A number of newly isolated clonal cell lines derived from diploid mouse embryo cells transformed by SV40 were examined in vitro and in vivo. Although these lines showed the properties that define transformation in vitro, they were not tumorigenic for many passages after their initial isolation.

Cells from tumours eventually produced by the SV40-transformed cells were fused with diploid mouse embryo cells. The hybrids formed were initially non-tumorigenic. This indicates that a normal diploid cell can suppress the malignant phenotype of a tumorigenic SV4O-transformed cell. The hybrid cells did, however, express the SV40 T antigen and they had a clearly transformed phenotype in vitro. It thus appears that neither the transformed phenotype nor the expression of the SV40 T antigen are enough to endow a cell with the ability to grow progressively in vivo.

The relationship between the transformed phenotype and tumorigenicity was further studied by fusing malignant mouse melanoma cells with non-tumorigenic SV4o-transformed cells. The hybrids expressed the transformed phenotype in vitro but were unable to form tumours in vivo. The changes that occur in cells after transformation by SV40 do not apparently affect the ability of these cells to suppress the malignant phenotype of tumour cells.

There are a number of reports in the literature that cells transformed in vitro by SV40 may fail to produce tumours in appropriately immuno-suppressed hosts. SV40-transformed cells of mouse (Black & Rowe, 1963; Kit, Kurimura & Dubbs, 1969; Wesslen, 1970; Tevethia & McMillan, 1974), hamster (Zuna & Lehman, 1977), and human origin (Stiles, Desmond, Sato & Saier, 1975; Stiles et al. 1976) have all been shown not to be tumorigenic immediately after transformation. However, the relationship between the transformed phenotype in vitro and tumorigenicity in immunosuppressed syngeneic hosts has not been systematically explored. We thought that it would be worthwhile to isolate a new set of SV4O-transformed mouse embryo cell lines in order to be able to study this question in the period immediately after transformation in vitro and before extensive secondary changes occur. We have also fused tumorigenic SV4o-transformed cells with diploid mouse embryo cells to examine the phenotype of the hybrid cells in vitro and their tumorigenicity in immuno-suppressed syngeneic hosts. Normal diploid cells of different tissue origins have been shown to be able to suppress the tumorigenicity of a wide range of spontaneous and experimentally induced tumours when the normal and the malignant cells were fused into hybrids (Wiener, Klein & Harris, 1971, 1974; Jonasson, Povey & Harris, 1977; Stanbridge, 1976; Shkolnik & Sachs, 1978; Marshall & Dave, 1978).

An apparent exception to the general finding that crosses between malignant tumour cells and normal diploid cells are initially non-tumorigenic is to be found in the work of Croce, Aden & Koprowski (1975a, b) and Koprowski & Croce (1977). These authors claim that crosses between SV/p-transformed human cells and mouse peritoneal exudate cells are tumorigenic ab initio, and this even when the hybrid cell lines contain only a single chromosome derived from the transformed human parental cell. They argue that a single chromosome bearing an active SV40 genome is enough to confer the malignant phenotype on the hybrid cell. Such a dramatic conclusion in conflict with much previous work on SV4O-transformed cells obviously warranted further study.

Cell culture

Cells were grown in 75-cm,2 plastic tissue culture flasks or 20-oz. (∼ 560 ml) glass medicine prescription bottles in Eagle’s Minimum Essential Medium (Eagle, 1959) plus 10% foetal calf serum (FCS) with added penicillin (100 units/ml), kanamycin (90 μg/ml), and streptomycin (120 μg/ml). The culture vessels were gassed with a mixture of 5 % CO2 in air, and the cells were grown in an incubator at 37 °C.

Cell lines

Embryo cells

Ten to fourteen-day-old embryos were minced and trypsinized (0·08% (w/v) trypsin in phosphate-buffered saline (PBS) to which 0·2 % (w/v) ethylenediaminetetra-acetic acid had been added). The cells were grown in plastic flasks. CBA/H-T6T6 embryos were chosen. These animals carry the T6 marker in the homozygous condition. This marker is a radiation-induced translocation chromosome easily identifiable in conventionally stained metaphase spreads (Carter, Lyon & Phillips, 1955).

Transformed cell

Plaque purified Hilleman strain (large plaque) SV40 and a small plaque derivative (Suarez et al. 1974) were used to transform the cells, essentially as described by Todaro & Green (1966).

C1SV-5

T6T6 embryo cells were infected with small plaque SV40 at a multiplicity of infection (MOI) of about 5. A colony containing morphologically transformed cells was isolated from the plastic culture surface and established into a line.

C3SV-6A, C3SF-10

These two cell lines were derived in the same way as C1SV-5. The three cell lines represent independent transformation events.

C12SV-1B, CfzSV-2

T6T6 embryo cells growing as a monolayer were treated with large plaque SV40 at an MOI of about 5, and the infected cells were suspended in 1·2 % Methocel. Two independent colonies of cells, selected in this way for their ability to clone in the absence of anchorage, were established as cell lines.

C13SV-1D

This cell line was established from a focus of transformed cells arising in a monolayer of T6T6 embryo cells infected with large plaque SV40 at an MOI of about 3 × 107.

C1SV-5 (EMS), C3SV-10 (EMS)

These two cell lines were derived from bulk cultures of C1SV-5 and C3SV-10 cells treated with ethyl methanesulphonate (EMS) for 18 h at 200 μg/ml essentially as described by Chasin (1973). Cell line C1SV-5 was treated in passage (p) 55; C3SV-10 was treated in p 12. The derived sub-lines were grown independently of the parental cells. (Passage numbers of the sub-lines refer to total number of passages from the origin of the parental lines.)

Assays for the transformed phenotype in vitro

Expression of SV40 T antigen

T antigen was detected essentially by the method of Pope & Rowe (1964). Sub-confluent cultures of cells growing on n-tnm glass coverslips were fixed in acetone at −20 °C for to min, washed extensively in PBS and stained with hamster anti-SV4O tumour serum (Flow Laboratories) for 45–60 min at 37 °C. After further washes in PBS, the cells were counter-stained with fluorescein isothiocyanate-conjugated swine anti-hamster globulin anti-serum (Nordic Laboratories) for 45–60 min at 37 °C. After final rinses in PBS, the coverslips were mounted in PBS: glycerol (1: 1) and examined under u.v. light with a Leitz Orthoplan microscope. At least 200 cells were scored in each assay.

Growth curves

Cells were seeded into replicate 35-mm plastic Petri dishes in MEM plus 10 % FCS at 2 to 4 × 104 cells per dish and incubated at 37 °C. On the following day the medium in half the dishes was replaced by MEM plus 1 % FCS, the other half received a change of MEM plus 10% FCS. The cell numbers in two replicate dishes were estimated in a Coulter counter Model ZB after the cells had been removed from the dishes with trypsin. On each following day, five replicate dishes from each of the two serum groups were counted. The medium in the cultures was changed every second day.

Serum dependence

The serum dependence of a number of cell lines was assayed in the following way. The cells were plated at one tenth confluency in MEM plus 10 % FCS. On the following day, the medium was changed to MEM plus 1 % serum. Cells able to grow to confluency in 1 % serum were scored as ‘serum-independent’.

Anchorage dependence

The ability of cells to form colonies in the absence of anchorage was estimated by suspending them at densities of 103–106 cells per 60-mm Petri dish in 5 ml of 0–3 % Bacto agar in MEM (MacPherson & Montagnier, 1964) or 1·2 % Methocel in MEM (Risser & Pollack, 1974) over a base layer of 0·9 % Bacto agar in MEM. The media were supplemented with 10% FCS and, in addition to the usual antibiotics, 40 units/ml of mycostatin. Each assay used a minimum of three dishes per cell line.

Three sub-lines of Ci SV-5 cells (in p 61) were established from colonies growing in suspension in 0·3% agar. The cell lines were designated C1SV-5/AS1, C1SV-5/AS2, C1SV-5/AS5.

Karyology

Chromosome spreads were prepared by the air-drying method of Rothfels & Siminovitch (1958). Some metaphase preparations were G-banded by the technique of Gallimore & Richardson (1973). In each case, at least 50 metaphases were counted.

Cell fusion

(1) Parental cells

Embryo cells

Rbz/Bnr strain mouse embryos were used to prepare diploid cells for the hybridization experiments. These animals carry a Robertsonian translocation involving chromosomes 4 and 6 in the homozygous condition.

PG 19 melanoma cells

The cells were derived from a spontaneous melanoma arising in a C57BI/6J mouse and were selected for resistance to 1·5 μg/ml 6-thioguanine (Jonasson et al. 1977). The cells contain metacentric marker chromosomes. An inoculum of 5 × 104 cells gives rise to tumours in 100% of syngeneic mice.

135-T-1, 137-T-1

These two lines were established from expiants of tumours arising from the inoculation of C3SV-10 (EMS) cells into immuno-suppressed newborn syngeneic mice. The C3SV-10 (EMS) cells used for the inoculation had been further mutagenized with N-methyl-NPR-nitro-nitrosoguanidine at 0·25 μg/ml for 24 h, essentially as described by Kao & Puck (1969). The cells were subsequently selected for resistance to 2 μg/ml 6-thioguanine. They were unable to survive in hypoxanthine-aminopterin-thymidine (HAT) medium (Littlefield, 1964).

1.5E-T-5

This line was obtained from a tumour arising from the inoculation of C1SV-5 (EMS) cells (in p 75) into immuno-suppressed syngeneic newborn mice.

(2) Hybrid cell lines

F14–4B, F14–5A

These two independently derived hybrid lines were isolated from hybrid colonies that arose from a mixture of 106 135-T-1 cells and 105 Rb2 embryo cells fused together by the method of Harris & Watkins (1965). HAT medium was used to select against the parental tumour cells. Embryo cells clone poorly if plated at low density, which facilitates selection against them.

F17–3C

A clonal hybrid line derived from the fusion of 106 137-T-1 cells and 105 Rbz cells. HAT medium was used for selection.

Fi 8–1

A clonal hybrid line derived from the fusion of 5 × 106 1.5 E-T-5 cells and 5×104 Rb2 cells. In the absence of a selectable marker in the parental tumour cells, γ-irradiation was used to depress their cloning efficiency (Pontecorvo, 1971). A dose of 14·5 J kg-1 of γ-irradiation depressed the cloning efficiency of the 1.5E-T-5 cells to 10−5.

Fg-2B, Fg-3A

Two clonal hybrid lines derived from the fusion of 106 PG19 cells and 106 C3SV-10 cells in p 29. C3SV-10 cells are not tumorigenic. To facilitate selection, they were irradiated prior to cell fusion and the hybrids were selected in HAT medium.

All assays were done as soon as possible after isolation of the hybrid cells, in order to reduce chromosome losses as far as possible.

Assay for tumorigenicity

Cells, in a volume of 0·1 ml of PBS, were inoculated subcutaneously into the supra-scapular region of irradiated genetically compatible newborn mice. The mice, aged 1–3 days, were given 4 or 4·5 J kg-1 of whole body y-irradiation from a Gammacell 40 Caesium 137 Irradiation Unit. After inoculation, the animals were examined at least once a week for the presence of tumours. Animals failing to produce a tumour within 12 weeks were scored as negative. Only progressive growths greater than 1 cm in diameter were scored as tumours.

Hybrid cells selected in HAT medium were passaged through aminopterin-free ‘HT’ medium before being inoculated. Hybrid cells were inoculated either from the second passage in HT medium or from MEM plus 10% FCS.

The following genetically compatible crosses were used as test animals: CBA/H-T6T6 × C57BI/6J, CBA/H-T6T6 × Rb2/Bnr, Rb2/Bnr × C57BI/6J.

Transformed cells

All the SV4o-transformed cell lines expressed the morphological features of the transformed phenotype, namely, growth in disorderly arrays, a high mitotic rate even when confluent and a tendency to form multilayers.

SK40 T antigen

The cell lines were assayed after a few passages in culture, to ensure that abortive transformants did not affect the assay. The results are shown in Table 1. All the cell lines expressed T antigen in more than 95% of their nuclei, thus indicating that they corresponded to the ‘standard transformants’ of Risser & Pollack’s (1974) classification.

Table 1.

Assays of the transformed phenotype in SV4.0-transf0rm.ed mouse embryo cell lines

Assays of the transformed phenotype in SV4.0-transf0rm.ed mouse embryo cell lines
Assays of the transformed phenotype in SV4.0-transf0rm.ed mouse embryo cell lines

Anchorage dependence

The results of assays of anchorage dependence of the cell lines are also given in Table 1. It will be seen that all except one of the lines were able to clone well in the absence of a solid substrate. Cell line C13SV-1D had a rather low efficiency of cloning compared to the other cell lines. Cell lines C12SV-1B and C12SV-3, although selected in 12% Methocel, had cloning efficiencies comparable to the small plaque SVqo-transformed cell lines.

Growth curves

Growth curve experiments were done on five of the cell lines to determine their sensitivity to density-dependent inhibition of growth and to assess their dependence on serum. The curves for C1SV-5, C3SV-6A, C3SV-10, C1SV-5 (EMS) and C3SV-10 (EMS) are given in Figs. 1–5. The curves for all the cell lines are roughly similar: the cells grew in both low and high serum concentrations and the growth rate was initially logarithmic. All the cell lines are clearly ‘serum transformed’ (Risser & Pollack, 1974). None of the lines shows a clear saturation density. Cell lines C1SV-5, C3SV-10 and C3SV-10 (EMS) continued to grow to the end of the assay, when the monolayers spontaneously detached from the dishes. These lines obviously lack density-dependent inhibition of growth. Cell lines C3SV-6A and C1SV-5 (EMS) showed some tendency to saturate, i.e. to show some density-dependent inhibition of growth.

Fig. 1.

Growth curves of small plaque SV4O-transformed cell lines. ○, mean counts of cells growing in MEM + 10% FCS (vertical bars indicate ± 1 8.E.); •, mean counts of cells growing in MEM ± 1 % FCS. The slopes were calculated from a linear regression of the logarithms of cell count against time made by the method of least squares. The lines for growth in 1 and 10% serum were constrained to meet at zero time.

Fig. 1.

Growth curves of small plaque SV4O-transformed cell lines. ○, mean counts of cells growing in MEM + 10% FCS (vertical bars indicate ± 1 8.E.); •, mean counts of cells growing in MEM ± 1 % FCS. The slopes were calculated from a linear regression of the logarithms of cell count against time made by the method of least squares. The lines for growth in 1 and 10% serum were constrained to meet at zero time.

Fig. 2.

Growth curves of small plaque SV4O-transformed cell lines. ○, mean counts of cells growing in MEM + 10% FCS (vertical bars indicate ± 1 8.E.); •, mean counts of cells growing in MEM ± 1 % FCS. The slopes were calculated from a linear regression of the logarithms of cell count against time made by the method of least squares. The lines for growth in 1 and 10% serum were constrained to meet at zero time.

Fig. 2.

Growth curves of small plaque SV4O-transformed cell lines. ○, mean counts of cells growing in MEM + 10% FCS (vertical bars indicate ± 1 8.E.); •, mean counts of cells growing in MEM ± 1 % FCS. The slopes were calculated from a linear regression of the logarithms of cell count against time made by the method of least squares. The lines for growth in 1 and 10% serum were constrained to meet at zero time.

Fig. 3.

Growth curves of small plaque SV4O-transformed cell lines. ○, mean counts of cells growing in MEM + 10% FCS (vertical bars indicate ± 1 8.E.); •, mean counts of cells growing in MEM ± 1 % FCS. The slopes were calculated from a linear regression of the logarithms of cell count against time made by the method of least squares. The lines for growth in 1 and 10% serum were constrained to meet at zero time.

Fig. 3.

Growth curves of small plaque SV4O-transformed cell lines. ○, mean counts of cells growing in MEM + 10% FCS (vertical bars indicate ± 1 8.E.); •, mean counts of cells growing in MEM ± 1 % FCS. The slopes were calculated from a linear regression of the logarithms of cell count against time made by the method of least squares. The lines for growth in 1 and 10% serum were constrained to meet at zero time.

Fig. 4.

Growth curves of small plaque SV4O-transformed cell lines. ○, mean counts of cells growing in MEM + 10% FCS (vertical bars indicate ± 1 8.E.); •, mean counts of cells growing in MEM ± 1 % FCS. The slopes were calculated from a linear regression of the logarithms of cell count against time made by the method of least squares. The lines for growth in 1 and 10% serum were constrained to meet at zero time.

Fig. 4.

Growth curves of small plaque SV4O-transformed cell lines. ○, mean counts of cells growing in MEM + 10% FCS (vertical bars indicate ± 1 8.E.); •, mean counts of cells growing in MEM ± 1 % FCS. The slopes were calculated from a linear regression of the logarithms of cell count against time made by the method of least squares. The lines for growth in 1 and 10% serum were constrained to meet at zero time.

Fig. 5.

Growth curves of small plaque SV4O-transformed cell lines. ○, mean counts of cells growing in MEM + 10% FCS (vertical bars indicate ± 1 8.E.); •, mean counts of cells growing in MEM ± 1 % FCS. The slopes were calculated from a linear regression of the logarithms of cell count against time made by the method of least squares. The lines for growth in 1 and 10% serum were constrained to meet at zero time.

Fig. 5.

Growth curves of small plaque SV4O-transformed cell lines. ○, mean counts of cells growing in MEM + 10% FCS (vertical bars indicate ± 1 8.E.); •, mean counts of cells growing in MEM ± 1 % FCS. The slopes were calculated from a linear regression of the logarithms of cell count against time made by the method of least squares. The lines for growth in 1 and 10% serum were constrained to meet at zero time.

Serum dependence

The serum dependence of cell lines C12SV-1B, C12SV-3 and C13SV-1D was assayed qualitatively by the simple method described. All three cell lines were able to grow from low density to confluence in MEM plus 1 % FCS, indicating a low dependence on serum.

Judged by the above criteria, all the cell lines show a clearly transformed phenotype. Although cell line C13SV-1D had a rather low efficiency of cloning in Methocel, its growth pattern, serum independence and the expression of SV40 T antigen warranted its classification also as a standard transformant.

Karyology

For each cell line chromosome counts were made on 50 metaphase spreads. All the lines were heteroploid and showed a wide range of chromosome numbers. Most of the cell lines were found to contain a majority of cells with hypo-tetra-ploid chromosome numbers. Cell line C13SV-1D, however, had an extremely wide range of chromosome numbers (78–172), with most cells falling within the hypooctaploid range. None of the lines showed a clear modal chromosome number.

Tumorigenicity

(1) Cell lines C1SV-5, C3SV-6A, C3SV-10, Ci2SV-iB, C12SV-3, C13SV-1D

The cell lines were tested at increasing passage levels, with inocula ranging from 9 ×105 to 8 × 106 per animal. Table 2 shows the results. None of the animals inoculated with these cell lines gave rise to progressively growing tumours during a 12-week period of observation. Cell lines C1SV-5, C3SV-6A and C3SV-10 were inoculated at higher passage levels, but produced no tumours. Cell line C3SV-10 at p 129 was inoculated into four mice via the intra-peritoneal route, at a dose of 3 × 106 cells, but no tumours were observed. This cell line was also tested at p 7 for the ability to give rise to tumours in athymie nude mice. Of ten animals given doses of cells of 101 or 107 each, none produced a tumour.

Table 2.

Tumorigenicity of transformed cell lines

Tumorigenicity of transformed cell lines
Tumorigenicity of transformed cell lines

Although no progressively growing tumours were observed in early passages, in a few instances, particularly after the inoculation of high doses of cells, transient swellings were noted at the site of inoculation. The swellings appeared 3–7 days after inoculation and usually completely regressed by the 15th day. A similar phenomenon has been described by Stiles et al. (1975,1976). Immunological rejection can hardly explain the complete resolution of these swellings in so short a time and, indeed, Stiles et al. (1975) have reported that animals that have borne such ‘abortive nodules’ are susceptible to subsequent challenge with frankly malignant cells.

(2) Cell lines C1SV-5AS1, C1SV-5IAS2 and C1SV-5IAS5

It has been reported that anchorage-independent growth is correlated with the ability to grow in vivo (Freedman & Shin, 1974; Shin, Freedman, Risser & Pollack, 1975). We examined this question in our material. Three cell lines, selected from colonies in soft agar, were assayed for their ability to give rise to tumours. The results are given in Table 3. It is clear that none of the lines gave rise to tumours. These results taken together with those obtained with the other anchorage-independent cell lines (Table 2), and particularly those obtained with cell lines C12SV-1B and C12SV-3, which were originally selected in suspension, give no support to the idea that anchorage-independence and tumorigenicity are invariably correlated.

Table 3.

Tumorigenicity of lines selected in agar

Tumorigenicity of lines selected in agar
Tumorigenicity of lines selected in agar
Table 4.

Tumorigenicity of cell line C1SV-5 (EMS)

Tumorigenicity of cell line C1SV-5 (EMS)
Tumorigenicity of cell line C1SV-5 (EMS)

(3) Cell lines C1SV-5 (EMS) and C3SV-10 (EMS)

After treatment with EMS the two cell lines were assayed for their ability to produce tumours in immuno-suppressed syngeneic newborn mice. The results are given in Tables 4 and 5. Immediately after treatment, both cell lines were still non-tumorigenic. However, after a period of subcultivation, both cell lines gave rise to progressively growing tumours. The tumours were identified histologically as invasive fibro-sarcomas.

Table 5.

Tumorigenicity of cell line C3SF-10 (EMS)

Tumorigenicity of cell line C3SF-10 (EMS)
Tumorigenicity of cell line C3SF-10 (EMS)

Hybrids between tumour cells derived from SV β-transformed cells and normal diploid embryo cells

Parental cells

(1) Rb2/Bnr embryo cells

The cells used for the fusions were in the second passage after explantation. They have a diploid number of 38 chromosomes, with two marker chromosomes.

(2) Tumour cells

The results of the assays for the parameters of the transformed phenotype in these cells are shown in Table 6. The cells are morphologically transformed, express T antigen and are serum- and anchorage-independent.

Table 6.

Assays of the transformed phenotype in SV40-transfarmed tumour cells

Assays of the transformed phenotype in SV40-transfarmed tumour cells
Assays of the transformed phenotype in SV40-transfarmed tumour cells

An inoculum of 106 cells of each of the tumour cell lines gives rise to tumours in 100% of immuno-suppressed syngeneic newborn mice. The mean latent periods before the appearance of tumours 1 cm in diameter were as follows: 1·5E-T-5, 9·8 ± 0·6 days; 135-T-1, 15 ±0 days; and 137-T-1, 8·6 ±0·5 days. (The deviations represent i S.E.)

Chromosome counts on 50 metaphases of each of the tumour cell lines are given in Figs. 6–8. A wide range of chromosome numbers was observed, with most of the cells of each line falling in the hypo-tetraploid range. A modal number of 2 T6 markers was found in 1.5E-T-5 (range 1–4) and 135-T-1 (range 1–3). A modal number of 3 T6 markers (range 1–4) was found in 137-T-1.

Fig. 6.

Chromosome constitutions of parental SV4o-transformed tumour lines.

Fig. 6.

Chromosome constitutions of parental SV4o-transformed tumour lines.

Fig. 7.

Chromosome constitutions of parental SV4o-transformed tumour lines.

Fig. 7.

Chromosome constitutions of parental SV4o-transformed tumour lines.

Fig. 8.

Chromosome constitutions of parental SV4o-transformed tumour lines.

Fig. 8.

Chromosome constitutions of parental SV4o-transformed tumour lines.

Properties of the hybrid cells

The colonies of hybrid cells arising in the fused cell populations were morphologically transformed. This indicated that the transformed phenotype was present in the immediate progeny of the fused cells, although some chromosome losses probably had occurred by the time the colonies could be identified.

Karyology

Chromosome spreads of each putative hybrid cell line were analysed as soon as possible. The results are shown in Figs. 9–12 and the number of marker chromosomes are given in Table 7. All the cell lines contained markers from both parents, thus confirming their hybrid nature. Cell lines F14–5A, F17–3C and F18–1 were probably bi-parental in origin. In cell line F18–1 the modal number of T6 markers was zero, which suggests that chromosome losses were occurring. It is a common experience that hybrids containing a γ-irradiated parent cell tend to lose the irradiated chromosomes preferentially (Pontecorvo, 1971). Some cells of line F14–4B contained more than two Rb2 markers and more than four T6 markers; this suggests that the cells may have been tetraparental in origin. However, since SV40 is known to cause polyploidization of infected cells (Cooper & Black, 1963; Lehman, 1974), it is possible that these cells of higher ploidy might have arisen from cells that were originally bi-parental hybrids.

Table 7.

Marker chromosomes in hybrids between SV40-transformed tumour cells and normal diploid embryo cells

Marker chromosomes in hybrids between SV40-transformed tumour cells and normal diploid embryo cells
Marker chromosomes in hybrids between SV40-transformed tumour cells and normal diploid embryo cells
Fig. 9.

Chromosome constitutions of hybrids between SV4o-transformed cells and normal diploid embryo cells.

Fig. 9.

Chromosome constitutions of hybrids between SV4o-transformed cells and normal diploid embryo cells.

In all the hybrid cell lines the Rbz marker was present in at least one copy in the majority of the cells. This indicates that there can have been no strong selection in these hybrids against the diploid chromosomes 4 (involved in the marker) during growth in vitro. Such selection is of importance in the generation of hybrid cell lines from other crosses between malignant and normal cells (Jonasson et al. 1977). Jonasson et al. observed systematic elimination of the chromosomes 4 derived from the diploid parent in all but one of the malignant × normal cell hybrids that they examined. The one exception was in crosses between the SEWA line, a sarcoma induced by polyoma virus, and diploid cells. No selection against the diploid cell chromosomes 4 could be demonstrated in this case. It seems that with SVzp-transformed tumour cells, as with the polyoma-induced SEWA tumour cells, selection against the chromosome 4 derived from the normal parent is not necessary in order to generate the hybrid cell lines.

Transformed phenotype

The hybrid cells were assayed for the expression of SV40 T antigen and for their serum- and anchorage-dependence. The results are shown in Table 8. All the cell lines expressed SV40 T antigen in 100% of their nuclei, and all but one could be classified as transformed by all the criteria tested for. Hybrid line F18–1 had a rather low cloning efficiency in suspension, but was transformed as judged by the other criteria.

Table 8.

Assays of the transformed phenotype in hybrids between SV40-transformed tumour cells and normal diploid embryo cells

Assays of the transformed phenotype in hybrids between SV40-transformed tumour cells and normal diploid embryo cells
Assays of the transformed phenotype in hybrids between SV40-transformed tumour cells and normal diploid embryo cells

Tumorigenicity

It can be seen that all the hybrid cell lines showed a greatly reduced take incidence (0 – 28%) compared to that of the parental tumour lines (100%). The latent periods were also obviously longer than those of the tumour cells, an indication that those tumours that did arise were probably not derived from the unselected growth of the bulk of the inoculum, but from selected sub-populations of the hybrid cells. The present experiments thus indicate that the malignant phenotype of the SV40-transformed tumour cell is suppressed by the normal diploid cell, as has been found to be the case with other malignant × normal diploid crosses.

Although few progressive tumours arose, several instances of ‘primary growth’ (Jonasson et al. 1977) were observed. Some hybrids between SV40-transformed tumour cells and normal diploid cells, like some melanoma × fibroblast hybrids previously described, appear to undergo a number of cell divisions in vivo and then abort without giving rise to progressive tumours. Histological examination of these ‘primary growths’ showed a central necrotic area, with no evidence of inflammation or lymphoid cell involvement.

A progressive tumour arising from the inoculation of the hybrid F14 – 5A cells was explanted, and the cells, designated 225-T-1, were examined further. These cells resembled the F14 – 5A cells and were morphologically transformed. 99% of them expressed T antigen. A karyological analysis showed that the cells contained a wide range of chromosomes. The chromosome profile is shown in Fig. 13. If this is compared with the profile of the F14 – 5A cells (Fig. 10), it will be seen that most of the tumour cells appear to have slightly lower chromosome numbers than the hybrid cells inoculated. When the explanted tumour cells were re-inoculated at a dose of 10® cells per animal, 100% of the mice developed tumours, with a mean latent period of 10 days. The differences in take incidence and latent period between the 225 – T-1 tumour cells and the F14 – 5A cells (hybrid cells originally injected) (Table 9) confirm that the tumours arose not by unselected growth of the bulk of the inoculum, but by selective overgrowth of a sub-population of malignant cells.

Table 9.

Tumorigenicity of hybrids between SV40-transformed tumour cells and normal diploid embryo cells

Tumorigenicity of hybrids between SV40-transformed tumour cells and normal diploid embryo cells
Tumorigenicity of hybrids between SV40-transformed tumour cells and normal diploid embryo cells
Fig. 10.

Chromosome constitutions of hybrids between SV4o-transformed cells and normal diploid embryo cells.

Fig. 10.

Chromosome constitutions of hybrids between SV4o-transformed cells and normal diploid embryo cells.

Fig. 11.

Chromosome constitutions of hybrids between SV4o-transformed cells and normal diploid embryo cells.

Fig. 11.

Chromosome constitutions of hybrids between SV4o-transformed cells and normal diploid embryo cells.

Fig. 12.

Chromosome constitutions of hybrids between SV4o-transformed cells and normal diploid embryo cells.

Fig. 12.

Chromosome constitutions of hybrids between SV4o-transformed cells and normal diploid embryo cells.

Fig. 13.

Chromosome constitution of hybrid tumour 225-T-1.

Fig. 13.

Chromosome constitution of hybrid tumour 225-T-1.

Hybrids between PGig melanoma cells and non-tumorigenic SV40-transformed cell line CβSV-xv

The hybrid nature of the cell lines F9 – 2B and F9 – 3A was confirmed by karyology. The chromosome profiles of the parental cells and of the hybrid clones are shown in Figs. 14–17. Both hybrid lines contained marker chromosomes from both parents. F9 – 2B is a bi-parental hybrid while F9–3 A is a tri-parental hybrid containing two PG19 genomes and one C3SV-10 genome.

Fig. 14.

Chromosome constitutions of PG19 cells, C3SV-10 cells and hybrids F9-2B and F9–3A.

Fig. 14.

Chromosome constitutions of PG19 cells, C3SV-10 cells and hybrids F9-2B and F9–3A.

Fig. 15.

Chromosome constitutions of PG19 cells, C3SV-10 cells and hybrids F9-2B and F9–3A.

Fig. 15.

Chromosome constitutions of PG19 cells, C3SV-10 cells and hybrids F9-2B and F9–3A.

Fig. 16.

Chromosome constitutions of PG19 cells, C3SV-10 cells and hybrids F9-2B and F9–3A.

Fig. 16.

Chromosome constitutions of PG19 cells, C3SV-10 cells and hybrids F9-2B and F9–3A.

Fig. 17.

Chromosome constitutions of PG19 cells, C3SV-10 cells and hybrids F9-2B and F9–3A.

Fig. 17.

Chromosome constitutions of PG19 cells, C3SV-10 cells and hybrids F9-2B and F9–3A.

Transformed phenotype

The two cell lines were morphologically transformed. F9–2B expressed SV40 T antigen in 100% of its nuclei, F9–3A in 99%.

Both cell lines were serum-independent, as measured by the simple assay; 0 · 35 ± 0 · 06% of F9–2B cells and 2,4± 0· 1% of F9–3A cells were able to form colonies in 1 · 2% Methocel.

Tumorigenicity

The two cell lines were inoculated into newborn syngeneic -γ -irradiated mice at a dose of 5 ×105 or 10 6 cells per animal. The results are shown in Table 10. No tumours arose in any of the inoculated mice. Non-tumorigenic SV40-transformed cells thus appear to be able to suppress the malignant phenotype of the melanoma cells.

Table 10.

Tumorigenicity of hybrids between PG19 melanoma cells and non-tumorigenic SV40-transformed cells

Tumorigenicity of hybrids between PG19 melanoma cells and non-tumorigenic SV40-transformed cells
Tumorigenicity of hybrids between PG19 melanoma cells and non-tumorigenic SV40-transformed cells

All the cell lines established after infection of the embryo cells with SV40 expressed a transformed phenotype in vitro. Only one of the cell lines was unable to grow well in the absence of anchorage, but this cell line could be classified as transformed by the criteria of growth pattern, serum-independence and SV40 T antigen production. None of the lines were, however, tumorigenic immediately after transformation. On one transformed cell line, tests were done by subcutaneous and intraperitoneal inoculation into genetically compatible newborn mice immunosuppressed by γ-irradiation and also by subcutaneous inoculation into athymie mice; but no tumours resulted. It is clear that cells transformed in vitro may be totally unable to grow progressively in vivo. It appears that the strain of virus, the multiplicity of infection and the selective conditions used to isolate the cells have no effect on their tumorigenicity. A correlation between multiplicity of infection and tumorigenic potential has been reported by Harwood & Gallimore (1975) for rat cells transformed by adenovirus type 2, but no such correlation was found in the present experiments. Two transformed cell lines eventually gave rise to tumours, but only after an extended period of cultivation in vitro. This observation confirms the findings of Kit et al. (1969), Wesslen (1970) and Tevethia & McMillan (1974).

Hybrids formed by fusing tumorigenic SV4O-transformed cells with normal diploid embryo cells were all found to have greatly reduced tumorigenicity compared with the parental tumour cells. The most striking results were those obtained with hybrids F14-4B and F18–1: with F14–4B only one tumour arose in 29 test animals; F18–1 failed to generate any tumours. Although there was evidence in the latter case that some chromosomes derived from the tumour parent cell were lost from the hybrid, all the cells continued to express T antigen and thus clearly retained the SV40 genome. Tumours derived from SV4O-transformed cells thus do not differ from other virus-induced tumours that we have studied in cell fusion experiments: when these tumour cells are fused with normal diploid cells malignancy is initially suppressed. The argument advanced by Croce et al. (1975a, b) and Koprowski & Croce (1977) that the presence of an active SV40 genome is itself enough to confer tumorigenicity finds no support in our experiments.

These authors make a further important claim: that, in their interspecific crosses between SV4O-transformed human cells and normal diploid mouse cells, the diploid cell is unable to suppress the malignant phenotype of the tumorigenic parent cell. In our intraspecific crosses, this suppression was unequivocal. We do not, however, believe that the differences between the two sets of results are due to the different behaviour of inter- and intra-specific crosses, for it has previously been shown in this laboratory that suppression of malignancy can be achieved across the species barrier (Jonasson & Harris, 1977). We think it more likely that the explanation of the discrepancy is to be found in differences of experimental design. In testing the tumorigenicity of their hybrids in athymie mice, Croce et al. (1975 a) and Koprowski & Croce (1977) used what we would regard as exceptionally high inocula: between 107 and 108 cells. Even so, the hybrids appear initially to have produced tumours in only a minority of the animals injected, although no formal data on take incidence or latent period were presented. Nor were any systematic comparisons made between the take incidences under comparable conditions of the original SV4O-transformed human cells, of the early passages of hybrids between these cells and the diploid mouse cells, and of the cells in the tumours derived from these hybrids. Since the use of high inocula maximizes the chance of selecting malignant variants in the hybrid cell population, it is difficult, without more systematic data, to exclude the possibility that the tumours initially produced by the injection of the hybrid cells might have arisen from the selective overgrowth of malignant variants present in the initial inoculum, a situation that has been frequently observed in other crosses between malignant and non-malignant cells. In any case, we see nothing in the data so far presented by these authors that would justify the conclusion that in hybrid cells the activity of the SV40 genome confers tumorigenicity in a dominant fashion.

On the other hand we concur with these authors, and some others (see for example Jha & Ozer, 1976), in finding a transformed phenotype in vitro in our hybrids between SV4o-transformed cells and normal diploid cells; and we found this phenotype as soon as the hybrid colonies could be identified. While the karyotypic instability of SV40-transformed cells and of hybrids made with them precludes any formal statement about the dominance or recessivity of the transformed phenotype in vitro, we would have no quarrel with the idea that hybrid cells that contain active SV40 genomes might show a transformed phenotype in vitro. This idea would, of course, be consistent with the now widely-held view that the A gene product of SV40 is responsible for the maintenance of a transformed phenotype in cells carrying this virus (Martin & Chou, 1975; Tegtmeyer, 1975; Brugge & Butel, 1975; Osborn & Weber, 1975; Brockman, 1978). It does not, however, appear to be generally true that crosses between transformed and diploid cells show a transformed phenotype in vitro. In some cases, where the transformation is not known to be induced by oncogenic viruses, many of the features of the transformed phenotype may be suppressed by fusion with untransformed or normal diploid cells (Sager & Kovac, 1978; Marshall & Dave, 1978).

We thank Mr Gareth Plant for technical assistance. C. J.G. held a Medical Research Council Scholarship for Training in Research Methods. The work was supported by the Cancer Research Campaign.

Black
,
P. H.
&
Rowe
,
W. P.
(
1963
).
Sv40-induced proliferation of tissue culture cells of rabbit, mouse and porcine origin
.
Proc. Soc. exp. Biol. Med
.
114
,
721
727
.
Brockman
,
W. W.
(
1978
).
Transformation of BALB/C-3T3 cells by tsA mutants of simian virus 40: temperature sensitivity of the transformed phenotype and retransformation by wild-type virus
,
Jt. Virol
.
25
,
860
870
.
Brugge
,
J. S
&
Butel
,
J. S.
(
1975
).
Role of simian virus 40 gene A function in maintenance of transformation
.
J. Virol
.
15
,
619
635
.
Carter
,
T. C.
,
Lyon
,
M. F.
&
Phillips
,
R. J. S.
(
1955
).
Gene-tagged chromosome translocations in eleven stocks of mice
.
J. Genet
.
53
,
154
166
.
Chasin
,
L.
(
1973
).
The effect of ploidy on chemical mutagenesis in cultured Chinese hamster cells
.
J. cell. Physiol
.
82
,
299
308
.
Cooper
,
H. L.
&
Black
,
P. H.
(
1963
).
Cytogenetic studies of hamster kidney cell cultures transformed by the simian vacuolating virus (SV40)
.
J. natn. Cancer Inst
.
30
,
1015
1043
.
Croce
,
C. M.
,
Aden
,
D.
&
Koprowski
,
H.
(
1975a
).
Somatic cell hybrids between mouse peritoneal macrophages and simian virus 40-transformed human cells. II. Presence of human chromosome 7 carrying simian virus 40 genome in cells of tumours induced by hybrid cells
.
Proc. natn. Acad. Sci. U.S.A
.
72
,
1397
1400
.
Croce
,
C. M.
,
Aden
,
D.
&
Koprowski
,
H.
(
1975b
).
Tumorigenicity of mouse-human diploid hybrids in nude mice
.
Science, N.Y
.
190
,
1200
1202
.
Eagle
,
H.
(
1959
).
Amino acid metabolism in mammalian cell cultures
.
Science, N.Y
.
130
,
432
437
.
Freedman
,
V. H.
&
Shin
,
S. I.
(
1974
).
Cellular tumorigenicity in nude mice: correlation with cell growth in semi-solid medium
.
Cell
3
,
355
359
.
Gallimore
,
P. H.
&
Richardson
,
C. R.
(
1973
).
An improved banding technique exemplified in the karyotype analysis of two strains of rat
.
Chromosoma
41
,
259
263
.
Harris
,
H.
&
Watkins
,
J. F.
(
1965
).
Hybrid cells derived from mouse and man: artificial heterokaryons of mammalian cells from different species
.
Nature, Lond
.
205
,
640
646
.
Harwood
,
L. M. J.
&
Gallimore
,
P. H.
(
1975
).
A study of the oncogenicity of adenovirus type 2 transformed rat cells
.
Int. J. Cancer
16
,
498
508
.
Jha
,
K. K.
&
Ozer
,
H. L.
(
1976
).
Expression of transformation in cell hybrids. I. Isolation and application of density inhibited Balb/3T3 cells deficient in hypoxanthine phosphoribosyltransferase and resistant to ouabain
.
Somat. Cell Genet
.
2
,
215
223
.
Jonasson
,
J
&
Harris
,
H.
(
1977
).
The analysis of malignancy by cell fusion. VIII. Evidence for the intervention of an extra-chromosomal element
.
J. Cell Sci
.
24
,
255
263
.
Jonasson
,
J.
,
Povey
,
S.
&
Harris
,
H.
(
1977
).
The analysis of malignancy by cell fusion. VII. Cytogenetic analysis of hybrids between malignant and diploid cells and of tumours derived from them
.
J. Cell Sci
.
24
,
217
254
.
Kao
,
F. T.
&
Puck
,
T. T.
(
1969
).
Genetics of somatic mammalian cells IX. Quantitation of mutagenesis by physical and chemical agents
.
J. cell. Physiol
.
74
,
245
248
.
Kit
,
S.
,
Kurimura
,
T.
&
Dubbs
,
D. R.
(
1969
).
Transplantable mouse tumor line induced by injection of SV4o-transformed mouse kidney cells
.
Int.J. Cancer
4
,
384
392
.
Koprowski
,
H.
&
Croce
,
C. M.
(
1977
).
Tumorigenicity of simian virus 40-transformed human cells and mouse-human hybrids in nude mice
.
Proc. natn. Acad. Sci. U.S.A
.
74
,
11421146
.
Lehman
,
J. M.
(
1974
).
Early chromosome changes in diploid Chinese hamster cells after infection with simian virus 40
.
Int.J. Cancer
13
,
164
172
.
Littlefield
,
J. W.
(
1964
).
Selection of hybrids from matings of fibroblasts in vitro and their presumed recombinants
.
Science, N.Y
.
145
,
709
710
.
Macpherson
,
I
&
Montagnier
,
L.
(
1964
).
Agar suspension culture for the selective assay of cells transformed by polyoma virus
.
Virology
23
,
291
294
.
Marshall
,
C. J.
&
Dave
,
H.
(
1978
).
Suppression of the transformed phenotype in somatic cell hybrids
.
J. Cell Sci
.
33
,
171
190
.
Martin
,
R. G.
&
Chou
,
J. Y.
(
1975
).
Simian virus 40 functions required for the establishment and maintenance of malignant transformation
.
J. Virol
.
15
,
599
612
.
Osborn
,
M
&
Weber
,
K.
(
1975
).
Simian virus 40 gene A function and maintenance of transformation
.
J. Virol
.
15
,
636
644
.
Pontecorvo
,
G.
(
1971
).
Induction of directional chromosome elimination in somatic cell hybrids
.
Nature, Lond
.
230
,
367
369
.
Pope
,
J. H.
&
Rowe
,
W. P.
(
1964
).
Detection of specific antigen in SV4O-transformed cells by immuno-fluorescence
.
J. exp. Med
.
120
,
121
128
.
Risser
,
R.
&
Pollack
,
R.
(
1974
).
A non-selective analysis of SV40 transformation of mouse 3T3 cells
.
Virology
59
,
477
489
.
Rothfels
,
K. H.
&
Siminovitch
,
L.
(
1958
).
An air-drying technique for flattening chromosomes in mammalian cells grown in vitro
.
Stain Technol
.
33
,
73
77
.
Sager
,
R.
&
Kovac
,
P. E.
(
1978
).
Genetic analysis of tumorigenesis. I. Expression of tumorforming ability in hamster hybrid cell lines
.
Somat. Cell Genet
.
4
,
375
392
.
Shin
,
S. I.
,
Freedman
,
V. H.
,
Risser
,
R.
&
Pollack
,
R.
(
1975
).
Tumorigenicity of virus-transformed cells in nude mice is correlated specifically with anchorage independent growth in vitro
.
Proc. natn. Acad. Sci. U.S.A
.
72
,
4435
4439
.
Shkolnik
,
T.
&
Sachs
,
L.
(
1978
).
Suppression of the in vivo malignancy and in vitro cell multiplication of myeloid leukemic cells by hybridization with normal macrophages
.
Expl Cell Res
.
113
,
197
204
.
Stanbridge
,
E. J.
(
1976
).
Suppression of malignancy in human cells
.
Nature, Lond
.
260
,
17
20
.
Stiles
,
C. D.
,
Desmond
,
W.
,
Chuman
,
L. M.
,
Sato
,
G.
&
Saier
,
M. H.
(
1976
).
Growth control of heterologous tissue culture cells in the congenitally athymie nude mouse
.
Cancer Res
.
36
,
I353
I36O
.
Stiles
,
C. D.
,
Desmond
,
W.
,
Sato
,
G.
&
Saier
,
M. H.
(
1975
).
Failure of human cells transformed by simian virus 40 to form tumours in athymie nude mice
.
Proc. natn. Acad. Sci. U.S.A
.
72
,
4971
4975
.
Suarez
,
H. G.
,
Cassingena
,
R.
,
Estrade
,
S.
,
Wicker
,
R.
,
Lavialle
,
C.
&
Lazar
,
P.
(
1974
).
Properties of SV40 rescued from actinomycin D sensitive and actinomycin D resistant transformed hamster cells. I. Lytic infection
.
Arch. ges. Virusforsch
.
46
,
93
104
.
Tegtmeyer
,
P.
(
1975
).
Function of simian virus 40 gene A in transforming infection
.
J. Virol
.
15
,
613
618
.
Tevethia
,
S. S.
&
Mcmillan
,
V. L.
(
1974
).
Acquisition of malignant properties by SV40-transformed mouse cells: relationship to type-C viral antigen expression
.
Intervirol
.
3
,
269
276
.
Todaro
,
G. J.
&
Green
,
H.
(
1966
).
High frequency of SV40 transformation of mouse cell line 3T3
.
J. Virol
.
28
,
756
763
.
Wesslen
,
T.
(
1970
).
SV40 tumorigenesis in mouse
.
Acta path, microbiol. scand. B
B
,
479
487
.
Wiener
,
F.
,
Klein
,
G.
&
Harris
,
H.
(
1971
).
The analysis of malignancy by cell fusion. HI. Hybrids between diploid fibroblasts and other tumour cells
.
J. Cell Sci
.
8
,
681
692
.
Wiener
,
F.
,
Klein
,
G.
&
Harris
,
H.
(
1974
).
The analysis of malignancy by cell fusion. VI. Hybrids between different tumour cells
.
J. Cell Sci
.
16
,
189
198
.
Zuna
,
R. E.
&
Lehman
,
J. M.
(
1977
).
Heterogeneity of karyotype and growth potential in simian virus 40-transtormed Chinese hamster cell clones
.
J. natn. Cancer Inst
.
58
,
1463
1472
.