Relationship between somatic mutation and neoplastic transformation.

5 downloads 0 Views 1MB Size Report
ulation doublings after carcinogen treatment. Although this frequency of transformation is comparable to that of somatic mutation, the detection time required is ...
Reprinted from Proc. Nati. Acad. Sci. USA

Vol. 75, No. 7, pp. 3297-3301, July 1978 Cell Biology

Relationship between somatic mutation and neoplastic transformation (chemical carcinogenesis/Syrian hamster/anchorage independence/morphological transformation/neoplastic progression)

J. CARL BARRETT* AND PAUL 0. P. Ts'o Division of Biophysics, School of Hygiene and Public Health, Johns Hopkins University, Baltimore, Maryland 21205

Communicated by James Bonner, April 10,1978

ABSTRACT Somatic mutation and neoplastic transformation of diploid Syrian hamster embyro cells were examined concomitantly. Mutations induced by benzo[a pyrene and Nmethyl-N'-nitro-N-nitrosoguanidine were quantitated at the hypoxanthine phosphoribosyltransferase and Na+/K+ ATPase loci and compared to phenotypic transformations measured by changes in cellular morphology and colony- formation in agar. Both cellular transformations ad characteristics distinct from the somatic mutations observed at the two loci. Mohological transformation was observed after a time comparale to that of somatic mutation but at a frequency that was 25- to 540-fold higher. Transformants capable of colony formation in agar were detected at a frequency of 10-5-106, but not until 32-75 population doublings after carcinogen treatment. Although this frequency of transformation is comparable to that of somatic mutation, the detection time required is much longer than the optimal expression time of conventionally studied somatic mutations. Neoplastic transformation of hamster embryo cells has been described as a multistep, progressive process. Various phenotypic transformations of cells after carcinogen treatment may represent different stages in this progressive transformation. The results are discussed in this context and the role of mutagenesis in the transition between various stages is considered. Neoplastic transformation may be initiated by a mutational change, but it cannot be described completely by a single gene mutational event involving a dominant, codominant, or X-linked recessive locus. Neoplastic transformation induced by chemical carcinogens is more complex than a single gene mutational process. Thus, this comparative study does not give experimental support to predictions of the carcinogenic potential of chemicals based on a simple extrapolation of the results obtained from conventional somatic mutation assays.

Since proposed by Boveri in 1914 (1), somatic mutation as a basis for the heritable alteration in malignant cells has been a popular hypothesis (2, 3). This hypothesis provides, at least in part, the rationale for the use of mutagenesis tests for the detection of biohazardous chemicals. The relationship, however, between somatic mutation and neoplastic transformation is unclear. An examination of this relationship requires that each process be quantitated and that the mechanism of each process be defined. The process of somatic mutation can be studied reliably by examining various heritable phenotypic alterations of mammalian cells, particularly resistance to certain drugs. Additionally, the basis of somatic mutation can be defined at the molecular level in biochemical terms (4). Neoplastic transformation, in contrast, is less well understood, a fact partially attributable to the lack of a definitive phenotypic alteration characteristic of malignancy (5). Although tumor formation in vivo serves to define neoplastic transformation of cells in vitro, tumorigenicity is a multifaceted phenomenon which is difficult to analyze at the molecular or cellular level. Accordingly, several other in vitro phenotypic characteristics

which are associated with transformed cells have been studied extensively (5). Thus, somatic mutation and neoplastic transformation can be investigated by the same experimental approach-i.e., by studying the heritable alterations of cells in culture. The elucidation of the significance of these cellular changes to tumorigenicity is crucial to an understanding of neoplastic transformation (5). To date, however, somatic mutation and neoplastic transformation have not been studied quantitatively in the same cellular system, thus preventing direct comparisons of the two processes. Recently, we reported the development of a mammalian cellular system, utilizing early passage, diploid Syrian hamster embryo cells, that is amenable to concomitant studies of neoplastic transformation (6, 7) and somatic mutation (8). We described the parameters involved in the quantification of mutants of Na+/K+ ATPase and hypoxanthine phosphoribosyltransferase (HPRT) functions of these cells. Syrian hamster embryo cells also have been utilized for quantitative studies of in vitro transformation by chemical carcinogens (9, 10). In these studies, transformation has been measured by the frequency of cells which either yield morphologically transformed colonies (9, 10) or are capable of anchorage-independent growth (11). Morphological transformation is an early alteration of Syrian hamster cells after exposure to chemical carcinogens (7, 9, 10), whereas anchorage-independent growth correlates very well with the ability of the cells to produce tumors in vivo (12, 13). In this report, these two phenotypic alterations associated with neoplastic transformation are compared with known somatic mutations in terms of observed frequency and time for detection after carcinogen treatment. Both morphological transformation and anchorage-independent growth have features distinct from conventionally studied somatic mutations. While such mutations can be characterized by a single-step mutation process, neoplastic transformation cannot be described adequately in such terms. Neoplastic development in vivo (14, 15) and in vitro (5, 6) has been described as a multistep progressive process. Although such a multistep process might be initiated by a mutational change, it cannot be completely described by a single gene mutational event involving a dominant, codominant, or X-linked recessive locus, because secondary changes must occur. The relationship between mutagenesis and carcinogenesis of Syrian hamster embryo cells in culture is therefore discussed with reference to the progressive multistep nature of neoplastic transformation. MATERIALS AND METHODS Cells. Syrian hamster embryo cell cultures were established Abbreviations: HPRT, hypoxanthine phosphoribosyltransferase; MNNG, N-methyl-N'-nitro-N-nitrosoguanidine. * Present address: National Institutes of Health, National Institute of Environmental Health Sciences, P.O. Box 12233, Research Triangle Park, NC 27709.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C §1734 solely to indicate this fact. 3297

8298

Cell

Proc. Natl. Acad. Sci. USA 75 (1978)

Biology: Barrett and Ts'o

from 13-day gestation fetuses collected aseptically by Caesarian section from inbred Syrian hamsters, strain LSH/ssLAK (Lakeview Hamster Colony, Newfield, NJ), or outbred hamsters from Engle Laboratories (Farmersburg, IN). Pools of primary cultures from littermates were stored in liquid nitrogen. Secondary cultures were initiated from the frozen stocks, and all experiments were performed with tertiary or later cultures. All cultures were routinely tested by Microbiological Associates and found free of mycoplasma contamination. Media and Growth Conditions. The cells were grown in IBR modified Dulbecco's Eagle's reinforced medium (Biolabs, Northbrook, IL) supplemented with 0.22% NaHCO0 (wt/vol) and 10% Rehatuin filter-sterilized fetal bovine serum (Reheis Chemical Company, Kankakee, IL) without antimicrobial agents. Cells were transferred by gentle trypsinization with 0.1% trypsin solution (1:250, GIBCO, Grand Island, NY) for 5 min at 370. Mutation Assays. Details of these methods have been published (8). For the respreading assay, tertiary passage Syrian hamster embryo cells were inoculated at a density of 5 X 105 cells in a 75-cm2 flask. After 15 hr, these cells were treated with N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) at 1 MiM or 5 MM for 2 hr, or they were treated with benzo[ajpyrene at 1 Ag/ml or 10 Mug/ml for 24 hr. After the exposure time, the cultures treated with MNNG were washed twice with phosphate-buffered saline and the flasks treated with benzo[a]pyrene were washed five times with complete medium (containing 5% serum). The cultures were then grown to confluency, subcultured at a split ratio of 1:10, and again grown to confluency. Untreated cultures normally attained confluency in 7 days, whereas cultures treated with 5 MtM MNNG or 10 ,g/inl benzo[a]pyrene required a longer time period. At each passage, 105 cells were plated in each of 10 100-mm petri plates. After 15 hr, selective medium containing either 8-azaguanine at 40 ug/ml, 6-thioguanine at 2 ug/ml, or ouabain at 1.25 mM was added to the plates. After 1 10-20 days incubation, with changes of the selective medium every 3-4 days, the colonies either were scored after fixation and staining or were isolated. For the direct assay of mutation, tertiary passage Syrian hamster embryo cells (105) were plated overnight in 100-mm petri plates, then treated with MNNG, benzo[a]pyrene, or solvent only for the specified period of time and washed as described. The cells were resupplied with complete medium in the absence of selective agent and incubated for a recovery/expression period of 1-3 days. After this period, 8-azaguanine at 40 Mg/ml was added to the medium and clones were grown in this selective medium for 3-4 weeks, with selective medium changes every 3-4 days. Transformation Assay. The frequencies of morphological and anchorage-independent transformants within the treated cultures were determined concomitantly with the somatic mutation frequencies. For the respreading assay, cells at each passage were plated at low density and incubated for eight days to form colonies. These colonies were enumerated to determine cytotoxity and also scored for morphological transformation with a stereodissecting microscope, using established criteria (9, 10). At each passage, 200-600 colonies were examined from each culture. Additionally, 106 cells from each culture were tested at each passage for colony formation in semisolid agar by using procedures described by Macpherson and Montagnier (16), as modified by Kakunaga and Kamahora (17). Suspensions of 105 cells in 4 ml of 0.3% Difco agar in complete medium supplemented with 0.1% bactopeptone were plated in 60-mm dishes over a basal layer of 0.6% agar in complete medium. All plates were incubated at 370 in a 5% C02-humidified atmo-

sphere for 28 days, after which time the number of colonies was determined. Cloning efficiency in soft agar was expressed as the percentage of plated cells that formed visible colonies containing over 25 cells. The population doublings were determined from the number of cells obtained at confluency, when a subculture was established. In the direct transformation assay, colonies formed eight days after treatment were examined for morphological transformation. (These plates were established and treated in parallel with those used for mutation assay.) RESULTS

The Detection Time of Morphological Transformation and Anchorage-Independent Growth in Syrian Hamster Embryo Cells after Exposure to Benzo[alpyrene. The temporal acquisition of various phenotypic transformations of Syrian hamster embryo cells after exposure to carcinogen or mutagen has been under active investigation in our laboratory (6, 7). Table 1 shows the temporal relationship between the appearance of morphologically transformed colonies and the appearance of colonies capable of growth in soft agar (anchor-

age-independent growth). As shown, morphologically transformed colonies appeared in the population 8 days after treatment. Anchorage-independent growth was measured at every passage after treatment with benzofa Ipyrene by testing 106 cells from each culture for growth in agar. Anchorageindependent growth was not observed until 6-15 passages after treatment, depending on individual experiments. The detection time of this transformation thus required 32-75 population doublings. When initially detected at 6-9 post-treatment passages, the frequency of this alteration was approximately 10-6 to 10-5 and increased slightly at later passages. Other studies involving isolated clones indicated that the long delay in expression of this transformed phenotype is not due to selection of a few transformed cells present early after treatment (6). Quantitative Comparisons among Somatic Mutation, Morphological Transformation, and Anchorage-Independent Growth. The effects of variation of the following parameters Table 1. The temporal relationship between the appearance of morphologically transformed colonies and the appearance of colonies grown in soft agar in Syrian hamster embryo cells exposed to benzo[alpyrene*

Total population Growth in Post- % morphologically doublings at soft agar, treatment transformed colonies/106 cells subculture§ coloniest passaget 0 0 1.1 3.8 0 1 1.1 2 3 4 5 6 7 8 9 12

0.75 0.5 2.0 1.7 2.7 7.2 11.8 25.0 >90

jtg

0 0 0 0 3 16 18 41 49

7.9 13.5 19.2 27.4 32.4 37.0 47.7 46.9 55.0

* Cells were exposed to 1 of benzo[alpyrene per ml for 24 hr, grown in mass culture, and assayed as described for each transformed phenotype when subcultured at various passages after treatment. t The cells were subcultured every 6-9 days. Number of colonies judged morphologically transformed by the criteria described per total number of colonies X 100%. § Number of population doublings calculated from number of initial cells per culture and number of cells obtained at time of subculture.

Cell Biology: Barrett and Ts'o

Proc.

the frequencies of somatic mutation and morphological transformation were examined: (i) The genetic loci for the somatic mutations-i.e., ATPase mutants vs. HPRT mutants; (ii) the selective agent for mutations of HPRT-i.e., 6-thioguanine vs. 8-azaguanine; (iii) the method of assay-i.e., respreading vs. direct; (iv) the animal source for the cell preparations; (v) the type of carcinogen; and (vi) the dose of carcinogen. The data for these variables are summarized in Tables 2 and 3. For the mutation studies the conditions employed have been shown to select for mutants that are stable in the absence of the selective agent and have a low reversion frequency. The AGr mutants are crossresistant to 6-thioguanine, and mutants selected with either purine analog have a low level of HPRT activity. The ouabain resistant mutants have Na+/K+ ATPase activity with an altered sensitivity to ouabain (8). The mutation frequencies in Tables 2 and 3 were determined at an optimal expression time (6-8 population doublings) after carcinogen exposure. The frequencies were also corrected for cytotoxicity and for the recovery efficiency of the mutant cells during the selective assay, which was 18-20%, as determined by a reconstitution experiment (8). Similar frequencies of induced mutations of HPRT were observed with either 6-thioguanine or 8-azaguanine selection. Isolated mutant colonies were also qualitatively similar regardless of selective agent. A small variation (1.3-4 fold) in the induced frequencies of ATPase vs. HPRT mutants was observed. The direct assay resulted in a higher frequency of mutants than the respreading assay, which may reflect the uncertainty of quantitating somatic mutation by this method (18). However, this experimental approach is used frequently for quantitating morphological transformation (9, 10); therefore, we have included the data for somatic mutation to allow

Morphological transformation frequency

Mutation frequency

(X 103)

Treatment

(X 105)

SHE*-1 SHE-2

1 AM MNNG (2 hr)