- Email: [email protected]
Induction of protein kinase Cζ-related protein kinase by growth suppression in carcinogen-initiated epidermal cell-line WYF31 cells
Cellular Signalling 12 (2000) 15–22 http://www.elsevier.com/locate/cellsig
Research Papers
Induction of protein kinase Cz-related protein kinase by growth suppression in carcinogen-initiated epidermal cell-line WYF31 cells Kiyotaka Nishikawaa,b,*, Junken Aokic, Satoshi Yamamotoa,d, Hisamitsu Samedaa, Hiroyasu Kudoa, Haruna Nagumoa, Jian Chun Wanga, Yasuhiko Tsuzukia, Hiroyuki Araia, Keizo Inouec, Ryuichi Katoa b
a Department of Pharmacology, School of Medicine, Keio University, 35 Shinanomachi, Shinjuku-ku, Tokyo 160, Japan Department of Clinical Pharmacology, International Medical Center of Japan, Research Institute, 1-21-1 Toyama, Shinjuku-ku, Tokyo 162, Japan c Department of Pharmaceutical Science, Tokyo University, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan d Department of Clinical Pharmacology, Banyu Co. Ltd., Meguro-ku, Tokyo, Japan Received 26 May 1999; accepted 31 August 1999
Abstract In primary cultured mouse epidermal cells, protein kinase C isozyme z (PKCz) consists of multiple forms, for example, low-salt eluted PKCz (l-PKCz; 79 and 85 kDa) and high-salt eluted PKCz (h-PKCz; 79 and 85 kDa) on anion-exchange column chromatography. In this study, biochemical and biophysical differences between l-PKCz and h-PKCz were examined by using carcinogeninitiated mouse epidermal cell-line WYF31 cells, whose growth is stimulated by tumour promoter phorbol 12-myristate 13-acetate (PMA). The binding efficiency of h-PKCz to anti-PKCz antibody-affinity column was 10 times higher than that of l-PKCz. T7tagged rat PKCz overexpressed in WYF31 cells was recovered only in the high-salt eluted area on the anion-exchange column. Furthermore, when rat PKCz was stably overexpressed in WYF31 cells, the content of h-PKCz increased 4 to 5 times compared to that of parental cells, but the content of l-PKCz was not altered. All of these results indicate that h-PKCz is the product of the PKCz gene (referred to as PKCz) and that l-PKCz is closely related but different from PKCz (referred to as PKCz-related kinase). Interestingly, serum starvation of WYF31 cells caused a marked increase of the content of PKCz-related kinase with a concomitant decrease of PKCz content. These changes were reversed by stimulating the cell growth with 10% foetal calf serum. Prolonged treatment of starved cells with PMA, which induces the proliferation of WYF31 cells, also caused the downregulation of PKCz-related kinase. These results suggest that the expression levels of PKCz-related kinase and PKCz are differently regulated, and that the increased expression of PKCz-related kinase might play a significant role in the growth-suppression processes of WYF31 cells. 2000 Elsevier Science Inc. All rights reserved. Keywords: Mouse epidermal cells; WYF31 cells; Protein kinase Cz; Growth suppression; Phosphorylation; Initiated cell; Downregulation; Overexpression
1. Introduction Protein kinase C (PKC) is a family of phospholipiddependent protein kinases involved in various signal transduction processes including cell growth and differentiation events [1]. Today, 12 isozymes of PKC have Abbreviations: PKC, protein kinase C; PMA, phorbol 12-myristate 13-acetate; PS, phosphatidylserine; DG, diacylglycerol; peptide e, protein kinase C-e substrate; MEM, minimum essential medium; FCS, foetal calf serum; PMSF, phenylmethylsulphonyl fluoride; TCA, trichloroacetic acid; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; l-PKCz, low-salt eluted PKCz; h-PKCz, high-salt eluted PKCz. * Corresponding author. Tel: 181-3-3202-7181, ext. 2801; Fax: 181-3-5273-3038.
been identified and classified into three groups based on structural and enzymatic properties. Conventional isozymes (cPKC: PKCa, bI, bII, and g) are activated by Ca21, phosphatidylserine (PS), and diacylglycerol (DG)/ phorbol esters [1,2]. Novel PKCs (nPKC: PKCd, e, h, m, and u) are activated by PS and DG/phorbol esters [1–3], and atypical PKC (aPKC: PKCı, l, and z) are activated by PS only [4–6]. Since the individual isozymes have a different tissue distribution, cofactor requirement, and substrate specificity [7], each isozyme may play a specific role in signal transduction processes. PKCz, which is classified as aPKC, is expressed in almost all cell types and tissues examined [2], suggesting that PKCz plays an important role(s) in cellular func-
0898-6568/00/$ – see front matter 2000 Elsevier Science Inc. All rights reserved. PII: S0898-6568(99)00062-5
16
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
tions. It has been demonstrated that PKCz plays significant roles in mitogenic signal transduction [8–12]. On the other hand, it has been shown that overexpression of PKCz stimulates leukemic U937 cell differentiation [13] and that PKCz plays significant roles in nerve growth factor-induced differentiation of PC12 cells [14]. Furthermore, recent reports suggest that overexpression of PKCz does not induce cell transformation or tumourigenicity in NIH3T3 cells [15,16]. Presently, the biological functions of PKCz are conflicting, and the precise mechanisms by which PKCz participates in the mitogenic or differentiational signal transduction still remains to be clarified. In our previous reports, we showed that PKCa, h, and z are present in primary cultured mouse epidermal cells and that the most predominant isozyme is PKCz [17,18]. In epidermal cells, PKCz consists of multiple forms, for example, low-salt eluted PKCz (l-PKCz; 79 kDa and 85 kDa) and high-salt eluted PKCz (h-PKCz; 79 kDa and 85 kDa) on anion-exchange column chromatography [18]. Immunoprecipitated l-PKCz and h-PKCz possessed PS-dependent protein kinase activity, but neither l-PKCz nor h-PKCz were further activated by 40 nM phorbol 12-myristate 13-acetate (PMA) in the presence of PS [18]. Interestingly, l-PKCz was found to be phosphorylated and activated in intact cells with PMA treatment, whereas h-PKCz did not respond to PMA treatment [18]. These results indicate that these two types of PKCz may undergo different regulations and play different roles in intact cells. Recently, we have established carcinogen-initiated cell-line WYF31 cells, which were prepared from epidermal cells of newborn mice transplacentally initiated with 7,12-dimethylbenz[a]anthracene [19]. Interestingly, the growth of WYF31 cells is stimulated by PMA [19], indicating that WYF31 cells are useful to investigate the role of various PKC isozymes in the regulation of cell proliferation induced by tumour promoters. In the present study, using immunological and molecular biological techniques, we found that l-PKCz and h-PKCz are closely related but different proteins that may play opposite roles in WYF31 cell growth.
gated to alkaline phosphatase was from Tago Inc., Burlingame, CA, USA; control rabbit IgG was from Zymed Lab. Inc., San Francisco, CA, USA; fluorotrans membrane was from Nihon Pall Ltd., Tokyo, Japan; alkaline phosphatase substrate kit I was from Vector Laboratories, Inc., Burlingame, CA, USA; protein kinase C-e substrate peptide (peptide e, PRKRQGSVRRRV) was from Upstate Biotechnology Incorporated, Lake Placid, NY, USA; anti-T7-Tag antibody and pET-21(a) vector were from Novagen, Madison, WI, USA; and G418 and Lipofectoamine reagents were from GIBCO BRL, Grand Island, NY, USA. Anti-PKCz antiserum was raised in rabbit against synthetic oligopeptide of the sequence 577-592 (GFEYINPLLLSAEESV), as described previously [17]. 2.2. Culture of WYF31 cells WYF31 cells were cultured as described previously [19]. Briefly, WYF31 cells equivalent to two 100-mm plastic dishes were cultured in a minimum essential medium (MEM) supplemented with 10% FCS. To suppress the growth of WYF31 cells, the medium of subconfluent cells was changed to Ca21-free MEM supplemented with 10% Chelex-treated (Ca21-deprived) FCS and 50 mM CaCl2 (final concentration). After 1 day, the medium was changed to Ca21-free MEM supplemented with 0.5% Chelex-treated FCS and 50 mM CaCl2 and cultured for 2 more days. 2.3. Fractionation of WYF31 cell 105,000-g supernatants and particulate fractions using RESOURCE Q column chromatography
2. Materials and methods
WYF31 cells were homogenised at 48C in buffer A (20 mM Tris–HCl, pH 7.5, 2 mM EGTA, 2 mM EDTA, 2 mM PMSF, 10 mM sodium orthovanadate); thereafter, 105,000-g supernatants and particulate fractions were prepared as described previously [20]. The 105,000-g supernatants and particulate fractions were loaded onto an anion-exchange column, RESOURCE Q column, equilibrated with buffer B (i.e., buffer A containing 50 mM 2-mercaptoethanol and 0.02% Triton X-100). Proteins were eluted with a 40-ml linear gradient of 0 to 500 mM NaCl at a flow rate of 2.0 ml/min. Fractionation was started from 0 mM NaCl. Each fraction consisted of 2 ml of eluate.
2.1. Materials
2.4. Western blot analysis
The sources of materials used in this study were as follows: phorbol 12-myristate 13-acetate (PMA), phenylmethylsulphonyl fluoride (PMSF), and phosphatidylserine (PS) were from Sigma Chemical Co., St Louis, MO, USA; Chelex-100 was from Bio-Rad Lab., Richmond, CA, USA; foetal calf serum (FCS) was from Iansa, Mexico; RESOURCE Q column (1 ml), pSVK3 vector, and pSV2-Neo vector were from Pharmacia, Piscataway, NJ, USA; [g-32P]ATP was from ICN Biomedicals Inc., Irvine, CA, USA; antirabbit IgG conju-
Western blot analysis was done as reported previously [17]. To the fractions of RESOURCE Q column chromatography, 100% ice-cold trichloroacetic acid (TCA) was added to a final concentration of 10%. The precipitated proteins were dissolved in 20 ml of sodium dodecyl sulfate (SDS) buffer (62.5 mM Tris, 2% SDS, 5% 2-mercaptoethanol, and 5% glycerol, pH 6.8) and boiled for 5 min. This solution was subjected to 7.5% SDS-polyacrylamide gel electrophoresis (PAGE). After electrophoresis, the proteins were electrophoreti-
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
17
cally transferred to fluorotrans membranes. Anti-PKCz specific antibody or anti-T7-Tag antibody was used as the primary antibody. Antirabbit IgG or antimouse IgG conjugated to alkaline phosphatase was used as a secondary antibody. The membranes were finally visualised by the alkaline phosphatase substrate kit I.
pressing G418 resistance were screened for PKCz expression using Western blotting. Three independent groups of cells stably expressing PKCz were obtained.
2.5. Binding of l-PKCf and h-PKCf to anti-PKCf antibody-affinity column
3.1. Structural differences between l-PKCf and h-PKCf in WYF31 cells
Anti-PKCz antiserum was affinity-purified with HiTrap Protein G column (Pharmacia, Piscataway, NJ, USA). The anti-PKCz IgG fraction obtained was conjugated to Sepharose beads by using HiTrap NHS-Activated column (Pharmacia, Piscataway, NJ, USA). The l-PKCz and h-PKCz fractions, which were prepared from 105,000-g particulate fractions of WYF31 cells cultured in 10% FCS, were loaded onto control rabbit IgG column and anti-PKCz IgG column sequentially. Elution from the columns was performed by an elution buffer (0.1% Triton X-100, 50 mM glycine, pH 2.5). The bound and unbound fractions were analyzed by Western blotting using anti-PKCz antibody.
In our previous reports, PKCz was shown to consist of multiple forms, for example, low-salt eluted PKCz (lPKCz; 79 kDa and 85 kDa) and high-salt eluted PKCz (h-PKCz; 79 and 85 kDa) on anion-exchange column chromatography by Western blotting using PKCz specific antibody in primary cultured mouse epidermal cells [18]. In the present study, we examined whether or not these proteins were also present in WYF31 cells. As shown in Fig. 1, both l-PKCz and h-PKCz were detected with PKCz specific antibody both in 105,000-g supernatants and particulate fractions of WYF31 cells cultured in 10% FCS (Fig. 1A). In the fractions of RESOURCE Q chromatography obtained from 105,000-g supernatants, anti-PKCz antibody recognised a doublet protein with molecular weights of 79 and 85 kDa in fraction 1 to 5 (corresponding to l-PKCz) and fraction 8 to 12 (corresponding to h-PKCz). The contents of h-PKCz were much higher than those of l-PKCz. In the fractions of RESOURCE Q chromatography obtained from particulate fractions, anti-PKCz antibody recognised only 79-kDa protein both in low-salt eluted and high-salt eluted fractions. These results are consistent with those of epidermal cells [18]. Immunoprecipitated l-PKCz and h-PKCz possessed PS-dependent protein kinase activity (data not shown), which was also consistent with the characteristics of l-PKCz and h-PKCz in epidermal cells [18]. To examine the structural differences between l-PKCz and h-PKCz, l-PKCz and h-PKCz prepared from 105,000-g particulate fractions were applied to anti-PKCz antibody-affinity column chromatography. For the sake of the quantitative merit of l-PKCz, we used 105,000-g particulate fractions but not 105,000-g supernatant as an enzyme source. As shown in Fig. 1B, over 60% of h-PKCz bound to the anti-PKCz antibodyaffinity column. In contrast, less than 5% of l-PKCz bound to the column. Neither PKCzs bound to the control IgG column. These results indicate that both PKCzs are structurally different in intact forms, even though they have the same epitopes, which are equally detected by Western blotting using the same antibody.
2.6. Construction of PKC-expression vectors and transfections Rat brain PKCz cDNA was prepared by the polymerase chain reaction (PCR) from a rat brain cDNA library with the following primers: 59-GCGGAGGAATT CATGCCCAGCAGGACGGACCCC-39 and 59-CGTG TCGACAACAGAGATGCTCATGG-39. These primers correspond to the regions of the known rat PKCz, which contain a 59-EcoRI site and a 39-SalI site, respectively. The EcoRI-SalI fragment obtained, which contained the complete coding sequence of rat PKCz cDNA, was cloned into the EcoRI-SalI site of the pET21(a) vector [pET-21(a)-PKCz]. From this pET-21(a)PKCz, a SmaI-SalI fragment containing T7-Tagged PKCz cDNA was prepared by PCR with the following primers: 59-CAAGCCCGGGCCACCATGGCTAGC ATGACTGGTGGACAGCAAATGGGTCGC-39 and 59-TTAGCAGCCGGATCTCAGTGGTGGTGGTGG TGG-39. These primers contained a 59-SmaI site and a 39-SalI site, respectively. The fragment obtained was ligated into the SmaI-SalI site of the pSVK3 vector to produce T7-Tagged PKCz (pSVK3-T7-PKCz). The EcoRI-SalI fragment containing the PKCz cDNA was also cloned into the EcoRI-SalI site of the pSVK3 vector to produce intact PKCz (pSVK3-PKCz). For the expression of T7-Tagged PKCz, pSVK3-T7PKCz was introduced into WYF31 cells using Lipofectoamine reagents according to the manufacturer’s instructions. For the isolation of stable transformants, pSVK3-PKCz was introduced into WYF31 cells with the pSV2-Neo vector containing a neomycin-resistant gene using Lipofectoamine reagents. Transfected cells were selected in complete MEM containing 0.8 mg/ml G418. After 10 to 16 days of selection, cells stably ex-
3. Results
3.2. Distribution of T7-Tagged PKCf transiently expressed in WYF31 cells on anion-exchange column chromatography To examine whether or not both l-PKCz and h-PKCz are derived from the same gene, T7-Tagged PKCz was expressed in WYF31 cells and its distribution was ex-
18
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
Fig. 1. Comparison of the binding efficiency of l-PKCz and h-PKCz to anti-PKCz antibody-affinity column. WYF31 cells were cultured in MEM supplemented with 10% FCS. The 105,000-g supernatant and particulate fractions of the cells were obtained from two 100-mm plastic dishes and loaded onto the RESOURCE Q column. Proteins were eluted with a linear gradient of 0 to 500 mM NaCl. Fractionation was started from 0 mM NaCl. Each fraction consisted of 2 ml of eluate. Each fraction was precipitated by TCA and analyzed by Western blot using anti-PKCz antibody (A). The l-PKCz and h-PKCz fractions were prepared from 105,000-g particulate fractions of WYF31 cells cultured in 10% FCS, as described above. Each fraction was loaded onto control rabbit IgG column. The proteins that did not bind to the column were collected and loaded onto the anti-PKCz IgG column. The proteins bound to the control IgG column (lane 1) and the anti-PKCz IgG column (lane 2), and the proteins that did not bind to the anti-PKCz IgG column (lane 3) were prepared and analyzed by Western blotting using anti-PKCz antibody.
amined. In the lysate of transfected WYF31 cells, the expression of T7-Tagged PKCz was detected as 82-kDa protein on the Western blot using anti-PKCz antibody and anti-T7-Tag antibody (Fig. 2A). The 105,000-g particulate fractions of WYF31 cells transfected with the pSVK3-T7-PKCz vector were prepared and separated on RESOURCE Q column chromatography. As shown in Fig. 2B, anti-T7-Tag antibody-reactive proteins were detected only in the fractions corresponding to the position of h-PKCz but not in the low-salt region. These results suggest that h-PKCz is the product of the PKCz gene. 3.3. Comparison of the amount of l-PKCf to h-PKCf in PKCf-overexpressed cell lines derived from WYF31 cells There may still be a possibility, however, that T7Tagged PKCz cannot distribute to the low-salt region because of the presence of T7-Tag in its molecule. To elucidate this point, stable transformants of WYF31 cells in which PKCz was stably overexpressed were prepared. pSVK3-PKCz was introduced into WYF31 cells with the pSV2-Neo vector. Three independent groups of cells stably expressing PKCz were obtained. Fig. 3A shows the Western blotting analysis of PKCz expression in control WYF31 cells and in the representative PKCzoverexpressed WYF31 cells. The amounts of PKCz in the cell lysate were increased 6- to 7-fold in all PKCzoverexpressed cell lines compared to parental WYF31 cells. The 105,000-g particulate fractions of these PKCzoverexpressed cell lines and parental WYF31 cells were
prepared and separated on RESOURCE Q column chromatography. As shown in Fig. 3B, the contents of h-PKCz increased two- to three-fold in all PKCz-overexpressed cell lines compared to parental WYF31 cells. In contrast, the amounts of l-PKCz in all of the PKCzoverexpressed cell lines were almost the same as those of parental WYF31 cells. These results further support the contention that h-PKCz is the product of the PKCz gene (referred to as PKCz) and that l-PKCz is a protein closely related but different from PKCz (referred to as PKCz-related kinase). 3.4. Induction of PKCz-related kinase during serum starvation in WYF31 cells The effects of growth suppression on the amounts of PKCz-related kinase and PKCz were examined. When the growth of WYF31 cells was suppressed by treatment with 0.5% FCS for 48 h, the content of PKCzrelated kinase increased three to four times compared to the cells cultured in 10% FCS, with a concomitant decrease of PKCz content both in 105,000-g supernatant and particulate fractions (Fig. 4A). These changes were reversed by stimulating the cell growth with 10% FCS for 1 h both in 105,000-g supernatant and particulate fractions (Fig. 4B). In contrast, treatment of the growth-suppressed cells with PMA, which induces proliferation of WYF31 cells [19], caused the apparent shift of the 79-kDa to the 85-kDa form of PKCz-related kinase in particulate fractions within 1 h (Fig. 4B). Prolonged treatment with PMA induced the downregulation of PKCz-related kinase in both fractions. Under
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
19
Fig. 2. The expression of T7-tagged PKCz in WYF31 cells and its distribution on RESOURCE Q column chromatography. (A) Control vector (lane 1) or pSVK3-T7-PKCz vector (lane 2) was introduced into WYF31 cells, as described in the text. After transfection, the particulate fractions were prepared and analyzed by Western blotting using anti-PKCz antibody or anti-T7-Tag antibody. (B) WYF31 cells transfected with pSVK3-T7-PKCz vector were cultured in MEM supplemented with 10% FCS. The particulate fractions of these cells were prepared and loaded onto RESOURCE Q column, as described in the legend to Fig. 1. The fractions obtained were analyzed by Western blotting using anti-T7-Tag antibody.
the same experimental condition, PKCz in 105,000-g supernatant and particulate fractions was not affected by PMA treatment. These results suggest that the expression levels of PKCz-related kinase and PKCz are differently regulated, and that the increased expression of PKCz-related kinase and decreased expression of PKCz might play a significant role in the growth-suppression processes of WYF31 cells. 4. Discussion In the present study, we examined the biophysical and biochemical differences between l-PKCz and h-PKCz using carcinogen-initiated cell-line WYF31 cells. l-PKCz and h-PKCz were detected in WYF31 cells
Fig. 3. Establishment of PKCz-overexpressed cell lines from WYF31 cells and comparison of the content of l-PKCz and h-PKCz. pSVK3PKCz was introduced into WYF31 cells with the pSV2-Neo vector using Lipofectoamine reagents. Transfected cells were selected in complete MEM containing 0.8 mg/ml G418. The cells stably expressing G418 resistance were screened for PKCz expression using Western blotting. Three independent groups of cells stably expressing PKCz were obtained (WYF31-P14, -P11, and -P20). Western blotting analysis of PKCz expression in the cell lysate of WYF31 cells and the PKCz-overexpressed cell lines was performed (A). The 105,000-g particulate fractions of WYF31 cells and these PKCz-overexpressed cell lines were prepared and separated on RESOURCE Q column chromatography, as described in the legend to Fig. 1. The fractions obtained were analyzed by Western blotting using anti-PKCz antibody. The amounts (OD 3 mm2) of anti-PKCz antibody reactive proteins in each fraction were measured by densitometry. Open circle, closed square, closed triangle, and closed circle indicate parental WYF31 cells, WYF31-P14, WYF31-P11, and WYF31-P20, respectively (B).
20
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
Fig. 4. Effect of growth suppression and subsequent treatment with 10% FCS or 100 nM PMA on the expression of l-PKCz and h-PKCz in 105,000-g supernatant and particulate fractions of WYF31 cells. WYF31 cells were cultured in MEM supplemented with 10% FCS. To suppress the growth of WYF31 cells, the medium of subconfluent cells was changed to Ca21-free MEM supplemented with 10% Ca21-deprived FCS and 50 mM CaCl2. After 1 day, the medium was changed to Ca21-free MEM supplemented with 0.5% Ca21-deprived FCS and 50 mM CaCl2 and cultured for 2 more days. Control WYF31 cells were cultured in MEM supplemented with 10% FCS. The 105,000-g supernatant and the particulate fractions of these cells were prepared and separated on RESOURCE Q column, as described in the legend to Fig. 1 (A). The medium of the growth-suppressed cells was replaced with MEM supplemented with 10% FCS (10% FCS) or the cells were stimulated by 100 nM PMA (PMA). After the indicated period (1 or 48 h), the 105,000-g supernatant and particulate fractions of these cells were prepared and subjected to RESOURCE Q column, as described in the legend to Fig. 1. Fraction no. 1 to 5 (l-PKCz fraction) and fraction no. 8 to 12 (h-PKCz fraction) were respectively mixed and analyzed by Western blot using anti-PKCz antibody (B).
with PKCz-specific antibody both in 105,000-g supernatant and particulate fractions, consistent with the results of parental primary cultured mouse epidermal cells. The binding efficiency of h-PKCz to the anti-PKCz antibody-affinity column was 10 times higher than that of l-PKCz, indicating that both l-PKCz and h-PKCz are structurally different in intact forms. Furthermore, both transiently expressed T7-tagged PKCz and stably transfected PKCz were detected only in the fractions corresponding to the position of h-PKCz on RESOURCE Q column chromatography. This suggests that h-PKCz is the product of the PKCz gene (referred to as PKCz) and that l-PKCz is a closely related but different protein from PKCz (referred to as PKCz-related kinase). We observed that transfected PKCzs in other types of cells, such as SF9 insect cells and mouse L-cells, were also expressed as h-PKCz (data not shown), consistent with the above contention. In our previous study it was demonstrated that immunoprecipitated PKCz-related kinase and PKCz
phosphorylated synthetic peptide derived from PKCe pseudosubstrate in a PS-dependent manner, but the cofactor dependency on autophosphorylation is different between these two PKCzs [18]. Autophosphorylation of PKCz-related kinase is not observed in the absence of PS but is markedly stimulated by PS, whereas autophosphorylation of PKCz is substantially detected even in the absence of PS [18]. This cofactor dependency of the autophosphorylation of PKCz is consistent with that of the purified PKCz from bovine kidney [21]. We observed the same characterisation for PKCz-related kinase and PKCz present in WYF31 cells (data not shown). These facts also suggest that PKCz-related kinase and PKCz are different kinases. It has been reported that anti-PKCz antibody raised against the C-terminal sequence of PKCz cross-reacts with other PKC isozymes, such as PKCa, b, g [22], ı [4], and l [5]. In primary cultured mouse epidermal cells and WYF31 cells, however, PKCb and g were not detected by Western blotting with any of the isozyme-spe-
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
cific antibodies [17]. PKCa was only present in 105,000-g supernatant fractions of the cell lysates, whereas PKCz-related kinase was present both in 105,000-g supernatant and particulate fractions [17]. Furthermore, the anti-PKCz antibody used in this study, which was also raised against the C-terminal sequence of PKCz, did not cross-react with PKCa (data not shown). These facts demonstrate that PKCz-related kinase is not derived from cPKCs. The molecular weights of PKCı [4] and PKCl [5] are 65 and 74 kDa, respectively. These weights are different from those of PKCz-related kinase (79 and 85 kDa). Furthermore, it has been reported that autophosphorylation of PKCl is observed in the absence of PS [5]. Therefore, it seems unlikely that PKCz-related kinase is one of these isozymes. At present, there may still be a possibility that PKCz-related kinase is derived from PKCz gene by alternative splicing, even though the genomic structure of PKCz gene has not been fully understood. Since the anti-PKCz antibody used in this study is a polyclonal antibody, it is possible that this antibody detects an alternative spliced form of PKCz and a highly homologous protein to PKCz as well as PKCz itself. To elucidate whether PKCz-related kinase is derived from PKCz gene or not, further studies must be performed. When WYF31 cells were cultured in the presence of 0.5% FCS for 48 h, the content of PKCz-related kinase increased three to four times compared to the cells cultured in 10% FCS, with a concomitant decrease of PKCz content, both in 105,000-g supernatant and in particulate fractions. We also found that the total PSdependent kinase activity of PKCz-related kinase obtained from serum-starved cells increased two- to threefold compared to that of the cells cultured in 10% FCS (data not shown). These results suggest that the increase of the expression of PKCz-related kinase might play a role in the growth-suppressive signals in WYF31 cells. In our previous study using primary cultured mouse epidermal cells, transforming growth factor b (TGFb), which has a potent growth-suppressive effect on this type of cell, strongly suppressed the Ca21-dependent PKC activity but slightly increased the Ca21-independent PKC activity both in 105,000-g supernatants and particulate fractions [23]. This Ca21-independent PKC activity primarily reflects PKCz activity [17]. It is possible that PKCz-related kinase may play a role in the common processes of growth-suppressive signals induced by serum starvation or TGFb treatment. The increase of PKCz-related kinase content and the concomitant decrease of PKCz content induced by growth suppression were reversed by treating the cells with 10% FCS. In contrast, treatment of the growthsuppressed cells with PMA, which is known to induce the proliferation of WYF31 cells [19], caused the apparent shift of 79-kDa PKCz-related kinase to the 85-kDa form in particulate fractions within 1 h. Prolonged treat-
21
ment with PMA induced the downregulation of PKCzrelated kinase, but PKCz was not affected by PMA treatment. These results suggest that the continuous presence of PKCz in the 105,000-g supernatants and the decrease of the amount of PKCz-related kinase both in 105,000-g supernatants and particulate fractions may be important in the growth-stimulating signals induced by either FCS or PMA in WYF31 cells. We have already found that the apparent shift of 79-kDa PKCz-related kinase to the 85-kDa form in particulate fractions by PMA treatment results at least in part from the phosphorylation of 79-kDa PKCz-related kinase in primary cultured mouse epidermal cells [18]. The total activity of PKCz-related kinase increases in association with the apparent shift from the 75- to 85-kDa form [18]. Therefore, it seems possible that the transient activation of PKCz-related kinase in particulate fractions in association with the apparent band shift plays a significant role in the early stage of the PMA-caused growth stimulation of WYF31 cells. In conclusion, we showed that PKCz-related kinase and PKCz are closely related but different proteins, and that the expression of PKCz-related kinase and PKCz are regulated differently in WYF31 cells. The exact mechanism responsible for the induction of PKCzrelated kinase and the reduction of PKCz during cell growth suppression has yet to be elucidated. To elucidate the independent signal transduction processes of PKCz-related kinase and PKCz, it is necessary to determine their specific endogenous substrate proteins. For this purpose, the determination of their specific substrate motifs using degenerate peptide library techniques [7] is now underway.
Acknowledgments We thank Mr. Kouichi Maruyama for his competent technical assistance. This work was supported in part by a general grant from the Japanese Ministry of Education, Science, and Culture, and by a grant from Keio Gijuku Academic Development Funds, Keio University, Tokyo, Japan.
References [1] Nishizuka Y. Science 1992;258:607–14. [2] Hug H, Sarre TF. Biochem J 1993;291:329–43. [3] Johannes FJ, Prestle J, Eis S, Oberhagemann P, Pfizenmaier K. J Biol Chem 1994;269:6140–8. [4] Selbie LA, Schmitzpeiffer C, Sheng YH, Biden TJ. J Biol Chem 1993;268:24296–320. [5] Akimoto K, Mizuno K, Osada S, Hirai S, Tanuma S, Suzuki K, Ohno S. J Biol Chem 1994;269:12677–83. [6] Dekker LV, Parker PJ. Trends Biochem Sci 1994;19:73–7. [7] Nishikawa K, Toker A, Johannes F-J, Songyang Z, Cantley LC. J Biol Chem 1997;272:952–60.
22
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
[8] Berra E, Diazmeco MT, Dominguez I, Municio MM, Sanz L, Lozano J, Chapkin RS, Moscat J. Cell 1993;74:555–63. [9] Diazmeco MT, Berra E, Municio MM, Sanz L, Lozano J, Dominguez I, Diazgolpe V, Delera MTL, Alcami J, Paya CV, Arenzanaseisdedos F, Virelizier JL, Moscat J. Mol Cell Biol 1993; 13:4770–5. [10] Dominguez I, Sanz L, Arenzanaseisdedos F, Diazmeco MT, Virelizier JL, Moscat J. Mol Cell Biol 1993;13:1290–5. [11] Diazmeco MT, Lozano J, Municio MM, Berra E, Frutos S, Sanz L, Moscat J. Journal of Biological Chemistry 1994;269:31706–10. [12] Lozano J, Berra E, Municio MM, Diazmeco MT, Dominguez I, Sanz L, Moscat J. J Biol Chem 1994;269:19200–2. [13] Ways DK, Posekany K, Devente J, Garris T, Chen JM, Hooker J, Qin WX, Cook P, Fletcher D, Parker P. Cell Growth Differentiation 1994;5:1195–203. [14] Wooten MW, Zhou GS, Seibenhener ML, Coleman ES. Cell Growth & Differentiation 1994;5:395–403.
[15] Crespo P, Mischak H, Gutkind JS. Biochem Biophys Res Commun 1995;213:266–72. [16] Montaner S, Ramos A, Perona R, Esteve P, Carnero A, Lacal JC. Oncogene 1995;10:2213–20. [17] Nishikawa K, Yamamoto S, Nagumo H, Ishihara M, Murayama N, Kato R. Cell Signal 1994;6:503–12. [18] Nishikawa K, Yamamoto S, Nagumo H, Maruyama K, Kato R. Cell Signal 1995;7:491–504. [19] Wang JC, Yamamoto S, Kato R. Carcinogenesis 1992;13:1301–5. [20] Nishikawa K, Yamamoto S, Otsuka C, Kato R. Cell Signal 1992; 4:757–76. [21] Nakanishi H, Exton JH. J Biol Chem 1992;267:16347–54. [22] Allen BG, Andrea JE, Walsh MP. J Biol Chem 1994;269: 29288–98. [23] Nishikawa K, Yamamoto S, Nagumo H, Otsuka C, Kato R. Biochem Biophys Res Commun 1993;193:384–9.
Research Papers
Induction of protein kinase Cz-related protein kinase by growth suppression in carcinogen-initiated epidermal cell-line WYF31 cells Kiyotaka Nishikawaa,b,*, Junken Aokic, Satoshi Yamamotoa,d, Hisamitsu Samedaa, Hiroyasu Kudoa, Haruna Nagumoa, Jian Chun Wanga, Yasuhiko Tsuzukia, Hiroyuki Araia, Keizo Inouec, Ryuichi Katoa b
a Department of Pharmacology, School of Medicine, Keio University, 35 Shinanomachi, Shinjuku-ku, Tokyo 160, Japan Department of Clinical Pharmacology, International Medical Center of Japan, Research Institute, 1-21-1 Toyama, Shinjuku-ku, Tokyo 162, Japan c Department of Pharmaceutical Science, Tokyo University, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan d Department of Clinical Pharmacology, Banyu Co. Ltd., Meguro-ku, Tokyo, Japan Received 26 May 1999; accepted 31 August 1999
Abstract In primary cultured mouse epidermal cells, protein kinase C isozyme z (PKCz) consists of multiple forms, for example, low-salt eluted PKCz (l-PKCz; 79 and 85 kDa) and high-salt eluted PKCz (h-PKCz; 79 and 85 kDa) on anion-exchange column chromatography. In this study, biochemical and biophysical differences between l-PKCz and h-PKCz were examined by using carcinogeninitiated mouse epidermal cell-line WYF31 cells, whose growth is stimulated by tumour promoter phorbol 12-myristate 13-acetate (PMA). The binding efficiency of h-PKCz to anti-PKCz antibody-affinity column was 10 times higher than that of l-PKCz. T7tagged rat PKCz overexpressed in WYF31 cells was recovered only in the high-salt eluted area on the anion-exchange column. Furthermore, when rat PKCz was stably overexpressed in WYF31 cells, the content of h-PKCz increased 4 to 5 times compared to that of parental cells, but the content of l-PKCz was not altered. All of these results indicate that h-PKCz is the product of the PKCz gene (referred to as PKCz) and that l-PKCz is closely related but different from PKCz (referred to as PKCz-related kinase). Interestingly, serum starvation of WYF31 cells caused a marked increase of the content of PKCz-related kinase with a concomitant decrease of PKCz content. These changes were reversed by stimulating the cell growth with 10% foetal calf serum. Prolonged treatment of starved cells with PMA, which induces the proliferation of WYF31 cells, also caused the downregulation of PKCz-related kinase. These results suggest that the expression levels of PKCz-related kinase and PKCz are differently regulated, and that the increased expression of PKCz-related kinase might play a significant role in the growth-suppression processes of WYF31 cells. 2000 Elsevier Science Inc. All rights reserved. Keywords: Mouse epidermal cells; WYF31 cells; Protein kinase Cz; Growth suppression; Phosphorylation; Initiated cell; Downregulation; Overexpression
1. Introduction Protein kinase C (PKC) is a family of phospholipiddependent protein kinases involved in various signal transduction processes including cell growth and differentiation events [1]. Today, 12 isozymes of PKC have Abbreviations: PKC, protein kinase C; PMA, phorbol 12-myristate 13-acetate; PS, phosphatidylserine; DG, diacylglycerol; peptide e, protein kinase C-e substrate; MEM, minimum essential medium; FCS, foetal calf serum; PMSF, phenylmethylsulphonyl fluoride; TCA, trichloroacetic acid; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; l-PKCz, low-salt eluted PKCz; h-PKCz, high-salt eluted PKCz. * Corresponding author. Tel: 181-3-3202-7181, ext. 2801; Fax: 181-3-5273-3038.
been identified and classified into three groups based on structural and enzymatic properties. Conventional isozymes (cPKC: PKCa, bI, bII, and g) are activated by Ca21, phosphatidylserine (PS), and diacylglycerol (DG)/ phorbol esters [1,2]. Novel PKCs (nPKC: PKCd, e, h, m, and u) are activated by PS and DG/phorbol esters [1–3], and atypical PKC (aPKC: PKCı, l, and z) are activated by PS only [4–6]. Since the individual isozymes have a different tissue distribution, cofactor requirement, and substrate specificity [7], each isozyme may play a specific role in signal transduction processes. PKCz, which is classified as aPKC, is expressed in almost all cell types and tissues examined [2], suggesting that PKCz plays an important role(s) in cellular func-
0898-6568/00/$ – see front matter 2000 Elsevier Science Inc. All rights reserved. PII: S0898-6568(99)00062-5
16
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
tions. It has been demonstrated that PKCz plays significant roles in mitogenic signal transduction [8–12]. On the other hand, it has been shown that overexpression of PKCz stimulates leukemic U937 cell differentiation [13] and that PKCz plays significant roles in nerve growth factor-induced differentiation of PC12 cells [14]. Furthermore, recent reports suggest that overexpression of PKCz does not induce cell transformation or tumourigenicity in NIH3T3 cells [15,16]. Presently, the biological functions of PKCz are conflicting, and the precise mechanisms by which PKCz participates in the mitogenic or differentiational signal transduction still remains to be clarified. In our previous reports, we showed that PKCa, h, and z are present in primary cultured mouse epidermal cells and that the most predominant isozyme is PKCz [17,18]. In epidermal cells, PKCz consists of multiple forms, for example, low-salt eluted PKCz (l-PKCz; 79 kDa and 85 kDa) and high-salt eluted PKCz (h-PKCz; 79 kDa and 85 kDa) on anion-exchange column chromatography [18]. Immunoprecipitated l-PKCz and h-PKCz possessed PS-dependent protein kinase activity, but neither l-PKCz nor h-PKCz were further activated by 40 nM phorbol 12-myristate 13-acetate (PMA) in the presence of PS [18]. Interestingly, l-PKCz was found to be phosphorylated and activated in intact cells with PMA treatment, whereas h-PKCz did not respond to PMA treatment [18]. These results indicate that these two types of PKCz may undergo different regulations and play different roles in intact cells. Recently, we have established carcinogen-initiated cell-line WYF31 cells, which were prepared from epidermal cells of newborn mice transplacentally initiated with 7,12-dimethylbenz[a]anthracene [19]. Interestingly, the growth of WYF31 cells is stimulated by PMA [19], indicating that WYF31 cells are useful to investigate the role of various PKC isozymes in the regulation of cell proliferation induced by tumour promoters. In the present study, using immunological and molecular biological techniques, we found that l-PKCz and h-PKCz are closely related but different proteins that may play opposite roles in WYF31 cell growth.
gated to alkaline phosphatase was from Tago Inc., Burlingame, CA, USA; control rabbit IgG was from Zymed Lab. Inc., San Francisco, CA, USA; fluorotrans membrane was from Nihon Pall Ltd., Tokyo, Japan; alkaline phosphatase substrate kit I was from Vector Laboratories, Inc., Burlingame, CA, USA; protein kinase C-e substrate peptide (peptide e, PRKRQGSVRRRV) was from Upstate Biotechnology Incorporated, Lake Placid, NY, USA; anti-T7-Tag antibody and pET-21(a) vector were from Novagen, Madison, WI, USA; and G418 and Lipofectoamine reagents were from GIBCO BRL, Grand Island, NY, USA. Anti-PKCz antiserum was raised in rabbit against synthetic oligopeptide of the sequence 577-592 (GFEYINPLLLSAEESV), as described previously [17]. 2.2. Culture of WYF31 cells WYF31 cells were cultured as described previously [19]. Briefly, WYF31 cells equivalent to two 100-mm plastic dishes were cultured in a minimum essential medium (MEM) supplemented with 10% FCS. To suppress the growth of WYF31 cells, the medium of subconfluent cells was changed to Ca21-free MEM supplemented with 10% Chelex-treated (Ca21-deprived) FCS and 50 mM CaCl2 (final concentration). After 1 day, the medium was changed to Ca21-free MEM supplemented with 0.5% Chelex-treated FCS and 50 mM CaCl2 and cultured for 2 more days. 2.3. Fractionation of WYF31 cell 105,000-g supernatants and particulate fractions using RESOURCE Q column chromatography
2. Materials and methods
WYF31 cells were homogenised at 48C in buffer A (20 mM Tris–HCl, pH 7.5, 2 mM EGTA, 2 mM EDTA, 2 mM PMSF, 10 mM sodium orthovanadate); thereafter, 105,000-g supernatants and particulate fractions were prepared as described previously [20]. The 105,000-g supernatants and particulate fractions were loaded onto an anion-exchange column, RESOURCE Q column, equilibrated with buffer B (i.e., buffer A containing 50 mM 2-mercaptoethanol and 0.02% Triton X-100). Proteins were eluted with a 40-ml linear gradient of 0 to 500 mM NaCl at a flow rate of 2.0 ml/min. Fractionation was started from 0 mM NaCl. Each fraction consisted of 2 ml of eluate.
2.1. Materials
2.4. Western blot analysis
The sources of materials used in this study were as follows: phorbol 12-myristate 13-acetate (PMA), phenylmethylsulphonyl fluoride (PMSF), and phosphatidylserine (PS) were from Sigma Chemical Co., St Louis, MO, USA; Chelex-100 was from Bio-Rad Lab., Richmond, CA, USA; foetal calf serum (FCS) was from Iansa, Mexico; RESOURCE Q column (1 ml), pSVK3 vector, and pSV2-Neo vector were from Pharmacia, Piscataway, NJ, USA; [g-32P]ATP was from ICN Biomedicals Inc., Irvine, CA, USA; antirabbit IgG conju-
Western blot analysis was done as reported previously [17]. To the fractions of RESOURCE Q column chromatography, 100% ice-cold trichloroacetic acid (TCA) was added to a final concentration of 10%. The precipitated proteins were dissolved in 20 ml of sodium dodecyl sulfate (SDS) buffer (62.5 mM Tris, 2% SDS, 5% 2-mercaptoethanol, and 5% glycerol, pH 6.8) and boiled for 5 min. This solution was subjected to 7.5% SDS-polyacrylamide gel electrophoresis (PAGE). After electrophoresis, the proteins were electrophoreti-
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
17
cally transferred to fluorotrans membranes. Anti-PKCz specific antibody or anti-T7-Tag antibody was used as the primary antibody. Antirabbit IgG or antimouse IgG conjugated to alkaline phosphatase was used as a secondary antibody. The membranes were finally visualised by the alkaline phosphatase substrate kit I.
pressing G418 resistance were screened for PKCz expression using Western blotting. Three independent groups of cells stably expressing PKCz were obtained.
2.5. Binding of l-PKCf and h-PKCf to anti-PKCf antibody-affinity column
3.1. Structural differences between l-PKCf and h-PKCf in WYF31 cells
Anti-PKCz antiserum was affinity-purified with HiTrap Protein G column (Pharmacia, Piscataway, NJ, USA). The anti-PKCz IgG fraction obtained was conjugated to Sepharose beads by using HiTrap NHS-Activated column (Pharmacia, Piscataway, NJ, USA). The l-PKCz and h-PKCz fractions, which were prepared from 105,000-g particulate fractions of WYF31 cells cultured in 10% FCS, were loaded onto control rabbit IgG column and anti-PKCz IgG column sequentially. Elution from the columns was performed by an elution buffer (0.1% Triton X-100, 50 mM glycine, pH 2.5). The bound and unbound fractions were analyzed by Western blotting using anti-PKCz antibody.
In our previous reports, PKCz was shown to consist of multiple forms, for example, low-salt eluted PKCz (lPKCz; 79 kDa and 85 kDa) and high-salt eluted PKCz (h-PKCz; 79 and 85 kDa) on anion-exchange column chromatography by Western blotting using PKCz specific antibody in primary cultured mouse epidermal cells [18]. In the present study, we examined whether or not these proteins were also present in WYF31 cells. As shown in Fig. 1, both l-PKCz and h-PKCz were detected with PKCz specific antibody both in 105,000-g supernatants and particulate fractions of WYF31 cells cultured in 10% FCS (Fig. 1A). In the fractions of RESOURCE Q chromatography obtained from 105,000-g supernatants, anti-PKCz antibody recognised a doublet protein with molecular weights of 79 and 85 kDa in fraction 1 to 5 (corresponding to l-PKCz) and fraction 8 to 12 (corresponding to h-PKCz). The contents of h-PKCz were much higher than those of l-PKCz. In the fractions of RESOURCE Q chromatography obtained from particulate fractions, anti-PKCz antibody recognised only 79-kDa protein both in low-salt eluted and high-salt eluted fractions. These results are consistent with those of epidermal cells [18]. Immunoprecipitated l-PKCz and h-PKCz possessed PS-dependent protein kinase activity (data not shown), which was also consistent with the characteristics of l-PKCz and h-PKCz in epidermal cells [18]. To examine the structural differences between l-PKCz and h-PKCz, l-PKCz and h-PKCz prepared from 105,000-g particulate fractions were applied to anti-PKCz antibody-affinity column chromatography. For the sake of the quantitative merit of l-PKCz, we used 105,000-g particulate fractions but not 105,000-g supernatant as an enzyme source. As shown in Fig. 1B, over 60% of h-PKCz bound to the anti-PKCz antibodyaffinity column. In contrast, less than 5% of l-PKCz bound to the column. Neither PKCzs bound to the control IgG column. These results indicate that both PKCzs are structurally different in intact forms, even though they have the same epitopes, which are equally detected by Western blotting using the same antibody.
2.6. Construction of PKC-expression vectors and transfections Rat brain PKCz cDNA was prepared by the polymerase chain reaction (PCR) from a rat brain cDNA library with the following primers: 59-GCGGAGGAATT CATGCCCAGCAGGACGGACCCC-39 and 59-CGTG TCGACAACAGAGATGCTCATGG-39. These primers correspond to the regions of the known rat PKCz, which contain a 59-EcoRI site and a 39-SalI site, respectively. The EcoRI-SalI fragment obtained, which contained the complete coding sequence of rat PKCz cDNA, was cloned into the EcoRI-SalI site of the pET21(a) vector [pET-21(a)-PKCz]. From this pET-21(a)PKCz, a SmaI-SalI fragment containing T7-Tagged PKCz cDNA was prepared by PCR with the following primers: 59-CAAGCCCGGGCCACCATGGCTAGC ATGACTGGTGGACAGCAAATGGGTCGC-39 and 59-TTAGCAGCCGGATCTCAGTGGTGGTGGTGG TGG-39. These primers contained a 59-SmaI site and a 39-SalI site, respectively. The fragment obtained was ligated into the SmaI-SalI site of the pSVK3 vector to produce T7-Tagged PKCz (pSVK3-T7-PKCz). The EcoRI-SalI fragment containing the PKCz cDNA was also cloned into the EcoRI-SalI site of the pSVK3 vector to produce intact PKCz (pSVK3-PKCz). For the expression of T7-Tagged PKCz, pSVK3-T7PKCz was introduced into WYF31 cells using Lipofectoamine reagents according to the manufacturer’s instructions. For the isolation of stable transformants, pSVK3-PKCz was introduced into WYF31 cells with the pSV2-Neo vector containing a neomycin-resistant gene using Lipofectoamine reagents. Transfected cells were selected in complete MEM containing 0.8 mg/ml G418. After 10 to 16 days of selection, cells stably ex-
3. Results
3.2. Distribution of T7-Tagged PKCf transiently expressed in WYF31 cells on anion-exchange column chromatography To examine whether or not both l-PKCz and h-PKCz are derived from the same gene, T7-Tagged PKCz was expressed in WYF31 cells and its distribution was ex-
18
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
Fig. 1. Comparison of the binding efficiency of l-PKCz and h-PKCz to anti-PKCz antibody-affinity column. WYF31 cells were cultured in MEM supplemented with 10% FCS. The 105,000-g supernatant and particulate fractions of the cells were obtained from two 100-mm plastic dishes and loaded onto the RESOURCE Q column. Proteins were eluted with a linear gradient of 0 to 500 mM NaCl. Fractionation was started from 0 mM NaCl. Each fraction consisted of 2 ml of eluate. Each fraction was precipitated by TCA and analyzed by Western blot using anti-PKCz antibody (A). The l-PKCz and h-PKCz fractions were prepared from 105,000-g particulate fractions of WYF31 cells cultured in 10% FCS, as described above. Each fraction was loaded onto control rabbit IgG column. The proteins that did not bind to the column were collected and loaded onto the anti-PKCz IgG column. The proteins bound to the control IgG column (lane 1) and the anti-PKCz IgG column (lane 2), and the proteins that did not bind to the anti-PKCz IgG column (lane 3) were prepared and analyzed by Western blotting using anti-PKCz antibody.
amined. In the lysate of transfected WYF31 cells, the expression of T7-Tagged PKCz was detected as 82-kDa protein on the Western blot using anti-PKCz antibody and anti-T7-Tag antibody (Fig. 2A). The 105,000-g particulate fractions of WYF31 cells transfected with the pSVK3-T7-PKCz vector were prepared and separated on RESOURCE Q column chromatography. As shown in Fig. 2B, anti-T7-Tag antibody-reactive proteins were detected only in the fractions corresponding to the position of h-PKCz but not in the low-salt region. These results suggest that h-PKCz is the product of the PKCz gene. 3.3. Comparison of the amount of l-PKCf to h-PKCf in PKCf-overexpressed cell lines derived from WYF31 cells There may still be a possibility, however, that T7Tagged PKCz cannot distribute to the low-salt region because of the presence of T7-Tag in its molecule. To elucidate this point, stable transformants of WYF31 cells in which PKCz was stably overexpressed were prepared. pSVK3-PKCz was introduced into WYF31 cells with the pSV2-Neo vector. Three independent groups of cells stably expressing PKCz were obtained. Fig. 3A shows the Western blotting analysis of PKCz expression in control WYF31 cells and in the representative PKCzoverexpressed WYF31 cells. The amounts of PKCz in the cell lysate were increased 6- to 7-fold in all PKCzoverexpressed cell lines compared to parental WYF31 cells. The 105,000-g particulate fractions of these PKCzoverexpressed cell lines and parental WYF31 cells were
prepared and separated on RESOURCE Q column chromatography. As shown in Fig. 3B, the contents of h-PKCz increased two- to three-fold in all PKCz-overexpressed cell lines compared to parental WYF31 cells. In contrast, the amounts of l-PKCz in all of the PKCzoverexpressed cell lines were almost the same as those of parental WYF31 cells. These results further support the contention that h-PKCz is the product of the PKCz gene (referred to as PKCz) and that l-PKCz is a protein closely related but different from PKCz (referred to as PKCz-related kinase). 3.4. Induction of PKCz-related kinase during serum starvation in WYF31 cells The effects of growth suppression on the amounts of PKCz-related kinase and PKCz were examined. When the growth of WYF31 cells was suppressed by treatment with 0.5% FCS for 48 h, the content of PKCzrelated kinase increased three to four times compared to the cells cultured in 10% FCS, with a concomitant decrease of PKCz content both in 105,000-g supernatant and particulate fractions (Fig. 4A). These changes were reversed by stimulating the cell growth with 10% FCS for 1 h both in 105,000-g supernatant and particulate fractions (Fig. 4B). In contrast, treatment of the growth-suppressed cells with PMA, which induces proliferation of WYF31 cells [19], caused the apparent shift of the 79-kDa to the 85-kDa form of PKCz-related kinase in particulate fractions within 1 h (Fig. 4B). Prolonged treatment with PMA induced the downregulation of PKCz-related kinase in both fractions. Under
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
19
Fig. 2. The expression of T7-tagged PKCz in WYF31 cells and its distribution on RESOURCE Q column chromatography. (A) Control vector (lane 1) or pSVK3-T7-PKCz vector (lane 2) was introduced into WYF31 cells, as described in the text. After transfection, the particulate fractions were prepared and analyzed by Western blotting using anti-PKCz antibody or anti-T7-Tag antibody. (B) WYF31 cells transfected with pSVK3-T7-PKCz vector were cultured in MEM supplemented with 10% FCS. The particulate fractions of these cells were prepared and loaded onto RESOURCE Q column, as described in the legend to Fig. 1. The fractions obtained were analyzed by Western blotting using anti-T7-Tag antibody.
the same experimental condition, PKCz in 105,000-g supernatant and particulate fractions was not affected by PMA treatment. These results suggest that the expression levels of PKCz-related kinase and PKCz are differently regulated, and that the increased expression of PKCz-related kinase and decreased expression of PKCz might play a significant role in the growth-suppression processes of WYF31 cells. 4. Discussion In the present study, we examined the biophysical and biochemical differences between l-PKCz and h-PKCz using carcinogen-initiated cell-line WYF31 cells. l-PKCz and h-PKCz were detected in WYF31 cells
Fig. 3. Establishment of PKCz-overexpressed cell lines from WYF31 cells and comparison of the content of l-PKCz and h-PKCz. pSVK3PKCz was introduced into WYF31 cells with the pSV2-Neo vector using Lipofectoamine reagents. Transfected cells were selected in complete MEM containing 0.8 mg/ml G418. The cells stably expressing G418 resistance were screened for PKCz expression using Western blotting. Three independent groups of cells stably expressing PKCz were obtained (WYF31-P14, -P11, and -P20). Western blotting analysis of PKCz expression in the cell lysate of WYF31 cells and the PKCz-overexpressed cell lines was performed (A). The 105,000-g particulate fractions of WYF31 cells and these PKCz-overexpressed cell lines were prepared and separated on RESOURCE Q column chromatography, as described in the legend to Fig. 1. The fractions obtained were analyzed by Western blotting using anti-PKCz antibody. The amounts (OD 3 mm2) of anti-PKCz antibody reactive proteins in each fraction were measured by densitometry. Open circle, closed square, closed triangle, and closed circle indicate parental WYF31 cells, WYF31-P14, WYF31-P11, and WYF31-P20, respectively (B).
20
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
Fig. 4. Effect of growth suppression and subsequent treatment with 10% FCS or 100 nM PMA on the expression of l-PKCz and h-PKCz in 105,000-g supernatant and particulate fractions of WYF31 cells. WYF31 cells were cultured in MEM supplemented with 10% FCS. To suppress the growth of WYF31 cells, the medium of subconfluent cells was changed to Ca21-free MEM supplemented with 10% Ca21-deprived FCS and 50 mM CaCl2. After 1 day, the medium was changed to Ca21-free MEM supplemented with 0.5% Ca21-deprived FCS and 50 mM CaCl2 and cultured for 2 more days. Control WYF31 cells were cultured in MEM supplemented with 10% FCS. The 105,000-g supernatant and the particulate fractions of these cells were prepared and separated on RESOURCE Q column, as described in the legend to Fig. 1 (A). The medium of the growth-suppressed cells was replaced with MEM supplemented with 10% FCS (10% FCS) or the cells were stimulated by 100 nM PMA (PMA). After the indicated period (1 or 48 h), the 105,000-g supernatant and particulate fractions of these cells were prepared and subjected to RESOURCE Q column, as described in the legend to Fig. 1. Fraction no. 1 to 5 (l-PKCz fraction) and fraction no. 8 to 12 (h-PKCz fraction) were respectively mixed and analyzed by Western blot using anti-PKCz antibody (B).
with PKCz-specific antibody both in 105,000-g supernatant and particulate fractions, consistent with the results of parental primary cultured mouse epidermal cells. The binding efficiency of h-PKCz to the anti-PKCz antibody-affinity column was 10 times higher than that of l-PKCz, indicating that both l-PKCz and h-PKCz are structurally different in intact forms. Furthermore, both transiently expressed T7-tagged PKCz and stably transfected PKCz were detected only in the fractions corresponding to the position of h-PKCz on RESOURCE Q column chromatography. This suggests that h-PKCz is the product of the PKCz gene (referred to as PKCz) and that l-PKCz is a closely related but different protein from PKCz (referred to as PKCz-related kinase). We observed that transfected PKCzs in other types of cells, such as SF9 insect cells and mouse L-cells, were also expressed as h-PKCz (data not shown), consistent with the above contention. In our previous study it was demonstrated that immunoprecipitated PKCz-related kinase and PKCz
phosphorylated synthetic peptide derived from PKCe pseudosubstrate in a PS-dependent manner, but the cofactor dependency on autophosphorylation is different between these two PKCzs [18]. Autophosphorylation of PKCz-related kinase is not observed in the absence of PS but is markedly stimulated by PS, whereas autophosphorylation of PKCz is substantially detected even in the absence of PS [18]. This cofactor dependency of the autophosphorylation of PKCz is consistent with that of the purified PKCz from bovine kidney [21]. We observed the same characterisation for PKCz-related kinase and PKCz present in WYF31 cells (data not shown). These facts also suggest that PKCz-related kinase and PKCz are different kinases. It has been reported that anti-PKCz antibody raised against the C-terminal sequence of PKCz cross-reacts with other PKC isozymes, such as PKCa, b, g [22], ı [4], and l [5]. In primary cultured mouse epidermal cells and WYF31 cells, however, PKCb and g were not detected by Western blotting with any of the isozyme-spe-
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
cific antibodies [17]. PKCa was only present in 105,000-g supernatant fractions of the cell lysates, whereas PKCz-related kinase was present both in 105,000-g supernatant and particulate fractions [17]. Furthermore, the anti-PKCz antibody used in this study, which was also raised against the C-terminal sequence of PKCz, did not cross-react with PKCa (data not shown). These facts demonstrate that PKCz-related kinase is not derived from cPKCs. The molecular weights of PKCı [4] and PKCl [5] are 65 and 74 kDa, respectively. These weights are different from those of PKCz-related kinase (79 and 85 kDa). Furthermore, it has been reported that autophosphorylation of PKCl is observed in the absence of PS [5]. Therefore, it seems unlikely that PKCz-related kinase is one of these isozymes. At present, there may still be a possibility that PKCz-related kinase is derived from PKCz gene by alternative splicing, even though the genomic structure of PKCz gene has not been fully understood. Since the anti-PKCz antibody used in this study is a polyclonal antibody, it is possible that this antibody detects an alternative spliced form of PKCz and a highly homologous protein to PKCz as well as PKCz itself. To elucidate whether PKCz-related kinase is derived from PKCz gene or not, further studies must be performed. When WYF31 cells were cultured in the presence of 0.5% FCS for 48 h, the content of PKCz-related kinase increased three to four times compared to the cells cultured in 10% FCS, with a concomitant decrease of PKCz content, both in 105,000-g supernatant and in particulate fractions. We also found that the total PSdependent kinase activity of PKCz-related kinase obtained from serum-starved cells increased two- to threefold compared to that of the cells cultured in 10% FCS (data not shown). These results suggest that the increase of the expression of PKCz-related kinase might play a role in the growth-suppressive signals in WYF31 cells. In our previous study using primary cultured mouse epidermal cells, transforming growth factor b (TGFb), which has a potent growth-suppressive effect on this type of cell, strongly suppressed the Ca21-dependent PKC activity but slightly increased the Ca21-independent PKC activity both in 105,000-g supernatants and particulate fractions [23]. This Ca21-independent PKC activity primarily reflects PKCz activity [17]. It is possible that PKCz-related kinase may play a role in the common processes of growth-suppressive signals induced by serum starvation or TGFb treatment. The increase of PKCz-related kinase content and the concomitant decrease of PKCz content induced by growth suppression were reversed by treating the cells with 10% FCS. In contrast, treatment of the growthsuppressed cells with PMA, which is known to induce the proliferation of WYF31 cells [19], caused the apparent shift of 79-kDa PKCz-related kinase to the 85-kDa form in particulate fractions within 1 h. Prolonged treat-
21
ment with PMA induced the downregulation of PKCzrelated kinase, but PKCz was not affected by PMA treatment. These results suggest that the continuous presence of PKCz in the 105,000-g supernatants and the decrease of the amount of PKCz-related kinase both in 105,000-g supernatants and particulate fractions may be important in the growth-stimulating signals induced by either FCS or PMA in WYF31 cells. We have already found that the apparent shift of 79-kDa PKCz-related kinase to the 85-kDa form in particulate fractions by PMA treatment results at least in part from the phosphorylation of 79-kDa PKCz-related kinase in primary cultured mouse epidermal cells [18]. The total activity of PKCz-related kinase increases in association with the apparent shift from the 75- to 85-kDa form [18]. Therefore, it seems possible that the transient activation of PKCz-related kinase in particulate fractions in association with the apparent band shift plays a significant role in the early stage of the PMA-caused growth stimulation of WYF31 cells. In conclusion, we showed that PKCz-related kinase and PKCz are closely related but different proteins, and that the expression of PKCz-related kinase and PKCz are regulated differently in WYF31 cells. The exact mechanism responsible for the induction of PKCzrelated kinase and the reduction of PKCz during cell growth suppression has yet to be elucidated. To elucidate the independent signal transduction processes of PKCz-related kinase and PKCz, it is necessary to determine their specific endogenous substrate proteins. For this purpose, the determination of their specific substrate motifs using degenerate peptide library techniques [7] is now underway.
Acknowledgments We thank Mr. Kouichi Maruyama for his competent technical assistance. This work was supported in part by a general grant from the Japanese Ministry of Education, Science, and Culture, and by a grant from Keio Gijuku Academic Development Funds, Keio University, Tokyo, Japan.
References [1] Nishizuka Y. Science 1992;258:607–14. [2] Hug H, Sarre TF. Biochem J 1993;291:329–43. [3] Johannes FJ, Prestle J, Eis S, Oberhagemann P, Pfizenmaier K. J Biol Chem 1994;269:6140–8. [4] Selbie LA, Schmitzpeiffer C, Sheng YH, Biden TJ. J Biol Chem 1993;268:24296–320. [5] Akimoto K, Mizuno K, Osada S, Hirai S, Tanuma S, Suzuki K, Ohno S. J Biol Chem 1994;269:12677–83. [6] Dekker LV, Parker PJ. Trends Biochem Sci 1994;19:73–7. [7] Nishikawa K, Toker A, Johannes F-J, Songyang Z, Cantley LC. J Biol Chem 1997;272:952–60.
22
K. Nishikawa et al. / Cellular Signalling 12 (2000) 15–22
[8] Berra E, Diazmeco MT, Dominguez I, Municio MM, Sanz L, Lozano J, Chapkin RS, Moscat J. Cell 1993;74:555–63. [9] Diazmeco MT, Berra E, Municio MM, Sanz L, Lozano J, Dominguez I, Diazgolpe V, Delera MTL, Alcami J, Paya CV, Arenzanaseisdedos F, Virelizier JL, Moscat J. Mol Cell Biol 1993; 13:4770–5. [10] Dominguez I, Sanz L, Arenzanaseisdedos F, Diazmeco MT, Virelizier JL, Moscat J. Mol Cell Biol 1993;13:1290–5. [11] Diazmeco MT, Lozano J, Municio MM, Berra E, Frutos S, Sanz L, Moscat J. Journal of Biological Chemistry 1994;269:31706–10. [12] Lozano J, Berra E, Municio MM, Diazmeco MT, Dominguez I, Sanz L, Moscat J. J Biol Chem 1994;269:19200–2. [13] Ways DK, Posekany K, Devente J, Garris T, Chen JM, Hooker J, Qin WX, Cook P, Fletcher D, Parker P. Cell Growth Differentiation 1994;5:1195–203. [14] Wooten MW, Zhou GS, Seibenhener ML, Coleman ES. Cell Growth & Differentiation 1994;5:395–403.
[15] Crespo P, Mischak H, Gutkind JS. Biochem Biophys Res Commun 1995;213:266–72. [16] Montaner S, Ramos A, Perona R, Esteve P, Carnero A, Lacal JC. Oncogene 1995;10:2213–20. [17] Nishikawa K, Yamamoto S, Nagumo H, Ishihara M, Murayama N, Kato R. Cell Signal 1994;6:503–12. [18] Nishikawa K, Yamamoto S, Nagumo H, Maruyama K, Kato R. Cell Signal 1995;7:491–504. [19] Wang JC, Yamamoto S, Kato R. Carcinogenesis 1992;13:1301–5. [20] Nishikawa K, Yamamoto S, Otsuka C, Kato R. Cell Signal 1992; 4:757–76. [21] Nakanishi H, Exton JH. J Biol Chem 1992;267:16347–54. [22] Allen BG, Andrea JE, Walsh MP. J Biol Chem 1994;269: 29288–98. [23] Nishikawa K, Yamamoto S, Nagumo H, Otsuka C, Kato R. Biochem Biophys Res Commun 1993;193:384–9.