- Email: [email protected]
“Normal values” for the lymphocyte cytokinesis-block micronucleus cytome parameters: Repeatability and reproducibility in a healthy reference population
Science of the Total Environment 652 (2019) 513–522
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
“Normal values” for the lymphocyte cytokinesis-block micronucleus cytome parameters: Repeatability and reproducibility in a healthy reference population Claudia Bolognesi a,⁎, Luigina Bonelli b, Andrea Compalati a, Valentina Ferla b, Laura Stagnaro a, Gianluca Ubezio c, Paolo Bruzzi b a b c
Unit of Environmental Carcinogenesis Ospedale Policlinico San Martino, Genova, Italy Unit of Clinical Epidemiology, Ospedale Policlinico San Martino Genova, Italy Transfusion Centre, Ospedale Policlinico San Martino Genova, Italy
H I G H L I G H T S
G R A P H I C A L
A B S T R A C T
• Good repeatability of the micronuclei and proliferative index was observed. • Consistency of MNBN values in individuals was confirmed in repeated blood samples. • Narrow ranges of normal reference group MNBN values were estimated. • Large variability of age- and genderspecific individual MNBN values was observed. • A model to derive MNBN “normal values” in a healthy population was established.
a r t i c l e
i n f o
Article history: Received 8 June 2018 Received in revised form 8 October 2018 Accepted 13 October 2018 Available online 15 October 2018 Keywords: Micronucleus Cytome assay Biomonitoring Occupational exposure Chromosomal instability Healthy subjects
a b s t r a c t The micronucleus test in peripheral blood lymphocytes is the most widely validated technique to evaluate the DNA damage and chromosomal instability in human populations. The test is largely applied in monitoring environmental and occupational exposure to genotoxic agents. It was also proposed as a biomarker of risk/susceptibility for cancer and other degenerative diseases. The availability of “normal values” in healthy populations is a main requisite for the assay application in human biomonitoring. Age and gender-related ranges of micronucleated binucleated cells (MNBN) baseline values were established in a group of 103 healthy platelet donors (50 males and 53 females) not recently exposed to genotoxic agents and characterized for demographic, lifestyle and dietary factors. Repeatability of the test by the same scorer was evaluated. Reproducibility was estimated through analysis of repeated blood samples. High correlation between the results of the three blood samplings in two separate scoring sessions of MNBN/1000BN (R2 values were 0.83, 0.74 and 0.68; P b 0.0001) and PI values (R2 values were 0.69, 0.62 and 0.65; P b 0.0001) was detected. High consistency among the values obtained in three different samplings in the same individual was observed (Intraclass Correlation Coefficient (ICC) = 0.905, (95% CI = 0.868–0.933, P b 0.0001) The range of “normal” values predicted on the basis of the results of the present study appears to be sufficiently narrow to warrant application of the assay in the comparison
⁎ Corresponding author at: Environmental Carcinogenesis Unit, Ospedale Policlinico San Martino, L.go Rosanna Benzi, 10, 16132 Genova, Italy. E-mail address: [email protected] (C. Bolognesi).
https://doi.org/10.1016/j.scitotenv.2018.10.187 0048-9697/© 2018 Elsevier B.V. All rights reserved.
514
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
of data obtained from groups of exposed or susceptible subjects, supporting its use in preventive programs. The large inter-individual variability predicted by the model used in the present study hampers a clinical application of the assay at individual level. The method applied in the present study represents a generally applicable model to derive “normal values” in any population, as an essential step before starting a biomonitoring study. © 2018 Elsevier B.V. All rights reserved.
1. Introduction The cytokinesis-block micronucleus assay in peripheral lymphocytes (L-CBMN assay), due to its ability to detect both structural and numerical chromosomal aberrations and some epigenetic effects, is one of the most widely applied methods to evaluate genome damage in humans (Fenech et al., 2016). L-CBMN assay is recommended by the International Atomic Energy Agency (IAEA) for biodosimetry of ionizing radiation exposure (IAEA, 2011) and is largely used in biomonitoring occupational and accidental exposure to genotoxic agents (Nersesyan et al., 2016). A recent analysis of N500 studies available in the scientific literature on the application of the L-CBMN assay in human populations carried out in the framework of the HUMN (HUman MicroNucleus) project, confirms the usefulness of the assay to assess the exposure to a large number of occupational or environmental genotoxic compounds and mixtures. However, a large variability was observed among the studies, partially due to the limitations in study design, different experimental protocols applied and heterogeneity of the populations considered in terms of genetic susceptibility and lifestyle/concurrent exposures impairing the detection of a dose-response relationship (Nersesyan et al., 2016; Mutation Research Special Issue, 2016). Furthermore L-CBMN assay was proposed as a marker of risk for cancer and other chronic degenerative diseases, based on the results of prospective studies in large cohorts of disease-free subjects (Bonassi et al., 2007; Murgia et al., 2008) and of large case-control studies showing significant increase of MN frequency in cancer patients as compared to control groups (Bonassi et al., 2011; El-Zein et al. 2006 and 2011; Cardinale et al., 2012; McHugh et al., 2013; Maffei et al., 2014; Migliore et al., 2011; Andreassi et al., 2011). Although the L-CBMN assay is considered a suitable and reliable biomarker to assess DNA damage in humans, its implementation is limited by the large inter- and intra-individual variability (Bonassi et al., 2011; Fenech et al., 2003; Vral et al., 2002) observed in the large majority of the available biomonitoring studies which prevented the definition of reliable ranges of “normal values” for the MN frequency in healthy populations against which the MN levels, detected in exposed groups or susceptible subjects, could be compared. The availability of “normal values” is important for the application of the assay in biomonitoring occupational exposure where the extent of the increase is evaluated at the group level but it's critical for a potential clinical use of the assay as a biomarker of risk/susceptibility at the individual level. The analysis of the key components of the variability, the experimental factors and the intrinsic characteristics of the individuals, is a necessary premise to establish a reliable baseline range of “normal” MN frequencies. The availability of standardized protocol and scoring criteria should minimize the variability due to experimental parameters. The comparison of repeated analyses carried out on the same samples under optimized experimental conditions gives the indication of the residual variability. The estimation of the intraindividual variability of the cytome parameters over time in healthy subjects, suitably selected as minimally exposed to environmental genotoxins, is the subsequent step to evaluate if a single measure of MN frequency is representative of the individual chromosomal instability.
To address these issues, we designed a longitudinal study in healthy subjects without current exposures to known carcinogenic or mutagenic agents, who accepted to undergo 3 blood drawings for research purposes at approximately 3–month intervals, and to complete a questionnaire focused on their dietary and life-style habits. Aims of study were: • To estimate the repeatability of the MN analysis in the same laboratory following standardized protocol and scoring criteria • To evaluate the reproducibility of the MN frequency in individual subjects through analyses of repeated blood samplings • To establish the range of normal values of baseline MN frequency
2. Materials and methods 2.1. Study design and population The study was approved by the Ethical Committee of the Liguria region (P.R.027REG2014) and was carried out at the Ospedale Policlinico San Martino of Genoa (Italy). Each participant received written detailed information on the study purpose and methods and signed a written informed consent. The study population was a sample of blood donors who volunteered to participate. The enrollment of blood donors guaranteed screening for health risks, such as several viral and bacterial infections, drug addiction, and chronic and degenerative diseases that could affect the results of the experimental tests. We selected platelet donors because the frequency of platelet donation fitted the time table of the study. Donors were recruited at the Transfusion Centre. They were contacted at the time of a platelet donation and evaluated for their interest in participating to the study and eligibility. We planned to enroll 100 individuals (50 males and 50 females). Eligible were donors aged 18–65 years selected to ensure adequate representativeness of age and gender. Individuals were excluded if they reported any recent exposure to organic solvents, pesticides, paints, welding fumes and documented past exposure to asbestos. Individuals who underwent radiological examinations in the previous 30 days and those who had had a diagnosis of diabetes mellitus were also ineligible, due to the high MN frequency reported in these patients (Corbi et al., 2014). In order to address the intra-individual variability of the single biomarkers, analyses were carried out in blood samples taken at enrolment and after three and six months. 2.2. Data collection At enrolment (first round) a detailed questionnaire (registration form in supplementary materials) was used to elicit information on demographic characteristics (years of education, marital status, occupation, area of residence), smoking habits, height and weight, physical activity (intensity and frequency), personal medical history, history of cancer in first and second-degree relatives, use of drugs and micronutrients supplementation. Women were also interviewed on their reproductive history. Dietary information on the frequency of consumption of foods and drinks during a 12-month period prior to enrolment was obtained by a quantitative Food Frequency Questionnaire FFQ
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
that was developed for Italian dietary habits and was used in the European Prospective Project Into Cancer and Nutrition (EPIC)—Italy study (Pala et al., 2003). The two questionnaires were administered by a trained interviewer at the Transfusion Centre immediately before a platelet donation. At the three and six month sampling (second and third round, respectively) the participants were asked to check information provided at the enrolment in order to record any change occurred afterwards. In addition, they were required to fill in a 3-day food diary before donation to control for substantial variations over time in the mean per-day intake of the most relevant food components as collected at the baseline interview. 2.3. Blood collection At each round, participants provided a 10 ml sample of peripheral blood that was taken at the moment of the platelet donation. Blood samples were collected in a heparinized vial and maintained at room temperature for a maximum of 24 h before being processed. 2.4. Cytokinesis-blocked micronucleus assay in peripheral lymphocytes Heparinized blood samples, sent to the lab within 24 h, were used for the establishment of the lymphocyte cultures. The standardized protocol for the cytokinesis-block micronucleus cytome assay was applied (Fenech, 2007). Three sterile whole blood cultures for each sample were prepared. 0.3 ml aliquot of whole blood per each culture was incubated at 37 °C in 4.7 ml of RPMI 1640 (Life Technologies, Milano, Italy) supplemented with 10% fetal bovine serum (Gibco BRL, Life Technologies SrL, Milano, Italy), 1.5% phytohemoagglutinin (Murex Biotech,
515
Dartford, UK), 100 Unit/ml penicillin and 100 μg/ml streptomycin. After 44 h, Cytochalasin B (Sigma, Milano, Italy) was added at a concentration of 6 μg/ml. At the end of incubation at 37 °C for 72 h, cells were centrifuged (1000 rpm, 10 min) then treated with 5 ml of 0.075 mM KCl for 3 min at room temperature to lyse erythrocytes. The samples were then treated with prefixative (methanol:acetic acid 3:1) and centrifuged. The cellular pellets were resuspended in 5 ml of methanol, then centrifuged. Treatment with fixative (methanol: acetic acid 5:1) followed by centrifugation was repeated twice for 20 min. Lymphocytes in fresh fixative were dropped onto clean iced slides, air-dried and stained in 2% Giemsa (Sigma, Milano, Italy). Two slides/sample were analysed. MN microscopic scoring was performed blindly only on lymphocytes with preserved cytoplasm by the same experienced scorer. Two thousand binucleated (BN) cells (1000 BN/slide) were scored for each sample. Cells were cytologically scored using the cytome approach to evaluate viability status (necrosis, apoptosis), mitotic status (mononucleated, binucleated, multinucleated) and chromosomal damage or instability status (presence of micronuclei, nucleoplasmic bridges, nucleoplasmic buds) (Fenech, 2007). The cytokinesis-block proliferation index (CBPI) was calculated as: CBPI = (M1 + 2 M2 + 3 M3) / N where M1–M3 represent the numbers of cells with 1–3 or more nuclei and N is the total number of viable cells scored (excluding necrotic and apoptotic cells). The results are reported as MNBN/1000 BN cells. All the experimental analyses were performed in the Laboratory of the Environmental Carcinogenesis. 2.5. Statistical analysis The arithmetic means and standard deviations (SD) were used to describe the distributions of the Cytome parameters. In all analyses, log-
Fig. 1. Flow diagram of the enrollment process.
516
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
trasformed MN/1000 (LnMN/1000) values were used because of a remarkable asymmetry and lack of homoscedasticity in MN/1000values, with an increased variability with increasing average values. The distribution resulting from the log-transformation approached a normal distribution, with a satisfactory stabilization of the Variance. However, for clarity, in the presentation of the results, the natural MN/1000 values are shown, obtained by exponentiating the results of the analyses. 2.5.1. Precision of the MN test In scientific terminology, the precision of a measurement system is the degree to which repeated measurements of the same object provide the same results. Precision includes repeatability and reproducibility. Repeatability measures the variation arising when experimental conditions are kept constant by using the same instrument and operator, and measurements are made during a short time period; reproducibility is the variation among different instruments and operators or different measurements on the same subjects over longer time periods when the same measurement process was used. 2.5.1.1. Repeatability. The repeatability of the reading values of MNBN1000BN and of the PI results obtained by the blinded scoring of two slides (1 for each culture/subject) was evaluated by using Bivariate correlation to assess the relationship between the two reading series for each sampling, calculating the Spearman's coefficient used as the variables are not normally distributed. The two series of readings were compared by means of the Wilcoxon test for paired samples.
3. Results From December 2014 to July 2015, the participation in the study was proposed to 187 platelet donors (102 males and 85 females); exclusion criteria were reported by 16 of them (10 males and 6 females). Of the 171 eligible individuals, 34 (19.9%; 20 males and 14 females) declined to participate because of the required long term commitment. In all, 137 individuals were enrolled: they completed the questionnaire on demographic and personal characteristics and the diet questionnaire and provided a 10 ml blood sample. After the first round, one individual (a man) dropped out the study because he had an invasive cancer diagnosed and 22 did no longer comply with the study, leaving 114 individuals who completed the 3-month protocol. The 6-month assessment protocol was completed by 108 individuals as 6 individuals (3 males and 3 females) dropped out the study. A final check before performing the statistical analyses revealed that five of the 108 individuals were ineligible and were excluded from the analyses: a female who worked as a radiotherapy technician and four men who reported previous jobs that were consistent with asbestos exposure. Thus 103 of 132 eligible participants (78.0%) were considered in the present analysis. The flow diagram of the enrollment process is reported in Fig. 1. The main characteristics of the 103 individuals who completed the study protocol are reported in Table 1: as compared to the Italian Table 1 Main characteristics of the study population. Males
2.5.1.2. Reproducibility. The reproducibility of the MN test was evaluated by computing a) the Mean and Standard Error of the Standard Deviation of the 3 sets of measurements in the 103 individuals undergone the MN test 3 times at approximately 3-months intervals, b) the intraclass correlation coefficient, which was used to test the correlation between the 3 different measurements in the same individual. 2.5.2. Determinants of MNBN frequency In order to assess the combined effect of age and gender, while taking into account measurements of MN replicated 3 times in each subject, a Mixed Model was fitted to the data. Mixed models are statistical models containing both fixed effects and random effects and are particularly useful in settings where repeated measurements are made on the same statistical units, because they allow to accommodate the intraindividual variability while evaluating the effects of individual characteristics (fixed effects). In the mixed model used in the present analysis, LnMN/1000 was the dependent variable, gender and age (in 4 categories) were included as fixed effects, and individual subjects as random effects. 2.5.3. Estimation of “normal” individual values and reference group values The results of the Mixed model analysis provided estimates of the coefficients measuring the difference in average Ln MN/1000 values between males and females, and between different age groups. Furthermore, it provided an estimate of the residual Variance in Ln MN/1000 values, after the effect of age and gender had been taken into account. This residual variance, that encompasses both the variability among individuals of the same age and gender, and the variability within the same individual when tested in different instances, was used to estimate the standard deviation (SD) of the distribution of Ln MN/1000 values. The gender- and age class-specific average Ln MN/1000 estimated by the model ±1.96 SD was used to estimate the range of values that are expected in 95% of normal individuals. Similarly, the Standard Error of the mean in each subgroup, computed as SD/√Ni, (where Ni is the number of subjects in that subgroup), was used to estimate the 95% Confidence Interval of age and Gender-specific average of LnMN/ 1000 values in that subgroup.
Females
N.
%
N
%
Total n. individuals
50
100.0
53
100.0
Age group ≤30 31–43 44–52 N52
14 13 10 13
28.0 26.0 20.0 26.0
12 13 13 15
22.6 24.5 24.5 28.3
Years of education ≤8 anni 9–13 N13
7 25 18
14.0 50.0 36.0
7 32 14
13.2 60.4 30.8
Marital status Unmarried Married Separated/divorced Widow
18 29 3 0
36.0 58.0 6.0 0
20 27 5 1
37.7 50.9 9.4 1.9
Place of residence Not industrialized Industrialized§
10 40
20.0 80.0
8 45
14.1 84.9
Tobacco smoking Never Current Former
28 8 14
56.0 16.0 32.1
26 10 17
49.1 18.9 32.1
BMI (weight/height) b25.0 25.0–29.9 ≥30.0
23 21 6
46.0 42.0 12.0
33 12 8
62.3 32.0 15.1
Physical activity None Moderate Vigorous
6 15 29
12.0 30.0 58.0
10 24 19
18.9 45.3 38.8
Family history of cancer None First degree only Second degree only First and second degree
12 12 16 10
24.0 24.0 32.0 20.0
14 15 18 6
26.4 28.3 34.0 11.3
§ The industrialized area is characterized by industries on an extensive scale. In the study area industries, are mainly represented by several oil refineries, a zinc oxide factory, storages of fats, vegetal oil, chemical and petrochemical products, production and storage of gas.
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
population, our study group shows a higher rate of graduated individuals (over 30% vs 20%), a slightly lower rate of never smokers (52% vs 64%) and a lower rate of individuals who reported no physical activity (15% vs 39%). The median intervals between the first and second blood sampling and between the second and third sampling were 3.2 months (IQ range 3.0–4.0) and 3.4 months (IQ range 3.1–4.1), respectively. The comparison of the questionnaires administered at the three blood sampling did not show any difference in the current life style habits and the analyses of the two food diaries did not show substantial variation in the mean per-day intake of the most relevant food components (i.e. proteins, fats, sugars, vitamins, micronutrients and total energy intake) [data not shown]. All the CBMN cytome parameters were considered in the scoring procedures. The mean values (with the corresponding SD) of mononucleated cells with MN (MNMono1000Mono), binucleated cells with MN (MNBN 1000BN), nucleoplasmic bridges (NPB1000BN) nucleoplasmic buds (NBUD1000BN) in binucleated cells and the proliferation index (PI), obtained in the three samplings, according to gender and agegroup are reported in Supplementary Table 1. No significant difference was detected in the mean MNBN frequency measured in the same subjects at three months intervals. A statistically significant difference was observed among the MNMono frequencies detected in the three samplings in males. The mean frequency of NPB1000BN and of NBUD1000BN were below 0.1 and 0.4 respectively in males and females. The PI values showed significant variability among the three sampling in females but not in males. Very rare necrotic or apoptotic cells were found in the samples.
517
The main parameters associated with the chromosomal damage, MNBN1000BN and PI were considered in the following analyses. 3.1. Repeatability In Fig. 2 (a, b) are reported the correlations between the values obtained in two separate scoring sessions for each of the three samplings and the mean values of MNBN1000BN and PI (±SE and SD) of the two slides for each sampling obtained by the scoring of 618 slides in two separate scoring sessions (2 slides/participant/ sampling). A strong correlation between the results obtained in the two scoring sessions for each of the three samplings was observed for both MNBN1000BN (R2 values were 0.83, 0.74 and 0.68; P b 0.0001) and PI values (R2 values were 0.69, 0.62 and 0.65; P b 0.0001). 3.2. Reproducibility The mean values of MNBN1000BN in the three samplings according to gender and age are reported in Table 2: no statistically significant difference in the mean values are observed in the three samplings for all individuals (β = −0.009; P = 0.82). In the three samplings the observed MNBN1000BN values were statistically significant higher in female compared to males (β = 0.411; P b 0.0001) and increased with increasing age (ß = 0.518; P b 0.0001). A high correlation between the 3 different measurements in the same individual was observed (Intraclass Correlation Coefficient (ICC) = 0.905, (95% CI = 0.868–0.933, P b 0.0001) (Fig. 3). The Mean Squared Error MSE in the analysis of variance was 9.354, corresponding to a
Fig. 2. a. Repeatability of the results of the scoring of the 618 slides performed by the same expert reader: MNBN1000BN b. Repeatability of the results of the scoring of the 618 slides performed by the same expert reader: Proliferation Index (PI).
518
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
Fig. 2 (continued).
Standard Deviation of 3.06, indicating that 95% of MN measurements lie within ±6 points of the true value. No improvement in the ICC of the measurements was observed when the logarithmic transformation was used (ICC = 0,874; 95% CL 0.826–0.911). When the Mixed Model was fitted to the data, a significant increase in average MN/1000 values with increasing age was seen [1.31/1000 (SE = 0.056)] in individuals age 18–30 to an average of 2.43 MN/1000 (SE = 0.054) in individuals 53+ years (P b 0.0001), (corresponding geometric mean 4.4/1.000, SE = 0.51 and 12.72/1.000, SE = 0.49). MN values were also significantly higher in females than in males (mean 2.286, SE = 0.039 and mean 1.729, SE = 0.041, respectively, P b 0.001) (corresponding geometric mean 11.92/1.000 MN, SE = 0.36 and 6.9/1.000 MN, SE = 0.37). Although a significant interaction was observed between gender and age class (P = 0.012), it appeared to be entirely due to the high values observed in the group of females age
41–52, while a similar difference between males and females was observed in the other age groups. (Fig. 4).
3.3. Normal values The mean MN values, with their 95% CI, expected in a group of healthy individuals without significant exposures is reported in Fig. 5 for each gender and age group. It increases with increasing age, and, at any age, is higher in females than in males. Noteworthy, the range of predicted mean values in each group is rather narrow, making these confidence intervals suitable for comparisons with the values observed in a group undergoing monitoring. In addition, the range of age/gender specific expected individual MN values was reported in Fig. 6.
Table 2 MNBN1000BN according to gender and age and for all individuals in the three samplings. N.
1st sampling
2nd sampling
3rd sampling
Mean ± SD (SE)
Mean ± SD (SE)
Mean ± SD (SE)
MNBN1000BN All individuals Males Females
103 50 53
9.64 ± 6.1 (0.60) 7.14 ± 3.95 (0.56) 11.99 ± 6.83 (0.94)
9.19 ± 6.37 (0.63) 6.53 ± 3.95 (0.56) 11.70 ± 7.18 (0.99)
9.49 ± 6.28 (0.62) 6.12 ± 3.98 (0.56) 12.67 ± 6.41 (0.88)
Age class ≤30 31–43 44–52 N52
26 26 23 28
3.92 ± 2.33 (0.46) 9.15 ± 4.54 (0.89) 12.28 ± 5.85 (1.22) 13.21 ± 6.14 (1.16)
3.76 ± 1.99 (0.39) 7.68 ± 3.49 (0.68) 13.26 ± 8.18 (1.71) 12.29 ± 5.39 (1.02)
4.90 ± 2.47 (0.48) 8.08 ± 4.63 (0.91) 12.04 ± 7.64 (1.59) 12.98 ± 5.94 (1.12)
B
P
−0.009
0.82
0.411
b0.0001
0.518
b0.0001
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
519
Fig. 3. Comparison of MNBN1000BN values across the three samplings.
4. Discussion The L-CBMN assay is the most validated technique to detect in human populations the expression of cumulative DNA damage induced by different molecular mechanisms associated with the exposure to genotoxic agents or as a result of genome instability in susceptible subjects (Fenech et al., 2016). However the requisite for the implementation of the assay in the routine surveillance of workers occupationally exposed to genotoxic agents and in clinics as a biomarker of risk for cancer and other chronic diseases, is the availability of reference “normal” values in healthy populations. The HUMN international research group performed several validation studies in order to establish reference range of normal values for MN frequency. An analysis of pooled data on thousands of subjects from 25 laboratories in 16 countries using the L-CBMN assay was carried out (Bonassi et al., 2001) and a N4-fold variation in average values for
Fig. 4. Range of MN values expected in 95% of the individuals of a given age group and gender.
MNBN cells was observed. This analysis revealed a large variability among the laboratories attributable to methods and scoring criteria, genotoxic factors, host factors, with 25% of the observed variability unexplained. Standardized protocol and scoring criteria were established and interlaboratory scoring exercises were performed to evaluate the repeatability and the reproducibility of the MN assay in the different laboratories (Fenech et al., 2003). Recently two micronucleus intercomparison exercises carried out in the framework of the “Realizing the European Network of Biodosimetry” (RENEB), with the aim of assuring a reliability in radiation dose estimation, showed a good agreement in the MN scores among the participant laboratories with N80% of satisfactory dose estimates (Depuydt et al., 2017). A number of studies were carried out to determine the baseline frequencies of MN and other nuclear anomalies measured with the CBMN cytome approach in standard healthy populations minimally exposed to genotoxic agents, in order to have a range of background values to use in comparing the extent of genomic damage in populations exposed to occupational or environmental genotoxic agents or at risk for cancer or chronic degenerative diseases. The main factors known to affect the frequency of MN and other nuclear anomalies were considered in recruiting the populations and were taken into account in the statistical analysed confirming the age and sex as the main factors positively associated with the frequency of the cytogenetic biomarkers (Cho et al. 2016; Coşkun et al., 2013; Donmez-Altuntas and Bitgen, 2012; GarajVrhovac et al., 2008; Gajski et al., 2018; Nefic and Handzic, 2013). Significant confounders associated with the conditions of sampling were also identified such as the sampling period, sampling season and different metereological parameters which were reported to differently modulate in men and women the frequency of MN and other nuclear anomalies (Gajski et al., 2018). Available data on background MN frequencies in different unexposed populations showed different results in terms of range of values and of correlation with age and gender being the main covariates affecting the MN frequency (Cho et al., 2017; Coşkun et al., 2013; DonmezAltuntas and Bitgen, 2012; Garaj-Vrhovac et al., 2008; Gajski et al., 2018; Nefic and Handzic, 2013). These differences possibly associated with difference in ethnicity, lifestyle, climate, socio-ecomical status, emphasize the need for baseline data for any population in a reference laboratory.
520
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
Fig. 5. Age and Gender specific AVERAGE – MNBN1000BN values expected in a group of normal subjects – The 95% CL of the mean showed overlap with those reported in the table.
Conversely, the reproducibility of the MN values in each individual, when measured in different times, has never been systematically assessed, although it is a critical information mainly for the clinical use of the test, that is for using it with the aim of classifying an individual at different levels of risk/exposure/susceptibility. The main aim of our study was to estimate the consistency of CBMN cytome parameters through repeated blood samples in a healthy population in order to establish a range of age- and gender-specific normal values. The choice of frequent platelet donors allows us to recruit individuals with healthy lifestyles. In addition stringent inclusion and exclusion criteria during and after the recruitment were applied to assure the selection of subjects not exposed to well-known chemical or physical genotoxic agents. Furthermore it is likely that repeated platelet donation with plateletpheresis technique minimally affect the leukocyte population and haemopoiesis (Strauss, 2003) and consequently the in vitro proliferative response of lymphocytes and the MN expression. As first step the repeatability of the analyses were evaluated carefully checking any single step of the method. The standardized protocol (Fenech, 2007) was applied: all the samples were processed using the
same batches for culture medium, serum and other reagents and all the slides were blindly scored by the same experienced scorer using validated scoring criteria. The comparison of the scoring values of two slides from different cultures of the same blood samples showed a good correlation for the MN frequency and PI, the main parameters relevant for the evaluation of chromosomal damage, which denotes good repeatability of test results in the same laboratory using standardized conditions. No significant correlation was observed for the frequencies of MNMONO cells detected by scoring two slides from the same samples. This result indicates that this parameter is susceptible to technical factors, such as storage of the samples and the conditions of culture. The other parameters of the cytome approach, nucleoplasmic bridges and buds, induced by nuclear alterations associated with specific exposure, necrosis and apoptosis, were observed only sporadically in the analysed samples. The reproducibility of the data was estimated by the evaluation of the level of consistency among the results of three different samplings at three month interval for the same subjects. A high correlation was
Fig. 6. Range of age/gender specific expected individual MNBN1000BN values.
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
observed among the three different measurements of MNBN frequency obtained in the same individuals (ICC = 0.905). An age related increase of the MNBN frequency and a significant age-gender interaction were observed apparently due to the high values detected in the group of females aged 41–52. These results are in agreement with the evidence from the scientific literature (Bonassi et al., 1995; Bolognesi et al., 1997, 1999). Statistically significant difference in the mean values of MNMONO among the three samplings was observed. The MONOMN cells is a mixed population including preexisting MN cells and cells expressing the DNA damage accumulated in lymphocytes in vivo and escaping the cytochalasin block. The low level of consistency observed for the frequencies of MNMONO among the values obtained in different scoring sessions and in different samplings confirms that the evaluation of this parameter does not yield robust results in biomonitoring studies. No significant effect of sampling season on the frequency of MN and other nuclear abnormalities was observed in the present study. The value of the present study is to provide a model to establish an age- and gender-specific range of acceptable, normal MN values in peripheral blood lymphocytes of healthy individuals which is a necessary step before starting any biomonitoring program. The age and gender range of “normal” CBMN mean values predicted for healthy subjects on the basis of the results of the present study appeared to be enough narrow to be applied for comparison of data obtained in exposed or susceptible subjects at a group level. This supports the use of L-CBMN assay in monitoring programs aimed at the timely detection of noxious exposures. Conversely, the range of age and gender individual MN values predicted by our model show a large variability increasing with age, which prevents the possible use of the L-CBMN assay at individual level to assess the exposure to genotoxic agents or the risk of developing cancer and other chronic diseases. 5. Conclusions Our results suggest that the inter and intraindividual variability commonly observed in MNBN frequency is not merely due to the experimental parameters, but it is related to the intrinsic characteristics of the individuals, such as possible differences in blood composition, in lymphocyte proliferation rate associated with conditions of blood sampling, immunological status, diet, hormone level. Further efforts are needed to identify additional confounding factors modulating the MN expression in order to define the conditions of applicability of the assay. Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2018.10.187. Conflict of interest Declarations of interest: none. Acknowledgments The authors thank Paola Roggieri for technical support. Funding This work was supported by the Compagnia di San Paolo [grant numbers 2013-0907]. References Andreassi, M.G., Barale, R., Iozzo, P., Picano, E., 2011. The association of micronucleus frequency with obesity, diabetes and cardiovascular disease. Mutagenesis 26, 77–83. Bolognesi, C., Abbondandolo, A., Barale, R., Casalone, R., Dalprà, L., De Ferrari, M., Degrassi, F., Forni, A., Lamberti, L., Lando, C., Migliore, L., Padovani, P., Pasquini, P., Puntoni, R., Sbrana, I., Stella, M., Bonassi, S., 1997. Age-related increase of baseline frequencies
521
of sister chromatid exchanges, chromosome aberrations, and micronuclei in human lymphocytes. Cancer Epidemiol. Biomark. Prev. 6, 249–256. Bolognesi, C., Lando, C., Forni, A., Landini, E., Scarpato, R., Migliore, L., Bonassi, S., 1999. Chromosomal damage and ageing: effect on micronuclei frequency in peripheral blood lymphocytes. Age Ageing 28 (4), 393–397. Bonassi, S., Bolognesi, C., Abbondandolo, A., Barale, R., Bigatti, P., Camurri, L., Dalprà, L., De Ferrari, M., Forni, A., Lando, C., 1995. Influence of sex on cytogenetic end points: evidence from a large human sample and review of the literature. Cancer Epidemiol. Biomark. Prev. 4, 671–679. Bonassi, S., Fenech, M., Lando, C., Lin, Y.P., Ceppi, M., Chang, W.P., Holland, N., KirschVolders, M., Zeiger, E., Ban, S., Barale, R., Bigatti, M.P., Bolognesi, C., Jia, C., Digiorgio, M., Ferguson, L.R., Fucic, A., Lima, O.G., Hrelia, P., Krishnaja, A.P., Lee, T.K., Migliore, L., Mikhalevich, L., Mirkova, E., Mosesso, P., Müller, W.U., Odagiri, Y., Scarfi, M.R., Szabova, E., Vorobtsova, I., Vral, A., Zijno, A., 2001. 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 (1), 31–45. Bonassi, S., Znaor, A., Ceppi, M., Lando, C., Chang, W.P., Holland, N., Kirsch-Volders, M., Zeiger, E., Ban, S., Barale, R., Bigatti, M.P., Bolognesi, C., Cebulska-Wasilewska, A., Fabianova, E., Fucic, A., Hagmar, L., Joksic, G., Martelli, A., Migliore, L., Mirkova, E., Scarfi, M.R., Zijno, A., Norppa, H., Fenech, M., 2007. An increased micronucleus frequency in peripheral blood lymphocytes predicts the risk of cancer in humans. Carcinogenesis 28 (3), 625–631. Bonassi, S., El-Zein, R., Bolognesi, C., Fenech, M., 2011. Micronuclei frequency in peripheral blood lymphocytes and cancer risk: evidence from human studies. Mutagenesis 26 (1), 93–100. Cardinale, F., Bruzzi, P., Bolognesi, C., 2012. Role of micronucleus test in predicting breast cancer susceptibility: a systematic review and meta-analysis. Br. J. Cancer 106, 780–790. Cho, N.Y., Kim, K.W., Kim, K.K., 2017. Genomic health status assessed by a cytokinesisblock micronucleus cytome assay in a healthy middle-aged Korean population. Mutat. Res. 814, 7–13. Corbi, S.C., Bastos, A.S., Orrico, S.R., Secolin, R., Dos Santos, R.A., Takahashi, C.S., ScarelCaminaga, R.M., 2014. Elevated micronucleus frequency in patients with type 2 diabetes, dyslipidemia and periodontitis. Mutagenesis 29, 433–439. Coşkun, M., Cayır, A., Coşkun, M., Tok, H., 2013. Evaluation of background DNA damage in a Turkish population measured by means of the cytokinesis-block micronucleus cytome assay. Mutat. Res. 757 (1), 23–27. Depuydt, J., Baeyens, A., Barnard, S., Beinke, C., Benedek, A., Beukes, P., Buraczewska, I., Darroudi, F., De Sanctis, S., Dominguez, I., Monteiro Gil, O., Hadjidekova, V., Kis, E., Kulka, U., Lista, F., Lumniczky, K., M'Kacher, R., Moquet, J., Obreja, D., Oestreicher, U., Pajic, J., Pastor, N., Popova, L., Regalbuto, E., Ricoul, M., Sabatier, L., Slabbert, J., Sommer, S., Testa, A., Thierens, H., Wojcik, A., Vral, A., 2017. RENEB intercomparison exercises analyzing micronuclei (cytokinesis-block micronucleus assay). Int. J. Radiat. Biol. 93 (1), 36–47. Donmez-Altuntas, H., Bitgen, N., 2012. Evaluation of the genotoxicity and cytotoxicity in the general population in Turkey by use of the cytokinesis-block micronucleus cytome assay. Mutat. Res. 748 (1–2), 1–7. El-Zein, R.A., Schabath, M.B., Etzel, C.J., Lopez, M.S., Franklin, J.D., Spitz, M.R., 2006. Cytokinesis-blocked micronucleus assay as a novel biomarker for lung cancer risk. Cancer Res. 66, 6449–6456. El-Zein, R., Vral, A., Etzel, C.J., 2011. Cytokinesis-blocked micronucleus assay and cancer risk assessment. Mutagenesis 26, 101–106. Fenech, M., 2007. Cytokinesis-block micronucleus cytome assay. Nat. Protoc. 2, 1084–1104. Fenech, M., Bonassi, S., Turner, J., Lando, C., Ceppi, M., Chang, W.P., Holland, N., KirschVolders, M., Zeiger, E., Bigatti, M.P., Bolognesi, C., Cao, J., De Luca, G., Di Giorgio, M., Ferguson, L.R., Fucic, A., Lima, O.G., Hadjidekova, V.V., Hrelia, P., Jaworska, A., Joksic, G., Krishnaja, A.P., Lee, T.K., Martelli, A., McKay, M.J., Migliore, L., Mirkova, E., Müller, W.U., Odagiri, Y., Orsiere, T., Scarfì, M.R., Silva, M.J., Sofuni, T., Surralles, J., Trenta, G., Vorobtsova, I., Vral, A., Zijno, A., 2003. HUman MicroNucleus project. Intra- and inter-laboratory variation in the scoring of micronuclei and nucleoplasmic bridges in binucleated human lymphocytes. Results of an international slide-scoring exercise by the HUMN project. Mutat. Res. 534 (1–2), 45–64. Fenech, M., Knasmueller, S., Bolognesi, C., Bonassi, S., Holland, N., Migliore, L., Palitti, F., Natarajan, A.T., Kirsch-Volders, M., 2016. Molecular mechanisms by which in vivo exposure to exogenous chemical genotoxic agents can lead to micronucleus formation in lymphocytes in vivo and ex vivo in humans. Mutat. Res. 770, 12–25. Gajski, G., Gerić, M., Oreščanin, V., Garaj-Vrhovac, V., 2018. Cytokinesis-block micronucleus cytome assay parameters in peripheral blood lymphocytes of the general population: contribution of age, sex, seasonal variations and lifestyle factors. Ecotoxicol. Environ. Saf. 148, 561–570. Garaj-Vrhovac, V., Durinec, M., Kopjar, N., Orescanin, V., 2008. A survey on the cytogenetic status of the Croatian general population by use of the cytokinesis-block micronucleus assay. Mutat. Res. 649 (1–2), 91–100. IAEA, 2011. Cytogenetic Dosimetry: Applications in Preparedness for and Response to Radiation Emergencies. EPR-Biodosimetry, International Atomic Energy Agency, Vienna. Maffei, F., Zolezzi Moraga, J.M., Angelini, S., Zenesini, C., Musti, M., Festi, D., Cantelli-Forti, G., Hrelia, P., 2014. Micronucleus frequency in human peripheral blood lymphocytes as a biomarker for the early detection of colorectal cancer risk. Mutagenesis 29 (3), 221–225. McHugh, M.K., Lopez, M.S., Ho, C.H., Spitz, M.R., Etzel, C.J., El-Zein, R.A., 2013. Use of the cytokinesis-blocked micronucleus assay to detect gender differences and genetic instability in a lung cancer case-control study. Cancer Epidemiol. Biomark. Prev. 22, 135–145. Migliore, L., Coppede, F., Fenech, M., Thomas, P., 2011. Association of micronucleus frequency with neurodegenerative diseases. Mutagenesis 26, 85–92.
522
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
Murgia, E., Ballardin, M., Bonassi, S., Rossi, A.M., Barale, R., 2008. Validation of micronuclei frequency in peripheral blood lymphocytes as early cancer risk biomarker in a nested case–control study. Mutat. Res. 639, 27–34. Mutation Research Special Issue, 2016. In Vivo Chemical Genotoxin Exposure and DNA Damage in Humans measured using the lymphocyte Cytokinesis-Block Micronucleus Assay. 770. Nefic, H., Handzic, I., 2013. The effect of age, sex, and lifestyle factors on micronucleus frequency in peripheral blood lymphocytes of the Bosnian population. Mutat. Res. 753 (1), 1–11. Nersesyan, A., Fenech, M., Bolognesi, C., Mišík, M., Setayesh, T., Wultsch, G., Bonassi, S., Thomas, P., Knasmüller, S., 2016. Use of the lymphocyte cytokinesis-block micronucleus assay in occupational biomonitoring of genome damage caused by in vivo exposure to chemical genotoxins: past, present and future. Mutat. Res. 770 (Pt A), 1–11.
Pala, V., Sieri, S., Palli, D., Salvini, S., Berrino, F., Bellegotti, M., Frasca, G., Tumino, R., Sacerdote, C., Fiorini, L., Celentano, E., Galasso, R., Krogh, V., 2003. Diet in the Italian EPIC cohorts: presentation of data and methodological issues. Tumori 89 (6), 594–607. Strauss, R.G., 2003. Safety of donating multiple products in a single apheresis collection: are we expecting too much? J. Clin. Apher. 18 (3), 135–140. Vral, A., Thierens, H., Baeyens, A., De Ridder, L., 2002. The micronucleus and G2-phase assays for human blood lymphocytes as biomarkers of individual sensitivity to ionizing radiation: limitations imposed by intraindividual variability. Radiat. Res. 157 (4), 472–477.
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
“Normal values” for the lymphocyte cytokinesis-block micronucleus cytome parameters: Repeatability and reproducibility in a healthy reference population Claudia Bolognesi a,⁎, Luigina Bonelli b, Andrea Compalati a, Valentina Ferla b, Laura Stagnaro a, Gianluca Ubezio c, Paolo Bruzzi b a b c
Unit of Environmental Carcinogenesis Ospedale Policlinico San Martino, Genova, Italy Unit of Clinical Epidemiology, Ospedale Policlinico San Martino Genova, Italy Transfusion Centre, Ospedale Policlinico San Martino Genova, Italy
H I G H L I G H T S
G R A P H I C A L
A B S T R A C T
• Good repeatability of the micronuclei and proliferative index was observed. • Consistency of MNBN values in individuals was confirmed in repeated blood samples. • Narrow ranges of normal reference group MNBN values were estimated. • Large variability of age- and genderspecific individual MNBN values was observed. • A model to derive MNBN “normal values” in a healthy population was established.
a r t i c l e
i n f o
Article history: Received 8 June 2018 Received in revised form 8 October 2018 Accepted 13 October 2018 Available online 15 October 2018 Keywords: Micronucleus Cytome assay Biomonitoring Occupational exposure Chromosomal instability Healthy subjects
a b s t r a c t The micronucleus test in peripheral blood lymphocytes is the most widely validated technique to evaluate the DNA damage and chromosomal instability in human populations. The test is largely applied in monitoring environmental and occupational exposure to genotoxic agents. It was also proposed as a biomarker of risk/susceptibility for cancer and other degenerative diseases. The availability of “normal values” in healthy populations is a main requisite for the assay application in human biomonitoring. Age and gender-related ranges of micronucleated binucleated cells (MNBN) baseline values were established in a group of 103 healthy platelet donors (50 males and 53 females) not recently exposed to genotoxic agents and characterized for demographic, lifestyle and dietary factors. Repeatability of the test by the same scorer was evaluated. Reproducibility was estimated through analysis of repeated blood samples. High correlation between the results of the three blood samplings in two separate scoring sessions of MNBN/1000BN (R2 values were 0.83, 0.74 and 0.68; P b 0.0001) and PI values (R2 values were 0.69, 0.62 and 0.65; P b 0.0001) was detected. High consistency among the values obtained in three different samplings in the same individual was observed (Intraclass Correlation Coefficient (ICC) = 0.905, (95% CI = 0.868–0.933, P b 0.0001) The range of “normal” values predicted on the basis of the results of the present study appears to be sufficiently narrow to warrant application of the assay in the comparison
⁎ Corresponding author at: Environmental Carcinogenesis Unit, Ospedale Policlinico San Martino, L.go Rosanna Benzi, 10, 16132 Genova, Italy. E-mail address: [email protected] (C. Bolognesi).
https://doi.org/10.1016/j.scitotenv.2018.10.187 0048-9697/© 2018 Elsevier B.V. All rights reserved.
514
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
of data obtained from groups of exposed or susceptible subjects, supporting its use in preventive programs. The large inter-individual variability predicted by the model used in the present study hampers a clinical application of the assay at individual level. The method applied in the present study represents a generally applicable model to derive “normal values” in any population, as an essential step before starting a biomonitoring study. © 2018 Elsevier B.V. All rights reserved.
1. Introduction The cytokinesis-block micronucleus assay in peripheral lymphocytes (L-CBMN assay), due to its ability to detect both structural and numerical chromosomal aberrations and some epigenetic effects, is one of the most widely applied methods to evaluate genome damage in humans (Fenech et al., 2016). L-CBMN assay is recommended by the International Atomic Energy Agency (IAEA) for biodosimetry of ionizing radiation exposure (IAEA, 2011) and is largely used in biomonitoring occupational and accidental exposure to genotoxic agents (Nersesyan et al., 2016). A recent analysis of N500 studies available in the scientific literature on the application of the L-CBMN assay in human populations carried out in the framework of the HUMN (HUman MicroNucleus) project, confirms the usefulness of the assay to assess the exposure to a large number of occupational or environmental genotoxic compounds and mixtures. However, a large variability was observed among the studies, partially due to the limitations in study design, different experimental protocols applied and heterogeneity of the populations considered in terms of genetic susceptibility and lifestyle/concurrent exposures impairing the detection of a dose-response relationship (Nersesyan et al., 2016; Mutation Research Special Issue, 2016). Furthermore L-CBMN assay was proposed as a marker of risk for cancer and other chronic degenerative diseases, based on the results of prospective studies in large cohorts of disease-free subjects (Bonassi et al., 2007; Murgia et al., 2008) and of large case-control studies showing significant increase of MN frequency in cancer patients as compared to control groups (Bonassi et al., 2011; El-Zein et al. 2006 and 2011; Cardinale et al., 2012; McHugh et al., 2013; Maffei et al., 2014; Migliore et al., 2011; Andreassi et al., 2011). Although the L-CBMN assay is considered a suitable and reliable biomarker to assess DNA damage in humans, its implementation is limited by the large inter- and intra-individual variability (Bonassi et al., 2011; Fenech et al., 2003; Vral et al., 2002) observed in the large majority of the available biomonitoring studies which prevented the definition of reliable ranges of “normal values” for the MN frequency in healthy populations against which the MN levels, detected in exposed groups or susceptible subjects, could be compared. The availability of “normal values” is important for the application of the assay in biomonitoring occupational exposure where the extent of the increase is evaluated at the group level but it's critical for a potential clinical use of the assay as a biomarker of risk/susceptibility at the individual level. The analysis of the key components of the variability, the experimental factors and the intrinsic characteristics of the individuals, is a necessary premise to establish a reliable baseline range of “normal” MN frequencies. The availability of standardized protocol and scoring criteria should minimize the variability due to experimental parameters. The comparison of repeated analyses carried out on the same samples under optimized experimental conditions gives the indication of the residual variability. The estimation of the intraindividual variability of the cytome parameters over time in healthy subjects, suitably selected as minimally exposed to environmental genotoxins, is the subsequent step to evaluate if a single measure of MN frequency is representative of the individual chromosomal instability.
To address these issues, we designed a longitudinal study in healthy subjects without current exposures to known carcinogenic or mutagenic agents, who accepted to undergo 3 blood drawings for research purposes at approximately 3–month intervals, and to complete a questionnaire focused on their dietary and life-style habits. Aims of study were: • To estimate the repeatability of the MN analysis in the same laboratory following standardized protocol and scoring criteria • To evaluate the reproducibility of the MN frequency in individual subjects through analyses of repeated blood samplings • To establish the range of normal values of baseline MN frequency
2. Materials and methods 2.1. Study design and population The study was approved by the Ethical Committee of the Liguria region (P.R.027REG2014) and was carried out at the Ospedale Policlinico San Martino of Genoa (Italy). Each participant received written detailed information on the study purpose and methods and signed a written informed consent. The study population was a sample of blood donors who volunteered to participate. The enrollment of blood donors guaranteed screening for health risks, such as several viral and bacterial infections, drug addiction, and chronic and degenerative diseases that could affect the results of the experimental tests. We selected platelet donors because the frequency of platelet donation fitted the time table of the study. Donors were recruited at the Transfusion Centre. They were contacted at the time of a platelet donation and evaluated for their interest in participating to the study and eligibility. We planned to enroll 100 individuals (50 males and 50 females). Eligible were donors aged 18–65 years selected to ensure adequate representativeness of age and gender. Individuals were excluded if they reported any recent exposure to organic solvents, pesticides, paints, welding fumes and documented past exposure to asbestos. Individuals who underwent radiological examinations in the previous 30 days and those who had had a diagnosis of diabetes mellitus were also ineligible, due to the high MN frequency reported in these patients (Corbi et al., 2014). In order to address the intra-individual variability of the single biomarkers, analyses were carried out in blood samples taken at enrolment and after three and six months. 2.2. Data collection At enrolment (first round) a detailed questionnaire (registration form in supplementary materials) was used to elicit information on demographic characteristics (years of education, marital status, occupation, area of residence), smoking habits, height and weight, physical activity (intensity and frequency), personal medical history, history of cancer in first and second-degree relatives, use of drugs and micronutrients supplementation. Women were also interviewed on their reproductive history. Dietary information on the frequency of consumption of foods and drinks during a 12-month period prior to enrolment was obtained by a quantitative Food Frequency Questionnaire FFQ
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
that was developed for Italian dietary habits and was used in the European Prospective Project Into Cancer and Nutrition (EPIC)—Italy study (Pala et al., 2003). The two questionnaires were administered by a trained interviewer at the Transfusion Centre immediately before a platelet donation. At the three and six month sampling (second and third round, respectively) the participants were asked to check information provided at the enrolment in order to record any change occurred afterwards. In addition, they were required to fill in a 3-day food diary before donation to control for substantial variations over time in the mean per-day intake of the most relevant food components as collected at the baseline interview. 2.3. Blood collection At each round, participants provided a 10 ml sample of peripheral blood that was taken at the moment of the platelet donation. Blood samples were collected in a heparinized vial and maintained at room temperature for a maximum of 24 h before being processed. 2.4. Cytokinesis-blocked micronucleus assay in peripheral lymphocytes Heparinized blood samples, sent to the lab within 24 h, were used for the establishment of the lymphocyte cultures. The standardized protocol for the cytokinesis-block micronucleus cytome assay was applied (Fenech, 2007). Three sterile whole blood cultures for each sample were prepared. 0.3 ml aliquot of whole blood per each culture was incubated at 37 °C in 4.7 ml of RPMI 1640 (Life Technologies, Milano, Italy) supplemented with 10% fetal bovine serum (Gibco BRL, Life Technologies SrL, Milano, Italy), 1.5% phytohemoagglutinin (Murex Biotech,
515
Dartford, UK), 100 Unit/ml penicillin and 100 μg/ml streptomycin. After 44 h, Cytochalasin B (Sigma, Milano, Italy) was added at a concentration of 6 μg/ml. At the end of incubation at 37 °C for 72 h, cells were centrifuged (1000 rpm, 10 min) then treated with 5 ml of 0.075 mM KCl for 3 min at room temperature to lyse erythrocytes. The samples were then treated with prefixative (methanol:acetic acid 3:1) and centrifuged. The cellular pellets were resuspended in 5 ml of methanol, then centrifuged. Treatment with fixative (methanol: acetic acid 5:1) followed by centrifugation was repeated twice for 20 min. Lymphocytes in fresh fixative were dropped onto clean iced slides, air-dried and stained in 2% Giemsa (Sigma, Milano, Italy). Two slides/sample were analysed. MN microscopic scoring was performed blindly only on lymphocytes with preserved cytoplasm by the same experienced scorer. Two thousand binucleated (BN) cells (1000 BN/slide) were scored for each sample. Cells were cytologically scored using the cytome approach to evaluate viability status (necrosis, apoptosis), mitotic status (mononucleated, binucleated, multinucleated) and chromosomal damage or instability status (presence of micronuclei, nucleoplasmic bridges, nucleoplasmic buds) (Fenech, 2007). The cytokinesis-block proliferation index (CBPI) was calculated as: CBPI = (M1 + 2 M2 + 3 M3) / N where M1–M3 represent the numbers of cells with 1–3 or more nuclei and N is the total number of viable cells scored (excluding necrotic and apoptotic cells). The results are reported as MNBN/1000 BN cells. All the experimental analyses were performed in the Laboratory of the Environmental Carcinogenesis. 2.5. Statistical analysis The arithmetic means and standard deviations (SD) were used to describe the distributions of the Cytome parameters. In all analyses, log-
Fig. 1. Flow diagram of the enrollment process.
516
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
trasformed MN/1000 (LnMN/1000) values were used because of a remarkable asymmetry and lack of homoscedasticity in MN/1000values, with an increased variability with increasing average values. The distribution resulting from the log-transformation approached a normal distribution, with a satisfactory stabilization of the Variance. However, for clarity, in the presentation of the results, the natural MN/1000 values are shown, obtained by exponentiating the results of the analyses. 2.5.1. Precision of the MN test In scientific terminology, the precision of a measurement system is the degree to which repeated measurements of the same object provide the same results. Precision includes repeatability and reproducibility. Repeatability measures the variation arising when experimental conditions are kept constant by using the same instrument and operator, and measurements are made during a short time period; reproducibility is the variation among different instruments and operators or different measurements on the same subjects over longer time periods when the same measurement process was used. 2.5.1.1. Repeatability. The repeatability of the reading values of MNBN1000BN and of the PI results obtained by the blinded scoring of two slides (1 for each culture/subject) was evaluated by using Bivariate correlation to assess the relationship between the two reading series for each sampling, calculating the Spearman's coefficient used as the variables are not normally distributed. The two series of readings were compared by means of the Wilcoxon test for paired samples.
3. Results From December 2014 to July 2015, the participation in the study was proposed to 187 platelet donors (102 males and 85 females); exclusion criteria were reported by 16 of them (10 males and 6 females). Of the 171 eligible individuals, 34 (19.9%; 20 males and 14 females) declined to participate because of the required long term commitment. In all, 137 individuals were enrolled: they completed the questionnaire on demographic and personal characteristics and the diet questionnaire and provided a 10 ml blood sample. After the first round, one individual (a man) dropped out the study because he had an invasive cancer diagnosed and 22 did no longer comply with the study, leaving 114 individuals who completed the 3-month protocol. The 6-month assessment protocol was completed by 108 individuals as 6 individuals (3 males and 3 females) dropped out the study. A final check before performing the statistical analyses revealed that five of the 108 individuals were ineligible and were excluded from the analyses: a female who worked as a radiotherapy technician and four men who reported previous jobs that were consistent with asbestos exposure. Thus 103 of 132 eligible participants (78.0%) were considered in the present analysis. The flow diagram of the enrollment process is reported in Fig. 1. The main characteristics of the 103 individuals who completed the study protocol are reported in Table 1: as compared to the Italian Table 1 Main characteristics of the study population. Males
2.5.1.2. Reproducibility. The reproducibility of the MN test was evaluated by computing a) the Mean and Standard Error of the Standard Deviation of the 3 sets of measurements in the 103 individuals undergone the MN test 3 times at approximately 3-months intervals, b) the intraclass correlation coefficient, which was used to test the correlation between the 3 different measurements in the same individual. 2.5.2. Determinants of MNBN frequency In order to assess the combined effect of age and gender, while taking into account measurements of MN replicated 3 times in each subject, a Mixed Model was fitted to the data. Mixed models are statistical models containing both fixed effects and random effects and are particularly useful in settings where repeated measurements are made on the same statistical units, because they allow to accommodate the intraindividual variability while evaluating the effects of individual characteristics (fixed effects). In the mixed model used in the present analysis, LnMN/1000 was the dependent variable, gender and age (in 4 categories) were included as fixed effects, and individual subjects as random effects. 2.5.3. Estimation of “normal” individual values and reference group values The results of the Mixed model analysis provided estimates of the coefficients measuring the difference in average Ln MN/1000 values between males and females, and between different age groups. Furthermore, it provided an estimate of the residual Variance in Ln MN/1000 values, after the effect of age and gender had been taken into account. This residual variance, that encompasses both the variability among individuals of the same age and gender, and the variability within the same individual when tested in different instances, was used to estimate the standard deviation (SD) of the distribution of Ln MN/1000 values. The gender- and age class-specific average Ln MN/1000 estimated by the model ±1.96 SD was used to estimate the range of values that are expected in 95% of normal individuals. Similarly, the Standard Error of the mean in each subgroup, computed as SD/√Ni, (where Ni is the number of subjects in that subgroup), was used to estimate the 95% Confidence Interval of age and Gender-specific average of LnMN/ 1000 values in that subgroup.
Females
N.
%
N
%
Total n. individuals
50
100.0
53
100.0
Age group ≤30 31–43 44–52 N52
14 13 10 13
28.0 26.0 20.0 26.0
12 13 13 15
22.6 24.5 24.5 28.3
Years of education ≤8 anni 9–13 N13
7 25 18
14.0 50.0 36.0
7 32 14
13.2 60.4 30.8
Marital status Unmarried Married Separated/divorced Widow
18 29 3 0
36.0 58.0 6.0 0
20 27 5 1
37.7 50.9 9.4 1.9
Place of residence Not industrialized Industrialized§
10 40
20.0 80.0
8 45
14.1 84.9
Tobacco smoking Never Current Former
28 8 14
56.0 16.0 32.1
26 10 17
49.1 18.9 32.1
BMI (weight/height) b25.0 25.0–29.9 ≥30.0
23 21 6
46.0 42.0 12.0
33 12 8
62.3 32.0 15.1
Physical activity None Moderate Vigorous
6 15 29
12.0 30.0 58.0
10 24 19
18.9 45.3 38.8
Family history of cancer None First degree only Second degree only First and second degree
12 12 16 10
24.0 24.0 32.0 20.0
14 15 18 6
26.4 28.3 34.0 11.3
§ The industrialized area is characterized by industries on an extensive scale. In the study area industries, are mainly represented by several oil refineries, a zinc oxide factory, storages of fats, vegetal oil, chemical and petrochemical products, production and storage of gas.
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
population, our study group shows a higher rate of graduated individuals (over 30% vs 20%), a slightly lower rate of never smokers (52% vs 64%) and a lower rate of individuals who reported no physical activity (15% vs 39%). The median intervals between the first and second blood sampling and between the second and third sampling were 3.2 months (IQ range 3.0–4.0) and 3.4 months (IQ range 3.1–4.1), respectively. The comparison of the questionnaires administered at the three blood sampling did not show any difference in the current life style habits and the analyses of the two food diaries did not show substantial variation in the mean per-day intake of the most relevant food components (i.e. proteins, fats, sugars, vitamins, micronutrients and total energy intake) [data not shown]. All the CBMN cytome parameters were considered in the scoring procedures. The mean values (with the corresponding SD) of mononucleated cells with MN (MNMono1000Mono), binucleated cells with MN (MNBN 1000BN), nucleoplasmic bridges (NPB1000BN) nucleoplasmic buds (NBUD1000BN) in binucleated cells and the proliferation index (PI), obtained in the three samplings, according to gender and agegroup are reported in Supplementary Table 1. No significant difference was detected in the mean MNBN frequency measured in the same subjects at three months intervals. A statistically significant difference was observed among the MNMono frequencies detected in the three samplings in males. The mean frequency of NPB1000BN and of NBUD1000BN were below 0.1 and 0.4 respectively in males and females. The PI values showed significant variability among the three sampling in females but not in males. Very rare necrotic or apoptotic cells were found in the samples.
517
The main parameters associated with the chromosomal damage, MNBN1000BN and PI were considered in the following analyses. 3.1. Repeatability In Fig. 2 (a, b) are reported the correlations between the values obtained in two separate scoring sessions for each of the three samplings and the mean values of MNBN1000BN and PI (±SE and SD) of the two slides for each sampling obtained by the scoring of 618 slides in two separate scoring sessions (2 slides/participant/ sampling). A strong correlation between the results obtained in the two scoring sessions for each of the three samplings was observed for both MNBN1000BN (R2 values were 0.83, 0.74 and 0.68; P b 0.0001) and PI values (R2 values were 0.69, 0.62 and 0.65; P b 0.0001). 3.2. Reproducibility The mean values of MNBN1000BN in the three samplings according to gender and age are reported in Table 2: no statistically significant difference in the mean values are observed in the three samplings for all individuals (β = −0.009; P = 0.82). In the three samplings the observed MNBN1000BN values were statistically significant higher in female compared to males (β = 0.411; P b 0.0001) and increased with increasing age (ß = 0.518; P b 0.0001). A high correlation between the 3 different measurements in the same individual was observed (Intraclass Correlation Coefficient (ICC) = 0.905, (95% CI = 0.868–0.933, P b 0.0001) (Fig. 3). The Mean Squared Error MSE in the analysis of variance was 9.354, corresponding to a
Fig. 2. a. Repeatability of the results of the scoring of the 618 slides performed by the same expert reader: MNBN1000BN b. Repeatability of the results of the scoring of the 618 slides performed by the same expert reader: Proliferation Index (PI).
518
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
Fig. 2 (continued).
Standard Deviation of 3.06, indicating that 95% of MN measurements lie within ±6 points of the true value. No improvement in the ICC of the measurements was observed when the logarithmic transformation was used (ICC = 0,874; 95% CL 0.826–0.911). When the Mixed Model was fitted to the data, a significant increase in average MN/1000 values with increasing age was seen [1.31/1000 (SE = 0.056)] in individuals age 18–30 to an average of 2.43 MN/1000 (SE = 0.054) in individuals 53+ years (P b 0.0001), (corresponding geometric mean 4.4/1.000, SE = 0.51 and 12.72/1.000, SE = 0.49). MN values were also significantly higher in females than in males (mean 2.286, SE = 0.039 and mean 1.729, SE = 0.041, respectively, P b 0.001) (corresponding geometric mean 11.92/1.000 MN, SE = 0.36 and 6.9/1.000 MN, SE = 0.37). Although a significant interaction was observed between gender and age class (P = 0.012), it appeared to be entirely due to the high values observed in the group of females age
41–52, while a similar difference between males and females was observed in the other age groups. (Fig. 4).
3.3. Normal values The mean MN values, with their 95% CI, expected in a group of healthy individuals without significant exposures is reported in Fig. 5 for each gender and age group. It increases with increasing age, and, at any age, is higher in females than in males. Noteworthy, the range of predicted mean values in each group is rather narrow, making these confidence intervals suitable for comparisons with the values observed in a group undergoing monitoring. In addition, the range of age/gender specific expected individual MN values was reported in Fig. 6.
Table 2 MNBN1000BN according to gender and age and for all individuals in the three samplings. N.
1st sampling
2nd sampling
3rd sampling
Mean ± SD (SE)
Mean ± SD (SE)
Mean ± SD (SE)
MNBN1000BN All individuals Males Females
103 50 53
9.64 ± 6.1 (0.60) 7.14 ± 3.95 (0.56) 11.99 ± 6.83 (0.94)
9.19 ± 6.37 (0.63) 6.53 ± 3.95 (0.56) 11.70 ± 7.18 (0.99)
9.49 ± 6.28 (0.62) 6.12 ± 3.98 (0.56) 12.67 ± 6.41 (0.88)
Age class ≤30 31–43 44–52 N52
26 26 23 28
3.92 ± 2.33 (0.46) 9.15 ± 4.54 (0.89) 12.28 ± 5.85 (1.22) 13.21 ± 6.14 (1.16)
3.76 ± 1.99 (0.39) 7.68 ± 3.49 (0.68) 13.26 ± 8.18 (1.71) 12.29 ± 5.39 (1.02)
4.90 ± 2.47 (0.48) 8.08 ± 4.63 (0.91) 12.04 ± 7.64 (1.59) 12.98 ± 5.94 (1.12)
B
P
−0.009
0.82
0.411
b0.0001
0.518
b0.0001
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
519
Fig. 3. Comparison of MNBN1000BN values across the three samplings.
4. Discussion The L-CBMN assay is the most validated technique to detect in human populations the expression of cumulative DNA damage induced by different molecular mechanisms associated with the exposure to genotoxic agents or as a result of genome instability in susceptible subjects (Fenech et al., 2016). However the requisite for the implementation of the assay in the routine surveillance of workers occupationally exposed to genotoxic agents and in clinics as a biomarker of risk for cancer and other chronic diseases, is the availability of reference “normal” values in healthy populations. The HUMN international research group performed several validation studies in order to establish reference range of normal values for MN frequency. An analysis of pooled data on thousands of subjects from 25 laboratories in 16 countries using the L-CBMN assay was carried out (Bonassi et al., 2001) and a N4-fold variation in average values for
Fig. 4. Range of MN values expected in 95% of the individuals of a given age group and gender.
MNBN cells was observed. This analysis revealed a large variability among the laboratories attributable to methods and scoring criteria, genotoxic factors, host factors, with 25% of the observed variability unexplained. Standardized protocol and scoring criteria were established and interlaboratory scoring exercises were performed to evaluate the repeatability and the reproducibility of the MN assay in the different laboratories (Fenech et al., 2003). Recently two micronucleus intercomparison exercises carried out in the framework of the “Realizing the European Network of Biodosimetry” (RENEB), with the aim of assuring a reliability in radiation dose estimation, showed a good agreement in the MN scores among the participant laboratories with N80% of satisfactory dose estimates (Depuydt et al., 2017). A number of studies were carried out to determine the baseline frequencies of MN and other nuclear anomalies measured with the CBMN cytome approach in standard healthy populations minimally exposed to genotoxic agents, in order to have a range of background values to use in comparing the extent of genomic damage in populations exposed to occupational or environmental genotoxic agents or at risk for cancer or chronic degenerative diseases. The main factors known to affect the frequency of MN and other nuclear anomalies were considered in recruiting the populations and were taken into account in the statistical analysed confirming the age and sex as the main factors positively associated with the frequency of the cytogenetic biomarkers (Cho et al. 2016; Coşkun et al., 2013; Donmez-Altuntas and Bitgen, 2012; GarajVrhovac et al., 2008; Gajski et al., 2018; Nefic and Handzic, 2013). Significant confounders associated with the conditions of sampling were also identified such as the sampling period, sampling season and different metereological parameters which were reported to differently modulate in men and women the frequency of MN and other nuclear anomalies (Gajski et al., 2018). Available data on background MN frequencies in different unexposed populations showed different results in terms of range of values and of correlation with age and gender being the main covariates affecting the MN frequency (Cho et al., 2017; Coşkun et al., 2013; DonmezAltuntas and Bitgen, 2012; Garaj-Vrhovac et al., 2008; Gajski et al., 2018; Nefic and Handzic, 2013). These differences possibly associated with difference in ethnicity, lifestyle, climate, socio-ecomical status, emphasize the need for baseline data for any population in a reference laboratory.
520
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
Fig. 5. Age and Gender specific AVERAGE – MNBN1000BN values expected in a group of normal subjects – The 95% CL of the mean showed overlap with those reported in the table.
Conversely, the reproducibility of the MN values in each individual, when measured in different times, has never been systematically assessed, although it is a critical information mainly for the clinical use of the test, that is for using it with the aim of classifying an individual at different levels of risk/exposure/susceptibility. The main aim of our study was to estimate the consistency of CBMN cytome parameters through repeated blood samples in a healthy population in order to establish a range of age- and gender-specific normal values. The choice of frequent platelet donors allows us to recruit individuals with healthy lifestyles. In addition stringent inclusion and exclusion criteria during and after the recruitment were applied to assure the selection of subjects not exposed to well-known chemical or physical genotoxic agents. Furthermore it is likely that repeated platelet donation with plateletpheresis technique minimally affect the leukocyte population and haemopoiesis (Strauss, 2003) and consequently the in vitro proliferative response of lymphocytes and the MN expression. As first step the repeatability of the analyses were evaluated carefully checking any single step of the method. The standardized protocol (Fenech, 2007) was applied: all the samples were processed using the
same batches for culture medium, serum and other reagents and all the slides were blindly scored by the same experienced scorer using validated scoring criteria. The comparison of the scoring values of two slides from different cultures of the same blood samples showed a good correlation for the MN frequency and PI, the main parameters relevant for the evaluation of chromosomal damage, which denotes good repeatability of test results in the same laboratory using standardized conditions. No significant correlation was observed for the frequencies of MNMONO cells detected by scoring two slides from the same samples. This result indicates that this parameter is susceptible to technical factors, such as storage of the samples and the conditions of culture. The other parameters of the cytome approach, nucleoplasmic bridges and buds, induced by nuclear alterations associated with specific exposure, necrosis and apoptosis, were observed only sporadically in the analysed samples. The reproducibility of the data was estimated by the evaluation of the level of consistency among the results of three different samplings at three month interval for the same subjects. A high correlation was
Fig. 6. Range of age/gender specific expected individual MNBN1000BN values.
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
observed among the three different measurements of MNBN frequency obtained in the same individuals (ICC = 0.905). An age related increase of the MNBN frequency and a significant age-gender interaction were observed apparently due to the high values detected in the group of females aged 41–52. These results are in agreement with the evidence from the scientific literature (Bonassi et al., 1995; Bolognesi et al., 1997, 1999). Statistically significant difference in the mean values of MNMONO among the three samplings was observed. The MONOMN cells is a mixed population including preexisting MN cells and cells expressing the DNA damage accumulated in lymphocytes in vivo and escaping the cytochalasin block. The low level of consistency observed for the frequencies of MNMONO among the values obtained in different scoring sessions and in different samplings confirms that the evaluation of this parameter does not yield robust results in biomonitoring studies. No significant effect of sampling season on the frequency of MN and other nuclear abnormalities was observed in the present study. The value of the present study is to provide a model to establish an age- and gender-specific range of acceptable, normal MN values in peripheral blood lymphocytes of healthy individuals which is a necessary step before starting any biomonitoring program. The age and gender range of “normal” CBMN mean values predicted for healthy subjects on the basis of the results of the present study appeared to be enough narrow to be applied for comparison of data obtained in exposed or susceptible subjects at a group level. This supports the use of L-CBMN assay in monitoring programs aimed at the timely detection of noxious exposures. Conversely, the range of age and gender individual MN values predicted by our model show a large variability increasing with age, which prevents the possible use of the L-CBMN assay at individual level to assess the exposure to genotoxic agents or the risk of developing cancer and other chronic diseases. 5. Conclusions Our results suggest that the inter and intraindividual variability commonly observed in MNBN frequency is not merely due to the experimental parameters, but it is related to the intrinsic characteristics of the individuals, such as possible differences in blood composition, in lymphocyte proliferation rate associated with conditions of blood sampling, immunological status, diet, hormone level. Further efforts are needed to identify additional confounding factors modulating the MN expression in order to define the conditions of applicability of the assay. Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2018.10.187. Conflict of interest Declarations of interest: none. Acknowledgments The authors thank Paola Roggieri for technical support. Funding This work was supported by the Compagnia di San Paolo [grant numbers 2013-0907]. References Andreassi, M.G., Barale, R., Iozzo, P., Picano, E., 2011. The association of micronucleus frequency with obesity, diabetes and cardiovascular disease. Mutagenesis 26, 77–83. Bolognesi, C., Abbondandolo, A., Barale, R., Casalone, R., Dalprà, L., De Ferrari, M., Degrassi, F., Forni, A., Lamberti, L., Lando, C., Migliore, L., Padovani, P., Pasquini, P., Puntoni, R., Sbrana, I., Stella, M., Bonassi, S., 1997. Age-related increase of baseline frequencies
521
of sister chromatid exchanges, chromosome aberrations, and micronuclei in human lymphocytes. Cancer Epidemiol. Biomark. Prev. 6, 249–256. Bolognesi, C., Lando, C., Forni, A., Landini, E., Scarpato, R., Migliore, L., Bonassi, S., 1999. Chromosomal damage and ageing: effect on micronuclei frequency in peripheral blood lymphocytes. Age Ageing 28 (4), 393–397. Bonassi, S., Bolognesi, C., Abbondandolo, A., Barale, R., Bigatti, P., Camurri, L., Dalprà, L., De Ferrari, M., Forni, A., Lando, C., 1995. Influence of sex on cytogenetic end points: evidence from a large human sample and review of the literature. Cancer Epidemiol. Biomark. Prev. 4, 671–679. Bonassi, S., Fenech, M., Lando, C., Lin, Y.P., Ceppi, M., Chang, W.P., Holland, N., KirschVolders, M., Zeiger, E., Ban, S., Barale, R., Bigatti, M.P., Bolognesi, C., Jia, C., Digiorgio, M., Ferguson, L.R., Fucic, A., Lima, O.G., Hrelia, P., Krishnaja, A.P., Lee, T.K., Migliore, L., Mikhalevich, L., Mirkova, E., Mosesso, P., Müller, W.U., Odagiri, Y., Scarfi, M.R., Szabova, E., Vorobtsova, I., Vral, A., Zijno, A., 2001. 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 (1), 31–45. Bonassi, S., Znaor, A., Ceppi, M., Lando, C., Chang, W.P., Holland, N., Kirsch-Volders, M., Zeiger, E., Ban, S., Barale, R., Bigatti, M.P., Bolognesi, C., Cebulska-Wasilewska, A., Fabianova, E., Fucic, A., Hagmar, L., Joksic, G., Martelli, A., Migliore, L., Mirkova, E., Scarfi, M.R., Zijno, A., Norppa, H., Fenech, M., 2007. An increased micronucleus frequency in peripheral blood lymphocytes predicts the risk of cancer in humans. Carcinogenesis 28 (3), 625–631. Bonassi, S., El-Zein, R., Bolognesi, C., Fenech, M., 2011. Micronuclei frequency in peripheral blood lymphocytes and cancer risk: evidence from human studies. Mutagenesis 26 (1), 93–100. Cardinale, F., Bruzzi, P., Bolognesi, C., 2012. Role of micronucleus test in predicting breast cancer susceptibility: a systematic review and meta-analysis. Br. J. Cancer 106, 780–790. Cho, N.Y., Kim, K.W., Kim, K.K., 2017. Genomic health status assessed by a cytokinesisblock micronucleus cytome assay in a healthy middle-aged Korean population. Mutat. Res. 814, 7–13. Corbi, S.C., Bastos, A.S., Orrico, S.R., Secolin, R., Dos Santos, R.A., Takahashi, C.S., ScarelCaminaga, R.M., 2014. Elevated micronucleus frequency in patients with type 2 diabetes, dyslipidemia and periodontitis. Mutagenesis 29, 433–439. Coşkun, M., Cayır, A., Coşkun, M., Tok, H., 2013. Evaluation of background DNA damage in a Turkish population measured by means of the cytokinesis-block micronucleus cytome assay. Mutat. Res. 757 (1), 23–27. Depuydt, J., Baeyens, A., Barnard, S., Beinke, C., Benedek, A., Beukes, P., Buraczewska, I., Darroudi, F., De Sanctis, S., Dominguez, I., Monteiro Gil, O., Hadjidekova, V., Kis, E., Kulka, U., Lista, F., Lumniczky, K., M'Kacher, R., Moquet, J., Obreja, D., Oestreicher, U., Pajic, J., Pastor, N., Popova, L., Regalbuto, E., Ricoul, M., Sabatier, L., Slabbert, J., Sommer, S., Testa, A., Thierens, H., Wojcik, A., Vral, A., 2017. RENEB intercomparison exercises analyzing micronuclei (cytokinesis-block micronucleus assay). Int. J. Radiat. Biol. 93 (1), 36–47. Donmez-Altuntas, H., Bitgen, N., 2012. Evaluation of the genotoxicity and cytotoxicity in the general population in Turkey by use of the cytokinesis-block micronucleus cytome assay. Mutat. Res. 748 (1–2), 1–7. El-Zein, R.A., Schabath, M.B., Etzel, C.J., Lopez, M.S., Franklin, J.D., Spitz, M.R., 2006. Cytokinesis-blocked micronucleus assay as a novel biomarker for lung cancer risk. Cancer Res. 66, 6449–6456. El-Zein, R., Vral, A., Etzel, C.J., 2011. Cytokinesis-blocked micronucleus assay and cancer risk assessment. Mutagenesis 26, 101–106. Fenech, M., 2007. Cytokinesis-block micronucleus cytome assay. Nat. Protoc. 2, 1084–1104. Fenech, M., Bonassi, S., Turner, J., Lando, C., Ceppi, M., Chang, W.P., Holland, N., KirschVolders, M., Zeiger, E., Bigatti, M.P., Bolognesi, C., Cao, J., De Luca, G., Di Giorgio, M., Ferguson, L.R., Fucic, A., Lima, O.G., Hadjidekova, V.V., Hrelia, P., Jaworska, A., Joksic, G., Krishnaja, A.P., Lee, T.K., Martelli, A., McKay, M.J., Migliore, L., Mirkova, E., Müller, W.U., Odagiri, Y., Orsiere, T., Scarfì, M.R., Silva, M.J., Sofuni, T., Surralles, J., Trenta, G., Vorobtsova, I., Vral, A., Zijno, A., 2003. HUman MicroNucleus project. Intra- and inter-laboratory variation in the scoring of micronuclei and nucleoplasmic bridges in binucleated human lymphocytes. Results of an international slide-scoring exercise by the HUMN project. Mutat. Res. 534 (1–2), 45–64. Fenech, M., Knasmueller, S., Bolognesi, C., Bonassi, S., Holland, N., Migliore, L., Palitti, F., Natarajan, A.T., Kirsch-Volders, M., 2016. Molecular mechanisms by which in vivo exposure to exogenous chemical genotoxic agents can lead to micronucleus formation in lymphocytes in vivo and ex vivo in humans. Mutat. Res. 770, 12–25. Gajski, G., Gerić, M., Oreščanin, V., Garaj-Vrhovac, V., 2018. Cytokinesis-block micronucleus cytome assay parameters in peripheral blood lymphocytes of the general population: contribution of age, sex, seasonal variations and lifestyle factors. Ecotoxicol. Environ. Saf. 148, 561–570. Garaj-Vrhovac, V., Durinec, M., Kopjar, N., Orescanin, V., 2008. A survey on the cytogenetic status of the Croatian general population by use of the cytokinesis-block micronucleus assay. Mutat. Res. 649 (1–2), 91–100. IAEA, 2011. Cytogenetic Dosimetry: Applications in Preparedness for and Response to Radiation Emergencies. EPR-Biodosimetry, International Atomic Energy Agency, Vienna. Maffei, F., Zolezzi Moraga, J.M., Angelini, S., Zenesini, C., Musti, M., Festi, D., Cantelli-Forti, G., Hrelia, P., 2014. Micronucleus frequency in human peripheral blood lymphocytes as a biomarker for the early detection of colorectal cancer risk. Mutagenesis 29 (3), 221–225. McHugh, M.K., Lopez, M.S., Ho, C.H., Spitz, M.R., Etzel, C.J., El-Zein, R.A., 2013. Use of the cytokinesis-blocked micronucleus assay to detect gender differences and genetic instability in a lung cancer case-control study. Cancer Epidemiol. Biomark. Prev. 22, 135–145. Migliore, L., Coppede, F., Fenech, M., Thomas, P., 2011. Association of micronucleus frequency with neurodegenerative diseases. Mutagenesis 26, 85–92.
522
C. Bolognesi et al. / Science of the Total Environment 652 (2019) 513–522
Murgia, E., Ballardin, M., Bonassi, S., Rossi, A.M., Barale, R., 2008. Validation of micronuclei frequency in peripheral blood lymphocytes as early cancer risk biomarker in a nested case–control study. Mutat. Res. 639, 27–34. Mutation Research Special Issue, 2016. In Vivo Chemical Genotoxin Exposure and DNA Damage in Humans measured using the lymphocyte Cytokinesis-Block Micronucleus Assay. 770. Nefic, H., Handzic, I., 2013. The effect of age, sex, and lifestyle factors on micronucleus frequency in peripheral blood lymphocytes of the Bosnian population. Mutat. Res. 753 (1), 1–11. Nersesyan, A., Fenech, M., Bolognesi, C., Mišík, M., Setayesh, T., Wultsch, G., Bonassi, S., Thomas, P., Knasmüller, S., 2016. Use of the lymphocyte cytokinesis-block micronucleus assay in occupational biomonitoring of genome damage caused by in vivo exposure to chemical genotoxins: past, present and future. Mutat. Res. 770 (Pt A), 1–11.
Pala, V., Sieri, S., Palli, D., Salvini, S., Berrino, F., Bellegotti, M., Frasca, G., Tumino, R., Sacerdote, C., Fiorini, L., Celentano, E., Galasso, R., Krogh, V., 2003. Diet in the Italian EPIC cohorts: presentation of data and methodological issues. Tumori 89 (6), 594–607. Strauss, R.G., 2003. Safety of donating multiple products in a single apheresis collection: are we expecting too much? J. Clin. Apher. 18 (3), 135–140. Vral, A., Thierens, H., Baeyens, A., De Ridder, L., 2002. The micronucleus and G2-phase assays for human blood lymphocytes as biomarkers of individual sensitivity to ionizing radiation: limitations imposed by intraindividual variability. Radiat. Res. 157 (4), 472–477.