Toxicology and Applied Pharmacology 252 (2011) 201210

Contents lists available at ScienceDirect

Toxicology and Applied Pharmacology

j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y t a a p

Identication of novel indicators of cyclosporine A nephrotoxicity in a CD-1 mouse model

Sin O'Connell, Craig Slattery, Michael P. Ryan, Tara McMorrow
UCD School of Biomolecular and Biomedical Research, UCD Conway Institute, University College Dublin, Ireland

a r t i c l e

i n f o

a b s t r a c t
The calcineurin inhibitor cyclosporine A (CsA) is a widely used immunosuppressive agent. However, nephrotoxicity is a serious side effect observed in patients which limits clinical use of CsA. CsA nephrotoxicity is associated with tubulointerstitial injury progressing to nephropathy. This is typically diagnosed by invasive renal biopsy and is often only detected when the disease process is well advanced. Therefore identication of novel, early indicators of CsA nephrotoxicity could be clinically advantageous. This study aimed to establish a murine model of CsA nephrotoxicity and to identify urinary proteins that may indicate the onset of CsAinduced nephropathy using 2-D gel electrophoresis. CsA nephrotoxicity was induced in CD-1 mice by daily CsA administration for 4 weeks. By week 4, elevated serum creatinine and proteinuria were observed after CsA treatment indicating signicant renal dysfunction. Decreased cadherin-1, increased -smooth muscle actin and broblast specic protein 1 in kidney tissue indicated disruption of normal tubular architecture. Alterations in podocin and uromodulin were also observed which may indicate damage to other segments of the nephron. Proteomic analysis of urine identied a number of differentially regulated proteins that may be involved in early CsA nephropathy including cadherin 1, superoxide dismutase and vinculin. These ndings suggest novel mechanisms of CsA nephrotoxicity and identify novel potential markers of the disease. 2011 Elsevier Inc. All rights reserved.

Article history: Received 6 December 2010 Revised 16 February 2011 Accepted 17 February 2011 Available online 23 February 2011 Keywords: Cadherin 1 (CDH-1) Connective tissue growth factor Fibroblast specic protein 1 Transforming growth factor beta 1 and 2-DE

Introduction The use of calcineurin inhibitors (CNIs), cyclosporine A (CsA) and tacrolimus has revolutionised solid organ transplantation over the last 30 years (Gaston, 2009). However, clinical use of CsA is associated with both acute and chronic nephrotoxicity which is a major limiting factor in its use. While alternative therapeutics have been sought, CNIs remain our most effective and widely used immunosuppressants (Gaston, 2009). While acute CsA nephrotoxicity is managed clinically through careful monitoring of renal function and appropriate regimen adjustment, the balance between preventing immunologic allograft loss and the management of chronic CNI nephrotoxicity (particularly CsA nephropathy) is still a major issue in renal transplantation (Bestard et al., 2005). Early diagnosis of nephropathy can greatly improve patient prognosis. However the initial stages of CsA nephropathy are largely asymptomatic making early diagnosis difcult. Therefore identication of novel, early disease indicators is currently a major research focus. Current diagnostic techniques employed to detect CsA nephropathy are inadequate. The primary method is estimation of glomerular ltration rate (eGFR) (Cockcroft and Gault, 1976). However this
Corresponding author at: School of Biomolecular and Biomedical Science, Conway Institute of Biomolecular & Biomedical Research, University College Dublin, Dublin 4, Ireland. Fax: +353 1 7166456. E-mail address: [Link]@[Link] (T. McMorrow). 0041-008X/$ see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/[Link].2011.02.015

technique is limited since eGFR varies greatly both between patients, and over time within a patient (Kwong et al., 2010). eGFR is the net result of the complex interaction of multiple factors including age, blood pressure and other diseases. Many of these factors are variable and so compensation can often occur, leading to stabilisation of eGFR, effectively masking early renal functional decline. These factors mean that eGFR can be a very insensitive indicator of renal damage. Determination of serum creatinine and blood urea nitrogen (BUN) are also used to estimate renal function although these tests can be insensitive and have poor diagnostic value (Dieterle et al., 2010). Measurement of albumin and/or protein in the urine to detect renal damage may be more sensitive than the determination of eGFR on an individual basis, especially in early disease states. However biopsy studies have clearly shown that intra-renal pathology often occurs well in advance of microalbuminuria (Rastaldi et al., 2002). Furthermore, the relationship between proteinuria and CsA nephropathy is complex (Li and Yang, 2009) limiting its power as an early indicator of CsA nephrotoxicity. CsA nephropathy is characterised by tubulointerstitial brosis (TIF), tubular vacuolisation, glomerulosclerosis, and arteriolopathy (Hara et al., 2009). Of these, TIF is thought to be the primary mechanism driving the progression of CsA nephropathy (Bobadilla and Gamba, 2007). TIF is characterised by the gradual loss of tubular epithelial cells, and progressive accumulation of broblasts and myobroblasts (-smooth muscle actin (-SMA) and broblast specic protein-1 (FSP-1) positive cells). The accumulation of myobroblasts results in excessive

202

S. O'Connell et al. / Toxicology and Applied Pharmacology 252 (2011) 201210

production and deposition of extracellular matrix (ECM) in the tubulointerstitium (Lan, 2003). Previous studies from this research group have demonstrated the direct toxic effects of CsA on renal tubular epithelial cells in vitro (Kiely et al., 2003; McMorrow et al., 2005; Slattery et al., 2005; Martin-Martin et al., 2010). It is clear that some of the downstream pathogenic effects of CsA are mediated by transforming growth factor beta 1 (TGF-1) and it's downstream effector connective tissue growth factor (CTGF) , (Grotendorst, 1997; Gupta et al., 2000; Shihab et al., 2003; Slattery et al., 2006). Conversely, the TGF-1 antagonist bone morphogenetic protein 7 (BMP-7) is downregulated in rat models of CsA nephrotoxicity (Tuglular et al., 2004). However the precise mechanism of CsA-induced nephrotoxicity remains to be fully elucidated. The aims of this study were to establish an in vivo mouse model of CsA nephropathy and evaluate a number of putative nephrotoxicity markers. These markers included TGF-, CTGF and BMP-7 which have been proposed in published studies, across a number of other models of nephrotoxicity, as being signicantly involved in disease progression and may be suitable as indicators of toxicity or therapeutic targets (El Chaar et al., 2007; Dudas et al., 2009; Phanish et al., 2010). TGF-, CTGF and BMP-7 were therefore evaluated in a CsA nephrotoxicity model, and a high-throughput proteomic screening methodology was utilised to identify novel, early indicators of CsA nephropathy. 2-DE has been widely used to identify potential urinary markers of disease in a range of settings including hepatocellular carcinoma (Jia et al., 2010) and juvenile idiopathic arthritis (Rosenkranz et al., 2010). Identication of discriminating urinary proteins in the current model may lead to novel markers of CsA nephrotoxicity and may also help to further elucidate the mechanisms underlying CsA nephrotoxicity. Methods Animal treatment Male CD-1 mice weighing 2535 g (68 week) were housed in the UCD biomedical facility according to ethical and legal guidelines in a temperature and light controlled environment. All experiments were approved by the UCD Animal Research Ethics Committee (P04-07). Government approval (B100/3539) was also granted by the Irish Department of Health under section 11 of the Cruelty to Animals Act. For the duration of the experiment mice were maintained on a low sodium diet (Harlan, UK Ltd.). CsA (Sigma-Aldrich, Cat no. C1832) was made up as a 1 mg/ml stock solution in olive oil (Sigma-Aldrich). CsA was administered by intraperitoneal injection (15 mg/kg/day), daily for 1 week or 4 weeks, as indicated. Control mice received 1 ml/kg of vehicle (olive oil) by intraperitoneal injection, daily for 1 week or 4 weeks, as indicated. These doses and time points were based upon data from other relevant in vivo studies (Thomas et al., 1998; Clarke and Ryan, 1999; Yang et al., 2002; Ling et al., 2003). Using this protocol ensured that by 4 weeks of CsA treatment signicant nephropathy had developed. The 1 week CsA group was utilised to examine early toxic effects of CsA before overt histological alterations had manifested (Ling et al., 2003; Chaaya et al., 2011). Mice had free access to food and water throughout the experiments. At the end of each treatment period mice were housed in group metabolism cages for 24 h for urine collection which was then frozen at 80 C. After this 24 h period mice were euthanised and blood samples obtained by cardiac puncture. Kidneys were collected for analysis of histology, gene and protein expression. Half a kidney was used for RNA isolation (Trizol method, T9424 Sigma-Alrdich) and half for protein isolation (RIPA buffer method Sigma-Aldrich, R0278). Half of the other kidney was snap frozen in liquid N2 and the other half was xed in neutral buffered formalin. Each treatment group i.e. 1 week control; 1 week CsA; 4 week control and 4 week CsA contained 6 mice. Urine was collected from each group of 6 mice and processed for analysis as one pooled sample for each group.

Renal function and histology Renal function was assessed by determination of serum creatinine and urinary protein (proteinuria). Serum creatinine was measured using the Quantichrom Creatinine Assay Kit (Cat no. DICT-500, Bioassay systems assay kit), according to the manufacturer's protocol. This colorimetric assay is based on the improved Jaffe method. Proteinuria was measured using the Bradford Assay for assessing total protein quantities in a biological sample (Bradford, 1976). Urine samples were normalised for urinary output. Half a kidney was xed in neutral buffered formalin, parafn embedded and sectioned at 5 m. After de-waxing, gross renal histology was examined using standard haematoxylin and eosin (H&E) staining (Sigma-Aldrich). Collagen staining of sections was performed using Masson's Trichrome stain (Sigma-Aldrich). Sections were stained using an automated slide stainer (Leica autostainer XL). Quantitative polymerase chain reaction (PCR). Total RNA was isolated using the trizol method from half a kidney stored at 80 C in RNAlater (Ambion Cat no. AM7020) according to the manufacturer's protocol. One microgram of total RNA was used to synthesise cDNA. A Real-Time PCR TaqMan assay was used to quantify the relative expression levels of genes of interest and has been described previously (Feighery et al., 2008). Briey, cDNA was amplied on the ABI 7900HT Sequence Detection System at default thermal cycling conditions: 2 min @ 50 C, 10 min @ 95 C for enzyme activation and then 40 cycles of 15 s @ 95 C for denaturation and 1 min @ 60 C for annealing and extension. Results were analysed using the delta Ct method of analysis. Primer sequences for murine TGF-1 were designed in the Conway Institute genomics core facility and synthesised by Applied Biosystems. Primer specicity was assessed by nBLAST in the NCBI database. Name: Mouse TGF-1 NM_011577.tx-529 Forward Sequence: AATTCCTGGCGTTACCTTGGT Name: Mouse TGF-1 NM_011577.tx-600 Reverse Sequence: GACGTCAAAAGACAGCCACTCA Commercially available gene expression assays were used for mouse CTGF (Mm00515790_g1), Cadherin 1 (Mm00486906_m1), SMA (Mm00426835_g1) and FSP-1 (Mm00803372_g1) all from Applied Biosystems.

Quantitative enzyme-linked immunosorbant assay (ELISA). A T GF 1 ELISA was used to determine the effect CsA had on urinary TGF1 protein levels. This was done according to the manufacturing company's (Cat no. DB100B, R&D systems) protocol. The specicity and sensitivity of the assay was assessed using 5 ng of TGF-1 as a positive control and sterile water as a negative control.

Western blot analysis. Total kidney protein was isolated using the RIPA buffer method (Sigma-Aldrich, R0278) from renal homogenates according to the manufacturer's protocol. The SDS-PAGE procedure used was that of Laemmli (1970). Expression levels of renal proteins following CsA treatment was determined by Western blot and has been described previously (McMorrow et al., 2005; Slattery et al., 2005; Feighery et al., 2008). Proteins of interest were detected using the following antibodies according to the manufacturer's protocol, mouse monoclonal anti--SMA antibody (A2547 Sigma-Aldrich), mouse monoclonal anti-Cadherin 1 (CDH-1) antibody (61081, BD biosciences) or rabbit anti-CTGF polyclonal antibody (A gift from Dr. John Crean, UCD Conway Institute). In urinary western blots colloidal coomassie blue stain (Sigma-Aldrich, B8522) which stains all proteins within a polyacrylamide gel was used as a loading control to ensure equal loading of proteins.

S. O'Connell et al. / Toxicology and Applied Pharmacology 252 (2011) 201210

203

2-dimensional gel electrophoresis (2-DE). Identi cation of novel proteins in urine that may be involved in CsA induced renal brosis was assessed using 2DE and mass spectrometry which has been described previously (Feighery et al., 2008). Briey, urine samples were prepared for 2-DE by centrifugation at 12,000 g for 5 min to remove any potential contaminants. The samples were then resuspended in a 1:1 volume of lysis buffer (9.5 M Urea; 2% CHAPS; 0.8% Pharmalyte pH 310; 1% DTT). Protein quantitation was measured using the Bradford assay prior to rst dimension isoelectric focusing (IEF) (Bradford, 1976). The pooled urine samples were divided into 500 g aliquots for subsequent 2-DE, mass spectrometry and database analysis. This analysis was repeated at least ve times. Confocal microscopy. CDH1 localisation was detected by indirect immuno-uorescence using an Alexa-546 conjugated secondary antibody (Invitrogen) (excitation 543 nm/emission 570 nm). CDH-1 localisation was detected by confocal microscopy (Zeiss LSM 510 Meta confocal microscope using 63 objective lens). Statistical analysis. Statistical analyses were performed using GraphPad Prism 4.0. Data was analysed by one-way analysis of variance (ANOVA) and multiple comparisons between control and treatment groups were made using the Bonferroni post-test. A student t-test was used for assessing statistical differences between two groups. A probability of 0.05 or less was deemed statistically signicant. Results were expressed as the mean SEM. The following scheme was used throughout the work; *p b 0.05, **p b 0.01, ***p b 0.001, compared to time matched control. Results Renal functional alterations in CD-1 mice following CsA treatment Serum creatinine and proteinuria were assessed to determine the effects of CsA on renal function (Table 1). After 1 week CsA treatment, no signicant change in either parameter was detected. After 4 weeks CsA treatment, signicant increases in both serum creatinine levels and in urinary protein levels were observed compared to timematched controls. Taken together these results suggested signicant decline in renal function of CsA treated mice after 4 weeks. Histopathological and gene expression alterations in CD-1 mice following CsA treatment H&E and Masson trichrome staining of kidney tissue was performed to assess the effect of CsA treatment on renal histology (Fig. 1A). After 1 week CsA treatment no signicant changes to gross renal histology were detectable. However, after 4 weeks CsA treatment, signicant histopathological alterations were observed compared to time-matched controls. These effects were particularly marked in glomerular and tubular compartments. Some renal tubules appeared disorganised,

irregular in shape and separated from neighbouring tubules. Masson trichrome staining indicated marked accumulation of collagen within the tubulointerstitium after 4 weeks of CsA administration. Increased interstitial collagen has been observed in other models of TIF (Murphy et al., 1999; Kattla et al., 2008). The expression of broblast markers FSP-1 and -SMA were assessed (Figs. 1C and D). Signicant increases in FSP-1 mRNA levels were detected at both 1 week and 4 weeks in CsA treated mice compared to time-matched controls. -SMA protein levels were signicantly increased in CsA treated mice at both 1 week and 4 weeks compared to time-matched controls. However, increased -SMA mRNA levels were only detected after 4 weeks CsA treatment. This may be due to altered protein processing and/or post-translational modications that may have resulted in increased protein without a detectable change in mRNA. Taken together these results provide strong evidence of TIF in CsA treated mice by week 4. The effect of CsA on mediators of TIF in CD-1 mice The effects of CsA on TGF-1, CTGF and BMP-7 in CD-1 mice were assessed as they are known mediators of TIF in other models (Ling et al., 2003; Shihab et al., 2003; Kattla et al., 2008; Veerasamy et al., 2009) (Fig. 2). CsA treatment caused signicant increases in TGF-1 mRNA levels in whole kidney RNA after 1 week and 4 week treatments (Fig. 2A i). TGF-1 protein was also signicantly upregulated in both whole kidney and urine (Figs. 2A ii and iii). However, in whole kidney protein, increased TGF-1 was only detected after 4 weeks, whereas urinary TGF- 1 levels were signicantly elevated in both 1 and 4 week CsA treated groups. The effect of CsA on CTGF mRNA levels in whole kidney RNA were similar to those on TGF-1 with signicant increases detected at both 1 and 4 weeks in the CsA groups (Fig. 2B i). In contrast however, increased CTGF protein levels were detected in whole kidney at both 1 and 4 weeks but urinary CTGF protein levels were not signicantly elevated until week 4 of CsA treatment (Figs. 2B ii and iii). BMP-7 is a negative regulator of brosis. A signicant reduction in BMP-7 gene expression in whole kidney RNA was detected following a 4 week CsA treatment (Fig. 2C). Together, these ndings provide further evidence that TIF is well established in CsA treated CD-1 mice by 4 weeks. Furthermore, these results indicate that the brotic process has initiated by the 1 week time point. Considering these results, it was decided that urine analysis would focus on the one week treatment group since identication of signicantly altered proteins at this stage would be more attractive as early nephrotoxicity indicators. Evaluation of podocin and uromodulin as markers of CsA nephrotoxicity Previous studies suggest that increased urinary levels of podocin, a slit-diaphragm protein, may be an early marker of glomerular damage in vivo (Sato et al., 2009). Therefore we examined podocin in the present model. Urinary podocin was signicantly increased following 1 week CsA treatment (Fig. 3A i). There was a corresponding decrease in podocin expression in whole kidney lysates after 1 week of CsA treatment (Fig. 3A ii). Uromodulin (also known as TammHorsfall protein) is normally detected at high levels in urine and has been proposed to play a protective role in the kidney (Prajczer et al., 2010). Urinary uromodulin levels were signicantly decreased following 1 week CsA treatment (Fig. 3B i). Coomassie blue stain (Sigma-Aldrich, B8522) was used as a loading control to ensure equal loading of urinary proteins. High throughput identication of novel urinary indicators of CsA nephropathy Urine collected from the 1 week CsA treatment group and timematch controls were analysed by 2-D gel electrophoresis, mass

Table 1 Functional alterations in CD-1 mice following CsA treatment. CD-1 mice were treated intraperitoneally with 15 mg/kg/day cyclosporine A or 1 ml/kg vehicle control for one week or four weeks. (A) Serum creatinine and proteinuria were used as indicators of renal function. Results are expressed as mean S.E.M. N = 6 for each treatment group. 1 week control Proteinuria (mg/24 h) Serum Cr (mg/dl) 3.24 0.19 0.43 0.02 1 week CsA 3.94 0.41 0.34 0.04 4 week control 2.86 0.04 0.56 0.04 4 week CsA 5.34 0.26 0.74 0.03

*Indicates statistically signicant difference to time matched control. * p b 0.05, *** p b 0.001.

204

S. O'Connell et al. / Toxicology and Applied Pharmacology 252 (2011) 201210

Fig. 1. Histopathological alterations in CD-1 mice following CsA treatment. CD-1 mice were treated with either CsA (15 mg/kg/day) or vehicle (1 ml/kg/day) control for one week or four weeks. (A) Kidney tissue was xed, sectioned and stained with H&E (1A; top panels) or with Masson's Trichrome stain (1A, bottom panels). Representative images are shown (n = 6 for each group) (magnication 200). Renal tubules appeared disorganised, irregular in shape and separated from neighbouring tubules (top row indicated by black arrows). Masson trichrome staining indicated marked accumulation of collagen (blue colour) within the tubulointerstitium after 4 weeks of CsA administration (bottom row indicated by black arrows). RNA and protein were extracted from each kidney. (B) FSP-1 and -SMA gene expression levels were examined by quantitative PCR. (C) -SMA protein expression was examined by Western blot analysis. Results are expressed as mean S.E.M (n = 6). *Indicates statistically signicant difference to time matched control (*p b 0.05, **p b 0.01, ***p b 0.001).

spectrometry and database analysis to identify discriminating proteins in the urine of CsA treated mice. This method allows for the rapid identication of multiple proteins within a biological sample. Representative micropreparative gels are shown in Figs. 4A and B. Circles indicate protein spots that were excised and analysed by mass spectrometry (MS). Fifteen distinct proteins were identied by MS. A number of the peptides identied included, cadherin 1 (CDH-1) and serum albumin precursor (highlighted in Fig. 4C). All of the protein spots identied and the associated database analysis are shown in Table 2. CsA-induced alterations in CDH-1 expression in CD-1 mice Following proteomic investigation CDH-1 was chosen for further analysis. CDH-1 was of interest as it was the only renal tubule specic marker identi ed from the proteomic analysis. Since tubular dysfunction is a major mechanistic feature of CsA nephrotoxicity, urinary CDH-1 was of particular interest as a potential early indicator

of tubular dysfunction. Since urinary CDH-1 levels were signicantly increased, the effect of CsA on CDH-1 whole kidney protein levels was examined to determine whether the alterations in urinary CDH-1 correlated with CDH-1 levels in the kidney. Signicant downregulation of CDH-1 was observed by Western blotting at both 1 and 4 weeks (Fig. 5A). This effect was also observed by immunouorescent microscopy where a marked reduction in the number of tubular epithelial cells expressing CDH-1 was observed following CsA treatment (4 weeks) (Fig. 5B). CDH-1 staining on a 5 M kidney cross-section appeared as punctuate red staining located at the junctions, on the apical sides of the epithelial cells in the tubules. This tubular staining of CDH-1 was markedly decreased following CsA treatment and is indicated by the white arrows. Discussion Despite its nephrotoxic side-effects, CsA avoidance regimens result in increased acute rejection and these avoidance strategies have been

S. O'Connell et al. / Toxicology and Applied Pharmacology 252 (2011) 201210

205

Fig. 2. Effect of CsA on mediators of TIF in CD-1 mice. CD-1 mice were treated with CsA or vehicle alone for one week or four weeks. Kidneys from each group were harvested, and RNA and protein were extracted. TGF-1 gene and protein expression was examined by quantitative PCR and ELISA (2A i and ii). TGF-1 protein expression in urine was examined by ELISA (2A iii). CTGF gene and protein expression was examined by quantitative PCR and Western blotting (B i and ii respectively). CTGF protein expression in urine was examined by Western blot analysis (B iii). Coomassie blue stain was performed to ensure equal protein loading. BMP-7 gene expression was examined by quantitative PCR (C). Results are expressed as mean S.E.M of a sample size of six mice per treatment group. *Indicates statistically signicant difference to time matched control (*p b 0.05, **p b 0.01).

discarded (Reis, 2010). Therefore, further elucidation of the mechanisms underlying CsA nephrotoxicity and identication of novel, early indicators of CsA nephropathy are of clinical importance. The complex and multi-factorial nature of CsA-induced nephrotoxicity makes early detection of this disease particularly challenging. In an effort to address this critical issue we established a murine model of CsA nephrotoxicity in CD-1 mice. Initially, we veried that CsA did in fact induce renal dysfunction in CD-1 mice. We examined a range of well established

patho-physiological features associated with CsA nephropathy. Dieterle et al. (2010) have demonstrated that total proteinuria is a specic marker of glomerular injury and that in conjunction with serum creatinine measurements are excellent indicators of renal dysfunction. In the current study, signicantly elevated proteinuria and serum creatinine levels were evident by 4 weeks CsA treatment. Histopathological analysis revealed signicant tubular atrophy and interstitial collagen accumulation in CsA treated mice by week 4. Fibroblast accumulation is also a

206

S. O'Connell et al. / Toxicology and Applied Pharmacology 252 (2011) 201210

Fig. 3. Evaluation of podocin and uromodulin as markers of CsA nephrotoxicity. CD-1 mice were treated with CsA or vehicle alone for one week. Urine collected from mice at the end of the treatment period. Kidneys from each group were harvested, and RNA and protein were extracted. Podocin protein levels were examined by Western blot analysis in urine (3A i), and in whole kidney protein lysates (3A ii). Uromodulin protein levels were examined by Western blot analysis in urine (3B i). Results are expressed as mean S.E.M of a sample size of six mice per treatment group. *Indicates statistically signicant difference to time matched control (*p b 0.05, **p b 0.01).

feature of TIF (Qi et al., 2006) and this was reected by increased levels of FSP-1 and -SMA protein in whole kidney lysates. For the purposes of this study putative nephrotoxicity markers were designated as those identied in the literature as having a signicant role in nephrotoxicity in a number of in vitro and in vivo models of renal disease. These included TGF-1, CTGF and BMP-7 (Zeisberg et al., 2003; Xu et al., 2009). The probrotic cytokine TGF-1 is a major contributor to TIF and nephropathy (Ling et al., 2003). Increased TGF-1 mRNA and protein expression after CsA treatment suggests a role for TGF-1 in the current model, and this is in agreement with previous studies (Shihab

et al., 2003, 2006; Lloberas et al., 2008). However, despite some initially promising results in experimental models of CsA nephropathy (Ling et al., 2003; El Chaar et al., 2007), TGF-1 blockade has not yet translated into an effective therapeutic strategy in human patients. Both CTGF and BMP-7 are downstream modulators of TGF-1 signalling (Veerasamy et al., 2009; Phanish et al., 2010). CTGF is pro-brotic and is upregulated in many models of TIF (Ito et al., 1998; Gupta et al., 2000; Yokoi et al., 2002; Wang and Hirschberg, 2003). In contrast, BMP-7 is anti-brotic and is decreased in models of renal brosis (Turk et al., 2009). The effects of CsA on CTGF and BMP-7 in the current study would likely favour the

S. O'Connell et al. / Toxicology and Applied Pharmacology 252 (2011) 201210

207

Fig. 4. High throughput identication of novel urinary indicators of CsA nephropathy. CD-1 mice were treated with either cyclosporine A or vehicle control for one week at which time urine was collected. Urinary proteins were separated by 2DE using a pH 310 gradient (4A control, 4B CsA). Circles indicate expression and locations of spots identied by mass spectrometry and UniProt database analysis (see Table 2 for details). Images are representative of ve analytical gels performed in duplicate. 2-DE images of selected upregulated and down-regulated proteins following CsA treatment (C). Spots of interest are indicated by the arrows. The images on the top represent urine samples from control mice subjected to 2-DE. The images on the bottom represent the corresponding spot location in urine from mice treated with CsA.

pro-brotic response and further underline the establishment of a brotic model after 4 weeks CsA treatment. Combining the functional and histological observations with the results of gene expression studies, we concluded that 4 weeks of CsA treatment was sufcient to induce alterations consistent with CsA nephropathy. Critically, a subset of these effects were detectable by 1 week of CsA treatment suggesting that the mechanisms driving the pathophysiological response were initiating at that early time point. It was therefore determined that the 1 week time point would by the focus of further proteomic analysis of urine to identify potential markers of CsA nephrotoxicity. In this study signicant changes in urinary podocin were observed even after a short CsA treatment period, prior to gross morphological changes and severe nephrotoxicity. This is a signicant and promising nding as there is emerging evidence from several experimental models and human diseases that podocyte damage and loss, can contribute to the initiation and progression of renal disease (Macconi et al., 2009; Sato et al., 2009; Wang et al., 2009). Podocytes are specialised epithelial cells covering the basement membrane of the glomerulus and form the nal barrier to serum protein loss. Podocin, a member of the stomatin family of membrane proteins, is exclusively expressed in the slit diaphragm that connects neighbouring podocytes in the glomerulus (Huber et al., 2001). The slit diaphragm is a key structure involved in maintaining

podocyte integrity. Measurement of urinary podocin along with other podocyte proteins has been proposed as a useful tool for detecting glomerular damage and renal disease development in other settings (Sakairi et al., 2010), but this study is the rst report associating podocin loss with CsA nephrotoxicity. Uromodulin (TammHorsfall protein) is the most abundant protein in normal urine (Hession et al., 1987). In the present study decreases in urinary TammHorsfall protein levels were detected following 1 week of CsA treatment. The biological role of Tamm Horsfall protein is unclear, however decreased urinary TammHorsfall protein is considered an indicator of tubular damage and has been proposed as a biomarker in other forms of renal disease (Kistler et al., 2009; Prajczer et al., 2010). Urinary TammHorsfall protein levels have previously been proposed as an indicator of renal function in transplant recipients (Kaden et al., 1994) and the results of this study suggest that urinary uromodulin levels may also be an indicator of CsA nephrotoxicity. 2-DE combined with mass spectrometry is a reliable high-throughput method which can be used to identify potential markers of toxicity. Using this technique, 15 urinary proteins that were signicantly altered by 1 week of CsA treatment were identied (Table 2). These included Heat shock protein 60 (Hsp60), superoxide dismutase (SOD), apolipoprotein A1, mouse transthyretin and serum albumin precursor. Several

208

S. O'Connell et al. / Toxicology and Applied Pharmacology 252 (2011) 201210

Table 2 Proteins identied by mass spectrometry and database analysis. The table contains the protein spots identied by mass spectrometry and UniProt database analysis. CD-1 mice were treated with either cyclosporine A or vehicle control for one week. Mouse urine was collected and subjected to 2D PAGE and subsequent gels visualised and analysed using Progenesis software. Protein spots were then identied following mass spectrometry on a Finnigan LTQ mass spectrometer connected to a Surveyor chromatography system. Raw mass spectrometry data was analysed using Proline, a proteomics analysis platform and the International Protein Index (IPI) for mouse version 3.16 used as the FASTA search database. A second platform was also used in which MS/MS spectra were analysed using TurboSEAQUEST. A full description of this process is described by Feighery et al. (2008). Protein identication Up-regulated proteins Cadherin 1 precursor Serum albumin precursor Actin 11 Kallikrein K1 precursor Vinculin Apolipoprotein A1 precursor Voltage dependent anion selective channel protein 1 Major urinary protein 1 Actin 1 Actin2 Mouse transthyretin Mouse superoxide dismutase Actin 4 Major urinary protein 11 Down-regulated proteins Heat shock protein 60 Accession no. P09803 P07724 P53496 P15947 Q64727 Q00623 Q60932 P11588 P60710 Q96292 P07309 P08228 P53494 P09438 M/W (kDa) 98.2 68.6 41.6 28.7 116.6 30.5 32.3 20.6 41.7 41.9 15.7 15.9 41.8 17.5 % Sequence coverage 12.3 24.5 14.0 31.6 1.8 4.4 14.7 69.7 6.0 6.22 5.9 14.6 139. 82.0 ID scores 1.6e-004 3.6e-005 1.4e-009 7.7e-008 2.4e-004 3.4e-007 3.7e-009 2.5e-010 3.8e-009 4.0e-007 1.7e-004 2.4e-006 1.8e-005 1.5e-009 Spot intensity 46.1 2.4 71.5 3.3 61.6 3.9 63.9 5.9 13.1 1.8 9.9 0.4 9.4 1.3 9.8 2.5 3.1 0.5 1.5 0.05 0.7 0.1 0.4 0.07 1.7 0.2 48.5 3.1 Fold change 2.00 2.47 2.75 2.11 2.86 3.69 3.74 4.75 8.49 16.81 65.49 70.88 43.72 0.99

P63038

60.9

1.6

9.4e-004

44.7 3.0

35.35

isoforms of actin were also detected along with vinculin, another cytoskeletal protein. Further analysis is required to determine the specic roles of these proteins and their usefulness as indicators and/or therapeutic targets in CsA-nephropathy. One protein, CDH-1, was selected for further analysis. CDH-1 was increased in the urine of CsA treated mice with a corresponding decrease in whole kidney protein expression by 1 week of CsA treatment. CDH-1 is a critical epithelial adhesion molecule that is highly expressed along the tubular epithelium. Urinary loss of CDH-1 suggests breakdown of the tubular junctions likely leading to loss of tubular integrity and function (Slattery et al., 2005). Increased urinary CDH-1 has been detected in patients with diabetic nephropathy and has been suggested to have clinical diagnostic value in that setting (Jiang et al., 2009). We propose that urinary CDH-1 may also be an early indicator of CsA nephropathy.

The results from this current study identify a number of novel aspects of CsA nephrotoxicity, highlighting roles for both the tubular epithelium and glomerular cells in establishing the pro-brotic environment. This research has identied a number of novel potential markers of CsA nephropathy. Further investigation will determine if these indicators translate to a clinical setting, and to what extent they may be useful as therapeutic targets.

Funding information This work was supported by the Health Research Board, Science Foundation Ireland, Enterprise Ireland and the Conway Institute funded by the programme for research in third level institutions administered

Fig. 5. CsA-induced alterations in CDH-1 expression in CD-1 mice. CD-1 mice were treated with CsA or vehicle alone for one week or four weeks. Kidneys from each group were harvested and total kidney protein extracted. CDH-1 protein expression was examined by Western blot analysis (A) and in kidney sections by immunouorescent microscopy (B) (magnication 400). Results are expressed as mean S.E.M of a sample size of six mice per treatment group. *Indicates statistically signicant difference to time matched control (*p b 0.05, **p b 0.01).

S. O'Connell et al. / Toxicology and Applied Pharmacology 252 (2011) 201210

209

by the Higher Education Authority and by the EU 7th Framework grant SysKid, HEALTHF22009241544. Conict of interest The authors have no conict of interest. The results presented in this manuscript have not been published previously in whole or part, except in abstract form. Acknowledgments We acknowledge the expert technical assistance of Catherine Moss, George Keating and Caitriona Scaife of the UCD Conway Institute Core Technology service. We acknowledge the UCD Biomedical Facility for their expert technical assistance and the expert assistance of Dr. Tom Crotty Department of Pathology, St Vincent's University Hospital. We acknowledge the donation of antibodies by Dr. John Crean, UCD Conway Institute and by Prof. Frank Strutz, University Hospital Goettingen. References
Bestard, O., Cruzado, J.M., Grinyo, J.M., 2005. Calcineurin-inhibitor-sparing immunosuppressive protocols. Transplant. Proc. 37, 37293732. Bobadilla, N.A., Gamba, G., 2007. New insights into the pathophysiology of cyclosporine nephrotoxicity: a role of aldosterone. Am. J. Physiol. Renal Physiol. 293, F2F9. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proteindye binding. Anal. Biochem. 72, 248254. Chaaya, R., Alfarano, C., Guilbeau-Frugier, C., Coatrieux, C., Kesteman, A.S., Parini, A., Fares, N., Gue, M., Schanstra, J.P., Bascands, J.L., 2011. Pargyline reduces renal damage associated with ischaemiareperfusion and cyclosporin. Nephrol. Dial. Transplant. 26, 489498. Clarke, H., Ryan, M.P., 1999. Cyclosporine A-induced alterations in magnesium homeostasis in the rat. Life Sci. 64, 12951306. Cockcroft, D.W., Gault, M.H., 1976. Prediction of creatinine clearance from serum creatinine. Nephron 16, 3141. Dieterle, F., Perentes, E., Cordier, A., Roth, D.R., Verdes, P., Grenet, O., Pantano, S., Moulin, P., Wahl, D., Mahl, A., End, P., Staedtler, F., Legay, F., Carl, K., Laurie, D., Chibout, S.D., Vonderscher, J., Maurer, G., 2010. Urinary clusterin, cystatin C, beta2-microglobulin and total protein as markers to detect drug-induced kidney injury. Nat. Biotechnol. 28, 463469. Dudas, P.L., Argentieri, R.L., Farrell, F.X., 2009. BMP-7 fails to attenuate TGF-beta1induced epithelial-to-mesenchymal transition in human proximal tubule epithelial cells. Nephrol. Dial. Transplant. 24, 14061416. El Chaar, M., Chen, J., Seshan, S.V., Jha, S., Richardson, I., Ledbetter, S.R., Vaughan Jr., E.D., Poppas, D.P., Felsen, D., 2007. Effect of combination therapy with enalapril and the TGF-beta antagonist 1D11 in unilateral ureteral obstruction. Am. J. Physiol. Renal Physiol. 292, F1291F1301. Feighery, R., Maguire, T., Ryan, M.P., McMorrow, T., 2008. A proteomic approach to immune-mediated epithelialmesenchymal transition. Proteomics Clin. Appl. 2, 11101117. Gaston, R.S., 2009. Chronic calcineurin inhibitor nephrotoxicity: reections on an evolving paradigm. Clin. J. Am. Soc. Nephrol. 4, 20292034. Grotendorst, G.R., 1997. Connective tissue growth factor: a mediator of TGF-beta action on broblasts. Cytokine Growth Factor Rev. 8, 171179. Gupta, S., Clarkson, M.R., Duggan, J., Brady, H.R., 2000. Connective tissue growth factor: potential role in glomerulosclerosis and tubulointerstitial brosis. Kidney Int. 58, 13891399. Hara, S., Umino, D., Someya, T., Fujinaga, S., Ohtomo, Y., Murakami, H., Shimizu, T., 2009. Protective effects of Mizoribine on Cyclosporine A nephropathy in rats. Pediatr. Res. 66, 524527. Hession, C., Decker, J.M., Sherblom, A.P., Kumar, S., Yue, C.C., Mattaliano, R.J., Tizard, R., Kawashima, E., Schmeissner, U., Heletky, S., et al., 1987. Uromodulin (TammHorsfall glycoprotein): a renal ligand for lymphokines. Science 237, 14791484. Huber, T.B., Kottgen, M., Schilling, B., Walz, G., Benzing, T., 2001. Interaction with podocin facilitates nephrin signaling. J. Biol. Chem. 276, 4154341546. Ito, Y., Aten, J., Bende, R.J., Oemar, B.S., Rabelink, T.J., Weening, J.J., Goldschmeding, R., 1998. Expression of connective tissue growth factor in human renal brosis. Kidney Int. 53, 853861. Jia, J., Wang, J., Teh, M., Sun, W., Zhang, J., Kee, I., Chow, P.K., Liang, R.C., Chung, M.C., Ge, R., 2010. Identication of proteins differentially expressed between capillary endothelial cells of hepatocellular carcinoma and normal liver in an orthotopic rat tumor model using 2-D DIGE. Proteomics 10, 224234. Jiang, H., Guan, G., Zhang, R., Liu, G., Cheng, J., Hou, X., Cui, Y., 2009. Identication of urinary soluble E-cadherin as a novel biomarker for diabetic nephropathy. Diabetes Metab. Res. Rev. 25, 232241.

Kaden, J., Oesterwitz, H., May, G., Strobelt, V., Schroder, K., Bohnke, C., Gellert, S., Groth, J., Schabel, J., Eismann, R., Templin, R., Sehland, J., 1994. Effects of PUVA therapy on kidney allografts: results of a randomized prospective double-blind study. Transpl. Int. 7 (Suppl 1), S275S280. Kattla, J.J., Carew, R.M., Heljic, M., Godson, C., Brazil, D.P., 2008. Protein kinase B/Akt activity is involved in renal TGF-beta1-driven epithelialmesenchymal transition in vitro and in vivo. Am. J. Physiol. Renal Physiol. 295, F215F225. Kiely, B., Feldman, G., Ryan, M.P., 2003. Modulation of renal epithelial barrier function by mitogen-activated protein kinases (MAPKs): mechanism of cyclosporine A-induced increase in transepithelial resistance. Kidney Int. 63, 908916. Kistler, A.D., Mischak, H., Poster, D., Dakna, M., Wuthrich, R.P., Serra, A.L., 2009. Identication of a unique urinary biomarker prole in patients with autosomal dominant polycystic kidney disease. Kidney Int. 76, 8996. Kwong, Y.T., Stevens, L.A., Selvin, E., Zhang, Y.L., Greene, T., Van Lente, F., Levey, A.S., Coresh, J., 2010. Imprecision of urinary iothalamate clearance as a gold-standard measure of GFR decreases the diagnostic accuracy of kidney function estimating equations. Am. J. Kidney Dis. 56, 3949. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680685. Lan, H.Y., 2003. Tubular epithelial-myobroblast transdifferentiation mechanisms in proximal tubule cells. Curr. Opin. Nephrol. Hypertens. 12, 2529. Li, C., Yang, C.W., 2009. The pathogenesis and treatment of chronic allograft nephropathy. Nat. Rev. Nephrol. 5, 513519. Ling, H., Li, X., Jha, S., Wang, W., Karetskaya, L., Pratt, B., Ledbetter, S., 2003. Therapeutic role of TGF-beta-neutralizing antibody in mouse cyclosporin A nephropathy: morphologic improvement associated with functional preservation. J. Am. Soc. Nephrol. 14, 377388. Lloberas, N., Torras, J., Alperovich, G., Cruzado, J.M., Gimenez-Bonafe, P., HerreroFresneda, I., Franquesa, M., Rama, I., Grinyo, J.M., 2008. Different renal toxicity proles in the association of cyclosporine and tacrolimus with sirolimus in rats. Nephrol. Dial. Transplant. 23, 31113119. Macconi, D., Sangalli, F., Bonomelli, M., Conti, S., Condorelli, L., Gagliardini, E., Remuzzi, G., Remuzzi, A., 2009. Podocyte repopulation contributes to regression of glomerular injury induced by ACE inhibition. Am. J. Pathol. 174, 797807. Martin-Martin, N., Ryan, G., McMorrow, T., Ryan, M.P., 2010. Sirolimus and cyclosporine A alter barrier function in renal proximal tubular cells through stimulation of ERK1/2 signaling and claudin-1 expression. Am. J. Physiol. Renal Physiol. 298, F672F682. McMorrow, T., Gaffney, M.M., Slattery, C., Campbell, E., Ryan, M.P., 2005. Cyclosporine A induced epithelialmesenchymal transition in human renal proximal tubular epithelial cells. Nephrol. Dial. Transplant. 20, 22152225. Murphy, M., Godson, C., Cannon, S., Kato, S., Mackenzie, H.S., Martin, F., Brady, H.R., 1999. Suppression subtractive hybridization identies high glucose levels as a stimulus for expression of connective tissue growth factor and other genes in human mesangial cells. J. Biol. Chem. 274, 58305834. Phanish, M.K., Winn, S.K., Dockrell, M.E., 2010. Connective tissue growth factor(CTGF, CCN2)a marker, mediator and therapeutic target for renal brosis. Nephron Exp. Nephrol. 114, e83e92. Prajczer, S., Heidenreich, U., Pfaller, W., Kotanko, P., Lhotta, K., Jennings, P., 2010. Evidence for a role of uromodulin in chronic kidney disease progression. Nephrol. Dial. Transplant. 25, 18961903. Qi, W., Chen, X., Poronnik, P., Pollock, C.A., 2006. The renal cortical broblast in renal tubulointerstitial brosis. Int. J. Biochem. Cell Biol. 38, 15. Rastaldi, M.P., Ferrario, F., Giardino, L., Dell'Antonio, G., Grillo, C., Grillo, P., Strutz, F., Muller, G.A., Colasanti, G., D'Amico, G., 2002. Epithelialmesenchymal transition of tubular epithelial cells in human renal biopsies. Kidney Int. 62, 137146. Reis, F.N., 2010. The unsolved cyclosporine-induced kidney injury: is paricalcitol a feasible new renoprotective option? Kidney Int. 77, 10551057. Rosenkranz, M.E., Wilson, D.C., Marinov, A.D., Decewicz, A., Grof-Tisza, P., Kirchner, D., Giles, B., Reynolds, P.R., Kolli, K., Thompson, S.D., Hirsch, R., 2010. Synovial uid proteins differentiate between the subtypes of juvenile idiopathic arthritis. Arthrit. Rheum. 62 (6), 18131823. Sakairi, T., Abe, Y., Kajiyama, H., Bartlett, L.D., Howard, L.V., Jat, P.S., Kopp, J.B., 2010. Conditionally immortalized human podocyte cell lines established from urine. Am. J. Physiol. Renal Physiol. 298, F557F567. Sato, Y., Wharram, B.L., Lee, S.K., Wickman, L., Goyal, M., Venkatareddy, M., Chang, J.W., Wiggins, J.E., Lienczewski, C., Kretzler, M., Wiggins, R.C., 2009. Urine podocyte mRNAs mark progression of renal disease. J. Am. Soc. Nephrol. 20, 10411052. Shihab, F.S., Bennett, W.M., Yi, H., Choi, S.O., Andoh, T.F., 2003. Mycophenolate mofetil ameliorates arteriolopathy and decreases transforming growth factor-beta1 in chronic cyclosporine nephrotoxicity. Am. J. Transplant. 3, 15501559. Shihab, F.S., Bennett, W.M., Yi, H., Andoh, T.F., 2006. Effect of cyclosporine and sirolimus on the expression of connective tissue growth factor in rat experimental chronic nephrotoxicity. Am. J. Nephrol. 26, 400407. Slattery, C., Campbell, E., McMorrow, T., Ryan, M.P., 2005. Cyclosporine A-induced renal brosis: a role for epithelialmesenchymal transition. Am. J. Pathol. 167, 395407. Slattery, C., McMorrow, T., Ryan, M.P., 2006. Overexpression of E2A proteins induces epithelialmesenchymal transition in human renal proximal tubular epithelial cells suggesting a potential role in renal brosis. FEBS Lett. 580, 40214030. Thomas, S.E., Andoh, T.F., Pichler, R.H., Shankland, S.J., Couser, W.G., Bennett, W.M., Johnson, R.J., 1998. Accelerated apoptosis characterizes cyclosporine-associated interstitial brosis. Kidney Int. 53, 897908. Tuglular, S., Gogas Yavuz, D., Cakalagaoglu, F., Citak, L., Arikan, H., Kocak, H., Ozener, C., Akoglu, E., 2004. Cyclosporine-A induced nephrotoxicity is associated with

210

S. O'Connell et al. / Toxicology and Applied Pharmacology 252 (2011) 201210 Xu, Y., Wan, J., Jiang, D., Wu, X., 2009. BMP-7 blocks the cyclosporine-A-induced epithelial-to-mesenchymal transition in renal tubular epithelial cells. Nephron Exp. Nephrol. 114, e23e31. Yang, C.W., Faulkner, G.R., Wahba, I.M., Christianson, T.A., Bagby, G.C., Jin, D.C., Abboud, H.E., Andoh, T.F., Bennett, W.M., 2002. Expression of apoptosis-related genes in chronic cyclosporine nephrotoxicity in mice. Am. J. Transplant. 2, 391399. Yokoi, H., Mukoyama, M., Sugawara, A., Mori, K., Nagae, T., Makino, H., Suganami, T., Yahata, K., Fujinaga, Y., Tanaka, I., Nakao, K., 2002. Role of connective tissue growth factor in bronectin expression and tubulointerstitial brosis. Am. J. Physiol. Renal Physiol. 282, F933F942. Zeisberg, M., Hanai, J., Sugimoto, H., Mammoto, T., Charytan, D., Strutz, F., Kalluri, R., 2003. BMP-7 counteracts TGF-beta1-induced epithelial-to-mesenchymal transition and reverses chronic renal injury. Nat. Med. 9, 964968.

decreased renal bone morphogenetic protein-7 expression in rats. Transplant. Proc. 36, 131133. Turk, T., Leeuwis, J.W., Gray, J., Torti, S.V., Lyons, K.M., Nguyen, T.Q., Goldschmeding, R., 2009. BMP signaling and podocyte markers are decreased in human diabetic nephropathy in association with CTGF overexpression. J. Histochem. Cytochem. 57, 623631. Veerasamy, M., Nguyen, T.Q., Motazed, R., Pearson, A.L., Goldschmeding, R., Dockrell, M.E., 2009. Differential regulation of E-cadherin and alpha-smooth muscle actin by BMP 7 in human renal proximal tubule epithelial cells and its implication in renal brosis. Am. J. Physiol. Renal Physiol. 297, F1238F1248. Wang, S., Hirschberg, R., 2003. BMP7 antagonizes TGF-beta-dependent brogenesis in mesangial cells. Am. J. Physiol. Renal Physiol. 284, F1006F1013. Wang, G., Lai, F.M., Kwan, B.C., Lai, K.B., Chow, K.M., Li, P.K., Szeto, C.C., 2009. Podocyte loss in human hypertensive nephrosclerosis. Am. J. Hypertens. 22, 300306.