buccal micronucleus cytome assay: results of an intra- and inter-laboratory scoring comparison

© The Author 2015. Published by Oxford University Press on behalf of the UK Environmental Mutagen Society. All rights reserved. For permissions, please e-mail: [email protected].

1

Mutagenesis, 2015, 1–11doi:10.1093/mutage/gev017

Original Article

Original Manuscript

Buccal micronucleus cytome assay: results of an intra- and inter-laboratory scoring comparisonClaudia Bolognesi*, Paola Roggieri, Monica Ropolo, Philip Thomas1, Maryam Hor1, Michael Fenech1, Armen Nersesyan2 and Siegfried Knasmueller2

Environmental Carcinogenesis Unit, IRCCS Azienda Ospedaliera Universitaria San Martino-IST Istituto Nazionale Ricerca sul Cancro, Largo Rosanna Benzi 10, Genova 16132, Italy, 1Commonwealth and Scientific Industrial Research Organization, Genome Health and Personalised Nutrition Laboratory, Food and Nutrition Flagship, Adelaide, South Australia 5000, Australia, 2Institute of Cancer Research, Department of Medicine I, Comprehensive Cancer Center, Medical University of Vienna, Borschkegasse 8a, A-1090 Vienna, Austria

*To whom correspondence should be addressed. Tel: +39 0105558304; Fax: +39 0105558344; Email: [email protected]

Abstract

The buccal micronucleus cytome (BMCyt) assay is a minimally invasive approach for measuring DNA damage, cell proliferation, cell differentiation and cell death in exfoliated buccal cells. The main limitation for its use is the lack of knowledge about inter- and intra-laboratory variability in scoring micronuclei and other end points included in the cytome approach. In order to identify the main sources of variability across the BMCyt biomarkers, a scoring exercise was carried out between three experienced laboratories using the same set of slides and an identical set of detailed scoring criteria and associated images for the different end points. Single batches of slides were prepared from pooled samples of four groups of subjects characterised by different frequencies of cell types and micronuclei, namely Down syndrome patients, head and neck cancer patients undergoing radiotherapy and two age- and gender-matched control groups. A good agreement among the laboratories in the identification of normal differentiated cells and of micronuclei was obtained. A 3-fold and 20-fold increase in the frequency of micronucleated cells and micronuclei in differentiated cells of Down syndrome patients and in cancer patients, respectively, compared to matched controls, was a consistent result in the three laboratories. The scores of other cell types and nuclear anomalies, such as basal, binucleated, condensed chromatin and karyorrhectic cells showed significant disagreement between and within laboratories indicating that their evaluation using the current visual scoring protocol does not yield robust results for these parameters. The guidelines for BMCyt assay application could be improved by combining the anomalies associated with cell death (condensed chromatin and karyorrhectic cells) in a single category and by defining more stringent criteria in classifying basal cell, binucleated cells and buds.

Introduction

The buccal micronucleus cytome (BMCyt) assay is a cost-effective minimally invasive approach for evaluating genomic damage, cell death and cytostasis in cells of the aerodigestive tract in humans (1,2). The method was initially restricted to the measurement of micronuclei (MNi) frequency, which is a valuable biomarker of DNA damage successfully applied since 1980 to evaluate the geno-toxic impact of inhalation and local exposure to hazardous environ-mental agents, malnutrition, lifestyle and inherited genetic defects

in DNA repair (1). The application of additional end points in the ‘cytome approach’ could further extend the scope and utility of the BMCyt assay by measuring a wider range of cellular pathologies. The association of specific nuclear anomalies and cytome profiles with Down syndrome (DS) (3,4) and Alzheimer’s disease (5), sug-gests a potential use of the BMCyt assay in identifying individuals at increased risk of developing these diseases. However, only a lim-ited number of studies in the scientific literature, including the most recent ones, report data on various cell types or nuclear anomalies, other than MNi.

Mutagenesis Advance Access published March 20, 2015 at L

ibrary MedU

ni Vienna (10076821) on M

arch 23, 2015http://m

utage.oxfordjournals.org/D

ownloaded from

http://www.oxfordjournals.org/

mailto:[email protected]?subject=

http://mutage.oxfordjournals.org/

The application of the ‘multiple end point’ cytome approach in the BMCyt assay requires a standardised protocol for buccal cell collection and slide preparation, an appropriate staining procedure and detailed scoring criteria to identify and quantify all the dif-ferent cell types and nuclear anomalies. The international Human Micronucleus (HUMN) project (www.humn.org) coordinating group launched the HUMNxl (‘XL’ designating eXfoLiated cells) project to validate the BMCyt assay, following the same strategy that was applied to the cytokinesis-block micronucleus (CBMN) assay in human peripheral blood lymphocytes (6,7). A standardised pro-tocol for the BMCyt assay was first established by determining and describing in detail the procedures for cell sampling, slide prepara-tion, staining and scoring criteria taking into account the strengths and weaknesses of available procedures, confounding factors and the possibility of staining artefacts (2). Recently, a worldwide survey of the current use and practice of the BMCyt assay, carried out by the HUMNxl project consortium, showed a large variability of the MNi frequencies within the individual data sets and indicated the need for an international inter-laboratory scoring exercise (8).

As a first step, it became evident that a validation study should be carried out applying the established standardised protocol and focusing exclusively on the scoring procedures. To meet this goal, a more detailed description of the scoring criteria to identify the dif-ferent biomarkers was developed and recently reported, providing a comprehensive set of photomicrographs with multiple examples of each cell type and nuclear anomaly (9).

The aims of the present study were: (i) to evaluate the inter-lab-oratory and intra-laboratory variations in scoring the same set of slides among three experienced laboratories using the more com-prehensive scoring criteria and (ii) to identify scoring parameters over which there is strongest agreement or disagreement in order to further improve the guidelines for slide scoring and to define acceptable degrees of variations between scorers. For this purpose, single batches of slides were prepared from pooled samples of four different groups of subjects who, based on previous studies, were expected to be characterised by different MNi frequency and cytome profiles. Head and neck cancer patients undergoing radiotherapy involving radiation exposure of the buccal mucosa with the conven-tional dose rates were considered to provide cells with an increased MNi frequency, >10 times the baseline level (10). Down syndrome cases (DSCs) were included because their MNi frequencies in buccal cells may represent an intermediate level (possibly up to ~5 times the level in controls) and have been shown to differ significantly with respect to the frequency of basal, binucleated and karyorrhectic cells (3). Two further groups were included, namely the age- and gender-matched control groups for both cancer and DS patients, i.e. healthy older and younger subjects, respectively.

Materials and methods

Subject recruitmentThis study was approved by the local Ethical Advisory Board at the National Cancer Research Institute in Genoa and by the Ethical Commission of the Institute of Cancer Research, Medical University of Vienna. The study involved a total of 30 participants including, young healthy controls, DSCs, old healthy controls and oral cancer patients undergoing radiotherapy (OCR).

Eight head and neck cancer patients (five males and three females aged 56–77 years, including four smokers, two ex-smokers and two non-smokers) and eight age-matched healthy controls (four males and four females aged 50–72 years, including two smokers, three ex-smokers

and three non-smokers) were recruited at the National Cancer Research Institute in Genoa (Italy). The range of the cumulative radiation dose received by the OCR patients was between 20 and 22 Gy.

Seven DSCs (four males and three females aged 18–26 years) and seven age-matched healthy controls (four males and three females aged 18–26 years) all non-smokers were recruited from the hospital ‘Krankenanstalt Rudolfstiftung’ and from the cultural centre ‘Ich bin OK’ for people with disabilities in Vienna (Austria). All participants (in case of DSCs also their parents) were informed about the aims and methods included in the study and gave a written consent to provide a sample of exfoliated buccal cells in order to perform the BMCyt assay.

Buccal cell collection, preparation and stainingBuccal cells were collected from both cheeks using a small-headed tooth brush following the standardised protocol (3). Samples were collected in Saccomanno’s fixative and stored at 4°C for 1–2 months until processing.

Samples were processed at the Unit of Chemical Carcinogenesis, National Cancer Research Institute in Genoa in collaboration with researchers from the Institute of Cancer Research (Medical University of Vienna) following the previously reported standardised proto-col (2). Briefly, cells from the right and the left cheeks were trans-ferred into separate centrifuge tubes and spun for 10 min at 580g at room temperature. The supernatants were removed and replaced with 10 ml of buccal cell buffer [0.01 M Tris–HCl (Sigma T-3253), 0.1 M EDTA tetra sodium salt (Sigma E-5391), 0.02 M sodium chlo-ride (Sigma S5886)] at pH 7.0. Cells were resuspended and spun at 580g for 10 min, then washed twice in the same buffer. Left and right cheek cell populations were pooled in a final volume of 5.0 ml of buccal cell buffer. The cell suspensions were briefly vortexed and then homogenised for 2–3 min in a hand homogeniser (Wheaton Scientific 0.1–0.15 mm gauge) at medium intensity to increase the number of single cells in the suspensions. Samples from subjects belonging to each group were pooled in a 50-ml tube and drawn into a syringe with a 21-G gauge needle, then expelled to improve cellular separation. Subsequently, the cells were passed through a 100 μm nylon filter (Millipore, code NYH02500) held in a swinex holder (Millipore, code SX0002500) to remove large cell aggregates. Cells were further spun at 580g for 10 min and the supernatant removed. They were resuspended in 10 ml of buccal cell buffer and the cell concentration determined using a counting chamber (Burker haemocytometer). Cell suspensions were prepared containing 80 000 cells/ml of buccal cell buffer. Fifty microlitres of dimethyl sulphoxide (Sigma D1435) per millilitre of cell suspension were added to fur-ther improve cell separation. One hundred and twenty microlitres of each cell suspension were added to each cytospin cup and spun at 40.65g (600 r.p.m.) for 5 min in a cytocentrifuge (Shandon Cytospin 3, Thermo Electron Corporation). Slides containing one spot of cells were air-dried for 10 min and then fixed in ethanol:acetic acid (3:1) for 10 min. Subsequently, slides were air-dried for at least 10 min. Fixed slides were treated for 1 min each in 50 and 20% ethanol and washed with deionised water for 2 min prior to staining. The slides were then treated with 5.0 M hydrochloric acid for 30 min followed by washing for 3 min in running tap water. Moist slides were treated with Schiff’s reagent (Sigma 3552016) at room temperature in the dark for 60 min, washed in running tap water for 5 min and rinsed in deionised water. Slides were stained in 0.2% Light green (Histoline Laboratories, Polyscience Inc., code 02753) for 30 sec, rinsed in deionised water and allowed to air-dry at room temperature prior to being mounted in DPX (Sigma 44581).

2 C. Bolognesi et al.

at Library M

edUni V

ienna (10076821) on March 23, 2015


Dow

nloaded from

http://www.humn.org


Scoring criteriaSlides were coded and scored (without knowledge about the group from which the cells were obtained) using transmitted light and flu-orescence microscopy with a far red filter at 1000 magnification as previously described (3,9). A stepwise approach was applied to score the slides. Firstly, the frequency of various cell types was evaluated in 1000 cells. A second step involved the determination of frequencies of nuclear abnormalities, MNi and nuclear buds (NBUDs), in 1000 nor-mal differentiated cells (which include both transitional and terminally differentiated cells). Cells containing MNi or NBUDs on bright field were also confirmed as being positive by examining the cells under far red fluorescence. Different cell types including basal, differentiated (DIFF) condensed chromatin (CC), karyorrhectic (KHC), pyknotic (PYC), karyolytic (KYL) and binucleated (BN) cells were identified fol-lowing the detailed criteria as recently described (Figure 1) (9). DIFF cells, including both mononucleated and binucleated cells, were then scored for the presence of nuclear abnormalities indicative of chromo-somal instability or DNA damage, specifically MNi and NBUDs and MNi were scored separately in mononucleated and BN DIFF cells.

Scoring processThree experienced laboratories who were involved in the develop-ment of the methodology (3,9,11) participated in the scoring exercise: the Environmental Carcinogenesis Unit, IRCCS AOU San Martino-IST-Istituto Nazionale Ricerca sul Cancro, Genova, Italy, the Institute of Cancer Research, Medical University of Vienna, Vienna, Austria and CSIRO Food and Nutrition Flagship, Adelaide, Australia.

Two experienced scorers from each laboratory (i.e. a total of six scorers) were involved in the exercise. Each scorer evaluated four slides prepared from each pool of cells that had been collected from the four groups of subjects. The schedule for each scorer is described in Figure 2. Each scorer analysed the same set of 16 slides twice. For each slide, 1000 cells were screened in the first step to identify the different cell types. Subsequently, in a second step, 1000 differentiated cells were scored for nuclear anomalies (MNi and NBUDs). A total number of 2000 cells and 2000 differentiated cells/slide were scored by each scorer in two sessions. This procedure allowed us to determine the intra-scorer, inter-scorer and inter-laboratory variability for each parameter.

Statistical analysisThe results are presented as means ± SDs for the different sets of slides from the cellular scores obtained in the three laboratories. One-way analysis of variance was applied to evaluate the signifi-cance of intra-scorer, inter-scorer and inter-laboratory variability. All analyses were conducted using the Statistical Package for Social Sciences for Windows, Version 20.0. P values < 0.05 were considered to be statistically significant.

Results

Combined results obtained with buccal cells from the four groupsThe results from the three laboratories concerning the frequencies of the various cell types from the four cohorts are summarised in Table 1.

Figure 1. Different cell types and nuclear anomalies evaluated in the buccal micronucleus cytome assay.

Inter-laboratory buccal MN scoring comparison 3

at Library M

edUni V

ienna (10076821) on March 23, 2015


Dow

nloaded from


These data are also shown graphically in Figure 3A–E. Different cytome profiles were found in the young and old control cohorts. Differentiated cell values were significantly higher (P < 0.001) and KYL cell frequen-cies were found to be lower (P < 0.001) in young control subjects com-pared with old controls in all laboratories. The data obtained with DSC samples show an altered cytome profile compared with the matched young controls with an increase in BN cells (P = 0.049) and KYL cells (P = 0.05) observed in all three laboratories. The results relating to OCR samples reveal a substantial increase in the frequency of basal

(P = 0.080), differentiated (P = 0.000) and BN cells (P = 0.000) and a decrease of KHC (P = 0.000) and KYL cells (P = 0.000) compared with old controls in all three laboratories.

A discrepancy among the scoring groups was evident in frequencies of some buccal cell types. A non-statistically significant increase in the basal cell frequency in DSC samples, compared with the matched con-trols was seen only in data from Vienna and Adelaide. Furthermore, a statistically significant increase in the BN cell frequency in the same group was recorded only in the Adelaide laboratory, while a

Figure 2. Scoring schedule.

Table I. Frequencies of buccal cell types (basal, differentiated, binucleated, condensed chromatin, karyorrhectic, pyknotic and karyolytic) from different groups of subjects: separate results from the three laboratories involved in the scoring exercise and their combined data

Status Laboratory Mean (SD)/1000 cells

Basal DIFF BN CC KHC PYC KYL

Younger healthy control (YHC)

Genoa 2.31 (1.08) 855.62 (12.95) 8.19 (2.71) 13.37 (3.86) 28.75 (7.62) 2.56 (1.15) 82.94 (21.38)Vienna 0.69 (0.70) 857.75 (18.31) 25.12 (10.65) 13.19 (7.48) 51.44 (20.98) 1.44 (0.96) 50.44 (15.09)Adelaide 5.37 (2.70) 888.75 (10.52) 11.12 (2.78) 14.69 (3.26) 38.12 (22.72) 2.44 (2.31) 44.50 (17.61)Combined data 2.79 (2.59) 867.37 (23.00) 14.81 (9.84) 13.75 (5.15) 39.44 (20.30) 2.14 (1.64) 59.29 (24.68)

Down syndrome cases (DSCs)

Genoa 2.37 (1.31) 838.56 (22.46) 7.31 (3.16) 12.93 (5.28) 32.75 (6.15) 1.62 (1.45) 103.81 (17.97)Vienna 1.06 (0.77) 852.43 (32.21) 33.25 (13.82) 6.12 (4.91)* 38.00 (21.65) 1.81 (1.33) 67.31 (25.33)Adelaide 6.62 (2.22) 859.12 (36.84) 18.56 (6.15)* 24.31 (14.42) 46.25 (31.54) 2.50 (3.42) 42.37 (25.42)Combined data 3.35 (2.84) 850.04 (31.63) 19.70 (13.84) 14.46 (11.85) 39.00 (22.60) 1.98 (2.26) 71.17 (34.12)

Older healthy controls (OHC)

Genoa 1.87 (1.78) 672.81 (31.75) 4.81 (2.69) 13.90 (6.55) 34.56 (8.30) 3.94 (3.09) 268.12 (27.18)Vienna 2.06 (1.34) 644.56 (59.30) 20.69 (9.31) 4.69 (3.61) 37.62 (19.80) 1.75 (0.93) 288.62 (54.15)Adelaide 7.87 (4.18) 684.87 (35.74) 10.31 (7.88) 12.44 (5.41) 57.31 (26.92) 2.56 (2.60) 222.44 (38.09)Combined data 3.94 (3.88) 667.42 (46.29) 11.94 (9.70) 10.34 (6.62) 43.17 (21.96) 2.75 (2.51) 259.73 (49.15)

Oral cancer radiotherapy (OCR)

Genoa 2.44 (1.26) 814.56 (10.62)* 54.00 (7.02)* 12.12 (2.28) 8.94 (1.48)* 1.37 (0.96)* 106.56 (6.31)*Vienna 6.44 (2.28)* 814.00 (50.27)* 56.12 (21.31)* 7.75 (4.61) 15.69 (9.84)* 2.31 (1.25) 90.69 (48.12)*Adelaide 6.69 (2.94) 804.94 (33.66)* 54.94 (9.63)* 12.00 (7.41) 26.56 (20.03)* 3.62 (2.19) 62.12 (23.27)*Combined data 5.19 (2.96) 810.0 (34.94)* 55.02 (9.72)* 10.62 (5.49) 17.06 (14.61)* 2.44 (1.78) 86.46 (35.63)*

Results shown are mean (SD) of the scores of two scorers from each laboratory for four slides in duplicate per group (total N = 16). Basal: basal cells; DIFF: differentiated cells; BN: binucleated cells; CC: condensed chromatin cells; KHC: karyorrhectic cells; PYC: pyknotic cells; KYL: karyolysis cells.

*P < 0.001 compared with the matched controls (i.e. OHC in the case of OCR; YHC in the case of DSC).


at Library M

edUni V

ienna (10076821) on March 23, 2015


Dow

nloaded from


significant decrease in the CC cells was recorded only in Vienna. Table 2 shows the results concerning the frequencies of nuclear anomalies including micronucleated cells (MNCs), MNi and NBUDs in both

mononucleated and BN cells from the same set of slides. No statisti-cally significant differences were observed in regard to the different nuclear anomalies between the young and old control group.

Figure 3. Box plots of frequencies of buccal cytome end points (A) basal cells, (B) differentiated (DIFF) cells, (C) binucleated (BN) cells; (D) karyolytic (KYL) cells; (E) mononucleated differentiated cells with micronuclei (MNC), from young healthy controls (YHC), Down syndrome cases (DSCs), old healthy controls (OHC) and oral cancer patients undergoing radiotherapy (OCR): results from Genoa (white box), Vienna (light-grey box) and Adelaide (dark-grey box). The results for Basal, DIFF, BN and KYL cells represent frequency per 1000 total cells. The results for MNC represent frequency per 1000 DIFF cells. Box plots: the centre horizontal line marks the median of the sample. The length of each box shows the range within which the central 50% of the values fall, with the top and bottom of the box at the first and third quartiles. The vertical T-lines represent the 75% + 1.5 × the interquartile range and the 25% − 1.5 × the interquartile range. Points for individuals that fall above or below 1.5 to 3.0 box heights from the top or bottom of the filled box are considered outliers marked with small ‘o’. Individual points above or below 3 box heights are considered extreme outliers and are marked with asterisks.


at Library M

edUni V

ienna (10076821) on March 23, 2015


Dow

nloaded from


A statistically significant increase (P < 0.000) in the frequencies of MNC, MNi and NBUDs in the DSC group relative to the matched young controls was evident when the mean of the results from the three laboratories was examined.

Data obtained with OCR samples show a large increase in the fre-quency of nuclear anomalies associated with the exposure to a high cumu-lative dose of ionising radiation. The frequency of MNC and MNi was ~20-fold higher than the value observed in the matched control group. In these samples, MNi were also detected in BN cells with a frequency that was ~11 times higher than that found in mononucleated cells. The MN frequency in BN cells is based on the observation that there were only 55 BN cells per 1000 differentiated cells and they contained 13.5 MNi that equate to a frequency of 245 MNi per 1000 BN cells.

Despite the difference among the laboratories in the mean MNi fre-quency in DSC samples (3.9, 1.81 and 0.37 MN/1000 cells in Genoa, Vienna and Adelaide, respectively), the frequency ratios relative to the values in the matched controls were very similar (3.66, 4.1 and 3.08, respectively). Higher MNC and MNi frequencies were observed in the old control subjects relative to the young controls in Genoa and Adelaide.

Inter-laboratory, inter-scorer and intra-scorer variationTable 3 and Figure 3A–E summarise the data on inter-laboratory variation on scoring different cell types. A statistically significant

difference was observed for different biomarkers, namely basal cells, BN, CC, KHC and PYC. The highest inter-laboratory difference was detected for basal cells with the mean frequency in Adelaide being 2.5 and 3.0 times higher compared with Vienna and Genoa, respectively. The mean frequencies of BN cells in Vienna were 1.4 and 1.8 times higher compared with the values recorded in Adelaide and Genoa, respectively. The frequency of CC cells was comparable in the Adelaide and Genoa laboratories and ~2 times higher as the value obtained in Vienna. Less marked inter-laboratory differences were observed for KHC and PYC frequency.

Table 4 and Figure 3E describe the inter-laboratory variations on scoring different nuclear anomalies. It can be seen that a statistically significant inter-laboratory difference was observed only in regard to NBUDs.

The results on intra-laboratory variations on scoring differ-ent cell types and nuclear anomalies are shown in Tables 5 and 6, respectively. Statistical significant differences between the scorers were observed in regard to basal cells (Genoa only), BN (Vienna only), CC (all laboratories), PYC and KHC (Vienna and Adelaide). A good intra-laboratory agreement was found for scoring DIFF ad KYL cells (Table 5). No intra-laboratory variations concerning MNi and nuclear abnormalities were observed (Table 6).

The results of the intra-scorer variation with regard to differ-ent cell types are reported in Table 7. No statistically significant

Table 2. Frequencies of nuclear abnormalities (micronucleated cells, micronuclei and nuclear buds) in mononucleated and binucleated buccal cells from different groups of subjects: results from the three laboratories involved in the scoring exercise

Status Laboratory Mean (SD)/1000 differentiated cells

MNC MN MNCBN MNBN NBUD

Younger healthy control (YHC)

Genoa 0.87 (0.88) 1.00 (0.97) 0.00 (0.00) 0.00 (0.00) 1.00 (0.73)Vienna 0.44 (0.51) 0.44 (0.51) 0.00 (0.00) 0.00 (0.00) 2.56 (1.71)Adelaide 0.12 (0.34) 0.19 (0.54) 0.00 (0.00) 0.00 (0.00) 2.44 (0.81)Combined data 0.48 (0.68) 0.54 (0.77) 0.00 (0.00) 0.00 (0.00) 2.00 (1.35)

Down syndrome cases (DSCs)

Genoa 3.19 (1.47)* 3.31 (1.49)* 0.00 (0.00) 0.00 (0.00) 6.44 (1.75)*Vienna 1.81 (0.83)* 1.87 (0.96)* 0.00 (0.00) 0.00 (0.00) 6.12 (2.36)*Adelaide 0.37 (0.62) 0.81 (1.64) 0.12 (0.34)* 0.12 (0.34)* 2.81 (1.94)Combined data 1.79 (1.54)* 2.00 (1.71)* 0.042 (0.20)* 0.042 (0.20)* 5.12 (2.59)*

Older healthy controls (OHC)

Genoa 1.44 (1.31) 1.62 (1.36) 0.00 (0.00) 0.00 (0.00) 0.81 (0.91)Vienna 0.37 (0.50) 0.50 (0.73) 0.00 (0.00) 0.00 (0.00) 3.31 (1.92)Adelaide 0.50 (0.52) 0.94 (1.34) 0.00 (0.00) 0.00 (0.00) 2.44 (1.75)Combined data 0.77 (0.97) 1.02 (1.25) 0.00 (0.00) 0.00 (0.00) 2.19 (1.87)

Oral cancer radiotherapy (OCR)

Genoa 24.25 (3.25)* 25.87 (4.04)* 12.18 (2.66)* 14.31 (4.01)* 1.06 (1.24)Vienna 19.37 (3.72)* 22.56 (4.62)* 10.06 (2.64)* 13.31 (2.82)* 8.31 (2.21)*Adelaide 16.62 (3.91)* 19.62 (4.99)* 10.31 (4.14)* 12.81 (6.26)* 6.18 (3.29)*Combined data 20.08 (4.78)* 22.69 (5.16)* 10.85 (3.30)* 13.47 (4.53)* 5.19 (3.87)*

MNC: mononucleated differentiated cells containing micronuclei; MN micronuclei in mononucleated differentiated cells; MNCBN: binucleated differentiated cells containing micronuclei; MNBN: micronuclei in binucleated differentiated cells; NBUD: nuclear buds in all differentiated cells.

*P < 0.001 compared with the matched controls (i.e. OHC in the case of OCR; YHC in the case of DSCs).

Table 3. Inter-laboratory variation of scoring different cell types in exfoliated buccal cells based on data from all of the scored slides, scored in duplicate, from the four groups combined for both scorers of each laboratory (N = 64)

Laboratory N Mean (SD)/1000 cells

Basal DIFF BN CC KRK PYC KYL

Genoa 64 2.25 (1.37) 795.39 (75.70) 18.58 (21.07) 13.08 (4.70) 26.25 (12.07) 2.37 (2.08) 140.36 (77.35)Vienna 64 2.56 (2.69) 791.69 (97.02) 33.80 (17.85) 7.94 (6.15) 35.69 (22.35) 1.83 (1.15) 124.26 (103.98)Adelaide 64 6.64 (3.15) 809.42 (84.66) 23.73 (19.69) 15.86 (9.86) 42.06 (27.50) 2.78 (2.66) 92.86 (80.29)P value <0.001 0.472 <0.001 <0.001 <0.001 0.034 0.009

Basal: basal cells; DIFF: differentiated cells; BN: binucleated cells; CC: condensed chromatin cells; KHC: karyorrhectic cells; PYC: pyknotic cells; KYL: kary-olysis cells.


at Library M

edUni V

ienna (10076821) on March 23, 2015


Dow

nloaded from


differences were observed in the results recorded in two independ-ent scoring sessions with the exception of two cases. A significant discrepancy between the two scoring sessions was detected in one scorer from Genoa in regard to the frequency of basal cells (P < 0.007) and in one scorer in Vienna for the frequency of KHC cells (P = 0.028). The highest level of intra-scorer agreement was

observed with biomarkers such as BN and KYL: the P values were ≥0.7 for all the scorers.

Table 8 describes the results of the intra-scorer variation of dif-ferent nuclear anomalies. A high level of agreement was observed for these data for all the scorers. In all cases, there was no difference in the data obtained for the two scoring sessions for each scorer.

Table 4. Inter-laboratory variation of scoring nuclear abnormalities in exfoliated buccal cells based on data from all of the scored slides, scored in duplicate, from the four groups combined for both scorers of each laboratory (N = 64)

Laboratory N Mean (SD)/1000 differentiated cells


Genoa 64 7.44 (10.00) 7.95 (10.70) 3.05 (5.47) 3.58 (6.55) 2.33 (2.67)Vienna 64 5.50 (8.31) 6.34 (9.74) 2.51 (4.58) 3.33 (5.97) 5.08 (3.07)Adelaide 64 4.40 (7.37) 5.39 (8.70) 2.61 (4.92) 3.23 (6.36) 3.47 (2.62)P value 0.135 0.325 0.814 0.950 <0.001

MNC: mononucleated differentiated cells containing micronuclei; MN: micronuclei in mononucleated differentiated cells; MNCBN: binucleated differentiated cells containing micronuclei; MNBN: micronuclei in binucleated differentiated cells; NBUD: nuclear buds in all differentiated cells.

Table 5. Intra-laboratory variation of scoring different cell types in exfoliated buccal cells based on comparison of data from all of the scored slides, scored in duplicate, from the four groups combined between single scorers of each laboratory (N=32).

N Mean (SD)/1000 cells

Basal DIFF BN CC KRK PYC KYL

Genoa GS1 32 2.03 (1.51) 797.72 (65.62) 18.66 (23.49) 11.53 (3.04) 27.06 (11.53) 2.47 (1.63) 140.53 (70.90) GS2 32 2.47 (1.19) 793.06 (85.62) 18.50 (18.72) 14.64 (5.54) 25.44 (12.72) 2.28 (2.48) 140.19 (84.46) P value 0.203 0.808 0.977 0.007 0.594 0.722 0.986Vienna VS1 32 2.12 (3.18) 781.34 (95.49) 25.56 (15.80) 5.19 (3.45) 48.62 (22.67) 2.19 (1.40) 127.47 (105.07) VS2 32 3.00 (2.06) 802.03 (98.95) 39.03 (18.48) 10.69 (7.02) 22.75 (12.48) 1.47 (0.67) 121.06 (104.46) P value 0.196 0.398 0.018 <0.001 <0.001 0.011 0.808Adelaide AS1 32 7.97 (3.04) 819.47 (82.57) 24.31 (21.73) 11.50 (5.20) 20.34 (10.00) 1.56 (1.46) 98.12 (84.06) AS2 32 5.31 (2.69) 799.37 (86.84) 23.16 (17.76) 20.22 (11.47) 63.78 (21.53) 4.00 (3.03) 87.59 (77.32) P value <0.001 0.347 0.816 <0.001 <0.001 <0.001 0.604

Basal: basal cells; DIFF: differentiated cells; BN: binucleated cells; CC: condensed chromatin cells; KHC: karyorrhectic cells; PYC: pyknotic cells; KYL: karyoly-sis cells. The codes GS1, GS2, VS1, VS2, AS1 and AS2 represent the different scorers from each laboratory.

Table 6. Intra-laboratory variation of scoring nuclear abnormalities in exfoliated buccal cells based on comparison of data from all of the scored slides, scored in duplicate, from the four groups combined between single scorers of each laboratory (N = 32)

N Mean (SD)/1000 differentiated cells


Genoa GS1 32 7.91 (10.92) 8.47 (11.62) 3.37 (6.09) 4.00 (7.39) 2.50 (2.82) GS2 32 6.97 (9.15) 7.44 (9.86) 2.72 (4.85) 3.16 (5.66) 2.16 (2.55) P value 0.711 0.703 0.635 0.610 0.611Vienna VS1 32 5.40 (8.42) 6.19 (9.83) 2.28 (4.30) 3.22 (5.91) 4.09 (2.56) VS2 32 5.59 (8.33) 6.50 (9.80) 2.75 (4.89) 3.44 (6.12) 6.06 (3.25) P value 0.929 0.899 0.685 0.885 0.009Adelaide AS1 32 4.34 (7.57) 5.19 (8.66) 2.84 (5.17) 3.50 (6.64) 2.53 (1.19) AS2 32 4.47 (7.30) 5.59 (8.88) 2.37 (4.72) 2.97 (6.16) 4.41 (3.28) P value 0.947 0.854 0.706 0.741 0.003

MNC: mononucleated differentiated cells containing micronuclei; MN: micronuclei in mononucleated differentiated cells; MNCBN: binucleated differentiated cells containing micronuclei; MNBN: micronuclei in binucleated differentiated cells; NBUD: nuclear buds in all differentiated cells. The codes GS1, GS2, VS1, VS2, AS1 and AS2 represent the different scorers from each laboratory.


at Library M

edUni V

ienna (10076821) on March 23, 2015


Dow

nloaded from


Discussion

One of the limitations in the practical application of the MN assay is the large variability of MN frequencies observed within individuals, within the same laboratory and between different laboratories. This variability, besides host factors and multiple exposures to endog-enous and exogenous genotoxic agents, is also caused by methodo-logical differences, i.e. sample processing, slide preparation, use of various staining techniques and differences in the scoring criteria as was shown previously for the CBMN assay with peripheral blood lymphocytes (12–14).

Visual scoring, based on the subjective evaluation of the cytologi-cal parameters, is the most critical methodological step in the BMCyt assay. Inconsistency in visual scoring leads to the intra- and inter-laboratory variability that weakens the possibility of achieving the optimal accuracy. The results of the first international slide-scoring

exercise on the application of the MN assay in human peripheral lymphocytes, carried out in the framework of the HUMN project, showed a substantial inter-laboratory variation in the frequency of MNC cells or MNi and even larger variations of nucleoplasmic bridges in binucleated cells that were harder to score because of lack of experience with this latter biomarker at that time (13). This varia-tion was evident despite the use of slides derived from the same cul-tures and using a common detailed set of scoring criteria. Therefore, it was anticipated that the intra- and inter-laboratory variation may also be even higher in the BMCyt assay because of the larger number of end points and differences in the experience and interpretation of scoring criteria. Furthermore, the BMCyt assay involves more com-plex criteria for identifying the various cell types associated with the differentiation process, as well as selecting cells to be scored for classification of MNi and other nuclear alterations commonly found in buccal cells.

Table 7. Intra-scorer variation for scoring different cell types of exfoliated buccal cells based on duplicate scores of four slides from each of the four groups

Genoa

Scorer and scoring session



GS1 1 16 2.75 (1.53) 797.00 (65.82) 17.62 (21.39) 10.69 (2.84) 27.44 (2.84) 2.81 (1.72) 141.69 (71.93) 2 16 1.31 (1.14) 798.44 (67.57) 19.69 (26.09) 12.37 (3.07) 26.69 (11.40) 2.12 (1.50) 139.37 (72.20) P value 0.007 0.952 0.808 0.118 0.857 0.238 0.928GS2 1 16 2.19 (1.22) 789.56 (87.54) 18.00 (19.33) 14.62 (5.39) 25.19 (13.66) 1.81 (1.17) 142.56 (92.64) 2 16 1.12 (2.47) 796.56 (86.36) 19.00 (18.71) 14.65 (5.86) 25.69 (12.16) 2.75 (3.30) 137.81 (78.41) P value 0.186 0.821 0.883 0.990 0.914 0.292 0.877

Vienna




VS1 1 16 1.68 (3.03) 761.81 (112.88) 28.94 (18.66) 5.81 (4.04) 57.31 (24.72) 2.25 (1.81) 139.37 (122.48) 2 16 2.56 (3.36) 800.87 (72.73) 28.19 (12.94) 4.56 (2.73) 39.94 (17.04) 2.12 (0.88) 140.19 (84.46) P value 0.445 0.254 0.896 0.313 0.028 0.805 0.530VS2 1 16 3.12 (1.89) 805.00 (108.18) 38.75 (17.55) 11.94 (7.65) 19.81 (10.81) 1.56 (0.63) 119.81 (109.79) 2 16 2.87 (2.28) 799.06 (92.26) 39.31 (19.93) 9.44 (6.31) 25.69 (13.66) 1.37 (0.72) 122.31 (104.42) P value 0.738 0.868 0.933 0.322 0.187 0.439 0.947

Adelaide




AS1 1 16 8.12 (3.20) 817.69 (82.00) 25.62 (24.66) 11.69 (5.04) 20.94 (10.28) 1.56 (1.59) 96.68 (82.43) 2 16 7.81 (2.97) 821.25 (85.78) 23.00 (19.08) 11.31 (5.51) 19.75 (10.00) 1.56 (1.36) 99.56 (88.35) P value 0.777 0.905 0.739 0.842 0.743 1.00 0.925AS2 1 16 5.37 (2.82) 802.62 (89.91) 23.00 (18.80) 21.75 (13.84) 63.62 (21.16) 3.69 (2.52) 83.37 (75.27) 2 16 5.25 (2.64) 796.12 (86.47) 23.31 (17.28) 18.69 (8.68) 63.93 (22.58) 4.31 (3.52) 91.81 (81.56) P value 0.898 0.836 0.961 0.459 0.968 0.568 0.763

Basal: basal cells; DIFF: differentiated cells; BN: binucleated cells; CC: condensed chromatin cells; KHC: karyorrhectic cells; PYC: pyknotic cells; KYL: karyoly-sis cells. The codes GS1, GS2, VS1, VS2, AS1 and AS2 represent the different scorers from each laboratory.


at Library M

edUni V

ienna (10076821) on March 23, 2015


Dow

nloaded from


The aim of the present study was to evaluate the inter- and intra-laboratory variation in scoring various cell types that have been char-acterised in the BMCyt assay between three experienced laboratories using the same set of slides, detailed scoring criteria and associated images for different cell types and nuclear anomalies. As a result of experience gained within these laboratories and because of their direct involvement in defining and publishing the latest set of detailed scoring criteria (9), the results reported in the current study reflect the repro-ducibility of visual scoring of the various assay biomarkers that can be obtained under the most optimal conditions that are currently possible.

Buccal cytome profiles of the four cohorts of subjectsThe combined results of the three labs allowed us to identify statisti-cally significant buccal cell cytome profile differences between the dif-ferent groups of subjects considered in this exercise, despite substantial

intra- and inter-laboratory variation. Different buccal cytome profiles were observed in the young and old control groups. The frequency of DIFF cells was significantly higher and the rate of KYL cells was lower in young subjects, but no statistically significant differences were detected in regard to different nuclear anomalies between the young and old con-trols. A significantly altered cytome profile characterised by higher fre-quencies of both BN and KYL cells and a 3-fold increase in the frequency of MNC and MNi was scored for the DSC samples compared with the matched controls. This was a consistent result in all three laboratories and is in agreement with the observations of previous studies (3).

Furthermore, data obtained with OCR samples show a marked increase in the frequency of basal, DIFF and BN cells and a decrease in KHC and KYL cells compared with the matched controls. The significant decrease of the KHC frequency in ORC samples has also been previously reported in Alzheimer’s patients (5) and it was hypothesised that apoptotic responses may be suppressed in an

Table 8. Intra-scorer variation for scoring nuclear abnormalities in exfoliated buccal cells based on duplicate scores of four slides from each of the four groups

Genoa

Scorer and scoring session N Mean (SD)/1000 differentiated cells


GS1 1 16 7.5 (11.06) 8.00 (11.82) 3.06 (5.49) 3.75 (6.83) 2.68 (2.98) 2 16 8.3 (11.12) 8.93 (11.78) 3.69 (6.81) 4.25 (8.14) 2.31 (2.72) P value 0.837 0.824 0.777 0.852 0.713GS2 1 16 6.81 (9.54) 7.12 (10.12) 2.44 (4.38) 2.75 (4.97) 2.31 (2.80) 2 16 7.12 (9.05) 7.75 (9.92) 3.00 (5.41) 3.56 (6.41) 2.00 (2.36) P value 0.925 0.861 0.749 0.692 0.735

Vienna

Scorer and scoring session N Mean (SD)/1000 cells


VS1 1 16 5.62 (9.08) 6.37 (10.63) 2.31 (4.42) 3.25 (6.13) 4.06 (2.26) 2 16 5.19 (8.00) 6.00 (9.32) 2.25 (4.33) 3.18 (5.89) 4.12 (2.89) P value 0.886 0.916 0.968 0.977 0.946VS2 1 16 5.37 (8.20) 6.19 (9.66) 2.50 (4.49) 3.37 (6.15) 5.44 (2.90) 2 16 5.81 (8.72) 6.81 (10.24) 3.00 (5.40) 3.50 (6.29) 6.69 (3.55) P value 0.885 0.860 0.778 0.955 0.284

Adelaide

Scorer and scoring session N Mean (SD)/1000 cells


AS1 1 16 4.06 (7.20) 5.00 (8.58) 2.87 (5.35) 3.50 (6.90) 2.69 (1.25) 2 16 4.62 (8.14) 5.37 (9.00) 2.81 (5.17) 3.50 (6.59) 2.37 (1.15) P value 0.837 0.905 0.973 1.00 0.467AS2 1 16 5.25 (8.43) 6.81 (10.26) 2.69 (5.41) 3.50 (7.42) 4.37 (2.60) 2 16 3.68 (6.14) 4.37 (7.39) 2.06 (4.08) 2.44 (4.77) 4.44 (3.93) P value 0.554 0.447 0.715 0.634 0.958

MNC: mononucleated differentiated cells containing micronuclei; MN: micronuclei in mononucleated differentiated cells; MNCBN: binucleated differentiated cells containing micronuclei; MNBN: micronuclei in binucleated differentiated cells; NBUD: nuclear buds in all differentiated cells. The codes GS1, GS2, VS1, VS2, AS1 and AS2 represent the different scorers from each laboratory.


at Library M

edUni V

ienna (10076821) on March 23, 2015


Dow

nloaded from


attempt to maintain the integrity of the buccal mucosa. It is also notable that a clear increase in the frequencies of nuclear anoma-lies in both mononucleated and BN cells was also detected in OCR samples, associated with the exposure to high cumulative doses of ionising radiations. The frequencies of MNC and MNi in the sam-ples were ~20-fold higher than the value observed in the controls. In these samples, MNi were detected also in BN cells with a frequency 11 times higher than that of the mononucleated cells. Data on MN frequency in BN cells of patients undergoing radiotherapy have not been previously reported in the literature, only MN in mononucle-ated cells was scored (10,15–17). Our results suggest that the use of MN in BN cells may prove to be a more sensitive biomarker of radiation exposure than mononucleated DIFF cells. However, one of the limitations may be the low frequency of BN cells (~5.5%) that was scored among differentiated cells.

Inter- and intra-laboratory variabilityThe evaluation of the inter-laboratory variability showed good agreement in scoring differentiated cells, KYL cells and MNi in mononucleated and in BN cells. The scores of basal, BN, CC and KHC cells indicate a strong disagreement both between laboratories and between scorers.

Consistent results of the individual scorers in the identification of cell types and nuclear anomalies, evaluated through the comparison of the results obtained by individual scorers in two separate scoring sessions, were observed, reflecting the sound knowledge of the crite-ria and the reproducible performance of the scorers. The inter- and intra-laboratory variability among experienced laboratories may be attributable to technical factors, such as the differences in micro-scopes that were used and their illumination set-up, the variations in the quality of lenses and also the differences in the interpretation of the scoring criteria between laboratories and between scorers.

Basal cells are identified on the basis of their size and nuclear to cytoplasm ratio. In some cases, the differential classification of basal cells and transitionally differentiated cells may be difficult due to the difficulty of assessing accurately the size of cells and the nuclear/cytoplasm area ratio visually. The characterisation of the basal cells through a tighter definition of the cytoplasmic size limits should allow the exclusion from the count of the larger tran-sitional cells and thus improve the stringency of basal cell identi-fication criteria.

Misclassification of two adjacent mononucleated cells as BN cells is possible when the cytoplasmic boundaries are not clearly evident either due to fading of the cytoplasmic stain or differences of the brightness of the microscope light source or poor focus and/or contrast provided by the microscope. Under such conditions subjective classifications may be increasingly difficult and may have lead to some of the inter-labora-tory differences. Our recommendation to reduce the misclassification of BN cells is to restrict scoring to clearly isolated cells or to carefully double-check that cells initially classified as binucleated cells are not in fact two overlapping cells that may be identified by carefully altering the focus to test continuity of their cytoplasmic boundaries.

The destructive fragmentation of the nucleus of a dying cell starts with the first step characterised by a striated condensed chromatin pattern (CC cells) that can be identified by parallel areas of differ-ent stain intensity in contrast to the more extensive and irregular nuclear chromatin aggregation in KHC cells. The discrimination between CC and KHC is difficult without meticulous examination of the chromatin condensation and fragmentation pattern, although it can be achieved with greater ease by using a higher power of mag-nification, as has been previously described (9). A possible solution

to this problem is to simply combine into a single category all the CC and KHC cells but this option should be clearly defined and stated if used as a modification of the standard protocol.

‘NBUDs’ in buccal mucosa cells, originally described by Tolbert et al. (18) as ‘broken eggs’, are small nuclear bodies attached to the nucleus ranging usually between 1/3 and 1/16 of the main nucleus. However, sometimes, NBUDs almost up to the size of the main nucleus may occur that appear to be indistinguishable from a binu-cleated cell in which the two nuclei are attached by a nucleoplasmic bridge. Due to the size and the atypical shape, the inclusion of large nuclear bodies in the NBUD category can be, in some cases, subjec-tive. It is recommended that more care be taken in scoring NBUDs in order to avoid confusion with BN cells.

Conclusions

Overall, the results of the present study show good agreement among the three laboratories in regard to the main parameters asso-ciated with the evaluation of chromosomal damage, such as MNi in mononucleated and BN DIFF cells.

Other end points associated with the differentiation process, such as the CC, KHC and nuclear alterations other than MNi, are poorly reproduced indicating that their evaluation using the current visual scoring protocol does not yield robust results for these parameters. Further improvements in scoring of these parameters are required before their use can be recommended without reservation. However, the results obtained with DSC samples, showing an altered cytome profile compared with the matched controls, suggest an added value of the evaluation of the different cellular biomarkers in identifying individuals at risk of developing these diseases.

The inconsistencies in the identification of basal cells, BN, CC and KHC cells compelled us to develop strategies to improve the guidelines for slide scoring by: (i) defining the nucleus:cytoplasm area ratio to characterise basal, transitional and differentiated cells, (ii) combining CC and KHC cells anomalies associated with cell death into a single CC/KHC and (iii) recommending special care in classifying NBUDs avoiding the analysis of overlapping cells and by checking that cells with large NBUDs are not misclassified as BN cells. An intercalibration exercise involving expert and non-expert labs will allow to further evaluate the feasibility for the application of the cytome approach in buccal cells.

The development of a robust and reproducible automated scor-ing system, in theory, could lead to improvements in the sensitiv-ity and accuracy of the BMCyt assay by overcoming many of the problems caused by subjective visual evaluation of slides. Important advances in the automated scoring of MNi in peripheral lympho-cytes and in cell cultures have been already obtained (19), but at present the automated detection of the different cell types, chromatin condensation/fragmentation status and nuclear anomalies in buccal mucosa cells is at an early stage of development with some notable progress achieved using laser scaning cytometry (20).

Funding

The work was in part supported by the Italian Association for Research on Cancer (AIRC) (Project IG 10458 2010).

AcknowledgementsThe authors should like to thank all participants of the study. Special thanks to Dr Bettina Baltacis from the hospital ‘Krankenanstalt Rudolfstiftung’ and


at Library M

edUni V

ienna (10076821) on March 23, 2015


Dow

nloaded from


the team of the cultural centre ‘Ich bin OK’ in Vienna for recruiting people with Down syndrome for our study. Furthermore, the authors also thank Mag. Alice Petschnig for her participation in the scoring exercise.Conflict of interest statement: None declared.

References 1. Holland, N., Bolognesi, C., Kirsch-Volders, M., Bonassi, S., Zeiger, E.,

Knasmueller, S. and Fenech, M. (2008) The micronucleus assay in human buccal cells as a tool for biomonitoring DNA damage: the HUMN pro-ject perspective on current status and knowledge gaps. Mutat. Res., 659, 93–108.

2. Thomas, P., Holland, N., Bolognesi, C., Kirsch-Volders, M., Bonassi, S., Zeiger, E., Knasmueller, S. and Fenech, M. (2009) Buccal micronucleus cytome assay. Nat. Protoc., 4, 825–837.

3. Thomas, P., Harvey, S., Gruner, T. and Fenech, M. (2008) The buccal cytome and micronucleus frequency is substantially altered in Down’s syndrome and normal ageing compared to young healthy controls. Mutat. Res., 638, 37–47.

4. Ferreira, F. L., Prá, D., Martino-Roth, M. G. and Garcias, G. L. (2009) Buccal micronucleus frequency is associated with age in Down syndrome. Genet. Mol. Res., 8, 1231–1237.

5. Thomas, P., Hecker, J., Faunt, J. and Fenech, M. (2007) Buccal micro-nucleus cytome biomarkers may be associated with Alzheimer’s disease. Mutagenesis, 22, 371–379.

6. Fenech, M., Bolognesi, C., Kirsch-Volders, M., Bonassi, S., Zeiger, E., Knasmüller, S. and Holland, N. (2007) Harmonisation of the micronucleus assay in human buccal cells–a Human Micronucleus (HUMN) project (www.humn.org) initiative commencing in 2007. Mutagenesis, 22, 3–4.

7. Bonassi, S., Biasotti, B., Kirsch-Volders, M., et al.; HUMNXL Project Con-sortium. (2009) State of the art survey of the buccal micronucleus assay–a first stage in the HUMN(XL) project initiative. Mutagenesis, 24, 295–302.

8. Bonassi, S., Coskun, E., Ceppi, M., et al. (2011) The HUman MicroNu-cleus project on eXfoLiated buccal cells (HUMN(XL)): the role of life-style, host factors, occupational exposures, health status, and assay proto-col. Mutat. Res., 728, 88–97.

9. Bolognesi, C., Knasmueller, S., Nersesyan, A., Thomas, P. and Fenech, M. (2013) The HUMNxl scoring criteria for different cell types and nuclear anomalies in the buccal micronucleus cytome assay - an update and expanded photogallery. Mutat. Res., 753, 100–113.

10. Hintzsche, H., Polat, B., Schewe, V., Djuzenova, C. S., Pfreundner, L., Flentje, M. and Stopper, H. (2012) Micronucleus formation kinetics in

buccal mucosa cells of head and neck cancer patients undergoing radio-therapy. Toxicol. Lett., 212, 33–37.

11. Nersesyan, A., Kundi, M., Atefie, K., Schulte-Hermann, R. and Knasmül-ler, S. (2006) Effect of staining procedures on the results of micronucleus assays with exfoliated oral mucosa cells. Cancer Epidemiol. Biomarkers Prev., 15, 1835–1840.

12. Bonassi, S., Fenech, M., Lando, C., et al. (2001) HUman MicroNucleus project: international database comparison for results with the cytokine-sis-block micronucleus assay in human lymphocytes: I. Effect of labora-tory protocol, scoring criteria, and host factors on the frequency of micro-nuclei. Environ. Mol. Mutagen., 37, 31–45.

13. Fenech, M., Bonassi, S., Turner, J., et al.; HUman MicroNucleus project. (2003) 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, 45–64.

14. Patino-Garcia, B., Hoegel, J., Varga, D., Hoehne, M., Michel, I., Jainta, S., Kreienberg, R., Maier, C. and Vogel, W. (2006) Scoring variability of micronuclei in binucleated human lymphocytes in a case-control study. Mutagenesis, 21, 191–197.

15. Moore, L. E., Warner, M. L., Smith, A. H., Kalman, D. and Smith, M. T. (1996) Use of the fluorescent micronucleus assay to detect the genotoxic effects of radiation and arsenic exposure in exfoliated human epithelial cells. Environ. Mol. Mutagen., 27, 176–184.

16. Cao, J., Liu, Y., Sun, H., Cheng, G., Pang, X. and Zhou, Z. (2002) Chro-mosomal aberrations, DNA strand breaks and gene mutations in naso-pharyngeal cancer patients undergoing radiation therapy. Mutat. Res., 504, 85–90.

17. Minicucci, E. M., Kowalski, L. P., Maia, M. A., Pereira, A., Ribeiro, L. R., de Camargo, J. L. and Salvadori, D. M. (2005) Cytogenetic damage in circulating lymphocytes and buccal mucosa cells of head-and-neck cancer patients undergoing radiotherapy. J. Radiat. Res., 46, 135–142.

18. Tolbert, P. E., Shy, C. M. and Allen, J. W. (1992) Micronuclei and other nuclear anomalies in buccal smears: methods development. Mutat. Res., 271, 69–77.

19. Fenech, M., Kirsch-Volders, M., Rossnerova, A., et al. (2013) HUMN project initiative and review of validation, quality control and prospects for further development of automated micronucleus assays using image cytometry systems. Int. J. Hyg. Environ. Health, 216, 541–552.

20. Leifert, W. R., François, M., Thomas, P., Luther, E., Holden, E. and Fenech, M. (2011) Automation of the buccal micronucleus cytome assay using laser scanning cytometry. Methods Cell Biol., 102, 321–339.


at Library M

edUni V

ienna (10076821) on March 23, 2015


Dow

nloaded from

http://www.humn.org


buccal micronucleus cytome assay: results of an intra- and inter-laboratory scoring comparison

Documents