Thyrotropin-releasing Hormone and Cyclic AMP ...

JOURNALOF BIOLOGICAL CHEMISTRY Val. 257, No. 6 , Issue of March 25. pp. 3306-3312. 1982 Printed in U.S.A.

THE

Thyrotropin-releasing Hormone and Cyclic AMP Activate Distinctive Pathways of Protein Phosphorylation inGH Pituitary Cells* (Received for publication, July 27, 1981, and in revised form, November 2, 1981)

Debra S. DrustS, Claudia A. Suttons, and ThomasF. J. Martin1 From the Department of Zoology, University of Wisconsin, Madison, Wisconsin 53706

Thyrotropin-releasing hormone was the first hypophysiotropic peptide to be purified, chemically characterized, and synthesized (1); however, themechanism by which TRH’

* This work was supported by grants from the National Institute of Arthritis, Metabolism and Digestive Diseases (AM 25861) and the American Diabetes Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ Predoctoral trainee supported by United States Public Health Service Training Grant GM 07215 to the University of Wisconsin. 9 Partially supported by a Biomedical Research Support Grant. 7 Requests for reprints andcorrespondence should be addressed to: Dr. T. F. J. Martin, Zoology Research Building, University of Wisconsin, Madison, WI 53706. I The abbreviations used are: TRH, thyrotropin-releasing hormone; VIP, vasoactive intestinal polypeptide; CAMP,adenosine 3’:5’-monophosphoric acid; 8-Br-cAMP, cyclic 8-bromo adenosine 3’:5”monophosphate; Bt2cAMP, dibutyryl cyclic adenosine 3’:5’-monophosphate; IEF, isoelectric focusing; PAGE, polyacrylamide gel electrophoresis; SDS, sodium dodecyl sulfate; MIX, methylisobutyl xan-

promotes pituitary hormonerelease remains obscure in spite of extensive investigation. Early studies on its thyrotropinand prolactin-releasing activity were directed toward implicating cAMP in its mechanism of action. Binding sites for TRH were localized to particulatefractions inadenohypophyseal homogenates and were found to partially co-purify with adenylate cyclase-enriched membranes (2).However, successful demonstration of cyclase stimulation by TRH has remained elusive and negative findings have been reported by several laboratories (3,4). Inour recent studies,2 it was found that cyclase activity in crude homogenates of GH pituitary cells could be stimulated by VIP andcholera toxin but not by TRH ( 5 ) . Efforts to demonstrate cAMP accumulation in pituitary cells, thyrotropic tumor cells, and GH cells in response to TRH have also been equivocal. Several workers reported a modest, slow increase in cAMPwith TRH treatment (6,9,10) whereas other efforts were unsuccessful (4, 7 , 8). In recent studies (5),2we directly compared cAMP accumulation and prolactin release promoted by TRH, VIP, and cholera toxin in GH cells. The lattertwo agents promotedlarge increases in cAMP compared with verysmallincreasesobservedwith equally effective prolactin-releasing concentrations of TRH. A similar comparison of VIP and TRH was also reported by Gourdji et al. (11). In summary, the studies reviewed here provide little conclusive support for a role of cAMP in TRH action. In contrast, there is ample evidence forthe existence of and role for a CAMP-dependent pathway in regulating pituitary thyrotropin andprolactin secretion. cAMP analogues inmany butnot all studies ( 7 ) exerted hormone-releasingactions. Agents which enhance cAMP levels were also found to be effective secretagoguesin pituitary tissue(14), thyrotropic tumors (4, 8), and GH cells (5, 11, 12). In addition, CAMPdependent protein kinase and endogenous substrates of the enzyme were studied in pituitary tissue by Lemay and coworkers (13). Inthestudiesreported here, we investigateda CAMPactivated pathway of protein phosphorylation in clonal GH strains of pituitary cells. Since these cells constitute a homogeneous TRH-responsivepopulation, we also delineateda pathway of TRH receptor-activated protein phosphorylation. A comparison of both pathways showed striking differences and constitutes strong evidence against a role for cAMP as the sole intracellular mediator of T R H action. A preliminary report of these findings has been presented (15). thine; ID, one-dimensional, 2D, two-dimensional; TCA, trichloroacetic acid; Hepes 4-(2-hydroxyethyl)-l-piperazineethanesulfonicacid; CT, cholera toxin; PKI, protein kinase inhibitor; C, catalytic subunit of CAMP-dependent protein kinase. 41 K, 45 K, etc., denote peptides of 41 kilodaltons (M, = 41,000). etc. * S. A. Ronning and T. F. J. Martin, manuscript submitted.

3306

Downloaded from www.jbc.org by guest, on July 12, 2011

The studies reported here were undertaken to clarify the cellular mechanism of the hypothalamic tripeptide, thyrotropin-releasing hormone (TRH), in clonal, hormone-responsive GH pituitary cells and to assess the possibility of a role for cyclic AMP as a mediator of TRH action. We investigated patterns of protein phosphorylation by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and autoradiography of high speed supernatant and pellet fractions from untreated and treated GH cells. Brief treatment of cells with agents which elevate or mimic cellular cyclic AMP (8-bromo cyclic AMP,dibutyryl cyclic AMP, vasoactive intestinal polypeptide or cholera toxin) stimulated thephosphorylation of five supernatant peptides (41,45,47, 72, and 82 kilodaltons) and one pellet peptide (135 kilodaltons) and decreased the phosphorylation of one supernatant peptide (55 kilodaltons). In contrast,TRH promoted the phosphorylation of four differentsupernatant peptides (two 59, 65, and 80 kilodaltons). In addition, TRH also stimulated the phosphorylation of cyclic AMP-responsive 41-, 45-, and 82-kilodalton supernatant peptides and 135-kilodalton pellet protein and decreased the phosphorylation of 55-kilodalton supernatant peptide. Altered labeling of 47- and 72-kilodalton supernatant peptides, however, was not observed with TRH. Time course studies, as well as the overlapping biological actions of TRH and vasoactive intestinal polypeptide, lead us toconclude that these peptide hormonesutilize distinct, parallel pathways which converge a t some late step. Furthermore, the results indicate that effects of TRH are mediated by a cyclic AMP-independent pathway.

Thyrotropin-releasingHormone and Protein Phosphorylation

acidic

3307

basic

I - \.

-

”

t

97

- 97

72

-72

59

- 59

47

- 47

41

-41

11

.*

- \

-47 -41

e

0

D I

control

VIP

.

TRH

FIG.2. Two-dimensional IEF/SDS polyacrylamide gels of VIP-and TRH-induced supernatant proteinphosphorylation in GH cells. Labeling of cultures with ‘”PO, for 60 min was as described in “Materials and Methods.” High speed supernatants from control (left), VIP-treated (middle),and TRH-treated (right) cultures were analyzed on two-dimensional gels as described. Hormone treatment was for 10 rnin at 10 p TRH and 0.1 p~ VIP. Downloaded from www.jbc.org by guest, on July 12, 2011

right). Asimilarexperimentanalyzed by two-dimensional PAGE indicated a maximal enhancement of 47 K (B) phosRESULTS phorylation by 1 min of VIP treatment (Fig. 5). In contrast, increasedphosphorylation of 70-80 K and 135 K peptides CAMP-activated Protein Phosphorylation in G H CellsOur previous studies (5)?showed that VIP acted in GH cells following VIP treatment occurred more slowly with maximal responses observed at 10-15 min (Fig. 4, left). to stimulate adenylate cyclase activity and promote a large Cholera toxin is known to increase cAMP levels in GH cells intracellular accumulation of CAMP.We utilized this hormone to define a cellular pathway of protein phosphorylation which following a characteristic lag (12).2 Treatment of cells with was CAMP-activated. VIP treatment of :”PO,-prelabeled GH cholera toxin for 1-2 h was found to promote a pattern of from cells for a brief period (10 min) was found to reproducibly changes in proteinphosphorylationindistinguishable enhance thephosphorylation of a discrete setof proteins (Fig. those described for VIP. In addition, exposure of cells to 8 Br1). In thehigh speed supernatant fraction, phosphorylation of or Bt’cAMP promoted similarly altered phosphorylation 41 K, 47 K, and 70-80 K peptides was enhanced by exposure (Figs. 1 and 6). These results support the conclusion that of cells to VIP. In the microsomal pellet fraction, only a 135 cellular effects of VIP are related to its action of increasing K phosphoprotein was reproducibly increased in intensity by CAMP.Hence, the pattern of proteinphosphorylationdea pathway which VIP treatment. Analysis of VIP-activated phosphorylation of scribed abovefor VIP was taken to represent was CAMP-activated. supernatant peptides by two-dimensional IEF/SDS PAGE TRH Receptor-activated Protein Phosphorylation in G H confirmed and extended these results asshown in Fig. 2. VIP treatment consistently increased labeling of 41 K (A), 47 K Cells-TRH and VIP have been found to exert a number of (B), 72 K (D), and 82 K (C) peptides. In addition, increased similar effects in GH cells (see “Discussion”). However, in contrast to VIP, TRH failed to stimulate adenylate cyclase phosphorylation of a 45 K (H) peptide and decreased phosphorylation of a 55 K peptide (I) with VIP treatment were activity (3-5)’ and increased cellular cAMP levels only modestly or not a t all (5, 7).’ An examination of the effects of TRH observed in the two-dimensional analysis. VIP increased phosphorylation of 41 K, 47 K, 70-80 K, and on protein phosphorylation revealed a number of differences 135 K peptides over a narrow dose-range (Fig. 3, left). The as well as similarities withthe effects of VIP. As shown in Fig. concentration of VIP for half-maximal response was similar 9, TRH increased the phosphorylation of 41 K, 59 K, and 70for each phosphopeptide and occurred at approximately 5 nM. 80 K supernatant peptides but not the 47 K peptide in a 10Increases in cAMP promoted by VIP in a 10-min incubation min treatment. Phosphorylation of microsomal 135 K protein were half-maximal at slightly higher concentrations (15 nM, was also increased by brief TRH treatment. The observed changes in phosphorylation were dose-dependent(Fig. lo), Fig. 3, right). Increasedproteinphosphorylationoccurred rapidly in response to VIP addition (Fig. 4, left). Two general and half-maximal responses occurred at 1-100 nM. A more detailed analysisof TRH action was undertaken by patterns were observed in the latenciesof responses. Increased 47 K phosphorylation was maximal by 1-2 min of VIP treat- analyzing supernatant phosphoproteins by 2-dimensional ment and appeared to parallel the rapid increase in cAMP IEF/SDS PAGE as shown in Fig. 2 (left and right). TRH which reachedapeak by 2 min of VIP treatment (Fig. 4, stimulated the phosphorylation of A (41 K ) and J (59 K ) ’’ Portions of this paper (including “Materials and Methods.” part peptides identified by one-dimensional PAGE. Increased phosphorylation of 82 K peptides (C) was also observed alof “Results,” Figs. 1, 3-5, and 7-10, and Table I) are presented in miniprint at the end of this paper. Miniprint is easily read with the though increased 72 K (D) peptide phosphorylation (in conaid of a standard magnifying glass. Fuilsize photocopies are available trast to VIP)was not observed. TRH also stimulated labeling from the Journal of Biological Chemistry, 9650 Rockville Pike, Beof 59 K (E), 65 K (F), 80 K (G), and 45 K (H) peptides and thcsda, MD20814. RequestDocument No. 81M-1815, cite the audecreased labeling of a 55 K (I) peptide. As previously found thors, andinclude a check or money order for $6.40 per set of photocopies. Full size photocopies are also included in the microfilm for VIP, changes in protein phosphorylation in response to TRH could be segregated into two classes on the basis of edition of the Journal that is available from Waverly Press. MATERIALS AND METHODS?

diBu

3308

acidic

Thyrotropin-releasingHormone and

--

Protein Phosphorylation

basic .

4.,

. -

"

-97

8-97

-97

-72 -59

-72 -59

-72 -59

-47 -41

-47 -41

control

-47 -4 1

,

CT

speed of onset. Increased phosphorylation of E (59 K) and F (65 K) peptides was extremely rapid (Fig.5) and was maximal by 30 s. In contrast, increased phosphorylation of the 80 K peptide (G) occurred more slowly, with a maximal response observed at 10 min following TRH addition (Fig. 5). Comparison of VIP- and TRH-induced alterations in supernatant protein phosphorylation (Fig. 2, middle and right) confirmed that both peptides exert effects incommon: increasing A, C, and H phosphorylation and decreasing I phosphorylation. However, TRH uniquely altered phosphorylation of E, F, G, and J whereas VIP uniquely altered phosphorylation of B and D. A summary of these results is presentedin Table I.

Although 47 K phosphorylation may be representative of a proximal step in a VIP receptor-activatedmechanism, we have no evidence that this peptide is actually a substrate for a CAMP-dependent protein kinase. In the present study, we identifiednineendogenous supernatantsubstrates of the CAMP-dependent enzyme by in vitro phosphorylation. However, we could not demonstrate VIP-responsive phosphoproteins tobe substrates of the enzyme. There areseveral possible interpretations of this result. 1) VIP-responsive phosphopeptides might besubstrates for CAMP-independent kinase/phosphatase systems which are distal elements of a cascade activated by the CAMP-dependent enzyme. 2) Conditions in vitro and in uiuo may notbe at all comparable. Apparent differences between CAMP-dependent protein kinase substrates observed DISCUSSION in uitro and CAMP-responsive phosphoproteins observed in reported by LeCam and coPrevious studies with somaticcell mutants have suggested vivo have alsobeenrecently that the cellular actions of cAMP may be attributable to the workers (22). TRH treatment altered protein phosphorylation in GH cells action of CAMP-dependent protein kinase(s) inpromoting protein phosphorylation (20). It should be possible to deter- in a reproducible and distinctive manner involving phosphomine a pattern of protein phosphorylation which is CAMP- proteins E, F, G, and J. In several experiments, maximally dependent by :"POj-labeling and gel electrophoresis. Here, we enhanced phosphorylation of E and F was observedby 30 s of present the results of such a study in hormone-responsive GH TRH treatment. This is at a time prior to maximal receptor E and F phosphorylation closely pituitary cells. Reproducible changes inlabeling following occupancy and indicates that treatment with agents which alter or mimic cellular cAMP follows TRH binding. Increased E, F, G, and J phosphorylain response to agents which alter were restricted to a small number of phosphoproteins. The tion was never observed cAMP levels. Hence, phosphorylation of these peptides must results indicated that altered protein phosphorylation promoted by VIP was indistinguishable from that observed with result from activation of a kinase/phosphatase system which is CAMP-independent. TRH treatmentalso consistently failed 8-Br-cAMP, Bt2cAMP, and cholera toxin. Such results are consistent with our recent studies (5)2and those from other to enhance phosphorylation of B or D, increased labeling of which was observed to be CAMP-activated. These results laboratories (11, 21), which have shown that VIP activates adenylate cyclase and stimulates cAMPaccumulation. These strongly supportthe conclusion that cellular responses to observationssuggest that cellular actions of VIP may be TRH occur independentlyof activation of a CAMP-dependent attributable toCAMP-dependent protein phosphorylation. In- pathway. TRH but not VIP rapidly stimulates phosphatidylinositol creases in protein phosphorylation observed with VIP were turnover (23).4cAMP increases and the phosphatidylinositol half-maximal at 5 nM, in reasonable agreement with similar response observed with VIP and TRH, respectively, may be estimates for other effects of VIP in GH cells (0.5 n ~prolactin , involved in initiating cellular responses to these peptide horrelease; 5 nM, uridine uptake; 15 nM, cAMP increase). VIP rapidly stimulated the phosphorylation of a 47 K ( B ) mones. The observation of distinctive patterns of protein peptide. In contrast, phosphorylation of 70-80 K and 135 K phosphorylation at early times of exposure to VIP or TRHis peptides increased more slowly. Increased phosphorylation of consistent with this view of independent distinctive pathways these latter peptides might be the result of the activation of of action. a cascade involving a number of kinases or phosphatases. ' Sutton, C. A,, and Martin, T. F. J. (1982) Endocrinology, in press.

Downloaded from www.jbc.org by guest, on July 12, 2011

FIG. 6. Two-dimensional IEF/SDS polyacrylamide gels of CAMP-activated supernatant protein phosphorylation in GH cells. Methods used were similar to those described in the legend to Fig. 2. Control (left),BkcAMI' (diBu CAMP)-treated (middle)and cholera toxin (C7')-treated (right) cultures were analyzed. BtfcAMP (3 mM) treatment was for 30 min and cholera toxin (10 ng/ml) treatment was for 120 min.

Thyrotropin-releasingHormone and Protein Phosphorylation

T. F. J. Martin, unpublished observations. D. S. Drust, and T. F. J. Martin, manuscript submitted.

nucleotide secretagogues influence their phosphorylation. Localization studies of cellular phosphoproteins and elements of the secretory apparatus may help to establish a functional significance for protein phosphorylation in GH pituitary cells. REFERENCES 1. Blackwell, R. E., and Guillemin, R. (1973) Annu. Rev. Physiol. 35,357-390 2. Poirier, G., Labrie, F., Barden, N., and Lemaire, S. (1972) FEBS Lett. 20, 283-286 3. Hinkle, P. M., and Tashjian,A. H., Jr. (1977) Endocrinology 100, 934-944 4. Gershengorn, M. C., Rebecchi, M. J., Geras, E., and Arevalo, C. 0. (1980) Endocrinology 107,665-670 5. Ronning, S. A., and Martin, T. F. J. (1981) Proceedings of the 63rd Annual Meeting of the Endocrine Society (Abstr. 956) 6. Labrie, F., Borgeat, P., Lemay, A., Lemaire, S., Barden, N., Drovin, J., Lemaire, I., Jolicoeur, P., and Belanger, A. (1975) Adu. Cyclic Nucleotide Res. 5, 787-801 7. Martin, T. F. J., and Tashjian, A. H., Jr. (1977) in Biochemical Actions of Hormones, (Litwack, G., ed) Vol. IV, pp. 269-312, Academic Press, New York 8. Eto, S., and Fleischer, N. (1976) Endocrinology 98, 114-122 9. Dannies, P. S., Gautvik, K. M., and Tashjian, A. H., Jr. (1976) Endocrinology 98, 1147-1159 10. Gautvik, K. M., and Kriz, M. (1976) Biochem. J. 156, 111-117 11. Gourdji, D., Bataille, D., Vauclin, N., Grouselle, D., Rosselin, G., and Tixier-Vidal, A. (1979) FEBS Lett. 104, 165-168 12. Dannies, P. S., and Tashjian, A. H., Jr. (1980) Endocrinology 106,1532-1536 13. Lemay, A., Deschenes, M., Lemaire, S., Poirier, G., Poulin, L., and Labrie, F. (1974) J.Bwl. Chem. 249, 323-328 14. Jutisz, M., and McKerns, K. W. (eds) (1980) Synthesis and Release of Adenohypophyseal Hormones, Plenum Press, New York 15. Martin, T. F. J., Sutton, C. A., and Drust,D. S. (1981)Proceedings of the Annual Meetingofthe Endocrine Society (Abstr. 951) 16. Martin, T. F. J. (1980) J. Cell Physiol. 103,489-502 17. Laemmli, U. K. (1970) Nature (Lond.)227,680-685 18. Schendel, P. F.,and Wells, R. D. (1973) J. Biol. Chem. 248,83198321 19. O'Farrell, P. H. (1975) J. Biol. Chem. 250,4007-4021 20. Bourne, H. R., Coffio, P., Melmon, K. L., Tomkins, G. M., and Weinstein, Y. (1975)Adu. Cyclic Nucleotide Res. 5, 771-786 21. Robberecht, P., Deschodt-Lanckman, M., Camus, J.-C., De Neef, P., Lambert, M., and Christophe, J. (1979) FEBS Lett. 103, 229-233 22. LeCam, A., Nicolas, J.-C., Singh, J. J., Cabral, F., Pastan, I., and Gottesman, M. M. (1981) J. Biol. Chem. 256, 933-941 23. Sutton, C . A. (1981) Proceedings ofthe 63rd Annual Meeting of the Endocrine Society (Abstr. 960)

Downloaded from www.jbc.org by guest, on July 12, 2011

In GHcells, TRH andVIP exertseveral effects in common, including enhanced prolactin secretion (5, ll),increased Ca2+ . ~regulating proefflux (5); and increased uridine ~ p t a k eIn lactin secretion, both TRH and VIP appear to be equally sensitive to inhibition by Ca2' depletion or antagonism, suggesting the presence of a common Ca2+-sensitivestep in the actions of both hormones (5).*Treatment of cells with either hormone increased the phosphorylation of a common set of proteins A, C, H, and P135 to approximately an equal extent and decreased phosphorylation of I. In the case of A phosphorylation, nonadditive stimulation with VIP and TRH at maximally effective concentrations was observed (not shown) and suggests that both hormonesutilize pathways which share some step in A phosphorylation. Similar considerations may be applicable to the phosphorylation of other common peptides by both hormones. Overlap in cellular phosphorylation mechanisms for TRH andVIP may indicateeither overlap in at least one early, initiating action promoted by either hormone, or convergence of independent, parallel pathways at some distal step. We favor the latter alternative since TRH and VIP promote biological responses in common in spite of seemingly distinctearly actions. Time course experiments support this suggestion since effects of hormones on B or E and F phosphorylation preceded those on A, C, and P135 phosphorylation. That the point of convergence for some responses to TRH andVIP may involve CaZ+is supported by our recent finding that phosphorylation of hormone-responsive C peptides is mediated in vitro by a Ca2'-dependent, phospholipid-activated protein kinase.' Effects of secretagogues such as VIP and TRHin increasing prolactin release can be mimicked by treatment of GH cells with CAMP analogues (7). Analogue-induced secretion probably occurs by a CAMP-dependent protein kinase-mediated pathway of protein phosphorylation. Presumably, the same pathway underlies prolactin-releasing actions of VIP. However, it is not known whether prolactin-releasing actions of TRH aremediated by protein phosphorylation. Whether any of the hormone-responsive phosphoproteins we have detected are involved in hormonal regulation of prolactin secretion cannot be determined from the present study. Clearly, A, C, H, I, and P135 peptides are candidates for regulatory phosphoproteins involved in secretion since peptide and cyclic

3309

33 10

Thyrotropin-releasingHormone and

Protein Phosphorylation

Supplemenr r o Thyr~rropin-relessing hormone(TRH) and cyclic AMP activate distinctive pathways of protein phosphorylation in GH pituitary cells Debra S. Drusr. Claudia A. Sutton and Thomas F.J. Martin PlATERlALS AND METHODS rat pituirary cell lines GH Materials Clonal prolacrin-producing ype Culture tolleccion. CCL 82.l)'and CH C (generously providad by A.H. Tashj:an. J r . ) were used in che present srudie4.l Similar results were cbrained vrrh either cell line. Cells were mainrained in Ham's F10 culcure medium supplemenred wlrh 157. horse and 2.57. fetal bovine serum. using prev ~ o u s i ydescribed methods (16). TRH. cyclic cyclic G W . 8-Br cyclic AMP. proceln klnare lnhibimr. carirlyric subunit and gel electrophoresis molecular werghr standards w e r e obtained from Sigma Chemical Corp. Cholera toxin and Reagencs for CAMP radiocryrrallinf urea were obcained from Schuarr-Ham izmunoassay were from Collaborative Research Inc. VIP utilized was either syn(American

AMP.

Falcon or Lux ScieniifLe. SDS. aerylamide and Coomassie Brilliant Blue R were obtained from Bio-Rad Laboratories. Ampholytea were obtained from LKB. Methods: 32P04 incorporation: CH cells were seeded at 2 x 104 cellsfcm' on I O f m d i s h e s In 3 ml cuIcurc medlum and refed ac 2-3d inremala for 10d. S rinsed wich Hepee-buffered (.02M. pH 7 . 5 ) . Prior to labeling. E U ~ L U T ~were rerum-free phosphate-free F10. Three to five ml of chis same F10 formulation concainmg'O.1 mC fml 12P04was then added for a 60-min incubation (37O. 57. COi). . Agents w e d dded for indicated times prior co cerminaiion of the 60-mh la e l m g period. 32P04 incorporation was terminated by aspiracing the culture medrum. rapidly rinsing cultures in ice-sold 0.03M Na phosphate (pH 7 . 6 ) . 0.15 H SaCl and freezing on P block of dry ice. Frozen cultures were thawed in 0.5-1 ml of cold 0.1 H NnF. 0.01 M EDTA. 0.025 M Hepes. pH 7 . 5 . 0.005 H :-mercaptoechanol. 0.002 M phenylmerhene sulfonic acid (PWF). h i s s i o n of PMSF did not a l r e r the r e s u l t s . Cells were removed from che dish with a rubber policeman and additional buffer was added LO rinse the ~ l a t e o . Comb m e d vblumes were homogenized i n a t i ht fitting DDunce homogenizer and the extract centrifuged at 15,000 x g for $0 :in. The supernaranr was centrifuged at 100.000 x g for 60 min in a Beckman type 65 rotor. The pellec was resus-

pended In L s e m l i (17) sample buffer (with I% instead of 2.37. SDS) and rhe supernatant adJusced to 0.4% SDS. 1% a-mercaptoethanol. Samples were heared Equal m o u n t s of TCA-insoluble redloactivity were loaded ongo either 1 D slab gels or isoelectric focusing gels.

. [VIP] (MI

[VIP] (M)

Fig. 3. Effect of VIP concentrecion on protein phoSphorylaiion end C A W a ~ ~ ~ m u l a ~in i oCHn cells: Left panel. Protein phosphorylacion was carried out as described in the iezend EO FIE. 1 with indicaced concentrationso f VIP for 10 min treatment. S c a k n g densitometry was used t o qumcicace peek heights for phosphoproceins present in aupernaranc (547. S 4 1 and S72) or pellec CAMP concenrrscions w e r e derermined in TCA ( P I X ) fraeciona. Righr panel: exfraccs o f CHhCl cells following 10 min of VIP treatment using a radioinmunoassay kic.

on a boiling H 0 bath for 5 mi".

In vitro phosphorylacion was conducred in 50 u l reaCtIon mixtures CODt a i n i e : m 2 M morpholino chanesu foni acid (pH 6.5) 0.01 I4 Q C l 0.008 M dichiorhreitol approx. IO-$ LO 1O-A M Is2-Pl ATP (prepired as descr?6ed by and 75 ug supernatant or pellet Schendel end &lls (18). approx. 5000 Cif-1) er preincubation prorein. Other agents as Indicated were also included. A on ice for 5 min. Z ~ B C L ~ O I I . were started by addition of Y-{$P ATP. incubated at 21' for indicated times and terminated by addition of 25 ill SDS-stop solution ( 0 . 3 M Trrs HC1 (pH 6.8). 1% SDS. 3% a-merceptoechanol and 6% glycerol). Samples were heated On P boiling Hz0 bath for mi" 5 and equal volumes loaded Onto gels a8 described above. Gel electrophoresis: One-dimensional SDS slab gels (14 x 20 x .15 cm) (19). Gels were stained with 0.05% Coomassie Brilliant Blue R. 25% isopropanol. 10% ecetlc acid. Following destaining with 10% P C ~ L I Sacid. 107.methmol. gels were dried onto filcer paper. Autoradiography of dried gels was performed nt -70° usIng Kodak XAR film and an Ilford fast-cungstats intensifying screen. Autoradiograms were scanned using P Zenith soft laser densitometer (LKB). Peak heights for indivldual bands w e r e decermined and normalized t o peaks corresponding to phosphoproreins whose labelingwas observed LO be consistently unchanged by expertmenrsl creatment. The following proceins were utilized to calibrate the SDSdimension of gels: rabbit muscle myosin (200 K). E. coli 0-gelacrosidase (130 K). phosphorylase B (97 K ) . bovine serum a l b m i n (66 K ) . catalase (57 K ) . ovalbumin ( 4 3 K) and carbonic anhydrase (30 K ) . were run as described by O'Fnrrell

SI00

PlOO

-W

Itr

72

-

fc

I35

-47 -41

c

VIP

c E-Er C T cAMP

minutes

at

In vlcro subscrates of the CAMP-dependent protein kinase: In preliminary studies. we arrempred L O identify endogenous substrates Of rhe c M - d e p e n d e n t protein kinase in v i t r o with subcellular fractions o f CH cells. Phosphorylation of severelTupern.ranc peptides (50 K and 120 K) was observed to be stirnulaced by CAMP addiLion ( I ,M) 8 9 shorn in Fig. 7 (left panel). The 120 K prorein was not well resolved from other high molecular welghc phosphoproteins in some experimencs. Stimulated phosphorylation of 50 K peptides was blocked by sddicion of the heat-stable p m t e l n kinase inhibitor. Addition of partially purified catalytic subunit of the 0 - d e p e n d e n r protein klnaoe did not reveal rhe presence o f additional peptide substraces of the enzyme. Addicion of a phosphodiesterase inhibitor. iBObUtylmeLhytx.nLhine. with e M promoted phosphorylation of an additional peptide (60 K). however. since phosphorylation o f 60 K was rrimulsced by HIX alone. it war noc considered co be a substrate o f the c M - d e p e n d e n t enzyme. r M - s r i m u l s r e d phosphorylation of 50 K peptides was maximal within 1 min of incubation at 210. Stirnulaced phosphorylation of supernatant peptides was observed to occur over a narrow range of CAHP c o n ~ e n ~ r s ~ i owith n s B halfmaximal response a t 2 x 10-7 M. Additional studies showed thac LCHP (2 .H) had no effecr on Bupernatant protein phosphorylation. Similar studies conducted with microsomal pellet fractions indicated che presence of TWO high molecular weight (-180 K and 210 K) wbsLraceS for the CAMP-dependent enZ?rme (Fig. 7. right). Phosphorylation of these peptides was enhanced by rrddltion o f C M or carslyric subunit. lnhibitim of stimulated phosphorylation by che hearstable kinase inhibitorwas also observed.

i'2

-

minutes

Fig. 4. Tim-course of VIP effects on protein phosphorylation (left panel) and CAW a c c m l e c i o n (right panel): Experimental methods wereas described i n the legends to Figs. 1 and 2. Left panel 1 24 IP was added rimes indiceced prior LO termination Of 60 min labeline with 3YPOa. Rizht

"

* a

i i i

In vitro phosphorylation reacrions were analyzed by 2D IEFfSDS PAGE t o -re z e q x l y resolve protein kinase substrates. As shorn in Fig. 8. CAW increased phosphorylarion of three 50K pepcides (labeled a. b and J) a s previously seen in ID gels. CAMP also increased phosphorylation of a number o f high molecular weighc p r o ~ e i n s (labeled d.e,f and g , appiox. 120 K ) . I n addition. phosphorylarron Of a 70 K peptide. k. and a low molecular weight protein. c (25 K). was seen t o be CAW-dependent. Several dephosphorylation events were also apparent (labeled h and I). The parcern of c M - d e p e n d e n t PrOLein phosphorylscion observed in ZD analysis was similar L O that defined by in the cyrosol t o ID analysis and increased the number of substrates observed nine.

c VIP

Fig. 1. Effeer o f VIP 8-Br C A M P and cho ra toxin on protein phosphorylation ~n CH cells: Cells bere labelled with $?PO for 6 0 mi" and fracrionared into high-speed supernsranr (S100) and pellet (PlOb) fraccione. SlOO and PlOO fracrione were subiecred LO SDS PACE as described in Methods. Parallel eulcures were either iefr untreated or were treated with VIP (1 .M. 10 min). 8-Br CAHP (1 d.b o min) or cholera toxin (10 ngfml. 60 min) as indicated. Apparent molecular weighre (in kilodalrono) of several bands is Indieaced with arrows.

& vitro counterparts whereas others did not (see Table 1). In addition.