Int. J . Cancer: 47, 929-932 (1991) 0 1991 Wiley-Liss, Inc.
Publication of the International Union Against Cancer Publication de I‘Union Internationale Contre le Cancer
THE EFFECT OF FETAL CALF SERUM ON GROWTH ARREST CAUSED BY ACTIVATORS OF PROTEIN KINASE C Tracey D. BRADSHAW’, Andreas GESCHER’~~ and George R. PETTIT’ ‘Cancer Research Campaign Experimental Chemotherapy Group, Pharmaceutical Sciences Institute, Aston University, Birmingham B47ET, UK, and ’Cancer Research Institute, Arizona State University, Tempe, AZ 85287, USA. The growth of human-derived A549 lung carcinoma cells is inhibited by activators of protein kinase C (PKC) such as 12Otetradecanoylphorbol- 13-acetate (TPA). In this study, the effect of serum deprivation on TPA-induced growth retardation has been investigated. Cells cultured with 10% FCS and TPA (I 0-8 M) stopped growing for 6 days, whereas inhibition of DNA synthesis caused by TPA in cells which were grown in medium containing the serum substitute ultraser lasted for less than 48 hr. The ability of cells to respond to the growthinhibitory potential of T P A decreased with decreasing amounts of FCS in the cellular medium. Addition of fetuin or epidermal growth factor (EGF) to incubates with serumdeprived cells increased the ability of TPA to affect growth, but addition of platelet-derived growth factor (PDGF), transforming growth factor beta (TGF-@)or retinoic acid (RA) was without effect. Growth arrest caused by bryostatin I, another PKC activator, was equally transitory in serum-supplemented and serum-deprived cells. Cytosol of serum-deprived cells contained only 32% of specific phorbol ester binding sites compared to cells grown with FCS; PKC enzyme activity and immunodectable protein were similarly reduced in cells grown without FCS. There was no difference in rate of TPAinduced down-regulation of PKC activity and cytosolic phorbol ester receptor sites between cells grown with or without serum.
Activators of PKC, such as the tumor-promoting phorbol ester TPA, cause a plethora of morphological and biochemical effects in cells (Blumberg, 1980; Sako et al., 1987). Exposure of some cell lines to PKC activators causes mitogenesis, yet in others it leads to inhibition of proliferation, in some cases as a consequence of induction of terminal differentiation (Gescher, 1985). For example, TPA and bryostatin 1, a PKC activator of marine origin and an experimental anti-tumor agent with a novel macrocyclic lactone structure (Pettit et al., 1982), retard the growth of human-derived A549 lung adenocarcinoma cells (Gescher and Reed, 1985; Dale and Gescher, 1989). The biochemical events which trigger the biological effects of PKC activators, for example growth inhibition, are only poorly understood. It appears increasingly likely that the activation of PKC on its own is not sufficient for some of these events to occur. Instead, a cellular response to exposure to phorbol esters seems to result from the interplay of a variety of factors in concert with PKC activation. The initial investigations on the ability of TPA to arrest the growth of A549 cells were conducted using conditions under which cells were maintained in 10% FCS (Gescher and Reed, 1985; Dale and Gescher, 1989). Serum provides cells with nutrients, such as amino acids and growth factors, which generally support proliferation, but certain constituents of serum are known to interfere with growth (Quinggang et al., 1990). In the work described here the hypothesis was tested that the presence of serum affects the influence of TPA and bryostatin 1 on growth rate and PKC activity in A549 cells. The overall objective of our work was to contribute to the understanding of the mechanism(s) through which PKC modulators inhibit growth, and especially of the role of PKC in the mediation of cytostasis. MATERIAL AND METHODS
Chemicals TPA, fetuin and retinoic acid were purchased from Sigma
(Poole, UK), and growth factors from British Biotechnology (Oxford, UK). Bryostatin 1 was isolated and purified as described previously (Pettit et al., 1982). Stock solutions of TPA or bryostatin 1 were prepared in DMSO and stored at - 20°C. The final concentration of DMSO in the medium did not exceed 0.5%, which alone did not affect cell growth. [3H]Phorbol dibutyrate ([3H]PDBu) and [3H]thymidine were obtained from NEN (Stevenage, UK), and ultraser and other tissue culture reagents from Gibco (Paisley, UK). Measurement of cell growth A549 human lung carcinoma cells which originated from the ATCC (Rockville, MD) were routinely cultured in Ham’s F-12 medium supplemented with 10% FCS. Details of the culture conditions have been described by Gescher and Reed (1985). In order to study the effect of FCS on TPA-induced growth modulation, cells were grown and subcultured in medium with 2% ultraser, a defined serum substitute, instead of FCS, for 12 weeks prior to experiments. Concentrations of agents added to incubates were: TPA 10 nM, PDGF, EGF and TGF-P-I 2 and 10 PM, respectively; retinoic acid 1 JLM, fetuin 2 mg/ml. At the end of the incubation period, cells were detached from dishes by trypsinization and counted using a Coulter Counter model ZM. Incorporation of [3H]thymidine into cells was measured as described by Dale and Gescher (1989). Measurement of PKC activity and cytosolic phorbol ester binding sites The cellular cytosolic fraction was prepared by disruption of cells by sonication and centrifugation for 30 min at 100,OOO g. Preparation of cellular cytosol for measurement of PKC activity, separation of enzyme activity by polyacrylamide gel electrophoresis, and PKC assay using protamine sulfate as substrate were conducted as published previously (Dale et al., 1989). Partial purification of cytosolic phorbol ester binding sites, formation of mixed micelles, and measurement of [3H]PDBu binding were performed as described by Hannun and Bell (1986). Binding occurred in the presence of phos) 50 n~ phatidylserine (20%) and free Ca2+ (0.2 m ~ with [3H]PDBu (specific activity: 13.2 CilmM) to which 50 JLM unlabelled PDBu was added for the determination of nonspecific binding. Phorbol ester binding determined in this way reflects cytosolic PKC activity. Radioactivity was counted using a Hewlett-Packard CA2000 scintillation counter. Nonspecific binding was between 7 and 25% of total binding. Values given in “Results” or shown in the Figures and in
3To whom correspondence and requests for reprints should be addressed. Abbreviations: AFP, alpha-fetoprotein; EGF, epidermal growth factor; TGF, transforming growth factor; FCS, fetal calf serum; PDGF, plateletderived growth factor; PDBu, phorbol dibutyrate; TPA, 12-0tetradecanoylphorbol-13-acetate.
Received: October 22, 1990 and in revised form December 24, 1990.
930
BRADSHAW ET A L .
*
Table I are the mean SD of 3 experiments, each performed in triplicate, unless otherwise stated. RESULTS
On incubation with TPA, the growth of A549 cells cultured in medium with 10% FCS was arrested (Fig. la). Growth inhibition lasted for 6 days, after which cells resumed growth at the same rate as control cells (Gescher and Reed, 1985). In contrast, growth arrest caused by TPA in cells maintained in medium with ultraser instead of FCS was much more transient (Fig. lb). The doubling time of A549 cells grown with ultraser was 3 6 4 8 hr, compared to 24 hr when cells were cultured with FCS. The presence of TPA did not appear to decrease the growth rate of cells in serum-free medium. However, DNA synthetic activity in these cells was potently arrested after incubation with TPA for 20 hr, after which it returned gradually to control levels (Fig. 1b). This short duration of inhibition of DNA synthesis explains why TPA hardly affected the doubling time of serum-deprived cells. Similarly, an effect of TPA on the morphology of A549 cells, which was dramatic in cells grown with FCS, was apparent in cells cultured without serum only during the initial 18 hr of exposure to TPA but not after 48 hr (result not shown). In order to investigate further the role of serum on the
A
'"I \ 6ol \ 80
40
+
,
OL
0
growth inhibition caused by TPA, cells were cultured in the presence of TPA with FCS at different concentrations, or with 10% newborn calf serum instead of FCS. DNA synthetic activity was measured after 2 days' exposure to TPA. Figure 2a shows that the inhibitory effect of TPA decreased with decreasing concentration of FCS. In the presence of newborn calf serum, TPA inhibited DNA synthesis by 48 -+ 2%, as compared to 86% in the case of 10% FCS. Addition of FCS (10%) to incubates of cells cultured in medium containing ultraser partially restored the inhibitory efficacy of TPA (Fig. 2b). In an attempt to identify the constituents of FCS which contributed to the growth-inhibitory effect of TPA, we supplemented cellular medium containing ultraser with factors (PDGF, EGF, TGF-P-1, fetuin, retinoic acid) which could be either absent from the serum substitute or present in amounts insufficient to affect TPA-induced growth inhibition. Addition to the medium of EGF (Table I) or fetuin (Figs. l b and 3) partially restored TPA-induced growth inhibition. Supplementation of medium with fetuin and EGF together made cells even more responsive to the effect of TPA (result not shown). Bryostatin 1 exerts only a very short-lived cytostatic effect on cells grown in medium with serum (Dale and Gescher, 1989). The time course of inhibition of DNA synthesis caused by bryostatin 1 in serum-deprived cells was very similar to that observed in cells which were cultured in serum-supplemented medium (Fig. lc), which contrasts sharply with the observations made in the case of TPA. Addition of fetuin to serumdeprived cells did not alter the pattern of bryostatin-induced growth inhibition. Furthermore, the partial restoration of TPAinduced growth retardation by fetuin was abolished when cells were incubated with TPA in the presence of bryostatin 1 (10 nM) (result not shown). This result is consistent with the ability of the bryostatins at high concentrations to abolish the growth arrest caused by phorbol esters in cells maintained with serum (Dale and Gescher, 1989). We tested the hypothesis that the change in responsiveness of cells to the growth-arresting properties of TPA on serum deprivation is related to changes in PKC levels. To that end we measured cytosolic phorbol ester binding sites after partial purification of PKC and PKC enzyme activity. The amount of phorbol dibutyrate bound per mg protein was 9.6 1.2 pmol in cells maintained in FCS and 3.0 0.6 pmol in cells cultured for 12 weeks in medium with ultraser. Total cellular PKC activity was similarly reduced. When cells grown in medium with ultraser fortified with fetuin for 14 days, cytosolic PDBu binding was slightly increased to 3.7 2 0.1 pmoVmg protein. In contrast to the differences in PKC activity and number of phorbol ester binding sites between naive cells grown in me-
*
*
8000
-
a: 0-
4
6000 -
@ 4000
-
$ 0 2000
-
;b Z a
sh I
0
10
I
I
I
I
20
30
40
50
L
,L
I
INCUBATION TIME (hr)
FIGURE1 - Time-course of the effect of TPA (a.b) or bryostatin 1 (c) (both 10 m) on [3H]thymidine incorporation into cells cultured with FCS (10%) (a, open squares in c) or with ultraser (2%) (b, filled diamonds in c), and in the presence or absence of fetuin ( 2 mg/ml) (b). Cells were seeded at 1-2 X 105/well, TPA or bryostatin 1 were added 4 hr later.
0'
10
5
1
0
0
10
FCS CONCENTRATION (%)
FIGURE 2 - Effect of different concentrations of FCS on inhibition by TPA (10 nM) of [3H]thymidine incorporation into cells grown in medium with FCS (10%)(a) or ultraser (2%) (b). Cells were seeded at 105/well. DNA synthetic activity was measured 48 hr after addition to the cells of TPA with or without FCS.
SERUM AUGMENTS TPA-I?VDUCED GROWTH ARREST
TABLE I - EFFJXT OF EGF ON GROWTH RETARDATION INDUCED BY TPA (10 nM) IN A549 CELLS GROWN IN MEDIUM SUPPLEMENTED WITH
ULTRASER (2%) Cell number
Treatment’
(% of control)
TPA
12.1 4 2.2 1.0 5.0
14.6 4 49.0 -t 103.4 48.1 4
EGF ( 2 p ~ ) EGF ( 2 p ~ )+ TPA EGF (1OPM) EGF ( 1 0 ~ + ~ )TPA
* 6.6
4.8
‘TPA and/or EGF were added 4 hr after cell seeding (2 X 104/dish). After 2 days, medium with TPNEGF was replenished and cells were counted after 4 days.
TPA TPA
0
+ FETUIN
0
; 2
w
0
“t 0
’ 0
I
I
I
I
I
J
1
2
3
4
5
6
INCUBATION TIME (days)
FIGURE3 - Time-course of the effect of fetuin (2 mgiml) on TPAinduced growth retardation in cells cultured in medium with ultraser (2%).Cells were seeded at 104/well. Results are expressed as percentage compared to number of control cells.
dium with either serum or ultraser, extent and rate of TPAinduced translocation of PKC activity to the particulate fraction and down-regulation of PKC and of cytosolic phorbol ester receptor sites were not affected by the absence of serum. PKC was also detected immunologically by a monoclonal antibody recognizing the cleavage domain of PKCa- and P-isozymes, and in the presence or absence of serum-TPA-induced changes which mimicked faithfully the alterations caused by TPA when PKC enzyme activity or cytosolic PDBu receptor sites were measured (results not shown). DISCUSSION
Retardation of the growth of certain human-derived cells is possible by exposure to TPA or other activators of PKC. Specific examples include MCF-7 breast carcinoma (Osborne et al., 1981; Roos et al., 1986), HL-60 erythroleukemia (Huberman and Callaham, 1979; Rovera et al., 1979) and A549 lung carcinoma cells (Gescher and Reed, 1985; Dale and Gescher, 1989). In this report we show that TPA-induced effects in A549 cells, inhibition of growth and alteration of cell morphology, are markedly affected by serum constituents. Maximal growth arrest caused by TPA was dependent on the presence of FCS in the cellular growth medium. The observation described here is similar to the growth retardation by TPA seen in 2 lung squamous carcinoma cell lines in the presence of 1% serum, which was, surprisingly, switched to TPA-induced growth stimulation under conditions of total serum deprivation (Sanchez et al., 1987). In contrast to the results obtained in A549 cells, the absence of FCS was found to be inconsequential for the inhibition of proliferation by TPA in MCF-7 cells (Osborne et al., 1981). The PKC activator bryostatin 1 is a much more ephemeral inhibitor of growth of A549 cells than is TPA (Dale and Gescher, 1989), and according to the results presented here the cytostatic potency of bryostatin 1, unlike that of TPA, was equally transient in either serum-supple-
93 1
mented or serum-deprived cells. This difference between bryostatin and TPA is consistent with the results of a number of studies in which subtle discrepancies between the 2 PKC activators have been observed (Kraft et al., 1987, 1989). We considered it possible that the decrease in growtharresting potential of TPA upon serum deprivation might be caused either by the absence of certain growth regulators from ultraser or by their presence at levels insufficient to amplify the effect of TPA. The “negative” growth factor TGF-P appeared to be a likely candidate for a serum constituent which cooperates with TPA in the cytostasis of A549 cells, as it is an effective inhibitor of the growth of A549 cells in vitro (Roberts et al., 1985) and in vivo (Twardzik et al., 1989) and can act synergistically with TPA (Kraft, 1986). However, addition of TGF-P to serum-deprived A549 cells did not augment the cytostatic effect of TPA, whereas supplementation of medium with EGF or fetuin did. Fetuin consists mainly of AFP, the fetal counterpart of serum albumin. Serum activity of AFP is greatly elevated in patients with primary hepatoma and sometimes in the case of malignancies of the stomach and pancreas, but is very low in normal serum (Abelev, 1971). The finding that FCS, which contains AFP, markedly augmented the effect of TPA, whereas newborn calf serum, which contains trace amounts of AFP, was much less effective, is consistent with a crucial role for AFP as an amplifier of the effect of TPA on A549 cells. AFP is known to co-operate with growth regulators such as EGF (Leak et al., 1990) and estradiol (Jacobson et al., 1990). Of course, it is possible that constituents of FCS, in addition to EGF and AFP, are responsible for the increase in TPA-induced growth inhibition in A549 cells. The results of 3 different measurements of PKC, by assay of cytosolic phorbol ester receptors, enzyme activity or immunodetectable protein, demonstrate that PKC in serum-deprived A549 cells constituted only a third of the PKC level observed in cells grown with FCS. This decrease in PKC caused by removal of serum in the culture medium may well be associated with the increase by a factor of almost 2 of the doubling time of the serum-deprived cells as compared to cells maintained with FCS. A link between rate of proliferation and PKC levels in A549 cells is consistent with recent observations according to which over-expression of PKC provoked an enhanced rate of growth in a number of cell lines (Krauss et al., 1989; Housey el al., 1988; Persons et al., 1988). Our results show that it is unlikely that the presence or absence of serum factors alters the susceptibility of the major portion of PKC present in the cell towards enzyme activation or down-regulation by exogenous agents. The decrease in duration of TPA-elicited cytostasis on serum deprivation could perhaps rather be explained in one of the following 2 ways. (i) Serum factors such as EGF and AFP might elicit recruitment of PKC activity or of a particular PKC isozyme from a cellular compartment which is inaccessible to exogenous PKC activators in the absence of FCS. Subsequent activation of this PKC by TPA, but not by bryostatin 1, would then be important for maximal growth inhibition. (ii) Serum factors might help to precipitate processes downstream of and completely unrelated to PKC activation, but they do this only in the presence of phorbol esters. The finding that A549 cells lost most of their PKC on serum deprivation compared to cells cultured with FCS is consistent with possibility (i). Biological effects of phorbol esters other than growth arrest are likely to be modulated by serum constituents. More studies of the interaction between tumor-promoting phorbol esters and serum proteins may lead to an improved understanding of the multi-step pathway of malignant transformation. Furthermore, when PKC modulators undergo clinical evaluation as chemotherapeutic agents, it might be possible and appropriate to exploit a better knowledge of their interaction with constituents in
932
BRADSHAW ET AL.
the tumor environment and thereby allow optimisation of therapeutic strategies. ACKNOWLEDGEMENTS
This work was supported by program grant SP 1518 from the Cancer Research Campaign of Great Britain and a studentship
from the UK Medical Research Council (to TDB). The isolation of bryostatin 1 was funded by PHS grant CA 16049-12, Outstanding Investigator Grant CA-44344-01A1 awarded by the NCI, DHHS, the Fannie E. Rippel Foundation, the Arizona Disease Control Research Commission and the Robert B. Dalton Endowment Fund.
REFERENCES ABELEV,G.I., Alpha-fetoprotein in oncogenesis and its association with malignant tumors. Advanc.Cancer Res., 14, 295-340 (1971). BLUMBERC, P.M., In vitro studies on the mode of action of the phorbol esters, potent tumor promoters. Crit. Rev. Toxicol., 8, 153-234 (1980). DALE,I.L., BRADSHAW, T.D., GESCHER, A. and PETTIT,G.R., Comparison of effects of bryostatins 1 and 2 and 12-0-tetradecanoylphorbol13-acetate on protein kinase C activity in A549 human lung carcinoma cells. Cancer Res., 49, 3242-3245 (1989). DALE,I.L. and GESCHER,A., Effects of activators of protein kinase C including bryostatin 1 and 2 on the growth of A549 human lung carcinoma cells. Int. J . Cancer, 43, 158-163 (1989). GESCHER, A,, Antiproliferative properties of phorbol ester tumor promoters. Biochem. Phnrmacol., 34, 2587-2592 (1985). GESCHER,A. and REED,D.J., Characterization of the growth inhibition induced by tumor-promoting phorbol esters and of their receptor binding in A549 human lung carcinoma cells. Cancer Res., 45,43154321 (1985). HANNUN, Y.A. and BELL,R.M., Phorbol ester binding and activation of protein kinase C on Triton X-100 micelles containing phosphatidylserine. J . biol. Chem., 261, 9341-9347 (1986). HOUSEY, G.M., JOHNSON, M.D., HSIAS,W.L.W., O'BRIAN,C.A., MURPHY, J.P., KIRSCHMEIER, P. and WEINSTEIN, I.B., Overproduction Of protein kinase C causes disordered growth control in rat fibroblasts. Cell, 52, 343-354 (1988). HUBERMAN, E. and CALLAHAM, M.F., Induction of terminal differentiation in human promyelmytic leukemia cells by tumor-promoting agents. Proc. nat. Acad. Sci. (Wash.), 76, 1293-1297 (1979). JACOBSON, H.I., BENNETT,J.A. and MIZEJEWSKI, G.J., Inhibition of estrogen-dependent breast cancer growth by a reaction product of alphafetoprotein and estradiol. Cancer Res., 50, 4 1 5 4 2 0 (1990). KRAFT,A.S., Effect of phorbol esters on the activity of purified transforming growth factors. Cancer Res., 46, 1764-1767 (1986). KRAFT,A . S . , BAKER,V.V. and MAY,W.S., Bryostatin induces changes in Drotein kinase C location and activitv without altering c-mvc gene expre'ssion in human promyelocytic leukthia cells (HL-66). O h G e n e , 1, 111-118 (1987). KRAFT,A.S., WILLIAM,F., PETTIT,G.R. and LILLY,M.B., Varied differentiation responses of human leukemias to bryostatin 1. Cancer Res., 49, 1287-1293 (1989). KRAUSS,R.S., HOUSEY,G.M., JOHNSON,M.D. and WEINSTEIN,I.B., Disturbances in growth control and gene expression in a C3H/T101/2 cell
line that stably overproduces protein kinase. C. Oncogene, 4, 991-998 (1989). LEAK,J.A., MAY,J.V. and KEEL, B.A., Human alpha-fetoprotein enhances epidermal growth factor proliferative activity upon porcine granulosa cells in monolayer culture. Endocrinology, 126, 669-671 (1990). OSBORNE, C.K., HAMILTON, B., NOVER,M. and ZIEGLER,J., Antagonism between epidermal growth factor and phorbol ester tumor promoters in human breast cancer cells. J . din. Invest., 67, 943-951 (1981). PERSONS,D.A., WILKISIN,W.O., BELL, R.M. and FINN,O.J., Altered growth regulation and enhanced tumorigenicity of NIH 3T3 fibroblasts transfected with protein kinase C-I c-DNA. Cell, 52, 447-458 (1988). PETTIT,G.R., HERALD,C.L., DOUBEK,S.L., HERALD,D.L., ARNOLD, J., Isolation and structure of bryostatin 1. J . Amer. chem. E. and CLARDY, Soc.,104, 6846-6848 (1982). QUINGCANC, L.I., BLACHER,R., ESCH,F. and CONGOTE,F., Isolation from fetal bovine serum of an apolipoprotein-H-like protein which inhibits thymidine incorporation in fetal calf erythroid cells. Biochem. J . , 267, 261-264 (1990). ROBERTS,A.B., ANZANO,M.A., WAKEFIELD,L.M., ROCHE,N.S., STERN,D.F. and SPORN,M.B., Type beta transforming growth factor: a bifunctional regulator of cellular growth. Proc. nat. Acad. Sci. (Wash.), 82, 11%123 (1985). ROVERA,G., SANTOLI,D. and DAMSKY, D., Human promyelmytic leukemia cells in culture differentiate into macrophage-like cells when treated with phorbol diester. Proc. nut. Acad. Sci. (Wash.), 75, 2779-2783 (1979). U., ROOS,W., FABBRO, D., KUNG,W., COSTA,S.D. and EPPENBERGER, Correlation between hormone dependency and the regulation of epidermal growth factor receptor by tumor promoters in human mammary carcinoma cells. Proc. nut. Acad. Sci. (Wash.), 83, 991-995 (1986). SAKO,T., YUSPA,S.H., HERALD,C.L., PETTIT,G.R. and BLUMBERG, P.M., Partial parallelism and partial blockade by bryostatin 1 of effects of phorbol ester tumor promoters on primary mouse epidermal cells. Cancer Res., 47, 5445-5450 (1987). T.W., SANCHEZ,J.H., BOREIKO,C.J., FURLONG,J . and HESTERBERG, Differential effects of tumor promoters on the growth of normal human bronchial epithelial cells and human lung tumor cell lines. Toxicol. in Vitro, 1, 183-188 (1987). TWARDZIK,D.R., RANCHALIS, J.E., MCPHERSON, J.M., OGAWA,Y., GENTRY, L., PURCHIO,A,, PLATA,E. and TODARO,G.J., Inhibition and promotion of differentiated-like phenotype of a human lung carcinoma in athymic mice by natural and recombinant forms of transforming growth factor-beta. J . nut. Cancer Inst., 81, 1182-1185 (1989).
Publication of the International Union Against Cancer Publication de I‘Union Internationale Contre le Cancer
THE EFFECT OF FETAL CALF SERUM ON GROWTH ARREST CAUSED BY ACTIVATORS OF PROTEIN KINASE C Tracey D. BRADSHAW’, Andreas GESCHER’~~ and George R. PETTIT’ ‘Cancer Research Campaign Experimental Chemotherapy Group, Pharmaceutical Sciences Institute, Aston University, Birmingham B47ET, UK, and ’Cancer Research Institute, Arizona State University, Tempe, AZ 85287, USA. The growth of human-derived A549 lung carcinoma cells is inhibited by activators of protein kinase C (PKC) such as 12Otetradecanoylphorbol- 13-acetate (TPA). In this study, the effect of serum deprivation on TPA-induced growth retardation has been investigated. Cells cultured with 10% FCS and TPA (I 0-8 M) stopped growing for 6 days, whereas inhibition of DNA synthesis caused by TPA in cells which were grown in medium containing the serum substitute ultraser lasted for less than 48 hr. The ability of cells to respond to the growthinhibitory potential of T P A decreased with decreasing amounts of FCS in the cellular medium. Addition of fetuin or epidermal growth factor (EGF) to incubates with serumdeprived cells increased the ability of TPA to affect growth, but addition of platelet-derived growth factor (PDGF), transforming growth factor beta (TGF-@)or retinoic acid (RA) was without effect. Growth arrest caused by bryostatin I, another PKC activator, was equally transitory in serum-supplemented and serum-deprived cells. Cytosol of serum-deprived cells contained only 32% of specific phorbol ester binding sites compared to cells grown with FCS; PKC enzyme activity and immunodectable protein were similarly reduced in cells grown without FCS. There was no difference in rate of TPAinduced down-regulation of PKC activity and cytosolic phorbol ester receptor sites between cells grown with or without serum.
Activators of PKC, such as the tumor-promoting phorbol ester TPA, cause a plethora of morphological and biochemical effects in cells (Blumberg, 1980; Sako et al., 1987). Exposure of some cell lines to PKC activators causes mitogenesis, yet in others it leads to inhibition of proliferation, in some cases as a consequence of induction of terminal differentiation (Gescher, 1985). For example, TPA and bryostatin 1, a PKC activator of marine origin and an experimental anti-tumor agent with a novel macrocyclic lactone structure (Pettit et al., 1982), retard the growth of human-derived A549 lung adenocarcinoma cells (Gescher and Reed, 1985; Dale and Gescher, 1989). The biochemical events which trigger the biological effects of PKC activators, for example growth inhibition, are only poorly understood. It appears increasingly likely that the activation of PKC on its own is not sufficient for some of these events to occur. Instead, a cellular response to exposure to phorbol esters seems to result from the interplay of a variety of factors in concert with PKC activation. The initial investigations on the ability of TPA to arrest the growth of A549 cells were conducted using conditions under which cells were maintained in 10% FCS (Gescher and Reed, 1985; Dale and Gescher, 1989). Serum provides cells with nutrients, such as amino acids and growth factors, which generally support proliferation, but certain constituents of serum are known to interfere with growth (Quinggang et al., 1990). In the work described here the hypothesis was tested that the presence of serum affects the influence of TPA and bryostatin 1 on growth rate and PKC activity in A549 cells. The overall objective of our work was to contribute to the understanding of the mechanism(s) through which PKC modulators inhibit growth, and especially of the role of PKC in the mediation of cytostasis. MATERIAL AND METHODS
Chemicals TPA, fetuin and retinoic acid were purchased from Sigma
(Poole, UK), and growth factors from British Biotechnology (Oxford, UK). Bryostatin 1 was isolated and purified as described previously (Pettit et al., 1982). Stock solutions of TPA or bryostatin 1 were prepared in DMSO and stored at - 20°C. The final concentration of DMSO in the medium did not exceed 0.5%, which alone did not affect cell growth. [3H]Phorbol dibutyrate ([3H]PDBu) and [3H]thymidine were obtained from NEN (Stevenage, UK), and ultraser and other tissue culture reagents from Gibco (Paisley, UK). Measurement of cell growth A549 human lung carcinoma cells which originated from the ATCC (Rockville, MD) were routinely cultured in Ham’s F-12 medium supplemented with 10% FCS. Details of the culture conditions have been described by Gescher and Reed (1985). In order to study the effect of FCS on TPA-induced growth modulation, cells were grown and subcultured in medium with 2% ultraser, a defined serum substitute, instead of FCS, for 12 weeks prior to experiments. Concentrations of agents added to incubates were: TPA 10 nM, PDGF, EGF and TGF-P-I 2 and 10 PM, respectively; retinoic acid 1 JLM, fetuin 2 mg/ml. At the end of the incubation period, cells were detached from dishes by trypsinization and counted using a Coulter Counter model ZM. Incorporation of [3H]thymidine into cells was measured as described by Dale and Gescher (1989). Measurement of PKC activity and cytosolic phorbol ester binding sites The cellular cytosolic fraction was prepared by disruption of cells by sonication and centrifugation for 30 min at 100,OOO g. Preparation of cellular cytosol for measurement of PKC activity, separation of enzyme activity by polyacrylamide gel electrophoresis, and PKC assay using protamine sulfate as substrate were conducted as published previously (Dale et al., 1989). Partial purification of cytosolic phorbol ester binding sites, formation of mixed micelles, and measurement of [3H]PDBu binding were performed as described by Hannun and Bell (1986). Binding occurred in the presence of phos) 50 n~ phatidylserine (20%) and free Ca2+ (0.2 m ~ with [3H]PDBu (specific activity: 13.2 CilmM) to which 50 JLM unlabelled PDBu was added for the determination of nonspecific binding. Phorbol ester binding determined in this way reflects cytosolic PKC activity. Radioactivity was counted using a Hewlett-Packard CA2000 scintillation counter. Nonspecific binding was between 7 and 25% of total binding. Values given in “Results” or shown in the Figures and in
3To whom correspondence and requests for reprints should be addressed. Abbreviations: AFP, alpha-fetoprotein; EGF, epidermal growth factor; TGF, transforming growth factor; FCS, fetal calf serum; PDGF, plateletderived growth factor; PDBu, phorbol dibutyrate; TPA, 12-0tetradecanoylphorbol-13-acetate.
Received: October 22, 1990 and in revised form December 24, 1990.
930
BRADSHAW ET A L .
*
Table I are the mean SD of 3 experiments, each performed in triplicate, unless otherwise stated. RESULTS
On incubation with TPA, the growth of A549 cells cultured in medium with 10% FCS was arrested (Fig. la). Growth inhibition lasted for 6 days, after which cells resumed growth at the same rate as control cells (Gescher and Reed, 1985). In contrast, growth arrest caused by TPA in cells maintained in medium with ultraser instead of FCS was much more transient (Fig. lb). The doubling time of A549 cells grown with ultraser was 3 6 4 8 hr, compared to 24 hr when cells were cultured with FCS. The presence of TPA did not appear to decrease the growth rate of cells in serum-free medium. However, DNA synthetic activity in these cells was potently arrested after incubation with TPA for 20 hr, after which it returned gradually to control levels (Fig. 1b). This short duration of inhibition of DNA synthesis explains why TPA hardly affected the doubling time of serum-deprived cells. Similarly, an effect of TPA on the morphology of A549 cells, which was dramatic in cells grown with FCS, was apparent in cells cultured without serum only during the initial 18 hr of exposure to TPA but not after 48 hr (result not shown). In order to investigate further the role of serum on the
A
'"I \ 6ol \ 80
40
+
,
OL
0
growth inhibition caused by TPA, cells were cultured in the presence of TPA with FCS at different concentrations, or with 10% newborn calf serum instead of FCS. DNA synthetic activity was measured after 2 days' exposure to TPA. Figure 2a shows that the inhibitory effect of TPA decreased with decreasing concentration of FCS. In the presence of newborn calf serum, TPA inhibited DNA synthesis by 48 -+ 2%, as compared to 86% in the case of 10% FCS. Addition of FCS (10%) to incubates of cells cultured in medium containing ultraser partially restored the inhibitory efficacy of TPA (Fig. 2b). In an attempt to identify the constituents of FCS which contributed to the growth-inhibitory effect of TPA, we supplemented cellular medium containing ultraser with factors (PDGF, EGF, TGF-P-1, fetuin, retinoic acid) which could be either absent from the serum substitute or present in amounts insufficient to affect TPA-induced growth inhibition. Addition to the medium of EGF (Table I) or fetuin (Figs. l b and 3) partially restored TPA-induced growth inhibition. Supplementation of medium with fetuin and EGF together made cells even more responsive to the effect of TPA (result not shown). Bryostatin 1 exerts only a very short-lived cytostatic effect on cells grown in medium with serum (Dale and Gescher, 1989). The time course of inhibition of DNA synthesis caused by bryostatin 1 in serum-deprived cells was very similar to that observed in cells which were cultured in serum-supplemented medium (Fig. lc), which contrasts sharply with the observations made in the case of TPA. Addition of fetuin to serumdeprived cells did not alter the pattern of bryostatin-induced growth inhibition. Furthermore, the partial restoration of TPAinduced growth retardation by fetuin was abolished when cells were incubated with TPA in the presence of bryostatin 1 (10 nM) (result not shown). This result is consistent with the ability of the bryostatins at high concentrations to abolish the growth arrest caused by phorbol esters in cells maintained with serum (Dale and Gescher, 1989). We tested the hypothesis that the change in responsiveness of cells to the growth-arresting properties of TPA on serum deprivation is related to changes in PKC levels. To that end we measured cytosolic phorbol ester binding sites after partial purification of PKC and PKC enzyme activity. The amount of phorbol dibutyrate bound per mg protein was 9.6 1.2 pmol in cells maintained in FCS and 3.0 0.6 pmol in cells cultured for 12 weeks in medium with ultraser. Total cellular PKC activity was similarly reduced. When cells grown in medium with ultraser fortified with fetuin for 14 days, cytosolic PDBu binding was slightly increased to 3.7 2 0.1 pmoVmg protein. In contrast to the differences in PKC activity and number of phorbol ester binding sites between naive cells grown in me-
*
*
8000
-
a: 0-
4
6000 -
@ 4000
-
$ 0 2000
-
;b Z a
sh I
0
10
I
I
I
I
20
30
40
50
L
,L
I
INCUBATION TIME (hr)
FIGURE1 - Time-course of the effect of TPA (a.b) or bryostatin 1 (c) (both 10 m) on [3H]thymidine incorporation into cells cultured with FCS (10%) (a, open squares in c) or with ultraser (2%) (b, filled diamonds in c), and in the presence or absence of fetuin ( 2 mg/ml) (b). Cells were seeded at 1-2 X 105/well, TPA or bryostatin 1 were added 4 hr later.
0'
10
5
1
0
0
10
FCS CONCENTRATION (%)
FIGURE 2 - Effect of different concentrations of FCS on inhibition by TPA (10 nM) of [3H]thymidine incorporation into cells grown in medium with FCS (10%)(a) or ultraser (2%) (b). Cells were seeded at 105/well. DNA synthetic activity was measured 48 hr after addition to the cells of TPA with or without FCS.
SERUM AUGMENTS TPA-I?VDUCED GROWTH ARREST
TABLE I - EFFJXT OF EGF ON GROWTH RETARDATION INDUCED BY TPA (10 nM) IN A549 CELLS GROWN IN MEDIUM SUPPLEMENTED WITH
ULTRASER (2%) Cell number
Treatment’
(% of control)
TPA
12.1 4 2.2 1.0 5.0
14.6 4 49.0 -t 103.4 48.1 4
EGF ( 2 p ~ ) EGF ( 2 p ~ )+ TPA EGF (1OPM) EGF ( 1 0 ~ + ~ )TPA
* 6.6
4.8
‘TPA and/or EGF were added 4 hr after cell seeding (2 X 104/dish). After 2 days, medium with TPNEGF was replenished and cells were counted after 4 days.
TPA TPA
0
+ FETUIN
0
; 2
w
0
“t 0
’ 0
I
I
I
I
I
J
1
2
3
4
5
6
INCUBATION TIME (days)
FIGURE3 - Time-course of the effect of fetuin (2 mgiml) on TPAinduced growth retardation in cells cultured in medium with ultraser (2%).Cells were seeded at 104/well. Results are expressed as percentage compared to number of control cells.
dium with either serum or ultraser, extent and rate of TPAinduced translocation of PKC activity to the particulate fraction and down-regulation of PKC and of cytosolic phorbol ester receptor sites were not affected by the absence of serum. PKC was also detected immunologically by a monoclonal antibody recognizing the cleavage domain of PKCa- and P-isozymes, and in the presence or absence of serum-TPA-induced changes which mimicked faithfully the alterations caused by TPA when PKC enzyme activity or cytosolic PDBu receptor sites were measured (results not shown). DISCUSSION
Retardation of the growth of certain human-derived cells is possible by exposure to TPA or other activators of PKC. Specific examples include MCF-7 breast carcinoma (Osborne et al., 1981; Roos et al., 1986), HL-60 erythroleukemia (Huberman and Callaham, 1979; Rovera et al., 1979) and A549 lung carcinoma cells (Gescher and Reed, 1985; Dale and Gescher, 1989). In this report we show that TPA-induced effects in A549 cells, inhibition of growth and alteration of cell morphology, are markedly affected by serum constituents. Maximal growth arrest caused by TPA was dependent on the presence of FCS in the cellular growth medium. The observation described here is similar to the growth retardation by TPA seen in 2 lung squamous carcinoma cell lines in the presence of 1% serum, which was, surprisingly, switched to TPA-induced growth stimulation under conditions of total serum deprivation (Sanchez et al., 1987). In contrast to the results obtained in A549 cells, the absence of FCS was found to be inconsequential for the inhibition of proliferation by TPA in MCF-7 cells (Osborne et al., 1981). The PKC activator bryostatin 1 is a much more ephemeral inhibitor of growth of A549 cells than is TPA (Dale and Gescher, 1989), and according to the results presented here the cytostatic potency of bryostatin 1, unlike that of TPA, was equally transient in either serum-supple-
93 1
mented or serum-deprived cells. This difference between bryostatin and TPA is consistent with the results of a number of studies in which subtle discrepancies between the 2 PKC activators have been observed (Kraft et al., 1987, 1989). We considered it possible that the decrease in growtharresting potential of TPA upon serum deprivation might be caused either by the absence of certain growth regulators from ultraser or by their presence at levels insufficient to amplify the effect of TPA. The “negative” growth factor TGF-P appeared to be a likely candidate for a serum constituent which cooperates with TPA in the cytostasis of A549 cells, as it is an effective inhibitor of the growth of A549 cells in vitro (Roberts et al., 1985) and in vivo (Twardzik et al., 1989) and can act synergistically with TPA (Kraft, 1986). However, addition of TGF-P to serum-deprived A549 cells did not augment the cytostatic effect of TPA, whereas supplementation of medium with EGF or fetuin did. Fetuin consists mainly of AFP, the fetal counterpart of serum albumin. Serum activity of AFP is greatly elevated in patients with primary hepatoma and sometimes in the case of malignancies of the stomach and pancreas, but is very low in normal serum (Abelev, 1971). The finding that FCS, which contains AFP, markedly augmented the effect of TPA, whereas newborn calf serum, which contains trace amounts of AFP, was much less effective, is consistent with a crucial role for AFP as an amplifier of the effect of TPA on A549 cells. AFP is known to co-operate with growth regulators such as EGF (Leak et al., 1990) and estradiol (Jacobson et al., 1990). Of course, it is possible that constituents of FCS, in addition to EGF and AFP, are responsible for the increase in TPA-induced growth inhibition in A549 cells. The results of 3 different measurements of PKC, by assay of cytosolic phorbol ester receptors, enzyme activity or immunodetectable protein, demonstrate that PKC in serum-deprived A549 cells constituted only a third of the PKC level observed in cells grown with FCS. This decrease in PKC caused by removal of serum in the culture medium may well be associated with the increase by a factor of almost 2 of the doubling time of the serum-deprived cells as compared to cells maintained with FCS. A link between rate of proliferation and PKC levels in A549 cells is consistent with recent observations according to which over-expression of PKC provoked an enhanced rate of growth in a number of cell lines (Krauss et al., 1989; Housey el al., 1988; Persons et al., 1988). Our results show that it is unlikely that the presence or absence of serum factors alters the susceptibility of the major portion of PKC present in the cell towards enzyme activation or down-regulation by exogenous agents. The decrease in duration of TPA-elicited cytostasis on serum deprivation could perhaps rather be explained in one of the following 2 ways. (i) Serum factors such as EGF and AFP might elicit recruitment of PKC activity or of a particular PKC isozyme from a cellular compartment which is inaccessible to exogenous PKC activators in the absence of FCS. Subsequent activation of this PKC by TPA, but not by bryostatin 1, would then be important for maximal growth inhibition. (ii) Serum factors might help to precipitate processes downstream of and completely unrelated to PKC activation, but they do this only in the presence of phorbol esters. The finding that A549 cells lost most of their PKC on serum deprivation compared to cells cultured with FCS is consistent with possibility (i). Biological effects of phorbol esters other than growth arrest are likely to be modulated by serum constituents. More studies of the interaction between tumor-promoting phorbol esters and serum proteins may lead to an improved understanding of the multi-step pathway of malignant transformation. Furthermore, when PKC modulators undergo clinical evaluation as chemotherapeutic agents, it might be possible and appropriate to exploit a better knowledge of their interaction with constituents in
932
BRADSHAW ET AL.
the tumor environment and thereby allow optimisation of therapeutic strategies. ACKNOWLEDGEMENTS
This work was supported by program grant SP 1518 from the Cancer Research Campaign of Great Britain and a studentship
from the UK Medical Research Council (to TDB). The isolation of bryostatin 1 was funded by PHS grant CA 16049-12, Outstanding Investigator Grant CA-44344-01A1 awarded by the NCI, DHHS, the Fannie E. Rippel Foundation, the Arizona Disease Control Research Commission and the Robert B. Dalton Endowment Fund.
REFERENCES ABELEV,G.I., Alpha-fetoprotein in oncogenesis and its association with malignant tumors. Advanc.Cancer Res., 14, 295-340 (1971). BLUMBERC, P.M., In vitro studies on the mode of action of the phorbol esters, potent tumor promoters. Crit. Rev. Toxicol., 8, 153-234 (1980). DALE,I.L., BRADSHAW, T.D., GESCHER, A. and PETTIT,G.R., Comparison of effects of bryostatins 1 and 2 and 12-0-tetradecanoylphorbol13-acetate on protein kinase C activity in A549 human lung carcinoma cells. Cancer Res., 49, 3242-3245 (1989). DALE,I.L. and GESCHER,A., Effects of activators of protein kinase C including bryostatin 1 and 2 on the growth of A549 human lung carcinoma cells. Int. J . Cancer, 43, 158-163 (1989). GESCHER, A,, Antiproliferative properties of phorbol ester tumor promoters. Biochem. Phnrmacol., 34, 2587-2592 (1985). GESCHER,A. and REED,D.J., Characterization of the growth inhibition induced by tumor-promoting phorbol esters and of their receptor binding in A549 human lung carcinoma cells. Cancer Res., 45,43154321 (1985). HANNUN, Y.A. and BELL,R.M., Phorbol ester binding and activation of protein kinase C on Triton X-100 micelles containing phosphatidylserine. J . biol. Chem., 261, 9341-9347 (1986). HOUSEY, G.M., JOHNSON, M.D., HSIAS,W.L.W., O'BRIAN,C.A., MURPHY, J.P., KIRSCHMEIER, P. and WEINSTEIN, I.B., Overproduction Of protein kinase C causes disordered growth control in rat fibroblasts. Cell, 52, 343-354 (1988). HUBERMAN, E. and CALLAHAM, M.F., Induction of terminal differentiation in human promyelmytic leukemia cells by tumor-promoting agents. Proc. nat. Acad. Sci. (Wash.), 76, 1293-1297 (1979). JACOBSON, H.I., BENNETT,J.A. and MIZEJEWSKI, G.J., Inhibition of estrogen-dependent breast cancer growth by a reaction product of alphafetoprotein and estradiol. Cancer Res., 50, 4 1 5 4 2 0 (1990). KRAFT,A.S., Effect of phorbol esters on the activity of purified transforming growth factors. Cancer Res., 46, 1764-1767 (1986). KRAFT,A . S . , BAKER,V.V. and MAY,W.S., Bryostatin induces changes in Drotein kinase C location and activitv without altering c-mvc gene expre'ssion in human promyelocytic leukthia cells (HL-66). O h G e n e , 1, 111-118 (1987). KRAFT,A.S., WILLIAM,F., PETTIT,G.R. and LILLY,M.B., Varied differentiation responses of human leukemias to bryostatin 1. Cancer Res., 49, 1287-1293 (1989). KRAUSS,R.S., HOUSEY,G.M., JOHNSON,M.D. and WEINSTEIN,I.B., Disturbances in growth control and gene expression in a C3H/T101/2 cell
line that stably overproduces protein kinase. C. Oncogene, 4, 991-998 (1989). LEAK,J.A., MAY,J.V. and KEEL, B.A., Human alpha-fetoprotein enhances epidermal growth factor proliferative activity upon porcine granulosa cells in monolayer culture. Endocrinology, 126, 669-671 (1990). OSBORNE, C.K., HAMILTON, B., NOVER,M. and ZIEGLER,J., Antagonism between epidermal growth factor and phorbol ester tumor promoters in human breast cancer cells. J . din. Invest., 67, 943-951 (1981). PERSONS,D.A., WILKISIN,W.O., BELL, R.M. and FINN,O.J., Altered growth regulation and enhanced tumorigenicity of NIH 3T3 fibroblasts transfected with protein kinase C-I c-DNA. Cell, 52, 447-458 (1988). PETTIT,G.R., HERALD,C.L., DOUBEK,S.L., HERALD,D.L., ARNOLD, J., Isolation and structure of bryostatin 1. J . Amer. chem. E. and CLARDY, Soc.,104, 6846-6848 (1982). QUINGCANC, L.I., BLACHER,R., ESCH,F. and CONGOTE,F., Isolation from fetal bovine serum of an apolipoprotein-H-like protein which inhibits thymidine incorporation in fetal calf erythroid cells. Biochem. J . , 267, 261-264 (1990). ROBERTS,A.B., ANZANO,M.A., WAKEFIELD,L.M., ROCHE,N.S., STERN,D.F. and SPORN,M.B., Type beta transforming growth factor: a bifunctional regulator of cellular growth. Proc. nat. Acad. Sci. (Wash.), 82, 11%123 (1985). ROVERA,G., SANTOLI,D. and DAMSKY, D., Human promyelmytic leukemia cells in culture differentiate into macrophage-like cells when treated with phorbol diester. Proc. nut. Acad. Sci. (Wash.), 75, 2779-2783 (1979). U., ROOS,W., FABBRO, D., KUNG,W., COSTA,S.D. and EPPENBERGER, Correlation between hormone dependency and the regulation of epidermal growth factor receptor by tumor promoters in human mammary carcinoma cells. Proc. nut. Acad. Sci. (Wash.), 83, 991-995 (1986). SAKO,T., YUSPA,S.H., HERALD,C.L., PETTIT,G.R. and BLUMBERG, P.M., Partial parallelism and partial blockade by bryostatin 1 of effects of phorbol ester tumor promoters on primary mouse epidermal cells. Cancer Res., 47, 5445-5450 (1987). T.W., SANCHEZ,J.H., BOREIKO,C.J., FURLONG,J . and HESTERBERG, Differential effects of tumor promoters on the growth of normal human bronchial epithelial cells and human lung tumor cell lines. Toxicol. in Vitro, 1, 183-188 (1987). TWARDZIK,D.R., RANCHALIS, J.E., MCPHERSON, J.M., OGAWA,Y., GENTRY, L., PURCHIO,A,, PLATA,E. and TODARO,G.J., Inhibition and promotion of differentiated-like phenotype of a human lung carcinoma in athymic mice by natural and recombinant forms of transforming growth factor-beta. J . nut. Cancer Inst., 81, 1182-1185 (1989).
