MTHFR and MTRR genotype and haplotype analysis and colorectal cancer susceptibility in a case-control study from the Czech Republic

(1)

Other uses, including reproduction and distribution, or selling or

licensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of the

article (e.g. in Word or Tex form) to their personal website or

institutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies are

encouraged to visit:

(2)

Contents lists available at

ScienceDirect

Mutation Research/Genetic Toxicology and

Environmental Mutagenesis

j o u r n a l h o m e p a g e :

w w w . e l s e v i e r . c o m / l o c a t e / g e n t o x

C o m m u n i t y a d d r e s s :

w w w . e l s e v i e r . c o m / l o c a t e / m u t r e s

MTHFR and MTRR genotype and haplotype analysis and colorectal cancer

susceptibility in a case–control study from the Czech Republic

Barbara Pardini

a

_{, Rajiv Kumar}

b

_{, Alessio Naccarati}

a

_{, Rashmi B. Prasad}

b

_{, Asta Forsti}

b

_,

Veronika Polakova

a

_{, Ludmila Vodickova}

a

,

c

_{, Jan Novotny}

d

_,

Kari Hemminki

b

_{, Pavel Vodicka}

a

,

∗

a_{Institute of Experimental Medicine, Academy of Sciences of the Czech Republic, Videnska 1083, 14200 Prague 4, Czech Republic} b_{German Cancer Research Center (DKFZ), Heidelberg, Germany}

c_{National Institute of Public Health, Prague, Czech Republic} d_{1st Medical Faculty, Charles University, Prague, Czech Republic}

a r t i c l e i n f o

Article history: Received 22 July 2010

Received in revised form 25 October 2010 Accepted 18 December 2010

Available online 4 January 2011 Keywords:

Colorectal cancer risk MTHFR

MTRR Genotype Haplotype analyses

a b s t r a c t

Polymorphic variants in genes involved in one-carbon metabolism, in particular of dietary folate, may modulate the risk for colorectal cancer through aberrant DNA-methylation and altered nucleotide syn-thesis and repair. In the present study, we have assessed the association of six polymorphisms and relative haplotypes in the MTHFR gene (rs1801133 and rs1801131) and in the MTRR gene (rs1801394, rs1532268, rs162036, and rs10380) with the risk for colorectal cancer in 666 patients and 1377 controls from the Czech Republic.

We found that the 677 C > T polymorphism in the MTHFR gene signiﬁcantly decreased the risk for colorectal cancer in homozygous carriers of the variant allele (OR, 0.58; 95% CI, 0.39–0.87). Also, we noted a signiﬁcantly different distribution of genotypes between cases and controls for the 66A > G polymorphism in the MTRR gene. In particular, homozygous carriers of the G-containing allele of this polymorphism were at an increased risk for colorectal cancer (OR, 1.39; 95% CI, 1.04–1.85).

Haplotype analysis of the two MTHFR polymorphisms showed a moderate difference in the distribution of the TA haplotype between cases and controls. In comparison to the most common haplotype (CA), the TA haplotype was associated with a decreased risk for colorectal cancer (OR, 0.84; 95% CI, 0.71–0.99). No difference in the distribution between cases and controls was observed for the haplotypes based on the four polymorphisms in the MTRR gene.

The present study suggests that the 677TT genotype and the TA haplotype in the MTHFR gene may also have a role in colorectal cancer risk in the Czech population, indicating the importance of genes involved in folate metabolism with respect to cancer risk. For MTRR, additional studies on larger populations are needed to clarify the possible role of variation in this gene in colorectal carcinogenesis.

1. Introduction

Colorectal cancer is one of the most common cancers in the

world and the third leading cause of cancer death. The incidence of

colorectal cancer varies substantially worldwide, with high rates in

Western countries

[1]

. Over the past decade, the role of folate and

genetic polymorphisms of enzymes involved in its metabolism has

attracted considerable interest in epidemiological research on this

cancer type

[2]

. In particular, imbalanced DNA-methylation,

char-acterized by genomic hypomethylation and methylation of usually

unmethylated CpG sites, has been consistently observed in

colorec-tal cancer

[3]

.

∗ Corresponding author. Tel.: +420 2 41062694; fax: +420 2 41062782. E-mail address:[email protected](P. Vodicka).

Methylene-tetrahydrofolate reductase (MTHFR) is a key

enzyme regulating folate metabolism, and it is thought to

inﬂuence DNA methylation and synthesis

[4,5]

. MTHFR

irre-versibly converts 5,10-methylenetetrahydrofolate (5,10-MTHF) to

5-methyltetrahydrofolate (5-MTHF), which provides the methyl

group that converts homo-cysteine to methionine, the precursor of

S-adenosylmethionine (SAM). SAM is the universal methyl-group

donor for methylation of a wide variety of biological substrates. It

has been hypothesized that folate/methyl depletion may not only

result in a global genomic hypomethylation, but also in aberrant

methylation of CpG clusters in the promoters of tumor-suppressor

and DNA-repair genes, probably via upregulation of DNA

methyl-transferase

[6]

. The substrate of MTHFR, 5,10-MTHF, is required

for conversion of deoxyuridylate to thymidylate. Depletion of the

thymidylate pool leads to uracil misincorporation into DNA, and

subsequently to single- and double-strand breaks

[6]

. Two

(3)

mon functional polymorphisms in the MTHFR gene are C677T

(rs1801133), which results in an alanine-to-valine substitution at

codon 222, and A1298C (rs1801131), which results in a

glutamate-to-alanine substitution at codon 429

[7,8]

. These most common

polymorphisms in the MTHFR gene have frequently been

inves-tigated for association with adenoma and colorectal cancer risk

[9,10]

. As reviewed recently, the 677TT genotype has been

associ-ated with a reduced risk for colorectal cancer

[11]

. However, results

are still inconclusive for the A1298C polymorphism, probably due

to the limited amount of data available

[6]

.

The role of polymorphisms in other genes in the folate metabolic

pathway has not yet been fully evaluated for association with the

risk for colorectal cancer. Methionine-synthase reductase (MTRR),

in particular, plays a crucial role in maintaining the active state

of methionine synthase (MTR), through reductive methylation of

cob(II)alamin, a cofactor of MTR

[12,13]

. Functionally, MTRR

main-tains adequate levels of methylcob(III)alamin. The active form of

MTR is essential for the regulation of appropriate levels of

methio-nine, which acts as a precursor for the universal methyl-donor SAM.

Mutations, including deletions and insertions in the MTRR gene

are associated with a rare autosomal recessive disorder, cblE-type

homocystinuria, leading to megaloblastic anemia and

developmen-tal delay in early childhood

[14]

. The most common polymorphism

in this gene is an isoleucine-to-methionine change at position 22

(A66G; rs1801394). Although the A66G polymorphism does not

appear to alter the catalytic activity of the protein, the 66GG

geno-type has been associated with a modest but signiﬁcant decrease in

plasma homocysteine levels

[12]

. Only a few studies have

investi-gated the role of MTRR variants in modulation of cancer risk, with

ambiguous outcome. The A66G polymorphism has been previously

associated with an increased risk for lung and esophageal cancer

and a decreased risk for squamous cell carcinoma of head and neck

and leukemia

[15–18]

. Kwak et al.

[19]

reported that carriers of the

G allele for this polymorphism had an increased risk to develop

hepatocellular carcinoma, while, more generally, the MTRR gene

(and in particular C1783T polymorphism) has been indicated as

a novel pancreatic cancer susceptibility factor

[20]

. On the other

hand, no association with prostate, breast or cervical cancer has

been reported

[21–24]

.

In case of colorectal cancer, the C25088T and T14208A

polymor-phisms in the gene were associated with increased adenoma and

colorectal cancer risk

[9,25–27]

. The data for the most common

A66G polymorphism in the MTRR gene, and its effect on the risk for

colorectal cancer have remained inconsistent

[9,25–30]

.

We have carried out a case–control association study to

evalu-ate the role of common polymorphisms and haplotypes within the

MTHFR and the MTRR genes in the risk for sporadic colorectal

can-cer. The study was conducted with a hospital-based case–control

population from the Czech Republic, where the reported incidence

of colon cancer is the third highest in the world and the highest

for rectal cancer worldwide

[31]

. To our knowledge, this is one of

the largest studies investigating the role of MTRR haplotypes in the

susceptibility to colorectal cancer.

2. Materials and methods

2.1. Study population

The study population comprised 666 patients with colorectal cancer and 1377 hospital-based healthy controls. Eligibility criteria for participation in the study included cases and controls who were of Czech origin, and consented to provide biological samples for genetic analysis. Patients with histologically conﬁrmed diag-nosis of colorectal cancer were recruited between September 2004 and February 2006 in nine different oncology departments in the Czech Republic (two in Prague, the others in the cities of Benesov, Brno, Liberec, Ples, Pribram, Usti nad Labem, and Zlin), as representative of the entire country. During the study period, a total of 968 patients with colorectal cancer provided blood samples from the above-mentioned hospitals. Sixteen individuals were initially excluded because they fulﬁlled the

Ams-terdam criteria I and II for hereditary colorectal cancer[32,33]. Another 286 cases were excluded because they did not meet eligibility criteria (age, ethnic origin, avail-ability of samples) or incomplete clinical information was available. The mean age at diagnosis of the colorectal cancer cases was 59.5 years (range 31–84 years).

Two control groups were included in the study. The first group was selected among individuals admitted to five large gastroenterological departments (Prague, Brno, Jihlava, Liberec, and Pribram) in the Czech Republic, at the same time period as the recruitment of cases took place. This group was undergoing colonoscopy for various gastrointestinal complaints (44.9%). The reasons for colonoscopical investi-gation were (i) macroscopic bleeding; (ii) positive fecal occult blood test (FOBT); (iii) abdominal pain of unknown origin. Due to the high incidence of colorectal cancer in the Czech Republic, colonoscopy is widely recommended and practiced. Subjects with negative colonoscopy results for malignancy or idiopathic bowel diseases were included in the control group. To reduce selection bias, only those subjects with no previous diagnosis of any chronic disease were included into the study. This crite-rion was used to avoid inclusion of individuals with chronic diseases who might have been repeatedly admitted to the hospital and modified their habits because of their disease. Among 739 recruited controls, a total of 610 (82.5%) were included in the study. The sex distribution among the controls excluded was similar to those included. The second group of controls consisted of 767 healthy individuals recruited by a blood-donor center in one hospital in Prague. They were cancer-free at the time of the sampling. The choice of two different control populations was done for two main reasons. Inclusion of ‘colonoscopically negative’ individuals as controls ensures disease-free control individuals because a negative colonoscopy result is the best available proof of the absence of colorectal cancer[34]. On the other hand, since this group of individuals may not necessarily represent the general population, we included also healthy, cancer-free individuals recruited among volunteers from blood centers.

The participating subjects were properly informed and provided written con-sent and approval of genetic analysis in agreement with the Helsinki declaration. The design of the study was approved by the Ethical Committee of the Institute of Experimental Medicine, Prague, Czech Republic.

2.2. Interviews

Cases and controls were personally interviewed by trained personnel, with the use of a structured questionnaire that was the same for all individuals in the study, to determine demographic characteristics and potential risk factors for colorectal cancer. Study subjects provided information on their lifestyle habits, body-mass index (BMI), diabetes, and family/personal history of cancer. Questions on lifelong or long-term (at least six consecutive months) drug use were also included in the questionnaire.

2.3. Genotyping

DNA was isolated from coded blood samples and stored at−80◦_{C. SNPs in}

the MTHFR gene (C677T (rs1801133) and A1298C (rs1801131)) were genotyped by means of a PCR-RFLP assay (PCR-RFLP assay reaction-conditions and primer sequences are available upon request).

SNPs in the MTRR gene (A66G (I22M), rs1801394; C524T (S175L), rs1532268; A1049G (K350R), rs162036; C1793T (H595Y), rs10380) were genotyped with the TaqMan allelic discrimination assays (Applied Biosystems, Foster City, CA; Assay-on-demand, SNP genotyping products: C 3068176 10, C 3068164 10, C 3068152 10, C 7580070 1, respectively) as described by[18].

The selection of SNPs was based on their reported functional effects or asso-ciation with cancer, available in the dbSNP database of the National Center for Biotechnology Information (NCBI;www.ncbi.nlm.nih.gov).

The genotype screening for cases and controls was carried out simultaneously in a blinded manner. The results were conﬁrmed by random re-genotyping of more than 10% of the samples for each polymorphism analysed. The concordance between repeated samples was 100%.

2.4. Statistical analysis

Genotype distribution for each polymorphism was tested in controls for Hardy–Weinberg equilibrium, and differences in expected and observed frequencies were tested for statistical signiﬁcance by the Pearson chi-squared test. Differences in baseline socio-demographic characteristics between cases and controls were anal-ysed by use of the chi-squared test and Student’s t-test.

Odds ratios (ORs), adjusted for gender and age, and the corresponding 95% con-ﬁdence intervals (95% CIs) for assessment of the association of colorectal cancer with different genotypes were calculated with SAS version 9.1 (SAS Institute, Cary, NC, USA). The haplotype procedure of SAS/Genetics Software was used to esti-mate haplotype frequencies in the cases and the controls separately, and to infer the possible haplotype combinations for each individual. Relationships between genotypes/haplotypes and colorectal cancer risk were summarized as global P-values, corrected for multiple-testing by the Westfall and Young permutation method. The association between the inferred genotype combinations and sus-ceptibility to colorectal cancer was explored using a forward stepwise approach: the analyses started by considering a single SNP, and likelihood-ratio tests were

(4)

used to assess whether the consideration of further genetic markers improved the ﬁt of the model[35]. Linkage disequilibrium was calculated with Haploview software (www.broad.mit.edu/mpg/haploview/documentation.php). Power calcu-lations were carried out with the power and sample size calculation software version 2.1.31.

LD was calculated with Haploview software (www.broad.-mit.edu/mpg/ haploview/documentation.php).

3. Results

3.1. Study population characteristics

The study included 666 CRC cases and 1377 controls (610

colonoscopy-negative controls and 767 blood center controls)

(

Table 1

). The distribution of the covariates considered did not differ

signiﬁcantly between patients and controls, even in the two groups

of controls.

3.2. Genotype and haplotype analysis

The distribution of genotypes within the selected genes in

the controls was in agreement with Hardy–Weinberg equilibrium

(

Table 2

). The analysis of linkage disequilibrium (LD) for the MTHFR

polymorphisms reported a

|D

_{| = −0.94 and r}

2

_{= 0.22. For the four}

loci in the MTRR gene, no strong LD was observed. Results are shown

in

Supplementary Fig. 1

.

Genotype and haplotype analyses were done with the

two control groups separately (data not shown) and

com-Table 1

Characteristics of the study population.

Controls (N = 1377) Cases (N = 666) Age at diagnosis (years)

Mean± SD 51.2± 11.69 59.5± 4.37 Range 32–85 31–84 Gender Females 608 (44.2%) 285 (42.8%) Males 769 (55.8%) 381 (57.2%) BMI ≤18.5 4(0.3%) 9 (1.4%) 18.6–24.99 462 (33.5%) 170 (25.5%) 25–29.99 53 (38.5%) 206 (30.9%) 30–34.99 161 (11.7%) 77 (11.5%) 35–39.99 30 (2.1%) 11 (1.7%) ≥40 10 (0.7%) 4 (0.6%) n.a. 182 (13.2%) 189 (28.4%) Smoking habit Nonsmokers 679 (49.3%) 321 (48.2%)

Former smokers (more than 10 years)

198 (14.4%) 144 (21.6%) Former smokers (less than

10 years)

45 (3.3%) 60 (9.0%)

Smokers 289 (20.9%) 97 (14.6%)

Missing 166 (12.1%) 44 (6.6%)

n.a., not available.

bined; the outcomes were the same. For this reason, we

have reported and discussed only the results from the

com-parison of the pooled controls with the colorectal cancer

cases.

Table 2

Distribution of MTHFR and MTRR genotypes in colorectal cancer patients and control subjects and results of binary logistic regression analysis.

Polymorphisms Controls (N = 1377) (%)a Cases (N = 666) (%)a OR 95% CI Global P-valueb 2_{and P-values} HWEc MTHFR 677 C > T (rs1801131) CC 613 (44.5%) 317 (47.6%) 1.00 (ref) – 0.03 1.76; 0.41 CT 627 (45.6%) 307 (46.1%) 0.96 0.78–1.17 TT 136 (9.9%) 42 (6.3%) 0.58 0.39–0.87 C-allele 1853 (67.3%) 941 (70.6%) 1.00 (ref) – 0.04 T-allele 899 (32.7%) 391 (29.4%) 0.86 0.74–0.99 1298 A > C (rs1801133) AA 583 (42.3%) 281 (42.2%) 1.00 (ref) – 0.87 0.88; 0.64 AC 638 (46.3%) 309 (46.4%) 1.06 0.86–1.30 CC 156 (11.4%) 76 (11.4%) 1.04 0.75–1.44 A-allele 1804 (65.5%) 871 (65.4%) 1.00 (ref) – 0.69 C-allele 950 (34.5%) 461 (34.6%) 1.03 0.89–1.19 MTRR 66 A > G (rs1801394) AA 291 (21.2%) 113 (17.1%) 1.00 (ref) – 0.08 0.29; 0.87 AG 671 (48.9%) 330 (49.9%) 1.21 0.92–1.58 GG 410 (29.9%) 218 (33.0%) 1.39 1.04–1.85 A-allele 1253 (45.7%) 556 (42.1%) 1.00 (ref) – 0.03 G-allele 1491 (54.3%) 766 (57.9%) 1.17 1.02–1.35 524 C > T (rs1532268) CC 554 (42.4%) 230 (38.9%) 1.00 (ref) – 0.24 0.46; 0.80 CT 603 (46.1%) 294 (49.7%) 1.20 0.96–1.50 TT 151 (11.5%) 67 (11.4%) 1.02 0.72–1.44 C-allele 1711 (65.4%) 754 (63.8%) 1.00 (ref) – 0.43 T-allele 905 (34.6%) 428 (36.2%) 1.06 0.91–1.24 1049 A > G (rs162036) AA 1044 (80.7%) 495 (83.9%) 1.00 (ref) – 0.25 1.60; 0.45 AG 231 (17.9%) 90 (15.3%) 0.85 0.64–1.12 GG 18 (1.4%) 5 (0.8%) 0.50 0.17–1.48 A-allele 2319 (89.7%) 1080 (91.5%) 1.00 (ref) 0.11 G-allele 267 (10.3%) 100 (8.5%) 0.81 0.62–1.05 1783 C > T (rs10380) CC 1076 (83.6%) 518 (87.3%) 1.00 (ref) – 0.09 0.64; 0.72 CT 204 (15.9%) 71 (12.0%) 0.71 0.52–0.96 TT 7 (0.5%) 4 (0.7%) 0.98 0.24–3.96 C-allele 2356 (91.5%) 1107 (93.3%) 1.00 (ref) 0.04 T-allele 218 (8.5%) 79 (6.7%) 0.74 0.56–0.99

OR, odds ratio; CI, 95% conﬁdence interval. Signiﬁcant P-values are in bold.

a_{Numbers may not add up to 100% of subjects due to genotyping failure. All samples that did not give a reliable result in the ﬁrst round of genotyping were resubmitted}

to up to three additional rounds of genotyping. Data points that were still not ﬁlled after this procedure were left blank.

b _{Adjusted for: gender and age.}

(5)

Table 3

Haplotype distribution of the four investigated MTHFR polymorphisms in colorectal cancer patients and control subjects.

Haplotypea _{Controls (%)}b _{Cases (%)}b _OR _{95% CI} _{Global P-value}

CA 925 (33.6%) 485 (36.4%) 1.00 – 0.08

CC 928 (33.7%) 456 (34.2%) 0.94 0.80–1.10 TA 877 (31.9%) 386 (29.0%) 0.84 0.71–0.99

TC 22 (0.8%) 5 (0.4%) 0.43 0.16–1.15

OR, odds ratio; CI, 95% conﬁdence interval. Signiﬁcant P-values are in bold.

a_{Loci: MTHFR 677 C > T (rs1801131); 1298 A > C (rs1801133).}

b_{n is the number of alleles. Because each individual has two alleles, the total number of alleles will be twice the total number of individuals. Individuals with missing}

haplotyping data were not included in the analyses.

Table 4

Haplotype distribution of the four investigated MTRR polymorphisms in colorectal cancer patients and control subjects.

Haplotypea,b _{Controls (%)}c _{Cases (%)}c _OR _{95% CI} _{Global P-value}

ACAC 601 (23.8%) 260 (22.3%) 1.00 – 0.56 GCAC 789 (31.3%) 385 (33.0%) 1.13 0.93–1.36 GTAC 556 (22.1%) 268 (23.0%) 1.11 0.91–1.37 ATAC 308 (12.2%) 149 (12.8%) 1.12 0.88–1.43 ACGT 186 (7.4%) 71 (6.1%) 0.88 0.65–1.20 ACGC 49 (1.9%) 21 (1.8%) 0.99 0.58–1.69 GCGT 18 (0.7%) 4 (0.3%) 0.51 0.17–1.53 GCGC 5 (0.2%) 2 (0.2%) 0.93 0.18–4.80 Othersb _{8 (0.4%)} _{6 (0.5%)}

OR, odds ratio; CI, 95% conﬁdence interval.

a_{Loci: MTRR 66 A > G (rs1801394); 524 C > T (rs1532268); 1049 A > G (rs162036); 1783 C > T (rs10380).}

b_{The following haplotypes are not presented in the table because in the study population each of them was represented only in less than 5 copies: ACAT, ATGC, GTGC,}

GCAT, ATGT, and ATAT.

c_{n is the number of alleles. Because each individual has two alleles, the total number of alleles will be twice the total number of individuals. Individuals with missing}

haplotyping data were not included in the analyses.

3.2.1. The MTHFR gene

Associations of the two polymorphisms in the MTHFR gene and

their related haplotypes were analysed. A signiﬁcant difference was

found between the cases and the controls in the TT-genotype

fre-quencies for the MTHFR C677T polymorphism (OR, 0.58; 95% CI,

0.39–0.87; global P = 0.03,

Table 2

). This different distribution was

also noted in the allele frequency analyses for the T-allele (OR, 0.86;

95% CI, 0.74–0.99, global P = 0.04;

Table 2

).

All four possible haplotypes for MTHFR were observed in both

cases and controls. Although there was some variation in haplotype

distribution between the cases and the controls, no statistically

signiﬁcant differences were observed (global P = 0.08) (

Table 3

).

However, compared with the most common haplotype (CA), the

TA haplotype was less frequent among the cases than among the

controls (OR, 0.84; 95% CI, 0.71–0.99).

3.2.2. The MTRR gene

A different distribution of the C1783T and the A66G genotypes

for the MTRR gene was found in cases and controls, although the

global P-values were not statistically signiﬁcant (global P = 0.08 and

global P = 0.09, respectively). In particular, carriers of the GG

geno-type of the A66G polymorphism had an increased risk for colorectal

cancer (OR, 1.39; 95% CI, 1.04–1.85), while those with the CT

geno-type of C1783T had a decreased risk (OR, 0.71; 95% CI, 0.52–0.96)

(

Table 2

). The allele frequency-analysis revealed that carriers of

the variant T allele of the MTRR C1783T polymorphism had a

decreased risk for colorectal cancer (OR, 0.74; 95% CI, 0.56–0.99,

global P = 0.04), while carriers of the G allele for the MTRR A66G

polymorphism had an increased risk (OR, 1.17; 95% CI, 1.02–1.35,

global P = 0.03) (

Table 2

).

Out of the 16 possible haplotypes for MTRR, 14 were noted in

the Czech study population. However, six haplotypes (ACAT, ATGC,

GTGC, GCAT, ATGT, and ATAT) were represented in the population

with less than ﬁve copies, and then were not reported. No

dif-ferences in the haplotype frequencies between the cases and the

controls were observed (

Table 4

).

4. Discussion

Colorectal cancer is a complex disease inﬂuenced by genetic

and environmental factors and their interactions

[36]

. Growing

evidence suggests that an appreciable component of the genetic

contribution to ‘sporadic’ colorectal cancer is due to common

vari-ants with individually small effects, thereby invoking the common

disease–common variant paradigm for this cancer

[37]

.

In the present case–control study, we have employed a

candidate-gene approach to investigate the associations between

six allelic variants distributed in two genes (MTHFR and MTRR)

drawn from the folate-metabolism pathway, and the risk for

col-orectal cancer. The results of this study showed a signiﬁcantly

different distribution of some alleles (T allele in MTHFR C677T, T

allele in MTRR C1783T and G allele in MTRR A66G) between cases

and controls. A signiﬁcant association of three polymorphisms (one

in MTHFR and two in MTRR) with susceptibility to colorectal cancer

was also observed. In particular, individuals carrying MTHFR 677TT

and MTRR 1783CT genotypes were at a decreased risk for

colorec-tal cancer, while the MTRR 66GG genotype was associated with an

increased risk.

Since the function of MTHFR and MTRR is essential for

provid-ing methyl groups, it is highly likely that enzymatic variants due

to functional polymorphisms may alter DNA methylation, which

would greatly affect carcinogenesis

[38]

. It is also biologically

plau-sible that polymorphisms or gene–environment interactions rather

than the folate intake alone would have an impact on the risk for

colorectal cancer since functional SNPs in folate-related genes

con-tribute to the alteration of folate metabolism

[2]

.

The association observed between the C677T SNP in the MTHFR

gene and susceptibility to colorectal cancer is in agreement with

the majority of studies published so far, providing good evidence

that homozygosity for the T-allele is associated with a modest, but

signiﬁcantly reduced risk for this cancer, also in the Czech

popu-lation. A recent meta-analysis

[39]

of 20 studies including 10,131

colorectal cancer patients and 15,362 controls, suggested that the

(6)

677T allele provides a protective effect against colorectal cancer

risk. Moreover, Taioli et al. published an interesting HuGE review

on the MTHFR C677T polymorphism, in which the 677TT genotype

appeared to be associated with a reduced risk for colorectal cancer

[11]

. The C677T SNP in the MTHFR gene inﬂuences the activity of

the enzyme, particularly in folate-deﬁcient conditions

[8]

.

Individ-uals with the 677TT genotype were observed to have no more than

30% of the normal enzyme activity, while heterozygotes (CT) had

65% of the normal enzyme activity

[8]

. Moreover, 677TT

homozy-gotes show a signiﬁcantly higher mean homocysteine level and

more elevated risk for stroke and coronary disease than people

who are 677CC homozygotes

[39]

. Decreased levels of MTHFR may

adversely affect the methylation of oncogenes and tumor

suppres-sor genes, contributing to carcinogenesis. In addition, depletion of

MTHFR interferes with thymidylate biosynthesis, which may lead

to an accumulation of deoxyuridylate in DNA. Subsequent removal

of this abnormal base may impair the integrity of DNA

[40]

. Thus,

the balance between DNA synthesis and DNA methylation, which

is also determined by the above-mentioned MTHFR polymorphism,

may modulate cancer risk. However, diet, and particularly folic acid

intake, can additionally modify the effects of the polymorphisms in

the MTHFR gene. No deﬁnite conclusion about the relation between

folate intake and risk for colorectal cancer has been possible at

present

[41]

. More careful stratiﬁcation analyses that take into

account the clinical character and diet, smoking status and alcohol

consumption are needed

[39]

.

Interestingly, the haplotype analysis based on the two

inves-tigated MTHFR polymorphisms (C677T and A1298C) showed that

haplotype TA was moderately less common in cases than controls.

This haplotype, when compared to the most common haplotype

CA, was associated with a decreased risk for colorectal cancer. This

outcome appears logical, if one considers the results of the

geno-type analysis for the individual C677T polymorphism in the MTHFR

gene. Only a few studies investigated the effect of haplotypes in

colorectal cancer susceptibility and the majority of them focused

on the potentially predictive role of C677T and A1298C variants on

toxicity and efﬁcacy of antifolate and ﬂuoropyrimidine agents (as

recently reviewed

[42]

).

An effect of the MTRR polymorphisms on adenoma/colorectal

cancer risk has not been consistently found, but the number of

stud-ies is quite small

[9,25–27]

. While functional effects are not yet fully

understood, the 66G allele is considered to decrease the enzyme

activity compared with the 66A allele

[43]

. It may be

hypothe-sized that subjects carrying the MTRR 66GG genotype have reduced

methionine levels compared with carriers of other genotypes. Our

observation of an increased risk for colorectal cancer among the

66GG genotype carriers is consistent with this hypothesis. Two

pre-vious studies have provided similar results for the association of

this SNP with colorectal cancer risk

[28,29]

.

Haplotype analysis for the polymorphisms in the MTRR gene

(A66G; C524T; A1049G; C1793T) did not show any association with

colorectal cancer. To our knowledge, this is one of the largest studies

investigating simultaneously the role of four MTRR polymorphisms,

and relative haplotypes, in modulating the susceptibility to

colorec-tal cancer. However, the relatively low numbers of observations for

this haplotype may limit the interpretations.

The present case–control study has several strengths. First, it

includes an adequate number of cases and controls recruited in

the same geographical area, collected during the same period of

time and from a country with a high incidence of colorectal cancer.

Additionally, the study uses two different groups of controls. The

ﬁrst control group, consisting of colonoscopy-negative individuals,

was used as a truly cancer-free control group. The negative result

of colonoscopy serves as the best available proof of the absence

of colorectal cancer

[34]

. Nevertheless, there is a concern that the

colonoscopy-negative control group comprised patients with

exist-ing medical conditions that required examination, and that they

may carry unknown risk factors for colorectal cancer. To avoid this

problem, we included a second control group of individuals who

had donated blood at the blood donor center and thus represented

a healthy, general population. The analysis of haplotypes represents

a much more powerful approach than those analysing only

individ-ual polymorphisms. This approach also ensures increased statistical

power. Assignment of alleles to chromosomes/haplotypes also

pro-vides important information on recombination (physical exchange

of DNA during meiosis), vital for locating disease-causing

muta-tions by linkage methods

[44]

. Moreover, our study includes one

of the largest populations with colorectal cancer investigated for

MTRR haplotypes so far. A few studies have previously looked at

the same genotype/haplotypes for the MTRR gene in association

with colorectal cancer/adenoma risk. In particular, Koushik et al.

analysed 24 SNPs (including all the variants included in the present

study) but on a relatively small population of colorectal cancer cases

(376) and controls (849)

[25]

. They found that carriers of the MTRR

C25088T and T14208A variants had an increased risk for

colorec-tal cancer. Similarly, Hazra et al. analysed 24 SNPs (including the

two MTHFR polymorphisms and the four MTRR variants included

in our study) in association with colorectal adenoma risk

[9]

. There

was some suggestion that the variant alleles of the C25088T and

T14208A polymorphisms in the MTRR gene were associated with

an increased risk of advanced colorectal adenoma.

Dietary folate and other nutrient intake have been shown to

inﬂuence the risk for different cancers, including colorectal

can-cer. On the other hand, this type of cancer has been associated

with variants in genes involved in the folate metabolic pathway

[2]

. However, as we could not reliably control for dietary folate

intake, the interaction between genes and nutrients remains to

be assessed, as postulated by

[45,46]

. The collection of

colorec-tal cancer cases and controls included in this study was intended

for genetic association studies in general

[47–49]

, which did not

strictly require this kind of data collection.

In conclusion, our study suggests that the 677TT genotype of

the MTHFR gene may provide a protective effect against

colorec-tal cancer risk. On the other hand, the MTRR 66GG genotype may

be associated with an increased risk for this cancer. These results

indicate a predominant role of the genes involved in the folate

metabolism in cancer risk.

Conﬂicts of interest statement

The authors declare that there are no conﬂicts of interests.

Acknowledgements

The authors would like to thank Thomas O’Hearn II for his

excel-lent technical support.

Funding: This work was supported by the Grant Agency of the

Czech Republic, grants GACR 305/09/P194, GACR P304/10/1286 and

by Internal Grant Agency of the Ministry of Health of the Czech

Republic, grant IGA NS 10230-3/2009.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in

the online version, at

doi:10.1016/j.mrgentox.2010.12.008

.

References

[1] F.Y. Li, M.D. Lai, Colorectal cancer, one entity or three, J. Zhejiang Univ. Sci. B 10 (2009) 219–229.

[2] L. Sharp, J. Little, Polymorphisms in genes involved in folate metabolism and colorectal neoplasia: a HuGE review, Am. J. Epidemiol. 159 (2004) 423–443.

(7)

[3] X.L. Xu, J. Yu, H.Y. Zhang, M.H. Sun, J. Gu, X. Du, D.R. Shi, P. Wang, Z.H. Yang, J.D. Zhu, Methylation proﬁle of the promoter CpG islands of 31 genes that may contribute to colorectal carcinogenesis, World J. Gastroenterol. 10 (2004) 3441–3454.

[4] M. Lucock, Is folic acid the ultimate functional food component for disease prevention? BMJ 328 (2004) 211–214.

[5] M. Lucock, Folic acid: nutritional biochemistry, molecular biology, and role in disease processes, Mol. Genet. Metab. 71 (2000) 121–138.

[6] S. Kono, K. Chen, Genetic polymorphisms of methylenetetrahydrofolate reduc-tase and colorectal cancer and adenoma, Cancer Sci. 96 (2005) 535–542. [7] N.M. van der Put, F. Gabreels, E.M. Stevens, J.A. Smeitink, F.J. Trijbels, T.K.

Eskes, L.P. van den Heuvel, H.J. Blom, A second common mutation in the methylenetetrahydrofolate reductase gene: an additional risk factor for neural-tube defects? Am. J. Hum. Genet. 62 (1998) 1044–1051.

[8] P. Frosst, H.J. Blom, R. Milos, P. Goyette, C.A. Sheppard, R.G. Matthews, G.J. Boers, M. den Heijer, L.A. Kluijtmans, L.P. van den Heuvel, et al., A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydro-folate reductase, Nat. Genet. 10 (1995) 111–113.

[9] A. Hazra, K. Wu, P. Kraft, C.S. Fuchs, E.L. Giovannucci, D.J. Hunter, Twenty-four non-synonymous polymorphisms in the one-carbon metabolic pathway and risk of colorectal adenoma in the Nurses’ Health Study, Carcinogenesis 28 (2007) 1510–1519.

[10] R.A. Hubner, R.S. Houlston, MTHFR C677T and colorectal cancer risk: a meta-analysis of 25 populations, Int. J. Cancer 120 (2007) 1027–1035.

[11] E. Taioli, M.A. Garza, Y.O. Ahn, D.T. Bishop, J. Bost, B. Budai, K. Chen, F. Gemignani, T. Keku, C.S. Lima, L. Le Marchand, K. Matsuo, V. Moreno, J. Plaschke, M. Pufulete, S.B. Thomas, G. Toffoli, C.R. Wolf, C.G. Moore, J. Little, Meta- and pooled analy-ses of the methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism and colorectal cancer: a HuGE-GSEC review, Am. J. Epidemiol. 170 (2009) 1207–1221.

[12] D.J. Gaughan, L.A. Kluijtmans, S. Barbaux, D. McMaster, I.S. Young, J.W. Yarnell, A. Evans, A.S. Whitehead, The methionine synthase reductase (MTRR) A66G polymorphism is a novel genetic determinant of plasma homocysteine con-centrations, Atherosclerosis 157 (2001) 451–456.

[13] D. Gemmati, A. Ongaro, G.L. Scapoli, M. Della Porta, S. Tognazzo, M.L. Serino, E. Di Bona, F. Rodeghiero, G. Gilli, R. Reverberi, A. Caruso, M. Pasello, A. Pel-lati, M. De Mattei, Common gene polymorphisms in the metabolic folate and methylation pathway and the risk of acute lymphoblastic leukemia and non-Hodgkin’s lymphoma in adults, Cancer Epidemiol. Biomarkers Prev. 13 (2004) 787–794.

[14] P. Zavadakova, B. Fowler, T. Suormala, Z. Novotna, P. Mueller, J.B. Hennermann, J. Zeman, M.A. Vilaseca, L. Vilarinho, S. Gutsche, E. Wilichowski, G. Horneff, V. Kozich, cblE type of homocystinuria due to methionine synthase reductase deﬁ-ciency: functional correction by minigene expression, Hum. Mutat. 25 (2005) 239–247.

[15] Z. Zhang, Q. Shi, Z. Liu, E.M. Sturgis, M.R. Spitz, Q. Wei, Polymorphisms of methionine synthase and methionine synthase reductase and risk of squamous cell carcinoma of the head and neck: a case–control analysis, Cancer Epidemiol. Biomarkers Prev. 14 (2005) 1188–1193.

[16] R.Z. Stolzenberg-Solomon, Y.L. Qiao, C.C. Abnet, D.L. Ratnasinghe, S.M. Dawsey, Z.W. Dong, P.R. Taylor, S.D. Mark, Esophageal and gastric cardia cancer risk and folate- and vitamin B(12)-related polymorphisms in Linxian, China, Cancer Epidemiol. Biomarkers Prev. 12 (2003) 1222–1226.

[17] Q. Shi, Z. Zhang, G. Li, P.C. Pillow, L.M. Hernandez, M.R. Spitz, Q. Wei, Polymor-phisms of methionine synthase and methionine synthase reductase and risk of lung cancer: a case–control analysis, Pharmacogenet. Genomics 15 (2005) 547–555.

[18] A. Gast, J.L. Bermejo, T. Flohr, M. Stanulla, B. Burwinkel, M. Schrappe, C.R. Bartram, K. Hemminki, R. Kumar, Folate metabolic gene polymorphisms and childhood acute lymphoblastic leukemia: a case–control study, Leukemia 21 (2007) 320–325.

[19] S.Y. Kwak, U.K. Kim, H.J. Cho, H.K. Lee, H.J. Kim, N.K. Kim, S.G. Hwang, Methylenetetrahydrofolate reductase (MTHFR) and methionine synthase reductase (MTRR) gene polymorphisms as risk factors for hepatocel-lular carcinoma in a Korean population, Anticancer Res. 28 (2008) 2807–2811.

[20] S. Ohnami, Y. Sato, K. Yoshimura, H. Sakamoto, K. Aoki, H. Ueno, M. Ikeda, C. Morizane, K. Shimada, Y. Sakamoto, M. Esaki, I. Saito, H. Hirose, D. Saito, H. Sugimura, T. Kosuge, T. Okusaka, T. Yoshida, His595Tyr polymorphism in the methionine synthase reductase (MTRR) gene is associated with pancreatic cancer risk, Gastroenterology 135 (2008) 477–488.

[21] S.M. Collin, C. Metcalfe, L. Zuccolo, S.J. Lewis, L. Chen, A. Cox, M. Davis, J.A. Lane, J. Donovan, G.D. Smith, D.E. Neal, F.C. Hamdy, J. Gudmundsson, P. Sulem, T. Rafnar, K.R. Benediktsdottir, R.A. Eeles, M. Guy, Z. Kote-Jarai, J. Morrison, A.A. Al Olama, K. Stefansson, D.F. Easton, R.M. Martin, Association of folate-pathway gene polymorphisms with the risk of prostate cancer: a population-based nested case–control study, systematic review, and meta-analysis, Cancer Epidemiol. Biomarkers Prev. 18 (2009) 2528–2539.

[22] J. Lissowska, M.M. Gaudet, L.A. Brinton, S.J. Chanock, B. Peplonska, R. Welch, W. Zatonski, N. Szeszenia-Dabrowska, S. Park, M. Sherman, M. Garcia-Closas, Genetic polymorphisms in the one-carbon metabolism pathway and breast cancer risk: a population-based case–control study and meta-analyses, Int. J. Cancer 120 (2007) 2696–2703.

[23] T. Suzuki, K. Matsuo, K. Hirose, A. Hiraki, T. Kawase, M. Watanabe, T. Yamashita, H. Iwata, K. Tajima, One-carbon metabolism-related gene polymorphisms and risk of breast cancer, Carcinogenesis 29 (2008) 356–362.

[24] S.Y. Tong, J.M. Lee, E.S. Song, K.B. Lee, M.K. Kim, Y.M. Yun, J.K. Lee, S.K. Son, J.P. Lee, J.H. Kim, S.Y. Hur, Y.I. Kwon, The effects of polymorphisms in methylenetetrahydrofolate reductase (MTHFR), methionine synthase (MTR), and methionine synthase reductase (MTRR) on the risk of cervical intraepithe-lial neoplasia and cervical cancer in Korean women, Cancer Causes Control 21 (2010) 23–30.

[25] A. Koushik, P. Kraft, C.S. Fuchs, S.E. Hankinson, W.C. Willett, E.L. Giovan-nucci, D.J. Hunter, Nonsynonymous polymorphisms in genes in the one-carbon metabolism pathway and associations with colorectal cancer, Cancer Epi-demiol. Biomarkers Prev. 15 (2006) 2408–2417.

[26] S.E. Steck, T. Keku, L.M. Butler, J. Galanko, B. Massa, R.C. Millikan, R.S. Sandler, Polymorphisms in methionine synthase, methionine synthase reductase and serine hydroxymethyltransferase, folate and alcohol intake, and colon cancer risk, J Nutrigenet Nutrigenomics 1 (2008) 196–204.

[27] E. Theodoratou, S.M. Farrington, A. Tenesa, G. McNeill, R. Cetnarskyj, R.A. Bar-netson, M.E. Porteous, M.G. Dunlop, H. Campbell, Dietary vitamin B6 intake and the risk of colorectal cancer, Cancer Epidemiol. Biomarkers Prev. 17 (2008) 171–182.

[28] L. Le Marchand, T. Donlon, J.H. Hankin, L.N. Kolonel, L.R. Wilkens, A. Seifried, B-vitamin intake, metabolic genes, and colorectal cancer risk (United States), Cancer Causes Control 13 (2002) 239–248.

[29] K. Matsuo, N. Hamajima, T. Hirai, T. Kato, M. Inoue, T. Takezaki, K. Tajima, Methionine synthase reductase gene A66G polymorphism is associated with risk of colorectal cancer, Asian Pac. J. Cancer Prev. 3 (2002) 353–359. [30] T. Otani, M. Iwasaki, T. Hanaoka, M. Kobayashi, J. Ishihara, S. Natsukawa, K.

Shaura, Y. Koizumi, Y. Kasuga, K. Yoshimura, T. Yoshida, S. Tsugane, Folate, vita-min B6, vitavita-min B12, and vitavita-min B2 intake, genetic polymorphisms of related enzymes, and risk of colorectal cancer in a hospital-based case–control study in Japan, Nutr. Cancer 53 (2005) 42–50.

[31] M.M. Center, A. Jemal, E. Ward, International trends in colorectal cancer inci-dence rates, Cancer Epidemiol. Biomarkers Prev. 18 (2009) 1688–1694. [32] H.F. Vasen, J.P. Mecklin, P.M. Khan, H.T. Lynch, The International Collaborative

Group on Hereditary Non-Polyposis Colorectal Cancer (ICG-HNPCC), Dis. Colon Rectum 34 (1991) 424–425.

[33] H.F. Vasen, P. Watson, J.P. Mecklin, H.T. Lynch, New clinical criteria for hered-itary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative group on HNPCC, Gastroenterology 116 (1999) 1453–1456.

[34] H. Singh, D. Turner, L. Xue, L.E. Targownik, C.N. Bernstein, Risk of devel-oping colorectal cancer following a negative colonoscopy examination: evidence for a 10-year interval between colonoscopies, JAMA 295 (2006) 2366–2373.

[35] D.J. Schaid, C.M. Rowland, D.E. Tines, R.M. Jacobson, G.A. Poland, Score tests for association between traits and haplotypes when linkage phase is ambiguous, Am. J. Hum. Genet. 70 (2002) 425–434.

[36] P.Y. Cheah, Recent advances in colorectal cancer genetics and diagnostics, Crit. Rev. Oncol. Hematol. 69 (2009) 45–55.

[37] A. Tenesa, M.G. Dunlop, New insights into the aetiology of colorectal cancer from genome-wide association studies, Nat. Rev. Genet. 10 (2009) 353–358. [38] S. Friso, S.W. Choi, Gene–nutrient interactions in one-carbon metabolism, Curr.

Drug Metab. 6 (2005) 37–46.

[39] Y. Huang, S. Han, Y. Li, Y. Mao, Y. Xie, Different roles of MTHFR C677T and A1298C polymorphisms in colorectal adenoma and colorectal cancer: a meta-analysis, J. Hum. Genet. 52 (2007) 73–85.

[40] R.S. Houlston, I.P. Tomlinson, Polymorphisms and colorectal tumor risk, Gas-troenterology 121 (2001) 282–301.

[41] S.J. Eussen, S.E. Vollset, J. Igland, K. Meyer, A. Fredriksen, P.M. Ueland, M. Jenab, N. Slimani, P. Boffetta, K. Overvad, A. Tjonneland, A. Olsen, F. Clavel-Chapelon, M.C. Boutron-Ruault, S. Morois, C. Weikert, T. Pischon, J. Linseisen, R. Kaaks, A. Trichopoulou, D. Zilis, M. Katsoulis, D. Palli, F. Berrino, P. Vineis, R. Tumino, S. Panico, P.H. Peeters, H.B. Bueno-de-Mesquita, F.J. van Duijnhoven, I.T. Gram, G. Skeie, E. Lund, C.A. Gonzalez, C. Martinez, M. Dorronsoro, E. Ardanaz, C. Navarro, L. Rodriguez, B. Van Guelpen, R. Palmqvist, J. Manjer, U. Ericson, S. Bingham, K.T. Khaw, T. Norat, E. Riboli, Plasma folate, related genetic variants, and colorectal cancer risk in EPIC, Cancer Epidemiol Biomarkers Prev 19 (2010) 1328–1340.

[42] E. De Mattia, G. Toffoli, C677T and A1298C MTHFR polymorphisms, a chal-lenge for antifolate and ﬂuoropyrimidine-based therapy personalisation, Eur. J. Cancer 45 (2009) 1333–1351.

[43] H. Olteanu, T. Munson, R. Banerjee, Differences in the efﬁciency of reductive activation of methionine synthase and exogenous electron acceptors between the common polymorphic variants of human methionine synthase reductase, Biochemistry 41 (2002) 13378–13385.

[44] D.C. Crawford, D.A. Nickerson, Deﬁnition and clinical importance of haplotypes, Annu. Rev. Med. 56 (2005) 303–320.

[45] O. Mayer Jr., J. Simon, H. Rosolova, A population study of the inﬂuence of beer consumption on folate and homocysteine concentrations, Eur. J. Clin. Nutr. 55 (2001) 605–609.

[46] K. Vesela, M. Pavlikova, B. Janosikova, M. Andel, J. Zvarova, J. Hyanek, V. Kozich, Genetic determinants of folate status in Central Bohemia, Physiol. Res. 54 (2005) 295–303.

[47] B. Pardini, A. Naccarati, J. Novotny, Z. Smerhovsky, L. Vodickova, V. Polakova, M. Hanova, J. Slyskova, E. Tulupova, R. Kumar, M. Bortlik, R. Barale, K. Hemminki, P. Vodicka, DNA repair genetic polymorphisms and risk of colorectal cancer in the Czech Republic, Mutat. Res. 638 (2008) 146–153.

(8)

[48] D. Landi, F. Gemignani, A. Naccarati, B. Pardini, P. Vodicka, L. Vodickova, J. Novotny, A. Forsti, K. Hemminki, F. Canzian, S. Landi, Polymorphisms within micro-RNA-binding sites and risk of sporadic colorectal cancer, Carcinogenesis 29 (2008) 579–584.

[49] V. Polakova, B. Pardini, A. Naccarati, S. Landi, J. Slyskova, J. Novotny, L. Vodick-ova, J.L. Bermejo, M. HanVodick-ova, Z. Smerhovsky, E. TulupVodick-ova, R. Kumar, K. Hem-minki, P. Vodicka, Genotype and haplotype analysis of cell cycle genes in spo-radic colorectal cancer in the Czech Republic, Hum. Mutat. 30 (2009) 661–668.