Matrix Metalloproteinase-2 (MMP-2) Gene Deletion Enhances MMP-9 Activity, Impairs PARP-1 Degradation, and Exacerbates Hepatic Ischemia and Reperfusion Injury in Mice

Abstract

Hepatic ischemia and reperfusion injury (IRI) is an inflammatory condition and a significant cause of morbidity and mortality after surgery. Matrix metalloproteinases (MMPs) have been widely implicated in the pathogenesis of inflammatory diseases. Among the different MMPs, gelatinases (MMP-2 and MMP-9) are within the most prominent MMPs detected during liver IRI. While the role of MMP-9 in liver damage has been fairly documented, direct evidence of the role for MMP-2 activity in hepatic IRI remains to be established. Due to the lack of suitable inhibitors to target individual MMPs in vivo, gene manipulation is as an essential tool to assess MMP direct contribution to liver injury. Hence, we used MMP-2-/- deficient mice and MMP-2+/+ wild-type littermates to examine the function of MMP-2 activity in hepatic IRI. MMP-2 expression was detected along the sinusoids of wild-type livers before and after surgery and in a small population of leukocytes post-IRI. Compared to MMP-2+/+ mice, MMP-2 null (MMP-2-/-) mice showed exacerbated liver damage at 6, 24, and 48 hours post-reperfusion, which was fatal in some cases. MMP-2 deficiency resulted in upregulation of MMP-9 activity, spontaneous leukocyte infiltration in naïve livers, and amplified MMP-9-dependent transmigration of leukocytes in vitro and after hepatic IRI. Moreover, complete loss of MMP-2 activity impaired the degradation of poly (ADP-ribose) polymerase (PARP-1) in extensively damaged livers post-reperfusion. However, the administration of a PARP-1 inhibitor to MMP-2 null mice restored liver preservation to almost comparable levels of MMP-2+/+ mice post-IRI. Deficient PARP-1 degradation in MMP-2-null sinusoidal endothelial cells correlated with their increased cytotoxicity, evaluated by the measurement of LDH efflux in the medium. In conclusion, our results show for the first time that MMP-2 gene deletion exacerbates liver IRI. Moreover, they offer new insights into the MMP-2 modulation of inflammatory responses, which could be relevant for the design of new pharmacological MMP-targeted agents to treat hepatic IRI.

Fig 1. Time course and sources of MMP-2 expression in MMP-2^+/+ livers.

MMP-2 mRNA expression (panel A) and MMP-2 activity (panel B) were readily detected in wild-type livers before and after hepatic IRI. MMP-2 positive staining was detected in the vasculature of naïve MMP-2^+/+ livers (panel C); MMP-2 in red (a; Alexa Fluor 594), PECAM-1 in green (b; Alexa Fluor 488), nuclear stain in blue (c; Dapi), and staining overlay (d). In addition to the vasculature, MMP-2 positive staining was also detected in Ly6G neutrophils in MMP-2^+/+ livers at 6h post-IRI (panel D); MMP-2 in red (a and b), Ly6G in green (c and d) and staining overlay (e and f); (arrows and arrow heads denote MMP-2 positive staining in the liver vasculature and in infiltrating leukocytes, respectively; n = 5/group; *p<0.05 relative to naïve livers).

Fig 2. Survival, liver histology and serum transaminases in MMP-2^-/- and MMP-2^+/+ mice after liver IRI.

MMP-2 expression (panel A) was detected in MMP-2^+/+ livers (lanes 1 and 2) and absent in MMP-2^-/- livers (lanes 3 and 4) post-IRI. Survival (panel B) of MMP-2^-/- deficient mice (dotted line) was significantly reduced at 7 days post-IRI, as compared with MMP-2^+/+ mice (solid line). Representative H&E staining (panel C) of MMP-2^+/+ (a, c, and e) and MMP-2^-/- (b, d, and f) livers at 6h (a, and b), 24h (c, and d), and 48h (e, and f) post-I/R injury showed that disruption of lobular architecture was significantly exacerbated in MMP-2^-/- livers post-reperfusion. The serum transaminase levels, sAST (panel D) and sALT (panel E), were markedly increased in MMP-2^-/- mice, particularly at 6h and 24h after IRI; (n = 5–6 mice/group *p<0.05).

Fig 3. Leukocyte infiltration in MMP-2^-/- and MMP-2^+/+ livers.

Representative immunostaining of Ly-6G+ neutrophils (panel A) and Mac-1+ leukocytes (panel B) in MMP-2^+/+ (a, and b) and MMP-2^-/- (c, and d) livers at 6h (a, and c) and 24h (b, and d) post-reperfusion; infiltrating leukocytes were detected in significantly higher numbers in livers of MMP-2^-/- mice after IRI (panel C). MPO enzymatic activity (panel D) was also markedly upregulated in MMP-2^-/- livers post-IRI (n = 5/group; *p<0.05).

Fig 4. Pro-inflammatory cytokine expression in MMP-2^-/- and MMP-2^+/+ livers.

IFN-γ, TNF-α, and IL-6, were upregulated in MMP-2^-/- deficient livers post-IRI, compared to respective controls (n = 5/group; *p<0.05).

Fig 5. Serum transaminases, liver histology and leukocyte infiltration/ activation in anti-MMP-2 antibody treated mice.

AST (panel A) and ALT (panel B) levels were elevated in blood samples of anti-MMP-2 antibody treated mice collected 6h after liver IRI. Liver histological preservation (panel C) was further impaired in anti-MMP-2 antibody treated mice (b), compared to IgG treated controls (a). Moreover, MPO activity (panel D), Mac-1 leukocyte infiltration (panel E) and proinflammatory cytokine expression (panel F) were all increased in anti-MMP-2 antibody treated mice, compared to IgG treated controls (n = 4/group; *p<0.05).

Fig 6. Effect of MMP-2 selective inhibition on MMP-9 activity and MMP-9-dependent leukocyte migration.

MMP-9 activity (panel A) was increased in MMP-2^-/- livers (lanes 3, 4, 7, 8, 11, and 12), compared to respective MMP-2^+/+ controls (lanes 1, 2, 5, 6, 9, and 10); naïve livers (lanes 1 to 4), and livers harvested at 6h (lanes 5 to 8) and 24h (lanes 9 to 12) post-IRI. Infiltrating MMP-9+ leukocytes (panel B) were detected in higher numbers in MMP-2^-/- livers than in MMP-2^+/+ livers. Representative immunostaining (panel C) shows that MMP-9+ leukocytes were virtually absent from MMP-2^+/+ naïve livers (a), but readily detected in MMP-2^-/- naïve livers (b); MMP-9+ leukocyte infiltration was also lower in MMP-2^+/+ livers (c) when compared to MMP-2^-/- livers (d) at 6h post-reperfusion. MMP-9 activity (panel D) was significantly increased in isolated MMP-2^+/+ neutrophils treated with a selective MMP-2 inhibitor in doses of 100nM (lane 2) and 500nM (lane 3) and in MMP-2^-/- neutrophils (lane 4), compared to respective MMP-2^+/+ controls (lane1). Moreover, in vitro cell migration (panel E) was significantly enhanced in MMP-2^+/+ neutrophils treated with a MMP-2 selective inhibitor and in MMP-2^-/- neutrophils; on the other hand, the addition of a MMP-9 selective inhibitor to MMP-2^-/- neutrophils significantly reduced their migration (n = 5/group; in vitro data is expressed as mean ± SD of four independent experiments; *p<0.05 relative to untreated MMP-2^+/+ samples and **p<0.05 relative to untreated MMP-2^-/- samples).

Fig 7. Effect of MMP-2 activity inhibition on PARP-1 degradation.

The cleaved form of PARP-1 (panel A) was detected in the MMP-2^+/+ livers (lanes 1 and 2) and virtually absent in MMP-2^-/- livers (lanes 3, and 4) at 6h post-IRI; the cleaved PARP/total PARP ratio was profoundly depressed in the MMP-2^-/- livers post-reperfusion. The cleaved PARP/total PARP ratios (panel B) were also markedly decreased in lysates of MMP-2^-/- SECs and MMP-2 inhibitor-treated MMP-2^+/+ SECs. The LDH levels (panel C) were increased in MMP-2^-/- SECs and in MMP-2 inhibitor-treated MMP-2^+/+ SECs, compared with controls (n = 4–5/group; in vitro data is expressed as mean ± SD of four independent experiments; *p<0.05 relative to MMP-2^+/+ samples and **p<0.05 relative to MMP-2^-/- samples).

Fig 8. Effect of PJ34 (PARP inhibitor) administration in hepatic IRI.

Administration of PJ34 to wild-type mice and to MMP-2^-/- mice at reperfusion reduced the serum AST levels (panel A) in these mice at 6h post-IRI. MMP-2^-/- mice treated with PJ34 showed improved liver histological preservation (panel B) post-IRI; H&E staining of vehicle treated-MMP-2^-/- (a) and PJ34 treated-MMP-2^-/- (b) livers post-reperfusion. MMP-9 activity was significantly depressed in MMP-2 null mice treated with PJ34 at 6h after IRI (n = 4–5/group; *p<0.05).

Fig 9. Anti-MMP-2 antibody therapy in MMP-9 null mice.

The serum AST levels (panel A) in MMP-9^-/- mice treated with anti-MMP-2 antibodies at 6h of liver IRI were significantly reduced when compared to wild-type mice treated with anti-MMP-2 antibodies and increased when compared to MMP-9^-/- mice treated with IgG. Histological preservation (panel B) was markedly improved in MMP-9 null livers treated with MMP-2 mAb at 6h after IRI (b), when compared to the extensively damaged anti-MMP-2 mAb treated WT livers (a), but inferior when compared to MMP-9^-/- livers treated with control IgG (c) (n = 4/group; *p<0.05 relative to WT+MMP-2mAb; **p<0.05 relative to MMP-9^-/-+IgG).