Effects of the dual PPAR-α/γ agonist aleglitazar on glycaemic control and organ protection in the...

orig

inal

artic

le

Diabetes, Obesity and Metabolism 15: 164–174, 2013.© 2012 Blackwell Publishing Ltdoriginal article

Effects of the dual PPAR-α/γ agonist aleglitazaron glycaemic control and organ protection in the Zuckerdiabetic fatty ratA. Benardeau1, P. Verry1, E.-A. Atzpodien2, J. M. Funk2, M. Meyer1, J. Mizrahi1, M. Winter2,M. B. Wright1, S. Uhles1 & E. Sebokova1

1pRED, Pharma Research & Early Development, DTA Cardiovascular & Metabolism, F. Hoffmann-La Roche AG, Basel, Switzerland2pRED, Non-Clinical Safety, F. Hoffmann-La Roche AG, Basel, Switzerland

Aims: To evaluate the effects of aleglitazar, a dual peroxisome proliferator-activated receptor-α/γ agonist, on the development ofdiabetes-related organ dysfunction, in relation to glycaemic and lipid changes, in Zucker diabetic fatty (ZDF) rats.Methods: Six-week-old, male ZDF rats received aleglitazar 0.3 mg/kg/day or vehicle as food admix for 13 weeks (n = 10 per group).Age-matched male Zucker lean rats served as non-diabetic controls. Plasma and renal markers were measured at several time points.Histopathology and quantitative immunohistochemistry were performed at 13 weeks.Results: Glycated haemoglobin (5.4 vs. 9.2%) and blood glucose (8.3 ± 0.3 vs. 26.1 ± 1.0 mmol/l) were significantly reduced at 12 weeks withaleglitazar versus vehicle-treated ZDF rats (both p < 0.01), while aleglitazar preserved near-normal plasma insulin levels. Aleglitazar preventedthe development of hypertriglyceridaemia (1.4 ± 0.1 vs. 8.5 ± 0.9 mmol/l) and reduced plasma non-esterified fatty acids (0.09 ± 0.02 vs.0.26 ± 0.04 mmol/l) relative to vehicle-treated animals (both p < 0.01). Urinary glucose and protein concentrations were significantly reducedat 13 weeks with aleglitazar versus vehicle-treated rats (both p < 0.01). Consistent with its effect on glycaemic control, aleglitazar protectedβ-cell morphology, as evidenced by preservation of islet integrity, and reduction of β-cell apoptosis and islet fibrosis. Aleglitazar preventedrenal glomerular hypertrophy, podocyte degeneration, glomerulosclerosis, tubulo-interstitial lesions and development of cataracts.Conclusions: Aleglitazar strongly improved glycaemic and lipid parameters while protecting key tissues, including the pancreas, kidneys andeyes, against diabetes-associated structural and functional changes in the ZDF rat.Keywords: β cell, diabetes complications, dual PPAR-α and -γ agonist, Zucker diabetic fatty rat

Date submitted 15 May 2012; date of first decision 12 June 2012; date of final acceptance 2 September 2012

IntroductionLandmark studies emphasize the need for comprehensivemanagement of multiple cardiovascular (CV) risk factors inpatients with type 2 diabetes mellitus (T2DM) to reduce CVmorbidity and mortality and microvascular complications [1].While agonists of peroxisome proliferator-activated receptor(PPAR)-γ (thiazolidinediones) provide effective glycaemiccontrol, and thus may reduce microvascular complications,only pioglitazone in the PROactive trial showed a reductionin CV events, as reflected by a reduction in the composite(secondary) end point of all-cause mortality, non-fatalmyocardial infarction and stroke [2]. PPAR-α agonists(fibrates) lower triglycerides (TG) and increase high-densitylipoprotein cholesterol (HDL-C) [3], although their effects toreduce risk for CV events have been modest [4].

Dual-selective agonists of PPAR-α and -γ are of interest asthey may combine the glucose-lowering and lipid-modifying

Correspondence to: Elena Sebokova, PhD, pRED Pharma Research & Early Development,DTA Cardiovascular & Metabolism, F. Hoffmann-La Roche AG, Grenzacherstrasse 124, BaselCH4070, Switzerland.E-mail: [email protected]

properties of both agents. Aleglitazar is a balanced dualPPAR-α/γ agonist designed to optimize effects on glucoseand lipids. In diabetic rodent models, aleglitazar significantlyimproved insulin sensitivity (Zucker fa/fa rats and db/dbmice) and improved plasma lipids (human apolipoproteinA1 transgenic mice) [5]. In a Rhesus monkey model of T2DM,aleglitazar more than doubled levels of HDL-C and significantlyreduced both low-density lipoprotein cholesterol (LDL-C) andTG, while lowering glucose and glycated haemoglobin (HbA1c)[6]. In a phase II dose-range-finding study (SYNCHRONY)in T2DM patients, aleglitazar dose-dependently improved thelipid profile (TG, LDL-C and HDL-C) and decreased glucose(HbA1c and fasting plasma glucose), as well as reduced othermarkers of CV risk, such as plasminogen activator inhibitor-1,fibrinogen and C-reactive protein [7].

Zucker diabetic fatty (ZDF) rat is a useful animal modelof T2DM disease progression and end-organ dysfunction[8–10]. Insulin resistance and slightly elevated plasma glucoseare typically evident by 7 weeks of age, with hyperinsulinaemiaoccurring between weeks 7 and 10. Insulin levels eventuallydecline, due to β-cell loss [11], with overt diabetes developingby 12 weeks. The disease course in the ZDF rat has similarities

DIABETES, OBESITY AND METABOLISM original articleto T2DM in humans, including declining renal function(glomerular and tubulo-interstitial alterations with proteinuriaand glucosuria) [10], presence of neuropathy (decreases inperipheral nerve conduction and alterations in sensoryparameters) [12] and retinopathy (retinal vascular changesand cataracts) [13].

The therapeutic effects of thiazolidinediones and fibrateshave been previously studied in the ZDF rat. Rosiglitazonewas shown to improve insulin sensitivity and glucosecontrol, leading to preservation of pancreatic β-cell massand insulin content [14,15]. Rosiglitazone had protectiveeffects on kidney function, resulting in significantly reducedproteinuria and improvements in glomerular and tubularmorphology – effects that were superior to the angiotensin-converting enzyme inhibitor cilazapril [16]. Pioglitazonehas also been shown to reduce proteinuria in ZDF ratsbecause of glucose-dependent and -independent mechanisms.Pioglitazone reduced glomerulosclerosis by reducing podocyteactivation and endothelial injury [17,18]. Effects were evidentdespite minor reductions in glucose, suggesting they may occurvia anti-fibrotic PPAR-dependent mechanisms directly in thekidney, where PPAR-γ is highly expressed [19].

Fibrate PPAR-α agonists have also been evaluated in ZDFrats. Fenofibrate reduced systolic blood pressure, and atten-uated glomerular hypertrophy and fibrosis [20]. Fenofibratealso reduced tubular injury, possibly by reducing interstitialmacrophages in the kidney, leading to suppression ofnuclear factor κB and transforming growth factor-β1/Smad3signalling [21].

In this study, we investigated effects of the dual PPAR-α/γagonist aleglitazar on glycaemic control and prevention ofdiabetes-related organ dysfunction in ZDF rats, employingquantitative immunohistochemistry with accepted andemerging structural and functional markers.

Materials and MethodsAnimals

Five-week-old male ZDF rats (ZDF fa/fa or ZDF) and lean,aged-matched male Zucker rats (ZDF +/fa or lean [ZL])from Charles River Laboratories (Wilmington, MA, USA)were fed a diabetogenic diet (KLIBA 2437; Provimi KLIBAAG, Kaiseraugst, Switzerland) immediately after weaning.The animal study was performed under an animal researchprotocol (ARP) authorization to F. Hoffmann-La Roche by theSwiss Cantonal Veterinary Office Basel-Stadt. Animals werehoused individually under standard hygienic conditions in anAAALACi-accredited facility with controlled temperature andhumidity, a 12 h/12 h light/dark cycle, and full access to foodand water.

Study Design and Blood/Urine Sampling

Rats were 6 weeks old when treatment commenced. Ani-mals received aleglitazar (0.3 mg/kg/day; n = 10; Roche, Basel,Switzerland) as food admix or vehicle (vehicle; n = 10) for13 weeks. For each treatment group, 50 kg of food (KLIBA2437) was homogenized in 10 l of vehicle buffer solution (0.89%

di-sodium hydrogenphosphate-dihydrate; 0.4% polysorbate80, 0.29% 1 N HCl in sterile water), with or without aleglitazar,and the fluid phase was carefully evaporated. Final food was pre-pared in pellets by KLIBA. Average daily food consumption was100 g/kg body weight during treatment (confirmed by record-ing of daily food intake), delivering aleglitazar at a daily dose of0.3 mg/kg/day. The potential food aversion risk was exploredprior to starting chronic treatment by monitoring food andwater intake and body weight on diet over a 1-week period. Nofood aversion behaviour was observed. Pharmacokinetic anal-ysis confirmed that the expected plasma exposure of aleglitazarwas achieved (data not shown). Age-matched male ZL rats(n = 10) received vehicle and served as non-diabetic controls.

Retro-orbital blood (200 μl of blood with EDTA) wastaken from anaesthetized rats (isoflurane) into tubes coatedwith EDTA, containing 1% aprotinin (to inhibit proteindegradation), and kept on ice. Animals were carefullymonitored both before and after blood sampling and nosigns of distress or complications were observed. Tubes werethen centrifuged (9279 g for 15 min at 4 ◦C) and plasma wasaliquoted in different tubes or 96-well plates for analysis.Blood samples were taken under postprandial conditions inthe morning 2 h following switch from dark to light cycle toassess glucose, HbA1c, insulin, TG, non-esterified fatty acids(NEFA), total cholesterol (TC) and adiponectin at several timepoints under non-fasting conditions. Glucose and HbA1c weremeasured in whole blood, while plasma was used for analysisof insulin, TG, NEFA, TC and adiponectin. Urine sampleswere taken after 5 weeks of treatment and at the end of thetreatment period (13 weeks); five representative animals pergroup according to similar body weight and plasma parameterswere placed individually into metabolic cages, without access tofood but with free access to water, for 16–17 h urine collectionusing ice-refrigerated tubes.

An oral glucose tolerance test (OGTT) was carried outat 12 weeks in overnight-fasted animals, with measurementstaken at 15, 30, 60 and 120 min after glucose challenge(1 g/kg). Homeostasis model assessment of insulin resistance(HOMA-IR) and β-cell function (HOMA-B) were estimatedafter overnight fasting and prior to performing OGTTusing the following calculations: HOMA-IR = fasting insulin(ng/ml)/0.04 × fasting blood glucose (mM)/22.5; HOMA-B = 20 × fasting insulin (ng/ml)/0.04/fasting blood glucose(mM) − 3.5 [22].

Blood glucose levels were measured using a glucometer(GlucoTrend®, Roche Diagnostics, Burgdorf, Switzerland).Plasma concentrations of insulin (Rat Insulin, Mercodia,Uppsala, Sweden) and adiponectin (Mouse/Rat ELISA kit,K1002-1, B-Bridge, Cupertino, CA, USA) were quantifiedby ELISA. Metabolic plasma profiles were established by theHitachi system using a fluorometric method: TC, TG, NEFA,HbA1c/Hb (turbimetric assay). HbA1c (%) was calculated asfollows: 87.6 × (HbA1c/Hb) + 2.27 [23].

Histopathology and Immunohistochemistry

Tissue samples were fixed in 10% neutral-buffered formalinfor at least 24 h. Eyes were fixed in modified Davidson’s fixativeconsisting of 13.3% ethyl alcohol, 6.3% glacial acetic acid and

Volume 15 No. 2 February 2013 doi:10.1111/dom.12006 165

original article DIABETES, OBESITY AND METABOLISM

13.9% formalin solution in distilled water, for at least 48 h.For histopathology examination, organs were embedded inParaplast. Tissue sections (2–3 μm) were stained with haema-toxylin eosin (pancreas, kidneys) or periodic acid-Schiff (PAS;kidneys). Glycogen within distal tubular epithelial cells wasdetermined in adjacent kidney sections on two separate slides.One section was pretreated with diastase, and both were stainedwith PAS. Glycogen was determined in the non-diastase-treatedsection. The following immunofluorescent stainings were car-ried out to assess islet morphology: anti-insulin (DAKO,Glostrup, Denmark), anti-glucagon (R&D Systems, Minneapo-lis, MN, USA) and 4′, 6-diamidino-2-phenylindole (DAPI;DAKO); to assess β-cell apoptosis: anti-insulin (DAKO),transferase-mediated dUTP nick-end labelling (TUNEL; RocheDiagnostics) and DAPI (DAKO); and to assess islet fibrosis:a cocktail of anti-insulin (DAKO), anti-glucagon (R&D Sys-tems), anti-pancreatic polypeptide (Millipore, Billerica, MA,USA), anti-somatostatin (Genetex, Irvine, CA, USA), anti-fibronectin (Sigma, St Louis, MO, USA) and DAPI (DAKO),followed by respective secondary fluorescent antibodies (Invit-rogen, Carlsbad, CA, USA). Digital imaging fluorescencemicroscopy of the pancreas was performed using a scan-ning platform (MetaSystems, Altlussheim, Germany) witha Zeiss Imager Z.2 microscope (Carl Zeiss MicroImaging,Inc., Jena, Germany). Immunohistochemical staining of kid-ney sections was carried out using anti-desmin (Millipore),anti-synaptopodin (Acris Antibodies, Herford, Germany),anti-collagen IV (Millipore), anti-fibronectin (GeneTex, SanAntonio, TX, USA), anti-vimentin (Roche/Ventana) andanti-T-cell immunoglobulin mucin/kidney injury molecule-1(TIM/KIM1) (R&D Systems) antibodies. Anti-desmin was usedto detect glomerular podocyte damage; anti-synaptopodin,anti-collagen IV and anti-fibronectin to detect further glomeru-lar alterations; and anti-vimentin, anti-TIM/KIM1, anti-collagen IV and anti-fibronectin to detect tubulo-interstitialdamage. Digital brightfield imaging of the kidneys was per-formed using the Aperio ScanscopeTM slidescanner (Aperio,Vista, CA, USA). Quantitative image analysis of islet andglomerular staining was performed using Definiens ArchitectXD (Definiens AG, Munich, Germany).

Statistical methods

Statistical analysis was performed by analysis of variancefollowed by Dunnett’s post hoc test. Any between-groupdifferences versus vehicle were considered significant atp ≤ 0.05. Data are reported as mean ± standard error of themean, unless otherwise stated.

ResultsGlucose and Lipid Effects of Aleglitazar

ZDF rats at baseline had slightly elevated blood glucose levelscompared with their non-diabetic ZL counterparts (Table 1).Blood glucose in untreated ZDF rats progressively increasedfrom baseline to 4 and 12 weeks, while blood glucose inaleglitazar-treated ZDF rats remained normal. Aleglitazarsignificantly lowered HbA1c compared with vehicle-treated

Table 1. Effect of aleglitazar on non-fasted metabolic parameters.

ZDF(N = 10)∗

ZDF +aleglitazar(N = 10)∗

ZL(N = 10)∗

Baseline (age = 6 weeks)Blood glucose, mmol/l 7.9 ± 0.2 8.0 ± 0.3 7.5 ± 0.1HbA1c, % 4.6 ± 0.1 4.7 ± 0.1 4.7 ± 0.1Plasma insulin, ng/ml 4.8 ± 0.5 5.0 ± 0.7 0.2 ± 0.0Plasma triglycerides, mmol/l 1.8 ± 0.1 1.8 ± 0.1 0.6 ± 0.0Plasma non-esterified fatty

acids, mmol/l0.07 ± 0.01 0.05 ± 0.02 0.21 ± 0.04

Plasma total cholesterol,mmol/l

2.6 ± 0.1 2.6 ± 0.0 2.2 ± 0.1

Body weight, g 204 ± 3 205 ± 3 160 ± 4At 4 weeks of treatment (age = 10 weeks)Blood glucose, mmol/l 17.5 ± 1.9 7.6 ± 0.1† 7.8 ± 0.2HbA1c, % 5.5 ± 0.1 4.9 ± 0.0† 4.8 ± 0.0Plasma insulin, ng/ml 8.3 ± 1.1 1.6 ± 0.3† 0.9 ± 0.1Plasma triglycerides, mmol/l 10.0 ± 0.5 0.6 ± 0.1† 1.0 ± 0.0Plasma non-esterified fatty

acids, mmol/l0.32 ± 0.03 0.05 ± 0.02† 0.32 ± 0.03


3.0 ± 0.0 3.6 ± 0.1† 2.0 ± 0.1

Body weight, g 354 ± 6 435 ± 7† 278 ± 4At 12 weeks of treatment (age = 18 weeks)Blood glucose, mmol/l 26.1 ± 1.0 8.3 ± 0.3† 7.3 ± 0.2HbA1c, % 9.2 ± 0.5 5.4 ± 0.1† 5.6 ± 0.1Plasma insulin, ng/ml 2.6 ± 0.5 1.9 ± 0.2 1.0 ± 0.0Plasma triglycerides, mmol/l 8.5 ± 0.9 1.4 ± 0.1† 1.6 ± 0.1Plasma non-esterified fatty

acids, mmol/l0.26 ± 0.04 0.09 ± 0.02† 0.26 ± 0.02


5.0 ± 0.1 4.5 ± 0.1 2.3 ± 0.1

Plasma adiponectin, ng/ml 3.3 17.9 4.2Glucosuria, mmol/l∗ 81.3 ± 0.4 0.5 ± 0.1† 1.4 ± 0.1Proteinuria, mg/ml∗ 4.4 ± 1.2 0.3 ± 0.1† 1.8 ± 0.2Body weight, g 428 ± 16 765 ± 16† 386 ± 5

ZDF, Zucker diabetic fatty; ZL, Zucker lean.Data are presented as means ± s.e.m.; ANOVA followed by Dunnett’s posthoc test.∗N = 5 per group for glucosuria and proteinuria, evaluated at 13 weeks.†p < 0.01 ZDF + aleglitazar versus ZDF.

ZDF rats at both 4 and 12 weeks. At the end of the study period,blood glucose and HbA1c were similar in both non-diabeticZL and aleglitazar-treated ZDF rats (Table 1). These dataindicate that aleglitazar effectively prevents the developmentof hyperglycaemia.

At baseline, ZDF rats were already hyperinsulinaemic(Table 1). In untreated ZDF rats, plasma insulin levels progres-sively increased up to 4 weeks and declined thereafter. Alegli-tazar reversed the development of early hyperinsulinaemia,maintaining near-normal plasma insulin levels at 4 weeks thatwere significantly lower than in control ZDF rats and onlyslightly elevated versus non-diabetic ZL rats. At 12 weeks, thedifference between plasma insulin levels in aleglitazar-treatedand vehicle-treated ZDF rats was not significant due to theadvanced state of disease in the untreated animals.

Aleglitazar strongly induced plasma adiponectin levels andshowed significant effects on non-fasted plasma lipids (Table 1).

166 Benardeau et al. Volume 15 No. 2 February 2013

DIABETES, OBESITY AND METABOLISM original articleTable 2. Effects on fasted parameters of glucose homeostasis and oralglucose tolerance test (OGTT) after 12 weeks of treatment.

ZDF(N = 10)

ZDF +aleglitazar(N = 10)

ZL(N = 10)

Fasting blood glucose, mmol/l 9.3 ± 1.1 5.8 ± 0.1∗ 4.4 ± 0.0Fasting plasma insulin, ng/ml 2.84 ± 0.86 1.70 ± 0.19 0.95 ± 0.03Glucose AUC0-120, mmol/

l.min2054 ± 265 971 ± 68∗ 883 ± 38

HOMA-IR 26.8 ± 5.9 11.0 ± 1.3∗ 4.6 ± 0.2HOMA-B 322 ± 143 379 ± 36 564 ± 22

HOMA-B, homeostasis model assessment of β-cell function; HOMA-IR,homeostasis model assessment of insulin resistance; ZDF, Zucker diabeticfatty; ZL, Zucker lean.Data are presented as means ± s.e.m. ANOVA followed by Dunnett’s posthoc test.∗p < 0.01 ZDF + aleglitazar versus ZDF.

At baseline, ZDF rats had elevated plasma TG and lower plasmaNEFA levels compared with non-diabetic ZL controls. In ZDFrats, plasma TG progressively increased over time – an effectthat was largely prevented by aleglitazar. At the end of thetreatment period, aleglitazar-treated ZDF rats had plasma TGcomparable to the ZL controls. Plasma cholesterol was elevatedin controlled ZDF rats compared with ZL rats but was notsignificantly changed by aleglitazar.

Fasting Parameters and OGTT. Aleglitazar significantlyreduced fasting blood glucose compared with vehicle-treatedZDF rats at 12 weeks (Table 2). Following an OGTT, aleglitazarsignificantly reduced glucose area under the curve comparedwith vehicle-treated ZDF rats at 12 weeks (Table 2) and normal-ized glucose excursions to a level comparable to vehicle-treatedZL rats. Fasting plasma insulin levels were slightly lowerwith aleglitazar compared with vehicle-treated ZDF rats after12 weeks of treatment, but were higher than in non-diabeticZL rats (Table 2; not significant vs. vehicle-treated ZDF rats).Reductions in HOMA-IR indicated a trend to improved insulinsensitivity with aleglitazar compared with vehicle-treated ZDFrats (Table 2; not significant vs. vehicle-treated ZDF rats).

Body Weight. At baseline, body weight was increased in ZDFrats compared with ZL controls (Table 1). Aleglitazar wasassociated with significantly increased body weight comparedwith vehicle-treated ZDF rats at both 4 weeks and 12 weeks.

Effects of Aleglitazar on Islet Morphology, β-CellMarkers and β-Cell Function

Severe β-cell destruction was notable in vehicle-treated ZDFrats. As shown in figure 1 and Table 3, loss of islet structuralintegrity and insulin content occurred in vehicle-treated ZDFrats after 13 weeks of treatment. Islets were enlarged, con-taining irregular projections into the exocrine pancreas, withα-cells scattered throughout individual islets and vacuolatedendocrine cells dispersed within the exocrine tissue. In contrast,aleglitazar preserved islet integrity, maintaining normal α-celllocalization at the islet periphery while preserving β-cell area

and number (figure 1 and Table 3). In aleglitazar-treated ZDFrats, β-cell area, number and insulin content were higher thanin non-diabetic ZL rats (figure 1).

Aleglitazar significantly prevented islet fibrosis (indicatedby percentage fibrotic islet area and fibronectin stainingintensity) to a level comparable to that of ZL rats (figure 2).Aleglitazar treatment significantly protected β-cells fromapoptosis (indicated by number of apoptotic β-cells andpercentage apoptotic β-cell number per islet cross-section)compared with vehicle-treated ZDF rats (figure 2).

Effect of Aleglitazar on Cataract Formation

Aleglitazar protected against cataract formation, whichoccurred in vehicle-treated ZDF rats (see figure 3 and Table 3).Clinical examination revealed vehicle-treated ZDF rats had lensopacity indicative of diabetic cataracts as compared with ZLrats. Histologically, cataracts were characterized by degenera-tion and swelling of lens fibres, with formation of Morgagnianglobules, abnormally retained nuclei (bladder cells) andhyperplasia of the lens epithelium underlying the anteriorcapsule. Lens opacity or histopathologic changes indicative ofcataract formation were absent in aleglitazar-treated rats.

Effects of Aleglitazar on Renal Function, andGlomerular and Tubulo-Interstitial Markers

ZDF rats developed renal dysfunction, as indicated by highlevels of urinary protein. At 13 weeks, aleglitazar significantlyreduced urinary protein and glucose compared with vehicle-treated ZDF rats to levels characteristic of non-diabetic ZL rats(Table 1).

Vehicle-treated ZDF rats exhibited evidence of glomeru-lar hypertrophy, PAS + glomerular absorption droplets withindegenerating podocytes, mesangial expansion and glomeru-losclerosis (Table 3). PAS staining revealed minimal focalsegmental glomerulosclerosis in vehicle-treated ZDF rats,which was prevented by treatment with aleglitazar (figure 4).Glomerular desmin staining was increased in vehicle-treatedZDF rats, suggesting podocyte damage, and was largely pre-vented by aleglitazar treatment (figure 4). Quantitative imageanalysis of desmin demonstrated that both desmin markerarea and percentage desmin marker area per glomerulumwas reduced in aleglitazar-treated ZDF rats to levels com-parable to those in non-diabetic ZL rats (figure 5). Othermarkers, most notably of fibrosis and glomerular dysfunc-tion [collagen IV and fibronectin (figure 4) and synaptopodin(data not shown)] were not significantly different in any ofthe groups.

Aleglitazar prevented tubular basophilia, dilation andlumenal proteinaceous casts accompanied by interstitialfibrosis, and vacuolation associated with PAS + dropletsindicative of glycogen storage within renal distal tubularepithelial cells, all of which were present in vehicle-treatedZDF rats (Table 3). Immunohistochemistry for vimentin,TIM/KIM1, collagen IV and fibronectin revealed increasedtubular expression of these markers in vehicle-treatedZDF rats (figure 6), correlating with tubular damage, alsoobserved with haematoxylin eosin and PAS staining. These



Figure 1. Preservation of islet structure in aleglitazar-treated rats. Assessment of islet morphology, β-cell area, β-cell number and insulin stainingintensity by digital imaging quantitative fluorescence microscopy using anti-insulin or anti-glucagon antibodies, and 4′, 6-diamidino-2-phenylindole(DAPI) staining. Scale bar, 100 μm. N = 179–291 islets/group, ***p < 0.001 versus vehicle; data are shown as mean ± s.e.m.; ANOVA followed by Dunnett’stest.


DIABETES, OBESITY AND METABOLISM original articleTable 3. Summary of histopathologic findings after 13 weeks of treatment.

ZDF (N = 10)ZDF + aleglitazar(N = 10) ZL (N = 10)

PancreasIslets enlarged with irregular projections into exocrine pancreas, endocrine cells scattered within

exocrine tissue (HE)+++ + +

Vacuolated endocrine cells (HE) corresponding to reduced insulin content (insulin IHC) ++++ − −α-cells at periphery of islet (glucagon IHC) − Present Presentα-cells scattered throughout islet (glucagon IHC) Present − −Eyes (HE)Cataract, cortical ++ − −Kidneys (HE & PAS)Kidney weight relative to brain weight 2.2 2.0 1.5Glomerular hypertrophy ++ − −PAS + absorption droplets within degenerating glomerular podocytes ++ − −Glomerular mesangial expansion + − −Glomerulosclerosis, focal, segmental + − −Vacuolation and PAS + droplets within distal tubular epithelial cells (corresponding to glycogen

storage)++ − −

Tubular dilation and lumenal proteinaceous casts +++ − −Tubular basophilia +++ (+) (+)

HE, haematoxylin eosin; IHC, immunohistochemistry; PAS, periodic acid-Schiff; ZDF, Zucker diabetic fatty; ZL, Zucker lean.− = absent; (+) = minimal, individual animals only; + = minimal; ++= slight; +++= moderate; ++++= marked.

markers were decreased in aleglitazar-treated animals to levelscomparable to non-diabetic ZL rats, indicating prevention oftubulo-interstitial fibrosis and pathologic changes.

DiscussionSeveral landmark studies have demonstrated a relationshipbetween improvements in metabolic abnormalities andprevention of diabetic complications in patients with T2DM.The direct relationship between glycaemia and diabetes-relatedcomplications was shown in an epidemiologic analysis fromthe UKPDS [24], while, in Steno-2, patients with T2DMand microalbuminuria who were receiving intensive therapyfor control of glucose, lipids and blood pressure had asignificantly lower risk of CV disease, nephropathy, retinopathyand autonomic neuropathy [25]. Results from these studiessuggest that a significant portion of the beneficial effect of themultifactorial intervention on CV disease was attributable tolipid lowering on top of effective blood pressure and glycaemiccontrol [26].

Dual PPAR-α/γ agonists that combine glycaemic controlwith improvements in lipid profile may provide a multifactorialapproach to reduce CV morbidity and mortality, as well as otherdiabetes-related complications. This study demonstrated theprotective effect of aleglitazar, a novel dual PPAR-α/γ agonist,on end-organ damage and the development of T2DM-relatedcomplications in ZDF rats. In concordance with previousstudies of PPAR ligands in the ZDF model [13–18,20,21],aleglitazar significantly improved both glycaemic and lipidparameters, with concurrent preservation of pancreatic isletstructure and renal structure and function. Aleglitazar was alsoassociated with significantly increased body weight comparedwith vehicle-treated ZDF rats, a typical effect of PPAR-γ

activation in rodents and humans [27,28]. PPAR-γ agonistsincrease fat mass, associated with a reduction in circulatingfree fatty acids (FFAs) due to FFA repartitioning towards fatrather than muscle and pancreas [28]. This may contribute toincreased insulin sensitivity and reduction of glucose, lipidsand inflammatory markers, which may in turn affect organprotection.

The strengths of this study are its evaluation of multipleimmunohistochemical markers of pancreatic β-cell health,and renal glomerular and interstitial changes, in a well-established rodent disease model of T2DM. A limitation ofthis study is inability to dissect the relative contributions ofimproved glycaemic control versus improved lipids, or ofthe contributions of direct mechanisms on β-cell or renalprotection conferred by aleglitazar.

Aleglitazar prevented β-cell apoptosis in this study by up to87%. Histologic and immunofluorescent examination of thepancreas showed that aleglitazar was able to maintain β-cellarea and number to levels comparable to prediabetic ZDFrats. After 13 weeks of treatment, aleglitazar preserved isletintegrity, maintained insulin content and significantly reducedislet fibrosis. These effects are likely the result of multiple factorsleading to reduced gluco- and lipotoxicity, related to improvedinsulin sensitivity, as well as significantly reduced glucose, TGand NEFA. These beneficial effects on β-cell morphology andfunction are in agreement with previous studies of PPAR-γagonists, which showed that rosiglitazone preserves β-cell massby preventing the rise in net β-cell death [14,15]. In one study,rosiglitazone, but not metformin, preserved β-cell mass whileboth agents reduced glucose levels, suggesting a protective effectof PPAR-γ that is independent of the effects on glucose [14].

Podocyte injury plays an early and important role inprogressive renal dysfunction in the ZDF model [10,29].



Figure 2. Aleglitazar inhibits islet fibrosis and β-cell apoptosis in the pancreas. Assessment of islet fibrosis and β-cell apoptosis by digital imagingquantitative fluorescence microscopy using anti-insulin, anti-glucagon, anti-pancreatic polypeptide, anti-somatostatin antibodies and anti-fibronectinantibody (for fibrosis) and anti-insulin antibody, transferase-mediated dUTP nick-end labelling (TUNEL) and 4′, 6-diamidino-2-phenylindole (DAPI)staining (for apoptosis). Scale bar, 100 μm. Fibrosis: N = 227–562 islets/group, Apoptosis: N = 190–344 islets/group, * p < 0.05, **p < 0.01 versus vehicle;data are shown as mean ± s.e.m.; ANOVA followed by Dunnett’s test.


DIABETES, OBESITY AND METABOLISM original article

Figure 3. Aleglitazar prevents development of cataracts. Histologic characterization of the lens following haematoxylin eosin (HE) staining, showingdevelopment of cortical cataracts in untreated Zucker diabetic fatty (ZDF) rats but prevention of cataracts in aleglitazar treated ZDF rats. Scale bar, 50 μm.

Figure 4. Aleglitazar prevents glomerulosclerosis. Assessment of glomerular pathology by periodic acid-Schiff (PAS) staining and immunohistochemicalstaining. Evidence of glomerulosclerosis indicated by black arrow. Evidence of podocyte damage indicated by PAS+ absorption droplets (open arrow)and by immunohistochemical staining with anti-desmin antibody. Extent of fibrosis was assessed by immunohistochemical staining with anti-collagen IVor anti-fibronectin antibodies. Scale bar, 50 μm.

Although early podocyte damage may be influenced by highglucose and dyslipidaemia [10], agonism of PPAR-α [20,21]and PPAR-γ [16–18] may also have direct renoprotectiveactions. PPAR-α and PPAR-γ are both expressed in the kidney,with PPAR-α predominantly expressed in proximal tubules andmedullary thick ascending limbs and PPAR-γ in medullarycollecting ducts, pelvic urothelium and glomerular mesangialcells [19]. In diabetic PPAR-α-knockout mice, diabeticnephropathy is more severe than in wild-type mice [30]. Thereis substantial evidence that PPAR-α and PPAR-γ agonists havedirect effects on several critical pathways contributing to thedevelopment and progression of diabetic kidney disease, asreviewed by Thomas et al. [30]. For example, both gemfibroziland rosiglitazone have been shown to reduce albuminuria,glomerulosclerosis, tubulointerstitial expansion and collagenIV deposition in streptozotocin-induced diabetes, where theeffects on glycaemic control and lipids are minimal, suggestingdirect protective actions [30]. Other glucose-independent

renoprotective mechanisms that have been attributed to PPAR-γ agonism include reductions in blood pressure, oxidativestress, inflammation, fibrosis and positive effects on podocytefunction, including increased expression of nephrin [30].

In this study, aleglitazar prevented renal glomerular hyper-trophy, podocyte degeneration and glomerulosclerosis, as wellas tubular basophilia, dilation and lumenal proteinaceouscasts accompanied by interstitial fibrosis. Aleglitazar also pre-vented vacuolation associated with PAS + droplets, indicativeof hyperglycaemia-related glycogen storage within renal distaltubular epithelial cells that may, in part, be consistent with the‘Armanni–Ebstein phenomenon’ in diabetic states [31]. Quan-titative immunohistochemistry using an extensive panel ofmarkers provided detailed insights regarding both glomerularand tubular protection. In the glomerulus, aleglitazar normal-ized desmin levels, suggesting prevention of podocyte dysfunc-tion. Our data also indicated a tubulo-interstitial-protectiveeffect of aleglitazar, evidenced by decreases in a range of markers



Figure 5. Aleglitazar decreases glomerular desmin expression. Quantitative image analysis of glomerular desmin staining measured as average desminmarker area (μm2/glomerulum) and percentage marker area per glomerulum. N = 240 glomeruli/group; ***p < 0.001 versus Zucker diabetic fatty (ZDF);data are shown as mean ± s.e.m.; ANOVA followed by Dunnett’s test.

Figure 6. Prevention of kidney fibrosis with aleglitazar. Assessment of tubular pathology by immunohistochemical staining of kidney sections usinganti-vimentin, anti-KIM-1, anti-collagen IV and anti-fibronectin antibodies. Scale bar, 200 μm.

for tubular damage, including vimentin, TIM/KIM1, collagenIV and fibronectin. These results concur with previous findingsin studies with both PPAR-α and -γ agonists [16–18,21,32],although, to our knowledge, previous investigations have notexplored the effect of PPAR agonists on such a comprehensivepanel of markers. In the case of fenofibrate, one mechanismthat has been hypothesized to explain its effects is the inhibitionof transforming growth factor β1/Smad3 and nuclear factor κBsignalling pathways, resulting in suppression of inflammationand fibrosis [21].

The reduction in renal fibrosis may, in part, be mediated bydecreased fibrinogen, which has been observed in preclinicaland clinical studies of aleglitazar [7,33]. This concurs with

recent data suggesting that fibrinogen promotes renal fibrosisby triggering resident fibroblast proliferation [34]. These effectsare similar to observations from studies [35] of fenofibratesuch as the Fenofibrate Intervention and Event Loweringin Diabetes (FIELD) trial, in which fenofibrate reducedalbuminuria and slowed estimated glomerular filtration rateloss over 5 years, indicating a potential to retard loss of kidneyfunction despite initial reversible increases in serum creatinine[36]. Serum creatinine levels were not monitored in thepresent study; however, aleglitazar has been shown to increaseserum creatinine in clinical studies [7,37]. Notwithstanding,these effects have been shown to be fully reversible, evenat the dose of 600 μg (four times the dose currently being


DIABETES, OBESITY AND METABOLISM original articleevaluated in phase III trials), with creatinine levels returningto pretreatment levels following cessation of treatment withina 4- to 8-week follow-up period [37].

In conclusion, aleglitazar strongly improves glycaemic andlipid parameters in ZDF rats while protecting key tissues,including the pancreas, eyes and kidneys, against diabetes-associated structural and functional changes. Aleglitazar pro-tected β-cell morphology, as evidenced by effects on isletintegrity, β-cell apoptosis and islet fibrosis. Aleglitazar pre-vented the development of cataracts and prevented renalglomerular hypertrophy, podocyte degeneration, glomeru-losclerosis and tubulo-interstitial lesions. Together, these datasuggest that aleglitazar may be effective in reducing the pro-gression of T2DM, and macrovascular and microvascularcomplications of the disease.

AcknowledgementsThis study was funded by F. Hoffmann-La Roche AG. We wouldlike to thank Mathieu Brecheisen and Claudia Richardson, F.Hoffmann-La Roche, Basel, for their assistance in preparationof immunofluorescence data, as well as Emmanuelle Hainautand Andree Roeckel for conducting the pharmacology studyand analysis. Editorial support was provided by CindyMacpherson, MediTech Media, UK, funded by F. Hoffmann-LaRoche AG.

Conflict of InterestA. B., M. M., J. M., E. S., P. V. and M. B. W. designed thestudy. E.-A. A., J. M. F., A. B., S. U., P. V. and M. W. conductedexperiments. E.-A. A., A. B., J. M. F., E. S., S. U., P. V., M. W.and M. B. W. analysed the study. E.-A. A., A. B., J. M. F., M. M.,J. M., E. S., S. U., M. W. and M. B. W. wrote the manuscript.All the authors are employees of F. Hoffmann-La Roche AG.

References

1. Skyler JS, Bergenstal R, Bonow RO et al. Intensive glycemic control andthe prevention of cardiovascular events: implications of the ACCORD,ADVANCE, and VA diabetes trials: a position statement of the AmericanDiabetes Association and a scientific statement of the American Collegeof Cardiology Foundation and the American Heart Association. DiabetesCare 2009; 32: 187–192.

2. Dormandy JA, Charbonnel B, Eckland DJ et al. Secondary prevention ofmacrovascular events in patients with type 2 diabetes in the PROactiveStudy (PROspective pioglitAzone Clinical Trial In macroVascular Events): arandomised controlled trial. Lancet 2005; 366: 1279–1289.

3. Staels B, Maes M, Zambon A. Fibrates and future PPARalpha agonists inthe treatment of cardiovascular disease. Nat Clin Pract Cardiovasc Med2008; 5: 542–553.

4. Jun M, Foote C, Lv J et al. Effects of fibrates on cardiovascular outcomes: asystematic review and meta-analysis. Lancet 2010; 375: 1875–1884.

5. Benardeau A, Benz J, Binggeli A et al. Aleglitazar, a new, potent, andbalanced dual PPARalpha/gamma agonist for the treatment of type IIdiabetes. Bioorg Med Chem Lett 2009; 19: 2468–2473.

6. Hansen BC, Tigno XT, Benardeau A, Meyer M, Sebokova E, Mizrahi J. Effectsof aleglitazar, a balanced dual peroxisome proliferator-activated receptoralpha/gamma agonist on glycemic and lipid parameters in a primatemodel of the metabolic syndrome. Cardiovasc Diabetol 2011; 10: 7.

7. Henry RR, Lincoff AM, Mudaliar S, Rabbia M, Chognot C, Herz M. Effect ofthe dual peroxisome proliferator-activated receptor-alpha/gamma agonistaleglitazar on risk of cardiovascular disease in patients with type 2 diabetes(SYNCHRONY): a phase II, randomised, dose-ranging study. Lancet 2009;374: 126–135.

8. Peterson RG, Shaw WN, Neel M-A, Little LA, Eichberg J. Zucker diabeticfatty rat as a model for non-insulin-dependent diabetes mellitus. ILAR1990; 32: 16–19.

9. Etgen GJ, Oldham BA. Profiling of Zucker diabetic fatty rats in theirprogression to the overt diabetic state. Metabolism 2000; 49: 684–688.

10. Coimbra TM, Janssen U, Grone HJ et al. Early events leading to renal injuryin obese Zucker (fatty) rats with type II diabetes. Kidney Int 2000; 57:167–182.

11. Pick A, Clark J, Kubstrup C et al. Role of apoptosis in failure of beta-cellmass compensation for insulin resistance and beta-cell defects in themale Zucker diabetic fatty rat. Diabetes 1998; 47: 358–364.

12. Brussee V, Guo G, Dong Y et al. Distal degenerative sensory neuropathy ina long-term type 2 diabetes rat model. Diabetes 2008; 57: 1664–1673.

13. Shibata T, Takeuchi S, Yokota S, Kakimoto K, Yonemori F, Wakitani K.Effects of peroxisome proliferator-activated receptor-alpha and -gammaagonist, JTT-501, on diabetic complications in Zucker diabetic fatty rats. BrJ Pharmacol 2000; 130: 495–504.

14. Atkinson LL, McDonald-Dyck C, Benkoczi C, Finegood DT. Effect of chronicrosiglitazone, metformin and glyburide treatment on beta-cell mass,function and insulin sensitivity in mZDF rats. Diabetes Obes Metab 2008;10: 780–790.

15. Smith SA, Lister CA, Toseland CD, Buckingham RE. Rosiglitazone preventsthe onset of hyperglycaemia and proteinuria in the Zucker diabetic fattyrat. Diabetes Obes Metab 2000; 2: 363–372.

16. Baylis C, Atzpodien EA, Freshour G, Engels K. Peroxisome proliferator-activated receptor [gamma] agonist provides superior renal protectionversus angiotensin-converting enzyme inhibition in a rat model of type 2diabetes with obesity. J Pharmacol Exp Ther 2003; 307: 854–860.

17. Toblli JE, Ferrini MG, Cao G, Vernet D, Angerosa M, Gonzalez-Cadavid NF.Antifibrotic effects of pioglitazone on the kidney in a rat model of type 2diabetes mellitus. Nephrol Dial Transplant 2009; 24: 2384–2391.

18. Toblli JE, Cao G, Giani JF, Angerosa M, Dominici FP, Gonzalez-Cadavid NF.Antifibrotic effects of pioglitazone at low doses on the diabetic rat kidneyare associated with the improvement of markers of cell turnover, tubularand endothelial integrity, and angiogenesis. Kidney Blood Press Res 2011;34: 20–33.

19. Guan Y, Breyer MD. Peroxisome proliferator-activated receptors (PPARs):novel therapeutic targets in renal disease. Kidney Int 2001; 60: 14–30.

20. Zhao X, Li LY. PPAR-alpha agonist fenofibrate induces renal CYP enzymesand reduces blood pressure and glomerular hypertrophy in Zucker diabeticfatty rats. Am J Nephrol 2008; 28: 598–606.

21. Li L, Emmett N, Mann D, Zhao X. Fenofibrate attenuates tubulointerstitialfibrosis and inflammation through suppression of nuclear factor-kappaBand transforming growth factor-beta1/Smad3 in diabetic nephropathy.Exp Biol Med (Maywood) 2010; 235: 383–391.

22. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, TurnerRC. Homeostasis model assessment: insulin resistance and beta-cellfunction from fasting plasma glucose and insulin concentrations in man.Diabetologia 1985; 28: 412–419.

23. Abadie JM, Koelsch AA. Performance of the Roche second generationhemoglobin A1c immunoassay in the presence of HB-S or HB-C traits. AnnClin Lab Sci 2008; 38: 31–36.

24. Stratton IM, Adler AI, Neil HA et al. Association of glycaemia withmacrovascular and microvascular complications of type 2 diabetes (UKPDS35): prospective observational study. BMJ 2000; 321: 405–412.



25. Gaede P, Vedel P, Larsen N, Jensen GV, Parving HH, Pedersen O.Multifactorial intervention and cardiovascular disease in patients withtype 2 diabetes. N Engl J Med 2003; 348: 383–393.

26. Gaede P, Pedersen O. Intensive integrated therapy of type 2 diabetes:implications for long-term prognosis. Diabetes 2004; 53: S39–47.

27. de Souza CJ, Eckhardt M, Gagen K et al. Effects of pioglitazone on adiposetissue remodeling within the setting of obesity and insulin resistance.Diabetes 2001; 50: 1863–1871.

28. Knouff C, Auwerx J. Peroxisome proliferator-activated receptor-γ calls foractivation in moderation: lessons from genetics and pharmacology. EndocrRev 2004; 25: 899–918.

29. Hoshi S, Shu Y, Yoshida F et al. Podocyte injury promotes progressivenephropathy in zucker diabetic fatty rats. Lab Invest 2002; 82: 25–35.

30. Thomas MC, Jandeleit-Dahm KA, Tikellis C. The renoprotective actions ofperoxisome proliferator-activated receptors agonists in diabetes. PPARRes 2012; 2012: 456529.

31. Zhou C, Gilbert JD, Byard RW. How useful is basal renal tubular epithelialcell vacuolization as a marker for significant hyperglycemia at autopsy?J Forensic Sci 2011; 56: 1531–1533.

32. Liao J, Soltani Z, Ebenezer P et al. Tesaglitazar, a dual peroxisomeproliferator-activated receptor agonist (PPAR alpha/gamma), improvesmetabolic abnormalities and reduces renal injury in obese Zucker rats.Nephron Exp Nephrol 2010; 114: e61–e68.

33. Corsini A, Raab S, Zhang X et al. Effect of aleglitazar, a balanceddual PPARalpha/gamma agonist, on pro-inflammatory biomarkers ofatherothrombotic vascular disease and HDL functionality in non-humanprimates with metabolic syndrome. Eur Heart J 2010; 31: P1544.

34. Sorensen I, Susnik N, Inhester T et al. Fibrinogen, acting as a mitogen fortubulointerstitial fibroblasts, promotes renal fibrosis. Kidney Int 2011; 80:1035–1044.

35. Sica DA. Fibrate therapy and renal function. Curr Atheroscler Rep 2009;11: 338–342.

36. Davis TM, Ting R, Best JD et al. Effects of fenofibrate on renal functionin patients with type 2 diabetes mellitus: the Fenofibrate Interventionand Event Lowering in Diabetes (FIELD) Study. Diabetologia 2011; 54:280–290.

37. Herz M, Gaspari F, Perico N et al. Effects of high dose aleglitazar on renalfunction in patients with type 2 diabetes. Int J Cardiol 2011; 151: 136–142.


Effects of the dual PPAR-α/γ agonist aleglitazar on glycaemic control and organ protection in the...

Documents

Transcript of Effects of the dual PPAR-α/γ agonist aleglitazar on glycaemic control and organ protection in the...