Beyond Proteinuria: VDR Activation Reduces Renal Inflammation in Experimental Diabetic Nephropathy

Am J Physiol Renal Physiol 302: F647–F657, 2012. First published December 14, 2011; doi:10.1152/ajprenal.00090.2011. Translational Physiology

Beyond proteinuria: VDR activation reduces renal inﬂammation in experimental diabetic nephropathy

Maria-Dolores Sanchez-Niño,3 Milica Bozic,1 Elizabeth Córdoba-Lanús,4 Petya Valcheva,1 Olga Gracia,1 Merce Ibarz,2 Elvira Fernandez,1 Juan F. Navarro-Gonzalez,4* Alberto Ortiz,3* and Jose Manuel Valdivielso1* 1Research Laboratory and Nephrology Department, 2Biochemistry Department, Hospital Universitari Arnau de Vilanova, IRBLLEIDA, Lleida; 3Nefrología, IIS-Fundación Jiménez Díaz, Universidad Autonoma de Madrid and Instituto Reina Soﬁa de Investigaciones Nefrológicas-IRSIN, Madrid; and 4Servicio de Nefrología y Unidad de Investigación, Hospital Universitario Nuestra Señora de Candelaria, Santa Cruz de Tenerife, Spain Submitted 8 February 2011; accepted in ﬁnal form 7 December 2011

Sanchez-Niño M, Bozic M, Córdoba-Lanús E, Valcheva P, eventual decrease in GFR. Damage to highly differentiated Gracia O, Ibarz M, Fernandez E, Navarro-Gonzalez JF, Ortiz A, glomerular podocytes is thought to be an early event in DN Valdivielso JM. Beyond proteinuria: VDR activation reduces renal (35). In recent years, a direct association between inflammatory inflammation in experimental diabetic nephropathy. Am J Physiol parameters and clinical markers of glomerular as well as Renal Physiol 302: F647–F657, 2012. First published December 14, tubulointerstitial damage has been demonstrated, suggesting 2011; doi:10.1152/ajprenal.00090.2011.—Local inflammation is thought to contribute to the progression of diabetic nephropathy. The vitamin D that inflammation may be a pathogenic factor for the development and progression of DN (22). Hyperglycemia, angiotensin receptor (VDR) activator paricalcitol has an antiproteinuric effect in ␤ human diabetic nephropathy at high doses. We have explored poten- II, transforming growth factor (TGF)- 1, and proteinuria itself tial anti-inflammatory effects of VDR activator doses that do not all play roles in stimulating renal inflammation and/or fibrosis modulate proteinuria in an experimental model of diabetic nephrop- and contribute to progression of DN. athy to gain insights into potential benefits of VDR activators in those Multiple factors have been implicated in the progression of patients whose proteinuria is not decreased by this therapy. The effect DN, but inflammatory cytokines may be critical in the devel- of calcitriol and paricalcitol on renal function, albuminuria, and renal opment of diabetic complications and nephropathy. Inflamma- inflammation was explored in a rat experimental model of diabetes tory markers such as IL-6, IL-18, monocyte chemoattractant induced by streptozotocin. Modulation of the expression of mediators protein-1 (MCP-1), and TNF-␣ are increased in the serum and of inflammation by these drugs was explored in cultured podocytes. urine of patients with diabetes and DN (22). Furthermore, At the doses used, neither calcitriol nor paricalcitol significantly urinary TNF-␣ is directly related to UAE, and strategies modified renal function or reduced albuminuria in experimental dia- ␣ betes. However, both drugs reduced the total kidney mRNA expres- modulating TNF- activity are associated with antialbuminuric sion of IL-6, monocyte chemoattractant protein (MCP)-1, and IL-18. effects. This increase in inflammatory parameters occurs early Immunohistochemistry showed that calcitriol and paricalcitol reduced in the disease, correlates with the degree of albuminuria, and MCP-1 and IL-6 in podocytes and tubular cells as well as glomerular even precedes the increase in UAE (18, 37). Resident cells, infiltration by macrophages, glomerular cell NF-␬B activation, apo- such as mesangial, tubular epithelial, and podocytes, can pro- ptosis, and extracellular matrix deposition. In cultured podocytes, duce cytokines and can express molecules that are part of the paricalcitol and calcitriol at concentrations in the physiological and costimulatory pathway (30). Experimental animal models have clinically significant range prevented the increase in MCP-1, IL-6, recently provided evidence that inflammatory molecules, in- renin, and fibronectin mRNA expression and the secretion of MCP-1 cluding TNF-␣ and MCP-1, may have a causative role in the to the culture media induced by high glucose. In conclusion, in development of DN (22). experimental diabetic nephropathy VDR activation has local renal anti-inflammatory effects that can be observed even when proteinuria The vitamin D receptor (VDR) is a ligand-activated tran- is not decreased. This may be ascribed to decreased inflammatory scription factor which, after activation, recruits cofactor mol- responses of intrinsic renal cells, including podocytes, to high glucose. ecules and binds to specific DNA binding sites to modify the expression of target genes (34). There is evidence that VDR is glomerular disease; interleukins; microalbuminuria a modulator of glomerular injury. Calcitriol, the endogenous VDR ligand, decreases the glomerulosclerosis index and UAE in rats with subtotal nephrectomy (13, 31), and mice lacking DIABETIC NEPHROPATHY (DN) is the leading cause of end-stage renal disease in the Western world. Diabetes is associated with the VDR are more susceptible to hyperglycemia-induced renal an initial increase in glomerular filtration rate (GFR). DN injury (39). The combination of a VDR activator and an ACE progression is characterized by the development of microalbu- inhibitor protected mice from developing DN (41). In the minuria in patients with normal renal function, followed by a clinical setting, chronic kidney disease (CKD) patients treated progressive increase in urinary albumin excretion (UAE) and with paricalcitol (a selective VDR activator) showed a significant reduction of proteinuria (measured by a semiquantitative dipstick method) after 23 wk of therapy, independently of * J. F. Navarro-Gonzalez, A. Ortiz, and J. M. Valdivielso share senior GFR, blood pressure, or ACE inhibition (1). However, in authorship. clinical trials the proteinuria response to VDR activators has Address for reprint requests and other correspondence: J. M. Valdivielso, Laboratorio de Investigación IRBLLEIDA, Hospital Universitari Arnau de not been uniform, raising the question of whether patients Vilanova, Rovira Roure 80, 25198 Lleida, Spain (e-mail: Valdivielso without reduced UAE might still be benefiting from VDR @medicina.udl.es). activator therapy. Thus, in Fishbane’s trial which included 50% http://www.ajprenal.org 1931-857X/12 Copyright © 2012 the American Physiological Society F647 Translational Physiology

F648 VDR ACTIVATION IN DIABETIC NEPHROPATHY Table 1. Weight and biochemical parameters in diabetic nephropathy animal model at the end of the 4-mo follow-up

Sham Diabetes DiabetesϩCalcitriol DiabetesϩParicalcitol Weight, g 393 Ϯ 10 345 Ϯ 10* 343 Ϯ 7* 343 Ϯ 13* Serum calcium, mg/dl 10.53 Ϯ 0.1 10.3 Ϯ 0.2 10.6 Ϯ 0.3 10.5 Ϯ 0.2 Serum phosphate, mg/dl 7.4 Ϯ 0.4 7.1 Ϯ 0.5 6.3 Ϯ 0.5 6.5 Ϯ 0.6 Serum creatinine, mg/dl 0.42 Ϯ 0.01 0.43 Ϯ 0.02 0.45 Ϯ 0.04 0.44 Ϯ 0.02 Serum glucose, mg/dl 91.4 Ϯ 2.08 490 Ϯ 5* 472 Ϯ 10* 505 Ϯ 9* Urine volume, ml/day 9.7 Ϯ 1 152 Ϯ 14* 148 Ϯ 19* 132 Ϯ 15* Urinary albumin/creatinine ratio, mg/mg 0.073 Ϯ 0.019 8.3 Ϯ 1.7* 6.7 Ϯ 1.6* 7.1 Ϯ 1.2* Creatinine clearance, ml/min 2.0 Ϯ 0.14 2.0 Ϯ 0.2 1.9 Ϯ 0.1 1.8 Ϯ 0.1 Values are means Ϯ SE. *P Ͻ 0.05 vs. Sham. diabetic subjects, paricalcitol decreased UAE Ͼ10% in only However, despite podocytes being key cells in proteinuria and 58% of patients (vs. 30% in the placebo group) (11). In the DN, potential direct VDR modulation of high glucose-induced VITAL trial of DN patients, differences in the primary out- inflammation in cultured podocytes had not being studied. come (% change in urinary albumin/creatinine ratio in the Thus the aim of our study was to investigate the regulation combined paricalcitol group vs. placebo) just fell short of of renal inflammatory parameters in experimental DN by statistical significance (8). However, even in patients without subantiproteinuric doses of VDR activators and the modulation reduced proteinuria, VDR activators might have beneficial by VDR activators of the inflammatory response elicited in effects. In this regard, another potential beneficial effect of cultured murine podocytes by high glucose. VDR activation is an anti-inflammatory action. In animal models of primary glomerular disease, VDR activators reduces METHODS glomerular infiltration of inflammatory cells (23). Consistently, higher serum vitamin D levels in CKD patients are associated Animal study. Thirty male Sprague-Dawley rats were rendered with decreased systemic inflammation (33). However, no ex- diabetic by intravenous administration of streptozotocin (65 mg/kg) perimental diabetes animal study had previously addressed (7). Experimental methods used with laboratory animals comply with Law 5/1995 of June 21 by “Generalitat de Catalunya” of protection of whether VDR activators may have beneficial effects on renal animals used for experimentation and other scientific finalities and the inflammation when proteinuria is not significantly reduced. In Royal Decree 1201/2005 of October 10 on the protection of animals tubular and mesangial cells, VDR activation inhibits NF-␬B, a used for experimentation and other scientific finalities. Moreover, the key transcription factor in inflammation that regulates cyto- present protocol was approved by the Ethic Animal Experimentation kine, chemokine, and adhesion molecule gene expression (4). Committee of the University of Lleida.

Fig. 1. Vitamin D receptor (VDR) activation with paricalcitol or calcitriol downregulates total kidney expression of inﬂammatory mediator mRNA in experimental diabetic nephropathy (DN), quantiﬁed by qRT-PCR. Values are means Ϯ SE. MCP-1, monocyte chemoattractant protein-1. *P Ͻ 0.05 vs. control. #P Ͻ 0.01 vs. diabetic vehicle. ϩP Ͻ 0.05 vs. diabetic vehicle.

AJP-Renal Physiol • doi:10.1152/ajprenal.00090.2011 • www.ajprenal.org Translational Physiology

VDR ACTIVATION IN DIABETIC NEPHROPATHY F649 After 48 h, the presence of hyperglycemia was confirmed and nondiabetic controls. At the end of the treatment period, animals were animals were randomized to different treatment groups: 10 animals placed in metabolic cages and a 24-h urine sample was collected. were maintained as diabetic controls and received 1 IU/day insulin Then, rats were euthanized after isoflurane anesthesia and a terminal subcutaneously and intraperitoneal (ip) injections of vehicle; 10 ani- blood sample was collected. mals received insulin and 250 ng/kg calcitriol three times a week ip; Kidneys were perfused with sterile isotonic saline and divided into two and 10 animals received insulin and 750 ng/kg paricalcitol three times parts. One part was fixed in 4% formaldehyde and embedded in paraffin, a week ip for 4 mo. Ten vehicle-injected animals were used as and the second part was kept in RNAlater and used to extract RNA.

Fig. 2. VDR activation with paricalcitol or calcitriol downregulates the glomerular expression of IL-6 protein in experimental DN. A: representative images by immunohistochemstry. Arrows point to stained glomerular cells with podocyte morphology. Original magnification ϫ200. B: quantification of glomerular IL-6 expression by immunohistochemistry. Values are means Ϯ SE. *P Ͻ 0.0001 vs. control. #P Ͻ 0.0001 vs. diabetic vehicle. C: colocalization of IL-6 and WT-1 staining identifies podocytes as sources of glomerular IL-6 (arrows).

AJP-Renal Physiol • doi:10.1152/ajprenal.00090.2011 • www.ajprenal.org Translational Physiology

F650 VDR ACTIVATION IN DIABETIC NEPHROPATHY Analytic determinations. Serum and urine biochemistries were Cells and reagents. Conditionally immortalized mouse podocytes analyzed using a multichannel autoanalyzer (Roche/Hitachi Modular were a kind gift by Peter Mundel (19). Podocytes were propagated on Analytics). The method used for calcium analysis was the ocresol- type I collagen (Biochrom, Berlin, Germany) at 33°C in the presence phthalein complexone method, for phosphate the ammonium molyb- of 10 U/ml mouse recombinant IFN-␥ (permissive conditions, Immu- date method, for creatinine the Jaffé reaction, and for microalbumin- genex, Los Angeles, CA) to enhance expression of a thermosensitive uria a turbidometric method. Blood glucose levels were tested with an T antigen. Once cells had reached 70–80% conﬂuence, differentiation Accu Check meter (Roche Diagnostics). and a quiescent phenotype were induced by culturing under nonper-

Fig. 3. VDR activation with paricalcitol or calcitriol downregulates the glomerular expression of MCP-1 protein in experimental DN. A: representative images by immunohistochemstry. Original magnification ϫ200. B: quantification of glomerular MCP-1 expression by immunohistochemistry. Values are means Ϯ SE. *P Ͻ 0.0001 vs. control diabetic vehicle. #P Ͻ 0.0001 vs. diabetic vehicle. C: colocalization of MCP-1 and WT-1 staining identifies podocytes as sources of glomerular MCP-1 (arrows).

AJP-Renal Physiol • doi:10.1152/ajprenal.00090.2011 • www.ajprenal.org Translational Physiology

VDR ACTIVATION IN DIABETIC NEPHROPATHY F651 missive conditions at 37°C without INF-␥ for Ͼ12 days, resulting in ELISA. Cells were stimulated with high glucose, paricalcitol, or disappearance of the T antigen. Culture medium was DMEM con- calcitriol for 6 h. The concentration of murine MCP-1 in the cell taining 5 mM glucose (GIBCO, Grand Island, NY), 10% heat- culture supernatants was determined by ELISA (BD Pharmingen, San inactivated FBS, 2 mM glutamine, 100 U/ml penicillin, and 100 Diego, CA). ␮g/ml streptomycin. Cells were cultured in serum-free media 24 h RNA extraction and real-time RT-PCR. Total RNA was extracted before the addition of the stimuli and throughout the experiment. from kidneys and cells by the TRIzol method (Invitrogen, Paisley, Paricalcitol (Abbott) or calcitriol (10Ϫ10 M, Sigma, St. Louis, MO) UK). One microgram RNA was reverse transcribed with a High was added 1 h before high glucose (DMEM containing 25 mM Capacity cDNA Archive Kit (Applied Biosystems, Foster City, CA) glucose) and murine TNF-␣ (30 ng/ml, Immugenex). This concentra- for 15 min at 25°C and2hat37°C. Real-time PCR reactions were tion of paricalcitol and calcitriol is close to the range of concentrations performed with the ABI Prism 7500 sequence detection PCR system reached with therapeutic administration of paricalcitol or physiolog- (Applied Biosystems) according to the manufacturer’s protocol using ical levels of calcitriol in humans (8). the ⌬⌬Ct method. Expression levels are given as ratios to GAPDH.

Fig. 4. VDR activation with paricalcitol or calcitriol reduces the number of glomerular infiltrating CD43- positive leukocytes in experimental DN. A: representative images by immunohistochemstry. Arrows point to glomerular CD43-positive cells. Original magnification ϫ200. B: quantification of glomerular CD43 staining by immunohistochemistry. Values are means Ϯ SE. *P Ͻ 0.0001 vs. control. #P Ͻ 0.0001 vs. diabetic vehicle.

AJP-Renal Physiol • doi:10.1152/ajprenal.00090.2011 • www.ajprenal.org Translational Physiology

F652 VDR ACTIVATION IN DIABETIC NEPHROPATHY Predeveloped primer and probe assays (PDAR) were obtained for with a nonspecific immunoglobulin of the same isotype as the primary GAPDH, MCP-1, IL-18, IL-6, renin, fibronectin, and TNF-␣ from antibody. MCP-1 and IL-6 staining was evaluated by a quantitative Applied Biosystems. The relative RNA amount was calculated by scoring system using Image-Pro Plus software (Media Cybernetics, standard formulas (24). Bethesda, MD ) in 20 randomly selected glomeruli/kidney. Anti- Immunohistochemistry. Immunohistochemistry was carried out in CD43 stains leukocytes (12). The total number of anti-CD43-stained paraffin-embedded tissue sections 5 ␮m thick (27). Primary antibodies cells was counted in 20 randomly glomeruli using Image-Pro Plus were rabbit polyclonal anti-RelA, goat polyclonal anti-MCP-1, anti- software. Samples from each animal were examined in a blinded IL6 or anti-WT1 (1:75, Santa Cruz Biotechnology) and mouse mono- manner. clonal anti-CD43 (1/100, Pharmingen). In brief, for costaining with Apoptotic cells were stained using terminal deoxynucleotidyl- anti-MCP-1 or anti-IL-6 antibodies and anti-WT1, slides were first transferase-mediated dUTP nick-end labeling (TUNEL; In Situ Cell incubated with anti-IL-6 or anti-MCP-1 and then with the respective Death Detection Kit; Promega) according to the manufacturer’s in- biotin-conjugated secondary antibody. After developing of the reac- structions and as previously described (27). tion with ABComplex, sections were incubated with anti-WT1 and Glomerular fibrosis was stained with Sirius red. Tissue sections then with a FITC secondary antibody. Sections were counterstained were deparaffinized and stained for 30 min in direct Red 80 (Sigma) with Carazzi’s hematoxylin. Negative controls included incubation dissolved in Picro-sirius acid and then dehydrated. The relative

Fig. 5. VDR activator effects on kidney NF-␬B activation and renin expression in experimental DN. A: VDR activation with paricalcitol or calcitriol reduces the number of glomerular cells with nuclear p65/RelA staining, an indi- cation of NF-␬B activation. Representative image is shown of NF-␬B p65/RelA immunohistochemistry in diabetic rats where nuclear RelA staining is observed in a glomerulus from a control diabetic rat. Original magniﬁ- cation ϫ200. B: VDR activation with paricalcitol or calcitriol tended to decrease kidney renin mRNA expression quantiﬁed by qRT-PCR. Values are means Ϯ SE.

AJP-Renal Physiol • doi:10.1152/ajprenal.00090.2011 • www.ajprenal.org Translational Physiology

VDR ACTIVATION IN DIABETIC NEPHROPATHY F653 fibrotic area, expressed as a percentage of glomerular area, was Despite the nonsignificant effect on UAE, calcitriol and pari- assessed by ImageProPlus software (Media Cybernetics), analyzing calcitol reduced glomerular IL-6 and MCP-1 protein expres- 20 glomeruli/kidney (2, 25). sion (Figs. 2 and 3). Statistics. Statistical analysis was performed using SPSS 11.0 DN was associated with increased glomerular infiltration by statistical software. Results are expressed as means Ϯ SE. Signifi- cance at the P Ͻ 0.05 level was assessed by Student’s t-test for two CD43-positive leukocytes (Fig. 4). Consistent with their effect groups of data and ANOVA for three of more groups, followed by a on chemotactic factors, calcitriol and paricalcitol also reduced Scheffé post hoc test. glomerular infiltration by CD43-positive leukocytes in DN (Fig. 4). RESULTS VDR activation may decrease the activation of NF-␬B (4) and may also decrease renin expression (14), and both effects Subantiproteinuric doses of calcitriol or paricalcitol de- may result in decreased inflammation. Decreased renal glomer- crease glomerular inflammation in experimental DN. Clinical ular inflammation was associated with decreased nuclear RelA, trials of paricalcitol in human DN have shown that paricalcitol a key component of the canonical NF-␬B activation pathway has a modest and variable effect on proteinuria at doses that do that promotes chemokine gene transcription (28, 29), in glo- not promote hypercalcemia or hyperphosphatemia (8). For our merular cells in rats treated with either calcitriol or paricalcitol studies, we chose doses of paricalcitol and calcitriol that did A not influence serum levels of calcium or phosphate (Table 1). (Fig. 5 ). In addition, a trend toward decreased kidney renin In addition, VDR activators did not modify glycemia or cre- mRNA expression was observed in VDR activator-treated DN atinine clearance. Finally, there were no significant differences rats (Fig. 5B). in UAE between animals receiving VDR activators and dia- Beyond inhibiting features of glomerular inflammation, betic controls. Thus we have developed an animal model that some key histological parameters of glomerular injury were allows exploring renal effects of VDR activators in DN in the improved by VDR activator in diabetic rats. Thus the presence absence of significant changes in proteinuria. DN was associ- of glomerular apoptotic cells, as assessed by TUNEL staining, ated with increased whole kidney mRNA expression of several was only elevated in nontreated DN rats (Fig. 6). An increased cytokines. Despite the nonsignificant effect on UAE, both extracellular matrix deposition is a key feature of DN. Indeed, calcitriol and paricalcitol reduced whole kidney expression of control DN rats displayed increased glomerular extracellular inflammatory cytokine mRNAs in DN (Fig. 1). matrix deposition as assessed by Sirius red staining (Fig. 7). At the protein level, increased expression of IL-6 and This was improved by both VDR activators (Fig. 7). MCP-1 protein was noted in DN glomeruli by immunohisto- Calcitriol and paricalcitol reduce expression of inflamma- chemistry (Figs. 2 and 3). Podocytes were among glomerular tory mediators fibronectin and renin in cultured podocytes. A cells expressing inflammatory cytokines as evidenced by high glucose concentration is the main initiator of DN. A high costaining with the podocyte marker WT-1 (Figs. 2C and 3C). glucose milieu increased the expression of IL-6 and MCP-1

Fig. 6. VDR activation with paricalcitol or calcitriol reduces the number of glomerular terminal deoxynucleotidyl-transferase-mediated dUTP nick-end labeling (TUNEL)-positive cells in experimental DN. Apoptotic nuclei were stained by TUNEL. The number of TUNEL-positive cells (arrow) was higher in control DN rats than in animals treated with VDR activators. Representative images are shown. Origi- nal magniﬁcation ϫ40.

AJP-Renal Physiol • doi:10.1152/ajprenal.00090.2011 • www.ajprenal.org Translational Physiology

F654 VDR ACTIVATION IN DIABETIC NEPHROPATHY

Fig. 7. VDR activation with paricalcitol or calcitriol reduces glomerular extracellular matrix deposition in experimental DN. Sirius red staining was used. A: representative images. Magniﬁcation ϫ40. B: the positive glomerular Sirius red-stained area was higher in control DN rats than in animals treated with VDR activators. Values are means Ϯ SE. *P Ͻ 0.05 vs. control. #P Ͻ 0.05 vs. diabetic vehicle.

mRNA in cultured podocytes (Fig. 8A). Podocytes are key cells VDR activators also prevented the increase in fibronectin in DN and in the genesis of proteinuria, and they express the mRNA expression induced by high glucose (Fig. 8C), suggest- VDR in culture and in vivo (15, 26). Despite this, the effect of ing that they may also exert a direct effect on extracellular VDR activation on the podocyte response to a diabetic milieu matrix production by podocytes as has been reported in the had not been studied at the cell culture level. Both paricalcitol context of Fabry disease (26). In addition, VDR activators also and calcitriol at concentrations relevant for the clinic prevented prevented the increase in renin expression induced by high the increase in IL-6 and MCP-1 mRNA expression in podo- glucose in cultured podocytes (Fig. 8D). cytes exposed to high glucose. Furthermore, both paricalcitol and calcitriol prevented the increase in MCP-1 protein released DISCUSSION to the supernatant by podocytes exposed to high glucose (Fig. 8B). Thus both VDR activators, at the same concentration, had The main observation of this study is that both calcitriol and a protective effect on podocytes, preventing the inflammatory paricalcitol, at doses that do not significantly modify calcemia, response to metabolic (high glucose) stress. phosphatemia, or urinary albumin excretion, have an anti-

AJP-Renal Physiol • doi:10.1152/ajprenal.00090.2011 • www.ajprenal.org Translational Physiology

VDR ACTIVATION IN DIABETIC NEPHROPATHY F655

Fig. 8. VDR activation modulates inﬂamma- tory mediator induction by high glucose in cultured podocytes. Paricalcitol (10Ϫ10 M) or calcitriol (10Ϫ10 M) prevents the upregula- tion of MCP-1 or Il-6 mRNA induced by high glucose (A) as well as the amount of MCP-1 released to the supernatant in the presence of high glucose (B) in murine cultured podocytes. Paricalcitol or calcitriol also prevents the increase in ﬁbronectin mRNA (C) and in renin mRNA (D) induced by high glucose. Values are means Ϯ SE of 4 inde- pendent experiments. *P Ͻ 0.05 vs. control (normal glucose). #P Ͻ 0.05 vs. high glucose vehicle.

inflammatory action on DN glomeruli and in cultured podo- in line with recent clinical trial results suggesting dissociation cytes exposed to high glucose. between an antialbuminuric effect and renal outcomes, at least Both calcitriol and paricalcitol have been previously studied in certain patients (16). Second, in clinical trials Ͼ40% of in the context of experimental kidney disease. In 1993 cal- patients did not benefit from paricalcitol regarding proteinuria citriol was reported to lower proteinuria in Heymann’s active (11). Indeed, high paricalcitol doses (those more likely to cause nephritis, at the expense of promoting hypercalcemia (3). More adverse effects on serum calcium and phosphate) were needed recently, both paricalcitol and calcitriol were shown to lower to decrease albuminuria in human DN. Recently reported proteinuria in the subtotal nephrectomy model of rat protein- results from the VITAL randomized controlled trial of parical- uric kidney disease (17). The involvement of VDR activation citol vs. placebo in proteinuric DN showed an antiproteinuric in DN has also been demonstrated. Diabetic VDR knockout effect that was statistically significant only for the higher mice developed more severe albuminuria and glomeruloscle- paricalcitol dose used (8). rosis with podocyte effacement than their wild-type counter- Both bloodborne cells (mainly monocytes and macrophages) parts (39). Paricalcitol reduced albuminuria when associated and intrinsic renal cells synthesize inflammatory cytokines in a with losartan in mice with streptozocin-induced diabetes, but diabetic context (10). The renal cortical mRNA expression of the effect of paricalcitol alone was moderate, not indicated to IL-6 and TNF is increased in rat DN (20, 21). A high glucose be statistically significant, and kidney inflammation or cal- milieu, albuminuria, and inflammatory factors released in re- citriol was not explored (41). Paricalcitol inhibited the induc- sponse to albuminuria or high glucose may all contribute to tion of kidney MCP-1, glomerulosclerosis, and glomerular further inflammation by promoting the secretion of chemokines macrophage infiltration in streptozotocin-diabetic mice (40). (22). Glomerular cells, including podocytes, have VDRs (15, However, proteinuria was not evaluated and the in vivo effects 26). Calcitriol inhibited high glucose-induced fibronectin pro- of calcitriol were not assessed. More recently, paricalcitol duction in cultured mesangial cells and increased nephrin reduced albuminuria and kidney inflammation in db/db dia- expression in cultured podocytes (39). Calcitriol also sup- betic mice (6). Thus none of the previous studies had addressed pressed high glucose-induced activation of the renin-angioten- whether VDR activation by either paricalcitol or calcitriol sin system (RAS) and TGF-␤ in mesangial and juxtaglomeru- might decrease renal inflammation even when proteinuria was lar cells (39). Furthermore, in cultured mesangial cells, cal- not significantly reduced. In the present study, we used doses citriol stabilized I␬B␣, leading to an inhibition of p65 of calcitriol or paricalcitol that did not significantly decrease translocation to the nucleus and subsequent reduction of UAE in diabetic rats. The present animal model allowed us to NF-␬B DNA binding (40). In tubular cells, paricalcitol pre- study additional effects of VDR activation, beyond proteinuria. vented RelA binding to DNA, but not nuclear translocation The anti-inflammatory action observed beyond proteinuria (32). However, a direct anti-inflammatory action of VDR may be clinically significant for two reasons. First, in human activation on podocytes in the context of a diabetic milieu had DN, the urinary presence of inflammatory cytokines differen- not been previously observed. We now show that VDR acti- tiated microalbuminuric patients with progressive vs. nonpro- vation protects cultured podocytes from the direct proinflam- gressive renal injury, suggesting both the importance of renal matory and profibrotic effects of hyperglycemia. In the present inflammation for kidney injury and the clinical dissociation model, an anti-inflammatory and antifibrotic effect of VDR between UAE and inflammation in some patients (36). This is activation was observed in vivo despite the lack of significant

AJP-Renal Physiol • doi:10.1152/ajprenal.00090.2011 • www.ajprenal.org Translational Physiology

F656 VDR ACTIVATION IN DIABETIC NEPHROPATHY reduction in albuminuria, and, based on cell culture results, we analyzed data; M.D.S.-N., M.B., A.O., and J.M.V. prepared figures; M.B., hypothesize that it can be ascribed to a lower inflammatory E.F., J.F.N.-G., A.O., and J.M.V. interpreted results of experiments; J.F.N.-G., A.O., and J.M.V. drafted manuscript; J.F.N.-G., A.O., and J.M.V. edited and response of podocytes to high glucose or inflammatory cyto- revised manuscript; J.F.N.-G., A.O., and J.M.V. approved final version of kines. Protection from high glucose-, cytokine-, or even albu- manuscript. minuria-induced inflammation might result in inhibition of inflammation-induced progression of DN by VDR activators, REFERENCES although this should be tested in clinical trials. NF-␬B (4) and renin (14) are key targets of VDR activators 1. Agarwal R, Acharya M, Tian J, Hippensteel RL, Melnick JZ, Qiu P, Williams L, Batlle D. Antiproteinuric effect of oral paricalcitol in chronic that may underlie the present observations. High glucose acti- kidney disease. Kidney Int 68: 2823–2828, 2005. vates a local RAS in podocytes by promoting transcription of 2. Bozic M, de Rooij J, Parisi E, Ortega MR, Fernandez E, Valdivielso the renin gene and increasing angiotensin II production in an JM. Glutamatergic signaling maintains the epithelial phenotype of prox- aliskiren-sensitive fashion (9). VDR activation may suppress imal tubular cells. J Am Soc Nephrol 22: 1099–1111, 2011. the RAS, renin and angiotensinogen being the main targets. 3. Branisteanu DD, Leenaerts P, Vandamme B, Bouillon R. Partial prevention of active Heymann nephritis by 1-alpha,25 dihydroxyvitamin The activated VDR negatively regulates renin biosynthesis by D-3. Clin Exp Immunol 94: 412–417, 1993. directly transrepressing renin gene transcription in different 4. Cohen-Lahav M, Shany S, Tobvin D, Chaimovitz C, Douvdevani A. cell types, including mesangial cells (38). Furthermore, VDR Vitamin D decreases NFkappaB activity by increasing IkappaBalpha activation may decrease the expression of other RAS pathway levels. Nephrol Dial Transplant 21: 889–897, 2006. genes through inhibition of NF-␬B activation, as recently 5. Deb DK, Chen YZ, Zhang ZY, Zhang Y, Szeto FL, Wong KE, Kong J, Li YC. 1,25-Dihydroxyvitamin D3 suppresses high glucose-induced shown for angiotensinogen in podocytes (5). As mentioned angiotensinogen expression in kidney cells by blocking the NF-␬B path- above, VDR activation may inhibit NF-␬B activation through way. Am J Physiol Renal Physiol 296: F1212–F1218, 2009. different mechanisms. An effect of VDR activation on NF-␬B 6. Deb DK, Sun T, Wong KE, Zhang Z, Ning G, Zhang Y, Kong J, Shi activation is supported by the negative regulatory effect of H, Chang A, Li YC. Combined vitamin D analog and AT1 receptor ␬ antagonist synergistically block the development of kidney disease in a VDR activators on the expression of the canonical NF- B model of type 2 diabetes. Kidney Int 77: 1000–1009, 2010. transcriptional targets MCP-1 and RANTES (29) in cultured 7. Deng A, Munger KA, Valdivielso JM, Satriano J, Lortie M, Blantz podocytes and the reduced evidence for p65/RelA-containing RC, Thomson SC. Increased expression of ornithine decarboxylase in complexes migrating to the nuclei of glomerular cells in vivo. distal tubules of early diabetic rat kidneys: are polyamines paracrine Furthermore, expression of renin in podocytes was also down- hypertrophic factors? Diabetes 52: 1235–1239, 2003. 8. de Zeeuw D, Agarwal R, Amdahl M, Audhya P, Coyne D, Garimella regulated by VDR activation, pointing to a possible dual effect T, Parving HH, Pritchett Y, Remuzzi G, Ritz E, Andress D. Selective reducing inflammation by decreasing both NF-␬B activation vitamin D receptor activation with paricalcitol for reduction of albumin- and renin expression. uria in patients with type 2 diabetes (VITAL study): a randomised A shortcoming of the study is that this model did not allow controlled trial. Lancet 376: 1543–1551, 2010. 9. Durvasula RV, Shankland SJ. Activation of a local renin angiotensin the evaluation of the effects of VDR activators on preservation system in podocytes by glucose. Am J Physiol Renal Physiol 294: of renal function, since renal function was not decreased in F830–F839, 2008. control DN rats. Our design, using daily administration of 10. Egido J, Gomez-Chiarri M, Ortiz A, Bustos C, Alonso J, Gomez- insulin to keep glucose levels under control, is closer to the Guerrero C, Gomez-Garre D, Lopez-Armada MJ, Plaza J, Gonzalez clinical settings, but does not allow the investigation of renal E. Role of tumor necrosis factor-alpha in the pathogenesis of glomerular diseases. Kidney Int Suppl 39: S59–S64, 1993. function deterioration beyond proteinuria. However, it did 11. Fishbane S, Chittineni H, Packman M, Dutka P, Ali N, Durie N. Oral provide a series of insights into actions on molecular mecha- paricalcitol in the treatment of patients with CKD and proteinuria: a nisms that are known to contribute to renal function deterio- randomized trial. Am J Kidney Dis 54: 647–652, 2009. ration. 12. Fukuda M, Carlsson SR. Leukosialin, a major sialoglycoprotein on human leukocytes as differentiation antigens. Med Biol 64: 335–343, In summary, our study provides insights into additional 1986. benefits (i.e., decreased inflammation) that might be expected 13. Kuhlmann A, Haas CS, Gross ML, Reulbach U, Holzinger M, from VDR activator treatment even in the absence of decreased Schwarz U, Ritz E, Amann K. 1,25-Dihydroxyvitamin D3 decreases albuminuria. These insights may help design new primary podocyte loss and podocyte hypertrophy in the subtotally nephrectomized outcome measures in future trials involving assessment of renal rat. Am J Physiol Renal Physiol 286: F526–F533, 2004. 14. Li YC, Kong J, Wei MJ, Chen ZF, Liu SQ, Cao LP. 1,25-Dihydroxyvi- inflammation. tamin D-3 is a negative endocrine regulator of the renin-angiotensin system. J Clin Invest 110: 229–238, 2002. GRANTS 15. Matsui I, Hamano T, Tomida K, Inoue K, Takabatake Y, Nagasawa Y, Kawada N, Ito T, Kawachi H, Rakugi H, Imai E, Isaka Y. Active This work was supported by FIS PS09/00447, PS09/0299, ISCIII-RETIC vitamin D and its analogue, 22-oxacalcitriol, ameliorate puromycin ami- REDinREN/RD06/0016, Comunidad de Madrid/FRACM/S-BIO0283/2006, nonucleoside-induced nephrosis in rats. Nephrol Dial Transplant 24: Sociedad Española de Nefrología, and Programa Intensificación Actividad 2354–2361, 2009. Investigadora (ISCIII/Agencia Laín-Entralgo/CM and ISCIII/Comunidad Au- 16. Messerli FH. The sudden demise of dual renin-angiotensin system block- tonoma de Canarias). J. M Valdivielso received a research grant from Abbott ade or the soft science of the surrogate end point. J Am Coll Cardiol 53: Laboratories. M.-D. Sanchez-Niño holds a Sara Borrell Fellowship. 468–470, 2009. 17. Mizobuchi M, Morrissey J, Finch JL, Martin DR, Liapis H, Akizawa DISCLOSURES T, Slatopolsky E. Combination therapy with an angiotensin-converting No conflicts of interest, financial or otherwise, are declared by the authors. enzyme inhibitor and a vitamin D analog suppresses the progression of renal insufficiency in uremic rats. J Am Soc Nephrol 18: 1796–1806, 2007. Moriwaki Y, Yamamoto T, Shibutani Y, Aoki E, Tsutsumi Z, Taka- AUTHOR CONTRIBUTIONS 18. hashi S, Okamura H, Koga M, Fukuchi M, Hada T. Elevated levels of Author contributions: M.D.S.-N., M.B., E.F., J.F.N.-G., A.O., and J.M.V. interleukin-18 and tumor necrosis factor-alpha in serum of patients with provided conception and design of research; M.D.S.-N., M.B., E.C.-L., P.V., type 2 diabetes mellitus: relationship with diabetic nephropathy. Metabo- O.G., and M.I. performed experiments; M.D.S.-N., J.F.N.-G., A.O., and J.M.V. lism 52: 605–608, 2003.

AJP-Renal Physiol • doi:10.1152/ajprenal.00090.2011 • www.ajprenal.org Translational Physiology

VDR ACTIVATION IN DIABETIC NEPHROPATHY F657

19. Mundel P, Kriz W. Cell culture of podocytes. Exp Nephrol 4: 263–266, 31. Schwarz U, Amann K, Orth SR, Simonaviciene A, Wessels S, Ritz E. 1996. Effect of 1,25(OH)2 vitamin D-3 on glomerulosclerosis in subtotally 20. Navarro JF, Milena FJ, Mora C, Leon C, Claverie F, Flores C, Garcia nephrectomized rats. Kidney Int 53: 1696–1705, 1998. J. Tumor necrosis factor-alpha gene expression in diabetic nephropathy: 32. Tan X, Wen X, Liu Y. Paricalcitol inhibits renal inflammation by relationship with urinary albumin excretion and effect of angiotensin- promoting vitamin D receptor-mediated sequestration of NF-kappaB sig- converting enzyme inhibition. Kidney Int Suppl S98–S102, 2005. naling. J Am Soc Nephrol 19: 1741–1752, 2008. 21. Navarro JF, Milena FJ, Mora C, Leon C, Garcia J. Renal pro- 33. Timms PM, Mannan N, Hitman GA, Noonan K, Mills PG, Synder- inflammatory cytokine gene expression in diabetic nephropathy: effect of combe-Court Aganna E, Price CP, Boucher BJ. Circulating MMP9, angiotensin-converting enzyme inhibition and pentoxifylline administra- vitamin D and variation in the TIMP-1 response with VDR genotype: mechanisms for inflammatory damage in chronic disorders? QJM 95: tion. Am J Nephrol 26: 562–570, 2006. 787–796, 2002. 22. Navarro-Gonzalez JF, Mora-Fernandez C. The role of inflammatory 34. Valdivielso JM. The physiology of vitamin D receptor activation. Contrib cytokines in diabetic nephropathy. J Am Soc Nephrol 19: 433–442, 2008. Nephrol 163: 206–212, 2009. 23. Panichi V, Migliori M, Taccola D, Filippi C, De Nisco L, Giovannini 35. Wolf G, Chen S, Ziyadeh FN. From the periphery of the glomerular L, Palla R, Tetta C, Camussi G. Effects of 1,25(OH)2D-3 in experimen- capillary wall toward the center of disease: podocyte injury comes of age tal mesangial proliferative nephritis in rats. Kidney Int 60: 87–95, 2001. in diabetic nephropathy. Diabetes 54: 1626–1634, 2005. 24. Parisi E, Almaden Y, Ibarz M, Panizo S, Cardus A, Rodriguez M, 36. Wolkow PP, Niewczas MA, Perkins B, Ficociello LH, Lipinski B, Fernandez E, Valdivielso JM. N-methyl-D-aspartate receptors are ex- Warram JH, Krolewski AS. Association of urinary inflammatory mark- pressed in rat parathyroid gland and regulate PTH secretion. Am J Physiol ers and renal decline in microalbuminuric type 1 diabetics. JAmSoc Renal Physiol 296: F1291–F1296, 2009. Nephrol 19: 789–797, 2008. 25. Rodriguez-Lopez AM, Flores O, Arevalo MA, Lopez-Novoa JM. 37. Wong CK, Ho AW, Tong PC, Yeung CY, Kong AP, Lun SW, Chan Tubular cell apoptosis and proliferation in the early phase of renal damage JC, Lam CW. Aberrant activation profile of cytokines and mitogen- in uninephrectomized SHR. Kidney Blood Pressure Res 25: 13–19, 2002. activated protein kinases in type 2 diabetic patients with nephropathy. Clin 26. Sanchez-Nino MD, Sanz AB, Carrasco S, Saleem MA, Mathieson PW, Exp Immunol 149: 123–131, 2007. Valdivielso JM, Ruiz-Ortega M, Egido J, Ortiz A. Globotriaosylsphin- 38. Yuan W, Pan W, Kong J, Zheng W, Szeto FL, Wong KE, Cohen R, gosine actions on human glomerular podocytes: implications for Fabry Klopot A, Zhang Z, Li YC. 1,25-dihydroxyvitamin D3 suppresses nephropathy. Nephrol Dial Transplant 26: 1797–1802, 2010. renin gene transcription by blocking the activity of the cyclic AMP 27. Sanchez-Nino MD, Sanz AB, Lorz C, Gnirke A, Rastaldi MP, Nair V, response element in the renin gene promoter. J Biol Chem 282: 29821–29830, 2007. Egido J, Ruiz-Ortega M, Kretzler M, Ortiz A. BASP1 promotes 39. Zhang Z, Sun L, Wang Y, Ning G, Minto AW, Kong J, Quigg RJ, Li apoptosis in diabetic nephropathy. J Am Soc Nephrol 21: 610–621, 2010. YC. Renoprotective role of the vitamin D receptor in diabetic nephropa- 28. Sanz AB, Justo P, Sanchez-Nino MD, Blanco-Colio LM, Winkles JA, thy. Kidney Int 73: 163–171, 2008. Kreztler M, Jakubowski A, Blanco J, Egido J, Ruiz-Ortega M, Ortiz 40. Zhang Z, Yuan W, Sun L, Szeto FL, Wong KE, Li X, Kong J, Li YC. A. The cytokine TWEAK modulates renal tubulointerstitial inflammation. 1,25-Dihydroxyvitamin D3 targeting of NF-[kappa]B suppresses high J Am Soc Nephrol 19: 695–703, 2008. glucose-induced MCP-1 expression in mesangial cells. Kidney Int 72: 29. Sanz AB, Sanchez-Nino MD, Ramos AM, Moreno JA, Santamaria B, 193–201, 2007. Ruiz-Ortega M, Egido J, Ortiz A. NF-kappa B in renal inflammation. J 41. Zhang Z, Zhang Y, Ning G, Deb DK, Kong J, Li YC. Combination Am Soc Nephrol 21: 1254–1262, 2010. therapy with AT1 blocker and vitamin D analog markedly ameliorates 30. Schena FP. Cytokine network and resident renal cells in glomerular diabetic nephropathy: blockade of compensatory renin increase. Proc Natl diseases. Nephrol Dial Transplant 14, Suppl 1: 22–26, 1999. Acad Sci USA 105: 15896–15901, 2008.

AJP-Renal Physiol • doi:10.1152/ajprenal.00090.2011 • www.ajprenal.org