Enforced expression of the antiapoptotic Bcl-2 family protein Mcl-1 promotes lymphomagenesis in the mouse; however, the functional role of Mcl-1 in human B-cell lymphoma remains unclear. We demonstrate that Mcl-1 is widely expressed in malignant B-cells, and high-level expression of Mcl-1 is required for B-lymphoma cell survival, since transfection of Mcl-1-specific antisense oligodeoxynucleotides was sufficient to promote apoptosis in Akata6 lymphoma cells. Mcl-1 was efficiently cleaved by caspases at evolutionarily conserved aspartic acid residues in vitro, and during cisplatin-induced apoptosis in B-lymphoma cell lines and spontaneous apoptosis of primary malignant B-cells. Overexpression of the Mcl-1 cleavage product that accumulated during apoptosis was sufficient to kill cells. Therefore, Mcl-1 is an essential survival molecule for B-lymphoma cells and is cleaved by caspases to a death-promoting molecule during apoptosis. In contrast to Mcl-1, Bcl-2 and Bcl-XL were relatively resistant to caspase cleavage in vitro and in intact cells. Interfering with Mcl-1 function appears to be an effective means of inducing apoptosis in Mcl-1-positive B-cell lymphoma, and the unique sensitivity of Mcl-1 to caspase-mediated cleavage suggests an attractive strategy for converting it to a proapoptotic molecule.
Mcl-1 is an antiapoptotic member of the Bcl-2 protein family, a critical regulator of apoptosis in normal and malignant cells (Packham 1998; Gross et al., 1999; Adams and Cory, 2001; Borner, 2003). Mcl-1 was discovered in myeloid cells where it is strongly induced following differentiation signals and plays an important role in cell survival (Kozopas et al., 1993; Moulding et al., 2000; Craig, 2002). Mcl-1 expression is highly regulated and is rapidly induced in response to a range of signals (Craig, 2002). Mcl-1 has a short half-life, and expression can also be rapidly downregulated (Kozopas et al., 1993; Akgul et al., 2000; Craig, 2002; Cuconati et al., 2003; Iglesias-Serret et al., 2003). In ultraviolet-irradiated HeLa cells, Mcl-1 downregulation is thought to be essential for the translocation of the proapoptotic Bax protein to the mitochondria and release of cytochrome c, implicating Mcl-1 as an apical regulator of the apoptosis cascade in this system (Nijhawan et al., 2003).
Several lines of evidence suggest that Mcl-1 may also be an important regulator of cell survival in human B-cell malignancies. Mcl-1 is highly expressed in some malignant B-cells (Ghia et al., 1998; Kitada et al., 1998; Soini et al., 1998; Agarwal and Naresh, 2002; Pagnano et al., 2002; Spets et al., 2002; Khoury et al., 2003), and this can correlate with clinical behaviour, such as failure to obtain complete remission in chronic lymphocytic leukemia (CLL) (Kitada et al., 1998). Importantly, enforced expression of Mcl-1 in transgenic mice promotes lymphomagenesis with a higher probability than Bcl-2 (approx. 80%) (Zhou et al., 2001). Consistent with the idea that Mcl-1 might play a significant role in regulating survival, Mcl-1 expression has also been reported to be downregulated during apoptosis in B-cells (Lomo et al., 1996; Altmeyer et al., 1997; Bellosillo et al., 1999; Byrd et al., 2002; Pepper et al., 2002). Despite these encouraging findings, the functional role of Mcl-1 in malignant human lymphocytes and mechanisms controlling its downregulation remain to be determined. Here, we investigated the role of Mcl-1 in apoptosis control in B-lymphoma cells.
Expression of Mcl-1 in malignant B cells
We first characterized the expression of Mcl-1 by immunoblotting in a panel of cell lines derived from B-cell malignancies, including Burkitt's lymphoma (BL) (Raji, Ramos, Akata6), diffuse large B-cell lymphoma (RL, DoHH2) and multiple myeloma (RPMI 8226, U-226). Mcl-1 was readily detected in all cell lines studied (Figure 1a). In ML-1 myeloid leukemia cells, Mcl-1 expression is rapidly induced by the differentiation agent 12-O-tetradecanoylphorbol-13-acetate (TPA) (Kozopas et al., 1993; Yang et al., 1995), and we compared the expression of Mcl-1 in B-lymphoma cell lines with the levels induced in ML-1 cells. Constitutive Mcl-1 expression in Akata6 and Ramos cells was equivalent to the expression in ML-1 cells stimulated with TPA (Figure 1b). Therefore, Mcl-1 is constitutively expressed in B-lymphoma cells at high levels, similar to those detected in differentiating myeloid cells. Finally, we analysed the expression of Mcl-1 in various primary B-cell malignancies, including follicular (Non-Hodgkin's lymphoma (NHL)1, 2), diffuse large B-cell (NHL3, 4) and mantle cell lymphoma (NHL5), and CLL1, 2. Mcl-1 was detected in each of the samples analysed, although relative expression levels varied between samples (Figure 1c, see also 3). We also analysed expression in normal B-cells derived from reactive lymph nodes (RLN1, 2). The expression of Mcl-1 was approximately equivalent in reactive and malignant lymph nodes (Figure 1c).
Effects of Mcl-1 antisense oligodeoxynucleotides (ASO) on B-lymphoma cell survival
The abundant expression of Mcl-1 in malignant B-cells suggested that Mcl-1 is an important survival protein in these cells. We therefore used ASO to ablate Mcl-1 expression to test directly the role of Mcl-1 in maintaining cell survival. The Mcl-1 ASO used in these experiments (ISIS 20408) was the most effective inhibitor of Mcl-1 expression identified from screening a series of different Mcl-1 ASO and has been shown to interfere effectively with Mcl-1 expression in other cell systems (Bannerman et al., 2001; Derenne et al., 2002). 2′-O-methoxyethyl/2′-deoxynucleotide chimeric ASO were used to support an RNAase H-dependent mechanism of action. Akata6 cells were transiently transfected with Mcl-1 ASO and Mcl-1 expression analysed by immunoblotting. Expression was reduced (by approx. 50%) within 4 h following transfection with ASO, compared to mock-transfected cells or cells transfected with a control oligodeoxynucleotide, and maintained at reduced levels for up to 24 h (Figure 2a and c). Although the downregulation of Mcl-1 expression was not complete, it is important to note that typically only 50–70% of Akata6 cells were transfected using these electroporation conditions. Therefore, considering that not all cells were transfected in these experiments, the Mcl-1 ASO was relatively effective at interfering with Mcl-1 expression. The effects of the ASO were specific for Mcl-1, since the expression of the cell cycle regulatory molecule PCNA and related antiapoptotic molecules, Bcl-2 and Bcl-XL, were unaltered by ASO (Figure 2b).
We determined whether decreasing Mcl-1 expression via ASO was sufficient to induce apoptosis by analysing PARP cleavage, a substrate for caspases activated during apoptosis. PARP was cleaved in cells transfected with Mcl-1 ASO, but not control oligodeoxynucleotides (Figure 2c). Although not completely cleaved in Mcl-1 ASO-transfected cells, the extent of PARP cleavage was extensive considering that only a subset of cells were transfectable. Consistent with the induction of PARP cleavage, the proportion of cells that had externalized phosphatidylserine (PS) to the outer face of the plasma membrane (a second specific marker of apoptosis) was also significantly increased at 24 h post-transfection in cells transfected with Mcl-1 ASO, but not missense oligodeoxynucleotide, relative to mock-transfected cells (16.0±4.7% PS-positive Mcl-1 ASO-transfected cells versus 5.9±0.7% PS-positive control cells, mean of three separate transfections±s.d.; P=0.02 independent Student's t-test). zVADfmk did not prevent the decrease in Mcl-1 expression in ASO-transfected cells, demonstrating that modulation of Mcl-1 expression is a direct consequence of antisense action and not due to the induction of apoptosis (data not shown). Therefore, Mcl-1 is required for the survival of Akata6 cells. Mcl-1 ASO-triggered apoptosis was associated with early activation of caspase-9 (data not shown), suggesting that like Bcl-2, Mcl-1 functions in cell survival to prevent cytochrome c release.
Regulation of the Mcl-1 protein during cisplatin-induced apoptosis
Mcl-1 is a rapidly turned over protein, and decreased expression of Mcl-1 has recently been shown to be a pivotal, initiating event in DNA damage-induced apoptosis (Nijhawan et al.,. 2003). Since, Mcl-1 was essential for the survival of Akata6 cells, we examined whether it was regulated during chemotherapy-induced apoptosis. Akata6 cells were treated with cisplatin, and the expression of Mcl-1 and other antiapoptotic Bcl-2 family proteins analysed by immunoblotting (Figure 3a). Mcl-1 expression was significantly reduced within 12–16 h following the addition of cisplatin (Figure 3a and data not shown). Notably, this was associated with the appearance of a novel Mcl-1 protein isoform of approximately 28 kDa, first detected at 8 h and more abundant at 16 h (Figure 3a and c). In contrast to Mcl-1, expression of Bcl-2 and Bcl-XL were not significantly altered in cisplatin-treated cells (Figure 3b). Similar results were obtained in BL40 BL cells (data not shown).
The appearance of a novel Mcl-1 protein suggested that Mcl-1 was a target for caspases during apoptosis in Akata6 cells. We therefore compared the kinetics of the activation of caspases and the accumulation of the 28 kDa Mcl-1 isoform during apoptosis (Figure 3a). The p35 form of caspase-9, generated by autoproteolysis, was detected within 4 h following the addition of cisplatin. This was followed by the activation of the effector caspase-3, at 8–16 h, evidenced by the accumulation of active caspase-3 isoforms, cleaved PARP (a caspase-3 substrate) and the p37 caspase-9 isoform generated by caspase-3-mediated cleavage. In contrast, expression of full-length caspase-8 was not decreased until 16 h, and there was no evidence for active caspase-8 isoforms activated by death receptor apoptosis pathways. This suggests that cisplatin predominantly triggers mitochondrial pathways of apoptosis via caspase-9. Consistent with the idea that Mcl-1 is cleaved by caspases, the accumulation of the 28 kDa Mcl-1 protein was concomitant with the activation of caspase-3 and the p37 caspase-9 cleavage product, and was prevented by the broad specificity caspase inhibitor zVADfmk (Figure 3c). Taken together, these results indicate that Mcl-1 protein, but not on Bcl-2 or Bcl-XL, is effectively cleaved by caspases during cisplatin-induced apoptosis in B-lymphoma cells.
Regulation of Mcl-1 protein in primary malignant B-cells
To extend this analysis, we also analysed the expression of the Bcl-2, Bcl-XL and Mcl-1 proteins during apoptosis of primary malignant B-cells. Spontaneous apoptosis of purified NHL and CLL cells cultured ex vivo (as demonstrated by PARP cleavage) was accompanied by the accumulation of the 28 kDa Mcl-1 protein in all samples analysed (two of five samples analysed are shown in Figure 4). Where tested, both PARP cleavage and the accumulation of the 28 kDa protein were prevented by zVADfmk. Similar to Akata6 cells, apoptosis in some samples was associated with a significant reduction (approx. 50%) in the expression of full-length Mcl-1 (Figure 4a). Interestingly, in other samples (Figure 4b), the levels of full-length Mcl-1 were not altered, although the 28 kDa protein accumulated. Primary samples can be heterogeneous and this observation might reflect preferential apoptosis of cells expressing relatively low levels of Mcl-1, such that cells expressing higher levels become relatively more abundant. An alternate explanation could be that some tumor samples contain modifications that slow the normally rapid turn over of Mcl-1, but do not affect cleavage by caspases. Yet, another possibility could be the induction of Mcl-1 by stress signals (Craig, 2002) in some samples. Whatever the case, in contrast to Mcl-1, Bcl-2 expression was not significantly altered in any samples, with the accumulation of only very low levels of a putative Bcl-2 caspase cleavage product (Cheng et al., 1997). Bcl-XL expression was decreased in some samples, but we did not detect Bcl-XL caspase cleavage products. Therefore, the appearance of a caspase cleavage product of Mcl-1 (but not Bcl-2 or Bcl-XL) was prominent in primary malignant B cells undergoing spontaneous apoptosis, providing a parallel to the observations in the B-lymphoma cell line.
Caspase cleavage of Mcl-1
To determine the potential caspase-3 cleavage site in Mcl-1, we incubated bacterially produced recombinant Mcl-1 lacking the TM region (Mcl-11–329 ΔCT) with caspase-3. N-terminal sequencing of the cleavage products obtained revealed that Mcl-1 was cleaved after Asp127 and Asp157 (Figure 5a). The relative intensities of the different fragments suggested that Asp157 was the preferred site of cleavage in this in vitro experiment. We also incubated full-length radiolabeled 9E10-epitope-tagged Mcl-1 with recombinant caspase-3 (Figure 5b). Consistent with results obtained with purified protein, Mcl-1 was efficiently cleaved by caspase-3 giving the same pattern of cleavage products of approximately 28, 23, 21 and 17 kDa. In reticulocyte lysates, Mcl-1 was cleaved at both sites at approximately equivalent efficiencies. In vitro cleavage was due to caspase activity since it was blocked by zVADfmk (data not shown).
To confirm that caspase-3 cleaved Mcl-1 following Asp127 and Asp157, we introduced aspartate to alanine substitutions at one or both of these positions (Mcl-1D127A, Mcl-1D157A and Mcl-1D127/157A). Mutant Mcl-1 proteins were generated in reticulocyte lysates and incubated with caspase-3 (Figure 5c). As predicted, mutation of Asp127 prevented the production of the 28 and 17 kDa cleavage products, and mutation of Asp157 prevented the production of the 23 and 21 kDa cleavage products. None of the cleavage products were detected when the double mutant, Mcl-1D127/157A, was incubated with caspase-3.
Bcl-2 and Bcl-XL have previously been described as targets for cleavage by caspases (Cheng et al., 1997; Clem et al., 1998), and we therefore compared the relative ability of caspase-3 to cleave these antiapoptotic proteins. Under conditions identical to analysis of Mcl-1, Bcl-2 and Bcl-XL were not cleaved by caspase-3 in reticulocyte lysates (Figure 5d) or using purified recombinant proteins (data not shown). Therefore, Mcl-1 is a relatively efficient target for cleavage by caspases compared to Bcl-2 and Bcl-XL. This is consistent with the stability of Bcl-2 and Bcl-XL in B-cells during apoptosis, and with the observation that Bcl-2 and Bcl-XL cleavage products were not present in abundance (Figures 3 and 4).
Induction of cell death by Mcl-1 cleavage products
It was possible that, similar to Bcl-2 and Bcl-XL (Cheng et al., 1997; Clem et al., 1998), Mcl-1 cleavage generated proteins with cell death-promoting activity. However, before initiating biological studies, it was important to determine what cleavage proteins accumulated in apoptotic cells. Specifically, although caspase-3 cleaved Mcl-1 at two sites in vitro, only one cleavage product of 28 kDa accumulated in apoptotic cells (Figure 3). The S-19 antibody used to detect Mcl-1 was raised against a peptide comprising amino-acid residues 121–138 of human Mcl-1, spanning the proximal cleavage site. Western blotting demonstrated that the S-19 antibody strongly recognized Mcl-11–157 and relatively weakly Mcl-1128–350 (data not shown). S-19 did not detect the other cleavage products because these proteins either lack completely (Mcl-1158–350) or contain only six amino acids (Mcl-11–127) of peptide sequences used to generate the antibody. The 28 kDa cleavage product detected by S-19 in cells comigrated precisely with the Mcl-1128–350 protein obtained in vitro (data not shown). Thus, Mcl-1 is cleaved after Asp127 in intact cells. In contrast, although the S-19 antibody can detect the Mcl-11–157 fragment derived from cleavage after Asp157, this fragment does not accumulate in apoptotic cells.
Based on the findings that Mcl-1 appeared to be predominantly cleaved at Asp127 in intact cells, we determined whether Mcl-1 cleavage products were sufficient to kill cells. We cotransfected NIH3T3 fibroblasts with expression plasmids for the Mcl-11–127 or Mcl-1128–350 proteins and enhanced green fluorescent protein (EGFP), to label transfected cells. The Mcl-1 expression plasmids were used at a fourfold excess relative to the EGFP expression plasmid, to ensure that cells expressing EGFP also received Mcl-1 expression plasmids. After 48 h, cells were analysed by fluorescence microscopy. The majority of cells transfected with control pcDNA3-9E10 or Mcl-11–127 expression plasmids were viable with a flat morphology typical of healthy cells (Figure 6a). In contrast, a significant proportion of cells expressing Mcl-1128–350 were rounded and were detached from the tissue culture plate, characteristic of dead cells, or were fragmented (Figure 6b and c, P<0.001, independent Student's t-test for Mcl-1128–350 expressing cells compared to control cells). Immunoblot analysis demonstrated the expression of 9E10-epitope-tagged Mcl-1 proteins in NIH3T3 cells (Figure 6d).
Mcl-1 is an antiapoptotic Bcl-2 family protein, first identified in differentiating myeloid cells where it plays an important role in cell survival (Kozopas et al., 1993; Moulding et al., 2000; Craig, 2002). Resistance to apoptosis is a hallmark of lymphomagenesis, and the enhanced frequency of lymphomas in mice with enforced Mcl-1 expression suggested that Mcl-1 contributed to human B-cell lymphoma (Zhou et al., 2001). Here, we show that Mcl-1 is widely expressed in B-lymphoma cell lines and primary malignant B-cells, and that antisense-mediated depletion of Mcl-1 is sufficient to promote B-lymphoma cell apoptosis. In contrast to Bcl-2 and Bcl-XL, Mcl-1 was efficiently cleaved by caspases in vitro and during apoptosis, generating a cell death-promoting molecule. Overall, Mcl-1 plays a key role in maintaining B-lymphoma cell survival, and Mcl-1 cleavage is likely to play an important role in the regulation of apoptosis.
The constitutive expression of Mcl-1 in malignant B-cell lines was similar to that detected in differentiating myeloid cells, where Mcl-1 has been shown to play a role in cell survival (Moulding et al., 2000). Abundant expression of Mcl-1 has been demonstrated previously in parental BL41 cells and a subline of BL41 cells, BL41-3, that have an amplification of the Mcl-1 gene (Vrana et al., 2002). However, amplification of the Mcl-1 gene was not detected in the other B-lymphoid cell lines analysed (data not shown) and the mechanism(s) responsible for abundant Mcl-1 expression remain to be determined. Mcl-1 expression was approximately equivalent in B-cells isolated from reactive and malignant lymph nodes. Therefore, high-level expression is not a unique feature of malignant cells. However, whereas Mcl-1 expression may be induced by signaling pathways in normal B-cells within nodal microenvironments, it will be important to determine whether distinct and/or constitutively activated signaling pathways drive Mcl-1 expression in malignant B-cells. Although recent data (Opferman et al., 2003) demonstrate that Mcl-1 is required for the development and maintenance of normal B- and T-lymphocytes, it will also be important to determine if malignant B-cells are particularly sensitive to agents that interfere with Mcl-1 compared to normal cells. The relative expression of Mcl-1 varied somewhat in primary malignant B-cell samples, but the number of samples analysed here is too small to determine whether this relates to malignancy type. We are currently performing a large-scale immunohistochemical study to determine the expression of Mcl-1 in various B-cell malignancies and its association with clinical parameters and outcome.
Mcl-1 is a highly regulated protein, and we also showed that Mcl-1 expression was downregulated during chemotherapy-induced apoptosis in B-lymphoma cells and during spontaneous apoptosis of primary malignant B cells ex vivo. Caspase-mediated cleavage of Mcl-1 appears to play a predominant role in this downregulation, since (i) the kinetics of the decrease in the expression of the full-length Mcl-1, the appearance of the cleavage product and activation of effector caspases and cleavage of PARP matched very closely, (ii) Mcl-1 cleavage was prevented by zVADfmk and (iii) the Mcl-1 cleavage product that accumulated in apoptotic cells comigrated with an Mcl-1 cleavage product generated in vitro. Although other proteases are activated during apoptosis in addition to caspases (Johnson, 2000), Mcl-1 cleavage was also detected during apoptosis induced by the proteasome inhibitor MG132, cathepsin inhibitor I or the aspartic protease inhibitor pepstatin A, demonstrating that these proteases are not required for Mcl-1 cleavage (data not shown). However, there are likely to be other mechanisms contributing to the downregulation of Mcl-1 expression during apoptosis (Iglesias-Serret et al., 2003). For example, we also have found that the activity of the Mcl-1 promoter and levels of Mcl-1 RNA are decreased during cisplatin-induced apoptosis (data not shown). Since Mcl-1 has a short half-life (Kozopas et al., 1993; Akgul et al., 2000; Craig, 2002), the decrease in Mcl-1 mRNA could result in a decrease in protein expression even in the absence of its cleavage by caspases. While these effects might reflect a general shut down in cellular function during apoptosis, Mcl-1 was relatively sensitive to these changes compared to controls. Therefore, reduced de novo synthesis of Mcl-1 might also play a role in the downregulation of the expression of the full-length protein. However, the abundance of the cleavage product that accumulates in apoptotic cells suggests that caspase-mediated cleavage must play a predominant role in our system.
Recently, the translation of Mcl-1 has been shown to be profoundly inhibited within 60 min following ultraviolet irradiation in HeLa cells, and this early downregulation of Mcl-1 expression is thought to be required for proximal events in the apoptosis cascade, including Bax translocation to the mitochondria and cytochrome c release (Nijhawan et al., 2003). However, in cisplatin-treated Akata6 cells, regulation of Mcl-1 did not precede the activation of caspase-9, and zVADfmk prevented both the appearance of the Mcl-1 cleavage product and the decrease in the expression of the full-length protein. Although it is not possible to exclude the possibility that there are small changes in Mcl-1 expression preceding caspase activation, these results demonstrate that the major regulation of Mcl-1 lies downstream of the mitochondria, and suggest that ablation of Mcl-1 is not essential for cytochrome c release to occur in these cells. Rather, in this system, Mcl-1 cleavage seems to be important to amplify the apoptosis cascade following initial cytochrome c release (see below).
Although caspase cleavage of Mcl-1 has been observed recently in other cell systems (Herrant et al., 2002; Snowden et al., 2003), the biological consequences of cleavage have not been reported. The caspase-3 cleavage sites that we identified (Asp127 and Asp157) are conserved in Mcl-1 proteins from human, mice, rats, dog and zebrafish (Figure 7a), consistent with an important biological role. Although Bcl-2 and Bcl-XL have previously been described as caspase substrates (Cheng et al., 1997; Clem et al., 1998), a relatively important role for Mcl-1 cleavage in apoptosis control is consistent with the fact that the Bcl-XL and Bcl-2 cleavage sites are not conserved in non-mammalian species (Cazals-Hatem et al., 1992; Chen et al., 2001).
Neither of the Mcl-1 caspase cleavage sites conformed precisely to optimal sites identified by screening a combinatorial peptide library (Thornberry et al., 1997). However, optimal caspase cleavage sites identified from libraries of peptides may not reflect the hierarchy of sites in intact proteins in cells, and many noncanonical sites have been reported. For example, the proximal site (EELD127) is similar to DELD sites in Acinus, D4-GDI and cPLA2 (Nicholson, 1999), and sites with a glutamate residue at the P4 position have been identified in caspase-2 (EESD) (Cohen, 1997) and topoisomerase I (EEED) (Samejima et al., 1999). Amino-acid residues N-terminal to Asp127 that are important for caspase recognition were very well conserved, including an acidic residue at the critical P4 position. Perhaps consistent with the less prominent role of cleavage at the distal site in intact cells, amino-acid residues immediately N-terminal to Asp157 were poorly conserved, even between mammals where overall identity of Mcl-1 proteins was high. Although both sites are cleaved in vitro by caspase-3, it appears that cleavage is largely limited to the proximal site following Asp127 in cells, since Asp157 cleavage products would be recognized by the S-19 Mcl-1 antibody but do not accumulate in cells. However, we cannot exclude the possibility that Mcl-1 is also cleaved following Asp157, but these products fail to accumulate due to further degradation.
A critical question is whether caspase cleavage of Mcl-1 is biologically significant. Caspase-mediated cleavage of Bcl-2 and Bcl-XL separates the BH1–3 and TM domains of these proteins from their N-terminal BH4 domains, and generates apoptosis-promoting molecules (Cheng et al., 1997; Clem et al., 1998). The Mcl-1 C-terminus is relatively closely related to Bcl-2, and cleavage of Mcl-1 at Asp127 and/or Asp157 within the PEST domain also generates a fragment containing BH1–3 and TM domains (Mcl-1128–350) with cell death-promoting activity (Figure 7b). Although Mcl-1 does not share significant homology with the BH4 domains of Bcl-2 and Bcl-XL at the primary sequence level, this suggests that the N-terminal parts of Mcl-1 perform a functionally equivalent role in protection from cell death. Taken together, with the results of the antisense experiments, these results demonstrate that Mcl-1 cleavage is likely to play a positive role in apoptosis, simultaneously depriving cells of an obligate survival molecule and generating a proapoptotic protein. Interestingly, the Mcl-1128–350 fragment was relatively highly expressed in NIH3T3 cells, suggesting that it may be stabilized relative to full-length Mcl-1, possibly via deletion of PEST sequences. Our results clearly demonstrate the ability of Mcl-1128–350 to kill cells, at least in overexpression studies; however, it will be interesting to determine to what extent this might be dependent on its high relative level of expression.
In an attempt to further investigate the role of Mcl-1 caspase cleavage, we transfected the wild-type Mcl-1 and Mcl-1D127/157A expression constructs into Akata6 cells and then treated the cells with cisplatin (data not shown). It was possible that the mutant form of the protein would be stabilized in apoptotic cells and therefore exhibit enhanced antiapoptotic activity versus the wild-type molecule. Examination of the expression of Mcl-1 protein in untreated transfected cells demonstrated that the transfected wild-type and Mcl-1D127/157A mutant proteins were expressed at equivalent levels. Interestingly, the expression of the exogenous wild-type protein was decreased following exposure to cisplatin, but this was not associated with significantly increased accumulation of caspase cleavage products. Moreover, the Mcl-1D127/157A mutant protein, which is resistant to caspase cleavage in vitro, was also decreased to the same extent upon exposure to cisplatin. This fact precluded a meaningful assessment of the relative antiapoptotic activity of the wild-type and mutant proteins in this system. It appears, therefore, that exogenously overexpressed Mcl-1 protein is not regulated in the same way as the endogenous protein: We have also noted that exogenously overexpressed Mcl-1 is relatively less affected by the inhibition of the proteasome compared to endogenous Mcl-1 (data not shown). It is possible that exogenous Mcl-1 is protected from proteolysis (via the proteasome or caspases) by association with binding proteins or localization to an alternate subcellular compartment. We are currently investigating the mechanisms controlling differential proteolysis of Mcl-1.
Our results support the hypothesis that Mcl-1 plays an important role in regulating apoptosis in B-lymphoma cells. Mcl-1 is required for survival in these cells, functioning upstream of cytochrome c release, since antisense depletion is sufficient to promote apoptosis. However, Mcl-1 downregulation is not obligate for the initiation of apoptosis since there was no evidence for an early change in Mcl-1 preceding caspase activation in cisplatin-treated cells. In contrast, during cisplatin-induced apoptosis, Mcl-1 functions downstream of cytochrome c release; caspase-mediated cleavage of Mcl-1 deprives cells of a survival molecule and generates a cell killing protein, stimulating further activation of caspases in a positive feedback loop. Although previously described as caspase substrates, Bcl-2 and Bcl-XL were relatively inefficient substrates for casapases, suggesting that Mcl-1 plays a particularly prominent role in such a feedback mechanism. Interfering with Mcl-1 expression or function via antisense strategies or small molecule inhibitors such as BH3 mimetics (Oxford et al., 2004) might be an attractive novel anticancer strategy for Mcl-1-positive B-NHL. Recent studies have also demonstrated a critical role for Mcl-1 for cell survival in multiple myeloma (Derenne et al., 2002; Zhang et al., 2002) and in the response of CLL cells to survival signals (Pedersen et al., 2002). Our work therefore extends the range of malignancies where targeting Mcl-1 might be an effective anticancer strategy. However, the potential benefits of interfering with Mcl-1 function should be considered in light of recent data demonstrating an essential role in the normal lymphoid system (Opferman et al., 2003).
Materials and methods
Cell lines and culture
Ramos, Raji and Akata6 cell lines were derived from BL biopsies (Lenoir et al. , 1985; Spender et al., 1999). Akata6 cells were derived by single-cell cloning of the original Akata cell line (Takada et al., 1991; Inman et al., 2001). ML-1 cells were derived from a human myeloid leukemia (Craig et al., 1984). RL and DoHH2 and RPMI 8226 and U-266 cell lines were derived from diffuse large B-cell lymphoma and multiple myeloma, respectively (see Drexler 2001 for original references). All B-cell lines were maintained in RPMI medium supplemented with 10% (v/v) fetal calf serum (FCS) (PAA, Yeovil, UK) and antibiotics. NIH3T3 mouse fibroblasts were maintained in DMEM supplemented with 10% (v/v) FCS and antibiotics. zVADfmk (Calbiochem) was dissolved in dimethylsulfoxide (DMSO) as a 10 mM stock. Cis-diaminedichloroplatinum (cisplatin) was from Roger Bull Laboratories (Warwick, UK). TPA was from Sigma.
Primary malignant and normal B-cells were obtained with Local Research Ethics Committee approval (Southampton and SW Hants LREC Number 158/00). NHL and normal B-cells were purified from malignant or RLN biopsies, respectively, by negative selection using the B-cell isolation kit (Miltenyi Biotec) as described (Dallman and Packham, 2004). CLL cells were isolated by negative selection from involved spleens. Cells were cultured in B-cell medium (IMDM supplemented with 10% heat-inactivated and filtered human serum, 2 mM L-glutamine/penicillin/streptomycin, 0.02 mg/ml gentamycin) and maintained at 37°C and 5% CO2 atmosphere in a humidified incubator. The proportion of B-cells in the purified sample and the expression of immunoglobulin light chains was determined by flow cytometry. Purified samples were >97% B-cells (except NHL3, 94% and CLL2, 83%) and malignant samples were light-chain restricted.
Western blotting was performed as described previously (Packham et al., 1997). Protein content of cell lysates was determined using the BioRad Protein Assay Reagent and equivalent amounts of protein were analysed in each lane. The antibodies used were: Mcl-1 (rabbit polyclonal S-19; Santa Cruz Inc.), Bcl-2 (hamster monoclonal 6C8, BD PharMingen), Bcl-XL (rabbit polyclonal AF800, R&D Systems), PARP (mouse monoclonal C-20, R&D Systems), caspase-3 (rabbit polyclonal #9662, Cell Signaling Technologies), caspase-8 (mouse monoclonal 3-1-9, BD Pharmingen) and caspase-9 (rabbit polyclonal #9502, Cell Signaling Technologies). Myc-epitope-tagged proteins were detected using 9E10 mouse monoclonal antibody hybridoma supernatant. A PCNA-specific mouse monoclonal antibody (a kind gift from Dr X Lu, Ludwig Institute for Cancer Research, London) and β-actin-specific rabbit polyclonal (20–33, Sigma) were used to control for protein loading.
Oligodeoxynucletoides were 2′-O-methoxyethyl-deoxynucleotide chimeric molecules with a uniform phosphorothioate backbone. Mcl-1 ASO (5′-IndexTermTTGGCTTTGTGTCCTTGGCG, ISIS 20408) has previously been shown to be effective in decreasing Mcl-1 expression (Bannerman et al., 2001; Derenne et al., 2002). Missense oligodeoxynucleotide (5′-IndexTermTGGGTCTGGTTTCTCTGTCG, ISIS 105232) was used as control. Akata6 cells were resuspended in serum-free medium at a density of 6 × 106 cells/360 μl in the presence of 50 μ M oligodeoxynucleotide or without added oligodeoxynucleotide (mock transfection control) and electroporated in 4 mm cuvettes using a EasyJect Plus (EQUIBIO) at 250 V and 900 μF (parameters). Transfected cells were cultured in 10 ml complete medium mixed 1 : 1 with conditioned medium.
Identification of caspase cleavage sites in Mcl-1ΔCT
BL21(DE3) bacteria (Novagen) were used to express protein from plasmid pET 15b-Mcl-1ΔCT (containing human Mcl-1 residues 1–329). In short, transformed cells were grown to an OD600 of 1.0 in the LB medium and expression was induced by the addition of isopropyl-beta-D-thiogalactopyranoside (0.4 mM) for 3 h at 37°C. Extraction via sonication and purification of Mcl-1ΔCT via Ni-affinity chromatography (QIAGEN) were performed in 25 mM 4-morpholinepropanesulfonic acid (MOPS) pH 7.5, 300 mM NaCl and 5 mM 2-mercaptoethanol. The resultant Mcl-1ΔCT was 95% pure. Mcl-1ΔCT (25 μg) was cleaved using recombinant caspase-3 (a kind gift from D Hockenbery, Fred Hutchinson Cancer Research Center, Seattle, USA) in 25 mM MOPS pH 7.5, 250 mM NaCl, 2 mM ethylenediaminetetraacetic acid, 2 mM dithiothreitol (DTT). Cleaved Mcl-1ΔCT was resolved by SDS–PAGE, transferred to polyvinylidine difluoride membrane and stained with 0.2% (w/v) Ponceau S Cleavage products were excised and submitted to N-terminal sequencing (Edman Degradation) using a Model 477A peptide sequencer/Model 120A Phenylthiohydantion Analyser (Applied Biosystems).
35S-radiolabeled Bcl-2, Bcl-XL and Mcl-1 proteins were generated by in vitro translation using TNT-coupled reticulocyte lysates (Promega) in the presence of Translabel (Amersham). Lysates (typically 1 μl in a total reaction volume of 10 μl) were incubated with recombinant human caspase-3 (R&D Systems, 125 ng) in cleavage buffer (25 mM N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid (HEPES), pH 7.5, 0.1% (w/v) cholamidopropyldimethylammonio-propanesulfonate (CHAPS), 10 mM DTT) at 37°C. Intact proteins and cleavage products were separated by SDS–PAGE and detected by phosphorimaging.
The Mcl-1 open-reading frame (ORF) was amplified by polymerase chain reaction (PCR) using a plasmid containing the human Mcl-1 cDNA as template. The product was cloned into plasmid pcDNA3-9E10-1 (a kind gift from Dr X Lu, Ludwig Institute for Cancer Research, London) with the 9E10 Myc-epitope tag fused in-frame to the N-terminus of Mcl-1 (to give (p9E10-Mcl-1). Mcl-1 mutants (Mcl-1D127A, Mcl-1D157A and Mcl-1D127/157A) were generated by Quikchange (Invitrogen Corporation) using p9E10-Mcl-1 as a template. Expression plasmids encoding 9E10-epitope-tagged truncated Mcl-1 proteins were generated by PCR using p9E10-Mcl-1 as a template and primers 5′-IndexTermGCCGGATCCTTTGGCCTCAAAAGAAACGCG and 5′-IndexTermCCGTCTAGACTAGTCCAGCTCCTCTTCGGGC for Mcl-11–127, and 5′-IndexTermGCCGGATCCGGGTACGAGCCGGAGCCTC and 5′-IndexTermCCGTCTAGACTATCTTATTAGATATGCCAAACCAGCTCC for Mcl-1128–350. PCR products were cloned into pcDNA3-9E10-1. All constructs were confirmed by direct sequencing. The pBK-CMV-Bcl-XL plasmid was generated by subcloning the human Bcl-XL ORF from pSFFV-Bcl-XL (a kind gift from Dr S Korsmeyer, Howard Hughes Medical Institute, Harvard Medical School, Dana-Farber Cancer Institute, Boston, USA) into pBK-CMV (Stratagene). The pcDNA3-Bcl-2 plasmid was a kind gift from Dr K Vousden (Beatson Institute for Cancer Research, Glasgow, UK). The EGFP expression plasmid pEGFP-N1 was from Clontech.
Cell death assay
NIH3T3 cell were transfected using Transfast (Promega) according to the manufacturer's instructions. Cells (5 × 104) were transfected for 1 h in 24-well plates in a total volume of 200 μl, containing 200 ng of pEGFP-N1 and 800 ng of wild-type or truncated Mcl-1 expression plasmid or pcDNA3-9E10 as a control, and 3 μl of Transfast. Cells were washed thoroughly to remove transfection reagent. Cells were cultured for 48 h and examined by fluorescence microscopy. The number of EGFP-positive cells with a rounded or fragmented phenoypte as a percentage of total EGFP-positive cells was recorded.
Adams JM and Cory S . (2001). Trends Biochem. Sci., 26, 61–66.
Agarwal B and Naresh KN . (2002). Am. J. Hematol., 70, 278–282.
Akgul C, Moulding DA, White MR and Edwards SW . (2000). FEBS Lett., 478, 72–76.
Altmeyer A, Simmons RC, Krajewski S, Reed JC, Bornkamm GW and Chen-Kiang S . (1997). Immunity, 7, 667–677.
Bannerman DD, Tupper JC, Ricketts WA, Bennett CF, Winn RK and Harlan JM . (2001). J. Biol. Chem., 276, 14924–14932.
Bellosillo B, Villamor N, Colomer D, Pons G, Montserrat E and Gil J . (1999). Blood, 94, 2836–2843.
Borner C . (2003). Mol. Immunol., 39, 615–647.
Byrd JC, Kitada S, Flinn IW, Aron JL, Pearson M, Lucas D and Reed JC . (2002). Blood, 99, 1038–1043.
Cazals-Hatem DL, Loie DC, Tanaka S and Reed JC . (1992). Biochim. Biophys. Acta, 1132, 109–113.
Chen M, Gong H-Y, Cheng C-Y, Wang J-P, Hong J-R and Wu J-L . (2001). Biochim. Biophys. Acta, 1519, 127–133.
Cheng EH, Kirsch DG, Clem RJ, Ravi R, Kastan MB, Bedi A, Ueno K and Hardwick JM . (1997). Science, 278, 1966–1968.
Clem RJ, Cheng EH, Karp CL, Kirsch DG, Ueno K, Takahashi A, Kastan MB, Griffin DE, Earnshaw WC, Veliuona MA and Hardwick JM . (1998). Proc. Natl. Acad. Sci. USA, 95, 554–559.
Cohen GM . (1997). Biochem. J., 326, 1–16.
Craig RW . (2002). Leukemia, 16, 444–454.
Craig RW, Frankfurt OS, Sakagami H, Takeda K and Bloch A . (1984). Cancer Res., 44, 2421–2429.
Cuconati A, Mukherjee C, Perez D and White E . (2003). Genes Dev, 17, 2922–2932.
Dallman CL and Packham G . (2004). Purification of primary malignant B-cells and immunoblot analysis of Bcl-2 family proteins. Molecular Methods in Cancer, Illidge T and Johnson PWM (eds) (in press).
Derenne S, Monia B, Dean NM, Taylor JK, Rapp MJ, Harousseau JL, Bataille R and Amiot M . (2002). Blood, 100, 194–199.
Drexler HG . 2001. The Leukemia–Lymphoma Cell Line Facts Book. Academic Press: London.
Ghia P, Boussiotis VA, Schultze JL, Cardoso AA, Dorfman DM, Gribben JG, Freedman AS and Nadler LM . (1998). Blood, 91, 244–251.
Gross A, McDonnell JM and Korsmeyer SJ . (1999). Genes Dev., 13, 1899–1911.
Herrant M, Luciano F, Loubat A and Auberger P . (2002). Oncogene, 21, 4957–4968.
Iglesias-Serret D, Pique M, Gil J, Pons G and Lopez JM . (2003). Arch. Biochem. Biophys., 417, 141–152.
Inman GJ, Binne UK, Parker GA, Farrell PJ and Allday MJ . (2001). J. Virol., 75, 2400–2410.
Johnson DE . (2000). Leukemia, 14, 1695–1703.
Khoury JD, Medeiros LJ, Rassidakis GZ, McDonnell TJ, Abruzzo LV and Lai R . (2003). J. Pathol., 199, 90–97.
Kitada S, Andersen J, Akar S, Zapata JM, Takayama S, Krajewski S, Wang HG, Zhang X, Bullrich F, Croce CM, Rai K, Hines J and Reed JC . (1998). Blood, 91, 3379–3389.
Kozopas KM, Yang T, Buchan HL, Zhou P and Craig RW . (1993). Proc. Natl. Acad. Sci. USA, 90, 3516–3520.
Lenoir GM, Vuillaume M and Bonnardel C . (1985). IARC Sci. Publ., 60, 309–318.
Lomo J, Smeland EB, Krajewski S, Reed JC and Blomhoff HK . (1996). Cancer Res., 56, 40–43.
Moulding DA, Giles RV, Spiller DG, White MR, Tidd DM and Edwards SW . (2000). Blood, 96, 1756–1763.
Nicholson DW . (1999). Cell Death Differ., 6, 1028–1042.
Nijhawan D, Fang M, Traer E, Zhong Q, Gao W, Du F and Wang X . (2003). Genes Dev., 17, 1475–1486.
Opferman JT, Letai A, Beard C, Sorcinelli MD, Ong CC and Korsmeyer SJ . (2003). Nature, 426, 671–676.
Oxford SME, Dallman CL, Johnson PWM, Ganesan A and Packham G . (2004). Current Med. Chem., (in press).
Packham G . (1998). Apoptosis, 326, 75–82.
Packham G, Brimmell M and Cleveland JL . (1997). Biochem. J., 328, 807–813.
Pagnano KB, Silva MD, Vassallo J, Aranha FJ and Saad ST . (2002). Acta Haematol., 107, 29–34.
Pedersen IM, Kitada S, Leoni LM, Zapata JM, Karras JG, Tsukada N, Kipps TJ, Choi YS, Bennett F and Reed JC . (2002). Blood, 100, 1795–1801.
Pepper C, Ali K, Thomas A, Hoy T, Fegan C, Chowdary P, Kell J and Bentley P . (2002). Eur. J. Haematol., 69, 227–235.
Samejima K, Svingen PA, Basi GS, Kottke T, Mesner Jr PW, Stewart L, Durrieu F, Poirier GG, Alnemri ES, Champoux JJ, Kaufmann SH and Earnshaw WC . (1999). J. Biol. Chem., 274, 4335–4340.
Snowden RT, Sun XM, Dyer MJ and Cohen GM . (2003). Leukemia, 17, 1981–1989.
Soini Y, Raunio H and Paakko P . (1998). Tumour Biol., 19, 176–185.
Spender LC, Cannell EJ, Hollyoake M, Wensing B, Gawn JM, Brimmell M, Packham G and Farrell PJ . (1999). J. Virol., 73, 4678–4688.
Spets H, Stromberg T, Georgii-Hemming P, Siljason J, Nilsson K and Jernberg-Wiklund H . (2002). Eur. J. Haematol., 69, 76–89.
Takada K, Horinouchi K, Ono Y, Aya T, Osato T, Takahashi M and Hayasaka S . (1991). Virus Genes, 5, 47–156.
Thornberry NA, Rano TA, Peterson EP, Rasper DM, Timkey T, Garcia-Calvo M, Houtzager VM, Nordstrom PA, Roy S, Vaillancourt JP, Chapman KT and Nicholson DW . (1997). J. Biol. Chem., 272, 17907–17911.
Vrana JA, Bieszczad CK, Cleaveland ES, Ma Y, Park JP, Mohandas TK and Craig RW . (2002). Cancer Res., 62, 892–900.
Yang T, Kozopas KM and Craig RW . (1995). J. Cell. Biol., 128, 1173–1184.
Zhang B, Gojo I and Fenton RG . (2002). Blood, 99, 1885–1893.
Zhou P, Levy NB, Xie H, Qian L, Lee CY, Gascoyne RD and Craig RW . (2001). Blood, 97, 3902–3909.
We thank Drs X Lu, K Vousden, DM Hockenbery and SJ Korsmeyer for the kind gift of reagents, and the FHCRC Proteomic Resources Center for N-terminal sequencing. This work was supported by NIH Grant RO1-CA57359 (RWC), American Cancer Society Grant RPG-97-173-01-LBC (KYJZ) and grants from Cancer Research UK and the Leukemia Research Fund.
About this article
Cite this article
Michels, J., O'Neill, J., Dallman, C. et al. Mcl-1 is required for Akata6 B-lymphoma cell survival and is converted to a cell death molecule by efficient caspase-mediated cleavage. Oncogene 23, 4818–4827 (2004). https://doi.org/10.1038/sj.onc.1207648
- non-Hodgkin's lymphoma
Expert Opinion on Investigational Drugs (2020)
Blood Reviews (2020)
Discovery of coumarin derivatives as potent and selective cyclin-dependent kinase 9 (CDK9) inhibitors with high antitumour activity
European Journal of Medicinal Chemistry (2020)
Biochemical and Biophysical Research Communications (2020)
Loss of FBXW7-mediated degradation of BRAF elicits resistance to BET inhibitors in adult T cell leukemia cells
Molecular Cancer (2020)