Regulation of steroidogenic acute regulatory protein transcription in largemouth bass by orphan nuclear receptor signaling pathways

Abstract

The steroidogenic acute regulatory (StAR) protein mediates the rate-limiting step of mitochondrial transport of cholesterol for steroid biosynthesis. To investigate the regulation of this protein in lower vertebrates, we cloned the StAR coding region from large-mouth bass for analysis. Induction of the mRNA corresponded with increasing levels of plasma sex steroids in vivo. Cultures of largemouth bass ovarian follicles were exposed to dibutyryl cAMP (dbcAMP), a potent signaling molecule for steroidogenesis. StAR mRNA expression was significantly up-regulated by dbcAMP signaling, suggesting that the 5' regulatory region of the gene is functionally conserved. To further analyze its transcriptional regulation, a 2.9-kb portion of the promoter was cloned and transfected into Y-1 cells, a steroidogenic mouse adrenocortical cell line. The promoter activity was induced in a dose-responsive manner upon stimulation with dbcAMP; however, deletion of 1 kb from the 5' end of the promoter segment significantly diminished the transcriptional activation. A putative retinoic acid-related receptor-alpha/rev-erb alpha element was identified between the -1.86- and -2.9-kb region and mutated to assess its potential role in dbcAMP regulation of the promoter. Mutation of the rev-erb alpha element significantly impeded dbcAMP-induced activation. Chromatin immunoprecipitation and EMSA results revealed rev-erb alpha and retinoic acid-related receptor-alpha enrichment at the site under basal and dbcAMP-induced conditions, respectively. These results implicate important roles for these proteins previously uncharacterized for the StAR promoter. Altogether these data suggest novel regulatory mechanisms for dbcAMP up-regulation of StAR transcription in the distal part of the largemouth bass promoter.

Figure 1

Seasonal expression of LMB StAR mRNA and plasma E₂. RNA was isolated from ovarian tissue of LMB previously collected every few weeks through the reproductive cycle (n = 7 for the first 10 time points and n = 6 for April 11, 2000) and in vivo expression of StAR mRNA was analyzed by Q-PCR. Data are reported on a logarithmic scale as mean copy number ± mean se. E₂ levels were quantified from the plasma of the same individuals by RIA as described previously (43).

Figure 2

dbcAMP induction of StAR mRNA in LMB ovarian follicle cultures. A, Hematoxylin-eosin histology of a vitellogenic LMB ovarian follicle. The following are labeled: germinal epithelium (GE), germinal vesicle (GV), yolk globules (YG), and oil droplets (OD). B, Representative ovarian follicle culture. Ten follicles were cultured per well in a 24-well culture plate and exposed to three doses of dbcAMP for 14 h. RNA was isolated from the follicles and reverse transcribed for analysis by real-time Q-PCR. This graph represents data obtained from one fish, with each treatment cultured in triplicate. Error bars, se between wells. This experiment was repeated three times with follicles from three individual fishes. Student’s t test was used to test for significance. *, P < 0.05.

Figure 3

A, Dose-response exposure of LMB StAR promoter to dbcAMP. Y-1 cells were transfected with the 2.9-kb LMB StAR promoter and exposed to increasing doses of dbcAMP from 0 to 2 mm dbcAMP. Graph is representative of one experiment done in triplicate that was repeated at least three individual times. B, Exposure of LMB StAR promoter to hCG. To further verify responses observed in Y-1 cells, MA-10 cells were transfected with the 2.9-kb LMB StAR promoter and exposed to 10 U/ml hCG. C, LMB StAR promoter constructs containing a deletion (∼1 kb from distal end of promoter) or a site-directed mutation (RORE/−1969) were evaluated for basal and dbcAMP stimulation of StAR promoter activity. C, Data collected from three to five individual experiments. In all experiments, values are normalized to renilla luciferase (internal control) and are reported in fold change from basal control. Student’s t test was used to determine significance. *, P < 0.05. D, In silico comparison of the StAR promoter across species. Promoter sequences (including the 5′ UTR) for the StAR gene promoters from LMB (accession no. DQ166819), brook trout (accession no. AY308064), rat (accession no. AB006007), mouse (39), and human (accession no. U29098) were aligned based on the transcriptional start sites for the gene. Putative ROREs were identified using Genomatix MatInspector online software. The sites selected for mapping had 80% or greater homology to the core mammalian transcription factor sequence.

Figure 4

Functional analysis of RORα and rev-erbα by ChIP and EMSA. ChIP assays were run using chromatin fixed from Y-1 cells transfected with the LMB StAR promoter and cultured under both basal and dbcAMP-induced conditions. ChIP was run on each sample with antibodies specific to mouse IgG (nonspecific control), RORα, or rev-erbα. Immunoprecipitation was accomplished by the addition of protein G agarose beads. The pull-downs were washed, complexes were eluted, and DNA was purified. Q-PCR was run on each sample and results are reported graphically as percent enrichment to a 1:10 dilution of each input control. Primers used in the Q-PCR encompassed the RORE/−1969 element in the LMB StAR promoter (A) or the RORE/−634 element in the mouse promoter (B), immunoprecipitated from the same transfected cells. Each figure is representative of one of three replicated experiments. Asterisks (B) indicate that DNA levels were below detection limits. To further investigate the capacity for the LMB RORE/−1969 element to bind RORα, an EMSA was run using recombinant human RORα4 protein (C). A probe encompassing the RORE/−1969 sequence was added to recombinant human RORα4 protein in 1× binding buffer. The reactions were separated on a native gel and transferred to a membrane for EMSA analysis using chemiluminescence. Addition of an antibody specific to RORα to the reactions diminished the banding pattern observed in the lanes containing the probe and protein only, verifying specificity of the protein-DNA interaction in vitro. The following are the sequences for the probes used: RORE/−1969 probe (5′-AAT AGG CAT ATG ACC TAC TTT GGC TC); perfect (human consensus) RORE probe (5′-TCG AGT CGT ATA ACT AGG TCA AGC GCT GGA C-3′); scrambled probe (5′-CCT CTA TAA CGG GTC GGA TAC TAT TA-3′). Lane 1, RORE/−1969 probe only; lane 2, RORE/−1969 probe with RORα4 protein; lane 3, RORE/−1969 probe, RORα4 protein, and RORα4 antibody; lane 4, RORE/−1696 probe, RORE/−1969 cold unlabeled probe, and RORα4 protein; lane 5, scrambled probe and RORα4 protein; lane 6, perfect (human consensus) RORE probe and RORα4 protein; and lane 7, perfect (human consensus) RORE probe, RORα4 protein, and RORα4 antibody.