Nuclear factor-Y (NF-Y) regulates transcription of mouse Dmrt7 gene by binding to tandem CCAAT boxes in its proximal promoter

Abstract

Dmrt7, a member of the Dmrt family of genes, is required for spermatogenesis. However, promoter functions of the gene Dmrt7 remain unknown. We have cloned and characterized the proximal promoter region of the mouse Dmrt7 gene. Functional analysis of the 5' flanking region by sequential deletion mutations revealed crucial positive elements between -60 and +1, in which two highly conserved and tandem CCAAT boxes: the CCAAT box1 (-48/-44) and the CCAAT box2 (-7/-3) are located. Site-directed mutagenesis studies demonstrated that both CCAAT boxes are indispensable to the promoter activity. Electrophoretic mobility shift assays (EMSAs) and gel-supershift assays indicated that transcription factor NF-Y binds to the promoter. Chromatin immunoprecipitation (ChIP) analysis demonstrated that NF-Y interacts in vivo with the promoter of the Dmrt7 gene in testis. Co-transfection and reporter analysis showed that over-expression of NF-Ys increased transcription of the Dmrt7-luc gene whereas expression of a dominant-negative NF-Ya decreased the transcription. This suggests that NF-Y can activate the Dmrt7 promoter. These results provide evidence of a transcription regulatory mechanism that controls Dmrt7 gene expression in mouse testis.

Keywords: DM domain; Endocrinology; Spermatogenesis; Transcriptional regulation.

Figure 1

Identification and function characterization of the mouse Dmrt7 promoter. a), Nucleotide sequence of the promoter region (from -407 to +116) of mouse Dmrt7. The putative binding sites for transcriptional factors are underlined. The CCAAT boxes are boxed. The numbering of the nucleotides starts at the first nucleotide of the cDNA (+1). b), Sequence alignment of 1 kb upstream of the first exon of Dmrt7 from mouse, rat and human. Two highly conserved CCAAT boxes are indicated. Conserved nucleotides among all three species are denoted by red letters and are shaded in yellow. The positions of the nucleotides in the mouse Dmrt7 gene are indicated. +1 represents the first nucleotide of the cDNA. c), Deletion analysis of the mouse Dmrt7 promoter using the luciferase assay. GC-1 or COS 7 cells were transiently transfected with 0.4 µg of the deletion constructs together with 10 ng Renilla luciferase plasmid, pRT-TK, which served as an inner control for transfection efficiency. The relative activities to the -60/+116 construct of a series of deletion mutants was determined by the luciferase assay. The results are the mean ± S.D. of three independent experiments. *, P < 0.05; **, P < 0.001. d), Mutation analysis in the CCAAT boxes of the Dmrt7 promoter by the luciferase assay. The -948/+116 and -60/+116 constructs were used as a template to generate point mutants. GC-1 or COS 7 cells were transiently transfected with 0.4 µg of these constructs together with 10 ng Renilla luciferase plasmid, pRT-TK. The relative activities of the -60/+116 construct of these mutants was determined by the luciferase assay. The results are the mean ± S.D. of three independent experiments. *, P < 0.05; **, P < 0.001 with compared to the -60/+116 wild-type construct. The box indicates the intact CCAAT box. The fork indicates the corresponding mutations.

Figure 2

Electrophoretic mobility shift assay of NF-Y binding to Dmrt7 promoter. The oligo1 corresponding to -58/-31 and oligo2 corresponding to -20/+8 were γ-32P-ATP labelled and incubated with 5µg of nuclear extract of mouse testis in the absence or presence of 50-fold excess various competitor DNA (mutant or non-labeled oligo) or anti-NFYa antibody, as indicated on the top of the gel image. Samples were resolved on 5% polyacrylamide gels in 0.5% TBE running buffer at 10 V/cm for 2 h. The specific DNA/protein complex and the super-shift bands were indicated by arrows. The shift bands cannot be competed with mutated DNA competitor (lane 7, 8). The sequences of oligo1, oligo2 and corresponding mutants are shown in the lower panel.

Figure 3

ChIP assay of NF-Y binding to the Dmrt7 promoter in testis. a), Western blot analysis of the Dmrt7 protein in the adult testis, heart, liver, spleen, lung, kidney, ovary, brain and epididymis. The Dmrt7 expression was only detected in the testis. β-actin was used as an internal control. b), Interaction of NF-Y with the Dmrt7 promoter in vivo was determined by chromatin immunoprecipitation analysis. Samples of mouse testis and liver were chopped into small pieces and cross-linked in 1% formaldehyde to cross-link endogenous proteins and DNA. Samples of sonicated chromatin were immunoprecipitated with anti-NFYA, no antibody (beads only) and preimmuno IgG (control) respectively. DNA isolated from immunoprecipitated material was amplified by PCR with primers to amplify the 160-bp sequence of mouse Dmrt7 promoter corresponding to the -101 to +59 region. Primers for an unrelated region were used as negative control. The amplified PCR fragments were analyzed on 2% agarose gel. c), A q-ChIP PCR was performed to quantify the binding extent of NF-Ya on the Dmrt7 promoter. All values were initially expressed relative to relevant IgG DNA content. d), The relative positions of the primers are shown.

Figure 4

Co-transfection analysis of the mouse Dmrt7 promoter. a), Effects of expression of NF-Y on the transcriptional activity of the Dmrt7 5'-flanking sequence (-60/+116). Cells were transfected with 0.2µg -60/+116 construct or mutants and 0.2µg NF-Y expression plasmids (pCX-NFYa, pCX-NFYb, pCX-NFYc). b), Effects of expression of the dominant negative mutant of NF-Y on the transcriptional activity of the Dmrt7 5'-flanking sequence (-60/+116). The dominant negative NF-Ya expression plasmid NF-YAm29 (0.1µg, 0.2µg, 0.3µg) was cotransfected with 0.2µg -60/+116 construct, which inhibits Dmrt7-luciferase activity. pGL3-basic was employed as a negative control. pRT-TK served as an inner control for transfection efficiency. The relative luciferase activities to the -60/+116 construct were determined by luciferase assay. Cells were harvested 24 h post-transfection, and luciferase activity was measured and normalized to Renilla luciferase activity. All transfection experiments were repeated at least three times and the activities are showed relative to that of the -60/+116 construct, which was set to 10. The results are the mean ± S.D. of three independent experiments. *, P < 0.05; **, P < 0.001 with compared to the -60/+116 wild-type construct.