SGLT2 inhibition with empagliflozin improves coronary microvascular function and cardiac contractility in prediabetic ob/ob-/- mice

Abstract

Background: Sodium-glucose cotransporter 2 inhibitors (SGLT2i) is the first class of anti-diabetes treatment that reduces mortality and risk for hospitalization due to heart failure. In clinical studies it has been shown that SGLT2i's promote a general shift to fasting state metabolism characterized by reduced body weight and blood glucose, increase in glucagon/insulin ratio and modest increase in blood ketone levels. Therefore, we investigated the connection between metabolic changes and cardiovascular function in the ob/ob^-/- mice; a rodent model of early diabetes with specific focus on coronary microvascular function. Due to leptin deficiency these mice develop metabolic syndrome/diabetes and hepatic steatosis. They also develop cardiac contractile and microvascular dysfunction and are thus a promising model for translational studies of cardiometabolic diseases. We investigated whether this mouse model responded in a human-like manner to empagliflozin treatment in terms of metabolic parameters and tested the hypothesis that it could exert direct effects on coronary microvascular function and contractile performance.

Methods: Lean, ob/ob^-/- untreated and ob/ob^-/- treated with SGLT2i were followed for 10 weeks. Coronary flow velocity reserve (CFVR) and fractional area change (FAC) were monitored with non-invasive Doppler ultrasound imaging. Food intake, urinary glucose excursion and glucose control via HbA1c measurements were followed throughout the study. Liver steatosis was assessed by histology and metabolic parameters determined at the end of the study.

Results: Sodium-glucose cotransporter 2 inhibitors treatment of ob/ob^-/- animals resulted in a switch to a more catabolic state as observed in clinical studies: blood cholesterol and HbA1c were decreased whereas glucagon/insulin ratio and ketone levels were increased. SGLT2i treatment reduced liver triglyceride, steatosis and alanine aminotransferase, an indicator for liver dysfunction. L-Arginine/ADMA ratio, a marker for endothelial function was increased. SGLT2i treatment improved both cardiac contractile function and coronary microvascular function as indicated by improvement of FAC and CFVR, respectively.

Conclusions: Sodium-glucose cotransporter 2 inhibitors treatment of ob/ob^-/- mice mimics major clinical findings regarding metabolism and cardiovascular improvements and is thus a useful translational model. We demonstrate that SGLT2 inhibition improves coronary microvascular function and contractile performance, two measures with strong predictive values in humans for CV outcome, alongside with the known metabolic changes in a preclinical model for prediabetes and heart failure.

Keywords: Coronary; Endothelial; Microvascular; Prediabetes; SGLT2.

Fig. 1

Study design, 55 mice arrived at 6 weeks of age and acclimatized for 1 week prior to entry into the study

Fig. 2

Food intake (a) and body weight gain (b) in lean and ob/ob^−/− with and without treatment at baseline and during empagliflozin treatment. Data were analyzed using Anova with Tukey’s multiple comparisons test and are presented as mean with standard deviation. *p < 0.05 and **p < 0.01 for 5 and 10 weeks compared with baseline, and ^‡p ≤ 0.05 for ob/ob^−/− treated vs. untreated group

Fig. 3

Analysis of a HbA1c b urine glucose levels, c urine creatine and d urine glucose/creatinine during the study. In the  treated group (n = 22) HbA1c decreased, increased in ob/ob^−/− untreated group (n = 21), and did not change in lean controls (n = 12) compared with baseline levels. Data were analyzed using repeated measure Anova with Tukey’s multiple comparisons test and are presented as mean with 95% confidence interval. *p < 0.05 and **p < 0.01 represents significant difference between lean controls and ob/ob^−/− groups. ^‡p ≤ 0.05 for ob/ob^−/− treated and untreated group

Fig. 4

Coronary flow was studied over time in lean, ob/ob^−/− treated and untreated mice using non-invasive transthoracic ultrasound. Coronary flow velocity reserve (CFVR) a was calculated as the ratio of b coronary hyperemic and c basal flow velocities. Statistical significance between groups was tested using Anova, and where there were differences Turkey’s multiple comparison test was carried out. Values are presented as mean with standard deviation, where *p < 0.05 and **p < 0.01 for significant difference between lean controls and ob/ob^−/− groups. ^‡p ≤ 0.05 for ob/ob^−/− treated and untreated group

Fig. 5

Triglyceride levels and steatotic count. Triglycerides levels in liver tissue in treated (n = 18) compared to untreated (n = 19) ob/ob^−/− group (a). Steatosis score in treated (n = 12) compared to untreated (n = 10) ob/ob^−/− group (b). Data were analyzed using two-tailed Mann–Whitney test for liver triglycerides and t-test for steatosis score and are presented as mean with standard deviation. ^‡p ≤ 0.05 represents significant difference between ob/ob^−/− treated and untreated group

Fig. 6

Representative images of liver sections stained with hematoxylin–eosin showing patches of vacuolated hepatocytes. Liver section from an untreated ob/ob^−/− mouse with highly vacuolated hepatocytes and a high steatosis score (a). Liver section from a treated ob/ob^−/− mouse with reduced steatosis score (b). The scale bar equals 200 µm in each figure

Fig. 7

Heart vessel area fraction was not significantly different between treated and untreated ob/ob^−/− mice (n = 11 per group). Data analyzed using two-tailed Mann–Whitney test, and are presented as mean with standard deviation