Intronic regulation of matrix metalloproteinase-2 revealed by in vivo transcriptional analysis in ischemia

Abstract

Matrix metalloproteinase-2 (MMP-2) plays an essential role in angiogenesis and arteriogenesis, two processes critical to restoration of tissue perfusion after ischemia. MMP-2 expression is increased in tissue ischemia, but the responsible mechanisms remain unknown. We studied the transcriptional activation of the MMP-2 gene in a model of hindlimb ischemia by using various MMP-2-lacZ reporter mice and chromatin immunoprecipitation. MMP-2 activity and mRNA were increased after hindlimb ischemia. Mice with targeted deletion of MMP-2 had impaired restoration of perfusion and a high incidence of limb gangrene, indicating that MMP-2 plays a critical role in ischemia-induced revascularization. Ischemia induced the expression and binding of c-Fos, c-Jun, JunB, FosB, and Fra2 to a noncanonical activating protein-1 (AP-1) site present in the MMP-2 promoter and decreased binding of the transcriptional repressor JunD. Ischemia also activated the expression and binding of p53 to an adjacent enhancer site (RE-1) and increased expression and binding of nuclear factor of activated T-cells-c2 to consensus sequences within the first intron. Deletion of either the 5' AP-1/RE-1 region of the promoter or substitution of the first intron abolished ischemia-induced MMP-2 transcription in vivo. Thus, AP-1 transcription factors and intronic activation by nuclear factor of activated T-cells-c2 act in concert to drive ischemia-induced MMP-2 transcription. These findings define a critical role for MMP-2 in ischemia-induced revascularization and identify both previously uncharacterized regulatory elements within the MMP-2 gene and the cognate transcription factors required for MMP-2 activation in vivo after tissue ischemia.

Fig. 1.

Hindlimb ischemia induces MMP-2 activity and expression. (A) Blood flow measurements by laser Doppler in nonischemic (N) and ischemic (I) limbs after 3 days of ischemia (n = 6 each) (*, P < 0.05). (B) MMP-2 mRNA levels in gastrocnemius muscle from nonischemic (N) and ischemic (I) limbs (n = 6–8) after 3 days of ischemia (*, P < 0.05). (C) MMP-2 activity by gelatin zymography in nonischemic (N) and ischemic (I) limbs after 3 days of ischemia. (D) Hindlimb blood flow expressed as ischemic/nonischemic ratio in MMP-2 (-/-) and C57Bl6 wild-type mice (n = 4–9 per group) after ischemia (*, P < 0.05 vs. C57Bl6).

Fig. 2.

Ischemia induces MMP-2 transcription. (A) Schematic of transgene in F8 MMP-2 reporter mouse: 5,086 bp of rat MMP-2 genomic DNA (through the second exon) with β-galactosidase reporter cassette. Arrow indicates MMP-2 transcription start site. *, mutation of MMP-2 translation start site. (B) Blood flow measured by laser Doppler in ischemic (open bars) and nonischemic (filled bars) hindlimbs (n = 6–8 each) at indicated time points after induction of ischemia. (*, P < 0.05). (C) MMP-2 transcription analyzed by β-galactosidase assay of ischemic (open bars) and nonischemic (filled bars) gastrocnemius muscles (n = 6 each) in F8 reporter mice at indicated time points after ischemia (*, P < 0.05). (D) Immunohistochemistry for β-galactosidase (brown) in nonischemic and ischemic gastrocnemius muscle sections. (Magnification: ×400.)

Fig. 3.

Ischemia-induced MMP-2 transcription requires the 5′ region and the first intron. (A) Schematic of transgenes in three different MMP-2 reporter mouse strains. F8 contains -1,686 to +3,400 bp of the rat MMP-2 promoter coupled to β-galactosidase. Arrow indicates MMP-2 transcription start site. *, mutated MMP-2 translation start site. F8-del strain contains promoter truncated at 5′ aspect (-1,241 bp), and F8-HGH strain is identical to F8 but with substitution of human growth hormone intron for first MMP-2 intron. (B) Blood flow by laser Doppler in ischemic (open bars) and contralateral limbs (filled bars) 3 days after ischemia for each strain. (*, P < 0.05; n = 6–8). (C) Zymography showing ischemia-induced endogenous (mouse) MMP-2 activity in gastrocnemius muscle in each strain 3 days after ischemia. I, ischemic; N, nonischemic limbs. Zymograms are representative of 3–5 similar experiments per strain. (D) β-galactosidase assay of gastrocnemius extracts from ischemic (open bars) and contralateral limbs (filled bars) 3 days after hindlimb ischemia demonstrates loss of either the 5′ region or the first intron abolishes ischemia-induced MMP-2 transcription (*, P < 0.05; n = 12 per strain).

Fig. 4.

Ischemia induces binding of AP-1 transcription factors to the 5′ aspect of the MMP-2 promoter. Chromatin immunoprecipitation of ischemic (I) and nonischemic (N) contralateral muscle on day 3 shows increased binding of JunB, FosB, c-Jun, c-Fos, and Fra2 to the 5′ AP-1 site and decreased JunD binding. Results shown are representative of 3–6 similar experiments per transcription factor.

Fig. 5.

Ischemia induces expression of AP-1 family transcription factors. Immunohistochemistry of ischemic and nonischemic muscle on day 3 after induction of ischemia shows increased expression of the indicated AP-1 transcription factors. (Magnification: ×400.)

Fig. 6.

Ischemia induces p53 expression and binding to 5′ RE-1 site of the MMP-2 promoter. (A) Chromatin immunoprecipitation 3 days after induction of ischemia shows increased binding of p53 to RE-1 site in the MMP-2 promoter in ischemic (I) versus nonischemic (N) limbs, no binding of YB-1, and no change in basal AP-2 binding with ischemia. Results shown are representative of 3–6 similar experiments per transcription factor. (B) Immunohistochemistry demonstrates increased expression of p53 and AP-2 in skeletal muscle 3 days after induction of ischemia. (Magnification: ×400.)

Fig. 7.

Ischemia induces NFATc2 expression and binding to the MMP-2 promoter. (A) Schematic of first intron of the mouse MMP-2 gene with potential binding sites for AP-2 (triangles), FosB (rectangle), NFATc2 (ovals), and the three primer sets (arrows below) used in chromatin immunoprecipitation. (B) Chromatin immunoprecipitation demonstrates no AP-2 binding (with primer set 1) in ischemic (I) or nonischemic (N) limbs. (C) Chromatin immunoprecipitation showing no change in FosB binding (with primer set 2) in ischemic (I) versus nonischemic (N) limbs. (D) Chromatin immunoprecipitation with NFATc2 antibody shows no binding with primer set 1 and increased binding with ischemia with primer set 3. No NFATc2 binding was found with primer set 2 (data not shown). Results shown are representative of three to six similar experiments per transcription factor. (E) Immunohistochemistry reveals increased NFATc2 expression in ischemic (I) versus nonischemic (N) muscle. (Magnification: ×400.)