- Email: [email protected]
Serum-dependent control of entry into S phase of next generation in rat 3Y1 fibroblasts
Experimental
Cell Research
163 (1986) 127-134
Serum-Dependent Control of Entry into S Phase of Next Generation in Rat 3Yl Fibroblasts Effect of Large T Antigen
ATSUYUKI Department
OKUDA*
of Simian
and GENKI
Virus 40
KIMURA
of Virology, Medical Institute of Bioregulation, Kyushu University 69, Fukuoka 812, Japan
When rat 3Y 1 tibroblasts are deprived of serum in S phase and/or G2 phase in the first generation, the cells delay entry into S phase in the second generation for the duration of the serum deprivation. We can now show that when resting 3Yl cells are infected with Simian virus 40 (SV40), the removal of serum through S and G2 phases in the first generation does not markedly delay entry into S phase in the second generation. These observations suggest that the serum-dependent control of entry into S phase of the second generation continues from the first generation, and that the abolition of this control by infection with SV40 in the first generation involves the mechanism operative when the resting cells are stimulated to enter S phase (of the first generation) by infection with sv40.
@ 1986 Academic
Press, Inc.
The proliferation of ‘normal’ tibroblastic cells in culture is regulated by cell density and by factors in serum supplemented in the culture medium [ 1, 21. Rat 3Yl fibroblastic cells enter the resting state when they reach a confluent cell density, or when they are deprived of serum. Resting 3Yl cells enter S phase when exposed to fresh serum [3,4] or to growth factors [5-71. When infected with Simian virus 40 (SV40), resting 3Y 1 cells enter S phase in the absence of serum or growth factors at an extremely high cell density [8], and under intracellular proliferation-inhibitory conditions caused by temperature-sensitive mutations P-111 (for other cell lines, see [12] and reviews [13-151). The length of time required for the start of S phase is longer in resting cells stimulated by serum than in mitotic cells in an extensively proliferating population (for review, see [16]). The premitotic preparation for the initiation of DNA synthesis of the next generation was predicted for eukaryotes by analogy with the case of prokaryotes [17]. Using 3Yl cells, we found that the absence of serum or the inhibition of anchorage to the substratum during S and G2 phases in the first generation prolongs the interval between cell division and entry into S phase (Gl period) in the second generation [6, 181. Conditions required in the first generation to shorten the length of the Gl period in the second generation, such as the presence of serum or growth factors and the anchorage to the substratum, are also required for the resting cells to be stimulated to enter S phase (in the first * To whom offprint requests should be addressed 9-868333
Copyright 0 1986 by Academic Press, Inc. All rights of reproduction in any form reserved 0014-4827/86 $03.00
128
Okuda and Kimura
generation) [6, 181. These findings support our proposal [6, 181 that the resting state in the ‘normal’ cells is the non-proliferating state in which cells have divided with a very low level of ‘physiological potential’ for the initiation of DNA synthesis. In the present work we show that when resting 3Y1 cells are infected with SV40, the removal of serum during S and G2 phases in the first generation does not lead to a marked delay in entry into S phase in the second generation, thereby providing further support for our afore-mentioned proposal. MATERIALS
AND
METHODS
Cell Culture A clonal isolate (clone 1-6) of rat 3Yl-B diploid fibroblast-like cells [19] (referred to as 3Yl) was used. Details of cell culture techniques were as described [3]. Resting 3Y 1 cultures were prepared by seeding 1x lO’/cells in 5.2-cm plastic dishes with 5 ml of Dulbecco’s modified Eagle medium (DEM) containing 10% fetal calf serum (FCS), followed by incubation for 5 days. The medium for ‘serumfree’ cultures were DEM containing 500 &ml of a-globulin (a-globulin fraction IV-I, ICN Nutritional Biochemicals, Cleveland, Ohio) (see also below). All cultures were placed in a CO2 incubator at 37°C.
Virus Wild-type (WT) SV40, strain SV68C [20], and the viable deletion mutant of SV40, dl-884 (deleted in the coding region for small t antigen) [21], were used. Procedures for preparation of virus stocks were as described [8].
Virus Infection Resting 3Yl cells (1.4x 10?5.2-cm dish) were inoculated with 0.5 ml of the undiluted virus stock suspension (WT : 1.7~ 10s PFU/ml; dl-884 : 1.6~ 109 PFU/ml), and incubated at 37°C for 90 min for virus adsorption.
Induction
of Entry into S Phase in Resting Cells
Infected or uninfected resting cells were dispersed with trypsin-EDTA, and 2x lo5 cells were seeded in 3.3-cm dishes with 2 ml of DEM supplemented with 10% serum or a-globulin. For autoradiography [3H]thymidine (1 @/ml, 20 Ci/mmol, Amersham International, UK) was added to media from the start of incubation.
Arrest at Early S Phase by Aphidicolin SV40-infected or uninfected cells were dispersed with trypsin-EDTA, and 1.4~10~ cells were seeded in 8.6~cm dishes with 10 ml of DEM containing 10% serum and aphidicolin (Wako Pure Chemical, Osaka) (2.5 pg/ml) (or 2x16 cells in 3.3-cm dishes with 2 ml of the medium for flow cytofluorometry) and then incubated for 20 h.
Kinetics
of Entry into S Phase in Second Generation
The infected and uninfected groups of cultures arrested at early S phase were incubated with DEM supplemented with 10% serum or a-globulin for 6 h after removal of aphidicolin. Mitotic cells were detached by vigorous pipetting and collected by centrifugation. When the SV40-infected cells arrested at early S phase were incubated with DEM (not containing a-globulin), a substantial number of nonmitotic cells were also dislodged. Since a-globulin (500 pg/ml) has little or no activity for induction of entry into S phase in uninfected resting 3Y 1 cells [8], and had no effect on the delay in entry into S phase in the second generation when added to medium lacking serum in the first generation (see Exp
Cell
Res
163 (1986)
Proliferation
control and SV40 T antigen
129
SWOtSerum
Serum
12 h
16 h
I. Flow cytofluorogram of resting 3Y 1 cells stimulated by SV40 infection, by serum exposure, and by both SV40 infection and serum exposure. SV40-infected or uninfected resting cells were reseeded sparsely at 0 h and incubated in the presence or absence of serum. At the indicated times, the DNA content distribution was determined by flow cytofluorometry. All flow cytofluorograms are normalized so that the total cell number is equal. Fig.
20h
24 h
28 h
Results), it was added to DEM (500 ug/ml) in the serum-free condition throughout this study (referred to as ‘serum-free’) (see also Results). The mitotic cells were reseeded at a cell density of 1x lo4 per 3.3-cm dish with DEM containing [3H]thymidine (1 t&i/ml) plus a-globulin or plus 10% serum. At 2-h intervals, the cultures were fixed for autoradiography.
Detection of T Antigen-positive Indirect immunofluorescence
Cells
for nuclear SV40 T antigen was performed as described [8].
Flow Cytojkorometry Cellular DNA was stained with propidium iodide (50 @ml in 0.1% sodium citrate), the nuclei were extracted and the DNA content of the individual nuclei was determined by flow cytofluorometry as described [18]. All cytograms were normalized, so that the total cell number would be the same.
RESULTS Analysis by Flow Cytojluorometry of Cell Cycle Progression in Serumstimulated Resting Cells and in SV40-infected Resting Cells (Jig. 1)
When the resting 3Yl cells infected or uninfected with SV40 were reseeded sparsely, and incubated in the presence of serum, most cells completed one round of cell-cycle phases within 24 h. The speed of the cell-cycle progression was more rapid in the infected cells. The SV40-infected resting cells also progressed through a round of cell cycle phases in the absence of serum. The uninfected resting cells did not progress through cell-cycle phases in the absence of serum (data not shown, and see below). Exp Cell Res 163 (1986)
130
Okuda and Kimura
2 12 I6 4 6 Hours afier reseeding
20
Fig. 2. Kinetics of entry into S phase in resting 3Yl cells stimulated by SV40 infection, by serum exposure, and by both SV40 infection and serum exposure. SV40-infected or uninfected resting cells were reseeded sparsely at 0 h, and were incubated in the presence or absence of serum with [3H]thymidine. At the indicated times the fractions of T antigen-positive cells and of i3H]thymidinelabelled cells were determined. T antigen-positive cells: 0, SV40, serum +; A, SV40, serum -. [3H]Thymidine-labelled cells: 0, SV40, serum +; A, SV40, serum -; 0, no infection, serum +; w, no infection, serum -. Fig. 4. Effect of serum removal during S and G2 phases from resting 3Y 1 cultures infected with the deletion mutant of SV40 @I-84) on the time course of entry into S phase in the second generation Resting cells were infected or uninfected with di-884 or WT SV40, reseeded sparsely, and were arrested at early S phase by aphidicolin as described in Materials and Methods. After release from the arrest the cells were incubated in the presence or absence of serum. Then, mitotic cells were selectively collected, and were incubated in the absence (for infected cells) or presence (for uninfected cells) of serum. The fraction of cells which had entered S phase by the indicated time was determined by continuously exposing cells to [3H]thymidine followed by processing for autoradiography. WT-SV40 infection in the 0, presence; 0, absence of serum during S and G2 phases of the first generation; dI-884 infection in the A, presence; A, absence of serum during S and G2 phase of the first generation; no infection in the n , presence; 0, absence of serum during S and G2 pahses of the first generation.
Kinetics of Entry into S Phase in Serum-stimulated Resting Cells (fig. 2)
or SV40-infected
Serum-stimulated and SV40-infected resting cells were each continuously exposed to L3Hlthymidine, and kinetics of entry into S phase was examined by autoradiography. SV40-infected cells incubated in the presence of serum, those incubated in the absence of serum, and uninfected cells incubated in the presence of serum entered S phase, in that order, with respect to the time course. In SV40infected cells, the emergence of T antigen-positive cells and that of S-phase cells were delayed in parallel in the absence of serum compared with its presence. Similarly, when SV40-infected resting 3Y 1 cells are reseeded sparsely with DEM alone, or at an extremely high cell density with DEM plus a-globulin, both Tantigen expression and entry into S phase are delayed in parallel, compared with the case in which they are reseeded sparsely with DEM plus a-globulin [8]. Uninfected cells did not enter S phase solely in the presence of a-globulin. In this sense, the culture with DEM plus a-globulin is referred to as the serum-free condition, even though a-globulin contained unidentified serum components (see Materials and Methods). Exp Cell Res 163 (1986)
Proliferation
control and SV40 T antigen
131
100 _ No infection
a
C b 0 Hours
10 after mitotic
Fig. 3. Effects of serum removal during S and G2 phases of the first generation from SV40-infected or uninfected resting 3Y 1 cultures, on the time course of entry into S phase in the second generation. Experimental design and the symbols used in (a, b) are presented schematically in (c). Incubation in the S+, presence; S-, absence of serum. AC, aphidicolin.
20 sekction
Effects of Serum Deprivation on Kinetics of Entry into S Phase in Second Generation in Uninfected and SV40-infected Resting Cells (figs 3, 4)
Uninfected and SV40-infected resting 3Y1 cells were each reseeded sparsely and incubated in the presence of both serum and aphidicolin for 20 h, so that they were arrested at early S phase [6]. Each of these two groups of cultures was incubated after removal of aphidicolin in the presence or absence of serum for 6 h. There was no significant difference in the progression through S and G2 phases between the infected and the uninfected cells (data not shown), although the speed of the progression was slightly more rapid in the absence than in the presence of serum, as noted in uninfected 3Yl cells [18]. Mitotic cells in each of these four groups of cultures were selectively collected and incubated in the presence or absence of serum. During this incubation cells were continuously labelled with [3H]thymidine, and the fraction of cells which had entered S phase was determined by autoradiography. These procedures are shown schematically in fig. 3 c. In the uninfected cultures, if serum was absent through S and G2 phases in the first generation, cells delayed entry into S phase in the presence of serum in the second generation (cf fig. 3 c, C vs A), in agreement with previous results [6, 181. A-substantial fraction of cells in the culture which had not been deprived of serum in the first generation, entered S phase in the absence of serum in the second Exp
Cell
Rev 163 (1986)
132
Okuda
and Kimura
generation. This was not the case with the culture deprived of serum through S and G2 phases in the first generation (cf fig. 3c, D vs B). When resting cells were infected with WT SV40, the serum deprivation through S and G2 phases in the first generation did not cause such a marked delaying effect on entry into S phase in the second generation (cf fig. 3 c, G vs E). In addition, SV40 infection in the first generation enabled cells to enter S phase in the second generation in the absence of serum, regardless of the absence of serum in the first generation (cf fig. 3 c, H vs F). Also in this case, entry into S phase in the second generation was not markedly delayed in the absence of serum through S and G2 phases in the first generation, though a slight delaying effect remained (cf fig. 4, 0, serum-free in the first generation with 0, serum-presence in the first generation). This was also the case when resting cells were infected with the deletion mutant of SV40 (df-884), which lacks the coding region for small t antigen (cf fig. 4, A, serum-free in the first generation with A, serum-presence in the first generation). In all cases, the entry into S phase in the second generation was more rapid in the cultures infected with SV40 (both WT and dl-884) in the first generation than in those not infected (fig. 4). DISCUSSION According to our model [6, 181, the regulation of proliferation of cultured ‘normal’ tibroblastic cells under the ordinary physiological conditions may be described as follows. The process(es) by which cells prepare for entry into S phase continues from the previous generation under the ordinary physiological conditions. The process(es) is controlled by conditions of the cells’ environment. The cellular ‘physiological potential’ for entry into S phase is elevated as the process(es) is executed under conditions optimum for proliferation. When the ‘physiological potential’ is elevated to or over a threshold level, cells are committed into S phase. It decreases to the minimum level as the cell traverses S and G2 phases, if no attempt is made to increase it, i.e., under suboptimal conditions for proliferation. The resting state is the one in which cells have divided with the minimum level of the ‘physiological potential’, or where the once-attained ‘physiological potential’ of relatively higher level is lowered to the minimum level during the Gl phase. When the resting cells are placed in the conditions optimum for proliferation, they require a lag phase before entry into S phase. The duration of the lag phase is longer than that of the Gl phase for the cells exposed to the conditions optimal for proliferation through S and G2 phases in the previous generation [6] (for review, see [16]). When uninfected cells were deprived of serum for a period of time in S and G2 phases in the first generation, cells delayed entry into S phase in the second generation (see also [6, 181). The delay of entry into S phase is due to the additional time required for the resting cells or the divided cells to elevate the Exp
Cell
Res
163 (1986)
Proliferation
control
and SV40 T antigen
133
‘physiological potential’ to the level that the cells could have attained by the time of mitosis in the environments more favourable for proliferation. The retardation of entry into S phase in the second generation, caused by the deprivation of serum in the first generation, diminished when resting cells were infected with SV40 in the first generation. Similar results were obtained with the deletion mutant of SV40 (dI-884) which lacks the coding region for small t antigen, suggesting that this effect is mediated by the viral large T antigen. SV40infected resting cells enter S phase in the first and second generations in the absence of serum with the aid of large T antigen accumulated in the cells [S]. Therefore, the marked shortening of delay in entry into S phase in the second generation seen in the SV40-infected cells can be explained if we assume that large T antigen increases the ‘physiological potential’ during S and/or G2 phases in the first generation, even in the absence of serum. In other words, large T antigen, in place of growth factors in serum, activates the process(es) which works in both the first and the second generation to prepare for entry into S phase of the second generation. Consequently, the cells expressing large T antigen are prevented from entering the resting state after mitosis. This explains why SV40transformed cells do not enter the resting state [2]. The length of delay in entry into S phase in the second generation in the uninfected cells deprived of serum in the first generation is much the same as the time length of serum absence in the first generation [6, 181. The length of time between the emergence of T antigen-positive cells and that of S-phase cells in the SV40-infected resting culture in the first generation (approx. 6 h, as shown in fig. 2) was almost the same as that between the beginning of S phase and mitosis in the first generation in uninfected and SV40-infected cultures (6 h). Therefore, if the explanation given above is valid, mitotic cells which have accumulated a sufficient amount of large T antigen by the early S phase in the first generation would enter S phase in the second generation immediately after cell division. However, the mitotic cells which had been infected with SV40 (both WT and dl884), and which therefore were already T-antigen-positive at the early S phase in the first generation, required 2-10 h to enter S phase in the second generation. Furthermore, in the SV40 (both WT and dl-884)-infected resting cells, the deprivation of serum through S and G2 phases in the first generation caused a slight delay in entry into S phase in the second generation. It is likely that large T antigen cannot elevate the ‘physiological potential’ during S and/or G2 phases as efficiently as it can during the phase between the resting state and S phase. If there are interactions between the two parallel processes (the progression through S and G2 phases and the elevation of the ‘physiological potential’), this can be justified, because no such interactions can exist when resting cells are stimulated to enter S phase. In fact, there was no delay in entry into S phase in the second generation, when the cells infected with SV40 in the first generation were deprived of serum in the second generation. Since the time required for either the resting or the mitotic cells to enter S Exp Cell
Res
163 (1986)
134
Okuda and Kimura
phase is shorter when mediated by large T antigen than when mediated by serum, the ‘power’ of the large T antigen to increase the ‘physiological potential’ exceeds that of serum. In the SV40-infected resting cells, entry into S phase in the first generation was more rapid in the presence than absence of serum. In the presence of serum, the lag time before expression of detectable T antigen shortened, while the interval between detectable T-antigen expression and entry into S phase was constant, irrespective of the presence or absence of serum. Therefore, the role of serum in the faster entry into S phase in the SV40-infected resting cells is in the process between SV40 virion adsorption and the detectable expression or accumulation of large T antigen. Similar situations were that the delay in production or accumulation of large T antigen accompanies a parallel delay in entry into S phase, when SV40-infected resting 3Yl cells are reseeded under completely serum-free conditions or under conditions of an extremely high cell density [8]. We thank M. Ohara for valuable comments on the manuscript. This work was supported in part by the Grants-in-Aid for Cancer Research from the Ministry of Education, Science and Culture of Japan, by the Fukuoka-Ken Cancer Society and by the Nakamura Foundation.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.
Holley, R W & Kieman, J A, Proc nati acad sci US 60 (1968) 300. Bartholomew, J C, Yokota, H & Ross, P, J cell physiol 88 (1976) 277. Okuda, A & Kimura, G, Exp cell res 111 (1978) 55. - Ibid 145 (1983) 155. Okuda, A, Kajiwara, Y & Kimura, G, In vitro 19 (1983) 376. Okuda, A & Kimura, G, Exp cell res 155 (1984) 24. Okuda, A, Yamada, K & Kimura, G, Growth and differentiation of cells in defined environment (ed H Murakami, I Yamane, D W Barnes, J P Mather, I Hayashi & G H Sato) p. 271. Kodansha, Tokyo and Springer-Verlag, Berlin, Heidelberg, New York, Tokyo (1985). Okuda, A, Shimura, H & Kimura, G, Virology 133 (1984) 35. Ohno, K & Kimura, G, Somat cell mol genet 10 (1984) 29. Mitsudomi, T & Kimura, G, J cell physiol 123 (1985) 353. Zaitsu, H & Kimura, G, J cell physiol 123 (1985) 305. Soprano, K J, Galanti, N, Jonak, G J, McKercher, S, Pipas, J M, Peden, K W C & Baserga, R, Mol cell biol 3 (1983) 214. Graessmann, A, Graessmann, M & Mueller, C, Adv cancer res 35 (1981) 111. Hand, R, Biochim biophys acta 651 (1981) 1. Martin, R G, Adv cancer res 34 (1981) 1. Baserga, R, Multiplication and division in mammalian cells, p. 175. Marcel Dekker, New York (1976). Cooper, S, Nature 280 (1979) 17. Okuda, A & Kimura, G, J cell physiol 110 (1982) 267. Kimura, G, Itagaki, A & Summers, J, Int j cancer 15 (1975) 694. Todaro, G J & Takemoto, K K, Proc natl acad sci US 62 (1%9) 1031. Shenk, T E, Carbon, J & Berg, P, J viral 18 (1976) 664.
Received July 2, 1985 Revised version received September 10, 1985 Exp
Cell
Rev
163 (1986)
F’rinted
in Sweden
Cell Research
163 (1986) 127-134
Serum-Dependent Control of Entry into S Phase of Next Generation in Rat 3Yl Fibroblasts Effect of Large T Antigen
ATSUYUKI Department
OKUDA*
of Simian
and GENKI
Virus 40
KIMURA
of Virology, Medical Institute of Bioregulation, Kyushu University 69, Fukuoka 812, Japan
When rat 3Y 1 tibroblasts are deprived of serum in S phase and/or G2 phase in the first generation, the cells delay entry into S phase in the second generation for the duration of the serum deprivation. We can now show that when resting 3Yl cells are infected with Simian virus 40 (SV40), the removal of serum through S and G2 phases in the first generation does not markedly delay entry into S phase in the second generation. These observations suggest that the serum-dependent control of entry into S phase of the second generation continues from the first generation, and that the abolition of this control by infection with SV40 in the first generation involves the mechanism operative when the resting cells are stimulated to enter S phase (of the first generation) by infection with sv40.
@ 1986 Academic
Press, Inc.
The proliferation of ‘normal’ tibroblastic cells in culture is regulated by cell density and by factors in serum supplemented in the culture medium [ 1, 21. Rat 3Yl fibroblastic cells enter the resting state when they reach a confluent cell density, or when they are deprived of serum. Resting 3Yl cells enter S phase when exposed to fresh serum [3,4] or to growth factors [5-71. When infected with Simian virus 40 (SV40), resting 3Y 1 cells enter S phase in the absence of serum or growth factors at an extremely high cell density [8], and under intracellular proliferation-inhibitory conditions caused by temperature-sensitive mutations P-111 (for other cell lines, see [12] and reviews [13-151). The length of time required for the start of S phase is longer in resting cells stimulated by serum than in mitotic cells in an extensively proliferating population (for review, see [16]). The premitotic preparation for the initiation of DNA synthesis of the next generation was predicted for eukaryotes by analogy with the case of prokaryotes [17]. Using 3Yl cells, we found that the absence of serum or the inhibition of anchorage to the substratum during S and G2 phases in the first generation prolongs the interval between cell division and entry into S phase (Gl period) in the second generation [6, 181. Conditions required in the first generation to shorten the length of the Gl period in the second generation, such as the presence of serum or growth factors and the anchorage to the substratum, are also required for the resting cells to be stimulated to enter S phase (in the first * To whom offprint requests should be addressed 9-868333
Copyright 0 1986 by Academic Press, Inc. All rights of reproduction in any form reserved 0014-4827/86 $03.00
128
Okuda and Kimura
generation) [6, 181. These findings support our proposal [6, 181 that the resting state in the ‘normal’ cells is the non-proliferating state in which cells have divided with a very low level of ‘physiological potential’ for the initiation of DNA synthesis. In the present work we show that when resting 3Y1 cells are infected with SV40, the removal of serum during S and G2 phases in the first generation does not lead to a marked delay in entry into S phase in the second generation, thereby providing further support for our afore-mentioned proposal. MATERIALS
AND
METHODS
Cell Culture A clonal isolate (clone 1-6) of rat 3Yl-B diploid fibroblast-like cells [19] (referred to as 3Yl) was used. Details of cell culture techniques were as described [3]. Resting 3Y 1 cultures were prepared by seeding 1x lO’/cells in 5.2-cm plastic dishes with 5 ml of Dulbecco’s modified Eagle medium (DEM) containing 10% fetal calf serum (FCS), followed by incubation for 5 days. The medium for ‘serumfree’ cultures were DEM containing 500 &ml of a-globulin (a-globulin fraction IV-I, ICN Nutritional Biochemicals, Cleveland, Ohio) (see also below). All cultures were placed in a CO2 incubator at 37°C.
Virus Wild-type (WT) SV40, strain SV68C [20], and the viable deletion mutant of SV40, dl-884 (deleted in the coding region for small t antigen) [21], were used. Procedures for preparation of virus stocks were as described [8].
Virus Infection Resting 3Yl cells (1.4x 10?5.2-cm dish) were inoculated with 0.5 ml of the undiluted virus stock suspension (WT : 1.7~ 10s PFU/ml; dl-884 : 1.6~ 109 PFU/ml), and incubated at 37°C for 90 min for virus adsorption.
Induction
of Entry into S Phase in Resting Cells
Infected or uninfected resting cells were dispersed with trypsin-EDTA, and 2x lo5 cells were seeded in 3.3-cm dishes with 2 ml of DEM supplemented with 10% serum or a-globulin. For autoradiography [3H]thymidine (1 @/ml, 20 Ci/mmol, Amersham International, UK) was added to media from the start of incubation.
Arrest at Early S Phase by Aphidicolin SV40-infected or uninfected cells were dispersed with trypsin-EDTA, and 1.4~10~ cells were seeded in 8.6~cm dishes with 10 ml of DEM containing 10% serum and aphidicolin (Wako Pure Chemical, Osaka) (2.5 pg/ml) (or 2x16 cells in 3.3-cm dishes with 2 ml of the medium for flow cytofluorometry) and then incubated for 20 h.
Kinetics
of Entry into S Phase in Second Generation
The infected and uninfected groups of cultures arrested at early S phase were incubated with DEM supplemented with 10% serum or a-globulin for 6 h after removal of aphidicolin. Mitotic cells were detached by vigorous pipetting and collected by centrifugation. When the SV40-infected cells arrested at early S phase were incubated with DEM (not containing a-globulin), a substantial number of nonmitotic cells were also dislodged. Since a-globulin (500 pg/ml) has little or no activity for induction of entry into S phase in uninfected resting 3Y 1 cells [8], and had no effect on the delay in entry into S phase in the second generation when added to medium lacking serum in the first generation (see Exp
Cell
Res
163 (1986)
Proliferation
control and SV40 T antigen
129
SWOtSerum
Serum
12 h
16 h
I. Flow cytofluorogram of resting 3Y 1 cells stimulated by SV40 infection, by serum exposure, and by both SV40 infection and serum exposure. SV40-infected or uninfected resting cells were reseeded sparsely at 0 h and incubated in the presence or absence of serum. At the indicated times, the DNA content distribution was determined by flow cytofluorometry. All flow cytofluorograms are normalized so that the total cell number is equal. Fig.
20h
24 h
28 h
Results), it was added to DEM (500 ug/ml) in the serum-free condition throughout this study (referred to as ‘serum-free’) (see also Results). The mitotic cells were reseeded at a cell density of 1x lo4 per 3.3-cm dish with DEM containing [3H]thymidine (1 t&i/ml) plus a-globulin or plus 10% serum. At 2-h intervals, the cultures were fixed for autoradiography.
Detection of T Antigen-positive Indirect immunofluorescence
Cells
for nuclear SV40 T antigen was performed as described [8].
Flow Cytojkorometry Cellular DNA was stained with propidium iodide (50 @ml in 0.1% sodium citrate), the nuclei were extracted and the DNA content of the individual nuclei was determined by flow cytofluorometry as described [18]. All cytograms were normalized, so that the total cell number would be the same.
RESULTS Analysis by Flow Cytojluorometry of Cell Cycle Progression in Serumstimulated Resting Cells and in SV40-infected Resting Cells (Jig. 1)
When the resting 3Yl cells infected or uninfected with SV40 were reseeded sparsely, and incubated in the presence of serum, most cells completed one round of cell-cycle phases within 24 h. The speed of the cell-cycle progression was more rapid in the infected cells. The SV40-infected resting cells also progressed through a round of cell cycle phases in the absence of serum. The uninfected resting cells did not progress through cell-cycle phases in the absence of serum (data not shown, and see below). Exp Cell Res 163 (1986)
130
Okuda and Kimura
2 12 I6 4 6 Hours afier reseeding
20
Fig. 2. Kinetics of entry into S phase in resting 3Yl cells stimulated by SV40 infection, by serum exposure, and by both SV40 infection and serum exposure. SV40-infected or uninfected resting cells were reseeded sparsely at 0 h, and were incubated in the presence or absence of serum with [3H]thymidine. At the indicated times the fractions of T antigen-positive cells and of i3H]thymidinelabelled cells were determined. T antigen-positive cells: 0, SV40, serum +; A, SV40, serum -. [3H]Thymidine-labelled cells: 0, SV40, serum +; A, SV40, serum -; 0, no infection, serum +; w, no infection, serum -. Fig. 4. Effect of serum removal during S and G2 phases from resting 3Y 1 cultures infected with the deletion mutant of SV40 @I-84) on the time course of entry into S phase in the second generation Resting cells were infected or uninfected with di-884 or WT SV40, reseeded sparsely, and were arrested at early S phase by aphidicolin as described in Materials and Methods. After release from the arrest the cells were incubated in the presence or absence of serum. Then, mitotic cells were selectively collected, and were incubated in the absence (for infected cells) or presence (for uninfected cells) of serum. The fraction of cells which had entered S phase by the indicated time was determined by continuously exposing cells to [3H]thymidine followed by processing for autoradiography. WT-SV40 infection in the 0, presence; 0, absence of serum during S and G2 phases of the first generation; dI-884 infection in the A, presence; A, absence of serum during S and G2 phase of the first generation; no infection in the n , presence; 0, absence of serum during S and G2 pahses of the first generation.
Kinetics of Entry into S Phase in Serum-stimulated Resting Cells (fig. 2)
or SV40-infected
Serum-stimulated and SV40-infected resting cells were each continuously exposed to L3Hlthymidine, and kinetics of entry into S phase was examined by autoradiography. SV40-infected cells incubated in the presence of serum, those incubated in the absence of serum, and uninfected cells incubated in the presence of serum entered S phase, in that order, with respect to the time course. In SV40infected cells, the emergence of T antigen-positive cells and that of S-phase cells were delayed in parallel in the absence of serum compared with its presence. Similarly, when SV40-infected resting 3Y 1 cells are reseeded sparsely with DEM alone, or at an extremely high cell density with DEM plus a-globulin, both Tantigen expression and entry into S phase are delayed in parallel, compared with the case in which they are reseeded sparsely with DEM plus a-globulin [8]. Uninfected cells did not enter S phase solely in the presence of a-globulin. In this sense, the culture with DEM plus a-globulin is referred to as the serum-free condition, even though a-globulin contained unidentified serum components (see Materials and Methods). Exp Cell Res 163 (1986)
Proliferation
control and SV40 T antigen
131
100 _ No infection
a
C b 0 Hours
10 after mitotic
Fig. 3. Effects of serum removal during S and G2 phases of the first generation from SV40-infected or uninfected resting 3Y 1 cultures, on the time course of entry into S phase in the second generation. Experimental design and the symbols used in (a, b) are presented schematically in (c). Incubation in the S+, presence; S-, absence of serum. AC, aphidicolin.
20 sekction
Effects of Serum Deprivation on Kinetics of Entry into S Phase in Second Generation in Uninfected and SV40-infected Resting Cells (figs 3, 4)
Uninfected and SV40-infected resting 3Y1 cells were each reseeded sparsely and incubated in the presence of both serum and aphidicolin for 20 h, so that they were arrested at early S phase [6]. Each of these two groups of cultures was incubated after removal of aphidicolin in the presence or absence of serum for 6 h. There was no significant difference in the progression through S and G2 phases between the infected and the uninfected cells (data not shown), although the speed of the progression was slightly more rapid in the absence than in the presence of serum, as noted in uninfected 3Yl cells [18]. Mitotic cells in each of these four groups of cultures were selectively collected and incubated in the presence or absence of serum. During this incubation cells were continuously labelled with [3H]thymidine, and the fraction of cells which had entered S phase was determined by autoradiography. These procedures are shown schematically in fig. 3 c. In the uninfected cultures, if serum was absent through S and G2 phases in the first generation, cells delayed entry into S phase in the presence of serum in the second generation (cf fig. 3 c, C vs A), in agreement with previous results [6, 181. A-substantial fraction of cells in the culture which had not been deprived of serum in the first generation, entered S phase in the absence of serum in the second Exp
Cell
Rev 163 (1986)
132
Okuda
and Kimura
generation. This was not the case with the culture deprived of serum through S and G2 phases in the first generation (cf fig. 3c, D vs B). When resting cells were infected with WT SV40, the serum deprivation through S and G2 phases in the first generation did not cause such a marked delaying effect on entry into S phase in the second generation (cf fig. 3 c, G vs E). In addition, SV40 infection in the first generation enabled cells to enter S phase in the second generation in the absence of serum, regardless of the absence of serum in the first generation (cf fig. 3 c, H vs F). Also in this case, entry into S phase in the second generation was not markedly delayed in the absence of serum through S and G2 phases in the first generation, though a slight delaying effect remained (cf fig. 4, 0, serum-free in the first generation with 0, serum-presence in the first generation). This was also the case when resting cells were infected with the deletion mutant of SV40 (df-884), which lacks the coding region for small t antigen (cf fig. 4, A, serum-free in the first generation with A, serum-presence in the first generation). In all cases, the entry into S phase in the second generation was more rapid in the cultures infected with SV40 (both WT and dl-884) in the first generation than in those not infected (fig. 4). DISCUSSION According to our model [6, 181, the regulation of proliferation of cultured ‘normal’ tibroblastic cells under the ordinary physiological conditions may be described as follows. The process(es) by which cells prepare for entry into S phase continues from the previous generation under the ordinary physiological conditions. The process(es) is controlled by conditions of the cells’ environment. The cellular ‘physiological potential’ for entry into S phase is elevated as the process(es) is executed under conditions optimum for proliferation. When the ‘physiological potential’ is elevated to or over a threshold level, cells are committed into S phase. It decreases to the minimum level as the cell traverses S and G2 phases, if no attempt is made to increase it, i.e., under suboptimal conditions for proliferation. The resting state is the one in which cells have divided with the minimum level of the ‘physiological potential’, or where the once-attained ‘physiological potential’ of relatively higher level is lowered to the minimum level during the Gl phase. When the resting cells are placed in the conditions optimum for proliferation, they require a lag phase before entry into S phase. The duration of the lag phase is longer than that of the Gl phase for the cells exposed to the conditions optimal for proliferation through S and G2 phases in the previous generation [6] (for review, see [16]). When uninfected cells were deprived of serum for a period of time in S and G2 phases in the first generation, cells delayed entry into S phase in the second generation (see also [6, 181). The delay of entry into S phase is due to the additional time required for the resting cells or the divided cells to elevate the Exp
Cell
Res
163 (1986)
Proliferation
control
and SV40 T antigen
133
‘physiological potential’ to the level that the cells could have attained by the time of mitosis in the environments more favourable for proliferation. The retardation of entry into S phase in the second generation, caused by the deprivation of serum in the first generation, diminished when resting cells were infected with SV40 in the first generation. Similar results were obtained with the deletion mutant of SV40 (dI-884) which lacks the coding region for small t antigen, suggesting that this effect is mediated by the viral large T antigen. SV40infected resting cells enter S phase in the first and second generations in the absence of serum with the aid of large T antigen accumulated in the cells [S]. Therefore, the marked shortening of delay in entry into S phase in the second generation seen in the SV40-infected cells can be explained if we assume that large T antigen increases the ‘physiological potential’ during S and/or G2 phases in the first generation, even in the absence of serum. In other words, large T antigen, in place of growth factors in serum, activates the process(es) which works in both the first and the second generation to prepare for entry into S phase of the second generation. Consequently, the cells expressing large T antigen are prevented from entering the resting state after mitosis. This explains why SV40transformed cells do not enter the resting state [2]. The length of delay in entry into S phase in the second generation in the uninfected cells deprived of serum in the first generation is much the same as the time length of serum absence in the first generation [6, 181. The length of time between the emergence of T antigen-positive cells and that of S-phase cells in the SV40-infected resting culture in the first generation (approx. 6 h, as shown in fig. 2) was almost the same as that between the beginning of S phase and mitosis in the first generation in uninfected and SV40-infected cultures (6 h). Therefore, if the explanation given above is valid, mitotic cells which have accumulated a sufficient amount of large T antigen by the early S phase in the first generation would enter S phase in the second generation immediately after cell division. However, the mitotic cells which had been infected with SV40 (both WT and dl884), and which therefore were already T-antigen-positive at the early S phase in the first generation, required 2-10 h to enter S phase in the second generation. Furthermore, in the SV40 (both WT and dl-884)-infected resting cells, the deprivation of serum through S and G2 phases in the first generation caused a slight delay in entry into S phase in the second generation. It is likely that large T antigen cannot elevate the ‘physiological potential’ during S and/or G2 phases as efficiently as it can during the phase between the resting state and S phase. If there are interactions between the two parallel processes (the progression through S and G2 phases and the elevation of the ‘physiological potential’), this can be justified, because no such interactions can exist when resting cells are stimulated to enter S phase. In fact, there was no delay in entry into S phase in the second generation, when the cells infected with SV40 in the first generation were deprived of serum in the second generation. Since the time required for either the resting or the mitotic cells to enter S Exp Cell
Res
163 (1986)
134
Okuda and Kimura
phase is shorter when mediated by large T antigen than when mediated by serum, the ‘power’ of the large T antigen to increase the ‘physiological potential’ exceeds that of serum. In the SV40-infected resting cells, entry into S phase in the first generation was more rapid in the presence than absence of serum. In the presence of serum, the lag time before expression of detectable T antigen shortened, while the interval between detectable T-antigen expression and entry into S phase was constant, irrespective of the presence or absence of serum. Therefore, the role of serum in the faster entry into S phase in the SV40-infected resting cells is in the process between SV40 virion adsorption and the detectable expression or accumulation of large T antigen. Similar situations were that the delay in production or accumulation of large T antigen accompanies a parallel delay in entry into S phase, when SV40-infected resting 3Yl cells are reseeded under completely serum-free conditions or under conditions of an extremely high cell density [8]. We thank M. Ohara for valuable comments on the manuscript. This work was supported in part by the Grants-in-Aid for Cancer Research from the Ministry of Education, Science and Culture of Japan, by the Fukuoka-Ken Cancer Society and by the Nakamura Foundation.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.
Holley, R W & Kieman, J A, Proc nati acad sci US 60 (1968) 300. Bartholomew, J C, Yokota, H & Ross, P, J cell physiol 88 (1976) 277. Okuda, A & Kimura, G, Exp cell res 111 (1978) 55. - Ibid 145 (1983) 155. Okuda, A, Kajiwara, Y & Kimura, G, In vitro 19 (1983) 376. Okuda, A & Kimura, G, Exp cell res 155 (1984) 24. Okuda, A, Yamada, K & Kimura, G, Growth and differentiation of cells in defined environment (ed H Murakami, I Yamane, D W Barnes, J P Mather, I Hayashi & G H Sato) p. 271. Kodansha, Tokyo and Springer-Verlag, Berlin, Heidelberg, New York, Tokyo (1985). Okuda, A, Shimura, H & Kimura, G, Virology 133 (1984) 35. Ohno, K & Kimura, G, Somat cell mol genet 10 (1984) 29. Mitsudomi, T & Kimura, G, J cell physiol 123 (1985) 353. Zaitsu, H & Kimura, G, J cell physiol 123 (1985) 305. Soprano, K J, Galanti, N, Jonak, G J, McKercher, S, Pipas, J M, Peden, K W C & Baserga, R, Mol cell biol 3 (1983) 214. Graessmann, A, Graessmann, M & Mueller, C, Adv cancer res 35 (1981) 111. Hand, R, Biochim biophys acta 651 (1981) 1. Martin, R G, Adv cancer res 34 (1981) 1. Baserga, R, Multiplication and division in mammalian cells, p. 175. Marcel Dekker, New York (1976). Cooper, S, Nature 280 (1979) 17. Okuda, A & Kimura, G, J cell physiol 110 (1982) 267. Kimura, G, Itagaki, A & Summers, J, Int j cancer 15 (1975) 694. Todaro, G J & Takemoto, K K, Proc natl acad sci US 62 (1%9) 1031. Shenk, T E, Carbon, J & Berg, P, J viral 18 (1976) 664.
Received July 2, 1985 Revised version received September 10, 1985 Exp
Cell
Rev
163 (1986)
F’rinted
in Sweden