toxicology (2) - ullmann

8/14/2019 Toxicology (2) - Ullmann

1/42

2 01 2 Wi le y- VC H V er la g G mb H & C o. K Ga A, W ei nh ei m

Toxicology, 2. Assessment Methods

WOLFGANG DEKANT, Institute of Toxicology, University of Wuerzburg, Germany

SPIRIDONVAMVAKAS, Institute of Toxicology, University of Wuerzburg, Germany

1. Toxicological Studies: General Aspects . . . 214

2. Acute Toxicity . . . . . . . . . . . . . . . . . . . . . . 216

2.1. Testing for Acute Toxicity by the Oral

Route: LD50Test and Fixed-Dose Method . 216

2.2. Testing for Acute Skin Toxicity . . . . . . . . . 219

2.3. Testing for Acute Toxicity by Inhalation . . 220

3. Repeated-Dose Toxicity Studies: Subacute,

Subchronic, and Chronic Studies . . . . . . . . 2224. Ophtalmic Toxicity . . . . . . . . . . . . . . . . . . 223

5. Sensitization Testing . . . . . . . . . . . . . . . . . 224

6. Phototoxicity and Photosensitization

Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226

7. Reproductive and Developmental Toxicity

Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226

7.1. Fertility and General Reproductive

Performance . . . . . . . . . . . . . . . . . . . . . . . 227

7.2. Embryotoxicity and Teratogenicity . . . . . . 227

7.3. Peri- and Postnatal Toxicity . . . . . . . . . . . 228

7.4. Multigeneration Studies . . . . . . . . . . . . . . . 228

7.5. The Role of Maternal Toxicity inTeratogenesis . . . . . . . . . . . . . . . . . . . . . . . 229

7.6. In VitroTests for Developmental Toxicity. 229

8. Bioassays to Determine the Carcinogenicity

of Chemicals in Rodents . . . . . . . . . . . . . . 229

9. In Vitroand In VivoShort-term Tests for

Genotoxicity . . . . . . . . . . . . . . . . . . . . . . . . 232

9.1. Microbial Tests for Mutagenicity. . . . . . . . 232

9.1.1. The Ames Test for Bacterial Mutagenicity . . 232

9.1.2. Mutagenicity Tests inEscherichia coli . . . . . . 239

9.1.3. Fungal Mutagenicity Tests . . . . . . . . . . . . . . 239

9.2. Eukaryotic Tests for Mutagenicity. . . . . . . 239

9.2.1. Mutation Tests inDrosophila melanogaster . 2399.2.2. In VitroMutagenicity Tests in Mammalian

Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239

9.3. In VivoMammalian Mutation Tests . . . . . 241

9.3.1. Mouse Somatic Spot Test . . . . . . . . . . . . . . 241

9.3.2. Mouse Specific Locus Test . . . . . . . . . . . . . 241

9.3.3. Dominant Lethal Test . . . . . . . . . . . . . . . . . 241

9.4. Test Systems Providing Indirect Evidence

for DNA Damage . . . . . . . . . . . . . . . . . . . . 241

9.4.1. Unscheduled DNA Synthesis (UDS) Assays . 241

9.4.2. Sister-Chromatid Exchange Test. . . . . . . . . . 242

9.5. Tests for Chromosome Aberrations

(Cytogenetic Assays) . . . . . . . . . . . . . . . . . 2439.5.1. Cytogenetic Damage and its Consequences . . 243

9.5.2. In VitroCytogenetic Assays. . . . . . . . . . . . . 244

9.5.3. In VivoCytogenetic Assays . . . . . . . . . . . . . 244

9.6. Malignant Transformation of Mammalian

Cells in Culture . . . . . . . . . . . . . . . . . . . . . 245

9.7. In VivoCarcinogenicity Studies of Limited

Duration . . . . . . . . . . . . . . . . . . . . . . . . . . 246

9.7.1. Induction of Altered Foci in the Rodent Liver 247

9.7.2. Inductionof Lung Tumorsin Specific Sensitive

Strains of Mice . . . . . . . . . . . . . . . . . . . . . . 247

9.7.3. Induction of Skin Tumors in Specific Sensitive

Strains of Mice . . . . . . . . . . . . . . . . . . . . . . 2479.8. Methods to Assess Primary DNA Damage . 247

9.8.1. Alkaline Elution Techniques . . . . . . . . . . . . 247

9.8.2. Methods to Detect and Quantify DNA

Modifications . . . . . . . . . . . . . . . . . . . . . . . 248

9.9. Interpretationof Results Obtained in Short-

Term Tests. . . . . . . . . . . . . . . . . . . . . . . . . 249

10. Evaluation of Toxic Effects on the Immune

System. . . . . . . . . . . . . . . . . . . . . . . . . . . . 250

11. Toxicological Evaluation of the Nervous

System. . . . . . . . . . . . . . . . . . . . . . . . . . . . 251

11.1. Functional Observational Battery . . . . . . . 251

11.2. Locomotor Activity . . . . . . . . . . . . . . . . . . 25212. Effects on the Endocrine System . . . . . . . . 253

References . . . . . . . . . . . . . . . . . . . . . . . . . 253

Abbreviations

AP: apurinic/apyrimidinic site

APS: adenosine 50-phosphosulfate

BHK: baby hamster kidney

CoA: Coenzym AED: effective dose

ELISA: enzyme-linked immunosorbent assay

FCA: Freunds complete adjuvant

GOT: glutamic acid oxalacetic transaminase

HGPRT: hypoxanthine guanine

phosphoribosyltransferaseLDH: lactate dehydrogenase

mRNA: messenger RNA

DOI: 10.1002/14356007.o27_015


2/42

1. Toxicological Studies: General

Aspects

The aim of toxicology is the assessment andmanagement of potential hazards from exposureof humans and the general environment includ-ing animals and plants to chemicals. To achievethis objective, detailed knowledge on the inher-ent hazard of a xenobiotic (for definition, see! Toxicology, 3. Evaluation of Toxic Effects,Section 2.1), that is, its acute and chronic toxic-ity, its no observed effect level (NOEL), and itsteratogenic, mutagenic and carcinogenic ef-fects, is required. This information can not beobtained from a single experiment. A battery ofin vivo and in vitro toxicity tests must be uti-lized. As required by law in most industrializedcountries, all toxicity testing must be performed

under the rules of good laboratory practice withexact documentation of all relevant conditionsand results. Adequate planning of toxicity testsfor obtaining optimal information from the ex-periments may greatly improve the basis for therisk assessment of a chemical and may reducethe number of animals needed and the financialexpense associated with toxicity studies. There-fore, all test batteries should be part of anintegrated approach to toxicity studies and in-clude not only methods to determine the toxic

effects of a chemical and their dose dependence,but also toxicokinetics, biotransformation, andmechanisms of action. This chapter provides anoverview of the currently used methods for theassessment of a chemicals toxic profile. Fordetails, specific guidelines on practical aspectsof toxicity studies and types of data required canbe obtained from web sites of national andinternational organisations (see ! Toxicology,1. Fundamentals, Section 1.4). OECD guide-lines for the toxicity testing of chemicals in vivo

are listed in Table 1.For the evaluation of a new chemicals toxic

effects in laboratory animals two types of studies

are carried out: acute-toxicity and repeated-dosing studies.

Acute Toxicity. Following administrationof a single dose of the test substance or ofmultiple doses given over a period of up to 24h, potentially adverse effects are usually moni-tored during the following 14 d. Acute toxicity

studies in animals aim to assess the human riskfrom single exposure to high doses, for example,in industrial accidents, after drug overdoses, orafter suicide attempts.

Repeated-Dosing Studies: Subacute,

Subchronic, and Chronic Toxicity. The pur-pose of repeated daily doses of a chemical for partof the animals life span is to study subchronicand chronic effects. Studies onsubacute toxicity

are carried out for two to four weeks, while studiesonsubchronic toxicityusually last for a period ofthree months. These studies are helpful in asses-sing the human risk resulting from frequent ex-posure to household or workplace chemicals andfrom intake of chemicals used for therapeuticpurposes. Studies to determine chronic toxiceffects are carried out for at least six months;studies aiming to investigate the carcinogeniceffects of a test compound are carried out overthe animals entire lifetime. Lifetime exposure of

humans may occur to widespread environmentalpollutants, food additives, or residues of agricul-tural chemicals in food.

In addition to the acute and repeated-dosetoxicity studies, the reproductive and develop-mental toxicity as well as the genotoxicity of anew chemical must be investigated in separateexperiments.

Many thousands of new and potentially toxiccompounds are synthesized every year. It wouldbe a waste of money, resources and manpower if

the entire battery of toxicity tests were automat-ically performed for every new chemical.Therefore, toxicity testing is rather undertaken

MTD: maximum tolerated dose

NADPH: nicotinamide dinucleotide phosphate

(H)

NOEL: no-observed-effect-level

SHE: Syrian hamster embryo

SMART: somatic mutation and recombination

test

TK: thymidine kinase

UDS: unscheduled DNA synthesis

214 Toxicology, 2. Assessment Methods Vol. 37


3/42

on the basis of a decision-point approach inseveral stages, as shown in Figure 1. At the endof every stage, the decision must be met, if thedevelopment will be continued or if, on the basisof the toxicity data available so far, the potentialhuman risk of the exposure to this chemical isunacceptable. If the latter is true, the develop-ment and consequently the toxicity testing isstopped.

Animal Husbandry. The use of standard-ized conditions for the housing of animals playsa major role in the planning, evaluation, and

interpretation of toxicity tests. Animals must bekept in a controlled environment, i.e., constanttemperature of 22 3 C, sufficient ventilation,relative humidity between 30 and 70% and a12 h light/dark cycle. Diet composition andquality of drinking water must also be standard-ized and controlled throughout the experiment.Only healthy young adult animals should beenrolled in the studies, and the animals shouldbe allowed to acclimatize to the experimental

conditions for at least one week prior to firstdosing. After the acclimatization period, ani-mals with poor health or body weights varying

Table 1. OECD guidelines on short- and long term toxicity testing in vivo

No. Title Original adoption Updated

401 Acute Oral Toxicity 12 May 1981 20 Dec. 2002*

402 Acute Dermal Toxicity 12 May 1981 24 Feb. 1987

403 Acute Inhalation Toxicity 12 May 1981

404 Acute Dermal Irritation/Corrosion 12 May 1981 24 April 2002

405 Acute Eye Irritation/Corrosion 12 May 1981 24 April 2002

406 Skin Sensitization 12 May 1981 17 July 1992

407 Repeated Dose 28-Day Oral Toxicity Study in Rodents 12 May 1981 27 July 1995

408 Repeated Dose 90-Day Oral Toxicity Study in Rodents 12 May 1981 21 Sept. 1998

409 Repeated Dose 90-Day Oral Toxicity Study in Non-Rodents 12 May 1981 21 Sept. 1998

410 Repeated Dose Dermal Toxicity:28-Day 12 May 1981

411 Subchronic Dermal Toxicity: 90-Day 12 May 1981

412 Repeated Dose Inhalation Toxicity: 28/14-Day 12 May 1981

413 Subchronic Inhalation Toxicity: 90-Day 12 May 1981

414 Prenatal Developmental Toxicity Study 12 May 1981 22 Jan. 2001

415 One-Generation Reproduction Toxicity 26 May 1983

416 Two-generation Reproduction Toxicity Study 26 May 1983 22 Jan. 2001

417 Toxicokinetics 4 April 1984

418 Delayed Neurotoxicity of Organophosphorus SubstancesFollowing Acute Exposure

4 April 1984 27 July 1995

419 Delayed Neurotoxicity of Organophosphorus Substances:

28-Day Repeated Dose Study

4 April 1984 27 July 1995

420 Acute Oral Toxicity Fixed Dose Procedure 17 July 1992 17 Dec. 2001

421 Reproduction/Developmental Toxicity Screening Test 27 July 1995

422 Combined Repeated Dose Toxicity Study with the

Reproduction/Developmental Toxicity Screening Test

22 March 1996

423 Acute Oral ToxicityAcute Toxic Class Method 22 March 1996 17 Dec. 2001

424 Neurotoxicity Study in Rodents 21 July 1997

425 Acute Oral Toxicity: Up-and-Down Procedure 21 Sept. 1998 17 Dec. 2001

426 Developmental Neurotoxicity Study Draft New Guideline, October 1999

427 Skin Absorption: In vivo method Expected, Approved by WNT (May 2002)

428 Skin absorption: In vitro method Expected, Approved by WNT (May 2002)

429 Skin Sensitization: Local Lymph Node Assay 24 April 2002

430 In Vitro Skin Corrosion: Transcutaneous Electrical

Resistance Test (TER)

Expected, Approved by WNT (May 2002)

431 In Vitro Skin Corrosion: Human Skin Model Test Expected, Approved by WNT (May 2002)

432 In Vitro 3T3 NRU phototoxicity test Expected, Approved by WNT (May 2002)

433 Acute Inhalation Toxicity: Fixed Dose Procedure Draft New Guideline, October 1999

451 Carcinogenicity Studies 12 May 1981

452 Chronic Toxicity Studies 12 May 1981

453 Combined Chronic Toxicity/Carcinogenicity Studies 12 May 1981

*Date of deletion.

Vol. 37 Toxicology, 2. Assessment Methods 215


4/42

by more than 20% of the the groups mean bodyweight are either excluded from the studies orrandomized to ensure a homogenous populationin the different control and treatment groups.The basic guidelines choice of species, num-

ber of animals, dosing regimens, duration andfrequency of observation, assessment of specificbody functions for acute, subchronic andchronic toxicity tests are summarized in Table 2.

A variety of in vitro methods are in develop-ment to reduce the numbers of animals used intoxicity testing. Some of the developed methodshave gained regulatory acceptance, and somemay be used for a priority-determining processfor further testing. All well-evaluated methodsfocus on local effects such as skin and eye

irritation, where most progress in the develop-ment of nonanimal methods has been made.Regarding replacement of toxicity studies on

systemic effects after repeated exposure by non-animal methods, due to the complexities ofinteractions resulting in toxic responses, accept-able nonanimal tests equaling the predictivepower of animal testing are unlikely to be avail-able in the near future.

2. Acute Toxicity

2.1. Testing for Acute Toxicity by the

Oral Route: LD50Test and Fixed-Dose

Method

The objectives of acute toxicity tests are

1. To assess the intrinsic toxicity of the testcompound

2. To identify target organs of toxicity affectedby the xenobiotic

3. To provide information concerning the doseselection and treatment regimens for repeated-dose studies

4. To provide information for human risk as-sessment after a single high-dose exposure tothe chemical

5. To provide essential data for the classifica-

tion, labeling, and transportation of the chem-ical (regulatory view report)

LD50 Test. The determination of the meanlethal dose (LD50) is still often considered as thefirst step in the evaluation of the acute oral orinhalation toxicity of a new chemical; with thepresent knowledge and recent experiences, theformal determination of the LD50 is no longerconsidered as necessary, and alternative methodsthat are also to be used for classification and

labeling have been developed. However, sincethe test is still widely used, it will be brieflydescribed before treating the newer methods usedin testing of acute toxicity and the reasons that ledto these changes.

In the LD50 test, groups of animals (usuallyfemale rats) are treated with graduated doses ofthe test compound, and by using mathematicalmodels the dose which causes death in 50% ormore of the population is determined. Interna-tionally accepted guidelines recommend the use

of at least three dose groups with five males andfive females for each dose or the use of three dosegroups with five animals of one sex and one dose

Figure 1. Evaluation of the toxicity profile of a new chemi-cal compound on the basis of a decision point approachAt every step of toxicity testing, further development may beinterrupted if, on the basis of data collected up to that point,the human risk is considered unacceptable



5/42

with five animals of the other sex [1]. The LD50 isthen determined from data obtained as describedin ! Toxicology, 1. Fundamentals, Section 1.7.Chemicals with LD50 values 25 mg/kg areconsidered very toxic, between 25 and 200 mg/kgas toxic, and between 200 and 2000 mg/kg asharmful. Some potentials and limitations of theLD50test follow:

Potential

1. Useful as a first approximation of hazards inthe workplace

2. Basis for the design of subchronic studies

3. Properly conducted test may give useful infor-mation on other relevant toxicity parameters

4. Rapid completion

Limitations and Problems

1. Lethality only criterion applied, other toxiceffects not considered

2. Animal welfare is major point of concernbecause large number of animals are requiredto obtain statistically acceptable values

3. Large variations in LD50 in different labora-tories with identical chemicals, many influ-encing factors

Table 2. Basic guidelines for acute, subchronic, and chronic oral toxicity tests

Acute oral Subchronic oral Chronic oral

Animals rats preferred rodent and nonrodent species rodent and nonrodent species

Sex males and females equally distributed per dose level

Age young adult, weight variation

within 20 % of mean

rodents, 6 weeks; dogs 4 6 months old

Number of animals at least 10 (5 per sex) at least 20 for rodents (10 per

sex)

50 per sex group for rodents

Number of treatment groups 3; mortality rates between 10

and 90 % should be produced

3; mortality shouldnot exceed

10 % in high-dose group

3; low dose should reflect

expected human exposure and

high dose must produce not

more than 10 % mortality

Untreated control not necessary yes yes

Vehicle control yes, if vehicle of unknown

toxicity is used

yes yes

Dosing gavage; single dose, same

dose of vehicle; if necessary

use divided doses over 24 h

diet, gavage, drinking water diet, gavage, drinking water

Duration of study at least 14-d observation

period

90 d 6 24 months in rats

Body weight determination before dosing, weekly there-

after, and at death

weekly and at termination weekly for first 13 weeks;

every 2 weeks thereafter and

at termination

Necropsy all animals all test animals; organ weights of liver, kidney, heart, lungs, brain,

gonads, adrenals, and spleen

Histopathology examination of organs show-

ing evidence of gross patho-

logical change

all tissues high-dose and

control groups; liver, kidney,

heart, lungs, target organs, and

any gross lesion in mid- and

low-dose groups

all tissues of animals

Frequency of cage-side

observations

frequently during day of dos-

ing; once each morning and

late afternoon thereafter

daily daily

Observations, assessments nature, onset, severity, and duration of any effect observed

ophthalmoscopy: pretest and at termination in control and high-dose

groups

hematology/clinical

chemistry:

pretest, dosing midpoint,

termination

pretest and at 3, 6,12, 18, 24

months

urine analysis: dosing mid-

point, termination

pretest and at 3, 6,12, 18, 24

months



6/42

4. Species and strain differences cause difficul-ties in extrapolation

5. No information on chronic toxicity obtained(chronic effects are more important for regu-lating exposure)

The scientific significance of the LD50 testhas been repeatedly questioned, not only be-cause the lethal dose is not relevant for humanrisk assessment but also on the basis of thevariability of the test results and last not butleast for reasons of animal welfare. Compara-tive assessment of LD50 values in 60 laborato-ries under controlled conditions resulted forexample in considerably different LD50 values(by a factor of up to 14). Therefore, numerousalternatives to the LD50 test have been proposedfor the evaluation of acute toxicity that rely onsigns of toxicity rather than on mortality. One ofthis procedures, the fixed-dose method - hasrecently gained acceptance by the OECD andthe EU

Fixed-Dose Method. The fixed-dose meth-od relies on the observations of clear signs oftoxicity developed at one of a series of fixed-dose levels (i.e. 5, 50, 300, and 2000 mg/kg

of the chemical oper kilogram body weight).The dose levels at which signs of toxicity butno deaths are detected are used to classify thetest compounds according to their toxic potential(Table 3).

As can be seen in Table 3, the fixed-dosemethod allows a classification identical to thatpreviously obtained in the LD50 test. Moreover,comparative investigations utilizing both the

LD50test and the fixed-dose procedure revealedthat in the majority of the test compunds (8090%), the toxicity class assigned by determiningthe LD50was identical to that determined by thefixed-dose method [1, 2].

In the fixed-dose method, at least ten animals

(five per sex) are used for each dose investigat-ed. The initial dose chosen (5, 50, 300, or 2000mg/kg body weight) is one that is judged likelyto produce evident toxic effects, but no mortali-ty. When such a judgement can not be made dueto lack of information on the potential toxiceffects of the xenobiotic, an initial sightingstudy should be carried out. If clear signs oftoxicity do not occur at the starting dose of 300mg/kg during the two weeks observation period,the dose is increased to the next level. A carefulclinical examination of the animals is performedat least twice on the day of administration andonce daily thereafter for the next two weeks.Animals obviously in pain or showing severesigns of distress or toxicity are humanely killed.Cage-side examinations include skin and fur,eyes and mucous membranes, respiratory sys-tem, blood pressure, somatomotor activity, andbehavior (for procedures, see Chapter 36). Par-ticular attention is directed to observation of

tremors, convulsions, hypersalivation, diar-rhoea, and coma as indices of neurotoxicity.Food consumption and weight development arealso monitored constantly. At the end of theobservation period, all animals in the study arekilled and subjected to gross autopsy. Organsshowing macroscopic evidence of gross pathol-ogy are further subjected to histopathologicalexamination.

Table 3. Classification of toxicity of a xenobiotic with the fixed-dose method

Dose (oral), mg/kg Results Classification

5 less than 90 % survival very toxic

90 % or more survival but evident toxicity toxic

90 % or more survival, no evident toxicity retest at 50 mg/kg

50 less than 90 % survival toxic, retest at 5 mg/kg

90 % or more survival, but evident toxicity harmful

90 % or more survival, no evident toxicity retest at 500 mg/kg

500 less than 90 % survival or evident toxicity and no death harmful, retest at 50 mg/kg

no evident toxicity retest at 2000 mg/kg

2000 less than 90 % survival harmful

90% or more survival, with or without evident toxicity unclassified, does not representa significant acute toxic risk if

swallowed, no further testing necessary



7/42

The fixed-dose method offers several impor-tant advantages as compared with the traditionalLD50test:

1. The available evidence suggest that thefixed-dose method produces more consistent

results without substantial interlaboratoryvariation.

2. It provides information on the type, time ofonset, duration, and consequences of toxiceffects. This information is more relevant forassessing the risks of human exposure to thechemical than the mean lethal doses of theLD50test.

3. It requires fewer animals than the LD50 test(roughly 50%) and subjects the animals to lesspain and distress.

4. It enables the classification of chemicals ac-cording to regulatory requirements.

For the standard acute oral and dermal teststhe LD50 should be determined, except whenthe substance causes no mortality at the limit dose(usually 2000 mg/kg). Similarly, for an acuteinhalation toxicity study the LC50 should be deter-mined, unless no mortality is seen at the limitconcentration (5 mg per L per 4 h for aerosols

and particulates, 20 mg per L per 4 h for gases andvapors). In the fixed-dose procedure, the discrimi-nating dose (the highest of the preset dose levelswhich can be administered without causing mor-tality) should be determined. For the acute toxicclass and the up-and-down methods the final doseused in the study should be determined followingthe testing protocol, except when the substancecauses no mortality at the limit dose.

Whichever approach is used in determiningacute toxicity critical information must be de-

rived from the data to be used in risk assessment.It is important to identify the dose levels at whichsigns of toxicity are observed, the relationship ofthe severity thereof with dose, and the level atwhich toxicity is not observed (i.e. the acuteNOAEL). However, note that a NOAEL is notusually determined in acute studies, partly be-cause of the limitations in study design.

2.2. Testing for Acute Skin Toxicity

Irrespective of whether a substance can becomesystemically available, it may cause changes at

the site of first contact (skin, eye, mucous mem-brane/gastrointestinal tract, or mucous mem-brane/respiratory tract). These changes are con-sidered local effects. A distinction can be madebetween local effects observed after single andafter repeated exposure. For local effects after

repeated exposure, see ! Toxicology, 1. Funda-mentals, Section 1.9.2. Only local effects aftersingle ocular, dermal, or inhalation exposure aredealt with in this section. Substances causinglocal effects after single exposure can be furtherclassified as irritant or corrosive substances,depending on the (ir)reversibility of the effectsobserved.

Irritants are noncorrosive substances whichthrough immediate contact with the tissue cancause inflammation. Corrosive substances arethose which can destroy living tissues with whichthey come into contact.

Knowledge on the dermal toxicity of a newchemical is one of the prerequisites for assess-ment of the risks associated with human expo-sure to the chemical, because skin contact mayrepresent a very important route of exposure inthe occupational setting and in the home. Test-ing for dermal toxicity is usually performed inrabbits. Three types of application of the test

chemical are employed: nonocclusive, semioc-clusive, and occlusive. The test compound isapplied uniformly to the back or a band aroundthe trunk (clipped free of hair); approximately10% of the body surface of the animal should becovered. Solid substances are pulverized andmoistened to a paste with physiological saline oranother appropiate solvent whose effects havebeen fully evaluated prior to the skin test. Forocclusive or semiocclusive testing, the applica-tion site is covered with a plastic sheet (or other

impervious material) or with a porous gauzedressing, respectively. For unocclusive expo-sure, the application site should be as close to thehead as possible to prevent ingestion of thechemical by the animal licking the site of appli-cation. The duration of exposure varies between4 and 24 h. If no test-chemical-related toxiceffects on the skin or systemic toxicity areobserved up after doses of up to 2 g/kg bodyweight, testing at higher doses is unnecessary.At the end of the exposure period, the compound

is removed with cotton wool soaked in an ap-propiate solvent and the skin irritation is scoredaccording to the Draize scoring system as shown



8/42

in Table 4. In addition, any adverse systemiceffects caused by percutaneous absorption of thetest compound are monitored.

However, substantial differences exist in skinanatomy between humans and experimental ani-mals. In general, the penetration of chemicalsthrough the human skin is similar to that of pig,miniature swine, and squirrel monkey and clearlyslower than that of the rat and rabbit. For exam-ple, administration of the insecticides lindane

and parathion to rabbit skin results in an absorp-tion of 51.2 and 99.5% of the dose, respectively;the corresponding absorption rates for humanskin are 9.3 and 9.7%.

Since the 1980s, in vitro studies using humanskin samples have been increasingly conductedto estimate percutaneous absorption of chemi-cals. The following experimental design is com-monly used: A piece of excised human skin isattached to a diffusion apparatus that has a topchamber for the test compound, an O-ring to hold

the skin in place, and a bottom chamber to collectsamples for analysis. The flow of a chemicalacross the skin can be calculated with modelsbased on chemical thermodynamics, taking intoconsideration the octanol/water partition coeffi-cients, the saturated concentration in aqueoussolution, and the molecular mass of the testcompound [3]. However, for routine applica-tions, the method has not been sufficientlyevaluated.

Furthermore, an increasing number of tox-

icokinetic models for estimating the extent ofpercutaneous absorption of chemicals has ap-peared in the literature. Among them, a physio-

logically based toxicokinetic model was recent-ly developed to describe the percutaneous ab-sorption of volatile and lipid-oluble organiccontaminants in dilute aqueous solution [4].This toxicokinetic model considers both physi-ological parameters such as volumes of bodycompartments and blood flow rate, as well as theproperties of the test compound. Presently, thesemodels do not play a role in regular toxicitytesting and are therefore not discussed in depthhere.

More recently extensive progress has beenalso made in developing in vitro systems forevaluating the dermal irritation potential of che-micals. An overview of the systems that have

been evaluated so far for a range of compoundsby comparison of their predictive accuracy withanimal test results is presented in Table 5 (for areview see [5]).

When evaluating these studies, attentionshould be given to the occurrence of persistingirritating effects, even those which do not lead toclassification. Effects such as erythema, oedema,fissuring, scaling, desquamation, hyperplasia,and opacity which are not reversible within thetest period may indicate that a substance will

cause persistent damage to the human skin andeye.

2.3. Testing for Acute Toxicity by

Inhalation

Studying the toxicity of a chemical by inhala-tion exposure requires a considerable techno-logical input. Therefore, inhalation exposure isusually not tested if this absorption pathway is

not expected to occur because the test chemicalis not volatile or the physicochemical proper-ties of solids do not allow the generation of

Table 4.Evaluation of skin reactions according to the Draize scoring

system

Erythema Score

No erythema 0

Very slight erythema (barely perceptible) 1

Well-defined erythema 2

Moderate to severe erythema 3

Severe erythema (beet redness) 4

No edema 0

Very slight edema (barely perceptible) 1

Slight edema (edges of area well

defined by definite raising)

2

Moderate edema (area raised ca. 1 mm) 3

Severe edema (raised more than

1 mm and extended beyond

area of exposure)

4

Table 5. In vitro test systems for detection of dermal irritation

potential

System End point

Mouse skin organ culture leakage of LDH* and GOT**

Human epidermal keratinocytes release of labeled

arachidonic acid, cytotoxicity

Cultured BHK21/C13 cells growth inhibition,

cell detachment

SKINTEX protein mixture protein coagulation

*LDH Lactate dehydrogenase.

**GOT Glutamic acid oxalacetic transaminase.



9/42

respirable particles. Particles with diametersgreater than 100 mm are unlikely to be inhaled,because they settle too rapidly. Particles withdiameters of 1050 mm are likely retained inthe nose and the upper parts of the respiratorytract, while particles with diameters of less than

7 mm can reach the alveoli of the human lung.When performing toxicity studies with inhala-tion exposure, the differences in respiratoryphysiology between humans and the smalllaboratory rodents must be considered. In con-trast to humans, the rat is an obligate nosebreather with a complex nasal turbinate struc-ture which filters many small particles. There-fore, the upper size limit for particles reachingthe alveolar region in rats is in the range of 34mm in diameter.

The duration of exposure in acute inhala-tion studies is usually 46 h, and they may beperformed either as whole-body or head(nose)-only procedures in specific exposure cham-bers. A number of important considerationsshould be taken into account in planning andevaluating inhalation studies. The frequentlymade assumption that on the basis of theirphysical form, gases and vapors will be ab-sorbed uniformly throughout the respiratory

tract is incorrect. Many gases or vapors, suchas ammonia, formaldehyde, and sulfur dioxidehave high solubility in water and are rapidly

absorbed by the humid epithelial surface of theupper respiratory tract. Therefore, toxic effectsobserved after inhalation of this type of chem-ical will generally be confined to these re-gions, especially to the nasal passages. Incontrast, chemicals with low solubility in wa-

ter such as nitrogen dioxide, phosgene, andozone will penetrate readily to the low pulmo-nary regions, even at relatively low concen-trations in the respiratory air. In mixed atmo-spheres containing both a gas or vapor andparticulates, the vapor or gas may be absorbedon the particulate fraction, so that the deposi-tion pattern of the vapor or gas is governed bythe size of the particulate fraction and not bythe water solubility of the vapor or gas.

The inhalation exposure of experimental an-imals may be performed in a dynamic or staticmode. In dynamic systems, the test atmosphereis continuously renewed, ensuring atmosphericstability and constant concentrations of thechemical in the gas phase. This mode of inhala-tion exposure is more complicated and requireslarger amounts of the test chemical and suitablesystems to ensure complete mixing of the con-tinuously applied test compounds with thestream of air flowing through the exposure

system (Fig. 2).Static systems are sealed, and the atmo-sphere is circulated. The concentration of the

Figure 2. Example of an open inhalation chamber for exposure to volatile liquidsa) Meter; b) Syringe coupled to infuser; c) Heated beads of silica or glass; d) Compressor; e) Metering valves; f ) Mixingchamber; g) Exposure chamber; h) Ventilation; i) Sampling valves for determination of atmospheric concentration; j) Exhaust



10/42

chemical in the gas phase decreases duringexposure due to its uptake and biotransforma-tion by the experimental animals. Static sys-tems are used predominantly in acute toxicitystudies and in research laboratories becausethey are relatively inexpensive and consume

only a small amount of the chemical in com-parison to the amounts need to generate adynamic exposure. In analogy to the LD50values obtained in the acute oral studies themean lethal air concentrations LC50 are as-sessed in acute inhalation studies, usually foran exposure time of 4 h (Fig. 3).

3. Repeated-Dose Toxicity Studies:

Subacute, Subchronic, and ChronicStudies

Repeated-dose toxicity studies assess the effectsresulting from the accumulation of a compoundor its toxic effects in the organism and, unlikeacute studies, they can also reveal toxic effectsthat appear after a latency period. A classicexample is the delayed neuropathy caused bysome organophosphorous insecticides and bycresyl phosphates, which is manifested several

weeks after the first administration of the testcompound. In contrast to the marked clinicalsymptoms observed in the course of delayed

neuropathy, these compounds hardly cause anyacute symptoms immediately after the first ad-ministration. Hence, false negative results maybe obtained if an assessment were based onlyacute toxicity testing. The major aims of repeat-ed-dose toxicity studies are the identification of

starting points for the extrapolations required inhuman risk assessment such as NOAEL andbenchmark doses, and the identification of criti-cal end points to be carried over into the riskassessment process.

Testing for subacute toxicity is usually per-formed over a time period of 24 weeks at threedose levels as an aid in selecting the dose levelsfor the subchronic studies. Studies to determinesubchronic effects are usually performed in ratsand dogs over 10% of the animals life-span

(3 months in rats, 12 months in dogs). At thestart of the study, rodents should be 68 weeksold and dogs 46 months old. The animal num-bers enrolled are between 10 and 20 rats and 6and 8 dogs per sex per dose group. Ideally, thelowest dose of the chemical should not inducetoxic effects, the intermediate dose should induceslight toxicity, and the high dose should induceclear signs of toxicity without causing death inmore than 10% of the animals.

In both subchronic and chronic toxicity studies

(see below), the test compound is often incorpo-rated into the diet or added to the drinking water.Food consumption varies from weanling to matu-

Figure 3. Closed exposure systema) Oxygen cylinder; b) Metering valve; c) Solenoid valve; d) Mixing chamber; e) Thermometer; f ) Oxygen sensor; g) Pressure

gauge; h) Exposure chamber; i) Oxygen monitor; j) Injection port; k) Condenser; l) Flow meter; m) Carbon dioxide absorber;n) Gas chromatograph; o) Pump



11/42

rity, with younger animals consuming more foodon a bodyweight basis. Therefore, it is necessary topredict the changes in body weight and foodconsumption on a weekly basis and adjust theconcentration of the test compound in the diet inorder to ensure constant dosing throughout the

study. Compounds not stable in diet or water ornot accepted by the animals may also be applied bygavage. Application by gavage directly into thestomach ensures constant dosing, but gavage stud-ies require skilled personnel, and gavage-relatedtrauma may reduce survival in all groups. Oralapplication with the feed is usually performed on a7 days per week basis, while a 5 days per weekscheme is frequently used for gavage administra-tion, skin application, and inhalation studies.

Studies on the chronic toxicity of chemicalsare usually performed for at least six months inrodents and 12 months in dogs, the chronictoxicity studies can be combined with a carcino-genicity study. Dose levels are usually selectedon the basis of the results of studies on acute andsubacute toxicity. The highest dose appliedshould be toxic, i.e., suppress body weight upto 10% (maximum tolerated dose, MTD). Thetwo other dose-levels are usually 1/4 and 1/8 ofthe MTD. Xenobiotics showing no adverse ef-

fects in the short-term studies are usually tested atdoses which are 100200 times higher than theexpected human exposure.

In both subchronic and chronic studies, cage-side examination and clinical chemistry are per-formed routinely during the test period. Aftertermination of the study, much emphasis isplaced on the histopathological evaluation oftreatment-induced adverse effects

Cage-side observations during the study

1. Body weight, food and water consumption2. Skin and fur, eyes, mucous membranes3. Respiration and blood circulation4. Motor activity and behavioral pattern

Clinical chemistry during the study

1. Blood: erythrocyte, leukocyte, and differen-tial leukocyte counts; hemoglobin concentra-tion; hematocrit, platelet, and reticulocyte

counts; electrolytes; inorganic phosphorusand alkaline phosphatase; glucose, protein,albumin, creatinine, urea, lipids, enzymes

2. Urine: volume and coloration/turbidity, os-molality and pH, glucose and protein, urineenzymes and cytology

Toxicologic pathology after termination of

the study and in animals dying during the

study

1. Organ weights and macroscopic evaluation2. Histopathological examination of brain, liver,

kidney, spleen, testes, and every organ withmacroscopic changes

4. Ophtalmic Toxicity

The majority of injuries to the eyes by direct

contact with a chemical occur with substanceswhich are handled in an uncontrolled manner,e.g,. by children in the home; and this type ofinjury is easily prevented at the occupationalsetting by using simple protective procedures.However, many irritant gases and vapors mayalso produce ophtalmic toxicity, and these toxiceffects are of practical importance in occupa-tional medicine.

The conventional in vivo eye irritation test in

the rabbit was formalized by Draize some 50years ago and still remains the only fully validat-ed method to assess ophtalmic toxicity. Increas-ing criticism primarily based on the discomfortand the deliberate injury caused to the animalsled to the development of several in vivo and invitro alternative methods, which will bedescribed at the end of this chapter.

In the conventional in vivo test the chemical(0.1 ml of liquid or 100 mg of solid chemical) isinstilled into one eye of each of the test rabbits,

the contralateral eye serving as control. Eyes arethen examined periodically (usually after 1, 24,48, 72 h and 7 dd) and the ocular lesions arescored essentially according to Draize et al.:

Cornea

A) Opacity

No opacity 0

Scattered or diffuse area, iris details

clearly visible

1

Easily discernible translucent areas, iris

details slightly obscured

2

Opalescent areas, iris details not visible,

pupil size barely discernible

3

Opaque, iris invisible 4



12/42

B) Area of cornea involved

One-quarter or less, but not zero 1

Greater than one-quarter, but less than half 2

Greater than half, but less than three-quarters 3

Greater than three-quarters, up to whole area 4

Corneal score (A)(B)5 (maximum total score 80)

Iris

A) Normal 0

Folds above normal, congestion, swelling,

circumcorneal injection (any or all), iris still

reacting to light

1

No reaction to light, hemorrhage, gross destruction 2

Iris score (A)5 (maximum total score 10)

Conjunctivae

A) Vessels normal 0

Vessels definitely injected, above normal 1

Diffuse, deep crimson red, individual vessels

not readily discernible

2

Diffuse beefy red 3

B) No chemosis (swelling) 0

Any swelling above normal (includes

nictitating membrane)

2

Obvious swelling with partial eversion of lids 2

Swelling with lids about half closed 3

Swelling with lids about half to completely closed 4

C) No discharge 0

Any amount of discharge different from normal 1

Discharge with moistening of lids and hairs adjacent to lids 2

Discharge with considerable moistening around eyes 3

Conjunctival score [(A) (B) (C)]2 (maximum total

score 20)

Total maximum score (cornea iris conjunctiva) 110

There is increasing evidence that a volume of0.01 ml of liquid xenobiotics is as sensitive as theconventionally used 0.1 ml and is probably moreappropiate for comparison with human exposuresituations. Ocular toxicity testing for exposure togases, vapors, and aerosols is carried out inappropiate exposure chambers.

A number of in vivo and in vitro alternatives tothe conventional eye irritation test have beensuggested. The in vivo alternatives aim at reduc-ing discomfort of the animals by employinglower doses of the test material and increasingthe sensitivity of the test by using noninvasiveobjective measurements. Among them, the as-sessment of corneal thickness and of the intra-vascular pressure seem to be sensitive parametersto identify mild to moderately irritant chemicals.In spite of the large number of suggested in vitro

tests, currently no single in vitro test has provedeffective in predicting the eye irritation. There-fore, in vitro tests can not replace the rabbit eye

test yet. However, in vitro assays are useful asscreens for product development to reduce thenumber of tests performed in animals later. Anumber of these tests are presented in [6]

5. Sensitization Testing

Chemicals that have the potential to elicit allergicreactions are continously introduced into thehuman environment. Therefore, allergic reac-tions of the skin are becoming an increasinglyimportant problem, especially in the workplace.Allergic contact dermatitis is one of the mostcommon occupational diseases and may becomedebilitating unless the causative agent is identi-fied and exposure stopped. While irritant derma-titis is generally produced by direct interaction ofthe chemical with skin constituents, allergic der-matitis is the result of a systemic immune reac-tion which in turn induces effects in the skin. Animportant characteristic of allergic reactions thatmust be taken into account when testing forallergenic potential, is that allergic responsesusually have a biphasic course. The inductionperiod between initial contact with the causativeagent and the development of skin sensitivity

may be as short as two days for strong sensitizerssuch as poison ivy extract, or may require severalyears for a weak sensitizer such as chromate; formost of the chemical compounds with allergenicpotential the induction period usually takes from10 to 21 d. After this initial development ofsensitivity to a certain allergenic chemical, thetime between reexposure to this agent and theoccurrence of clinical allergic symptoms is gen-erally between 12 and 48 h; in animal testing, thisperiod is called the challenge phase.

The general objectives are to determinewhether there are indications from human expe-rience of skin allergy or respiratory hypersensi-tivity following exposure to the agent and wheth-er the agent has skin sensitization potential basedon tests in animals. There are two methodscurrently described in EU Annex V and OECDguidelines for skin sensitization in animals: theguinea pig maximisation test (GPMT) and theBuehler test. The GPMT is an adjuvant-type testin which the allergic state (sensitization) is po-

tentiated by the use of Freunds Complete Adju-vant (FCA). The Buehler test is a non-adjuvantmethod involving topical application for the



13/42

induction phase rather than the intradermal in-jections used in the GPMT (Table 6; for reviewssee references [7, 8]. Although they differ byroute and frequency of treatment, they all utilizethe guinea pig as test species. In general, for theinduction phase the chemical is administered tothe shaved skin intradermally, epicutaneously, orby both routes several times over a period of twoto four weeks. Freunds complete adjuvant(FCA, a mixture of heat-killed Mycobacteriumtuberculosis, paraffin oil, and mannide monoole-ate) is often included to increase the immuno-logical response. During the challenge phase, anonirritating concentration of the chemical isapplied. The concentration of the test chemical

and the application route (epi- or intradermal) areoften different between the two phases. Sensiti-zation is assessed by examining the skin reac-tions (edema, erythema) following the challengephase and comparing them with any skin reac-tions observed immediately after the inductionphase; the latter reactions are considered to resultfrom direct irritating (toxic) properties of the testchemical. Hence, the difference between thesymptoms observed after the induction and afterthe challenge phase is attributed to the allergenic

effects of the chemical.The guinea pig maximization test is the most

widely used and is considered to be very sensi-tive. The first part of the induction phase includessimultaneous injection of FCA alone, the testcompound in saline, and the test compound inFCA into three different areas in close proximityto each other. The second part of the inductionphase 7 d later employs epicutaneous applicationof the chemical on a filter paper, which is oc-cluded and left in place for 48 h. The challenge

phase is conducted epicutaneously for 24 h, twoweeks after the induction phase. The maximiza-tion test is very sensitive and may produce false

positive results. The original procedure (injec-tion of the test compound) does not allow testingof final product formulations. Therefore, a modi-fied procedure has been developed. In the firstweek, the FCA is injected four times and the testproduct formulation is administered epidermally,as in the second induction week.

Both the GPMT and the Buehler test havedemonstrated the ability to detect chemicals withmoderate to strong sensitization potential, as wellas those with relatively weak sensitization po-tential. These guinea pig methods provide infor-mation on skin responses, which are evaluatedfor each animal after several applications of thesubstance, and on the percentage of animals

sensitized.The murine local lymph node assay (LLNA) isanother accepted method for measuring skinsensitization potential. It has been validated in-ternationally and has been shown to have clearanimal welfare and scientific advantages com-pared with guinea pig tests. In June 2001, theOECD recommended that the LLNA should beadopted as a stand-alone test as an addition to theexisting guinea pig test methods.

Respiratory hypersensitivity is a term that is

used to describe asthma and other related respi-ratory conditions, irrespective of the mechanismby which they are caused. When directly consid-ering human data in this document, the clinicaldiagnostic terms asthma, rhinitis, and alveolitishave been retained.

There are currently no internationally recog-nised test methods to predict the ability of che-micals to cause respiratory hypersensitivity. Po-tentially useful test methods based on allergicmechanisms are the subject of research and

development. However, there are currently notest methods under development which are de-signed specifically to identify chemicals that

Table 6. Guinea pig sensitization tests

Test Induction: route/number of applications Challenge: route/number of applications

Draize intradermal/10 intradermal/1

Open epicutaneous epidermal open/20 epidermal open/1

Buehler epidermal occlusive/3 epidermal occlusive/1

FCA*

intradermal in FCA*/3 epidermal open/1

Split adjuvants epidermal occlusive/4 FCA* intradermal/1 epidermal occlusive/1Optimization intradermal FCA*/10 intradermal/1 epidermal occlusive/1

Maximization intradermal FCA*/1 epidermal occlusive/1

epidermal occlusive/1

*FCA Freunds complete adjuvant.



14/42

cause respiratory hypersensitivity by nonimmu-nological mechanisms.

6. Phototoxicity and

Photosensitization Testing

The biologically active spectrum of light can bedivided into UV (220400 nm) and visible light(400760 nm). The UV spectrum is further di-vided into UVA (315400 nm), UVB (280315 nm), and UVC (220280 nm); the last-namedis absorbed in the stratosphere and does not reachthe surface of the earth. The primary source oftoxic effects on the skin is UVB, although UVAmay also play a critical role in some reactions.

Xenobiotics localized within the skin may beactivated by UVB and induce phototoxicity and/or photosensitization (photoallergy). Photoal-lergy is similar, both mechanistically and clini-cally, to allergic contact dermatitis, the onlydifference being that the chemical must reactwith light to become allergenic. Photoallergicreactions are not necessarily dose-dependent andshow great variability between individuals. Inanalogy, phototoxicity may be compared withirritant dermatitis. Many phototoxic reactions

may be caused by the formation of free radicalsfollowed by lipid peroxidation and localizedinflammation. In addition to phototoxicity andphotoallergy, light-induced activation of chemi-cals may cause depigmentation, induction of anendogenous photosensitizer or of a disease char-acterized by photosensitization such as lupuserythematodes or pellagra. However, these reac-tions are comparatively rare. The tests to identifyphotoallergenic and phototoxic chemicals areperformed in analogy to the tests for skin sensi-

tization and irritation, respectively.Photoallergenic potential is evaluated by re-

peated application of the test compound on theskin of guinea pigs and exposure of the treatedarea with UV light after every application; theUV treatment should cause a very slight erythe-ma. Several days after this induction phase, thechallenge phase is conducted by treatment with alow dose of the test compound together with UVlight. Phototoxic reactions can usually by ob-served after the first exposure to the test com-

pound together with UV light; in addition toguinea pigs, mice and rabbits are also utilizedin tests for phototoxicity.

7. Reproductive and Developmental

Toxicity Tests

The general objectives of reproductive and de-velopmental toxicity testing are to establishwhether exposure to the chemical may be asso-ciated with adverse effects on reproductivefunction or capacity, and whether administra-tion of the substance to males and/or femalesprior to conception and during pregnancy andlactation causes adverse effects on reproductivefunction or capacity. Another focus of thesestudies are induction of nonheritable adverseeffects in the progeny and whether the pregnantfemale is potentially more susceptible to generaltoxicity.

Reproductive and developmental toxicity is avery broad term including any adverse effect onany of the following aspects:

1. Male or female sexual structure and function(fertility)

2. Development of the new organism throughthe period of major organ formation, organo-genesis (embryotoxicity and teratogenicity)

3. Development of the new organism during theperi- and postnatal periods

The field of reproductive toxicology has be-come increasingly important with the recogni-tion that viral and bacterial infections andxenobiotic chemicals can produce severe andirreversible defects in the offspring. The firstreports on malformations due to rubella virusinfections, ionizing radiation, hormones, dietarydeficiencies, and chemicals appeared in the1930s and 1940s, but the potential impact ofreproductive toxicity on public health was only

recognized more than two decades later. In 1960,a large increase in newborns with specific limbmalformations, which are rarely seen otherwise,was recorded in Germany and in other parts of theworld. One year later, the sedative/hypnotic drugthalidomide was recognized as the causativeagent. The thalidomide epidemic resulted in10 000 malformed children and subsided afterthe drug was withdrawn from the market at theend of 1961. One important consequence of thethalidomide disaster was the introduction of re-

quirements for the testing of potential new drugsfor reproductive toxicity. All test batteries re-quired include detailed tests for reproductive



15/42

toxicity to prevent a repetition of the thalidomidedisaster.

The first specific reproductive toxicity test tobe conducted is usually the two-generationstudy. which should be initiated after the rat90-d subchronic repeated-exposure study, since

the results obtained may provide informationnecessary for selecting dose levels for the two-generation study. Additionally, repeated-expo-sure studies that can provide information rele-vant to reproductive toxicity should be used inthe design of the two-generation study. Forexample, the observation of neurological effectsmay indicate the need to evaluate developmen-tal neurotoxicity.

The first developmental toxicity study is nor-mally performed after completion of the two-generation study. The design of the developmen-tal toxicity study should use all informationderived from the repeated-exposure and two-generation studies, in particular doseresponserelationships and information on maternal toxic-ity. The preferred species for the two-generationstudy is the rat; the necessity of a developmentaltoxicity study in the rabbit is dependent on theoutcome of the first study.

7.1. Fertility and General

Reproductive Performance

Segment I experiments are usually conductedin rats (20 animals of each sex per dose) withthree doses of the test chemical, most oftenadministered with diet. The treatment must notcause general systemic toxicity in the parentalorganism; therefore, in dose selection the doselevels are chosen according to observations in

studies of subacute and subchronic toxicity,which are usually performed before testing forreproductive toxicity. Young adult male ratsare treated for 6080 d prior to mating to covera whole period of spermatogenesis. Female ratsare pretreated for 14 d to cover three estrouscycles. Treatment of both sexes is continuedduring the mating period and that of femalesthroughout pregnancy. Half of the females aresacrificed just before term, and the numbers ofresorbed and dead fetuses as well as structural

abnormalities in the developed fetuses are as-sessed. In the United States, pregnancy is inter-rupted midterm in half of the females. Treat-

ment of the remaining females is continuedthrough parturition and lactation until weaningof the newborns, usually 21 d after birth. Theyoung animals (F1generation) are reared with-out receiving the test compound until sexualmaturity, when their fertility is assessed. Dur-

ing the rearing period, the development of theyoung animals is monitored with cage-sidecarefully clinical observations. If an adverseeffect on fertility, pregnancy, or developmentof the offspring is observed, it is necessary toevaluate whether the effect is due to toxicity tothe male or female reproductive system or both.This information can be obtained by separatemating of treated males with untreated femalesand vice versa.

Furthermore, evaluation of toxic effects onthe male and female reproductive systems withspecific test systems may be required. The effectof the test compound on male reproductiveperformance may be evaluated by monitoringmating behavior (e.g., frequency of copulation).Structural and functional impairment of themale reproductive organs is assessed by con-ducting gross pathology and histology of thetestes and sperm analysis (viability, motility,and morphology). Histological examination of

the ovaries plays an important role in assess-ment of toxic effects on the female reproductivesystem.

7.2. Embryotoxicity and

Teratogenicity

Segment II studies assess adverse effects duringthe period of organogenesis. Xenobiotics inter-fering with the developing organism during this

extremely sensitive period may cause severe andirreversible structural malformations. Thesestudies are carried out in two species, usuallyrats (20 per dose) and rabbits (10 per dose); inmost cases, two dose levels and an untreatedcontrol group are included. Pregnant animals aretested during the period of organogenesis: days 6to 15 for rats and 6 to 18 for rabbits. The fetusesare delivered by cesarean section one day prior tothe estimated time of delivery: day 21 for rats andday 31 for rabbits. The main reason for avoiding

natural delivery is to prevent loss of deformed ordead fetuses by cannibalism, which happens inrodents and rabbits. The uterus of the maternal



16/42

animal is excised, weighed, and examined forimplantation sites and resorbed fetuses. The pupsare weighed, and one-half of each litter is usuallyexamined for skeletal defects and the remainingone-half for soft-tissue defects.

7.3. Peri- and Postnatal Toxicity

For segment III studies, treatment of pregnant ratswith three dose levels (1012 animals per dose)begins on day 16 of gestation and is carried onthrough delivery and lactation, until weaning of theoffsprings, normally on day 21 postpartum. Treat-ment during parturition and lactation is also per-formed in segment I studies; however, segment IIIstudies may offer additional information sincehigher doses can be used than in segment I. Theperi- and postnatal segment evaluates effects onbirth weight and survival as well as developmentofthe offspring in the postnatal period. However,extrapolation of results of segment III studies fromthe rat to the human situation should be performedwith care and consideration of the specific circum-stances in each species. In contrast to the humansituation, where in addition to the mother the socialenvironment takes care of the newborns, young

rats depend completely on the functional integrityof the maternal organism. Furthermore, the organ-ism of newborn rats is significantly less mature bythe time of delivery than that of newborn humans.Hence, treatments that affect the maternal organ-ism simply by causing sedation or fatigue maysignificantly impair the development of newbornrats in the first 21 days of life. Such effects shouldnot be automatically interpreted as relevant forpostnatal toxicity in humans.

7.4. Multigeneration Studies

Multigeneration studies assess the cumulativeeffects of continuous application of the testchemical on reproduction and development dur-ing two or three generations. This applicationmode is relevant for long-term exposure to che-micals in the environment, such as pesticideresidues in food or contamination of drinkingwater with agricultural chemicals or nonbiode-

gradable solvents.The two-generation study is a general test

which allows evaluation of the effects of the test

substance on the complete reproductive cycleincluding libido, fertility, development of theconceptus, parturition, postnatal effects in bothdams (lactation) and offspring, and the repro-ductive capacity of the offspring. The two-gen-eration study is preferable to the one-generation

study because the latter has some limitationregarding assessment of post-weaning develop-ment, maturation, and reproductive capacity ofthe offspring. Thus, some adverse effects suchas oestrogenic- or antiandrogenic-mediated al-terations in testicular development may not bedetected. The two-generation study provides amore extensive evaluation of the effects onreproduction because the exposure regime cov-ers the entire reproductive cycle, permitting anevaluation of the reproductive capabilities ofoffspring that have been exposed from concep-tion to sexual maturity. The prenatal develop-mental toxicity study only provides a focusedevaluation of the potential effects on prenataldevelopment.

Three dose levels are usually given to groups of25 female and 25 male rats shortly after weaning atdays 30 to 40 of age. In the multigeneration study,these rats are referred to as the F0generation. TheF0 generation is treated throughout breeding,

which occurs at about 140 d of age, and the femaleanimals also during pregnancy and lactation.Hence, the offspring (F1 generation) has beenexposed to the test compound in utero, via thematernal milk, and thereafter in the diet. In manyprotocols, the F1 generation is standardized toinclude certain numbers of animals, e.g., eightanimals per litter. In analogy to the F0generation,the F1generation is bred at about 140 d of age toproduce the F2generation. In some of the parents(F0and F1generations), gross necropsy and histo-

pathology is conducted with greatest emphasis onthe reproductive organs. In addition, necropsy andhistopathology are carried out in all animals dyingduring the study.

The percentage of F0 and F1 females thatbecome pregnant, the number of pregnanciescarried to full term, the litter size, and numberof resorptions, stillborns, and live births arerecorded. Viability counts and pup weights arerecorded at birth, and at days 4, 7, 14, 21 and28 of age. With these data, the following

parameters are calculated for assessment ofthe long-term reproductive toxicity of the testcompound:



17/42

Fertility index % Number of pregnancies

Number of matings 100

Gestation index % Number of litters

Number of bred females 100

Birth index % Number of pregnancies resulting in live off spring

Number of pregnancies 100

Viability index % Number of animals alive at day 4 after birth

Number of new borns 100

Lactation index % Number of animals alive at day 28 after birth

Number of animals alive at day 4 after birth 100

7.5. The Role of Maternal Toxicity in

Teratogenesis

If an agent with selective developmental toxicityis administered throughout the organogenesisperiod (days 6 to 15 in the rat), identifying themost sensitive target organs becomes difficult. Inaddition, teratogenic effects may be masked byembryolethality with repeated dosing during theorganogenesis period. Developmental toxicity inthe form of increased resorption and decreasedfetal body weight is generally accepted to occur

at maternally toxic dose levels. The role ofmaternal toxicity in causing congenital malfor-mations, however, is not clear. Doses causingmaternal toxicity, as indicated by reduced ma-ternal body weight, clinical signs of toxicity, ordeath, commonly cause reduction in fetal bodyweight, increased resorption, and rarely, fetaldeaths. Three patterns of association betweenmaternal toxicity and malformations can be ob-served: (1) for some compounds, maternal tox-

icity is not associated with malformations; (2) forothers, maternal toxicity is associated with adiverse pattern of malformations, often includingcleft palate; and (3) the maternal toxicity of stillothers is associated with a characteristic patternof malformations.

Compounds in the second category are themost difficult to classify in terms of teratogenicpotential. Cleft palate is the principal malforma-tion resulting from food and water deprivationduring pregnancy in mice; however, cleft palate

is also a malformation specifically induced inmice by a number of teratogens, most notably theglucocorticoids, without apparent maternal toxi-

city. Complete determinations of food and waterconsumption, maternal body weights, and im-pairment of the maternal organism are necessaryto distinguish between cleft palate caused by theteratogenic effect of a chemical on the embryoand that resulting from systemic maternal toxici-

ty, which secondarily affects embryonic devel-opment. The association of maternal toxicitywith major malformations, such as exencephalyand open eyes, is not generally accepted, al-though most investigations agree that maternaltoxicity can cause minor structural abnormalitiessuch as variants in the ribs.

7.6. In Vitro Tests for Developmental

Toxicity

Models to elucidate the mechanism of embryo-genesis have been under development for severaldecades, and therefore developmental toxicologyis a fields in which alternative methods to animalexperimentation are available. However, be-cause of the complicated, multistep nature of thedevelopment of a new life, none of the in vitrosystems presently available can replace the ani-mal tests. In vitro tests rather serve for screening

purposes, i.e., to preclude the extensive tradition-al whole-animal test protocol for compoundswith marked toxicity on reproduction and devel-opment. The existing alternative test systems fallinto six groups: lower organisms, cell-culturesystems, organ-culture systems, whole-embryocultures, embryos, and others (Table 7). Sincenone of the in vitro methods is sufficiently vali-dated for a set of compounds for which the effectson humans or animals are known and the field ismuch too extensive to be comprehensively re-

viewed here, the reader is referred to two com-prehensive reviews of this field [9].

8. Bioassays to Determine the

Carcinogenicity of Chemicals in

Rodents

Despite the many available short-term in vivoand in vitro tests to determine the genotoxic and

carcinogenic potential of chemicals and the vastamount of literature on the subject, the lifelongcarcinogenicity bioassay remains the main



18/42

instrument for reliable evaluation of the carcino-genic properties of a xenobiotic. The principalguidelines for performing bioassays were estab-lished some 25 a ago by the U.S. National CancerInstitute and have essentially been adopted withslight alterations by all regulatory authorities(Table 8).

A number of factors may interfere with theanalysis and interpretation of data from animalcarcinogenicity studies. A variety of statisticaltechniques has been developed to adjust for con-founding factors and to estimate confidence inter-vals and significance of results. Significance testsare used to assess neoplastic response in treatedgroups as compared to control groups or historicalcontrols (cancer incidence in the identical strainand species observed in control groups for othercancer bioassays under identical housing condi-tions in the same facility). To estimate the abso-lute cancer risk posed by a specific chemical,background or spontaneous cancer incidences(induction of neoplasms not related to the admin-

istration of the test chemical) must be well de-fined. In general, high background incidences ofcancer such as liver cancer observed in specificstrains of mice requires larger number of animalsin the treatment groups to detect increases incancer incidence induced by the administrationof the test chemical and to obtain statisticallysignificant results. The demonstration of a doseresponse curve for the cancer incidence in groupsof animals treated with different doses of thecarcinogen will increase the confidence in posi-

tive results of an animal cancer bioassay. Thesame holds for identical results observed withcancer as an endpoint in an independent study.

Table 7. In vitro test systems for developmental toxicity

Group Test system/organisms End points monitored

Lower organisms and

small animals

sea urchins growth

drosophila

trout, medaka (fish species)

plania

brine shrimp

animal virus

Cell culture pregnant mouse and chick

lens epithelial cell

protein synthesis

avian neural crest cell differentiation

neuroblastoma cell differentiation

Organ culture frog limb regeneration

mouse embryo limb bud morphological and biochemical

differentiation, toxicity

metanephric kidney organ

culture from day 11 mouse embryos

morphological and biochemical differentiation

Whole embryo cultures chick embryo embryotoxicity, malformations

frog embryo teratogenesis assay lethality, no observed-effect-level, developmentstage, attained growth, motility, pigmentation,

gross anatomical malformations

rat embryo culture

(postimplantation embryo)

viability, growth and macromolecular

content, gross structural and histological abnormalities

Table 8. Basic procedures of rodent carcinogenicity bioassays

Species rat (Fischer 344, Sprague

Dawley, Wistar)

mouse (B6C3F1, CD)

Age of animals

at the beginning

4 to 6 weeks (shortly after weaning)

Number of animals 50 per sex per dose for

carcinogenicity

10 20 for additional studies

during the course of experiment

Dose at least three doses and vehicle control

maximum tolerated dose

intermediate dose

nontoxic dose

Duration 24 months

Application gavage, in feed, drinking water,

inhalation (only if absolutely necessary)

Toxicologic all animals: gross necropsy

pathology weight of all important organs

histopathology of all tissues (ca. 40)and all tumors and preneoplastic lesions

by two independent pathologists



19/42

However, despite the importance of animalcancer bioassays for characterizing chemicalcarcinogens, this approach has been criticizedrecently (see also ! Toxicology, 3. Evaluationof Toxic Effects, Section 2.4.4). Due to theinfluence of rodent carcinogenicity assay on the

development of new chemicals and pharmaceu-tical drugs, the pros and cons of this type of assayare discussed in depth in the following.

Practically all chemicals identified as humancarcinogens produce tumors in the rodent bioas-say. Hence, the test has a very good predictivevalue, and every chemical exerting carcinoge-nicity in rodents should be handled as a potentialhuman carcinogen. Considering the many newchemicals developed each year, two major dis-advantages of the life-long assay are its high costand long duration. A two-year gavage study inonly one species amounts to approximatelye 106

and takes 34 years or longer for completeevaluation. For optimized evaluation of carcino-genic properties, it is important to use animalspecies that are closest to humans with regard tobiotransformation and toxicokinetics of the testcompounds. However, for practical and financialreasons, only rodents can be used. Long-termcarcinogenic bioassays in dogs or primates, for

example, require seven to ten years for comple-tion and are much more costly.The application of the maximum tolerated

dose (MTD) in rodent bioassays has been thesubject of much controversy [1016], but theexperimental design limitations of in vivo studiesmake the application of high doses necessary. Forexample, if a specific dose of a chemical causes a0.5% increase in human cancer incidence, thiswould result in several hundred thousands ofadditional cancer cases in a country such as

Germany each year and would thus definitelypose an unacceptable risk. However, the identi-fication of this 0.5% increase in cancer incidencewith statistical confidence in the rodent bioassaywould require a minimum of 1000 animals,provided the incidence of spontaneous tumorsis zero. Therefore, there seems to be generalagreement that the use of the MTD, althoughnot an optimum solution, is necessary for riskassessment. According to the U.S. National Can-cer Institute, the MTD is defined as the highest

dose that can be predicted not to alter theanimals normal longevity from effects otherthan carcinogenicity. In practical terms MTD

is the dose which, in the subchronic three-monthstoxicity study causes not more than a 10% weightdecrement as compared to the control groups anddoes not produce mortality, clinical signs oftoxicity, or pathological lesions other than thosewhich may be related to a neoplastic response

that would be predicted to shorten an animalslife span. As stated above, the MTD is deter-mined in the preliminary three-month studies onsubchronic toxicity, where it fulfills the aboverequirements. Ideally, this is exactly what shouldalso happen in the 24-month carcinogenicitystudy. However, due to cumulation of toxiceffects and/or alterations in toxicokinetics of thexenobiotics during the study, for example, induc-tion of toxification or detoxification pathways byapplication of the xenobiotic in high doses, theMTD dose group often shows reduced survivalrates in the life-long bioassay. This may invali-date the study, that is, make it inadequate forevaluation of the carcinogenic potential. Indeed,this is not a rare event in carcinogenicity studies.The opposite effect may also occur: due to tox-icokinetic differences between the three- and 24-month studies, the MTD chosen may turn out toolow in the long-term bioassay. In spite of all theseproblems and because of the absence of a satis-

factory alternative solution, the use of the MTD iscurrently the only method to compensate for thefact that in relation to the human populationexposed to potential carcinogens, the numbersof rodents used in the carcinogenicity bioassayare extremely low. The legitimate argumentagainst the MTD is that any chemical given ata sufficiently high dose level will induce adverseeffects. This understanding, which is beyonddispute in toxicology, has been tentatively gen-eralized by several scientists in recent years by

the notion that carcinogenic effects obtained atthe MTD may exclusively result from target-organ toxicity, and the increased cell prolifera-tion may contribute to tumor formation by in-creasing the rate of spontaneous mutations, sinceDNA replication does not take place with 100%fidelity. Furthermore, during increased cell turn-over, the time available to repair DNA damage isreduced, so that an increased number of damagedDNA sites may be converted to heritable muta-tions. Although this may be the mechanism

underlying the carcinogenic effects of some non-genotoxic chemicals, it can not be generalized toevery tumor observed at the MTD. Toxicity and



20/42

cell proliferation do not necessarily result intumor formation. Table 9 summarizes the impor-tant differences between rodent bioassays andhuman exposure to carcinogens.

In addition to these problems, which are inher-ent in the bioassay procedure, the evaluation of the

toxicological pathology has repeatedly become anissue of debate, since differences in evaluationsbetween pathologists are frequent. This does notnecessarily indicate incompetence of one of thepathologists. The different evaluations may be theresult of difference in terminology. Also, some-times evaluations are conducted years apart, and inthe interim period understanding of the pathogen-esis of lesions may have changed. Thus, even thesame pathologist may not come to the same con-clusion when reevaluating tissue slices severalyears after the first examination.

Due to the uncertainties of rodent bioassaysand the extremely high costs and personnel re-quirements, a multitude of short-term tests hasbeen developed in recent years. These tests aim topredict the carcinogenic potential of chemicals.Most of these in vitro tests are based on damageto the genetic material (genotoxicity) by thechemical or its metabolites. Genotoxicity is with-out doubt the field in toxicology with the best

established and validated in vivo and in vitroshort-term tests.

9. In Vitro andIn Vivo Short-term

Tests for Genotoxicity

Genetic toxicology a comparatively new field ofresearch that has rapidly grown since the 1960s,deals with mutagenicity and genotoxicity.

Mutagenicity is the induction of permanent

transmissible changes in the genetic material ofcells or organisms. Changes may involve a singlegene or gene segment, a block of genes, or whole

chromosomes. Effects on whole chromosomesmay be structural and/or numerical.

Genotoxicity is a broader term and refers topotentially harmful effects on genetic materialwhich may not be associated with mutagenicity.Thus, tests for genotoxicity include systems

which give an indication of damage to DNA (nodirect evidence of mutation). End points deter-mined here are unscheduled DNA synthesis(UDS), sister-chromatid exchange (SCE), DNAstrand breaks, formation of DNA adducts andmitotic recombination.

Evidence has increasingly accumulated thatmany carcinogens are also mutagenic, and a largenumber of short-term in vitro and in vivo testswere developed as predictive tools. Most of thesetests are well validated and aim to assess geno-toxic properties of chemicals. Today, the majori-ty of potential carcinogens are first identified asmutagens or chromosome-damaging agents inshort-term tests and subsequently as carcinogensin the rodent carcinogenicity bioassay. The nu-merous in vivo and in vitro assays can be cate-gorized into two major groups:

1. Short-term tests detecting gene mutations2. Short-term tests detecting structural and/or

numerical chromosomal aberrations

A number of these in vitro test procedureshave gained regulatory acceptance for toxicolo-gy testing (Table 10).

In vitro tests to detect gene mutations can becategorized into two groups: microbial and mam-malian cell assays. An important step in thehistory of modern genetic toxicology was thedevelopment of genetically precisely definedstrains of bacteria carrying mutations in particu-

lar genes coding for enzymes involved in thebiosynthesis of amino acids. Among the numer-ous tests, the most widely used and best validatedis the assay in Salmonella typhimurium devel-oped by AMESet al.

9.1. Microbial Tests for Mutagenicity

9.1.1. The Ames Test for Bacterial

Mutagenicity

The most common method to detect mutationsin microorganisms is selecting for reversions in

Table 9.Some important differences between carcinogenicity tests in

rodents and human exposure to potential carcinogens

Rodent carcinogenicity test Human exposure

High doses (usually) low doses

Continuous exposure (often) infrequent or not

regular exposure

Single compound, no

interactions

simultaneous exposure to several

carcinogenic chemicals,

interactions probable

Homogeneous population heterogeneous population



21/42

strains that have a specific nutritional (i.e.,amino acid) requirement differing from wild-type members of the species; the tester strainsare auxotroph for this particular nutrient.The Salmonella typhimurium mutant strainsdeveloped by AMES can not synthesize histi-dine, because each strain carries one of a num-ber of mutations in in the operon (group ofgenes) coding for histidine biosynthesis [17,18]. The result of this mutation is that the testerstrains can not grow and form colonies inhistidine-free medium. The mutation may re-vert to the wild-type sequence or a functionallyequal sequence either spontaneously (a rareevent) or by exposure of the tester strains togenotoxic compounds. The revertant coloniesare, like the wild-type bacteria, capable ofsynthesizing histidine and form colonies inhistidine-free medium. For the common tester

strains, the DNA sequence at the site of theoriginal mutation in the relevant histidine genehas been determined. According to the type ofmutation leading to the inability to synthesizehistidine, the strains can be categorized into twogroups: base-substitution and the frameshiftstrains. The difference between these two cate-gories can be illustrated with the followingsentence, in which each letter represents a DNAbase, each word a triplet coding for an aminoacid, and the whole sentence a gene coding for a

protein (i.e. enzyme of histidine biosynthesis).The correct sentence represents the gene in thewild-type strain.

THE NUN SAW OUR CAT

EAT THE RAT

original sentence (wild-type)

Base-pair substitution:

THESUN SAW OUR CAT

EAT THE RAT

missense mutation coding

for a wrong amino acid

THE NSN SAW OUR CAT

EAT THE RAT

nonsense mutation resulting

in interruption of gene transcription

Frameshift mutation:

THE NUN SAS WOU RCA

TEA TTH ERA T

1-frameshift mutation

This example illustrates that a base substitutioncan be reverted by another base substitution andin analogy a frameshift mutation can be revertedby another frameshift mutation. Hence, the Amestest not only provides information on the geno-toxic potential of the test compound but also onthe nature of the DNA damage.

Genetic Makeup of the Salmonella typhi-

muriumStrains Used in the Ames Test. His-tidine mutations of the tester strains. The base-pair substitution strains can be categorized intotwo families.

1. The Salmonella typhimurium strains TA100and TA1535 carry the sequence CCC (forleucine) instead of the wild-type sequenceCTC (for proline). These missense mutationsmay be efficiently reverted by mutagens withalkylating properties.

2. The second group of base-pair substitutionstrains carry a nonsense mutation (TAA in-stead of CAA). These strains (TA2638,

Table 10. OECD guidelines on genetic toxicology testing and guidance on the selection and application of assays

No. Title Original adoption Updated

471 Bacterial Reverse Mutation Test 26 May 1983 21 July 1997

472 Genetic Toxicology:Escherichia coli, Reverse Assay 26 May 1983 21 July 1997*

473 In Vitro Mammalian Chromosome Aberration Test 26 May 1983 21 July 1997

474 Mammalian Erythrocyte Micronucleus Test 26 May 1983 21 July 1997

475 Mammalian Bone Marrow Chromosome Aberration Test 4 April 1984 21 July 1997

476 In Vitro Mammalian Cell Gene Mutation Test 4 April 1984 21 July 1997

477 Genetic Toxicology: Sex-Linked Recessive Lethal Test inDrosophilia melanogaster 4 April 1984

478 Genetic Toxicology: Rodent dominant Lethal Test 4 April 1984

479 Genetic Toxicology: In Vitro Sister Chromatid Exchange assay in Mammalian Cells 23 Oct. 1986

480 Genetic Toxicology:Saccharomyces cerevisiae, Gene Mutation Assay 23 Oct. 1986

481 Genetic Toxicology:Saccharomyces cerevisiae, Mitotic Recombination Assay 23 Oct. 1986

482 Genetic Toxicology: DNA Damage and Repair, Unscheduled

DNA Synthesis in Mammalian Cells In Vitro

23 Oct. 1986

483 Mammalian Spermatogonial Chromosome Aberration Test 23 Oct. 1986 21 July 1997

484 Genetic Toxicology: Mouse Spot Test 23 Oct. 1986

*Date of deletion (method merged with TG 471).



22/42

TA100) may detect mutations induced byradicals or oxidizing agents such as hydrogenperoxide and reactive oxygen matabolites.

The commonly used frameshift strains TA98and TA1538 carry a 1 frameshift mutation near

a GCGCGCGC sequence. This strain may beused to detect frameshift mutagens such as poly-cyclic aromatic hydroxycarbons, certain aroma-tic amines, and certain aromatic nitro compounds.

In addition to the mutation in one of the genesof histidine biosynthesis, the Ames strains carryadditional gen

toxicology (2) - ullmann

Documents