Environmental Mutagenesis 8:727-740 (1986)

Evaluation of the Effect of Agar on the Results Obtained in the L5178Y Mouse Lymphoma Assay Mary Meyer, Karen Brock, Kay Lawrence, Bruce Casto, and Martha M. Moore

Environmental Health Research and Testing, Inc. (M. M., K. B., B.C.) and Mutagenesis and Cellular Toxicology Branch, Genetic Toxicology Division, Health Effects Research Laboratory, U.S. Environmental Protection Agency (K. L., M. M. M.), Research Triangle Park, North Carolina

The L5178Y TK+’- mouse lymphoma assay is widely used in mutagenicity testing. Trifluorothymidine-resistant (TFT‘) mutants are quantitated following growth in agar-supplemented cloning medium. In an attempt to evaluate the effect of agar on plating efficiency, we have tested several lots of Difco Noble agar (cat. No. 0142-01-8; normally used in this assay) and compared it with Baltimore Biological Laboratory (BBL) agar (cat No. 11849). We find that BBL agar gives a higher and less variable plating efficiency than any of the Noble lots tested. Colonies plated in BBL agar tend to appear significantly earlier on the plates than those cloned in Noble agar. The absolute mutant number and the induced mutant frequency quantitated from a treated culture is generally higher in BBL compared to Noble agar. To determine if this higher frequency is due to increased mutant recovery rather than “sneak through” of nonmutant cells, we isolated 97 mutants from treated cultures (44 large colonies and 53 small colonies) and 69 mutants from untreated cultures (24 large colonies and 45 small colonies) and tested them for TFT resistance. All but one (a large colony from an untreated culture) were found to be TFT‘, indicating that the mutant frequency is due to an increased mutant recovery. The spontaneous mutant frequency was quantitated for 122 untreated cultures. Showing little variation within and between experiments, the spontaneous mutant frequency yielded a mean of 57.7, with a standard deviation

Abbreviations used: 2-AAF, 2-acetylaminofluorene; B(a)P, benzo(a)pyrene; BBL, Baltimore Biological Laboratory; A, large colony (TK-’- or TFT’); MMS, methyl methanesulfonate; u, small colony (TK-’- or TFT’); TIT, trifluorothymidine (5-trifluoromethylthymidine); TFT‘, TFT resistant; TK, thymidine kinase.

This manuscript has been reviewed by the Health Effects Research Laboratory, U.S. Environmental Protection Agency, and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.

Received November 3, 1985; revised and accepted April 21, 1986.

Address reprint requests to Martha M. Moore, MD-68, Mutagenesis and Cellular Toxicology Branch, Genetic Toxicology Division, U.S. Environmental Protection Agency, Research Triangle Park, NC 27711.

0 1986 Alan R. Liss, Inc.

728 Meyeretal

of 14.4. Under our laboratory conditions, BBL agar gave reliable results, and we prefer it for use in cloning L5178Y mouse lymphoma cells.

Key words: thymidine kinase locus, L5178Y mouse lymphoma assay

INTRODUCTION

The L5178Y TK+/- mouse lymphoma assay is a short-term in vitro test that has been extensively used to evaluate compounds for their mutagenicity [Aaron et al, 1980; Amacher et al, 1979, 1980; Amacher and Paillet, 1980, 1981; Clive, 1973; Clive and Spector, 1975; Clive et al, 1972, 1979, 1980; Clive and Moore-Brown, 1979; Gold et al, 1980; Jacobson et al, 1981; Jacobson and Krell, 1979; Jotz and Mitchell, 1981; Krell and Jacobson, 1980; MacGregor et al, 1979; Meltz and Mac- Gregor, 1981; Moore et al, 1985a,b; Moore-Brown and Clive, 19791. In performing this assay, a TK+/- cell culture is treated with the test chemical. After a 4-hr exposure period, the cells are washed free of compound and incubated at 37°C to allow for phenotypic expression of newly induced mutations. Following this 48-hr expression period, the treated cells are cloned in TFT-supplemented medium, from which mutants are quantitated by their resistance to TIT. After 9-11 days of incubation at 37”C, colonies are counted on an automatic counter to determine the plating effi- ciency and mutant frequency.

Because the clonal phase provides the data for the quantitated mutant frequency of a chemical, conditions allowing the maximum number of cells to grow into colonies are essential. It is imperative that the semi-soft agar cloning medium maintain an environment for the optimal growth of TKdeficient mutant cells as well as normal cells.

We evaluated the effect of two different types of agar on the plating efficiency of mouse lymphoma cells. Several lots of the two agars were compared using four parameters: (1) plating efficiency of treated and nontreated cells, (2) relative number of countable colonies on the plates at various incubation times, (3) spontaneous and mutant frequencies quantitated, and (4) relative size distribution of the colonies.

MATERIALS AND METHODS Cells and Media

The TIC+’- 3.7.2C heterozygote of L5178Y mouse lymphoma cells [Clive and Voytek, 19771 was used. Cells were maintained in Fischer’s Medium for Leukemic Cells of Mice (GIBCO, Grand Island, NY) The medium was supplemented with penicillin-streptomycin [Clive and Spector, 19751, sodium pyruvate, sodium bicar- bonate, and pluronic F68 [Clive et al, 19791. After the medium was filter-sterilized by positive pressure (5% COZ-in-air), it was heat-inactivated for 45 min at 57°C. To support normal growth of cells, the medium was supplemented with 10% horse serum (GIBCO) that was heat-inactivated for 30 min at 57°C. Thirteen lots of Difco Noble agar and five lots of BBL agar were evaluated for their ability to support cell growth during the clonal phase. Cells were cloned in 20% horse serum with 0.37% Noble or 0.22% BBL agar (except as noted).

Effect of Agar on the Mouse Lymphoma Assay 729

Chemicals TFT, MMS, B(a)P, 2-AAF, NADP, isocitric acid, and 1 N NaOH were obtained

from Sigma Chemical Company (St. Louis, MO). A >99.9% sample of C.I. Solvent Yellow No. 33 (2-[2-quinolyl]-l,3-indanone) was obtained from the U.S. Army Inhalation Toxicology Research Institute, Lovelace Biomedical and Environmental Research Institute, Inc. (Albuquerque, NM). Aroclor 1254-induced rat liver S-9 was purchased from Microbiological Associates (Bethesda, MD).

Mutagen Treatment and Mutant Selection BBL and Noble agars were used to compare the effect of agar on cell growth

during the clonal phase. Nontreated cells as well as cells treated with MMS (15 pg/ ml), B(a)P (1.5 to 4.0 pg/ml with S-9 activation), or C.I. Solvent Yellow No. 33 (2 to 20 pg/ml with S-9 metabolic activation, and 0.1 to 50 pg/ml without exogenous activation) were used. Plating efficiency, mutant frequency, and colony size distribu- tion were evaluated for each experiment.

Three MMS and three yellow dye experiments were conducted for agar com- parison. Four production lots of Noble and one lot of BBL agar were used for experiments performed with the two test compounds. TK+'- 3.7.2C cells (6 X lo6) were treated with an appropriate stock solution of the test compound for 4 hr. For the yellow dye treatments, experiments were conducted both with and without exogenous activation. Aroclor-induced rat liver S-9 with isocitric acid and NADP cofactors were used for metabolic activation [Clive et al, 19791. Following treatment, cells were washed free of the chemical and maintained in log-phase growth for 48 hr to allow for phenotypic expression of newly induced mutants. Samples (15 ml) of treated or nontreated cells (2 x I d cells/ml) were cloned in 0.37% Noble agar or 0.22% (except as noted) BBL agar supplemented growth medium containing 20% horse serum. TK-deficient mutants were quantitated by supplementing medium with 1 pg/ ml TFT. Treated and nontreated cells were cloned without selection (200 celldplate) to determine the plating efficiency.

Colony Counting and Sizing After 9-11 days of incubation at 37"C, colonies were counted on an Artek

model 880 automatic colony counter using settings that optimize colony counts, compared to hand counts, without producing a spurious background. This counter has a size discriminator that allows the quantitation of the colony size distribution. Colony sizing curves were generated by calculating the difference between adjacent counts of up to 16 uniformly spaced settings on the size discriminator. Each of these differences was converted to a mutant frequency by correcting for the viability of the cells plated. The difference in mutant frequency between adjacent size settings was plotted as a histogram.

Evaluation of Mutants for TFT Resistance Since a larger number of TFT' colonies were quantitated in BBL agar than in

Noble agar, the stability of the TFT' phenotype was evaluated for colonies grown in BBL agar. Ninety-seven mutants from treated cultures .(53 u colonies and 44 X colonies) and 69 spontaneous mutants (45 u colonies and 24 X colonies) were analyzed.

730 Meyer et al

TFT‘ mutants were classified by eye as u or X and isolated with a Pasteur pipette. Colonies identified as u were less than approximately 0.6 mm in diameter. Colonies identified as X were greater than approximately 0.6 mm in diameter. Care was taken to withdraw only single colonies. Newly isolated mutants were immediately placed into fresh growth medium without TFT-selective pressure and allowed to incubate without agitation until judged sufficiently concentrated by visual inspection to be placed in a 50-ml centrifuge tube and incubated on a roller drum. Cells were then tested for TFT resistance; each culture was adjusted to a density of 4 X 104 cells/ml and divided in half. TFT (1 pg/ml) was added to one of each pair. Cell counts were taken after 2 days growth, and growth in the presence and absence of TFT was compared. For each test, a TK+’- control culture was used to test the ability of the TFT to completely arrest cell growth in TK-competent cells.

RESULTS AND DISCUSSION

We have been investigating the effect of varying the type of agar used in the TK+’- mouse lymphoma assay. In this analysis we considered not only the potential effects on the quantitated mutant frequency but also on the relative size distribution of the colonies. An evaluation of mutant colony size distribution may give insight into the type of genetic damage caused by the compound being evaluated [Hozier et al, 1981, 1982, 1985; Moore et al, 1985a,b]. In the assay the quantitated mutant fre- quency is determined by the number of TFT‘ mutants corrected by the percent plating efficiency. It is imperative that the plating efficiency of cells plated for viable count (200 celldplate, 3 plates) be representative of the plating efficiency of cells plated in TlT-selection (1 X lo6 celldplate, 3 plates). In considering the effect of agar on the results obtained with the TK+’- 3.7.2C mouse lymphoma assay, four parameters were considered: (1) plating efficiency of treated and nontreated cells, (2) relative number of countable colonies on the plates at various incubation times, (3) quantitated mutant frequency, and (4) relative size distribution of the colonies.

Historically, we have experienced inconsistent plating efficiencies with mouse lymphoma cells. Preliminary experiments tended to indicate excessive variability both within and between different lots of Noble agar. Because it appeared that the agar might be a significant factor in plating efficiency, we began a systematic evaluation. Table I shows the mean and standard deviation for the plating efficiency of untreated cells cloned in the 13 different production lots of Noble agar that were used at various times over approximately a 2-yr time period. Approximately 1 yr of data was obtained historically before the initiation of the comparative studies, and 1 yr of data was obtained during the comparative studies. We determined that only 2 of the 13 lots of Noble agar gave a mean plating efficiency greater than 70%. (One of these two lots had only one sample cloned.) Several of those lots used in more than one cloned sample showed a large standard deviation, reflecting the inconsistency in experimental data when cells were cloned in Noble agar.

To evaluate the effect of agar on the plating efficiency of mouse lymphoma cells, we obtained five different production lots of BBL agar. An initial experiment using 0.37% BBL agar yielded only very small colonies in a very dense, stiff medium. Based on previous experience with high densities of Noble agar, >0.45%, these results indicated that the colonies growing in the 0.37% BBL agar may have been unable to attain “normal” size because of the stiffness of the medium. A study was

Effect of Agar on the Mouse Lymphoma Assay 731

TABLE I. Evaluation of the Plating Efficiency of Untreated TK+'- 3.7.2C Cells in W Lots of Noble Agar

Percent plating efficiency" Lot x f l SD (n)

68575 1 58.8 19.4 42 706263 79.0 12.5 39 718415 47.4 22.4 7 7 18869 34.1 16.5 7

1 632345 74.5 - 1 706287 3.3 - 1 706250 7.3 - 1 70627 1 9.9

722950 55.2 19.5 10 706264 40.7 28.2 7 706265 62.5 15.9 26 722471 52.8 11.8 9

1 722887 13.5 -

'Percent plating efficiency for n samples from an untreated culture. SD, standard deviation.

-

TABLE II. Density Comparison of BBL Agar

Percent plating efficiency' - Density

Lot (96) X *1 SD (n)

C4DNMJ 0.22 75.3 6.1 7 1 0.24 80.0 -

0.26 78.7 3.1 3 0.27 82.1 5.7 6

1 0.37 21.0 -

F4DNBQ 0.22 82.4 7.1 9 0.24 79.3 4.0 6 0.26 77.2 3.7 6 0.27 74.6 11.8 6

"Percent plating efficiency for n samples from an untreated culture. SD, standard deviation.

performed using two lots of BBL agar to determine what density of agar would be optimal for clonal growth. From the results (Table II), we decided to use 0.22% BBL agar because it gave consistently high plating efficiencies. It should be noted that previous experience with lowered densities (from 0.37% to 0.20%) of Noble agar altered neither plating efficiency nor quantitated mutant frequency (Moore, unpub- lished data) [Moore et al, 1985al. Furthermore, plating medium containing 0.22% BBL agar was approximately the same stiffness as plating medium containing 0.37% Noble agar. Thus, the high plating efficiency seen with 0.22% BBL agar could not be attained by simply lowering the density of Noble agar. During our evaluation of BBL agar, five different lots were used (Table III) at various times over 1 yr. The average plating efficiency for untreated cells plated in the five lots ranged from 91.2% to 83.9%. These data, which are compiled from a large number of different cloning experiments, indicated that BBL agar might result in more consistent experimental data.

732 Meyeret al

TABLE JII. Evaluation of the hting Efficiency of Untreated TK”- 3.7.2C Cells in Five Lots of BBL Agar

Percent plating efficiency’ -

Lot X * I SD (n)

C4DNM.l 81.8 9.8 98 F4DNBQ 81.4 12.0 51 A6DOQF 81.2 4.9 30 A6DOKL 82.0 4.2 30 A6DOFM 83.9 4.2 30

‘Percent plating efficiency for n samples from an untreated culture. SD. standard deviation.

DAYS, POST PLATING

Fig. 1. Evaluation of the average number of countable colonies/plate (three plates counted/point). Artek colony counter counts were begun on day 4 and performed daily to day 12 of the incubation period. Cells plated in BBL agar are indicated as open circles. Cells plated in the two different lots of Noble agar are indicated as open squares or triangles. Untreated cells were plated at a density of 200/ plate.

In evaluating the differences between the two agars, a number of experiments were conducted comparing the two agars in concurrent tests. Visual observation of colonies plated in BBL agar indicated that colonies seemed to appear at shorter incubation times compared to cells plated in Noble agar. To adequately evaluate these initial observations, untreated cells were plated either in BBL or Noble agar supple- mented medium at a density of 200 cells/plate (as is normally done for viable count plates). Colonies plated in BBL agar were visible to the eye by day 4 post plating as compared to day 5 post plating for Noble agar. By day 6 (Fig. l), colonies plated in both agars could be counted by the Artek colony counter. The number of colonies countable was significantly higher on day 6 for those plated in BBL agar than those

Effect of Agar on the Mouse Lymphoma Assay 733

plated in Noble agar. The colonies plated in BBL agar or in one of the two Noble agar lots all reached countable size by day 8, whereas those plated in one of the Noble lots required 11 days. This particular experiment was performed three separate times before an acceptable Noble agar plating efficiency was obtained. We felt that it was critical to the interpretation of the experiment that the cells plated in Noble agar eventually attain approximately the same plating efficiency as those cloned in BBL agar.

To evaluate the effect that agar might have on the plating efficiency and mutant frequency of treated and nontreated cells, three experiments were conducted. In these experiments a single MMS-treated (4 hr, 15 pglml) or untreated culture was divided and multiple samples cloned in each of the two to four different agar lots. Figure 2 (left panels) depicts the percent plating efficiency, number of mutants, and sponta- neous mutant frequency for nontreated cells. Experiments 2 and 3 included six samples from a single culture of untreated cells cloned in each of the agar lots evaluated, and experiment 1 included four samples. Cells plated in BBL agar gave a plating efficiency greater than 85% in all three experiments compared to Noble agar

200 m E

f 1501:; m7

8 100

c =O

0

EXPE~MENT

! 40 d 20 E

0

EXPERIMENT -

Fig. 2 . Evaluation of the effect of BBL (lot C4DNM.l) and Noble agar (lots 706265, 722950, 722471) on (A) the plating efficiency, (B) numbers of mutants (expressed as the total number for three plates), and (C) quantitated spontaneous mutant frequency of untreated TK+’- 3.7.2C cells (left panels) and quantitated total mutant frequency of cells treated with 15 pglml of MMS (right panels). Each data point is expressed as the mean *1 SD of four separate determinations for experiment 1 and six separate determinations for experiments 2 and 3. For each of the three separate experiments, a single culture was divided and cloned in the various agar lots.

734 Meyeret al

which ranged from 39% to 78%. In all three experiments, a larger absolute number of mutants was quantitated in BBL agar than in Noble agar (Fig. 2B). This was the case with both untreated and with MMS-treated cells. The spontaneous mutant frequency was not significantly different in the BBL agar or three lots of Noble agar for experiments 1 and 3. In experiment 2, the mutant frequency quantitated in BBL agar was significantly higher than that quantitated in the Noble agar. In these three experiments, there appeared to be somewhat more variability in the mutant frequency quantitated when cells were cloned in either of the three lots of Noble agar than when cells were cloned in BBL agar.

Figure 2 (right panels) shows the percent plating efficiency and mutant fre- quency for the MMS-treated cells. The MMS-treated cultures were also divided and cloned in BBL or one of three lots of Noble agar. The plating efficiency was observed to be significantly higher in two of three experiments when cells were plated in BBL agar (Fig. 2A, right panel, experiments 1 and 3) rather than in Noble agar. More interestingly, the quantitated mutant frequency was significantly higher for all three experiments when cells were cloned in BBL rather than in Noble agar (Fig. 3C, right panel).

To further investigate the apparent difference in the quantitated mutant fre- quency, cells treated with C.I. Solvent Yellow No. 33 were cloned in BBL or Noble agar. For these three experiments, single-treated cultures were divided just before cloning and were cloned either in BBL or Noble agar. In two of the experiments cells were treated with Aroclor-induced rat liver S-9, and in the third experiment cells were treated without exogenous activation. In one experiment we found BBL agar to give a significantly higher plating efficiency for each dose point evaluated compared to Noble agar (Fig. 3A, left panel). In the other two experiments the differences were small (Fig. 3A, middle and right panels), with BBL agar giving perhaps a slightly higher plating efficiency. The same lot of BBL agar was used in all three experiments. The same lot of Noble agar was used in the experiments shown in Figure 3 (left and middle panels), while a second lot was used for the experiment shown in Figure 3 (right panel). The absolute number of mutants per plate was always higher in BBL agar than in Noble agar. The mutant frequency quantitated in all three experiments was higher for those samples cloned in the BBL agar instead of the Noble agar (Fig. 3B).

In evaluating the significance of this higher number of mutants per plate and thus the higher quantitated mutant frequency, the relative proportion of u and X colonies for each cloning condition was considered. Histograms were generated from representative samples of TFT-selected colonies and viable count colonies. For untreated cells from the three experiments shown in Figure 2, the most significant difference in colony-size distribution appears to occur on the TFT-selected plates (representative results seen in Fig. 4). TFT' mutants (Fig. 4, right panels) grew to form larger colonies in BBL than in Noble agar. The magnitude of the large colony peak on the viable count plates (Fig. 4, left panels) is indicative of the larger number of colonies on the plates, and thus the higher observed plating efficiency (experiment 1, Fig. 2A, left panel). For MMS-treated cells (Fig. 5 , left panels), the histograms for viable count colonies indicate a higher plating efficiency in BBL agar (as well as a slightly larger size for the larger colonies). The histograms for the TFT-selected colonies (Fig. 5 , right panels) reflect not only the higher quantitated mutant frequency in BBL agar but also a significantly higher number of small colonies (small colony

I 01

. .

0'

0 5

10 1

2 14

16

I8

20

22 2

4 DO

SE

(ug

h11

x - > 0

z

W 3

0

W

LL a

I- z

3

L 2

250

200

150

100 50 "0

2.5

5

10

12

14

16

18 2

0 22 2

4 DO

SE

(ug/

ml)

>

0 z

W

W

Y a a

10

20

30

40

50

DOSE

(u

gh

11

Fig.

3. E

valu

atio

n of

the

eff

ect o

f usi

ng BBL (l

ot C

4DN

M.J)

or

Nob

le a

gar (

lot 7

0626

5, le

ft an

d m

iddl

e pa

nels

; lot

632

345,

rig

ht p

anel

) on

(A)

the

plat

ing

effic

ienc

y an

d (B

) qua

ntita

ted

mut

ant f

requ

ency

for

3.7.

2C c

ells

trea

ted

with

C.I.

Sol

vent

Yel

low

No. 33

in th

e pr

esen

ce of

aro

clor

-indu

ced r

at li

ver

=+

/-

S-9

(left

and

mid

dle

pane

ls) o

r with

out e

xoge

nous

act

ivat

ion

(rig

ht p

anel

). Si

ngle

trea

ted

cultu

res w

ere

divi

ded

and

clon

ed e

ither

in BBL o

r Nob

le a

gar s

uppl

emen

ted

clon

ing

med

ium

.

736 Meyeretal

6 - >

w 3

C LL

I- 2

< I-

Y P

s Q

l 0 K h - m L 0

40

20

BBL

SUE SElTlNO SIZE SETTING

Fig. 4. Evaluation of the effect of BBL or Noble agar on the colony size distribution for both viable count plates (left panels) and TFT selected plates (right panels) of untreated cells. The colony size increases with increasing size settings, ie, small colonies are on the left and large colonies are on the right. Plates were incubated for 10 days at 37”, 5% C@.

peak falls to the left of size setting 250) quantitated for MMS-treated cells plated in BBL rather than in Noble agar. Histograms were also obtained from C.I. Solvent Yellow No. 33 treated cells. More u TFT‘ mutant colonies were quantitated in BBL agar than in Noble agar (Fig. 6). As with the MMS-treated cells, the greatest difference was seen in the larger number of u colonies quantitated in BBL agar when compared to Noble agar.

To determine if the higher number of mutants observed in BBL agar was due to increased mutant recovery rather than “sneak through” of nonmutant cells, we isolated 97 mutants from treated cultures (53 u colonies and 44 X colonies) and 69 mutants from untreated cultures (45 u colonies and 24 X colonies) and tested them for TFT resistance. These TFT-resistance tests were performed after at least 6 days (generally 7 days or more) of growth in nonselective medium. This represents a minimum of ten generations post isolation. The normal doubling time for the cell line is about 9 hr. The newly isolated u mutants generally grow more slowly [Moore et al, 1985bl. All but one (a X colony from an untreated culture) were found to be TFT‘ (Table IV). It appears, therefore, that in the BBL agar more u mutants can grow to a colony size detected by the automatic counter.

Because the separate quantitation of u and X mutants has been shown to be important to the evaluation of data obtained from the mouse lymphoma assay [Moore et al, 1985a,b], and because the magnitude of the quantitated mutant frequency is used to define “potency” of any potential mutagen, it is critical to use optimal growth conditions during the clonal phase of the assay. Throughout the course of these

Effect of Agar on the Mouse Lymphoma Assay 737

vc NOBLE

SIZE SETTINQ

250 TFT

NOBLE 2oo

mms

loot -m 0 100

SUE SETflNO I

Fig. 5. Evaluation of the effect of BBL or Noble agar on the colony size distribution. Viable count colonies are shown in the left panels, and TITr mutant colonies are shown in the right panels. For this experiment cells were treated for 4 hr with 15 pgglml MMS. Plates were incubated for 10 days at 37"C, 5 % cq.

experiments, we found that agar can have a profound effect on the plating efficiency, u mutant frequency, and total quantitated mutant frequency.

In our laboratory, BBL agar appears to be a more reliable agar for use in cloning L5178Y mouse lymphoma cells. This has been demonstrated with higher and less variable plating efficiency of both treated and untreated cells. We observed that cells plated in BBL agar appear at an earlier time during the incubation period compared to cells plated in Noble agar. The mutant frequency quantitated for BBL agar indicated an increased recovery of both spontaneous and induced TFT' mutants, which upon random isolation were found to be TFT'. Evaluation of histograms indicated a higher quantitated mutant frequency when treated cells are cloned in BBL instead of Noble agar.

It should be stressed that the results of these experiments are based on the work of one laboratory and that different sources of medium, serum, or water may give different results. Some laboratories using other assay systems have improved plating efficiency with washing the Noble agar (Casto, unpublished data; de Serres, personal communication). We were, however, unable to obtain higher plating efficiencies by washing the Noble agar with either water, acetone, or 95% ethanol (data not shown).

ACKNOWLEDGMENTS

the careful typing and editing of the manuscript. We thank Patrick Strong for expert technical assistance and Shirley Milton for

738 Meyeret al

150-

125

100

75

c 50- 0 = 25 -

.-I I NOBLE YELLOW DYE - 24 &nl

- -

12S

im[

n

Y

f 125- I-

1 100-

a 75

YELLOW DYE 10 */mi

YELLOW DYE 24 &nl

BBL

-

a YELLOW DYE

I im a

0 im

SIZE SETTING SIZE SETTING

Fig. 6 . Evaluation of the effect of BBL or Noble agar on TFT' colony size distribution. TFT' colonies were evaluated for size following treatment with 10 p g l m l (left panels) or 24 pglml (right panels) C.I. Solvent Yellow No. 33 (with aroclor-induced rat liver S-9). The u mutants are seen to the left of each panel; A mutants are to the right. Single treated cultures were divided and cloned either in Noble agar supplemented cloning medium (top panels) or BBL agar supplemented cloning medium (bottom panels). Plates were incubated for 10 days at 37°C. 5% COz.

TABLE IV. Analysis of TFTp

Percent total growthb

Colony - Mutagen size No. X f 1 SD

MMS U 33 98.1 9.7 A 9 98.3 8.4

B W U 9 92.5 10.7 A 35 97.5 8.6

(2.1. Solvent U I 1 99.7 18.4

Untreated U 45 97.7 12.4 A 23 97.9 9.2

Yellow Dye No. 33

1 57.1 15.3'

aColonies selected with 1 pg/ml TFT were isolated, grown, and retested for TFT' as described in Materials and Methods. bPercent total growth of clonal Line with 1 pglml TFT as compared to growth without TFT. SD, standard deviation. 'This clonal line was not TFT'.

Effect of Agar on the Mouse Lymphoma Assay 739

REFERENCES

Aaron CA, van k l a n d AA, Mohn GR, Natarajan AT, Knapp AGAR, Tates AD, Glickman BW (1980): Molecular dosimetry of the chemical mutagen ethyl methanesulfonate. Quantitative comparison of mutation induction in Escherichia coli, V79 Chinese hamster cells, and L5178Y mouse lymphoma cells, and some cytological results in vitro and in vivo. Mutat Res 69:201-216.

Amacher DE, Paillet S, Ray V (1979): Point mutation at the thymidme kinase locus in L5178Y mouse lymphoma cells, I. Application to genetic toxicology testing. Mutat Res 64:391-406.

Amacher DE, Paillet SC, Turner GN, Ray VA, Salsburg DS (1980): Point mutations at the thymidine kinase locus in L5178Y mouse lymphoma cells, II. Test validation and interpretation. Mutat Res 72M7-474.

Amacher DE, Paillet SC (1980): Induction of trifluorothymidine-resistant mutants by metal ions in W178Y TK+’- cells. Mutat Res 78:279-288.

Amacher DE, Paillet SC (1981): Trifluorothymidine resistance and colony size in L5178YITK”- cells treated with methyl methanesulfonate. J Cell Phys 106349-360.

Clive D (1973): Recent developments with the L5178Y TK heterozygote mutagen assay system. Environ Health Perspect 6:119-126.

Clive D, Batson AG, Turner NT (1980): The ability of the L5178Y/TK+’- mouse lymphoma cells to detect single gene and viable chromosomal mutation: Evaluation and relevance to mutagen and carcinogen screening. In Williams GM, Kroes R, Woaijers HW, van de Pall KW (eds): “The Predictive Value of Short-Tern Screening Tests in Carcinogenicity Evaluation.” Amsterdam: Elsevier, pp 103-123.

Clive D, Flamm WG, Machesko MR, Bernheim NJ (1972): A mutational assay system using the thymidine kinase locus in mouse lymphoma cells. Mutat Res 16:77-87.

Clive D, Johnson KO, Spector JFS, Batson AG, Brown MMM (1979): Validation and characterization of the L5178YITK+’- mouse lymphoma mutagen assay system. Mutat Res 59:61-108.

Clive D, Moore-Brown MM (1979): The L5178YITK”- mutagen assay system: Mutant analysis. In Hsie AW, O’Neill JP, McElheny V (eds): “Banbury Report 2: Mammalian Cell Mutagenesis: The Maturation of Test Systems.” New York: Cold Spring Harbor Laboratory, pp 421-430.

Clive D, Spector JFS (1975): Laboratory procedure for assessing specific locus mutations at the TK locus in cultured L5178Y mouse lymphoma cells. Mutat Res 31:17-29.

Clive D, Voytek P (1977): Evidence for chemically-induced structural gene mutations at the thymidine kinase locus in cultured L5178Y mouse lymphoma cells. Mutat Res 44:269-278.

Gold A, Nesnow S, Moore M, Garland H, Curtis G, Howard B, Graham D (1980): Mutagenesis and morphological transformation of mammalian cells by a non-bay-region polycyclic cyclo- penta(c,d)-pyrene and its 3,4-0xide. Cancer Res 40:4482484.

Hozier J, Sawyer J, Clive D, Moore M (1982): Cytogenetic distinction between the TK+ and TK- chromosomes in the L5178Y TK+’- 3.7.2C mouse-lymphoma cell line. Mutat Res 105:451-456.

Hozier J, Sawyer J, Clive D, Moore MM (1985): Chromosome 11 aberrations in small colony L5178Y TK-’- mutants early in their clonal history. Mutat Res 147:237-242.

Hozier J, Sawyer J, Moore M, Howard B, Clive D (1981): Cytogenetic analysis of the L5178Y/TK+’- -+ TK-’- mouse lymphoma mutagenesis assay system. Mutat Res 84: 169-181.

Jacobson E, Krell K (1979): Ultraviolet wavelength regions implicated in toxic and mutagenic effects of broad spectrum radiation from fluorescent lamps on L5178Y mouse lymphoma cells. Mutat Res

Jacobson ED, KreU K, Dempsey MJ (1981): The wavelength dependence of ultraviolet light-induced cell killing and mutagenesis in L5178Y mouse lymphoma cells. Photcchem Photobiol 33:257- 260.

Jotz MM, Mitchell AD (1981): Effects of 20 coded chemicals on the forward mutation frequency at the thymidine kinase locus in W178Y mouse lymphoma cells. In de Serres FJ, Ashby J ( 4 s ) : ‘‘Evaluation of Short-Tern Tests for Carcinogens.” New York Elsevier, pp 580-593.

Krell K, Jacobson ED (1980): Sunlight-induced mutagenesis and toxicity in L5178Y mouse cells: Detefinination and comparison with other light sources. Environ Mutagen 2:389-394.

MacGregor JT, Gould DH, Mitchell AD, Sterling GP (1979): Mutagenicity test of diflubenzuron in the micronucleus test in mice, the L5178Y mouse lymphoma forward mutation assay, and the Ames Salmonella reverse mutation test. Mutat Res 66:45-53.

Meltz ML, MacGre or JT (1981): Activity of the plant flavonol quercetin in the mouse lymphoma L5178Y/TK+’- mutation, DNA single-strand break, and BALB-C-373 chemical transformation assays. Mutat Res 88:317-324.

62:533-538.

740 Meyer e t a l

Moore M, Clive D, Howard BE, Batson AG. Turner NT (1985a): In situ analysis of trifluorthymidine- resistant (TFT‘) mutants of L5178YITK”- mouse lymphoma cells. Mutat Res 151: 141-159.

Moore MM, Clive D, Hozier JC, Howard BE, Batson AG, Turner NT, Sawyer J (1985b): Analysis of trifluorothymidine-resistant (TFT‘) mutants of L5178Y/TK+’- mouse lymphoma cells. Mutat Res 151:161-174.

Moore-Brown MM, Clive D (1979): The L5178Y/TK+’- mutagen assay system: In situ results. In Hsie AW, O’Neill JP, McElheny V (eds): “Banbury Report 2: Mammalian Cell Mutagenesis: The Maturation of Test Systems.” New York: Cold Spring Harbor Laboratory, pp 71-88.