Review

The Comet assay for the evaluation of genotoxic impact in aquatic environments

G. Frenzilli a,*, M. Nigro a, B.P. Lyons b

a Department of Human Morphology and Applied Biology, University of Pisa, Pisa, Italyb Cefas Weymouth Laboratory, Barrack Road, The Nothe, Weymouth, Dorset DT4 8UB, UK

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

2. Marine environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

2.1. Invertebrates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

2.2. Vertebrates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

3. Freshwater environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

3.1. Invertebrates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

3.2. Vertebrates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

4. General remarks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

4.1. Protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

4.2. Apoptotic cells detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

4.3. Cell type/target organs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

4.4. Comet assay and other genotoxicity tests: a comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

4.5. Indigenous versus transplanted species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

Mutation Research 681 (2009) 80–92

A R T I C L E I N F O

Article history:

Received 31 August 2007

Received in revised form 29 February 2008

Accepted 3 March 2008

Available online 13 March 2008

Keywords:

Environmental impact

Biomonitoring

Comet assay

Aquatic organisms

DNA damage

A B S T R A C T

This review considers the potential of the Comet assay (or Single Cell Gel Electrophoresis, SCGE) to

evaluate the environmental impact of genotoxins in aquatic environments. It focuses on in vivo and in situ

studies that have been carried out in various marine and freshwater sentinel species, published in the last

5 years. A large number of the studies reviewed report that the Comet assay is more sensitive when

compared with other biomarkers commonly used in genetic ecotoxicology, such as sister chromatid

exchanges or micronucleus test. Due to its high sensitivity, the Comet assay is widely influenced by

laboratory procedures suggesting that standard protocols are required for both fish and mussel cells.

However, there are still a wide variety of personalised Comet procedures evident in the literature

reviewed, making comparison between published results often very difficult. Standardization and inter-

laboratory calibration of the Comet assay as applied to aquatic species will be required if the Comet assay

is to be used routinely by national bodies charged with monitoring water quality.

� 2008 Elsevier B.V. All rights reserved.

Contents l is ts ava i lab le at ScienceDirec t

Mutation Research/Reviews in Mutation Research

journal homepage: www.e lsev ier .com/ locate / rev iewsmrCommuni ty address : www.e lsevier .com/ locate /mutres

* Cor

Applica

fax: +3

E-m

1383-5

doi:10.1

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

1. Introduction

The demand for a clean and safe supply of water for drinking,agriculture and recreation has rapidly increased over the last fewdecades. Receiving waters, such as lakes, rivers and marine coastal

responding author at: Dipartimento di Morfologia Umana e Biologia

ta, via Volta n.4, 56126 Pisa, Italy. Tel.: +39 050 2219111;

9 050 2219101.

ail address: giada@biomed.unipi.it (G. Frenzilli).

742/$ – see front matter � 2008 Elsevier B.V. All rights reserved.

016/j.mrrev.2008.03.001

areas are the receptacles for huge amounts of wastes derived directlyfrom industry, agriculture and urban settlements or indirectly fromthe atmospheric deposition of airborne emissions. Present amongstthese waters are a complex environmental mixture of well-knowntoxicants along with an increasing number of emerging contami-nants, which pose a threat to both aquatic ecosystems and the healthand welfare of human populations [1]. It is known that a number ofchemicals present are highly persistent and have mutagenic and/orclastogenic properties [2,3]. The relevance of detecting themutagenic/genotoxic risks associated with water pollution wasfirstly perceived in the late 1970s, when methods based on

G. Frenzilli et al. / Mutation Research 681 (2009) 80–92 81

Salmonella bioassay [4] or sentinel species, such as mussels [5] andfish [6,7] were set up for monitoring the presence of mutagens andgenotoxicants in aquatic environments. Since that time several testshave been developed for evaluating DNA alterations in aquaticanimals, these are based on potentially pre-mutagenic lesions suchas, DNA adducts, base modifications, DNA–DNA and DNA–proteinscross-linking and DNA strand breaks [8].

The analysis of DNA alterations in aquatic organisms has beenshown to be a highly suitable method for evaluating the

Table 1Assessment of DNA damage by Comet assays after in vivo exposure of aquatic animals

Organism Cell type Agent Exposure time

Invertebrates

D. polymorpha Haemocytes Lake water (Italy) 3 h; 20 days

D. polymorpha Haemocytes Lake water (Italy) 20 days

L. fortunei Haemocytes Sediment samples

(urban sites, Brazil)

7 days

D. polymorpha Haemocytes 3 strains MC

toxins (Microcystis

aeruginosa)

7–14–21 days

U. tumidus Haemocytes,

gill cells,

digestive

gland cells

B(a)P 6 days

Fe3+

U. tumidus Digestive

gland cells

Polyphenols 24 h

48 h

L. fortunei Haemocytes PCP 2 h + repair

CuSO4

P. felina Whole animals Norflurazon

(herbicide)

7 days

G. schubarti Tail Diluvio’s Basin

(Brazil)

13 days

C. gigas Embryos B(a)P 16 h

EE2

ES

S. sachalinensis Digestive

gland cells

MNNG 2 days

Haemocytes B(a)P

M. edulis Haemocytes Tritiated water, HTO 96 h

P. viridis Haemocytes Water-borne B(a)P 3–12 days

P. viridis Male gonad cells Extracts of

cigar tobacco

2–16 days

M. edulis Haemocytes TBT 7 days

M. edulis Gill cells Cd 10 days

Cr 7 days

Cr VI Injection

M. edulis Haemocytes Styrene 7 days

T. semidecussatus Haemocytes Estuarine

sediments

7–21 days

Gill cells

Digestive cells

S. droebachiensis Coelomocytes Dispersed

crude oil

4–5 weeks

M. edulis Haemocytes

(+) Positive response, (�) negative response, (D–R) dose–response.

genotoxic contamination of environments, being able to detectexposure to low concentrations of contaminants in a wide rangeof species. In general, these methods have the advantage ofdetecting and quantifying the genotoxic impact without requir-ing a detailed knowledge of the identity and the physical/chemical properties of the contaminants present. Tests directlyassessing DNA strand breaks, or downstream alterationsfollowing DNA strand damage, are commonly used to assessgenotoxic impact in aquatic animals. The early procedures for

to genotoxicants.

Concentration range Parameter Response Reference

+ or � disinfectants

(NaCl, PAA, ClO2)

LDR (migration

length/head diameter)

PAA � NaCl

and ClO2 +

(reduction)

[48]

Different seasons

(Autumn, Winter,

Summer)

LDR, TL + [49]

100 � 5 g of

sediment sample

DI (damage index);

DF (damage frequency)

+ [50]

104 cells/ml

freshwater for

each strain

%Tail DNA + [52]

50–100 mg/l %Tailed DNA cells + [44]

20–40 mg/l

60–500 mM TM + [45]

10–150 mg/l IL (image

length, mm); DI

D–R [51]

3.75–20 mg/ml

0.2–2 mM TL + [54]

TM

%Tail DNA

Urban waste water

(products of

automobile

fumes, human

urban activities)

Damage index (DI) + [55]

0.2 nM to 2 mM %Tail DNA + [20]

0.02–1.70 nM OTM �+

0.01–1 ppm(mg/l) TL + [63]

0.1–1 ppm TM

12–485 mGy/hr TM D–R [21]

0.3–30 mg/l TL, OTM, %Tail DNA D–R [17]

2.5–15 mg/ml %Tail DNA + [18]

0.1–5 mg/l TBTO %Tail DNA + [19]

10–200 mg/l %Tail DNA + [29]

10–200 mg/l +

10.4 mg/animal +

2 mg/l %Tail DNA + [36]

1 kg sediment

added to 2 l

of seawater

%Tail DNA + [37]

0.06–0.25 mg/l %Tail DNA + [24]

0.15–0.25 mg/l

G. Frenzilli et al. / Mutation Research 681 (2009) 80–9282

measuring DNA strand breaks were based on the separation ofdouble-stranded DNA, assessed by centrifugation or filtration; oron the denaturation rate, under alkaline conditions, determinedby the incorporation of a fluorescent dye by the double-strandedDNA. Alternative procedures based on cytogenetic investiga-tions, such as the sister chromatid exchanges (SCE) assay and themicronucleus (MN) test have also been widely used. The SCEassay is capable of detecting the exchange of chromosomalfragments between sister chromatids which follow DNA strandbreakage [9]. The micronucleus test is based on the assessmentof chromosomal fragments or whole chromosomes, which arenot incorporated into daughter nuclei after cell division. Such

Table 2Assessment of DNA damage by Comet assays after in vivo exposure of aquatic animals

Organism Cell type Agent Exposure tim

Vertebrates

C. carpio Erythrocytes Lake water (Italy) 20 days

C. auratus Erythrocytes PBTA-6 (dye) 3–9 h

ADDB (dye) 3 h

D. rerio Freshly fertilized eggs Sediments from

Laguna lake

(the Philippines)

96 h

C. auratus Erythrocytes Leachate from l

andfill sites (Japan)

9 days

C. carpio Erythrocytes Disinfectants: 20 days

NaClO

ClO2

CH3COO2H

C. batrachus Erythrocytes Herbicides:

2,4-D 48–96 h

Butachlor

S. mellops Erythrocytes Styrene 7 days

C. carpio Gill and liver cells MNNG 7 days

O. mykiss Gill and liver cells B(a)P 2 days

P. promelas Haepatocytes Biosolids

(PAHs, PBDEs, NPs)

3–28 days

C. punctatus Gill and kidney

tissues

Endosulfan (pesticide) 24–96 h

U. pygmaea L. Gill cells Rhine ground water

(The Netherlands)

3–11 days

A. anguilla Erythrocytes B(a)P

b-BF

Arochlor 1254

TCDD

O. mykiss Erythrocytes Algal extracts

(P. fucoides)

7 days

C. auratus Erythrocytes Glyphosate 48–144 h

P. lineatus Erythrocytes Diesel water soluble

fraction (DWSF)

6–96 h; 15 da

O. mykiss Erythrocytes Sediments from

Biobio River (Chile)

21 days

S. maximus L. Liver cells Cork Harbour

sediments (Ireland)

3 weeks

X. Laevis Larvae erythrocytes Leachates and bottom

ash from

municipal solid

waste incineration

(percolate)

1–12 days

X. Laevis; P. waltl Erythrocytes Captan (fungicide) 1–12 days

X. Laevis; P. waltl Erythrocytes Cd 1–12 days

G. schubarti Tail Water from Diluvio’s

Basin (Porto Alegre)

13 days

(+) Positive response, (�) negative response, (D–R) dose–response.

damage results in the production of a micronucleus outside themain nucleus [10,11]. The MN test is widely used with fish oraquatic invertebrates, whose numerous and tiny chromosomesoften inhibit the use of metaphase based assays (e.g. SCE). TheSingle Cell Gell Electrophoresis (SCGE) or Comet assay was firstapplied to ecotoxicology about 15 years ago, and it has becomeone of the most popular tests for detecting strand breaks inaquatic animals through in vitro, in vivo and in situ exposures [8]and has been review by Mitchelmore and Chipman [12], Cotelleand Ferard [13], Lee and Steinert [14]. The advantages of theComet assay include the following: (a) genotoxic damage isdetected at the single cell level; (b) most eukaryotic cell types are

to genotoxicants.

e Concentration range Parameter Response Reference

Different seasons

(Autumn, Winter, Summer)

LDR, TL + [49]

1–100 mg/kg TM, TL + [57]

Both

430 mg/ml (100%)

(1:1); 215 mg/ml

(50%); 107.5 mg/ml (25%)

TM + [56]

TM + [58]

TM [65]

14.5–15.5% +

8% NaCl + 10% HCl �15% +

TL D–R [61]

25–75 ppm

1–2.5 ppm

2 mg/l %Tail DNA + [36]

0.01–1 ppm (mg/l) TL + [63]

0.1–1 ppm TM +

0.5–2.5 g/l %Tail DNA + [72]

4–10 ppb %Tail DNA D–R [62]

In both

tissues

TL + [66]

0.1–50 mg/kg %Tail DNA + [34]

0.1–50 mg/kg

0.1–50 mg/kg

0.01–2 mg/kg

0.5% %Tail DNA + [43]

5–15 ppm D–R [64]

ys Diesel oil added to 4 parts

water. Water phase diluted

to 50% DWSF with

dechlorinated water

Classes of damage

(damage score)

+ [60]

PAHs (2000–7000 ng/g d.w.) TL (mm) + [71]

Oxygenated surface

layer of sediment (1–2 cm)

%Tail DNA + [37]

1.5–25% ETM (extent

tail moment)

+ [74]

TL

2–125 mg/l; 62.5 mg/l tDNA, TL, OTM + [75]

2–50 mg/l tDNA, TL, OTM + [76]

Urban waste water Damage index (DI) + [55]

Table 3Assessment of DNA damage by the Comet assay on cells from aquatic animals collected from contaminated sites.

Animal Tissue/cells Site location Principal contaminants Parameter Study type Response Reference

M. edulis Gill cells Chemical dumping

sites, Denmark

Hg, Ni, Zn, Pb, As, Cu TM Native and

transplanted

+ [26]

P. viridis Digestive

gland cells

South West coast of India Fe > Zn > Mn > Cu >

Cd > Ni > Cr > Pb

%Occurrence Native Low [27]

M. galloprovincialis Gill cells Galician coast (NW of Spain) Prestige oil (50% aromatic

hydrocarbons,

22% saturated hydrocarbons,

28% resins and asphalthene).

Metals (Cu, Cd, Pb)

TL Native + [23]

M. galloprovincialis Gill cells Genoa harbour (Italy) PAHs, heavy metals %tDNA Transplanted + [30]

M. galloprovincialis Gill cells River estuarine (Cecina, Italy) Heavy metals %tDNA Native and

transplanted

+ [31]

M. edulis Haemocytes,

gill cells

Reykjavik harbour (Iceland) PAHs %tDNA Transplanted + [28]

M. edulis Haemocytes,

gill cells

Denmark coast Wastewater from

industries (Cr, Ni, Cd)

%tDNA Native + [25]

C. ccidentalis Erythrocytes San Joaquin River (CA, USA) Agricultural runoff events %tDNA Field-caged + [70]

Z. viviparous Erythrocytes Goteborg harbour (Sweden) Oil spill (24.7% PAHs) %tDNA Field-caged + [39]

G. holbrooki Erythrocytes Sarno River, Italy Heavy metals (Cr, Mn),

pesticides, detergents

TL Feral + [59]

Mugil sp., Netuma sp Erythrocytes Two Brazilian estuaries Hydrocarbons, metals %Damage

frequency (DF)

Feral + [38]

L. cephalus Liver cells Three rivers (UK) PAHs, PCBs, OCPs,

heavy metals

%tlDNA Feral and caged + [68]

At high, intermediate

and poor water quality.

G. aculeatus L Erythrocytes 3 locations with different amount

of sewage-treatment effluent

(Northern Germany)

Sewage and

agricultural run-off

%tDNA Feral � [93]

T. rendalli Erythrocytes Lake Igapo II (Brazil) Heavy metals Classes of damage Feral + [69]

B. raddei Erythrocytes,

liver cells

Lanzhou Region (China) Petrochemical

contaminants

%tDNA Feral + [77]

(+) Positive response, (�) marginally significant.

G. Frenzilli et al. / Mutation Research 681 (2009) 80–92 83

suitable for the assay; (c) only a small number of cells arerequired; (d) it is generally faster to conduct and more sensitivethan other available methods for the assessment of strandbreaks; (e) DNA strand breaks form quickly following genotoxicexposure, allowing for an early response evaluation on biota.

In this review we will focus on a synopsis of the genotoxicitydata based on the use of the Comet assay in aquatic animals,published between 2004 and 2007, 5 years after the publication ofother detailed reviews on this topic [12–14]. Furthermore, generalremarks regarding the requirement for suitable guidelines forstandardizing Comet assay protocols in order to achieve a bettercomparability and interpretation of results are discussed.

The aquatic environment provides a sink for many natural andanthropogenically derived chemicals. The genotoxic effects onaquatic organisms have been reviewed by Lee and Steinert in 2003[14]. In the last 4 years different in vivo, and in situ studies havebeen carried out on both invertebrates and vertebrates, as shownin Tables 1–3.

2. Marine environments

2.1. Invertebrates

Hartl et al. [15] used clams as an indicator for the presence ofpotentially genotoxic substances in estuarine environments,demonstrating an increase of DNA damage in haemocytes, gilland digestive gland cells of animals exposed to contaminatedsediments. These studies found gill and digestive cells to be themost sensitive tissue type in detecting genotoxic exposure. Thesame authors [16] also compared the effects of a 3 week exposureto sediment samples (marine and estuarine) in clams and turbot(Scophthalmus maximus), indicating that the marine flatfish turbotis potentially a better indicator species due to naturally lower

levels of background DNA damage. Green-lipped mussels (Perna

viridis) have been exposed to water-borne benzo(a)pyrene (B(a)P,0, 0.3, 3, 30 mg/l) for 0, 1, 3, 6 and 12 days [17]. A time and dose-dependent increase in DNA damage was observed up to 12 daysfollowed by a reduction in DNA damage at the end of the exposure,which was thought to be related to DNA repair. Similar to thestudies of Hartl et al. [15], digestive gland cells were again found torespond better than haemocytes and a good correlation betweenComet and MN test data was observed over the exposure period.The in vivo effects of the extracts of cigar tobacco [18] and oftributyltin (TBT) [19] were studied in mussels by the combined useof Comet assay long with the presence of deformed nuclei and MN,respectively. A common trend was displayed by the twogenotoxicity biomarkers evaluated after exposing green musselsto tobacco products [18]. In addition, a strong correlation wasfound between DNA strand breaks and the formation of micro-nuclei after treating marine molluscs with TBT [19]. Genotoxicityand embryotoxicity of three environmentally relevant pollutantswere studied in oyster embryos (Crassostrea gigas) [20]. While DNAstrand breaks were observed following exposure to B(a)P and theorganochlorine pesticide endosulfan (ES), the synthetic estrogenichormone 17a-ethinylestradiol (EE2) was not found to be genotoxicover the concentrations tested. Moreover, a significant correlationwas observed between genotoxicity and embryotoxicity.

While the majority of environmental studies have investigatedthe impacts of chemical pollutants, some studies have focused onnon-chemical genotoxins in the marine environment. For example,Jha et al. [21] showed how low doses of tritiated water (<500 mGy/h) are able to induce DNA fragmentation and micronucleiformation in mussel haemocytes.

The Comet assay has been widely used to investigate theimpacts of environmental exposure to contaminants. The effects ofcrude oil were investigated in mussels (Mytilus galloprovincialis)

G. Frenzilli et al. / Mutation Research 681 (2009) 80–9284

[22,23] and on both mussels (Mytilus edulis) and sea urchins(Strongylocentrotus droebachiensis) [24]. In the first study theeffects on mussels exposed for 12 days to spilled Prestige crude oil,rich in polycyclic aromatic hydrocarbons (PAHs), were evaluated ingill cells. Comet assay data reflected an increase in the DNAdamage associated with oil exposure, which was higher in musselsexposed to oil with a greater total polycyclic aromatic hydro-carbons (TPAHs) content. An increase in DNA damage was alsofound [23] in gill cells of mussels directly exposed to the Prestigeoil spill. While the levels of DNA damage in mussels collected fromoil-impacted areas reduced following a 7-day recovery period thelevels were still elevated when compared to a clean reference site.Blue mussels (M. edulis) from five locations in Denmark known toreceive waste waters [25] showed higher levels of DNA damage ingill cells compared with haemolymph cells and positive correla-tions were found between tail moment and the concentrations ofchromium, nickel and cadmium present in the waste water.Mussels have also been used in field studies using both native andtransplanted mussels (see Table 3). One such study using a suite ofbiomarkers, including the Comet assay in gill cells, the Neutral RedRetention Time (NRRT) assay and the acetylcholinesterase (AchE)inhibition, investigated a costal area in Denmark impacted by adisused chemical dump site [26]. The results indicated that all thebiomarkers responded to pollution at the impacted sites, but theresults were inconsistent. Seasonal fluctuations were observed inthe levels of DNA damage as detected by the Comet assay and theresults highlight the need to consider the effects natural factorsmay have on biomarker response. It was also speculated that theadaptation of populations to local contaminations may beresponsible for some of the DNA damage patterns observed, andit was recommended that the Comet assay should be used as partof a battery of biomarkers when undertaking environmentalmonitoring programmes. Further studies have also used a multi-biomarker approach when using the Comet assay to assessgenotoxic exposure in the marine environment. Studies usingthe green mussel (P. viridis) investigated DNA damage as detectedby the Comet assay with that observed using chromosomalaberrations (AC), SCE, and the MN assay in samples collected fromboth clean and contaminated sites [27]. Another study investigatedthe genotoxic effect in mussels deployed at polluted harbour sitesin Reykjavık, Iceland [28]. Intertidal mussels from Reykjavıkharbour had higher levels of DNA damage in haemocytes whencompared with subtidal mussels from the same site. While thestudy supports the use of DNA strand breaks as a measure ofenvironmental pollution it also indicated that high levels of intra-site variability in DNA damage can occur. It was suggested that thelocation within a site from which the mussels were collected cansignificantly effect biomarker results due differences in contami-nant exposure or physiological and biochemical responses tovariations in oxygen availability and temperature (time out ofwater prior to processing). As such the study further serves tounderline the importance of standardizing deployment andsampling protocols when using both transplanted and naturalpopulations of sentinel species in biomonitoring studies.

An in vivo study on the effects of cadmium and chromium hasalso been performed using gill cells of mussels (M. edulis) [29]. Theauthors used the modified version of the Comet assay utilizing therestriction enzyme formamidopyrimidine glycosilase (FPG), whichis able to recognise ring-opened purines, as well as 8-oxo-guanine,in order to detect oxidized DNA bases. The presence of oxidizedDNA purines has also been evaluated in a field translocationexperiment [30]. In this study several biomarkers were investi-gated in marine mussels (M. galloprovincialis), which had beencaged for 4 weeks at an industrialized harbour in North West Italy.Mussels were collected at different time intervals (1, 3, 7, 14, 21, 30days) in order to characterize sensitivity and detect temporal

variations in the levels of DNA damage. Besides a rapid increase inthe presence of oxidized DNA bases after 3 days, likely due to thedeployment procedure itself, a statistically significant increase inoxidative DNA damage was observed in organisms transplantedinto the harbour for 30 days. Such a time interval wasrecommended for future transplantation studies using Mytilus

species. The data obtained also confirmed the complex cause–effect relationship between exposure to pollutants, oxyradicalmetabolism and cellular damage. The same authors [31] undertooka study investigating genotoxicity and lysosomal alterations inboth native mussels and transplanted mussel (M. galloprovincialis).The degree of DNA damage detected in gills cells indicated that ittook at least 30 days for mussels transplanted to a contaminatedestuary to display levels of damage comparable to that seen innative populations. Mussels transplanted in parallel at a referencelocation had significantly lower levels of DNA damage. This studyalso observed a significant correlation between the amount of DNAdamage and the lysosomal membrane destabilization as detectedby NRRT assay. The detection of rapidly induced biomarkers, suchas the loss of DNA integrity evaluated by Comet assay andlysosomal membrane instability were induced at similar degree innative and transplanted mussels. In contrast, alterations resultingfrom cumulative events, such as the increase of MN frequency wasmore elevated in native specimens when compared withtransplanted and reference samples. Thus, suggesting that theparallel use of caged and native mussels in environmentalbiomonitoring can improve the characterization of the study area.It is noteworthy that in this study the authors used the same gillcell suspension to perform both Comet assay and MN test.

A limited number of studies have investigated the use ofpolychaetes as sentinel species in genotoxicological studies, as theyare in direct contact with the contaminants present in the sediments.However, those studies undertaken have not found them to be asuitable species for the assessment of the genotoxic risk ofcontaminant exposure. This is likely due to their high tolerance[32], as previously assessed by De Boeck and Kirsch-Volders [33].

2.2. Vertebrates

There are a limited number of studies utilizing the comet assaywith marine fish species in comparison to those using fresh waterspecies (see Table 2). This maybe due to logistical problemsassociated with performing the comet assay at sea, whereexcessive moment due to weather may adversely affect electro-phoresis conditions. Those studies undertaken have often focusedon flatfish, or those species with closely associated with sediments,where the majority of contaminants tend to accumulate.

In vivo studies have been conducted to investigate thesusceptibility to oxidative stress under laboratory conditionsusing the European eel (Anguilla anguilla) [34]. Eels were exposedto increasing concentrations of common environmental pollutants,including B(a)P (0, 0.1, 1, 10, 50 mg/kg), beta-naphto flavone (b-NF, 0, 0.1, 1, 10, 50 mg/kg); Arochlor 1254 (0, 0.1, 1, 10, 50 mg/kg)and 2,3,7,8-tetrachlorodibenzo p-dioxin (TCDD) (0, 0.01, 0.1, 1,2 mg/kg). Analysis of total oxyradical scavenging capacity sug-gested the presence of metabolic oxidative pathways leading thevery reactive OH� on exposure to the chemicals studied. Enhanced7-ethoxyresorufin-o-deethylase (EROD) activity correlated withDNA strand breaks was only observed when the fish were exposedto the nonhalogenated hydrocarbons B(a)P and b-NF. The highestdegree of genotoxic damage was observed in those individuals inwhich the capacity to absorb or scavenge OH� was lowest, thussuggesting a general relationship between oxidative stress and lossof DNA integrity.

The comet assay has also proven to be a useful tool forinvestigating the genotoxic effects of compounds which do not

G. Frenzilli et al. / Mutation Research 681 (2009) 80–92 85

readily bio-accumulate in the marine environment. For example,the environmental effects of styrene, which is a known mutagenand potential carcinogenic compound [35], but not considered tobe harmful to marine fauna due to its high volatility and lowcapacity to bioaccumulate, has been studies in mussel (M. edulis)and fish (Symphodus melleops) [36]. At the end of a 7 day exposureexperiment the measured concentration was 10 times less than theexpected nominal dose (2 mg/l), which was thought to be due tovolatilization and biodegradation of styrene in the test system.However, it was shown to cause a statistically significant increasein DNA damage in blood cells from both mussels (1.8�) and fish(1.6�) over control samples, probably due to the formation ofradical styrene metabolite, which is thought to have a potentoxidative capacity. Hatchery-reared turbot (S. maximus L.) exposedto sediment collected from PAH and heavy metal polluted sites inCork Harbour (Ireland), displayed higher levels of DNA damage inliver cells of fish exposed to contaminated sediments [37].

Mullet and sea catfish have been employed to monitor twoSouthern Brazilian estuaries [38]. In this study an associationbetween hydrocarbons, metals, pH and water temperature and thelevel of damaged cells was observed. Moreover, a generalcorrelation was observed between the levels of DNA damagedetected by the Comet assay and that detected by the MN test. Bothassays were able to detect differences in genotoxic exposurebetween impacted and reference locations. However, the Cometassay appeared to be more sensitive than MN test as it was alsoable to detect intra-site differences due to seasonality. Eelpout(Zoarces viviparus) have been used to investigate a bunker oil spillthat took place in Goteborg harbour, Sweden. The Comet assay andDiffusion assay along with the detection of PAHs metabolites in thebile were investigated and 3 weeks after the spill high levels ofdamaged DNA, paralleled by a peak in bile PAH metabolites weredetected in fish from the most impacted site. Moreover, asignificant recovery in terms of DNA strand break and % apoptoticcells in fish from the spill site was observed after 5 months. Apositive correlation was also shown between the frequency ofapoptotic erythrocyte cells as detected by the Diffusion assay andthe levels of PAHs in the bile [39]. The marine flatfish dab (Limanda

limanda) has been used in a number of studies investigating theimpacts of contaminants in coastal and estuarine waters [40–42].Akcha et al. [40,41] demonstrated that sex and age of the fish had asignificant effect on the presence of DNA strand breaks. Moreover,they observed a significant interaction between age and sex thatmight indicate the complex influence of other factors (i.e.reproductive status) on the extent of DNA damage.

Red tide events represent a risk for different organisms andhuman health as shellfish contaminated with toxins result inneurotoxic shellfish poisoning. The filamentous red marinemacroalgae Polysiphonia fucoides occur along the Swedish Eastcoast and have increased in abundance during recent years. Thelarge amounts of algae result in noxious effects occurring in thoseanimals in close contact with intertidal areas and as such causeproblems for local fishing and tourism industry. Erythrocytes inrainbow trout (Oncorhynchus mykiss) were examined after an in

vivo exposure with a 0.5% marine algal extract for 7 days.Genotoxicity was evaluated by the mild alkaline version of Cometassay (pH 12.1) in order to detect single-strand breaks (ssb); theneutral version at pH 8 to detect double-strand breaks (dsb), theDiffusion assay to evaluate the presence of apoptotic cells and MNtest for the detection of micronucleated cells [43]. Trout treatedwith algal extracts showed increased levels of ssb, comparable tothose induced by the in vivo exposure to the positive control B(a)P(20 mg/kg). The study failed to detect an increase the level of dsb,apoptotic cells and antioxidant parameters and further underlinesthe utility of Comet assay in conjunction with other tests toelucidate the mechanistic action of contaminants.

It is of concern to verify the bioavailability to consumers ofgenotoxic pollutants in contaminated food. Rats (Rattus norvegicus)have been fed with mussels (Mytilus sp.) contaminated by PAHsreleased into the marine environment after the ‘‘Erika’’ shipwreckalong the coast of South Brittany (France) [44]. Animals were feddaily for 2 and 4 weeks with different concentrations of PAHscontained in the contaminated sea-food (312, 569, 870 mg of16 PAHs/kg, corresponding to 33.8, 83.6, 180.7 mg toxic equivalentquantity (TEQ)/kg, respectively). The Comet assay was assessed inliver, bone marrow and blood. A dose–effect–time relationship wasobserved between the amount of DNA damage in the liver andbone marrow of the rats and the PAHs contamination level of themussels, while no damage was detected in blood cells. This studyalso established the threshold of contamination of mussels belowwhich no genotoxicity could be expected (range of 6–34 mg TEQs/kg d.w.).

3. Freshwater environments

3.1. Invertebrates

The majority of freshwater studies using invertebrates havefocused on the use of filter feeding organisms, such as bivalves astest species in both field and laboratory based experiments (seeTables 1 and 3). The freshwater mussel Unio tumidus has been usedin a several in vitro studies investigating the genotoxicogicalproperties of tannins and polyphenols, common contaminants inthe aquatic environment. In vivo studies exposing U. tumidus topolyphenols (60, 200, 500 mM) for 24 and 48 h detected thegreatest effect of DNA damage and oxidative protein modificationafter 24 h [45]. The fresh water mussel has also been used toinvestigate the potential for B(a)P (50–100 mg/l) and ferric iron(20–40 mg/l), exposed in combination or alone, to induce DNAdamage [46]. In parallel oxidative damage was also evaluated byHPLC detection of 8-oxodGuo and malondialdehyde (MDA) tissuelevels used as a marker of lipid peroxidation. While no synergisticeffects were observed a higher level of oxidative damage was foundafter B(a)P treatment and a good correlation was shown between8-oxodGuo and Comet data, suggesting that oxidative stress wasthe mediator of genotoxicity. Moreover, digestive gland cells werefound to be more sensitive to DNA damage when compared withgill cells or haemocytes.

It is worthwhile to mention a recent study [47] whichinvestigated genotoxicity of tap drinking water samples by usingyeast (Schizosaccharomyces pombe) as a model organism. Highlysignificant effects of DNA damage were detected in treated water(tap water) when compared to negative control (distilled water)samples for both 1 h and 2 h exposure, demonstrating thatchlorinated disinfectant mixtures used in drinking water have apotent cumulative genotoxic effect in eukaryotic cells. Such dataobviously raises concerns over the potential genotoxic risk forhuman health following long-term consumption of drinking watertreated this way. The effect of classic (sodium hypochloride andchlorine dioxide) and alternative (peracetic acid, PAA) disinfec-tants on the formation of mutagens in surface waters used forhuman consumption were also investigated in the zebra mussels(Dreissena polymorpha) exposed in experimental basins suppliedwith treated lake water [48]. Mussels maintained in PAA treatedwater failed to show a difference when compared with controlsanimals, whereas those treated with the two chlorinated disin-fectants generally displayed a reduction in DNA damage, thoughtto be associated with the induction of detoxification processes orcross-links inhibiting DNA migration. The same organism and ateleost (Cyprinus carpio) were used in situ to assess seasonalvariations in the water quality of an Italian lake [49]. Seasonalvariations in temperature and oxygen level along with different

G. Frenzilli et al. / Mutation Research 681 (2009) 80–9286

pollutant contents in the lake water appeared to affect the DNAmigration in both carp and zebra mussel cells.

The golden mussel (Limnoperna fortunei) has been used toassess the genotoxic potential of Brazilian lake waters using acombination of the MN test and Comet assay [50]. Mussels wereexposed for 7-days to either water or sediment samples from arange of impacted areas and it was noted that micronucleiformation was responsible for almost 60% of the positive Cometresults observed. The golden mussel has also been used toinvestigate genotoxic effects in vivo of the biocide pentachlor-ophenol (PCP) and the pesticide copper sulfate (CuSO4) [51]. Adose–response relationship was observed for both chemicals andmussels exposed to PCP showed 100% repair 2 h after the exposureperiod. Moreover, the exposure to an environmental sample over 7days confirmed the species sensitivity to water-borne contami-nants and indicated its potential use as a biomonitoring organism.The genotoxicity of naturally occurring toxins has been investi-gated by feeding zebra mussels (D. polymorpha) for 0, 7, 14 and 21days with biotoxin (microcystin) producing strains of cyanobac-teria [52]. Strain specific DNA damage profiles were inducedthrough the formation of ROS, with damage persisting over the 3-week study period confirming the sublethal toxicity of thesetoxins.

Recent studies have suggested the use of the flatwormplanarian (Polycelis felina) an useful organism for evaluatingenvironmental genotoxicity, due to its sensitivity to environmentalcontaminants, low husbandry costs, high reproductive capacityand basal evolutionary position in relation to complex Metazoans.In vivo experiments using copper sulfate have been conducted tovalidate and develop a Comet protocol for this species [53].Furthermore, planarians have been used as convenient non-targeted aquatic bioindicator organisms for short-term aquatictests for herbicides commonly known to contaminate ground-water. Locomotive, morphological changes and mortality havebeen observed following the exposure of P. felina to the herbicideNorflurazon (0.2, 2, 20, 200 mM) for 4 days [54]. The authors founda significant increase in DNA damage at two sublethal doses of 0.2and 2 mM. The planarian Girardia schubart has been used inlaboratory studies to study the toxicity of Brazilian urban streamwater [55]. An increasing gradient of DNA damage was observedmoving towards the rivers mouth, which indicated a correlationbetween water genotoxicity and urbanization.

3.2. Vertebrates

Embryos of zebrafish (Danio rerio) have been used to assessfreeze-dried sediment and sediment extracts from Laguna Lake inThe Philippines [56]. Sediments extracts were found to be mainlycontaminated by perylene and copper and able to induce aconsiderable genotoxicity in the zebra fish embryos. In anotherstudy two compounds PBTA-6 and ADDB identified as majormutagens in river water in Japan were injected in goldfish(Carassius auratus) [57]. Both compounds which are commonlyused in the textile industry, showed significant genotoxic potentialfollowing evaluation by both the MN test and Comet assay. It isinteresting to note that while Comet assay revealed a genotoxiceffect in peripheral erythrocytes, MN test detected positiveresponses in gill cells but not in peripheral erythrocytes, indicatinga difference in sensitivity between the two tests employed.Goldfish have also been used along with the Comet assay and MNtest to study the effects of leachates from landfill sites [58]. Fishwere exposed to the both modified leachate (chemical andbiological treatment) or unmodified leachate. Both the Cometassay (peripheral erythrocytes) and MN test (gill cells) demon-strated the higher genotoxic activity of unmodified leachates whencompared with modified leachates. Similar responses were

observed between the Comet assay and MN test when experimentswere carried out on erythrocytes of the mosquitofish (Gambusia

holbrooki) exposed to polluted waters of an Italian rivercontaminated by heavy metals, pesticides and detergents [59].The combination of Comet and MN assays also proved to besuitable in the evaluation of the genotoxicity water solublefractions of diesel to a neotropical fish species (Prochilodus

lineatus), living in Brazilian rivers [60].The effects of two herbicides (2,4-D and butachlor, both potent

generators of aldehydes) have been studied on the freshwatercatfish (Clarias batrachus) erythrocytes, after exposing organismsfor 48, 72, and 96 h at three sublethal concentrations of the twocompounds. Both concentration and time dependent increases inDNA damage were observed [61]. Gill and kidney cells of the airbreathing freshwater teleost Channa punctatus have also beenutilized to test the time dependent (24, 48, 72 and 96 h) gentoxicityof the organochlorine pesticide endosulfan [62]. Analysis of thedata revealed that gill cells were the most sensitive to exposure butboth cell types responded in a dose-dependent manner.

Two reference compounds, the direct agent N-methyl-N0-nitro-N-nitrosoguanidine (MNNG) and the indirect agent B(a)P wereused to treat carp (C. carpio), rainbow trout (Oncorhyncus mykiss)and clams (Spisula sachalinensis) [63]. The Comet assay wasperformed on gill and liver (or digestive gland); MN on blood cells.The authors found clear distinctions between all concentrations inthe Comet assay but not in the MN and a higher correlationbetween dose and response in the Comet when compared withthat observed in the MN test.

The genotoxicity of the herbicide roundup (R) containingisopropylamine salt of glyphosate has been investigated in vivo

utilizing goldfish (Carassus auratus) [64]. Fish were exposed tothree concentrations of chemical (5, 10, 15 ppm) and erythrocytesexamined using both the Comet assay and MN test 48, 96, 144 hpost-treatment. Dose-dependent increases in the frequencies ofMN, nuclear abnormalities as well as DNA sb were found,underlining the high concordance of the genotoxicity biomarkersused. Comet assay and MN test have been used to assess C. carpio

erythrocytes exposed for 20 days in vivo to lake waters treated withdisinfectants (sodium hypochlorite, peracetic acid, chloridedioxide) for potabilization [65]. After 3 h an increase in DNA sb

was detectable indicating DNA damage was induced directly incirculating erythrocytes. In contrast MN frequencies reached theirpeak at later sampling times, which was thought to be a result ofgenotoxic damage in stem cells of the cephalic kidney beingexpressed in circulating erythrocytes. Eastern mudminnow(Umbra pygmea L.) has been exposed for 3 and 11 days in flow-through aquaria to Rhine groundwater. An increase of both SCE andDNA damage was detected, indicating the presence of genotoxinsin the river, even if by comparison the genotoxic potential asdetected by the SCE assay had markedly decreased compared with27 years ago [66].

The chub (Leuciscus cephalus) has been used to study thepresence of oxidized purines (FPG) in liver, gills, and blood in fishfollowing exhausting exercise [67]. Chub have also been selected toinvestigate the genotoxicity of UK rivers contaminated with arange of organochlorine pesticides (OCPs), heavy metals, PAHs andPCBs [68]. An increase of liver cell DNA sb was observed to correlatewith a decrease in chemical water quality in both wild and cagedchubs. Little evidence of oxidative DNA damage, evaluated by theuse of FPG modified Comet assay was observed in any of the feralfish groups. Comet assay carried out in erythrocytes of Tilapia

rendalli showed significant differences between the fish from aBrazilian lake mainly contaminated by heavy metals (lead andaluminium) and the control animals [69].

In order to study the effects of agriculture chemical runoff, field-caging experiments and laboratory exposures have utilized the

Table 4DNA migration (mm) evaluated in fish erythrocytes at two different pH conditions.

Cell type Fish

erythrocytes

Human

sperm cells

Human

sperm cells

Comet parameter TL (�2.27 mm) Tail DNA TM

pH 12.1 11.21 � 10.61 10.64 � 9.28 11.79 � 18.78

pH > 13 21.8 � 7.47 88.22 � 9.38 279.76 � 43.17

pH 8 30.29 � 14.54 81.92 � 49.39

ANOVA P < 0.001

DNA migration (tail length, tail moment) evaluated in human sperms at two

different pHs. Adapted from [89] and unpublished data of G. Frenzilli.

G. Frenzilli et al. / Mutation Research 681 (2009) 80–92 87

Sacramento sucker (Catastomus occidentalis) [70]. In this study alinkage between induction of DNA sb and the timing of agriculturalrunoff, as well as a causal relationship between runoff events andgenotoxicity as detected by Ames test were observed. The effects ofsediments from four sampling sites along Biobio River (Chile),collected along a known pollution gradient from a petrochemicalindustrial discharge have been assessed in rainbow trouterythrocytes following a 21 days exposure period [71]. Fatheadminnow (Pimephales promelas) have been used in conjunction withthe Comet assay to investigate the effects of exposure to biosolids(sludge that is left following sewage treatment) [72]. Althoughorganic contaminants are known to be present in biosolids, theseare not currently regulated and little data exist on their potentialtoxicity to aquatic organisms. Chemical analysis of the biosolidsrevealed the presence of PAHs, polybrominated diphenyl ethers(PBDEs) and nonylphenols (NPs), and an induction of DNA damagewas found in fish hepatocytes following 14 and 28 days ofexposure. A significant correlation was also found between CYP1Ainduction and DNA damage.

Increasingly amphibians have been used as sentinel species dueto their perceived sensitivity to environmental conditions.Tadpoles of Rana hexadactyla have been exposed to increasingconcentrations of sulfur dyes used in the tannery industry andstatistically significant increases in DNA damage were observed.The level of DNA damage was seen to decrease following a recoveryperiod of 24 h in dechlorinated tap water [73]. Amphibian larvaefrom Xenopus laevis have been used to evaluate the genotoxicpotential of aqueous extracts of two soils (leachates) and of bottomash resulting from municipal solid waste incineration (percolate)[74]. The authors concluded that their studies confirmed theecotoxicological relevance of the amphibian model and high-lighted the importance of bioassays, such as the Comet assay, as acomplement to physico-chemical data, for risk evaluation. Thesame authors have also used amphibian larvae of X. laevis andPleurodeles waltl to study the effects of the fungicide captan and ofcadmium using both the Comet assay and MN test [75,76]. In bothstudies captan (62.5 mg/l) and Cd (2 mg/l) were found to inducemicronuclei in Xenopus but not in P. waltl over all the concentra-tions tested, while Comet assay showed captan and cadmium to begenotoxic from the first day of exposure with the degree ofgenotoxicity both time and dose dependent. Toads (Bufo raddei)caught from areas polluted by petrochemical contaminants inChina and analyzed for Comet assay, MN test and global DNAmethylation status revealed an increased DNA damage in botherythrocytes and liver cells when compared with control animals[77]. The authors determined that liver cells were more sensitivethan blood for assessing genotoxicity. In fact, a positive correlationbetween the increased DNA damage in liver cells and theconcentrations of oil and/or phenol at the sampling locationswas observed. A similar correlation between contaminantconcentration and decrease in global DNA methylation was alsoobserved, suggesting that both methods were suitable formonitoring the genotoxicity effects on amphibians exposed topetrochemical pollutants.

4. General remarks

In general, after 10 years of the Comet assay being almostexclusively applied to mammalian studies across numerousresearch fields [78–83], SCGE started to be applied to environ-mental studies revolutioning the field of genetic ecotoxicology[12–14] and providing the opportunity to study DNA damage,repair and cell death (apoptosis) in different cell types withoutprior knowledge of their karyotype and their turn-over rate [84].The Comet assay has proved to be highly suitable for aquaticgenotoxicity monitoring due to its simplicity and high sensitivity

[63]. However, unlike mammalian genotoxicology, where the focusis centred on a limited number of model species, efforts in theaquatic field have generally lacked coordination, with the resultthat progress towards standardised procedures has been slow andfragmented. In spite of this, compared with other methods SCGEprovides a well-researched and widely utilized tool for addressingvarious aspects relating to genotoxicity in environmental studies.While guidelines relating to the use of the Comet assay have beenpublished for mammalian genotoxicology [85–87], no standardi-zation protocols currently exist for environmental studies.Consequently, the variations in protocols can lead to majordifferences in results, misinterpretations and an inability todirectly compare studies.

4.1. Protocols

In 1989 Singh et al. published a study [88], which investigatedthe presence of alkali-sensitive sites using several tests includingthe Comet assay in both human and mouse cells. A higher level ofssb (106 to 107) were detected in human and mouse sperm cells,but these were not seen in human lymphocytes or mouse bonemarrow cells. Such ssb were not due to pre-existing single- and/ordouble-strand breaks, but were associated with alkali labile sites(als) formed by functional condensation. Such characteristiccondensed chromatin was also observed in chicken erythrocyteDNA and is thought to be typical of nucleated erythrocytes. Tenyears later Frenzilli et al., while proposing an adaptation of SCGEfor monitoring marine ecosystems compared the basal levels ofDNA migration in mussel gill cells, mussel haemocytes and fisherythrocytes following both mild alkaline (pH 12.1) and alkalineversions (pH > 13) of Comet assay [89]. Results demonstrated thepresence of alkali labile sites in fish erythrocytes, which wasfurther supported by studies undertaken in trout erythrocytes [90].Similar results were obtained with human sperm cells (seeTable 4), indicating that the mild alkaline version of the assayshould be considered when dealing with highly condensedchromatin cells, in order to prevent higher background levels ofDNA strand breaks. Initial studies applying the alkaline version ofthe Comet assay to human sperm cells observed that the presenceof als confounded DNA migration patterns detected [91]. Otherresearch groups have also reported detecting high control values offish erythrocytes when compared to control values obtained forhuman blood samples when using the alkaline version of Cometassay (pH > 13) [92]. Indeed this problem has been highlighted inother studies using teleost species where excessive DNA tailmigration has inhibited the interpretation of results [93].

In addition to the variation in response depending on cell type itis also apparent a wide array of Comet assay protocols (differing interms of agarose concentrations, lysing parameters and electro-phoresis parameters) have been utilized in environmentalmonitoring studies. This is in contrast to those undertaken inmammalian research, where standardised protocols are available[85–87]. Therefore, effort is required to establish standardisedprotocols for each key species and cell types commonly used in

G. Frenzilli et al. / Mutation Research 681 (2009) 80–9288

environmental studies to produce control cells displaying around20% tail DNA, as described by Tice et al. [85]. Such improvements interms of standardised protocols and inter-laboratory ring testingare essential if the Comet assay is to develop as an environmentalmonitoring tool.

4.2. Apoptotic cells detection

Another ‘‘hot issue’’ is related to the important and criticalaspect represented by the role of apoptotic cells within eco-genotoxicity. Since 1993 apoptotic cells have been identified as the‘‘large fan-like tails and small head cells’’ (i.e. so-called hedgehogs),while necrotic cells were concluded to form comets with relativelylarge heads and narrow tails of varying lengths (i.e. cometsindistinguishable from those resulting from genotoxic damage)[94]. Other studies identified the Comet assay as a useful tool formeasuring by appearance the late stages of apoptosis [95] and todetect ‘‘hedgehog’’ comet images even after liquid holding [96].Such an approach, similar to that applied to mussel cells [97] hasbeen recently still reported in a methodological review of Cometassay published on Nature Protocols [98]. However, due to the verylow molecular weight (LMW) of DNA in terminal apoptotic andnecrotic cells, Vasquez and Tice [99] suggested that the DNA ofmany of these cells may be lost from the gels under the typicalelectrophoretic conditions. They suggested the development of aLMW diffusion assay for detecting apoptotic/necrotic cells andindicated the usefulness of this tool in helping to set the maximumdose of a test substance to test in the Comet assay or in theinterpretation of a positive response [85]. A detailed protocol ofdiffusion assay has been published after validating the techniquewith inducers of apoptosis and necrosis and comparing the resultsobtained with other methods of apoptosis measurement [100,101].Apoptotic cell nuclei were revealed by presenting a hazy orundefined outline without any clear boundary, due to nucleoso-mal-sized DNA diffusing into agarose. It has been verified thatapoptosis does not necessarily need to correlate or coincide withDNA damage observed with genotoxic substances in the Cometassay [102], while the evaluation of the presence of apoptotic cells,alongside genotoxicity, can give a better picture about the effect ofa compound. The apoptosis assessment by the DNA Diffusion assayhas been widely applied, as reported in a recent review [103].Indeed a small number of environmental studies have also shownpromising results using the Diffusion assay [104,39,43]. An in vivo

approach studied the ability of chemicals to induce DNA sb byComet assay and apoptosis by Diffusion assay in the European eel(A. anguilla) [104]. Arochlor 1254 exhibited a general enhancementof apoptotic cell frequency over the control levels, while apoptoticcells occurred at the lowest doses after exposure to b-NF. Dioxin(TCDD) caused apoptosis in the highest and lowest concentrationsused. An interesting behaviour was observed for B(a)P, whichbehaved as a real carcinogenic compound, with a significantinduction of apoptotic cells only evident after significant levels ofDNA damage had already been induced. Eelpout erythrocyteswere also investigated using the Diffusion assay to study theeffects related with a bunker oil spill that occurred at Goteborgharbour [39]. In this study the levels of apoptotic cells showed asignificant recovery in fish from the oil-impacted site 5 monthsafter the spill. It is suggested that the Diffusion assay should beconsidered as a complementary tool to be used along side theComet assay to explore different aspects of genotoxicity inenvironmental studies.

4.3. Cell type/target organs

More than the half of the articles reported here used circulatingblood cells (either molluscs haemocytes or fish erythrocytes), as

target cells for Comet assay analysis. This is likely to be due to thepractical advantages of processing tissues containing nucleatedcells, which are already separated. Solid tissues such as gill, fishhepatocytes or mussel digestive gland require cell dissociationprior to the Comet procedure, with the potential of introducingdamage through enzymatic and or mechanical manipulation.However, when comparing cells types it is usually reported thatcirculating cells are usually less sensitive than hepatocytes or gillcells [15,17,44,62,63,77]. This is likely to be due to the fact that gillsare in direct contact with the potentially stressing environmentand represent an important tissue for the uptake of contaminants.Similarly, mussel digestive gland and fish liver are more directlyimpacted by contaminants, which are accumulated through thefood. Hepatocytes and mussel digestive cells are known to bothbioaccumulate and biotransform pollutants, two processes thatcan lead to the activation of genotoxins or promoting theenhancement of reactive oxygen species and other radicals [17].Blood and to a lesser extent the haemolymph are ‘‘buffered’’tissues, in which contaminants arrive having crossed numerousbiological barriers. Moreover, mature erythrocytes have a lowmetabolising capacity and turn over rates, which act to minimiseboth the accumulation of pollutants and DNA damage [17]. This isparticularly apparent when reviewing the results from MN tests infish, where the frequency of micronucleated erythrocytes isextremely low, requiring the scoring of several thousands of cellsin order to achieve any sort of statistically significant result. In afew articles the Comet assay and MN test have been performed onthe same cell population such as mussel gill cells. This approach[31] offers the opportunity for investigating the relationshipbetween the formation of strand breaks (early and reversible) andchromosomal alterations (cumulative and irreversible). Moreover,data by Kim and Hyun [63] indicated that different cell typesresponded with different sensitivities, depending on the con-taminant used. Gill cells appeared to be the most sensitive forMNNG, while liver and digestive gland were more sensitive toB(a)P exposure, suggesting that uptake routes and bioaccumula-tion mechanisms need to be taken into account when designingexperiment systems.

4.4. Comet assay and other genotoxicity tests: a comparison

The application of biomarkers is an important approach wheninvestigating the causal relationship between exposure toenvironmental pollutants and the observation of long-term effectsin individuals and populations. The use of a battery of biomarkersin field monitoring has been increasing over the past 15 years.Genotoxicity biomarkers are now considered to be an integral partof this approach as the exposure to genotoxic agents may exertdamage beyond that of individuals and may be detected throughseveral generations. As has already been shown for the associationbetween chromosomal aberrations and cancer risk [105] and thecorrelation between MN frequency in peripheral blood lympho-cytes and cancer risk in human populations [106].

SCGE is currently the most widely employed method to detectDNA lesions in eco-genotoxicology. The impact of genotoxicchemicals on the integrity of cellular DNA is one of the first eventsin organisms exposed to contaminants. The production of DNA sb

correlates well with the mutagenic and carcinogenic properties ofenvironmental pollutants with diverse structures. Chromosomaldamage expressed after cell replication represents an accumulatedeffect associated with long-term exposure. In fact while the MNdetects non-repairable damage, such as clastogenic and aneugeniclesions, SCGE detects recent lesions that can be repaired, such asbreaks and alkali labile sites.

Several studies have been carried out using both the Cometassay and MN test and/or other genotoxicity tests to evaluate the

Table 5Studies using both Comet assay and MN test and/or other genotoxicity test to evaluate the impact of environmental pollutants.

Organism Exposure SCGE SCE MN Reference

P. viridis Water-borne B(a)P + + [17]

P. viridis Low contaminated sites � � � [27]

M. edulis Tritiated water + + [21]

M. edulis Tributyltin + + [19]

M. galloprovincialis Heavy metals impacted estuarine environment + + [31]

L. fortunei PCP (biocide); CuSO4 + + [51]

D. polymorpha Disinfectants + + [48]

C. carpio, O. mykiss, S. sachalinensis MNNG; B(a)P + + [63]

Mugil sp., Netuma sp Biomonitoring (two Brazilian rivers) + + [38]

G. aculeatus L. Sewage-treatment effluent � + [93]

U. pygmea L. Rhine water + + [7]

C. auratus Glyphosate + + [64]

C. auratus PBTA-6, ADDB (dyes) + + [57]

C. auratus Leachates from landfill sites + + [58]

G. holbrooki River water (heavy metals, pesticides, detergents) + + [59]

P. lineatus Diesel water soluble fraction (DWSF) + + [60]

X. laevis Aqueous extracts of two soils (leachates) and of bottom

ash from municipal solid waste incineration (percolate)

+ + [74]

X. laevis Captan (fungicide) + + [75]

P. waltl + �

X. laevis Cadmium + + [76]

P. waltl + �

B. raddei Petrochemical contaminants + � [77]

G. Frenzilli et al. / Mutation Research 681 (2009) 80–92 89

impact of environmental pollutants (see Table 5). An agreementbetween SCE test and Comet assay data was found in Easternmudminnow (U. pygmea L.) exposed for 3 and 11 days in flow-through aquaria to Rhine groundwater [66]. A high concordancebetween the genotoxic effects detected by SCGE and MN test wasreported by different authors for several organisms[45,6,8,21,49,59], even if the higher sensitivity of the Comet assayhas been often underlined [48,50,63]. Masuda et al. [57] foundpositive Comet results on erythrocytes of goldfish injected withtwo mutagenic dyes present in river waters, but did not find anymicronucleus in the same cell type. However, they found a dose–response in terms of MN frequencies in gill cells. Similar resultswere obtained by Deguchi et al. [58], after exposing goldfish toleachates from landfill sites. Such findings might indicate thatdifferent genotoxicity biomarkers should be carried out ondifferent target organs to determine the best approach, beforeany large-scale experiments are conducted.

Some researchers found negative responses of MN test inpresence of Comet positivity [77]. Another study took intoconsideration the evaluation of DNA damage detected by Cometassay (whose electrophoresis was run at pH 10) together withother genotoxicity biomarkers like AC, SCE, MN in green mussels(P. viridis) from clean and contaminated sites [27]. However, lowlevels of genotoxic damage were detected, making comparisonsbetween biomarkers difficult to undertake. Wirzinger et al. evenreported MN to possess more potential to differentiate genotoxiceffects than SCGE, demonstrating a negative correlation betweenstrand breakage and micronuclei for the three-spined stickleback(Gasterosteus aculeatus L.) from the most polluted locations studiedand for the pooled data of all locations [93]. However, it should betaken into account that in this work the alkaline version (pH > 13)of the Comet assay in fish erythrocytes was chosen and nucleidisplaying >75% DNA in the tail were not scored, possibly leadingto an underestimation of the actual damage.

Overall, when taking into account complementary informationarising from the use of both the Comet and MN assays severalauthors have suggested that both assays should be employed dueto their abilities to detect different aspects of genotoxicity andclastogenicity [75,74,31,65,59,60].

4.5. Indigenous versus transplanted species

The vast majority of genotoxicity biomarkers are unable toidentify the class of chemicals responsible for the effects observed,but results from the interactions, namely antagonism, additive orsynergism, occurring within environmental mixture of pollutantsare evident. In this respect, environmentally complex mixtures canbe tested with the Comet assay, which has proven to be an effectivebiomarker of genotoxicity through in situ approaches using eitherindigenous or transplanted specimens. The use of indigenousorganisms species is logistically easy and often representative of amore stable situation. Chub selected to investigate the UK riverscontaminated with various levels of contaminants [68] showed anincrease of liver cell sb correlated with a decrease in chemicalwater quality in both wild and caged chubs. These studies alsodemonstrated that in feral fish the % tail DNA detected in cometspositively correlated with EROD, total muscle OCPs and withspecific PCBs congeners, while in caged fish % tail DNA onlycorrelated with EROD activity. On the other hand, little evidence ofoxidative DNA damage, evaluated by the use of FPG modifiedComet assay was observed in any of the feral fish groups.

Nevertheless the use of indigenous organisms may behampered by the absence of a suitable sentinel species or bymigrational movements in case of fish. Moreover, the genotoxicresponses obtained from indigenous specimens may be influ-enced by physiological adaptations, which potentially renderchronically exposed specimens less responsive to genotoxicimpacts. Apart from precluding the above difficulties, the use ofcaged sentinel specimens (mainly mussels and fish) also avoidsthe possibility that results from sampling programmes spreadover relatively wide geographical areas might be influenced byfactors such as differences in genetic background, develop-mental stage or reproductive status. Finally, genotoxic biomar-kers in transplanted specimens can be referred to a preciseexposure window, while the responses of native specimensresult from an integration of exposure over the life of theorganism [26].

A few articles have dealt with the comparison of Comet assayresults between indigenous and transplanted specimens

G. Frenzilli et al. / Mutation Research 681 (2009) 80–9290

[26,30,31,68]. These studies indicate that after a 4 weekexposure, the results of Comet assay are comparable intransplanted and indigenous mussel, while biomarkers basedon the accumulation of damage, such as the Micronucleus test,are more effective (i.e. discriminate better between impactedand reference sites) when analysed in indigenous specimens[30,31]. The Comet assay has also been applied in situ toevaluate the time-course variations in DNA damage followingfield translocation experiments. Such studies demonstrate thatwithin the first 7 days DNA damage fluctuates, which is mostlikely attributed to manipulation disturbance. After 2 weeksDNA migration reaches a plateau, eventually decreasing after 1month from deployment. These features suggest that very shortlasting (less than 10 days) transplantation experiment may giveinvalid results in environmental monitoring studies as resultsmay be biased by artefacts associated with the samplingprocedure.

In conclusion, the studies reviewed here demonstrate that theComet assay has broad applicability when applied to aquaticbioindicator organisms, providing a sensitive, rapid and versatilesystem for the study of environmental genotoxicity. Its combineduse with other biomarkers as well as standardization and inter-laboratory calibration are recommended to further strengthen itsuse in environmental assessment studies.

References

[1] N. Pollack, A.R. Cunningham, H.S. Rosenkrantz, Environmental persistenceof chemicals and their carcinogenic risks to human, Mutat. Res. 528 (2003)81–91.

[2] M.D. Waters, H.F. Stack, N.E. Garrett, M.A. Jackson, The genetic activity profiledatabase, Environ. Health Perspect. 96 (1991) 41–45.

[3] M.D. Waters, H.F. Stack, M.A. Jackson, Genetic toxicology data in the evaluation ofpotential human environmental carcinogens, Mutat. Res. 437 (1999) 21–49.

[4] B.N. Ames, J. McCann, E. Yamasaki, Methods for detecting carcinogens andmutagens with the Salmonella/mammalian-microsome mutagenicity test,Mutat. Res. 31 (1975) 347–364.

[5] J.M. Parry, D.J. Tweats, M.A.J. Al-Mossawi, Monitoring the marine environmentfor mutagens, Nature 264 (1976) 538–540.

[6] A.E. Prein, G.M. Thie, G.M. Alink, C.L.M. Poels, J.H. Koeman, Cytogenetic changesin fish exposed to water of the river Rhine, Sci. Total Environ. 9 (1978) 287–291.

[7] G.M. Alink, E.M.H. Frederix-Wolters, M.A. van der Gaag, J.F.F. van der Kerkhoff,C.L.M. Poels, Induction of sister-chromatid exchanges in fish exposed to Rhinewater, Mutat. Res. 78 (1980) 369–374.

[8] T. Ohe, T. Watanabe, K. Wakabayashi, Mutagens in surface waters: a review,Mutat. Res. 567 (2004) 109–149.

[9] S.A. Latt, J.W. Allen, In vitro and in vivo analysis of sister-chromatid exchangeformation, in: B.J. Kilbey, M. Legator, W. Nichols, C. Ramel (Eds.), Handbook ofMutagenicity Test Procedures, Elsevier, Amsterdam, 1977, pp. 275–291.

[10] D. Jenssen, C. Ramel, The micronucleus test as a part of a short-term mutageni-city test program for the prediction of carcinogenicity evaluated by 143 agentstested, Mutat. Res. 75 (1980) 191–202.

[11] C.K. Grisolia, A comparison between mouse and fish micronucleus test usingcyclophosphamide, mitomycin C, and various pesticides, Mutat. Res. 518 (2002)145–150.

[12] C.L. Mitchelmore, J.K. Chipman, DNA strand breakage in aquatic organisms andthe potential value of the comet assay in environmental monitoring, Mutat. Res.399 (1998) 135–147.

[13] S. Cotelle, J.F. Ferard, Comet assay in genetic ecotoxicology: a review, Environ.Mol. Mutagen. 34 (1999) 246–255.

[14] R.F. Lee, S. Steinert, Use of the single cell gel electrophoresis/comet assay fordetecting DNA damage in aquatic (marine and freshwater) animals, Mutat. Res.544 (2003) 43–64.

[15] M.G. Hartl, B.M. Coughlan, D. Sheehan, C. Mothersill, F.N. van Pelt, S.J. O’Reilly, J.J.Heffron, J. O’Halloran, N.M. O’Brien, Implications of seasonal priming andreproductive activity on the interpretation of Comet assay data derived fromthe clam, Tapes semidecussatus Reeves 1864, exposed to contaminated sedi-ments, Mar. Environ. Res. 57 (2004) 295–310.

[16] M.G. Hartl, M. Kilemade, B.M. Coughlan, J. O’Halloran, F.N. van Pelt, D. Sheehan, C.Mothersill, N.M. O’Brien, A two-species biomarker model for the assessment ofsediment toxicity in the marine and estuarine environment using the cometassay, J. Environ. Sci. Health A Tox. Hazard. Subst. Environ. Eng. 41 (2006) 939–953.

[17] W.H. Siu, J. Cao, R.W. Jack, R.S. Wu, B.J. Richardson, L. Xu, P.K. Lam, Application ofthe comet and micronucleus assays to the detection of B[a]P genotoxicity inhaemocytes of the green-lipped mussel (Perna viridis), Aquat. Toxicol. 66 (2004)381–392.

[18] A. Nagarajappa, U. Ganguly, Goswami, DNA damage in male gonad cells of Greenmussel (Perna viridis) upon exposure to tobacco products, Ecotoxicology 15(2006) 365–369.

[19] J.A. Hagger, M.H. Depledge, T.S. Galloway, Toxicity of tributyltin in the marinemollusc Mytilus edulis, Mar. Pollut. Bull. 51 (2005) 811–816.

[20] N. Wessel, S. Rousseau, X. Caisey, F. Quiniou, F. Akcha, Investigating therelationship between embriotoxic and genotoxic effects of benzo[a]pyrene,17a-ethinylestradiol and endosulfan on embryos, Aquat. Toxicol. 85 (2007)133–142.

[21] A.N. Jha, Y. Dogra, A. Turner, G.E. Millward, Impact of low doses of tritium on themarine mussel, Mytilus edulis: genotoxic effects and tissue-specific bioconcen-tration, Mutat. Res. 586 (2005) 47–57.

[22] B. Perez-Cadahia, B. Laffon, E. Pasaro, J. Mendez, Evaluation of PAH bioaccumu-lation and DNA damage in mussels (Mytilus galloprovincialis) exposed to spilledPrestige crude oil, Comp. Biochem. Physiol. C: Toxicol. Pharmacol. 138 (2004)453–460.

[23] B. Laffon, T. Rabade, E. Pasaro, J. Mendez, Monitoring of the impact of Prestige oilspill on Mytilus galloprovincialis from Galician coast, Environ. Int. 32 (2006) 342–348.

[24] I.C. Taban, R.K. Bechmann, S. Torgrimsen, T. Baussant, S. Sanni, Detection of DNAdamage in mussels and sea urchins exposed to crude oil using comet assay, Mar.Environ. Res. 58 (2004) 701–705.

[25] J. Rank, K. Jensen, P.H. Jespersen, Monitoring DNA damage in indigenous bluemussels (Mytilus edulis) sampled from coastal sites in Denmark, Mutat. Res./Genet. Toxicol. Environ. Mutagen. 585 (2005) 33–42.

[26] J. Rank, K.K. Lehtonen, J. Strand, M. Laursen, Aquatic toxicology, DNA damage,acetylcholinesterase activity and lysosomal stability in native and transplantedmussels (Mytilus edulis) in areas close to coastal chemical dumping sites inDenmark, Aquat. Toxicol. 84 (2007) 50–61.

[27] P.K. Krishnakumar, G. Sasikumar, G.S. Bhat, D.P. Asokan, Biomarkers of environ-mental contaminants in field population of green mussel (Perna viridis) fromKarnataka–Kerala coast (South West coast of India), Ecotoxicology 15 (2006)347–352.

[28] H.P. Halldorsson, G. Ericson, J. Svavarsson, DNA strand breakage in mussels(Mytilus edulis L.) deployed in intertidal and subtidal zone in Reykjavik harbour,Mar. Environ. Res. 58 (2004) 763–767.

[29] C. Emmanouil, T.M. Sheehan, J.K. Chipman, Macromolecule oxidation and DNArepair in mussel (Mytilus edulis L.) gill following exposure to Cd and Cr(VI), Aquat.Toxicol. 82 (2007) 27–35.

[30] F. Regoli, G. Frenzilli, R. Bocchetti, F. Annarumma, V. Scarcelli, D. Fattorini, M.Nigro, Time-course variations of oxyradical metabolism, DNA integrity andlysosomal stability in mussels, Mytilus galloprovincialis, during a field transloca-tion experiment, Aquat. Toxicol. 68 (2004) 167–178.

[31] M. Nigro, A. Falleni, I. Del Barga, V. Scarcelli, P. Lucchesi, F. Regoli, G. Frenzilli,Cellular biomarkers for monitoring estuarine environments: transplanted ver-sus native mussels, Aquat. Toxicol. 77 (2006) 339–347.

[32] L. Bach, A. Palmqvist, L.J. Rasmussen, V.E. Forbes, Differences in PAH tolerancebetween Capitella species underlying biochemical mechanisms, Aquat. Toxicol.74 (2005) 307–319.

[33] M. De Boeck, M. Kirsch-Volders, Nereis virens (Annelida: Polychaeta) is not anadequate sentinel species to assess the genotoxic risk (comet assay) of PAHexposure to the environment, Environ. Mol. Mutagen. 30 (1997) 82–90.

[34] F. Regoli, G.W. Winston, S. Gorbi, G. Frenzilli, M. Nigro, I. Corsi, S. Focardi,Integrating enzymatic responses to organic chemical exposure with total oxy-radical absorbing capacity and DNA damage in the European eel Anguillaanguilla: toward development of a more holistic biomarker assessment, Environ.Toxicol. Chem. 22 (2003) 2120–2129.

[35] International Agency for Research on Cancer (IARC)—Summaries & Evaluations,82 (2002) 437.

[36] E. Mamaca, R.K. Bechmann, S. Torgrimsen, E. Aas, A. Bjornstad, T. Baussant, S.L.Floch, The neutral red lysosomal retention assay and Comet assay on haemo-lymph cells from mussels (Mytilus edulis) and fish (Symphodus melops) exposedto styrene, Aquat. Toxicol. 75 (2005) 191–201.

[37] M.G.J. Hartl, M. Kilemade, D. Sheehan, C. Mothersill, J. O’Halloran, N.M. O’Brien,F.N.A.M. van Pelt, Hepatic biomarkers of sediment-associated pollutionin juvenile turbot, Scophthalmus maximus L, Mar. Environ. Res. 64 (2007)191–208.

[38] V.M. de Andrade, J. da Silva, F.R. da Silva, V.D. Heuser, J.F. Dias, M.L. Yoneama, T.R.de Freitas, Fish as bioindicators to assess the effects of pollution in two SouthernBrazilian rivers using the Comet assay and Micronucleus test, Environ. Mol.Mutagen. 44 (2004) 459–468.

[39] G. Frenzilli, V. Scarcelli, I. Del Barga, M. Nigro, L. Forlin, C. Bolognesi, J. Sturve,DNA damage in eelpout (Zoarces viviparus) from Goteborg harbour, Mutat. Res.552 (2004) 187–195.

[40] F. Akcha, G. Leday, A. Pfohl-Leszkowicz, Potential value of the comet assay andDNA adduct measurement in dab (Limanda limanda) for assessment of in situexposure to genotoxic compounds, Mutat. Res. 534 (2003) 21–32.

[41] F. Akcha, F. Vincent Hubert, A. Pfohl-Leszkowicz, Measurement of DNA adductsand strand breaks in dab (Limanda limanda) collected in the field: effects of biotic(age, sex) and abiotic (sampling site and period) factors on the extent of DNAdamage, Mutat. Res. 552 (2004) 197–207.

[42] B.P. Lyons, G.D. Stentiford, J. Bignell, F. Goodsir, D.B. Sivyer, M.J. Devlin, D. Lowe,A. Beesley, C.K. Pascoe, M.N. Moore, E. Garnacho, A biological effects monitoringsurvey of Cardigan Bay using flatfish histopathology, cellular biomarkers andsediment bioassays: findings of the Prince Madog Prize 2003, Mar. Environ. Res.62 (Suppl) (2006) S342–S346.

G. Frenzilli et al. / Mutation Research 681 (2009) 80–92 91

[43] I. Del Barga, G. Frenzilli, V. Scarcelli, M. Nigro, A. Malmvarn, L. Asplund, L. Forlin, J.Sturve, DNA strand breaks induced by algal extracts (Polysiphonia fucoides) inrainbow trout, Mar. Environ. Res. 2 (2006) 283–286.

[44] S. Lemiere, C. Cossu-Leguille, A. Bispo, M.J. Jourdain, M.C. Lanhers, D. Burnel, P.Vasseur, DNA damage measured by the single-cell gel electrophoresis (Comet)assay in mammals fed with mussels contaminated by the ‘Erika’ oil-spill, Mutat.Res. 581 (2005) 11–21.

[45] M. Labieniec, M. Biernat, T. Gabryelak, Response of digestive gland cells offreshwater mussel Unio tumidus to phenolic compound exposure in vivo, Cell.Biol. Int. 31 (2007) 683–690.

[46] S. Lemiere, C. Cossu-Leguille, A.M. Charissou, P. Vasseur, DNA damage (cometassay) and 8-oxodGuo (HPLC-EC) in relation to oxidative stress in the freshwaterbivalve Unio tumidus, Biomarkers 10 (2005) 41–57.

[47] P. Banerjee, S.N. Talapatra, N. Mandal, G. Sundaram, A. Mukhopadhyay, D.Chattopadhyay, S.K. Banerjee, Genotoxicity study with special referenceto DNA damage by comet assay in fission yeast, Schizosaccharomycespombe exposed to drinking water, Food Chem. Toxicol. 46 (2008) 402–407.

[48] C. Bolognesi, A. Buschini, E. Branchi, P. Carboni, M. Furlini, A. Martino, M.Monteverde, P. Poli, C. Rossi, Comet and micronucleus assays in zebra musselcells for genotoxicity assessment of surface drinking water treated with threedifferent disinfectants, Sci. Total Environ. 333 (2004) 127–136.

[49] C. Pellacani, A. Buschini, M. Furlini, P. Poli, C. Rossi, A battery of in vivo and invitro tests useful for genotoxic pollutant detection in surface waters, Aquat.Toxicol. 77 (2006) 1–10.

[50] I.V. Villela, I.M. de Oliveira, J.C. Silveira, J.F. Dias, J.A. Henriques, J. da Silva,Assessment of environmental stress by the micronucleus and comet assays onLimnoperna fortunei exposed to Guaiba hydrographic region samples (Brazil)under laboratory conditions, Mutat. Res. 628 (2007) 76–86.

[51] I.V. Villela, I.M. de Oliveira, J. da Silva, J.A.P. Henriques, DNA damage and repair inhaemolymph cells of golden mussel (Limnoperna fortunei) exposed to environ-mental contaminants, Mutat. Res. 605 (2006) 78–86.

[52] G. Juhel, J. O’Halloran, S.C. Culloty, R.M. O’riordan, J. Davenport, N.M. O’Brien, K.F.James, A. Furey, O. Allis, In vivo exposure to microcystins induces DNA damage inthe haemocytes of the zebra mussel, Dreissena polymorpha, as measured with thecomet assay, Environ. Mol. Mutagen. 48 (2007) 22–29.

[53] T. Guecheva, J.A. Henriques, B. Erdtmann, Genotoxic effects of copper sulphate infreshwater planarian in vivo, studied with the single-cell gel test (comet assay),Mutat. Res. 497 (2001) 19–27.

[54] T. Horvat, N. Kalafatic, G. Kopjar, Kovacevic, Toxicity testing of herbicide nor-flurazon on an aquatic bioindicator species—the planarian Polycelis felina (Daly),Aquat. Toxicol. 73 (2005) 342–352.

[55] D. Pra, A.H. Lau, T. Knakievicz, F.R. Carneiro, B. Erdtmann, Environmentalgenotoxicity assessment of an urban stream using freshwater planarians, Mutat.Res. 585 (2005) 79–85.

[56] T. Kosmehl, A.V. Hallare, T. Braunbeck, H. Hollert, DNA damage induced bygenotoxicants in zebrafish (Danio rerio) embryos after contact exposure tofreeze-dried sediment and sediment extracts from Laguna Lake (The Philippines)as measured by the comet assay, Mutat. Res. 650 (2008) 1–14.

[57] S. Masuda, Y. Deguchi, Y. Masuda, T. Watanabe, H. Nukaya, Y. Terao, T. Takamura,K. Wakabayashi, N. Kinae, Genotoxicity of 2-[2-(acetylamino)-4-[bis(2-hydro-xyethyl)amino]-5-methoxyphenyl]-5-amino-7-bromo-4-chloro-2H-benzotria-zole (PBTA-6) and 4-amino-3,30-dichloro-5,40-dinitro-biphenyl (ADDB) ingoldfish (Carassius auratus) using the micronucleus test and the comet assay,Mutat. Res. 560 (2004) 33–40.

[58] Y. Deguchi, T. Toyoizumi, S. Masuda, A. Yasuhara, S. Mohri, M. Yamada, Y. Inoue, N.Kinae, Evaluation of mutagenic activities of leachates in landfill sites by micro-nucleus test and comet assay using goldfish, Mutat. Res. 627 (2007) 178–185.

[59] C. Russo, L. Rocco, M.A. Morescalchi, V. Stingo, Assessment of environmentalstress by the micronucleus test and the Comet assay on the genome of teleostpopulations from two natural environments, Ecotoxicol. Environ. Saf. 57 (2004)168–174.

[60] T.P. Vanzella, C.B.R. Martinez, I.M.S. Colus, Genotoxic and mutagenic effects ofdiesel oil water soluble fraction on a neotropical fish species, Mutat. Res. 631(2007) 36–43.

[61] B. Ateeq, M. Abul Farah, W. Ahmad, Detection of DNA damage by alkaline singlecell gel electrophoresis in 2,4-dichlorophenoxyacetic-acid- and butachlor-exposed erythrocytes of Clarias batrachus, Ecotoxicol. Environ. Saf. 62 (2005)348–354.

[62] S. Pandey, N.S. Nagpure, R. Kumar, S. Sharma, S.K. Srivastava, M.S. Verma,Genotoxicity evaluation of acute doses of endosulfan to freshwater teleostChanna punctatus (Bloch) by alkaline single-cell gel electrophoresis, Ecotoxicol.Environ. Saf. 65 (2006) 56–61.

[63] I.Y. Kim, C.K. Hyun, Comparative evaluation of the alkaline comet assay with themicronucleus test for genotoxicity monitoring using aquatic organisms, Ecotox-icol. Environ. Saf. 64 (2006) 288–297.

[64] T. Cavas, S. Konen, Detection of cytogenetic and DNA damage in peripheralerythrocytes of goldfish (Carassius auratus) exposed to a glyphosate formulationusing the micronucleus test and the comet assay, Mutagenesis 22 (2007) 263–268.

[65] A. Buschini, A. Martino, B. Gustavino, M. Monfrinotti, P. Poli, C. Rossi, M. Santoro,A.J. Dorr, M. Rizzoni, Comet assay and micronucleus test in circulating erythro-cytes of Cyprinus carpio specimens exposed in situ to lake waters treated withdisinfectants for potabilization, Mutat. Res. 557 (2004) 119–129.

[66] G.M. Alink, J.T. Quik, E.J. Penders, A. Spenkelink, S.G. Rotteveel, J.L. Maas, W.Hoogenboezem, Genotoxic effects in the Eastern mudminnow (Umbra pyg-

maea L.) after exposure to Rhine water, as assessed by use of the SCE andComet assays: a comparison between 1978 and 2005, Mutat. Res. 631 (2007)93–100.

[67] S.O. Aniagu, N. Day, J.K. Chipman, E.W. Taylor, P.J. Butler, M.J. Winter, Doesexhaustive exercise result in oxidative stress and associated DNA damage in thechub (Leuciscus cephalus)? Environ. Mol. Mutagen. 47 (2006) 616–623.

[68] M.J. Winter, N. Day, R.A. Hayes, E.W. Taylor, P.J. Butler, J.K. Chipman, DNA strandbreaks and adducts determined in feral and caged chub (Leuciscus cephalus)exposed to rivers exhibiting variable water quality around Birmingham, UK,Mutat. Res. 552 (2004) 163–175.

[69] N.G. Lemos, A.L. Dias, A.T. Silva-Souza, M.S. Mantovani, Evaluation of environ-mental waters using the comet assay in Tilapia rendalli, Environ. Toxicol.Pharmacol. 19 (2005) 197–201.

[70] K.M. Whitehead, J.L. Kuivila, S. Orlando, S.L. Kotelevtsev, Anderson, Genotoxicityin native fish associated with agricultural runoff events, Environ. Toxicol. Chem.23 (2004) 2868–2877.

[71] B. Inzunza, R. Orrego, M. Penalosa, J.F. Gavilan, R. Barra, Analysis of CYP4501A1,PAHs metabolites in bile, and genotoxic damage in Oncorhynchus mykiss exposedto Biobio River sediments, Central Chile, Ecotoxicol. Environ. Saf. 65 (2006) 242–251.

[72] C. Sullivan, C.L. Mitchelmore, R.C. Hale, P.A. Van Veld, Induction of CYP1A andDNA damage in the fathead minnow (Pimephales promelas) following exposureto biosolids, Sci. Total Environ. 384 (2007) 221–228.

[73] P. Rajaguru, R. Kalpana, A. Hema, S. Suba, B. Baskarasethupathi, P.A. Kumar, K.Kalaiselvi, Genotoxicity of some sulfur dyes on tadpoles (Ranahexadactyla) measured using the comet assay, Environ. Mol. Mutagen. 38(2001) 316–322.

[74] F. Mouchet, L. Gauthier, C. Mailhes, M.J. Jourdain, V. Ferrier, G. Triffault, A.Devaux, Biomonitoring of the genotoxic potential of aqueous extracts of soilsand bottom ash resulting from municipal solid waste incineration, using thecomet and micronucleus tests on amphibian (Xenopus laevis) larvae andbacterial assays (Mutatox and Ames tests), Sci. Total Environ. 355 (2006)232–246.

[75] F. Mouchet, L. Gauthier, C. Mailhes, V. Ferrier, A. Devaux, Comparative evaluationof genotoxicity of captan in amphibian larvae (Xenopus laevis and Pleurodeleswaltl) using the comet assay and the micronucleus test, Environ. Toxicol. 21(2006) 264–277.

[76] F. Mouchet, L. Gauthier, M. Baudrimont, P. Gonzalez, C. Mailhes, V. Ferrier, A.Devaux, Comparative evaluation of the toxicity and genotoxicity ofcadmium in amphibian larvae (Xenopus laevis and Pleurodeles waltl) usingthe comet assay and the micronucleus test, Environ. Toxicol. 22 (2007)422–435.

[77] D. Huang, Y. Zhang, Y. Wang, Z. Xie, W. Ji, Assessment of the genotoxicity in toadBufo raddei exposed to petrochemical contaminants in Lanzhou Region, China,Mutat. Res. 629 (2007) 81–88.

[78] N.P. Singh, M.T. McCoy, R.R. Tice, E.L. Schneider, A simple technique for quanti-tation of low levels of DNA damage in individual cells, Exp. Cell. Res. 175 (1988)184–191.

[79] V.J. McKelvey-Martin, M.H. Green, P. Schmezer, B.L. Pool-Zobel, M.P. De Meo, A.Collins, The single cell gel electrophoresis assay (comet assay): a Europeanreview, Mutat. Res. 288 (1993) 47–63.

[80] D.W. Fairbairn, P.L. Olive, K.L. O’Neill, The comet assay: a comprehensive review,Mutat. Res. 339 (1995) 37–59.

[81] R.R. Tice, G.H.T. Strauss, The single cell gel electrophoresis/comet assay: apotential tool for detecting radiation-induced DNA damage in humans, Stem.Cells 13 (1995) 207–214.

[82] A. Collins, M. Dusinska, M. Franklin, M. Somorovska, H. Petrovska, S. Duthie, L.Fillion, M. Panayiotidis, K. Raslova, N. Vaughan, Comet assay in human biomo-nitoring studies: reliability, validation, and applications, Environ. Mol. Mutagen.30 (1997) 139–146.

[83] E. Rojas, M.C. Lopez, M. Valverde, Single cell gel electrophoresis assay: meth-odology and applications, J. Chromatogr. B Biomed. Sci. Appl. 722 (1999) 225–254, Review.

[84] A.N. Jha, Genotoxicological studies in aquatic organisms: an overview, Mutat.Res. 552 (2004) 1–17 (Review).

[85] R.R. Tice, E. Agurell, D. Anderson, B. Burlinson, A. Hartmann, H. Kobayashi, Y.Miyamae, E. Rojas, J.C. Ryu, Y.F. Sasaki, Single cell gel/comet assay: guidelines forin vitro and in vivo genetic toxicology testing, Environ. Mol. Mutagen. 35 (2000)206–221.

[86] A. Hartmann, E. Agurell, C. Beevers, S. Brendler-Schwaab, B. Burlinson, P. Clay, A.Collins, A. Smith, G. Speit, V. Thybaud, R.R. Tice, Recommendations for conduct-ing the in vivo alkaline Comet assay, in: 4th International Comet Assay Work-shop, Mutagenesis, vol. 18, 2003, 45–51.

[87] B. Burlinson, R.R. Tice, G. Speit, E. Agurell, S.Y. Brendler-Schwaab, A.R. Collins, P.Escobar, M. Honma, T.S. Kumaravel, M. Nakajima, Y.F. Sasaki, V. Thybaud, Y. Uno,M. Vasquez, Andreas Hartmann, Fourth International Workgroup on Genotoxi-city testing: results of the in vivo Comet assay workgroup, Mutat. Res. 627 (2007)31–35.

[88] N.P. Singh, D.B. Danner, R.R. Tice, M.T. McCoy, G.D. Collins, E.L. Schneider,Abundant alkali-sensitive sites in DNA of human and mouse sperm, Exp. Cell.Res. 184 (1989) 461–470.

[89] G. Frenzilli, V. Scarcelli, F. Taddei, M. Nigro, Adaption of SCGE as a candidate formonitoring marine ecosystems, Neoplasma 46 (1999) 6–7.

[90] M. Moretti, M. Villarini, G. Scassellati-Sforzolini, A.M. Santroni, D. Fedeli, G.Falcioni, Extent of DNA damage in density-separated trout erythrocytes assessedby the ‘comet’ assay, Mutat. Res. 397 (1998) 353–360.

G. Frenzilli et al. / Mutation Research 681 (2009) 80–9292

[91] C.M. Hughes, S.E. Lewis, V.J. McKelvey-Martin, W. Thompson, A comparison ofbaseline and induced DNA damage in human spermatozoa from fertile andinfertile men, using a modified comet assay, Mol. Hum. Reprod. 2 (1996) 613–619.

[92] V.M. de Andrade, T.R. de Freitas, J. da Silva, Comet assay using mullet(Mugil sp.) and sea catfish (Netuma sp.) erythrocytes for the detectionof genotoxic pollutants in aquatic environment, Mutat. Res. 560 (2004)57–67.

[93] G. Wirzinger, L. Weltje, J. Gercken, H. Sordyl, Genotoxic damage in field-collectedthree-spined sticklebacks (Gasterosteus aculeatus L.): a suitable biomonitoringtool? Mutat. Res. 628 (2007) 19–30.

[94] P.L. Olive, G. Frazer, J.P. Banath, Radiation-induced apoptosis measured in TK6human B lymphoblast cells using the comet assay, Radiat. Res. 136 (1993) 130–136.

[95] R.C. Wilkins, B.C. Kutzner, M. Truong, J. Sanchez-Dardon, J.R. McLean, Analysis ofradiation-induced apoptosis in human lymphocytes: flow cytometry usingAnnexin V and propidium iodide versus the neutral comet assay, Cytometry48 (2002) 14–19.

[96] M.S. Rundell, E.D. Wagner, M.J. Plewa, The comet assay: genotoxic damage ornuclear fragmentation? Environ. Mol. Mutagen. 42 (2003) 61–67.

[97] S.A. Steinert, Contribution of apoptosis to observed DNA damage in mussel cells,Mar. Environ. Res. 42 (1996) 253–259.

[98] P.L. Olive, J.P. Banath, The comet assay: a method to measure DNA damage inindividual cells, Nat. Protoc. 1 (2006) 23–29.

[99] M. Vasquez, R.R. Tice, Detecting genotoxic activity against high molecularweight DNA using the alkaline single cell gel (SCG) assay, Environ. Mol. Mutagen29 (S28) (1997) 53.

[100] N.P. Singh, A simple method for accurate estimation of apoptotic cells, Exp. Cell.Res. 256 (2000) 328–337.

[101] N.P. Singh, Microgels for estimation of DNA strand breaks, DNA protein cross-links and apoptosis, Mutat. Res. 455 (2000) 111–127.

[102] S. Roser, B.L. Pool-Zobel, G. Rechkemmer, Contribution of apoptosis to responsesin the comet assay, Mutat. Res. 497 (2001) 169–175.

[103] N.P. Singh, Apoptosis assessment by the DNA diffusion assay, Methods Mol. Med.111 (2005) 55–67.

[104] M. Nigro, G. Frenzilli, V. Scarcelli, S. Gorbi, F. Regoli, Induction of DNA strandbreakage and apoptosis in the eel Anguilla anguilla, Mar. Environ. Res. 54 (2002)517–520.

[105] L. Hagmar, U. Stromberg, S. Bonassi, I.L. Hansteen, L.E. Knudsen, C. Lindholm, H.Norppa, Impact of types of lymphocyte chromosomal aberrations on humancancer risk: results from Nordic and Italian cohorts, Cancer Res. 64 (2004) 2258–2263.

[106] S. Bonassi, A. Znaor, M. Ceppi, C. Lando, W.P. Chang, N. Holland, M. Kirsch-Volders, E. Zeiger, S. Ban, R. Barale, M.P. Bigatti, C. Bolognesi, A. Cebulska-Wasilewska, E. Fabianova, A. Fucic, L. Hagmar, G. Joksic, A. Martelli, L. Migliore,E. Mirkova, M.R. Scarfi, A. Zijno, H. Norppa, M. Fenech, An increased micro-nucleus frequency in peripheral blood lymphocytes predicts the risk of cancer inhumans, Carcinogenesis 28 (2007) 625–631.