Hyperglycemia and Renin-Dependent Hypertension Synergize to Model Diabetic Nephropathy

BRIEF COMMUNICATION www.jasn.org

† † † Bryan R. Conway,* Jillian Rennie,* Matthew A. Bailey, Donald R. Dunbar, † † † Jonathan R. Manning, Christopher O. Bellamy, Jeremy Hughes,* and John J. Mullins*

*MRC Centre for Inﬂammation Research and †University of Edinburgh/British Heart Foundation Centre for Cardiovascular Science, University of Edinburgh, Edinburgh, Scotland

ABSTRACT Rodent models exhibit only the earliest features of human diabetic nephropathy, in a number of hypertensive rodent which limits our ability to investigate new therapies. Hypertension is a prerequisite models.8–10 for advanced diabetic nephropathy in humans, so its rarity in typical rodent models The renin-dependent hypertensive may partly explain their resistance to nephropathy. Here, we used the Cyp1a1mRen2 (mRen-2)27 rat has been extensively rat, in which the murine renin-2 gene is incorporated under the Cytochrome P4501a1 use to model DN9; however, it is limited promoter. In this transgenic strain, administration of low-dose dietary indole-3-carbinol by the development of malignant phase induces moderate hypertension. In the absence of hypertension, streptozotocin- hypertension.11,12 To determine how induced diabetes resulted in a 14-fold increase in albuminuria but only mild changes hyperglycemia and hypertension inter- in histology and gene expression despite 28 weeks of marked hyperglycemia. In the act at a molecular level, we used the presence of induced hypertension, hyperglycemia resulted in a 500-fold increase in Cyp1a1mRen2 rat, which harbors the albuminuria, marked glomerulosclerosis and tubulointerstitial fibrosis, and induction of murine Ren2 cDNA under the control many of the same pathways that are upregulated in the tubulointerstitium in human of the cytochrome P4501a1 promoter,13 diabetic nephropathy. In conclusion, although induction of diabetes alone in rodents such that hypertension may be induced has limited utility to model human diabetic nephropathy, renin-dependent hyperten- by dietary supplementation with indole- sion and hyperglycemia synergize to recapitulate many of the clinical, histological, and 3-carbinol (I-3-C). Unlike the constitu- gene expression changes observed in humans. tive (mRen-2)27 rat, hypertension can be induced after the onset of diabetes to J Am Soc Nephrol 23: ccc–ccc, 2012. doi: 10.1681/ASN.2011060577 mimic the natural history of human DN and the I-3-C dose may be titrated to avoid malignant phase hypertension. Diabetic nephropathy (DN) is the single artery stenosis there may be no evidence Cyp1a1mRen2 rats were allocated largest cause of end stage renal failure in of nephropathy in the kidney down- into four groups: controls (n=6), strep- the Western world.1 Although the devel- stream of the stenosis, despite severe ne- tozotocin-induced diabetes (DM; n=6), opment of novel therapeutic strategies phropathy in the contralateral kidney, I-3-C–induced hypertension (HTN; for DN remains a research priority, we suggesting that transmission of systemic n=7), and combined hypertension and are constrained by the fact that current hypertension to the diabetic glomerulus rodent models replicate only the earliest is a prerequisite for the development of stages of human DN.2 One potential ex- advanced nephropathy.6,7 Received June 15, 2011. Accepted October 13, 2011. planation for the resistance of rodents to Despite the crucial role of hyperten- DN is that they tend not to develop hy- sion in the pathogenesis of DN, there is J.H. and J.J.M. contributed equally to this work. pertension, which is critical for progres- a paucity of data regarding how high Published online ahead of print. Publication date sive DN in humans. Abnormalities in BP, BP and hyperglycemia interact at a mo- available at www.jasn.org. such as loss of nocturnal dipping, occur lecular level to promote nephropathy. Correspondence: Dr. Bryan Conway, Centre for early in the course of human DN3 and Whereas it is difficult to dissect the rel- Inflammation Research, Queen’s Medical Research rigorous BP control is at least as effective ative contribution of hypertension and Institute, University of Edinburgh, 47 Little France Crescent, Room W3.06, Edinburgh EH16 4TJ, UK. as glycemic control in retarding disease diabetes in humans, rodent studies may Email: [email protected] progression.4,5 Indeed, in patients with be informative as hyperglycemia and high Copyright © 2012 by the American Society of diabetes and co-existing unilateral renal BP synergize to promote nephropathy Nephrology

J Am Soc Nephrol 23: ccc–ccc, 2012 ISSN : 1046-6673/2303-ccc 1 BRIEF COMMUNICATION www.jasn.org diabetes (DN+HTN; n=8). During the There was very mild histological in- Tubulointerstitial fibrosis (TIF) and subsequent 28 weeks, blood sugar levels jury in the DM group; however, induc- inflammation are key components in the were 20–30 mM in both diabetic groups tion of hypertension alone promoted pathogenesis of DN; indeed, the severity with no significant difference between FSGS and a nonsignificant increase in of TIF more accurately predicts progno- the DM and DM+HTN animals (Figure the glomerulosclerosis index (GSI; Fig- sis than the glomerular findings.14 The 1A). Dietary I-3-C induced an equiva- ure 2, A and B). Concurrent diabetes and absence of overt TIF in rodent models of lent increase in tail-cuff BP in both hy- hypertension significantly increased the DN compromises their ability to effec- pertensive groups compared with their GSI compared with all other groups tively model human DN. Indeed, even in nonhypertensive counterparts (Figure (Figure 2B) and resulted in the develop- the endothelial nitric oxide synthase 1B). The tail-cuff readings were consistent ment of intraglomerular fibrin caps, knockout mouse, which develops mod- with those obtained by arterial cannula- which were rarely observed with either erate hypertension and significant glo- tion under terminal anaesthesia (mean DM or HTN alone but are typical of hu- merular pathology and is arguably the arterial pressure of 12762.3, 13662.8, man DN (Figure 2, A and C). Impor- most convincing model of DN to date, 18166.4, and 16968.7 mmHg in con- tantly, there was no histological evidence there is scant evidence of TIF.15,16 As trols, DM, HTN, and DM+HTN animals, of malignant phase hypertension, such anticipated, there was no evidence of respectively). as onion-skinning of the renal arterioles TIF after induction of diabetes alone; DM animals exhibited a modest in- in either hypertensive group. Only the however, overt TIF developed in the crease in albuminuria, with a 14-fold DM+HTN rats had a significant increase DM+HTN animals as indicated by a sig- higher median albumin/creatinine ratio in mesangial cell activation as indicated nificant increase in collagen deposition than that of controls at 28 weeks, equiv- by a-smooth muscle actin (a-SMA) (Figure 3, A and C) and myofibroblast alent to microalbuminuric levels in staining (Figure 2D). There was an in- activation (Figure 3, B and D). The in- humans (Figure 1C). Hypertension and crease in glomerular macrophage infil- nate immune system plays a major role diabetes synergized to promote albu- tration in all of the intervention groups, in the pathogenesis of DN,17 and al- minuria, such that by 28 weeks the me- which reached significance in the DM+ though this was not activated by diabetes dian albumin/creatinine ratio in the HTN animals (Figure 2E). Few glomer- alone, marked macrophage infiltration DM+HTN group was 500-fold higher ular lymphocytes were observed with no was observed in the tubulointerstitium than controls and significantly greater significant differences between the of both hypertensive groups (Figure than that in either the DM or HTN groups. groups. 3E). The role of the adaptive immune system in DN is less well characterized; however, tubulointerstitial T cell and B cell infiltration is observed in human DN18 and T cells may be pathogenic in rodent DN.19 There was an increase in tubulointerstitial T lymphocytes in the hypertensive animals, which was not ev- ident with diabetes alone (Figure 3F). In addition, focal B cell aggregates were observed solely in the DM+HTN group, often adjacent to blood vessels (Figure 3G). Lymphocyte recruitment may be mediated by the increase in chemokines and chemokine receptors observed predominantly in the DM+HTN group (Supplemental Table 1). To determine the molecular signature of the interaction between hypertension and hyperglycemia we performed microarray analysis on whole kidney tissue (n=4 per group). Remarkably, despite prolonged severe hyperglycemia in the Figure 1. Diabetes and hypertension synergise to promote albuminuria. (A) Mean (6SD) DM group, only 8 and 15 genes were sig- fi early morning nonfasting blood sugar level, (B) mean (6SD) tail-cuff systolic BP, and (C) ni cantly upregulated and downregu- median (interquartile range) albumin/creatinine ratio in the four groups of rats over the lated (corrected P,0.01), respectively, 28-week course of the experiment. ***P,0.001 versus control; #P,0.05, ##P,0.01, and versus controls. Indeed, the vast major- ###P,0.001 versus diabetic alone; $P,0.05 and $$P,0.01 versus HTN alone. ity of genes were dysregulated only by

2 Journal of the American Society of Nephrology J Am Soc Nephrol 23: ccc–ccc,2012 www.jasn.org BRIEF COMMUNICATION

Figure 1), the antigen presenting cell- mediated regulation of the cell cycle (Sup- plemental Figure 2) and an extracellular matrix gene network (Supplemental Figure 3). It is, however, worth noting that because the microarray was performed on terminal tissue samples, many of the changes in gene expression will be sec- ondary to the presence of an inflamma- tory cell infiltrate or to modification of the intrinsic cells due to anchorage to a scarred extracellular matrix, rather than reflect the primary causal pathways of hyperglycemic and hypertensive damage. Because the glomeruli comprise a small proportion of the total renal mass, whole-kidney microarray predominantly reflects gene expression in the tubulointerstitium. Hence, to determine whether the pattern of gene expression observed in our study reflects the molecular pathophysiology of human DN, we compared the upregulated genes from each group with those preferentially expressed in the tubulointerstitium of patients with DN.20 Remarkably, none of the genes that were upregulated in human DN were induced by diabetes alone, whereas 21% and 42% were significantly upregulated in the HTN and DM+HTN groups, respectively (Supplemental Table 4). Conversely, 43% and 39% of the top 60 upregulated and downregulated genes in the DM+HTN rats were similarly dys- Figure 2. Diabetes and hypertension together promote glomerulosclerosis. (A) Repre- regulated in the tubulointerstitium in sentative images of the glomeruli from kidneys from each group. Arrowhead indicates human DN (Supplemental Tables 2 and example of a fibrin cap. Bars represent 25 mM. (B) Mean (6SEM) glomerulosclerotic index 3; http://www.nephromine.org). The in- (GSI) for each group. (C) Mean (6SEM) percentage of glomeruli from each group that ability of hyperglycemia alone to activate exhibit fibrin caps. (D) Mean (6SEM) percentage a-SMA positive area in the glomeruli for many of the pathways that promote DN each group. (E) Mean (6SEM) number of macrophages (ED-1 positive cells per glomerular in humans suggests that a “second hit” ## hilar cross-sectional area). *P,0.05, **P,0.01, and ***P,0.001 versus control; P,0.01 such as hypertension is required and ### , $ , $$$ , and P 0.001 versus DM; P 0.05 and P 0.001 versus HTN. this is readily illustrated by examining the expression of individual genes. concurrent diabetes and hypertension pathways that contain a significant Havcr1, which encodes the tubular (Figure 4). The top 60 upregulated and over-representation of genes that were injury marker kidney injury molecule-1, downregulated genes in DM+HTN ani- upregulated (Figure 4B) or downreg- is upregulated early in the course of mals compared with controls are given ulated(Figure4C)intheDM+HTN human DN.21 Although it was highly in Supplemental Tables 2 and 3. Analysis group. Examples of pathways and net- expressed in the DM+HTN animals using the Database for Annotation, Vi- works identified by GeneGo Metacore (Supplemental Table 2 and Supplemen- sualization and Integrated Discovery pathway analysis and Network Building tal Figure 4), the lack of induction by (National Institute of Allergy and In- software from the upregulated genes in diabetes alone implies minimal tubular fectious Diseases, National Institutes the DM+HTN rats include the classical injury despite 28 weeks of marked hy- ofHealth,Frederick,MD)identified complement pathway (Supplemental perglycemia. The Janus kinase-signal

J Am Soc Nephrol 23: ccc–ccc, 2012 Hypertension Exacerbates Rodent DN 3 BRIEF COMMUNICATION www.jasn.org

Figure 3. Diabetes and hypertension together promote tubulointerstitial inflammation and fibrosis. Representative images of interstitial (A) picrosirius red and (B) a-SMA staining in kidneys from each group. Bars represent 50 mM. Mean (6SEM) percentage of tubulointerstitial area in each group staining with (C) picrosirius red and (D) a-SMA. Mean (6SEM) number of (E) macrophages (ED-1+ve) or (F) T cells (CD-3+ve) per 200-power tubulointerstitial field in each group. (G) B cells (CD45RA+ve) were found solely in the DM+HTN animals in (i) aggregates and (ii) adjacent to blood vessels, in particular. Bars represent 25 mM. ***P,0.001 versus control; #P,0.05, ##P,0.05, and ###P,0.001 versus DM; $P,0.05 and $$P,0.01 versus HTN.

4 Journal of the American Society of Nephrology J Am Soc Nephrol 23: ccc–ccc,2012 www.jasn.org BRIEF COMMUNICATION

were absent such as arteriolar hyalino- sis and a signiﬁcant (.50%) decline in renal function. There was, however, a trend toward a reduction in mean (6SD) inulin clearance in both hypertensive groups (1.960.6 ml/min [n=6] and 2.660.6 ml/min [n=8] in the HTN and HTN+DM groups, respectively) compared with the control and DM animals (3.261.4 ml/min [n=6] and 3.460.9 ml/min, [n=5], respectively). At the level of BP used in this study, hypertension seems to be at least as important as hyperglycemia in mediating renal injury; this may also be the case in human DN, as emphasized by the case reports of patients with diabetes and unilateral renal artery stenosis in which the failure of transmission of systemic hypertension to the kidney prevents development of nephropathy.6,7 In conclusion, this study reafﬁrms that induction of diabetes alone in rodents is of limited utility in modelling human DN. However, superimposing moderate renin-dependent hypertension results in many of the clinical, Figure 4. Patterns of renal gene expression changes with diabetes and hypertension. (A) histological, and molecular features of Venn diagrams demonstrating number of genes upregulated or downregulated .1.5-fold human DN and is a relevant model in in the DM, HTN, or DM+HTN groups compared with control animals at a corrected which to test novel therapies and dissect P,0.01. Kegg pathways that contain an over-representation of genes that are (B) upreg- the pathogenesis of the disease. ulated and (C) downregulated in DM+HTN rats compared with controls.

CONCISE METHODS transducer and activator of transcrip- in human DN, VEGF-A expression is re- tion (JAK-STAT) pathway plays a key duced20 and VEGF-A antagonists may Animal Studies role in human DN and the absence of be detrimental and promote protein- Cyp1a1mRen2 rats were generated as de- TIF in standard rodent models of DN uria.24 Likewise in this study, whereas scribed.13 Diabetes was induced by a single may reflect a failure to activate this path- hyperglycemia alone tended to increase intravenous injection of 20 mg/kg streptozo- way.22 In keeping with this theory, mul- VEGF-A expression, the combination tocin and blood sugars were maintained in tiple JAK-STAT pathway genes, such as of hypertension and hyperglycemia the 20–30 mM range by serial subcutaneous JAK1, JAK2, and STAT1, were induced reduced VEGF-A expression (Supple- insulin implants (Research Pack, Linshin, by the combination of diabetes and hy- mental Figure 6), suggesting that super- Canada). Two weeks after onset of diabetes, pertension, but not by diabetes alone imposing hypertension on diabetes more hypertension was induced by dietary supple- (Supplemental Figure 5). closely reflects the pathophysiology of mentation of 0.125% by mass I-3-C (Sigma, Gene expression analysis from con- human DN. UK). Tail-cuff BP measurements were per- ventional rodent models of DN may not One limitation of this study is that formed twice weekly in conscious, trained be reliably informative regarding the it cannot determine whether the de- animals and 24-hour urine collections were pathogenesis of human DN. For exam- velopment of nephropathy is due to obtained every 4 weeks. After 28 weeks, ani- ple, the increase in vascular endothelial hypertension or activation of the renin- mals were anesthetized (Inactin; 120 mg/kg growth factor-A (VEGF-A) expression angiotensin-aldosterone-system per se. intraperitoneally) and inulin clearance was observed in rodent models of incipient In addition, whereas the DM+HTN rats determined as described.25 All procedures DN23 implicated VEGF-A inhibition as an mimicked many of the hallmarks of were preformed under a UK Home Office attractive therapeutic strategy. However, human DN, some cardinal features license.

J Am Soc Nephrol 23: ccc–ccc, 2012 Hypertension Exacerbates Rodent DN 5 BRIEF COMMUNICATION www.jasn.org

Albumin and Creatinine module,29 and analyzed with the Limma30 Diabetes and Digestive and Kidney Diseases, Measurements and Rank Products packages.31 Gene ontol- 2009 2. Brosius FC 3rd, Alpers CE, Bottinger EP, Albumin and creatinine measurements were ogy32 and Kyoto Encyclopedia of Genes and Breyer MD, Coffman TM, Gurley SB, Harris performed on a Cobas Fara centrifugal ana- Genomes pathway enrichment analysis was RC, Kakoki M, Kretzler M, Leiter EH, Levi M, lyzer (Roche Diagnostics Ltd, Welwyn Garden performed using the Database for Annota- McIndoe RA, Sharma K, Smithies O, Susztak K, City, UK) using a commercial immunotur- tion, Visualization and Integrated Discovery Takahashi N, Takahashi T Animal Models of bidimetric assay (Microalbumin Kit, Olym- (National Institute of Allergy and Infectious Diabetic Complications Consortium: Mouse models of diabetic nephropathy. JAmSoc pus Diagnostics Ltd, Watford, UK) and a Diseases, National Institutes of Health, Nephrol 20: 2503–2512, 2009 creatininase-based enzymatic method26 (Al- 33 Frederick, MD). Metacore pathway analysis 3. Lurbe E, Redon J, Kesani A, Pascual JM, pha Laboratories Ltd, Eastleigh, UK), respec- software (GeneGo, St. Joseph, MI) was used Tacons J, Alvarez V, Batlle D: Increase in tively. to identify functional links in the differen- nocturnal blood pressure and progression to tially expressed genes and networks were microalbuminuria in type 1 diabetes. NEngl JMed347: 797–805, 2002 Immunohistochemistry built using Metacore’s default network build- m 4. UK Prospective Diabetes Study Group: Tight Staining was performed on 4- Mmethacarn- ing algorithm. Microarray data are available blood pressure control and risk of macro- fixed, paraffin-embedded sections using in the ArrayExpress database (www.ebi.ac. vascular and microvascular complications in standard protocols with the following pri- uk/arrayexpress;accessionno.E-MEXP- type 2 diabetes: UKPDS 38. BMJ 317: 703– mary antibodies: mouse anti-rat a-SMA 3165). The relative expression levels of 713, 1998 5. Mogensen CE: Combined high blood pres- (1:500; Sigma-Aldrich, Dorset, UK), mouse selected genes from the microarray were val- sure and glucose in type 2 diabetes: Double anti-ratED-1(macrophagemarker,1:100; idated by real-time PCR using inventoried jeopardy. British trial shows clear effects of -DDCt AbD Serotec, Kidlington, UK), rabbit anti- Taqman gene expression assays and the 2 treatment, especially blood pressure re- rat CD3 (T cell marker, 1:1000; Abcam, Cam- method per the manufacturer’s instructions duction. BMJ 317: 693–694, 1998 bridge, UK), anti-rat CD45RA (B cell marker, (Applied Biosystems, Cheshire, UK). 6. Berkman J, Rifkin H: Unilateral nodular diabetic glomerulosclerosis (Kimmelstiel-Wilson): 1:500; AbD Serotec), goat anti-rat kidney Report of a case. Metabolism 22: 715–722, injury molecule-1 (1:100; R&D Systems, Statistical Analyses 1973 Abingdon, UK), or a species-specific isotype. Data are presented as mean 6 SD and me- 7. Béroniade VC, Lefebvre R, Falardeau P: Uni- Sections were also stained for periodic acid– dian (interquartile range) where the data lateral nodular diabetic glomerulosclerosis: Schiff and picrosirius red using standard pro- are normal or skewed, respectively. The groups Recurrence of an experiment of nature. Am J Nephrol 7: 55–59, 1987 tocols. The glomerulosclerosis index for each were compared by one-way ANOVA (after 8. Cooper ME, Allen TJ, Macmillan P, Bach L, animal was determined from all glomeruli in log-transformation for nonparametric data Jerums G, Doyle AE: Genetic hypertension one kidney cross-section (mean 180 glomer- where necessary). accelerates nephropathy in the streptozoto- uli/animal) as described previously.27 The cin diabetic rat. Am J Hypertens 1: 5–10, mean number of ED-1 and CD3+ve cells 1988 9. Kelly DJ, Wilkinson-Berka JL, Allen TJ, were obtained from 60 hilar glomerular ACKNOWLEDGMENTS Cooper ME, Skinner SL: A new model of cross-sections per animal and twenty 200- diabetic nephropathy with progressive renal power tubulointerstitial fields per animal. Wethank Dr. Forbes Howie, Dr. Linda Mullins, impairment in the transgenic (mRen-2)27 rat The mean glomerular and tubulointerstitial Dr.ChrisKenyon,LouiseEvans,Robert (TGR). Kidney Int 54: 343–352, 1998 area staining for a-SMA or picrosirius red Menzies, and the staff of the animal facility 10. Janssen U, Riley SG, Vassiliadou A, Floege J, Phillips AO: Hypertension superimposed was calculated from 30 hilar glomerular for useful discussions and technical support. on type II diabetes in Goto Kakizaki rats cross-sections per animal or 20 interstitial B.R.C. was supported by a Medical Re- induces progressive nephropathy. Kidney fields using Adobe Photoshop software. search Council Clinician Scientist Fellowship. Int 63: 2162–2170, 2003 D.R.D. and J.R.M. are supported by the Uni- 11. Mullins JJ, Peters J, Ganten D: Fulminant versity of Edinburgh BHF Centre for Research hypertension in transgenic rats harbouring Microarray Analysis – Excellence. the mouse Ren-2 gene. Nature 344: 541 RNAwas extracted using the Nucleospin RNA 544, 1990 II kit (Macherey-Nagel, Duren, Germany) 12. Hartner A, Cordasic N, Klanke B, Wittmann M, from snap-frozen, homogenized whole kid- Veelken R, Hilgers KF: Renal injury in strep- ney tissue from four representative animals DISCLOSURES tozotocin-diabetic Ren2-transgenic rats is mainly dependent on hypertension, not on from each group. RNA was processed using None. diabetes. Am J Physiol Renal Physiol 292: standard Affymetrix protocols, including one F820–F827, 2007 fi round of cDNA ampli cation, and was hy- 13. Kantachuvesiri S, Fleming S, Peters J, Peters bridized to the Affymetrix Rat Genome 230 REFERENCES B, Brooker G, Lammie AG, McGrath I, 2.0 GeneChip. Data were extracted by us- Kotelevtsev Y, Mullins JJ: Controlled hyper- ing GeneChip Operating Software software, 1. US Renal Data System: USRDS2009An- tension, a transgenic toggle switch reveals fi nual Data Report: Atlas of Chronic Kidney differential mechanisms underlying vascular and Celestia les were used for further data Disease and End-Stage Renal Disease in disease. J Biol Chem 276: 36727–36733, 2001 28 processing in Bioconductor, normalized the United States, Bethesda, MD, National 14. Gilbert RE, Cooper ME: The tubulointer- by Robust Multiarray Average in the Affy Institutes of Health, National Institute of stitium in progressive diabetic kidney disease:

6 Journal of the American Society of Nephrology J Am Soc Nephrol 23: ccc–ccc,2012 www.jasn.org BRIEF COMMUNICATION

More than an aftermath of glomerular injury? and N-acetyl-b-D-glucosaminidase. Kidney 28. Gentleman RC, Carey VJ, Bates DM, Bolstad Kidney Int 56: 1627–1637, 1999 Int 79: 464–470, 2011 B, Dettling M, Dudoit S, Ellis B, Gautier L, Ge 15. Zhao HJ, Wang S, Cheng H, Zhang MZ, 22. Berthier CC, Zhang H, Schin M, Henger A, Y, Gentry J, Hornik K, Hothorn T, Huber W, Takahashi T, Fogo AB, Breyer MD, Harris RC: Nelson RG, Yee B, Boucherot A, Neusser Iacus S, Irizarry R, Leisch F, Li C, Maechler M, Endothelial nitric oxide synthase deficiency MA, Cohen CD, Carter-Su C, Argetsinger LS, Rossini AJ, Sawitzki G, Smith C, Smyth G, produces accelerated nephropathy in diabetic Rastaldi MP, Brosius FC, Kretzler M: En- Tierney L, Yang JY, Zhang J: Bioconductor: mice. JAmSocNephrol17: 2664–2669, 2006 hanced expression of Janus kinase-signal Open software development for computa- 16. Kanetsuna Y, Takahashi K, Nagata M, Gannon transducer and activator of transcription tional biology and bioinformatics. Genome MA, Breyer MD, Harris RC, Takahashi T: De- pathway members in human diabetic ne- Biol 5: R80, 2004 ficiency of endothelial nitric-oxide synthase phropathy. Diabetes 58: 469–477, 2009 29. Gautier L, Cope L, Bolstad BM, Irizarry RA: confers susceptibility to diabetic nephropa- 23. Cooper ME, Vranes D, Youssef S, Stacker SA, Affy—Analysis of Affymetrix GeneChip data thy in nephropathy-resistant inbred mice. Cox AJ, Rizkalla B, Casley DJ, Bach LA, Kelly at the probe level. Bioinformatics 20: 307– Am J Pathol 170: 1473–1484, 2007 DJ, Gilbert RE: Increased renal expression of 315, 2004 17. Tesch GH: Macrophages and diabetic ne- vascular endothelial growth factor (VEGF) 30. Smyth GK: Linear models and empirical phropathy. Semin Nephrol 30: 290–301, 2010 and its receptor VEGFR-2 in experimental bayes methods for assessing differential 18. Cohen CD, Lindenmeyer MT, Eichinger F, diabetes. Diabetes 48: 2229–2239, 1999 expression in microarray experiments. Stat Hahn A, Seifert M, Moll AG, Schmid H, Kiss E, 24. Eremina V, Jefferson JA, Kowalewska J, Appl Genet Mol Biol 3: Article 3, 2004 Gröne E, Gröne HJ, Kretzler M, Werner T, Hochster H, Haas M, Weisstuch J, Richardson 31. Breitling R, Armengaud P, Amtmann A, Nelson PJ: Improved elucidation of biolog- C, Kopp JB, Kabir MG, Backx PH, Gerber HP, Herzyk P: Rank products: A simple, yet ical processes linked to diabetic nephropa- Ferrara N, Barisoni L, Alpers CE, Quaggin SE: powerful, new method to detect differen- thy by single probe-based microarray data VEGF inhibition and renal thrombotic mi- tially regulated genes in replicated micro- analysis. PLoS ONE 3: e2937, 2008 croangiopathy. NEnglJMed358: 1129– array experiments. FEBS Lett 573: 83–92, 19. Lim AK, Ma FY, Nikolic-Paterson DJ, Kitching 1136, 2008 2004 AR, Thomas MC, Tesch GH: Lymphocytes 25. Liu X, Bellamy CO, Bailey MA, Mullins LJ, 32. Ashburner M, Ball CA, Blake JA, Botstein D, promote albuminuria, but not renal dys- Dunbar DR, Kenyon CJ, Brooker G, Butler H, Cherry JM, Davis AP, Dolinski K, function or histological damage in a mouse Kantachuvesiri S, Maratou K, Ashek A, Clark Dwight SS, Eppig JT, Harris MA, Hill DP, model of diabetic renal injury. Diabetologia AF, Fleming S, Mullins JJ: Angiotensin- Issel-Tarver L, Kasarskis A, Lewis S, Matese 53: 1772–1782, 2010 converting enzyme is a modifier of hyper- JC, Richardson JE, Ringwald M, Rubin GM, 20. Lindenmeyer MT, Kretzler M, Boucherot A, tensive end organ damage. JBiolChem284: Sherlock G The Gene Ontology Consortium: Berra S, Yasuda Y, Henger A, Eichinger F, 15564–15572, 2009 Gene ontology: Tool for the unification of Gaiser S, Schmid H, Rastaldi MP, Schrier RW, 26. Keppler A, Gretz N, Schmidt R, Kloetzer HM, biology. Nat Genet 25: 25–29, 2000 Schlöndorff D, Cohen CD: Interstitial vascu- Groene HJ, Lelongt B, Meyer M, Sadick M, 33. Huang W, Sherman BT, Lempicki RA: Sys- lar rarefaction and reduced VEGF-A expres- Pill J: Plasma creatinine determination in tematic and integrative analysis of large sion in human diabetic nephropathy. JAm mice and rats: An enzymatic method com- gene lists using DAVID bioinformatics re- Soc Nephrol 18: 1765–1776, 2007 pares favorably with a high-performance sources. Nat Protoc 4: 44–57, 2009 21. Vaidya VS, Niewczas MA, Ficociello LH, liquid chromatography assay. Kidney Int 71: Johnson AC, Collings FB, Warram JH, 74–78, 2007 Krolewski AS, Bonventre JV: Regression of 27. Rodríguez-Iturbe B, Quiroz Y, Shahkarami A, microalbuminuria in type 1 diabetes is asso- Li Z, Vaziri ND: Mycophenolate mofetil ameli- This article contains supplemental material online ciated with lower levels of urinary tubular orates nephropathy in the obese Zucker rat. at http://jasn.asnjournals.org/lookup/suppl/doi:10. injury biomarkers, kidney injury molecule-1, Kidney Int 68: 1041–1047, 2005 1681/ASN.2011060577/-/DCSupplemental.

J Am Soc Nephrol 23: ccc–ccc, 2012 Hypertension Exacerbates Rodent DN 7 Supplementary data

Double jeopardy for the kidney: combinatorial hyperglycaemia and renin-dependent hypertension promotes nephropathy in rodents

1,2 B.R.Conway, 1J.Rennie, 2M.A.Bailey, 2D.R.Dunbar, 2J.R.Manning 2C.O.Bellamy,

1,3 J.Hughes, 2,3 J.J.Mullins

1MRC Centre for Inflammation Research, University of Edinburgh

2University of Edinburgh/British Heart Foundation Centre for Cardiovascular

Science, University of Edinburgh

3These authors contributed equally to this work

1 Suppl Table 1. Mean relative expression of genes encoding adaptive immune system chemokines and their receptors in the DM, HTN and DM+HTN groups compared with control animals. *p<0.05, **p<0.01, ***p<0.001

Fold increase v Control Gene Name DM HTN DM+HTN Ccl19 1.31 1.67 2.32** Ccl21 2.03 2.10* 4.66*** Ccr7 1.20 8.77 7.93*** Cxcl11 1.43 1.90 3.09** Cxcl13 3.42 1.61 4.12*** Cxcr3 1.04 1.33 1.88*** Cxcl12 1.32 -1.02 -1.28 Cxcr4 1.59 1.86 3.51*** Ccl5 1.22 1.55 1.81** Ccr5 1.02 1.26 1.48*

2 Suppl Table 2. The 60 genes most significantly up-regulated in the DM+HTN animals compared with control animals. *Genes that are also significantly up-regulated in human

DN and #genes that were not represented in the human microarray

(www.nephromine.org ).

Corrected Fold increase v Cont p-value Gene Name Description DM+HTN v DM HTN DM+HTN Cont Havcr1* hepatitis A virus cellular receptor 1 -1.06 11.88 33.16 4.08E-04 Fgb fibrinogen beta chain 1.49 5.32 17.73 7.70E-05 Fga fibrinogen alpha chain 1.29 4.84 13.14 1.49E-04 Car3 # carbonic anhydrase 3 4.55 14.08 12.59 6.37E-03 Cfi complement factor I 1.15 5.40 11.49 7.70E-05 Tnc* tenascin C 1.91 3.64 10.81 3.43E-06 Spp1 secreted phosphoprotein 1 1.16 6.11 9.59 4.18E-04 Clu* clusterin 1.74 4.47 8.66 1.92E-05 Gpnmb* glycoprotein (transmembrane) nmb 1.38 3.40 7.96 6.76E-05 serine (or cysteine) peptidase inhibitor; Serpina3n 1.03 2.27 7.92 1.08E-04 clade A; member 3N Col3a1* collagen; type III; alpha 1 1.43 3.74 7.72 2.34E-06 Ubd ubiquitin D (FAT10) 1.04 4.81 7.13 1.53E-03 protein tyrosine phosphatase; receptor- Ptprz1 1.00 3.78 6.87 1.11E-04 type; Z polypeptide 1 Fgg fibrinogen gamma chain 1.14 3.16 6.64 4.27E-04 Lamc2* laminin; gamma 2 1.60 3.11 6.47 3.37E-03 Cp* ceruloplasmin 1.26 2.70 6.34 3.99E-05 Fc fragment of IgG; low affinity IIb; Fcgr2b* 1.50 3.55 6.08 4.07E-04 receptor (CD32) Mrc1* mannose receptor; C type 1 1.76 2.96 6.07 5.68E-05 C3 complement component 3 1.61 4.02 6.04 5.95E-04 solute carrier family 34 (sodium Slc34a2 -1.68 3.02 5.78 7.50E-03 phosphate); member 2 macrophage galactose N-acetyl- Mgl1 # 1.41 4.21 5.47 2.58E-04 galactosamine specific lectin 1 cell division cycle 2; G1 to S and G2 to Cdc2 # 1.04 2.09 5.43 3.54E-05 M Lox lysyl oxidase 2.10 1.78 5.31 6.04E-05 Fn1* fibronectin 1 1.53 2.54 5.15 3.88E-06 Ccna2 cyclin A2 -1.16 1.70 4.99 1.76E-04 membrane-spanning 4-domains; Ms4a6b # 1.28 2.84 4.94 7.70E-05 subfamily A; member 6B six transmembrane epithelial antigen of Steap1 1.35 2.91 4.76 2.93E-04 the prostate 1 chemokine (C-C motif) ligand 21b Ccl21b 2.03 2.10 4.66 3.61E-05 (serine) Cdkn3 cyclin-dependent kinase inhibitor 3 -1.22 1.58 4.53 4.08E-04 Tpbg* trophoblast glycoprotein -1.37 1.85 4.46 5.64E-04 Ect2 epithelial cell transforming sequence 2 -1.46 1.69 4.44 1.26E-04

3 oncogene Cd53* Cd53 molecule (OX44) 1.44 2.76 4.41 1.23E-05 Reg3b # regenerating islet-derived 3 beta 1.07 2.18 4.36 9.07E-04 Col5a2 collagen; type V; alpha 2 1.30 2.19 4.22 1.84E-06 antigen identified by monoclonal Mki67 -1.06 1.61 4.13 2.57E-04 antibody Ki-67 Cxcl13 chemokine (C-X-C motif) ligand 13 3.42 1.61 4.12 6.65E-04 Col4a1* collagen; type IV; alpha 1 1.78 2.56 3.98 1.28E-05 Fbn1* fibrillin 1 1.25 2.20 3.98 2.04E-04 similar to hypothetical protein RGD1308694 # 2.20 3.28 3.95 2.16E-03 MGC47256 Fcrls # Fc receptor-like S; scavenger receptor 1.24 1.89 3.93 8.63E-05 Prc1 protein regulator of cytokinesis 1 1.02 1.56 3.91 5.00E-04 Cd44* Cd44 molecule -1.15 2.53 3.89 1.15E-03 Rrm2 ribonucleotide reductase M2 -1.24 1.69 3.88 2.96E-04 Gpx2 glutathione peroxidase 2 1.19 2.29 3.88 5.07E-04 NS5A (hepatitis C virus) transactivated Ns5atp9* -1.12 1.48 3.84 7.33E-05 protein 9 protein tyrosine phosphatase; receptor Ptprc* 1.30 2.58 3.83 1.87E-04 type; C Crlf1 cytokine receptor-like factor 1 1.17 2.12 3.81 2.94E-05 serine (or cysteine) peptidase inhibitor; Serpina10 clade A (alpha-1 antiproteinase; -1.19 1.71 3.78 4.00E-03 antitrypsin); member 10 Cdca3 cell division cycle associated 3 -1.06 1.33 3.77 7.74E-05 Tcf7* transcription factor 7; T-cell specific 1.14 2.30 3.76 1.11E-04 Ccnb2 cyclin B2 -1.13 1.92 3.74 2.78E-04 Col1a1* collagen; type I; alpha 1 1.15 1.76 3.73 3.09E-05 Mmp12 matrix metallopeptidase 12 1.19 2.13 3.73 2.20E-04 Timp1* TIMP metallopeptidase inhibitor 1 1.10 2.14 3.73 2.42E-03 proteasome (prosome; macropain) Psmb8 subunit; beta type 8 (large 1.06 2.16 3.71 3.99E-05 multifunctional peptidase 7) CDC28 protein kinase regulatory Cks2 1.43 1.76 3.70 9.66E-04 subunit 2 Col1a2* collagen; type I; alpha 2 1.22 2.31 3.65 5.68E-05 Glipr1 GLI pathogenesis-related 1 1.36 2.14 3.62 7.42E-06 Vim* vimentin 1.06 1.94 3.59 2.70E-05 Coro1a* coronin; actin binding protein 1A 1.24 2.19 3.56 1.26E-04

Suppl Table 3. The 60 genes most significantly down-regulated in the DM+HTN animals compared with control animals. *Genes that are also significantly downregulated in human DN and #genes that were not represented in the human microarray

(www.nephromine.org ).

Corrected Fold increase v Cont p-value Gene name Description DM HTN DM+HTN DM+HTN v Cont # Shisa3 shisa homolog 3 (Xenopus laevis) -16.64 -10.85 -19.53 1.30E-03 KLK1* Kallikrein -3.39 -2.34 -18.35 6.75E-06 Ren renin 1.49 -9.71 -9.25 7.21E-04 Klks3* kallikrein; submaxillary gland S3 -2.19 -1.77 -5.09 1.44E-04 synuclein; alpha (non A4 component of Snca* -4.33 -3.69 -4.66 1.54E-03 amyloid precursor) Lipg lipase; endothelial -3.58 -1.81 -3.88 8.50E-05 hydroxysteroid (17-beta) dehydrogenase Hsd17b1 -1.85 -1.40 -3.46 1.45E-03 1 RGD1304644 # similar to RIKEN cDNA 2310046K01 -2.94 -1.88 -3.38 3.36E-04 Nat8 N-acetyltransferase 8 -2.12 -1.89 -3.32 5.07E-03 Fem1b* feminization 1 homolog b (C. elegans) -1.81 -1.60 -3.09 2.32E-04 CCR4 carbon catabolite repression 4-like Ccrn4l -2.28 -2.03 -3.09 5.74E-03 (S. cerevisiae) similar to MIR-interacting saposin-like # protein precursor (Transmembrane LOC685001 -1.95 -1.81 -3.01 7.38E-03 protein 4) (Putative secreted protein ZSIG9) Ank2* ankyrin 2; neuronal -2.11 -1.19 -2.86 4.08E-04 Vwf von Willebrand factor homolog -2.66 -1.37 -2.81 1.69E-03 Tsku* tsukushin -1.29 -1.23 -2.81 7.92E-05 protein phosphatase 1; regulatory Ppp1r1a* -1.78 -1.50 -2.78 8.40E-04 (inhibitor) subunit 1A Nefm neurofilament; medium polypeptide -2.08 -1.20 -2.68 2.20E-04 similar to MIR-interacting saposin-like # protein precursor (Transmembrane LOC685001 -1.57 -1.62 -2.57 4.79E-03 protein 4) (Putative secreted protein ZSIG9) Dph3 # DPH3; KTI11 homolog (S. cerevisiae) -1.89 -2.38 -2.50 7.63E-03 Casr calcium-sensing receptor -1.54 -1.46 -2.50 3.47E-03 Guca2b guanylate cyclase activator 2B -1.32 -1.48 -2.33 1.55E-03 Susd3 # sushi domain containing 3 -1.64 -1.67 -2.32 5.83E-03 Neto2 neuropilin (NRP) and tolloid (TLL)-like 2 -2.71 -1.54 -2.27 1.99E-03 potassium voltage-gated channel; shaker- Kcnab1 1.06 -1.10 -2.26 3.87E-03 related subfamily; beta member 1 malic enzyme 1; NADP(+)-dependent; Me1 -1.72 -1.59 -2.24 5.79E-03 cytosolic Sult2b1* sulfotransferase family; cytosolic; 2B; -1.93 -1.48 -2.23 5.99E-03

5 member 1 5-hydroxytryptamine (serotonin) receptor Htr5b # -1.85 -1.27 -2.18 2.31E-03 5B Mat2a methionine adenosyltransferase II; alpha -2.00 -1.42 -2.12 2.04E-03 Neb nebulin -1.44 -1.11 -2.04 2.04E-03 SWI/SNF related; matrix associated; actin Smarca2 dependent regulator of chromatin; -1.59 -1.77 -2.01 5.64E-03 subfamily a; member 2 Mylk3 myosin light chain kinase 3 -1.51 -1.24 -2.00 6.20E-04 Cml5 # camello-like 5 -1.75 -1.32 -2.00 7.65E-03 RGD1310587* similar to hypothetical protein FLJ14146 -1.89 -1.29 -1.98 5.56E-03 Calcr calcitonin receptor -1.52 -1.26 -1.98 4.39E-04 Ryr1 ryanodine receptor 1; skeletal muscle -1.81 -1.72 -1.95 2.74E-03 potassium channel; subfamily T; member Kcnt1 # -1.69 -1.63 -1.95 3.80E-03 1 Gcgr* glucagon receptor -1.39 -1.52 -1.93 6.78E-04 Chst2 carbohydrate sulfotransferase 2 -1.67 -1.47 -1.91 1.14E-03 Dclk3 # doublecortin-like kinase 3 -1.49 -1.46 -1.90 3.97E-04 Tef thyrotrophic embryonic factor -1.44 -1.38 -1.88 4.83E-04 C1q and tumor necrosis factor related C1qtnf4 # -1.02 -1.18 -1.87 2.16E-03 protein 4 Tcfap2b # transcription factor AP-2 beta -1.49 -1.21 -1.84 1.25E-03 Acy3 # aspartoacylase (aminocyclase) 3 -1.33 -1.68 -1.83 5.04E-03 Tmem150 # transmembrane protein 150 -1.58 -1.38 -1.79 2.34E-03 family with sequence similarity 81; Fam81a # -1.14 -1.25 -1.79 9.07E-04 member A # solute carrier family 2 (facilitated glucose Slc2a13 -1.62 -1.31 -1.78 4.88E-04 transporter); member 13 Usp2* ubiquitin specific peptidase 2 -1.30 -1.26 -1.78 6.01E-04 Egf* epidermal growth factor -1.14 -1.15 -1.77 7.40E-04 LOC685742 # hypothetical protein LOC685742 -1.63 -1.42 -1.76 7.94E-03 Lace1 # lactation elevated 1 -1.27 -1.08 -1.76 9.75E-03 aspartate dehydrogenase domain Aspdh # containing -1.03 -1.42 -1.75 2.20E-03 Fgf9 fibroblast growth factor 9 -1.04 -1.53 -1.75 9.96E-04 Prr5* proline rich 5 (renal) -1.10 -1.25 -1.74 2.06E-04 Tcfap2b # transcription factor AP-2 beta -1.35 -1.18 -1.73 1.36E-03 Pim3 # pim-3 oncogene -1.21 -1.50 -1.73 3.04E-03 Pecr* peroxisomal trans-2-enoyl-CoA reductase -1.92 -1.27 -1.72 2.31E-03 Vill villin-like -1.25 -1.41 -1.72 2.46E-03 oxidoreductase NAD-binding domain Oxnad1 # containing 1 -1.43 -1.42 -1.71 4.11E-03 RGD1308772* similar to KIAA0564 protein -1.25 -1.55 -1.70 4.47E-03 solute carrier family 2 (facilitated glucose Slc2a4 transporter); member 4 -1.50 -1.47 -1.70 8.71E-03

Suppl Table 4. List of published genes that are up-regulated in human DN and their degree of up-regulation in the diabetic and/or hypertensive animals with respect to controls. *p<0.05, **p<0.01 and ***P<0.001 v control group.

Fold increase v Control Gene name Description DM HTN DM+HTN ADAM10 ADAM metallopeptidase domain 10 1.04 1.02 1.09 ADAM17 ADAM metallopeptidase domain 17 1.27 1.29 1.41*** ADAM9 ADAM metallopeptidase domain 9 -1.14 -1.05 1.08 aldo-keto reductase family 1, member B1 (aldose AKR1B1 reductase) 1.20 1.41 1.77** BMP1 bone morphogenetic protein 1 1.22 1.22 1.54* BMP6 bone morphogenetic protein 6 1.02 -1.02 1.00 CASP1 caspase 1 1.20 1.45* 1.89*** CCL21 chemokine (C-C motif) ligand 21 2.02 2.10* 4.66*** COL1A2 collagen, type I, alpha 2 1.26 1.96 3.36*** COL4A1 collagen, type 4, alpha 1 1.78 2.56** 3.98*** COL4A2 collagen, type 4, alpha 2 1.31 1.51* 2.18*** ELF1 E74-like factor 1 (ets domain transcription factor) -1.02 1.07 1.15 ELF2 E74-like factor 2 (ets domain transcription factor) -1.01 -1.02 1.10

ELF3 E74-like factor 3 (ets domain transcription factor) -1.04 -1.17 1.09 ELK3 ETS-domain protein (SRF accessory protein 2) 1.19 1.41 1.84** v-ets erythroblastosis virus E26 oncogene homolog ERG (avian) 1.24 1.66 2.05** v-ets erythroblastosis virus E26 oncogene homolog ETS2 2 (avian) 1.11 1.06 1.12 FLII flightless I homolog (Drosophila) -1.04 1.13 1.22** FN1 fibronectin 1 1.53 2.54** 5.15*** ICAM-1 intercellular adhesion molecule 1 1.18 -1.79 -1.37 ICAM-2 intercellular adhesion molecule 2 1.29 1.33 1.23 Met proto-oncogene (hepatocyte growth factor MET -1.18 1.03 1.22 receptor) MMP14 matrix metallopeptidase 14 1.22 1.37 1.95*** MMP2 matrix metallopeptidase 2 1.26 1.80* 2.73*** MMP7 matrix metallopeptidase 7 1.00 1.15 1.05 nuclear factor of kappa light polypeptide gene NFKB-1 -1.01 1.14 1.25 enhancer in B-cells 1 PECAM-1 platelet/endothelial cell adhesion molecule 1 1.39 1.04 1.12 PLAT plasminogen activator, tissue 1.24 1.05 1.06 SOD2 superoxide dismutase 2, mitochondrial 1.11 1.03 1.02 SPP1 secreted phosphoprotein 1 (osteopontin) 1.16 6.11* 9.59***

7 TGFB1 transforming growth factor, beta 1 1.06 1.08 1.24 TGFBR2 transforming growth factor, beta receptor II -1.01 1.00 1.02 TIMP1 TIMP metallopeptidase inhibitor 1 1.10 2.14 3.73** TIMP2 TIMP metallopeptidase inhibitor 2 1.00 -1.02 -1.02 TIMP3 TIMP metallopeptidase inhibitor 3 -1.10 1.20 1.00 VCAM-1 vascular cell adhesion molecule 1 1.30 -1.46 -1.61 VIM vimentin 1.06 1.94* 3.59*** VWF von Willebrand factor -2.66* -1.37 -2.81**

8 Suppl figure legends

Suppl Fig 1. GeneGo custom pathway map of the classical complement pathway demonstrating enrichment for genes that are up-regulated in DM+HTN rats compared with control animals (identified by the presence of an adjacent red bar annotated with 1).

Suppl Fig 2. GeneGo custom pathway map of the role of the antigen presenting cell in cell-cycle regulation demonstrating enrichment for genes that are up-regulated in

DM+HTN rats compared with control animals (identified by the presence of an adjacent red bar annotated with 1).

Suppl Fig 3. The Metacore network building algorithm developed a network of extracellular matrix genes from inputed genes that were over-expressed in the DM+HTN animals compared with control rats (identified by the presence of an adjacent red circle).

Suppl Fig 4a. Representative images of renal cortical KIM-1 staining from each group.

Bars represent 50 µM. b. Real-time PCR analysis of KIM-1 gene expression relative to

18S in each group.

Suppl Fig 5. Relative expression of JAK1, JAK2 and STAT1 genes in each of the groups as determined by microarray ( a,c,e , respectively) and then confirmed by rtPCR analysis

(b,d,f , respectively) *p,0.05, **p<0.01 ***p<0.001 v Cont; #p<0.05, ## p<0.01, ### p<0.001 v DM; $p<0.05 v HTN.

Suppl Fig 6. Relative expression of VEGF-A in each of the groups as determined by (a) microarray and then (b) confirmed by rtPCR *p<0.05 v Cont; #p<0.05 v DM.