ABCB1) and risk of multiple myeloma

Available online at www.sciencedirect.com

Leukemia Research 33 (2009) 332–335

Brief communication

Polymorphisms and haplotypes in the multidrug resistance 1 gene (MDR1/ABCB1) and risk of multiple myeloma Krzysztof Jamroziak a,∗ , Ewa Balcerczak b , Karolina Calka b , Sylwester Piaskowski c , Halina Urbanska-Rys a , Aleksandra Salagacka b , Marek Mirowski b , Tadeusz Robak a b

a Department of Hematology, Medical University of Lodz, Ciolkowskiego 2, 93-513 Lodz, Poland Department of Pharmaceutical Biochemistry, Laboratory of Molecular Biology and Pharmacogenomics, Medical University of Lodz, Poland c Department of Molecular Pathology and Neuropathology, Medical University of Lodz, Poland

Received 6 May 2008; received in revised form 6 May 2008; accepted 9 June 2008 Available online 18 July 2008

Abstract MDR1(ABCB1) gene encodes for P-glycoprotein (P-gp, MDR1, ABCB1), an ATP-binding cassette superfamily member involved in the transport of xenobiotics. Here, we investigated whether common MDR1 single nucleotide polymorphisms (1236C>T, 2677G>A/T and 3435C>T) affect predisposition to multiple myeloma. Genotyping was performed in 111 myeloma patients and 96 controls by PCR-based assays. Haplotypes were inferred using PHASE algorithm. We found comparable allele and genotype frequencies among myeloma patients and controls. Moreover, patient and control groups did not differ regarding MDR1 haplotype distribution (p = 0.18). In conclusion, our results do not support major influence of MDR1 variants on the risk of myeloma in Caucasians. © 2008 Elsevier Ltd. All rights reserved. Keywords: P-glycoprotein; MDR1; ABCB1; SNP; Haplotype; Multiple myeloma; Genetic predisposition

1. Introduction The etiology of multiple myeloma (MM) is poorly defined. There is some evidence that environmental exposure to radiation, herbicides, insecticides, benzene, and other organic solvents may be involved [1]. On the other hand, racial differences in MM incidence as well as familial clusters of the disease underlie the role of genetic predisposition to MM [1]. P-glycoprotein (P-gp, MDR1, ABCB1) is an ATP-binding cassette superfamily transporter encoded by MDR1 (ABCB1) gene [2]. Characteristic tissue distribution and ability to export a wide spectrum of substrates indicate that P-gp is involved in the body defense against endogenous and exogenous toxins [2]. A number of single nucleotide polymorphisms (SNPs) in MDR1 were identified, of which the most attention is paid to the silent SNP 3435C>T SNP in exon 26 because of its association with altered P-gp expres∗

Corresponding author. Tel.: +48 426895191; fax: +48 426895192. E-mail address: [email protected] (K. Jamroziak).

0145-2126/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.leukres.2008.06.008

sion and function [3]. Two other common MDR1 SNPs, silent 1236>T in exon 12 and non-synonymous 2677TG>T/A (Ala893Ser/Thr) in exon 21, were shown to be in linkage disequilibrium with 3435C>T, and correlated with functional effects in some studies [2,3]. Given the putative role of interaction between genetic and environmental factors in the etiology of MM, we investigated whether common MDR1 SNPs (1236C>T, 2677G>T/A and 3435C>T) affect predisposition to MM in Caucasians.

2. Materials and methods 2.1. Patients and controls A total of 111 MM patients, 53 (47.7%) males and 58 (52.3%) females, with median age of 62 years, diagnosed in the Department of Hematology, Medical University of Lodz between 1992 and 2002 were included. The patients’ clinical characteristics at diagnosis are shown in Table 1. The control group consisted of 96 healthy blood donors, 39 (41%) males and 57 (59%) females, covering

K. Jamroziak et al. / Leukemia Research 33 (2009) 332–335 Table 1 Basic clinical characteristics at diagnosis of 111 multiple myeloma patients included to the study Parameter

n (%)

Age Median Range

62 years 40–87 years

Sex Males Females

53 (47.7) 58 (52.3)

Myeloma type IgG IgA IgD Light chain disease

57 (51.4) 27 (24.3) 1 (0.9) 26 (23.4)

Clinical stage I II III

333

using the following parameters: 30 s at 95 ◦ C, 30 s at 50 ◦ C and 1 min at 70 ◦ C, after that 3 ␮l of stop/loading buffer to each reaction were added. Denaturated seqPCR products were separated in polyacrylamide gel. 2.3. Haplotype analysis Haplotypes were statistically inferred from MDR1 position 1236, 2677 and 3435 genotype data using an algorithm based on Bayesian inference, PHASE, version 2.1.1. (http://www.stat.Washington.edu/stephens/) [5]. PHASE calls were made a total of 10 times, separately for patients and controls groups, and relative standard deviations of haplotype frequencies were consistently less than 2.5%. 2.4. Statistical analysis

8 (7.2) 12 (10.8) 91 (82.0)

comparable age range to MM patients. All study subjects were unrelated Polish Caucasians. The research was approved by the Bioethical Committee of Medical University of Lodz, and a written informed consent was obtained from every study participant. 2.2. Genotyping of MDR1 polymorphisms DNA was isolated from peripheral blood samples, obtained at diagnosis or during follow-up from MM patients and provided by the controls, using Blood Mini Kit (A&A Biotechnology, Gdansk, Poland) according to the manufacturer’s protocol. MDR1 3435C>T SNP was identified by polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP) method as reported previously [4]. Genotyping of 1236C>T and 2677G>T/A SNPs was performed by direct sequencing with the use of automated sequencer LI-COR® 4000 (Li-COR Biosciences, Lincoln, NE), and reagents and conditions as described below. The reaction mixture for PCR amplification included 200–500 ng of DNA template, 0.4 ␮M of each primer (for 1236C>T SNP: TCA GTT ACC CAT CTC GAA AAG AA and ACA TCA GAA AGA TGT GCA ATG TG; for 2677T>G/A: TAT GGT TGG CAA CTA ACA CT and CAT GAA AAA GAT TGC TTT GA), 1× Buffer PCR for AccuTaq LA DNA Polymerase, 0.05 U/␮l of JumpStart AccuTaq LA DNA Polymerase Mix, 500 ␮M dNTP mix, according to the protocol of AccuTaqTM LA DNA Polymerase Kit (Sigma Aldrich, Germany). PCR amplification consisted of three steps: denaturation at 94 ◦ C for 90 s, annealing at 54 ◦ C (for 2677G>T/A SNP) or 56 ◦ C (for 1236C>T SNP) for 60 s, and extension at 72 ◦ C for 90 s followed by 30 cycles. A sample without DNA template was included in each experiment as a negative control. After checking PCR product by electrophoresis in 2% agarose gel, DNA were amplified in sequencing PCR by SequiTherm EXCEL TM II DNA Sequencing Kit-LC for 25–41 cm gels (Epicentre Technologies, Madison, WI). At first, the following components were combined: 1–2 pmol of IRD800 labeled primer (1236seq TGT TTT CTT GTA GAG ATT ATA A, 2677seq TAT GGT TGG CAA CTA ACA CT), 3.5X SequiTherm EXCEL II Sequencing buffer, 5 U/␮l SequiTherm EXCEL II DNA Polymerase, DNA template and deionized water. This mixture was divided into four tubes in which proper SequiTherm EXCEL IILC Termination Mix was added. Then 30 cycles were performed

The observed genotype frequencies were compared to expected according to Hardy–Weinberg rule. Statistical significance of differences in the allele frequencies and distributions between MM patients and controls was assessed using χ2 -test. Disease association was described using odds ratio (OR) with 95% confidence interval (95% CI) estimated by logistic regression. Difference in haplotype distribution among patients and controls was assessed by case-control permutation test using PHASE, -c option [5]. A number of permutations was set to 1000. For all computations p < 0.05 was considered significant.

3. Results Genotyping of MDR1 polymorphisms at positions 1236C>T, 2677G>T/A and 3435C>T was successful in all included 111 MM patients and 96 healthy controls. A detailed comparison of the observed allele and genotype frequencies is provided in Table 2. Polymorphisms were distributed in accordance with Hardy-Weinberg equilibrium within patient and control cohorts except MDR1 1236C>T in MM patients (Table 2). Overall, the prevalence of all three analyzed MDR1 SNPs was comparable between patients and controls. The frequencies of variant alleles were as follows: for MDR1 1236T-allele 0.46 in patients and 0.43 in controls (p = 0.45), for 2677Tallele 0.45 in MM patients and 0.40 in controls (p = 0.47), for 2677A-allele 0.014 in MM patients and 0.0 in controls (p = 0.47), and for 3435T 0.46 in patients and 0.47 in controls (p = 0.92). The logistic regression analysis did not show significant impact of variant alleles and genotypes on likelihood of MM development (Table 2). In agreement with previous studies we confirmed linkage disequilibrium between tested SNPs (p < 0.01 for each pair of 1236C>T, 2677G>T/A and 3435C>T SNPs in patient and control groups). Therefore, we used PHASE software to verify the influence of MDR1 haplotypes on MM [5]. Estimated frequencies of the most likely haplotypes inferred by PHASE calls separately for MM patient and control cohorts are listed in Table 3. Wild-type 1236C–2677G–3435C haplotype dominated in both MM patients and controls, and

334

K. Jamroziak et al. / Leukemia Research 33 (2009) 332–335

Table 2 Comparison of allele frequency and distribution of MDR1 gene SNPs at positions 1236C>T, 2677G>T/A and 3435C>T among patients with multiple myeloma and normal healthy individuals of Caucasian origin MDR1 variants

Myeloma, n = 111

1236CC 1236CT 1236TT 1236C 1236T HWE: p

95% CI

p

26 (23.4%) 67 (60.4%) 18 (16.2%)

Controls, n = 96 28 (29.1%) 54 (56.3%) 14 (14.6%)

1.00 1.34 1.38

– 0.70–2.54 0.57–3.33

0.61 0.32

119 (53.6%) 103 (46.4%)

110 (57.3%) 82 (42.7%)

1.00 1.16

– 0.79–1.71

0.45

–

–

–

0.02

2677GG 2677GT 2677GA 2677TT 2677G 2677T 2677A HWE: p

0.14

32 (28.8%) 52 (46.8%) 3 (2.7%) 24 (21.7%)

32 (33.3%) 51 (53.2%) 0 (0%) 13 (13.5%)

1.00 1.02 492.6 1.85

– 0.55–1.90 0.0–4 × 1014 0.80–4.25

0.43 0.88 0.69

119 (53.6%) 100 (45.0%) 3 (1.4%)

115 (59.9%) 77 (40.1%) 0 (0%)

1.00 1.25 476.0

– 0.85–1.86 0.0–4 × 1013

0.47 0.80

0.74

3435CC 3435CT 3435TT 3435C 3435T HWE: p

–

–

–

27 (24.3%) 65 (58.6%) 19 (17.1%)

27 (28.1%) 48 (50%) 21 (21.9%)

0.30

1.00 1.35 0.90

– 0.71–2.60 0.40–2.05

0.81 0.25

119 (53.6%) 103 (46.4%)

102 (53.1%) 90 (46.9%)

1.00 0.98

– 0.67–1.44

0.92

–

–

–

0.06

0.97

constituted 19% of all haplotype variants (Table 3). PHASE case-control permutation testing did not show significant difference in haplotype distribution between MM patients and controls (p = 0.18).

4. Discussion It is generally accepted that inherited variation in the transport and metabolism of environmental toxins may determine the likelihood of malignant transformation. Because of wide spectrum of drugs and carcinogens being P-gp substrates, a significant literature has been published on the relation of MDR1 SNPs and haplotypes with pharmacokinetics, outcomes of pharmacotherapy and predisposition Table 3 The most likely haplotype frequencies in investigated populations of myeloma patients and healthy controls inferred using PHASE software Position

Odds ratio

Multiple myeloma, n = 111

Controls, n = 96

1236

2677

3435

Frequency

S.E.

Frequency

S.E.

C C C T T C T T

G G T T G T G T

C T T T C C T C

0.189 0.119 0.122 0.112 0.083 0.117 0.112 0.083

0.015 0.013 0.013 0.014 0.011 0.013 0.014 0.012

0.194 0.124 0.157 0.073 0.071 0.104 0.073 0.071

0.017 0.021 0.024 0.014 0.016 0.017 0.014 0.014

Haplotypes with frequency lower than 1% were not shown.

to diseases (reviewed in [2]). We and others showed that MDR1 3435C>T SNP associates with susceptibility to cancer including pediatric acute lymphoblastic leukemia and renal epithelial tumors [4,6]. In this study, we found that allele frequencies and distributions of three MDR1 SNPs are comparable in MM patients and healthy controls (Table 2). The observed frequencies of variant alleles in control group are in agreement with previous reports concerning healthy Caucasian populations [2]. Furthermore, the prevalence of 2677G>T\A and 3435C>T SNPs in our patient group is similar to that described by Maggini et al. in their recent pharmacogenetic study in MM [7]. The variant allele frequencies were 0.45 and 0.39 for 2667T, 0.014 and 0.031 for 2677A, and 0.46 and 0.47 for 3435T, in our study and as reported by Maggini et al. [7], respectively. Since some reports attributed functional effects to MDR1 haplotypes involving 1236–2677–3435 positions, we performed haplotype analysis using PHASE software [5]. Based on the result of permutation testing, we could not reject the null hypothesis that the MM cases and controls are random draws from a common set of population haplotype frequencies. Therefore, our data do not suggest major role of MDR1 variants in predisposition to MM. There are several factors related to the design of such case-control study that could influence our results. First, the putative effect of MDR1 SNPs on P-gp activity described in some previous reports is modest, thus the study could be underpowered to detect a real difference [2]. Moreover, the tested cohorts were relatively small. However, this last effect is partially reduced by high frequency of variant MDR1 alle-

K. Jamroziak et al. / Leukemia Research 33 (2009) 332–335

les in Caucasian population (between 0.4 and 0.5 for 1236T, 2677T and 3435T alleles). The only rare 2677A allele is unlikely to confer predisposition to MM because its frequency is significantly higher in Asian populations that have lower incidence of MM than Caucasians [2]. Furthermore, it is clear that MM patients differ in type and quantity of past carcinogen exposure. Importantly, a recent ex vivo study showed that silent polymorphisms in MDR1 (in particular 3435C>T) can alter P-gp conformation, and thus the polymorphisms’ effect may be substrate specific [8]. This raises the possibility that functional effect of MDR1 SNPs may be relevant only among individuals with specific carcinogen exposure. Interestingly, recent report showed that survival of MM patients uniformly treated with regiments containing Pgp substrate drugs is dependent on 2677G>T and 3435C>T SNPs and haplotypes [7]. Therefore, larger genotyping studies that include history of carcinogenic exposure may be needed to detect possible effect of MDR1 SNPs in myeloma.

5. Conclusions In conclusion, the results obtained in this study do not provide support for the hypothesis that common inherited variants in MDR1 gene have major influence on the likelihood of MM development in Caucasians.

Acknowledgements This work was supported in part by a grant from the Polish Ministry of Science (no. 2PO5B14528) and a grant from the Medical University of Lodz (no. 502-11-680).

335

Contributions. K-J designed the study and wrote the report, E.B. performed the research and wrote the report, K.C. performed the research, S.P. performed the research, H.U-R, performed the research, A.S. performed the research, M.M performed the research, T.R. wrote the report.

References [1] Kyle RA, Rajkumar SV. Epidemiology of the plasma-cell disorders. Best Pract Res Clin Haematol 2007;4:637–64. [2] Sakaeda T. MDR1 genotype-related pharmacokinetics: fact or fiction? Drug Metab Pharmacokinet 2005;6:391–414. [3] 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 Pglycoprotein expression and activity in vivo. Proc Natl Acad Sci USA 2000;97:3473–8. [4] Jamroziak K, Mlynarski W, Balcerczak E, Mistygacz M, Trelinska J, Mirowski M, et al. Functional C3435T polymorphism of MDR1 gene: an impact on genetic susceptibility and clinical outcome of childhood acute lymphoblastic leukemia. Eur J Haematol 2004;72:314– 21. [5] Stephens M, Smith NJ, Donnelly P. A new statistical method for haplotype reconstruction from population data. Am J Hum Genet 2001;68:978–89. [6] Siegsmund M, Brinkmann U, Scháffeler 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;13:1847–54. [7] Maggini V, Buda G, Martino A, Presciuttini S, Galimberti S, Orciuolo E, et al. MDR1 diplotypes as prognostic markers in multiple myeloma. Pharmacogenet Genomics 2008;18:383–899. [8] Kimchi-Sarfaty C, Oh JM, Kim IW, Sauna ZE, Calcagno AM, Ambudkar SV, et al. A “silent” polymorphism in the MDR1 gene changes substrate specificity. Science 2007;315:525–8.