Changes in Deoxyribonucleic Acid Synthesis ... - Journal of Virology

Vol. 6, No. 6 Printed in U.S.A.

JOURNAL OF VIROLOGY, Dec. 1970, p. 738-749 Copyright © 1970 American Society for Microbiology

Changes in Deoxyribonucleic Acid Synthesis Regulation in Chinese Hamster Cells Infected with Simian Virus 401 JOHN M. LEHMAN'

VITTORIO DEFENDI3

AND

The Wistar Institute of Anatomy and Biology, Philadelphia, Pennisylvanlia 19104

Received for publication 6 July 1970

Infection of primary or secondary cultures of Chinese hamster embryo cells with simian virus 40 at a multiplicity of 20 to 50 induced synthesis of the virus-specific intranuclear T antigen in 80 to 90% of the cells within 48 to 72 hr. In the infected cultures, 30 to 50%O more cells were recruited into deoxyribonucleic acid (DNA) synthesis than in the controls, whether or not the cultures were confluent. The newly synthesized DNA was mostly cellular, since little virus was produced (as shown by various techniques: immunofluorescence for viral antigen, virus growth curves, and isolation of viral DNA from infected cultures). Transformed cells could be detected a few weeks after infection and produced tumors when inoculated into irradiated animals. Chromosomal changes were observed soon after infection (24 hr). Initially, there was a marked increase in the proportion of polyploid cells (8 to 14%), most of which were chromosomally normal. In a few weeks, a large majority of the infected population was polyploid (30 to 50%). Thus, the polyploid cells have the ability to proliferate. Evidence is presented to suggest that polyploid cells arise by stimulation of cells in the G1, G2, or S phases to undergo two or more successive periods of DNA synthesis without an intervening mitosis. With a subsequent loss or redistribution of chromosomal material, this may lead eventually to a biologically transformed cell; thus, it is suggested that the initial event(s) relevant to transformation occurs at the level of control of cellular DNA synthesis. in preparation), or contact inhibition (13, 35). Thus, viral DNA has rescue capabilities, possibly by activation of initiation sites that have been shut off or impaired. Whether initiation is produced by direct interaction of the viral DNA with the cellular DNA or by a virusdirected diffusible product is not yet known; the recent finding that mitochondrial DNA is also stimulated to replicate in SV40-infected cells concomitant to nuclear DNA (A. Levine, personal communication) favors the latter hypothe-

Small oncogenic deoxyribonucleic acid (DNA) viruses, such as simian virus 40 (SV40) or polyoma virus, have the unique property of inducing cellular DNA synthesis in infected permissive or nonpermissive cells [for review, see Winocour (34)]. Cellular DNA synthesis can be induced in cells that are in a G, or Go phase, such as human cells at the final step in their life span in vitro (28), myoblasts (12, 37), or macrophages (Mauel and Defendi, Exp. Cell Res., in press)-that is, cells whose state of differentiation is such that DNA replication no longer occurs normally. DNA synthesis can also be restored by these viruses in cells that have been blocked in a particular period of the cell cycle by drugs (3), elevated temperature (25), X ray (13; H. Frey,

sis.

1 Submitted by John M. Lehman in partial fulfillment of the requirements for the Ph.D. degree to the Department of Pathology, University of Pennsylvania, Philadelphia. 2 Present address: Department of Pathology, University of Colorado Medical School, Denver, Colo. 80220. 3 Department of Pathology, School of Medicine, University of Pennsylvania. Recipient of Faculty Research Award PRA-47 from the American Cancer Society.

One of the central problems in understanding the oncogenic capacity of these viruses is the relevance of the cellular DNA induction to the phenomenon of cell transformation. In this paper we present evidence suggesting that SV40 infection of Chinese hamster (ChH) cells causes an alteration of the normal pattern of cellular DNA synthesis, as a result of which polyploid cells are formed. From this population, transformed cells may originate.

738

VOL. 6, 1970

CHANGES IN DNA SYNTHESIS IN SV40-INFECTED CELLS

MATERIALS AND METHODS Virus. The SV40 (strain RH-911) pool, prepared in CV-1 cells (18) at a multiplicity of infection (MOI) of 0.01 plaque-forming unit (PFU) per cell, titered at 3 X 10' PFU/ml in CV-1 cells and was negative for mycoplasma. The virus was assayed by the plaque method on CV-1 cells (6). The SV40 was inactivated by ultraviolet (UV) light as a thin film in an uncovered 100-mm petri dish at a distance of 22 cm from a Westinghouse germicidal lamp. Cell cultures. ChH cultures were prepared by trypsinization of 19- to 20-day-old embryos (10). They were maintained in 2X Eagle's basal medium and 1 X Earle's balanced salt solution, supplemented with 5% fetal calf serum and antibiotics (100 units of penicillin per ml and 50 jug of streptomycin per ml). In most experiments, primary or secondary cultures were used. The cultures were split twice weekly at a ratio of 1:2. One cell line (97SV cl 21) was derived from a culture of ChH cells transformed by SV40. Growth curve of SV40. Confluent monolayers of primary ChH cells in 60-mm petri dishes were infected with 0.1 ml of SV40. After adsorption for 1 hr at 37 C, the cells were washed three times with phosphate-buffered saline (PBS), and 4 ml of medium was added. Samples were collected at 12-hr intervals by removing the supernatant, scraping the cells off the plastic with a rubber policeman, and resuspending the cells in 2 ml of the previously removed supernatant. The samples were frozen and thawed three times, and the virus was assayed. nmmunofluorescence techniques. Cover-slip cultures were washed in PBS, fixed, and stained with a fluorescein isothiocyanate-conjugated hamster anti-SV40 T-antigen serum (Flow Laboratories, Rockville, Md.) to demonstrate T antigen, as previously described (14), or with rabbit anti-SV40 serum followed by fluorescein isothiocyanate anti-rabbit baboon gamma globulin to demonstrate capsid antigen. The percentage of T antigen-positive cells was determined from counts of 500 cells; that of V antigen-positive cells, by counting fluorescent cells in the whole cover slip containing approximately 105 cells. Extraction of SV40 DNA from infected cells. Confluent ChH or CV-1 cells in 120-mm petri dishes were infected with 40 PFU of SV40/cell; after a 1.5-hr adsorption period, 10 ml of medium with 0.5 ,uCi of 3H-thymidine per ml (specific activity, 6.0 Ci/mmole; Schwarz BioResearch, Inc., Orangeburg, N.Y.) was added. The cultures were harvested at 72 hr after infection. One hour before harvesting, the medium was removed, and fresh medium was added to exhaust the intracellular pool of 3H-thymidine. The extraction procedure for viral DNA described by Hirt (16) was followed. The sodium dodecyl sulfate (SDS) lysate was mixed with NaCl to a final concentration of 1 M NaCl and stored at 4 C overnight. The solution was centrifuged in an SW 40 rotor, model L Spinco centrifuge, at 100,000 X g for 30 min. For a more critical detection of the presence of SV40 DNA component I (1, 9), the supernatant from

739

the Hirt extraction was added to CsCl (1.566 g/ml) in a solution of 0.01 M tris(hydroxymethyl)aminomethane (Tris), pH 8.0, 0.01 M ethylenediaminetetraacetate (EDTA), and 100 yg of ethidium bromide per ml. It was centrifuged at 37,500 rev/min for 48 hr in an SW 39 rotor at 20 C. Drops were collected from the bottom of the tube on ifiter-paper discs and dried; radioactivity was measured in a Beckman -200 scintillation counter. 32P-SV40 DNA component I was used as a marker. The proportion of viral DNA (component I) in the total newly synthesized DNA was calculated from the 3H-DNA in the supematant and pellet fractions. Transformation assay. Two overlay media were used for the selection of transformants from the infected population: 1.2% methyl cellulose (Methylcel, Dow type MC, 4,000 cps) (31) and 0.3% Difco Noble agar (20) in growth medium containing 20% fetal calf serum. For another detection method, the cells were plated at a low dilution (100 to 500 per 60-mm petri dish) and allowed to grow at 37 C for 7 days. After fixation and staining with Giemsa, colony morphology and

cloning efficiency were determined. Transplantation of ChH-transformed cells. Cells from the SV-40-transformed cell line, 97SV cl 21, were inoculated into both normal and irradiated (400 r from a cesium-137 gamma source) 2- to 3-monthold Chinese hamsters. The tissues from autopsy were fixed in Tellyesniczky's fluid, sectioned, and stained with hematoxylin and eosin. Cytogenetic analysis. Metaphase chromosomes were prepared by pretreatment of cultures for 4 to 6 hr with Colcemid (Ciba, Summit, N.J.) at a concentration of 0.05 lAg/ml. The cells were harvested as described previously (19), stained with Giemsa for 5 min, and mounted in Permount. The karyotype arrangement of Hsu and Zenzes (17) was followed. Polyploid cells were scored by counting 500 or more metaphases. DNA synthesis in ChH cells infected with SV40. Secondary ChH cells from 1-week confluent cultures were seeded into petri dishes containing cover slips (7 X 105 to 1 X 106 cells) and allowed to attach for 3 hr. The medium was removed and the cells were infected with 60 to 80 PFU of SV40/cell. After a 1-hr adsorption period, the medium was returned to the cultures, and 3H-thymidine was added at a concentration of 0.2 ,Ci/ml for continuous labeling or 2 MCi/ml for pulse labeling. At various intervals, two cover slips were removed from infected and mock-infected cultures and stained for T antigen. After the percentage of T antigenpositive cells was determined, the cover slips were removed from the slides, washed in PBS, and fixed in Carnoy's solution. They were mounted face up on slides, dipped into Kodak NTB-3 liquid emulsion at 42 C, and exposed for 5 to 7 days prior to developing. The slides were stained with Toluidine Blue at pH 3.5. The percentage of labeled nuclei was determined on the basis of 1,000 cells counted. Cytophotometry. The relative DNA content in single cells was estimated by optical density measurements within CYDAC (Cytophotometric Data Con-

740

LEHMAN AND DEFENDI

version), described by B. H. Mayall and M. L. Mendelsohn (22a). Cells were grown on cover slips, washed in PBS, and fixed in methanol for 10 min. They were stained by a modified Feulgen procedure described by Deitch (11) and Mayall (22). The optical density at a wavelength of 566 nm, as measured with CYDAC, was plotted on graph paper to give the distribution of relative DNA content of the population sampled. Cells from the same population could be classified as 2n (GI phase) or 4n (G2 phase) by measuring cells in each phase of the mitotic cycle. The DNA content of cells in the S phase was between that of cells in the G, and G2 phases. Approximately 200 cells were measured for each interval studied. Analysis of the cell cycle. The analysis of the ChH cell cycle was determined by the labeled mitoses method of Quastler and Sherman (26). Cells were seeded in 60-mm petri dishes containing 12 small round 18-mm cover slips. The cells were allowed to attach for 24 hr, and the medium was removed and saved (conditioned medium). A 2-ml amount of conditioned medium containing 3H-thymidine (2 MCi/ml) was added to the cultures and removed after 15 min. The cultures were washed with warm PBS, and equal parts of prewarmed conditioned medium and fresh medium containing cold thymidine (5 jug/ml) were added. All operations were performed at 37 C. Two cover slips were removed at hourly intervals, washed in PBS, and fixed for autoradiography. Between 50 and 150 metaphases were scored for the ratio of labeled to unlabeled mitoses at each time interval. Time-lapse cinemicrography of SV40-infected ChH cells. Secondary ChH cells were infected with 50 PFU of SV40/cell for 2 hr and were then transferred into 30-mm petri dishes containing 18-mm round cover slips. Mitoses were recorded cinemicrographically on 16-mm film at 50 frames per hr through an inverted microscope with a lOX objective. The cultures were maintained in a 37 C incubator in the presence of 5% carbon dioxide in air. The timelapse cinemicrography apparatus developed by LaRoy N. Castor of the Institute for Cancer Research, Philadelphia, Pa. (7), was used. After 48 to 72 hr, the cultures were fixed and the cells were prepared for cytophotometric determination (CYDAC) of the DNA content of nuclei.

RESULTS Growth curve of SV40 in ChH cells. No cell lysis was noted after infection of ChH cells with various multiplicities of SV40 under conditions in which destruction of 5 to 10% of the cell population could have easily been detected. When ChH cells were infected at an MOI of 20, the amount of virus recovered decreased with time (Fig. 1). This is in marked contrast to permissive monkey cells in which at an MOI of 3 to 10 there is destruction of the cell sheet in 5 days and an increase of at least two logs of virus at 3 to 5 days after infection (6).

J. V IROL.

-J

103

24

48

20 144 168 92 216 96 HOURS AFTER INFECTION

72

240

FIG. 1. Growth curves of SV40 in confluent ChH

cells. The curves represent two different experiments.

The fact that virus could still be recovered from the ChH cells 10 days after infection may indicate some virus replication. This amount could be higher than indicated in Fig. 1, if the virus is thermolabile at 37 C. This possibility was excluded by incubating virus at 37 C and titering samples at several intervals for 22 days. The initial titer was 107 PFU/ml, and after 22 days it was 3 x 106 PFU/ml. The amount of SV40 recovered from the ChH cultures may be accounted for by the persistence of the initial input virus, by a few cells producing a large amount of virus, or by many cells producing only few viral progeny. Induction of viral antigens. At an MOI of 50, T antigen was first detected 12 to 19 hr after infection (Fig. 2b), and the percentage of positive cells reached 80 to 95% at 72 hr. The T antigen was localized in the nucleus with no staining of the cytoplasm or the nucleolus. At an MOI below 30, the percentage of T antigen-positive cells progressively decreased, but, even at an MOI of 3, 50 to 60% of the cells were positive by 72 hr. One week after infection, there was no reduction in the percentage of T antigen-positive cells even after two subcultures. This is at variance with the interaction of SV40 with 3T3 cells (4). High susceptibility to infection was lost rapidly after several passages in vitro. Thus, when, after about 6 to 10 passages, cultures were infected even at an MOI of 50 or higher, only about 10 to 30% of the cells produced T antigen. For this reason, primary or secondary cells were used in all subsequent experiments.

VOL. 6, 1970


741

x

ZIow ow z _

25

0

z 0Cv I-J

DiJ

z

J

C-,v z 0

hi .4

Iz

hi

r10~ hi

.

2

4

6

8

10

12

:

0 0

O 0 00 0 00

00

.

14 16 I8 20 22 24 26 26 30 32 343 HOURS AFTER INFECTION

48

FIG. 2. DNA synthesis in control and SV40-infected ChH cells as measured by 3H-thymidine incorporation. (a) Mitotic index was determined by counting 1,000 cells (experiment 175). (b) Cumulative labeling experiment (no. 175). (c) One-hour pulse-labelinzg experimenzt (no. 181).

When infected cells were stained with antisera directed against the viral coat protein, no positive cells were detected at first, but by 7 days after infection a few of the cells (,1 %) were positive. Viral DNA replication. In these experiments, ChH cells were infected with SV40 (40 PFU/ cell), and 3H-thymidine (0.5 ,uCi/ml) was added after adsorption. At 72 hr, the culture was harvested by the Hirt procedure and the supernatant was centrifuged in CsCl in the presence of ethidium bromide. As shown in Table 1, the band of SV40 DNA represents only about 2 %

of the newly synthesized DNA. In CV-1, a permissive cell line, at 72 hr after infection and 24 hr after addition of 3H-thymidine, a large peak of viral DNA was detected which represents about 50 to 80% of the total DNA synthesized (Table 1). It can be concluded that the primary or secondary ChH cell-SV40 system is essentially nonpermissive, although a few cells do replicate the viral genome and produce infectious virus.

Transformation assay. To determine whether

quantitative assays of transformation could be applied to this cell system, secondary ChH cells were infected with SV40 (MOI of 20 to 50) and plated at very low densities (100 to 500 cells per 60-mm petri dish). Colonies were visible in 7 to 14 days, with the infected-cell population having twice the cloning efficiency of the control cells (3.4% versus 1.7%). However, it was not possible to score transformed colonies unequivocally, because of the variation in the control population, including colonies which showed the phenomenon of piling-up. Clearer evidence of morphological transformation was obtained by plating ChH cells that had been infected 1 to 3 months previously. These cultures had a high cloning efficiency (20 to 50%), whereas control cultures carried for the same length of time had a cloning efficiency of only 2 to 4%. All of the colonies from the infected cultures had the characteristic appearance of transformed cells, whereas the control colonies were normal-appearing, contact-inhibited, fibroblast-like cells. When primary or secondary ChH cells were

742

LEHMAN AND DEFENDI

TABLE 1. Percentage of viral DNA replicated in various cell systems8

J. VIROL.

cultures reached a plateau at about 22 to 24 hr: the infected cultures with 85 to 90% labeled nuclei and the control with 50%. Total labeling Percentage of total label in When a 1-hr pulse label experiment was done Cells time prior to harvest Cellular DNA SV40 DNA under similar conditions, the percentage of nuclei labeled in control and infected cultures from 0 hr to 18 hr was essentially the same-0 to 5% (Fig. 2c). In the control cultures at 19 to 36 hr, 5 to CV-l....... 24 25 .75 CV-1 ........ 24 46 12% of the cells were positive at any one sam54 ChH ........ 72 98 2 ple time; however, in the infected cultures, there was an increase in the percentage of labeled a Cultures of ChH and CV-1 cells were infected nuclei from 12 to 60 during this interval. This with SV40 (MOI, 10 to 30 PFU) and harvested at large increase of cells in DNA synthesis in the 72 hr by the Hirt procedure (see Materials and Methods). 3H-thymidine was added to the cultures infected population could be due to a rapid 24 or 72 hr prior to harvest. The percentages of recruitment of cells into S or to a prolonged S total label present in the supernatant (SV40 com- period. T antigen-positive cells appeared between ponent I DNA) and pellet (cellular DNA) indi- 12 and 18 hr, and the number continued to rise cate the relative amount of cellular or SV40 DNA for the duration of the experiment (Fig. 2b). replicated during the labeling period. Mitoses were seen in both control and infected cultures soon after plating, a result of the presinfected with 20 to 50 PFU/cell and plated in ence of G2 cells in the initial population. There agar or methyl cellulose, no colonies grew. was a burst of mitotic activity in both control However, colonies were obtained in the semi- and infected cultures at 18 to 20 hr, as usually solid medium when cells were plated several occurs in freshly explanted cell populations weeks after infection; no colonies appeared with (Fig. 2a). Mitoses were more numerous in the cells from noninfected cultures carried for the infected culture, but they appeared at about the same time as in the control cultures. Furthersame number of passages. Transplantation of ChH-transformed cells. more, in the infected cultures, the mitotic index Twenty animals, 1 to 4 months old, were inocu- rose to a level of between 1 and 3% and stayed lated subcutaneously and intraperitoneally with at that level; in the controls, the mitotic index 15 x 106 cells of 97SV cl 21 at the 62nd passage; returned to the initial value after 18 to 26 hr. To reduce the background proliferation of the no tumors had appeared after 8 months. Ten irradiated Chinese hamsters, 2 to 3 months old, control cultures (Fig. 2), the experiments were which received 50 X 106 cells of 97SV cl 21 at repeated with cultures 7 to 10 days after they the 67th passage, developed nodules at the site had reached confluency. Similar results, that is, of inoculation in 3 weeks. Histologically, the an increased proportion of cells in DNA syntumors were small-cell fibrosarcomas with thesis in the infected cultures, were observed. numerous mitoses and little collagen deposition. However, in the control population there was Infiltration into the neighboring muscle tissue still a considerable number of cells incorporatwas observed in the majority of the tumors ex- ing 3H-thymidine in pulse experiments (3 to amined. Attempts to transplant fragments from 8%), indicating that primary ChH cells do not the primary tumor into unirradiated hamsters respond too well to contact inhibition. were unsuccessful; however, in three of five The question whether the increased recruiting irradiated hamsters, tumors developed from of cells in DNA synthesis was accompanied by fragments of transplanted primary tumors. an alteration of the cell cycle, i.e., a prolongation Cytology of SV40-infected ChH cells. The of the S phase, was resolved by making a cytofollowing experiments were designed to deter- photometric analysis of the infected cultures. mine whether SV40 induces cellular DNA syn- The chronology of the changes of DNA content thesis in a growing culture. Confluent cultures at the single-cell level was determined. Confluent were held for 1 week, subcultured into petri monolayer cultures were held for 1 week to dishes, and infected after 3 hr. Figure 2b shows accumulate the majority of cells in the G1 phase. the percentage of nuclei labeled when 3H-thymi- The cells were trypsinized, seeded in petri dishes dine (0.2 ,uCi/ml) was added at the beginning containing cover slips, and infected as deof the experiment (no. 175). The infected cul- scribed in Materials and Methods. A control tures accumulated labeled nuclei at the same cover slip at 3 hr was examined under CYDAC: rate as the control cultures; however, more cells the proportion of cells in the G1 (2n) and G2-S were recruited into DNA synthesis in the in- (4n) phases was 86 and 13%, respectively (Fig. 3). fected population. Both control and infected Samples of control and infected cells (experi-

743


VOL. 6, 1970

ment no. 175) were stained with Feulgen at 16, 20, 24, and 34 hr, and the DNA content of individual nuclei was measured. These times were chosen so that the DNA content of nuclei could be determined on the rise of the DNA stimulation curve, at the peak, and a few hours after 2n

the plateau (Fig. 2b). The relative DNA content distribution of nuclei at these intervals is shown in Fig. 4-6. Table 2 summarizes the frequency of cells in 2n (G ), 4n (late part of S and G2), and 8n (polyploid), or greater, classes of nuclei. From Table 2 and Fig. 3-6, it is evident that a

on

120

180

600 5

.

RELATIVE DNA CONTENT (LOG)

FIG. 3. DNA content of nuclei as measured cytophotometrically by CYDAC (experiment 175) at 3 hr.

2n

4am

s0-

on

70-

60-

20 HOURS CONTROL 200 CELLS

5O-

n

U) IL

0 Ia.

'lx0

CD

80-

a

I

-, 1~~~~

III

70

20 HOURS INFECTED 200 CELLS

60-

SO. 40 30 20

10.

_

_ _--1

//

.~~~~~~~~~~~~~~~~~.1


FIG. 4. DNA content of nuclei as measured cytophotometrically by CYDAC (experiment 175) at 20 hr.

744

LEHMAN AND DEFENDI

J. VIROL.

4f

2

Sn

80S 7060-


SO. 40

()

20

-j

J

e. U

IO-

IL

I

m

a 00 D 70z

24 HOURS INFECTED

200 CELLS

50' 40

30

20

..

.

.

~~~~~._ec-,-,-.

~

.

.


FIG. 5. DNA content of nuclei as measured cytophotometrically by CYDAC (experiment 175) at 24 hr. 2n

4n

Sn


U) 4 4

20-

o

10-

LiJ

L

2V

4n

Sn

0

34 HOURS INFECTED 200 CELLS

5040-

302010-

mmDi

It

I..I|

-:

l.:


FIG. 6. DNA

of niuclei as measured cytophotometrically by CYDAC (experiment 175) at 34 hr. rapid increase in the class of 8n nuclei occurred. nuclei at 16 hr and then a marked increase to 8 % In the control, the number of polyploid cells at 20 hr, 16% at 24 hr, and 18% at 36 hr. remained at a level of 2%'. The infected cultures Between 16 and 24 hr, there was an increase began to show a twofold increase of polyploid in 4n cells and a concomitant decrease in 2n cells content

VOL. 6, 1970


TABLE 2. Percentage of cells with 2n, 4n, and 8n relative amounts of DNA as determinied by cytophotometry (CYDAC)a Cells

infection

2n

4n

......

3b

86.8

13.2

0

......

3

Control ...... Infected ......

16 16

57.5 50.5

40.5 45.0

2.0 4.5

125


20 20

53.0 25.5

44.5 66.5

2.5 8.0


24 24

45.0 18.0

52.5 66.0

2.5 16.0


34 34

68.5 34.0

29.5 48.0

2.0 18.0

Data from Fig. 3-6 were arbitrarily broken into classes. The 2n class was the G, phase. The 4n class included most of S and G2 phase. The 8n included all tetraploid and greater than tetraploid cells. b Based on count of 400 cells; all other times based on counts of 200 cells each. a

Polyploid mitoses (%) Infected

Control

63.0 96.0 96.0 96.0

8 .4b 11.8b 10.5b 14.6b

1.4 2.4 2.4 2.4

8.06 74.0 86.0 100.0

11 ob 10.6b 17.8b

2.6 3.2 2.4 3.9

95.0 96.0 100.0

4.Ob 6 .2b 5 .8c 7 .4b

0.8 0.8 3.2 2.0

8n

hr

Control Infected

TABLE 3. Percentage of polyploid cellsa in Chinese hamster cells infected with SV40 after T antigenExpt no.| Time infection Ipositive cells

Percentage of cells

Time after

745

161

162

24 48 72 96

hr hr hr

31 53 72 16

hr hr hr days

26 76 97 8

hr hr hr

hr

days

a Based on counts of 500 cells. b p 4n persistent increase in the number of cells. the Thus, population polyploids present to undergo division and give rise to polyploid soon after infection of ChH cells with SV40 cells had to be demonstrated before these cells survived and had the ability to proliferate. In could be considered relevant to the transformapreparations, no endoreduplicated tion process. Therefore, a cytogenetic analysis all of thesewere metaphases observed.t of the infected culture was made. To test whether the induction of polyploidy Cytogenetic analysis. In three separate experiments (Table 3), the percentage of polyploid cells was a result of viral gene effect, the viral coat in the infected cultures was three to five times protein, or other factors in the medium, as that in the controls by 30 hr postinfection. In reported with some other viruses (23), a cytothe control cultures, the percentage remained genetic analysis was made of cells infected with at a low level, but in the infected cultures it SV40 that had been irradiated for 15 or 30 min. The unirradiated (zero-time) virus was used continued to increase. By 24 hr postinfection,

746

LEHMAN AND DEFENDI

an M OI of 20, and each culture received an equal volume. A rise in polyploidy was observed in cultures infected with the zero-time virus, as described previously, but no rise was observed with the 30-min irradiated sample; in fact, it stayed at the same level as the control. The 15-min irradiated virus did induce a slight rise in the number of polyploid cells, but this returned to control values (Fig. 7). It is apparent that polyploidy is not the result of an effect of the viral coat protein or of some nonspecific cellular changes induced by virus, but rather the result of an event directly related to some expression of the viral genome. There are four possibilities for the source of the polyploid cell in the infected cultures: (i) tetraploid cells already existing in the normal population, (ii) cell fusion or incomplete cytokinesis initially yielding binucleate cells, (iii) faulty karyokinesis or spindle abnormalities that result in a mononuclear cell with 4n DNA content (this cell would have a subsequent DNA synthesis period followed by a mitosis; in these mitoses, the chromosomes may be arranged as diplochromosomes), and (iv) stimulation of cells in the G2 or S phase causing repetition of the S period without an intervening mitosis. Hypothesis one can be eliminated on the basis of the following. In the normal population, only a 2% background of tetraploid cells exists at any time (Table 2), and these cells would have to undergo at least three generation cycles for the infected cultures to attain 16% polyploidy at 24 hr. This appears unlikely, as the generation time of both normal and transformed ChH cells under these culture conditions is about 22 hr with an S period of 13 hr (Fig. 8). The second hypothesis was tested by counting the number of binucleate cells. Polyploids arising by this mechanism would first form binucleate cells by either cell fusion or karyokinesis, with no cytokinesis; then the two nuclei would fuse to form a mononucleate cell. But there was no increase of binucleate cells (4 to 6 per 1,000) in the infected versus control cultures at 0, 8, 12, 22, 24, and 28 hr after infection. The third and fourth hypotheses were tested by combining cinemicrography and cytophotometry. During 3 days after infection, 670 mitoses were observed in time-lapse movies in several microscopic fields; once mitosis began, all cells underwent complete cytokinesis. Seven cells were observed to undergo tripolar mitosis. Slides on which fields had been recorded on film were fixed, and the cells with DNA content > 8n were identified by cytophotometry. By backtracking the life history of these cells in the film, a clear determination could be made

at

10

J. VIROL.

1A

8

A

6

z

Q

* is' UV * 30 UV o CONTROL

9

A

w

5.

C.) w

a.

4-

A

32

t

A

I 0

2-

. *4 o8

0

6

24 48 72 96 120 144 168 192 216 HOURS

FIG. 7. Percentage ofpolyploid cells in ChH cultures infected with unirradiated (108 PFUIm), IS-min irradiated (10155 PFU/ml), and 30-min irradiated (108 PFU/ml) S V40. Cytogenetic preparations were prepared at the various time intervals indicated, and the percentage of polyploid metaphases was scored on the basis of500 mitoses counted.

2

4

6

8

I0 12 14 16

18

20 22 24 26 28 30 32 34 36

HOURS

FIG. 8. Cell cycle analysis of ChH cells (0) at the 12th passage and the transformed line 97SV cl 21 (0) at the 54th passage. The generation time of both cells is approximately 20 to 22 hr with an S period of 13 hr.

whether the polyploid cells arose by a faulty karyokinesis. In the selected fields, eight nuclei were found that had a DNA content > 8n. In the film, none of these cells had undergone a faulty mitosis. Thus, we can conclude that the fourth hypothesis is correct, namely, that the polyploid cells arise by an induction of two successive rounds of DNA synthesis in a G1, S, or G2 cell before mitosis. DISCUSSION In the experiments reported, we have established the following. (i) Primary or secondary ChH embryo cells are highly susceptible to SV40 infection, as demonstrated by T-antigen as to

VOL. 6, 1970


induction. (ii) The ChH cells are essentially nonpermissive, as viral replication can be detected in only a small fraction (