Skip to main content
NCBI home page
Cell Proliferation logoLink to Cell Proliferation
. 2003 Nov 17;36(6):293–306. doi: 10.1046/j.1365-2184.2003.00286.x

The presence of 19‐kDa Bcl‐2 in dividing cells

R W M Hoetelmans 1,1,✉, C J H Van de Velde 1, J H Van Dierendonck 1
PMCID: PMC6496489  PMID: 14710848

Abstract

Abstract.  The 26‐kDa bcl‐2 gene product inhibits apoptosis and cell proliferation. Cleavage of Bcl‐2 into a 22‐kDa fragment inactivates its anti‐apoptotic activity and is a key event in apoptosis. Here, and in recent work, we describe massive 19‐kDa Bcl‐2 immunoreactivity in non‐apoptotic cells, suggesting a link with viability rather than cell death. Loss of 19 kDa Bcl‐2 in adriamycin‐induced apoptotic cells underlines this. G2/M‐phase accumulation of cells by nocodazole‐treatment also results in loss of 19 kDa Bcl‐2. Next to its well‐documented cytoplasmic localization, a substantial pool of Bcl‐2 resides in nuclei. Hampered nuclear localization of Bcl‐2 leads to a loss of cell cycle repression. This has led us to point at a pivotal role for nuclear Bcl‐2 in cellular proliferation. In this report, cellular fractionation of bcl‐2 transfected cells in various phases of the cell cycle reveals a constitutive cytoplasmic pool of 19 kDa Bcl‐2. Nuclear 19‐kDa Bcl‐2 immunoreactivity is far more pronounced in rapidly dividing nuclei compared with more quiescent nuclear fractions. This implicates that ongoing cell proliferation involves cleavage of nuclear Bcl‐2 with a 19‐kDa fragment.

INTRODUCTION

Apoptosis plays an essential role in developmental processes in the immune and nervous system, raising considerable interest in genes that are capable of regulating apoptosis. One of the most important advances in understanding the basic mechanisms of apoptosis in vertebrates has come from studies of the anti‐apoptotic proto‐oncogene bcl‐2, which was found deregulated in B‐cell lymphomas due to a chromosomal t(14,18) translocation (Tsujimoto et al. 1984). Bcl‐2 initially represented an alternative class of oncoproteins, predisposing cells to exhibit a potent cell cycle inhibitory effect, as shown in vitro and in vivo (Vaux et al. 1988; McDonnell et al. 1990), allowing survival of a variety of apoptotic stimuli (Borner 1996; Mazel et al. 1996). The product of the bcl‐2 gene is still the founding member of an extended family of proteins that have either anti‐apoptotic properties (e.g. Bcl‐2, Bcl‐XL, Mcl‐1, Bcl‐w, A1) or pro‐apoptotic properties (Bax, Bid, Bcl‐XS, Bak, Bad, Bik, Diva, Bim, Bok, Bod, Hrk, Blk).

A key event in apoptosis regulation is dimerization among Bcl‐2 family members mediated by evolutionarily conserved homologous regions, i.e. the Bcl‐2 homology (BH) domains BH1, BH2, BH3, and BH4 (Korsmeyer 1999). The N‐terminal BH4 domain is required for interaction with a panel of non‐structurally related proteins, but was also found to participate in binding the pro‐apoptotic Bax protein (Hirotani et al. 1999). Like Bcl‐2, most Bcl‐2 family members contain a C‐terminal amino acid sequence that allows formation of channels in lipid membranes in vitro (Brenner et al. 2000). Bcl‐2 family members orchestrate the release of mitochondrial apoptogenic factors (e.g. cytochrome c) into the cytosol (Kluck et al. 1997; Brenner et al. 2000). Inhibition of the release of cytochrome c by Bcl‐2 blocks the activation of the apoptosome, a complex residing at the outer mitochondrial membrane and responsible for triggering downstream effector caspases (e.g. caspases‐3 and ‐9) (Zou et al. 1999). Caspases are highly specific proteolytic enzymes that are executioners of the apoptotic machinery by cleaving their target protein at an aspartate (Asp) residue.

The emergence of a 22‐kDa Bcl‐2 form in apoptotic cells has been described (Cheng et al. 1997; Grandgirard et al. 1998; Yamamoto et al. 1998; Fadeel et al. 1999; Fortney et al. 2002). It is generally accepted that this product results from cleavage at the Asp34 residue in the Bcl‐2 protein (with a pivotal role for caspase‐3), thereby releasing the N‐terminal BH4 domain. Yamamoto et al. (1998) described the presence of a 19‐kDa Bcl‐2 fragment in lymphoid cells constitutive, presumably the result of cleavage at a different site than the Asp34 residue. We confirmed recently that the human Bcl‐2 protein, overexpressed in rat CC531 cells (CCbcl2 cells), is present in several forms on western blots, i.e. full‐length 26 kDa, as a minor band at 22 kDa and a relatively abundant 19 kDa expression (Hoetelmans et al. 2003a). Post‐transcriptional modification – such as caspase‐(like)‐mediated proteolysis – is not restricted to the Bcl‐2 protein: Bcl‐XL, a functional homologue of Bcl‐2, is converted into a pro‐apoptotic molecule upon cleavage (Fujita et al. 1998) while the pro‐apoptotic activity of Bax is restrained upon cleavage (Wood & Newcombe 2000).

It has been demonstrated numerous times that an overexpression of 26 kDa Bcl‐2 in cell cultures inhibits cellular proliferation. Following a series of reports, we posulated that nuclear‐residing Bcl‐2 might control, or be under control of, cell cycle kinetics (2000, 2003a). Massive presence of the 19 kDa Bcl‐2 form in viable cell cultures led us to conclude that 19 kDa might originate from specific cleavage of the full‐length 26‐kDa Bcl‐2 under conditions not relating to control of the apoptotic process, but rather reflecting cell cycle kinetics.

To exclude any connection of 19‐kDa Bcl‐2 expression with apoptotic events, its expression was evaluated in CCbcl2 cells exposed to a range of sublethal‐to‐apoptosis‐inducing concentrations of the anthracyclin adriamycin. Adriamycin arrests cells in the G2/M‐phase. We therefore evaluated 19‐kDa Bcl‐2 expression when cells were transiently accumulated in the G2/M phase by non‐toxic action of the specific compound nocodazole. Expression of 19‐kDa Bcl‐2 was verified in predominant G0/G1‐phase or S‐phase residing CCbcl2 cell populations growing at 100 or 50% confluence, respectively (2003a, 2003b). In addition, the expression of 19‐kDa Bcl‐2 was determined in cytoplasmic and nuclear fractions of these variously growing bcl‐2 transfectants.

Furthermore, to explore whether cleavage of the N‐terminal BH4 domain from the Bcl‐2 protein (resulting in 22‐kDa Bcl‐2) could progressively undergo further cleavage into a 19‐kDa Bcl‐2 fragment, or to verify for other sequence requirements for cleavage of Bcl‐2, 19 kDa Bcl‐2 expression was evaluated in CC531 cells that overexpress the human Bcl‐2 protein missing an evolutionary BH domain: i.e. the BH1 (CCΔBH1 cells), BH3 (CCΔBH3 cells), or BH4 domain (CCΔBH4 cells) or the transmembrane region (CCΔTM cells).

We present data on the subcellular localization of the rarely described 19‐kDa Bcl‐2 fragment in various cell cycle stages of viable cells. These data implicate a link between 19‐kDa Bcl‐2 residing in nuclei and cell cycle control.

MATERIALS AND METHODS

Cell culture

Culturing conditions for parental rat CC531 colorectal cancer cells, CC531 cells stably transfected with the full‐length bcl‐2 gene (CCbcl2 cells), the corresponding neo‐construct (CCneo cells) and truncated bcl‐2 gene constructs, as well as for parental human MCF‐7 breast cancer cells have been described in detail (Hoetelmans et al. 2003a). Stably transfected cells were used for a maximum of 10 passages after seeding.

Cell synchronization

Cells were grown to approximately 50 or 100% confluency by seeding, respectively, 0.25 × 106 or 2.0 × 106 cells in 25‐cm2 culture flasks. After 48 h, cells were harvested for analysis of cell cycle distribution, protein detection or thiol analysis. Nocodazole‐induced synchronization of CC531 and CCbcl2 cells in the G2/M phase was performed as described by Ling et al. (1998). Briefly, we seeded 1.0 × 106 cells in 25‐cm2 culture flasks; after 24 h cells were exposed for a total of 16 h to 200 ng/ml of the G2/M cell cycle arresting compound nocodazole (Sigma, St Louis, MO, USA) dissolved in dimethylsulfoxide (DMSO). Cells were collected by the gentle shake‐off method, washed twice in fresh medium and immediately prepared for flow cytometric analysis or protein analysis. One part of the synchronized cells was relieved from G2/M‐phase arrest by replating in fresh medium and viability was confirmed light microscopically in the next 24 h.

Transfection, plasmids, western blot analysis and antibodies

Stably transfected rat CC531 single‐cell colonies were generated with the liposomal transfection reagent Fugene (Roche Diagnostics, Indianapolis, USA) as described (Hoetelmans et al. 2003a). The full‐length human bcl‐2 gene and neo‐construct have previously been characterized (Hoetelmans et al. 2003a). The bcl‐2 gene constructs lacking the coding sequences for the BH1, BH3, BH4, TM domain or the amino acid region 30–80 were generously provided by Dr Parslow (University of California, San Francisco, California) and were described in detail (Hoetelmans et al. 2003a). Preparation of whole cell lysates, separation of nuclear and cytoplasmic fractions, electrophoretical separation, transfer to nitrocellulose membranes and staining of filters was performed as described by Hoetelmans et al. (2003a). The nuclear lysate was loaded in a 3 : 1 ratio (based on the Bradford assay) compared with the corresponding cytoplasmic lysate to assure equal loading of fractional protein mass.

Rat Bcl‐2 was detected with the pAb #N‐19 diluted 1 : 250 (Santa Cruz Biotechnologies, Santa Cruz, CA, USA). Analysis of human Bcl‐2 with the monoclonal antibody (mAb) #6C8 (Pharmingen, San Diego, USA), the PARP cleavage product (see Apoptosis detection) with mAb #C‐2–10 (Oncogene, Darmstadt, Germany) and α‐actin with mAb #C4 (1 : 15 000) (Roche Diagnostics) was recently described (Hoetelmans et al. 2003a). The polyclonal anti‐thioredoxin antibody was kindly donated by Dr Nakamura (Department of Biological Responses, Institute for Virus Research, Kyoto University, Kyoto, Japan). Secondary rabbit anti‐mouse and swine anti‐rabbit antibodies labelled with horseradish peroxidase were both purchased from Dako (Carpinteria, CA, USA).

Glutathione analysis

Nuclear fractions of 50 and 100% confluent CC531 and CCbcl2 cells were isolated, as recently described, by including MgCl2 for stabilization of nuclear membranes (Hoetelmans et al. 2000). Cellular glutathione (GSH) was extracted from these nuclear pellets by adding 5% (w/v) sulfosalicylic acid and GSH (and other thiol‐containing components) were assayed by the method of Ellman (1959). Total protein was determined by the method of Lowry et al. (1951). The thiol content was expressed as nmol GSH per mg of total protein.

Flow cytometric analysis

Flow cytometric analysis was performed on 50% confluent, growing CC531 and CCbcl2 cells, on CCbcl2 cells directly after nocodazole treatment and on CCbcl2 cells, 24 h after the end of the exposure period to 5 µm of adriamycin (see Apoptosis detection). Cells were washed briefly in phosphate‐buffered saline (PBS) and fixed for 10 min in a freshly prepared 1% paraformaldehyde solution and were subsequently permeabilized in ice‐cold (−20 °C) methanol for 10 min. Cells were washed extensively in PBS and PBS/1% bovine serum albumin (BSA) and were incubated for 30 min with Rnase A (1 mg/ml) in PBS at 37 °C and were subsequently put on ice for 30 min and incubated with propidium iodide (75 µg/ml) in water (both Rnase A and propidium iodide were purchased from Sigma). A minimum of 10 000 events was measured on a FACscan flow cytometer (Becton Dickinson, San Jose, CA, USA). We approached actual sizes of cell cycle fractions by the Modfit 2 program, the Synchronization Wizard. The G2/G1‐ratio was set at 1.91 and S‐phase was defined as one rectangle compartment.

Apoptosis detection

Stock solution of adriamycin (a generous gift from Pharmacia‐Upjohn) was prepared in saline and further diluted in medium. In 25 cm2 culture flasks, 0.25 × 106 CC531 or CCbcl2 cells were grown for 48 h. Cells were briefly trypsinized then diluted to 0.2 × 106 cells per ml of medium, containing various concentrations of adriamycin, respectively, 1, 2, 5, 10, 12, 15, 20, 25, 35, 40 and 50 µm, and were subsequently transferred to sealed tubes. Cells were gently shaken in a waterbath for 1 hour. At the end of the incubation period, cells were pelleted at 500 g and were washed twice with fresh medium.

PARP cleavage: 0.1 × 106 CC531 or CCbcl2 cells, treated with adriamycin as described above, were seeded in 6‐well plates. After 24 h, medium was spun down at 6000 g and the pellet was pooled with the cellular pellets. Both fractions were lysed in 1% NP‐40 lysis buffer supplemented with protease inhibitors as previously described (Hoetelmans et al. 2000).

Cytokeratin and TUNEL staining: from the same batch of adriamycin‐treated cells, 0.1 × 106 CCbcl2 cells were seeded on coverslips in 24‐well plates for anti‐cytokeratin‐18 staining and TUNEL staining (both purchased from Roche Diagnostics) according to manufacturer's specifications.

Proteasome inhibitor assay

CCbcl2 cells were treated daily for a total period of 30 days with sublethal concentrations of the tripeptide aldehyde proteasome inhibitors PSI (0.01 nm) in DMSO, or MG132 (1 nm) in DMSO or the general serine protease inhibitor phenylmethylsulfonyl fluoride (PMSF) (1 µm) in ethanol. PSI and MG132 were purchased from Peptides International (Louisville, USA), PMSF was purchased from Sigma. After 28 days of treatment, 0.25 × 106 cells were seeded in a 25‐cm2 culture flask and cells were harvested after 48 h to prepare whole cell lysates for western blot analysis.

RESULTS

Absence of 19‐kDa human Bcl‐2 in apoptotic and G2/M‐phase accumulated rat CCbcl2 cells

In parental CC531 cells exposed to (apoptosis‐inducing) 5 µm of adriamycin, complete loss of 26 kDa rat Bcl‐2 immunoreactivity (the only detectable form) was observed (Fig. 1a; lane 5). In contrast to rat Bcl‐2, several forms of human Bcl‐2 were evident in non‐treated CCbcl2 cells (Fig. 1b; lane c): 26‐kDa, poor 22‐kDa and substantial 19‐kDa Bcl‐2 immunoreactivity. In CCbcl2 cells exposed to (sublethal) 5 µm adriamycin, 26 kDa Bcl‐2 immunoreactivity appeared unchanged, while a decline in 19‐kDa (and loss of 22‐kDa) Bcl‐2 immunoreactivity was already apparent (Fig. 1b; lane 5). As reflected in parental CC531 cells (Fig. 1a), only at apoptosis‐inducing concentrations of adriamycin (25 and 50 µm), 26‐kDa Bcl‐2 immunoreactivity declined in CCbcl2 cells, while other Bcl‐2 forms were no longer detectable (Fig. 1b; lanes 25 and 50). Apoptosis in CCbcl2 cells exposed to 25 and 50 µm of adriamycin was confirmed by appearance of the 85‐kDa PARP cleavage product (Fig. 1b), and for cells exposed to 25 µm of adriamycin by positive identification of DNA fragmentation and cleaved cytokeratin‐18 (Fig. 1c).

Figure 1.

Figure 1

Open in a new tab

Forms of Bcl‐2 under apoptotic conditions and cell cycle distribution of CCbcl2 cells following exposure to adriamycin. (a, b) Expression of characteristic 26‐kDa rat Bcl‐2 in parental CC531 cells (a), 26‐kDa human Bcl‐2 and 22‐kDa and 19‐kDa Bcl‐2 fragments in CCbcl2 cells (b), 24 h after 1‐h exposure to the anthracyclin adriamycin, expressed in µM. (a) CC531 cells are non‐treated (lane C), treated with 2 µm (lane 2) or 5 µm adriamycin (lane 5). Similarly for CCbcl2 cells in (b), lane C – non‐treated; lanes 5, 12, 25 and 50 represent CCbcl2 cells exposed to 5, 12, 25 and 50 µm of adriamycin, respectively. Equal protein loading of CCbcl2 lysates is confirmed by α‐actin staining (b). Upon increasing adriamycin dosages, 19‐kDa (and 22‐kDa) Bcl‐2 immunoreactivity is lost in CCbcl2 cells (b). Apoptosis in CCbcl2 cells − 24 h after exposure to 25 and 50 µm of adriamycin is identified in (b) by detection of cleavage of PARP (PARP of 113 kDa is cleaved into an approximately 85‐kDa fragment); in (c) (exposure to 25 µm of adriamycin) by DNA fragmentation (TUNEL assay), a characteristically late‐apoptotic event (arrows in c, top panels; shrinkage of cytoplasm and nuclear fragmentation is evident) and cleaved cytokeratine‐18 (an early apoptotic event) (arrows in c, lower panels: arrows indicate immunostaining of cleaved cytokeratin in cytoplasmic flame‐like structures). (d) At a sublethal concentration of adriamycin (5 µm), in CCbcl2 cells a characteristic adriamycin‐induced cell cycle deregulation is observed by flow cytometric DNA analysis (arrows to G0/G1‐phase and G2/M‐phase).

Loss of 19‐kDa Bcl‐2 in CCbcl2 cells was observed exposing these cells to a sublethal (5 µm) concentration of adriamycin, a condition characterized by extensive G2/M‐phase accumulation (85%) based upon flow cytometric analysis (Fig. 1d). In cell lysates containing high fractions of G2/M‐phase residing CCbcl2 cells (following treatment with nocodazole), 19‐kDa Bcl‐2 immunoreactivity was absent (Fig. 2). No apoptosis occurred in these G2/M‐phase synchronized cells after replating, confirming the non‐toxic action of nocodazole (data not shown).

Figure 2.

Figure 2

Open in a new tab

Histogram of expression of Bcl‐2 and DNA: CCbcl2 cells after nocodazole treatment. CCbcl2 cells treated with the mitosis‐arresting compound nocodazole and characteristic G2/M‐phase accumulation verified by flow cytometric DNA analysis, directly at the end of the incubation period: DNA histogram, arrows show G0/G1‐phase (1%) and G2/M‐phase (75%). These mitotically arrested CCbcl2 cells and control CCbcl2 cells underwent western blot analysis. In contrast to control CCbcl2 cells, no 19‐kDa Bcl‐2 immunoreactivity was found in mitotically arrested CCbcl2 cells.

Does 19‐kDa Bcl‐2 result from specific cleavage at the Asp64 residue?

Transfecting (foreign) human Bcl‐2 into a rat environment may induce non‐specific cleavage of Bcl‐2. However, parental and human bcl‐2 transfected MCF‐7 and ZR‐75–1 breast cancer cells also reflect faint 22‐kDa Bcl‐2 and more pronounced 19‐kDa Bcl‐2 immunoreactivity (depicted for parental MCF‐7 cells in Fig. 4a).

Figure 4.

Figure 4

Open in a new tab

(a) Western blot analysis of Bcl‐2 in whole cell lysates of parental MCF‐7 human breast cancer cells. Expression of several forms of human Bcl‐2 immunoreactivity evaluated in parental MCF‐7 breast cancer cells (using mAb #6C8). (b) Western blot analysis of Bcl‐2 in nuclear fractions of CC531 cells transfected with bcl‐2 construct lacking amino acids 30–80. Bcl‐2 expression evaluated in whole cell lysates of CCbcl2 cells and CC531 cells overexpressing human Bcl‐2 protein missing the amino acids 30–80 (CC30/80) (using mAb #6C8). (c) Western blot analysis of whole cell lysates of CCbcl2 cells treated with sublethal concentrations of several proteasome inhibitors. Bcl‐2 immunoreactivity evaluated in CCbcl2 cells treated with sublethal concentrations of the proteasome inhibitors PSI or MG132 or the general protease inhibitor PMSF (using mAb #6C8). (d) Western blot analysis of Bcl‐2 in nuclear and cytoplasmic lysates of 50 and 100% confluent parental CC531 cells and CC531 cells transfected with full‐length or truncated bcl‐2 gene constructs. Expression of full‐length or truncated Bcl‐2 determined in nuclear and cytoplasmic fractions of 50 and 100% confluent CCbcl2 cells, CCΔBH1, CCΔBH3, CCΔBH4 or CCΔTM cells. Equal protein loading is verified by α‐actin staining in nuclear and cytoplasmic fractions (depicted here for nuclear and cytoplasmic fractions of 100% confluent cells). Full‐length Bcl‐2 immunoreactivity in nuclear fractions of 100% confluent cells was partly published in earlier work (Hoetelmans et al. 2003a). (e) Western blot analysis of thioredoxin in nuclear and cytoplasmic lysates of 50 and 100% confluent CC531 and CCbcl2 cells. Expression of thioredoxin (TRX) evaluated in nuclear and cytoplasmic fractions of 50 and 100% confluent CC531 and CCbcl2 cells.

CC531 cells were transfected with a bcl‐2 construct lacking the coding sequence for amino acids 30–80, i.e. devoid of the putative Asp64 cleavage site (yielding an approximately 21‐kDa protein) (Fig. 4b). In whole cell lysates, we could not detect the 19‐kDa Bcl‐2 form using the mAb #6C8, in contrast to full‐length bcl‐2‐transfected cells (Fig. 4b).

We evaluated the presence of 19‐kDa Bcl‐2 in CCbcl2 cells continuously treated with sublethal concentrations of the proteasome inhibitors PSI and MG132 and a general protease inhibitor PMSF. No effect was observed on 26‐kDa and 19‐kDa Bcl‐2 expression after exposure to PSI, whereas PMSF and MG132 exposure resulted in declined 26‐kDa Bcl‐2 and simultaneously increased 19‐kDa Bcl‐2 immunoreactivity (Fig. 4c).

Subcellular localization of 19‐kDa Bcl‐2 in variably growing bcl‐2 transfectants and sequence requirements for Bcl‐2 cleavage

In order to discern nuclear and cytoplasmic fractions of 50 and 100% confluent growing CCbcl2 cells in terms of proliferation status, these fractions were initially screened for the expression profile of a compartimentalized proliferation marker, i.e. thioredoxin (TRX): TRX translocates from the cytoplasm into the nucleus upon mitogenic stimulation.

In cytoplasm of 100% confluent CC531 and CCbcl2 cells, a comparable amount of TRX was found, while TRX was absent in the corresponding nuclear fractions (Fig. 4e). In both 50% confluent cell populations, TRX expression was abundant in cytoplasm and to a similar extent in nuclear fractions (Fig. 4e).

The nuclear and cytoplasmic lysates of 50 and 100% confluent growing CCbcl2 cells were subsequently screened for Bcl‐2 expression profiles. As we recently demonstrated for full‐length human Bcl‐2, truncated Bcl‐2 expression was proportionally distributed among the nuclear and cytoplasmic fractions of the bcl‐2 transfectants, except for CCΔBH4 cells (Fig. 4d) (Hoetelmans et al. 2003a). The 50 and 100% confluent cytoplasmic fractions of CCbcl2, CCΔBH3 and CCΔTM cells displayed additional 19‐kDa Bcl‐2 immunoreactivity. The 19‐kDa Bcl‐2 band became evident in the cytoplasm of 50% confluent CCΔBH1 cells, but CCΔBH4 cells did not display either 19‐kDa or 22‐kDa Bcl‐2 immunoreactivity in the cytoplasmic fractions of 50 and 100% confluent cells (Fig. 4d). Additional 22‐kDa banding was lost in the cytoplasm of 50% confluent CCbcl2 and CCΔBH3 cells, except for the CCΔTM cells in which 22‐kDa banding was increasingly apparent under growing conditions (Fig. 4d).

In nuclear fractions of 100% confluent cells, 19‐kDa Bcl‐2 was expressed in CCbcl2, CCΔBH3 (poor) and CCΔTM cells, and was absent in CCBH1 and CCΔBH4 nuclear fractions (Fig. 4d): a similar pattern was observed for their corresponding cytoplasmic fractions. Intensity of 19‐kDa Bcl‐2 banding increased or became evident in all nuclear fractions of 50% confluent cells, while concomitant immunoreactivity of full‐length or truncated bcl‐2 gene products had decreased compared with 19‐kDa immunoreactivity. No 22‐kDa immunoreactivity was detected in any of the nuclear fractions, supporting the concept of hampered nuclear transport of the Bcl‐2 protein missing its N‐terminal BH4 domain.

DISCUSSION

Absence of 19‐kDa human Bcl‐2 in apoptotic and G2/M‐phase accumulated rat CCbcl2 cells

Parental CC531 and CCbcl2 cells were exposed to the apoptosis‐inducer adriamycin as described recently (Hoetelmans et al. 2003b). This earlier work had demonstrated that, under these specific conditions, CCbcl2 cells were significantly more resistant to adriamycin‐induced apoptosis than were parental CC531 cells (IC50 values were 9.1 ± 1.6 versus 33.8 ± 2.6 µm for, respectively, CC531 and CCbcl2 cells).

Adriamycin treatment is ultimately characterized by G2/M‐phase accumulation, which is already apparent at sublethal concentrations (for CCbcl2 cells at 5 µm) (Fig. 1d) (Hoetelmans et al. 2003b). To explore this for mitotically arrested cells, CCbcl2 cells were exposed to the specific G2/M cell cycle arresting compound nocodazole (Ling et al. 1998). Loss of 22‐kDa and 19‐kDa Bcl‐2 immunoreactivity (Fig. 1b) in apoptotic cells seems contradictive, at least for the well‐documented cleavage of Bcl‐2 into a 22‐kDa fragment, and points at putative cell cycle stage specific appearance of 19‐kDa Bcl‐2. Upon entering the G2/M‐phase, Bcl‐2 is phosphorylated and it has been postulated that the anti‐apoptotic function of Bcl‐2 is thereby abolished, allowing progress of the apoptotic process (Haldar et al. 1995) or even enhanced in growth factor‐stimulated apoptosis (May et al. 1994).

Post‐translational protein phosphorylation has evolved as a major field of research. Ling et al. (1998), among others, conjectured that phosphorylation of Bcl‐2 merely reflects the cell cycle state. Interestingly, Brichese et al. (2002) recently suggested phosphorylation of Bcl‐2 to prevent it from being degraded. Phosphorylation of Bcl‐2 takes place at several serine/threonine residues in the so‐called flexible loop region spanning between amino acids 30–90 (Fig. 3a) and is identified as a 30‐kDa band on western blot. Until now, we were unable to detect this hyperphosphorylated Bcl‐2 protein, either in adriamycin‐ or nocodazole‐induced G2/M‐phase accumulated CCbcl2 cells when using a panel of commercially available mAbs (#124, #100, #4D7 or #6C8), or in human HT‐29 colon cancer cells that solely contain phosphorylated Bcl‐2 (Guan et al. 1996) (data not shown). Both 22‐kDa and 19‐kDa Bcl‐2 retain some serine/threonine residues that may be phosphorylated upon entering the G2/M‐phase. Steric hindrance by phosphorylation might interfere with the specific recognition of epitopes by antibodies, hampering identification of the various Bcl‐2 products. This might be an explanation for the decline in 26‐kDa Bcl‐2 immunoreactivity in adriamycin‐treated apoptotic parental CC531 cells and CCbcl2 cells. However, massive G2/M‐phase accumulation of CCbcl2 cells after exposure to sublethal 5 µm adriamycin did not influence 26 kDa immunoreactivity compared with non‐treated CCbcl2 cells (Fig. 1b). Also, phosphorylation of Bcl‐2 fragments appears to be unlikely: Blagosklonny et al. (2000) demonstrated 22‐kDa Bcl‐2 fragments in G2/M‐phase residing cells using an antibody raised against the 41–54 region of the Bcl‐2 protein, within the region of extensive phosphorylation (Fig. 3a). Moreover, 19‐kDa Bcl‐2 expression has been detected so far only by applying the mAb#6C8 (Yamamoto et al. 1998; Hoetelmans et al. 2003a). Although the exact epitope(s) of the mAb6#8 antibody is (are) not yet specified, identification of the Bcl‐2 protein missing amino acids 1–64 (almost the complete flexible loop region including some of the phosphorylation sites) suggests epitopes for this specific antibody to reside outside this region (Fig. 3a). Therefore, it is not likely that phosphorylation of the Bcl‐2 cleavage products – if any – would hamper identification with the mAb#6C8 antibody.

Figure 3.

Figure 3

Open in a new tab

(a) Structure of the human Bcl‐2 protein including phosphorylation sites, epitopes of anti‐Bcl‐2 mAbs and (putative) cleavage sites. The N‐terminal BH4 domain of the human Bcl‐2 protein resides between the amino acids (aa) 10–30, the BH3 domain resides between aa 93–107, the BH1 domain between aa 133–155, the BH2 domain between aa 187–202 and the carboxyl‐terminal transmembrane region TM spans between aa 219 and 236. The so‐called flexible loop region lies between the BH4 and BH3 domain. Cleavage by caspase‐3 is targeted at the aspartate residue 34 (Asp34) yielding a 22‐kDa Bcl‐2 fragment. Cleavage (by the unknown factor X) of the Asp64 residue yields a 19‐kDa Bcl‐2 fragment. The aliphatic Valine 61 residue determines that caspases‐6, ‐8, ‐9 and ‐10 are putative candidates to cleave Bcl‐2 at the Asp64 residue. In the loop region lie several phosphorylation sites: serine 70 and 87, threonine 56 and 74. Phosphorylation at these residues may interfere with epitope recognition by specific monoclonal antibodies due to steric hindrance by the phosphate molecule itself or by perturbation of electrostatic interactions. The epitope for the mAbs #124 and #100 resides between aa 41–54, for the mAb #4D7 between aa 61–76, and epitope(s) for mAb #6C8 is (are) unknown. A part of (a) was reproduced after Hoetelmans et al. (2003a). (b) Putative cleavage of Bcl‐2 in the nucleus and the cytoplasm based on subcompartimental expression profiles of 22‐kDa and 19‐kDa Bcl‐2. Bcl‐2 in the cytoplasm (CP) may be cleaved in two manners: release of cytochrome c from the mitochondrial innermembranous space (M) activates caspase‐3 (via activation of caspase‐9 involved in the apoptosome) and caspase‐3 subsequently may cleave the N‐terminal BH4 domain from the Bcl‐2 protein at the Asp34 residue (Bcl‐2Δ1‐34). As no 19‐kDa Bcl‐2 can be traced in cytoplasmic fractions of CCΔBH4 cells, this indicates that cleavage of the BH4 domain from cytoplasmic Bcl‐2 does not allow further cleavage into 19‐kDa Bcl‐2. Activation of factor X (possibly a subclass III caspase, e.g. caspase‐9) might cleave Bcl‐2 at the Asp64 residue (Bcl‐2Δ1‐64). In the nucleus (N), presence of the Bcl‐2Δ1‐64 fragment may result from activation of factor X as in the CP. As both fragments appear in nuclear fractions of proliferating CCΔBH4 cells, the BH4 domain is not required for cleavage at the Asp64 residue by factor X and thus may be the result of cleavage of the Bcl‐2Δ1‐34 fragment by another factor Y.

Does 19‐kDa Bcl‐2 result from specific cleavage at the Asp64 residue?

Studies in the nematode Caenorhabditis elegans demonstrated cleavage of the functional Bcl‐2 homologue Ced‐9 by a homologue of caspase‐3 (Xue & Horvitz 1997). Cleavage of Bcl‐2‐like molecules seems conserved throughout evolution. Caspases cleave a wide range of biological molecules and recognize tetrapeptide sequences with an absolute requirement for aspartate (Asp) residues, referring to the well‐documented caspase‐3‐mediated cleavage of the N‐terminal domain from the Bcl‐2 protein at the Asp34 residue. Based on amino acid weights, cleavage at the Asp64 residue would yield a 19‐kDa Bcl‐2 fragment (Fig. 3a) as earlier suggested by Yamamoto et al. (1998) and demonstrated here in Fig. 4(b).

The Valine 61 residue of the human Bcl‐2 protein determines the specificity of caspase cleavage and makes caspases of the so‐called subclass III (including caspase‐9) potential inducers of Bcl‐2 cleavage at the Asp64 residue. Pro‐caspase‐9 is involved in the apoptosome complex and is cleaved into its active form upon cytochrome c release, thereby activating, among others, caspase‐3. Zhivotovsky et al. (1999) demonstrated that both the inactive pro‐caspase‐9 and its activated form reside in trace amounts in the mitochondrial fraction of non‐apoptotic and apoptotic Jurkat cells. Constantini et al. (2002) reported pro‐caspase‐9 to be released upon apoptotic stimuli from mitochondrial fractions in HeLa cells and to translocate into the nucleus. Also, caspase‐9 was visualized in nuclei of apoptotic hippocampal neurones (Krajewski et al. 1999). The mitochondrial membranes and nucleoplasm are subcellular sites that are preferred by Bcl‐2 and the close proximity of Bcl‐2 to (pro‐)caspase‐9 pools points at an interplay. Zhivotovski et al. (1999) suggested that pro‐caspase‐9 might undergo mitochondrial activation without interference of cytochrome c and the apoptosome pathway, confirming findings of Stennicke et al. (1999). They demonstrated caspase‐9 activation, which was not preceded by any cleavage activity. The concept of activation not requiring any proteolytic processing, but rather depending on cytosolic factors was hereby introduced. This would fit the concept of caspase activity outside the field of apoptosis, such as differentiation, proliferation or signal transduction (Zeuner et al. 1999). The abundant presence of 19‐kDa Bcl‐2 in viable cells may be due to the fact that mitochondrial caspase‐9 may directly induce cleavage of Bcl‐2 (located in or near the mitochondrial membrane fraction or in the nucleoplasm) into 19 kDa, thereby lacking any apoptotic stimuli or features.

Next to proteolysis by caspases, much interest is taken in the ubiquitin‐proteasome protein degradation pathway which is crucial in controlling intracellular levels of a variety of short‐lived proteins, maintaining a myriad of cellular events. A multitude of regulatory proteins (cyclins, CDK inhibitors, tumour suppressors) are degraded by this route. Suppression of proteasome functioning may induce apoptosis via the release of cytochrome c from mitochondria and activation of caspase‐3‐(like)‐proteases resulting in accelerated cleavage of Bcl‐2 into a 22‐kDa fragment (Zhang et al. 1999).

Specific proteasome inhibition – and presumably subsequent caspase activation – can affect cleavage of Bcl‐2 into a 19‐kDa fragment as we demonstrate here for PMSF and MG132 (Fig. 4c). Dimmeler et al. (1999) described that dephosphorylation of Bcl‐2 at specific MAP kinase sites residing within the flexible loop region targets the Bcl‐2 protein for proteasome‐mediated degradation. Entering the G2/M‐phase and subsequent phosphorylation of full length Bcl‐2 may provide an effective way to protect cells from breakdown of Bcl‐2, as recently suggested by Brichese et al. (2002).

Subcellular localization of 19‐kDa Bcl‐2 in variably growing bcl‐2 transfectants and sequence requirements for Bcl‐2 cleavage

Here, and in a series of reports, Bcl‐2 was demonstrated to reside in interphase nuclei (Fig. 4d) (2000, 2003a). The N‐terminal BH4 domain of the Bcl‐2 protein appeared to be a prerequisite for the nuclear translocation of Bcl‐2 (Hoetelmans et al. 2003a) and for the reduced cellular proliferation of cells overexpressing Bcl‐2 (Huang et al. 1997; 2003a, 2003b). This led us to conclude that nuclear‐residing Bcl‐2 affects cellular proliferation. To extend this hypothesis, the expression of 19‐kDa Bcl‐2 was evaluated in nuclear and cytoplasmic fractions of exponentially growing (50% confluent) and mitotically retarded (100% confluent) CCbcl2 cell populations. Cells in 50% confluence have a statistically significantly larger portion of S‐phase cells and a smaller fraction of G0/G1‐phase cells compared with 100% confluent cell populations (Hoetelmans et al. 2003b). To support this concept for nuclear and cytoplasmic fractions of 100 and 50% confluent cultures, 50 and 100% confluent parental CC531 and CCbcl2 cells were characterized for the expression of the oxidoreductor thioredoxin (TRX) in nucleus and cytoplasm. TRX is a major contributor to redox regulation and translocates from the cytoplasm into the nucleus upon mitogenic stimulation (Nakamura et al. 1997). This expression profile was confirmed in 50 and 100% confluent CCbcl2 fractions (Fig. 4e).

To explore sequence requirements for appearance of the 19‐kDa Bcl‐2 fragment, we additionally included nuclear and cytoplasmic fractions of 50 and 100% confluent growing bcl‐2 transfectants overexpressing the Bcl‐2 protein lacking a Bcl‐2 homology domain: the BH1 (CCΔBH1 cells), BH3 (CCΔBH3 cells), or the BH4 domain (CCΔBH4 cells) or the transmembrane (TM) domain (CCΔTM cells) (2003a, 2003b).

The 19‐kDa Bcl‐2 form is present in the cytoplasm of bcl‐2‐transfectants regardless of growing status, but its appearance requires the presence of a functional N‐terminal BH4 domain (Fig. 3b). It can be concluded that in the cytoplasm, cleavage of Bcl‐2 into a 22‐kDa fragment (Fig. 3b; Bcl‐2Δ1‐34) by caspase‐3 does not result in further cleavage into a 19‐kDa fragment (Fig. 3b; Bcl‐2Δ1‐64) by a yet unkown factor X (a caspase (‐like) factor). Cleavage of 19‐kDa Bcl‐2 in nuclear fractions requires a Bcl‐2 domain residing outside the BH1, BH3, BH4 or the TM region. This cleavage might result from action of the same cellular factor responsible for cleavage of this type of cytoplasmic Bcl‐2 (factor X) or could be another factor (Y) as well. Moreover, proliferation‐dependent nuclear, and not cytoplasmic, 19‐kDa Bcl‐2 expression suggests distinct functionalities of compartimentalized Bcl‐2. In Fig. 3(b) we summarized the above findings concerning Bcl‐2 domain dependency and compartimentalization of 19‐kDa Bcl‐2.

In line with findings of others, we have demonstrated an increase in cellular glutathione (GSH) levels when the Bcl‐2 protein has been overexpressed (2003a, 2003b). Focusing on the nuclear compartment, we recently demonstrated elevated GSH levels in isolated nuclear fractions of exponentially growing parental CC531 and neo‐transfected cells compared with mitotically retarded cells (Hoetelmans et al. 2003a). Nuclear GSH levels in CCbcl2 cells were higher than in control cells in agreement with the findings of Voehringer et al. (1998) who described an increase in nuclear GSH levels in bcl‐2 transfected HeLa cells, but nuclear GSH levels were consolidated at 50 and 100% confluence. Presence of Bcl‐2 in quiescent nuclear fractions and its cleavage (in actively dividing nuclei) into a 19‐kDa fragment might orchestrate or be orchestrated by these nuclear GSH levels. Alterations in the nuclear redox environment may thus have a profound effect on the progression of the cell cycle or, alternatively, may modulate apoptotic pathways by influencing the activity of nuclear apoptotic enzymes such as AP‐24 (Wright et al. 1998) or enzymes capable of cleaving Bcl‐2. Validation of retainment of the in vivo distribution of GSH in nuclear fractions and availability of more specific molecular GSH probes should be a first step towards elucidation of these pathways.

In conclusion, a rarely described 19‐kDa (human) Bcl‐2 form is present in nuclear and cytoplasmic fractions of viable cells that overexpress the (human) Bcl‐2 protein. Immunoreactivity of 19‐kDa Bcl‐2 is intensified in nuclear fractions of more rapidly cycling cell populations and does not depend on the presence of a functional BH1, BH3, BH4, or TM domain. Cytoplasmic 19‐kDa Bcl‐2 requires presence of the N‐terminal BH4 domain and is not influenced by proliferation control. As demonstrated earlier, localization of the full‐length Bcl‐2 in the nuclear compartment might represent inhibition of cellular proliferation with a potential role for nuclear GSH levels. Here, we demonstrate that cleavage of nuclear full‐length 26‐kDa Bcl‐2 into a smaller 19‐kDa fragment coincides with cellular proliferation; we thus suggest that this cleavage overcomes 26‐kDa Bcl‐2 repression of cellular proliferation.

ACKNOWLEDGEMENTS

The authors are indebted to Dr Parslow for generously providing the truncated bcl‐2 gene constructs and Dr Nakamura for kindly donating the anti‐TRX antibodies. I. M. Leeflang is acknowledged for manuscript processing.

REFERENCES

  1. Blagosklonny MV, Bishop PC, Robey R, Fojo T, Bates SE (2000) Loss of cell cycle control allows selective microtubule‐active drug‐induced Bcl‐2 phosphorylation and cytotoxicity in autonomous cancer cells. Cancer Res. 60, 3425. [PubMed] [Google Scholar]
  2. Borner C (1996) Diminished cell proliferation associated with the death‐protective activity of Bcl‐2. J. Biol. Chem. 271, 12695. [DOI] [PubMed] [Google Scholar]
  3. Brenner C, Cadiou H, Vieira HL, Zamzami N, Marzo I, Xie Z, Leber B, Andrews D, Duclohier H, Reed JC, Kroemer G (2000) Bcl‐2 and Bax regulate the channel activity of the mitochondrial adenine nucleotide translocator. Oncogene 19, 329. [DOI] [PubMed] [Google Scholar]
  4. Brichese L, Barboule N, Heliez C, Valette A (2002) Bcl‐2 phosphorylation and proteasome‐dependent degradation induced by paclitaxel treatment: consequences on sensitivity of isolated mitochondria to Bid. Exp. Cell Res. 278, 101. [DOI] [PubMed] [Google Scholar]
  5. Cheng EH, Kirsch DG, Clem RJ, Ravi R, Kastan MB, Bedi A, Ueno K, Hardwick JM (1997) Conversion of Bcl‐2 to a Bax‐like death effector by caspases. Science 278, 1966. [DOI] [PubMed] [Google Scholar]
  6. Constantini P, Bruey JM, Castedo M, Metivier D, Loeffler M, Susin SA, Ravagnan L, Zamzami N, Garrido C, Kroemer G (2002) Pre‐processed caspase‐9 contained in mitochondria participates in apoptosis. Cell Death Differ. 9, 82. [DOI] [PubMed] [Google Scholar]
  7. Dimmeler S, Breitschopf K, Haendeler J, Zeiher AM (1999) Dephosphorylation targets Bcl‐2 for ubiquitin‐dependent degradation: a link between the apoptosome and the proteasome pathway. J. Exp. Med. 189, 1815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Ellman GL (1959) Tissue sulfhydryl groups. Arch. Biochem. Biophys. 82, 70. [DOI] [PubMed] [Google Scholar]
  9. Fadeel B, Hassan Z, Hellstrom‐Lindberg E, Henter JI, Orrenius S, Zhivotovsky B (1999) Cleavage of Bcl‐2 is an early event in chemotherapy‐induced apoptosis of human myeloid leukemia cells. Leukemia 13, 719. [DOI] [PubMed] [Google Scholar]
  10. Fortney JE, Hall BM, Bartrug L, Gibson LF (2002) Chemotherapy induces bcl‐2 cleavage in lymphoid leukemic cell lines. Leuk. Lymphoma 43, 2171. [DOI] [PubMed] [Google Scholar]
  11. Fujita N, Nagahashi A, Nagashima K, Rokudai S, Tsuruo T (1998) Acceleration of apoptotic cell death after the cleavage of Bcl‐XL protein by caspase‐3‐like proteases. Oncogene 17, 1295. [DOI] [PubMed] [Google Scholar]
  12. Grandgirard D, Studer E, Monney L, Belser T, Fellay I, Borner C, Michel MR (1998) Alpha viruses induce apoptosis in Bcl‐2‐overexpressing cells: evidence for a caspase‐mediated, proteolytic inactivation of Bcl‐2. EMBO J. 17, 1268. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Guan RJ, Moss SF, Arber N, Krajewski S, Reed JC, Holt PR (1996) 30‐kDa phosphorylated form of Bcl‐2 protein in human colon. Oncogene 12, 2605. [PubMed] [Google Scholar]
  14. Haldar S, Jena N, Croce CM (1995) Inactivation of Bcl‐2 by phosphorylation. Proc. Natl Acad. Sci. USA 92, 4507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hirotani M, Zhang Y, Fujita N, Naito M, Tsuruo T (1999) NH2‐terminal BH4 domain of Bcl‐2 is functional for heterodimerization with Bax and inhibition of apoptosis. J. Biol. Chem. 274, 20415. [DOI] [PubMed] [Google Scholar]
  16. Hoetelmans RWM, Van Slooten HJ, Keijzer R, Erkeland S, Van de Velde CJH, Van Dierendonck JH (2000) Bcl‐2 and Bax proteins are present in interphase nuclei of mammalian cells. Cell Death Differ. 7, 384. [DOI] [PubMed] [Google Scholar]
  17. Hoetelmans RWM, Vahrmeijer AL, Van Vlierberghe RLP, Keijzer R., Van de Velde CJH, Mulder GJ, Van Dierendonck JH (2003a) The role of various Bcl‐2 domains in the anti‐proliferative effect and modulation of cellular glutathione levels: a prominent role for the BH4 domain. Cell Prolif. 36, 35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hoetelmans RWM, Vahrmeijer AL, Mulder GJ, Van de Velde CJH, Nagelkerke JF, Van Dierendonck JH (2003b) Bcl‐2 overexpression does not prevent but retards adriamycin toxicity in CC531 colon carcinoma cells. Chemotherapy in press . [DOI] [PubMed]
  19. Huang DC, O'Reilly LA, Strasser A, Cory S (1997) The anti‐apoptosis function of Bcl‐2 can be genetically separated from its inhibitory effect on cell cycle entry. EMBO J. 16, 4628. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kluck RM, Bossy‐Wetzel E, Green DR, Newmeyer DD (1997) The release of cytochrome c from mitochondria: a primary site for Bcl‐2 regulation of apoptosis. Science 275, 1132. [DOI] [PubMed] [Google Scholar]
  21. Korsmeyer SJ (1999) Bcl‐2 gene family and the regulation of programmed cell death. Cancer Res. 59, 1693. [PubMed] [Google Scholar]
  22. Krajewski S, Krajewska M, Ellerby LM, Welsh K, Xie Z, Deveraux QL, Salvesen GS, Bredesen DE, Rosenthal RE, Fiskum G, Reed JC (1999) Release of caspase‐9 from mitochondria during neuronal apoptosis and cerebral ischemia. Proc. Natl Acad. Sci. USA 96, 5752. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Ling YH, Tornos C, Perez‐Soler R (1998) Phosphorylation of Bcl‐2 is a marker of M phase events and not a determinant of apoptosis. J. Biol. Chem. 273, 18984. [DOI] [PubMed] [Google Scholar]
  24. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein measurement with the folin‐phenol reagent. J. Biol. Chem. 193, 265. [PubMed] [Google Scholar]
  25. May WS, Tyler PG, Ito T, Armstrong DK, Qatsha KA, Davidson NE (1994) Interleukin‐3 and bryostatin‐1 mediate hyperphosphorylation of BCL2α in association with suppression of apoptosis. J. Biol. Chem. 269, 26865. [PubMed] [Google Scholar]
  26. Mazel S, Burtrum D, Petrie HT (1996) Regulation of cell division cycle progression by bcl‐2 expression: a potential mechanism for inhibition of programmed cell death. J. Exp. Med. 183, 2219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. McDonnell TJ, Nunez G, Platt FM, Hockenberry D, London L, McKearn JP, Korsmeyer SJ (1990) Deregulated Bcl‐2‐immunoglobulin transgene expands a resting but responsive immunoglobulin M and D‐expressing B‐cell population. Mol. Cell Biol. 10, 1901. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Nakamura H, Nakamura K, Yodoi J (1997) Redox regulation of cellular activation. Annu. Rev. Immunol. 15, 351. [DOI] [PubMed] [Google Scholar]
  29. Stennicke HR, Deveraux QL, Humke EW, Reed JC, Dixit VM, Salvesen GS (1999) Caspase‐9 can be activated without proteolytic processing. J. Biol. Chem. 274, 8359. [DOI] [PubMed] [Google Scholar]
  30. Tsujimoto Y, Finger LR, Yunis J, Nowell PC, Croce CM (1984) Cloning of the chromosome breakpoint of neoplastic B cells with the t (14; 18) chromosome translocation. Science 226, 1097. [DOI] [PubMed] [Google Scholar]
  31. Vaux DL, Cory S, Adams JM (1988) Bcl‐2 gene promotes haemopoietic cell survival and cooperates with c‐myc to immortalize pre‐B cells. Nature 335, 440. [DOI] [PubMed] [Google Scholar]
  32. Voehringer DW, McConkey DJ, McDonnell TJ, Brisbay S, Meyn RE (1998) Bcl‐2 expression causes redistribution of glutathione to the nucleus. Proc. Natl Acad. Sci. USA 95, 2956. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Wood DE, Newcombe EW (2000) Cleavage of Bax enhances its cell death function. Exp. Cell Res. 256, 375. [DOI] [PubMed] [Google Scholar]
  34. Wright SC, Wang H, Wei QS, Kinder DH, Larrick JW (1998) Bcl‐2‐mediated resistance to apoptosis is associated with glutathione‐induced inhibition of AP24 activation of nuclear DNA fragmentation. Cancer Res. 58, 5570. [PubMed] [Google Scholar]
  35. Xue D, Horvitz HR (1997) Caenorhabditis elegans CED‐9 protein is a bifunctional cell‐death inhibitor. Nature 390, 305. [DOI] [PubMed] [Google Scholar]
  36. Yamamoto AM, Eloy L, Bach JF, Garchon HJ (1998) N‐terminus cleavage of bcl‐2 by a novel cellular non‐ICE cysteine proteinase. Leukemia 12, 1467. [DOI] [PubMed] [Google Scholar]
  37. Zeuner A, Eramo A, Peschle C, De Maria R (1999) Caspase activation without death. Cell Death Differ. 6, 1075. [DOI] [PubMed] [Google Scholar]
  38. Zhang XM, Lin H, Chen C, Chen BD (1999) Inhibition of ubiquitin‐proteasome pathway activates a caspase‐3‐like protease and induces Bcl‐2 cleavage in human M‐07e leukaemic cells. Biochem. J. 340, 127. [PMC free article] [PubMed] [Google Scholar]
  39. Zhivotovsky B, Samali A, Gahm A, Orrenius S (1999) Caspases: their intracellular localization and translocation during apoptosis. Cell Death Differ. 6, 644. [DOI] [PubMed] [Google Scholar]
  40. Zou H, Li Y, Liu X, Wang X (1999) An APAF‐1 cytochrome c multimeric complex is a functional apoptosome that activates procaspase‐9. J. Biol. Chem. 274, 11549. [DOI] [PubMed] [Google Scholar]

Articles from Cell Proliferation are provided here courtesy of Wiley

RESOURCES