Empagliflozin Attenuates Non-Alcoholic Fatty Liver Disease (NAFLD) in High Fat Diet Fed ApoE(-/-) Mice by Activating Autophagy and Reducing ER Stress and Apoptosis

Aims/hypothesis: SGLT-2 inhibitors (SGLT-2i) have been studied as potential treatments against NAFLD, showing varying beneficial effects. The molecular mechanisms mediating these effects have not been fully clarified. Herein, we investigated the impact of empagliflozin on NAFLD, focusing particularly on ER stress, autophagy and apoptosis. Methods: Five-week old ApoE(-/-) mice were switched from normal to a high-fat diet (HFD). After five weeks, mice were randomly allocated into a control group (HFD + vehicle) and Empa group (HFD + empagliflozin 10 mg/kg/day) for five weeks. At the end of treatment, histomorphometric analysis was performed in liver, mRNA levels of Fasn, Screbp-1, Scd-1, Ppar-γ, Pck-1, Mcp-1, Tnf-α, Il-6, F4/80, Atf4, Elf2α, Chop, Grp78, Grp94, Χbp1, Ire1α, Atf6, mTor, Lc3b, Beclin-1, P62, Bcl-2 and Bax were measured by qRT-PCR, and protein levels of p-EIF2α, EIF2a, CHOP, LC3II, P62, BECLIN-1 and cleaved CASPASE-8 were assessed by immunoblotting. Results: Empagliflozin-treated mice exhibited reduced fasting glucose, total cholesterol and triglyceride serum levels, as well as decreased NAFLD activity score, decreased expression of lipogenic enzymes (Fasn, Screbp-1c and Pck-1) and inflammatory molecules (Mcp-1 and F4/80), compared to the Control group. Empagliflozin significantly decreased the expression of ER stress molecules Grp78, Ire1α, Xbp1, Elf2α, Atf4, Atf6, Chop, P62(Sqstm1) and Grp94; whilst activating autophagy via increased AMPK phosphorylation, decreased mTOR and increased LC3B expression. Finally, empagliflozin increased the Bcl2/Bax ratio and inhibited CASPASE-8 cleavage, reducing liver cell apoptosis. Immunoblotting analysis confirmed the qPCR results. Conclusion: These novel findings indicate that empagliflozin treatment for five weeks attenuates NAFLD progression in ApoE(-/-) mice by promoting autophagy, reducing ER stress and inhibiting hepatic apoptosis.


Supplementary Materials
Data S1. The effect of empagliflozin on atherosclerotic plaque formation.

Background
It has been shown that patients with severe NAFLD have a 2-to 3-fold increased risk of developing T2D and CVD events [1,9] while empagliflozin, an antidiabetic agent, treatment for 3.1 years (median time) in type 2 diabetes patients with prevalent CVD, lowered the combined CVD endpoint of CVD death, non-fatal stroke and non-fatal myocardial infarction [11]. Interestingly, the observed cardio-protective effect of empagliflozin was irrespective of glycemic control [8]. To this end, we also investigated the effect of this treatment on atherogenesis, focusing on the expression of pro-inflammatory and adhesion molecules that promote the atherogenic process.

Quantification of atherosclerotic lesion area
At the end of the 5-weeks of treatment, all mice were sacrificed under isoflurane anesthesia by transection of the diaphragm and the aorta was rapidly removed, fixed and embedded in paraffin. 4μm-thick sections were stained with hematoxylin-eosin (H&E) and used for histopathological analysis. The degree of pathological changes was evaluated microscopically by measuring the area of atherosclerotic plaques. Results are reported as the percentage of the neointima area containing the lesion. Plaque area analysis was carried out using Image Pro Plus software version 5.1 (Media Cybernetics, Inc.).

Empagliflozin administration for 5 weeks did not significantly affect the atherosclerotic plaque formation
Two out of 8 mice in the Empa-group developed aortic atherosclerotic plaque lesions, while in the control-group three out of 8 mice presented with such atherosclerotic plaques (representative Figure S1.a). The mean atherosclerotic lesion area (as measured by the percentage of lumen area covered by total plaque area in all aortic root sections) was lower in the Empa-group, albeit not significantly (p = 0.17) ( Figure S1.b).
Empagliflozin treatment had no significant effect on F4/80, Icam-1 and Timp-1 mRNA levels ( Figures S1.c,d,e). The mRNA Timp-1/Mmp-2 ratio (the balance between MMPs and TIMPs is a known indicator of MMPs overall collagenolytic activity) at the end of the 5-week empagliflozin treatment was found to be higher in Empa-group compared to control-group, approaching significance (p = 0.05) ( Figure S1.f). Figure S1. Atherosclerotic plaque and mRNA expression of molecules implicated in atherogenesis in ApoE (-/-) treated with either 10 mg/kg/day Empagliflozin (Empa-group) or vehicle (control-group) for 5 weeks. a. Selected 4 μm section images from the aortic root stained with H&E (Original magnification ×40).b. Quantification of plaque area is shown as a percentage of lumina stenosis by thickened intima. Values are shown as mean ± SD. c. IL-6, TNF-α and F4/80 (inflammatory markers) mRNA expression in thoracic aorta tissues of mice. IL-6 mRNA expression was significantly decreased in Empa-group as compared to control-group. d. ICAM -1 and VCAM-1 (adhesion molecules) mRNA expression in thoracic aorta tissues of mice. A reduced expression of VCAM-1 in Empa-group approaching significance (p = 0.072), as compared to control group. e. The mRNA levels of MMP-2 and its inhibitor TIMP-1 (gelatinolytic activity) in thoracic aorta tissues. MMP-2 mRNA expression was significantly decreased in Empa-group compared to control-group. f. TIMP-1/MMP-2 mRNA ratio was also significantly increased in Empa-group as compared to control-group. qPCR data are shown as mean ± SEM. * : p < 0.05. Table S1. List of primer sequences used for RT-PCR analysis in this study.