In Vitro Toxicity Testing Protocols

CHAPTER1

Alternative

Method

of Assessing

Toxicity

Chris K Atterwill 1. Introduction The safety assessment of new chemicals and pharmaceuticals and the incorporation of these data into a human risk assessmentpackage requires a large number of expensive, regulated tests in animal species including, in some cases, nonhuman primates (I-3). There are currently a wide range of animal replacement alternative opportunities in industrial chemical and drug development (Table 1) (4,.5). Although in vitro methodology has long been used as a basic laboratory tool for defining biological and toxicological processes in different cellular systems, application and use alternatives in industrial compound discovery (i.e., research and development) is slow. Coupled with a relatively low innovation rate in the design of new in vivo tests for the toxicological and safety evaluation of new compounds, this has both ethical and resource implications. From a basic scientific viewpoint in vitro toxicological models have made important contributions in elucidating, e.g., the cellular and molecular mechanisms involved in apoptotic and necrotic cell death and in carcinogenesis and the role of mediators, such as free radicals and oncogenes, in these processes (6). The “take-up” of in vitro systems in toxicity testing is, however, now gradually occurring in industry and resources are being invested slowly into the area for both ethical and scientific reasons. A lot of emphasis currently lies on the ethical question as public sensitivity to animal use in safety testing increases. This has occurred most significantly in the case From Methods m Molecular Wology, Edited by S. O’Hare and C. K Atterwlll

Vol 43 In Wtro Tox/c/ty Testmg Protocols Copyright Humana Press Inc , Totowa, NJ

1

Atterwill Table 1 Ammal Replacement Alternatives Improved storage, exchange, and use of information, so that unnecessary repetition of experiments on animals could be avoided. Maximum use of predictions based on physical and chemical properties of molecules Mathematical modeling of quantitative structure-activity relationships. Molecular modeling and the use of computer graphics. Mathematical modeling of biochemical, physiological, pharmacological, and toxlcological systems and processes. The use of lower organisms not protected by legislation, including invertebrates, plants, and microbes.a The use of embryonic and larval vertebrates before they reach the developmental stage at which time they become protected by law a The use of in vitro methods including subcellular fractions, perfused organs, tissue shces, and cell suspensions; and cell organotypic cultures * Human studies, including epidemiology, postmarketing surveillance, and the properly regulated use of volunteers.a % vitro areas

of cosmetic and toiletry safety assessmentand the safety testing of chemical intermediates used in drug synthesis, which have classically used the controversial Draize, guinea pig, and rabbit tests, and for which valid alternatives now exist. Furthermore, in drug development the classic LD,, tests for acute toxicity have not only been partially replaced by the “fixed dose” procedure, but much work is being carried out by Multicentre Evaluation of In Vitro Cytotoxicity (MEIC) on new and rapid in vitro predictors for acute cytotoxicity using human cell lines. 2. Compound Development Utilizing Alternative Test Models The development and registration of both new drugs and chemicals currently requires the submission of a large battery of in vivo toxicity data derived from a number of species for the “risk-assessment” process (1-3). The compilation and validation of the animal batteries has been largely empirical over the years and, although being fairly well-proven for detecting toxic phenomena in animal species, can have limited predictive value for human safety assessment for some of the reasons listed in Table 2 (a,b). There are large lists of chemicals with good animalhuman toxic correlations, but equally, lists of compounds exist that have been withdrawn from the market because of the increasing number of clinically reported adverse reactions (ADRs).

Alternative

Method of Assessing Toxicity

3

Table 2a Reasons for Incorrect Predictions from Animal Toxicity Studies False negative responses Effect not looked for Use of inappropriate assay methoda Improper timing of assay Insufficient target organ exposure0 Incorrect evaluation of an experimental finding0 Failure to consider absence of preexisting pathological condition Inability to identify and measure adverse effect0 False negative and false positive responses Failure to consider differences in metabolic activation and detoxrfrcationa Disregard of anatomrcal and physiological differences between species“ Inability of animals to express human-specific reaction pattern9 Table 2b Conversion Factors for Predictmg from Animal Studies to Individuals Toxic effects Molecular Subcellular Cellular Tissue Organ Whole animal

From experiment to humans! Bioavailability Pharmacokinetics Protein bindinga Metabolism0 Dose and time Concentrationa Receptor sensitivitya Anatomical characteristics Physiological characteristicsa Repau mechanismsa Species-specific responsea Mechanisms of actiona

at Risk

From humans to individuals at risk Healthy men and women Babies and children Pregnant women Elderly People at genetic risk Diabetics People with infections Immunosuppressed individuals Alcohol or drug abusers Smokers Patients with organic disease Occupationally exposed indivrduals

OAreas in which alternatlves tests can have an impact. (Data taken from table compiled by G Zbinden, personal communication.)

It is also well accepted that during the development of a “safe” and effective pharmaceutical (or agrochemical agent) there is a large attrition rate throughout the safety assessment process with massive financial implications. When one superimposes on this the varying worldwide regulatory requirements for administration of a new compound to humans, one can see a number of important reasons for developing in vitro toxicity testing systems either as prescreens or as adjuncts to current in vivo test packages. This, together with harmonization of the cur-

4

Atterwill

rent regulatory requirements, will hopefully improve the sensitivity and specificity of animal tests (3).

3. Summary

of Gains

1. Financial gains: Reduce attrition rate by use of prescreening strategies prior to full regulatory animal study packages and develop better predictors of human toxic phenomena. 2. Scientific gains: Describe more effectively the lesions seen m vivo from regulatory studies and give better definition of safe concentration and clinical dose. Define “direct toxicant effects” on target organs as opposed to indirect effects, and give human reaction mdication using primate cells. “Adjunct” studies will improve sensitivity and specificity of animal studies. 3. Ethical gains: Implementation of 3R strategy (reduce, refine, replace amma1testing). Supplement and reduce current m vivo toxicity tests, particularly those involving distressing procedures and the use of large mammals and primates.

4. Principles, and Types of In Vitro

Aims, Toxicity

Testing

The general advantages and disadvantages of in vitro testing for toxicity are described in Table 3. 5. Validation The successful use and industrial and regulatory acceptance of a new in vitro test model depends on a certain degree of validation (7,s). Detailed validation is generally required if the in vitro test is to replace an in vivo test or is to be used as a prescreen where financial factors are critical. However, when the result from an in vitro procedure is submitted to a regulatory authority along with that from an in vivo test package in order to explain a lesion, then full validation is not formally necessary as long as good laboratory practice (GLP) procedures have been adopted in the execution of that test and the test and endpoints have acceptable relevance. For example, the gradual replacement of the Draize procedure by tests such as the EYETEX or SKIN2 tests has required extensive validation of that test. So, for example, would the use of a prescreen for a new immunosuppressant with adrenal toxicity where there were a limited number of available backup compounds or they were very expensive to synthesize. On the other hand, if a drug company were trying to confirm to the Food and Drug Administration (FDA) that a particular drug had no

Alternative

Method of Assessing Toxicity

5

Table 3 Advantages and Disadvantages of In Vitro Systems for Detection of Xenobiotic-Mediated Toxicity Advantages (general) Detect direct (vs Indirect) toxrc/cytotoxic effects on target organ. Use controlled conditions of exposure-concentration of toxrcant known. Study parent compound vs metabolite (rt liver S9 metabolizing fractions from different species) Study effects on cells vs subcellular organelles. Has resource implications (time, animals, number of compounds tested). Disadvantages (general) Systems not always representative of mature, differentiated target organs (cells dedifferentiated in cell lines?). Xenobiotic concentrations not representative of those in vivo (e.g., plasma protein bindmg factors). Biological barriers absent (e.g., blood-brain barrier in neural cultures of CNS). Metabolite profiles differ Difficulty of culturing/maintaining certain target organs in vitro.

direct neurotoxic effect in humans despite some minor behavioral changes detected in the rat, then submission of data from cultured human exposed neurons would be acceptable, probably without full validation of that particular culture model. Accepted validation criteria for an in vitro system are described in Table 4 and include definition of the specificity, sensitivity, and predictive value of such a test. The validation parameters are obtained by conducting blind validation trials. It is my belief that the requirements for in vitro test validation can be summarized as follows: 1. Full validation involving multicenter coordination: replacement.

To support in vrvo test

2. In-house validation: In vivo test reduction, to support,e.g., development

of a prescreen, 3. Limited inter- or intralaboratory: To supportrefinement or supplementa-

of in vivo toxicity test data,to develop adjunct tests;use of the model to define basic scientific toxic phenomena.

tion

It is believed that validation should not be used as an excuse for nonadoption, nondevelopment, or nonacceptance of in vitro methodology. Sadly, and largely for political reasons, this scenario still exists in many companies and countries.

Atterwill Table 4 Validation Criteria for In Vitro Test Models A formal validation study will require: Careful selection of chemicals (mimmum 20-40?) Use of chemical pairs Toxicological classification from m vivo data “Blind” testing to be performed Method for evaluation of test outcome (absolute values) Method for evaluation of test performance Methods of expressing test performance Other points Are there good in vivo comparative data for compounds chosen? Which kind of in vivo assay trying to emulate/evaluate

in vitro?

Agree with collaborating centers in validation trial at beginning who will be organizing and collatmg data.

6. Spectrum of Available In Vitro Toxicity Tests The currently available models in In Vitro Toxicology (Table 5) (5-8) span six main areas: reproductive toxicity, mutagenicity, irritancy testing, immunotoxicity, target organ toxicity (including endocrine and neurotoxicity), and ecotoxicity involving the use of fish, invertebrates, and so on. Within these main areas there are also important subareas. As alluded to above there are various modes in which to operate these tests in an industrial setting and generally the mode predominance varies significantly according to both scientific area and whether or not one is operating in the drug, chemical, or cosmetic industry. For example, a test system might progress from unvalidated use in the fine description of a pathologically identified lesion for a lead development compound, to the subsequent, semivalidated use of this system in a prescreening mode for second-generation drug candidate compounds. Alternatively, the agrochemical industry has developed a tiered in viva/in vitro hierarchical model for the labeling of industrial chemicals as skin irritants. This latter development was performed under the auspices of the British Toxicological Society, showing how the scientific and industrial communities can interact so well on such issues. Here, a chemical for irritancy classification would proceed from tests on isolated skin or cells in vitro to tests in a limited number of animals in vivo depending on negative or positive outcomes in the initial in vitro tests.

Alternative

Method of Assessing Toxicity

7

Table 5 In Vttro Models Currently Available in Toxicology Mutagenicity testing Irritancy testing Reproductive toxicity testing Quality-Structure Activity Relationship (QSAR) Target organ toxicity Immunotoxicity Hemtc system Endocrine toxicity Neurotoxtctty Acutelcytotoxicity testing

7. Recent Successes and Developments in In Vitro Toxicity

Testing

It is refreshing to observe the momentum that is now gathering in this field (see Table 6) and the way in which “in the face of adversity” some tests are being accepted as full replacement alternatives. It is noteworthy to say that a lot of this energy has been provided by the public and by academic research centers. Apart from the mutagenicity test area where many innovations continue to occur, some of the following recent developments in other areas warrant attention. 1. Eyetex, Skmtex, and Corrosrtex tests for eye and skin irritancy (Ropak Corporation Ltd) and the SKIN2 Model (Advanced Tissue Sciences). More recently, vitro.

the Ropak

Solatex

system for predicting

photoirritation

in

2. The use of hepatocyte “couplets” for in vitro investigation of xenobiotic effects on bile flow. Together with measurements of hepatotoxicity and fatty acid accumulation by these cells, rt may now be possible to obtain a

complete hepatoxicological profile in one in vitro model. 3. Luminescent bacteria (Microtox test) for measuring the ecotoxic potential of industrial effluent. 4. More sensitive in vitro toxicity measurementsusing, for example, the mitochondrial MTT test for succurate dehydrogenase activity. This test gives a more sensitive and earlier prediction of toxicity than classical LDH or neutral red measurements.

Atterwill Table 6 Orgamzatlons Involved m the Development of Alternative Testing Bodies for promotion of alternatlve nonanimal testing FRAME-Fund for Replacement of Animals in Medical Experiments ERGAT-European Research Group for Alternative Testing EURONICHE-European group for alternatlve methods for biology teaching CAAT-Center for Alternatives to Animal Testings (Johns Hopkins Medical School, Baltimore, MD) Dr. Hadwen Trust-Nonammal research and testing strategies (UK-based) Societies, conferences, and journals advancing alternative testing PIVT-Practical In Vitro Toxicology conference IVTS-In Vitro Toxicology Society (UK) Scandinavian

Cellular

Toxicology

Society

FRAME Toxicity Committee and Conference TIV-Toxicology In Vitro, Journal ATLA-Alternatwes to Laboratory Animals, Journal Hildegard Doerenkamp and Gerhard Zbinden Foundation for Reahstlc Animal Protection and Scientific Research, Switzerland

5. Measurements of calcium accumulation in single cultured neurons for the measurement of neurotoxicity. 6. Tlered tests involving both simple and organotyplc organ systems; hierarchical proceeds involving a battery of m vitro and in vivo models.

8. Conclusions Industry and academia have come far in developing in vitro alternatives, and bodies such as Fund for Replacement of Animals in Medical Experiments (FRAME), European Centre for Validation in Alternative Methods (ECVAM), and Center for Alternatives to Animal Testing (CAAT) (USA), have simultaneously enhanced public and regulatory awareness (Table 6). The regulatory and industrial acceptance of new alternative tests depends on proper, well-coordinated validation trials at a level befitting the intended use of the alternative test. This has started through FRAME and European Community (EC) initiatives and good examples have been set by the cosmetics and toiletry industries. More commercial “takeup” is still required for these new tests at the toxicological prescreening and in vivo adjunct testing level. Regulatory harmonization of in vivo animal testing is occurring for both ethical and resource reasons. The gradual replacement of the LD,, test by the “Fixed

Alternative

Method of Assessing Toxicity

9

Dose” procedure for acute toxicity testing and the realization that 6 mo chronic testing (I) is sufficient to identify the most important pathology (excluding carcinogenicity) are most welcome changes. The potential risks for humans in adopting alternative toxicity tests are few and benefits great if the data generated is used correctly. The imminent replacement of all in vivo tests is unlikely but in the future may gradually occur. In the meantime in vitro tests will continue to supplement the somewhat “impirical” animal tests for human toxicity. References 1. Volans, G. N., Sims, J., Sullivan, F. M., and Turner, P. (eds.) (1989) Basic Science in Toxzcology, V International Congress of Toxicology (ICTV), Taylor & Francis. 2 Poole, A. and Leslie, G. B , eds. (1989) A Practical Approach to Toxological Investigations, Cambridge University Press. 3. Lumley, C., Parkmson, C, and Walker, S. R. (1992) An international appraisal of the minimum duration of chronic animal toxicity studies. Hum. Exp. Toxicol. 11, 155-162 4. Parish, W. E. and Hard, G. C. (eds.), Toxicology In Vitro. Proceedings of Second International Conference on Practical In Vitro Toxicology 4/5. 5. Atterwill, C. K. and Steele, C. E., eds. (1987) In-Vitro Methods in Toxicology, Cambridge University Press, Cambridge. 6. Walum, E., Stenberg, K., and Jenssen, D. (1990) Understanding Cell ToxicologyPrinciples and Practice, Ellis Horwood, New York, London, Sydney. 7. OECD Environment Monograph No. 36. Scientific Criteria for Validation of In Vitro Toxicity Tests, Sept. 1990. 8. FRAME 21st Anniversary Issue (1990) ATLA, 18.

CHAPTER 2

LLC-RKl

Cell Screening for Nephrotoxicity

David

X&t

J. White and Chris Seaman

1. Introduction In this test, kidney-derived cells are cultured in the presence of test compounds whose cytotoxicity is then determined by the Neutral Red method, and serves as an indicator of potential nephrotoxicity (1). Healthy LLC-RKl cells (an established cell line, ATCC CCL) maintained in culture continuously divide and multiply over time but still retain certain characteristics of kidney cells in vivo (2). Compounds that have a deleterious effect on these cells may, therefore, be considered as potential in vivo nephrotoxins. In this test system, harmful effects on cell viability are determined by monitoring the uptake of the vital dye Neutral Red into the lysosomes of healthy cells. LLC-RKl cells are maintained in culture and exposed to varying concentrations of test compounds, The cultures are incubated for 48 h. The cultures are then rinsed and incubated for 3 h in medium containing Neutral Red that is taken up by viable cells. After rinsing, the dye present in the cell population is liberated and the amount is quantified using a spectrophotometer, in order to obtain an indication of cell number. Comparison of the number of cells in control and test cultures provides an index of cytotoxicity and an indication of potential nephrotoxicity in vivo. The maintenance and culture of a cell line such as LLC-RKl cells is a relatively simple and inexpensive technique. Additionally, LLC-RKl cells exhibit many features in common with kidney cells in vivo. Among these is the unidirectional transport of solutes via the Na+K+ ATPase From. Methods m Molecular Biology, Ed&d by. S. O’Hare and C. K Atterwtll

Vol. 43 In Wtro Tox/c/ty Testing Protocols Copyright Humana Press Inc , Totowa, NJ

11

12

White and Seaman

system. As a result of this, one-way transport “blisters” are formed in the monolayer, a feature in common with primary kidney cells and other cell lines in culture (2). The application of such cultures to determine nephrotoxicity may potentially allow the rapid, highly reproducible testing of many chemicals on a routine basis. There are, however, disadvantages associated with using a cell line in culture. The cells grow rapidly and are nondifferentiating. Additionally, the cells in culture inevitably lose many characteristics of those in vivo. In particular, the loss of xenobiotic metabolizing activity may affect the sensitivity of the cells to certain chemicals and may raise concern when trying to directly extrapolate results to the in vivo situation. 1.1. Neutral Red Uptake Assay Neutral Red is preferentially taken up into the lysosomes/endosomes of the cell. Absorbances obtained using the Neutral Red assayshave been shown to correlate linearly with cell number over the specific optical density range obtained using this method. Any chemical having a localized effect on the lysosomes/endosomes will, therefore, result in an artificially low (or possibly high) reflection of cell viability and cell number. This factor does, however, make the system useful to detect other chemicals that selectively affect the lysosomes, especially when it is used in conjunction with other tests capable of determining cell number (3).

1. 2. 3. 4. 5. 6. 1. 2. 3. 4.

2. Materials 2.1. Equipment 37°C incubator,hurmdified, 5% CO,/95% air. 75 cm2tissue culture flasks. 24-well tissueculture plates. Inverted phasecontrastmicroscope. Spectrophotometer. Hemocytometer. 2.2. Reagents Cell line rabbit kidney LLC-RKl cells. Dulbecco’s formulation tablets,without magnesiumand calcium. PBS Trypsin/EDTA, 1X cone (dispenseas 10 rnL aliquots into universals and storeat -2O’C), Glbco Europe Ltd. (Paisley, Scotland). Eagle’s minimum essentialmedium (MEM) supplementedwith 1% penicillin/streptomycin, 5% fetal calf serumN.B. Omit penicillin and streptomycin if the test compoundis an antibiotic.

LLC-RKl

Cell Screening

Test

5. 10,000 U/mL penicillm/l0,000 pg/mL streptomycin solution in saline Neutral Red stock solution; 100 mL 0.4% w/v Neutral Red in distilled water, filter sterilized. Store at 4°C until required. 6. Neutral Red medium: dilute the stock dye solutron (1 in SO)just prior to use with culture medium to give a final concentration of 50 pg/rnL. 7. Formaldehyde. 8. Glacial acetic acid. 9. Ethanol. 10. Calcium chloride. 11. Trypsin/EDTA solution: 0.5 g trypsin (1:250) and 0.2 g EDTAIL of Puck’s salme A. 12. Ca2+- and Mg2+-free PBS: Make up Dulbecco’s tablets as indicated by supplier. 13. Neutral Red wash solution: 10% CaC12in formaldehyde. 14. Neutral Red resorb solutron: 1% glacial acetic acid, 50% ethanol, 49% distilled water. 15. Test compounds: These should be drssolved in sterile water, ethanol, methanol, or drmethylsulfoxrde (DMSO), as approprrate at loo-fold the required final concentratron. The final solvent concentration should be kept at a constant level of 1% in the culture medium.

3. Methods 3.1. Cell Maintenance 3.1.1. Preparation of Cells for Freezing 1. Count the cells and dilute to l/O.9 of the intended final concentration of l-2 x 106/mL in complete culture medium. 2. Add DMSO to a final concentration of 10% to the cell suspensron immediately prior to adding to the vials. 3. Aliquot 1.8 mL of cell suspension per vial and freeze at a rate of l”C/min in liquid nitrogen, 1. 2. 3. 4. 5.

3.1.2. Thawing and Culture of Cells When required, thaw the cells rapidly in a 37°C water bath to avoid damage owing to the high DMSO concentration. Transfer immediately to a 75 cm2 tissue culture flask containing -30 mL medium (i.e., l-2 x lo6 cells/flask). After 24 h, rinse the culture with 5-10 mL of PBS at 37OC. Add -30 mL fresh medium. Subculture the cells 2-3 times following thawing before using for test purposes.

14

1. 2. 3. 4. 5. 6. 7.

White and Seaman 3.1.3. Subculture of Cells When the cultures approach confluence remove the cells from the dish by trypsinization. (N.B. If the cells are not subcultured or used for test purposes, the medium should be changed every 3-4 d.) Decant the medium and rinse the cultures with 5-10 mL of PBS at 37OC. Add 10 mL trypsin/EDTA (37°C) and incubate at 37°C. Remove the flasks after 20-30 s and examine visually to ensure the cells have begun to detach (i.e., round up). Discard the trypsin/EDTA solution and return the flask to 37°C. After a further 1 mm, examine the cells and if necessary tap the side of the flask to aid detachment. Add 10 mL of complete medium to neutrahze the trypsin activity and spht or use for test purposes.

3.2. Test Procedure 1. After growing up the cells and preparing a cell suspension as described above, remove an aliquot of suspension and count the number of cells using a hemocytometer and dilute to a concentration of lo5 cells/ml medium. 2. Add 1 mL of the diluted suspension to all the wells of a 24-well plate. Incubate overnight to allow adherence and recovery from the trypsm exposure. 1.

2. 3. 4. 5.

3.2.1. Range Finder Prepare the following test chemical concentrations (diluting the stock solutions 1: 100 in medium) immediately before use: a. 0.5,5,50,500, and 5000 pg/rnL. b. 1% solvent control. Remove growth medium and replace with 1 mL of each chemical dilution in the appropriate wells in 24-well plates. Shake the plates gently to ensure an even distribution. Incubate for 48 h at 37°C. Remove the medium and determine the cell number by the Neutral Red assay (see Note 3). From the preliminary results select six concentrations, spanning the range of O-100% cell death, for an accurate determination of cytotoxicity.

3.2.2. Determination of IDzO, ID,, and IDgO 1, Test each chemical concentration m triplicate on three separate occasions. 2. Prepare: a. The appropriate solvent controls. b. Six concentrations of the test chemical.

LLC-RKl

Cell Screening

Test

15

3. Prepare the 24-well plates as before. 4. After overmght incubation, remove growth medium and replace with 1 mL of the test chemical or the control to random wells (thus minimizing bias), but ensuring that a careful note is made of the treatment received by the cells in each well. Shake the plates gently to ensure even distribution. Incubate for 48 h. 5. Estimate the cell number using the Neutral Red assay. 3.2.3. Neutral Red Assay (1) 1. After 48 h, remove the medium from all the wells. Wash gently with PBS. Add 1 mL of Neutral Red medium per well. Incubate for 3 h at 37OC,5% CO2 in a humidified atmosphere. 2. After 3 h, remove the Neutral Red medium. Wash quickly with the Neutral Red wash solution. Add 1 mL of resorb solution to each well. Agitate the plates intermittently for a period of 15-20 min. Transfer the solutions to cuvets and measure their absorbance at 540 nm using the resorb solution as a blank.

3.2.4. Results 1. Determine the mean value for the absorbance of the control cultures and adjust all indivtdual absorbances accordingly. Mean the values for each treatment group and plot graphically. Determine the ID,,, ID,,, and ID,, values from the curve. 2. Mean the ID values from three separate experiments and give the final concentrations expressed as pg/mL or mmol/L. Rank the chemicals for toxicity using the ID,, value (the section of the curve most likely to be linear and subject to least variation).

4. Notes 1. Volatile chemicals tend to evaporate under the conditions of the test, thus the ID,, value may be variable, especially when the toxicity of the compound is fairly low. Chemicals that are unstable or explosive in water are also difficult to test (4J). Neutral Red is preferentially taken up into the lysosomes/endosomesof the cell. Absorbances obtained using the Neutral Red assayshave been shown to correlate linearly with cell number over the specific optical density range obtained using this method. Any chemical having a localized effect on the lysosomes/endosomeswill, therefore, result in an artificially low (or possibly high) reflection of cell viability and cell number. This factor does, however, make the system useful to detect other chemicals that selectively affect the lysosomes, especially when it is used in conjunction with other tests capable of determining cell number.

16

White and Seaman

2. One major drawback of the assay is the precipitation of the Neutral Red dye into visible, fine, needle-like crystals. When this occurs it is almost impossible to reverse, thus producing inaccurate readings. Some chemicals induce this precipitation therefore a visual inspection stage in the procedure 1svery important. 3. If the intensity of the color is too great, it may be necessary to add a further 2 n-L of resorb solution to some wells. If this is required, carry out the procedure for all the wells. References 1, Borenfreund, E. and Puerner, J. A. (1985) Toxicity determined in vitro by morphological alterations and Neutral Red absorption. Toxicol. Lett. 24, 118. 2 Williams, P. D., Laska, D A., Tag, L K., and Hottendorf, G. H. (1988) Comparative toxicities of cephalosporin antibiotics in a rabbit kidney cell line (UC-RKl) Antimicrob. Agents Chemother. 32(3), 314. 3. Riddell, R. J., Clothier, R. H , and Balls, M. (1986) An evaluation of three in vitro cytotoxicity assays. Fd. Chem. Toxic01 24,469-47 1 4 Riddell, R. J., Panacer, D. S , Wilde, S M., Clothier, R. H., and Balls, M. (1986) The importance of exposure period and cell type in in vitro cytotoxictty tests. ATM 14,86-92.

5 Knox, P., Uphill, P. F., Fry, J. R., Benford, D. J., and Balls, M. (1986) The FRAME multicentre project on in vitro cytotoxicology. Fd Chem. Tox. 24,457-463.

CHAPTER3

Preparation and Use of Cultured Astrocytes for Assay of Gliotoxicity Mark R. Cookson, R. McClean, and Vie tor W. Pentreatk 1. Introduction Cultured astrocytes provide a valuable and important system for predictive testing and mechanistic analysis of neurotoxic compounds. The culture procedures allow relatively rapid assessment of different chemicals or their metabolites over a range of concentrations, using cells derived from a restricted source. The use of multiwell plates for the cultural astrocytes means that multiple samples can be analyzed with a high degree of statistical accuracy and the cell environment can be carefully monitored or manipulated for content of nutrients, ions, agonists, antagonists, or modulators. On the other hand, cultured astrocytes are devoid of their normal integrative functions, and the lack of a blood-brain barrier, the absence of neuron-glial metabolic interactions, and metabolism of substances outside the CNS, together with local regional astrocyte heterogeneity and limited survival time (about 3 mo) are potential important shortcomings that require correlative reference to other neuronal, coculture, and in vivo studies. A recent review on astrocyte culture for evaluation of neurotoxic-induced injury can be found in ref. 1. The preparation and use of astrocytes is a well established and documented procedure for which there is general consensus regarding the principal steps (see refs. 2,3). There are, however, many variations in the From- Methods m Molecular B/ology, Edited by: S. O’Hare and C K Atterwill

Vol. 43 In Vitro Toxrcity Testing Protocols Copyright Humana Press Inc , Totowa, NJ

17

18

Cookson, M&lean,

and Pentreath

detailed methodology of, for example, cell separation and media composition, and our descriptions are appropriate for several tests for gliotoxicity. Cell purity of the primary cultures are checked by immunostaining for glial fibrillary acidic protein (GFAP), with typical values in the 9095% positive range. For toxicological studies, subcultured astrocytes are advantageousbecausehigher cell densities and purities may be obtained. Keys to the assay and understanding of gliotoxic mechanisms will lie in the accurate measurement of cell membrane perturbations together with the subsequent intracellular biochemical and metabolic effects. Because a large number of physiological and biochemical properties of astrocytes have been measured in culture systems across different disciplines in neurobiology, a considerable range of potential targets or endpoints are available for neurotoxicological evaluation. Some examples are described in ref. 1. The choice of procedures that may be useful as a preliminary predictive screen is therefore critical and is currently the subject of extensive studies by us. To date we have analyzed cell viability, total cell protein, energy utilization, and membrane integrity as likely key indicators. These assays can be completed relatively quickly, with cell viability and total protein combined in the same experimental procedure and with the energy utilization and membrane integrity also evaluated together by uptake and backflux of radiolabeled 2-deoxyglucose (2-DC). The findings show that these have a high degree of validity with, for example, a variety of gliotoxic substances causing increases in total protein (measured in pg/104 cells) at certain critical concentrations. However, additional valuable procedures will undoubtedly be described in the future. Below we describe convenient methodologies for the preparation of astrocyte cultures and the application of procedures. 2. Materials 1. Poly-L-lysine (mol wt 75,000-150,000): 0.1 mg/mL in sterile deionized water. 2. BME: 0.05% TrypsmIEDTA (5.3 mm EDTA) solution in with 25 mm HEPES and Hank’s Salts without L-glutamine (BME).

3. 0.1% Trypsin mhibrtor in BME (asabove)plus 200 uL deoxyribonuclease (Type 1, from bovine pancreas)per 5 mL of solution, 4. Serum Supplemented Growth Medium: Dulbecco’s Modified Eagle’s

Medium (DMEM) containing 4500 mg glucose,110 mg sodium pyruvate, and 110 mg sodium bicarbonatewithout L-glutamine plus 10% fetal calf serum and 1% antibiotic solution (see Note 1).

Astrocytes

for Gliotoxicity

Assay

19

5. Hank’s Balanced Salt Solution with sodium bicarbonate, without calcium chloride or magnesium sulfate (HBSS). 6. TrypsmiEDTA solution: 0.1% made up in HBSS as above. 7. Sodium hydroxide: 0.2M. 8. Trypan Blue dye solution: 0.4% in 0.8% NaCl and 0.06% K2HP04. 9. Coomassie blue based protem assayreagent kit plus bovine serum albumin (BSA) standards (1 mg/mL). 10. 2-deoxy-D-[13H] glucose ([3H]2-DG). 11. Ultima Gold High Flashpoint Liquid Scintillation Cocktail. 12. Pony vial H/L. Miniature polyethylene “Rangin” vials. 13. Phosphate buffered saline (PBS), pH 7.4, consisting of in wt/vol; 0.8% NaCl, 0.02% KCl, 0.02% KHZ P04, and 0.115% Na2HP04 made up both with and without 5.6 mM n-glucose (see ref. 6). 14. 0.2M HCI.

3. Methods of Astrocyte Cultures 1. Poly-L-lysine (PLL) coating: Cells can be grown either on glass coverslips (13-mm diameter; No. I thickness) in 24 well multidishes or in 25 cm2 flasks (see Note 2). Coverslips are flamed after dipping in 95% ethanol to sterilize, then placed in the wells of the culture dish. To each well, 100 pL of poly+lysine solution is added and allowed to dry for 5 min. The coverslips are then rinsed with sterile water and allowed to dry thoroughly before use. If coverslips are not required, the PLL can be added directly to the cells of the plate. 2. All instruments are sterilized by being flamed in ethanol and are supported on a piece of aluminum foil in a sterile laminar flow hood. Neonatal rat pups are obtained and the heads are quickly wiped in alcohol to sterilize them. Pups are sacrificed by decapitation and the heads are placed in a sterile Petri dish with a few milliliters of BME. 3. A small cut is made m the skin at the posterior of the skull and the skin removed. The skull is very fragile and can be easily removed using two pairs of forceps. The whole brain is then removed from the skull cavity by carefully scooping it out. 4. The neocortical tissue is isolated by pinching off the olfactory lobes at the anterior and the superior colliculi and the developing cerebellum at the posterior of the brain. The meninges are teased off and are separated carefully from the cortical hemispheres using fine forceps. 5. The cortical tissue is chopped into about eight pieces. These are placed in a sterile tube containing 10 mL of Trypsin/EDTA solution in BME, then are covered and incubated in a water bath at 37°C for 25 min.

3.1.

Production

and

Maintenance

20

Cookson, MeClean, and Pentreath

6. The cells are then triturated using a sterile Pasteur pipet and centrifuged at 1000 r-pm in a benchtop centrifuge. 7. The supernatant is removed and 2.5 mL of trypsin inhibitor plus DNase added. The tissue is triturated using a fresh Pasteur pipet. This suspension is then recentrifuged at 1000 rpm for 5 mm. 8. Step 7 is repeated. 9. After discarding the supernatant, 2 mL of supplemented DMEM are added to resuspend the cells. This suspension can be plated out at 50 pL/well or 1 n&/25 cm2 flask. 10. After incubating the cells for l-2 h at 37°C in a 5% CO2 humidified atmosphere, the cells are fed with 500 pWwel1 or 5 ml/flask of serum supplemented DMEM. These are then incubated for 1 wk and thereafter are fed twice weekly (see Note 3). 11. Staining for Glial Fibrillary Acidic Protein (GFAP) after 1 wk culture (4) should demonstrate that over 90% of such cultures are astrocytes. Both anti-GFAP and secondary, fluorescein-conjugated antibodies are commercially available. However, for various reasons it is often preferable to subculture at least once before usmg the cells in toxicological evaluations, mainly to give a greater yield of cultures per rat and to ensure even plating densities across the wells of the multiwell plates. 3.2. Astrocyte Subculture 1. Coverslips with attached cells are rinsed three times in HBSS using sterile Pasteur pipets. 2. Trypsin/EDTA solution in HBSS is added to cover the cells: 200 pL/ multiwell is sufficient. The covered plates are incubated at 37°C until the cells round up and float off the coverslips. The amount of time required to do this depends largely on the density of the cultures and is generally around 5-10 min (see Note 4). 3. Samples from the wells are removed and pooled in a sterile tube, centrrfuged for 5 min at 1000 rpm, and the supernatant replaced with serum supplemented DMEM as above and plated out at a density of about 2 x lo4 cells/well, as assessedby a hemocytometer count. 3.3. Measurement of Cell Viability and Total Protein 1. Cells on coverslips are washed in HBSS three times and are trypsinized wrth 200 /.tL of Trypsm/EDTA m HBSS per well until the cells detach from the coverslips. 2. Trypsin inhibitor (150 pLWwel1)IS added and the contents of each removed and placed in labeled microcentrifuge tubes,

Astrocytes for Gliotoxicity

Assay

21

3. Of this sample, 50 pL is removed and mixed with 50 l.tL of Trypan blue dye solution, and this sample is counted in a hemocytometer. Viability is expressed as percent dye excluding cells divided by the total number of cells. 4. The remainder of the cell suspension is centrifuged at 2000 rpm and the supernatant is removed. The small pellet of cells is resuspended in 500 pL of 0.2M NaOH, vortexed to disperse the pellet, and left overnight at 4OC. 5. BSA standards are made up in the following series; 1, 2.5, 5, 10, 15, 25, and 50 l.tg/mL. These are pipeted in duplicate into wells of a microtiter plate (150 pL/well). Likewise, 150 ltL of each sample is added in duplicate to a series of wells in the plate, followed by 150 pL of protein assay reagent to each well of the plate (5). The color develops within 5 min at room temperature and lasts for several hours. The plates are read at 570 nm using a plate reader (see Note 5).

3.4. Measurement 1. 2. 3. 4. 5. 6.

7.

of Cell Membrane Integrity with PHl2-DG Following the specified incubation with toxicant (see Notes 6 and 7), the medium is aspirated and each well washed twice with 1 mL PBS (37OC). PBS with 5.6 mM o-glucose (450 u,L/well) (37OC) is added followed by 50 PL 0.5 l.tCi/rnL [3H]2-DG. Incubation is carried out at 37OCin a 5% CO,/ 95% air humidified incubator for 15-45 min (7). Incubation is terminated by aspiration (with legitimate disposal) of the medium (see Note 8) followed by three washes with 1 mL ice-cold PBS. Cells are digested using 300 l.tL 0.2it4 NaOH and left overnight. The solution is neutralized with 0.2M HCI. Two 150 pL samples are taken directly from the well to the microtiter plate for protein determination and treated as in Section 3.3.5. BSA standards are made up in 0.2M NaCl. The remaining solution is transferred to a miniature scintillation vial containing 3 mL scintillation cocktail. A serial dilution of [3H]2-DG (0.50.0005 l&i) is made up in parallel to act as a standard. Samples are shaken and left overnight in the dark at 4OC. Samples are counted on a liquid scintillation counter (see Note 9). Backflux is determined by incubating cells in PBS with 5 rnM o-glucose containing 0.5 l&i [3H]2-DG as described above. At the end of the incubation time the medium is aspirated and each well washed three times with 1 mL PBS (37°C). A further incubation is carried out (15-45 min) in 500 FL PBS without [3H]2-DG. This [3H]2-DG solution is treated as in Section 3.4.6. (see Note 10).

22

Cookson, l&Clean,

and Pentreath

8. The remaining cells are treated for protein determination as in Section 3.4., steps 4 and 5 and scmtillation counting as in Section 3.4., step 6.

4. Notes 1. Suitable antibiotics are either penicillin/streptomycin (stock 10,000 p,Penicillin G, 10 mg streptomycm/mL) or gentamycin (10 mg/mL stock solution). We commonly use the latter. 2. Multiwell dishes are convenient for toxicological evaluations since one can perform experiments at five concentrations, plus a control, m quadruplicate from a 24-well plate. 3. It takes around 2 wk for these cells to reach confluency, depending on the age of the rat. The fastest growing cultures are prepared from neonates of ~24 h old. It is possrble to grow cells from older rats (we have used up to 5 d), but in older rats the more developed meninges are difficult to remove and may contaminate the culture with fibroblast-like cells. 4. Time of trypsinization should be kept to a minimum to prevent excess cellular damage. The progress of the reaction can be monitored using an inverted microscope. 5. Control values for total protein are typically around l-2 l.tg/104 cells. 6. We have used 6, 12, and 25 h incubation periods with toxicants. Cells are fed within 48 h preceding dosing with toxicant. 7. The uptake of [3H]2-DG in older cultures may become reduced. We recommend that primary cultures be used soon after confluence (2-3 wk) or 4-5 d after the first subculture. 8. Uptake and backflux can be measured together rf Section 3.4., step 7 is proceeded to. The total uptake of [3H]2-DG is equal to that contained in the cells plus that in the medium. 9. Typical control values for uptake are in the range of l-10 pmol/mg protein/min. 10. Typical control values for backflux are between 5 and 10% of uptake.

References 1. Aschner, M. and Kimelberg, H. K. (1991) The use of astrocytes in culture as model systems for evaluating neurotoxic-induced-injury. Neurotoxicology 12,505-518. 2 Shahar, A., de Vellis, J., Vernadakis, A., and Haber, B (eds.) (1989) A Dissection and Tissue Culture Manual of the Nervous System Lrss, New York. 3. Hertz, L., Juurlink, B. H. J., Szuchet, S., and Walz, W. (1986) In Neuromethods, Vol. I, GeneralNeurochemical Techniques. (Boulton, A. A and Baker, G. B., eds.), Humana, Clifton, NJ, pp. 117-167. 4. Raff, M., Fields, K., Mirsky, R., Press, R., and Winter, J. (1979) Cell type specific markers for distinguishing and studying neurons and the maJor classes of glial cells in culture. Brain Res. 174,283-308

Astrocytes for Gliotoxicity

Assay

23

5 Redinbaugh, M. G. and Campbell, W. H. (1985) Adaptation of the dye binding protein assay to microtitre plates. Anal. Biochem. 147,144-147 6. Brookes N. and Yarowsky P. J. (1985) Determinants of deoxyglucose uptake in cultured astrocytes. the role of the sodium pump. J. Neurochem. 44,473-479. 7. Yarowsky P. J., Boyne A. F., Weirwille R., and Brookes N. (1986) Effect of monensin on deoxyglucose uptake in cultured astrocytes: energy metabolism is coupled to sodium entry J Neurosci. 6,859-866.

CHAPTER4

Human

Thyroid

Carmel

Culture

Mothersill

1. Introduction The technique described in this chapter enables the culturing of thyroid cells without loss of differentiation and medium change. It is potentially useful for the long-term study of drug effects on the thyroid gland. Human thyroid cells obtained during surgery can be maintained in culture for periods of up to 2 mo without losing morphological or functional differentiation (1). In the clinical situation the thyroid may be exposed to long-term drug or radiation treatment that may have adverse effects on the functioning of the thyroid gland. These adverse effects can be assessed in culture by studying morphological and biochemical changes after exposure to test chemicals, cytotoxics, or radiation and extrapolated to the likely toxicity in humans. Sections of human thyroid are incubated in a trypsin/collagenase solution. The resulting supernatant is filtered and centrifuged twice. The cells are collected and resuspended in growth medium and any undigested thyroid tissue is reincubated in the trypsin/collagenase solution on two further occasions, Each supematant is filtered and centrifuged. The digests are pooled and plated out in flasks containing Eagle’s Medium. The cultures are incubated for 48 h and are then exposed to test chemicals and morphology, epithelial cell growth, and biochemical parameters. A long-term culture system for sheepthyroid has been established that retains many of the characteristic functional and morphological features of the gland. The human thyroid culture has been adapted from this sheep culture with only minor modifications. Some morphological differentiaFrom Methods m Molecular Slotogy, Edlted by. S O’Hare and C. K. Atterw~ll

Vol 43. In Vltfo Toxmty Testing Protocols Copyright

25

Humana

Press Inc., Totowa,

NJ

Mothersill tion time discrepancies occur. Follicles develop in sheep cultures in 5-8 d and in human cultures after 15-20 d. Undifferentiated areas are more common in human cultures and are visualized as patches of epithelial cells devoid of follicles. The unusual glucose and lactate metabolism of the sheep system permits a prolonged culture period. Glucose is rapidly metabolized to lactate, and then the lactate is utilized by the cultures over their remaining life-span. Exhaustion of lactate in the culture medium coincides with cell death, but the latter can be delayed by adding concentrated glucose to the medium just before this occurs. The metabolism of glucose to lactate and subsequentlactate utilization, follows the same pattern in human cultures, but at a much slower rate because of the lesser degree of differentiation (lactate use correlates strongly with morphological differentiation). The major factor in establishing a human thyroid culture is the amount and character of healthy tissue obtained. The best cultures are from 5-10 g samples of thyroid tissue, although even samples of 0.05 g have been cultured. The slower rate of differentiation of the human culture system is advantageous when long-term studies of drugs or radiation effects on the human thyroid are required. The test chemical can be added directly above the differentiated monolayer without disturbing the media or the degree of differentiation that occurs. Although human thyroid cultures have been established and utilized by other scientists, they have been short-term systems or subcultures maintained by the use of hormones or growth enhancers. In general, these have been used to study the biochemical behavior of the cultures in the short-term or in the characteristics of the subculture. In these cases the primary culture was not maintained for more than 7 d. This culture system correlates well with the in vivo situation. Thyroid cultures have a limited life-span in humans (15-20 doublings), which equates to the deterioration of cells in culture after the third or fourth subculture. The best endpoints for determining that the cells are functioning correctly are the T4 assay or 1251trapping ability. The thyroid culture shows a progressive loss of differentiation after repeated subculture. It is postulated that this may be a result of the effects of the trypsination, which causes the release of a receptor component into the medium that binds thyrotrophin. This receptor is regenerated when the cells are replated, but it is thought that the regeneration declines after repeated subcultures.

Human

Thyroid Culture

27

A large number of scientists use whole animals or animal cell culture systems (mostly rodent) that have limited use in relation to the study of human disease and toxicity. The heterogeneity of the source material in terms of genetic makeup and previous history of cytotoxic insult is a disadvantage in relation to the development of a standardized routine test system. 2. Materials 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

Normal and diseased human thyroid tissue excised during surgery. Blood administrator set. 100 j.t.mfine surgical gauze. Centrifuge. 40 mL sealed flasks. Sterile scalpel. Radioimmunoassay kit for T4. Liquid scintillation counter. Perchloric actd. 2 #Zi/mL Na ‘?odtde. Scintillation fluid. PAS stain. Collagenase (Type IV from Clostridium vibrio) solution:Make up a 1 mg/mL solution in BSS. Trypsin/collagenase solution: Prepare a 0.25% w/v trypsin in 1 mg/mL collagenase solution. Glucose solution: Prepare a 10 mg/mL solution in Eagle’s balanced salt solution. Methimazole solution: Prepare a 2 mM solution in Eagle’s balanced salt solution. Balanced salt solution: Ca2+ and Mg2+ free balanced salt solution (BSS) containing 200 IU/mL penicillin, 200 pg/rnL streptomycin, and 40 IU/mL gentamicin. Growth medium: 500 mL Eagle’s medium, 2 mM L-glutamine, 20% v/v lamb serum, 0.1 ug/mL hydrocortisone, 10 mL U/n-& insulin, 1 @4 potassium iodide, 20 IU/mL penicillin, 20 pg/mL streptomycin, 4 IU/mL gentamicin, 1 pg/rnL fungizone, 40 mL U/r& thyrotrophin, 12.5 mL 1M HEPES buffer.

3. Methods 3.1. Culture Procedure 1. Add pieces of human thyroid to ice-cold BSS continumg antibiotrcs. Chop the thyroid into small pieces, preferably of 5-10 g, using a sterile scalpel.

28

Mothersill

2. 3. 4. 5. 6.

Add the chopped tissue to 12 mL trypsin/collagenase solution at 37°C. Incubate for 30 min at 37OC. Filter the supernatant (containing freed cells) through a blood admmistration set. Filter again through a fine surgical gauze, thus removing any fat and fibrous debris. Centrifuge the filtrate at 4OOg. Resuspend the tissue digest in an equal volume of growth medium containing 40% serum, neutralizing trypsin activity. Incubate any undigested thyroid pieces m the trypsin/collagenase solution and incubate for two further periods of incubation (depending on the size of the tissue) followed by filtration, Centrifuge. Pool the cells from all three digests. Count the cells and adjust a 0.5 mL cell suspension to -1 x lo6 cells. Add the adjusted cell suspension to 5 mL of growth medium in a 40 mL flask. Plate directly into a flask containing the minimum amount of medium necessary to wet the surface of the plastic. Leave for 24 h (the explant should have adhered). Add further medium up to a total volume of 5 mL.

The following

3.2. Testing assays can be performed to evaluate any thyrotoxicity.

1. After 48 h expose the cultures to test chemicals (0.05-0.1 mL in culture medium or DMSO/ethanol), cytotoxics, or irradiation as follows: a. 5 replicates for each chemical. b. 5 or 10 controls. c. 3 replicate experiments. 2. Assessthe toxic effect using the following parameters: a. Lactate/glucose levels (2,3). b. Monitor the perchloric acid extracts of media samples throughout the culture period and compare to the control. c. T4 measurement. d. Use a T4 radioimmunoassay kit to monitor T4 release into the medium. e. Iodine trapping ability: Add 0.1 mL of 2 mM methimazole and 0.1 mL of 2 pCi/mL sodium 125I to the medium of the cell culture at room temperature, Incubate for 90 min. Wash the culture thoroughly with 4 aliquots of 5 mL Earle’s BSS to remove all traces of free radioactivity. Trypsinize the cells with 5 mL trypsin/EDTA solution and resuspend in -10 mL fresh growth medium. Count the cells and determine the 1251levels by liquid scintillation. Calculate the 1251counts/106 cells. Plot a graph of 1251counts/lO”cells as a function of time in culture for the differentiated cultures.

Human

Thyroid Culture

1

0

6

29

18

12

DAYS

IN

2~

30

36

12

CULTURE

Fig. 1. The levels of T, (ng/106 cells) detected in medium from differentiated huan thyroid cultures (dertved from multinodular goiter) over a 40-d life span. (n = 6). 3. Epithelial cell outgrowth: Monitor the tissue outgrowth by measuring its area and by performing autoradiography. Study the effects of the toxic agents on the different cell types present using immunocytochemical analysis for intermediate filaments or surface antigens.

3.3. Results 3.3.1. T4 Plot a standard curve for T4. Read off the values for T4 found in the medium samples at regular time intervals. Plot a graph of T4 levels over 40 d (Fig. 1).

30

Mothersill

z r \

2x10L

1= r= 0

lxlOL

"-

5 x103

\ ul ; 2x103 2 0 1x103 DAYS

IN

CULTURE

Fig. 2. The iodine trapping (count&O6 cells) of human thyroid cells over a 40-d life span (n = 6). Solid circle-multinodular goiter; open circle-adenoma; triangle-fibroblast culture. 3.3.2. Iodine Trapping Ability Calculate the 125Icounts/106 cells. Plot a graph of 1251counts/106 cells as a function of time in culture for the differentiated culture (Fig. 2).

4. Notes 1. If the tissue sample is small, incubate with the trypsin/collagenase solution and then plate directly mto a flask. After 48 h, expose the cultures to test chemicals as above. 2. Fibroblast contamination can occur in cultures derived from low initial cell numbers. Seeding high numbers of cells probably inhibits the proliferation of any fibroblasts present. 3. If the amount of tissue is very small (co.5 cm3), incubate the tissue sample with trypsin/collagenase solutron for 30 mm at 37°C. 4. Cultures rapidly metabolize the available glucose after which cell death occurs, therefore, it may be necessary to prolong the culture life-span by adding 0.1 mL of a concentrated glucose solution. This can be assessed initially by using a glucose assay,or once experienced with the technique, by judging the color of the medium.

Human

31

Thyroid Culture References

1 Mothersill, C., Seymour, C., Moriarty, M. .I., and Cullen, M. J. (1985) Long-term culture of differentiated human thyroid tissue. Acta Endocrinologica 108, 192-199. 2. Schmidt, F. H (1961) Enzymatic determination of glucose in biological samples. Klin. Wochenschr.

39,1244

3. Gutmann, I. and Wahlefel, A. W. (1974) Enzymatic measurement of lactic acid in samples of serum and plasma, in Methods of Enzymatic Analysis (Bergmeyer, H U , ed ), Academic, New York, pp. 1464-1467

CHAPTER5

Thyroid Follicular Cells in Monolayer Culture In Vitro Models for Thyroid

Stephen

Toxicity

Testing

I? Bidey

1. Introduction The development of in vitro models that preserve the functional characteristics of the normal thyroid gland has been a challenging objective of recent experimental thyroidology. A major aim of such endeavors has been to facilitate the development of simple, reliable, reproducible testing strategies for compounds interacting with, and perturbing the function of, the thyroid follicular cell (TFC). The earliest experimental thyroid models were based on organ culture or tissue slice preparations or, alternatively, short-term cell suspensions. Subsequently, however, it has become possible to maintain TFCs as monolayer cultures in which a high level of thyroid-specific differentiation and responsiveness may be preserved over a prolonged period. Such characteristics, which allow experimental intervention and the subsequent study of cell function and morphology, have facilitated the development of biological assays for thyrotrophin (TSH) (I) and thyroid autoantibodies (2) in serum, and have recently begun to encompass applications in the field of cellular toxicology, where the application of these new investigative tools has enabled the identification of the sites and mechanisms of action of agents demonstrating direct thyroid toxicity in vivo (3). The preservation of TSH-dependent responses in cultured TFCs, together with the relative ease with which large numbers of identical, From Methods in Molecular Biology, Edited by S O’Hare and C K Atterwill

Vol. 43: In Vitro Toxicity Testing Protocols Copynght

33

Humana

Press

Inc., Totowa,

NJ

Bidey

34

replicate cultures may be maintained, has made the TFC monolayer the model of choice for the quantitative assessment of the agents interfering with or modifying TSH-receptor interaction, transmembrane iodide movement, or cell proliferation. The most widely adopted in vitro functional markers of TFC stimulation have included intracellular CAMP accumulation and thyroid-radioiodide uptake, the latter being a response unique to the thyroid cell. Although cell proliferation within the normal adult thyroid gland is minimal, growth is enhanced by dietary goitrogens, such as the cyonogenie glucosides and thioglucosides, which inhibit iodiode uptake, leading to an impaired synthesis of thyroid hormones and a compensatory rise in pituitary TSH secretion. In vitro strategies for estimating the rate of cell proliferation involve the determination of the incorporation of C3H] thymidine into subconfluent TFC monolayers or, alternatively, assessment of the metaphase index of the culture (i.e., determination of the percentage of cells with chromosomes visibly in the “S” phase) (4). Despite the widespread use of primary thyroid cultures derived directly from thyroid tissue as a fundamental tool in thyroid cell biology, the inherent viability in agonist and antagonist responsiveness between preparations of cells derived from different individual thyroid tissues, coupled with the progressive dedifferentiation of cultures with increasing duration of in vitro maintenance, has limited the use of this system to essentially short-term, qualitative investigations of thyroid function and proliferation, Recently, however, a number of stable cell lines have been isolated that retain major features of the differentiated follicular cell. Foremost among these is FRTL-5, a cloned Fischer rat thyroid cell (5-7). 2. Materials 2.1. Basic

Maintenance

Medium

FRTL-5 cell monolayers are maintained in Ham’s F- 12 medium (Coon’s modification) containing various supplements,asdescribedbelow. Each liter of working medium is prepared by adding 2.5 g NaHC03 to 12.08 g of powdered medium and making this to 1 L with triple glass-distilled water. The medium is then filtered (0.22 pm) into presterilized 100 mL bottles and may be stored at 4°C for up to 3 mo. Immediately prior to use, a supplement of 4 hormones (as detailed below), MEM nonessential amino acids, penicillin and streptomycin are filtered into the medium.

Thyroid Toxicity Testing In Vitro

35

2.2. Hormone Supplements The hormones comprising this supplement are somatostatin (1 mg/L), hydrocortisone (0.33 mg/L), transferrin (OS g/L), and glycyl-histidyllysine acetate (2 mg/L). The stock supplement is prepared as follows: 1. Somatostatin:50 ug is dissolvedin 500 uL of Ca2+h4g2+free Hank’s balancedsalt solution (HBSS), and then madeto 5 mL with Ca2+/Mg2+free HBSS. 2. Hydrocortisone: 1 mg is dissolved into 1.5 rnL absoluteethanol, and 100 uL then is addedto 10.9mL Ca2+/Mg2+free HBSS. 3. Glycyl-histidyl lycine acetate:1 mg is dissolved in 1 mL Ca2+/Mg2+free HBSS. 4. Transferrin: 25 mg is dissolved into 5 mL Ca2+/Mg2+free HBSS. To prepare stock aliquots of the combined hormone supplement, 5 rnL of each of the solutions detailed above is added to 25 rnL Ca2+h4g2+free HBSS, mixed and stored as 1 mL aliquots at -70°C until required. After thawing, each is diluted in 100 mL of basic maintenance medium. The preparation of complete maintenance medium also requires addition of 200 mmol glutamine/L, nonessential amino acids (stored as 1 mL aliquots at 4”C), penicillin (100 U/mL) and streptomycin (100 pg/niL) (stored as 1 mL aliquots at -2OOC). Immediately before use, supplements are filtered (0.22 pm pore size) into 90 rnL of base medium, and sterile newborn calf serum (NCS) added. 2.3. TSH Preparations A number of biologically active preparations of TSH are available from commercial sources, which may be used as reference thyroid cell stimulators. Bovine TSH (First International Standard of Thyrotrophin) (pituitary TSH, bovine, for bioassay; coded 53/11) and human TSH (Second International Standard, coded 80/558) are specifically recommended for this purpose, and may be obtained from the National Institute for Biological Standards and Control (South Mimms, Potters Bar, Herts, UK). Both are supplied as ampuled, lyophilized preparations having a uniform, stated activity, and are stable at -20°C over prolonged periods (i.e., years). After reconstitution, preparations are stored as aliquots at -7O”C, and used within 6 mo. It is particularly important that partially used aliquots are not refrozen, since this will diminish their biological activity.

36

Bidey

Fig. 1. Phase-contrast photomicrograph of a monolayer colony of the rat thyroid follicular cell strain FRTL-5,48 h after passaging (200x magnification). 2.4. FRTL-5 Cells FRTL-5 cells are available from the American Type Culture Collection (Rockville, MD). They are routinely passaged in Coon’s modified Ham’s F12 medium supplemented with the “4H” hormone mixture described above, together with 10 pg/mL insulin, 100 pU/mL TSH, and 5% (v/v) NCS. Cells are grown in lo-cm diameter Petri dishes in 5% COz in air at 37°C. In the presence of TSH, the cells proliferate as uniform, round colonies (Fig. 1). 1. 2. 3. 4.

2.5. Giemsa 1 mL Giemsa stock reagent. 40 mL distilled water. 1.25 mL methanol. 2 drops 1M NaHC03.

Stain

3. Methods 3.1. Preparation of Cell Suspension from Stock FRTL-5 Cultures 1. Aseptically aspirate the growth medium from stock cultures of FRTL-5 cells. Rinse the cultures with prewarmed Ca2+/Mg2+-free HBSS.

Thyroid

37

Toxicity Testing In Vitro

2. To each culture, add a sterile solution of trypsm (1 mg/mL) and collagenase (20 U/n& in Ca*+/Mg*+-free HBSS). Ensure that the liquid covers the monolayer, Return the cultures to the incubator for 2-5 min. 3. Use a sterile Pasteur pipet to transfer the suspension of detached cells into a sterile 25 mL universal tube. Clumped cells should be dispersed by repeatedly pipetmg the suspension. 4. Add calf serum (0.5% v/v) to inactivate the trypsin. 5. Close the tube and shake gently to obtain a uniform cell suspension. 6. Centrifuge (lOOg, 5 min at room temperature) to obtain a cell pellet. 7. Remove the supernatant solution with a sterile Pasteur pipet, and resuspend the cells to an appropriate density in a small volume (e.g., 5 mL) of the plating medium.

3.2. Preparation

of Replicate

FRTL-5

Cell Monolayers

1, Having prepared a suspensionof single, viable FWI’L-5 cells from stock cultures, initiate the replicate monolayers that will form the bioassay “target” tissueby adding aliquots of the cell suspensionto 24-well tissueculture dishes. 2. After initiating the test cultures maintain the bioassay plates at 37°C under 5% CO2 in air, in a water-saturated atmosphere to prevent evaporation of the culture medium.

3.3. Prebioassay

Treatment

of Cells

1. A change of culture medium will be necessary 3-4 d after initiation of cultures. Remove the exhausted medium from the monolayers using a sterile Pasteur pipet attached to a vacuum suction pump and collection jar. 2. Add fresh medium (500 pLWwel1)to each monolayer with a minimum of delay, so that the cultures do not become dry. 3. Imtiate the bioassay by the addition of test solutions at the time of, or shortly after the first medium change.

3.4. Procedure for Extracting and Determining Intracellular CAMP Assessment of the effect of a compound on TSH-dependent adenylate cyclase activity is made by determining the final intracellular CAMP levels attained in the presence of a serial dilution of that compound, in cells simultaneously exposed to a standard dose of TSH, compared with the CAMP level attained in cells exposed to TSH alone. 1. Remove the maintenance medium from subconfluent monolayers by aseptic aspiration, and replace with 500 pL Leibovitz (L-15) medium. 2. Add 3-isobutyl-1-methylxanthine (IBMX) to a final level of 0.4 mM m all wells, to inhibit CAMP-dependent phosphodiesterase activity.

3. Add the test compound in multiple dilution, and, if appropriate, control diluent to triplicate sets of incubation wells. 4. Add a standard dose of TSH (e.g., 100 pU/mL) to each well. 5. After incubation for 15 min, remove the medium, and add 500 ltL of icecold absolute ethanol to each culture. This treatment both arrests the incubation reaction and releases intracellular CAMP from the lysed cells. 6. Seal the culture plates in wrapping film to prevent evaporation of ethanol, and transfer them to a -20°C freezer for 24 h. 7. Remove 200 p,L aliquots of the ethanolic fractions and transfer these to small glass test tubes. 8. Evaporate the tube contents to dryness under a stream of nitrogen. 9. Redissolve the dried residues in 25 mM Tris, 50 mM NaCl, 8 n&f or other appropriate assaybuffer. 10 Determine the CAMP content by conventional radioimmunoassay (e.g., usmg a commercial kit). 11, Express the final CAMP level attained within each set of triplicate cultures as mean + SD level/culture. 3.5. Procedure for Determining Cellular Iodide Uptake The accumulation of inorganic iodide against a concentration gradient is a thyroid specific CAMP-dependent process that is maintained by the FRTL-5 line. 1, Remove the maintenance medium from subconfluent monolayers by aseptic aspiration, and replace with a fresh 500 l.tL ahquot of the same medium in each well. 2. Add 3-isobutyl-1-methylxanthine (IBMX) to a final level of 0.1 mM m all wells, to inhibit CAMP-dependent phosphodiesterase activity. 3. Add the test substance or control diluent, in multiple dilution, to triplicate sets of incubation wells, leaving a set of wells as controls. 4. Add a standard dose of TSH (e.g., 100 pU/mL) to both test and control incubation wells. 5. After incubation for 24 h, remove the medium, and add a fresh aliquot of maintenance medium containing 1 ltCr Na1251.Continue to incubate for a further 24 h at 37°C. 6. After incubation, remove the radioactive media and carefully discard. Briefly rinse (X2) each culture with 50 pL ice-cold HBSS. 7. After washing, add 500 l.tL 100 pJ4 sodium perchlorate to each well. This inhibits iodide pump activity, allowing intracellular iodide to discharge into the medium. 8. After 20 min, remove duplicate 100 l.tL portions of the sodium perchlorate solution, and determine 1251content using a y-scintillation counter. Express

Thyroid

Toxicity

Testing

In Vitro

results as a percentage of the mean 1251uptake value obtained in replicate cultures exposed to the standard TSH dose alone. 3.6. PH] Thymidine Incorporation as a Marker of Cell Proliferation 1. Remove the maintenance medium from subconfluent monolayers by aseptic aspiration, and replace with a fresh 500 pL aliquot of the same medium in each well. 2. Add the test substance,in multiple dilution, to triplicate sets of incubation wells, leaving a set of wells as controls. 3. Followmg a 24 h incubation period in the presence of appropriate combinations of test substances,dilute an aliquot of [methyl 3H] thymidine stock solution to 5 mL with serum-free “4H” medium, and add 50 pL to each well (1 pWwel1). Continue to incubate the cultures at 37OC,under 5%C02/ 95% air for a further 8 h. 4. Terminate the incubation by removal of the medium. 5. Briefly rinse each culture (X2) with ice-cold 10% (w/v) trichloroacetic acid (TCA), followed by a further addition of 500 pL 10% TCA/well. Leave the cultures for 3-4 h at 4°C to allow protein precipitation. 6. Remove the acid supernatants using a fresh Pasteur pipet for each well. Add 250 pL of 1N NaOH/L to each well. Seal the plates, wrap in aluminum foil, and leave overnight at room temperature to allow cellular digestion. 7. Remove duplicate 100 pL aliquots, and transfer these to scintillation vials. Add liquid scintillant (e.g., “Hisafe 2” scintillant, Pharmacia, Uppsala, Sweden) (4 mL) to each vial and determine [3H] thymidine activity using a p-scintillation counter. 3.7. Metaphase Index Determination 1. Follow the procedure shown in 3.5 for iodide uptake to include step 4. 2. After incubation for 44 h, remove the medium and add 200 PL fresh, serum-free maintenance medium together with 30 pL colcemid, a mitoticspindle arresting agent. 3. Incubate cultures for a further 3 h at 37°C. 4. Remove the incubation medium, and add 200 pL of freshly-prepared 30% (w/v) glacial acetic acid/70% ethanol to fix the cells. 5. After 15 min, remove the fixative with a Pasteur pipet, and rinse the monolayers twice in 70% ethanol. Leave to dry overnight. 6. After drying, add 100 pL diluted Giemsa stain to each well. Leave for 150 mm, replenishing the stain after 1 h. 7. Rmse the cell layers twice in 70% ethanol, then once with 95% ethanol. 8. Dry the cell layers, and observe for metaphase figures under a 40x objective.

40

Bidey

9. Calculate the Metaphase Index (X/100) x lOO%, where X= no. of cells/100 displaying figures.

4. Notes 1. Since the uniformity of the monolayer test cultures is crucial in obtaming a high level of precision and sensitivity of the assay system, cells must be dispensed into the culture wells with the aid of a fixed-volume repeating pipet, using disposable micropipet tips previously sterilized by autoclaving or y-irradiation. Assuming an even distrtbution of smgle cells within the starting suspension, tt 1spossible to obtain a between-culture variation in cell plating density approaching & l-2%. In the case of standard 24-well plates, a suitable starting inoculum may consist of lo5 cells/well. 2. L-15 medium does not require an equilibrating CO2 gas phase, so that incubations may be performed in room air at 37°C. 3. It is important to recogmze that the calculation in Section 3.4., step 11, assumes that the population densities of cells within replicate cultures are closely identical (i.e., within Z!Zl-2% limit). The between-culture variation in density may be estimated on the basis of cell protem or DNA estimations in a separate series of cultures wtthin each batch of assay plates. 4. One of the major advantages of the FRTL-5 is that unlike primary cultures, these cells may be maintained in long-term culture, having a reproducible, fully characterized behavioral pattern and population-doubling time. Given stable culture conditions, therefore, the responses of sequential passages of FRTL-5 cells should be entirely predictable. This has the important advantage of enabling large numbers of replicate and uniformly responsive test monolayers to be established, while also generating the inocula of FRTL-5 cells required to initiate subsequent stock cultures. 5. It may be desirable to investigate, m parallel cultures, for actions of TFC function on both CAMP accumulatton and transmembrane iodide uptake. Thus, although the latter is dependent on the functtonal and structural integrity of the cell membrane, specific inhibition of iodide pump or Na+/ K+-dependent ATPase activity may not necessarily, at least in the short term, adversely affect adenylate cyclase activity in a cell membrane that is otherwise structurally intact. However, if both CAMP accumulation and iodide uptake are diminished after incubation of cells with the test material, an effect on cell viability should be suspected. 6. In order to investigate for thyroid-directed toxic actions of nonwater soluble molecules, after solubilizing these in a nonpolar solvent, it is important that equivalent levels of the solvent are also included within TSHcontaining control cultures.

Thyroid Toxicity Testing In Vitro 7. The potential toxicity of any given compound toward TFCs may result from either a change in the normal functioning of that cell, through direct and specific interference (e.g., with TSH receptors or iodide pump activity) or, alternatively, through a more fundamental perturbation of basal cell activity owing to generalized cytotoxic actions. In the latter case, thyroid-specific function will also be impaired as a consequence of the interaction of the compound with the basal cell membrane, although the fundamental toxicity of the compound may, of course, not be restricted to TFCs. Thus, for example, since accumulation of iodide is dependent on the correct functionmg of a Na+/K+ ATPase and an iodide pump m the basolateral TFC membrane, in addition to the TSH receptor itself, nonspecific damage to the basolateral membrane may also impatr TFC function. The converse is also true, and if iodide accumulation remains unimpaired in the presence of a given compound, then the basolateral membrane is unlikely to have suffered any fundamental damage. The question as to whether a drug-induced decrease in iodide uptake is secondary to a nonthyroid specific impairment in cell viability, can, in extreme cases,be ascertained by morphological observation. However, more subtle changes may be uncovered by investigating for the release into the medium of membrane-associated enzymes, such as lactate dehydrogenase. References 1. Robertson, W. R. and Bidey, S. P. (1991) The in vctro bioassayof peptide hormones, in Peptide Hormone Secretion, A Practical Approach (Hutton, J. C and Siddle, K., eds ), IRL, Oxford, UK, pp 121-157. 2 Vitti, P., Rotella, C. M., Valente, W. A , Cohen, J., Aloj, S. M., Laccetti, P., Ambesi-Impiombato, F. S , Grollman, E. F., Pinchera, A., Toccafondi, R., and Kohn, L. D. (1983) Characterisation of the optimal stimulatory effects of Graves’ monoclonal and serum IgGs on cyclic AMP production m FRTL-5 thyroid cells: a potential climcal assay J. Clin. Endocrinol. Metabol. 57,782-79 1. 3. Fowler, K. L. and Atterwill, C. K. (1989) Potential use of thyroid FRTL-5 cells for predicting the toxic effects of xenobiotics, in FRTL-5 Today: Proceedings of the First International

Workshop

on the Characterization

and Standardization

of an

In Vitro Thyrozd Cell System (Ambesi-Impiombato, F. S and Perrild, H., eds.), Exerpta Medica, Amsterdam, pp. 3 l-35. 4. Ealey, P. A , Emmerson, J , Bidey, S P., and Marshall, N. J. (1985) Thyrotrophin stimulation of mitogenesis of the rat thyroid cell strain FRTL-5: a metaphase index assay for the detection of thyroid growth stimulators. J. Endocrinol. 106,203-210 5. Ambesi-Impiombato, F. S., Picone, R., and Tramontano, D. (1982) Influence of hormones and serum on growth and differentiation of the thyroid cell strain FRTL, in Growth of Ceils in Hormonally-Defined Media, vol. 9. (Sato, G. H., Pardee, A., and Sirbasku, D. A., eds ), Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, pp. 483-492.

Bidey 6. Ambesi-Impiombato, F. S. (1989) The FRTL-5 cells, an in vitro thyroid system: introduction and development, in FRTL5 Today: Proceedings of the First International Workshop on the Characterization and Standardization of an In Vitro Thyroid Cell System, (Ambesi-Impiombato, F. S. and Perrild, H., eds.), Exerpta Medica, Amsterdam, pp. 3-7 7. Bidey, S. P., Lambert, J., and Robertson, W. R. (1988) Thyroid cell growth, differentiation and function in the FRTL-5 cell line: a survey J. Endocrinol. 119, 365-376.

CHAPTER6

Dust Toxicity in Rat Alveolar Macrophage Cultures Yrjii

Collan

and

Veli-Matti

Kosma

1. Introduction Macrophage cells in culture may be exposed to particulate matter, and resultant effects on cell viability are determined by vital dye exclusion and enzyme leakage assays. Many compounds, when inhaled as dust particles, have been found to be toxic to the respiratory system, with long-term exposure resulting in the development of pneumoconiotic fibrotizing lung disease. One of the initiating factors in fibrogenic lung disease is believed to be the direct damage inflicted on the alveolar macrophages, hence, these cells may represent a suitable in vitro screening system to investigate whether particulate matter is likely to be harmful when inhaled over an extended period (l-4). Macrophages can be isolated, cultured, and exposed to suspensions of particulate matter. Damage may then be assessedin two ways. The fast is simply by estimating cell death. The second parameter, i.e., lactate dehydrogenase (LDH) activity in the supernatant, provides a means of determining cell membrane damage as indicated by leakage of the enzyme, LDH, out of the cells into the medium (5). The damageto the cells as a result of exposure can then be usedto assesswhether inhalation of the dust may be harmful in vivo. This test system provides a rapid, sensitive, and relatively inexpensive means of assessing the harmful effects of dust particles, In vitro hemolysis, an acute toxicity test, has been used in the past to detect the potential toxicity of particulate matter. However, since alveoFrom- Methods m Molecular Brology, Vol. 43 In V&o Toxicity Testmg Protocols Edlted by- S. O’Hare and C K Atterw~ll Copynght Humana Press Inc , Totowa, NJ

43

Collan and Kosma

44

lar macrophages appear to be one of the primary sites of damage in the initial stages of fibrogenic lung disease, they are probably a more suitable cell to be used in an in vitro screening system and are likely to reflect the in vivo situation more closely. A number of enzyme activities have been examined by the authors including aspartate amino transferase, acid phosphatase, and alanine amino transferase activity. Although a certain degree of reproducibility and correlation to cell damage were found, LDH activity was the most sensitive and readily detectable of the enzyme activities examined (1,2,6). Determination of this enzyme activity is, therefore, the preferred indicator of cell damage. There are three possible methods of examining the cells after exposure to potential toxins (6): 1. The medium of the cultures with the naturally detached floating macrophages is collected and analyzed. 2. The medium of the cultures IS removed, fresh medmm is added, macrophages are detached from the bottom of the well, then new medium wrth the harvested cells IS collected and analyzed. 3. The macrophages are detached from the bottom of the well without adding fresh medium; these are collected and analyzed. The third option has proved to be the most effective and consistent in identifying cell damage and best correlates to the occurrence of cell death. Modifications of this test system that use shorter incubation times are currently being developed by colleagues of the authors.

2. Materials 1. Animal-rat, e.g., BN-Kuo, Osborne-Mendel, Wistar: A number of different strains of rat have been used to provide cells for this procedure. Although the response of the cells to test compounds does not appear to be dependent on strain, it should be noted that the yield of cells may vary greatly. 2. Sterile dissection equipment. 3. Petri dishes. 4. Lammar flow cabinet. 5. Venflon-catheter. 6. 10 mL syringe. 7. Sterile centrifuge tubes. 8. Centrifuge, capable of 1000-3000 rpm. 9. Burker’s chamber. 10. 6 well plates with 1” diameter wells.

Dust Toxicity 11, 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23, 24. 25. 26. 27. 28. 29. 30.

31.

1. 2. 3. 4.

in Cultures

45

Incubator, 37”C, 5%/95% COZ in au, 95% humidity. Rubber policeman. Filters-1.2,0.8,0.45, and 0.22 pm pore. Automatic Multistat III analyzer-Intr. Lab. Microc. Analyzer, USA (for LDH determination). Sodium thiopental. Medium 199. 2.9% L-glutamine. Hanks’ balanced salt solution, pH 7.4. Hanks’ balanced salt solution, Ca2+- and Mg2+-free, pH 7.1-7.4. Phosphate buffered saline, Ca2+-and Mg2*-free, pH 7.2. Trypan blue. 2M hydrochloric acid. 2M NaOH. UV-LDH working solution. Na-pyruvate starter for LDH assay. Sodium chloride solutton: 0.9% NaCl, containing 100 U/mL penicillin, 100 I.t.g/mL streptomycin. Physiological saline: 0.9% NaCl, 10 U/mL heparin, 100 U/mL penicillin, 100 I.tg/mL streptomycin. 0.5% solution of Trypan blue dye in 0.9% NaCl solution. Culture medium: 1X medium 199 supplemented with 100 mL/L FCS, 100 U/mL penicillin, 100 p,g/mL streptomycin, and 10 mL/L 2.9% L-glutamine. Add 100 U/mL penicillin and 100 pg/mL streptomycin to the following: a. Ca2+- and Mg2+-free Hanks’ balanced salt solution. b. Hanks’ solution containing Ca2+and Mg2+. c. Ca2+- and Mg2+-free phosphate buffered saline. d. Heat treatment of fetal calf serum (see item 31). To free the FCS of LDH activity: a. Heat the serum at 56°C for 30 min. b. Keep at pH 3.5 (adjust with 2M hydrochloric acid) for 2 h. c. Adjust the pH to 7.4 (add 2M NaOH). d. Filter through 1.2,0.8,0.45, and 0.22 i.trn Millex nitrocellulose filters. 3. Methods 3.1. Cell Preparation Remove the lungs from anesthetized animals and lavage. Centrifuge the alveolar macrophages thus obtained and suspend m culture medium. Estimate cell number, dilute to 1 x lo6 cells/n& and plate out. Once the cells have adhered rinse the cultures and add fresh medium.

46

Collan and Kosma

5. After 24 h incubation the cells should be exposed to medium containing particulate matter for a further 24 h period. After this time the medium and harvested cells should be collected and the percent cell death established (using a Trypan blue staining technique). 6. The cells should then be centrifuged and an aliquot of the supernatant (culture medium) be removed for assessmentof LDH activity. 1. 2. 3. 4.

3.2. Macrophage Collection Anesthetize the rats with a subcutaneous injection of sodium thiopental (1.2 g/kg). Remove the lungs, ensuring the trachea and bronchi remain intact. Wash the outside of the lungs with sterile ice-cold sodmm chloride solution (containing antibiotics). Keep in covered Petri dishes on crushed ice for 10-20 min if not processed immediately.

Carry out the following procedures in the laminar flow cabinet. 1. 2. 3. 4. 5. 6. 7. 8. 9.

Cannulate the trachea with the catheter. Draw out the needle from within the catheter. Push the catheter deeper into the lumen of the trachea. Slowly introduce 10 mL of the physiological saline solution (4°C) into the lungs through the catheter using a 10 mL syringe. When filled, gently compress the lungs with the fingertips to encourage flow between lung compartments. Withdraw the fluid back into the syringe. Transfer to a sterile centrifuge tube. Keep tubes on ice for up to a maximum of 30 mm. Repeat the lavage 5-7 times for each animal.

3.3. Macrophage Culture 1. Centrifuge the cells at 4°C for 3 min at 2000 rpm, followed immediately by 7 min at 1000 rpm. Suspend in Ca2+- and Mg2+-free Hanks’ balanced salt solution (containing antibiotics) (4°C). Centrifuge the cells (3 min at 2000 rpm, 7 min at 1000 rpm). 2. The cells obtained from several animals may be pooled and suspended in a volume of culture medium (4°C) equivalent to 2.5 mL/ammal. 3. Transfer to plastic tubes kept on ice. Estimate cell number using a Burker’s Chamber and adjust to 1 x lo6 cells/ml. Perform a Trypan dye exclusion test at this point (200 p,L cell suspension, 200 FL 0.5% Trypan blue). 4. Add 2 mL of the cell suspension to each well of a multidish. Incubate for l-2 h at 37°C (5% CO2 m air, 95% humidity). Once the cells have adhered carefully remove the supernatant. Wash the cells for 5 min with Hanks’

Dust Toxicity

47

in Cultures

solution (containmg antibiotics) (room temperature). Aspirate off. Add 2 mL of culture medium/well (room temperature). Incubate for 24 h.

3.4. Preparation

of Dust Samples

1, Add 2 mg of the dust samples to be tested to 1 mL of Ca2+-and Mg2+-free PBS (containing antibiotics). Mix thoroughly. 2. Add 100 pL or 200 pL dust concentrate per 2 mL tissue culture fluid to give a final concentration of 100-200 pg/mL. 3. Incubate for 30 min at 37°C. 4. Place in an ultrasound bath for 3 min.

3.5. Testing Remove the culture medium from the wells. Wash each well carefully with Hanks’ solution containing antibiotics (room temperature). Set up

the plate as follows: 1. 2. 3. 4.

5. 6. 7. 8. 9. 10.

Zero control: culture medium only, no cells. Cell-free control: culture medium containing dust particles, no cells. Dust-free control: dust-free culture medium. Dust control: a. Titanium dioxide dust in culture medium. b. Chrysotile dust in culture medium. c. Additional control dusts in culture medium. d. Test condition-dust sample to be tested in culture medium. Add 2 mL of relevant medium (room temperature) to each well. Incubate for 24 h at 37°C 5% COz.in air. At the appropriate time detach the cells from the dish with a rubber policeman. Collect the cell-containing medium for analysis. Cell death: Remove a 200 pL aliquot of the cell suspension and add 200 pL of 0.5% Trypan blue. Count the number of viable and nonviable cells using a Burker’s chamber. LDH activity: Spin down the detached cells at 3000 rpm for 15 min. Remove 9 pL of supernatantfor assessmentof lactic dehydrogenase activity. Add 91 pL UV-LDH working solution and 5.2 pL Na-pyruvate starter.

The assay proceeds automatically to determine absorption at 340 nm and 37°C and express the result as IU/L LDH. 3.6. Results 3.6.1, Cell Death

Calculate the number of dead cells present in each situation as a percentage of the total number of cells, The dust-free situation represents the control, i.e., cell wastage, which occurs naturally. The positive con-

48

Collan

and Kosma

trols and test situations may be compared to assess whether a significant increase in cell death has occurred. 3.6.2. LDH Activity Zero control provides an indication of LDH activity present in medium. Cell-free control indicates whether the dust particles are affecting the LDH assay. Dust-free control indicates the level of cell death and LDH activity present in control, nonexposed cultures. Dust control provides a positive control, i.e., a standard against which the harmful effect of test compounds may be judged (I--3,6).

4. Notes 1. Care should be taken to treat the serum used to supplement culture medium so that any lactate dehydrogenase activity can be removed. It should also be noted that the serum may, to some extent, exert an mhibitory effect on the enzyme assay,although the incorporation of appropriate controls in the experimental protocol will allow for this eventuality. 2. Under ideal conditions, the zero control and the cell-free control will be the same. 3. Each situation should be corrected with respect to any activity found in the cell-free control. An increase in LDH activity over that observed in the dust-free cultures indicates that damage has occurred to the cells. The degree of damage may be estimated by comparing to standard positive controls. 4. Cells can be left for up to 3 h without markedly influencing the results but the larger the number of cells present the higher the enzyme value; therefore, ensure careful standardization of cell number and technique. 5. There should be a good correlation between the fraction of dead cells and the results of the enzyme assays.

References 1. Holopainen, M., Collan, Y , Kosma V.-M , Kalliokoski, P , Kulju, T., Anttonen, H , Tossavainen,A , and Kauppinen, H. (1986) Evidence for toxicity of phlogopite in hemolysis and macrophagetests,in Proceedings of the 2nd International Symposium on Occupattonal Health and Safety in Mining and Tunneling. Prague. pp. 65-72. 2. Committee on Enzymesof the ScandinavianSociety for Clinical Chemistry and Clinical Physiology (1974) Recommendedmethodsfor the determination of four enzymesin blood. Stand. J. Clin. Lab. Invest. 33,290-306 3 Pasanen,J. T. (1982) Alveolaarlset ja peritoneaalisetmakrofagit pdlyJen sytotoksisuudentestauksessa in vitro. Pro gradu-tutkielma.Jyvaskyllin yliopisto. Mom&e, Tybterveyslaitos, Helsinki.

Dust

Toxicity

in Cultures

49

4. Collan, Y., Kosma, V -M , Kalhokoski, P., Seppa, A., Kulju, T., Vaantinen, I., Remola-Pgirssmen, E., Mlettmen, R., Pretila, L., Gidlund-Marjanen, A.-L., Manninen, R., Anttonen, H , Tossavainen, A., Husman, K., Huuskonen, M. S., Rytkiinen, E., Lehtinen, A., Kauppinen, H., Mikkonen, A., Koistinen, S., Karjalainen, T., and Harm&i, 0. (1985) Siilinjiirven Apatiittiesiintymiin Richteriitti: Biologinen Vaikutus ja Tybhygieeninen Merkitys. Kuopio 5. Collan, Y., Kosma V.-M , Anttonen, H., and Kulju, T (1986) Toxicity of richterite in hemolysls test and macrophage cultures. Toxic interfaces of neurones, smoke and genes. Arch. Toxic01 Suppl. 9,292-293. 6 Collan, Y., Kosma, V.-M., Kulju, T., Vaananen, I., Remola-Parssmen, E., Pesonen, E., Puhakainen, R., Rytoluoto-Kfirkkainen, R., Manninen, R., and Pasanen, J. (1988) Estimation of dust toxicity in rat alveolar macrophage cultures, in Safety Evaluation of Chemicals on Laboratory Animals, Proceedmgs of the Finnish-Soviet Symposium, Kuoplo, 20-22 May 1986 (Nevalainen, T., Voipio, H -M., and Haataja, H., eds.), Kuopio, Finland, pp. 105-123

CHAPTER7

Hepatoma Cell Cultures as In Vitro Models for Hepatotoxicity Margherita

Ferro

1. Introduction This test is designed to detect irreversible toxic effects on both cell growth and survival, by theevaluation of colony-forming (CF) efficiency, in hepatoma cell lines derived from humans, rat, and mouse. The liver is a major target organ for the cytotoxicity of many xenobiotics. It has been suggested that hepatoma cell lines may provide an appropriate in vitro model for the assessment of likely hepatotoxicity in vivo. It should be noted, however, that the usefulness of such systems largely depends on the ability of the cells to maintain differentiated functions. The procedure presented here suggests a simple means of assessing the cytotoxicity of compounds to hepatoma cell lines. The method involves exposing the cells to xenobiotics, after which colony formation is monitored and compared to that of nonexposed control cultures. The method can easily be adapted for many hepatoma cell lines. In this particular procedure, six hepatoma cell lines have so far been used: 1. HepG2, a human hepatoblastoma cell line that shows basal and inducible levels of monooxygenases; 2. MH&, a rat hepatoma cell line, that shows basal and inducible levels of monooxygenases; 3. 7777, a rat hepatoma cell line that shows basal and inducible levels of monooxygenases; 4. HTC, a rat hepatoma cell line that does not show monooxygenase activities; 5. JM2, a rat hepatoma cell line that does not show monooxygenase activities; and From’ Methods m Molecular Biology, Edited by S O’Hare and C K Atterwlll

Vol. 43’ In Vitro Toxmty Tesbng Protocols Copyright

51

Humana

Press

Inc , Totowa,

NJ

Ferro 6. Hepa 1~1~7,a murine hepatomacell line that shows basal and inducible levels of monooxygenases. A small number of cells (250-500) are plated into complete medium in 6 well (35~mm diameter) culture plates. After 24 h, the medium is removed and the cells are exposed to different concentrations of the test compounds in serum-free medium. After 1,6, or 24 h, the experimental medium is removed and the cells are grown in standard conditions for 8-15 d. The colonies are then stained and those at least 0.3 mm in diameter are scored with the naked eye and the percentage survival is calculated with reference to control cultures. A decrease in the number of colonies formed is an indication of test compound toxicity. Colony formation may provide a much more sensitive measure of toxicity than certain other commonly employed methods. For example, it was found to be more sensitive than the LDH leakage assay (I), because it depends on cell growth-related mechanisms rather than cell membrane damage. Colony formation was also found to be a more sensitive parameter of toxicity than cell viability, assessed by total macromolecular content of the attached monolayer (2,3). This increased sensitivity could be because colony formation is assessedwhile the cells are in a state of proliferation, and thus more susceptible to toxic effects. Moreover, the measurement of total macromolecular content is carried out in a larger number of cells, which may mask some dose-dependent effects since the test compound is distributed over a large population of target cells. The sensitivity of the colony-formation assay, and the fact that dose and time-dependent effects are detectable, enables acute and chronic exposure periods to be investigated as well as permitting recovery studies. It has been found that exposure of the cells to test compounds before seeding rather than 24 h post attachment is not an effective way of assessing toxicity. Colony formation may be employed as an endpoint in many different hepatoma cell lines. The method must be standardized, however, for each cell type. There are cell lines, for example H4IIEC3, whose cells spread out from the colony, thus impairing distinction of separate colonies. Ideally the cells should be plated out at a suitable level so that over a period of approx 7-15 d they form between 50 and 100 colonies. The level of plating and period of incubation for any particular cell line can be established this way.

Hepa toma Cell Cd t ures Basal Colony-Forming Cell line JM2 7777

HTC Hep G2 MI-WI Hepa lclc7

53

Table 1 Abilities (CFA) of the Various Hepatoma Cell Lines No. cells/well 250 250 250 500 300 200

CFA (%) 37f 30*

6 2

39 f 10 19 * 12 16f 4 37 Ik 15

The different hepatoma cell lines vary in their sensitivity to different chemicals. A comparison of HTC (rat hepatoma) and Hep G2 (human hepatoma) cells (3) showed differences that arise from the fact that Hep G2 cells retain some cytochrome P450-dependentfunctions, whereas HTC cells are practically devoid of bioactivating enzymes. Thus, the IC,a values in mM for paracetamol, amitryptiline and nicotine (see Table 1) are approximately ten times greaterfor HTC cells than for Hep G2 cells, showing that metabolism by the latter contributes to the toxicity. However, Hep G2 cells are also more sensitive than HTC to iron sulfate, digoxin, and potassium cyanide, substances that are assumed to have a direct toxic effect. This could be caused by less specific mechanisms operating in the human cell line. Similarly, a difference in sensitivity to benzaldehyde was found between the two rat cell lines, MHiCr and HTC (I), because aldehyde dehydrogenase is present in much greater quantities in HTC cells. Thus, although the effects of a chemical on colony formation in a hepatoma cell line will provide an indication of cytotoxicity, interpretation of the results depends on many factors, including the metabolic competence of the cells with respect to the test compound. The degree of expression of the various monooxygenase isozymes by different hepatoma cell lines must therefore be taken into consideration, The relevance of results obtained in various hepatoma cell lines to human toxicity in vivo, and therefore the definition of the most appropriate cell lines to use for specific testing purposes has yet to be established. The inclusion of this system into the Multicenter Evaluation of In Vitro Cytotoxicity (MEIC) program may begin to answer these questions. The cells are exposed to the test compounds in serum-freemedium so that the direct toxic effect can be estimated in the absenceof any confounding effects arising from the interaction of the compounds with serum proteins.

Ferro

54

2. Materials 2.1. Cell Lines 1. Human hepatoma cell line: Hep G2 (#), ECACC: 85011430, ATCC: HB8065. Obtained from B. Knowles, Wistar Institute of Anatomy and Biology, Philadelphia, PA. 2. Rat hepatoma cell lines. a. MHiCt (5), ECACC: 85112702, ATCC: CCL 144. Obtained from the American Type Culture Collection, Rockville, MD. b. McA-RH 7777 (6), ECACC: 90021504, ATCC: CRL 1601. Obtained from R. Lindahl, Dept. Biochemistry and Molecular Biology, Umversity of South Dakota, Vermillion, SD. c. JM2 (7). Obtamed from R. Lmdahl, Dept. Brochemrstry and Molecular Biology, University of South Dakota. d. HTC (8). Obtained from Flow Laboratories (Irvine, Scotland). 3. Murine hepatoma cell line: Hepa lclc7 (9). Obtained from G. Bellomo, Dept. of Experimental Medicine and Oncology, Universrty of Turm, Italy. 2.2. Equipment 1. Incubator: 37”C, humidrfied atmosphere, 5% CO,/95% air. 2. COSTAR Nucleopore 6 well (35-mm diameter) cluster dishes. 3. Corning tissue culture flasks, area 25 or 75 cm2. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

12.

2.3. Materials D-MEM/F12 (1: 1) (Grbco, Paisley, UK). MEM (Boehringer Mannheim GmbH, Germany). Nonessential ammo acids supplement for culture medium (Gibco). Vitamin supplement for culture medmm (Gibco). Newborn calf serum (Sera-Lab Ltd., Crawley Down, Sussex, UK). Fetal bovine serum (Sera-Lab Ltd.). Gentamycin (Gibco). Trypsm-EDTA (O.OS-0.02%) solution (Seromed, Berlin, Germany). HBSS or PBS. Stammg solution: 1% crystal violet m a 3: 1 v/v mixture of acetic acid and 97% ethanol. Complete medium for MHtC,, HTC, JM2, McA-RH 7777, and Hepa lclc7 cells: D-MEM/F12 (1: 1) supplemented with 0.1% NEA (by volume from a 100X solution), 0.1% vrtamms (by volume from a 100X solution), 10% newborn calf serum, 50 pg/mL gentamycin. Complete medmm for HepG2 cells: MEM supplemented with 0.1% NEA (by volume from a 100X solution), 10% fetal bovine serum, 50 ,uglrnL gentamycin.

Hepatoma

55

Cell Cultures

2.4, Test Compounds All test compounds/drug solutions should be prepared in serum-free medium on the day of use. Any test compounds/drug solutions that are insoluble in water should be dissolved in dimethylsulfoxide (DMSO), methanol, or ethanol. The final maximum solvent concentration should not exceed 0.25% v/v. 3. Methods 3.1. General Maintenance of the Hepatoma Cell Lines 1. Thaw cells rapidly (i.e., in a few minutes) from frozen stock. Suspend the cells in the appropriate medium and seed in tissue culture flasks (37”C, 5% CO2 in air) as follows: in 25 cm* flasks: a. 3 x lo5 MH,CI cells in 5 mL medium. b. 1 x lo5 HTC cells in 5 mL medium. c. 1 x lo5 7777 cells in 5 mL medium. d. 1 x lo5 JM2 cells in 5 mL medium. e. 1 x lo5 Hepa lclc7 cells in 5 mL medium. f. 3 x lo5 Hep G2 cells in 5 mL medium. in 75 cm* flasks: g. 1 x lo6 MH,H, cells in 15 mL medium. h. 3 x lo5 HTC cells in 15 mL medium. i. 3 x lo5 7777 cells in 15 mL medium. J. 3 x lo5 JM2 cells in 15 mL medium. k. 3 x lo5 Hepa lclc7 cells in 15 mL medium. 1. 1 x lo6 Hep G2 cells m 15 mL medium. 2. Change the medium every 2 d. Add 5 or 15 mL fresh medium to 25 and 75 cm* flasks, respectively. Grow as monolayer cultures. 3. To subculture the cells: Aspirate off medium and rinse cells with either HBSS or PBS. Add 1 (or 2) mL trypsin-EDTA to detach cells. Incubate flasks at 37°C for l-5 min. Monolayers should detach after 2-5 mm. Resuspend the cells m the appropriate complete medium and seed as above unless the cells are to be used for cytotoxiclty assessment.

3.2. Cytotoxicity

Test by Colony-Forming

Efjkiency

1. Cells should be harvested at the late-log phase of growth (i.e., when the cells are Just approaching confluency) in the manner described m Section 3.1., item 3. This ensures maximum colony-forming efficiency. 2. Count the cell suspensions and aliquot correct volumes (1.5 mL medium/ well) into 6-well cluster plates to seed the following numbers of cells/well:

56

Ferro 200 Hepa lclc7 cells, 250 HTC, 7777, and JM2 cells, 300 MH,C1 cells, 500 Hep G2 cells.

3.3. Exposure

to Test Compound

1. Ikventy four hours after the mmal seedmg,replace growth medium with complete (or serum-free) medium containmg the xenobiotic under investigation. 2. Acute exposure: After 1, 6, or 24 h, remove the xenobiotic-containing medium as described below and replace with fresh growth medium. Do not disturb the cultures until the end of the incubation period. 3. Chronic exposure: Leave cells exposed to the xenobiottc-containing medium. Do not disturb the cultures until the end of the incubation period. 4. Colony formation: At the end of the exposure period, aspirate off the experimental medium. Wash cells briefly with fresh medium. Grow cells in complete medium in standard conditions for 8-15 d. At the end of the culture period, rinse dishes with cold PBS. Stain with crystal violet staining solution. Score stained colonies over 0.3 mm in diameter (20-50 cells/ colony).

5. Calculate the number of colonies formed m the presence of test compounds as a percentage of those occurring m control cultures. Use these results to produce a dose-response curve from which the IC5a value (i.e., the concentration of test compound

that reduces colony formatton

to 50% of that in

controls) can be obtained.

4. Notes 1. Some serum-free batches of these media will also be required for dilution of test compounds. 2. It is important to handle colony-forming cultures with great care. All mediakolutrons

should be prewarmed before use.

3. The system should be standardized so that 70-100 colonies are obtained after 7 (HTC, 7777, JM2, and Hepa lclc7

cells) or 15 (MHIC1,

Hep G2

cells) d of incubation using control cultures. 4. After certain treatments,

the colonies produced are very small but visible

to the naked eye, thus, only colonies that can be seen without magnification are considered.

References 1. Ferro, M., Bassi, A. M., and Nanni, G. (1988) Hepatoma cell cultures as in vitro models for the hepatotoxlcity of xenobtottcs. ATLA 16(l), 32-37. 2. Bassi, A. M., Piana, S., Pence, S., Bosco, O., Brenci, S., and Ferro, M. (1991) Use of an established cell line in the evaluation of the cytotoxlc effects of various chemicals. Boll. Sot. It. Biol. Sper. 8, 809-816.

Hepatoma

Cell Cultures

57

3. Bassi,A. M., Bosco, 0 , Brenci, S , Adamo, D., Pence, S , Piana, S., Ferro, M., and Nanni, G. (1993) Evaluation of the cytotoxicity of the first 20 MEIC chemicals m two hepatoma cell lines with different xenobiotic metabolism capacities. ATLA 21, 65-72. 4. Knowles, B., et al (1980) SCWW 209,497-499. 5. Richardson, U. I , et al. (1969) J. Blol. Chem. 40,236-247. 6. Becker, J. E., et al (1976) in Oncodevelopmental Gene Expression. Academrc, New York, pp. 259-270 7 Novicki, D., et al (1983) In Vitro 19, 191-202. 8 Thompson, E. B., et al. (1966) Proc. Nat1 Acad Scl USA 56,296-303 9 Hankinson, 0. (1979) Proc. Natl. Acad. Sci. USA 76,373-376

CHAPTER8

Cytotoxicity, DNA Fragmentation, and DNA Repair Synthesis in Primary Human Hepatocytes Giovanni

Brambilla

and Antonietta

Martelli

1. Introduction The well-documented existence of interspecies differences in the responses elicited by exposure to chemical agents (1-3) represents the main difficulty in the extrapolation to humans of the results of toxicity studies carried out in laboratory animals or in in vitro systems employing as targets prokaryotes, lower eukaryotes, or mammalian cells. Growing evidence indicates that the most important cause of these species-related differences is the different handling of xenobiotics by metabolic activation/detoxification processes.Primary cultures of human hepatocytes that can be prepared from material discarded during the course of prescribed surgery offer the unique advantage of directly assessing the cytotoxic and genotoxic effects of chemicals in cells of our species possessing a comprehensive metabolic capability. Isolation and culture of human hepatocytes were first described by Strom and others in 1982 (4,5). In this chapter, we describe our procedure for hepatocyte isolation and culture, as well as the techniques for the evaluation of cytotoxicity, DNA fragmentation, and DNA repair induced by the exposure to chemical agents. Examples of the use of human hepatocyte primary cultures in cytotoxicity and genotoxicity studies are offered by several papers (6-l 1). Up to now chemicals tested for their cyto- and genotoxic effects in human hepaFrom Methods m Molecular Biology, Vol 43 ln V/fro Toxmty Teshng Protocols Edlted by S O’Hare and C K Atterw~ll Copynght Humana Press Inc , Totowa, NJ

59

60

Brambilla

and Martelli

tocytes are only a minimal fraction of those potentially hazardous for our species, but some provisional considerations can already be drawn. Human hepatocytes from different donors display a great variability in response. This can be ascribed to the fact that human donors differ in age, sex, diet, drug intake, lifestyle, and pathological conditions of the liver. For the large majority of compounds tested on human hepatocytes, the great interindividual variability of the response results in an overlapping of the genotoxic potencies with those observed in rat hepatocytes, but some chemicals, such as cimetidine, tripelennamine, P-naphtylamine, and unleaded gasoline are negative in humans and positive in rats; the converse occurs for 2,4-diaminotoluene and 5methylchrysene. Our survey of the literature data suggests that a standardization of the methods employed is necessary in order to avoid differences of results attributable rather to different experimental conditions than to true interspecies differences. Human liver may be obtained from discarded surgical material. Hepatocytes are isolated by collagenase perfusion. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

2. Materials Heating water bath, 42OC. Peristaltic pump, Mastefflex model 7014, tubing set. Plastic polythene catheters, 18-20 gage. Large glass Petri dish. Stainless steel dog comb and scissors. Inox or nylon mesh (74 pm pore size). Refrigerated centrifuge, swinging-bucket type. Hemocytometer and microscope. Hanks’ balanced salt solution (HBSS). Disodmm EDTA. Ca2+- and Mg2+-free HBSS, Collagenase (type I or IV), Sigma (St. Louis, MO). Williams’ E medium (WE). Fetal bovine serum (FS). Gentamicin, 50 pg/mL m WE. Trypan blue, 0.4% in saline. Plastic culture dishes, 60- and 35-mm diameter. Slide flasks. Collagen, 25 pg/mL. Inverted microscope. Plate shaker. Spectrophotometer (or microplate reader photometer),

Cytotoxicity

in Human Hepatocytes

61

23. 24. 25. 26. 27.

Trypan blue, 0.4% in saline. Neutral red dye, 50 pg/mL WE. Fixative: 1% formaldehyde, 1% CaC& in distilled water. Destain solution: 1% acetic acid, 50% ethanol, 49% distilled water. Stainless steel filter holder wrth an upper cylindrical funnel section and a lower collecting section. 28. Tubing set. 29. Peristaltic pump Gilson (Paris, France) Miniplus 2. 30. Spectrophotofluorometer.

2.1. Solutions

for DNA El&ion

1, Merchant’s solutron: 0.14M NaCl, 2.7 mIt4 KCI, 1.47 mZt4KH,PO,, 8.1 nuI4 Na2HP04, 0.53 miI4 disodium EDTA, pH 7.5. 2. Lysing solution: 0.2% sodium lauroyl sarcosinate, 2A4NaCl, 20 mZt4disodium EDTA, pH 10.0. 3. Eluting solution: 0.06M tetraethylammonium hydroxide, 20 mM disodium EDTA, pH 12.3. 4. Washing solution: 20 mIt4 disodium EDTA, pH 10.0. 5. Millipore mixed estersof cellulose filters, 25 mm diameter, 5 pm pore size.

2.2. Solutions

for DNA Determination

1. 33258 Hoechst dye stock solution (0.15 miI4 m distilled water); prepare and store in the dark; stable for at least 1 wk at 4°C. 2. Citrate buffer: 0.154MNaCl,O.O15M sodium citrate, pH 7.0. 3. 0.2M KH2P04 solution. 4. Dark room. 5. Heating water bath at 38°C for meltmg of photographic emulsion. 6. Black hermetic box. 7. Microscope (x 1250 magnification). 8. [Methyl-3H] thymidine (specific activity 20-40 Ci/mmol). 9. Sodium citrate, 1% solution. 10. 1:3 Acetic acid: ethanol solution. 11. Glycerol jelly. 12. Kodak NTB-2 emulsion. 13. Dl 1 Kodak developer. 14. Unifix Kodak fixer. 15. May Grtinwald-Giemsa stain.

3. Methods 1. After surgical removal the tissue is immediately placed in me-cold saline. Perfusion should start within 1 h from excision; a longer time interval reduces the hepatocyte viability.

Brambilla

and Martelli

2. Remove any damaged areas of the liver sample. The fragment for use should be at least 10 g m weight, and cut 2 cm apart from the damaged areas. It is better if the fragment is enclosed m the hepatic capsula on all sides, except for the one cut surface. 3. Place this fragment m a large sterile glass Petri dish, and insert plastic polythene catheters mto vascular orifices on the cut surface. Usually a mmimum of 2 or a maximum of 4 catheters are placed; the canulae may be moved from one cut vessel to another in order to obtain a more homogeneous blanching of the liver. 4. The perfusion starts with Ca2+- and Mg2+-free HBSS supplemented with 0.5 mM disodmm EDTA. Set up the peristalttc pump so as to perfuse at a rate of 14 mL/min for about 20 min. Dissolve collagenase m complete HBSS to a concentratton of 0.4-0.5 mg/mL, and continue the perfusion with this solutton, at the same rate, for a further 20-40 mm. The length of this step is dependent on the hver conditions. Since the buffers should be delivered to the hepatic site at a temperature of 37”C, the buffers should be maintained durmg perfusion m a heating water bath at 42°C; a loss of temperature occurs over the tubing set. 5. A few minutes after startmg the perfusion, wedge-shaped areas of blanched tissue should become evident on the surface of the liver. By the end of about 40 min perfusion approx 70% of the surface should have this appearance. 6. The following steps are carried out at 4°C in order to stop the collagenase digest and endonuclease activity. The Glisson membrane is cut with scissors, and the perfused hepatic parenchyma is gently combed in HBSS with the dog comb to obtain a liver cell suspension. Filter this suspension through a 74 pm pore mesh. The volume of HBSS used m this step should be sufficiently large to dilute bile that will prevent the subsequent cell sedimentation performed in a refrigerated centrifuge at 5Og for 4 mm. The pellet is resuspended m serum-free WE medium and centrifuged again Repeat this procedure twice, and finally resuspend the hepatocytes in complete WE medium, i.e., WE supplemented with 10% FS, and 50 yglmL gentamicm. Add 0.1 mL sample of cell suspension to 0.4 mL of 0.4% trypan blue solution; count the number of viable and nonviable (blue-stained) cells in a hemocytometer after 3 min resting. The fraction of viable cells usually varies from 70-90%, but since human liver fragments are not so easily obtamed, even hepatocyte suspensions 60% viable may be used. 7. The hepatocyte suspension is diluted m complete WE medium to the required concentration of 5 x lo5 cells/ml. The cell yield obtained with this procedure usually ranges from 5 x lo6 to 10 x 107/g of liver.

Cytotoxicity

in Human Hepatocytes 3.1. Cell Culture

and

63 Treatment

The method of seeding the cells depends on the assayto be carried out. 1. For cytotoxicity evaluation, 2 mL cell suspension (-1 X lo6 cells) are seeded in 35 mm dishes precoated with collagen; the total number of dishes depends on the test performed; for the trypan blue exclusion test only 1 dish/dose may be sufficient; for the neutral red test 3 dishes/dose are required. For the DNA damage/alkaline elution assay, 4 mL cell suspension (-2 x lo6 cells) are seeded in 60 mm dishes not coated with collagen, in order to allow an easy cell detachment; 2-3 dishes/dose are needed. For unscheduled DNA synthesis (UDS), 2 mL cell suspension are seeded in 35 mm plastic dishes or in slide flasks; the collagen coating of the flasks may be difficult and is usually avoided; 2 dishes or flasks/dose are needed. 2. The cultures are incubated in a 95% sir/5% CO2 humidified atmosphere for about 3 h, in order to allow cell attachment. Then remove the medium, rinse the culture with WE serum-free medium, and start the exposure to the test compound. 3. Appropriate concentrations of the compound to be tested are prepared m the medium used to set up the cultures. If the test compound is not directly soluble m medmm, use a solvent vehicle, such as ethanol or dimethyl sulfoxide, at maximum concentrations of 3 and 0.5%, respectively. Solvent control cultures receive equal concentrations of the solvent alone. The length of treatment is usually 20 h. 3.2. Cytotoxicity

Assays

Since not all laboratories are equipped with a microplate reader photometer suitable to allow the use of multiwell plates, the procedure for the neutral red assay will be described for cultures in 35 mm plastic Petri dishes. Multiwell plates are more convenient where lower numbers of cells are required. 1. Trypan blue assay: At the end of the treatment the cultures are washed with two changes of salme. Then 0.8 mL 0.4% trypan blue solution are added. After 2-3 min incubation, determine the fraction of nonviable (bluestained) cells in 10 randomly chosen fields with a graduated inverted microscope. 2. Neutral red (NR) assay:At the end of the treatment the medium is removed and replacedwith 2 mL serumfree medrumcontaining 50 ug NR/mL. The cultures are returned to the Incubator for further 2 h to allow for the uptake of the vital dye into the lysosomes of viable cells. Thereafter the medium is removed, and the cells are rapidly washed with 3 changes of fixative, fol-

64

Brambilla

and Martelli

lowed by addition of 2 ml/culture of destain solution. Plates are shaken for 10 min, and the extracted dye is read at 540 nm. Results are expressed as a percentage of the optical density determined with extract from control cultures at 540 nm. 3.3. DNA FragmentationlAlkaline El&ion Assay 1. At the end of the treatment wash cultures with cold Merchant’s solution and incubate dishes for 4-5 min at 4°C in 2 mL of the same solution. Merchant’s solution, containing EDTA that binds Ca2+ ions, allows the detachment of hepatocytes without scraping, which could cause mechanical DNA damage. Cells will be gently harvested in the same solution with a Pasteur pipet, and about 1 x lo6 cells of this suspension are loaded onto a 25 mm diameter, 5 urn pore size Millipore filter supported by a stainless steel holder with an upper cylinder funnel section and a lower collecting section terminatmg with an outflow to which a flexible tube can be attached. After an additional wash with 5 mL Merchant’s solution, pass 4.5 mL of lysing solution through the filter to lyse the cells. Filters are then washed with 3 mL 20 mM disodium EDTA at 22OC.Single-strand DNA is eluted from the lysate m the dark with a controlled flow (0.13 mL/min) of 13 mL eluting solution, The eluate is collected at lo-min intervals (10 samples). Once the elution is complete, break up the filter in 3 mL of eluting solution. 2. The DNA content of each eluate and that remammg on the filter is determined by the microfluorometric procedure of Cesarone et al. (12). Bring to pH 7.2 with 0.2M KH2P04 1 mL aliquots of DNA samples and blank (eluting solution), and dilute to a final volume of 2 mL with citrate buffer. Add to each sample 2 mL of a 1:50 dilution m citrate buffer of the 33258 Hoechst stock solution, Shake and mcubate for 10 min in the dark. The fluorescence spectra of the dye-DNA complex and of the free dye solution are recorded in a spectrophotofluorometer with an excitation wavelength at 360 nm and emission at 450 nm. 3. Calf thymus DNA diluted to an mitial concentration of 1 mg/mL m citrate buffer is used as standard solution, A DNA standard is run through the system after every 20 elution assays. 3.4. Unscheduled DNA Synthesis (UDS) 1. In the case of the UDS assay, treatments are carried out in serum-free medium supplemented with [methyl-3H] thymidine (10 mCi/mL). The length of exposure is 18-20 h. At the end of incubation, wash the cultures with cold saline; then add 2 ml/dish of 1% sodium citrate solution. After 5-10 min incubation, remove the citrate solution, add 2 ml/dish 1:3 acetic acid:ethanol, and incubate for 10 min. Repeat the exposure to the acetic acid:ethanol a further two times in order to fix the cells.

Cytotoxicity

in Human Hepatocytes

65

2. Cultures are air-dried; then a section from the bottom of the dish is cut out and glued to a microscope slide using glycerol jelly. Alternatively, plastic disposable slrde flasks may be used. 3. Dip the slides in Kodak NTB-2 emulsion for autoradiography, and expose for 7 d at 4OC in a black hermetic box. Develop autoradiographs in Dll Kodak developer at room temperature for 2-5 min in the dark. Wash the slides for 30 s in 1% acetic acid, and fix for 4 min m Umfix Kodak fixer. Stain the slides with May Grtmwald-Giemsa. 4. Count autoradiographic grains in the nucleus and m an equal-sized randomly chosen area of the cytoplasm of 50 consecutive cells of each slide using a microscope at 1250x magnification. The net number of nuclear grains owing to DNA repair synthesis is calculated by subtracting the cytoplasmic grain count from each nuclear count.

4. Notes 1. A variety of standard media such as Leibowitz-15, Waymonth MB 752/l, or Dulbecco’s modification of MEM have been successfully used for these cultures, but WE medium is the most convenient and the most often employed. Human hepatocyte primary cultures need a CO, incubator. 2. The NR contammg medium must be preincubated overnight at 37°C and centrifuged prior to use to remove precipitates of dye crystals. 3. In Section 3.4., step 1 acetic acid is essential to remove the excessof radioactive thymidme. 4. Results should report the net nuclear count but also nuclear and cytoplasmic counts separately. As a matter of fact, some chemicals could influence cytoplasmic labeling and interfere in the evaluation of DNA repair.

References 1.

Langenbach,R., Nesnow, S.,andRice, J. M. (1983) Organ and Species Specificuy

in Chemical Carcinogenesis Plenum, New York. 2 Hsu, L. C , Harris, C C , Lipsky, M. M , Snyder, S , and Trump, B F. (1987) Cell and species differences in metabolic activation of chemical carcinogens. Mutat. Res. 177, l-7. 3. Philhpson, C. E. and Ioannides, C. (1984) A comparative study of the broactivatlon of nitrosamines to mutagens by various animal species includmg man Carcinogenesis 5,1091-1094 4. Strom, S., Jirtle, R. L , Jones, R. S., Novicki, D L , Rosenberg, M. R , Novotny, A., Irons, G., MC Lain, J. R., and Michalopoulos, G. (1982) Isolation, culture, and transplantation of human hepatocytes. J. Natl. Cancer Inst. 68,771-778 5 Maekubo, H., Ozaki, S., Mitmaker, B., and Kalant, N (1982) Preparation of human hepatocytes from primary culture In Vitro 18,483-49 1. 6 Butterworth, B E., Bermudez, E , Smith-Oliver, T , Earle, L , Cattley, R , Martin, J., Popp, J. A , Strom, S , Jirtle, R., and Mrchalopoulos, G. (1984) Lack of genotoxic

66

Brambilla

and Martelli

activity of di(2-ethylhexyl) phthalate (DEHP) in rat and human hepatocytes. Curcmogenesis 5, 1329-1335 7 Robbiano, L., Gazzamga, G. M., Martelh, A., Pino, A , and Brambilla, G. (1986) DNA-damagmg activity of tripelennamine m primary cultures of human hepatocytes. Mutat. Res. 173,229-232. 8. Martelli, A., Robbiano, L., Ghia, M., Giuliano, L., Angelim, G., and Brambilla, G. (1986) A study of the potential genotoxicity of cimetidme using human hepatocyte primary cultures; discrepancy from results obtained in rat hepatocytes. Cancer Lett

30,11-16. 9. Martelli, A., Robbiano, L., Gazzaniga, G M., and Brambilla, G (1988) Comparative study of DNA damage and repair induced by ten N-mtroso compounds in primary cultures of human and rat hepatocytes. Cancer Res. 48,4144-4152. 10 Butterworth, B E , Smith-Oliver, T , Earle, L , Loury, D J , White, R D , Doolittle, D. J., Working, P K., Cattley, R C , Jlrtle, R., Michalopoulos, G., and Strom, S. (1989) Use of primary cultures of human hepatocytes in toxicology studies. Cancer Res 49,1075-1084 11 Robbiano, L., Martelh, A., Allavena, A., Mazzei, M., Gazzaniga, G. M., and Brambilla, G. (1991) Formation of the N-nitroso derivatives of SIX P-adrenergicblocking agents and their genotoxic effects m rat and human hepatocytes Cancel Res. S&2273-2279. 12. Cesarone, C. F., Bolognesi, C., and Santa, L. (1979) Improved microfluorometric DNA determmation in biological material using 33258 Hoechst. Anal. Biochem

100,188-197.

CHAPTER9

Model

Cavity

Method

INVITTOX 1. Introduction The model cavity method enables the in vitro cytotoxicity testing of dental restorative materials that may then be related to dental toxicity likely to occur in vivo. The test system has been designed to mimic, as closely as possible, the contact that occurs between dental restorative and tooth pulp when cavities are filled in the clinical situation, A monolayer of cells is maintained in culture, the test material being in contact with the medium. A reduction in cell number, compared to control (nonexposed) cultures, indicates that the compound is cytotoxic and, hence, may cause damage to a tooth in the in vivo situation. A hole is bored through the center of a Petri dish lid and a borosilicate glass cylinder; the lower aperture, sealed with a microfilter, is attached to the underside. The resultant chamber, which represents the cavity, is packed with test material. The lid is placed over a culture dish containing a monolayer of either fibroblasts or macrophages (which represents the pulp), in such a way that the filter is just in contact with the medium. After a set exposure period the number of cells in the culture is determined and compared to that of control cultures. A reduction in cell number provides an indication of the cytotoxicity of the compound. If the reduction is 10% the compound is tested again in the presence of dentine. Artificial cavities are prepared as above with the addition of powdered dentine (obtained from noncarious human teeth) compacted into a 0.5 mm layer at the base of the cylinder. (Or, alternatively, the filter can From: Methods m Molecular Biology, Vol 43. In V&o Toxfaty Tesbng Protocols Edlted by S O’Hare and C K Atterwlll Copynght Humana Press Inc , Totowa, NJ

67

INVITTOX be replaced by an intact dentine slice.) The cytotoxicity of test compounds is then reassessedas before. This system provides a very reproducible, simple technique for the screening of large numbers of compounds. It is less time-consuming and relatively inexpensive, especially in terms of animal usage, compared to in vivo tests. 1.1. Comparison with the In Vivo Situation In vivo testing of dental restoratives is performed on the teeth of primates, dogs, or ferrets. The wide variation in results, together with species differences, complicates any extrapolation to the situation in humans. It has been suggested that this method, which attempts to mimic cavities in human teeth, may provide a more appropriate test system for comparing the relative toxicities of compounds, especially in view of its reproducibility. At the present time, however, the test cannot be categorically stated to be superior to in vivo tests. 1.2. Comparison with Other In Vitro Test Systems This test has several advantagesover other in vitro test systems because it conforms more accurately to the conditions that exist in the clinical situation. Certain in vitro systems produce inaccuracies, often because of the unrealistically large volume of material being testedproducing toxic effects that would not be seen in vivo. In this test the ratio of volume and surface area of the test material, to the volume of the culture medium, is similar to that found in the clinical situation, thus producing more applicable results. Contact between the test material and underlying cells is through a permeable filter enabling the materials to be introduced into the system in their freshly mixed state when they are most irritant, as occurs in use. In conventional in vitro test systems the test chemical is often in direct contact with the target cells (representing the pulp cells of the tooth). In humans, however, when a tooth is prepared for a filling it is usual for a certain amount of dentine to remain intact. This layer of dentine would, therefore, separatethe material used to fill the cavity from the pulp of the tooth. This test system has been adapted so that, should a compound appear toxic, it can be retested in the same system, but one that incorporates a dentine component. The test compound must, therefore, penetrate the dentine before coming into contact with the medium bathing the pulp (i.e., macrophages or fibroblasts).

Model

Cavity

Method

The dentine component is prepared from noncarious human teeth (usually extracted wisdom teeth), which are readily available and again increase the applicability of the system for direct comparison to the clinical situation in humans. Dentine can be prepared as a powder or in slices. In its powdered form the dentine provides an alternative to intact dentine as a chemically reactive substrate,moderating toxicity and binding potentially toxic elements released from dental material. It is unsatisfactory, however, when the permeability characteristics of the dentine are important. In this instance, a dental slice provides a better simulation of the clinical situation. The slices also possessan intact smear layer (a layer of cutting debris that serves as a natural cavity liner) that limits the access of certain toxic agents to the pulp. The choice between using slices or powdered dentine ultimately depends on the importance of standardization against structural normality. However, if a material is to be tested according to the British Standard, dentine powder must be used. of Cell Type The dental pulp is a specialized area of connective tissue composed of a peripheral layer of odontoblasts and a central mass of fibroblastic tissue. Although odontoblasts are the most likely to be initially affected by a toxic material, they are highly specialized and difficult to grow in culture. Therefore, the cells of choice are either fibroblasts or macrophages. Mouse macrophages can be used as the target cells. The pulp beneath a carious dentine lesion may contain macrophages. Although much of this lesion is removed prior to placing a restoration, the inflammation would still be present. In addition, trauma induced during cavity preparation prior to placing a restoration can result in the presence of inflammatory cells. Macrophages from two strains of mice have been compared (7) and no difference in responsiveness was observed. The authors stress, however, that this does not guarantee that differences would not occur if other strains were used. On a routine basis, fibroblasts rather than macrophages are used for several reasons, e.g., ease of handling and maintaining a cell line rather than having to obtain fresh cells, lack of animal involvement, and so on. Pulp fibroblasts can be cultured as a primary diploid culture but this takes time, and the characteristics of the cells may vary from culture to culture. Fibroblasts are probably the most important cell type. It should be stressed that, since the two types of cell respond differently to certain 1.3. Choice

INVITTOX toxic constituents of dental restoratives, results obtained using fibroblasts are not necessarily comparable to those obtained with macrophages. The model cavity was recently adopted as a British Standard. The model cavity system is now under consideration as an international testing method. Dental materials have been tested from many different manufacturers.

2. Materials 2.1. Tissue Culture 1. Incubator, temperature 37 f l”C, humidified, 5% COJ95% air. 2. 50-mm Diameter sterile disposable vented Petrt dishes (tissue culture grade) 3. 35-mm Diameter sterile disposable vented Petn dishes (tissue culture grade). 1. 2. 3. 4. 5.

2.2. Model Cavity System 35 mm sterile plastic disposable vented Petri dishes (tissue culture grade). Dental wax. Borosilicate glass cylinder, BS 2598 (7 mm outside diameter x 3 mm inside diameter x 4 mm hrgh). Microfilter cement. Cellulose acetate filter (retainmg particles 2 0.45 ~JV!in size). 2.3. Preparation

1. 2. 3. 4. 5.

of Dentine

Phers. Steel bur. Ball mill. Liquid N2. Packing tool (2.5-mm diameter). 2.4. Solutions

1. 2. 3. 4. 5. 6. 7.

95% ethanol solution. Filter sterilized salme solution, 9 g/L NaCI. Penicillin-streptomycin solution (100 U/mL and 0.1 mg/mL, respectively). Trypsm-EDTA solution (0.5 g/L and 0.2 g/L, respectively). Dulbecco’s phosphate-buffered saline solution (D-PBS). Conditioning liquid, e.g., 50% citric acrd, acids or EDTA. Fibroblast growth medium Composition, mL/lOO mL Eagle’s minimal essential medium Tryptose phosphate broth Glutamine (29.24 g/L) Penicillin-streptomycin Newborn calf serum

76 10 2 2 10

Model

Cavity

Method

71

8. Macrophage culture medium Composition, mL/lOO mL RPM1 1640 medium (HEPES buffered) 87 1 Glutamine 2 Penicillin-streptomycin Newborn calf serum 10 Composition, mWlO0 mL 9. Macrophage lavage medium D-PBS 90 Heparin BP solution (500 U/mL) 0.4 10 Newborn calf serum 10. Test materials: Prepare any test material immediately before use according to the manufacturer’s instructions. The material should be processed under consistent conditions of temperature and humidity. 11. Teeth: Noncarious freshly-extracted human wisdom teeth. Store in saline solutron containing 1 g/L thymol at 0-4”C, until required. Although it has been suggested that teeth can be stored for several months, the author prefers to use them within 1 mo. 12. Animal: BKW mice, 6-8 wk old. 3. Methods

3.1. Cell Culture 3.1.1. Preparation

of Mouse Macrophages

Use aseptic techniques throughout. Kill mice by asphyxiation with carbon dioxide, Subject each animal to peritoneal lavage using 2.5 mL of macrophage lavage medium. Pool the aspirated material from each mouse. Distribute -1.5 mL aliquots into 35mm diameter Petri dishes to give 3.5 x lo6 cells (including erythrocytes and small lymphocytes) per dish. Incubate (37”C, humidified, 5% CO,/95% air) for 2 h to allow the cells to attach. Remove nonadherentcells by washing twice with D-PBS solution (37°C). Add 2 mL vol of macrophage culture medium to each dish and incubate. 3.1.2. Routine Fibroblast Cell Maintenance 1. BHK-21 (C13) fibroblasts: Passagerecently thawed BHK-21 (C13) cells twice per week for 6-8 wk. Culture the cells m sterile disposable vented Petri dishes (50 mm diameter) containing 5 mL of fibroblast growth medium (37OC, humidified, 5% COJ95% air). 2. Subculture: For routine subculture BHK-21 (C13) fibroblasts are passaged at 1:5 ratio. 3. Wash the cell monolayer with D-PBS (37”C), to remove cell debris. Repeat. Add 0.3 mL of the trypsm/EDTA solution per Petri dish. Incubate

72

IIVVITTOX the dishes at 37°C for 10 min, to detach all the cells. Add 5 mL of fibroblast growth medium to each dish and disaggregate the cells by vigorous pipetmg to produce a fibroblast cell suspensron. Add 1 mL of the cell suspension to 4 mL of fresh medium m 50 mm culture dishes.

3.2. Model

Cavity

System

Drill a 5 mm diameter hole in the center of a vented 35 mm sterile tissue culture grade plastic Petri dish lid. Fix a borosilicate glass cylinder directly beneath the hole, using dental wax. Stick the cellulose acetate filter to the underside of the cylinder using microfilter cement.

3.3. Preparation

of Test Cultures

1. Prepare a fibroblast suspensionculture as described previously. Dilute with fibroblast growth medium until rt contams 1 x lo5 cells/ml. Plate out 4 mL of the suspensron m the base of the Petri dish; or 2. Prepare a monolayer of macrophage cells as described previously (3.5 x lo6 macrophages/dtsh in 4 mL of macrophage culture medium). Incubate for 24 h, remove the supernatant medium, and replace it with 4 mL of fresh macrophage culture medium. Then incubate monolayers for 24 h at 37”C, humidified 5% CO,/95% air.

3.4. Exposure

to Dental

Restoratives

Pack the cylinder (i.e., model cavity) with freshly mixed test material until it is flush with the top of the lid. Sterilize the lid assembly by swabbing the lid and filter with 95% ethanol. Allow to evaporate. Place the lid

assembly over the Petri dish. Agitate the dish so that the fluid contacts the filter. Expose the fibroblast or macrophage cultures for 24 h (incubate at 37”C, humidified 5% CO,/95% air) to the test material. 3.5. Control

Cultures

1. Use lid assemblies that do not contain test materials. 2. Test each material in quadruplicate. 3. Repeat each test on three separate occasrons.

3.6. Assessment

of Cell Damage

1. Frbroblast culture: After incubation for 24 h, remove the supernatant medmm. Wash the cells twice with prewarmed (37°C) D-PBS medium. Add 100 pL of trypsin/EDTA solutron. Incubate at 37°C for 30 min. Occasionally agitate the dishes. Add 2 mL of saline solutron that contams 10% v/v newborn calf serum, to each dish. Suspend and disaggregate the cells

Model

Cavity

Method

73

by vigorous pipeting. Dilute cell suspension with 9.6 mL sterilized saline solution. Determme the number of cells present, 2. Macrophage culture: After incubation for 24 h, remove the supernatant medium. Wash the cells twice with prewarmed (37°C) D-PBS solution, Add 2 mL of a lignocaine hydrochloride in RPM1 1640 medium to each dish and incubate for 12-15 min at 23 Z!Z2°C (this has been shown to aid the detachment of macrophage cells). Scrape the cells off the bottom of the dish and suspendthem by pipetmg. Add 1.5 mL of the suspendedcells to 8.5 mL of filter sterilized saline solution. Determine the number of cells present. Determine cell number of control cultures. Determine cell number of test cultures and present as a percentage of that present in controls. Statistically compare control results to test results. Test any material that gives a statistically significant reduction (p < 0.05) in cell number, or a reduction of 10% compared with controls, in the presence of dentine.

3.7. Test Procedure

in the Presence

of Dentine

1. Preparation of dentine powder: Take noncarious freshly-extracted human permanent wisdom teeth from storage. Wash the teeth in water and air-dry. Place m absolute ethanol and redry. Break off the crowns with pliers and remove the cementum and root canal contents with a steel bur. Grind the remaining dentine m a ball mill under liquid nitrogen to a particle size of ~0.2 mm. (Ground dentine can be stored at -2OOC;wash before use). Using a packing tool, apply a pressure of 4.2Nlm2 to compact 4.7 X 10m3-1.2 X lo-4 g of dentine powder to a depth of 0.5 f 0.1 mm in the bottom of a borosilicate glass cylinder. Repeat the test procedure as before. 2. Preparation of dentine slice: Cut 100, 500, or 1000 pm sections transversely through the crowns of freshly extracted teeth (down to the level of the pulp horns) using a ground section machine. Float the sections on conditioning liquid for 30 s to remove the lower pulpal smear layer. (N.B. The upper smear layer may be removed if it is applicable to the clinical use of the test material.) Attach the slices to the base of the cavity m place of the Millipore filter. Repeat the test procedure as before. 4. Notes 1, If using an electronic cell counter determine values for amplification, aperture diameter, current, and lower threshold value. 2. A reduction in fibroblasts indicates direct toxicity and/or effects on cell growth. 3. A reduction m the number of macrophages is indicatrve of irreversible damage.

74

IN??ITTOX References

1. Rabinovitch, M. and Destefano, M J. (1975) In Vitro 11,379-381. 2. Tyas, M. J. (1977) A method for the in vitro toxicity testmg of dental restorative materials. J. Dent Res. 56, 1285-1290 3. Meryon, S. D. and Browne, R. M. (1983) Evaluation of the cytotoxicity of four dental materials in vitro assessed by cell viability and enzyme cytochemistry. J. Oral Rehab. 10,363-372 4. Meryon, S. D. and Browne, R. M. (1983) Test methods for assessing the cytotoxrcity of dental restorative materials using an in vitro model cavity system, in Ceramics in Surgery (Vincenzini, P., ed.), Elsevier, Amsterdam, pp 127-135. 5. Meryon, S D., Stephens, P. G., and Browne, R. M. (1983) A comparison of in vitro cytotoxiclty of two glass monomer cements J. Dent. Res. 6769-773. 6. Meryon, S. D. and Browne, R. M (1984) in vitro cytotoxicity of a glass ionomer cement of a new generation. Cell Biochem. Funct. 2,43-48 7. Meryon, S. D., Uphill, P F., Cordery, A. D., and Browne, R. M (1985) A reproducibility study of the model cavity method for the in vitro toxicity testmg of dental restorative materials. ATLA 12,215-223 8. Meryon, S. D. (1988) Model cavity method incorporating dentine. Znt. Endo. .Z 21, 79-84.

CHAPTER10 Human

Esophageal Cannel

Culture

Mothersill

1. Introduction This chapter describes a method for establishing short-term explant cultures of esophageal mucosa. Adverse effects produced by exposure to radiation or test compounds can be detected as an inhibition of cell outgrowth. When mucosa explants are plated into culture there is an initial migration of cells outward, followed by a period of mitotic activity resulting in a pronounced outgrowth. Such explant cultures can be exposed to test compounds and radiation, after the initial period of migration has occurred, and the effects of these can then be quantified as an inhibition on the rate of cell growth. As the cells grow outward they form a definite area surrounding the tissue initially plated that, after fixing and staining, can readily be quantified by visual examination. Mucosa samples are dissected from the esophagus; the pieces of tissue are placed in growth medium and cut into small sections. They are then incubated in a trypsinkollagenase mixture for a short period, then suitable tissue samples are selected and cultured in growth medium until required for testing. After -2 d, test chemicals are added to the cultures and, after the required exposure period, toxicity is assessedby fixing, staining, and measuring the area of outgrowth of the epithelial cells. Primary cultures of epithelial cells are very difficult to achieve, require a high degree of technical expertise, and can be very time-consuming to maintain. Although a number of primary “normal” cell lines have been establishedfrom adenomatoustissue,it is quite common for theseto become From Methods m Molecular Brology, Edlted by S O’Hare and C K Atterwill

Vol 43: In Vitm Toxmty Testmg Protocols Copynght Humana Press Inc , Totowa, NJ

75

Mothersill contaminated with fibroblasts and, thus, they can no longer be considered “normal.” A further complication with regard to maintaining esophageal mucosa cells in culture is their lack of clonogenicity. The epithelial nature of the outgrowth was confirmed in representative samples of cultures using a low mol-wt general cytokeratin antibody with indirect peroxidase development. The presence of stromal or endothelial elements can be monitored using antivimentin and antihuman endothelium. This method, however, provides a relatively simple, fast, and easy technique for establishing short-term cultures of relatively pure populations of mucosal epithelial cells. Should contamination with fibroblasts occur, these cells are readily discernable after staining and can be allowed for in the calculation of cell growth. Unfortunately, the major drawback of this technique is that the cultures are only viable for a limited period of -2 wk (but up to 4 wk is possible). The viability of the culture can be extended by changing half the medium weekly. In addition, each new culture is dependent on a fresh supply of human tissue, which may not always be readily available. It should also be noted, however, that the technique is readily adapted to healthy and diseased tissue, thus enabling the differential response to compounds to be examined in both tissue types. Cell outgrowth is measured directly by quantifying the area of cells around the explant. This may provide a more reliable measure of cell growth and survival than other methods currently used, e.g., growth curve extrapolation in monolayer cell culture systems. The possibility that the outgrowth of cells from the original explant is simply a result of a process of cell migration has been considered. This is largely true in the initial stages of culture, however, the contribution of migrating cells to overall outgrowth at later stages is insignificant. This has been confirmed through experiments designed to measure the incorporation of tritiated thymidine (coupled with autoradiographic analysis) and examination of cultures by electron microscopy, which have illustrated that the cells present in the area of outgrowth have a high rate of mitotic

activity and are actively dividing (1). Cell proliferation in cultures canbe studied using the monoclonal antibody, Ki 67 (Dako, Santa Barbara, CA) and an indirect peroxidase immunocytochemical development technique. The effects of irradiation can also be studied using this cell system either in the presence or absence of the test compounds. Irradiation is carried out 2 d after the explant is established and/or 12 h after exposure

Human

Esophageal

77

Culture

to the test compound. Preliminary results using this culture system have shown that a reduction in the outgrowth of cells from the explant occurs after radiation treatment, The dose-response relationship that is obtained is within normal mammalian limits and tissue specific differences can be detected (2).

2. Materials 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11,

12. 13.

14. 15. 16.

Tissue: Normal or cancerous human esophageal mucosa explants. Sterile scissors. Sterile forceps. 90 mm Sterile Petri dish. 25 cm2 Tissue culture flasks. Incubator, 37”C, humidified atmosphere, 5%/95% CO2 in air. Macroscopic or microscopic grid: Either a grid is drawn on transparent paper or the flasks/Petri dishes are bought with grids on their surfaces. Eagle’s Balanced Salt Solution (EBSS). Eagle’s Basal Medium containing L-glutamme (EBM). Trypsin/collagenase solution: 0.25% trypsin and 10 mg/mL collagenase type IV in EBSS. Complete growth medium: EBM supplemented to give final concentration of: 20% fetal calf serum (FCS) 0.1 pg/mL hydrocortisone, 10 U/mL insulin, 20 U/r& penicillin, 20 pg/mL streptomycin, 1 ug/mL fungizone, 4 U/mL gentamicin sulfate, 50 U/n& mycostatin. Fixative: Make up a solution of formalm m 10% formaldehyde solution Test compounds: Test compounds should be dissolved in culture medium if water soluble. If insoluble in water, use DMSO or ethanol. Prepare a concentrated stock solution and dilute by at least 1: 100 in culture medium prior to addition to cells. Hematoxylm. Eosin. Periodic Acid Schiff Reagent (PAS).

3. Methods 3.1. Tissue

Preparation

The most common source of tissue is that available when patients are undergoing surgery for esophageal tumor removal. 1. Fix 2 mm3 samples of tissues for histological examination using\ 10 mL formalin at 20°C. Place the main sections (not fixed) into complete growth medium at 20°C (room temperature).

Mothersill

78 3.2. Culture

Technique

1. Rapidly chop the tissue into -2-3 mm2 pieces using sterile scrssors and forceps. 2. Place all the pieces m 10 mL of the trypsin/collagenase mixture at room temperature and incubate for 30 mm at 37°C. 3. Place the tissue segments m a 90 mm sterrle Petrt dish and select appropriate pieces for culture (nonfibrous, nonfatty, and of a uniform size). 4. Place each piece of explant into a separate 25 cm2 tissue culture flask containing 2 mL growth medium at room temperature. Maneuver the explant to the middle of the flask and carefully transfer to an incubator (37OC, 5%/ 95% CO, m au) without dislodging it. 5. Leave for 2 d after plating before commencmg with testmg, to ensure that the explant is attached and the mmal mrgratlon of cells has started.

3.3. Exposure

to Test Compounds

1, Discard any flasks that are heavily contaminated with frbroblasts or show no discernible migration of cells. 2. Establish the base line counts of cell numbers at rusk (usmg the method in Section 3.4.5., but without fixation and staining of the culture). 3. Set up 5 flasks/test concentration and controls. 4. Add 0.05-o. 1 mL of test compound, or vehicle, to each flask and return to the mcubator. 5. After the required period of exposure, remove the cultures and assess cell growth. 6 Remove the compound after the exposure period and replace wrth fresh medium. 7. Incubate for 2 wk.

3.4. Area

Measurement

1. Drain the flask of medium and rinse the culture wtth BSS at 20°C. 2 Add 5 mL of acetic acid, methanol, or formalin to fix the cultures at 20°C. 3. Leave the cultures for 15 mm m the case of acetic acid/methanol or 24 h for formalin. 4. Add 5 mL of either hematoxylin and eosm or Periodic Acid Schtff Reagent (H and E standard concentrations for hrstologlcal sections). 5. Measure the area of the flask covered by cells using a macroscopic or mrcroscoprc grid, depending on the size of the explant culture. Exclude the size of the initial explant and any frbroblastic areas from the calculation (fibroblastrc areas are vlsuahzed as fibrous, densely stained areas). 6. Express the results as a percentage of the control area The results can be converted to cell numbers by countmg the number of cells m a representative number of grid squares (normally 25% of total outgrowth).

79

Human Esophageal Culture 4. Notes

1. No more than 1 h should elapse between removal of the sample and plating out in culture flasks. 2. The medmm should not be changed at any stage, since thus leads to the death of eprthehal cells and prohferation of frbroblasts. 3. The exposure period will depend on the half life of the compound/drug but should be in the range of l-24 h. 4. PAS ISused to illustrate secretory activity, for example, in adenocarcmomas.

References 1. Mothersill, C., Cusack, A., MacDonnell, M., Hennessy, T. P., and Seymour, C. B. (1988) Differential response of normal and tumour oesophageal explant cultures to radiation. Acta Oncologica 27,275-280 2. Mothersill, C., Cusack, A., and Seymour, C. B. (1989) Enhanced proliferation of cells from human tissue explants following Irradiation m the presence of envrronmental carcinogens Radiat. Enwon Biophys. 28,203-212.

CHAPTER11 The Application of In Vitro Models of Anterior Pituitary Function in Toxicity Testing Glenda E. Gillies and Julia

C. Buckingkam

1. Introduction Together, the hypothalamus and the pituitary gland form the functional unit called the neuroendocrine system. For the purposes of introducing this system, therefore, the two will be considered together, although the methods for assessing their function will be considered separately (the pituitary gland in this chapter and the hypothalamus in Chapter 12). The hormones of the anterior pituitary gland, or adenohypophysis, namely growth hormone (GH), thyroid stimulating hormone (TSH), the gonadotrophins (luteinizing hormone, LH; follicle stimulating hormone, FSH), prolactin (Prl), and the pro-opiomelanocortin (POMC) family of peptides (adrenocorticotrophic hormone, ACTH; P-lipotrophin, P-LPH; P-endorphin; and N-terminal POMC peptides) play a key role in the maintenance of homeostasis. Their secretion is controlled by neurohormones (neuropeptides and dopamine), which are secreted by hypothalamic neurons mto the hypophyseal portal vessels and thereby conveyed directly to the anterior pituitary gland. The activity of the hypothalamopituitary axis is tightly regulated by a variety of mechanisms, including ascending and descending neural inputs to the hypothalamus, local regulatory mechanisms operating within the hypothalamus and pituitary gland, and blood-borne factors acting at the level of both the hypothalamus and pituitary gland (for review, see ref. I). The major secretory prodFrom. Methods m Molecular B/ology, Edited by S O’Hare and C K Atterw~ll

Vol. 43. In V/fro Toxmfy Testrng Protocols Copyright

81

Humana

Press

Inc , Totowa,

NJ

82

Gillies

and Buckingham

ucts of the posterior pituitary gland, or neurohypophysis, vasopressin, and oxytocin, will not be addressed in this chapter. Disturbances in hypothalamo-pituitary function result invariably in a spectrum of disorders of growth, physical and mental development, reproductive function, metabolism, and osmotic balance, many of which may not become apparent until some time after the original insult, by which time irreversible damage may have already occurred. In addition, the ability of the individual to respond appropriately to and/or cope with acute or long-term stress may be impaired with consequent insidious effects on both physical and mental health. Clinically, such conditions may be precipitated by pathological lesions (2) or by drugs and other xenobiotic agents, including environmental pollutants (I). Thus, there is growing awareness within the discipline of toxicology of the importance of experimental models for assessing hypothalamic and pituitary functional responsesover extended periods of time, and that simple measurements of endocrine organ weight or histological examination, which were once thought to be sufficient, may no longer be an adequate reflection of these parameters. Investigations of drug action on the hypothalamo-pituitary system in vivo are complicated by the inaccessibility and sensitivity of the tissues to experimental manipulation. Thus, although gross changes in neuroendocrine function induced by acute or repeated exposure to drugs may be readily assessedindirectly in vivo, by measuring changes in blood concentrations of the hormones produced by the pituitary gland or their peripheral target organs (e.g., the steroids of the adrenal cortex or gonads; thyroid hormones), such studies provide only limited insight as to the mode of action of the active agents or the point in the axis (hypothalamus, pituitary, or peripheral endocrine organ) at which they may act (1). These difficulties have prompted the development of a variety of in vitro models, which unlike their in vivo counterparts, permit direct examination of drug action within the anterior pituitary gland and hypothalamus at the cellular and molecular level. This and Chapter 12 describe the methodology of a number of in vitro preparations that may be used to mvestigate drug action on the anterior pituitary gland and the hypothalamus, drawing attention to their various advantages and limitations. Models of posterior pituitary function in vitro are less widely used and will not be covered here.

In Vitro Models

in Toxicity

Testing

83

It should be noted that all the methods described here involving the use of animal tissue may be subject to governmental approval and that appropriate permission should be obtained before the work is undertaken. 1.1. Basic Principles

of In Vitro Models

The techniques used to maintain adenohypophysial and hypothalamic tissue in vitro fall into two broad categories, short- and long-term. With respect to the anterior pituitary gland, short-term preparations (minuteshours) utilize tissue segments or enzymatically dispersed cells maintained in static or dynamic conditions, whereas long-term preparations (days-weeks) include primary cultures of mixed or “purified” cells and cell lines (e.g., ACTH producing AtT20 cells or growth hormone/prolactin, producing GH, cells). Short-term hypothalamic preparations include isolated nerve endings (synaptosomes) and whole hypothalami as well as tissue slices and fragments, whereas the most important long-term preparations are organotypic cultures derived from hypothalamic fragments and dissociated cell cultures. In all cases, the function of the tissue is assessedby measuring the output of the hormone under investigation, using, in most instances, biological or immunological assays. In designing studies and interpreting data several important points must be taken into account. For short-term studies, it is critical that the tissue should be derived from animals of the same strain, age, weight, sex, and, if female, stage of the estrous/menstrual cycle (which in the rat normally spans a 4-d period). In addition, the diet, housing, lighting, and handling regimes prior to autopsy should be tightly controlled, since these too may influence the subsequent activity of hypothalamic and pituitary tissue in vitro. Ideally, such stringent criteria should also be applied to tissue collection for long-term study. The viability of the tissue in vitro is obviously of paramount importance and, since small changes in the physicochemical environment (e.g., pH, ionic balance, O2 tension, and endotoxin content of commercially available media) may influence the level of cellular activity, the incubation conditions should be rigorously monitored. Finally, medium containing the pituitary or hypothalamic secretions and tissue samples should be stored in appropriate conditions (snap frozen and kept at -20°C for peptides or -70°C for dopamine) prior to stringent analysis by well validated, specific, precise methods. Individual drug treatments may be tailored as required and will

84

Gillies and Buckingham

depend on whether acute or long-term effects are to be investigated. In principle, the animals may be exposed to the drug under consideration in vivo and the respective glands removed and their functional activity assessedin vitro; alternatively, the tissues may be removed from naive animals and exposed to the drug in vitro. Ideally, responses to a range of doses/concentrations of drugs should be examined in two or more models. Particular care should be taken to include suitable vehicle controls, because in many instances, solvents (e.g., ethanol, polyethylene glycol, dimethyl sulfoxide) have been shown to exert marked effects on the basal and neurochemically evoked secretory activity of hypothalamic and pituitary tissue in vitro. 1.2. Advantages and Limitations of In Vitro Models of Anterior Pituitary

Function

The static incubation system utilizing pituitary segments has several advantages. First, the tissue retains its three dimensional structure and thus the cell-cell interactions are representative of those in vivo. Second, the system is highly precise and therefore lends itself to quantitative analysis of drug action. Third, it is simple and inexpensive to perform, requiring little specialized equipment. Its main disadvantage is that tissue viability may be limited by poor diffusion of nutrients into and of metabolites out of the tissue, and experiments should therefore not be continued for longer than 3 h. Tissue viability may be improved by the use of a perifusion system in which pituitary segments from 3-4 rats are incubated in a single chamber (3). Such preparations may provide valuable information about the dynamics of pituitary hormone release. They also permit detailed examination of the rate of onset, intensity, and duration of acute responsesto drugs and are particularly valuable in dynamic studies on drugs, such as steroids, that may exert biphasic actions. However, the potential of perifused tissue to exhibit pronounced variations in the magnitude of the secretory responses to repeated stimulation (3-5) limits their use in quantitive pharmacological studiesinvolving, for example, assessment of agonist/antagonist potency. Models utilizing cells dispersed enzymatically prior to incubation do not suffer from diffusion problems and cell viability is generally good. However, production of a single cell suspension inevitably destroys cellcell contacts that may be important physiologically, and reducesthe opportunity for transmission of autocrine and paracrine influences, which

In Vitro Models

in Toxicity

Testing

85

growing evidence suggests are present within the intact gland. Furthermore, when maintained in static incubation conditions, dispersed cells are often relatively insensitive to physiological secretagogs (6). Nevertheless, the static system described above, in which aliquots of cells derived from a single pool are used, readily lends itself to quantitative pharmacology, since inherent variation is minimal and the responses to a wide range of drug treatments (duration of contact/concentration) and/or secretagogs may be examined in parallel. In contrast to the static incubations, responsesfrom the perifused anterior pituitary cell column are virtually immediate, and this system provides a means whereby nutrient supply is continuously replenished and has rapid accessto all cells while metabolites are readily removed. These features undoubtedly contribute to the fact that this preparation is robust and the most sensitive of all the in vitro preparations described here (7). It is also particularly well-suited to the study of the dynamics of the secretory response. Because the pituitary gland contains large stores of hormones (unlike the hypothalamus), the perifused cells respond to repeated stimulation for periods of up to 8 h (in the region of 50 stimuli), so that several variables may be tested in replicate in one experiment. Since all the responses are produced by a single pool of cells, biological variation is minimized and standard errors are relatively small. However, because one stimulus could possibly influence the magnitude of a subsequent response, it is crucial to randomize doses with at least four replicates per dose. It is also important to realize that, although a single experiment may yield many data, the results are representative of only a single pool of cells, and that repetition of the experiment is essential for comprehensive statistical analysis. This point is also relevant to dispersed cells in static incubation. A criticism often leveled at the use of single cell suspensions is the possibility that the enzymes used for cell dispersal may perturb receptor structure. However, our studies, which substituted finely chopped anterior pituitary gland tissue for the enzyme dispersed cells in a perifusion system, show that responses to a variety of stimuli and combinations of stimuli are unaffected, except for a significant time lag in responsiveness, which we attributed to the altered diffusion characteristics. Cultured pituitary cells (primary culture or cell lines) permit examination of not only the acute but also the more long-term effects of drugs on pituitary hormone secretion in tightly controlled conditions. In many

Gillies

86 I Anterior segment

pttuitary (I/, or’/,

and Buckingham

gland cl ) per well

Lid of multiwell mul~iwell dish with holes bored in it Plastic tubing sealed glue Into lhe holes 1OOpm pore nylon

with mesh

>

FlnOnQ

nunYnG

f

Base portion of multiwell dash

I

/ lml of incubation medium

Incubate

at 37 ‘C asdescribed

in the text

Fig. 1. Adaptation of multiwell tissue culture plates, as described in the text, permits easy, rapid transfer of the pituitary gland segments to fresh medium without the need to handle the segmentsthemselves.

respects the advantages and limitations of cultured cells resemble those of static incubates of acutely dispersed cells. In addition, the cells may undergo some degree of (de)differentiation, and thus their functional activity may not be directly comparable with that of either freshly removed tissue or of the cells in vivo. Before utilizing cultured cells in toxicological studies it is therefore advisable to verify the integrity of the secretory system to be investigated, for example, by comparison with a well established acute in vitro model. 1. 2. 3. 4. 5. 6. 7. 8. 9.

2. Materials For dissection: guillotine, forceps (fine, straight, and bent watchmakers forceps are ideal). Multiwell tissue culture plates adapted as in Fig. 1. Sealed perspex box. 95% 02/5% cop 37°C incubator. Cell dispersal apparatus as described in Frg. 2. Collagenase or trypsin. 37OCwater bath. 37°C shaking water bath.

In Vitro Models in Toxicity Testing Teflon pulley and stirring paddle

Teflon bearing Polythene vial __)

b Tissue fragments

+c&

0 *

--J

d 4 d

f-------

D

I

h b

b

b . .

c

Trypsin solution

3.2 cm-

Fig. 2. Cell dispersal apparatus, 10. 11. 12. 13.

Bio-Gel P2,200-400 mesh. Perifuston apparatus as described in Frg. 3. Media and buffers: Earle’s balanced salt solution (EBSS) (see Table 1). Incubation medium for anterior pituitary segments dispersed cells and perifused cell column. Bovine serum albumin (0.25% w/v) Benzyl penicillin (100 IU/mL,) Streptomycin ( 100 pg/mL) Aprotonin (100 Kalhkrein inactivator U/mL) Earle’s balanced salt solution (100 mL) Note: Artificial cerebrospinal fluid (CSF) may also be used for incubation of pituitary tissue (8). Antimicrobtal drugs are not necessary for mcubation of the segments. 14. “Growth” medium for cultured anterior pituitary cells: Charcoal stripped fetal bovine serum (20%) Nonessential amino acids (1%) L-Glutamme (4 mM) Penicillin (100 U/nL)

Gentamycin (10 pg/mL) Dulbecco’s Modified Eagle’s Medium (100 mL)

Gillies

88 WASTE

-

and Buckingham

0 15ml/mr -+-y-l

TEST SUBSTANCES

I

FRACTION -COLLECTOR

Fig. 3. Perifusedisolated rat anterior pituitary cell bioassay. 3. Methods 3.1. Anterior Pituitary Segments For Materials see Section 2., numbers l-5.

(8,9)

1. Remove the anterior pituitary glands postmortem from 20-30 rats (agematched and of the same strain and sex) that have been handled regularly and housed for at least 1 wk before the experiment in a temperature- and lightcontrolled room (21-23”C, lights on 7 a.m.-7 p.m.) with food and water available ad libitum. Decapitatethe rats, remove the dorsal surface of the skull, lift the frontal lobes, cut the optic nerves, and reflect the whole brain to expose the prtuitary gland. Dissect away the overlymg diaphragma sellae, remove the gland, and discard the centrally positioned neurointermediate lobe. 2. Divide the anterior lobe into 4-6 pieces (dependent on the age and weight of the rat) of approx 1 mm3 size (0.8-1.0 mg wet wt). 3. Pipet an equal volume (1 .O n-L) of the (prewarmed) incubatron medium (see Section 2., number 12b), pH 7.4, into each well of a multiwell tissue culture plate (Fig. 1). Place one segment in each well and then transfer the plates into a sealed perspex box connected to a 95% O&5% CO, gas sup-

In Vitro Models

in Toxicity

Testing

89

Table 1 Earle’s Balanced Salt Solution (EBSS) Concentration,

mg/lOO mL Elevated K+

Basal K+ Constituent NaCl KC1 CaCl, . 2Hz0 MgS04. 7Hz0 NaH2P04 - 2Hz0 NaHC03 D-Glucose

56&K+ 680 40 36 20 15.8 220 100

14mA4K+ 615 104 36 20 15.8 220 100

28 mMK+ 534 209 36 20 15.8 220 100

56mMK+ 370 417 36 20 15.8 220 100

aNaCl is reduced as KC1 increases m order to maintain isotomcity. Ca2+/Mg2+free 56 mM K” as above, excluding Ca2+Cl, s 2H,O and Mg2+S0, s 7H,O Ca2+ chelatmg agent (e g , EGTA, 2 mM) may also be added to “Ca2+ free” medium

ply. Saturate atmosphere with 95% 0,/5% CO2 gas (5 min). Incubate for 2.5 h at 37”C, replacing the medium and stopper gas mletjoutlet and replenishmg 95% 02/5% CO2 atmosphere after 2 h and 2 h 15 min. Note: These incubations may also be performed m an incubator at 37’C in 5% CO,/ 95% air and 100% humidity. 4. Transfer the segments to fresh medium containing the test substancesand/ or appropriate secretagogs (e.g., releasing hormones) or, in the case of controls, an equal volume (1 .OmL) of medium alone or appropriately diluted vehicle, and then incubate in 95% 02/5% CO* saturated conditions as described above for a further 30 min. Note: If required, test substances/ vehicle may also be included m the premcubatlon period. 5. Collect aliquots (0.2 mL) of medium m plastic tubes, snap freeze, and store at -2OOC for hormone assay, or assay immediately. Weigh the pituitary tissue, chill rapidly (-7O”C), and store for subsequent biochemical or histological analysis. Results are normally expressed per mg tissue. 3.2. Dispersed Anterior Pituitary Cells in Static Incubation 3.2.1. General Method for Dispersal The method for cell dispersal is essentially common to all models, although many modifications may be found in the literature. In some cases these variations may depend on personal preference rather than scientific arguments.

90

Gillies

and Buckingham

1. Collect anterior pituitary glands postmortem as described above, cut each lobe into approx 10 pieces with a fresh scalpel blade, and transfer the combined fragments of 5-8 anterior pituitary glands to a dispersal apparatus (Fig. 2) containing 10 mL of Earle’s balanced salt solution supplemented with collagenase (0.1% w/v) or trypsin (0.25% w/v) over which watersaturated 95% 02/5% CO2 gas is bubbled continuously. 2. Place the vial in a water bath maintained at 37°C and incubate for 30 min, agitating constantly by rotating the paddle at approx 200 rpm; aid dispersal with periodic gentle repetitive pipetmg using a clean 1 mL plastic pipet tip or a fire polished, siliconized glass Pasteur pipet. 3. Allow the prtmtary pieces to settle and pour the supernatant fluid mto a plastic tube and store at room temperature. Add a further 10 mL aliquot of enzyme containing medium to the pituitary fragments and repeat the dispersal procedure two more times until all the tissue is dispersed. 4. Centrifuge at 1OOgfor 10 min to recover the cells. Discard the supernatant fluids and resuspend each cell pellet in the medium (10 mL) employed for the subsequent incubation (see Section 2., number 12). 5. Repeat the centrifugation procedure and resuspend the cells from each harvest in medium (1 mL), pool, and filter through 100 pm pore nylon gauze (previously washed in sterile saline) to remove any remaining connective tissue or clumps of cells. 6. Take an aliquot (10 pL) of the cell suspension to determine the cell count. Verify cell viability at this stage using the trypan blue exclusion test (make a 1: 1 mix of cell suspension with 0.4% trypan blue solution m 0.9% NaCl, wait 5 min, and calculate the percentage of cells that do not take up the dye using a hemocytometer). Viability is usually >90% and should not be ~85%. 3.2.2. Static Incubation of Dispersed Anterior Pituitary Cells (6) 1. Dilute the cells to a known concentration, normally 3.3 x 105-5.0 x lo5 cells/ml in incubation medium (see Section 2., number 12), and transfer aliquots of the cell suspension (0.6 mL) to plastic tubes. 2. Incubate for 2 h at 37OCin a shaking water bath, gassing the medium continuously with water saturated 95% 02/5% C02. 3. Centrifuge the cell suspensions (lOOg, 10 min, 4”C), discard the supernatant fluids, and resuspend the cell pellets either in medium (0.6 mL) contaming appropriate drugs and/or secretagogs or, m the case of controls, an equal volume of medium alone or appropriately diluted vehicle. Incubate the cell suspensions for a further 60 min under the conditions described above. 4. Centrifuge the cells (lOOg, 10 min, 4”C), collect the supernatant fluid, snap freeze, and store at -20°C for hormone assay.

In Vitro Models

1.

2.

3.

4.

in Toxicity

Testing

91

3.2.3. For Preparation of the Perifused Anterior Pituitary Cell Column (7) Mix the filtered cell suspension (approx 1 mL) with 0.5 g Bio-Gel P2, 200-400 mesh, which had been previously swollen overnight in 0.9% NaCl, containing antibiotics, and then washed twice with the medium (see Section 2., number 12b). Draw the slurry into a column constructed from a plastic disposable syrmge (Fig. 3) (7), connect to a variable speed peristaltic pump, clamp vertically, and perifuse at a rate of 1 mL/min with medium containing trypsin lima bean inhibitor (0.1% w/v) if trypsin was used for dispersal. When the cells and Bio-Gel mixture have packed down to a volume of 1.5 mL (2-3 min), disconnect the column from the pump, and, taking care that the gel does not dry out, reduce the dead volume to 0.1-O-2 mL and recommence perifuslon with medium at a rate of 0.5 mL/min. Clamp the column vertically in a water bath at 37°C and perifuse for 1.5-2 h to allow hormone secretion to reach a steady baseline level. Connect the column outlet to a fraction collector and collect the eluate as 2-min fractions (Fig. 3). When 2-4 min pulses of hypothalamic regulatory factors, diluted in the perifusion buffer, are passed through the column a response occurs almost immediately and secretion rapidly returns to baseline such that a subsequent stimulus may be given 10 min later (7). Fractions containing the column effluent may either be assayed immediately or snap frozen and stored at -20°C.

3.3. Cultured

Anterior

Pituitary

Cells (10)

Use materials listed in Section 2., numbers 1,4, and 6-9. In addition, at all stages use sterile buffers, media, and plastics and perform the relevant manipulations in a sterile atmosphere provided by an appropriate microbiological safety cabinet. 3.3.1. Culture Preparation 1. Collect anterior pituitary glands postmortem from 9-15 rats, as described earlier, but under aseptic conditions. 2. Pool the tissue and wash several times in sterile Ca2+/Mg2+ free Earle’s balanced salt solution until all the blood has been removed. Place in a Petri dish and chop into cubes (1 mm3) using a sterile scalpel blade. 3. Transfer the pituitary fragments to a plastic vial containing 5 mL Ca2+/ Mg2+ free Earle’s balanced salt solution containing 25% (v/v) pancreatin 4X and 10% (v/v) collagenase (stored as 0.1% [w/v] in Tris buffered CaCl,), and incubate in a water bath at 37°C for 20-30 mm. At the end of

92

4. 5.

6. 7.

Gillies and Buckingham this period, aid dispersion by gentle repetitive pipetmg. Allow the pituitary fragments to settle and transfer the supernatant fluid to a plastic tube containing 5 mL unstripped fetal bovine serum, stopper, and store at room temperature. Add a further 5 mL Ca2+/Mg2+free Earle’s balanced salt solution containing pancreatin 4X and collagenase to the remammg tissue and repeat the procedure. Collect four such harvests and centrifuge (lOOg, 5-10 min). Discard the supernatant fluids and resuspend each cell pellet in 5 mL “growth medium” (see Section 2., number 12~). Pool the cell suspensions and centrifuge (lOOg, 5-10 min). Discard the supernatant fluid and resuspend the cell pellet in 10 mL growth medium. Repeat the centrifugation, again suspending the resulting cell pellet in 10 mL growth medium. Take an ahquot (10 p,L) to determine the cell count and to verify cell viability using the trypan blue exclusion test as described earlier. Dilute the cells in the growth medium to a concentration of lo5 cells/ml. Add 200 PL aliquots of this cell suspension to each well of a 96-well culture plate and incubate m sterile conditions at 37°C in a 5% C02/95% air atmosphere for 72 h or until the cells grow to confluence.

3.3.2. Assessment of Function 1. Decant the growth medium and wash the cultured cells twice with incubation medium (see Section 2., number 12~). 2. Add a further 200 p.L medium to each well and incubate at 37°C for 1 h in an atmosphere saturated with 5% CO,/95% air. Decant the medium and replace with an equal volume (200 pL) containing appropriate drugs and/ or secretagogs, or m the case of controls, an equal volume of medium alone or appropriate diluted vehicle. 3. Incubate at 37°C in 5% CO,/95% an for 2 h. 4. Remove the supernatant fluid from each well and store at -20°C for assay. 4. Notes 1. All buffers, incubation media, and culture media are to be used prewarmed to 37°C. Media and additives are readily obtained from major suppliers. Unless indicated, additives are most conveniently stored at -20°C for up to 3 mo in 0. l-l .OmL vol aliquots at concentrations that, when added to 100 mL of medium, produce the desired concentration. For cell culture, solutions containing additives that are not originally sterile (e.g., powdered additives) need to be filter sterilized through 0.2 km pore disks that fit on to sterile disposable plastic syringes (commercially available). The osmolality of all media should be within the range 290-300 mosM/L.

93

In Vitro Models in Toxicity Testing

2. In order to obtain a high index of precision m the anterior pituitary seg3. 4. 5. 6.

ment model we recommend that n = 5 for each treatment and thus rt is feasible for a single operator to assess approx 16 variables/experiment. Owing to diffusion problems, static pituitary segment incubations should not be continued for more than 3 h. For static incubations of dispersed cells, relatively long periods of contact with secretagog (>l h) may be required before a response is seen. When using perrfused isolated cells, a response may be seen within mmutes. When cutting anterior pituitary glands into fragments use a clean, sterile scalpel blade wrth a smooth, not a dragging, motion to make the cut.

References 1. Buckingham, J C and Gillies, G. E. (1992) Hypothalamus and pituitary gland. xenobiotic induced toxicity and models for its investrgatton, in Endocrine Toxicology (Flack, J. D and Atterwrll, C., eds.), Cambridge University Press, pp 83-114 2. Besser, G M and Cudworth, A G. (eds.) (1987) Clinical Endocrinology Chapman and Hall, London. 3. Busbridge, N. J , Chamberlain, G. V. P , Grrffiths, A., and Whitehead, S. A. (1990) Non-steroidal follicular factors attenuate the self-prrmmg action of gonadotrophin releasing hormone on the pmutary gonadotroph. Neuroendocrinology 51,493-499. 4. Buckingham, J C and Cover, P. 0. (1986) Changes in the responsiveness of perifused rat adenohypophyseal cells to luteinizing hormone releasing hormone. Acta Endocrinol.

(Copenhag.)

113,479-486.

5. Cover, P 0 and Buckingham, J. C. (1989) Effects of estradiol and tamoxifen on GnRH self priming m perifused rat adenohypophysial cells. Acta Endocnnol. 121, 365-373 6. Cowell, A.-M., Flower, R. J., and Buckmgham, J. C. (1991) Studies on the roles of phospholipase A2 and elcosonotds in the regulation of corticotrophin secretion by rat pituitary cells m vitro. J. Endocr. 130,21-32.

7. Gilhes, G. E. and Lowry, P. J. (1978) Perfused rat isolated anterior pituitary cell column as bioassay for factor(s) controllmg release of adrenocortlcotropin:

valida-

tion of a technique. Endocrinology 103,521-527. 8. Buckingham, J. C. and Hodges, J. R. (1977) The use of corticotrophin production by adenohypophysial tissue in vitro for the detection and estimation of potential corticotrophin releasing factors. J. Endocr. 72, 187-193.

9. Hadley, A. D., Flack, J. D., and Buckingham,

J. C. (1992) Effects of selective

phosphodlesterase mhlbttors on the release of ACTH and LH from rat anterior pituitary segments in vitro. Pharmacol. Commun. 3,283-295. 10. Stone, M A., Carey, F , Cowell, A -M , and Buckingham, J C (1989) Etcosanoids

and piturtary function: a role for phospholipase AZ in the stimulation from cultured pituitary cells. Br. J. Pharmacol. 96, 162P.

of ACTH

CHAPTER12

The Application of In Vitro Models of Hypothalamic Function in Toxicity Testing GZenda E. GiZZies and Julia

C. Buckingham

1. Introduction The reader is referred to the previous chapter in which the background and rationale for monitoring hypothalamic function in vitro is discussed. 1.1. Advantages and Limitations of In Vitro Models of Hypothalamic Function Like other in vitro systems, the isolated whole hypothalamus permits the examination of hormone release in a precisely controlled physical and chemical environment. It also has the advantage that it largely retains the three-dimensional relationships between neurons and non-neuronal supporting cells and, thus, it is approximately representative of the prior state in vivo. Several workers have expressed concern about the viability of a tissue block of the size employed. This is an important consideration since, according to Lumsden (I), 1 mm3 is the maximum volume of brain tissue that will permit adequate diffusion of nutrients and metabolites to and from the center of the tissue and thereby maintain cellular integrity. Cells located near the edge of the larger explants retain their morphological and biosynthetic characteristics in vitro, but those at the center may undergo necrosis. The viability studies of Bradbury et al. (2) showed that the O2 consumption of the isolated hypothalamus is linear over a 3 h period (range 68.9-120 w/g/h, n = 5). Similar data have been reported by Berelowitz et al. (3), who also observed raised O2 consumption in the From Methods m Molecular B/ology, Edlted by: S O’Hare and C. K. Atterwlll

Vol. 43: In V/fro Toxmty Testmg Protocols Copyrtght Humana Press Inc , Totowa, NJ

95

96

Gillies and Buckingham

presence of an elevated K+ concentration (178 * 45 p,M/glh, mean + SE, n = 3). Examination histologically of the tissue after a 3 h incubation revealed signs of deterioration but, despite a variable degree of perineuronal edema, there was no indication of neuronal death. It thus appears that the tissue viability is limited. If one allows a 60 mm pre-incubation period to minimize leakage from severedneuronal surfaces and to enable the tissue to recover from the trauma of excision, the preparation is suitable only for short-term experimental studies involving at the most two successive stimulations. Nevertheless, there are reports of more prolonged incubations and one group (4) has successfully maintained the explants for more than 24 h. Thus, the isolated whole hypothalmus provides an invaluable tool with which to examine the physiology, pharmacology, and biochemistry of the hypophysiotrophic neurons. It is currently employed in this and other laboratories in studies on the maturation and agmg of hypothalamic neurons, the second messenger systems effecting peptide/transmitter release, the modes of action of steroids and other neuroactive drugs, and the interactions between the immune and neuroendocrine systems. As mentioned earlier, the problems of diffusion and viability may be reduced by using small tissue pieces and/or perifusion systems. Inevitably, the improvements obtained with tissue fragments are at the expense of maintaining cellular integrity and the three-dimensional structure, which limits the usefulness of the models. Perhaps the most effective compromise 1sthe bissected organ (5) in which the improvements in diffusion are coupled with preservation of a high degree of cellular integrity together with the normal intrinsic network and supporting glia, and so forth. This technique has been successfully exploited in several laboratories. Superfusion procedures appear to improve viability, and hrstological studies indicate that there is little deterioration for periods of up to 4-5 h. Theoretically, therefore, such preparations may be expected to lend themselves to repetitive stimulation. However, in practice they frequently exhibit marked tachyphylaxis and, since very small quantities of neuropeptides are released, it is often necessary to concentrate/lyophylize the eluate fractions in order to make reliable measurements, even though as many as 12 hypothalami may be used in each perifusion chamber. In our hands, therefore, we feel that the disadvantages of the perifusion technique outweigh its advantages.

In Vitro Models The advantages of using hypothalamic cells in culture include the virtual removal of diffusion problems and the provision of an opportunity to study cellular responses directly under controlled conditions (especially when using defined medium) without interference from the many homeostatic mechanisms that operate through the hypothalamus. This model, therefore, allows one to investigate repeated responses over many weeks in vitro and also the effects of prior treatments (acute vs chronic) on subsequent responses in a manner that cannot be achieved in vivo or in other in vitro experiments. Because fetal or early neonatal tissue has to be used, hypothalamic cultures are particularly useful for the study of factors that influence the development of the neuroendocrine system. It should be noted, however, that the development of the cells may proceed in a distinctly different manner, depending on the culture medium used. In particular, many cultures are grown in the presence of serum that, although promoting growth and survival, is a variable, undefined, and unphysiological cocktail of bioactive substances that do not normally come into contact with brain cells. Thus, we have shown that the responsiveness of hypothalamic peptidergic neurons becomes distinctly suppressed when grown in a serum-supplemented medium compared with those grown in a defined, serum-free medium and that the cell types that survive in each of these media are distinctly different (6-8). In fetal/early neonatal tissue the levels of certain neuropeptides may be low (e.g., corticotrophin-releasing hormone and vasopressin) and this could therefore, present a problem if detection systems are not sufficiently sensitive. Another potential disadvantage of using isolated hypothalamic cells is the loss of the normal cytoarchitecture found in VWO, but studies using cortical cultures suggest that a certain degree of histotypical reorgamzation occurs in vitro. Generally, however, cultured hypothalamic cells behave in many ways, as would be expected in vivo. They exhibit both a morphological and functional maturation (6-12) and they also exhibit electrical activity and synaptogenesis as well as Ca2+ dependence of their secretory activity. In context, therefore, they offer certain advantages over other systems. However, like all experimental models, this preparation should not be used in isolation, but results should be compared with other in vitro models as well as ultimately being verified in vivo.

Gillies

1. 2. 3. 4. 5. 6. 7. 8.

9. 10. 11. 12. 13.

14.

15.

and Buckingham

2. Materials Guillotme, scissors, and forceps for dissection, Shaking water bath. 95% o,/co,. Anesthetizing box. Sterile culture facilities (aseptic working area, autoclave, sterilizing oven, humidified CO, incubator, Class II microbiologtcal safety cabinet, sterile plastics). Inverted microscope. Buffers and media. Incubation medium for hypothalamic tissue (artificial cerebrospinal fluid [CSFJ): 126 mM NaCl, 6 mM KCl, 1mM NaJ-IPQ, 0.877 mM MgSO,7H,O, 22 mMNaHCOs, 1.45 n&f CaC12,and 200 mg/% glucose. When madeup in distilled and deionized water, osmolarity should be 298 mosM/L. Sterile saline: 100 IU/mL pencillm, 100 ug/mL streptomycin, and 100 mL sterile saline (0.9% NaCl). Collection buffer: 0.01% bovine serum albumin (BSA Fraction V), 100 IU/mL pencillin, 100 pg/mL streptomycm, 20 mJ4 HEPES, and 100 mL Hanks’ Balanced Salt solution (HBSS, Ca2+/Mg2+free). Enzyme solution: 0.04% BSA, 100 IU/mL pencillin, 100 pg/mL streptomycin, 0.2 mg/mL Dispase I, 0.5 mg/mL DNase I, 20 rnil4 HEPES, and 40 mL HBSS (Ca2+/Mg2+free). DNase solution: 0.01% BSA, 100 IU/mL pencillin, 100 pg/rnL streptomycin, 20 mA4HEPES, 1 mg/mL DNase, and 20 mL HBSS. Serum supplemented culture medium: 10 mL heat inactivated fetal calf serum (FCS), 10 mL heat inactivated donor horse serum (HS), 100 IU/mL penicillin, 100 pg/mL streptomycin, 0.5 pg/mL fungizone, and 100 mL Dulbecco’s Modified Eagle’s Medium (DMEM). Defined medium: 50 mL DMEM, 50 mL Ham’s F12 nutrient medium, 100 IU/mL penicillin, 100 pg/mL streptomycin, 0.5 pg/mL fungizone, 10 n&f HEPES, 5 pg/mL insulin, 100 pg/mL transferrin, 2 x lo-*M progesterone, 3 x lo-*M selenmm, lOAM putrescme, 10-12M P-estradiol, and 10e9Mtriodothyronme. Release medium: 30 pg/mL ascorbic acid, 30 pg/mL bacitracin, 0.01% w/v bovme serum albumin, 0.1% w/v glucose, 10 n&f HEPES buffer, 100 IU/mL penicillin, 100 ug/mL streptomycin, 80 KIU/mL trasylol, and 100 mL EBSS.

2.9. Siliconization of Pasteur Pipets 1. Wash 230 mm unplugged glass Pasteur pipets m distilled water for 1 h in a gas chromatography Jar, rinse in methanol, and dry at 100°C in an oven.

In Vitro Models 2. When cool, rinse the pipets individually in the repel coat, dimethyldichlorosilane solution, in a 100 mL measuring cyclinder for approx 10 s and dry at 100°C before the final distilled water wash (1 h), followed by a brief rinse in methanol and final drying in the oven. 3. When cooled again, fire polish the Pasteur tips in a Bunsen flame to give tips of varying diameters with no rough edges. With non-absorbent cotton wool, pack into a metal box and heat sterilize at 180°C for 2 h. Once opened, sterile Pasteur storage boxes are kept in the microbio-

logical safety cabinet. 3. Methods 3.1. Short Term Incubation of Hypothalamic Fragments (13,14) Use materials in Section 2., items l-3. 1. Collect hypothalami post mortem from rats (age-matched and of the same strain and sex) that have been handled regularly and housed for at least a week before the experiment in a temperature- and light-controlled room (21-23”C, lights on 7 AM-~ PM)with food and water available ad libitum. Remove the dorsal area of the skull immediately after decapitation and lift the frontal lobes gently. Cut the exposed optic nerves and reflect the whole brain. Dissect out the hypothalamus using fine scissors;the tissue block taken is bordered rostrally by the optic chiasma, laterally by the hypothalamic fissures, caudally by the mammillary bodies, and dorsally by the ventral surface of the thalamus. 2. Lift the tissue carefully handling only by the cut ends of the optic nerves and transfer to a plastic incubation vial containing 1.0 mL of medium (Section 2.1.), pH 7.4., pre-warmed to 37OC and pregassed with watersaturated 95% 02/5% C02. 3. Incubate for 1 h at 37°C in a shaking water bath in an atmosphere saturated with 95% 0,/5% COZ. Replace the medium after 30 min. 4. Transfer the hypothalami either to medium (1 .OmL) containing the test substance and/or appropriate secretagogs or to a corresponding volume of medium alone or appropriate diluted vehicle (controls). Incubate for a further 30 min in the conditions described above, Collect medium, freeze, and store at -70°C for assay of peptides and/or neurotransmitter substances.Freeze the tissue in liquid Nz or on solid CO2 and store (-70°C) for subsequent biochemical studies or histological examination.

If the actions of a drug with a slow onset is to be investigated (e.g., steroids), the compound may also be included during earlier incubation periods.

100

Gillies and Buckingham

3.2. Modifications Utilizing Hypothalamic Fragments andfor Perifision Systems The viability of hypothalamic tissue in vitro may be prolonged for more long-term studies (up to 12 h) by utilizing smaller tissue pieces and/or perfusion systems. Thus, the hypothalamic blocks (removed as described above) may be bisected by a mid-sagittal incision, quartered by perpendicular mid-sagittal and coronal cuts, or sliced (200-250 pm) in the sagittal plane. Alternatively, a smaller total hypothalamic mass, for example the medial basal hypothalamus or median eminence, may be employed, the dissections of which are described in detail elsewhere (15,16). In addition, perifusion techniques may be applied to whole hypothalami (17,18) and to hypothalamic fragments (15,19-22) or slices (23,24) in a manner essentially similar to that described above for pituitary tissue. 3.3. Primary Cultures of Dissociated Hypothalamic Cells (6,7) Use materials in Section 2., items 1-6. All procedures described below involve aseptic techniques. All dissection instruments are either heat sterilized (large scissors, forceps, and so forth) or autoclaved (fine forceps, scissors, and so forth that have autoclavable protective plastic tips, and so on) before use. The polythene cell dispersal pot (as described in Fig, 2 of Chapter 11) is sterilized by soaking in 70% ethanol for 24 h with thorough rinsing with sterile saline before use. The Teflon head of the dispersal apparatus is autoclaved before use. Cells and sterile medium are handled only in the microbiological safety cabinet with sterile pipets. 3.3.1. Tissue Dissection 1. Removal of fetal rats from the pregnant mothers IS carried out m a nonsterile, clean area outside the trssueculture laboratory. Place the timed 18 d pregnant rats, one at a time, in an anesthetizing box. After inhalation of ether vapor, dams rapidly become anesthetized and removal of fetuses is performed under contmued anesthesia (after whrch the mothers are killed by methods as approved by the Home Office). 2. Soak the abdomen with 70% ethanol and make a large rmd-lme incision. Hold back the skm and abdommal walls wrth Spencer Wells clasps, ensuring that the fur does not touch the uterus.

In Vitro Models

101

3. Remove the uterus containing the fetuses and place it in a 70-mm bacteriological sterile plastic Petri dish containing sterile saline supplemented with antibiotics (Section 2.2.). 4. Transfer the uterus to the tissue culture laboratory and remove the fetuses one by one from the uterus. 5, Decapitate the fetuses and place the heads in a sterile Petri dish containing a piece of filter paper, moistened with sterile saline supplemented with antibiotics to provide a non-slip surface for the subsequent dissection. 6. While securing the head with forceps in the eye sockets, make one midline cut through the dorsal fetal skull, ease out the whole brain, and invert it through 180°C so that the hypothalmus is visible. 7. Remove the hypothalami using fine watch-makers’ forceps to take a “pinch” of tissue that includes the “hypophysiotrophic area” bordered by the hypothalamic sulci laterally, the mamillary bodies caudally, and the optic chiasm rostrally. The area removed, therefore, includes the secretory nuclei known to contain the hypothalamic factors that control anterior pituitary hormone secretion and weighs approx 1 mg (wet weight). 8. Place the dissected hypothalami into a sterile Petri dish containing the collection buffer (Section 2.3.) that has been warmed previously to 37°C in an incubator. If more than five pregnant rats are used, a second Petri dish containing warmed and gassed collection buffer is used for subsequent collection of hypothalami while the first dish is kept in a humidified CO2 incubator at 37°C. With practice, up to 100 hypothalami (approx 10 mothers) may be collected by two workers in 80 min. 3.3.2. Cell Dispersal From this stage onward all procedures take place in the sterile environment provided by a microbiological safety cabinet. 9. Transfer all hypothalami to a 30 mL plastic sterile Universal vessel using an automatic pipet fitted with a sterile disposable pipet and add fresh collection buffer. 10. Gently pellet the tissue at 800 rpm for 3 min and then resuspend in 15 mL of previously warmed (37°C) enzyme solution (Section 3.4.) and transfer to the sterile dispersal pot (Fig. 2). 11. Place the dispersal apparatus into a clean water bath at 37OCcontaining fresh tap water and Roccal disinfectant. The contents of the dispersal apparatus are gently gassedwith 5% COZ: 95% O2 and mechanically agitated by the paddle of the apparatus, driven by a motor, rotating at 100 rpm for 30 min at 37°C.

102

Gillies

and Buckingham

12. Remove the supernatant containing the dtspersed cells mto a Universal container and store at 37OC. 13. Add 2 mL. of fresh enzyme solution to the tissue fragments and gently triturate 10 times with a fire polished siliconized sterile Pasteur pipet with a tip bore diameter of approx 1 mm. Add a further 13 mL of enzyme solution and reconnect the dispersal apparatus to the motor for a further 15-20 min at 37°C. 14. Collect the supernatant as above. Any tissue fragments remaining after this second dispersal procedure can be gently trtturated with siliconized Pasteurs of decreasing tip bore diameters until all tissue appears to be dispersed. 15. Pool andcentrifuge dispersed cells at 1100 rpm for 8 min in a 30 mL Universal container; resuspendthe pellet in 2 mL of DNase solution (Section 2.5.). 16. Gently disperse the cell pellet using a wide diameter Pasteur pipet and then make up the volume of the DNase solution to 20 mL. This suspension is then allowed to stand at 37°C in the CO2 Incubator for 15 min. The aim of this step is to remove nucleic acid released from damaged cells because it would make the subsequent cell pellets very “sticky” and diffrcult to redisperse. 17. Centrifuge at 1100 rpm for 8 min and resuspend the pelleted cells in 2 mL of collection buffer previously kept at 37°C. 18. Gently layer the cell suspension onto 20 mL of collection buffer (Section 2.3.) containing an extra 4% w/v BSA and centrifuge at 1100 rpm for 8 min. 19. Resuspend the pelleted cells in normal collection buffer (Section 2.3.) and estimate the total number of cells and cell viabihty (trypan blue exclusion test as described m the prevrous chapter) using a hemocytometer. 20. Centrifuge at 1100 rpm for 8 min and resuspend in an appropriate volume of either serum supplemented or defined culture medium (Sections 2.6. and 2.7.) to yield 2.5 x lo6 cells/mL. 21. While continually gently swirling the cell suspension transfer 1 mL to 35-mm diameter sterile, plastic tissueculture dishes,swirling each dish gently to ensure even distribution of cells. All 35 mm dishes have been previously coated with filter sterilized poly-L-lysine (10 pg/mL) for 30 min at room temperature, followed by two rinses with HBSS before incubating at 37°C for 2-3 h with serum supplemented culture medium until ready for use. 22. Maintain the cells in an incubator at 37°C in 5% CO*, 95% air, and 100% hurmdity. 23. Change the medium after 4 d, decreasing the sera concentrations, if used, to 5%. Thereafter the medium ISchanged every 4 d or after experimentation.

In Vitro Models

103

3.3.3. Functional Assessment Studies on the release of peptides and neurotransmitters (GABA and dopamine) that influence anterior pituitary hormone secretion will be described here, Release under basal or control conditions is compared to the release in the presence of the substance(s) under investigation. 24. Remove the growth medium and rinse the cells (on/off) with 1 ml/dish of basal release medium (Section 2.8.) prewarmed to 37OC.The cells are then equilibrated for 15 mm at 37°C again with release medium before the final timed mcubation. 25. For basal release, 500 PL of release medium is added to each culture dish for the required period of time. A further control should be included if the drug to be studied requires dilution m a vehicle other than the release medium. For stimulated peptide release, release medium containing 56 mM K+ (with Na+ concentration reduced to maintain isotonicity, as described m the previous chapter) or various concentrations of the substance(s) under investigation are added m 500 lrL vol to the dishes and incubated at 37OCfor the required period of time. 26a. If peptides are to be measured, proceed as follows: At the end of the experimental epoch, collect the medium from each dish into plastic tubes containing 50 pg Polypep and HCl(50 mL of l.OM). The samples are then heated to 70°C for 5 min to destroy peptidase activity and centrifuged at 11OOgfor 5 min to remove any precipitate or cell debris. Providing sensitive immunoassays are available, the neuropeptides may be measured directly in the release medium, which does not interfere significantly in the majority of assays(although this should always be checked). However, it is also possible to concentrate samplesfor pepttde detection (6,7) (see Note 1). b. If GABA is to be measured, proceed as follows: Using HPLC with electrochemical detection, GABA can be monitored after derivatization with o-phthaldehyde 2 min prior to sample injection in order to produce electrochemically detectable reactants (9). Incubation medium may be stored at -20°C so that measurements can be done at a later date. c. If dopamine is to be measured, proceed as follows: Collect the incubation medium onto a small volume of perchloric acid (PCA) to give a final concentration of 0.M. Flash freeze and maintain at -70°C for storage. Before measurement by HPLC with electrochemical detection, return the sample to pH 8.5 with Tris-EDTA buffer (pH 8.6, 1.5it4) and load onto alumina cartrrdges. Elute the catecholamines with 200 mL PCA (0.M) and inject onto the HPLC column (9). (N. B. This procedure is not as easy as it sounds, but further discussion is beyond the scope of thts chapter.)

104

Gillies

and Buckingham

Notes 1. Make up fresh each month. 2. Store as lyophyhzed powder. 3. Ethanol is required for the initial dissolution of a few milligrams. Subsequent dilutions to prepare stock solutions may be done in an aqueous vehicle (0.9% NaCl or EBSS) and hrgh drlution ensures that the vehicle has no effect on the cultures. 4. Concentration of samples for peptrde detection (6). To detect all three peptides (CRF-41, AVP, and SRIF), a concentration of medium 1s required using silica columns (C, and Cs) prepared in 1 mL disposable plastic syringes. The bottom of each column 1s sealed with porous Teflon disks cut from a sheet using a No. 2 cork borer. Silica is poured into the syringe to a depth of approx 6 mm and sealed on top with another Teflon disk. Newly prepared columns are washed with 5 x 1 mL methanol, pushed through the column usmg a 1 mL plasttc syringe plunger with a tip resistant to solvent corrosion. Columns are then washed with 5 x 1 mL 10 mM HsPO,, followed by 2 x 1 mL 80% acetomtrile/20% 10 mM H3P04 before presaturation of nonspecific bmding sites of the silica with polypep (1 mg/mL in 10 mM HsPO,). Samples are loaded first onto the C4 column and eluted directly onto the Cs column since no one resin gives optimal recovery for all 3 peptides. Peptides are eluted with 80% CH$ = N/20% H3P04 (10 mill), which is then evaporated to dryness under vacuum. Samples may be stored at -20°C prior to reconstitution m radtoimmunoassay buffer and measurement. 5. In general, hypothalamic trssue is more delicate and requrres more careful handling than pituitary tissue. 6. Although an “arttfictal CSF” has been recommended as the mcubation medium for hypothalamtc tissue (2), Earle’s balanced salt solution may also be used (26). These media may either be purchased or made up m deionized water immediately prior to collection of the hypothalamic tissue. The media may be supplemented with protease mhrbitors (e.g., aprotinin, 0.5%), reducing agents (e.g., ascorbic acid 10M3m, and albumm (e.g., bovine serum albumin, 0.25%) to minimize degradation or adsorptton of the released hormones. However, the necessity for such additives is disputed because, although the hypothalamus is rich in proteolytic enzymes (27), Berelowitz et al. (3) reported good recovery of somatostatin from supplement-free medium, possibly because the enzymes released mto the medium are inactive or effectively diluted out by the relatively large incubation volume. Drugs may be Included in the medium during the initial and/or final mcubation period. They are normally diluted in medium immediately prior to use and pH adjusted to 7.4

105

In Vitro Models

7. When making up media, dissolve all salts except CaC12in deiomzed water at room temperature: bubble water saturated 95% 0,/S% CO* gas through the solution for 5-10 min before adding CaCl, and, subsequently, glucose (this procedure prevents the CaCIZ from precipitating out). Check pH (7.4) and warm to 37°C while continuing to pass the 95% 02/5% CO2 gas through the solution. (N. B. The K+ concentration of this solution IS approximately twice that of mammalian CSF; this improves the viability of the isolated hypothalamus preparation 121). 8. When using acute hypothalamic incubations, we recommend that n = 5 per treatment group for reliable statistics. This enables a single operator to analyze up to 5 variables per experiment. 9. When usmg hypothalamic cultures, for reliable statistics we recommend that it = 4 per treatment and, since 28 wells may be plated for every 10 mothers used, 7 variables may be tested on any given day and retested at 48 h intervals over 4-6 wk in vitro. However great care must be taken with randomization of the cultures between treatments and each result should be repeated several times on different batches of cells. 10. As well as measurmg neurotransmitter/neuromodulator release or content, hypothalamic cultures may also be used to investigate the cellular localization of the substances under mvestlgation using immunocytochemistry (8,9) and we are currently investigating the use of in situ hybridization methods to identify their cells of synthesis. References 1. Lumsden, C L. (1986) Nervous tissue in culture, in Structure and Function of Nervous Tissue (Bourne, G H , ed.), Academic, New York, pp 67-140 2. Bradbury, M W. B., Burden, J , Hillouse, E W., and Jones, M. T (1974) Stimulation electrically and by acetylchohne of the rat hypothalamus in vitro. J. Physiol. 239,269-283. 3. Berlowitz, M., Kronheim, S., Pimstone, B , and Sheppard, M. (1978) Potassmmstimulated calcium-dependent release of lmmunoreactlve somatostatin from incubated rat hypothalamus. J. Neurochem. 3, 1537-1539 4 Calogero, A E , Gallucci, W. T , Bernardini, R., Saoutis, C , Gold, P W., and Chrousos, G. P (1988) Effect of cholinergic agonists and antagonists on rat-hypothalamic corticoctrophm-releasing hormone secretion in vitro. Neuroendocrinology47,303-308. 5 Maeda, K and Frohman, L A. (1980) Release of somatostatin and thyrotropmreleasing hormone from rat hypothalamic fragments in vitro. Endocrinology 106, 1837-1842. 6 Clarke, M J O., Lowry, P J , and Gdlies, G. E. (1987) Assessment of corticotropinreleasing factor, vasopressin and somatostatin secretion by fetal hypothalamic neurons in culture Neuroendocrinology 46,147-154.

106

Gillies

and Buckingham

7 Clarke, M. .I. 0. and Gilhes, G. E. (1988) Comparrson of peptide release from fetal rat hypothalamic neurones cultured in defined media and serum-containmg media. J. Endocrinol.

116,349-3X

8. Davidson, K. and Gillies, G E (1993) Neuronal vs ghal somatostatin in the hypothalamus: a cell culture study of the ontogenesis of cellular locatron, content and release. Brain Res. 624,15-84 9 Murray, H. E. and Gilles, G. E. (1993) Investigation of the ontogenetic patterns of rat hypothalamic dopammergic neurone morphology and function in vitro J Endocnnol.

139,4031114.

10. Murray, H. E. and Gillies, G. E. (1992) Maturation of morphologrcally distinct subsets of dopaminergic neurons in primary cultures of dissociated hypothalamic cells. J. Endocrinol 132, Abstract 198 11. Davidson, K. and Gillies, G. E. (1992) GABAergic influences on somatostatm secretion from hypothalamic neurons cultured m defined medium. Neuroendocrznology 55,248-256

12. Gillies, G. E , Cover, P O., Loxley, H. D , and Buckmgham, J C. (1992) Evidence for a hypothalamic involvement durmg the stress hyporesponsiveness period in the rat using in vivo and in vitro models. (manuscript submitted). 13. Buckingham, J. C. and Hodges, J. R. (1977) Production of corticotrophin releasing factor by the isolated hypothalamus of the rat. J. Physiol. (Lond.) 272,469-479. 14 Loxley, H. D., Cowell, A.-M., Flower, R. J , and Buckingham, J C (1992) Modulation of the hypothalamic-pituitary-adrenocortical response to cytokmes m the rat by Lipocortin-1 and glucocorticoids: a role for lipocortin-1 in the feedback inhibiton of CRF41 release Neuroendocrinology 57,801-814. 15 Terry, L. C., Rorstad, 0. P., and Martin, J. B. (1980) The release of biologically and rmmunologically reactive somatostatin from perifused hypothalamus fragments. Endocrinology 107,794-800. 16. Negro-Vilar, A., Ojeda, S. R., Arimura, A., and McCann, S. M. (1978) Dopamme and norepine-phrine stimulate somatostatm release by median emmence fragments in vitro. Life Sci. 23, 1493-1498. 17. Kim, K. and Ramirez, V. D. (1982) In vitro progesterone stimulates the release of luteinizmg hormone-releasing hormone from superfused hypothalamic tissue from ovariectomized estradiol-primed prepubertal rats. Endocrinology 111,750-756 18. Redei, E., Branch, B. J., Gholami, S., Lin, E. Y. R., and Taylor, A. N. (1988) Effects of ethanol on CRF release in vitro. Endocrinology 123,2736-2743. 19. Gallardo, E. and Ramirez, V D. (1977) A method for superfusion of rat hypothalami: secretion of lutemizing hormone releasing hormone (LH-RH) (39749). Proc. Sot. Exp. Biol. Med.

155,79-84.

20. Drouva, S. V., Epelbaum, I., Heri, M , Tapia-Arancibia, L., Laplante, E , and Kordon, C. (1981) Ionic channels involved m the LHRH and SRIF release from rat medrobasal hypothalamus. Neuroendocinology 32, 155-162. 21. Shimatsu, A., Kato, Y., Matsushita, N , Katakami, H , Yanaihara, N , and Imura, H. (1982) Effects of glucagon, neurotensisn and vasoactive intestmal polypeptrde on somatostatin release from perrfused rat hypothalamus Endocrmology 110, 2113-2117.

In Vitro Models

107

22. Gillies, G. E., Puri, A., Hodgkison, S., and Lowry, P. J. (1984) Involvement of rat corticotrophin-releasing factor-41-released peptide and vasopressin in adrenocorticotrophin-releasing activity from super-fused rat hypothalamic in vitro. J. Endocrinol.

103,25-29

23. Kelly, M. J., Condon, T. P , Levine, J. E., and Ronnekleiv, 0. K. (1985) Combined electro-physiological, immunocytochemical and peptide release measurements in the hypothalamic slice. Brain Res. 345,2&I-270 24. Nikolarakis, K. E., Almeida, 0. F. X., and Herz, A. (1986) Corticotropm-releasing factor (CRF) ‘inhibits gonadotropin releasing hormone (GnRH) release from superfused rat hypothalami in vitro. Brain Res. 377,388-390. 25. Gillies, G. E., Anderson, R., Davidson, K., and Cross, A. (1992) Release of endogenous GABA from primary cultures of drssociated hypothalamic neurons. J Endoc 132, Abstract 220. 26. Leposavic, G., Dashwood, M. R., Ginsberg, J., and Buckingham, J. C. (1990) Peripubertal changes in the nature of the GnRH response to alpha-adrenoceptor stimulation in vitro and their modulation by testosterone Neuroendocrinology 52, 82-89.

27. Griffiths, E C., Jeffcoate, S. L., and Holland, D. T (1977) Inactivation of somatostatin by peptidases in different areas of the rat brain. Acta Endocrinol. (Copenhagen) 85, l-10.

CHAPTER13 The FRAME Cytotoxicity (Kenacid Blue) Richard

Test

H. Clothier

1. Introduction The cytotoxic effect of chemicals on cells in culture is measured by the change in total cell protein (Kenacid Blue R dye binding method). Healthy 3T3-Ll cells (an established cell-line, ATCC CCL92. l), when maintained in culture continuously divide and multiply over time, The

basis of this test is that a cytotoxic chemical (regardless of site or mechanism of action) will interfere with this process and, thus, result in a reduction of the growth rate as reflected by cell number. The degree of inhibition of growth, related to the concentration of the test compound, provides an indication of toxicity. 3T3-Ll Cells are maintained in culture and exposed to test compounds over a range of concentrations. The cultures are visually examined after 24, 48, and 72 h, and the number of viable cells and/or the total cell protein content are determined, after either 24 or 72 h exposure, by the

Kenacid Blue method. This assay may be performed on cells previously used for the Neutral Red uptake assay. The number of cells in the presenceof test chemicals is compared with that observed in control cultures and the percent inhibition of growth calculated. The IDSa, ID,,, and IDso concentrations (i.e., the concentrations producing 20, 50, and 80% inhibition of growth) are determined and expressed as jtg/mL or n-H. These values enable a comparison of the relative cytotoxicity of the test compounds. From Methods m Molecular Biology, Vol. 43 In V/fro Toxioty Testrng Protocols Edlted by* S O’Hare and C K Atterw~ll CopyrIght Humana Press Inc , Totowa, NJ

109

210

Clothier

The maintenance and culture of a cell line, such as 3T3-Ll cells, is a relatively simple and inexpensive technique. The application of such cultures to determine cytotoxicity enables the rapid, highly reproducible testing of many chemicals on a routine basis. There are certain limitations of the technique, some of which concern the character of the compounds to be tested: Volatile chemicals tend to evaporate under the conditions of the test, thus the ID,, value may be variable, especially when the toxicity of the compound is fairly low. This has been overcome to some extent by adapting the procedure for use in 96- rather than 24-well plates (1,2) since the smaller surface area of the well in these dishes reduces the extent of evaporation. Mineral oil can also be used on the medium to reduce evaporation (3). Other chemicals that are difficult to test include those that are unstable or explosive in water. Insoluble substances are also unsuitable for testing, although the author has adapted the method for use with some compounds, using mineral oil as the solvent. Other difficulties are related to the nature of the cell line, i.e., rapidly growing, nondifferentiating cells of very low metabolic activity, hence raising problems of direct extrapolation of results to the in vivo situation. The system is likely to underestimate the toxicity of chemicals that require metabolic activation to a toxic intermediary or product. Substances that specifically attack dividing cells may appear to be of a much higher order of toxicity than they would be in vivo. The toxicity of substances that bind to serum proteins (i.e., such as those found in newborn calf serum) may be also underestimated. 1.1. 24 us 72 h Exposure

Period

The procedure may be adapted to enable determination of cytotoxicity of chemicals after an exposure period of either 24 or 72 h. The authors would stress, however, that they believe the longer exposure period should be used routinely (4). 1.2. Kenacid

Blue

R Dye Binding

Assay

One of the drawbacks of this assay is that the Kenacid Blue dye may, on occasion, precipitate out. The likelihood of this occurring increases as the length of handling time increases, therefore 96-well plates should be agitated regularly and inspected visually for uneven blue color. The process is, however, readily reversed by agitation, so any odd reading should be retested after trituration to obviate the possibility of precipitation.

The FRAME

Cytotoxicity

Test

111

Another problem that may occur is the deposition of a ring of dried protein around the walls of the well, at the air/medium interface. This arises through excessive evaporation or if the culture medium is not properly removed. Such precipitated protein will give an inaccurate assessment of total cellular protein. Total protein measurement does not make allowances for necrotic cells that may still be attached to the culture dish and, therefore, may underestimate the toxicity of a compound. It should be noted, however, that the occurrence of adhering dead or dying cells is very rare. Advantages of this system include: 1, It can be repeatedmore than once on the samecells. 2. Cells can be fixed and the staining performed later. 3. The cell distribution can easily be seenwith the naked eye when stained with the Kenacid Blue before desorbing,thus giving a rapid indication of the successof the assay. Despite the limitations of the system, it provides a simple screen for the rapid assessment of the toxicity of compounds. The current validation of the system is producing results for a wide range of compounds, 150 of which are published (5). It would appear that the correlation to the in vivo situation (especially when problems concerning metabolic activation, and so on, are taken into account) is very good. When the in vitro cytotoxicities, i.e., IDS0 values, of 59 chemicals were compared with rat oral and mouse intraperitoneal LD,, values, correlation coefficients of 0.76 and 0.80, respectively, were obtained (6). The advantages and disadvantages of the Kenacid Blue protein assay are presented above. Under certain conditions a direct comparison of the Kenacid Blue and the Neutral Red methods may be of value. For example, certain chemicals, such as Chloroquine sulfate, and other antimalarial agents that target lysosomes may give different results (4). In such cases, performing both methods will yield extra information. The cytotoxicity test system has undergone a period of in-house development and an investigation into its interlaboratory validation has been performed in a collaborative study involving four different centers. The results of this study are presented in the paper by Knox et al. (I). The authors note that the initial cytotoxicity test employed a human embryonic cell line, BCL-Dl, but now uses a mouse embryonic cell line, 3T3-Ll . At present a number of research groups are evaluating the cur-

112

Clothier

rent procedure and comparing its performance to a variety of other test systems. The test is being used in cooperative schemes to compare different results, including those run by the European Commission, and the Multicentre Evaluation of In Vitro Cytotoxicity (MEIC) scheme organized by the Scandinavian Society of Cell Toxicology. These authors have now tested 225 pure chemicals and 100 formula-

tions in the system. The experimental data for 150, together with in vivo comparisons for 59 of these chemicals, is published (3-5). The cytotoxicity test system will be presented in two sections. The first outlines the maintenance of the cell-line and culture procedures, and describes the exposure of the cells to test chemicals. The second section presents the methodologies for the Kenacid Blue assay.

2. Materials of Cell Cultures to Test Compounds Cell line: 3T3-Ll cells, obtainable from the American Tissue Culture Collection (ATCC), code CCL92.1 or the European Collection of Animal Cell Cultures (ECACC) (Porton Down, UK). Incubator: 37”C, humidified, 5% C02/95% an. Tissue culture flasks, 80 cm. 96-Well tissue culture plates. Phosphate buffered saline (PBS), calcium and magnesium free. Prepared from Dulbecco’s formulation. PBS tablets, Trypsin-EDTA: 0.05% w/v trypsin, 0.02% w/v EDTA. Dulbecco’s supplemented medium; composition per liter: 730 mL stenle double distilled water, 100 mL Dulbecco’s minimum essential medmm (x10 concentrate), 100 mL newborn calf serum, heat inactivated. 50 mL Sodium bicarbonate (7.5%) 10 mL L-glutamme (200 mM), 100 mg. Streptomycin sulfate, 100,000 IU benzylpenicillin, UK, 2 mg fungizone. Solvents: Ethanol, methanol, dimethyl sulfoxide (DMSO), and, if appropriate, mmeral oil. All solutions, glassware, and so on, are sterile and all procedures are carried out under aseptic conditions and in the sterile environment of a laminar flow cabinet (biological hazard standard). 2.1.

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

Exposure

2.2.

Determination

of Cell

Growth

Inhibition

1. Rotatest shaker. 2. Micro ELISA Kontron SLT 210 plate reader: readings taken at 577 nm, reference blank at 404 nm, or Anthos 2001 plate reader: readings taken at 570 nm, reference blank at 404 nm. Note: Filters of 577 and 600 nm may also be used.

The FRAME

Cytotoxicity

Test

113

3. Phosphate buffered saline (PBS), pH 7.4. 4. Fixative: 1% glacial acetic acid, 50% ethanol, 49% distilled water. 5. Kenacid Blue stock solution: 0.4 g Coomassie Brilliant Blue R-250 stain, 250 mL ethanol, 630 mL distilled water. 6. Kenacid Blue stain: 12 mL glacial acetic actd, 88 mL Kenactd Blue stock solution. Prepared immediately prior to use. 7. Washing solution: 10% ethanol, 5% glacial acetic acid, 85% distilled water. 8. Desorbing solution: 98.15 g potassium acetate (lM), 300 mL distilled water, 700 mL ethanol.

3. Methods 3.1. Exposure of Cell Culture to Test Compound 3.1.1. Cell Maintenance and Culture Procedures 1. Stocks of 3T3-Ll cells can be stored in sterile, heat sealable ampules in liquid nitrogen or -8OOC freezer, following suspension in complete medmm containing 5% DMSO as a cryoprotective agent. The cell concentration used is approx 1 x lo6 cells/ml. Cells can be centrifuged for short periods, i.e., 5 min at low speeds (approx 1OOg). 2. When required, the cells should be thawed rapidly in a water bath at 37°C and suspended in sufficient medium in a tissue culture flask to ensure that the DMSO is diluted to yield a final concentration of ~0.1%. Once the cells have attached overnight, the medium should be replaced, i.e., approx 30 mL in a 80 cm tissue culture flask. 3. When the cells approach confluence they should be removed from the flask by trypsinization: Decant the medium and rinse the cultures with PBS at 37OC. Add 1 mL of Trypsin-EDTA. After 2-3 min give the flask 2 or 3 sharp taps to detach the cells into a single suspension (rocking the flask causes the cells to clump). When this occurs (judge by visual examination) add -10 mL of medium (37OC) to prevent enzymatic damage. Gentle trituration may be employed to ensure a single cell suspension is obtained. 4. Count a sample of the cell suspension (using a hemocytometer). Dilute the cell suspension with medium (37OC) to give a final concentration of 1.5-2 x lo4 cells/n& for the 72 h exposure period assay or 4-5 x lo4 cells/ml for the 24 h exposure period assay. 5. Dispense 150 pL of the diluted cell suspension into 95 of the 96 wells of each multiwell plate. Add 150 l.tL of cell-free medium only to the remaining well. Incubate the cells overnight to allow adherence and recovery from the exposure to trypsin.

114

Clothier 3.1.2. Range Finding

Experiment

1. Assess the solubility of each test compound in the following order: Highest possible in medium; 100 + mg/mL m methanol; 100 + mg/mL ethanol; 100 + mg/mL DMSO. 2. A preliminary mvestigatron should be performed for each chemical. Test eight different conditions: a. Control, medium only. b. Solvent control, medium containing 1% solvent (if necessary). c. Stx test chemical concentrations, e.g., 0.01, 0.1, 1.0, 10, 100, 1000 l.tg/ mL or up to the limit of solubility in medium (keeping solvent at 1%). Solutions >lOO mg/mL, or if a substance takes up 10% of the final volume, should be made up volumetrically. Results from wells containing any chemical precipitate should be disregarded. Two chemicals are tested per plate. 3. Prepare 96-well plates as outlmed above. (One row of 3 wells on each plate contains cell-free control medium only.) The outer 36 wells are not used for test chemrcals. 4. After the overnight mcubation aspirate the medium from the cells. Add 150 pL of medium containing the appropriate concentration of test chemical. Incubate for either 24 or 72 h. 5. Visual examination should be performed after 24, 48, and 72 h and an value noted. This enables detection of time-related approximate IDsa changes in cytotoxicity. 6. After the fmal visual examination remove the medium and determine protein content by the Kenacid Blue method, see Section 3.1. The results of this preliminary study enable the concentration range over which O-100% inhibition occurs to be identified. 7. If no toxicrty is found at 1000 pg/mL, another range finding expertment is carried out with a top concentration of 100,000 pg/mL or maximal solubility is obtained, i.e., the concentration at which the chemical begins to precipitate out of solution. 8. At least six concentrations spanmng this range can then be selected for an accurate determination of cytotoxicity.

3.1.3. Accurate Determination

of IDzO, IDsO, and IDS0 Values

1. Each chemical should be tested, which may be performed blind, on three separate occasions (allows for variation in weighing of chemical and plating of cells) and each of the six concentrations of a chemical and controls should be tested in triplicate. It is important that the same range of concentrations be tested on three separate occasions.

The FRAME

Cytotoxicity

115

Test

2. The following conditions should be set up: a. Nonsolvent control, medium only. b. Solvent control, medium + 1% solvent (where necessary). c. At least six compound concentrations. d. Positive controls should also be set up to ensure that the cells behave as expected. It is suggested that 70 p,g/mL dinitrophenol or Tween 20 at 250 pg/mL be employed in this respect. 3. Prepare 96-well plates as outlined above. (All corner wells on each plate contam cell-free control medium only.) After the overnight mcubation aspirate the medium from the cells. Add 150 j-tL of appropriate drug-containing or control medium. It is recommended that drug treatment of the outer wells be avoided. Incubate for either 24 or 72 h. An indrcation of the number of cells present at the initiation of the experiment should be obtained by removing, but not replacing, the medium from three of the wells at the treatment stage. They will then air dry and the protein content can be determined at the end of the incubation period. After 24 or 72 h exposure the Kenacid Blue method may be employed to give an mdication of cell number. 3.2. Determination of Cell Growth Inhibition 3.2.1. The Frame Kenacid Blue R Dye Binding Method (Nottingham Modification)

This method is basedon that of ref. I. The measurementof total cell protein provides a quantitative indication of cell number present in a culture. 3.2.1.1, MODIFICATION

OF ORIGINAL

METHOD

This method of protein determination has been adapted for use in 96-, rather than 24-multiwell plates, hence, certain parameters are different from those given in previously published papers (e.g., 8). Other modifications of the original Kenacid Blue method include: 1. Glutaraldehyde is no longer used as the frxattve because it has a heavy vapor, is not a rapid fixative, and is a penetrative, accumulative toxin. 2. A Kontron or Anthos 2007 plate reader 1snow used, which has a 577 or 570 nm filter read agamst a 404 nm filter (reference readmg). 3.2.2. Method 1. After 24 or 72 h exposure to the chemical, discard medium from cells, Rinse cultures twice wrth warm PBS (-37”C), to remove any remaining medium. Aliquot approx 150 pL of fixative to each well. Shake multiwell plates gently for 20 mm on a Rotatest Shaker. Remove fixative. If the

116

2.

3. 4.

5.

6. 7.

Clothier Kenacid Blue assay is to be performed on cells that have been tested with the Neutral Red method, the destain is removed and the Kenacld Blue assay 1scontinued as below. Prepare Kenacld Blue stain immediately prior to use: Add 12 mL glacial acetic acid to every 88 mL of stock Kenacid Blue R solution. Add 150 p.L of Kenacid Blue stain per well. Shake multiwell plates for a further 2-3 h on the Rotatest shaker. Remove stain. Fill each well with washing solution to remove excess stain. Repeat. Leave for 1 min. Replace this washing solution with fresh washing solution. Shake plates for a further 20 min. Remove washing solution and replace with precisely 150 PL desorbing solution. Rapidly agitate the plates for 20 min until the dye has completely gone mto solution and gives a homogeneous colored solution. Read the absorbance of each well directly at 577 nm using the plate reader with 404 nm reference blank, against the reference well contaming no cells. Set this well at zero. (Note: 570 or 600 nm measurement filters can be used.) The absorbance correlates linearly with cell number over specific optical density ranges 0.6-l .8 at 570 nm without subtraction of a reference filter using 24-well plates. When 96-well plates are employed and the optical density is 570 nm minus the 404 nm reference, the readings should be m the range of 0.4-1.2. The increase in optical density value between the original protein level and the value 72 h later should be 1.8x or more. If either of these conditions are not satisfied, the results should be discarded. The positive control should give a total protein value of 50% of the medium control. (Thirty to seventy percent is acceptable. Outside this range the results should be discarded.) The standard error of the mean for triplicate wells should not be more than 20% of the mean for results to be valid.

3.2.3. Calculation of Results 1. An estimation of the total cell protein (measured using the Kenacid Blue method) is made on each culture dish as outlined above. The results obtained under test conditions are compared to the appropriate control and converted to a percentage value. The six concentrations of each compound tested should span the range of no effect up to 95-100% inhibition of cell growth. 2. The mean result of trlphcates (i.e., 3 wells exposed to the same concentration) are plotted on a graph (usually sigmoidal or exponential in shape) as concentration vs percent inhibition. Precise ID,,, ID,,,, and ID,, values can be calculated from the curve. The concentration of test compound producmg just less than, and that producing Just over, the required level of inhlbi-

The FRAME

Cytotoxicity

Test

tion are joined on the graph by a straight line. The appropriate ID value is then calculated or read directly from the graph at the point the line intersectsthe 20, 50, or 80% inhibition mark. 3. The ID values from three separate runs are averaged to give final concentrations, usually expressed as pg/mL. Chemicals are normally ranked for toxicity using the ID,, value, since this is the section of the curve most likely to be linear. Ranking in terms of IDso mM values is preferred, since this reflects relative toxicity in terms of the same number of molecules to which the cells are exposed.

When linear regression analysis was used to compare in vitro (log ID,,) values and rat and mouse in vivo (log LDSO)acute toxicities, the following correlation coefficients (r values) were obtained: Rat oral/in vitro Mouse ip/in vitro Most toxic (rat or mouse)/in vitro Least toxic (rat or mouse)/in vitro

r r r r

= 0.76 = 0.80 = 0.81 = 0.78

Data on the comparison of the in vitro cytotoxicities and acute in vivo toxicities of 59 chemicals were taken from ref. 6. 4. Notes 1. The concentration m solvent should always be loo-fold greater than that required in the medium. This enables the concentration of solvent present in the medium to be maintained at a constant level, i.e., 1%. It should be noted that 1% ethanol or methanol exert little effect on the cells in culture but that 1% DMSO may cause a reduction in total protein to between 515% compared with solvent-free cultures. (If a larger decrease occurs in response to the solvent alone, the results of that plate should be discarded.) 2. All test chemical solutions should be made up immediately prior to addition to avoid problems of stability and to avoid problems of precipitation of medium proteins. 3. The authors note that they have detected a decrease in cell growth in the wells on the outer perimeter of the plate, therefore they recommend the use of the inner wells only. (That is only two chemicals per plate.) 4. Cells are plated out m 95 of the wells, whereas 1 well is left without cells, but with medium, to act as a reference. This well is set at zero when reading the absorbance. 5. The cells do not stain well if the acetic acid is omitted from the Kenacid blue stain.

118

Clothier

6. If any dark blue rings of Kenacid Blue stain are left on the sides of the wells at the orrgmal atr/medmm interface, carefully wipe these away between washes. This can be drffrcult rf not impossible in 96-well plates so if it occurs the results should be discarded.

Acknowledgments This is based on the INVITTOX protocol 3B, ISSN 0960-2194, INVITTOX, 34 Stoney St., Nottingham, NGl lNB, UK. References 1 Knox, P., Uphill, P. F , Fry, J. R , Benford, D. J., and Balls, M. (1986) The FRAME Multicentre ProJect on in vitro Cytotoxicology. Fed. Chem. Tox. 24,457463. 2 Riddell, R. J., Clothier, R. H , and Balls, M. (1986) An evaluation of three m vitro cytotoxicity assays. Fed. Chem. Tox~ol. 24,469-471 3. Smith, L. M., Clothier, R. H., Hilligde, S., and Balls, M. (1992) Modification of the FRAME Kenacid Blue method for cytotoxmity tests on volatile materials, ATLA 20,230-234. 4 Riddell, R J , Panacer, D S , Wilde, S M., Clothier, R H , and Balls, M (1986) The importance of exposure period and cell type in in vitro cytotoxicity tests ATLA 14,86-92. 5 Clothier, R. H , Hulme, L , Ahmed, A. B., Reeves, H L , Smith, M., and Balls, M. (1988) In vitro cytotoxicity of 150 chemicals to 3T3-Ll cells, assessed by the FRAME Kenacid Blue Method. ATLA 16,84-95. 6. Clothier, R. H., Hulme, L., Smith, M., and Balls, M. (1987) A comparison of the m vitro cytotoxicities and acute in vivo toxicities of 59 chemicals. Mol. Toxic01 1, 571-577. 7. Hulme, L., Reeves, H. L., Clothier, R. H., Smith, M., and Balls, M (1987) An assessment of two alternative methods for predicting the in vivo toxicities of Metallic compounds. Mol. Toxicol. 1,589-596. 8 Balls, M., Riddell, R. J , Horner, S. A, and Clothter, R H (1987) The FRAME approach to the development, validation, and evaluation of in vitro alternative methods, in In Vitro Methods tn Tox~ology - Approaches to Validation (Goldberg, A. M., ed.), Mary Ann Liebert, New York, pp 45-58

CHAPTER14 Allium

Test

Geirid Fiskev% 1. Introduction The Allium test provides a rapid screening procedure for chemicals, pollutants, contaminants, and so on that may represent environmental hazards. Root growth inhibition and adverse effects on chromosomes provide an indication of likely toxicity. The root tip is often the first part of any plant that is likely to come into contact with chemicals and pollutants found in soil and water supplies. Observation of the root tip system of the onion, AZ&m cepa, has shown that this plant is particularly sensitive to the harmful effects of such environmental contaminants. Gross effects can be quantified by measurement of inhibition of growth of the newly developing root system, whereas examination of the chromosomes of the individual cells of the root tip can indicate likely mutagenic effects. Twelve onions (A&m cepa) are prepared by removal of the outer scales and brownish bottom plate and put into test tubes filled with test liquids for 4 d, the liquid being changed every day. A further series of twelve onions are similarly prepared and maintained in pure water to provide a control population. The ten onions that appear to be developing the best in each series are selected for examination. On d 2 one or two root tips from each of 5 onions is prepared for microscopic examination, One hundred mitoses are scored from each of the 5 slides, as is the mitotic index (MI) for 400 cells. On d 4 the root length of each bulb is measured and the series photographed. (A recovery experiment can be performed by changing the medium for 5 of the 10 onions of each test From Methods m Molecular B/ology, EcMed by S O’Hare and C K Atterwlll

Vol 43 In Vitro Toxmty Testmg Protocols Copyright Humana Press Inc , Totowa, NJ

119

120

Fiskesjti

series to control water after measurement on d 4, replenishing the liquid on d 5 and, finally, measuring the root length and photographing the series on d 6). Toxicity is measured by both macroscopic parameters (e.g., growth inhibition), where the degree of damage is used to assess the toxic status of the chemical tested and microscopic parameters, where the rate of chromosome breakage and damage may be used to predict mutagenesis. Plants are easy to store and handle and are plentiful and inexpensive. In general, the chromosome condition of plant cells is good, thus providing a high standard in control conditions. The Allium test is a relatively rapid, easy test. It is also both highly sensitive and reproducible. It also provides comparable results to a number of other test systems. Both macroscopic and microscopic effects may be observed, and there appearsto be a good correlation between the two. The macroscopic effect (inhibition of root growth) appears to be the most sensitive parameter. This is to be expected since any deleteriouseffect, direct or indirect, is likely to result in inhibition of growth. Microscopic examination allows assessment of chromosome damage and cell division disturbances, thus providing additional information as to the severity or mechanism of the toxic effect, or potential mutagenicity. The root cells possesscertain enzymes, the mixed function oxidases, that are instrumental in the activation of many promutagens to mutagens. This activating system will improve the detection of those chemicals that exert their toxic effect via a reactive metabolite. The system has a wide range of applications, e.g., for testing pure chemicals, drinking water, natural water, industrial waste, and so on, and is useful for evaluating and ranking environmental chemicals with reference to toxicity. The test can also be used to measure the relative toxicities of nonwater soluble compounds, provided they can be dissolved in a suitable solvent and then diluted in water so that the final concentration does not exceed certain limits. In cases such as these, solvent controls must also be incorporated into the test regime. The system operates over a wide pH range (3.5-11.0) without any obvious effects on the growth of the root systems. Thus, moderately acidic/alkali water samples, chemical solutions, and so on can be tested readily without pH correction being necessary. Although the pH itself may not affect the growth of the roots, it should be taken into consideration when assessing the toxicity of compounds

Allium

Test

121

since, in many cases, the pH can dramatically alter the toxic potential of these, e.g., by changing the state of ionization. Disadvantages of the system concern problems associated with the state of compounds being tested. The influence of pH on compounds present in solution and the resultant change in toxic profiles has already been mentioned. Another problem concerns the presence of insoluble compounds in waterways, industrial effluent, and so on. It is very difficult to look at the biological effect of such complex mixtures in the Allium system since particulate matter may exert indirect harmful effects, such as the prevention of uptake of nutrients. It is therefore recommended that samples such as these are also chemically analyzed. The Allium test is highly sensitive and, as such, positive toxic effects may result for a number of compounds that would not necessarily be deemed harmful when tested in other systems (especially higher organisms, such as fish). Although this may occasionally result in false positives, it also ensures that contaminations will not be overlooked. This is especially important when complex mixtures are to be tested. A positive result in this test system should, therefore, be taken to indicate a potential biological hazard. False negatives, on the other hand, have been shown to rarely occur in either the Allium test or other similar plant tests (I), therefore, any compound tested giving a negative result can be reliably considered nonmutagenic. An extrapolation of results from one test system to another (and eventually to humans) should, however, be based on the results of a battery of tests and with due consideration to the metabolic pathways of the compound tested. The Allium test system was first used in 1938 (2) and 1944 (3) to examine the effect on chromosomes of colchicine. It has received much attention since that time (for review see ref. 4). Certain modifications of the basic test system have been introduced to enable environmental monitoring of various chemicals, and also of complex mixtures such as those present in river water, industrial waste, and so on (5). The major modifications include the use of a series of bulbs (i.e., 10) for each condition tested (thus permitting an EC5, determination), and the immediate exposure of bulbs to test solutions. The old test allowed an initial growth period in pure water until roots reached an appropriate length, (l-2 cm), after which they were exposed to test compounds,

122

Fiskesjii

1.1. Comparison to Other Short-Term Alternative Toxicity Test Systems This test has shown good agreement with results from other test systems, using many different organisms, eukaryotic as well as prokaryotic. Results of such comparisons are summarized below (5-8). 1.1.1. Chinese Hamster

Cell Line V79

In the absence of a metabolic activating system, V79 cells appear less sensitive than the Allium test in response to benzo(a)pyrene. The relative sensitivity is reversed when the cells are incubated in the presence of mixed function oxidases. Despite changes in sensitivities, the overall results between the two systems are comparable. 1.1.2. Human

Lymphocytes

Human lymphocytes seem slightly more sensitive than Allium root tip cells to the effects of organic mercury compound, although the overall ranking of chemicals by toxicity is similar. It should also be noted that when studied microscopically both cell types respond in a similar way (c-mitosis). 1.1.3. Autotrophic

Algae, Heterotrophic Microorganisms, and Activated Sludge

A number of chemicals have been tested in the Allium test and the results compared with those found using 16 different plankton algae (green algae and silicious algae), yeast (Succharomyces cerevisiae), protozoa (Tetruhymena pyriformis), and activated sludge (composed of bacteria, yeast, and protozoa). The tests were all comparable when the rank ordering of chemicals according to toxicity was examined, although differences in sensitivities were apparent. The majority of the algae were more sensitive than the Allium test, whereas the yeast, protozoa, and activated sludge were less sensitive. 1.1.4. Aquatic Animals and Plants A number of aquatic animals (Duphnia magna, Brachydario

rerio-

eggs or spawn, and Microtox test bacteria) and plants (lens and unicellular algae) appear to be less sensitive to certain classes of compounds compared with the Allium test, e.g., fish (Gusterosteus aculeatus). The Allium test, in this case, is probably a better test for environmental screening because of its higher sensitivity. Other animals (e.g., the crus-

Allium

Test

123

tacean Nitocru spinz’pes)and plants (e.g., lens and unicellular algae) give comparable results to the Allium test. A wide range of metals, industrial pollutants, compounds, and so on have been tested. Water from a variety of sources has also been examined. For examples, see refs. 9-15. 2. Materials 1, Onions: Equal size bulbs (l-5-2.0 cm diameter) of the speciesAZZiumcepa. After harvest, the bulbs should be stored in dry and well aerated conditions for a few months at lo-20°C before use (winter rest). Discard any onions that are dry, moldy, or have started shooting green leaves. 2. Test tubes: 1.5 cm diameter, 10 cm length. 3. Test tube racks. 4. Microscope, camera, slides, coverslips, and coverslip cement (to preserve the slides). 5. Materials: Fixation and maceration/hydrolysis: 9 parts 45% HAc and 1 part 1N HCl; 2% orcein in 45% HAc. a. Nutrient solutron for plant growth: Nutrient Stock Final concentratron Ca(N0,)2 . 4H20 KNOs MgS04. 7H20 KH2P04 Fe-EDTA e3H20 b. Trace elements MnS04 CuCl, NazMo04 ZnSO, H3BO3

l.Omh4

0.1 mM

2.0 mM

0.2 n-&l

l.Omh4 1.0 mlI4

0.1 mM 0.1 r&i

0.2 mM

0.02 mM

3.64pM 0.48 /.tM 0.0078 pA4 0.0042 piI4 3.7 ~.LM

0.364 PM 0.048 yM 0.00078pM 0.00042pM 0.37 pM

A stock solution (as above) should be prepared at tenfold the required concentration. Before use the solution should be diluted with distilled water. (Colchicine may be used for a specific purpose; see Section 4.2.) 6. Test samples: Control-Normal fresh tap water or prepare the nutrient solution for plant growth (see above). The water should be relatively hard,

e.g., Ca + Mg = 50-70 mg/L, with relatively neutral pH. No toxic ions should be present, e.g., copper ions from copper pipes (Cu2+ should not exceed 0.05 mg/L). Do not use distilled or deionized water alone. Insoluble compounds may be dissolved in certain solvents (i.e., ethanol, methanol, or acetone) and diluted to the required concentration in pure water. Final

Fiskesjti solvent concentration must not exceed 1% v/v. Chemicals should be tested initially over a concentration range of 1P2-lOAM, although in certain casesit may also be necessary to test higher or lower concentrations. Test materials may be stored under dry and well aerated conditions at lO-20°C until bulbs from the next harvest become available. All test/control solutions should be brought to room temperature before commencement of the experiment.

3. Methods 3.1. Preparation 0fAllium cepa Remove the outer scales and brownish bottom plate of the bulbs, leaving the root primordia

intact. If a large number of bulbs are being pre-

pared, place the peeled bulbs into fresh pure water to protect the primordia from drying. A number of bulbs in any population will be natu-

rally slow or poor growing (should not exceed 20%). To allow for this, a series of 12 bulbs for each control and test condition should be prepared, the 10 best bulbs being chosen at the appropriate time (d 1 or 2) for examination and continued observation. 3.2. Exposure

to Test Solutions

Remove bulbs from water and place on a soft layer of paper to remove place in test tubes filled with test liquids or control solutions and incubate for 2-4 (or up to 6) d, changing the test

excess water. Immediately

liquid every day. All tests should be performed at a relatively

constant

room temperature of about +2O”C and protected against direct sunlight. Run one control per each set of experiments and, where the chemical

is dissolved in a solvent, a second control series where pure water is supplemented with the relevant concentration of solvent. 3.3. Day 2

On d 2, prepare slides for microscopic examination. One or two root tips is removed from 5 bulbs out of each series of ten. These are then

used to prepare 5 slides in accordance with the standard procedure for orcein staining of squashed material (the method of choice because of its proven high and rapid performance).

Fixation and maceration/hydroly-

sis is a mixture of 9 parts 45% HAc and 1 part 1NHCl at 50°C for 5 min. Squash the material in 2% orcein in 45% HAc on slide and place coverslip over. Place a piece of blotting paper over the coverslip and press down to fix into position. With the other hand, press two fingers from

Allium

125

Test

left to right over the coverslip to remove superfluous stain. This also helps to arrange the chromosomes in one plane. If coverslip cement is used, the slides may be stored in a refrigerator for 2 mo or more. Examine under a microscope. Score 100 mitoses for each slide; 5 slides for each series. Score the mitotic index for 400 cells on each of 5 slides. 3.4. Microscopic Parameters 1. Mitotic index: the number of dividing cells per 1000 observed cells. 2. Characterization of mitosis: 100 mitoses/slide. Normal metaphase and anaphase. it: Early normal anaphase: A higher number compared with the control indicates lower rate of cell division. Normal metaphase and anaphase. :: Observation of stickiness: This occurrence indicates a highly toxic, usually irreversible effect, probably leading to cell death. e and f. Clastogenic effects (chromosome fragments or bridges): This occurs as a result of chromosome- and chromatid-breaks, and is an indication of mutagenicity. g and h. Vagrant chromosomes (weak c-mitotic effect): This indicates a risk of aneuploidy. i and j. C-mitosis: A relatively weak toxic effect that may be reversible if the occurrence rate is low. It indicates a risk of aneuploidy. k. Other lessfrequently seen aberrations: Multipolar anaphases(weak c-mitotic effect); C-mitosis “with spindle;” “Banded” chromosome (6). 3.5. Day 3 or 4 1. On d 3 or 4, the root length and root appearanceshould be noted. Control roots will usually have reached a length of -5 cm. Root length can be measured in two ways: a. The most accurate method is to measure every root from each bulb requiring the removal of the roots and the termination of the experiment. Those roots that are mechanically bent or damaged are discarded. b. The length of the whole root bundle is measured outside the test tube by a ruler giving one value for each bulb. The second method is preferred because it allows the continuation of the experiment and does not appear to cause any significant reduction in

sensitivity or accuracyof the test (5). 2. Photograph the test series. 3.6. Macroscopic

Parameters

The most important parameter to measure is root length.

126

Fiskesjii 3.7. Other

Parameters

Change of color: The root tip as well as the whole plant may change color either with a certain salt (e.g., blue-green from copper sulfate), or the root tips may turn brownish because of toxic effects causing cell death. 3.8. Root Form

1. Normal roots. 2. “Crochet hooks”: Bending of the roots or root tips that may occur especially after treatment with certain metal salts. 3. c-Tumor formation: This may be apparentafter 3-5 d of cultivation after various types of treatments,but is more obvious after a longer period of time and is observedas a swelling of the root tips 4. Broken root tips. 4. Modifications of Basic Procedure 4.1. Recovery Study

On d 4 (after measurement of root length) change the test liquid to control water in 5 of the 10 test tubes. Change the liquid on d 5. Measure root lengths and/or photograph the series on d 6. If the first 4-d treatment was not irreversible, the five onions in pure water will have new roots or roots longer than those in test liquids. 4.2. Colchicine

Treatment

This is performed specifically to study chromosome breaks in more detail; these are sometimes more evident in c-mitotic chromosomes. Add 0.1% colchicine (to one or two extra onions in those test series that are to be studied) l-2 h prior to the preparation of slides. In those cases where environmental monitoring is being performed, colchicine should not be used, because it may mask possible c-mitotic effects induced by the test chemicals themselves. 5. Results

Calculate the mean root length for each onion as a percentage of the control. Plot treatment concentration against root length as a percentage of the control. Calculate the EClO, ECSO,and ECgOfrom the curve (representing the Effect Concentration) causing 10,50, and 90% growth restriction in relation to the control.

Allium

Test

127

Some sample ECsO values are presented below: Methyl mercury chloride-g.0 x 1c7M. Mercury chloride-3.3 x 1O”M. Copper sulfate-2.7 x lOA. Nickel chloride-l .7 x 10m5A4. Cadmium chloride-3.1 x 10e5M. Beryllium sulfate4.8 x ltiiI4. Aluminum chloride-8 .Ox 1Od4M. Manganese chloride-5.2 x 10m3M. Lithium chloride-2.0 x 10M2M. References 1. Ennever, F K , Andreano, G , and Rosenkranz, H S (1988) The ability of plant genotoxrcity assays to predict carcinogenicity Mutat Res 205,99-105 2. Levan, A. (1938) The effect of colchicine on root mitoses m Alhum. Heredztas 24, 471-486. 3. Ostergren, G (1944) Colchrcme mitosis, chromosome contraction, narcosis and protein cham folding Heredztas 30,429-467. 4. Grant, W. F. (1982) Chromosome aberration assays m Allium. Mutat Res. 99, 273-29 1. 5. Fiskesjo, G (1985) The Allium as a standard m environmental monitoring. Heredztas 102,99-102. 6. FiskesJG, G. (1981) Benzo(a)pyrene and N-methyl-N-nitro-N-nitrosguanidine in the Allium test. Heredztas 95, 155-162. 7. Fiskesjo, G. (1988) The Allium test-an alternative m environmental studies, the relative toxicity of metal ions Mutat. Res 197(2), 243-260 8. Fiskesjo, G. (1993) The Allium test-a potential standard of assessment of environmental toxrcity Environmental Toxicology and Risk Assessment: 2nd Vol., ASTM STP 1216. (Gorsuch, J. W., Dwyer, F. J., Ingersoll, C G., and La Pomt, T W , eds.), American Society for Testing and Materials, Philadelphia, PA. 9 Berggren, D. and Fiskesjo, G (1987) Aluminum toxicity and speciation m soil liquids-experiments with Allzum cepa L. Environ. Tox. Chem. 6,771-779. 10 Fiskesjo, G. (198 1) Allium test on copper in drmkmg water. Vatten 17(3), 232-240 11. Fiskesjo, G., Lassen, C , and Renberg, L. (1981) Chlorophenoxyacetic acids and chlorophenols m the modified Allium test. Chem.-Biol. Znterac. 34,333-344. 12. Fiskesjd, G. (1982) Evaluation of short-term tests for toxicity and mutagenicity with special reference to mercury and selenium. PhD. thesis, Institute of Genetics, University of Lund, Sweden 13 Fiskesjo, G. (1983) Nucleolar dissolution induced by aluminium in root cells of Allium. Physiologzca plantarum 59, 508-5 11 14. Fiskesjo, G. (1985) Allmm test on river water from BraUn and SaxUn before and after closure of a chemical factory Ambio 14(2), 99-103 15 Fiske@, G. (1987) The Alhum test-an alternative ATLA 15,33-35

CHAPTER 15

Measurement of Cell Membrane Toxicity by Means of Z-Deoxy-D-Glucose Erik

Walum and Anna For&y

1. Introduction The membrane theory of toxicity (I) emanates from the fact that the plasma membrane is the first barrier met by a toxic agent on reaching the cell, and was first advanced to explain the toxic effects of heavy metals. The vital regulatory mechanisms inherent in the plasma membrane and its chemical composition make it susceptible to many toxic compounds (2). A toxic substance may cause structural alterations in the membrane by binding to or crosslinking proteins or by dissolving in the lipid matrix and thereby disturbing its order. The formation of functional holes in the membrane through the action of chemicals with detergent properties is common. Structural alterations often lead to specific functional changes, since the function of both integral and peripheral proteins are greatly influenced by their lipid environment. Furthermore, chemicals may induce toxic effects by direct interaction with functional proteins. Membrane leakage tests are frequently used to screen for cytotoxic effects in cultured cells. In one way or another these tests are designed to measure an increased permeability of the plasma membrane, and they are based on the assumption that leakage of intra- or extracellular substances indicates damage to the membrane. Furthermore, the molecular size of the leaked material is often regarded as indicative of the character of membrane damage, in terms of size of the lesions induced by the test compound (3). From. Methods m Molecular B/ology, Edited by- S. O’Hare and C K Attenwll

Vol 43’ In Vitro Toxicity Testing Protocols Copyright Humana Press Inc , Totowa, NJ

129

Walum and Forsby

Tritiated 2-deoxy-o-glucose has proved suitable as a tool for determining general cell membrane permeability changes in monolayer cultures (4). As an analog of o-glucose, 2-deoxy-n-glucose is readily taken up into cells via the glucose transport system and it is then phosphorylated by hexokinase. The product, 2-deoxy-D-glucose-&phosphate, accumulates in the cell for several reasons: Hexokinase is not feedbackinhibited, the product is metabolically inert, and the cell membrane permeability for this compound is very low. By terminating the uptake and then measuring the efflux of radioactivity in the presence of a test compound, it is possible to quantify membrane impairment. In a study of membrane lesions in cultured neuroblastoma cells exposed to metal compounds, a membrane toxic concentration (MTC) was defined as the concentration of the tested chemical, giving rise to an increase in the relative efflux from 1.O to 1.2 in 60 min (5). Changes in the kinetics of efflux of tritiated 2-deoxy-o-glucose are well correlated with alterations in the structure of the cell surface as observed in the scanning electron microscope (6). Furthermore, it is possible to detect more subtle changes in the physical properties of the cell membrane by the use of 2-deoxy-o-glucase. This was shown in an investigation, where the influence of benzene and phenol on the activation energies for the efflux were studied in a number of cell lines (7). In a more recent paper, the effect of six sesquiterpenes on neuroblastoma cell membrane permeability is described. Using a quantitative structure-activity relationship analysis, it was possible to show a very close correlation between the chemical characteristics of the terpenesand their ability to alter 2-deoxy-o-glucose-efflux (8). Below, two different procedures are given for the application of 2deoxy-o-glucose in membrane toxicity studies. The first procedure is essentially identical to that recorded by the in vitro toxicology data bank INVITTOX (9) and involves the continuous perfusion of monolayer cultures during determinations (perfusion procedure). The second procedure has recently been published (8) and requires no perfusion (static procedure). 2. Materials

1. 2. 3. 4. 5.

CO2incubator, 37°C. Incubator, 37*C. Water bath, 37OC. Electronic cell counter. Scintillation counter.

Measurement

of Cell Membrane

Toxicity

131

Fig. 1. Perfusion chamber consisting of a standard culture dish (A) and a perfusion block (B) with inlet (C) and outlet (D) channels. 6. Tissue culture dishes, diameter 60 mm, culture surface 22 cm2. 7. Scintillation vials. 8. Perfusion chamber consisting of the tissue culture dish and a cylindershaped polycarbonate perfusion block (Sweden pat. 8101564-6, Eur pat. 0073773, US pat. 4,530,907; Peterson and Walum, 1983; Walum et al., 1990, see Fig. 1). Viton rubber O-rings. 9. Luer connections. 10. Silicone rubber tubings. 11. Multichannel peristaltic pump. 12. Perfusion buffer reservoirs. 13. Fraction collector. 14. 2-deoxy-o-[l-3H]glucose, aqueous solution, 15a. Phosphate buffered saline (PBS 1): 8.00 g NaCl, 0.20 g KCI, 0.10 g CaC12 * 2H20, 0.10 g MgCl2 * 6H2O, 1.15 g Na2HP0, * 12H20, 0.20 g/L KH2P04, pH 7.0, 1 mL of water. b. PBS 1 containing 0.5 pC!i [3H]2-deoxy-n-glucose/mL. c. PBS 1 supplemented with n-glucose (1 mg/mL) and containing appropriate concentrations of test compound. 16. 1M NaOH. 17. 1MHCI. 18. Scintillation cocktail.

Walum and Forsby

232 3. Methods 3.1. Perfusion

Procedure

1. Four days before an experiment detach cells in stock cultures, pool, and determme cell number using an electronic cell counter. Dilute and plate out cells in 22 cm* tissue culture dishes at a density of 1 x lo6 cells/dish. Maintain cells in a growth-promoting medium at 37OC, in a humidified atmosphere with approprrate amounts of CO,. 2. Remove growth medium from cell cultures and wash twice with 5 mL PBS 1 (37°C). Add immediately thereafter 4 mL of PBS 1 containmg tritiated deoxyglucose to each culture dish. Incubate dishes at 37OC, in a hunndtfied atmosphere without extra COZ, for 2 h. 3. Remove the incubation solution and wash cultures 3 times with 5 mL aliquots of ice-cold PBS 1. Mount culture dishes on perfnsron block (see Fig. 1). 4. The perfusion chamber consists of two parts; the standard tissue culture dish to which the cells are attached, and the perfusion block. The latter part is a cylinder-shaped polycarbonate block that has an outer diameter equal to the inner diameter of the culture dish. The top of the perfusion block has a turned counter-sink determining the volume of the perfusion chamber, Channels, which serve as in- and outlets for the perfusion solution, are placed at the periphery of the counter-smk (see Fig. 1). 5. Connect silicone tubmgs to the m- and outlets of the perfusion blocks, via simple luer connections. Connect inlet tubings to reservoirs containing perfusion solutions (PBS 1 with glucose and toxic substance). Adapt inlet tubings to peristaltic pump and outlet tubings to fraction collector. 6. Insert an O-ring around the top of the perfusion blocks. Press the culture dishes, containing adherent preloaded cells, onto the top of the perfusion blocks. Submerge the perfusion chambers and perfusion solution reservoirs in a water bath at 37°C (see Fig. 2). Start perfusion and continue for 1 h, at a rate of 1 r&/mm. Collect fractions of the perfusate mto scintillation vials at 5-min intervals (vials no. 1-12). 7. After 1 h stop the perfusion and remove buffer from each culture dish. Place that portion of buffer in a scintillation vial (no. 13). Add 1 mL of 1M NaOH solution to each dish and leave at room temperature for 30 min. Transfer the solubilized cell suspension into scmtillation vial no. 13. Wash each dish with 1 mL 1M HCl and transfer the acid into vial no. 13. 8. Add scintillation cocktail to vials containing perfusate fractions or solubilized cells and determine radioactivity as disintegrations per minute (DPM) by liquid scintillation counting. 9. Efflux kinetic calculations (perfusion procedure):

Measurement

of Cell Membrane

Toxicity

133

Fig. 2. Perfusion apparatus consisting of buffer reservoir (A), peristaltic pump (B), perfusion chamber (C; cf., Fig. l), scintillation vial (D), and fraction collector (E). a. Time dependence: The radioactivity remaining in the cells (calculated by subtracting the cumulative radioactivity released by the cells from the total radioactivity incorporated) is divided by the radioactivity in the cells at the beginning of the experiment (the total radioactivity incorporated) and plotted logarithmically vs time according to Kotyk and Jancek (10). The graph obtained (4) is resolved in its components, and the least-square fit of the lines calculated. Pool sizes, halflives, and rate constants are calculated from the equations of the lines. b. Temperature dependence: Rate constants are plotted vs the reciprocal of the absolute temperature (7) and activation energies determined from the straight lines obtained according to Arrhenius (11). c. Concentration dependence: For determinations of concentrationeffect curves the relative efflux is calculated according to: p, = a-/b, where u is the efflux, a the amount of radioactivity released in 60 mm, and b the total amount of radioactivity incorporated in the cells. The ratios of lt in the presence and absence of the test substance are then plotted vs the concentration of substance. In this graph (5) the membrane toxic concentration (MTC) may be determined as the concentration of the tested chemical giving rise to an increase in p,-ratio from 1.0 to 1.2.

3.2. Static Procedure Materials are the same as those outlined in Section 3.1. However, no perifusion apparatus is involved. 1. The cultures are prepared and the preloadmg of cells 1scarried out as in the perfusion procedure.

Walum

and Forsby

2. When the incubation with trrtiated deoxyglucose IS terminated, 2 mL of 37°C PBS 1 contanung glucose and test substance are added to each culture dish and the cultures incubated at 37°C m a humidified atmosphere without extra COz, for 1 h. 3. One minute after addition of the test substance, 50 pL IS taken from each dish and put into scintillation vials (no. 1). This procedure is then repeated 5, 10, 20, and 30 mm after the start of the incubatron (vials no. 2-5). At 30 min the remammg PBS 1 (1.75 mL) IS collected from the dishes and put into vials (no. 6). 4. The cells are then dissolved in 1 mL 1M NaOH, at room temperature for 30 min. The cell fraction IS transferred to vial no. 7, and the dish rinsed with 1 mL 1M HCl, which is also transferred to vial no. 7. 5. Scmtrllatron cocktail IS added to each 7-set of vials and the radioactivity determined in a liquid scintillation counter as DPM. 6. Efflux kinetics calculations (static procedure): The relative remaining radioactivity m the cells and concentratron dependence in each dish after each time interval (y,) is calculated as: y, = (DPMtot - [(DPM, . v,j50) + (DPMI to DPM, _ r)])/DPMt,r DPMtot is the total radioactivity taken up by the cells, DPM, is the amount of radioactrvrty in the 50 pL fraction taken from the incubation solution at time X, and DPM,- 1is the radioactivity in the fraction taken before that at time x. The expression cell membrane impermeability denotes the resistance of the membrane to radroactrvrty penetration and is calculated as yXin the presence of test substance divided by yX m the absence of test substance and multiplied by 100 to obtain a value as percent of control. This value is then plotted vs concentration of test compound for each time interval used (8). From these graphs values for the concentratrons that give a 5 or 20% increase in permeability (EC5 and EC& can be obtained. References 1. Rothstein,A. (1959) Cell membraneas a site of action of heavy metals. Fed. Proc l&1026-1035.

2. Kinter, B and Pritchard,J. B. (1977) Altered permeabrhtyof cell membranes,rn Handbook ofPhysiology (Lee, D H K , ed ), 9,563-576. 3 Thelestam, M. and Mollby, R. (1976) Cytotoxic effects on the plasma membrane of human diploid fibroblasts-a comparative study of leakage tests Med. Biol. 54, 39-49. 4 Walum, E. and Peterson, A (1982) Tritlated 2-deoxy-o-glucose as a probe for cell membrane permeablhty studies. Anal Bzochem. 120,8-l 1. 5. Walum, E (1982) Membrane lesions in cultured mouse neuroblastoma cells exposed to metal compounds. Toxicology 2567-74

Measurement

of Cell Membrane

Toxicity

135

6. Walum, E. and Marchner, H. (1983) Effects of mercuric chloride on the membrane integrity of cultured cell hnes. Toxic01 Lett. l&89-95. 7. Walum, E (1982) Temperature dependence of membrane permeability in cultured cells exposed to benzene and phenol. Biochem. Biophys. Rex Comm. 108,948-952. 8. Forsby, A., Walum, E., and Sterner, 0. (1992) The effect of six pungent sesquiterpenes with anttfeedant activity on cell membrane permeability in human neuroblastoma SH-SYSY cells. Chem. Biol. Interact. 84,85-95. 9. Warren, M., Atkinson, K., and Steer, S. (1989) Introducing INVITTOX: the ERGATT/FRAME in vitro toxicology data bank. ATLA 16,332-343. 10. Kotyk, A. and JanaZcek, K. (1970) Cell Membrane Transport: Principles and Techniques, Plenum, New York, pp. 240-244. 11 Arrhenms, S. (1915) Quantitative Laws in Biological Chemistry, Bell, London. 12. Peterson, A. and Walum, E. (1987) Growth and morphology of neuronal cell lines cultured in perfusion. In Vitro 19, 875-880. 13 Walum, E. and Jenssen, D. (1990) Understanding Cell Toxicology: Principles and Practice, Ellis Horwood, Chichester, p. 203.

CHAPTER16

MTT Assays Rosa Supino 1. Introduction The MTT calorimetric assay determines the ability of viable cells to convert a soluble tetrazolium salt [3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide] (MTT) into an insoluble formazan precipitate. Tetrazolium salts accept electrons from oxidized substrates or appropriate enzymes, such as NADH and NADPH. In particular, MTT is reduced at the ubiquinone and cytochrome b and c sites of the mitochondrial electron transport system and is the result of succinate dehydrogenase activity. This reaction converts the yellow salts to blue-colored formazan crystals that can be dissolved in an organic solvent whose concentration can be spectrophotometrically determined (Fig. 1). Owing to the many advantagesof the assay,it is today considereda significant advance over traditional techniques. In fact, it is rapid, versatile, quantitative, and highly reproducible with a low intratest variation between datapoints (+15% SD); it is useful in a large-scale, antitumor drug-screening program (1-3). Moreover, the test can also be used for floating cells, such as leukemias and small cell lung carcinoma, and always allows sufficient time for cell replication, drug-induced cell death, and loss of enzymatic activity, which generatesthe formazan product from the M’IT substrate(4). The MTT assay has to be set up for each cell line. The optimal cell number seeding (since cells during the assay have to be actively metabolizing), the duration of the experiment, and the time of MTT incubation necessary for an evaluable final optical density need to be selected in preliminary experiments. From: Methods m Molecular Biology, Edlted by: S. O’Hare and C K At&twill

Vol 43: In V/fro ToxrCrty Testing Protocols Copynght Humana Press Inc , Totowa, NJ

137

138

Supino 160,

1.20. x ‘?: .!! z a.s

060.

0” 0.40

*

ooo200

, 300

400

500 Wavelength

I 600

I 700

(nm)

Fig. 1. Absorption spectraof MTT formazanreagent(25 ug/mL) in DMSO (A) and 0.04N HCVisopropanol (B). A number of key points are relevant for the choice of the MTT assay. It is valid for a number of cell lines derived from a broad spectrum of solid tumors by a variety of isolation techniques. Most cell lines tested (106/l 11) exhibited acceptable calorimetric profiles (control growth absorbance > 0.500 U). Measurements of cell growth by MTT reduction correlated well with indices of cellular protein and viable cell number. At specific culture conditions and appropriate assay parameters, it provides reproducible indices of drug sensitivity. It is very sensitive since 32 human cells/well give an optical density of 0.05 U (optical density developed by murine cells is lower than that of human cells) (Fig. 2). It is comparable with other in vitro methods of drug cytotoxicity evaluation (Fig. 3).

1. 2. 3. 4. 5.

2. Materials Cells: exponentially growing cell line with recognizable levels of mitochondrial activity. 96-well tissue culture plates,U-bottomed or flat bottomed. Scanning multiwell spectrophotometer. Automatic plate shaker. Incubater, 37OC,humidified, 5% CO2atmosphere.

139

MTT Assays

cell number

(~10~)

Fig. 2. Relationship between cell number and optical density in B16V melanoma cells (0) and N592 small cell lung carcinoma (A). 6. Centrifuge. 7. 50 mL tubes. Make up the following: 8. Drugs made up to solutions 10x more concentrated than required. 9. Hank’s salt solution, composition (in g/L): KC1 (0.4), KH2P04 (O-06), NaCl (8), NaHCO, (0.35), Na2HP04. 7H,O (0.09). 10. MTI’ solution: 5 mg/mL in Hank’s salt solution. Note: MIT solution may be stored at 4°C for about 1 wk. 11. Culture medium, dependent on the cell line. 12. Dimethylsulfoxide (DMSO). 13. RPMI-1640. 14. Fetal calf serum. 15. 0.45 p filters. 16. Agarose. 17. HCl. 18. Isopropanol. 19. Collagenase. 20. Hyaluronidase. 21. Ficoll-Hypaque. 22. Percoll.

Supino

-0

“\*

“\

sPo *+

0 A

0

1

2 Adrmmycin

3 (IN)

I 0

4

3

"

20

40

60 60 ctsplatln (UM)

100

120

: : : : : 1

5

20

25

30

Fig. 3. Clonogenic assay and MTT assay curves: effect of adriamycin, cisplatin, and vinblastine on V79 cells. 0 MTT assay;A Clonogenic assay. 3. Methods 3.1. MTT Method For the chemosensitivity test, exponentially growing cells were harvested, counted, and inoculated (at the appropriate concentrations in a vol of 100 pL) into 96-well microtiter plates; 8 replicates were prepared for each dose. U-bottom microplates were used for suspension-growing cells, whereas flat-bottom microplates were used for plastic-adherent cell cultures. Immediately or 24 h after cell seeding, 10 PL of different dilutions of drugs, prepared 10x more concentrated than requested, was added to each well. After different incubation times at 37°C in a humidified 5% CO, atmosphere, the MTT assay was performed. MTT (Sigma,

MTT Assays St. Louis, MO) was dissolved at a concentration of 5 mg/mL in Hank’s salt solution and filtered with a 0.45 l.~filter (in order to avoid MTT aggregates). Ten microliters of MTT solution was added to each well and also to the control wells without cells. In fact, additional controls consisted of media alone with no cells, with or without the various drugs. After 4-6 h of incubation, microtiter plates were centrifuged at 2000 rpm for 10 min; medium was then removed, and 100 PL of DMSO was added to each well. After thorough mixing with a mechanical plate mixer, absorbance of the wells was read in a scanning well microculture plate reader at test and reference wavelengths of 550 and 620 nm, respectively, that are approximately the peak and the lowest MTT wavelengths of absorption required to avoid quenching from growth medium, in particular phenol red. Absorbance values from all wells were corrected against these control absorbance levels, and the IDS0 was defined as the concentration of drug that produced 50% reduction of absorbance compared with untreated control cells. 1. Harvest, count, and inoculate, in 100 uL of complete medium, the appropriate number of cells. 2. Add 10 pL of drug solution 10x more concentratedthan requested. 3. Incubatefor different times in a humidified atmospherein 5% CO2at 37°C. 4. Add 10 FL of MTT solution (5 mg/mL). 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Incubate 4-6 h in 5% CO2 at 37°C. Centrifuge the plates at 2000 rpm for 10 min. Remove the medium. Add 100 pL of DMSO. Mechanically mix the plate until formazan crystals are dissolved. Read in a microculture plate reader at test and reference wavelengths of 550 and 660 nm, respectively. Calculate the mean of the optical density of different replicates of the same sample. Evaluate the percentage of each value vs the control. Plot in a semilog chart the percentage of optical density against drug concentration. Determine the ID5e from the dose-response curve. 3.2. HTCA-MTT

Method

Different authors have reported that the assay yields results very similar to the clonogenic tumor cell assay in Chinese Hamster Ovary, lung

142

Supino

cancer cells, and mouse melanoma cell lines and their derived pleiotropic drug-resistant counterparts (5). However, data obtained with one technique should be compared only with data obtained with the same technique and the same treatment schedule. The MTT assay is successfully used, since other valid tests such as HTCA (human tumor colony assay) often have several technical problems, such as low plating efficiency and longer assay time. However, the use of the MTT assay for drug sensitivity testing of tumor samples may result in a higher background, because contamination by normal cells may reduce tetrazolium dye as well. To reduce such contamination, an HTCA-MTT mixed test was thus set up. After treatment for 1 h with the drugs, samples were seeded (in 96well microtiter plates prefilled with a 50 PL underlayer of RPMI-1640 and 15% fetal calf serum [FBS] in 0.4% agarose) in 50 PL of RPMI1640 and 15% FBS in 0.25% agarose.Followmg incubation for l-7 d at 37°C in 5% COZ, 10 ltL of MTT (5 mg/mL) was added to each well, and the plates were incubated at 37°C for 4 h. Then, 100 ~JLof 0.04N HCI in isopropanol was added to each well, and the solution was mixed vigorously to solubilize the formazan product. Air bubbles were sucked out, and after 1 h at room temperature, the absorbance of the wells was measured in a 2-wavelength microplate photometer at test and reference wavelengths of 550 and 660 nm, respectively. 1. Fill 96-well microplates with 50 l&L of complete medium in 0.4% agarose. 2. Allow agarose to solidify. 3. Add 50 ~.LLof complete medium contammg the appropriate number of drug-treated cells in 0.25% agarose. 4. Incubate for l-7 d in a humidified atmosphere of 5% CO2 at 37°C. 5. Add 10 FL of MIT (5 mg/mL). 6. Incubate for 4-6 h at 37°C in 5% CO*. 7. Add 100 l.tL of 0.04/V HCI m isopropanol. 8. Mix vigorously. 9. Leave for 1 h at room temperature. 10. Read m a microplate photometer at test and reference wavelengths 550 and 660 nm, respectively. 11. Calculate the mean of the optical density of different replicates of the same sample. 12. Evaluate the percentage of each value vs the control.

MTT Assays

143

13. Plot in a semilog chart the percentageof optical density againstdrug concentration. 14. Determine the IDS0from the dose-responsecurve. The HTCA-MTT hybrid assay is successful, since the HTCA assay is time-consuming, complicated, costly, and applicable for a limited number of human solid tumors. However, the MTT assay results in a high background owing to contamination by normal cells (especially in solid tumors), which may reduce tetrazolium dye as well. Moreover, it has been reported that tumor tissues are more sensitive to various antitumor drugs than are adjacent normal mucosal tissues. In contrast, the HTCAMTT hybrid assay is much shorter than the HTCA assay (4 d vs 2-3 wk) and is also effective to measure cells with low colony-forming efficiency. Furthermore, normal cells (fibroblasts, lymphocytes, and endothelial cells) do not grow in the double layer of agarose. 3.3. Purification

of Tumor

Cells

Other authors have reported a technique for purification from normal cells of tumor cells using discontinuous Ficoll-Hypaque and Percoll gradients. Using this method they obtained a higher correlation between in vitro results and clinical response (6) (Fig. 4). Tumor cells were dispersed in complete medium containing collagenase (2 mg/mL, type V-S; Sigma, hyaluronidase (10 U/r& type IV-S; Sigma), and DNase-1 (0.4 mg/rnL; Sigma). After a 40-min incubation at 37OC, the cells were harvested, washed, and suspended in complete medium. In the case of ascites, cells after centrifugation at 400g for 5 min were suspended in complete medium. Cells were then centrifuged on Ficoll-Hypaque (specific gravity 1.077, Pharmacia, Uppsala, Sweden) gradients at 400g for 30 min in 50-mL tubes (400g at the bottom of tubes). Mononuclear and tumor cells at the interface were collected, washed, and suspended at 106/mL in complete medium. The cells were then layered on discontinuous gradients consisting of 10 mL of 100% and 15 mL of 75% Ficoll-Hypaque in 50-mL plastic tubes. After centrifugation at 400g for 30 min (400g at the bottom of tubes), a tumor cell-rich fraction was collected from the 75% interface. The tumor-cell-enriched suspension was then layered onto discontinuous gradients containing 4 mL each of 25, 15, and 10% Percoll (Pharmacia) in complete medium in 15-n& plastic tubes. Centrifugation was performed at 25g for 7 min (25g at the

144

Supino Tumor

twiues I

enzymatic

suspended

dIgestion

In medium .. 100% Rcoll-Hypaque

& 1 4009 30 mm Tumor

I

cell-rich

fraction

400g 30 mm Tumor

cell-rich

fraction

Percoll

Tumor

cells (purity

more than 90%)

Fig. 4. Technique for Ficoll-Hypaque tumor cell purification from normal cells. bottom of tubes), and tumor cells depleted of lymphoid cells were col-

lected from the bottom and the 25% interface, washed, and suspended in complete medium at a concentration of 1 x 106/nL. The cells thus prepared were mainly tumor cells, with ~10% contamination by nonmalignant cells, as judged by morphological examination using Papanicolaou staining or carcinoembryonic antigen staining for CEA-positive tumor cells. The cells were found to be more than 90-95% viable by the Trypan blue dye exclusion test. The mean yield of purified tumor cells was 2.3 x 106, and the tumor cell count at the beginning of preparation was 13 x lo6 (rate of yield = 17.7).

145

MTT Assays 3.0

T

A

A--lr---.-A/A /Al A--A--A-A--A-a

--a------v~

-----A

o~--o-o/---o

Hours

Fig. 5. Effect of MTT concentration and incubation time on MTT reduction optical den&y. 0 MTT 0.5 mA4and A MTT 1 mM on Ml9 melanoma cells; 0 MTT 0.5 mM and A MTT 1 mM on SNB56 cells.

4. Notes 1. Considerable changes in optical density are brought about by the presence of different volumes of reaction medium. Low and constant evaporation microwell plates therefore have to be used (7), and variations over a medium volume htgher than 20 pL are unacceptable. 2. Phenol red at 10 mg/mL does not change the optical density of formazan in DMSO. 3. The MTT reductron is dependent on the o-glucose concentration in the culture medium and is independent of pH (8). 4. MTT formazan production is dependent on the MTT concentration in the culture medium (9) (Fig. 5). 5. The kinetics of MTT formazan production and the degree of saturation vary in a cell-line-specific manner. 6. Drug cytotoxrcrty evaluation may be influenced by the length of exposure to MTT (9) (Fig. 6). 7. Since MTT is cleaved by active mitochondria, the assayis effective, but to a lesser extent, also in the absence of cell proliferation. 8. For any given tumor cell line, the optical density of the solubilized formazan product is directly proportional, over a wide range, to the number of cells per well.

146

Spin0

01

L 0

-+--y--+1

----+

.--

2

3

-

,-.

--

--A

4

5

Hours

Fig. 6. Dependence of adrtamycin ID,, from MIT-time and concentration exposure. 0 M’IT 0.5 mA4and A M’IT 1 mA4on Ml9 melanoma cells; 0 MIT 0.5 n-H and A MTT 1 mA4 on SNB56 cells.

01

3E-1

1

10 Adrlamyan

100 ( rig/ml

1000

)

Fig. 7. Dose-effect curves of adriamycin on N592 (0) and N592DX (0) cell lines. 9. The MTT assay is effective on parental cell lines and their pleiotropic drug-resistant counterpart, thus leading to a correlation between comparable cell lines (Fig. 7). 10. The assay is effective on many different clmmal anticancer drugs (I) (Table 1).

MTT Assays

147

Table 1 Chemosensitivity of Cells Based on ICscs @tg/mL) of Some Clinical Anticancer Drugs in the MTT Reduction Assay Cells B16FlO B16 B16Fl L929 Mouse spleen HPBMC LOVO K562 RPM17272 COL0320HSR L1210 Clone A WiDr COLO205 MCF7 COLO201 Cells

Dactinomycin

Doxorubicin

Ara-C

<0.0001 <0.0001 <0.0001 <0.0001
0.01 0.002 0.001 0.1 0.05 0.02 002 0.003 0 008 0 17 0 02 0.05 0 17 0.35 001 >l

0.006 0.005 0.0002 0 46 0.07 0.04 0.10 0.01 0.44 0.03 0.001 0.14 >lO >lO >lO >lO

Cycloheximide

B16FlO B16 B16Fl L929 Mouse spleen HPBMC LOVO K562 RPM17272 COL0320HSR L1210 Clone A WiDr COLO205 MCF7 COLO20 1

0.04
11. To choose the optimal

5-FU 0.01 0.01 0 002 0.29 0.02 0.04 0.44 0.02 0.99 0.44 0.004 021 1.50 >lO 049 >lO

Mitomycin 0.008 0.002 001 0.08 0.03 0.03 0.04 0.03 0.02 0.19 0.006 0.06 0.46 011 0 04 9.20

cell seeding concentration,

Bleomycin

Cisplatin

0.01 0.05 0.11 0.04 0.32 >I 0.07 0.10 3.60 7 90 >lO >lO >lO >lO >lO >lO

0.001 0.06 0.35 0.30 >l >1 >lO >lO >lO 9.50 >lO >lO >lO >lO >lO >lO

Methotrexate

Vincristine

<0.0001 <0.0001 <0.0001 <0.0001 0.003 0.007 0.45 1.56 0.0003 0.46 lO >lO 3 70 >lO >lO

lO

cells need to be seeded

and tested in replicate 96-well plates to maximize the number of cell doublmgs over the drug-exposure period between d 2 and 5.

148

Supino

12. MIT m solution can be stored and protected from hght at 4°C for several weeks, during which time it has been demonstrated that the absorption spectrum and the intensity of the solution does not change.

Some limitations

4.1. Limitations of MTT assay:

1. It cannot distmgmsh between a cytostatic and a cytocrdal effect, but this limitation can be overcome by changes in experimental design. 2. Individual cell numbers are not quantitated, and the results have to be expressed as a percentage of control absorbance. However, m order to know approximately the cell number, an optical density/cell number curve can be obtained immediately after the seeding of different cell concentrations in multiwell plates under the same experimental conditions. 3. The test is least efficient when carried out in medium that has supported cellular growth for several days, thus leading to an underestimation of control, untreated samples. 4. Formazan production, as well as the number of mitochondria per cell and/ or mitochondrial activity, can be induced by the drugs that cause perturbations in cell cycle phase distribution, resulting in an increase in cell dimensions. In some of these cases, further studies should be performed to evaluate the eventual incorrect evaluation of the experiment. (This is the case of interferon [IO]).

References 1 Ruben, R L. and Neubauer, R. H. (1987) Semiautomated calorimetric assay for in vitro screening of anticancer compounds. Cancer Treat. Rep. 71, 1141-l 149. 2. Alley, M. C , Scudiero, D. A., Monks, A , Hursey, M. L., Czerwinski, M J , Fine, D. L , Abbott, B J., Mayo, J. G., Shoemaker, R. H., and Boyd, M. R. (1988) Feasibility of drug screening with panels of human tumor cell lines using a microculture tetrazolium assay. Cancer Rex 48,589-601. 3. Carmichael, J., Mitchell, J. B., DeGraff, W G., Gamson, J., Gazdar, A. F , Johnson, B. E , Glatstein, E., and Minna, J D (1988) Chemosensitivity testing of human lung cancer cell lines using the MTT assay. Br. J. Cancer 57,540-547 4 Mosmann, T (1983) Rapid colonmetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J. Immunol Meth. 65,55-63 5. Carmichael, J , DeGraff, W. G., Gazdar, A. F , Minna, J. D., and Mitchell, J. B. (1987) Evaluation of a tetrazolium-based semiautomated calorimetric assay: assessment of chemosensitivity testing. Cancer Res. 47,936-942. 6. Yamaue, H., Tanimura, H., Tsunoda, T., Tani, M , Iwahashi, M., Noguchi, K , Tamai, M., Hotta, T., and Arii, K (1991) Chemosensitivity testing with highly purified fresh human tumour cells with the MTT calorimetric assay Eur J Cancer 27,1258-1263.

MTT Assays

149

7. Twentyman, P. R. and Luscombe, M. (1987) A study of some variables in a tetrazolium dye (MTT) based assay for cell growth and chemosensitivity. Br. J. Cancer 56,279-285.

8. Plumb, J. A., Milroy, R., and Kaye, S. B. (1989) Effects of the pH dependence of 3-(4,5-d~methylthiazol-2-yl)-2,5-diphenyltetrazolium bromide-formazan absorption on chemosensitivity determined by a novel tetrazolium-based assay. Cancer Res. 49,4435-4440.

9. Vistica, D T , Skehan, P., Scudiero, D., Monks, A., Pittman, A., and Boyd, M. R. (1991) Tetrazolium-based assays for cellular viability: a critical examination of selected paramenters affecting formazan production. Cancer Rex 51,25 15-2520. 10. Jabbar, S. A. B., Twentyman, P. R., and Watson, J. V. (1989) The M’IT assay underestimates the growth inhibitory effects of interferons. Br. J. Cancer 60,523-528.

CHAPTER 17

V79 Cytotoxicity Test for Membrane Damage Vera Bianchi 1. Introduction The cytotoxic effect of test chemicals in V79 cell culture can be determined by assessing damage to the plasma membrane as determined by a nucleic acid leakage assay. The plasma membrane is the first point of contact between cells and xenobiotics. Impairment of its molecular organization with consequent changes in permeability is a direct indicator of damage. The size of the intracellular components that leak out from the cells relates to the degree of damage that has occurred and can therefore form the basis of tests designed to assessmembrane-directed toxic effects. Preincubation of cells with 3H adenine results in the tracer being incorporated into the ATP pool via the salvage enzyme adenine phosphoribosyl-transferase, and ultimately into RNA and DNA following the reduction of ADP to dADP by ribonucleotide reductase. Radioactivity is thus distributed between the soluble nucleotide pool and the macromolecular nucleic acids. By varying the duration and timing of this preincubation with respect to exposure of cells to the test substance,it is possible to assessthe degree of membrane damage caused (I). If exposure follows on immediately after a short preincubation, the label will be present mainly within the soluble nucleotides and will be released into the medium after only moderate membrane damage. However, if preincubation is timed so as to allow most of the isotope to be incorporated into macromolecular nucleic acids, the presence of radioactivity in the medium will indicate more From Methods m Molecular B/ology, EcMed by S O’Hare and C K Atterwill

Vol 43 In Wtro Toxmty Testing Protocols Copynght Humana Press Inc , Totowa, NJ

151

152

Bianchi

extensive lesions. Furthermore, a comparison of isotope distribution between the various subcellular fractions after exposure to the test substance may provide information on the mechanism of cytotoxicity. Hamster cell monolayers are subjected to one of two protocols. In protocol 1 they are preincubated with 3H adenine for 1 h and then exposed immediately to the test substance. In protocol 2 they are preincubated with radioactive adenine for 24 h and left in nonradioactive medium for a further 24 h before being exposed to the test substance. After exposure the distribution of radioactivity is determined in the following fractions: medium (both in total medium and in the acid-precipitate); intracellular soluble nucleotide pool; intracellular macromolecules. The distribution of radioactivity between these fractions in test and control cultures is compared in order to estimate the degree of membrane damage. The use of tritiated adenine in the procedure described here results in an assay sensitive enough to detect toxic effects even at very low levels of membrane damage that would not be detectable by other methods. The protocol is technically simple, highly reproducible, and suited to any chemical that affects the integrity of the plasma membrane. However, the procedure does necessarily bear the disadvantages associated with the handling of radioactive materials. The use of a cultured cell line allows multiple assays to be performed rapidly and under consistent conditions. When using the protocol that requires only a short preincubation with the adenine, followed immediately by exposure to the test substance, the use of exponentially growing cells is recommended. This is because nucleotide metabolism is more active in such cells, and thus precursor uptake is more efficient. In addition, we have noticed that V79 cells treated with LAS are more resistant to the detergent at high cell density, probably because less of the cell surface is actually exposed. of Membrane Damage Tests that identify membrane damage by assessingchanges that have occurred in membrane permeability may be grouped into two main categories: dye exclusion/retention tests,and tests measuring the releaseof cellular components into the incubation medium. The useof, e.g., Trypan blue staining will provide gross estimatesof alterations in membranepermeability in a population of cells, but changesmay occur with respect to small molecules before the cells lose their ability to exclude the dye. Furthermore, cells that are taking in the dye may exhibit different degreesof membrane lesion. 1.1. Assessment

V79 Cytotoxicity

Test

153

The second category is able to provide more information since the size of the leaked material indicates the size of the “holes” in the membrane, and identification of the types of molecules that are being lost may add to interpretation of the toxic effect that is taking place. Although some currently available techniques, such as high pressure liquid chromatography and atomic absorption spectrometry, can detect leakage of small molecules and ions without the use of a tracer, such methods are time-consuming and thus not appropriate for routine use in large-scale screening of chemicals. More suitable for this purpose are assays that measure radioactivity released into the medium from cells that have been preincubated with labeled physiological precursors, such as nucleosides, bases, amino acids, or nonmetabolizable analogs. A nucleic acid precursor was chosen as a tracer for several reasons. There is no competition with medium ingredients, as occurs, for example, with the glucose analogs that have been used in leakage tests (2,3). The test may therefore be carried out in culture medium rather than in saline, which ensures that the cell metabolism remains as normal as possible. The spontaneous release back into the medium, which occurs with tracers that are not metabolized, e.g., the amino acid analog aminoisobutyric acid (3,4), is not a problem here because this tracer is quickly phosphorylated and further metabolized. Although polymerization of the labeled precursor into macromolecular nucleic acids removes radioactivity from the soluble nucleotide pool, this may be used to advantage since a further group of damage indicators, i.e., labeled nucleic acids of high molecular weight, is available to detect more extensive membrane damage. This may substitute for the use of enzymes, especially where low endogenous levels of the indicator enzyme or the exposure conditions for a given test chemical do not allow the use of an enzyme leakage assay, e.g., in the case of V79 cells exposed to LAS (5). Adenine was chosen in preference to other nucleic acid precursors because it is incorporated into the ATP pool, which is the largest nucleotide pool, and therefore makes a larger quantity of radioactivity available for release into the medium after membrane damage. In itself the presence of radioactivity in the extracellular environment does not prove loss of adenine nucleotides. However, this may be inferred from the presence of acid-precipitable, RNase-sensitive labeled material in the medium that demonstrates RNA leakage accompanied by nucleotide release.

154

Bianchi

of Nucleic Acid Release The two experimental protocols described allow the detection of different degrees of membrane damage. They are modified from those of Thelestam and Mollby (6,7). The modified procedure is faster, and in addition to indicating the size of “functional holes” in the membrane, it also provides a detailed picture of cytotoxicity with respect to the stability and metabolism of nucleic acids. In protocol 1, exposure to test chemicals takes place immediately after the cells have been incubated for a short period with tritiated adenine. At this stage and under the described conditions, 75% of total cell radioactivity is found within the soluble ATP pool. Radioactivity in the medium after exposure to the test substance indicates leakage of small molecules, i.e., minor membrane damage. It was found that the counts in the medium fraction exceeded those lost from the nucleotide pool becauseleakage of macromolecular nucleic acids had also occurred. However, since exposure had taken place while the cells were still metabolizing the tracer, a direct estimate of the amount of RNA lost was not possible. For this reason, a second protocol was developed. In protocol 2, the cells are preincubated for 24 h with the label, and then kept for a further 24 h in a nonradioactive medium before being exposed to the test substance. This allows for about 85% of the cell radioactivity to be incorporated into the macromolecular nucleic acids (DNA and RNA) and for the metabolism of the radioactive precursor to reach a plateau. Since incorporation of the label is complete, its distribution between the various cell fractions will be the same in all cells. Changes in this distribution will provide information on the effects of the test substance on the various cell components (RNA, DNA, soluble nucleotides). The amount of radioactivity present in the DNA of the cell monolayers is a direct measure of the cells remaining attached to the substrate, i.e., it indicates whether any cells became detached after exposure to the test substance. The counts lost from the RNA fraction are directly related to the degree of cytotoxrcity, becauseexposure occurs when RNA labeling is homogenous in all the cultures. The acid-precipitated labeled material in the medium indicates more extensive membrane damage, however, the counts of this fraction would in themselves tend to underestimate the damage. If the counts of this fraction cannot account for the counts lost

1.2. Quantitative

Determination

V79 Cytotoxicity

Test

155

from intracellular macromolecular RNA, it is possible to assume that exposure to the test substance has resulted in degradation of RNA, for example, by the release of lysosomal nucleases. 2. Materials

V79 cell line derived from hamster lung fibroblasts: The cells used by the author came from S. Bonatti, Laboratorio di Mutagenesi e Differenziamento de1CNR, Pisa, Italy, and originated with C. F. Arlett (Brighton, UK). Available from European Collection of Animal Cell Cultures (Porton Down, UK) (ECACC No. 86041102). 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

13 14 15 16 17 18 19 20 21

Incubator: 37’C, 5% CO2 atmosphere. 35 mm Petri dishes (Nunc [Denmark]). Scmtillation counter, e.g., Packard TriCarb 300C. Vortex mixer. Whatman GF/C glass fiber filters (Maidstone, UK). Filter chamber (cat. no. Xx2702550), Mrlltpore (Bedford, MA). Tritiated adenme (2-3H Ade, 24 Ci/mmol, supplied as 1 mCi/mL, i.e., 42 @4) (Amersham, UK). Dulbecco’s modified Eagle’s medium (DMEM), Gibco (Gatthersburg, MD). Heat inactivated fetal calf serum (Gtbco). Scmtillation fluid. Culture medmm. Dulbecco’s modified Eagle’s medium with 5% heat mactivated FCS. Hank’s balanced salt solution (components m g/L): CaClz (0.14 g), KC1 (0.40 g), KH,PO, (0.06 g), MgCl, . 6H,O (0.10 g), MgSO, q 7Hz0 (0.10 g), NaCl(8.00 g), NaHC03 (0.35 g), Na2HP04 .7H,O (0.09 g), o-glucose (1 .OO g). Adjust to pH 7.4. Versene solution (components in g/L): EDTA (0.2 g), NaCl (8.0 g), KC1 (0.2 g), Na2HP04 (1.75 g), KH,PO, (0.2 g). Trypsin solution, 0.25%. 60% methanol stock solution. 0.3M NaOH. Salmon sperm DNA stock solution, 1 mg/mL (Sigma, St. Louis, MO). tRNA stock solution, 1 mg/mL. 10% TCA solution contammg O.lM sodium pyrophosphate. 5% TCA solution. Test compound solutions made up in HBSS. Substances poorly soluble in

water are initially dissolved to 0 1M in DMSO

156

Bianchi

3. Methods 3.1. Cell Maintenance and Culture 1, Cells are grown from frozen stock: Thaw by dipping the ampules in water at 37OC. 2. Plate out 0.5 x lo6 cells in 10 mL culture medium into 10 cm diameter Petri dishes. 3. Passage on Mondays, Wednesdays, Fridays, with no replacement of medium m between. To passagecells: Remove medium from plates. Wash monolayer once with prewarmed versene solution, pH 7.2. Add 2 mL of 0.25% prewarmed trypsin solution. Incubate at 37°C for 5-10 mm until the cells detach from the plate (see Notes 1 and 2). Neutralize the trypsin with 4 mL of culture medium. 4. Make up cell suspension and seed cells into Petri dishes according to protocol requirements. Use exponentially growing cells in both protocols. 5. To replenish frozen stock: Centrifuge cells at 1000 rpm for 5-8 min to remove trypsin. Prechill the freezing medium (culture medium + 10% DMSO) by dipping the bottle m ice. Resuspend the pellet at lo6 cells/ml in freezing medium. Freeze 1.5 mL aliquots in ethanol-dry ice and store in liquid nitrogen.

3.2. Test Protocols Two protocols have been established to allow for different adenine incorporation into the cells.

1. 2. 3. 4. 5. 6. 7.

levels of

3.2.1. Protocol 1 (to Incorporate Adenine into Soluble Nucleotide Pool) Make up cell suspension of lo4 cells/ml. Seed 2 x lo4 cells/dish, i.e., 2 mL/drsh. Incubate 47 h. Aspirate off 1 mL medium, to leave 1 mL in dish. Incubate 1 h. Add 12 PL 3H Ade + 18 p.L double distilled water to each dish (final cont. Ade = 0.5 @4). Incubate 1 h. Rinse with HBSS (37”C), using about 2-3 ml/plate. Rinse the plates twice and aspirate off the saline carefully with a vacuum pump to remove extracellular radioactivity.

3.2.2. Protocol 2 (to Incorporate Adenine into Macromolecular 1. Make up cell suspension of 5 x lo3 cells/ml. 2. Seed lo4 cells/dish, i.e., 2 ml/dish.

Nucleic Acids)

V79 Cytotoxicity

157

Test

3. Incubate 12 h. 4. Add 4 pL 3H Ade + 26 p,L double distilled water to each dish (final cont. Ade = 0.05 @4). 5. Incubate 24 h. 6. Collect all the radioactive medium from each plate and place in individual tubes. Use two aliquots, 100 and 200 p,L, from each sample for scintillation counting. 7. Rinse with HBSS (37”(Z), using about 2-3 ml/plate. Rinse the plates twice and aspirate off salme carefully with a vacuum pump to remove extracellular radioactivity. 8. Add fresh culture medium (2 ml/plate) and incubate for 24 h. 9. Collect all the medium from each plate and place in individual tubes. Use two aliquots, 100 and 200 FL, from each sample for scintillation counting. 10. Rinse with HBSS (37°C). The remainder of the procedure is now the same for both protocols.

3.3. Exposure

to Test Compounds

Expose for 2 h to test substance dissolved in warm HBSS (use HBSS alone for controls). Use duplicate cultures for each concentration tested.

3.4. HBSS Fraction This fraction contains soluble nucleotides and macromolecules were released because of changes in membrane permeability.

that

1. Put the plates on ice to halt cell metabolism. 2. Collect HBSS into mdlvldual tubes and retam for scmtlllatlon counting.

3.5. Methanol

Fraction

This fraction contains the soluble intracellular nucleotide pool. Rinse the plates twice with ice-cold HBSS, using enough to remove all traces of the incubation mixture. Add 500 PL 60% methanol for 30 min at 4°C. Note: Make sure that the whole surface of the plate has been wetted with methanol. Collect the methanol fraction into individual tubes. Rinse the plates with 500 PL of methanol and add the fractions to the relevant tubes containing the first fraction.

3.6. AZkaZi Fraction This fraction contains the intracellular macromolecular (DNA and RNA) obtained on solubilization of the cells.

nucleic acids

Bianchi 1. Add 500 yL 0.3M NaOH for at least 2 h at 37°C. Alternatively the plates may be left overnight at room temperature. Decant the contents of each plate into individual tubes. 2. Count two aliquots for each fraction* a. HBSS fraction, 100 and 200 yL. b. Methanol fraction, 100 and 200 pL. c. Alkali fraction, 50 and 100 ~.LL. 3. In addition, under protocol B count the ahquots of medium removed at steps 7 and 10. 3.7. Nucleic Acid Precipitation 1. DNA: Take 2 aliquots, 25 and 50 pL, from the alkali fraction and place m tubes contammg 280 ltL double-dtstilled water. Add 20 p.L salmon sperm DNA stock solution. Add 1.5 mL 10% TCA. Vortex and keep on ice for 30 min. Collect precipitate on glass fiber filters. Wash the precipitate twice with 5% TCA and once with absolute ethanol (fill the filter chamber with the washing agent). Dry the filters by placing under an infrared lamp for not more than 10 mm. Count. 2. RNA: Take 2 aliquots, 100 and 200 l.tL, from the HBSS fraction and place mto tubes contammg 180 FL water. Add 20 FL tRNA stock solution. Add 1.5 mL 10% TCA. Vortex and keep on ice for 30 mm. Collect the precipitate on glass fiber filters. Wash the precipitate twice with 5% TCA and once with absolute ethanol. Dry the filters. Count. 3.8. Calculation

of Results

From dpm obtained for each aliquot calculate the total radioactivity m the respective fraction: HBSS, methanol, alkali, culture medium. In addition to calculating the total radioactivity of the HBSS fraction, calculate the radioactivity of the macromolecules that were released, i.e., the radioactivity left on the filters after RNA precipitation. Subtract this value from the total radioactivity of the HBSS fraction in order to determine the quantity of leakage from the intracellular soluble nucleotide pool. The radioactivity of the alkali fraction is equivalent to the total intracellular nucleic acid radioactivity (i.e., DNA and RNA). The radioactivity of the filters after DNA precipitation gives the value for intracellular DNA, which in protocol B is an index of the number of cells that remained attached to the dishes, The value for cellular RNA is obtained by subtracting DNA radioactivity from that of the alkali fraction. Check the recovery rate from the different plates by adding together the values

V79 Cytotoxicity

159

Test

Table 1 Distribution of the Radioactivity Counts Among the Cellular Fraction and the Medium m Cultures0

Treatment SDS, mg/L 0 20 30 40

Recovered radioactivity, dpm x 1O-3 Medium Macromolecular nuclear acids Acid Nucleotide Total prec. pool Total DNA 49 94 182 306

12 31 69 119

207 186 152 64

471 328 247 158

98 91 67 63

Sumb 727 889 581 528

Treated for 2 h with SDS accordmg to protocol B. bSum of radioactivity counts m medium + nucleotlde pool + macromolecular nucleic acids

for the HBSS, methanol, and alkali fractions. The sum total should be the same for all cultures. 3.9. Experimental Data Table 1 lists the counts obtained in the various fractions of cells exposed to SDS according to protocol B. The counts of the medium fraction increase with increasing quantity of detergent. At the highest concentration of detergent, less than half of the radioactivity in the medium fraction came from the acid-precipitable material. Since this medium fraction contained far more radioactivity than was found in the soluble nucleotide fraction of control cultures, it may be taken as an index of RNA degradation that was caused by the detergent. The DNA fraction shows relatively constant results, indicating that the decrease in the total macromolecular nucleic acid fraction results from loss of RNA. This indicates that most of the cells are attached to the plate and are not completely solubilized by the detergent. Variations in the total sum of radioactivity from all the fractions combined arise from losses during processing of the samples. 4. Notes

Cells should be passagedat least once after being thawed before they are usedin testing. 2. Cells should remam outsidethe incubatorfor as short a time as possrbleto minimize chilling. 1.

160

Bianchi

References 1. Fortunati, E. and Bianchr, V. (1989) Plasma membrane damage detected by nuclerc acid leakage. Mol. Toxic01 1,27-38. 2. Walum, E. and Peterson, A. (1982) Tritiated 2-deoxy-D-glucose as a probe for cell membrane permeability studies. Anal. Biochem 120,8-l 1. 3 Malik, J. K., Schwartz, L. R., and Wiebel, F. J. (1983) Assessment of membrane damage in continuous culture of mammalian cells. Chem. Biol. Interact 45,29-42. 4. Thelestam, M. and Mollby, R (1975) Sensmve assay for detectron of toxin-induced damage to the cytoplasmic membrane of human drploid fibroblasts. Infect. Zmmun. 12,225-232.

5. Bianchr, V. and Fortunatr, E (1990) Cellular effects of an anionic surfactant detected in V79 fibroblasts by different cytotoxicity tests. Toxicol. In Vitro 4,4-16. 6. Thelestam, M. and Mollby, R (1975) Determination of toxin-induced leakage of different-sized nucleotides through plasma membrane of human drplord fibroblasts Infect Immun. 11,640-648 7. Thelestam, M. and MBllby, R. (1976) Cytotoxic effects on the plasma membrane of human diploid tibroblasts: a comparative study of leakage tests. Med. Biol. 54,39-49.

CHAPTER 18

SIRC Cytotoxicity Odile Blein-Sella

Test

and Monique

Adolphe

1. Introduction In this test, rabbit-derived cornea1cells are cultured in the presence of test compounds, the toxicity of which are determined by their effects on cell viability. A decrease in cell number, as measured by uptake of the dye Neutral Red, serves as an indicator of potential cytotoxicity. This test has been proposed as a potential replacement alternative for the Draize Eye Irritation test (I). Healthy SIRC cells (an established cell line, ATCC CCL60), when maintained in culture continuously, divide and multiply over time. The basis of this test is that a cytotoxic chemical (regardless of site or mechanism of action) will interfere with this process and, thus, result in a reduction of the growth rate as reflected by cell number. The degree of inhibition of growth, related to the concentration of the test compound, provides an indication of toxicity (2,3). SIRC cells are maintained in culture and exposed to a range of concentrations of test compound for 24 h. After a visual examination, the cultures are rinsed and incubated for 3 h in medium containing Neutral Red, which is taken up by viable cells. After rinsing, the dye present in the cell population is liberated and the amount quantified using a spectrophotometer in order to obtain an indication of cell number. The number of cells in the presence of the test compound is compared with that observed in control cultures and the percentage inhibition of growth is calculated. The IC5, (i.e., the concentration producing 50% inhibition of growth) is determined and expressed as mg/mL. These valFrom- Methods m Molecular Biology, Edited by S O’Hare and C K. Atterw~ll

Vol. 43: In Wro Toxroty Testing Protocols Copyright

161

Humana

Press

Inc , Totowa,

NJ

162

BLein-Sella

and Adolphe

ues enable a comparison of the relative toxicity of the test compounds to be made (4). 1.1. Cell Culture Procedure The maintenance and culture of a cell line, such as SIRC cells, is a relatively simple and inexpensive technique. The application of such cultures to determine cytotoxrcrty may potentially allow rapid, highly reproducible testing of many chemicals on a routine basis (5). This assay is currently being evaluated in a validation study organized by Oeuvre pour l’assistance aux ammaux de laboratoire (OPAL) (the French association for the welfare of laboratory animals), as a potential replacement alternative for the Draize Eye Irritation test. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

18 19.

2. Materials Cell lme: SIRC (rabbit cornea1 cell line). 37’C mcubator, humidified, 5% C02/95% air Phase-contrast microscope. Cryotubes. Centrifuge. -80°C freezer. Liquid mtrogen dewar Hemocytometer. 25 cm2 tissue culture flasks. 75 cm2 tissue culture flasks. 96-Well tissue culture microtiter plate. Reagent basin. Multichannel pipet. Microtiter plate shaker Microplate reader, measurement filter 540 nm. Phosphate buffered salme without calcmm or magnesium (PBS). Culture medium: MEM supplemented with: a. For routine culture: glutamme, 2 mM; gentamicm, 4 pg/mL; fetal calf serum, lo%, and sodmm bicarbonate. b. For treatment: glutamme, 2 mM; gentamicm, 4 pg/mL; fetal calf serum, 5%; and sodium bicarbonate. The concentration of FCS IS reduced to 5% in treatment medmm to munmlze possible mteractions between the tested product and serum components. Note: Complete medium may be kept for l-2 wk at 4°C Trypsm-EDTA solution: 0.1% trypsm/0.02% EDTA m PBS Neutral Red stock solution: 4 mg/mL solution of Neutral Red m distilled water It is recommended that this solution be replaced frequently, I e.,

NRC

20.

21.

22. 23.

Cytotoxzcity

163

Test

every 2 wk, since crystals of Neutral Red appear if the solution 1s kept for a long time. Neutral Red medium. Dilute the Neutral Red solution 1:80 m treatment culture medium to give a fmal concentration of 50 pg/mL. Prepare Neutral Red medium and incubate at room temperature m a dark place, overnight. Centrifuge twice at 1500g for 10 mm before use to remove any fine, needlelike precipitate of dye crystals. Formol-calcmm solution: 40% 1 mL formaldehyde, 10 mL calcium chloride, 10% 89 mL distilled water. Note: This solution may be kept for several months at 4°C. Acid-ethanol solutton’ 1 mL acetic acid, 99 mL ethanol. Note: This solution may be kept for several months at 4°C. Test compounds should be dissolved in sterde medium, ethanol, or dimethyl sulfoxlde (DMSO), as appropriate. The final solvent concentration should be kept at l%, or less, m culture medium, since this level of solvent does not adversely affect the growth of SIRC cells (see Note 1).

All solutions, glassware, and so on, are sterile and all procedures are carried out under aseptic conditions and in the sterile environment of a laminar flow cabinet (biological hazard standard). 3. Methods 3.1.

Cell

Maintenance

3.1.1. Preparation

of Cells for Freezmg 1. Detach cells from the bottom of the flask by trypsmlzatlon (see Section 3.1.3.) and wash with complete medium. Count the resulting cell suspension using a hemocytometer. 2. Centrifuge at 4°C for 7 min at 400g. Resuspend the pellet at a concentration of 3 x lo6 cells/ml m complete medium containing 10% DMSO. Allquot 1 mL of the cell suspension/cryotube. Transfer cryotubes lmmediately mto a -80°C freezer. After 2 h, rapidly transfer the cryotubes mto a liquid nitrogen dewar. The cells may be stored m hquld mtrogen for several years.

3.1.2. Thawing

1.

2. 3. 4.

and Culture

of Cells When required, thaw cells rapidly by immersing frozen cryotubes directly into water bath at 37°C. This avoids damage to the cells from the high concentration of DMSO. Immediately resuspend the cells in complete growth medium. Centrifuge for 7 mm at 400g in order to wash cells. Repeat procedure. Transfer cells mto a 25 cm2 culture flask m complete growth medium.

164

1. 2. 3. 4. 5. 6. 7. 8. 9.

Blein-Sella

and Adolphe

3.1.3. Subculture of Cells When the cultures approach confluence, remove the cells from the flask by trypsinization. The flask is trypsinized once a week. Decant the medium and add 5 mL trypsin-EDTA solution at room temperature. Leave in contact with the cells for 5 s. Remove and repeat the procedure. Finally, remove the trypsin solution and incubate the flask at 37°C untrl the cells round up. Add 10 mL of complete growth medium to neutralize the trypsin activity. Disperse the cells by trituration. Count the cell suspension and either use for splittmg into new flasks or for test purposes. For routine subculturing, seed cells at 1.5 x lo6 cells/75 cm2 flask/l5 mL culture medium (see Note 2).

3.2. Test Procedure 1. After growing up the cells, prepare a cell suspension of 7.5 x lo3 cells/ml in complete culture medium. Cells are kept suspended by the use of a magnetic stirrer. 2. Transfer 10 mL of cell suspension into a reagent basin. Leave the first column, i.e., wells Al-H1 (Table 1) free of cells. These wells act as a blank. 3. Using a multichannel pipet aliquot 0.2 mL of cell suspension into the remaining wells of a 96-well microtiter plate to give a final concentration of cells of 1.5 x lo3 cells/well. 4. After aliquoting the cell suspension to a row of wells, triturate the remaining suspension using the multichannel pipet. 5. Transfer the plates to a microtlter plate shaker. Shake the plate gently (speed 6, 1000 rpm), on a Dynatech vari-shaker, to ensure an even distribution of cells in each well. 6. Incubate at 37°C in an humidified 5% CO2 atmosphere for 24 h. This overnight incubation allows the cells to adhere and recover from trypsm exposure. 3.2.1. Range Finder 1. Prepare a series of 7 test chemical concentrations immediately before use. a. 0.0001, 0.001, 0.01, 0.1, 1, 10, and 100 mg/mL of test chemical. b. 1% Solvent control (see Note 3). 2. Invert plate, tap out growth medium, and replace with 0.2 mL of each chemical dilution to the appropriate wells. Eight wells are treated per concentration of test chemical. Aliquot treatment medium alone into 2 rows of

SIRC Cytotoxicity

165

Test

Table 1 Plating-Out Scheme for SIRC Cytotoxicity Test Performed on 96-Well Plates

A B C D E F G H

1

2

3

4

5

6

7

8

9

10

11

12

B B B B B B B B

C C C C C C C C

SC SC SC SC SC SC SC SC

Tl Tl Tl Tl Tl Tl Tl Tl

T2 T2 T2 T2 T2 T2 T2 T2

T3 T3 T3 T3 T3 T3 T3 T3

T4 T4 T4 T4 T4 T4 T4 T4

T5 T5 T5 T5 T5 T5 T5 T5

T6 T6 T6 T6 T6 T6 T6 T6

T7 T7 T7 T7 T7 T7 T7 T7

-

C C C C C C C C

B. Control blank, i e , no cells added to the well, C: Control cultures, I e., cells are not exposed to test compound, but only to treatment medium; SC. Solvent control, I.e., cells are not exposed to test compound, but only to treatment medium containing 1% solvent; Tn: Test cultures, 8 wells exposed to the same concentration of test compound

3. 4. 5. 6.

wells on either srde of the plate, i.e., wells A2-H2 and A12-H12 (Table 1). These wells serve as controls. Incubate the treated plates for 24 h in a humidified 5% CO* atmosphere at 37OC. After 24 h, visually examine the cells by phase contrast microscopy. Record any morphological changes in the cells and visually estimate the concentration of test chemical at which toxicity occurs. Remove the medmm and determine cell number by the Neutral Red Uptake assay (see ref. 3). From the preliminary results, select several concentrations, spanning the range O-100% cell death, for an accurate determination of cytotoxicity.

3.2.2. Determination

of IC,,

IC,,

and ICaO

1. Test each chemical concentration in 8 wells on three separate occasions. 2. Prepare: a The appropriate range of 7 test chemical concentrations. b. The appropriate solvent controls. For example, from the range-finder, if the toxicity is found between 0.1 and 1 mg/mL, the following range of concentrations would be tested: 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, and 0.8 mg/rnL. 3. Prepare the 96-well plate as before. 4. After the overnight mcubation, remove growth medmm and replace with 0.2 mL of the test chemical or the control into the appropriate wells. Incubate for 24 h in a humidified 5% COz atmosphere at 37°C.

Blein-Sella

and Adolphe

5. After 24 h, visually examine the cells by phase contrast microscopy. Record any morphological changes in the cells and visually estimate the concentration of test chemical at which toxicity occurs. Determine cell number by the Neutral Red Uptake assay (see ref. 3).

3.3. Neutral

Red Uptake

Assay

1. At the end of the 24 h treatment period, invert plate and tap out medium. 2. Wash cells with 0.2 mL volume of 0.9% sodium chloride solution (at room temperature) per well. 3. Add 0.2 mL Neutral Red medium to each well. 4. Incubate for 3 h m a humidified 5% COz atmosphere at 37OC. 5. After 3 h remove the Neutral Red medium. 6. Wash rapidly (withm 1 mm) with 0.2 mL formol-calcium fixative per well. 7. Add 0.2 mL acetic acid-ethanol solution to each well to extract the dye from the cells, 8. Shake the plates rapidly (speed 6, 1000 rpm) on a Dynatech van-shaker for 15 min. 9. Measure the absorbance of the resulting colored solution at 540 nm on a microplate reader using the blanks as a reference (see Note 4).

3.4. Results 1. Express the optical density values obtained from those wells that have been treated with chemicals, as a percentage of those values obtained with control cultures. Typical optical density values for control wells is 0.5-0.6 (see Note 5). 2. Mean the values for each treatment group and plot graphically. Determine the IC2e, I&,, and ICsOvalues from the curve. 3. Mean IC values form 3 separate experiments and express the final concentrations as mg/mL or rnJ4 values. Rank the chemicals using the I&, value, since this is the section of the curve most likely to be linear and subject to least variation, 4. The ICse% is the concentration of toxtcant that induces a 50% decrease m absorbance.

4. Notes 1. Testing of volatile and msoluble compounds. Volatile substancestend to evaporate under the conditions of the test, which may lead to interference between wells and to variable IC,, values. This is especially true for those compounds that have a fatrly low toxicity. 2. If the cells are not subcultured or used for test purposes, the medium should be changed every 2 d, and 3 d for the weekend.

SIRC Cytotoxicity

Test

167

3. Solutions of liquids and solids are weighed and made up on a v/v or w/v basis, and expressed m mg/mL. It IS recommended that viscous solutions should also be weighed to avoid problems of pipeting. It is the authors’ opinion that chemicals that are insoluble in medium or solvents compatible with cell culture should not be tested by this technique. 4. Neutral Red is preferentially taken up into the lysosomes/endosomes of the cell. Absorbances obtained using the Neutral Red Uptake assays have been shown to correlate linearly with cell number over the specific optical density range obtamed using this method. Any chemical having a localized effect on the lysosomes/endosomes will, therefore, result in an artificially low (or possibly high) reflection of cell viability and cell number. This factor does, however, make the system useful to detect other chemicals that selectively affect the lysosomes, especially when it is used in conjunction with other tests capable of determining cell number. 5. One major drawback of the assay is the precipitation of the Neutral Red dye into visible, fine, needle-like crystals. When this occurs, it is almost impossible to reverse, thus producing inaccurate readings. Some chemicals induce this precipitation, therefore a visual inspection stage in the procedure is very important.

References 1. Borenfreund, E., Babich, H., and Martin-Alguacil, N. (1988) Comparison of two in vitro cytotoxicity assays. The Neutral red (NR) and tetrazolium MTT test Toxicol In Vitro 2, 1-6 2. Borenfreund, E. and Borrero, 0. (1984) In vitro cytotoxicity assays Potential alternatives to the Draize ocular irritancy test. Cell Biol. Toxicol. 1,33-39. 3. Borenfreund, E and Puerner, J. A. (1984) A simple quantitative procedure using monolayer cultures for cytotoxicity assay (HTDM 90). J. Tiss. Cult. Meth. 9,7-9. 4. Borenfreund, E. and Puerner, J. A. (1985) Toxicity determined in vitro by morphological toxicology. Toxicol. Lett. 24,119-124. 5. Borenfreund, E. and Shopsrs, C. (1985) Toxicity monitored with correlated set of cell culture assays Xenobiotica 115,7057 11

CHAPTER19

Rabbit Articular Chondrocyte Functional Toxicity Test OdiZe BZein-SeZZa

and Monique

AdoZphe

1. Introduction In this test, rabbit articular chondrocytes are cultured in the presence of a test compound, the toxicity of which is then determined by its effect on the production of proteoglycan by the cells, as detected by the dye Alcian Blue. Healthy, freshly isolated, articular chondrocyte cells can be successfully maintained for a short period of time but still retain certain characteristics of the cells in vivo, such as their ability to secrete the components that make up the background matrix of cartilage. In this test system, deleterious effects on cell function are determined by monitoring the secretion of proteoglycans by healthy cells, which can be stained by the dye Alcian Blue (I). Freshly isolated rabbit articular chondrocytes are cultured and exposed to varying concentrations of test compounds. The cultures are incubated for 72 h, after which time the cells are stained with Alcian Blue. After rinsing, the dye taken up by the proteoglycan matrix is liberated and measured spectrophotometrically (2). Comparison between the optical density values of cells exposed to test compounds and control cultures provides an index of toxicity. This procedure requires the isolation of cells from rabbit joints, since: 1. At present,chondrocytecell lines are unavailable.

From. Methods in Molecular Bology, Vol 43’ In Wtro Toxmty Tesbng Protocols Edlted by. S O’Hare and C. K Atterw~ll Copyright Humana Press Inc , Totowa, NJ

169

Adolphe

170

and Blein-Sella

2. Although chondrocytes can be maintained m culture for several weeks, they rapidly lose their differentiated phenotype, i.e., their ability to secrete type II collagen and proteoglycans. It should be noted that the isolation of such cells is a time-consuming procedure that requires fresh animal tissue and considerable practical experience. 1.1. Alcian Blue Staining Alcian Blue is a very specific dye that associates with the sulfate

groups found along the length of proteoglycan molecules. Absorbance obtained after Alcian Blue staining is proportional to the amount of proteoglycans secreted in chondrocyte monolayers. Thus, any chemical having an effect on proteoglycan synthesis can be detected with this method.

2. Materials 1. Rabbit articular chondrocyte cells freshly isolated from the shoulder and knee joints of 2-3-mo-old rabbits Young rabbits are sacrificed to mcrease the yield of chondrocytes isolated 2. 37°C incubator, humidified, 5% CO,/95% air. 3. Two-compartment digestion chamber. 4. Magnetic stirrer. 5. Centrifuge. 6. 96-Well tissue culture microtiter plate. 7. 75 cm* tissue culture flask. 8. Reagent basin. 9. Multrchannel pipet. 10. Hemocytometer. 11. Phase contrast microscope. 12. Microtiter plate shaker. 13. Microplate reader, measurement filter 630 nm. 14. Gey’s balanced salt solution (g/L): NaCl (go), KC1 (0.375), CaCl* s2H,O (0.225), MgC12 +6Hz0 (0.210), MgS04 . 7Hz0 (0.070), Na2HP04 s7Hz0 (0.226), KH2P04 (0.030), NaHCOs (0.277), glucose (1 .O). 15. 0.2% Trypsin in Gey’s solution, 16. 0.2% Collagenase in Gey’s solution. 17. Culture medium: HAM F12 medmm supplemented with 2 mM glutamme, 10% fetal calf serum(FCS),4 pg/mL gentamlcm, and sodium bicarbonate. Complete medium can be stored for l-2 wk at 4°C. 18. Trypsin-EDTA solution: 0.1% trypsin/0.02% EDTA m phosphate buffered saline (PBS) without calcmm and magnesium. 19. Alcian Blue solution: 0.5% w/v Alcian Blue in O.lN hydrochloric acid. This solution can be kept for several months at room temperature,

Rabbit

Articular

Chondrocyte

Test

171

-

Class

stopper

-

Glass

cylinder

-

Magnetic

-

Nylon

filter

-

Ring

holding

-

Glass

-

Bottle

Fig. 1. Two-compartment

bar

the

Filter

triangle

digestion

chamber.

20

Acid-ethanol solutton 2% glacial acetic acid in absolute alcohol. This solution can be kept for several months at 4°C. 21, 4M Guamdine hydrochloride. This solution can be kept for several months at 4°C. 22. Test compounds should be dissolved m sterile medium, ethanol, or dtmethyl sulfoxide (DMSO), as appropriate. The final solvent concentration should be kept at l%, or less, m culture medium, since this level of solvent does not adversely affect the growth of chondrocyte cells.

All solutions, glassware, and so on, are sterile, and all procedures are carried out under aseptic conditions and in the sterile environment of a laminar flow cabmet (biological hazard standard) (see Note 1). 3. Methods

3.1. Isolation

of Articular

Chondrocytes

1. Sacrrfrce rabbits and dissect the cartilage from the shoulder and knee joints. Cut the cartilage up into small pieces in HAM F12 medium. 2. Place the dtssected cartilage mto the mternal compartment of a two-compartment digestion chamber (Fig. 1).

172

Adolphe and Blein-Sella

3. Add 0.2% trypsin in Gey’s solution and digest the tissue at 37°C for 30 min. Throughout successrve digestions, continually stir the digestion mixture using a stirrer and magnetic flea. 4. Dissociate the chondrocytes by adding 0.2% collagenase in Gey’s solution to the central well of the two-compartment digestion chamber. 5. Incubate the digestion chamber for 1 h at 37°C. Repeat the digestion twice more, but decrease the incubation time to 45 mm. The collagenase drgestion releases isolated chondrocytes that pass from the internal to the external compartment. 6. Centrifuge the resulting cell suspension. 7. Wash the cells in supplemented HAM F12 medium. 8. Count the cell suspension using a hemocytometer. 9. Seed the cells at a concentration of 1.5 x lo6 cells/75 cm2 flask. 10. Change the medmm m the flask 2 d after seeding. The cells can be used for testing purposes 5 d after imtial isolation. Primary cultures of chondrocytes are used in this procedure. Cells should be used not longer than a week after initial isolation since rabbit articular chondrocytes lose their differentiated phenotype in culture if they are grown for a prolonged period. 1. 2. 3. 4. 5. 6.

1. 2. 3.

4.

3.2. Cell Maintenance When the cultures approach confluence, remove the cells from the flask by trypsinization. Decant the medium and add 5 mL TrypsinEDTA solutron at room temperature. Leave in contact with the cells for 20 s. Remove the trypsmEDTA solutron and incubate the flask at 37°C until the cells round up. Add 10 mL of complete growth medium to neutralize the trypsin activity. Disperse the cells by trituration. Count the cell suspension.The cells may now be used for the test procedure (see Note 2). 3.3. Test Procedure Prepare a cell suspension of 3.5 x lo5 cells/ml in complete culture medium. Cells are kept suspended by the use of a magnetic stirrer. Transfer 10 mL of cell suspension mto a reagent basin. Using a multichannel pipet, aliquot 0.2 mL of cell suspension into every well of a 96-well microtiter plate to give a fmal cell concentration of 7 x lo4 cells/well. After the addition of cell suspension to a row of wells, triturate the remaining suspension using the multichannel prpet. Transfer the plate to a microtiter plate shaker.

Rabbit Articular

A B C D E F G H

173

Chondrocyte Test

Table 1 Plating-Out Schemefor Rabbit Articular Functional Toxicity Test Performed on 96-Well Plates 123456789 10 11

12

C C C C C C C C

C C C C C C C C

SC SC SC SC SC SC SC SC

Tl Tl Tl Tl Tl Tl Tl Tl

T2 T2 T2 T2 T2 T2 T2 T2

T3 T3 T3 T3 T3 T3 T3 T3

T4 T4 T4 T4 T4 T4 T4 T4

T5 T5 T5 T5 T5 T5 T5 T5

T6 T6 T6 T6 T6 T6 T6 T6

T7 T7 T7 T7 T7 T7 T7 T7

-

-

C: Control cultures, i.e., cells are not exposed to test compound, but only to treatment medium, SC: Solvent control, i e., cells are not exposed to test compound, but only to treatment medmm supplemented with solvent; Tn. Test cultures, 8 wells exposed to the same concentration of test compound; B: Blanks are not required since each well 1sblanked during measurement of Alcian Blue stain.

5. Shake the plate gently (speed 6, 1000 rpm) to ensure an even distribution of cells in each well. Incubate at 37°C in a humidified 5% CO2 atmosphere for 24 h. This overnight exposure allows the cells to adhere and recover from the trypsm exposure.

3.4. Range Finder 1. Prepare a series of 7 test chemical concentrations immediately before use. a. 0.0001,0.001,0.01,0.1, 1, 10, and 100 mg/mL. b. 1% solvent control. A fixed dose may also be tested. 2. Invert plate, tap out culture medium, and replace with 0.2 mL of each chemical dilution to the appropriate wells. 3. Carefully add solutions of test chemical to each well since the cell monolayer is very fragile. Eight wells are used per concentratron of chemical (see Table 1). 4. Aliquot medmm alone mto 2 rows of wells on either side of the plate, i.e., wells Al-HI and A12-H12. These wells serve as controls. 5. Incubate the treated plate for 72 h at 37°C. 6. After 72 h, visually examine the cells using phase contrast microscopy. 7. Record any morphological changes in the cells and visually estimate the concentration of test chemical at which toxicity occurs. 8. At the end of the 72 h Incubation period, remove the medium and determine the secretion of proteoglycan by Alcian Blue staining (3).

174

Adolphe and Blein-Sella

9. From the preliminary results select several concentrations, spannmg the range O-100% proteoglycan secretron, for an accurate determination of cytotoxicity. 1. 2.

3. 4. 5. 6. 7.

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

3.5. Definitive Determination of Toxicity Test each chemical concentration in 8 wells on three separate occasions. Prepare: a. The appropriate range of 7 test chemical concentrations. b. The appropriate solvent controls. For example, from the range-finder, if the toxicity is found to be between 0.1 and 1 mg/mL, the followmg range of concentratrons would be tested: 0.2,0.3,0.4,0.5,0.6,0.7, and 0.8 mg/mL-prepare the 96-well plate as before. After the overmght mcubatron, remove the growth medmm and replace with 0.2 mL of the test chemical or the control, in the appropriate wells. Incubate for 72 h at 37°C. After 72 h, visually examme the cells using phase contrast microscopy. Record any morphological changes in the cells and visually estimate the concentration of test chemrcal at whrch toxicity occurs. At the end of the 72 h mcubation determine the secretton of proteoglycan by Alcian Blue stainmg (see Section 3.3). 3.6. Alcian Blue Staining After 72 h, visually examme the cells using phase contrast microscopy. Record any morphological changes m the cells and visually estimate the concentration of test chemical at which toxicity occurs. Remove medium and carefully wash the cells twice with a fixed volume of PBS. Add 0.2 mL acid-ethanol solution to each well to fix the cells. Fix for 5 min. If necessary, the plates may be kept at this step at 4°C. Rehydrate the cells through 95% ethanol for 5 min. Repeat using 70% ethanol. After fixation, add Alcian Blue solution to each well and leave to stain overnight, i.e., 18-20 h. Rinse the plates twice with 200 FL vol of O.lNHCl to remove any unbound dye. Finally, rinse once with 200 PL distilled water. Add 100 pL of 4M guamdine HCl to each well, to extract the Alcian Blue from the cells. Seal the plates with “parafilm,” and allow to stand overnight at 4°C. Shake the plates rapidly (speed 6, 1000 rpm) on a rmcrotiter plate shaker at room temperature. Read the absorbance at 630 nm on a microplate reader.

Rabbit

Articular

Chondrocyte

Test

175

12. After obtammg a gross reading, remove the stain and wash the cells twice with distilled water. 13. Add 100 l.tL of 4M guanidme HCl to each well. 14. Reread the plate to obtam blank values. 15. Subtract these latter values from the gross values to give the net absorbance of extracted Alcian Blue stain.

3.7. Results 1, Express the optical density (at 630 nm) of the cells that have been exposed to test chemical as a percentage of control culture. 2. Mean the values for each treatment group and plot graphically. Typical optical density of controls may range between 0.25 and 0.4. 4. Notes 1. Solutions of liquids and sohds are weighed and made up on a v/v or w/v basis, and expressed in mg/mL. It is recommended that viscous solutions also be weighed to avoid problems of pipeting It is the authors’ opinion that chemicals that are insoluble m medium or in solvents that are incompatible with cell culture should not be tested by thts technique. 2. In order to express their differentiated phenotype, chondrocytes must have reached confluence. Thus the cells are seeded at a high cell density. References 1 Hassell, J. R. and Horigan, E. A. (1982) Chondrogenesls: a model developmental system for measurmg the teratogemc potential of compounds Terutogenesis, Carcmogenesis Mutagenesis 2,325-33 1 2 Paulsen, D. F. and Solrush, M. (1988) Microtiter micromass cultures of limb-bud mesenchymal cells. In Vitro CelE Dev Blol. 24(2), 138-147

CHAPTER20

Balbh

3T3 Cytotoxicity

H. M. Liebsck

and Horst

Test Spielmann

1. Introduction The cytotoxic effect of chemicals on Balb/c 3T3 cells in culture is measured by cell viability (Neutral Red Uptake) and total cell protein (Kenacid Blue R dye binding method). Healthy Balb/c 3T3 cells (an established cell line), when maintained in culture, continuously divide and multiply over time. The basis of this test is that a cytotoxic chemical (regardlessof site or mechanism of action) will interfere with this process and, thus, result in a reduction of the growth rate as reflected by cell number. The degree of inhibition of growth, related to the concentration of the test compound, provides an indication of toxicity. Balb/c 3T3 cells are maintained in culture and exposed to test compounds over a range of concentrations. The cultures are visually examined after 24 h, the highest tolerated dose (HTD) is estimated, and the number of viable cells and/or the total cell protein content are determined after 24 h exposure by the Neutral Red Uptake and Kenacid Blue methods, respectively. The nature of the assays are such that both may be performed on the same cultures, provided that the Neutral Red Uptake determination is performed first (an indication of the number of viable cells). The number of cells in the presence of test chemicals is compared with that observed in control cultures and the percentage inhibition of growth calculated. The I& concentration (i.e., the concentration producing 50% inhibition of growth) is determined and expressed as pg/mL or mmol/L. From Methods m Molecular Biology, Vol. 43. In Wro Toxrcity Testmg Protocols Edited by: S O’Hare and C. K Atterw~ll Copynght Humana Press Inc , Totowa, NJ

177

178

Liebsch and Spielmann

These values enable a comparison of the relative cytotoxicities of the test compounds to be performed. The maintenance and culture of a cell line such as Balb/c 3T3 cells is a relatively simple and inexpensive technique. The application of such cultures to determine cytotoxicity enables the rapid, highly reproducible testing of many chemicals on a routine basis. There are certain limitations of the technique, some of which concern the character of the compounds to be tested. The system is likely to underestimate the toxicity of chemicals that require metabolic activation to a toxic intermediary or product. However, metabolic activation is not an essential factor for assessing the irritation potential of chemicals. Substances that specifically attack dividing cells may appear to be of a much higher order of toxicity than they would in vivo. The toxicity of substances that bind to serum proteins (i.e., such as those found in newborn calf serum) may be also underestimated. 1.1. Neutral Red Uptake Assay Neutral Red is preferentially taken up into the lysosomes/endosomes of the cell. Any chemical having a localized effect on the lysosomes/ endosomes will, therefore, result in an artificially low (or possibly high) reflection of cell viability and cell number. This factor does, however, make the system useful to detect those chemicals that selectively affect the lysosomes, especially when it is used in conjunction with other tests that are capable of determining cell number. 1.2. Kenacid Blue R Dye Binding Assay Advantages of this system include: 1. It can be repeated more than once on the same cells. 2. Cells can be fixed and the stammg performed later. 3. The cell distribution can easily be seen with the naked eye when stained with the Kenacid Blue before desorbmg, thus giving a rapid indication of the successof the assay. Despite the limitations of the system, it provides a simple screen for the rapid assessment of the toxicity of compounds. The advantages and disadvantages of both the Neutral Red Uptake and the Kenacid Blue protein assays are presented above. A direct comparison of the Kenacid Blue and the Neutral Red methods may also be of value to someone considering choice of endpoint.

Bulb/c

3T3 Cytotoxicity

Test

179

Once initiated, the Neutral Red Uptake assay must be completed, i.e., once the cells have been incubated with the Neutral Red and the dye is taken up into the lysosomes, the process of fixing and destaining must follow immediately. With the Kenacid Blue assay the cells can be fixed and the staining/destaining performed later. The Kenacid Blue test can be repeated more than once on the same cells, i.e., once the cells are fixed the procedure of dye addition and destaining can be repeated several times. This is obviously out of the question with the Neutral Red assay, which is dependent on live cell uptake of the dye. Although there is a danger of the Kenacid Blue dye precipitating out, it is readily solubilized by agitating the dishes. However, if Neutral Red crystals form they are almost impossible to resolubilize without removing the stain from the cells as well. This greatly alters the reliability of any readings obtained and has proven to be an occasional occurrence, induced by some chemicals. Another disadvantage of the Neutral Red assay is the possibility that deceptively low cell viability or cell number readings will result in those caseswhere a chemical has a relatively selective effect on the lysosomes/endosomes of the cell. An example of this would be chloroquine sulfate, which alters the pH of lysosomes/ endosomes, an effect that inhibits Neutral Red uptake. One advantage of the Neutral Red assay is that it detects only viable cells. Total protein measurement does not make allowances for necrotic cells that may still be attached to the culture dish and, therefore, may underestimate the toxicity of a compound. It should be noted, however, that the occurrence of dead or dying cells adhering to the culture dish is very rare. It is possible to perform both the Kenacid Blue and the Neutral Red assays on the same culture (see Table l), i.e., Neutral Red estimates can be obtained and, since the cells are by then fixed, protein determination can be made using the Kenacid Blue method. Performing both assays would provide a means of checking the sensitivity of the Neutral Red assay, when a chemical is suspected of affecting lysosomes. This test, along with several other in vitro systems, is presently undergoing validation as an alternative test to replace the Draize Rabbit Eye Test in a national interlaboratory study started in June 1988, by the Federal Health Office (BGA) of the Federal Republic of Germany (FRG) (1,2).

180

Liebsch

and Spielmann

Table 1 Scheme of Neutral Red and Kenacid Blue Tests Time (h)

Procedure

0.0

Seed 96-well plates with 1 x lo4 cells in 100 pL growth medium and incubate at 37.5”C for 24 h. Remove medium. Add 100 pL test solution or medmm (zero control) with 5% NBCS and incubate at 37.5”C for 24 h. Detection of HTD value. Remove test solution Wash once wrth 100 p,L PBS. Add 100 pL Neutral Red stock solution and incubate at 37.5”C for 3 h Discard NR medium. Wash once with 100 pL PBS. Centrifuge plate for 5 min at 600g. Add 150 j.tL fixative (ethanol/acetic acid solution). Shake plate for 10 mm. Detection of NR absorbance at 540 nm, i.e., cell viability. Discard fixative. Wash once with 100 p,L fixative. Add 100 l,tL Kenacid Blue solution. Shake plate for 10 min. Remove surplus Kenacid Blue by washing twice with 150 l,tL Kenacid Blue washing solution. Shake plate for 10 min. Discard Kenacid Blue washing solution. Centrifuge plate for 5 min at 600g. Add 150 l,tL measuring solution. Shake plate for 10 min. Detection of Kenacid Blue at 570 nm, i.e., cellular protein Calculation of results. It& values

24.0 48.0 51.0 51.5 52.0 52.5 53.0 53.5

2. Materials 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

Balb/c 3T3 cells. Incubator: 37.5OC, humidified, 7.5% CO,/air. Clean bench. Water bath, 37.5”C. Phase contrast microscope. Laboratory burner. Centrifuge with microtiter rotor. Laboratory balance. Osmometer. Immuno reader. Shaker for mtcrotiter plates. Cell counter or hemocytometer. Pipeting aid. Pipets, g-channel-pipets, dilution block. Cryotubes. 80 cm* tissue culture flasks or Petri dishes, 96-well tissue culture microtiter plates,

Balblc

3T3 Cytotoxicity

Test

181

18. Permeable plastic film, Greiner, No. 676001. 19. Computer wrth evaluation program “Symphony.” 20. 1X DMEM fluid medium without r.-glutamine (buffered with sodium bicarbonate) supplemented with (Final concentrations of the compounds, m medium, are quoted.): a. For routine culture: 10% newborn calf serum (NBCS), 4 n&! glutamine, 100 IU penicillin/100 l.tg/mL streptomycin. b. For treatment: 5% NBCS, 4 mM glutamine, 100 IU penicillin/100 p,g/ mL streptomycin. c. For freezing: Growth medium wtth 20% NBCS and 7-10% dimethylsulfoxide (DMSO). Complete medium should be kept at 4°C and stored for no longer than 2 wk. 21. Neutral red stock solution: Neutral red dye (0.4 g), 100 mL phosphate buffered saline (PBS) with Ca2+and Mg 2+.Make up prior to use and store for up to 2 mo. 22. Neutral red (NR) medium: 1 mL of Neutral Red stock solution, 79 mL assay medium, Make up prior to use. The NR medium should be incubated overnight at 37°C and centrifuged at 6OOgfor 10 min (to remove NR crystals) before adding to cells. 23. Ethanol/acetic actd solution (neutral red desorb): 1% Glacial acetic acid solution, 50% ethanol, 49% double distilled water. Prepare immediately prior to use. Do not store for longer than 1 h. 24. Stock Kenacid Blue stock solution: Kenacid Blue stain (0.4 g), 250 mL glacial acetic acid, 630 mL double distilled water. Make up prior to use. 25. Kenacid Blue solution: Kenacid Blue stock solution (88 mL), 12 mL glacial acetic acid. Prepare immediately prior to use. 26. Kenacid Blue washing solution: Ethanol (lo%), 5% glacial acetic acid, 85% double distilled water. Prepare immediately prior to use. 27. Kenacrd Blue desorbing solution: Potassium acetate (1M) (98.15 g), 700 mL ethanol, 300 mL double distilled water. 28. Trypsin/EDTA solution: Make up a 0.05% trypsin/0.02% EDTA m a salt solution. 29. Test compounds: Dissolve in sterile treatment medium, ethanol, or dimethyl sulfoxide (DMSO), as appropriate, at loo-fold the desired final concentration in the case of solvents. The final solvent concentration should be kept at a constant level of 1% in culture medium. Volatile chemicals should be tested in mtcrotiter plates sealed by a plastic film permeable to C02. Insoluble and viscous substances are very difficult to test. In such cases,the supernatant of a saturated solution is used as the highest concentration m the test. All solutions, glassware, and so on, are sterile and all

Liebsch

182

and Spielmann

procedures should be carried out under aseptic condmons and in the sterile environment of a laminar flow cabinet (biological hazard standard) (see Notes 1 and 2).

3. Methods 3.1. Cell Maintenance and Culture Procedures Balb/c 3T3 cells are routinely grown as a monolayer in 80 cm2 tissue culture grade flasks, at 375°C in a humidified atmosphere of 7.5% C02. The cells should be examined on a daily basis under a phase contrast microscope, and any changes in morphology or their adhesive properties noted (see Note 3). 3.1.1.

Routine

Cuhre

of Bulb/c

3T3

Cells

When the cells approach confluence they should be removed from the flask by trypsinization: 1. Decant the medmm and rinse the cultures with PBS-without Ca*+ and Mg2+, e.g., 5 mL is used for a 25 cm2 flask. 2. Wash the cells by gentle agitation to remove any remaining serum that might otherwise inhibit the action of the trypsin. 3. Discard the washing solution. 4. Add l-2 mL trypsm-EDTA solution to the monolayer. 5. Incubate for -1 min, at 375°C. 6. Remove excess trypsin/EDTA solution and incubate the cells at 37.5”C. 7 After 2-3 min, lightly tap the flask to detach the cells into a single cell suspension. 3.1.2. Cell Counting 1. After detaching the cells, add 0.1-0.2 mL of routme culture medium/cm2 flask, i.e., 2.5 mL for a 25 cm* flask. Disperse the monolayer by gentle trituration. It is important to obtain a smgle cell suspension for exact counting. 2 Count a sample of the cell suspension obtained using a hemocytometer or cell counter. 3.1.3. Subculture After determination of cell number, the culture can be subcultured into another flask or seeded into a 96-well microtiter plate. Balb/c 3T3 cells are routinely passaged at a cell density of 1 x lo5 cells/ml in 80 cm2 flasks every 3-4 d (average doubling time, 20-24 h). 3.1.4.

Freezing

Stocks of Balb/c 3T3 cells can be stored in sterile, freezing tubes in liquid nitrogen. Dimethylsulfoxide is used as a cryoprotective agent.

Balblc

3T3 Cytotoxicity

Test

183

1. Centrifuge trypsinized cells at 200g. 2. Suspend the cells in routine culture medium, containing 20% NBCS, at a concentration of l-5 x lo6 cells/ml. 3. Aliquot 120-180 l.tL of cooled DMSO into freezing tubes and fill to 1.8 mL with the cell suspension. 4. Place the tubes mto a freezer at -80°C for 24 h. This gives a freezing rate of 1°C/min. 5. Place the frozen tubes mto liquid nitrogen for storage. 1, 2. 3. 4. 5. 6.

3.1.5. Thawing Thaw the cells by putting the ampules into a water bath at 37SOC. Leave for as brief a time as possible. Resuspend the cells and transfer into routine culture medium. Incubate at 37.5OC m a humidified 7.5% CO2 atmosphere. When the cells have attached to the bottom of the flask (this may take up to -4 h), decant the supernatant and replace with fresh medium. Culture as described above. Passage two to three times before using the cells in a cytotoxicity test. A fresh batch of frozen cells should be thawed out approx every 2 mo.

3.2. Determination 1. 2. 3. 4.

of Cytotoxicity

3.2.1. Preparation of Cell Cultures After growing up the cells, prepare a cell suspension of 10 x lo4 cells/ml in routine culture medium. Using a multichannel pipet, dispense medium only into the peripheral wells of a 96-well tissue culture microtiter plate (see Note 4). In the remaining wells, dispense 100 pL vol of cell suspension (in routine culture medium) at a cell concentration of 1 x lo4 cells/well. Incubate the cells for 24 h in a humidified atmosphere with 7.5% CO2 at 37.5”C, until they form a half confluent monolayer. This incubation period allows for cell recovery and adherence.

3.2.2. Treatment 1. Make up a range of doses of test substance by diluting with treatment medium. The first run for each chemical should have concentrations covering a large range, e.g., O.Ol-lo/100 mg/mL. In subsequent runs, the concentration range should be narrower, e.g., 0.5-5 mg/mL. The final concentrations are reached with a constant dilution factor (e.g., 10%). It is important that cells are in an exponential growth phase when they are taken for the cytotoxictty assay.Eight concentrations, with six replicates, should be run for each chemical on two separate occasions, excluding the initial range-finder.

184

Liebsch and Spielmann

2. The serum concentration of treatment medium is reduced to 5%, since the toxicity of the test substance may be masked by serum proteins. Serum NBCS cannot be totally excluded because cell growth is markedly reduced in its absence. 3. After 24 h incubation, aspirate the medium from the cells. Add 100 yL of treatment medium containing the appropriate concentration of test chemical. Incubate at 37S”C for 24 h. 4. After treating the cells for 24 h, examme the cells under a phase contrast microscope. Record any changes in growth behavior or morphology of the cells owing to the cytotoxic effect of the test chemical. Estimate and record, microscopically, the Highest Tolerated Dose (HTD), i.e., the test substance concentration that causesthe minimum morphological defects to the cells. 3.3. Determination of Cell Growth Inhibition 3.3.1. Neutral Red Measurement This method is based on that of Ellen Borenfreund (3). The uptake of the vital dye Neutral Red into the lysosomes/endosomes and vacuoles of living cells is used as a quantitative indication of cell number and viability. 1. At the end of the 24 h treatment period, wash the cells with 100 pL prewarmed PBS-with Ca*+ and Mg*+. 2. Remove the washing solution by gentle tapping. 3. Add 100 PL NR medium and incubate at 37.5OC, in a hurmdlfied atmosphere of 7.5% COz for 3 h. 4. After 3 h mcubation, remove the NR medium, and wash the cells with 100 PL PBS-with Ca*+ and Mg*+. 5. Add exactly 150 p,L NR desorb (ethanol/acetic acid) solution. 6. Shake the microtiter plate rapidly on a microtiter plate shaker for 10 min until NR has been extracted from the cells and forms a homogeneous solution. 7. Measure the absorbance of the resulting colored solution at 540 nm (reference filter 380 nm) on a microtiter plate reader, using the blank as a reference (see Note 5). 3.3.2. Kenacid Blue Measurement This method is based on that of Knox et al. (4). The measurement of total cell protein provides a quantitative indication of cell number in a culture (see Notes 6-9). 1, After assessingNeutral Red uptake, remove the NR desorb solution. 2. Wash the cells once with 100 pL NR desorb (ethanol/acetic acid) solution.

Bulb/c

3T3 Cytotoxicity

Test

185

3. Add 100 FL KB solution. Make up on day of use. 4. Shake plate for 10 mm. 5. Remove unbound stain by washing twice with 150 pL Kenacid Blue washing solution. 6. Shake the plates for a further 10 mm. 7. Centrifuge the plates at 600g for 5 min to remove all the remaining washing solution, 8. Add 150 PL Kenacid Blue desorb solution and rapidly shake the plates for 10 min, until KB has been extracted from the cells and formed a homogeneous solution. 9. Measure the absorbance of the resulting colored solution at 570 nm (reference filter 380 nm) on a microtiter plate reader, using the blank as a reference. 3.4. Calculation of Results An estimation of the number of viable cells (determined by the Neutral Red Uptake method) or total cell protein (measured using the Kenacid Blue method) is made on each culture dish as outlined above. The results obtained under test conditions are compared with the appropriate control and converted to a percentage value. The 8 concentrations of each compound tested should span the range of no effect up to 95100% inhibition of cell growth. A Hill function (percentage inhibition being a function of the concentration) is fitted to the results of the eight concentrations using the method of least squares. Hill functions are sigmoidal in shape and represent a good model for many dose-response curves. The average optical densities obtained with untreated 3T3-cells are 0.6 for NR and 1.O for KB (5).

4. Notes 1. Volatile chemicals tend to evaporate under the conditions of the test, thus the IC,, value may be variable, especially when the toxicity of the compound is fairly low. This has been overcome to some extent by adapting the procedure for use in 96-well plates since the smaller surface area of the well in these dishes reduces the extent of evaporation. In addition, plates can be sealed with CO2 permeable plastic film that is impermeable to volatile chemicals, thus decreasing evaporation. 2. Other chemicals difficult to test mclude those that are unstable or explosive in water. 3. Other difficulties are related to the nature of the cell hne, i.e., rapidly growmg, nondifferentiating cells of very low metabolic activity, hence raising problems of direct extrapolation of results to the in vivo situation.

186

Liebsch

and Spielmann

4. Several laboratorles have shown that the peripheral wells of 96-well plates do not sustain cell growth at the same rate as the inner wells. As a result, many laboratories use the peripheral wells as blanks only (containing medium). 5 One maJor drawback of the assay 1s the precipitation of the Neutral Red dye mto readily visible, fine, needle-like crystals. When this occurs it 1s almost impossible to reverse, thus producing inaccurate readings. To avoid this precipitation, NR medium 1s incubated overnight and centrifuged before bemg added to the cells. Additionally, some chemicals induce this preclpltatlon, therefore, a visual mspectlon stage m the procedure 1s very important. 6. One of the drawbacks of this assay 1s that the Kenacld Blue dye may, on occasion, precipitate out. The hkehhood of this occurrmg increases as the length of handling time increases, therefore 96-well plates should be agltated regularly and inspected visually for uneven blue color. The process is, however, readily reversed by agitation, so any odd reading should be retested after trlturation to obviate the posslblllty of precipitation. 7. Another problem that may occur 1s the deposition of a ring of dried protein around the walls of the well, at the alr/medmm interface. This arises if the culture medium is not properly removed or through excessive evaporation. Such precipitated protein will give an inaccurate assessment of total cellular protein. 8. It should be noted that m certain cases care should be taken m the mterpretatlon of results. In the Kenacld Blue assay, cells treated with organic acids or alcohols become fixed to the bottom of the plate-including dead cells. As a result, although the protein content determined decreases dosedependence at lower concentrations, It returns to higher values at higher concentrations of test chemical. 9. It 1s important that cells are in an exponential growth phase when they are taken for the cytotoxlcity assay.

References 1, Spielmann, H , Gerner, I , Kalweit, S , and Besoke, R. (1990) Vahdatlon proJect of alternatives to the Dralze eye test in West Germany: first results (Abstract). Naunyn Schmiedebergs Arch. Pharmacol 341(Suppl.), R23 2 Spielmann, H., Gerner, I , Kalwelt, S , Ewe, S , Lausen, A , and Besoke, R. (1989) Zum Draize Test am Kaninchenauge Erste Ergebnisse des BMFT ForschungsproJekt zur Valldlerung von Alternatlvmethoden, in Wege zur Bewertung und Anerkennung von Alternativmethoden zum Tierversuch (Bulling, H E and Splelmann, R., eds.), MMV, Medlzinverlag, Bass, Munich 3. Borenfreund, E. and Puerner, J. A (1985) Toxicity determined in vitro by morphological alterations and Neutral Red absorption. Toxic01 Lett. 24, 119-124

Bulb/c

3T3 Cytotoxicity

Test

187

4. Knox, P , Uphill, P. F , Fry, J. R., Benford, D J., and Balls, M. (1986) The FRAME multicentre proJect on m vrtro cytotoxicology. Fd. Chem. Tax. 24,457-463. 5. Spielman, H , Gerner, I , Kalweit, S., Moog, R., Wirnsberger, T , Krauser, K., et al. (1991): Interlaboratory assessment of alternatives to the Draize eye irrrtation test in Germany. Toxic01 In Vitro 5,539-542 6. Borenfreund, E. and Borrero, 0. (1984) In vitro cytotoxicity assays potential alternatives to the Drarze ocular irrltancy test Cell Biol. Toxic01 1,55-65 7. Kalweit, S., Besoke, R , Gerner, I., and Spielmann, H. (1990) A national validation proJect of alternatrve methods to the Draize rabbrt eye test Toxzcol In Vitro 4, 702-706 8 Kalwelt, S., Gerner, I., and Spielmann, H (1987) Vahdatron proJect of alternatives for the Dralze Eye Test Mol. Tox~ol 1,589-596

CHAPTER21

The Pollen Udo K&ten

Tube Growth and

Rolf

Test

Kappler

1. Introduction In monitoring possible cytotoxic effects of bioactive chemicals, it is desirable to have easy and sensitive test systems. The in vitro culture of pollen can provide a sensitive indication of toxicity at the cellular level, since germination and growth of pollen tubes are inhibited in the presence of toxic substances. Quantification of pollen tube growth will allow this inhibitory effect to be expressed as a numerical value: EDS0(or IC&), i.e., th& concentration of a test compound that reduces pollen tube growth to 50% of control. In the past two decades,pollen grains and pollen tubes of various plant species have been used to determine the cytotoxic effects of environmental pollutants. These determinations include pesticides and adjuvants (1,2), acid rain and inorganic chemicals (3-51, inorganic and triethyl lead (6), and many other compounds (7,8). In all these studies, the pollen tube growth inhibition was microscopically estimated by measuring the lengths of several hundred tubes. This method is time consuming because the tubes growing in a culture medium are usually very bent, making measurement more difficult. The pollen tube length can be measured faster using a method in which the pollen grains are applied to an agar plate in a straight line (9). This method gives relatively good estimates of the inhibitory effect of watersoluble compounds, but provides no accurate evaluations. The pollen tube growth (PTG) test is designed to determine toxicity at the cellular level, although pollen tubes do not represent single cells but From* Methods m Molecular Biology, Vol. 43: In Wro Toxmfy TesOng Protocols Edlted by: S. O’Hare and C K Atterw~ll Copynght Humana Press Inc , Totowa, NJ

189

190

Kristen

and Kappler

are tricellular organisms for the sole purpose of transporting two sperm cells directly to the plant ovule. It must further be emphasized, especially when assessing the environmental impact of pollutants by the PTG test, that pollen tubes represent a different physiological system from that of a root, a leaf, or a whole plant. The suitability of the in vitro growing pollen tube for testing a wide range of substance classes is based primarily on the wide range of organelles present in pollen and the resultant diversity of metabolic processes as well as the extremely high growth rate of the tube. The pollen tube has no chloroplasts, thus it is incapable of photosynthesis;consequently, in some ways it resembles an animal more than a plant organ and is therefore also suitable as a model for the toxicological assessmentof compounds harmful to animals and humans. Because of the lack of chloroplasts, the PTG test will not identify toxic effects that target the photosynthetic apparatus. This may be used to positive advantage to test whether herbicides developed to interfere with the electron transport chains of plant photosystems are harmless to nonphotosynthetic organisms. The PTG test is fast, easy to handle, inexpensive, reproducible, and requires little equipment. Moreover, it has been demonstrated by rank correlations (Fig. 1 shows an example) that ED5a values of the PTG test correlate well with the data of in vitro toxicity assays (10). This is especially true in regard to the Draize eye irritation test to which the PTG test has proved to be a suitable alternative (II). One disadvantage of the PTG test is the need for DMSO to dissolve test substances of low water solubility in the culture medium. DMSO was shown to have no effect on pollen germination and tube growth when present in the culture medium at a concentration not exceeding 1%. However, it cannot be fully excluded that the initial solvent changes the chemical constitution of the test compound, thus influencing its inhibitory effect. The PTG test is currently being automated. Automation will reduce the duration of a parallel test run of 100 samples by 50%. The PTG test used as a routine assayin cooperation of one of the authors (U. Kristen) with the British American Tobacco Company, B.A.T. Cigarettenfabriken GmbH, Hamburg, Germany, and with Beiersdorf AG (Cosmetics), Hamburg, Germany. Through the latter cooperation, the PTG test is successfully mvolved in the CTFA Evaluation of Alternatives Program (USA) (12). One of the authors (U. Kristen) also participates with the PTG test in another international test program, the

191

Pollen Tube Growth Test ED50 (ppm) PTG-Test 1.0

1

10

100

1000

10000

Oinoseb (ONEPI ONOC C- Oimtrophenol CTAB 4 - Ni trophenol Benzolc

acid

Furfural Methylphenylketon Glyphosate Phenol Toluol Aceton DMSO 5

50

500

5000

LO 50 (mg / kg) Fig. 1. Rank correlation of the test parametersof PTG test (black bars) and the rat LDsO(oral) assay(striped bars) considering chemicals of various substanceclasses. Multicenter Evaluation of In vitro Cytotoxicity (MEIC project) of the Scandinavian Society of Cell Toxicology (13,14). Hitherto, around 250 compounds have been examined with the PTG test, including detergents, pesticides, gases, and organic gas mixtures, airborne organic pollutants, and chemicals with data on human systemic toxicity (15). The EDs0 values of the most interesting compounds of these classes have recently been published (16). The PTG test is based on photometric quantification of the tube mass production in vitro by measuring the turbidity of the pollen tube suspension after ultrasonic treatment (17). The efficiency of this technique can be increased even further if the concentration of a dye that binds chemi-

192

Kristen

and Kappler

tally to the pollen tube wall but dissociates at lower pH is measured instead of the turbidity. Since the main metabolic activity of the growing pollen tube is the production of tube wall material, such a quantification may be achieved by measuring the amount of tube wall material present by determining the binding of Alcian blue, which has an affinity for water-insoluble polysaccharides, a major component of plant cell walls. For the PTG test, the authors prefer the Alcian blue staining method for the photometric quantification of pollen tube mass production (18). It is more accurate and easier to handle than the ultrasonic treatment of the tubes. For some types of test compounds, however, the chemical binding of the dye to the pollen tube walls may be impeded. This is especially true if the test compound itself occupies the binding sites for Alcian blue or if it covers or agglutinates the pollen tubes, thus preventing them from being stained by the dye. The latter property is characteristic for fatty, predominantly water-insoluble substances or mixtures, e.g., creams or milky lotions. In such cases, the photometric quantification of the pollen tube production should be carried out by the turbidity measurement via ultrasonic treatment of the pollen suspension. Volatile substances, gases and gas mixtures can also be examined by the PTG test. To this end, only a simple modification of the test flasks is necessary that renders an intense, long-time contact of the gas to the pollen tube suspension. The PTG test is not suited for water-insoluble compounds which cannot be dissolved or at least emulgated with the help of nontoxic solubilizers or emulgators. In the present chapter, we will describe how the PTG test is used to achieve dose-responsecurves and ED,, data for the toxicological valuation of a wide range of chemicals. 2. Materials

1. Pollen from anthersof tobacco flowers: Tobacco plants of the species Nicotiana sylvestris Speg. and Comes, cultivated from seedsin a greenhouseandset out to flower in the field, areusedasa pollen source(Fig. 2). The anthersof flowers that are not fully open are collected, spreadin a Petri dish, and kept at 20-25°C in a dry place until they have opened.This shouldnot takelonger than 2 d in orderto avoid bacterial andfungal infection. The pollen is separatedfrom the dry anthersby shaking, and moreover, by using a sieve and a small brush. The pollen is then frozen in Eppendorf tubes (1.5 rnL) at -18 to -25°C. It can be maintained in this

Pollen Tube Growth Test

2.

3.

4. 5. 6. 7.

8. 9. 10. 11. 12. 13. 14.

193

state for at least 3 yr with practically no loss of its ability to germinate and can thus be used in the PTG test at times other than the flowering period (July-October) of the tobacco plants. Pollen culture medium: Double-distilled water containing 10% sucrose (w/v), 0.01% boric acid (w/v), 3 rnII4 Ca(NOs) *, 10 mMMES (2-[N-morpholinol-ethane-sulfonic acid), analytical grade (SERVA Chemical Company, Heidelberg, Germany), adjusted to pH 5.6 with KOH. The culture medium can be frozen for storage. Ethanolic Alcian blue stock solution: 0.5% solution made by dissolving 05 g Alcian blue 8GX (dye content 57%; Sigma, Deisenhofen, Germany) m 100 mL absolute ethanol. The stock solution is stable for some months in the dark. Aqueous Alcian blue stock solution: A 0.05% aqueous solution prepared by diluting the ethanolic stock solution with double-distilled water. Prepare on day of use, since the dye tends to precipitate out on standing. 40% citric acid (w/v). Dimethyl-sulfoxide (DMSO, analytical grade; Sigma). Stock solutions of the test compounds: Dissolve water-soluble compounds in water to yield a stock solution 2X the required final test concentration, Make up nonwater-soluble compounds in DMSO at 100X the required final test concentration. Dilute the resultant solution 1 in 50 with water to yield a stock solution 2X the required final test concentration. 50 mL screw-capped flasks. The screw cap must have a silicon septum when gaseous substancesare to be injected. Stirrer. Sonicator, e.g., B12 Sonifier Cell Disruptor (Branson Sonic Power Company, Danburg, CT). The sonicator is not necessary if only the Alcian blue method is used. Photometer, e.g., Beckman Photometer Model 34). Centrifuge capable of 1OOOg. Incubator. Personal computer (optional for comfortable storage and calculation of the results. For on-line use with the photometer, a special adaptor-card and software may be necessary).

3. Method 1. Preparation of the pollen suspension: Prepare a pollen suspension at a concentration of 5 mg/l mL culture medium in the following manner: a. Place the pollen into a test tube and add a trace of culture medium. b. Mix with a glass rod until a homogenous pulp is obtained. c. Dilute the pulp with the rest of the culture medium.

294

Kristen

and Kappler

d. Place 100 pL aliquots of the pollen suspension mto 50 mL screwcapped flasks. The suspension should be constantly stirred for homogenization while the ahquots are being taken. 2. Exposure to test compounds: Set up four rephcate flasks for each control and two for each test condition: Aqueous control. Add 100 FL water. Solvent control: Add 100 pL 2% DMSO m water. Zero-time control: Add 100 PL 20% ethanol. Test condition (aqueous): Add 100 PL test compound stock solution. Test condition (gaseous): Add 100 PL water and, using a syringe, inject various quantities of gas through the septum of the flask. Note: The solublhzer, DMSO, IS used at a final concentration of 1%, which has no influence on pollen germination or pollen tube growth. Ethanol 1s used at a final concentration of lo%, which 1s sufficient to reduce pollen tube growth to zero. Incubate the flasks for 18 h at 27OC in the dark 3 Range-finding experiment: Usually, a range-finding experiment should be performed first, followed by a more precise determination of the ED,, values. Six different final concentrations of the test compound between 10% and 1 ppm should be employed m order to estimate the critical range of tube growth inhibition including the ED,, value. This may be detected with the naked eye by checking the size of the pellet of the test suspension against that of the control after centrlfugatlon or by microscopic checking of the pollen tube length (compare Figs. 3A and B). 4. Determination of pollen tube production: A narrower range of concentrations, chosen as a result of the range-finding experiment, is used to determme the exact point of the EDS,-, value. In practice, the test compound concentrations increasing by a factor of two have proved to be very useful. Each test series should be run on at least five separate occasions m order to obtain at least five series of dose-response data. a. Alcian blue method* At the end of the 18 h incubation, transfer the contents of each flask separately mto centrifuge tubes. By rinsing the flasks twice with 4 mL portions of double-distilled water and adding the wash water to their respective centrifuge tubes, it 1s assured that no pollen material remains m the flasks. Centrifuge at 1OOOg for 1 mm and asptrate off the supernatant. Resuspend the pellet m 2 mL 0.05% Alclan blue for 30 mm and add 7 mL of double-distilled water. Centrifuge again at 1OOOg for 1 mm and aspirate off the supernatant Check the staining intensity of the test pellet against that of the control. Insufficient staining or over-stammg of the test pellet indicate an influence of the test compound on the cell wall-dye interaction. This

Pollen

Tube Growth

Test

Fig. 2, Flowers of the tobacco speciesNicotiana sylvestris Speg. and Comes. For the PTG test, pollen must be taken before anthesis, i.e., from flowers that are not fully open (suited developmental stages are marked by arrows). Fig. 3. Microscopic feature of in vitro grown pollen tubes at the end of an 18 h incubation. (A) Tube growth in the aqueous control medium. (B) Tube growth inhibited by test compound. may lead to an over- or underestimation of the test compound toxicity. There are two possible ways to overcome this problem. First, perform at least five washings of the pellets after the first centrifugation in order to remove residues of the test compound that may have attached to the pollen tubes. If this is not successful, use the turbidity method as described below instead of the Alcian blue method. If the pellet is well stained, resuspend it in 9 mL double-distilled water. Repeat the centrifugation and wash again. After the second wash, resuspend each pellet in 2 mL 40% citric acid for 10 min to redissolve the bound dye. Thereafter, centrifuge at 1OOOgfor 1 min and determine the extinction of the supernatant at 607 nm with the photometer. b. Turbidity method: At the end of 18 h incubation, transfer the contents of each flask separately into centrifuge tubes. By rinsing the flasks twice with 4 mL portions of double-distilled water and adding the wash water

Kristen and Kappler

196

Acetylsalicylic

acid

ED 50 = 23.63 +- 2.20 mg/L Inhibition

(“/a 1

Repetitions 6

1

10

100

0 1000

Cont. Img/ll

Fig. 4. PTG test dose-response curve (drawn with the aid of HARVARD GRAPHICS computer program) of acetylsalicylic acid with human systemic toxicity. Vertical bars = standard deviation, filled circles = mean values, squares = number of repetitions. to their respective centrifuge tube, it is assured that no pollen material remains in the flasks. Sonicate for 30 s on setting no. 4 in order to break the pollen tubes into short pieces. Centrifuge for 1 min at 1OOOgand aspirate off the supematant. Resuspend the pellet in 3 mL double-distilled water, whirl up, and measure turbidity at 500 nm without delay. Because the readings at the photometer are unstable it is necessary to integrate over the first 30 s. Alternatively, take five readings in 30 s and calculate the mean value. 5. Calculation of the results: a. Generation of dose-response curve: Subtract extmction values of zerotime control from those of test and control experiments. Use corrected

Pollen Tube Growth Test

197

Thallium sulfate ED 50 = 71.N +- 10.85 mg / I Inhtblfion

I

1

(%I

I I111111

Repetitions

I

10

I I111111

I

I I111111

100

1000

I

-6

I I1111

10000

Cont. Img/ll

Fig. 5. PTG test dose-response curve (drawn with the aid of HARVARD GRAPHICS computer program) of thallium sulfate with human systemic toxicity. Vertical bars = standard deviation, filled circles = mean values, squares = number of repetitions. values to calculate percent growth inhibition, A plot of these percentage values vs the logarithmic scale of test concentration will produce a dose-response curve. Graphic presentations (Figs. 4 and 5) can be obtained with the aid of commercial graphic programs. b. Calculation of EDsO:From at least five dose-response curves, the ED,, for each curve is estimated by interpolation from the data of the two concentrations confining the EDso. The mean and standard deviation may then be calculated.

References 1. Gentile, A. G., Richman, S M., andEaton,A. T. (1973) Corn pollen germination and tube elongation inhibited or reducedby commercial andexperimentalformulations of pesticidesand adjuvants.Envir. Entomol. 2,473-476.

198

Kristen

and Kappler

2. Gentile, A. G , Vaughan, A W , and Pfeiffer, D. G. (1978) Cucumber pollen germination and tube elongatron inhibited or reduced by pesticides and adjuvants. Envy. Entomol. 7,689-69 1. 3. Masarau, N., Katsuhisa, F., Sankichi, T., and Yugata, W (1980) Effects of inorganic components in acid ram on tube elongation of Camellia pollen Envlr Pollut (Ser A) 21,51-57

4 Cox, R. M. (1988) The sensitivity of pollen from various coniferous and broadleafed trees to combinanons of acidity and trace metals. New Phytol. 109, 193-201. 5. MeJnartowicz, L. and Lewandowski, A. (1985) Effects of fluorides and sulfur dioxide on pollen germination and growth of the pollen tube. Actu Sot. Bot Pol 54,125-129.

6 Roderer, G. and Reiss, H D (1988) Different effects of inorganic and triethyllead on growth and ultrastructure of lily pollen tubes. Protoplasma 144, 101-109. 7. Pfahler, P. (1981) In vitro germination of maize pollen to detect biological activity of environmental pollutants Envir. Health Perspect 37, 125-132. 8. Wolters, J. H. B. and Martens, M. J. M. (1987) Effects of air pollutants on pollen Bot. Rev. 53,372-414

9 Martin, F W. (1972) In vitro measurement of pollen tube growth inhibition.

Plant

Physiol. 49,924-925.

10. Strube, K , Janke, D., Kappler, R., and Knsten, U. (1991) Toxicity of some herbicides to ln vitro growing tobacco pollen tubes (the pollen test) Envir. Exp Bot 31,217-222 11 Kristen, U , Hoppe, U , and Pape, W (1993) The pollen tube growth test: a new alternative to the Draize eye irritation assay J Sot. Cosmet. Chem 44, 153-162. 12 Gettings, S D , Dipasquale, L. C., Bagley, D. M., Chudkowski, M., Demetrulias, J L., Feder, P. I , Hintze, K. L , Marenus, K D., Pape, W , Roddy, M., Schnetzinger, R , Silber, P , Teal, J. J., and Weise, S L (1990) The CTFA evaluation of alternatives program: an evaluation of zn vitro alternatives to the Drarze primary eye irritation test (phase I) hydro-alcohohc formulations; a preliminary communmation. In Vitro Toxicol. 3,293-302

13. Bondesson, I., Ekwall, B , Hellberg, S , Romert, L , Stenberg, K., and Walum, E. (1989) MEIC-A new international multicenter project to evaluate the relevance to human toxicity of m vitro cytotoxicity tests. Cell Blol Toxic01 5,331-347 14. Ekwall, B., Barile, F , BJarregaard, H., Chesne, C., Clothier, R , Dierickx, P., et al. (1991) The first 10 MEIC chemicals tested m 70 cytotoxicity assays-the influence of cell type and toxicity criteria on cytotoxicity IXth Scandinavian Workshop on In Vitro Toxicology. Nagu, Finland 15 Barile, F. A , Dierickx, P J., and Kristen, U (1994) In vitro cytotoxicity testing for prediction of acute human toxicity. Cell Biol Toxic01 10, 155-162. 16. Kristen, U., Joos, U., van Aken, J P., and Kappler, R (1992) Bestimmung der Zytotoxizitat von Detergentien, Pestiziden und anderen Chemikahen mit dem Pollentest (Pollen Tube Growth Test) VDI-Berichte 901, 1191-1210 17 Kappler, R. and Kristen, U (1987) Photometric quantification of in vitro pollen tube growth: a new method suited to determine the cytotoxicrty of various envrronmental substances Envir. Exp Bot. 27,305-309. 18. Kappler, R and Kristen, U (1988) Photometrrc quantification of water-insoluble polysacchartdes produced by in vitro grown pollen tubes. Envir. Exp Bot. 8,33-36

CHAPTER22 HET-CAM Horst

Test

Spielmann

1. Introduction The potential irritancy of compounds may be detected by observing adverse changes that occur in the chorionallantoic membrane of the egg after exposure to test chemicals (1). Chemicals are placed directly onto the chorionallantoic membrane of the hen’s egg. The occurrenceof vascular injury or coagulation in response to a compound is the basis for employing this technique as an indication of the potential of a chemical to damage mucous membranes (in particular the eye) in vivo. Hen’s eggs are rotated in an incubator for 9 d, after which time any defective eggs are discarded. The shell around the air cell is removed and the inner membranes are extracted to reveal the chorionallantoic membrane. Test chemicals are added to the membrane and left in contact for 5 min. The membrane is examined for vascular damage and the time taken for injury to occur is recorded. Irritancy is scored according to the severity and speed at which damage occurs. The test has several advantages, including its simplicity, rapidity, sensitivity, ease of performance, and relative cheapness. A factor to consider is the fertility and the ability of the eggs to hatch. The survival of chickens is dependent on a complex interrelationship of ecological factors (e.g., the genetic background and the age of the mated birds, the nutritional status and general management of the flock, and in part, seasonal variations). Eggs should, therefore, be obtained from reliable local contractors. (The authors have produced some empirical data From Methods In Molecular Blotogy, Edited by S O’Hare and C K Atterwlll

Vol. 43 In Vitro Toxmty Testing Protocols Copyrrght Humana Press Inc , Totowa, NJ

199

200

Spielmann

on the fertility of the particular flocks they use. The fertility of middleaged flocks is approx 90% with lO-15% defective eggs. On average, there are 20% lesions produced during preparation.) The major disadvantage of the procedure is the subjective nature of the evaluation of the results. This is overcome to a certain extent by the inclusion of positive standards and by using a comprehensive scheme for scoring the irritant effects of the chemicals. The exposure period to the test chemical of 5 min has been found to be sufficient to reveal irritant/toxic effects (longer exposure does not appear to yield any additional information). A factor for consideration is whether the hen’s egg test may be considered as an animal experiment. At present the test is often considered borderline, although it has potential to be used in a manner likely to reduce the number of mammals used in conventional testing and also to contribute toward a reduction in the associated suffering. This test, along with two cytotoxicity assays underwent validation as an alternative test to replace the Draize Rabbit Eye Test, in a national interlaboratory study started in June 1988, by the Federal Health Office (BGA) of the Federal Republic of Germany (FRG) (2-4). Preliminary findings indicate that data from the HET-CAM test appears to correlate better than the two cytotoxicity tests when compared with in vivo Draize scores. The cytotoxicity tests give a greater number of false positives and negatives compared to the HET-CAM test. The HET-CAM test will provide reproducible results if carried out under routine conditions with well trained operators. 1.

2. 3, 4. 5. 6.

2. Materials Animals: Whrte Leghorn chicken eggs (Shaver Starcross 288A). The White Leghorn chicken has been selected for several reasons; hatching of the eggs of this breed is very consistent and reproducible, and there do not appear to be any hereditary defects m this breed. Incubator with an automatic rotating device: Optimum temperature 375°C (&0.5”C), relative humidity 62.5% (k 7.5%). Candling light. Dentist’s rotating sawblade. Computer with appropriate software (HET-CAM evaluation program)not commercially available (authors will give assistance to interested scientists). Cold-light lamp.

HET-CAM 7. 8. 9. 10. 11. 12. 13. 14. 15.

201

Test

pH-meter. Thermometers. Tapered forceps. Pipets (300 j,rL application). Stopclock. O.lN NaOH (standard 1). 0.9% NaCl solution m distilled water. 1% SDS solution in distilled water (standard 2). Test chemicals: Make up the chemicals m 0.9% NaCl solution or olive oil.

3. Methods 3.1. Incubation of Eggs 1. Select fresh fertile 50-60 g eggs. 2 Candle the eggs and discard any that are defective. 3. Place the eggs flat onto incubator trays in a 37.5”C incubator and rotate for 8 d to prevent the attachment of the embryo to one side of the egg. 4. Check the temperature and humidrty at the same time each day. 5. Candle the eggs on d 9 and discard any nonviable eggs. 6. Replace in the incubator with the large end upward but do not rotate, thus ensuring accessibility to the chorronallantorc membrane. 7. On d 10, prepare the eggs for assaying (see Note 1).

3.2. Assay Preparation 1. Candle each egg to ensure that all are viable. Use a cold lamp to ensure optimal illuminatron of the chorianallantorc membrane. 2. Carry out in a fume cupboard with safety goggles to prevent inhalation and contact with the fine eggshell powder. Mark the air cell using a rotating dentist-sawblade and pare the section of shell off. 3. Carefully moisten the membrane with 0.9% NaCl solution at 37°C. 4. Replace the eggs m the incubator until ready for assaying (maximum of 30 mm between opening the eggs and starting the assay). 5. Freshly prepare standards and test solution (in the appropriate solvents) before each assay at room temperature. Measure and record the pH.

3.3. Assay Procedure 1. Table the opened egg out of the incubator, pour off the NaCl solution, and carefully remove the membrane (without iqurmg any underlying blood vessels) using tapered forceps. 2. Add 0.3 mL of the standard or test chemical to the chorionallantoic membrane (CAM). 3. Observe the reactions on the CAM over a period of 5 min. Monitor the appearance of:

Spielmann

202

a. Hemorrhage (bleeding); b. Vascular lysis (blood vessel disintegration); and c. Coagulation (protein denaturation intra- and extravascular). 4. Record in seconds the time for each reaction to occur and calculate an irritation

score (IS).

IS = [(301- set H)/(300)] .5 + [(301- set L)/(300)] B7 + [(301- set C)/(300)] -9 where H = hemorrhage; L = vessel lysis; C = coagulation; set = start second. When determining the threshold, the degree of severity of each reaction after treatment time has to be recorded according to the following scheme: 0 = No reaction 1 = Slight reaction 2 = Moderate reaction 3 = Severe reaction The threshold is then defined as the highest concentration at which slight reactions occur. To determine the threshold apply 0.3 mL of the starting concentration to each of three eggs. A good choice is 5% if no further information is given. Graduate the severity of the main reaction after 5 min: If the observed reaction is slight, double the concentration. If the reaction is moderate or severe, divide the concentration by two or ten to get the next test concentration. Proceed further until the threshold concentration is found. 3.4. Test Scheme For a given chemical, the procedure consists of four steps: 1. Determine the irritation score (IS) for the two standardswith two eggs each, The1%SDSshouldgiveanISof10+2andthe0.1NNaOHanISof15f3. 2. Determine the threshold concentration of the testchemical asdescribed above. 3. Determine the IS for a 10% solution for three eggs. For insoluble substances take the supernatant of a saturated solution, 4. Determine the IS for the pure substance (100%). If the test chemical is an insoluble solid substance, proceed a follows: Instead of determining the IS, put some grains of the substance onto the CAM to cover approximately half of its surface. After 5 mm, carefully rinse off the test material with NaCl solution and record the severity of each of the three reactions (hemorrhage, lysis, coagulation) according to:

HET-CAM

203

Test

Threshold (TH concentration) TH
Table 1 Calculations andClassifications Irritation score Severity (10%) >I6 16

Classification Severeicorr. Severe/corr.

Severereaction after 1 min

cl6 <16

Severe reaction after 5 min Weak or no reaction

>16 <16
Severereaction

Severe/corr. Irritant Irritant Irritant Moderate Moderate Moderate No/slight

0 = No reaction 1 = Slight reaction 2 = Moderate reaction 3 = Severe reaction If any reaction of degree 3 is observed, repeat the procedure with three new eggs, rinsing after 1 min. At the end of the assay kill the embryos as quickly as possible (e.g., by placing the eggs into a freezer at -20°C).

3.5. Calculations

and Classifications

Calculate the mean value of the IS for the three eggs for each of the two runs and both concentrations as well as the mean over both runs of the IS and threshold concentration. A classification of the irritating potential can be carried out according to the preliminary classification scheme shown in Table 1.

4. Note 1. Avoid any shakmg, unnecessary tilting, knocking, and all other mechanical irritation of the eggs when preparing them for the assay.

References 1. Ldpke, N. P. (1986) HET (hen’s eggtest)in toxicological research,in Skin Models. Models to Study Function and Disease of Skin (Marks, R. and Plewig, G., eds.), Springer-Verlag, pp. 282-29 1,

204

Spielmann

2. Kalweit, S., Besolke, R., Gerner, I., and Spielmann, H (1990) A national validation project of alternative methods to the Dratze rabbit eye test. Toxic. In Vitro 4,702-706 3. Spielmann, H., Gerner, I., Kalwelt, S., Moog, R., Wirnsberger, T , Krauser, K , et al. (199 1) Interlaboratory assessment of alternatives to the Dratze eye irritation test in Germany. Toxic In Vitro 5, 539-542 4. Spielmann, H., Kalwert, S , Liebsch, M., Wirnsberger, T , Gerner, I., BertramNeis, E., et al. (1993) Validation study of alternatives to the Drarze eye irritation test in Germany: cytotoxicity testing and HET-CAM test with 136 industrial chemrcals. Toxicol. In Vitro 7,505-5 10.

CHAPTER23

The Use of the Bovine Isolated Cornea as a Possible In Vitro Test for Ocular Irritancy Ann M. Northover 1. Introduction Over the last few years several groups of workers have endeavored to devise a replacement for the Draize eye test (1) in order to reduce the numbers of experiments carried out in vivo in fulfillment of safety requirements for medicinal preparations, toiletries, and cosmetics that might, either by intention or by accident, come into contact with human eyes. The method described below uses slaughterhouse material, namely bovine isolated corneas, thus obviating the necessity to use laboratory animals. The test is based on the premise that corneas opacify when exposed to deleterious substances. Preliminary experiments comparing the results obtained for the same compounds using both the Draize eye test and the bovine isolated cornea test were encouraging (2) (see Note 1).

2. Materials 1. Bovine eyes collected from a slaughterhouse. It IS not necessary to refrigerate the eyes, but they should be used within 4 h of removal. 2. Dissecting instruments. 3. Physiological salt solution (PSS), e.g., Tyrode’s solution: 137 mA4 NaCI, 2.68 rnA4 KCl, 1.8 mA4CaCl,, 0.87 n&J MgQ, 0.68 rnA4 NaH2P04, 11.9 mkl NaHC03, 5.56 mA4 glucose. 4. Incubator at 32°C. This temperature was considered to be a suitable compromise between the body temperature of 37°C and the atmospherrc temFrom Methods III Molecular Biology, Edlted by S O’Hare and C K Atterwlll

Vol. 43. In VI&O Joxmty Jestmg frotoco/s Copynght Humana Press Inc , Totowa, NJ

205

Northover

206

A

4rnm 9

nl’

I i cm I

I

II II II II

I

II II II II II .

B

2

t

cm

Int

dlam

‘11 mm

. 3 mm-

I Perspex

Fig. 1. Diagrammatic representations of a cornea1 holder. (A) End-on view showing positions of screws and mlet for fluid. (B) Side view showmg two identical haIves with cornea fixed in the middle. Measurements are for the bovine cornea. perature to which the cornea would be exposed in life. It IS not necessary to humidify the incubator. 5. Perspex cornea1 holder as shown in Figs. 1 and 2, checked for possible leakage of fluid from joints. The dimensions given are appropnate for bovine isolated corneas. The cornea1 holder shown in Fig. 2 was designed after the publication of the initial work (2). However, a similar model,

Test for Ocular

A Possible

207

Irritancy 3 mm Perspex

I

A

4 I

3.9

cm I i +-¤

5.5cm--)

tTT

B 1

2.5 cm

4.0 cm

1 I 1.3 ‘cm 0.3’ cm

1

A Cornea 4%

0.9

cm

3

lmm I mm-Perspex

Fig. 2. Diagrammatic representation of modified cornea1 holder showing curved inner surfaces designed to accept the endothelial and epithelial surfaces of the cornea. This design is intended to remove the problem of wrinkling of the cornea, which sometimes interferes with the measurement of opacity when using the cornea1 holder depicted in Fig. 1. (A) End-on internal view. (B) Sideon view. modified to accommodate porcine isolated corneas, has been used by other workers (3). 6. Muir Opacitometer (4) as shown in Fig. 3. This consists of a matt black, light-tight box (with lid) into which the comeal holder fits, so that white light from a suitable source passes through the cornea, impinges on a photocell, and is transformed into a voltage that can be recorded on a voltameter. The dark current should be small, e.g., co.005 V when the lamp is off.

Northover

208

,

\

2 Volt AC power SUPPlY

5 cm -pathlengthLight

Fig. 3. Diagrammatic representation of the Muir opacitometer. The cornea1 holder, light source, slit, and photocell are enclosed within a light-tight black box. 7. Solvents: Some compounds may have to be dissolved m a solvent other than water. Concentrated solutions should be prepared and diluted as required with PSS to ensure maximum dilution of the solvent. Any solvent, at all concentrations used, should first be tested for any deleterious effects on the cornea. Care should be taken to avoid solvents that might affect the jomts of the Perspex chambers.

3. Methods 1. Prepare the cornea as shown m Fig. 4. First cut 4-5 mm external to the limbus. Then cut the iris out from its insertion Pull away the lens using forceps, and finally wash the cornea m PSS (see Notes 2 and 3). 2. Calibrate the opacitometer by usmg a holder contammg PSS but no cornea. The recorded voltage should be adjusted to a suitable baseline reading, e.g., 2.5 V (4) or 5.0 V (3). 3. Fix a cornea m the Perspex holder and expose both surfaces to 5 mL PSS. Immediately place the holder m the opacitometer, epithehal surface nearest the light source, and record the voltage (Vi) (see Note 4). 4. Remove the holder from the opacitometer and replace the PSS on the eptthelial side with either fresh PSS or PSS contaming a known concentration of test compound. Incubate the cornea1 holder plus cornea at 32°C for the

A Possible

Test for Ocular

209

Irritancy Vitreous

CC

Fig. 4. Diagrammatic representation of a bovine eye showing where cuts are made m order to remove the cornea. chosen time. Four to SIX hours IS recommended. Incubatton for <4 h may not produce measurable opacity (4), and for >6 h may give rtse to bacterial growth that mtght cause opacity (see Note 5). 5. After incubation replace the PSS or PSS plus test compound on both sides of the cornea with fresh PSS at 32°C. Return the holder to the opacttometer and record the voltage (V,). 6. A fall in voltage is indicative of thi development of opacity in the cornea (4). Thus % opacity = VI - Vz x 100 V, when VI - V2 = 0 the opacity is zero; when VI - V, = 2.5 the opacity 1s100%. 4. Notes 1. A good correlatton was found between the results of the bovine isolated cornea test and the Dratze eye test for the followmg compounds (2): Sodmm decyl sulfate, sodium lauryl sulfate, triethanolamine decyl sulfate, triethanolamme lauryl sulfate, triethanolamme myristyl sulfate, cetyl trimethylammonium bromide, lauryl trimethylammonium bromide, myristyl

210

2.

3. 4. 5.

Northover trimethylammonium bromide, ally1 alcohol, droxane, ethanol, carbitol, and propylene glycol. The cornea1 preparation can be used to investigate the effects of test compounds on intact corneas as described above, or on corneas from which the epithelium and/or the endothelium plus Descemet’s membrane has been removed. Test compounds can be applied to the epithelial surface and/or to the endothelial surface. Likewise, the effects of pH or osmolarity can be tested. The cornea1 holders can be used to enable corneas to be incubated with varrous test solutions to determine possible effects on cornea1 thickness. This can be measured using a micrometer screw gage ($6). The cornea1 holders could also be used to determine the passage of test compounds across the cornea, by placing the test solutron on the epithelial surface and, after incubation, assaying the solution on the endothelial surface for the presence of the test compound.

References 1. Draize, J H., Woodward, G., and Calvery, H. 0. (1944) Methods for the study of irritation and toxicity of substances applied topically to the skin and mucous membranes J. Pharmacol. 82,377-390. 2. Muir, C. K. (1985) Opacity of bovme cornea in vitro induced by surfactants and industrial chemicals compared with ocular imtancy in vivo. Toxicol. L&t. 24,157-162. 3. Igarashi, I., Katsuta, Y , Matsuno, H., Nakazato, Y., and Kawasaki, T. (1989) Carbachol HCl-induced opacity of porcine isolated cornea. ATLA 16,322-330. 4. Muir, C. K. (1984) A simple method to assesssurfactant-induced bovine cornea1 opacity in vitro: preliminary findings. Toxicol. Lett. 22, 199-203. 5. Muir, C. K. (1985) The effects of surfactants and hypotonicity on bovine cornea in vitro: comparison between opacity and thickness. ATLA 12, 137-144 6. Igarashi, H. and Northover, A. M. (1987) Increases in opacity and thickness induced by surfactantsand other chemicals in the bovme isolated cornea. Toxicol. Lett 39, 249-254.

CHAPTER24

Cytotoxicity in an AnchorageIndependent Fibroblast Cell Line Measured by a Combination of Fluorescent Dyes Richard

B. Kemp

1. Introduction In the search for valid alternatives to animals in assessing the acute toxicity of substances foreign to humans, the use of cells in vitro is a sound proposition. The validity of any in vitro model is limited, however, because the physiological, immunological, and inflammatory conditions in vivo cannot be reproduced in vitro (1). In fact, these elements also mean that difficulties occur when extrapolating results from whole animals to events in humans. The simplest and most elementary endpoint is cell death, which has been defined as the irreversible cessation of cellular activity and function coupled with irreparable disorganization of structure (2). The initial phases of injury are represented by fragmentation of the plasma membrane and changes in mitochondria and endoplasmic reticulum (3,4). Damage to the cell membrane leads to release of intracellular enzymes (4), for instance lactate dehydrogenase (5), the occurrence of which often has been used as an indicator of cellular damage. It has been speculated, however, that over 100 enzymes may be essential for cell life, Thus, monitoring the absence or presence in culture medium of one of these is unlikely to provide an unequivocal indication of the cell’s vital function. From Methods In Molecular Biology, Edited by S O’Hare and C K Atterwill

Vol, 43: In Vitro Toxioty Testing Protocols Copyright

211

Humana

Press

Inc , Totowa,

NJ

212

Kemp

Methods of determining cell death include dye exclusion (6), changes in cell morphology (7), radioactive chromium release (a), and tritiated thymidine incorporation (9), each of which present problems for the researcher. The first of these techniques introduces intraobserver variability in visually assessing whether or not the dye has been taken up by a cell. The second suffers from difficulties of interobserver subjectivity, whereas the other techniques often involve complex, expensive, or time consuming assays, without great advantage. In advocating the use of a combination of fluorescent dyes in determining cell death (I, 10) the needs for relatively low cost, high degree of accuracy, strict reliability, simplicity, and potential for automation were borne in mind. The two complementary dyes are fluorescein diacetate (FDA) and ethidium bromide (EB). Both have been used as viability stains for many years. FDA is a nonpolar compound that readily diffuses into the cell when incorporated into the culture medium. Intracellular esterases hydrolyze the dye to produce fluorescein (II), a negatively charged and, therefore, highly polar molecule that only very slowly diffuses from intact cells. It fluoresces green under ultraviolet excitation. Rotman and Papermaster (II) postulated that influx of FDA was much faster than extrusion of fluorescein, resulting in its accumulation in cells with intact membranes-healthy cells. It was shown that Saponin-treated cells, as well as those unable to form clones, that is, to divide, were nonfluorochromatic because they had plasma membranes that were physically punctured and thus unable to retain the dye. Such cells were deemed to be dead. It is interesting to note that erythrocytes and cells in primary culture originating from certain tissues always failed to exhibit fluorochromasia. A useful property of the reaction producing fluorescein from FDA is a broad pH maximum, pH 6.7-8.0 (II). Fluorescein liberation only allows the number of viable cells to be determined, but of course does not provide an indication of nonviable cells because they cannot be seen under ultraviolet light. For this purpose, the cells are also exposed to a red fluorescent compound, EB. This binds covalently to nucleic acids and, in particular, intercalates with DNA (12), making it a powerful frame-shift mutagen. Thus, it stains nuclear material, especially chromatin and nucleoli, and is not present on nuclear and plasma membranes (13). It is taken up by both living and dead cells but its rate of penetration into the latter is, of course, faster, a fact that has been exploited in a spectrofluorimetric assay for cell viability using EB

Cytotoxicity

Test

213

alone (14). In combination with FDA (1.5), however, viable cells also contain fluorescein, the fluorescence of which partially masks the red color. Quantitation of nonviable cells is achieved by the selective use of filters. The choice of cell type is, to a certain extent, arbitrary, once it is conceded that this test is a first-order one (16); that is, designed to rank the acute lethality of pure chemicals and chemical formulations to living matter. Cells are carefully chosen for specific physiological and biochemical properties to form the basis for second-order tests. It has been suggested that primary cultures of human cells would provide ideal systems for assessing cytotoxicity, if results are to be extrapolated to the situation in humans. For instance, primary cultures of human eye tissue should be used when potential ocular irritancy is the important parameter under consideration. Quite clearly, the widespread use of such a tissue source is not practicable. Primary culture of cells from selected tissues of other embryonic and adult animals might seem the best substitute because it could be reasoned that such cells would retain their characteristic physiological properties, This proposal has been criticized, however, because the necessary enzyme dissociation and other procedures can result in variability and friability without any scientific certainty of the validity of the model that the properties of animal tissues closely represent those of human tissues. Since cells in primary culture are likely to differ from those in vivo, and obtaining them requires considerable use of animals, consideration should be given to utilizing established cells lines (17), especially for first-order tests. Although such cells do randomly lose certain physiological properties during the course of establishment, they subsequently remain constant, having been adapted to specific culture conditions. The choice then lies between anchorage-dependent (monolayer) and anchorage-independent (suspension) types of cell. The former generally requires enzyme treatment for release from the substratum and a growth period of time before harvesting at confluency. The latter can be grown as a continuous culture with daily “milking” (generation time 22 h) for experimentation. It is recommended that an anchorage-independent, suspension-adapted cell line be employed in first-order studies of acute lethality. The particular cell type, that is, LS-L929 mouse fibroblasts originally cloned from areolar and adipose tissue, was selected for the ease with which it is maintained in continuous suspension culture.

214 Many cytotoxicity tests require 48-72 h of culture before determination of endpoint. This test was consciously designed for completion within the working day, to give a rapid turnover of data. The active ingredient of many commercial products is contained in a complex carrier system, e.g., lotion, cream, tablet, and so on, at least some of which is insoluble and/or immiscible in aqueous culture medium. These excipients cause technical difficulties to the assay, visually interfering with endpoint estimations. This has necessitated modifications to the procedure involving extraction of water-soluble components of the formulation into culture medium before the 4-h incubation. Extraction can be carried out in two ways (18), by a variable volume technique or by a bulk method. Mostly, the former is preferred, but both rely on mobilization of polar compounds into the culture medium. In summary, mouse fibroblasts are maintained in continuous suspension, diluted when required, and known volumes incubated for 4 h in the presence of test material in a range of concentrations for 4 h. Aliquots are then removed and added to an equal volume of physiological solution containing FDA and EB. The cells are then examined under epifluorescent illumination, An image analyzer, together with the correct filter systems, enables differentiation between viable and nonviable cells. The endpoint is taken as 50% cell death (CD,,) and this is used to rank the potential acute toxicity of test substances. 1. 2. 3. 4. 5. 6. 7.

2. Materials The cell line: LS cells derived from NCTC L929 fibroblasts (Flow Laboratories, McLean, VA). The culture vessel is a 1 dm3 round siliconized flask. Agitation is by the Bernoulli effect (19) using punctured plates attached to a stirring rod that 1spowered by a Vibromixer (Chemap, Mannedorf, Switzerland). 5 cm3 Erlenmeyer flasks (slhconized). Orbital shaker, 90 rev/min, 37OC. 5channel micropipet. 60-well Terasaki plate. Inverted fluorescence mrcroscope with eprfluorescence source, suitable range of filters, and motorized

mechanical

stage for plates (e.g., Leitz).

8. Mrcroscope image analyzer with television camera (e.g., Lertz or Analytical Measuring

System [UK]).

9. IBM-compatible microcomputer as “bolt-on” to image analyzer, with a surtable statistical package including linear regression and probit analyses(20).

Cytotoxicity

Test

215

10. Eagle’s modified Minimal Essential Medium for suspension cells (MEMS, Flow Laboratones) supplemented with 5% (v/v) fetal calf serum, 0.85 g! dm3 sodium bicarbonate, 2 mM L-glutamine, 100 U/mL penicillin, and 100 pg/mL streptomycin. Medium gassed with 5% (v/v) carbon dioxrde in air should have a pH of 7.4 at 37°C. 11. Physrological solution: Eagle’s modified Minimum Essential Medium, Dulbecco A and B, or 0.9% (w/v) NaCl. 12. Internal standard: Sodium lauryl sulfate (Analar Grade). Prepare a stock solution of 6 mg/mL which should give a CDS, value of 60 p,g/mL. If the value deviates from this by more than lo%, the run should be discarded. 13. Fluorescent dyes: The stock solution of fluorescein diacetate is 5 mg/mL in acetone. This can be stored at -2OOC.Dilute 1:lOO in ethidium bromide solution, 300 pg/mL physiological solution. 14. Test materials: Stock solution is 100 mg/mL in MEMS or other appropnate solvent (e.g., dimethylsulfoxide, ethanol, methanol).

3. Methods 3.1. Exposure to Test Compound 1. Maintain suspension cultures of LS-L929 mouse fibroblasts in round-bottomed siliconized flasks (a vibrating stirrer rod penetrating the flask bung via a glassfunnel and rubber flange) at 37°C in MEMS gassedwith 5% (v/v) COZ in air. 2. Exposure to soluble test materials: When the cell culture stock reaches a concentration of 0.8-l .2 x lo6 cells/cm3 (22 h generation time), transfer 1 mL aliquots into 5 mL siliconized flasks. Add an aliquot of test compound (or internal standard) solution, the volume not exceeding 60 uL to produce the required concentration. 3. Exposure to immiscible/insoluble test materials: Extract the immiscible/ insoluble compounds into culture medium using one of the following methods. 4. Gas all the flasks (containing both the soluble and extracted insoluble compounds) with 95% sir/5% CO;! and incubate m an orbital shaker, 90 rev/ min, at 37OCfor 4 h. The gas is introduced to the flask from a cylinder by a tube ending m a Pasteur pipet. After gassing, the flasks are sealed with silicone bungs.

3.2. Variable

Volume

Technique

1. Place the test material in 25 mL Erlenmeyer flasks. Add culture medium to each flask to provide the concentrations required. Gas the flasks with 5% (v/v) CO2 in air and seal with a silicone bung. Incubate the flasks at 37°C on an orbital shaker, 90 rev/min for 24 h.

216

Kemp

2. After the incubation period, centrifuge the material at 4300g for 30 mm at 18OCto separate the two phases. Attach a Teflon@ tube to a syringe and insert the tube through the upper oily layer without disturbing the mterface. Withdraw the lower aqueous layer for testmg. 3. For both extractions, count the cells m the cell culture stock and remove sufficient medium to give a total lo6 cells per concentration of extract to be tested. Centrifuge at 800g for 5 min at 18°C. Discard the growth medium, resuspend the cells in 1 mL aliquots of both the concentrated and diluted extract, and add to 5 mL Erlenmeyer flasks. 1.

2.

3.

4. 5.

6.

7.

3.3. Bulk Method Place the test material in Erlenmeyer flasks. Add culture medium to each flask to produce the concentrations required. Ensure that the concentration of test substance is at least double that of the approximate CD,, value obtained from a standard run. Gas the flasks with 5% (v/v) CO, in air and incubate at 37°C on an orbital shaker at 90 rev/min for 24 h. After the incubation period centrifuge the material at 43OOgfor 30 min at 18°C to separate the two phases. Attach a thin Teflon tube to a syringe and insert the tube through the upper oily layer without disturbing the interface. Wlthdraw the lower aqueous layer for testing. For both extractions, count the cells in the cell culture stock and remove sufficient medium to give a total lo6 cells per concentration of extract to be tested. Centrifuge at 800g for 5 mm at 18OC. Discard the growth medium, resuspend the cells m 1 mL aliquots of both the concentrated and diluted extract, and add to 5 mL Erlenmeyer flasks. After the incubation period, remove 0.5 mL aliquots of the cell suspension and mix with an equal volume of physiological solution contaming the fluorescent dyes. Automated image analysis of fluorescent cells from controls, test cultures, and internal standard: Using a six-channel micropipet, add 5 p.L of untreated cell suspension to each of 6 wells in a 60-well Terasaki plate. Repeat the procedure for each of the cell suspensions exposed to sodium lauryl sulfate and test compounds. Place the plate on the automated stage of the inverted microscope. Examine the cells under epifluorescent illumination using a 10x objective, the resultant image being split between the binocular microscope head and the monitor of an Image Analyzer. Count the number of green fluorescent cells using a blue filter, and red fluorescent cells using a green filter. The stage moves automatically to the next sample and the procedure 1srepeated for all 60 wells of each plate.

Cytotoxicity

Test

217

Image focusmg is performed manually or with the aid of an autofocus device. 8. Store the data on a microcomputer. 9. Each substance should be tested m a minimum of 4 separate experiments, but up to 10 runs may be necessary if there is a large amount of statistical variation. 10. CD,, figures (i.e., concentration of test material required to damage 50% of the cell population) should be computer-generated from linear regression plots. 11. Results: Carry out a linear regression analysis on the data usmg the method of least squares. Transform the data to the probit-log dose relationship for cumulative frequencies (21-23). 12. To differentiate between products of similar cytotoxic potential, the regressional analyses from the slopes can be statistically compared for evidence of colinearity and coincidence on the ordinate. Axiomatically, lines that are both colmear and coincident are stattstically equivalent. Owing to the inherent biological variability of the test procedure, the latter rigorous statistical treatment is not always possible (see Note 1).

4. Note 1. A biochemical mdication of toxicity has been incorporated within the procedure that increases the sensitivity, provides more accurate figures, and allows automation ATP measurement was chosen because ATP is the prime energy donor m the cell and is thus an ideal indicator of cellular health (10,24). This method detects effects at lower irritant concentrations compared to the viability dye techmque and, coupled with the greater mtrinsic sensitivity of the luciferase bioluminescence system, provides a more efficient test procedure. Statistically reproducible effects can be obtained from as little as lo4 cells (10 pL) leading to savings m culture costs and an increase in data output. Spectroscopic machines are now fully automated and can be interfaced with microcomputers, reducmg labor costs and thus compensating for the greater cost of reagents.

References 1 Kemp, R B., Meredith, R W. J., Gamble, S., and Frost, M (1983) A rapid cell culture technique for assessing the toxrcity of detergent-based products m vitro as a possible screen for eye imtancy in uivo Cytobios 36, 153-159. 2 Dixon, K. C. (1967) Events in dying cells. Proc. Roy Sot. Med 60,271-275 3 Mahnin, T I and Perry, P. (1967) A revrew of tissue and organ viability assays. Cryobiology 4, 104-l 15. 4. Trump, B F., Valigorsky, J., and Dees, J. (1973) The modernisation of the autopsy applications of ultrastructural and biochemical methods to human disease. Med. Co11 VA Q. 9,323-333

218

Kemp

5. Cornelis, M., DuPont, C , and Wepierre, J. (1991) In vitro cytotoxicity tests on cultured human skm fibroblasts to predict the irritation potential of surfactants ATLA 19,324-336.

6. Kemp, R. B , Jones, B. M., Cunningham, I., and James, M C. M (1967) Quantttatlve investigation on the effect of puromycm on the aggregation of trypsin- and versene-dissociated chick fibroblast cells J. Cell Sci 2,323-340. 7. Reinhardt, C. A , Pelh, D. A., and Zbinden, G. (1985) Interpretation of cell toxicity data for the estimation of potential irritation. Fed Chem. Toxicol. 23,247-252. 8. Parish, W. E. (1985) Relevance of in vitro tests to in vivo acute skin mflammation* potential zn vitro applications of skin keratome slices, neutrophils, tibroblasts, mast cells and macrophages. Fed Chem. Toxic01 23,275-288. 9 Baserga, R (1989) Cell Growth and Division, IRL, Oxford, pp. 39-41. 10 Kemp, R. B., Cross, D. M , and Meredith, R. W J (1988) Comparison of cell death and adenosine triphosphate as indicators of acute toxicrty In vitro Xenobiotica 18, 633-639.

11. Rotman, B. and Papermaster, B W. (1966) Membrane properties of living mammalian cells as studied by enzymatic hydrolysis of fluorogenic esters Proc. Nat1 Acad. Sci. USA 55,134-141. 12 Gabby, E. J. and Wilson, W D. (1978) Intercalating agents as probes of chromatin structures Meth Cell. Biol. 18,351-384. 13. Burns, V. W. (1972) Localization and molecular charactensttcs of fluorescent complexes of ethidmm bromide in the cell Exp. Cell Res. 75,200-206 14 Edidin, M. (1970) A rapid quantitative fluorescence assay for cell damage by cytotoxic antibodies J Immunol. 104, 1303-1306 15. Takasugi, M. (1971) An improved fluorochromatic cytotoxic test Transplantation 12,148-151.

16 Balls, M. and Horner, S. A. (1985) The FRAME interlaboratory programme on in vitro cytotoxicity. Fed. Chem. Toxicol. 23,209-213 17 Freshney, R. I. (1994) Culture of Animal Cells and Manual of Basic Techniques, 3rd ed., Liss, New York, 492 p. 18. Kemp, R. B., Meredith, R. W. J., and Gamble, S. H. (1985) Toxictty of commercial products on cells in suspension culture. a possible screen for the Draize Eye Irritation Test. Fed. Chem. Toxicol. 23,267-270. 19. Ulrich, K. and Moore, G. E (1965) A vibrating mixer for agitation of suspension cultures of mammalian cells. Biotech. Bioeng. 7,417. 20. Krewski, D. and Franklm, S., eds. (1991) Statistics in Toxicology, Gordon and Breach, New York. 21. Bliss, C. I. (1937) The calculation of the time-mortality curve. Ann. Appl. Biol. 24, 815-852. 22. Litchfield, J. T., Jr. (1949) A method of time-per cent effect curves J. Pharm. Exp. Ther. 97,399-406

23 Litchfield, J. T., Jr. and Wilcoxon, F (1949) A simplified method of evaluating dose-effect curves. J. Pharm Exp. Ther. 96,99-l 15 24 Kemp, R B., Cross, D. M , and Meredith, R W. J. (1986) Adenosine triphosphate as an indicator of cellular toxicity in vitro. Fed. Chem. Tox~ol. 24,465-466.

CHAPTER25 Cell Culture

Phototoxicity

Test

Paul A. Duff) 1. Introduction Human A431 cells and mouse 3T3 cells are exposed in culture to UV light both in the presence and absenceof test compound. Phototoxicity is expressed as a decrease in cell viability as determined by the 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) assay. Phototoxicity can be classified as an abnormal cutaneous response that occurs as a result of exposure to ultraviolet (UV) light or visible radiation and is associated with the presence of xenobiotic (which may be present locally or systemically in the body). This test is designed to mimic such a response by subjecting cell cultures to increasing periods of exposure to UV illumination in the absence and presence of test compounds (at concentrations previously demonstrated to have no cytotoxic effect). Subsequent, postexposure, cell growth is determined using the MTT assay (I) (the basis for which is outlined below). Increased exposure to UV light leads to an inhibition of cell growth and a decrease in cell viability. If this detrimental effect is significantly potentiated in the presence of a test chemical it may be considered a likely phototoxic agent. 1.1. Use

of the

MTT

Assay

The tetrazolium salt, MTT, is taken up into cells and reduced by a mitochondrial dehydrogenase enzyme to yield a purple formazan product that is largely impermeable to cell membranes, thus resulting in its accumulation within healthy cells. Solubilization of the cells results in the liberation of the product that can readily be detected using a simple calorimetric assay. The intracellular reduction of MTT is thus indicative From. Methods m Molecular B!ology, Edrted by S O’Hare and C K Atterwlll

Vol 43 In Wtro Toxmty Testmg Protocols Copyright Humana Press Inc , Totowa, NJ

219

220

DUffY

of the normal functional biochemistry of energy-requiring mitochondrral enzyme reactions and, more broadly, provides a means to estimate the number of viable cells. 1.2. Cytotoxicity

Assay

Monolayers of human A431 epidermal cells or mouse 3T3 cells are exposed to increasing concentrations of test chemical for 24 h, after which time the cell number/viability is determined by the calorimetric MTT assay. The highest concentration of each chemical at which no reduction in cell viability is observed (compared to the control situation) is recorded as being the “no effect” concentration. 1.3. Phototoxicity

Assay

Monolayers are incubated for 4 h, then exposed to UVA or a combination of UVA and UVB light for increasing periods of time, both in the absence and presence (at the “no effect” concentration) of test chemical. The cultures are incubated for a further 20 h, after which viability is determined by the MTT assay. The formation of formazan product is plotted against the time of exposure to UV and the ID,, value, i.e., that dose of UV that reduces cell viability to 50% of control levels, calculated from the curves. The IDS0 values in the absence and presence of test compounds are then compared. A compound is considered of phototoxic potential when the IDS0value in the presence of test compound is significantly less than that which occurs in its absence. This method provides a rapid, relatively inexpensive, simple to perform method for screening many compounds for potential phototoxicity. The test conditions can be strictly controlled with regard to time period of exposure and wavelength of light. UV light is used because it contains the wavelengths most often associated with phototoxic reactions. There is, however, a very small number of chemicals activated by light from within the visible part of the spectrum. The test avoids a number of problems associated with human testing, such as variability because of skin thickness, pigmentation, and pharmacokinetic and metabolic handling of the compound. These factors must, however, be considered when attempting to extrapolate and interpret results in relation to likely in vivo phototoxicity. The main difference to the in vivo situation is that the test is based on the exposure of test chemicals in contact with only a monolayer of one type of cell in any assay plate.

Cell Culture Phototoxocity

Test

221

The cell lines used reflect the types of cell present at the site of activity, i.e., the skin, Although the phototoxic insult is a biophysical one, and might therefore be expected to produce the same reaction in most types of cell, significant differences have nevertheless been observed in the sensitivities of A43 1 epidermal cells and 3T3 fibroblasts. This could be a result of cell lineage specific biochemical characteristics. The use of two cell types reduces the chances of missing a specific phototoxin. A further consideration is the lack of a metabolic activation component. However, it appears from the literature that hepatic metabolism is not greatly involved in the production of molecules that are subsequently activated by UV light to phototoxic products. The porphyria type of photoinduced dermal reactions are one exception. If, however, hepatic metabolism was considered to be of potential significance in relation to a given test compound, the inclusion of a hepatic S9 microsomal metabolic activation system would only require slight modification of the protocol. 1.4. The MTT Assay as an Endpoint

Formation of the formazan product has been found to correlate well with cell viability in terms of normal functioning of mitochondrial energy-requiring biochemical reactions. The assay compares favorably with several other methods used in the determination of cell number/ viability, e.g., dye exclusion. It is rapid, sensitive, relatively simple to perform, and lends itself to semiautomation (I), It should be noted, however, that the MTT assay is not readily adaptable for use with static cell populations or those of low mitochondrial activity. Certain compounds may selectively affect the mitochondria of the cells, resulting in a greatly overestimated level of toxicity. A recent paper (2) draws attention to the possible influence of pH on the MTT reaction. However, this effect should not be of any significance in the assay as used in the Phototoxicity Test. 2. Materials 1. A43 1 Human epidermal carcinoma cell line (3). 2. 3T3 Swiss mousefibroblast cell line (4). 3. Incubator: 37°C humidified atmosphere,5% COJ95% air. 4. Standard80 cm2tissue culture flasks. 5. 96-Well microtiter plates. 6. Oriel 1000 W solar simulator enhancedfor UV output. The exposure should be modified in such a way that each column of wells can be irra-

DUffY

7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

diated separately. Thus, the period of exposure for each column can be carefully controlled, enabling increasing doses to be applied across the plate. One way to do this is to screen the plates by a piece of cord or metal plate containing a slit that would allow one S-well column to be exposed alone at any one time. Alternatively, moving a screen across the plate column by column will result in the achievement of exposure times by a cumulative process. UVA region filter, 320-400 nm wavelength. UVA + B region filter, 280-400 nm wavelength. Oriel enhanced UV radiometer, Model 8 1021. Titertek Multiskan. Earl’s balanced salt solution (Ca2+and Mg2+ free). MlT reagent: 5 mg/mL in PBS. The solution should be protected from light and stored at +4”C for not more than 1 mo. Acidified isopropanol: 4% v/v 1M HCl in isopropanol. Complete culture medium: DMEM (carbonate-buffered and containing phenol red) supplemented with (final concentrations shown): 10% FCS; 2 mfJ4L-glutamine; 100 IU/mL penicillin; 100 pg/mL streptomycin. Treatment culture medium: MEM supplemented with 10% FCS. MTT medium: Phenol red- and supplement-free Modified Eagle’s Medium (MEM) without FCS or supplements. Test compounds: Test compounds should be made up in complete culture medium, treatment medium, or solubilized in dimethylsulfoxide (DMSO) and diluted to required concentration in appropriate medium (max. acceptable DMSO cont. = 10 pL/mL medmm). Compound solubthty will influence the final test concentrations, but the usual range is between 0.1 and 1000 pg/mL. Solutions in complete medium are used for the initial cytotoxicity assay. Solutions in UV treatment medium, which does not contain phenol red, are used for the phototoxicity assay, because phenol red is known to absorb UV light. 3. Methods

3.1. General

Cell Maintenance

of A431 and 3T3 Cell Lines

1. Both 3T3 and A431 are kept as frozen stock at 2-5 x lo6 cells/ml in DMEM (without FCS) containing 10% DMSO. Thaw frozen stocks rapidly at 37°C and then dilute to 10 mL with fresh prewarmed DMEM (all supplements). 2. Centrifuge at 1000 rpm for 10 mm and discard supernatant. 3. Resuspend in appropriate volume of fresh prewarmed complete DMEM and seed into culture flasks containing DMEM (25 mL/80 cm2 flask). The approximate seeding ratio from frozen stock is one ampule to two flasks.

Cell Culture

Phototoxocity

Test

223

4. Grow cells to 90% confluence, replacing medium every 3-4 d as required. 5. Subculture cells at 90% confluence every 3-4 d at a subcultive ratio of 1:3 (or 4) for A431 cells, 1:6 (or 7) for 3T3 cells. 6. Rinse l-2 times with 25 mL of a Ca2+,Mg2+ free buffer (type not critical) per flask. 7. Add 10 mL trypsm solution per flask and leave 2-5 min at room temperature (see Note 1). 8. Confirm cell detachment by microscopic observation of cells rounding up. 9. Decant off approx 8 mL of the trypsin and leave cells for a further 5 min. 10. Dislodge the cells by tapping the flask gently.

3.2. Testing Compounds for Cytotoxic and Phototoxic Effects 1. 2. 3.

4.

3.2.1. Cytotoxicity Assay Suspend cells in DMEM to a concentration of 5 x 104/mL. Plate out 100 p.L cell suspension into each well of a 96-well microtiter plate (i.e., 5 x lo3 cells/well). Incubate plates overnight at 37°C. The next day, discard medium and replace with 100 uL of either complete culture medium (containmg appropriate level of vehicle, where necessary) or complete medium containing test compound over a range of concentrations-8 microwell replicates for each concentration tested. Reincubate the plates at 37OCfor a further 24 h.

3.2.2. MTT Assay 1. At the end of the incubation period remove the medium, rinse cells once with 100 p.L of Ca-, Mg-free Earl’s balanced salt solution, and then add 100 p,L of MTT medium prewarmed to 37°C to each well. 2. Add 10 pL of MTT reagent to each well. Stain cells for 4 h at 37OC.At the end of this period, add 100 pL acidified isopropanol to solubilize the purple formazan crystals produced. Measure the absorbance at 500 nm with a Titertek Multiskan. 3. For each test compound determine the highest “no effect” concentration, i.e., the highest concentration of test compound to which the cells were exposed that did not result in a decrease in absorbance compared to the control situation, 3.2.3. Phototoxicity Assay 1. Suspend cells in DMEM to a concentration of 5 x 104/mL. Plate out 100 pL cell suspension into each well of a 96-well microtiter plate (i.e., 5 x lo3 cells/well). Incubate plates overnight at 37°C. The next day, discard medium and replace with either treatment medium (containing appropriate

MEDIUM ONLY

-)(-w)(e)< TREATED )ooc:

G

qoo(

)oocwuu

CELLS

MEDIUM TEST CO+MPO”ND

Fig. 1. The 96-well “split-plate” dosing format. The upper four rows (A-D) of the 96-well test plate contain cells and media only (control). The lower four rows (E-H) contain cells exposed to the no-effect concentratron of compound. level of vehicle, where necessary) or treatment medium containing the test compound at the previously determined “no effect” concentration (see Note 2). 2. The plates should be set up m the “split plate” format (see Fig. 1). This format results in the first 4 rows of the plate being exposed to the treatment medium alone (i.e., control), whereas the bottom 4 rows are exposed to medium containing the test compound. 3. Incubate the plates for 4 h to allow equilibratron. Four hours has been established as the minimum period required for the compound to enter the cells to a degree sufficient for the demonstration of photoactivity. 4. After the 4 h eqmlibratron period, expose each column (1-12) of the plate to either UVA alone or a combination of UVA + UVB light using an Orrel 1000 W solar simulator with the appropriate filters to enable emission at the following wavelengths: 320-400 nm UVA regron UVA + B region 280-400 nm 5. One column on each plate should not be irradiated to provide a control. During the course of each experiment the output of the solar simulator should be recorded. To accomplish this, place a UV radiometer at set distance from the simulator, the same distance as that of the cultures from the source, and record the output at both wavelengths. The values obtained should be m the range of 20-30 mW cm* (UVA) and 0.9-1.1 mW cm* (UVA and UVB). 6. After exposure to UV, incubate cells for a further 20 h. 7. Repeat the MTT assay as outlined above.

Cell Culture Phototoxocity

Test

225

3.3. Results 1. Plot a graph of mean absorbance of MTT product vs exposure time of UV for both compound treated and nontreated cells. From the resulting curves determine the IDsa value, i.e., that dose of UV that reduces the viability to 50% of the control levels. 2. The IDsa values in the absence and presence of test compound should be compared. Using this system in repeated experiments, it has been established ($6) that the IDsa values for cells not treated with compound produce a coefficient of variation of maximum 25%. These authors therefore suggest that a reduction in IDS0 of more than 30% in compound-treated as compared to untreated cells is of biological significance, i.e., that in the presence of the compound the cells were significantly more suscepttble to the detrimental effect of exposure to UV light. 3. This in vitro phototoxicity test has been validated with 30 compounds. They were categorized as strong, idiosyncratic, and negative phototoxic compounds based on the frequency of reported adverse effects in patients (7). For each of the compounds the occurrence of a brologrcally significant effect, i.e., a reduction in IDSo >30%, was scored as negative or positive for each of the cell lines (A431 or 3T3) and each of the exposure conditions (UVA or UVA + B).

4. Notes 1. A431 cells may require more extensive washing with buffer and/or mcubation with trypsin-EDTA. This must be determined by trial and error. In carrying out the cytotoxicity and phototoxicity assays,the passagenumber of the cells is not critical. 2. Compound solubihty will influence the final test concentrations but the usual range is between 0.1 and 1000 ~g/rnL. Solutions in complete medium are not used for the initial cytotoxicity assay. Solutions in treatment medium, which does not contain phenol red, are used for the cytotoxicity assay, because phenol red is known to absorb UV light.

References 1. Mosmann, T. (1983) Rapid calorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays.J. Immunol. Meth. 65,55-63.

Plumb, J. A., Milroy, R., and Kaye, S. B. (1989) Effects of the pH dependenceof 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide-Formazan absorption on chemosenativity. Cancer Res. 49,4435-4440 3. Giard, D. J., Aranson,S. A., and Todaro, G. J. (1973) In vitro cultivation of human tumours:establishedcell lines derived from a seriesof solid tumours.J. Natl. Cancer Inst. 1, 1417-1427. 2

226

DUffY

4 Todaro, G J. and Green, H. (1963) Qualitative studies of the growth of mouse embryo cells in culture and their development into estabhshed cell lines. J Cell Biol. 17,299-313.

5. Duffy, P A , Bennett, A., Roberts, M., and Flint, 0. P. (1987) Prediction of phototoxic potential using human A431 cells and mouse 3T3 cells. MoZ Toxicol. 1, 579-587

6 Duffy, P A., Bennett, A., Roberts, M , and Flint, 0. P. (1989) The prediction of phototoxrc potential using human A431 cells and mouse 3T3 cells, in In Vitro Toxicology: New Directions, Alternative Methods in Toxicology, vol. 7 (Goldberg, A. M., ed.), Mary Ann Liebert, New York, pp. 327-335. 7. Hawk, J. L. M. (1984) Photosensltizmg agents used in the United Kingdom. Clin Exp. Dermatol.

9,300-302.

CHAPTER26

The Murine Jennifer

Local Lymph Hilton

Node Assay

and Ian Kimber

1. Introduction Allergic conta :t dermatitis is a delayed-type hypersensitivity reaction in the skin characterized by a tissue damaging inflammatory response. A wide variety of natural and synthetic chemicals have been shown to be capable of causing contact dermatitis (I). A number of guinea pig tests are available for evaluating the sensitizing potential of chemicals. All such tests employ a biphasic protocol in which animals are sensitized (induction phase) and subsequently challenged (elicitation phase) with the test chemical. Sensitization potential is determined by the visual assessment of challenge-induced erythematous skin reactions. In 1944, Draize et al. (2) introduced a scoring system to assessthese erythematous reactions, thus developing the first method designed specifically for measurement of sensitizing potential. Modifications to the Draize test have been introduced (3) and a number of new procedures developed, including the occluded patch test of Buehler (4-6) and the Magnusson and Kligman guinea pig maximization test (7-10) that currently have the widest application for predictive assessmentof contact sensitizing activity. The methods used for sensitization of guinea pigs vary considerably. Some tests employ epicutaneous application (open or occluded), others intradermal injection, and some a combination of both. Some of the more sensitive assays use adjuvant to augment responses. Although there is no doubt that such assays play an important role in predicting sensitizing activity, it has been recognized that they have important limitations (11-13). All guinea pig tests require the visual From Methods m Molecular Biology, Vol 43 In Wtro Toxmty Testing Protocols Edlted by S O’Hare and C K Attetw~ll CopyrIght Humana Press Inc , Totowa, NJ

227

228

Hilton

and Kimber

assessment of erythema, an endpoint that is subjective and may lead to interpretive difficulties when colored or irritant chemicals are examined. Assays are also relatively time-consuming and costly to perform and require substantial amounts of test material. It is normal practice to challenge with the maximum subirritant concentration of the chemical. Therefore, with highly irritant materials, the concentration selected for challenge may be too low to elicit a dermal hypersensitivity reaction. It is against such a background that this and other laboratories have speculated that an increased understanding of the mnnunological events that mediate and regulate contact sensitization may provide opportunities to develop new predictive test methods. Major advancesin our understandingof contact sensitivity have derived from studies in the mouse. In 1968 Asherson and Ptak (14) demonstrated that elicitation reactions could be measured quantitatively in mice as a function of challenge-induced increases in ear thickness. Several groups have attempted to develop predictive test methods based on such measurements. Examples include the mouse ear swelling test (MEST) (15,16), the mouse ear sensitization assay (MESA) (I7), and a variant of these in which mice are fed a diet supplemented with vitamin A; a regimen considered to potentiate cell-mediated immune function (18-20). Alternative methods for evaluating challenge reactions in the mouse have included measurement of increases in wet weight of treated ears (21,22), assessment of the cellular infiltrate using radioactive labels (23,24), and optimized lymphocyte blastogenesis assays that have been used to assessboth contact and photocontact allergy (25-27). Of these methods, it is the MEST assay that has been evaluated most thoroughly. Problems regarding the reliability and sensitivity of the assay (24,28) have largely been overcome by the introduction of a vitamin A supplemented diet, which results in enhancedear swelling responses(29). In this laboratory we adopted a different approach. We considered that it might be possible to identify contact allergens on the basis of events occurring during the induction phase, rather than the elicitation phase, of sensitization. The primary immunological response to epicutaneously applied skin sensitizing agents is dependent on the recognition of, and response to, chemical antigens by T lymphocytes (30-32) and is characterized by hyperplasia in the draining lymph nodes (33-35). Lymph node activation is reflected by an increase in node weight, the appearance of large

Murine Lymph Node Assay

229

pyroninophilic cells, and the induction of lymphocyte proliferation. In our preliminary studies, each of these parameters were examined following topical exposure to chemical (36,37). It was concluded that lymphocyte proliferative activity measured by the incorporation of [3H] thymidine (3HTdR) provided the most sensitive and reliable correlate of skin sensitizing potential. It now seems likely that the vigor and duration of lymphocyte proliferation influences directly the extent to which sensitization occurs and consequently the severity of the dermatitic reaction that will develop following challenge (38). These preliminary studies (36,37) defined the optimum conditions required to provoke and measure lymph node cell proliferation, including investigation of the kinetics of the response and strain comparisons. It was found that exposure of mice to skin allergens on the dorsum of the ears causedhyperplasia in draining (auricular) lymph nodes; nodes that could be readily identified and excised. Daily exposure for 3 consecutive days produced a more vigorous response than did a single application of the same total amount of chemical. In comparative investigations, mice of the CBAKa strain were found to exhibit greater responses than did other strains examined (37). These factors were incorporated into the first version of the local lymph node assay, where proliferation was measured in vitro following culture of isolated draining lymph node cells with 3HTdR. The advantage offered by this method is that the proliferative activity of the draining lymph node cell populations prepared from sensitized mice may be enhanced selectively by inclusion in culture of an exogenous source of the T-cell growth factor interleukin 2 (IL-2). Addition of IL-2 has little or no effect on the low levels of background proliferation observed in cultures of lymph node cells prepared from naive or vehicle-treated control mice. This version of the assay was used to assessa wide variety of chemicals for which there exists information regarding sensitizing activity in guinea pigs and/or humans (37). With the objective of simplifying the assay and obviating the requirement for tissue culture, a modified local lymph node assay has been developed in which lymph node cell proliferation is measured in situ following intravenous injection of 3HTdR (39). Since its development, this version of the local lymph node assay has been the subject of internal and interlaboratory validations and of comparisons with existing guinea pig predictive test methods (4045).

230

Hilton

and Kimber

Interlaboratory evaluations have found the in situ local lymph node assay to be robust and to yield equivalent predictions of sensitization potential when performed independently in separatelaboratories (4245). Compared with currently available guinea pig predictive tests, the local lymph node assay offers a number of significant advantages. The assay is rapid and cost effective. Exposure is via the relevant route, there are no requirements for adjuvant, and relatively small amounts of test material are needed. The endpoint is objective, quantitative, and not influenced by the color of the test chemical. The weight of available evidence indicates also that in most casesirritant, nonsensitizing chemicals fail to elicit positive responses. Internal and interlaboratory trials have compared the sensitivity and selectivity of the local lymph node assaywith the most commonly used guinea pig tests,the occluded patch test of Buehler and the guineapig maximization test (4M5). The indications are that the local lymph node assay correlates well with the guinea pig tests and is able to identify accurately those chemicals classified as having moderate or greater sensitizing potential. At present the local lymph node assay identifies contact sensitizers as a function of LNC proliferation at a defined period following exposure. Another option may be to measure relative sensitizing potential as a function of the concentration of chemical required to elicit a positive response. Such an approach has recently been employed successfully in a comparative evaluation of the skin sensitizing potential of biocides (46). 2. Materials 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

Sterile 200 gage stainless steel mesh. Phosphate-buffered saline (PBS). RPMI- 1640 culture medium. HEPES buffer. Ampicillin. Streptamycm. Heat-treated fetal calf serum, 56OCfor 30 min. Trypan blue. 96-well microtiter plates. CO2 incubator. Tritlated thymidine: 2 Wmmol. Scintillation counter. Mice: Young adult (8-16 wk) CBA/Ca stram mice are allowed to acchmatize for at least 2 d. Although either sex may be used, it is recommended

Murine

Lymph

231

Node Assay

that animals of a single sex be used in each experiment. Experimental groups, each comprising four mice, are caged separately. 14. Test concentrations: Three concentrations of the test material, together with an appropriate vehicle control, are evaluated. One approach 1sto select 3 consecutive concentrations from the following: 50, 25, 10, 5, 2.5, 1, 0.5, 0.25, and 0.1% (w/v). Although, in the context of hazard Identification, it may be considered desirable to select the highest test concentrations possible, this is not always practical. Poor solubility and/or concerns regarding acute local or systemic toxicity may dictate a more conservative approach. 15. Vehicle: Many organic vehicles may be used. Water, however, is mappropriate. The question of vehicle selection has been addressed durmg recent interlaboratory evaluations of the local lymph node assay.Experience indicates that, in order of preference, the vehicles of choice are: 4:l acetone:ohve oil (AOO), methylethyl ketone (MEK), drmethylformamide (DMF), propylene glycol (PG), and dimethylsulfoxrde (DMSO). Vehrcle selection is dictated by the relative solubility of the test material. For most purposes A00 is smtable.

3. Methods 3.1. In Situ Local Lymph

Node Assay

3.1.1. Exposure

All mice receive 25 p.L of the test material, or of vehicle alone, on the dorsum of both ears. The solution is delivered using an automatic pipet, with a disposable tip, ensuring even distribution over the surface of the ear. Identical treatment is performed daily for a total of three consecutive days. 3.1.2. In Situ Labeling Five days following the initiation of treatment, all mice receive an intravenous injection of 250 PL of PBS containing 20 pCi of [3H]-methyl thymidine CH’I’dR; specific activity 2 CMnmol). Five hourslater mice aresacrificed. 3.1.3. Preparation

of Lymph Node Cells

Draining (auricular) lymph nodes are excised and pooled for each experimental group in a small volume of PBS. The nodes in PBS are placed onto a 4 cm2 grid of stainless steel gauze (200 mesh) with the sides of the gauze turned up in order to retain the nodes. The gauze is contained within a 60-mm plastic Petri dish. A single cell suspension of lymph node cells (LNC) is prepared by mechanical disaggregation using the flat-ended plunger of a 5 mL syringe. Cell suspensions are trans-

Hilton

232

and Kimber

ferred from the Petri dish into a 10 mL round-bottom plastic tube. LNC are washed twice in PBS by centrifugation at 19Ogfor 10 min. After the final wash the cells are resuspendedin 3 mL of trichloroacetic acid (TCA) and stored overnight at 4°C. Precipitates are recovered by centrifugation, resuspended in 1 mL of TCA, and transferred to 10 mL of scintillation fluid. 3HTdR incorporation is measured by p-scintillation counting. 3.1.4. Analysis

of Results

Results may be recorded either as total disintegrations/min/node for each experimental group, or as a stimulation index using the values derived from vehicle controls as the comparator. It is currently our policy to classify chemicals on the basis of local lymph node assay data as sensitizers or not strong sensitizers. For a chemical to be classified as a sensitizer it must fulfill two criteria: 1. A threefold or greaterelevation in 3HTdR incorporationrelative to vehicle controls must be recordedwith at least one test concentration. 2. The data must not be incompatible with a conventional biological dose response,unless aberrantresults at higher concentrationscan be reconciled with the possibility of local or systemrctoxicity. 3.2. In Vitro Local Lymph Node Assay Groups of mice (n = 4) receive three consecutive daily applications (25 FL) of various concentrations of the test chemical or an equal volume of vehicle alone to the dorsum of both ears. Animals are sacrificed l-3 d following the final exposure. The draining (auricular) lymph nodes are excised and pooled for each experimental group. A single cell suspension of LNC is prepared under aseptic conditions by mechanical disaggregation through sterile 200-mesh stainless steel gauze. Lymphocyte suspensions are washed once in PBS and resuspended in RPMI-1640 culture medium supplemented with 25 mA4 HEPES, 400 pg/rnL ampicillin, 400 p.g/mL streptomycin, and 10% heat inactivated (56°C for 30 min) fetal calf serum (RPMI-FCS). Viable cell counts are performed by exclusion of 0.5% trypan blue and the cell concentration adjusted to working values (7.5 x lo6 cells/ml) in RPMI-FCS. Lymphocyte suspensions are seededinto 96-well microtiter plates at a concentration of 1.2 x lo6 cells/well (five wells per group) and cultured for 24 h at 37°C in a humidified atmosphere of 5% CO, in air with 2 @i 3H-methyl thymidine (specific activity 2 Wmmol). Parallel cultures may be supplemented

Murine Lymph Node Assay with a source of IL-2. Culture is terminated by automatic cell harvesting and 3HTdR incorporation is determined by p-scintillation counting. References 1. Cronm, E. (1980) Contact Dermatztts, Churchill Livmgstone, London. 2. Draize, J. H., Woodward, G., and Calvery, H. 0. (1944) Methods for the study of irritation and toxicity of substances applied topically to the skin and mucous membranes. J. Pharm. Exp. Ther 82,377-390 3 Johnson, A. W. and Goodwin, B. F. J. (1985) The Draize test and modrfications, in Contact Allergy Predtcttve Tests in Gutnea Pigs. Current Problems in Dermatol-

ogy, vol 14 (Andersen, K E. and Marbach, H. I, eds.), Karger, Basel, pp 31-38 4. Buehler, E. V. (1965) Delayed contact hypersensitivity in the guinea pig. Arch. Dermatol.

91,171-177.

5. Buehler, E. V. (1985) A rationale for the selection of occlusion to Induce and elicit delayed contact hypersensitivity in the guinea pig. A prospective test, in Contact Allergy Predictive Tests m Guinea Pigs* Current Problems in Dermatology,

vol.

14 (Andersen, K E and Maibach, H. I., eds ), Karger, Basel, pp 39-58 6. Ritz, H. L. and Beuhler, E V. (1980) Planning, conduct and interpretation of guinea pig sensitization patch tests, in Current Concepts in Cutaneous Toxicity (Drill, V. A and Lazar, P , eds.), Academic, New York, pp 25-40. 7. Magnusson, B. and Kligman, A. M. (1969) The identification of contact allergens by animal assay. The guinea pig maximization test method. J. Invest. Dermatol 52,268-276. 8. Magnusson, B. and Kligman, A. M. (1970) Allergic Guinea Pig, Charles C Thomas, Springfield, IL.

Contact Dermatitis

tn the

9. Wahlberg, J. E. and Boman, A. (1985) Gumea pig maximization

test, m Contact Allergy Predictive Tests in Guinea Pigs: Current Problems in Dermatology, vol.

14 (Andersen, K. E. and Maibach, H. I., eds.), Karger, Basel, pp. 59-106. 10. Maurer, T. and Hess, R. (1989) The maximrzation test for skin sensitization potential-updating the standard protocol and validation of a modified protocol. Fed. Chem. Toxicol. 27,807-g 11. 11. Andersen, K. E. and Maibach, H. I. (1985) Gumea pig sensitization assays An overview, in Contact Allergy Predictive Tests in Guinea Pigs: Current Problems in Dermatology, vol. 14 (Andersen, K E. and Maibach, H. I., eds ), Karger, Basel, pp. 263-290 12. Oliver, G. J. A., Botham, P. A., and Kimber, I (1986) Models for contact sensitization-novel approaches and future developments. Br. J Dermatol. 115,53-62. 13. Kimber, I. (1989) Aspects of the immune response to contact allergens. opportunities for the development and modification of predtctive test methods. Fed. Chem. Toxicol. 27,755-762

14. Asherson, G. L. and Ptak, W (1968) Contact and delayed hypersensittvity in the mouse. I. Active sensitization and passive transfer Immunology 15,405-416, 15. Gad, S. C., Dunn, B. J., Dobbs, D. W , Reilly, C , and Walsh, R. D. (1986) Developmentand validation of an alternative dermal sensitization test* the mouse ear swelling test (MEST). Toxtcol. Appl. Pharmacol. 8,93-l 14.

Hilton

and Kimber

16 Gad, S. C (1988) A scheme for the prediction and ranking of relative potencies of dermal sensitizers based on data from several systems J Appl. Toxic01 8,361-368. 17 Descotes, J. (1988) Identification of contact allergens: the mouse ear sensitization assay. J. Toxic01 Cut. Ocular Tox~ol. 7,263-272 18 Miller, K , Maisey, J., and Malkovsky, M. (1984) Enhancement of contact sensitization in mice fed a diet enriched in vitamm A acetate. Int Arch. Allergy Appl Immunol. 75,120-125. 19 Massey, J. and Miller, K. (1986) Assessment of the ability of mice fed on vitamin A supplemented diet to respond to a variety of potential contact sensitizers. Contact Derm. 15, 17-23 20. Maisey, J., Purchase, R., Robbins, M. C , and Miller, K (1988) Evaluation of the sensitizing potential of 4 polyammes present in technical triethylenetetramme using 2 animal species. Contact Derm 18, 133-137 21 Corsml, A C , Bellucci, S B , and Costa, M G (1979) A simple method of evaluating delayed type hypersensitivity In mice J. Immunol Meth. 30, 195-200. 22. Moller, H. (1984) Attempts to induce contact allergy to nickel in the mouse Contact Derm 10,65-68. 23. Back, 0. and Larsen, A. (1982) Contact sensitivity in mice evaluated by means of ear swelling and a radiometric test. J Invest. Dermatol 78,309-312 24 Cornacoff, J. B., House, R V., and Dean, J H (1988) Comparison of a radioisotopic incorporation method and the mouse ear swelling test (MEST) for contact sensitivity to weak sensitizers. Fund Appl Toxic01 10,4M4 25 Robinson, M K (1989) Optimization of an m vitro lymphocyte blastogenesis assay for predictive assessment of immunologic responsiveness to contact sensitizers. J Invest. Dermatol. 92,860--867. 26. Robinson, M. K. and Sneller, D. L (1990) Use of an optimized in vitro lymphocyte blastogenesis assay to detect contact sensitivity to nickel sulfate m mice. Toxicol. Appl Pharmacol 104,106-l 16 27 Gerberick, G F , Ryan, C. A , Fletcher, E R., Sneller, D L., and Robinson, M. K (1990) An optimized lymphocyte blastogenesis assay for detecting the response of contact sensitized or photosensitized lymphocytes to hapten or prohapten modified antigen presenting cells Toxic In Vitro 4,289-292. 28 Dunn, B. J., Rusch, G M., Slglm, J. C , and Blaszcak, D L (1990) Variability of a mouse ear swelling test (MEST) m predicting weak and moderate contact sensmzatron. Fund Appl. Toxicol. 15,242-248. 29. Thorne, P S., Hawk, C , Kaliszewski, S. D , and Gumey, P. D. (1991) The nonmvasive mouse ear swelling assay. I. Refinements for detecting weak contact sensitizers. Fund. Appl Toxccol 17,790-806 30. Davies, A J S , Carter, R. L., Leuchars, E , and Wallis, V (1969) The morphology of immune reactions m normal, thymectomized and reconstituted mice II The response to oxazolone. Immunology 17, 111-126. 31. De Sousa, M A B and Parrott, D. M V (1969) Induction and recall in contact sensitivity, changes in skin and drammg lymph nodes of Intact and thymectomized mice J. Exp. Med. 130,671-686. 32. Pritchard, H. and Micklem, H S (1972) Immune responses m congenitally thymusless mice I. Absence of response to oxazolone Clin. Exp Immunol 10, 15 l-l 61

Murine

Lymph

Node Assay

235

33. Oort, J. and Turk, J L (1965) A histological and autoradiographic study of lymph nodes during the development of contact sensitivity in guinea pigs. Br. J. Exp. Pathol. 46, 147-154. 34. Parrott, D. M. V. and de Sousa, M. A. B. (1966) Changes m the thymus dependent areas of lymph nodes after rmmunologrcal stimulation. Nature 212, 1316-1317. 35 Asherson, G L. and Barnes, R. M R (1973) Contact sensitivity in the mouse. XII. The use of DNA synthesis in vitro to determine the anatomical location of immunological responsiveness to picryl chloride. Immunology 25,495-508. 36 Kimber, I, Mitchell, J. A., and Griffin, A. C. (1986) Development of a murine local lymph node assay for the determination of sensitizing potential Fed Chem. Toxicol. 24,585-586.

37. Kimber, I. and Weisenberger, C. (1989) A murine local lymph node assay for the identification of contact allergens. Assay development and results of an initial validation study. Arch. Toxicol. 63,274-282. 38. Kimber, I. and Dear-man, R. J. (1991) Investigation of lymph node cell prohferation as a possible immunological correlate of contact sensitizing potential. Fed. Chem. Toxicol. 29, 125-129. 39. Kimber, I., Hilton, J., and Weisenberger, C. (1989) The murme local lymph node assay for identification of contact allergens: a preliminary evaluation of in srtu measurement of lymphocyte proliferation. Contact Derm. 21,215-220 40 Kimber, I , Hilton, J , and Botham, P. A (1990) Identification of contact allergens using the murine local lymph node assay: comparisons with the Buehler occluded patch test in guinea pigs. J Appl Toxic01 10, 173-180. 41. Basketter, D. A. and Scholes, E. W. (1992) A comparison of the local lymph node assay with the guinea prg maximization test for the detection of a range of contact allergens. Fed Chem Toxicol. 30,65-69. 42. Basketter, D. A., Scholes, E. W , Kimber, I., Botham, P. A., Hilton, J., Miller, K., Robbins, M C., Harrison, P. T. C., and Waite, S. J. (1991) Interlaboratory evaluation of the local lymph node assay with 25 chemicals and comparison with guinea pig test data. Tox~ol Meth. 1,30-43 43. Scholes, E W , Basketter, D. A., Sarll, A. E., Kimber, I , Evans, C. D., Miller, K., Robbins, M. C , Harrison, P. T. C., and Waite, S J. (1992) The local lymph node assay: results of a final inter-laboratory validation under field conditions. J. Appl. Toxicol. 12,217-222.

44. Kimber, I., Hilton, J., Botham, P A , Basketter, D. A., Scholes, E W , Miller, K., Robbins, M. C., Harrison, P T. C , Gray, T. J. B., and Waite, S. J (1991) The murine local lymph node assay: results of an inter-laboratory trral. Toxlcol. Lett. 55,203-2 13 45. Kimber, I. and Basketter, D. A. (1992) The murme local lymph node assay. A commentary on collaborative studies and new directions. Fed. Chem. Toxicol. 30, 165-169. 46. Botham, P A , Hilton, J., Evans, C. D , Lees, D., and Hall, T J. (1991) Assessment of the relative skm sensitizing potency of three biocides using the murine local lymph node assay. Contact Derm. 25, 172-177

CHAPTER27

Polymorphonuclear Leukocyte Locomotion Matte0

VaZentino

1. Introduction Chemotaxis and random migration of polymorphonuclear leukocytes can be determined in the presence of test chemicals. A change in the chemotactic index provides an indication of the cytotoxicity of compounds.

The migration of polymorphonuclear leukocytes to the site of injury is an integral part of the inflammation process in vivo. Any compound that inhibits this locomotion, either by a direct cytotoxic effect or by interference with the chemotactic stimulus, would severely compromise the inflammation process of an individual. The Boyden chamber filter membrane method has been shown to be a simple in vitro method for studying the migration of PMN and examining the influence of chemicals on that migration. The chemotactic index gives an indication of the level of toxic effects of compounds on locomotion. Polymorphonuclear leukocytes are isolated from heparinized blood of healthy volunteers by Ficoll-Hipaque separation and suspended in RPM1 1640. The suspension is centrifuged against a micropore filter to deposit a layer of cells over one surface. Zymosan activated plasma (ZAP) or formyl-methionyl-leucyl-phenylalanine peptide (F-MLP) are prepared as the chemotactic stimulus. 1.1. Chemotaxis

Diluted ZAP or diluted F-MLP is placed in the lower compartment of a Boyden chamber (in the absence or presence of test compound). The From Methods m Molecular Brology, Eduteri by. S O’Hare and C K AtterM

Vol 43 In V&o Toxmty Testmg Protocols CopyrIght Humana Press Inc , Totowa, NJ

237

Valentino

cell-coated filter is placed over the lower compartment (cell-side up), the chamber assembled, and the top compartment filled with RPM1 1640. 1.2. Random

Migration

RPM1 is placed m the lower compartment of a Boyden chamber (in the absenceor presenceof test compound). The cell-coated filter is placed over the lower compartment (cell-side up), the chamber assembled, and the top compartment filled with RPM1 1640. The chambers are incubated for a fixed period of time at 37°C. The filters are then removed and subjected to various rinsing, fixing, and staining stages, after which they are mounted on slides (cell side down). The chemotactic index for chemotactic locomotion can be calculated by determining the number of cells that have passed through the filter in relation to the number of cells originally delivered onto the filter. The leading front for random migration can be determined by measuring the distance from the top of the filter to the furthest plane of focus. The degree of toxicity is related to a reduction in the chemotactic index and/or leading front. This system allows the simultaneous study of both chemotaxis and random migration. In chemotaxis the chemotactic stimulus induces a polarization and a receptor redistribution to the front of the cell; in random unstimulated migration there is no membrane receptor redistribution. The test is simple, inexpensive, and can be performed within 1 d. It can be used to study two populations of cells at once. Depositing a layer of cells on the surface of the filter is preferable to using a cell suspension, since all cells will begin the incubation, and hence be subject to migration, from the same starting position. Two types of filters can be employed in this method. The cellulose-ester membrane filters are used in preference to polycarbonated membranes to allow the measurement of both chemotaxis and random migration (the polycarbonate membrane being too thin). Drawbacks of the system include: 1. The test gives an indication of the responseof a population of cells, but less mformation about the behavior of single cells or how they actually reachedthe attractant. 2. It is difficult to confirm the uniformity of the chemotactic concentration gradientthrough the membranefilter.

Polymorphonuclear

Leukocyte

Locomotion

239

3. In a gradient of an attractant m vitro, the migration of PMN can be shown not only to be accelerated, but also to become orientated and directional. This is not easy to observe using a Boyden chamber, therefore direct observation may also be necessary. 4. Proteins that are nonchemotactic (e.g., albumin) can influence the random migration and also bind to test chemicals, affecting the results obtained. 1.3. Modifications of the Technique This method investigates the effect of chemicals on the response of PMN to a chemotactic stimulus. It can be modified to study both the chemotactic and antichemotactic properties of a test substance. 1.3.1, Chemotactic Properties Add the chemical to be tested to the lower compartment of the Boyden chamber in the absence of the chemotactic stimulant, ZAP or F-MLP, and incubate for 3 h. Stain and count the cells using the method outlined. 1.3.2. Antichemotactic Properties Add the chemical to the lower compartment of the Boyden chamber together with the chemotactic stimulant, ZAP or F-MLP, and incubate for 3 h. Stain and count the cells using the method outlined. The antichemotactic effect is owing to a direct interference with the chemotactic stimulant, rather than the effect on the response of the cells to the stimulus. 1.3.3. Comparison with Other Techniques PMN locomotion can be measured in vitro using several methods other than the filter membrane method outlined below. 1. Direct microscopic method: This method does not allow mvestigations of mixed cell preparations, for example when studying the dose-effect relationship of a drug, since only one type of cell can be observed at any one time. 2. Capillary tube method: When whole blood is centrifuged, the buffy coat cells tend to migrate upward into the overlying plasma. This method allows effects on random migration to be determined but does not permit simultaneous measurement of chemotaxis. 3. Agarose method: PMN are placed in a well cut vertically in an agarose plate. A chemotactic solution is placed in a well 3 mm from the first well and chemotactic factor-free medium is placed 3 mm away on the opposite side. The distance that the cells migrate toward each of these wells gives

240

Valentino an indication of chemotaxis and random migration, respectively. There are two main drawbacks to this system, however. First, the concentration gradient of the factor affecting the cells varies during the period of incubation, leading to errors in interpretation. Second, PMN stick to glass and plastic, thereby affecting their movement.

2. Materials 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

Boyden chamber. 13-mm Micropore filter: 5 w cellulose-ester membrane filters. Cytocentrifuge. Incubator, 37”C, 5% CO* m air. Pasteur pipets. Slides and thm cover shps. Freshly obtained blood from healthy humans. Preservative-free lithium heparin solution: 5000 IU/mL. Ficoll-Htpaque solution. RPM1 1640. 5 mg/mL desiccated zymosan. Formyl-methionyl-leucyl-phenylalanine peptide (F-MLP). Human albumin. 70,90, and 100% isopropyl alcohol. Acid alcohol: 3 drops of HC1/200 mL 70% isopropyl alcohol. Bluing agent: 20 g MgS04 + 2 g NaHCOs m 1 L of water. Xylene. Hematoxylin. Dimethylsulfoxide (DMSO). Hank’s balanced salt solution (HBSS): (g/L), CaCIZ (anhyd.) (0.14), KC1 (0.40), KH,PO, (0.06), MgCl, .6H,O (0. lo), MgS04 . 7Hz0 (0. lo), NaCl (8.00), NaHCO, (0.35), Na2HP04 .7H,O (0.09), o-glucose (1 .OO),Phenol Red (0.01). 21. Test chemicals: All chenncals must be dissolved m solution before apphcation to the Boyden chamber. Dissolve all water-soluble compounds in RPM1 1640 medium. Any insoluble compounds should be dissolved in ~0.5% DMSO m RPM1 1640 medium.

3. Methods 3.1. PMN Isolation 1. Take a 5 mL sample of blood from a healthy volunteer. Place into a tube containing 5 pL of heparm solution and store at room temperature. 2. Separate the cells using Ficoll-Hipaque by the method of Ferrante and Thong (I): Carefully add 3.5 mL of the heparinized blood to 3 mL of Ficoll-

Polymorphonuclear

Leukocyte

Locomotion

241

Hipaque solution in a sterile 10 mL centrifuge tube (this should form 2 distmct layers). Centrifuge, using a swinging bucket rotor, at 300g for 30 min at room temperature. 3. Differential migration during centrifugation results in the formation of 2 layers of cells and a red blood cell pellet. The top layer, at the plasma:M-PRM interface, should be approx 13 mm wide and contain mononuclear cells. The second layer (directly beneath the mononuclear cell fraction) consists of 90-96% pure polymorphonuclear leukocytes. 4. Carefully transfer the second fraction to fresh tubes and wash the cells with HBSS at room temperature. Centrifuge at 25Og for 10 min at room temperature and repeat the wash and centrifugation. 5. Resuspend cells, count, and dilute to a density of 2 x 106/mL in RPM1 1640. Carry out a viability test (e.g., Trypan blue exclusion [22]; fluorochromatic test [3]). Note: No decrease in viability should be observed over a 3 h incubation at 37°C even in the presence of the chemical to be tested. 3.2. PMN Deposition on a Filter Centrifuge 0.5 mL of the PMN suspension against a 5 pm filter using a cytocentrifuge at 500 rpm for 1.5 rnin (see Note 1). 3.3. Preparation of Chemotactic Stimulus 1. ZAP is prepared by the method of Colditz and Movat (4): Obtain hepannized (10 U/r&) fresh plasma from healthy volunteers (sufficient for 5 mL of ZAP). Incubate at 37°C for 30 min with 5 mg/mL desiccated zymosan. Agitate occasionally. Centrifuge at 800g for 10 min at room temperature to remove zymosan. Store activated plasma in aliquots at -70°C for up to 4 mo. Thaw out gradually at room temperature when required. Before use dilute ZAP 1: 10 in RPM1 1640. 2. F-MLP peptide IS dissolved at a concentration of 10-8M in a solution of RPM1 1640 contaming 2% human albumin according to Sasagawaet al. (5). 3.4. Preparation of the Blind Well Boyden Chamber Duplicate chambers should be run for each control and test condition. See Fig. 1 and 2. 3.4.1. Chemotaxis Inject diluted ZAP or F-MLP into the lower compartment of a Boyden Chamber using a Pasteur pipet (this prevents air bubbles forming underneath the micropore filter), Insert the filter into the chamber with the cellular side facing upward. Screw on the cap. Fill the upper chamber with RPM1 1640 (control) or with RPM1 1640 containing the required concentration of test compound (no protein should be added).

Valentino

242

Fig. 1. Blind well Boyden chambers. 3.4.2. Random Migration 1. Inject RPM1 1640 into the lower chamber of the Boyden chamber. Insert the filter into the chamber with the cellular side facing upward. Screw on the cap, Fill the upper chamber with RPM1 1640 (control) or RPM1 1640 containing the required concentration of test compound (no protein should be added). 2. Incubate the chambers for 3 h at 37OC,5% CO2 in air.

3.5. Filter 1. 2. 3. 4. 5. 6. 7. 8. 9.

Staining

Remove filters at the end of the incubation period. Immediately place in 100% isopropyl alcohol and fix for 30 s. Transfer the filters to hematoxylin and stain for 2 min. Rinse in water. Decolorize in acid alcohol for 30 s. Rinse in water. Immerse in bluing agent for 30 s. Rinse in water. Dehydrate each filter with 70,90, and 100% isopropyl alcohol for 2 min in succession. 10. Place filters in xylene for at least 10 min.

Polymorphonuclear

Leukocyte

rI

243

Locomotion

Q A--

L

Filter

PMN

ill ‘II

Fig. 2. Schemeof blind well Boyden chamber. 11. Mount the stainedfilters, cell-side down, on slides using Permount. 12. Cover with a thin cover slip. 3.6. Results

Count the total number of cells that have completely migrated through the filter within 10 random microscopic fields (400x) using a 5 x 5 mm photographic reticule. Calculate the chemotactic index according to Hill (6): Chemotactic index = PMN in 10 random fields/ PMN delivered to the filter in millions

The denominator is 1 (1 x lo6 cells added) in this case, but should be adjusted according to the number of cells added. In random migration there is an incomplete migration of cells through the filter, therefore, calculate the leading front according to Zigmond and Hirsch (7). The leading front is measured in five fields across the filter using the optimal micrometer on the fine focus knob of the microscope. Focus on the top of the filter (starting plane). Adjust the fine focus

Valentino

knob until at least 2 PMN come into focus (arrival plane). Measure the distance from the arrival plane to the starting plane. Repeat this measurement in four other fields across the filter. The leading front is the mean of these five measurements. 4. Note 1. When the cells rmgrate through the filter a certain percentage (0.90-0.38) fall off the underside of the filter mto the fluid contained in the lower chamber.

References 1, Ferrante, A. and Thong, Y H (1980) Optimal conditions for simultaneous punfication of mononuclear and polymorphonuclear leukocytes from human peripheral blood by the Ficoll-Hipaque method J. Zmmunol. Meth. 36, 109-l 17 2 Bhuyan, B K., Loughman, B E., Fralser, T. J , and Day, K. J (1976) Comparison of different methods of determining cell viability after exposure to cytotoxic compounds Exp. Cell Res. 97,275280. 3. Aeschbacher, M., Remhardt, C. A., and Zbinden, G. (1986) A rapid cell membrane permeability test using fluorescent dyes and flow cytometry. Cell. Biol. Toxic01 2, 247-255. 4. Colditz, I. G and Movat, H Z. (1984) Kmetics of neutrophil accumulation in acute Inflammatory lesions induced by chemotaxins and chemotaxinogens. J Immunol.

133,2169-2173 5. Sasagawa, S , Suzulu, K., Sakatani, T., and Fujikura, T. (1985) Effects of nicotine on the functions of human polymorphonuclear leukocytes m vitro. J. Leuk. Biol

37,493-502. 6. Hill, H R., Hogan, N. A , and Mitchell, T G. (1975) Evaluation of a cytocentrifuge method for measuring neutrophll granulocyte chemotaxis. J Lab Clin. Med. 86, 703-710. 7. Zlgmond, S. H. and Hirsch, J. G (1973) Leukocyte locomotion and chemotaxisnew m methods for evaluation and demonstration of a cell-derived chemotactic factor J. Exp. Med. 137,387-410.

CHAPTER28 Immune

Function

Assays

Steve Nicklin 1. Introduction The increasing interest in immunotoxicology as a subtopic of toxicology acknowledges the growing recognition that the immune system and immunocompetence per se may be a sensitive and perhaps early indicator of toxicity. Indeed, accumulating evidence based on human and experimental animal studies now link exposure to a variety of drugs and chemical agents with various immunotoxic processes, including immunosuppression and hypersensitization (1,2). The primary objective of this chapter is to describe a minimum battery of standardized techniques that allows the routine assessment of potentially immunotoxic compounds (3,4). The chapter is divided into three main sections: the first covers humoral immunity, namely the plaque assay; the second is concerned with cell-mediated immunity and covers in vitro T-cell mitogenicity; the third covers the mixed lymphocyte assay and in vitro cytotoxicity. 1.1. Source of Cells for In Vitro Assessments In immunotoxicological studies, as with other studies utilizing some form of in vitro assessment, cells must still be taken from animals following a standardized dosing regimen (5). Whereas the inbred mouse is usually considered the animal of choice for fundamental immunological research, the rat is now increasingly favored for immunotoxicity studies. This stems from the fact that the rat is traditionally favored as the standard rodent under existing toxicological legislation. As with other aspects of toxicology, the donor animals, whether rats or mice, must always be From. Methods m Molecular Biology, Edtted by S. O’Hare and C. K Atterwlll

Vol. 43 In Wro Tax/city Testmg Protocols Copynght Humana Press Inc , Totowa, NJ

245

246

Nicklin

healthy, specified pathogen free, and obtained under license from an approved and accredited supplier. Prior to study, the animal(s) should be held quarantined for a minimum of 7 d in order to acclimatize them. Following experimentation, the animals must be killed humanely using a British Home Office approved method (e.g., cervical dislocation or CO, intoxication). 1.2. Assessment

of Humoral

Immunity

The qualitative and quantitative assessment of the humoral response to a test antigen challenge currently represents one of the most reliable indicators of humoral immune competence. The plaque assay was developed by Jerne and Nordin to detect and enumerate cells producing antibody against erythrocyte antigens (6). Briefly, splenic lymphocytes from test/control donors previously immunized with sheep erythrocytes are incubated in an agar gel with an aliquot of the irnmunizmg erythrocytes. Antibody produced by the immune plasma cells defuses into the gel and binds to the red cells in the locality. Following the addition of complement, these erythrocytes are specifically lysed, producing “holes” in the erythrocyte lawn. These holes are referred to as plaques that are counted. Each plaque thus identifies the position of a plaque forming cell (PFC), also referred to as an antibodyforming cell (AFC). By counting the plaques produced by a known concentration of spleen cells it is possible to determine the number of PFCs/ spleen. The PFC response to sheep red blood cells (a T-dependent antigen) is a particularly pertinent indicator of immune competence since the response requires the cooperation of B cells, T cells, and macrophages. Current research indicates that the majority of xenobiotics that significantly decrease this response will also significantly alter host resistance to various microbial or tumor cell challenges.

1.3. Lymphocyte

Transformation

Assays

Lymphocytes can proliferate in vitro following stimulation with a variety of agents, including antigen, plant lectins, and constituents of bacterial cell membranes (3,8,9). Lymphocyte transformation assays have proven very useful in detecting xenobiotics that affect cell-mediated immunity. However, it is important to note that such assaysmonitor the effect of a xenobiotrc on the afferent arm of the response since the cells are removed from the treated donor and assessedin vitro, Short-

Immune

Function Assays

247

acting or localized in vivo affect may be lost in vitro. It is, however, possible to add the xenobiotic to the culture system or perhapsmore appropriately the active metabolites, if known. 1.4. Lymphocyte Transformation by Mitogens Substances that evoke significant cellular proliferation in a nonimmunological fashion are known collectively as mitogens. Specific mitogens have been shown to selectively stimulate different lymphocyte subpopulations. For example, phytohemagglutinin (PHA) and concanavalin A (con A) stimulate T-cells, whereas bacterial lipopolysaccharides (LPS) stimulate B-cell populations. Because of their selectivity, mitogens are frequently used to evaluate the ability of lymphocytes to undergo blastogenesis and proliferation. Mitogens are also employed to monitor the adverse effects of xenobiotics on cell membrane receptors/intracellular metabolic processes involved in cell division. For logistical reasons as well as convenience, B and T cell studies are performed simultaneously using single-cell suspensions from donors treated as described in Section 3.2. 1.5. Mixed Lymphocyte Reaction (MLR) A mitotic response is also elicited when cells taken from two inbred strains or from outbreds from the same species are cocultured together in vitro. The mixed lymphocyte response is now a recognized technique for detecting xenobiotics that affect T-cell mediated immune reactivity. Indeed, a good correlation has been demonstratedbetween the suppression of this reaction and increased susceptibility in classical host resistance models, e.g., listeria and PYB6 sarcoma challenge, for which cell-mediated immunity is known to play a major role in host defense (3,4). In order to ensure a unidirectional response, one population, the allogenic lymphocytes, which serve as stimulator cells, are inhibited from proliferation by mitomycin C or irradiation to arrest cell division, alternatively it is possible to use a Fl x Parent combination, i.e., strain A vs strain (AxB)Fl, which avoids the need for the mitomycin C or irradiation. Thus, only proliferation by the responder population is measured. Suitable strain combinations include PVG vs DA for rat studies and C3H vs DBA/2 for murine studies. In both instances, the responder population is quoted first, i.e., responder vs stimulator,

Nicklin

248 1.6. Cell Mediated

Cytotoxicity

Assay This assay measures the capacity of lymphocytes from test and control donors to kill 51Crlabeled target cells. The assay addressesthe following functional capacities: 1. The ability of the host cells to recognize allogenic cells as foreign; 2. Their capacity to undergo clonal expansion; and 3. Their abihty to differentiate mto functional cytotoxic cells.

2. Materials 1. Mitomycin C. 2. Lyophilized guinea pig complement. 3. Ice. 4. Anti IgG, see Section 3.7. 5. Tritiated thymidme solution. 6. Scmtillation liquid. 7. Concanavalin A. 8. Triton X-100. 9. Sheep erythrocytes. 10. P815 mastocytoma cells. 11. Sodmms’ chromate. 12. Dissection equipment-mcludmg scissors, watchmaker’s forceps, and scalpel blades. 13. Silicon rubber bung. 14. Stainless steel or nylon gauze mesh, size 400. 15. 20 mL plastic “V’‘-bottomed centrifuge tubes. 16. 0.22 or 0.45 pm filters. 17. Flat-bottomed microtiter tissue culture plates. 18. Hemocytometer. 19. Multichannel pipets. 20. Centrifuge. 21. CO2 incubator. 22. Autoclave. 23. Plaque viewer. 24. Scintillation counter. 25. Microscope. 26. Earle’s Balanced Salt Solution. 27. Oxoid no. 1 agar. 28. 30-mm plastic Petri dishes. 29. Eagle’s Minimal Essential Medium. 30. RPM1 1640 + L-glutamine + 40 nQt4HEPES.

Immune

249

Function Assays

31, White Blood Cell Diluting Fluid: 0.15% Toluidine blue, 1.O%glacial acetic acid in deionized water. 32. Alsever’s solution: 20.50 g Dextrose, 4.20 g sodium chloride, 8.00 g sodmm citrate. Dissolve in 1 L distilled water. Sterilize by membrane filtration (0.22 or 0.45 pm Millipore umt or equivalent). Store at 4OC. 33. DEAE dextran (stock solution): 3.00 g DEAE dextran, 8.40 g sodium chloride. Dissolve m 1 L distilled water. Buffer to pH 6.90 with 6.9 HEPES, 1.15 g disodium hydrogen phosphate, 0.15M phosphate buffered saline (PBS), pH 7.2, 8.00 g sodmm chloride, 0.20 g potassium chloride, 0.20 g potassium dihydrogen phosphate, Dissolve rn 1 L distilled water and autoclave. It is convenient to make up 10x for storage and dilute as required. PBS tablets are available also from chemical supply houses, e.g., Sigma (St. Louis, MO). 34. Physiological salme (normal salme): 0.14M. Dissolve 8.5 g sodium chloride in 1 L distilled water and autoclave. It is convenient to make up 10x for storage and dilute as required. Sterile saline is available also from chemical/pharmaceutical suppliers. 35. Tissue culture media: RPM1 1640 plus L-glutamine and 40 mil4 HEPES. This medium should be used as supplied when washing cells. For cell culture, the basic medium is supplemented with 60 U/mL penicillin/ streptomycin mix and 10 heat-inactivated fetal calf serum (56°C for 30 min). Where stated, culture media may be further supplemented with 2.5 x 10m5M2-mercaptoethanol.

3. Methods 3.1. Single Cell Lymphocyte Suspensions 1. Single cell lymphocyte suspensions may be prepared from the spleen, lymph nodes, or thymus. The lymphoid organ or organs, as required, must be removed using aseptic procedures. Briefly, the abdomen/thorax of the donor is swabbed and wet with 70% ethanol or industrial methylated spirit (IMS), the skin is cut laterally and pulled to expose the thorax and/or peritoneal wall, the barrier is cut, the tissues are retracted and reflected as necessary, and the test organ(s) removed and placed in a sterile Petri dish (plastic, approx 60 x 15 mm, disposable). All procedures should be performed under sterile filtered/laminar airflow with sterile instruments, Any adhering fat should be removed from the organ(s) using fine (no. 4 or 5 size) watchmaker’s forceps. The organ should then be transferred to a fresh Petri dish containing approx 2 mL cold (4OC) PBS, pH 7.2 (see Note 1). 2. In order to obtam the lymphocytes it is necessary to disrupt the organ by either repeatedly teasing/tearing the organ using forceps, or repeatedly cuttmg the organ usmg a pair of scalpel blades. This is usually sufficient to

250

Nicklin

disrupt thymus and lymph nodes, however, for spleens (especially from older rats) it is often necessary to press the spleen fragments further, employmg a firm crushing and twisting action using a sterile silicon rubber bung. Alternative methods of obtaming single cell suspensionsused by other investigators include repeatedly squeezing the tissues between the frosted ends of two microscope slides, pressing the organs through steel grids or “tea stramers,” or using a loosely fitting ground-glass tissue homogemzer 3. The resuitmg suspension is filtered through stainless steel or nylon gauze to yield a debris-free suspension (coarse/medium fme is adequate, but final mesh size should be 400 or equivalent). The cells should then be transferred to a centrifuge tube (sterile, 20 mL, plastic, “V” bottom, disposable) to be washed twice by centrifugation (5OOgfor 10 mm at 4°C) m PBS and once in RPM1 tissue culture media. Gently resuspend the cells and disperse any clumps between the washes by using a Pasteur pipet and rubber bulb. Finally, resuspend the cells in cold tissue culture medium (RPM1 164O/FCS), perform a cell viability count (see Section 3.2.), and adjust the cell concentration required for the assay. Hold on ice until required. 3.2. Estimation of Viable Cell Number 1. An aliquot of the cell suspension to be evaluated is mixed 5050 with white blood cell diluting fluid. View 20 pL of suspension in a hemocytometer (e.g., Improved Neubauer) using a 40 x objective. Viable cells exclude the Trypan blue and shme like beads of morning dew; dead cells are stained dark blue/black. Count at least 200 cells and calculate the percentage viability (see Note 2). 2. A standard immunotoxrcological assessmentof a compound requires that rodents be dosed daily with vehicle or set doses of the test article via an appropriate route for up to 28 d (trmes may, of course, be varied depending on circumstances). The animals are immunized intravenously via a lateral tail vein with a set concentration of sheep erythrocytes (2 x lo* washed cells in 0.5 mL physiological saline) on d 25 for IgM-producmg plaque cells (primary response) and d 23 or 24 for the IgG response (for the secondary response, see Section 3.5.). 3. One day after the last exposure to the test agent the animals are killed and splenocytes are prepared as described in Section 3.1. and adjusted to a working concentration of 2 x lo6 cells/ml. 3.3. Preparation of the Agar 1. Dispense 100 mL of Earle’s Balanced Salt Solution (EBSS) into a sterile glass flask and add 1M HEPES droowise until the color iust turns vellow.

Immune

251

Function Assays

2. Add 0.5 g Oxotd No. 1 agar and heat until boiling; mix while heatmg to avoid charring. 3. Add 1.6 mL stock DEAE dextran, mix well, and hold at 47OC (to keep molten) until required.

3.4. The Assay 1. Into a 5.0 mL glass tube mix: a. 1.4 mL agar/dextran solution. b. 25 pL 20% SRBCs. c. 100 pL of the splenocyte suspension. 2. Mix the contents by flicking the tube and pour rapidly, avoiding bubbles, into a 30 mm plastic Petri dish. Allow the agar to set, then incubate for 2 h at 37°C in a 5% CO2 incubator. 3. Meanwhile reconstitute 2-5 mL lyophilized guinea pig complement according to supplier’s instructions; absorb against packed SRBCs (25 pL/ 1 mL complement) on ice for 30 min. After 30 min, centrifuge the complement&RBC mixture at 500g for 10 min at 4°C and recover the complement as supernatant. Dilute complement l/20 or l/40 in culture media and hold on ice until required. After incubation time is complete, add 500 pL complement to “flood” each Petri dish and incubate for a further 1 h, then examine the plates for plaques (see Notes 3 and 4).

3.5. Enumeration The plates should be coded and read blind. (A number of commercial plaque viewers are available). The plaques may also be enumerated using a bacteriological colony counter, however, it is also possible to count the plaques unaided using indirect lighting and “under marking” each plaque using a felt tip pen. The results of such an assay are expressed as specific activity (PFCs/106 spleen cells) or PFCs per spleen (see Notes 5 and 6).

3.6. Indirect

Plaque

Response

1. Measurement of IgG-defined PFC response (the secondary response) to the antigen challenge is performed in an identical fashion to that described above except that an appropriate dilution of an anti-IgG (e.g., 500 pL of l/500 dilution of rabbit antimouse or antirat IgG, depending on the species under study) is added to the plates for 30 min prior to the addition of the complement. Since the activity of the anti-IgG antibodies vary between suppliers it 1salways necessary to perform a titration expenment in order to determine the most effective concentration for use. The number of IgG PFCs is determined by subtracting the number of plaques produced in the absence of the secondary antibody (i.e., the IgM plaque count) from the

Nicklin number produced in the presence of the secondary antibody (i.e., the combined sum of the IgG and IgM plaques). As with IgM data, results are expressed as IgG PFCs/106 spleen cells or IgG PFCs/spleen (see ref. 7 for alternative methods) (see Note7). 2. Lymphocyte transformation by mitogens: Reconstitute and prepare graded dilutions of each mitogen as required using tissue culture media. The optimal concentration of mitogen, however, often varies between suppliers and strain and species under test. Thus, it is usual to use a range of doses (see Notes 8-10). 3. Suggested doses for LPS and Con A are 0.5, 1.0, 5.0, and 10 pg/well and 0.1, 0.5, 1.0, and 5.0 pg PHA/well. Once prepared, the solutions have a shelf life of 1 wk at 4°C. However, aliquots of 10x mitogen solutions may be stored for extended periods m a -20°C deep freeze. Add 100 pL of culture media plus graded concentration of each mitogen to be studied to each well of a flat-bottomed microtiter tissue culture plate. Use a mmimum of three rephcates per mitogen concentration (we routinely use five). It is convenient to plate out, starting with the “culture medium alone” control wells and working up the concentration gradient. Multichannel pipets are strongly recommended for this purpose. Plates may be prepared up to a week in advance but should be maintained at a minimum temperature of -2OOC. 3.7. Culture 1, Prepare cells as in Section 3.1. above and adjust to a working concentration of 2 X lo6 cell/mL. 2. Add 100 pL of cells/well and incubate the plate(s) at 37°C with 5% COz. Peak proliferation occurs between 48 and 72 h, dependmg on mitogen and the nature of the cells under study. The magnitude of proliferation is assessedby tritiated thymidine (H-TdR) mcorporation (see Sections 3.8. and 3.9.). 3.8. Preparation of H-TdR Dilute H-TdR solution 1:20 in tissue culture media using appropriate care. Remove plates from incubator and pulse each well with 20 FL of the H-TdR solution. Incubate as above for a further 6 h at 37°C. 3.9. Cell Harvesting Harvest cells using a cell harvester and count radioactivity by liquid scintillation counting. 3.10. Data Handling Data may be expressed as disi;ltegrations per minute (dam), net dam = mean dam in replicate treated cultures - mean dam of replicate con-

Immune

253

Function Assays

trol cultures. The use of stimulation indices (S) = mean of mitogentreated cultures/mean of control cultures are often misleading and should be avoided. 3.11. Preparation of Stimulator Cells for MLR Unidirectional Format

Spleen cells from the stimulator population are prepared first since they require mitomycin C treatment. Use 50 p,g mitomycin C/mL of 2 X lo7 stimulator cells. Incubate at 37°C. Wash 4X by centrifugation prior to use. of Responder Cells The responder cells from the various treatment groups are prepared as described in Section 3.1, Both stimulator and responder cell populations are adjusted to a final concentration of 2 x lo6 cells/ml in RPM1 tissue culture media containing 10% FCS. 3.12. Preparation

of Controls Appropriate controls are also required to ensure the successful outcome and interpretation of MLR assays. The stimulator cells, after mitomycin C treatment, should be incubated with a standard dose of C. A or PHA to ensure that their ability to proliferate has been inhibited. Similarly, it is also usual to incubate the responder cells under identical conditions to ensure that they are capable of proliferation. 3.13. Choice

3.14. Culture

The various lymphocyte populations are established in culture in “U” bottomed microtiter plates by mixing 100 l.tL aliquots of the responder and stimulator cells (see Notes). The various culture combinations should be set up in quadruplicate and the plates are incubated at 37°C with 5% CO2 for 4 d. The degree of proliferation is assessedby tritiated thymidine (H-TdR) incorporation using the procedure described in Sections 3.8. and 3.9. 3.15. Data

Handling

The data from control animals and animals receiving treatment is usually recorded as mean dpm of responder cells cultured in the presence and absenceof stimulator cells. The use of stimulation indices (S) = mean of responder cultures plus stimulator cells/mean of responder cultures alone are often misleading and should be avoided.

254

Nicklin 3.16. Cell Mediated Cytotoxicity 3.16.1. Cell Preparation

Assay

One day after the last exposure to the test agent the animals are killed, splenocytes prepared as described in Section 3., and adjusted to a working concentration of 2 x lo6 cells/ml using Eagle’s MEM supplemented with 10% fetal calf serum, 25 W/L HEPES, 1 n-N/L L-glutamme, 50 pg/mL gentamicin, and 1 x lo5 mM 2-mercaptoethanol. 3.16.2. Sensitization

In this assay, donor cells are first sensitized to the target cells by coculture. Thus, 19.5 mL of the above cells are admixed with 0.5 mL mitomycin C-treated P8 15 mastocytoma cells in a 25 cm2 tissue culture flask to yield a final 50: 1 responder:sensitizer ratio. Use one flask/donor. Incubate the cultures under standard conditions for 5 d; harvest the cells by gentle agitation, decanting, and rinsing the flasks with PBS. Pool the washings by centrifugation (500g for 10 min), wash once in fresh PBS, and resuspend in EMEM. Hold on ice until required. 3.16.3. Preparation

of Labeled

Target Cells

Label fresh P815 cells with 51Cr.This IS most easily achieved by incubating 5 x lo7 cells/ml with 100 @i of sodium 51Cr chromate at 37°C for 40 min. Remove free radioactivity by centrifugation two washes in medium, rest the cells on ice for a further 30 min, then wash 3 times, adjust to 2 x lo5 cells/ml, and hold on ice. 3.16.4. Cytotoxicity

Assay

To initiate the assay, 100 PL of the labeled cells are cocultured with 100 PL of graded numbers of splenic effector cells in “U’‘-bottom microtiter plates to yield a serial log 2 dilution of effector:target ratio from 25: 1 to 0.75: 1. The plates are incubated at 37°C and 5% CO, for 4 h. 3.16.5. Harvesting

After incubation, the plates are centrifuged (3OOgfor 10 min) and 100 PL of the supernatant is removed from each well for gamma-counting. Target cells are treated with 0.1% Triton X- 100, and the media alone serve as maximum and spontaneous release controls. 3.16.6. Data Handling

The data are expressed as the percentage of specific release as comparedto a function of the effector:target ratio. Results may also be expressed

Immune

Function Assays

255

in terms of LYs, (lytic 50) U, which refer to the number of splenocytes required to kill 50% of the target cells at any given ratio. 1.

2. 3. 4.

5.

6.

7. 8.

9.

4. Notes It must be emphasized that at all stages of the procedures the experimenter will be handling living cells. Maximum viability is only mamtamed if the cells arekept at 4”C, i.e., ice water. Speedis also essentialuntil the cells are safely in culture. Get help with the preparation if required. Improve your technique if you score less than 90% viable. It is recommended that laboratories optimize the time of assessment for this procedure, because it is influenced by doses of SRBCs, species, strain, and age of the experimental animals. The source, age, and dose of the sheep erythrocytes admirustered can dramatically affect the magnitude of the PFC response. The erythrocytes must be stored at 4°C m Alsever’s solution, used within 2 wk of collection, and washed (by centrifugation in saline until the supernatant is free of hemolysis) prior to administration. Precise cell counting is crucial to successand it is strongly recommended that each laboratory conduct a dose-response study with their SRBCs on each strain and speciesof experimental animal prior to initiating any xenobiotic studies. Routinely, at least two dilutions of spleen cells are used to ensure that the number of plaques are in a range that can be accurately counted. This is particularly important with test agents that may be immunostimulatory or affect immunoregulatory suppressor cells, thus resultmg m an enhanced or stimulated response. Whereas the PFC response peaks 4-5 d postantigen challenge, measurement of responses after 3-7 d could reveal more subtle xenobiotic-induced affects on response kinetics. All mitogens should be considered as hazardous; take appropriate precautions. Gloves and face mask should be worn for handling tntiated thymidine. All syringes, needles, tubes, and plastics contaminated with radioactivity must be soaked in detergent (e.g., lipsol) overnight, rinsed thoroughly, and disposed of m accordance with local arrangements. Mitomycin C treatment: Mitomycin C is highly toxic Take care. Alternatively, use irradiated (3000 R) cells. However, please note that the stimulatory capacity of irradiated cells falls within a few hours if they are allowed to stand. Some laboratories use a higher stimulator to responder cell ratio, e.g., 4 or 5:l. This appears to depend on the particular strain combmations selected for the MLR assay and is a parameter that needs to be optimized in each laboratory conducting the assay.

Nicklin

10. Mitomycin C treatment: Note: Mitomycin C is highly toxic. Take care! Select P815 cells from log phase rather than confluent cultures for labeling. Use appropriate precautions with the 51Cr since it 1sa gamma-emitter. It is important to note that this assay only monitors the effect of a xenobiotic on the afferent arm of the response, i.e., cells are removed from the treated donor and assessedin vitro (10).

References 1. ECETOC 10 (1987) Zdentification oflmmunotoxic Effects of Chemicals and Assessment of Their Relevance to Man ECETOC Monograph 10, Brussels, Belgium 2. Murray, M. and Thomas, P. T. (1992) Toxic Consequences of Chemical Inteructions with the Immune System (Miller, K., Turk, J., and Nlcklm, S , eds ), Blackwell Scientific, Oxford, UK, pp 65-85 3. Dean, J., Lauer, R., House, R., Ward, E., and Murray, M (1987) Experience with validation of methodology for immunotoxity assessment in rodents, in Zmmunotoxicology (Berlin, A., Dean, J., Draper, M. H , Smith, E M B., and Spreafico, F , eds.), Martinus Nijhoff, The Netherlands, pp 135-158. 4. Luster, M. I., Munson, A E., Thomas, P., Hossapple, M. P., Fenters, J., White, K. L., Jr., Laurer, L. D., and Dean, J. H. (1988) Development of a testmg battery to assesschemical-induced immunotoxiclty. Fund. Appl. Tox~ol. 10,2-19. 5. Nicklin, S. and Miller, K. (1991) Dose-effect and dose responses in immunotoxicology. Problems and conceptual conslderatlons, in Zmmunotoxicology and Zmmunotoxicity of Metals (Dayan, A D , Hertel, R. F., Heseltine, E , Kazantzis, G., Smith, E. M , and Van der Venne, M T., eds.), Pergamon, New York, pp. 43-56. 6. Jerne, N K., Nordin, A. A, and Henry, C. (1963) The agar plaque technique for recognising antibody producing cells, in Cell-Bound Antibody (Amos, F. J. and Koprowski, U., eds.), Wistar, Phihadelphia, pp. 26-3 1. 7 Hudson, L., and Hay, F. C. (1976) Advanced techniques in cellular Immunology, in Practical Zmmunology (Hudson, L. and Hay, F C., eds.), Blackwell, Oxford, pp. 213-218 8 Janossy, G. and Greaves, M. F. (1972) Lymphocyte activation I. Response of T and B lymphocytes to phytomltogens Clm. Exp. Immunol 10,525-536. 9 Thorpe, P. E. and Knight, S. C. (1974) Microplate culture of mouse lymph node cells. I. Quantitatlon of response to allogemc lymphocytes, endotoxin and phytomltogens. J. Immunol. Meth. 5,387-404 10. Bradley, S. G. and Morahan, P. S (1980) Approaches to assessing host resistance. Environ. Health Persp. 43,62-69.

CHAPTER29

The DNA Alkaline Genotoxicity

Unwinding Test

Gem-g Bolcsfoldi 1. Introduction DNA is the target molecule for chemical and physical mutagens and carcinogens. These agents may attack DNA directly or modify other cellular processes associated with the integrity of the genome. The resulting alterations in the structure of DNA may either be irreversible, giving rise to mutations and cell death, or be repaired by various DNA repair enzymes without producing toxic effects. DNA damage can also be caused by agents that primarily affect the viability of the cell (1). Several methods have been developed for the detection of DNA damage (2-9). These measure one or more of the following parameters: 1. 2. 3. 4.

Conformational changes; A decreasem molecular size or weight of DNA; The formation of single- and/or double-strandbreaks; The formation of alkali labile sites.

The alkaline unwinding technique exploits the finding that the DNA double-strand unwinds in an alkaline solution from free ends (10,11). The rate of unwinding is constant, but the length of unwound DNA will depend on the number of points where unwinding is occurring. The amount of unwound DNA is therefore proportional to the number of strandbreaks and is measured by separating the unwound single-stranded DNA fragments from the intact double-stranded DNA and quantifying the two. A convenient way to achieve this is by eluting the DNA through a hydroxyapatite column with solutions of differing ionic strength. From Methods !n Molecular B/o/ogy, Vot 43 In Vitro Toxmty Testmg Protocols EdIted by S O’Hare and C K Atterwlll Copynght Humana Press Inc , Totowa, NJ

257

Bolcsfoldi DNA strand breaks result from a number of different types of reactions (12). These include: 1, 2. 3. 4. 5. 6.

Base and nucleotide excision repair; Direct scission of the DNA backbone by chemical or radical attack; Scission following the binding of intercalating agents; Alkali labile DNA adducts; Endonuclease and topoisomerase action; and DNA hydrolase release from lysosomes.

Thus, most types of chemical attack on DNA will result in strand

breaks and lead to increased unwinding of DNA. Notable exceptions can be found among the chemicals that produce DNA crosslinks that may interfere with the unwinding process and those compounds that affect DNA only during its synthesis, such as metabolic inhibitors, which may cause DNA damage after prolonged exposure allowing for the synthesis of DNA in a majority of the cells. 2. Materials 1.

2.

1, 2. 3.

2.1. Metabolic Activation Solution S9 fraction: S9 fraction is the supernatant obtained after centnfugation at 9OOOgof a rat liver homogenate from male Sprague-Dawley rats pretreated with Aroclor 1254. The S9 fraction may be prepared in-house according to Ames et al. (16), or purchased, e.g., as a freeze-dried preparation that is temperature stable, from Molecular Toxicology (Annapolis, MD). The S9 fraction is stored at -70°C or lower and is stable for 1 yr. The freeze-dried preparation can be stored in a refrigerator for up to 2 yr. The metabolic activation capacity of the preparation is normally checked by testing reference mutagens m the Ames test (16). In addition, the protein content should be measured to give an mdication of the concentration of the metabolizing enzymes in the preparation. S9-nnx: Prepare a cofactor mix containing NADP (8 mg/mL) and sodium isocitrate (15 mg/mL) in Fischer’s medium without serum. Add the cofactor mix to the S9 m the proportion of 3: 1 and filter sterilize through a 0.45 nm Millipore filter. The S9-mix should be kept on ice at all times and used within 4 h of preparation. 2.2. Equipment Microbiological safety cabinet. Incubator, 37°C. M&pore filters (0.45 pm pore size).

DNA Alkaline 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Unwinding

Genotoxicity

Test

259

Culture flasks or tubes. Dark cupboard. Soniprep 150 or equivalent somcator. Freezer -20°C. 5 mL glass columns or plastic syringes. Glass wool. Watson Marlow 502 S peristaltic pump with three 501-DX multichannel delta pumpheads: provides 30 channels in all. Scintillation counter. Refrigerated table-top centrifuge. Block heater Grant BT3 with block removed and 13 mm holes bored through the bottom. Microscope equipped with phase contrast optics, 100x magnification. 2.3. Reagents

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

13.

Aroclor 1254. [3H-methyl]thymidme, specific activity 5 Ci/mmol. Scintillation cocktail (Instagel). Hank’s balanced salt solution (HBSS), without phenol red. Sodmm dodecyl sulfate. BIO-GEL HTP hydroxyapatite. 50 mM solution of NaH2P0,. Trypan-blue solution: 0.5% in 0.9% saline, should be filtered after preparation. Complete culture medium: Fop supplemented with (final concentrations shown): 1 mM pyruvate, 0.1% Pluronic F68, 10% horse serum, 20 U each of penicillin and streptomycin, Alkali solution: 20 mM NaOH and 1M NaCl: Made up fresh on day of use from a stock solution of 1M NaOH, kept at 4°C for up to 1 mo, and solid NaCl. Sodium phosphate buffer: 10 mM sodium phosphate buffer, pH 6.8, prepared by mixing 10 mM NaH,P04 and 10 rniV Na2HP0, until the desired pH is reached (can be stored for up to 1 mo at 4OC). Potassium phosphate buffer: 60 mM potassmm phosphate buffer, pH 6.8 and 160 mM potassium phosphate buffer, pH 6.8. Prepare by mixing 60 mM KH2P04 and 60 mM K2HP04, or 160 mM KH2P04 and 160 mJ4 K2HP04, respectively, until the desired pH is reached (can be stored for up to 1 mo at 4°C). Solvents: Test chemicals should be dissolved in either Fop, water, ethanol, or dimethyl sulfoxide at 100X the greatest concentration to be tested and appropriate dilutions made. (Generally, the maximum concentration of

Bolcsfoldi solvent allowed in the culture medium should not exceed 1:lOO.) If the solubility does not enable a 100X concentrated solution to be made, a 5X concentrated solution should be made up in Fop. The protocol is then modified as follows: 0.5 mL of testcompound solution is added to 1.OmL cells at a density of 6 x 105/mL and 1.0 rnL of Fop or 1.OmL of S9-mix.

3. Methods The performance of the alkaline unwinding test is based on the work of Ahnstrijm and Erixon (13). The test has been developed as a screen for genotoxicity as a predictor of the mutagenic activity of chemicals in the mouse lymphoma thymidine kinase locus assay (14). As such, the cells and treatment conditions were chosen to mimic those used in this assay. However, the method has been used for the measurement of DNA damage in other cells and in tissues (14). 3.1. Cells and Maintenance Mouse lymphoma L5178Y/TK* cells may be obtained from the American type culture collection ATCC CRL 95 18. They are grown in Fischer’s medium supplemented with 10% horse serum, 1 mM sodium pyruvate, 0.1% Pluronic F68, 1.13 g/L of sodium bicarbonate, and 20 U each of penicillin and streptomycin/L. The cells can be grown in stationary cultures in tissue culture flasks or in tissue culture tubes rotated on a roller under an atmosphere of 5% CO,/95% air at 37°C or in closed flasks that have been equilibrated with this gas mixture. The doubling time under these conditions is 10-l 1 h and the cultures should be diluted when the cell density reaches lo6 cells/ml. 3.2. Test Performance 1. The day before the cells are to be exposed to the test compound they are allowed to incorporate radioactively labeled thymidine into their DNA. Approximately 40 rnL of cell culture at a density of 2 x lo5 cells/ml are incubated with 0.3 pCi of (methyl-3H) thymidine overnight, centrifuged for 10 min at 2OOg,after which the radioactive medium is removed, The cell pellet is resuspended by repeated pipetmg in 5 mL Hank’s basal salt solution (HBSS) and centrifuged once more. The cells are resuspended in 40 rnL, of warm Fischer’s medium as above but containing 5% horse serum at a density of approx 4 x lo5 cells/ml and incubated for l-2 h. 2. The culture is then subdivided into 1.5 mL aliquots to which either 1 mL of Fischer’s medium without serum or 1 rnL metabolic activation mix (S9mtx) is added followed by 25 pL of a 100X concentrated solutron of the

DNA Alkaline

Unwinding

Genotoxicity

Test

test compound. Incubation is for 3 h, after which time an 0.5 mL sample is removed from each culture for viability determination and kept on ice for up to 2 h. The remaining 2 mL of culture are centrifuged as above at 4°C and the supernatant is discarded. The pellet is washed by resuspension in HBSS at 4OC and the cells are centrifuged once again. The washing is repeated one more time. The cells are then resuspended in 50 p,L HBSS pipeted forcefully mto 2 mL of a solution of 20 mM NaOH and 1M NaCl to lyse the cells and start DNA unwinding. The cultures are immediately transferred to a dark place without undue shaking and allowed to remain for 30 min at room temperature. The unwinding process is stopped by the addition of approx 1 mL of a 50 mM solution of NaH2P04 to neutralize the solution to pH 6.8 and the cultures are sonicated for 15 s at an amplitude of 10 CL,to hinder the reannealing of single-strand fragments. After sonication 2.5% sodium dodecyl sulfate (SDS) is added to a final concentration of 0.25% to denature proteins in the solution. At this stage, the samples can be frozen at -20°C until hydroxyapatite separation is to be performed. 3.3. Hydroxyapatite Separation Separation of single- and double-stranded DNA is achieved by adsorbing the DNA in the sample to hydroxyapatite, a calcium phosphate salt, to which charged macromolecules are bound and can be eluted with buffers of different ionic strength (15). 1. Prepare 5 mL glass columns or syringes by plugging with glass wool. 2. Form columns of hydroxyapatite by suspending 0.6 g of hydroxyapatite in 10 rnM sodium phosphate buffer, pH 6.8, and allowing to settle in the column. These are held at 60°C in a modified block heater with holes in the bottom. Air bubbles may form in the hydroxyapatite after it is heated. These decrease the reproducibility of the elution and should be removed by gentle stirring of the hydroxyapatite with a Pasteur pipet or thin glass rod. 3. The samples of DNA are thawed, and, if necessary, are heated to 60°C to dissolve any precipitated SDS and are allowed to elute through the columns. 4. The columns are then washed twice with 5 mL of 10 mM sodium phosphate buffer that 1sdiscarded. 5. Single-stranded DNA (ssDNA) is eluted by adding 8 mL of a 60-100 mZt4 solution of potassium phosphate buffer, pH 6.8, and double-stranded DNA (dsDNA) with the same volume of a 160-220 mM buffer. The columns can either be eluted by gravity or by using a peristaltic pump, which gives a more even flowrate. The flowrate should be 0.4 mL/min or less. The concentration of the potassium phosphate buffer that gives the best separation of ss- and dsDNA varies for different

Bolcsfoldi

batches of hydroxyapatite. This should be determined by eluting a control sample (see below) with stepped concentrations of potassium phosphate across the range lo-220 mM. For routine separations, the molarity of the first buffer is selected to be that which elutes all ssDNA and none of the dsDNA, whereas the second buffer should elute all of the dsDNA. 3.4. Measurement of DNA The relative amount of DNA in the ssDNA and dsDNA fractions is quantitated by measuring the radioactivity in an 0.5 rnL aliquot of the ss- and dsDNA fractions. Samples of the eluates are mixed with 3.8 mL scmtillation cocktail, thoroughly mixed, and the radioactivity is measured using a liquid scintillation counter. Nonradioactive methods may also be used in the cases where nondividing cells are used (12). 3.5. Measurement of Viability Viability is determined by staining with Trypan blue, a stain that is excluded by cells having an intact cell membrane. Trypan blue solution is mixed with the cell sample in equal proportions and is allowed to stand for a few minutes on ice before the live and dead cells are counted under the microscope using a hemocytometer. The samples are counted in order of decreasing concentration of the test compound, to avoid cells dying while waiting to be counted. Thus the samples in which the cells have the highest viability are counted last. 3.6. Test Protocol Two separate trials are normally performed on test compounds that have not been previously tested. Each trial consists of a series of treatments without and with metabolic activation. Duplicate cultures treated with the solvent in which the test compound is dissolved and a single culture treated with a known DNA-damaging chemical are prepared without and with metabolic activation. Suggested positive controls and final concentrations are 5 p,M of 4-NitroquinolineN-oxide, which is a direct acting genotoxin, and 100 piI of 9,10dimethyl-1,2-benzanthracene, which requires the presence of S9-mix for activity, Note: These are potent carcinogens that should be handled and disposed of accordingly!

DNA Alkaline

Unwinding

Genotoxicity

263

Test

Single cultures are prepared for each concentration of the test chemical without and with metabolic activation. The concentrations of the test chemical are chosen to encompass a wide range, providing data on both toxicity and DNA damaging effects. The highest concentration should be at least 10 m.M, which has been shown to be required for the detection of some genotoxins (14). Eight threefold dilutions of this concentration can be used routinely to test compounds of unknown toxicity. The concentrations in the second trial should provide at least three data sets in the interval giving 5-50% increases in relative toxicity. 3.7, Calculation

and Interpretation

of Results

Concentrations of chemicals that are toxic will generally produce some degree of DNA damage irrespective of whether the chemical is genotoxic. To be able to differentiate between the two categories of compound, a method was devised that compares the relative effect of the test compound on cell viability and DNA integrity. The validity of this method is based on data obtained from the testing of a large number of chemicals with a wide spectrum of structures and mechanisms of action (14). The quantitative comparison of these effects is calculated as follows. 3.7.1. Relative Toxicity The viability of the control and treated cells is calculated as the fraction of the total number of cells that is counted, which is not stained by Trypan blue. The relative toxicity is expressed as viability of the controls minus viability of the treated cells at each concentration. The values are expressed as percentages. of ssDNA The fraction of ssDNA in the controls and treated cultures is calculated as a percentage of the total DNA present. The relative fraction of ssDNA is expressed as the fraction of ssDNA m the treated - control cultures: 3.7.2. Relative

Fraction

[(ssDNA)/(ss + dsDNA) treated- (ssDNA)/(ss + dsDNA) control] x 100 The effects of a test compound that is both toxic and induces DNA damage can be quantitatively compared since both the relative toxicity and relative fraction of ssDNA will show increasing values, A decision on whether to classify a chemical as genotoxic or not is arrived at using the following criteria: By the inclusion of duplicate control cultures, con-

Bolcsfoldi fidence intervals for viability, and fraction of ssDNA have been established based on a large number of experiments (14). 3.7.3. Statistical Criteria Each laboratory should establish statistical criteria for the test as follows. Duplicate solvent controls are included without and with metabolic activation in each experiment. The viability and fraction of ssDNA are calculated. The distribution of the differences between the duplicates should follow a one-tailed normal curve. The standard deviation of the differences between duplicates is calculated and used as the basis for deciding whether a chemical has a true effect, i.e., an increase of twofold or more of the standard deviation of the duplicates represents a true effect of the test compound at the probability level of p c 0.05. 3.7.4. Positive Effect For a compound to be considered to have a positive DNA damaging effect the following criteria must be fulfilled: 1. The relative fraction of ssDNA was significantly increased. 2. The relative toxicity at the corresponding concentration of the test compound was less than the relative fraction of ssDNA. 3.7.5. Negative Effect For a compound to be considered to have a negative DNA damaging effect the following criteria apply: 1. The relative fraction of ssDNA was not increased. 2. The relative fraction of ssDNA was significantly increased but was less than the relative toxlclty at the correspondmg concentration. Experience has shown that the fraction of ssDNA reaches a maximum at approx 50%, whereas the relative toxicity can be 100%. Therefore, values obtained at relative toxicities above 50% are not included in the evaluation. A compound is considered to be genotoxic if: 1. It has a positive effect, as described above, in two independent trials. 2. The effect is dose related in at least one of the trials. A compound is classified as nongenotoxic if: 1. There is no significant increase in the relative fraction of ssDNA and a significant increase in relative toxicity is found; or

DNA Alkaline

Unwinding

Genotoxicity

Test

265

2. There IS no significant increase m the relative fraction of ssDNA at a concentration of the test compound of at least 10 rnM. 3. If the highest concentration tested was below 10 mM, the test compound is classified as nongenotoxic up to the limits of solubility.

References 1. Scott, D., Galloway, S. M., Marshall, R. R., Ishidate, M., Jr., Brusick, D., Ashby, J., et al. (1991) Genotoxicity under extreme culture conditions. A report from ICPEMC Task Group 9 Mutut. Res 257,147-204. 2. Kohn, K. W. (1979) DNA as a target m cancer chemotherapy: measurement of macromolecular DNA damage produced in mammalian cells by anticancer agents and carcinogens, in Methods m Cancer Research, vol. XVI. Cancer Drug Development, Part A (De Vita, V. T. and Busch, H., eds.), Academic, New York, pp. 291-345. 3. Brrnboim, H. C. and Jevcak, J. J. (1981) Fluorometric method for rapid detection of DNA strand breaks in human white blood cells produced by low doses of radiation Cancer Res 41,1889-1892 4 Morris, S. R. and Shertzer, H. G. (1985) Rapid analysis of DNA strand breaks in soft tissues Envzron Mut. 7,871-880 5. Freeman, S. E., Blackett, A. D , Monteleone, D. C., Setlow, R. B., Sutherland, B. M., and Sutherland, J C. (1986) Quantitation of radiation-, chemical-, or enzymeinduced single strand breaks in nonradioactive DNA by alkaline gel electrophoresis. application to pynmidine dimers. Anal Biochem. 158, 119-129. 6. Snyder, R. D. and Matheson, D. W (1985) Nick translation-A new assay for monitoring DNA damage and repair in cultured human fibroblasts Environ. Mut 7,267-279 7 Chow, S C., McConkey, D. J., Orrenius, S., and Jondal, M. (1989) Quantitation of DNA fragmentation using fiberglass filters. Anal Biochem. 183,42-45. 8. Olive, P. L., Chan, A P. S., and Cu, C. S. (1988) Comparison between the DNA

precipitation and alkali unwinding assays for detecting DNA strand breaks and cross-links. Cancer Res. 48,6444-6449. 9. Singh, N. P., McCoy, M. T , Tice, R. R , and Schnerder, E. L. (1988) A simple technique for quantitatron of low levels of DNA damage in individual cells. Exp. Cell. Res 175, 184-191. 10. Ahnstrom, G. and Erixon, K (1973) Radiation induced strand breakage in DNA from mammalian cells. Strand separation in alkaline solution. Znt J. Radiut. Biol 23(3), 285-289

11. Rydberg, B. (1975) The rate of strand separation in alkali of DNA of irradiated mammahan cells Radiat Res 61,274-287 12. Bradley, M. 0. and Sina, J F. (1984) Methods for detecting carcinogens and mutagens with the alkaline elution/rat hepatocyte assay, in Handbook of Mutagenicity Test Procedures, 2nd ed. (Kilbey, B. J., Legator, M., Nichols, W., and Ramel, C., eds.), Elsevier, Amsterdam, pp. 77-82. 13. Ahnstriim, G. and Erixon, K (1981) Measurement of strand breaks by alkaline denaturation and hydroxyapatite chromatography, in DNA Repair. A Laboratory

266

Bolcsfoldi

Manual of Research Procedures (Priedberg, E. C. and Hanawalt, P. C., eds.), Marcel Dekker, New York, pp 403-419 14. Garberg, P., Akerblom, E.-L, and Bolcsfoldi, G. (1988) Evaluatton of a genotoxicity test measuring DNA-strand breaks-m mouse lymphoma cells by alkaline unwinding and hydroxyapatite elution. Mutat. Res. 203, 155-176. 15 Bernardi, G. (1971) Chromatography of nucleic acids on hydroxyapatite columns, m Methods in Enzymology, vol XXI (Grossman, L. and Moldave, K , eds ), Academic, New York, pp. 95-126. 16. Ames, B. N., McCann, J., and Yamasakt, E. (1975) Methods for detecting carcinogens and mutagens with the Salmonella/mammal~an microsome mutagenicity test. Mutat. Res 31,347-364

CHAPTER30

Measurement of Unscheduled DNA Synthesis In Vitro Using Primary Rat Hepatocyte Cultures Steve Dean 1. Introduction The in vitro unscheduled DNA synthesis (UDS) assay performed in primary cultures of rat hepatocytes is a useful screen for DNA damage and repair. Although it gives no direct information about the number or type of DNA lesions or about the consequences of repair, it does detect repair of the type of DNA lesions (adducts) caused by chemical carcinogens, many of which are thought to act by interaction with macromolecules within the cell, particularly DNA (1). This interaction results from the metabolism of the compound in vivo by, for example, liver monooxygenases, which produces a reactive electrophilic species. Such molecules may interact with nucleophilic sites within the cell (2). Since animals and humans are being exposed continuously to low levels of DNA-damaging agents, the cell has developed DNA repair systems in order to maintain the integrity of the genome. Cancer is thought to be a somatic mutational event arising either directly or indirectly from DNA damage. Agents that damage DNA, therefore, have the potential of being carcinogenic. Examination of the cell for DNA repair can provide a means of studying carcinogen-DNA interactions indirectly and thus provides a useful in vitro test. This is the principle behind the in vitro rat hepatocyte UDS assay. Hepatocytes are isolated from the livers of rats after dissection, by From Methods m Molecular Bology, Edlted by S. O’Hare and C K Atterwlll

Vol 43 In Wtro Toxmty Testmg Protocols Copynght Humana Press Inc , Totowa, NJ

267

perfusion with collagenase. Primary hepatocyte cultures are then exposed to test chemical in the presence of [3H] thymidme that is incorporated into the DNA if UDS is occurring. Normal S-phase synthesis is rare in hepatocytes and can be distinguished readily from UDS autoradiographically. Incorporation is followed by autoradiography of the hepatocytes and grain counting. The technique described here was developed by Williams (3) and evaluated more recently by Ashby et al. (4). The advantage of using hepatocytes over other cell types is that hepatocytes tend to be nondividing, which reduces the incorporation of radiolabel because of semiconservative DNA synthesis. Furthermore, these cells contain the enzymes necessary for the metabolic activation of many compounds without the need for exogenous activation systems.

2. Materials 1. Hepatocyte buffer I : 150 mkf NaCl, 3.73 mM NaI-IC03, 4.84 mM Na2HP04, 4.97 mM KCI, 1.24 n-&f KH2P04, 0.62 mM MgSO‘,, 0.62 mM MgQ, 10 pg/mL phenol red indicator. 2. Hepatocyte buffer 2: 142 mJ4 NaCl, 24 mM NaHC03, 4.37 mM KCl, 1.24 mM KH2P04, 0.62 nuJ4MgS04, 0.62 mM MgC&, 10 pg/mL phenol red indicator. 3. 769 miJ4 calcmm chloride solution, 4. 50 mg (approx 75 U) collagenase A (batch testing IS advised to ensure high viabihty preparations). 5. Williams medium E-complete (WE-C): Williams medium E with 2.2 mg/ mL sodium bicarbonate, supplemented to 4 mM with L-glutamme, 100 pg/ mL gentamycin, and 10% fetal calf serum. 6. Williams medmm E-incomplete (WE-I): Williams medium E with 2.2 mg/ mL sodium bicarbonate, supplemented to 4 mM wrth L-glutamme and 100 pg/mL gentamycin. 7. 0.4% (w/v) Trypan blue. 8. 18 and 14 G catheters. 9. Suture thread. 10. 70% (v/v) ethanol. 11. Variable flow peristaltic pump (e.g., Autoclude). 12. Sterile 150 pm mesh nylon boltmg cloth. 13. Sterile 50-mL centrifuge tubes. 14. Hemocytometer. 15. Binocular microscope.

Unscheduled 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 3 1. 32. 33. 34. 35. 36. 37. 38. 39.

DNA Synthesis

269

Dimethyl sulfoxide (DMSO). 5% (v/v) CO, in air gassed incubator set at 37°C. 6-well sterile multiwell plates or 3 cm Petri dishes. Coverslips, 25 nUJ4diameter round, Thermanox. r3H]-thymidine, 1 mCi/mL. Phosphate-buffered saline (PBS). Freshly prepared 1:3 (v/v) glacral acetic acid:absolute ethanol. 2-acetamrdofluorene (ZAAF). Distilled water. Microscope slides. DPX mountant. Glycerol. Ilford K2 emulsion. Darkroom safelight, Ilford 902 filter, or equivalent. 100 mL measuring cylmders cut to 30 mL mark. Silica gel. Kodak D 19 developer. Ilford Hypam fixer. Meyers hemalum (1 g hematoxylin dissolved in 800 mL distilled water with warming, cooled, supplemented with 50 g ammonium alum, 0.2 g sodium iodate, 1 g citric acid, and 50 g chloral hydrate and made up to 1 L). 1% (w/v) aqueous eosin Y. 90% (v/v) ethanol. Absolute ethanol. Xylene. Coverslips, 22 x 50 mm.

3. Methods 3.1. Perfusion of the Liver 1. The rat is anesthetized with ether or an appropriate anesthetic to prevent any possibility of recovery. The surface of the abdomen is then disinfected by swabbing with 70% (v/v) ethanol. 2. A “V”-shaped incision is made with scissors,through both skm and muscle from the center lower abdomen to the ribcage. The skin and muscle are then removed. 3. The inferior vena cava is located and is clamped superior to the kidney. 4. A cotton tie is placed loosely around the hepatic portal vein, and the vein is cannulated wrth a suitable size (e.g., 18 G) catheter. The inner needle is removed and the hgature trghtened.

270

Dean

5. The dtaphragm is cut and the ribcage removed. A cotton tie 1s placed loosely around the superior vena cava and the vein IS cannulated wrth a suitable size (e.g., 14 G) catheter. The inner needle is removed and the ligature tightened. 6. The hepatic portal vem cannula 1sconnected to a variable flow perrstaltic pump and the liver flushed through with Buffer 1 at a flowrate of approx 40 mL/min. 7. The vena cava cannula is connected to a waste lme and the liver, with gentle massage if required, is washed free of blood. 8 After approx 400 mL Buffer 1 has flowed through, the liver is perfused with Buffer 2 (40 mL/min) until about 200 mL 1s left. Approximately 50 mg collagenase IS dissolved m 1 mL 769 mM calcmm chlortde, and 10 mL of Buffer 2 is added to this. The collagenase solutton IS then poured into the remaining reservoir of Buffer 2. After approx 1 min or when the darker red color caused by the collagenase has reached the liver, the waste lme IS placed directly into the Buffer 2 reservoir so that the solution rectrculates and the pump speed IS reduced to approx 20 mL/min. 9. When the reticular pattern of the liver begins to break up and the liver becomes “spongy” the perfusion is stopped. This generally takes about lo-15 min. The liver is then cut from the carcass and placed in a beaker with approx 10 mL of the recirculated Buffer 2 containing calcium and collagenase. 1. 2. 3.

4. 5. 6.

3.2. Primary Culture of Hepatocytes The liver is transferred to a sterile, plastic Petri dish, cut open, and the hepatocytes are carefully teased out using a spatula. All Williams-E media should be prewarmed to 37OC. The resulting hepatocyte suspension ISgently filtered through 150 pm nylon bolting cloth mto a glass beaker. The residue 1sthen washed through the bolting cloth with WE-C to give a final volume of approx 120 mL; approx 40 mL ahquots of this well mixed suspension are decanted into each of three 50-mL polycarbonate centrifuge tubes. The cell suspension is centrrfuged at approx 40g for 2-3 min at room temperature to pellet the hepatocytes. The resultant pellets are gently resuspended in approx 20 mL WE-C, The centrifugation and resuspension steps are performed twice more. The three pellets are then pooled in approx 50 mL of WE-C and a 0.5 mL aliquot of hepatocyte suspensiondiluted with an equal volume of 0.4% (w/v) Trypan blue.

Unscheduled

DNA Synthesis

271

7. The suspension 1scounted on a hemocytometer. The nuclei of nonviable cells are stained blue, viable cells appear yellowish. Four different areas are counted and the viable and total cell counts averaged. The percentage viability = No. cells not containing dye/Total no. of cells counted x 100 8. If necessary, hepatocytes can be obtained from further ammals and liver homogenates may be pooled. Vlabilities should be rechecked to ensure enough viable cells are available for culture preparation. Note: Hepatocyte suspensions yielding ~70% viable cells should be reJected. 9. Hepatocytes are plated mto 6-well multiplates or 3-cm dishes at a density of 1.5 x lo5 viable cells/ml of WE-C, 3 rnL/dlsh/well. The quantity of diluted cells prepared should always be more than is required. a. The total number of viable cells required (D) is calculated: = B x 3 (mL) x 1.5 x lo5 (cells/mL) (D) where B = the number of cultures for which dilution is being prepared. b. The number of viable cells present/ml of hepatocyte preparation is calculated: =Ax2x

lo4

where A = the viable cell count of 0.1 L of cell suspension/Trypan blue mix. c. The total volume of the hepatocyte preparation needed 1scalculated: = D/(A x 2 x 104) d. This volume of hepatocyte suspension is then diluted with WE-C to provide the required volume (3 x B) of cells at 1.5 x 105/mL. 10. Three milliliters of this suspension 1splated out into setsof nine 3-cm Petri dishes/wells, each containing a 25 mm round plastic Thermanox coverslip, for each dose of test solution. Thirteen dishes/wells containing coverslips are set up for the solvent controls and a further six for the positive control. Note: The coverslips should be pressed down to exclude air. 11. These cultures are placed in a 5% CO* gassedincubator at 37°C for approx 1S h to allow cell attachment.

3.3. Treatment of Hepatocyte Cultures This procedure should be carried out aseptically. Cultures are prepared for UDS analysis with radioactivity present. Additional cultures to allow survival determination are prepared without radiolabel.

272 1. Test chemical is weighed out and dissolved in the appropriate solvent (usually anhydrous dimethylsulfoxide) at its solubility limit or 500 mg/mL. Normally the formulated test agent is filter-sterilized. A series of fivefold dilutions are made from this, to grve a total of 8 concentrations. When diluted l/100 m the test cultures, the final concentrations of the test compound will then be: 5000, 1000, 200,40,8, 1.6,0.32,0.064 l.tg/rnL. Alternatively, the test agent may be made up at another multiple of its final concentration (e.g., x20) to a maximum of 5% (v/v) of test solution added, if this yields a higher final concentration than the method above. The negative control is the chosen solvent. The positive control is usually 2-AAF, although other suitable compounds can be used. Typically, solutions of 0.25 mg/mL and 0 5 mg/mL are prepared, but it is good practice to check the efficacy of each batch of the control chemical All solutions are prepared within 4 h of treatment. 2. For UDS cultures, using normal radiochemical safety precautions; for each dose, a labeled, sterile, 50-n& Falcon tube is prepared contaming WE-I, [3H] thymidine (at 10 uCi/mL final concentration) and either the test compound dilution, solvent, or positive control dilution (at 1% v/v). Media containing [3H] radtolabel are kept in labeled trays lined with absorbent material, e.g., Benchkote. All radioactive waste solutions are collected in a vessel held within a lined tray and are subsequently disposed of down a sink designated for radiochemical use. Radtoacttve chemical disposal forms should be completed, e.g., to prepare a final volume of 20 mL: a. 19.6 mL WE-I; b. 0.2 mL [3H] thymtdine; c. 0.2 mL test compound dtlution/solvent/posittve control. For survival determmation, for each dose a labeled, sterile, 50-mL Falcon tube is prepared containing WE-I and either the test compound dilution or solvent (at l%, v/v). Positive control cultures are not required for viability checks, e.g., to prepare a final volume of 10 mL. a. 9.9 rnL WE-I; b. 0.1 mL test compound dilution/solvent. 3. The medium is removed from the hepatocyte cultures prepared approx 90 min earlier. 4. The hepatocytes are washed with 2 mL of prewarmed WE-I. 5. From each tube, 2 mL of the treatment media is dispensed into each well of the corresponding plate. 6. Cultures are incubated overnight (at least 12 h) at 37°C.

Unscheduled

DNA Synthesis

273

3.4. Assessment of Viability of Treated Hepatocytes 1. Following overnight incubation of survival plates containing treated hepatocytes, 1 mL of medium is removed from each dish, and 1 mL of 0.4% (w/v) Trypan blue is added to the 1 mL of medium remaining. 2. After a minimum of 5 min, the Trypan blue/ medium is replaced with 2 mL PBS and the cells are rinsed with gentle agitation. 3. Starting with the negative control and progressing down the dose range from the highest dose, the coverslips are scored for viable and total cells. Each dish is placed on an inverted microscope fitted with a suitable graticule. Areas are selected at random, although regions of very sparse or very heavily populated coverslip are avoided. Four areas are scored for viable (unstained) and total (unstained plus stained) hepatocytes. 4. Mean counts and viabilities for each dish are calculated, and the mean viability per treatment is compared to that of the negative control. Doses are scored until the nontoxic dose range is established. 3.5. Fixation of Hepatocytes and Slide Preparation 1. Hepatocytes are attached onto round coverslips having been exposed to tritiated thymtdme the previous day. The medium is removed and the coverslips are washed with 2 mL PBS, then 2 mL of a freshly prepared mixture of 1:3 glacial acetic acid:absolute ethanol (v/v) is added to each plate or well for approx 10 min. Washing with 1:3 glacial acetic acid:absolute ethanol 1srepeated twice more. 2. Wells/dishes are washed 4 times with distilled water. 3. The coverslips are dried by placing them on the edge of the appropriate plates/dishes or on any clean surface in dust-free conditions, e.g., in paper towel-lined plastic boxes with lids. 4. When dry, coverslips are mounted, cell side up, onto mtcroscope slides with DPX, approx 0.2 cm from the unfrosted end of the slide that bears the study details. 5. Slides are left flat overnight for the DPX to set, m dust-free conditions, prior to autoradiography.

of Fixed Hepatocyte Slides All stepsinvolving emulsion areperformed in a darkroom under safe light conditions. Usually only 3 of the 6 available treatment or positive control slides and 5 of the 10 solvent control slides are prepared. The remainder are held in reserve. 3.6. Autoradiography

1. Twelve milliliters distilled water is placed into a 100 mL measuring cylinder cut off at the 30 mL mark. 0.5 mL glycerol is added, and the cylinder is then placed in a water bath at 43°C.

Dean

274

2. Under safe lighting, enough Ilford K2 emulsion is removed from the stock bottle, with a plastic teaspoon, to reach approx the 20-25 mL mark on a second, cutoff measuring cylinder. 3. The cylinder containing the emulsion is placed at 43°C for 10 min, with periodic gentle stirring with a clean nonferrous implement. Care is taken during stirring not to cause bubbles to form m the emulsion. 4. Both measuring cylinders are removed from the water bath, and the sides are wiped to remove water. The molten emulsion is gently poured into the cylinder containing water/glycerol until the 25 mL mark is reached. The cylinder is then returned to the water bath for 2 min and is stirred very gently with a clean nonferrous implement. 5. The emulsion mixture is removed from the water bath. Each slide is kept vertical and dipped into the emulsion, then withdrawn slowly and steadily. It is then held vertically for 2 s and the underside then wiped with a paper tissue. If dipping takes longer than approx 15 min, the emulsion may be replaced in the water bath to prevent excessive thickening. 6. Each slide is placed face up on a tray set on ice for at least 10 min. 7. When all the slides are gelled, they are placed in light-proof boxes and allowed to dry for at least 90 min at room temperature. 8. In a darkroom, under safe light conditions, a small quantity of silica gel is added to slide boxes, behind a clean microscope slide Inserted about 1 cm from one end. Slides are slotted into the boxes, which are sealed, labeled with the date and study details, and stored in a refrigerator at 4°C. 9. Autoradiographs are exposed at 4°C for approx 12-14 d.

and Staining

3.7. Development of Exposed Hepatocyte

Slides

1. After autoradiography exposure time, hepatocyte slides are developed and stained. Treatments are most easily accomplished in staining troughs containing the appropriate solutions. Until the emulsion is fixed, these procedures are carried out in a darkroom under safe light conditions. Development is achieved by immersing racks of slides as follows: a. 5 min in Kodak D19 developer diluted 1:l (v/v) with distilled waterdeveloper temperature between 19 and 21°C (this is critical). b. 1 min distilled water wash. c. 5 mm m 1:3 (v/v) Ilford Hypam fixer:distilled water. d. 10 mm tapwater wash. 2. Cell staining is performed directly after developing slides treated as follows: a. 3 min in Meyers hemalum. b. Gentle rinse in tapwater until nuclei are just pale blue. If staimng is inadequate, the slides are restained in Meyers hemalum for, e.g., a further minute.

Unscheduled

275

DNA Synthesis

c. Stain in 1% (w/v) aqueous eosin Y for 1.5 mm. d. Wash gently in tapwater (up to 2 min). e. Dehydrate m 70% (v/v) ethanol (approx 2 min)-if counterstain is too weak, the slides may be restained as in stage 2c. f. Dehydrate in 90% (v/v) ethanol (2 min). g. Dehydrate in absolute ethanol (2 min). h. Dehydrate again in absolute ethanol (2 mm). i. Clear twice in xylene (3 min). j. Mount coverslips onto the cell preparations with DPX. All times given are approximate since staining efficiency varies and must be monitored, using a microscope, at stages 2b, 2d, and 2e (in the latter casebecause 70% [v/v] aqueous alcohol weakens the counterstain if left in contact too long). k. Cell preparations are left at least overnight before assessment.

3.8. Scoring

of Slides for UDS

1 Examination of the survival data is used to select treatments suitable for scoring. The top dose is usually 5000 pg/mL, the solubility limit or a dose that reduces survival by 50-80%, whichever is appropriate. This plus four consecutive doses are scored. The positive control slides are examined and the lower dose preferred if this is satisfactory. 2. Scoring of slides is performed using a microscope equipped with at least x10 and xl00 objectives. Positive control slides are scored first to confirm that the test systemis working. Only cells of normal morphology are scored and multmuclear or touching cells should be scored only when it is certain where the cytoplasmic boundaries lie. Heavily labeled cells that are in Sphase should not be scored. Slides are scanned in a manner that ensures that the same cells cannot be scored twice. Using either an eyepiece graticule of concentric circles or an image analyzer, the number of grains over the nucleus and over one or more similarly sized areas of cytoplasm are counted. Between 25 and 100 cells (50 is typical) can be scored on each of the 3 slides and the acceptability of the positive control response is determined. 3. Solvent and treatment slides are scored only after coding by a person other than the scorer. Slides are scored in sequence and the study decoded only when all the slides have been scored. 3.9. Data Evaluation 1. For each cell, the net grain count (NG) is calculated by subtractmg the (mean) cytoplasmic grain count from the nuclear grain count. 2. For each slide, a mean NG count, cytoplasmic count, and nuclear count is calculated.

276 3. For each treatment, usmg the slide means, a mean NG count, cytoplasmic count, and nuclear count IS calculated. 4. A cell is deemed to be in repair if it has a NG count of +5 or more. Solvent control cultures are expected to yield negative NG counts with only a few percent cells having NG counts of +5 or more. Acceptable positive control values should be upward of +5 NG with 50% of cells or more having NG counts of 5 or more. 5. The test compound is clearly negative rf the mean NG is 0 or less and ~20% cells are in repair. A clear posttive response requrres a mean NG of

+5 or more and 20% or more of cells in repair. Any treatment giving a positive mean NG count between 0 and +5 would be regarded with suspicion and may warrant further investrgatton.

3.10. Analysis

of Slides for S-Phase Response

Slides can be scored for the percentage of cells in S-phase by recording the number of nuclei with dense black graining, often completely obscuring

the nucleus.

A total of 1000 or more cells per slide should

be

scored. Reduction in the proportion of S-phase cells, compared with concurrent controls, is indicative of toxic effects. An increase in S-phase cells could suggest mitogenic properties. References 1 Miller, E. C. and Miller, J. A. (1981) Searches for ultrmate chemical carcmogens and their reactlons with macromolecules. Cancer 47,2327-2345. 2 Garner, R. C. (1979) Carcinogen prediction in the laboratory: a personal view, in Long Term Hazards from Environmental

Chemicals: A Royal Society Discusnon,

The Royal Society, London, pp. 121-124 3. Williams, G. M. (1978) Further improvements in the hepatocyte primary culture DNA repair test for carcmogens: detection of carcmogenic biphenyl derrvatrves Cancer Lett. 4,69-75.

4. Ashby, J., Lefevre, P. A., Burlinson, B., and Penman, M G (1985) An assessment of the in vivo hepatocyte DNA repair assay. Mutat. Res 156, 1-18.

CHAPTER31

Gene Mutation in Mammalian Julie

Assays Cells

Clements

1. Introduction Gene mutation assays in cultured mammalian cells may be used to give a measure of the response of the mammalian genome to potential mutagens, and yet they are rapid and simple to carry out when compared to the use of whole mammals. Many mammalian cell gene mutation assays are available, but only four cell lines and three genetic loci are well validated and widely used (I). These are V79 and CHO Chinese hamster cells, human lymphoblastoid TK6 and mouse lymphoma L5 178Y cells, and the following genetic loci: hprt, tk, and the cell membrane Na+/K+ ATPase. The use of cells in suspension culture is advantageous because cell numbers are not restricted by problems of metabolic cooperation between cells. Therefore the mouse lymphoma L5 178Y systems are preferable. The tk and hprt loci are perhaps the most commonly used. The mutation systems work by placing treated cells under selective pressureso that only mutant cells areable to survive. For example, resistance to 6-thioguanine (6TG) results from lack of hypoxanthine phosphoribosyl transferase (HPRT) activity and resistance to 5trifluorothymidine (TFT) from lack of thymidine kinase (TK) activity. Thus, the mutants (HPRTor TK-) do not incorporate the toxic analogs 6TG or TFT and survive in their presence. The HPRT and TK assays detect a different range of mutation types. The hprt gene is X-linked and is not expected to detect large genetic From Methods m Molecular B/ology, Ed&d by S. O’Hare and C. K Atterwlll

Vol 43. In V/fro Toxmty Testrng Protocols Copynght Humana Press Inc., Totowa, NJ

277

Clements

278

changes that involve neighboring genes, since these are likely to be lethal. Nevertheless, hprt is a large gene and changes of 30-40 kb of DNA have been detected (2) as well as point mutations. In contrast, the tk locus is autosomal and two types of TFT-resistant colonies are selected: large colonies and small slow-growing colonies. Molecular analysis has indicated that the large colonies tend to represent events within the gene (basepair substitutions and deletions), whereas small colony mutants often involve large genetic changes frequently visible as chromosome

aberrations (3,4). Thus, in this system, gene mutations within the tk gene (1 l-l 3 kb) and chromosomal events involving the gene may be detected. The TK system has a high spontaneous mutant frequency and because of

the high numbers of cells that can be treated and sampled it is the most satisfactory mammalian cell mutation assay from the statistical point of view. The methods described in this chapter are for the performance of a standard TK assay in mouse lymphoma L5 178Y cells, often the system

of choice, using a fluctuation protocol originally developed by Jane Cole (1-S). For more information

on variations of this assay and on other cell

types, the reader is recommended to consult the literature (I-6). 2. Materials 1. Tissue culture cabinet. 2. Coulter counter (or hemocytometer).

3. Inverted microscope. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

Incubators, non-CO2 and CO*. Rocker platform. Waterbath. Bottle top dispensers. Adjustable micropipets and disposable tips, multichannel pipets. Light box. Osmometer. 96-well plates. Tissue culture flasks (250 mL). Sterile disposable 50 mL conical tubes. Sterile containers (to hold 10 or 100 mL vol). Trifluorothymidine; 0.3 mg/mL in unsupplemented RPM1 medium. Filtered isotomc salme. Phosphate buffered salme (PBS). 4-nitroquinoline l-oxide (NQO) solution.

Gene Mutation

Assays

Chemical 4-Nitroquinoline 1-oxide

(NQO) Benzo(a)pyrene(BP)

279 Table 1 Positive Controls Stockaconcentrations Final concentration s-9 04&L) WmL) 5 0.05 10 0.10 200 300

2.00 3 00

+ +

aAll stock solutions are prepared m anhydrous analytical grade dimethylsulfoxide (DMSO). NQO and BP solutions, if not used immediately, may be stored as frozen aliquots at -8OT.

19. Benzo(a)pyrene (BP) solution (see Table 1). 20. The test chemical: Solutions of the test chemical should be freshly prepared and care should be taken to dispose of any unused solutions safely. The widely accepted maximum treatment concentration for the assay is 5000 pg/mL, 10 mA4, or the limit of solubility or toxicity. Preliminary solubility assessmentsare made, a solvent chosen, and dilutions arranged that allow maximum exposure. The solvents of choice are dimethyl sulfoxide (DMSO), tissue culture medium, and water (or saline) but acetone or dimethylformamide may be used. If other solvents or volumes of the organic solvents exceeding 1% (v/v) need to be used, the effects on viability and spontaneous mutant frequencies will need to be checked first. Ideally, a loo-fold dilution of a solution of the test chemical is used. If the final concentration in the medium exceeds 10 rnA4 then the osmolality of the medium needs to be measured, because fluctuations in osmolality of more than 50 mOsm/kg have been responsible for increased mutation (7). 21. Controls: Negative controls comprise treatments with the selected prime solvent diluted to the same extent (loo-fold) as the test chemical solutions. The positive control chemicals are used as shown in Table 1. 22. Metabolic activation system: The mammalian liver postmitochondrial fraction (S-9) used for metabolic activation may be prepared in house (8,9). However, the enzyme inducer Aroclor 1254 is widely used in this procedure and, m view of the fact that obtaining and disposing of this carcinogen poses problems, it may be preferable to obtain S-9 from a commercial supplier. The batches of S-9 can be stored frozen at -80°C prior to use. In the assay,treatment is carried out both in the absence and presence of S-9 by using the following solutions:

Clements

280 Quantity (mL) Solution A with S-9 1.0 1.0 1.0 2.0

Glucose-6-phosphate (180 mg/mL) NADP (25 mg/mL) 150 rnfV KC1 Rat liver S-9

Solution B without S-9 5.0 -

These solutions (A or B) are used at the rate of 1.0/19.0 mL of cell culture containing the test chemical (to achieve the required final concentration in a total of 20 mL). 24 Growth media: Three types of RPM1 1640 medium are prepared as follows: Final concentratton m Horse serum (heat inactivated) Gentamycin Fungizone Pluronic

RPM1 A

RPM1 10

RPM1 20

0% v/v 100 l.tg/mL 2.5 pg/mL 0.5 mg/mL

10% v/v 100 pg/mL 2.5 pg/mL 0.5 mg/mL

20% v/v 100 pg/mL 2.5 pg/mL -

Batches of horse serum are stored frozen and, before use, should be tested for effects on cell growth, cloning efficiency, and spontaneous and induced mutant frequencies. Heat inactivated horse serum is used when TK mutants are to be selected, m order to eliminate a factor that degrades TFT. The horse serum is allowed to thaw at 37°C in a waterbath, the temperature is then increased to 56OC, and this temperature is maintained for 30 min, with occasional swirlmg of the bottles. The water temperature must be closely momtored throughout. The heat inactivated horse serum is cooled before use and, if not required immediately, may be stored in a refrigerator for up to 2 wk. Prior to use, batches of medium should be tested for acceptable clonmg efficiency. Following preparation, RPM1 medium may be stored refrigerated (except RPM1 20, which must be stored at -20°C) for up to 3 mo. 24. Cell cultures: L5178Y TK+/- mouse lymphoma cells may be obtained from the American Type Culture Collection or from laboratories routinely using the cells. The cultures can be stored as frozen stocks m liquid mtrogen. Each batch of frozen cells should be purged of TK- mutants using “THAG” or equivalent (medium containmg thymidine, hypoxanthme, aminopterin, and glycine), and checked for spontaneous mutant frequency. The cells should also be mycoplasma free. This may be checked usmg

Gene Mutation

281

Assays Table 2 Typical Dose Ranges Concentration of treatment solution

Expenment Range-finder

1 and2

s-9

- and +

- and +

(w/mL)

1.58 5.00 15 8 50.0 158.0 500.0 31.25 62.50 125.0 250.0 500.0

Final concentration WmL) 15.8 50.0 158 500 1580 5000 312 5 625 1250 2500 5000

commercial kits or by sendingthe cells for testing. For eachexperiment, 1 or more vials are thawed rapidly, the cells diluted in RPM1 10 m a flask, and incubated in a humidified atmosphere of 5% (v/v) CO, in air. When cells are growing well, subcultures can be established m an appropriate number of flasks (for example, 6 flasks each containing 100 mL cell culture) .

3. Methods

Following the selection of a suitable vehicle for the test chemical, a preliminary range-finding cytotoxicity experiment is performed to establish an appropriate concentration range for the mutation experiments. The range-finder is performed both in the absence and presence of metabolic activation, since toxicity is often observed at different concentrations under these two test conditions. Normally, at least six doses separated by half-log or twofold intervals ranging down from the solubility limit or 5000 pg/mL would be used in the cytotoxicity range-finder. When the toxic range has been determined, a minimum of five doses are usually selected for the first mutation experiment, ranging from nontoxic to toxic (>lO% relative survival) where possible. Normally, a minimum of four doses are carried through all stages of the assay. A second experiment will be carried out, although not necessarily using the same dose-range as the first experiment. An example of the dose intervals that can be used and the methods of achieving the final concentrations is given in Table 2.

282

Clements

When possible, stock test solutions are filter-sterilized further dilution or before use.

3.1. Cytotoxicity

either before

Range-Finder

Treatment of cell cultures is as described below (see Section 2.2.) for the mutation experiments. However, single cultures only may be used for the range-finder and positive controls do not need to be included. Following treatment, cells are washed and resuspended in 20 rrL tissue culture medium. Cell concentrations are adjusted to 8 cells/ml and, for each dose, 0.2 mL is plated into 32 microtiter wells. The plates are incubated at 37°C in a humidified incubator gassed with 5% v/v COZ in air for a minimum of 3 d. Wells containing viable clones are identified, either under a microscope or by eye, and counted. A mimmum of five doses are then usually selected for the mutation experiment.

3.2. Mutation

Assay

3.2.1. Treatment of Cell Cultures 1, Usually at least lo7 cells are placed in each of a servesof sterile disposable 50 mL centrifuge tubes. Treatment medium may contain a reduced serum level of 5% (v/v) and to achieve this cells may be pelleted by centrifugation, the culture medium

removed,

and the cells resuspended

in a final

volume of treatment medium containing 5% (v/v) horse serum. 2. A suitable volume of solvent, test compound, or positive control solution is added as described in points 20 and 21 of Section 2., and 1.OmL, S-9 mix or 150 mM KC1 added as in point 22 of Section 2., such that each tube is at a final volume of 20 mL in the absence or presence of S-9, and is in dupli-

cate (single cultures may be used for each dose of positive control). 3. After 3 h on a rocker platform at 37”C, the tubes are centrifuged at 200g for 5 min, the cells washed with tissue culture medium or phosphate buffered salme, and resuspended further m 20 mL RPM1 lo/tube. 4. Cell densities may be determined using a Coulter counter or hemo-

cytometer and, where sufficient cells survive, the concentratron adjusted to 2 x 1oQnL. 5. Cells are transferred to flasks for growth, gassed with 5% CO2 in air, and

incubated at 37°C for the expression perrod or diluted to be plated for survival as described in Section 2.2.2.

3.2.2. Plating for Survival Following adjustment of the cultures to 2 x lo5 cells/n-L after treatment, samples from these are diluted to 8 cells/ml as in Table 3.

Gene Mutation

Assays

283 Table 3 Plating for Survival

Initial cell Intermediate cell concentration Dilution concentration Dilution mL A mL medium mLB mLRPMI20 (A) (B) Survival

2 x 105/mLa

0.1

99

2 x 103/mL

0.2

50

Final cell concentration (Cl 8/mL

aWhere fewer than 2 x lo5 cells/ml survive treatment, an alternattve dilution scheme can be adopted to gwe 2 X lo3 cells/ml. Table 4 Plating for Viability Initial cell Intermediate cell concentration Dilution concentration Dilution mL A mL medium mLB mLRPMI20 (A) 09 Viability

lx 104/mL 0.1

99

5 x 102/mL

0.8

50

Final cell concentration 0 8/mL

Using a multichannel pipet, 0.2 mL of concentration C of each culture is placed into each well of 2 x 96-well microtiter plates (192 wells, averaging 1.6 cells/well). The plates are incubated at 37°C in a humidified incubator gassed with 5% v/v CO, in air for l-2 wk. Wells containing viable clones are identified by eye using background illumination and counted. 3.2.3. Expression

Period

Cultures are maintained in flasks for a total of 2 or 3 d during which time the TK- mutation will be expressed. Throughout this period cell densities are maintained at or below 1 x lo6 cells/rnL while retaining a total of at least 1 x lo7 cells/flask where possible. From observations on recovery and growth of the cultures during the expression period, normally at least four test dose levels plus negative and positive controls will be selected to be plated for viability and 5-trifluorothymidine (TFT) resistance. 3.2.4. Plating

for Viability

At the end of the expression period the cell densities in the selected cultures are determined using a Coulter counter or hemocytometer and adjusted to 1 x 104/mL with RPM1 20 in readiness for plating for TFT resistance. Samples from these are diluted to 8 cells/mL as shown in Table 4.

284

Clements

Using a multichannel pipet, 0.2 mL of concentration C of each culture is placed into eachwell of 2 x 96-well microtiter plates (192 wells, averaging 1.6 cells/well). The plates are incubated at 37°C in a humidified incubator gassed with 5% v/v CO2 in air for l-2 wk. Wells containing viable clones are identified by eye using background illumination and counted. 3.2.5. Plating

for 5-Trifluorothymidine

(TFT) Resistance

At the end of the expression period, cell concentrations are adjusted to give 1 x 104/mL (see Section 2.2.4.). TFT (300 pg/rnL) is diluted lOOfold into these suspensions to give a final concentration of 3 Fg/mL. It should be noted that solutions of TFT are light sensitive and unstable in medium at 37°C. They should be stored frozen and thawed immediately before use. Using a multichannel pipet, 0.2 mL of each suspension is placed into each well of 4 x 96-well microtiter plates (384 wells at 2 x lo3 cells/well). Plates are incubated for 11-12 d and wells containing clones are identified as above and counted. In addition, scoring of large and small colonies is recommended since the additional information obtained may contribute to an understanding of the mechanism of action of the test chemical (I). The number of wells containing large colonies and the number containing small colonies can be scored for the negative and positive controls and for doses of test chemical showing a significant increase in mutant frequency over the negative control. In this way small colony and large colony mutant frequencies may be estimated. of Results Software has been developed specifically for the analysis of tk data and is commercially available (e.g., York Electromc Research, Huntington, UK). 3.3. Analysis

3.3.1. Determination

of Survival

or Viability

From the zero term of the Poisson distribution the probable number of clones/well (P) on microtiter plates in which there are EW empty wells (without clones) out of a total of TW wells is given by: P = -In (EW/TW)

The plating efficiency (PE) in any given culture is therefore: = P/no. of cells plated per well and, since an average of 1.6 cells/well are plated on all survival and viability plates, PE

Gene Mutation

Assays

285 PE = P/1.6

The percentage relative survival (%RS) in each test culture will therefore be determined by comparing plating efficiencies in test and control cultures thus: % RS = [PE (test)/PE (control)] x 100 3.3.2. Determination of Mutant Frequency It is usual to express mutant frequency (MF) as “mutants/lo6

viable

cells.” In order to calculate this, the plating efficiencies of both mutant and viable cells in the same culture are calculated: MF = [PE (mutant)/PE (viable)] x lo6

From the formulae given in Section 3.1. and with the knowledge that 2 x lo3 cells are plated/well for mutation to 5trifluorothymidine resistance, PE (mutant) = P (mutant)/2 x lo3 PE (viable) = P (vtable)/l.6

where, in each case, P = - In (EW/TW). Therefore, MF = [P (mutant)/2 x 103] x [1.6/P (viable)] x lo6 = {-ln [EW/TW (mutant)]/-ln [EWAV (viable)]} x 800 3.3.3. Assessment of Statistical Significance of Mutation Frequency Statistical significance of mutant frequencies (total wells with clones)

may be carried out according to the UKEMS guidelines (IO). Thus, the control log mutant frequency (LMF) is compared with the LMF from each treatment dose, and second, the data are checked for a linear trend in mutant frequency with treatment dose. These tests require the calculation of the heterogeneity factor to obtain a modified estimate of variance. 3.3.4. Acceptance Criteria The assay will be considered valid if the following

criteria are met:

1. The mutant frequencies m the negative (solvent) control cultures fall within the normal range (not more than 3 times the historical mean value). 2. At least 1 concentration of each of the positive control chemicals induces a clear increase m mutant frequency (the difference between the positive and negative control mutant frequencies 1sgreater than half the historical mean value)

286

Clements

3.3.5. Evaluation Criteria The test substance will be considered to be mutagenic if: 1. The assay IS valid (see Section 3.4.). 2. The mutant frequency at 1 or more doses IS srgmfrcantly greater than that of the negative control. 3. There is a significant dose relationship, as indicated by the linear trend analysis. 4. The effects described above are reproducible. Results that only partially satisfy the above criteria are dealt with on a case-by-case basis. Similarly, positive responses seen only at high levels of cytotoxicity require careful interpretation when assessing their biological significance.

References 1. Cole, J., Fox, M., Garner, C., McGregor, D. B., and Thacker, J. (1990) Gene mutation assays in cultured mammahan cells, in Basic Mutagenicity Tests, UKEMS Recommended Procedures (Kirkland, D. J., ed.), Cambridge University Press, pp. 87-l 14. 2. Thacker, J. (1985) The molecular nature of mutations in cultured mammalian cells: a review Mutat. Res. 150,431442. 3. Moore, M. M., Clive, D., Hozier, J. C., Howard, B. E., Gail Batson, A., Turner, N. T , and Sawyer, J. (1985) Analysis of TFTr mutants of L5178Y/TK”mouse lymphoma cells. Mutaf. Res. 151, 161-174. 4 Applegate, M. L., Moore, M. M., Broder, C. B , Burrell, A , Juhn, G , Kasweck, K. L., Lin, P.-F., Wadhams, A., and Hozier, J. C. (1990) Molecular dissection of mutations at the heterozygous thymidine kinase locus in mouse lymphoma cells. Proc. Natl. Acad. Sci. USA 87,51-55. 5. Cole, J., Arlett, C. F., Green, M. H. L., Lowe, J., and Muriel, W. (1983) A comparison of the agar cloning and microtitration techniques for assaying cell survival and mutation frequency in L5178Y mouse lymphoma cells Mutat Res. 111,3 17-386. 6 Cole, J. and Arlett, C. F. (1984) The detection of gene mutations in cultured mammalian cells, in Mutagenicity Testing. A Practical Approach (Venitt, S. and Parry, J. M , eds.), IRL, Oxford, pp. 233-273. 7 Brusick, D. (1986) Genotoxic effects in cultured mammalian cells produced by low pH treatment conditions and increased ion concentrations Envwon Mutagen. f&879-886. 8. Ames, B. N., McCann, J , and Yamasaki, E (1975) Methods for detecting carcinogens and mutagens with the Salmonellalmammalian microsome mutagenicity test. Mutat. Res. 31,347-364. 9 Maron, D. M. and Ames, B. N. (1983) Revised methods for the Salmonella mutagenicity test. Mutat. Res. 113, 173-215 10. Robinson, W D , Green, M. H L., Cole, J., Garner, R C , Healy, M J R., and Gatehouse, D (1990) Statistical evaluation of bacterial/mammalian fluctuation tests, m Statistical Evaluation of Mutagenicity Test Data (Kirkland, D J , ed.), Cambridge University Press, Cambridge, UK, pp. 102-140.

CHAPTER32

Measurement of Chromosome Aberrations In Vitro Using Human Peripheral Blood Lymphocytes Richard

Marshall

1. Introduction Chromosomal aberrations are implicated in the induction of both cancer and birth defects (1,2). The types of aberrationthat causehuman disease are usually subtle rearrangements, e.g., reciprocal exchanges, which are not cell lethal. These are frequently difficult to detect microscopically but their presence is indicated by the appearance of more gross structural damage, e.g., deletions. Although these may be of less biological significance because they are rarely survivable for more than a few divisions, they are easier to detect. Many chemical agents are capable of inducing structural and numerical chromosome changes and it is for this reason that the in vitro chromosomal aberration test is usually recommended as part of the test battery required by many regulatory bodies. Human peripheral blood lymphocytes provide a good test system for this assay. They are readily cultured, grow rapidly to provide large numbers of mitotic cells, and maintain a diploid karyotype during short-term culture. The techniques described here have been developed over many years in a large number of different laboratories. The reader is referred in particular to Verma and Babu (3) and Savage (4) for technical details, and to Scott et al. (5) and Preston et al. (6) for discussion of protocol.

From: Methods m Molecular Bjology, Vol. 43 In Wtro Toxmty Testing Protocols Edited by. S. O’Hare and C. K. Atterwlll Copynght Humana Press Inc , Totowa, NJ

287

288

Marshall

2. Materials 1. HEPES buffered RPM1 1640 medium containing 50 pg/mL gentamycm, 20% (v/v) fetal calf serum, and 0.375% phytohemagglutinin (PHA, reagent grade). Stored at 28°C. 2. Sterile physiological saline. Stored at room temperature. 3. Liver S-9. Stored at -90°C. 4. Sterile KC1 solution, 150 mM. Stored at room temperature. 5. Sterile glucose-6-phosphate solution, 180 mg/mL. Stored at -20°C. 6. Sterile NADP solution, 25 mg/mL. Stored at -2OOC. 7. Sterile colchicine solution, 100 mg/mL. Stored at 2-8°C. 8. KC1 solution, 0.075 mM. Stored at room temperature (if sterile). 9. Carnoy’s fixative, 3: 1 methanol:glacial acetic acid. Freshly prepared. 10. Giemsa stain. Stored at room temperature. 11. Phosphate buffer, pH 6.8. Stored at room temperature. 12. DPX mountant. 13. Sterile 10 mL centrifuge tubes, autoclavable dispensers and adjustable pipets with sterile tips, precleaned microscope slides, covershps. 3. Methods 3.1. Collection of BZood Donors used to provide blood for clastogenicity studies should be screened, Young, adult males or females should be used who do not

smoke and who are not exposed to chemicals or radiation through employment or lifestyle. They should be in good health when the blood is taken. All potential donors should be pretested for frequencies of spontaneous chromosome aberrations that should be within the laboratory historical negative control range (see Section 3.10.) and their cells should respond well to PHA, that is, regularly giving yields of mitotic cells (mitotic indices) of 5% or more under experimental conditions. Venous blood is withdrawn into a heparinized (sodium heparin preferred) syringe and stored refrigerated until cultures are established. Blood can be stored under these conditions for several days but to prevent significant loss of viability, should be cultured within 48 h of withdrawal. 3.2. Establishment of Cultures Although it is possible to establish cultures from isolated lymphocytes, it is much more convenient to use whole blood. The most commonly used medium is HEPES-buffered RPM1 1640 supplemented with 20% (v/v) fetal calf serum, gentamycin (an antibiotic), and phytohemagglutinin (PHA). A total of 0.4 mL blood in a final volume (at the start of

Measurement

of Chromosome

Aberrations

289

treatment) of 10 mL will give a good yield of cells (although the relative amount of blood added per culture can be varied). Cultures are set up and treated under aseptic conditions. They are initiated by dispensing 9.0 mL culture medium into each sterile culture tube and adding 0.4 mL blood. Tubes should have conical bottoms and be suitable for centrifugation. The tubes are inverted several times to ensure adequate mixing and then placed in racks on a rocking platform in a 37°C incubator. A rocking platform is not essential but if it is not used then the tubes should be incubated slanting on their sides. Cultures are incubated for approx 48 h before treatment. During this time, the PHA stimulates the T lymphocytes to enter the cell cycle and by 48 h, when treatment commences, the cells have escaped the Go resting phase and are asynchronous because the cells proliferate at different rates. 3.3. Test Chemical The test chemical should, where possible, be dissolved in an appropriate primary solvent prior to addition to cultures. A preliminary solubility assessment is therefore usually necessary. Culture medium is the solvent of choice although if this cannot be used, distilled water or dimethylsulfoxide (DMSO) are the best alternatives with dimethylformamide or acetone as a last resort. Ethanol should not be used. If organic solvents are necessary, their final concentration in the cultures should not exceed 1%. Distilled water can be added to a limit of 20% without marked effects on cell proliferation. The starting volume of the cultures should be adjusted to permit the addition of the necessary aliquot of test chemical preparation at the time of treatment (see Note 1). 3.4. Controls Negative control cultures should receive treatment with the primary solvent diluted to the same extent as the test chemical solutions. Positive controls should also be included for treatments in both the absenceand presenceof S-9.4-Nitroquinoline l-oxide and cyclophosphamide at final concentrations of 5 and 25 pg/mL respectively, are suitable alternatives. Both can be dissolved in DMSO at 100x these concentrations prior to use. 3.5. Metabolic Activation Most chemicals are biologically inert unless metabolically activated. In in vitro tests, the capacity for metabolic activation is normally provided

290

Marshall

in the form of “S-9 mix.” S-9 is the supernatant after centrifugation of a liver homogenate (usually rat) at 9OOOgand comprises microsomes, which carry the enzymes required for metabolic activation and a cytosolic fraction. A source of NADPH is also required and is generated in the S9 mix by cofactors, i.e., NADP and either G-6-P or isocitrate. The rats used to provide the S-9 are induced with (normally) Aroclor 1254 a few days prior to sacrifice. Other inducing agents can also be used, the most common of which is a combination of P-naphthoflavone and sodium phenobarbitone. S-9 is available from a number of commercial sources, e.g., Molecular Toxicology (Annapolis, MD) who will provide a quality control statement for each batch of S-9 they supply. Many different formulae have been suggested for S-9 mix. None can be described as being optimal. The following is a well tried alternative: Glucose-6-phosphate(180 mg/mL) NADP (25 mg/mL) KC1 (150 mit4) Liver S-9

Quantity (mL) 1 1 1 2

The above is used a rate of 0.5 rnIJ9.5 mL of cell culture containing the test chemical (to achieve the required final concentration in a total of 10 mL). Cultures treated in the absence of S-9 receive an equivalent volume of 150 mM KCl. 3.6. Treatment Cultures should be treated in duplicate. It is advisable to set up negative controls in quadruplicate, in which case, cells from the additional cultures can be analyzed in the event of a marginal result to give extra power to the statistical evaluation. S-9 mix or 150 rnA4 KC1 (0.5 mL) is added first, followed by the test chemical at each of its concentrations in the primary solvent. The contents of each tube are mixed by inverting and tapping to dislodge the cell pellet. Any evidence of precipitation (which should redissolve on mixing) when the chemical is added should be recorded (see Notes 2-5). The sensitivity of cells to clastogens in terms of the production of chromosomal aberrations varies according to the phase of the cell cycle during which they are treated. For this reason, continuous treatments (i.e., to the time of harvest) are preferable. This is not a problem for treatment in

Measurement

of Chromosome Aberrations

291

the absence of S-9, but it is not possible for treatment in its presence because over extended periods of time the S-9 itself becomes cytotoxic. An optimal exposure period in the presence of metabolism is 3-4 h. After this time, the cultures are centrifuged (2OOg,10 min), 10 mL sterile saline is added, and the tubes are recentrifuged as before. This washing process is repeated. The addition of 10 mL culture medium (without PHA) and reincubation on the rocking platform completes the process (see Note 6). 3.7. Harvesting

and Slide

Preparation

The appropriate time(s) at which cells should be sampled is far from well established. Cell cycle effects cause the aberration frequency to vary with time after treatment (8) and aberrations continue to be produced at relatively long times after treatment. Ideally, therefore, multiple sampling times should be used, but this would make chromosomal aberration assays prohibitively large. Fortunately, most authors would agree that the majority of clastogenic agents can be detected at an optimal sampling time of about 20 h (approximately one and a half cell cycles) from the beginning of treatment. There is no doubt, however, that for some clastogens 20 h is too early, and many investigators now suggest that a second, later, sampling time is included. For convenience, this can be at 44 h, although the experimental evidence to support this is limited. Cultures must be treated l-2 h prior to harvest with a spindle inhibitor to arrest cells in metaphase. Several spindle poisons are available, e.g., vinblastine or colcemid, but the most commonly used is colchicine, which should be addedto each culture as a 0.1 nL aliquot of a 100 clg/mL solution in water to give a final concentration of 1 p,g/mL. All procedures from this stage to the time of slide making are nonsterile. Cell pellets should be resuspended (as described in Section 3.5.) following the addition of colchicine and the tubes returned to the incubator, Harvesting starts by centrifugation at 200g for 10 min. The supernatant is then carefully removed so as not to dislodge the cell pellet. The cells are resuspended in the remaining culture medium by tapping the tube and 0.4 rnL 0.075M KC1 solution (prewarmed to 37°C) is added. This makes the cells swell, and if hypotonic treatment is too long they will burst. Exposure to hypotonic solution should be for approx 15 min in the incubator before fixation with 3: 1 methanol:acetic acid. It is at the fixation stage where cell clumping can occur, so it is better not to remove the hypotonic solution before adding fixative. A convenient way of

292

Marshall

avoiding clumping is to suck the bulk of the hypotonic solution into a Pasteur pipet, add 4 mL fixative (freshly prepared and ice-cold) to the tubes, and then slowly run the cell suspension into the fixative. The tubes should then be centrifuged (2OOg,10 min), the supernatant removed, the pellet knocked into suspension, and fresh fixative (4 mL) added. By this time the cells are not so fragile and centrifugation can be carried out at faster speeds, e.g., 800g for 2-3 min. The fixative should be changed at least twice more. Slides can be prepared at this stage, but better preparations will result if cells are left in fixative and refrigerated overnight. Tubes are then centrifuged and the cell pellet resuspended m a small volume (approx 0.2 mL) of fresh fixative to give a milky suspension. Several drops of 45% (v/v) aqueous acetic acid added to each suspension at this stage will improve chromosome spreading. Drop 3 or 4 drops of each suspension evenly on to a wet (precleaned) slide and allow to dry. It is good practice to examine the first few slides that are prepared using a low power (e.g., 10x) phase objective to check cell density and chromosome spreading. See Note 6. Cells are stained in filtered, 4% (v/v) Giemsa stain in pH 6.8 buffer. The slides are rinsed in tap water and mounted with coverslips. 3.8.

Selection

of Doses

for Analysis

It is important that the highest dose at which cells are examined for chromosomal aberrations induces some evidence of toxicity (or is the highest dose tested). There are several ways of measuring toxicity in cultured cells, the most relevant of which is in terms of loss of reproductive potential or colony forming ability (CFA). Measurement of CFA in human lymphocytes is not easy because in the absence of growth factors, lymphocytes do not undergo sufficient divisions in culture to produce visible colonies. A relevant indication of toxicity, however, is given by mitotic inhibition, that is, the reduction in the proportion of mitotic cells (mitotic index) as compared with concurrent negative controls. Although depression of mitotic index may be the consequence of cell death, it is more often the result of cell cycle delay. It is, however, very pertinent to cytogenetic assays insofar as mitotic inhibition must not be so severe as to preclude analysis. With a typical mitotic index of 5-10% in negative controls, a reduction of 75-80% can be tolerated and still leave sufficient cells to score. Thus, a reasonable upper limit for a maximum analyzable

Measurement

of Chromosome Aberrations

293

dose is often quoted as one that induces SO-80% mitotic inhibition. It is important that this is the mitotic inhibition suffered by the cells that are analyzed for chromosomal aberrations rather than as estimated by a range-finder. Mitotic indices are usually measured using a 40x objective on a total of at least 1000 cells/culture. Mitotic inhibition is then calculated and a dose that induces 50-80% is selected as the top dose for analysis. Slides from the duplicate cultures receiving this dose are then “coded” with those from negative controls and the next two lower doses in the range. Coding can be simply a matter of covering the treatment details on the slide with tape and assigning a random letter or number code. Whatever the method used, analysis must be carried out “blind.” 3.9. Chromosome Analysis It is beyond the scope of this chapter to discuss types and classification of chromosomal aberrations,but the ISCN (1985) schemeof classification is most widely used. The reader is referred to the paper by Savage (4). Slides are scanned at low power to select cells suitable for analysis. To avoid bias, once a cell has been selected at low power, every effort should be made to score it, so the initial choice should be taken with care. The chromosome number is counted at high power using an objective with at least a 63x magnification. Some loss of chromosomes can be expected to occur during preparation and it is acceptable to analyze cells that have lost up to 2 chromosomes (i.e., having 44 or 45). Hyperdiploid cells (47-68 chromosomes) and polyploid or endoreduplicated cells (~68 chromosomes) should be recorded but not analyzed for structural aberrations. A minimum of 100 cells/culture should be scored for chromosomal aberrations, although it is permissible to analyze fewer from positive controls so long as a positive effect is clearly demonstrated. The microscope coordinates of every aberrant cell (at least) should be recorded. 3.10. Analysis of Results The laboratory should establish an internal historical negative control range. This should be based on negative control data from about 20 or so recent, consecutive experiments. Aberration data for individual cultures from male and female donors treated in the absence and presence of S-9 should be accumulated and a range for each determined. The historical control (normal) range should be regularly updated.

294

Marshall

For any experiment, the number of cells with aberrations in each culture is calculated and split into 3 categories: 1. Cells with aberrations, including gaps; 2. Cells with aberrations, excluding gaps; 3. Cells that are hyperdiploid, polyploid, or endoreduplicated.

The numbers of cells with structural aberrations in individual negative control cultures should fall within or close to the normal range, otherwise the study should be rejected. The statistical evaluation of in vitro cytogenetics assaysis discussed in detail in Richardson et al. (10). The cell should be considered as the biological unit rather than the chromosome and data evaluated using numbers of cells with aberrations rather than numbers of aberrations. It is recommended that, initially, the variation between treatments is compared with the variation between replicate culturesusing a binomial dispersion test. The variation between replicates should conform to a binomial distribution for an acceptable assay. The proportion of cells with aberrations at each dose level is then compared with the proportion in negative controls using Fisher’s exact test. Probability values of p I 0.05 should be accepted as significant. Fisher’s exact test is very sensitive and it is sometimes dangerous to call a positive result on the basis of marginal statistical significance at a single dose level alone. It is accepted practice in some laboratories to impose a second criterion, i.e., that the incidence of aberrant cells in at least one replicate culture at a significant data point exceeds the historical negative control range. If the proportion of aberrant cells at any one dose fulfils both the above criteria then a positive result may be concluded. In some cases equivocal results may be resolved with reference to the types of aberrations that are observed. Chromatid exchanges, for example, are extremely rare in negative controls and their appearanceshould therefore be regarded as significant. Similarly, cells with more than one aberration are more likely to be seenfollowing treatment with a clastogen. These observations, however, are usually of only limited value and it is highly recommended that any result should be confirmed in a repeat experiment. 4. Notes 1. For compounds of limited solubility or short supply, it may only be possible to achieve a maximum exposure by dissolving the chemical directly

into culture medium andaddingthis to the cell pellet following centrifugation (2OOg, 10 min).

Measurement

of Chromosome Aberrations

295

2. The chemical should be added to the test cultures as soon as possible after dissolution and, if dissolved in an aqueous medium, should be membrane filter sterilized. 3. The highest dose level should be its limit of solubility or 10 rnM (whichever is lower). Testing at precipitating doses is not recommended insofar as it is difficult to see how this could be relevant to exposures in vivo. 4. A “range-finder” may be used to select doses for the chromosomal aberration study. It should be noted, however, that the predictivity of range finding experiments is limited. Even if a range-finder is used, rt is always necessary to treat at more doses than will be analyzed in the main study (see Section 3.8.), and it is possible to use sufficient, closely spaced doses so that a range finding experiment is precluded. A total of 12-15 doses, following a geometric progression and separated by a factor of 0.65 is convenient. 5. Extremes of culture conditions such as pH and ionic strength can cause chromosomal aberratrons (7). It is therefore wise to check the pH of the culture medium containing the test chemical at its highest concentration. A shift of >l pH unit may be enough to cause clastogenicity and, if this occurs, neutralization of the test chemical prior to addition to the cultures should be considered. A maximum treatment level of 10 rnM has been recommended to avoid the problems associated with osmolality. Nevertheless, it is good practice to check that the presence of the test chemical does not increase the osmolality of the treatment solutions by more than about 50 mOsm/kg. 6. Spreading 1svery dependent on relative humidity. Low humidity causes rapid evaporation and sometimes overspreading with consequent loss of chromosomes from the metaphase cells. High humidity produces the opposite effect, that is, cells where the membrane has not burst or where the cytoplasm persists around the chromosomes. Such preparations may be very difficult if not impossible to analyze, particularly if the chromosomes are overlapping. In such a case,a change of fixative may help. If not, leaving the tubes refrigerated for one or more days (or until the humidity decreases) should be considered. Tubes of cells in fixative can be stored for several months in the refrigerator and still provide adequate preparations. Two slides per culture will usually provide sufficient cells for analysis. References 1. Solomon,E , Borrow, J., andGoddard,A. D. (1991) Chromosomeaberrationsand cancer.Science 254,1153-l 160. 2. Boue, J., Boue, A., andLazar, P. (1975) Retrospectiveand prospective epidemiological studiesof 1500 karyotypedspontaneoushuman abortions. Teratology 12, 1l-26.

Marshall

296

3. Verma, R. S. and Babu, A. (1989) Human Chromosomes Manual of Basic Techniques, Pergamon, New York. 4. Savage, J. R. K. (1976) Annotation: classification and relationships of induced chromosome structural change. J. Med. Genet. 13, 103-122. 5. Scott, D., Dean, B J., Danford, N. D., and Krrkland, D J (1990) Metaphase chromosome aberration assays in vitro, m Basic Mutagenicity Tests: UKEMS Recommended Procedures (Kirkland, D. J., ed ), Cambridge University Press, Cambridge, pp. 62-86. 6. Preston, R. J., San Sebastion, J. R., and McFee, A. F. (1987) The in vitro human lymphocyte assay for assessing the clastogenicity of chemical agents. Mutat. Res 189,143-188.

7. Scott, D., Galloway, S. M., Marshall, R. R., Ishidate, M., Brusick, D., Ashby, J., and Myhr, B. C. (1991) Genotoxicity under extreme culture conditions. Mutat Res. 257,147-204.

8. Scott, D. and Evans, H. J (1967) X-ray induced chromosomal aberrattons m vrcza faba: changes m response during the cell cycle. Mutat. Res 4,579-599. 9. ISCN (1985) An International System for Human Cytogenetzc Nomenclature (Harnden, D. J., et al., eds.), Karger, Basel. 10. Richardson, C., Williams, D A., Allen, J. A., Amphlett, G., Chanter, D. 0 , and Phillips, B (1989) Analysts of data from in wtro cytogenetic assays, m Statistical Evaluation ofMutagenicity Test Data (Kirkland, D. J., ed.), Cambridge University Press, Cambridge, pp 141-154.

CHAPTER33

Mutation

Assays in Bacteria

Christopher

Glare

1. Introduction Although there are many variations of assays measuring the ability of a test chemical to induce mutation in bacterial cells, by far the best known (and probably the best known of all mutagenicity tests) is the Ames test. The methods described in this chapter are largely based on those developed by Bruce Ames and his coworkers in the 1970s and early 1980s (1,2). Since then, this test has gained worldwide acceptance as a rapid, reliable, and economical method for screening compounds for potential genetic activity at the nucleotide level. Many studies have been undertaken that have examined the correlation between bacterial mutation assay results (3-5) and carcinogenicity (usually in rodents), and the results show that these assays are worthy of a place in a battery of short-term tests for assessinga potential mutagenic or carcinogenic hazard to humans. The Ames test, as it has become known, examines the ability of a test compound to induce mutation in strains of SuZmoneZZatyphimurium. These strains are already mutant at a site in a gene required for histidine biosynthesis, and a compound inducing mutations in the bacteria will revert some of them to a nonhistidine-requiring state. The number of revertant colonies appearing on plates after treatment is therefore a measure of the number of mutations induced by the test compound. Testing is carried out both in the absence and presence of a mammalian liver (normally rat) metabolizing system, since a number of comFrom Methods m Molecular Brology, Edited by’ S O’Hare and C K Atterwtll

Vol. 43 In V&o Toxroty Testing Protocols Copynght Humana Press Inc , Totowa, NJ

297

Glare

298

pounds are known to become activated to carcinogens by mammalian metabolism. The methods described in this chapter are those for the performance of a standard Ames plate test. Numerous variations of this have been developed, for example, for the improved detection of certain classes of test chemical, or for the testing of gases or insoluble materials. It would be impossible to detail all the variations of the assay in this chapter, and for more information the reader is recommended to consult the literature. Probably a good starting point would be refs. 2,6 at the end of this chapter. 2. Materials 2.1. Reagents 1. Nutrient broth (Oxoid [Basingstoke,UK] Number 2). 2. Difco (Surrey, UK) bacto-agar. 3. Bacterial tester strains. Ames originally requested that these be obtained directly from his laboratory (1). Alternatively, any laboratory already performing these tests should be able to provide them. 4. Potassium chloride (KCl).

5. 6. 7. 8.

Sodium phosphatebuffer, pH 7.4. Glucose-&phosphate(disodium). NADP (disodium). Magnesium chloride (MgC12).

9. Suitable solvent, e.g., dimethylsulfoxide (DMSO), acetone. 10. Positive control chemicals, 2-nitrofluorene (2NF), sodium azide (Nal$), 9-aminoacridine (AAC), glutaraldehyde, and 2-aminoanthracene (AAN). 11. Mammalian liver S9 fraction.

Other materials may be required if agar plates and liver S9, for example, are not obtained from external suppliers. 12. 13. 14. 15. 16. 17. 18. 19.

37OCbacteriological incubator. 90-mm plastic Petri dishes. Water bath or heating block. Test tubes. Adjustable micropipets and disposable tips, Vortex mixer. Inverted microscope. Bacterial culture flasks.

20. Shaking water bath or incubator. 21. Colony countmg system (not essential).

Mutation

299

Assays in Bacteria 2.2. Minimal

Davis

Agar

To make 1 L. 1, Ammonium sulfate AR grade,1.Og. Dipotassium hydrogenorthophosphateAR grade,7.0 g. Potassiumdihydrogen orthophosphateAR grade,2.0 g. Trisodium citrate AR grade,0.5 g. Magnesium sulfate AR grade,0.1 g. Dissolve in distilled water and make up to a volume of 400 mL. 2. Agar No. 1, 15.0 g. Distilled water, 600 mL. Sterilize both (1) and (2) by autoclaving (121°C for 15 min). Allow the solutions to cool to approx 56°C and add solution (2) aseptically to solution (1). Finally, add 8 mL of a sterile 50% w/v glucose solution and mix thoroughly. Dispense aseptically into 90 mm plastic Petri dishes (25 mL/ plate) and allow the agar to set. Vogel Bonner agar is also suitable for use in the test (2). Both types of media can be purchased as ready-poured plates from commercial suppliers. To prepare 1 L soft agar: Bacto-agar (Difco Laboratories), 9.0 g. NaCl, 5.0 g. Distilled water, 1 L. The agar and NaCl are weighed into a suitable vessel and the water is added. The mixture is heated with stirring until completely dissolved and is then dispensed in 2.5~mL aliquots into racks of test tubes. The tubes are capped and autoclaved ( 121“C for 15 min). 2.3. Mammalian Liver Postmitochondrial Fraction

(S9)

The S9 is usually prepared from the livers of rats that have been administered with an enzyme inducer prior to sacrifice, The method of preparation is not complicated (1,2,6), but does require careful technique to ensure that a sterile and active S9 fraction is obtained. In addition, the use of the enzyme inducer most widely used in S9 preparations, Aroclor 1254, requires particular care since it is a known carcinogen and is difficult to dispose of because of its stability. For these reasons, it may be preferable to purchase S9 from a commercial supplier.

300

Clare Table 1 Components of the 10% S9 Mix and Buffer Solutron (Per 100 mL Volumes) Quantity (mL)

Component

Concentration

Sodmm phosphate, pH 7.4 Glucose-6-phosphate NADP Magnesium chloride Potassium chloride L-histidme HCl (in 250 mM MgCl,) D-biotin Liver S9 fraction Distilled water

500 180 25 250 150

10% S9 mix

mM mg/mL mg/mL mM mM

20 0 845 12 6 3.2 22

1 mg/mL 1 mg/mL -

4 4 88 10 to volume

2.3.1. Preparation

Buffer solution 20

-

4 4 88 to volume

of 10% S9 Mix

Treatments are carried out in both the absence and presence of an S9 mix. For those treatments without S9, the buffer solution detailed below is added instead of the S9 mix. One hundred milliliter quantities are prepared as detailed in Table 1. All components except the liver S9 fraction are filter-sterilized prior to use. The S9 preparation should be sterile and thawed immediately prior to use. 2.4. Storage and Checking of Bacterial Test Strains Many different tester strains are in existence. However, it is usual to use a battery of about five strains that have been selected so that between them they will detect a range of different types of mutation (and therefore different mutagens). A widely accepted approach would be to use strains TA98, TAlOO, TA1535, TA1537, and TA102 of Salmonella typhimurium. 2.4.1. Preparation

of Frozen Permanents

To prepare a frozen stock culture, grow up a nutrient broth culture of the strain (a 10-h incubation period m a shaking water bath is suitable). Add DMSO to the culture at the rate of 0.09 mL DMSO/mL of culture. Mix thoroughly and aliquot into 1 mL capacity sterile freezer vials. The tubes can then be transferredto storagein liquid nitrogen or a -80°C freezer. Cultures should remain viable for many months if prepared and stored in this way.

Mutation

Assays in Bacteria

301

Table 2 Expected Results of Test Strain Characterization (Phenotype) Checks Expected results with stram Test

TA98

TAlOO

TA1535

TA1537

TA102

Histidine requirement Ampicillin resistant Tetracycline resistant Crystal violet sensitive UV sensitive

Yes Yes No Yes Yes

Yes Yes No Yes Yes

Yes No No Yes Yes

Yes No No Yes Yes

Yes Yes Yes Yes No

2.4.2. Preparation of Cultures for Testing Each time a test culture is prepared, 1 vial of stock culture is thawed and a few drops are used to inoculate a flask containing nutrient broth (the flask volume should be 3-5 times the culture volume). Ampicillin (final concentration 25 pg/mL) can also be added when preparing cultures of strains TA98, TAlOO, and TA102, and tetracycline (final concentration, 2 pg/mL) is added for cultures of strain TA102 only. The flask containing the broth culture is then placed in a shaking water bath and incubated for 10 h at 37°C. At the end of this period the culture should contain approx 2 x lo9 bacteria/ml, and is ready for use in the test. 2.4.3. Checking of Test Strain Characteristics The test strains described here are thought to be relatively genetically stable (7). However, it is nonetheless very important that the test strains are regularly checked for their genetic characteristics (2). Table 2 shows the tests commonly used and the expected results with the 5 test strains recommended here. 1. Histidine requirement can be tested by spreading some of the culture over the surface of a minimal agar plate, onto which already has been spread 0.1 mL of 0.5 miW brotin solution (strains possessing the uvrB mutation also have a requirement for biotm). After incubation overnight at 37°C there should be little or no growth of bacteria on the plate. A control plate containing histidine (0.1 mL of O.lM L-histidine spread onto the plate surface with the biotin) should be prepared and compared with the biotin-only plate (the control should have heavy growth). 2. Tester strains possessing the pKMlO1 R-factor plasmid are resistant to amprcillin. This should be tested sincethe plasrmd can be lost from the bacterra. Test by spreading the culture over the surface of a nutrient agar plate

and adding an ampicillin sensitivity testing disk (Oxoid Ltd.) before incubating at 37OC overnight. There should be no inhibition zone around the disk on plates with test strains possessingthe plasnud. By contrast, parallel testmg of a strain not possessing the plasmrd should show a clear zone of growth inhibition. Strain TA102 also possessesthe pAQ1 plasmid that has a tetracycline resistancemarker. A similar method to that described for testing ampicillin sensitivity (but substituting a tetracycline sensitivity disk) can be used to demonstrate the presence of the plasmid in the TA102 strain. 3. The 5 tester strains described here also have the rfa, or “deep rough,” mutation, This mutation, affecting the cell wall of the bacteria, allows large molecules to enter the cell. The presence of this mutation can be confirmed by demonstrating sensitivity to crystal violet (this is lethal to the bacteria only if it penetrates the cell wall). A nutrient agar plate IS spread with a sample of the culture to be tested, and a sterile filter paper disk (5 mm), to which has been added 10 PL of 1 mg/mL crystal violet solutron, is placed on the seeded plate. Following overnight incubation at 37OC, a zone of inhibition should surround the filter paper drsk. 4. All strains possessing the uvrB mutation can be confirmed by demonstrating UV sensitivity. Test strain culture should be streaked across the surface of a nutrient agar plate. Then, while covering half of the plate, irradiate the plate with a UV source (a 15 W germicidal lamp at a distance of 33 cm was originally recommended by Ames). Strains possessing the pKMlO1 plasmid (ampicillin resistant) should be exposed for 8 s, the other strains for 6 s. TA102 can be tested in parallel on the plate as a control (it does not have the uvrB mutation). Once again, the plates should be incubated at 37°C overnight. Strains with the uvrB mutation should only grow on the unirradiated part of the plate. 5. Each tester strain also has a characteristic number, or range, of revertant colonies on negative- or solvent-treated plates (representmg those mutants that arise spontaneously within the culture). Thusshould be determmed for each laboratory over a period of time (months or years), but might be expected to fall within the limits shown in Table 3. The ranges are fairly wide but do give some indication that tester strains are functioning correctly.

3. Methods 3.1. Treatment of Bacteria Each test compound is normally tested initially in a dose range-finder experiment (where only one of the strains is used), and then in two inde-

pendent mutation experiments with all five test strains. The experimental procedure is the same for each of these experiments.

Mutation

303

Assays in Bacteria Table 3 Normal Rangesof Revertant Counts on Negative Control Platesfor Each Test Strain Expectedrange of revertant counts Test strain on negative control plates lo-55 TA98 65-200 TAlOO TA1535 3-40 2-20 TA1537 280-500 TA102

Before treatments commence, the following

must be prepared:

1. Bacterial cultures. Cultures of each tester strain required are normally prepared on the day prior to testing, according to the method described elsewhere. For convenience, a time switch can be used to control the operation of the water bath, so as to provide 10-h cultures when required on the day of the test. 2. S9 mix and buffer solution: Prepared as detailed in Table 1. Solutions are stored on ice after preparation. 3. Test compound preparation: The,widely accepted maximum treatment dose for these tests is 5 mg/plate, or the limit of solubility or toxicity, whichever is the lowest. To achieve 5 mg/plate, a 50 mg/mL solution of the test compound in a suitable solvent (8) is required, since it is usual to add no more than 0.1 mL of an organic solvent solution per plate. A range of further dilutions are made for treatment, so as to try to demonstrate a dose-relationship for any mutagens tested. The range-finder and first experiment are often performed with fairly widely spaced doses (e.g., fivefold intervals between doses), with the dose range narrowed for the second experiment in order to investigate any possible dose-relationship. All treatment solutions should be sterile (filter sterilization using disposable units is most convenient) and protected from light once prepared. 4. Soft agars: Prepared as described elsewhere. One tube of agar for each plate to be made should be melted (autoclaving is suitable) and held at 46°C for treatment (using either a water bath or heating block). 5. Agar plates: Plates containing Minimal Davis or Vogel Bonner agar should be labeled before treatment. Each plate should be labeled on the side of the base with a code that fully identifies the strain and treatment to be made on that plate. Each strain is normally treated with five concentrations of the test chemical, plus a solvent and positive control, and each treatment is

304

Glare Table 4 Suggested Positive Control Compounds, Their Treatment, and Final Plate Concentrations for Each Test Strain

Compound 2NF Nfl3

AAC Glutaraldehyde AAN

Use TA98 TAlOO TA1535 TA1537 TA102 All strains

S9 +

Treatment concentration WmL) 500 20 20 500 250 50

Final concentration Wdplate) 50 2 2 50 25 5

carried out both in the absence and presence of S9 mix. To maximize the

statistical sensitivity of the test, as many replicate plates as possible should be used for each treatment. Obviously, there is a practical limit as to how many plates can be used in terms of the size of each experiment, but a reasonable compromise 1sto use five replicate plates for solvent controls, three for each concentration of testcompound, and three for positive controls. For an experiment using all five test strains, this amounts to 230 plates. 6. Positive controls: In addition to preparing solutrons of the test compound for treatment, a solutron of each positive control compound is required. However, m many cases these may be prepared as stock solutions and stored under appropriate conditions for a number of weeks (9). Table 4 gives information regarding the preparation and use of these compounds. Once the required preparations have been completed, treatment of the bacteria with the test agent is carried out. This is a relatively simple process but does require care to ensure that the correct additions are made for each treatment. It is convenient to carry out treatments in batches of lo-20 plates. An aliquot of bacterial culture (0.1 rnL) is added to one of the test tubes containing soft agar (held at 46°C in a water bath or heating block). The required treatment or control solution (0.1 mL) is added, quickly followed by the S9 mix or buffer solution (0.5 rnL). The contents of the tube are then mixed by gentle vortexing and are poured onto the minimal agar plate labeled for that treatment. Tilt and rotate the plate to spread the agar over the plate surface and then stand on a level surface until it has set. Plates are then inverted and placed in the incubator. Incubation should be for 48-72 h at 37°C.

Mutation

305

Assays in Bacteria 3.2. Plate

Counting

Plates may be scored by eye, but many laboratories now use automated colony counting systems for this purpose. Apart from the obvious time-saving advantages of automated counting, the use of such systems also allows the direct capture and transfer of data to a personal computer for statistical analysis. Software has been developed specifically for the analysis of Ames test data, and is commercially available (e.g., York Electronic Research, Huntington, England). 3.3. Toxicity

The trace of histidine added in the S9 mix or buffer solution allows all the bacteria present to undergo a few divisions, and results in the appearance of a background lawn of bacteria. Examination of this lawn, often with the aid of an inverted microscope, provides information as to the toxicity of the test agent. If a large proportion of bacteria have been killed, then a thinning of the background lawn should be apparent. In cases of severe toxicity, the increased “share” of the available histidine to the few surviving bacteria may allow small colonies (or microcolonies) to develop, which can be mistaken for true revertants. Observation of the background lawn should provide a clue to this, but the nature of the colonies should also be verified by cross-streaking the colonies to minimal agar containing biotin but no histidine. Only true revertant colonies will grow on these plates, microcolonies formed because of toxicity will not. 3.4. Interpretation

of Data

Following counting of plates, mean revertant colony counts obtained at each treatment dose are compared with those obtained for the corresponding solvent control. A number of statistical tests have been recommended for analysis of data from these assays (10). A compound mutagenic in this test system should give rise to statistically significant increases in revertant numbers in one or more test strains that are both dose-related and reproducible in an independent experiment. Many mutagenic compounds, such as those detailed as positive control chemicals previously, will give increases in revertant numbers that are clearly significant without the need for statistical analysis to reinforce this fact. However, the interpretation of results may well be less clear than this and in such cases statistical analysis will provide a useful guide to interpretation. Additional experiments, possibly using a modified protocol,

Clare

306

often useful if equivocal or conflicting results have been obtained in independent experiments. References

are

1 Ames, B N., McCann, J., and Yamasaki, E. (1975) Methods for detecting carcmogens and mutagens with the Salmonella/mammalian microsome mutagenicity test Mutat. Res. 31,347-364.

2. Maron, D. M. and Ames, B. N. (1983) Revised methods for the Salmonella mutagenicity test. Mutat. Res. 113, 173-215. 3. McCann, J. and Ames, B. N. (1976) Detection of carcinogens as mutagens in the Salmonella/microsome test: assay of 300 chemicals* discussion Pruc NatE. Acad. Sci. USA 73,950-954.

4 Tennant, R. W., Margolin, B. H , Shelby, M. D., Zeiger, E., Haseman, J. K., Spalding, J., Caspary, W., Resmck, M , Staserwicz, S., Anderson, B., and Minor, R. (1987) Prediction of chemical carcmogenicity in rodents from m vitro genetic toxicity assays. Science 236,933-941. 5 Zeiger, E. (1987) Carcinogenicity of mutagens: predictive capability of the Salmonella mutagenesis assay for rodent carcinogenicity Cancer Res. 47, 1287-1296. 6 Gatehouse, D. G., Wilcox, P., Forster, R., Rowland, I , and Callander, R. D. (1990) Bacterial mutation assays, in Basic Mutagenrctty Tests. UKEMS Recommended Procedures, Report of the UKEMS Subcommittee on Guidelines for Mutagemcity Testing (Kirkland, D. J., ed.), Cambridge Umversrty Press, Cambridge, pp. 13-61. 7. Margolin, B. H., Risko, K. J., Shelby, M. D., and Zerger, E. (1983) Sources of variability in Ames Salmonella typhimurium tester strains: analysis of the International Collaborative Study on “Genetic Drift. ” Mutat. Res 130, 1 l-25 8 Maron, D., Katzenellenbogen, J., and Ames, B. N. (1981) Compatibility of organic solvents with the Salmonelldmicrosome test. Mutat. Res. S&343-350 9. Pagano, D. A. and Zeiger, E. (1985) The stability of mutagenic chemicals stored in solution. Environ. Mutagen. 7,293-302. 10 Mahon, G. A. T., Green, M H. L., Middleton, B , Mitchell, I. de G , Robinson, W. D., and Tweats, D J. (1989) Analysis of data from microbial colony assays, in Statistical Evaluation of Mutagenicity Test Data, Report of the UKEMS Sub-committee on Guidelines for Mutagenicity Testing, Part III. (Kirkland, D J., ed.), Cambridge University Press, Cambridge, pp. 26-65.

CHAPTER34

Chick

Embryotoxicity Screening (CHEST I and II)

Wendy J. Davies

and Stuart

Test

J. Freeman

1. Introduction The chick embryo has been used for the study of experimental teratogenicity by various techniques (l-4). This chapter describes a method developed by Jelinek (.5,6) that combines a standardized technique (CHEST I) with other techniques (CHEST II); it allows the administration of small amounts of test compound and the measurement of a quantitative endpoint (CHEST I). 2. Materials 1. Fertilized eggs: Morphologically normal and at specific stages of embryo development (7) 2. Micropipet glass: Bent and with beveled ground tip, calibrated to 3 ltL (10 or 100 PL). 3. Polyethylene tube and mouthpiece (with exchangeable cotton wool filter): Connected to micropipet. 4. Glass slides and paraffin. 5. Incubator (or thermostatic oven). 6. Dental drill. 7. Dissecting mtcroscope wtth ocular micrometer.

3. Methods 3.1. CHEST

I

1. Eggs are incubated at 37-38°C and 50-60% relative humidity throughout. Sterile conditions are employed when possible. From. Methods WI Molecular B/o/ogy, Vol 43 In Wtro Toxroty Testrng Protocols E&ted by S O’Hare and C K Atterwlll Copynght Humana Press Inc , Totowa, NJ

307

Davies

and Freeman

Fig. 1. Diagramatic representation of chick embryos. (A) Stage 11 HH; (B) stage 16 HH. 2. Eggs are incubated for approx 40 h (until embryos reach stagesHH+ 10-l 1) (Fig. 1A). Blunt ends are perforated before the eggs are candled and then a small window is cut m the shell directly above the embryo using a dental drill (8). Underlying shell membranes are removed with forceps. 3. Six eggs are used for each concentration of test compound (concentrations in whole-log intervals). 4. Test compounds are dissolved in saline or distilled water. The followmg solvents can also be used: 30% (v/v) ethanol, 10% (v/v) DMSO, sunflower oil, and the suspending agent carboxymethylcellulose (1% w/v). pH is adjusted to 5-9. If the substance is poorly soluble, dose volumes of 10 or 100 pL can be used instead of 3 l.tL. 5. Three microliters l.tL of a solution of test compound (or a dilution of) is injected directly below the caudal region of the embryo. The tip of the micropipet is carefully inserted on the outside of the area pellucida. The procedure is carried out under a dissecting microscope.

CHEST I and II 6. The vitelline membrane (blastoderm) is then moistened with one drop of saline before the windows are covered with glass slides and sealed with paraffin. 7. Eggs are returned to the incubator for a further 24 h. 8. Eggs are then reopened. The distance between the vitellme arteries and the tip of the tail (the caudal trunk) is measured with an ocular micrometer (Fig. 1B). 9. Embryos are stained with neutral red (0.05%) to enable detection of abnormalities. 10. The mean lengths for each concentration (n = 6) are plotted against log concentration. 11, The interval between the maximum tolerated and lowest effective concentrations (the latter being a significant p < 0.05 by one sided t-test, shortening of the mean trunk length) is considered to be the beginning of the embryotoxic range.

3.2. CHEST

II

12. Three groups of ten eggs (HH 11-14, d 2; HH 17-20, d 3, and HH 21-24, d 4) are used for each concentration of test compound. 13. Three or four concentrations are selected from the beginning of the embryotoxic range found with CHEST I. 14. Solutions are iqected subgerminally on d 2 and intraamniotically on d 3 and 4. 15. Before administration, eggs are turned daily starting on d 2. Following administration, eggs are incubated to 8 d without turning. 16. Fetal membranes are then removed from the embryos and they are weighed and examined. 17. Malformations of the following organs are assessed:head, face, eye, body wall, trunk, and heart (and internally) (6).

3.3. Evaluation 18. The determination of the beginning of the embryotoxic range using CHEST I has been described in steps 10-l 1. 19. The number of dead, malformed, and growth retarded fetuses (weight ~650 mg and with no malformation) are totalled for each concentration and stage of embryo at the time of exposure. 20. The proportion of affected embryos is plotted against: a. Concentration of test compound; and b. The embryo-stage of administration, A more accurate embryotoxic range can now be established. 21. The proportion of effects on particular organs in surviving fetuses is considered separately to establish a profile of effects for each compound.

Davies

310

and Freeman

22. Studies have shown that there is a positive correlation between the magmtude of effective concentratrons m CHEST and those m other systems, including mammals (9,lO). Results suggest that CHEST is able to rank agents in order of their teratogenic potency m mammals (II). 23. Embryotoxic and therapeutic ranges for drugs can be compared for an assessmentof human risk.

4. Notes 1. A refinement of the windowing technique used to expose embryos has reduced the incidence of nonspecific anomalies (4). 2. It is possible to allow chicks to go to hatching to confirm a negative embryotoxic effect.

References 1. McLaughlin, J., Jr., Marhac, J., Verret, M. J., Mutcler, M K , and Fitzhugh, 0. G (1963) The injection of chemicals into the yolk sac of fertile eggs prior to incubation as a toxicity test. Toxicol. Appl. Pharm. 5,760-771. 2. Gebhardt, D. 0. E. (1972) The use of the chick embryo in applied teratology, m Advances in Teratology, vol. 5 (Woollam, D. H M., ed ), Academy, New York, pp. 97-l 11 3. Verrett, M. J., Scott, W. F., Reynaldo, E. F., Alterman, E. K., and Thomas, C A. (1980) Toxrcity and teratogenicity of food additive chemicals in the developing chicken embryo. Toxicol. Appl. Pharm. 56,26.5-273. 4 Fisher, M and Schoenwolf, G. C. (1983) The use of early chick embryos in experrmental embryology and teratology: improvements in standard procedures. Teratology 27,65-72. 5 Jelinek, R. (1977) The chick embryotoxicity screenmg test (CHEST), in Methods in Prenatal Toxrcology, (Neubert, D , Merker, H. J., and Kwaugroch, T. E., eds.), G Thieme, Stuttgart, pp. 381-386. 6. Jelinek, R , Peterka, M , and Rychter, Z (1985) Chick embryotoxicity screening test-130 substances tested Znd J Exp. Biol. 23,588-595 7 Hamburger, V and Hamilton, H. L. (1951) A serves of normal stages in the development of the chick embryo. J. Morphol. 88,49-92. 8 Hamburger, V. (1960) A Manual of Experimental Embryology, rev ed , University of Chicago Press, Chicago. 9 Vesely, D , Vesela, D., and Jelinek, R (1982) Nineteen mycotoxins tested on chicken embryos. Toxic01 Lett 13,239-245. 10 Vesely, D., Vesela, D , and Jelinek, R (1984) Use of chick embryo m screening for toxm-producing fungi MycopathoZogia 88, 135-140. 11. Jelinek, R and Rychter, Z. (1979) Morphogenetic systems and the central phenomena of teratology, in Advances in the Study of Bu-th Defects, vol. 11, Teratologzcal Testing (Persaud, T V. M., ed.), Umversrty Park Press, Balumore, MD, pp. 41-67

CHAPTER35 Frog Embryo

Teratogenesis

Assay

Xenopus (FETAXI

Wendy J. Davies

and Stuart

J. Freeman

1. Introduction

The method described was developed by Dumont and colleagues (I). 2. Materials 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

Xenopus Zaevis, mature.

Aquaria, polycarbonate or glass. Petri dishes (6 cm diameter). Polypropylene micropestle and microfuge tube (1.5 mL). Lyophilized pregnant mare serum (3.4 x lo3 IU gonadotropin/mg). Human chorionic gonadotropm. Ethyl m-aminobenzoate methanesulfonate (Tricame). Sodium phosphate. L-cysteine HCl. Tris phosphate buffer: 15 mM Trts base, 0.5 m/t4 Na2HP04, 2 rn.k! KCl, 2 mM NaCI, 1 nG’l4NaJOs, cont. MgS04/7Hz0, adjusted to pH 7.6 with dilute acetic acid. 11. FETAX medium (artificial pondwater), 10.7 rmWL NaCl, 1.14 mWL NaHCOs, 0.40 miWL KCl, 0.14 W/L CaC12,0.35 rniWL CaS04, 0.62 mM/L MgS04, in demineralized distrlled water, pH 7.6-7.9 (2). 12. Formalin or gluteraldehyde. 13. Dissecting microscope with ocular micrometer.

From Methods UI Molecular Bfology, Vol. 43’ In Vjtro Toxrcrfy Testmng Protocols EdIted by. S O’Hare and C K Alterw~ll Copynght Humana Press Inc., Totowa, NJ

311

Davies

312

and Freeman

3. Methods 3.1. Maintenance 1. Xenupus are housed at 20 f 10°C m aquaria filled to a depth of 10 cm with sodium chloride solution (10 n-M/L in deminerahzed water). 2. They are fed three times a week with either pelleted frog chow or a ground mixture of beef organs (liver, heart, lung). Food is supplemented with a vitamin mixture. 1. 2. 3. 4. 5. 6.

3.2. Breeding On the evening before the assay, one male and female Xenopus (at least 30-d-old) are injected with 500 and 1000 IU, respectively, of human chorionic gonadotropin into the dorsal lymph sac. Amplexus usually takes place within 2-6 h, and deposrtion of eggs from 9-12 h after injection. After breeding, adults and fecal material are removed and the embryos are collected into a Petri dish. The jelly is removed from the eggs by swirlmg for 3 mm m FETAX medium contaming L-cysteine HCl (0.13 mol/L) adjusted to pH 8.0 with dilute NaOH solution. Cysteine is then removed by washing the eggs four times with FETAX medium, transferring them to a clean Petri dish, and washing another four times. The developing embryos are kept in an incubator at 23 * l°C until the FETAX assaystartsat approx 5 h postfertilization. By this stagethe embryos should have completed the seventh cleavage and reached the large-cell blastula stage.Normally-cleaving embryos are selectedfor the assay(Fig. 1).

3.3. “ArtificiaZ” Fertilization An alternative to the breeding method described above is one that involves the preparation of an artificial sperm solution (3). It enables a more controlled ovulation and fertilization and is described in steps l-l 1. 1. Lyophilized pregnant mare serum is dissolved in NaCl solution (140 n&l/ L) such that 0.1 mL contains 50 IU gonadotropin. 2. Three or four days before each FETAX assay,an adult female is primed to ovulate by an injection of the reconstituted pregnant mare serum (0.1 mL) into the dorsal lymph sac. 3. On the evening before the assay,the female is given an injection of human chorionic gonadotropin (600 IU) dissolved in 0.3 mL of sodium phosphate buffer (10 mM/L, pH 7.2) into the dorsal lymph sac. 4. The female is kept overnight at 18OC.

Fig. 1. Xerwpus laevis embryonic development. Adapted from Weisz (4).

314

Davies

and Freeman

5. On the mormng of the assaya fresh sperm suspension 1sprepared. A male Xenopus is anesthetized by immersion for approx 10 mm m an aqueous solutton of ethyl m-ammobenzoate methanesulfonate (20 n&!/L). 6. A testicle is excised and kept m a chilled beaker contammg 5 mL Tris phosphate buffer. 7. Immediately prior to fertilization, one-third of the testicle is minced with a polypropylene mtcropestle in a n-ncrofuge tube contammg FETAX medium. 8. On the morning of the assay,gentle pressure 1sapphed to the female’s lower back. Batches of 300-400 eggs are expelled into a Petri dish. They are fertilized immediately by adding a fresh suspension of the prepared sperm. 9. Sperm suspensionis ptpeted dropwise over eggs in the Petri dish. After allowmg 45 s for sperm attachment, FETAX medium is added to cover the eggs. 10. Within 20 mm, fertilized eggs should rotate to show the darkly pigmented animal pole at the surface. Unfertilized eggs usually show the cream-colored vegetal pole. 11. This process (steps 8-10) is repeated at 15-mm intervals so that sufficient eggs from each female are harvested. The prevtous method from steps 4-6 1sthen followed. 1. 2. 3. 4. 5. 6. 7. 8.

3.4. The Assay Between 20 and 25 embryos are used per Petri dish. Dishes are covered to prevent evaporation but allow aeratton. Duplicate Petri dishes are exposed to each concentration of test substance in FETAX medium (quadruplet dishes are exposed to FETAX medium only as controls). (See Notes 1-3.) The embryos are viable in low concentrations of vehicles, including DMSO, propylene glycol, acetone, and alcohol (I). Assays are repeated as many times as are necessary-at least three times. Different adult pairs are used for each. During the assaysembryos are kept at 23 t- 1°C in a tissue culture mcubator. They can also be kept at room temperature tf this IS constant for each assay. Media are exchanged and dead embryos are removed at 29, 53, and 75 h postfertilization. At 101 h postfertilization, surviving tadpoles are counted, fixed m 3% formalin, and examined using a dissectmg microscope. (See Note 4.) A variety of endpoints can be measured. development stages (4,5), head-

to-tail length, motlhty (behavior), pigmentation (I), malformations of the major organs, notably gut, eyes, and brain (3), chromosomal damage, and btochemical markers, including RNA, DNA, protein synthesis,and enzyme levels (6). (See Notes 5 and 6.)

FETAX

315

9. Malformed embryos that die are not included in the total number of malformed. Death at 24 and 48 h is determined by skin pigmentation, structural integrity, and irritability. After 72 h, the lack of heartbeat is an unambiguous sign of death. 10. Embryos can also be prepared for paraffin sectioning and electron micros-

COPY(7). 3.5. Evaluation 1. The lethal concentrations LC,,, LCse, and LC,, (concentration lethal to 10,50, or 90% of the embryos over the 4 d of the assay) are calculated by probit analysis from mortality vs concentration plots (8). 2. The effect concentrations EClo, EC,,, and EC9e(concentration producing a 10, 50, or 90% incidence of abnormality by the end of the assay) are calculated by probit analysis (8). These can be cumulative or compared for each individual parameter measured. 3. The Teratogenic Index is calculated in the followmg way: TI = LCs&CsO 4. The Minimum Concentration to Inhibit Growth (MCIG) IS calculated by t-tests between grouped observations to find a significant reduction m growth at p < 0.05. 5. The smaller the MCIG, and the greater the TI, the more likely is the compound to have teratogenic potential. The interpretation and relevance of the FETAX assay is discussed in many articles (2,3,6,9). 1. 2. 3. 4 5. 6.

4. Notes Exposure periods can be altered as required. The sequelae of compoundinduced malformations can be determined by allowing tadpoles to complete metamorphosis for organ and skeleton inspection (10). The removal of jelly from the embryos is optional (I). It can be removed by manual (1) as well as chemical means (12). Some workers have used 100 U/mL pemcillin and 100 U/mL streptomycm in each dish (13) to prevent microbial contammation. A variety of different fixing methods are used: OS-3% formalin, 3-6% glutaraldehyde in pH 7.2, and O.lM phosphate buffer. It is good practice for the same individual to score all embryos from the same assay to reduce variability of subjective assessments, It is possible to introduce a metabolic activation system to the FETAX to more accurately assessthe teratogenic risk of proteratogemc compounds (13,14)

316

Davies

and Freeman

References 1. Dumont, J. N., Schultz, T. W., Buchanan, M. V., and Kao, G. L. (1983) Frog embryo teratogenesis assay Xenopus: FETAX-a short-term assay applicable to complex environmental mixtures, in Symposium on the Application of Short-Term Bioassays in the Analysis of Complex Environmental Mixtures: III (Waters, M. D., Sandhu, S. S , Lewtas, J., Claxton, L , Chernoff, N , and Nesnow, S., eds.), Plenum, New York, pp. 393405. 2. Dawson, D. A. and Bantle, J. A. (1987) Development of a reconstituted water medium and prehminary validatron of the frog embryo teratogenesis assayXenopus (FETAX). J. Appl. Toxicol. 7(4), 237-244 3. Plowman, M. C., Peracha, H., Hopfer, S. M., and Sunderman, F. W., Jr. (1991) Teratogenicity of cobalt chloride in Xenopus laevis, assayed by the FETAX procedure. Terat. Cart. Mutat. 11,83-92. 4. Weisz, P. B. (1945) The normal stages in the development of the south african clawed toad, Xenopus laevis. Anat. Rec. 93, 161-169. 5. Neiuwkoop, P. D. and Faber, J. (1975) Normal Tables of Xenopus laevis (Daudin), 2nd ed., North Holland Publishing, Amsterdam. 6. Courchesne, C. L. and Bantle, J. A. (1985) Analysis of the activity of DNA, RNA and protein synthesis inhibitors on Xenopus embryo development. Terat. Cart. Mutat. 5, 177-193. 7. Schultz, T. W., Dumont, J. N., and Epler, R. G. (1985) The embryotoxic and osteolathyrogenic effects of semicarbazide. Toxicology 36, 183-198. 8. Sachs, L. (1984) Applied Statistics: A Handbook of Techniques, 2nd ed., Springer Verlag, New York. 9. Dawson, D. A. and Wilke, T. S. (1991) Evaluatron of the frog embryo teratogenesis assay: Xenopus (FETAX) as a model system for mixture toxicity hazard assessment. Environ. Toxicol. Chem. 10,941-948. 10. Hopfer, S. M., Plowman, M C., Sweeney, K. R., Bantle, J. A., and Sunderman, F. W , Jr. (199 1) Teratogemcity of Ni2+ in Xenopus laevis, assayed by the FETAX procedure. Biol. Trace Elem. Res. 29,203-216. 11. Rugh, R. (1962) Techniques m experimental embryology, in Experimental Embryology (Rugh, R., ed.), Burgess, MN, pp. l-30. 12. Gusseck, D J. and Hedrick, J. L. (1971) A molecular approach to fertilization, I: disulphide bonds m Xenopus laevis jelly coat and a molecular hypothesis for fertilization Dev. Biol. 25,337-347. 13. Bantle, J. A. and Dawson, D. A. (1988) Uninduced rat liver microsomes as a metabolic activation system for the frog embryo teratogenesis assay-Xenopus (FETAX), in Aquatic Toxicology and Hazard Assessment, vol. 10 (Adams, W. J., Chapman, G. A., and Landis, W. G., eds.), American Society for Testing and Materials, Philadelphia, pp 3 16-326 14 Fort, D. J , Dawson, D. A., and Bantle, J. A. (1988) Development of a metabolic activation system for the frog embryo teratogenesis assay: Xenopus (FETAX) Terat. Cart. Mutat $25 l-263.

CHAPTER36

The Drosophila Wendy J. Davies

melanogaster and Stuart

Assay

J. Freeman

1. Introduction The method described is based on that developed by Schuler and colleagues (1,2) and recently refined (3). 2. Materials 1. Drosophila melanogaster: Oregon-R wild type. 2. Drosophila medium. 3. Yeast: Baker’s live.

4. 5. 6. 7.

Flint-glass shell vials (30 mL). Refined cotton. Cheesecloth. Dissectmg microscope.

3. Methods 3.1. Maintenance 1. Instant plain Drosophila medium is reduced to a powder using a mortar and pestle. One gram is placed m a flint-glass vral and dissolved in 5 mL distilled deionized water. A small amount (~1 mg) of yeast is sprinkled onto the surface of the setting medium. Vials are stoppered with refined cotton and covered with cheesecloth. 2. Drosophila cultures in vials are incubated at 25 f 1“C, at 60-70% relative humidity, and with a 12 h light/dark diurnal cycle. 3.2. Mating 1. Newly enclosed males and females (l-24-h-old)

from stock cultures are

CO:! anesthetizedbeforecollection into holding vials (single sex). Virgin flies arecollected, usually for 3 d (Wednesdayto Friday). From Methods m Molecular Srology, Edited by: S O’Hare and G K. Attetwlll

Vol 43. In Wro Toxdty Testing Protocols Copyright Humana Press Inc , Totowa, NJ

317

318

Davies

and Freeman

2. On the 3rd d or when sufficient flies have been collected, they are anesthetized and transferred to culture flasks (about 20 of each sex, from different stocks) for mass-mating. 3. After 3-5 d (usually over the weekend), the flies are anesthetized again, and inseminated females are removed and introduced singly mto each assay vial (freshly prepared). They are also introduced mto stock vials for the maintenance of stocks. 1. 2. 3. 4. 5. 6. 7. 8. 9.

10. 11. 12.

3.3. The Assay Assay vials are prepared as descrtbed in step 1 above, but using a solutron of test chemical (or a dilution of) instead of water. Following 20 h each female is removed from the assay vial and discarded. Deposited eggs are counted using a dissectmg microscope. After about 9 d (thuscan vary with test compound, and is carefully checked for each vial), adult flies begin to emerge from puparmm (eclosure). Adult flies are harvested daily for 10 d following the first day of eclosure and counted. Flies are not counted after 10 d, in case a subsequent generation of flies are counted. On the 10th d for each vial, survival data is totaled (number of harvested flies/number of eggs deposited). Initial range-finding tests are carried out. Three vials at each of 14 concentrations from O-80 mg test compound/vial are used. A closer range and fewer concentrations are then used to produce a gradient of mortality ranging from 100 to 0%. This can be analyzed by probit analysts (4) to establish concentrations for further tests. These next screenmg test(s) use half-logarithmic mterval concentrations including and below the LC& concentration (concentration lethal to 50% of the developing flies), and controls. Flies used for screenmg tests are selected usmg a dissectmg microscope: males before mass-mating and females before egg deposition. Vigorous and morphologically normal males and the most fecund females are selected. The number of vials is adjusted to yield approx 200 offspring for every concentration tested: The number of vials can be estimated from data obtained in the range-finding tests. If insufficient flies are produced at a concentration, more vials are used m a subsequent repeat. Concentrations are adjusted upward if necessary so that a 50% mortality is always achieved at the highest concentration. Flies are exposed to a continuous flow of CO2 gas after harvestmg. They are scored on the day of eclosure if possible, and not later than 1 wk afterward.

The Drosophila

melanogaster

319

Assay nna orbltals L

Fig. 1. Drosphila

melanogaster

morphology. Adapted from Schuler et al. (2).

13. Each fly is exammed m several orientations for external morphological abnormalities (Fig. 1). All major bristles, legs, segmentation halteres, wings, antennae, eyes, and mouth parts are examined; also, the general appearance of the thorax, abdomen, small bristles, and entire body (includmg size and color). A check for body alignment, excesstissue growth, and extra or absent body parts is also carried out. 3.4. Evaluation 1. Incidence data for the same concentrations from different runs are pooled. 2. If a stattstically stgnlficant concentration-dependent mcidence of any malformation is observed (chi-square, p < 0.05) (4) repeat testing is terminated. The result 1sthen considered to be positive: The compound is predicted to have teratogenic potential. Note: Concentrations testedare lessthan the LCso. 3. If this criterion is not met after challenging with increasing concentrations and obtaining data from a minimum of 500 pooled flies, the result is considered to be negative and testing stops. 4. Notes 1. A simpler method is to allow 3 pairs of adult flies to oviposit in a series of vials (2 d/vial) containing different concentrations of test compound (5).

Davies and Freeman Emerging flies are examined as usual. Larvae produced in this way in untreated vials can also be collected at selected developmental stages for subsequent exposure to compound. 2. Combined adult and developmental toxicity assessments can be made by allowing an adult pair to remain in a vial for 7 d before scormg (6). 3. An important discovery recently has demonstrated that it may only be necessary to assess bent bristles and wing notches for an improved screenmg test for developmental toxicity (3). 4. Drosophila embryonic cell cultures have also been used to detect teratogens (7). These shall not be discussed except to report that they have been used to demonstrate the presence of a metaboltc activating system m Drosophila

(8).

References 1 Schuler, R. L , Hardm, B. D , and Niemerer, R. W. (1982) Drosophzlu as a tool for the rapid assessment of chemicals for teratogenicity Terat. Cart. Mutat. 2,293-301 2. Schuler, R. L., Radike, M. A., Hardm, B. D., and Nremeier, R. W. (1985) Pattern of response of intact Drosophila to known teratogens. J Am. Co11 Toxic01 4(4), 291-303. 3. Lynch, D. W., Schuler, R. L , Hood, R. D., and Davis, D G. (1991) Evaluation of Drosophila for screening developmental toxicants: test results with eighteen chemicals and presentation of a new Drosophila bioassay. Terat. Cart. Mutat. 11,147-173 4. Sachs, L. (1984) Applied Statutics: A Handbook of Techniques, 2nd ed., SprmgerVerlag, New York. 5. Ranganathan, S., Davis, D. G., and Hood, R. D. (1987). Developmental toxicity of ethanol in Drosophila melanogaster. Teratology 36,45-49. 6. Goldstem, S. H. and Babrch, H. (1989) Drfferential effects of arsemte and arsenate to DrosophiZu melanogaster in a combined adult/developmental toxicity assay. Bull. Environ. Contam. Toxicol. 42,276-282.

7 Bournias-Vardiabasis, N., Teplitz, R. L., Chernoff, G. F , and Seecof, R. L. (1983) Detection of teratogens in the Drosophila embryonic cell culture test: assay of 100 chemicals. Teratology 28,109-122 8. Bournias-Vardiabasis, N. and Flores, J. (1983) Drug metabolising enzymes in Drosophila melanogaster teratogenicity of cyclophosphamide in vitro. Terat. Care. Mutat. 3,255-262

CHAPTER37

The Hydra Wendy

J. Davies

attenuata

Assay

and

J. Freeman

Stuart

1. Introduction The method described was devised by Johnson and coworkers (1,2).

2. Materials 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

Hydra attenuatu: cloned males. Artemia nauplii. Amikacin. Glass tanks: semiautomated. Glass wells (5 mL). Centrifuge tubes (15 mL). Eppendorf tubes. Polyethylene tubing, mternal diameter (ID) 0.58 mm. Hamilton syringe. Hydra medium: 0.147 g CaCl, e2Hz0, 0.115 g N-tris-(hydroxymethyl)methyl-2-ammoethane sulfonic acid, 0.004 g (EDTA) dissolved in 1 L distilled water (5 mosM). 11. Reaggregation medium: 0.290 g KCI, 0.968 g CaC12/2Hz0, 0.156 g MgS04, 1.940 g Na citrate, 0.726 g Na pyruvate, 3.003 g N-Tris (hydroxymethyl)methyl-Zaminoethane sulfonic acid, 0.1 g phenol red, dissolved in 1 L distilled water (70 mosM, pH 6.9). 12. Dissecting microscope.

3. Methods 3.1. Maintenance 1. Hydra are grown m Hydra medium in glass trays. A population of 100,000 can be housed in an area of 1.5 m2. From Methods m Molecular B/ology, Edlted by: S O’Hare and C K. Atterwill

Vol 43’ In Wro Toxmty Testmg Protocols Copynght

321

Humana

Press Inc , Totowa,

NJ

Davies

32‘2

and Freeman

2. The temperature IS regulated between 10 and 20°C. Amikacin (150 mg/L) is added to the medium to deter bacterial growth. 3. They are fed once or twice a day with freshly hatched Artemia nauplil. These should be rmsed free of salt beforehand Bacterra should also be killed either by 3-mm exposure of the eggs to 0 1% potassium permanganate, or 20 min exposure of the hatchlmgs to 20 ppm tetraglycine periodide. 4. Uneaten feed and waste products of digestion are flushed away at the end of each feeding cycle. 5. Adult animals scheduled for dissociation (Section 3 3., step 12) are starved for 24 h before use. 3.2. Adult Toxicity Assay 1. Single adult Hydra are placed in glass wells contammg 4 mL Hydra

2. 3. 4.

5.

medium, 150 mg/L Amikacm, and logarithmic interval concentrations of test substance. (See Notes 1 and 2.) Solvents and suspending agentscompatible with the assayare 1% (v/v) ethyl alcohol, 0.5% (v/v) DMSO, and 0.1% (w/v) carboxymethyl cellulose (3). Wells should be covered to prevent evaporation of medmm, although sufficient aeration should be permitted. Two range-finding and two confrrmatron tests are carried out. Intact adults are exposed chronically (no fixed time) until the endpoint is achieved at whole-log concentrations ranging from 1c3-lo3 mg/L. The lowest concentration to produce an effect IS confirmed (Expt. II) and then subdivided into l/10 logarrthmic intervals (Expt. III) before a repeat (Expt. IV) to establish the mmimum effective concentration (MEC) of test substance required to produce the endpomt toxic response (Table 1). The “tuhp” stage is the single endpoint measured It is an irreversible adult toxic event and immediately precedes death of the polyp (Fig. 1). 3.3.

“Embryo”

Development

Assay

1. An artificial embryo is created that is composed of cells from adult PolYPs* 2. Approximately 300 adults are rinsed and placed in a conical 15-mL centrifuge tube containing 3 mL reaggregation medium for 30 mm at room temperature. 3. Polyps are dissocrated by repeat pipetmg. Cells and fragments are allowed to settle for 7 mm before the supernatant is transferred to a second tube 4. Remaining polyps and undissociated fragments are resuspended and step 3 is repeated twice.

The Hydra attenuata Assay

323

Table 1 Hydra Toxicity Test Sequence

Experiment I TC No 0 10-3 1o-2 lo-’ 1 10 102 103

1 1 1 1 1 1 1 1

Experiment II TC 0 closest lower whole-log concentration lowest effective whole-log concentration closest higher whole-log concentration

Experiment III No 1 2 2 2

TC

Experiment IV No.

2/10of effective whole-log concentratton 3/10 of effective whole-log concentration 4/10 of effective whole-log concentration YlO of effective whole-log concentration 6110 of effective whole-log concentration 7/10 of effective whole-log concentration 8/10 of effective whole-log concentration 9/10 of effective whole-log concentratton

1 1 1

1

TC

No.

0 2110 log concentration lower l/10 log concentration lower l/10 effective log concentration l/10 log concentration higher

1 1 1

1

TC Test substanceconcentration in mg/L. No Number of Hydra dosed.

5. The three supernatants are pooled and centrrfuged for 3 mm at 200g. This time the supernatant IS decanted away to leave fluid approximately equal to the pellet volume, which is then resuspended.

6. Using a Hamilton syringe the slurry 1sdrawn mto 6.5 cm-long segments of ID 0.58 mm polyethylene tubing (about 20). Tubes are filled with 3.5 cm of slurry with 0.5 cm of air at the bottom. 7. The tubmg is placed mto small Eppendorf tubes, pinched at one end by the cap, and centrifuged for 3 mm at 1OOg.The Eppendorfs are stood upright afterward for 10 mm before removmg

tubes.

1 3

3

3 3

Adult

Toxicity

Assay

Clubbed tentacles

Normal adult POlYP

ii!

shortened body and tentacles

Tulip (smgle endpomt)

Disintegrated

“Embryo” development (assay evaluation time in hours)

(4)

Pellet

(18)

Laminar and hollowmg

Fig. 1. Hydru attenuata:

(26)

Tentacle buds

(42)

Elongated tentacle buds

(66)

Hypostomes tentacles

on

Polyps

stages of normal embryo development and adult toxicity.

The Hydra attenuata Assay

325

8. The tubes are cut below the pinched area and a Hamilton syringe is reattached. 9. The short columns of randomly associated Hydra cells are expelled into a test well (2/well) containing 4 mL reaggregation medium, 150 mg/L Amikacin, and logarithmic interval concentrations of test substance. 10. Exposure is as for the adult I-Iy&u (Table 1). Fresh test compound is added to wells (and Amikacin) with each medium change. 11. The process is carried out in high molarity reaggregation medium that is undesirable for adult Hydra. At 4 h after pellet formation the 70 mosM medium is diluted to 35 mosM. The irregular clump of cells should now be a solid, smooth sphere. 12. At 18 h, the medium is diluted again to 17.5 mosM. Now the sphere (unaffected) shows signs of becoming hollow. 13. At 26 h, the medium 1sreplaced with Hydra medium (5 mosM). The hollowing process should be complete. 14. Once a day from now on (at 42 and 66 h) a fresh solution is provided. During this period tentacle buds emerge on the surface of the sphere. Under normal conditions they elongate, hypostomes are formed by d 4, and body columns elongate and are sculpted by d 5 (Fig. 1). The assay is stopped after 90 h. 15. The endpoint measured IS total dissolution of the “embryo.” This can (or cannot) occur at any stage, depending on the test substance.The minimum effective concentration (MEC) is established.

3.4. Evaluation 1. Level IV experiments (Section 3.2., step 4) are often repeated to ensure that the minimal effective concentrations, to the nearest l/10 log, for both the adult and “embryo” are accurate. 2. The ratio of the adult MEC (A; adult) to the “embryo” MEC (D; development) is calculated. A/D ratio = index of developmental toxicity hazard 3. An A/D ratio
where (3-5).

and relevance of the Hydra assay is discussed else-

Davies and Freeman

326 4. Notes

1, An alternatrve version of the Hydra assayhas been proposed by Wilby and coworkers (6,7). Effects of test substanceson intact polyps are compared with those on the regenerating isolated adult digestive regron. They claim that this method measures effects on redifferentiation (regeneration) rather than on reorganization. 2. More recently it has been possible to introduce a mtxed-function oxrdase (MFO) system to the Hydra assay to test for agents that require metabolism to become biologtcally active (8).

References 1 Johnson,E. M , Got-man,R M., Gabel, B. E G., and George, M. E (1982) The Hydra attenuata system for detection of teratogemc hazards. Terat. Cart

Mutat.

2,263-276. 2. Johnson, E. M and Gabel, B. E. G (1983) The role of an artificial embryo m detecting potential teratogemc hazards, in Handbook of Experimental Pharmacology, vol. 65 (Johnson, E M. and Kochhar, D M., eds ), Springer-Verlag, New York, pp. 335-347 3 Johnson, E M., Newman, L. M., Gabel, B. E. G , Boerner, T. F , and Dansky, L. A (1988) An analysis of the Hydra assay’s applicability and reliability as a developmental toxicity prescreen. J. Am Call. Toxicol. 7(2), 111-126 4 Collins, T. F. X. (1987) Teratologmal research using in vitro systems V.

Nonmammalian model systems.Environ. Health Perspect 72,237-249. 5 Newman, L M., Giacobbe, R. L , Fu, L -J., and Johnson, E M. (1990) Developmental toxicity evaluation of several cosmetic ingredients m the Hydra assay. J. Am. Coil. Toxicol. 9(3), 361-365. 6. Wilby, 0 K., Newall, D. R., and Tesh, J M. (1986) A Hydra assay as a prescreen for teratogenic potential. Fd. Chem. Toxicol. 24(6/7), 651-652. 7 Wilby, 0. K. and Tesh, J. M (1990) The Hydra assay as an early screen for teratogenie potential. Tox~ol. In Vitro 4(4/5), 582-583. 8. Newman,L. M., Johnson,E. M., Giacobbe,R. L., and Fu, L.-J. (1990) The in vitro activation of cyclophosphamide m the Hydra developmental toxicology assay Fund Appl. Toxicol. 15,488-499