Communicated by Leon A. Heppel, April 3, 1981. ABSTRACT. Addition ofcholera ..... Pastan, I. H., Johnson, G. S. & Anderson, W. B. (1975) Annu. Rev. Biochem.
Proc. Natl. Acad. Sci. USA Vol. 78, No. 7, pp. 4392-4396, July 1981 Cell Biology
Cyclic AMP: A mitogenic signal for Swiss 3T3 cells (DNA synthesis/choleva toxin/insulin/serum-free medium/growth control)
ENRIQUE ROZENGURT, ANN LEGG, GEORGE STRANG, AND NIGEL COURTENAY-LUCK Imperial Cancer Research Fund, P.O. Box 123, Lincoln's Inn Fields, London WC2A 3PX, England
Communicated by Leon A. Heppel, April 3, 1981
Addition ofcholera toxin (100 ng/ml) to quiescent cultures of Swiss 3T3 cells acts synergistically with serum (2-4%), insulin, phorbol esters, epidermal growth factor, and fibroblastderived growth factor to stimulate DNA synthesis. In the presence of insulin, cholera toxin caused a dose-dependent increase in cumulative [3H]thymidine incorporation into acid-insoluble material and in the intracellular cyclic AMP (cAMP) level. The dose-response curves for the two processes were similar. Furthermore, addition of 1-methyl-3-isobutylxanthine (15-500 FLM) or of 4-(3butoxy-4-methoxybenzyl)-2-imidazolidinone (5-100 gLM), both of which are potent inhibitors of cyclic nucleotide phosphodiesterase activity, stimulated DNA synthesis and increased cAMP levels in Swiss 3T3 cells. These compounds strikingly potentiated the effect of cholera toxin on DNA synthesis and on cAMP levels. When quiescent Swiss 3T3 cells were exposed to cholera toxin (100 ng/ ml) and insulin at 10 pg/ml (4- to 7-fold increase in cAMP level) or to these agents and 1-methyl-3-isobutyl xanthine at 50 gIM (35fold increase in cAMP level), DNA synthesis began after a lag of 16 hr. These results indicate that cAMP acts as a mitogenic signal for Swiss 3T3 cells and differ from the widely held view that cyclic AMP inhibits the proliferation of fibroblast cells. ABSTRACT
complex nature of serum and its multiplicity of actions have frequently rendered difficult the interpretations of experiments with cultured animal cells grown in medium containing serum. In recent years, certain combinations of insulin, epidermal growth factor (EGF), vasopressin, phorbol esters, and fibroblast-derived growth factor (FDGF), a cell-derived growth factor, have been shown to completely replace serum in stimulating DNA synthesis in quiescent cultures of 3T3 cells (1, 1416). We used such mitogenic factors and defined culture conditions to assess the effect of cAMP elevating agents on the stimulation of DNA synthesis in cultures of Swiss 3T3 cells. In contrast to previous reports (2-10), we found that increased cellular concentrations of cAMP act synergistically with growth-promoting agents to stimulate DNA synthesis in quiescent cultures of 3T3 cells. MATERIALS AND METHODS Cell Culture. Stock cultures of Swiss 3T3 cells (17) maintained as described (18) were subcultured in 33-mm Nunc Petri dishes with 2 ml of Dulbecco's modified Eagle's medium/10% fetal bovine serum, refed after 2 days, and used 5 days after the last change of medium. These cells were confluent and arrested in G1/G0 (16, 18). Assays of Growth-Promoting Activity. Confluent and quiescent cultures of Swiss 3T3 cells in 33-mm dishes were washed twice with Dulbecco's modified Eagle's medium at 37VC to remove residual serum. Then, the cultures were incubated in 2 ml of Dulbecco's modified Eagle's medium/Waymouth medium (19) (1:1)/1 AM [3H]thymidine (1 pCi/ml; 1 Ci = 3.7 x 1010 becquerels) and various additions as indicated. The incorporation of radioactivity into acid-precipitable material was measured as described (18). cAMP Determination. Confluent and quiescent cultures of Swiss 3T3 cells in 33-mm dishes were washed twice with Dul.becco's modified Eagle's medium at 370C and incubated in 2 ml of Dulbecco's modified Eagle's medium/Waymouth medium in the presence of various additions and for various times. At the end of the specified incubation period, the medium was rapidly removed and cAMP was extracted with 200 1.l of 0.1 M HCl (20). After 20 min, the HCl solution was transferred to tubes and the precipitated protein remaining on the dish was dissolved in 1 ml of 0.1 M NaOH/2% Na2CO3. Appropriate dilutions of the neutralized HCl extract were acetylated (20), and cAMP was determined in duplicate by radioimmunoassay. Materials. Bovine insulin (26 international units/mg) and cholera toxin were obtained from Sigma. 1-Methyl-3-isobutylxanthine (IBMX) was obtained from Aldrich, and 4-(3-butoxy4-methoxybenzyl)-2-imidazolidinone (Ro.20-1724) was generously supplied by Hoffman-La Roche. The serum used was fetal
Quiescent cells can be stimulated to synthesize DNA and.proliferate by the addition of fresh serum or combinations of growth-stimulating factors (1). The possibility that the cyclic nucleotides, cyclic AMP (cAMP) and cyclic GMP, may.regulate the proliferative response of quiescent fibroblastic cells is the subject of a large and controversial literature (2-5). In 3T3 cells and other fibroblast cells, increased levels of cAMP are widely thought to reduce the rate of growth and inhibit the stimulation of DNA synthesis promoted by adding serum to quiescent cells (6-10). An objection to many of these studies has been that these effects were elicited by high concentrations of analogues of cAMP and could be regarded as nonspecific (2, 4). Furthermore, a few other studies with 3T3 cells suggested that cAMP promotes proliferation, but this contention remained unproven. Thus, addition of cAMP enhanced the initiation of DNA synthesis produced by serum, but noncyclic purine nucleotides or nucleosides were also effective (11). In addition, cholera toxin, an irreversible activator of the adenylate cyclase of eukaryote cells (12) was reported to stimulate, rather than inhibit, cell division in 3T3 cells maintained in-the presence of serum (13). However, the mitogenic effect of cholera toxin was not mimicked by other agents that elevated cAMP levels of 3T3 cells (13). All these inconsistent findings preclude any definitive conclusion concerning the effect of cAMP on the initiation of DNA synthesis of fibroblast cell lines. A complicating factor in most early studies concerning the effect of cyclic nucleotides on the initiation of DNA synthesis is that they have been conducted in cultures maintained in nutrient medium supplemented with unfractionated serum. The
Abbreviations: cAMP, cyclic AMP; EGF, epidermal growth factor; FDGF; fibroblast-derived growth factor; IBMX; 1-methyl-3-isobutylxanthine; Ro.20-1724, 4-(3-butoxy-4-methoxybenzyl)-2-imidazolidinone; PMA, phorbol 12-myristate 13-acetate.
The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.
4392
Cell Biology: Rozengurt et al.
Proc. Natl. Acad. Sci. USA 78 (1981)
bovine (Flow Laboratories, Rockville, MD). [3H]Thymidine was from the Radiochemical Centre (Amersham, England). Antigens and antibodies for radioimmunoassay of cAMP were from New England Nuclear. All other chemicals were of the purest grade commercially available.
4393
Table 1. Synergistic stimulation of DNA synthesis in quiescent Swiss 3T3 cells by cholera toxin and growth factors
[3H]Thymidine incorporation, cpm x 10-3 per dish Cholera toxin
RESULTS Effect of Cholera Toxin on Serum-Stimulated DNA Synthesis. Fig. 1 shows that addition of cholera toxin at 100 ng/ ml did not inhibit the induction of DNA synthesis promoted by saturating concentrations of serum (8% and 10%) in Swiss 3T3 cells. In fact, cholera toxin further enhanced [3H]thymidine incorporation and caused a 5- to 7-fold increase in the cellular content of cAMP (Fig. 1 Inset) when added in the presence of 2%, 3%, or 5% fetal bovine serum. The results suggest that increased levels of cAMP promote rather that inhibit entry into DNA synthesis in 3T3 cells. Synergistic Effects Between Cholera Toxin and GrowthPromoting Factors in Serum-Free Medium. As shown in Table 1, cholera toxin added with hormones such as insulin, EGF, or vasopressin (16); or with the potent tumor-promoting agent phorbol 12-myristate 13-acetate (PMA) (18); or with FDGF (14) synergistically stimulated DNA synthesis in cultures of Swiss 3T3 cells. All these potentiating effects between cholera toxin and the growth-promoting agents occurred in serum-free medium. Dose-Response of Cholera Toxin Effect on DNA Synthesis and cAMP Level. In the presence of insulin, cholera toxin stimulated DNA synthesis in a concentration-dependent fashion (Fig. 2 Upper). A significant effect was observed at a concentration as low as 0.1 ng/ml; the effect was almost maximal at 50
Addition None EGF (5 ng/ml) PMA (100 ng/ml)
(100 ng/ml) 4.8 36.6 62.9 81.3 182.5 290.1
Vasopressin (10 ng/ml) FDGF (10 pg/ml) Insulin (10 u~g/ml) [3H]Thymidine incorporation into acid-precipitated material was determined after a 40-hr incubation. =100 ng/ml. In the presence of cholera toxin at 100 ng/ml, addition of insulin stimulated DNA synthesis in a dose-dependent manner (results not shown). As insulin can decrease the level of cAMP in 3T3 cells (21), we tested the effect of cholera toxin on cAMP levels in the absence and presence ofthis hormone. Table 2 shows that addition of cholera toxin (100 ng/ml) caused a 4- to 6-fold increase in cAMP levels, measured after 3 hr of incubation, either in the absence or in the presence of insulin. As the mitogenic effect of cholera toxin was shown in the presence of insulin, most of the measurements of cAMP level were performed in the presence of this hormone. As shown in Fig. 2 Lower, cholera toxin elevated the cAMP level in a concentration-dependent fashion. The shapes of the dose-response curves for the two processes, induction of DNA synthesis (Fig. 2 Upper) and increase in cAMP level (Fig. 2 Lower) were similar.
1 0
20
Z
e 16
00
0.= 50
x
30 -
Control 2.1 3.7 13.4 4.9 122.7 19.3
v =
12
0
4I 0 ~~~~~~~~~~~~~~~~0
E.a -
4
lw OU
200
60 (u). quiescent Ncultures of .) ConfluentCand Siuaino FI.1. ytei yvroscnetatin
4a)
° 50 %:o
C6.-
3
2
0
5
tw
40
0
30
I/
I
I
I
0.
0
1
3
2
4
5
6
7
8
9
10
0. 20
oi t10n/l fe 2 ra 7C hnoprto of' [3 0hlr ltyidn 1) inoai-noul0aeilwsdtrie Serum, %
wi
10
FIG.
1.
Stimulation
of DNA
synthesis by various concentrations or presence of cholera toxin
in the absence (o)
of fetal bovine
serum
Confluent and quiescent cultures of Swiss 3T3 cells
were exposed (vol/vol) of fetal bovine serum without or ng/ml. After 28 hr at 370C, the incorporation
to various concentrations
with cholera toxin at 100 of
[3H~thymidine
into acid-insoluble material
(Inset) cAMP levels (mean for 3 hr with medium the absence
(o)
or
was
determined (18).
3) in Swiss 3T3 cells incubated SEM; n containing various concentrations of serum in ±
presence
=
of cholera toxin.
0
10-2 10-1 1
10
102
Cholera toxin, ng/ml FIG. 2. Dose-response of cholera toxin effect on DNA synthesis (Upper) and cAMP level (Lower) in the presence of insulin (10 yg/ml). cAMP levels (mean SEM; n = 9) and DNA synthesis were measured at 3 hr and 40 hr, respectively.
4394
Proc. Natl. Acad. Sci. USA 78 (1981)
Cell Biology: Rozengurt et al.
Table 2. Cholera toxin increases cAMP levels of Swiss 3T3 cells in the absence or in the presence of insulin cAMP, pmol/mg of protein Addition 14.4 ± 0.78 (6) None 12.2 ± 0.53 (6) Insulin 61.0 ± 4.9 (6) Cholera toxin 69.5 ± 4.6 (14) Insulin + cholera toxin
Quiescent Swiss 3T3 cells were incubated in the absence or in the presence of cholera toxin (100 ng/ml), insulin (10 pg/ml), or both for 3 hr. Results represent mean ± SEM. Values in parentheses are the numbers of independent cultures that were treated identically.
Time Course of Stimulation of DNA Synthesis and cAMP Level by Cholera Toxin. Fig. 3 shows that cholera toxin added to quiescent cultures of 3T3 cells elevated cAMP levels after a lag of 15 min. The cellular content of cAMP reached a maximum (7-fold increase) after 3 hr of incubation and remained elevated for at least 24 hr. Fig. 3 (Inset) shows that the combination of cholera toxin and insulin stimulated DNA synthesis after a lag of -16 hr, which is similar to the effect of addition of serum or growth factors to cultures of 3T3 cells (15, 16, 18). Similar results were obtained when the incorporation of [3H]thymidine was monitored by autoradiography (results not shown). Effect of Cholera Toxin and Insulin on 3T3 Cell Proliferation. Confluent and quiescent cultures of Swiss 3T3 cells were exposed to various concentrations of fetal bovine serum (2-10%) with or without insulin, cholera toxin, or both. As shown in Fig. 4, addition of insulin and cholera toxin caused a synergistic increase in cell number at all concentrations of serum tested. Dose-Response of IBMX and Ro.20-1724 Effects on DNA Synthesis and cAMP Level. IBMX, a potent inhibitor of cyclic nucleotide phosphodiesterase (22) caused a dose-dependent stimulation of [3H]thymidine incorporation into acid-insoluble material when tested in the presence of insulin and increased the intracellular levels of cAMP (Fig. 5). Both parameters were affected in the same range of IBMX concentration (15-500 AmM). 70
60 .,-
N
o
50
04 4.4
0
b
40
0
:6
30 30
¢ 20 10
Time, hr FIG. 3. Time course of elevation of cAMP level and stimulation of DNA synthesis by cholera toxin in cultures of Swiss 3T3 cells. The cultures were incubated in medium containing insulin (10 zg/ml) and cholera toxin (100 ng/ml). At the times indicated, the cellular content of cyclic AMP (mean + SEM; n = 3) was measured. (Inset) Cumulative [3Hlthymidine incorporation into acid-insoluble material was measured in cultures of Swiss 3T3 cells exposed to insulin at 10 yg/ml (o) or to insulin at 10 t.g/ml plus cholera toxin at 100 ng/ml (u).
20
-
*10
15 k
-f.,
._ u: 10
6 kax
l10
6
a)
Q
5
*.
L000
_0
0
0:00 o0
0
L
OL
2
4
I[
6 Seum
0a 00 0a 0a 0a 0
0a 0a 0a 0a 0a 0a 00 04 0a
'. 10
04 0
04 0 04 04 a
8
FIG. 4. Increase in cell number produced by insulin, cholera toxin, or both in the presence of various concentrations of fetal bovine serum. Quiescent cultures of Swiss 3T3 cells in 33-mm dishes were incubated in the absence (o) or in the presence of insulin at 1 ug/ml (), cholera toxin at 100 ng/ml (E, orboth (i) and various concentrations of serum. After 3 days of incubation, the cells were removed from the dishes and assayed as described (16). Data represent the mean of two independent cultures treated identically; values agree within 10%. Arrows indicate the cell number that would be expected if the effects of insulin and cholera toxin were strictly additive.
To further check the specificity of action of IBMX, we also studied the effect of Ro.20-1724, a potent and selective inhibitor of cAMP phosphodiesterase (22). This compound stimulated DNA synthesis in the presence of insulin and elevated the cAMP level in the same range of concentration (5-100 AM) (Fig. 5). These results suggest a close correlation between cAMP level and stimulation of DNA synthesis. IBMX and Ro.20-1724 Potentiate the Effect of-Cholera Toxin on DNA Synthesis and cAMP Level. Ifthe primary effect of cholera toxin on the initiation of DNA synthesis by 3T3 cells is due to its activation of adenylate cyclase and cellular accumulation of cAMP, inhibitors of cyclic nucleotide phosphodiesterase activity should potentiate the stimulation of DNA synthesis and the cellular accumulation of cAMP produced by cholera toxin. As shown in Fig. 6 Left, 50 ,uM IBMX markedly potentiated the stimulation of DNA synthesis by cholera toxin; maximal effect was obtained at a concentration of cholera toxin as low as 10 pg/ml (10 fM). Fig. 6 Right shows that IBMX strikingly potentiates the dose-dependent increase in cellular cAMP level promoted by cholera toxin. In the presence of IBMX, cholera toxin at 10 pg/ml caused a 7-fold increase in cAMP level after a 3-hr incubation. Similar results were obtained when 10 ,uM Ro.20-1724 was added instead of 50 ,M IBMX (results not shown). Fig. 7 shows the kinetics of stimulation of DNA synthesis and of cellular accumulation of cAMP in quiescent 3T3 cells exposed to cholera toxin and IBMX in the presence of insulin. It can be
Proc. Natl. Acad. Sci. USA 78 (1981)
Cell Biology: Rozengurt et al.
4395
CD)._
20
50
15
40
0.
300
30 x
"a
E 0.
30
Go
.
a) 0
S. to
0
X 10 CD)
20
.° 20
5 "-4
10
0
*r 0
0 5-
C
200 E
Ca
le
0
5
0.cow
x
0. Ci) 10
0. 10
6 0.0
-
I
r. ._
Cd
t0 0
0. $4
0
0
._..
12
60
20
0
1
10
10-210-1 1
100 0 0.5 5 10 50 100 Ro.20-1724, ,uM
FIG. 5. Dose-response of IBMX effect (Upper) and Ro.20-1724 effect (Lower) on DNA synthesis and cAMP level. Quiescent cultures of Swiss 3T3 cells were incubated with (a) or without (0) insulin at 10 fig/ ml and various concentrations of IBMX or Ro.20-1724. Some of the cultures received [3H]thymidine and cumulative incorporation of radioactivity into acid-insoluble material was measured after 40 hr of incubation. The cellular content of cAMP (mean ± SEM; n = 3) was measured after 4 hr of incubation.
10 102
10-210-1 1
10 102
Cholera toxin, ng/ml
""
40
4
0
E-
FIG. 6. Dose-response of cholera toxin effect on stimulation of DNA synthesis and cyclic AMP level in the absence (e) or presence (a) of 50 uM IBMX. (Left) Quiescent Swiss 3T3 cells were incubated in medium containing insulin at 10 ,ug/ml [3H]thymidine, and various concentrations of cholera toxin for 40 hr. In this experiment, addition of medium containing 10% fetal bovine serum produced 4.66 x 105 cpm per culture. (Right) cAMP levels (mean ± SEM; n = 3) were determined under conditions similar to those used for the DNA synthesis measurements except that [3H]thymidine was omitted from the medium. The cultures were incubated for 3 hr.
6 and 7) initiate DNA synthesis after a lag of 16 hr, which is similar to the lag observed when DNA synthesis is induced by combinations of hormonal growth-promoting factors (15, 16, 18). Other studies performed with mixed cell populations (26, 30, 31) or in vivo (26) are difficult to interpret because changes in cAMP levels cannot be properly documented and because it cannot be ruled out that cAMP is acting indirectly by promoting the release of a growth factor. Finally, in contrast to most previous studies, our experiments with 3T3 cells were con"a 400
I-
seen that DNA synthesis is markedly stimulated after a lag of 16 hr. The level of cAMP measured in parallel cultures increased 35-fold after a lag of less that 1 hr and remained elevated for at least 30 hr. DISCUSSION Our findings demonstrate that cAMP is a mitogenic signal for Swiss 3T3 cells and differ from the widely held view (2-10) that cAMP inhibits the proliferation of fibroblast cells. This conclusion has been further substantiated by recent experiments showing that derivatives of cAMP are mitogenic for Swiss 3T3
cells (unpublished data). A number of reports have suggested a role for cAMP in stimulating proliferation in cell types other than fibroblasts (22-31). However, there are important differences between these studies and the present findings with 3T3 cells. In epidermal keratinocytes (23, 24) and in mammary epithelial cells (25), addition of cholera toxin or cAMP derivatives appears to decrease the doubling time of the population or to delay the onset of terminal differentiation, rather than to induce quiescent cells in G/G1 to divide. In T5LB liver epithelioid cells arrested near the end ofthe G1 phase by Ca deprivation, a transient elevation in cAMP level is followed by initiation of DNA synthesis within 1-2 hr (26, 27). We found that Swiss 3T3 cells stimulated by cholera toxin and insulin (4- to 7-fold increase in cAMP, Figs. 2 and 3) or by these agents and IBMX (35-fold increase in cAMP, Figs.
S4
-0 ._'.
300 -S ca0
x
0. 0
C4-,
20 Si 20 Q
a-6l
200
0
C.
14
0
E
0, $.
OQ
1._
,-I 10
100 2!4
0)
a0 "a 10
20 Time, hr
30
40
FIG. 7. Time course of elevation of cAMP level (s) and stimulation of DNA synthesis (s) by cholera toxin and IBMX. Quiescent Swiss 3T3 cells were incubated in medium containing insulin at 10 Ag/ml, cholera toxin at 100 ng/ml, and 50 ;LM IBMX. At the times indicated, cellular cAMP (mean ± SEM; n = 3) was measured. Incorporation of [3Hlthymidine into acid-insoluble material was measured in other cultures exposed to cholera toxin at 100 ng/ml and 50 ItM IBMX (a) or to cholera toxin at 100 ng/ml, 50 pM IBMX, and insulin at 10 /Ag/ ml (e and *). In this experiment, addition of medium containing 10% fetal bovine serum resulted in 3.2 x 105 cpm incorporated per dish after a 40-hr incubation.
4396
Cell Biology: Rozengurt et al.
ducted in serum-free medium supplemented with defined growth-promoting factors. The present findings raise the possibility that cAMP plays a role as a "second messenger" in mediating the action of growth-promoting factors in 3T3 cells. This possibility is, however, unlikely for many growth factors because increased cellular levels of cAMP act synergistically with EGF, FDGF, PMA, vasopressin, and insulin to stimulate DNA synthesis and because our preliminary results show that none of these factors increase the basal levels of cAMP in 3T3 cells. Although further studies on the effect of these agents and their mitogenic combinations on cAMP levels are needed, our findings suggest that there are cAMP-dependent and cAMP-independent pathways that synergistically lead to 3T3 cell proliferation. Whether the cAMP-dependent mitogenic pathway is independent from, or converges with, that activated by other growth-promoting factors is an important question that requires further experimental work. 1. Rozengurt, E. (1980) Curr. Top. CelL Regul 17, 59-88. 2. Chlapowski, F. J., Kelly, L. A. & Butcher, R. W. (1975) Adv. Cyclic Nucleotide Res. 6, 245-338. 3. Pastan, I. H., Johnson, G. S. & Anderson, W. B. (1975) Annu. Rev. Biochem. 44, 491-522. 4. Friedman, D. L., Johnson, R. A. & Zeilig, C. E. (1976) Adv. Cyclic Nucleotide Res. 7, 69-114. 5. Rozengurt, E. (1979) in Surfaces of Normal and Malignant Cells, ed. Hynes, R. (Wiley, New York), pp. 323-353. 6. Johnson, G. S. & Pastan, I. (1972) J. Natl Cancer Inst. 48, 13771387. 7. Willingham, M. C., Johnson, G. S. & Pastan, I. (1972) Biochem. Biophys. Res. Commun. 48, 743-748. 8. Burger, M. M., Bombik, B. M., Breckenridge, B. McL. & Seppard, J. R. (1972) Nature (London) 239, 161-163. 9. Bombik, B. M. & Burger, M. M. (1973) Exp. Cell Res. 80, 88-94.
Proc. Natl. Acad. Sci. USA 78 (1981) 10. Kram, R., Mamont, P. & Tomkins, G. M. (1973) Proc. Natl Acad. Sci. USA 70, 1432-1436. 11. Schor, S. & Rozengurt, E. (1973) J. Cell Physiol 81, 339-346. 12. Gill, D. M. (1977) Adv. Cyclic Nucleotide Res. 8, 85-118. 13. Pruss, R. M. & Herschman, H. R. (1979)J. Cell Physiol. 98, 469474. 14. Bourne, H. R. & Rozengurt, E. (1976) Proc. Natl. Acad. Sci. USA 73, 4555-4559. 15. Dicker, P. & Rozengurt, E. (1978) Nature (London) 276, 723726. 16. Rozengurt, E., Legg, A. & Pettican, P. (1979) Proc. Natl Acad. Sci. USA 76, 1284-1287. 17. Todaro, G. J. & Green, H. (1963) J. Cell Biol 17, 299-313. 18. Dicker, P. & Rozengurt, E. (1980) Nature (London) 287, 607612. 19. Mierzejewski, K. & Rozengurt, E. (1977) Exp. Cell Res. 106, 394396. 20. Brooker, G., Harper, J. F., Terasaki, W. L. & Moylan, R. D. (1979) Adv. Cyclic Nucleotide Res. 10, 1-33. 21. Otten, J., Johnson, G. S. & Pastan, I. (1972) J. Biol Chem. 247, 7082-7087. 22. Bergstrand, H., Kristofferson, J., Lundquist, B. & Schurmann, A. (1977) Mol Pharnacol 13, 38-43. 23. Green, H. (1978) Cell 15, 801-811. 24. Marcelo, C. L. (1979) Exp. Cell Res. 120, 201-209. 25. Taylor-Papadimitriou, J., Purkis, P. & Fentiman, I. S. (1980) J. Cell Physiol 102, 317-321. 26. Whitfield, J. F., Boynton, A. L., MacManus, J. P., Siroska, M. & Tsang, B. K. (1979) Mol Cell Biochem. 27, 155-179. 27. Boynton, A. L. & Whitfield, J. F. (1979)J. Cell Physiol 101, 139148. 28. Pawelek, J., Halaban, R. & Christie, G. (1975) Nature (London) 258, 539-540. 29. Raff, M. C., Hornby-Smith, A. & Brockes, J. P. (1978) Nature (London) 278, 672-673. 30. Wang, T., Sheppard, J. R. & Foker, J. E. (1978) Science 201, 155-157. 31. Rabinovitch, A., Blondel, B., Murray, T. & Mintz, D. H. (1980) J. Clin. Invest. 66, 1065-1071.
