cAMP-PKA-Creb signal pathway

The cAMP signalling pathway activates CREB through PKA,

p38and MSK1in NIH 3T3cells

Marit Pedersen Delghandi,Mona Johannessen,Ugo Moens *

Department of Biochemistry,Institute of Medical Biology,University of Troms ?,N-9037Troms ?,Norway

Received 14January 2005;accepted 9February 2005

Available online 16March 2005

Abstract

Cyclic adenosine 3V ,5V -monophosphate (cAMP)was originally shown to induce gene transcription through activation of cAMP-dependent protein kinase (PKA),and subsequent phosphorylation of the transcription factor cAMP response element-binding protein,CREB,at serine-133.However,elevated cAMP levels may activate multiple signalling pathways with protein kinases that can phosphorylate CREB at serine-133.We analysed the pathways involved in CREB phosphorylation and activation in NIH 3T3cells exposed to the cAMP elevating agent forskolin.PKA represented the predominant pathway during the burst phase,while the mitogen-activated protein kinase p38pathway became activated in a PKA-dependent fashion in forskolin treated cells.The phosphorylation kinetics of p38was delayed compared to PKA activation.Activated p38stimulated CREB-mediated transcription and potentiated the transcriptional strength of CREB provoked by forskolin.The p38-mediated activation of CREB was inhibited by dominant negative mutants of MSK-1and by the PKA/MSK-1inhibitor H89,but not by dominant negative mutants of MSK-2/RSK-B and MAPKAPK2.Our results suggest that forskolin-induced CREB phosphorylation and activation in NIH 3T3cells is mediated directly by PKA and by a time-delayed PKA-dependent p38/MSK-1pathway.This bifurcation and time-dependent regulation of the cAMP-responsive signalling pathways may enable the cell to endure and/or enforce a cellular response provoked by a cAMP-elevating stimulus.D 2005Published by Elsevier Inc.

Keywords:CREB;PKA;p38;MSK1;Forskolin;NIH 3T3cells

1.Introduction

It is generally accepted that cyclic adenosine 3V ,5V -mono-phosphate (cAMP)-induced gene transcription is mediated through activation of cAMP-dependent protein kinase or protein kinase A (PKA),and subsequent phosphorylation of the transcription factor cAMP response element-binding protein,CREB.CREB binds the conserved cAMP response element TGACGTCA typically found in the promoter of many cAMP-responsive genes.The bipartite transactivator domain of CREB is involved in basal and signal-inducible transcription activation by CREB.The glutamine-rich,constitutive activation domain (Q2or CAD)interacts with the general transcription factor TAF4(formerly known as TAF II 130/135)and allows recruitment of the RNA polymer-ase II initiation complex.The activity of the kinase-inducible domain (KID)of CREB’s transactivation domain was found to be regulated by PKA,thus forming the missing link in a linear signalling cAMP–PKA–CREB pathway.It was shown that serine residue 133in KID was a unique phosphoacceptor site for PKA in vitro and in vivo,and that phosphorylation of CREB at this residue promotes recruitment of the general co-activator CBP and its paralogue p300.CBP and p300stimulate CREB-mediated transcription through their inter-action with the RNA polymerase II complex and through their intrinsic histon acetyltransferase activity [1–4].

Although correct,cAMP/PKA-dependent phosphorylation of CREB at Ser-133is not the only mechanism that activates CREB-mediated transcription.Ramification of the cAMP pathway contributes to the diversity of signalling pathways that can modulate CREB activity.Indeed,it was originally believed that PKA was the only target for cAMP .However,cAMP was also found to bind and activate the guanine

0898-6568/$-see front matter D 2005Published by Elsevier Inc.doi:10.1016/j.cellsig.2005.02.003

T Corresponding author.Tel.:+4777644622;fax:+4777645350.E-mail address:ugom@fagmed.uit.no (U.Moens).

Cellular Signalling 17(2005)

1343–1351

https://www.360docs.net/doc/149120346.html,/locate/cellsig

nucleotide-exchange factors Epac1and Epac2.These proteins can activate the GTPase Rap1,which in turn binds B-Raf. cAMP-induced activation of Rap1through Epac is clearly independent of PKA[5,6].The B-Raf–Rap1complex has been shown to promote activation of the mitogen-activated protein kinases(MAPK)ERK1/2in different cells and to regulate CREB-dependent gene expression[7–10].ERK1/2can activate mitogen-activated protein kinase activated protein kinases.Some of these,including mitogen and stress-activated kinase1(MSK1),RSK2(also called MAPKAPK1b),and MSK2(=RSK-B)have been shown to phosphorylate CREB at serine133[11–13].Elevated cAMP levels can also activate the p38and the MEK5/ERK5MAPK pathways and CREB has been shown to be a substrate for the p38-regulated MAP-KAPK2and MAPKAPK3[14–16].Although ERK5has been shown to be involved in neurotrophin-induced CREB phos-phorylation[17],a role for MEK5/ERK5in cAMP-induced CREB phosphorylation is missing.Serine133turned out to be a target for protein kinase B(PKB),but the exact signalling pathway that leads to PKB-induced CREB phosphorylation remains unclear[18].Interestingly,cAMP activates PKB through a PI3-K dependent mechanism in a cell-type-specific manner[19–21],opening for an alternative pathway for cAMP-induced CREB phosphorylation.Another target for cAMP are the cyclic-nucleotide-gated ion channels[22].They conduct the release of Ca2+,which may result in stimulation of calmodulin-dependent protein kinases(CaMK)and conven-tional protein kinases C(PKC).Both CaMK and PKC can modulate CREB phosphorylation at ser-133and enhance CREB-mediated transcription,although the exact role of these kinases in CREB phosphorylation remains elusive[1].

The adenylyl cyclase activator forskolin is an extremely valuable tool to study cellular processes in response to increased intracellular cAMP levels[23].Hence,forskolin has often been employed in studies with CREB.Because of the ramification of the cAMP signalling pathway,we wanted to identity which of the cAMP transduction pathways that could mediate forskolin-induced CREB phosphorylation in NIH3T3cells.Our results show that cAMP/PKA is the major pathway explored in forskolin-treated NIH3T3cells,but that PKA may also activate p38,which in turn can phosphorylate and activate CREB through the MAPKAPK MSK1.The PKA-induced activation of the p38-MSK1is delayed compared to the cAMP–PKA activation,suggesting that the former pathway may be used when the cAMP–PKA pathway is attenuated.The cross talk between PKA and p38may allow the cells to prolong the effect of an incoming signal.

2.Material and methods

2.1.Reagents

Forskolin was purchased from Sigma(MO,USA),and newborn Calf Serum(NCS)was from Bio Whittaker (Verviers,Belgium).Cell culture medium was obtained from Gibco/BRL(MD,USA).CDP-Star and MagicMark Western Standard were from Applied Biosystems(CA, USA)and Invitrogen(Oslo,Norway),respectively.Anti-bodies against CREB,phospho-CREB,p38MAPK, phospho-p38MAPK,and the monoclonal antibody phospho-p44/42MAPK were all obtained from Cell Signaling Technology(CA,USA).Luciferase assay kit was from Perkin-Elmer-Tropix(MA,USA).H89and phospho-ERK5/BMK1antibody were purchased from Biosource International(Belgium).LY294002,SB203580 and KN-62were obtained from Alexis Biochemicals (Lausen,Switzerland).Alkaline phosphate conjugated anti-rabbit and anti-mouse antibodies were purchased from DAKO(Denmark)and Sigma-Aldrich(MO,USA), respectively.The specific Epac activator8-pMeOPT-2V-O-Me-cAMP was obtained from Biolog Life Science Institute (Bremen,Germany).

2.2.Cells

NIH3T3cells(ATCC CRL1658)were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10%(v/v)NCS,penicillin(100U/ml),and100A g/ml streptomycin(Gibco/BRL)in a CO2incubator(5%CO2) at378C.All experiments with NIH3T3cells were performed with cell passages between130and140.

2.3.Plasmids

The GAL4–CREB expression plasmid was kindly provided by Dr.R.A.Maurer[24].The coding sequence of CREB was subcloned in the GAL4–DBD expression vector pM(Clontech)and subsequently verified by sequencing.The luciferase reporter plasmid pG5E1bLuc, which contains five binding sites for the yeast tran-scription factor GAL4upstream of an Adenovirus E1b TATA-box and a luciferase gene,was obtained from Dr. R.J.Davis[25].The dominant negative mutants of MSK1,N-terminal kinase dead MSK1and C-terminal kinase dead MSK1,were kindly provided by Dr.M.Deak [26].The empty expression vector pSR and the plasmid pSR C a,containing the catalytic C a subunit from PKA, were a kind gift from Dr.T.Jahnsen and Dr.K.Taske′n [27].Wild type p38in the pcDNA3-HA vector was kindly provided by Dr. A.Brunet[28],and the constitutively active MKK6,MKK6E/E,was obtained from Dr. D. Zechner[29].The plasmid pcDNA3.1was purchased from Invitrogen,while pCMV5was kindly provided by Dr.E.Sontag[30].

2.4.Sequencing

Cycle sequencing was performed using the Big Dye sequencing kit(Perkin-Elmer-Tropix).Sequencing reac-tions were analysed on an ABI377Prism Sequencer (Perkin-Elmer).

M.P.Delghandi et al./Cellular Signalling17(2005)1343–1351 1344

2.5.Transient transfection and luciferase assay

Transfection of NIH3T3cells and luciferase assays were as previously described[31].

2.6.Immunoblot analysis

All immunoblot analyses were performed on extracts derived from cell grown in6-welled trays.Cells were washed twice in PBS,harvested in80A l lysis buffer(0.25 M DTT,and1:1NuPAGE LDS Sample Buffer(4?)and ddH2O),heated to708C for10min and sonicated for4s. Samples were analysed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis(SDS-PAGE)(4–12% NU-PAGE;Invitrogen)and transferred to0.45-A m-pore-size polyvinylidene difluoride membrane(Millipore).The blots were blocked with5%non-fat dry milk(Nestleˉ)/ 0.1%Tween-20(blocking buffer)for1h at room temperature and incubated with primary antibodies (1:1000)overnight at48C in blocking buffer.The blots were subsequently incubated for1h in room temperature with alkaline phosphate conjugated anti-rabbit antibodies (DAKO,Denmark)in blocking buffer before exposure to chemiluminescence’s substrate CDP-Star.To verify equal protein loading,some of the membranes were stripped for 5min in0.2M NaOH,washed three times in PBS with 0.1%Tween-20(PBST),blocked and reprobed with anti-CREB antibody.Densitometry quantification of the hybrid-ization signals was performed with a Lumi-Imager F1k using LumiAnalyst k Software(Boehringer Mannheim, Germany).

2.7.Statistics

Statistical significance of results was determined by Student’s t-test with p b0.05as a significant difference. 2.8.Densitometry

The densitometric analysis was performed with the BIO-RAD Model GS-700Imaging densitometer,using the program Multi-Analyst/PC(BIO-RAD Laboratories, USA).

3.Results

3.1.Specific inhibition of cAMP-regulated signalling path-ways indicates that PKA is the major pathway that mediates forskolin-induced CREB phosphorylation in NIH3T3cells The adenylate cyclase activator forskolin(FSK)is generally applied as a cAMP-elevating agent to induce cAMP-dependent protein kinase or PKA[23].However, cAMP has been demonstrated to activate also the MAP kinase p38,MEK1/2-ERK1/2and MEK5-ERK5modules,and the PI3K-PKB/Akt1pathway[32–35].Moreover, cAMP-induced release of Ca2+from cyclic-nucleotide-gated ion channels may stimulate CaMK and conventional PKC[36,37].We first set out to determine which signalling pathways can mediate forskolin-induced phos-phorylation of CREB at serine residue133in NIH3T3 cells.For this purpose,we monitored the levels of phosphoserine-133-CREB in extracts of FSK-stimulated NIH3T3cells in the absence or presence of one of the following protein kinase inhibitors H89(PKA),LY294002 (Akt/PKB),SB203580(p38a/h)or KN-62(CaMKII). Previously,we have showed that forskolin fails to activate the MEK1/2-ERK1/2pathway in NIH3T3cells and that neither the MEK inhibitor,PD98059,nor the PKC inhibitor,GF109203X,were able to influence FSK-provoked CREB phosphorylation at Ser-133in these cells [38,39].H89completely abrogated FSK-induced CREB phosphorylation,while the other inhibitors did not signifi-cantly influence the phosphorylation status of CREB(Fig. 1A).Similar results were obtained in an independent experiment.Densitometric analysis of the relative phos-phoCREB levels were done and the result is presented in a bar chart(Fig.1A).To ensure that these chemical inhibitors on their own did not influence CREB phosphorylation,cells were also treated with the inhibitors in the absence of FSK and phosphoserine-133CREB levels were monitored.None of the inhibitors affected the basal phosphoCREB levels in the absence of FSK(results not shown).Taken together, these results indicate that PKA mediates forskolin-induced CREB phosphorylation in NIH3T3cells.

Several studies have shown that distinct stimuli may provoke CREB phosphorylation with the same stoichiometry and kinetics,yet their ability to stimulate CREB-mediated transcription may differ considerably.However,substitution of serine-133into non-phosphorable alanine completely abolishes signal-induced activation of CREB.Hence,phos-phorylation of serine-133is necessary,but not sufficient for stimulus-induced activation of CREB-dependent transcrip-tion(reviewed in Ref.[40]).This prompted us to test whether specific inhibitors of signalling pathways known to convert to CREB affected GAL4–CREB dependent transcription in forskolin treated cells.H89,the only inhibitor that prevented forskolin-induced CREB phosphorylation,also completely abrogated CREB activation by FSK(Fig.1B),but had no influence on GAL4-mediated transcription(results not shown).As expected,the specific inhibitors for PKB,p38 or CaMKII signalling pathways did not affect forskolin-induced CREB-dependent transcription(Fig.1B and results not shown).

Until recently,PKA was thought to be the main target of cAMP in eukaryotic cells.However,Epac(exchange factor directly activated by cAMP),a widely expressed exchange factor for the small GTPases Rap1and Rap2,has been shown to be a receptor for cAMP as well[5,6].Because specific Epac inhibitors are lacking so far,we investigated the putative role of Epac in mediating forskolin-induced

M.P.Delghandi et al./Cellular Signalling17(2005)1343–13511345

CREB phosphorylation by monitoring whether a specific Epac-activator,8-pMeOPT-2V -O -Me-cAMP,could provoke CREB phosphorylation in NIH 3T3cells [41].The Epac activator failed to induce CREB phosphorylation and to stimulate GAL4–CREB-mediated transcription (Fig.2A

and B),arguing against a role for Epac in mediating phosphorylation and transactivation of CREB in forskolin-treated NIH 3T3cells.

3.2.Activated MAPK p38stimulates CREB-mediated tran-scription and potentiates forskolin-induced CREB activity The activity of several protein kinases involved in cAMP-induced signalling transduction is regulated by phosphorylation [42].Therefore,we compared the phos-phorylation profile of different protein kinases in non-stimulated and forskolin-treated cells by immunoblot studies with phosphospecific antibodies.We found that neither MAP kinases ERK1/2nor ERK5were activated by forskolin treatment of NIH 3T3cells (Fig.3A).Forskolin also failed to provoke phosphorylation of the upstream regulators of ERK1/2,Raf-1and MEK1/2,and to induce phosphorylation of the ERK1/2substrate Elk-1(results not shown and Ref.[39]).These findings confirm that the Raf-MEK1/2-ERK1/2pathway is not involved in FSK-induced CREB phosphorylation in NIH 3T3cells.Neither Akt1/PKB,PKC,nor CaMK became phosphorylated after forskolin stimulation (results not shown).The MAPK p38,however,was clearly phosphorylated upon forskolin stimuli,and this FSK-induced phosphorylation of p38MAP kinase is mediated by PKA,since the presence of H89clearly diminished the activation of p38(Fig.3

B).

Fig.1.Forskolin-induced phosphorylation and activation of CREB are mediated by PKA in NIH 3T3cells.(A)Serum-starved cells were incubated with the different inhibitors;H89(10A M),LY294002(50A M),SB203580(10A M),and KN-62(10A M)for 1h,and then treated with 10A M forskolin (FSK)for 30min.Each experiment was performed in duplicate,except for the control cells where only one extract was analysed.PhosphoCREB levels were determined by Western blot analysis using phospho-specific anti-bodies that recognize phosphoser-133CREB (top panel).The arrowhead indicates the position of phosphoCREB,while the lower band is phosphoATF-1.To verify equal loading and blotting of the samples,the membrane was stripped and re-incubated with anti-CREB antibodies (middle panel).The ration of phosphoCREB:CREB in each was determined by densitometry and the average of the duplicate samples was calculated.The ratio in untreated control cells was arbitrary set as 1and the values obtained for the treated cells were expressed as fold induction.The asterisk (*)indicates a significant difference (p b 0.05).(B)Cells were transfected with G5-E1b-LUC reporter plasmid (1A g)and expression plasmid for chimeric GAL4–CREB (0.5A g).The chimeric protein contains the DNA binding domain of GAL4(residues 1–147)fused to full-length CREB.Serum-starved cells were either untreated (unstimulated)or pretreated with the PKA inhibitor H89(10A M),the CaMKII inhibitor KN-62(10A M)or the p38(a /h )inhibitor SB203580(10A M)for 1h,and then treated with FSK (10A M).The luciferase activity in cell extracts was determined 3h after FSK-stimulation.The luciferase activity in untreated control cells was arbitrary set as 1and the luciferase activities in treated cells are represented as fold induction.The results are the average of three independent parallels F S.D.The experiment was performed in triplicate and the results of a single representative experiment are

shown.

Fig.2.Epac is not an activator of CREB in NIH 3T3cells.(A)The specific Epac-activator,8-pMeOPT-2V -O -Me-cAMP failed to induce phosphoryla-tion and activation of CREB.Serum-starved cells were either untreated or stimulated with forskolin (FSK)(10A M),or the specific Epac-activator (0.1or 0.3mM)for 30min.PhosphoCREB levels was determined by Western blot analysis using phosphoserine-133CREB antibodies.To verify equal loading and blotting of the samples,the membrane was stripped and re-incubated with anti-CREB antibodies (lower panel).(B)The cells were transfected as described in the legend of Fig.1,and untreated or stimulated with FSK (10A M,3h),or the Epac-activator (0.1or 0.3mM,1h).Luciferase activities were determined as described in the legend of Fig.1.The results are the average of three independent parallels F S.D.

M.P .Delghandi et al./Cellular Signalling 17(2005)1343–1351

1346

These results suggest that p38MAPK may mediate forskolin-induced activation of CREB.We therefore tested whether ectopic expression of p38could stimulate GAL4–CREB-mediated transcription.Thereto,NIH 3T3cells were co-transfected with p38and with a constitutive active form of its upstream activator,MKK6E/E.Co-expression of MKK6E/E plus p38strongly enhanced CREB-dependent transcription,while ectopically expressed p38alone failed to stimulate CREB-driven transcription (Fig.4A and B).Moreover,p38potentiated forskolin-,as well as PKA-induced GAL4–CREB driven transcription,(Fig.4B and C,respectively).This suggests that additional stimulation of PKA-induced CREB transcriptional activity by p38requires a PKA-dependent activation of p38.

3.3.MSK1mediates increased CREB-dependent transcrip-tion by activated MAPK p38

The mitogen-activated protein kinase-activated protein kinase MSK1has been shown to be a genuine CREB kinase and can be activated by the MAPK p38[11].Hence,we wanted to investigate whether MSK1was involved in mediating p38-induced CREB activation.Overexpression of dominant negative mutants of MSK1(C-terminal kinase dead MSK1(dnMSK1-CT)or/and N-terminal kinase dead MSK1(dnMSK1-NT))strongly reduced MAPK p38-induced stimulation of CREB-directed transcription (Fig.5A).Concentrations of H89that inhibit PKA,have also been reported to inhibit the kinase activity of MSK1[43].To confirm a role for MSK1in p38-induced activation of CREB,we assayed the effect of H89on the transcriptional

potency of GAL4–CREB in the presence of activated MKK6plus p38.As shown in Fig.5B,addition of H89significantly (p b 0.01)reduced the p38-induced transcrip-tional activity of CREB.MAPKAPK2and RSK-B

(also

Fig.4.Activated p38strongly enhanced CREB-dependent transcription and enhances forskolin-and PKA-induced CREB-driven transcription.(A)Cells were co-transfected with the reporter plasmid G5-E1b-LUC (1A g),the empty GAL4expression vector (0.5A g)or GAL4–CREB expression plasmid (0.5A g)and either 2A g empty control vector or 1A g p38-and 1A g MKK6E/E expression plasmid.(B)Serum-starved cells were co-transfected with the reporter plasmid G5-E1b-LUC (1A g),the empty GAL4expression vector (0.5A g)or GAL4–CREB expression plasmid (0.5A g)and either 1A g empty control vector or 1A g p38-expression plasmid.Where indicated,cells were stimulated with forskolin (FSK)(10A M)for 3h.(C)The cells were co-transfected with the reporter plasmid G5-E1b-LUC (1A g),the GAL4–CREB expression plasmid (0.5A g),the constitutive active subunit of PKA (C a )(1A g),and empty control plasmids (1A g)or p38-expression plasmid (1A g).Luciferase activities were determined as described in the legend of Fig.1.Statistical significance was calculated by Student’s t -test,*p b 0.05.The results are the average of three independent parallels F S.D.Similar results were obtained in corresponding

experiments.

Fig.3.Forskolin provoked phosphorylation of p38MAPK in NIH 3T3cells depends on PKA.(A)Serum-starved cells were incubated with either forskolin (FSK)(10A M)or newborn calf serum (NCS)(10%)for 30min.Each experiment was performed in duplicate.PhosphoERK1/2(top panel)and phosphoERK5(lower panel)levels were determined by Western blot analysis using phospho-specific antibodies.(B)Pre-treatment of serum-starved cells with H89(10A M)prevents FSK-induced https://www.360docs.net/doc/149120346.html,ne 1:control cells;lane 2:cells treated with FSK (10A M)for 30min;lane 3:cell pretreated for 1h with H89(10A M)and then stimulated with FSK for 30min;lane 4:cells exposed to NCS (10%)for 30min.To verify equal loading and blotting,the membrane was stripped and re-incubated with anti-CREB antibodies (lower panel).

M.P .Delghandi et al./Cellular Signalling 17(2005)1343–13511347

referred to as MSK-2),two other p38-regulated protein kinases,have also been shown to phosphorylate CREB at serine-133[13,44].However,overexpression of dominant negative mutants of these kinases did not interfere with forskolin-induced activation of GAL4–CREB-mediated transcription(results not shown).In conclusion,our results show that CREB phosphorylation and activation in for-skolin-treated NIH3T3cells is mediated by PKA and by PKA-dependent activation of the p38-MSK1pathway.

3.4.Forskolin-induced activation of p38MAPK is delayed compared to PKA activation

Activation of the cAMP/PKA-signalling pathway follows the characteristic b burst–attenuation–refractory Q phases. During the burst phase,PKA becomes activated and the kinetics of PKA activation and CREB phosphorylation are remarkably similar[45].However,our time-dependent studies of forskolin-treated NIH3T3cells demonstrated that deterioration of PKA activity during the attenuation phase occurred much faster than CREB phosphorylation.Indeed, while forskolin-induced PKA activity rapidly decreased after30min[39],phosphoCREB levels remained elevated for at least2h[31].These observations made us speculate that the activated p38-MSK1pathway could be involved in enduring the phosphorylation profile of CREB during attenuation of PKA.Therefore,we examined the kinetics of p38phosphorylation in forskolin exposed NIH3T3cells. Increased phospho-p38levels were detected30min after FSK-stimulation and remained elevated for up to120min (Fig.6).These findings suggest that the activated p38-MSK1pathway may sustain CREB phosphorylation even when PKA enters the refractory phase.

4.Discussion

The adenylate cyclase activator forskolin is commonly used to increase intracellular cAMP levels and is generally applied as a specific activator of the PKA pathway[23]. However,other pathways such as the PI3K Y PKB,the Epac Y B-Raf Y MEK1/2Y ERK1/2,the MEK5Y ERK5,the c-Raf Y MEK1/2Y ERK1/2,and the p38pathways can also be induced by cAMP in a cell-specific fashion[5–21,32–35].Moreover,cAMP-induced release of Ca2+from cyclic-nucleotide-gated ion channels may stimulate CaMK and conventional PKC[36,37].As some of these pathways can converge to the transcription factor CREB,we wanted to identify which of them are involved in forskolin-induced CREB phosphorylation and activation in NIH3T3cells. Our results showed that of the known cAMP-regulated pathways,PKA and p38MAP kinase were the only kinases that became activated upon forskolin treatment.No increase in phosphorylation levels of PKB,CaMKII,ERK5,or ERK1/2were observed after forskolin treatment,and in agreement with the group of Enserink et al.Epac was not found to be activated upon forskolin stimulation of NIH3T3 cells[41].The failure of forskolin to induced phosphor-ylation of ERK1/2in NIH3T3cells is in accordance with the observations of several other groups[46–48].In contrast,two other groups found that ERK1/2became activated in forskolin stimulated NIH3T3cells[32,49].

The Fig.6.Kinetics of forskolin-induced p38phosphorylation in NIH3T3cells. Serum-starved cells were either untreated(lane1)or treated with forskolin (10A M)for5,15,30,60,120,180,240,or300min(lanes2–8, respectively).The level of phospho-p38was determined by using the phospho-specific antibody.The membrane was stripped and re-incubated with anti-CREB antibodies to assure equal loading and blotting of the individual samples(lower

panel). Fig.5.MSK1can mediate p38induced GAL4–CREB mediated tran-

scription in NIH3T3cells.(A)The cells were co-transfected with the

reporter plasmid G5-E1b-LUC(1A g),the GAL4–CREB expression

plasmid(0.5A g)and either empty control vector(1A g),p38-,MKK6E/E-

expression plasmids(1A g)or dominant negative MSK1mutants(1A g)as

indicated.(B)The cells were co-transfected with the reporter plasmid G5-

E1b-LUC(1A g),the empty GAL4expression vector(0.5A g)or the

GAL4–CREB expression plasmid(0.5A g),and either empty control

vector(1A g)or MKK6E/E expression plasmid(1A g).Serum-starved

cells were untreated or treated with the chemical inhibitor H89(10A M)

for3h.Luciferase activities were determined as described in the legend

of Fig.1.Statistical significance was calculated by Student’s t-test,

*p b0.01compared with untreated cells.The results are the average of

three independent parallels F S.D.

M.P.Delghandi et al./Cellular Signalling17(2005)1343–1351 1348

authors explain the discrepancies between the different groups by subtle variations in cell lines used in different laboratories.Forskolin has been shown to induce ERK5 activation in both NIH3T3cells and C2C12myoblasts[32], but we were not able to reproduce these results,again maybe due to subtle variations in cell lines.cAMP has also been implicated in modulating PKB activity and forskolin was shown to provoke phosphorylation of PKB in hepatocytes,EBNA-transformed HEK293cells,and immor-talized Schwann cells[34,50–52].We did not,however, observe any forskolin-induced PKB phosphorylation in our NIH3T3cells,which is in accordance with the findings of the group of Fang et al.[53].The group of Makhinson reported that forskolin-induced phosphorylation of CaMKII in the CA1region of the hippocampus.Increased phospho-CaMKII levels resulted from PKA-dependent inactivation of protein phosphatase1[54].We did not observe any forskolin-induced CaMKII phosphorylation in our NIH3T3 cells,and previously we have failed to find a role for PKC in forskolin-induced CREB phosphorylation and activation [38].

Previous studies have shown that forskolin can induce phosphorylation of the p38MAP kinase in a PKA-depend-ent mode in a variety of cells,including macrophages,SK-N-MC,PC12,MC3T3-E1,Chinese hamster ovary cells, colon cancer cells,adipocytes and adult mouse cardiomyo-cytes[33,55–62].However,forskolin has also been shown to inhibit p38MAP kinase phosphorylation in HUVEC cells and in thyrocytes[63,64].We demonstrated that p38was phosphorylated in a time-dependent manner in NIH3T3cells exposed to forskolin.Forskolin-induced p38phosphor-ylation was PKA-dependent since the chemical inhibitor of PKA,H89,prevented this phosphorylation.In agreement with our findings,Schulte and colleagues demonstrated that cAMP-induced phosphorylation of p38in Chinese hamster ovary cells(CHO)required PKA[56].However,Nishihara et al.reported PKA-independent activation of p38in colon cancer cells.They found that cAMP-induced phosphoryla-tion of CREB was mediated by activation of p90RSK and MSK1,which followed the PKA-independent ERK1/2and p38activation[57].Our study shows that ectopic expression of p38alone was not sufficient to stimulate GAL4–CREB-driven transcription in serum-starved cells.However, activation of endogenous or ectopic expressed p38by a constitutive active MKK mutant(MKK6E/E)stimulated CREB-dependent transcription,while p38potentiated both forskolin-and PKA-induced CREB activity.Of the known CREB kinases that are activated by p38,MAPKAPK2, RSK-B,and MSK-1,we found that only MSK-1was involved in mediating p38-induced CREB activation.H89, a specific inhibitor of both PKA and MSK1,completely abrogated forskolin-induced CREB phosphorylation and activation.On the other hand,specific inhibition of p38by SB203580,which will also affect its downstream targets such as MSK1and hence CREB phosphorylation,had no measurable effect on forskolin-induced CREB phosphory-lation and CREB-directed transcription.These results indicate that activation of p38,either through MKK6or through a PKA-dependent mechanism,can stimulate CREB-mediated transcription,but that this pathway is

not Fig.7.A schematic overview of the described cAMP-signalling pathways in mammalian cells.The continuous lines indicate the signalling pathways operational in NIH3T3cells as identified in this study.The dashed lines indicate signalling pathways functional in other mammalian cell lines.The predominance of the PKA pathway in NIH3T3cells is shown by the bold arrows.

M.P.Delghandi et al./Cellular Signalling17(2005)1343–13511349

the major pathway applied in forskolin-treated NIH3T3 cells.Activation of the p38-MSK1pathway occurs after the burst phase of PKA activation and may represent a relay mechanism where PKA passes the signal onto a new signalling pathway that converge to a common substrate (i.e.CREB).

The mechanism by which PKA activates p38in NIH3T3 cells remains elusive.Studies with T lymphocytes have shown that PKA phosphorylates the haematopoietic protein tyrosine phosphatase(HePTP)at serine23.HePTP is a negative regulator of the MAP kinase p38,and it binds to p38through a kinase-interaction motif(KIM)[65].A similar mechanism seems to be operational in COS-7cells where the protein tyrosine phosphatase PTP-SL can be phosphorylated by PKA at serine residue231,resulting in impaired binding of the protein phosphatase to the MAP kinase p38.The authors suggest a mechanism by which PKA can regulate the activity of p38and its translocation to the nucleus.Such a mechanism would involve the existence, in certain cell types,of a pool of inactive MAP kinases outside of the nucleus,which would be complexed with PTP-SL or the other KIM-containing PTPs.The dissociation equilibrium of the stimulus-specific conditions of PKA activity,and the lack of association would be favoured by the PKA-mediated phosphorylation of the KIM regulatory residue on the PTP.Thus upon conditions of PKA activation,both the tyrosine phosphorylation and the entry into the nucleus of the MAP kinase would be prevalent[66]. It is tempting to speculate that a similar mechanism is operational in NIH3T3cells,and obviously the identifica-tion of a putative p38-interacting tyrosine phosphatase in NIH3T3cells would support this hypothesis.

5.Conclusions

The second messenger cAMP can activate multiple signalling pathways that can converge to CREB.The aim of this study was to identify the cAMP signalling pathways involved in forskolin-induced CREB activation in NIH3T3 cells.Our findings,which are summarized in Fig.7, demonstrate that PKA is the major CREB kinase exploited by NIH3T3cells after forskolin treatment.Forskolin also induces phosphorylation of p38in a PKA-dependent way, and the forskolin-induced p38phosphorylation kinetics is delayed compared to PKA activation.Maximum phos-phorylation levels of p38occurred at time points when activated PKA levels are gradually ebbing.Moreover,we showed that activated p38stimulates CREB-mediated tran-scription through MSK1.

Our results suggest that the lagging activation of p38 may ensure the cell that the stimulus converging to CREB is transmitted by another pathway even when the original activated transduction pathway enters the refractory state. The transcriptional activity of CREB seems to depend on the duration of CREB phosphorylation[40].Cell-specific cross talk between distinct pathways may therefore form a mechanism to prolong the phosphorylation of CREB and to ensure a cellular response by activated CREB. Acknowledgements

The authors thank Dr.R.A.Maurer,Dr.R.J.Davis,Dr. M.Deak,Dr.T.Jahnsen,Dr.K.Taske′n,Dr.A.Brunet,Dr.

D.Zechner,and Dr.

E.Sontag for kindly providing plasmids used in this study.

This work was supported by grants from the Norwegian Cancer Society(Kreftforeningen,project A01037/004),the Norwegian Research Council(NFR projects S5168and S5228),and the Olav and Aakre foundation(A5048). References

[1]A.J.Shaywitz,M.E.Greenberg,Annu.Rev.Biochem.68(1999)821.

[2]B.Mayr,M.Montminy,Nat.Rev.,Mol.Cell Biol.2(2001)599.

[3]D.De Cesare,P.Sassone-Corsi,Prog.Nucleic Acid Res.Mol.Biol.

64(2000)343.

[4]P.G.Quinn,Prog.Nucleic Acid Res.Mol.Biol.72(2002)269.

[5]H.Kawasaki,G.M.Springett,N.Mochizuki,S.Toki,M.Nakaya,M.

Matsuda,D.E.Housman,A.M.Graybiel,Science282(1998)2275.

[6]J.de Rooij,F.J.Zwartkruis,M.H.Verheijen,R.H.Cool,S.M.Nijman,

A.Wittinghofer,J.L.Bos,Nature396(1998)474.

[7]S.S.Grewal,D.M.Fass,H.Yao,C.L.Ellig,R.H.Goodman,P.J.

Stork,J.Biol.Chem.275(2000)34433.

[8]T.Fujita,T.Meguro,R.Fukuyama,H.Nakamuta,M.Koida,J.Biol.

Chem.277(2002)22191.

[9]L.Iacovelli,L.Capobianco,L.Salvatore,M.Sallese,G.M.

D V Ancona,B.A.De,Mol.Pharmacol.60(2001)924.

[10]H.Takahashi,M.Honma,Y.Miyauchi,S.Nakamura,A.Ishida-

Yamamoto,H.Iizuka,Arch.Dermatol.Res.296(2004)74.

[11]J.S.Arthur,P.Cohen,FEBS Lett.482(2000)44.

[12]J.Xing,D.D.Ginty,M.E.Greenberg,Science273(1996)959.

[13]G.R.Wiggin,A.Soloaga,J.M.Foster,V.Murray-Tait,P.Cohen,J.S.

Arthur,Mol.Cell.Biol.22(2002)2871.

[14]Y.Tan,J.Rouse, A.Zhang,S.Cariati,P.Cohen,https://www.360docs.net/doc/149120346.html,b,

EMBO J.15(1996)4629.

[15]S.Pugazhenthi,T.Boras, D.O V Connor,M.K.Meintzer,K.A.

Heidenreich,J.E.Reusch,J.Biol.Chem.274(1999)2829.

[16]E.T.Maizels,A.Mukherjee,G.Sithanandam,C.A.Peters,J.Cottom,

K.E.Mayo,M.Hunzicker-Dunn,Mol.Endocrinol.15(2001)716.

[17]F.L.Watson,H.M.Heerssen,A.Bhattacharyya,L.Klesse,M.Z.Lin,

R.A.Segal,Nat.Neurosci.4(2001)981.

[18]K.Du,M.Montminy,J.Biol.Chem.273(1998)32377.

[19]L.A.Cass,S.A.Summers,G.V.Prendergast,J.M.Backer,M.J.

Birnbaum,J.L.Meinkoth,Mol.Cell.Biol.19(1999)5882.

[20]N.Iida,K.Namikawa,H.Kiyama,H.Ueno,S.Nakamura,S.Hattori,

J.Neurosci.21(2001)6459.

[21]I.J.Gonzalez-Robayna,A.E.Falender,S.Ochsner,G.L.Firestone,

J.S.Richards,Mol.Endocrinol.14(2000)1283.

[22]U.B.Kaupp,R.Seifert,Physiol.Rev.82(2002)769.

[23]P.A.Insel,R.S.Ostrom,Cell.Mol.Neurobiol.23(2003)305.

[24]P.Sun,H.Enslen,P.S.Myung,R.A.Maurer,Genes Dev.8(1994)

2527.

[25]A.Seth,E.Alvarez,S.Gupta,R.J.Davis,J.Biol.Chem.266(1991)

23521.

[26]M.Deak,A.D.Clifton,L.M.Lucocq,D.R.Alessi,EMBO J.17

(1998)4426.

M.P.Delghandi et al./Cellular Signalling17(2005)1343–1351 1350

[27]K.B.Foss,https://www.360docs.net/doc/149120346.html,ndmark,B.S.Skalhegg,K.Tasken,E.Jellum,V.

Hansson,T.Jahnsen,Eur.J.Biochem.220(1994)217.

[28]A.Brunet,J.Pouyssegur,Science272(1996)1652.

[29]D.Zechner,R.Craig, D.S.Hanford,P.M.McDonough,R.A.

Sabbadini,C.C.Glembotski,J.Biol.Chem.273(1998)8232. [30]E.Sontag,S.Fedorov,C.Kamibayashi,D.Robbins,M.Cobb,M.

Mumby,Cell75(1993)887.

[31]M.Johannessen,M.P.Delghandi,O.M.Seternes,B.Johansen,U.

Moens,Cell.Signal.16(2004)1187.

[32]G.W.Pearson,M.H.Cobb,J.Biol.Chem.277(2002)48094.

[33]T.V.Hansen,J.F.Rehfeld,F.C.Nielsen,S.Nagano,M.Takeda,L.Ma,

B.Soliven,S.Tiwari,K.Felekkis,E.Y.Moon,A.Flies,D.H.Sherr,

A.Lerner,G.Schulte,

B.B.Fredholm,Mol.Endocrinol.13(1999)

466.

[34]S.Nagano,M.Takeda,L.Ma,B.Soliven,S.Tiwari,K.Felekkis,

E.Y.Moon, A.Flies, D.H.Sherr, A.Lerner,G.Schulte, B.B.

Fredholm,J.Neurochem.77(2001)1486.

[35]M.Frodin,P.Peraldi,E.Van Obberghen,J.Biol.Chem.269(1994)

6207.

[36]M.Sheng,G.McFadden,M.E.Greenberg,Neuron4(1990)571.

[37]M.Antoine,C.Gaiddon,J.P.Loeffler,Mol.Cell.Endocrinol.120

(1996)1.

[38]O.M.Seternes,B.Johansen,U.Moens,Mol.Endocrinol.13(1999)

1071.

[39]O.M.Seternes,R.Sorensen, B.Johansen,T.Loennechen,J.

Aarbakke,U.Moens,Biochim.Biophys.Acta1395(1998)345. [40]M.Johannessen,M.P.Delghandi,U.Moens,Cell.Signal.16(2004)

1211.

[41]J.M.Enserink, A.E.Christensen,J.de Rooij,M.van Triest, F.

Schwede,H.G.Genieser,S.O.Doskeland,J.L.Blank,J.L.Bos,Nat.

Cell Biol.4(2002)901.

[42]B.Nolen,S.Taylor,G.Ghosh,Mol.Cell15(2004)661.

[43]S.Thomson,A.L.Clayton,C.A.Hazzalin,S.Rose,M.J.Barratt,L.C.

Mahadevan,EMBO J.18(1999)4779.

[44]M.Iordanov,K.Bender,T.Ade,W.Schmid,C.Sachsenmaier,K.

Engel,M.Gaestel,H.J.Rahmsdorf,P.Herrlich,EMBO J.16(1997) 1009.

[45]M.Montminy,Annu.Rev.Biochem.66(1997)807.

[46]J.M.Schmitt,P.J.Stork,Mol.Cell.Biol.21(2001)3671.[47]B.M.Burgering,G.J.Pronk,P.C.van Weeren,P.Chardin,J.L.Bos,

EMBO J.12(1993)4211.

[48]V.Calleja, E.P.Ruiz, C.Filloux,P.Peraldi,V.Baron, E.Van

Obberghen,Endocrinology138(1997)1111.

[49]M.Klinger,O.Kudlacek,M.G.Seidel,M.Freissmuth,V.Sexl,

J.Biol.Chem.277(2002)32490.

[50]N.Filippa,C.L.Sable,C.Filloux,B.Hemmings,E.Van Obberghen,

Mol.Cell.Biol.19(1999)4989.

[51]C.L.Sable,N.Filippa,B.Hemmings,E.Van Obberghen,FEBS Lett.

409(1997)253.

[52]C.R.Webster,M.S.Anwer,Am.J.Physiol.277(1999)G1165.

[53]X.Fang,S.X.Yu,Y.Lu,R.C.Bast Jr.,J.R.Woodgett,https://www.360docs.net/doc/149120346.html,ls,

Proc.Natl.Acad.Sci.U.S.A.97(2000)11960.

[54]M.Makhinson,J.K.Chotiner,J.B.Watson,T.J.O V Dell,J.Neurosci.

19(1999)2500.

[55]M.Zheng,S.J.Zhang,W.Z.Zhu,B.Ziman,B.K.Kobilka,R.P.Xiao,

J.Biol.Chem.275(2000)40635.

[56]G.Schulte,B.B.Fredholm,Exp.Cell Res.290(2003)168.

[57]H.Nishihara,M.Hwang,S.Kizaka-Kondoh,L.Eckmann,P.A.Insel,

J.Biol.Chem.279(2004)26176.

[58]X.Mao,I.G.Bravo,H.Cheng,A.Alonso,Exp.Cell Res.292(2004)

304.

[59]T.O.Hansen,J.F.Rehfeld,F.C.Nielsen,J.Neurochem.75(2000)

1870.

[60]A.Kakita, A.Suzuki,Y.Ono,Y.Miura,M.Itoh,Y.Oiso,

Prostaglandins Leukot.Essent.Fat.Acids70(2004)469.

[61]W.Cao,A.V.Medvedev,K.W.Daniel,S.Collins,J.Biol.Chem.276

(2001)27077.

[62]C.C.Chio,Y.H.Chang,Y.W.Hsu,K.H.Chi,W.W.Lin,Cell.Signal.

16(2004)565.

[63]A.Rahman,K.N.Anwar,M.Minhajuddin,K.M.Bijli,K.Javaid,A.L.

True,A.B.Malik,Am.J.Physiol.,Lung Cell.Mol.Physiol.287 (2004)L1017.

[64]F.Vandeput,S.Perpete,K.Coulonval, https://www.360docs.net/doc/149120346.html,my,J.E.Dumont,

Endocrinology144(2003)1341.

[65]M.Saxena,S.Williams,K.Tasken,T.Mustelin,Nat.Cell Biol.1

(1999)305.

[66]C.Blanco-Aparicio,J.Torres,R.Pulido,J.Cell.Biol.147(1999)

1129.

M.P.Delghandi et al./Cellular Signalling17(2005)1343–13511351