- Email: [email protected]
Mutation of the hydrophobic motif in a phosphorylation-deficient mutant renders protein kinase C delta more apoptotically active
Archives of Biochemistry and Biophysics 493 (2010) 242–248
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi
Mutation of the hydrophobic motif in a phosphorylation-deficient mutant renders protein kinase C delta more apoptotically active Mrigendra B. Karmacharya a, Jung-Ie Jang a, Yoon-Jin Lee b, Yun-Sil Lee b, Jae-Won Soh a,* a b
Biomedical Research Center for Signal Transduction Networks, Department of Chemistry, Inha University, Incheon 402-751, Korea Laboratory of Radiation Effect, Korea Institute of Radiological and Medical Sciences, Seoul 139-706, Korea
a r t i c l e
i n f o
Article history: Received 13 October 2009 and in revised form 6 November 2009 Available online 13 November 2009 Keywords: PKCd Hydrophobic motif Phosphorylation Apoptosis
a b s t r a c t Protein kinase C delta (PKCd) is one of the important isoforms of PKCs that regulate various cellular processes, including cell survival and apoptosis. Studies have shown that activation of PKCd is correlated with apoptosis in various cell types, depending upon various stimuli. Phosphorylation of Thr505, Ser643 and Ser662 is crucial in activation of PKCd. Furthermore, phosphorylation of tyrosine residues, in particular that of Tyr311, is associated with PKCd activation and induction of apoptosis. Here, we generated a hydrophobic motif phosphorylation-deficient mutant of PKCd (PKCd-S662A) by mutating Ser662 to Ala, and studied the effect of this mutation in inducing apoptosis in L929 murine fibroblasts. We report that this mutation renders PKCd apoptotically more active. Furthermore, we found that the mutant PKCdS662A is tyrosine-phosphorylated and translocated to the membrane faster than its wild-type counterpart. Ó 2009 Elsevier Inc. All rights reserved.
Introduction Protein kinase C delta (PKCd),1 originally discovered in mouse epidermis [1], has been extensively studied, and its roles in cell signaling as a pro- as well as an anti-apoptotic protein have been described [2,3]. Studies show that kinase-active PKCd is involved in up-regulating cell apoptosis in response to various stimuli such as H2O2 [4,5], Tumor Necrosis Factor-a (TNF-a) [6], UV radiation [7], and DNA-damaging agents [8]. Activation of PKCd was found to be associated with inhibition of cell cycle progression [9]. PKCd contains an N-terminal regulatory domain and a C-terminal catalytic domain with a variable V3 region interposed between them (Fig. 1). The V3 region, which is highly accessible to proteolytic cleavage, contains a caspase-3 cleavage site [10] flanked by two important phosphorylation sites, Tyr311 and Tyr332. Upon cleavage at the V3 region, a constitutively active catalytic domain is released. The catalytic domain contains two conserved regions, C3 (ATP-binding region) and C4 (catalytically active/substrate-binding region), interspersed by two variable regions, V4 and V5. The C4 domain contains the activation loop residue Thr505 (Fig. 2). The V5 region contains two conserved phosphorylation sites: the turn motif Ser643 (Ser645 in humans) and the hydrophobic motif Ser662 (Ser664 in humans). Phosphorylation of the activation loop, the turn motif,
* Corresponding author. Fax: +82 32 872 6698. E-mail address: [email protected] (J.-W. Soh). 1 Abbreviations used: PKCd, protein kinase C delta; TNF-a, Tumor Necrosis Factor-a; PMA, phorbol 12-myristate 13-acetate; MBP, myelin basic protein; PCR, polymerase chain reaction; EGF, epidermal growth factor; PDGF, platelet-derived growth factor. 0003-9861/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2009.11.007
and the hydrophobic motif residues are considered essential steps in the activation of PKCd (see below). The N-terminal regulatory domain contains two conserved domains: a DAG/phorbol ester-binding C1 domain and a C2-like domain that lacks key sequences essential for Ca2+-coordination, which are present in the authentic C2 domains of cPKCs [11]. The C1 domain consists of a tandem repeat of two non-equivalent cysteine-rich zinc-finger-like domains, C1A and C1B, with an isolated C1A domain having a much higher affinity for DAG than that of C1B [12], unlike in PKCc and PKCe, where both C1A and C1B domains have comparable affinities for DAG [13]. The reason for the non-equivalence of C1A and C1B domains of PKCd has been attributed to an acidic residue in the C1A domain, Glu177, that plays a key role in regulating DAG accessibility. Furthermore, the affinity of the C1 domain of PKCd for DAG is much higher than that of cPKCs, which compensates for the lack of key Ca2+-binding sequences in the C2-like domain in PKCd [14]. Additionally, an auto-inhibitory pseudosubstrate domain precedes the C1 domain, and two variable regions V1 and V2 flank the C2-like domain. Among other PKC isoforms, PKCd poses a unique case in that it can be activated without the activation loop of Thr505-phosphorylation [15], although such activation is quite low. Stempka et al. have shown that Thr505-phosphorylation is not important, while the acidic residue Glu500 and autophosphorylation at Ser643 are important, for the activation of PKCd [16]. It has been proposed that PKCd has a unique structural mechanism, where twenty residues N-terminal to the kinase domain and a pair of phenylalanine residues Phe500 and Phe527 play a critical role in stabilizing its activation loop, which enables it to be catalytically active even in the absence of activation-loop phosphorylation [17]. Moreover, it
M.B. Karmacharya et al. / Archives of Biochemistry and Biophysics 493 (2010) 242–248
243
Fig. 1. Schematic structure of PKCd, showing conserved and variable regions.
Fig. 2. Alignment of PKC isoforms, showing conserved phosphorylation sites.
has been reported in cardiomyocytes that the activation loop Thr505 of PKCd is autophosphorylated, the event being triggered by PKCe overexpression and PKCd-Tyr311/Tyr332 phosphorylation [18]. Furthermore, while the turn motif Ser643 in PKCd is autophosphorylated, the hydrophobic motif Ser662 is transphosphorylated by an upstream kinase PKCn in vitro and in vivo [19], as is evident from the fact that expression of an active PKCn induces the hydrophobic motif Ser662-phosphorylation of PKCd [20]. In addition, Ser299, Ser302 and Ser304 of the V3 region have been reported as in vitro autophosphorylation sites in PKCd [21]. In addition to these serine/threonine-phosphorylations, PKCd is activated by tyrosine-phosphorylation [22]. Several tyrosine residues have been identified as correlating with the activation of PKCd in response to various stimuli. Tyr52, Tyr155 and Tyr187 in the regulatory domain, Tyr311 and Tyr332 in the V3 region and Tyr512 and Tyr523 in the catalytic domain are some of the important tyrosine-phosphorylation sites [23]. UV enhances tyrosinephosphorylation of PKCd, which contributes to the induction of apoptosis in HaCaT cells [7]. In response to H2O2, PKCd is phosphorylated at Tyr311, Tyr332, and Tyr512 [24]. Phosphorylation at Tyr311 and Tyr332 induces proteolytic activation of PKCd [25,26]. Furthermore, PKCd has been found to interact with heat shock proteins (HSP90, HSP25 and HSP27) [27–29]. Specifically, we have shown elsewhere that PKCd interacts with HSP27 though the V5 region [30]. It is evident from these studies that the V5 region plays a critical role in activation and interaction of PKCd. In this study, we investigated the role of the hydrophobic motif residue Ser662 in the V5 region of PKCd in activation, translocation, and status of tyrosine-phosphorylation of the enzyme. Contrary to the prevailing notion that phosphorylation of the hydrophobic motif is associated with the activation of PKCd, we report that the mutation of a hydrophobic motif to a phosphorylation-deficient mutant (PKCd-S662A) renders PKCd apoptotically more active. Furthermore, consistent with this result, the mutant PKCd-S662A was found to be more tyrosine-phosphorylated, and faster translocated to the membrane, than its wild-type counterpart.
substrate were purchased from Upstate Biotechnology. [c-32P]ATP was purchased from Sigma. H2O2 was purchased from Calbiochem. Protein assay reagent was purchased from Bio-Rad. Phosphop53(Ser15) antibody was purchased from NEB. Plasmid construction Wild-type (WT) mouse PKCd (GenBank Accession No. AY545076) was cloned into pHANE that contains an N-terminal HA-tag. The hydrophobic motif phosphorylation-deficient mutant of PKCd (PKCd-S662A) was constructed by polymerase chain reaction (PCR) using overlap extension primers to replace the hydrophobic motif residue Ser662 with Ala. The sequences of the PCR-constructed cDNAs were confirmed by DNA sequencing. The PCR products were digested with EcoRI and cloned into the EcoRI site of the pHANE vector. All primer sequences are available upon request. Cell culture L929 murine fibroblasts were cultured in Dulbecco’s minimal essential medium, DMEM (Invitrogen) supplemented with heatinactivated 10% fetal bovine serum, FBS (Invitrogen), and antibiotics at 37 °C in a 5% CO2 humidified incubator. Irradiation Cells were plated on 50 mm dishes and incubated at 37 °C in 5% CO2 in a culture medium and were grown until the cells were 70– 80% confluent. Cells were then exposed to c-rays from a 137Cs c-ray source (Atomic Energy of Canada, Ltd., Canada) at 10 Gy/min. Flow cytometric analysis Cells were cultured, harvested at the indicated times, stained with propidiumiodide (PI), according to the manufacturer’s protocol, and then analyzed using a FACScan flow cytometer (BD Biosciences).
Materials and methods Polyacrylamide gel electrophoresis (PAGE) and Western blotting Reagents Antibodies of hemagglutinin (HA) and PKCd were purchased from Santa Cruz Biotechnology. Phosphotyrosine antibody, phorbol 12-myristate 13-acetate (PMA) and myelin basic protein (MBP)
For PAGE and Western blotting, cells were solubilized with lysis buffer (120 mM NaCl, 40 mM Tris (pH 8.0), 0.1% Nonidet P-40) and boiled for 5 min, and equal amounts of proteins were electrophoresed on 7.5% SDS–PAGE. After electrophoresis, proteins were trans-
244
M.B. Karmacharya et al. / Archives of Biochemistry and Biophysics 493 (2010) 242–248
ferred to a nitrocellulose membrane and processed for immunoblotting with an HA antibody. Blots were further incubated with horseradish peroxidase-conjugated secondary antibody diluted at 1:2000, and specific bands were visualized by chemiluminescence (ECL, Amersham Biosciences). Autoradiographs were recorded on X-omat AR film (Eastman Kodak Co.). Western blot analyses of all protein samples in this study were performed by b-actin as a loading control (data not shown) and confirmed that there were no changes in the total amounts of proteins in each sample.
were treated with TPA for 3 h to a final concentration of 100 ng/ mL. Dimethyl sulfoxide (DMSO) was used as a control. Luciferase Assays were done with serum-starved or TPA-treated cells using a Luciferase Assay System (Promega). Luciferase activities were normalized by b-gal activities. b-gal assays were performed using b-Galactosidase Enzyme Assay System (Promega). Results Construction and expression of HA-tagged PKCd mutants
PKC kinase assay Cellular proteins were extracted using PKC extraction buffer (50 mM HEPES (pH 7.5), 150 mM NaCl, 0.1% Tween20, 1 mM EDTA, 2.5 mM EGTA, and 10% glycerol) containing protease inhibitors (10 lg/mL aprotinin, 10 lg/mL leupeptin and 0.1 mM phenylmethylsulfonyl fluoride) and phosphatase inhibitors (1 mM NaF, 0.1 mM Na3VO4, and 10 mM b-glycerophosphatate). HA-tagged PKC proteins from 300 lg of cell extracts were immunoprecipitated with 2 lg of HA antibody and 30 lL of protein G-Sepharose for 3 h at 4 °C. The immunoprecipitates were washed twice with PKC extraction buffer and PKC reaction buffer (50 mM HEPES (pH 7.5), 10 mM MgCl2, 1 mM dithiotheitol, 2.5 mM EGTA, 1 mM NaF, 0.1 mM Na3VO4, and 10 mM b-glycerophosphate) and then resuspended in 20 lL of PKC reaction buffer. The kinase assay was initiated by adding 40 lL of PKC reaction buffer containing 5 lg of dephosphorylated MBP (Upstate Biotechnology, Inc.) and 5 lCi of [c-32P]ATP. Reactions were carried out for 30 min at 30 °C and terminated by adding SDS sample buffer. Mixtures were then boiled for 5 min, and the reaction products were analyzed by SDS–PAGE and autoradiography. Immunoprecipitation Cells were lysed in immunoprecipitation buffer (50 mM HEPES (pH 7.6), 150 mM NaCl, 5 mM EDTA, 0.1% Nonidet P-40). After centrifugation (15,000g for 10 min) and removal of the particulate materials, the supernatant was incubated with 2 lg of HA antibody (mouse monoclonal antibody from Santa Cruz Biotechnology) with constant agitation at 4 °C. Immunocomplexes were precipitated with protein G-Sepharose (Sigma) and analyzed by SDS–PAGE with enhanced chemiluminescence (Amersham Biosciences). Fractionation of cytosolic and membrane proteins Cells were washed twice with PBS and then sonicated in buffer containing 100 mM Tris–HCl (pH 7.5), 50 mM MgCl2, 10 mM dithiotheitol, and 10 mM phenylsulfonyl fluoride. The sonicated cells were then incubated on ice for 1 h and ultracentrifuged for 1 h at 100,000g to separate the membranous and cytosolic fractions. The cell pellet was then sonicated in a buffer containing 1% Triton X-100, incubated on ice for 1 h, and then ultracentrifuged for 1 h at 10,000g to separate the membrane fraction from cellular debris.
The hydrophobic motif residue Ser662 of PKCd is unique among PKCs in that it is transphosphorylated rather than autophosphorylated. Here, we generated the hydrophobic motif phosphorylationdeficient mutant of PKCd, PKCd-S662A, by PCR using overlap extension primers. L929 murine fibroblast cells were transfected with empty vector pHANE, pHANE-PKCd-WT, and the mutant pHANE-PKCd-S662A separately. We found that the PKCd-WT and PKCd-S662A have similar transfection efficiencies, which are much higher than the control vector (Fig. 3). PKCd-S662A mutant has higher kinase activity than PKCd-WT First of all, we examined the difference in PKC kinase activities of PKCd-WT and PKCd-S662A. We observed that the in vitro kinase activity of PKCd-S662A mutant was higher than that of PKCd-WT when assayed using MBP as an in vitro kinase substrate (Fig. 4). Apoptotic activity of PKCd-S662A mutant is higher than PKCd-WT While PKCd has been reported to enhance cell survival [31], a majority of studies indicate that it is involved in inducing apoptosis [2,3,32]. Ionizing radiation is one of the stimuli that induce apoptosis in various cell lines. We have previously shown that overexpression of PKCd leads to decreased clonogenic survival and increased apoptosis after radiation in NIH3T3 cells [33]. To examine the effects of the mutation in radiation-induced apoptosis, we tested whether Ser662-to-Ala mutation in PKCd enhances apoptotic cell death in L929 cells. We observed that the mutant PKCd-S662A induced cell death more than PKCd-WT (Fig. 5A). In addition, it has been shown that PKCd induces apoptosis in U-937 cells when activated by H2O2 [34], so we examined and compared the apoptotic activity of PKCd-WT with that of PKCd-S662A in H2O2-stimulated L929 cells. We observed that PKCd-S662A induced more cell death after H2O2-treatment when
Fig. 3. PKCd-WT and the mutant PKCd-S662A have similar transfection efficiencies. The transfection efficiencies of HA-tagged vectors were confirmed by Western blotting using an anti-HA antibody.
Luciferase assay L929 cells were grown in DMEM containing 10% calf serum. Triplicate samples of 1 105 cells in 35 mm plates were transfected using Lipofectin (Gibco BRL) with 1 lg of reporter plasmid, 0.05–5 lg of various expression vectors, and 1 lg of pCMV-b-gal. pcDNA3 plasmid DNA was added to the transfections as needed to achieve the same total amount of plasmid DNA per transfection. Six hours after transfection, cells were fed with new media (DMEM with 10% calf serum) overnight. Cells were then serum-starved for 24 h in DMEM with 0.5% calf serum. For TPA experiments, cells
Fig. 4. PKCd-S662A has higher kinase activity than PKCd-WT. Lysates from L929 cells after transfecting HA-tagged PKCd-WT and PKCd-S662A vectors were immunoprecipitated with an anti-HA antibody. Cellular proteins were extracted after lysing with PKC extraction buffer. HA-tagged PKC proteins from 300 lg of cell extracts were immunoprecipitated using an anti-HA antibody and protein A-Sepharose. Immune-complex kinase reactions were performed in the presence of MBP substrate and [c-32P]ATP.
M.B. Karmacharya et al. / Archives of Biochemistry and Biophysics 493 (2010) 242–248
245
Fig. 5. Apoptotic activity of PKCd-S662A mutant is higher than PKCd-WT. (A) The cell deaths of L929 cells transfected with the indicated PKCd-WT or PKCd-S662A mutant with or without treatment of 10-Gy c-ray for 48 h were measured by flow cytometry after propidiumiodide (PI) staining. Results are the means ± SD. (B) The cell death of L929 cells transfected with the indicated PKCd-WT or PKCd-S662A mutant with or without treatment of H2O2 (2 mM) for 5–6 h were measured by flow cytometry using PI staining. Results are the means ± SD.
compared with PKCd-WT (Fig. 5B). These findings suggest that PKCd-S662A mutant has higher apoptotic activity than PKCd-WT.
PKCd-S662A mutant is more tyrosine-phosphorylated than PKCd-WT after H2O2-treatment PKCd is known to be regulated by tyrosine-phosphorylation. PKCd is phosphorylated on tyrosine residues in cells stimulated by H2O2, PMA, epidermal growth factor (EGF), or platelet-derived growth factor (PDGF). Tyrosine-phosphorylation of the catalytic domain increases the kinase activity of PKCd in cells treated with H2O2, so we tested whether PKCd-S662A is more tyrosine-phosphorylated than PKCd-WT. We found that the extent of tyrosinephosphorylation of the control, PKCd-WT, and PKCd-S662A were comparable before H2O2-treatment. After 30 min of H2O2-treatment, PKCd-WT and the mutant were more tyrosine-phosphorylated than the control. After 60 min, tyrosine-phosphorylation of PKCd-S662A mutant increased more than that of PKCd-WT (Fig. 6A). These findings demonstrate that the PKCd-S662A mutant is more tyrosine-phosphorylated in response to H2O2 than PKCdWT is. It can be anticipated that an increase in tyrosine-phosphorylation of the PKCd-S662A mutant is likely to result in an increase in its kinase activity.
PKCd-S662A mutant is more phosphorylated at Tyr311 than PKCd-WT After observation that the mutant PKCd-S662A is more tyrosinephosphorylated than PKCd-WT, we decided to examine the tyrosine-phosphorylation status of a specific tyrosine residue in PKCd-WT and PKCd-S662A. As phosphorylation of Tyr311 has been reported to be indispensable for the proteolytic cleavage and activation of PKCd and the subsequent induction of apoptosis [25], we examined whether PKCd-S662A is more tyrosine-phosphorylated at Tyr311 than PKCd-WT. Using a phospho-PKCd-Tyr(311) antibody, we observed that PKCd-S662A is more tyrosine-phosphorylated at Tyr311 than PKCd-WT (Fig. 6B). Since phosphorylation of Tyr311 has been shown to correlate with the kinase activity of PKCd, and as the mutant PKCd-S662A showed more phosphorylation at Tyr311 compared with PKCd-WT, this observation further
Fig. 6. PKCd-S662A mutant is more tyrosine-phosphorylated than PKCd-WT after H2O2-treatment. (A) L929 cells were transfected with PKCd-WT and PKCd-S662A mutant and treated with or without H2O2 (2 mM). Cells were then lysed, and cellular proteins extracted. HA-tagged PKCd proteins were immunoprecipitated (IP) from 300 lg of cell extracts by using an anti-HA antibody and protein A-Sepharose. Immunodetection was then performed using anti-phosphotyrosine antibody. (B) PKCd-S662A mutant has more phosphorylation on Tyr311 than PKCd-WT. L929 cells were transfected with PKCd-WT and PKCd-S662A mutant. Cells were then lysed, and cellular proteins extracted. HA-tagged PKCd proteins were immunoprecipitated from 300 lg of cell extracts by using an anti-HA antibody and protein A-Sepharose. Immunodetection was then performed using anti-phospho-PKCd(Y311) antibody.
confirms that the Ser662-to-Ala mutation makes PKCd enzymatically more active.
246
M.B. Karmacharya et al. / Archives of Biochemistry and Biophysics 493 (2010) 242–248
in fact, due to its faster activation. The mutant PKCd-S662A is activated faster and thus moved faster to the membrane than PKCdWT. These findings thus demonstrate that the PKCd-S662A mutant is more active than PKCd-WT and is activated faster than PKCd-WT. PKCd-S662A degrades faster than PKCd-WT after PMA or cyclohexamide treatment
Fig. 7. p53 is more phosphorylated in PKCd-S662A mutant expressing cells than PKCd-WT expressing cells. The cell lysates of L929 cells transfected with the indicated PKCd-WT and PKCd-S662A mutant and Western blotting was performed using anti-p53 antibody and anti-phospho-p53(Ser15) antibody .
p53 is more phosphorylated at Ser15 in PKCd-S662A expressing cells than in PKCd-WT expressing cells Activation of PKCd has been reported to induce basal transcription of p53, a tumor suppressor gene [35]. In addition, phosphorylation of Ser15 of p53 correlates with induction of apoptosis [36]. We examined the effect of the mutation of the hydrophobic motif of PKCd in the expression of p53 and phosho-p53(Ser15). We observed that phosphorylation level of p53 at Ser15 is higher in cells expressing PKCd-S662A than cells expressing PKCd-WT, while the level of p53 protein expressed is similar in both cells expressing PKCd-WT and PKCd-S662A (Fig. 7). As activation of the p53 though phosphorylation at Ser15 can lead to G1/S cell cycle arrest and apoptosis, our finding that PKCd-S662A-expressing cells exhibited more phospho-p53(Ser15) proteins further suggests that the PKCd-S662A mutant has more apoptotic activity than PKCd-WT.
PKCd-S662A mutant translocates to membrane faster than PKCd-WT after PMA treatment PKCd translocates to membrane upon activation by various stimuli. To examine differences in translocation patterns between PKCd-WT and PKCd-S662A, we fractionated the cytosolic and membrane proteins and observed the translocation patterns of PKCd-WT and PKCd-S662A in various time intervals after treatment with PMA, a PKCd-activator. PKCd-WT and the PKCd-S662A mutant werre found to translocate to the membrane in natural conditions. We observed that PKCd-WT and PKCd-S662A translocated to the membrane after 30 and 60 min of PMA treatment, but PKCdS662A translocated to membrane faster than PKCd-WT (Fig. 8). As membrane-translocation of protein kinase C is considered to be the hallmark of its activation, it is reasonable to conjecture that the faster translocation of the PKCd-S662A mutant to membrane is,
As PKCd is known to be degraded after its activation, we next examined the stability of PKCd-S662A compared with PKCd-WT. To examine the stability of PKCd proteins, we harvested cells hourly after treatment with PMA for Western blot analysis. We found that the amount of both PKCd-WT and PKCd-S662A decreased after PMA treatment, but PKCd-S662A decreased faster than PKCd-WT. This shows that the degradation of PKCd-S662A is faster than that of PKCd-WT, and that PKCd-S662A is less stable than PKCd-WT. To further confirm this, we performed an experiment with cyclohexamide, a protein synthesis inhibitor. We again observed that PKCd-S662A mutant was degraded faster than PKCdWT (Fig. 9B). Thus, these results suggest that the PKCd-S662A mutant not only is activated faster but also degrades faster than PKCd-WT in response to stimuli such as PMA or cyclohexamide. Discussion In this study, we examined the role of the hydrophobic motif residue Ser662 in the V5 region of PKCd in the activation mechanism of the enzyme. Here, we mutated the residue Ser662 to Ala and generated the hydrophobic motif phosphorylation-deficient mutant PKCd-S662A. We studied the effect of this mutation in the activation of PKCd. As activation of PKCd has been shown to be involved in inducing apoptosis, we specifically examined the effect of this mutation in apoptotic activity of PKCd. We found that the mutant PKCd-S662A is apoptotically more active than PKCdWT. Consistent with this result, we found that the mutant PKCdS662A translocates to the membrane faster than the wild-type isoform. Activation of PKC isoforms in response to physiological agonists is marked by the membrane-translocation of the enzyme. In targeting the enzyme to the plasma membrane, the C1A domain of the regulatory region plays a crucial role. It is the C1A domain that binds with DAG, which is buried in the plasma membrane because of its hydrophobic nature, and thus facilitates the membranetranslocation of the enzyme. In the close and inactive conformation of PKC isoforms, the C1A domain is not exposed due to interaction between the C1 and C2 domains, and the C1A domain is therefore masked from binding with DAG. For activation of PKC isoforms, the C1 domain has to be exposed so that the C1A domain can bind with
Fig. 8. PKCd-S662A mutant translocates to membrane faster than PKCd-WT after PMA treatment. The cells were transfected with PKCd-WT and PKCd-S662A mutant. The cytosolic and membrane proteins were fractionated and the translocation patterns of PKCd-WT and PKCd-S662A were observed in various time intervals after PMA treatment (100 nM) using an anti-HA antibody.
M.B. Karmacharya et al. / Archives of Biochemistry and Biophysics 493 (2010) 242–248
247
Fig. 9. PKCd-WT is more stable than PKCd-S662A mutant. A. PKCd-S662A mutant degrades faster than PKCd-WT after PMA treatment. The cells were transfected with the indicated PKCd-WT and PKCd-S662A mutant with or without treatment of PMA (100 nM). Then, membrane and cytosolic proteins were fractionated from the cells. Immunodetection was done using anti-HA antibody in each of the cell lysates. B. PKCd-WT and PKCd-S662A mutant in lysates from L929 cells were immunodetected at 0, 1, 2, and 4 h after adding cyclohexamide (80 lg/mL) using anti-HA antibody. Lysates from L929 cells 0 h, 3 h, 6 h, 9 h, 12 h after adding PMA (100 nM) were immunodetected using anti-HA antibody.
DAG. It should be noted that the process of exposing the C1 domain is tightly regulated because C1 domains with the exposed hydrophobic patch are subject to protein aggregation in solution [37]. Indeed, in classical PKCs this domain is exposed and becomes accessible to DAG only after agonist-induced Ca2+-dependent membrane binding. Furthermore, as the tight interaction between the C1 and C2 domains occurs only after phosphorylation of the turn and hydrophobic motifs in the V5 region following the activation-loop phosphorylation, it is plausible to suppose that following conformational change there is an interaction between the V5 region and the regulatory domain during activation. Such an interaction between the V5 region and the C2 domain has indeed been reported in PKCe [38]. In addition, it has been shown that the phosphorylation at the V5 region in PKCbII affects the Ca2+-affinity of the enzyme, suggesting a possible interaction between the V5 region and the C2 domain [39]. In agreement with these findings, our current study too implies, albeit indirectly, that during activation of PKCd, there is an interaction between the V5 region and the C2-like domain in the regulatory region of the enzyme. We found that the mutation in the hydrophobic motif residue Ser662 affects the activation of PKCd and the membrane-targeting C1 domain has been shown to interact with the C2-like domain. Thus, we hypothesize that during activation the hydrophobic motif is phosphorylated and triggers a conformational change in which the V5 region interacts with the C2-like domain, which masks the C1 domain from being exposed, thus making it less accessible to DAG. When hydrophobic motif is mutated from serine to a phosphorylation-deficient alanine residue, it abrogates the interaction between the V5 region and the C2-like domain, rendering the V5 region unable to mask the C2-like domain. The C1 domain, which is otherwise present in a tight interaction with the C2-like domain forming a complex, is then more exposed and thus more accessible to DAG. Consequently, faster membrane-translocation of the phosphorylationdeficient mutant results which makes it more active than the wild-type. This result thus suggests that the phosphorylation of the hydrophobic motif prevents PKCd from being active. A similar result, where hydrophobic motif phosphorylation opposes the membrane-translocation of PKC, has been reported elsewhere [40]. They showed that the membrane-targeting function of C1 region is facilitated by the C2 region, but the carboxy-terminal autophosphorylation exerts an inhibitory effect on translocation that is most predominant when the C2 region is not functional. Furthermore, our previous study revealed that the phosphorylation-defi-
cient mutants of PKCd can bind to HSP25, whereas the phosphorylation-mimicking mutants cannot, suggesting that the non-phosphorylated form of PKCd-V5 is important for HSP25-binding [29]. Thus, it seems that while membrane-translocation of PKCd is facilitated by the C1 and C2-like domains, phosphorylation of the hydrophobic motif has an inhibitory effect on translocation and subsequent activation. The fact that the hydrophobic motif phosphorylation-deficient mutant is more active is clearly manifested in the apoptotic activity of the enzyme. As compared with the cells transfected with the wild-type isoform of PKCd, the cells transfected with the mutant PKCd-S662A suffered more apoptotic cell death after treatment with H2O2 and/or ionizing radiation. Consistent with these results, we found that the mutant PKCd-S662A is more tyrosine-phosphorylated, specifically at Tyr311, than PKCd-WT. As phosphorylation of various tyrosine residues, particularly that of Tyr311, has been correlated with the activation of PKCd, our finding that the mutant PKCd-S662A is more tyrosine-phosphorylated than PKCd-WT further supports our hypothesis that the mutant PKCd-S662A is more active than PKCd-WT. In addition, we found that the phosphorylation level of phospho-p53 at Ser15 is higher in cells transfected with the mutant PKCd-S662A than those transfected with PKCdWT, providing additional support to our notion that the mutant PKCd-S662A is apoptotically more active than PKCd-WT. Acknowledgments This work was supported by INHA UNIVERSITY Research Grant. References [1] M. Gschwendt, W. Kittstein, F. Marks, Biochem. Biophys. Res. Commun. 137 (1986) 766–774. [2] C. Brodie, P.M. Blumberg, Apoptosis 8 (2003) 19–27. [3] D.N. Jackson, D.A. Foster, FASEB J. 18 (2004) 627–636. [4] H. Konishi, H. Matsuzaki, H. Takaishi, T. Yamamoto, M. Fukunaga, Y. Ono, U. Kikkawa, Biochem. Biophys. Res. Commun. 264 (1999) 840–846. [5] K. Niwa, O. Inanami, T. Yamamori, T. Ohta, T. Hamasu, T. Karino, M. Kuwabara, Free Radic. Res. 36 (2002) 1147–1153. [6] Y. Emoto, Y. Manome, G. Meinhardt, H. Kisaki, S. Kharbanda, M. Robertson, T. Ghayur, W.W. Wong, R. Kamen, R. Weichselbaum, et al., EMBO J. 14 (1995) 6148–6156. [7] M. Fukunaga, M. Oka, M. Ichihashi, T. Yamamoto, H. Matsuzaki, U. Kikkawa, Biochem. Biophys. Res. Commun. 289 (2001) 573–579. [8] A. Basu, M.D. Woolard, C.L. Johnson, Cell Death Differ. 8 (2001) 899–908. [9] T. Watanabe, Y. Ono, Y. Taniyama, K. Hazama, K. Igarashi, K. Ogita, U. Kikkawa, Y. Nishizuka, Proc. Natl. Acad. Sci. USA 89 (1992) 10159–10163. [10] K. Yoshida, Cell. Signal. 19 (2007) 892–901.
248
M.B. Karmacharya et al. / Archives of Biochemistry and Biophysics 493 (2010) 242–248
[11] H. Pappa, J. Murray-Rust, L.V. Dekker, P.J. Parker, N.Q. McDonald, Structure 6 (1998) 885–894. [12] R.V. Stahelin, M.A. Digman, M. Medkova, B. Ananthanarayanan, J.D. Rafter, H.R. Melowic, W. Cho, J. Biol. Chem. 279 (2004) 29501–29512. [13] R.V. Stahelin, M.A. Digman, M. Medkova, B. Ananthanarayanan, H.R. Melowic, J.D. Rafter, W. Cho, J. Biol. Chem. 280 (2005) 19784–19793. [14] J.R. Giorgione, J.H. Lin, J.A. McCammon, A.C. Newton, J. Biol. Chem. 281 (2006) 1660–1669. [15] L. Stempka, A. Girod, H.J. Muller, G. Rincke, F. Marks, M. Gschwendt, D. Bossemeyer, J. Biol. Chem. 272 (1997) 6805–6811. [16] L. Stempka, M. Schnolzer, S. Radke, G. Rincke, F. Marks, M. Gschwendt, J. Biol. Chem. 274 (1999) 8886–8892. [17] Y. Liu, N.V. Belkina, C. Graham, S. Shaw, J. Biol. Chem. 281 (2006) 12102– 12111. [18] V.O. Rybin, J. Guo, Z. Gertsberg, H. Elouardighi, S.F. Steinberg, J. Biol. Chem. 282 (2007) 23631–23638. [19] W.H. Ziegler, D.B. Parekh, J.A. Le Good, R.D. Whelan, J.J. Kelly, M. Frech, B.A. Hemmings, P.J. Parker, Curr. Biol. 9 (1999) 522–529. [20] D.B. Parekh, W. Ziegler, P.J. Parker, EMBO J. 19 (2000) 496–503. [21] J. Durgan, N. Michael, N. Totty, P.J. Parker, FEBS Lett. 581 (2007) 3377–3381. [22] M. Gschwendt, Eur. J. Biochem. 259 (1999) 555–564. [23] S.F. Steinberg, Biochem. J. 384 (2004) 449–459. [24] H. Konishi, E. Yamauchi, H. Taniguchi, T. Yamamoto, H. Matsuzaki, Y. Takemura, K. Ohmae, U. Kikkawa, Y. Nishizuka, Proc. Natl. Acad. Sci. USA 98 (2001) 6587–6592. [25] S. Kaul, V. Anantharam, Y. Yang, C.J. Choi, A. Kanthasamy, A.G. Kanthasamy, J. Biol. Chem. 280 (2005) 28721–28730.
[26] W. Lu, H.K. Lee, C. Xiang, S. Finniss, C. Brodie, Cell. Signal. 19 (2007) 2165– 2173. [27] E.T. Maizels, C.A. Peters, M. Kline, R.E. Cutler Jr., M. Shanmugam, M. HunzickerDunn, Biochem. J. 332 (Pt. 3) (1998) 703–712. [28] S. Takai, R. Matsushima-Nishiwaki, H. Tokuda, E. Yasuda, H. Toyoda, Y. Kaneoka, A. Yamaguchi, T. Kumada, O. Kozawa, Life Sci. 81 (2007) 585–591. [29] Y.J. Lee, D.H. Lee, C.K. Cho, S. Bae, G.J. Jhon, S.J. Lee, J.W. Soh, Y.S. Lee, J. Biol. Chem. 280 (2005) 18108–18119. [30] E.H. Kim, H.J. Lee, D.H. Lee, S. Bae, J.W. Soh, D. Jeoung, J. Kim, C.K. Cho, Y.J. Lee, Y.S. Lee, Cancer Res. 67 (2007) 6333–6341. [31] Z. Liu, X. Liu, K.I. Nakayama, K. Nakayama, K. Ye, J. Neurochem. 100 (2007) 1278–1288. [32] T.A. DeVries-Seimon, A.M. Ohm, M.J. Humphries, M.E. Reyland, J. Biol. Chem. 282 (2007) 22307–22314. [33] Y.J. Lee, J.W. Soh, N.M. Dean, C.K. Cho, T.H. Kim, S.J. Lee, Y.S. Lee, Cell Growth Differ. 13 (2002) 237–246. [34] P.K. Majumder, N.C. Mishra, X. Sun, A. Bharti, S. Kharbanda, S. Saxena, D. Kufe, Cell Growth Differ. 12 (2001) 465–470. [35] T. Abbas, D. White, L. Hui, K. Yoshida, D.A. Foster, J. Bargonetti, J. Biol. Chem. 279 (2004) 9970–9977. [36] T. Shono, P.J. Tofilon, T.S. Schaefer, D. Parikh, T.J. Liu, F.F. Lang, Cancer Res. 62 (2002) 1069–1076. [37] W. Cho, J. Biol. Chem. 276 (2001) 32407–32410. [38] V. Kheifets, D. Mochly-Rosen, Pharmacol. Res. 55 (2007) 467–476. [39] A.S. Edwards, A.C. Newton, J. Biol. Chem. 272 (1997) 18382–18390. [40] X. Feng, K.P. Becker, S.D. Stribling, K.G. Peters, Y.A. Hannun, J. Biol. Chem. 275 (2000) 17024–17034.
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi
Mutation of the hydrophobic motif in a phosphorylation-deficient mutant renders protein kinase C delta more apoptotically active Mrigendra B. Karmacharya a, Jung-Ie Jang a, Yoon-Jin Lee b, Yun-Sil Lee b, Jae-Won Soh a,* a b
Biomedical Research Center for Signal Transduction Networks, Department of Chemistry, Inha University, Incheon 402-751, Korea Laboratory of Radiation Effect, Korea Institute of Radiological and Medical Sciences, Seoul 139-706, Korea
a r t i c l e
i n f o
Article history: Received 13 October 2009 and in revised form 6 November 2009 Available online 13 November 2009 Keywords: PKCd Hydrophobic motif Phosphorylation Apoptosis
a b s t r a c t Protein kinase C delta (PKCd) is one of the important isoforms of PKCs that regulate various cellular processes, including cell survival and apoptosis. Studies have shown that activation of PKCd is correlated with apoptosis in various cell types, depending upon various stimuli. Phosphorylation of Thr505, Ser643 and Ser662 is crucial in activation of PKCd. Furthermore, phosphorylation of tyrosine residues, in particular that of Tyr311, is associated with PKCd activation and induction of apoptosis. Here, we generated a hydrophobic motif phosphorylation-deficient mutant of PKCd (PKCd-S662A) by mutating Ser662 to Ala, and studied the effect of this mutation in inducing apoptosis in L929 murine fibroblasts. We report that this mutation renders PKCd apoptotically more active. Furthermore, we found that the mutant PKCdS662A is tyrosine-phosphorylated and translocated to the membrane faster than its wild-type counterpart. Ó 2009 Elsevier Inc. All rights reserved.
Introduction Protein kinase C delta (PKCd),1 originally discovered in mouse epidermis [1], has been extensively studied, and its roles in cell signaling as a pro- as well as an anti-apoptotic protein have been described [2,3]. Studies show that kinase-active PKCd is involved in up-regulating cell apoptosis in response to various stimuli such as H2O2 [4,5], Tumor Necrosis Factor-a (TNF-a) [6], UV radiation [7], and DNA-damaging agents [8]. Activation of PKCd was found to be associated with inhibition of cell cycle progression [9]. PKCd contains an N-terminal regulatory domain and a C-terminal catalytic domain with a variable V3 region interposed between them (Fig. 1). The V3 region, which is highly accessible to proteolytic cleavage, contains a caspase-3 cleavage site [10] flanked by two important phosphorylation sites, Tyr311 and Tyr332. Upon cleavage at the V3 region, a constitutively active catalytic domain is released. The catalytic domain contains two conserved regions, C3 (ATP-binding region) and C4 (catalytically active/substrate-binding region), interspersed by two variable regions, V4 and V5. The C4 domain contains the activation loop residue Thr505 (Fig. 2). The V5 region contains two conserved phosphorylation sites: the turn motif Ser643 (Ser645 in humans) and the hydrophobic motif Ser662 (Ser664 in humans). Phosphorylation of the activation loop, the turn motif,
* Corresponding author. Fax: +82 32 872 6698. E-mail address: [email protected] (J.-W. Soh). 1 Abbreviations used: PKCd, protein kinase C delta; TNF-a, Tumor Necrosis Factor-a; PMA, phorbol 12-myristate 13-acetate; MBP, myelin basic protein; PCR, polymerase chain reaction; EGF, epidermal growth factor; PDGF, platelet-derived growth factor. 0003-9861/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2009.11.007
and the hydrophobic motif residues are considered essential steps in the activation of PKCd (see below). The N-terminal regulatory domain contains two conserved domains: a DAG/phorbol ester-binding C1 domain and a C2-like domain that lacks key sequences essential for Ca2+-coordination, which are present in the authentic C2 domains of cPKCs [11]. The C1 domain consists of a tandem repeat of two non-equivalent cysteine-rich zinc-finger-like domains, C1A and C1B, with an isolated C1A domain having a much higher affinity for DAG than that of C1B [12], unlike in PKCc and PKCe, where both C1A and C1B domains have comparable affinities for DAG [13]. The reason for the non-equivalence of C1A and C1B domains of PKCd has been attributed to an acidic residue in the C1A domain, Glu177, that plays a key role in regulating DAG accessibility. Furthermore, the affinity of the C1 domain of PKCd for DAG is much higher than that of cPKCs, which compensates for the lack of key Ca2+-binding sequences in the C2-like domain in PKCd [14]. Additionally, an auto-inhibitory pseudosubstrate domain precedes the C1 domain, and two variable regions V1 and V2 flank the C2-like domain. Among other PKC isoforms, PKCd poses a unique case in that it can be activated without the activation loop of Thr505-phosphorylation [15], although such activation is quite low. Stempka et al. have shown that Thr505-phosphorylation is not important, while the acidic residue Glu500 and autophosphorylation at Ser643 are important, for the activation of PKCd [16]. It has been proposed that PKCd has a unique structural mechanism, where twenty residues N-terminal to the kinase domain and a pair of phenylalanine residues Phe500 and Phe527 play a critical role in stabilizing its activation loop, which enables it to be catalytically active even in the absence of activation-loop phosphorylation [17]. Moreover, it
M.B. Karmacharya et al. / Archives of Biochemistry and Biophysics 493 (2010) 242–248
243
Fig. 1. Schematic structure of PKCd, showing conserved and variable regions.
Fig. 2. Alignment of PKC isoforms, showing conserved phosphorylation sites.
has been reported in cardiomyocytes that the activation loop Thr505 of PKCd is autophosphorylated, the event being triggered by PKCe overexpression and PKCd-Tyr311/Tyr332 phosphorylation [18]. Furthermore, while the turn motif Ser643 in PKCd is autophosphorylated, the hydrophobic motif Ser662 is transphosphorylated by an upstream kinase PKCn in vitro and in vivo [19], as is evident from the fact that expression of an active PKCn induces the hydrophobic motif Ser662-phosphorylation of PKCd [20]. In addition, Ser299, Ser302 and Ser304 of the V3 region have been reported as in vitro autophosphorylation sites in PKCd [21]. In addition to these serine/threonine-phosphorylations, PKCd is activated by tyrosine-phosphorylation [22]. Several tyrosine residues have been identified as correlating with the activation of PKCd in response to various stimuli. Tyr52, Tyr155 and Tyr187 in the regulatory domain, Tyr311 and Tyr332 in the V3 region and Tyr512 and Tyr523 in the catalytic domain are some of the important tyrosine-phosphorylation sites [23]. UV enhances tyrosinephosphorylation of PKCd, which contributes to the induction of apoptosis in HaCaT cells [7]. In response to H2O2, PKCd is phosphorylated at Tyr311, Tyr332, and Tyr512 [24]. Phosphorylation at Tyr311 and Tyr332 induces proteolytic activation of PKCd [25,26]. Furthermore, PKCd has been found to interact with heat shock proteins (HSP90, HSP25 and HSP27) [27–29]. Specifically, we have shown elsewhere that PKCd interacts with HSP27 though the V5 region [30]. It is evident from these studies that the V5 region plays a critical role in activation and interaction of PKCd. In this study, we investigated the role of the hydrophobic motif residue Ser662 in the V5 region of PKCd in activation, translocation, and status of tyrosine-phosphorylation of the enzyme. Contrary to the prevailing notion that phosphorylation of the hydrophobic motif is associated with the activation of PKCd, we report that the mutation of a hydrophobic motif to a phosphorylation-deficient mutant (PKCd-S662A) renders PKCd apoptotically more active. Furthermore, consistent with this result, the mutant PKCd-S662A was found to be more tyrosine-phosphorylated, and faster translocated to the membrane, than its wild-type counterpart.
substrate were purchased from Upstate Biotechnology. [c-32P]ATP was purchased from Sigma. H2O2 was purchased from Calbiochem. Protein assay reagent was purchased from Bio-Rad. Phosphop53(Ser15) antibody was purchased from NEB. Plasmid construction Wild-type (WT) mouse PKCd (GenBank Accession No. AY545076) was cloned into pHANE that contains an N-terminal HA-tag. The hydrophobic motif phosphorylation-deficient mutant of PKCd (PKCd-S662A) was constructed by polymerase chain reaction (PCR) using overlap extension primers to replace the hydrophobic motif residue Ser662 with Ala. The sequences of the PCR-constructed cDNAs were confirmed by DNA sequencing. The PCR products were digested with EcoRI and cloned into the EcoRI site of the pHANE vector. All primer sequences are available upon request. Cell culture L929 murine fibroblasts were cultured in Dulbecco’s minimal essential medium, DMEM (Invitrogen) supplemented with heatinactivated 10% fetal bovine serum, FBS (Invitrogen), and antibiotics at 37 °C in a 5% CO2 humidified incubator. Irradiation Cells were plated on 50 mm dishes and incubated at 37 °C in 5% CO2 in a culture medium and were grown until the cells were 70– 80% confluent. Cells were then exposed to c-rays from a 137Cs c-ray source (Atomic Energy of Canada, Ltd., Canada) at 10 Gy/min. Flow cytometric analysis Cells were cultured, harvested at the indicated times, stained with propidiumiodide (PI), according to the manufacturer’s protocol, and then analyzed using a FACScan flow cytometer (BD Biosciences).
Materials and methods Polyacrylamide gel electrophoresis (PAGE) and Western blotting Reagents Antibodies of hemagglutinin (HA) and PKCd were purchased from Santa Cruz Biotechnology. Phosphotyrosine antibody, phorbol 12-myristate 13-acetate (PMA) and myelin basic protein (MBP)
For PAGE and Western blotting, cells were solubilized with lysis buffer (120 mM NaCl, 40 mM Tris (pH 8.0), 0.1% Nonidet P-40) and boiled for 5 min, and equal amounts of proteins were electrophoresed on 7.5% SDS–PAGE. After electrophoresis, proteins were trans-
244
M.B. Karmacharya et al. / Archives of Biochemistry and Biophysics 493 (2010) 242–248
ferred to a nitrocellulose membrane and processed for immunoblotting with an HA antibody. Blots were further incubated with horseradish peroxidase-conjugated secondary antibody diluted at 1:2000, and specific bands were visualized by chemiluminescence (ECL, Amersham Biosciences). Autoradiographs were recorded on X-omat AR film (Eastman Kodak Co.). Western blot analyses of all protein samples in this study were performed by b-actin as a loading control (data not shown) and confirmed that there were no changes in the total amounts of proteins in each sample.
were treated with TPA for 3 h to a final concentration of 100 ng/ mL. Dimethyl sulfoxide (DMSO) was used as a control. Luciferase Assays were done with serum-starved or TPA-treated cells using a Luciferase Assay System (Promega). Luciferase activities were normalized by b-gal activities. b-gal assays were performed using b-Galactosidase Enzyme Assay System (Promega). Results Construction and expression of HA-tagged PKCd mutants
PKC kinase assay Cellular proteins were extracted using PKC extraction buffer (50 mM HEPES (pH 7.5), 150 mM NaCl, 0.1% Tween20, 1 mM EDTA, 2.5 mM EGTA, and 10% glycerol) containing protease inhibitors (10 lg/mL aprotinin, 10 lg/mL leupeptin and 0.1 mM phenylmethylsulfonyl fluoride) and phosphatase inhibitors (1 mM NaF, 0.1 mM Na3VO4, and 10 mM b-glycerophosphatate). HA-tagged PKC proteins from 300 lg of cell extracts were immunoprecipitated with 2 lg of HA antibody and 30 lL of protein G-Sepharose for 3 h at 4 °C. The immunoprecipitates were washed twice with PKC extraction buffer and PKC reaction buffer (50 mM HEPES (pH 7.5), 10 mM MgCl2, 1 mM dithiotheitol, 2.5 mM EGTA, 1 mM NaF, 0.1 mM Na3VO4, and 10 mM b-glycerophosphate) and then resuspended in 20 lL of PKC reaction buffer. The kinase assay was initiated by adding 40 lL of PKC reaction buffer containing 5 lg of dephosphorylated MBP (Upstate Biotechnology, Inc.) and 5 lCi of [c-32P]ATP. Reactions were carried out for 30 min at 30 °C and terminated by adding SDS sample buffer. Mixtures were then boiled for 5 min, and the reaction products were analyzed by SDS–PAGE and autoradiography. Immunoprecipitation Cells were lysed in immunoprecipitation buffer (50 mM HEPES (pH 7.6), 150 mM NaCl, 5 mM EDTA, 0.1% Nonidet P-40). After centrifugation (15,000g for 10 min) and removal of the particulate materials, the supernatant was incubated with 2 lg of HA antibody (mouse monoclonal antibody from Santa Cruz Biotechnology) with constant agitation at 4 °C. Immunocomplexes were precipitated with protein G-Sepharose (Sigma) and analyzed by SDS–PAGE with enhanced chemiluminescence (Amersham Biosciences). Fractionation of cytosolic and membrane proteins Cells were washed twice with PBS and then sonicated in buffer containing 100 mM Tris–HCl (pH 7.5), 50 mM MgCl2, 10 mM dithiotheitol, and 10 mM phenylsulfonyl fluoride. The sonicated cells were then incubated on ice for 1 h and ultracentrifuged for 1 h at 100,000g to separate the membranous and cytosolic fractions. The cell pellet was then sonicated in a buffer containing 1% Triton X-100, incubated on ice for 1 h, and then ultracentrifuged for 1 h at 10,000g to separate the membrane fraction from cellular debris.
The hydrophobic motif residue Ser662 of PKCd is unique among PKCs in that it is transphosphorylated rather than autophosphorylated. Here, we generated the hydrophobic motif phosphorylationdeficient mutant of PKCd, PKCd-S662A, by PCR using overlap extension primers. L929 murine fibroblast cells were transfected with empty vector pHANE, pHANE-PKCd-WT, and the mutant pHANE-PKCd-S662A separately. We found that the PKCd-WT and PKCd-S662A have similar transfection efficiencies, which are much higher than the control vector (Fig. 3). PKCd-S662A mutant has higher kinase activity than PKCd-WT First of all, we examined the difference in PKC kinase activities of PKCd-WT and PKCd-S662A. We observed that the in vitro kinase activity of PKCd-S662A mutant was higher than that of PKCd-WT when assayed using MBP as an in vitro kinase substrate (Fig. 4). Apoptotic activity of PKCd-S662A mutant is higher than PKCd-WT While PKCd has been reported to enhance cell survival [31], a majority of studies indicate that it is involved in inducing apoptosis [2,3,32]. Ionizing radiation is one of the stimuli that induce apoptosis in various cell lines. We have previously shown that overexpression of PKCd leads to decreased clonogenic survival and increased apoptosis after radiation in NIH3T3 cells [33]. To examine the effects of the mutation in radiation-induced apoptosis, we tested whether Ser662-to-Ala mutation in PKCd enhances apoptotic cell death in L929 cells. We observed that the mutant PKCd-S662A induced cell death more than PKCd-WT (Fig. 5A). In addition, it has been shown that PKCd induces apoptosis in U-937 cells when activated by H2O2 [34], so we examined and compared the apoptotic activity of PKCd-WT with that of PKCd-S662A in H2O2-stimulated L929 cells. We observed that PKCd-S662A induced more cell death after H2O2-treatment when
Fig. 3. PKCd-WT and the mutant PKCd-S662A have similar transfection efficiencies. The transfection efficiencies of HA-tagged vectors were confirmed by Western blotting using an anti-HA antibody.
Luciferase assay L929 cells were grown in DMEM containing 10% calf serum. Triplicate samples of 1 105 cells in 35 mm plates were transfected using Lipofectin (Gibco BRL) with 1 lg of reporter plasmid, 0.05–5 lg of various expression vectors, and 1 lg of pCMV-b-gal. pcDNA3 plasmid DNA was added to the transfections as needed to achieve the same total amount of plasmid DNA per transfection. Six hours after transfection, cells were fed with new media (DMEM with 10% calf serum) overnight. Cells were then serum-starved for 24 h in DMEM with 0.5% calf serum. For TPA experiments, cells
Fig. 4. PKCd-S662A has higher kinase activity than PKCd-WT. Lysates from L929 cells after transfecting HA-tagged PKCd-WT and PKCd-S662A vectors were immunoprecipitated with an anti-HA antibody. Cellular proteins were extracted after lysing with PKC extraction buffer. HA-tagged PKC proteins from 300 lg of cell extracts were immunoprecipitated using an anti-HA antibody and protein A-Sepharose. Immune-complex kinase reactions were performed in the presence of MBP substrate and [c-32P]ATP.
M.B. Karmacharya et al. / Archives of Biochemistry and Biophysics 493 (2010) 242–248
245
Fig. 5. Apoptotic activity of PKCd-S662A mutant is higher than PKCd-WT. (A) The cell deaths of L929 cells transfected with the indicated PKCd-WT or PKCd-S662A mutant with or without treatment of 10-Gy c-ray for 48 h were measured by flow cytometry after propidiumiodide (PI) staining. Results are the means ± SD. (B) The cell death of L929 cells transfected with the indicated PKCd-WT or PKCd-S662A mutant with or without treatment of H2O2 (2 mM) for 5–6 h were measured by flow cytometry using PI staining. Results are the means ± SD.
compared with PKCd-WT (Fig. 5B). These findings suggest that PKCd-S662A mutant has higher apoptotic activity than PKCd-WT.
PKCd-S662A mutant is more tyrosine-phosphorylated than PKCd-WT after H2O2-treatment PKCd is known to be regulated by tyrosine-phosphorylation. PKCd is phosphorylated on tyrosine residues in cells stimulated by H2O2, PMA, epidermal growth factor (EGF), or platelet-derived growth factor (PDGF). Tyrosine-phosphorylation of the catalytic domain increases the kinase activity of PKCd in cells treated with H2O2, so we tested whether PKCd-S662A is more tyrosine-phosphorylated than PKCd-WT. We found that the extent of tyrosinephosphorylation of the control, PKCd-WT, and PKCd-S662A were comparable before H2O2-treatment. After 30 min of H2O2-treatment, PKCd-WT and the mutant were more tyrosine-phosphorylated than the control. After 60 min, tyrosine-phosphorylation of PKCd-S662A mutant increased more than that of PKCd-WT (Fig. 6A). These findings demonstrate that the PKCd-S662A mutant is more tyrosine-phosphorylated in response to H2O2 than PKCdWT is. It can be anticipated that an increase in tyrosine-phosphorylation of the PKCd-S662A mutant is likely to result in an increase in its kinase activity.
PKCd-S662A mutant is more phosphorylated at Tyr311 than PKCd-WT After observation that the mutant PKCd-S662A is more tyrosinephosphorylated than PKCd-WT, we decided to examine the tyrosine-phosphorylation status of a specific tyrosine residue in PKCd-WT and PKCd-S662A. As phosphorylation of Tyr311 has been reported to be indispensable for the proteolytic cleavage and activation of PKCd and the subsequent induction of apoptosis [25], we examined whether PKCd-S662A is more tyrosine-phosphorylated at Tyr311 than PKCd-WT. Using a phospho-PKCd-Tyr(311) antibody, we observed that PKCd-S662A is more tyrosine-phosphorylated at Tyr311 than PKCd-WT (Fig. 6B). Since phosphorylation of Tyr311 has been shown to correlate with the kinase activity of PKCd, and as the mutant PKCd-S662A showed more phosphorylation at Tyr311 compared with PKCd-WT, this observation further
Fig. 6. PKCd-S662A mutant is more tyrosine-phosphorylated than PKCd-WT after H2O2-treatment. (A) L929 cells were transfected with PKCd-WT and PKCd-S662A mutant and treated with or without H2O2 (2 mM). Cells were then lysed, and cellular proteins extracted. HA-tagged PKCd proteins were immunoprecipitated (IP) from 300 lg of cell extracts by using an anti-HA antibody and protein A-Sepharose. Immunodetection was then performed using anti-phosphotyrosine antibody. (B) PKCd-S662A mutant has more phosphorylation on Tyr311 than PKCd-WT. L929 cells were transfected with PKCd-WT and PKCd-S662A mutant. Cells were then lysed, and cellular proteins extracted. HA-tagged PKCd proteins were immunoprecipitated from 300 lg of cell extracts by using an anti-HA antibody and protein A-Sepharose. Immunodetection was then performed using anti-phospho-PKCd(Y311) antibody.
confirms that the Ser662-to-Ala mutation makes PKCd enzymatically more active.
246
M.B. Karmacharya et al. / Archives of Biochemistry and Biophysics 493 (2010) 242–248
in fact, due to its faster activation. The mutant PKCd-S662A is activated faster and thus moved faster to the membrane than PKCdWT. These findings thus demonstrate that the PKCd-S662A mutant is more active than PKCd-WT and is activated faster than PKCd-WT. PKCd-S662A degrades faster than PKCd-WT after PMA or cyclohexamide treatment
Fig. 7. p53 is more phosphorylated in PKCd-S662A mutant expressing cells than PKCd-WT expressing cells. The cell lysates of L929 cells transfected with the indicated PKCd-WT and PKCd-S662A mutant and Western blotting was performed using anti-p53 antibody and anti-phospho-p53(Ser15) antibody .
p53 is more phosphorylated at Ser15 in PKCd-S662A expressing cells than in PKCd-WT expressing cells Activation of PKCd has been reported to induce basal transcription of p53, a tumor suppressor gene [35]. In addition, phosphorylation of Ser15 of p53 correlates with induction of apoptosis [36]. We examined the effect of the mutation of the hydrophobic motif of PKCd in the expression of p53 and phosho-p53(Ser15). We observed that phosphorylation level of p53 at Ser15 is higher in cells expressing PKCd-S662A than cells expressing PKCd-WT, while the level of p53 protein expressed is similar in both cells expressing PKCd-WT and PKCd-S662A (Fig. 7). As activation of the p53 though phosphorylation at Ser15 can lead to G1/S cell cycle arrest and apoptosis, our finding that PKCd-S662A-expressing cells exhibited more phospho-p53(Ser15) proteins further suggests that the PKCd-S662A mutant has more apoptotic activity than PKCd-WT.
PKCd-S662A mutant translocates to membrane faster than PKCd-WT after PMA treatment PKCd translocates to membrane upon activation by various stimuli. To examine differences in translocation patterns between PKCd-WT and PKCd-S662A, we fractionated the cytosolic and membrane proteins and observed the translocation patterns of PKCd-WT and PKCd-S662A in various time intervals after treatment with PMA, a PKCd-activator. PKCd-WT and the PKCd-S662A mutant werre found to translocate to the membrane in natural conditions. We observed that PKCd-WT and PKCd-S662A translocated to the membrane after 30 and 60 min of PMA treatment, but PKCdS662A translocated to membrane faster than PKCd-WT (Fig. 8). As membrane-translocation of protein kinase C is considered to be the hallmark of its activation, it is reasonable to conjecture that the faster translocation of the PKCd-S662A mutant to membrane is,
As PKCd is known to be degraded after its activation, we next examined the stability of PKCd-S662A compared with PKCd-WT. To examine the stability of PKCd proteins, we harvested cells hourly after treatment with PMA for Western blot analysis. We found that the amount of both PKCd-WT and PKCd-S662A decreased after PMA treatment, but PKCd-S662A decreased faster than PKCd-WT. This shows that the degradation of PKCd-S662A is faster than that of PKCd-WT, and that PKCd-S662A is less stable than PKCd-WT. To further confirm this, we performed an experiment with cyclohexamide, a protein synthesis inhibitor. We again observed that PKCd-S662A mutant was degraded faster than PKCdWT (Fig. 9B). Thus, these results suggest that the PKCd-S662A mutant not only is activated faster but also degrades faster than PKCd-WT in response to stimuli such as PMA or cyclohexamide. Discussion In this study, we examined the role of the hydrophobic motif residue Ser662 in the V5 region of PKCd in the activation mechanism of the enzyme. Here, we mutated the residue Ser662 to Ala and generated the hydrophobic motif phosphorylation-deficient mutant PKCd-S662A. We studied the effect of this mutation in the activation of PKCd. As activation of PKCd has been shown to be involved in inducing apoptosis, we specifically examined the effect of this mutation in apoptotic activity of PKCd. We found that the mutant PKCd-S662A is apoptotically more active than PKCdWT. Consistent with this result, we found that the mutant PKCdS662A translocates to the membrane faster than the wild-type isoform. Activation of PKC isoforms in response to physiological agonists is marked by the membrane-translocation of the enzyme. In targeting the enzyme to the plasma membrane, the C1A domain of the regulatory region plays a crucial role. It is the C1A domain that binds with DAG, which is buried in the plasma membrane because of its hydrophobic nature, and thus facilitates the membranetranslocation of the enzyme. In the close and inactive conformation of PKC isoforms, the C1A domain is not exposed due to interaction between the C1 and C2 domains, and the C1A domain is therefore masked from binding with DAG. For activation of PKC isoforms, the C1 domain has to be exposed so that the C1A domain can bind with
Fig. 8. PKCd-S662A mutant translocates to membrane faster than PKCd-WT after PMA treatment. The cells were transfected with PKCd-WT and PKCd-S662A mutant. The cytosolic and membrane proteins were fractionated and the translocation patterns of PKCd-WT and PKCd-S662A were observed in various time intervals after PMA treatment (100 nM) using an anti-HA antibody.
M.B. Karmacharya et al. / Archives of Biochemistry and Biophysics 493 (2010) 242–248
247
Fig. 9. PKCd-WT is more stable than PKCd-S662A mutant. A. PKCd-S662A mutant degrades faster than PKCd-WT after PMA treatment. The cells were transfected with the indicated PKCd-WT and PKCd-S662A mutant with or without treatment of PMA (100 nM). Then, membrane and cytosolic proteins were fractionated from the cells. Immunodetection was done using anti-HA antibody in each of the cell lysates. B. PKCd-WT and PKCd-S662A mutant in lysates from L929 cells were immunodetected at 0, 1, 2, and 4 h after adding cyclohexamide (80 lg/mL) using anti-HA antibody. Lysates from L929 cells 0 h, 3 h, 6 h, 9 h, 12 h after adding PMA (100 nM) were immunodetected using anti-HA antibody.
DAG. It should be noted that the process of exposing the C1 domain is tightly regulated because C1 domains with the exposed hydrophobic patch are subject to protein aggregation in solution [37]. Indeed, in classical PKCs this domain is exposed and becomes accessible to DAG only after agonist-induced Ca2+-dependent membrane binding. Furthermore, as the tight interaction between the C1 and C2 domains occurs only after phosphorylation of the turn and hydrophobic motifs in the V5 region following the activation-loop phosphorylation, it is plausible to suppose that following conformational change there is an interaction between the V5 region and the regulatory domain during activation. Such an interaction between the V5 region and the C2 domain has indeed been reported in PKCe [38]. In addition, it has been shown that the phosphorylation at the V5 region in PKCbII affects the Ca2+-affinity of the enzyme, suggesting a possible interaction between the V5 region and the C2 domain [39]. In agreement with these findings, our current study too implies, albeit indirectly, that during activation of PKCd, there is an interaction between the V5 region and the C2-like domain in the regulatory region of the enzyme. We found that the mutation in the hydrophobic motif residue Ser662 affects the activation of PKCd and the membrane-targeting C1 domain has been shown to interact with the C2-like domain. Thus, we hypothesize that during activation the hydrophobic motif is phosphorylated and triggers a conformational change in which the V5 region interacts with the C2-like domain, which masks the C1 domain from being exposed, thus making it less accessible to DAG. When hydrophobic motif is mutated from serine to a phosphorylation-deficient alanine residue, it abrogates the interaction between the V5 region and the C2-like domain, rendering the V5 region unable to mask the C2-like domain. The C1 domain, which is otherwise present in a tight interaction with the C2-like domain forming a complex, is then more exposed and thus more accessible to DAG. Consequently, faster membrane-translocation of the phosphorylationdeficient mutant results which makes it more active than the wild-type. This result thus suggests that the phosphorylation of the hydrophobic motif prevents PKCd from being active. A similar result, where hydrophobic motif phosphorylation opposes the membrane-translocation of PKC, has been reported elsewhere [40]. They showed that the membrane-targeting function of C1 region is facilitated by the C2 region, but the carboxy-terminal autophosphorylation exerts an inhibitory effect on translocation that is most predominant when the C2 region is not functional. Furthermore, our previous study revealed that the phosphorylation-defi-
cient mutants of PKCd can bind to HSP25, whereas the phosphorylation-mimicking mutants cannot, suggesting that the non-phosphorylated form of PKCd-V5 is important for HSP25-binding [29]. Thus, it seems that while membrane-translocation of PKCd is facilitated by the C1 and C2-like domains, phosphorylation of the hydrophobic motif has an inhibitory effect on translocation and subsequent activation. The fact that the hydrophobic motif phosphorylation-deficient mutant is more active is clearly manifested in the apoptotic activity of the enzyme. As compared with the cells transfected with the wild-type isoform of PKCd, the cells transfected with the mutant PKCd-S662A suffered more apoptotic cell death after treatment with H2O2 and/or ionizing radiation. Consistent with these results, we found that the mutant PKCd-S662A is more tyrosine-phosphorylated, specifically at Tyr311, than PKCd-WT. As phosphorylation of various tyrosine residues, particularly that of Tyr311, has been correlated with the activation of PKCd, our finding that the mutant PKCd-S662A is more tyrosine-phosphorylated than PKCd-WT further supports our hypothesis that the mutant PKCd-S662A is more active than PKCd-WT. In addition, we found that the phosphorylation level of phospho-p53 at Ser15 is higher in cells transfected with the mutant PKCd-S662A than those transfected with PKCdWT, providing additional support to our notion that the mutant PKCd-S662A is apoptotically more active than PKCd-WT. Acknowledgments This work was supported by INHA UNIVERSITY Research Grant. References [1] M. Gschwendt, W. Kittstein, F. Marks, Biochem. Biophys. Res. Commun. 137 (1986) 766–774. [2] C. Brodie, P.M. Blumberg, Apoptosis 8 (2003) 19–27. [3] D.N. Jackson, D.A. Foster, FASEB J. 18 (2004) 627–636. [4] H. Konishi, H. Matsuzaki, H. Takaishi, T. Yamamoto, M. Fukunaga, Y. Ono, U. Kikkawa, Biochem. Biophys. Res. Commun. 264 (1999) 840–846. [5] K. Niwa, O. Inanami, T. Yamamori, T. Ohta, T. Hamasu, T. Karino, M. Kuwabara, Free Radic. Res. 36 (2002) 1147–1153. [6] Y. Emoto, Y. Manome, G. Meinhardt, H. Kisaki, S. Kharbanda, M. Robertson, T. Ghayur, W.W. Wong, R. Kamen, R. Weichselbaum, et al., EMBO J. 14 (1995) 6148–6156. [7] M. Fukunaga, M. Oka, M. Ichihashi, T. Yamamoto, H. Matsuzaki, U. Kikkawa, Biochem. Biophys. Res. Commun. 289 (2001) 573–579. [8] A. Basu, M.D. Woolard, C.L. Johnson, Cell Death Differ. 8 (2001) 899–908. [9] T. Watanabe, Y. Ono, Y. Taniyama, K. Hazama, K. Igarashi, K. Ogita, U. Kikkawa, Y. Nishizuka, Proc. Natl. Acad. Sci. USA 89 (1992) 10159–10163. [10] K. Yoshida, Cell. Signal. 19 (2007) 892–901.
248
M.B. Karmacharya et al. / Archives of Biochemistry and Biophysics 493 (2010) 242–248
[11] H. Pappa, J. Murray-Rust, L.V. Dekker, P.J. Parker, N.Q. McDonald, Structure 6 (1998) 885–894. [12] R.V. Stahelin, M.A. Digman, M. Medkova, B. Ananthanarayanan, J.D. Rafter, H.R. Melowic, W. Cho, J. Biol. Chem. 279 (2004) 29501–29512. [13] R.V. Stahelin, M.A. Digman, M. Medkova, B. Ananthanarayanan, H.R. Melowic, J.D. Rafter, W. Cho, J. Biol. Chem. 280 (2005) 19784–19793. [14] J.R. Giorgione, J.H. Lin, J.A. McCammon, A.C. Newton, J. Biol. Chem. 281 (2006) 1660–1669. [15] L. Stempka, A. Girod, H.J. Muller, G. Rincke, F. Marks, M. Gschwendt, D. Bossemeyer, J. Biol. Chem. 272 (1997) 6805–6811. [16] L. Stempka, M. Schnolzer, S. Radke, G. Rincke, F. Marks, M. Gschwendt, J. Biol. Chem. 274 (1999) 8886–8892. [17] Y. Liu, N.V. Belkina, C. Graham, S. Shaw, J. Biol. Chem. 281 (2006) 12102– 12111. [18] V.O. Rybin, J. Guo, Z. Gertsberg, H. Elouardighi, S.F. Steinberg, J. Biol. Chem. 282 (2007) 23631–23638. [19] W.H. Ziegler, D.B. Parekh, J.A. Le Good, R.D. Whelan, J.J. Kelly, M. Frech, B.A. Hemmings, P.J. Parker, Curr. Biol. 9 (1999) 522–529. [20] D.B. Parekh, W. Ziegler, P.J. Parker, EMBO J. 19 (2000) 496–503. [21] J. Durgan, N. Michael, N. Totty, P.J. Parker, FEBS Lett. 581 (2007) 3377–3381. [22] M. Gschwendt, Eur. J. Biochem. 259 (1999) 555–564. [23] S.F. Steinberg, Biochem. J. 384 (2004) 449–459. [24] H. Konishi, E. Yamauchi, H. Taniguchi, T. Yamamoto, H. Matsuzaki, Y. Takemura, K. Ohmae, U. Kikkawa, Y. Nishizuka, Proc. Natl. Acad. Sci. USA 98 (2001) 6587–6592. [25] S. Kaul, V. Anantharam, Y. Yang, C.J. Choi, A. Kanthasamy, A.G. Kanthasamy, J. Biol. Chem. 280 (2005) 28721–28730.
[26] W. Lu, H.K. Lee, C. Xiang, S. Finniss, C. Brodie, Cell. Signal. 19 (2007) 2165– 2173. [27] E.T. Maizels, C.A. Peters, M. Kline, R.E. Cutler Jr., M. Shanmugam, M. HunzickerDunn, Biochem. J. 332 (Pt. 3) (1998) 703–712. [28] S. Takai, R. Matsushima-Nishiwaki, H. Tokuda, E. Yasuda, H. Toyoda, Y. Kaneoka, A. Yamaguchi, T. Kumada, O. Kozawa, Life Sci. 81 (2007) 585–591. [29] Y.J. Lee, D.H. Lee, C.K. Cho, S. Bae, G.J. Jhon, S.J. Lee, J.W. Soh, Y.S. Lee, J. Biol. Chem. 280 (2005) 18108–18119. [30] E.H. Kim, H.J. Lee, D.H. Lee, S. Bae, J.W. Soh, D. Jeoung, J. Kim, C.K. Cho, Y.J. Lee, Y.S. Lee, Cancer Res. 67 (2007) 6333–6341. [31] Z. Liu, X. Liu, K.I. Nakayama, K. Nakayama, K. Ye, J. Neurochem. 100 (2007) 1278–1288. [32] T.A. DeVries-Seimon, A.M. Ohm, M.J. Humphries, M.E. Reyland, J. Biol. Chem. 282 (2007) 22307–22314. [33] Y.J. Lee, J.W. Soh, N.M. Dean, C.K. Cho, T.H. Kim, S.J. Lee, Y.S. Lee, Cell Growth Differ. 13 (2002) 237–246. [34] P.K. Majumder, N.C. Mishra, X. Sun, A. Bharti, S. Kharbanda, S. Saxena, D. Kufe, Cell Growth Differ. 12 (2001) 465–470. [35] T. Abbas, D. White, L. Hui, K. Yoshida, D.A. Foster, J. Bargonetti, J. Biol. Chem. 279 (2004) 9970–9977. [36] T. Shono, P.J. Tofilon, T.S. Schaefer, D. Parikh, T.J. Liu, F.F. Lang, Cancer Res. 62 (2002) 1069–1076. [37] W. Cho, J. Biol. Chem. 276 (2001) 32407–32410. [38] V. Kheifets, D. Mochly-Rosen, Pharmacol. Res. 55 (2007) 467–476. [39] A.S. Edwards, A.C. Newton, J. Biol. Chem. 272 (1997) 18382–18390. [40] X. Feng, K.P. Becker, S.D. Stribling, K.G. Peters, Y.A. Hannun, J. Biol. Chem. 275 (2000) 17024–17034.