Nitric oxide regulates matrix metalloproteinase-9 activity by guanylyl-cyclase-dependent and -independent pathways

Abstract

Matrix metalloproteinases (MMPs) are of central importance in the proteolytic remodeling of matrix and the generation of biologically active molecules. MMPs are distinguished by a conserved catalytic domain containing a zinc ion, as well as a prodomain that regulates enzyme activation by modulation of a cysteine residue within that domain. Because nitric oxide (NO) and derived reactive nitrogen species target zinc ions and cysteine thiols, we assessed the ability of NO to regulate MMPs. A dose-dependent, biphasic regulatory effect of NO on the activity of MMPs (MMP-9, -1, and -13) secreted from murine macrophages was observed. Low exogenous NO perturbed MMP/tissue inhibitor of metalloproteinase (TIMP)-1 levels by enhancing MMP activity and suppressing the endogenous inhibitor TIMP-1. This was cGMP-dependent, as confirmed by the cGMP analog 8-bromo-cGMP, as well as by the NO-soluble guanylyl cyclase-cGMP signaling inhibitor thrombospondin-1. Exposure of purified latent MMP-9 to exogenous NO demonstrated a concentration-dependent activation and inactivation of the enzyme, which occurred at higher NO flux. These chemical reactions occurred at concentrations similar to that of activated macrophages. Importantly, these results suggest that NO regulation of MMP-9 secreted from macrophages may occur chemically by reactive nitrogen species-mediated protein modification, biologically through soluble guanylyl-cyclase-dependent modulation of the MMP-9/TIMP-1 balance, or proteolytically through regulation of MMP-1 and -13, which can cleave the prodomain of MMP-9. Furthermore, when applied in a wound model, conditioned media exhibiting peak MMP activity increased vascular cell migration that was MMP-9-dependent, suggesting that MMP-9 is a key physiologic mediator of the effects of NO in this model.

Fig. 1.

NO-mediated cell migration in an ex vivo model of wound-driven angiogenic response is MMP-9-dependent. (A) Conditioned media of untreated cells or cells treated with 10 μM Sper/NO were applied to an ex vivo model of wound-driven angiogenesis, demonstrating increased vascular cell migration as indicated by the outgrowth of vascular cells (arrow) away from the perimeter of explanted tissue. (B) Quantitation of vascular cell migration in explanted tissue exposed to conditioned media of control and Sper/NO-treated ANA-1 cells shown in A, as well as deactivated Sper/NO control media and 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ)-treated cells with or without Sper/NO. (C) Enhanced vascular cell migration is suppressed in conditioned media from MMP-9 knockdown cells treated with Sper/NO. (D) Knock down of MMP-9 translation by using MMP-9 antisense morpholino completely inhibits MMP-9 translation, as demonstrated by gel zymography. In contrast, the scrambled control oligo does not affect MMP-9 translation. Symbols indicate statistical significance when compared with conditioned media of untreated control: *, P < 0.001; **, P < 0.05.

Fig. 2.

MMP regulation is NO/sGC-dependent and sGC-independent. (A) Steady-state nanomolar NO as a function of Sper/NO concentration in 10 ml of serum-free, phenol-red-free RPMI medium 1640 (pH 7.4; n = 3). (B) Exposure of resting macrophages to increasing concentrations of Sper/NO for 4 h results in biphasic and dose-dependent activation/inactivation of total MMP activity in the cell-conditioned media. sGC dependence is indicated by the ability of 10 μM 8-bromo-cGMP to increase MMP activity ≈2-fold beyond basal levels. The results are representative of at least five independent experiments. (C) Gel zymography showing MMP-9 in conditioned media of ANA-1 cells treated with Sper/NO. Suppressed signals in media samples from 100 to 1,000 μM Sper/NO suggest enzyme inactivation by high flux of NO/RNS. Sample loading was normalized per milligram of protein. (D) Basal MMP activity levels are suppressed by reducing constitutive NO levels in the presence of N^G-nitro-l-arginine methyl ester (L-NAME) (n = 3). Symbols indicate statistical significance when compared with conditioned media of untreated control at P < 0.001 (*) and P < 0.01 (**) or when compared with 10 μM Sper/NO at P < 0.001 (†).

Fig. 3.

NO suppression of endogenous TIMP-1 inhibitor is sGC-dependent. (A) Exogenous NO suppresses TIMP-1 levels in conditioned media from Sper/NO-treated ANA-1 cells. (B) Dose-dependent TIMP-1 suppression by 8-bromo-cGMP suggests that low-dose NO effects are sGC-dependent. (C) Inhibition of NO and sGC signaling by thrombospondin-1 (TSP-1) abolished NO-mediated TIMP-1 suppression. All results are representative of n = 3 blots. (D–F) Quantitation of results shown in A–C, respectively. (G) Inverse modulation of MMP activity and TIMP-1 protein levels in the conditioned media of Sper/NO-treated cells suggests perturbation of MMP-9/TIMP-1 balance by NO. Symbols indicate statistical significance when compared with untreated control: *, P < 0.05; **, P < 0.01.

Fig. 4.

NO/RNS regulation of MMP-9 activity. Biphasic and dose-dependent activation/inactivation of pro-MMP-9 by exogenous NO released from Sper/NO. The x axis shows both Sper/NO concentration and nanomolar steady-state NO measured by chemiluminescence. The results represent the mean of n = 3 measurements and are representative of two independent experiments. Symbols indicate statistical significance when compared with untreated control (*, P < 0.01) or when compared with 10 μM Sper/NO (**, P < 0.05; ***, P < 0.001). ND, not done.

Fig. 5.

iNOS inhibition by Ag augments activity levels of MMP-9 secreted from IFN-γ/LPS-stimulated ANA-1 cells. (A) Gel zymography demonstrating NO/RNS modulation of MMP-9 activity in the conditioned media of control untreated cells (lane 3), IFN-γ/LPS-stimulated cells supplemented with 1 mM l-arginine to promote generation of endogenous NO (lane 4), IFN-γ/LPS-stimulated cells supplemented with 5 mM Ag to inhibit iNOS generated NO (lane 5), or IFN-γ/LPS-stimulated cells supplemented with 5 mM Ag plus increasing concentrations of Sper/NO (lanes 6–9). Lanes 1 and 2 contain purified standard MMP-2 and MMP-9 proteins. (B) Quantitation of NO/RNS-mediated effects on MMP-9 activity measured by using internally quenched fluorescent MMP substrates (multiple enzyme multiple reagent assay) as described in Materials and Methods and ref. . MMP activity is expressed as the mean value determined from duplicate linear regression plots (fluorescence vs. time) (n = 18), as described in Materials and Methods. All sample linear regression correlation coefficients (r) were within 0.88–1.00 at P < 0.0001.