Proximal Tubular Cannabinoid-1 Receptor Regulates Obesity-Induced CKD

BASIC RESEARCH www.jasn.org

† Shiran Udi,* Liad Hinden,* Brian Earley, Adi Drori,* Noa Reuveni,* Rivka Hadar,* † Resat Cinar, Alina Nemirovski,* and Joseph Tam*

*Obesity and Metabolism Laboratory, Institute for Drug Research, School of Pharmacy, Faculty of Medicine, The Hebrew University of Jerusalem, Jerusalem, Israel; †Laboratory of Physiological Studies, National Institute on Alcohol Abuse and Alcoholism, Bethesda, Maryland

ABSTRACT Obesity-related structural and functional changes in the kidney develop early in the course of obesity and occur independently of hypertension, diabetes, and dyslipidemia. Activating the renal cannabinoid-1 receptor (CB1R) induces nephropathy, whereas CB1R blockade improves kidney function. Whether these effects are mediated via aspecific cell type within the kidney remains unknown. Here, we show that specific deletion of CB1R in the renal proximal tubule cells did not protect the mice from obesity, but markedly attenuated the obesity-induced lipid accumulation in the kidney and renal dysfunction, injury, inflammation, and fibrosis. These effects associated with increased activation of liver kinase B1 and the energy sensor AMP- activated protein kinase, as well as enhanced fatty acid b-oxidation. Collectively, these findings indicate that renal proximal tubule cell CB1R contributes to the pathogenesis of obesity-induced renal lipotoxicity and nephropathy by regulating the liver kinase B1/AMP-activated protein kinase signaling pathway.

J Am Soc Nephrol 28: 3518–3532, 2017. doi: https://doi.org/10.1681/ASN.2016101085

The prevalence of obesity and its comorbidities has in the renal proximal tubular cells (RPTCs),12,18,19 risen dramatically within the past two decades. Re- and their blockade with CB1R antagonists improves cently, more attention has been devoted to obesity- renal function and reduces albuminuria and glo- associated renal structural and functional changes, merular lesions in obese and diabetic mouse which develop early in the course of obesity.1–3 In models.14,15,20–23 In vitro,themaineCB(arachidonoyl fact, obese individuals have a greater risk to develop ethanolamide) increases RPTC hypertrophy, whereas 4 19 ESRD, especially in combination with diabetes and CB1R blockade reduces it, and ameliorates palmitic hypertension, which account for .70% of ESRD.5,6 acid-induced apoptosis of these cells.18 However, most Obesity induces hemodynamic and morphologic of these studies failed to determine whether the changes in the kidney,7 which together with renal eCB system also plays a role in obesity-associated inflammation8 and oxidative stress,9 may lead to re- renal pathologies. If it does, is it mediated centrally, duced renal function and ultimately, glomerulosclerosis peripherally, or via a specificcelltypewithinthe and tubulointerstitial fibrosis.2,7,10,11 Although multi- kidney? ple metabolic factors have been proposed to contribute to obesity-induced nephropathy, the underlying signaling mechanisms are not completely understood. Received October 10, 2016. Accepted June 21, 2017.

Endocannabinoids (eCBs) are endogenous lipid Published online ahead of print. Publication date available at ligands that interact with the cannabinoid-1 recep- www.jasn.org. tor (CB R), abundantly expressed in the brain and 1 Correspondence: Dr. Joseph Tam, Obesity and Metabolism 12 periphery, including the kidney. Accumulating Laboratory, The Institute for Drug Research, School of Pharmacy, evidence shows the potential role of the eCB system Faculty of Medicine, The Hebrew University of Jerusalem, PO Box 13 12065, Jerusalem 91120, Israel. Email: [email protected] in various renal pathologies. CB1Rs are expressed – in podocytes,14 16 mesangial cells,17 and particularly Copyright © 2017 by the American Society of Nephrology

3518 ISSN : 1046-6673/2812-3518 JAmSocNephrol28: 3518–3532, 2017 www.jasn.org BASIC RESEARCH

Figure 1. High-fat diet-induced obesity results in increased renal eCB tone. CB1R gene locus containing ﬂox sites (red triangles) with a forward G50 primer and reverse G51 and G53 primers (black arrows). Cre recombination results in a 600 bp product, whereas no 2/2 recombination results in a 500 bp product (A). Cre recombination is shown only in RPTCs isolated from RPTC-CB1R mice and not in cells extracted from wild-type mice or in tissues such as liver, brain, muscle, inguinal, and retroperitoneal fat pads and pancreas.

(B). Loss of CB1R protein expression speciﬁcally in RPTCs (black arrows), with its normal expression in the distal tubule and glo- 2/2 merulus in RPTC-CB1R mice compared with their wild-type littermate control animals (C). Whole renal CB1RmRNAexpression 2/2 levels were greatly reduced in RPTC-CB1R mice compared with their wild-type littermate controls (D). HFD feeding induced a parallel increase in renal arachidonoyl ethanolamide and 2-arachidonoylglycerol levels in both mouse strains (E). CB1RmRNA 2/2 expression levels were greatly reduced in RPTC-CB1R mice under both diets (F). The HFD-induced increase in CB1R protein expression levels in wild-type control animals was limited to the RPTCs (black arrows) and not to distal tubules or glomerulus (G). 2/2 NotetheabsenceofHFD-inducedupregulationinCB1R protein expression in the RPTC-CB1R mice. Scale bar, 20 mm. Data represent the mean6SEM from 8 to 10 mice per group. *P,0.05 relative to animals on STD of the same genotype; #P,0.05 relative to wild-type animals on the same diet. HET, heterozygous; NC, negative control; PC, positive control.

To address these questions, we focused on the RPTCs re- manipulating RPTC-CB1R may be a novel therapeutic ap- sponsible for active renal reabsorption by a mechanism proach for treating obesity-induced nephropathy. requiring a large amount of energy derived from fatty acid b-oxidation.24 RPTCs are particularly sensitive to the accumulation of lipids, an effect called lipotoxicity, which is regu- RESULTS lated by the cellular energy and redox sensor, AMP-activated 25,26 protein kinase (AMPK). AMPK inactivates acetyl-CoA Generating RPTC-SpecificCB1R-Null Mice carboxylase (ACC) by its phosphorylation, consequently re- To explore the role of CB1R on the RPTCs in mediating obesity- ducing malonyl CoA, leading to increased activity of carnitine induced renal dysfunction, we generated RPTC-specific 2/2 palmitoyl transferase-1. This increases the translocation of CB1R-null mice (RPTC-CB1R ) by crossing CB1R-floxed fatty acids across the mitochondrial membrane, thereby in- mice with transgenic iL1-sglt2-Cre mice,28 specifically ex- creasing fatty acid b-oxidation.27 Our findings suggest that pressing Cre recombinase in the brush border membrane of

J Am Soc Nephrol 28: 3518–3532, 2017 Cannabinoid-1 Receptor Regulates Renal Lipotoxicity 3519 BASIC RESEARCH www.jasn.org

2/2 2/2 Figure 2. RPTC-CB1R mice are sensitive to HFD-induced obesity and have similar metabolic abnormalities. RPTC-CB1R mice gained a similar body weight (A and B) and displayed equal fat (C) and lean (D) body masses as their wild-type littermate controls. Loss of CB1Rin RPTCs did not alter glucose homeostasis, as measured by fasting blood glucose (E), serum insulin levels (F), and the glucose tolerance test

(G). An insulin sensitivity test revealed a signiﬁcant improvement in the null mice on HFD (H). The absence of CB1R in RPTCs did not change 2/2 the lipid proﬁle, as determined by the serum cholesterol levels (I) as well as the HDL/LDL cholesterol ratio (J). RPTC-CB1R mice exhibited a similar liver dysfunction on HFD, as measured by the serum levels of ALT (K) and AST (L). Data represent the mean6SEM from 8 to 10 mice pergroup.*P,0.05 relative to animals on STD of the same genotype.

2/2 the S1 segment of the proximal tubule. Genomic DNA anal- was significantly reduced in the RPTC-CB1R mice (Figure ysis revealed that CB1R gene deletion occurred specifically in 1, F and G). However, this did not affect the susceptibility of 2/2 RPTCs isolated from mouse kidneys, whereas its deletion was the mice to develop obesity because both the RPTC-CB1R excluded from the liver, brain, muscle, fat pads, and pancreas mice and their wild-type littermates developed a similar de- (Figure 1, A and B). Additionally, RPTCs were completely gree of obesity (Figure 2, A and B), associated with increased devoid of CB1R protein expression, whereas its expression fat and reduced lean body mass (Figure 2, C and D). Both pattern remained normal in the distal tubular cells and obese mouse strains showed equal disturbances in glucose glomerulus (Figure 1C). Likewise, the mRNA expression of homeostasis (Figure 2, E and G). Surprisingly, the obese CB1Rinthewholekidneywassignificantly reduced (Figure knockouts were more sensitive to insulin (Figure 2H). A sim- 1D), indicating its substantial expression in RPTCs.12 ilar degree of hypercholesterolemia, a low HDL/LDL ratio, and liver injury were measured in both strains (Figure 2, I–L). An 2/2 RPTC-CB1R Mice Are Sensitive to Obesity Induced exacerbated metabolic phenotype was found when the mice by High-Fat Diet maintained on HFD for 43 weeks, although slight metabolic 2/2 When maintained on a high-fat diet (HFD) for 14 weeks, improvements were noted in RPTC-CB1R mice (Supple- an equal elevation in renal arachidonoyl ethanolamide and mental Figure 1, A–H). No differences in BP were found between 2-arachidonoylglycerol levels was documented in both mouse wild-type mice on standard diet (STD) and HFD, as well as be- strains (Figure 1E). Importantly, the renal expression of CB1R tween the two mouse strains (Supplemental Figure 2).

3520 Journal of the American Society of Nephrology J Am Soc Nephrol 28: 3518–3532, 2017 www.jasn.org BASIC RESEARCH

2/2 Figure 3. RPTC-CB1R mice are protected from obesity-induced morphologic changes in the kidney, renal dysfunction, and injury. The HFD-induced morphologic changes in the kidney, such as glomerular enlargement (A and G), an increased Bowman’s space area 2/2 (B and G), and a mesangial expansion (C and G) were attenuated in obese RPTC-CB1R mice. Similarly, the HFD-induced kidney dysfunction, as manifested by the increased albumin-to-creatinine ratio (D), creatinine clearance (E), and BUN (F), was markedly attenuated in 2/2 the obese RPTC-CB1R mice. The HFD-induced upregulation in renal mRNA and protein expression levels as well as the urine excretion 2/2 levels of the kidney injury markers TIMP-1 (H and J–L), and KIM-1 (I and M–O) were attenuated or normalized in RPTC-CB1R mice on the same diet. Scale bar, 20 mm. Original magniﬁcation, 33 in G, H, and I. Data represent the mean6SEM from 8 to 10 mice per group. *P,0.05 relative to animals on STD of the same genotype; #P,0.05 relative to wild-type animals on the same diet.

J Am Soc Nephrol 28: 3518–3532, 2017 Cannabinoid-1 Receptor Regulates Renal Lipotoxicity 3521 BASIC RESEARCH www.jasn.org

2/2 Figure 4. RPTC-CB1R mice are protected from obesity-induced morphologic changes in the kidney, renal dysfunction and injury even after a very prolonged (43 weeks) duration of HFD feeding. The HFD-induced morphologic changes in the kidney, such as glomerular enlargement (A and B), an increased Bowman’s space area (A and C), and mesangial expansion (A and D) were attenuated 2/2 in obese RPTC-CB1R mice. The HFD-induced kidney dysfunction, as manifested by the increased albumin-to-creatinine ratio (E), 2/2 creatinine clearance (F), and increased BUN (G) was signiﬁcantly attenuated in the obese RPTC-CB1R mice. The HFD-induced upregulation in the renal mRNA expression levels of the kidney injury markers Timp-1 (H), Lcn2 (I), and Clu (J), as well as in the urinary

3522 Journal of the American Society of Nephrology J Am Soc Nephrol 28: 3518–3532, 2017 www.jasn.org BASIC RESEARCH

Metabolically, both obese mouse strains exhibited a similar with others,26,29 quantifying vacuolated proximal tubules in alteration in the basal metabolic rate compared with those on HFD-fed mice revealed a high accumulation of lipid droplets 2/2 STD (Supplemental Figure 3, A–D). Combined with a similar in wild-type mice, but not in the obese RPTC-CB1R ani- activity profile and food and water intake (Supplemental Figure mals, after either 14 (Figure 6, A and D) or 43 (Supplemental 3, E–H), these findings suggest that CB1RinRPTCsdoesnot Figure 7, A and C) weeks on HFD. Molecularly, CB1R is a key contribute to obesity and its associated comorbidities in mice. regulator of AMPK in different tissues,30–32 and modulating this signaling pathway affects both fatty acid b-oxidation and 2/2 25,26,29 RPTC-CB1R Mice Are Protected from Obesity- lipogenesis in the kidney. Likewise, we found a signifi- Induced Kidney Injury cant reduction in the activated/phosphorylated form of AMPK Enlarged glomerular and Bowman’sspace(Figure3,A,B,andG) after 7 days (Figure 6, G and I), 14 weeks (Figure 6, B and E), and areas, increased mesangial expansion (Figure 3, C and G), as well even 43 weeks (Supplemental Figure 7, B and D) on HFD. More- as markedly elevated albumin-to-creatinine ratio (Figure 3D), over, the expression of the nonactivated/phosphorylated form of creatinine clearance (Figure 3E), and BUN levels (Figure 3F) its downstream substrate, ACC, was dramatically downregulated were found in obese wild-type control mice. These changes under the same conditions (14 weeks: Figure 6, C and F; 7 days: were associated with marked elevations in several kidney injury Figure 6, H and J), an effect that may promote lipid accumula- markers (Figure 3, H–O, Supplemental Figure 4). All of these tion in the kidney. Importantly, these effects were completely 2/2 2/2 effects were normalized/attenuated in the obese RPTC-CB1R absent in the obese RPTC-CB1R mice, as well as in obese animals. Despite increased obesity after a prolonged HFD feeding wild-type mice treated with the peripherally restricted CB1R (43 weeks), little further progression of renal damage was ob- antagonist, JD5037 (Figure 6, G–J).33 Increasing the in vivo pro- served in obese wild-type mice (Figure 4, A–J, L, and M). Kid- duction of eCBs by JZL19534 resulted in a significant reduction ney-to-body weight ratios were reduced in both mouse strains on in the phosphorylated AMPK and ACC in the kidney (Supple- HFD, although significantly less reduction was documented in the mental Figure 8). 2/2 RPTC-CB1R mice, probably attributable to their larger kid- Exposing RPTCs to a mixture of oleate/palmitate for 24 neys on baseline conditions (Supplemental Figure 5). hours in vitro greatly elevated the levels of 2-arachidonoylglycerol and its endogenous precursor and degrading ligand, AA 2/2 RPTC-CB1R Mice Are Protected from Obesity- (Figure 7A), and the expression of CB1R (Figure 7, B and C). Induced Renal Inflammation and Tubulointerstitial These changes were associated with elevated expression levels of Fibrosis several proinflammatory cytokines (Figure 7D). Likewise, di- Next, measuring the renal expression levels of several proinflam- rectly activating CB1R by the synthetic agonist arachidonyl-2’- matory (Figure 5, A–F) and profibrotic markers (Figure 5, G, H, chloroethylamide (ACEA) upregulated these markers (Figure and J–L) revealed marked increases in obese wild-type controls 7E). Both effects were completely normalized by pretreating 2/2 and normal expression in obese RPTC-CB1R mice. Moreover, the cells with JD5037. In accordance with the in vivo findings, quantifying renal fibrosis revealed a significant collagen deposition we also found that JD5037 completely reversed the oleate/pal- 2/2 in wild-type animals on HFD, but not in RPTC-CB1R mice mitate– andACEA-induceddownregulationinboththe (Figure 5, I and M). A similar pattern was documented in wild- pAMPK/AMPK and pACC/ACC ratios (Figure 7, F, H, I, K, type mice fed HFD for 43 weeks (Figure 4, K and N–R). Enhanced and L), and consequently reduced the accumulation of fat drop- renal expression and localization of the proliferative and macro- lets in the cells (Figure 7M). phage markers, Ki67 and F4/80, respectively, was only found in obese wild-type mice (Supplemental Figure 6), suggesting a key Role of the Gi-PKA-LKB1 Axis in Activating AMPK by fl role of RPTC-CB1R in mediating HFD-induced renal in amma- CB1R Deletion/Blockade in RPTCs tion, macrophage infiltration, and tubulointerstitial fibrosis. Two upstream kinases, the tumor suppressor liver kinase B1 (LKB1) and the Ca2+/calmodulin-dependent kinase kinase

CB1R Deletion/Blockade Ameliorates Lipid b (CaMKKb), modulate AMPK activity by inducing its phos- Accumulation in RPTCs via an AMPK/ACC Signaling phorylation.35,36 In vivo, HFD reduced the phosphorylated Pathway (activated) form of LKB1 on Ser428 residue in the kidney, whereas Asigniﬁcant feature of obesity-related renal dysfunction is the genetic deletion or pharmacologic blockade of CB1Rameliorated accumulation of lipid droplets in the kidney. In accordance it (Figure 6, K–N, Supplemental Figure 7, E and F). Similarly,

2/2 excretion levels of TIMP-1 (L) and KIM-1 (M) were normalized in RPTC-CB1R mice on the same diet. Similarly, the HFD-induced renal 2/2 fibrosis was attenuated in RPTC-CB1R mice, as ascertained by measuring the renal mRNA expression levels of the profibrotic markers Col1 (N), Col3 (O), aSma (P), and Fn1 (Q), as well as by quantifying collagen deposition by Sirius Red staining (K and R). Scale bar, 20 mm. Original magnification, 33inAandK.Datarepresentthemean6SEM from 5 to 10 mice per group. *P,0.05 relative to animals on STD of the same genotype; #P,0.05 relative to wild-type animals on the same diet.

J Am Soc Nephrol 28: 3518–3532, 2017 Cannabinoid-1 Receptor Regulates Renal Lipotoxicity 3523 BASIC RESEARCH www.jasn.org

2/2 Figure 5. RPTC-CB1R mice are protected from obesity-induced tubulointerstitial inflammation and fibrosis. The HFD-induced upregulation in the renal mRNA and the protein expression levels of the inflammatory markers IL-18 (A, B, and C), TNFa (A, B, and D), 2/2 iNOS (A, B, and E), and MCP-1 (A, B, and F) were attenuated or normalized in the RPTC-CB1R mice on the same diet. The HFD- induced upregulation in the renal mRNA and protein expression levels of the fibrogenic markers collagen-1 (G, H, and J), collagen-3 (G, 2/2 H, and K), and aSMA (G, H, and L) were attenuated or normalized in the RPTC-CB1R mice on the same diet. Similarly, an elevated 2/2 collagen deposition, measured by Sirius Red staining, was evident in the HFD-fed wild-type controls compared with the RPTC-CB1R mice on the same diet (I and M). Scale bar, 20 mm. Original magnification, 33 for all inserts. Data represent the mean6SEM from 8 to 10 mice per group. *P,0.05 relative to animals on STD of the same genotype; #P,0.05 relative to wild-type animals on the same diet.

3524 Journal of the American Society of Nephrology J Am Soc Nephrol 28: 3518–3532, 2017 www.jasn.org BASIC RESEARCH

Figure 6. Genetic deletion or pharmacologic blockade of CB1R ameliorates lipid accumulation in RPTCs via aLKB1/AMPK/ACCsignaling 2/2 pathway. A high accumulation of lipid droplets was found in obese wild-type mice, and not in RPTC-CB1R animals (A and D). Significant reductions in the expression of the activated/phosphorylated form of AMPK after both long (14 weeks; B and E) and short (7 days; G and I) periods of HFD feeding and in the nonactivated/phosphorylated form of ACC (14 weeks: C and F; 7 days: H and J) as well as in the activated/ phosphorylated form of LKB1 (14 weeks: K and L; 7 days: M and N) were found in obese wild-type mice. Genetic deletion of CB1R in RPTCs (B,C,E,F,K,andL)ortheperipheralblockadeofCB1R by JD5037 (3 mg/kg, administered orally) (G–J, M, and N) significantly normalized the expression levels of pLKB1, pAMPK, and pACC. Scale bar, 20 mm. Original magnification, 33 for all inserts. Data represent the mean6SEM from 8to10micepergroup.*P,0.05 relative to animals on STD of the same genotype; #P,0.05 relative to wild-type animals on the same diet.

J Am Soc Nephrol 28: 3518–3532, 2017 Cannabinoid-1 Receptor Regulates Renal Lipotoxicity 3525 BASIC RESEARCH www.jasn.org

Figure 7. Peripheral CB1R blockade reverses inflammation and lipid accumulation in RPTCs via AMPK signaling. Increased activity of the eCB system was reflected by the elevated levels of 2-arachidonoylglycerol and its endogenous precursor and degraded ligand AA, but not arachidonoyl ethanolamide (A) as well as the upregulation of mRNA (B) and the protein (C) expression levels of CB1RinHK-2 cells exposed to 0.5 mM oleate/palmitate for 24 hours. This elevation was associated with upregulating the mRNA expression levels of several kidney injury and inflammatory markers, such as Tnfa, Il-1a, Il-6, Ip-10, Il-18, Lcn2, Tgf-b,andTimp-1 (D). Likewise, a direct activation of CB1R by the synthetic agonist ACEA resulted in similar upregulation of these genes (E). Both effects were completely

3526 Journal of the American Society of Nephrology J Am Soc Nephrol 28: 3518–3532, 2017 www.jasn.org BASIC RESEARCH

JD5037 completely prevented the oleate/palmitate– and ACEA- RPTC-CB1R should be considered as a therapeutic approach for induceddownregulationinpLKB1onRPTCsin vitro (Figure 7, F, treating obesity-induced nephropathy. G, and J). Conversely, a specific blocker of CaMKKb (STO-609, 1 Although the eCB/CB1R system was shown to regulate renal mg/ml) failed to inhibit the activation of AMPK by JD5037 in the hemodynamics, inflammation, and fibrogenesis during path- presence of oleate/palmitate (Supplemental Figure 9). ologic conditions,13 it was not determined whether these Next, we tested the role of Gai/o-dependent protein kinase abnormalities are mediated via CB1R on a specific cell type A (PKA), which was shown to regulate the activity of LKB1 within the kidney. By developing a transgenic mouse line, 37 elsewhere. Adding 100 ng/ml of pertussis toxin (a Gai/o which specifically lacks the CB1R gene in RPTCs (driven by blocker) to RPTCs treated with ACEA caused a significant in- SGLT2-Cre recombination), we could answer this question. crease in phosphorylation of LKB1, AMPK, and ACC compared The absence of the SGLT2 gene and protein in extrarenal or- with the effect of ACEA alone (Figure 8, A and C), suggesting Gi gans was reported by others.28,38–40 However, a recent report protein involvement. Moreover, the increased phosphorylation demonstrated that SGLT2 protein is expressed in glucagon- by JD5037 with oleate/palmitate was abolished by pretreating expressing a cells of the human pancreas.41 Because the ex- 42–45 RPTCs with 500 nM H-89 (a PKA blocker; Figure 8, B and D). pression of CB1R in pancreatic a cells is still controversial, Collectively, these findings imply that CB1R modulates obe- we cannot exclude the possible absence of its protein in these 2/2 sity-induced lipotoxicity in the RPTCs via regulating the cells in RPTC-CB1R mice. Nevertheless, we did not LKB1/AMPK/ACC signaling pathway. detect a Cre-Lox recombination in pancreas collected from the null mice, and the contribution of such a loss (if present)

CB1R Deletion/Blockade Increases Fatty Acid to the metabolic and/or renal phenotypes of these mice may b-Oxidation in RPTCs be minimal because most mouse and human a cells do not 46 Next, we found that the HFD-induced reduction in the renal express SGLT2 and CB1R. 47 expression of Ppara, Slc27a,andFabp1, which encode proteins Global CB1R knockout mice and animals with a selective 48 49 important for utilizing fatty acids in the mitochondria, is ame- genetic deletion or downregulation of CB1R in CaMKKa 2 2 / 50 liorated in RPTC-CB1R mice (Figure 8E). Similarly, JD5037 or hypothalamic neurons are lean and resistant to HFD- completely reversed the oleate/palmitate– or ACEA-induced re- induced obesity, suggesting that hypothalamic and/or pe- duction in the expression of these genes in RPTCs (Figure 8, F ripheral sympathetic CB1R is required for developing obesity. and G). Accordingly, the basal and maximal respiratory rates However, this does not negate the crucial role of CB1Rin were markedly elevated (Figure 8, H–J), and intracellular ATP peripheral organs during obesity. Indeed, mice lacking hepatic production was vastly increased (Figure 8K) in JD5037-treated CB1R develop obesity on HFD similarly to their littermates; RPTCs. Taken together, these findings suggest that deleting/ however, they are protected from obesity-induced hepatic stea- 51 blocking CB1R in RPTCs utilizes fatty acids by b-oxidation tosis. Likewise, both mouse strains used here become obese 2/2 rather than by lipogenesis (Figure 8L), and protects the kidney and metabolically impaired; however, RPTC-CB1R mice are from obesity-induced dysfunction and injury. largely protected from the deleterious effect of a prolonged fatty diet on the kidneys. These findings indicate that activating RPTC-CB1Rs is crucial for producing nephropathy during obe- DISCUSSION sity. Nevertheless, our findings do not exclude the possibility that other intrarenal or extrarenal CB1Rs, such as CB1R on glomer- Obesity-related renal dysfunction develops early in the course of ular podocytes, which contribute to glomerular dysfunction and obesity, justifying the search for novel regulators that could then damage in type 2 diabetic nephropathy,52 may also contribute to be targeted for therapy. Overactivation of the eCB/CB1Rsystem nephropathy. In addition, the slight improvements in some of in obesity promotes tubulointerstitial lesions and glomerulop- the metabolic parameters after a very prolonged HFD feeding in 2/2 athy; however, the exact underlying molecular mechanism is not RPTC-CB1R mice are possibly related to an undefined role of fully understood. This study demonstrates the role of proximal CB1R in RPTCs in modulating these parameters. tubule CB1R in renal lipotoxicity and consequently, in obesity- Independently of the metabolic syndrome, obesity induces induced renal inflammation, fibrosis, and injury by governing structural changes in the kidney, leading to reduced renal the LKB1/AMPK/ACC signaling pathway. Hence, manipulating function.1–4 Our findings in obese wild-type mice appear to

normalized by pretreating the cells with the peripherally restricted CB1R inverse agonist JD5037 (100 nM). Similarly, JD5037 reversed the oleate/palmitate–induced downregulation in the phosphorylated forms of LKB1, AMPK, and ACC (F–I), as well as increased the phosphorylation of these proteins despite the presence of ACEA (F, and J–L). Pretreating HK-2 cells with JD5037 also reversed the accumulation of fat droplets in cells with oleate/palmitate or ACEA (M). Scale bar, 20 mm. Data represent the mean6SEM from three independent experiments. *P,0.05 relative to vehicle-treated cells under normal conditions; #P,0.05 relative to vehicle-treated cells under oleate/palmitate conditions or ACEA treatment.

J Am Soc Nephrol 28: 3518–3532, 2017 Cannabinoid-1 Receptor Regulates Renal Lipotoxicity 3527 BASIC RESEARCH www.jasn.org

Figure 8. CB1R modulates lipogenesis or fatty acid b-oxidation in RPTCs via the Gi-PKA axis. PretreatingHK-2cellswith100ng/ml pertussis toxin (PTX; a Gai/o blocker) with the CB1R agonist ACEA signiﬁcantly increased the phosphorylation of LKB1, AMPK, and ACC (A and C). JD5037 loses the ability to increase the phosphorylation of these proteins while incubating HK-2 cells with 500 nM H-89 (a PKA inhibitor) (B and D). Reduced mRNA expression levels of Ppara, Slc27a,andFabp1 were documented in HFD-induced obese wild-type mice (E) as well as in HK-2 cells treated with oleate/palmitate (F) or ACEA (G). These reductions were normalized by either genetic ablation or pharmacologic blockade of CB1R in RPTCs. Increased basal and maximal respiratory rates (H–J) as well as

3528 Journal of the American Society of Nephrology J Am Soc Nephrol 28: 3518–3532, 2017 www.jasn.org BASIC RESEARCH be consistent with this notion. Remarkably, obese RPTC- their synthesis, and breakdown, as well as their oxidation and 2 2 / 26,63 CB1R mice are not only resistant to the HFD-induced efflux. Previous studies showed that this process and its re- structural and functional changes in the kidney; they also do lated kidney damage are associated with reduced activity of not develop an inflammatory response, and are resilient to de- AMPK in the kidney.25,26,29 AMPK, abundantly expressed in veloping renal fibrosis even after 43 weeks on HFD. Moreover, the kidney, restores energy balance by promoting energy- the fact that both mouse strains have similar levels of eCBs in the generating pathways. Together with ACC, regulation of fatty kidney but a significant difference in renal CB1R expression, acid b-oxidation and synthesis occurs. Our study clearly sup- indicates the pivotal role of CB1R in RPTCs for maintaining ports these observations, and, for the first time, highlights the renal homeostasis and function. pivotal role of blocking or deleting CB1R in enhancing fatty acid Surprisingly, only a minor progression in renal damage was b-oxidation rather than promoting lipid accumulation in observed in wild-type mice consuming approximately 30% RPTCs. saturated fat for 43 weeks. Similar findings were also reported As reported in hepatocytes,37 we show that activating 53 by others, demonstrating that HFD-induced obesity (24 AMPK by inhibiting CB1R is controlled upstream by a Gi- weeks) aggravates age-dependent kidney dysfunction, with PKA-LKB1 axis, without the involvement of CaMKKb.This no major increase in kidney injury after 40 weeks of feeding. axis, identified here in RPTCs, shows that CB1R, coupled with 64 It is reasonable to assume that higher saturated fat content may G proteins of the Gai/o, triggers a signaling pathway that promote renal dysfunction over time. However, HFD may en- inhibits the activation of PKA, which directly phosphorylates hance kidney dysfunction of different etiologies independently LKB1, the upstream kinase of AMPK, responsible for its activity. of hypertension,54 including intrauterine growth restriction,55 The importance of LKB1 in modulating intracellular metabo- type 1 diabetic nephropathy,56 aging,53 IgA GN,57 and unineph- lism in our model is supported by a recent work demonstrating rectomy.58 Moreover, under certain clinical conditions, such as that its deletion in renal distal tubular cells alters fatty acid me- diabetic nephropathy,14,20,59 IgA nephropathy,60 and acute in- tabolism via AMPK signaling.65 terstitial fibrosis,60 there is increased activation of the renal Collectively, we demonstrated that lipid accumulation and re- eCB/CB1R system. Although none of these studies determined duced fatty acid b-oxidation in RPTCs, associated with obesity- whether it enhances susceptibility to HFD-induced renal in- induced renal abnormalities, are governed by a CB1R-coupled jury, it is reasonable to suggest that superimposing HFD un- Gai/o-PKA axis, which mediates the downstream activation der these conditions most likely would exacerbate kidney of the LKB1/AMPK/ACC signaling pathway (Figure 8L). damage. Indeed, genetic analyses implicate lipotoxicity in Thus, manipulating CB1R specifically in the RPTCs may the susceptibility to type 2 diabetic nephropathy.61 Because provide a novel therapeutic intervention for treating obesity- evidence for the role of the eCB/CB1R system in diabetic induced nephropathy. nephropathy has been reported in humans and murine models,15,52,62 we also may speculate that HFD would probably augment type 2 diabetes–induced kidney damage by en- CONCISE METHODS hanced lipotoxicity. HFD-induced obesity results in the excessive accumulation Please see Supplemental Material for a detailed description of the of lipid vacuoles in RPTCs. Similarly, exposing RPTCs to oleate/ methods. palmitate resulted in increased intracellular fat accumulation associated with upregulating eCBs and CB1R. A similar effect Animals and Experimental Protocol was also found by directly activating CB1R in RPTCs, suggesting The experimental protocol was approved by the Institutional Animal 2/2 that the eCB system plays a key role in mediating lipid metab- Care and Use Committee of the Hebrew University. RPTC-CB1R olism in RPTCs. These findings are in accordance with others, mice were generated by crossing mice containing two loxP sites flank- 28 who reported increased RPTC apoptosis in the presence of ing the open reading frame of CB1R with the iL1-sglt2-Cre line. 2/2 palmitic acid, an effect that depended on CB1R-induced stimu- Six-week-old male RPTC-CB1R mice and their wild-type litter- lation of the endoplasmic reticulum stress response.18 Altered mates were fed either STD or HFD for 14 or 43 weeks. Two additional lipid metabolism and deposition have also been reported in mouse cohorts of C57Bl/6J mice were used: (1) mice on HFD treated obese humans with CKD.63 Intracellular lipid accumulation is with JD5037 (3 mg/kg, administered orally) or vehicle; and (2)mice normally regulated by a balance between the influx of fatty acids, treated with JZL195 (20 mg/g, administered intraperitoneally).

ATP production (K) were measured in HK-2 cells treated with JD5037 (100 nM) with oleate/palmitate. The suggested molecular mechanism by which CB1R in RPTCs regulates renal lipotoxicity (L). In vitro data represent the mean6SEM from three independent experiments. *P,0.05 relative to vehicle-treated cells under normal conditions; #P,0.05 relative to vehicle-treated cells under oleate/palmitate conditions or ACEA treatment; &P,0.05 relative to JD5037-treated cells under oleate/palmitate conditions. In vivo data represent the ∧ mean6SEM from 8 to 10 mice per group. P,0.05 relative to animals on STD of the same genotype; &P,0.05 relative to wild-type animals on the same diet.

J Am Soc Nephrol 28: 3518–3532, 2017 Cannabinoid-1 Receptor Regulates Renal Lipotoxicity 3529 BASIC RESEARCH www.jasn.org

Cell Culture Experiments ACKNOWLEDGMENTS Human immortalized RPTCs were cultured in REGM BulletKit me- fi dium at 37°C in a humidi ed atmosphere of 5% CO2/95% air. Lipotox- We are grateful to Dr. Dinorah Barasch for her technical assistance icity was achieved by 0.5 mM oleate/palmitate (2:1). Activating CB1Rwas with the liquid chromatography tandem-mass spectrometry (LC-MS/ done by ACEA (10 mM). Extracellular flux analysis was recorded by Sea- MS) analysis, as well as Michele Allen (NHLBI core facility), Anna horse XF24. ATP content was quantified by using a commercial kit. Permyakova, and Asaad Gammal for their technical assistance. This work was supported by the German-Israeli Foundation grant eCB Measurements (no. I-2345-201.2/2014), and an European Research Council (ERC)- eCBs were extracted, purified, and quantified in kidney and cell su- 2015-Starting Grant (no. 676841) to J.T. pernatant, as described previously.66

Multiparameter Metabolic Assessment DISCLOSURES Mouse phenotypic analysis was assessed by the Promethion High- None. Deﬁnition Behavioral Phenotyping System, as described previously.66

Biochemistry REFERENCES Serum and urinary biochemical analyses were determined by Cobas C-111. Urinary albumin, KIM-1, TIMP-1, LCN2, and serum insulin 1. Wang Y, Chen X, Song Y, Caballero B, Cheskin LJ: Association between levels were measured by ELISAs. Glucose homeostasis tests were done obesity and kidney disease: A systematic review and meta-analysis. – as described previously.66 Kidney Int 73: 19 33, 2008 2. Hsu CY, McCulloch CE, Iribarren C, Darbinian J, Go AS: Body mass index and risk for end-stage renal disease. Ann Intern Med 144: 21–28, 2006 Genomic DNA Extraction 3. Nguyen S, Hsu CY: Excess weight as a risk factor for kidney failure. Curr Mouse kidneys were harvested using a Percoll extraction and ultra- Opin Nephrol Hypertens 16: 71–76, 2007 centrifugation method. DNA was extracted and RT-PCR analysis for 4. Ejerblad E, Fored CM, Lindblad P, Fryzek J, McLaughlin JK, Nyrén O: 67 Obesity and risk for chronic renal failure. JAmSocNephrol17: 1695– the CB1R gene locus was performed as described previously. 1702, 2006 5. de Zeeuw D, Ramjit D, Zhang Z, Ribeiro AB, Kurokawa K, Lash JP, Chan Western Blotting J, Remuzzi G, Brenner BM, Shahinfar S: Renal risk and renoprotection 66 Kidney and cell homogenates were blotted as described previously. among ethnic groups with type 2 diabetic nephropathy: A post hoc analysis of RENAAL. Kidney Int 69: 1675–1682, 2006 Histopathology 6. Remuzzi G, Macia M, Ruggenenti P: Prevention and treatment of di- Paraffin-embedded kidney sections (2–3 mm) were stained with pe- abetic renal disease in type 2 diabetes: The BENEDICT study. JAmSoc – riodic acid–Schiff and Sirius Red. The glomerular and Bowman’s Nephrol 17[Suppl 2]: S90 S97, 2006 fi 7. Deji N, Kume S, Araki S, Soumura M, Sugimoto T, Isshiki K, Chin-Kanasaki space cross-sectional areas and mesangial expansion were quanti ed M, Sakaguchi M, Koya D, Haneda M, Kashiwagi A, Uzu T: Structural and using ZEN imaging software. Measuring tubular vacuole accumula- functional changes in the kidneys of high-fat diet-induced obese mice. tion was done on periodic acid–Schiff-stained sections, as described Am J Physiol Renal Physiol 296: F118–F126, 2009 previously.26 8. Stemmer K, Perez-Tilve D, Ananthakrishnan G, Bort A, Seeley RJ, Tschöp MH, Dietrich DR, Pfluger PT: High-fat-diet-induced obesity causes an inflammatory and tumor-promoting microenvironment in the fl Immunohistochemistry and Immuno uorescence rat kidney. Dis Model Mech 5: 627–635, 2012 fi RPTCs and paraf n-embedded kidney sections were used to deter- 9. Lee W, Eom DW, Jung Y, Yamabe N, Lee S, Jeon Y, Hwang YR, Lee JH, mine protein expression using commercially available antibodies. Kim YK, Kang KS, Kim SN: Dendrobium moniliforme attenuates high-fat diet-induced renal damage in mice through the regulation of lipid- induced oxidative stress. Am J Chin Med 40: 1217–1228, 2012 Real-Time PCR 10. Pai R, Kirschenbaum MA, Kamanna VS: Low-density lipoprotein stim- Total mRNA from mouse kidney and RPTCs was extracted as described ulates the expression of macrophage colony-stimulating factor in glo- 66 previously. Primer information is described in Supplemental Tables merular mesangial cells. Kidney Int 48: 1254–1262, 1995 1 and 2. 11. Coimbra TM, Janssen U, Gröne HJ, Ostendorf T, Kunter U, Schmidt H, Brabant G, Floege J: Early events leading to renal injury in obese Zucker (fatty) rats with type II diabetes. Kidney Int 57: 167–182, 2000 Materials 68 12. Larrinaga G, Varona A, Pérez I, Sanz B, Ugalde A, Cándenas ML, Pinto JD5037 was purchased from Haoyuan Chemexpress Co., Ltd. FM, Gil J, López JI: Expression of cannabinoid receptors in human kidney. Histol Histopathol 25: 1133–1138, 2010 Statistical Analyses 13. Tam J: The emerging role of the endocannabinoid system in the Values are expressed as the mean6SEM. Unpaired two-tailed t test pathogenesis and treatment of kidney diseases. J Basic Clin Physiol – was used to determine differences between groups. Results in multi- Pharmacol 27: 267 276, 2016 14. Barutta F, Corbelli A, Mastrocola R, Gambino R, Di Marzo V, Pinach S, ple groups and time-dependent variables were compared by ANOVA, Rastaldi MP, Perin PC, Gruden G: Cannabinoid receptor 1 blockade followed by a Bonferroni test (GraphPad Prism v6 for Windows). ameliorates albuminuria in experimental diabetic nephropathy. Diabetes 59: Significance was set at P,0.05. 1046–1054, 2010

3530 Journal of the American Society of Nephrology J Am Soc Nephrol 28: 3518–3532, 2017 www.jasn.org BASIC RESEARCH

15. Nam DH, Lee MH, Kim JE, Song HK, Kang YS, Lee JE, Kim HW, Cha JJ, tissue, muscle, and liver: The role of eNOS, p38 MAPK, and AMPK Hyun YY, Kim SH, Han SY, Han KH, Han JY, Cha DR: Blockade of pathways. Diabetes 59: 2826–2836, 2010 cannabinoid receptor 1 improves insulin resistance, lipid metabolism, 32. Cavuoto P, McAinch AJ, Hatzinikolas G, Cameron-Smith D, Wittert GA: and diabetic nephropathy in db/db mice. Endocrinology 153: 1387– Effects of cannabinoid receptors on skeletal muscle oxidative path- 1396, 2012 ways. Mol Cell Endocrinol 267: 63–69, 2007 16. Lim SK, Park SH: The high glucose-induced stimulation of B1R and B2R 33. Tam J, Cinar R, Liu J, Godlewski G, Wesley D, Jourdan T, Szanda G, expression via CB(1)R activation is involved in rat podocyte apoptosis. Mukhopadhyay B, Chedester L, Liow JS, Innis RB, Cheng K, Rice KC, Life Sci 91: 895–906, 2012 Deschamps JR, Chorvat RJ, McElroy JF, Kunos G: Peripheral cannabinoid-1 17. Lim JC, Lim SK, Park MJ, Kim GY, Han HJ, Park SH: Cannabinoid re- receptor inverse agonism reduces obesity by reversing leptin receptor 1 mediates high glucose-induced apoptosis via endoplasmic sistance. Cell Metab 16: 167–179, 2012 reticulum stress in primary cultured rat mesangial cells. Am J Physiol 34. Long JZ, Nomura DK, Vann RE, Walentiny DM, Booker L, Jin X, Burston Renal Physiol 301: F179–F188, 2011 JJ, Sim-Selley LJ, Lichtman AH, Wiley JL, Cravatt BF: Dual blockade 18. Lim JC, Lim SK, Han HJ, Park SH: Cannabinoid receptor 1 mediates of FAAH and MAGL identifies behavioral processes regulated by palmitic acid-induced apoptosis via endoplasmic reticulum stress in endocannabinoid crosstalk in vivo. Proc Natl Acad Sci U S A 106: human renal proximal tubular cells. J Cell Physiol 225: 654–663, 2010 20270–20275, 2009 19. Jenkin KA, McAinch AJ, Grinfeld E, Hryciw DH. Role for cannabinoid 35. Hawley SA, Pan DA, Mustard KJ, Ross L, Bain J, Edelman AM, Frenguelli receptors in human proximal tubular hypertrophy. Cell Physiol Bio- BG, Hardie DG: Calmodulin-dependent protein kinase kinase-beta is chem 26: 879–886, 2010 an alternative upstream kinase for AMP-activated protein kinase. Cell 20. Mukhopadhyay P, Pan H, Rajesh M, Bátkai S, Patel V, Harvey-White J, Metab 2: 9–19, 2005 Mukhopadhyay B, Haskó G, Gao B, Mackie K, Pacher P: CB1 canna- 36. Shaw RJ, Kosmatka M, Bardeesy N, Hurley RL, Witters LA, DePinho RA, binoid receptors promote oxidative/nitrosative stress, inflammation Cantley LC: The tumor suppressor LKB1 kinase directly activates AMP- and cell death in a murine nephropathy model. Br J Pharmacol 160: activated kinase and regulates apoptosis in response to energy stress. 657–668, 2010 Proc Natl Acad Sci U S A 101: 3329–3335, 2004 21. Janiak P, Poirier B, Bidouard JP, Cadrouvele C, Pierre F, Gouraud L, 37. Wu HM, Yang YM, Kim SG: Rimonabant, a cannabinoid receptor type 1 Barbosa I, Dedio J, Maffrand JP, Le Fur G, O’Connor S, Herbert JM: inverse agonist, inhibits hepatocyte lipogenesis by activating liver ki- Blockade of cannabinoid CB1 receptors improves renal function, nase B1 and AMP-activated protein kinase axis downstream of Ga i/o metabolic profile, and increased survival of obese Zucker rats. Kidney inhibition. Mol Pharmacol 80: 859–869, 2011 Int 72: 1345–1357, 2007 38. Sabolic I, Vrhovac I, Eror DB, Gerasimova M, Rose M, Breljak D, 22. Jenkin KA, O’Keefe L, Simcocks AC, Grinfeld E, Mathai ML, McAinch Ljubojevic M, Brzica H, Sebastiani A, Thal SC, Sauvant C, Kipp H, Vallon AJ, Hryciw DH: Chronic administration of AM251 improves albuminuria V, Koepsell H: Expression of Na+-D-glucose cotransporter SGLT2 in and renal tubular structure in obese rats. J Endocrinol 225: 113–124, rodents is kidney-specific and exhibits sex and species differences. Am 2015 J Physiol Cell Physiol 302: C1174–C1188, 2012 23. Barutta F, Grimaldi S, Gambino R, Vemuri K, Makriyannis A, Annaratone 39. You G, Lee WS, Barros EJ, Kanai Y, Huo TL, Khawaja S, Wells RG, Nigam L, di Marzo V, Bruno G, Gruden G. Dual therapy targeting the endo- SK, Hediger MA: Molecular characteristics of Na(+)-coupled glucose cannabinoid system prevents experimental diabetic nephropathy transporters in adult and embryonic rat kidney. J Biol Chem 270: [published online ahead of print April 6, 2017]. Nephrol Dial Transplant 29365–29371, 1995 doi:10.1093/ndt/gfx010 40. Chen J, Williams S, Ho S, Loraine H, Hagan D, Whaley JM, Feder JN: 24. Mount P, Davies M, Choy SW, Cook N, Power D: Obesity-related Quantitative PCR tissue expression profiling of the human SGLT2 gene chronic kidney disease-The role of lipid metabolism. Metabolites 5: and related family members. Diabetes Ther 1: 57–92, 2010 720–732, 2015 41. Bonner C, Kerr-Conte J, Gmyr V, Queniat G, Moerman E, Thévenet J, 25. Declèves AE, Mathew AV, Cunard R, Sharma K: AMPK mediates the Beaucamps C, Delalleau N, Popescu I, Malaisse WJ, Sener A, Deprez B, initiation of kidney disease induced by a high-fat diet. JAmSoc Abderrahmani A, Staels B, Pattou F: Inhibition of the glucose transporter Nephrol 22: 1846–1855, 2011 SGLT2withdapagliflozin in pancreatic alpha cells triggers glucagon 26. Declèves AE, Zolkipli Z, Satriano J, Wang L, Nakayama T, Rogac M, Le secretion. Nat Med 21: 512–517, 2015 TP, Nortier JL, Farquhar MG, Naviaux RK, Sharma K: Regulation of lipid 42. Starowicz KM, Cristino L, Matias I, Capasso R, Racioppi A, Izzo AA, Di Marzo accumulation by AMP-activated kinase [corrected] in high fat diet- V: Endocannabinoid dysregulation in the pancreas and adipose tissue of induced kidney injury. Kidney Int 85: 611–623, 2014 mice fed with a high-fat diet. Obesity (Silver Spring) 16: 553–565, 2008 27. Yamauchi T, Kamon J, Minokoshi Y, Ito Y, Waki H, Uchida S, Yamashita 43. Malenczyk K, Jazurek M, Keimpema E, Silvestri C, Janikiewicz J, Mackie S, Noda M, Kita S, Ueki K, Eto K, Akanuma Y, Froguel P, Foufelle F, Ferre K, Di Marzo V, Redowicz MJ, Harkany T, Dobrzyn A: CB1 cannabinoid P, Carling D, Kimura S, Nagai R, Kahn BB, Kadowaki T: Adiponectin receptors couple to focal adhesion kinase to control insulin release. stimulates glucose utilization and fatty-acid oxidation by activating J Biol Chem 288: 32685–32699, 2013 AMP-activated protein kinase. Nat Med 8: 1288–1295, 2002 44. Tharp WG, Lee YH, Maple RL, Pratley RE: The cannabinoid CB1 receptor 28. Rubera I, Poujeol C, Bertin G, Hasseine L, Counillon L, Poujeol P, Tauc is expressed in pancreatic delta-cells. Biochem Biophys Res Commun M: Specific Cre/Lox recombination in the mouse proximal tubule. 372: 595–600, 2008 JAmSocNephrol15: 2050–2056, 2004 45. Li C, Bowe JE, Jones PM, Persaud SJ: Expression and function of can- 29. Szeto HH, Liu S, Soong Y, Alam N, Prusky GT, Seshan SV: Protection of nabinoid receptors in mouse islets. Islets 2: 293–302, 2010 mitochondria prevents high-fat diet-induced glomerulopathy and 46. Avrahami D, Li C, Zhang J, Schug J, Avrahami R, Rao S, Stadler MB, proximal tubular injury. Kidney Int 90: 997–1011, 2016 Burger L, Schübeler D, Glaser B, Kaestner KH: Aging-dependent 30. Tedesco L, Valerio A, Cervino C, Cardile A, Pagano C, Vettor R, Pasquali demethylation of regulatory elements correlates with chromatin state R, Carruba MO, Marsicano G, Lutz B, Pagotto U, Nisoli E: Cannabinoid and improved b cell function. Cell Metab 22: 619–632, 2015 type 1 receptor blockade promotes mitochondrial biogenesis through 47. Ravinet Trillou C, Delgorge C, Menet C, Arnone M, Soubrié P: CB1 endothelial nitric oxide synthase expression in white adipocytes. cannabinoid receptor knockout in mice leads to leanness, resistance to Diabetes 57: 2028–2036, 2008 diet-induced obesity and enhanced leptin sensitivity. Int J Obes Relat 31. Tedesco L, Valerio A, Dossena M, Cardile A, Ragni M, Pagano C, Metab Disord 28: 640–648, 2004 Pagotto U, Carruba MO, Vettor R, Nisoli E: Cannabinoid receptor 48. Quarta C, Bellocchio L, Mancini G, Mazza R, Cervino C, Braulke LJ, stimulation impairs mitochondrial biogenesis in mouse white adipose Fekete C, Latorre R, Nanni C, Bucci M, Clemens LE, Heldmaier G,

J Am Soc Nephrol 28: 3518–3532, 2017 Cannabinoid-1 Receptor Regulates Renal Lipotoxicity 3531 BASIC RESEARCH www.jasn.org

Watanabe M, Leste-Lassere T, Maitre M, Tedesco L, Fanelli F, Reuss S, 59. Jenkin KA, McAinch AJ, Zhang Y, Kelly DJ, Hryciw DH: Elevated can- Klaus S, Srivastava RK, Monory K, Valerio A, Grandis A, De Giorgio R, nabinoid receptor 1 and G protein-coupled receptor 55 expression in Pasquali R, Nisoli E, Cota D, Lutz B, Marsicano G, Pagotto U: CB(1) proximal tubule cells and whole kidney exposed to diabetic conditions. signaling in forebrain and sympathetic neurons is a key determinant of Clin Exp Pharmacol Physiol 42: 256–262, 2015 endocannabinoid actions on energy balance. Cell Metab 11: 273–285, 60. Lecru L, Desterke C, Grassin-Delyle S, Chatziantoniou C, Vandermeersch 2010 S, Devocelle A, Vernochet A, Ivanovski N, Ledent C, Ferlicot S, Dalia M, 49. Jung KM, Clapper JR, Fu J, D’Agostino G, Guijarro A, Thongkham D, Saïd M, Beaudreuil S, Charpentier B, Vazquez A, Giron-Michel J, Avanesian A, Astarita G, DiPatrizio NV, Frontini A, Cinti S, Diano S, Azzarone B, Durrbach A, François H: Cannabinoid receptor 1 is a major Piomelli D: 2-arachidonoylglycerol signaling in forebrain regulates mediator of renal fibrosis. Kidney Int 88: 72– 84, 2015 systemic energy metabolism. Cell Metab 15: 299–310, 2012 61. Murea M, Freedman BI, Parks JS, Antinozzi PA, Elbein SC, Ma L: Lipotoxicity 50. Cardinal P, Bellocchio L, Guzmán-Quevedo O, André C, Clark S, in diabetic nephropathy: The potential role of fatty acid oxidation. Clin Elie M, Leste-Lasserre T, Gonzales D, Cannich A, Marsicano G, Cota JAmSocNephrol5: 2373–2379, 2010 D: Cannabinoid type 1 (CB1) receptors on Sim1-expressing neurons 62. Buraczynska M, Wacinski P, Zukowski P, Dragan M, Ksiazek A: Common regulate energy expenditure in male mice. Endocrinology 156: polymorphism in the cannabinoid type 1 receptor gene (CNR1) is as- 411–418, 2015 sociated with microvascular complications in type 2 diabetes. JDi- 51. Osei-Hyiaman D, Liu J, Zhou L, Godlewski G, Harvey-White J, Jeong WI, abetes Complications 28: 35–39, 2014 Bátkai S, Marsicano G, Lutz B, Buettner C, Kunos G: Hepatic CB1 receptor is 63. Herman-Edelstein M, Scherzer P, Tobar A, Levi M, Gafter U: Altered required for development of diet-induced steatosis, dyslipidemia, and in- renal lipid metabolism and renal lipid accumulation in human diabetic sulin and leptin resistance in mice. J Clin Invest 118: 3160–3169, 2008 nephropathy. J Lipid Res 55: 561–572, 2014 52. Jourdan T, Szanda G, Rosenberg AZ, Tam J, Earley BJ, Godlewski G, 64. Howlett AC: Pharmacology of cannabinoid receptors. Annu Rev Phar- Cinar R, Liu Z, Liu J, Ju C, Pacher P, Kunos G: Overactive cannabinoid 1 macol Toxicol 35: 607–634, 1995 receptor in podocytes drives type 2 diabetic nephropathy. Proc Natl 65. Han SH, Malaga-Dieguez L, Chinga F, Kang HM, Tao J, Reidy K, Susztak Acad Sci U S A 111: E5420–E5428, 2014 K: Deletion of Lkb1 in renal tubular epithelial cells leads to CKD by 53. van der Heijden RA, Bijzet J, Meijers WC, Yakala GK, Kleemann R, altering metabolism. J Am Soc Nephrol 27: 439–453, 2016 Nguyen TQ, de Boer RA, Schalkwijk CG, Hazenberg BP, Tietge UJ, 66. Knani I, Earley BJ, Udi S, Nemirovski A, Hadar R, Gammal A, Cinar R, Heeringa P: Obesity-induced chronic inflammation in high fat diet Hirsch HJ, Pollak Y, Gross I, Eldar-Geva T, Reyes-Capo DP, Han JC, challenged C57BL/6J mice is associated with acceleration of age-de- Haqq AM, Gross-Tsur V, Wevrick R, Tam J: Targeting the endocanna- pendent renal amyloidosis. Sci Rep 5: 16474, 2015 binoid/CB1 receptor system for treating obesity in Prader-Willi syn- 54. Foster MC, Hwang SJ, Porter SA, Massaro JM, Hoffmann U, Fox CS: drome. Mol Metab 5: 1187–1199, 2016 Fatty kidney, hypertension, and chronic kidney disease: The Framing- 67. Marsicano G, Wotjak CT, Azad SC, Bisogno T, Rammes G, Cascio MG, ham Heart Study. Hypertension 58: 784–790, 2011 Hermann H, Tang J, Hofmann C, Zieglgänsberger W, Di Marzo V, Lutz 55. Aliou Y, Liao MC, Zhao XP, Chang SY, Chenier I, Ingelfinger JR, Zhang SL: B: The endogenous cannabinoid system controls extinction of aversive Post-weaning high-fat diet accelerates kidney injury, but not hypertension memories. Nature 418: 530–534, 2002 programmed by maternal diabetes. Pediatr Res 79: 416–424, 2016 68. Chorvat RJ, Berbaum J, Seriacki K, McElroy JF: JD-5006 and JD-5037: 56. Yokoyama M, Tanigawa K, Murata T, Kobayashi Y, Tada E, Suzuki I, Peripherally restricted (PR) cannabinoid-1 receptor blockers related to Nakabou Y, Kuwahata M, Kido Y: Dietary polyunsaturated fatty acids SLV-319 (Ibipinabant) as metabolic disorder therapeutics devoid of CNS slow the progression of diabetic nephropathy in streptozotocin-in- liabilities. Bioorg Med Chem Lett 22: 6173–6180, 2012 duced diabetic rats. Nutr Res 30: 217–225, 2010 57. Bonnet F, Deprele C, Sassolas A, Moulin P, Alamartine E, Berthezène F, Berthoux F: Excessive body weight as a new independent risk factor for clinical and pathological progression in primary IgA nephritis. Am J See related editorial, “Obesity-Related CKD: When Kidneys Get the Munchies,” – Kidney Dis 37: 720 727, 2001 on pages 3429–3432. 58. Rayner HC, Ward L, Walls J: Cholesterol feeding following unilateral nephrectomy in the rat leads to glomerular hypertrophy. Nephron 57: This article contains supplemental material online at http://jasn.asnjournals. 453–459, 1991 org/lookup/suppl/doi:10.1681/ASN.2016101085/-/DCSupplemental.

3532 Journal of the American Society of Nephrology J Am Soc Nephrol 28: 3518–3532, 2017 Supplementary Information

Proximal Tubular Cannabinoid-1 Receptor Regulates Obesity-induced Chronic

Kidney Disease

Shiran Udi, Liad Hinden, Brian Earley, Adi Drori, Noa Reuveni, Rivka Hadar Resat

Cinar, Alina Nemirovski, and Joseph Tam

Inventory of Supplemental Information:

The following Supplemental Figures and Complete Methods provide additional information supporting the role of proximal tubule CB1R in the pathogenesis of obesity- induced nephropathy:

Supplementary Figure 1. A very prolonged (43 weeks) duration of HFD feeding induced

-/- a similar degree of obesity and its associated metabolic abnormalities in RPTC-CB1R mice and their wild-type controls.

Supplementary Figure 2. Similar blood pressure parameters in STD- and HFD- fed mice.

Supplementary Figure 3. Similar metabolic and activity profiles, as well as food and

-/- water consumption of RPTC-CB1R mice compared with their wild-type littermate controls after long-term (14 weeks) HFD feeding.

-/- Supplementary Figure 4. RPTC-CB1R mice are protected from long-term (14 weeks)

HFD-induced kidney injury.

-/- Supplementary Figure 5. Kidney weights of RPTC-CB1R mice and their wild-type littermates on STD and HFD for 14 and 43 weeks.

-/- Supplementary Figure 6. RPTC-CB1R mice are protected from long-term (14 weeks)

HFD-induced macrophage infiltration in the kidney.

Supplementary Figure 7. Genetic deletion of CB1R in RPTC ameliorates HFD (43 weeks)-induced renal lipid accumulation via a LKB1/AMPK signaling pathway.

Supplementary Figure 8. Pharmacologically increased in vivo production of eCBs inhibits renal AMPK and ACC phosphorylation via the CB1R.

Supplementary Figure 9. The CaMKK pathway does not play a significant role in the

CB1R/AMPK/ACC signaling pathway in RPTCs.

Supplementary Figure 1. A very prolonged (43 weeks) duration of HFD feeding induced a

-/- similar degree of obesity and its associated metabolic abnormalities in RPTC-CB1R mice and their wild-type controls. An almost identical metabolic phenotype was documented in RPTC-

-/- CB1R mice and their littermate controls following a very prolonged (43 weeks) STD and HFD feeding. Both mouse strains gained a similar body weight (A), displayed glucose intolerance (C), reduced insulin sensitivity (D), and hypercholesterolinemia (E). Improved circulating glucose (B),

-/- HDL/LDL ratio (F), ALT (G), and AST (H) levels were noted in obese RPTC-CB1R mice. Data represent the mean ± SEM from 5-10 mice per group. *P<0.05 relative to animals on STD of the # same genotype, P<0.05 relative to wild-type animals on the same diet.

Supplementary Figure 2. Similar blood pressure parameters in STD- and HFD- fed mice. Identical systolic and diastolic blood pressures were documented in wild-type mice fed either STD or HFD for 20 weeks (A). A genetic deletion of CB1R from RPTCs did not alter blood pressure in null mice in comparison with their littermate controls fed STD (B). Data represent the mean ± SEM from 5-10 mice per group.

Supplementary Figure 3. Similar metabolic and activity profiles, as well as food and water

-/- consumption of RPTC-CB1R mice compared with their wild-type littermate controls after long-term (14 weeks) HFD feeding. As measured by indirect calorimetry, both mouse strains on HFD exhibited a similar reduction in respiratory quotient (RQ; A), an equal resting energy expenditure (REE; B), increased fat oxidation (FO; C), and decreased carbohydrate oxidation (CHO; D), compared with mice fed a STD. Both mouse strains exhibited a similar ambulatory activity (E) and wheel running capabilities (F). Similar reductions in food (G) and water (H) intakes were recorded in both mouse strains on HFD. Data represent the mean ± SEM from 8-10

# mice per group. *P<0.05 relative to animals on STD of the same genotype, P<0.05 relative to wild- type animals on the same diet.

-/- Supplementary Figure 4. RPTC-CB1R mice are protected from long-term (14 weeks) HFD- induced kidney injury. The HFD-induced upregulation in renal Clusterin (Clu) mRNA expression

-/- (A), as well as protein levels (B, C) were attenuated/normalized in the RPTC-CB1R mice on the same diet. Similarly, the renal mRNA (D) and protein (E, G) levels of lipocalin 2 as well as its urinary excretion rate (F) were significantly upregulated in obese wild-type mice and attenuated in

-/- the RPTC-CB1R animals. Data represent the mean ± SEM from 8-10 mice per group. *P<0.05 # relative to animals on STD of the same genotype, P<0.05 relative to wild-type animals on the same diet.

-/- Supplementary Figure 5. Kidney weights of RPTC-CB1R mice and their wild-type littermates on STD and HFD for 14 and 43 weeks. Improved kidney-to-body weight ratios (A,

-/- C), and high kidney weights (B, D) were documented in RPTC-CB1R mice fed either STD or HFD for 14 and 43 weeks. Data represent the mean ± SEM from 5-10 mice per group. *P<0.05 relative to animals on STD of the same genotype, #P<0.05 relative to wild-type animals on the same diet.

-/- Supplementary Figure 6. RPTC-CB1R mice are protected from long-term (14 weeks) HFD- induced macrophage infiltration in the kidney. The HFD-induced upregulation in the renal mRNA (A) and protein (B, D) expression levels of Ki67 as well as the protein expression levels of

-/- F4/80 (C, E) were significantly attenuated/normalized in the RPTC-CB1R mice on the same diet. Data represent the mean ± SEM from 8-10 mice per group. *P<0.05 relative to animals on STD of # the same genotype, P<0.05 relative to wild-type animals on the same diet.

Supplementary Figure 7. Genetic deletion of CB1R in RPTC ameliorates HFD (43 weeks)- induced renal lipid accumulation via a LKB1/AMPK signaling pathway. A significantly higher

-/- lipid droplet accumulation was found in obese wild-type mice in comparison with RPTC-CB1R animals (A, C). A very prolonged HFD-feeding resulted in a significant reduction in the expression levels of the activated/phosphorylated form of AMPK (B, D) and LKB1 (E, F). These changes

-/- were greatly attenuated in RPTC-CB1R mice on the same diet. Scale bar, 20 µm. Data represent the mean ± SEM from 5-10 mice per group. *P<0.05 relative to animals on STD of the same genotype, #P<0.05 relative to wild-type animals on the same diet.

Supplementary Figure 8. Pharmacologically increased in vivo production of eCBs inhibits renal AMPK and ACC phosphorylation via the CB1R. A single dose of 20 µg/g JZL195 (a dual inhibitor of the hydrolysis of AEA and 2-AG) resulted in significantly elevated levels of AEA and its precursor and degraded ligand AA, but not 2-AG (A). This increase was associated with reduced expression in the phosphorylated forms of both AMPK (B, D) and ACC (C, E) in the kidney. Data represent the mean ± SEM from 5 mice per group. *P<0.05 relative to animals treated with vehicle.

Supplementary Figure 9. The CaMKK pathway does not play a significant role in the

CB1R/AMPK/ACC signaling pathway in RPTCs. A specific blocker of CaMKKβ (STO-609, 1 µg/mL) failed to completely inhibit the activation of AMPK by JD5037 in the presence of oleate/palmitate. Data represent the mean ± SEM from 2 independent experiments. *P<0.05 relative to oleate:plamitate-treated cells.

Complete Methods

Animals. The experimental protocol used was approved by the Institutional Animal Care and Use Committee of the Hebrew University, which is an AAALAC International accredited institute. To elucidate the role of proximal tubular CB1R in the pathogenesis of obesity-induced metabolic and renal complications, we generated a novel mouse strain that

fl/fl;sglt2Cre lacks CB1Rs specifically in the RPTCs (CB1R ) by crossing mice containing two

fl/fl loxP sites flanking the open reading frame of the CB1R gene (CB1R ; a kind gift of Prof.

Beat Lutz, University Medical Center of the Johannes Gutenberg University Mainz,

Germany) with the iL1-sglt2-Cre line1, which specifically expresses Cre recombinase in the brush border membrane of the S1 segment of the proximal tubule. Mice were genotyped by PCR. For the iL1-sglt2-Cre line, the primers 5’-CAG GGT GTT ATA AGC AAT CCC

- 3’, and 5’-CCT GGA AAA TGC TTC TGT CCG - 3’ produced a band of 350 bp, when the transgene was present. For the genotyping of the CB1R gene locus, G50, G51, and G53 primers were used as described previously2 (Figure 1A). Cre-mediated recombination of the CB1R gene explicitly in RPTCs was confirmed by cre-lox recombination detection in

fl/fl;sglt2Cre genomic DNA and by the mRNA and protein expression analyses in CB1R

-/- (RPTC-CB1R ) mice (Figure 1B-D).

Experimental protocol. To generate diet-induced obesity (DIO; body weight > 42g), six-

-/- week-old male RPTC-CB1R mice and their wild-type littermate controls were fed either a high-fat diet (HFD; 60% of calories from fat, 20% from protein, and 20% from carbohydrates; Research Diet, D12492), or a standard diet (STD; 14% fat, 24% protein,

62% carbohydrates; NIH-31 rodent diet) for 14 weeks or 43 weeks. Body weight was monitored every week. Total body fat and lean masses were determined by EchoMRI-

100H™ (Echo Medical Systems LLC, Houston, TX, USA). 24 h urine was collected one week before euthanasia using mouse metabolic cages (CCS2000 Chiller System, Hatteras

Instruments, NC, USA). At weeks 20 and 49, mice were euthanized by a cervical dislocation under anesthesia, the kidneys, brain, liver, fat pads, and muscles were removed and weighed, and samples were either snap-frozen or fixed in buffered 4% formalin. Trunk blood was collected for determining the biochemical parameters.

To assess the effect of peripherally restricted CB1R blockade on the kidney, a separate cohort of male C57Bl/6J mice were fed a HFD or a STD for one week while being treated with JD50373 (3 mg/kg, po) or vehicle (1% Tween80, 4% DMSO, 95% Saline).

One hour after the last dose, the animals were euthanized by a cervical dislocation under anesthesia and their kidneys were collected and processed for immunohistochemistry or

Western blotting analyses.

To assess the effect of CB1R activation by its endogenous ligands, 20 µg/g JZL195

(Cayman Chemical) was intraperitoneally injected into male C57Bl/6J mice, and four hours later the animals were euthanized by a cervical dislocation under anesthesia, and their kidneys were collected and processed for immunohistochemistry analysis and eCB measurements by LC/MS-MS.

Cell culture. HK-2 (Human immortalized RPTCs) were cultured in REGM BulletKit medium (Lonza, USA) supplemented with 10 μg/mL recombinant human epidermal growth factor, 5 µg/mL insulin, 0.5 mg/mL hydrocortisone, 0.5% FBS, 0.5 µg/mL epinephrine, 6.5 μg/mL T3, 10 mg/mL transferrin, 10 mg/mL gentamicin, and 50 μg/mL amphotericin-B and penicillin/streptomycin. Cells were cultured at 37°C in a humidified atmosphere of 5% CO2/95% air. Cell experiments were conducted at >80% confluence in

6-well plates. To establish lipotoxic conditions, cells were incubated with REGM BulletKit medium (Lonza, USA) containing 1% fatty acid BSA (Amresco, 0332-TAM), 0.5 mM sodium oleate (Sigma Aldrich, O7501), and sodium palmitate (Sigma Aldrich, P9767) in a ratio of 2:1, dissolved in 11% free-fatty acid BSA (Sigma Aldrich, A7030), and ultra-pure

DDW (Biological Industries, 01-866-1A). In addition, CB1R in RPTCs was stimulated by using 10 µM arachidonyl-2-chloroethylamide (ACEA), a selective CB1R agonist, dissolved in EtOH. To inhibit CaMKK, Gαi, and PKA, cells were treated with 1 µg/mL

STO-609 (Abcam, ab141591), 100 ng/mL PTX (Sigma Aldrich, P7208), and 500 nM H-

89 (Abcam, ab143787), respectively. JD5037 (100 nM) was dissolved in DMSO. Cells were harvested using Bio-Tri RNA lysis buffer (Bio-Lab, Israel). Lipids were stained using the Oil-Red-O kit (Abcam, ab150678). Briefly, medium was removed, cells were fixed with 10% PFA, followed by the addition of 60% isopropanol. Staining quantification was analyzed using the Spectra Max Plus plate reader (Molecular Devices, CA, USA) at a wave length of 500 nm.

Extracellular flux analysis. The oxygen consumption rate (OCR) was measured using a

Seahorse XF24 Extracellular Flux Analyzer (Seahorse Bioscience, Agilent Technologies).

Briefly, HK-2 cells were seeded on an XF24 cell culture plate (4×104 cells/well) and incubated overnight in a 5% CO2 incubator at 37°C. Oleate:palmitate (0.5 mM; 2:1, respectively) and JD5037 (100 nM) were added to cells the next morning, in a fresh serum- free media for the indicated time. Basal and maximal OCR were measured throughout the incubation time.

ATP measurement. ATP content in RPTCs was quantified using an ATP assay kit

(Abcam, ab83355) following the manufacturer’s instructions. Briefly, cells (3x106 cells/well) were harvested by trypsin and re-suspended with 100 µL ATP assay buffer.

Then, cells were deproteinized using TCA (Abcam, ab204708). To each 50 µL sample, a

50 µL reaction mix was added. Outputs were measured by a microplate reader at 570 nm.

Eight replicates from each group were tested.

Endocannabinoid measurements by LC-MS/MS. eCBs were extracted, purified, and quantified in kidney and cell supernatant, as described previously4, with modifications.

Briefly, total proteins were precipitated with ice-cold acetone/Tris buffer (50 mM, pH 8.0).

Then, the samples were homogenized in ice-cold methanol/Tris buffer (50 mM, pH 8.0,

2 2 1:1) containing [ H4]arachidonoyl ethanolamide ([ H4]AEA) as an internal standard.

Homogenates were then extracted with CHCl3:MeOH (2:1, vol/vol) and washed three times with ice-cold CHCl3, dried under nitrogen flow, and reconstituted with MeOH.

LC-MS/MS was analyzed on an AB Sciex (Framingham, MA, USA) Triple Quad™

5500 mass spectrometer coupled with a Shimadzu (Kyoto, Japan) UHPLC System. Liquid

15 chromatographic separation was achieved using 5 μL injections of samples onto a Kinetex

2.6µm C18 (100*2.1 mm) from Phenomenex. The autosampler was set at 4°C and the column was maintained at 40°C during the entire analysis.

Gradient elution mobile phases consisted of A (0.1% formic acid in water) and B

(0.1% formic acid in acetonitrile). Gradient elution (250 μL/min) was held at 10% B for the first 0.5 min, followed by a linear increase towards 70% B at 1 min, and a linear increase towards 75% B at 6 min, maintained until 16 min and then increased linearly to 98% B at

16.3 min; this was maintained until 20 min.

eCBs were detected in a positive ion mode using electron spray ionization (ESI) and the multiple reaction monitoring (MRM) mode of acquisition. Air was produced (SF

2 FF compressor, Atlas Copco, Belgium) and purified using an NM20Z nitrogen generator

(Peak Scientific, Inchinnan, Scotland). Purified air was used as source and exhaust gases and purified nitrogen as curtain and collision gases. A receiver was placed between the compressor and the nitrogen generator for a large and stable supply of air.

The molecular ion of the compounds [M+H]+ was selected in the first mass analyzer and fragmented in the collision cell followed by detecting the products of fragmentation in the second mass analyzer. The TurboIonspray® probe temperature was set at 650°C with the ion spray voltage at 4500V. The curtain gas was set at 40.0 psi. The nebulizer gas (Gas

1) was set to 50 psi, the turbo heater gas (Gas 2) was set to 70 psi, and the collision gas

(CAD) was set to 8 psi.

Levels of each compound were analyzed by multiple reactions monitoring. The molecular ion and fragment for each compound were measured as follows: m/z

348.3→62.1 (quantifier) and 91.1 (qualifier) for AEA, m/z 379.3→287.3 (quantifier) and

91.1(qualifier) for 2-AG, m/z 305.2 →91.1 (quantifier) and 77.1 (qualifier) for AA and m/z

2 352.3→66.1 (quantifier) and 91.1 (qualifier) for [ H4]AEA. Levels of AEA, 2-AG, and AA in the cell medium and kidney samples were measured against standard curves.

Multi-parameter metabolic assessment. Metabolic and activity profiles of the mice were assessed by using the Promethion High-Definition Behavioral Phenotyping System (Sable

Instruments, Inc., Las Vegas, NV, USA). Data acquisition and instrument control were performed using MetaScreen software version 2.2.18.0, and the obtained raw data were processed using ExpeData version 1.8.4 using an analysis script detailing all aspects of data transformation. Mice with free access to food and water were subjected to a standard

12 h light/12 h dark cycle, which consisted of a 48 h acclimation period followed by 24 h of sampling. Respiratory gases were measured by using the GA-3 gas analyzer (Sable

Systems, Inc., Las Vegas, NV, USA) using a pull-mode, negative-pressure system. Air flow was measured and controlled by FR-8 (Sable Systems, Inc., Las Vegas, NV, USA), with a set flow rate of 2000 mL/min. Water vapor was continuously measured and its dilution effect on O2 and CO2 was mathematically compensated. Effective mass was calculated by [body mass]0.75. Respiratory quotient (RQ) was calculated as the ratio of

VCO2/VO2. Resting energy expenditure (REE) was expressed as kcal/h. Fat oxidation (FO) and carbohydrate oxidation (CHO) were calculated as FO = 1.69 x VO2 – 1.69

eff.Mass x VCO2 and CHO = 4.57 x VCO2– 3.23 x VO2 and expressed as g/d/kg . Ambulatory activity and position were monitored simultaneously with the collection of the calorimetry data using XYZ beam arrays with a beam spacing of 0.25 cm.

Blood and urine biochemistry. Serum levels of cholesterol, HDL, LDL, ALT, AST, creatinine, and urea as well as urine levels of creatinine were determined by using the

Cobas C-111 chemistry analyzer (Roche, Switzerland). Blood urea nitrogen (BUN) was calculated by serum urea levels (BUN mg/dL= Urea mM × 2.801). Creatinine clearance was calculated using urine and serum creatinine levels (CCr mL/h= Urine creatinine mg/dL ×

Urine volume / Serum creatinine mg/dL × 24 hrs). Urine levels of albumin, kidney injury marker (KIM-1), tissue inhibitor of metalloproteinases-1 (TIMP-1), lipocalin 2 (LCN2), as well as serum insulin were measured by ELISA kits (Albumin, Bethyl Laboratories, TX,

USA; KIM-1 & TIMP-1, R&D Systems, MN, USA; LCN2, BOSTER Biological

Technology, CA, USA; Insulin, Crystal Chem, Inc., Downers Grove, IL, USA). Fasting blood glucose was measured using the Elite glucometer (Bayer, Pittsburgh, PA).

Glucose tolerance (ipGTT) and insulin sensitivity tests (ipIST). Mice that fasted overnight were injected with glucose (1.5 g/kg, ip), followed by a tail blood collection at 0, 15, 30,

45, 60, 90, and 120 minutes. Blood glucose levels were determined using the Elite glucometer (Bayer, Pittsburgh, PA). The following day, mice underwent fasting for 6 h before receiving insulin (0.75 U/kg, ip; Eli Lilly), and blood glucose levels were determined at the same intervals as above.

-/- Primary RPTC extraction. Mouse kidneys from wild-type and RPTC-CB1R were harvested. While the inner medullary portion was excised, cortices were digested by 2

18 mg/mL Collagenase IV (Worthington, NJ, USA) and 1 mg/mL DNase I (Sigma Aldrich,

DN25) in 1x PBS containing 10% FCS for 1 hour at 37oC. Then, 0.5 M EDTA was added and the samples were incubated for an additional 10 min at 37oC. Lysates were filtered using a 40 µM cell strainer. Next, 1.5 mL lysate was added to a designated ultra-centrifuge tube containing 10 mL 45% Percoll and 1.5 mL 90% Percoll (Sigma Aldrich, GE17-0891-

01) solutions. Samples were ultra-centrifuged using the SW-41 rotor at 20,000 x g at 4 oC for 30 min with break-off. Centrifugation resulted in the separation of four banded layers.

The third layer, containing kidney single cell suspension, was transferred, washed, and cultivated in REGM BulletKit medium (Lonza, USA).

Genomic DNA extraction. DNA was extracted following the QIAGEN DNA extraction kit protocol. Briefly, RPTCs and mouse brain, liver, muscle, pancreas, inguinal, and retroperitoneal fat pads were incubated with proteinase K for 1 hour at 56oC. Lysates were pipetted to DNeasy Mini spin columns containing DNA binding membrane. After two membrane washing steps, an elution buffer was added to detach the genomic DNA from the membrane into the tube. RT-PCR analysis for CB1R receptor gene locus was performed using the G50, G51, and G53 primers2.

Western blotting. Kidney homogenates were prepared in a RIPA buffer (25mM Tris-HCl pH 7.6, 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS) using

BulletBlender® and zirconium oxide beads (Next Advanced, Inc., NY, USA). Cells were

19 harvested using Bio-Tri RNA lysis buffer (Bio-Lab, Israel) and proteins were extracted by

Isopropanol, 0.3 M guanidine hydrochloride, and 0.1% SDS. Protein concentrations were measured with the Pierce™ BCA Protein Assay Kit (Thermo Scientific, IL, USA). Samples were resolved by SDS-PAGE (4-15% acrylamide, 150V) and transferred to PVDF membranes using the Trans-Blot® Turbo™ Transfer System (Bio-Rad, CA). Membranes were then incubated for 1 h in 5% milk (in 1x TBS-T) to block unspecific binding. Blots were incubated overnight with antibodies against LCN2 (Abcam, ab70287), AMPK (Cell

Signaling Technology, #2532,), phosphorylated AMPK (Cell Signaling Technology,

#2535), acetyl-CoA carboxylase (ACC; Cell Signaling Technology, #3662), phosphorylated ACC (Cell Signaling Technology, #3661), phosphorylated LKB1 (Abcam, ab138386 and ab63473) and LKB1 (Abcam, ab15095) at 4°C. Anti-rat or rabbit horseradish peroxidase (HRP)-conjugated secondary antibodies were used for 1 h at room temperature and followed by chemiluminescence detection using the Clarity™ Western

ECL Blotting Substrate (Bio-Rad, CA). Densitometry was quantified using Image J.

Quantification was normalized to anti-β actin antibody (ab49900, Abcam).

Histopathological Analyses. 3 µm paraffin-embedded kidney sections from 5 animals per group were stained with periodic acid–schiff (PAS), followed by hematoxylin staining. In addition, kidney sections were stained for collagen types I and III deposition by using Sirius

Red staining (Abcam, ab150681). Kidney images were captured with a Zeiss AxioCam

ICc5 color camera mounted on a Zeiss Axio Scope.A1 light microscope and taken from 10 random 40x fields of each animal. The mesangial expansion, glomerular and Bowman’s

20 space cross-sectional areas were quantified using ZEN imaging software (Zeiss, Germany).

The Sirius Red positive area was quantified using Image J with a minimum of 10 random kidney sections per mouse.

The tubular vacuole accumulation was characterized on PAS stained sections, as described previously5. Briefly, 10 randomly selected 94.125 mm2 areas in the outer cortex of each kidney section were evaluated by counting cells containing vacuoles.

Immunohistochemistry. Kidney tissue was fixed in buffered 4% formalin for 48 h and then embedded in paraffin. Sections were deparaffinized and hydrated. Heat-mediated antigen was retrieved with 10 mM citrate buffer pH 6.0 (Thermo Scientific, IL, USA). Endogenous peroxide was inhibited by incubating with a freshly prepared 3% H2O2 solution in MeOH.

Unspecific antigens were blocked by incubating sections for 1 h with 2.5% horse serum

(VE-S-2000, Vector Laboratories). For assessment of tubule-interstitial injury, inflammation, and fibrosis, 2 µm kidney sections were stained with rabbit anti-mouse IL-

18 (Abcam, ab71495), Clusterin (Proteintech, 12289-1-AP), KIM-1 (Abcam ab78949),

TIMP-1 (Proteintech, 10753-1-AP), TNFα (Abcam, ab6671), iNOS (Abcam, ab3523),

MCP-1 (Abcam, ab25124), F4/80 (Abcam, ab6640), Ki-67 (Sigma Aldrich,

SAB4501880), collagen-1 (Abcam, ab34710), collagen-3 (Abcam, ab7778), αSMA

(Abcam ab5694), pAMPK (Abacm, ab133448) and pACC (Cell Signaling Technology,

#3661) antibodies, followed by a goat anti-rabbit or anti-rat HRP conjugate

(ImmPRESS™, Vector laboratories). Color was developed after an incubation with 3,3'-

Diaminobenzidine (DAB) substrate (ImmPACT DAB Peroxidase (HRP) Substrate, SK-

4105, Vector Laboratories), followed by hematoxylin counterstaining and mounting

(Vecmount H-5000, Vector laboratories). Stained sections were photographed as described above. The positive (stained) area for each marker was calculated using color thresholding and measuring area fractions with the Image J software (NIH Public Domain), with a minimum of 5-6 random kidney sections per mouse. Images are presented in the figures, showing the animal with the median value for each group.

Immunofluorescence staining. HK-2 cells were seeded on poly-L-lysine precoated glass coverslips in 24-well plates. At the end of the experiment, cells were fixed for 10 min in

4% paraformaldehyde, washed 3 times with PBS 1x, and permeabilized using 0.25% Triton

X-100 for 7 min. Cells were then washed 3 times and blocked in PBS 1x containing 2%

BSA for 2 h. All preceding steps were conducted at room temperature. Following blocking, cells were stained with anti-CB1R rabbit IgG (Immunogenes, Hungary), overnight at 4°C.

Alexa flour 647-conjugated goat anti-rabbit IgG (Invitrogen molecular probes, A21244) was used as a secondary antibody for 1 h. Cover slips were mounted using Vectashield mounting medium (H-1200, Vector laboratories). Images were taken with an OLYMPUS

IX-73 fluorescent microscope and analyzed using CellSens Dimension software.

Real-time PCR. Total mRNA from mouse kidney and HK-2 cells was extracted using Bio-

Tri RNA lysis buffer (Bio-Lab, Israel) followed by DNase I treatment (Thermo Scientific,

IL, USA), and reverse transcribed using the Iscript cDNA kit (Bio-Rad, CA). Real-time

PCR was performed using the iTaq Universal SYBR Green Supermix (Bio-Rad, CA) and the CFX connect ST system (Bio-Rad, CA). The following Mus musculus genes were detected: Cnr1, Col1a1, Col3a1, α-Sma, Fn1, Clu, Kim-1, Timp-1, Lcn2, Il-18, Tnfα, Nos2,

Mcp-1, Lrp2, Pparα, Slc27a and Fabp1. All genes were normalized to Mus musculus β-

Actin (primer information is described in Supplementary Tables 1 and 2).

The expression of the following human genes was analyzed: TNFα, IL-1α, IL-6,

IP-10, IL-18, LCN2, TGF-β, TIMP-1, CNR1, PPARα, SLC27A and FABP1. All genes were normalized to human RPLP0 (primer information is described in Supplementary Tables

1 and 2).

Non-invasive blood pressure measurement. Mouse blood pressure was measured by the

CODA 4.1 noninvasive blood pressure monitoring system (Kent scientific) on conscious restrained mice connected to tail-cuff after placing the animals on a warming platform set to 35-38°C. Each animal session carried out 5 acclimation cycles and 15 measurement cycles per set. Time between sets was 5 seconds, deflation time was 15 seconds and minimum volume was 15 uL.

Materials. JD5037 was synthesized as described previously6, and purchased from Haoyuan

Chemexpress Co., Ltd.

Statistics. Values are expressed as the mean ± SEM. Unpaired two-tailed Student’s t-test was used to determine differences between groups. Results in multiple groups and time- dependent variables were compared by ANOVA, followed by a Bonferroni test (GraphPad

Prism v6 for Windows). Significance was set at P < 0.05.

Supplementary Table 1. Primer information.

Gene Forward Reverse mus musculus AAGTCGATCTTAGACGGCCTT TCCTAATTTGGATGCCATGTCTC Cnr1 mus ATGTGGACCCCTCCTGATAGT GCCCAGTGATTTCAGCAAAGG musculusFn1 mus GCATTAGCTTCAGATTTA TTAAAAACCTGGATCGGAACCAA musculusMcp1 mus musculus CCACCAGCTCACCATTGTACT TCTGCTTGGACTTCCTTTTCA Kim1 mus musculus AAAATGGAAACGGGGTGACTT GGCTGCATACATTGGGTTTTCA Lrp2 mus musculus GGC TGT ATT CCC CTC CAT CG CCA GTT GGT AAC AAT GCC ATG β-Actin homo sapience GAATGACGCCCTCAATCAAAGT TCATCTTGGGCAGTCACATACA IL1α homo sapience TAGCCGCCCCACAGACAG GGCTGGCATTTGTGGTTGGG IL6 homo sapience TGAAATTATTCCTGCAAGCCAA CAGACATCTCTTCTCACCCTTCTTT IP10 (CXCL10) homo sapience GCCTAGAGGTATGGCTGTAA GCGTCACTACACTCAGCTAA IL18 homo sapience GTAGGCCTGGCAGGGAATG GGAACAAAAGTCCTGATCCAGTAGT LCN2 homo sapience CCCAGCATCTGCAAAGCTC GTCAATGTACAGCTGCCGCA TGFβ homo sapience AGACCTACACTGTTGGCTGTGAG GACTGGAAGCCCTTTTCAGAG TIMP1 homo sapience AAGCCCGCATGGACATTAGGTTAG AGCAGAGGGCCCCAGCAGAT CNR1 homo sapience CCAACTACTTCCTTAAGATCATCCAA ACATGCGGATCTGCTGCTGCA RPLP0

Supplementary Table 2. Primer information.

Gene QuantiTect Primer Assays (Qiagen, Germany) mus musculus Col1a1 QT00162204 mus musculus Col3a1 QT01055516 mus musculus αSma QT00140119 mus musculus Timp1 QT00996282 mus musculus Lcn2 QT00113407 mus musculus Tnfα QT00104006 mus musculus Il18 QT00171129 mus musculus Clu QT00146174 mus musculus Nos2 QT00100275 mus musculus Ppara QT001379 mus musculus Slc27a QT00157766 mus musculus Fabp1 QT00112427 homo sapience PPARα QT00017451 homo sapience SLC27A QT00047817 homo sapience FABP1 QT01000405 homo sapience TNFα QT01079561

References

1. Rubera I, Poujeol C, Bertin G, Hasseine L, Counillon L, Poujeol P, Tauc M. Specific

Cre/Lox recombination in the mouse proximal tubule. Journal of the American

Society of Nephrology : JASN 15:2050-2056, 2004

2. Marsicano G, Wotjak CT, Azad SC, Bisogno T, Rammes G, Cascio MG, Hermann

H, Tang J, Hofmann C, Zieglgansberger W, Di Marzo V, Lutz B. The endogenous

cannabinoid system controls extinction of aversive memories. Nature 418:530-534,

2002

3. Tam J, Cinar R, Liu J, Godlewski G, Wesley D, Jourdan T, Szanda G,

Mukhopadhyay B, Chedester L, Liow JS, Innis RB, Cheng K, Rice KC, Deschamps

JR, Chorvat RJ, McElroy JF, Kunos G. Peripheral cannabinoid-1 receptor inverse

agonism reduces obesity by reversing leptin resistance. Cell metabolism 16:167-179,

2012

4. Knani I, Earley BJ, Udi S, Nemirovski A, Hadar R, Gammal A, Cinar R, Hirsch HJ,

Pollak Y, Gross I, Eldar-Geva T, Reyes-Capo DP, Han JC, Haqq AM, Gross-Tsur V,

Wevrick R, Tam J. Targeting the endocannabinoid/CB1 receptor system for treating

obesity in Prader-Willi syndrome. Mol Metab 5:1187-1199, 2016

5. Decleves AE, Zolkipli Z, Satriano J, Wang L, Nakayama T, Rogac M, Le TP,

Nortier JL, Farquhar MG, Naviaux RK, Sharma K. Regulation of lipid accumulation

by AMP-activated kinase [corrected] in high fat diet-induced kidney injury. Kidney

international 85:611-623, 2014

6. Chorvat RJ, Berbaum J, Seriacki K, McElroy JF. JD-5006 and JD-5037: peripherally

restricted (PR) cannabinoid-1 receptor blockers related to SLV-319 (Ibipinabant) as

27 metabolic disorder therapeutics devoid of CNS liabilities. Bioorganic & medicinal chemistry letters 22:6173-6180, 2012