A protocol and guide for the in vitro rat hepatocyte DNA-repair assay

Mutation Research, 189 (1987) 113-121

113

Elsevier

M T R 01170

A protocol and guide for the in vitro rat hepatocyte DNA-repair assay Byron E. Butterworth 1, John Ashby 2, Edilberto Bermudez 1, Daniel Casciano 3, Jon Mirsalis 4, Gregory Probst 5 and Gary Williams 6 Chemical Industry Institute of Toxicology, P.O. Box 12137, Research Triangle Park, NC 27709 (U.S.A.), 2 Central Toxicology Laboratory, ICI, Alderley Park, Nr. Macclesfield, Cheshire SKIO 4TJ (Great Britain), 3 National Center for Toxicological Research, Jefferson, A T 72079 (U.S.A.), 4 S R I International, 333 RavenswoodAvenue, Menlo Park, CA 94025 (U.S.A.), 5 Division of Toxicology, Lilly Research Laboratories, Eli Lilly and Company, Greenfiela~ IN 46104 (U.S.A.) and 6 American Health Foundation, Naylor Dana Institute, 1 Dana Road, Valhalla, N Y 10595 (U.S.A.) (Received 13 January 1987) (Accepted 16 January 1987)

Keywords: Protocol; Guide, DNA-repair assay; Hepatocyte, in vitro rat; Hepatocyte DNA-repair assay.

Contents Summary ................................................................................... Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Liver perfusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Preparation of hepatocyte cultures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Labelling the cultures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Autoradiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Grain counting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Calculations and data analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WEI (Williams medium E, incomplete) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . W E C (Williams medium E, complete) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.5 m M E G T A perfusion solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Collagenase perfusion solution (100 u n i t s / m l ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . [3H]Thymidine solution (10 g C i / m l ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methyl-green pyronin Y solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

113 114 114 114 115 116 116 117 118 119 120 120 120 120 120 120 121

Summary The in vitro rat-hepatocyte DNA-repair assay is a valuable tool in assessing the genotoxic activity of chemical agents. An advantage of the assay is that the target cells themselves are metabolically competent, so that the patterns of metabolic activation and detoxification closely reflect those in the whole animal. This article provides a typical procedure and guidelines for conducting the rat in vitro hepatocyte DNA-repair assay.

EMS liaison: Dr. Robert Naismith, Chairman, Committee E-47.09. A S T M liaison: A n n e M. McKlindon, Staff Manager ASTM, 1916 Race St., Philadelphia, PA 19103 (U.S.A.).

Correspondence: Dr. Robert W. Naismith, P h a r m a k o n Research International, Inc., Waverly, PA 18471 (U.S.A.).

0165-1218/87/$03.50 © 1987 Elsevier Science Publishers B.V. Biomedical Division)

114 Introduction

Measurement of chemically induced DNA repair is a means of assessing the ability of a chemical to reach and alter the DNA. DNA repair is an enzymatic process that involves the recognition and excision of a DNA-chemical adduct followed by DNA-strand polymerization and ligation to restore the original primary structure of the DNA (Roberts, 1978). This process can be quantitated by measuring the amount of labeled thymidine incorporated into the nuclear DNA of cells that are not in S-phase and is often called unscheduled D N A synthesis (UDS) (Rasmussen and Painter, 1966). Numerous assays have been developed for the measurement of chemically induced DNA repair in various cell lines and primary call cultures from both rodent and human origin (reviewed by Mitchell and Mirsalis, 1984). The primary rathepatocyte DNA-repalr assay developed by Williams (1976) has proven to be particularly valuable in assessing the genotoxic activity and potential carcinogenicity of chemicals (Williams et al., 1982; Mitchell et al., 1983). Genotoxic activity is often produced by reactive metabolites of a chemical. The in vitro rat-hepatocyte assay provides a system in which a metabolically competent cell is itself the target cell for measured genotoxicity. Most other short-term tests for genotoxicity employ a rat-liver homogenate ($9) for metabolic activation, which differs markedly in many important ways from the patterns of activation and detoxification that actually occur in hepatocytes. An extensive literature is available on the use of the in vitro hepatocyte DNA-repair assay (Bermudez et al., 1979; Butterworth et al., 1983; Casciano and Gaylor, 1983; Conaway et al., 1984; Kornbrust and Dietz, 1985; Lefevre and Ashby, 1985; Mori et al., 1984a,b; Probst et al., 1981; Tong et al., 1981; Waters et al., 1984; West et al., 1984; Williams, 1977, 1978, 1984; Williams and Laspia, 1979; Williams et al., 1977). The purpose of this article is to provide a typical procedure and guidelines for conducting the rat in vitro hepatocyte DNA-repair assay. The procedures presented here are based on similar protocols that have been shown to be reliable and are also acceptable (Butterworth, 1986; Casciano and Gaylor, 1983; Probst et al., 1981; Maslansky

and Wil-liams, 1985; McQueen and Williams, 1985; Waters et al., 1984). In some cases brands of supplies are mentioned by name. These are meant only as examples and not as an endorsement. Other manufacturers or suppliers of equivalent media, reagents, or equipment are acceptable. Procedures

Liver perfusion (1) All personnel must be knowledgeable in the procedures for safe handling and proper disposal of carcinogens, potential carcinogens, and radiochemicals. Disposable gloves and lab coats must be worn. (2) Any strain or sex of rat may be used. The largest database is for male Fischer-344 rats. Young adult animals are preferred. (3) The rat is anesthetized by i.p. injection with 0.2 ml Nembutal Sodium Solution (50 m g / m l ) per 100 g body weight 10 min prior to the perfusion procedure. Other proven anesthetics are also acceptable. Make sure that the animal is completely anesthetized, so that it feels no pain. (4) The abdomen is thoroughly wetted with 70% ethanol and wiped with gauze for cleanliness to discourage loose fur from getting on the liver when the animal is opened. (5) A "V"-shaped incision is made through both skin and muscle from the center of the lower abdomen to the lateral aspects of the rib cage. Do not puncture the diaphragm or cut the liver. The skin and attached muscle are folded back over the chest to reveal the abdominal cavity. (6) A tube approximately 1 cm in diameter is placed under the back to make the portal vein more accessible. (7) The intestines are gently moved out to the right to reveal the portal vein. The tube under the animal is adjusted so that the portal vein is horizontal. (8) A suture is put in place (but not tightened) in the center of the portal vein and another around the vena cava just above the right renal branch. (9) Perfusions are performed with a peristaltic pump, the tubing of which is sterilized by circulation of 70% ethanol followed by sterile water. A valve is placed in the line so that one may switch from the EGTA solution to the collagenase solu-

115 tion without disrupting the flow. Solutions are kept at 3 7 ° C with a minimal distance to the animal and shortest possible tube length to minimize cooling in the line. (10) A typical pump would be the Cole-Parmer Masterflex pump (K7562-10) with a Masterflex pump head (K-7020-50) and Cole-Parmer silicone tubing (K-6411-43). (11) The flow of the 3 7 ° C E G T A solution is begun at 8 m l / m i n and run to waste. (12) The portal vein is cannulated with a 20 GA 1 1 / 4 " Angiocath (Deseret No. 2878) about 3 mm below the suture. The inner needle is removed and the plastic catheter is further inserted to about 1 / 3 the length of the vein and tied in place by the suture. Blood should emerge from the catheter. The tube with the flowing E G T A is then inserted in the catheter (avoid bubbles) and taped in place. (13) The vena cava is immediately cut below the right renal branch and the fiver is allowed to drain of blood for 1.5 min. The liver should rapidly clear of blood and turn a tan color. If all lobes do not clear uniformly, the catheter has probably been inserted too far into the portal vein. (14) The suture around the vena cava is then tightened and the flow increased to 20 m l / m i n for 2 min. The fiver should swell at this point. In some cases gentle massaging of the fiver or adjusting the orientation of the Angiocath may be necessary for complete clearing. At this point the vena cava above the suture may be clipped to release some of the pressure in the fiver. (15) The flow is then switched to the 3 7 ° C collagenase solution for 12 min. During this period the fiver is covered with sterile gauze wetted with sterile saline or WEI and a 40-W lamp is placed 2" above the fiver for warming. It is valuable to screen each new batch of collagenase to be assured that it will function properly. (16) The perfusate is allowed to flow onto the paper and is collected by suction into a vessel connected via a trap to the vacuum fine. (17) After the perfusion is over, the catheter and gauze are removed. The fiver is carefully removed by (1) cutting away the membranes connecting it to the stomach and lower esophagus, (2) cutting away the diaphragm, and (3) cutting any remaining attachments to veins or tissues in the abdomen.

(18) The fiver is held by the small piece of attached diaphragm and rinsed with sterile saline or WEI. (19) The fiver is placed in a sterile petri dish and taken to a sterile hood to prepare the cells.

Preparation of hepatocyte cultures (1) The perfused liver is placed in a 60-mm petri dish and rinsed with 37 ° C WEI. Extraneous tissues (fat, muscle, etc.) are removed. (2) The liver is placed in a clean petri dish and 30 ml of fresh collagenase solution at 3 7 ° C is added. (3) Carefully make several incisions in the capsule of each lobe of the liver. Large rips in the capsule lead to large unusable clumps of hepatocytes. (4) Gently comb out the cells, constantly swirling the liver while combing. A sterile metal doggrooming comb with teeth spaced from I to 3 mm or a hog bristle brush work well. (5) When only fibrous and connective tissue remain, add 20 ml cold WEI and transfer the cell suspension to a sterile 50-ml centrifuge tube (Coming No. 25331) using a wide-bore sterile pipet (Falcon No. 7536). Some laboratories report successful hepatocyte preparations when steps 5-9 are conducted with media at room temperature or heated to 37 ° C. (6) Allow the cells to settle on ice for 5-10 min until a distinct interface is seen. Carefully remove and discard the supernatant by suction. (7) Bring the cells to 50 ml with cold WEI. Resuspend the cells by pipeting with a wide bore pipet. Gently pipet the suspension through a 4-ply layer of sterile gauze (Johnson and Johnson No. 8-2317, 4 x 4) into a sterile 50-ml centrifuge tube. (8) Centrifuge the cells at 50 x g for 5 rain and discard the supernatant. Gently resuspend the pellet in ice-cold WEI with a wide-bore pipet. (9) Some laboratories keep the cells on ice until ready for use. The cells should be used as soon as possible. (10) Determine viability and cell concentration by the method of trypan blue exclusion. The preparation should be primarily a single-cell suspension with a viability of over 60% for control cultures. With the proper technique, viabifities of about 90% can routinely be obtained.

116

(11) A Thermanox No. 5415 25-mm round plastic coverslip No. 1 1 / 2 (Lux Scientific Corp.) is placed into each well of Linbro 6-well culture dishes (Flow Labs). Thermanox No. 5408 10.5 × 22 mm plastic coverslips and Lux No. 5218 26 × 33 mm 8-chamber culture dishes can also be used. Be sure to keep the proper side up as marked on the package. 4 ml of WEC is added to each well. Hepatocytes will not attach to glass unless the slides have been boiled. The use of collagen-coated thermanox coverslips improves cell attachment and morphology. (12) These procedures yield preparations consisting primarily of hepatocytes. Approximately 400 000 viable cells are seeded into each well and distributed over the coverslip by shaking or stirring gently with a plastic 1-mi pipet. Glass pipettes can scratch the coverslips. (13) The cultures are incubated for 90-120 min in a 3 7 ° C incubator with 5% CO2, 95% relative humidity, to allow the cells to attach.

Labelling the cultures (1) After the attachment period the cultures are washed once with 4 ml WEI per well to remove unattached cells and debris. This is done by tilting the culture slightly, aspirating the media, and adding the fresh media at 37 ° C. Be careful not to direct the stream from the pipet directly onto the cells. (2) Chemical solutions are prepared in 3H-WEI (10 g C i / m l [3H]thymidine). Serial dilutions are generally employed. Any solvent, such as dimethyl sulfoxide (DMSO) or ethanol, should not exceed a 1% final concentration. Concentrations should be chosen that go just beyond cytotoxicity to about 1000-fold below the cytotoxic concentration. Typical concentration ranges are from 1 m M to 0.001 mM. Relatively insoluble substances should be tested up to their limit of solubility. Freely soluble, non-toxic chemicals should not be tested at concentrations beyond 100 mM. Dimethylnitrosamine produces a strong response at the noncytotoxic dose of 10 raM. In contrast, 1,6-dinitropyrene induces D N A repair at concentrations as low as 0.00005 mM. (3) The WEI is removed and replaced with 2 ml of [3H]thymidine solution containing the dissolved test chemical. The cultures are then placed in the

incubator for 16-24 h. During this period the compound will be metabolized. If D N A damage occurs, it will be repaired, resulting in incorporation of the [3H]thymidine. (4) Cultures are washed twice with 4 ml WEI per well. (5) media is replaced with 4 ml of a 1% sodium citrate solution and the cultures are allowed to stand for 10 min to swell the nuclei. This step can be omitted, but the nuclei will be slightly smaller. (6) The sodium citrate solution is replaced with 3 ml of a 1 : 3 acetic acid : absolute ethanol solution and let stand for 10 min to fix the cells. This is repeated twice more for a total fixing time of at least 30 rain. (7) Wells are washed 2-6 times each with deionized distilled water. (8) Coverslips are removed from the wells and placed cell-side-up on the edge of the dish covers to dry in a dust-free location. (9) When dry, coverslips are mounted cell-sideup on microscope slides with Krystalon or Permount. Coverslip should be mounted about 1 cm from the unfrosted end of the slide. Each slide is given a unique identifying number. (10) At this point the cultures can be examined for gross cytotoxicity. In pilot experiments if chemical treatment at the higher doses has resulted in there being no cells on the slides, they need not be subjected to autoradiography.

Autoradiography (1) Any proven autoradiographic technique that yields silver grains in proportion to the amount of incorporated labeled thymidine may be used. Presented here is a typical procedure. (2) All steps involving emulsion should be done in total darkness. If absolutely necessary a No. 1 red (Kodak) safelight filter may be used sparingly. (3) Shdes for each dose are mounted in plastic slide grips (Lipshaw). Duphcate slides may be held in reserve. (4) A 50-ml disposable plastic beaker is mounted with tape into a slightly larger jar full of water. This assembly is then placed into a 4 2 ° C water bath and allowed to reach the bath temperature. (5) Kodak NTB-2 emulsion is most commonly used. The emulsion is used undiluted or can be

117 used diluted 1 : 1 with distilled water. If the emulsion is diluted, care should be taken to use ultrapure water, thoroughly mix the solution and avoid formation of air bubbles. Undiluted emulsion saves a step and provides slightly higher grain counts. Emulsion is melted in a 3 7 ° C incubator for at least 3 h. Gently pour 40-50 ml of the emulsion into the dipping cup. The unused portion can be resealed and stored under refrigeration. (6) Dip a test slide. Briefly turn on the No. 1 Red (Kodak) Safelight and hold the slide up to it to make sure that there is enough emulsion in the cup to cover the cells and that there are no bubbles in the emulsion. Air bubbles can be removed from the surface of the emulsion by skimming the surface with a glass slide. Turn off the Safelight. (7) Dip each group of slides by lowering them into the cup until they touch the bottom. Pull the slides out of the emulsion with a smooth action to a 5-sec count. Touch the bottom ends of the slides to a pad of paper towels to remove the bead of emulsion on the bottom. Remember, all of these steps are taking place in total darkness. Do not reuse the used emulsion. (8) Hang the slide holders in a vertical position in a rack in a light-tight box for 3-12 h to let the emulsion dry. Pack the slides into 'light-tight slide boxes that contain a false bottom packed with desiccant. Seal the boxes with black electrical tape and wrap them in aluminum foil to assure no light leaks. (9) Store the slides absolutely dry at 4 ° C to - 2 0 ° C for a set amount of time. 7-14 days are most common. Shorter times yield lower backgrounds. Longer times produce higher counts. (10) After the exposure allow the slide boxes to thaw at room temperature for at least 3 h. In the dark, place the slides into a rack suitable for developing and staining the slides. (11) Develop the slides at 1 5 ° C (56°F) for 3 min in Kodak D-19 developer. Tap the rack gently to the bottom of the developing dish several times to dislodge any air bubbles on the slides. (12) Rinse slides 30 sec in 1 5 ° C water, then fix in Kodak Fixer (not Rapid-Fix) for 5 min with agitation every 60 sec. Wash the slides in a bat.h with gently running water for 25 min. Do not hit the slides directly with the water stream. (13) Slides can be stained while still wet from

development. Dip into methyl green Pyronin Y solution for 10-20 sec. Follow this immediately with repeated washings in water and a final rinse in distilled water. Do not overstain the cells. Cells should have faint blue nuclei and pink sytoplasms. Overstained cells make automatic grain counting difficult. Other stains are also acceptable. Remember, the cells are still exposed at this point. Take care not to touch the slide surface. (14) Allow the slides to air dry for a few hours. Mount a 25-mm square coverslip (Coming) over the round coverslip using a thin layer of Krystalon or Permount. Keep the slides flat overnight to dry. The are now ready for grain counting.

Grain counting (1) Grain counting can be done by hand. This, however, is a tedious process. If the assay will be used routinely, an automated counting system is recommended. (2) Grain counting is best accomplished with an automated system such as the A R T E K model 982 Colony Counter interfaced to a Zeiss Universal microscope with an A R T E K high resolution TV camera. Data can be fed directly into a computer such as the VAX 11/780 via an A R T E K BCD-RS232 Onmi-interface. Any proven system that accurately counts the grains, however, is acceptable. (3) Normally 20-50 cells are counted per slide, 1-3 slides per treatment, 3 Expts. per data point. In an initial screening experiment in which multiple doses are being examined, only noncytotoxic doses need be counted and repeats only need cover the active, noncytotoxic concentrations. (4) Counting usually requires a 100 × objective under oil immersion. An optivar (Zeiss) can be employed to further increase magnification. (5) Each slide is examined to make sure that the culture as a whole was viable. Signs of toxicity are the absence of cells or pyknotic (small, darkly stained) cells. Note the highest concentration of chemical that was not toxic to the cultures as a whole. (6) A patch of cells is selected as a starting point and cells are scored in a regular fashion by bringing new cells into the field of view, moving only the X-axis. If the desired number of cells have not been scored before coming to the edge of

118 the slide, the stage is moved 1-2 fields on the Y-axis and counting resumes in the opposite X-direction, parallel to the first line. If upon visual scanning of the slide there appears to be any difference in response in different areas of the slide, then the counting should be done selecting patches of cells from several areas of the slide. (7) The following criteria are used to determine if a cell should or should not be counted. (a) Cells with abnormal morphology, such as those with pyknotic or lysed nuclei, should not be counted. (b) Isolated nuclei not surrounded by cytoplasm should not be counted. (c) Cells with unusual staining artifacts or in the presence of debris should not be counted. (d) heavily labeled cells in S-phase should not be counted. (e) All other normal cells encountered while moving the stage must be counted without regard as to their apparent response. (8) Counts are generally made in the mode that counts the area of the grains. This allows patches of grains that are touching to be counted without being mistaken by the machine as a single grain. When using the area mode, a correction factor to convert to grain counts must be used. This conversion factor must be determined for the particular counting set-up and configuration being used. To do so, count a number of areas (10-30 discrete grains/aperture) on the count mode and manually to determine the actual number of silver grains. Now, perform a machine area count on the same aperture area. After counting 20-30 areas from at least two different slides add all the actual counts and all the area counts. The conversion factor is calculated as: C=

actual number of grains (total) measured area of grains (total)

Then, machine counts can be converted to actual grains by multiplying by C. (9) For each cell the following procedure is used: (a) The sensitivity of the machine is adjusted so that only grains are being counted and so that the configuration is the same as when the conversion factor was calculated. (b) Place the aperture directly over the nucleus and adjust to the same size as the nucleus.

(c) Press the count button to record the nuclear counts. (d) Keeping the aperture the same size, count an area over the cytoplasm that is adjacent to the nucleus and which appears to have the highest grain counts. Press the count button to record the cytoplasmic counts. Alternatively, one may use the highest of 3 independent cytoplasmic counts as the cytoplasmic background. In general, cytoplasmic counts are relatively uniform throughout the cytoplasm. Accordingly, experience shows that the procedure of taking a single cytoplasmic count saves time is consistent, and yields the same answer that would be obtained using the highest of 3 counts. The conservative approach of using the highest cytoplasmic background is the one most often used in the published literature. Using the average of 3 independent cytoplasmic counts will yield a cytoplasmic count that is only slightly lower than the previous method. Both methods yield consistent results. The single cytoplasmic count method is recommended because of the ease of counting. (e) Subtract the cytoplasmic count from the nuclear count to give the net grains/nucleus (NG) for that cell. In the case of control cells, there will usually be more grains per unit area in the cytoplasm than in the nucleus so that the N G will be a negative number. This must be reported as such. Continue counting the desired number of cells in the same manner. (f) At all times remember that actual grains over the nucleus are being determined. If a spurious count is observed, it should be corrected before the data are calculated. If Good Laboratory Practice is being followed, any change in the dataset should be noted for the record.

Calculations and data analysis (1) Alterations in the following parameters will affect the N G observed: (a) The concentration and specific activity of [3H]thymidine used. (b) The length of cell labeling. (c) The type and dilution of emulsion. (d) The autoradiographic exposure time. (2) Thus, in data analysis one must either use a published procedure where the expected values for negative and positive responses are known, or this must be determined for the conditions being used.

119 (3) For the conditions described here, greater than or equal to 5 N G has been chosen as a conservative estimate as to whether a particular cell is responding or is " i n repair". In chosing a N G cutoff for designating an individual cell in repair for other conditions, one should select a value where only about 2% of the cells in historical control cultures would be classified as responding. (4) The following should be calculated for each slide: (a) The population average N G _+ S.D. (cell to cell). (b) The percent of cells responding or in repair. (c) The population average N G _+ S.D. (cell to cell) for the subpopulation of cells that are in repair (optional). (5) The following should be calculated for each cell preparation: (a) The population average N G + S.E. (slide to slide). (b) The percent of cells responding or in repair + S.D. (slide to slide). (c) The population average N G + S.E. (slide to slide) for the subpopulation of ceils that are in repair (optional). (6) Data from one cell preparation from one animal represents an n of one. Experiments must be repeated at least once and preferably twice with different cell preparations from different animals. (7) Repeat values should be presented side by side in reports or publications. Alternatively the following can be calculated for each data point. (a) The population average N G + S.E. (preparation to preparation). (b) The percent of cells responding or in repair + S.E. (preparation to preparation). (c) The population average N G + S.E. (preparation to preparation) for the subpopulation of cells that are in repair (optional). (8) A negative (solvent or untreated) control and a genotoxic positive control chemical should be included in every experiment. (9) Using conditions similar to those described here with a cutoff of at least 5 N G to classify an individual cells as in repair, control values in the range from - 1 0 to - 2 N G (population average) with 0-10% in repair should be expected. (10) For the positive controls 10 m M dimethylnitrosamine or 0.001 mM 2-acetylaminofluorene, one might expect values of 15-30 N G with 70-100% of the cells with greater than or equal to 5 NG. For those conditions described here, if a

chemical yields greater than or equal to 5 N G (population average) and greater than or equal to 20% of cells responding, the response is considered positive. A population average between 0 N G and 5 N G would be considered a marginal response. A positive dose-response relationship in both N G and the percentage of cells in repair are required additional information to confirm a positive response for counts below 5 NG. (11) One of the most troublesome aspects of this assay is that high cytoplasmic backgrounds are often observed that obscure any realistic evaluation of nuclear counts. Some laboratories consider an experiment invalid if cytoplasmic background counts of control cultures exceed 30 grains per nuclear sized area. Procedures to lower the background include lowering the concentration of [3H]thymidine, incubating the cells for a shorter period of time, using fresh [3H]thymidine, and decreasing the autoradiographic exposure time. (12) If the chemical treatment has removed the cytoplasm or has lowered the cytoplasmic grain counts to zero, this must be noted in the final report and the results viewed with caution. In this case, any small increase in N G (less than 5 NG) may have been due only to the lowered cytoplasmic background and the results are suspect. (13) Within an experiment, the cytoplasmic grain count usually remains reasonably constant over a wide range of test agent concentrations, even when the nuclear grain count increases due to DNA repair. Under some test conditions, the cytoplasmic grain count increases in a dose-related manner, independent of the nuclear count, which may also increase. In such situations, it is difficult to apply the normal criteria for a positive response, especially if weak (less than 5 NG). In cases where the cytoplasmic count increases by more than double the control level, this must be noted in the final report and derived N G data should not be considered in isolation as valid evidence for chemically induced UDS. Increases in the cytoplasmic grain count may be due to mitochondrial D N A synthesis or repair. In the absence of a precise understanding of these effects, caution in the interpretation of such data should be exercised. (14) Casciano and Gaylor (1983) have de-

120 scribed statistical criteria for evaluating chemicals as positive or negative in the in vitro hepatocyte DNA-repair assay for the conditions that they employ in their laboratory. The criteria involve both the N G count and the percent of cells in repair. It is important to run several control cultures over a period of time so that the historical control baseline can be determined for those conditions being used in any particular laboratory. In this assay, the control cells come from the same preparation as the treated cells, and thus represent a true concurrent control. The unpaired t-test for the equality of two means, with the N G counts (population average) of the individual slides as the unit of measure, is an appropriate statistical test for these data. (15) The probable reason that control N G values tend to be less than zero is that the cytoplasm (and the components therein producing the cytoplasmic background) is slightly thinner over the nucleus compared to the rest of the cell as it sits on the substrate. N G counts may vary as the result of compound-related effects on cytoplasmic grain counts. Consequently, no result may be considered positive unless the compound actually produces more grains over the nucleus than over the cytoplasm, i.e. a N G value greater than zero. Knowledge of the biology of this assay dictates that in order to have any confidence in a positive D N A repair response, the treatment must produce nuclear counts beyond the cytoplasmic background. Thus, for any statistical test employed, a lower limit of at least 0 N G is required for a positive response. (16) In any final report the authors should clearly state their conclusions as to whether a compound was positive or negative.

Solutions Presented here are solutions that work using the procedures described above. Other suppliers of these media or reagents are acceptable. Any specific reference to a particular product or manufacturer is meant only as an example and not an endorsement.

WEI (Williams medium E, incomplete) 500 ml Williams medium E (Flow Labs 12-502-54) 5 ml sterile 200 mM L-glutamine (Flow Labs 16-801-49)

0.5 ml of 50 m g / m l gentamycin sulfate (Sigma G-7507) Some laboratories use lower concentrations of gentamycin to minimize toxicity of the antibiotic to the cells.

WEC (Williams medium E, complete) 180 ml WEI 20 ml heat-inactivated fetal bovine serum (Reheis or Gibco 200-6140) Heat inactivation 30 min at 56 ° C, Freeze aliquots.

0.5 mM EGTA perfusion solution 100 ml Hanks balanced salt solution without Ca 2+ or Mg 2+ (Flow 18-104-54) 19 mg EGTA (ethylene glycol-bis (fl-amino ethyl ether) N, N'-tetraacetic acid (Sigma E-4378) Dissolve the E G T A in 0.1 ml 2 N NaOH. 0.5 ml 2 M HEPES (Calbiochem 391338) 0.1 ml of 50 m g / m l gentamycin sulfate (Sigma G-7507) Filter sterilize.

Collagenase perfusion solution (100 units / ml) 500 ml WEI 2.5 ml 2 M HEPES 0.1 ml 2 N N a O H 50,000 units Type 1 collagenase (Worthington or Sigma C-0130) Leave in 37 ° C bath until dissolved (20-30 min) Filter sterilize. This solution should be made up no more than 24 h prior to use.

[3H]thymidine solution (10 ~tCi/ml) 100 ml WEI 1000 t~Ci (1 ml) [3H]thymidine([methyl-3H] thymidine, Amersham TRK.418, 40-60 C i / mmole) 0.5 ml sterile 2 M HEPES Make up just prior to use. [3H]thymidine should be stored refrigerated and be no older than two months.

Methyl-green pyronin Y solution Add 7.45 g N a 2 H P O 4 to 66 ml methanol. Add distilled water to bring volume to 263 ml. Stir until dissolved. Add 5 g citric acid to 59 ml methanol. Add distilled water to bring volume to

121 240 ml. W h e n b o t h s o l u t i o n s a r e d i s s o l v e d , m i x a n d a d d 12.5 g l p h e n o l , 125 m g r e s o r c i n o l a n d 5 g m e t h y l - g r e e n p y r o n i n Y ( R o b o z S u r g i c a l Co.). A l l o w to sit for 2 w e e k s b e f o r e use. K e e p in a d a r k b o t t l e a n d filter b e f o r e e a c h use. D i s c a r d a f t e r a p p r o x i m a t e l y 6 m o n t h s o r w h e n a n o b v i o u s dec r e a s e in s t a i n i n g i n t e n s i t y is o b s e r v e d .

References Bermudez, E., B.E. Butterworth and D. Tillery (1979) The effect of 2,4-diaminotoluene and isomers of dinitrotoluene on unscheduled DNA synthesis in primary rat hepatocytes, Environ. Mutagen., 1, 391-398. Butterworth, B.E. (1986) measurement of chemically induced DNA repair in hepatocytes in vivo and in vitro as an indicator of carcinogenic potential, in: E.J. Rauckman and G.M. Padilla (Eds.), The Isolated hepatocyte: Use in Toxicology and Xenobiotic Biotransformations, Academic Press, New York,. Butterworth, B.E., L.L. Earle, S. Strom, R. Jirtle and G. Michalapoulos (1983) Induction of DNA repair in human and rat hepatocytes by 1.6-dinitropyrene, Mutation Res., 122, 73-80. Casciano, D.A., and D.W. Gaylor (1983) Statistical criteria for evaluating chemicals as positive or negative in the hepatocyte/DNA repair assay, Mutation Res., 122, 81-86. Conaway, C.C., C. Tong and G.M. Williams (1984) Evaluation of morpholine, 3-morpholinone, and N-substituted morpholines in the rat hepatocyte primary culture/DNA repair test, Mutation Res., 136, 153-157. Kornbrust, D., and D. Dietz (1985) Aroclor 1254 pretreatment effects on DNA repair in rat hepatocytes elicited by in vivo or in vitro exposures to various chemicals, Environ. Mutagen., 7, 857-870. Lefevre, P.A., and J. Ashby (1985) Investigations into the reported abihty of cimetidine to initiate UDS in rat hepatocyte primary cultures, Environ. Mutagen., 7, 833-837. Maslansky, C.J., and G.M. Williams (19850 Methods for the initiation and use of hepatocyte primary cultures from various rodent species to detect metabolic activation of carcinogens, in: M. Webber and L. Sekely (Eds.), In Vitro Models for Cancer Research, CRC Press, pp. 43-60. McQueen, C., and G.M. Williams (1985) Methods and modifications of the hepatocyte primary culture/DNA repair test, in: H.A. Milman and E.K. Weisburger (Eds.), Handbook of Carcinogen Testing, Noyes Publications, Park Ridge, N J, pp. 116-129. Mitchell, A.D., and J.C. Mirsalis (1984) Unscheduled DNA synthesis as an indicator of genotoxic exposure, in: A.A. Ansari and F.J. de Serres (Eds.), Single-Cell Mutation Monitoring Systems, Plenum, New York, pp. 165-216. Mitchell, A.D., D.A. Casciano, M.L. Meltz, D.E. Robinson, R.H.C. San, G.M. Williams and E.S. von Halle (1983) Unscheduled DNA synthesis test: A report of the Gene-Tox Program, Mutation Res., 123, 363-410. Moil, H., K. Kawai, F. Ohbayashi, T. Kuniyasu, M. Yamazaki,

T. Hamasaki, and G.M. Williams (1984a) Genotoxicity of a variety of mycotoxins in the hepatocyte primary culture/ DNA repair test using rat and mouse hepatocytes, Cancer Res., 44, 2918-2923. Mori, H., F. Ohbayashi, I. Hirono, T. Shamada and g.M. Williams (1984b) Absence of genotoxicity of the carcinogenic sulfated polysaccharides carrageenan and dextran sulfate in mammalian DNA repair and bacterial mutagenicity assays, Nutrition and Cancer, 6, 92-97. Probst, G.S., R.E. McMahon, L.E. Hill, C.Z. Thompson, J.K. Epp and S.B. Neal (1981) Chemically-induced unscheduled DNA synthesis in primary rat hepatocyte cultures: a comparison with bacterial mutagenicity using 218 compounds, Environ. Mutagen., 3, 33-43. Rasmussen, R.E., and R.B. Painter (1966) Radiation-stimulated DNA synthesis in cultured mammalian cells, J. Cell Biol., 9, 11-19. Roberts, J.J. (1978) The repair of DNA modified by cytotoxic, mutagenic and carcinogenic chemicals, Adv. Radiat. Biol., 7, 211-436. Tong, C., M.F. Laspia, S. Telang and G.M. Williams (1981) The use of adult rat liver cultures in the detection of the genotoxicity of various polycyclic aromatic hydrocarbons, Environ. Mutagen., 3, 477-487. Waters, R., J. Ashby, B. Burlinson, P. Lefevre, R. Barrett and C. Martin (1984) Unscheduled DNA synthesis, in: UKEMS (The United Kingdom Environmental Mutagen Society), Report of the UKEMS Sub-committee on Guidelines for Mutagenicity Testing, UKEMS, Swansea, United Kingdom, pp. 63-87. West, WR., P.A. Smith, P.W. Stokes, G.M. Booth, T. SmithOliver, B.E. Butterworth and M.L. Lee (1984) Analysis and genotoxicity of a PAC-polluted river sediment, in: M. Cook and A.J. Dennis (Eds.), Proceeding of the 8th International Symposium on Polynuclear Aromatic Hydrocarbons, Battelle Press, Columbus, OH, pp. 1395-1411. Williams, G.M. (1976) Carcinogen induced DNA repair in primary rat liver cell cultures; a possible screen for chemical carcinogens, Cancer Lett.., 1, 231-236. Williams, G.M. (1977) Detection of chemical carcinogens by unscheduled DNA synthesis in rat liver primary cell cultures, Cancer Res., 37, 1845-1851. Williams, G.M. (1978) Further improvements in the hepatocyte primary culture DNA repair test for carcinogens: Detection of carcinogenic biphenyl derivatives, Cancer Lett., 4, 69-75. Williams, G.M. (1984) DNA damage and repair tests for the detection of genotoxic agents, Food Additives and Contaminants, 1, 173-178. Williams, G.M., and M.F. Laspia (1979) The detection of various nitrosamines in the hepatocyte primary culture/ DNA repair test, Cancer Lett., 66, 199-206. Williams, G.M., E. Bermudez and D. Scaramuzzino (1977) Rat hepatocyte primary cell cultures, III. Improved dissociation and attachment techniques and the enhancement of survival by culture medium, In Vitro, 13, 809-817. Williams, G.M., M.F. Laspia and V.C. Dunkel (982) Reliability of the hepatocyte primary culture/DNA repair test in testing of coded carcinogens and noncarcinogens, Mutation Res., 97, 359-370.