- Email: [email protected]
Interrelation between cellular rRNA content and regulation of the cell cycle of normal and transformed mouse cell lines
Cell Biology
International
Interrelation of the Cell
Reports,
Vol. 9, No.
between Cellular Cycle of Normal and Beate
Kleuser
and
11, November
rRNA Content Transformed Gerold
1985
985
and Regulation Mouse Cell Lines
Adam
FakultZt fiir Biologie Universitat Konstanz D-7750 Konstanz, FRG ABSTRACT The relation between cellular rRNA content, as a measure of cell size, and the regulation of the cell cycle has been investigated for Swiss 3T3 and the spontaneously transformed Swiss 3T6 cell line. It is shown that the characteristic of percent of quiescent cells stimulated into the cell cycle versus cellular rRNA content is basically different for 3T3 and 3T6 cells: 3T3 cells do not enter the cell cycle below a certain threshold of cellular rR.NA content, whereas 3T6 cells start proliferation without a substantial increase of rRNA. These data are interpreted as consistent with transformation of 3T6 cells being in essence their uncoupling from the requirement of normal cells of passing over a threshold of cellular rRNA content (cell size) before initiating DNA-replication. INTRODUCTION The concept of cell size requirements governing the cell cycle of eukaryotic cells has been attractive since its inception by Hertwig (1908). Modern models of the mammalian cell cycle postulate a continuous growth cycle superimposed to, and sufficient for the regulation of, the chromosome cycle (Cooper, 1982; Prescott, 1982). The somewhat protean notion of "cell size" may be specified in terms of cellular rRNA content as a relevant parameter describing cellular growth (Prescott, 1976; Baserga, 1976, 1981, 1982). More recent data on the relation between cellular rRNA content and regulation of mammalian cell proliferation appeared to contradict the basic concept that the single parameter of cell size (viz. cellular rRNA) governs entry into the cell cycle. Following the dissection of specific growth factor requirements of initiation of the fibroblast cell cycle in terms of "competence" and "progression" (Scher et al., 1979), it was shown that progression, but not induction of competence leads to an increase of cellular rRNA content (Seuwen and Adam, 1983; Adam et al., 1983; Seuwen et al., 1984). Consequently, an increase of cell size (viz. rRNA) only did not appear sufficient for entry into the cell cycle. Furthermore, in a recent review of pertinent data, quite a number of cellular systems were compiled, showing entry into the cell cycle without previous accumulation of cellular rRNA (Baserga, 1984). In order to gain more insight into these phenomena it appeared useful to investigate in more detail the 0309-I
651/85/l
10985-08/$03.00/O
@ 1985
Academic
Press
Inc.
(London)
Ltd.
986
Cell Biology
international
Reports,
Vol. 9, No. 11, November
7985
effect of different defined growth factors (in particular those inducing progression) on cellular rRNA content, and furthermore characterize cell lines transformed to different extent with respect to these features. Along these lines we have determined cellular rRNA content (which to 80-90% represents rRNA) in different states of proliferation of Swiss 3T3 cells and Swiss 3T6 cells, a spontaneously transformed cell line which, however, still exhibits several characteristics of normal proliferation control (Gl-arrest, Ca-dependent growth, etc.). Our results are consistent with competence and progression factors contributing in different modes to accumulation of cellular rRNA content to some threshold, above which 3T3 cells are committed for entry into the cell cycle. According to this interpretation, our data on 3T6 cells indicate that their proliferation behaviour is largely independent of a limiting rRNA content. An abstract on part of this work appeared recently (Kleuser and Adam, 1985) . MATERIALS AND METHODS Cells. Swiss 3T3 and Swiss 3T6 were obtained from Flow Laboratories (Bonn-Bad Godesberg, FRG). Cells were propagated as described previously (Seuwen et al., 1984). Details regarding experimental protocols are given in the legends to the figures. Epidermal growth factor (EGF) was obtained from Collaborative Research , Lexington, Massachusetts, insulin from Sigma, Taufkirchen (FRG), and the calf sera from Gibco Europe, Karlsruhe (FRG). Population-Dynamical Analyses. Cell densities were determined after trypsinization using a Coulter counter (Seuwen et al., 1984). They are reproducible within 5%. Fractions of cells stimulated to proliferation from the quiescent state were evaluated using the 5bromodeoxyuridine/Hoechst 33258 method (Rabinovitch, 1983) employing 30 uM 5-bromodeoxyuridine in the presence of 30 IJM deoxycytidine. Staining and flow-cytometric analysis of Hoechst 33258 fluorescence was performed as detailed previously (Seuwen et al., 1984). Determination of Cellular RNA-Content. After alkaline extrac1977), cellular RNA contents were determined tion (Thilly et al., using the two-wavelength method of W-absorption (Fleck and Begg, Since 80 to 90% oftotalcellular RNA is rRNA, these measure1965). ments were calibrated using rRNA from rabbit liver (Sigma, Taufkirchen, FRG). For the protein standard, bovine serum albumin (Serva, Heidelberg, FRG) was used. RESULTS Dependence of Cellular RNA on Cell Density. The dependence of cellular RNA content on cell density has been characterized previously,measuring flow-cytometrically the red fluorescence of 3T3 cells stained by acridine orange (Adam et al., 1983; Seuwen and Adam, 1983;
Cell Biology
International
Reports,
Vol. 9, No. 11, November
1985
987
imeldays-
Growth curves Fig.1. Upper: lar RNA-content versus time calf serum. Right: 3T6 cells
with daily medium renewal. Lower: Celluof growth. Left: 3T3 cells at 5% new-born at 1% new-born calf serum.
‘“OF
C-
MosE 3
E*l
FCS
75
E
1
E*l
C’
xARNA 50 tt
I,
I
1
E.3
FCS
c-
25 0
PI MOS
E
1
E
1
E*l
Responses of 3T3 cells relative to unstimulated control after Fig.2. stimulation with different combinations of growth factors. Upper: percent of cells stimulated to divide (at 40-47 h after start of stimulation).Lower: percent increase of cellular RNA (at 15 h after start of stimulation). Cells were grown to confluency in 2.5% new-born calf serum (medium renewal every second day) and left without medium renewal for 2-3 days before stimulation with or without competence treatment by 15% fetal calf serum for 4-4.5 h denoted by C+ or C-, respectively, followed by different progression treatments until measurement. Notations for these (with numbers of independent expts., each done in duplicate, in parentheses) are: C+: MOS = medium without (5), E+I = EGF + insulin (5), FCS serum (6), E = EGF (4), I = insulin = 15% fetal calf serum (6); C-: E = EGF (l), I = insulin (3), E+I = EGF + insulin (4). Concentrations: 20 rig/ml EGF, 10-5 M insulin. Given are averages f SEM.
988
Ceil Biology
International
Reports,
Vol. 9, No. 7 1, November
7985
Seuwen et al., 1984). These results have been confirmed and detailed by the present direct method (see Fig-l, left). In addition, analogous data for the spontaneously transformed 3T6 cell line are shown (Fig.1, right), exhibiting also a down-regulation of cellular RNA content which, however, is less pronounced than that for 3T3 cells. Stimulation by Different Combinations of Growth Factors. The results of stimulation of quiescent 3T3 and 3T6 cells by different conbinations of growth factors are shown in Figs. 2 and 3, respectively. The upper parts of these figures show the effect of stimulation in terms of percentage of cells stimulated to divide..The results on 3T3 cells (Fig.2, upper) essentially reproduce the basic phenomena of competence and progression as demonstrated earlier (Scher et al., 1979; Leof et al., 1982; Seuwen and Adam, 1983; Seuwen et al., 1984), indicating that the functions of at least three components of growth stimulation are required for a substantial fraction of 3T3 cells to enter the cell cycle. These are: (i) competence treatment (here effected by a 4-h exposition to 15% fetal calf serum) , (ii) a progression function elicited by epidermal growth factor (EGF) , and (iii) another progression function induced by insulin (at superphysiological concentration to substitute for insulinlike growth factors). In contrast, 3~6 cells are substantially stimulated into proliferation if only EGF is applied; a competence-treatment is not required (Fig.3, upper). In order to gain some insight
11
C’
-I. AN
MO5
E
I
E*l
F
E
1
E*l
Fig.3. Response of 3T6 cells relative to unstimulated control after stimulation with different combinations of growth factors. Upper: percent of cells stimulated to divide (determined 30-40 h after beginning of stimulation). Lower: percent increase of cellular RNA (determined 12 h after stimulation beginning). Cells were grown to confluency in 1% new-born calf serum (daily medium renewal) and left without medium renewal for 2-3 days before start of stimulation with or without competence treatment by 15% fetal calf serum for 2-2,5 h, Given are averages f SEM from 3 denoted by C+ or C-, respectively. independent experiments each done in duplicate. Further notations and concentrations are as given in the legend to Fig.1.
Cell Biology
loot %aN
International
Reports,
$ 3T3
Vol. 9, No. 11, November
1985
100
3T6
%h N
;t 50 -
989
Q’ 50-
4t 7 pg RNA/cell * “mn”al
0
0
40
7-+-&, 80,)
,
,
,
120
Plot of percent cells stimulated to divide versus cellular Fig.4. RNA content according to the data given in Figs. 2 and 3. Left: 3T3 Error bars indicate SEM. Progression treatcells. Right: 3T6 cells. ments are for experiments with competence: 0, 15% fetal calf serum; o, EGF+insulin; A, insulin; 0, EGF; 0, medium without serum; without compentence: o, EGF+insulin; &insulin; m, EGF; without control before stimulation. treatment:@, into the intracellular effects of these distinct components of stimulation of proliferation, we have determined cellular RNA contents for these same regimes of stimulation. The results are shown in the lower parts of Fig. 2 and 3. It is obvious that all the components of eventual stimulation of quiescent 3T3 cells into the cell cycle contribute, albeit in different modes, to a rise of cellular RNA. For both cell lines a treatment by only EGF or competence factors results only in a minor increase of cellular RNA. However, a competence pretreatment greatly sensitizes both cell lines for the effects of EGF and/or insulin on cellular RNA. With regard to stimulation of RNA metabolism, both cell lines thus react very similar to different combinations of growth factors. However, the relation between cellular RNA contents and percent of cells entering the cell cycle turns out drastically different. This is shown more clearly in the plots of percent of cells stimulated to divide versus cellular RNA content (Figs. 4 and 5). Fig.4 gives the data from Figs. 2 and 3 in that reduced form. For 3T3 cells this characteristic is strongly non-linear, indicating a threshold behaviour (Fig-Q, left). The corresponding characteristic for the spontaneously transformed 3T6 cell line is entirely different: a substantial stimulation into proliferation is found without an appreciable rise of cellular RNA. In order to ascertain whether these characteristics for 3T3 and 3T6 cells depend on the particular (and somewhat artificial) mode of progression induction by the defined growth factors EGF and/or insulin, we have replaced these progression treatments by analogous treatments employing a graded series of concentrations of fetal calf serum. The results of these experiments in terms of the characteristic of
990
Ceil Biology
International
100
=T
%A N
3T3
Reports,
Vol. 9, No. 11, November
1985
100
3T6
%AN
I
50 -
50 -
I
pg RNA/cell
pg RNA
--
/cell e
+/
120 80 -f Fig.5. Plot of percent cells stimulated to divide versus cellular RNA-content according to experiments with competence treatments as in Figs. 2-4, and progression treatments by different concentrations of fetal calf serum (in percent): l , 15; o, 10; A , 5; A, 2.5; I, 1; R , 0.5; q , 0.3 (3T3); 0, 0.1 (3T6), H, medium without serum; QJ, before stimulation. Left: 3T3 cells. Right: 3T6 cells. Given are averages & SEM from two independent experiments for each cell line. 0
0
,‘l-
“I’
40
+
‘80
Oo
120
40
stimulation versus cellular RNA are shown in Fig.5 and found to be very similar to those on the progression induced by defined growth Again the results for 3T6 cells differ characteristically factors. from those for 3T3 cells. DISCUSSION Although we have employed very different regimes of stimulation of 3T3 cells (i.e. with or without competence induction, progression treatment by different defined growth factors or by serum), the resulting relation between percentage of cells stimulated to divide and cellular rRNA content is remarkably simple (see Figs. 4, left a strongly non-linear characteristic indicating a and 5, left): Very similar results have been obtained for threshold.behaviour. human diploid fibrobalsts IMR90 (A. Simm and G. Adam, to be published elsewhere) and rat embryo fibroblasts (Adam, Braun and Kleuser, to These results appear very interesting in be published elsewhere). the context of models of the mammalian cell cycle stipulating only as governing the initiation of the cycle. one parameter, cell size, In fact, the single-valued dependence of stimulation of cell division on cellular rRNA content for normal cells suggests that a single rRNA) yields a sufficient parameter proportional to cell size (viz. description. This interpretation does appear to give a specification the competence/progression ccnthan a contradiction to, of, rather as the mode of contribution of competence cept (Scher et al., 1979), induction to a rise in cellular rRNA is entirely different from that and the rise of proliferaof progression (in fact depends on it), tion stimulation with rRNA above threshold is fairly steep. Contributions by competence factors or RGF to proliferation stimulation
Cell Biology
International
Reports,
Vol. 9, No. 1 I, November
1985
991
contribution to regulain addition to, and independent of, their tion via cellular rRNA-content clearly can not be excluded and are, suggested by the observed regulation of 3T6 proliferation in fact, by EGF independent of cellular rRNA content (Figs. 3, 4 right, and 5 right). The distinct characteristica of 3T6 cells are of considerable interest as 3T6 cells share several features of "normal" proliferation control with 3T3 cells: Gl-arrest (Kleuser et al., 1985), Ca-dependent proliferation (van der Bosch et al., 1979; Kleuser and Adam, unpubl.), and cell-density dependence of rRNA content (Fig.1). The present results on 3T6 cells can be summarized by stating the following features (as different to 3T3): (i) proliferation is independent of competence induction (" competence is constitutive", Figs. 3 and 4,5 right), (ii) proliferation does not require transgression of a threshold of cellular rRNA, and (iii) proliferation is still controlled by supply of EGF. It is of interest to note that these cells still exhibit cell-density dependent inhibition of growth, although their requirement for competence induction (usually regarded as the growth-limiting condition) is lost. Some of these features of 3T6 cells correspond closely to those reported for a class of spontaneously or chemically transformed cells (Pardee, 1982). Our data on 3T6 cells suggest that the examples of cell lines exhibiting entry into the cell cycle without substantial increase of cellular RNA (Baserga, 1984) belong to this class of partially transformed cells. ACKNOWLEDGMENT We wish to thank Ms. A. Kesper for excellent technical help with regard to propagation of the stem cultures. This work was supported by grants from Stiftung Volkswagenwerk (Schwerpunkt Synergetik) and Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 156). REFERENCES Adam, G., Steiner, U. and Seuwen, K. (1983) Proliferative activity and ribosomal RNA content of 3T3 and SV40-3T3 cells. Cell Biology International Reports 7-, 955-962. Baserga, R. (1976) Multiplication and Division in Mammalian Cells. Marcel Dekker, New York-Basel. Baserga, R. (1981) Introduction to Cell Growth: Growth in Size and DNA Replication. In: R. Baserga (ea.) Tissue Growth Factors, pp. l-12. Springer, Berlin. Baserga, R. (1982) Protein and RNA Synthesis. In: C. Nicolini (ea.) Cell Growth, pp. 337-345. Plenum Press, New York. Baserga, R. (1984) Growth in size and cell DNA replication. Experimental Cell Research 151 l-5. -'
992
Cell Biology
International
Reports,
van
Vol. 9, No.
11, November
1985
der Bosch, J., Sommer, I., Maier, H. and Rahmig, W. (1979) Density-dependent growth adaptation kinetics in 31'3 cell populations following sudden [Ca2+] and temperature changes. Zeitschrift fiir Naturforschung z, 279-283. Cooper, S. (1982) The Continuum Model: Application of Gl-Arrest and In: C. Nicolini G(o). (ea.) Cell Growth, pp. 315-336. Plenum Press, New York. Fleck, A. and Begg, D. (1965) The estimation of ribonucleic acid using uitraviolet absorption measurements. Biochimica Biophysica Acta 108, 333.339. Hertwig, 0. (1908) Neue Probleme der Zellenlehre. Archiv fur Zellforschung I, l-32. Kleuser, B. and Adam, G. (1985 ) Characteristics of the progression signal sufficient for stimulation of proliferation-competent 3T3 cells. European Journal of Cell Biology 36, Suppl. 7, 34. Kleuser, B., Rieter, H. and Adam, G. (1985) Seleczve effects by valinomycin and cytotoxicity and cell cycle arrest of transformed versus non-transformed rodent fibroblasts in vitro. Cancer Research 45, in press. Leof, E.B., Wharton, W., Van Wyk, J.J. and Pledger, W.J. (1982) Epidermal growth factor (EGF) and somatomedin C regulate Gl progression in competent Balb/c 3T3 cells. Experimental Cell Research 141, 107-115. Pardee, A.B. (1982) Molecular Mechanisms of the control of cell growth in cancer. In: C. Nicolini (ea.) Cell Growth, pp. 673714, Plenum Press, New York. Prescott, D.M. (1976) Reproduction of Eukaryotic Cells. Academic Press, New York. Prescott, D.M. (1982) Initiation of DNA Synthesis and Progression (ea.) Cell Growth, pp. through the S Period. In: C. Nicolini 355-364. Plenum Press, New York. Rabinovitch, P.S. (1983) Regulation of human fibroblast growth rate by both non-cycling cell fraction and transision probability is shown by growth in 5-bromodeoxyuridine followed by Hoechst 33258 flow cytometry. Proceedings of the National Academy of Sciences USA 80, 2951-2955. Scher, C.D., Shepard, R.C., Antoniades, H.N. and Stiles, C.D. (1979) Platelet,derived growth factor and the regulation of the mamBiochimica Biophysics Acta 560, malian fibroblast cell cycle. 217-241. Seuwen, K. and Adam, G. (1983) Only one of the signals required for initiation of the cell cycle is associated with cellular accumulation of ribosomal RNA. Biochemical and Biophysical Research Communications 117, 223-230. Cellular content of Seuwen, K., Steiner, U. and Adam, G. (1984) ribosomal RNA in relation to the progression and competence signals governing proliferation of 3T3 and SV40-3T3 cells. EXperimental Cell Research 154, 10-23. Thilly, W.S., Arkin, D.I. and Wogan, G.N. (1977) Nucleic acid content of HeLa-S3 cells during the cell cycle: variations between cycles. Cell and Tissue Kinetics 10, 81-88.
Received:
29.7.85
Accepted:
20.8.85
International
Interrelation of the Cell
Reports,
Vol. 9, No.
between Cellular Cycle of Normal and Beate
Kleuser
and
11, November
rRNA Content Transformed Gerold
1985
985
and Regulation Mouse Cell Lines
Adam
FakultZt fiir Biologie Universitat Konstanz D-7750 Konstanz, FRG ABSTRACT The relation between cellular rRNA content, as a measure of cell size, and the regulation of the cell cycle has been investigated for Swiss 3T3 and the spontaneously transformed Swiss 3T6 cell line. It is shown that the characteristic of percent of quiescent cells stimulated into the cell cycle versus cellular rRNA content is basically different for 3T3 and 3T6 cells: 3T3 cells do not enter the cell cycle below a certain threshold of cellular rR.NA content, whereas 3T6 cells start proliferation without a substantial increase of rRNA. These data are interpreted as consistent with transformation of 3T6 cells being in essence their uncoupling from the requirement of normal cells of passing over a threshold of cellular rRNA content (cell size) before initiating DNA-replication. INTRODUCTION The concept of cell size requirements governing the cell cycle of eukaryotic cells has been attractive since its inception by Hertwig (1908). Modern models of the mammalian cell cycle postulate a continuous growth cycle superimposed to, and sufficient for the regulation of, the chromosome cycle (Cooper, 1982; Prescott, 1982). The somewhat protean notion of "cell size" may be specified in terms of cellular rRNA content as a relevant parameter describing cellular growth (Prescott, 1976; Baserga, 1976, 1981, 1982). More recent data on the relation between cellular rRNA content and regulation of mammalian cell proliferation appeared to contradict the basic concept that the single parameter of cell size (viz. cellular rRNA) governs entry into the cell cycle. Following the dissection of specific growth factor requirements of initiation of the fibroblast cell cycle in terms of "competence" and "progression" (Scher et al., 1979), it was shown that progression, but not induction of competence leads to an increase of cellular rRNA content (Seuwen and Adam, 1983; Adam et al., 1983; Seuwen et al., 1984). Consequently, an increase of cell size (viz. rRNA) only did not appear sufficient for entry into the cell cycle. Furthermore, in a recent review of pertinent data, quite a number of cellular systems were compiled, showing entry into the cell cycle without previous accumulation of cellular rRNA (Baserga, 1984). In order to gain more insight into these phenomena it appeared useful to investigate in more detail the 0309-I
651/85/l
10985-08/$03.00/O
@ 1985
Academic
Press
Inc.
(London)
Ltd.
986
Cell Biology
international
Reports,
Vol. 9, No. 11, November
7985
effect of different defined growth factors (in particular those inducing progression) on cellular rRNA content, and furthermore characterize cell lines transformed to different extent with respect to these features. Along these lines we have determined cellular rRNA content (which to 80-90% represents rRNA) in different states of proliferation of Swiss 3T3 cells and Swiss 3T6 cells, a spontaneously transformed cell line which, however, still exhibits several characteristics of normal proliferation control (Gl-arrest, Ca-dependent growth, etc.). Our results are consistent with competence and progression factors contributing in different modes to accumulation of cellular rRNA content to some threshold, above which 3T3 cells are committed for entry into the cell cycle. According to this interpretation, our data on 3T6 cells indicate that their proliferation behaviour is largely independent of a limiting rRNA content. An abstract on part of this work appeared recently (Kleuser and Adam, 1985) . MATERIALS AND METHODS Cells. Swiss 3T3 and Swiss 3T6 were obtained from Flow Laboratories (Bonn-Bad Godesberg, FRG). Cells were propagated as described previously (Seuwen et al., 1984). Details regarding experimental protocols are given in the legends to the figures. Epidermal growth factor (EGF) was obtained from Collaborative Research , Lexington, Massachusetts, insulin from Sigma, Taufkirchen (FRG), and the calf sera from Gibco Europe, Karlsruhe (FRG). Population-Dynamical Analyses. Cell densities were determined after trypsinization using a Coulter counter (Seuwen et al., 1984). They are reproducible within 5%. Fractions of cells stimulated to proliferation from the quiescent state were evaluated using the 5bromodeoxyuridine/Hoechst 33258 method (Rabinovitch, 1983) employing 30 uM 5-bromodeoxyuridine in the presence of 30 IJM deoxycytidine. Staining and flow-cytometric analysis of Hoechst 33258 fluorescence was performed as detailed previously (Seuwen et al., 1984). Determination of Cellular RNA-Content. After alkaline extrac1977), cellular RNA contents were determined tion (Thilly et al., using the two-wavelength method of W-absorption (Fleck and Begg, Since 80 to 90% oftotalcellular RNA is rRNA, these measure1965). ments were calibrated using rRNA from rabbit liver (Sigma, Taufkirchen, FRG). For the protein standard, bovine serum albumin (Serva, Heidelberg, FRG) was used. RESULTS Dependence of Cellular RNA on Cell Density. The dependence of cellular RNA content on cell density has been characterized previously,measuring flow-cytometrically the red fluorescence of 3T3 cells stained by acridine orange (Adam et al., 1983; Seuwen and Adam, 1983;
Cell Biology
International
Reports,
Vol. 9, No. 11, November
1985
987
imeldays-
Growth curves Fig.1. Upper: lar RNA-content versus time calf serum. Right: 3T6 cells
with daily medium renewal. Lower: Celluof growth. Left: 3T3 cells at 5% new-born at 1% new-born calf serum.
‘“OF
C-
MosE 3
E*l
FCS
75
E
1
E*l
C’
xARNA 50 tt
I,
I
1
E.3
FCS
c-
25 0
PI MOS
E
1
E
1
E*l
Responses of 3T3 cells relative to unstimulated control after Fig.2. stimulation with different combinations of growth factors. Upper: percent of cells stimulated to divide (at 40-47 h after start of stimulation).Lower: percent increase of cellular RNA (at 15 h after start of stimulation). Cells were grown to confluency in 2.5% new-born calf serum (medium renewal every second day) and left without medium renewal for 2-3 days before stimulation with or without competence treatment by 15% fetal calf serum for 4-4.5 h denoted by C+ or C-, respectively, followed by different progression treatments until measurement. Notations for these (with numbers of independent expts., each done in duplicate, in parentheses) are: C+: MOS = medium without (5), E+I = EGF + insulin (5), FCS serum (6), E = EGF (4), I = insulin = 15% fetal calf serum (6); C-: E = EGF (l), I = insulin (3), E+I = EGF + insulin (4). Concentrations: 20 rig/ml EGF, 10-5 M insulin. Given are averages f SEM.
988
Ceil Biology
International
Reports,
Vol. 9, No. 7 1, November
7985
Seuwen et al., 1984). These results have been confirmed and detailed by the present direct method (see Fig-l, left). In addition, analogous data for the spontaneously transformed 3T6 cell line are shown (Fig.1, right), exhibiting also a down-regulation of cellular RNA content which, however, is less pronounced than that for 3T3 cells. Stimulation by Different Combinations of Growth Factors. The results of stimulation of quiescent 3T3 and 3T6 cells by different conbinations of growth factors are shown in Figs. 2 and 3, respectively. The upper parts of these figures show the effect of stimulation in terms of percentage of cells stimulated to divide..The results on 3T3 cells (Fig.2, upper) essentially reproduce the basic phenomena of competence and progression as demonstrated earlier (Scher et al., 1979; Leof et al., 1982; Seuwen and Adam, 1983; Seuwen et al., 1984), indicating that the functions of at least three components of growth stimulation are required for a substantial fraction of 3T3 cells to enter the cell cycle. These are: (i) competence treatment (here effected by a 4-h exposition to 15% fetal calf serum) , (ii) a progression function elicited by epidermal growth factor (EGF) , and (iii) another progression function induced by insulin (at superphysiological concentration to substitute for insulinlike growth factors). In contrast, 3~6 cells are substantially stimulated into proliferation if only EGF is applied; a competence-treatment is not required (Fig.3, upper). In order to gain some insight
11
C’
-I. AN
MO5
E
I
E*l
F
E
1
E*l
Fig.3. Response of 3T6 cells relative to unstimulated control after stimulation with different combinations of growth factors. Upper: percent of cells stimulated to divide (determined 30-40 h after beginning of stimulation). Lower: percent increase of cellular RNA (determined 12 h after stimulation beginning). Cells were grown to confluency in 1% new-born calf serum (daily medium renewal) and left without medium renewal for 2-3 days before start of stimulation with or without competence treatment by 15% fetal calf serum for 2-2,5 h, Given are averages f SEM from 3 denoted by C+ or C-, respectively. independent experiments each done in duplicate. Further notations and concentrations are as given in the legend to Fig.1.
Cell Biology
loot %aN
International
Reports,
$ 3T3
Vol. 9, No. 11, November
1985
100
3T6
%h N
;t 50 -
989
Q’ 50-
4t 7 pg RNA/cell * “mn”al
0
0
40
7-+-&, 80,)
,
,
,
120
Plot of percent cells stimulated to divide versus cellular Fig.4. RNA content according to the data given in Figs. 2 and 3. Left: 3T3 Error bars indicate SEM. Progression treatcells. Right: 3T6 cells. ments are for experiments with competence: 0, 15% fetal calf serum; o, EGF+insulin; A, insulin; 0, EGF; 0, medium without serum; without compentence: o, EGF+insulin; &insulin; m, EGF; without control before stimulation. treatment:@, into the intracellular effects of these distinct components of stimulation of proliferation, we have determined cellular RNA contents for these same regimes of stimulation. The results are shown in the lower parts of Fig. 2 and 3. It is obvious that all the components of eventual stimulation of quiescent 3T3 cells into the cell cycle contribute, albeit in different modes, to a rise of cellular RNA. For both cell lines a treatment by only EGF or competence factors results only in a minor increase of cellular RNA. However, a competence pretreatment greatly sensitizes both cell lines for the effects of EGF and/or insulin on cellular RNA. With regard to stimulation of RNA metabolism, both cell lines thus react very similar to different combinations of growth factors. However, the relation between cellular RNA contents and percent of cells entering the cell cycle turns out drastically different. This is shown more clearly in the plots of percent of cells stimulated to divide versus cellular RNA content (Figs. 4 and 5). Fig.4 gives the data from Figs. 2 and 3 in that reduced form. For 3T3 cells this characteristic is strongly non-linear, indicating a threshold behaviour (Fig-Q, left). The corresponding characteristic for the spontaneously transformed 3T6 cell line is entirely different: a substantial stimulation into proliferation is found without an appreciable rise of cellular RNA. In order to ascertain whether these characteristics for 3T3 and 3T6 cells depend on the particular (and somewhat artificial) mode of progression induction by the defined growth factors EGF and/or insulin, we have replaced these progression treatments by analogous treatments employing a graded series of concentrations of fetal calf serum. The results of these experiments in terms of the characteristic of
990
Ceil Biology
International
100
=T
%A N
3T3
Reports,
Vol. 9, No. 11, November
1985
100
3T6
%AN
I
50 -
50 -
I
pg RNA/cell
pg RNA
--
/cell e
+/
120 80 -f Fig.5. Plot of percent cells stimulated to divide versus cellular RNA-content according to experiments with competence treatments as in Figs. 2-4, and progression treatments by different concentrations of fetal calf serum (in percent): l , 15; o, 10; A , 5; A, 2.5; I, 1; R , 0.5; q , 0.3 (3T3); 0, 0.1 (3T6), H, medium without serum; QJ, before stimulation. Left: 3T3 cells. Right: 3T6 cells. Given are averages & SEM from two independent experiments for each cell line. 0
0
,‘l-
“I’
40
+
‘80
Oo
120
40
stimulation versus cellular RNA are shown in Fig.5 and found to be very similar to those on the progression induced by defined growth Again the results for 3T6 cells differ characteristically factors. from those for 3T3 cells. DISCUSSION Although we have employed very different regimes of stimulation of 3T3 cells (i.e. with or without competence induction, progression treatment by different defined growth factors or by serum), the resulting relation between percentage of cells stimulated to divide and cellular rRNA content is remarkably simple (see Figs. 4, left a strongly non-linear characteristic indicating a and 5, left): Very similar results have been obtained for threshold.behaviour. human diploid fibrobalsts IMR90 (A. Simm and G. Adam, to be published elsewhere) and rat embryo fibroblasts (Adam, Braun and Kleuser, to These results appear very interesting in be published elsewhere). the context of models of the mammalian cell cycle stipulating only as governing the initiation of the cycle. one parameter, cell size, In fact, the single-valued dependence of stimulation of cell division on cellular rRNA content for normal cells suggests that a single rRNA) yields a sufficient parameter proportional to cell size (viz. description. This interpretation does appear to give a specification the competence/progression ccnthan a contradiction to, of, rather as the mode of contribution of competence cept (Scher et al., 1979), induction to a rise in cellular rRNA is entirely different from that and the rise of proliferaof progression (in fact depends on it), tion stimulation with rRNA above threshold is fairly steep. Contributions by competence factors or RGF to proliferation stimulation
Cell Biology
International
Reports,
Vol. 9, No. 1 I, November
1985
991
contribution to regulain addition to, and independent of, their tion via cellular rRNA-content clearly can not be excluded and are, suggested by the observed regulation of 3T6 proliferation in fact, by EGF independent of cellular rRNA content (Figs. 3, 4 right, and 5 right). The distinct characteristica of 3T6 cells are of considerable interest as 3T6 cells share several features of "normal" proliferation control with 3T3 cells: Gl-arrest (Kleuser et al., 1985), Ca-dependent proliferation (van der Bosch et al., 1979; Kleuser and Adam, unpubl.), and cell-density dependence of rRNA content (Fig.1). The present results on 3T6 cells can be summarized by stating the following features (as different to 3T3): (i) proliferation is independent of competence induction (" competence is constitutive", Figs. 3 and 4,5 right), (ii) proliferation does not require transgression of a threshold of cellular rRNA, and (iii) proliferation is still controlled by supply of EGF. It is of interest to note that these cells still exhibit cell-density dependent inhibition of growth, although their requirement for competence induction (usually regarded as the growth-limiting condition) is lost. Some of these features of 3T6 cells correspond closely to those reported for a class of spontaneously or chemically transformed cells (Pardee, 1982). Our data on 3T6 cells suggest that the examples of cell lines exhibiting entry into the cell cycle without substantial increase of cellular RNA (Baserga, 1984) belong to this class of partially transformed cells. ACKNOWLEDGMENT We wish to thank Ms. A. Kesper for excellent technical help with regard to propagation of the stem cultures. This work was supported by grants from Stiftung Volkswagenwerk (Schwerpunkt Synergetik) and Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 156). REFERENCES Adam, G., Steiner, U. and Seuwen, K. (1983) Proliferative activity and ribosomal RNA content of 3T3 and SV40-3T3 cells. Cell Biology International Reports 7-, 955-962. Baserga, R. (1976) Multiplication and Division in Mammalian Cells. Marcel Dekker, New York-Basel. Baserga, R. (1981) Introduction to Cell Growth: Growth in Size and DNA Replication. In: R. Baserga (ea.) Tissue Growth Factors, pp. l-12. Springer, Berlin. Baserga, R. (1982) Protein and RNA Synthesis. In: C. Nicolini (ea.) Cell Growth, pp. 337-345. Plenum Press, New York. Baserga, R. (1984) Growth in size and cell DNA replication. Experimental Cell Research 151 l-5. -'
992
Cell Biology
International
Reports,
van
Vol. 9, No.
11, November
1985
der Bosch, J., Sommer, I., Maier, H. and Rahmig, W. (1979) Density-dependent growth adaptation kinetics in 31'3 cell populations following sudden [Ca2+] and temperature changes. Zeitschrift fiir Naturforschung z, 279-283. Cooper, S. (1982) The Continuum Model: Application of Gl-Arrest and In: C. Nicolini G(o). (ea.) Cell Growth, pp. 315-336. Plenum Press, New York. Fleck, A. and Begg, D. (1965) The estimation of ribonucleic acid using uitraviolet absorption measurements. Biochimica Biophysica Acta 108, 333.339. Hertwig, 0. (1908) Neue Probleme der Zellenlehre. Archiv fur Zellforschung I, l-32. Kleuser, B. and Adam, G. (1985 ) Characteristics of the progression signal sufficient for stimulation of proliferation-competent 3T3 cells. European Journal of Cell Biology 36, Suppl. 7, 34. Kleuser, B., Rieter, H. and Adam, G. (1985) Seleczve effects by valinomycin and cytotoxicity and cell cycle arrest of transformed versus non-transformed rodent fibroblasts in vitro. Cancer Research 45, in press. Leof, E.B., Wharton, W., Van Wyk, J.J. and Pledger, W.J. (1982) Epidermal growth factor (EGF) and somatomedin C regulate Gl progression in competent Balb/c 3T3 cells. Experimental Cell Research 141, 107-115. Pardee, A.B. (1982) Molecular Mechanisms of the control of cell growth in cancer. In: C. Nicolini (ea.) Cell Growth, pp. 673714, Plenum Press, New York. Prescott, D.M. (1976) Reproduction of Eukaryotic Cells. Academic Press, New York. Prescott, D.M. (1982) Initiation of DNA Synthesis and Progression (ea.) Cell Growth, pp. through the S Period. In: C. Nicolini 355-364. Plenum Press, New York. Rabinovitch, P.S. (1983) Regulation of human fibroblast growth rate by both non-cycling cell fraction and transision probability is shown by growth in 5-bromodeoxyuridine followed by Hoechst 33258 flow cytometry. Proceedings of the National Academy of Sciences USA 80, 2951-2955. Scher, C.D., Shepard, R.C., Antoniades, H.N. and Stiles, C.D. (1979) Platelet,derived growth factor and the regulation of the mamBiochimica Biophysics Acta 560, malian fibroblast cell cycle. 217-241. Seuwen, K. and Adam, G. (1983) Only one of the signals required for initiation of the cell cycle is associated with cellular accumulation of ribosomal RNA. Biochemical and Biophysical Research Communications 117, 223-230. Cellular content of Seuwen, K., Steiner, U. and Adam, G. (1984) ribosomal RNA in relation to the progression and competence signals governing proliferation of 3T3 and SV40-3T3 cells. EXperimental Cell Research 154, 10-23. Thilly, W.S., Arkin, D.I. and Wogan, G.N. (1977) Nucleic acid content of HeLa-S3 cells during the cell cycle: variations between cycles. Cell and Tissue Kinetics 10, 81-88.
Received:
29.7.85
Accepted:
20.8.85