Loss of bone morphogenetic protein-binding endothelial regulator causes insulin resistance

Abstract

Accumulating evidence suggests that chronic inflammation of metabolic tissues plays a causal role in obesity-induced insulin resistance. Yet, how specific endothelial factors impact metabolic tissues remains undefined. Bone morphogenetic protein (BMP)-binding endothelial regulator (BMPER) adapts endothelial cells to inflammatory stress in diverse organ microenvironments. Here, we demonstrate that BMPER is a driver of insulin sensitivity. Both global and endothelial cell-specific inducible knockout of BMPER cause hyperinsulinemia, glucose intolerance and insulin resistance without increasing inflammation in metabolic tissues in mice. BMPER can directly activate insulin signaling, which requires its internalization and interaction with Niemann-Pick C1 (NPC1), an integral membrane protein that transports intracellular cholesterol. These results suggest that the endocrine function of the vascular endothelium maintains glucose homeostasis. Of potential translational significance, the delivery of BMPER recombinant protein or its overexpression alleviates insulin resistance and hyperglycemia in high-fat diet-fed mice and Lepr^db/db (db/db) diabetic mice. We conclude that BMPER exhibits therapeutic potential for the treatment of diabetes.

Fig. 1. BMPER depletion causes mice to develop hyperinsulinemia, glucose intolerance, and insulin resistance.

a The generation of the BMPER iKO mouse model. The gRNA-guided CRISPR/Cas9 strategy was used for targeted deletion of bmper gene. b BMPER depletion was examined with Western blotting. c–e Fasted insulin and glucose, HOMA-IR. f, g Glucose and insulin tolerance tests. h Glucose infusion rate (GIR). i Glucose disposal rate (GDR). j, k Hepatic glucose production (HGP; j) and glucose uptake in peripheral tissues (k) were analyzed with hyperinsulinemic-euglycemic clamps. l Insulin signaling was blunted in BMPER iKO mice. Insulin (Ins, 0.5 h) was injected (i.p.) into BMPER iKO and WT mice. Indicated tissues were used for Western blotting. GM gastrocnemius muscle, BAT brown adipose tissue, WAT white adipose tissue, Hrt heart. WT, BMPER^flox/flox; CAG-CreER^–/–. iKO, BMPER^flox/flox; CAG-CreER^+/–. n = 4 mice (b), 6 mice (c–e WT), 7 mice (c–e iKO), 7 mice (f, h–k) and 5 mice (g). Data are presented as mean values ± SEM. NS not significant. Analysis was two-way ANOVA followed by Fisher’s LSD multiple comparison test (for f, g, j) or unpaired two-tailed Student’s t test (for b–e, h, i, k).

Fig. 2. BMPER depletion in ECs leads to glucose dysregulation.

Metabolic studies were performed with BMPER eKO and eWT mice at 4 months after tamoxifen injection. a Body weight (BW). b–d Fasted glucose and insulin, HOMA-IR. e Fed insulin. f, g Glucose and insulin tolerance tests. eWT, BMPER^flox/flox; Cdh5-CreER^–/–. eKO, BMPER^flox/flox; Cdh5-Cre ER^+/–. n = 9 mice (a eWT), 10 mice (a eKO), 5 mice (b–g). Data are presented as mean values ± SEM. NS not significant. Analysis was unpaired two-tailed Student’s t test (for a–e) or two-way ANOVA followed by Fisher’s LSD multiple comparison test (for f, g).

Fig. 3. BMPER plasma level decreases in metabolic syndrome patients and DIO mice.

a, b BMPER plasma levels were decreased in metabolic syndrome (MS) patients. c–f BMPER plasma level was associated with body weight (BW), insulin, and triglyceride (TG) plasma levels in MS patients. Correlation was computed for Pearson correlation coefficients with an assumption of Gaussian distribution. g, h BMPER plasma level was decreased in HFD-fed mice. n = 11 individuals (b–f) and 8 mice (h). Data are presented as mean values ± SEM. Analysis was unpaired two-tailed Student’s t test (for b, h).

Fig. 4. BMPER promotes insulin signaling through IR but not BMPR2.

a BMPER (B, 1 h) or insulin (Ins, 0.5 h) was injected (i.p.) into mice. Tissues, including liver, GM (gastrocnemius muscle), and Hrt (heart), were used for Western blotting. b Hepatocytes were transduced with adenovirus of flag-tagged BMPR2 full-length (FL) or kinase-dead mutant (KD). Cells were then treated with BMPER (1 h) and harvested for Western blotting. c Hepatocytes were isolated from IR-iKO and their littermate control (IR-WT) mice. Cells were treated with insulin (Ins, 0.5 h) or BMPER (B, 1 h) and harvested for Western blotting. IR-WT, IR^flox/flox; CAG-CreER^−/−. IR-iKO, IR^flox/flox; CAG-CreER^+/−.

Fig. 5. BMPER promotes insulin signaling through NPC1 and endocytosis.

a Co-immunoprecipitation (IP) of GFP-tagged BMPER and flag-tagged NPC1 in HEK293 cells. b Co-immunoprecipitation of GFP-tagged NPC1 constructs containing different lumenal domains (NTD, N-terminal domain, a.a. 25-164; LP2, loop 2, a.a. 371-615; LP3, loop 3, a.a. 855-1098) and flag-tagged BMPER in HEK293 cells. SP signal peptide. c Hepatocytes were treated with flag-tagged BMPER for 30 min and staining of BMPER and endogenous NPC1 was performed. d, e Hepatocytes were transduced with NPC1 shRNA lentivirus and then treated with flag-tagged BMPER (B, 1 h) or insulin (Ins, 30 min). Western blotting was performed and band intensity was quantified (e). f, g Hepatocytes were transduced with NPC1 shRNA lentivirus and then treated with flag-tagged BMPER (B, 1 h). Membrane fractions were separated and then subjected for IP with flag antibody. The associated IR with BMPER was quantified (g). h, i Hepatocytes were treated with chlorpromazine (CPM, 50 μM) or methyl-β-cyclodextrin (MCD, 10 mM). Thirty minutes later, cells were pulsed with flag-BMPER (100 nM) for 1 h and cell lysates were used for Western blotting. Internalized BMPER was quantified in (i). j, k Hepatocytes were treated with CPM for 30 min, and then treated with BMPER for the detection of IRS1 phosphorylation. n = 3 repeated experiments (e, g, i, k). Data are presented as mean values ± SEM. NS not significant. Analysis was two-way ANOVA (for e, k) or one-way ANOVA (for i) followed by Fisher’s LSD multiple comparison test and unpaired two-tailed Student’s t test (g).

Fig. 6. AAV-BMPER improves glucose responses in diabetic mice.

a BMPER plasma levels in C57BL/6 (WT), AAV-GFP, or AAV-BMPER (B)-injected mice that were fed HFD or CC diet. b, c Fasted insulin and glucose. d Glucose and insulin tolerance tests. e BMPER plasma levels in C57BL/6 (WT), AAV-GFP, or AAV-BMPER (B)-injected db/db mice. f, g Fasted insulin and glucose. h Glucose and insulin tolerance tests. i Urinary glucose levels. j Urinary albumin levels. k Urine volume. l Glucose and insulin tolerance tests. IR-WT, IR^flox/flox; CAG-CreER^–/–. IR-iKO, IR^flox/flox; CAG-CreER^+/–. n = 3 mice (a, e), 6–8 mice (b–d AAV-GFP), 6–8 mice (b–d AAV-BMPER), 5 mice (f, g, i–k), 5 mice (h AAV-GFP), 6–7 mice (h AAV-BMPER), 4 mice (l IR-WT;AAV-GFP, IR-iKO;AAV-GFP, IR-WT;AAV-BMPER), and 5 mice (l IR-iKO;AAV-BMPER). Data are presented as mean values ± SEM. Analysis was two-way ANOVA (for a–d, h, l) or one-way ANOVA (for e) followed by Fisher’s LSD multiple comparison test or unpaired two-tailed Student’s t test (f, g, i–k).

Fig. 7. Recombinant BMPER protein improves glucose responses in DIO mice.

C57BL/6 mice were fed HFD or control chow (CC) for 4 weeks and then injected with recombinant BMPER protein at 0.1 mg/kg/mouse every other day for 6 weeks. Metabolic parameters were measured, including a Body weight. b Fasted insulin. c Fasted glucose. d Glucose tolerance tests. e Insulin tolerance tests. n = 4 mice (a CC), 5 mice (a HFD), and 5 mice (b–e). Data are presented as mean values ± SEM. NS not significant. Analysis was one-way ANOVA (for a–c) or two-way ANOVA (for d, e) followed by Fisher’s LSD multiple comparison test.