Cyclic AMP and Phorbol Esters Separately Induce Growth Inhibition ...

2 downloads 60 Views 3MB Size Report
Cyclic AMP and Phorbol Esters Separately Induce Growth Inhibition,. Calcitonin Secretion, and Calcitonin Gene Transcription in Cultured. Human Medullary ...
Vol. 261, No. 17, Issue of June 15, pp. 8036-8041,1986 Printed in U.S.A.

THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1986 by The American Society of Biological Chemists, Inc.

Cyclic AMP and Phorbol EstersSeparately Induce Growth Inhibition, Calcitonin Secretion, and Calcitonin Gene Transcription in Cultured Human Medullary Thyroid Carcinoma* (Received for publication, December 23, 1985)

AndrQ deBustrosS, Stephen B. Baylin, Michael A. Levine, and Barry D. Nelkin From the Oncology Center and the DeDartmentof Medicine, The Johns Hopkins UniversitySchool of Medicine, Baltimore, Maryland 21205

We have previously reported that thephorbol ester, 12-0-tetradecanoyl phorboll3-acetate(TPA) induces, in the TT cell line of human medullary thyroid carcinoma, decreased cellular proliferation, increased calcitonin secretion, and enhanced calcitonin gene transcription (deBustros, A., Baylin, s. B., Berger, C. L., ROOS, B.A., Leong, S. S., and Nelkin, B. D. (1985)J. Biol. Chem. 260, 98-104). The cellular responses evoked by TPA are thought to be mediated by protein kinase C. In the present study, we have investigated whether protein kinase A, another key mediator of extracellular signal transduction, may also alter the differentiation status of the TTcells. We find that the effects of CAMP, an activator of protein kinase A, on cellular growth, calcitonin secretion, and calcitonin gene transcription almost parallel those of TPA. We also show that both TPA and cAMP lead to similar increases of both major mRNA species encoded within the calcitonin gene, calcitonin itself and theneuropeptide calcitonin gene-related peptide. In addition, cAMP increases nuclear calcitonin and calcitonin gene-related peptide mRNA precursors toa greater extent(810-fold) than it does the mature cytoplasmic mRNA species (2-4-fold). The effects of TPA and cAMP on the TTcells are additive rather thansynergistic. Furthermore, TPA evokes no increaseinintracellular CAMP. We thus conclude that TPA and cAMP can trigger, independently, in the TT cells, a similarly programmed set of events resulting in a more differentiated phenotype. These cells provide a model system to explore how these two pathwaysof signal transduction converge to regulatemolecular events such as the transcription of the calcitonin gene.

The molecular events underlying cellular responses to external stimuli represent a major focus of investigation of the regulation of cell differentiation. In many systems, the flow of information from the cell surface to thenucleus is mediated by two major biochemical pathways. One is related to the turnover of phosphatidylinositol with subsequent proteinkinase C activation and the other is via the CAMP-dependent protein kinase A (1,2). We have used the humanmedullary thyroid carcinoma TT * This work was supported hy Grant PDT-207 from the American Cancer Society and by the W. W. Smith Foundation. 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. $Supported by a Daland Fellowship from the American Philosophical Society and funds from the Hodson Trust.

cell line as a model system for studying the role of these two major pathways of extracellular signal transduction in endocrine cell differentiation and peptide hormone gene expression. This cell line exhibits several useful features for this purpose. First, the TT cells retain in long-term culture their endocrine characteristics as evidenced by their continued synthesis of large amounts of calcitonin, the main product of the C-cell of the thyroid gland from which the tumor arises (3). Second, the TT cells produce, by alternative splicing of a primary transcript of the single multiexonic calcitonin gene, both calcitoninand CGRP’ messenger RNAs ( 4 ) , thus providing a good in. vitro system for exploring the transcriptional and post-transcriptional regulation of this endocrine gene. Third, calcitonin production by the TT cells exhibits a distinct relationship to thegrowth dynamics of these cells. Thus, calcitonin production is highest when the cells are in stationary growth phase and lowest when the cells are growing exponentially ( 5 ) . Furthermore, this inverse relationship between growth and calcitonin production can be amplified by treatment with the tumor promoter TPA, which produces a marked decrease in DNA synthesis, and stimulation of calcitonin secretion and calcitonin gene transcription (6). The complex series of events induced by TPA in the TT cells is presumably due to activation of the protein kinase C pathway of signal transduction (7). We wondered whether activation of the other major mediator of cellular response, CAMP-dependent protein kinase A, might also modulate the differentiation status of the TT cells. We have thus examined the response of the TT cells to CAMP, including the effects on transcription of the two mRNA species derived from the calcitonin gene, calcitonin, and theneuropeptide, CGRP. We now report that cAMP leads separately, in the TT cells, to cellular responses similar to those induced by TPA, and that the effects of TPA andcAMP are additive. EXPERIMENTALPROCEDURES

Materials Reagents for these studies were purchased as follows: RPMI 1640 medium from B&B Scott, Fiskeville, RI; fetal bovine serum from GIBCO; TPA from Chemicals for Cancer Research, Inc., Eden Prairie, MN; dihutyryl cAMP ((B&cAMP), 8-bromo-CAMP (8-BrCAMP), and sodium butyrate(NaB), from Sigma; forskolin from Calhiochem-Behring; [32P]dCTPand [32P]UTP(3000 Ci/mmol), as well as [3H]thymidine (50 Ci/mmol) from Amersham Corp., Klenow DNA polymerase from New EnglandBiolabs, Beverly, MA, oligo(dT)cellulose from Collaborative Research, Inc., Lexington, M A RNasefree DNase I from Worthington; and nitrocellulose filter paper from Schleicher & Schuell. The abbreviations used are: CGRP, calcitonin gene-related peptide; (Bu),cAMP, dibutyryl cAMP 8-Br-cAMP, 8-bromo-CAMP; NaB, sodium butyrate; TPA, 12-0-tetradecanoyl phorbol 13-acetate.

8036

CAMP and Methods

TPA Increase Calcitonin Expression Gene

8037

their effects are studied, TPA and cAMP can either inhibit or stimulatecellular proliferation (13-15). We hadpreviously Tissue Culture-The TT cell line has been previously described TT cells, TPA inhibits DNA synthesis (6). (3). TT cells were grownin monolayers in RPMI 1640 supplemented shown that, in the with 16% heat-inactivated fetal bovine serum, 2 mM glutamine, 100 We now find that dibutyryl cAMPalso decreases the rate of units/ml penicillin, and 100 pg/ml streptomycin. TPAwas stored in growth of the TT cells. However, the maximal effects of TPA dimethylsulfoxide, at a concentration of1.6 X M at -20 "C. (80%decrease at 1.6 X M) are more profound than those Forskolin was storedin ethanol at a concentration of M. TPA and forskolin from stock solutions were diluted into sterile culture of (Bu),cAMP (60% decrease), even when (Bu),cAMP is used (data not shown). medium immediately prior to use. Final concentrations of dimethyl at the maximal effective dose of2.5mM sulfoxide and ethanol were no greater than 0.1%; these concentrations The time course for inhibition of DNA synthesis for both of dimethyl sulfoxideand ethanol alone had no effects on any of the drugs is shown in Fig. l a . [3H]Thymidine incorporation deparameters of cell function studied. (BU)2cAMP, 8-Br-CAMPand creases 78% in TT cells cultured in thepresence of TPA (1.6 NaB were dissolved in sterile medium immediately prior to use. X M) and 37% in cells treatedwith (Bu)zcAMP ( 1 mM). Assay of DNA Synthesis-Rates of DNA synthesis were evaluated Analysis of each of the individual data pointsfor these doses by measuring the incorporation of [3H]thymidineinto acid-insoluble of drugs suggests that the joint effects of TPA and (BU)~CAMP material, as described previously (6). of DNA synthesisinthe Secretion of Calcitonin-Freshmedium alone or fresh medium ontheinhibition TT cells are containing TPA (1.6 X 10"' M to 1.6 X lo-' M), forskolin (1-100 pM), additive, and notsynergistic. TPA (1.6 X lo-'' M) + forskolin (1 pM), or (Bu)zcAMP,8-Br-CAMP, Dibutyryl cAMP can be metabolized by the cell to cAMP NaB at concentrations of 1 and 2.5 mM was added to the cells. Two and butyrate,a compound known to exert, byitself, profound hours later, aliquots of culture medium in contact with the cells were aspirated and stored at -20 "C for later calcitonin measurements. effects on cell growth (16, 17). We therefore assessed the role Calcitonin was determined by a previously described radioimmunoas- of each of these components of the dibutyryl cAMPmolecule by examining theeffects of another cAMPanalogue, 8-bromosay (8). CAMP Measurements-Fresh medium aloneor medium containing cAMP (8-Br-cAMP),as well as that of sodium butyrate (NaB) TPA, forskolin, or both, at the same concentrations used for calcion theproliferative response of the TT cells. TT cells cultured tonin secretion studies, was added to TT cells grown to confluency in the presence of (Bu),cAMP and 8-Br-CAMP at the maximal in 24-well plates. Incubations were carried out at 37 "C for 20 min. The mediumwas then removed and the cAMPwas extracted by effective concentration of2.5 mM exhibited a comparable adding 0.3 ml of 0.1 N HCl to each well. Samples from each well were decrease in the rate of incorporation of [3H]thymidine into diluted in50 mM sodium acetate (pH 6.3) and acetylated (9).Aliquots DNA (62 and 63% decrease from controlat 6 h for (B&cAMP of each sample were assayed for cAMP by radioimmunoassay, using and 8-Br-cAMP,respectively; Fig. lb), indicating thatcAMP a double-antibodymethod to separate antibody-bound fromfree ligand. Each samplewas assayed in duplicate. Intra-assay and inter- can, by itself, decrease growth in the TT cells. The effects of NaB at 2.5mM on cellgrowth are alsoimpressive. The assay variationwas less than 5 and 8%, respectively. by NaB has cDNA Probes-The calcitonin- and CGRP-specific cDNA probes decrease in [3H]thymidine incorporation induced have been previously described (4).pTT1062 contains only calcitonin- a slower onset than either (Bu),cAMP or 8-Br-CAMP but is specific sequences and no 5' common region or CGRP sequences. It more profound at 48 h. The more prolonged inhibition of wasused as a probe to detect calcitonin-specific mRNA. A Ban11 DNA synthesis in cells treated with (Bu)~cAMP rather than fragment of clone pTT42, containing only CGRP-specific sequences, was used as a probe to detect CGRP-specific mRNA. Recombinant 8-Br-CAMPcould then be due to the effects of the butyrate plasmids containing chicken 8-actin sequences (pAl) were provided released from (Bu),cAMP. Alternatively, 8-Br-CAMPmay be TT cells than (BU)~CAMP. conclude We that by Dr. DonCleveland, the Johns HopkinsUniversitySchool of less stable in the Medicine (10).These probes were labeled with [32P]dCTPto a specific both moieties of the dibutyryl cAMP molecule can decrease activity ofup t o 1 X lo9 cpm/pg by random priming with Klenow TT cells, although the alterations cellular proliferation in the DNA polymerase as described previously (6). in growth dynamics in response to the cAMPanalogues and Quuntitation of Calcitonin, CGRP, and 8-Actin mRNAs-In some experiments, the cytoplasmic dot-blot hybridization methodof White sodium butyrate are somewhat different. Effects of CAMP and TPA on Calcitonin Secretion-We and Bancroft (11)was used to analyze levels of mRNA. Cytoplasmic cell extracts from triplicate wells were serially dilutedand applied to next studied theeffects of cAMP on calcitoninsecretion. We nitrocellulose filters with a Hybridot manifold (Bethesda Research treated the TT cells for a period of 2 h with the two cAMP Laboratories). These nitrocellulose filters were hybridized to the 32P- analogues, (Bu),cAMP and 8-Br-cAMP, andfound that both labeled calcitonin, CGRP,and 8-actin-specificcDNA probes, as previously described (6). The intensity of the hybridization signals was analogues, at 1 and 2.5 mM, increased calcitonin secretion intothemedium by 2-3-fold above control (Fig. 2a). In quantitated by densitometry. In some experiments, the above mRNA species were characterized contrast to its effects on DNA synthesis, NaB alone was by Northern blotting procedures. Cytoplasmic and nuclear RNA were ineffective in promoting calcitonin secretionat doses as high prepared accordingto Maniatis et al. (12). as 2.5 mM. Thus, the effects of (BU)~CAMP on calcitonin In some instances, poly(A)+ RNAwasisolatedbypassage of secretion by the TT cells are similar to thoseobserved previcytoplasmicRNA through an oligo(dT)-cellulosecolumn prior to fractionation on agarose gels. Gel electrophoresiswas carried out in ously for TPA. We then used forskolin, a diterpene which directlyactivates 1.5% agarose containing formaldehyde. RNA was then transferred to a nitrocellulose filter whichwashybridized under the conditions adenylate cyclase and increases intracellular cAMP (18), in described previously (6). order tosee whether calcitonin secretionby the TT cells can Nuclear Rumff Transcription-Isolation of nuclei, in vitro nuclear be stimulated by increasingendogenous CAMP. Forskolin ( 1 transcription, and isolation of 32P-labeledRNA were performed as ) t o a marked increase in calcitonin secretion previously described (6). Equal numbersof counts (10-30 X lo6 cpm) and 10 y ~ led from treated and control cells were hybridized to plasmids containing into the medium (3- and 5-fold increase over control, respeccalcitonin, CGRP, and P-actin-specific sequences immobilized on a tively) (Fig. 2b), accompanied by a dramatic increase in intranitrocellulose filter. Hybridization conditions wereas described pre- cellular cAMP(Table I). Increasingtheconcentration of viously (6). forskolin to 100 p~ did not lead t o a further increase in calcitonin secretion (data notshown). RESULTS Analysis of the results for the joint treatment of the TT TPA and cAMP Decrease Cellular Proliferation in the TT cells with a low dose of forskolin (1 PM) and a low dose of Cells-In many types of cells, both TPA and cAMP have TPA (1.6 X 10-l' M), shows that thetwo compounds have an profound effects on growth. Depending on the systemwhich in additive effect on calcitoninsecretion. This suggests that TPA

8038

cAMP and TPA Increase Calcitonin Gene Expression a

--e FIG. 1. Effectof (a)(BU)~CAMP (1 lo-' M), or both, and (b) (Bu)*cAMP, Ei-Br-cAMF', andNaB (2.5 mM), on DNA synthesis in the TT cells. Ateachtimepoint, incorporation of [3H]thymidine into acid-insoluble material was calculated as a percentage of the incorporation in untreated controls. The means of triplicate wells are shown. Standard errors were less than +.20% of the mean for each time point.

8

m), TPA (1.6 x

-

k '"t I\

-

.._...... No6 2.5mM

c

$

; :.

f

f

$ E! e =, u

I

I

I

24

48

Hours

b a

24 Hours

48

140

60 -

,5

5

FIG. 2. Effectof (a)(BU)~CAMP, p 8-Br-CAMP, and NaB, and (b) forskolin, TPA,orforskolin + TPAon F calcitoninrelease into the medium. .- = Equalnumbers of TT cells in 24-well plates were incubated for 2 h with 1 ml 2 of medium with orwithout the various compounds mentioned, at the concentra- y tions indicated. The means and standard 5 errors of triplicate wells are shown. Cal- 5 citonin secretion intothe medium is in 8 nanograms/milliliter.

P

100

4 0-

!

62

'0

-

TABLE I Intracellular CAMPaccumulation in response to forskolin and TPA Equal numbers of TT cells in 24-well plates were incubated with 1 ml of medium with or without forskolin, TPA, or forskolin + TPA at the concentrations indicated. At 20 min, the medium was removed and the cells were harvested for intracellular cAMP measurements. The means and standarderrors of triplicate wells are shown.

cells were mediated by an increase in intracellular CAMP. Whereas forskolin, within 15 min, increased calcitonin secretion 3-&fold and intracellular cAMP more than 300-fold, TPA, in doses of 1.6 x 10-l' to 1.6 X IO-', increased calcitonin secretion 2-3-fold, but had no effect on intracellular cAMP levels (Fig. 2b and Table I). We conclude that TPA affects the growth and calcitonin secretion of the TT cells by a cAMP mechanism independent of CAMP. pmol/well (BU)~CAMP Increases Calcitonin and CGRP mRNA in the Control 9 f 1.0 TT Cells-Our previous studies demonstrated that TPA, in addition to stimulation of calcitonin secretion, also increases 8 f 0.6 TPA (1.6 X 10"' M) 11f 1.7 the transcription of the calcitonin gene (6). We have now TPA (1.6 x 10-9 M I 8 f 1.2 TPA (1.6 X M) investigated the effects of cAMP on the production of calcitonin mRNA by the TT cells. Treatment of the TT cells with 182 f 21.0 Forskolin (1 p M ) (BU)~CAMP (1mM) for 48 h led to a marked increase in the 1503 j r 33.0 Forskolin (10 p M ) 1-kilobase calcitonin mRNA species, as shown by Northern f 127.0 3113 Forskolin (100 p M ) hybridization to thecalcitonin-specific probe, pTT1062 (Fig. TPA (1.6 X 10-lo M) + forskolin (1 pM) 194 +_ 10.7 3). A 2.2-kilobase actin mRNA species remained unchanged by the (BU)~CAMP treatment. Actin was thus used as a control for the subsequent series of experiments, as was done in our and CAMP,in regard to their effect on DNA synthesis in the TT cells, may act independently to bring about increased previous study of TPA (6). We used a dot-blot hybridization assay to determine calcitonin secretion by these cells. whether the increase in calcitonin mRNA induced by Effects of TPA on Intracellular CAMP-Both TPA and is due to the cAMP moiety. We used the CAMP cAMP thus appear to alter similarly the growth and calcitonin (BU)~CAMP secretion pattern of the TT cells. Yet, the additive effects of analogue 8-Br-CAMP as well as sodium butyrate alone, since the two compounds administered simultaneously, suggested butyrate by itself is known to alter the expression of several separatepathways of action. We assessed this possibility genes (19-21). Both (Bu)+AMP and 8-Br-CAMPled, within further by determining whetherthe effects of TPA on theTT 24 h of treatment, to a 2-4-fold increase in calcitonin-specific

CAMPand TPA Increase Calcitonin Gene Expression (BulzcAMP Control

I mM

8039

Control (C) Forskolin(F1 "

Ratio

F/ c = 3

FIG. 5. Dot-blot hybridization to pTT1062 of cytoplasmic extracts from control cells and fromcells treated with forskolin (10 PM) for 3 h. No change in calcitonin-specific mRNA in FIG. 3. Northern blot of 1 fig of poly(A)+ RNA extracted from control cells and from cells treated with(Bu)zcAMP (1 mM) for 48 h. Hybridization is to the pTT1062 calcitonin-specific probe, and then to the pAl @-actin probe, without removal of the calcitonin probe. CT,calcitonin.

response to forskolin was observed at earlier time points (1/z h, 1%h), and calcitonin-specific mRNA was back to control levels after 6 h of treatment (data not shown). TPA(T) *

Control TPAITI (CI

(BU1zcAMP(B1 (BUlpcAMP(BI

C T * . @ . . . . O . . . . Rolros

CGRP

T/C:?

35

B/C:24

T+B/I,:3

3

*.....e.#.. RO~IOS TIC: 2 3

B/C - 2 8

T*B/C:4

FIG. 6. Calcitonin- and CGRP-specific mRNA response to treatment of the TT cells with (Bu)*cAMP (1mM), TPA (1.6 X lo-'), or both. Duplicate nitrocellulose filters were hybridized to either the calcitonin-specific pTT1062or to the CGRP-specific Ban11 fragment of pTT42.

expression exists in the rat and is mediated by alternative splicing of a primary RNA transcript of the calcitonin gene (22-24), we examined whether TPA and/or cAMP treatment would result in a selective increase in calcitonin or CGRP mRNA. 0 0 Using the previously described calcitonin- and CGRP-spe0 . cific probes (4), we found that both TPA (1.6 x lo-' M ) and (Bu),cAMP (1mM) lead to a 2-4-fold increase in both mRNA species as detected by dot-blot hybridization (Fig. 6).We conclude that neither TPA nor cAMPaffects the alternative FIG.4. Expression of calcitonin-specific mRNA in response processing choice of the calcitonin gene primary RNA tranto treatment of the TT cells for 24 h with (BU)~CAMP, 8-BrmRNA in theTT cells. CAMP,and NaBat the concentrations indicated.Hybridization script to calcitonin or CGRP

(Bu),cAMP and TPA Increase Both Calcitonin and CGRP mRNA atthe TranscriptionalLevel-In order to further define the level of regulation of calcitonin and CGRP mRNA by CAMP,we examined by Northern blotting the production, by the TT cells, of both nuclear and cytoplasmic calcitonin and CGRP transcripts. We found that (BU)~CAMP (1 mM) leads within 24 h to a dramatic and comparable accumulation of nuclear precursors (8-10-fold mRNA (Fig. 4). NaB at similar concentrations did not affect bothcalcitoninandCGRP 7). Interestingly, theincreaseinthe above controls) (Fig. the levels of calcitonin mRNA after 24 h of treatment. We cytoplasmic calcitonin and CGRPspecies in this same study also used forskolin in order to see whether theeffects of the is more modest (2-3-fold above controls for both calcitonin exogenous cAMP derivatives on calcitonin mRNA can be and CGRP), thus raising thepossibility of a regulatory step mimicked by an increase in endogenous intracellular CAMP. in calcitonin gene processing involving RNA transport from We found thatforskolin at a concentration of 10 p~ led to a the nucleus to thecytoplasm. transient increase in calcitonin mRNA that peaked after 3h T o determine whether the accumulation of nuclear and of treatment (Fig. 5). By contrast, the increase in calcitonin cytoplasmic calcitonin and CGRP mRNA after cAMP stimmRNA induced by (B&cAMP and 8-Br-CAMP peaked after ulation is due to stabilization of the mRNA or is a direct 6 h of treatment and was maintained up to 48 h (data not consequence of enhanced calcitonin gene transcription, we shown). The reasons for the different calcitonin mRNA dy- examined in vitro transcription in nuclei isolated from cells namics in response to the above agents are unclear to us at treated for 3 h with medium aloneor medium to which present, but may involve a differential stability of the calci- (BU)~CAMP (1 mM) was added. The radiolabeled nascent tonin mRNA under the different conditions of treatment. The RNA was hybridized to calcitonin and CGRPsequence-coneffect of forskolin documents, however, that endogenous as taining plasmids immobilized on nitrocellulose filters. A 8well as exogenous cAMP can promote an increase in cytoactin plasmidwasused as control. As shownin Fig. 8, plasmic calcitonin mRNA in theTT cells. (B&cAMP led to a marked and comparableincrease in both Since an inverse relationship between calcitonin and CGRP calcitonin- and CGRP-specific hybridization relative to the

was to the pTT1062 calcitonin-specificprobe. Here, as well as in Fig. 5 and 6, quantitation of each dot was done by densitometry. The values given represent the means of triplicate dots, usually the second dilution. Rehybridization of the same nitrocellulose filter to a 8-actin probe, after removal of the calcitonin-related probes used, revealed no change in the 8-actin-specific signal in treated versus control cells (data not shown).

cAMP and TPA Increase CalcitoninGene Expression

8040 b

d

sulting indecreased cellularproliferation, increased calcitonin secretion, and enhanced transcriptionof the calcitonin gene, with accumulation of both the calcitonin and CGRP mRNA species derived from this gene. The effects of TPA and cAMP on individual cell systems CT are complex, and vary with the type of cell studied. In turn, RNA 4 the interactions between TPA and cAMP can be dependent upon interactions between components of the two pathways k “Ikb which these two agents activate (25-27). CAMP, a key mediator of transmembrane signaling, is believed to exert its cellular effects by activating acascade of protein phosphorylation b C d events, which are mediated by the CAMP-dependent protein kinase A (2). TPA, on the other hand, is thought to act as a substitute for diacylglycerol, a product of the phosphatidylinositol breakdown induced by physiological extracellular signals and an activator of another kinase pathwaymediated by protein kinase C (7). In many cellular systems, the protein kinases A and C pathways are antagonistic. Thus, cAMP inhibits phosphatidylinositol breakdown-induced cellular processes such as serotinin release in platelets (28).However, several other endocrine systems are described in the literature, FIG.7. Northern blot of 2.5 pg of nuclear ( A and B, lanes a in which TPA and cAMP act independently or, in an additive and b) and cytoplasmic ( A and B , lanes c and 6)RNA extracted manner, to evoke similar cellularresponses. Of particular from TT cells treated with medium alone ( A and B. lanes a and c) or medium with (Bu)zcAMP (1 mM) for 24 h ( A and B , relevance to our present study is the observation that both secrelanes b and 6).Hybridization is to the calcitonin-specific probe ( A ) TPA and cAMP appear to stimulate peptide hormone culture. Thus, in the and to the CGRP-specific probe (R).The arrows indicate the major tionin several endocrinetumorsin nuclear RNA species hybridizing to the calcitonin and CGRP probes. mouse pituitary tumor cells, At-T20, TPA as well as several The lower bands co-migrate with the mature mRNAs, and the higher secretagogues known to increase intracellular cAMP such as molecular weight bands are precursorswe and othershave previously corticotropin-releasing factor or forskolin (29, 30), increase described (4, 24). The signals were quantitated by densitometry. Kb, adrenocorticotropin hormonesecretion into themedium, Simkilohase. ilarly, in primary pituitarycell cultures, both TPA and growth CGRP CT BActin hormone-releasing hormone, which activates the adenyl cyclase pathway, increasegrowth hormone release intothe medium (31). Most interestingly, in the well-studied rat pia Control tuitarytumor GH,cells, bothTPAandcAMPstimulate prolactin secretion as well as prolactin gene transcription (3234). The TT cells then represent asecond example of an endocrine cell system in which the TPA-dependent protein kinase C, and CAMP-dependent protein kinase A are shown to act in concert toincrease the secretion and production, a t the transcriptionallevel, of a small polypeptide hormone. In addition to the stimulation of calcitonin secretion and calcitonin gene transcription, both TPA and cAMP decrease Ratios CGRP( BUI~CAMP); Con+rol CT ((BU Control I~cAMP DNA synthesis in the TT cells. There is a well-recognized, FIG.8. Nuclear runoff transcription in nuclei isolated from frequent inverse relationship between growth and theexprescells incubated with medium alone or in medium containing sion of the differentiated function of a cell. However, i t is (Bu)zcAMP(1 mM) for 3 h. The intensityof the signal of calcitonin, often difficult to establish whether the changes in gene tranCGRP, and P-actin-specific nascent RNA was quantitated by densi- scription are directly dependent upon a decrease in overall tometry. The alteration of the calcitonin- and CGRP-specific signals in response to (Ru)$AMP is expressed relativeto the 6-actin-specific DNA synthesis or whether these two events are only temporally related to the differentiation process. It is then of imsignal used as control. portance in our system that butyrate hasa profound inhibiP-actin-specific signal (6- and %fold, respectively). The mag- tory effect on the growth of the TT cells, but evokes no nitude of this increase is close to that just shown for the increase in calcitonin secretion or calcitonin gene transcripnuclear calcitonin and CGRP RNA species following tion under the experimental conditions used in this study. Thus, the effects of TPA and cAMP oncalcitonin secretion (Bu)~cAMP treatment. The data thus indicate that (Bu),cAMPaffectscalcitoninandCGRPproductionto a and calcitonin gene transcription do not appear to be simple consequences of their ability to slow down cell growth. Insimilar extent at the transcriptional level. The increases in stead, TPA and cAMP seem to elicit a series of temporally calcitonin mRNA and CGRP mRNA induced by TPA are similarly due to an increase in the transcription of the calci- linked cellular responses. tonin gene by TPA (nuclear runoff transcription data for The effects of TPA and cAMP on each component of the calcitonin previously published (6) and data for CGRP, not above-mentionedintegrated cellularprogram are additive. shown). Since TPA evokes no increase in intracellular cAMP in the TT cells, we conclude that the interaction between the two DISCUSSION biochemical pathwaysactivated by cAMP and TPA must The data presented here suggest that both TPA and cAMPoccur at a step distal to the generation of CAMP, such as protein phosphorylation. It is conceivable that a cascade of elicit, in the TT cells, an integrated biological response re~o

C

F

e

I

CAMPand TPA Increase Calcitonin Expression Gene cellular phosphorylation events, triggered by the two kinase systems, could eventually bring aboutthe phosphorylation of common proteins which, in turn, trigger events such as peptide secretion, growth inhibition and gene-specific transcription. Two observations regarding protein phosphorylation are of special relevance to our system. First, the TPA-induced growth inhibition of both the HL60 leukemic and A431 epidermoid carcinoma cell lines is associated with the phosphorylation of proteins which are substrates of the CAMP-dependent protein kinaseA (35). Second, in theGH, cells, when transcription of the prolactin gene is enhanced by either TPA or CAMP, a non-histone basic nuclear protein is concomitantly phosphorylated (33). The phosphorylation in the TT cells of identical proteinsby the two different kinasesystems activated by TPA and and cAMP could then explain the similarity of cellular responses induced by these two compounds. An alternative explanation to our findings that TPA and cAMP act separately to initiate similar responses in the TT cells may be that these compounds could exert effects independent of theirprotein kinase activating properties. For example, in eukaryotic cells, cAMP may regulate gene transcription by direct binding to a receptor which interacts with specific DNA sequences, as occurs in prokaryotes (36). Interestingly, a consensus nucleotide sequence in the 5' flanking area is shared by CAMP-responsive prokaryotic as well as eukaryotic genes such as rat prolactin, rat phosphoenolpyruvate carboxykinase, and rat tyrosine aminotransferase (37, 38). The availability of cloned DNA fragments of the human calcitonin gene, and more recently, of a genomic 5' calcitonin gene flanking region should allow us to identify one or several DNA regions responsible for the regulation of calcitonin gene transcription by TPA or CAMP. Finally, our current results directly address regulation of the complex expression of the calcitonin gene, including posttranscriptional events. Thus, in our cell system, in addition to the stimulation of calcitonin gene transcription, TPA and cAMP both enhance the production, in comparable amounts, of the two mRNA species encoded within the calcitonin gene, calcitonin itself, and CGRP. These data are consistent with previous observations that transcription always proceeds through the entire coding region of the calcitonin gene, but that the final calcitonin to CGRP mRNA ratio is determined by post-transcriptional processing events(39). In our cell system, nuclear precursor RNAs for calcitonin and CGRP are increased by cAMP to the same extent. Regulation then occurs such that a lesser amount of these transcripts appears in thecytoplasm. Also, the splicing system of the cells is able to maintain similar ratios of calcitonin and CGRP mRNAs during both basal state and after stimulation. Our system is then an excellent model to study the tissue-specific factors that regulate each component of calcitonin gene expression, especially during stimulation with agents such as TPA or CAMP. Acknowledgments-We thank Sara Lyles for technical assistance and Alverta Fields for typing the manuscript. REFERENCES 1. Kishimoto, A., Takai, Y., Mori, T., Kikkawa, U., and Nishizuka, Y. (1980) J. Biol. Chem. 255,2273-2276 2. Kuo, J. F., and Greengard, P. (1969) Proc. Natl. Acud. Sci. U. S. A. 6 4 , 1349-1355 3. Leong, S.S., Horoszewicz, J. S., Shimaoka, K., Friedman, M., Kawinski, E., Song, M. J., Zeigel, R., Chu, T. M., Baylin, S. B.,

8041

and Mirand, E. A. (1981) in Advances i n Thyroid Neoplasia 1981 (Andreoli, M., Monaco, F., and Robbins, J., eds) pp. 95108, Field Educational Italia, Rome, Italy 4. Nelkin, B. D., Rosenfeld, K. I., de Bustros,A., Leong, S. S., Roos, B. A., and Baylin, S. B. (1984) Biochem. Biophys.Res. Commun. 123,648-655 5. Berger, C. L., deBustros, A., Roos, B. A., Leong, S. S., Mendelsohn, G., Gesell, M. S., and Baylin, S. B. (1984) J. Clin. Endocrinol. Metub. 59,338-343 6. deBustros, A., Baylin, S. B., Berger, C. L., Roos, B. A., Leong, S. S., and Nelkin, B. D. (1985) J. Biol. Chem. 260,98-104 7. Castagna, M., Takai, Y. Kaibuchi, K., Sano, K., Kikkawa, U., and Nishizuka, Y. (1982) J. Biol. Chem. 2 5 7 , 7847-7851 8. Baylin, S. B., Wieman, K. C., O'Neil, J. A., and Roos,B. A. (1981) J. Clin. Endocrinol. Metab. 5 3 , 489-497 9. Harper, J. F.,and Brooker, G. (1975) J. Cyclic Nucleotide Res. 1 , 207-218 10. Cleveland, D. W., Lopata, M. A., MacDonald, R. J., Cowan, N. J., Rutter, W. J., and Kirschner, M. W. (1980) Cell 20,95-105 11. White, B.A., and Bancroft, F. C. (1982) J. Biol.Chem. 2 5 7 , 8569-8572 12. Maniatis, T., Fritsch, E. F., and Sambrook, J. (eds) (1982) Molecular Cloning, A Laboratory Manual,pp. 191-192, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 13. Abrahm, J., and Rovera, G. (1980) Mol. Cell. Biochem. 3 1 , 165175 14. Dicker, P., and Rozengurt, E. (1978) Nature 2 7 6 , 723-726 15. Friedman, D. L. (1976) Physiol. Reu. 56, 652-707 16. Prasad, K. N., and Sinha, P. K. (1976). I n Vitro (Rockuille) 12,

"_

"_.?x

125-1

17. Hagopian, H. K., Riggs, M. G., Swartz, L. A,, and Ingram, V. M. (1977) Cell 12,855-860 18. Seamon, K. B., Padgett, W., and Daly, J. W. (1981) Proc. Natl. Acad. Sci. U. S. A. 78,3363-3367 19. Reeves, R., and Csejesi, P. (1979) J. Biol. Chem. 2 5 4 , 42834290 20. Green, R., and Shields, D. (1984) Endocrinology 1 1 4 , 1990-1994 21. Stanley, F., and Samuels, H. H. (1984) J. Biol. Chern. 259,97689775 22. Rosenfeld, M. G., Mermod, J-J., Amara, S. G., Swanson, L. W., Sawchenko, P. E., Rivier, J., Vale, W. W., and Evans, R.M. (1983) Nature 3 0 4 , 129-135 23. Steenbergh, P. H., Hoppener, J. W. M., Zandberg, J., Van de Ven, W. J. M., Jansz, H. S., and Lips, C. J. M. (1984) J. Clin. Endocrinol. Metab. 5 9 , 358-360 24. Edbrooke, M. R., Parker, D., McVey, J. H., Riley,J. H., Sorenson, G. D., Pettengill, 0. S., and Craig, R. K. (1985) EMBO J. 4 , 715-724 25. Nishizuka, Y. (1984) Science 2 2 5 , 1365-1370 26. Kaibuchi, K., Takai, Y., Ogawa,Y., Kimura, S., Nishizuka, Y., Nakamura, T., Tomomura, A., and Ichihara, A. (1982) Biochem. Biophys. Res. Commun. 1 0 4 , 105-112 27. Kelleher, D. J., Pessin, J. E., Ruoho, A. E., and Johnson, G.L. (1984) Proc. Natl. Acud. Sci. U. S. A. 8 1 , 4316-4320 28. Takai, Y., Kaibuchi, K., Sano, K., and Nishizuka, Y. (1982) J. Biochem. (Tokyo) 91,403-406 29. Heisler, S. (1984) Eur. J. Phurmacol. 98, 177-183 30. Heisler, S., and Reisine, T. (1984) J. Neurochem. 4 2 , 1659-1666 31. Ohmura, E., and Friesen, H. G. (1985) Endocrinology 116, 728733 32. Osborne, R., and Tashjian, A. H., Jr. (1981) Endocrinology 108, 1164-1170 33. Murdoch, G. H.,Rosenfeld, M. G., and Evans, R.M. (1982) Science 218,1315-1317 34. Supowit, S. C., Potter, E., Evans, R.M., and Rosenfeld, M. G . (1983) Proc. Natl. Acad. Sci. U. S. A. 8 1 , 2975-2979 35. Feuerstein, N., Sahai, A., Anderson, W. B., Salomon, D. S., and Cooper, H. L. (1984). Cancer Res. 44, 5227-5233 36. Peterkofsky, A. (1976). Adu. Cyclic Nucleotide Res. 7 , 1-48 37. Wynshaw-Boris, A., Gross Lugo, T., Short, J. M., Fournier, R. E. K., and Hanson, R.W. (1984). J. Biol.Chem. 2 5 9 , 1216112169 38. Nagamine, Y., and Reich, E. (1985) Proc. Natl. Acud. Sci. U. S. A. 82,4606-4610 39. Amara, S. G., Evans, R.M., and Rosenfeld, M. G . (1984) Mol. Cell. Biol. 4, 2151-2160