NQO1 rs1800566 (C609T), PON1 rs662 (Q192R), and PON1 rs854560 (L55M) polymorphisms segregate the risk of childhood acute leukemias according to age range distribution

Page 1

1 23

Cancer Causes & ControlAn International Journal of Studies ofCancer in Human Populations ISSN 0957-5243Volume 23Number 11 Cancer Causes Control (2012)23:1811-1819DOI 10.1007/s10552-012-0060-5

NQO1 rs1800566 (C609T), PON1 rs662(Q192R), and PON1 rs854560 (L55M)polymorphisms segregate the risk ofchildhood acute leukemias according to agerange distributionBruno Alves de Aguiar Gonçalves, GiseleM. Vasconcelos, Luiz Claudio SantosThuler, Camilla Andrade, AlessandraFaro, et al.

Page 2

1 23

Your article is protected by copyright and

all rights are held exclusively by Springer

Science+Business Media B.V.. This e-offprint

is for personal use only and shall not be self-

archived in electronic repositories. If you

wish to self-archive your work, please use the

accepted author’s version for posting to your

own website or your institution’s repository.

You may further deposit the accepted author’s

version on a funder’s repository at a funder’s

request, provided it is not made publicly

available until 12 months after publication.

Page 3

ORIGINAL PAPER

NQO1 rs1800566 (C609T), PON1 rs662 (Q192R), and PON1rs854560 (L55M) polymorphisms segregate the risk of childhoodacute leukemias according to age range distribution

Bruno Alves de Aguiar Goncalves • Gisele M. Vasconcelos • Luiz Claudio Santos Thuler •

Camilla Andrade • Alessandra Faro • Maria S. Pombo-de-Oliveira •

Brazilian Collaborative Study Group of Infant Acute Leukemia

Received: 16 March 2012 / Accepted: 28 August 2012 / Published online: 14 September 2012

� Springer Science+Business Media B.V. 2012

Abstract

Purpose The risk of developing childhood leukemia has

been associated with gene polymorphisms that decrease the

activity of detoxifying metabolic enzymes and enzymes

involved in systemic oxidative stress. We investigated the

NQO1 and PON1 polymorphisms for associations with

susceptibility to childhood leukemia.

Methods Samples from 1,027 Brazilian children (519

acute lymphoblastic leukemia, ALL; 107 acute myeloid

leukemia, AML; 401 controls) were analyzed. TaqMAN

real-time assays were used to determine the NQO1

rs1800566 (C609T), PON1 rs662 (Q192R), and PON1

rs854560 (L55M) frequencies. Logistic regression was

used to evaluate the association of polymorphisms with

cases and controls, with age and somatic fusion genes

(MLL-r and ETV6-RUNX1) as covariables.

Results Children with at least one NQO1 variant allele

were at lower risk for developing infant AML (odds ratio

(OR) 0.26, 95 % confidence interval (CI) 0.10–0.68); no

association was detected for ALL. PON1 rs854560 (L55M)

was associated with an increased risk of developing

childhood leukemia (LM ? MM, OR 1.93, 95 % CI

1.32–2.81). The PON1 rs662 R192R genotype had a sta-

tistically significant decreased frequency in ALL (OR 0.64,

95 % CI 0.43–0.93). Infant ALL cases were more likely to

harbor homozygous PON1 rs854560 alleles than controls

(OR 1.72, 95 % CI 1.03–2.89); at least one M allele was

associated with an increased risk of ALL in children older

than 1 year (OR 1.99, 95 % CI 1.17–3.3).

Conclusions The NQO1 rs1800566 (C609T), PON1

rs854560 (L55M), and PON1 rs662 (Q192R) polymor-

phisms modified risk depending on leukemia subtype

(decreased in AML, increased and decreased in ALL,

respectively), age strata, and variant genotype combinations.

Keywords Infant leukemia � Acute lymphoblastic

leukemia � NQO1 � PON1 � Polymorphisms

Introduction

Epidemiological studies of childhood leukemia have

sought risk factors in an effort to determine the etiology of

childhood leukemia [1, 2]. Childhood leukemia, as for most

human malignancies, has been postulated to result from

combinatorial exposure factors influenced by inherited

genetic susceptibility [3, 4]. It is unlikely that all childhood

leukemia cases share a common or exclusive cause [2]; to

date, only ionizing radiation exposure has been consistently

and significantly associated with either acute lymphoid

leukemia (ALL) or acute myeloid leukemia (AML). Other

environmental risk factors such as parental cigarette

smoking, maternal alcohol consumption, and hydrocarbon

exposure have been inconsistently associated with child-

hood leukemia [1].

The details of the Brazilian Collaborative Study Group of Infant

Acute Leukemia are given in Appendix.

Electronic supplementary material The online version of thisarticle (doi:10.1007/s10552-012-0060-5) contains supplementarymaterial, which is available to authorized users.

B. A. de Aguiar Goncalves � G. M. Vasconcelos � C. Andrade �A. Faro � M. S. Pombo-de-Oliveira (&)

Pediatric Hematology-Oncology Program, Research Center,

Instituto Nacional de Cancer, Rua Andre Cavalcanti 37, Rio de

Janeiro, RJ 20231050, Brazil

e-mail: [email protected]

L. C. S. Thuler

Clinical Research, Research Center, Instituto Nacional de

Cancer, Rio de Janeiro, Brazil

123

Cancer Causes Control (2012) 23:1811–1819

DOI 10.1007/s10552-012-0060-5

Author's personal copy

Page 4

Gene fusions arising at the stem-cell level are associated

with distinct lineage-affiliated cell subtypes, and more

relevant acquired gene abnormalities in early childhood

leukemia occur during fetal hematopoiesis [5]. For

instance, infant leukemia usually harbors MLL gene rear-

rangements (MLL-r) as the major acquired genetic abnor-

mality; in common ALL (incidence peak at 2–5 years of

age), the ETV6-RUNX1 fusion gene is the most frequent

abnormality [2, 5]. Maternal exposure to DNA topoiso-

merase II inhibitors during pregnancy is proposed to be

associated with an increased risk of infant leukemia [6–8],

whereas common ALL has been associated with the delay

of infectious exposure [9, 10].

Among other factors, maternal exposure to pesticides

before conception or throughout pregnancy is potentially

associated with acute leukemia, ALL, and AML, in early

childhood [1, 11]. Furthermore, the risk of childhood leu-

kemia has also been associated with polymorphic variants of

genes encoding enzymes involved in the activation and

detoxification xenobiotic processes [3, 4]. We previously

demonstrated the population-attributable risk of estrogens,

pesticides, and benzene via maternal exposure during preg-

nancies of early childhood leukemia in Brazil [8, 11]. Some

chemicals that contain these compounds include the semiq-

uinones and quinones, which are substrates for the detoxifi-

cation enzyme NAD(P)H: quinone oxidoreductase (NQO1),

a flavoenzyme that detoxifies benzene metabolites, qui-

nones, and other topoisomerase II inhibitors. NQO1 protects

cells against mutagenesis due to free radicals and oxygen

metabolites [12, 13]. In addition, paraoxonase 1 (PON1) is

reported to exhibit antioxidant properties and functional

activity with systemic measures of oxidative stress; PON1

appears to oxidize lipids and hydrolyze active metabolites of

several organophosphorus pesticides [14]. The most com-

mon PON1 polymorphisms, rs662 (Q192R) and rs854560

(L55M), are known to cause variability in enzyme activity,

affecting its sensitivity to pesticides as well as the pharma-

cokinetics of the metabolism of some drugs [14, 15].

The NQO1 rs1800566 (C609T) polymorphism has been

associated with the risk of childhood ALL, particularly for

infant leukemia with MLL-r [16], whereas the PON1 poly-

morphism affecting the antioxidant properties of the

enzyme, with deficiency conferring organophosphorus tox-

icity, has not been studied in childhood leukemia. We were

motivated to examine the mechanisms and implications of

the NQO1 and PON1 variants in young children with acute

leukemia. We sought to evaluate a cohort of childhood leu-

kemias stratified according to age range for several reasons.

First, age-related differences in susceptibility to environ-

mental toxicants are known to occur in early childhood

(children younger than 24 months) [17, 18]; infants and

young children are at greater risk of bioaccumulation due to

immaturity of the xenobiotics system. Second, leukemia

subtypes with different somatic gene fusions and age ranges

have been associated with distinct leukemogenesis pathways

[5]. In the present study, we investigated whether the NQO1

and PON1 polymorphisms were associated with childhood

leukemia, focusing on leukemia subtypes and specific

genetic aberrations. Age subdivisions were used to enforce a

surrogate for leukemia subtypes of distinct etiopathology.

Materials and methods

Subjects

This study includes samples from 1,027 Brazilian children

(519 ALL, 107 AML, and 401 controls) that were diagnosed

between January 2000 and December 2010. Cases were

randomly selected from Brazilian regions through hospital-

based cancer registries and from the Brazilian Collaborative

Study Group of Infant Leukemia, both of which had bio-

logical material available with good-quality DNA [8, 19].

All controls were age-matched with cases selected among

children without malignancies from the same regional areas

as the leukemia cases. Samples from these children were

evaluated in the laboratory after exams that ruled out any

diagnosis of malignancies or myelodysplasia. Patients with

leukemia were classified as ALL or AML according to

standard immunophenotyping and molecular biology tech-

niques [20]. Demographic characteristics of cases and

controls, leukemia subtypes, and allele frequencies of

NQO1 rs1800566 (C609T), PON1 rs662 (Q192R), and

PON1 rs854560 (L55M) appear in Supplementary Table 1.

The pronounced level of admixture in the trihybrid

Brazilian population, with European, Amerindian, and

African roots, poses special challenges to ethnic classifi-

cations in this population. In this study, the race definition

was applied according to Instituto Brasileiro de Geografia e

Estatıstica (www.ibge.gov.br), which relies on skin color

self-definition. Then, we categorized subjects into two

major groups: whites (cases, n = 294; controls, n = 148)

and non-whites (cases, n = 320; controls, n = 255). The

non-whites group encompasses the participants self-cate-

gorized as brown or black. This categorization was also

based on the observation that European descent was pre-

dominant and similar in all geographical regions of Brazil

and that the proportions of African and Amerindian

ancestry did not differ between regions [21].

Ethical aspects

All collaborating Brazilian institutions approved the study,

and written informed consent was obtained from the par-

ents of the study subjects for both the interview and the

additional samples obtained after diagnostic procedures.

1812 Cancer Causes Control (2012) 23:1811–1819

123

Author's personal copy

Page 5

Ta

ble

1N

QO

1rs

18

00

56

6(C

60

9T

),P

ON

1rs

66

2(Q

19

2R

),an

dP

ON

1rs

85

45

60

(L5

5M

)g

eno

typ

esfr

equ

enci

esac

cord

ing

tosk

inco

lor

inch

ild

ho

od

acu

tele

uk

emia

,B

razi

l2

00

0–

20

11

Gen

oty

pes

Tota

lW

hit

esN

on-w

hit

es

Cas

es

n(%

)

Contr

ols

n(%

)

OR

(CI

95

%)

Cas

es

n(%

)

Contr

ols

n(%

)

OR

(CI

95

%)

Cas

es

n(%

)

Contr

ols

n(%

)

OR

(CI

95

%)

Cru

de

Adju

sted

bC

rude

Adju

sted

cC

rude

Adju

sted

c

NQ

O1

C609T

rs1800566

CC

325

(59.5

)190

(56.5

)1.0

a1.0

a153

(64.6

)79

(58.5

)1.0

a1.0

a172

(63.7

)111

(59.7

)1.0

a1.0

a

CT

182

(33.3

)131

(39.0

)0.8

0(0

.61–1.0

7)

0.8

0(0

.60–1.0

7)

84

(35.4

)56

(41.5

)0.7

8(0

.50–1.1

9)

0.7

4(0

.46–1.1

4)

98

(36.3

)75

(40.3

)0.8

4(0

.57–1.2

4)

0.7

9(0

.53–1.1

7)

TT

39

(7.2

)15

(4.5

)1.6

9(0

.91–3.1

4)

1.2

3(0

.89–1.7

0)

21

(12.1

)5

(6.0

)2.1

7(0

.79–5.9

7)

1.4

5(0

.87–2.4

2)

18

(9.5

)10

(8.3

)1.1

6(0

.52–2.6

1)

1.0

9(0

.71–1.6

8)

CT

?T

T221

(40.5

)46

(43.5

)0.8

9(0

.68–1.1

6)

0.8

7(0

.66–1.1

5)

105

(40.7

)61

(43.6

)0.8

9(0

.59–1.3

5)

0.8

4(0

.56–1.2

9)

116

(40.3

)85

(43.4

)0.8

8(0

.61–1.2

7)

0.8

3(0

.57–1.2

2)

PO

N1

Q192R

rs662

QQ

96

(40.3

)74

(31.6

)1.0

a1.0

a51

(40.2

)33

(35.1

)1.0

a1.0

a45

(40.5

)41

(29.3

)1.0

a1.0

a

QR

102

(42.9

)106

(45.3

)0.7

8(0

.52–1.1

5)

0.7

7(0

.51–1.1

6)

58

(45.7

)45

(47.9

)0.8

3(0

.46–1.5

0)

0.9

2(0

.50–1.6

9)

44

(39.6

)61

(43.6

)0.6

6(0

.37–1.1

7)

0.6

5(0

.37–1.1

7)

RR

40

(16.8

)54

(23.1

)0.5

9(0

.36–0.9

7)

0.7

7(0

.60–1.0

1)

18

(14.1

)16

(17.0

)0.7

3(0

.33–1.6

3)

0.8

6(0

.58–1.3

0)

22

(19.9

)38

(27.1

)0.5

3(0

.27–1.0

4)

0.7

1(0

.51–1.0

0)

QR

?R

R142

(59.7

)160

(68.4

)0.7

1(0

.49–1.0

3)

0.7

1(0

.49–1.0

5)

76

(59.8

)61

(64.9

)0.8

1(0

.46–1.4

0)

0.8

7(0

.50–1.5

4)

66

(59.5

)99

(70.7

)0.6

1(0

.36–1.0

2)

0.6

0(0

.35–1.0

2)

PO

N1

L55M

rs854560

LL

104

(43.9

)131

(58.2

)1.0

a1.0

a64

(58.2

)53

(64.6

)1.0

a1.0

a40

(43.0

)78

(62.9

)1.0

a1.0

a

LM

99

(41.8

)75

(33.3

)1.6

6(1

.13–2.4

4)

1.7

9(1

.20–2.6

9)

46

(41.8

)29

(35.4

)1.3

1(0

.73–2.3

7)

1.3

3(0

.72–2.4

6)

53

(57.0

)46

(37.1

)2.2

5(1

.30–3.8

9)

2.2

8(1

.31–4.0

0)

MM

34

(14.3

)19

(8.5

)2.2

3(1

.23–4.0

5)

1.5

9(1

.16–2.1

9)

13

(16.9

)6

(10.2

)1.7

9(0

.64–5.0

4)

1.4

0(0

.82–2.4

0)

21

(34.4

)13

(14.3

)3.1

5(1

.43–6.9

4)

1.8

0(1

.21–2.7

0)

LM

?M

M133

(56.1

)94

(41.8

)1.7

8(1

.24–2.5

5)

1.9

3(1

.32–2.8

1)

59

(48.0

)35

(39.8

)1.3

9(0

.80–2.4

3)

1.4

4(0

.80–2.5

8)

74

(64.9

)59

(51.4

)2.4

5(1

.47–4.0

8)

2.5

2(1

.49–4.2

6)

OR

odds

rati

o,

CI

confi

den

cein

terv

als

a1.0

set

asre

fere

nce

bA

dju

sted

by

skin

colo

ran

dag

ec

Adju

sted

by

age

Cancer Causes Control (2012) 23:1811–1819 1813

123

Author's personal copy

Page 6

The Ethics and Scientific Committees of Instituto Nacional

de Cancer approved this study.

DNA purification

Genomic DNA was isolated from peripheral blood cells or

from buccal cells with the QIAamp DNA Blood Mini Kit

(Qiagen�, USA) or with Oragene DNA technology

(Genotek, Ontario, Canada), respectively, according to the

manufacturer’s instructions.

Genotyping

NQO1 and PON1 polymorphisms were detected by allelic

discrimination using TaqMAN� probes (TaqMAN� SNP

Genotyping Assays; Applied Biosystems, Foster City, CA,

USA) using the BioRad C1000 Thermal Cycler (CFX96

Real-Time System). Three different primer/probe systems

in three different reactions were used to detect the NQO1

polymorphism rs1800566 (C609T) and the PON1 poly-

morphisms rs662 (Q192R) and rs854560 (L55M). Poly-

merase chain reactions were set up in a total volume of

10 lL, including 1X TaqMAN� Universal PCR Master-

mix, 1X TaqMAN� SNP Genotyping Assay (Applied

Biosystems), and 10 ng of template DNA. The TaqMan

probes were synthesized and labeled by Applied Biosys-

tems. After an initial denaturation step for 10 min at 95 �C,

each cycle consisted of denaturation for 15 s at 92 �C and

annealing and primer extension for 60 s at 60 �C for a total

45 cycles. The PON1 variant pattern was set up with DNA

from positive controls kindly provided by Rios et al. [22].

Statistical analysis

Allele frequencies were derived by gene counting. The Chi-

square (v2) method was used for comparisons of NQO1 and

PON1 alleles and genotype distributions across population

subgroups. The 95 % confidence intervals (95 % CI) were

determined for pair-wise comparisons of allele frequencies

between subgroups. The v2 test for goodness-of-fit was used

to check whether the distribution of NQO1 and PON1

genotypes in the overall population deviated from the

Hardy–Weinberg equilibrium. First, we analyzed the dis-

tributions of the NQO1 and PON1 genotypes in our study

population. Data were then stratified by skin color (whites

and non-whites), leukemia subtype, and age range (B1 year

old; [1–10 years old; older than 10 years). Data from

children 13–24 months of age were evaluated to test the

effect of the distributions of the NQO1 and PON1 genotypes

in childhood acute leukemia subtypes.

The adjusted odds ratios (ORs) and 95 % CIs of the

association between polymorphism and leukemia risk were

estimated by unconditional binary logistic regression

models after controlling for ethnicity/color classification.

In all cases, wild-type genotypes were used as the reference

category. All statistical analyses were performed using

PASW Statistic 18 with statistical significance defined as

p \ 0.05 (SPSS 18, Chicago, Il). Acquired fusion genes

(MLL-r and ETV6-RUNX1) were also assessed as covari-

ables. For the MLL analyses, the OR was generated from

case–control and case-only (MLL rearranged versus MLL

germ line) approaches for both subgroups of age strata

(0–12 and 13–24 months) and MLL status. Cases were not

stratified by age range for the ETV6/RUNX1 analyses.

Results

All children were B12 years old, with a median age of

51 months at the time of diagnosis of acute leukemia and

identification of control cases. The distributions of cases

and controls by age and gender were comparable, but there

were proportionally fewer whites in the control group than

in the acute leukemia cases (p = 0.01, Supplementary

Table 1). The rates of positivity for the ETV6-RUNX1

fusion gene (16.2 %) and MLL-r (31.4 %) are due to the

identification of infant leukemia across the country by the

Brazilian Collaborative Study Group of Infant Leukemia.

For all subjects, the allele frequencies for NQO1 rs1800566

(C609T), PON1 rs662 (Q192R), and PON1 rs854560

(L55M) were 35, 41, and 34 %, respectively. The NQO1

and PON1 genotype distributions among the controls were

in Hardy–Weinberg equilibrium.

Table 1 contains the genotype frequency distributions of

cases and controls according to skin color, crude, and

adjusted by age; statistically significant differences were

detected in the frequencies of the PON1 genotypes between

cases and controls. On the whole, presence of the PON1

rs854560 (L55M) allele was associated with an increased

risk of developing acute leukemia (LM ? MM, OR 1.93,

95 % CI 1.32–2.81), even when adjusted for skin color and

age. When data were stratified by skin color and adjusted

by age, we observed that the presence of at least one var-

iant M allele increased the risk of developing leukemia by

2.5-fold in non-white children. We tested whether Brazil-

ian regional differences in the population accounted for the

PON1 genotype distributions (Supplementary Table 2).

The presence of the PON1 M allele was increased among

acute leukemia cases in the Northeast (PON1 rs854560

LM ? MM, OR 1.91, 95 % CI 1.13–3.23) and Middle East

(OR 5.74, 95 % CI 1.59–20.7) regions when the data were

adjusted for skin color.

The distributions of the NQO1 rs1800566 (C609T)

allele in childhood ALL, AML, and control cases accord-

ing to age range and adjusted by skin color are shown in

1814 Cancer Causes Control (2012) 23:1811–1819

123

Author's personal copy

Page 7

Table 2. Children with at least one NQO1 variant allele

were at a lower risk of developing infant AML (OR 0.26,

95 % CI 0.10–0.68), whereas no association was observed

for ALL. An increased crude risk was detected in children

older than 1 year, although this association lacked statis-

tical significance (OR 1.83, 95 % CI 0.82–4.06).

The distributions of PON1 rs662 (Q192R) and PON1

rs854560 (L55M) in ALL, AML, and control cases

according to age range and adjusted by skin color are given

in Table 3. The PON1 rs662 R192R genotype was signifi-

cantly decreased in ALL cases in children older than 1 year

and less than 10 years of age (OR 0.64, 95 % CI 0.43–0.93).

The same association was observed with the sum of the

PON1 rs662 (QR ? RR) genotypes (OR 0.57, 95 % CI

0.33–0.97). Children with ALL were prone to harbor the

PON1 rs85450 (L55M) variant; infant ALL cases were

more likely to be PON1 rs854560 homozygous than controls

(OR 1.72, 95 % CI 1.03–2.89), and in ALL cases

[1–10 years of age, at least one M allele was associated

with an increased risk of developing ALL (OR 1.99, 95 % CI

1.17–3.3). Further, we tested the differences in the genetic

susceptibility in early childhood (younger than 24 months)

by evaluating the distribution of the NQO1 rs1800566

(C609T), PON1 rs662 (Q192R), and PON1 rs854560

(L55M) genotypes in children aged 13–24 months (Sup-

plementary Table 3). The presence of at least one variant

allele of PON1 rs854560 (L55M) was associated with an

increased risk of developing ALL (OR 2.47, 95 % CI

1.07–5.69) when the data were adjusted by skin color.

Since infants and young children (\24 months of age) have

been associated with an increased risk of acute leukemia with

MLL-r, we tested the distribution of genotypes for age ranges

B12 months and 13–24 months (Table 4) adjusted by skin

color. A protective effect of the NQO1 rs1800566 (C609T)

variant was found for infants with acute leukemia with MLL in

the germ line (OR 0.36, 95 % CI 0.16–0.81). NQO1

rs1800566 (C609T) was associated with an increased risk of

acute leukemia in children older than 13–24 without MLL-r

(OR 2.36, 95 % CI 1.02–5.72). The NQO1 rs1800566

(C609T) variant allele was not significantly associated with

the ETV6-RUNX1 fusion gene in common ALL (OR 0.89,

95 % CI 0.64–1.22; data not shown). Synergistic effects of the

NQO1 rs1800566 (C609T), PON1 rs662 (Q192R), and

rs854560 (L55M) alleles were not significantly associated

with acute leukemia (data not shown).

Discussion

We assessed the genotype frequencies of the NQO1

rs1800566 (C609T) and PON1 rs662 (Q192R) and rs854560

(L55M) variants in a series of Brazilian children with an

emphasis on early infancy. Adding genetic polymorphismTa

ble

2N

QO

1rs

18

00

56

6(C

60

9T

)g

eno

typ

esd

istr

ibu

tio

nin

chil

dh

oo

dac

ute

leu

kem

iasu

bty

pes

acco

rdin

gto

age

ran

ges

Ag

e

(yea

rs)

Gen

oty

pe

Co

ntr

ols

n(%

)

AL

n(%

)

OR

(95

%C

I)

Cru

de

OR

(95

%C

I)

Ad

just

ed*

AL

L

n(%

)

OR

(95

%C

I)

Cru

de

OR

(95

%C

I)

Ad

just

ed

AM

L

n(%

)

OR

(95

%C

I)

Cru

de

OR

(95

%C

I)

Ad

just

ed*

B1

CC

52

(52

.0)

75

(61

.5)

1.0

a1

.0a

50

(55

.6)

1.0

a1

.0a

25

(78

.1)

1.0

a1

.0a

CT

43

(43

.0)

43

(35

.2)

0.6

9(0

.40

–1

.20

)0

.61

(0.3

4–

1.1

1)

36

(40

.0)

0.8

7(0

.48

–1

.57

)0

.77

(0.4

1–

1.4

4)

7(2

1.9

)0

.34

(0.1

3–

0.8

6)

0.2

9(0

.11

–0

.76

)

TT

5(5

.0)

4(3

.3)

0.5

5(0

.14

–2

.16

)0

.67

(0.3

2–

1.3

9)

4(4

.4)

0.8

3(0

.21

–3

.27

)0

.82

(0.3

9–

1.7

1)

0–

–

CT

?T

T4

8(4

8.0

)4

7(3

8.5

)0

.67

(0.4

0–

1.1

6)

0.6

0(0

.34

–1

.06

)4

0(4

4.4

)0

.87

(0.4

9–

1.5

4)

0.7

6(0

.41

–1

.40

)7

(21

.9)

0.3

0(0

.12

–0

.77

)0

.26

(0.1

0–

0.6

8)

[1

–1

0C

C1

26

(57

.5)

21

6(6

0.8

)1

.0a

1.0

a1

84

(59

.0)

1.0

a1

.0a

32

(74

.4)

1.0

a1

.0a

CT

84

(38

.3)

11

2(3

1.5

)0

.78

(0.5

4–

1.1

1)

0.8

1(0

.56

–1

.17

)1

04

(33

.3)

0.8

5(0

.59

–1

.22

)0

.88

(0.6

1–

1.2

9)

8(1

8.6

)0

.38

(0.1

6–

0.8

5)

0.4

1(0

.18

–0

.95

)

TT

9(4

.2)

27

(7.7

)1

.75

(0.8

0–

3.8

4)

1.3

1(0

.88

—1

.93

)2

4(7

.7)

1.8

3(0

.82

–4

.06

)1

.33

(0.9

0–

2.0

0)

3(7

.0)

1.3

1(0

.34

–5

.13

)1

.16

(0.5

8–

2.3

0)

CT

?T

T9

3(4

2.5

)1

39

(39

.2)

0.8

7(0

.62

–1

.23

)0

.91

(0.6

4–

1.2

9)

12

8(4

1.0

)0

.94

(0.6

6–

1.3

4)

0.9

8(0

.68

–1

.40

)1

1(2

5.6

)0

.47

(0.2

2–

0.9

7)

0.5

1(0

.24

–1

.08

)

[1

0C

C3

0(6

1.2

)4

0(5

1.9

)1

.0a

1.0

a3

2(5

0.8

)1

.0a

1.0

a8

(57

.2)

1.0

a1

.0a

CT

19

(38

.8)

29

(37

.7)

1.1

4(0

.54

–2

.42

)1

.85

(0.7

7–

4.4

5)

26

(41

.3)

1.2

8(0

.59

–2

.78

)1

.05

(0.4

3–

2.5

9)

3(2

1.4

)0

.59

(0.1

4–

2.5

1)

0.5

7(0

.12

–2

.63

)

TT

08

(10

.4)

––

5(7

.9)

––

3(2

1.4

)–

–

CT

?T

T1

9(3

8.8

)3

7(4

8.1

)1

.46

(0.7

1–

3.0

2)

1.1

8(0

.50

–2

.78

)3

1(4

9.2

)1

.53

(0.7

2–

3.2

6)

1.2

4(0

.52

–2

.99

)6

(42

.8)

1.1

8(0

.35

–3

.94

)1

.14

(0.3

1–

4.2

0)

*A

dju

sted

by

skin

colo

ra

1.0

set

asre

fere

nce

;A

LL

Acu

teL

ym

ph

ob

last

icL

euk

emia

;A

ML

Acu

teM

yel

oid

Leu

kem

ia;

OR

od

dra

tio

;C

Ico

nfi

den

cein

terv

al

Cancer Causes Control (2012) 23:1811–1819 1815

123

Author's personal copy

Page 8

Ta

ble

3P

ON

1rs

66

2(Q

19

2R

)an

dP

ON

1rs

18

00

56

6(L

55

M)

gen

oty

pes

freq

uen

cies

inch

ild

ho

od

acu

tele

uk

emia

sub

typ

esac

cord

ing

toag

era

ng

e

Ag

e

(yea

rs)

Gen

oty

pe

Con

tro

ls

n(%

)

AL n(%

)

OR

(95

%C

I)

cru

de

OR

(95

%C

I)

adju

sted

AL

L

n(%

)

OR

(95

%C

I)

cru

de

OR

(95

%C

I)

adju

sted

AM

L

n(%

)

OR

(95

%C

I)

Cru

de

Ad

j.O

R

(95

%C

I)

adju

sted

B1

rs6

62

QQ

31

(32

.3)

39

(39

.4)

1.0

a1

.0a

29

(42

.0)

1.0

a1

.0a

10

(33

.3)

1.0

a1

.0a

QR

42

(43

.7)

44

(44

.4)

0.8

3(0

.44

–1

.57

)0

.91

(0.4

6–1

.78

)2

9(4

2.0

)0

.74

(0.3

7–1

.48

)0

.73

(0.3

5–1

.54

)1

5(5

0.0

)1

.11

(0.4

4–2

.79

)1

.40

(0.5

2–3

.81

)

RR

23

(24

.0)

16

(16

.2)

0.5

5(0

.25

–1

.22

)0

.80

(0.5

3–1

.21

)1

1(1

6.0

)0

.51

(0.2

1–1

.23

)0

.74

(0.4

7–1

.18

)5

(16

.7)

0.6

7(0

.20

–2

.24

)0

.93

(0.5

0–1

.75

)

QR

?R

R6

5(6

7.7

)6

0(6

0.6

)0

.73

(0.4

1–1

.32

)0

.81

(0.4

3–1

.52

)4

0(5

8.0

)0

.66

(0.3

5–1

.25

)0

.67

(0.3

4–1

.33

)2

0(6

6.7

)0

.95

(0.4

0–2

.28

)1

.21

(0.4

7–3

.10

)

rs8

54

56

0

LL

50

(54

.9)

43

(44

.8)

1.0

a1

.0a

30

(45

.5)

1.0

a1

.0a

13

(43

.3)

1.0

a1

.0a

LM

33

(36

.3)

37

(38

.5)

1.3

0(0

.70

–2

.42

)1

.56

(0.8

0–3

.05

)2

6(3

9.4

)1

.31

(0.6

6–2

.60

)1

.47

(0.7

0–3

.09

)1

1(3

6.7

)1

.28

(0.5

1–3

.20

)1

.62

(0.6

2–4

.21

)

MM

8(8

.8)

16

(16

.7)

2.3

2(0

.91

–5

.96

)1

.72

(1.0

3–2

.89

)1

0(1

5.1

)2

.08

(0.7

4–5

.86

)1

.72

(0.9

8–3

.03

)6

(20

.0)

2.8

8(0

.85

–9

.79

)1

.87

(0.9

2–3

.78

)

LM

?M

M4

1(4

5.1

)5

3(5

5.2

)1

.50

(0.8

4–2

.68

)1

.83

(0.9

8–3

.42

)3

6(5

4.5

)1

.46

(0.7

7–2

.77

)1

.75

(0.8

7–3

.49

)1

7(5

6.7

)1

.59

(0.6

9–3

.66

)1

.95

(0.8

0–4

.76

)

[1

–1

0rs

66

2

QQ

42

(32

.6)

55

(42

.0)

1.0

a1

.0a

48

(44

.9)

1.0

a1

.0a

7(2

9.2

)1

.0a

1.0

a

QR

55

(42

.6)

54

(41

.2)

0.7

5(0

.43

–1

.30

)0

.71

(0.4

1–1

.25

)4

3(4

0.2

)0

.68

(0.3

8–1

.22

)0

.66

(0.3

7–1

.18

)1

1(4

5.8

)1

.20

(0.4

3–3

.36

)1

.15

(0.4

1–3

.26

)

RR

32

(24

.8)

22

(16

.8)

0.5

3(0

.27

–1

.03

)0

.71

(0.5

0–1

.01

)1

6(1

4.9

)0

.44

(0.2

1–0

.91

)0

.64

(0.4

3–0

.93

)6

(25

.0)

1.1

2(0

.34

–3

.67

)1

.17

(0.6

3–2

.18

)

QR

?R

R8

7(6

7.4

)7

6(5

8.0

)0

.67

(0.4

0–1

.11

)0

.64

(0.3

8–1

.07

)5

9(5

5.1

)0

.59

(0.3

5–1

.01

)0

.57

(0.3

3–0

.97

)1

7(7

0.8

)1

.17

(0.4

5–3

.04

)1

.21

(0.4

6–3

.19

)

rs8

54

56

0

LL

74

(59

.7)

58

(43

.6)

1.0

a1

.0a

47

(43

.1)

1.0

a1

.0a

11

(45

.8)

1.0

a1

.0a

LM

40

(32

.3)

57

(42

.8)

1.8

2(1

.07

–3

.09

)1

.91

(1.1

1–3

.28

)4

6(4

2.2

)1

.81

(1.0

3–3

.17

)1

.89

(1.0

7–3

.34

)1

1(4

5.8

)1

.85

(0.7

4–4

.64

)1

.92

(0.7

5–4

.88

)

MM

10

(8.0

)1

8(1

3.6

)2

.30

(0.9

9–5

.35

)1

.52

(0.9

9–2

.34

)1

6(1

4.7

)2

.51

(1.0

5–6

.01

)1

.56

(1.0

1–2

.43

)2

(8.4

)1

.35

(0.2

6–6

.97

)1

.18

(0.5

1–2

.74

)

LM

?M

M5

0(4

0.3

)7

5(5

6.4

)1

.91

(1.1

7–3

.14

)1

.97

(1.1

8–3

.27

)6

2(5

6.9

)1

.95

(1.1

6–3

.29

)1

.99

(1.1

7–3

.39

)1

3(5

4.2

)1

.75

(0.7

5–4

.21

)1

.80

(0.7

4–4

.39

)

[1

0rs

66

2

QQ

8(3

8.1

)2

(22

.2)

1.0

a1

.0a

1(1

4.2

)1

.0a

1.0

a1

(50

.0)

1.0

a1

.0a

QR

10

(47

.6)

4(4

4.4

)1

.60

(0.2

3–1

1.0

8)

0.8

0(0

.07

–8

.47

)3

(42

.9)

2.4

0(0

.21

–2

7.7

2)

1.2

(0.0

7–1

9.6

3)

1(5

0.0

)0

.80

(0.0

4–1

4.8

8)

0.8

2(0

.23

–1

7.8

2)

RR

3(1

4.3

)3

(33

.4)

4.0

0(0

.43

–3

7.1

1)

1.6

8(0

.43

–5

.85

)3

(42

.9)

8.0

0(0

.58

–1

10

.27

)2

.44

(0.5

7–1

0.4

3)

0–

–

QR

?R

R1

3(6

1.9

)7

(77

.8)

2.1

5(0

.36

–1

3.0

5)

1.3

4(0

.18

–1

0.2

9)

6(8

5.8

)3

.69

(0.3

7–3

6.5

7)

2.3

3(0

.20

–2

7.5

7)

1(5

0.0

)0

.61

(0.0

3–1

1.2

8)

0.7

1(0

.06

–9

.28

)

rs8

54

56

0

LL

15

(65

.2)

3(3

3.3

)1

.0a

1.0

a3

(42

.9)

1.0

a1

.0a

01

.0a

1.0

a

LM

5(2

1.7

)5

(55

.5)

5.0

0(0

.87

–2

8.8

6)

3.2

9(0

.51

–2

1.4

8)

3(4

2.9

)3

.00

(0.4

5–1

9.9

3)

2.0

0(0

.27

–1

4.8

6)

2(1

00

.0)

––

MM

3(1

3.1

)1

(11

.2)

1.6

7(0

.13

–2

2.0

0)

1.6

6(0

.33

–8

.31

)1

(14

.2)

1.6

7(0

.13

–2

2.0

0)

1.6

6(0

.33

–8

.31

)0

––

LM

?M

M8

(34

.8)

6(6

6.7

)3

.75

(0.7

3–1

9.1

4)

3.3

0(0

.54

–2

0.1

4)

4(5

7.1

)2

.50

(0.4

5–1

4.0

4)

2.2

1(0

.33

–1

4.4

0)

2(1

00

.0)

––

Ad

just

edb

ysk

inco

lor

AL

Lac

ute

lym

ph

ob

last

icle

ukem

ia,

AM

Lac

ute

my

elo

idle

ukem

ia,

OR

od

dra

tio

,C

Ico

nfi

den

cein

terv

ala

1.0

set

asre

fere

nce

1816 Cancer Causes Control (2012) 23:1811–1819

123

Author's personal copy

Page 9

analysis to childhood leukemia framework studies is greatly

beneficial because gene variants reflect different levels of

association risk within distinct leukemia pathways. Since

young children are more vulnerable than adults to the effect

of xenobiotics, we examined the risk of childhood acute

leukemia (ALL and AML) associated with genes that con-

tribute to the metabolism of substances that may be involved

in the etiology of childhood acute leukemia. To our

knowledge, no previous study has explored the effect of

PON1 polymorphisms on childhood leukemia. Our obser-

vations are in agreement with the NQO1 and PON1 distri-

butions reported by other series of Brazilians and Hispanic

genotyped populations [22, 23]. Throughout our analysis,

we adjusted the relatively non-homogeneous Brazilian eth-

nic/racial populations for skin color in both cases and con-

trols. Differences between cases and controls were observed

in non-white children harboring the PON1 rs854560 L55M

variant genotype. As a whole, the association between

PON1 rs854560 L55M and childhood leukemia remained

when the data were adjusted by skin color and age.

In the present report, no significant association was

detected between NOQ1 s1800566 and ALL risk. The

association between NOQ1 (C609T) and ALL was

described by Krajinovic et al. in a relatively homogeneous

Quebec population (restricted to white patients) [3]; we

detected this effect among white Brazilians, although it

was not statistically significant (Table 1; OR 2.17, 95 % CI

0.79–5.97). In another series of ALL patients carrying

distinct fusion genes (MLL/AF4, ETV6/RUNX1, and BCR/

ABL), the NQO1 rs1800566 (C609T) polymorphism did

not confer an increased risk of childhood ALL [24]. We did

not detect associations with MLL/AF4 and ETV6-RUNX1 in

a subset of ALL cases. According to some studies (mainly

of subjects of European white ancestry), lower NQO1

activity was associated with an increased risk of infant

ALLs carrying MLL/AF4 fusion genes [16, 25]. Therefore,

conflicting issues remain regarding the association of

NQO1 rs1800566 and the risk of childhood ALL.

Following adjustment for skin color, we identified a

significant protective association between the NOQ1

rs1800566 (C609T) variant genotype and infant AML.

Sirma et al. reported a borderline association between

AML and NQO1 variant allele [23]. Nevertheless, a recent

meta-analysis revealed no evidence of association between

the NQO1 polymorphism and childhood AML [26–28]. We

uncovered a protective association in AML in children

older than 12 months with one NQO1 rs1800566 (C609T)

variant allele (OR 0.41, 95 % CI 95 % 0.18–0.95).

NQO1 rs1800566 heterozygous individuals (C/T) have

intermediate enzyme activity, and homozygotes for the

variant allele (T/T) are deficient in NQO1 activity. NQO1

reduces quinones to hydroquinones in a single two-electron

step, bypassing potentially toxic semiquinone radical

intermediates. Not all hydroquinones are chemically stable,

and in some cases, metabolism by NQO1 yields a more

active product that autoxidizes to produce reactive oxygen

species or causes DNA rearrangement [29]. This dual

function may explain the conflicting reports of NQO1 and

the lower risk associated with AML in the present study.

According to some studies, the NQO1 polymorphism does

not appear to confer an increase risk of childhood leuke-

mia, but this effect may be modified by the presence of

other gene variants in xenobiotic systems [27–29].

Since PON1 detoxifies organophosphorus, a risk factor for

childhood leukemia [30], we have speculated that PON1

variant genotype would increase the risk of developing leu-

kemia. Regarding the two PON1 polymorphisms, we

observed opposite results in relation to ALL. As long as PON1

rs662 (Q192R) was associated with a decreased risk of ALL,

at least one M allele of PON1 rs854560 (L55M) increased the

risk of ALL, but this effect was not found in AML. PON1

rs854560 (L55M) is associated with susceptibility to

Table 4 NQO1 rs1800566 (C609T) genotypes distribution according to age ranges and MLL rearrangement in acute lymphoblastic leukemia

Age

(months)

NQO1Genotype

MLL-r

N (%)

MLL-GL

N (%)

Controls

N (%)

Controls 9 MLL-r

OR (95 % CI) Crude

OR (95 % CI)

Adjusted

Controls 9 MLL-GL

OR (95 % CI) crude

OR (95 % CI)

adjusted

B12 CC 33 (55.0) 32 (71.0) 52 (52.0) 1.0a 1.0a 1.0a 1.0a

CT 24 (40.0) 12 (26.7) 43 (43.0) 0.88 (0.45–1.71) 0.81 (0.41–1.62) 0.45 (0.21–0.98) 0.37 (0.16–0.87)

TT 3 (5.0) 1 (2.3) 5 (5.0) 0.95 (0.21–4.22) 0.89 (0.40–1.97) 0.32 (0.04–2.91) 0.49 (0.15–1.56)

CT ? TT 27 (45.0) 13 (29.0) 48 (48.0) 0.88 (0.47–1.68) 0.82 (0.42–1.59) 0.44 (0.21–0.94) 0.36 (0.16–0.81)

CC 21 (65.6) 24 (58.5) 46 (76.6) 1.0a 1.0a 1.0a 1.0a

[12–24 CT 7 (21.8) 13 (31.7) 12 (20.0) 1.28 (0.44–3.71) 1.28 (0.44–3.77) 2.08 (0.82–5.25) 2.14 (0.83–5.56)

TT 4 (12.6) 4 (9.8) 2 (3.4) 4.38 (0.74–25.82) 2.22 (0.89–5.49) 3.83 (0.65–22.45) 1.88 (0.76–4.65)

CT ? TT 11 (34.4) 17 (41.5) 14 (23.4) 1.72 (0.67–4.42) 1.77 (0.68–4.57) 2.33 (0.98–5.51) 2.36 (1.02–5.72)

ORs were adjusted by skin color

MLL-r, MLL rearranged; MLL-GL, MLL germ line; OR odds ratio, CI confidence intervala 1.0 set as reference

Cancer Causes Control (2012) 23:1811–1819 1817

123

Author's personal copy

Page 10

childhood acute leukemia with discrimination between ALL

and AML. Furthermore, the magnitude of ALL risk was

modified according to age group older than 1 year.

There is a huge range of the PON1 activity variability in

young children, being particularly low in newborns. It is

known that children with the PON1192 R allele experience

pronounced, increased activity over time [15]; the R allo-

enzyme hydrolyzes paraoxon and fenitroxon substrates

more rapidly than the Q alloenzyme, and the PON155 M

allele does not affect the catalytic efficiency of substrate

hydrolysis by the enzyme, but it is correlated with

decreased mRNA. Therefore, PON1 gene exhibited a wide

variation in enzyme activities both within and between

genotypes which implied insights into a potential differ-

ence in sensitivity to organophosphorus toxicity [31].

PON1 enzyme activity and level vary broadly in humans,

and the determinants of PON1 variation, including genetics

and age, may further explain the role this enzyme plays in

relation to exposure and malignant diseases [32, 33].

The current investigation has some limitations. First, the

small numbers of children with acute leukemia subtype

(e.g., AML) stratified by different fusion genes and geno-

types (Type II error) generated insufficient statistical power

to detect significant differences. Second, the selection of

controls may have been biased, as the biological samples

were obtained from hospitalized children without malig-

nant disorders; these children could not represent the

general population. Other limitations of the study were our

inability to analyze PON1 activity measurements. Mean-

while, it should be highlighted that this is the first report

that found the association of genetic variability in PON1

gene that determines susceptibility risk to lymphoid or

myeloid leukemia in childhood. The fact that this study

included individuals from different Brazilian geographical

regions in which the analysis were adjusted by skin color

confers greater reliability on the results. Also, the differ-

ences in the mean ages of patients and controls are fully

comparable and should not influence our findings.

In conclusion, the present data suggest that NQO1

rs1800566 (C609T) confers protection against the risk of

AML. PON1 rs854560 (L55M) was associated with

increased risk factors by age and ALL subset, whereas

PON1 rs662 (Q192R) conferred a decreased risk to ALL.

Therefore, further analysis with haplotypes, enzyme activ-

ities, and environmental exposition should be performed to

clarify how the low-penetrance metabolic genes act in the

mechanistic pathway leading to leukemia in childhood.

Acknowledgments This investigation was supported by the Bra-

zilian National Research Council (CNPq), Instituto Nacional de

Cancer and Fundacao do Cancer Ary Frauzino. MSPO is supported by

CNPq research scholarship #309091/2007. The project was granted

by INCT-Controle do Cancer; CNPq #573806/2008-0 and FAPERJ

E026/2008. We thank Dr. Davy Rapozo for helping with the setup of

the technical procedures. We are grateful to Dr. Joseph L. Wiemels

for helpful comments and reviewing the manuscript.

Conflict of interest All authors disclose that no financial or per-

sonal relationships with other individuals or organizations have

inappropriately influenced this study.

Appendix: Brazilian Collaborative Study Group

of infant acute leukemia that contributed to the study

and listed as co-authors

Mariana Emerenciano1, Beatriz de Camargo1, Luna Bern-

stain1, Cynthia Curvello Neves2, Jozina Maria de Andrade

Agareno2, Lilian Maria Burlacchini de Carvalho3, Flavia

Nogueira Serafim Araujo2, Nilma Pimentel de Brito3, Isis

Q. Magalhaes4, Jose Carlos Cordoba4, Flavia Pimenta5,

Andreia Gadelha5, Eloısa Cartaxo5, Rosania Maria Basegio7,

Atalla Mnayarji7, Marcelo S. Souza7, Alejandro Arencibia8,

Renato Melaragno8, Virgınia Maria Coser9,Thereza Chris-

tina Lafayete9, Sergio Koifman10.

Affiliations: 1. Research Center of Instituto Nacional de

Cancer, Rio de Janeiro; 2. Sociedade de Oncologia da Bahia,

Salvador-Bahia; 3. Hospital Martagao Gesteira, Salvador-

Bahia; 4. Hospital de Apoio Brasılia, Unidade de Onco-

Hematologia Pediatrica, Brasılia, DF; 5. Hospital Napoleao

Laureano, Joao Pessoa, Paraıba; 6. Departamento de Pe-

diatria, Faculdade de Medicina,Universidade Federal de

Minas Gerais, Belo Horizonte, MG; 7. Hospital Rosa Ped-

rossian, Campo Grande, MS; 8. Hospital Santa Marcelina,

Sao Paulo, Sao Paulo; 9. Departamento de Pediatria, Fac-

uldade de Medicina, Universidade Federal de Santa Maria,

Santa Maria, Rio Grande do Sul; 10. Escola Nacional de

Saude Publica, FIOCRUZ, Rio de Janeiro, Brazil.

References

1. Belson M, Kingsley B, Holmes A (2007) Risk factors for acute

leukemia in children: a review. Environ Health Perspect

115:138–145

2. Greaves M (1999) Molecular genetics, natural history and the

demise of childhood leukaemia. Eur J Cancer 35:1941–1953

3. Krajinovic M, Sinnett H, Richer C, Labuda D, Sinnett D (2002)

Role of NQO1, MPO and CYP2E1 genetic polymorphisms in the

susceptibility to childhood acute lymphoblastic leukemia. Int J

Cancer 97:230–236

4. Bolufer P, Barragan E, Collado M, Cervera J, Lopez JA, Sanz

MA (2006) Influence of genetic polymorphisms on the risk of

developing leukemia and on disease progression. Leuk Res

30:1471–1491

5. Greaves MF, Wiemels J (2003) Origins of chromosome translo-

cations in childhood leukaemia. Nat Rev Cancer 3:639–649

6. Ross JA, Potter JD, Reaman GH, Pendergrass TW, Robison LL

(1996) Maternal exposure to potential inhibitors of DNA topoi-

somerase II and infant leukemia (United States): a report from the

Children’s Cancer Group. Cancer Causes Control 7:581–590

1818 Cancer Causes Control (2012) 23:1811–1819

123

Author's personal copy

Page 11

7. Alexander FE, Patheal SL, Biondi A et al (2001) Transplacental

chemical exposure and risk of infant leukemia with MLL gene

fusion. Cancer Res 61:2542–2546

8. Pombo-de-Oliveira MS, Koifman S et al (2006) Infant acute

leukemia and maternal exposures during pregnancy. Cancer Ep-

idemiol Biomarkers Prev 15:2336–2341

9. Greaves M (2006) Infection, immune responses and the aetiology

of childhood leukaemia. Nat Rev Cancer 6:193–203

10. Simpson J, Smith A, Ansell P, Roman E (2007) Childhood leu-

kaemia and infectious exposure: a report from the United Kingdom

Childhood Cancer Study (UKCCS). Eur J Cancer 43:2396–2403

11. Ferreira JD, Couto AC, Pombo-de-Oliveira MS, Koifman S et al.

(2011) Pregnancy, exposure to pesticides and infant leukemia in

Brazil. Environ Health Perspect. doi:http://dx.doi.org/10.1289/

ehp.isee2011

12. Vasiliou V, Ross D, Nebert DW (2006) Update of the

NAD(P)H:quinone oxidoreductase (NQO) gene family. Hum

Genomics 2:329–335

13. Rothman N, Smith MT, Hayes RB et al (1997) Benzene poi-

soning, a risk factor for hematological malignancy, is associated

with the NQO1 609C- [ T mutation and rapid fractional excre-

tion of chlorzoxazone. Cancer Res 57:2839–2842

14. Costa LG, Vitalone A, Cole TB, Furlong CE (2005) Modulation

of paraoxonase (PON1) activity. Biochem Pharmacol 69:541–550

15. Cole TB, Jampsa RL, Walter BJ et al (2003) Expression of

human paraoxonase (PON1) during development. Pharmacoge-

netics 13:357–364

16. Smith MT, Wang Y, Skibola CF, et al. (2002) Low

NAD(P)H:quinone oxidoreductase activity is associated with

increased risk of leukemia with MLL translocations in infants and

children. Blood 100:4590–4593

17. Perera FP (1999) Molecular epidemiologic research on the effects

of environmental pollutants on the fetus. Environ Health Perspect

107(Suppl 3):451–460

18. Huen K, Harley K, Brooks J et al (2009) Developmental changes

in PON1 enzyme activity in young children and effects of PON1

polymorphisms. Environ Health Perspect 117:1632–1638

19. Pombo de Oliveira MS, Koifman S, Vasconcelos GM, Emeren-

ciano M et al (2009) Development and perspective of current

Brazilian studies on the epidemiology of childhood leukemia.

Blood Cells Mol Dis 42:121–125

20. So CWE (2009) Leukemia Methods and Protocols. Methods in

molecular biology. Humana, London, pp 538–544

21. Pena SD, Di Pietro G, Fuchshuber-Moraes M et al (2011) The

genomic ancestry of individuals from different geographical regions

of Brazil is more uniform than expected. PLoS ONE 6(2):e17063

22. Rios DL, D’Onofrio LO, Cerqueira CC et al (2007) Paraoxonase

1 gene polymorphisms in angiographically assessed coronary

artery disease: evidence for gender interaction among Brazilians.

Clin Chem Lab Med 45:874–878

23. da Silva Silveira V, Canalle R, Scrideli CA et al (2009) Poly-

morphisms of xenobiotic metabolizing enzymes and DNA repair

genes and outcome in childhood acute lymphoblastic leukemia.

Leuk Res 33:898–901

24. Kracht T, Schrappe M, Strehl S et al (2004) NQO1 C607T

polymorphism in distinct entities of pediatric hematological

neoplasms. Haematologica 89:1492–1497

25. Wiemels JL, Pagnamenta A, Taylor GM, Eden OB, Alexander

FE, Greaves MF (1999) A lack of a functional NAD(P)H: qui-

none oxidoreductase allele is selectively associated with pediatric

leukemias that have MLL fusions. United Kingdom Childhood

Cancer Study Investigators. Cancer Res 59:4095–4099

26. Sirma S, Agaoglu L, Yildiz I et al (2004) NAD(P)H:quinone

oxidoreductase 1 null genotype is not associated with pediatric de

novo acute leukemia. Pediatr Blood Cancer 43:568–570

27. Guha N, Chang JS, Chokkalingam AP, Wiemels JL, Smith MT,

Buffler PA (2008) NQO1 polymorphisms and de novo childhood

leukemia: a HuGE review and meta-analysis. Am J Epidemiol

168:1221–1232

28. Nebert DW, Roe AL, Vandale SE, Bingham E, Oakley GG

(2002) NAD(P)H:quinone oxidoreductase (NQO1) polymor-

phism, exposure to benzene, and predisposition to disease: a

HuGE review. Genet Med 4:62–70

29. Gaikwad NW, Rogan EG, Cavalieri EL (2007) Evidence from

ESI-MS for NQO1-catalyzed reduction of estrogen ortho-qui-

nones. Free Radic Biol Med 43:1289–1298

30. Buffler PA, Gunier RB, Dahl G, Smith MT, Reinier K, Reynolds

P (2002) Critical windows of exposure to household pesticides

and risk of childhood leukemia. Environ Health Perspect 110:

955–960

31. Sirivarasai J, Kaojarern S, Yoovathaworn K, Sura T (2007)

Paraoxonase (PON1) polymorphism and activity as the determi-

nants of sensitivity to organophosphates in human subjects. Chem

Biol Interact 168:184–192

32. Nielsen SS, Mueller BA, De Roos AJ, Viernes HMA, Farin FM,

Checkoway H (2005) Risk of brain tumors in children and sus-

ceptibility to organophosphorus insecticides: the potential role of

paraoxonase (PON1). Environ Health Perspect 113:909–913

33. Martınez C, Molina JA, Alonso-Navarro H, Jimenez-Jimenez FJ,

Agundez JA, Garcıa-Martın E (2010) Two common nonsynony-

mous paraoxonase 1 (PON1) gene polymorphisms and brain

astrocytoma and meningioma. BMC Neurol 19:10

Cancer Causes Control (2012) 23:1811–1819 1819

123

Author's personal copy