SREBP1 induction mediates long-term statins therapy related myocardial lipid peroxidation and lipid deposition in TIIDM mice

Abstract

Statins therapy is efficacious in diminishing the risk of major cardiovascular events in diabetic patients. However, our research has uncovered a correlation between the prolonged administration of statins and an elevated risk of myocardial dysfunction in patients with type II diabetes mellitus (TIIDM). Here, we report the induction of sterol regulatory element-binding protein 1 (SREBP1) activation, associated lipid peroxidation, and the consequent diabetic myocardial dysfunction after statin treatment and explored the underlying mechanisms. In db/db mice, we observed that 40 weeks atorvastatin (5 and 10 mg/kg) and rosuvastatin (20 mg/kg) administration exacerbated diabetic myocardial dysfunction by echocardiography and cardiomyocyte contractility assay, increased myocardial inflammation and fibrosis as shown by CD68, IL-1β, Masson's staining and Collagen1A1 immunohistochemistry (IHC) staining, increased respiratory exchange ratio (RER) by metabolic cage system assessment, exacerbated mitochondrial structural pathological changes by transmission electron microscopy (TEM) examination, increased deposition of lipid and glycogen by TEM, Oil-red and periodic acid-schiff stain (PAS) staining, which were corresponded with augmented levels of myocardial SREBP1 protein and lipid peroxidation marked by 4-hydroxynonenal (4-HNE) staining. Comparable myocardial fibrosis was also observed in KK-ay and low-dose streptozotocin (STZ)-induced TIIDM mice. Elevated SREBP1 levels were observed in the heart tissues from diabetic patients, which was positively correlated with their myocardial dysfunction. To elucidate the role of statin induced SREBP1 in lipid peroxidation and lipid deposition and related mechanism, we cultured neonatal mouse primary cardiomyocytes (NMPCs) and treated them with atorvastatin (10 μM, 24 h), tracing with [U-¹³C]-glucose and evaluating for SREBP1 expression and localization. We found that statin treatment elevated de novo lipogenesis (DNL) and the levels of SREBP1 cleavage-activating protein (SCAP), reduced the interaction of SCAP with insulin-induced gene 1 (Insig1), and enhance SCAP/SREBP1 translocation to the Golgi, which facilitate SREBP1 cleavage leading to its nuclear trans-localization and activation in NMPCs. Ultimately, SREBP1 knockdown or l-carnitine mitigated long-term statins therapy induced lipid peroxidation and myocardial fibrosis in low-dose STZ treated SREBP1^+/- mice and l-carnitine treated db/db mice. In conclusion, we demonstrated that statin therapy may augment DNL by activating SREBP1, resulting in myocardial lipid peroxidation and lipid deposition.

Keywords: Myocardial lipid peroxidation; Statins; Sterol regulatory element-binding protein 1; Type 2 diabetes mellitus.

Fig. 1

Elevated blood lipids, myocardial lipid deposition, and myocardial SREBP1 upregulation in TIIDM Mice. (A) Serum lipid profile in db/db mice at 8 week and 16 weeks. n = 6 in each group. (B) Representative TEM images in the myocardium. Original magnification × 26,500, scale bar = 0.5 μm. (C) Representative Oil-Red O staining images in the myocardium. (D) Immunohistochemistry staining of SREBP1 in the heart from each group. Original magnification × 400, scale bar = 50 μm. (E–F) Representative immunoblot images of SREBP1 in mouse heart lysates and quantification. n = 6 in each group. (G) Analysis of SREBP1 protein expression in the myocardium according to immunohistochemical staining. n = 6 in each group. (H) Immunofluorescence staining of protein SREBP1 (red), Nile red (gray), and cardiomyocyte marker Troponin T (green) in the myocardium. Red arrows indicate myocardial lipid droplets; white arrows indicate SREBP1 is transferred into the nucleus. Original magnification × 1000, scale bar = 20 μm. All results were presented as the means ± SEM. Student's t-test was used for statistical analysis. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 2

SREBP1 is elevated in cardiac tissues of diabetic patients. (A–D) Representative Periodic acid–Schiff (PAS) staining, Masson's trichrome staining, SREBP1 immunohistochemical staining, and 4-HNE immunohistochemical staining images in the myocardium from normal and TIIDM patients. Black arrows indicate glycogen deposition (A), collagen deposition (B), SREBP1 nuclear translocation (C), and 4-HNE positive expression (D) in the myocardium. Original magnification × 400, scale bar = 50 μm. (E–H) Quantitative analysis of glycogen, fibrosis, SREBP1 expression, and 4-HNE expression within the myocardium. (I–K) Linear correlation analysis. Red dots indicate normal individuals and blue dots indicate TIIDM patients. n = 10 in the Normal group and n = 12 in the TIIDM group. Data are expressed as means ± SEM. A two-tailed unpaired Student's t-test was used for analysis in (E–H). For correlation analysis, linear regression models were performed and the goodness of fit for regression models was assessed using R values. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 3

Long-term statins treatment exacerbated myocardial dysfunction and elevated myocardial lipid peroxidation, inflammation and fibrosis in TIIDM mice. (A) Timeline of experimental design. Investigation of the effect of long-term administration of statins on cardiac function using the model of db/db mice, STZ mice, and KK-ay mice. In brief, three types of TIIDM mice were administered Atorvastatin at dosages of 5 or 10 mg/kg and Rosuvastatin at a dosage of 20 mg/kg daily by gavage for 40 weeks, beginning after a stable increase in blood glucose levels was observed at the start of the 10th week. Echocardiography was performed at the indicated time points.(B) Representative left ventricular M-mode echocardiographic tracings from each group of db/db mice. (C–D) Quantification of ejection fraction and fractional shortening. n = 6 in each group. (E) Detection of BNP in serum in statin-treated db/db mice. n = 6 in each group. (F) The effect of long-term administration of statins on cardiomyocyte contractility assay in db/db mice. (F–I) Representative sarcomere shortening. (F-II) Peak shortening. (F-III) Contraction velocity. (F-IV) Relaxation velocity. (F–V) Time(s) to peak (50 %). (F-VI) Time(s) to baseline (50 %). Violin plots show lines at the median (solid) and quartiles (dashed). Two-tailed unpaired t-test (A–H). n = 3 in each group. p-values are indicated. (G) Representative HE staining images in the myocardium. Black arrows indicate vacuolar degeneration and inflammatory infiltrate of cardiomyocytes. Original magnification × 400, scale bar = 50 μm. (H) Representative TEM images in the myocardium. Red arrows indicate cardiac mitochondrial swelling and condensation; blue arrows indicate myofibril breakage, lysis, and necrosis. Original magnification × 8000 or × 20,000, scale bar = 5 or 2 μm. Data are expressed as means ± SEM. One-way ANOVA with Tukey post hoc test was used for statistical analysis. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 4

Upregulation of cardiac SREBP1 lipogenesis in statin-treated db/db Mice. (A) Respiratory exchange ratio (RER) of db/db mice after statin treatment at the end of the study. n = 5 in each group. (B–C) Representative immunoblot images and quantification of ATGL, CD36, ACC2, SREBP1, ACC1, SCD1, and FASN in the heart tissues. HSP90 was used as an internal control. n = 6 in each group. (D) Immunofluorescence staining of protein SREBP1 (green) and cardiomyocyte marker Troponin T (pink) in the myocardium. White arrows indicate SREBP1 nuclear translocation. Original magnification × 1000, scale bar = 20 μm. (E–F) Immunohistochemistry staining of SREBP1 and ACC1 in the heart from each group of db/db mice. Original magnification × 400, scale bar = 50 μm. (G–H) Analysis of SREBP1 and ACC1 protein expression in the myocardium according to immunohistochemical staining. n = 6 in each group. Data are expressed as means ± SEM. One-way ANOVA with Tukey post hoc test was used for statistical analysis. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 5

Amelioration of statin-exacerbated myocardial lipid deposition in TIIDM mice by SREBP1 genetic knockdown. (A) Metabolic flux analysis of atorvastatin-treated NMPCs labeled by [U–¹³C] glucose, utilized for de novo fatty acid synthesis (left panel). The incorporation of ¹³C into Hexadecanoic acid and Octadecanoic acid are shown in the middle and right panels. NMPCs were subjected to starvation in a serum-free medium overnight, then divide into 3 groups, respectively cultured in [U–¹³C]-glucose 1.0 mmol/L (group NG), cultured in a [U–¹³C]-glucose 4.5 mmol/L (group HG), the group HG + ATO treated with [U–¹³C]-glucose 4.5 mmoL/l and atorvastatin 10 μmol/L (Sigma-Aldrich, PHR1422). All groups NMPCs were cultured for 24 h. The experiment was repeated three times. (B) Representative Oil-Red O staining images in the myocardium from each group of db/db mice. (C) Representative TEM images of left ventricular walls of the hearts from each group of db/db mice. LDs indicate lipid droplets. Yellow arrows represent lipid droplets deposited in cardiomyocytes. Scale bar = 0.5 μm. (D) Determination of cardiac tissue TCHO, LDL-C, TG, and FFA contents from each group of db/db mice. n = 5 in each group. (E) Representative left ventricular M-mode echocardiographic tracings from each group of low doses STZ-induced TIIDM mice. S–CON indicates SREBP1-deficient mice. S-STZ indicates low doses STZ-induced TIIDM mice of SREBP1-deficient mice. S-STZ + ATO10 indicates low doses STZ-induced TIIDM mice of SREBP1-deficient mice treated with atorvastatin for 30 weeks. (F–G) Representative Masson's trichrome staining and 4-HNE (IHC) images of left ventricular walls of the hearts from each group. Original magnification × 400, scale bar = 50 μm. (H–I) Quantification of ejection fraction and fractional shortening from each group. n = 6 in each group. (J) Semi-quantification analysis of fibrotic areas in the myocardium from each group. n = 6 in each group. (K) Semi-quantification analysis of 4-HNE expression from each group. n = 6 in each group. Data are expressed as means ± SEM. One-way ANOVA with Tukey post hoc test was used for statistical analysis. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 6

Long-term statins administration increases myocardial glucose accumulation in db/db mice via enhanced glycogenesis. (A–B) Representative PAS staining and the receptor of advanced glycation endproducts (RAGE) immunohistochemical staining images of left ventricular walls of the hearts from each group of db/db mice. Black arrows indicate glycogen deposition (A) and RAGE positive expression (B). (C) Heart sections treated with diastase as a negative control for PAS staining. Original magnification × 400, scale bar = 50 μm. (D) Cardiac glucose uptake was evaluated with ¹⁸F-fluorodeoxyglucose (18 F-FDG). Representative images of ¹⁸F-FDG positron emission tomography/computed tomography (PET/CT) from each group of db/db mice. The mean value standard uptake value (SUV_mean) based on region of interest (ROI) analysis was quantified via PET/CT. (E) Determination of cardiac tissue glucose contents from each group of db/db mice. n = 5 in each group. (F) Analysis of glycogen content in the heart according to PAS staining. n = 5 in each group. (G) Analysis of RAGE protein expression in the myocardium according to immunohistochemical staining. n = 5 in each group. (H) SUV_mean of mouse heart uptake of ¹⁸F-FDG. n = 3 in each group. (I) Representative immunoblot images of HK2, PFKM, PKM2, GAPDH, and GYS1 in the heart tissues. HSP90 was used as an internal control. (J) Quantification of HK2, PFKM, PKM2, GAPDH, and GYS1 protein expression in the myocardium according to immunoblot. n = 6 in each group. (K) Schematic diagram showing the changes in glycolysis-related enzymes following long-term administration of statins in db/db mice. Data are expressed as means ± SEM. One-way ANOVA with Tukey post hoc test was used for statistical analysis.

Fig. 7

Statin enhance intracellular glucose accumulation, promoting SCAP N-glycosylation and trafficking to the Golgi, leading to SREBP1 nuclear translocation. NMPCs were subjected to starvation in a serum-free medium overnight, then divided into six groups: group 1 cultured in glucose 1.0 mmol/L, group 2 cultured in glucose 4.5 mmol/L, group 3 treated with glucose 4.5 mmol/L and atorvastatin 10 mmol/L, group 4 treated with tunicamycin (1 μg/mL), group 5 treated with GlcNAc (20 mM), and group 6 treated with glucose 4.5 mmol/L, atorvastatin 10 mmol/L, and tunicamycin (1 μg/mL). All groups of NMPCs were cultured for 24 h (A–B). Representative immunoblot images and quantification of SCAP, INSIG1, SREBP1, SCD1, FASN, and ACC1. HSP90 was used as an internal control. The demonstrated bands were typical from three experimental repeats. (C) Immunofluorescence images of SCAP (green) sub-cellular localization in relation to the endoplasmic reticulum protein marker PDI (red) and nuclear DAPI staining (blue) in NMPCs with the same treatment procedure as panel A. White arrows indicate merged SCAP and PDI signals (yellow). (D) Immunofluorescence images of SCAP (green) sub-cellular localization in relation to the Golgi protein marker Golgin (red) and nuclear DAPI staining (blue) in NMPCs with the same treatment procedure as panel A. White arrows indicate merged SCAP and Golgin signals (yellow), indicating translocation of endogenous SCAP into the Golgi. (E) Immunofluorescence images of SREBP1 (red) sub-cellular localization in relation to the Golgi protein marker Golgin (green) and nuclear DAPI staining (blue) in NMPCs with the same treatment procedure as panel A. White stars indicate translocation of endogenous SREBP1 into the nucleus; white arrows indicate merged SREBP1 and Golgin signals (yellow). Original magnification × 1000, scale bars = 20 μm. (F) N-glycans of SCAP were analyzed as described in the Supplementary materials and methods section. Data are expressed as means ± SEM. One-way ANOVA with Tukey post hoc test was used for statistical analysis. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 8

l-Carnitine supplementation with statins therapy mitigated statin-induced myocardial dysfunction and lipid peroxidation. Db/db mice were administered a combination of statin at a dosage of 10 mg/kg and l-carnitine at 300 mg/kg over a period of 40 weeks. Upon completion of the study, the hearts and plasma of the subjects were analyzed. Cardiac function was assessed using echocardiography. (A) Detection of l-carnitine content in heart tissue from each group. n = 5 in each group. (B–C) Quantification of ejection fraction and fractional shortening of left ventricular M-mode echocardiographic tracings. n = 6 in each group. (D–H) Representative HE staining, Masson's trichrome staining, PAS staining, SREBP1 staining, and 4-HNE staining images of heart sections. (I–K) Semi-quantification analysis of PAS staining, SREBP1 staining, and 4-HNE staining. n = 6 in each group. Data are expressed as means ± SEM. One-way ANOVA with Tukey post hoc test was used for statistical analysis.