Empagliflozin modulates the expression of renal ion transporters and channels and partially reverses metolazone-induced hypomagnesemia

Background: Hypomagnesemia can be caused by thiazide and thiazide-like diuretics that inhibit the renal NaCl cotransporter (NCC) expressed along the distal convoluted tubule (DCT). The mechanism is believed to be primarily induction of DCT atrophy that lowers Mg 2+ reabsorption through the TRPM6/7 channel. Inhibitors of the sodium-coupled glucose transporter 2 (SGLT2), normally used to treat diabetes, have been reported to increase plasma Mg 2+ in humans. We tested the hypothesis that the SGLT2 inhibitor empagliflozin (EMPA) suppresses hypomagnesemia induced by the thiazide-like diuretic metolazone (MET) and modulates expression of renal ion transport proteins. Methods: 64 male C57BL/6 mice were randomized into 4 groups of 16 mice (1: control, 2: MET, 3: EMPA, 4: MET+EMPA) and placed on a 0.08% Mg 2+ diet. For the entire study, mice in group 1 received no drugs, while mice in groups 3 & 4 were given MET (250 mg/kg diet) in their diet. From day 7 to day 28, EMPA (300 mg/kg diet) was added to the diets of the mice in groups 2 and 4 only. On day 28, 24-hour urine was collected and on day 29 mice were euthanized. Plasma and urine Mg 2+ and other electrolytes were measured. Kidneys were homogenized for Western blots for total NCC, phosphorylated NCC, and TRPM6 (expressed along DCT), and total NKCC2, phosphorylated NKCC2, γ-ENaC, α-ENaC, Calbindin, phosphorylated AQP2, total AQP2, NHE3, and NAPI2 (expressed in other segments). Results: Compared to controls, plasma Mg 2+ was significantly lower in MET mice and higher in EMPA mice; in MET+EMPA mice it was significantly higher than in MET mice and not different to controls, suggesting EMPA reversed MET-induced hypomagnesemia. Compared with control mice, plasma K + trended lower in MET mice. Plasma [TCO 2 ] was significantly higher in both MET mice and MET+EMPA mice than in controls and EMPA mice suggesting EMPA does not reverse MET-induced metabolic alkalosis. Consistent with strong diuretic effects, urine volume and urinary Na + , K + , and Mg 2+ excretion were higher in all drug groups compared with control. Compared with controls, abundances of total NCC, TRPM6, and calbindin were significantly lower in both MET and MET+EMPA mice but did not differ for EMPA mice, suggesting effects on other segments contribute to correction of hypomagnesemia. Abundance of total NKCC2 was significantly lower in MET mice compared with EMPA mice and controls, but did not differ from that of the MET+EMPA mice. Abundance of phosphorylated NKCC2 did not differ between groups. Uncleaved γ-ENaC was higher in EMPA mice but lower in MET and MET+EMPA mice. Abundance of phosphorylated AQP2 was significantly lower in MET-treated mice, and higher in EMPA mice; in MET+EMPA mice it did not differ from controls. Abundances of total AQP2, NHE3, and NAPI2 did not differ in any group compared with control mice. These data suggest that while EMPA may be a helpful approach to correct drug-induced hypomagnesemia in patients, it exerts complex effects on renal ion channels and transporters due to increased water and electrolyte excretion. American Heart Association 23EIA1037600 This abstract was presented at the American Physiology Summit 2025 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.