MMP12 Inhibits Corneal Neovascularization and Inflammation through Regulation of CCL2 - PubMed (original) (raw)
MMP12 Inhibits Corneal Neovascularization and Inflammation through Regulation of CCL2
Marie Wolf et al. Sci Rep. 2019.
Abstract
Following corneal injury, coordinated cellular and protein interactions occur at the wound site to restore tissue homeostasis. Regulation of this response is required to prevent the development of chronic inflammation, abnormal neovascularization, and fibrosis. The chemokine CCL2 and its primary receptor CCR2 are key regulators of the inflammatory and neovascular responses to injury. In this study, we investigated the role of macrophage-associated matrix metalloproteinase 12 (MMP12) in the regulation of CCL2 and CCR2 after corneal wounding. Using two corneal injury models, we examined the temporal and spatial expression of CCL2 and CCR2 in Mmp12-/- and wild-type (WT) mice. Our data showed that MMP12 downregulated CCL2 and CCR2 expression in a manner dependent on the timing and mechanism of injury. We also examined the effect of CCL2 on the injury response in Mmp12-/- and WT corneas. We found that macrophage infiltration and neovascularization following CCL2 blockade was significantly reduced in Mmp12-/- corneas as compared with WT corneas. These findings indicate that MMP12 inhibits corneal inflammation and neovascularization after injury through its regulation of CCL2.
Conflict of interest statement
The authors declare no competing interests.
Figures
Figure 1
MMP12 inhibits expression of CCL2 and CCR2 following chemical corneal injury. (A–C) Relative expression levels of CCL2 mRNA in unwounded corneas and chemically wounded corneas of WT and _Mmp12_−/− mice, as determined by qRT-PCR. Expression levels are relative to uninjured WT corneas. Results are relative expression levels (means ± s.e.m.) at time points unwounded (Un; WT N = 12; KO N = 10), 1 day (1D; WT N = 12, KO N = 12), 4 days (4D; WT N = 4; KO N = 4), and 6 days (6D; WT N = 12; KO N = 12) after injury. (D–F) Relative expression levels of CCR2 mRNA in unwounded corneas and chemically wounded corneas of WT and _Mmp12_−/− mice, as determined by qRT-PCR. Expression levels are relative to uninjured WT corneas. Results are relative expression levels (means ± s.e.m.) at time points unwounded (Un; WT N = 6; KO N = 6), 1 day (1D; WT N = 6, KO N = 6), 4 days (4D; WT N = 4; KO N = 4), and 6 days (6D; WT N = 6; KO N = 6) after injury. ****P < 0.0001, **P < 0.05 and *P < 0.05.
Figure 2
Expression patterns of CCL2 and CCR2 in unwounded and wounded corneas of WT and MMP12 KO mice. (A) Immunofluorescence of CCL2 chemokine and (B) its receptor CCR2 in unwounded and chemically wounded (2-days after injury) WT and _Mmp12_−/− mouse corneas. Control images represent mouse corneas stained with secondary antibody only and without primary antibody. Nuclei were visualized by staining with DAPI (blue). Scale bars: 50 µm. A magnified image of a wounded WT cornea shows perinuclear expression of CCL2 (orange box). CCL2 staining was visualized in epithelial, stromal, and endothelial layers of wounded WT and _Mmp12_−/− corneas. CCR2 staining was visualized in epithelial and stromal layers of unwounded and wounded WT and _Mmp12_−/− corneas.
Figure 3
Temporal expression of CCL2 and CCR2 in epithelial injured corneas of WT and MMP12 KO mice. (A–C) Relative expression levels of CCL2 mRNA in unwounded corneas and epithelial wounded corneas of WT and _Mmp12_−/− mice, as determined by qRT-PCR. Expression levels are relative to uninjured WT corneas. Results are relative expression levels (means ± s.e.m.) at time points unwounded (Un; WT N = 12; KO N = 10), 1 hour (1 H, WT N = 15; KO N = 3), 2 hours (2 H, WT N = 15; KO N = 4), 16 hours (16 H, WT N = 3; KO N = 3), 1 day (1D; WT N = 3, KO N = 3), 2 days (2D; WT N = 14; KO N = 3), 4 days (4D; WT N = 3; KO N = 5), and 6 days (6D; WT N = 5; KO N = 7) after injury. (D–F) Relative expression levels of CCR2 mRNA in unwounded corneas and epithelial wounded corneas of WT and _Mmp12_−/− mice, as determined by qRT-PCR. Expression levels are relative to uninjured WT corneas. Results are relative expression levels (means ± s.e.m.) at time points unwounded (Un; WT N = 6; KO N = 10), 1 hour (1 H, WT N = 9; KO N = 3), 2 hours (2 H, WT N = 9; KO N = 4), 16 hours (16 H, WT N = 3; KO N = 3), 1 day (1D; WT N = 3, KO N = 3), 2 days (2D; WT N = 8; KO N = 3), 4 days (4D; WT N = 3; KO N = 5), and 6 days (6D; WT N = 5; KO N = 7) after injury. ****P < 0.0001, ***P < 0.005, **P < 0.05 and *P < 0.05.
Figure 4
Blocking CCL2 by subconjunctival injection reverses the increased macrophage recruitment in MMP12 KO corneas. (A) Primary antibody against CCL2 was delivered adjacent to the cornea into the subconjunctival space. (B) Effect of CCL2 neutralization on macrophage infiltration into the corneas of WT and _Mmp12_−/− mice. PBS control or antibody to CCL2 was injected into the subconjunctival space and 2 hours later the corneas were wounded chemically. Corneas were collected 7 days after chemical injury. Representative whole-mount images of F4/80+ levels in central corneas of PBS-treated and anti-CCL2-treated WT and _Mmp12_−/− corneas. Scale bar: 10 µm. (C) Quantification of F4/80 levels (pixels) in corneas of PBS-treated and anti-CCL-treated mice. The mean number (±s.e.m.) of F4/80+ cells are shown (WT PBS N = 9; WT anti-CCL2 N = 12; KO PBS N = 9; KO anti-CCL2 N = 15). **P < 0.05 and *P < 0.05.
Figure 5
Blocking CCL2 by subconjunctival injection reverses the increased neovascularization in MMP12 KO corneas. (A) The effect of CCL2 neutralization on angiogenic endothelium (CD31) in wounded corneas of WT and _Mmp12_−/− mice. PBS control or antibody to CCL2 was injected into the subconjunctival space and 2 hours later the corneas were wounded chemically. Corneas were collected 7 days after chemical injury. Representative whole-mount images of CD31 + limbal vessels in corneas of PBS-treated and anti-CCL2-treated WT and _Mmp12_−/− corneas. Scale bar: 10 µm. (B) Quantification of angiogenesis, showing vessel lengths in corneas of PBS-treated and anti-CCL-treated mice. The mean lengths (±s.e.m.) are shown (WT PBS N = 33; WT anti-CCL2 N = 44; KO PBS N = 38; KO anti-CCL2 N = 65). ***P < 0.005 and *P < 0.05.
Figure 6
Neutralization with anti-CCL2 antibody partially reverses VegfA expression in MMP12 KO corneas. The effect of CCL2 neutralization on VegfA expression in wounded corneas of WT and _Mmp12_−/− mice. (A) VegfA is 1.5 times more highly expressed in wounded corneas of _Mmp12_−/− mice treated with PBS control compared with WT mice (WT PBS N = 8; KO PBS N = 7). (B) Treatment of WT mice with anti-CCL2 antibody results in 1.5 times higher VegfA expression (WT PBS N = 8; WT anti-CCL2 N = 9). (C) Treatment of _Mmp12_−/− mice with anti-CCL2 antibody results does not affect VegfA expression (1.5 versus 1.6 respectively; KO PBS N = 7 and KO anti-CCL2 N = 9). (D) Treatment with anti-CCL2 antibody partially reverses the increased VegfA expression in MMP12 KO corneas compared with WT corneas (1.6 versus 1.5 respectively; KO anti-CCL2 N = 9 and WT anti-CCL2 N = 9). *P < 0.05.
Figure 7
Neutralization with anti-CCL2 antibody partially reverses VegfB expression in MMP12 KO corneas. The effect of CCL2 neutralization on VegfB expression in wounded corneas of WT and _Mmp12_−/− mice. (A) VegfB is 1.5 times more highly expressed in wounded corneas of _Mmp12_−/− mice treated with control PBS compared with WT mice (WT PBS N = 8; KO PBS N = 8). (B) Treatment of WT mice with anti-CCL2 antibody results in 1.5 times higher VegfB expression (WT PBS N = 8; WT anti-CCL2 N = 9). (C) Treatment of _Mmp12_−/− mice with anti-CCL2 antibody does not affect VegfB expression (1.5 versus 1.7 respectively; KO PBS N = 8 and KO anti-CCL2 N = 9). (D) Treatment with anti-CCL2 antibody partially reverses the increased VegfB expression in MMP12 KO corneas. (1.7 versus 1.5 respectively; KO anti-CCL2 N = 9 and WT anti-CCL2 N = 9). *P < 0.05.
Figure 8
Expression patterns of CCR2, VEGFA, and VEGFB protein in wounded corneas of WT and MMP12 KO mice. (A) Protein expression levels of CCR2 and actin in unwounded and wounded corneas of WT (N = 8 per lane) and _Mmp12_−/− (N = 8 per lane) mice 1 and 6 days post-chemical injury, as determined by Western blot analysis. Full-length blots are presented in Supplementary Fig. 1A. (B,C) The effect of CCL2 neutralization on VEGFA and VEGFB protein expression in WT and _Mmp12_−/− mice at 7 days post-chemical injury. Treatment of WT and _Mmp12_−/− mice with PBS or anti-CCL2 antibody had no significant effect on VEGFA expression. Treatment of _Mmp12_−/− mice with anti-CCL2 significantly decreased VEGFB protein expression compared with PBS-treated _Mmp12_−/− mice (0.42 versus 0.065 respectively). VEGFB expression was decreased more in WT mice compared with _Mmp12_−/− mice following treatment with anti-CCL2 (0.24 versus 0.065 respectively). *P < 0.05. Full-length blots are presented in Supplementary Fig. 1B,C. While we had to use several gels to fit all samples, they all derive from the same experiment and gels/blots were processed in parallel.
Figure 9
A model of MMP12 regulation of the CCL2-CCR2 signaling axis. Our data suggest that MMP12 and CCL2 are both secreted after epithelial and chemical injuries of the cornea. MMP12 inhibits CCL2 expression and reduces CCL2-CCR2 binding. This leads to reduced monocyte recruitment and reduced angiogenesis which prevents the development of corneal fibrosis.
References
- Dean RA, et al. Macrophage-specific metalloelastase (MMP-12) truncates and inactivates ELR+ CXC chemokines and generates CCL2, −7, −8, and −13 antagonists: potential role of the macrophage in terminating polymorphonuclear leukocyte influx. Blood. 2008;112:3455–3464. doi: 10.1182/blood-2007-12-129080. -DOI -PubMed
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Molecular Biology Databases
Miscellaneous