Haplotype Analysis of MDR1 and Risk for Cervical Cancer in Northeastern Thailand

Objective: The aim of this study was to investigate the association between genotype and haplotype of MDR1 (C1236T, G2677T/A and C3435T) and the risk for cervical cancer in Northeastern Thai women.
Methods: An age-matched case-control study involving squamous cell cervical cancer (SCCA) patients (n=204) and healthy controls (n=204) was enrolled for MDR1 genotyping by real-time PCR method.
Results: The genotype distribution of MDR1 in both patients and controls was not significantly different (p>0.05). The haplotype analysis showed that T-T-T was the most common haplotype in this population. Significantly increased risk of cervical cancer was observed in carriers of T-T-C and C-G-T haplotypes with ORs of 1.86 (95%CI=1.02-3.39, p=0.0416) and 2.00 (95%CI=1.18-3.40, p=0.0140), respectively. Analysis of 2677-3435 haplotype showed increased risk for cervical cancer in G-T (OR=1.55; 95% CI=1.12-2.13, p=0.0432) and T-C (OR=1.91; 95%CI=1.05-3.47, p=0.0325).
Conclusion: The results provide evidence that haplotype of MDR1 may be an important risk factor for cervical cancer development in Northeastern Thai women. Creative Commons Attribution License

Introduction

The multidrug resistance 1 (MDR1) gene located on chromosomal region 7q21 encodes P-glycoprotein (P-gp). P-gp is a membrane transporter of an ATP-dependent efflux pump which contributes to protecting the body from environmental toxins and carcinogenic compounds. It is widely expressed on the membrane of normal tissues, such as breast, colon, gastric and cervical tissues (Marzolini et al., 2004; Nicol et al., 2014; Riou et al., 1990).

The MDR1 is a highly polymorphic gene with at least 50 single nucleotide polymorphisms (SNPs) (Fung and Gottesman, 2009). The most common three SNPs are C1236T, G2677T/A and C3435T in the exon 12, 21 and 26, respectively (Kimchi-Sarfaty et al., 2007(a); Panczyk et al., 2009). C1236T and C3435T are silent mutations that do not change the encoded amino acid, whereas non-synonymous G2677T/A results in an amino acid substitution from alanine to serine or threonine. Polymorphisms of the MDR1 may lead to inter-individual variation of P-gp expression and function, and may influence inter-individual carcinogenic susceptibility (Wang et al., 2005).

Among the identified MDR1 SNPs, C3435T was the first SNP correlated with expression levels and function of P-gp (Hitzl et al., 2001; Hoffmeyer et al., 2000). Several studies have reported that genetic variations of the MDR1 alter the risk of cancer development which is more apparent in haplotype analysis with other two, C1236T and G2677T/A, SNPs (Abuhaliema et al., 2016; Panczyk et al., 2009; Potocnik et al., 2008; Wu et al., 2012). Carriers of the T allele of C3435T were mostly found with the increased risk for cancer in colon, renal and breast (Kurzawski et al., 2005; Siegsmund et al., 2002; Turgut et al., 2007), while some studies reported that carriers of the C allele had an increased risk of colorectal cancer and breast cancer (Abuhaliema et al., 2016; Yue et al., 2013). However, no significant relationships between the polymorphism and increased colorectal cancer (Petrova et al., 2008), breast cancer (Rubis et al., 2012) and gastric cancer (Zebrowska et al., 2014) have also reported.

Since the C3435T polymorphism is in linkage disequilibrium with C1236T and G2677T/A polymorphisms (Kimchi-Sarfaty et al., 2007(b)), haplotype analyses for the three polymorphisms may provide more useful information than simple genotypic analyses. Therefore, our study was designed to determine whether the MDR1 polymorphisms and their haplotype are associated with the risk for developing cervical cancer among Northeastern Thai women.

Materials and Methods

Study population

This case-control study was conducted in Khon Kaen Hospital and Srinagarind Hospital, Khon Kaen Province, Northeastern Thailand. Participants were 204 women with pathologically diagnosed squamous cell carcinoma of the cervix (SCCA) and 204 healthy women without cervical cancer and/or pre-invasive lesion of the cervix confirmed by cytological and histological examination. The controls were matched to the cases on age within 5 year interval. The study protocol was approved by the Ethics Committee of Khon Kaen University (HE 561382) and a signed informed consent was obtained from each participant.

Detection of MDR1 polymorphism

Genomic DNA was extracted from buffy coat using GF-1 Blood DNA Extraction Kits (Vivantis, USA). The real-time polymerase chain reaction with TaqMan® probe (Applied Biosystems, USA) was used for the detection of MDR1 polymorphisms. Taqman® probe was designed for specific SNP, VIC fluorescence dye’s probe was used for wild type allele (C allele) whereas FAM fluorescence dye’s probe was used for mutant type allele (T allele) of C1236T and C3435T. For G2677T/A SNP which provided two types of variant allele (T and A alleles) performed in two assays, one for the detection of G and A alleles and another for the detection of G and T alleles, VIC probe was specific with the G allele and FAM probe was specific with the A allele or T allele. The reactions were set under the following conditions: for C1236T (rs1128503) and C3435T (rs1045642) SNPs, holding stage at 95 °C for 10 minutes, follow by 40 cycles each of denaturation at 95 °C for 15 seconds, annealing and extension at 60 °C for 60 seconds. For G2677T/A (rs2032582) SNP, holding stage at 95 °C for 10 minutes, follow by 50 cycles each of denaturation at 95 °C for 15 seconds, annealing and extension at 60 °C for 60 seconds.

Statistical analyses

Statistical analyses were performed using STATA software. Deviation from Hardy-Weinberg equilibrium (HWE) for the genotype of each SNP was tested by Pearson’s chi-square. The haplotypes were inferred using PHASE algorithm version 2.1.1. Association between genotype or haplotype and cervical cancer risk was tested by calculating the odd ratios (OR) with 95% confidence intervals (CI) by logistic regression analyses. P values less than 0.05 were considered statistically significant.

Results

The genotype distribution and allele frequency of the MDR1 polymorphisms (C1236T, G2677T/A and C3435T) are summarized in Table 1. The genotype distribution for each locus among the cases and controls was in agreement with Hardy-Weinberg equilibrium (p > 0.05). The genotype distribution of C1236T, G2677T/A and C3435T in both patients and controls were not different and no association was found between these genotype and cervical cancer risk. For the allele distribution, only the C3435T polymorphism showed significantly different distribution between cases and controls (p=0.042), which indicated that the presence of T allele gave a higher risk for cervical cancer.

Table 1

Genotype Distribution and Allele Frequency of The MDR1 in Cervical Cancer

Position	Exon	Amino acid change	Genotype/Allele	Number/Frequency
				Case[a]	Control[b]
			CC	24	23
1236	12	Gly412Gly	CT	96	97
			TT	84	84
			C	0.35	0.35
			T	0.65	0.65
2677	21	Ala893Ser/Thr	GG	55	59
			GT	84	92
			GA	10	6
			TT	47	32
			TA	7	13
			AA	1	2
			G	0.50	0.53
			T	0.45	0.41
			A	0.05	0.06
3435 [c]	26	Ile1145Ile	CC	47	55
			CT	84	97
			TT	73	52
			C	0.44	0.51
			T	0.56	0.49

, no deviation from Hardy–Weinberg equilibrium;

, no deviation from Hardy–Weinberg equilibrium;

, allele distribution of Ile1145Ile (C/T) was different between case and control (p=0.042)

Table 2 presents the distribution of haplotype for the three loci of MDR1 (1236-2677-3435). The haplotype T-T-T was most frequently observed both in cases and controls. T-T-C and C-G-T haplotypes significantly increased the risk for cervical cancer compared to T-T-T with OR=1.86 (95%CI=1.02-3.39, p=0.042) and OR=2.00 (95%CI=1.18-3.40, p=0.014), respectively. Furthermore, carriers of C-G-T had 2.5 times higher risk of cervical cancer compared to the Caucasian and African common C-G-C haplotype (95%CI=1.42-4.45, p=0.021).

Table 2

Association between MDR1 Haplotypes (1236-2677-3435) and Risk for Cervical Cancer

MDR1 1236-2677-3435	Case n (%)	Control n (%)	OR [95%CI, p] T-T-T as standard	OR [95%CI, p] in haplotype pair
T-T-T	148 (36.27)	145 (35.54)	1
T-T-C	36 (8.82)	19 (4.66)	1.86 [1.02-3.39, 0.042*]
C-G-C	73 (17.89)	90 (22.06)	0.80 [0.54-1.17, 0.241]	1
C-G-T	51 (12.50)	25 (6.13)	2.00 [1.18-3.40, 0.014*]	2.52 [1.42-4.45, 0.021*]
C-A-C	9 (2.21)	17 (4.17)	0.52 [0.22-1.20, 0.120]	1
C-A-T	10 (2.45)	6 (1.47)	1.63 [0.58-4.61, 0.350]	3.15 [0.86-11.49, 0.113]
T-G-C	59 (14.46)	81 (19.85)	0.71 [0.48 – 1.16, 0.103]	1
T-G-T	21 (5.15)	20 (4.90)	1.03 [0.54 -1.98, 0.932]	1.44 [0.72-2.90, 0.303]
C-T-C	1 (0.26)	0 (0.00)	b	b
C-T-T	0 (0.00)	5 (91.26)	b	b

p<0.05; b, drop because confidence levels not possible with zero count cells

Discussion

The MDR1 polymorphisms have been intensively studied so far, mostly focusing on the role of C3435T (Fung and Gottesman, 2009; Wang et al., 2005). C3435T correlated with other SNPs i.e. G2677T/A and C1236T that have been shown to be a part of common haplotypes, is the main determinant of functional differences in P-gp (Kimchi-Sarfaty et al., 2007(b); Panczyk et al., 2009). Analysis of MDR1 haplotype likely provide a better understanding of MDR1 polymorphism on cancer risk rather than their allelic frequency distribution for each individual SNP.

In our result, the genotype distributions for the three MDR1 SNPs in the controls were in Hardy-Weinberg equilibrium which confirmed they represented the given population. Among healthy Thais, the commonest allele for each locus was T (C1236T), G (G2677T/A) and C (C3435T) as in other Asian populations but was different from that in European-Americans and Caucasians (Ameyaw et al., 2001; Cascorbi et al., 2001, Kim et al., 2001; Komoto et al., 2006; Li et al., 2007; Tang et al., 2002; Wang et al., 2015), showing inter-ethnic differences.

As for C3435T, risk for several cancers, such as breast, endometrial and colorectal have been documented (Cizmarikova et al., 2010; Mrozikiewicz et al., 2007; Khedri et al., 2011; Wu et al., 2012), whereas in this study, T allele itself showing significantly higher risk for cervical cancer was observed nevertheless genotype was not involved in.

Because of the linkage disequilibrium, the haplotype frequencies were not of the representative of allelic frequencies; T-T-T haplotype was the most common haplotype in our population. It is documented that this haplotype is a predominant haplotype among Asian populations (Fung and Gottesman, 2009; Xu et al., 2008), whereas C-G-C haplotype is a predominant haplotype in Caucasian and African populations (Fung and Gottesman, 2009; Gumus-Akay et al., 2010; Jeannesson et al., 2007); here observed is inter-ethnic differentce, too.

Participation of MDR1 polymorphisms in the alteration of risk for cervical cancer might be due to the change of P-gp function which is likely influenced by the 2677-3435 haplotype (Kwon et al., 2009, Ni et al., 2015, Pang et al., 2014). A study of 2677-3435 haplotype demonstrated that G-T haplotype contributed to diffuse large B cell lymphoma development (Ni et al., 2015) which was in agreement with our findings; significant increased risk for cervical cancer was observed in carriers of G-T haplotype (C-G-T and T-G-T) with OR 1.546 (95%CI=1.12-2.13, p=0.043) compared to T-T haplotype. A significant increased risk for cervical cancer was also observed in T-C haplotype (OR=1.908 95% CI=1.05-3.47, p=0.033); however, A-T haplotype did not show an increased risk. It is suggested that the presence of T allele in these two loci may alter susceptibility to cervical cancer. The presence of variant alleles has been associated with lower mRNA expression and less production of P-gp which results in carcinogen accumulation (Lamba et al., 2006, Johne et al., 2002). In addition, a possible explanation for different P-gp function observed within the MDR1 haplotype may result from the interaction of C1236T, G2677T/A and C3435T leading to the presence of rare codon. Alteration of codon influences the co-translational folding time and insertion of P-gp into the membrane, thereby change of P-gp structure and substrate specificity (Kimchi-Sarfaty et al., 2007(b)).

In conclusion, our finding demonstrates the association between haplotype of MDR1 and cervical cancer risk in Northeastern Thai women. The haplotype testing to identify genetic susceptibility may provide more information for the cancer prevention.