Matrix metalloproteinase-9 (MMP-9) deletion has been shown to improve remodeling of the left ventricle post-myocardial infarction (MI), but the mechanisms to explain this improvement have not been fully elucidated. MMP-9 has a broad range of in vitro substrates, but relevant in vivo substrates are incompletely defined. Accordingly, we evaluated the infarct regions of wild-type (wt) and MMP-9 null (null) mice using a proteomic strategy. Wt and null groups showed similar infarct sizes (48+/-3 in wt and 45+/-3% in null), indicating that both groups received an equal injury stimulus. Left ventricle infarct tissue was homogenized and analyzed by 2-DE and MS. Of 31 spot intensity differences, the intensities of 9 spots were higher and 22 spots were lower in null mice compared to wt (all p<0.05). Several extracellular matrix proteins were identified in these spots by MS, including fibronectin, tenascin-C, thrombospondin-1, and laminin. Fibronectin was observed on the gels at a lower than expected molecular weight in the wt group, which suggested substrate cleavage, and the lower molecular weight spot was observed at lower intensity in the MMP-9 null group, which suggested cleavage by MMP-9. Immunoblotting confirmed the presence of fibronectin cleavage products in the wt samples and lower levels in the absence of MMP-9. In conclusion, examining infarct tissue from wt and MMP-9 null mice by proteomic analysis provides a powerful and unique method to identify in vivo candidate MMP substrates.