Z. Zhuang L. Dufau M.L. J. Biol. Chem. 1996; 271: 10242-10246Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar), we could derive that the order of these 5′-flanking/promoter regions of the PRLR gene is 5′-PIII-PI-PII-3′, which spans over 20 kb in the genome. Due to the lack of appropriate rat gonadal stable cell lines, we have utilized a mouse Leydig tumor cell line (MLTC-1) (18Rebois R.V. J. Cell Biol. 1982; 94: 70-76Crossref PubMed Scopus (109) Google Scholar) for investigation of the transcriptional control of the PRLR gene promoters. For these studies, it was necessary to initially characterize the endogenous expression of PRLR alternative exons in these cells. Northern blot analysis showed a major mRNA transcript of 9.5 kb comparable to the 9.7 kb of the rat species using 5′-coding region of the rat PRLR as the probe (Fig. 2 A, left). In addition, minor transcripts of 4.2, 2.4, and 1.4 kb were also revealed in MLTC. Based on the high similarity between the rat and the mouse PRLR cDNAs, primers derived from the rat E11and E13 were used for RT-PCR analyses of mRNA from rat ovaries and MLTC cells. PCR products of identical sizes for E11 (325 bp) (Fig. 2 B, lanes 2–4) and for E13 (229 bp) (lanes 5–7) were observed for the rat ovary and MLTC (mouse), indicating analogous expression of alternative first exons E11 and E13 in both species. Furthermore, recent genomic cloning of the mouse E13 and its 5′-flanking region showed 94% sequence similarity between the rat and the mouse within the noncoding exon E13 and the proximal 5′-flanking region (not shown). These results indicate that as in rat gonads, the PRLR expression is also under the control of putative promoters PI and PIII in MLTC cells. The promoter utilization of the PRLR gene in rat mammary gland was also examined. Northern blot analyses showed that the major mRNA transcript in this tissue is the 1.8-kb species (Fig. 2 C,left), which is also seen in the Leydig cells (right) and the liver (1Hu Z. Zhuang L. Dufau M.L. J. Biol. Chem. 1996; 271: 10242-10246Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar), and corresponds to the short form of the receptor (1Hu Z. Zhuang L. Dufau M.L. J. Biol. Chem. 1996; 271: 10242-10246Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar, 4Hu Z.Z. Dufau M.L. Biochem. Biophys. Res. Commun. 1991; 181: 219-225Crossref PubMed Scopus (24) Google Scholar). 5′-RACE PCR analyses of mRNA from the rat mammary gland showed that only E13 sequences were present in this tissue by sequencing the PCR products (not shown) and by Southern blot analysis of the PCR products (Fig. 2 D). Neither E11 nor E12 sequences were found to be present in the mammary tissue by Southern hybridization by oligonucleotide probes derived from E11 (lane 3 and 4) or E12 (lane 5 and6). This result indicated that PIII is the sole promoter utilized in the rat mammary gland and further demonstrated that the usage of PI is restricted to the gonads among the tissues examined. A luciferase reporter gene construct containing the 5′-flanking region with adjacent E11 sequence [PI(−1566/−124)/LUC] was examined for its promoter activity in the steroidogenic MLTC and the non-steroidogenic cell line HepG2. This genomic fragment exhibited strong promoter activity, which was 43-fold over that of the promoterless vector and about 60% that of the SV40 promoter in MLTC. In contrast, only minimal induction was observed in HepG2 cells, where the SV40 promoter displayed activities comparable to those observed in MLTC cells (Fig. 3, left). This finding indicated the specificity of activation of promoter I in MLTC and is consistent with the pattern of promoter utilization in vivo, where PI activity was found only in gonads but not in the liver (1Hu Z. Zhuang L. Dufau M.L. J. Biol. Chem. 1996; 271: 10242-10246Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar) and the mammary gland (Fig. 2 D). The transcription initiation site from the PI(−1566/−124)/LUC construct transfected in MLTC was determined by the 5′-RACE PCR method. An extended product of ∼540 bp was revealed on Southern hybridization of the PCR products by a nested probe (GS3, Fig. 3, right, above), and the transcriptional initiation site derived from this experiment was consistent with that reported for the rat ovary (−549 relative to translation initiation codon at +1) (1Hu Z. Zhuang L. Dufau M.L. J. Biol. Chem. 1996; 271: 10242-10246Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar) (Fig. 3, right, below). These results indicate that MLTC is an appropriate cell line for characterization of the regulatory components of this gonad-specific promoter (PI), since the PI/LUC construct is highly active and faithfully transcribed in MLTC. The PRLR PI/LUC constructs with serial deletion of the 5′-flanking region were expressed in MLTC to localize the minimal promoter domain. Deletion of the 5′-flanking sequences beyond −700 had no significant effect on luciferase activity (Fig. 4). The region between −700 and −549 was required for promoter activity, whereas deletion of the downstream exon 1 sequences (−124 to −549) had no significant effect on promoter activity. Thus, the 152-bp region between −700 and −549 retained full promoter activity in transfected MLTC cells. In contrast, the PI(−1566/−124)/LUC construct (Fig. 3) and the deletion mutants of PI including the minimal promoter domain −700/−549 (not shown) exerted only minor promoter activities in HepG2 cells, further indicating the tissue specificity of promoter I for expression in gonadal cells. The regulatory cis-elements and transcriptional activators required for this promoter were further analyzed by the DNA-protein binding analyses and site-directed mutagenesis of regulatory elements within the promoter domain. DNase I footprinting analysis using nuclear extracts from MLTC was employed to examine the specific nu

Referência(s)
Altmetric
PlumX