Mutation Research 544 (2003) 243–254
Micronuclei frequency in children exposed to environmental mutagens: a review Monica Neri a , Aleksandra Fucic b , Lisbeth E. Knudsen c , Cecilia Lando a , Franco Merlo a , Stefano Bonassi a,∗ a
c
Environmental Epidemiology and Biostatistics, National Cancer Research Institute, Largo R. Benzi 10, 16132 Genoa, Italy b Institute for Medical Research and Occupational Health, P.O. Box 291, Zagreb, Croatia Institute of Public Health, University of Copenhagen, Blegdamsvej 3, DK-2200 N, Copenhagen, Denmark Received 5 May 2003; received in revised form 11 June 2003; accepted 12 June 2003
Abstract Cytogenetic monitoring has been traditionally used for the surveillance of populations exposed to genotoxic agents. In recent years sensitivity problems emerged in surveys of populations exposed to low levels of mutagens, and therefore alternative approaches have been explored. Biomonitoring studies in children are a promising field, since because of evident differences in the uptake, metabolism, distribution and excretion of mutagens this population seems to be more susceptible than adults. Further, the effect of major confounders such as cigarettes smoking, occupation, life-style, and dietary factors plays a minor role. Among cytogenetic assays, the micronucleus assay (MN) has several advantages and is increasingly used. A review was then carried out to synthesize the published data on the occurrence of MN in children and adolescents (age range 0–18 years), and to assess the impact of genotoxic exposure on MN frequency. Overall, 20 papers from international literature and 8 Russian papers were included. An effect of age was found within this age range, while the influence of gender on MN frequency was irrelevant. These results were confirmed by the re-analysis of data for 448 children selected from the HUMN database. An effect of chronic and infectious diseases on MN levels has been reported by various authors. Most studies describing the effect of exposure to genotoxic agents (ionizing radiation, chemicals, drugs, environmental tobacco smoke) found an increase of MN in exposed children. The limited number of published papers indicates that the conduct of properly designed studies on the effect of environmental pollutants in children may be difficult. This review confirmed the usefulness of MN assay in biomonitoring studies conducted in children, revealing that in many circumstances investigating children increases the sensitivity of the study, even with low dose exposures. © 2003 Elsevier B.V. All rights reserved. Keywords: Children; Genotoxicity; Micronucleus test; Review; Molecular epidemiology
1. Introduction
∗ Corresponding author. Tel.: +39-010-5600924; fax: +39-010-5600501. E-mail address:
[email protected] (S. Bonassi).
Cytogenetic monitoring has been traditionally used for the surveillance of populations exposed to environmental mutagens including ionizing radiation (IR), chemicals and life-style factors, and to medical treatments. The validity of these studies has been rein-
1383-5742/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.mrrev.2003.06.009
244
M. Neri et al. / Mutation Research 544 (2003) 243–254
forced by recent findings from European cohort studies, which have validated chromosomal aberrations as predictors of cancer risk, supporting their use in populations exposed to genotoxic agents [1,2]. Children, as developing individuals, may express increased susceptibility to environmental hazards due to differences in the uptake, metabolism, distribution and excretion of mutagens. This hypothesis is mostly supported by epidemiological findings, reporting increasing trends of cancer incidence in children, and by experimental data showing an increased risk of cancer in rodents exposed in utero to carcinogens as compared to exposure occurring at mature age [3]. Specifically, the incidence of leukemia and central nervous system cancers has significantly increased in the US, Canada, and parts of Europe [4]. Although an increased incidence may reflect improvements in diagnosis, intrauterine and postnatal exposure to environmental xenobiotics is considered to be etiologically relevant [5]. Biomonitoring studies in children are affected to a lower extent by confounders like cigarette smoking and drinking habits, occupational exposure and life-style (mainly dietary factors), which are factors of great concern in adults. Conversely, the role played by infectious diseases in pediatric populations needs to be carefully taken into account. Among the bioassays used to evaluate the impact of environmental, genetic and life-style factors on genomic stability in humans, the micronucleus assay (MN) has gained increased attention, and a growing number of studies have been published. The key advantage of the MN assay is the relative ease of scoring, the limited costs and person-time required, and the precision obtained from scoring larger numbers of cells. MN is generally performed in peripheral blood lymphocytes (PBL) and, to a lesser extent, in epithelial cells, erythrocytes, alveolar macrophages and fibroblasts. This systematic review performed within the framework of the ChildrenGenoNetworkEuropean Community funded program aims at: (1) summarizing the available data on the occurrence of MN in different tissues in children (age range 0–18 years) exposed to known or suspected environmental mutagens or medical treatments; and (2) to assess the role of these exposures in MN frequency while accounting for the effect of individual host factors, and potential confounders.
2. Materials and methods 2.1. Review protocol and search strategy Individuals from newborns to late adolescence (age range 0–18 years) were considered as children. The choice of this age range allowed, theoretically, the evaluation of potential effects of hormonal changes occurring during puberty and that of cigarette smoking, often starting immediately after adolescence. All published studies were identified by systematically searching the MedLine database (US National Library of Medicine). The search strategy was the following: the term “Micronucleus Tests” was used as Medical Subject Headings (MeSH) keyword, and “All Child: 0–18 years”, “Human” and “Publication Date: from 1980 to 2002” were set as limits. One hundred nineteen citations were retrieved and all abstracts were manually reviewed. Both English and non-English-language papers were collected and reviewed. Relevant papers cited in the retrieved articles were also obtained. Studies including less than 10 subjects in each group (e.g., exposed and referents) were excluded. Overall, 20 papers from the international literature and 8 Russian papers were included in the review. MN frequency reported in the text should be always intended per 1000 binucleated cells (BN), unless otherwise specified. Since the most popular method is the cytokinesis block approach [6], the results reported in the text refer to this technique unless otherwise specified. In order to test the effect of age and gender on MN frequency, data for 448 subjects selected from the HUMN database [7] were analyzed using uni- and multi-variate techniques based on log-transformed MN data and using the SPSS 11.0 statistical software. 3. Results and discussion 3.1. Effect of host factors and confounding variables We found 14 studies evaluating the effect of gender and/or age on MN. Some of them have been specifically designed for this purpose while others included MN frequency by age and gender as secondary study outcome.
M. Neri et al. / Mutation Research 544 (2003) 243–254
3.1.1. Gender Bonassi et al. analyzed a pooled database of nearly 7000 subjects from 25 international laboratories participating in the HUMN project [7]. They reported an increase in micronucleated cells (MNed) frequency in adult females compared to males, but not in the age range 0–19 years. Among children, no effect of age was reported (or could not be evaluated) by the other nine studies reviewed (Table 1). Barale et al. [8] conducted a population-based study among 1650 Italian subjects. In the subgroup aged ≤19 years (no. = 136), males (55%) and females showed the same frequency of MN (mean ± S.D. = 2.20 ± 2.41 and 2.20 ± 2.04, respectively). The effect of age was not investigated specifically in this subgroup, but in the whole population it was strongly and positively associated with MN frequency. The authors concluded that . . . newborns should show the same MN frequency regardless of sex. Ganguly [9] studied 32 children 0–18 years old and found that MN frequency in males and females was not significantly different. Gender had no statistical effect on MNed frequency in a study of 20 healthy Chinese children 0–10 years old [10]. Similar mean levels of MNed esophageal cells were measured in 106 young Chinese aged 15–20 years: males 5.8 ± 4.6, females 6.0 ± 4.1, mean ± S.D. [11]. MN frequencies did not differ in 12 boys (mean ± S.D. = 4.8 ± 3.1) and 13 girls (4.5 ± 3.1) who migrated from the Chernobyl area to the USA and in USA referents (5.4 ± 2.9, 16 boys and 4.9 ± 2.015 girls, [12]). No significant effect of gender on MNed frequencies was detected in 42 Belarus children (16 affected by thyroid tumor) exposed to radioactive contamination following the Chernobyl accident and 30 healthy Italian unexposed kids [13]. A study of 149 children aged 4–17 years living in Belgian villages close to chemical waste sites or in reference villages failed to reveal any relationship between gender and MN frequencies [14]. Gender had no clear influence on MNed frequency assessed in PBL of 21children with various malignancies and 20 controls [15] and in MN scored in bone marrow erythroblasts of 41 acute lymphoid leukaemia (ALL) patients (age range 1–13 years) at diagnosis, during and after chemo-radiotherapy [16]. The change in MN frequency across treatment showed no significant gender effect. The consistency of the findings reported by the reviewed studies, is against any clear effect of gender on MN frequency. Gender
245
Fig. 1. MN frequency (standardized) according to age class in 448 children selected from the HUMN database.
was found to have no effect on MN by the re-analysis of 448 children selected in the HUMN dataset, even in the age class after puberty (>14 years) (Fig. 1). 3.1.2. Age The extent of the association between MN frequency and age seems rather unclear in the age classes examined here. No effect at all was described in three studies, while four papers reported the presence of a positive association between age and MN frequency (Table 1). No effect of age was shown in children exposed to environmental tobacco smoke (ETS) or IR, and in neoplastic patients. Baier et al. [17] studied the effects of ETS in 77 children aged 3–15 years. No correlation of MN frequency with age was found in the whole study group (r = 0.12, P = 0.47), nor when ETS exposed and referent subjects were analysed separately. Age showed no significant effect on MN in Belarus children exposed to IR and in Italian referents [13]. No age effect was detected by multiple regression analysis in ALL patients [16]. MN frequency increased with age in 36 healthy children not exposed to IR [9]. An age-dependent increase in MN was observed in exfoliated urothelial cells but not in buccal cells of 81 healthy Afro-American children aged 5–12 years (P < 0.01)
246
Table 1 Studies reporting on the effect of age and gender on MN or MNed frequency in children Study aim (country)
No. of children
Age range (years)
Males (%)
Effect of gender
Effect of age
Bonassi et al., 2001 [7]
HUMN data set (international) Air pollution effect (Italy) Effect of age (India)
484
0–19
53
No effect
n.a.
136 population based 36 healthy unexposed
≤19 0–20
55 53
No effect No effect
20 healthy referents
0–10
65
No effect
n.a. Positive relation with age in the whole group (127 healthy subjects 0–70 years) P < 0.0000 MN increased in 11–20 years vs. 0–10 years old n.a.
106 (83 normal, 23 esophagitis) 56 (25 exposed, 31 US referents) 72 (26 exposed, 16 exposed with thyroid tumor, 30 referents) 149 (59 exposed, 90 referents) 41 (21 cancer cases, 20 controls) 41 ALL (followed through therapy) 77 (45 exposed, 32 referents) 81 healthy Afro-American
15–20
54
No effect
n.a.
Exposed: 4–14, referents: 4.0a 13.0 ± 6.0a
50
No effect
n.a.
53
No effect
No effect
4–17
48
No effect
n.a.
0–18
49
No effect
n.a.
1–13
44
No effect
No effect
3–15
n. sp.
n.a.
No effect of age in referents
5–12
n. sp.
n.a.
8–12
56 (exposed)
n.a.
ALL: 1–7 referents: 1–12
53
n.a.
Age-dependent increase in MN in exfoliated urothelial cells, no effect in buccal cells No effect of age in referents effect in exposed only (cumulative dose) Referents: positive correlation with age (P < 0.01)
Barale et al., 1998 [8] Ganguly, 1993 [9]
Shi et al., 2000 [10] Chang-Claude et al., 1992 [11] Livingston et al., 1997 [12] Zotti-Martelli et al., 1999 [13]
Vleminckx et al., 1997 [14]
Age and non-disjunction of chromosome 21 Risk of esophageal cancer (China) Immigrants to USA from Chernobyl (Russia) Chernobyl effects (Russia)
Slavotinek et al., 1993 [16]
Effect of chemical dump site (Belgium) Effect of health status (various cancers) ALL on RT, CT
Baier et al., 2002 (in German) [17]
ETS (Germany)
Holland et al., 2001 (abstract) [18]
Development of a non-invasive methods (USA) Radon in elementary school (Slovenia) ALL on RT, CT
Fellay-Reynier et al., 2000 [15]
Bilban and Vaupotic, 2001 [19] Migliore et al., 1991 [20]
105 (85 exposed, 20 referents) 45 (15 on therapy, 15 end therapy, 15 controls)
Not assessed (n.a.), not specified (n. sp.), acute lymphoid leukemia (ALL), radiotherapy (RT), chemotherapy (CT), environmental tobacco smoke (ETS). a Mean ± S.D.
M. Neri et al. / Mutation Research 544 (2003) 243–254
Author, year
M. Neri et al. / Mutation Research 544 (2003) 243–254
enrolled in a study aimed at developing non-invasive methods for the evaluation of genetic damage [18]. Bilban and Vaupotic [19] compared 85 pupils attending a radon-contaminated elementary school in Slovenia and 20 children from a non-contaminated school (age range 8–12 years). No correlation between age and MN was found in unexposed scholars while a significant correlation was detected in radon exposed pupils (P = 0.032), although age may reflect a higher cumulative exposure to radon. Conversely, a study of children with ALL showed a significant correlation (P < 0.01) between age and MNed in referent subjects (age range 1–12 years) [20]. The main limitation of the studies reviewed is their small sample size (compared to studies in adults), and a low statistical power of detecting an effect if present. The re-evaluation of a subset of the HUMN database after categorization of age in four classes indicated a clear increase of MN with increasing age (β = 0.03; P < 0.05) (Fig. 1). 3.1.3. Health status and therapeutic treatment Some studies were conducted among sick children (Table 2). They generally revealed an increased MN frequency both in children with non-neoplastic (two studies) and neoplastic diseases (four studies) compared to healthy controls. Chemo-radiotherapeutic treatments were also associated with increased MN levels. A three-fold increase in micronucleated PBL frequency was detected in acute and in convalescent children with mononucleosis [21]. MNed frequency was significantly lower (but still twice that in controls) in these children 18 months after recovery (the study reported unusually high figures: mean MNed = 28.7‰ cells in control subjects). Maluf and Erdtmann [22] compared MN frequencies in children affected by Fanconi anemia (FA), in infants with Down syndrome (DS) and in healthy children. The authors reported more than doubled levels of MN in FA patients than in both DS and control children, who showed very similar MN frequencies. DS subjects were younger than the other two groups. Children affected by various types of neoplasms (lymphoma, nephroblastoma, neuroblastoma, etc.), recruited before any treatment, had significantly higher MNed frequency than healthy controls (most of them were feverish at the time of blood sampling) [15]. Belarus children (no. = 42) exposed to IR following the Cher-
247
nobyl accident were monitored by Zotti-Martelli et al. [13]. Children affected by thyroid tumor (no. = 16) had significantly increased MNed frequencies with respect to Italian controls. Two studies examined the effects of antileukemic chemo-radiotherapy (CT–RT) in children with ALL. Migliore et al. [20] showed five-fold higher levels of MN in CT–RT treated patients than in controls, with MN frequency decreasing at the end of treatment. Slavotinek et al. [16] counted MN in 500 (or less) erythroblasts from bone marrow slides of 41 ALL patients followed up from diagnosis through three phases of CT–RT to the completion of 2 years of treatment. MN frequencies were markedly higher in ALL children than in controls. A marked and significant rise in MN frequency was shown throughout the treatment period. Wuttke et al. [23] reported on the effects of radio iodine therapy (RIT) for thyroid cancer on MN frequency in PBL from children aged 7–18 years from Belorussian and Ukranian republics exposed to fall-out of the Chernobyl accident. The MN level assessed after RIT was 2.5 higher than before treatment, and MN were found to be useful in determining individual radionuclide dose after therapy. 3.2. Effect of environmental exposure to mutagens 3.2.1. Ionizing radiation Exposure to IR as a consequence of accidents has been the most common reason for studying MN in children. An outline of these studies is reported in Table 3. MN frequency was counted using four different techniques in children from four regions of Belarus [24,25]. Contamination levels in soil measured with 137 Cs and 90 Sr Ci km−2 were used as measures of exposure to IR. Higher values were found in Vetka, intermediate in Yazvenki, and low in Kalinkovichi. Minsk, characterized by a very low contamination, was considered as the reference region. MN were counted in binucleated and mononucleated cultured blood cells, and in blood smears in lymphocytes and in erythrocytes. Children from Yazvenki had higher MN frequencies than Minsk controls with all four techniques, from Vetka in blood smears only, and from Kalinkovichi only in mononucleated cells. Twenty-five immigrants (age range 4–14 years) from the Chernobyl region to the USA [12] showed slightly
248
Table 2 Studies reporting on the effect of health status and therapeutic treatments on MN frequency in children Author, date
Disease, treatment
No. of children
Age range (years)
MNed frequency (10−3 cells), mean ± S.D. (no.) Sick
Statistical significance
Frequency ratio
Healthy
Mononucleosis
58
3–15
Acute stage Convalescent Recovery
89.3 ± 18.1 (21) 93.3 ± 9.7 (7) 57.5 ± 5.2 (17)
28.7 ± 3.8 (13)
ac/ref P < 0.001 conv/ref P < 0.001 rec/ref P < 0.001 conv/rec P < 0.01
3.1 3.3 2.0 1.6
Maluf et al., 2001 [22]
Fanconi anemia Down syndrome
74
FA 3–16, DS 0–2 Ref. 0–17
FA DS
10.9a (14) 5.1a (30)
4.7a (30)
FA/ref P < 0.001 DS/ref n.s.
2.3 1.1
Fellay-Reynier et al., 2000 [15]
Various cancers
41
0–18
2.4 ± 2.3 (20)
P < 0.05
2.1
Zotti-Martelli et al., 1999 [13]
Thyroid tumor (Chernobyl)
72
13.0 ± 6.0b
IR exp + tumor
IR exposed Referents
3.6 ± 2.7 (26) 2.0 ± 2.1 (30)
Exp + tum/exp < 0.05 Exp + tum/ref < 0.05
2.1 3.7
Migliore et al., 1991 [20]
ALL on RT, CT
45
ALL 1–7, Ref 1–12
On therapy End therapy
20.0 ± 12.9 (15) 13.2 ± 8.4 (15)
Referents
3.7 ± 1.6 (15)
On/ref P < 0.001 End/ref P < 0.001 On/end n.s.
5.4 3.6 1.5
Slavotinek et al., 1993 [16]
ALL on RT, CT
51
ALL 1–13
At diagnosis I phase II phase III phase End therapy
11.8/0–56a,c (12) 13.6/2–10.4a,c (38) 64.1/2–194a,c (32) 28.5/0–200a,c (13) 51.6/0–258a,c (36)
2.6/0–6a,c (10)
End/diagn P < 0.001 I/II P < 0.001
4.4 4.7
Before RIT After RIT
22d (43) 55d (41)
Wuttke et al., 1996 [23]
Thyroid cancer
56 7–18
5.1 ± 3.9 (21) 7.4 ± 4.9 (16)
Diagn/ ref = 4.5
2.5 The percentages of males in the studies ranged from 44 to 62% (not specified in Urazova et al., 2001). Methods: CBMN in blood or PBL culture; non-CBMN in Urazova et al., 2001; bone marrow erythroblasts on stored slides in Slavotinek et al., 1993. FA: Fanconi anemia; DS: Down syndrome; IR: ionizing radiation; ALL: acute lymphoid leukemia; RT: radiotherapy; CT: chemotherapy; RIT: radio-iodine therapy; ref: referents. a MN frequency (10−3 cells). b Mean ± S.D. c Mean/range. d MN frequencies estimated from Fig. 1 [23].
M. Neri et al. / Mutation Research 544 (2003) 243–254
Urazova et al., 2001 [21]
Table 3 Studies reporting on the effect of exposure to ionizing radiation on MN frequency in children Type of exposure (site)
Exposure assessment
No. of children
Age range Biol sample (assay) (years)
Fenech et al., 1997 [24]; Mikhalevic et al., 2000 [25]
Radioactive sources (Chernobyl)
Quantitative group level
116
12–15
MN frequency (10−3 cells), mean ± S.D. (no.) Exposed
Blood (CBMN), binucleated cells Blood (CBMN) mononucleated cells Blood smears lymphocytes
Livingston et al., 1997 [12] Zotti-Martelli et al., 1999 [13] da Cruz et al., 1994 [26] Tsai et al., 2001 [27] Bilban and Vaupotic, 2001 [19]
Radioactive sources (Chernobyl) Radioactive sources (Chernobyl) 137 Cs radiological accident (Brazil) 60 Co contaminated buildings (Taiwan) Radon contaminated school (Slovenia)
Quantitative, individual level Quantitative, individual level Quantitative, group level Quantitative, individual level Quantitative, group level
56
4–14b
Blood smears, erythrocytes PBL(CBMN)
72
13 ± 6.0
Blood (CBMN)
68
1–18
Blood (CBMN)
18d
4–18
PBL (CBMN)
105
8–12
PBL (CBMN)
Yazvenki Vetka Kalinkovichi Yazvenki Vetka Kalinkovichi Yazvenki Vetka Kalinkovichi Yazvenki Vetka Males Females Thyroid tumor Healthy Directly exposed Probably exposed First assay
Statistical significance
Frequency ratio
Referents 46.4 14.3 7.5 57.0 23.2 27.1 35.6 26.9 9.7 0.20 0.60 4.8 4.5 7.4 3.6 11.6 10.0 20.7
± 3.8a (10) ± 1.9a (10) ± 0.5a (20) ± 6.8a (10) ± 3.4a (10) ± 1.6a (20) ± 10.1a (4) ± 3.4a (24) ± 1.4a (20) ± 0.02a (22) ± 0.05a (10) ± 3.1 (12) Males ± 3.1 (13) Females ± 4.9c (16) ± 2.7c (26) (24) (14) Second assay
6.5 ± 2.5e (85)
9.8 ± 1.0a (11) Y/ref P < 0.01
10.7 ± 2.2a
7.3 ± 1.0a
0.13 ± 0.3a
4.7
Y/ref P < 0.01 K/ref P < 0.01
5.3 2.5
P < 0.01 P < 0.01 P < 0.01 P < 0.01 P < 0.01
4.9 3.7
Y/ref Y/ref V/ref Y/ref V/ref n.s.
1.5 4.6 5.4 ± 2.9 (16) 0.89 4.9 ± 2.0 (15) 0.92 2.0 ± 2.1c (30) Tum/ref P < 0.05 3.7 exp/ref n.s. 1.8 7.3 (30) – 1.6 1.4 8.8 2.3 4.5 ± 1.9e (20) P = 0.001
1.4
M. Neri et al. / Mutation Research 544 (2003) 243–254
Author, date
The percentages of males in the studies ranged from 50 to 67%. Y, Yazvenky; V, Vetka; K, Kalinkovichi. a Mean ± S.E. b US ctrl: mean age 4.0. c MNed frequency (10−3 cells). d All exposed subjects tested twice, interval 11–36 months. e MN frequency/500 binucleated cells.
249
250
M. Neri et al. / Mutation Research 544 (2003) 243–254
lower MN levels than 31 US pre-school referent children. Forty-two children from the severely contaminated area of Gomel (16 of whom with thyroid tumor and 5 with thyroid nodules) had higher MN frequency than healthy Italian children [13]. According to the authors, the increased MN frequency observed in children with thyroid tumor was likely to be attributable to 137 Cs contamination rather than to the presence of the tumor itself. Sixty-eight children aged 1–18 were studied following a radiological accident involving 137 Cs in Goiania, Brazil. IR dose was estimated in a subgroup of exposed subjects by urine and faeces bioassays and whole-body counting. Estimation of MN frequency was performed from published individual data [26], and indicated higher levels in exposed than in unexposed children. The presence of steel rods contaminated with 60 Co in civilian buildings in Taiwan resulted in the exposure of thousands of civilians to ␥-radiation. Forty-eight subjects (annual average exposure 32.8 ± 28.1 mSv) were evaluated twice for MN: 9 months after relocation from the contaminated buildings and 26 months later [27]. Eighteen of the 48 subjects were aged 18 or less and their mean MN frequency (calculated from individual data) showed higher levels at 9 months than at 26 months. Pupils attending a radon-contaminated elementary school [19], had higher MN per 500 binucleated cells (6.5 ± 2.5) than matched children from a non-contaminated school (4.5 ± 1.9). 3.2.2. Chemicals The MN test was used in two studies evaluating children exposed to environmental chemicals (Table 4). In the village of Mellery, Belgium, alkanes and unsaturated chlorinated hydrocarbons were detected in air, water and soil [28,14] near a chemical dumping site. A cytogenetic survey was performed in 45 children 3–17 years old after the decontamination of the area; 12 children (age range 7–16 years) from Hensies, a Belgian village situated close to an active chemical dumping site; and 89 referents (aged 6–15) selected from different areas. MN frequency was significantly increased in the exposed children from Hensies compared to the other groups. MN in children from Mellery did not differ from referents. MN levels were similar in children from two urban areas while those from a semi-rural village had the lowest MN frequency. Ten native Andean children exposed to drinking water heavily con-
taminated with arsenic (As concentration = 200 g/l in water and 9.0 g/l in blood) were compared with 12 unexposed children in the same age range (age range 8–15 years) (As concentration = 0.7 g/l in water and 0.8 g/l in blood) [29]. The frequency of MN in PBL of the exposed children (35 ± 4.6) was six times higher than in the reference subjects (5.6 ± 1.6). 3.2.3. Air pollution The evaluation of MN frequency in human populations exposed to air pollution (AP) or ETS is one of the most challenging applications of the assay in biomonitoring studies. The majority of studies conducted in adults failed to detect any clear difference between subjects exposed to AP and referents, attributable to polluted ambient air. The higher sensitivity of children may improve the performance of these field studies. In a population study carried out in 1640 individual living in Pisa, Italy, and surrounding municipalities [8], selected respiratory function parameters and cytogenetic endpoints, including MN in PBL, were assessed. The study population was stratified in subgroups according to the distance of their residence from traffic polluted roads. No specific results were provided for the 119 children included in the study. 3.2.4. Environmental tobacco smoke (ETS) and cigarette smoking Despite the great concern for the health effects (both acute and delayed) of ETS exposure, the number of published studies conducted in children is scanty. A German study analyzed the relation between ETS exposure and respiratory or atopic diseases and genotoxic damage measured as MN frequency in PBL in children aged 2–15 years [17]. MN frequency was significantly increased in 45 ETS exposed when compared to 32 unexposed children. The difference was observed also when children were stratified in two age groups (age range ≤7 and >7 years). Children with both parents smoking had slightly higher MN than those with only one parent smoking. MN were counted in esophageal cells of 186 “young adults” from a high-incidence area for esophageal cancer in China [11]. No effect of smoking on MN was detected (mean values ± S.D. = 6.2 ± 4.6 and 6.2 ± 4.1) in 121 non-smoker and 65 current smokers (usually < 1 pack per day), respectively. Unfortunately, these data refer to the whole group of subjects,
Author, year
Vleminckx et al., 1997 [14]
Study aim (country) No. of Age range Exposure children (years) assessment
MN frequency (10−3 cells), mean ± S.D. (no.)
Effect of chemical 149 dump site (Belgium)
Mellery 4.8 ± 6.5 (47) Semi-rural area
4–17
Quantitative, group level
Exposed
Referents 2.9 ± 2.5 (34)
Hensies 7.0 ± 3.8 (12) Non-industrial city 4.2 ± 3.5 (31) Industrial city Dulout et al., 1996 [29]
Arsenic in drinking water (Argentina)
22
8–15
Quantitative, individual level
35 ± 4.6a (10)
The percentages of males in the studies were about 50%. Method: CBMN in blood cultures. a Mean ± S.E.
Statistical significance
Frequency ratio
Hensies/semi-rural 1.65 area P < 0.01 Hensies/Mellery 1.45 P < 0.05
4.8 ± 5.0 (25) 5.6 ± 1.6a (12) P < 0.001
6.25
M. Neri et al. / Mutation Research 544 (2003) 243–254
Table 4 Studies reporting on the effect of chemical exposure on MN frequency in children
251
252
M. Neri et al. / Mutation Research 544 (2003) 243–254
aged 15–26, as no further categorization by age was attempted. 3.2.5. Review of the Russian literature on MN in children Most reports published in Russian and Ukrainian literature on biodosimetry and genetic toxicology are by large based on data from the Chernobyl accident and others accidental exposures of the general or working population. Pelevina et al. [30,31] performed the in vitro MN assay in lymphocytes of 860 children (age range 2–16 years) exposed to urban air pollution stratified into subgroups according to severity of air pollution (the size of the subgroups ranged between 22 and 96 children). The results showed an increased MN frequency in children exposed to environmental pollution (23.0‰) compared to referents (10.0‰) followed by a reduced capability to develop adaptive response and an increased radio-sensitivity. Illynskikh et al. [32,33] conducted a study of adult and pediatric populations using the in vivo and in vitro MN assay to estimate genetic damage following accidental overexposures to IR in the population living in an atomic testing region. Specifically, the capacity of two drugs, pentoxylum and leucogenum to decrease MN frequency in reticulocytes was evaluated. A decreased MN frequency was observed in children after 3 days of therapy. The same effect was observed in adults following a 2 weeks treatment. Fedoretsova et al. [34] addressed the issue of damage persistence in 54 children (age range 6–16 years) evacuated from the Chernobyl area and in two referent groups from Sosnovy Bor (no. = 53, age range 3–6 years) and S. Petersburg (no. = 41, age range 3–6 years), sampled 5 years after the accident. The frequency of micronucleated erythrocytes (FME) was higher in IR exposed children (mean = 0.19) than in referents (0.017 and 0.008), and exposed children had also a higher inter-individual variability. Interestingly, exposed children had higher FME than exposed adults (no. = 51, P = 0.06), possibly because of a higher radiosensitivity of erythropoietic cells in children. Children chronically exposed to internal radiation 9 years after Chernobyl accident [25] showed an increased frequency of MN in mononucleated lymphocytes, but not in binucleated lymphocytes. This finding addresses the issue of monitoring MN frequency, both in mono and binucleated lymphocytes. An interesting study has
been conducted in several Ukrainian regions with water and soil contaminated by metals (Fe, Al, Mn, Pb, Zn, Ni). MN frequency in buccal epithelium was measured in children 6 years old, and detailed analyses of metals in soil and water were performed. A correlation between the concentration of metals (Al and Fe) and frequency of MN is reported, although not enough information on the study group was provided [35]. More than 1,000,000 of children in the age range 0–14 are potentially exposed to IR in the Russian Federation alone; most of them live in territories contaminated by 137 Cs with an intensity of 5 Ci km−2 or more [36]. As a direct consequence of this evidence, scientists from the Russian Federation, Ukraine, Byelorussia, and from the rest of the world, have shown a growing interest to evaluate genetic damage in children living in this area.
4. Conclusions The present review of MN studies in children exposed to environmental mutagens reveals the very limited application of the MN assay to detect early biological outcomes of exposures to environmental genotoxic agents. The review provides important information that should be considered in the design of studies on environmental pollutants, such as the need to control for confounding factors, to understand the impact of effect modifiers of the exposure–effect relations, and to define the appropriate number of children required in biomonitoring studies, i.e., sample size. An evident effect of chronic and infectious diseases on MN levels was reported in most studies, making clear that any research on environmental mutagens and MN frequency should always consider carefully the role of health status, preferably during the design of a study protocol. An age-related effect is present in children (as in adults), while the influence of gender in MN frequency seems to be irrelevant. Although the limited number of published studies prevents from reaching definitive conclusions on the relationship between environmental exposure and early biological effect, most studies describing the effect of exposure to genotoxic agents found an increase of MN in exposed children. A certain contribution of the so-called publication bias should be considered, e.g., the negative result from a large study [8] was not published.
M. Neri et al. / Mutation Research 544 (2003) 243–254
However, the internal consistency of some studies lends support to the presence of this association, describing significant dose-related effects in exposed populations. The MN assay seems to work well in children exposed to low doses of IR, such as in the case of residents in contaminated areas. In some of the studies evaluated, various biomarkers were measured in the same subjects, and a multi-endpoints approach seems to be more efficient. Data concerning chemical pollution and exposure to ETS, although suggestive, are too sparse to provide any sound conclusion. In conclusion, the review confirmed the usefulness of MN assay in biomonitoring studies conducted in children, confirming a great extent of sensitivity, even for exposure to low doses of environmental agents. The limited number of published papers may indicate that the conduct of properly designed population-based studies in children is difficult. More research in this segment of the human population is recommended.
Acknowledgements The study was supported by grants funded by the Associazione Italiana per la Ricerca sul Cancro (AIRC) and the European Union 5th FP (QLRT2001-02831 and QLRT-2001-02198). References [1] Nordic Study Group on the Health Risk of Chromosome Damage, A Nordic data base on somatic chromosome damage in humans, Mutat. Res. 241 (1990) 325–337. [2] S. Bonassi, L. Hagmar, U. Strömberg, A.H. Montagud, H. Tinnerberg, A. Forni, P. Heikkilä, S. Wanders, P. Wilhardt, I.L. Hansteen, L.E. Knudsen, H. Norppa,. European Study Group on Cytogenetic Biomarkers and Heath Chromosomal aberrations in lymphocytes predict human cancer indipendently of exposure to carcinogens, Cancer Res. 60 (2000) 1619–1625. [3] L.M. Anderson, B.A. Diwan, N.T. Fear, E. Roman, Critical windows of exposure for children’s health: cancer in human epidemiological studies and neoplasms in experimental animal models, Environ. Health Perspect. 3 (Suppl. 108) (2000) 573– 594. [4] J.G. Gurney, S. Davis, R.K. Severson, J.Z. Fang, J.A. Ross, L.L. Robinson, Trends in cancer incidence among children in the US, Cancer 78 (1996) 532–541. [5] J. Olsen, Significance of prenatal exposure for long-term health effects, Lakartidningen 98 (2001) 3798–3802.
253
[6] M. Fenech, A.A. Morley, The effect of donor age on spontaneous and induced micronuclei, Mutat. Res. 148 (1985) 99–105. [7] S. Bonassi, M. Fenech, C. Lando, Y.P. Lin, M. Ceppi, W.P. Chang, N. Holland, M. Kirsch-Volders, E. Zeiger, S. Ban, R. Barale, M.P. Bigatti, C. Bolognesi, C. Jia, M. Di Giorgio, L.R. Ferguson, A. Fucic, O.G. Lima, P. Hrelia, A.P. Krishnaja, T.K. Lee, L. Migliore, L. Mikhalevich, E. Mirkova, P. Mosesso, W.U. Muller, Y. Odagiri, M.R. Scarf`ı, E. Szabova, I. Vorobtsova, A. Vral, A. Zijno, HUman MicroNucleus project: international database comparison for results with the cytokinesis-block micronucleus assay in human lymphocytes. I. Effect of laboratory protocol, scoring criteria, and host factors on the frequency of micronuclei, Environ. Mol. Mutagen 37 (2001) 31–45. [8] R. Barale, L. Chelotti, T. Davini, S. Del Ry, M.G. Andreassi, M. Ballardin, M. Bulleri, J. He, S. Baldacci, F. Di Pede, F. Gemignani, S. Landi, Sister chromatid exchange and micronucleus frequency in human lymphocytes of 1,650 subjects in an Italian population. II. Contribution of sex, age, and lifestyle, Environ. Mol. Mutagen 31 (1998) 228–242. [9] B.B. Ganguly, Cell division, chromosomal damage and micronucleus formation in peripheral lymphocytes of healthy donors: related to donor’s age, Mutat. Res. 295 (1993) 135– 148. [10] Q. Shi, J. Chen, I. Adler, J. Zhang, R. Martin, S. Pan, X. Zhang, X. Shan, Increased nondisjunction of chromosome 21 with age in human peripheral lymphocytes, Mutat. Res. 452 (2000) 27–36. [11] J. Chang-Claude, H. Shimada, N. Munoz, J. Wahrendorf, Q.S. Liang, Y.G. Rei, M. Crespi, R. Raedsch, P. Correa, Micronuclei in esophageal cells of Chinese youths in a highincidence area for Esophageal Cancer in China, Cancer Epidemiol. Biomarkers Prev. 1 (1992) 463–466. [12] G.K. Livingston, R.H. Jensen, E.B. Silberstein, J.D. Hinnefeld, G. Pratt, W.L. Bigbee, R.G. Langlois, S.G. Grant, R. Shukla, Radiobiological evaluation of immigrants from the vicinity of Chernobyl, Int. J. Radiat. Biol. 72 (1997) 703–713. [13] L. Zotti-Martelli, L. Migliore, G. Panasiuk, R. Barale, Micronucleus frequency in Gomel (Belarus) children affected and not affected by thyroid cancer, Mutat. Res. 440 (1999) 35–43. [14] C. Vleminckx, W. Klemans, L. Schriewer, I. Joris, N. Lijsen, M. Ottogali, A. Pays, C. Planard, G. Rigaux, Y. Ros, M. Vande Riviere, J. Vandenvelde, P. De Plaen, T.H. Lakhanisky, A. Maes, L. Verschaeve, Performance of cytogenetic biomarkers on children exposed to environmental pollutants, Toxicol. Ind. Health 13 (1997) 219–230. [15] I. Fellay-Reynier, T. Orsiere, I. Sari-Minodier, P. Auquier, H. Zattara-Cannoni, A.M. Capodano, C. Coze, M. De Meo, J.L. Bernard, A. Botta, Evaluation of micronucleated lymphocytes, constitutional karyotypes and anti-p53 antibodies in 21 children with various malignancies, Mutat. Res. 467 (2000) 31–39. [16] A. Slavotinek, A. Thomson, P. Eynaud, P. Perry, C.M. steel, O.B. Eden, The frequency of micronuclei in bone-narrow erythroblast during the treatment of childhood acute lymphoblastic leukemia, Mutat. Res. 303 (1993) 11–18.
254
M. Neri et al. / Mutation Research 544 (2003) 243–254
[17] G. Baier, H. Stopper, C. Kopp, U. Winkler, I. Zwirner-Baier, Respiratory diseases and genotoxicity in tobacco smoke exposed children, Laryngorhinootologie 81 (2002) 217–225. [18] N.T. Holland, A. Hubbard, L. Schumacher, L. Gunn, D. Golden, P. Duramad, M.T. Smith, Factors affecting micronucleus frequency in exfoliated buccal and urothelial cells of adults and children, Environ. Mol. Mutagen 37 (Suppl. 32) (2001) 39. [19] M. Bilban, J. Vaupotic, Chromosome aberrations study of pupils in high radon level elementary school, Health Phys. 80 (2001) 157–163. [20] L. Migliore, P. Guidotti, C. Favre, M. Nardi, M.R. Sessa, E. Brunori, Micronuclei in lymphocytes of young patients under antileukemic therapy, Mutat. Res. 263 (1991) 243–248. [21] O.I. Urazova, L.S. Litvinova, V.V. Novitskii, A.P. Pomogaeva, Cytogenetic impairments of peripheral blood lymphocytes during infectious mononucleosis, Bull. Exp. Biol. Med. 131 (2001) 392–393. [22] S.W. Maluf, D.F. Ferreira Passos, A. Bacelar, G. Speit, B. Erdtmann, Genomic unstability in Down syndrome and Fanconi anemia assessed by micronucleus analysis and single-cell gel electrophoresis, Cancer Genet. Cytogenet. 124 (2001) 71–75. [23] K. Wuttke, C. Streffer, W.U. Muller, C. Reiners, J. Biko, E. Demidchik, Micronuclei in lymphocytes of children from the vicinity of Chernobyl before and after 131I therapy for thyroid cancer, Int. J. Radiat. Biol. 69 (1996) 259–268. [24] M. Fenech, G. Perepetskaya, L. Mikhalevich, A more comprehensive application of the micronucleus technique for biomonitoring of genetic damage rates in human populations—experiences from the Chernobyl catastrophe, Environ. Mol. Mutagen 30 (1997) 112–118. [25] L.S. Mikhalevich, F.A. De Zwart, G.A. Perepetskaya, N.V. Chebotareva, E.A. Mikhalevich, A.D. Tates, Radiation effects in lymphocytes of children living in a Chernobyl contaminated region of Belarus, Int. J. Radiat. Biol. 76 (2000) 1377–1385. [26] A.D. da Cruz, A.G. Mc Arthur, C.C. Silva, M.P. Curado, B.W. Glickman, Human micronucleus counts are correlated with age, smoking, and cesium-137 dose in the Goiania (Brazil) radiological accident, Mutat. Res. 313 (1994) 57–68. [27] M.H. Tsai, J.S. Hwang, K.C. Chen, Y.P. Lin, W.A. Hsieh, W.P. Chang, Dynamics of changes in micronucleus frequencies in subjects post cessation of chronic low-dose radiation exposure, Mutagenesis 16 (2001) 251–255. [28] W. Klemans, C. Vleminckx, L. Schriewer, I. Joris, N. Lijsen, A. Maes, M. Ottogali, A. Pays, C. Planard, G. Rigaux, Y. Ros,
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
M. Vande Riviere, J. Vandenvelde, L. Verschaeve, P. Deplaen, Th. Lakhanisky, Cytogenetic biomonitoring of a population of children allegedly exposed to environmental pollutants. Phase 2. Results of a three-year longitudinal study, Mutat. Res. 342 (1995) 147–156. F.N. Dulout, C.A. Grillo, A.I. Seoane, C.R. Maderna, R. Nilsson, M. Vahter, F. Darroudi, A.T. Natarajan, Chromosomal aberrations in peripheral blood lymphocytes from native Andean women and children from Northwestern Argentina exposed to arsenic in drinking water, Mutat. Res. 370 (1996) 151–158. I.I. Pelevina, A.V. Aleshchenko, M.M. Antoshchina, O.V. Kudriashova, L.E. Kurneshova, V.Y.A. Gotlib, L.A. Noskin, V.A. Noskin, L.P. Semenova, A.M. Serebrianyi, Level of spontaneous and radiation-induced cytogenetic damage in blood lymphocytes of children depending on age and life style, Radiat. Biol. Radioecol. 41 (2001) 573–579. I.I. Pelevina, G.G. Afanasiev, A.V. Aleshchenko, V.Y. Gotlib, L.E. Kurneshova, V.A. Noskin, L.A. Noskin, L.P. Semenova, A.M. Serebrynyi, Radioadaptive response in children and influence of external and internal factors, Radiat. Biol. Radioecol. 39 (1999) 106–112. N.N. Ilyinskikh, A.V. Eremich, I.I. Ivanchuk, I.N. Ilyinskikh, Micronucleus test of erythrocytes and lymphocytes in the blood of the Altai region residents living near the Semipalatinsk atomic proving ground, Mutat. Res. 392 (1997) 223–228. N.N. Ilyinskikh, E.N. Ilyinskikh, I.N. Ilyinskikh, A decrease in the level of erythrocytes with micronuclei under the influence of pyrimidine and thiazolidine derivatives in the blood of persons who came under radiation exposure as a result of the accident at the Siberian Chemical Combine, Tsitol. Genet. 32 (1998) 26–29. R.F. Fedoretsova, V.Y. Kravtsov, O.V. Belyakov, I. Vorobtsova, V.N. Nekrasov, A.M. Nikiforov, Late cytogenetic effects of the Chernobyl accident in children, Int. J. Pediatr. Hematol. Oncol. 4 (1997) 295–299. S.S. Rudenko, T.V. Morozova, S.S. Kostyshyn, V.F. Bezrukov, A.I. Gorova, Micronuclear index of somatic cells in the population of Chernivtsi region and its ecological conditionality, Tsitol. Genet. 36 (2002) 23–29. I.N. Lounga, E.V. Zlatkina, Assessment of the consequences of ionizing radiation exposure to human generations, in: Proceedings of the International Conference on Genetic Consequences of Emergency Radiation Situation, Moscow, 2002, pp. 167–169.
