Gene Polymorphisms Involved In One Carbon Metabolism Biology Essay

Published:

Polymorphisms in the genes encoding enzymes in the folate-mediated one-carbon metabolism pathway have been implicated as maternal risk factors for giving rise to Down syndrome but results have been observed to vary with different populations and ethnicity. The SNPs in two key enzymes, one each of folate-dependent and folate-independent pathways were evaluated as risk factors for giving rise to babies with Down syndrome. The frequency of the MTHFR 677C>T and BHMT 742G>A polymorphisms was analysed in an Indian cohort consisting of 102 case-mothers and 98 control-mothers. The frequency of the variant 'T' allele for the MTHFR SNP was 0.11 in case-mothers and 0.14 in control-mothers. The frequency of the variant 'A' allele for the BHMT SNP was 0.30 in case-mothers and 0.28 in control-mothers. Since the frequencies did not differ significantly, our findings did not support the association of the SNPS with the risk of Down syndrome in Indian population.

Key words: Down syndrome, polymorphisms, one-carbon metabolism, MTHFR, BHMT

Introduction

Lady using a tablet
Lady using a tablet

Professional

Essay Writers

Lady Using Tablet

Get your grade
or your money back

using our Essay Writing Service!

Essay Writing Service

The most common form of aneuploidy sufficiently viable to survive to term in relatively large numbers is Down syndrome (DS) [Cuckle, 2005]. DS is a type of birth defect associated with chromosomal abnormality (extra copy of chromosome 21) in the affected children. A congenital disorder, it is the most common cause of mental retardation [James et al., 1999]. The association between DS and a third copy of chromosome 21 was discovered in 1959 but the molecular basis is still unclear [Lejeune et al., 1959]. Meiotic non-disjunction (abnormal segregation of chromosomes during meiosis) resulting in disomic maternal gamete (ovum) more than 90% of the time, has been addressed as the pocause for the extra copy of chromosome 21 [James et al., 1999]. Fertilization of an abnormal (disomic) maternal gamete with normal (monosomic) paternal gamete gives rise to three copies of chromosome 21 - i.e. chromosome 21 trisomy in the fetus [Saenz, 1999]. The excess genetic material arising as a result of a third copy of chromosome 21 has a profound effect on multiple systems and leads to mental retardation and a host of malformations such as congenital heart disease, thyroid disease, loss of hearing, opthalmological anomalies amongst others [Antonarakis and Epstein, 2006; Committee on Genetics, 2001; Bromham et al., 2002]. The incidence of DS is ~1 in 600 to 1 in 1,000 live births [Hobbs et al., 2000]. In the Indian population, the prevalence of DS was 0.98 per 1,000 according to the data obtained from the National Neonatal - Perinatal Database, 1995 (India) [NNPD, 1997] whereas regional studies carried out in some parts of India saw a prevalence range of 0.66 to 1.17 per 1,000[Bharucha, 1998; Verma et al., 1998; Modi et al., 1998; Isaac et al., 1985].

The complexities associated with this birth defect are huge, by and large and unravelling of the mystery is of critical importance taking into consideration that there is no medical cure. The strongest epidemiological variable linked to DS is advanced maternal age [Cuckle, 2005]. But, of late children with DS are born to young mothers.

One-carbon metabolism

Folates play an important role in the acceptance and the transfer of one-carbon units in the synthesis of thymidylate and purines, amino acid inter-conversions including conversion of serine to glycine, methionine regeneration by remethylation of homocysteine, catabolism of histidine, and the formation of S-adenosylmethionine (SAM) - the primary methylating agent in the body required for many biological methylation reactions [Gregory, 2001; Lucock et al., 2005].

The absorbed folate from the diet (monoglutamate form- folic acid) gets reduced to 5, 6, 7, 8-tetrahydrofolate (THF) via 7, 8-dihydrofolate (DHF) by the enzyme dihydrofolate reductase, and later to 5,10-methylene THF and eventually to 5-methyl THF [Figueiredo et al., 2008]. The irreversible reduction of 5, 10-methylene THF to 5-methyl THF is brought about by the flavin adenine dinucleotide (FAD)-dependent 5,10-methylene tetrahydrofolate reductase enzyme (MTHFR, EC 1.5.1.20). The product resulting from the MTHFR enzyme activity; 5-methylTHF is the methyl donor in the remethylation of homocysteine to methionine catalyzed by vitamin B12-dependent methionine synthase, (MS; EC 2.1.1.13) [Hustad et al., 2007]. The remethylation of homocysteine to methionine can take place by both, the folate- dependent and -independent pathways. In the folate dependent pathway, the activity of 5, 10- methylene tetrahydrofolate reductase leads to the availability of a methyl group for which is utilized by methionine synthase enzyme for remethylation. In the folate independent pathway, the methyl group is donated by betaine, in a reaction catalyzed by betaine homocysteine methyltransferase (BHMT; EC 2.1.1.5) [da Costa et al., 2006]. After donating the methyl group for the remethylation of homocysteine, betaine gets transformed into dimethylglycine [Forges, 2007]. The dimethylglycine so formed, gets oxidized to form glycine. This leads to the introduction of one - carbon units into the folate pool that stimulates the folate - dependent homocysteine remethylation, thus resulting in decreased homocysteine levels [Heil, 2000]. Methionine formed by remethylation of homocysteine then gets converted into S - Adenosyl Methionine (SAM) by accepting adenosine from ATP molecule in a reaction catalyzed by methionine adenosyl transferase [Lucock et al., 2005]. SAM donates the methyl group during the various methylation reactions, catalyzed by specific methyltransferases, and then forms S-adenosyl homocysteine (SAH). SAH gets hydrolyzed into homocysteine in a reversible reaction. Homocysteine may undergo trans-sulphuration to give rise to cystathionine or it may undergo remethylation to form methionine [Steegers-Theunissen, 1995].

Lady using a tablet
Lady using a tablet

Comprehensive

Writing Services

Lady Using Tablet

Plagiarism-free
Always on Time

Marked to Standard

Order Now

SAM is involved in methylation of cytosine residues in DNA. DNA methylation makes a gene transcriptionally inactive. Thus, gene expression is controlled by DNA methylation. When there is dietary folate deficiency and/or reduced activity of MTHFR enzyme, the generation of SAM is limited. This may result in DNA hypo-methylation and consequential inappropriate extent of gene inactivation [Narayanan et al., 2004]. DNA hypo-methylation has been linked to abnormal chromosomal segregation [Rosenblatt, 1999] that may be a risk factor for DS-birth. Chromosomal instability and higher homocysteine transsulphuration may be linked to defects in the genes involved in one-carbon metabolism [Gueant et al., 2005].

MTHFR 677C>T SNP

A 677C>T transition at the position 677 in the exon 4 on chromosome 1 was identified in the coding region of MTHFR, which resulted in the change of an alanine residue, which is highly conserved, to a valine residue in the enzyme at the position 222. This mutation led to the introduction of restriction site for the enzyme Hinf I. The homozygous variant of this gene showed decreased enzyme activity and increased susceptibility to heat in vitro [Frosst et al., 1995].

The variant ('T') allele frequency has shown marked variation according to the geographic location as well as ethnicity; for example, 33.3% in the Portugal study [Castro et al., 2003], 42% in Spain [Guillen et al., 2001] and 43.8% in Italy [Botto & Yang, 2000]. Among the African- Americans, the mutation was found to be relatively infrequent [Stevenson et al., 1997]. In a meta-analysis based on data obtained from 40 studies by Klerk and colleagues, the allelic frequencies ranged from 3.2% (UK Indians) to 30.2% (Italy in 1998) [Klerk et al., 2002].

James and colleagues hypothesized that the MTHFR mutation would be a predisposing factor for abnormal DNA methylation and lead to a higher risk of meiotic non-disjunction which was thought to be the major cause of the presence of 3 copies in Trisomy 21. Their study was the first to link abnormal folate metabolism and MTHFR 677 C>T mutation with DS. Positive association of the MTHFR SNP as a risk factor for giving rise to children with Down syndrome was obtained in subsequent studies [Hobbs et al., 2000; Meguid et al., 2008; Wang et al., 2007]. Nevertheless, negative findings have also been reported [Petersen et al., 2001; Stuppia et al., 2002; Yanamandra et al., 2003; Chango et al., 2005; Gueant et al., 2003].

BHMT 742G>A SNP

Another polymorphism of interest was the BHMT 742G>A SNP. Deficiency of vitamins B12 and folate, mild hyperhomocysteinemia, has been reported in several studies in different parts of India [Refsum et al., 2001; Chandrasekhar et al., 1980; Vijayalakshmi & Shobana, 1982; Balaji & Dustagheer, 2000]. In case folate metabolic pathway gets compromised as a result of gene defects or low availability of micronutrients such as vitamin B12 and folate, the folate-independent pathway of remethylation of homocysteine would be operational and BHMT would have a crucial role to play [Chiuve et al., 2007; Ananth et al., 2007]. The Guanine to Adenine change at the 742 position in exon 6 leads to the replacement of an arginine with a glutamine residue at codon 239 (R239Q) in the protein [Weisberg et al., 2003]. In vitro studies carried out on the 742G>A SNP did not find any differences in thermostability of the enzyme nor any effect on the plasma homocysteine concentrations [Weisberg et al., 2003]. BHMT mutation has been subsequently associated with cardiovascular disease [Weisberg et al., 2003], placental abruption [Ananth et al., 2007] and neural tube defects [Boyles et al., 2006; Morin et al., 2003]. To the best of our knowledge, no studies have been carried out to demonstrate the association of the BHMT 742 G>A SNP with DS. Hence, we decided to analyze it as well. Thus, the primary purpose of the study was to analyse mutations in genes coding for key enzymes of both the folate-dependent as well as folate-independent pathways (MTHFR and BHMT) for the remethylation of homocysteine to methionine as maternal risk factors for giving birth to children with DS in the Indian population.

Materials & methods

Subjects:

The study was approved by the Ethics Committee of Bai Jerbai Wadia Hospital for Children, Mumbai (B.J.W.H.C.). The study participants (case-and control-mothers) were unrelated and of Indian origin. Enrolment of all study participants was made with their written informed consent. The inclusion criteria were as follows: (a) mothers who had given birth to offspring affected with Down syndrome (case-mothers) (b) mothers of healthy offspring (control-mothers) (c) a written informed consent to participate in the study. The exclusion criteria included: (a) history of major illness- namely cancer, renal/liver disease (b) participation in another clinical trial within a month of enrolment in the present study. A detailed case record form (CRF) was filled out for every participant in the study. The CRF contained information on demographics, the medical and obstetric/gynaecology history of the mother, information about her affected as well as unaffected children and any current medication, and dietary information using a 24-hour recall and food frequency questionnaire method. The diagnosis of children affected with DS was done on the basis of their karyotype report and/or clinical signs such as facial features, presence of mental retardation or congenital heart disease confirmed by echocardiography. In all cases diagnosis of DS was confirmed by the physician.

Lady using a tablet
Lady using a tablet

This Essay is

a Student's Work

Lady Using Tablet

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Examples of our work

Enrolment:

102 case-mothers and 104 control-mothers were enrolled in the study. The enrolment of case-mothers was done by visiting the wards and OPDs of B.J.W.H.C., as well as meeting the mothers at special schools. Out of a total of 102 case-mothers, 37 were enrolled from Sulbha School for mentally retarded children (36%), 33 at Centre for Research in Mental Retardation (CREMERE) (32%), 20 at BJWHC (20%), 7 at Saraswati Mandir School (7%), 4 at K.M.S. Shirodkar School (4%) and 1 at blood donation camp held in the premises of Larsen and Toubro Ltd. (1%). Out of 104 control-mothers, 19 were enrolled at the Institute of Chemical Technology (Mumbai), 17 at Larsen and Toubro (Mumbai); and the remaining at several residential societies: 17 in Borivali, 16 in Ghatkopar, 15 at Worli Dairy, 13 at Nagpada, 4 at Dahisar, 2 in Bandra and 1 in SEC school.

Sample collection & processing

8 ml sample of venous blood was withdrawn from each participant in two EDTA tubes. Genomic DNA was extracted from whole blood using salt-extraction method as well as using miniprep spin columns (MB504 HiPurA Blood Genomic DNA MiniPrep Purification Spin Kit-HiMedia, India).

Genotype analysis

Genotype analysis of the MTHFR gene was carried out by polymerase chain reaction in a heated lid thermal cycler (Progene-Techne, UK) followed by restriction enzyme digestion of the amplified product with Hinf I using conditions as previously established [Frosst et al., 1995]. For the genotype analysis of the BHMT SNP, PCR was carried out using the primers TGCTGGTTTCTGGTGCATCCCTAA and AAGGGCTGGCTCATCAGGTGAGCTTTGAGT in a heated lid thermal cycler (MyCycler™ - BIO - RAD, USA). The amplification protocol followed was a modification of the method described elsewhere wherein a final extension of 5min, 72°C was added [Ananth et al., 2007]. The amplified PCR product of 171-bp was digested overnight at 37oC with Hinf I. The presence of normal G allele established a restriction site, producing 141-bp and 30-bp fragments of whereas the mutant 'A' allele abolished the restriction site. The digest was sized on a 3% agarose gel with ethidium bromide and visualized under UV-light. Genotypes were recorded. The gel was photographed for record using a digital camera (Canon G5-Canon, Asia) and a UV proof hood. Genotype results were obtained for all case-mothers but 98 out of 104 control-mothers. This was because 4 control-mothers did not report for blood collection whereas DNA yield was poor for other 2 samples.

Statistical analysis

The allelic and genotypic frequencies were determined by direct count. Association of MTHFR and BHMT genotypes between case and control mothers was estimated by using an odds ratio with 95% confidence intervals. Mean and standard deviations were presented for the continuous variables. A 'p' value of 0.05 or less was considered to be significant statistically.

Results

The case- and control-mothers were matched for age and region of origin. They consisted of Maharashtrians, Gujaratis, North Indians and South Indians. The frequency of the variant 'T' allele of the MTHFR 677C>T SNP was 0.14 in control-mothers and 0.11 in case-mothers. The frequency of the 'A' allele of the BHMT 742G>A SNP was 0.28 in control-mothers and 0.30 in case-mothers. Genotype distribution followed Hardy-Weinberg equilibrium for MTHFR 677C>T and BHMT 742G>A SNP's.

The BMI for control-mothers was 25±4 kg/m2 (Mean±SD) and 24±5 kg/m2 for case-mothers (Mean±SD). The mean age of the case-mothers at the time of delivery of a child with DS was found to be 28±6 years (Mean±SD). In 38 cases (37%), the first born child had DS, whereas the second born child with DS was seen in 35 cases (34%). A family history of a child with DS was seen in 12 case-mothers (12%). Documents related to DS were obtained in 79 cases (77%). Karyotype reports were available for 45 cases (4 had mosaic DS whereas the rest had Trisomy 21). In all cases diagnosis of DS was confirmed by the physician. The gender of DS child was found to be male in 64 cases (62%) and female in 38 cases (38%). Premature delivery was seen in 14 cases (14%) whereas the birth of the DS child was found to be full term in 88 cases (86%). Consanguineous marriage was seen in 7 case-mothers (7%) and in 6 control-mothers (6%). History of fever during gestation was reported by 11% of case-mothers and in 4% of control-mothers. Hypertension was reported in 13 case-mothers (13%) and in 8 control-mothers (8%). Type II diabetes mellitus was reported in 3 case-mothers (3%) and in 5 control-mothers (5%). Nine case-mothers (9%) and 3 control-mothers (3%) reported on the habit of tobacco consumption (97%). Vitamin supplements were reported only in 2 case-mothers (2%) and in 10 control-mothers (10%). Calcium supplements were reported only in 7 case-mothers (7%) and in 23 control-mothers (22%).

Discussion

Made up of more than 4,500 culturally and anthropologically well defined populations with limited gene flow in-between, India represents the largest human diversity. India has a population of more than one billion citizens, coupled with a plethora of castes, sub castes and tribes, high degree of endogamy and consanguinity in various sects [Malini & Ramachandra, 2006].

Studies carried out in Indian population have reported relatively low frequency of T allele in Indians [Mukherjee et al., 2002; Devi et al., 2004; Kumar et al., 2005; Refsum et al., 2001; Nair et al., 2002; Angeline et al., 2004]. In the present study, the homozygous variant genotype (i.e. 'TT' genotype) was present in 2% control-mothers as against 1% case-mothers. In a study on North Indian mothers, the homozygous variant genotype was present in 6% of control-mothers and absent in case-mothers [Kohli, et al., 2003]. In the study on Gujarati women, more control-mothers (8%) had the homozygous variant genotype as compared to the case-mothers (4%) [Sheth et al., 2003] but contradictory results were reported in a study in women mostly from Eastern India where 8% of case-mothers and 1.2% of control-mothers had the homozygous variant genotype [Rai et al., 2006]. Another study carried out in South India also reported MTHFR as a risk factor with T allele frequency of 4% in case-mothers and 0% in control-mothers [Cyril et al., 2009]. In the present study, the frequency of the variant 'T' allele for the MTHFR 677C>T SNP was 0.11 (11%) for case-mothers and 0.14 (14%) for control-mothers. In the West Bengal study, slightly higher values were reported, with 0.15 (15%) and 0.16 (16%) for the case-mothers and control-mothers respectively [Dutta et al., 2007]. It is thought that in a country like India with low levels of folate, low frequency of T allele would be an offshoot of natural selection [Rai et al., 2006].

There is no data on Indian population with respect to the BHMT 742G>A SNP and maternal risk of DS. In the present study, the variant genotype ('AA' genotype) was similar for case-mothers (8%) and control-mothers (6%). The frequency of the variant 'A' allele for the BHMT 742G>A SNP was comparable in case-mothers (0.30) and control-mothers (0.28). The frequency of 'A' allele of BHMT 742G>A SNP in a study on the Indian population carried out at the Foods Department of ICT was 0.25 for controls [Mukherjee et al, 2009]. Hence the BHMT 742G>A SNP was not associated as a maternal risk factor of DS.

A known risk factor for DS is the age of the mother above 35 years (advanced maternal age). But, only 17% of the case-mothers were found to be over 35 years of age at the time of birth of the affected DS child. Similar low findings (8.4%) have been reported [Sheth et al., 2007]. Hence our findings do not support advanced maternal age as risk factor for birth of DS child. Also, maximum DS children were firstborns in the present study. Comparable results have been reported [Kaur et al., 2003].

Consanguinity has been reported to increase the risk (3%) of having children with congenital malformations [Sogaard & Vedsted - Jakobsen, 2003]. Also, it has been observed that there is an approximately four fold increase in DS occurrence (p < 0.005) in closely related parents [Alfi et al., 1980]. An increased DS risk in South India has also been linked with consanguinity [Malini & Ramachandra, 2007]. But in the present study, the frequency of consanguineous marriage was relatively low and comparable in both control- and case-mothers, indicating little or no association between consanguinity and DS. Similar findings related to consanguinity have also been reported [Hamamy et al., 1990; Basaran et al., 1992; Zlotogora, 1997].

A higher percentage of case-mothers showed history of fever compared to the control-mothers (11% versus 4%). Though the difference was not statistically significant, it could be said that there is a trend of history of high fever during pregnancy in the case-mothers. Thus, there might be a possibility that history of fever during pregnancy may be associated as a risk factor for DS.

In this study, an excess of males were found as compared to females. Higher sex ratio in individuals with DS compared to the general population has been observed [Isaac et al., 1985]. More number of males has been found among live - born and aborted conceptions with Trisomy 21 [Hassold et.al, 1983]. Very early selection against female conceptions with an extra chromosome 21 as compared to male conceptions could be the reason for the males among all clinically recognized trisomy 21 conceptions.

Even though the chosen polymorphisms did not appear to be linked as possible maternal risk factor of DS, the association of DNA polymorphisms at different loci in the same or other genes cannot be ruled out. Another thing to be considered is that there are a large number of ethnic groups in India, and a particular finding should not be used to draw any significant conclusion as the gene-pool is highly conserved. So we hereby recommend further studies with larger sample sizes with people of different geographic origin so as to make a better judgement on the association.

Conclusion

Our study did not support the association of the MTHFR 677C>T nor the BHMT 742G>A SNP's as maternal risk factors predisposing to babies with DS in the Indian population. DNA polymorphisms at different loci in the same or other genes may be associated.

Tables

Table 1 MTHFR genotypes and alleles in case-mothers and control-mothers

Genotype

Case-mothers

Control-mothers

CC, n (%)

82 (80)

73 (74.5)

CT, n (%)

19 (19)

22 (23.5)

TT, n (%)

1 (1)

2, (2)

'T' allele frequency

0.11

0.14

Table 2 BHMT genotypes and alleles in case-mothers and control-mothers

Genotype

Case-mothers

Control-mothers

GG, n (%)

50 (49)

42 (43)

GA, n (%)

44 (43)

50 (51)

AA, n (%)

8 (8)

6 (6)

'A' allele frequency

0.30

0.28