Received: 21 June 2017 Revised: 17 September 2017 Accepted: 27 September 2017

DOI: 10.1002/jgm.2990

RESEARCH ARTICLE

Association between MTHFR microRNA binding site

polymorphisms and methotrexate concentrations in Chinese
pediatric patients with acute lymphoblastic leukemia
Shu‐Mei Wang1 | Wei‐Xin Zeng1 | Wan‐Shui Wu2 | Lu‐Lu Sun1 | Dan Yan1

1
Department of Pharmacy, Beijing Shijitan
Hospital, Capital Medical University, Beijing, Abstract
China Background: The pharmacokinetics and therapeutic response to methotrexate (MTX) display
2
Department of Pediatrics, Beijing Shijitan
large variability in the treatment of acute lymphoblastic leukemia (ALL). The aim of the present
Hospital, Capital Medical University, Beijing,
China
study was to investigate the association of two microRNA (miRNA) binding site polymorphisms

Correspondence
Shu‐Mei Wang and Dan Yan, Department of
Pharmacy, Beijing Shijitan Hospital, Capital
Medical University, No 10 Tieyi Road,
Yangfangdian, Haidian District, Beijing
100038, China.
Email: wangshumei1980@126.com;
yd277_1@126.com
Funding information
Beijing Key Laboratory of Bio‐characteristic
Profiling for Evaluation of Rational Drug Use,
Grant/Award Number: BZ0439; Beijing
Municipal Administration of Hospitals’ Youth
Programme, Grant/Award Number:
QML20160703; National Natural Science
Foundation of China, Grant/Award Number:
81503135
(rs3737966 G > A and rs35134728 DEL/TTC) in the 3′‐untranslated region of MTHFR with serum
MTX concentrations, in a Chinese pediatric population with ALL.
Methods: Genotyping for MTHFR rs3737966 and rs35134728 in 144 children with ALL was
performed using the Sequenom MassArray system (Sequenom, San Diego, CA, USA). Serum MTX
concentrations were measured by a fluorescence polarization immunoassay 24 h (C24h) and 42 h
(C42h) after administration. The effects of the polymorphisms on concentration‐to‐dose (C/D)
ratios of MTX were assessed.
Results: Complete linkage disequilibrium between rs3737966 and rs35134728 poly-
morphisms (r2 = 1) was found in the study population. The minor allele frequency observed in
the present study (17.4%) was significantly lower than those in European and African samples
reported in the 1000 Genomes Project (42.9% and 63.9%, respectively; p < 0.01). The C/D ratios
of MTX at 24 and 42 h for the TTC/TTC‐A/A haplotype carriers (11.74 and 0.07 μmol/l per g/
m2, respectively) were significantly lower than those in DEL/DEL‐G/G or DEL/TTC‐G/G
haplotype carriers (12.49 and 0.09 μmol/l per g/m2, respectively; p < 0.05). Computational
predictions suggested that the two polymorphisms overlapped with putative binding sites of
several miRNAs.

Conclusions: The rs3737966 and rs35134728 polymorphisms in MTHFR were associated

with serum MTX concentrations. The findings of the present study indicate that miRNAs might
be involved in the post‐transcriptional regulation of MTHFR.

KEY W ORDS

acute lymphoblastic leukemia, genetic polymorphism, methotrexate, methylenetetrahydrofolate

reductase, microRNA

1 | I N T RO D U CT I O N resistance, which may require dose reduction or drug withdrawal. There

High‐dose (HD)‐methotrexate (MTX) plays an essential role in the
consolidation and maintenance therapy for acute lymphoblastic leukemia
(ALL).1,2 The introduction of MTX‐based chemotherapy has significantly
improved the cure rates of childhood ALL.3,4 Nevertheless, there is
marked inter‐individual variability in MTX pharmacokinetics and
responses to MTX.5-7 Some patients can develop severe side effects or
are well‐established associations between the exposure and response to
MTX.8 In general, the therapeutic window and safety window of MTX
are 20–40 μmol/l at 24 h after administration (C24h) and <0.5 μmol/l at
48 h after administration (C48h), respectively. It is necessary to increase
the dose and frequency of leucovorin rescue when C24h or C48h is above
the upper limit of the therapeutic threshold of MTX.9 Therefore, the MTX
concentration can serve as an objective marker of toxicity.10

J Gene Med. 2017;19:e2990. wileyonlinelibrary.com/journal/jgm Copyright © 2017 John Wiley & Sons, Ltd. 1 of 7
https://doi.org/10.1002/jgm.2990
2 of 7
The antitumor activity of MTX depends on its inhibition of
dihydrofolate reductase (DHFR). 11
Inter‐individual differences in
MTX pharmacokinetics can largely be attributed to the genetic
polymorphisms of transporters and enzymes involved in folate
metabolism.12 It may be valuable to identify genetic polymorphisms
associated with serum MTX concentrations to optimize the
treatment outcome of pediatric patients with ALL. 5,10‐
Methylenetetrahydrofolatereductase (MTHFR) is one of the key
enzymes in one‐carbon metabolism, converting 5,10‐methylenetetra-
hydrofolate to 5,10‐methyltetrahydrofolate.13 Several studies have
shown that genetic polymorphisms of MTHFR may affect the kinetics
and effects of MTX.14-16 The incidence of hepatic or hematological
toxicity after MTX therapy was higher among patients with the
rs1801133 TT genotype.14 Radtke et al.15 found that the MTHFR
rs1801131 single‐nucleotide polymorphisms (SNP) was significantly
associated with event‐free survival. The half‐life and area under
the curve of plasma MTX significantly increased, whereas the MTX
elimination rate and clearance significantly decreased in rs1801133
TT genotype carriers.15
Nonsynonymous SNPs in the coding region of MTHFR attracted
the most attention in previous studies. Among them, rs1801133
C > T and rs1801131 A > C were the most widely studied in ALL.
Relatively less attention was paid to SNPs within the 3′‐untranslated
region (3′‐UTR) of MTHFR, which contains multiple microRNA
(miRNA) binding sites.17 miRNAs are a class of short noncoding RNA
that act as post‐transcriptional regulators of mRNA expression by
specifically binding to the 3′‐UTR of target mRNAs.18 Therefore,
genetic variations in miRNA target sites may affect miRNA–mRNA
19
interactions, resulting in altered expression of target genes. Recent
studies suggested that SNPs located in miRNA binding sites were
associated with cancer risk and drug response. Mishra et al.20 identi-
fied a miR‐24 binding site polymorphism, 829 C > T, in the 3′‐UTR
of DHFR that led to the loss of miR‐24 function and resulted in DHFR
over‐expression and MTX resistance.
In the present study, we focused on two miRNA binding site
polymorphisms (rs3737966 G > A and rs35134728 DEL/TTC) in
the 3′‐UTR of MTHFR, which have rarely been studied previously.
We investigated the distribution of the two polymorphisms and their
association with serum MTX level in a Chinese pediatric population
with ALL.
2 | MATERIALS AND METHODS
2.1 | Study population
Blood samples were collected from 144 patients with ALL treated at
the Pediatric Department of Beijing Shijitan Hospital, Capital Medical
University, Beijing, China, from May 2009 to November 2013. All
patients were administered HD‐MTX and then leucovorin rescue
treatment. Prior to MTX administration, all children were in complete
remission. Patients were assigned to the low‐risk group, intermediate
risk group or high‐risk group according to their clinical characteristics,
the biologic profile of leukemic cells and the early response to induc-
tion treatment. Children received MTX 1–3 g/m2 (low‐risk group and
WANG ET AL.
intermediate risk group) or 3–5 g/m2 (high‐risk group). The first dose
(1/6 of the full dose; ≤ 500 mg) was intravenously transfused within
30 min. The remaining dose was infused during the following
23.5 h. Leucovorin rescue (15 mg/m2) was started 36 h after HD‐
MTX and continued for at least six doses every 6 h until the MTX
concentration was below 0.1 μmol/l. Patients received intravenous
hydration and alkalization in accordance with standard protocols. All
procedures performed in studies were in accordance with the ethical
standards of the institutional and/or national research committee and
with the 1964 Helsinki Declaration and its later amendments or
comparable ethical standards. The study was approved by The Ethics
Committee of Beijing Shijitan Hospital, Capital Medical University
(Approval No. 2010‐C16). Informed consent was obtained from all
participants or their guardians before inclusion in the study.
Regarding inclusion/exclusion criteria, patients who received HD‐
MTX would be included in the present study if their informed
consents were obtained and their complete demographics informa-
tion, MTX chemotherapy protocols and MTX concentrations at 24
and 42 h were available from the medical records. Patients whose
genotyping for the two polymorphisms investigated were not
successful, or who had severe hepatic or renal injury before MTX
chemotherapy, were excluded from the study.
2.2 | MTX determination
Each blood sample was centrifuged (3000 g for 5 min) at room
temperature to obtain serum. The serum concentrations of MTX were
measured by a fluorescence polarization immunoassay on a TDxFLx
analyzer (Abbott Laboratories, Abbott Park, IL, USA). MTX serum con-
centrations were determined 24 and 42 h after the start of infusion,
comprising routine determinations performed for all patients receiving
HD‐MTX. The lowest measurable concentration was 0.01 μmol/l. The
measurements of quality control samples with low, middle and high
concentrations were between 90% and 110% of the stated sample
concentrations. The intra‐ and inter‐day variations were all within
10%. Dose‐adjusted serum concentrations (C/D ratio) of MTX were
calculated by dividing the concentration (μmol/l) by the corresponding
24‐h dose (g/m2).
2.3 | Polymorphism selection
MTHFR rs3737966 G > A and rs35134728 DEL/TTC polymorphisms
were selected for further investigation based on certain con-
siderations. Using Patrocles (http://www.patrocles.org/) and
PolymiRTS (http://compbio.uthsc.edu/miRSNP/) Databases, it was
predicted that the two polymorphisms had functional effects that lead
to the disruption/creation of miRNAs targets. Additionally, the minor
allele frequencies of the two polymorphisms were above 10% in
Chinese samples of the dbSNP database, which indicated they might
have pharmacogenetic relevance for Chinese populations. Finally,
there are data available on the two polymorphisms in the literature
so far. There were only the two polymorphisms in the MTHFR gene
that met the inclusion standards at the moment of our selection. Con-
sidering that the number of annotated miRNAs has been increasing,
there may be other polymorphic miRNA binding sites for this gene.
WANG ET AL. 3 of 7

2.4 | MTHFR rs3737966 G > A and rs35134728 DEL/ used for comparisons among three non‐normally distributed groups.
TTC genotyping p < 0.05 was considered statistically significant.

Genomic DNA was extracted from frozen blood samples using a

QIAamp DNA Blood Mini Kit (Qiagen, Valencia, CA, USA) in accor-
dance with the manufacturer's instructions. DNA concentration and
purity were determined by absorbance at 260 and 280 nm. DNA
samples were diluted to working concentrations of 30–50 ng/ml
before genotyping. Approximately 20 ng of genomic DNA was used
to genotype each sample. The SNPs rs3737966 G > A and
rs35134728 DEL/TTC were genotyped using the Sequenom
MassArray system (iPLEX assay; Sequenom, San Diego, CA, USA).
Locus‐specific polymerase chain reaction and detection primers were
designed using assay design software (MassARRAY, version 3.1;
Sequenom). Allele detection was performed using matrix‐assisted
laser desorption/ionization time‐of‐flight mass spectrometry. To
evaluate the genotyping accuracy for the two polymorphisms, 10
samples were genotyped repeatedly and the results were 100%
consistent. The genotyping call rates for rs3737966 G > A and
rs35134728 DEL/TTC were above 95%.
2.5 | Statistical analysis
Statistical analyses were two‐tailed and performed using Prism,
version 4.00 (GraphPad Software Inc., San Diego, CA, USA).
Distributions of genotypes and alleles were assessed for deviation
by Hardy–Weinberg equilibrium, and the percentages of patients
with MTX concentrations above defined thresholds were compared
in different genotypes using the chi‐squared test or the Fisher's exact
test. The Mann–Whitney test was used for comparisons between
two non‐normally distributed groups. The Kruskal–Wallis H‐test was
3 | RESULTS
3.1 | Baseline characteristics of study population
The baseline characteristics of the study population on admission
are shown in Table 1. A total of 144 Chinese children with ALL
(male/female, 87/57; B‐cell/T‐cell/mixed‐lineage ALL, 114/13/17)
were included in the present study. The mean ± SD age of participants
was 7.51 ± 3.98 years. The median MTX dosage received was 2.50 g/
m2. The mean ± SD MTX C24h and C42h were 30.53 ± 12.52 and
0.20 ± 0.127 μmol/l, respectively.
3.2 | Genotype and allele frequencies
The genotype and allele frequencies for MTHFR rs35134728 DEL/TTC
and rs3737966 G > A polymorphisms are summarized in Table 2. Inter-
estingly, the two polymorphisms were found to be in complete linkage
disequilibrium (r2 = 1) for the patients included in the study. The
observed frequencies for rs35134728 DEL/DEL, DEL/TTC and TTC/
TTC genotypes, as well as rs3737966 GG, GA and AA genotypes, were
6, 38 and 100, respectively. The theoretical frequencies calculated
were 4.34, 41.32 and 98.34, respectively. No significant deviations
from Hardy–Weinberg equilibrium were observed for the polymor-
phisms (χ2 = 0.93, p > 0.05). The minor allele frequency (MAF)
observed in the present study (17.4%) was similar to those in the
Han Chinese (in Beijing, China) (22.8%) and Japanese (in Tokyo, Japan)
(17.8%) samples of the 1000 Genomes Project. However, it was

TABLE 1 Patient demographic and clinical characteristics

Characteristics Value Median Range

Age (years), mean ± SD 7.51 ± 3.98 6.5 1–17
Sex, % (n) Male 60.4 (87)
Female 39.6 (57)
Immunotype, % (n) B lineage 79.2 (114)
T lineage 9.0 (13)
Mixed 11.8 (17)
Cytogenetics, % (n) t (9; 22)+ 3.5 (5)
t (4; 11)+ 1.4 (2)
t (1; 19)+ 3.5 (5)
t (12; 21)+ 14.6 (21)
Other fusion genes+ 31.2 (45)
Fusion genes‐ 45.8 (66)
Karyotype, % (n) Hyperdiploid 15.3 (22)
Nonhyperdiploid 84.7 (122)
Risk, % (n) Low 47.2 (68)
Intermediate 32.0 (46)
High 20.8 (30)
MTX dose (g/m2), mean ± SD 2.43 ± 0.36 2.50 2.00–4.00
MTX C24 (μmol/l), mean ± SD 30.53 ± 12.52 28.83 10.17–74.37
MTX C42 (μmol/l), mean ± SD 0.20 ± 0.12 0.18 0.01–0.65
4 of 7 WANG ET AL.

TABLE 2 Genotypes and allele frequencies for MTHFR rs35134728 DEL/TTC and rs3737966 G > A in Chinese ALL children and samples in the
1000 genomes project
Frequency, % (n)
Genotypes and alleles Chinese ALL children CEUa HCBa JPTa YRIa

rs35134728 rs3737966 n = 144 n = 99 n = 103 n = 104 n = 108

Del/del G/G 4.2 (6) 16.2 (16)** 2.9 (3) 2.9 (3) 40.7 (44)**
Del/TTC G/A 26.4 (38) 53.5 (53)** 39.8 (41)* 29.8 (31) 46.3 (50)**
TTC/TTC A/A 69.4 (100) 30.3 (30)** 57.3 (59)* 67.3 (70) 13.0 (14)**
DEL G 17.4 (50) 42.9 (85)** 22.8 (47) 17.8 (37) 63.9 (138)**
TTC A 82.6 (238) 57.1 (113)** 77.2 (159) 82.2 (171) 36.1 (78)**
a
Genotype and allele frequencies are available at: http://grch37.ensembl.org/index.html.
*p < 0.05 and
**p < 0.01 compared to Chinese ALL children group by the chi‐square test.
HCB, Han Chinese in Beijing, China; JPT, Japanese in Tokyo, Japan; CEU, Utah residents with ancestry from northern and western Europe; YRI, Yoruba in
Ibadan, Nigeria.

significantly lower than those in the Utah residents with northern and DEL/TTC‐G/A or DEL/DEL‐G/G haplotype carriers (25.0%),
western Europe ancestry (42.9%) and Yoruba (in Ibadan, Nigeria) although the differences were not statistically significant. The above
(63.9%) samples (p < 0.01), indicating marked ethnic differences in findings indicated that MTHFR rs35134728 DEL/TTC and
the distributions of MTHFR rs35134728 DEL/TTC and rs3737966 rs3737966 G > A polymorphisms had significant effects on serum
G > A polymorphisms. MTX concentrations.

3.3 | Effects of MTHFR polymorphisms on serum 3.4 | Effects of MTHFR polymorphisms on the
concentrations of MTX prognosis of ALL

Table 3 shows the univariate analysis of the C/D ratios of MTX at 24
and 42 h. Sex, cytogenetics and risk stratification were not associated
with the two parameters. The association between age and C/D ratios
of MTX at 24 and 42 h was marginally significant (p = 0.079 and
0.088, respectively). The highest median C/D ratios were in patients
aged 9–17 years (12.74 μmol/l per g/m ). There were marginally sig-
2 83.7% and 76.2%, respectively. The differences were not statistically
nificant differences in the C/D ratios of MTX at 24 h among three
immunotype groups (p = 0.082) and the highest median C/D ratio
was in patients with T‐cell lineage ALL (15.22 μmol/l per g/m ). How- 2
ever, the median C/D ratios of MTX at 42 h were identical among the
three immunotype groups (0.07 μmol/l per g/m2). The median C/D
ratios of MTX at 24 h in patients with hyperdiploid karyotype
(10.71 μmol/l per g/m2) were significantly lower than those in
patients with nonhyperdiploid karyotype (12.20 μmol/l per g/m2,
p = 0.017). However, there were no significant differences between
the median C/D ratios of MTX at 42 h of the two karyotypes. The
median C/D ratios of MTX at 24 and 42 h for the TTC/TTC‐A/A
haplotype carriers (11.74 and 0.07 μmol/l per g/m2, respectively)
were significantly lower than those in DEL/TTC‐G/A or DEL/DEL‐
G/G haplotype carriers (12.49 and 0.09 μmol/l per g/m2, respectively)
(p = 0.046 and 0.017, respectively).
We also analyzed the associations of the polymorphisms with
MTX C24h > 40 μmol/l (therapeutic threshold) and
C42h > 0.5 μmol/l (safety threshold). As shown in Table 4, the per-
centage of patients with MTX C42h > 0.5 μmol/l was significantly
lower in TTC/TTC‐A/A haplotype carriers (0.4%) compared to that
in DEL/TTC‐G/A or DEL/DEL‐G/G haplotype carriers (13.6%,
p = 0.036). The proportion of patients with MTX C24h > 40 μmol/l
in TTC/TTC‐A/A haplotype carriers (18.0%) was lower than that in
We performed preliminary survival analysis of the two polymorphisms
interested and different doses of MTX. The event free survival rates of
patients with TTC/TTC‐A/A haplotype and DEL/DEL‐G/G or DEL/
TTC‐G/A haplotypes were 82.0% and 84.1%, respectively. The event
free survival rates of patients with MTX dose ≥3 and <3 g/m2 were
significant by log rank test. These findings indicated there were no sig-
nificant effects of the two polymorphisms investigated and MTX dose
on the event free survival of patients in the present study.
3.5 | Computational prediction
The polymorphisms rs35134728 DEL/TTC and rs3737966 G > A are
located in the 3′‐UTR of MTHFR. It is possible that the two polymor-
phisms create or disrupt miRNA binding sites. Table 5 shows predicted
miRNAs whose putative binding sites overlap with rs35134728 DEL/
TTC and rs3737966 G > A polymorphisms (http://compbio.uthsc.
edu/miRSNP). The interaction of the predicted miRNAs with MTHFR
is likely to be affected by the polymorphisms. However, this needs to
be validated in further studies.
4 | DISCUSSION
a
MTX is a widely used agent in the therapy of ALL worldwide.2 How-
ever, the therapeutic responses to MTX show large variability among
individuals, sexes, and racial or ethnic groups.6 This may be partially
attributed to the genetic polymorphisms in the coding genes of metab-
olizing enzymes, transporters and targets involved in the cellular path-
way of MTX.12 In the present study, we investigated the distribution of
WANG ET AL. 5 of 7

TABLE 3 Univariate analysis of MTX C/D ratios at 24 h and 42 h after the two polymorphisms occurred simultaneously in each patient in
administration in Chinese ALL children our population. The genotype of rs35134728 DEL/TTC would be
C/D (μmol/l per g/m2) known if the genotype of rs3737966 G > A was available, vice versa.
n 24 h 42 h The haplotype of the two polymorphisms would be obtained when
Sex genotyping for only one polymorphism was performed. The MAF

Male 87 11.94 (5.09–29.75) 0.07 (0.00–0.25)

found in the present study was similar to those in Chinese and
Japanese samples. However, a contrasting distribution pattern was
Female 57 12.18 (4.51–28.44) 0.08 (0.00–0.26)
observed in European and African samples of the 1000 Genomes
p‐valuea 0.707 0.371
Project. These results suggest that a wide inter‐ethnic variability
Age (years)
existed in the genotype and allele frequencies of rs3737966 G > A
1–4 42 10.67 (4.65–27.86) 0.07 (0.00–0.25)
and rs35134728 DEL/TTC polymorphisms, and this might contribute
5–8 48 11.83 (4.76–29.75) 0.06 (0.03–0.20)
to variations in responses to MTX in different ethnic populations.
9–17 54 12.74 (4.51–28.44) 0.08 (0.00–0.26)
MTHFR is an important enzyme in the folate–homocysteine
p‐valueb 0.079 0.088
Immunotype
cycle.21 The functional variability in MTHFR may affect both the effi-

T‐lineage 13 15.22 (8.06–20.67) 0.07 (0.04–0.18) cacy and toxicity of MTX. It was found that the MTHFR gene was
B‐lineage 114 12.02 (4.51–29.75) 0.07 (0.00–0.26) highly polymorphic. Thus, the association between genetic polymor-
Mixed‐lineage 17 10.71 (4.76–18.15) 0.07 (0.00–0.19) phisms in MTHFR with therapeutic responses to MTX in patients with
p‐valueb 0.082 0.785 ALL has attracted significant attention. The classic examples were
Cytogenetics rs1801133 C > T and rs1801131 A > C, which resulted in the reduc-
t (9; 22)+ 5 10.94 (5.35–11.78) 0.08 (0.04–0.14) tion of MTHFR enzyme activity.22 Most studies indicated the strong
t (4; 11)+ 2 9.68 (6.89–12.47) 0.17 (0.11–0.22) association between the above two SNPs with the occurrence of
t (1; 19)+ 5 14.52 (6.48–22.31) 0.09 (0.05–0.18) toxicity during HD‐MTX treatment of ALL patients. The frequencies
t (12; 21)+ 21 11.94 (7.11–18.24) 0.08 (0.03–0.25) of hepatic, hematological, gastrointestinal and dermal toxicities were
Other fusion genes+ 45 11.99 (6.05–28.44) 0.07 (0.03–0.26) higher in patients with the rs1801133 TT or rs1801131 CC geno-
Fusion genes– 66 12.48 (4.51–29.75) 0.07 (0.00–0.20) type.23 Some studies focused on the association of the two SNPs with
p‐valueb 0.355 0.154 the risk of relapse in ALL patients. Eissa and Ahmed16 found that the
Karyotype rs1801133 TT genotype was associated with higher relapse rate in
Hyperdiploid 22 10.71 (4.65–19.14) 0.08 (0.03–0.20)
adult ALL patients. The effects of the two SNPs on the concentrations
Nonhyperdiploid 122 12.20 (4.51–29.75) 0.07 (0.00–0.26)
of MTX could account for their association with worse treatment out-
p‐valuea 0.017 0.469
come in ALL populations to some extent. Prolonged high serum MTX
Risk
concentrations were observed in patients with rs1801133 T allele.24
Low 68 11.49 (4.51–29.75) 0.07 (0.00–0.25)
In our previous study, it was found that the MTX C/D ratio and per-
Intermediate 46 12.94 (5.09–28.44) 0.07 (0.02–0.26)
centage of serum MTX above the therapeutic threshold in carriers of
High 30 11.98 (4.76–21.34) 0.07 (0.00–0.22)
rs1801133 CT or TT were higher than those in CC carriers.25
p‐valueb 0.135 0.772
However, there were also conflicting results on the association of poly-
MTHFR haplotype
rs35134728‐ morphisms in MTHFR with MTX pharmacokinetics and toxicity as a
rs3737966 result of a difference in sample size and the selected populations.26,27
TTC/TTC‐A/A 100 11.74 (4.51–29.75) 0.07 (0.00–0.25)
In the present study, significantly lower C/D ratios of MTX at 24
Del/TTC‐G/A + 44 12.49 (6.41–25.74) 0.09 (0.00–0.26)
and 42 h were observed in patients with the rs3737966 A/A or
DEL/del‐G/G
p‐valuea 0.046 0.017
rs35134728 TTC/TTC genotype. In addition, the proportion of
patients with MTX C42h > 0.5 μmol/l was significantly lower in
Data are the median (range).
rs3737966 A/A or rs35134728 TTC/TTC genotype carriers. These
a
p‐value determined by the Mann–Whitney test.
findings indicate that rs3737966 A or rs35134728 TTC may be a
b
p‐value determined by the Kruskal–Wallis H‐test.
functional genetic variant with significant effects on serum MTX con-
centrations. We also investigated the influence of clinical factors on
two miRNA binding site polymorphisms (rs3737966 G > A and the serum concentration of MTX, such as sex, age, immunotype, cyto-
rs35134728 DEL/TTC) in the 3′‐UTR of MTHFR and their effects on genetics, karyotype and risk stratification. Marginally higher C/D ratios
the serum MTX level in a cohort of Chinese pediatric patients with of MTX at 24 h were seen in children with older age and T‐cell line-
ALL. To our knowledge, this is the first report evaluating the frequen- age. Significant lower C/D ratios of MTX at 24 h were observed in
cies and functionality of the two polymorphisms in Chinese children patients with hyperdiploid karyotype. There were no significant
with ALL. effects on the C/D ratios of MTX with respect to sex, cytogenetics
The high linkage disequilibrium between the two investigated and risk stratification. MTX concentrations above the therapeutic
polymorphisms predicted in the 1000 Genomes Project was confirmed threshold might be associated with greater toxicity and worse out-
in the present study. rs35134728 DEL/TTC and rs3737966 G > A come.8 However, the preliminary results of survival analysis did not
polymorphisms were in complete linkage disequilibrium, which meant show the significant effects of the two polymorphisms investigated
6 of 7 WANG ET AL.

TABLE 4 Proportions of patients with MTX C24h > 40 μmol/l and MTX C42h > 0.5 μmol/l
Haplotype n MTX C24h > 40 μmol/l, % (n) MTX C42h > 0.5 μmol/l, % (n)
MTHFR rs35134728‐ rs3737966
TTC/TTC‐A/A (reference) 100 18.0 (18) 0.4 (4)
Del/TTC‐G/A 38 26.3 (10) 15.8 (6)
p‐value 0.278 0.027
Del/del‐G/G 6 16.7 (1) 0.0 (0)
p‐value 1.000 1.000
Del/TTC‐G/A + DEL/del‐G/G 44 25.0 (11) 13.6 (6)
p‐value 0.335 0.036

p‐value determined by the chi‐squared test or Fisher's exact test.

TABLE 5 Prediction of miRNAs binding sites affected by rs3737966 and rs35134728 polymorphisms in 3′‐UTR of MTHFR
Location dbSNP ID Variant type Ancestral allele Allele miR ID miRSitea

11847334 rs35134728 INDEL – – hsa‐miR‐1293 gcctcttCCACCCAttc

hsa‐miR‐3190‐3p gcctCTTCCACccattc
hsa‐miR‐363‐5p gcctctTCCACCCAttc
hsa‐miR‐4483 gcctcttCCACCCAttc
Hsa‐miR‐6745 gcctctTCCACCCAttc
hsa‐miR‐6756‐5p gcctcttCCACCCAttc
hsa‐miR‐6766‐5p gcctcttCCACCCAttc
hsa‐miR‐7‐5p gccTCTTCCAcccattc
TTC hsa‐miR‐3185 gcctCTTCTTCcacccattc
11847759 rs3737966 SNP G A hsa‐miR‐1262 cctgcCACCCAAa
hsa‐miR‐6736‐5p cctgcCACCCAAa
hsa‐miR‐6741‐5p cctgcCACCCAAa

Prediction results were available in the PolymiRTS Database 3.0 (http://compbio.uthsc.edu/miRSNP).

a
Sequence context of the miRNA site. Bases complementary to the seed region are shown in uppercase and SNPs are underlined and shown in bold.

and MTX dose on the event‐free survival of patients in the present
study. This may be a result of the short follow‐up times for some
subjects. Future studies will evaluate the effects of specific genetic
and clinical factors on the toxicity and outcome of MTX treatment
in our study population.
The polymorphisms rs3737966 G > A and rs35134728 DEL/
TTC were not located in the coding region; thus, they would not
result in amino acid sequence changes within the MTHFR enzyme.
However, computational analysis predicted that the two poly-
morphisms were present within miRNA binding sites of MTHFR.
miRNAs post‐transcriptionally regulate gene expression through
3′‐UTR‐binding of the target mRNA. Polymorphisms in 3′‐UTR over-
lapping with miRNA seed sequences can interfere with miRNA binding
to the target mRNA, potentially disrupting or creating miRNA binding
sites.18 Therefore, SNPs in miRNA genes or their target binding sites
could affect phenotypes and disease susceptibility.28-30 No functional
information regarding the identified polymorphisms was found in the
literature. However, a certain number of such miRNA binding site
polymorphisms have been functionally evaluated in recent years. Chin
et al.31 identified a functional SNP in the binding site of miRNA let‐7
within the 3′‐UTR of KRAS, which resulted in altered let‐7‐mediated
regulation of KRAS expression and an increased the risk of non‐small
cell lung cancer. Ma et al.32 found that rs3208684 A > C enhanced
5‐fluorouracil sensitivity and the expression of Bcl‐xl by disrupting
the binding of let‐7b to the 3′‐UTR of Bcl‐xl. Computational analysis
suggested that the interactions between multiple miRNAs and MTHFR
could potentially be affected by rs3737966 G > A and rs35134728
DEL/TTC. It was possible that the two polymorphisms affected the
binding affinities of certain miRNAs to the MTHFR mRNA, which
resulted in an altered protein expression of MTHFR. These might be
the potential mechanism of the two polymorphisms associated with
MTX concentrations. Further functional studies will validate whether
the polymorphisms lead to altered expression of MTHFR.
The present study had several limitations. The relatively small
sample size that we used probably resulted in a low power and bias
with respect to the interpretation of the results. The primary objective
of the present study was to compare the difference of C/D ratio of
MTX at 42 h between TTC/TTC‐A/A and other haplotypes carriers.
Group sample sizes of 100 and 44 achieve 65% power to detect a
difference of 0.02 with a significance level (alpha) of 0.05.
Furthermore, data on toxicity and outcome of MTX treatment were
not collected in the present study. Therefore, further detailed studies
are required to evaluate the effects of the polymorphisms on
responses to MTX in our population. Finally, we did not provide
evidence supporting the computationally predicted biological function
of the polymorphisms. This also warrants further investigation.
In conclusion, two polymorphisms associated with MTX pharma-
cokinetics were identified in the 3′‐UTR of MTHFR. Their genotyping
may help to optimize the MTX treatment of ALL patients. Further
studies are needed to evaluate the biological function of the genetic
variations identified, as well as their effects on clinical outcomes in
these patients.
WANG ET AL.
ACKNOWLEDGEMEN TS
This work was supported by National Natural Science Foundation of
China (No. 81503135), Beijing Municipal Administration of Hospitals’
Youth Programme (No. QML20160703) and Beijing Key Laboratory
of Bio‐characteristic Profiling for Evaluation of Rational Drug Use
(No. BZ0439). The authors declare that there are no conflict of interest.
ORCID
Shu‐Mei Wang http://orcid.org/0000-0001-9121-8168
RE FE R ENC E S
1. Jolivet J, Cowan KH, Curt GA, Clendeninn NJ, Chabner BA. The
pharmacology and clinical use of methotrexate. N Engl J Med.
1983;309:1094‐1104.
2. Pui CH, Evans WE. Treatment of acute lymphoblastic leukemia. N Engl J
Med. 2006;354:166‐178.
3. Stanulla M, Schrappe M. Treatment of childhood acute lymphoblastic
leukemia. Semin Hematol. 2009;46:52‐63.
4. Pui CH, Robinson LL, Look AT. Acute lymphoblastic leukemia. Lancet.
2008;371:1030‐1043.
5. Schmiegelow K. Advances in individual prediction of methotrexate
toxicity: a review. Br J Haematol. 2009;146:489‐503.
6. Imanishi H, Okamura N, Yagi M, et al. Genetic polymorphisms associ-
ated with adverse events and elimination of methotrexate in
childhood acute lymphoblastic leukemia and malignant lymphoma. J
Hum Genet. 2007;52:166‐171.
7. Rau T, Erney B, Gores R, Eschenhagen T, Beck J, Langer T. High‐dose
methotrexate in pediatric acute lymphoblastic leukemia: impact of
ABCC2 polymorphisms on plasma concentrations. Clin Pharmacol Ther.
2006;80:468‐476.
8. Stoller RG, Hande KR, Jacobs SA, Rosenberg SA, Chabner BA. Use of
plasma pharmacokinetics to predict and prevent methotrexate toxicity.
N Engl J Med. 1977;297:630‐634.
9. Yang SL, Zhao FY, Song H, Shen DY, Xu XJ. Methotrexate associated
renal impairment is related to delayed elimination of high‐dose metho-
trexate. Sci World J. 2015;2015:751703
10. Lopez‐Lopez E, Martin‐Guerrero I, Ballesteros J, et al. Polymorphisms
of the SLCO1B1 gene predict methotrexate‐related toxicity in
childhood acute lymphoblastic leukemia. Pediatr Blood Cancer.
2011;57:612‐619.
11. Rudin S, Marable M, Huang RS. The promise of pharmacogenomics in
reducing toxicity during acute lymphoblastic leukemia maintenance
treatment. Genomics Proteomics Bioinformatics. 2017;15:82‐93.
12. Huang L, Tissing WJ, de Jonge R, van Zelst BD, Pieters R.
Polymorphisms in folate‐related genes: association with side effects
of high‐dose methotrexate in childhood acute lymphoblastic leukemia.
Leukemia. 2008;22:1798‐1800.
13. Kantar M, Kosova B, Cetingul N, et al. Methylenetetrahydrofolate
reductase C677T and A1298C gene polymorphisms and therapy‐
related toxicity in children treated for acute lymphoblastic leukemia
and non‐Hodgkin lymphoma. Leuk Lymphoma. 2009;50:912‐917.
14. Xiao Y, Deng TR, Su CL, Shang Z. Methylenetetrahydrofolate reduc-
tase polymorphisms and susceptibility to acute lymphoblastic
leukemia in a Chinese population: a meta‐analysis. Oncol Res Treat.
2014;37:576‐582.
15. Radtke S, Zolk O, Renner B, et al. Germline genetic variations in meth-
otrexate candidate genes are associated with pharmacokinetics,
toxicity, and outcome in childhood acute lymphoblastic leukemia.
Blood. 2013;121:5145‐5153.
16. Eissa DS1, Ahmed TM. C677T and A1298C polymorphisms of the
methylenetetrahydrofolate reductase gene: effect on methotrexate‐
related toxicity in adult acute lymphoblastic leukaemia. Blood Coagul
Fibrinolysis 2013; 24: 181–188.
7 of 7
17. Ryan BM, Robles AI, Harris CC. Genetic variation in microRNA
networks: the implications for cancer research. Nat Rev Cancer.
2010;10:389‐402.
18. Lai EC, Micro RNA. are complementary to 3′ UTR sequence motifs that
mediate negative post‐transcriptional regulation. Nat Genet.
2002;30:363‐364.
19. Teo MT, Landi D, Taylor CF, et al. The role of microRNA‐binding site
polymorphisms in DNA repair genes as risk factors for bladder cancer
and breast cancer and their impact on radiotherapy outcomes. Carcino-
genesis. 2012;33:581‐586.
20. Mishra PJ, Humeniuk R, Mishra PJ, Longo‐Sorbello GS, Banerjee D,
Bertino JRA. miR‐24 microRNA binding‐site polymorphism in
dihydrofolate reductase gene leads to methotrexate resistance. Proc
Natl Acad Sci USA. 2007;104:13513‐13518.
21. Tong N, Fang Y, Li J, et al. Methylenetetrahydrofolate reductase
polymorphisms, serum methylenetetrahydrofolate reductase levels,
and risk of childhood acute lymphoblastic leukemia in a Chinese
population. Cancer Sci. 2010;101:782‐786.
22. Frosst P, Blom HJ, Milos R, et al. A candidate genetic risk factor for vas-
cular disease: a common mutation in methylenetetrahydrofolate
reductase. Nat Genet. 1995;10:111‐113.
23. Chiusolo P, Reddiconto G, Farina G, et al. MTHFR polymorphisms'
influence on outcome and toxicity in acute lymphoblastic leukemia
patients. Leuk Res. 2007;31:1669‐1674.
24. Yanagimachi M, Goto H, Kaneko T, et al. Influence of pre‐hydration and
pharmacogenetics on plasma methotrexate concentration and renal
dysfunction following high‐dose methotrexate therapy. Int J Hematol.
2013;98:702‐707.
25. Wang SM, Sun LL, Zeng WX, Wu WS, Zhang GL. Influence of genetic
polymorphisms of FPGS, GGH, and MTHFR on serum methotrexate
levels in Chinese children with acute lymphoblastic leukemia. Cancer
Chemother Pharmacol. 2014;74:283‐289.
26. Shimasaki N, Mori T, Samejima H, et al. Effects of methylenetetrahy-
drofolate reductase and reduced folate carrier 1 polymorphisms on
high‐dose methotrexate‐induced toxicities in children with acute
lymphoblastic leukemia or lymphoma. J Pediatr Hematol Oncol.
2006;28:64‐68.
27. Seidemann K, Book M, Zimmermann M, et al. MTHFR 677 (C → T)
polymorphism is not relevant for prognosis or therapy‐associated
toxicity in pediatric NHL: results from 484 patients of multicenter trial
NHL‐BFM 95. Ann Hematol. 2006;85:291‐300.
28. Slaby O, Bienertova‐Vasku J, Svoboda M, Vyzula R. Genetic polymor-
phisms and microRNAs: new direction in molecular epidemiology of
solid cancer. J Cell Mol Med. 2012;16:8‐21.
29. Zhang L, Liu Y, Song F, et al. Functional SNP in the microRNA‐367
binding site in the 3'UTR of the calcium channel ryanodine receptor
gene 3 (RYR3) affects breast cancer risk and calcification. Proc Natl
Acad Sci USA. 2011;108:13653‐13658.
30. Nicoloso MS, Sun H, Spizzo R, et al. Single‐nucleotide polymorphisms
inside microRNA target sites influence tumor susceptibility. Cancer
Res. 2010;70:2789‐2798.
31. Chin LJ, Ratner E, Leng S, et al. A SNP in a let‐7 microRNA
complementary site in the KRAS 3′ untranslated region increases
non‐small cell lung cancer risk. Cancer Res. 2008;68:8535‐8540.
32. Ma J, Guo R, Wang T, Pan X, Lei X. Let‐7b binding site polymorphism in
the B‐cell lymphoma‐extra large 3′ UTR is associated with fluorouracil
resistance of hepatocellular carcinoma. Mol Med Rep. 2015;11:677‐681.
How to cite this article: Wang S‐M, Zeng W‐X, Wu W‐S, Sun
L‐L, Yan D. Association between MTHFR microRNA binding
site polymorphisms and methotrexate concentrations in
Chinese pediatric patients with acute lymphoblastic leukemia.
J Gene Med. 2017;19:e2990. https://doi.org/10.1002/jgm.2990