Cardiac sodium-glucose co-transporter 1 (SGLT1) contributes to heart failure in a mouse model of diabetic cardiomyopathy

Figures

• 

  Fig. 1. Full-length SGLT1 Is expressed in cardiomyocytes. a RT-PCR of SGLT1 mRNA from isolated cardiomyocytes and whole tissue from C57BL/6 J mice using primers to amplify the 5’ end (lane A, exons 1–2), the middle (lane B, exons 9–11) and the 3’ end (lane C, exons 14–15) of the transcript. Full length SGLT1 mRNA comprising all 15 exons was present in all specimens, including isolated cardiomyocytes and total cardiac tissue. The ladder (1 Kb Plus DNA Ladder, Invitrogen) sizes 100–15,000 bp double-stranded DNA. b RT-PCR products for exons 5–6 were subjected to Sanger sequencing to confirm the specificity of the amplification

• 

  Fig. 2. Cardiac SGLT1 expression is upregulated in diabetes. Cardiac membrane SGLT1 and total phosphorylated (p-SGLT1) protein was upregulated in a human subjects with type 2 diabetes (n = 3/group) and b db/db (n = 4/group) and c high fat diet (HFD) (n = 3/group) mouse models of type 2 diabetes. Eight-week-old WT and TGSGLT1-DOWN mice were fed a HFD (60% kcal from fat) or control chow (4% kcal from fat) for 20 weeks (n = 5/group). The upregulation of SGLT1 was attenuated in TGSGLT1-DOWN mice fed a HFD. α-tubulin or lamin B were used as loading controls. Each sample shown is a separate biological replicate representing an individual mouse or human

• 

  Fig. 3. Transgenic knockdown of cardiomyocyte SGLT1 attenuates cardiomyopathy in vivo. Eight-week-old TGSGLT1-DOWN mice and WT Littermates were fed a HFD or control chow for 20 weeks (n = 4–6/group, with equal numbers of males and females, as shown in the figures). HFD caused similar increases in a fasting blood glucose and b body mass in WT and TGSGLT1-DOWN mice. No significant sex differences were observed. Relative to WT mice exposed to HFD, TGSGLT1-DOWN mice were protected against c cardiac hypertrophy as assessed by the ratio of cardiac mass to tibial length after sacrifice, d left ventricular dilation and e deterioration in left ventricular ejection fraction (LVEF) on echocardiography before sacrifice, and f deterioration of diastolic function on ex vivo hemodynamic studies. TGSGLT1-DOWN mice had lower cardiac expression of Nppb on QPCR g, less cardiac fibrosis on Masson trichrome staining h, and less cardiac expression of collagen I on immunohistochemistry i and immunoblot j. Scale bars on histology images represent 100 μm. α-tubulin was used as a loading control on the immunoblot

• 

  Fig. 4. Transcriptional regulatory mechanisms of cardiomyocyte SGLT1 following exposure to high glucose and insulin. a WT mice exposed to a hyperinsulinemic-euglycemic clamp for 1 h showed an increase in cardiac sarcolemma-bound SGLT1 expression (n = 4/group). b In cardiomyocytes exposed to insulin (10 µg/ml) or high glucose (25 mM), there was a time-dependent increase in SGLT1 expression that was associated with an increase in HuR and/or HNF-1 expression. Densitometry of 3 blots is shown. *, P < 0.05 compared to respective control (n = 3/group). c–e Exposure of cardiomyocytes cells to insulin and glucose increased binding of Sp1 and HNF-1 to the SGLT1 gene promoter as assayed by chromatin immunoprecipitation (ChIP) and HuR to SGLT1 transcript assayed by RNA immunoprecipitation (RIP). *, P < 0.05 compared to respective control (n = 4/group). α-tubulin, cyclophilin, or lamin B were used as loading controls on the immunoblots. Representative immunoblots are shown

• 

  Fig. 5. ERK regulation of cardiomyocyte SGLT1 in diabetes. a Phosphorylated ERK (p-ERK) expression in cardiac tissue from WT and TG.SGLT1-DOWN mice fed a control diet or HFD. α-tubulin was used as a loading control on the immunoblot (n = 3/group). b Total cardiac protein from wildtype mice was immunoprecipitated by anti-Sp1 or anti-SGLT1 primary antibody and the immunoprecipitate probed on immunoblots for ERK1. ERK1 directly interacts with SGLT1 and Sp1 (n = 3/group). c SGLT1 phosphorylation in cardiomyocytes exposed to control (physiological) glucose, mannitol (25 mM), or high glucose (25 mM) (n = 3/group). d SGLT1 phosphorylation was decreased in cardiomyocytes exposed to the ERK inhibitor U0126 (n = 3/group). e SGLT1 expression in the sarcolemma of cardiomyocytes exposed to control (physiological) or high glucose in the absence or presence U0126 (1 μM for 24 h). Representative immunoblots are shown (n = 3/group)

• 

  Fig. 6. SGLT1 mediates PKC-Nox2 pathway in diabetic cardiomyopathy. Relative to WT mice, TGSGLT1-DOWN mice exposed to HFD exhibited less cardiac expression of Nox2 a, that was associated with less expression of active, membrane-associated PKC-β b (n = 4–7/group)

• 

  Fig. 7. SGLT1 mediates oxidative stress in the presence of high glucose and insulin. Cardiomyocytes were exposed to SGLT1 siRNA (knockdown) or scrambled siRNA (control) for 24 h, followed by exposure to glucose at concentrations 5–25 mM in the culture medium for 24 h. a There was no association between glucose concentration and SMIT1 expression. b In contrast, there was a glucose concentration-dependent increase in SGLT1 expression that was attenuated with SGLT1 knockdown. When SGLT1 was knocked down by siRNA, there was b decreased expression of phosphorylated, active PKC (p-PKC) at all glucose concentrations (*, P < 0.05 relative to control at glucose 5 mM; ✝, P < 0.05 relative to control cells at the equivalent glucose concentration) (n = 5/group), c decreased PKC activity as assayed by a commercially available colorimetric kit (*, P < 0.05 relative to control cells at the equivalent glucose concentration), and d decreased superoxide production as assayed by a commercially available fluorescent kit (n = 4/group). e Pharmacological inhibition of SGLT1 by sotagliflozin (sota, 100 nM for 24 h) abrogated the production of malondialdehyde (MDA), a marker of lipid peroxidation, in response…

• 

  Fig. 8. PKC mediates the effects of SGLT1 on oxidative stress. a Expression of SGLT1, p-PKC, p-p47 and Nox2 in cardiomyocytes after exposure to high glucose or SGLT1 cDNA transfection, with and without the PKC inhibitor chelerythrine. Densitometry reflects three separate experiments. *, P < 0.05 compared to control; ✝, P < 0.05 compared to glucose 25 mM; ‡, P < 0.05 compared to SGLT1 cDNA (n = 3/group). b Production of malondialdehyde (MDA), a marker of oxidative stress, following SGLT1 cDNA transfection with and without chelerythrine (n = 4/group)