Conservation of mechanisms mediating gonadotrophin-releasing hormone 1 stimulation of human luteinizing hormone beta subunit transcription

Abstract

Gonadotrophin-releasing hormone (GNRH1) regulates pituitary luteinizing hormone (LH). Previous studies have delineated a mechanism for GNRH1-induced LHbeta subunit gene (Lhb) transcription, the rate-limiting step in LH production. GNRH1 induces expression of early growth response 1 (EGR1), which interacts with steroidogenic factor 1 (SF1) and paired-like homeodomain transcription factor 1 (PITX1) to regulate Lhb promoter activity. Though the cis-elements for these factors are conserved across species, regulation of human LHB transcription has not been thoroughly investigated. We therefore characterized LHB transcriptional regulation by GNRH1 using promoter-reporter analyses in LbetaT2 cells. GNRH1 stimulated LHB transcription via an extracellular signal-regulated kinase 1/2 pathway. EGR1 bound to two binding sites on the LHB promoter and this binding was increased by GNRH1. Mutation of either site or knockdown of endogenous EGR1 decreased basal and/or GNRH1-regulated promoter activity. The human LHB promoter also contains low and high affinity SF1 binding sites. Mutation of these elements or depletion of endogenous SF1 impaired basal and ligand-induced transcription. Knockdown of PITX1 or PITX2 isoforms impaired GNRH1 induction, and endogenous PITX1 bound to the candidate PITX binding site on the LHB promoter. Thus, the mechanism described for GNRH1 regulation of Lhb in other species is largely conserved for human LHB. We also uncover a previously unappreciated role for PITX2 isoforms in this process.

Figure 1

Aligment of proximal Lhb/LHB promoters from human, rat and cow. In all cases, +1 refers to the transcription start site. Nucleotides that differ from the consensus are shaded. The conserved SF1, EGR and PITX elements are boxed. ‘d’: distal, ‘p’: proximal.

Figure 2

Schematic representations of the proximal LHB promoter are shown at the left of each graph. The SF1, EGR and PITX elements are represented by squares, triangles and a circle, respectively. Black symbols indicate mutated sites. (A) LβT2 cells were transfected with 450 ng/well of the indicated LHB-luc reporters. WT, wild-type; xdEGR, mutated distal EGR site; xpEGR, mutated proximal EGR site; 2xEGR, both EGR elements mutated. Cells were treated or not with 10⁻⁷ M GNRH1 for 6 h. (B) LβT2 cells were transfected as above with either WT 0.2 kb LHB-luc reporter or mutant constructs with the inactivated distal (xdSF1), proximal (xpSF1) or both (2xSF1) SF1 sites. Where indicated, GNRH1 treatment was given for 6 h. (C) LβT2 cells were transfected as above with either WT LHB-luc reporter or a construct with a mutated PITX element (xPITX). Differences in reporter activity were measured after 6 h GNRH1 treatment. The fold induction by GNRH1is indicated at the bottom of the graph. Bars with different symbols differ significantly. n = 3 for all treatments.

Figure 3

(A) Nuclear extracts from LβT2 cells treated (+) or not (−) with 10⁻⁷ M GNRH1 for 1 h were incubated with a radio-labeled probe corresponding to −66/−33 of the LHB promoter. Where indicated, the binding reactions contained 100-fold excess of cold homologous wild-type probe (WT; lane 3) or probes with mutated proximal SF1 (xpSF1; lane 4) or EGR (xpEGR; lane 5) elements. Control IgG (lane 6), or SF1 (lane 7) or EGR1 (lanes 8 and 9) antibodies were added as indicated. Asterisks denote super-shifted complexes. (B) Nuclear extracts from LβT2 cells treated (+) or not (−) with 10⁻⁷ M GNRH1 for 1 h were incubated with a radio-labeled probe corresponding to the −66/−33 region of the LHB promoter. Ten, 50, 100 or 500-fold excess homologous cold probe (−66/−33; lanes 3–6), cold probe containing the putative distal SF1 and EGR elements (−134/−103; lanes 7–10), or cold probe with mutated proximal or distal SF1 or EGR elements (500× only; lanes 11–14) were added where indicated.

Figure 4

(A) Nuclear extracts from LβT2 cells treated (+) or not (−) with 10⁻⁷ M GNRH1 for 1 h (lanes 1–10), or nuclear extracts from CHO cells transfected with empty vector (pcDNA3; lane 11) or Myc-PITX1 (lanes 12–14) were incubated with a radio-labeled probe corresponding to the −104/−79 region of the LHB promoter. Where indicated, the binding reactions contained 100-fold excess of cold homologous WT probe (lanes 3 and 4) or probe with a mutated PITX site (PITX mut; lanes 5 and 6), control IgG (lanes 7, 8 and 13), PITX1 antibody (lanes 9 and 10) or Myc antibody (lane 14). (B) DNAP was performed using the probes described above. Whole cell lysates (total) or proteins interacting with the probes were subjected to immunoblot (IB). Cells were treated (+) or not (−) with 10⁻⁷ M GNRH1 for 1 h.

Figure 5

(A) LβT2 cells were co-transfected with 450 ng/well of WT 0.2 kb LHB-luc reporter and 10⁻⁸ M of (A) Egr1 or Sf1, (B) Pitx1 or (C) Pitx2 siRNAs. In all cases, 1× siRNA buffer was used as control. Cells were treated or not with 10⁻⁷ M GNRH1 for 6 h prior to collection of lysates for luciferase assays. Fold induction by GNRH1 is indicated at the bottom of the graphs. Bars with different symbols differ significantly. n = 3 per treatment.

Figure 6

(A) CV-1 cells were transfected with 900 ng/well of the 0.2 kb LHB-luc reporter along with 30 ng/well of EGR1 and/or PITX1 expression vectors or empty vector (pcDNA3). After overnight recovery, reporter activity was measured. The average fold stimulation, indicated at the bottom of the graph, was normalized to the reporter activity measured in presence of only the empty vector. (B) CV-1 cells were transfected as in panel (A) with 30 ng/well of SF1 and/or PITX1 expression vectors or empty vector (pcDNA3). Reporter activity was measured and normalized as above. Bars with different symbols differ significantly. n = 3 per treatment.