We previously observed that high-dose angiotensin receptor blocker (ARB) can induce regression of existing glomerulosclerosis. We also found that proliferator-activated recepto-γ (PPARγ) agonist can attenuate glomerulosclerosis in a nondiabetic model of kidney disease, with specific protection of podocytes. We now assessed effects of combination therapy with ARB and pioglitazone on established glomerulosclerosis. Sprague-Dawley male rats underwent 5/6 nephrectomy (5/6 Nx) at week 0 and renal biopsy at week 8. Rats were randomized to groups with equal starting moderate glomerulosclerosis, and treated with ARB, PPARγ agonist (pioglitazone), combination or vehicle from weeks 8 to 12. Body weight, systolic blood pressure (SBP), and urinary protein (UP) were measured at intervals. In rats with established sclerosis, SBP, UP, and GS were equal in all groups at week 8 before treatment by study design. Untreated control rats had hypertension, decreased GFR, and progressive proteinuria and glomerulosclerosis at week 12. Only combination therapy significantly ameliorated hypertension and proteinuria. ARB alone or pioglitazone alone had only numerically lower SBP and UP than vehicle at week 12. Both pioglitazone alone and combination had significantly less decline in GFR than vehicle. Combination-induced regression of glomerulosclerosis in more rats from weeks 8 to 12 than ARB or pioglitazone alone. In parallel, combination treatment reduced plasminogen activator inhibitor-1 expression and macrophage infiltration, and preserved podocytes compared with vehicle. These results were linked to increased AT2 receptor and Mas1 mRNA in the combination group. PPARγ agonists in combination with ARB augment regression of glomerulosclerosis, with downregulation of injurious RAAS components vs PPARγ alone, with increased anti-fibrotic/healing RAAS components, enhanced podocyte preservation, and decreased inflammation and profibrotic mechanisms.
Chronic kidney disease (CKD) is characterized by a progressive loss in renal function with progressive scarring that affects all structures of the kidney. Angiotensin II, a key component of the renin-angiotensin-aldosterone system (RAAS), binds to AT1 or AT2 receptors. The classic AT1 receptor mediates vasoconstriction and promotes extracellular matrix synthesis both directly and via profibrotic factors such as transforming growth factor-β (TGF-β) and plasminogen activator inhibitor-1 (PAI-1).1 The AT2 receptor is counterregulatory, contributing to vasodilatation and apoptosis. Various clinical studies have shown superior effects on progression of RAAS blockade with angiotensin I converting enzyme inhibitor (ACEI) and/or angiotensin II type I receptor blockers (ARB) vs other nonspecific anti-hypertensive treatments.2 Unfortunately, administration of ARBs in doses considered sufficient to robustly block AT1 receptors often fails to provide effective blood pressure (BP) control and to protect against adverse cardiovascular events and diabetes.3, 4 In our previous studies in the rat 5/6 Nx model, we showed that high-dose ARB-induced regression of existing glomerulosclerosis over 4 weeks, although 6-month treatment only ameliorated progression.5 Unfortunately, combination ACEI and ARB may be associated with adverse events, and renin inhibitors also had side effects in clinical trials.6, 7 Thus, RAAS blockade alone is not sufficient to prevent ongoing progression of CKD and hypertension. Approaches to target additional pathways of sclerosis and in particular podocyte injury are therefore of particular interest.
Peroxisome proliferator-activated receptor-γ (PPARγ) agonists are classically used to treat dyslipidemia and enhance insulin sensitivity in type II diabetic patients. PPARγ agonists, such as thiazolidinediones, activate PPARγ, a ligand-activated nuclear transcription factor that dimerizes with the retinoid X receptor and then binds to the PPAR response element of target genes in the nucleus.8 We previously found that PPARγ agonists protect against glomerulosclerosis in nondiabetic models of CKD by reducing TGF-β, PAI-1, and macrophage infiltration.9, 10 PPARγ agonists directly protected podocytes against injury, and maintained expression of nephrin and other key slit-diaphragm proteins.11
The multiple additional signal pathways potentially modulated by PPARγ agonists, beyond RAAS targets, point to possible benefits of combination interaction. We therefore assessed potential additional benefits and mechanisms of combination PPARγ agonist and ARB therapy in existing glomerulosclerosis.
MATERIALS AND METHODS
Adult male Sprague-Dawley rats (200–250 g; Charles River, Chicago, IL, USA) were studied. All protocols were approved by the Institutional Animal Care and Use Committee according to the National Institutes of Health guidelines. Rats were housed under normal conditions with a 12:12-h light–dark cycle at 21 °C with 40% humidity and 12 air exchanges per hour, standard rat chow (‘5001’ diet; rodent diet, 23.4% protein, 4.5% fat, 6.0% fiber, 0.40% sodium; Purina Laboratory, St Louis, MO, USA) and water ad libitum. Rats underwent 5/6 nephrectomy (Nx) by uninephrectomy on the right side and ligation of two or three branches of the left renal artery. This model is characterized by the development of early sclerosis, proteinuria, and hypertension by week 4, with established sclerosis by week 8, with ongoing progression to uremic death by 12–16 weeks.
Rats were randomized at week 8 after 5/6 Nx to achieve similar levels in each group of systemic BP, urinary protein levels, and glomerulosclerosis, assessed by open renal biopsy, before treatment was started. Rats were then treated from weeks 8 to 12 after 5/6 Nx with the ARB TAK-491 (ARB, 3 mg/kg/day in drinking water, n=6), pioglitazone (Pio, 2.5 mg/kg/day in chow, n=7), combination drug therapy (ARB+PIO, n=7), or vehicle (vehicle, n=7). Doses were based on previous studies in hypertensive kidney disease models.12
Functional studies were performed at baseline, weeks 4, 8, and 12 after 5/6 Nx. Systolic BP (SBP) was measured using a conscious tail-cuff method. Animals were placed in metabolic cages for collection of 24 h urine samples. Urine protein excretion was measured by Bio-Rad Protein Assay Kit (Bio-Rad Laboratories, Hercules, CA, USA). Plasma and urine creatinine were measured at week 12 by Vitros CREA slides (Johnson & Johnson Clinical Diagnostics, Rochester, NY, USA).
Kidney tissues were fixed in 4% paraformaldehyde/phosphate-buffered saline solution and routinely processed, and 3 μm sections were cut and stained.
A semiquantitative score (glomerulosclerosis index, sclerosis index (SI)) was used to evaluate the degree of glomerulosclerosis. SI in each glomerulus was graded from 0 to 4+ as follows: 0 defined as no lesion; 1+ sclerosis of <25% of kidney; 2+ sclerosis between 25 and 50% of the glomerulus; 3+ sclerosis between 50 and 75% of the glomerulus; and 4+ sclerosis >75% of the glomerulus. A whole kidney average sclerosis index for each rat was obtained by averaging scores from all glomeruli on PAS-stained sections. All sections were examined without knowledge of the treatment protocol. Change of sclerosis from biopsy to autopsy was calculated for each animal as previously described.10
Sections were deparaffinized and hydrated. For WT1 and ED1, sections were microwaved in citrate buffer (pH 6.0) for 5 min. Endogenous peroxidase was quenched with 3% hydrogen peroxidase for 10 min, and slides were then exposed to Power Block (BioGenex Laboratories, San Ramon, CA, USA) for 45 min. The primary antibodies used were rabbit anti-rat WT1 (1:800; Santa Cruz), rabbit anti-rat PAI-1 (20 μg/ml; American Diagnostica, CT, USA), mouse anti-rat ED1 (1:50; Dako) and mouse anti-rat collagen I (1:400; Abcam, Cambridge, MA, USA). Sections were incubated overnight at 4 °C. Immunoperoxidase staining was performed with the Vectastain ABC kit (Vector Laboratories, Burlingame, CA, USA) with diaminobenzidine as a chromogen. Hematoxylin was used as a counterstain.
WT1 and ED1-positive nuclei per glomerulus were counted, and localization was assessed based on cell anatomical location and morphology. Collagen I stained slides were scanned by using the Aperio CSO system (Leica Biosystems, Buffalo Grove, IL, USA). The positive area was analyzed by using Aperio Imagescope software. All sections were examined without knowledge of the treatment protocol.
Frozen kidney tissue samples were transferred in RIPA buffer containing 1:100 phosphatase inhibitor cocktail I, 1:100 phosphatase inhibitor cocktail II (Sigma, St Louis, MO, USA), and 1:100 proteinase inhibitor cocktail tablet (Roche Diagnostics, Mannheim, Germany). Samples were homogenized and centrifuged, and the protein concentration was measured using the DC Protein Assay kit (Bio-Rad Laboratories, Hercules, CA, USA). PAI-1 (1:400; R&D Systems, Minneapolis, MN, USA), AT1R (1:400; Abcam), AT2R (1:1000; Abcam), and GAPDH (1:500; Cell Signaling) were detected by using the corresponding antibody overnight at 4 °C. Horseradish peroxidase-labeled anti-goat or anti-rabbit IgG secondary antibody (1:2500 in 5% milk TBS-Tween20) was added and incubated at room temperature for 45 min. Protein bands on western blots were visualized by ECL Plus (Amersham, Arlington Heights, IL, USA) according to the manufacturer’s instructions and were developed on film.
Frozen kidney tissue samples were transferred in 1 × hypotonic buffer supplemented with 1M DTT and detergent. Samples were homogenized and centrifuged, and nuclear material was extracted with Complete Lysis Buffer using a Nuclear Extraction Kit (Active Motif, Carlsbad, CA, USA). Protein concentration was measured using a BSA assay (Active Motif) and PPARγ activity was measured using a TransAM PPARγ kit (Active Motif).9
8-isoprostane expression in kidney was measured by the EIA kit (Oxford Biomedical Research, Rochester Hills, MI, USA).
Quantitative Real-Time PCR
Reverse transcription was performed using the TaqMan Reverse Transcription Kit (Applied Biosystems, Branchburg, NJ, USA) in a total reaction volume of 80 μl (2 μg total RNA, 5 mM MgCl2, 1 mM dNTPs, 1 U/μl RNase inhibitor, 2.5 μM random hexamers, and 2.5 U/μl MuLV-RT) for 30 min at 42 °C. Quantitative real-time PCR was performed in a total reaction volume of 25 μl using iQ SYBR Green Supermix (Bio-rad, Hercules, CA, USA), 1 μl cDNA, and 500 nM forward and reverse primers. IL-1, IL-6, renin, angiotensinogen, ACE, AT1R, AT2R, and Mas1 primers were from SABiosciences Corporation (Frederick, MD, USA). Quantitative real-time PCR was carried out using the iCycler iQ Multi-Color Real-Time PCR Detection System (Bio-Rad) with the following cycling parameters: polymerase activation for 3 min at 95 °C and amplification for 40 cycles of 20 s at 95 °C, 20 s at 60 °C, and 40 s at 72 °C. Following amplification, a final melting curve was obtained for each sample by cooling the reaction mixture to 55 °C and then slowly heating it to 95 °C at 0.2 °C/s. Experimental cycle threshold (Ct) values were normalized to GAPDH measured on the same plate, and fold differences in gene expression were determined by using the 2−△△Ct method.
Data are expressed as means±s.e., and as change in each rat from start of intervention till killing. Differences among groups were examined by one-way ANOVA followed by Bonferroni’s test. Nonparametric data were compared by Mann–Whitney U-test. P-values<0.05 were considered significant.
Combination of PPARγ Agonist with ARB-Protected Renal Function
SBP increased from 194.6±13.2 at week 8 to 217.8±13.3 mm Hg at week 12 in the vehicle group. Both ARB and pioglitozone numerically blunted this increase in BP (ARB 2.0±4.1%, Pio 3.3±6.0%, vs vehicle 14.4±7.2% change week 12 vs week 8, pNS). Only combination treatment significantly decreased SBP compared with baseline at week 8 (ARB+Pio −10.1±3.2%, vs vehicle, P<0.05) (Figure 1a and Table 1).
Proteinuria increased from 258.3±37.3 to 428.8±69.5 mg per 24 h in the vehicle group. Combination treatment, but not ARB or pioglitazone alone, blunted the increase in proteinuria (ARB+Pio 44.6±10.8%, vs vehicle 112.7±29.8% change vs wk 8, P<0.05) (Figure 1b and Table 1).
Serum creatinine increased from weeks 8 to 12 in all groups. Only combination treatment significantly ameliorated the rise of Scr (ARB+Pio 21.5±7.7%, vs vehicle 47.6±6.4% change vs wk 8, P<0.05) (Figure 2a and Table 1). Similarly, creatinine clearance decreased from weeks 8 to 12 in vehicle. Both pioglitazone and combination treatment protected renal function (change in Ccr, Pio −27.8±10.5%, ARB+Pio −29.0±10.3%, vs vehicle −58.2±5.2%, P<0.05), while ARB alone did not have significant effect vs vehicle (−34.9±14.1%) (Figure 2b and Table 1).
Effects on Progression
Glomerulosclerosis was similar at biopsy in all treatment groups by study design. At week 12 after 5/6 Nx, vehicle rats showed progressive glomerulosclerosis, as evidenced by an average 176.7% increase in the SI from biopsy to autopsy. In contrast, ARB alone or Pio alone ameliorated progression of glomerulosclerosis vs vehicle (ARB 98.4±59.3%, Pio 59.8±25.9% vs vehicle 176.7±76.2%, change in SI from biopsy to autopsy, pNS). Combination therapy reduced the extent of glomerulosclerosis at week 12 (ARB+Pio -11.7±24.8%, vs vehicle, P<0.05) (Figures 3a and b and Table 1). Regression, defined as reduction of sclerosis from biopsy to autopsy in the same rat, occurred in 5/7 rats in the ARB+Pio group, compared with 2/6 in ARB and 1/7 in vehicle and Pio groups (ARB+Pio vs vehicle, P<0.05; ARB vs vehicle, pNS) (Figure 3c).
Interstitial fibrosis, measured by collagen I staining, was significantly reduced by combination therapy (ARB+Pio 9.90±0.53, vs vehicle 12.97±0.92%, P<0.05), but not by ARB or Pio alone (ARB 12.64±1.20, Pio 12.42±1.04%, vs vehicle, pNS) (Figure 4a).
PAI-1, a molecule that can inhibit matrix degradation, was not significantly changed by ARB or Pio alone vs vehicle, while combination therapy resulted in significantly less PAI-1 expression when compared with vehicle (ARB+Pio 0.04±0.01, vs vehicle 0.14±0.03, PAI-1 ratio to β–actin, P<0.05) (Figure 4b). 8-isoprostane, a marker for reactive oxygen species, increased in vehicle-treated 5/6 Nx kidneys (894.63±7.74 pg/mg) compared with normal kidneys (207.22±71.28 pg/mg), with no further effect by treatment (ARB 800.19±117.99, Pio 978.95±275.08, and ARB+Pio 913.60±90.83 pg/mg, respectively).
Effects on Inflammation and Podocytes
Macrophage staining was most prominent in glomeruli of vehicle rats at week 12 (5.35±1.76 ED1+ cells/glom). Treatment with pioglitazone did not significantly change infiltrating glomerular macrophages (3.43±1.18 ED1+ cells/glom), in contrast to a significant decrease in macrophage infiltration in the ARB+Pio rats (1.54±0.41 ED1+ cells/glom, vs vehicle, P<0.05) (Figure 5a). IL-1 and IL-6, measured in renal cortex by real-time PCR, did not show any difference among groups (data not shown).
Density of WT1-positive cells in the glomeruli was taken as an indicator of differentiated podocytes. At week 8, glomerular WT1-positive cell density was reduced similarly in all groups vs normal baseline. At week 12, the vehicle group showed 0.84±0.07 × 10−3/μm2 WT1+ cells in glomeruli. ARB alone or Pio alone was not different vs vehicle (ARB 0.90±0.09, Pio 0.88±0.12 × 10−3/μm2, vs vehicle, pNS), but the ARB+Pio combination resulted in higher density of WT1+ glomerular cells (1.08±0.10 × 10−3/μm2, vs vehicle, P<0.05) (Figure 5b).
Effects on PPARγ Activity and RAAS Components
Kidney cortical PPARγ activity was numerically increased in ARB compared with vehicle at week 12 (ARB 0.19±0.03, vs vehicle 0.11±0.02, pNS). Pioglitazone alone and combination treatment showed significantly more activated PPARγ in the kidney (Pio 0.24±0.03, ARB+Pio 0.25±0.05 vs vehicle, P<0.05) (Figure 6a).
Combination treatment numerically decreased AT1R protein expression (ARB+Pio 0.49±0.09 vs vehicle 1.09±0.49, pNS) and pioglitazone numerically increased AT2R (ARB+Pio 1.85±0.34 vs vehicle 1.34±0.22, pNS), while ARB did not change the AT1R (0.71±0.20) and AT2R (1.37±0.21) protein expression. Pioglitazone alone significantly enhanced renin mRNA expression and reduced ACE mRNA expression vs vehicle. The combination group had lower renin mRNA than pioglitazone alone. AT2R and Mas1, an Ang(1–7) receptor, mRNA expressions were increased only in the combination groups compared with the vehicle. Angiotensinogen and AT1R mRNA were not different among groups (Figure 6b).
Treatment of CKD has relied on reducing proteinuria, normalizing BP, and antagonism of the RAAS. However, these approaches have mainly slowed progression of CKD in most patients. Our previous studies have shown that high-dose ARB, beyond BP control, can induce regression of glomerulosclerosis in animal models over 4 weeks.13 However, over 6-month follow-up, high-dose ARB merely ameliorated progression in the rat 5/6 Nx model.5 In this study, PPARγ agonist in addition to ARB, was more effective than either alone in ameliorating progression and even regressing existing glomerulosclerosis. In particular, only combination treatment had beneficial effects on systemic BP, proteinuria, GFR, glomerulosclerosis index, and PAI-1. Mechanisms of combination therapy included decreased inflammation and preservation of differentiated podocytes.
ARB affects the PPARγ system. Classic ARBs have little direct effect on PPARγ expression or activity, but some ARBs, such as telmisartan, are also partial agonists of PPARγ, which may provide more protection in patients with complicated diabetes or evidence for coronary artery disease without heart failure.14 Azilsartan, the ARB used in this study, is a potent AT1 receptor antagonist that is structurally similar to candesartan except for bearing a 5-oxo-1,2,4-oxadiazole moiety in place of a tetrazole ring. Azilsartan is less acidic and more lipophilic than candesartan and in preclinical studies has been shown to be significantly more potent than candesartan in improving glucose tolerance and tissue glucose metabolism.12 In cell-based receptor transactivation studies in vitro, azilsartan appears to have little or no direct effect on PPARγ activity.12 However, in a mouse model of type II diabetes, oral administration of very low doses of azilsartan significantly increased PPARγ gene expression in adipose tissue.12 In our study, this ARB showed a trend to increase PPARγ activity, but still far lower than that seen with pioglitazone. The combination of azilsartan and pioglitazone showed enhanced renal protection vs the ARB alone, suggesting maximal PPARγ activity may provide extra benefit.
Conversely, PPARγ agonists affect the RAAS. In vitro, PPARγ activation reduced AngII production in mesangial cells.15 PPARγ agonists also downregulated AT1R expression in proximal tubular epithelial and vascular smooth muscle cells with inhibition of ERK1/2 activation and attenuated NF-κ B activation.16, 17, 18 Further, in AT1a receptor-overexpressing HEK 293 cells, PPARγ agonist stimulated this receptor to induce endogenous βarr1/2 recruitment and AT1aR internalization without Gq protein activation.19 In our study, we did not detect change of AT1R expression, but can not exclude a potential effect of PPARγ on AT1R activity through other pathways, such as G protein signaling.20 A change in balance of AT1 vs AT2 receptor could also influence progression of renal disease. We showed previously that AT2 blockade prevented beneficial effects of AT1 blockade by ARB, supporting a beneficial role of the AT2 receptor.21 In the current study, PPARγ agonist showed a trend to reduce AT1R and upregulate AT2R protein levels, but had no effect on mRNA. We also found that PPARγ agonist significantly induced renin expression. Two newly described mechanisms are of relevance in this regard. First, the recently identified PPARγ target sequence Pal3 is specific for the human renin gene and mediates exceptionally high sensitivity to transactivation.22 In addition, PPARγ has potentiating effects on cAMP signaling in renin-producing cells.23 The reduced ACE by PPARγ agonist is also consistent with previous reports.24
Combination of ARB and PPARγ agonist enhanced renal protection in the current study. All resident cells in glomeruli, including podocytes, express PPARγ, AT receptors and PAI-1 after glomerular injury. Some infiltrating cells, such as macrophages, also express AT receptors and PAI-1. The enhanced renal protection by combination intervention may be related to interactions of the RAAS and PPARγ, and involve effects on both systemic and cell functions affected by ARB and PPARγ agonist.
Of note, the anti-hypertensive effect of the combination therapy was significantly greater than that of the monotherapy. ARBs have direct anti-hypertensive effects. However, PPARγ agonist attenuated the development of hypertension in Dahl salt-sensitive rats and inhibited proliferation of renal arteriolar smooth muscle cells.25 Furthermore, PPARγ agonist also ameliorated endothelial dysfunction of afferent arterioles in Zucker obese rats and diet-induced type 2 diabetic mice.26, 27 PPARγ agonist dilated not only the afferent but also the efferent arterioles in an ex vivo study.28 Improved endothelial function in intrarenal arteries and arterioles, accompanied by a fall in systemic BP, may account for the enhanced effects on proteinuria by the ARB plus PPARγ agonist combination therapy. ARBs may reduce proteinuria through functional hemodynamic effects, and also have direct beneficial effects on podocytes.29, 30 We also observed additional benefits on structural injury by combination therapy, with less progression and preserved glomerular WT1 positivity, a marker of differentiated podocytes, compared with monotherapy. PPARγ agonists have direct effects on podocytes, with reduced apoptosis and injury and improved podocyte differentiation in vitro.31 Although combination treatment did not induce additional PPARγ activity, renin was decreased compared with PPARγ agonist alone. Combination treatment also further increased AT2R and Mas1 expression, which may thus be another mechanism for the further benefit, with AT2 and Ang(1–7) implicated in repair/anti-fibrotic responses.
Combination therapy with losartan and pioglitazone has previously shown added effects in improving structure and function in rats with metabolic syndrome.32 PPARγ agonist also enhanced the renoprotective effects of an ARB in patients with diabetic nephropathy, which could in part relate to improved control of diabetes.33 Of note, our study shows that combination of PPARγ agonist and ARB induced more regression of glomerulosclerosis than monotherapy, linked to enhanced podocyte preservation, decreased glomerular inflammation and decreased profibrotic PAI-1, and increased AT2R and Mas1 expression, potential mediators of healing. We conclude that a rational choice of combined interventions to augment dampening of the multiple injury pathways activated in progressive kidney disease can achieve additional benefits on structure and function. It was also supported by O’Brien center grant (P30 DK079341, ABF).
We thank Ellen Donnert for expert technical assistance. This work was supported in part by a grant from Takeda Pharmaceuticals USA and by a grant from NIDDK DK 56942 (ABF).
About this article
Current Opinion in Nephrology and Hypertension (2017)