The Proteolytic Cleavage of Protein Kinase C Isotypes, Which Generates Kinase and Regulatory Fragments, Correlates with Fas-Mediated and 12-O-Tetradecanoyl-Phorbol-13-Acetate-Induced

Eur. J. Biochem. 250, 7-18 (1997) 0 FEBS 1997

The proteolytic cleavage of protein kinase C isotypes, which generates kinase and regulatory fragments, correlates with Fas-mediated and 12-0-tetradecanoyl-phorbol-13-acetate-induced apoptosis Keiko MIZUNO’, Kumi NODA’, Tamao ARAKI’ ’, Tomomi IMAOKA?, Yuko KOBAYASHI’, Yoshiko AKITA4, Motoyuki SHIMONAKA’, Shuji KISH13 and Shigeo OHNO’ ’ Department of Molecular Biology, Yokohama City University School of Medicine, Yokohama, Japan * Science University of Tokyo, Tokyo, Japan ’ Pharmaceutical Basic Rerearch Laboratories JT Inc., Yokohama, Japan

Department of Molecular Biology, Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan

(Received 29 July 1997) - EJB 97 1083/1

Protein kinase C (PKC) has been implicated in signaling induced by diverse sets of stimuli regulating growth, differentiation, and apoptosis. The present study focused on the fate of PKC isotype proteins during Fas-mediated apoptosis of human leukemic cell lines. Among the PKC isotypes expressed in different cell types, such as Jurkat, HPB-ALL, U937, and HL60, all the nPKC isotypes including nPKC6, nPKCE, and nPKCB, but not cPKCa and fl1 and aPKCC (n, c, and a represent novel, conventional and atypical, respectively), showed limited proteolytic cleavage during Fas-mediated apoptosis. The limited proteolysis of nPKC isotypes means the disappearance of the intact protein band concomitant with the appearance of two fragments, most likely containing the kinase and regulatory domains, in contrast to the so-called down-regulation known for both cPKC and nPKC isotypes following exposure to stimuli such as 12-0-tetradecanoyl-phorbol 13-acetate (TPA). The time course of Fas-mediated apoptosis in Jurkat cells parallels that of the activation of a 32-kDa cysteine protease (CPP32)-like protease and also closely parallels the proteolytic cleavage of nPKC isotypes. A peptide inhibitor of the CPP32-like protease, Ac-DEVD-CHO, blocked the proteolytic cleavage of nPKC isotypes as well as apoptosis mediated by Fas. Transfection of recombinant protein coding for the catalytic fragment of nPKCG to COSl cells resulted in the apoptotic morphology of cells and nuclei.

The effect of TPA on apoptosis depends on the cell type. TPA significantly suppressed Fas-mediated apoptosis in Jurkat, whereas TPA alone caused apoptosis in HPB-ALL, U937, and HL60, only slight apoptosis in Jurkat. The proteolytic fragmentation of nPKC isotypes again closely correlated with the degree of apoptosis even in apoptosis induced by TPA. Separation of TPA-treated cells into apoptotic and non-apoptotic differentiating cells revealed that the proteolytic fragmentation of nPKC isotypes occurs only in apoptotic cells and, in adherent differentiating cells, nPKC isotypes as well as cPKCa were down- regulated without the generation of nPKC fragments. These results are consistent with the idea that nPKC isotypes meet two different fates, down-regulation and proteolytic cleavage generating kinase and regulatory fragments, and that the proteolytic cleavage of nPKC isotypes is a step in the signaling pathway involved in Fas-mediated and TPA-induced apoptosis.

Keywords: protein kinase C ; apoptosis; limited proteolysis; caspase.

Protein kinase C (PKC) is a serinehhreonine kinase implicated in various physiological responses, such as cell growth and differentiation [l, 21. Recent studies have demonstrated that PKC also modulates apoptosis, although the results are contra- dictory even in studies involving the same cell type. For example, treatment with PKC inhibitor induces apoptosis in lympho-

Correspondence to K. Mizuno and S. Ohno, Department of Molecu- lar Biology, Yokohama City University School of Medicine, Fuku-ura 3-9, Kanazawa-ku, Yokohama, Japan 236

Fax: +81 45 785 4140. E-mail: [email protected] Abbreviations. PKC, protein kinase C; nPKC, novel PKC; cPKC,

conventional PKC ; aPKC, atypical PKC ; TPA, 12-0-tetradecanoyl- phorbol 13-acetate; ICE, interleukin-ID-converting enzyme; CPP32, 32-kDa cysteine protease; CD3, cluster of differentiation antigen 3 ; GFP, green fluorescent protein; FITC, fluorescein isothiocyanate.

Enzyme. Protein kinase (EC 2.7.1.37).

cytes 131, the HL60 human leukemia cell line [4], and Swiss 3T3 fibroblasts [ 5 ] , but antagonizes CD3-mediated apoptosis of 2B4.11 T-cell hybrydomas [6]. Treatment with potent PKC activator phorbol esters, such as a 12-0-tetradecanoyl-phorbol 13- acetate (TPA), induces apoptosis in MCF-7 breast cancer cells expressing cPKCa [7], U937 constitutively expressing aPKCC [8] and a T-cell lymphoma Jurkat derivative expressing activated v-Ha-ras [9]. TPA also enhances the 1 -P-D-arabinofuranosylcytosine-induced apoptosis of HL60 [lo]. TPA and related com- pounds, however, inhibit CTLL cell apoptosis induced by interleukin withdrawal [ 1 I], ceramide-induced apoptosis of HL60 and U937 [12], and the VP-16 (etoposide)-induced apoptosis of DU-145 human carcinoma cells 1131.

The apparent conflict in the reports regarding the role of PKC in apoptosis seems to be due in part to the existence of multiple PKC isotypes. The PKC family consists of 11 different polypeptides divided into three subfamilies based on their struc-

8 Mizuno et al. ( E m J . Biochern. 250)

turd and biochemical properties, conventional PKC (cPKCa, PI, PII, and y ) , novel PKC (nPKC6, E, 0, and v), and atypical PKC (aPKC<, j h , and p ) [ l , 21. Among them, cPKC and nPKC mem- bers respond to TPA in vitro and i n vivo. The PKC isotypes show different activator sensitivities, substrate specificities, and tissue and cellular distributions, suggesting their specific cellular functions 11, 21.

The regulatory domain of PKC contains a sequence, the pseudosubstrate, that interacts with the catalytic domain and blocks substrate binding, hence the enzyme remains inactive in unstimulated cells [2, 141. The binding of activating cofactors (Ca2+, phospholipid, diacylglycerol/phorbol ester) induces a conformational change in the enzyme and the release of the pseudosubstrate from the substrate-binding pocket. Another PKC activation mechanism has been proposed where proteolysis occurs between the regulatory and kinase domains, producing the active (cofactor-independent) catalytic C-terminal fragment and an N-terminal regulatory fragment [2, 14, 151. However, there are few studies showing the appearance of the catalytic fragment of PKC in vivo [16- 181.

Recent reports demonstrated the cleavage of nPKCG and the production of its catalytic fragment during apoptosis induced by the treatment of U937 with ionizing radiation [I91 or DNA-damaging agents [20]. The proteolytic cleavage of nPKC6 is inhibited by the expression of the anti-apoptotic proteins Bcl-2 and Bcl-X, and pre-treatment with the interleukin-ID-converting enzyme (ICE)/caspase-1-inhibitor peptide, YVAD. These results suggest that the proteolytic activation of nPKCG may play a role in apoptosis.

Several lines of evidence reveal the importance of the ICE/ caspase family proteases in signal transduction leading to apoptosis [21-231. Apoptosis is induced by the overexpression of several ICE/caspase family protease genes, and viral or chemical inhibitors of these proteases inhibit apoptosis in various cells. To date, several proteins have been reported to be cleaved during apoptosis including poly(ADP-ribose) polymerase and nPKCG, although it remains unknown which of these substrates is responsible for the phenomena that occur during the apoptotic process [22-261.

In the present study, we addressed whether the generation of the nPKCG catalytic fragment observed in U937 during apoptosis induced by ionizing radiation or DNA-damaging agents is a general phenomenon associated with Fas-mediated apoptosis in a variety of cell lines, and examined the fate of the PKC isotypes expressed in leukemic cells during Fas-mediated and TPA-induced apoptosis.

MATERIALS AND METHODS

Cell culture and the induction of apoptosis. Human T lymphoma cells, Jurkat and HPB-ALL, and human myeloid leukemia cells, U937 and HL60, were maintained in RPMI 1640 medium (Gibco BRL) supplemented with 10% fetal bovine serum (ICN Biomedicals) using standard tissue culture procedures. Apoptosis was induced by treatment with 100 ng/ml anti-Fas antibody C H l l (MBL) in the presence (U937 and HL60) or absence (Jurkat and HPB-ALL) of 5 pglml cycloheximide. To examine the effect of protease-inhibitor peptides, cells were pretreated for 1 h with ‘100 pM or 500 pM Ac-DEVD-CHO, Ac- DMQD-CHO, leupeptin (Peptide Institute Inc.), or calpain inhibitor 1 (Nakarai Tesque Inc.), and subsequently treated with CHl l for 6 h. Apoptosis was quantitated by counting the num- bers of apoptotic and non-apoptotic cells, with the nuclear morphology visualized by fluorescence microscopy of cells stained with Hoechst 33258 (Polysciences Inc.).

Immunoblot analysis. After apoptotic treatment, the cells were collected by gentle centrifugation, washed 2 or 3 times with cold 10 mM sodium phosphate, 150 mM NaC1, pH 7.5 (NaCVP,) and frozen in liquid N,. Frozen cell pellets were suspended directly in Laemmli’s sample buffer [27] (0.5 - 1 X lo’ cells/ml), sonicated, and boiled. Aliquots (10 pl) of total cell extracts were fractionated by SDS-PAGE using 10% acrylamide gel and transferred to polyvinylidene fluoride membrane (ATTO corporation). After blocking with 5 % non-fat dry milk in wash- ing buffer (NaClIP,, pH 7.5, containing 0.1 % Tween 20), the membranes were probed with specific anti-PKC antibodies.

Anti-peptide antibodies against nPKCG (6SC) and 8 (BSC) were obtained from Santa Cruz and antibodies against nPKCE (EGB) and cPKCa (aGB) were from Gibco BRL. Monoclonal antibodies against the regulatory domains of nPKCG (STL), E

(ETL) and 0 (fRL) were purchased from Transduction Laborato- ries. Antisera against nPKC6 (65) [28], nPKCO (82/3-I) [29], and cPKCa (73) [30] were previously reported. Anti-cPKCpI (DIICl) and -aPKC[ ((Rb2) antibodies were raised against synthetic peptides corresponding to C-terminal sequences of cPKCp1 and aPKC[, respectively. The antibodies were diluted with the blocking solution (1:200 for 6SC, HSC, HTL, and ETL; 1 :500 for STL, 65, and aGB; 1: ‘1000 for EGB, 0213, lfiICl and (‘Rb2; 1 : 4000 for y3). The primary antibody complexes were detected using the ECL kit (Amersham) after treatment with the horseradish-peroxidase-conjugated anti-rabbit (Bio Source In- ternational Inc., Tago Products, 1 : 4000 dilution), anti-mouse (Amersham, 1 : 2000), and anti-guinea pig IgG (Vector Laborato- ries Inc., 1 : 2000) secondary antibodies.

Protease assay. Cells (1 X lo’) were incubated in 1 ml serum-free medium with 100 ng/ml of C H l l for the indicated times and immediately frozen in liquid N, after gentle centrifugation. Cell extracts were prepared by incubation for 30 min on ice in 500 p1 lysis buffer (25 mM Hepes, pH 7.5, 5 mM MgCI,, 5 mM EDTA, 1 mM phenylmethanesulfonyl fluoride, 50 pg/ml leupeptin, 20 pM phosphoramidon, 1.5 pM pepstatin A, 5 niM dithiothreitol), and subsequently diluted with 25 mM Hepes buffer (100 pg protein/ml) and incubated at 37°C for 30 min with 100 pM fluorescent substrates Ac-YVAD-Mec or Ac- DEVD-Mec (Peptide Institute Inc.). The preparation of total extracts and immunoblot analyses were performed as described above.

Expression vectors. The mammalian expression vectors for mouse nPKCG, M241 [28], and constitutive active nPKCG, DR144/145A [31], have been described elsewhere. The expression vector encoding a catalytic fragment of nPKCG (positions 328-674) was constructed with the SRD vector [30], synthetic oligo DNA linker (EcoRI site-CCACCATGAACAACGG- GACCTATGGCA-BglII site), and BglII-EcoRI fragment of M241. The kinase-defective mutant of the 6 catalytic fragment was generated from 6 cat by substitution of the BglII-PvuII fragment for that of kinase-defective nPKCG, DK376A [31]. The expression vector (pGFP) encoding marker protein, green fluorescent protein (GFP), was constructed with the human elonga- tion factor-la promoter and Not1 fragment of pGreen Lantern-1 (Gibco BRL).

Transfection and immunofluorescence microscopy. COS-1 cells were grown in Dulbecco’s modified Eagle’s medium (Nis- sui Pharmaceutical Co.) supplemented with 10% fetal calf serum (Gibco BRL). Cells were transfected by an electroporation method using 16 pg DNA for 6X106 cells according to the previous method [29]. After 24h of standard culture, cells were treated with trypsin and seeded at a density of 4x1O4 cells in each well of 24-well plates. Cells were grown in growth medium for another 24 h and fixed with 3.7% HCOH in NaCI/P,, pH 7.5, for 10 min. The cells were permeabilized by treatment with

Mizuno et al. (ELIY. J . B i o c k ~ ~ 250) 9

0.1 % Triton X-100 in NaCI/P, for 10 min, and blocked with 10% goat normal serum in NaCUP, containing 0.05% Tween 20. Immunofluorescence was performed with 65 (1 : 200 dilution) as primary antibody, fluorescein-isothiocyanate (F1TC)-conjugated anti-rabbit 1gG (EY Laboratories, Inc.) as the secondary antibody, and Hoechst 33258. Immunofluorescence was observed under a fluorescence microscope (Olympus, BX60) and analyzed with SenSys 0400 (Photometrics).

RESULTS Limited proteolysis of nPKCG during Fas-mediated apoptosis in human leukemic cell lines. Treatment of a variety of human leukemic cells with the anti-Fas monoclonal antibody, CH11, results in apoptosis as characterized by morphology, visualized by phase contrast microscopy, and nuclear morphology, as visualized by fluorescent microscopy, after staining with Hoechst 33258 [22]. We confirmed this in Jurkat cells, which showed apoptosis in 25 % and 58 % of cells within 3 h and 6 h, respectively, after C H l l treatment as judged by Hoechst staining. Another leukemic cell line, HPB-ALL, also showed a similar pattern of induction of apoptosis after treatment with CHI 1 (25 % and 64 % cells apoptotic after 3 h and 6 h, respectively). Other leukemic cell lines such as U937 and HL60 also showed apoptotic induction when the cells were treated with C H l l in the presence of cycloheximide (22% and 27% apoptotic cells 6 h after treatment for U937 and HL60 cells, respectively).

Previous studies have revealed the generation of a catalytic fragment of nPKCG in U937 cells during apoptosis induced by ionizing radiation, Fas-activation, tumor necrosis factor a, [ 191 or DNA-damaging agents [20]. To examine whether this limited proteolysis is a general or specific phenomenon during Fas-mediated apoptosis, we used the four human leukemic cell lines

A esc Jurkat HPB-ALL

'T 0 3 6"T 0 3 6' 10-

80-

50. W

34-

D EGB

105- 82 -

49 -

33.3 -

B BTL

- cat

p42 -

A 6SC B 65 c 6TL P C T O 3 6 P C T O 3 6 PC T 0 3 6 &as Ab (h)

80- I

50.-

34-

- r c a t

- 4 reg p38

D Juikat HPBALL HL60

PCT 0 3 6 T 0 3 6 T 0 3 6aFasAb(h) ---

B ~72174

cat p41

reg p38

Fig. 1. Fas-mediated apoptosis induces the cleavage of nPKCS and the production of its catalytic and regulatory fragments. U937 cells (A-C) and other leukemic cells (D) were treated with anti-Fas antibody, CHl1, for the indicated times and total cell extracts were prepared as described in the Materials and Methods section. nPKC6 and its derivatives were detected by immunoblot analyses using three independent antibodies against nPKCG. (D) Intact and regulatory fragment of nPKCG were detected using bTL and catalytic fragment of nPKC6 were observed using 6SC. To examine the down-regulation of PKC isotypes, cells were treated with 100 ng/ml TPA (Sigma) for 24 h (indicated as T). The putative catalytic (41 kDa) and regulatory (38 kDa) fragments of nPKCG are indicated as cat and reg, respectively, with an arrowhead. PC indicates the positive control, which was recombinant nPKCG expressed in COSl cells as previously reported [28]. Bars indicate the prestained markers (ma, Bio-Rad). The data shown are representative of five independent experiments.

Jurkat HPB-ALL Jurkat HPB-ALL I T 0 3 6"T 0 3 6'PC IT 0 3 6"T 0 3 6' aFasAb(h)

E ETL Jurkat HP 8-A LL HP B-A LL

I T 0 2 3 4 5 6 8"T 0 2 3 4 5 6 8' 0 3 6 aFas Ab (h)

- 4 reg p46-49

P ~ ~ ] cat 4 P36 -

4 reg p37

Fig.2. Proteolytic cleavage of nPKC6J (A-C) and E (D and E) during Fas-activated apoptosis. Jurkat and HPB-ALL cells were treated with anti-Fas antibody (CH11) for the indicated times and total extracts were prepared. Immunoblot analyses were performed using three anti-nPKCB (A, BSC; B, RTL; C, B2/3-1) and two anti-nPKCE (D, EGB; E, cTL) antibodies. T indicates TPA treatment (100 ng/ml, 24 h) and PC indicates recombinant protein expressed in COSl cells as positive control [29, 301. p41 catalytic (cat) and p37 regulatory (reg) fragments of nPKCB are indicated with arrowheads (A-C). Two catalytic fragments of nPKCI: (cat, 36 kDa and 44 kDa) were observed using &B (D) and regulatory fragments (reg, 46-49 kDa) were detected using cTL (E). The data shown are representative of three independent assays.

10 Mizuno et al. (Eur: J. Biochem. 250)

A Jurkat HPB-ALL

C P K C ~ IT 0 2 3 4 5 6 8"T 0 2 3 4 5 6 8' aFas Ab (h)

105- 82 - 4 c I

49 -

33.3-

B HPB-ALL U937 HL60 cPKCpll IT 0 3 6"T 0 3 6I'T 0 3 6 ' cIFasAb(h)

105- 82-

49-

33.3-

C HPB-ALL U937 HL60 aPKCC 'T 0 3 6'IT 0 3 6"T 0 3 6'

aFas Ab (h) 105- 82 - 49-

33.3-

Fig. 3. cPKCa (A), f l I (B) and aPKC< (C) were not affected by Fas- activated apoptosis. Cells were prepared as described in the legend of Figs 1 and 2. cPKCu (A) and PI (B) and *KC[ (C) were detected using anti-cPKCu (y3) , -cPKCflI (/j'IICl) and -aPKC[ ([RbZ), respectively. T indicates TPA treatment (100 nglml, 24 h). The data shown are representative of three independent assays.

listed above and analyzed nPKC6 and its derivatives in total cell lysates. For this purpose, the cells were suspended directly in SDS sample buffer and analyzed by western blotting using three different anti-nPKC6 antibodies. 6SC and 65 are antibodies raised against the C-terminal peptide of rat nPKC6 and dTL is a monoclonal antibody raised against the regulatory domain of rat nPKC6. Typical blots are shown in Fig. 1.

Two protein bands, 72 kDa and 74 kDa, correspond to intact nPKC6 for the following reasons : a) only these bands were usu- ally detected by all three antibodies; b) the bands showed the same mobilities as those of recombinant nPKC6 expressed in COSl cells (indicated as PC in Fig. 1) and endogenous nPKCG in rat 3Y1 fibroblasts (data not shown), where the slower migrating band corresponds to the phosphorylated form [32] ; c) treatment of cells with TPA for 24 h results in an upward-shift of the bands and a decrease in their intensity (Fig. 1, lane T). This is consistent with previous observations that TPA induces the autophosphorylation of nPKCG resulting in a mobility shift followed by down-regulation [32].

After treatment of the cells with the anti-Fas antibody CHI 1, the intensity of the 72/74-kDa bands declined within 6 h in U937 cells (Fig. 1A-C) and within 3 h in others (D). Concomitant to the decrease in the 72/74-kDa bands, 41-kDa (Fig. IA, B and D) and 38-kDa (Fig. 1 C and D) bands began to appear. The 41- kDa band most likely corresponds to a catalytic fragment of nPKCG because of its molecular size and because it is recognized by two independent antibodies raised against a C-terminal

peptide of nPKC6. This is consistent with the results observed in apoptotic U937 cells [ 19, 201 and indicates that the generation of a catalytic fragment of nPKCG during Fas-mediated apoptosis occurs in a variety of leukemic cell lines. Importantly, the 38- kDa band most likely corresponds to the regulatory fragment of nPKC6, because the antibody STL, but not antibodies 6SC and 65, recognize the band. These results raise the interesting possibility that the proteolytic cleavage of nPKCG during Fas-mediated apoptosis generates not only its catalytic fragment but also its regulatory fragment. It should also be noted that the treatment of cells with TPA for 24 h results in the appearance of both the 41-kDa and 38-kDa fragments of nPKCG, although the inten- sities of the bands differ significantly among the cell lines tested (Fig. 1, lane T). TPA treatment does not efficiently produce the catalytic fragments in Jurkat (Fig. 1 D), whereas treatment generates both the nPKC6 fragments in HPB-ALL, U937, and HL60 cells, as discussed below.

Limited proteolysis of all the nPKC isotypes expressed in human leukemic cells during Fas-mediated apoptosis. Leuke- mic cell lines express multiple PKC isotypes including cPKCa, cPKC,!?, nPKC6, nPKCH, nPKCc, and aPKC[ [33-351. Among them, nPKCU is structurally most closely related to nPKCG [29]. As previously described, nPKCG contains the sequence motif DMQD/N in the hinge region between the catalytic and regulatory domains and this might be the recognition/cleavage site for the ICE-like protease [19]. Interestingly, nPKCH also contains a similar sequence DEVDK (positions 351 -355, resembling the PARP cleavage site DEVDG [36]) in the hinge region between the kinase and regulatory domains. Thus, we examined the effect of Fas-mediated apoptosis on other PKC isotypes including nPKCH.

We used three independent antibodies for the identification of nPKCH. One of them, HSC was raised against the C-terminal peptide of human nPKCH, and recognizes the 72-kDa protein band migrating to the same position as COSI-cell-expressed recombinant mouse nPKCB, in extracts from Jurkat and HPB-ALL (data not shown), indicating that the band corresponds to nPKCH (Fig. 2A). Treatment of cells with TPA results in an upward- shift and a decrease in the intensity of the 72-kDa band, consistent with the notion that the 72-kDa band corresponds to nPKCB. It is also clear that treatment of cells with anti-Fas antibody results in a decline in the amount of the 72-kDa band and the appearance of 42-kDa bands, presumably corresponding to the catalytic fragment of nPKC0. The monoclonal anti-nPKCO antibody (BTL) was raised against bacterially synthesized regulatory fragment of nPKCO and recognizes the same 72-kDa nPKCB but not the 42-kDa fragment (Fig. 2B). Concomitant to the decline in the 72-kDa bands, 37-kDa bands began to appear. The other antibody (B2/3) was raised against a peptide corresponding to the sequence 332-355 of mouse nPKC0 [29]. As the putative cleavage site described above is located at the C-terminal end of the antigen peptide, this antibody seems to recognize the regulatory domain. Besides the decrease in 72-kDa nPKCQ, 82/3 recognized the appearance of the same 37-kDa band as that detected using BTL, suggesting that the 37-kDa protein corresponds to a regulatory fragment of nPKCB. The strong staining of the 37-kDa bands by 02/3 compared with the intact 72-kDa band and with staining by BTL might reflect the exposure of the antigenic sequence in the C-terminus of the 37-kDa putative regulatory domain, supporting the notion that the 37-kDa regulatory fragment results from proteolytic cleavage at the site, DEVD/K. It should also be noted that TPA treatment also results in the generation of the 37-kDa regulatory fragment (Fig. 2B and C) as well as the 42-kDa catalytic fragment (Fig. 2A).

Mizuno et al. (EUK J. Biochem. 250)

Nuclear fragmentation Protease activity

- - 8 0 -

. .- c , 0

- i 4 0 -

1 &

E ' 0

11

F - ' " ' . ' " .

0 1 2 3 4 5 0 1 2 3 4 5 nPKCS Time (h) nPKCE Time (h)

0 1 2 3 4 5 0- 0 1 2 3 4 5

Fig.4. Proteolysis of nPKC isotypes is accompanied by the activation of a CPP32-like activity. Time courses for nuclear fragmentation (A) and protease activity of ICE-like (0) and CPP32-like (0) protease (B) were measured using Jurkat cells treated with anti-Fas antibody CH11 in serum- free medium as described in the Materials and Methods section. The values are means of triplicate determinations. (C-F) Using the same cells, immunoblot analyses were performed with isotype-specific antibodies to examine time courses for the reduction of intact isozymes and the production of proteolytic fragments of nPKCG (C, GSC), e (D, EGB), and H (E, H2/3). The expression of cPKCa was not affected by Fas-activation (F, y3) . The amount of protein was measured by densitometric analysis of intact isozymes (0) and proteolytic fragment (0, p41 catalytic fragment of nPKC6, p44 catalytic fragment of nPKCe, and p37 regulatory fragment of nPKCH). The results were normalized against the quantity of intact PKC at time 0. The values are means of duplicate determinations. Insets show proteolytic fragment production of nPKCG (C), t' (D), and 0 (E) and intact cPKCa (F). The data shown are representative of three independent assays.

To identify nPKCe, we used two different antibodies. One, eGB, was raised against the C-terminal sequence of rat nPKCE and the other, ETL, was raised against the regulatory domain of rat nPKCE. Fig. 2 D shows a typical blot obtained using eGB, showing that the amounts of 90-kDa and 92-kDa nPKCc decrease whereas the 36-kDa and 44-kDa fragments, putative catalytic fragments, are generated in both Jurkat and HPB-ALL cells during Fas-mediated apoptosis. Similar results were also obtained using ETL, where 46-kDa to 49-kDa putative regulatory fragment bands were generated instead of the 36/44-kDa bands (Fig. 2 E). These results strongly suggest that proteolytic cleavage of nPKCE occurs and generates both the regulatory and the catalytic fragments during Fas-mediated apoptosis in the two cell lines. Similar proteolytic cleavage of nPKCe also occurred in U937 and HL60 after Fas activation (Fig. 8 C and D).

In contrast to the results for the three nPKC isotypes, the results on cPKCa and @I and aPKC[ showed no change in the pattern of the blots during Fas-mediated apoptosis (Fig. 3), consistent with the results in apoptotic U937 cells [19, 201. These results show that Fas-mediated apoptosis is accompanied by the

proteolytic cleavage of the nPKC isotypes nPKCG, nPKCB, and nPKCE, but not cPKCa and PI and aPKC[. Further, the cleavage of nPKC isotypes most likely generates both the catalytic and regulatory fragments.

Proteolytic cleavage of nPKC isotypes is accompanied by the activation of a CPP32kaspase-3-like protease. To clarify the physiological meaning of the proteolytic cleavage of nPKC isotypes during Fas-mediated apoptosis. we next compared the time course for the cleavage of nPKC isotypes with that for the activation of ICE and ICE-related proteases in Jurkat cells. Protease activities were measured using cell extracts against specific peptide substrates, YVAD-Mec for the ICE-like protease and DEVD-Mec for the ICE-related protease, the CPP32-like protease [37]. Fig. 4 shows a comparison of the percentage of apoptotic cells, as judged by nuclear morphology visualized by fluorescence microscopy (Fig. 4 A), with other parameters including protease activities (Fig. 4B) and the cleavage of nPKC isotypes (Fig. 4C-E). The treatment of Jurkat cells with anti-Fas antibody C H l l results in a fivefold activation of the CPP32-like

12 Mizuno et al. ( E m J . Biochern. 250)

i nn I I ""

80 h

60 0 0

0 .- c

40

2 20

0

B DEVD DMQD LLL Leu -- r l O O 500' '100 5OO''lOO 500' '100 500' nnnnnnnnn - + - + - + - + - + - + - + - + - +

intact 6

6 cat

6 reg

intact E

E cat Fig. 5. Cleavage of nPKC isotypes is inhibited by pretreatment of Jurkat with a peptide inhibitor, Ac-DEVD-CHO, of the CPP32-like protease. Jurkat cells were pretreated with 100 pM or 500 pM of each peptide for 1 h and apoptosis was induced by adding anti-Fas antibody, CH11, Counts of nuclear fragmented cells (A) and immunoblot analyses using 6SC (for catalytic fragment of nPKC6) and bTL (for intact protein and regulatory fragment of nPKC6) and eGB (for nPKCe) (B) were performed as described in the Materials and Methods section. LLL and Leu indicate calpain inhibitor-1 and leupeptin, respectively. The data shown are representative of two or three independent experiments.

protease while the ICE-like protease is unaffected (Fig. 4B), consistent with previous observations in human carcinoma cell lines [37]. The cleavage of nPKC isotypes was slightly induced after 2 h of CHl l treatment and the amounts of the catalytic fragments of nPKCG and E (indicated in the insets of Fig. 4C and D, respectively) and the regulatory fragment of nPKC6' (indicated in the inset of Fig. 4E) increased gradually. However, cPKCa did not change (Fig. 4F, inset indicating intact cPKCa). Importantly, the time course for the cleavage of all three nPKC isotypes also parallels those of Fas-mediated apoptosis and the activation of the CPP32-like proteases. Similar results were obtained for HPB-ALL (data not shown). These results support the idea that the cleavage of nPKC isotypes might play some role in Fas-mediated apoptosis.

A peptide inhibitor of the CPP32-like protease blocks the proteolytic cleavage of nPKC isotypes as well as apoptosis. The similarity in the time courses for the activation of the CPP32-like protease and the proteolytic cleavage of nPKC isotypes suggests the possible involvement of the CPP3Zlike protease in the proteolytic cleavage of nPKC isotypes. If this is the case, then the inhibition of the CPP32-like protease activity should suppress the proteolytic cleavage of nPKC isotypes. Fig. 5 shows the effect of protease inhibitors on Fas-mediated apoptosis in Jurkat. An inhibitor of the CPP32-like protease, Ac- DEVD-CHO [36, 371, almost completely suppresses Fas-mediated apoptosis in Jurkat cells (Fig. 5A), consistent with previous reports that the CPP32-like protease is involved in Fas-mediated apoptosis [37]. However, Ac-YVAD-CHO, an inhibitor of the

ICE-like protease, has no effect on Fas-mediated apoptosis (data not shown). An inhibitor, Ac-DMQD-CHO, synthesized based on the sequence of the cleavage site of nPKCG, also suppresses Fas-mediated apoptosis, suggesting that the protease responsible for the cleavage of nPKCG is also involved in the apoptotic process. Furthermore, Ac-DEVD-CHO and Ac-DMQD-CHO also suppress the cleavage of nPKCG, nPKCc (Fig. 5 B), and nPKC6' (data not shown). Based on these results and those in Fig. 4, it appears that the CPP32-like protease might be responsible for the proteolysis of all three nPKC isotypes i n Jurkat during Fas- mediated apoptosis.

The strong correlation between the degree of apoptosis and the proteolytic cleavage of nPKC isotypes shown for the two inhibitors extends also to the effect of calpain inhibitor-1 . Since calpain has been shown to cleave cPKC and nPKCE to generate both catalytic and regulatory fragments in vitro [15, 381, we examined the effect of calpain inhibitors, calpain inhibitor-I (Ac- LLL-CHO) and leupeptin, on the cleavage of nPKC isotypes as well as apoptosis mediated by Fas. As shown in Fig. 5, leupeptin fails to inhibit the cleavage of nPKC isotypes or apoptosis, consistent with the observation for nPKCG in U937 cells [19]. Cal- pain inhibitor-l also fails to inhibit completely, and slightly stimulates both the cleavage of nPKC isotypes and apoptosis mediated by Fas. Moreover, calpain inhibitor-1 causes apoptosis without CH11 treatment (Fig. 5A) . It has been reported that calpain inhibitor-I causes apoptosis in human prostate cancer cells 1391, while various calpain inhibitors prevent apoptosis [40- 421. Although the mechanisms of action of these agents remain to be clarified, it is clear that there is a very strong correlation

Mizuno et al. (Ectr: J. Biochcnz. 250) 13

60 I E 1 F - &at DRA

105-

82-

-- * DRA

4* + &at

33.3-

1 2 3 4 5 6 7 Fig. 6. Transfection of nPKCS catalytic-fragment (&at)-induced apoptosis in COSl cells. (A-D), COSl cells were transfected with &at (A and B), wild-type nPKCG (C), or &at-kn (D), visualized by immunofluorescent microscopy using 85 as primary antibody (A, C, and D, green) and Hoechst 33258 (B-D, blue), and analyzed by a SenSys 0400 instrument (Olympus). The bar indicates 20 pin. (E) pGFP (4 pg) was cotransfected into COSl cells with indicated amounts of control vector (-), wild-type nPKCG (WT), &at, DR144/14SA (DRA), or &at-kn (c-kn). The amounts of transfected DNA were equalized by adding vector DNA. The cells were stained with Hoechst 332.58 48 h after transfection. The percentage of apoptotic cells was determined by the number of green cells with condensed and rounded morphology (closed bar) or with condensed and fragmented nuclei (open bar) divided by the total number of green cells (about 200 cells). (F) The same samples as in E were analyzed by western blotting using 6.5 as primary antibody. The data shown are representative of two or three independent assays.

between the occurrence of apoptosis and the cleavage of nPKC isotypes (Fig. 5).

Induction of apoptosis by overexpression of nPKCG catalytic fragment. To clarify the role of the catalytic fragment of nPKCd in apoptosis, expression plasmid encoding nPKCG catalytic fragment, &at (encoding amino acids 328-674 of mouse nPKCG, corresponding to positions 330-673 of human nPKCG), was generated and transiently transfected into COSl cells. The expression of &at was detected by immunofluorescence using S5 as a primary antibody. COSl cells transfected with Scat showed a dramatic reduction in the number of FITC-positive cells, as

compared with cells transfected with a control vector, wild type nPKCG, or kinase-inactive mutant, &at-kn (data not shown). Transfection of nPKCG or &at-kn had little effect on cell morphology (Fig. 6C and D), although most of the cells transfected with &at showed condensed and rounded morphology (Fig. 6A). Hoechst staining of cells transfected with &at resulted in condensed and fragmented nuclei (Fig. 6B). However, transfection of nPKCd or &at-kn showed little effect on nuclear morphology (Fig. 6 C and D). Almost the same results were obtained using HeLa cells (data not shown). Our findings described above are good agreement with previous report by Ghayur et al. using HeLa and NIH3T3 cells [43].

14 Mizuno et al. (Eul: J . Biochem. 250)

HPB-ALL Jurkat

- E. - 40 - a 0 0 .- c 2 20 a 0 a U

6 6 12 24 6 6 12 24 F A Time(h) + - . . + . - - . - aFasAb + - - - + - - - aFasAb

+ TPA + TPA

u937 HL60 100 80

5 6 0 - 80 - - - . - 2 60 a

; 40 u .- c a g 20 a

6 6 12 24 6 6 12 24 F A Time(h) + - - - + _ _ . _ - aFasAb

0

+ TPA

Fig. 7. TPA inhibits apoptosis mediated by Fas activation in Jurkat cells but activates in other cell lines. Cells were treated with 50 ng/ml TPA in the presence or absence of anti-Fas antibody for the indicated times. Floating (F) and adherent (A) cells were separated after treatment with TPA for 24 h. Apoptotic cells were counted using a fluorescence microscope. The data shown are representative of two or three independent experiments.

Transfection of nPKCH catalytic fragment, &at (encoding amino acids 355-707 of mouse nPKCB), or constitutively active nPKCG mutant, DR144/145A [30], also resulted in apoptotic morphology and nuclei (data not shown). To examine the dose/ response of induction of apoptosis by nPKCG mutants, we used the pGFP-cotransfection assay where each expression plasmid was cotransfected with pGFP and the cells expressing GFP were evaluated by means of morphology (Fig. 6E). The proportion of cell with rounded shape and fragmented nuclei increased (Fig. 6E) with an increase in the amount of Gcat or DR144/145A expressed (Fig. 6F) Transfection of wild-type nPKCG or Gcat- kn, however, could not induce apoptosis (Fig. 6E) in spite of their extensive expression (data not shown). These results suggest that an excess expression of kinase activity of nPKCG might cause apoptosis. Since transfection of &at as well as DR144/ 145A enhanced the induction of the TPA-response-element -lu- ciferase reporter gene [31, 321 in HeLa cells, &at is active without cell stimulation (data not shown).

TPA-induced apoptosis accompanies the proteolytic cleavage of nPKC isotypes but not their down-regulation. Previous studies have demonstrated that the treatment of a variety of cells with TPA results in the translocation followed by the down-regulation of cPKC and nPKC isotypes 132-35, 44-46]. However, there have been few demonstrations of the transient appearance of the catalytic and regulatory fragments of any of the PKC isotypes during TPA-induced down-regulation [ 16, 17, 471. As noted above, however, the treatment of cells with TPA for 24 h results in the accumulation of the same nPKC fragments as those generated by anti-Fas antibody CHI 1 (Figs 1 and 2). Phase contrast and fluorescent microscopy confirmed that such treatment actually causes apoptosis (data not shown). Previous reports demonstrated that TPA treatment of U937 and HL60 attenuated

the apoptosis induced by various stimuli [12, 48-50], but that exposure to TPA alone resulted in apoptosis in some of the cells [49, 501. However, sensitivity to TPA differs depending on the sublines of cells [S, 49, 501. Importantly, the occurrence of apoptosis in cells treated with TPA roughly correlates with the intensity of the bands generated by the cleavage of nPKC isotypes (Figs 1 and 2). It seems that there are two different fates for nPKC isotypes after TPA treatment, down-regulation and proteolytic cleavage to generate both the catalytic and regulatory fragments. Down-regulation is the dominant fate in Jurkat cells, while proteolytic cleavage dominates in other cells as shown in Figs 1, 2, and 8. Although the mechanism to select which of the two fates the nPKC isotypes meets after TPA treatment remains unknown, the cell type specificity of the effect of TPA on apoptosis seems to be involved in this determination.

To confirm this assumption, we next examined the effect of TPA on Fas-mediated apoptosis. The addition of TPA and anti- Fas antibody CHl l to Jurkat cells results in a 94% inhibition of Fas-mediated apoptosis (Fig. 7A), consistent with previous observations [ 5 ] . The proteolytic cleavage of nPKC isotypes as visualized by the appearance of nPKC fragments induced by C H l l is almost completely inhibited by the addition of TPA during CHI 1 treatment (Fig. 8A, lanes 1 and 5) . Further, TPA also attenuates the spontaneous apoptosis induced by high cell densities after 24 h incubation (Fig. 7A) and the generation of nPKC fragments is suppressed by the addition of TPA (Fig. 8A, lanes 4 and 8). In the case of HPB-ALL, TPA inhibits Fas-mediated apoptosis to 45% (Fig. 7B). Consistent with these results, TPA fails to suppress completely the generation of nPKC fragments (Fig. SB, lanes 1 and 5). In U937 and HL60 cells, the inhibition of Fas-mediated apoptosis by TPA is only 14% and 0%, respectively (Fig. 7 C and D), and TPA suppresses the generation of nPKC fragments slightly but not completely (Fig. 8C

Mizuno et al. ( E m J. Biochem. 250)

A Jurkat B HPB-ALL c u937 D HL60

*A *A -A -A 6 6 12246 6 1224F A 6 6 12246 6 1224 6 612 2 4 6 6 1 2 2 4 F A 6 6 1 2 2 4 6 6 1224F A + - - - + - _ _ _ - + - - - + - - - + - - - + - - - - - + - - - + _ . _ _ .

intact

nPKCij[ ;t

1 2 3 4 5 6 7 6 9 1 0 1 2 3 4 5 6 7 8 9 1 0

15

TP A Time (h) aFas Ab

8

E

1 2 3 4 5 6 7 8 9 1 0 1 2 3 4 5 6 7 8

Fig. 8. TPA induces the accumulation of nPKC fragments in HPB-ALL, U937, and HL60, but not in Jurkat cells. Cells were the same as described in the legend of Fig. 7. Immunoblot analyses were performed using nPKC6 (6SC for intact 6 in B-D and catalytic fragment; and dTL for intact 6 in A and regulatory fragment in A and B), anti-nPKCe (EGB), and anti-nPKCB (OX) antibodies. The data shown are representative of two or three independent experiments

and D, lanes 1 and 5) . Furthermore, TPA alone induces apoptosis (Fig. 7B-D) and the appearance of nPKC fragments in HPB- ALL, U937, and HL60 (Fig. 8B-D, lanes 4 and 8).

Previous studies have demonstrated that the treatment of cells with TPA results in the growth arrest of Jurkat cells [46] and the differentiation of U937 [44] and HL60 cells [Sl]. 6 h after TPA treatment, some of these cells become attached to the dishes. The floating cells and the attached cells were separated after 24 h of TPA treatment and examined for their apoptotic phenotype (Fig. 7) and the cleavage of nPKC isotypes (generation of nPKC fragments) (Fig. 8). Fig. 7 shows that floating cells are rich in apoptotic cells in all the cell lines, especially U937 and HL60, whereas apoptotic cells are a minor population among attached cells. Blots of Jurkat cell lysates indicate a clear difference between the floating and adherent cells in terms of their generation of nPKC fragments, i.e. floating cells, rich in apoptotic cells, contain significant amounts of nPKC fragments whereas adherent cells do not (Fig. 8, lanes 9 and 10). This is most striking for U937 and HL60 cells. These results support the possibility that there are two possible fates for cells after TPA treatment, apoptosis accompanied by the cleavage of nPKC isotypes, and differentiation accompanied by the down-regulation of nPKC isotypes (Fig. 8) and cPKCa and /?I1 (Fig. 3A and B).

DISCUSSION

Emoto et al. has reported a very interesting observation that the treatment of U937 cells with ionizing radiation and DNA- damaging agents results in the appearance of myelin basic protein kinase activity [19, 201. The purified myelin basic protein kinase is the catalytic fragment of nPKCG whose N-terminal sequence starts from N331 in the nPKCS sequence DMQDN. Ionizing radiation and treatment with 1 -P-D-arabinofuranosyky- tosine results in the disappearance of intact nPKCG and the generation of the catalytic fragment, while other PKC isotypes such as cPKCa, cPKCP, nPKCc, and aPKCi do not change. This raises the intriguing possibility of the specific involvement of the catalytic fragment of nPKCG in apoptosis induced by these stimuli. Thus, we examined the specificity and generality of the fragmentation of nPKCG using several leukemic cell lines during Fas-mediated apoptosis.

Fas-mediated apoptosis results in proteolytic cleavage of nPKCG with the generation of a kinase fragment in all the leukemic cell lines tested including U937 (Fig. 1). The proteolytic cleavage of nPKCG results in the generation of the putative regu-

latory domain (Fig. 1 C and D) in addition to the kinase domain, suggesting the possibility that the regulatory domain of nPKC isotypes might also be involved in apoptotic signaling. The anti- nPKCG antibody used by Emoto et al. was GSC, which recognizes only the C-terminal sequence of nPKCG, resulting in the failure to detect the regulatory domain. Further, in contrast to Emoto's results, the proteolytic cleavage to generate both the kinase and regulatory domains was found to occur for all nPKC isotypes expressed in the cell lines tested, including nPKCG, nPKCE, and nPKC0 (Figs 1 and 2). cPKCa, and PI, and aPKC< remain intact during Fas-mediated apoptosis (Fig. 3). This apparent inconsistency seems to reflect the difference in the detec- tion protocol for nPKC fragments. Emoto et al. used a fraction of the Q-Sepharose eluate of the cell lysate supernatant fraction, instead of the total cell lysate to analyze the fate of PKC isotypes, and so failed to detect the nPKCE fragments. In fact, the proteolytic fragments of the nPKC isotypes, especially the catalytic fragments, are recovered in the particulate rather than the supernatant fraction (data not shown). Further, Emoto et al. did not check the fate of nPKC0. It is also possible that the contra- diction is due to the difference in the apoptosis inducer used or the sublines of U937 cells.

The accumulation of nPKC fragments correlates well with apoptosis: (a) The time course for the production of nPKC fragments is almost the same as the time course for apoptosis as determined by fluorescent microscopy of cell nuclei (Fig. 4). (b) The proteolytic cleavage of nPKC is accompanied by the activation of a CPP32-like protease (Fig. 4) and inhibited by pre-treatment with a DEVD peptide inhibitor (Fig. S) , suggesting that the CPP3Zlike protease might be involved in the cleavage not only of nPKCG but also nPKCE and 8. (c) The sensitivity to protease inhibitors coincides with the proteolytic cleavage of nPKC and apoptosis (Fig. 5). (d) In Jurkat, TPA inhibits apoptosis and the production of nPKC fragments resulting from Fas activation, while it stimulates both in HPB-ALL, U937, and HL60 (Figs 7 and 8). (e) After treatment with TPA for 24 h, the accumulation of nPKC fragments can be detected in apoptotic cells (floating cells), but not in non-apoptotic cells (adherents cells) (Fig. 8). (f) Transfection of expression vector for catalytic fragment of nPKCG to COSl cells resulted in rounded cell shape and condensed and fragmented nuclei, exhibiting typical apoptotic morphology (Fig. 6). Taken together, these results support the idea that the proteolytic cleavage of nPKC isotypes might be a criti- cal step involved in the transduction of the apoptotic signal.

A decline in nPKC isotype levels and the generation of fragments was detected 2 h after Fas-activation (Fig. 4), indicating this is not an early response to apoptotic stimuli. Pre-treatment


of Jurkat with 100 pM Ac-DEVD-CHO inhibits the production of the proteolytic fragments of nPKC and nuclear fragmentation, a late apoptotic event, but does not inhibit nuclear condensation, an early apoptotic event (data not shown). These results suggest that the proteolytic cleavage of nPKC occurs after commitment to apoptosis and that the generation of proteolytic fragments is involved in the transition after commitment to the execution stage of apoptosis. There are reports suggesting a positive role of PKC in the late stages of apoptosis. Rusnak and Lazo have reported that a depletion of PKC by phorbol-12,13-dibutyrate treatment attenuates VP-16-induced apoptosis in DU-145 human carcinoma cells [13]. This inhibition by PKC depletion occurs during the transition between the pre-apoptotic and apoptotic stage, because chronic exposure to phorbol-12,13-dibutyrate inhibits the generation of short DNA fragments but not that of long DNA fragments by VP-16. These results, consistent with our findings, suggest that PKC is involved in the late stages of the apoptotic process. Another example of proteolytic cleavage occurring at a late stage of apoptosis is retinoblastoma protein, which is cleaved into p68 and p48 fragments in VP-16-induced apoptosis during the transition between the commitment and execution stages [25].

The generation of both the regulatory and kinase fragments of all nPKC isotypes suggests their possible involvement in apoptosis. Based on their size, the kinase fragments of nPKCe and nPKCO most likely possess kinase activity as indicated for the kinase fragment of nPKCG [19, 20, 431. As shown in Fig. 6, transfection of &at or &at resulted in apoptosis in COSl and HeLa cells, suggesting that the production of catalytic fragment of nPKC isozymes might not be byproducts but one of the key processes in Fas-mediated apoptosis. Wild-type nPKCG or &at- kn could not induce apoptosis in spite of extensive expression, suggesting that induction of apoptosis by &at or DR144/145A might depend on their kinase activity, The inducibility of apoptosis by &at was greater than that by DR144/145A despite its lower protein level (Fig. 6 E and F). That is not due to potent kinase activity of Gcat compared to DR144/145A. These results support the idea that &at plays a specific role besides having high kinase activity which was observed when DR144/145A was transfected. The catalytic fragment of rat brain PKC obtained by proteolysis using trypsin has been reported to exhibit an altered substrate specificity toward smooth muscle myosin light chain in vitro [52]. The regulatory domain of nPKCG has been shown to be involved in the recognition of the protein substrate, myristoylated alanine-rich C kinase substrate [53]. Thus, it is quite possible that the kinase fragment possesses a substrate specificity different from that of intact nPKC in leukemic cells and hence causes a distinct fate in cells.

On the other hand, there are several lines of evidence to suggest that the regulatory domain of PKC contains the phorbol ester and phospholipid-binding activity [54, 551, and that PKC interacts with proteins including substrates in part through the regulatory domain [53, 56, 571. This suggests the possibility that the regulatory fragment is also involved in the signal transduction systems of PKC. The regulatory fragment of aPKCl has been shown to act in a dominant negative manner against the intact isotype (581. In any case, the target of the nPKC fragments remains to be clarified.

Our study clarified that a CPP32-like protease is mostly responsible for the proteolytic cleavage of nPKC isotypes, including nPKCe (Figs 4 and 5) . Although nPKCG and nPKC0 contain a sequence in their hinge regions that can be recognized by CPP32-like proteases, there is no typical sequence motif in the hinge region of nPKCe. It has been reported that the limited proteolysis of cPKC isotypes and nPKCE by calpain in vitro generates an active catalytic fragment and the regulatory fragment

115, 381, although the calpain inhibitors fail to protect against nPKC cleavage (Fig. 5B), suggesting no involvement of calpain in the cleavage of nPKC during Fas-mediated apoptosis. The possible involvement of calpain in the down-regulation of PKC after TPA treatment has been demonstrated, since calpain inhibitors suppress down-regulation and the transient production of the catalytic fragment [17, 181. We have demonstrated that the treatment of leukemia cell lines with TPA induces the down- regulation or limited fragmentation of PKC, although the protease(s) involved i n the cleavage of PKC isotypes remains to be determined. Is the CPP32-like protease involved in the generation of the catalytic fragment of nPKC during TPA-induced apoptosis? What is the element responsible for the difference between complete digestion and the limited proteolysis of nPKC by TPA? Are distinct proteases involved in these processes, or does the same protease cleave nPKC in both processes while the inhibition of proteolysis occurs only during the latter process?

There have been numerous reports dealing with the effects of TPA on the fate of PKC isotypes. In most cases, TPA causes the disappearance of all cPKC and nPKC isotypes, whereas the degree and time course differ depending on the isotype and cell type. Despite numerous efforts to detect the catalytic fragment of PKC recognized by specific antibodies during down-regulation induced by TPA, the catalytic fragment has rarely been found (16-181. Shea et al. reported the generation of the catalytic fragment of cPKCa induced by the treatment of SH-SY5Y cells with calcium ionophore but not TPA, although the catalytic fragment once formed was degraded approximately ten times faster than intact cPKCa [ 181. Importantly, no accumulation of intermediate proteolytic products was observed during the down-regulation of PKC, in clear contrast to the proteolytic cleavage of accumu- lating kinase and regulatory fragments shown in the present study. For example, TPA causes the differentiation of U937 [44] and HL60 cells [51]. In differentiated cells, the PKC isotypes are down-regulated (Fig. 8, lane lo), although a portion of the TPA-treated cells did not differentiate but underwent apoptosis as judged by nuclear fragmentation showing proteolytic cleavage generating the kinase and regulatory fragments (Fig. 8, lane 9). These results indicate that nPKC isotypes can meet two different fates, proteolytic cleavage or down-regulation. This suggests the intriguing possibility that the selection between these two fates is in involved in the determining cell fate, i.e. apoptosis or differentiation.

This work was supported in part by grants from the Ministry of Education, Science, Sports, and Culture of Japan, from the Cell Science Foundation, and from the Uehara Foundation.

REFERENCES 1. Nishizuka, Y. (1992) Intracellular signaling by hydrolysis of phos-

pholipids and activation of protein kinase C, Science 258, 607- 614.

2. Dekker, L. V. & Parker, P. J. (1994) Protein kinase C - a question of specificity, Trends Biochem. Sci. 19, 73-77.

3. Lucas, M., Sanchez, M. V., Sanz, A. & Solano, F. (1994) Protein kinase C activation promotes cell survival in mature lymphocytes prone to apoptosis, Biochern. Pharmacol. 47, 667-672.

4. Jarvis, W. D., Turner, A. J., Povirk, L. F., Traylor, R. S. & Grant, S. (1994) Induction of apoptotic DNA fragmentation and cell death in HL-60 human promyelocytic leukemia cells by pharmacological inhibitors of protein kinase C, Carzcrr Res. 54, 1707- 1714.

5. Cuvillier, O., Pirianov, G., Kleuser, B., Vanek, P. G., Coso, 0. A,, Gutkind, J. S . & Spiegel, S. (1996) Suppression of ceramide-mediated programmed cell death by sphingosine 1 -phosphate, Nature 381, 800-803.

Mizuno et al. (Eur: J. Biochem. 250) 17

6. Jin, L. W., Inaba, K. & Saitoh, T. (1992) The involvement of protein kinase C in activation-induced cell death in T-cell hybridoma, Cell. Immunol. 144, 217-227.

7. de Vente, J. E., Kukoly, C. A., Bryant, W. O., Posekany, K. J., Chen, J., Fletcher. D. J., Parker, P. J., Pettit, G. J., Lozano, G., Cook, P. P. & Ways, D. K. (1995) Phorbol esters induce death in MCF-7 breast cancer cells with altered expression of protein kinase C isoforms, J. Clin. Invest. 96, 1874- 1886.

8. de Vente, J., Kiley, S., Garris, T., Bryant, W., Hooker, J., Posekany, K., Parker, P., Cook, P., Fletcher, D. & Ways, D. K. (1995) Phor- bol ester treatment of U937 cells with altered protein kinase C content and distribution induces cell death rather than differentiation, Cell Growth Diffeer: 6, 371-382.

9. Chen, C. Y. & Faller, D. V. (1995) Direction of p21r"'-generated signals towards cell growth or apoptosis is determined by protein kinase C and Bcl-2, Oncogene 11, 1487-1498.

10. Jarvis, W. D., Povirk, L. F., Turner, A. J., Traylor, R. S., Gewiritz, D. A., Pettit, G. R. & Grant, S. (1994) Effects of bryostatin 1 and other pharmacological activators of protein kinase C on ~-[P-D- arabinofuranosyl]cytosine-induced apoptosis in HL-60 human promyelocytic leukemia cells, Biochem. Pharmacol. 47, 839- 852.

11. Walker, P. R., Kwast, W. J., Gourdeau, H., Leblanc, J., Neugebauer, W. & Sikorska, M. (1993) Relationship between apoptosis and the cell cycle in lymphocytes: roles of protein kinase C, tyrosine phosphorylation, and API, Exp. Cell Rex 207, 142-151.

12. Jarvis, W. D., Fornari, F. A,, Browning, J . L., Gewirtz, D. A., Koles- nick, R. N. & Grant, S. (1994) Attenuation of ceramide-induced apoptosis by diglyceride in human myeloid leukemia cells, J . Biol. Chem. 269, 31 685-31692.

13. Rusnak, J. M. & Lazo, J. S. (1996) Downregulation of protein kinase C suppresses induction of apoptosis in human prostatic carcinoma cells, Exp. Cell Res. 224, 189-199.

14. Sando, J. J. , Maurer, M. C., Bolen, E. J. & Grisham, C. M. (1992) Role of cofactors in protein kinase C activation, Cell. Signaling

15. Kishimoto, A,, Mikawa, K., Hashimoto, K., Yasuda, I., Tanaka, S., Tominaga, M., Kuroda, T. & Nishizuka, Y. (1989) Limited proteolysis of protein kinase C subspecies by calcium-dependent neu- tral protease (calpain), J. B i d . Chem. 264, 4088-4092.

16. Pontremoli, S., Michetti, M., Melloni, E., Sparatore, B., Salamino, F. & Horecker, B. L. (1990) Identification of the proteolytically activated form of protein kinase C in stimulated human neutro- phils, Pmc. Nut1 Acad. Sci. USA 87, 370553707,

17. Ai, Z. & Cohen, C. M. (1993) Phorbol 12-myristate 13-acetate-stimulated phosphorylation of erythrocyte membrane skeletal protein is blocked by calpain inhibitors: possible role of protein kinase M, Biochem. J. 296, 675-683.

18. Shea, T. B., Beermann, M. L., Griffin, W. R. & Leli, U. (1994) Degradation of protein kinase Cn and its free catalytic subunit, protein kinase M, in intact human neuroblastoma cells and under cell-free conditions. Evidence that PKM is degraded by mM calpain-mediated proteolysis at a faster rate than PKC, FEBS Left. 350, 223 - 229.

19. Emoto, Y., Manome, Y., Meinhardt, G., Kisaki, H., Kharbanda, S., Robertson, M., Ghayur, T., Wong, W. W., Kamen, R., Weichsel- baum, R. & Kufe, D. (1995) Proteolytic activation of protein kinase Cd by an ICE-like protease in apoptotic cells, EMBO J. 14,

20. Emoto, Y., Kisaki, H., Manome, Y., Kharbanda, S. & Kufe, D. (1996) Activation of protein kinase Cd in human myeloid leukemia cells treated with 1 -B-D-arabinOfUranoSylCytOSine, Blood 87, 1990- 1996.

21. Alnemri, E. S., Livingston, D. J., Nicholson, D. W., Salvesen, G., Thornberry, N. A., Wong, W. W. & Yuan, J. (1996) Human ICE/ CED-3 protease nomenclature, Cell 87, 171.

22. Whyte, M. (1996) ICE/CED-3 proteases in apoptosis, Trends Cell Biol. 6, 245-248.

23. Fraser, A. & Evan, G. (1996) A license to kill, Cell 85, 781-784. 24. Martin, S. J., O'Brien, G. A,, Nishioka, W. K., McGahon, A. J.,

Mahboubi, A,, Saido, T. C. & Green, D. R. (1995) Proteolysis of fodrin (non-erythroid spectrin) during apoptosis, J. B i d . Chewz.

4 , 595-609.

6148-6156.

270, 6425-6428.

25. An, B. & Dou, Q. P. (1996) Cleavage of retinoblastoma protein during apoptosis : an interleukin 1P-converting enzyme-like protease as candidate, Cancer Res. 56, 438-442.

26. Liu, X., Zou, H., Slaughter, C. & Wang, X. (1997) DFF, a hetero- dimeric protein that functions downstream of caspase-3 to trigger DNA fragmentation during apoptosis, Cell 89, 175 - 184.

27. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nuture 227, 680-685.

28. Mizuno, K., Kubo, K., Saido, T. C., Akita, Y., Osada, S., Kuroki, T., Ohno, S. & Suzuki, K. (1991) Structure and properties of a ubiquitously expressed protein kinase C, nPKC6, Eur: J. Biochem.

29. Osada, S., Mizuno, K., Saido, T. C., Suzuki, K., Kuroki, T. & Ohno, S. (1992) A new member of the protein kinase C family, nPKC0, predominantly expressed in skeletal muscle, M d . Cell. Bid. 12,

30. Ohno, S., Akita, Y., Konno, Y., Imajoh, S. & Suzuki, K. (1988) A novel phorbol ester receptor/protein kinase, nPKC, distantly related to the protein kinase C family, Cell 53, 731 -741.

31. Hirai, S., Izumi, Y., Higa, K., Kaibuchi, K., Mizuno, K., Osada, S., Suzuki, K. & Ohno, S. (1994) Ras-dependent signal transduction is indispensable but not sufficient for the activation of APl/Jun

32. Ohno, S., Mizuno, K., Adachi, Y., Hata, A., Akita, Y., Akimoto, K., Osada, S., Hirai, S. & Suzuki, K. (1994) Activation of novel protein kinase C6 and CE upon mitogenic stimulation of quiescent rat 3Y1 fibroblasts, J. Bid. Chem. 26Y, 17495-17501.

33. Tsutsumi, A., Kudo, M., Fuji, H., Freire, M. J., Turck, C. W. & Ransom, J. T. (1993) Regulation of protein kinase C isoform proteins in phorbol ester-stimulated Jurkat T lymphoma cells, J. Im- munol. 150, 1746-1754.

34. Baier, G., Baier, B. G., Meller, N., Coggeshall, K. M., Giampa, L., Telford, D., Isakov, N. & Altman, A. (1994) Expression and biochemical characterization of human protein kinase-0, Eur: J . Bio- chem. 225, 195-203.

35. Baier, G., Telford, D., Giampa, L., Coggeshall, K. M., Baier, B. G., Isakov, N. & Altman, A. (1993) Molecular cloning and chardcter- ization of PKC-0, a novel member of the PKC gene family expressed predominantly in hematopoietic cells, J. Bid. Chem. 268,

36. Lazebnik, Y. A,, Kaufmann, S. H., Desnoyers, S., Pokier, G. G . & Earnshaw, W. C. (1994) Cleavage of poly(ADP-ribose) polymerase by a proteinase with properties like ICE, Nature 371, 346- 347.

37. Hasegawa, J., Kamada, S., Kamiike, W., Shimizu, S., Imazu, T., Matsuda, H. & Tsujimoto, Y. (1996) Involvement of CPP32/ Yama(-like) proteases in Fas-mediated apoptosis, Cancer Rex 56,

38. Saido, T. C., Mizuno, K., Konno, Y., Osada, S., Ohno, S. & Suzuki, K. (1992) Purification and characterization of protein kinase CE from rabbit brain, Bio~henii.\tr). 31, 482-490.

39. Zhu, W., Murtha, P. E. & Young: C. Y. F. (1995) Calpain inhibitor- induced apoptosis in human prostate adenocarcinoma cells, Bio- chem. Biophys. Res. Comnz~m. 214, 11 30- 1 137.

40. Bruno, S., Lassota, P., Giarette, W. & Dai-zynkiewicz, Z. (1992) Apoptosis of rat thymocytes triggered by prednisolone, campto- thecin, or teniposide is selective to GO cells and is prevented by inhibitors of proteases, Oncol. Res. 4, 29-35.

41. Squier, M. K. T., Miller, A. C. K., Malkinson, A. M. & Cohen, J. J. (1994) Calpain activation in apoptosis, J . Cell. Physiol. 159, 229- 237.

42. Sarin, A,, Clerici, M., Blatt, S. P., Hendrix, C. W., Shearer, G. M. & Henkart, P. A. (1994) Inhibition of activation-induced programmed cell death and restoration of defective immune responses of HIV ' donors by cysteine protease inhibitors, J. Immu- nol. 153, 862-872.

43. Ghayur, T., Hugunin, M., Talanian, R. V., Ratnofsky, S., Quinlan, C., Emoto, Y., Pandey, P., Datta, R., Huang, Y., Kharbanda, S., Allen, H., Kamen, R., Wong, W. & Kufe, D. (1996) Proteolytic activation of protein kinase C6 by an ICEKED3-like protease induces characteristics of apoptosis, J. Exp. Med. 184, 2399- 2404.

44. Ways, D. K., Dodd, R. C. & Earp, H. S. (1987) Dissimilar effects of phorbol ester and diacylglycerol derivative on protein kinase

202, 931 -940.

3930 - 3938.

by PKC6, EMBO J. 13, 2331 -2340.

4997-5004.

1713-1718.


activity in the monoblastoid U937 cell, Cancer Rex 47, 3344- 3350.

45. Ways, K., Riddle, R., Ways, M. & Cook, P. (1991) Effect ofphorbol esters on cytosolic protein kinase C content and activity in the human monoblastoid U937 cell, J. B id . Chem. 266, 1258-1264.

46. Isakov, N., McMahon, P. & Altman, A. (1990) Selective post-tran- scriptional down-regulation of protein kinase C isoenzymes in leukemic T cells chronically treated with phorbol ester, J. Biol. Chem. 265, 209 1-2097.

47. Fournier, A. & Murray, A. W. (1987) Application of phorbol ebter to mouse skin causes a rapid and sustained loss of protein kinase C, Nature 330, 761-769.

48. Masuda, Y., Yoda, M., Ohizumi, H., Aiuchi, T., Watabe, M., Nakajo, S. & Nakaya, K. (1997) Activation of protein kinase C prevents induction of apoptosis by geranylgeraniol in human leukemia HL60 cells, hi. J. Canccr 71, 691 -697.

49. Hass, R., Meinhardt, G., Hadarn, M. & Bartels, H. (1994) Character- ization of human TUR leukemia cells: continued cell cycle pro- gression in the presence of phorbol ester is associated with resis- tance to apoptosis, Eur J. Cell Bid . 65, 408-416.

SO. Aoki, K., Nakashima, H., Hattori, T., Shiokawa, D., Ni-imi, E., Tanimoto, Y., Maruta, H., Uchiumi, F., Kochi, M., Yamamoto, N. & Tanuma, S. (1994) Sodium benzylideneascorbate induces apoptosis in HIV-replicating U1 cells, FEBS Lett. 351. 105-108.

51. Huberman, E. & Callaham, M. F. (1979) Induction of terminal differentiation in human promyelocytic leukemia cells by tumor-pro- rnorting agents, Proc. Nut1 Acad. Sci. USA 76, 1293 - 1297.

52. Nakabayashi, H., Sellers, J. R. & Huang, K. P. (1991) Catalytic fragment of protein kinase C exhibits altered substrate specificity

toward smooth muscle myosin light chain, FEBS Lett. 294, 144- 148.

53. Fujise, A,, Mizuno, K., Ueda, Y., Osada, S., Hirai, S., Takayanagi, A, , Shimizu, N., Owada, M. K., Nakajima, H. & Ohno, S. (1994) Specificity of the high affinity interaction of protein kinase C with a physiological substrate, myristoylated alanine-rich protein kinase C substrate, J. B id . Chem. 269, 31 642-31 648.

54. Lee, M. H. & Bell, R. M. (1986) The lipid binding, regulatory domain of protein kinase c. A 32-kDa fragment contains the calcium- and phosphatidylserine-dependent phorbol diester binding activity, J . Biol. Chern. 261, 14867-14870.

55. Huang, K. P. & Huang, F. L. (1986) Conversion of protein kinase C from a Ca2'-dependent to an independent form of phorbol ester- binding protein by digestion with trypsin, Biochern. Biophys. Rex Commun. 139, 320- 326.

56. Liao, L., Hyatt, S. L., Chapline, C. & Jaken, S. (1994) Protein kinase C domains involved in interactions with other proteins, Biochem- isrp 33, 1229-1233.

57. Diaz-Meco, M. T., Municio, M. M., Frutos, S., Sanchez, P., Lozano, J., Sanz, L. & Moscat, J. (1996) The product of pur-4, a gene induced during apoptosis, interacts selectively with the atypical isoforms of protein kinase C, Cell 86, 777-786.

58. Akimoto, K., Takahashi, R., Moriya, S., Nishioka, N., Takayanagi, J., Kimura, K., Fukui, Y., Osada, S., Mizuno, K., Hirai, S., Kaz- lauskas, A. & Ohno, S. (1996) EGF or PDGF receptors activate atypical PKCi through phosphatidylinositol 3-kinase, EMBO J. 15, 788-798.

Documents

The Proteolytic Cleavage of Protein Kinase C Isotypes, Which Generates Kinase and Regulatory Fragments, Correlates with Fas-Mediated and 12-O-Tetradecanoyl-Phorbol-13-Acetate-Induced