1

PPAR� ligand protects during cisplatin-induced Acute Renal Failure by preventing inhibition of renal FAO and PDC activity

Shenyang Li*, Pengfei Wu§, Padma Yarlagadda*, Nicole M. Vadjunec*, Alan D. Proia#, Robert A. Harris§, Didier Portilla*

*Division of Nephrology, Departments of Internal Medicine, University of Arkansas for Medical Sciences and Central Arkansas Veterans Healthcare System 4300 W. Seventh

Street, Little Rock, Arkansas 72205, §Department of Biochemistry and Molecular Biology Indiana University, #Department of Pathology, Duke University

RUNNING TITLE: PDC activity in Toxic Acute Renal Failure

Address correspondence to: Didier Portilla, M.D. University of Arkansas for Medical Sciences Department of Medicine, Slot 501 4301 W. Markham St. Little Rock AR 72205 Office: 501-257-5833 FAX: 501-257-5827 e-mail: portilladidier@uams.edu

Copyright (c) 2003 by the American Physiological Society.

Articles in PresS. Am J Physiol Renal Physiol (November 11, 2003). 10.1152/ajprenal.00190.2003

2

ABSTRACT Previous studies demonstrated that during cisplatin-induced acute renal failure there is a

significant reduction in proximal tubule fatty acid oxidation. We now report on the

effects of PPARα ligand Wy-14,643(WY) on the abnormalities of medium chain fatty

acid oxidation and Pyruvate Dehydrogenase Complex (PDC) activity in kidney tissue of

cisplatin-treated mice. Cisplatin causes a significant reduction of mRNA levels and

enzyme activity of mitochondrial Medium Chain Acyl CoA Dehydrogenase (MCAD).

PPARα ligand Wy-14643 ameliorated cisplatin-induced acute renal failure and prevented

cisplatin induced reduction of mRNA levels and enzyme activity of MCAD. In contrast,

in cisplatin treated PPARα null mice, Wy-14,643 did not protect kidney function and did

not reverse cisplatin induced decreased expression of MCAD. Cisplatin inhibited renal

Pyruvate Dehydrogenase Complex activity prior to the development of acute tubular

necrosis, and PDC inhibition was reversed by pretreatment with PPARα agonist Wy-

14,643. Cisplatin also induced increased mRNA and protein levels of Pyruvate

Dehydrogenase Kinase-4 (PDK4), and PPARα ligand Wy-14643 prevented cisplatin

induced increased expression of PDK4 protein levels in wild type mice. We conclude that

PPARα agonists have therapeutic potential for cisplatin induced acute renal failure. Use

of PPARα ligands prevent acute tubular necrosis by ameliorating cisplatin-induced

inhibition of two distinct metabolic processes MCAD mediated fatty acid oxidation and

PDC activity.

Keywords: Peroxisome Proliferator Activated Receptor alpha ; Pyruvate Dehydrogenase

Complex ; Fatty acid oxidation; Acute Renal Failure, Cisplatin.

3

INTRODUCTION

We have demonstrated previously that during Ischemia/Reperfusion (I/R) and cisplatin induced

acute renal failure (ARF) there is a significant reduction in proximal tubule fatty acid oxidation

(FAO) which leads to the accumulation of toxic fatty acid amphiphiles (21,22,25). Furthermore,

our studies also have shown that the reduction in FAO observed in kidney tissue during ARF

occurs as a result of a direct inhibition of Peroxisome Proliferator Activated Receptor-alpha

(PPARα) DNA binding activity to its target genes and also by reduced expression of its tissue

specific co-activator PGC-1 (21). The use of PPARα ligands such as clofibrate, Wy-14,643, and

etomoxir significantly ameliorated proximal tubule cell injury and renal function during I/R

injury. These results underscore the importance of PPARα mediated regulation of FAO as a

potential therapeutic target that could be used in various models of ARF.

Previous studies in muscle tissue have documented abnormalities in glucose homeostasis

during ARF (4). The potential presence of similar metabolic abnormalities in kidney tissue

during ARF has not been explored. PPARα has been recently shown to modulate glucose

metabolism via regulation of pyruvate dehydrogenase complex (PDC) activity in skeletal muscle

tissue (9). Regulation of the activity of PDC is an important component of the regulation of

glucose homeostasis (7). PDC is a multienzyme complex that catalyzes the conversion of

pyruvate to acetyl coenzyme A. This reaction is the first irreversible step in the oxidation of

carbohydrate-derived carbon, and regulates the entry of carbohydrate into the tricarboxylic acid

cycle. PDC activity is regulated by reversible phosphorylation and dephosphorylation reactions

catalyzed by an intrinsic kinase and phosphatase (7,9,12,20). Phosphorylation of E1 catalytic

subunits of PDC by pyruvate dehydrogenase kinase (PDK) causes inactivation of the enzyme,

whereas pyruvate dehydrogenase phosphatase (PDP) removes phosphate and returns the enzyme

to its active form (PDCa) (20). The relative activities of the PDPs and PDKs determine the

proportion of the complex in its active form. There are four known isoforms of PDKs, which

differ significantly in their tissue distribution, specific activity and sensitivity to their effectors

(24, 27). To date PDK4 and PDK2 are the two most abundant PDK isoforms expressed in

rodents and human kidney tissue (28,29,31).

4

In the present study we examined the effects of PPARα ligand Wy-14,643 on

cytoprotection in the toxic model of cisplatin induced ARF. More specifically, we examined the

effects of cisplatin on renal function and histological parameters of proximal tubule injury in the

absence and presence of WY. To address the potential mechanisms of protection by PPARα

ligands we also examined the effects of this ligand on MCAD-mediated fatty acid oxidation,

PDC complex activity and mRNA and protein levels of PDK isoforms in kidney tissue. Our

studies corroborate previous observations that cisplatin induced ARF is accompanied by

significant attenuation of mRNA, and enzyme activity of mitochondrial FAO enzyme Medium

Chain Acyl CoA Dehydrogenase (MCAD)( 21). The use of PPARα ligand Wy-14,643

ameliorated both proximal tubule cell injury and kidney function in the PPARα wild type mice

but not in the PPARα null mice. Increased mRNA and enzyme activity of renal FAO enzyme

MCAD accompanied this protective effect. In addition, we observed that cisplatin induced a

profound and sustained inhibition of PDC activity detected on the second day after cisplatin

injection, and prior to the development of acute tubular necrosis (ATN). PDC inhibition was

accompanied by increased expression of protein and mRNA levels of renal mitochondrial PDK4

isoform. Pretreatment with PPARα ligand Wy-14,643 reversed the inhibition of PDC activity by

cisplatin and prevented the observed increased expression of kidney PDK4 protein levels in

cisplatin treated mice. Our results demonstrate for the first time in kidney tissue that cisplatin

induced ARF is accompanied by the presence of significant metabolic changes, which include

inhibition of MCAD-mediated FAO, as well as the inhibition of glucose metabolism via

inhibition of PDC activity. Pretreatment with PPARα ligand Wy-14,643 reversed the inhibition

of MCAD and PDC activities, and correlated with protection of kidney function. Further studies

are needed to delineate the nephron segments, and the mechanisms by which PPARα activation

modulate substrate oxidation, and protect kidney function during ARF.

5

METHODS

In vivo model of cisplatin induced acute renal failure: Male sv129 mice approximately

8-10 weeks old, weighting 25 to 30 grams were assigned to treatment groups (eight

animals per group). Animals received single intraperitoneal doses of vehicle (saline) or

cisplatin (20 mg/kg body weight). Following treatment the animals were sacrificed and

the kidney tissue was frozen in liquid nitrogen for RNA or protein isolation. Mice were

housed in a temperature and light controlled environment and provided food and water.

Pelleted mouse chow was prepared containing either zero or 0.1 % (wt/wt) Wy-14,643.

To investigate the effects of PPAR� activation, mice were fed the Wy-14,643 containing

diet for 2 weeks. For histopathological evaluation kidneys were collected in 10% neutral

buffered formalin. Blood Urea Nitrogen and creatinine were measured by an enzymatic

colorimetric assay as previously described (34,35).

Histopathological alterations: We evaluated histopathologic alterations in the kidneys 4

days after the mice were treated with cisplatin with or without Wy-14,643. Kidneys were

bisected, fixed in 3.7% phosphate-buffered neutral formaldehyde, dehydrated with serial

alcohols, and embedded in paraffin. We stained three-micrometer-thick paraffin sections

with hematoxylin and eosin and a periodic acid-Schiff method (17). The twelve

morphological features described by Solez and coworkers (26) were evaluated in a

masked fashion: leukocyte accumulation in the vasa recta: tubular necrosis (presence of

necrotic cells, apparently denuded areas of tubular basement membrane, or ruptured

tubular basement membranes); tubular regeneration; mitotic figures in tubular cells;

dilatation of Bowman’s space with retraction of the glomerular tuft (“acute glomerular

ischemia”); loss of PAS-positive tubular brush border; “tubularization” of the parietal

epithelium of Bowman’s capsule; tubular casts; interstitial inflammation; interstitial

edema; tubular dilatation; and prominence of the juxtaglomerular apparatus. We graded

the morphological changes on a scale from 0 to 2 where 0 = none; 1/2 = minimal; 1 =

mild; 1 1/2 = moderate; and 2 = marked (11).

6

RNA isolation. Mice were sacrificed following previously described experimental

conditions, and the kidney tissue was rapidly snap-frozen in liquid nitrogen and stored at

-75�C. Total RNA was extracted with TRIZOL Reagent (Invitrogene Life Technologies)

according to the manufacturer’s directions.

RT-PCR. Total RNA extract was treated with 1 U of RQ1 RNase-free DNase (Promega)

per microgram of total RNA, at 37�C for 1 h. Reverse transcription was performed at

42�C for 50 min in a total volume of 20 µl containing 5 µg RNA, 0.5 µg of oligo (dT)12-

18, 200 U of SUPERSCRIPT II RNase H- reverse transcriptase (RT) (Invitrogene Life

Technologies). Subsequently, RT was inactivated by incubation at 70�C for 15 min,

followed by treatment with 1.2 U of RNase H at 37�C for 30 min. PCR was performed

with 1/20 of the RT reaction in a total volume of 50 µl using the Taq DNA Polymerase

(Invitrogene). To control for the generation of PCR products due to residual

contamination of genomic DNA, an aliquot of RNA, not treated with RT, was also tested

in parallel. Amplification was performed using the following primer pairs for 25 cycles

(denaturation at 94�C for 30 s, annealing at 57�C for 25 sec, and extension at 72�C for 30

sec); PDk4 sense (5’-ACCGCATTTCTACTCGGATG-3’) and antisense

(5’-CCTCCTCGGTCAGAAATCTT-3’) primers, GAPDH ense

(5’-AACTTTGGCATTGTGGAAGG -3’) and antisense (5’-

AGATCCACGACGGACACATT-3’) primers, MCAD (medium chain acetyl-Coenzyme

dehydrogenase) sense (5’- GTACCCGTTCCCTCTCATCA-3’) and antisense (5’-

CGTGCCAACAAGAAATACCA-3’) primers. The sense primer was end-labeled using

[γ-32P] ATP (Perkin-Elmer.) and T4 polynucleotide kinase (New England Biolabs). Five

microliters from the PCR reactions were resolved on a 4% acrylamide gel. The gels were

dried, analyzed on a 445 SI PhosphoImager with ImageQuantNT (Molecular Dynamic),

and then subjected to autoradiography. Results are presented as the ratio of the signal for

PDK4 or MCAD band to that of the GAPDH signal.

7

Western blots: PDK2 and PDK4 protein levels were estimated by Western Blot analysis

of mouse kidney tissue extracts obtained from several experimental conditions, as

described previously (31).

PDC activity. Mice were sacrificed after experimental conditions, and the kidney tissue

was rapidly snap-frozen in liquid nitrogen and stored at -75�C. Frozen kidney tissue was

extracted as described previously by Goodwin et al (5) with modification. In brief, 50 mg

tissue was homogenated in 0.5 ml of homogenization buffer (see below) with Teflon

homagenizer (7-10 strokes). The resulting homogenate was subsequently centrifuged at

10,000 g, 4 °C for 10 min to remove tissues debris and the supernatant was removed,

placed on ice and enzyme activity was assayed immediately. To determine the fraction of

PDC in the active form (PDCa), tissue was extracted using a homogenization buffer

containing 30 mM K-HEPES (pH 7.5), 3% Triton-100, 5 mM EDTA, 10 mM EGTA, 10

mM DCA, 50 mM KF, 2% bovine serum, 0.01 mM TPCK, 10 µg/ml Trypsin inhibitor, 1

µM, and 5 mM DTT. Under these conditions both pyruvate dehydrogenase kinase and

phosphatase were inhibited. To determine the total activity of PDC (PDCt), tissue was

extracted under conditions where PDC phosphatase (PDP) was stimulated (and also with

addition of exogenous PDP) and PDC kinase (PDK) was inhibited, using a

homogenization buffer consisted with 30 mM K-HEPES (pH 7.5), 3 % Triton-100, 2%

bovine serum, 0.01 mM TPCK, 10 µg/ml Trypsin inhibitor, 1 µM leupeptin, 5 mM DTT,

20 mM DCA, 15 mM MgCl2, and 1 mM CaCl2. Enzyme activity assays of PDC complex

in above extracts was determined as described previously (17) with a SPECTRAmax®

190 Micro plate Spectrophotometer (Molecular Devices, Sunnyvale, CA).

MCAD activity: MCAD activity assay was performed in kidney tissue extracts following

protocol previously described by Lehman et al (14) with minor modifications. Kidneys

were homogenized in cold 100 mM Hepes with 0.1 mM EDTA (pH 7.6). The

homogenates were then centrifuged briefly at 4 �C and the MCAD activity was measured

immediately on the supernatant at 37 �C. 2 µl of supernatant was added to 200 µl of

reaction solution of 100 mM Hepes buffer (pH 7.6), 0.1 mM EDTA, 200 µM ferricenium

hexafluorophosphate, 0.5 mM sodium tetrathionate and 50 µM octanoyl-CoA. The

8

absorbance decrease at 300 nm in ferricenium ion was determined by SPECTRAmax

microplate spectrophotometer (Molecular Devices) over the initial 60-second period. The

values were corrected by subtracting the background absorbance of a tissue blank,

measured in the absence of octanoyl-CoA in the reaction solution. Results are presented

as Mean � SE of MCAD activity relative to that obtained for control mice, which was set

arbitrary as 100 % in each experiment and were calculated from at least three

independent experiments

Statistics analysis. Results are presented as the mean � standard error (SE). Statistical

analysis was performed using unpaired Student t tests. A p value of � 0.05 was

considered to be statistically significant.

9

RESULTS

Effect of PPARα ligand WY on body weight in PPAR� wild-type (+/+) and null (-/-)

mice

Mean body weight was significantly lower in PPARα wild-type mice fed Wy-14,643-

containing diet when compared with wild-type mice that received a similar regular chow

diet. The PPARα wild-type mice lost 10% of their body weight after receiving WY diet

for two weeks. On the contrary, the PPARα null mice fed Wy-14,643-containing diet

gained 3% of their body weight when compared to null mice receiving regular chow diet.

The loss of body weight in PPARα wild-type mice fed with Wy-14,643 is consistent with

previous reports, and is likely related to ligand mediated increased fatty acid oxidation by

liver and kidney tissue (1).

PPARα ligand WY protects kidney function during cisplatin-induced ARF in

PPAR� wild-type mice, but not in PPAR� null (-/-) mice

Kidney function was monitored for 4 days after IP injection of saline or cisplatin, by

measuring blood urea nitrogen (BUN) and serum creatinine. Figure 1A and 1B present

the changes on BUN and creatinine seen in PPARα wild type mice treated with and

without WY ligand in the absence (Control) and presence of cisplatin.

Comparison of the renal function between PPARα wild-type mice fed for two weeks with

either regular diet or 0.1% WY containing diet did not show differences in BUN and

creatinines when treated with single injection of vehicle alone (saline IP). Values for this

control group C as shown in Figure 1A and 1B, were not significantly different at day 1

(BUN 24 mg/dl, creatinine 0.2 mg/dl) versus day 4 (BUN 30 mg/dl, creatinine 0.2

mg/dl). Mice treated with regular diet and cisplatin developed ARF at day 3 and 4 (BUN

went up from 27 to 264 mg/dl and creatinine went up from 0.2 to 1.5 mg/dl at day 4 after

single cisplatin injection). The group of PPARα wild type mice that received the WY diet

and cisplatin did not develop significant ARF when compared to mice treated with

cisplatin alone (BUN went from 23 on day 1 to 40 mg/dl at day 4, and creatinine was

unchanged at 0.3 mg/dl at day 4.

In contrast PPARα null mice (-/-), pretreated with WY diet prior to cisplatin

administration did not exhibit any protection in renal function at day 4 following cisplatin

10

injection, BUN went from 30 mg/dl on day 1 to 250 mg/dl day 4, and creatinine went

from 0.3 at day 1 to 1.7 mg/dl at day 4). This is shown in Figures 1C and 1D. These

observations suggest that the protective effect of PPARα ligand WY on renal function

seen only on PPARα wild type mice was dependent on having an intact PPARα gene.

PPARα null mice treated with PPARα ligand exhibit less FAO due to their lack of

response to PPARα ligands (13).

Histopathological alterations of mouse kidneys subjected to cisplatin induced ARF

in the absence and presence of PPARα ligand WY

Results of histopathologic abnormalities in the kidneys are given in table 1. Dilatation of

Bowman’s space with retraction of the glomerular tuft (“acute glomerular ischemia”),

“tubularization” of the parietal epithelium of Bowman’s capsule, interstitial

inflammation, interstitial edema, tubular dilatation, and prominence of the

juxtaglomerular apparatus were not seen in any of the kidneys (results not shown). The

control mouse kidneys exhibited no pathological changes, whereas those from mice

treated with Wy-14,643 consistently had a few necrotic epithelial cells (score = 0.5) and

scattered tubules with loss of their PAS-positive brush border (score = 0.5). Three of the

four kidneys from mice treated with Wy-14,643 also had occasional mitotic figures

evident in the tubular epithelium (score = 0.5). All four mice treated with cisplatin alone

had moderate to marked tubular necrosis and loss of the epithelial cell brush border,

numerous casts, and three of the four had regenerative tubular changes manifest as

flattened, hypereosinophilic epithelial cells. There was a minimal infiltrate of leukocytes

in the vasa recta of one mouse and a mild infiltrate in one mouse (data not shown), but

none had tubular epithelial cell mitoses. All four mice treated with Wy-14,643 together

with the cisplatin were readily distinguishable from the mice treated with cisplatin alone

since they had only minimal epithelial cell necrosis, no or only minimal loss of the brush

border, and no casts. The kidneys from mice receiving cisplatin and Wy-14,643 were

distinguishable histologically from those treated with only Wy-14,643 by the absence of

epithelial cell mitotic figures. An example of a kidney from a control (untreated) mouse

and typical mice treated with Wy-14,643, cisplatin, and cisplatin+Wy-14,643 are shown

11

in Figure 2. Also shown in this figure is the study of the effects of WY ligand on PPARα

null mice. A histologic section from a PPARα -/- mouse demonstrates subtle loss of PAS-

positive brush border in few tubules, while a histological section of a PPARα -/- mouse

treated with WY and cisplatin shows necrosis of the tubular epithelium. Again these

histopathological alterations are consistent with the results shown on Figure 1 on renal

function. Altogether our data suggest that PPARα ligand WY protects PPARα wild type

mice and not PPARα null mice from cisplatin induced ARF likely by a mechanism

dependent on having an intact PPARα gene.

Effects of PPARα ligand WY on the expression of MCAD

To determine the mechanisms by which PPARα ligand protects renal function and

ameliorates histological parameters in cisplatin induced ATN, we examined the effects of

cisplatin and WY ligand on renal MCAD activities of both PPARα wild type and null

mice. As shown in figure 3A-B-C by representative autoradiograms and phosphor-

imaging analysis, cisplatin caused a progressive decline on the mRNA expression of

MCAD, a rate limiting enzyme in the metabolism of medium chain fatty acids by

mitochondria in kidney tissue. At day 4 of renal failure there was a 63% inhibition of

MCAD expression when compared to control mice (P � 0.005). In PPARα null mice (see

Figure 3B) cisplatin at day 4 had a similar effect on MCAD mRNA levels. At day 4

following cisplatin injection, there was 65% inhibition of MCAD activity (P<0.005).

Pretreatment with WY ligand, reversed cisplatin-induced inhibition of renal MCAD

activity in wild type mice, but this effect was not observed in the PPARα null mice (see

figure 3C). These results are similar to our previously published observations in which

the use of PPARα ligands etomoxir and clofibrate resulted in up-regulation of renal FAO

during IRI. Therefore, again these observations further corroborate the cytoprotective role

of PPARα ligands on renal function during ARF and further underscore the importance of

mitochondrial FAO on the preservation of structure and function of the proximal tubule

during ARF.

12

Effects of PPARα ligand WY on the enzyme activity of MCAD

Since PPAR� ligand WY prevented cisplatin induced reduction of MCAD mRNA levels

on the PPAR� wild type mice but not on the PPAR� null mice, we next examined the

effects of cisplatin and WY on MCAD enzyme activity. As shown on figure 4 cisplatin

at day 4 caused a profound decline in the enzyme activity of renal MCAD on both

PPARα wild type and null mice. Pretreatment with WY prevented cisplatin induced

reduction of renal MCAD activity in PPARα wild type mice. In contrast to the effects of

WY on PPARα wild type mice, pretreatment with WY did not affect cisplatin induced

reduction of MCAD activity in PPARα null mice These data further supports our

previous results showing that the protective effect of PPAR� ligand Wy-14,643 on

mRNA levels and activity of FAO enzyme MCAD was dependent on having an intact

and functionally active PPAR� gene.

Cisplatin induced inhibition of PDC activity is prevented by PPAR� ligand WY

Previous studies support the notion that ARF is accompanied by abnormal glucose

oxidation in muscle tissue(4,16). There is no experimental evidence that renal pathways

involved in glucose homeostasis are affected during ARF. Therefore we have examined

first the effect of cisplatin induced ARF on PDC activity. Cisplatin led to a rapid

inhibition of mitochondrial PDC activity detected within the first 48 hours after cisplatin

administration (p<0.001) (results not shown). This effect was sustained and by day 4 as

shown in Figure 5A, and 5C there was more than 70% inhibition of mitochondrial PDC

activity by cisplatin (*** p<0.0005). Pretreatment of PPARα wild type mice with WY

prevented cisplatin induced inhibition of PDC activity as shown in Figure 5C. In addition

WY treatment alone although decreased the fraction of both PDCa and PDCt in kidney

tissue did not significantly affect PDC activity measured as percentage of PDCa/PDCt as

shown in Figure 5C. These studies suggest that in addition to preventing cisplatin

inhibition of FAO, WY also to some extent restores the capacity for the kidney to oxidize

glucose. Therefore our studies further support the role of PPARα as a master regulator of

substrate oxidation in kidney tissue.

13

Effects of PPAR� ligand and cisplatin on the expression of PDK4 mRNA levels

There are recent reports that PPAR� may play a role in the regulation of glucose

metabolism (33). Previous studies have shown that increased expression and activity of

pyruvate dehydrogenase kinase accounts for the inhibition of PDC activity in

pathophysiological states such as starvation, diabetes and hyperthyroidism (8,30). To

determine the mechanisms by which PDC activity is inhibited by cisplatin we examined

the effects of cisplatin and WY ligand on PDK4 mRNA levels of both PPARα wild type

and null mice. As shown in figure 6A-B-C by representative autoradiograms and

phosphor-imaging analysis, cisplatin caused a progressive increase on the mRNA

expression of PDK4mRNA levels. As shown in Figures 6A and C in PPARα wild type

mice at day 4 of renal failure there was a profound up-regulation (8-fold) of PDK4

mRNA levels when compared to control mice (** p � 0.005). In PPARα null mice (see

Figure 6B and 6C) cisplatin at day 4 had only a modest effect causing a 2.8 fold

stimulation of PDK4 mRNA levels. Pretreatment with WY ligand also led to a fivefold

increased on PDK4 mRNA levels in wild type mice, but this effect was again

significantly lower (only twofold increase) in the PPARα null mice (see figure 6C). We

next examined the effects of WY on cisplatin-induced increased expression of PDK4. As

shown in Figure 6A and 6C cisplatin was unable to further increase PDK4 mRNA

expression when compared to cisplatin treated mice. In contrast, in WY-fed PPARα null

mice cisplatin significantly increased PDK4 mRNA levels in kidney tissue of PPARα

null mice when compared to control, cisplatin or WY treated null mice.

Effects of cisplatin and WY on the expression of renal PDK4 protein

Since changes in PDK4 mRNA levels induced by WY or cisplatin did not correlate with

changes on PDC activity, in the next series of studies we investigated the effects of

cisplatin and WY on the protein levels of PDK4 using western blot analysis of mouse

kidney tissue. As shown in Figure 7 by a representative autoradiogram and by

densitometric quantification, cisplatin induced a profound upregulation (8-fold) of PDK4

protein levels in wild type mice. In WY treated wild type mice, as well as in WY treated

mice that received cisplatin, PDK4 protein was not expresssed. We conclude that

increased protein levels of PDK4 induced by cisplatin in the wild type mice represent

14

one of the mechanisms by which PDC activity is inhibited, and that the inhibition of the

expression of PDK4 protein by PPARα ligand WY helps preserve PDC activity during

ARF.

15

DISCUSSION

In the present studies we have extended previous observations made in the IRI model

of acute renal failure. We demonstrated herein: 1) similar to IRI, cisplatin induced ARF is

accompanied by reduced expression and inhibition of enzyme activity of mitochondrial FAO

enzyme MCAD; 2) pretreatment of PPAR� wild type mice with a specific PPAR� ligand

protected renal function, and this protection in renal function was substantiated by

histopathological examination that demonstrated, amelioration of cisplatin-induced acute

tubular necrosis, loss of brush border, and cast formation; 3) pretreatment with PPAR�

ligand also reversed reduced expression and enzyme activity of kidney MCAD induced by

cisplatin; 4) the persistent inhibition of FAO, and lack of protection of renal function in

PPAR� null mice treated with cisplatin and WY, further underscore the importance of

having an intact PPAR� gene on the observed cytoprotective effect of PPAR� ligands during

ARF (21,22). These results corroborate the importance of having a preserved renal PPAR�

modulated FAO pathway in the observed protection of renal function during ARF.

Our histopathological analysis of mice treated with PPARα ligand WY also revealed

the presence of proliferation of few isolated proximal tubular cells, and minimal loss of brush

border, effects that were not accompanied by significant changes in renal function. The

significance of these findings is unknown, but perhaps represent a toxic effect of this ligand

on kidney tissue. The use of more potent PPARα ligands that could be given at the time of

cisplatin injection in future studies should allow us to avoid these side effects. In addition our

studies cannot completely rule out the possibility of a potential effect of the PPARα ligand on

pre-conditioning, an effect which can protect kidney tissue from further injury.

Since the observed protection in kidney function in PPARα wild type mice that

received a PPARα ligand, likely involves modification of various metabolic pathways, other

than FAO in kidney tissue, we also examined the effect of PPARα ligand on PDC actvity.

Abnormalities in glucose metabolism have been previously described in skeletal muscle of

rats subjected to ARF (4). Carbohydrate metabolism is deranged in ARF as a result of

impaired insulin mediated actions. Previous studies suggest that insulin resistance accounts

16

for one of the hormonal abnormalities involved in the deranged muscle protein catabolism in

ARF. Those studies have demonstrated decreased insulin mediated glucose uptake, glycogen

synthesis and glucose oxidation in the muscle of acutely uremic rats (16). Clark and Mitch

(4) have suggested that the abnormal net protein degradation in ARF may be linked to the

altered carbohydrate metabolism and might be a consequence of defective glucose oxidation.

In the present study we examined the effects of cisplatin induced acute renal failure on

pyruvate dehydrogenase complex activity, the enzyme that catalyzes the conversion of

pyruvate to acetyl CoA in the mitochondria. Our studies demonstrate the inhibition of PDC

activity as early as 48 hrs after cisplatin injection, a time point where we could not detect

significant changes in BUN or creatinine, or histopathological parameters of tubular injury.

In addition, prior administration of PPARα ligand WY reversed cisplatin induced inhibition

of PDC activity. The significance of this observation is high, since preservation of PDC

activity perhaps represents an important metabolic adaptation of distal nephron segments to

the lack of energy production during ARF.

The differential regulation of pyruvate dehydrogenase kinase and phosphatase is the

key to the overall regulation of the activity of the pyruvate dehydrogenase complex. Three

specific serine residues in each alpha subunit of the E1 domain namely, site 1 Ser 264, site 2

Ser 271, and site 3 Ser203, are subject to ATP-dependent phosphorylation and inactivation

by PDKs (7). Pyruvate dehydrogenase phosphatase dephosphorylates these three serine

residues and reactivates PDC (3,15). Mammalian PDK isoenzymes differ in their catalytic

activity, responsiveness to modulators like NADH and acetyl CoA and tissue specific

expression. PDK1 is present mostly in heart, whereas PDK2 is found in most tissues. PDK3

is predominantly expressed in testis, whereas heart, skeletal muscle and kidney have the

highest amount of PDK4 (2,6,29). In pathophysiological states such as starvation, diabetes

and hyperthyroidism (8,28,32) PDC activity is reduced and mRNA and protein levels of

PDK4 are increased. Our studies showing a temporal relationship between cisplatin induced

inhibition of PDC activity and increased expression of renal PDK4 protein suggest a similar

mechanism of regulation of PDC activity. In addition our data also suggest that increased

phosphorylation of PDC by free PDK4 could be a potential mechanism that accounts for the

observed inhibition of PDC activity during ARF. However, our current data cannot

conclusively establish a cause-effect relationship between these two metabolic abnormalities,

17

since we also observed decreased PDC activity in cisplatin treated PPARα null mice and

these mice did not express PDK4 protein when treated with cisplatin. Therefore, other

potential pathways by which PDC activity could be down-regulated during cisplatin could

include decreased expression of PDPs, or changes on the levels of NAD, NADH. In fact

recent studies suggest that starvation and streptozotocin-induced diabetes cause decreases in

Pyruvate Dehydrogenase Phosphatase-2 (PDP2) mRNA abundance, PDP2 protein amount

and PDP activity in rat heart and kidney. Re-feeding and insulin treatment effectively

reversed these effects of starvation and diabetes, respectively. Those findings indicate that

opposite changes in expression of specific PDK and PDP isoenzymes contribute to

hyperphosphorylation and therefore inactivation of the PDC in heart and kidney during

starvation and diabetes (19). More studies will be needed to completely clarify all the

mechanisms that contribute to renal PDC inhibition during ARF.

Our studies also suggest that by day 4 cisplatin inhibits gene expression of enzymes

associated with FAO and pyruvate oxidation, two important metabolic substrates for the

generation of acetyl CoA, a three carbon molecule which is utilized by normal kidney tissue

via the TCA cycle for the generation of energy in the form of ATP. Metabolic abnormalities

resulting from the inhibition of these two important sources of energy, are likely to further

contribute to the catabolic state of ARF, and perhaps explain the metabolic need for

increased protein degradation during ARF. Therefore, increased oxidation of amino acids by

kidney tissue, as a maladaptive mechanism, is likely to provide the substrate needed for the

generation of acetyl CoA and energy production via the TCA cycle in the setting of

decreased energy production. Of interest, a recent study (10) has demonstrated that inhibition

of protein degradation using a specific proteasome inhibitor results in cytoprotection during

ischemic-reperfusion injury to the kidney.

In the present study we did not examine whether PPARα ligand had an effect on the

apoptotic response to cisplatin. In preliminary studies not shown here, we only found a

positive tunnel staining in the kidneys of PPARα wild type mice treated with cisplatin which

represented less than 5% of cellular injury seen in the cortico-medullary junction, with

necrosis representing the major form of cell death. These similar findings have been

previously reported by other investigators (18). Of interest, our most recent studies done in

proximal tubule cells in culture have shown that PPARα ligand inhibits caspase 3 activation

18

by cisplatin, and this results in amelioration of proximal tubule cell death (23), suggesting

that PPARα-mediated inhibition of the caspase cascade might represent also a mechanism of

cytoprotection.

In conclusion, our results clearly indicate the ability of PPARα ligand to ameliorate

cisplatin induced ARF. In addition, our studies further support the role of inhibition of

MCAD-mediated fatty acid oxidation and inhibition of PDC activity in the pathogenesis of

cisplatin induced ARF. Further studies are needed to localize the nephron segment(s) where

these abnormalities in PDC and fatty acid oxidation take place in the kidney, to further

examine the cellular mechanisms of substrate inhibition, and to determine whether PPARα

ligands have similar effects on these cellular pathways, as our studies suggest it is the case

in this in vivo study.

19

Figure legends

Figure 1. Effects of PPARα ligand WY on renal function in PPARα wild-type and null

mice after single dose of cisplatin. Mice were fed with either a regular diet or a diet

containing 0.1% Wy-14,643 (WY) for two weeks prior to cisplatin administration, as

described in methods. For wild-type mice, BUN and creatinine levels were measured

every day for a period of four days. For null mice, BUN and creatinine levels were

measured only at day 4 after either saline (control) or cisplatin IP injection. Bars

correspond to mean � SE of at least four independent experiments under each condition.

Figure 1A and 1B show the changes on BUN and creatinine levels respectively after

cisplatin or saline administration in PPARα wild type mice (+/+). Figure 1C and 1D

show the comparison between PPARa +/+ and PPARa -/- mice, of changes on BUN (C)

and creatinine (D)levels, at day 4 after cisplatin or saline injection � p � 0.05, � �p �

0.0005 when animals were compared to control in unpaired student t test, † indicates p �

0.005 when comparisons were made between PPARα wild type and null mice in unpaired

student t test.

Figure 2. Histopathological alterations in mice treated with PPARα ligand and

cisplatin : PPARα +/+: A histological section from a control (untreated) kidney (A)

discloses a normal glomerulus and tubules, while a typical kidney from a mouse treated

with Wy-14,643 (B) demonstrates an epithelial cell mitosis (large arrowhead) and loss of

the PAS-positive brush border in some tubules (arrow). A typical kidney from a mouse

20

treated with cisplatin (C) shows extensive necrosis of the tubular epithelium (�); many of

the tubules shown on the left side of figure C are necrotic and similar in appearance to

that denoted with the arrow. A section from a typical cisplatin+Wy-14,643-treated

kidney (D) has only a rare necrotic epithelial cell (large arrowhead) and loss of the PAS-

positive brush border in some tubules (arrow).

PPARα -/- : A. Histological sections from a PPARα null mice treated with WY also

demonstrates loss of the PAS-positive brush border in few tubules(see arrow). B. Kidney

sections from PPARα -/- mice treated with cisplatin and WY showed marked tubular

necrosis of the tubular epithelium (*). All sections were stained with periodic acid-Schiff

reagent and photographed at the same magnification (magnification bar = 50 �m).

Figures 3. Effect of cisplatin and PPARα ligand on the mRNA levels of MCAD :

Figure 3A shows representative autoradiographs of RT-PCR analysis performed using

total RNA isolated from kidney tissue of PPARα wild type mice who received a normal

or WY containing diet, and that were treated with either saline or cisplatin. Figure 3B

shows representative autoradiographs of RT-PR analysis from kidney tissue of PPARα

null mice treated with cisplatin, WY and WY+cisplatin.

Figure 3C shows quantitative densitometry of mRNA levels determined using labeled

primers for MCAD and GAPDH as described in methods. Bars represent means � SE

MCAD mRNA levels. Data was obtained from at least four independent experiments �

indicates p � 0.0005 compared to control in unpaired student t test. † indicates p � 0.005

21

when comparisons were made between PPARα wild type and null mice in unpaired

student t test.

Figure 4. Effect of cisplatin and PPARα ligand on the enzyme activity of MCAD:

Mice fed with either a regular or WY diet received IP dose of saline (groups control and

WY) or cisplatin (groups cisplatin and WY+ cisplatin). Following treatment the animals

were sacrificed at day 4. MCAD enzyme activity was determined as described in

methods. Bars represent means � SE MCAD enzyme activity levels when compared to

control conditions (renal MCAD activity of 1). Data was obtained from at least four

independent experiments. � indicates p � 0.0005 compared to control, † indicates p �

0.005 when comparison was made between PPARα wild type and null mice in unpaired

student t test.

Figure 5. Effect of cisplatin and WY on renal PDC activity. PPARα wild type mice

were fed with either a regular or a WY diet, and then they received single IP dose of

saline (groups control and WY) or cisplatin (groups cisplatin and WY+cisplatin).

Following treatment the animals were sacrificed at day 4. Values for the active form of

PDC (PDCa) are shown in Figure 5A, total PDC (PDCt) are shown in figure 5B and

percentage of PDCa/PDCt are shown in figure 5C. Data was obtained from at least 4

independent experiments. � indicates p � 0.05, �� indicates p � 0.005 and *** p � 0.0005

22

when comparison was made between PPARα wild type and null mice in unpaired student

t test.

Figure 6. Effect of cisplatin and WY on PDK4 mRNA levels:

Figure 6A shows a representative autoradiographs of RT-PCR analysis performed using

total RNA isolated from kidney tissue of PPARα wild type mice who received a normal

or WY containing diet, and that were treated with either saline or cisplatin. Analysis was

done after 4 days of receiving saline (control), or after 1,2,3, and 4 days of receiving

cisplatin. For quantitative RT-PCR measurements mRNA levels were determined using

labeled primers for PDK4 and GAPDH. Figure 6B shows representative autoradiographs

of RT-PCR analysis from kidney tissue of PPARα null mice treated with either a regular

or WY containing diet, then were given a single IP dose of saline (groups control and

WY) or cisplatin (groups cisplatin and WY+cisplatin) Following treatment the animals

were sacrificed at day 4. Figure 6C shows quantitative densitometry of PDK4 mRNA

levels in the control group(saline), cisplatin day 4, WY, and WY+cisplatin day 4 in both

PPARα wild type and null mice. Bars represent means � SE PDK4 mRNA levels. Data

was obtained from at least four independent experiments � indicates p � 0.05 , **

indicates p � 0.005 compared to control in unpaired student t test. † indicates p � 0.05

when comparisons were made between PPARα wild type and null mice in unpaired

student t test.

23

Figure 7. Effect of cisplatin and WY on PDK4 protein levels.

PPARα wild type mice fed either a regular or WY- diet and then received single IP dose

of saline(groups control and WY) or cisplatin(groups cisplatin and cisplatin+WY).

Following treatment the animals were sacrificed at day 4. Figure 7A shows a

representative western blot analysis of mouse kidney tissue treated with saline (control)

or cisplatin for 4 days using PDK4 antibody. Figure 7B shows the quantitative changes in

relative amount of PDK4 protein levels under conditions given in figure 7A. Bars

correspond to PDK4 protein levels � SE of at least four independent experiments. **

indicates p � 0.0005, compared to control in unpaired student t test.

24

ACKNOWLEDGEMENTS

We acknowledge Dr Robert L. Safirstein for his advice and for sharing his preliminary

data from microarray analysis. This work was supported by National Institutes of Health

Grant PO1-DK58324-01A1 and a VA Merit Award to Dr. Didier Portilla, and NIH grants

DK47844 and DK61594 to Dr Robert A. Harris.

25

Table 1. Grading of lesions in kidneys after cisplatin administered with or without Wy-14,643 pretreated

Tubular Tubular Brush Border Group Necrosis Regneration Mitoses Loss Casts

Control 0 0 0 0 0 0 0 0 0 0

Mean 0 0 0 0 0 SD 0 0 0 0 0

Wy-14,643 treated 0.5 0 0.5 0.5 0

0.5 0 0.5 0.5 0 0.5 0 0.5 0.5 0 0.5 0 0.5 0.5 0

Mean 0.5 0 0.5 0.5 0 SD 0 0 0 0 0

Cisplatin treated 2 0 0 2 2 2 0.5 0 2 2 2 0.5 0 2 2 1.5 1 0 1.5 1.5 Mean 1.9 0.5 0 1.9 1.9 SD 0.3 0.4 0 0.3 0.3 Cisplatin + Wy-14,643 0.5 0 0 0.5 0 treated 0.5 0 0 0.5 0 0.5 0 0 0 0 0.5 0 0 0 0 Mean 0.5 0 0 0.3 0 SD 0 0 0 0.3 0 Grading system is that of Solez et al. (13) and scoring system is that of Kelleher etal. (14) where 0 = none, 0.5 = minimal, 1 = mild, 1.5 = moderate, and 2 = marked.

26

FIGURES

Figure. 1

Figure 1

27

Figure 2

Figure 2

28

Figure 3

29

Figure 4

30

Figure 5

Figure 5

31

Figure 6

32

Figure 7

33

REFERENCES

1. Akiyama TE, Nicol CJ, Fievet C, Staels B, Ward JM, Auwerx J, Lee SS,

Gonzalez FJ, and Peters JM. Peroxisome proliferator-activated receptor-alpha

regulates lipid homeostasis, but is not associated with obesity: studies with

congenic mouse lines. J Biol Chem. 276:39088-39093,2001.

2. Bowker-Kinley MM, Davis WI, Wu P, Harris RA, and Popov KM. Evidence for

existence of tissue-specific regulation of the mammalian pyruvate dehydrogenase

complex. Biochem J 329:191-196, 1998.

3. Chen G, Wang L, Liu S, Chuang C, and Roche TE. Activated function of the

pyruvate dehydrogenase phosphatase through Ca2+-facilitated binding to the

inner lipoyl domain of the dihydrolipoyl acetyltransferase. J Biol Chem

271:28064-28070, 1996.

4. Clark AS and Mitch WE. Muscle protein turnover and glucose uptake in acutely

uremic rats. Effects of insulin and the duration of renal insufficiency. J Clin Invest

72:836-845, 1983.

5. Goodwin GW, Zhang B, Paxton R, and Harris RA. Determination of activity and

activity state of branched-chain alpha-keto acid dehydrogenase in rat tissues. Methods

Enzymol 166:189-201, 1988.

6. Gudi R, Bowker-Kinley MM, Kedishvili NY, Zhao Y, and Popov KM. Diversity

of the pyruvate dehydrogenase kinase gene family in humans. J Biol Chem.

270:28989-28994, 1995.

34

7. Harris RA, Bowker-Kinley MM, Huang B, and Wu P. Regulation of the activity

of the pyruvate dehydrogenase complex. Adv Enzyme Regul 42:249-259, 2002.

8. Holness MJ, Smith ND, Bulmer K, Hopkins T, Gibbons GF, and Sugden MC.

Evaluation of the role of peroxisome-proliferator-activated receptor alpha in the

regulation of cardiac pyruvate dehydrogenase kinase 4 protein expression in

response to starvation, high-fat feeding and hyperthyroidism. Biochem J 364:687-

694, 2002.

9. Huang B, Wu P, Bowker-Kinley MM, and Harris RA. Regulation of pyruvate

dehydrogenase kinase expression by peroxisome proliferator-activated receptor-alpha

ligands, glucocorticoids, and insulin. Diabetes 51:276-283, 2002.

10. Itoh M, Takaoka M, Shibata A, Ohkita M, and Matsumura Y. Preventive effect of

lactacystin, a selective proteasome inhibitor, on ischemic acute renal failure in

rats. J Pharmacol Exp Ther 298:501-507, 2001.

11. Kelleher SP, Robinette JB, Miller F, and Conger JD. Effect of hemorrhagic reduction

in blood pressure on recovery from acute renal failure. Kidney Int. 31: 725-730, 1987

12. Kolobova E, Tuganova A, Boulatnikov I, and Popov KM. Regulation of pyruvate

dehydrogenase activity through phosphorylation at multiple sites. Biochem J 358:69-

77, 2001.

13. Lee SS, Pineau T, Drago J, Lee EJ, Owens JW, Kroetz DL, Fernandez-Salguero

PM, Westphal H, and Gonzalez FJ. Targeted disruption of the alpha isoform of

the peroxisome proliferator-activated receptor gene in mice results in abolishment

of the pleiotropic effects of peroxisome proliferators. Mol Cell Biol. 15:3012-

3022, 1995.

35

14. Lehman TC, Hale D E, Bhala A, and Thorpe C. Anal Biochem 186:280–284, 1990.

15. Liu S, Baker JC, and Roche TE. Binding of the pyruvate dehydrogenase kinase to

recombinant constructs containing the inner lipoyl domain of the dihydrolipoyl

acetyltransferase component. J Biol Chem 270:793-800, 1995.

16. May RC, Clark AS, Goheer A Mitch WE Specific defects in insulin mediated

muscle metabolism in acute uremia Kidney Int 28:490-497, 1985.

17. McManus J.F.A. 1948. Histological and histochemical uses of periodic acid. Stain

Technol. 23:99-108.

18. Megyesi J, Safirstein RL, and Price PM. Induction of p21WAF1/CIP1/SDI1 in

kidney tubule cells affect the course of cisplatin-induced acute renal failure. J

Clin Invest 101:777-782, 1998.

19. Huang B., Wu P., Popov KM., Harris R.A. Starvation and diabetes reduce the

amount of pyruvate dehydrogenase phosphatase in rat heart and kidney Diabetes

vol.2 , 1371-1376, 2003.

20. Patel MS and Korotchkina LG. Regulation of mammalian pyruvate dehydrogenase

complex by phosphorylation: complexity of multiple phosphorylation sites and

kinases. Exp Mol Med 33:191-197, 2001.

21. Portilla D, Dai G, McClure T, Bates L, Kurten R, Megyesi J, Price P, and Li S.

Alterations of PPARalpha and its coactivator PGC-1 in cisplatin-induced acute renal

failure. Kidney Int 62:1208-1218, 2002.

22. Portilla D, Dai G, Peters JM, Gonzalez FJ, Crew MD, and Proia AD. Etomoxir-

induced PPAR�-modulated enzymes protects during acute renal failure. Am J

Physiol Renal Physiol 278:F667-F675, 2000.

36

23. Portilla D, Kurten R, and Kaushal GP. PPAR� but not PPAR� ligands protect

against cisplatin induced renal tubular epithelial cell injury. (Abst) J Am Soc

Nephrol 13:138A, 2002.

24. Roche TE, Baker JC, Yan X, Hiromasa Y, Gong X, Peng T, Dong J, Turkan A, and

Kasten SA. Distinct regulatory properties of pyruvate dehydrogenase kinase and

phosphatase isoforms. Prog Nucleic Acid Res Mol Biol 70:33-75, 2001.

25. Schonefeld M, Noble S, Bertorello AM, Mandel LJ, Creer MH, and Portilla D.

Hypoxia-induced amphiphiles inhibit renal Na+, K(+)-ATPase. Kidney Int. 49:1289-

1296, 1996.

26. Solez K, Morel-Maroger, L, and Sraer J-D. The morphology of “acute tubular

necrosis” in man: analysis of 57 renal biopsies and a comparison with the glycerol

model. Medicine 58:362-376, 1979.

27. Sugden MC, Bulmer K, and Holness MJ. Fuel-sensing mechanisms integrating lipid

and carbohydrate utilization. Biochem Soc Trans 29:272-278, 2001.

28. Sugden MC, Bulmer K, Gibbons GF, and Holness MJ. Role of peroxisome

proliferator-activated receptor-alpha in the mechanism underlying changes in

renal pyruvate dehydrogenase kinase isoform 4 protein expression in starvation

and after refeeding. Arch Biochem Biophys 395:246-252, 2001.

29. Sugden MC, Holness MJ, Donald E, and Lall H. Substrate interactions in the short-

and long-term regulation of renal glucose oxidation. Metabolism 48:707-715, 1999.

30. Wu P, Inskeep K, Bowker-Kinley MM, Popov KM, and Harris RA. Mechanism

responsible for inactivation of skeletal muscle pyruvate dehydrogenase complex in

starvation and diabetes. Diabetes.48:1593-1599, 1999.

37

31. Wu P, Blair PV, Sato J, Jaskiewicz J, Popov KM, and Harris RA. Starvation increases

the amount of pyruvate dehydrogenase kinase in several mammalian tissues. Arch.

Biochem Biophys 381:1-7, 2000.

32. Wu P, Sato J, Zhao Y, Jaskiewicz J, Popov KM, and Harris R A. Starvation and

diabetes increase the amount of pyruvate dehydrogenase kinase isoenzyme 4 in rat

heart. Biochem J. 329:197-201, 1998.

33. Xu J, Xiao G, Trujillo C, Chang V, Blanco L, Joseph SB, Bassilian S, Saad MF,

Tontonoz P, Lee WN, and Kurland IJ. Peroxisome proliferator-activated receptor

alpha (PPARalpha) influences substrate utilization for hepatic glucose production.

J Biol Chem 277:50237-50244, 2002.

34. Crocker C.L. Rapid determination of urea nitrogen in serum or plasma without

deproteinization Am. J. Med Technol 33:361, 1967.

35. Heinegard D., Tiderstrom G. Determination of serum creatinine by a direct

colorimetric method Clin. Chim. Acta 43:305, 1973.