Sage Journals: Discover world-class research

Abstract

BACKGROUND:

Remethylation of homocysteine is catalyzed by B12 dependent methionine synthase (MTR) in all types of cells and by B12 non-dependent betaine homocysteine methyltransferase (BHMT) in liver and kidney cells. Of many etiologies of cancer, an unexplored area is the variations of genes implicated in methylation reaction.

OBJECTIVE:

The study evaluated the association of BHMT (rs3733890) with acute lymphoblastic leukemia (ALL), followed by in-silico characterization of variations in BHMT gene.

METHODS:

BHMT [rs3733890; c.742G $>$ A, which substitutes an arginine by a glutamine at codon 239 (R239Q)] was screened by Tetra-primer Amplification Refractory Mutation System PCR (T-ARMS-PCR) and confirmed using DNA sequencing. In-silico analysis was conducted using bioinformatics tools.

RESULTS:

BHMT (rs3733890) showed an insignificant association with both childhood and adult ALL. Bioinformatics analysis showed that 18 nsSNPs are deleterious, 3 SNPs in 3 ${}^{\prime}$ -UTR (rs59109725, rs116634518 and rs138578732) alter the miRNA-binding site, and 11 CNVs are present in the BHMT gene. As consequence of BHMT (rs3733890) polymorphism the free energy changes from $-$ 101210.1 kJ/mol to $-$ 200021.8 kJ/mol.

CONCLUSIONS:

BHMT (rs3733890) polymorphism showed no association with ALL. Hence this investigation needs further evaluation in larger sample size and effect of other SNPs, CNVs and miRNA’s is required to elucidate the role of BHMT gene in ALL development.

Keywords

Acute lymphoblastic leukemia BHMT gene T-ARMS-PCR SNPs

1. Introduction

Betaine-homocysteine methyltransferase (BHMT), one of the zinc-dependent cytosolic enzymes, is involved in the remethylation of homocysteine by B12 non-dependent manner especially in liver and kidney cells [1]. It has been predicted that BHMT protein can perform at least half of the total homocysteine remethylation reaction [2]. Though there were many pieces of evidence of elevated homocysteine in neural tube defect, cardiovascular disorders and cancers like acute lymphoblastic leukemia [3, 4, 5]; surprisingly there is little effort to evaluate the influence of BHMT gene variations on ALL. It has been observed that variations in other folate metabolism genes involved in the methylation reaction such as methylene tetrahydrofolate reductase (MTHFR), 5-methyltetrahydrofolate homocysteine methyltransferase (MTR) and 5-methylte- trahydrofolate homocysteine methyltransferase reductase (MTRR) can majorly contribute to the pathophysiology of ALL and other cancers [6, 7, 8, 9]. Scientific attention to exploring the possible role of BHMT in ALL development has been somewhat lacking. The human BHMT gene resides on chromosome 5q13.1-q15 [10]. BHMT gene has a common single nucleotide polymorphism (c.742G $>$ A; rs3733890), which substitutes an arginine by a glutamine at codon 239 (R239Q). It is one of the well-studied polymorphism located in the sixth exon of the gene with the evidence of its association with Down syndrome, neural tube defects (NTD), orofacial clefts, cervical cancer, breast cancer and head and neck cancer [11, 12, 13, 14, 15, 16]. To the best of our knowledge, no attempt has been made to study the association of this polymorphism with ALL.

ALL is a hematological malignancy with 80% estimates of total childhood leukemia and 20% of all adult leukemia [17, 18]. The knowledge about the most prominent etiology for ALL has been improved, but the molecular mechanism behind the disease development is not entirely known except for the role of ionizing radiations and chemicals like benzene [19, 20, 21]. It has been shown that the combination of many genetic variations and environmental elements can play a significant part in the ALL development as well as in the therapy [22, 23]. There has been an intense scientific focus on folate metabolism gene variations on ALL, but its unsustainable association with ALL in different populations [8, 9, 11] has impeded its establishment as a specific and prominent etiology for ALL. Furthermore, repeated analysis and new studies on population-specific genetic association may defiantly reveal the idea of the actual role of folate metabolism gene polymorphisms in the ALL development and therapy, which can be implemented to prevent and cure the disease in the future. Along with this the functional characterization of polymorphism by genetic expression analysis may add additional support for the disease management.

In India, existing data on ALL predisposing and drug response markers is limited and insufficient to draw a remarkable conclusion. Also, considering the heterogeneity in the populations, ALL predispositions and its therapy have always remained a challenge. Hence, it becomes necessary to evaluate the association of genetic polymorphisms with ALL. To decipher the associated risk of BHMT (rs3733890) with ALL, we conducted a case-control association analysis in South Indian population. An in-silico evaluation of other BHMT gene variants was also performed to predict the possible mechanisms of gene regulation.

2. Materials and methods

2.1 Case-control association analysis

2.1.1 Subjects

The study included 199 ALL individuals and 245 normal healthy individuals from South Indian population. The ALL was diagnosed through a morphological and immune phenotypic approach and subjects were characterized by World Health Organization (WHO) classification. The study protocols were approved by Kasturba Hospital institutional ethics committee, Manipal. The study conformed to The Code of Ethics of the World Medical Association. The written consent form was taken from both ALL individuals and normal healthy individuals. The written consent was acquired from parent or guardian in case of childhood ALL. Blood samples were collected from 125 childhood ALL individuals and 74 adults ALL individuals who were enrolled in the study from Kasturba Hospital, Manipal and Kasturba Hospital, Mangalore, Karnataka, India. The mean age of childhood and adult ALL participants were 6.57 years and 40.58 years respectively. Healthy normal subjects comprised of 191 young adults and 54 adults with a mean age 19.15 years and 47.78 years respectively.

Figure 1.

Bioinformatic analysis of nsSNPs in BHMT gene. (A) Percentage of deleterious and tolerated nsSNPs of BHMT gene predicted by 7 different bioinformatics tools. (B) Schematic representation of BHMT gene structure displaying the position of 18 deleterious nsSNPs along with rs3733890. Abbreviations: SIFT, Sorting Intolerant from Tolerant; PolyPhen-2, Polymorphism Phenotyping v2; nsSNPanalyzer, non-synonymous protein coding Single Nucleotide Polymorphisms analyzer; PROVEAN v1.1, Protein Variation Effect Analyzer; PhD-SNP; Predictor of human Deleterious Single Nucleotide Polymorphisms.

2.1.2 Genotyping of BHMT (rs3733890)

The genotyping of BHMT (rs3733890) in ALL individuals and normal healthy individuals were performed by tetra-primer amplification refractory mutation system-polymerase chain reaction (T-ARMS-PCR). The primers were designed using PRIMER1 web tool [24, 25]. The 25 $\mu$ L reaction mixture having 100 ng of genomic DNA, 2.5 $\mu$ L of all four 4 mM dNTPs, 2.5 $\mu$ L of 10X buffer, 1 $\mu$ L of 10 pm/ $\mu$ L each of outer forward primer (5 ${}^{\prime}$ -TGATGAAGAGAGGAG AGCTGGCCC-3 ${}^{\prime}$ ), outer reverse primer (5 ${}^{\prime}$ -GCAGGG ACAGAAACCAGGGACAAG-3 ${}^{\prime}$ ), inner forward pri- mer (5 ${}^{\prime}$ -AGGAGGGCTTGGAGGCTGCACA-3 ${}^{\prime}$ ) specific for mutant allele and inner reverse primer (5 ${}^{\prime}$ -CTGGCTCATCAGGTGAGCTTTCATTC-3 ${}^{\prime}$ ) specific for wild-type allele were used to amplify the target gDNA by using thermocycler. PCR was performed with initial denaturation of 95 ${}^{\circ}$ C for 5 minutes followed by 35 cycles of denaturation at 95 ${}^{\circ}$ C for 30 seconds, annealing at 62 ${}^{\circ}$ C for 40 seconds and extension at 72 ${}^{\circ}$ C for 30 seconds and a final extension of 72 ${}^{\circ}$ C for 10 minutes. PCR products were resolved on 1.5% agarose gel along with a 100 bp marker. Homozygotes profiles showed respectively, a 402 bp product of outer primers (control band) with either a 259 bp product for BHMT 742AA genotype or a 143 bp product for BHMT 742GG genotype (Supplementary Fig. 1). The heterozygote profile showed 3 PCR product bands (402, 259 and 143 bp) for BHMT 742AG genotype. The results were validated by gold standard DNA sequencing technology.

2.1.3 Validation of genotyping by Sanger assay

The genotyping results of case-control populations captured by the T-ARMS-PCR were confirmed by gold standard Sanger DNA sequencing method by using ABI Prism 3130 genetic analyzer automated DNA sequencer (Applied Biosystems, USA) using ABI Prism BigDye Terminator v3.1 cycle sequencing kit. A total of 3 samples with three different genotypes of BHMT (rs3733890) were validated, and the T-ARMS-PCR genotype results were confirmed. The results of Sanger DNA sequencing were analyzed by Finch TV tool (http://jblseqdat.bioc.cam.ac.uk/gnmweb/download/soft/FinchTV_1.4/doc/).

2.1.4 Statistical analysis

Table 1
Distribution of genotype and allele frequencies of BHMT (rs3733890) in individuals with childhood ALL and normal healthy individuals

SNP/rsID Genotype/Allele ALL Cases $n=$ 125 (%) Controls $n=$ 191 (%) OR (95% CI) ${}^{*}$ $p$ value ${}^{{\dagger}}$

rs3733890 742GG 64 (51.2) 100 (52.35) 1.00 (Reference) –

742GA 50 (40.0) 78 (40.84) 1.002 [0.62–1.60] 0.99

742AA 11 (8.80) 13 (6.81) 1.322 [0.55–3.13] 0.52

742GA $+$ AA 61 (48.8) 91 (47.65) 1.047 [0.66–1.64] 0.84

G allele 0.71 0.73 1.00 (Reference)

A allele 0.29 0.27 1.08 [0.75–1.54] 0.66

SNP/rsID	Genotype/Allele	ALL Cases $n=$ 125 (%)	Controls $n=$ 191 (%)	OR (95% CI) ${}^{*}$	$p$ value ${}^{{\dagger}}$
rs3733890	742GG	64 (51.2)	100 (52.35)	1.00 (Reference)	–
	742GA	50 (40.0)	78 (40.84)	1.002 [0.62–1.60]	0.99
	742AA	11 (8.80)	13 (6.81)	1.322 [0.55–3.13]	0.52
	742GA $+$ AA	61 (48.8)	91 (47.65)	1.047 [0.66–1.64]	0.84
	G allele	0.71	0.73	1.00 (Reference)
	A allele	0.29	0.27	1.08 [0.75–1.54]	0.66

BHMT, BetaineHomocysteineMethyltransferase; ALL, Acute Lymphoblastic Leukemia; SNP, Single Nucleotide Polymorphism; OR, odds ratio; CI, confidence interval; *OR and 95% CI were calculated with BHMT 742GG genotype as reference group. ${}^{{\dagger}}$ $p$ -value less than 0.05 is significant.

Table 2

Distribution of genotype and allele frequencies of BHMT (rs3733890) polymorphism in individuals with adult ALL and normal healthy individuals

SNP/rsID	Genotype/Allele	ALL cases $n=$ 74 (%)	Controls $n=$ 54 (%)	OR (95% CI) ${}^{*}$	$p$ value ${}^{{\dagger}}$
rs3733890	742GG	37 (50.0)	32 (59.26)	1.00 (Reference)	–
	742GA	28 (37.84)	18 (33.33)	1.345 [0.63–2.87]	0.44
	742AA	9 (12.16)	4 (7.41)	1.946 [0.54–6.92]	0.29
	742GA $+$ AA	37 (50.0)	22 (40.74)	1.455 [0.71–2.95]	0.29
	G allele	0.69	0.76	1.00 (Reference)
	A allele	0.31	0.24	0.70 [0.40–1.23]	0.21

Pearson’s Goodness of fit chi-square test was used to calculate Hardy–Weinberg equilibrium (HWE) for observed genotypes in normal healthy subjects and ALL individuals. The risk association of BHMT (rs3733890) with ALL individuals was estimated by calculating odds ratio (OR) and 95% CI through chi-square test using Graph Pad Intent 3. Software.

2.2 Bioinformatic survey of BHMT gene

2.2.1 Selection of SNPs for in silico analysis

For in silico analysis we have obtained all the information such as rsID, mRNA accession number, transcript ID, protein accession number, protein ID, protein sequences and Swiss-prot ID from National Centre for Biotechnology Information (NCBI), dbSNP database (http:/www.ncbi.nlm.nih.gov/projects/SNP) and Ense- mbl Genome Browser (http://asia.ensembl.org/index.html).

2.2.2 Prediction of nsSNPs, miRNASNPs, and CNVs by in-silico tools

In silico examination of non-synonymous SNPS (nsSNPs) were performed to predict the role of SNPs on protein structure and stability using bioinformatics tools such as Sorting Intolerant from Tolerant v. 1.03 (SIFT v. 1.03) [26], Polymorphism Phenotyping v2 (PolyPhen-2) [27], I-MUTANT 2.0 [28], Protein Variation Effect Analyzer (PROVEAN v1.1) [29], SNP&GO [30], nsSNPs protein coding single nucleo- tide polymorphisms analyzer (nsSNPanalyzer) [31], and Predictor of human Deleterious Single Nucleotide Polymorphisms (PhD-SNP) [32].

Furthermore, we analyzed the SNPs in the 3 ${}^{\prime}$ UTR regions for prediction of influence of SNPs on miRNA-binding efficiency (miRNA gain and loss) by using microRNA-related single nucleotide polymorphism V2 (miRNASNP V2) [33]: data updated based on miRBase 19 and dbSNP V137), database of polymorphisms altering miRNA target sites (mirSNP data-base) [34] and Polymorphisms in microRNAs and their target sites (PolymiRTS database 3.0) [35]. SNPs predicted to influence the miRNA binding by at least two tools out of three tools were considered for confirmation analysis by RNAhybrid [36]. The Copy number variations (CNVs) in the BHMT gene and its frequency were retrieved from DGV database using hg19 genome assembly [37].

Table 3
List of polymorphisms which alter the binding of miRNA in 3 ${}^{\prime}$ UTR region of BHMT gene and free energy change predicted using RNAhybrid

rsID/miRNA/MAF Allele Energy change miRNA/Target duplex

rs59109725 (hsa-miR-1283) Gain MAF:0.0008 C $-$ 14.8 kcal/mol target 5 ${}^{\prime}$ A UA GA A 3 ${}^{\prime}$ GAAGC UUUUU UG UUUCG AAAGG AC miRNA 3 ${}^{\prime}$ UC CG AA AUCU 5 ${}^{\prime}$

G $-$ 20.0 kcal/mol target 5 ${}^{\prime}$ A UA GAUGAA U 3 ${}^{\prime}$ GAAGC UUUUU UUUGUAGG UUUCG AAAGG AAACAUCU miRNA 3 ${}^{\prime}$ UC CG 5 ${}^{\prime}$

rs116634518 (hsa-miR-1243) Gain MAF:0.0032 A $-$ 12.8 kcal/mol target 5 ${}^{\prime}$ A A C A 3 ${}^{\prime}$ UCC AUA CCAG AGG UAU GGUC miRNA 3 ${}^{\prime}$ GUG A UAACUA AA 5 ${}^{\prime}$

G $-$ 15.0 kcal/mol target 5 ${}^{\prime}$ A AU A 3 ${}^{\prime}$ AUAA AUCCAGU UAUU UAGGUCA miRNA 3 ${}^{\prime}$ GUGAGGA AAC A 5 ${}^{\prime}$

rs116634518 (hsa-miR-1225-5p) Gain MAF:0.0032 A $-$ 20.6 kcal/mol target 5 ${}^{\prime}$ U AAUA GA UC CU G 3 ${}^{\prime}$ CC CCCA G CUG CUCA GG GGGU C GGC GGGU miRNA 3 ${}^{\prime}$ G GA CC AU G 5 ${}^{\prime}$

G $-$ 22.0 kcal/mol target 5 ${}^{\prime}$ A GA AUAAAUAU A G 3 ${}^{\prime}$ CCU UCA CC GUACCCA GGG GGU GG CAUGGGU miRNA 3 ${}^{\prime}$ G GACCC G 5 ${}^{\prime}$

rs138578732 (hsa-miR-369-3p) Loss MAF: 0.0002 G $-$ 10.8 kcal/mol target 5 ${}^{\prime}$ A G 3 ${}^{\prime}$ GUAUUAU CAUAAUA miRNA 3 ${}^{\prime}$ UUUCUAGUUGGUA A 5 ${}^{\prime}$

A $-$ 9.2 kcal/mol target 5 ${}^{\prime}$ U AA CU G 3 ${}^{\prime}$ GGA UCA GCU UCU AGU UGG miRNA 3 ${}^{\prime}$ UU UACAUAAUAA 5 ${}^{\prime}$

rsID/miRNA/MAF	Allele	Energy change	miRNA/Target duplex
rs59109725 (hsa-miR-1283) Gain MAF:0.0008	C	$-$ 14.8 kcal/mol	target 5 ${}^{\prime}$ A UA GA A 3 ${}^{\prime}$ GAAGC UUUUU UG UUUCG AAAGG AC miRNA 3 ${}^{\prime}$ UC CG AA AUCU 5 ${}^{\prime}$
	G	$-$ 20.0 kcal/mol	target 5 ${}^{\prime}$ A UA GAUGAA U 3 ${}^{\prime}$ GAAGC UUUUU UUUGUAGG UUUCG AAAGG AAACAUCU miRNA 3 ${}^{\prime}$ UC CG 5 ${}^{\prime}$
rs116634518 (hsa-miR-1243) Gain MAF:0.0032	A	$-$ 12.8 kcal/mol	target 5 ${}^{\prime}$ A A C A 3 ${}^{\prime}$ UCC AUA CCAG AGG UAU GGUC miRNA 3 ${}^{\prime}$ GUG A UAACUA AA 5 ${}^{\prime}$
	G	$-$ 15.0 kcal/mol	target 5 ${}^{\prime}$ A AU A 3 ${}^{\prime}$ AUAA AUCCAGU UAUU UAGGUCA miRNA 3 ${}^{\prime}$ GUGAGGA AAC A 5 ${}^{\prime}$
rs116634518 (hsa-miR-1225-5p) Gain MAF:0.0032	A	$-$ 20.6 kcal/mol	target 5 ${}^{\prime}$ U AAUA GA UC CU G 3 ${}^{\prime}$ CC CCCA G CUG CUCA GG GGGU C GGC GGGU miRNA 3 ${}^{\prime}$ G GA CC AU G 5 ${}^{\prime}$
	G	$-$ 22.0 kcal/mol	target 5 ${}^{\prime}$ A GA AUAAAUAU A G 3 ${}^{\prime}$ CCU UCA CC GUACCCA GGG GGU GG CAUGGGU miRNA 3 ${}^{\prime}$ G GACCC G 5 ${}^{\prime}$
rs138578732 (hsa-miR-369-3p) Loss MAF: 0.0002	G	$-$ 10.8 kcal/mol	target 5 ${}^{\prime}$ A G 3 ${}^{\prime}$ GUAUUAU CAUAAUA miRNA 3 ${}^{\prime}$ UUUCUAGUUGGUA A 5 ${}^{\prime}$
	A	$-$ 9.2 kcal/mol	target 5 ${}^{\prime}$ U AA CU G 3 ${}^{\prime}$ GGA UCA GCU UCU AGU UGG miRNA 3 ${}^{\prime}$ UU UACAUAAUAA 5 ${}^{\prime}$

MAF, Minor Allele Frequency.

2.2.3 Modeling of the mutant protein structure

The available three-dimensional (3D) structure of BHMT protein (4M3P) was retrieved from PDB (Protein Data Bank) and modeled for mutation at 239 position from arginine to glutamine (rs3733890; MAF: A $=$ 0.2907) by using PyMOL 1.3 and RMSD (root mean square deviation) was calculated using SuperPose version 1.0 [38].

2.2.4 Bioinformatic network analysis

A network analysis to evaluate the closely related functional genes of BHMT were performed by using STRING 10.0 a functional protein association database by using seven different active prediction methods which include neighborhood, co-occurrence, co-expression, gene fusion, databases, experiments and text mining [39].

3. Results

3.1 BHMT (rs3733890) association analyses

The validation of genotypes by Sanger DNA sequencing showed 100% similarity with the genotype results by T-ARMS-PCR technique. The allele frequencies and genotype frequencies of normal healthy individuals, childhood ALL individuals and adult ALL individuals are represented in Tables 1 and 2. The genotype frequency of BHMT (rs3733890) followed HWE in childhood ALL individuals, adults ALL individuals and normal healthy individuals. BHMT (rs3733890) showed an insignificant association with ALL in South Indian population (Tables 1 and 2).

3.2 Bioinformatic analysis

3.2.1 Non-synonymous SNPs prediction

Table 4
Details of copy number variations in BHMT gene from DGV database

Accession number Location information Variant type Sample size Observed gain Observed loss

1 esv2762527 chr5:77921199..78548196 CNV gain 1109 1 0

2 nsv519156 chr5:78351512..78468004 CNV gain 2026 1 0

3 nsv1022839 chr5:78352894..78507368 CNV gain 29084 1 0

4 nsv1019727 chr5:78401434..78445459 CNV gain 29084 1 0

5 nsv525398 chr5:78410564..78413504 CNV gain 2026 1 0

6 esv1010320 chr5:78425473..78426966 CNV gain 3 1 1

7 nsv4892 chr5:78425901..78446715 CNV insertion 9 1 0

8 esv3442711 chr5:78426074..78426982 CNV duplication 185 1 0

9 esv3308186 chr5:78426519..78426580 CNV mobile element insertion 185 2 0

10 esv3408955 chr5:78426536..78426567 CNV insertion 185 5 0

11 nsv512853 chr5:78426528..78426753 CNV insertion 1 1 0

	Accession number	Location information	Variant type	Sample size	Observed gain	Observed loss
1	esv2762527	chr5:77921199..78548196	CNV gain	1109	1	0
2	nsv519156	chr5:78351512..78468004	CNV gain	2026	1	0
3	nsv1022839	chr5:78352894..78507368	CNV gain	29084	1	0
4	nsv1019727	chr5:78401434..78445459	CNV gain	29084	1	0
5	nsv525398	chr5:78410564..78413504	CNV gain	2026	1	0
6	esv1010320	chr5:78425473..78426966	CNV gain	3	1	1
7	nsv4892	chr5:78425901..78446715	CNV insertion	9	1	0
8	esv3442711	chr5:78426074..78426982	CNV duplication	185	1	0
9	esv3308186	chr5:78426519..78426580	CNV mobile element insertion	185	2	0
10	esv3408955	chr5:78426536..78426567	CNV insertion	185	5	0
11	nsv512853	chr5:78426528..78426753	CNV insertion	1	1	0

Note: nsv (NCBI; dbVar) andesv (EBI; DGVa) are accession IDs assigned to structural variations.

We observed 174 nsSNPs in BHMT gene from Ensembl database (http://asia.ensembl.org/Homo_sapiens/Transcript/Variation_Transcript/Table?db=core;g=ENSG00000145692;r=5:79111779-79132290;t=ENST00000274353). The prediction of effect of nsSNPs revealed that 98 nsSNPs from SIFT analysis, 34 nsSNPs from PolyPhen 2 analysis, 124 nsSNPs from I-Mutant analysis, 77 nsSNPs from nsSNP analyzer analysis, 99 nsSNPs from SNP&GO analysis, 87 nsSNPs from PROVEAN analysis and 64 SNPs in PhD-SNP were deleterious for BHMT protein (Supplementary Table 1). Out of 174 nsSNPs, only 18 polymorphisms were commonly predicted as deleterious for BHMT protein by these seven different tools (Fig. 1).

3.2.2 3

{}^{\prime}

-UTR SNPs and miRNA-binding

Figure 2.

Modeled structures (cartoon shape) of BHMT (rs3733890). Native (A), Mutant (B) and Superimposition of native and mutant (C) using PyMOL 1.3.

The database search by using Ensembl showed 57 SNPs in 3 ${}^{\prime}$ -UTR of the BHMT gene (supplementary Table 2). The probable gain and loss of miRNA-binding sites due to SNPs in 3 ${}^{\prime}$ -UTR were investigated by three different databases namely, miRNASNP V2, mirSNP database and PolymiRTS database 3.0 (supplementary Table 3) and re-confirmed by RNAhybrid tool showed that here are two SNPs which led to gain of three miRNA binding sites (rs59109725 [hsa-miR-1283] and rs116634518 [hsa-miR-1243 and hsa-miR-1225-5p]) and there was one SNP which accounts for a loss of miRNA binding site (rs138578732 [hsa-miR-369-3p]). However, the minor allele frequency of rs59109725 (MAF: G $=$ 0.0008), rs116634518 (MAF: G $=$ 0.0032) and rs138578732 (MAF: A $=$ 0.0002) was very less. Hence we believe none of these SNPs play a major role in the population (Table 3).

3.2.3 Copy number variations

Analysis by using DGV and UCSC databases revealed that BHMT gene has 11 copy number variations (CNVs) (Table 4). A total of 6 copy number gains (esv2762527, nsv519156, nsv1022839, nsv1019727, nsv525398, esv1010320) were observed. Further, there were four insertions (nsv4892, esv3308186, esv34089 55 and nsv512853) and one duplication (esv3442711) (Supplementary Fig. 2).

3.2.4 Protein modeling

The impact of nsSNP rs3733890 on BHMT protein was assessed by performing mutagenesis on BHMT protein structure at position 239, where glutamine was introduced in place of arginine as a consequence of rs3733890 (Fig. 2) using PyMOL. It was observed that the total energy of the native and mutant type structure is $-$ 101210.1 kJ/mol and $-$ 200021.8 kJ/mol, respectively. The modeled native (wild type) (Fig. 2A), mutant (Fig. 2B) and superimposed structures (Fig. 2C) were having RMSD 0.4Å analyzed from Superpose Version 1.0.

3.2.5 Bioinformatic network analysis

The network analysis of BHMT gene by STRING 10.0 web tool showed eleven functional partners Cys- tathionine-beta-synthase [CBS], Methylene tetrahydrofolatereductase [MTHFR], Methionine adenosyltransferase 1 A [MATIA], Adenosylhomocysteinase [AHCY], Methionine adenosyltransferase 2 A [MAT IIA], Dimethylglycinedehydrogenase [DMGDH], Sarcosinedehydrogenase [SARDH], Adenosylhomocysteinaselike 2 [AHCYL2], Adenosylhomocysteinaselike 1 [AHCYL1], Cystathionine Gamma-Lyase [CTH], and Methionine Adenosyltransferase 2B [MATIIB]) of BHMT gene were observed and its confidence score is mentioned in Fig. 3 and Table 5. The BHMT had the highest confidence score of 0.981 with CBS and lowest confidence score of 0.937 with MTHFR.

4. Discussion

Table 5
Confidence score of interactions between BHMT and its closely related functional genes by STRING 10.0 tool

MTHFR MATIIA MATIA DMGDH CTH CBS BHMT AHCYL2 AHCYL1 AHCY MATIIB

MTHFR – 0.724 0.789 0.674 0.851 0.895 0.937 0.599 0.577 0.836 –

MATIIA 0.724 – 0.651 – 0.691 0.687 0.963 0.929 0.921 0.958 0.999

MATIA 0.789 0.651 – 0.468 0.8 0.744 0.967 0.928 0.919 0.964 0.461

DMGDH 0.674 – 0.468 – 0.493 0.429 0.976 – – 0.577 –

CTH 0.851 0.691 0.8 0.493 – 0.997 0.948 0.888 0.882 0.942 –

CBS 0.895 0.687 0.744 0.429 0.997 – 0.981 0.94 0.937 0.976 –

BHMT 0.937 0.963 0.967 0.976 0.948 0.981 – 0.946 0.947 0.98 0.942

AHCYL2 0.599 0.929 0.928 – 0.888 0.94 0.946 – 0.571 – 0.478

AHCYL1 0.577 0.921 0.919 – 0.882 0.937 0.947 0.571 – – 0.601

AHCY 0.836 0.958 0.964 0.577 0.942 0.976 0.98 – – – –

MATIIB – 0.999 0.461 – – – 0.942 0.478 0.601 – –

	MTHFR	MATIIA	MATIA	DMGDH	CTH	CBS	BHMT	AHCYL2	AHCYL1	AHCY	MATIIB
MTHFR	–	0.724	0.789	0.674	0.851	0.895	0.937	0.599	0.577	0.836	–
MATIIA	0.724	–	0.651	–	0.691	0.687	0.963	0.929	0.921	0.958	0.999
MATIA	0.789	0.651	–	0.468	0.8	0.744	0.967	0.928	0.919	0.964	0.461
DMGDH	0.674	–	0.468	–	0.493	0.429	0.976	–	–	0.577	–
CTH	0.851	0.691	0.8	0.493	–	0.997	0.948	0.888	0.882	0.942	–
CBS	0.895	0.687	0.744	0.429	0.997	–	0.981	0.94	0.937	0.976	–
BHMT	0.937	0.963	0.967	0.976	0.948	0.981	–	0.946	0.947	0.98	0.942
AHCYL2	0.599	0.929	0.928	–	0.888	0.94	0.946	–	0.571	–	0.478
AHCYL1	0.577	0.921	0.919	–	0.882	0.937	0.947	0.571	–	–	0.601
AHCY	0.836	0.958	0.964	0.577	0.942	0.976	0.98	–	–	–	–
MATIIB	–	0.999	0.461	–	–	–	0.942	0.478	0.601	–	–

Abbreviations: CBS, Cystathionine-beta-synthase; MTHFR, Methylenetetrahydrofolatereductase; MATIA, Methionine Adenosyltransferase 1A; AHCY, Adenosylhomocysteinase; MATIIA, Methionine Adenosyltransferase 2A; DMGDH, Dimethylglycine Dehydrogenase; SARDH, Sarcosine Dehydrogenase; AHCYL2, Adenosylhomocysteinase Like 2; AHCYL1, Adenosylhomocysteinase Like 1; CTH, Cystathionine Gamma-Lyase; MATIIB, Methionine Adenosyltransferase 2B. “–”: represents no interaction.

Figure 3.

Network analysis of BHMT gene by STRING 10.0 tool. The network image represents the interaction between BHMT protein with closely related protein. Abbreviations: cystathionine-beta-synthase [CBS], Methylenetetrahydrofolatereductase [MTHFR], Methionine Adenosyltransferase1A [MATIA], Adenosylhomocysteinase [AHCY], Methionine Adenosyltransferase2A [MATIIA], Dimethylglycine Dehydrogenase [DMGDH], Sarcosine Dehydrogenase [SARDH], Adenosylhomocysteinase Like 2 [AHCYL2], AdenosylhomocysteinaseLike 1 [AHCYL1], Cystathionine Gamma-Lyase [CTH], and Methionine Adenosyltransferase 2B [MATIIB].

The folate and choline metabolisms are closely coupled pathways for homocysteine remethylation resulting in cellular methionine production. The remethylation of homocysteine to methionine by using 5-methyltetrahydrofolate as a methyl donor is catalyzed by methionine synthase and remethylation reaction by using betaine as a methyl donor is catalyzed by zinc-dependent cytosolic betaine-homocysteine methyltr- ansferase. However, the less availability of 5-methy- ltetrahydrofolate and betaine can cause the abnormal DNA methylation due to an imbalance in the development of SAM (S-adenosylmethionine) from methionine. There is no consistent association of polymorphisms in genes of folate metabolism including 5-methyltetrahydrofolate homocysteine methyltransferase (MTR) rs1805087 and methylene tetrahydrofolate reductase gene (MTHFR) rs1801133, rs1801131 in the ALL individuals [40, 41, 42, 43]. Hence, it is mandatory to focus on other genes involved in the homocysteine remethylation to draw a conclusion for the link between the ALL development and homocysteine remethylation.

The observations of the present study include an insignificant association of BHMT (rs3733890) with the risk of ALL in South Indian population and effect of other variations in the BHMT gene (including SNPs in the coding region, 3 ${}^{\prime}$ UTR region, and CNVs) through computational analysis.

In recent past, few evidence of association of BHMT (rs3733890) with few other cancers has been highlighted. In the case of uterine cervical cancer, BHMT(rs3733890) polymorphism acts as protective fac-tor [14], in the event of breast cancer this polymorphism is associated with reduced mortality [15], and in the event of head and neck cancer this polymorphism along with tobacco consumption increases the risk of head and neck squamous cell carcinoma [16]. BHMT (rs3733890) perhaps has also shown association with many other diseases. In the case of Down’s syndrome BHMT 742AA genotype reduces the risk in Brazilian population [11], in the case of neural tube defect of Caucasian American population the rs3733890 polymorphism acts as a significant risk factor [12], in regard to orofacial clefts of Polish population this polymorphism works in a protective fashion [13]. The pattern of association of this rs3733890 polymorphism conceivably different for different diseases but however it has been focused that the choline supplement can reduce the mortality of breast cancer thus it explored a hopeful strategy to adopt new lifestyle after cancer is diagnosed [15]. In our population, both MTHFR (rs1801133 and rs1801131) and BHMT (rs3733890) gene polymorphisms showed insignificant association with ALL [43]. This outcome may also be attributed to small sample size and distribution of minor allele in the study population.

Further, it may be necessary to evaluate the combined effect, and different haplotypes of pathogenic nsSNPs reported to be associated with the ALL and other cancers. The expansion of informatics into the field of biology has opened the windows for much graphical stimulation, one of which includes the prediction of disturbances in protein stability or structure caused due to changes in the sequence. Since the association analysis of the present study failed to reveal a significant association of BHMT (rs3733890) with ALL, we performed in-silico search for other non-synonymous SNPs and SNPs in 3 ${}^{\prime}$ UTR region was conducted which highlighted miR-1283 binding at BHMT (rs59109725). Besides SNP analysis, 11 CNVs were present in BHMT gene which also needs to be evaluated for association with ALL along with the close interacting MTHFR, MATIIA, MATIA, DMGDH, CTH, CBS, BHMT, AHCYL2, AHCYL1, AHCY genes identified through our STRING analysis. The frequency of these variants is not known in ALL individuals. Hence these are important loci to take it forward for functional characterizations. Although BHMT (rs3733890) was predicted as tolerated, there was a change in total energy for the native and mutant type structure from $-$ 101210.1 kJ/mol to $-$ 200021.8 kJ/mol which supports the importance of future evaluation of BHMT (rs3733890) polymorphism in the larger cohort.

5. Conclusion

In summary, BHMT (rs3733890) showed insignificant association with ALL in South Indian population and the molecular mechanisms underlying the modulation in folate levels and ALL development in South Indian population is still unresolved. We further propose that the effective SNPs including rs59109725, rs116634518, rs138578732 and 11 CNVs which are present BHMT gene needs to be characterized in an attempt to better interpret the molecular mechanism of folate and cobalamin-dependent association of ALL for effective treatment and diagnosis in the future.

Supplementary data

The supplementary files are available to download from http://dx.doi.org/10.3233/CBM-160186.

Footnotes

Acknowledgments

This study was supported by Department of Biotechnology (DBT), Indian Council of Medical Research (ICMR), Technology Information Forecasting and Assessment Council-Centre of Relevance and Excellence (TIFAC-CORE) in Pharmacogenomics under Department of Science and Technology (DST), Government of India and Dr TMA Pai Endowment Chair for Pharmacogenomics, Manipal University.

References

Shinohara

Kuhlenkamp

Chan

and Kaplowitz

, Mechanisms of protection by the betaine-homocysteinemethyltransferase/betaine system in HepG2 cells and primary mouse hepatocytes, Hepatology 46 (2007), 1586–1596.

Finkelstein

J.D.

and Martin

J.J.

, Methionine metabolism in mammals, distribution of homocysteine between competing pathways, J BiolChem 259 (1984), 9508–9513.

Selhub

, Public health significance of elevated homocysteine, Food Nutr Bull 29(2) (2008), S116–125.

Nygard

Vollset

S.E.

Refsum

et al., Total plasma homocysteine and cardiovascular risk profile, The hordalandhomocysteine study, JAMA 274 (1995), 1526–1533.

Ruud

Holmstrom

Brosstad

and Wesenberg

, Children with acute lymphoblastic leukaemia have high plasma levels of total homocysteine at time of diagnosis, Scand J Clin Lab Invest 66 (2006), 67–78.

Wang

Meng

Zhao

Wang

Liu

and Mi

, Methylene tetrahydrofolatereductase polymorphism C677T is a protective factor for pediatric acute lymphoblastic leukemia in the Chinese population: A meta-analysis, Genet Test Mol Biomarkers 16 (2012), 1401–1407.

Wang

Zhao

Liu

and Mi

, Methylene tetrahydrofolatereductase polymorphisms and risk of acute lymphoblastic leukemia-evidence from an updated meta-analysis including 35 studies, BMC Med Genet 13 (2012), 77.

Zhang

Shi

Liu

Sturgis

E.M.

Spitz

M.R.

and 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.

Jokic

Brcic-Kostic

Stefulj

et al., Association of MTHFR, MTR, MTRR, RFC1, and DHFR gene polymorphisms with susceptibility to sporadic colon cancer, DNA Cell Biol 30 (2011), 771–776.

10.

Sunden

S.L.

Renduchintala

M.S.

Park

E.I.

Miklasz

S.D.

and Garrow

T.A.

, Betaine-homocysteinemethyltransferase expression in porcine and human tissues and chromosomal localization of the human gene, Arch BiochemBiophys 345 (1997), 171–174.

11.

Zampieri

B.L.

Biselli

J.M.

Goloni-Bertollo

E.M.

and Pavarino

E.C.

, BHMT G742A and MTHFD1 G1958A polymorphisms and Down syndrome risk in the Brazilian population, Genet Test Mol Biomarkers 16 (2012), 628–631.

12.

Boyles

A.L.

Billups

A.V.

Deak

K.L.

et al., Neural tube defects and folate pathway genes: Family-based association tests of gene-gene and gene-environment interactions, Environ Health Perspect 114 (2006), 1547–1552.

13.

Mostowska

Hozyasz

K.K.

Wojcicki

Dziegelewska

and Jagodzinski

P.P.

, Associations of folate and choline metabolism gene polymorphisms with orofacial clefts, J Med Genet 47 (2010), 809–815.

14.

Mostowska

Myka

Lianeri

Roszak

and Jagodzinski

P.P.

, Folate and choline metabolism gene variants and development of uterine cervical carcinoma, ClinBiochem 44 (2011), 596–600.

15.

Gammon

M.D.

Zeisel

S.H.

et al., High intakes of choline and betaine reduce breast cancer mortality in a population-based study, FASEB J 23 (2009), 4022–4028.

16.

da Silva

L.M.

Galbiatti

A.L.

Ruiz

M.T.

al.,

, MolBiol Rep 39 (2012), 887–893.

17.

Pathak

Hess

and Weiss

W.A.

, Liposomal vincristine for relapsed or refractory Ph-negative acute lymphoblastic leukemia: A review of literature, Ther Adv Hematol 5 (2014), 18–24.

18.

Belson

Kingsley

and Holmes

, Risk factors for acute leukemia in children: A review, Environ Health Perspect 115 (2007), 138–145.

19.

Rushton

and Romaniuk

, A case-control study to investigate the risk of leukaemia associated with exposure to benzene in petroleum marketing and distribution workers in the United Kingdom, Occup Environ Med 54 (1997), 152–166.

20.

Schnatter

A.R.

Armstrong

T.W.

Nicolich

M.J.

et al., Lymphohaematopoietic malignancies and quantitative estimates of exposure to benzene in Canadian petroleum distribution workers, Occup Environ Med 53 (1996), 773–781.

21.

Buffler

P.A.

Kwan

M.L.

Reynolds

and Urayama

K.Y.

, Environmental and genetic risk factors for childhood leukemia: Appraising the evidence, Cancer Invest 23 (2005), 60–75.

22.

deJonge

Tissing

W.J.

Hooijberg

J.H.

et al., Polymorphisms in folate-related genes and risk of pediatric acute lymphoblastic leukemia, Blood 113 (2009), 2284–2289.

23.

Wall

A.M.

and Rubnitz

J.E.

, Pharmacogenomic effects on therapy for acute lymphoblastic leukemia in children, Pharmacogenomics J3 (2003), 128–135.

24.

Dhillon

Collins

A.R.

and Day

I.N.

, An efficient procedure for genotyping single nucleotide polymorphisms, Nucleic Acids Res 29 (2001), E88–88.

25.

Medrano

R.F.

and de Oliveira

C.A.

, Guidelines for the tetra-primer ARMS-PCR technique development, MolBiotechnol 56 (2014), 599–608.

26.

Kumar

Henikoff

and Ng

P.C.

, Predicting the effects of coding non-synonymous variants on protein function using the SIFT algorithm, Nat Protoc 7 (2009), 1073–1081.

27.

Adzhubei

I.A.

Schmidt

Peshkin

et al., A method and server for predicting damaging missense mutations, Nat Methods 7 (2010), 248–249.

28.

Capriotti

Fariselli

and Casadio

, I-Mutant 2.0: Predicting stability changes upon mutation from the protein sequence or structure, Nucleic Acids Res 33 (2005), W306–310.

29.

Choi

and Chan

A.P.

, PROVEAN web server: A tool to predict the functional effect of amino acid substitutions and indels, Bioinformatics (2015), btv195.

30.

Calabrese

Capriotti

Fariselli

et al., Functional annotations improve the predictive score of human disease-related mutations in proteins, Hum Mutat 30 (2009), 1237–1244.

31.

Bao

Zhou

and Cui

, nsSNPAnalyzer: Identifying disease-associated nonsynonymous single nucleotide polymorphisms, Nucleic Acids Res 33 (2005), W480–482.

32.

Capriotti

Calabrese

and Casadio

, Predicting the insurgence of human genetic diseases associated to single point protein mutations with support vector machines and evolutionary information, Bioinformatics 22 (2006), 2729–2734.

33.

Gong

Tong

Zhang

H.M.

et al., Genome-wide identification of SNPs in microRNA genes and the SNP effects on microRNA target binding and biogenesis, Hum Mutat 33 (2012), 254–263.

34.

Liu

Zhang

et al., MirSNP, a database of polymorphisms altering miRNA target sites, identifies miRNA-related SNPs in GWAS SNPs and eQTLs, BMC Genomics 13 (2012), 661.

35.

Bhattacharya

Ziebarth

J.D.

and Cui

, PolymiRTS Database 3.0: Linking polymorphisms in microRNAs and their target sites with human diseases and biological pathways, Nucleic Acids Res 42 (2014), D86–91.

36.

Rehmsmeier

Steffen

Hochsmann

and Giegerich

, Fast and effective prediction of microRNA/target duplexes, RNA 10(10) (Oct 2004), 1507–1517.

37.

MacDonald

J.R.

Ziman

Yuen

R.K.

et al., The database of genomic variants: A curated collection of structural variation in the human genome, Nucleic Acids Res 42 (2014), D986–992.

38.

Maiti

Van Domselaar

G.H.

Zhang

et al., SuperPose: A simple server for sophisticated structural superposition, Nucleic Acids Res 32(2) (2004), W590–594.

39.

Szklarczyk

Franceschini

Wyder

et al., STRING v10: Protein-protein interaction networks, integrated over the tree of life, Nucleic Acids Res (2014), gku1003.

40.

Nikbakht

MalekZadeh

Jha

A.K.

et al., Polymorphisms of MTHFR and MTR genes are not related to susceptibility to childhood ALL in North India, Exp Oncol 34 (2012), 43–48.

41.

Xia

Wang

Zhang

and Hu

, Association between MTR A2756G polymorphism and childhood acute lymphoblastic leukemia: A meta-analysis, Leuk Lymphoma 55 (2014), 1388–1393.

42.

Jiang

Hou

Zhang

et al., The MTHFR C677T polymorphism and risk of acute lymphoblastic leukemia: An updated meta-analysis based on 37 case-control studies, Asian Pac J Cancer Prev 14 (2013), 6357–6362.

43.

Bellampalli

Phani

N.M.

Bhat

K.G.

et al., Significance of 5, 10-methylenetetrahydrofolate reductase gene variants in acute lymphoblastic leukemia in Indian population: An experimental, computational and Meta-analysis, Leuk Lymphoma 56 (2015), 1450–1459.

Acute lymphoblastic leukemia and genetic variations in BHMT gene: Case-control study and computational characterization

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

2. Materials and methods

2.1 Case-control association analysis

2.1.1 Subjects

2.1.3 Validation of genotyping by Sanger assay

2.1.4 Statistical analysis

2.2.1 Selection of SNPs for in silico analysis

2.2.2 Prediction of nsSNPs, miRNASNPs, and CNVs by in-silico tools

2.2.4 Bioinformatic network analysis

3. Results

3.1 BHMT (rs3733890) association analyses

3.2 Bioinformatic analysis

3.2.1 Non-synonymous SNPs prediction

3.2.4 Protein modeling

3.2.5 Bioinformatic network analysis

4. Discussion

Supplementary data

Footnotes

Acknowledgments

References