The Adiponectin Receptor Agonist Adiporon Ameliorates Diabetic Nephropathy in a Model of Type 2 Diabetes

BASIC RESEARCH www.jasn.org

The Adiponectin Receptor Agonist AdipoRon Ameliorates Diabetic Nephropathy in a Model of Type 2 Diabetes

Yaeni Kim,1,2 Ji Hee Lim,1,3 Min Young Kim,1,3 Eun Nim Kim,1,3 Hye Eun Yoon,1,2 Seok Joon Shin,1,2 Bum Soon Choi,1,3 Yong-Soo Kim,1,3 Yoon Sik Chang,1,4 and Cheol Whee Park1,3

1Division of Nephrology, Department of Internal Medicine, College of Medicine, The Catholic University of Korea, Seoul, Korea; 2Division of Nephrology, Department of Internal Medicine, Incheon St. Mary’s Hospital, Incheon, Korea; 3Division of Nephrology, Department of Internal Medicine, Institute for Aging and Metabolic Diseases, Seoul St. Mary’sHospital, Seoul, Korea; and 4Division of Nephrology, Department of Internal Medicine, Yeouido St. Mary’s Hospital, Seoul, Korea

ABSTRACT Adiponectin exerts renoprotective effects against diabetic nephropathy (DN) by activating the AMP-activated protein kinase (AMPK)/peroxisome proliferative-activated receptor–a (PPARa) pathway through adiponectin receptors (Adi- poRs). AdipoRon is an orally active synthetic adiponectin receptor agonist. We investigated the expression of AdipoRs and the associated intracellular pathways in 27 patients withtype2diabetesandexaminedtheeffectsofAdipoRonon DNdevelopmentinmaleC57BLKS/Jdb/db mice, glomerular endothelial cells (GECs), and podocytes. The extent of glomerulosclerosis and tubulointerstitial fibrosis correlated with renal function deterioration in human kidneys. Expres- sion of AdipoR1, AdipoR2, and Ca2+/calmodulin-dependent protein kinase kinase–b (CaMKKb) and numbers of phosphorylated liver kinase B1 (LKB1)–and AMPK-positive cells significantly decreased in the glomeruli of early stage human DN. AdipoRon treatment restored diabetes-induced renal alterations in db/db mice. AdipoRon exerted renoprotective effects by directly activating intrarenal AdipoR1 and AdipoR2, which increased CaMKKb, phosphorylated Ser431LKB1, phosphorylated Thr172AMPK, and PPARa expression independently of the systemic effects of adiponectin. AdipoRon- induced improvement in diabetes-induced oxidative stress and inhibition of apoptosis in the kidneys ameliorated relevant intracellular pathways associated with lipid accumulation and endothelial dysfunction. In high-glucose–treated human GECs and murine podocytes, AdipoRon increased intracellular Ca2+ levels that activated a CaMKKb/phosphorylated Ser431LKB1/phosphorylated Thr172AMPK/PPARa pathway and downstream signaling, thus decreasing high- glucose–induced oxidative stress and apoptosis and improving endothelial dysfunction. AdipoRon further produced cardioprotective effects through the same pathway demonstrated in the kidney. Our results show that AdipoRon ameliorates GEC and podocyte injury by activating the intracellular Ca2+/LKB1-AMPK/PPARa pathway, suggesting its efficacy for treating type 2 diabetes–associated DN.

J Am Soc Nephrol 29: 1108–1127, 2018. doi: https://doi.org/10.1681/ASN.2017060627

Received June 8, 2017. Accepted December 7, 2017. Although diabetic nephropathy (DN) is tradi- tionally characterized by hyperglycemia-induced Published online ahead of print. Publication date available at www.jasn.org. metabolic and hemodynamic changes, accumu- lating evidences suggest that derangements in Correspondence: Dr. Cheol Whee Park, Division of Nephrology, lipid metabolism play a crucial role in DN devel- Department of Internal Medicine, Seoul St. Mary’sHospital, College of Medicine, The Catholic University of Korea, 222, opment and progression. Accumulation of free Banpo-daero, Seocho-gu, Seoul 06591, Republic of Korea. fatty acids, which are otherwise used as an energy Email: [email protected] source, in tubular epithelial cells of diabetic kid- Copyright © 2018 by the American Society of Nephrology neys indicates a state of energy surplus.1 This state

1108 ISSN : 1046-6673/2904-1108 JAmSocNephrol29: 1108–1127, 2018 www.jasn.org BASIC RESEARCH of energy surplus (referred to as lipotoxicity) is character- Significance Statement ized by the deposition of fatty acid metabolites such as diacylglycerols and ceramides in nonadipose organs Adiponectin is an adipokine secreted by adipocytes known to exert favor- and leads to toxicity and cell death.2 Although mainstream able effects in the milieu of diabetic and metabolic syndrome through its anti- fl fi research indicates that slowdown of fatty acid b-oxidation in ammatory, anti brotic, and antioxidant effects. In human and animal diabetic nephropathy, the expression of adiponectin receptors is reduced. 3 leads to the accumulation of different lipid products, This manuscript reveals a favorable effect of AdipoRon, an orally active emerging evidence indicates that rapid but incomplete synthetic adiponectin receptor agonist, in protection, even reversal, of di- b-oxidation in the nonadipose tissue and by-products of abeticnephropathyproducedbythe activationof adiponectinreceptors and ++ oxidative stress promote the formation of toxic lipid inter- downstream targets through increased intracellular Ca /AMPK-LKB1/ mediates.4 PPARa pathway. AdipoRon may be a promising drug for restoration of diabetic nephropathyintype2diabetes. Adiponectin is one of the numerous adipokines secreted by adipocytes.5 It exerts favorable effects in the milieu of metabolic syndrome through anti-inflammatory, antifi- RESULTS brotic, and antioxidant effects.6 Adiponectin mediates fatty acid metabolism by inducing AMP-activated protein Human Diabetic Kidneys Show Decreased kinase (AMPK) phosphorylation and increasing peroxi- Intraglomerular AdipoR1/AdipoR2 Expression and some proliferative-activated receptor–a (PPARa)expres- Decreased CaMKKb/Phosphorylated Ser431LKB1/ sion, which in turn upregulate the expression of acyl CoA Phosphorylated Thr172AMPK/PPARa Pathway oxidase and uncoupling proteins involved in energy con- The patient group included 27 patients with biopsy-proven DN. sumption.7,8 Low circulating adiponectin levels in obese Clinical characteristics of these patients are shown (Supplemental patients with a risk of insulin resistance, type 2 diabetes, Table 1). Nondiabetic subjects included healthy controls (n=6) and cardiovascular disease9 and increased adiponectin ex- with minor urinary abnormalities. In diabetes, the extent of glo- pression in state of albuminuria indicate the protective and merulosclerosis and tubulointerstitial fibrosis markedly increased compensatory role of adiponectin10 to mitigate further with renal function deterioration (Figure 1, A and B). Immuno- renal injury against the development of overt nephropa- fluorescence analysis showed that AdipoR1/AdipoR2, CaMKKb, thy.5,11 These beneficial effects of adiponectin have promp- andnephrinexpressionandphosphorylatedAMPK-andphos- ted research on drugs that mimic adiponectin, yet this phorylated LKB1-positive cell number significantly decreased in manner of increasing the level of circulating adiponectin the glomerulus of human diabetic kidneys compared with that of is not a panacea. Moreover, adiponectin overexpression is nondiabetic control kidneys even in the earliest CKD stage. There associated with adverse effects such as reduced bone den- were no significant differences in the expression of AdipoRs and sity, left ventricular hypertrophy, weight gain, tumor their relevant downstream molecules with increasing stages of growth, and infertility.12–14 Therefore, there is a need to CKD (Figure 1, C–L) and in those with or without angiotensin develop a novel agent that could deliver the favorable ef- II receptor blocker use in the kidneys (Supplemental Figure 1). fects of adiponectin but not the detrimental pitfalls due to adiponectin excess. AdipoRon Improves Renal Function without Affecting AdipoRon is an orally active synthetic adiponectin receptor Serum Adiponectin and Glucose Levels in db/db Mice agonist developed by Okada-Iwabu et al.15 In db/db Serum adiponectin level was higher in nondiabetic mice than in di- mice, AdipoRon binds to AdipoRs AdipoR1 and AdipoR2 abetic mice. AdipoRon treatment did not affect body weight or food to activate AMPK and PPARa, respectively, and induces intake. Moreover, it did not affect plasma glucose, HbA1c, and serum such prometabolic effects as improved insulin sensitivity, creatinine levels in both nondiabetic and diabetic mice. There were no weight neutrality, and expanded life span.15 Diminished adi- differences in systolic BP among study groups and creatinine clearance ponectin-associated metabolic effect in AdipoR1-/AdipoR2- decreased in db/db mice treated with AdipoRon. Moreover, albumin- knockout mice and restoration of obesity-induced metabolic uria, urinary nitric oxide metabolites (NOx), and homeostatic model alterations in high-fat-diet–fed mice by AdipoRon adminis- assessment–insulin resistance (HOMA-IR) index improved in db/db tration with implication on fatty acid combustion suggest that mice treated with AdipoRon compared with those in db/db control AdipoRon is a promising agent for treating type 2 diabe- andindiabeticmicebeforethetreatment(Table1).Theseresults tes.15,16 This instigated us to investigate the favorable effects suggest that AdipoRon ameliorates metabolic and renal function pa- of AdipoRon against DN in db/db mice, human glomerular rameters independently of serum adiponectin and glucose levels. endothelial cells (GECs), and murine podocytes. In addition, we examined renal AdipoR1/AdipoR2 expression and AdipoRon Activates the CaMKKb/Phosphorylated Ser CaMKKb/phosphorylated Ser431LKB1/phosphorylated Thr172 431LKB1/Phosphorylated Thr172AMPK/PPARa Pathway AMPK pathway activation in type 2 diabetes and hypothesized by Increasing Intrarenal AdipoR1/AdipoR2 Expression that AdipoRon treatment ameliorated lipotoxicity and oxidative in db/db Mice stress by activating the AMPK/PPARa pathway by increasing AdipoRon treatment restored diabetes-induced decrease in AdipoR1/AdipoR2 expression. renal AdipoR1 and AdipoR2 expression to the levels present

J Am Soc Nephrol 29: 1108–1127, 2018 AdipoRon Ameliorates Diabetic Nephropathy 1109 BASIC RESEARCH www.jasn.org in control db/m mice, as indicated by abundant AdipoR1 and expression was associated with increased expression of their AdipoR2 distribution throughout the renal ultrastructure of downstream effectors such as PGC-1a, phosphorylated ACC, diabetic mice (Figure 2, A–F). AdipoRon treatment increased and phosphorylated Ser1177eNOS and decreased expression CaMKKb and phosphorylated LKB1 expression (Figure 2, D of SREBP-1c in HGECs cultured in high-glucose medium and G–I) and activated phosphorylated AMPK and PPARa, (Figure 5, M–Q). Moreover, AdipoRon treatment attenuated which are established to be the primary downstream targets of DHE expression and decreased TUNEL-positive cell number AdipoR1 and AdipoR2, respectively (Figure 2, J–L). (Figure 5, R–T). HGECs cultured in high-glucose medium were transfected with small interfering RNAs (siRNAs) against AdipoRon Ameliorates Diabetes-Induced Renal genes encoding AdipoR1 and AdipoR2 to verify which one of Damage by Reducing Intrarenal Lipotoxicity and the adiponectin receptors was responsible for activating its Oxidative Stress in db/db Mice downstream effectors. Transfection of HGECs with AdipoR1 AdipoRon treatment restored PGC-1a, phosphorylated Ser75 siRNA suppressed AdipoR1 expression and increased ACC, and phosphorylated Ser1177eNOS expression and de- AdipoR2 expression to the level present in siRNA control– creased SREBP-1c expression in diabetic mice to the degree transfected cells treated with AdipoRon and vice versa of those present in nondiabetic control mice (Figure 3, A–E). (Figure 6, A–G). AdipoRon treatment increased phos- Even distribution of red lipid droplets throughout the glomer- phorylated AMPK, total AMPK, and PGC-1a expression ulus of diabetic mice disappeared after AdipoRon treatment in HGECs transfected with AdipoR2 siRNA but not in (Figure 3F). Consistently, AdipoRon treatment decreased in- HGECs transfected with AdipoR1 siRNA; moreover, Adi- trarenal triacylglycerol (TG) and FFA levels (Figure 3, G–I). poRon treatment increased PPARa expression in HGECs Immunofluorescence analysis showed that AdipoRon treat- transfected with AdipoR1 siRNA but not in HGECs transfected ment attenuated red chromophore (DHE) formation and with AdipoR2 siRNA (Figure 6, E and H–K). Moreover, Adi- transformed blackout into green (nephrin) in the glomerulus poRon treatment did not increase phosphorylated ACC, total of db/db mice (Figure 3, J–L). Renal phenotypes of diabetic ACC, phosphorylated endothelial nitric oxide synthase control and those before the treatment exhibited similar de- (eNOS), total eNOS, and NOx expression in HGECs transfec- gree of diabetes-induced alterations; increased expansion of ted with AdipoR1 or AdipoR2 siRNA (Figure 6, J and L–N). the mesangial area and numbers of glomerular collagen IV-, ++ TGF-b–, and F4/80-positive cells. These renal alterations were AdipoRon-Induced Increase in Intracellular Ca Level Acti- 431 restored by AdipoRon administration to the level present in non- vates the CaMKKb/Phosphorylated Ser LKB1/Phosphory- 172 diabetic control mice (Figure 4, A–E). Electron microscopy anal- lated Thr AMPK/PPARa Pathway and Ameliorates ysis revealed recovery of podocyte injury, including attenuation of Lipotoxicity and Oxidative Stress in Murine Podocytes ++ glomerular basement membrane thickness, foot process efface- AdipoRon treatment increased intracellular Ca level in murine ment, and slit diaphragm width, in db/db mice (Figure 4, F–I). podocytes cultured in both low- and high-glucose media in a Moreover, AdipoRon treatment reduced the amount of TUNEL- dose-dependent manner (Supplemental Figure 2, A–D). Immu- positive endothelial (PECAM-1–positive) cells (Figure 4, J and K) nofluorescence analysis showed that AdipoRon treatment and TUNEL-positive (WT-positive) podocytes (Figure 4, L–N) in increased phosphorylated LKB1 and phosphorylated AMPK ex- diabetic mice. AdipoRon treatment decreased the levels of urinary pression in murine podocytes cultured in high-glucose medium oxidative stress markers 8-OH-dG and isoprostane in 24-hour (Supplemental Figure 2, E–G). Moreover, AdipoRon-induced urine collected from diabetic mice (Figure 4, O and P). Overall, increase in CaMKKb, phosphorylated LKB1, phosphorylated AdipoRon-AdipoR1/AdipoR2 interaction–induced activation of AMPK, and PPARa expression was associated with further in- the AMPK/PPARa pathway ameliorated lipotoxicity, apoptosis, crease in the expression of their downstream effectors such as 1177 and oxidative stress, which in turn alleviated the features of DN. PGC-1a, phosphorylated ACC, and phosphorylated Ser eNOS and decrease in the expression of SREBP-1c in murine In Vitro Studies podocytes cultured in high-glucose medium (Supplemental Fig- AdipoRon-Induced Increase in Intracellular Ca++ Level Acti- ure 2, H–Q). Moreover, AdipoRon treatment attenuated DHE vates the CaMKKb/Phosphorylated Ser431LKB1/Phosphory- expression and decreased TUNEL-positive cell number in mu- lated Thr172AMPK/PPARa Pathway and Ameliorates rine podocytes (Supplemental Figure 2, R–T). Lipotoxicity and Oxidative Stress in HGECs AdipoRon treatment increased intracellular Ca++ level in AdipoRon-Induced Increase in Cardiac AdipoR1/AdipoR2 HGECs cultured in both low- and high-glucose media in a Expression Activates the CaMKKb/Phosphorylated Ser431 dose-dependent manner (Figure 5, A–D). Immunofluores- LKB1/Phosphorylated Thr172AMPK/PPARa Pathway and cence analysis showed that AdipoRon treatment increased Ameliorates Cardiac Function and Phenotype phosphorylated LKB1 and phosphorylated AMPK levels in AdipoRon treatment improved systolic and diastolic functions as HGECs cultured in high-glucose medium (Figure 5, E–L). represented by increased fractional shortening, velocity of cir- Moreover, AdipoRon-induced increase in CaMKKb,phos- cumferential shortening, and E/A ratio without affecting cardiac phorylated LKB1, phosphorylated AMPK, and PPARa mass in diabetic mice (Supplemental Figure 3, A and B).

1110 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 1108–1127, 2018 www.jasn.org BASIC RESEARCH

Figure 1. Human diabetic kidneys show characteristic diabetic alterations in line with decreased expression of intraglomerular AdipoRs and relevant molecules according to CKD stages. Representative sections stained with (A and B) PAS reagent, (C–L) immunoﬂuorescence staining, and quantitative analyses of AdipoR1, AdipoR2, CaMKKb, phosphorylated Ser431LKB1, phosphorylated Thr172AMPK, and nephrin levels in the human diabetic and nondiabetic kidneys according to CKD stages. *P,0.05; #P,0.001 compared with other groups.

AdipoRon also decreased expression of trichrome-, collagen IV-, DISCUSSION TGF-b1–, and CTGF-positive area and reduced TUNEL- and F4/80-positive cells, indicating improvement in cardiac fibrosis, Significant decrease in circulating adiponectin level with a con- apoptosis, and inflammation (Supplemental Figure 3, C–J). comitant decrease in AdipoR1/AdipoR2expression in the mus- These favorable alterations were in line with increased expres- cle and adipose tissues of insulin-resistant ob/ob mice may be sion of AdipoR1/AdipoR2 and CaMKKb/phosphorylated Ser431 causally related to insulin resistance, dysfunctional lipid me- LKB1/phosphorylated Thr172AMPK/PPARa pathway in the tabolism, and obesity.17 This is accompanied with decreased same study group (Supplemental Figure 3, K–R). AMPK and PPARa activation, which further downregulates

J Am Soc Nephrol 29: 1108–1127, 2018 AdipoRon Ameliorates Diabetic Nephropathy 1111 BASIC RESEARCH www.jasn.org

Figure 1. Continued.

AdipoR expression. The resultant decrease in fatty acid oxida- staining of relevant receptors and molecules in the human tion and increase in fatty acid synthesis contribute to the pro- diabetic kidneys performed in this study were consistent gression of type 2 diabetes.18 Results of immunoﬂuorescence with those of previous studies, which showed a decrease in

1112 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 1108–1127, 2018 www.jasn.org BASIC RESEARCH

Table 1. Biochemical and physical characteristics of mice in the four groups before and after 4 weeks of AdipoRon treatment Characteristics db/m Cont db/m AdipoRon db/db Cont db/db AdipoRon 16 w db/db Cont Body weight, g 36.261.7 37.261.8 53.067.3a 51.164.8a 49.963.4a Kidney weight, g 0.2060.04 0.2160.03 0.2160.04 0.2160.01 0.2260.01 Heart weight, g 0.1860.02 0.1860.02 0.1860.04 0.1760.02 0.1760.03 FBS, mg/dl 152614 13867534689a 532693a 6406150a HbA1c, % 4.060.2 3.960.2 10.560.49a 11.560.8a 8.860.7b a b a HOMAIR 0.0660.01 0.0660.01 2.3760.02 0.2160.01 2.1160.11 Total cholesterol, mmol/L 2.7060.41 2.6560.51 3.5160.42c 3.3960.55c 3.8060.50c Triacylglycerols, mmol/L 1.2160.21 1.3260.19 2.1860.33a 1.6060.22c 1.9660.29a Nonesteriﬁed fatty acid, mmol/L 0.6260.22 0.6760.19 1.4260.20c 1.2160.15c 1.4160.21c 24-h albuminuria, mg/d 10.062.0 8.262.8 223.0651.9a 100.0618.8b 163.2630.3a Urine volume, ml 0.860.2 1.060.2 14.564.5a 10.363.4a,c 11.464.3a,c Serum Cr, mmol/L 17.163.1 17.361.7 19.363.7 17.563.9 18.363.1 Ccr, ml/min 0.3660.11 0.3260.23 1.1360.34a 0.7460.25c 1.1560.33a Mean systolic BP, mm Hg 10167 1026510667 1036410564 Serum adiponectin, mg/ml 10.661.2 10.861.4 4.660.5a 4.760.9a 4.560.7a

Data are mean6SD. Cont, control; FBS, fasting blood sugar; HbA1c, Hemoglobin A1c; HOMAIR, homeostatic model assessment of insulin resistance; Cr, Cre- atinine; Ccr, creatinine clearance. aP,0.001 compared with other groups. bP,0.01 compared with other groups. cP,0.05 compared with other groups.

AdipoRs and CaMKKb, LKB1, and AMPK expression, signi- study showed that AdipoR2 expression was comparable to fying the implication of AdipoR agonism in diabetes. Several AdipoR1 expression and recovered after AdipoRon treatment. studies have reported the favorable renal effect of adiponectin Because decreased AdipoR1 or AdipoR2 expression may de- against albuminuria development in rodent models. Sharma crease adiponectin sensitivity, upregulation of AdipoR1 or et al.19 verified the renoprotective effects of adiponectin by AdipoR2 expression after AdipoRon treatment may restore using adiponectin-knockout mice. They showed that adipo- the renoprotective function of adiponectin. Regarding the nectin treatment restored renal oxidative stress and that this mechanism through which AdipoR1/R2 become activated, it was achieved through the activation of AMPK, which in turn is reported that AdipoR1/AdipoR2 expression is inversely cor- was responsible for decreased podocyte permeability and al- related with plasma insulin levels in vivo. Moreover, previous buminuria.19 Identification of AdipoR1/AdipoR2 prompted in vitro studies indicate that insulin decreases AdipoR1/Adi- the need to develop the synthetic AdipoR agonist AdipoRon poR2 expression in hepatocytes through a phosphoinositide that mimicked the prometabolic effects of adiponectin but did 3–kinase/FOXO1-dependent pathway.17 In line with this, im- not exert adverse effects associated with the chronic upregu- provement in the HOMA-IR index of diabetic mice suggests lation in serum adiponectin level.16 that increased insulin sensitivity is involved in AdipoR1/Adi- Whereas AdipoR1 is abundantly expressed in the skeletal poR2 activation. muscle,20 AdipoR2 is mainly expressed in the liver.16 Both The established concept on the role of AdipoRs and their AdipoR1/AdipoR2 serve as the major AdipoRs for regulating associated pathway indicates that AdipoR1 activates the AMPK glucose and lipid metabolism in different organs. In the view pathway, whereas AdipoR2 is associated with PPARa activation of renal structure, AdipoR1 is abundantly expressed in prox- thus comprehensively reducing oxidative stress and increasing imal tubular cells and endothelial cells, podocytes, mesangial fatty acid oxidation.15 We investigated the AdipoR responsible cells, and Bowman’s capsule epithelial cells of the glomerulus; for activating its downstream effectors by transfecting cul- in contrast, AdipoR2 expression is relatively scarce across the tured HGECs with AdipoR1 or AdipoR2 siRNA. Consistent renal structure.21,22 Models of chronic renal failure show in- with the established role-play assigned to each AdipoR, inhi- creased renal AdipoR1/AdipoR2 expression with consistently bition of either one of the two receptors increased the expres- upregulated serum and urinary adiponectin levels.23 This ev- sion of the other receptor to a greater extent as a compensatory idence suggests that upregulation of AdipoR expression is a mechanism. Moreover, suppression of either one of the two compensatory mechanism to prevent ongoing renal damage. receptors affected the expression of one’s assumable down- This study showed that AdipoRon treatment restored diabe- stream effectors, confirming that phosphorylated AMPK/ tes-induced downregulation of AdipoR1/AdipoR2 expression. PGC1a pathway and PPARa were the primary downstream However, this finding was challenging to the conventional targets of AdipoR1 and AdipoR2, respectively. However, ex- conceptthatrenalAdipoR2expressionisscarce,lessan- pression of further downstream effectors, including phos- nounced than renal AdipoR1 expression.22 Results of this phorylated ACC, total ACC, phosphorylated eNOS, total

J Am Soc Nephrol 29: 1108–1127, 2018 AdipoRon Ameliorates Diabetic Nephropathy 1113 BASIC RESEARCH www.jasn.org

Figure 2. AdipoRon activates the CaMKKb/phosphorylated Ser431LKB1/phosphorylated Thr172AMPK/PPARa pathway by increasing intrarenal AdipoR1/AdipoR2 expression in db/db mice. (A–C) Representative images of immunoﬂuorescence staining and quantitative analyses of AdipoR1 and AdipoR2 expression. (D–I) Representative images of western blotting and quantitative analyses of AdipoR1, AdipoR2, CaMKKa,CaMKKb, phosphorylated Ser431LKB1, total LKB1, and b-actin levels. (J–L) Representative images of western blotting and quantitative analyses of phosphorylated AMPK Thr172, total AMPK, PPARa,andb-actin levels. *P,0.05 and **P,0.01 versus db/db mice. Cont, control.

1114 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 1108–1127, 2018 www.jasn.org BASIC RESEARCH

Figure 2. Continued. eNOS, and NOx, did not increase in HGECs transfected with albuminuria and glomerulosclerosis in patients with diabe- AdipoR1 or AdipoR2 siRNA even after AdipoRon treatment, tes,29 we explored changes in relevant signaling pathways in suggesting that dual activation of AdipoR1/AdipoR2 is re- both HGECs and podocytes. The decreased number of func- quired for exerting prometabolic effects through the AMPK/ tioning podocytes is an independent risk factor for the de- PPARa axis. velopment and progression of DN.29 In this study, results of AdipoR activation stimulates AMPK and PPARa to pro- in vitro experiments were consistent with those of in vivo duce antidiabetic effects through their downstream effectors experiments, suggesting that AdipoRon exerted favorable ef- PGC-1a,ACC,andSREBP-1c.AMPKandPPARa modulate fects on cellular levels as well by activating AMPK. The net lipid metabolism by regulating fatty acid oxidation. ACC, a effect was reduced oxidative stressandapoptosisindifferent key enzyme that catalyzes the rate-limiting steps of fatty acid cell types, which improved the renal phenotype in terms of synthesis, is inactivated upon phosphorylation by AMPK, fur- restored podocyte number and decreased podocyte foot pro- ther accelerating fatty acid oxidation. On the other hand, cess widening. These alterations ultimately decreased uri- AMPK and PPARa activation suppresses the expression of nary albumin excretion rate. Moreover, decreased creatinine such lipogenesis-associated enzymes as SREBP-1c. Moreover, clearance observed in AdipoRon-treated db/db mice pro- PPARa-activated PGC-1a/b/estrogen-related receptor poses its possible renoprotective mechanism in which im- (ERR)–1a axis reduces oxidative stress by enhancing mito- proved hyperinsulinemia-induced hyperﬁltration would chondrial oxidative capacity.24,25 Increased eNOS level is ex- further enhance reciprocal regulation between HGECs and pected to neutralize ROS, reduce adhesion molecule synthe- podocytes, resulting in reduced albuminuria. Collectively, sis, and suppress cell proliferation, which collectively help the results of in vitro experiments suggest that endothelial confer protective effects on endothelial cells against albumin- dysfunction affects podocyte injury and resultant albuminuria development.26–28 uria by altering paracrine communication or crosstalk be- In keeping with an expanding body of evidence that endo- tween HGECs and podocytes. thelial dysfunction isa majorcontributor in the pathogenesisof Next,weexploredtheexpression ofLKB1andCaMKK,which DNand because glomerular podocyte injury is a major cause of are upstream AMPK kinases,30 to determine the missing link

J Am Soc Nephrol 29: 1108–1127, 2018 AdipoRon Ameliorates Diabetic Nephropathy 1115 BASIC RESEARCH www.jasn.org

Figure 3. AdipoRon ameliorates diabetes-induced intrarenal lipotoxicity and oxidative stress in db/db mice. (A–E) Representative images of western blotting and quantitative analyses of PGC-1a, phosphorylated ACC, total ACC, SREBP-1c, phosphorylated Ser1177eNOS, total eNOS, and b-actin levels. (F–I) Representative images of oil red O staining of the renal cortex and quantitative analyses of intrarenal NEFA, TG, and TC levels. (J–L) Representative images of immunoﬂuorescence staining and quantitative analyses of DHE and nephrin expression. *P,0.05 and **P,0.01 versus db/db mice. Cont, control.

between AdipoRs and key AdipoR-activated molecules. En- Viewed in this light, both the upstream signaling pathways hanced cytosolic LKB1 localization and Ca++-induced CaMKK involved in AdipoRon-induced AMPK activation provide a activation promote adiponectin-stimulated AMPK activation in mechanism in which regulation of AMPK activity could be dy- muscle cells.31 Among two isoforms of CaMKK, AdipoRon namically yet ﬁnely tuned in accordance with the surrounding seems to activate CaMKKb rather than CaMKKa in the kidneys environmental and nutritional requirements.34 Moreover, by stimulating intracellular Ca++ release. AdipoRon treatment of CaMKKb activation by AdipoRon-induced Ca++ inﬂux serves HGECs and murine podocytes increased intracellular Ca++ con- as exercise mimetics further enhancing PGC-1a expression in- centration in a dose-dependent manner. Although LKB1 acti- dependently of AMPK. vates AMPK primarily through an AMP-dependent mechanism Several studies have reported that AdipoRon exerts a glu- under high cellular energy stress,32 CaMKKs phosphorylate cose-lowering effect. However, this study did not show changes AMPK through Ca++/calmodulin independent of AMP.33 in serum glucose and HbA1c levels in AdipoRon-treated db/db

1116 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 1108–1127, 2018 www.jasn.org BASIC RESEARCH

Figure 3. Continued.

mice.15 This discrepancy may be because of the use of different serum adiponectin level. Human diabetic kidneys exhibit de- doses (50 mg/kg versus 30 mg/kg) and treatment durations creased AdipoR1/AdipoR2, CaMKKb, phosphorylated AMPK, (10 days versus 4 weeks). Nevertheless, these favorable results phosphorylated LKB1, and nephrin expression even in the early reinforce to attest to the role of AdipoRon in attenuating renal stages of CKD and show negligible changes in relevant molecular lipotoxicity independent of its systemic glucose-lowering expression with increasing stages of CKD. Moreover, significant effect. improvements in renal function and phenotype before and after In terms of hemodynamic effect, AdipoRon improved both the treatment of AdipoRon to the degree comparable to that of db/ systolic and diastolic heart functions without affecting cardiac m control mice represent its reversal effect in the progression of DN. mass that was accompanied by decreased cardiac fibrosis, in- These findings suggest that when translated into the clinical field, flammation, and apoptosis. These favorable effects on diabetic AdipoR agonism in the kidneys is a promising therapeutic strategy heart were delivered through the activation of both AdipoRs for activating AMPK. In this context, AdipoRon may be a novel and their downstream pathway that was delineated in the kid- promising candidate to start with in greeting the new era of DN. ney. This is in accordance with previous studies that associate reduced expression of AdipoR1 in the heart muscle to intol- erance of ischemic injury35 and consequent pathologic CONCISE METHODS processes36 and that reduced activities of mitochondrial oxidative phosphorylation complexes; i.e., of AMPK, and PGC1 Study Subjects signaling contribute to diabetic cardiomyopathy.37 The study recruited a total of 33 healthy subjects (n=6) and type 2 Results of this study clearly demonstrated that AdipoRon im- diabetic patients (n=27) with various stages of CKD according to the proved high-glucose–induced oxidative stress and lipotoxicity by modified GFR (Supplemental Table 1). We reviewed biopsy samples activating intrarenal AdipoR1/AdipoR2, which in turn activated of human diabetic kidneys and nondiabetic kidneys according to the intracellular Ca++/AMPK/PPARa pathway without affecting CKD stages. Experimental research involving human specimens

J Am Soc Nephrol 29: 1108–1127, 2018 AdipoRon Ameliorates Diabetic Nephropathy 1117 BASIC RESEARCH www.jasn.org

Figure 4. AdipoRon ameliorated features of DN through decreased intrarenal fibrosis, inflammation, apoptosis, and recovered podocyte injury. (A) Representative sections stained with PAS reagent are shown to estimate the mesangial fractional area (%) (B) together with the results of quantitative analysis according to groups. Immunohistochemical staining and quantitative analyses of (A and C) type IV collagen-, (A and D) TGF-b1-, and (A and E) F4/80-positive area. (F–I) Representative electron microscopic images of the glomerulus and quantitative analysis according to groups. (J and L, respectively) Representative images of immunofluorescence staining of TUNEL- positive endothelial cells and podocytes and (K, M, and N, respectively) quantitative analyses of the results. (O and P) Twenty-four-hour urinary 8-OH-dG and isoprostane levels in the study mice; *P,0.05, **P,0.01, and #P,0.001 compared with other groups. Col IV, type IV collagen; Cont, control.

was approved by the Institutional Review Board at the College of miceweredividedintofourgroups,andreceivedeitheraregulardiet Medicine, Catholic University of Korea (KC16SISI0158). chow or a diet containing AdipoRon. AdipoRon (30 mg/kg; Sigma, St Louis, MO) was mixed into the standard chow diet and provided to db/db Experimental Methods mice (db/db+AdipoR, n=8) and age- and sex-matched db/m mice (db/m Six-week-old male C57BLKS/J db/m and db/db mice were purchased from +AdipoR, n=8) from 16 weeks of age for 4 weeks. Control db/db (db/db Jackson Laboratories (Bar Harbor, ME). Male C57BLKS/J db/m and db/db cont, n=8) and db/m mice (db/m cont, n=8) were fed normal diet chow.

1118 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 1108–1127, 2018 www.jasn.org BASIC RESEARCH

Figure 4. Continued.

J Am Soc Nephrol 29: 1108–1127, 2018 AdipoRon Ameliorates Diabetic Nephropathy 1119 BASIC RESEARCH www.jasn.org

Figure 5. AdipoRon-induced increase in intracellular Ca++ concentration activates its downstream signaling and ameliorates lipotoxicity and oxidative stress in HGECs. (A–D) Intracellular Ca++ concentration in HGECs cultured in low- or high-glucose medium with or without AdipoRon. (E–G) Representative images of immunoﬂuorescence staining and quantitative analyses of phosphorylated Ser431LKB1 and phosphorylated Thr172AMPK levels. (H) Representative images of western blotting analysis of CaMKKb, phosphorylated Ser431LKB1, phosphorylated Thr172AMPK, total AMPK, PPARa,andb-actin levels and (I–L) their quantitative analyses. (M–Q) Representative images of western blotting and quantitative analyses of PGC-1a, phosphorylated ACC, total ACC, SREBP-1c, phosphorylated Ser1177eNOS, total eNOS, and b-actin levels. (R–T) Representative images of immu- noﬂuorescence staining and quantitative analyses of DHE expression and TUNEL-positive cells. *P,0.05 compared with LG 0; **P,0.001 compared with LG and HG control, respectively; #P,0.001 compared with LG+10 and HG+10, respectively. Cont, control; HG, high-glucose; LG, low-glucose.

1120 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 1108–1127, 2018 www.jasn.org BASIC RESEARCH

Figure 5. Continued.

At weeks 16 and 20, all animals were anesthetized by intraperitoneal in- To evaluate oxidative stress, we measured the 24-hour urinary 8-OH- jection of 30 mg/kg tiletamine plus zolazepam (Zoletil; Virbac, Carros, dG (8-hydroxy-29-deoxyguanosine; OXIS Health Products, Inc., France)and10mg/kgxylazinehydrochloride(Rompun;Bayer,Leverku- Portland, OR), 8-epi-PG F2a (OXIS Health Products, Inc., Portland, sen, Germany) to compare the effect of AdipoRon before and after the OR), and NOx (BioVision, Mountain View, CA). The kidney lipids treatment. Blood was collected from the left ventricle and the plasma was were extracted by the method of Bligh and Dyer with slight modifi- stored at 270°C for subsequent analyses. All animal experiments were cationsaspreviouslydescribed.38 Total cholesterol and TG concen- performed in accordance with the Laboratory Animals Welfare Act, Guide trations were measured by an autoanalyzer (Hitachi 917, Tokyo, Ja- for the Care and Use of Laboratory Animals, and approved by the In- pan) using commercial kits (Wako, Osaka, Japan). Nonesterified stitutional Animal Care and Use Committee (IACUC) at the College of fatty acid levels were measured with a JCA-BM1250 automatic ana- Medicine, Catholic University of Korea (CUMC-2017–0251–01). lyzer (JEOL, Tokyo, Japan). In addition, we performed oil red O staining to evaluate the effect of AdipoRon on lipid accumulation Assessment of Renal Function and Oxidative Stress in the glomerulus. and Intrarenal Lipids A 24-hour urine collection was obtained using metabolic cages at Light Microscopic Study weeks 16 and 20 and urinary albumin concentration was measured Kidney samples were fixed in 10% buffered formalin and embedded in by an immunoassay (Bayer, Elkhart, IN). Plasma and urine creatinine paraffin. The histology was assessed after periodic acid–Schiff staining concentrations were measured by an enzymatic creatinine assay (PAS) and trichrome staining. The mesangial matrix and glomerular (Samkwang Medical Laboratory, Seoul, Korea). The concentration of tuft areas were quantified for each glomerular cross-section using serum adiponectin was determined by ELISA (Biosource, Camarillo, CA). PAS-stained sections as previously reported.39 More than 30

J Am Soc Nephrol 29: 1108–1127, 2018 AdipoRon Ameliorates Diabetic Nephropathy 1121 BASIC RESEARCH www.jasn.org

Figure 5. Continued. glomeruli which were cut through the vascular pole were counted per Immunohistochemistry kidney and the average was used for analysis. Heart samples collected We performed immunohistochemistry for TGF-b1, type IV collagen, after systemic perfusion with PBS were ﬁxed in 4% paraformalde- F4/80, TUNEL, and CTGF. Renal (4-mm thick) and cardiac (5-mm hyde. Ten consecutive heart cross-sections stained in trichrome were thick) tissue sections were incubated overnight with anti–TGF-b1 analyzed. (1:100; R&D Systems, Minneapolis, MN), anti-COL IV (1:200;

1122 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 1108–1127, 2018 www.jasn.org BASIC RESEARCH

Figure 6. Effect of AdipoR1 and AdipoR2 siRNAs on downstream signaling of AdipoRon-treated HGECs indicates that AdipoR1 and AdipoR2 activate the AMPK and PPARa pathway, respectively. (A–I) Representative images of western blotting and quantitative analyses of AdipoR1, AdipoR2, phosphorylated Thr172AMPK, total AMPK, PPARa,andb-actin levels. (J–N) Representative images of western blotting and quantitative analyses of PGC-1a, phosphorylated ACC, total ACC, phosphorylated Ser1177eNOS, total eNOS, NOx, and b-actin levels; *P,0.05 and **P,0.01 compared with other groups. HG, high glucose; LG, low glucose.

J Am Soc Nephrol 29: 1108–1127, 2018 AdipoRon Ameliorates Diabetic Nephropathy 1123 BASIC RESEARCH www.jasn.org

Figure 6. Continued.

Biodesign International, Saco, ME), and anti-F4/80 (1:200; Serotek, Ox- nephrin by using tyramide signal amplification fluorescence system (Per- ford, UK) antibodies with additional cardiac tissue staining for anti- kin Elmer, Waltham, MA) and counterstained with the 4,6-diamidino2- CTGF (1:250 in blocking solution; Abcam, Cambridge, UK), and the phenylindole (DAPI). Detection of apoptotic cells in the formalin-fixed, in situ TUNEL assay (Apoptag Plus; Intergen, New York, NY) in a hu- paraffin-embedded tissue was performed by in situ TUNEL, using an midified chamber at 4°C. The antibodies were localized using a perox- ApopTag In Situ Apoptosis Detection Kit (Chemicon-Millipore, Billerica, idase-conjugated secondary antibody and the Vector Impress kit (Vector MA). The TUNEL reaction was assessed in the whole glomeruli biopsy Laboratories, Burlingame, CA) with a 3,3-diamninobenzidine substrate sample under 2003 and 4003 magnifications. The fluorescent images solution with nickel chloride enhancement. The sections were then de- were examined under a laser scanning confocal microscope system (Carl hydrated in ethanol, cleared in xylene, and mounted without counter- Zeiss LSM 700, Oberkochen, Germany). staining. The sections were examined in a blinded manner under light microscopy (Olympus BX-50; Olympus Optical, Tokyo, Japan). To Electron Microscopy quantify the stained areas, approximately 20 views (2003 and 4003 For transmission electron microscopy, kidney specimens were fixed in magnifications) were used, which were located randomly in the renal 4% paraformaldehyde and 2.5% glutaraldehyde in 0.1 M phosphate cortex and corticomedullary junction of each slide (Scion Image Beta buffer overnight at 4°C. After washing in 0.1 M phosphate buffer, the 4.0.2, Frederick, MD). For cardiac tissues, 20 views were randomly lo- specimens were postfixed with 1% osmium tetroxide in the same cated in the middle portion of the myocardium of each slide (Scion buffer for 1 hour. The specimens were then dehydrated using a series Image Beta 4.0.2). of graded ethanol, exchanged through acetone, and embedded in Epon 812. Ultrathin sections (approximately 70–80 nm) were ob- Immunofluorescence Analysis tained by ultramicrotome (Leica Ultracut UCT, Leica, Germany) We performed immunofluorescence analysis for AdipoR1, AdipoR2, and were double stained with uranyl acetate and lead citrate and CaMKKb, pLKB1, pAMPKa, PECAM-1, WT1, DHE, TUNEL, and examined using a transmission electron microscope (JEM 1010,

1124 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 1108–1127, 2018 www.jasn.org BASIC RESEARCH

Tokyo, Japan) at 60 kV. We measured GBM thickness, slit diaphragm The 340-nm fluorescence of fura-2 increases and the 380-nm fluo- diameter, and foot process width within at least four glomeruli from rescence decreases with increasing intracellular Ca2+ ([Ca2+]i). For each mouse, and four mice from each study group were examined. [Ca2+]i measurements, HGECs and podocytes (20,000 cells/well) were plated on black 96-well plates with a clear bottom in complete Western Blot Analysis and Enzyme Activity medium. After 1 day the cells were serum-starved for 2 hours. In the Determination last 45 minutes of calcium-free serum-starvation, 5 mM FURA-2AM The total proteins of the renal cortical and cardiac tissues were extracted was added to the cells, then rinsed with Hanks Balanced Salt Solution with a Pro-Prep Protein Extraction Solution (Intron Biotechnology, (Gibco BRL, Grand Island, NY). FURA-2AM–loaded cells were Gyeonggi-Do, Korea), following the manufacturer’s instructions. Western sequentially excited at 340 and 380 nm by spectrophotometer micro- assay was performed with specific antibodies for AdipoR1 (Abcam, Cam- plate reader (Synergy MX; BioTek, Winooski, VT). AdipoRon- bridge, UK), AdipoR2 (Abcam, Cambridge, UK), phospho-Thr172 induced [Ca2+]i were quantified by measurement of area under curve AMPK (Cell Signaling Technology, Danvers, MA), total AMPK (Cell Sig- and peak amplitude for the rise in relative [Ca2+]i. naling Technology, Danvers, MA), PPARa (Abcam, Cambridge, UK), PPARg coactivator (PGC)–1a (1:2000; Novus Biologicals, Littleton, Assessment of Cardiac Function by Echocardiographic CO), sterol regulatory element–binding protein (SREBP)–1c (Santa Parameters Cruz Biotechnology, Santa Cruz, CA), phosphorylated acetyl-CoA car- Cardiac size and function were assessed by an echocardiogram boxylase (pACC) (Santa Cruz Biotechnology, Santa Cruz, CA), total ACC using a Hewlett-Packard Sonus 4500 ultrasound machine (Agilent (Santa Cruz Biotechnology, Santa Cruz, CA), phospho-Ser1177 eNOS Technologies, Edmonton, Alberta, Canada) while the animals were (Cell Signaling Technology, Danvers, MA), total eNOS (Cell Signaling anesthetized. Technology, Danvers, MA), and b-actin (Sigma-Aldrich, St. Louis, – MO). After incubation with horseradish peroxidase conjugated anti- Statistical Analyses mouse or anti-rabbit IgG (Cell Signaling Technology, Danvers, MA), tar- The data are expressed as means6SD. Differences between the groups get proteins were visualized by an enhanced chemiluminescence substrate were examined for statistical significance using ANOVAwith Bonfer- (ECL Plus; GE Healthcare Bio-Science, Piscataway, NJ). roni correction using SPSS version 11.5 (SPSS, Chicago, IL). A P value ,0.05 was considered statistically significant. Cell Culture and siRNA Transfection HGECs (Angio-Proteomie, Boston, MA) were cultured in endothelial growth medium (Angio-Proteomie, Boston, MA) at 37°C in a hu- midified, 5% CO2/95% air atmosphere. Passages 4–8wereusedinall ACKNOWLEDGMENTS experiments. We also cultured conditionally immortalized mouse podocytes that were kindly provided by Dr. Kang (Yonsei University, This study was supported by grants from the Basic Science Research College of Medicine, Seoul, Korea) as previously described.40 The Program through the National Research Foundation of Korea funded HGECs and podocytes were then exposed to low glucose (5 mmol/ by the Ministry of Education, Science and Technology (J.H.L.: L D-glucose) or high glucose (35 mmol/L D-glucose), with or with- 2015R1D1A1A01056984, and C.W.P.: 2016R1A2B2015980). out the additional 6-hour application of AdipoRon (5, 10, and The abstract of this study appeared on a poster for the American 50 nM). Western blot analysis was performed with specific antibodies Society of Nephrology Kidney Week 2017, held in New Orleans, LA, for AdipoR1, AdipoR2, phospho-Thr172 AMPK, total AMPK, and on November 2, 2017. b-actin. siRNAs, targeted to AdipoR1, AdipoR2, and scrambled Y.K., J.H.L., M.Y.K., E.N.K., H.E.Y., S.J.S., B.S.C., Y.S.K., Y.S.C., siRNA (siRNA cont), were complexed with transfection reagent (Lip- and C.W.P.designed and conceptualized the experiments. Y.K., J.H.L., ofectamine 2000; Invitrogen, Carlsbad, CA), according to the man- M.Y.K., and C.W.P. conducted the experiments and analyzed the ufacturer’s instructions. The sequences of the siRNAs were as follows: data. Y.K. and C.W.P. wrote the manuscript. All of the authors fi AdipoR1, 59-GGACAACGACUAUCUGCUACATT-39; AdipoR2, 59- critically analyzed the manuscript and approved its nal version for CCAACUGGAUGGUACACGA-39; and nonspecific scrambled publication. siRNA, 59-CCUACGCCACCAAUUUCGU-39 (Bioneer, Daejeon, Korea). HGECs in six-well plates were transfected with a final concentration of 50 nM AdipoR1 and AdipoR2 siRNAs using Lip- DISCLOSURES ofectamine2000 in Opti-MEM(R) I reduced serum medium (Gibco None. Invitrogen, Carlsbad, CA) for 24 hours and then the medium was changed back to growth medium for additional incubation. After transfection, cells were treated with AdipoRon (50 nM) in high- REFERENCES glucose media to evaluate the effects of siRNAs on HGECs reactions. 1. Jiang T, Wang Z, Proctor G, Moskowitz S, Liebman SE, Rogers T, Lucia MS, Li J, Levi M: Diet-induced obesity in C57BL/6J mice causes increased renal 2+ Intracellular Ca Measurement lipid accumulation and glomerulosclerosis via a sterol regulatory element- Calcium concentrations were determined from the ratio of fura-2 binding protein-1c-dependent pathway. JBiolChem280: 32317–32325, fluorescence intensity at 340-nm excitation and 380-nm excitation. 2005

J Am Soc Nephrol 29: 1108–1127, 2018 AdipoRon Ameliorates Diabetic Nephropathy 1125 BASIC RESEARCH www.jasn.org

2. Goldberg IJ, Trent CM, Schulze PC: Lipid metabolism and toxicity in the in nonalcoholic steatohepatitis through impaired mitochondrial bio- heart. Cell Metab 15: 805–812, 2012 genesis. Hepatology 60: 133–145, 2014 3. Liu L, Shi X, Bharadwaj KG, Ikeda S, Yamashita H, Yagyu H, Schaffer JE, 19. Sharma K, Ramachandrarao S, Qiu G, Usui HK, Zhu Y, Dunn SR, Yu YH, Goldberg IJ: DGAT1 expression increases heart triglyceride Ouedraogo R, Hough K, McCue P, Chan L, Falkner B, Goldstein BJ: content but ameliorates lipotoxicity. JBiolChem284: 36312–36323, Adiponectin regulates albuminuria and podocyte function in mice. J 2009 Clin Invest 118: 1645–1656, 2008 4. Yilmaz M, Hotamisligil GS: Damned if you do, damned if you don’t: The 20. Iwabu M, Yamauchi T, Okada-Iwabu M, Sato K, Nakagawa T, Funata M, conundrum of adipose tissue vascularization. Cell Metab 17: 7–9, 2013 Yamaguchi M, Namiki S, Nakayama R, Tabata M, Ogata H, Kubota N, 5. Hotta K, Funahashi T, Arita Y, Takahashi M, Matsuda M, Okamoto Y, Takamoto I, Hayashi YK, Yamauchi N, Waki H, Fukayama M, Nishino I, Iwahashi H, Kuriyama H, Ouchi N, Maeda K, Nishida M, Kihara S, Sakai Tokuyama K, Ueki K, Oike Y, Ishii S, Hirose K, Shimizu T, Touhara K, N, Nakajima T, Hasegawa K, Muraguchi M, Ohmoto Y, Nakamura T, Kadowaki T: Adiponectin and AdipoR1 regulate PGC-1alpha and mi- Yamashita S, Hanafusa T, Matsuzawa Y: Plasma concentrations of a tochondria by Ca(2+) and AMPK/SIRT1. Nature 464: 1313–1319, 2010 novel, adipose-specific protein, adiponectin, in type 2 diabetic pa- 21. Cammisotto PG, Londono I, Gingras D, Bendayan M: Control of gly- tients. Arterioscler Thromb Vasc Biol 20: 1595–1599, 2000 cogen synthase through ADIPOR1-AMPK pathway in renal distal tu- 6. de Jong PE, Curhan GC: Screening, monitoring, and treatment of al- bules of normal and diabetic rats. Am J Physiol Renal Physiol 294: buminuria: Public health perspectives. J Am Soc Nephrol 17: 2120– F881–F889, 2008 2126, 2006 22. Perri A, Vizza D, Lofaro D, Gigliotti P, Leone F, Brunelli E, Malivindi R, 7. Kahn BB, Alquier T, Carling D, Hardie DG: AMP-activated protein ki- De Amicis F, Romeo F, De Stefano R, Papalia T, Bonofiglio R: Adipo- nase: Ancient energy gauge provides clues to modern understanding nectin is expressed and secreted by renal tubular epithelial cells. J of metabolism. Cell Metab 1: 15–25, 2005 Nephrol 26: 1049–1054, 2013 8. Yamauchi T, Kamon J, Waki H, Terauchi Y, Kubota N, Hara K, Mori Y, Ide T, 23. Yu Y, Bao BJ, Fan YP, Shi L, Li SQ: Changes of adiponectin and its re- Murakami K, Tsuboyama-Kasaoka N, Ezaki O, Akanuma Y, Gavrilova O, ceptors in rats following chronic renal failure. Ren Fail 36: 92–97, 2014 Vinson C, Reitman ML, Kagechika H, Shudo K, Yoda M, Nakano Y, Tobe K, 24. Hong YA, Lim JH, Kim MY, Kim TW, Kim Y, Yang KS, Park HS, Choi SR, Nagai R, Kimura S, Tomita M, Froguel P, Kadowaki T: The fat-derived Chung S, Kim HW, Kim HW, Choi BS, Chang YS, Park CW: Fenofibrate hormone adiponectin reverses insulin resistance associated with both improves renal lipotoxicity through activation of AMPK-PGC-1a in db/ lipoatrophy and obesity. Nat Med 7: 941–946, 2001 db mice. PLoS One 9: e96147, 2014 9. Tamba S, Nakatsuji H, Kishida K, Noguchi M, Ogawa T, Okauchi Y, 25. Kim MY, Lim JH, Youn HH, Hong YA, Yang KS, Park HS, Chung S, Ko SH, Nishizawa H, Imagawa A, Nakamura T, Matsuzawa Y, Funahashi T, Shin SJ, Choi BS, Kim HW, Kim YS, Lee JH, Chang YS, Park CW: Re- Shimomura I: Relationship between visceral fat accumulation and uri- sveratrol prevents renal lipotoxicity and inhibits mesangial cell gluco- nary albumin-creatinine ratio in middle-aged Japanese men. Athero- toxicity in a manner dependent on the AMPK-SIRT1-PGC1a axis in db/ sclerosis 211: 601–605, 2010 db mice. Diabetologia 56: 204–217, 2013 10. Kacso IM, Bondor CI, Kacso G: Plasma adiponectin is related to the 26. Davis BJ, Xie Z, Viollet B, Zou MH: Activation of the AMP-activated progression of kidney disease in type 2 diabetes patients. Scand J Clin kinase by antidiabetes drug metformin stimulates nitric oxide synthesis Lab Invest 72: 333–339, 2012 in vivo by promoting the association of heat shock protein 90 and en- 11. Christou GA, Kiortsis DN: The role of adiponectin in renal physiology dothelial nitric oxide synthase. Diabetes 55: 496–505, 2006 and development of albuminuria. J Endocrinol 221: R49–R61, 2014 27. Ouedraogo R, Wu X, Xu SQ, Fuchsel L, Motoshima H, Mahadev K, 12. Cha DR, Zhang X, Zhang Y, Wu J, Su D, Han JY, Fang X, Yu B, Breyer Hough K, Scalia R, Goldstein BJ: Adiponectin suppression of high- MD, Guan Y: Peroxisome proliferator activated receptor alpha/gamma glucose-induced reactive oxygen species in vascular endothelial cells: dual agonist tesaglitazar attenuates diabetic nephropathy in db/db Evidence for involvement of a cAMP signaling pathway. Diabetes 55: mice. Diabetes 56: 2036–2045, 2007 1840–1846, 2006 13. Holland WL, Scherer PE: Cell Biology. Ronning after the adiponectin 28. Rakatzi I, Mueller H, Ritzeler O, Tennagels N, Eckel J: Adiponectin receptors. Science 342: 1460–1461, 2013 counteracts cytokine- and fatty acid-induced apoptosis in the pancre- 14. Yang J, Zhang D, Li J, Zhang X, Fan F, Guan Y: Role of PPARgamma in atic beta-cell line INS-1. Diabetologia 47: 249–258, 2004 renoprotection in Type 2 diabetes: Molecular mechanisms and thera- 29. Siddiqi FS, Advani A: Endothelial-podocyte crosstalk: The missing peutic potential. Clin Sci (Lond) 116: 17–26, 2009 link between endothelial dysfunction and albuminuria in diabetes. 15. Okada-Iwabu M, Yamauchi T, Iwabu M, Honma T, Hamagami K, Diabetes 62: 3647–3655, 2013 Matsuda K, Yamaguchi M, Tanabe H, Kimura-Someya T, Shirouzu M, 30. Yamauchi T, Kamon J, Minokoshi Y, Ito Y, Waki H, Uchida S, Yamashita Ogata H, Tokuyama K, Ueki K, Nagano T, Tanaka A, Yokoyama S, S, Noda M, Kita S, Ueki K, Eto K, Akanuma Y, Froguel P, Foufelle F, Ferre Kadowaki T: A small-molecule AdipoR agonist for type 2 diabetes and P, Carling D, Kimura S, Nagai R, Kahn BB, Kadowaki T: Adiponectin short life in obesity. Nature 503: 493 –499, 2013 stimulates glucose utilization and fatty-acid oxidation by activating 16. Yamauchi T, Nio Y, Maki T, Kobayashi M, Takazawa T, Iwabu M, AMP-activated protein kinase. Nat Med 8: 1288–1295, 2002 Okada-Iwabu M, Kawamoto S, Kubota N, Kubota T, Ito Y, Kamon J, 31. Zhou L, Deepa SS, Etzler JC, Ryu J, Mao X, Fang Q, Liu DD, Torres JM, Tsuchida A, Kumagai K, Kozono H, Hada Y, Ogata H, Tokuyama K, Jia W, Lechleiter JD, Liu F, Dong LQ: Adiponectin activates AMP- Tsunoda M, Ide T, Murakami K, Awazawa M, Takamoto I, Froguel P, activated protein kinase in muscle cells via APPL1/LKB1-dependent HaraK,TobeK,NagaiR,UekiK,KadowakiT:Targeteddisruptionof and phospholipase C/Ca2+/Ca2+/calmodulin-dependent protein kinase AdipoR1 and AdipoR2 causes abrogation of adiponectin binding and kinase-dependent pathways. JBiolChem284: 22426–22435, 2009 metabolic actions. Nat Med 13: 332–339, 2007 32. Sanders MJ, Grondin PO, Hegarty BD, Snowden MA, Carling D: In- 17. Tsuchida A, Yamauchi T, Ito Y, Hada Y, Maki T, Takekawa S, Kamon J, vestigating the mechanism for AMP activation of the AMP-activated Kobayashi M, Suzuki R, Hara K, Kubota N, Terauchi Y, Froguel P, Nakae protein kinase cascade. Biochem J 403: 139–148, 2007 J, Kasuga M, Accili D, Tobe K, Ueki K, Nagai R, Kadowaki T: Insulin/ 33. Suter M, Riek U, Tuerk R, Schlattner U, Wallimann T, Neumann D: Foxo1 pathway regulates expression levels of adiponectin receptors Dissecting the role of 59-AMP for allosteric stimulation, activation, and and adiponectin sensitivity. J Biol Chem 279: 30817–30822, 2004 deactivation of AMP-activated protein kinase. J Biol Chem 281: 32207– 18. Handa P, Maliken BD, Nelson JE, Morgan-Stevenson V, Messner DJ, 32216, 2006 Dhillon BK, Klintworth HM, Beauchamp M, Yeh MM, Elfers CT, Roth CL, 34. Woods A, Dickerson K, Heath R, Hong SP, Momcilovic M, Johnstone Kowdley KV: Reduced adiponectin signaling due to weight gain results SR, Carlson M, Carling D: Ca2+/calmodulin-dependent protein kinase

1126 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 1108–1127, 2018 www.jasn.org BASIC RESEARCH

kinase-beta acts upstream of AMP-activated protein kinase in mam- barley and oats are similarly antiatherogenic in hypercholesterolemic malian cells. Cell Metab 2: 21–33, 2005 Syrian golden hamsters. JNutr133: 468–475, 2003 35. Li J, Su S, Zong X: Analysis of the association between adiponectin, 39. Shin SJ, Lim JH, Chung S, Youn DY, Chung HW, Kim HW, et al.: Per- adiponectin receptor 1 and diabetic cardiomyopathy. Exp Ther Med 7: oxisome proliferator-activated receptor-alpha activator fenoﬁbrate 1023–1027, 2014 prevents high-fat diet-induced renal lipotoxicity in spontaneously hy- 36. Wang Y, Ma XL, Lau WB: Cardiovascular adiponectin resistance: The pertensive rats. Hypertens Res. 32: 835–845, 2009 critical role of adiponectin receptor modiﬁcation. Trends Endocrinol 40. Mundel P, Reiser J, Zúñiga Mejía Borja A, Pavenstädt H, Davidson GR, Metab 28: 519–530, 2017 Kriz W, Zeller R: Rearrangements of the cytoskeleton and cell contacts 37. Koentges C, König A, Pfeil K, Hölscher ME, Schnick T, Wende AR, induce process formation during differentiation of conditionally im- Schrepper A, Cimolai MC, Kersting S, Hoffmann MM, Asal J, Osterholt mortalized mouse podocyte cell lines. Exp Cell Res 236: 248–258, 1997 M, Odening KE, Doenst T, Hein L, Abel ED, Bode C, Bugger H: Myo- cardial mitochondrial dysfunction in mice lacking adiponectin receptor 1. Basic Res Cardiol 110: 37, 2015 38. Delaney B, Nicolosi RJ, Wilson TA, Carlson T, Frazer S, Zheng GH, This article contains supplemental material online at http://jasn.asnjournals. Hess R, Ostergren K, Haworth J, Knutson N: Beta-glucan fractions from org/lookup/suppl/doi:10.1681/ASN.2017060627/-/DCSupplemental.

J Am Soc Nephrol 29: 1108–1127, 2018 AdipoRon Ameliorates Diabetic Nephropathy 1127 Supplementary Table 1. Biochemical and physical characteristics of healthy normal control subjects (Cont) and patients with DN according to CKD stages.

Characteristics Cont Stage 1 Stage 2 Stage 3 Stage 4 Stage 5

No. of patient (M) 6 (3) 6 (3) 6 (3) 6 (4) 6 (3) 3 (2)

Age (years) 31 ± 5 32 ± 4 53 ± 5 47 ± 6 60 ± 4 57 ± 5

Medication NA ARB (3) ARB (4) ACEi (2) ARB (6) ARB (5) ACEi (1) ARB (3)

Serum Cr (mg/dl) 0.81 ± 0.08 0.82 ± 0.11 1.25 ± 0.22 1.98 ± 0.73 2.53 ± 0.70 6.45 ± 1.21

eGFR (ml/min/1.73 92.3 ± 2.1 91.2 ± 1.8 72.0 ± 3.8 37.9 ± 4.5 19.0 ± 4.4 8.9 ± 1.1 m2)

Proteinuria (g/day) 0.1 ± 0.2 1.3 ± 1.3 3.2 ± 1.9 4.3 ± 1.1 3.2 ± 1.7 5.6 ± 2.2

SBP (mmHg) 129 ± 10 135 ± 12 133 ± 9 136 ± 11 136 ± 17 137 ± 11

DBP (mmHg) 74 ± 8 74 ± 9 73 ± 9 83 ± 5 77 ± 5 75 ± 8

DM duration (years) NA 1.2 ± 1.5 2.2 ± 1.8 6.5 ± 2.4 10.0 ± 5.4 8 ± 3.4

Abbreviation: ACEi; angiotensin-converting enzyme inhibitor, ARB; angiotensin II receptor blocker, Cont; healthy normal control, Cr; creatinine, DM; diabetes mellitus, DBP; diastolic blood pressure, eGFR; estimated glomerular filtration rate, NA; not available, No.; number, SBP; systolic blood pressure. Supplementary Figure 1.

The expression of adiponectin receptor 1 (AdipoR1) and 2 (AdipoR2) in stage 1 patients with or without ARB treatment Supplementary Figure 2.

A C

Low glucose (5 mM) High glucose (35 mM) 3 2.5 2.5 2 2 1.5 1.5

1 1 100 nM AdipoRon 0.5  10 nM AdipoRon 0.5 (fold) ratio 340/380

340/380 ratio (fold) ratio 340/380

0 0 1 4 7 1013 1619 2225 2831 3437 4043 4649 52 1 4 7 101316192225283134374043464952 Time interval (30 sec) Time interval (30 sec)

B D

# # # #

** # # #

ratio)

level

380 380

Peak Peak

340

(

Area curve under Area

LG LG LG HG HG HG LG LG LG HG HG HG +10 +100 +10 +100 +10 + 100 +10 +100 E LG Cont LG AdipoRon HG Cont HG AdipoRon Negative p-AMPK/-LKB1 DAPI Phospho- Ser431 LKB1 (x200)

Phospho- Ser431 LKB1 (x400)

Phospho- Thr172 AMPK (x200)

Phospho- Thr172 AMPK (x400) F

Phospho-LKB1 positive cell (x 400/HPF) Cont LG

+ AdipoRon LG

Cont HG **

+ AdipoRon HG

Phospho-AMPK positive cell (x 400/HPF) G

Cont LG

+ AdipoRon LG

Cont HG **

+ AdipoRon HG

CaMKK 66 kDa

Phospho-Ser431 LKB1 54 kDa

Total LKB1 54 kDa

Phospho-Thr172- 62 kDa AMPK

Total AMPK 62 kDa

PPAR 54 kDa

β-actin 43 kDa

D-glucose (nM/L) 5 5 5 5 30 30 30 30

AdipoRon (nM/L) - 5 10 50 - 5 10 50 I J

* 1 ratio * * * *

actin



/total LKB

(Fold)



(Fold) *

LKB * * -

ratio *

CaMKK

LG LG LG LG HG HG HG HG LG LG LG LG HG HG HG HG

5 10 50 5 10 50 Phospho 5 10 50 5 10 50 K L

* * * *

* * * * * * * ratio

actin -

* 

(Fold)

* 

(Fold) *

AMPK/total AMPK/total AMPK ratio

- PPAR LG LG LG LG HG HG HG HG LG LG LG LG HG HG HG HG

5 10 50 5 10 100 5 10 50 5 10 100 Phospho M N PGC-1 90 kDa

Phospho-Ser79 ACC 265 kDa * *

Total ACC 265 kDa

ratio(Fold) actin actin

SREBP-1c 68 kDa - * 

/ * 

1177 1

Phospho-Ser - - 140 kDa eNOS

LG LG LG LG HG HG HG HG PGC 5 10 50 5 10 100 Total eNOS 140 kDa

β-actin 43 kDa

D-glucose (nM/L) 5 5 5 5 30 30 30 30

AdipoRon (nM/L) - 5 10 50 - 5 10 50

O P Q

* * * * * * ratio * *

* * eNOS * * *

actin actin



/total

(Fold)

* - eNOS

(Fold) *

- ratio

ACC/total ACC/total ACC ratio *

- SREBP LG LG LG LG HG HG HG HG LG LG LG LG HG HG HG HG LG LG LG LG HG HG HG HG 5 10 50 5 10 100 5 10 100 Phospho

5 10 50 5 10 50 5 10 100 Phospho R LG Cont HG Cont HG AdipoRon 10 HG AdipoRon 50 Negative

DAPI DHE

DAPI TUNEL

S T

# #

(Fold)

/HPF)

positive cell cell positive

200

TUNEL

Dihydroethidium

LG HG HG HG LG HG HG HG Cont 0 10 50 Cont 0 10 50 Supplementary Figure 3.

septal

LV mass (mg) mass LV

thickness (mm) thickness

LV posterior wall posterior LV Ventricular Ventricular

db/m db/m db/db db/db db/m db/m db/db db/db db/m db/m db/db db/db Cont AdipoRon Cont AdipoRon Cont AdipoRon Cont AdipoRon Cont AdipoRon Cont AdipoRon

* *

E/A ratio E/A

shortening(sec)

db/m db/m db/db db/db Fractional shorteningFractional (%)

db/m db/m db/db db/db Velocity ofcircumferential Velocity Cont AdipoRon Cont AdipoRon db/m db/m db/db db/db Cont AdipoRon Cont AdipoRon Cont AdipoRon Cont AdipoRon

C D * area (%) area

db/m Cont db/m AdipoRon db/db Cont db/db AdipoRon

positive positive

Trichrome db/m db/m db/db db/db

Cont AdipoRon Cont AdipoRon

Trichrome

E **

Col IV

Collagen IV (Fold) CollagenIV

db/m db/m db/db db/db Cont AdipoRon Cont AdipoRon F

* (Fold)

TGF-1



TGF

db/m db/m db/db db/db Cont AdipoRon Cont AdipoRon G CTGF

CTGF(Fold)

db/m db/m db/db db/db Cont AdipoRon Cont AdipoRon I H

db/m Cont db/m AdipoRon db/db Cont db/db AdipoRon - **

)

400

positive -

TUNEL

(x400) (x cell/HPF

TUNEL

db/m db/m db/db db/db Cont AdipoRon Cont AdipoRon

J F4/80 (x400)

) **

400

cell/HPF (X cell/HPF

80 80

db/m db/m db/db db/db Cont AdipoRon Cont AdipoRon K AdipoR1 43 kDa L

AdipoR2 44 kDa

CaMKK 66 kDa ratio(Fold) Phospho- 431 54 kDa

Ser LKB1 ** /GAPDH

Total LKB1 54 kDa 1

GAPDH 37 kDa db/m db/m db/db db/db AdipoR Cont AdipoRon Cont AdipoRon

db/m db/m db/db db/db Cont AdipoRon Cont AdipoRon

M N O *

(Fold)

ratio(Fold)

/total LKB

1 (Fold) * *

LKB ** /GAPDH

* -

/GAPDH rat /GAPDH

ratio



AdipoR Phospho

db/m db/m db/db db/db db/m db/m db/db db/db db/m db/m db/db db/db CaMKK Cont AdipoRon Cont AdipoRon Cont AdipoRon Cont AdipoRon Cont AdipoRon Cont AdipoRon P

Phospho- 62 kDa Thr172 AMPK

Total AMPK 62 kDa

PPAR 54 kDa

GAPDH 37 kDa

db/m db/m db/db db/db Cont AdipoRon Cont AdipoRon

(Fold)

ratio

(Fold) * **

AMPK/total AMPK/total AMPK

ratio

/GAPDH 

db/m db/m db/db db/db PPAR

Phospho db/m db/m db/db db/db

Cont AdipoRon Cont AdipoRon Cont AdipoRon Cont AdipoRon Supplementary Figure Legends

Figure 1. The expression of adiponectin receptors in CKD stage 1 patients with or without ARB treatment. Figure 2. Effect of AdipoRon on intracellular Ca++ concentration and downstream signaling in murine podocytes cultured in low- or high-glucose medium with or without AdipoRon. Intracellular Ca++ concentration in murine podocytes cultured in low- or high-glucose medium with or without AdipoRon (A–D). Representative images of immunofluorescence staining and quantitative analyses of phosphorylated Ser431LKB1 and phosphorylated Thr172AMPK levels (E–G). Representative images of western blotting analysis of CaMKK, phosphorylated Ser431LKB1, phosphorylated Thr172AMPK, total AMPK, PPARα, and β-actin levels (H) and their quantitative analyses (I–L). Representative images of western blotting and quantitative analyses of PGC-1α, phosphorylated ACC, total ACC, SREBP-1c, phosphorylated Ser1177eNOS, total eNOS, and β-actin levels (M–Q). Representative images of immunofluorescence staining and quantitative analyses of DHE level and TUNEL- positive cells (R–T) Figure 3. Changes in the cardiac systolic and diastolic function and its phenotypes in db/m and db/db mice treated with or without AdipoRon. Echocardiographic changes as assessed by ventricular septal thickness, left ventricular (LV) wall thickness, LV mass, fractional shortening, velocity of circumferential shortening, and E/A ratio in cardiac tissues of db/m and db/db mice treated with or without AdipoRon (A and B). Representative sections stained with trichrome are shown to estimate the trichrome positive area (%) (C) together with the results of quantitative analysis according to groups (D). Immunohistochemical staining and quantitative analyses of type IV collagen- (C and E), TGF-β1- (C and F), and CTGF- positive area (C and G). Representative images of immunohistochemical staining of TUNEL- and F4/80-positive cardiac myocytes (H) and quantitative analyses of the results (I and J). Representative images of western blotting and quantitative analyses of AdipoR1, AdipoR2, CaMKK, phosphorylated Ser431LKB1, total LKB1, and GAPDH levels (K–O). Representative images of western blotting and quantitative analyses of phosphorylated AMPK Thr172, total AMPK, PPARα, and GAPDH (P-R).

SIGNIFICANCE STATEMENT

Adiponectin is an adipokine secreted by adipocytes known to exert favorable effects in the milieu of diabetic and metabolic syndrome through its anti-inﬂammatory, antiﬁbrotic,andantioxidant effects.In human andanimal diabetic nephropathy, the expression of adiponectin receptors is reduced. This manuscript reveals a favorable effect of AdipoRon, an orally active synthetic adiponectin receptor agonist, in protection, even reversal, of diabetic nephropathy produced by the activation of adiponectin receptors and downstream targets through increased intracellular Ca++/AMPK-LKB1/PPARa pathway. Adi- poRon may be a promising drug for restoration of diabetic nephropathy in type 2 diabetes.