Protease Signals in Heart Failure

Proteases are enzymes that cut proteins. In heart failure research, that activity can be approached from opposite directions: investigators can inhibit selected proteases to test whether disease-related cell behavior changes, or measure fragments produced by proteolysis as possible signals of stress and prognosis. Two recent studies illustrate those strategies. One tested aloxistatin, a broad inhibitor of cysteine proteases, in human cardiac cells and rat heart-tissue slices. The other measured N-terminal insulin-like growth factor binding protein-4 (NT-IGFBP-4), a fragment associated with PAPP-A-mediated proteolysis, in people hospitalized with acute heart failure. The studies do not test the same enzyme system, but together clarify the large gap between a laboratory mechanism and a clinically useful biomarker.

Evidence

The aloxistatin study used human cardiac fibroblasts derived from ischemic or dilated cardiomyopathy, human induced-pluripotent-stem-cell-derived macrophages, reporter cells, and living myocardial slices from rats. Most fibroblast experiments exposed cells to 100 micromolar aloxistatin for 48 hours. Aloxistatin reduced migration in fibroblasts from dilated cardiomyopathy but not ischemic cardiomyopathy. It reduced proliferation in both backgrounds without lowering general cell viability in the reported assay.

RNA sequencing identified 197 genes that differed between aloxistatin-treated fibroblasts and solvent controls. In a model of acute profibrotic activation, transforming growth factor beta-1 altered 1,527 protein-coding genes. Comparing that response with aloxistatin treatment identified 156 overlapping genes, which the drug regulated almost entirely in the opposite direction. Follow-up assays reported reduced expression of fibroblast-activation markers ACTA2 and CTGF, reduced secretion of fibronectin-1, and attenuation of the TGF-beta-1-associated rise in matrix metalloproteinase-2 secretion.

The tissue-slice experiments captured a separate inflammatory signal. Aloxistatin altered 179 genes after four hours and 1,178 after 24 hours relative to controls. Inflammation-related genes including IL6 and IL11 were among those downregulated at both time points. Reporter cells showed reduced NF-kappa-B activity, and stimulated stem-cell-derived macrophages showed lower expression of the activation marker HLA-DR. Knockdown experiments suggested that calpain-2 contributed to some effects: removing CAPN2 abolished aloxistatin's inhibition of CTGF and ACTA2, but did not account for all changes. These are preclinical results from small experimental replicates, not evidence of improved heart function or outcomes in people.

The NT-IGFBP-4 study examined a clinical signal rather than an intervention. Investigators measured baseline NT-IGFBP-4 and NT-proBNP in 309 people with acute heart failure and followed them for as long as one year for rehospitalization or all-cause death. The ingested abstract reports 81 events: 71 readmissions and 10 deaths. NT-IGFBP-4 correlated moderately with NT-proBNP and only weakly with high-sensitivity C-reactive protein and high-sensitivity cardiac troponin I.

After adjustment for clinical factors and both biomarkers, elevated NT-IGFBP-4 was associated with 30-day events, with a hazard ratio of 10.07 and a 95% confidence interval of 3.10 to 32.68. The corresponding NT-proBNP estimate was 2.68, with a confidence interval of 0.72 to 10.01, which crossed the null. For one-year events, both biomarkers retained associations: hazard ratios were 3.27 for NT-IGFBP-4 and 2.04 for NT-proBNP. A combined classification had a reported hazard ratio of 5.72. These estimates describe associations within this cohort; they do not show that measuring the marker changes outcomes.

Analysis — Mechanism and marker remain disconnected

This cross-study connection is analysis, not an established biological or clinical conclusion. Both papers make proteolysis observable, but at different levels. The aloxistatin experiments perturb broad cysteine-protease activity and then track fibroblast, inflammatory, and extracellular-matrix responses. The clinical study starts downstream, measuring a circulating fragment generated through PAPP-A-related cleavage and testing its association with later events. PAPP-A is not the calpain or cathepsin system targeted by aloxistatin, and neither study reports that aloxistatin changes NT-IGFBP-4. The useful convergence is therefore methodological: a proposed remodeling mechanism becomes more informative when paired with a reproducible human measure, but that pairing must be demonstrated rather than assumed. A stronger translational program would identify which protease produces which measurable fragment, confirm target engagement in human heart-failure samples, and prospectively test whether the measure adds calibrated information in an independent cohort. These studies supply separate pieces of that sequence, not the connecting evidence.

Limitations

The aloxistatin study did not administer the compound to people or to a living animal model of heart failure. Its human fibroblasts came from commercial donor lots, most functional experiments had roughly three to six replicates, and the rat slices modeled short-term culture stress rather than the full course of human cardiac remodeling. The 100-micromolar exposure may not represent an achievable or safe concentration in patients. Aloxistatin inhibits several calpains and cathepsins, while CAPN2 knockdown explained only part of the response. Short viability and electrophysiology assays cannot establish clinical safety.

The NT-IGFBP-4 source was available to this review only as an abstract. It does not provide the cohort's recruitment setting, baseline characteristics, event definitions, biomarker thresholds, missing-data handling, model calibration, or external validation. Only 10 deaths occurred, and the wide 30-day confidence interval indicates imprecision. Data-derived cutoffs can overstate performance in the same sample. Most importantly, association does not establish that the fragment participates in disease or that acting on its value provides benefit. Independent prospective cohorts and direct experiments linking specific proteases to circulating fragments would be needed to connect these findings.