osmotic stress variants in chinese hamster cells

Somatic Ceil and Molecular Genetics, VoL 19, No. 5, 1993, pp. 439-447

Osmotic Stress Variants in Chinese Hamster Cells

Morgan Harris

Department of Molecular and Cell Biology, Division of Cell and Developmental Biology, University of Califbmia, Berkeley, California 94720

Received 6 July 1993

Abstract--Stable variants resistant to h~pertonic stress have been obtained from V79 cells by one-step selection in media supplemented with graded concentrations of NaCL Such variants retain a potential for resistance when isolated and propagated in isotonic media. On replating in graded NaCl, a family of dose-response curves is obtained, rising in level of resistance according to the degree of hypertonicity used to isolate the variants initially. In hybrids between variants and sensitive cells, phenotypic expression of resistance to hypertonic NaCI is recessive. Stable variants can also be isolated by one-step selection in media made hypertonic with o-mannitot. Clonal sublines selected with mannitol, as well as those obtained with NaCI, are res&tant to both types of hypertonic media. Fluctuation tests in media supplemented with NaCI show that resistance arises spontaneously and at random, with measured rates of variation that depend on the concent1~tion of NaCI used for selection. Treatment of sensitive cells with 5-azacytidine increases the frequency of resistant variants in assays with high levels of added NacI but is less effective when selection is performed at lower concentrations. Exposure to ethyl methane sulfonate has little or no effect on variant frequency.

INTRODUCTION

Osmotic stress can be readily imposed on cells in culture and offers a model for the study of adaptation at the population level. The effects of hypertonic media on animal cells for short periods have often been reported, and in more extended exposure, adaptation of mouse cells to hypertonic NaCI (1) and to hypertonic sorbitol (2) have been described. Normal sensitivity was regained, however, when these cells were returned to isotonic media. In more recent work, Hen- nessy and Rubin (3) showed that stable variants resistant to osmotic stress could be obtained from transformed BALB/3T3 mouse cells by prolonged subculture in media supplemented with 100 mM NaC1. The potential for resistance was retained

after several hundred generations in normal media. Stable variants in hypertonic NaCI have also been reported for populations of rabbit kidney cells (4). This is not surprising, since cells from the renal medulla of several species are known to possess specialized mechanisms for survival in a hypertonic milieu, e.g., accumulation of high concentrations of compatibIe osmolytes such as beta- ine, myoinositol, and sorbitol (5--7), as well as inducible activity of Na+,K+-ATPase (8). There is no evidence that these mechanisms operate in nonrenal cells, and the basis for resistance in other cell types is yet to be determined.

In the present experiments we have obtained information previously lacking on the origin and properties of stable variants

439

0740-7750/93/0900-0439507.00/0 © 1993 Plenum Publishing Corporation

440 Harris

isolated under hypertonic stress, using V79 Chinese hamster cells and media supplemented with NaC1 or D-mannitol as the added osmolyte. Topics to be considered include stability and clonal differences between individual variants, the spectrum of resistance to added NaC1 exhibited by variants isolated at rising levels of hypertonicity, phenotypic expression of resistance in hybrids with sensitive cells, rates of variation to resistance as measured by fluctuation tests, and assays for induction of resistance by treatment with ethyl methane sulfonate and 5-azacytidine.

MATERIALS AND METHODS

Culture Procedures. The experiments to be described were performed with V79 Chinese hamster cells, for the most part with subline 204-7. The latter lacks thymidine kinase and was derived by clonal isolation after long-term culture in BrdU (9). For construction of hybrids between resistant and sensitive cells we combined 204-7 with V79 subline 129-5, which lacks HPRT (10). Stock cultures were maintained in a basal medium containing 10% fetal bovine serum (Flow Laboratories) and 90% Dulbecco's modification of Eagle's medium with 4.5 mg/ml glucose. This nutrient (10FCSDB) was supplemented before use with sodium penicillin (60 ixg/ml), streptomycin (50 p.g/ ml), and L-glutamine (100 txg/ml). Hyper- tonic media were prepared by incorporating appropriate volumes of stock 500 mM NaC1 or D-mannitol, made up in Dulbecco's nutrient, into the Dulbecco fraction of 10FCSDB. HAT medium was prepared by adding a lyophilized concentrate (Sigma) to 10FCSDB. Plating experiments were performed in 60-ram Petri dishes maintained at 37°C in a humidified CO2 incubator. When colonies were well formed, experiments were terminated by staining the cultures for 30 rain in a saturated solution of crystal violet in

0.85% NaCI, after which the dishes were rinsed in tap water and air-dried.

Fluctuation Tests. Each assay was initiated from a four-day colony of approximately 200 cells. This primary clone was dissociated and replated in 10FCSDB. After seven days of incubation, cells from each of 15 colonies were isolated and transferred to flasks to be grown in 10FCSDB as mass populations. Aliquots from 12 sublines were finally assayed in 10FCSDB supplemented with 50 mM or 100 mM NaC1. After 21 days of incubation without fluid change, the cultures were terminated and stained as described above.

Induction of Variants. In experiments with ethyl methane sulfonate (EMS), mass populations of 204-7 cells were treated for 24 h in 10FCSDB containing 300 ixg/ml EMS. After recovery and subculture in normal medium for five days, EMS-treated and control cells were plated in graded numbers for assay in media containing 50 mM or 100 mM added NaCI. In other experiments, 204-7 cells were treated for 24 h in 10FCSDB containing freshly prepared 5-azacytidine (5-aza-CR) at 1.0 ~g/ml. After recovery in normal medium for two days, treated and untreated cells were plated in graded numbers in 10FCSDB containing 50 mM or 100 mM added NaC1. Following incubation for 21 days, the cultures were terminated and stained.

RESULTS

Isolation of Variants in Hypertonic NaCl. In an early study, 204-7 cells were propagated serially in 10FCSDB supplemented with 80 mM NaC1. Growth at first was slow but gradually increased, and clone 2715-2 was eventually isolated in this medium. Preliminary tests showed that plating efficiency of the 2715-2 cells was approximately the same in medium supplemented with 80 mM NaC1 as in 10FCSDB alone. By contrast,

Osmotic Stress Variants 441

the plating efficiency of stock 204-7 cells in medium with 80 mM added NaC1 was less than 0.1%. Experiments were performed to test the stability of resistance in 2715-2 cells. The latter were returned to unsupplemented 10FCSDB and maintained in this medium for five passages. At each subculture, platings were made in 10FCSDB and in media supplemented with 80 mM and 100 mM NaC1. Table 1 presents the data obtained. The plating efficiency of 2715-2 cells in media with 80 mM NaCI and in 10FCSDB was approximately the same at each transfer. Although the plating efficiency in media with 100 mM NaC1 was lower, there was no systematic decline over the test period. These results show that resistance to hypertonic NaC1 was stably maintained by 2715-2 cells in the absence of selection.

One-Step Assays for Osmotic Resistance. Studies were performed to see whether stable variants resistant to hypertonic NaCI could be isolated in one-step plating experiments. A practical problem at high cell numbers is that population density strongly enhances the survival of sensitive ceils in test plates. For example, when stock 204-7 cells

Table 1. Stability of Resistance to Hypertonic NaCI during Subculture of 2715-2 Cells in Isotonic Media a

Plate counts

80 mM 100 mM Population added added

Passage doublings 10FCSDB NaC1 NaC1

0 0 66.0 60.7 60.0 1 5 69.0 57.0 24.0 2 12 71.3 61.0 42.0 3 17 65.5 47.3 18.3 4 24 66.7 56.3 30.0 5 31 68.3 62.0 36.3

~Cells of subtine 2715-2, isolated as a clone in 10FCSDB with 80 mM added NaCI, were transferred to isotonic 10FCSDB for five passages. At each subculture, groups of three Petri dishes each were set up at 100 cells in 10FCSDB, and in media supplemented with 80 mM NaCI and 100 mM NaCI. Figures shown represent average colony counts per 100 cells plated.

were plated directly into media with 100 mM NaC1, the limiting number that could be used for resistance studies in 60-mm dishes was 8 x 10 s. At higher cell numbers, prominent central masses and background foci prolifer- ated, but if these cells were removed and tested at lower densities, sensitMty to NaCI was found to be unaltered. The limitations imposed by population density thus require a pilot test at each level of hypertonicity to determine the maximum ceil density that can be used for isolation of variants with true resistance.

Guided by these restrictions, we obtained stably resistant colonies by one-step plating in media containing several concentrations of added NaC1. A number of these variants were transferred to 10FCSDB and, when replated in the original media contin- ued to show resistance. Experiments were carried out to compare resistance in sublines isolated from media containing 50 mM, 80 raM, and 100 mM NaC1. Variants from these media were isolated and propagated for several passages in unsupplemented 10FC- SDB. Each variant was then replated in 10FC- SDB with graded concentrations of NaCI. Figure I illustrates the results of these assays. A series of dose-response curves is obtained in which each variant shows increased resistance at all concentrations of NaC1. Differ- ences between curves for individual variants correspond to the varying levels of hypertonicity in media from which the variants were initially derived. Resistance to osmotic stress thus is graduated in accordance with the selective screen used for variant isolation.

Another dimension of variability is seen when clones isolated in the same hypertonic medium are compared individually. Figure 2 shows dose-response curves for 204-7 cells and three derivative clones originating in 10FCSDB with 50 mM added NaC1. The level of resistance between the three clones varies significantly, and these differentials are reproducible, as a second plating of the

442 Harris

I00'

10 .0

b" .£ .o

.~ 1.0

rw

. . . . . ~ . \ . .

0"1 i

0 . 0 1 0 ~ ~ _ ~ 8~0 ' ~ _ J _ ° J 4 0 1 2 0 1 6 0 2 0 0

NaC} added, mM

Fig. 1. Resistance to osmotic stress in 204-7 cells and derivatives isolated from media with graded supple- ments of NaCI. Stock 204-7 cells, O; clone 8 isolated from 10FCSDB with 50 mM added NaC1, [~; clone 2715-2 isolated from 10FCSDB with 80 mM added NaC1, A; clone 17 isolated from 10FCSDB with 100 mM added NaCI, @. All points shown represent average plate counts for three to six Petri dish cultures.

clonal populations indicates. A similar diver- gence in resistance was observed for clones isolated in media supplemented with 100 mM NaC1 (data not shown). It appears that stock populations represent a mosaic of cells with differing potentials for resistance to hypertonic NaC1. Interclonal variability of this type has received little attention, but it is similar to the population mosaicism we have described for BrdU resistance in V79 cells (11) and variability in response to 5-aza-CR in CHO cells (12).

Characterization of Osmotic Resistance. To see whether resistance to osmotic stress was specific for NaCI as an osmolyte, additional experiments were performed in media made hypertonic with D-mannitol. Cultures of 204-7 cells were propagated initially in 10FCSDB supplemented with 150 mM mannitol. The growth rate soon rose to normal, and the population could then be

100

10.0

-~ 1.0

rr"

o . I L -

i ..................... i f 0.01 ~// 50 60 70

NaCI added, mM

Fig. 2. Interclonal differences in resistance to hypertonic NaC1. Stock 204-7 cells, Q; clone 1, A; clone 2, KI; and clone 3, V. Clones 1-3 were isolated as single colonies from 204-7 cells in 10FCSDB supplemented with 50 mM NaC1. Solid lines show the initial assays, and dotted lines, the repeat platings. All points shown represent average plate counts for three to six Petri dish cultures.

maintained indefinitely in media with 200 mM added mannitol. Resistant variants could also be obtained by one-step plating of 204-7 cells in graded mannitol, with log- linear survival curves, as with NaC1. Studies on cross-resistance were carried out, using clone 2500-2, isolated from media containing 200 mM added mannitoI, and line 2715-2, which originated from medium supplemented with 80 mM NaC1. For assay, both lines were plated in graded concentrations of NaC1 and in graded concentrations of manni- toI, along with stock 204-7 cells. Figure 3 shows the results of these experiments. Each variant is resistant to both NaC1 and D- mannitol, although neither line had been previously exposed to both types of hypertonic media. Resistance therefore is corre- lated with hypertonicity as such rather than the specifics of osmolyte permeability, since ionic transport mechanisms control the uptake of NaCI, while the uptake of mannitol is

O s m o t i c S t r e s s V a r i a n t s 443

I00,

~' l o E

2

0.1

O.O1

ill ~OO 200

i i

5 0 0 4 0 0

Added NaCt or D-mannitol, rnM

Fig. 3. Tests for cross-resistance between 204-7 sublines isolated in hypertonic NaCI and hypertonic D-mannitol. Stock 204-7 cells, O, 0; clone 2715-2 isolated originally in 10FCSDB with 80 mM added NaC1, G, A; clone 2500-2, isolated from 10FCSDB supplemented with 200 mM mannitol, D, II. Open symbols show platings in media with graded NaCI, and closed symbols, platings in media supplemented with mannitol. All points shown represent average plate counts for three to six Petri dish cultures.

governed by the very different process of pinocytosis (13).

In other experiments hybrids were con- structed between cells sensitive and resistant to NaC1 to examine the phenotypic expression of resistance to hypertonic media. Using a previously described procedure (14), we hybridized cells from sensitive line 129-5, deficient in HPRT, with resistant 2715-2 cells, which lack thymidine kinase. A series of hybrids were isolated from H A T medium as single colonies. To test for resistance, cells from the parent lines, and 10 clonal hybrids were plated in 10FCSDB alone, and in 10FCSDB supplemented with 100 mM NaC1. Table 2 shows the results of this study. All 10 hybrids were sensitive, with plate counts in selective medium that were similar to or less than those of the sensitive 129-5 parent cells.

Table 2. Expression of Resistance to Hypertonic NaC1 in 129-5 (NaCl-Sensitive) and 2715-2 (NaCI-Resistant)

Cells and in Hybrids 1-10 between Parent Lines ~

Plate counts

100 mM Cell l i n e 10FCSDB added NaCI

129-5 50.0 0.13 2715-2 64.7 12.4 Clone 1 38.0 < 0.03

2 47.7 < 0.01 3 70.0 0.20 4 60.7 0.19 5 45.0 0.08 6 38.0 0.18 7 62.3 0.03 8 34.6 0.01 9 46.7 0.01

10 59.7 0.01

aFor each cell line, groups of three Petri dish cultures were set up at 100 cells in 10FCSDB, and at 100, 1000, and 10 4 cells in 10FCSDB with 100 mM added NaC1. All counts shown represent averages per 100 cells plated.

These data indicate that resistance to hypertonic NaC1 is recessive in 2715-2 cells.

Fluctuation Tests in Hypertonic NaCl. We used fluctuation analysis (15) to determine whether resistance to hypertonic NaC1 originates from a population shift or by the selection of individual variants. To minimize background variation, each fluctuation test was performed with a series of clonal sublines, all derived initially from a common small colony of stock 204-7 cells. Assays were carried out by plating cells from each subline into 10FCSDB supplemented by 50 or 100 mM NaCI. After 21 days of incubation without fluid change, the cultures were terminated and stained for colony counts.

Table 3 summarizes the results of this work. The calculated chi-square values for resistance between subtines are significant, and the variance clearly exceeds the mean values in all experiments. As in the classic experiments of Luria and Delbrfick (15), this implies that the potential for resistance to osmotic stress is preexisting and independent from the later process of selection. The data

444 Harris

Table 3. Fluctuation Tests for Variant Formation by 204-7 Cells under One-Step Selection in Media Supplemented with 100 mM and 50 mM NaCI a

Selection with Selection with 100 mM NaC1 50 mM NaCI

Experiment 1 2 3 4

No. of sublines 12 12 12 12 Initial cell No. 1 1 1 1 Mean final cell

No. x 106 31.2 16.0 28.8 15.4 Total cells tested

per subline x 103 480 810 12.9 9.9 Mean No. resistant

colonies in sublines 4.1 x 10 -5 7.2 × 10 .5 7.0 × 10 .3 11.7 X 10 -3 Variance 23 94 51 57 X 2 62 144 84 53 P < 0.001 < 0.001 < 0.001 < 0,001 Variation rate per

cell per generation Lea and Coulson (16) 2.4 x 10 .6 6.1 x 10 -6 5.2 x 10 4 8.2 x 10 4 Capizzi and Jameson (17) 5.4 x 10 .6 9.5 x 10 -6 5.8 x 10 4 9.7 × 10 -4

~Fluctuation tests were performed as described previously (11). Each experiment was initiated from a single colony of approximately 200 cells, which was dissociated and replated in 10FCSDB. After seven days, single colonies were isolated and grown as sublines to mass populations. Assays were performed by plating graded numbers of cells into media with 100 mM or 50 mM added NaC1.

can accordingly be used to ca lcu la te ra tes of va r i a t ion to res i s tance us ing the m e t h o d s of L e a and Coulson (16) and Capizz i and J a m e s o n (17). T h e va lues o b t a i n e d d e p e n d on the level o f NaC1 a d d e d for se lec t ion . F o r cu l tu res m a i n t a i n e d with 100 m M a d d e d NaC1, the ra te of va r i a t ion falls b e t w e e n 10 .5 and 10 .6 . W h e n se lec t ion is ca r r i ed out in m e d i a s u p p l e m e n t e d with 50 m M NaC1, the ra te is app rox ima te ly 100-fold higher , 10 .3 to 10 .4 . T h e s e resul t s a re in l ine wi th f r equen- cies obse rved for d i rec t p la t ings of s tock 204-7 cel ls in g r a d e d NaCI (Fig. 1) and suggest tha t only a few of the many var ian ts a p p e a r i n g spon t aneous ly in m e d i a with 50 m M a d d e d NaC1 have a high enough res i s tance to pass the se lec t ion sc reen in m e d i a s u p p l e m e n t e d wi th 100 m M NaC1.

Assays for Induction of Osmotic Resis- tance. Since E M S is a we l l -known mutagen , we t r e a t e d s tock 204-7 cel ls wi th this agent , fol lowing a s t a n d a r d p ro toco l known to p r o d u c e mu tan t s in conven t iona l assay sys- t ems (14). Tab le 4 shows the resul ts of

severa l expe r imen t s in which 204-7 cells were exposed ini t ial ly to 300 jxg/ml E M S in 1 0 F C S D B for 24 h. A f t e r a f ive-day recovery per iod , the cells were p l a t e d in m e d i a wi th 50 or 100 m M a d d e d NaC1, a long with u n t r e a t e d popula t ions . T h e r e was no c learcu t indica- t ion of induc t ion by E M S in e i the r of these assay systems. Smal l bu t cons is ten t d i f feren- t ials were observed , wi th a twofold average increase in va r ian t f r equency for cells ex- p o s e d to E M S as c o m p a r e d to u n t r e a t e d

popu la t ions . M o r e s t r ik ing effects were o b t a i n e d

when we t r e a t e d s tock 204-7 cells with 5 -aza-CR, an agent known to p r o d u c e s table changes in gene express ion (18, 19). In a first ser ies of exper iments , cells were exposed to 1.0 Ixg/ml 5 - a z a - C R in 10FCSD B for 24 h and, a f te r a two-day recovery pe r iod , were p l a t e d into m e d i a s u p p l e m e n t e d with 100 m M NaC1, a long with con t ro l popu la t ions . U n d e r these condi t ions , an induct ive effect is d e a r l y ev ident (Tab le 5). As c o m p a r e d with u n t r e a t e d cells, the average increase in


Table 4. Induction of Variants Resistant to Hypertonic NaC1 by Treatment of 204-7 Cells with EMS a

Selection Variant frequency Variant frequency Mean ratio Experiment medium untreated cells after EMS treated/untreated

1 100 mM NaCI 5.8 x 10 --5 6.8 x 10 -5 2 5.0 10.3 2.08 3 12.2 27.6 4 4.0 11.3

5 50 mM NaCI 1.72 x 10 5̀ 1.76 × 10 .5 6 0.75 2.25 2.06 7 1.38 2.66 8 0,44 1.00

aCells were treated with EMS (300 Ixg/ml in 10FCSDB for 24 h) and assayed after a five-day recovery period in 10FCSDB, with one subculture. Treated and untreated cells were plated out in graded cell numbers in 10FCSDB with 100 mM or 50 mM added NaC1 (12--24 Petri dish cultures per series). An additional group of six Petri dish cultures for each series was set up at 100 cells in 10FCSDB, to correct for differences in plating efficiency.

var iant f r equency for four exper iments with

5 -aza -CR was 85-fold, but w h e n the assay

m e d i u m was 10FCSB s u p p l e m e n t e d with 50

m M NaC1, a very different resul t was

ob ta ined . With less s t r ingent select ion, the

appa ren t induct ion with 5 -aza -CR was great ly

reduced , with the f requency of var iants for

t r ea t ed cells less than double that of un-

t r e a t ed popula t ions (Table 5). This does not

necessar i ly m e a n that the absolute n u m b e r of

var iants induced by 5 -aza -CR was less. T h e

large increase in spon taneous var iants w h e n

select ion is p e r f o r m e d with 50 m M added

NaC1 may serve mere ly to obscure the

induced variants that stand out m o r e clearly

when select ion is car r ied out with 100 m M

added NaC1.

D I S C U S S I O N

We have shown that s table var iants

res is tant to osmot ic stress can be ob ta ined

f rom Chinese hams te r ceils by one-s tep

plat ing in med ia m a d e hyper ton ic wi th added

NaCI or D-mannitol . T h e f requency of

resis tant colonies decl ines as the concen t ra -

t ion of added osmolyte is increased, but

var iants at all levels re ta in a po ten t ia l for

Table 5. Induction of Variants Resistant to Hypertonic NaC1 by Treatment of 204-7 Cells with 5-aza-CR ~

Selection Variant frequency Variant frequency Mean ratio Experiment medium untreated cells after 5-aza-CR treated/untreated

t 100mM NaC1 2.4 × 10 .5 113 x 10 -s 2 0.5 41 84.8 3 1.1 117 4 0.8 83

5 50 mM NaC1 0.51 x 10 -2 1.54 x 10 -s 6 2.62 5.55 1.81 7 1.20 1.53 8 2.30 1.92

"Cel~s were treated with 5-aza-CR (1.0 ~g/ml in 10FCSDB for 24 h) and assayed after a two-day recovery period in 10FCSDB. Treated and untreated cells were plated out in graded numbers in 10FCSDB supplemented with 100 mM or 50 mM NaCI (12-24 Petri dish cultures per series). An additional group of six Petri dish cultures for each series was set up at 100 cells in 10FCSDB, to correct for differences in plating efficiency.

446 Harris

resistance when isolated and grown in isotonic media. The magnitude of this potential is a function of the conditions used for initial selection. Thus, when isolated variants are replated in media of graded hypertonicity, a family of dose-response curves is seen, rising progressively in resistance according to osmolyte concentrations used for the original selection. Resistance to osmotic stress therefore seems to be incremental, increasing by a tandem or multistep series of changes. Models for multistep variation are well known from drug resistance studies with bacteria (20). In the so-called streptomycin pattern, clonal isolates from primary cultures give a spectrum of variants ranging from partial to complete resistance. This contrasts with penicillin-type resistance, in which the initial isolates show partial resistance only, but if these are reexposed sequentially to increasing concentrations of the drug, highly resistant forms finally emerge. Whether osmotic resistance is based similarly on unit repeats or sequential change remains to be s e e n .

At a more general level the question is whether resistance to hypertonic stress is mediated by genetic or epigenetic change. Fluctuation tests demonstrate that the shift is a random one but do not indicate what the mechanism may be. Genetic alterations in the DNA coding system clearly meet the criterion of random change, but epigenetic shifts in DNA methylation patterns do so as well and may result in stable alterations in gene expression that have been documented as the basis for a number of cell variations (21). There is little in the present study to support a genetic mechanism for resistance to hypertonic stress. The spontaneous rate of variation for cells in media supplemented with 100 mM NaCI falls within the range expected for mutation, but the rate determined in media supplemented with 50 mM NaC1 (100-fold higher) is questionable. Induc- tion of variations by EMS as a known

mutagen is often involved to support a hypothesis of genetic change, but recent studies show" that under appropriate circum- stances EMS may promote epigenetic shifts as well. In the experiments in question, EMS was found to induce the reexpression of asparagine synthetase in mutant CHO cells by causing hypomethylation at the 5' end of the AS gene (22). A global reduction of 10% in DNA methylation after EMS treatment was also noted. Considering the uncertainty in mode of action posed by these findings, the marginal increases in resistance to hypertonic stress that we have observed on exposure to EMS seem equivocal in significance.

A more convincing case can be made for epigenetic shifts as the basis for resistance to osmotic stress. We show that exposure to 5-aza-CR followed by stringent selection in media supplemented with 100 mM NaC1 consistently yields a large increase in variants over background frequencies observed with untreated cells. Since 5-aza-CR is a well- known DNA demethylating agent (18, 23), it is reasonable to assume that the transition to resistance under these conditions may be implemented by epigenetic change. The results obtained by 5-aza-CR treatment followed by selection in media with 50 mM added NaC1 are more difficult to interpret. Conceivably, high-level resistance to osmotic stress is directly dependent on DNA methylation changes, while such shifts are less important for resistance at lower levels. In this case, the inductive effects of 5-aza-CR may well be masked by the high frequency of spontaneous variants when selection is carried out in media of lower hypertonicity. Additional experiments will be required to resolve this question.

ACKNOWLEDGMENTS

This work was supported in part by grants from the Committee on Research, University of California, Berkeley.


L I T E R A T U R E C I T E D

1. Schachtschabel, D.O., and Foley, G.E. (1972). Exp. Cell Res. 70:317-324.

2. Clegg, J.S., Gallo, G., and Gordon, E. (1986). Exp. CeIl Res. 163:35-46.

3. Hennessy, T.L., and Rubin, H. (1986). Cancer Res. 46:6041-6048.

4. Uchida, S., Green, N., Coon, H., Triche, T., Mires, S., and Burg, M.B. (1987).Am Z Physiol. 253:C230- C242.

5. Maury, J. (1988).Arch. Microbiol. 149:232-239. 6. Nakanishi, T., Turner, R.J., and Burg, M.B. (1990).

Am J. Physiol. 258:F1061-F1067. 7. Nakanishi, T., Balaban, R.S., and Burg, M.B.

(1988). Am J. Physiol. 255:C181-C191. 8. Bowen, J.W. (1992).Am Z PhysioL 262:C845-C853. 9. Harris, M. (1982). Cell29:483-492.

10. Harris, M. (1979). Somat. Cell Genet. 5:793-808. 11. Harris, M., and Collier, K. (1980). Proc. Natl. Acad.

Sci. U.S.A. 77:4206-4210.

12. Harris, M. (1984). Somat. CellMol. Genet. 10:615- 624.

13. De Courcy, K., and Storrie, B. (1991). Exp. Cell Res. 192:52-60.

14. Harris, M. (1986). Somat. Cell MoL Genet. 12:459- 466.

15. Luria, S.E., and Delbr/ick, M. (1943). Genetics 28:491-511.

16. Lea, D.E., and Coulson, C.A. (1949). J. Genet. 49:264-285.

17. Capizzi, R.L., and Jameson, J.W. (1973). Mutat. Res. 17:47-148.

18. Taylor, S.M., and Jones, P.A. (1979). Cell 17:771- 779.

19. Riggs, A.D., and Jones, P.A. (1983). Adv. Cancer Res. 40:1-30.

20. Bryson, V., and Demerec, M. (1950). Adv. Genet. 7:1-46.

21. Holliday, R. (1990). PhiL Trans. R. Soc. London B. 326:320-338

22. Worton, K.S., Kerbel, R.S., and Andrulis, J.L. (1991). CancerRes. 51:985-989.

23. Jones, P.A., and Taylor, S.M. (1980). Cell20:85-93.

osmotic stress variants in chinese hamster cells

Documents