Abstract BACKGROUND AND AIMS Klotho (KL) is a transmembrane protein mainly expressed by the tubular cells of the kidneys, with an additional soluble form that is present in blood and urine. Preserved KL has been related to physiological protection of the kidneys. Sodium–glucose cotransporter type-2 inhibitors (SGLT2i) are a new class of anti-diabetic drugs with important renal benefits. The aim of the present study was to analyze the effect of treatment with SGLT2i on soluble KL concentrations as well as on mRNA expression levels of the KL gene in renal tubular cells. METHOD Fifty-seven patients (30 males and 27 females; mean age 61 ± 5 years) with type 2 diabetes and chronic kidney disease stage G2-A2, all treated with metformin and blockers of the renin–angiotensin system, were included in the study. All patients were in need of intensified anti-diabetic therapy at the discretion of their physicians in accordance with the routine clinical practice. Forty-five received SGLT2i during 3 months (15 empagliflozin, 15 canagliflozin and 15 dapagliflozin), and the evolution was compared with a group of 12 patients matched by age, sex and stage of CKD who received DPP4 inhibitors (control group). Serum and urine levels of soluble KL and tumor necrosis factor-alpha (TNFa) were determined by ELISA. In addition, the effect on mRNA expression levels of KL was assessed in vitro in renal tubular cells cultures. RESULTS Baseline values of serum and urinary KL and TNFa were similar in both groups. Urinary KL was inversely correlated with albuminuria (r = −0.45, P < 0.001) and urine TNFa excretion (r = −0.40, P < 0.01). At the end of the study, there was a similar improvement in metabolic control in both groups; however, only patients treated with SGLT2i showed a significant reduction in albuminuria and urinary TNFa (P < 0.01). Serum and urinary KL increased by 5.2% and 38.9%, respectively, only in patients receiving SGLT2i (P < 0.001 versus baseline), without changes in controls (P < 0.01 between groups). There were no differences among the three drugs in the SGLT2i group. The reduction in albuminuria (β = −0.35, P < 0.01) and urinary TNFa (β = −0.57, P < 0.001) were independently associated with the increase in urine KL, while the change in albuminuria was related to the increase in serum KL (β = −0.39, P < 0.05). Cultured renal tubular cells under high glucose, inflammatory stimulus with TWEAK or serum bovine albumin decreased Klotho mRNA and protein expression within 24 h, an effect that was prevented by all the SGLT2i when added to the cultures. CONCLUSION In type 2 diabetic patients, treatment with SGLT2i induces a reduction in albuminuria and modulates inflammation (as reflected by a decrease in the urinary excretion of TNFa), which is associated with a significant increase in soluble KL concentrations. In addition, SGLT2i prevent the reduction in protein and mRNA expression levels of the KL gene in renal tubular cells. The preservation of KL by SGLT2i may be an important mechanism of renal protection.