Role of Regulatory Subunits and Protein Kinase Inhibitor (PKI) in Determining Nuclear Localization and Activity of the Catalytic Subunit of Protein Kinase A

Abstract

Regulation of protein kinase A by subcellular localization may be critical to target catalytic subunits to specific substrates. We employed epitope-tagged catalytic subunit to correlate subcellular localization and gene-inducing activity in the presence of regulatory subunit or protein kinase inhibitor (PKI). Transiently expressed catalytic subunit distributed throughout the cell and induced gene expression. Co-expression of regulatory subunit or PKI blocked gene induction and prevented nuclear accumulation. A mutant PKI lacking the nuclear export signal blocked gene induction but not nuclear accumulation, demonstrating that nuclear export is not essential to inhibit gene induction. When the catalytic subunit was targeted to the nucleus with a nuclear localization signal, it was not sequestered in the cytoplasm by regulatory subunit, although its activity was completely inhibited. PKI redistributed the nuclear catalytic subunit to the cytoplasm and blocked gene induction, demonstrating that the nuclear export signal of PKI can override a strong nuclear localization signal. With increasing PKI, the export process appeared to saturate, resulting in the return of catalytic subunit to the nucleus. These results demonstrate that both the regulatory subunit and PKI are able to completely inhibit the gene-inducing activity of the catalytic subunit even when the catalytic subunit is forced to concentrate in the nuclear compartment. Regulation of protein kinase A by subcellular localization may be critical to target catalytic subunits to specific substrates. We employed epitope-tagged catalytic subunit to correlate subcellular localization and gene-inducing activity in the presence of regulatory subunit or protein kinase inhibitor (PKI). Transiently expressed catalytic subunit distributed throughout the cell and induced gene expression. Co-expression of regulatory subunit or PKI blocked gene induction and prevented nuclear accumulation. A mutant PKI lacking the nuclear export signal blocked gene induction but not nuclear accumulation, demonstrating that nuclear export is not essential to inhibit gene induction. When the catalytic subunit was targeted to the nucleus with a nuclear localization signal, it was not sequestered in the cytoplasm by regulatory subunit, although its activity was completely inhibited. PKI redistributed the nuclear catalytic subunit to the cytoplasm and blocked gene induction, demonstrating that the nuclear export signal of PKI can override a strong nuclear localization signal. With increasing PKI, the export process appeared to saturate, resulting in the return of catalytic subunit to the nucleus. These results demonstrate that both the regulatory subunit and PKI are able to completely inhibit the gene-inducing activity of the catalytic subunit even when the catalytic subunit is forced to concentrate in the nuclear compartment. The subcellular localization of multifunctional protein kinases is emerging as a critical regulatory mechanism to target activity to sites of second messenger generation or to a specific subset of intracellular substrates (1Mochly R.D. Science. 1995; 268: 247-251Crossref PubMed Scopus (835) Google Scholar, 2Coghlan V.M. Bergeson S.E. Langeberg L. Nilaver G. Scott J.D. Mol. Cell. Biochem. 1993; 128: 309-319Crossref Scopus (55) Google Scholar). One of the important actions of PKA 1The abbreviations used are: PKA, protein kinase A; R, regulatory; C, catalytic; PKI, protein kinase inhibitor; NLS, nuclear localization signal; NES, nuclear export signal; PCR, polymerase chain reaction; NPC, nuclear pore complex; CRE, cAMP response element; RSV, Rous sarcoma virus; AKAP, A kinase-anchoring protein; CMV, cytomegalovirus; CREB, cAMP responsive element binding protein; hGH, human growth hormone. involves the phosphorylation of transcription factors and the induction of specific genes. However, the PKA holoenzyme is normally found to be cytoplasmic, and only the catalytic subunit is thought to be able to enter the nucleus after its release from the holoenzyme. The translocation of the C subunit into the nucleus is the key event associated with the initiation of PKA-mediated induction of transcription. After translocation to the nucleus, the C subunit phosphorylates transcription factors such as CREB, specifically stimulating transcription (3Mellon P.L. Clegg C.H. Correll L.A. McKnight G.S. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 4887-4891Crossref PubMed Scopus (201) Google Scholar, 4Gonzalez G.A. Montminy M.R. Cell. 1989; 59: 675-680Abstract Full Text PDF PubMed Scopus (2064) Google Scholar, 5Sun P. Schoderbek W.E. Maurer R.A. Mol. Endocrinol. 1992; 6: 1858-1866PubMed Google Scholar). The C subunit of PKA has been shown to translocate into the nucleus after the microinjection of purified protein into the cytoplasm of cultured cells (6Fantozzi D.A. Taylor S.S. Howard P.W. Maurer R.A. Feramisco J.R. Meinkoth J.L. J. Biol. Chem. 1992; 267: 16824-16828Abstract Full Text PDF PubMed Google Scholar, 7Fantozzi D.A. Harootunian A.T. Wen W. Taylor S.S. Feramisco J.R. Tsien R.Y. Meinkoth J.L. J. Biol. Chem. 1994; 269: 2676-2686Abstract Full Text PDF PubMed Google Scholar). However, microinjection of preformed holoenzyme injected into the cytoplasm fails to traverse the nuclear membrane and remains cytoplasmically sequestered (8Meinkoth J.L. Alberts A.S. Went W. Fantozzi D. Taylor S.S. Hagiwara M. Montminy M. Feramisco J.R. Mol. Cell. Biochem. 1993; 128: 179-186Crossref Scopus (124) Google Scholar). The nuclear pore complex (NPC) is the primary route for proteins to diffuse back and forth across the nuclear membrane. Studies measuring pore permeability suggest that proteins larger than 23–26 nm in diameter, which corresponds approximately to a 40–60-kDa globular protein, are excluded from passage through the aqueous pore of the NPC (9Davis L.I. Annu. Rev. Biochem. 1995; 64: 865-896Crossref PubMed Scopus (288) Google Scholar, 10Newmeyer D.D. Curr. Opin. Cell Biol. 1993; 5: 395-407Crossref PubMed Scopus (61) Google Scholar). Because the C subunit is approximately 40 kDa, this places it at the upper size limit of proteins that can diffuse passively into the nucleus. Nevertheless, investigation into the mechanism of C subunit translocation into the nucleus has suggested that passive diffusion is sufficient to explain the kinetics and apparent nuclear accumulation of C (11Harootunian A.T. Adams S.R. Wen W. Meinkoth J.L. Taylor S.S. Tsien R.Y. Mol. Biol. Cell. 1993; 4: 993-1002Crossref PubMed Scopus (149) Google Scholar). The partitioning of C between cytoplasm and nucleus is further complicated by the recent discovery of the nuclear export properties of PKI. Residues 37–46 of PKI contain a leucine-rich region that has been shown to function as a nuclear export signal (NES) (12Wen W. Meinkoth J.L. Tsien R.Y. Taylor S.S. Cell. 1995; 82: 463-473Abstract Full Text PDF PubMed Scopus (1006) Google Scholar). PKI appears to transport the C subunit out of the nucleus in a C/PKI complex utilizing an ATP and temperature-dependent mechanism (6Fantozzi D.A. Taylor S.S. Howard P.W. Maurer R.A. Feramisco J.R. Meinkoth J.L. J. Biol. Chem. 1992; 267: 16824-16828Abstract Full Text PDF PubMed Google Scholar, 7Fantozzi D.A. Harootunian A.T. Wen W. Taylor S.S. Feramisco J.R. Tsien R.Y. Meinkoth J.L. J. Biol. Chem. 1994; 269: 2676-2686Abstract Full Text PDF PubMed Google Scholar). PKI is capable of freely entering the nucleus (6Fantozzi D.A. Taylor S.S. Howard P.W. Maurer R.A. Feramisco J.R. Meinkoth J.L. J. Biol. Chem. 1992; 267: 16824-16828Abstract Full Text PDF PubMed Google Scholar, 8Meinkoth J.L. Alberts A.S. Went W. Fantozzi D. Taylor S.S. Hagiwara M. Montminy M. Feramisco J.R. Mol. Cell. Biochem. 1993; 128: 179-186Crossref Scopus (124) Google Scholar, 13Wen W. Taylor S.S. Meinkoth J.L. J. Biol. Chem. 1995; 270: 2041-2046Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar) and actively shuttling the C subunit back into the cytoplasm where the C subunit can recombine with R subunits to form inactive holoenzyme and restore cAMP regulation to the cell. In this study, we utilize epitope-tagged C subunit to evaluate the ability of either the RII subunit or PKI to prevent nuclear accumulation and inhibit gene induction of a CRE-luciferase reporter. By adding an artificial nuclear localization signal (NLS) to the tagged C subunit we are able to compare the ability of RII and PKI to affect a strong nuclear localization signal and inhibit C subunit activity. Our results are consistent with the current model developed from microinjection experiments and further demonstrate that the entire RII/C holoenzyme complex would be transported to the nucleus if the C subunit contained an exposed NLS. In contrast to RII, PKI is capable of exporting a C subunit with a strong NLS, although export is not a prerequisite for inhibition of gene expression. The export process that is engaged by PKI appears to be saturable, suggesting that free PKI may interact with the export machinery. JEG3 cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (Life Technologies, Inc.) in a 10% CO2incubator at 37 °C. Before transfection, the cells were passaged onto 24-well plates at 2 × 104 cells/well. The cells were allowed to grow to approximately 50% confluence. The medium was reduced to 250 μl/well 3–6 h before transfection. The cells were transfected by standard calcium phosphate techniques (14Matthews R.P. Guthrie C.R. Wailes L.M. Zhao X. Means A.R. McKnight G.S. Mol. Cell. Biol. 1994; 14: 6107-6116Crossref PubMed Scopus (498) Google Scholar), adding 25 μl of the calcium phosphate/DNA precipitate to each well. The α168-luciferase reporter, containing 168 base pairs of the human gonadotropin α promoter (3Mellon P.L. Clegg C.H. Correll L.A. McKnight G.S. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 4887-4891Crossref PubMed Scopus (201) Google Scholar, 15Correll L.A. Woodford T.A. Corbin J.D. Mellon P.L. McKnight G.S. J. Biol. Chem. 1989; 264: 16672-16678Abstract Full Text PDF PubMed Google Scholar), was used at a concentration of 2.5 ng/well in all transfections as a cAMP responsive reporter. CTAG, NLS-CTAG, and RIITAG were used at a concentration of 2.5 ng/well unless otherwise stated. The RSV-PKIα construct was generated by synthesis of the nucleotide sequence inferred from the PKIα amino acid sequence (16Thomas J. Van Patton S.M. Howard P. Day K.H. Mitchell R.D. Sosnick T. Trewhella J. Walsh D.A. Maurer R.A. J. Biol. Chem. 1991; 266: 10906-10911Abstract Full Text PDF PubMed Google Scholar). RSV-PKIα was used at 37.5 ng/well unless otherwise designated. To control for promoter competition between MT-1 driven constructs, ZEM3 (an empty vector containing the metallothionein promoter) was added to each well to keep levels constant across conditions in each experiment. Bluescript KS+ was added to each reaction to bring the total DNA to 250 ng/well. Transfection efficiency was normalized by the addition of 50 ng/well of the constitutive reporter RSV-β-galactosidase to correct for variability between wells and transfection conditions. After the addition of DNA, the cells were placed in a 3% CO2incubator for 24 h. To induce expression of the MT-1-driven constructs, the cells were grown in 2.5% fetal bovine serum supplemented Dulbecco's modified Eagle's medium with 100 μm ZnCl2 for 12–15 h at 10% CO2. After zinc treatment, cells used in the reporter assays were washed once with cold phosphate-buffered saline, lysed, and The assays were as (14Matthews R.P. Guthrie C.R. Wailes L.M. Zhao X. Means A.R. McKnight G.S. Mol. Cell. Biol. 1994; 14: 6107-6116Crossref PubMed Scopus (498) Google Scholar). activity was for transfection efficiency by to activity in each of were in and the for each was The was by a containing the region and the of the sequence adding the to the The is to the of and the acid with an added for of the expression vector was used as a to the that was with and and into McKnight G.S. Proc. Natl. Acad. Sci. U. S. A. 1992; PubMed Scopus Google Scholar), a MT-1-driven C expression vector with the from the human growth The was a containing a by the sequence for the amino of the nuclear localization signal and the of and the as was used to the The was and the resulting was into The was used to the to the of with a to the region of by an to the RIITAG was a to the region of to the used in the of and with a to the region that a at the The construct was generated by the and and the in the nuclear export signal to were generated to the PKIα a and a with 25 of between the of the and the of the The purified were and of to the synthesis of the PKIα before the addition of the and of containing and were added to the and were The resulting purified was into the vector The generation of was the and with The region was into the of both and were and After 12–15 h of zinc treatment, the cells were with in phosphate-buffered and with in phosphate-buffered The cells were with the at a of for h. The cells were with the at by with at in × After the cells were in μm to the were as (14Matthews R.P. Guthrie C.R. Wailes L.M. Zhao X. Means A.R. McKnight G.S. Mol. Cell. Biol. 1994; 14: 6107-6116Crossref PubMed Scopus (498) Google Scholar). The was a The localization of either free C subunit or holoenzyme was by JEG3 cells with an epitope-tagged C subunit with and When the construct was transfected expressed C subunit was throughout the cell. Because the region of the C subunit has been shown to affect activity Zhao X. McKnight G.S. J. Biol. Chem. 1993; 268: Full Text PDF PubMed Google Scholar, S.S. W. J. 1993; Google Scholar), it was critical to adding the to the of the C subunit activity of the expressed of the CRE-luciferase reporter into cells that expression of C was associated with of suggesting that the C was However, it was that the activity was of C subunit from its with R in holoenzyme. To this were a tagged mutant C subunit that not with R McKnight G.S. Proc. Natl. Acad. Sci. U. S. A. 1992; PubMed Scopus Google Scholar). In levels of of the CRE-luciferase reporter as C subunit and CTAG, suggesting that the addition of the to the region of the protein not activity not When and RII were to form the to the cytoplasm C, and this with a in of the CRE-luciferase reporter that the expressed C and R subunits to form the holoenzyme and that this into the nucleus, consistent with that the holoenzyme is to diffuse between and nuclear D.A. Harootunian A.T. Wen W. Taylor S.S. Feramisco J.R. Tsien R.Y. Meinkoth J.L. J. Biol. Chem. 1994; 269: 2676-2686Abstract Full Text PDF PubMed Google Scholar, 8Meinkoth J.L. Alberts A.S. Went W. Fantozzi D. Taylor S.S. Hagiwara M. Montminy M. Feramisco J.R. Mol. Cell. Biochem. 1993; 128: 179-186Crossref Scopus (124) Google Scholar). In of the cells transfected with appeared to be nuclear accumulation of the was C subunit by microinjection D.A. Harootunian A.T. Wen W. Taylor S.S. Feramisco J.R. Tsien R.Y. Meinkoth J.L. J. Biol. Chem. 1994; 269: 2676-2686Abstract Full Text PDF PubMed Google Scholar, 8Meinkoth J.L. Alberts A.S. Went W. Fantozzi D. Taylor S.S. Hagiwara M. Montminy M. Feramisco J.R. Mol. Cell. Biochem. 1993; 128: 179-186Crossref Scopus (124) Google Scholar). One is that the C subunit a or NLS, although and it (11Harootunian A.T. Adams S.R. Wen W. Meinkoth J.L. Taylor S.S. Tsien R.Y. Mol. Biol. Cell. 1993; 4: 993-1002Crossref PubMed Scopus (149) Google Scholar). We the ability of a well to both the subcellular localization of C and its activity in a gene induction The was onto the amino of to After transfection into JEG3 was to the nucleus 2 C, upper expression in levels of of the CRE-luciferase reporter, demonstrating that the expressed protein was In experiments with the RII regulatory subunit, it was that in the nucleus C, However, although the R subunit not prevent nuclear it prevent of the CRE-luciferase construct 2 the C subunit was to the nucleus but was not suggested that the the C subunit translocation of the entire holoenzyme into the nucleus. To this an RIITAG expression vector was RIITAG expressed was only found in the and with a expression vector the localization of the RIITAG However, when RIITAG was with in was nuclear translocation of the RIITAG protein that an and exposed the C subunit was sufficient to translocate the entire holoenzyme to the nucleus. PKI has been shown to actively export the C subunit from the nucleus the nuclear export signal PKI between 37 and (12Wen W. Meinkoth J.L. Tsien R.Y. Taylor S.S. Cell. 1995; 82: 463-473Abstract Full Text PDF PubMed Scopus (1006) Google Scholar). To correlate export of the C subunit with the ability of PKI to inhibit C subunit of CRE-luciferase were with and PKI. were to the of to inhibit the of C subunit activity not with sufficient PKI to inhibit C subunit activity by than in of the protein We the of and export by in the presence or of PKI. Transfection with in nuclear localization as shown in of with PKI from completely nuclear to completely Because PKI has been shown to be expressed in a cell we that be cell expression of export factors to the variability in localization However, when cells were at in the cell the variability in localization demonstrating it was not a cell not One for export is that of PKI and expression results in a of PKI to C protein in was by increasing PKI and both the CRE-luciferase and the subcellular localization of each transfection A that increasing of PKI a in CRE-luciferase However, of a localization such that increasing of PKI localization the cytoplasm only of the cells the concentration of PKI this in a in the localization and a return to nuclear localization such that of transfected cells were as nuclear that is an of PKI to C for export and that further in PKI with the C/PKI complex for binding to the export machinery. has been that PKI C subunit activity and actively the C subunit out of the nucleus in a C/PKI These are in of the However, the of C gene expression has been The localization of and its ability to CRE-luciferase was of either PKI or mutant PKI in which the critical the nuclear export signal been to of the PKI a both with to export and inhibition of of the CRE-luciferase reporter with (12Wen W. Meinkoth J.L. Tsien R.Y. Taylor S.S. Cell. 1995; 82: 463-473Abstract Full Text PDF PubMed Scopus (1006) Google Scholar), the mutant PKI to export the C subunit out of the nucleus A of mutant and PKI that mutant PKI was a inhibitor of CRE-luciferase at However, with levels of PKI, both the mutant and proteins were in activity. that export of the C subunit may a important in the regulation of C subunit activity when PKI is at to C subunit but that export is not essential for inhibition of C gene expression. The subcellular localization and of PKA has a of and been that may this The localization of holoenzyme is suggested to be a mechanism to bring PKA to increasing as well as the of is that to PKA to sites of cAMP generation and in cAMP to PKA for PKA localization been A of proteins for PKA been that with nm to the regulatory and are associated with or Scott J.D. J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar, R.Y. J. Biol. Chem. PubMed Scopus Google Scholar). been that can to regulatory subunits although with J. Taylor S.S. Proc. Natl. Acad. Sci. U. S. A. PubMed Scopus Google Scholar, J. Taylor S.S. J. Biol. Chem. PubMed Scopus Google Scholar). The and of C been shown to be C.R. McKnight G.S. J. Biol. Chem. PubMed Scopus Google Scholar), and this a in membrane although for this to The inhibitor of PKA, PKI, has been shown to nuclear export of C subunits back to the cytoplasm D.A. Harootunian A.T. Wen W. Taylor S.S. Feramisco J.R. Tsien R.Y. Meinkoth J.L. J. Biol. Chem. 1994; 269: 2676-2686Abstract Full Text PDF PubMed Google Scholar, W. Harootunian A.T. Adams S.R. Feramisco J. Tsien R.Y. Meinkoth J.L. Taylor S.S. J. Biol. Chem. 1994; 269: Full Text PDF PubMed Google Scholar), and this may to the We used an to C subunits that were free or to either RII or PKI. The nuclear of C subunit was with the of a reporter Our results demonstrate that C subunit to both and nuclear and gene However, when C was into an nuclear C was and transcription from the CRE-luciferase reporter to base These results are in with microinjection of tagged C and R proteins (11Harootunian A.T. Adams S.R. Wen W. Meinkoth J.L. Taylor S.S. Tsien R.Y. Mol. Biol. Cell. 1993; 4: 993-1002Crossref PubMed Scopus (149) Google Scholar) as well as the C subunit by J. 4: PubMed Scopus Google Scholar, G. Cell. Full Text PDF PubMed Scopus Google Scholar). The holoenzyme the size that is capable of at through the In are capable of binding holoenzyme and its to the nucleus. When cAMP the holoenzyme and C subunit, this is to be to diffuse into the nucleus where it can nuclear such as and transcription activity. The C subunits not to contain an NLS, although a sequence the human is at amino acid of this sequence been shown to nuclear accumulation or activity of L. A. L. and G. S. We an onto the amino of C to the ability of either RII or PKI to the localization and with a well signal for nuclear The presence of an the C subunit transported of the protein into the nucleus, and was an of reporter gene activity. that the C subunit is completely to the nucleus when an exposed is to the protein and that is in the free C When an RII expression vector was added to the transfections with the NLS-CTAG, the ability of C to induce gene activity was completely but the C subunit that the is holoenzyme and that the is capable of this entire holoenzyme to the nucleus where it remains We this by a tagged RII expression vector and by the localization of RII in the presence or of RII when transfected its or when transfected with However, in the presence of the RII was transported into the nucleus, the of which RII with a strong was capable of all of the holoenzyme to the nucleus. all cells to we not which are in and it is that if been of the holoenzyme in the compartment. The of C out of the nucleus has been shown to be by PKI, which not only and C subunit activity but as an to target C to the nuclear export (12Wen W. Meinkoth J.L. Tsien R.Y. Taylor S.S. Cell. 1995; 82: 463-473Abstract Full Text PDF PubMed Scopus (1006) Google Scholar). are of PKI and J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar, J. Biol. Chem. 1993; 268: Full Text PDF PubMed Google Scholar), and in addition to the C subunit each has a leucine-rich These leucine-rich nuclear export been shown to interact with and with form a complex that can be through the NPC M. M. Cell. Full Text Full Text PDF PubMed Scopus Google Scholar, M. M. M. Cell. Full Text Full Text PDF PubMed Scopus Google Scholar). in the proteins and the of the complex to process the C subunit to the cytoplasm where it is to to R subunits when cAMP has to PKI may function to the nuclear actions of C subunit and the PKA to When was transfected with PKI, the subunit not in the nucleus to a and gene expression was inhibited. The complex of C and PKI may be to diffuse through the NPC, and the C/PKI that enter the nucleus be However, when the was transfected with of PKI, a complicated the of PKI to expression vector the induction of CRE-luciferase to The nuclear localization of at with increasing PKI, but as PKI expression vector was added to the the tagged C subunit in the nucleus and in the We this to that of PKI with the nuclear export the to export the C/PKI with suggesting that the PKI is and only when C to PKI (12Wen W. Meinkoth J.L. Tsien R.Y. Taylor S.S. Cell. 1995; 82: 463-473Abstract Full Text PDF PubMed Scopus (1006) Google Scholar, W. Harootunian A.T. Adams S.R. Feramisco J. Tsien R.Y. Meinkoth J.L. Taylor S.S. J. Biol. Chem. 1994; 269: Full Text PDF PubMed Google Scholar). One be that free PKI with of the export and function a target for export in the of Because PKI as both an inhibitor and of C subunit, we the of the export process in the control of gene expression. A PKI expression vector was with in the (12Wen W. Meinkoth J.L. Tsien R.Y. Taylor S.S. Cell. 1995; 82: 463-473Abstract Full Text PDF PubMed Scopus (1006) Google Scholar) that it of with the export but not affect C subunit the mutant PKI not the of CTAG, which distributed in both the cytoplasm and nucleus. However, when the ability of this mutant PKI to inhibit gene induction by C subunit was with PKI, we found only a in which was at levels of PKI transfection. Our results suggest that the export function of PKI is not essential for gene induction by PKA and that the of PKI export is to the for gene induction The ability of the C subunit of PKA to gene expression is well and thought to be important in the regulation of and the of gene expression in In at the C subunit to the nucleus to nuclear substrates such as and of the of transcription However, C be able to and and it is not that the C subunit not a strong nuclear localization signal. To gene expression and the the C subunit be to a holoenzyme The that PKI can to transport C out of the nucleus that it may an essential in the regulation of gene expression by to the function of PKI in may the of PKI as both an inhibitor and of C subunit in the cell.