Redox-sensitive gene-regulatory events controlling aberrant matrix metalloproteinase-1 expression

Abstract

Aberrant matrix metalloproteinase-1 (MMP-1) expression contributes to the pathogenesis of many degenerative disease processes that are associated with increased oxidative damage or stress. We and others have established that shifts in steady-state H2O2 production resulting from enforced antioxidant gene expression, senescence, or UV irradiation control MMP-1 expression. Here we establish that histone deacetylase-2 (HDAC2) protein levels and its occupancy of the MMP-1 promoter are decreased in response to enforced manganese superoxide dismutase (Sod2) expression. Inhibition of HDAC activity further accentuates the redox-dependent expression of MMP-1. Sod2-dependent decreases in HDAC2 are associated with increases in a proteasome-sensitive pool of ubiquitinylated HDAC2 and MMP-1-specific histone H3 acetylation. Sod2 overexpression also enhanced recruitment of Ets-1, c-Jun, c-Fos, and the histone acetyltransferase PCAF to the distal and proximal regions of the MMP-1 promoter. Furthermore, the Sod2-dependent expression of MMP-1 can be reversed by silencing the transcriptional activator c-Jun. All of the above Sod2-dependent alterations are largely reversed by catalase coexpression, indicating that the redox control of MMP-1 is H2O2-dependent. These findings identify a novel redox regulation of MMP-1 transcription that involves site-specific promoter recruitment of both activating factors and chromatin-modifying enzymes, which converge to maximally drive MMP-1 gene expression.

Keywords: Acetylation; Epigenetic modification; Free radicals; Histone deacetylase-2; Hydrogen peroxide; Manganese superoxide dismutase; Matrix metalloproteinase-1; Oxidative stress; Redox-dependent; Steady-state H(2)O(2) concentration.

Figure 1. Redox-dependent MMP-1 expression is associated with decreased HDAC and increased HAT recruitment that is accompanied by enhanced histone acetylation

Cells were subjected to ChIP as described in materials and methods. Protein-DNA complexes were pulled down using specific antibodies to HDAC2, PCAF, and acetylated H3. IgG antibody was used as a negative control for non-specific binding. A) Schematic diagram of transcription elements on the MMP-1 promoter indicating primer spanning regions used in ChIP. Distal and proximal primers span −1491 to −1696 and −245 to −37 upstream of the transcription start site respectively. B) HDAC2 promoter occupancy at the distal and proximal MMP-1 promoter. *p=0.0149 compared to control, ** p<0.0001 control vs. Sod2 or Sod2Cat, Sod2 vs. Sod2cat C) Acetylation of Histone H3 at the distal and proximal regions of the MMP-1 promoter. * p=0.0007 or **p<0.0001 compared to control D) Promoter occupancy of PCAF to the distal and proximal MMP-1 promoter. * p<0.0001 or ** p<0.05 for Sod2 compared to control, or Sod2 compared to Sod2Cat. N=3 independent immunoprecipitations. For data analysis, ChIP samples were all normalized to the amount of input DNA for each sample. Data is presented as enrichment after ChIP, with the IgG control sample set to 1. Mean values are plotted with error bars representing standard error of the mean.

Figure 2. HDACs contribute to the redox-responsiveness of MMP-1 transcription

A) The indicated cell lines were treated with or without TSA (500 ng/ml) for 18 hrs and resulting real-time RT-PCR analysis of MMP-1 mRNA levels normalized to β-actin is shown. Data is representative of two independent experiments, *=p<0.005 compared to untreated. Means are plotted with error bars indicating standard deviation. B) Representative western blots are shown for cytosolic and nuclear HDAC2 and Actin in the indicated cell lines treated with 10 µM MG-132 for 4 hours. C) Relative densitometric intensity (RDI) of HDAC2 normalized to Actin is presented. RDI of HDAC2 was obtained and expressed as % of untreated control group from three independent experiments, *=p<0.05 and **=p<0.01.

Figure 3. HDAC2 ubiquitination is increased in response to oxidant stress and regulates MMP-1 expression

A) Cells were either treated with 10 µM MG-132 or left untreated for 18 hours followed by immunoprecipitation with anti HDAC2 antibody and blotted for anti-ubiquitin. Blot is representative of three separate experiments. B) Relative densitometric intensity of ubiquitinated HDAC2 normalized to total HDAC2 levels. The control was set to 1. Data is representative of three independent experiments. *, p<0.05 as compared to control. Means are plotted with error bars indicating standard error of the mean.

Figure 4. Redox-dependent recruitment of Ets-1, c-Jun and c-Fos to the MMP-1 promoter

Cells were subjected to ChIP as described in materials and methods. Protein-DNA complexes were pulled down using specific antibodies to Ets-1, c-Jun, and c-Fos. IgG antibody was used as a negative control for non-specific binding. A) Relative promoter occupancy at the distal region of the MMP-1 promoter by Ets-1, c-Jun and c-Fos. B) Relative promoter occupancy of Ets-1, c-Jun and c-Fos at the distal region of the MMP-1 promoter. * p<0.0001 for Sod2 compared to control or Sod2 compared to Sod2Cat. N=3 independent immunoprecipitations. For data analysis, ChIP samples were all normalized to the amount of input DNA for each sample. Data is presented as enrichment after ChIP, with the IgG control sample set to 1. Mean values are plotted with error bars representing standard eror of the mean.

Figure 5. Redox-regulated MMP-1 transcription is c-Jun dependent in response to Sod2 overexpression

A) Western blot analysis of c-Jun and GAPDH in Sod2 overexpressing cells transfected with c-Jun specific or scrambled siRNA. Lower panel, Relative c-Jun levels in siRNA transfected cells normalized to that of scrambled control (n=3). Average percentage in reduction in c-Jun protein levels is indicated. B) Western blot of MMP-1 protein levels in media from Sod2 overexpressors treated as indicated. Lower Panel, Relative densitometric analysis of MMP-1 levels in siRNA transfected cells as compared to scrambled controls (n=3). C) Real-time RT-PCR analysis of MMP-1 and c-Jun in Sod2 overexpressors treated as above (n=3). mRNA levels normalized to β-actin in the indicated cells D) MMP-1 promoter luciferase reporter assay in Sod2 overexpressing cells treated as indicated (n=3). All immunoblots are representative of three independent experiments. *, p<0.007 as compared to scrambled control. **, p=0.0002 as compared to scrambled control. Means are plotted with error bars indicating standard error of the mean.