Multiparametric CMR identifies macrophage NOS2-mediated benefits of preventive SGLT2 inhibition in a mouse model of metabolic heart disease.

Background: Sodium-glucose cotransporter 2 (SGLT2) inhibitors improve metabolic and cardiovascular outcomes, but the mechanisms remain incompletely understood. We utilized cardiac magnetic resonance (CMR) and complementary methods to investigate whether preventive SGLT2 inhibitor administration attenuates the development of metabolic heart disease in a high-fat, high-sucrose diet (HFHSD) mouse model.

Methods: Male wild type (WT) C57BL/6J mice were fed an HFHSD for 18 weeks to induce obesity, coronary microvascular disease, and diastolic dysfunction. WT mice treated preventively with an SGLT2 inhibitor, empagliflozin (EMPA), were compared to untreated WT mice, and mice fed either an HFHSD or standard chow diet with myeloid cell-specific knockout of the Nos2 gene (Nos2^LysMCre) were compared to floxed controls (Nos2^fl/fl). CMR assessed epicardial adipose tissue (EAT) volume, fatty acid composition (FAC), proton density fat fraction (PDFF), and T1, and myocardial perfusion, and strain. EAT FAC, PDFF, and T1 were quantified using an inversion-recovery multi-echo gradient echo sequence and a multi-resonance triglyceride model. EAT volume was quantified using cine images. Myocardial perfusion reserve (MPR) and strain were measured using arterial spin labeling, and displacement encoding with stimulated echoes (DENSE), respectively. Histology and flow cytometry assessed EAT remodeling and macrophage polarization.

Results: EMPA treatment reduced EAT volume (0.36 ± 0.18µL/g vs 0.61 ± 0.25µL/g, p < 0.01) and saturated fatty acid fraction (38.81 [32.83-47.71]% vs 48.06 [43.82-52.65]%, p < 0.05), increased EAT T1 (0.799 [0.764-0.859] s vs 0.755 [0.678-0.772] s, p < 0.05), and decreased EAT NOS2⁺ macrophages (34.74 [21.38-42.098^]% vs 46.36 [38.08-61.30]%, p < 0.05) compared to controls. EMPA improved diastolic strain rate (2.96 [2.61-3.99] s^-1 vs 1.68 [1.21-2.80] s^-1, p < 0.01) and adenosine MPR (2.00 ± 0.54 vs 1.37 ± 0.40, p < 0.01) compared to controls. Myeloid cell NOS2 knockout mice fed an HFHSD exhibited improved adenosine MPR (1.90 ± 0.47 vs 1.39 ± 0.38, p < 0.01) compared to floxed controls.

Conclusions: In this obesity-related metabolic heart disease model, EMPA treatment prevents cardiometabolic dysfunction by improving EAT quantity and quality, coronary microvascular function, and diastolic function. These benefits are mediated in part through macrophage NOS2.