Influence of polymorphisms of ABCB1 and ABCC2 on mRNA ... - Nature

The Pharmacogenomics Journal (2007) 7, 56–65 & 2007 Nature Publishing Group All rights reserved 1470-269X/07 $30.00 www.nature.com/tpj

ORIGINAL ARTICLE

Influence of polymorphisms of ABCB1 and ABCC2 on mRNA and protein expression in normal and cancerous kidney cortex S Haenisch1,4, U Zimmermann2, E Dazert3, CJ Wruck1, P Dazert4, S Siegmund4, HK Kroemer4, RW Warzok3 and I Cascorbi1,4 1 Institute of Pharmacology, University Hospital Schleswig-Holstein, Kiel, Germany; 2Department of Urology, University of Greifswald, Greifswald, Germany; 3Institute of Pathology, University of Greifswald, Greifswald, Germany and 4Institute of Pharmacology, University of Greifswald, Greifswald, Germany

Correspondence: Dr I Cascorbi, Institute of Pharmacology, University Hospital Schleswig-Holstein, Campus Kiel, Hospitalstrasse 4, D-24105 Kiel, Germany. E-mail: [email protected]

There is increasing evidence that polymorphisms of the adenosine 50 triphosphate membrane transporters ABCB1 (P-glycoprotein, MDR1) may affect expression and function, whereas less information is available about the impact of ABCC2 (multidrug resistance-associated protein (MRP2)) single-nucleotide polymorphisms . Particularly, their role in human kidney for drug elimination and in the etiology of renal cell carcinoma is poorly understood. ABCB1 and ABCC2 mRNA and protein expression levels were determined by real-time polymerase chain reaction or immunohistochemistry in kidney cancer and adjacent unaffected cortex tissue of 82 nephrectomized renal cell cancer (RCC) patients (63 clear-cell RCC (CCRCC), 19 non-CCRCC). The DNA of all patients was genotyped for ABCB1 2352G4A, 692T4C, 2677G4T/A (Ala893Ser/Thr), and 3435C4T, and ABCC2 24C4T, 1249G4A (Val417Ile) and 3972C4T. ABCB1 and ABCC2 were less expressed in CCRCC than in normal cortex on mRNA as well as on protein level. Although the overall genotype frequency distribution did not differ between the patients and a matched control group, ABCB1 2677T/ A and 3435T genotypes were associated with higher (P ¼ 0.02 and P ¼ 0.04) and ABCC2 24 T with lower mRNA levels in normal tissues (0.03). The expression of ABCB1 and ABCC2 was not related to genetic variants in RCC tissue. In a reporter gene assay in HepG2 cells, the ABCC2 24T construct showed an 18.7% reduced activity (P ¼ 0.003). In conclusion, ABCB1 and ABCC2 genotypes modulate the expression in the unaffected renal cortex of RCC patients, possibly contributing to inter-individual differences in drug and xenobiotics elimination. Their role in RCC cancer susceptibility or chemotherapy resistance needs further elucidation. The Pharmacogenomics Journal (2007) 7, 56–65. doi:10.1038/sj.tpj.6500403; published online 20 June 2006 Keywords: ATP membrane transporter; P-glycoprotein; multidrug resistance; SNP; renal cell carcinoma; xenobiotics

Introduction

Received 5 December 2005; revised 28 March 2006; accepted 2 May 2006; published online 20 June 2006

The adenosine 50 triphosphate-binding cassette transmembrane proteins Pglycoprotein (P-gp, MDR1) and multidrug resistance-associated protein (MRP2) are expressed at certain compartment barriers like liver, small intestine, placenta and the blood–brain barrier,1–4 where they play an important role in absorption, distribution and excretion.5 In the human kidney, these transporters are expressed in the apical luminal membrane of epithelial proximal tubules cells,6,7 where they actively excrete

Polymorphic expression of ABCB1 and ABCC2 in kidney cancer S Haenisch et al

57

various substances and their organic anion conjugates. P-gp and MRP2 play a relevant role for detoxification, thus protecting tubule cells against potential carcinogens, but are also believed to contribute to chemotherapy resistance.8,9 P-gp is encoded by the ABCB1 (multidrug resistance protein 1 (MDR1)) gene, localized on chromosome 7q21, and MRP2 by the ABCC2 gene, localized on chromosome 10q23–24. Both genes exhibit various genetic polymorphisms. There was a significant correlation of the C3435T polymorphism in exon 26 of the ABCB1 gene with lower intestinal P-gp expression levels and elevated oral bioavailability of digoxin.10 The differences were more evident considering haplotypes of 2677G4T and 3435C4T.11,12 Very recently, it was shown that the 3435C4T polymorphism may influence mRNA stability.13 The influence of ABCB1 polymorphisms on expression and function in kidney tissue was examined in a Japanese sample, showing no influence on mRNA expression in normal kidney cortex as well as in renal cell carcinoma (RCC) tissue,14 whereas in a German sample, a strong association of ABCB1 3435C4T to the risk of non-clear-cell renal cell carcinoma (non-CCRCC) was described.15 To our knowledge, there is no data available on the significance of ABCC2 variants to MRP2 expression or function in normal renal tissue or in renal cancer. Kidney cancers contribute to 2% of all malignant tumors and 85% occur as RCCs in adults. These neoplasms originate mostly from mutant epithelial proximal tubule cells, rarely from collecting duct cells and are characterized by lack of early warning signs and relatively strong resistance against radiotherapy and chemotherapy.16 RCCs may be histologically subdivided into two groups: CCRCC, having a prevalence of 75–85%, and non-CCRCC that consist of chromophilic (papillary) (12–14%), chromophobic (5%), oncocytic (2–4%) and collecting duct (Bellini’s duct) (1%) tumors.17 Environmental factors like smoking and occupational exposure to industrial chemicals are also considered to be involved in the development of RCC.18–20 Hence, the question arises, whether differences of detoxification processes like metabolizing enzymes and membrane transporters may contribute to inter-individually varying disease risk. In this study, we determined the distribution of major ABCB1 and ABCC2 hereditary polymorphisms, including the recently described ABCB1 2352G4A and 692T4C polymorphisms and the protein and mRNA expression of these membrane transporters in the normal kidney cortex and adjacent cancerous tissue obtained from 82 Caucasian RCC patients. Additionally, we investigated the functional significance of the ABCC2 24C4T polymorphism using a reporter gene assay.

Materials and methods Subjects Eighty-two RCC patients with a mean age of 61.8 years (45 men, 37 women) of the Department of Urology, University Hospital Greifswald were enrolled. Patients were German

Whites from North-East Germany. Seventeen patients were current smokers. Five underwent treatment with cytostatics 6 months before nephrectomy. After independent histological observation, the tumor tissues were classified into 63 CCRCC and 19 non-CCRCC, subdivided in chromophilic (papillary) (n ¼ 7), mixed (n ¼ 11) and chromophobic types (n ¼ 1). Thirty-five tumors were assigned to tumor node metastasis stage I, 29 stage II, six stage III and nine stage IV. For comparison of the genotype distribution, subjects were matched 1:2 by gender to 164 controls (90 men and 74 women), being healthy Caucasian volunteers with a mean age of 26 years from the Department of Clinical Pharmacology, University of Greifswald. Cancerous and adjacent noncancerous tissue of each patient and blood samples of controls were genotyped for the ABCB1 polymorphisms 2352G4A, 692T4C, 2677G4T/A (Ala893Ser/Thr), 3435C4T (silent), and for 24C4T, 1249G4A (Val417Ile) and 3972C4T (silent) in the ABCC2 gene. DNA/RNA isolation Venous blood was obtained from 164 controls and DNA was extracted using a QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany). DNA from all 82 kidney cancer patients was isolated from homogenized frozen cancerous and adjacent non-cancerous tissues of kidney cortex using DNeasy Tissue Kit (Qiagen). RNA could be prepared from 79 non-cancerous and 71 cancerous (54 CCRCC and 17 nonCCRCC) tissues by RNeasy Tissue Kit (Qiagen). Genotyping ABCB1 polymorphisms 2677G4T/A and 3435C4T genotyping was conducted by polymerase chain reaction/restriction fragment length polymorphism (PCR/RFLP) according to Cascorbi et al.21 Moreover, the recently described ABCB1 2352G4A and 692T4C polymorphisms22 were genotyped under same conditions with the following polymerase chain reaction (PCR) profile: 35 cycles of 941C 30 s, 601C 30 s and 721C 30 s (primers and restriction enzymes are given in Table 1). DNA amplifications for ABCC2 fragments were performed under similar conditions for each reaction with 5 pmol of specific primers (TIB Molbiol, Berlin, Germany), 1.0 mmol/l of desoxyribonucleoside triphosphate (Biozym, Hessisch Oldendorf, Germany), 2.5 mmol/l MgCl2 (3972 C4T, 2.0 mmol/l), 1 ml dimethyl sulfoxide 99.9% (SigmaAldrich, Steinheim, Germany) (24C4T, 3972C4T) and 0.75 U of Taq DNA polymerase (Invitrogen, Karlsruhe, Germany) in a total volume of 25 ml. The initial denaturation was performed for 2 min at 941C (C3972T: 5 min), followed by 35 cycles of 30 s at 941C, 30 s at 541C (24C4T), 601C (1249G4A), or 571C (3972C4T), and 30 s at 721C. All PCR amplifications were carried out using a GeneAmp 9700 thermocycler (Applied Biosystems, Darmstadt, Germany). DNA fragments generated after restriction enzyme digestion (New England Biolabs, Frankfurt/Main, Germany) were separated on a 2.5% agarose gel (AppliChem, Darmstadt, Germany) and visualized after ethidium bromide staining by a digital image station (Kodak, Stuttgart, Germany). Correctness of genotyping was confirmed by DNA sequencing for

The Pharmacogenomics Journal


58

Table 1 Primers and restriction endonucleases used for genotyping of novel ABCB1 (MDR1) and ABCC2 (MRP2) variants and primers and fluorescent probes used for cDNA real-time quantification Gene

SNP

Name of primer

Primer sequence

Restriction enzyme

Fragment length (bp) mut/wt

ABCB1

T-692C

MDR1-692f MDR1-692r MDR1-2352f MDR1 -2352r* mdr1 forward mdr1 reverse Probe MRP2-Pro MRP2-ProR2 MRP2-10* MRP2-10R MRP2-28 MRP2-28R* mrp2 forward mrp2 reverse Probe

50 -CTA GAG AGG TGC AAC GGA AAG 50 -TAG TAG CTC CCA GCT TTG CG 50 -TTT ACC TGA TGC TCA AGA TTG TAG 50 - TCA CTT TTG TTT TGC TTT GTT GCT TGA T 50 -TTC GCA ACC CCA AGA TCC TC 50 -ACA ATG GTG GTC CGA CCT TT 50 -6FAM-ATC CAG AGC CAC CTG AAC CAC TGC T XT p 50 -TAA ATG GTT GGG ATG AAA GG 50 -GCT TTA GAC CAA TTG CAC ATC 50 -AAC TTG GCC AGG AAG GAG TAC AAC 50 -CTG GGT GAC TTT TTC TTT ACC TGA ATG 50 -AAC TTA CTT CTC ATC TTG TCT CCT TGC 50 -CTC CAC CTA CCT TCT CCA TGC TAG 50 -CTG GGA ACA TGA TTC GGA AGC 50 -GAG GAT TTC CCA GAG CCG AC 50 -6FAM-CAG TCC GAG ATG TGA ACC TGG ACA T XT p

MspA1I

32 117 208 117 240 179 144 35

G-2352A ABCB1

ABCC2

C-24T G1249A C3972T ABCC2

MboI

BbsI AclI BsrDI

301 188 113 260 236 24 156 28 184

Abbreviations: MDR1, multidrug resistance protein 1; MRP2, multidrug resistance-associated protein. Exchanged DNA nucleotides of designed primers are marked as bold characters.

all genotypes investigating two cases of each. Moreover, 10 randomly selected DNA samples of RCC patients and 20 of controls were blindly repeated by PCR/RFLP achieving identical results, respectively. ABCB1 and ABCC2 mRNA quantification ABCB1 and ABCC2 mRNA expression was determined by real-time PCR using TaqMan technology (ABI Prism 7700 SDS, Applied Biosystems) as described recently.23,24 Primers and fluorescent probes for cDNA quantification are listed in Table 1. In order to select the most suitable reference gene, we quantified 18S ribosomal RNA (rRNA), b-actin and RPLP0 (ribosomal protein, large, P0) in 14 cancerous and adjacent non-cancerous tissue samples. After analysis using the GeNorm version 3.2c program,25 18S rRNA was identified as the most constitutively expressed internal standard gene in these non-cancerous and cancerous tissues. Every experiment was conducted in duplicate. P-gp and MRP2 immunohistochemistry Immunohistochemistry was carried out in cancerous and adjacent non-cancerous paraffin-embedded kidney cortex tissue sections (2 mm) of 23 patients, with the exception of one subject, where only tumorous tissue was available. Renal P-pg and MRP2 were stained with the monoclonal antibodies JSB-1 (Alexis, Gruenberg, Germany) and MRP2 III-6 (Alexis, Gruenberg, Germany) with the labeled streptavidin– biotin detection method. Prepared slides were deparaffinized, rehydrated and steamed for 5 min in sodium citrate buffer (pH 7). After a 5-min incubation in 3% H2O2, samples were blocked for 10 min with avidin and biotin (Avidin–Biotin-Blocking-Kit


SP-2001, Linaris, Wertheim-Bettigen, Germany) and for 30 min in 3% horse serum (Linaris). Samples were incubated overnight with P-gp (JSB-1 1:50) and MRP2 (MRP2 III-6 1:20) primary antibodies and subsequently for 30 min with horse anti-mouse antibody (rat adsorbed, 1:100, Linaris). From each section, five digital pictures were made randomly at different places (3CCD color camera, Hitachi HV-C20M; ¨ sseldorf, Tokyo, Japan or Nikon Eclipse 400, Nikon, Du Germany) (magnification, 200) (Figure 1). Optical density was measured using a KS400 imaging system (V3.0; Zeiss, ¨ nchen, Germany). The ratios of protein-specific pixels to Mu total pixels of background were multiplied by the intensity of the specific signal. Values are given as means of five different regions of each tissue. Staining and evaluation were carried out in a blinded manner in one run with identical staff, equipment and chemicals. Cloning of reporter gene vector constructs To investigate the functional significance of the ABCC2 24C4T polymorphism, PCR products including 1229 bp of the 50 -untranslated region (50 -UTR) of ABCC2 gene were amplified from human genomic DNA (Accession No. AL133353, positions 130 178–131 406) using the forward primer 50 -CTC GGT AC*C TCT AGA ATT TGA CCA GAT TTA AGG CCA ATT T containing a KpnI restriction site and the reverse primers 50 - GAG C*CA TGG ATT CCT GGA CTG CGT CTG GAA CGA AGA CTC and 50 - GAG C*CA TGG ATT CCT GGA CTG CGT CTG GAA CAA AGA CTC containing NcoI restriction sites each (* represents the location of a restriction site and bold characters represent wild type or variant SNP). The wild-type (ABCC2-C) and the variant (ABCC2-T) fragment were ligated into the pGL3 basic vector (Promega, Mannheim, Germany) being in-frame of the translation


59

Figure 1 P-gp expression (JSB-1 stained) in normal kidney cortex (a) and in clear-cell carcinoma tissue (b) of the same patient. (c) and (d) Demonstrate MRP2 expression (MRP2-III stained) in non-affected (c) and CCRCC tissue (d) of kidney cortex of another patient. Black arrows indicate positive reaction of JSB1 and MRP2-III antibodies in brush-border membranes of proximal tubule epithelial cells, respectively.

start of the firefly luciferase gene. Correctness of plasmids was verified by control sequencing using the sequence primers 50 -CTA GCA AAA TAG GCT GTC CC (RV3) and 50 ATC TTC CAG CGG ATA GAA TGG (GL2plus). Cell culture and transient transfection Human hepatoblastoma HepG2 cells (DSMZ, Braunschweig, Germany) were cultured in Eagle’s minimal essential medium (MEM) with Earle’s salts with L-glutamine spiked with 10% fetal bovine serum, 1% essential amino acids, 1% ¨ lbe, Germany) and 1% pyruvate (PAA Laboratories, Co penicillin/streptomycin (Gibco Invitrogen, Karlsruhe, Germany), at 371C in a 5% CO2 atmosphere. Cells were transfected by using Lipofectamine 2000 (Invitrogen, Groningen, the Netherlands) according to the manufacturer’s recommendations. Four hours after transfection, the medium was changed. Reporter gene assay HepG2 cells were plated on 96-well plates, 5000 cells/well, in 100 ml medium. Reporter plasmids pGL3-ABCC2-C or pGL3ABCC2-T (0.1 mg/well) containing the firefly luciferase reporter gene and 0.02 mg/well of the phRL-TK plasmid (Promega), containing the renilla luciferase reporter gene under the control of the herpes simplex virus thymidine kinase promoter as an internal control, were co-transfected. Additional transfections of pGL3basic vector or pGL3control vector (Promega), both also co-transfected with phRL-TK,

were used as negative and positive control, respectively. The activities of both firefly and renilla luciferases were determined 24 h after transfection with the Dual-Luciferase Reporter Assay System (Promega). The luciferase activities were normalized to the renilla luciferase activity of the internal control. Samples were analyzed by Veritas Software and Veritas microplate luminometer (Tuner Biosystems, Sunnyvale, CA, USA). We conducted three experiments with identical conditions. In each experiment, each transfection was carried out eight times. Statistical analysis Kruskal–Wallis test and Mann–Whitney U-test (multiple comparisons) were performed to test mRNA and protein expression differences dependent on genotype or histology using computer software SPSS (version 12.0) and Statcalc (version 6) for calculation of odds ratios, respectively. A Pvalue below 0.05 was considered to be significant and calculated as two-sided significance. Haplotype analysis was performed by using the EH program.26 Adjustment of multiple testing was performed by Bonferroni–Holm method. Results Expression of membrane transporters The expression of both membrane transporters, ABCB1 (MDR1) and ABCC2 (MRP2), was significantly lower in



60

cancerous tissues compared to normal cortex tissues on immunohistochemically determined protein level (P-gp: Po0.001; MRP2: Po0.001). This was also evident on mRNA level standardized to ribosomal 18S rRNA for ABCB1 (P ¼ 0.026), and with a tendency for ABCC2 (P ¼ 0.056). However, the ABCB1 mRNA/18S rRNA median values in normal tissue were higher than in CCRCC (2.2 103 vs 1.0 103, P ¼ 0.017). A P-value of 0.016 was considered to be significant after adjustment for multiple testing. On the other hand, there was no difference compared to nonCCRCC (P ¼ 0.51) (Figure 2a). The P-gp content in the 19 CCRCC samples was 70 times lower compared to normal cortex tissue (Po0.001), but there was no difference between the four non-CCRCC samples and the normal tissue samples (P ¼ 0.62) (Figure 2b). In contrast, there was no difference on ABCC2 mRNA level between CCRCC and tumor-free adjacent tissue (P ¼ 0.282), but a 1.55-fold lower ABCC2 mRNA concentration in the 17 non-CCRCC samples (P ¼ 0.009) (Figure 2c). The MRP2 protein levels were significantly lower in CCRCC samples and non-CCRCC compared to the normal tissue, Po0.001 and P ¼ 0.002, respectively (Figure 2d). ABCB1 and ABCC2 mRNA expression in non-cancerous tissue differed significantly dependent on the tumor stage. Tumor stages I and IV exhibited higher relative mRNA levels

than stages II and III (ABCB1: P ¼ 0.049 and ABCC2: P ¼ 0.004, respectively, Kruskal–Wallis test), but these findings did not reach significance on protein level. Smoking and chemotherapy did not influence the expression of ABCB1 and ABCC2 in this study. However, there was no correlation of relative mRNA expression and protein content neither in normal (ABCB1 r ¼ 0.142, P ¼ 0.528; ABCC2 r ¼ 0.407, P ¼ 0.06) nor in tumorous tissue (ABCB1 r ¼ 0.275, P ¼ 0.215; ABCC2 r ¼ 0.148, P ¼ 0.511). Distribution of genetic polymorphisms The genotype distribution of all polymorphisms investigated was in Hardy–Weinberg equilibrium. The ABCB1 2352G4A and 692T4A polymorphisms in the promoter region had an allele frequency of 0.6 and 2.0% in patients, and 0.9 and 3.0% in controls, respectively. The allelic frequency of the putatively functional relevant exon 26 3435C4T polymorphism was 53.7% in controls, 60.3% in CCRCC (P ¼ 0.20) and 47.4% in non-CCRCC (P ¼ 0.46). The odds ratio of genotypes 3435 TT/CT vs CC in all cancer cases vs controls was 1.30 (95% CI 0.64–2.69; P ¼ 0.44) (Table 2). Considering both, ABCB1 2677G4T/A and 3435C4T, we could identify four haplotypes in patients and six haplotypes in controls (Table 3). Owing to the low frequency of

Figure 2 Relative ABCB1 mRNA expression (a) and P-gp expression (b), relative ABCC2 mRNA expression (c) and ABCC2 protein expression (d) in non-cancerous and cancerous tissues subdivided in CCRCC and non-CCRCC tissues. (P-values calculated by Mann–Whitney U-test.)



61

Table 2

Allele frequencies and frequencies of genotypes in 82 patients and in 164 controls

Location

Allele

Allele frequency RCC

Controls

MDR1 G-2352A

G A

99.4 0.6

99.1 0.9

T-692C

T C

98.0 2.0

97.0 3.0

G2677T/A

G T A

58.5 39.6 1.8

54.3 43.6 2.1

C3435T

C T

42.7 57.3

46.3 53.7

C T

78 22

81.4 18.6

G1249A

G A

83.5 16.5

77.8 22.2

C3972T

C T

62.8 37.2

69.2 30.8

MRP2 C-24T

Genotype

Frequency of genotypes

Odds ratio

P-value

RCC

controls

95% CI

GG GA AA TT TC CC GG GT TT GA TA AA CC CT TT

98.8 1.2 0.0 96.3 3.7 0.0 35.4 42.7 18.3 3.7 0.0 0.0 18.3 48.8 32.9

98.2 1.8 0.0 93.3 6.7 0.0 32.3 41.5 22.0 2.4 1.8 0.0 22.6 47.6 29.9

0.66 (0.03–7.29)

0.72

0.53 (0.11–2.12)

0.33

0.87 (0.48–1.59)

0.63

1.3 (0.64–2.69)

0.44

CC CT TT GG GA AA CC CT TT

64.6 26.8 8.5 67.1 32.9 0.0 41.5 42.7 15.9

68.9 25.0 6.1 61.6 32.3 6.1 51.2 36.0 12.8

1.21 (0.67–2.20)

0.50

0.79 (0.43–1.43)

0.40

1.48 (0.84–2.62)

0.15

Abbreviations: MDR1, multidrug resistance protein 1; MRP2, multidrug resistance-associated protein; RCC, renal cell carcinoma. Odds ratios were calculated with wild types as reference.

2352A and 692C, we did not include these singlenucleotide polymorphisms (SNPs) into haplotype calculation. ABCB1 haplotypes H1 (wild-type 2677G/3435C), H2 (2677G/3435T) and H4 (2677T/3435T) represented 98.2% of all haplotypes in patients and 91.8% in controls. Comparison of haplotype frequencies showed a slight difference between patients and controls for ABCB1 (w2 ¼ 11.63, P ¼ 0.04), but not for ABCC2 (w2 ¼ 6.98, P ¼ 0.128). Interestingly, we could not find any ABCB1 H3 haplotype in RCC patients. In RCC cases, these three ABCB1 haplotypes occurred in six diplotypes: H1/H1 15.8%, H1/H2 18.3%, H1/H4 29.3%, H4/H4 18.3%, H2/H4 13.4% and H2/H2 1.2%. In contrast, the estimation model did not enable exact allocation of defined diplotypes to every control subject, as some heterozygous genotypes could resolve from different combinations of the six haplotypes estimated in the volunteers. Analyzing the ABCC2 SNPs 24C4T, 1249G4A (Val417Ile) and 3972C4T, a lower frequency of the 1249A variant, coding for isoleucin, was found in clear-cell carcinomas (15.1 vs 22.2%), however without reaching statistical significance (P ¼ 0.09). Interestingly, no homozygote 1249A carrier was found in RCC patients (Table 2). Haplotype analysis resulted in four ABCC2 haplotypes in patients and six haplotypes in controls (Table 3). ABCC2

haplotypes H1, H3, H5 and H6 formed nine diplotypes in patients: H1/H1 23.2%, H1/H3 18.3%, H1/H5 12.2%, H1/H6 15.9%, H3/H5 9.8%, H3/H6 4.9%, H5/H5 1.2%, H5/H6 6.1% and H6/H6 8.5%. In controls, owing to the diversity of haplotypes, definite diplotypes could not be assigned to using this estimation model.

Impact of polymorphisms on transporter expression There was a clear correlation between ABCB1 polymorphisms 2677G4T/A and 3435C4T with mRNA expression in the non-cancerous kidney cortex tissue (P ¼ 0.02 and P ¼ 0.04, respectively, Kruskal–Wallis test). For example, samples carrying at least one 2677T/A allele showed higher mRNA expression levels than samples homozygous for 2677G (data not shown). Tissue samples homozygote for 3435C exhibited 1.8-fold lower ABCB1 mRNA/18S rRNA ratios than heterozygotes or 4.2-fold lower ratios than homozygote 3435T carriers (Figure 3a). Considering haplotypes of G2677T/A and C3435T, the 15 samples with the variant diplotype H4/H4 exhibited higher ABCB1 mRNA/ 18S rRNA ratios than H1/H1. The difference, however, was statistically not significant (P ¼ 0.134), but there were increased mRNA expression levels among ABCB1 H1/H4 (P ¼ 0.044) and ABCB1 H2/H4 (P ¼ 0.01) (Mann–Whitney



62

Table 3 Frequencies of ABCB1 and ABCC2 haplotypes in RCC patients and controls estimated with the EH program26 ABCB1 G2677T/ haplotype A

C3435T

Frequency RCC

1 2 3 4 5 6

ABCC2 haplotype

G G T T A A

C T C T C T

C-24T

G1249A

40.9 38.8 17.7 15.5 0.0 6.1 39.6 37.5 1.8 1.5 0.0 0.6 w2 ¼ 11.63, d.f. ¼ 5, P ¼ 0.04 C3972T

Frequency RCC

1 2 3 4 5 6 7 8

C T C T C T C T

G G A A G G A A

Controls

C C C C T T T T

Controls

45.8 47.0 0.0 1.4 16.4 20.8 0.0 0.0 15.8 12.2 22.0 17.2 0.0 1.4 0.0 0.0 w2 ¼ 8.55, d.f. ¼ 5, P ¼ 0.128

Twelve square w2 test was conducted to compare differences between haplotype frequencies of RCC and controls.

U-test) (Figure 4a). The changes of P-gp protein staining were without any statistical significance (P ¼ 0.46) (Figure 4b). In contrast to the data in normal renal tissue, there were no genotypes or diplotype-related differences of mRNA expression in both types of RCC (CCRCC and non-CCRCC) tissues (data not shown). On protein level, we observed no significant association of ABCB1 variants with P-gp expression neither in noncancerous nor in cancerous tissues. In normal cortex tissues, ABCC2 SNP 24C4T, but not 1249G4A and 3972C4T, correlated significantly with mRNA levels. Homozygote 24C allele carriers had 1.94 and 1.43 times higher median ABCC2 mRNA/18S rRNA ratios than heterozygotes and homozygotes 24T samples (P ¼ 0.03 Kruskal–Wallis test) (Figure 3b). This association could not be proven in CCRCC and non-CCRCC tissues (data not shown). There was also no correlation between ABCC2 SNPs and protein expression in all tissue samples investigated. Considering diplotypes, only ABCC2 H1/H1 had a tendency to express higher mRNA levels than homozygote carriers of ABCC2 H6 in non-tumorous tissue (P ¼ 0.053). In vitro investigation of functional significance of ABCC2 24C4T To investigate the functional influence of the 24C4T SNP in ABCC2, we created two reporter gene vector constructs pGL3-ABCC2-C and pGL3-ABCC2-T. Both constructs showed


Figure 3 ABCB1 mRNA expression in ABCB1 C3435T genotypes in non-cancerous tissues (a) (P ¼ 0.043), (Kruskal–Wallis test). Relative ABCC2 mRNA expression in ABCC2 C-24T genotypes in non-cancerous tissues (b) (P ¼ 0.032) (Kruskal–Wallis test).

a significant increase of luciferase activity compared to the negative control vector pGL3 basic. The pGL3-ABCC2-T provided a relative mean luciferase activity of 81.3% (720.0% s.d.) as compared to pGL3-ABCC2-C (100%) (P ¼ 0.003).

Discussion In this study, we examined the expression of the ABC membrane transporters ABCB1 (P-gp, MDR1) and ABCC2 (MRP2), the genotype frequency distribution of major polymorphisms and their association to mRNA and protein expression in kidney cortex tissue of nephrectomized RCC patients. Interestingly, the median ABCB1 mRNA and protein expression differed between healthy and cancer tissue, with lower levels among RCC tissue. These results are in line with the finding of Uwai et al.,14 who reported a similar difference


63

Figure 4 Relative (a) ABCB1 mRNA expression (P ¼ 0.067) and (b) ABCB1 protein expression (P ¼ 0.46) in ABCB1 diplotypes in noncancerous tissues (Kruskal–Wallis test).

in Japanese RCC tissue. However, splitting the data for the different histological subtypes, it became obvious that this difference was only owing to clear-cell carcinomas, but not for non-CCRCC tissue. There is some evidence that xenobiotics may interfere with signaling pathway involved in the activation of P-gp and MRP227–29 or lead to an elevated expression via PXR-transactivating elements.27,30 However, the lower ABCB1 expression of CCRCC tissue or ABCC2 in non-CCRCC could be also the consequence of altered gene regulation or low cell differentiation and not necessarily causative in the genesis of cancer.31,32

In contrast, in both RCC subtypes a significantly decreased MRP2 protein expression was observed as demonstrated before, particularly for clear-cell carcinoma with less differentiation status.7 However, this was not confirmed on mRNA level. These data suggest that P-gp and MRP2 are less expressed in tumor tissues than in healthy kidney tissue, but CCRCC and non-CCRCC differ by expression of P-gp expression. Cytogenetic characterization of RCCs demonstrated chromosomal aberrations such as trisomy 7 in chromophilic carcinomas.33 As the ABCB1 gene is localized on chromosome 7, it may be speculated whether chromosomal aberrations could contribute to higher expression rates in non-CCRCC tissues compared to CCRCC. The genotype of DNA from cancer tissue was identical with the genotype identified from normal tissue, indicating absence of novel mutations in the tumor. Moreover, our data did not indicate differences of allele or genotype frequencies between RCC patients and a control group of healthy volunteers, but the overall ABCB1 haplotype frequency differed between RCC and controls. However, this study was not designed and had no statistical power to investigate the association of ABCB1 and ABCC2 SNPs to renal cancer, but was intended to examine the influence of SNPs on transporter expression. Therefore, these data do not necessarily contradict the findings on an association of ABCB1 3435C4T and cancer risk that was obtained in a large sample of German non-CCRCC patients.15 Moreover, a limitation of our study was the fact that controls were not matched by age; however, the frequency distribution of the entire sample (extended of a previously published sample34) was identical to another large German sample of healthy volunteers.21 There was a clear gene–dose effect on ABCB1 mRNA expression in non-cancerous tissues, but not in the tumor itself. We found a significant trend of elevated mRNA expression among carriers of the variant genotype 2677T/A (P ¼ 0.02) and 3435T (P ¼ 0.04). A similar tendency was observed before in myocardial tissue35 and in the intestine of Japanese subjects.36 Overall ratios from mRNA/18S rRNA did not correlate with protein content neither in noncancerous nor in cancerous tissue. However, translation efficiency and membrane trafficking is not necessarily proportional to mRNA transcription efficiency37 and the estimated half-time of P-gp is reportedly 3.7 days,38 but the mRNA stability was shown to be much shorter.13 As a consequence, based on our findings, we can conclude that ABCB1 2677G4T/A, 3435C4T and ABCC2 24C4T influence the overall mRNA content in non-cancerous kidney tissue, but we could not detect an alteration of immunohistologically determined protein expression. However, the current data on the functional impact of ABCB1 polymorphisms is partly conflicting. A number of studies have found an association of higher P-gp activity with the 3435 C-variant,10,39,40 and very recently, a higher mRNA stability of the 3435C variant was reported,13 whereas some others showed lack of evidence or a weak association.34,41 The allelic frequencies of ABCB1 promoter polymorphisms 2352A and 692C were much lower in our



64

samples than in a Japanese population22 and did not contribute to expression variability. There are several studies showing that consideration of particular ABCB1 haplotypes may improve the approach to explain the inter-individual variability of digoxin bioavailability.11 2677G4T/A and 3435C4T are known to be in linkage disequilibrium,21,42 and in our study, 2677T was always linked with 3435T. The variant diplotype H4/H4 (homozygous 2677T/3435T) showed 4.1-fold elevated mRNA levels as compared to H1/H1 in non-cancerous tissue. As 2677G/3435T was earlier shown to be associated with elevated bioavailability of digoxin, suggesting impaired P-gp function,11 our data suggest that the regulation of ABCB1 expression may be different in renal cortex tissue or is altered owing to factors derived from adjacent cancer tissue. The genotype distribution ABCC2 SNPs investigated was identical in normal cortex tissue and in RCC samples and there was no evidence for an association to disease risk. The earlier described ABCC2 SNPs, namely 2302C4T (Arg768Trp), 2366C4T (Ser789Phe) and 4348G4A (Ala1450Thr),43 could not be found in our Caucasian volunteers (data not shown). 24C4T and 3972C4T were in strong linkage disequilibrium (100% in patients and 92.5% in controls) as described before for a Japanese sample.44 Interestingly, carriers homozygous for 24T exhibited 1.4 times lower median mRNA/ 18S rRNA ratios in non-cancerous cortex tissue than 24CC carriers (P ¼ 0.03). Moreover, subjects, homozygous for the haplotype 24T/3972T, exhibited a tendency of decreased ABCC2 mRNA expression than homozygote 24C/3972C carriers (P ¼ 0.053). The reporter gene assay in vitro data confirmed the in vivo finding, although the difference was less pronounced. Currently, however, the mechanism how the 24C4T exchange in the 50 -UTR of ABCC2 affects mRNA expression remains open. Possibly, the SNP could affect mRNA stability, as we could show previously for the endothelin gene.45 On the other hand, transcription efficiency could be altered by differential binding affinity of potential transcription factors. Although the promoter region of rat ABCC2 was thoroughly characterized, showing the importance of Y-box protein-1, binding 186 bp before translation start,46 the role of the human proximal ABCC2 50 -UTR for transcription efficacy remains to be elucidated. A limitation of our study was the fact that we were not able to study the physiological expression of transporters in kidney tissue of healthy individuals. Therefore, it remains open whether our findings in unaffected cortex tissue really reflect physiological conditions, or whether the expression might be influenced by the adjacent tumor. In general, RCC is considered to be highly resistant toward cytostatics therapy and P-gp and MRP2 overexpression was shown to be a cause of chemoresistance in several cell lines.8,9 In conclusion, ABCB1 (P-gp, MDR1) and ABCC2 (MRP2) were less expressed in tumor tissues than in non-cancerous adjacent renal cortex tissues on mRNA as well as on protein level. However, CCRCC and non-CCRCC tissues seem to differ not only histologically, but also non-CCRCC showed a significantly higher expression of P-gp as compared to CCRCC. It is unclear, however, whether the decreased


function of P-gp is the cause or consequence of tumor development.47 Although there was no evidence of an association between genotypes and cancer risk, we could show in the sample of 82 RCCs that ABCB1 genotypes correlated with ABCB1 mRNA expression and ABCC2 24C4T was associated with decreasing ABCC2 mRNA expression in non-cancerous renal cortex tissue, which could be confirmed by in vitro experiments. Further studies are required to elucidate whether there is a difference in cancer susceptibility or response to chemotherapy dependent on genotypes, especially with lower transporter expression.

Duality of Interest No duality of interest. Abbreviations 50 -UTR ABC MDR1 MRP2 P-gp SNP PCR/RFLP

50 -untranslated region adenosine 50 triphosphate-binding cassette multidrug resistance protein 1 multidrug resistance-associated protein P-glycoprotein single-nucleotide polymorphism polymerase chain reaction/restriction fragment length polymorphism

Acknowledgments The skilful technical assistance of Ingrid Geissler is gratefully acknowledged. We also thank Dr Amke Caliebe and Professor Michael Krawczak, Institute of Medical Informatics and Statistics, UKSH Campus Kiel, Germany, for valuable statistical advice. This work was supported by the German Federal Ministry for Education and Research (01 ZZ 0103).

References 1 Borst P, Evers R, Kool M, Wijnholds J. The multidrug resistance protein family. Biochim Biophys Acta 1999; 1461: 347–357. 2 Klein I, Sarkadi B, Varadi A. An inventory of the human ABC proteins. Biochim Biophys Acta 1999; 1461: 237–262. 3 Konig J, Nies AT, Cui Y, Leier I, Keppler D. Conjugate export pumps of the multidrug resistance protein (MRP) family: localization, substrate specificity, and MRP2-mediated drug resistance. Biochim Biophys Acta 1999; 1461: 377–394. 4 Young AM, Allen CE, Audus KL. Efflux transporters of the human placenta. Adv Drug Deliv Rev 2003; 55: 125–132. 5 Ambudkar SV, Dey S, Hrycyna CA, Ramachandra M, Pastan I, Gottesman MM. Biochemical, cellular, and pharmacological aspects of the multidrug transporter. Annu Rev Pharmacol Toxicol 1999; 39: 361–398. 6 Schaub TP, Kartenbeck J, Konig J, Vogel O, Witzgall R, Kriz W et al. Expression of the conjugate export pump encoded by the mrp2 gene in the apical membrane of kidney proximal tubules. J Am Soc Nephrol 1997; 8: 1213–1221. 7 Schaub TP, Kartenbeck J, Konig J, Spring H, Dorsam J, Staehler G et al. Expression of the MRP2 gene-encoded conjugate export pump in human kidney proximal tubules and in renal cell carcinoma. J Am Soc Nephrol 1999; 10: 1159–1169. 8 Cui Y, Konig J, Buchholz JK, Spring H, Leier I, Keppler D. Drug resistance and ATP-dependent conjugate transport mediated by the


65

9

10

11

12

13

14

15

16 17 18

19

20

21

22

23

24

25

26 27

apical multidrug resistance protein, MRP2, permanently expressed in human and canine cells. Mol Pharmacol 1999; 55: 929–937. Juliano RL, Ling V. A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim Biophys Acta 1976; 455: 152–162. Hoffmeyer S, Burk O, von Richter O, Arnold HP, Brockmoller J, Johne A et al. Functional polymorphisms of the human multidrug-resistance gene: multiple sequence variations and correlation of one allele with P-glycoprotein expression and activity in vivo. Proc Natl Acad Sci USA 2000; 97: 3473–3478. Johne A, Kopke K, Gerloff T, Mai I, Rietbrock S, Meisel C et al. Modulation of steady-state kinetics of digoxin by haplotypes of the P-glycoprotein MDR1 gene. Clin Pharmacol Ther 2002; 72: 584–594. Verstuyft C, Schwab M, Schaeffeler E, Kerb R, Brinkmann U, Jaillon P et al. Digoxin pharmacokinetics and MDR1 genetic polymorphisms. Eur J Clin Pharmacol 2003; 58: 809–812. Wang D, Johnson AD, Papp AC, Kroetz DL, Sadee W. Multidrug resistance polypeptide 1 (MDR1, ABCB1) variant 3435C4T affects mRNA stability. Pharmacogenet Genom 2005; 15: 693–704. Uwai Y, Masuda S, Goto M, Motohashi H, Saito H, Okuda M et al. Common single nucleotide polymorphisms of the MDR1 gene have no influence on its mRNA expression level of normal kidney cortex and renal cell carcinoma in Japanese nephrectomized patients. J Hum Genet 2004; 49: 40–45. Siegsmund M, Brinkmann U, Schaffeler E, Weirich G, Schwab M, Eichelbaum M et al. Association of the P-glycoprotein transporter MDR1(C3435T) polymorphism with the susceptibility to renal epithelial tumors. J Am Soc Nephrol 2002; 13: 1847–1854. Motzer RJ, Bander NH, Nanus DM. Renal-cell carcinoma. N Engl J Med 1996; 335: 865–875. Storkel S, van den BE. Morphological classification of renal cancer. World J Urol 1995; 13: 153–158. Brauch H, Weirich G, Hornauer MA, Storkel S, Wohl T, Bruning T. Trichloroethylene exposure and specific somatic mutations in patients with renal cell carcinoma. J Natl Cancer Inst 1999; 91: 854–861. Bruning T, Weirich G, Hornauer MA, Hofler H, Brauch H. Renal cell carcinomas in trichloroethene (TRI) exposed persons are associated with somatic mutations in the von Hippel–Lindau (VHL) tumour suppressor gene. Arch Toxicol 1997; 71: 332–335. Dhote R, Pellicer-Coeuret M, Thiounn N, Debre B, Vidal-Trecan G. Risk factors for adult renal cell carcinoma: a systematic review and implications for prevention. BJU Int 2000; 86: 20–27. Cascorbi I, Gerloff T, Johne A, Meisel C, Hoffmeyer S, Schwab M et al. Frequency of single nucleotide polymorphisms in the P-glycoprotein drug transporter MDR1 gene in white subjects. Clin Pharmacol Ther 2001; 69: 169–174. Taniguchi S, Mochida Y, Uchiumi T, Tahira T, Hayashi K, Takagi K et al. Genetic polymorphism at the 50 regulatory region of multidrug resistance 1 (MDR1) and its association with interindividual variation of expression level in the colon. Mol Cancer Ther 2003; 2: 1351–1359. Meissner K, Sperker B, Karsten C, Zu Schwabedissen HM, Seeland U, Bohm M et al. Expression and localization of P-glycoprotein in human heart: effects of cardiomyopathy. J Histochem Cytochem 2002; 50: 1351–1356. Vogelgesang S, Kunert-Keil C, Cascorbi I, Mosyagin I, Schroder E, Runge U et al. Expression of multidrug transporters in dysembryoplastic neuroepithelial tumors causing intractable epilepsy. Clin Neuropathol 2004; 23: 223–231. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 2002; 3: RESEARCH0034.1–RESEARCH0034.11. Terwilliger JDaOJ. Handbook of Human Genetic Linkage. Johns Hopkins University Press: Baltimore, MD, 1994. Ernest S, Rajaraman S, Megyesi J, Bello-Reuss EN. Expression of MDR1 (multidrug resistance) gene and its protein in normal human kidney. Nephron 1997; 77: 284–289.

28

29

30

31

32

33

34

35

36

37 38

39

40

41 42

43

44

45

46

47

Maher JM, Cheng X, Slitt AL, Dieter MZ, Klaassen CD. Induction of the multidrug resistance-associated protein family of transporters by chemical activators of receptor-mediated pathways in mouse liver. Drug Metab Dispos 2005; 33: 956–962. Miller DS. Xenobiotic export pumps, endothelin signaling, and tubular nephrotoxicants – a case of molecular hijacking. J Biochem Mol Toxicol 2002; 16: 121–127. Bauer B, Hartz AM, Fricker G, Miller DS. Pregnane X receptor up-regulation of P-glycoprotein expression and transport function at the blood–brain barrier. Mol Pharmacol 2004; 66: 413–419. Denson LA, Auld KL, Schiek DS, McClure MH, Mangelsdorf DJ, Karpen SJ. Interleukin-1beta suppresses retinoid transactivation of two hepatic transporter genes involved in bile formation. J Biol Chem 2000; 275: 8835–8843. Miller DS, Sussman CR, Renfro JL. Protein kinase C regulation of p-glycoprotein-mediated xenobiotic secretion in renal proximal tubule. Am J Physiol 1998; 275: F785–F795. Kovacs G, Fuzesi L, Emanual A, Kung HF. Cytogenetics of papillary renal cell tumors. Genes Chromosome Cancer 1991; 3: 249–255. Siegmund W, Ludwig K, Giessmann T, Dazert P, Schroeder E, Sperker B et al. The effects of the human MDR1 genotype on the expression of duodenal P-glycoprotein and disposition of the probe drug talinolol. Clin Pharmacol Ther 2002; 72: 572–583. Meissner K, Jedlitschky G, Meyer zu SH, Dazert P, Eckel L, Vogelgesang S et al. Modulation of multidrug resistance P-glycoprotein 1 (ABCB1) expression in human heart by hereditary polymorphisms. Pharmacogenetics 2004; 14: 381–385. Nakamura T, Sakaeda T, Horinouchi M, Tamura T, Aoyama N, Shirakawa T et al. Effect of the mutation (C3435T) at exon 26 of the MDR1 gene on expression level of MDR1 messenger ribonucleic acid in duodenal enterocytes of healthy Japanese subjects. Clin Pharmacol Ther 2002; 71: 297–303. Day DA, Tuite MF. Post-transcriptional gene regulatory mechanisms in eukaryotes: an overview. J Endocrinol 1998; 157: 361–371. Petriz J, Gottesman MM, Aran JM. An MDR-EGFP gene fusion allows for direct cellular localization, function and stability assessment of Pglycoprotein. Curr Drug Deliv 2004; 1: 43–56. Fellay J, Marzolini C, Meaden ER, Back DJ, Buclin T, Chave JP et al. Response to antiretroviral treatment in HIV-1-infected individuals with allelic variants of the multidrug resistance transporter 1: a pharmacogenetics study. Lancet 2002; 359: 30–36. Hitzl M, Drescher S, van der KH, Schaffeler E, Fischer J, Schwab M et al. The C3435T mutation in the human MDR1 gene is associated with altered efflux of the P-glycoprotein substrate rhodamine 123 from CD56+ natural killer cells. Pharmacogenetics 2001; 11: 293–298. Kim RB. MDR1 single nucleotide polymorphisms: multiplicity of haplotypes and functional consequences. Pharmacogenetics 2002; 12: 425–427. Kim RB, Leake BF, Choo EF, Dresser GK, Kubba SV, Schwarz UI et al. Identification of functionally variant MDR1 alleles among European Americans and African Americans. Clin Pharmacol Ther 2001; 70: 189–199. Ito S, Ieiri I, Tanabe M, Suzuki A, Higuchi S, Otsubo K. Polymorphism of the ABC transporter genes, MDR1, MRP1 and MRP2/cMOAT, in healthy Japanese subjects. Pharmacogenetics 2001; 11: 175–184. Itoda M, Saito Y, Soyama A, Saeki M, Murayama N, Ishida S et al. Polymorphisms in the ABCC2 (cMOAT/MRP2) gene found in 72 established cell lines derived from Japanese individuals: an association between single nucleotide polymorphisms in the 50 -untranslated region and exon 28. Drug Metab Dispos 2002; 30: 363–364. Popowski K, Sperker B, Kroemer HK, John U, Laule M, Stangl K et al. Functional significance of a hereditary adenine insertion variant in the 50 -UTR of the endothelin-1 gene. Pharmacogenetics 2003; 13: 445–451. Geier A, Mertens PR, Gerloff T, Dietrich CG, En-Nia A, Kullak-Ublick GA et al. Constitutive rat multidrug-resistance protein 2 gene transcription is down-regulated by Y-box protein 1. Biochem Biophys Res Commun 2003; 309: 612–618. del Moral RG, Olmo A, Aguilar M, O’Valle F. P glycoprotein: a new mechanism to control drug-induced nephrotoxicity. Exp Nephrol 1998; 6: 89–97.
