Food and Chemical Toxicology 66 (2014) 107–112

Contents lists available at ScienceDirect

Food and Chemical Toxicology

journal homepage: www.elsevier .com/locate / foodchemtox

Sister chromatid exchange, (SCE), High-Frequency Cells (HFCs) and SCEdistribution patterns in peripheral blood lymphocytes of Spanish adultsmokers compared to non-smokers

http://dx.doi.org/10.1016/j.fct.2014.01.0110278-6915/Crown Copyright � 2014 Published by Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +34 963543056; fax: +34 963544954.E-mail address: jose.soriano@uv.es (J.M. Soriano).

Natividad Sebastià a, David Hervás b, Miguel Almonacid c, Juan Ignacio Villaescusa c,José Miguel Soriano a,⇑, Vicenta Sahuquillo c, Valentín Esteban d, Joan Francesc Barquinero e,Gumersindo Verdú f, José Cervera g, Esperanza Such g, Alegría Montoro c

a Department de Medicina Preventiva i Salut Pública, Facultat de Farmàcia, Universitat de Valencia, Burjassot, Spainb Unidad de Bioestadística IIS La Fe, Instituto de Investigación Sanitaria La Fe, Valencia, Spainc Servicio de Protección Radiológica, Hospital Universitario Politécnico La Fe, Valencia, Spaind Servicio de Salud Laboral, Dirección General de Salud Pública, Consellería de Sanitat, Generalitat Valenciana, Spaine Institut de Radioprotection et SûretéNucléaire (IRSN), PRP-HOM, SRBE, LDB.BP017, 92262 Fontenay-aux-Roses, Francef Departamento de Ingeniería Química y Nuclear, Univeristat Politécnica de Valencia, Camino de Vera s/n, Spaing Servicio de Hematología, Hospital Universitario y Politécnico La Fe, Valencia, Spain

a r t i c l e i n f o

Article history:Received 24 October 2013Accepted 6 January 2014Available online 18 January 2014

Keywords:TobaccoCytogenetic assessmentBiomarkerSCEs

a b s t r a c t

According to the International Agency for Research on Cancer, smoking tobacco is a major cause of cancerin humans. It causes about half of all male cancer deaths and an ever increasing number of cancer deathsin females. The aim of this study was to establish whether cigarette smoking increases sister chromatidexchanges (SCEs) in peripheral blood lymphocytes in two Spanish population groups; light and heavysmokers. The mean number of High-Frequency Cells (HFCs) was determined and, the SCE distributionpattern among the chromosomes was analysed represented by a ratio described below. A local sampleof 101 adult smokers (n = 48) and non-smokers (n = 53), aged from 18 to 49 years, was studied usingSCE levels in peripheral lymphocytes. Heavy smoking (P10 cigarettes per day) increased significantlythe SCE frequency and the HFC parameters. Neither age nor sex significantly influenced the frequenciesin the groups studied.

Crown Copyright � 2014 Published by Elsevier Ltd. All rights reserved.

1. Introduction

Tobacco smoking is a worldwide habit and one of the majorinternational health problems. Tobacco is a natural product thatconsists of more than 3800 constituents (4800 compounds in to-bacco smoke) ranging from small molecules such as hydrocarbons,terpenes, phenols, or nitriles, to macromolecules such as cellulose,lignin or nucleic acids. These compounds suffer transformationsdue to the high temperatures achieved during tobacco consump-tion and some become toxic and carcinogenic. Among thesecompounds more than 50 are known carcinogens, such as polycy-clic aromatic hydrocarbon, N-nitrosamines, aromatic amines, andtrace metals (Khabour et al., 2011). It has been stated that tobaccosmoking is the single largest cause of cancer worldwide (Secretanet al., 2009). Data indicate that 5–6 million deaths each year are

attributed to tobacco use and this annual toll may increase to 10million within the next 20–30 years (Khabour et al., 2011). Focus-ing on tobacco carcinogenicity, and tumour sites, for which the evi-dence of causality is sufficient, have been compiled in a specialreport of the International Agency for Research on Cancer (oral cav-ity, oropharynx, nasopharynx, and hypopharynx, oesophagus (ade-nocarcinoma and squamous-cellcarcinoma), stomach, colorectum,liver, pancreas, nasal cavity and paranasal sinuses, larynx, lung,uterine cervix, ovary (mucinous), bladder, kidney (body and pel-vis), ureter, bone marrow (myeloid leukaemia)) (Secretan et al.,2009).

In recent decades, the genotoxicity of cigarette smoking hasbeen assessed by several assays. Our study aimed to assess theimpact of cigarette smoking on the Spanish population, thereforewe follow the Prevention Technical Note 354 emitted by theNational Institute of Hygiene and Occupational Health from theSpanish Ministry of Labour and Social Affairs entitled ‘‘Biologicalcontrol of exposure to genotoxic agents: cytogenetic techniques’’.

108 N. Sebastià et al. / Food and Chemical Toxicology 66 (2014) 107–112

According to this Technical Note the exposure to tobacco smokingcan be evaluated using two different cytogenetic techniques, chro-mosomal aberrations and sister chromatid exchanges (SCEs). How-ever, it is suggested that due to the presence of certain chemicalagents, induction of SCEs may have a greater sensitivity to chromo-somal aberrations, and may result in genotoxic effects at concen-trations of up to 100 times lower than those required to producechromosomal aberrations (National Institute of Hygiene and Occu-pational Health, 1994). The SCE assay has been widely used to as-sess the genotoxic potential of mutagenic and carcinogenic agents,although it has to be taken into account that SCEs do not predictcancer as good as other biomarkers like chromosomal aberrations(Norppa et al., 2006). SCEs occur, in particular, when there is asymmetrical exchange of DNA segments between two sister chro-matids of a duplicating chromosome formed during the S phase ofthe cell cycle and can be induced by chemicals that are S phase-dependent DNA-damaging agents (Johnson et al., 2009).

In vitro toxicology studies of tobacco and tobacco smoke havebeen a useful tool to understand why tobacco consumption causescancer and a plethora of in vitro assay sare available to assess to-bacco smoke modes of action, mostly using non-human cell mod-els. The SCEs is an assay considered for this assessment; Johnsonet al. (2009) reviewed the literature available for these assaysand observed that it has been demonstrated that cigarette smokecondensates and total particulate matter is capable of inducingSCEs in in vitro studies in a concentration-dependent manner.The SCE biomarker is still being used as an indicator of tobaccoexposure (Khabour et al., 2011; Ben Salah et al., 2011). The analysisof SCEs is frequently complemented by the analysis of High-Fre-quency Cells (HFCs) which are defined as cells whose SCE fre-quency exceeds the 95th percentile of the SCE distribution in apooled data set from control individuals (Ponzanelli et al., 1997).This HFC measure is often more sensitive to the evaluation of theexposure to genotoxic agents, and so increases the sensitivity ofthe SCE assay when the effect of an exposure is not obviously de-tected by differences in mean SCE values (Ben Salah et al., 2011).

Some authors have postulated that HFCs could represent a sub-population of DNA repair-deficient lymphocytes. Others, however,have stated that HFCs may represent a subpopulation of longer livinglymphocytes which have an increased likelihood of accumulating alarger number of lesions during the G0 phase in vivo, thus showinga predisposition to SCEs. In the latter case, HFCs would be more prop-erly used for evaluation of exposure to genotoxic agents (Ponzanelliet al., 1997). We also analysed the SCE distribution patterns in meta-phases which allowed checking for the uniform or non-uniform dis-tribution of SCEs among the chromosomes in each metaphase.

The aim of this work is to assess the tobacco related genotoxiceffects in peripheral blood lymphocytes in the Spanish local popu-lation by using the SCE biomarker of exposure with an improvednew ratio which assists in describing the aggregation pattern ofSCEs among the chromosomes.

2. Subjects and methods

2.1. Participants

A sampling frame of one hundred and one adults (55 females and 46 males) aged18–49 years were sourced from an existing internal local panel supplied by the localhealth authorities. The control group (non-smokers) included 48 subjects and thesmoking group 53; all the smokers included in this study had smoked at least duringthe last five years. The survey was conducted from December 2003 to November2004 and approved by the Ethical Committee of Hospital La Fe (Valencia, Spain).

2.2. Culture conditions and stain technique

Human peripheral blood samples were collected in sterile vacutainer tubes(Becton, Dickinson and Company, Franklin Lakes, NJ, USA) containing lithium hep-arin as anticoagulant, after informed consent. For each treatment, separate cultures

were set up by mixing 0.75 mL of whole blood with 5 mL of PB-Max™ Karyotipingmedium (Gibco, Barcelona, Spain) and incubated 72 h at 37 �C. 150 lL of Colcemid�

(Gibco, Barcelona, Spain) from a stock solution of 10 lg mL�1 were added 2 h beforeharvesting to stop the cell culture in metaphase. Carnoy’s lymphocytes fixativesolution was prepared with methanol (Merck, Ramstadt, Germany) and acetic acid(Panreac, Barcelona, Spain). Cytogenetic analyses were carried out by using a con-ventional microscope (Izasa, Barcelona, Spain) and an image analysis system withthe IKAROS-software (MetaSystems). Three-day-old slides were stained with Fluo-rescence plus Giemsa stain technique. Old slides were treated 20 min in Hoechst33258 at room temperature. The slides were washed with distillated water. Theslides were covered with 2xSSC and treated with UV (300 W) for 2 min. Oncewashed with water and dried 30 min, the slides were stained with Leishman(Merck, Ramstadt, Germany) for 5 min this being useful for the identification offirst-, second- and third-division metaphases (Fig. 1).

2.3. Cytogenetic analysis

The incidence of SCE, HFC and SCE distribution patterns (Fig. 2) was determinedfrom the analysis of 50 s division metaphases for each individual. The evaluation ofSCE scores in lymphocytes included scoring total exchanges for the total number ofanalysed cells for each treatment to establish the SCE frequency (YSCE) in 46 chro-mosomes per cell.

2.4. Statistical methods

For each cell, the SCE frequency was calculated. We also computed the meannumber of HFCs per individual according to the method proposed by Carrano andMoore (1982): an HFC was defined as a cell which exceeds the 95th percentile ofthe SCE distribution from control individuals. As a measure of differences in SCEdistribution patterns among the 46 cell chromosomes, we estimated, for each cell,the number of SCEs per each affected chromosome. Using this estimate a ratio

can then be established, n2 of SCEn2 affected chromosomes, with a higher ratio considered as a more

clustered SCE distribution pattern which means that, instead of having one SCEper chromosome, a higher number of SCEs is present in some of the affectedchromosomes and some of these chromosomes will be, therefore, more damagedthan others. Since the estimated ratio for a random distribution of SCEs in a cellgrows as a function of the number of SCEs (Fig. 3), we subtracted the expectedratio from our observed ratio as a measure of deviation from a random distribu-tion pattern.

According to Laugesen and Swinburn (2000), subjects were categorized intothree categories: light smokers (who smoke less than 10 cigarettes per day), heavysmokers (defined as those who smoke 10 or more cigarettes per day) and non-smokers. SCE frequencies and ratio distribution patterns were modelled in termsof age, sex and smoking status using a linear mixed model approach. These modelsregarded individuals as a random factor in order to correct for intra-individual cor-relation among cells. Counts of HFCs per individual were modelled with the sameexplicative variables using a negative binomial distribution. The negative binomialdistribution was preferred over the Poisson due to the presence of over dispersionin the data. All interactions between variables were assessed (age-sex, age-smoking,sex-smoking). After fitting of the saturated models with all the variables and itsinteractions, a model selection procedure was performed using second-orderAkaike Information Criterion (AICc) to find the best model. All statistical analyseswere carried out using the R software (version 2.15.2).

3. Results

3.1. SCE analysis

The estimate of SCE frequency for the population mean was8.19 and a 95% CI from 7.88 to 8.5. The best model, according tothe model selection procedure, included only the smoking variable,rejecting the effect of age and sex (Table 1). The effect of smoking,with a p-value <0.001, was statistically highly significant. Both,heavy smoking and light smoking status were associated with asignificant increase in the SCE frequency (Fig. 4). According tothe model estimates for the coefficients, the mean frequency ofSCE in non-smoker population was estimated to be 7.46 (95% CI[7.12; 7.81]). The light smoking group has a SCE estimate 1.21(95% CI [0.6; 1.88]) units higher than the non-smoker population(mean SCE for light smokers, 8.67). Finally the heavy smokinggroup had an SCE estimate 2.02 (95% CI [1.27; 2.78]) units higherthan the non-smoker population (mean SCE for heavy smokers,9.48).

Fig. 1. Image illustrating 3 consecutive divisions of a lymphocyte after Fluorescence plus Giemsa staining. In the image it can be observed (a) lymphocytes in first division, (b)lymphocytes in second division (SCE counting) and (c) lymphocytes in third division.

Fig. 2. Metaphase in second division showing chromosomes containing differentnumber of SCEs: (a) 1 SCE, (b) 2 SCE, (c) 3 SCE. The ratio is calculated according tothis formula: ratio = a�1 + b�2 + c�3 (total SCEs)/a + b + c (affected chromosomes).For this metaphase: ratio = 29/20 = 1.45. Adjusted ratio = 1.45 – expected ratio for29 SCE (1.35) = 0.1.

Fig. 3. Estimated ratio for a random distribution of SCEs in a cell.

Table 1Selected linear mixed model for SCE frequency by AICC criterion.

Estimate Std. error t-Value p-Value

Intercept 7.46 0.21 35.82 <0.001Light smoking 1.21 0.34 3.52 <0.001Heavy smoking 2.02 0.41 4.92 <0.001

Fig. 4. Comparison of SCE frequencies distribution between non-smokers, light andheavy smoker cells and towards the population mean. Distribution of the meanfrequency of SCE in the population.

N. Sebastià et al. / Food and Chemical Toxicology 66 (2014) 107–112 109

3.2. HFC analysis

The results of our second analysis, regarding the count of HFCsper individual, were similar. The distribution of SCEs had a 95thpercentile of 16 (Fig. 5). Mean number of HFCs/individual was 3with a 95% CI [2.35; 3.85]. The best model, according to the modelselection procedure, included only the smoking variable, rejectingthe effect of age and sex (Table 2).

The effect of smoking, with a p-value <0.001, was statisticallyhighly significant. Both, the heavy smoking and light smoking sta-tuses were associated with a significant increase in HFC count perindividual. According to the model estimates for the coefficients,

Fig. 5. Distribution of SCEs with a 95th percentile of 16.

Table 2Selected negative binomial model for HFC count by AICC criterion.

Estimate Std. error t-Value p-Value

Intercept 0.33 0.17 1.92 0.055Light smoking 1.08 0.25 4.30 <0.001Heavy smoking 1.44 0.29 5.04 <0.001

Table 3Selected linear mixed model for ratio by AICC criterion.

Estimate Std. error z-Value p-Value

Intercept 0.011 0.003 3.17 0.0015Light smoking 0.011 0.006 2.04 0.042Heavy smoking 0.015 0.007 2.26 0.024

110 N. Sebastià et al. / Food and Chemical Toxicology 66 (2014) 107–112

the mean HFC count per individual in the non-smoking group was1.38 (95% CI [0.99; 1.93]), in the light smoking group it was 4.07(95% CI [2.5; 6.69]) and in the heavy smoking group the meanwas 5.83 (95% CI [3.38; 10.36]) (Fig. 6).

3.3. SCE distribution pattern

Results of our analysis regarding the SCE distribution patternagree with the SCE frequency and mean HFC number for that var-iable, which statistically influences this parameter. This means thatagain, the best model included, only the variable ‘‘smoking status’’(p-value = 0.029) (Table 3) rejecting the effect of age and sex. Both,heavy and light smoking status, were associated with a significantincrease (p < 0.05) in the ratio representing the SCE distributionpattern, indicating a more clustered damage.

4. Discussion

It is known that SCEs are one of the most extensively used bio-markers to assess the genotoxic potential of mutagenic and carcin-ogenic agents because its expression reflects possible alterationsduring the cell cycle or genetic damaging events at the chromo-somal level (Mudry et al., 2011). Very few articles (Carbonellet al., 1995) have assessed the tobacco-related SCE and HFC in-crease in the Spanish population. This is the reason we aimed toimprove this analysis with the new SCE distribution patternparameter described.

Our results concerning the influence of age and sex on the SCEfrequency match with some groups (Husum et al., 1982; Hedneret al., 1983; Carbonell et al., 1995) who studied a group of bothsmokers and non-smokers and observed that, male/female differ-ences did not significantly influence the frequency of SCEs. How-ever, totally different results have also been found; Andersonet al. (1986) observed that females had significantly higher SCE fre-quencies than males, and cigarette smoking significantly increasedsuch values after allowing for sex. Lazutka et al. (1994) reportedsimilar results, indicating that SCEs increased with age and ciga-rette smoking intensity, and higher SCE frequencies were observedin females. Bearing in mind the contradictory results showing the

Fig. 6. Comparison of HFCs count per individual between non-smokers, light andheavy smokers and towards the population mean.

influence of age and gender on the SCE frequency, it is necessarytherefore to consider these two variables in the statistical models.

After discarding the sex and age variables, our statistical modelonly considered the tobacco smoking variable (non-smokers, lightsmokers, and heavy smokers). Both smoking classifications, heavyand light smokers were associated with a significant increase(p < 0.001) in SCE frequencies compared to the non-smoking groupwhich means that even the light smokers had an SCE frequencysignificantly higher than the non-smokers.

Since the 1970s, a number of studies have attributed the effectsof the tobacco habit on the increase in the individual SCE frequen-cies. In 1978 Lambert et al. reported that, among 43 subjects, thosewho were cigarette smokers had significantly higher SCE frequen-cies than non-smokers. They observed a stepwise increase of about15% in the average SCE frequency among moderate (<10 cigarettesper day) and heavy smokers (P10 cigarettes per day) and they as-cribed their results to benzo(a)pyrene (BP) which, as they stated,increased the frequency of SCEs almost two-fold in human lym-phocytes in vitro. BP is classified as carcinogen for humans (Group1) (International Agency for Research on Cancer, 2012) and in factit has been demonstrated that BP can induce SCEs in human lym-phocytes (Hatzi et al., 2011). Wulf et al. (1983) studied a group ofcigarette smokers, cheroot or pipe smokers. The cigarette groupconsisted of smokers of high-tar cigarettes with filter, high-tar cig-arettes without filter and low-tar cigarettes with filter, all of themwith the same cigarette consumption. The SCE mean values in 3categories of cigarette smokers were statistically higher than themean value observed in the non-smoking group. Moreover, theyalso detected that even the mean value of the SCEs for pipe andcheroot smokers was always higher than the non-smoking group.Smokers of all types of tobacco had an increased SCE frequency.The Spanish work carried out by the group of Carbonell et al.,1995 observed that the cigarette smoking factor increased theSCE rates as well as the HFC values. The group of Kao-Shan et al.(1987) examined the SCEs in peripheral blood and bone marrowin 18 smokers with an average cigarette use corresponding to 6pack years. They found a significant increase in the mean numberof SCEs in both bone marrow cells (p < 0.001) and peripheral bloodlymphocytes (p < 0.005) of the smoker group.

An increase in the frequency of SCEs in the peripheral lympho-cytes of smoking volunteers (6.5 ± 0.3) compared to non-smokers(4.1 ± 0.2) was observed by Sardas� et al. (1991) where both theduration of smoking and the number of cigarettes smoked perday appeared to influence SCE frequencies since those who smokedmore than 10 cigarettes per day and those who had habituallysmoked for over 10 years had a higher SCE frequency. An interest-ing study was performed by Milic et al. (2008) which aimed toassess the genotoxic hazard for workers in the cigarette manufac-turing industry. Since it had been reported that tobacco workersproducing cigarettes are exposed to a wide range of chemicalcompounds that have been shown to be genotoxic or carcinogenic,such as nicotine, nitrosamines, formaldehyde, acetaldehyde, cro-tonaldehyde, hydrazine, arsenic, nickel, cadmium, and benzopy-rene. They divided the groups into those exposed and those notexposed to genotoxic agents and both groups were then dividedinto smokers and non-smokers. As expected, smoking subjects, ingeneral, had a higher frequency of SCEs than non-smokers in boththe control and exposed groups. Furthermore, exposed smokers

N. Sebastià et al. / Food and Chemical Toxicology 66 (2014) 107–112 111

showed a significantly higher frequency of SCEs than the controlsmokers. Finally, their data showed that exposure to tobacco dustcaused a significant increase in SCEs.

Recently, Ben Salah et al. (2011) carried out a study in Tunisia assmoking continues to increase in low- and middle income coun-tries. The authors mentioned that in their country, an epidemiolog-ical study was conducted and data published in the Tunisian cancerregistry which showed an alarming 54% increase in lung cancerincidence over the last 10 years. They indicated that the mean fre-quency of SCEs per cell was significantly higher in smokers than innon-smokers. They divided the smoker group into heavy smokers(smoking >10 years) and light smokers (smoking 610 years). Re-sults from both studies suggest that the mean frequency of SCEsmight increase not only because of the number of cigarettessmoked but also by the length of time an individual has smoked.Another recent study of Khabour et al. (2011) also evaluated thegenotoxicity of cigarette smoking in volunteers from Jordan usingthe SCE assay. Our results agreed with theirs since SCE frequenciesin the cigarette smoking group significantly increased comparedwith those of non-smokers. Furthermore, they compared SCE val-ues of cigarette and water pipe smokers and found that SCE fre-quencies in the water pipe smoking group were higher than inthe control group and even higher than in the cigarette smokinggroup. In a similar study of Alsatari et al. (2012) they investigatedthe genotoxicity of water pipe smoking in the lymphocytes ofwater pipe smokers using another cytogenetic biomarker, thechromosomal aberrations (CAs) assay. They observed that cigarettesmoking, like water pipe smoking significantly increased the fre-quencies of CAs (p < 0.01) and even that, the frequency of CAs in-creased with the greater the water pipe use, confirming thegenotoxicity of both cigarette and water pipe smoking. It seemsclear that smoking increases the SCE frequencies but our aimwas to test this result in the Spanish population. We performed astatistical analysis based on models which consider the variablestogether and not by comparing them two by two, similar to themajority of tests.

In an attempt to carry out a more accurate analysis, we calcu-lated the mean number of HFCs since it represents a more sensi-tive criterion for assessment of exposure to chemicals particularlythose due to smoking, rather than only the mean frequency ofSCEs (Ben Salah et al., 2011). According to the analysis of HFCs,we found that a heavy and light smoking status was associatedwith a significant (p < 0.001) increase in the mean number ofHFCs per individual. Some authors have postulated that HFCscould represent a subpopulation of DNA repair deficient lympho-cytes. Ponzanelli et al. (1997) interpreted SCEs as a signal of DNAdamage and that SCE frequency in HFCs decreased during re-peated cell cycles as a consequence of DNA damage removal. Itcan be inferred that HFCs, although initially more damaged, suc-ceeded in removing most SCEs inducing lesions over three cell cy-cles, especially after the first cycle. We did not check the SCEfrequency in the third cycle but only the second and our resultswere similar to theirs. The larger number of SCEs that we ob-served in HFCs could be attributable to a higher level of initialdamage, as they claimed, and not to repair deficiencies. Thishypothesis agrees with other authors who stated that HFCs mayrepresent a subpopulation of longer living lymphocytes whichhave an increased likelihood of accumulating a larger numberof lesions during the G0 phase in vivo, thus showing a predispo-sition to SCEs (Ponzanelli et al., 1997). In cases of repair-capacitydeficiency, HFCs should be more properly used for the evaluationof individual DNA repair capacity, whereas in the case of longerliving lymphocytes, it would be more correctly used for the eval-uation of exposure to genotoxins. We consider, therefore, that thesecond hypothesis is more appropriate for explaining the increasein the mean HFC number in our smoking group.

From the analysis of the SCE distribution pattern, both the hea-vy and light smoking statuses were associated with a significant in-crease in the SCE/chromosome ratio described in this study whichimplies a more clustered pattern in SCE distribution. This meansthat SCEs are not uniformly distributed among the 46 chromo-somes in the cell but it seems there exist a tendency to accumulatethe SCEs in fewer chromosomes that would correspond with a uni-form distribution for all of them. The mechanisms by which somecompounds can induce SCEs are different; some polyphenols dam-age the DNA and may produce SCEs by arresting the S phasethrough cleaving to the DNA (Matsuoka et al., 2001), others, suchas antitumor antibiotics (i.e., mytomycin), damage DNA in thepresence of a redox-active metal ion such as iron (Fe2+) or Cooper(Cu2+), as well as molecular oxygen (O2) (Hecht, 2000).

Considering this last mechanism, it may be possible that amongthe wide variety of tobacco compounds some of them can act likeantitumor antibiotics so they would suffer activation by a redoxcomplex with metal ions from DNA. These active products, usuallywith a short life, would, therefore, produce a clustering damage inthe same chromosome. However, it has been described that somemolecules such as nitroimidazole derivates did not affect all thechromosomes equally, supporting the idea of the ‘‘specific genomictargets’’. Analyzing SCE frequencies by chromosome, these authorsobserved that certain chromosomes showed higher susceptibilityto such treatments and moreover, they specified which region in-side the band was more affected emphasizing the importance ofthe so-called ‘‘hot spots’’ and the validity of SCE biomarkers ascytogenetic indicators of genomic fragility (Carballo et al., 2009;Mudry et al., 2011). Those chromosomes with a higher ratio of SCEsare not known and so we cannot specify whether SCEs were local-ized in a region with hot spots, but it could be hypothesised thatsome compounds from tobacco could affect the chromosomes inthe same way as some other molecules such as nitroimidazole.

In the Spanish study of Marcilla et al., 2012, the authors carriedout an extensive analysis using 10 tobacco brands to determine themain compounds found in mainstream tobacco smoke. The analy-sis of the vapour fraction showed that major compounds were car-bon monoxyde (CO) and carbon dyoxide (CO2). In addition, thesmall aldehyde molecules present in tobacco smoke are particu-larly harmful, especially acetaldehyde. Another important com-pound found in the study was 1,3-butadiene; this compoundpresents the highest cancer risk index of all the constituents of cig-arettes smoke and among which metabolite 1,2:3,4-diepoxybutanewas shown to induce SCEs in vitro in human lymphocytes (Kliger-man and Hu, 2007). Furthermore, in the Spanish study, 85 compo-nents were identified in the particulate matter such as nicotine,polycyclic aromatic compounds or nitrosamines which accordingto the International Agency for Research in Cancer (2004) areknown carcinogens. In vitro studies are correlated with thosein vivo studies mentioned in this paper with our results.

According to our research, it can be stated that the frequenciesof SCEs and HFCs from the cigarette smoking group significantly in-creased compared with those of non-smokers. Moreover, with ournew developed parameter, we can observe an SCE distribution pat-tern among the cells so that SCEs concentrate in some chromo-somes instead of distributing homogeneously which could beinterpreted as some kind of genomic susceptibility to tobacco com-pounds. The findings are in agreement with other authors whenusing the SCE assay to confirm smoking induced chromosomedamage and to justify the need for promoting campaigns and topressure for more legislation to reduce cigarette consumption.

Conflict of Interest

The authors declare that there are no conflicts of interest.

112 N. Sebastià et al. / Food and Chemical Toxicology 66 (2014) 107–112

Transparency Document

The Transparency document associated with this article can befound in the online version.

References

Alsatari, E.S., Azab, M., Khabour, O.F., Alzoubi, K.H., Sadiq, M.F., 2012. Assessment ofDNA damage using chromosomal aberrations assay in lymphocytes ofwaterpipe smokers. Int. J. Occup. Med. Environ. Health 25, 218–224.

Anderson, D., Jenkinson, P.C., Dewdney, R.S., Francis, A.J., Godbert, P., Butterworth,K.R., 1986. Chromosome aberrations, mitogen-induced blastogenesis andproliferative rate index in peripheral lymphocytes from 106 controlindividuals of the U.K. population. Mutat. Res. 204, 407–420.

Ben Salah, G., Kamoun, H., Rebai, A., Ben Youssef, A., Ayadi, H., Belghith-Mahfoudh,N., Fourati, A., Ayadi, H., Fakhfakh, F., 2011. Sister chromatid exchange (SCE) andhigh frequency cells (HFCs) in peripheral blood lymphocytes of healthyTunisian smokers. Mutat. Res. 719, 1–6.

Carbonell, E., Peris, F., Xamena, N., Creus, A., Marcos, R., 1995. SCE analysis in humanlymphocytes of a Spanish control population. Mutat Res. 335, 35–46.

Carballo, A.M., Martinez, R.A., Mudry, M.D., 2009. Nitroimidazole derivatives: nonrandomness sister chromatid exchanges in human peripheral bloodlymphocytes. J. Appl. Toxicol. 29, 248–254.

Carrano, A.V., Moore, D.H., 1982. The rationale methodology for quantifying sisterchromatid exchanges in humans. In: Heddie, J.A. (Ed.), New Horizons in GeneticToxicology. Academic Press, New York, pp. 267–304.

Hatzi, V.I., Terzoudi, G.I., Stavropoulou, C., Malik, S.I., Makropoulos, V., Pantelias,G.E., 2011. Lack of association between GSTT1 polymorphism and endogenousor benzo[a]pyrene induced sister chromatid exchanges as analyzed inmetaphase or G2-phase lymphocytes. Mol. Biol. Rep. 38, 3959–3966.

Hecht, S.M., 2000. Bleomycin: new perspectives on the mechanism of action. J. Nat.Prod. 63, 158–168.

Hedner, K., Högstedt, B., Kolnig, A.M., Mark-Vendel, E., Strömbeck, B., Mitelman, F.,1983. Sister chromatid exchanges and structural chromosome aberrations inrelation to smoking in 91 individuals. Hereditas 98, 77–81.

Husum, B., Wulf, H.C., Niebuhr, E., 1982. Increased sister chromatid exchangefrequency in lymphocytes in healthy cigarette smokers. Hereditas 96, 85–88.

International Agency for Research on Cancer (IARC), 2004. Tobacco smoke andinvoluntary smoking. IARC Monographs, vol. 83.

International Agency for Research on Cancer (IARC), 2012. Chemical agents andrelated occupations. A review of human carcinogens. IARC Monographs, vol.100.

Johnson, M.D., Schilz, J., Djordjevic, M.V., Rice, J.R., Shields, P.G., 2009. Evaluation ofin vitro assays for assessing the toxicity of cigarette smoke and smokelesstobacco. Cancer Epidemiol. Biomarkers Prev. 18, 3263–3304.

Kao-Shan, C.S., Fine, R.L., Whang-Peng, J., Lee, E.C., Chabner, B.A., 1987. Increasedfragile sites and sister chromatid exchanges in bone marrow and peripheralblood of young cigarette smokers. Cancer Res. 47, 6278–6282.

Khabour, O.F., Alsatari, E.S., Azab, M., Alzoubi, K.H., Sadiq, M.F., 2011. Assessment ofgenotoxicity of water pipe and cigarette smoking in lymphocytes using thesister chromatid exchange assay: a comparative study. Environ. Mol. Mutagen.52, 224–228.

Kligerman, A.D., Hu, Y., 2007. Some insights into the mode of action of butadiene byexamining the genotoxicity of its metabolites. Chem. Biol. Interact. 166, 132–139.

Lambert, B., Lindblad, A., Nordenskjöld, M., Werelius, B., 1978. Increased frequencyof sister chromatid exchanges in cigarette smokers. Hereditas 88, 147–149.

Laugesen, M., Swinburn, B., 2000. New Zealand’s tobacco control programme 1985–1998. Tob. Control. 9, 155–162.

Lazutka, J., Dedonyte, V., Krapavickaite, D., 1994. Sister-chromatid exchanges andtheir distribution in human lymphocytes in relation to age, sex and smoking.Mutat. Res. 306, 173–180.

Marcilla, A., Martínez, I., Berenguer, D., Gómez-Siurana, A., Beltrán, M.I., 2012.Comparative study of the main characteristics and composition of themainstream smoke of ten cigarette brands sold in Spain. Food Chem. Toxicol.50, 1317–1333.

Matsuoka, A., Furuta, A., Ozaki, M., Fukuhara, K., Miyata, N., 2001. Resveratrol, anaturally occurring polyphenol, induces sister chromatid exchanges in aChinese hamster lung (CHL) cell line. Mutat. Res. 494, 107–113.

Milic, M., Kasuba, V., Orescanin, V., Zeljezic, D., Kopjar, N., Rozgaj, R., 2008.Chromosome damage in workers in cigarette manufacturing industry. J. Appl.Toxicol. 28, 399–404.

Mudry, M.D., Martinez, R.A., Nieves, M., Carballo, M.A., 2011. Biomarkers ofgenotoxicity and genomic instability in a non-human primate, Cebuslibidinosus (Cebidae, Platyrrhini), exposed to nitroimidazole derivatives.Mutat. Res. 721, 108–113.

Norppa, H., Bonassi, S., Hansteen, I.L., Hagmar, L., Stromberg, U., Rossner, P., Boffetta,P., Lindholm, C., Gundy, S., Lazutka, J., Cebulska-Wasilewska, A., Fabiánová, E.,Srám, R.J., Knudsen, L.E., Barale, R., Fucic, A., 2006. Chromosomal aberrationsand SCEs as biomarkers of cancer risk. Mutat Res. 600, 37–45.

National Institute of Hygiene and Occupational Health, 1994. Biological control ofexposure to genotoxic agents: cytogenetic techniques. Spanish Ministry ofLabour and Social Affairs. <http://www.insht.es/InshtWeb/Contenidos/Documentacion/FichasTecnicas/NTP/Ficheros/301a400/ntp_354.pdf>.

Ponzanelli, I., Landi, S., Bernacchi, F., Barale, R., 1997. The nature of high frequencysister chromatid exchange cells (HFCs). Mutagenesis 12, 329–333.

Sardas�, S., Gök, S., Karakaya, A.E., 1991. Increased frequency of sister chromatidexchanges in the peripheral lymphocytes of cigarette smokers. Toxicol. In Vitro5, 263–265.

Secretan, B., Straif, K., Baan, R., Grosse, Y., El Ghissassi, F., Bouvard, V., Benbrahim-Tallaa, L., Guha, N., Freeman, C., Galichet, L., Cogliano, V., 2009. WHOInternational Agency for Research on Cancer Monograph Working Group. Areview of human carcinogens- -Part E: tobacco, areca nut, alcohol, coal smoke,and salted fish. Lancet Oncol. 10, 1033–1034.

Wulf, H.C., Husum, B., Niebuhr, E., 1983. Sister chromatid exchanges in smokers ofhigh-tar cigarettes, low-tar cigarettes, cheroots and pipe tobacco. Hereditas 98,225–228.