Matrix metalloproteinase-2 (MMP-2) is a key protease in several oxidative stress related pathologies in the heart and vasculature. Activated MMP-2 does not only proteolyze extracellular matrix proteins, but also degrades susceptible intracellular substrates in the heart, including the sarcomeric protein troponin I (TnI). Fluorescence resonance energy transfer (FRET) based biosensors have been used to study the activity of many proteases. We therefore wished to apply this tool to determine intracellular MMP-2 activity.Methods
We constructed MMP-2 FRET-based biosensors by inserting amino acid sequences containing MMP-2 cleavage sites derived from human cardiac TnI between a FRET pair. Both donor (CyPet) and acceptor (YPet) fluorescent proteins were amplified by PCR and subsequently ligated into pBAD/His B vector. DNA fragments encoding potential MMP-2 cleavage sites were prepared by overlap PCR with primers appending with appropriate restriction sites and were ligated into pBAD/His B between the FRET pair. In total we designed three FRET constructs, two of which were based on TnI cleavage sites identified in our lab, containing N-terminal TnI amino acid sequences (TnI-1 and TnI-2). The third sensor contained the sequence GGSGG IPVSLRSG GGSGG, reported to be selective for MMP-2 (Control) (Turk et al. Nat Biotechnol. 2001). Following purification of FRET proteins, we determined their activity by assessing the changes in the CyPet/YPet ratio recorded upon digesting the biosensors with human recombinant MMP-2 at 37°C. We tested the specificity of these substratesfor MMP-2 by comparing the ability of other proteases (MMP-3, -9 and -13; calpain-1; caspase-3) to cleave them.Results
Kcat/Km values were determined to be 3372 M-1s-1 for TnI-2 and 700 M-1s-1 for Control. TnI-1 was not digested by MMP-2. TnI-2 and Control were selective substrates for all MMPs, but were not highly specific for MMP-2 versus other MMPs. Neither biosensor was cleaved by either calpain-1 or caspase-3. The fluorescence ratio (475/527 nm; RFU) vs. time observed for MMPs -3, -9 and -13 was essentially identical, when comparing TnI-2 and Control biosensors. However, for MMP-2 the TnI-2biosensor showed higher RFU vs. time than Control biosensor. The biosensors are now being expressed in neonatal cardiomyocytes in order to track intracellular MMP-2 activity following oxidative stress stimuli.Conclusions
A TnI-based FRET biosensor is a valuable substrate for detecting MMP activities. It will allow us to visualise and quantify MMP-2 activation within cardiomyocytes.