Intervirology 9 : 28-38 (1978)

Transformation of Human Cells by Temperature-Sensitive Mutants of Simian Virus 40 C hristopher A. Lomax1, Edward Bradley2, J oseph W eber and P ierre Bourgaux Département de Microbiologie, Faculté de Médecine, Université de Sherbrooke, Sherbrooke, Qué.

Key Words. SV40 temperature-sensitive mutants • Transformation of human cells • Temperature-sensitive mutants, cell transformation • Mutants of SV40, temperaturesensitive ■Papovavirus SV40 Summary. Conditions necessary for the establishment and maintenance of transfor­ mation of human cells by wild type and temperature-sensitive mutants o f SV40 were examined. For both early and late mutants, the frequency of transformation was found to be up to 5-fold higher, and virus yield 100-fold lower, at 39 than at 33 . No such effect was observed with the wild type virus under the same conditions. This observation is apparently at variance with previously published work, but may be explained by the semipermissive nature of the cells that we used. Increasing the temperature to 40.5 caused cells transformed by the early mutant, tsA30, to lose T-antigcn as detectable by staining, and also to lose the abilit to grow to high density, while it produced no effect on cells transformed by wild type virus.

Although the phenomenon has been recognized for many years, the mech­ anism by which simian virus 40 (SV40) effects the neoplastic transformation of cultured mammalian cells remains obscure. The isolation of mutant viruses, temperature-sensitive for viral replication [1], and their subsequent genetic and functional analysis [2-4] has greatly enhanced the knowledge of the SV40 functions involved in the transformation event. Studies with these 1 Present address: 3 Bailey Fold, Westhaughton Bolton BL5 3H4 (England). 2 Present address: Institut du Cancer de Montréal, Hôpital Notre-Dame, Montréal,

Received: October 4, 1976; revised: March 1, 1977.

Downloaded by: Stockholm University Library 130.237.165.40 - 1/19/2019 10:04:41 PM

Québec (Canada). Address inquiries to: Dr. P ierre Bourgaux, Département de Microbiologie, Faculté de Médecine, Université de Sherbrooke, Sherbrooke, Québec J!H 5N4 (Canada)

L omax /B radley /W eber/B ourgaux

29

mutants, in a variety of mammalian cell lines, have indicated the necessity for a functional ‘early’ gene A in order to establish and maintain the trans­ formed slate [2,5-8], A converse approach to the same problem was taken by Y amaguchi and K uchino [9] who selected mutants of SV40 on the basis of their inability to transform rat cells at a restrictive temperature. Examination of other phenotypic properties of these mutants suggested that they are probably defective in gene A, but their map position still remains to be de­ termined. A brahams et at. [10] have been able to transform cells with SV40 DNA fragments as small as 59% of the entire genome, provided the ‘early’ region was present and intact. These studies emphasize that the early region of SV40 DNA, and particularly gene A, is the transforming region of the viral genome. There is mounting evidence that the product of the SV40 gene A is Tantigen [5,7,11]. Since a functional gene A product is necessary for the initia­ tion, but not the completion, of a round of viral DNA synthesis [2], specula­ tions have been raised concerning the role T-antigen may play in the trans­ formed cell. A model has been advanced in which T-antigen would initiate DNA synthesis at an aberrant host site, leading to a complete round of host DNA replication and to the other phenotypic changes characteristic of trans­ formed cells [12], We have studied the transformation of human skin cells by wild type and temperature-sensitive mutants of SV40. In this system, semi-permissive for viral replication, we find an increased frequency of transformation at a tem­ perature restrictive for viral replication, compared to that at the non-restrictive temperature.

Cells and culture conditions. Skin fragments from healthy donors were cut up and trypsinized. Cultures were obtained by plating the resulting cell suspensions in plastic Petri dishes or flasks (Falcon Plastics, Oxnard, Calif.) containing Dulbecco's modified Eagle’s medium (DMEM) supplemented with 20% fetal calf serum (Flow Labs.. Inc., Rockville, Md.). These cell cultures were incubated at 37 in a humidified atmosphere containing 5% CO-2. Cultures referred to later as M l, M2 and M5 originated from male Caucasian donors of ages 12, 28, and 24 years, respectively. Cell cultures were tested routinely for mycoplasma contamination by autoradiography following tritiated thymidine incorpo­ ration [13] and were always found to be negative. Transformation experiments were generally performed when the cells were at passage 5 (10-15 generations from the initiation o f the culture). The life spans o f cultures M l, M2 and M5 were of approximately 50 generations, while transformation by SV40 increased

Downloaded by: Stockholm University Library 130.237.165.40 - 1/19/2019 10:04:41 PM

Materials and Methods

30

Lomax /B radley /W eber/B ourgaux

this figure by about 20 generations. No cell line was isolated from either normal or trans­ formed cultures. Viruses. Wild type SV40 and the mutants tsA30 and tsBI were kindly provided by P .T egtmeyer. They were grown on CV-I cells from an initial infection of 0.01 PFU/cell, at 37 for the wild type and 33 for the mutants. After I week at 37 or 2 weeks at 33 , virus was harvested by three freeze-thaw cycles followed by centrifugation at 3,000 rpm for 10 min. This crude stock was stored at -9 0 and used directly without further purification. Virus was plaque-titrated on confluent sheets of CV-1 cells. Fresh medium was added at 5-day intervals, and after 14 days at 37 or 21 days at 33 the cells were stained by adding 5 ml of medium containing 0.01% neutral red. Plaques were counted after a further 24 h incubation. Most of the experiments described were performed with stocks of the following titers: 2 x 10s PFU/ml for wild type virus, 1.2 x 107 PFU/ml for tsA30 mutant, and 7 x 10" PFU/ml for tsBI mutant. Transformation experiments. Confluent cells grown in a 6-cm dish were infected with 0.2 ml of virus suspension of known titer. Adsorption proceeded at 33 for 3 h after which the cells were washed free of unattached virus, detached with a trypsin-versene mixture and replated in 6-cm dishes at known density, usually between 10i| and 105 cells per dish. The medium used was DMEM supplemented with 10% fetal bovine serum and contained 0.25% horse anti-SV40 antiserum (Flow Labs.; titer 640 vs. 100 TCD 50). The dishes were incubated without subculturing, but with a weekly change o f medium and antiserum until transformed foci were visible (4-8 weeks, depending on the temperature). As a permissive temperature, 33 was selected. Even though it was known to be only partially restrictive for mutant growth in primate cells [2], 39 had to be used as restrictive temperature since the human cells employed were found to be unable to multiply for extended periods of time at a higher temperature. Transformed foci were counted without staining, and a representative selection was picked and subcultured. Growth studies. Cells were plated at 10s viable cells per 35-mm dish in DMEM plus 10% fetal bovine serum, and incubated at 33, 39 and 40:. At this density, all cells studied appeared to plate with approximately the same efficiency. Duplicate samples were washed, trypsinized and counted every 2 days. On the day of sampling, the growth medium was changed for the remaining cultures. T-antigen staining. Cells to be assayed were grown on glass coverslips ( 12 mm diameter) in 60-mm dishes until they were 30-60% confluent. The coverslips were then removed, dipped twice into phosphate-buffered saline (PBS) to wash the cells, and allowed to dry. Cells were fixed in dry acetone at -2 0 for 10 min. The immunofluorescence assay for T-antigen was performed with commercially prepared fluorescein isothiocyanate(FlTC)conjugated anti-T ascitic fluid (Flow Labs.), using a single-step version o f the standard procedure [14].

Transformation experiments. The effects of temperature on the frequencies of transformation of several strains of human cells, by wild type and tempera­ ture-sensitive mutants of SV40, are shown in table 1. The ratio of the fre-

Downloaded by: Stockholm University Library 130.237.165.40 - 1/19/2019 10:04:41 PM

Results

31

Transformation of Human Cells by SV40

Table I. Effect of temperature on the frequency of transformation by wild type (WT) SV40 virus and by two mutants Cells

Virus

M Ol1

Transformation frequency2 33°

MI M5 M2

WT tsA30 WT tsA30 WT tsBl

2

2.0 0.6

26 25

1.4 4.2

40

Ratio

39'

2.0

1.0

2.7 1.4

4.5 3.0

10

1.0

12.6 1.6

55

3.0

7.9

1.0

1.6 2.6

quencies of transformation by wild type at 39 and 33° is close to unity, that is to say, the wild type virus transforms equally well at both temperatures. However, the frequency of transformation by the temperature-sensitive mutant tsA30 is increased at 39° by a factor of between 3 and 5 over that seen at 33°. The frequency of transformation by tsBl is similarly increased by a factor of 2.6 at the restrictive temperature. It is also apparent that, even at 33°, the transformation frequency is about 4 times higher for tsA30 than for wild type virus, at comparable MOI. As transformation was apparently affected by the temperature of incuba­ tion, it was of interest to determine the length of the critical period. This was determined by measuring the transformation frequency of cultures incubated at 33° for increasing periods prior to shifting to 39°. As shown in table II, there is an inverse relationship between the transformation frequency by tsA30 and the time of incubation at 33° preceding that at 39°, with the trans­ formation frequency being reduced 25% by as little as 1-2 days incubation at 33°. The maximal length of time during which the transformation frequency may be affected by the temperature of incubation is of the order of 8 days, for the transformation frequency after this period differs little from that observed after continual incubation at 33°. In contrast, transformation by wild type virus is insensitive to temperature shifts. As already noted in table I, the trans­ formation frequency of cells maintained continuously at 33° is higher for the mutant than for wild type virus.

Downloaded by: Stockholm University Library 130.237.165.40 - 1/19/2019 10:04:41 PM

1 Plaque-forming units/cell. 2 Number of transformed colonies per 10' cells plated. The transformation frequency is the mean of a group of at least 8 dishes.

32

L omax /B radley /W eber/B ourgaux

Table II. Effect of a temperature shift on transformation frequency by wild type (WT) and mutant SV40 virus Conditions

Transformation frequency1 tsA30

WT

0

12.6

I

8.7 9.6 6.7 4.7 4.2

1.4 1.7

Days at 33 before shifting to 39

2 4

8 Cells at 33 continually

1.2 1.7 1.3 1.4

Temperature sensitivity o f tsA30 in human cells. Various reports have indi­ cated that group A mutants fail to transform cultured cells at the restrictive temperature [2,6,8], The temperature selected as restrictive varied among these publications, but was in most cases higher than 39°. In view of the published evidence that the replication of tsA30 in primate cells is only par­ tially suppressed at 39° [2], it could have been argued that the results obtained with this mutant were due to a lack of expression of the mutation in human cells at that temperature. We therefore measured viral yield during productive infection of human cells at 33 and 39° (table III). We found little difference in yield between the two temperatures for wild type virus, while we observed a 96-fold reduction for tsA30 at the restrictive temperature. Since these results are from single time points, rather than from a growth curve, the difference is a minimum estimate and could be even greater. These results suggest that, in human cells, the leakiness of tsA30 is not sufficient to account for a normal ability of transformation, much less an increase in this ability. It is also strik­ ing that the infectious titer reached by tsA30 in these human cells was, even at 33°, much lower than that reached by wild type virus. Properties o f transformed cells: growth kinetics. Individual foci, trans­ formed by wild type and tsA30 virus, were picked and subcullured. Their growth kinetics were studied at 33° and at two distinct ‘restrictive’ tempera­ tures, 39 and 40° (fig. 1).

Downloaded by: Stockholm University Library 130.237.165.40 - 1/19/2019 10:04:41 PM

1 Number of transformed colonies per IO'* cells plated, detected at 4-5 weeks (shifted cultures) or 8 weeks (control cultures) after infection. Human M5 cells were infected with virus at a multiplicity of 25 PFU/cell. The transformation frequency represents the mean of a group of at least 8 dishes.

Transformation o f Human Cells by SV40

33

Table III. Effect o f temperature on growth of SV40 in human M5 cells1 Virus

WT tsA30

Yield, PFU/culture 33°

39°

2 .6 x 10» 9.6 x 103

1.6 x 105 1.0 x 102

Ratio 33°/39°

1.6 96

1 Confluent monolayers of M5 cells (about 5 x 105 cells) were infected with wild type (WT) or tsA30 virus at an MOI o f 20 PFU/cell for 3 h at 33°. The cultures were washed twice with medium containing anti-SV40 antiserum and incubated in DMEM + 10% fetal bovine serum at 33 or 39° for 16 days. The medium and cells were harvested by three freeze-thaw cycles, and the yield was titered on CV-1 cells at 37° for WT and 33° for tsA30. Lysis of cells, similarly infected, after 24 h incubation produced no infectious virus.

IN C U B A T IO N

T IM E (O A T S )

Untransformed cells were found to grow to low densities at all three tem­ peratures, while cells transformed by wild type virus grew to densities 3- to 5-fold greater. Similarly, cells transformed by tsA30 grew to densities 3- to 4-fold greater than untransformed cells at both 33 and 39°. At 40°, however,

Downloaded by: Stockholm University Library 130.237.165.40 - 1/19/2019 10:04:41 PM

Fig. I. Cells were plated at 105 cells per 35-mm dish in DMEM with 10% fetal bovine serum. At 48-hour intervals duplicate dishes were washed, trypsinized and counted. A = 33°; 0 = 39°; □ = 40°.

34

L omax/B radley /W eber/B ourgaux

these latter cells failed to grow to a density higher than that of untransformed cells. Increasing the temperature to 40° therefore affects at least one para­ meter of the transformed phenotype, but only in those transformed cells carrying the tsA30 mutant lesion. Temperature sensitivity o f T-antigen stainability. Every population of cells isolated, at either 33 or 39°, on the basis of density-independence of growth, was T-antigen-positive in the immunofluorescence test, although some populations were less than 100% positive prior to cloning. The distribu­ tion of the staining was clearly distinct from that observed by us in SV40transformed mouse or hamster cells; a fluorescent region following the inner boundary of the nuclear membrane was characteristic of the cells of all human transformed strains examined. It has been reported that T-antigen immunofluorescence in cells trans­ formed by group A mutants of SV40 is reduced at the restrictive temperature [5]. To investigate this phenomenon using our system, transformed cells were grown at 33 or 39° for 1 week and then shifted, without subcultivation, to a higher temperature, as indicated in table IV. The T-antigen activity could not be detected in any tsA30 transformants after 6 h of a shift from 33 to 40.5°, or after 24 h of a shift from 39 to 40.5°. There was no such effect after a shift from 33 to 39°, although virus yield was reduced 100-fold at this tem­ perature (table III). In wild type virus-transformed cells, T-antigen stainability remained unchanged. Mutant-transformed cells could recover stainability, along with the characteristic visual appearance of T-antigen, within 18 h after a subsequent shift-down to either 39 or 33°.

In the experiments described here, transformation by SV40 was performed using human cells which had undergone a limited number of passages in vitro. Such cells were found to be semi-permissive for wild type virus replication, similar virus yields being obtained at either 33 or 39°. In contrast, the replica­ tion of early mutant tsA30 was clearly temperature-sensitive in the same cells. As it is known that the expression of temperature-sensitive mutations in the SV40 A gene may vary with the origin of the host cell [6], this finding is of great significance with respect to the observations made regarding transforma­ tion (see below). Furthermore, the infectious titers reached at 33° were much lower for mutant tsA30 than for wild type virus, even when similar input multiplicities had been used. This situation may or may not indicate that the

Downloaded by: Stockholm University Library 130.237.165.40 - 1/19/2019 10:04:41 PM

Discussion

35

Transformation of Human Cells by SV40 Table IV. Effect o f temperature shift onT-antigen activity1 Initial temperature of incubation °C

Temperature shift regimen

33 39 33 33 39 33 33

no shift no shift 6 h at 39° 6 h2 at 40.5° 24 h2 at 40.5° 6 h at 40.5°, 18 h at 33° 6 h at 40.5°, 18 h at 39°

T-antigen in cells as shown by immuno­ fluorescence staining WT

tsA30

yes yes yes yes yes yes yes

yes yes yes no no yes yes

A30 mutation was already partially expressed in our cells at 33°, a temperature often considered as non-restrictive. It is perhaps worth keeping this suggestion in mind when considering the efficiency of transformation by tsA30. The latter is not only higher at 39° than at 33°, but also higher than that registered for wild type virus at both temperatures. Such findings regarding transformation by temperature-sensi­ tive mutants of SV40, although without precedent, do not necessarily contra­ dict what was published recently on the subject [5-8]. Indeed, one should not lose sight of the fact that the cells used here were semi-permissive. Thus, in our experiments, the mutated A function may have been inactivated, espe­ cially at 39°, to such an extent that virus multiplication - and thus killing of susceptible cells - was considerably lower for tsA30 than for wild type virus. Conceivably, a reduction in the potential of the viral genome for cell killing could represent additional opportunities for the expression of its transforming potential. Such a situation could be accounted for strictly in terms of functions of the A gene product. While many functional molecules of that product would be expected to be required for the numerous rounds of replication of the viral genome, a few molecules of it might suffice to establish and main­

Downloaded by: Stockholm University Library 130.237.165.40 - 1/19/2019 10:04:41 PM

1 Cells transformed by either wild type (WT) or tsA30 were plated on coverslips and incubated for 1 week at the indicated initial temperature, at which time they were 50-80% confluent. Upon completion of the temperature shift sequence, cells were stained with anti-T ascitic fluid as described in ‘Materials and Methods’. Three independent cloned isolates of «-transformed cells were tested, each giving the same results. 2 Minimum incubation time required for disappearance o f T-antigen activity.

L omax /B radley /W eber/B ourgaux

tain transformation. Yet, with the late mutant tsBl, transformation frequen­ cies were also higher at 39° than at 33°. Thus, the unusual behavior of these mutants in transformation is likely to be due to a nonspecific restriction in their ability to replicate and thereby kill susceptible cells, rather than to a change in the expression of one specific function. The explanation we offer would imply that in the human cultures we have studied, most of the cells susceptible to transformation belong to a subpopu­ lation in which the virus is also able to replicate. Under conditions where virus replication would not be restricted, such cells should undergo a lytic cycle which would prevent them from giving rise to transformed colonies. It is only because SV40 mutants were characterized using either permissive cells or nonpermissive cells that this concept may appear new in this instance, while being an old concept in the papovavirus field. An infectious virus genome cannot be rescued from permissive mouse cells, or semi-permissive hamster cells, transformed by wild type polyoma virus [15,16], Yet, mouse cells transformed by early temperature-sensitive mutants of polyoma virus and maintained at the restrictive temperature, yield virus after a simple transfer to the non-restrictive temperature [17], while similarly transformed hamster cells release virus upon fusion with permissive cells [18]. It has already been proposed that, unless conditionally defective viral genomes are used under restrictive conditions, the infection of permissive or semi-permissive cells by polyoma virus generates transformed cell lines that contain unconditionally defective, and therefore non-rescuable, viral genomes [18]. It is thus not unlikely that the interaction between SV40 and the human cells that we have described is analogous to that known to take place between polyoma virus and hamster cells. Our characterization of the cells transformed by the tsA30 mutant extends earlier studies indicating that cells transformed by such mutants exhibit a temperature-sensitive phenotype [5-8]. Temperature sensitivity of T-antigen in cells transformed by A group mutants of SV40 is a controversial matter, however. Our observations indicating that T-antigen actually is temperaturesensitive in such cells, while in agreement with the results of O sborn and W eber [5], are at variance with those of T egtmeyer [6] and Brugge and Butel [7], Careful examination of the conditions used by O sborn and W eber [5] and by ourselves does suggest that, depending on the temperature shift protocol selected, the result of the immunofluorescence test could vary considerably. Unfortunately, the investigators who have reported T-antigen stainability not to be affected by the temperature of incubation give relatively little detail about the protocol they have used [6,7]. Whatever the reason for these dis­

Downloaded by: Stockholm University Library 130.237.165.40 - 1/19/2019 10:04:41 PM

36

Transformation of Human Cells by SV40

37

crepancies, there is little doubt that the T-antigen of A group mutants dis­ plays altered properties at restrictive temperatures, as indicated by several observations on either productively infected [11] or transformed [19] cells. It is, therefore, interesting that in our experiments, T-antigen immunofluor­ escence correlates with another parameter of the transformed phenotype, namely the ability of the cells to grow to high density. Finally, to put our results in a broader context, we should like to point out that they provide another example of a situation where, as a result of mutation, a viral genome has been generated which has reduced infectivity, but also increased oncogenic potential with respect to certain host cells. A cknowledgments This work was supported by the National Cancer Institute of Canada. J. W. is a Scholar of the NCIC. E.B. held a post-doctoral fellowship from the Cancer Research Society, Montreal. We thank Peter T egtmeyer for the generous gift of wild type and temperaturesensitive mutants o f SV40, and D anielle Bourgaux and J ean-P aul T hirion for many stimulating discussions.

1 T egtmeyer, P .; D ohan , C., jr., and R eznikoff, C .: Inactivating and mutagenic effects of nitrosoguanidine on simian virus 40. Proc. natn. Acad. Sci. USA

Transformation of human cells by temperature-sensitive mutants of simian virus-40.

Intervirology 9 : 28-38 (1978) Transformation of Human Cells by Temperature-Sensitive Mutants of Simian Virus 40 C hristopher A. Lomax1, Edward Bradl...
1MB Sizes 0 Downloads 0 Views