(A–G) βOHB levels, HMGCS2 activity, and metabolic parameters in C57BL/6J mice fasted for 6 hours and treated with vehicle (VC) or empagliflozin (EMPA; 30 mg/kg). (H and I) βOHB levels and HMGCS2 activity in 6-hour-fasted Alb^Cre and Hmgcs2^Liver–/– mice treated with VC or EMPA. (J) DARTs assay of ketogenic enzymes in liver lysates ± EMPA (3 mM). HSP90 served as loading control. (K) HMGCS2 activity in human hepatocytes treated with VC or EMPA (100 μM). (L) Dose–response activation of recombinant human HMGCS2 by 3 SGLT2i (EMPA, dapagliflozin [DAPA], ertugliflozin [ERTU]); EC₅₀ values indicated. (M and N) Kinetic analyses of HMGCS2 with SGLT2i; V_max and K_m values are shown to the right and expressed as mean ± SEM. (O) Molecular docking of EMPA highlighting interacting residues. (P) Molecular dynamics simulations showing reduced His301–Cys166 distance upon EMPA binding; top, unbound; bottom, EMPA-bound. (Q) Phosphorylation sites identified in HMGCS2 immunoprecipitates from fasted mice treated with VC or EMPA. (R) Global hepatic phosphatase activity following VC or EMPA treatment. (S) HMGCS2 activity in mice treated with OA (150 μg/kg) ± EMPA. (T) V_max of purified human WT, T91A, and T91D HMGCS2 mutants ± EMPA. Data are mean ± SEM. Statistical significance was determined using 2-tailed Student’s t test or 1-way ANOVA. *P < 0.05 vs. VC.