HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 21, 18810-18816, May 24, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/21/18810    most recent M112107200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Constantinescu, A. Articles by Diamond, I. Search for Related Content PubMed PubMed Citation Articles by Constantinescu, A. Articles by Diamond, I. Social Bookmarking                      What's this?

cAMP-dependent Protein Kinase Types I and II Differentially Regulate cAMP Response Element-mediated Gene Expression

IMPLICATIONS FOR NEURONAL RESPONSES TO ETHANOL^*

Anastasia Constantinescu^§, Adrienne S. Gordon^¶, and Ivan Diamond^¶

From the  Ernest Gallo Clinic and Research Center, Departments of Neurology, ^¶ Cellular and Molecular Pharmacology, and  Neuroscience Graduate Program and Center for the Neurobiology of Addiction, University of California at San Francisco, Emeryville, California 94608

Received for publication, December 18, 2001, and in revised form, February 22, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We have shown that ethanol induces translocation of cAMP-dependent protein kinase (PKA) to the nucleus, cAMP response element-binding protein (CREB) phosphorylation, and cAMP response element-mediated gene transcription in NG108-15 cells. However, little is known about which PKA types regulate this process. We show here that under basal conditions NG108-15 cells contain type I PKA (CRI) primarily in cytosol and type II PKA (CRII) in the particulate and nuclear fractions. Antagonists of both type I and type II PKA inhibit forskolin- and ethanol-induced cAMP response element-mediated gene transcription. However, only the type II PKA antagonist inhibits forskolin-induced C and ethanol-induced C and RII translocation to the nucleus and CREB phosphorylation; the type I antagonist is without effect. Our data suggest that forskolin- and ethanol-induced CREB phosphorylation and gene activation are differentially mediated by the two types of PKA. We propose that type II PKA is translocated and activated in the nucleus and induces CREB phosphorylation that is necessary but not sufficient for gene transcription. By contrast, type I PKA is activated in the cytoplasm, turning on a downstream pathway that activates other transcription cofactors that interact with phosphorylated CREB to induce gene transcription.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

cAMP signaling is a multiple component system that requires production of cAMP and induces activation of PKA,¹ phosphorylation of CREB, and cAMP-dependent gene expression. This pathway has been implicated in learning and memory (1, 2) as well as addictive behaviors (3) and responses to ethanol (4). Ethanol activates the cAMP pathway to its full extent by stimulating adenylyl cyclase activity (4), C translocation to the nucleus (5), CREB phosphorylation (6), and CRE-mediated gene transcription (7).

The primary intracellular receptors for cAMP in mammalian cells are various isoforms of PKA. In the absence of cAMP, PKA is a tetrameric holoenzyme consisting of two catalytic subunits (C) bound to a regulatory subunit (R) dimer. After binding of two molecules of cAMP to each R monomer, the two C subunits are released and activated to phosphorylate intracellular substrates. The PKA family consists of four regulatory subunits (RI, RI, RII, RII) and three catalytic subunits (C, C, and C), each encoded by a unique gene (8, 9). Type I or type II PKA are classically defined by their specific regulatory subunits. The mechanisms by which the cAMP-signaling pathway achieves specificity include 1) compartmentalization of PKA via binding to scaffolding proteins such as A kinase-anchoring protein (AKAP) near target substrates; 2) regulated expression of distinct R and C subunit isoforms in cells and tissues; and 3) differential combinations of R and C subunit isoforms. All these mechanisms can affect PKA interaction with cAMP, substrates, and inhibitors. Studies with knockout and transgenic mouse models have shed some light on the physiological functions of specific isoforms of PKA in vivo (8). Models used in ethanol studies provided the first evidence for a differential role of PKA types in ethanol consumption and sedation. For example, RII/ mice show increased ethanol consumption and reduced sedation, whereas RI / or C / mice show normal voluntary consumption of ethanol (10). In contrast, transgenic mice that express a form of RI mutated at both cAMP binding sites have decreased PKA activity in forebrain and hippocampus and show decreased ethanol consumption (11).

We have shown previously that ethanol has differential effects on PKA types in NG108-15 cells. Ethanol induces translocation of the C and RII subunits of PKA from the Golgi area to the nucleus in NG108-15 cells (5, 6). Ethanol-induced translocation of C to the nucleus is accompanied by a persistent phosphorylation of the nuclear transcription factor CREB (6). This increase in CREB phosphorylation leads to a striking increase in CRE-mediated gene expression in cells transfected with the CRE-luciferase reporter gene (7). C and RI were not translocated after ethanol exposure (5, 6). In the present study we analyze the specific localization and association of PKA types and their differential role in forskolin- and ethanol-induced CREB phosphorylation and gene activation. We demonstrate that type II PKA is activated in the nucleus and induces CREB phosphorylation that is necessary but not sufficient for gene transcription. By contrast, type I PKA is activated in the cytoplasm, turning on a downstream pathway that induces activation of other transcription cofactors that interact with phosphorylated CREB to induce gene transcription.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- All chemicals were purchased from Sigma except where indicated.

Cell Culture-- NG108-15 neuroblastoma × glioma hybrid cells obtained from the cell culture facility at the University of California (San Francisco) were grown in 10% Serum Plus (JRH Biosciences) at 37 °C in a 10% CO₂ incubator as described previously (12). All cells were maintained for 3 days in complete defined medium (13). On day 3, the media were changed, and the cells were maintained in the absence or presence of 100 mM ethanol for various times. Treatment with 1 µM forskolin was carried out for 10 min. 300 µM Rp-Cl-cAMPS (RpI) or Rp-CPT-cAMPS (RpII) (BIOLOG) were added 2 h before ethanol or forskolin exposure and remained throughout the experiments.

Immunocytochemistry-- Cells were fixed with 4% paraformaldehyde, blocked in 4% normal goat serum (Jackson Immuno Research Laboratories, Inc.), and incubated overnight with primary antibodies. Monoclonal antibodies for C and RII (BD Transduction Laboratories) were diluted 1:100. Rabbit anti-mouse IgG2b (Zymed Laboratories Inc.) and fluorescein isothiocyanate-conjugated goat anti-mouse IgG1 (Roche Molecular Biochemicals) were diluted 1:250. Polyclonal antibody against CREB phosphorylated at Ser-133 was purchased from Cell Signaling Technology, Inc. and diluted 1:200. Secondary antibodies (Texas Red- or fluorescein isothiocyanate-conjugated goat anti-rabbit) were purchased from Jackson ImmunoResearch Laboratories, Inc. and diluted 1:250. No fluorescence was detected when cells were incubated with fluorophor-conjugated secondary antibody only.

Cell Fractionation-- Cytosolic, nuclear, and particulate fractions were isolated from NG108-15 cells in hypertonic sucrose as follows. Cells were homogenized in buffer A (10 mM Tris-HCl pH 7.4, 2 M sucrose, 1% Triton X-100, 3 mM MgCl₂, 2 mM dithiothreitol, 0.1 mM EDTA, proteases inhibitor cocktail (Roche Molecular Biochemicals), phosphatases inhibitor cocktail (Sigma)) and layered on a cushion of buffer A minus Triton X-100. After centrifugation for 1 h at 28,500 rpm in a swinging bucket rotor, the top layer containing cytosol and particulate fractions was diluted 3 times with buffer B (buffer A minus sucrose) and centrifuged for 1 h at 100,000 × g. The resulting supernatant was cytosol, and the pellet was the particulate fraction. The nuclear fraction (pellet from the first centrifugation) was resuspended in buffer B supplemented with 0.5 M NaCl and shaken at 4 °C for 60 min to lyse the nuclei. After centrifugation for 20 min at 67,000 × g the supernatant contained the nuclear extract; the pellet, consisting of mostly DNA, was discarded. Protein content in all fractions was determined by the Bradford Bio-Rad protein assay.

Immunoprecipitation-- Cell fractions or total lysate were resuspended in IP buffer (20 mM Tris, pH 7.4, 2 mM EDTA, 2 mM EGTA, pH 8, 300 mM NaCl, 0.4 mM sodium orthovanadate, 0.4 mM 4-(2-aminoethyl)benzenesulfonyl fluoride, 2% Triton X-100, 1% IGHEPAL CA-630) to 0.5 mg of protein and precleared for 30 min with protein G-conjugated agarose. Supernatant from a 10-min centrifugation at 10,000 rpm was incubated overnight with 2 µg of monoclonal RI or RII antibodies (BD Transduction Laboratories) and further incubated with protein G-conjugated agarose for 1 h. The immunoprecipitate was centrifuged and washed four times with phosphate buffered saline, pH 7.0. All incubations were at 4 °C.

Western Blots-- Cellular fractions and immunoprecipitates were subjected to SDS-PAGE, and proteins were transferred to polyvinylidene difluoride membranes as previously described (6). Blots were probed with antibodies against C (Santa Cruz Biotechnology Inc.), C (a generous gift from Dr. G. S. McKnight, University of Washington, Seattle, WA), RII, and AKAP-95 (Transduction Laboratories) diluted 1:1000. Secondary antibodies (Cell Signaling Technology, Inc.) were horseradish peroxidase-linked goat anti-rabbit (1:1000) for C and C and rabbit anti-mouse (1:1000) for RII. Blots were incubated with ECL Plus detection reagent (Amersham Biosciences), immunoreactive bands were visualized, and densities were calculated using the Storm 860 and ImageQuant software (Molecular Dynamics).

PKA Assay-- Nuclear extracts (0.6 mg/ml protein) were incubated in 50 mM phosphate buffer in the presence or absence of 5 µM cAMP. PKA holoenzyme types I and II from either rabbit muscle (Sigma) (0.05 mg/ml) or immunoprecipitated from total lysate of NG108-15 cells were preincubated with 50 µM Rp-Cl-cAMPS (RpI) or Rp-CTP-cAMPS (RpII) for 30 min before incubation in 50 mM phosphate buffer in the presence or absence of 5 µM cAMP. The reaction was carried out using the PKA-specific substrate Leu-Arg-Arg-Ala-Ser-Leu-Gly (Kemptide) (14) as previously described (6). Phosphorylation of other potential substrates due to all kinases in the cell lysate was determined in the absence of Kemptide and was subtracted from the values obtained in the presence of Kemptide. Phosphorylation of Kemptide due to kinases other than PKA was also determined in the presence of 20 µg/µl PKI from rabbit muscle and was not significantly different from backgound phosphorylation measured in the absence of Kemptide.

Luciferase Assay-- NG108-15 cells (5 × 10⁴) were transfected in defined media with 150 ng of the pFC-CRE-luciferase plasmid from Stratagene using Effectene (Qiagen) as described by the manufacturer. Media were changed 24 h after transfection. To activate CRE-mediated gene expression, the cells were incubated in either 1 µM forskolin for 4 h or 100 mM ethanol for 14 h, where ethanol-induced CRE-luciferase activity was previously found to be maximal (7). Cell extracts were prepared, and luciferase activity was measured as described (7). Each point represents three separate experiments with six replicate samples for each point.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We have previously shown that chronic ethanol exposure differentially affects the localization of PKA subunits expressed in NG108-15 cells. Chronic exposure to ethanol induces translocation of C to the nucleus of NG108-15 cells in a time- and dose-dependent manner (5). RII also translocates to the nucleus, but localization of C and RI is not altered (6). We hypothesized that this differential translocation might be due partly to a type-specific association and localization of PKA subunits in NG108-15 cells. Therefore, to address this question we first analyzed the distribution of regulatory PKA subunits in cytosol, particulate, and nuclear fractions.

Type-specific Association and Localization of PKA Subunits in NG108-15 Cells-- NG108-15 cells were grown, lysed, and fractionated as described under "Experimental Procedures." Fractions were then solubilized and incubated overnight with monoclonal antibodies against RI and RII. The monoclonal antibody for RII is specific for the  subunit of RII. The monoclonal antibody against RI recognizes both RI and RI regulatory subunits of PKA. However, we were not able to detect either RI or RII in NG108-15 cells by using specific monoclonal antibodies against these subunits (6); hence, the RI antibody should selectively immunoprecipitate RI in these cells. The immunoprecipitates were subjected to SDS-PAGE electrophoresis, blotted, probed with polyclonal antibodies against C, stripped, and reprobed with polyclonal antibodies against C. Results in Fig. 1 show that C associates with RI only in the cytosol and does not associate with RII in these cells. Furthermore, C associates with RII only in particulate and nuclear fractions and does not associate with RI. Before immunoprecipitation, C and RI subunits are also exclusively localized in the cytosol, whereas C and RII are found in particulate and nuclear fractions (data not shown). These results demonstrate that NG108-15 cells contain type I PKA (CRI) primarily in cytosol and type II PKA (CRII) in the particulate and nuclear fractions under basal conditions.

View larger version (50K): [in this window] [in a new window]   Fig. 1.   Type-specific association and localization of PKA subunits in NG108-15 cells. Cytosol (Cyt), particulate (Par), and nuclear extract (Nuc) fractions were prepared from NG108-15 cells as described under "Experimental Procedures." RI and RII were immunoprecipitated (IP) with specific monoclonal antibodies, and Western blots of the immunoprecipitates were probed with polyclonal antibodies against C and C. The blot shown is representative of three experiments.

Ethanol Activates PKA Type II and Stimulates Its Translocation to the Nucleus-- Using immunocytochemistry, we previously showed that C and RII are colocalized before and after nuclear translocation in response to ethanol exposure (15). To determine whether these subunits are indeed associated, we asked whether the activity of translocated CRII is dependent on cAMP. Nuclear PKA activity was assayed in nuclear extracts prepared from cells exposed to ethanol for various periods of time and from control cells not exposed to ethanol. Extracts were analyzed for phosphotransferase activity using the PKA-specific substrate Kemptide. Western blots of the same nuclear extracts were used to estimate the relative amounts of C and RII in the nucleus. To control for protein loading, we used antibodies against nuclear protein AKAP-95 (16, 17). Fig. 2, A and B, shows that both C and RII levels are significantly increased in the nucleus after 1 and 3 h of ethanol exposure. After 6 h of ethanol treatment, C and RII returned to control levels. By 12 h and continuing for 24 h, C and RII were again translocated to the nucleus. Nuclear PKA activity parallels the time course of PKA subunit accumulation in the nucleus, with increased activity observed after 1 and 3 h of ethanol exposure (Fig. 2C). This activity is cAMP-dependent only during the first 12 h of ethanol exposure, suggesting that C and RII are associated during that time. However, after 24 h of ethanol exposure, nuclear PKA activity became cAMP-independent.

View larger version (39K): [in this window] [in a new window]   Fig. 2.   Ethanol-induced PKA activation in the nucleus. NG108-15 cells were incubated with ethanol for the indicated times, and nuclear extracts were prepared as under "Experimental Procedures." A, Western blot (40 µg of protein) of nuclear extracts prepared from NG108-15 cells exposed to 100 mM ethanol for the indicated times. Blots were probed with antibodies against C, RII, and AKAP95 as an internal control. The blot shown is representative of three experiments. B, densitometric quantitation of proteins in A. A.U., arbitrary units. *, significantly different (p < 0.05) from the corresponding protein level in the untreated cells (Student's t test). C, Kemptide phosphorylation by nuclear extracts in the absence (striped bars) or presence (solid bars) of 5 µM cAMP. Data are means of three independent experiments. *, significantly different (p < 0.05) from the basal activity in untreated cells (Student's t test).

Rp-cAMPS Analogs Differentially Block cAMP Activation of Type I and Type II PKA in Vitro-- The above results, demonstrating that C and RII preferentially associate in NG108-15 cells and selectively translocate to the nucleus after ethanol treatment, suggested that there may be a specific role of PKA type II in ethanol-induced CREB phosphorylation. To test this hypothesis, we used the commercially available antagonists of cAMP, Rp-Cl-cAMPS (RpI) and Rp-CPT-cAMPS (RpII), specific for type I and type II PKA, respectively (18, 19). To ensure that the cAMP antagonists were specific for their corresponding PKA types, we performed kinase assays using type I and type II holoenzymes that were either purified from skeletal muscle or immunoprecipitated with antibodies against RI and RII from NG108-15 cells. Type I and type II holoenzymes purified from rabbit skeletal muscle (Sigma) were preincubated in assay buffer with or without 50 µM RpI or RpII. These concentrations were previously reported to be sufficient to completely abolish cAMP activation of the purified holoenzymes in vitro (18). After preincubation for 30 min, the samples were tested for Kemptide phosphorylation in the absence and presence of cAMP. As shown in Fig. 3A, both RpI and RpII had no effect on the basal activity of either type I or type II holoenzymes in the absence of cAMP. In the presence of cAMP, however, the activity of the type I holoenzyme was inhibited by RpI but not RpII, whereas the activity of the type II holoenzyme was abolished by RpII but not RpI. Similar results were obtained with PKA isozymes immunoprecipitated from NG108-15 cells (Fig. 3B). Note that there was ~6 times more activity of type II than type I PKA immunoprecipitated from NG108-15 cells (compare y axis scales in Fig. 3B). This may account for the incomplete inactivation induced by 50 µM RpII of the immunoprecipitated type II PKA as compared with the effect of the same concentration of RpI on type I PKA. Taken together these results demonstrate that the inhibitors used in this study are potent and specific for their respective PKA types expressed in NG108-15 cells.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   PKA activity of type I and type II holoenzymes is inhibited by the corresponding isozyme-specific inhibitor. A, type I and type II holoenzymes purified from rabbit skeletal muscle were incubated with or without 50 µM RpI or RpII for 30 min and then tested for Kemptide phosphorylation in the presence and absence of cAMP. B, type I and type II holoenzymes immunoprecipitated (IP) with antibodies against RI or RII from total lysate of NG108-15 cells were incubated with or without 50 µM RpI or RpII for 30 min and then tested for Kemptide phosphorylation in the presence and absence of cAMP. The data shown are the means of three experiments. *, significantly different (p < 0.005) from the corresponding type without inhibitor (Student's t test); **, not statistically different from the corresponding type without inhibitor (Student's t test).

The Specific Inhibitor of Type II PKA Blocks Ethanol-induced C and RII Translocation to the Nucleus-- The PKA subunits C and RII preferentially associate in NG108-15 cells, and they both translocate to the nucleus after ethanol treatment (Figs. 1 and 2). Therefore, we hypothesized that the antagonist of cAMP (RpII) that specifically inhibited dissociation of regulatory RII from catalytic subunits would block ethanol-induced translocation of C and RII to the nucleus. To test this possibility, we prepared nuclear fractions from NG108-15 cells pretreated with 300 µM RpI or RpII for 2 h before exposure to 100 mM ethanol for 3 and 24 h. This concentration of the Rp-cAMPS analogs was previously reported to completely abolish cAMP activation of type I and type II in cultured cells (18, 20). Nuclear extracts were subjected to SDS electrophoresis, blotted, and probed with antibodies against C and RII (Fig. 4). As expected, RpII but not RpI, blocked C and RII translocation to the nucleus after ethanol treatment.

View larger version (64K): [in this window] [in a new window]   Fig. 4.   RpII but not RpI inhibits ethanol-induced translocation of C and RII to the nucleus. A, Western blot (40 µg protein) of nuclear extracts prepared from NG108-15 cells pretreated with 300 µM RpI or RpII for 2 h and with 100 mM ethanol for an additional 3 or 24 h. Blots were probed at the same time with antibodies against C and RII. The blot shown is representative of three experiments. B, densitometric quantitation of proteins in A. Data are the means of three experiments. *, significantly different (p < 0.05) from the corresponding control cells without ethanol (Student's t test). A.U., arbitrary units.

Forskolin-induced Translocation of C to the Nucleus Is Blocked by Type II PKA Inhibitor-- We next asked whether forskolin, a widely used adenylyl cyclase activator, also induces translocation of type II PKA to the nucleus. We performed Western blots of nuclear extracts to answer this question. Nuclear extracts from cells incubated with or without forskolin for 10 min were subjected to SDS electrophoresis, blotted, and probed with antibodies against C and RII. The results in Fig. 5, A and B, show that forskolin induces a robust increase in nuclear C, but the amount of RII is not changed. We used immunocytochemistry to investigate whether forskolin-induced translocation of C would be inhibited by RpII. NG108-15 cells were pretreated with 300 µM RpI or RpII for 2 h, and then 1 µM forskolin was added for another 10 min. Cells were fixed, blocked, and incubated with monoclonal antibodies against C and RII as described under "Experimental Procedures." The results in Fig. 5C show that C (red) and RII (green) are co-localized (yellow) mainly in the Golgi area in NG108-15 cells that were either untreated (a) or treated with RpI (c) or RpII (e) alone. After 10 min of forskolin treatment, C but not RII translocated to the nucleus (b). C translocation to the nucleus after forskolin treatment was blocked by RpII (f) but not RpI (d). Taken together the results in Figs. 4 and 5 demonstrate that both ethanol- and forskolin-induced C translocation is dependent on type II PKA and that RII translocates to the nucleus after ethanol, but not forskolin, treatment.

View larger version (45K): [in this window] [in a new window]   Fig. 5.   Forskolin-induced translocation of C to the nucleus is inhibited by RpII but not RpI. A, Western blot (40 µg of protein) of nuclear extracts prepared from NG108-15 cells treated without (C) or with 1 µM forskolin (F) for 10 min. Blots were probed at the same time with antibodies against C and RII. The blot shown is representative of three experiments. B, densitometric quantitation of proteins in A. Data are the means of three experiments. *, significantly different (p < 0.05) from the corresponding control cells without forskolin (Student's t test). A.U., arbitrary units. C, NG108-15 cells were either untreated (a) or preincubated with 300 mM RpI (c and d) or RpII (e and f) for 2 h followed by 1 µM forskolin for 10 min (b, d, and f). Cells were fixed, blocked, and probed with monoclonal antibodies against C (red) and RII (green) as described under "Experimental Procedures." All images are ×60 magnification, obtained with a Bio-Rad 1024 confocal microscope. The data shown are representative of three experiments.

Ethanol- and Forskolin-induced CREB Phosphorylation Is Regulated by Type II PKA in NG108-15 Cells-- The results presented above, demonstrating that C and RII preferentially associate in NG108-15 cells and selectively translocate to the nucleus after ethanol treatment, suggest that PKA type II may be responsible for the ethanol-induced CREB phosphorylation that we previously observed (6). To test this hypothesis, we exposed NG108-15 cells to forskolin for 10 min or to ethanol for various times ranging from 1 to 24 h in the presence or absence of RpI or RpII. We performed immunocytochemistry using antibodies specific for the phosphorylated form of CREB. Preincubation with 300 µM RpII abolished forskolin- and ethanol-induced CREB phosphorylation, whereas preincubation with the same concentration of RpI failed to inhibit both forskolin and ethanol-induced CREB phosphorylation at all time points tested (Fig. 6 and data not shown). These results demonstrate that phosphorylation of CREB after ethanol or forskolin treatment depends only on type II PKA.

View larger version (49K): [in this window] [in a new window]   Fig. 6.   Ethanol-induced CREB phosphorylation is due to type II PKA. Cells were preincubated for 2 h with 300 µM RpI or RpII and then further incubated with either 1 µM forskolin for 10 min or with100 mM ethanol for 3 or 14 h. Cells were fixed, blocked, and probed with polyclonal antibodies against CREB phosphorylated at Ser-133 as described under "Experimental Procedures." All images are ×60 magnification obtained with a Bio-Rad 1024 confocal microscope. Data shown are representative of six experiments.

Ethanol- and Forskolin-induced CRE-mediated Luciferase Expression Is Regulated by Both Type I and Type II PKA-- Increased expression of genes with an upstream CRE is an important functional consequence of PKA translocation to the nucleus and phosphorylation of the transcription factor CREB (21). We have shown that 14 h of ethanol exposure produces a maximal increase in luciferase activity in cells transiently transfected with a CRE-luciferase reporter construct (7). This increase was PKA-dependent since various PKA inhibitors including H89, Rp-cAMPS, and overexpression of a dominant negative RI construct abolished ethanol-induced luciferase activity (7). To evaluate the role of type I and type II PKA in ethanol-induced CRE-mediated gene expression, we transiently transfected NG108-15 cells with a CRE-luciferase reporter plasmid followed by incubation of the cells in 300 µM RpI or RpII for 2 h and subsequent addition of either 1 µM forskolin for another 4 h or 100 mM ethanol for 14 h. As shown in Fig. 7 both RpI and RpII totally abolished forskolin and ethanol-induced luciferase activity, suggesting that both types of PKA are required for forskolin and ethanol-induced gene transcription. Hence, although C translocation as well as phosphorylation of CREB after ethanol or forskolin treatment appears to be dependent only on type II PKA, gene expression in response to these compounds requires both type I and type II PKA.

View larger version (21K): [in this window] [in a new window]   Fig. 7.   Both type I and type II PKA are required for ethanol-induced CRE-mediated luciferase expression in NG108-15 cells. NG108-15 cells were transfected with a CRE-luciferase construct (150 ng) as described under "Experimental Procedures," and luciferase expression was measured in cells incubated in either 1 µM forskolin for 4 h or 100 mM ethanol (EtOH) for 14 h. Cells were preincubated with 300 µM RpI or RpII for 2 h before forskolin (F) or ethanol (E) treatment; the inhibitors remained during the course of the experiment. The results are expressed as absolute luciferase units (ALU)/mg of protein/min. Each point represents three separate experiments with six replicate samples for each point. Data are presented as means ± S.D. of the mean (S.E.). *, significantly different (p < 0.005) from the corresponding untreated cells (Student's t test).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The major finding in this study is that forskolin- and ethanol-induced CREB phosphorylation and gene activation are differentially regulated by type I and type II isoforms of PKA expressed in NG108-15 cells. Phosphorylation of CREB is solely dependent on type II PKA translocation to and activity in the nucleus. However, the consequent CRE-mediated increase in gene expression requires an additional step(s) that depends on type I PKA activity in the cytoplasm (see the model in Fig. 8).

View larger version (29K): [in this window] [in a new window]   Fig. 8.   Schematic representation of forskolin (red arrows) and ethanol (black arrows) modulation of PKA isoforms in NG108-15 cells. Ethanol (EtOH) inhibits adenosine uptake, leading to an increase in extracellular adenosine and activation of the A2 receptor. This in turn leads to an increase in cAMP (4). Ethanol also induces release of RII from its AKAP. Forskolin directly activates adenylate cyclase, leading to increases in cAMP level. Increased cAMP activates both types of PKA. Type II PKA translocates to the nucleus, inducing phosphorylation of CREB that is necessary but not sufficient for gene transcription. Type I PKA is activated in cytoplasm phosphorylating a protein(s) (black box) that in turn activates CREB-binding protein (CBP) or other cofactors required for gene transcription.

We show here that there is a specific association of PKA subunits in NG108-15 cells. Thus, RII associates only with C, whereas RI binds exclusively to the C catalytic subunit (Fig. 1). To the best of our knowledge, these results demonstrate for the first time a preferential combination of catalytic and regulatory subunits in living cells. The subunit composition of a holoenzyme contributes to its biochemical properties. Specifically, RI-containing holoenzymes need a 5-fold higher concentration of cAMP for activation than RI-containing holoenzymes (22, 23). Moreover, type II holoenzymes containing C are also more sensitive to dissociation by cAMP than are the C-containing type II holoenzymes (24). Thus, the preferential combination of C-RI in NG108-15 cells may confer a greater sensitivity to cAMP for this holoenzyme than for the C-RII holoenzyme. We show that the activity of PKA type I is 80% less than PKA type II at saturating cAMP concentrations (Fig. 3B). Nevertheless, the type I isoforms will still be a major contributor to phosphorylation at the low level of cAMP produced by ethanol (25). We also report here a specialized localization of PKA subunits in NG108-15 cells. RII and C reside in the particulate and nuclear fractions, whereas RI and C are found only in the soluble cytosolic fraction (Fig. 1). These results are consistent with other studies showing differential localization of type I and type II PKA in mammalian cells. Indeed, numerous studies have shown that RI is found in the soluble fraction (26), whereas RII is localized to membranes (27), the Golgi area (28), and nucleus (29-31) through binding to specific A kinase-anchoring proteins, AKAPs.

Only C and RII translocate to the nucleus after ethanol exposure. Both subunits accumulate in the nucleus with a similar time course. However, it is unlikely that the two subunits enter the nucleus as a holoenzyme, because its size probably would not permit passage through the nuclear pore. C is 40 kDa, and RII is 54 kDa, and the upper size limit for a protein to enter through the nuclear pore by diffusion is 40 kDa (32, 33). Studies from different laboratories have shown that increases in cAMP induce translocation of C to the nucleus, but regulatory subunits do not change their location inside the cells (33, 34). Therefore, the ethanol-induced translocation of RII to the nucleus cannot be explained solely by diffusion and increases in cAMP; several events must take place for type II PKA to move to the nucleus. First, an increase in cAMP is required to initiate dissociation of C from RII, because Rp-cAMPS analogs block this step. Then C should phosphorylate RII and unmask the nuclear localization signal present in RII. Finally, the binding of RII to its AKAP as well as the dimerization domain must be disrupted. All these events may well take place as a result of ethanol treatment. First, we and others (35, 36) have shown that acute ethanol exposure potentiates receptor-activated cAMP production mediated by several G protein-coupled receptors. Second, point mutation of the RII autophosphorylation site or in the nuclear localization signal abolishes its translocation to the nucleus (37). Third, the three-dimensional structure of the dimerization domain of the two RII molecules is a highly hydrophobic X-type four-helix bundle making contact with the hydrophobic helix domain of one AKAP molecule (38, 39). Therefore, it is possible that ethanol could simultaneously disrupt both RII-AKAP and RII-RII binding by interfering with this highly hydrophobic structure. This explanation is also supported by data in Fig. 5 showing that forskolin-induced increases in cAMP do not cause RII translocation to the nucleus, because RII-AKAP hydrophobic binding is not affected.

Once in the nucleus, C and RII appear to re-associate since the phosphotransferase activity of nuclear extracts is cAMP-dependent in the first 12 h of ethanol exposure (Fig. 2). We reconfirm here that after 24 h of ethanol exposure, nuclear PKA activity becomes cAMP-independent, suggesting two mechanisms in ethanol-induced translocation of type II PKA to the nucleus. Our preliminary data suggest that the loss of cAMP independence at 24 h requires protein synthesis during the first 2 h of ethanol exposure.²

Ethanol-induced translocation of C to the nucleus causes a significant increase in CREB phosphorylation, with a peak at 3 h followed by a sustained increase even after 24 h of ethanol exposure (6). In contrast, forskolin induces a robust but transient phosphorylation of CREB (40). Increased CREB phosphorylation is due primarily to PKA, since specific inhibitors of PKA like Rp-cAMPS (6) and H89 (not shown) inhibit forskolin and ethanol-induced CREB phosphorylation. The results showing that C and RII preferentially associate in NG108-15 cells (Fig. 1) and selectively translocate to the nucleus after ethanol treatment (Fig. 2) suggested a specific role for PKA type II in ethanol-induced CREB phosphorylation. We used the Rp-cAMPS analogs that have been shown to be specific for type I and type II PKA (18, 19) to investigate this possibility. Our results reconfirm that these analogs are specific for the corresponding PKA types in vitro (Fig. 3). Moreover, in intact NG108-15 cells, both ethanol-induced PKA type II (C and RII) and forskolin-induced C translocation to the nucleus are blocked by RpII, the specific inhibitor of PKA type II, but not by RpI, even after 24 h of ethanol exposure. Likewise, RpII but not RpI inhibited CREB phosphorylation, demonstrating that type II PKA is solely responsible for forskolin- and ethanol-induced CREB phosphorylation. In contrast, however, both RpI and RpII inhibited forskolin and ethanol-induced CRE-mediated transcription of the luciferase reporter gene (Fig. 7), suggesting that PKA type I and type II together are required for CRE-mediated gene expression. These results are in agreement with data from other laboratories showing that cerebellar granule cells that do not express RII fail to transmit cAMP signals to the nucleus as measured by CREB phosphorylation and subsequent gene transcription (41, 42). This is also consistent with findings that overexpression of RII restores cAMP-dependent transcription in a cAMP-unresponsive cell line (43) and that neural gene expression and c-fos induction by cAMP are defective in the striatum of RII/ mice (44, 45), possibly because there may not be CREB phosphorylation in this brain area of these animals.

Phosphorylation of CREB at serine 133 is an essential step for inducing transcription of genes containing a CRE in their promoter region (21). Another important step in this process is recruitment of CREB-binding protein, which links the DNA binding factor CREB to the basal transcription machinery (46, 47) and Fig. 8. There is increasing evidence that positive regulation (enhancement of cAMP-dependent transcription) is attained through PKA-dependent phosphorylation of a factor other than CREB that may be CREB-binding protein (CBP) itself (48, 49) or downstream of CBP (50, 51). Our results show that, in NG108-15 cells, ethanol and forskolin induce phosphorylation of CREB and transcription of the luciferase reporter gene that is PKA-dependent. However, we found that PKA types expressed in NG108-15 cells differentially regulate this process. RpI inhibits CRE-mediated gene transcription but not C translocation to the nucleus or CREB phosphorylation induced by ethanol or forskolin. RpII inhibits both steps. Therefore, we propose (Fig. 8) that after ethanol or forskolin treatment, type II PKA is translocated to the nucleus and phosphorylates CREB. This step is necessary but not sufficient for gene transcription. In addition, we propose that type I PKA is activated in the cytoplasm, turning on a pathway that promotes activation of other transcription cofactors that interact with CREB phosphorylated at Ser-133 to induce gene transcription. This is an important finding that may partially resolve apparent differences between different laboratories assessing the role of PKA in ethanol drinking behavior and sensitivity to ethanol. Drinking behavior could be viewed as a final consequence of transcription of new genes that regulate sensitivity to ethanol. Our results show that both type I and type II PKA are required for gene transcription. Therefore, the absence of RII in striatum of RII/ mice (10) or in mushroom bodies of Drosophila (52) may abolish ethanol-induced transcription of a gene responsible for increased sensitivity to the intoxicating effects of ethanol. Similarly, the expression of dominant negative RI in the forebrain and hippocampus of R(AB) transgenic mice (11) may abolish ethanol-induced transcription of another group of genes responsible for the decreased sensitivity to the hypnotic effects of ethanol. We present data showing that ethanol-dependent increases in CRE-mediated gene transcription requires both type I and type II PKA activity in distinct cellular compartments. Taken together with genetic data discussed above, it is likely that both types of PKA play an important role in molecular mechanisms that underlie addictive behaviors. This suggests that it may be possible to develop new drugs to treat the pathologic consequences of alcoholism that focus on one specific type of PKA without dramatically altering other cellular functions.

	ACKNOWLEDGEMENTS

We thank Drs. Dorit Ron and Jennifer Whistler for helpful discussions and critical review of the manuscript.

	FOOTNOTES

^* This work was supported in part by National Institutes of Health Grant R37 AA10030 and funds provided by the state of California for medical research on alcohol and substance abuse through the University of California at San Francisco.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ To whom correspondence should be addressed: Ernest Gallo Clinic and Research Center, 5858 Horton St., Suite 200, Emeryville, CA 94608. Tel.: 510-985-3142; Fax: 510-985-3101; E-mail: anconst@itsa. ucsf.edu.

Published, JBC Papers in Press, March 8, 2002, DOI 10.1074/jbc.M112107200

² A. Constantinescu, unpublished observations.

	ABBREVIATIONS

The abbreviations used are: PKA, cAMP-dependent protein kinase; AKAP, A kinase-anchoring protein; CREB, cAMP response element-binding protein; CRE, cAMP response element; R(AB), type I regulatory subunit of PKA mutated at cAMP binding sites A and B; Rp-Cl-cAMPS, (RpI) and Rp-CPT-cAMPS, (RpII).

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore