Caballero B. (ed.) Encyclopaedia of Food Science, Food Technology and Nutrition. Ten-Volume Set

Подождите немного. Документ загружается.

mutations in oncogenes and tumor suppressor genes

have been reported in various types of HCA-induced

tumors. Recent evidence in animal models suggests

that spontaneously or artificially induced inherited

mutations in cancer-related genes and certain dietary

factors increase the susceptibility to cancer induction

by HCAs.

Nongenotoxic Carcinogens and Epigenetic

Mechanisms

0031 The diversity of epigenetic mechanisms that may con-

tribute to cancer causation by nongenotoxic food-

borne carcinogens can be illustrated with various

pesticides, and the mycotoxin, fumonisin B

,as

examples (Table 3). The carcinogenic pesticides can

be found amongst each category of pesticide, that is,

herbicides, insecticides, or fungicides, and belong to

several chemical classes such as triazines, organo-

phosphates, and organochlorines. Animal studies

indicate that some pesticides are classic genotoxic

agents that are metabolically activated to react with

DNA and cause mutations or other DNA alterations,

as shown above for other groups of foodborne

carcinogens. Other pesticides are largely nongeno-

toxic and function as tumor promoters through the

differing mechanisms described below. With more

available evidence, it is being realized that, although

these diverse epigenetic mechanisms may be impli-

cated only during the promotion or progression

stages, this does not necessarily preclude their role(s)

during the initiation stage or throughout the carcino-

genesis process as well.

0032 Enhanced cellular proliferation and inhibition of

apoptosis Some substances have been classified as

tumor promoters because they greatly increase

cell proliferation in one or more tissues without

necessarily exerting any other effects at a significant

rate. Studies in rodent liver carcinogenesis models

have shown that cell proliferation in response to a

chemical carcinogen may be classified as cytotoxic or

mitogenic. Cytotoxic carcinogens induce cell death,

which is then followed by cell proliferation to replace

lost cells. A continued cell proliferation can result

in an increase in levels of mutations through DNA

replication errors. Mitogenic chemicals in the rodent

liver have been observed to produce a sustained

increase in organ weight during exposure to the

mitogen. However, cell proliferation rates return to

normal once the organ has reached its new size. Mito-

genic agents provide a selective growth advantage to

initiated cells by decreasing apoptosis, inhibiting

normal liver cell proliferation or increasing cell pro-

liferation in precancerous cells. Hence, depending on

the type of carcinogen and the cell population in-

volved, cell proliferation may affect the initiation of

a tumor, whereas proliferative changes in initiated cell

populations may influence tumor development and

progression.

0033Similarly, in recent years, it has become known that

programmed cell death or apoptosis contributes

during multiple stages of cancer development. The

number of DNA-damaged cells may be increased by

mitosis or lost through apoptosis. Thus, chemicals

that decrease the rates of apoptosis may favor initi-

ation, in addition to tumor promotion and progres-

sion, or vice versa. For example, agents such as

peroxisome proliferators, which induce a transient

mitogenic activity, also inhibit apoptosis. Tumor pro-

moters such as 2,3,7,8-tetrachloro-dibenzo-p-dioxin

may promote initiated cell growth by inhibiting apop-

tosis in the initiated cell populations. Current data

suggest that responses of cells to proliferation and

apoptosis signals may change throughout cancer

NCH

dG-C8-MelQx

dG-N

-MelQx

fig0006 Figure 6 Chemical structures of IQ and MeIQx-DNA adducts formed at the C8 and N

atoms of guanine.

916 CARCINOGENS/Carcinogenic Substances in Food: Mechanisms

development and that, at each stage, a delicate bal-

ance between cell proliferation and apoptosis is ne-

cessary to prevent cancer induction.

0034 Oxidative DNA damage Recent evidence implicates

oxidative DNA damage as having similar conse-

quences for the genome as electrophiles generated

through metabolic activation. Fat metabolism, for

example, produces a number of reactive oxygen

species of which the hydroperoxide radicals (RO

and ROO

) are stable and reactive enough to interact

with DNA. About 20 different DNA lesions caused

by these radicals have been described, including

single- and double DNA strand breaks, and the for-

mation of mono- or dihydroxylated derivatives of the

DNA bases. Of the adducts, the most studied to date

is the 8-hydroxy-2

-deoxyguanosine, which can lead

to base mispairing at the same site or at an adjacent

base in the DNA. Nitrogen radical-induced damage

has also been reported but less frequently than

oxygen radical damage. Even though carcinogens

producing oxidative DNA damage are similar to

the electrophile-generating substances in causing

mutations, these chemicals cannot be identified as

genotoxic by the current battery of genotoxicity

tests, because they do not contain the necessary

enzyme functions necessary to permit their detection.

Among the foodborne carcinogens, the peroxisome

proliferator class of pesticides (chlorophenols and

di(2-ethylhexyl)phthalate) and the mycotoxin fumo-

nisin B

are likely to generate lipid peroxidation

and oxygen radicals, and therefore may be expected

to operate through the oxidative DNA damage

pathway. Much work is required before this can be

confirmed.

0035 Gap junctional intracellular communication Some

tumor-promoting organochlorines perhaps operate

through mechanisms such as inhibition of gap

junctional intercellular communication (GJIC). Gap

junctions are clusters of intercellular aqueous chan-

nels that allow communications between adjoining

cells. They are formed by two semichannels, called

connexons, each coming from two adjacent cells.

Each connexon consists of six identical or similar

transmembrane protein molecules called connexins,

which are oligomerized within a plasma membrane.

Gap junctions are the only means by which cells in

organized tissue can share different messengers of

low molecular weight and therefore maintain tissue

homeostasis. The nature of these messengers can

range from ions, nutrients, nucleotides and metabol-

ites, either of endogenous or exogenous origin.

During cancer induction, it has been found that cer-

tain tumor cells show high levels of communication

with each other but not with surrounding normal

cells, thus facilitating rapid selective clonal expansion

and malignant transformation of abnormal cells. Two

basic mechanisms of altered connexin functions may

contribute to carcinogenesis at the genetic and func-

tional levels. Genetic alteration of GJIC may occur at

either initiation stage when genotoxic carcinogens

induce mutations in connexin or other GJIC-related

genes, or at later stages of tumor progression as a

result of cumulative genetic alterations due to wide-

spread genetic instability. Findings in cell cultures

suggest that GJIC can affect cell-cycle-related gene

expression supporting the earliest hypothesis that

the role of GJIC may be to provide a channel for

distribution of an intracellular factor to control cell

growth.

0036Hormonal mimicry Other pesticides act by mimick-

ing hormones, leading to disruption in endocrine

responsive organs such as mammary gland, testis,

prostrate, and uterus. For example, the nongenotoxic

herbicide atrazine is associated with an increased in-

cidence of mammary tumors and uterine adenocarci-

nomas in animals This is probably related to

atrazine’s ability to lengthen the estrous cycle, thus

prolonging exposure to endogenous estrogens.

Several organochlorines, including chlordecone and

DDT, also have demonstrated estrogenic activity. A

receptor-mediated mechanism has been suggested

to explain the role of hormonal effects in cancer

induction. Estrogens stimulate DNA replication and

increase the mitotic rate of target cells. This is

achieved through interaction of estrogens or estro-

gen-mimetic substances with the estrogen receptor

protein, which may then react with the target cell

DNA to modify the expression of genes involved in

cell growth and differentiation. Sustained occupation

of the estrogen receptors by estrogenic compounds is

necessary for alteration of the genome leading to

enhanced cell proliferation. Progestins oppose this

effect by decreasing the concentration of the estrogen

receptors. There is evidence that estrogenic sub-

stances may further generate oxidative DNA damage

and lipid peroxidation.

0037Peroxisome proliferation Peroxisome proliferation

is another nongenotoxic, receptor-mediated mechan-

ism whereby some pesticides (e.g., chlorophenols

and phenoxyacetic acid herbicides) and a packaging

material residue (e.g., di(2-ethylhexyl)phthalate)

have been shown to induce cancer in rodents. Peroxi-

somes are subcellular organelles responsible for fat

metabolism in various tissues. For the peroxisome

proliferator class of compounds, the primary target

organ is liver where they cause proliferation of

CARCINOGENS/Carcinogenic Substances in Food: Mechanisms 917

peroxisomes, induce hypertrophy and hyperplasia of

liver, stimulate growth factors, activate oncogenes,

and increase the activity of the enzymes catalase,

cytochrome P450 monooxygenases, epoxide hydrase,

glucuronyl transferase, and those involved in fatty

acid b-oxidation. While the precise mechanism of

cancer induction by these chemicals is unknown, the

pleitropic response appears to be related to activation

of a novel steroid hormone receptor, the peroxisome

proliferator-activated receptor (PPAR). In addition,

two more theories of cancer induction have been

proposed. First, induction of fatty acid b-oxidation

enzymes by the peroxisome proliferators results in the

generation of hydrogen peroxide and reactive oxygen

species that interact with DNA to form an oxidative

DNA damage-related adduct. In support of this

theory, increases in levels of 8-hydroxy-2

-deoxygua-

nosine have been detected in peroxisome proliferator

treated rats. The second theory relates to an increase

in liver size and weight that follows peroxisome pro-

liferator administration. This effect has been demon-

strated to be due to both enhanced cell replication

and a decrease in apoptosis. The possibility then

exists that the process of DNA replication itself

causes mutations. However, peroxisome proliferators

may promote the growth of cells that are already

initiated either spontaneously or by other environ-

mental stimuli. Operationally, it is likely that some

or all of these mechanisms play a role in cancer in-

duction by peroxisome proliferators. The role of

peroxisome proliferators in human cancer is not

clear, since peroxisome proliferation does not occur

in humans, although they respond to these agents in

hyperlipidemia, for example. Preliminary data sug-

gest that PPARs are expressed at low, variable levels

in human liver and may also be polymorphic.

0038 Disruption of the sphingolipid pathway The myco-

toxin, fumonisin B

, unlike aflatoxin B

and ochra-

toxin A, does not undergo extensive metabolism. In

orally treated rats, the majority of fumonisin B

excreted in the feces in unmetabolized form. In female

vervet monkeys, products of hydrolysis of the tricar-

ballylic acid groups of fumonisin B

have been ob-

served in feces, suggesting that the only metabolism

of fumonisin B

is by ester hydrolysis to the less polar

aminopentol. Fumonisin B

persists in livers and

kidneys of rats longer than any other tissues, and this

is consistent with the observed toxicity and carcino-

genic effects of this mycotoxin in these organs.

Although fumonisin B

is a liver carcinogen, it lacks

mutagenic activity in the Ames assays, and genotoxic

effects in in vivo and in vitro DNA repair assays. This

mycotoxin is a weak initiator, as demonstrated in

in vivo initiation-promotion assays, but has effects

on tumor promotion and selection of initiated

cells. Lipid peroxidation and oxidative damage may

be involved in the genesis of fumonisin B

-induced

initiated cells. While the molecular mechanism of

action of fumonisins is not completely known, the

chemical structural similarity of these compounds to

sphingosine and sphinganine (Figure 7) has suggested

the following mechanism. Sphingolipids are believed

to be involved in the regulation of cell growth, differ-

entiation, and cell transformation through effects on

cell–cell communication, cell-membrane interactions,

cell receptors, and signaling systems (Figure 8).

0039It has been postulated that fumonisins may exert

their effects by disrupting sphingosine metabolism,

and current data support this hypothesis. The main

target for fumonisins is the inhibition of the enzyme

sphinganine (sphingosine) N-acyltransferase (or cer-

amide synthase), resulting in the accumulation of

free sphingoid bases in serum, and in the target

tissues, which in the rat are the liver and kidney.

The exact mechanistic role of the sphingolipid

pathway in liver-cancer induction in rats remains to

be elucidated.

0040Altered immune function Alteration of immune

function is a known cause of many human cancers.

Many pesticides and other foodborne carcinogens are

also known to disrupt immune function, which may

additionally contribute to their carcinogenic potency.

For example, the insecticide, DDT, causes thymus

atrophy, and 2,4-dichlorophenoxyacetic acid en-

hances the T- and B-cell immune response in the

mouse and reduces lymphocyte subsets and natural

R COCH

CH(COOH)CH

COOH

OH CH

Fumonisin B

Sphinganine

Sphingosine

fig0007Figure 7 Structures of sphinganine, sphingosine, and fumoni-

sin B

918 CARCINOGENS/Carcinogenic Substances in Food: Mechanisms

killer cells in farmers. Organophosphates may play a

role in cancer induction through inhibition of serine

esterases, which are critical components for the cyto-

lytic activities of T-lymphocytes and natural killer

cells. Ochratoxin A, N-nitrosodimethylamine, 7,12-

dimethylbenz[a]anthracene, benzo[a]pyrene and the

antioxidant butylated hydroxytoluene have also been

observed as immunosuppressants in animal studies.

Sphingomyelin:

Ceramide:

Sphingosine:

OH OH

P OCH

N(CH

)

OH OH

OH O PO

Membrane structure

interaction with cholesterol

Activation of protein kinase/phosphatase(s)

induction of apoptosis, cellular differentiation

stimulation/inhibition of cell growth

mediator of some actions of:

Alteration of diacylglycerol and cholesterol

metabolism

Inhibition of retinoblastoma protein

dephosphorylation

Activation of phospholipase C/inhibition

of phosphatidic acid phosphatase

Inhibition of cell transformation

Sphingosine 1-phosphate:

Activation of phospholipase D

stimulation of cell growth

mobilization of Ca

from intracellular

stores

Stimulation/inhibition of cell growth

and differentiation

Inhibition of protein kinase C and other

kinases (activation of some kinases)

Tumor necrosis factor α

interleukin 1β

nerve growth factor

1α, 25-dihydroxyvitamin D

fig0008 Figure 8 Examples of the products of sphingolipid hydrolysis and some of the biological activities that have been attributed to these

compounds.

CARCINOGENS/Carcinogenic Substances in Food: Mechanisms 919

Future Considerations

0041 It is clear that the majority of evidence for the types of

specific mechanisms in cancer induction is from

animal studies. Evidence to date suggests that the

most identified genotoxic human carcinogens react

with DNA through electrophile generation, but

for nongenotoxic carcinogens, the mechanisms are

unclear. Much research is required (1) to understand

the precise role(s) of the currently known mechan-

isms, and (2) to discover any new mechanistic path-

ways that may contribute to human cancer induction

by foodborne carcinogens. In addition, there is a need

to define the relationship to cancer induction by

factors that may interact with carcinogenic sub-

stances in food such as macro- and micronutrient

levels in the diet, concurrent environmental exposure

to multiple chemicals, and genetic differences in re-

sponses to these substances.

See also: Amines; Carcinogens: Carcinogenicity Tests;

Food Additives: Safety; Mutagens; Mycotoxins:

Classifications; Occurrence and Determination;

Toxicology; Nitrosamines

Further Reading

Ames BN, Gold LS and Willett WC (1995) The causes and

prevention of cancer. Proceedings of the National

Academy of Sciences of the United States of America

92: 5258–5265.

Clayson DB (2001) Toxicological Carcinogenesis. Boca

Raton, FL: Lewis Publishers.

Food and Agricultural Organisation of the United Nations

(1996) Worldwide Regulations for Mycotoxins. Rome:

FAO.

Keifer MC (1997) Human health effects of pesticides. In:

Occupational Medicine: State of Art Reviews, vol. 12.

Philadelphia, PA: Hanley & Belfus.

Kitchin KT (1999) Carcinogenicity – Testing, Predicting

and Interpreting Chemical Effects. New York: Marcel

Dekker.

Lawley PD (1990) N-nitroso compounds. In: Cooper CS

and Grover PL (eds) Chemical Carcinogenesis and

Mutagenesis I, pp. 409–470. New York: Springer Verlag.

Lockshin RA, Zakeri Z and Tilly JL (1998) When Cells

Die. A Comprehensive Evaluation of Apoptosis and

Programmed Cell Death. Chichester, UK: Wiley-Liss.

Merrill AH, Schmelz E-M, Dillehay DL et al. (1997)

Sphingolipids, the enigmatic lipid class: Biochemistry,

physiology and pathophysiology. Toxicology and

Applied Pharmacology 142: 208–225.

Nagao M and Sugimura T (2000) Food Borne Carcinogens.

Heterocyclic Amines. Current Toxicology Series.

Chichester, UK: Wiley-Liss.

Schafer KA (1998) The cell cycle: a review. Veterinary

Pathology 35: 461–478.

Scheuplein RJ (1990) Perspectives on toxicological risk – an

example: Food-borne carcinogenic risk. In: Clayson DB,

Munro IC, Shubik P and Swenberg JA (eds) Progress

in Predictive Toxicology, pp. 351–371. Amsterdam:

Elsevier.

Carcinogenicity Tests

T J Schrader, Sir Frederick Banting Research Centre,

Ottawa, Ontario, Canada

Introduction

0001Carcinogens may be present in foodstuffs as: (1) nat-

ural components; (2) byproducts of cooking, heating,

or processing; (3) food additives (preservatives,

colors, flavors); (4) leachates from packaging mater-

ials; and (5) pesticides. The testing of new foods and

food-associated chemicals for carcinogenicity is an

important aspect of an overall toxicological assess-

ment for determining product safety and allowable

exposure limits. Government regulatory decisions on

existing products can also be revisited and modified

to include new knowledge gained in the understand-

ing of carcinogenic processes and the results of new

testing methods.

0002In excess of 200 assays have been developed to

screen substances for carcinogenic potency and

to characterize their modes of action. However,

many of the processes involved in carcinogenesis

remain poorly understood and a perfect assay for the

identification of human carcinogens remains elusive.

Confounding factors include: (1) the long latency

period (frequently decades) required for metastatic

tumor development in humans; (2) the number and

types of molecular mechanisms governing the many

cellular changes involved in carcinogenesis; (3) the

mechanisms involved in tissue-specific susceptibilities,

often showing species differences; (4) variations in

genetic make-up and metabolic capabilities between

individuals in a population not reflected within inbred

strains of animals or tissue culture cell lines; (5) the

effects of different exposure routes and timing of ex-

posure; and (6) the contribution of coexposure to

other chemicals in mixtures or other agents which

could modify the carcinogenicity of a test chemical.

0003A tiered approach can be used to screen chemicals

for carcinogenic activity and obtain supportive evi-

dence. The first clues may arise from an examination

of the test chemical for structural features which are

920 CARCINOGENS/Carcinogenicity Tests

common with known carcinogens. Other evidence

may arise from epidemiological studies relating

human cancer incidence with chemical exposure. Ex-

perimental evidence may be gained from a variety

of short-term chemical and cell-based (in vitro) assays

used for screening test chemicals for DNA damage

and repair, mutagenesis, chromosomal aberrations,

and morphological transformation. However, while

these assays can be important in identifying chemicals

positive for their respective endpoints and for provid-

ing important mechanistic data, they are not repre-

sentative of the entire carcinogenic process and may

lack important endpoints or metabolic processes.

Therefore, testing may proceed to mutagenesis and

chromosomal aberration assays in animal (in vivo)

models. Ultimately, the most validated and reliable

assay for the evaluation of human carcinogens is the

long-term rodent carcinogenicity assay, in which

animals are exposed to the test chemical over a large

part of their lifespan (generally 18 months for mice

and 2 years for rats). Even so, the rodent carcinogen-

icity assay suffers some major drawbacks which

could lead to the misinterpretation of results: (1)

species differences in chemical metabolism which

could cause the chemical to act in a way not possible

in humans; (2) difficulties in determining the human

risk of chemicals which affect the formation of

tumors which otherwise occur spontaneously with

age in some rodent species; and (3) errors which are

introduced in extrapolating or estimating the risk to

humans at normal exposure limits when chemicals

are tested at far higher levels. Therefore, data from

the rodent carcinogenicity assay are usually combined

with that from some complementary shorter-term

tests to confirm the classification of a test chemical.

A number of testing procedures can also be used on

human blood and urine samples to monitor if expos-

ure to certain classes of carcinogens has occurred. Of

course, this tiered approach may also work in the

opposite direction, where a chemical found positive

in vivo is then subjected to a number of in vitro tests

to obtain information on its mechanism of action.

DNA Damage and Repair Assays

0004 Many assays have been developed to characterize the

DNA damage induced by carcinogen exposure as

well as the induction of DNA repair and changes in

cellular gene expression. With slight modifications,

most of the assays can be used for samples arising

from both in vitro and in vivo studies.

Strand Breakage Assays

0005 Comet assay The comet assay has become one of

the most convenient and widespread assays for DNA

strand breakage. Cells exposed to test agent are

embedded in agarose which is spread over a micro-

scope slide to form a thin gel. The cells are then lysed

and the DNA electrophoresed under either neutral

conditions to detect double strand breaks or alkaline

conditions to detect single and double strand breaks

as well as apurinic/apyrimidinic sites. When subse-

quently visualized by staining with a fluorescent

dye, the DNA appears as a ‘comet’ with a compact

‘head’ of larger-molecular-weight DNA and a ‘tail’ of

smaller fragments, the size of which depends upon the

amount of strand breakage present. Various imaging

and software packages are available to quantitate the

distribution of DNA in the images. The assay can be

extended to detect specific DNA adducts and specific

forms of oxidative damage by using damage-specific

DNA endonucleases to cleave the DNA at the damage

sites, thus creating strand breaks.

0006Other assays for strand breakage include: alkaline

gel electrophoresis in which fragmented genomic

DNA separates as a smear instead of the discrete

band of high-molecular-weight material observed

with intact DNA; alkaline elution, in which the time

taken for alkaline solutions of DNA to pass through

filters is related to the extent of fragmentation; and

alkaline unwinding techniques, in which the fluores-

cent dye ethidium bromide is used to intercalate into

the DNA, causing it to unwind. The unwinding in

turn causes a decrease in fluorescence intensity which

can be related to DNA fragmentation. As well, the

induction of strand breakage in compact supercoiled

plasmids causes a conformational change to an open

circular form, which can be detected as a molecule

with slower mobility by agarose gel electrophoresis.

Adduct Assays

0007

P-Postlabeling The

P-postlabeling method is

extremely sensitive for detecting chemical adducts

and is theoretically sensitive to any DNA-damaging

agent. Isolated DNA is first broken down into normal

and damaged 3

monophosphate deoxynucleosides.

radioactive phosphate residue is then added

by the enzyme T4 polynucleotide kinase to allow

detection of the deoxynucleosides. The radioactive

products are separated by multidirectional anion

exchange thin-layer chromatography and the

presence of adducts detected and quantitated.

0008Chemical adducts in DNA can also be detected by a

variety of other assays. If the chemical is available as

a radioisotope, the presence of adducts may be

detected as radioactivity copurifying with DNA

from treated cells. The adducts may then be isolated

and characterized using the radioactive signal as a

tracer. As well, particular types of DNA damage can

be recognized by damage-specific antibodies which

CARCINOGENS/Carcinogenicity Tests 921

can be used to quantitate the amount of damage or

isolate the damaged sites for further study. Some

adducts can be identified by their altered fluorescent

properties or mobilities when separated by high-

pressure liquid chromatography or gas chromatog-

raphy. Further chemical identification may also be

made by mass spectrometry.

Unscheduled DNA Synthesis or Repair Replication

0009 Unscheduled DNA synthesis refers to the DNA syn-

thesis which occurs in mammalian cells during the

repair of DNA damage by the excision repair path-

way. During excision repair, damage is removed in the

form of an oligonucleotide which must be replaced by

a ‘repair patch’ synthesized using the opposite un-

damaged strand as a template. This form of repair

can be followed in treated cell cultures by including a

labeled nucleotide precursor, such as tritiated thymi-

dine or bromodeoxyuridine, in the culture medium.

Tritiated thymidine incorporation can be followed by

autoradiography of the labeled nuclei or by liquid

scintillation counting while bromodeoxyuridine in-

corporation can be quantitated by immunostaining

with antibromodeoxyuridine antibodies.

0010 The repair of DNA damage can be followed in

specific genes and comparisons made in transcribed

and nontranscribed strands using a variety of molecu-

lar biological techniques. Damage-specific endonu-

cleases can be used to cleave the site of damage

which can be detected by its increased mobility

during gel electrophoresis and different hybridization

patterns. Other techniques including strand-specific

quantitative polymerase chain reaction (PCR) in

which the amount of product is proportional to the

undamaged template and strand-specific ligation

PCR which allows sites of damage to be identified

with DNA sequencing gels.

Gene Expression Assays

0011 Assays which measure carcinogen-induced changes in

gene expression are becoming an increasingly import-

ant method for characterizing test chemical effects

and a variety of assays have been developed. DNA

arrays are available which make the analysis of gene

expression changes in hundreds or thousands of

genes possible by using gene-specific complementary

oligonucleotides arranged on either nylon membranes

or glass slides. Several reporter gene vectors have been

developed in which the expression of a variety of

genes coding for easily measured products (such as

firefly luciferase, or green fluorescent protein) is

controlled by the promoter sequences of genes

important for the control of carcinogenesis. A

chemical-induced change in the expression of the

reporter gene then indicates possible carcinogenic

activity. Nuclease protection assays can also be used

to quantitate changes in the expression of specific

genes involved in carcinogenesis.

Mutagenicity Assays

In vivo

Mutagenesis Assays

0012The mouse specific locus test The mouse specific

locus test is used to detect and quantitate mutations

in the germ line of a mammalian species. Animals

are chosen for mating which differ at certain easily

assayed genetic loci such that mutations will

produce an offspring which differs from that

expected. Males are usually exposed to the test chem-

ical. Three variations of the method have been used

for detecting the induction of point mutations in

mouse germ cells: (1) a visible specific locus test

examining either five or seven loci; (2) a biochemical

specific locus test surveying up to 20 enzymes; and (3)

a test for histocompatibility loci mutations.

0013Transgenic mouse mutagenesis assays Transgenic

mouse strains have been developed which have had

several copies of either LacI or LacZ gene sequences

from Escherichia coli incorporated into their

genomes, either as part of a shuttle vector or recom-

binant l-phage backbone. Animals are exposed to

the test chemical and the DNA isolated from tissues.

For shuttle vector-based systems, the shuttle vector

DNA is excised, recovered, religated, and introduced

into the proper E. coli host which are then plated

on selective medium allowing growth of mutant

colonies. Reconstitution of the l-based systems is

somewhat simpler and more efficient, where the l-

DNA can be cleaved and repackaged into phage using

commercially available extracts. The phage are used

to infect the proper E. coli host, which are then plated

in the presence of a selective agent. In a positive

selection version of the LacZ system, only mutant

plaques will grow while mutant plaques in the LacI

system are identified by a blue/colorless selection

procedure. Mutation frequencies can be examined in

all somatic tissues as well as germline tissues using

these transgenic systems and the defined sequences

provide a well-characterized target for sequence

analysis of the mutations.

Assays in Bacteria

0014Ames Salmonella test The Ames Salmonella test

remains the most validated mutagenesis assay for

reasons of speed, economy, and dependability. A

series of Salmonella strains have been developed

which possess different mutations in the operon

922 CARCINOGENS/Carcinogenicity Tests

responsible for histidine synthesis, making it impos-

sible for the bacteria to grow in the absence of histi-

dine. The Salmonella test is a reverse mutation assay

in which back-mutation of the his operon will allow

the bacteria to grow in the absence of histidine. Bac-

teria are exposed to the test chemical with and with-

out a metabolic activation system (usually rat liver

S9, a cofactor-supplemented postmitochondrial prep-

aration of rat liver microsomes prepared from rats

exposed to Aroclor 1254, a PCB (polychlorinated

biphenyl) inducer of the cytochrome P450 hydroxyla-

tion enzymes) and spread on to minimal medium agar

plates. After a suitable period of incubation, revertant

colonies are counted and compared to those found on

control plates containing either vehicle alone or a

known mutagen. Several strains with different sensi-

tivities are available with TA98 and TA100 detecting

frameshift and basepair mutations, respectively, while

strains TA102 and TA104 respond to oxidative

damage. Other frameshift-specific strains such as

TA1535 and TA1537 are also available. Recently,

related strains have been developed which are each

reverted by specific basepair changes, allowing a

mutational spectrum of test chemical activity to be

generated. Specialized strains have also been created

which express various metabolic activation enzymes

to examine specific classes of chemicals. A lumines-

cent version of the assay is also commercially avail-

able which substitutes luminescence in place of

colony formation as a measure of revertant cell

growth.

0015 E. coli tryptophan (trp) reversion assay The E. coli

trp reversion assay using E. coli WP2 and related

strains is a microbial assay which measures trp



trp

reversion induced by mutagenic chemicals.

Various strains are available which are deficient in

different aspects of DNA repair, making them more

susceptible to reversion. As well, both frameshift and

basepair mutations can be detected using different

strains. The assays are carried out in a similar manner

to the Ames Salmonella test.

Assays in Somatic Cells in Culture

0016 Mammalian cell culture systems may also be used to

detect mutations induced by chemical substances.

Widely used cell lines include L5178Y mouse lymph-

oma cells as well as the Chinese hamster ovary (CHO)

and V-79 lines of Chinese hamster cells. The most

commonly used systems measure mutation at the thy-

midine kinase (TK, L5178Y cells), hypoxanthine-

guanine-phosphoribosyl transferase (HGPRT, CHO,

and V-79 cells) and Na

ATPase (V-79) genetic

loci. The TK and HGPRT mutational systems detect

basepair mutations, frameshift mutations, and small

deletions, while the Na

ATPase assay detects

only basepair mutations.

0017Cells are exposed to the test agent with and with-

out metabolic activation for a suitable period of time.

They are then subcultured to determine cytotoxicity

and to allow phenotypic expression prior to mutant

selection. The cultures are analyzed for mutant fre-

quency at the end of the expression time by seeding

known numbers of cells in medium with and without

a selective agent which is lethal to the parental cells

but ineffective against the mutant cells. After a suit-

able incubation time, cell colonies are counted. The

number of mutant colonies in selection medium is

adjusted by the number of colonies in normal medium

to derive the mutant frequency. Cell lines are now

available which express a variety of metabolic

enzyme activities introduced by genetic engineering.

These cell lines can be used to probe the effects of

various metabolic activities on carcinogen activation

and to avoid the need for adding an exogenous meta-

bolic activation system.

Fungal Mutagenesis Assays

0018A number of fungal systems have been developed for

mutagenicity assays, including Aspergillus nidulans

and Neurospora crassa, which can be used to detect

both forward and reverse gene mutations by nutri-

tional, biochemical, or morphological changes in the

treated population. Similar methods have been de-

veloped using the yeast Saccharomyces cerevisiae to

measure forward and reverse mutations as well as

mitotic gene conversion, the nonreciprocal exchange

of DNA between sister chromatids.

Characterization of Mutations

0019Mutations in specific DNA sequences can be

detected by heteroduplex analysis. Heteroduplexes

are formed when wild-type and mutated sequences

are annealed to form double-stranded mismatched

structures. Heteroduplexes can be distinguished

from perfectly matched homoduplexes by techniques

such as denaturing gradient gel electrophoresis,

single-stranded conformation polymorphism, con-

stant denaturant gel electrophoresis, chemical mis-

match cleavage, Rnase A cleavage, and cleavage

with bacteriophage resolvases. The ligase chain reac-

tion is one method which allows the detection of

specific DNA sequences and point mutations. Primers

to specific wild-type and mutated gene sequences are

designed such that they anneal side by side when the

complementary sequence is present. The primers are

then ligated together and subsequent cycles of

annealing and ligating amplify the amount of product

produced, which can then be detected by several

means.

CARCINOGENS/Carcinogenicity Tests 923

Chromosomal Analysis

Detection of Chromosomal Aberrations in Bone

Marrow

0020 Assays for chromosomal aberrations can detect

changes in either the structure or number of chromo-

somes. However, in practice, cells are analyzed at the

first mitosis after treatment, and therefore number

changes are not usually found. Chromosome prepar-

ations are made from the bone marrow cells of ex-

posed animals, stained and examined under a

microscope for metaphase cells which are scored for

chromosomal aberrations. A more recent develop-

ment is fluorescent in situ hybridization (FISH) in

which chromosomal aberrations are detected with

fluorescent-labeled DNA probes which hybridize to

specific chromosomal regions.

Micronucleus Assay

0021 The micronucleus test is a mammalian in vivo and

in vitro test which detects chromosomal or mitotic

apparatus damage. Micronuclei are small cytoplas-

mic particles consisting of acentric fragments of

chromosomes or entire chromosomes, which become

separated from the nuclei of daughter cells. Polychro-

matic erythrocytes in the bone marrow of rodents

form a useful model for this assay since the nuclei

are lost during development from erythroblasts. The

lack of a nucleus makes the scoring of micronuclei,

either by microscope or flow cytometry, easier in

these cells.

Rodent Dominant Lethal Assay

0022 A dominant lethal mutation is one which occurs in a

germ cell and results in the death of the fertilized egg

or developing embryo. Dominant lethals are generally

accepted to be the result of both structural and

numerical chromosomal aberrations. In this assay,

male animals are usually exposed to the test substance

and mated to untreated females. The females are

sacrificed after an appropriate period of time and

the numbers of live and dead embryos in the uteri

determined. The ratio of dead to live embryos from

the treated group compared to the ratio of dead to live

embryos from the control group is used as a measure

of dominant lethality. Several treatment protocols are

available. The most widely used require either single

administration of the test substance or treatment on

five consecutive days.

The Rodent Heritable Translocation Assay

0023 In this assay, males are treated with the test chemical,

mated, and the male progeny screened either for

chromosomal translocations, or for sterility, which

is indicative of translocations between nonhomolo-

gous chromosomes in the treated male gametes.

Sister Chromatid Exchange Assay

0024The sister chromatid exchange (SCE) assay detects

the ability of a chemical to enhance the exchange of

DNA between two sister chromatids of a duplicating

chromosome. The test may be performed in vitro,

using continuous cell lines, or in vivo using

animal models such as mice, rats, and hamsters. For

in vitro assays, cell cultures are exposed to test chem-

icals, and allowed to replicate in the presence of

bromodeoxyuridine. The cells are then treated

with colchicine or colcemid to arrest cells in a meta-

phase-like stage of mitosis, and following harvesting

chromosome preparations made. Preparations are

stained and metaphase cells analyzed for SCEs using

a microscope.

Transformation Assays

0025Both in vivo and in vitro assay systems have been

developed which assay chemicals for their abilities

to induce morphological transformation, changing

the morphology of cells from normal or near-normal

monolayer growth-inhibited cultures to multilayered

altered foci resembling neoplastic cell growth.

Depending upon the assay, test chemicals may be

tested for activity as complete carcinogens, tumor

initiators, or tumor promoters.

Rodent Skin

in vivo

Assay

0026When chemicals are tested for activity as complete

carcinogens in the rodent skin transformation assay,

they are applied repetitively over several weeks to a

shaved area on the backs of the animals, and the area

observed for tumor formation. Tumor initiators can

be detected by a single application followed by repeti-

tive application of a known tumor promoter such as

benzoyl peroxide or chrysarobin. Similarly, tumor

promoters can be assayed by first applying a known

initiator such as 3-methylcholanthrene to the shaved

area followed by repeated applications of the test

chemical.

0027In vitro methods for assaying cellular transform-

ation include: the Syrian hamster embryo (SHE) cell

transformation assay, using primary cultures derived

from 2-week-old SHEs, epithelial cells, the BALB/

c-3T3 and C3H-10T½ cell lines and a number of

cells which have been infected by a transforming

virus, such as Rauscher leukemia virus-infected Fisher

rat embryo cells or simian adenovirus SA7-infected

SHE cells. While the SHE cells maintain may of

their metabolic capabilities, many of the other cell

types require the addition of a metabolic activation

924 CARCINOGENS/Carcinogenicity Tests

system such as primary hepatocytes and microsomal

fractions. Many of these systems can also be used for

identifying tumor promoters by first treating the cells

with a low concentration of known initiator.

Gap Junctional Communication

0028 Adjacent cells form contact points called gap junc-

tions which allow the exchange of small molecules in

a form of communication. The exposure of cells to

carcinogens causes a blockage of gap junctional com-

munication. Presumably the loss of communication

with neighbors alters cells in such a way that they act

in isolation and ignore neighboring signals to limit

growth, a process thought to be involved in the

development of carcinogenesis. The activity of gap

junctions can be measured by several systems.

Metabolic Cooperation

0029 Wild-type and 6-thioguanine-resistant V79 Chinese

hamster cells are cultured together in the presence of

6-thioguanine. With active gap junctions, the lethal

metabolites formed from 6-thioguanine in the wild-

type cells can disperse into the otherwise resistant

cells, causing cell death. If gap-junctional communi-

cation is blocked, treated resistant cells survive to

form colonies, which can be enumerated.

Dye Transfer

0030 The fluorescent dye lucifer yellow, a fluorescent dye,

is introduced into cells either by preloading or

through a scratch in the monolayer. In the presence

of intercellular communication, the dye spreads to

neighboring cells in the culture and the fluorescent

area increases. When chemical treatment causes gap

junctions to be blocked, fluorescence remains concen-

trated at the original site.

Gene Expression

0031 The blockage of intracellular communication can be

reflected by the altered expression of mRNA coding

for gap junction components, such as connexin 43,

which is involved in gap junction formation.

Peroxisome Proliferation

0032 Peroxisomes are cytoplasmic organelles responsible

for the oxidative degradation of fatty acids and

other molecules. It has been observed that several

nonmutagenic rodent liver carcinogens, including

some lipid-lowering drugs and plasticizers, cause

liver enlargement and the proliferation of numbers

of peroxisomes within hepatocytes. The enumera-

tion of peroxisomes in liver tissue is sensitive for

several nongenotoxic rodent hepatocarcinogens but

peroxisome proliferation does not appear to occur in

humans and its relevance is not known.

In vivo

Assays for Tumor Formation

0033A number of animal models have been used to assess

test chemical carcinogenicity. However, considerations

of time, cost, and ease of handling have favored the use

of rodents instead of other animals such as dogs or

primates since assays require animals to be exposed to

the test agent over a significant portion of their lifespan.

Although animal models provide the most complete

and dependable surrogates for the identification of

human carcinogens, they are not entirely representative

of the human situation due to a number of factors,

including species differences in metabolism and suscep-

tible tissues, differences in time and route of exposure

as well as amount of chemical tested, and lack of ex-

posure to other chemicals which may act as cocarcino-

gens. The interpretation of results obtained when

chemicals are tested in rodents at the maximum toler-

ated dose (MTD, the largest dose possible which does

not shorten the lifespan or produce clinical signs of

toxicity) may be difficult to relate to human exposures.

Many chemicals identified as rodent carcinogens have

been found to cause inflammation and oxidative

damage leading to tissue death and regeneration only

at these elevated levels far removed from any expected

human exposure.

Rodent Carcinogenicity Assay

0034The rodent carcinogenicity assay remains the most

reliable and validated system for the identification

of possible human carcinogens. With minor vari-

ations, the methods used for the rodent carcinogen-

icity assay have been standardized both nationally

and internationally, and by law all new chemicals

introduced for human use must be tested for long-

term effects using the assay. In order to address pos-

sible species differences in carcinogenic response, a

chemical is tested using two species, usually rats and

mice, for comparison, although hamsters may

sometimes be substituted. Animals of both sexes are

treated with increasing doses of chemical, which

are included in the diet or given orally by gavage for

food-related studies. Each dose group may contain 50

or more animals and the dosing is maintained over

most of the animal’s remaining lifespan (18 months

for mice and 2 years for rats), after which tissues are

collected at necropsy and examined by a pathologist

for malignant tumors and preneoplastic lesions.

0035The duration of the study and number of animals

required make the rodent carcinogenicity assay an

extremely expensive and time-consuming undertak-

ing. Several other in vivo assays have been developed

CARCINOGENS/Carcinogenicity Tests 925