Modulation of Renal GLUT2 by the Cannabinoid-1 Receptor: Implications for the Treatment of Diabetic Nephropathy

Abstract

Altered glucose reabsorption via the facilitative glucose transporter 2 (GLUT2) during diabetes may lead to renal proximal tubule cell (RPTC) injury, inflammation, and interstitial fibrosis. These pathologies are also triggered by activating the cannabinoid-1 receptor (CB₁R), which contributes to the development of diabetic nephropathy (DN). However, the link between CB₁R and GLUT2 remains to be determined. Here, we show that chronic peripheral CB₁R blockade or genetically inactivating CB₁Rs in the RPTCs ameliorated diabetes-induced renal structural and functional changes, kidney inflammation, and tubulointerstitial fibrosis in mice. Inhibition of CB₁R also downregulated GLUT2 expression, affected the dynamic translocation of GLUT2 to the brush border membrane of RPTCs, and reduced glucose reabsorption. Thus, targeting peripheral CB₁R or inhibiting GLUT2 dynamics in RPTCs has the potential to treat and ameliorate DN. These findings may support the rationale for the clinical testing of peripherally restricted CB₁R antagonists or the development of novel renal-specific GLUT2 inhibitors against DN.

Keywords: CB1 Receptor; Endocannabinoids; GLUT2; Renal Proximal Tubule Cells; diabetic nephropathy.

Graphical abstract

Figure 1.

Peripheral CB₁R blockade reverses diabetes-induced renal dysfunction. (A) Body weight changes in control animals treated orally with Veh in comparison with diabetic mice treated orally with JD5037 (3 mg/kg), SLV319 (3 mg/kg), or Veh for 16 weeks. (B) Fat and (C) lean body masses in mice. (D) Serum glucose and (E) insulin levels after 16 weeks of treatment. (F) Islets to pancreas area and (G) representative insulin staining of the pancreas from each treatment group. Original magnification, ×40. Scale bar, 50 μm. (H) Urine glucose levels. (I) Kidney-to-body weight ratio. (J) Urine albumin-to-creatinine ratio (ACR) and (K) BUN levels. Quantification of (L) glomerular and (M) Bowman’s space cross-sectional areas as well as (N) tubular damage. (O) Representative periodic acid–Schiff staining of the glomeruli and tubules from each treatment group. Data represent the mean±SEM from eight to ten mice per group. Scale bars, 50 μm in O, upper panel; 20 μm in O, lower panel. *P<0.05 relative to the corresponding control group treated with Veh; ^#P<0.05 relative to the corresponding STZ group treated with Veh.

Figure 2.

Peripheral CB₁R blockade abolishes diabetes-induced renal injury, inflammation, and tubule-interstitial fibrosis. The diabetes-induced elevations in urinary excretion levels of (A) clusterin, cystatin C, and IP-10 and (B) kidney TNFα mRNA and (C) renal protein expression of these injury and inflammatory markers were normalized in mice treated chronically with JD5037 or SLV319. (D) Representative renal IHC staining of clusterin, cystatin C, TNFα, and IP-10 from each group. Original magnification, ×40. Scale bar, 50 μm. Fibrosis was determined by measuring the (E) renal mRNA and (F and G) protein expression levels of collagen-1, collagen-3, TIMP-1, and fibronectin-1. Note that diabetes-induced renal fibrosis was completely ameliorated by chronic JD5037 and SLV319 treatment. Representative renal IHC staining from each group. Data represent the mean±SEM from eight to ten mice per group. Original magnification, ×40. Scale bar, 50 μm. *P<0.05 relative to the corresponding control group treated with Veh; ^#P<0.05 relative to the corresponding STZ group treated with Veh.

Figure 3.

CB₁R regulates diabetic renal tubular cell injury by affecting glucose uptake by GLUT2. The diabetes-induced increases in (A and C) GLUT2 and (B and D) PKC-β1 protein expression in the kidney were ameliorated by JD5037 and SLV319 treatment. (A) Representative renal IHC staining of GLUT2 from each group. Data represent the mean±SEM from seven to eight mice per group. Original magnification, ×40. Scale bars, 50 μm in left panel; 20 μm in right panel. *P<0.05 relative to the corresponding control group treated with Veh; ^#P<0.05 relative to the corresponding STZ group treated with Veh. (E) High glucose levels (75 mM) and (F) stimulating CB₁R by ACEA (10 μM) induced an increase in [Ca²⁺]i in hRPTCs and DhRPTCs, respectively. These effects were attenuated by preincubating the cells with JD5037 (100 nM) and/or in calcium-free media. Data represent the mean±SEM from three to four independent experiments. Elevated expression levels of (G) CB₁R and (H) GLUT2 in DhRPTCs are associated with (I) higher levels of glucose uptake. (J) Reduced GLUT2 expression in hRPTCs by siRNA resulted in (K) lower levels of glucose uptake. Either (L) high-glucose (30 mM) levels or (M) stimulating CB₁R by ACEA (10 μM) increased glucose uptake in hRPTCs, effects that were attenuated by JD5037 (100 nM) or cytochalazin B. (N) Stimulating CB₁R by ACEA (10 μM) resulted in elevated GLUT2 protein expression levels, an effect that was abolished by JD5037 (100 nM). Data represent the mean±SEM from three to four independent experiments. Real-time quantitative PCR mRNA expression levels of (O) CB₁R, (P) TNFα, and (Q) IL-18 in hRPTCs exposed to high-glucose conditions in the absence/presence of CB₁R activation by ACEA and/or its blockade by JD5037. (R) JD5037 inhibited the ACEA-reduced cell viability in hRPTCs in the absence/presence of high glucose. Data represent the mean±SEM from six independent replicates. *P<0.05 relative to normal glucose concentrations (5.5 mM); ^#P<0.05 relative to the corresponding ACEA treatment.

Figure 4.

Peripheral CB₁R blockade reduces diabetes-induced renal dysfunction in type 1 diabetic Akita mice. (A) Islet to pancreas area and (B) representative insulin staining in the pancreas in 3.5-month-old Ins2+/C96Y Akita mice treated with JD5037 (3 mg/kg orally) or Veh for 7 weeks in comparison with their wild-type nondiabetic control animals. Original magnification, ×40. Scale bar, 50 μm. Akita mice were characterized by (C) elevated serum glucose and (D) reduced serum insulin levels as well as (F) glycosuria. (E) Whereas JD5037 was unable to reduce the increased kidney-to-body weight ratio, both (G) the increased albumin-to-creatinine ratio (ACR) and (H) the high levels of BUN were significantly reduced in JD5037-treated Akita diabetic mice. JD5037 also normalized (I) creatinine clearance and (J) the urine excretion-to-water consumption ratio. Diabetes-induced (K) renal injury, (L) fibrosis, and (M) inflammation were attenuated by JD5037 treatment. These changes were associated with modulating the mRNA and protein expression levels of (N–R) GLUT2 and (S–U) PKC-β1 in the kidney. Data represent the mean±SEM from five to eight mice per group. *P<0.05 relative to wild-type nondiabetic littermate controls. ^#P<0.05 relative to Akita diabetic mice treated with Veh.

Figure 5.

CB₁R regulates the basal to apical translocation of GLUT2 in RPTCs. Fluorescent images of MDCK II cell cysts, expressing GLUT2-mCherry fusion protein and cultured in Matrigel, show that both (A and B) high-glucose levels (75 mM) and (C and D) CB₁R stimulation by ACEA (10 μM) induced basal to apical translocation of GLUT2, which was inhibited by pre-exposing the cysts to JD5037 (100 nM). Data represent the mean±SEM from five to ten cells in each cyst and three to four cysts per treatment. Scale bar, 20 μm. (E and F) Fluorescent images of kidney sections from high STZ (185 mg/kg intraperitoneally)–induced diabetic mice treated with JD5037 (3 mg/kg orally) or Veh in comparison with their Veh-treated nondiabetic controls indicate that hyperglycemia induced the apical translocation of GLUT2, which was inhibited by JD5037 treatment. Data represent the mean±SEM from eight to ten tubules per treatment. Arrows indicate BLM, and arrowheads indicate BBM.*P<0.05 relative to the corresponding basal group from all treatments; ^#P<0.05 relative to the corresponding apical group from all treatments.

Figure 6.

RPTC-CB₁R^−/− mice are protected from low-dose, STZ-induced diabetes and DN. RPTC-CB₁R^−/− mice underwent the same experimental procedure as mice treated chronically with CB₁R blockers described in Figure 1. RPTC-CB₁R^−/− mice had reduced (A and C) GLUT2 and (B and D) PKC-β1 expression levels in the kidney. (A) Representative renal IHC staining of GLUT2 from each group. Original magnification, ×40. Scale bars, 50 μm in left panel; 20 μm in right panel. (B) Representative blots from eight to ten mice per group. (E) Serum glucose and (F) insulin levels after 16 weeks of treatment. Reduced (G) glycosuria, (H) kidney-to-body weight ratio, (I) BUN levels, and (J) albumin-to-creatinine ratio (ACR). These improvements were associated with lower (K) renal injury, (L) fibrosis, and (M) inflammation in RPTC-CB₁R^−/− mice. Data represent the mean±SEM from eight to ten mice per group. *P<0.05 relative to animals treated with Veh within the same genotype; ^#P<0.05 relative to wild-type animals receiving the same treatment.

Figure 7.

RPTC-CB₁R^−/− mice develop a similar degree of diabetes after a high dose of STZ but are protected from the development of DN. RPTC-CB₁R^−/− mice were treated with a single high dose of STZ (185 mg/kg intraperitoneally). Comparable serum levels of (A) glucose and (B) insulin were measured in both mouse strains exposed to STZ. Reduced (C and E) GLUT2 and (D and F) PKC-β1 expression levels were observed in the kidneys of the RPTC-CB₁R^−/− mice. (C) Representative renal IHC staining of GLUT2 from each group. Original magnification, ×40. Scale bars, 50 μm in left panel; 20 μm in right panel. (D) Representative blots from four to six mice per group. (G) Elevated urine glucose levels, (H) reduced BUN levels, and (I) reduced albumin-to-creatinine ratio (ACR) were noted in the RPTC-CB₁R^−/− mice. (J) Normal mRNA expression levels of renal injury (lipocalin and TGF-β), fibrotic (fibronectin 1), and inflammatory (IP-10, IL-18, and MCP-1) markers were noted in the diabetic RPTC-CB₁R^−/− mice. (K) The survival rate of the high-dose STZ-injected RPTC-CB₁R^−/− mice was higher compared with their wild-type littermate animals. Data represent the mean±SEM from four to six mice per group. *P<0.05 relative to animals treated with Veh within the same genotype; ^#P<0.05 relative to wild-type animals receiving the same treatment.

Figure 8.

Proposed mechanism for the involvement of CB₁R in regulating hyperglycemia-induced GLUT2 translocation to the BBM in RPTCs. (Upper panel) High glucose levels induce the activation CB₁R-G_q/11 protein, which results in the release of intracellular stores of calcium from the endoplasmic reticulum (ER). Increased calcium activates PKC-β1, which mediates the apical translocation of GLUT2, and the increased entry of glucose from the tubule lumen into the RPTCs, leading to tubular injury and DN. (Lower panel) Pharmacologic blockade or genetic deletion of CB₁R in the RPTCs results in reduced activation of G_q/11-coupled protein, and consequently, fewer glucose molecules enter the RPTCs. This, in turn, protects the cells from the deleterious effects of hyperglycemia.