Sub-nephrotoxic cisplatin sensitizes rats to acute renal failure and increases urinary excretion of fumarylacetoacetase

Toxicology Letters 234 (2015) 99–109

Contents lists available at ScienceDirect

Toxicology Letters journal homepage: www.elsevier.com/locate/toxlet

Sub-nephrotoxic cisplatin sensitizes rats to acute renal failure and increases urinary excretion of fumarylacetoacetase Laura Vicente-Vicente a,b,c , Fernando Sánchez-Juanes b,d, Omar García-Sánchez a , Víctor Blanco-Gozalo e, Moisés Pescador a , María A. Sevilla a , José Manuel González-Buitrago b,d,f , Francisco J. López-Hernández a,b,c,d,g , José Miguel López-Novoa a,b,c , Ana Isabel Morales a,b,c, * a

Unidad de Toxicología and Unidad de Fisiopatología Renal y Cardiovascular, Departamento de Fisiología y Farmacología, Universidad de Salamanca, Spain Instituto de Investigación Biomédica de Salamanca (IBSAL), Salamanca, Spain c Fundación Renal Íñigo Álvarez de Toledo, Instituto Reina Sofía de Investigación Nefrológica, Madrid, Spain d Unidad de Investigación, Hospital Universitario de Salamanca, Salamanca, Spain e Bio-inRen SL, Salamanca, Spain f Departamento de Bioquímica y Biología Molecular, Universidad de Salamanca, Salamanca, Spain g Instituto de Estudios de Ciencias de la Salud de Castilla y León (IECSCYL), Soria, Spain b

H I G H L I G H T S








Sub-nephrotoxic administration of cisplatin did not alter renal function or renal tissue integrity. Sub-nephrotoxic cisplatin administration predisposed rats to AKI. The urinary level of fumarylacetoacetase is increased in rats predisposed to AKI by cisplatin. Increased urinary fumarylacetoacetase results from an altered renal handling of the filtered blood-borne protein. Fumarylacetoacetase appears in the urine of patients treated with cisplatin.

A R T I C L E I N F O

A B S T R A C T

Article history: Received 1 December 2014 Received in revised form 19 January 2015 Accepted 6 February 2015 Available online 9 February 2015

Nephrotoxicity limits the therapeutic efficacy of the antineoplastic drug cisplatin. Due to dosage adjustment and appropriate monitoring, most therapeutic courses with cisplatin produce no or minimal kidney damage. However, we studied whether even sub-nephrotoxic dosage of cisplatin poses a potential risk for the kidneys by predisposing to acute kidney injury (AKI), specifically by lowering the toxicity threshold for a second nephrotoxin. With this purpose rats were treated with a single sub-nephrotoxic dosage of cisplatin (3 mg/kg, i.p.) and after two days, with a sub-nephrotoxic regime of gentamicin (50 mg/kg/day, during 6 days, i.p.). Control groups received only one of the drugs or the vehicle. Renal function and renal histology were monitored throughout the experiment. Cisplatin treatment did not cause any relevant functional or histological alterations in the kidneys. Rats treated with cisplatin and gentamicin, but not those under single treatments, developed an overt renal failure characterized by both renal dysfunction and massive tubular necrosis. In addition, the urinary excretion of fumarylacetoacetase was increased in cisplatin-treated animals at subtoxic doses, which might be exploited as a cisplatininduced predisposition marker. In fact, the urinary level of fumarylacetoacetase prior to the second nephrotoxin correlated with the level of AKI triggered by gentamicin in predisposed animals. ã 2015 Elsevier Ireland Ltd. All rights reserved.

Keywords: Acute kidney injury Acquired predisposition Cisplatin Urinary biomarkers Fumarylacetoacetase

1. Introduction * Corresponding author at: Universidad de Salamanca, Edificio Departamental, Lab. 226, Campus Miguel de Unamuno, 37007 Salamanca, Spain.

http://dx.doi.org/10.1016/j.toxlet.2014.11.033 0378-4274/ ã 2015 Elsevier Ireland Ltd. All rights reserved.

Cisplatin is an antineoplastic drug widely used in the treatment of a variety of solid tumours. However, nephrotoxicity represents a common side effect that limits its therapeutic efficacy. Cisplatin

100

L. Vicente-Vicente et al. / Toxicology Letters 234 (2015) 99–109

causes acute tubular necrosis leading to a reduction in glomerular filtration rate (GFR; Sanchez-Gonzalez et al., 2011; Winston and Safirstein, 1985). The tubular damage inflicted by cisplatin occurs mainly in the S3 segment of the proximal tubule, but also in the thick ascending limb of the loop of Henle and the distal tubule (Dobyan et al., 1980; Liu and Baliga, 2003; Megyesi et al., 1998; Ramesh and Reeves, 2003, 2005). Clinically, the manifestations of kidney damage produced by high amounts of cisplatin administration consist in increases in serum concentrations of creatinine and urea as well as decrease in creatinine clearance (Crcl). Also an increased excretion of enzymes as N-acetyl-b-glucosaminidase (NAG) and electrolyte disturbances have been shown in cisplatininduced acute kidney injury (AKI; Goldstein and Mayor, 1983). The initial experience with this drug showed that approximately 25–30% of treated patients underwent nephrotoxicity even after a single dose of cisplatin (Kovach et al., 1973; Ries and Klastersky, 1986). Several strategies were used to avoid cisplatin nephrotoxicity, with limited and heterogeneous results, including (i) patient hydration before and during treatment (Arany and Safirstein, 2003), (ii) co-administration of platinium-binding molecules (Bodenner et al., 1986), and (iii) co-treatment with inhibitors of its cytotoxicity, mainly antioxidants such as amifostine and N-acetylcisteine (Hartmann et al., 2000; SanchezGonzalez et al., 2011). Also, the incidence of AKI after cisplatin treatment might be even higher because it might not be correctly estimated by plasma creatinine concentration, hitherto the gold standard indicator (Skinner et al., 1991; Vaidya et al., 2010; Womer et al., 1985). Mild forms of AKI may not result in sufficient renal dysfunction to cause an increment in plasma creatinine. Yet even mild, transitory, self-repairing episodes of AKI may have important health consequences. Accumulated AKI episodes may lead to progression to chronic kidney disease (Singbartl and Kellum, 2012). In fact, repeated cycles of chemotherapy with cisplatin may lead to chronic tubulointerstitial fibrosis (Guinee et al., 1993). A small fraction of AKI patients need dialysis for life (Van Berendoncks et al., 2010); and finally, AKI episodes are related to an increase in medium and long term mortality (Van Berendoncks et al., 2010). Therefore, detection of traditionally undetected mild forms of AKI, as well as earlier detection of AKI episodes are unmet goals for a better handling of this disease. A new generation of biomarkers, including neutrophil gelatinaseassociated lipocalin (NGAL), kidney injury molecule (KIM-1) and others, are under development, which show earlier and higher sensitivity than plasma creatinine (Gautier et al., 2010; McDuffie et al., 2010). Complementary to early detection to reduce AKI severity, diagnostic prevention poses a new strategy to reduce AKI incidence. We have recently demonstrated that gentamicin, in a sub-nephrotoxic regime, sensitizes rats to AKI by lowering the threshold of nephrotoxicity for other nephrotoxins (Quiros et al., 2010). Gentamicin-induced sensitization correlates with the increased excretion of ganglioside M2 activator protein, which might serve to detect this condition and, prospectively, to stratify patients in a pre-emptive manner according to their individual risk, for their most appropriate clinical handling. Thereafter we hypothesized that other drugs such as cisplatin might also induce sensitization to AKI, even at sub-nephrotoxic doses. Indeed, several studies have revealed that the incidence of AKI due to aminoglycosides is increased in patients treated with cisplatin (Milovic et al., 2010; Salem et al., 1982). These data suggest that cisplatin might have predisposed patients to aminoglycoside-triggered AKI. With this background, we studied if sub-nephrotoxic cisplatin sensitized rats to the nephrotoxic effect of gentamicin. Our results indicate that, administration of a subnephrotoxic dose of cisplatin does not produce renal damage or renal dysfunction, but sensitizes rats to developing AKI. This sensitization correlates with a subtle

alteration of the urinary proteome resulting in the increased excretion of fumarylacetoacetase (FAA) which is presented as a potential marker of this condition. 2. Materials and methods Except where otherwise indicated, all reagents were purchased from Sigma–Aldrich (Madrid, Spain). 2.1. Animals and drugs Male Wistar rats (weighing 200–250 g initially) were housed under controlled environmental conditions in metabolic cages and had free access to standard rat chow and drinking water. The studies have been approved by the Institutional Animal Experimentation Ethical committee. Animals were handled according to the guidelines of the European Communities Council Directive 2010/63/UE and to the current Spanish legislation for the use and care of animals, RD 53/2013. Animals were divided into four groups (n = 12 rats per group) (i) control group (C) which received vehicle (saline) via i.p. daily for 8 days; (ii) Cisplatin group (CDDP): rats were given a single i.p. injection of low dose (3 mg/kg) cisplatin, and then i.p vehicle (saline) daily for 6 days; iii) Gentamicin group (G): rats were treated with i.p. vehicle (saline) daily for 2 days, and then with a low dosage regime of gentamicin (50 mg/kg/day) i.p. for 6 days; and (iv) Cisplatin + Gentamicin (CDDP-G) group: rats were treated first with cisplatin (3 mg/kg), and 2 days later with gentamicin (50 mg/kg/day) for another 6 days. The doses of the drugs used in this article were obtained in pilot studies. They correspond to the maximum dose that did not produce nephrotoxicity (i.e. the maximal sub-nephrotoxic dose). At day 2, four animals of each group were euthanized and the remaining animals at day 8. Twenty four hours urine output was collected in metabolic cages at days 0, 2, 4, 6, and 8, cleared by centrifugation and stored at 80  C. At these times, blood samples (200 mL) were also obtained in heparinized capillaries from a small incision in the tail tip. Plasma was separated by centrifugation and kept at 80  C. At the time of sacrifice, rats were anesthetized, and the kidneys dissected. One half of each kidney was fixed in 3.7% para-formaldehyde and further used for histological studies. 2.2. Human samples The urine and blood from six unselected patients treated with cisplatin, was obtained from volunteers from the Oncology Service of the Hospital Universitario de Salamanca (Spain). The urine was collected at 0 h (before cisplatin administration), 24 h and 48 h after cisplatin treatment. Glomerular filtration rate was estimated by the Cockroft–Gault equation (Herget-Rosenthal et al., 2007). Patients’ data were encrypted and treated as confidential information according to the protocols of the Oncology Service of the Hospital Universitario de Salamanca. 2.3. Biochemical analysis Rat plasma samples were analyzed for urea and creatinine (Crp) concentration, and urine samples for creatinine (Cru) concentration with an analyzer (Reflotron plus1; Roche Diagnostics, Barcelona, Spain; lower detection limit 0.5 mg/dL). Creatinine clearance (Clcr) was calculated using the equation Clcr = (Cru X UF)/Crp (were UF stands for urinary flow). Urine was also assayed for protein concentration using the Bradford technique (Bradford, 1976), for glucose concentration using the o-toluidine method (Passey, 1983) and for NAG activity using a commercial, colorimetric kit based on the conversion of

L. Vicente-Vicente et al. / Toxicology Letters 234 (2015) 99–109

101

Table 1 Sub-nephrotoxic cisplatin did not cause kidney damage. Plasma urea and creatinine concentration, creatinine clearance (Clcr), glucosuria, proteinuria, and urinary N-acetyl-b-D-glucosaminidase (NAG) activity at basal time (immediately before cisplatin administration; d0) and two days after (immediately before gentamicin administration; d2) in control, and cisplatin treated animals. AU, arbitrary units. Data show the mean  standard error of the mean of n = 12 animals per group. Urea mg/dL

Creatinine mg/dL

Clcr mL/min

Glucosuria mg/day

Proteinuria mg/day

NAG AU/day

Control

d0 d2

48.6  5.2 41.2  1.2

<0.5 <0.5

>0.88  0.04 >0.95  0.05

9.46  1.19 9.02  1.48

4.28  0.42 3.56  0.5

0.62  0.10 0.73  0.13

Cisplatin

d0 d2

38.1  2.0 54.0  11.0

<0.5 <0.5

>0.90  0.04 >0.90  0.07

8.55  0.79 14.04  2.7

4.85  0.38 3.47  0.35

0.70  0.15 0.76  0.13

3-cresolsufonphtaleinyl-N-acetyl-b-D-glucosaminide into the purple 3-cresolsufonphthaleinyl and N-acetyl-b-D-glucosamine (Roche Diagnostics, Barcelona, Spain) following the manufacturers’ instructions. 2.4. Western blot Tissue protein extracts were obtained after kidney homogenization as previously described (Morales et al., 2006). Renal kidney homogenate (100 mg protein), a volume of urine proportional to the 24 h urine output of each rat or 21 mL or human urine were separated by polyacrylamide gel electrophoresis (PAGE; Mini Protean II system, BioRad Madrid, Spain). Proteins were electro-

transferred to Immobilon P membranes (Millipore, Madrid, Spain), which were then incubated with goat polyclonal antibodies against kidney injury molecule-1 (KIM-1; 1:500 dilution; RD Systems, Minneapolis, MN, USA) FAA (1:200 dilution; Santa Cruz Biotechnology, Santa Cruz, CA, USA), with mouse polyclonal antibody against plasminogen activator inhibitor-1 (PAI-1; 1:1000 dilution; Santa Cruz Biotechnology, Santa Cruz, CA, USA), and with rabbit polyclonal antibodies against Lipocalin-2 (NGAL; 1:500 dilution; MBL international, Wobrun, MA, USA) and clusterin (1:500 dilution, Santa Cruz Biotechnology, Santa Cruz, CA, USA). Membranes were then incubated with horseradish peroxidase (HRP)-coupled secondary antibodies, subsequently incubated with the quimioluminiscent HRP-substrate (ECL; Millipore, Madrid,

[(Fig._1)TD$IG]

Fig. 1. Sub-nephrotoxic cisplatin does not produce kidney damage. a) Representative photographs of the cortical and medullary areas of renal slices stained with hematoxilyn and eosin from control and cisplatin treated rats, 2 days after cisplatin administration. Scale bar = 50 mm. b) Western blot images of the urinary excretion of the proteins lipocalin 2 (NGAL), kidney injury molecule 1 (KIM 1), Clusterin and plasminogen activator inhibitor-1 (PAI-1) in control and cisplatin treated animals at day 2 (n = 8 animals per group).

102

L. Vicente-Vicente et al. / Toxicology Letters 234 (2015) 99–109

Table 2 Subnephrotoxic cisplatin did not alter renal hemodynamics. Glomerular filtration rate (measured as inulin clearance), renal vascular resistance (RVR), renal plasma flow (RPF; measured as p-aminohippuric acid clearance), and systolic blood pressure (SBP), and renal blood flow (RF) both measured by laser doppler flowmetry, in control and cisplatin treated rats at day 2. AU, arbitrary units. Data show the mean  standard error of the mean of n = 5 animals per group. Inulin clearance (mL/min)

RVR (mmHg  mL/min)

RPF (mL/min)

SBP (mmHg)

RF (AU)

Control

1.53  0.3

10.51  1.61

8.52  2.77

102.1  6.99

297.41  112.41

Cisplatin

1.15  0.48

13.6  0.35

5.08  3.05

102.1  2.63

301.40  113.78

Spain), and exposed to photographic films (Kodak, Rochester, NY, USA). Bands were quantified with the Scion Image software (Scion Corporation; Frederick, MD, USA).

2.5. Histological studies Paraffin blocks were made with para-formaldehyde fixed tissue and 5-mm tissue sections were stained with hematoxilin and eosin. Photographs were taken under an Olympus BX51 microscope connected to an Olympus DP70 colour, digital camera (Olympus, Madrid, Spain). 2.6. Laser Doppler flowmetry Right before gentamicin regime onset, kidney cortical perfusion was measured by laser Doppler flowmetry (moor LDSL laser Doppler line scanner, Wilmington, DE, USA), in four animals of each group, by applying the beam of light directly on the kidney surface. Renal perfusion was expressed as arbitrary units (AU; (GonzálezEscalada et al., 1999; Scheeren et al., 2011)).

2.7. Inulin and p-aminohippuric acid (PAH) clearance Four animals of each group were anesthetized with sodium pentobarbital (5 g/kg, i.p.) and their carotid artery, jugular vein and urinary bladder were cannulated in order to perform renal clearance studies as previously described (Morales et al., 2010). Inulin and PAH were measured in plasma and urine using a twochannel Liquid Scintillation Counter (allac 1409 DSA, Turku, Finland). Glomerular filtration rate (GFR) was measured as clearance of 3H-inulin and renal plasma flow (RPF) was measured as clearance of 14C-PAH. The calculation of GFR and RPF was performed by standard formulae. Renal blood flow (RBF) was calculated from RPF and packed cell volume. Filtration fraction (FF) was calculated as GFR/RPF. Renal vascular resistance was calculated as RBF/Mean arterial pressure. 2.8. Renal functional reserve (RFR) Another four animals of each group were used for this study, also at day 2 (before gentamicin administration). The surgical method was the same as described for PAH-inulin clearance experiments, but in this case two period studies were conducted in

[(Fig._2)TD$IG]

Fig. 2. Sub-nephrotoxic cisplatin does not alter renal hemodynamic. a) Glomerular filtration rate (GFR) progression during glycine perfusion and renal functional reserve, calculated as percentage of GFR increment (with respect to the basal value), in control and cisplatin groups. Experiments were performed at day 2. b) Renal vascular reactivity from ex vivo perfused kidneys from control and cisplatin-treated rats. Left panel, basal perfusion pressure; central panel, response to KCl (K) and phenylephrin (PE); and right panel, acetylcholine concentration-relaxation curves. Experiments were performed 2 days after cisplatin injection. Data are expressed as mean  standard error of the mean of n = 4 rats per group. * p < 0.05 vs control at the same time point.

L. Vicente-Vicente et al. / Toxicology Letters 234 (2015) 99–109

[(Fig._3)TD$IG]

103

Fig. 3. Sub-nephrotoxic cisplatin predisposes rats to AKI: biochemical parameters. Plasma creatinine and urea concentration, creatinine clearance, proteinuria, glucosuria, and N-acetyl-beta-D-glucosaminidase (NAG) activity at times basal (d0), days 2, 4, 6 and 8 (d2, d4, d6 and d8). Data are expressed as mean  standard error of the mean of n = 8 animals per group (# p < 0.05 vs basal in the same group; * p < 0.05 vs control at the same time point; & p < 0.05 vs cisplatin at the same time point; g p < 0.05 vs gentamicin at the same time point).

each rat as previously described (Slomowitz et al., 2002), with some modifications. Renal vascular reserve was calculated as the percent increase in the GFR value induced by glycine with respect to the basal GFR value.

2.9. Ex vivo perfused kidney for renal vascular reactivity studies Kidneys from four rats treated with cisplatin (3 mg/kg) or from four rats treated with vehicle (control) were used 2 days after cisplatin dose injection. The left kidney was cannulated through the renal artery as previously described (Monroy-Ruiz et al., 2011). Renal vasoconstriction and vasodilatation were detected through changes in perfusion pressure. KCl, phenylephrine and acetylcholine were incorporated in the Krebs solution perfusing the kidneys. The contracting responses to KCl and phenylephrine were expressed as mm Hg of increment over basal perfusion pressure. The relaxant responses to acetylcholine were expressed as percentages of phenylephrine contraction.

2.10. Proteomics The proteome of day 2 urine samples from animals treated with cisplatin or vehicle (as controls) was analyzed by two-dimensional electrophoresis (2-DE), as previously described (Ferreira et al., 2011; Quiros et al., 2010). The spots of interest were analyzed with the Image Master Platinum software (GE Healthcare, Madrid, Spain) and in-gel digested with porcine trypsin (Promega,

Barcelona, Spain). Tryptic peptides were analyzed by MALDI-TOF on an Autoflex III instrument (Bruker Daltonics, Madrid, Spain). Protein identification was performed with the MASCOT software (www.matrixscience.com). In all protein identifications, the probability scores were greater than the score fixed by MASCOT as significant with a p-value lower than 0.05.

2.11. Extracorporeal circuit for kidney perfusion Two days after subtoxic cisplatin treatment, four rats were anesthetized and an extracorporeal circuit for kidney perfusion was set up, as described elsewhere (Lopez-Novoa et al., 1986), with some modifications. Briefly, the renal artery, vein and ureter of the right kidney were ligated. The renal artery and vein, of the left kidney the carotid and the urinary bladder were canulated. A heparin bolus (1000 UI/kg) was administered via the femoral vein. The carotid and renal arteries were connected through a catheter to allow blood to flow directly from the carotid to renal artery and urine fractions were collected from a catheter placed in the (LopezNovoa et al., 1986) urinary bladder every 10 min during 60 min. After this time, blood flow from the carotid was interrupted, and oxygenated and warm (37  C) Krebs–dextran (40 g/L of dextran (molecular weight 64 K–76 K) in Krebs solution (120 mM NaCl, 4.8 mM KCl, 1.8 mM CaCl2, 1.2 mM MgSO4, 1.2 mM KH2PO4, 25 mM NaHCO3, 0.026 mM EDTA, 11.1 glucose, pH 7.4)] was perfused through the renal artery at 2.4 mL/min. The effluent from the renal vein was discarded and urine fractions were collected from the catheter placed in the urinary bladder, every 10 min during 60 min.

104

L. Vicente-Vicente et al. / Toxicology Letters 234 (2015) 99–109

All urine samples were kept at 80  C until assayed by Western blot for the presence of specific proteins. 2.12. Gene expression analysis Renal tissue RNA was purified with the commercial kit Nucleospin RNAII (Macherey-Nagel, Düren, Germany) according to the manufacturer’s instructions. The sequence of the FAA primers used were: sense 50 -GCACGCTGCCGAAGCTCCTT-30 and antisense 30 -CTCGCGAAGCGGCATGGAA-50 . The reactions were run on an iQ5 Real-Time PCR detection system (Bio-Rad Madrid, Spain).

2.13. Statistical analysis In order to analyze the differences between the parameters measured we performed ANOVA and post hoc analyses with Scheffe’s test for normally distributed data, and Kruskal–Wallis test when data were not normally distributed. Differences were considered statistically significant when p < 0.05. We used exponential regression to test the predictive relationships among the excreted proteins albumin, transferrin and FAA and the level of renal damage taken as plasma urea concentration 2 days after gentamicin regime onset. Data were analyzed using the Number Cruncher Statistical System (NCSS) software, version 6.0.10 for Windows and IBM SPSS Statistics 20.0. Data are shown as mean  standard error of the mean.

3. Results 3.1. Sub-toxic administration of cisplatin did not alter renal function or renal tissue integrity As shown in Table 1, two days after treatment, administration of 3 mg/kg cisplatin did not produce any alterations in plasma urea and creatinine concentration, urinary excretion of glucose and proteins, and NAG activity. Congruently, GFR measured as Clcr, did not show differences between animals receiving cisplatin and animals without the chemotherapeutic agent. Renal histological examination revealed that 3 mg/kg cisplatin did not produce any gross morphologic alterations in the kidneys (Fig. 1a). Similarly, no increased urinary excretion of the novel and sensitive markers NGAL, KIM-1, clusterin and PAI-1 was observed in cisplatin-treated rats (Fig. 1b). These results indicate that the dose of 3 mg/kg cisplatin was apparently bereft of renal toxicity in our model. It has been previously shown that, in pigs, low dose (2.5 mg/kg) of cisplatin causes no alterations in GFR or RBF. However, it induces a reduction in RFR, which is manifested only when renal function is challenged, i.e. by unilateral nephrectomy (Robbins et al., 1990, 1992). In order to study the potential nephrotoxic effect of cisplatin, we evaluated RBF and RFR in our model. Clearance experiments (Table 2) corroborated that low dose cisplatin did not modify GFR, estimated by inulin clearance. Two different techniques, PAH clearance and laser Doppler flowmetry showed that RBF was not significantly altered by cisplatin (Table 2). PAH clearance also indicated that cisplatin administration induced no

[(Fig._4)TD$IG]

Fig. 4. Sub-nephrotoxic cisplatin predisposes rats to AKI: renal histology. Representative photographs of the cortical area of renal slices stained with hematoxilyn and eosin from control, cisplatin, gentamicin and cisplatin + gentamicin treated rats in day 8. Scale bars = 50 mm. Black arrows means tubular dilatation, white arrows means hyaline deposits.

L. Vicente-Vicente et al. / Toxicology Letters 234 (2015) 99–109

[(Fig._5)TD$IG]

105

Fig. 5. Differential proteomic profiling of the urine from control and cisplatin-treated rats. Representative images of 2-DE gels, n = 8 animals per group. The marked spots correspond to proteins significantly increased in the urine of cisplatin treated rats as compared with control rats. The table below reports protein identification parameters obtained by MALDI-TOF analysis and data base blast for the selected spots. Sc, score, MW, molecular weight (in kDa), pI, isoelectric point.

significant differences in renal vascular resistance between control and cisplatin groups. RFR (the ability of the kidneys to induce a compensatory increase in GFR by means of increasing RBF) was not modified by cisplatin (Fig. 2). Finally, renal vascular reactivity was not altered by cisplatin treatment. As shown in Fig. 2, ex vivo perfused kidneys from cisplatin-treated rats had not higher vascular resistance than those from control animals in basal conditions. They even showed lower vascular resistance than controls, and responded similarly to vasoconstrictors (e.g. high KCl and phenylephrine) and the vasodilator acetylcholine. These results reinforce the idea that, in our model, 3 mg/kg cisplatin causes no renal vascular alterations. 3.2. Sub-nephrotoxic cisplatin administration predisposed rats to AKI In animals previously treated with 3 mg/kg cisplatin, gentamicin administration (two days after cisplatin) produced an overt AKI, as shown in Fig. 3. Compared with control group, a significant increase in all biochemical parameters associated to kidney injury was detected only in the CDDP-G group, including plasma urea and creatinine concentration, proteinuria, glucosuria and urinary NAG

activity. GFR, measured as Clcr was lower in CDDP-G than in control group. Renal histology revealed a massive tubular necrosis in rats co-treated with cisplatin and gentamicin, in which extensive tubular atrophy and tubular obstruction were evident (Fig. 4). Kidneys from control animals (C) or treated only with one drug (CDDP and G groups) did not show any gross morphological alterations. 3.3. Identification of urinary biomarkers of acquired susceptibility to AKI caused by sub-nephrotoxic cisplatin. Differential proteomics was used to search for proteins with increased urinary excretion in rats treated with cisplatin, compared to controls. Urine samples from day 2 (right before gentamicin administration) were used. The urinary proteome of both groups seems to be highly similar (Fig. 5), but subtle differences were observed. Of note, these differences did not translate in altered proteinuria, as evidenced in Table 1 and Fig. 3. However, four proteins were found to be consistently and significantly increased in the urine of rats treated with cisplatin, namely albumin, transferrin, transthyretin and FAA. The increased

106

[(Fig._6)TD$IG]

L. Vicente-Vicente et al. / Toxicology Letters 234 (2015) 99–109

3.4. Increased urinary FAA results from an altered renal handling of the filtered blood-borne protein Albumin and transferrin are known serum borne proteins. However, the source of urinary FAA is unknown. We studied whether it goes into the urine from the blood (passing through the glomerular filtration barrier) or from the renal parenchyma. An increased plasmatic level might explain the increased urinary excretion. However, no differences in plasma levels of FAA were detected by western blot between cisplatin-treated and control rats (Fig. 7a). Renal tissue FAA protein level was also identical in both groups (Fig. 7b). Gene expression analysis carried out on renal tissue by qPCR showed that cisplatin treatment did not modify FAA gene expression in the kidneys (Fig. 7c). These results suggested that urinary FAA might come from the blood. Further confirmation was obtained from the experiments using the extracorporeal circuit for kidney perfusion. Urinary excretion of FAA was determined by western blot before and after the beginning of kidney perfusion with Krebs solution. As depicted in Fig. 7d, when the blood irrigating the kidneys was substituted by Krebs solution, FAA disappeared from the urine. 3.5. FAA appears in urine of patients treated with cisplatin Finally, as a proof of concept that FAA can be detected in human urine samples in the context of therapeutic courses with cisplatin, urinary FAA was evaluated in six oncologic patients treated for first time with this drug (Fig. 8) at 0 (before cisplatin administration), 24 and 48 h after cisplatin treatment. FAA can be detected in the urine of these patients. Furthermore, it can be seen that different patterns of urinary FAA evolve after cisplatin administration, potentially corresponding with a predisposition or sensitization to AKI, which needs to be thoroughly investigated in clinical studies. The patients 2, 4, 5 and 6 showed an increase in FAA excretion 24 h after cisplatin administration. Three of them (i.e. 2, 4 and 5) had no alteration in the renal function parameters. Patient 6 already has high basal urinary FAA levels, corresponding to a certain degree of renal injury, as indicated by high NAG excretion and elevated proteinuria. On the other hand, FAA was not increased in the urine of patients 1 and 3, hypothetically indicating that renal susceptibility was not altered by cisplatin treatment. 4. Discussion

Fig. 6. FAA correlation with the level of predisposition to AKI. a) Confirmation by western blot of the increased urinary excretion of FAA identified by proteomics 2 days after cisplatin administration in control and cisplatin-treated animals. Bands were quantified with the Scion Image software. b) Correlation between FAA urinary excretion at day 2 (before gentamicin administration) and plasma urea concentration at day 4 (after two doses of gentamicin) when damage is already evident. Each bar represents the mean  standard error of the mean, n = 8 animals per group (* p < 0.05 vs control). AU, arbitrary units.

urinary excretion of these proteins was confirmed by western blot analysis, which revealed statistical differences between control and cisplatin-treated rats for albumin, transferrin and FAA (data shown only for FAA, Fig. 6a). Interestingly, FAA showed a statistically significant relation between its excretion level at day 2 (immediately prior to gentamicin administration), and the level of renal dysfunction produced by gentamicin (evaluated as plasma urea concentration 2 days after gentamicin regime onset; R = 0.729, p = 0.011; Fig. 6b). Excretion of albumin and transferrin did not correlate with subsequent renal damage. Thus, further studies were only focused on FAA.

Predisposition or sensitization to AKI caused by subnephrotoxic courses of nephrotoxic drugs has been recently demonstrated to occur in experimental animals with low-dose regimes of gentamicin (Quiros et al., 2010). Although related, this concept differs from that of potentiation of nephrotoxicity, known to occur upon the concomitant administration of two potentially nephrotoxic drugs, where a certain level of renal damage inflicted by a drug is amplified by the other. On the contrary, predisposition or sensitization to AKI occurs at sub-nephrotoxic doses of both drugs, i.e. both of the predisposing and of the triggering drug. Accordingly, the sensitization is a silent effect that, notwithstanding increases the risk of undergoing an overt renal failure (Quiros et al., 2010). Herein we show that the predisposing effect is not a particular characteristic of gentamicin. Our present results demonstrate that a sub-nephrotoxic dose of cisplatin also predisposes animals to developing an acute renal failure, when they are re-challenged with a subnephrotoxic course of gentamicin that causes no evidence of renal injury in non-predisposed animals. This is a proof of concept for the capacity of cisplatin to sensitize animals, and potentially patients, to AKI, even under therapeutic courses that apparently have no renal side effects. It is also the base for a potential application in the human being that will call into

L. Vicente-Vicente et al. / Toxicology Letters 234 (2015) 99–109

[(Fig._7)TD$IG]

107

Fig. 7. Increases in the urinary excretion of FAA caused by an altered renal handling. a) Western blot analysis of FAA in plasma (n = 6 per group) and b) renal tissue (n = 4 peer group) in control and cisplatin groups. c). Quantitative reverse transcriptase (qPCR) amplification of the renal mRNA of FAA from control and cisplatin treated rats (n = 6 per group). d) Representative images of western blot analysis of the time-course level of FAA in the urine of rats treated with cisplatin (3 mg/kg) perfused with Krebs solution (n = 4 treated rats). Data are expressed as mean  standard error of the mean. (* p < 0.05 vs basal time). AU, arbitrary units.

considering further and more detailed clinical monitoring and special care and handling of patients treated with cisplatin, even when they do no show signs of renal injury. The fact that two unrelated molecules (i.e. gentamicin and cisplatin) subclinically predispose or sensitize animals to AKI opens up the possibility for a wider generalization of this concept to other drug families, which needs further investigation. Specifically, the drug combination used in our study has, beyond the proof-of-concept utility, a direct correspondence with a real clinical scenario. It is known that chemotherapeutic drugs are myelosuppressive (Engineer et al., 1987). Patients treated with

cisplatin often suffer from infections, neutropenia and febrile episodes. Antibiotics, including aminoglycosides, are necessary in the treatment of urinary and pulmonary infections in patients with metastatic lesions (Engineer et al., 1987). In these circumstances, potentiation of nephrotoxicity has been evidenced in courses in which patients were simultaneously treated with cisplatin and gentamicin therapy (Dentino et al., 1987; Gonzalez-Vitale et al., 1978; Milovic et al., 2010; Salem et al., 1982). This potentiation has also been reproduced in rats treated with cisplatin and different aminoglycoside antibiotics, such as gentamicin, amikacin and tobramicin (Engineer et al., 1987; Jongejan et al., 1989; Kawamura

[(Fig._8)TD$IG]

Fig. 8. Urinary excretion of FAA in patients treated with cisplatin without apparent kidney damage. Western blot analysis of FAA in urine of six patients and their kidney function parameters evaluated: plasma creatinine and urea concentration, glomerular filtration rate (GFR) calculated with the Crockoft–Gault equation, proteinuria and NAG activity in urine. Times evaluated 0, 24 and 48 h after cisplatin administration. AU, arbitrary units.

108

L. Vicente-Vicente et al. / Toxicology Letters 234 (2015) 99–109

et al., 1980). In all these cases, the doses of cisplatin were lightly or overtly nephrotoxic; and, in those circumstances, concurrent aminoglycoside administration increased the severity of the renal damage. In our model, cisplatin dosage was completely sub-nephrotoxic, as revealed not only by parameters used in the clinical practice, but also by newer and more sensitive biomarkers as KIM-1 and NGAL urinary excretion. Studies of renal hemodynamics supported the absence of kidney damage after two days of cisplatin administration. Renal inulin clearance, RBF and RFR were undistinguishable from controls. Also, no differences were observed in renal vascular resistance between groups in vivo neither in vitro. The urinary proteomes of our cisplatin-treated and control rats were highly similar. Yet our urinary differential proteomic study identified 4 proteins with statically significant, increased urinary excretion as a result of the treatment with cisplatin, namely albumin, transferrin, transthyretin and FAA. This might indicate that the sieving properties of the glomerular filtration barrier (GFB) might be affected by cisplatin. As a consequence, some medium molecular weight proteins mostly excluded from filtration in normal circumstances by size or electrical reasons are now capable of spanning the GFB. FAA (also known as fumarylacetate hydrolase) is a cellular enzyme required for tyrosine metabolism. Dysfunctional mutations in the FAH gene have been related to hereditary tyrosinemia type I, a disease coursing with liver and renal injury, characterized by tubular dysfunction (Pierik et al., 2005; Rootwelt et al., 1994). Interestingly, increased plasma level of FAA has also been linked to tubular nephrotoxicity induced by D-Serine (Williams and Lock, 2004). This observation contrasts with our study, in which plasma levels of FAA was not modified by cisplatin treatment. This difference in furmarylacetoacetase plasma levels might be exploited as a marker to distinguish evident tubular necrosis from cisplatin-induced predisposition to AKI. In this latter case, the increased urinary excretion of this protein appears to be the result of an altered renal handling caused by sub-nephrotoxic cisplatin. Altered renal handling might be the consequence of a subtle alteration (i) in the GFB sieving properties leading to increased filtration of certain proteins; or (ii) in specific reabsorption mechanisms. Our results show for first time the detection of FAA in urine, and the possibility to use it as a marker of predisposition to kidney injury by sub-toxic administration of cisplatin. Clearly, the relation of this enzyme and its plasma and urine levels with different aspects of the pathophysiology and diagnosis of nephrotoxicity need to be further explored. Our in situ renal perfusion experiments clearly indicate, however, that the FAA found in the urine does not come from the renal tissue. On the contrary, FAA appears to reach urine as a consequence of a specific glomerular or tubular defect in the renal handling of the bloodborne protein. In fact, when kidneys are perfused with Krebs– dextran solution (bereft of proteins), FAA is no longer excreted with the urine. Most interestingly, urinary FAA appears to correlate with the degree of predisposition induced by sub-nephrotoxic cisplatin. Indeed, the level of FAA in the urine predicts the extent of renal dysfunction when AKI is produced by a second nephrotoxin (in our case gentamicin). The other urinary proteins found to be increased by cisplatin showed no correlation with the level of predisposition. Also, FAA was detected in the urine of patients treated with cisplatin, with different evolution patterns. We might only hypothesize that different FAA patterns may correspond to different patterns of renal susceptibility to AKI, which needs to be deeply evaluated in appropriate clinical studies. In conclusion, FAA is a potential biomarker candidate of predisposition to kidney injury induced by sub-toxic administration of cisplatin. As a near future challenge, clinical investigation

might open new useful means (based on these and other biomarkers) for detecting and most conveniently handling individuals at risk, in order to pre-emptively reduce the incidence and severity of drug-related AKI. 5. Conflict of interest The authors declare that there are no conflict of interest. 6. Transparency document The Transparency document associated with this article can be found in the online version. Acknowledgments This work was supported by grants from Junta de Castilla y León (BIO/SA20/14 and SA359U14 to AIMM) and from Instituto de Salud Carlos III (PI081900 and PI11/02278 to FJL-H; PI14/01776 to AIMM). JML-N’s research group holds the Excellence mention from the Junta de Castilla y León (Group GR100) with associated funds, which were partly used for this study. References Arany, I., Safirstein, R.L., 2003. Cisplatin nephrotoxicity. Semin. Nephrol. 23, 460–464. Bodenner, D.L., Dedon, P.C., Keng, P.C., Borch, R.F., 1986. Effect of diethyldithiocarbamate on cis-diamminedichloroplatinum(II)-induced cytotoxicity, DNA cross-linking, and gamma-glutamyl transpeptidase inhibition. Cancer Res. 46, 2745–2750. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. Dentino, M., Luft, F.C., Yum, M.N., Williams, S.D., Einhorn, E.H., 1987. Long therm effect of cis-diamminedichloride platinium (CDDP) on renal function and structure in man. Cancer 41, 1274–1281. Dobyan, D.C., Levi, J., Jacobs, C., Kosek, J., Weiner, M.W., 1980. Mechanism of cisplatinum nephrotoxicity: II. Morphologic observations. J. Pharmacol. Exp. Ther. 213, 551–556. Engineer, M.S., Bodey Sr., G.P., Newman, R.A., Ho, D.H., 1987. Effects of cisplatininduced nephrotoxicity on gentamicin pharmacokinetics in rats. Drug Metab. Dispos. 15, 329–334. Ferreira, L., Quiros, Y., Sancho-Martinez, S.M., Garcia-Sanchez, O., Raposo, C., LopezNovoa, J.M., Gonzalez-Buitrago, J.M., Lopez-Hernandez, F.J., 2011. Urinary levels of regenerating islet-derived protein III beta and gelsolin differentiate gentamicin from cisplatin-induced acute kidney injury in rats. Kidney Int. 79, 518–528. Gautier, J.C., Riefke, B., Walter, J., Kurth, P., Mylecraine, L., Guilpin, V., Barlow, N., Gury, T., Hoffman, D., Ennulat, D., Schuster, K., Harpur, E., Pettit, S., 2010. Evaluation of novel biomarkers of nephrotoxicity in two strains of rat treated with Cisplatin. Toxicol. Pathol. 38, 943–956. Goldstein, R.S., Mayor, G.H., 1983. Minireview. The nephrotoxicity of cisplatin. Life Sci. 32, 685–690. González-Escalada, J.R., de la Calle, J.L., Perucho, A., 1999. Un nuevo procedimiento diagnóstico y evaluativo del dolor. Rev. Soc. Esp. Dol. 6, 187–198. Gonzalez-Vitale, J.C., Hayes, D.M., Cvitkovic, E., Sternberg, S.S., 1978. Acute renal failure after cis-dichlorodiammineplatinum(II) and gentamicin–cephalothin therapies. Cancer Treat. Rep. 62, 693–698. Guinee Jr., D.G., van Zee, B., Houghton, D.C., 1993. Clinically silent progressive renal tubulointerstitial disease during cisplatin chemotherapy. Cancer 71, 4050– 4054. Hartmann, J.T., Knop, S., Fels, L.M., van Vangerow, A., Stolte, H., Kanz, L., Bokemeyer, C., 2000. The use of reduced doses of amifostine to ameliorate nephrotoxicity of cisplatin/ifosfamide-based chemotherapy in patients with solid tumors. Anticancer Drugs 11, 1–6. Herget-Rosenthal, S., Bokenkamp, A., Hofmann, W., 2007. How to estimate GFRserum creatinine, serum cystatin C or equations? Clin. Biochem. 40, 153–161. Jongejan, H.T., Provoost, A.P., Molenaar, J.C., 1989. Potentiated nephrotoxicity of cisplatin when combined with amikacin comparing young and adult rats. Pediatr. Nephrol. 3, 290–295. Kawamura, J., Soeda, A., Yoshida, O., 1980. Nephrotoxicity of cisdiamminedichloroplatinium (II) (cis-platinium) and the additive effect of antibiotics: morphological and functional observation in rats. Toxicol. Appl. Pharmacol. 58, 475–482. Kovach, J.S., Moertel, C.G., Schutt, A.J., Reitemeier, R.G., Hahn, R.G., 1973. Phase II study of cis-diamminedichloroplatinum (NSC-119875) in advanced carcinoma of the large bowel. Cancer Chemother. Rep. 57, 357–359.

L. Vicente-Vicente et al. / Toxicology Letters 234 (2015) 99–109 Liu, H., Baliga, R., 2003. Cytochrome P450 2E1 null mice provide novel protection against cisplatin-induced nephrotoxicity and apoptosis. Kidney Int. 63, 1687–1696. Lopez-Novoa, J.M., Santos, J.C., Villamediana, L.M., Garrote, F.J., Thim, L., Moody, A.J., Valverde, I., 1986. Renal catabolism of 125I-glicentin. Am. J. Physiol. 250, E545–550. McDuffie, J.E., Sablad, M., Ma, J., Snook, S., 2010. Urinary parameters predictive of cisplatin-induced acute renal injury in dogs. Cytokine 52, 156–162. Megyesi, J., Safirstein, R.L., Price, P.M., 1998. Induction of p21WAF1/CIP1/SDI1 in kidney tubule cells affects the course of cisplatin-induced acute renal failure. J. Clin. Invest. 101, 777–782. Milovic, M., Popov, I., Jezdic, S., Stojanovic, S., Stankovic, V., Radic, S., 2010. Monitoring levels of nephrotoxicity of different aminoglycosides during febrile neutropenia caused by nephrotoxic chemotherapy: a single centre study. J. BUON 15, 297–302. Monroy-Ruiz, J., Sevilla, M.A., Carron, R., Montero, M.J., 2011. Astaxanthin-enricheddiet reduces blood pressure and improves cardiovascular parameters in spontaneously hypertensive rats. Pharmacol. Res. 63, 44–50. Morales, A.I., Detaille, D., Prieto, M., Puente, A., Briones, E., Arevalo, M., Leverve, X., Lopez-Novoa, J.M., El-Mir, M.Y., 2010. Metformin prevents experimental gentamicin-induced nephropathy by a mitochondria-dependent pathway. Kidney Int. 77, 861–869. Morales, A.I., Vicente-Sanchez, C., Jerkic, M., Santiago, J.M., Sanchez-Gonzalez, P.D., Perez-Barriocanal, F., Lopez-Novoa, J.M., 2006. Effect of quercetin on metallothionein, nitric oxide synthases and cyclooxygenase-2 expression on experimental chronic cadmium nephrotoxicity in rats. Toxicol. Appl. Pharmacol. 210, 128–135. Passey, R.B., 1983. Glucose method comparison criticized. Clin. Chem. 29, 1320– 1321. Pierik, L.J., van Spronsen, F.J., Bijleveld, C.M., van Dael, C.M., 2005. Renal function in tyrosinaemia type I after liver transplantation: a long-term follow-up. J. Inherit. Metab. Dis. 28, 871–876. Quiros, Y., Ferreira, L., Sancho-Martinez, S.M., Gonzalez-Buitrago, J.M., Lopez-Novoa, J.M., Lopez-Hernandez, F.J., 2010. Sub-nephrotoxic doses of gentamicin predispose animals to developing acute kidney injury and to excrete ganglioside M2 activator protein. Kidney Int. 78, 1006–1015. Ramesh, G., Reeves, W.B., 2003. TNFR2-mediated apoptosis and necrosis in cisplatin-induced acute renal failure. Am. J. Physiol. Renal Physiol. 285, F610–618. Ramesh, G., Reeves, W.B., 2005. p38 MAP kinase inhibition ameliorates cisplatin nephrotoxicity in mice. Am. J. Physiol. Renal Physiol. 289, F166–174. Ries, F., Klastersky, J., 1986. Nephrotoxicity induced by cancer chemotherapy with special emphasis on cisplatin toxicity. Am. J. Kidney Dis. 8, 368–379.

109

Robbins, M.E., Bywaters, T.B., Jaenke, R.S., Hopewell, J.W., Matheson, L.M., Tothill, P., Whitehouse, E., 1992. Long-term studies of cisplatin-induced reductions in porcine renal functional reserve. Cancer Chemother. Pharmacol. 29, 309–315. Robbins, M.E., Campling, D., Whitehouse, E., Hopewell, J.W., Michalowski, A., 1990. Cisplatin-induced reductions in renal functional reserve uncovered by unilateral nephrectomy: an experimental study in the pig. Cancer Chemother. Pharmacol. 27, 211–218. Rootwelt, H., Chou, J., Gahl, W.A., Berger, R., Coskun, T., Brodtkorb, E., Kvittingen, E.A., 1994. Two missense mutations causing tyrosinemia type 1 with presence and absence of immunoreactive fumarylacetoacetase. Hum. Genet. 93, 615–619. Salem, P.A., Jabboury, K.W., Khalil, M.F., 1982. Severe nephrotoxicity: a probable complication of cis-dichlorodiammineplatinum (II) and cephalothingentamicin therapy. Oncology 39, 31–32. Sanchez-Gonzalez, P.D., Lopez-Hernandez, F.J., Lopez-Novoa, J.M., Morales, A.I., 2011. An integrative view of the pathophysiological events leading to cisplatin nephrotoxicity. Crit. Rev. Toxicol. 41, 803–821. Scheeren, T.W., Martin, K., Maruschke, M., Hakenberg, O.W., 2011. Prognostic value of intraoperative renal tissue oxygenation measurement on early renal transplant function. Transpl. Int. 24, 687–696. Singbartl, K., Kellum, J.A., 2012. AKI in the ICU: definition, epidemiology, risk stratification, and outcomes. Kidney Int. 81, 819–825. Skinner, R., Pearson, A.D., Coulthard, M.G., Skillen, A.W., Hodson, A.W., Goldfinch, M. E., Gibb, I., Craft, A.W., 1991. Assessment of chemotherapy-associated nephrotoxicity in children with cancer. Cancer Chemother. Pharmacol. 28, 81–92. Slomowitz, L.A., Deng, A., Hammes, J.S., Gabbai, F., Thomson, S.C., 2002. Glomerulotubular balance, dietary protein, and the renal response to glycine in diabetic rats. Am. J. Physiol. Regul. Integr. Comp. Physiol. 282, R1096–1103. Vaidya, V.S., Ozer, J.S., Dieterle, F., Collings, F.B., Ramirez, V., Troth, S., Muniappa, N., Thudium, D., Gerhold, D., Holder, D.J., Bobadilla, N.A., Marrer, E., Perentes, E., Cordier, A., Vonderscher, J., Maurer, G., Goering, P.L., Sistare, F.D., Bonventre, J.V., 2010. Kidney injury molecule-1 outperforms traditional biomarkers of kidney injury in preclinical biomarker qualification studies. Nat. Biotechnol. 28, 478–485. Van Berendoncks, A.M., Elseviers, M.M., Lins, R.L., 2010. Outcome of acute kidney injury with different treatment options: long-term follow-up. Clin. J. Am. Soc. Nephrol. 5, 1755–1762. Williams, R.E., Lock, E.A., 2004. D-serine-induced nephrotoxicity: possible interaction with tyrosine metabolism. Toxicology 201, 231–238. Winston, J.A., Safirstein, R., 1985. Reduced renal blood flow in early cisplatininduced acute renal failure in the rat. Am. J. Physiol. 249, F490–496. Womer, R.B., Pritchard, J., Barratt, T.M., 1985. Renal toxicity of cisplatin in children. J. Pediatr. 106, 659–663.