Stearate triggers metaflammation in human non-inflammatory macrophages: mitigation by empagliflozin?

Background: An oversupply of saturated fatty acids (e.g. stearic acid, SA) can trigger lipotoxicity in several cell types, a process thought to underly cardiometabolic diseases. In the latter ones, the balance between M1 (pro-inflammatory) and M2 (anti-inflammatory) macrophages (MF) plays a pivotal pathophysiologic role. Empagliflozin (EMPA) reduces the risk of cardiovascular events, possibly by also decreasing lipotoxicity. Aims: To assess in non-inflammatory subtypes (M0 and M2-IL4) of human MF the effects of 1. SA on phenotype and immunometabolism and 2. EMPA in curbing SA-induced lipotoxicity. Methods: MF from healthy donors’ monocytes were polarized to M0 and M2-IL4 and pre-incubated with/without EMPA (100µM) in the presence/absence of SA (200µM). Polarization, oxidant (OX) stress and inflammatory biomarkers were assessed by FACS and qPCR analyses. Jun kinase (JNK) activation (by western blot) and role (by JNK inhibitor) were also assessed. Steady-state bioenergetics (fluxes: pmol/min in 5×104 cells) were obtained by cell indirect microcalorimetry, which integrates the measurements of four independent fluxes [O2 consumption and H+ production by Seahorse (Agilent), lactate and NH4 releases by microfluorimetry] with the net stoichiometric equations of the main metabolic pathways. Results: SA decreased CD209 (M2-specific; -7.8±1 fold) and increased CD86 (M1-specific; +5.8±0.2 fold) polarization markers in M2 vs control (both p<0.001). Compared to control, SA increased JNK-mediated pro-inflammatory TNF-α (4.2±0.6 and 2.9±0.6 fold increases in M0 in M2, respectively; both p<0.005), IL-8 (76±16 and 25±4 fold increases in M0 in M2, respectively; both p<0.001) and IL-6 (11±3 and 2.2±0.4 fold increases in M0 in M2, respectively; both p<0.05) expression, and OX stress (p<0.05). Also, SA increased the fluxes (pmol/min) of anaerobic glycolysis (+58±12 in M0; +56±5 in M2) and ATP production (+117±24 in M0; +112±10 in M2) vs control (all p<0.05), indicating a M1-like metabolic switch. EMPA curbed IL-8, TNF-α (only in M0) and IL-6 expression (all p<0.05) – but not OX stress – in M0 and M2. Conclusions: SA remodels M2 – and to a lesser extent M0 – towards a M1-like immunometabolic profile, highlighting a possible novel pathophysiologic link between lipotoxicity and cardiometabolic risk. EMPA counteracts, at least in part, SA effects on M0 and M2, pinpointing a possible novel mechanism of action for cardiometabolic protection by this agent.