Publications

Although enhanced cardiac matrix metalloproteinase (MMP)-2 synthesis has been associated with ventricular remodeling and failure, whether MMP-2 expression is a direct mediator of this process is unknown. We generated transgenic mice expressing active MMP-2 driven by the alpha-myosin heavy chain promoter. At 4 mo MMP-2 transgenic hearts demonstrated expression of the MMP-2 transgene, myocyte hypertrophy, breakdown of Z-band registration, lysis of myofilaments, disruption of sarcomere and mitochondrial architecture, and cardiac fibroblast proliferation. Hearts from 8-mo-old transgenic mice displayed extensive myocyte disorganization and dropout with replacement fibrosis and perivascular fibrosis. Older transgenic mice also exhibited a massive increase in cardiac MMP-2 expression, representing recruitment of endogenous MMP-2 synthesis, with associated expression of MMP-9 and membrane type 1 MMP. Increases in diastolic [control (C) 33 +/- 3 vs. MMP 51 +/- 12 microl; P = 0.003] and systolic (C 7 +/- 2 vs. MMP 28 +/- 14 microl; P = 0.003) left ventricular (LV) volumes and relatively preserved stroke volume (C 26 +/- 4 vs. MMP 23 +/- 3 microl; P = 0.16) resulted in markedly decreased LV ejection fraction (C 78 +/- 7% vs. MMP 48 +/- 16%; P = 0.0006). Markedly impaired systolic function in the MMP transgenic mice was demonstrated in the reduced preload-adjusted maximal power (C 240 +/- 84 vs. MMP 78 +/- 49 mW/microl(2); P = 0.0003) and decreased end-systolic pressure-volume relation (C 7.5 +/- 1.5 vs. MMP 4.7 +/- 2.0; P = 0.016). Expression of active MMP-2 is sufficient to induce severe ventricular remodeling and systolic dysfunction in the absence of superimposed injury.

The theoretical dependence of the mean and standard deviation of ADC values on signal-to-noise ratio (SNR) was derived and compared to measured values in porous phantoms and the lungs of human subjects using diffusion-weighted hyperpolarized helium-3 MRI. For SNR values below 15, mean ADC values were highly SNR-dependent due to a combination of noise and choice of noise thresholding. Above SNR values of 15 and for mean ADC values within ranges relevant for evaluating lung disease (<0.6 cm2/s), the mean ADC was largely independent of SNR. The standard deviation, by contrast, was highly dependent on SNR over a much larger range, but this dependence was well predicted by theory, suggesting the histogram of ADC values might be corrected for these stochastic processes to more accurately evaluate disease using restricted diffusion measures in the lungs.