- Email: [email protected]
Protein kinase C activation modulates pro- and anti-apoptotic signaling pathways
EJCB
824 European Journal of Cell Biology 79, 824 ± 833 (2000, November) ´ Urban & Fischer Verlag ´ Jena http://www.urbanfischer.de/journals/ejcb
Protein kinase C activation modulates pro- and anti-apoptotic signaling pathways Gerold Meinhardt1), Jeannette Roth, Gabriela Totok Department of Hematology/Oncology, Medizinische Klinik Innenstadt of the Ludwig-Maximilians-University, Munich/Germany Received July 29, 1999 Received in revised version April 3, 2000 Accepted June 15, 2000
PKC ± apoptosis ± phorbol ester Activation of protein kinase C (PKC) by TPA in human U937 myeloid leukemia cells is associated with induction of adherence, differentiation, and G0/G1 cell cycle arrest. In this study, we demonstrate that in addition to these differentiating cells about 25% of U937 cells accumulated in the subG1 phase after TPA treatment. This effect proved to be phorbol ester-specific, since other compounds such as retinoic acid or vitamin D3 failed to induce apoptosis in conjunction with differentiation. Only a specific inhibitor of PKC, GF109203X, but not the broad-spectrum kinase inhibitor staurosporine or a tyrosine kinase inhibitor genistein could reverse the induction of apoptosis. Bryostatin-1, another specific PKC activator with distinct biochemical activity failed to induce apoptosis. Moreover, bryostatin-1 completely abolished the induction of apoptosis in U937 cells even if added 8 hours after TPA treatment. Apart from apoptosis induced by various chemotherapeutic drugs, TPA-related cell death is not mediated by an autocrine Fas-FasL loop and could not be prevented by a blocking antibody to the Fas receptor. However, a 75% reduction in the number of apoptotic cells after TPA stimulation was achieved by preincubation with a blocking antibody to the TNFa receptor. Tetrapeptide cleavage assays revealed a four-fold increase in the DEVD-cleavage activity in U937 cells compared to a three-fold increase in TUR cells. Immunoblotting demonstrated that TUR cells did not activate significant levels of caspase-3 or -7, whereas in U937 cells a 20-kDa cleavage product corresponding to activated caspase-3 was detectable after 3 d TPA exposure. Moreover, immunoblots revealed a strongly reduced expression of the adaptor molecule APAF-1, which is required for cytochrome c-dependent activation of caspase-9 and subsequently caspase-3. APAF-1 proved to be inducible after PKC activation with phorbol ester in U937, but not in TUR cells. Thus, APAF-1 expression may, at least in part, be regulated by PKC activity and reduced Dr. Gerold Meinhardt, Medizinische Klinik Innenstadt, Department of Hematology/Oncology, Ludwig-Maximilians-University, Ziemssenstrasse 1, D-80336 Munich/Germany, e-mail: geroldmeinhardt@yahoo. de, Fax: 49 89 5160 4412.
1)
APAF-1 levels are associated with resistance to various inducers of apoptosis. Furthermore, TPA exposure of U937 cells is associated with increased levels of the pro-apoptotic proteins Bak and Bcl-xS, whereas simultaneously a decline in the Bcl-2 expression was noticable.
Abbreviations. PI Propidium iodide. ± PKC Protein kinase C. ± TPA 12-OTetradecanoylphorbol-13-acetate. ± RA Retinoic acid. ± vitD3 1,25Dihydroxyvitamin D3.
Introduction Proliferation, differentiation, retrodifferentiation, and apoptosis represent mutually exclusive cellular responses to distinct stimuli (Hass, 1994). The molecules contributing to the respective signaling pathways include proteins that may participate in diverse signaling cascades. In this context, particularly the PKC family of serine/threonine kinases has been implicated in the regulation of such processes in many cell types. Activation of PKC by the endogenous substrate diacylgylcerol or pharmacological compounds such as phorbol esters that bind to most PKC isoforms leads to induction of transcription factors like AP-1, Ets-1 (Naito et al., 1998), NFkB (Liu et al., 1991) and others, resulting in a cellular response mediated by a variety of other signaling pathways linked to PKC. These PKC-mediated responses have been extensively studied with respect to proliferation and differentiation (Nishizuka, 1988, 1992). Nonetheless, current knowledge about the role of PKC in the regulation of factors crucial to the apoptotic machinery is still sparse. Functional studies have demonstrated that activation of PKC in many cell types relays resistance to or delay of apoptosis in response to many different stimuli (Lotem et al., 1991). This protective effect may result from induction of anti-apoptotic signals or modulation of anti-apoptotic pathways. Apoptosis is a tightly regulated process involving an increasingly complex network of proteins inducing, regulating and executing programmed cell death. (Cryns and Yuan, 1998)
0171-9335/00/79/11-824 $15.00/0
EJCB
Among them, the family of Bcl-2-related proteins with proand anti-apoptotic members forming homo- or heterodimers plays a central role (Yang and Korsmeyer, 1996). Several reports have focussed on the role of the Bcl-x protein, which includes the long (L) splice variant preventing apoptosis and the short (S) splice product promoting apoptosis. Treatment of K562 cells with phorbol ester resulted in differentiation towards the megakaryocytic lineage associated with an increase in the levels of Bcl-x. Moreover, phorbol ester-induced megakaryocytic differentiation was facilitated by the overexpression of Bcl-x in K562 cells, whereas the overexpression of Bcl-xS induced differentiation in the same cell line (Ray et al., 1996; Terui et al., 1998). In the erythropoietin-dependent cell line AS-E2, treatment with Epo or TPA resulted in upregulation of Bcl-xL (Tsushima et al., 1997). Furthermore, in HL-60 and U937 cells 12,13-phorbol dibutyric acid has been postulated to induce up-regulation of Bcl-xL (Chatterjee et al., 1997). During early stages of human hematopoietic progenitor maturation, Bcl-x protein expression was shown to increase with continuance of this high expression level in monocytes/ macrophages, but not in granulocytes (Sanz et al., 1997). Another protein of the Bcl-2 family, Mcl1, can be induced in response to PKC activation or GM-CSF stimulation in leukemic cells (Chao et al., 1998; Townsend et al., 1998, 1999). In contrast, in human leukemic HL-60 cells, activation of PKC by phorbol ester leads to differentiation coupled with down-regulation of the anti-apoptotic Bcl-2 protein (Benito et al., 1995; Sakakura et al., 1996). Previous work has demonstrated that features of apoptosis such as internucleosomal DNA fragmentation occur during the PKC-mediated differentiation and retrodifferentiation process of leukemic cells induced to differentiate with phorbol ester (Gunji et al., 1992). In this context, retrodifferentiation and apoptosis represent alternative means to evade terminal differentiation: Whereas the majority of differentiated cells can enter a cellular program resulting in reversal of the differentiated phenotype, up to 25% of differentiated cells undergo programmed cell death (Gunji et al., 1992). Furthermore, recent data indicate that activation of PKC by phorbol ester can induce differentiation and apoptosis in the same leukemic clone in parallel (manuscript submitted). Accordingly, a certain population of leukemic cells undergoes apoptosis without prior acquisition of a differentiated phenotype. Among the Ca2-independent PKC isoforms, activation of the d form of PKC has been implicated in the induction of differentiation, apoptosis, and cell cycle regulation in leukemic cells (Watanabe et al., 1992; Mischak et al., 1993; Emoto et al., 1995). In particular, phorbol ester-activated PKCd in conjunction with PKCa has been linked to the maturation of murine 32D leukemic precursor cells, whereas caspase-activated PKCd was shown to coincide with and induce apoptosis in U937 myeloid leukemic and other cells. Recently, PKCd has been linked to activation of caspase-3, suggesting a role both downstream and upstream of the caspase cascade (Reyland et al., 1999). Other PKC isoforms, particularly PKCq, have been shown to be involved in later stages of apoptosis as substrates of caspases (Datta et al., 1997; Mizuno et al., 1997). However, little is known on how activation of PKC contributes to the differential regulation of apoptosis and differentiation in leukemic cells. Therefore, we examined whether PKC-mediated induction of differentiation and cell cycle arrest is associated with concomitant induction of anti- or pro-apoptotic pathways.
PKC activation and apoptosis 825
Material and methods Cell lines and reagents
Human U937 myeloid leukemic cells (American Type Culture Collection, Rockville, MD) and the differentiation-resistant subclone TUR were grown in RPMI 1640 medium containing 10% heat-inactivated fetal calf serum (FCS) (BioWhittaker, Belgium) supplemented with 2 mM L-glutamine without antibiotics. The cells were incubated in a humidified atmosphere (37 8C, 5% CO2) and treated with various concentrations of the phorbol ester 12-O-tetradecanoylphorbol-13acetate (TPA) (Sigma, Germany) for the time points indicated. For some experiments, the cells were preincubated with specific inhibitors of caspase-1, YVAD-CHO, and caspase-3/-7, DEVD-CHO (Biomol, Germany), respectively. GF109203X, bryostatin-1, retinoic acid, and vitamin D3 were purchased from Biomol, Germany. The anti-Fas IgG antibody and the anti-TNFa-receptor IgG antibody used to block the receptors were from Upstate Biotechnology, Germany.
Flow cytometry
For cell cycle analysis, untreated and stimulated U937 and TUR cells (106 cells/ml) were fixed with 70% (w/v) ice-cold ethanol overnight. Adherent cells were first scraped from the plastic surface of the culture flask. After two washes with ice-cold PBS, the fixed cells were resuspended in 1 ml of PBS containing 40 mg/ml propidium iodide (Sigma, Germany) and 500 U/ml RNase A (Boehringer Mannheim). Following incubation for 30 min in the dark the cells were analysed on an EPICS XL-MCL flow cytometer (Coulter, Germany) using System II software.
Caspase-3/-7 activity assay
The assay was performed according to the manufacturer s instructions (Clontech, Heidelberg, Germany). Briefly, cells were resuspended in 50 ml lysis buffer and incubated for 10 min on ice. Fifty ml of 2 reaction buffer containing 1,4-dithiothreitol (DTT) and 1 ml of the caspase-3/-7 inhibitor were added to the inhibitor sample and incubated for 30 min at 37 8C, while the other samples were kept on ice. Subsequently, 50 ml of 2 reaction buffer (with DTT) were added to the remaining samples. Five ml of the conjugated substrate were added to each sample and incubated at 37 8C for 1 h. Thereafter, sample volumes were adjusted to 1.0 ml with distilled water and the absorbance at 405 nm was measured with a spectrophotometer.
Immunoblot analysis
Untreated and stimulated U937 and TUR cells were washed three times in ice-cold PBS and lysed in a buffer containing 10 mM Tris-HCl (pH 7.6), 137 mM NaCl, 1 mM Na3VO4, 10 mM NaF, 10 mM EDTA, 1% (v/v) NP-40 with the addition of 10 mg/ml aprotinin, 10 mg/ml leupeptin, and 1 mM phenylmethylsulfonylfluoride (PMSF). Protein concentration was adjusted using a colorimetric assay. Controls for equal loading of protein were performed using an antibody against bactin. Proteins were subjected to SDS-polyacrylamide gel electrophoresis and transferred to a PVDF membrane (Millipore, Germany). The transfer buffer contained 25 mM Tris, 192 mM glycine, 0.037% (w/v) SDS and 20% (v/v) methanol. The membranes were blocked with PBS containing 5% dried milk and 0.05% Tween-20 (PBS/Tween). After washing four times with PBS/Tween, the membranes were incubated with appropriate primary antibodies (all from Santa Cruz Biotechnology, Santa Cruz, CA) and visualized by autoradiography using the ECL detection kit (Amersham, Germany). For preparation of membrane and cytosolic fractions, cells were swollen on ice in a buffer containing 25 mM Tris-HCl, pH 7.6, 10 mM NaF, 10 mM EDTA, 1 mM Na3VO4, with the addition of 10 mg/ml aprotinin, 10 mg/ml leupeptin, 1 mM phenylmethylsulfonylfluoride (PMSF), lysed with 30 strokes of a homogenizer type S, and centrifuged at 15 000g for 15 min at 4 8C. After separation of the supernatant containing cytosolic proteins, the pellet, containing the membrane fraction, was incubated with buffer containing 1% NP-40 for an additional 30 min.
826 G. Meinhardt, J. Roth, G. Totok
EJCB
Results Induction of differentiation-sensitive U937 leukemia cells with TPA in contrast to the differentiation-resistant TUR subclone leads to pronounced apoptosis
Treatment of U937 myeloid leukemia cells with 5 nM TPA was associated with a complete growth arrest after 72 hours. Whereas most cells enter a non-proliferative G0/G1 cell cycle phase coupled with induction of differentiation markers (Hass et al., 1989, 1990; Meinhardt and Hass, 1995), a fraction of the cells undergoes apoptosis with chromatin condensation, membrane blebbing, appearance of apoptotic bodies, annexin V-binding (data not shown) and DNA fragmentation in 8% of the cells after 24 hours and 25% after 72 hours as determined by analysis of propidium iodide stained cells by flow cytometry (Figure 1). In comparison, we tested the induction of apoptosis in the related, U937-derived human leukemic TUR cell line, which is resistant to phorbol ester-induced differentiation and cell cycle arrest by a mechanism involving a reduced expression of distinct PKC isoforms (Hass et al., 1993, 1994). Incubation of TUR cells with 5 nM TPA for 24 and 72 hours resulted in no increase in the number of apoptotic cells (Figure 1). Preincubation of U937 cells with 1 mM of the specific PKC inhibitor GF109203X completely abolished apoptosis induced by treatment of the cells with 5 nM TPA for 48 hours (data not shown).
Retinoic acid or 1,25-dihydroxyvitamin D3 (vitD3), inducers of leukemia cell differentiation, cause cell cycle arrest, but not induction of apoptosis
Fig. 1. Induction of differentiation-sensitive U937 leukemia cells with TPA, but not the differentiation-resistant TUR subclone, leads to pronounced apoptosis. The cells were treated with 5 nM TPA for the indicated time points. The cells were fixed in ethanol and subjected to flow cytometry following digestion of RNA and staining of DNA with propidium iodide. The values represent the number of cells displaying hypodiploid DNA content (subG1-DNA) and are the mean of at least three independent experiments SD.
The ultimate fate of differentiated cells is either terminal commitment and subsequent apoptotic cell death or activation of a cellular program, termed retrodifferentiation, resulting in a reversal of the differentiated phenotype and reentry into the cell cycle. To investigate whether the apoptotic population of cells after TPA treatment is related to activation of signaling pathways leading to differentiation in general, or specifically arises from TPA-mediated PKC-activation, U937 cells were
treated with two distinct inducers of myeloid cell maturation, namely, retinoic acid (RA) and 1,25-dihydroxyvitamin D3 (vitD3). Accordingly, 1 mM RA for 4 days caused a G0/G1arrest of 81% of the population, whereas 500 nM vitD3 induced a cell cycle arrest in 66% of the cells, compared to 53% of proliferating U937 cells, demonstrating the efficacy of the concentrations applied (Figure 2a). Both agents failed to
Fig. 2. Retinoic acid or 1,25-dihydroxyvitamin D3 (vitD3), inducers of leukemia cell differentiation, cause cell cycle arrest, but not induction of apoptosis. U937 cells were treated with 5 nM TPA, 1 mM retinoic acid (RA), or 500 nM 1,25-dihydroxyvitamin D3 (vitD3) for the indicated time points. For the co-incubations, TPA was added without prior removal of RA or vitD3. The number of cells in subG1
(apoptotic) or G1 cell cycle phase was determined by flow cytometry following digestion of RNA and staining of DNA with propidium iodide. The values represent the percentage of cells in a given cell cycle phase and are the mean of at least three independent experiments SD. (a) Percentage of cells in G1 cell cycle phase. (b) Percentage of apoptotic cells.
EJCB
PKC activation and apoptosis 827
induce apoptosis above a background level of 3%, which is in the range of untreated U937 cells. Preincubation with both RA and vitD3, however, decreased the amount of apoptotic cells after TPA-mediated PKC activation by 20% and 50%, respectively, indicating that the activation of pathways mediating RA- and vitD3-related differentiation is distinct from and interferes with PKC-induced cell death (Figure 2b).
PKC activation by bryostatin-1 does not induce apoptosis
Bryostatin-1, a macrocyclic lactone, is another specific activator of PKC, which has recently entered clinical trials for the treatment of various cancer types. Bryostatin-1 is known to induce isozyme-specific down-regulation of PKCs, thus partially antagonizing the differentiating effects of phorbol ester. However, little is known about the effects of this compound on the regulation of apoptosis. 5 to 100 nM of bryostatin-1 for 72 hours induced no significant accumulation of subG1 cells. Moreover, pretreatment of cells with bryostatin-1 for 1 hour followed by 5 nM TPA for 72 hours largely abolished the apoptosis-inducing effects of the phorbol ester. This inhibitory effect of bryostatin-1 was also observed when bryostatin-1 was added up to eight hours after incubation of U937 cells with TPA. Bryostatin-1 added after 24 hours, however, could no longer reverse the activation of the pro-apoptotic PKCmediated pathway, indicating that signaling intermediates downstream of PKC become activated after 8 to 24 hours of incubation with phorbol ester (Figure 3a).
Differential regulation of PKC isoforms after treatment with TPA or bryostatin-1 in U937 and TUR cells
Activation of PKC leads to translocation of isoforms to different cellular compartments and subsequent down-regulation of the kinases. We investigated this differential regulation after stimulation with TPA or bryostatin-1 for 72 hours by Western blotting. Whereas little if any change in the expression and translocation of PKCq could be observed, incubation of U937 cells with TPA and bryostatin-1 induced membrane translocation of the PKC isoforms b and d, respectively (Figure 3b). In contrast to TPA, however, bryostatin-1 led to a down-regulation of PKC b and d after 72 hours. Interestingly, a different pattern of expression and translocation could be detected in TUR cells. Most notably, bryostatin-1 led to downregulation of PKCd, but failed to induce membrane translocation of this particular isoform (Figure 3B). These findings suggested, that bryostatin-1 in comparison to TPA is capable to translocate and down-regulate PKC with distinct isoform specificity.
PKC inhibition and activation during early retrodifferentiation
PKC-activated, differentiated cells down-regulate PKC. Restauration of PKC content, in addition to other, only partially identified factors, leads to resumed proliferation and retrodifferentiation. A significant number of cells, however, fail to undergo retrodifferentiation and thus become committed to apoptosis. To address the question of PKC involvement in the process of apoptosis following differentiation, cells were treated with 5 nM TPA for 3 days. The adherent, differentiated population was then thoroughly washed and challenged with either the specific PKC inhibitor GF109203X, the phorbol ester-unrelated PKC activator bryostatin-1 or another course
Fig. 3. a PKC activation by bryostatin-1 does not induce apoptosis. U937 cells were treated with the indicated concentrations of TPA or bryostatin-1 (BRYO) alone or as co-incubations. The analysis of apoptotic cells was performed as described in Figure 1. b. Translocation of PKC isoforms after treatment with TPA or bryostatin-1. U937 and TUR cells were treated with 5 nM TPA or 100 nM bryostatin-1 for 72 hours. Cytosolic and membrane fractions were subjected to Western blotting with the indicated antibodies using enhanced chemiluminescence for detection of proteins. Actin controls indicated equal loading of cytosolic and membrane proteins, respectively.
of TPA, each for additional 3 days. Whereas approximately 25% of cells treated with TPA for 72 h underwent apoptosis, treatment of the differentiated, vital cells with TPA for another 3 days resulted in an additional loss of 23% of cells due to apoptosis. This compares to 17% of cells cultured only in the presence of medium and 11% or 12% of cells, respectively, further challenged with bryostatin or GF109203X (Figure 4). This is consistent with a model of restauration of PKC activity in retrodifferentiating cells with PKC function being part of the signals leading to apoptosis following differentiation. Taken together, the observed cellular responses indicate an
828 G. Meinhardt, J. Roth, G. Totok
EJCB
apoptotic response specifically related to phorbol estermediated PKC activation rather than to a common differentiation program.
therapy-induced apoptosis. Preincubation of U937 cells with the caspase-3/-7-related inhibitor DEVD-CHO for 1 hour significantly impaired the TPA-mediated pro-apoptotic signal generated by PKC activation, as demonstrated by a moderate to strong reduction of subG1 cells (manuscript submitted). Further analysis revealed, that 500 mM DEVD added 12 hours after TPA treatment still caused a 60% reduction in the number of apoptotic cells by 60 hours. YVAD-CHO, a specific inhibitor of caspase-1, however, was unable to reduce the generation of subG1 cells if added 12 h after TPA (Figure 5a). Since both caspase inhibitors penetrate membranes rather poorly (manufacturer s information), we used a concentration that proved sufficient for DEVD to reduce the amount of apoptosis in U937 cells induced by VP-16 (etoposide), a potent cytotoxic agent, by 75% after a 4-h incubation period (data not shown). Thus, in the time-frame of TPA-induced differentiation with adherence of the majority of cells occurring 12 h after stimulation (data not shown), caspase-3-mediated cleavage of vital cellular substrates does not seem to occur within the first 12 h of PKC activation. Direct measurement of the activity of caspase-3/-7 in U937 and resistant TUR cells was performed using a colorimetric tetrapeptide substrate. 5 nM TPA for 72 hours caused a four-fold increase in the cleavage activity of U937 cells compared to a three-fold increase in TUR cells, indicating that a certain threshold of caspase activation seems necessary to proceed with the execution phase of the programmed cell death (Figure 5b).
Caspase-3 activation during TPA-mediated PKC activation
Evidence for TNF/TNF-receptor, but not for Fas/ FasL interaction during TPA-induced cell death
Fig. 5. a Caspase-3 activation during TPA-mediated PKC activation. U937 cells were treated with 5 nM TPA for 12 hours. Subsequently, 500 mM of the caspase-1 inhibitor YVAD or the caspase-3/-7 inhibitor DEVD was added and the cells were harvested at the indicated time points. The analysis of apoptotic cells was performed as described in Figure 1. b. Measurement of the activity of caspase-3/-7 in U937 and
resistant TUR cells using a colorimetric tetrapeptide substrate assay. U937 and TUR cells were treated with 5 nM TPA for 3 days. Cytosolic extracts were prepared and the activity of DEVD-caspases was measured using a colorimetric tetrapeptide substrate assay. The cleavage activity is depicted in arbitrary units. The values represent the mean of three independent experiments SD.
Fig. 4. PKC inhibition and activation during early retrodifferentiation. U937 cells were treated with 5 nM TPA for 3 days. Subsequently, the differentiated adherent cells were washed several times to remove the TPA and then either treated with the indicated concentrations of bryostatin-1, the PKC inhibitor GF109203X, TPA, or left untreated for an additional 3 days. The analysis of apoptotic cells was performed as described in Figure 1.
The caspases, a family of related proteases, have been implicated in virtually all aspects of apoptotic cell death. Among them, caspase-1 (ICE) as well as caspase-3 and -7 play important roles in inflammatory, death receptor- and chemo-
Recent evidence suggests that chemotherapy-induced apoptosis may at least in part be mediated by an autocrine mechanism involving interaction of induced Fas ligand with its receptor (Friesen et al., 1996). Consequently, we examined the
EJCB
possible involvement of a FasL-Fas autocrine loop in U937 cells which constitutively express Fas (data not shown). Blockage of Fas in untreated U937 control cells with an excess of a non-activating antibody (Fas IgG) (McGahon et al., 1998) caused no apparent cell death. Moreover, preincubation of cells with this Fas IgG for 15 min did not reduce the number of subG1 cells after TPA stimulation, excluding the possibility of an autocrine loop. In contrast, an increase in the number of apoptotic cells was observed after 72 hours of TPA following Fas IgG incubation (Figure 6a). Accordingly, a possible autocrine TNF/TNF-receptor loop was investigated. Whereas an excess of the non-activating TNFa-receptor antibody did not result in enhanced cell death, pre-incubation of this antibody 15 min prior to TPA stimulation significantly reduced apoptosis by more than 75% (Figure 6b). Thus, TPA-induced apoptosis is at least in part mediated by activation of the TNF/TNF-receptor pathway, but not the Fas/FasL pathway. Preincubation with up to 150 mM of the tetrapeptide inhibitor Z-IETD-fmk, however, did not reverse the proapoptotic effect of TPA (data not shown).
PKC activation and apoptosis 829
differentiation, an early increase in expression of Mcl1 could be detected as early as 3 h following treatment with TPA. The levels of Mcl1 remained elevated throughout the observed 72h period. A similar increase in Mcl1 expression was observed in the resistant TUR cells with a time-course paralleling that of U937 cells, indicating that TUR cells retain at least some suceptibility to TPA stimulation despite their failure to undergo phorbol ester-induced differentiation or apoptosis. Whereas in TUR cells a slight increase in the levels of Bcl-2 protein could be detected, U937 cells down-regulated Bcl-2 after 72 h independent of phorbol ester concentration (Figure 7a).
Expression of APAF-1 and cleavage of caspases during TPA treatment
Bcl-2 proteins govern the initiation or execution of apoptotic signaling cascades. Therefore, we determined the expression of relevant members of this protein family at various time points after incubation with 5 ± 500 nM TPA. No relevant change in the expression levels of the pro-apoptotic proteins Bax and Bad was observed at all time points and TPA concentrations in U937 cells. There was a significant increase, however, in the levels of Bak and the short form of the Bcl-x protein after 72 h TPA. Interestingly, the enhanced expression of both proapoptotic proteins was observed at a time, where the cells were already in the late execution phase of either differentiation or apoptosis. Another pro-apoptotic protein, Bag, showed little if any variation in expression levels. Concomitant with induced
The APAF-1 protein is a central component of the apoptosome complex, being comprised of cytochrome c, released from its mitochondrial origin, pro-caspase-9, and ATP. Formation of this complex leads to activation of caspase-9 and subsequently downstream caspases, such as caspase-3 or -7. There is very little knowledge about the regulation of APAF-1 expression in stimulated and apoptosis-resistant cells. Consequently, APAF-1 expression was analysed in response to PKC activation in U937 and TUR cells. Treatment with TPA for up to 72 h resulted in enhanced expression of APAF-1 in U937 cells (Figure 7b). The resistant TUR cells demonstrated a highly reduced level of APAF-1 expression as determined by Western blotting of membranes overexposed after the chemiluminescence reaction (data not shown) and no apparent APAF induction after TPA stimulation (Figure 7b). In addition, the activation of caspase-3 was observed only in the TPAstimulated U937 cells by detecting an approximately 20-kDa protein corresponding to a cleaved fragment of caspase-3. In this context, no activation of caspase-7 could be observed in either cell line after TPA treatment (Figure 7b).
Fig. 6. Evidence for TNF/TNF-receptor, but not for Fas/FasL interaction during TPA-induced cell death. (a) U937 cells were treated with a blocking antibody against Fas (Fas-IgG) or 5 nM TPA for 1 or 3 days. For the co-incubations, the cells were stimulated with TPA 1 hour after the addition of Fas-IgG. (b) U937 cells were treated with a blocking
antibody against the TNFa receptor (TNF-IgG) or 5 nM TPA for 1 or 3 days. For the co-incubations, the cells were stimulated with TPA 1 hour after the addition of TNF-IgG. The values represent the mean of three independent experiments SD. The analysis of apoptotic cells was performed as described in Figure 1.
Expression of Bcl-2 familiy proteins
830 G. Meinhardt, J. Roth, G. Totok
EJCB
Fig. 7. a Expression of Bcl-2 familiy proteins. U937 and TUR cells were treated with TPA for the indicated time points and concentrations. Lysates were subjected to Western blotting with the indicated antibodies using enhanced chemiluminescence for detection of proteins. The actin probe served as control to confirm equal loading of
proteins. The results shown are representative of several independent experiments. b. Expression of APAF-1 and cleavage of caspases during TPA treatment. U937 and TUR cells were treated with 5 nM TPA for 3 days and subjected to Western blotting as described in (a).
Discussion
(data not shown). Moreover, the acquisition of a differentiated phenotype requires approximately 2 ± 3 days in U937 (Hass et al., 1992), whereas inhibition of caspases more than 24 h following TPA treatment is no longer effective in preventing apoptosis (unpublished data). Previous work has confirmed, that stimulation of U937 and related leukemic cell lines with compounds such as retinoic acid or vitD3 results in a G0/G1 cell cycle arrest with concomitant induction of differentiation, which is independent of direct PKC activation (Hass et al., 1993). We have demonstrated, that both retinoic acid and vitD3 do not lead to a simultaneous appearance of apoptotic cells beyond the level of untreated controls, thus suggesting that the observed cell death following TPA treatment is intrinsically linked to activation of PKC rather than induction of a general cellular differentiation program. This is underscored by the ability of the specific PKC inhibitor GF109203X to fully prevent TPA-mediated apoptosis.
Programmed cell death is a multi-step process involving an increasingly complex network of proteins contributing to the induction, regulation and execution of apoptosis. We attempted to examine the contribution of PKC activation upon the regulation of certain aspects of the apoptotic process. Activation of PKC in myeloid leukemic cells by phorbol ester treatment is commonly associated with differentiation along the monocytic lineage (Hass et al., 1989). We have demonstrated in this paper, that approximately 25% of a monoclonal U937 population derived from single cell suspensions fails to undergo G0/G1 cell cycle arrest and exhibits features of apoptotic cells. This induction of apoptosis is apparently not a consequence of prior differentiation with subsequent cell death, since the fraction of apoptotic cells does not exhibit markers commonly associated with monocytic cell maturation
EJCB
Some insight into the mechanisms of PKC activation and function stems from the observation that bryostatin-1 in contrast to phorbol ester fails to induce differentiation or proliferation in distinct cell lines, at least in part by differential down-regulation of PKC isoforms such as a, d and e (Szallasi et al., 1994). Our data support the hypothesis of differential PKC activation by demonstrating that byrostatin-1 significantly blocks TPA-mediated apoptosis in U937 cells, while by itself inducing neither differentiation nor apoptosis. The observed differences in the expression levels and translocation of the PKC isoforms suggest involvement in the apoptotic response. Most recently, PKCd has been shown to translocate to the mitochondrial membrane and has been implicated in the regulation of caspase-3 activation (Li et al., 1999; Reyland et al., 1999). Our data, demonstrating down-regulated PKCd especially in the membrane fraction in bryostatin-1-treated cells, is consistent with this model. Thus, the antagonistic effect of bryostatin-1 upon the TPA-mediated induction of apoptosis may be caused by a reduced amount of membrane-associated PKCd being required for caspase activation. It is unknown whether phorbol ester-related differentiation and apoptosis may require target proteins other than PKC in myeloid cells. However, this possibility has not been formally ruled out and additional phorbol ester acceptors have been identified (Areces et al., 1994). In view of recent results establishing the regulation of caspase-9 and Bcl-2 proteins such as Bad and Bcl-2 by phosphorylation (Haldar et al., 1995; Zha et al., 1996; Cardone et al., 1998), it is conceivable that PKC isoforms may have a role upstream of caspase activation. In this context, both Bad and Bcl-2 are phosphorylated on serine residues, possibly by a mechanism involving the Raf-1 serine/threonine-kinase. (Wang et al., 1994, 1996) Bcl-2 can target Raf-1 kinase to mitochondria, resulting in phosphorylation of Bad (Wang et al., 1996). Phosphorylated Bad is unable to bind to and inactivate Bcl-2 or Bcl-xL, thus allowing a proapoptotic signal to proceed possibly towards caspase activation (Zha et al., 1996). Raf-1 kinase was shown to be activated following disruption of microtubular architecture and can bind via its catalytic domain to Bcl-2 in a BH4 domain-dependent manner. While Bcl-2 is not phosphorylated by Raf-1 directly, this effect requires the activity of another serine-/threoninekinase (Wang et al., 1994). In this context, PKCa has been shown to co-localize with Bcl-2 in mitochondrial membrane preparations of leukemic HL-60 cells as well as to phosphorylate Bcl-2 in vitro at serine 70. Moreover, overexpression of PKCa induces enhanced Bcl-2 phosphorylation and a substantial increase in resistance to drug-induced apoptosis (Ruvolo et al., 1998). Previous as well as unpublished results show that TUR cells contain reduced levels of several PKC isoforms relative to U937 cells, among them bII and d (manuscript submitted) (Hass et al., 1993). Therefore, enhanced Bcl-2 phosphorylation is less likely to be a mechanism of apoptosis resistance in TUR cells compared to differences in the levels of protein expression. Differences in the expression of certain Bcl-2-like proteins have been linked to an increase in the suceptibility to cytotoxic agents. Increased expression of Bax has been shown to be sufficient to induce the release of mitochondrial cytochrome c and activation of pro-caspase-9 without prior opening of the mitochondrial permeability transition pore (Eskes et al., 1998). TPA treatment of U937 cells, however, is not associated with increased levels of Bax, but induces other pro-apoptotic factors, namely Bak and Bcl-xS. Conversely, the anti-apoptotic
PKC activation and apoptosis 831
Bcl-2 is down-regulated in U937 cells. The data presented here show an increase in Bak and Bcl-xS expression coupled with reduced Bcl-2 only after more than 24 h, thus making it less likely, that the induction of these factors alone leads to caspase activation, which is observed after 12 h. Enhanced expression of Bak and Bcl-xS may be a necessary step in an amplification process leading to an enhancement of the pro-apoptotic signals (Thompson, 1998). This might be in concert with observations (including ours) that activation of caspases and appearance of an apoptotic phenotype can be separated by a prolonged lag period of unknown function. The induction of Bcl-xS expression in U937 cells with a constant level of Bcl-xL after 72 h independent of TPA concentration as consistently observed by us is in contrast to previously published results showing enhanced expression of Bcl-xL after treatment with the related phorbol ester 12,13-phorbol dibutyric acid (Chatterjee et al., 1997). Recent results have established that non-receptor-mediated apoptosis involves release of cytochrome c from mitochondria. Cytosolic cytochrome c then associates with APAF-1 and caspase-9 in the presence of ATP, allowing for the autoprocessing of the caspase-9 zymogen form to the active dimeric protease and subsequent activation of downstream effectors (Li et al., 1997). In addition, APAF-1 is able to associate with other caspases like -4 und -8. Binding of APAF-1 to the antiapoptotic protein Bcl-xL prevents caspase-9 activation, suggesting that APAF-1 serves as adaptor protein linking the caspase executioners to upstream regulators (Hu et al., 1998; Pan et al., 1998). Homozygous mutants of murine APAF-1 die at embryonic day 16 and are unable to activate caspase-3 (Cecconi et al., 1998). Thus, it is conceivable, that the severely reduced expression of APAF-1 in TUR cells might play a central role in the observed resistance to a variety of apoptotic inducers. There is little information about the regulation of APAF-1 expression as well as its role in the resistance to apoptosis. We provide evidence that APAF-1 protein levels are upregulated during induced differentiation of U937 cells. Interestingly, TPA-treated TUR cells show an increase in DEVD-tetrapeptide cleavage activity, which extends to 75% of the respective activity of TPA-stimulated U937 cells. DEVD is a consensus cleavage site for both caspase-3 and -7 as well as a potential cleavage site for other caspases (Kidd, 1998). We could not detect processing of caspase-3 and -7 in TPA-treated TUR cells by immunoblotting, whereas the typical 20-kDa protein band of cleaved caspase-3 could be readily detected in U937 cells. Thus, TUR cells may activate other DEVDsensitive caspases, but apparently below a threshold that is required for proceeding with the execution phase of apoptosis. Activation of the downstream caspases can also be achieved by stimulation of members of the TNF receptor superfamily. Recently, activation of the Fas receptor by an autocrine FasL/ Fas loop has been postulated as a possible mechanism of chemotherapy-induced apoptosis (Friesen et al., 1996). Our results rule out an involvement of Fas in TPA-mediated apoptosis. In contrast, we demonstrate employment of the TNF/TNF-receptor pathway in the process of PKC-activated cell death. This is in concert with a recent study linking phorbol ester-induced TNFa production and apoptosis in the same cell line (Takada et al., 1999). The mechanisms of TNF-induced cell death in U937 cells needs to be elucidated, since our data could not prove involvement of caspase-8. Taken together, our data demonstrate that activation of PKC in U937 leukemic cells is accompanied by marked
832 G. Meinhardt, J. Roth, G. Totok
alterations in the expression of proteins governing apoptosis. These alterations are largely absent in the PKC-deficient U937 subclone TUR. It will be necessary to determine the exact contribution of individual PKC isoforms to the induction of apoptosis as well as to regulation of the expression of certain apoptosis proteins. Acknowledgement. This work was supported by a grant from the Deutsche Forschungsgemeinschaft (Me 1189/2) to G. Meinhardt. Some experiments are part of the thesis work of G. Totok.
References Areces, L. B., Kazanietz, M. G., Blumberg, P. M. (1994): Close similarity of baculovirus-expressed n-chimaerin and protein kinase C alpha as phorbol ester receptors. J. Biol. Chem. 269, 19553 ± 19558. Benito, A., Grillot, D., Nunez, G., Fernandez-Luna, J. L. (1995): Regulation and function of Bcl-2 during differentiation-induced cell death in HL-60 promyelocytic cells. Am. J. Pathol. 146, 481 ± 490. Cardone, M. H., Roy, N., Stennicke, H. R., Salvesen, G. S., Franke, T. F., Stanbridge, E., Frisch, S., Reed, J. C. (1998): Regulation of cell death protease caspase-9 by phosphorylation. Science 282, 1318 ± 1321. Cecconi, F., Alvarez-Bolado, G., Meyer, B. I., Roth, K. A., Gruss, P. (1998): Apaf1 (CED-4 homolog) regulates programmed cell death in mammalian development. Cell 94, 727 ± 737. Chao, J. R., Wang, J. M., Lee, S. F., Peng, H. W., Lin, Y. H., Chou, C. H., Li, J. C., Huang, H. M., Chou, C. K., Kuo, M. L., et al. (1998): mcl-1 is an immediate-early gene activated by the granulocyte-macrophage colony-stimulating factor (GM-CSF) signaling pathway and is one component of the GM-CSF viability response. Mol. Cell. Biol. 18, 4883 ± 4898. Chatterjee, D., Han, Z., Mendoza, J., Goodglick, L., Hendrickson, E. A., Pantazis, P., Wyche, J. H. (1997): Monocytic differentiation of HL-60 promyelocytic leukemia cells correlates with the induction of Bcl-xL. Cell Growth Differ. 8, 1083 ± 1089. Cryns, V., Yuan, J. (1998): Proteases to die for. Genes Dev. 12, 1551 ± 1570. Datta, R., Kojima, H., Yoshida, K., Kufe, D. (1997): Caspase-3mediated cleavage of protein kinase C theta in induction of apoptosis. J. Biol. Chem. 272, 20317 ± 20320. Emoto, Y., Manome, Y., Meinhardt, G., Kisaki, H., Kharbanda, S., Robertson, M., Ghayur, T., Wong, W. W., Kamen, R., Weichselbaum, R., et al. (1995): Proteolytic activation of protein kinase C delta by an ICE-like protease in apoptotic cells. EMBO J. 14, 6148 ± 6156. Eskes, R., Antonsson, B., Osen-Sand, A., Montessuit, S., Richter, C., Sadoul, R., Mazzei, G., Nichols, A., Martinou, J. C. (1998): Baxinduced cytochrome c release from mitochondria is independent of the permeability transition pore but highly dependent on Mg2 ions. J. Cell Biol. 143, 217 ± 224. Friesen, C., Herr, I., Krammer, P. H., Debatin, K. M. (1996): Involvement of the CD95 (APO-1/FAS) receptor/ligand system in druginduced apoptosis in leukemia cells. Nature Med. 2, 574 ± 577. Gunji, H., Hass, R., Kufe, D. (1992): Internucleosomal DNA fragmentation during phorbol ester-induced monocytic differentiation and G0/G1 arrest. J. Clin. Invest. 89, 954 ± 960. Haldar, S., Jena, N., Croce, C. M. (1995): Inactivation of Bcl-2 by phosphorylation. Proc. Natl. Acad. Sci. USA 92, 4507 ± 4511. Hass, R. (1994): Retrodifferentiation and cell death. Crit. Rev. Oncog. 5, 359 ± 371. Hass, R., Bartels, H., Topley, N., Hadam, M., Kohler, L., GoppeltStrube, M., Resch, K. (1989): TPA-induced differentiation and adhesion of U937 cells: changes in ultrastructure, cytoskeletal organization and expression of cell surface antigens. Eur. J. Cell Biol. 48, 282 ± 293. Hass, R., Giese, G., Meyer, G., Hartmann, A., Dork, T., Kohler, L., Resch, K., Traub, P., Goppelt-Strube, M. (1990): Differentiation and retrodifferentiation of U937 cells: reversible induction and suppression of intermediate filament protein synthesis. Eur. J. Cell Biol. 51, 265 ± 271.
EJCB
Hass, R., Gunji, H., Datta, R., Kharbanda, S., Hartmann, A., Weichselbaum, R., Kufe, D. (1992): Differentiation and retrodifferentiation of human myeloid leukemia cells is associated with reversible induction of cell cycle-regulatory genes. Cancer Res. 52, 1445 ± 1450. Hass, R., Hirano, M., Kharbanda, S., Rubin, E., Meinhardt, G., Kufe, D. (1993): Resistance to phorbol ester-induced differentiation of a U937 myeloid leukemia cell variant with a signaling defect upstream to Raf-1 kinase. Cell Growth Differ. 4, 657 ± 663. Hass, R., Meinhardt, G., Hadam, M., Bartels, H. (1994): Characterization of human TUR leukemia cells: continued cell cycle progression in the presence of phorbol ester is associated with resistance to apoptosis. Eur. J. Cell Biol. 65, 408 ± 416. Hu, Y., Benedict, M. A., Wu, D., Inohara, N., Nunez, G. (1998): Bcl-XL interacts with Apaf-1 and inhibits Apaf-1-dependent caspase-9 activation. Proc. Natl. Acad. Sci. USA 95, 4386 ± 4391. Kidd, V. J. (1998): Proteolytic activities that mediate apoptosis. Annu. Rev. Physiol. 60, 533 ± 573. Li, L., Lorenzo, P. S., Bogi, K., Blumberg, P. M., Yuspa, S. H. (1999): Protein kinase C delta targets mitochondria, alters mitochondrial membrane potential, and induces apoptosis in normal and neoplastic keratinocytes when overexpressed by an adenoviral vector. Mol. Cell. Biol. 19, 8547 ± 8558. Li, P., Nijhawan, D., Budihardjo, I., Srinivasula, S. M., Ahmad, M., Alnemri, E. S., Wang, X. (1997): Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 91, 479 ± 489. Liu, J. L., Chiles, T. C., Sen, R. J., Rothstein, T. L. (1991): Inducible nuclear expression of NF-kappa B in primary B cells stimulated through the surface Ig receptor. J. Immunol. 146, 1685 ± 1691. Lotem, J., Cragoe, E. J., Jr., Sachs, L. (1991): Rescue from programmed cell death in leukemic and normal myeloid cells. Blood 78, 953 ± 960. McGahon, A. J., Costa Pereira, A. P., Daly, L., Cotter, T. G. (1998): Chemotherapeutic drug-induced apoptosis in human leukaemic cells is independent of the Fas (APO-1/CD95) receptor/ligand system. Br. J. Haematol. 101, 539 ± 547. Meinhardt, G., Hass, R. (1995): Differential expression of c-myc, max and mxi1 in human myeloid leukemia cells during retrodifferentiation and cell death. Leukemia Research 19, 699 ± 705. Mischak, H., Pierce, J. H., Goodnight, J., Kazanietz, M. G., Blumberg, P. M., Mushinski, F. J. (1993): Phorbol ester induced myeloid differentiation is mediated by protein kinase C-alpha and -delta and not by protein kinase c-betaII, -epsilon, -zeta and -eta. J. Biol. Chem. 268, 20110 ± 20115. Mizuno, K., Noda, K., Araki, T., Imaoka, T., Kobayashi, Y., Akita, Y., Shimonaka, M., Kishi, S., Ohno, S. (1997): The proteolytic cleavage of protein kinase C isotypes, which generates kinase and regulatory fragments, correlates with Fas-mediated and 12-O-tetradecanoylphorbol-13-acetate-induced apoptosis. Eur. J. Biochem. 250, 7 ± 18. Naito, S., Shimizu, S., Maeda, S., Wang, J., Paul, R., Fagin, J. A. (1998): Ets-1 is an early response gene activated by ET-1 and PDGF-BB in vascular smooth muscle cells. Am. J. Physiol. 274, C472 ± C480. Nishizuka, Y. (1988): The molecular heterogeneity of protein kinase C and its implications for cellular regulation. Nature 334, 661 ± 665. Nishizuka, Y. (1992): Intracellular signaling by hydrolysis of phospholipids and activation of protein kinase C. Science 258, 607 ± 614. Pan, G., O Rourke, K., Dixit, V. M. (1998): Caspase-9, Bcl-XL, and Apaf-1 form a ternary complex. J. Biol. Chem. 273, 5841 ± 5845. Ray, S., Bullock, G., Nunez, G., Tang, C., Ibrado, A. M., Huang, Y., Bhalla, K. (1996): Enforced expression of Bcl-XS induces differentiation and sensitizes chronic myelogenous leukemia-blast crisis K562 cells to 1-beta-D-arabinofuranosylcytosine-mediated differentiation and apoptosis. Cell Growth Differ. 7, 1617 ± 1623. Reyland, M. E., Anderson, S. M., Matassa, A. A., Barzen, K. A., Quissell, D. O. (1999): Protein kinase C delta is essential for etoposide-induced apoptosis in salivary gland acinar cells. J. Biol. Chem. 274, 19115 ± 19123. Ruvolo, P. P., Deng, X., Carr, B. K., May, W. S. (1998): A functional role for mitochondrial protein kinase C alpha in Bcl2 phosphorylation and suppression of apoptosis. J. Biol. Chem. 273, 25436 ± 25442.
EJCB
Sakakura, C., Sweeney, E. A., Shirahama, T., Hakomori, S., Igarashi, Y. (1996): Suppression of bcl-2 gene expression by sphingosine in the apoptosis of human leukemic HL-60 cells during phorbol esterinduced terminal differentiation. FEBS Lett. 379, 177 ± 180. Sanz, C., Benito, A., Silva, M., Albella, B., Richard, C., Segovia, J. C., Insunza, A., Bueren, J. A., Fernandez-Luna, J. L. (1997): The expression of Bcl-x is downregulated during differentiation of human hematopoietic progenitor cells along the granulocyte but not the monocyte/macrophage lineage. Blood 89, 3199 ± 3204. Szallasi, Z., Smith, C. B., Pettit, G. R., Blumberg, P. M. (1994): Differential regulation of protein kinase C isozymes by bryostatin 1 and phorbol 12-myristate 13-acetate in NIH 3T3 fibroblasts. J. Biol. Chem. 269, 2118 ± 2124. Takada, Y., Hachiya, M., Osawa, Y., Hasegawa, Y., Ando, K., Kobayashi, Y., Akashi, M. (1999): 12-O-tetradecanoylphorbol-13acetate-induced apoptosis is mediated by tumor necrosis factor alpha in human monocytic U937 cells. J. Biol. Chem. 274, 28286 ± 28292. Terui, Y., Furukawa, Y., Kikuchi, J., Iwase, S., Hatake, K., Miura, Y. (1998): Bcl-x is a regulatory factor of apoptosis and differentiation in megakaryocytic lineage cells. Exp. Hematol. 26, 236 ± 244. Thompson, E. B. (1998): Special topic: apoptosis. Annu. Rev. Physiol. 60, 525 ± 532. Townsend, K. J., Trusty, J. L., Traupman, M. A., Eastman, A., Craig, R. W. (1998): Expression of the antiapoptotic MCL1 gene product is regulated by a mitogen activated protein kinase-mediated pathway triggered through microtubule disruption and protein kinase C. Oncogene 17, 1223 ± 1234.
PKC activation and apoptosis 833
Townsend, K. J., Zhou, P., Qian, L., Bieszczad, C. K., Lowrey, C. H., Yen, A., Craig, R. W. (1999): Regulation of MCL1 through a serum response factor/Elk-1-mediated mechanism links expression of a viability-promoting member of the BCL2 family to the induction of hematopoietic cell differentiation. J. Biol. Chem. 274, 1801 ± 1813. Tsushima, H., Urata, Y., Miyazaki, Y., Fuchigami, K., Kuriyama, K., Kondo, T., Tomonaga, M. (1997): Human erythropoietin receptor increases GATA-2 and Bcl-xL by a protein kinase C-dependent pathway in human erythropoietin-dependent cell line AS-E2. Cell Growth Differ. 8, 1317 ± 1328. Wang, H. G., Miyashita, T., Takayama, S., Sato, T., Torigoe, T., Krajewski, S., Tanaka, S., Hovey, L. R., Troppmair, J., Rapp, U. R., et al. (1994): Apoptosis regulation by interaction of Bcl-2 protein and Raf-1 kinase. Oncogene 9, 2751 ± 2756. Wang, H. G., Rapp, U. R., Reed, J. C. (1996): Bcl-2 targets the protein kinase Raf-1 to mitochondria. Cell 87, 629 ± 638. Watanabe, T., Ono, Y., Taniyama, Y., Hazama, K., Igarashi, K., Ogita, K., Kikkawa, U., Nishizuka, Y. (1992): Cell division arrest induced by phorbol ester in CHO cells overexpressing protein kinase C-delta subspecies. Proc. Natl. Acad. Sci. USA 89, 10159 ± 10163. Yang, E., Korsmeyer, S. J. (1996): Molecular thanatopsis: A discourse on the Bcl2 family and cell death. Blood 88, 386 ± 401. Zha, J., Harada, H., Yang, E., Jockel, J., Korsmeyer, S. J. (1996): Serine phosphorylation of death agonist BAD in response to survival factor results in binding to 14 ± 3 ± 3 not BCL-X(L). Cell 87, 619 ± 628.
824 European Journal of Cell Biology 79, 824 ± 833 (2000, November) ´ Urban & Fischer Verlag ´ Jena http://www.urbanfischer.de/journals/ejcb
Protein kinase C activation modulates pro- and anti-apoptotic signaling pathways Gerold Meinhardt1), Jeannette Roth, Gabriela Totok Department of Hematology/Oncology, Medizinische Klinik Innenstadt of the Ludwig-Maximilians-University, Munich/Germany Received July 29, 1999 Received in revised version April 3, 2000 Accepted June 15, 2000
PKC ± apoptosis ± phorbol ester Activation of protein kinase C (PKC) by TPA in human U937 myeloid leukemia cells is associated with induction of adherence, differentiation, and G0/G1 cell cycle arrest. In this study, we demonstrate that in addition to these differentiating cells about 25% of U937 cells accumulated in the subG1 phase after TPA treatment. This effect proved to be phorbol ester-specific, since other compounds such as retinoic acid or vitamin D3 failed to induce apoptosis in conjunction with differentiation. Only a specific inhibitor of PKC, GF109203X, but not the broad-spectrum kinase inhibitor staurosporine or a tyrosine kinase inhibitor genistein could reverse the induction of apoptosis. Bryostatin-1, another specific PKC activator with distinct biochemical activity failed to induce apoptosis. Moreover, bryostatin-1 completely abolished the induction of apoptosis in U937 cells even if added 8 hours after TPA treatment. Apart from apoptosis induced by various chemotherapeutic drugs, TPA-related cell death is not mediated by an autocrine Fas-FasL loop and could not be prevented by a blocking antibody to the Fas receptor. However, a 75% reduction in the number of apoptotic cells after TPA stimulation was achieved by preincubation with a blocking antibody to the TNFa receptor. Tetrapeptide cleavage assays revealed a four-fold increase in the DEVD-cleavage activity in U937 cells compared to a three-fold increase in TUR cells. Immunoblotting demonstrated that TUR cells did not activate significant levels of caspase-3 or -7, whereas in U937 cells a 20-kDa cleavage product corresponding to activated caspase-3 was detectable after 3 d TPA exposure. Moreover, immunoblots revealed a strongly reduced expression of the adaptor molecule APAF-1, which is required for cytochrome c-dependent activation of caspase-9 and subsequently caspase-3. APAF-1 proved to be inducible after PKC activation with phorbol ester in U937, but not in TUR cells. Thus, APAF-1 expression may, at least in part, be regulated by PKC activity and reduced Dr. Gerold Meinhardt, Medizinische Klinik Innenstadt, Department of Hematology/Oncology, Ludwig-Maximilians-University, Ziemssenstrasse 1, D-80336 Munich/Germany, e-mail: geroldmeinhardt@yahoo. de, Fax: 49 89 5160 4412.
1)
APAF-1 levels are associated with resistance to various inducers of apoptosis. Furthermore, TPA exposure of U937 cells is associated with increased levels of the pro-apoptotic proteins Bak and Bcl-xS, whereas simultaneously a decline in the Bcl-2 expression was noticable.
Abbreviations. PI Propidium iodide. ± PKC Protein kinase C. ± TPA 12-OTetradecanoylphorbol-13-acetate. ± RA Retinoic acid. ± vitD3 1,25Dihydroxyvitamin D3.
Introduction Proliferation, differentiation, retrodifferentiation, and apoptosis represent mutually exclusive cellular responses to distinct stimuli (Hass, 1994). The molecules contributing to the respective signaling pathways include proteins that may participate in diverse signaling cascades. In this context, particularly the PKC family of serine/threonine kinases has been implicated in the regulation of such processes in many cell types. Activation of PKC by the endogenous substrate diacylgylcerol or pharmacological compounds such as phorbol esters that bind to most PKC isoforms leads to induction of transcription factors like AP-1, Ets-1 (Naito et al., 1998), NFkB (Liu et al., 1991) and others, resulting in a cellular response mediated by a variety of other signaling pathways linked to PKC. These PKC-mediated responses have been extensively studied with respect to proliferation and differentiation (Nishizuka, 1988, 1992). Nonetheless, current knowledge about the role of PKC in the regulation of factors crucial to the apoptotic machinery is still sparse. Functional studies have demonstrated that activation of PKC in many cell types relays resistance to or delay of apoptosis in response to many different stimuli (Lotem et al., 1991). This protective effect may result from induction of anti-apoptotic signals or modulation of anti-apoptotic pathways. Apoptosis is a tightly regulated process involving an increasingly complex network of proteins inducing, regulating and executing programmed cell death. (Cryns and Yuan, 1998)
0171-9335/00/79/11-824 $15.00/0
EJCB
Among them, the family of Bcl-2-related proteins with proand anti-apoptotic members forming homo- or heterodimers plays a central role (Yang and Korsmeyer, 1996). Several reports have focussed on the role of the Bcl-x protein, which includes the long (L) splice variant preventing apoptosis and the short (S) splice product promoting apoptosis. Treatment of K562 cells with phorbol ester resulted in differentiation towards the megakaryocytic lineage associated with an increase in the levels of Bcl-x. Moreover, phorbol ester-induced megakaryocytic differentiation was facilitated by the overexpression of Bcl-x in K562 cells, whereas the overexpression of Bcl-xS induced differentiation in the same cell line (Ray et al., 1996; Terui et al., 1998). In the erythropoietin-dependent cell line AS-E2, treatment with Epo or TPA resulted in upregulation of Bcl-xL (Tsushima et al., 1997). Furthermore, in HL-60 and U937 cells 12,13-phorbol dibutyric acid has been postulated to induce up-regulation of Bcl-xL (Chatterjee et al., 1997). During early stages of human hematopoietic progenitor maturation, Bcl-x protein expression was shown to increase with continuance of this high expression level in monocytes/ macrophages, but not in granulocytes (Sanz et al., 1997). Another protein of the Bcl-2 family, Mcl1, can be induced in response to PKC activation or GM-CSF stimulation in leukemic cells (Chao et al., 1998; Townsend et al., 1998, 1999). In contrast, in human leukemic HL-60 cells, activation of PKC by phorbol ester leads to differentiation coupled with down-regulation of the anti-apoptotic Bcl-2 protein (Benito et al., 1995; Sakakura et al., 1996). Previous work has demonstrated that features of apoptosis such as internucleosomal DNA fragmentation occur during the PKC-mediated differentiation and retrodifferentiation process of leukemic cells induced to differentiate with phorbol ester (Gunji et al., 1992). In this context, retrodifferentiation and apoptosis represent alternative means to evade terminal differentiation: Whereas the majority of differentiated cells can enter a cellular program resulting in reversal of the differentiated phenotype, up to 25% of differentiated cells undergo programmed cell death (Gunji et al., 1992). Furthermore, recent data indicate that activation of PKC by phorbol ester can induce differentiation and apoptosis in the same leukemic clone in parallel (manuscript submitted). Accordingly, a certain population of leukemic cells undergoes apoptosis without prior acquisition of a differentiated phenotype. Among the Ca2-independent PKC isoforms, activation of the d form of PKC has been implicated in the induction of differentiation, apoptosis, and cell cycle regulation in leukemic cells (Watanabe et al., 1992; Mischak et al., 1993; Emoto et al., 1995). In particular, phorbol ester-activated PKCd in conjunction with PKCa has been linked to the maturation of murine 32D leukemic precursor cells, whereas caspase-activated PKCd was shown to coincide with and induce apoptosis in U937 myeloid leukemic and other cells. Recently, PKCd has been linked to activation of caspase-3, suggesting a role both downstream and upstream of the caspase cascade (Reyland et al., 1999). Other PKC isoforms, particularly PKCq, have been shown to be involved in later stages of apoptosis as substrates of caspases (Datta et al., 1997; Mizuno et al., 1997). However, little is known on how activation of PKC contributes to the differential regulation of apoptosis and differentiation in leukemic cells. Therefore, we examined whether PKC-mediated induction of differentiation and cell cycle arrest is associated with concomitant induction of anti- or pro-apoptotic pathways.
PKC activation and apoptosis 825
Material and methods Cell lines and reagents
Human U937 myeloid leukemic cells (American Type Culture Collection, Rockville, MD) and the differentiation-resistant subclone TUR were grown in RPMI 1640 medium containing 10% heat-inactivated fetal calf serum (FCS) (BioWhittaker, Belgium) supplemented with 2 mM L-glutamine without antibiotics. The cells were incubated in a humidified atmosphere (37 8C, 5% CO2) and treated with various concentrations of the phorbol ester 12-O-tetradecanoylphorbol-13acetate (TPA) (Sigma, Germany) for the time points indicated. For some experiments, the cells were preincubated with specific inhibitors of caspase-1, YVAD-CHO, and caspase-3/-7, DEVD-CHO (Biomol, Germany), respectively. GF109203X, bryostatin-1, retinoic acid, and vitamin D3 were purchased from Biomol, Germany. The anti-Fas IgG antibody and the anti-TNFa-receptor IgG antibody used to block the receptors were from Upstate Biotechnology, Germany.
Flow cytometry
For cell cycle analysis, untreated and stimulated U937 and TUR cells (106 cells/ml) were fixed with 70% (w/v) ice-cold ethanol overnight. Adherent cells were first scraped from the plastic surface of the culture flask. After two washes with ice-cold PBS, the fixed cells were resuspended in 1 ml of PBS containing 40 mg/ml propidium iodide (Sigma, Germany) and 500 U/ml RNase A (Boehringer Mannheim). Following incubation for 30 min in the dark the cells were analysed on an EPICS XL-MCL flow cytometer (Coulter, Germany) using System II software.
Caspase-3/-7 activity assay
The assay was performed according to the manufacturer s instructions (Clontech, Heidelberg, Germany). Briefly, cells were resuspended in 50 ml lysis buffer and incubated for 10 min on ice. Fifty ml of 2 reaction buffer containing 1,4-dithiothreitol (DTT) and 1 ml of the caspase-3/-7 inhibitor were added to the inhibitor sample and incubated for 30 min at 37 8C, while the other samples were kept on ice. Subsequently, 50 ml of 2 reaction buffer (with DTT) were added to the remaining samples. Five ml of the conjugated substrate were added to each sample and incubated at 37 8C for 1 h. Thereafter, sample volumes were adjusted to 1.0 ml with distilled water and the absorbance at 405 nm was measured with a spectrophotometer.
Immunoblot analysis
Untreated and stimulated U937 and TUR cells were washed three times in ice-cold PBS and lysed in a buffer containing 10 mM Tris-HCl (pH 7.6), 137 mM NaCl, 1 mM Na3VO4, 10 mM NaF, 10 mM EDTA, 1% (v/v) NP-40 with the addition of 10 mg/ml aprotinin, 10 mg/ml leupeptin, and 1 mM phenylmethylsulfonylfluoride (PMSF). Protein concentration was adjusted using a colorimetric assay. Controls for equal loading of protein were performed using an antibody against bactin. Proteins were subjected to SDS-polyacrylamide gel electrophoresis and transferred to a PVDF membrane (Millipore, Germany). The transfer buffer contained 25 mM Tris, 192 mM glycine, 0.037% (w/v) SDS and 20% (v/v) methanol. The membranes were blocked with PBS containing 5% dried milk and 0.05% Tween-20 (PBS/Tween). After washing four times with PBS/Tween, the membranes were incubated with appropriate primary antibodies (all from Santa Cruz Biotechnology, Santa Cruz, CA) and visualized by autoradiography using the ECL detection kit (Amersham, Germany). For preparation of membrane and cytosolic fractions, cells were swollen on ice in a buffer containing 25 mM Tris-HCl, pH 7.6, 10 mM NaF, 10 mM EDTA, 1 mM Na3VO4, with the addition of 10 mg/ml aprotinin, 10 mg/ml leupeptin, 1 mM phenylmethylsulfonylfluoride (PMSF), lysed with 30 strokes of a homogenizer type S, and centrifuged at 15 000g for 15 min at 4 8C. After separation of the supernatant containing cytosolic proteins, the pellet, containing the membrane fraction, was incubated with buffer containing 1% NP-40 for an additional 30 min.
826 G. Meinhardt, J. Roth, G. Totok
EJCB
Results Induction of differentiation-sensitive U937 leukemia cells with TPA in contrast to the differentiation-resistant TUR subclone leads to pronounced apoptosis
Treatment of U937 myeloid leukemia cells with 5 nM TPA was associated with a complete growth arrest after 72 hours. Whereas most cells enter a non-proliferative G0/G1 cell cycle phase coupled with induction of differentiation markers (Hass et al., 1989, 1990; Meinhardt and Hass, 1995), a fraction of the cells undergoes apoptosis with chromatin condensation, membrane blebbing, appearance of apoptotic bodies, annexin V-binding (data not shown) and DNA fragmentation in 8% of the cells after 24 hours and 25% after 72 hours as determined by analysis of propidium iodide stained cells by flow cytometry (Figure 1). In comparison, we tested the induction of apoptosis in the related, U937-derived human leukemic TUR cell line, which is resistant to phorbol ester-induced differentiation and cell cycle arrest by a mechanism involving a reduced expression of distinct PKC isoforms (Hass et al., 1993, 1994). Incubation of TUR cells with 5 nM TPA for 24 and 72 hours resulted in no increase in the number of apoptotic cells (Figure 1). Preincubation of U937 cells with 1 mM of the specific PKC inhibitor GF109203X completely abolished apoptosis induced by treatment of the cells with 5 nM TPA for 48 hours (data not shown).
Retinoic acid or 1,25-dihydroxyvitamin D3 (vitD3), inducers of leukemia cell differentiation, cause cell cycle arrest, but not induction of apoptosis
Fig. 1. Induction of differentiation-sensitive U937 leukemia cells with TPA, but not the differentiation-resistant TUR subclone, leads to pronounced apoptosis. The cells were treated with 5 nM TPA for the indicated time points. The cells were fixed in ethanol and subjected to flow cytometry following digestion of RNA and staining of DNA with propidium iodide. The values represent the number of cells displaying hypodiploid DNA content (subG1-DNA) and are the mean of at least three independent experiments SD.
The ultimate fate of differentiated cells is either terminal commitment and subsequent apoptotic cell death or activation of a cellular program, termed retrodifferentiation, resulting in a reversal of the differentiated phenotype and reentry into the cell cycle. To investigate whether the apoptotic population of cells after TPA treatment is related to activation of signaling pathways leading to differentiation in general, or specifically arises from TPA-mediated PKC-activation, U937 cells were
treated with two distinct inducers of myeloid cell maturation, namely, retinoic acid (RA) and 1,25-dihydroxyvitamin D3 (vitD3). Accordingly, 1 mM RA for 4 days caused a G0/G1arrest of 81% of the population, whereas 500 nM vitD3 induced a cell cycle arrest in 66% of the cells, compared to 53% of proliferating U937 cells, demonstrating the efficacy of the concentrations applied (Figure 2a). Both agents failed to
Fig. 2. Retinoic acid or 1,25-dihydroxyvitamin D3 (vitD3), inducers of leukemia cell differentiation, cause cell cycle arrest, but not induction of apoptosis. U937 cells were treated with 5 nM TPA, 1 mM retinoic acid (RA), or 500 nM 1,25-dihydroxyvitamin D3 (vitD3) for the indicated time points. For the co-incubations, TPA was added without prior removal of RA or vitD3. The number of cells in subG1
(apoptotic) or G1 cell cycle phase was determined by flow cytometry following digestion of RNA and staining of DNA with propidium iodide. The values represent the percentage of cells in a given cell cycle phase and are the mean of at least three independent experiments SD. (a) Percentage of cells in G1 cell cycle phase. (b) Percentage of apoptotic cells.
EJCB
PKC activation and apoptosis 827
induce apoptosis above a background level of 3%, which is in the range of untreated U937 cells. Preincubation with both RA and vitD3, however, decreased the amount of apoptotic cells after TPA-mediated PKC activation by 20% and 50%, respectively, indicating that the activation of pathways mediating RA- and vitD3-related differentiation is distinct from and interferes with PKC-induced cell death (Figure 2b).
PKC activation by bryostatin-1 does not induce apoptosis
Bryostatin-1, a macrocyclic lactone, is another specific activator of PKC, which has recently entered clinical trials for the treatment of various cancer types. Bryostatin-1 is known to induce isozyme-specific down-regulation of PKCs, thus partially antagonizing the differentiating effects of phorbol ester. However, little is known about the effects of this compound on the regulation of apoptosis. 5 to 100 nM of bryostatin-1 for 72 hours induced no significant accumulation of subG1 cells. Moreover, pretreatment of cells with bryostatin-1 for 1 hour followed by 5 nM TPA for 72 hours largely abolished the apoptosis-inducing effects of the phorbol ester. This inhibitory effect of bryostatin-1 was also observed when bryostatin-1 was added up to eight hours after incubation of U937 cells with TPA. Bryostatin-1 added after 24 hours, however, could no longer reverse the activation of the pro-apoptotic PKCmediated pathway, indicating that signaling intermediates downstream of PKC become activated after 8 to 24 hours of incubation with phorbol ester (Figure 3a).
Differential regulation of PKC isoforms after treatment with TPA or bryostatin-1 in U937 and TUR cells
Activation of PKC leads to translocation of isoforms to different cellular compartments and subsequent down-regulation of the kinases. We investigated this differential regulation after stimulation with TPA or bryostatin-1 for 72 hours by Western blotting. Whereas little if any change in the expression and translocation of PKCq could be observed, incubation of U937 cells with TPA and bryostatin-1 induced membrane translocation of the PKC isoforms b and d, respectively (Figure 3b). In contrast to TPA, however, bryostatin-1 led to a down-regulation of PKC b and d after 72 hours. Interestingly, a different pattern of expression and translocation could be detected in TUR cells. Most notably, bryostatin-1 led to downregulation of PKCd, but failed to induce membrane translocation of this particular isoform (Figure 3B). These findings suggested, that bryostatin-1 in comparison to TPA is capable to translocate and down-regulate PKC with distinct isoform specificity.
PKC inhibition and activation during early retrodifferentiation
PKC-activated, differentiated cells down-regulate PKC. Restauration of PKC content, in addition to other, only partially identified factors, leads to resumed proliferation and retrodifferentiation. A significant number of cells, however, fail to undergo retrodifferentiation and thus become committed to apoptosis. To address the question of PKC involvement in the process of apoptosis following differentiation, cells were treated with 5 nM TPA for 3 days. The adherent, differentiated population was then thoroughly washed and challenged with either the specific PKC inhibitor GF109203X, the phorbol ester-unrelated PKC activator bryostatin-1 or another course
Fig. 3. a PKC activation by bryostatin-1 does not induce apoptosis. U937 cells were treated with the indicated concentrations of TPA or bryostatin-1 (BRYO) alone or as co-incubations. The analysis of apoptotic cells was performed as described in Figure 1. b. Translocation of PKC isoforms after treatment with TPA or bryostatin-1. U937 and TUR cells were treated with 5 nM TPA or 100 nM bryostatin-1 for 72 hours. Cytosolic and membrane fractions were subjected to Western blotting with the indicated antibodies using enhanced chemiluminescence for detection of proteins. Actin controls indicated equal loading of cytosolic and membrane proteins, respectively.
of TPA, each for additional 3 days. Whereas approximately 25% of cells treated with TPA for 72 h underwent apoptosis, treatment of the differentiated, vital cells with TPA for another 3 days resulted in an additional loss of 23% of cells due to apoptosis. This compares to 17% of cells cultured only in the presence of medium and 11% or 12% of cells, respectively, further challenged with bryostatin or GF109203X (Figure 4). This is consistent with a model of restauration of PKC activity in retrodifferentiating cells with PKC function being part of the signals leading to apoptosis following differentiation. Taken together, the observed cellular responses indicate an
828 G. Meinhardt, J. Roth, G. Totok
EJCB
apoptotic response specifically related to phorbol estermediated PKC activation rather than to a common differentiation program.
therapy-induced apoptosis. Preincubation of U937 cells with the caspase-3/-7-related inhibitor DEVD-CHO for 1 hour significantly impaired the TPA-mediated pro-apoptotic signal generated by PKC activation, as demonstrated by a moderate to strong reduction of subG1 cells (manuscript submitted). Further analysis revealed, that 500 mM DEVD added 12 hours after TPA treatment still caused a 60% reduction in the number of apoptotic cells by 60 hours. YVAD-CHO, a specific inhibitor of caspase-1, however, was unable to reduce the generation of subG1 cells if added 12 h after TPA (Figure 5a). Since both caspase inhibitors penetrate membranes rather poorly (manufacturer s information), we used a concentration that proved sufficient for DEVD to reduce the amount of apoptosis in U937 cells induced by VP-16 (etoposide), a potent cytotoxic agent, by 75% after a 4-h incubation period (data not shown). Thus, in the time-frame of TPA-induced differentiation with adherence of the majority of cells occurring 12 h after stimulation (data not shown), caspase-3-mediated cleavage of vital cellular substrates does not seem to occur within the first 12 h of PKC activation. Direct measurement of the activity of caspase-3/-7 in U937 and resistant TUR cells was performed using a colorimetric tetrapeptide substrate. 5 nM TPA for 72 hours caused a four-fold increase in the cleavage activity of U937 cells compared to a three-fold increase in TUR cells, indicating that a certain threshold of caspase activation seems necessary to proceed with the execution phase of the programmed cell death (Figure 5b).
Caspase-3 activation during TPA-mediated PKC activation
Evidence for TNF/TNF-receptor, but not for Fas/ FasL interaction during TPA-induced cell death
Fig. 5. a Caspase-3 activation during TPA-mediated PKC activation. U937 cells were treated with 5 nM TPA for 12 hours. Subsequently, 500 mM of the caspase-1 inhibitor YVAD or the caspase-3/-7 inhibitor DEVD was added and the cells were harvested at the indicated time points. The analysis of apoptotic cells was performed as described in Figure 1. b. Measurement of the activity of caspase-3/-7 in U937 and
resistant TUR cells using a colorimetric tetrapeptide substrate assay. U937 and TUR cells were treated with 5 nM TPA for 3 days. Cytosolic extracts were prepared and the activity of DEVD-caspases was measured using a colorimetric tetrapeptide substrate assay. The cleavage activity is depicted in arbitrary units. The values represent the mean of three independent experiments SD.
Fig. 4. PKC inhibition and activation during early retrodifferentiation. U937 cells were treated with 5 nM TPA for 3 days. Subsequently, the differentiated adherent cells were washed several times to remove the TPA and then either treated with the indicated concentrations of bryostatin-1, the PKC inhibitor GF109203X, TPA, or left untreated for an additional 3 days. The analysis of apoptotic cells was performed as described in Figure 1.
The caspases, a family of related proteases, have been implicated in virtually all aspects of apoptotic cell death. Among them, caspase-1 (ICE) as well as caspase-3 and -7 play important roles in inflammatory, death receptor- and chemo-
Recent evidence suggests that chemotherapy-induced apoptosis may at least in part be mediated by an autocrine mechanism involving interaction of induced Fas ligand with its receptor (Friesen et al., 1996). Consequently, we examined the
EJCB
possible involvement of a FasL-Fas autocrine loop in U937 cells which constitutively express Fas (data not shown). Blockage of Fas in untreated U937 control cells with an excess of a non-activating antibody (Fas IgG) (McGahon et al., 1998) caused no apparent cell death. Moreover, preincubation of cells with this Fas IgG for 15 min did not reduce the number of subG1 cells after TPA stimulation, excluding the possibility of an autocrine loop. In contrast, an increase in the number of apoptotic cells was observed after 72 hours of TPA following Fas IgG incubation (Figure 6a). Accordingly, a possible autocrine TNF/TNF-receptor loop was investigated. Whereas an excess of the non-activating TNFa-receptor antibody did not result in enhanced cell death, pre-incubation of this antibody 15 min prior to TPA stimulation significantly reduced apoptosis by more than 75% (Figure 6b). Thus, TPA-induced apoptosis is at least in part mediated by activation of the TNF/TNF-receptor pathway, but not the Fas/FasL pathway. Preincubation with up to 150 mM of the tetrapeptide inhibitor Z-IETD-fmk, however, did not reverse the proapoptotic effect of TPA (data not shown).
PKC activation and apoptosis 829
differentiation, an early increase in expression of Mcl1 could be detected as early as 3 h following treatment with TPA. The levels of Mcl1 remained elevated throughout the observed 72h period. A similar increase in Mcl1 expression was observed in the resistant TUR cells with a time-course paralleling that of U937 cells, indicating that TUR cells retain at least some suceptibility to TPA stimulation despite their failure to undergo phorbol ester-induced differentiation or apoptosis. Whereas in TUR cells a slight increase in the levels of Bcl-2 protein could be detected, U937 cells down-regulated Bcl-2 after 72 h independent of phorbol ester concentration (Figure 7a).
Expression of APAF-1 and cleavage of caspases during TPA treatment
Bcl-2 proteins govern the initiation or execution of apoptotic signaling cascades. Therefore, we determined the expression of relevant members of this protein family at various time points after incubation with 5 ± 500 nM TPA. No relevant change in the expression levels of the pro-apoptotic proteins Bax and Bad was observed at all time points and TPA concentrations in U937 cells. There was a significant increase, however, in the levels of Bak and the short form of the Bcl-x protein after 72 h TPA. Interestingly, the enhanced expression of both proapoptotic proteins was observed at a time, where the cells were already in the late execution phase of either differentiation or apoptosis. Another pro-apoptotic protein, Bag, showed little if any variation in expression levels. Concomitant with induced
The APAF-1 protein is a central component of the apoptosome complex, being comprised of cytochrome c, released from its mitochondrial origin, pro-caspase-9, and ATP. Formation of this complex leads to activation of caspase-9 and subsequently downstream caspases, such as caspase-3 or -7. There is very little knowledge about the regulation of APAF-1 expression in stimulated and apoptosis-resistant cells. Consequently, APAF-1 expression was analysed in response to PKC activation in U937 and TUR cells. Treatment with TPA for up to 72 h resulted in enhanced expression of APAF-1 in U937 cells (Figure 7b). The resistant TUR cells demonstrated a highly reduced level of APAF-1 expression as determined by Western blotting of membranes overexposed after the chemiluminescence reaction (data not shown) and no apparent APAF induction after TPA stimulation (Figure 7b). In addition, the activation of caspase-3 was observed only in the TPAstimulated U937 cells by detecting an approximately 20-kDa protein corresponding to a cleaved fragment of caspase-3. In this context, no activation of caspase-7 could be observed in either cell line after TPA treatment (Figure 7b).
Fig. 6. Evidence for TNF/TNF-receptor, but not for Fas/FasL interaction during TPA-induced cell death. (a) U937 cells were treated with a blocking antibody against Fas (Fas-IgG) or 5 nM TPA for 1 or 3 days. For the co-incubations, the cells were stimulated with TPA 1 hour after the addition of Fas-IgG. (b) U937 cells were treated with a blocking
antibody against the TNFa receptor (TNF-IgG) or 5 nM TPA for 1 or 3 days. For the co-incubations, the cells were stimulated with TPA 1 hour after the addition of TNF-IgG. The values represent the mean of three independent experiments SD. The analysis of apoptotic cells was performed as described in Figure 1.
Expression of Bcl-2 familiy proteins
830 G. Meinhardt, J. Roth, G. Totok
EJCB
Fig. 7. a Expression of Bcl-2 familiy proteins. U937 and TUR cells were treated with TPA for the indicated time points and concentrations. Lysates were subjected to Western blotting with the indicated antibodies using enhanced chemiluminescence for detection of proteins. The actin probe served as control to confirm equal loading of
proteins. The results shown are representative of several independent experiments. b. Expression of APAF-1 and cleavage of caspases during TPA treatment. U937 and TUR cells were treated with 5 nM TPA for 3 days and subjected to Western blotting as described in (a).
Discussion
(data not shown). Moreover, the acquisition of a differentiated phenotype requires approximately 2 ± 3 days in U937 (Hass et al., 1992), whereas inhibition of caspases more than 24 h following TPA treatment is no longer effective in preventing apoptosis (unpublished data). Previous work has confirmed, that stimulation of U937 and related leukemic cell lines with compounds such as retinoic acid or vitD3 results in a G0/G1 cell cycle arrest with concomitant induction of differentiation, which is independent of direct PKC activation (Hass et al., 1993). We have demonstrated, that both retinoic acid and vitD3 do not lead to a simultaneous appearance of apoptotic cells beyond the level of untreated controls, thus suggesting that the observed cell death following TPA treatment is intrinsically linked to activation of PKC rather than induction of a general cellular differentiation program. This is underscored by the ability of the specific PKC inhibitor GF109203X to fully prevent TPA-mediated apoptosis.
Programmed cell death is a multi-step process involving an increasingly complex network of proteins contributing to the induction, regulation and execution of apoptosis. We attempted to examine the contribution of PKC activation upon the regulation of certain aspects of the apoptotic process. Activation of PKC in myeloid leukemic cells by phorbol ester treatment is commonly associated with differentiation along the monocytic lineage (Hass et al., 1989). We have demonstrated in this paper, that approximately 25% of a monoclonal U937 population derived from single cell suspensions fails to undergo G0/G1 cell cycle arrest and exhibits features of apoptotic cells. This induction of apoptosis is apparently not a consequence of prior differentiation with subsequent cell death, since the fraction of apoptotic cells does not exhibit markers commonly associated with monocytic cell maturation
EJCB
Some insight into the mechanisms of PKC activation and function stems from the observation that bryostatin-1 in contrast to phorbol ester fails to induce differentiation or proliferation in distinct cell lines, at least in part by differential down-regulation of PKC isoforms such as a, d and e (Szallasi et al., 1994). Our data support the hypothesis of differential PKC activation by demonstrating that byrostatin-1 significantly blocks TPA-mediated apoptosis in U937 cells, while by itself inducing neither differentiation nor apoptosis. The observed differences in the expression levels and translocation of the PKC isoforms suggest involvement in the apoptotic response. Most recently, PKCd has been shown to translocate to the mitochondrial membrane and has been implicated in the regulation of caspase-3 activation (Li et al., 1999; Reyland et al., 1999). Our data, demonstrating down-regulated PKCd especially in the membrane fraction in bryostatin-1-treated cells, is consistent with this model. Thus, the antagonistic effect of bryostatin-1 upon the TPA-mediated induction of apoptosis may be caused by a reduced amount of membrane-associated PKCd being required for caspase activation. It is unknown whether phorbol ester-related differentiation and apoptosis may require target proteins other than PKC in myeloid cells. However, this possibility has not been formally ruled out and additional phorbol ester acceptors have been identified (Areces et al., 1994). In view of recent results establishing the regulation of caspase-9 and Bcl-2 proteins such as Bad and Bcl-2 by phosphorylation (Haldar et al., 1995; Zha et al., 1996; Cardone et al., 1998), it is conceivable that PKC isoforms may have a role upstream of caspase activation. In this context, both Bad and Bcl-2 are phosphorylated on serine residues, possibly by a mechanism involving the Raf-1 serine/threonine-kinase. (Wang et al., 1994, 1996) Bcl-2 can target Raf-1 kinase to mitochondria, resulting in phosphorylation of Bad (Wang et al., 1996). Phosphorylated Bad is unable to bind to and inactivate Bcl-2 or Bcl-xL, thus allowing a proapoptotic signal to proceed possibly towards caspase activation (Zha et al., 1996). Raf-1 kinase was shown to be activated following disruption of microtubular architecture and can bind via its catalytic domain to Bcl-2 in a BH4 domain-dependent manner. While Bcl-2 is not phosphorylated by Raf-1 directly, this effect requires the activity of another serine-/threoninekinase (Wang et al., 1994). In this context, PKCa has been shown to co-localize with Bcl-2 in mitochondrial membrane preparations of leukemic HL-60 cells as well as to phosphorylate Bcl-2 in vitro at serine 70. Moreover, overexpression of PKCa induces enhanced Bcl-2 phosphorylation and a substantial increase in resistance to drug-induced apoptosis (Ruvolo et al., 1998). Previous as well as unpublished results show that TUR cells contain reduced levels of several PKC isoforms relative to U937 cells, among them bII and d (manuscript submitted) (Hass et al., 1993). Therefore, enhanced Bcl-2 phosphorylation is less likely to be a mechanism of apoptosis resistance in TUR cells compared to differences in the levels of protein expression. Differences in the expression of certain Bcl-2-like proteins have been linked to an increase in the suceptibility to cytotoxic agents. Increased expression of Bax has been shown to be sufficient to induce the release of mitochondrial cytochrome c and activation of pro-caspase-9 without prior opening of the mitochondrial permeability transition pore (Eskes et al., 1998). TPA treatment of U937 cells, however, is not associated with increased levels of Bax, but induces other pro-apoptotic factors, namely Bak and Bcl-xS. Conversely, the anti-apoptotic
PKC activation and apoptosis 831
Bcl-2 is down-regulated in U937 cells. The data presented here show an increase in Bak and Bcl-xS expression coupled with reduced Bcl-2 only after more than 24 h, thus making it less likely, that the induction of these factors alone leads to caspase activation, which is observed after 12 h. Enhanced expression of Bak and Bcl-xS may be a necessary step in an amplification process leading to an enhancement of the pro-apoptotic signals (Thompson, 1998). This might be in concert with observations (including ours) that activation of caspases and appearance of an apoptotic phenotype can be separated by a prolonged lag period of unknown function. The induction of Bcl-xS expression in U937 cells with a constant level of Bcl-xL after 72 h independent of TPA concentration as consistently observed by us is in contrast to previously published results showing enhanced expression of Bcl-xL after treatment with the related phorbol ester 12,13-phorbol dibutyric acid (Chatterjee et al., 1997). Recent results have established that non-receptor-mediated apoptosis involves release of cytochrome c from mitochondria. Cytosolic cytochrome c then associates with APAF-1 and caspase-9 in the presence of ATP, allowing for the autoprocessing of the caspase-9 zymogen form to the active dimeric protease and subsequent activation of downstream effectors (Li et al., 1997). In addition, APAF-1 is able to associate with other caspases like -4 und -8. Binding of APAF-1 to the antiapoptotic protein Bcl-xL prevents caspase-9 activation, suggesting that APAF-1 serves as adaptor protein linking the caspase executioners to upstream regulators (Hu et al., 1998; Pan et al., 1998). Homozygous mutants of murine APAF-1 die at embryonic day 16 and are unable to activate caspase-3 (Cecconi et al., 1998). Thus, it is conceivable, that the severely reduced expression of APAF-1 in TUR cells might play a central role in the observed resistance to a variety of apoptotic inducers. There is little information about the regulation of APAF-1 expression as well as its role in the resistance to apoptosis. We provide evidence that APAF-1 protein levels are upregulated during induced differentiation of U937 cells. Interestingly, TPA-treated TUR cells show an increase in DEVD-tetrapeptide cleavage activity, which extends to 75% of the respective activity of TPA-stimulated U937 cells. DEVD is a consensus cleavage site for both caspase-3 and -7 as well as a potential cleavage site for other caspases (Kidd, 1998). We could not detect processing of caspase-3 and -7 in TPA-treated TUR cells by immunoblotting, whereas the typical 20-kDa protein band of cleaved caspase-3 could be readily detected in U937 cells. Thus, TUR cells may activate other DEVDsensitive caspases, but apparently below a threshold that is required for proceeding with the execution phase of apoptosis. Activation of the downstream caspases can also be achieved by stimulation of members of the TNF receptor superfamily. Recently, activation of the Fas receptor by an autocrine FasL/ Fas loop has been postulated as a possible mechanism of chemotherapy-induced apoptosis (Friesen et al., 1996). Our results rule out an involvement of Fas in TPA-mediated apoptosis. In contrast, we demonstrate employment of the TNF/TNF-receptor pathway in the process of PKC-activated cell death. This is in concert with a recent study linking phorbol ester-induced TNFa production and apoptosis in the same cell line (Takada et al., 1999). The mechanisms of TNF-induced cell death in U937 cells needs to be elucidated, since our data could not prove involvement of caspase-8. Taken together, our data demonstrate that activation of PKC in U937 leukemic cells is accompanied by marked
832 G. Meinhardt, J. Roth, G. Totok
alterations in the expression of proteins governing apoptosis. These alterations are largely absent in the PKC-deficient U937 subclone TUR. It will be necessary to determine the exact contribution of individual PKC isoforms to the induction of apoptosis as well as to regulation of the expression of certain apoptosis proteins. Acknowledgement. This work was supported by a grant from the Deutsche Forschungsgemeinschaft (Me 1189/2) to G. Meinhardt. Some experiments are part of the thesis work of G. Totok.
References Areces, L. B., Kazanietz, M. G., Blumberg, P. M. (1994): Close similarity of baculovirus-expressed n-chimaerin and protein kinase C alpha as phorbol ester receptors. J. Biol. Chem. 269, 19553 ± 19558. Benito, A., Grillot, D., Nunez, G., Fernandez-Luna, J. L. (1995): Regulation and function of Bcl-2 during differentiation-induced cell death in HL-60 promyelocytic cells. Am. J. Pathol. 146, 481 ± 490. Cardone, M. H., Roy, N., Stennicke, H. R., Salvesen, G. S., Franke, T. F., Stanbridge, E., Frisch, S., Reed, J. C. (1998): Regulation of cell death protease caspase-9 by phosphorylation. Science 282, 1318 ± 1321. Cecconi, F., Alvarez-Bolado, G., Meyer, B. I., Roth, K. A., Gruss, P. (1998): Apaf1 (CED-4 homolog) regulates programmed cell death in mammalian development. Cell 94, 727 ± 737. Chao, J. R., Wang, J. M., Lee, S. F., Peng, H. W., Lin, Y. H., Chou, C. H., Li, J. C., Huang, H. M., Chou, C. K., Kuo, M. L., et al. (1998): mcl-1 is an immediate-early gene activated by the granulocyte-macrophage colony-stimulating factor (GM-CSF) signaling pathway and is one component of the GM-CSF viability response. Mol. Cell. Biol. 18, 4883 ± 4898. Chatterjee, D., Han, Z., Mendoza, J., Goodglick, L., Hendrickson, E. A., Pantazis, P., Wyche, J. H. (1997): Monocytic differentiation of HL-60 promyelocytic leukemia cells correlates with the induction of Bcl-xL. Cell Growth Differ. 8, 1083 ± 1089. Cryns, V., Yuan, J. (1998): Proteases to die for. Genes Dev. 12, 1551 ± 1570. Datta, R., Kojima, H., Yoshida, K., Kufe, D. (1997): Caspase-3mediated cleavage of protein kinase C theta in induction of apoptosis. J. Biol. Chem. 272, 20317 ± 20320. Emoto, Y., Manome, Y., Meinhardt, G., Kisaki, H., Kharbanda, S., Robertson, M., Ghayur, T., Wong, W. W., Kamen, R., Weichselbaum, R., et al. (1995): Proteolytic activation of protein kinase C delta by an ICE-like protease in apoptotic cells. EMBO J. 14, 6148 ± 6156. Eskes, R., Antonsson, B., Osen-Sand, A., Montessuit, S., Richter, C., Sadoul, R., Mazzei, G., Nichols, A., Martinou, J. C. (1998): Baxinduced cytochrome c release from mitochondria is independent of the permeability transition pore but highly dependent on Mg2 ions. J. Cell Biol. 143, 217 ± 224. Friesen, C., Herr, I., Krammer, P. H., Debatin, K. M. (1996): Involvement of the CD95 (APO-1/FAS) receptor/ligand system in druginduced apoptosis in leukemia cells. Nature Med. 2, 574 ± 577. Gunji, H., Hass, R., Kufe, D. (1992): Internucleosomal DNA fragmentation during phorbol ester-induced monocytic differentiation and G0/G1 arrest. J. Clin. Invest. 89, 954 ± 960. Haldar, S., Jena, N., Croce, C. M. (1995): Inactivation of Bcl-2 by phosphorylation. Proc. Natl. Acad. Sci. USA 92, 4507 ± 4511. Hass, R. (1994): Retrodifferentiation and cell death. Crit. Rev. Oncog. 5, 359 ± 371. Hass, R., Bartels, H., Topley, N., Hadam, M., Kohler, L., GoppeltStrube, M., Resch, K. (1989): TPA-induced differentiation and adhesion of U937 cells: changes in ultrastructure, cytoskeletal organization and expression of cell surface antigens. Eur. J. Cell Biol. 48, 282 ± 293. Hass, R., Giese, G., Meyer, G., Hartmann, A., Dork, T., Kohler, L., Resch, K., Traub, P., Goppelt-Strube, M. (1990): Differentiation and retrodifferentiation of U937 cells: reversible induction and suppression of intermediate filament protein synthesis. Eur. J. Cell Biol. 51, 265 ± 271.
EJCB
Hass, R., Gunji, H., Datta, R., Kharbanda, S., Hartmann, A., Weichselbaum, R., Kufe, D. (1992): Differentiation and retrodifferentiation of human myeloid leukemia cells is associated with reversible induction of cell cycle-regulatory genes. Cancer Res. 52, 1445 ± 1450. Hass, R., Hirano, M., Kharbanda, S., Rubin, E., Meinhardt, G., Kufe, D. (1993): Resistance to phorbol ester-induced differentiation of a U937 myeloid leukemia cell variant with a signaling defect upstream to Raf-1 kinase. Cell Growth Differ. 4, 657 ± 663. Hass, R., Meinhardt, G., Hadam, M., Bartels, H. (1994): Characterization of human TUR leukemia cells: continued cell cycle progression in the presence of phorbol ester is associated with resistance to apoptosis. Eur. J. Cell Biol. 65, 408 ± 416. Hu, Y., Benedict, M. A., Wu, D., Inohara, N., Nunez, G. (1998): Bcl-XL interacts with Apaf-1 and inhibits Apaf-1-dependent caspase-9 activation. Proc. Natl. Acad. Sci. USA 95, 4386 ± 4391. Kidd, V. J. (1998): Proteolytic activities that mediate apoptosis. Annu. Rev. Physiol. 60, 533 ± 573. Li, L., Lorenzo, P. S., Bogi, K., Blumberg, P. M., Yuspa, S. H. (1999): Protein kinase C delta targets mitochondria, alters mitochondrial membrane potential, and induces apoptosis in normal and neoplastic keratinocytes when overexpressed by an adenoviral vector. Mol. Cell. Biol. 19, 8547 ± 8558. Li, P., Nijhawan, D., Budihardjo, I., Srinivasula, S. M., Ahmad, M., Alnemri, E. S., Wang, X. (1997): Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 91, 479 ± 489. Liu, J. L., Chiles, T. C., Sen, R. J., Rothstein, T. L. (1991): Inducible nuclear expression of NF-kappa B in primary B cells stimulated through the surface Ig receptor. J. Immunol. 146, 1685 ± 1691. Lotem, J., Cragoe, E. J., Jr., Sachs, L. (1991): Rescue from programmed cell death in leukemic and normal myeloid cells. Blood 78, 953 ± 960. McGahon, A. J., Costa Pereira, A. P., Daly, L., Cotter, T. G. (1998): Chemotherapeutic drug-induced apoptosis in human leukaemic cells is independent of the Fas (APO-1/CD95) receptor/ligand system. Br. J. Haematol. 101, 539 ± 547. Meinhardt, G., Hass, R. (1995): Differential expression of c-myc, max and mxi1 in human myeloid leukemia cells during retrodifferentiation and cell death. Leukemia Research 19, 699 ± 705. Mischak, H., Pierce, J. H., Goodnight, J., Kazanietz, M. G., Blumberg, P. M., Mushinski, F. J. (1993): Phorbol ester induced myeloid differentiation is mediated by protein kinase C-alpha and -delta and not by protein kinase c-betaII, -epsilon, -zeta and -eta. J. Biol. Chem. 268, 20110 ± 20115. Mizuno, K., Noda, K., Araki, T., Imaoka, T., Kobayashi, Y., Akita, Y., Shimonaka, M., Kishi, S., Ohno, S. (1997): The proteolytic cleavage of protein kinase C isotypes, which generates kinase and regulatory fragments, correlates with Fas-mediated and 12-O-tetradecanoylphorbol-13-acetate-induced apoptosis. Eur. J. Biochem. 250, 7 ± 18. Naito, S., Shimizu, S., Maeda, S., Wang, J., Paul, R., Fagin, J. A. (1998): Ets-1 is an early response gene activated by ET-1 and PDGF-BB in vascular smooth muscle cells. Am. J. Physiol. 274, C472 ± C480. Nishizuka, Y. (1988): The molecular heterogeneity of protein kinase C and its implications for cellular regulation. Nature 334, 661 ± 665. Nishizuka, Y. (1992): Intracellular signaling by hydrolysis of phospholipids and activation of protein kinase C. Science 258, 607 ± 614. Pan, G., O Rourke, K., Dixit, V. M. (1998): Caspase-9, Bcl-XL, and Apaf-1 form a ternary complex. J. Biol. Chem. 273, 5841 ± 5845. Ray, S., Bullock, G., Nunez, G., Tang, C., Ibrado, A. M., Huang, Y., Bhalla, K. (1996): Enforced expression of Bcl-XS induces differentiation and sensitizes chronic myelogenous leukemia-blast crisis K562 cells to 1-beta-D-arabinofuranosylcytosine-mediated differentiation and apoptosis. Cell Growth Differ. 7, 1617 ± 1623. Reyland, M. E., Anderson, S. M., Matassa, A. A., Barzen, K. A., Quissell, D. O. (1999): Protein kinase C delta is essential for etoposide-induced apoptosis in salivary gland acinar cells. J. Biol. Chem. 274, 19115 ± 19123. Ruvolo, P. P., Deng, X., Carr, B. K., May, W. S. (1998): A functional role for mitochondrial protein kinase C alpha in Bcl2 phosphorylation and suppression of apoptosis. J. Biol. Chem. 273, 25436 ± 25442.
EJCB
Sakakura, C., Sweeney, E. A., Shirahama, T., Hakomori, S., Igarashi, Y. (1996): Suppression of bcl-2 gene expression by sphingosine in the apoptosis of human leukemic HL-60 cells during phorbol esterinduced terminal differentiation. FEBS Lett. 379, 177 ± 180. Sanz, C., Benito, A., Silva, M., Albella, B., Richard, C., Segovia, J. C., Insunza, A., Bueren, J. A., Fernandez-Luna, J. L. (1997): The expression of Bcl-x is downregulated during differentiation of human hematopoietic progenitor cells along the granulocyte but not the monocyte/macrophage lineage. Blood 89, 3199 ± 3204. Szallasi, Z., Smith, C. B., Pettit, G. R., Blumberg, P. M. (1994): Differential regulation of protein kinase C isozymes by bryostatin 1 and phorbol 12-myristate 13-acetate in NIH 3T3 fibroblasts. J. Biol. Chem. 269, 2118 ± 2124. Takada, Y., Hachiya, M., Osawa, Y., Hasegawa, Y., Ando, K., Kobayashi, Y., Akashi, M. (1999): 12-O-tetradecanoylphorbol-13acetate-induced apoptosis is mediated by tumor necrosis factor alpha in human monocytic U937 cells. J. Biol. Chem. 274, 28286 ± 28292. Terui, Y., Furukawa, Y., Kikuchi, J., Iwase, S., Hatake, K., Miura, Y. (1998): Bcl-x is a regulatory factor of apoptosis and differentiation in megakaryocytic lineage cells. Exp. Hematol. 26, 236 ± 244. Thompson, E. B. (1998): Special topic: apoptosis. Annu. Rev. Physiol. 60, 525 ± 532. Townsend, K. J., Trusty, J. L., Traupman, M. A., Eastman, A., Craig, R. W. (1998): Expression of the antiapoptotic MCL1 gene product is regulated by a mitogen activated protein kinase-mediated pathway triggered through microtubule disruption and protein kinase C. Oncogene 17, 1223 ± 1234.
PKC activation and apoptosis 833
Townsend, K. J., Zhou, P., Qian, L., Bieszczad, C. K., Lowrey, C. H., Yen, A., Craig, R. W. (1999): Regulation of MCL1 through a serum response factor/Elk-1-mediated mechanism links expression of a viability-promoting member of the BCL2 family to the induction of hematopoietic cell differentiation. J. Biol. Chem. 274, 1801 ± 1813. Tsushima, H., Urata, Y., Miyazaki, Y., Fuchigami, K., Kuriyama, K., Kondo, T., Tomonaga, M. (1997): Human erythropoietin receptor increases GATA-2 and Bcl-xL by a protein kinase C-dependent pathway in human erythropoietin-dependent cell line AS-E2. Cell Growth Differ. 8, 1317 ± 1328. Wang, H. G., Miyashita, T., Takayama, S., Sato, T., Torigoe, T., Krajewski, S., Tanaka, S., Hovey, L. R., Troppmair, J., Rapp, U. R., et al. (1994): Apoptosis regulation by interaction of Bcl-2 protein and Raf-1 kinase. Oncogene 9, 2751 ± 2756. Wang, H. G., Rapp, U. R., Reed, J. C. (1996): Bcl-2 targets the protein kinase Raf-1 to mitochondria. Cell 87, 629 ± 638. Watanabe, T., Ono, Y., Taniyama, Y., Hazama, K., Igarashi, K., Ogita, K., Kikkawa, U., Nishizuka, Y. (1992): Cell division arrest induced by phorbol ester in CHO cells overexpressing protein kinase C-delta subspecies. Proc. Natl. Acad. Sci. USA 89, 10159 ± 10163. Yang, E., Korsmeyer, S. J. (1996): Molecular thanatopsis: A discourse on the Bcl2 family and cell death. Blood 88, 386 ± 401. Zha, J., Harada, H., Yang, E., Jockel, J., Korsmeyer, S. J. (1996): Serine phosphorylation of death agonist BAD in response to survival factor results in binding to 14 ± 3 ± 3 not BCL-X(L). Cell 87, 619 ± 628.