Zinc Release from Protein Kinase C as the Common Event during Activation by Lipid Second Messenger or Reactive Oxygen

Abstract

Zinc is a structural component of many regulatory molecules including transcription factors and signaling molecules. We report that two alternate signaling pathways of protein kinase C (PKC) activation involving either the lipid second messengers (diacylglycerol and its mimetics, the phorbol esters) or reactive oxygen converge at the zinc finger of the regulatory domain. They all trigger the release of zinc ions. An increase in intracellular free Zn2+ was observed by confocal microscopy in intact cells treated with phorbol ester or by mild oxidation. The source of liberated Zn2+ was traced to PKC and particularly the zinc finger domains. The activated form of native PKCα contained significantly less Zn2+ than the resting form. Furthermore, purified recombinant PKC protein fragments shed stoichiometric amounts of Zn2+ upon reaction with diacylglycerol, phorbol ester, or reactive oxygen in vitro. Our results offer new insight into the regulation of PKC. Far from cementing rigid structures, zinc actually is the linchpin that orchestrates dynamic changes in response to specific signals, allowing kinase activity to be turned on or off. Zinc is a structural component of many regulatory molecules including transcription factors and signaling molecules. We report that two alternate signaling pathways of protein kinase C (PKC) activation involving either the lipid second messengers (diacylglycerol and its mimetics, the phorbol esters) or reactive oxygen converge at the zinc finger of the regulatory domain. They all trigger the release of zinc ions. An increase in intracellular free Zn2+ was observed by confocal microscopy in intact cells treated with phorbol ester or by mild oxidation. The source of liberated Zn2+ was traced to PKC and particularly the zinc finger domains. The activated form of native PKCα contained significantly less Zn2+ than the resting form. Furthermore, purified recombinant PKC protein fragments shed stoichiometric amounts of Zn2+ upon reaction with diacylglycerol, phorbol ester, or reactive oxygen in vitro. Our results offer new insight into the regulation of PKC. Far from cementing rigid structures, zinc actually is the linchpin that orchestrates dynamic changes in response to specific signals, allowing kinase activity to be turned on or off. Zinc is essential to the structure of numerous signaling proteins. Transcription factors and signaling molecules share zinc finger structures as a common motif, although they differ in composition and function (1Ahmed S. Kozma R. Lee J. Monfries C. Harden N. Lim L. Biochem. J. 1991; 280: 233-241Crossref PubMed Scopus (91) Google Scholar). The C2H2 zinc finger is believed to provide rigidity to transcription factors for proper DNA binding capacity. By comparison, the structure/function relationship of the components in the C3H1 zinc fingers of signaling molecules is not fully understood. We report here that Zn2+ ions of the cysteine-rich domain play a dynamic role in the function of protein kinase C. Protein kinase C (PKC) 1The abbreviations used are: PKC, protein kinase C; PMA, phorbol 12-myristate 13-acetate; TSQ, N-(6-methoxy)-8-quinolyl-toluene sulfonamide; PBS, phosphate-buffered saline; TPEN, N,N,N′,N′-tetrakis-(2-pyridylmethyl)ethylenediamine; PAR, 4-(2-pyridylazo-resorcinol); GST, glutathioneS-transferase; PMPS, p-hydroxy-mercuri-phenylsulfonate; FACS, fluorescence-activated cell sorter 1The abbreviations used are: PKC, protein kinase C; PMA, phorbol 12-myristate 13-acetate; TSQ, N-(6-methoxy)-8-quinolyl-toluene sulfonamide; PBS, phosphate-buffered saline; TPEN, N,N,N′,N′-tetrakis-(2-pyridylmethyl)ethylenediamine; PAR, 4-(2-pyridylazo-resorcinol); GST, glutathioneS-transferase; PMPS, p-hydroxy-mercuri-phenylsulfonate; FACS, fluorescence-activated cell sorterisoforms function as central signal amplifiers. They are engaged by two alternate pathways. Most isoforms with the exception of atypical ones (2Kazanietz M.G. Areces L.B. Bahador A. Mischak H. Goodnight J. Mushinski J.F. Blumberg P.M. Mol. Pharmacol. 1993; 44: 298-307PubMed Google Scholar) are activated by the second messenger, diacylglycerol (3Nishizuka Y. Science. 1992; 258: 607-614Crossref PubMed Scopus (4219) Google Scholar,4Nishizuka Y. FASEB J. 1995; 9: 484-496Crossref PubMed Scopus (2357) Google Scholar), or its mimetics, the phorbol esters (5Kazanietz M.G. Bustelo X.R. Barbacid M. Kolch W. Mischak H. Wong G. Pettit G.R. Bruns J.D. Blumberg P.M. J. Biol. Chem. 1994; 269: 11590-11594Abstract Full Text PDF PubMed Google Scholar). Independent of this classic pathway, PKC is also controlled by a redox mechanism where oxidation converts the protein to the catalytically competent form (6Gopalakrishna R. Anderson W.B. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 6758-6762Crossref PubMed Scopus (368) Google Scholar, 7Konishi H. Tanaka M. Takemura Y. Matsuzaki H. Ono Y. Kikkawa U. Nishizuka Y. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 11233-11237Crossref PubMed Scopus (535) Google Scholar, 8Knapp L.T. Klann E. J. Biol. Chem. 2000; 275: 24136-24145Abstract Full Text Full Text PDF PubMed Scopus (167) Google Scholar), whereas reduction reverses this process (9Imam A. Hoyos B. Swenson C. Levi E. Chua R. Viriya E. Hammerling U. FASEB J. 2001; 15: 28-30Crossref PubMed Scopus (68) Google Scholar). Cofactors are required for enzyme regulation by both pathways. Calcium and phosphatidylserine are important for amplification of the diacylglycerol signal, whereas vitamin A, as we have shown recently, is needed for efficient redox activation of several PKC isoforms (9Imam A. Hoyos B. Swenson C. Levi E. Chua R. Viriya E. Hammerling U. FASEB J. 2001; 15: 28-30Crossref PubMed Scopus (68) Google Scholar). The lipid binding sites, accommodating diacylglycerol or phorbol ester on one hand (10Burns D.J. Bell R.M. J. BIol. Chem. 1991; 266: 18330-18338Abstract Full Text PDF PubMed Google Scholar, 11Ono Y. Fuji T. Igarashi K. Takayoshi K. Tanaka C. Kikkawa U. Nishizuka Y. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 4868-4871Crossref PubMed Scopus (392) Google Scholar) and retinol on the other (12Hoyos B. Imam A. Chua R. Swenson C. Tong G.C. Levi E. Noy N. Hammerling U. J. Exp. Med. 2000; 192: 835-846Crossref PubMed Scopus (67) Google Scholar) are located at the non-overlapping regions within the twin cysteine-rich structures (referred to as Cys domains) of the regulatory domains. These 50 amino acid-long highly homologous stretches contain 6 conserved cysteine and 2 conserved histidine residues, tetrahedrically coordinated by two Zn2+ ions into a composite zinc finger (1Ahmed S. Kozma R. Lee J. Monfries C. Harden N. Lim L. Biochem. J. 1991; 280: 233-241Crossref PubMed Scopus (91) Google Scholar, 13Hurley J.H. Newton A.C. Parker P.J. Blumberg P., M. Nishizuka Y. Protein Sci. 1997; 6: 477-480Crossref PubMed Scopus (317) Google Scholar, 14Zhang G. Kazanietz M.G. Blumberg P.M. Hurley J.H. Cell. 1995; 81: 917-924Abstract Full Text PDF PubMed Scopus (595) Google Scholar, 15Mott H.R. Carpenter J.W. Zhong S. Ghosh S. Bell R.M. Campbell S.L. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 8312-8317Crossref PubMed Scopus (157) Google Scholar). We show that the diacylglycerol phorbol ester pathway and the alternate redox pathway converge at this zinc finger structure. The common event is the release of the chelated zinc ions. NIH 3T3 cells were grown to semi-confluence in minimum essential high glucose medium supplemented with 7% fetal bovine serum. Trichoplusia ni insect cells (Invitrogen) were cultured in TNM-FH insect medium with 7% fetal bovine serum. At semi-confluence, the cells were infected with baculovirus constructs that harbored the full-length human PKCα or ζ-genes or empty vector. Cells were loaded with 1 μm N-(6-methoxy)-8-quinolyl-toluene sulfonamide (TSQ) (Molecular Probes) for 20 min, washed intensively with PBS, and moved to the stage of confocal laser scanning microscope LSM 510. Images were acquired with a Zeiss/LSM510 system equipped with the Enterprise UV laser. The 351-nm line was used for the excitation of TSQ, and fluorescence below 545 nm was captured. The focal plane was set close to the nuclear center. Fluorescence intensity was recorded on-line for several minutes to obtain background fluorescence, and after the addition of activator, the acquisition of images at 30-s intervals was continued for 30 min or longer until saturation was reached. Morphometric analysis was performed using the MetaMorph image analysis software. The confocal projection images present one section at 1.6-μm optical resolution in the Z direction at 630-fold optical magnification. Because laser irradiation may stimulate various changes in redox-sensitive systems, control-bleaching experiments were performed. Exposing the loaded cells three times for 45 s at the maximal laser power did not change TSQ fluorescence (35.9 ± 6.8 and 31.3 ± 7.5 units in the control and postbleached cells, respectively). To ascertain the specificity of TSQ for Zn2+, the specific zinc chelatorN,N,N′,N′-tetrakis-(2-pyridylmethyl)ethylenediamine (TPEN) was used in competition experiments. TSQ was added to 2–10-μm protein solutions in phosphate-buffered saline to a final concentration of 10 μm. Protein solutions were excited at 334 nm, and fluorescence emission was recorded at 465 nm in a Jasco FP-777 spectrofluorometer. A standard zinc solution (Sigma) was used to calibrate the assay system. For zinc determination by 4-(2-pyridylazo-resorcinol) (PAR), 40 μm of solution of recombinant GST fusion protein in 50 mm Tris buffer, pH 7.5, was mixed with 100 μm PAR and oxidized with indicated peroxide concentrations for 6 min at 37 °C or treated with phorbol 12-myristate 13-acetate (PMA). Optical density was read at 500 nm. Maximum zinc release was achieved by the addition of 500 μm p-hydroxy-mercuri-phenylsulfonate (PMPS). Suspensions of thymus or spleen cells of 4–6-week-old C57Bl/6 mice were activated by incubation with 166 μm PMA or 100 μmhydrogen peroxide for 15 min at 37 °C. Resting or activated cells were lysed at 0 °C in a standard lysis buffer (50 mmTris-HCl, pH 7.5, 1 mm EDTA, 10 mmdithiothreitol, 1 mm orthovanadate, 10 μg/ml aprotinin, 10 μg/ml leupeptin, 1% Triton X-100). PKCα was precipitated with PKCα-specific monoclonal antibody (Transduction Laboratories), and the immunoprecipitates were collected on protein G-agarose beads of standard size or of average size of 34 μm in diameter for FACS analysis (Amersham Biosciences). The beads were washed once with lysis buffer and three times with 50 mm Tris-HCl, pH 7.5. Samples were divided into two equal portions, which were analyzed for 1) PKC amount by Western blot and phosphotransfer activity as described previously (9Imam A. Hoyos B. Swenson C. Levi E. Chua R. Viriya E. Hammerling U. FASEB J. 2001; 15: 28-30Crossref PubMed Scopus (68) Google Scholar) and for 2) zinc content by a modified TSQ assay (vide infra). In the TSQ assay, the beads coated with PKC immunoprecipitates of activated cells were analyzed for residual zinc content in comparison with those of resting control cells. Beads were stained with 10 μm TSQ solution for 5 min in the cold and then inspected by confocal fluorescence microscopy as described above. After acquiring background fluorescence, the thiol-reactive reagent PMPS was added and the proportional increase in fluorescence emission caused by the displaced Zn2+ was recorded. The zinc content of non-activated cells was taken as reference point, and that of activated PKC samples was computed as a fraction thereof. Additionally, the 34-μm TSQ-stained beads were analyzed by the fluorescence-activated cell sorter. After determining background fluorescence emission, the beads were treated with PMPS and the proportional increase in fluorescence was measured by FACS. The reported values represent the fractions of zinc retained in the activated form compared with non-activated PKCα. Because no method exists to quantitatively separate activated PKC species from inactive forms, no attempts were made to determine molar ratios. The control- and PMA-treated insect cells were rinsed with PBS and fixed for 10 min at room temperature with 2% (v/v) paraformaldehyde in PBS. Formaldehyde excess was neutralized with 0.1 m glycine in PBS for 15 min at room temperature. Cells were washed with PBS and permeabilized with 0.1% Triton X-100 (w/v) in PBS containing 1% bovine serum albumin for 10 min at room temperature. After the washing and blocking of nonspecific binding sites by 1% bovine serum albumin, the cells were stained for 1 h at room temperature with anti-PKCα antibody (Transduction Laboratories) followed by fluorescein-conjugated goat anti-mouse IgG antibody. The immunostained cells were observed in a confocal laser scanning microscope as described above with the modification that the 488-nm line was used for the excitation of fluorescein. Intact NIH 3T3 fibroblasts treated with PMA (Fig.1, A andB) as well as neonatal cardiomyocytes and COS cells (data not shown) released within minutes substantial amounts of Zn2+ as visualized by confocal microscopy using as indicator the membrane-permeable Zn2+-sensitive probe TSQ. The selective zinc-chelating agent TPEN quenched the TSQ fluorescence, indicating Zn2+ specificity (Fig. 1). To trace Zn2+ release to PKC, we probed T. ni insect cells infected with baculovirus constructs that harbored the full-length human PKCα or ζ-genes or empty vector. PMA produced a strong increase in fluorescence in the cells that overexpress PKCα but not in the cells expressing PKCα- or PKCζ-encoding genes or the empty vector (Fig. 1,C–F), in agreement with the known preference of PMA for the α-isoform (5Kazanietz M.G. Bustelo X.R. Barbacid M. Kolch W. Mischak H. Wong G. Pettit G.R. Bruns J.D. Blumberg P.M. J. Biol. Chem. 1994; 269: 11590-11594Abstract Full Text PDF PubMed Google Scholar). Both α- and ζ-isoforms responded to hydrogen peroxide (Fig. 1, E and F). Monitoring zinc release over time after peroxide administration revealed a noticeable signal in 60 s and a gradual increase to saturation within 15 min. PMA triggered the release of Zn2+ equally and rapidly in human PKCα-expressing insect cells. By comparison, PMA liberated a modest amount of Zn2+ in empty vector control insect cells, probably from insect PKC. The morphological appearance of TSQ staining changed from homogeneous cytoplasmic distribution of fluorescence in control toward a patchy distribution in PMA-treated cells, reminiscent of the vesicle-associated zinc accumulation seen in neurons (16Velazquez R.A. Cai Y. Shi Q. Larson A.A. J. Neurosci. 1999; 6: 2288-2300Crossref Google Scholar). However, the TSQ fluorescence pattern did not overlap with that of PKC. As in mammalian cells, PMA caused PKCα to translocate to the plasma membrane as shown by immunofluorescence (Fig. 1 G). Different locations of TSQ and PKC indicate that released Zn2+ was responsible for enhanced TSQ fluorescence and not the binding of TSQ to PKC. Altogether, the results of Fig. 1 indicated that PKCα was the source of free intracellular Zn2+. The concept that zinc movements were important for PKC to gain enzymatic capacity was validated by comparing the Zn2+contents in resting and active PKC with the expectation that the former should contain more Zn2+. To test this assumption, native PKCα was immunoprecipitated from lysates of mouse splenocytes and thymocytes before and after PMA or hydrogen peroxide treatment. The immune complexes were collected on protein G coupled to agarose beads. Taking advantage of its hydrophobic nature, we loaded protein G-agarose beads with TSQ as a zinc-sensitive probe. Beads were analyzed by confocal microscopy and flow cytometry, the two methods giving concordant results. This assay system was validated by the addition of external Zn2+, which resulted in a strictly dose-dependent fluorescence emission. The fluorescence signal was quenched by TPEN. To measure the relative amounts of thiol-bound Zn2+ in PKC, we treated the precipitates with the thiol-reactive compound PMPS, which quantitatively displaced Zn2+ (17Hunt J.B. Neece S.H. Schachman H.K. Ginsburg A. J. Biol. Chem. 1984; 259: 14793-14803Abstract Full Text PDF PubMed Google Scholar). The liberated Zn2+ partitioned to TSQ, increasing the fluorescence of the beads. Fig.2 displays the confocal images of TSQ-stained beads containing activated and resting PKC immunoprecipitates before and after PMPS addition together with control beads. Fig. 2, bar graphs, shows the results of quantitations of the confocal images by MetaMorph (Fig. 2 B) and of the FACS analyses (Fig. 2 D). Statistically significant differences in mean fluorescence values are evident. PMPS displaced far less Zn2+ from in vivo activated PKCα than from the resting ester caused a of of thiol-bound zinc contained in inactive PKC, whereas oxidation over of PMA as well as peroxide was also of in of immunoprecipitates from resting cells and less from activated cells, the in vivo (data not Beads were in for activity and for PKC amount by Western blot to equal (Fig. 2 the zinc content with PKC To the that PKC was the source of liberated Zn2+ and to trace Zn2+ release to the zinc finger we in the GST fusion containing the human PKCα or J.H. Newton A.C. Parker P.J. Blumberg P., M. Nishizuka Y. Protein Sci. 1997; 6: 477-480Crossref PubMed Scopus (317) Google Scholar), using a assay on TSQ as a probe. PKCα and Cys released Zn2+ upon PMA whereas and Cys did However, all of the Zn2+ after treatment. of the are for the Both and PMA triggered the release of stoichiometric amounts of of lipid one of free Zn2+. was of the zinc was PMPS was used to determine Zn2+ content in the GST fusion The measured was as high as that by a of However, the molar of zinc to protein was below the of indicating that not all molecules were and of active and inactive by also to Zn2+ release from the protein that PMA saturation within a minutes but as Zn2+. the domain as a control protein not known to PMA S. J. Bell R.M. J. Biol. Chem. 1994; 269: Full Text PDF PubMed Google Scholar), we did not Zn2+ release with PMA, whereas oxidation was (Fig. The results achieved with TSQ were by the used assay with PAR (17Hunt J.B. Neece S.H. Schachman H.K. Ginsburg A. J. Biol. Chem. 1984; 259: 14793-14803Abstract Full Text PDF PubMed Google Scholar). To the of fully the of retinol binding sites was as described previously (12Hoyos B. Imam A. Chua R. Swenson C. Tong G.C. Levi E. Noy N. Hammerling U. J. Exp. Med. 2000; 192: 835-846Crossref PubMed Scopus (67) Google Scholar), average of or of the of The zinc content measured after PMPS also of the of the protein to represent the form. To PMA and oxidation with hydrogen peroxide the or zinc two were and in The results (Fig. that agent displaced of the whereas caused zinc Our results of zinc ions in the PKC zinc The function of zinc finger structures as was described for the In were to zinc and the of the protein to the active enzyme S. U. J. Biol. Chem. 2000; 275: Full Text Full Text PDF PubMed Scopus Google Scholar, U. M. J. Biol. Chem. 2000; 275: Full Text Full Text PDF PubMed Scopus Google Scholar). the Cys of mammalian have known to as important regulation the mechanism of control not understood. We that two activation more two alternate pathways converge at this structure and the that the release of Zn2+ ions from the is of triggered by PMA or by reactive oxygen In both experiments that the inactive kinase Zn2+ with that of TSQ and oxidation or PMA binding to the Cys the of Zn2+ was allowing Zn2+ to to the Our results are in agreement with those of and L.T. Klann E. J. Biol. Chem. 2000; 275: 24136-24145Abstract Full Text Full Text PDF PubMed Scopus (167) Google Scholar) previously the of zinc release PKC activation by the redox We this concept to the pathway of PKC activation by lipid second The is that both pathways converge on the zinc finger and in the the release of Zn2+. The of oxidation are the of L.T. Klann E. J. Biol. Chem. 2000; 275: 24136-24145Abstract Full Text Full Text PDF PubMed Scopus (167) Google Scholar, S. U. J. Biol. Chem. 2000; 275: Full Text Full Text PDF PubMed Scopus Google Scholar). As high for zinc on the of three a of one of the C3H1 zinc finger and the of changes that have by for the of PKC These the of the regulatory from the of hydrophobic that to and to and A.C. 1993; PubMed Scopus Google Scholar, J. J. Biol. 1997; PubMed Scopus Google Scholar, FASEB J. PubMed Scopus Google Scholar). a be made for redox activation of PKC, the of lipid second on PKC structure is not Our Zn2+ in a linchpin role in which its to the acquisition of PMA probably in which hydrogen and G. Kazanietz M.G. Blumberg P.M. Hurley J.H. Cell. 1995; 81: 917-924Abstract Full Text PDF PubMed Scopus (595) Google Scholar). PMA the by the binding of one Zn2+ is of that PMA, in molar liberated more than 1 of the 2 zinc contained in the composite zinc finger and did mild oxidation. Because agent produced the release of of the whereas the to concentration of agent and PMA caused zinc the two to zinc ions. Altogether, the above provide new into the role of Zn2+ in signaling and particularly in the regulation of protein kinase C. We for the time that a PKC, is the source of free Zn2+ in the The two known activation lipid second messengers and mild converge on the zinc finger domain and Zn2+ far from a rigid the zinc finger domain of PKC as a The of the tetrahedrically coordinated zinc to a linchpin the to and the PKC to We for the with the and on the and and for We are to and A. at the for with acquisition and analysis and Blumberg PKC