Bax translocation is crucial for the sensitivity of leukaemic ... - Nature

Oncogene (2001) 20, 4817 ± 4826 ã 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00 www.nature.com/onc

Bax translocation is crucial for the sensitivity of leukaemic cells to etoposide-induced apoptosis Li Jia*,1, Yasmeen Patwari1, Srinivasa M Srinivasula2, Adrian C Newland1, Teresa Fernandes-Alnemri2, Emad S Alnemri2 and Stephen M Kelsey1 1

Department of Haematology, St. Bartholomew's and The Royal London School of Medicine and Dentistry, London, E1 2AD, UK; Department of Microbiology and Immunology, Kimmel Cancer Institute, Thomas Jeerson University, Philadelphia, Pennsylvania, PA 19107, USA

2

Bax translocation from cytosol to mitochondria is believed to be a crucial step for triggering cytochrome c release from mitochondria. However, it is unclear whether Bax translocation is associated with Bax induction by DNA damaging agents. The induction of Bax in response to DNA damaging agents has been considered to be linked with p53. In this study, we used the p53 negative human chronic myeloid leukaemia K562 cell line. Bax up-regulation occurred at the whole cell level after DNA damage induced by etoposide. However, after incubation with etoposide, Bax failed to translocate to mitochondria and as a result, the apoptotic process was blocked. A Bax stable transfectant, the K/Bax cell line, expressed more Bax protein in the cytosol, mitochondria and nuclei. This Bax overexpression induced cytochrome c release, a reduction of cytochrome c oxidase activity and mitochondrial membrane potential (DCm). However, Bax-induced apoptosis was blocked downstream of mitochondria in K562 cells. The increased levels of mitochondrial Bax sensitized cells to etoposideinduced activation of caspases-2, -3 and -9 and apoptosis. However, after transient transfection with the Apaf-1 gene, K/Bax cells were sensitized to etoposide-induced caspase activation and apoptosis to a larger extent compared with Bax or Apaf-1 transfection alone. We therefore conclude that two mechanisms contribute to the resistance of K562 cells to etoposide-induced apoptosis; ®rstly failure of Bax targeting to mitochondria and, secondly, de®ciency of Apaf-1. Uncoupling of Bax translocation from Bax induction can occur in response to etoposide-induced DNA damage. Oncogene (2001) 20, 4817 ± 4826. Keywords: apoptosis; Bax expression and translocation; caspases; chemotherapy; etoposide; leukaemia

*Correspondence: L Jia, Department of Haematology, St. Bartholomew's and The Royal London School of Medicine and Dentistry, Turner Street, London E1 2AD, UK; E-mail: [email protected] Received 23 February 2001; revised 9 May 2001; accepted 10 May 2001

Introduction Resistance to apoptosis is an important cause of failure of chemotherapy in human leukaemia. Cytotoxic drugs such as etoposide and anthracycline antibiotics can inhibit the DNA repair enzyme, topoisomerase II. As a result, exposure to cytotoxic drug results in the induction of double strand breaks in leukaemic cell DNA. DNA damage is sensed by dedicated machinery, and responses including DNA repair, cell cycle arrest or apoptosis (Evan and Littlewood 1998; Fearnhead et al., 1998). The treatment of leukaemic cells with these DNA damaging agents is associated with the induction of apoptosis via the so-called `mitochondrial pathway'. The pathway is initiated when mitochondria, in response to DNA damage or other signals, release cytochrome c (Liu et al., 1996; Yang et al., 1997; Jia et al., 2001). Cytochrome c binds to a cytosolic adapter protein, Apaf-1, to activate caspase-9 through autoproteolysis (Li et al., 1997). Active caspase-9 proteolyses and activates caspase-3 and the resulting activated caspase-3 then activates other downstream caspases causing the apoptotic phenotype. It is unclear how DNA damage is linked to cytochrome c release from mitochondria. The p53 tumour suppressor protein transmits signals arising from DNA damage to genes and factors that induce cell cycle arrest or apoptosis. Activated p53 acts as a transactivator of several other genes including the cyclin-dependent-kinase inhibitor p21, IGF-BP3, and the pro-apoptotic protein Bax (Buckbinder et al., 1995), which plays an important role in triggering cytochrome c release from mitochondria. Bax is present mainly in the cytosol during growth arrest and localizes to mitochondria during apoptosis (Deng and Wu, 2000). Thus, Bax induces cytochrome c release only when it targets to the mitochondrial outer membrane (Wolter et al., 1997; Jia et al., 1999). As a target of p53, Bax protein expression can be up-regulated in a number of systems during p53-mediated apoptosis (Miyashita and Reed, 1995; Zhan et al., 1994). P53 induces mitochondrial cytochrome c release by a pathway requiring cytosolic Bax (Schuler et al., 2000). However, p53 does not directly induce Bax translocation to mitochondria (Deng and Wu, 2000).

Bax induction and translocation L Jia et al

4818

Antiapoptotic members of the Bcl-2 protein family act to prevent cytochrome c release from the mitochondria (Yang et al., 1997; Kluck et al., 1997). Chronic myeloid leukaemia (CML)-associated Bcr ± abl tyrosine kinase also inhibits chemotherapeutic druginduced mitochondrial membrane potential (DCm) collapse and cytochrome c release, thereby blocking the activation of the downstream caspases and apoptosis (Martins et al., 1997; Amarante-Mendes et al., 1998). The K562 cell line, which was derived from human CML, displays a relatively high level of resistance to most cytotoxic drugs, probably due to a combination of Bcr ± abl oncogene expression and p53 protein de®ciency (Prokocimer et al., 1986; Martins et al., 1997; Amarante-Mendes et al., 1998; Landowski et al., 1999; Fang et al., 2000). However, this cell line has been con®rmed to be Bcl-2 protein negative (Perkins et al., 2000). In previous studies, we have found that K562 cells are resistant to UV light-induced apoptosis. This is associated with resistance to cytochrome c-dependent activation of caspases-9 and -3 in a cell-free system. The resistance to UV light can be partially overcome by transfection of the Apaf-1 gene or the addition of puri®ed Apaf-1 protein in a cell-free system (Jia et al., 2001). In this study, we found that Apaf-1 transfection alone could not increase the sensitivity of K562 cells to etoposide-induced apoptosis. Exposure to etoposide can increase Bax expression, which was independent of p53 in the K562 cell line. However, following DNA damage in K562 cells Bax failed to translocate to mitochondria. Enforced Bax targeting to mitochondria by developing a Bax stable transfectant in K562 cells (K/Bax) sensitized these cells to etoposide-induced apoptosis. Apaf-1 can sensitize K562 cells to etoposide-induced apoptosis only in stable Bax transfected cells.

Results Increased levels of Apaf-1 protein failed to sensitize K562 cells to etoposide-induced apoptosis The topoisomerase II (topo II) inhibitor etoposide induced rapid DNA double strand breaks. After treatment with 10 mg/ml etoposide for 2 h, nearly 90% duplex DNA was unwound to single-stranded DNA (Figure 1a). Using ¯ow cytometry, we con®rmed that K562 cells were resistance to apoptosis and only 37% of cells had died by apoptosis after 72 h exposure to etoposide. We have previously reported that K562 cells were resistant to UV light-induced apoptosis due to a de®ciency of functional Apaf-1 protein or an existing inactive isoform, Apaf-1 XL (Jia et al., 2001; Fu et al., 2001), and that this resistance can be overcome by transfection of the Apaf-1 gene into the K562 cell line (Jia et al., 2001). We were therefore interested in whether transfection of Apaf-1 could increase the sensitivity of K562 cells to etoposideinduced apoptosis. Increased levels of Apaf-1 were seen Oncogene

Figure 1 Eect of Apaf-1 on etoposide-induced apoptosis. (a) Etoposide-induced DNA damage. K562 cells were treated with 10 mg/ml etoposide. After 2 h, DNA double strand breaks were assessed by the alkaline unwinding assay. DNA damage was expressed by decreased ratio of duplex DNA to total DNA. (b) Etoposide-induced apoptosis in the parental K562 and Apaf-1 transfected K562 cells. K562 cells were transiently transfected with Apaf-1 plasmid DNA. The inserted Western blot in (b) is the increased expression of the Apaf-1 protein which was detected after 24 h transfection using polyclonal anti-Apaf-1 antibody at 1 : 1500 dilution for 50 mg protein/lane. 1 represents K562 cells and 2 indicates Apaf-1 transfected K562 cells. After 24 h, cells were washed and resuspended with culture medium. Cells were then treated with 10 mg/ml etoposide. Etoposide-induced apoptosis was analysed by ¯ow cytometry

after transfection for 24 h (Figure 1b insert). However, Apaf-1 transfected cells were not signi®cantly more sensitive to etoposide-induced apoptosis (Figure 1b, P40.05, ANOVA). This result indicated a block conferring resistance to etoposide-induced apoptosis in K562 cells which is upstream of Apaf-1. Etoposide-induced Bax overexpression and Bax targeting It has been reported that the K562 cell line does not express p53. P53 expression in K562 cells is blocked at both mRNA and protein levels without any change in DNA (Lubbert et al., 1988). We have con®rmed that there was no p53 protein expression in our K562 cells by Western blot analysis (data not shown). We initially proposed that etoposide might not induce Bax expression in the K562 cell line due to the absence of the p53 protein. However, we observed that etoposide did induce Bax expression which started from 12 h and continued to increase until 48 h (Figure 2a). However, Bax expression was reduced again after 60 and 72 h. This result demonstrated that etoposide-induced Bax expression is p53 independent in the K562 cell line. Bax overexpression appeared to be an early response of K562 cells to etoposide-induced DNA damage. We further tested whether increased Bax would target to mitochondria. Mitochondria were isolated at 24, 48 and 72 h after treatment with etoposide. Bax protein was not increased at the mitochondrial level, in contrast to the whole cell level, but in fact decreased at 48 and 72 h (Figure 2b). This suggests that etoposide can induce Bax overexpression but not Bax targeting to mitochondria in the K562 cell line. Moreover, mitochondrial Bax levels started to fall after exposure to etoposide.


Figure 2 Etoposide-induced Bax overexpression and translocation. (a) Etoposide-induced Bax overexpression. Bax expression was detected by Western blotting using whole cell lysate (50 mg protein/lane). Polyclonal rabbit anti-Bax antibody (D21) was used at a 1 : 200 dilution. b-actin at a 1 : 10 000 dilution served as protein loading control. (b) Bax expression on mitochondria. Mitochondrial proteins (100 mg protein/lane) from samples 0, 24, 48 and 72 h treatment of etoposide were subjected to 12% SDS ± PAGE. Anti-Bax antibody (D21) was used at a 1 : 200 dilution

Mitochondrial alterations induced by Bax are insufficient to cause apoptosis in the K562 cell line We then tested whether Bax overexpression by transfection rather than etoposide exposure might induce apoptosis in K562 cells. We transfected the Bax-pRSC-lacZ plasmid DNA into K562 cells. Transient transfection of Bax did not induce apoptosis in K562 cells even though Bax transfection induced apoptosis in the apoptosis sensitive cell lines, such as the CEM/VLB100 and the U937 cell line after 24 h transfection (data not shown). We therefore developed a Bax stable transfectant cell line, named the K/Bax cell line. Increased Bax expression in K/Bax cells was detected at cytosolic, mitochondrial and nuclear levels by the Western blot analysis (Figure 3a). The transient transfection of Bax induced some degree of cytochrome c release (Figure 3b). The DCm was decreased after Bax transfection when compared with parental K562 cells (Figure 3c). Decreased cytochrome c content in the mitochondrial intermembrane space caused signi®cantly reduced activity of the cytochrome c oxidase (COX) or complex IV (Figure 3d) in the K/Bax cell line, as measured by the oxygen electrode. However, Bax overexpression-induced cytochrome c release, did not activate downstream caspases (data not shown). Thus, Bax translocation to mitochondria, cytochrome c release and the reduction of DCm are not sucient to induce apoptosis in the K562 cell line. K/Bax cells showed increased sensitivity to etoposideinduced apoptosis and activation of caspases We therefore tested whether Bax overexpression or Bax mitochondrial targeting could sensitize K562 cells to etoposide-induced apoptosis. As expected, K/Bax cells signi®cantly increased their susceptibility to etoposideinduced apoptosis (Figure 4a). Etoposide-induced Bax overexpression also occurred in the K/Bax cell line. The Bax protein levels continued to increase up to 72 h (Figure 4b). Again, Bax translocation was not observed

in K/Bax cells after treatment with etoposide as analysed by Western blotting (Figure 4c). Using immuno-¯uorescence staining on Bax, we further con®rmed that the intracellular distribution of Bax in both K562 and K/Bax cells did not change after treatment with etoposide for 48 h (Figure 5a ± d). The mitochondrial location was stained with a red mitochondrion-selective vital dye, Mito-Tracker, and Bax was stained green. The cytosolic Bax was stained a green colour and mitochondrial Bax was shown as yellow/orange when the green merged with red. There was no distinct dierence in Bax localization before (Figure 5a and c) and after (Figure 5b and d) treatment with etoposide. Mito-Tracker is also a DCm-sensitive dye. Loss of DCm, as shown by loss of red colour, was observed in some cells after treatment with etoposide in both K562 (Figure 5b) and K/Bax cells (Figure 5d). Etoposide-induced activation of caspases-2, 3, and 9 was compared between K562 parental and K/Bax cells by both ¯uorogenic assay and Western blot analysis. Both caspases-2 and -3 are downstream of caspases-9 when they respond to cytochrome c (Slee et al., 1999). Signi®cantly increased activity of these caspases, as shown by the ability to cleave ¯uorogenic substrates, was observed in K/Bax cells compared with K562 parental cells following exposure to etoposide (Figure 6a ± c, *P50.0001 to **0.001, ANOVA). Therefore, although Bax overexpression alone does not induce apoptosis, it sensitizes K562 cells to etoposide-induced apoptosis upstream of caspases-2, -3 and -9, probably via facilitating mitochondrial cytochrome c release.

4819

Apaf-1 increases the susceptibility to etoposide-induced apoptosis and activation of caspases in the cells overexpressing Bax Transient transfection of Apaf-1 into stable transfected K/Bax cells caused apoptosis (about 10%). This also induced signi®cantly increased sensitivity to etoposideinduced apoptosis compared with the K562 parental cells or those transfected with Bax or Apaf-1 alone (Figure 7a). Cells underwent apoptosis at 24 h, which could not be observed with Bax overexpression alone. Apaf-1 transfection signi®cantly increased the sensitivity of K/Bax cells to etoposide-induced activation of caspase-2 (P50.0001, t-test), caspase-3 (P50.001), and caspase-9 (P50.0001) at 24 h (Figure 8a ± c), at least in an additive manner. Discussion A critical determinant of the ecacy of chemotherapy is the response of leukaemic cells to DNA damage induced by anticancer agents. We have previously shown that leukaemic cells respond dierently to UV light-induced DNA damage. The mitochondrial barrier is less eective on UV light-induced cytochrome c release. After exposure to UV light, the major block to apoptosis is downstream of mitochondria, and it can be overcome by increasing Apaf-1 levels in resistant Oncogene


4820

Figure 3 Bax overexpression induced mitochondrial alterations. (a) Increased Bax protein levels at whole cells, cytosol, mitochondria (Mito) and nuclei. Fifty mg protein from each cellular fraction were loaded onto 12% SDS ± PAGE. The monoclonal mouse anti-Bax antibody (YTH-2D2, R&D) was used at a 1 : 1000 dilution. (7) Indicates K562 parental cells and (+) means Bax transfected cells. (b) Bax transfection induced cytochrome c release from mitochondria into cytosol. Fifty mg cell lysates and cytosol protein and 25 mg mitochondrial protein were subjected to 14% SDS ± PAGE. The monoclonal mouse anti-cytochrome c antibody (7H8.2C12) was used at a 1 : 1000 dilution. (c) Bax induced DCm reduction. DiOC6 (3) stained cells were analysed by ¯ow cytometry. Decreased DCm in K/Bax cells was shown by a shifted peak in the FL1-H channel. (d) Bax-induced decrease in the activity of COX. The activity of COX was measured by the oxygen electrode after addition of electron donors TEMD/ascorbate in the presence of antimycin A. Data shown are means+s.d. from six samples. Signi®cant dierence *P50.0001 was analysed by the t-test

leukaemic cells, such as the K562 cell line (Jia et al., 2001). It has been recently reported, however, that levels of Apaf-1 protein alone have no relationship with the response to induction chemotherapy in acute leukaemia (Svingen et al., 2000). In this study, we found that increased Apaf-1 level alone could not sensitize K562 cells to etoposide-induced apoptosis. In fact, the major block, that confers resistant to etoposide-induced apoptosis on K562 cells, was due to the failure of the cytosolic Bax translocation to mitochondria, despite the fact that Bax expression was upregulated on exposure to etoposide. Mitochondrial Bax and cytosolic Apaf-1 levels together are both crucial for the sensitivity of K562 cells to etoposide-induced apoptosis. The K562 cell line is resistant to etoposide-induced apoptosis but not DNA damage. Following exposure to etoposide more than 90% of the double-stranded DNA was broken down at 2 h. This response is similar to etoposide-sensitive cell lines, such as CEM and CEM/VLB100 (data not shown) but was not associated with apoptosis. The resistance to etoposide-induced apoptosis in K562 cells is known to be associated with the induction of cell cycle arrest in G2/M phases (Amarante-Mendes et al., 1998) and the block is upstream of cytochrome c release or reduction of Oncogene

DCm. This has been suggested to be due to Bcr ± abl expression in K562 cells (Martins et al., 1997 and Amarante-Mendes et al., 1998). In addition to the upstream block, we also found that K562 cells are resistant to cytochrome c-induced activation of caspases-9 and -3 in a cell-free system which is due to the lower level of functional Apaf-1 protein (Jia et al., 2001). However, transfection of the Apaf-1 gene failed to increase the sensitivity of K562 cells to etoposide-induced apoptosis. The link between DNA damage and cytochrome c release remains obscure. Bax is an important factor for cytochrome c release from mitochondria and it has been shown that Bax translocates from the cytosol to mitochondria when overexpressed by transfection or in response to certain cell death stimuli (Hsu et al., 1997; Wolter et al., 1997; Rosse et al., 1998; Jia et al., 1999; Deng and Wu, 2000). Bax induction has been linked with the p53-mediated DNA damage response (Miyashita and Reed 1995; Adams and Cory 1998; Deng and Wu, 2000). We initially proposed that the K562 cell line might fail to induce Bax expression in response to etoposide-induced DNA damage due to absent p53 protein. Using Western blot analysis, however, we detected remarkably increased Bax protein expression in response to etoposide at the whole cell level. Our


4821

Figure 4 Eects of etoposide on apoptosis, Bax expression and localization in K562 parental and K/Bax cells. K562 and K/Bax cells (56105/ml) were treated with 10 mg/ml etoposide and incubated for up to 72 h. Cells were collected at each of the indicated time points. After ®xing and staining procedures, DNA contents were analysed using ¯ow cytometry. (a) Apoptosis. The signi®cantly increased apoptosis (*P50.001) in K/Bax cells were statistically analysed by ANOVA in comparison to the K562 parental cells. (b) Bax expression in K/Bax cells. Fifty mg protein extracted from whole cells was loaded onto 12% SDS ± PAGE. The monoclonal anti-Bax antibody was used at a 1 : 1000 dilution. b-actin served as a loading control. (c) Bax expression on mitochondria in K/Bax cells. One hundred mg/ml mitochondrial protein was loaded onto SDS ± PAGE. The monoclonal anti-Bax antibody was used at a 1 : 1000 dilution

result has demonstrated that DNA damage-associated Bax induction can occur in cells which lack p53 protein. Etoposide-mediated Bax induction in the K562 cell line was also reported by Fukumi et al. (2000). However, mitochondrial Bax levels were not signi®cantly increased compared with the whole cell level. It is unclear why induction of Bax protein is uncoupled from its translocation to mitochondria in response to etoposide. It has been reported that Bax translocation from cytosol to mitochondria requires a protein, called Peg3/Pw1. This protein is up-regulated during p53-mediated apoptosis and functions as a modulator downstream of p53 to regulate Bax redistribution in the cells. However, Bax translocation has been observed in a p53-independent system, and Peg3/Pw1-induced Bax translocation does not require p53 (Deng and Wu, 2000). It is unclear about the role of Bcr ± abl protein on Bax induction and translocation. Bcr ± abl transfected HL-60 cells did not show any dierence in Bax expression compared with HL-60 parental cells (Amarante-Mendes et al., 1998). The Bcr-abl inhibitor, STI571, did not change Bax levels in both K562 and HL60/Bcr ± abl cells (Fang et al., 2000). Our own studies could not exclude the role of Bcr ± abl on the translocation of Bax. Bax translocation can occur without any change to its expression (Jia et al., 1999). In unstimulated cells, Bax is located in the cytosol and in peripheral association with intracellular membranes including mitochondria, but inserts into the mitochondrial membrane after a death signal (Wolter et al., 1997). However, Bax translocation seems to be independent on levels of Bax in the cell. Unlike DNA damage-induced Bax overexpression, stable transfected

Bax can distribute more or less evenly in cytosol, mitochondria and nuclei in the K/Bax cell line. However, mitochondrial Bax can be detected in some cell lines in the resting states, such as the CEM/VLB100 cell line (Jia et al., 1999). Both naturally expressed and transfected Bax can translocate to mitochondria in K562 cells. Bax stable transfectant, K/Bax cells, were sensitized at the mitochondria. Bax localization to the mitochondria is a vital aspect of its death promoting ability. Enforced Bax overexpression causes cytochrome c release, decreased COX activity and reduced DCm, which are associated with Bax levels in the mitochondria. These mitochondrial alterations are sucient to induce apoptosis if the downstream caspase cascade is intact. However, mitochondrial localization of overexpressed Bax causes neither apoptosis nor cell growth arrest in the K562 cell line. This result is consistent with our previous ®nding in which Apaf-1 de®ciency in the K562 cell line confers resistance to cytochrome cinduced caspase activation. However, K/Bax cells are more sensitive to etoposide-induced activation of caspases-2, -3 and -9 and apoptosis. We suggest there may be two possibilities. First, the higher level of mitochondrial Bax may increase the amount of cytochrome c release in response to etoposide and thus enhance the reaction of the cytochrome cdependent mitochondrial pathway. Even in the presence of less functional Apaf-1 protein, the increased cytochrome c levels could facilitate apoptosomeinduced cleavage of procaspase-9. Second, mitochondria in the K/Bax cells are partially dysfunctional which may make them more sensitive to etoposideinduced apoptosis. Oncogene


4822

Figure 5 Immuno¯uorescence analysis of intracellular localisation of Bax. K562 cells (a and b) and K/Bax cells (c and d) were cultured in the absence (a and c) and in the presence of 10 mg/ml etoposide for 48 h. Cells were labelled with Mito-Tracker, and stained with anti-Bax antibody. The Bax antibody was visualized with an FITC-conjugated anti-mouse IgG and assigned the colour green, whereas, mitochondria labelled with Mito-Tracker were assigned the colour red. The cytosolic or nuclear Bax was shown as detected a green colour and the mitochondrial Bax was yellow/orange

Although K/Bax cells have increased sensitivity to etoposide-induced caspase activation and apoptosis, it is limited by comparison with that seen after transfection with the Apaf-1 gene. Apaf-1 can sensitize K562 cells to etoposide-induced apoptosis when the mitochondria are activated. Together with mitochondrial Bax, Apaf-1 sensitizes K562 cells to etoposide-induced apoptosis. We also found that Apaf-1 de®ciency exists not only in leukaemia cell lines but also occurs in the primary leukaemia blasts (Jia et al., 2001). In summary, we proposed that the failure of Bax translocation to mitochondria combined with insucient levels of Apaf-1 protein are largely responsible for the resistance of these cells to chemotherapy-induced apoptosis. The reason for the failure of Bax to translocate appropriately requires further study. Materials and methods Materials Etoposide was obtained from PCH Pharmachemie (Haarlem, The Netherlands). E. coli DH5a strain, DMRIE-C reagent, Oncogene

serum-free OPTI-MEM medium and Geneticin (G418 sulphate) were purchased from Gibco ± BRL (West Sussex, UK). Plasmid DNA MiniPrep Kit was purchased from QIAGEN (Valencia, CA, USA). b-Gal Assay Kit was purchased from Invitrogen (Leek, The Netherlands). DiOC6(3) and Mito-TrackerTM red CMXRos were purchased from Molecular Probes (Leiden, The Netherlands). Fluorescent substrates, Z-Val-Asp-Val-Ala-Asp-AFC (Z-VDVADAFC), Ac-Asp-Glu-Val-Asp-AFC (Ac-DEVD-AFC), and Ac-Leu-Glu-His-Asp-AFC (Ac-LEHD-AFC), and Hoechst 33258 were obtained from Calbiochem (Nottingham, UK). Rabbit anti-Apaf-1 antibody was from QED Bioscience Inc. (San Diego, CA, USA). Mouse anti-caspase-3 antibody was from BD Transduction Laboratories (Lexington, KY, USA). Rabbit anti-caspase-2 and mouse anti-Bax (clone YTH-2D2) antibodies were from R&D Systems (Oxon, UK). Rabbit anti-Bax antibody (D21) was from Santa Cruz (Wiltshire, UK). Rabbit anti-caspase-3 antibody was from StressgenBioquote (North Yorkshire, UK). Rabbit anti-caspase-9 and denatured mouse anti-cytochrome c (clone 7H8.2C12) antibodies were from BD PharMingen (San Diego, CA, USA). SuperSignal enhanced chemiluminescence (ECL) was from Pierce (Rockford, IL, USA). Propidium iodide (PI), mouse anti-b-actin antibody and all chemicals were purchased from Sigma (Dorset, UK).


4823

Figure 6 Etoposide-induced activation of caspases and cleavage of procaspases. (a) Caspase-2. Z-VDVAD-AFC was used as substrate for caspase-2. Procaspase-2 cleavage was assessed using whole cell lysate (50 mg/lane) and polyclonal anti-caspase-2 antibody was used at a 1 : 1000 dilution. (b) Caspase-3. Ac-DEVD-AFC was used as substrate for caspase-3. Monoclonal antiprocaspase-3 antibody was used at a 1 : 500 dilution and b-actin was co-probed at a 1 : 10 000 dilution. Procaspase-3 band was detected at 32 kD and b-actin was at 41 kD. Activated caspases-3 bands were probed by rabbit anti-caspase-3 antibody (StressgenBioquote) at a 1 : 7000 dilution and were shown at 20 and 18 kD. (c) Caspase-9. Ac-LEHD-AFC was used as substrate. Polyclonal anti-caspase-9 antibody (PharMingen) was used at a 1 : 1000 dilution and procaspase-9 bands were detected at 45 kD. The activity of caspases was expressed as mM/h/mg protein. Data shown are means+s.d. from three independent experiments. Signi®cant dierence (*P50.0001; **P50.001) was analysed by ANOVA in comparison to non-transfected K562 cells

Cell lines and gene transfection The human chronic myeloid leukaemic K562 cell line was used in this study. These cell lines were cultured in RPMI1640 medium as described previously (Jia et al., 1996).

Flag-tagged Apaf-1L pCMV2 plasmid and Bax- pRSC-Lac Z plasmid were grown in E. coli DH5a strain and were puri®ed using Plasmid DNA MiniPrep Kit. Four mg plasmid DNA was transfected into 26106 leukaemic cells using DMRIE-C reagent in serum-free OPTI-MEM Oncogene


4824

medium. After culturing at 378C for 5 h, cell culture conditions were recovered by the addition of 15% fetal calf serum containing RPMI-1640 medium. For transient transfection cells were harvested after 48 h of transfection. Transfection eciency was 80 ± 100% as measured by cotransfection of 2 mg pRSC-GFP plasmid and observed by ¯uorescence microscopy. For stable transfection of the Bax gene in the K562 cell line, transfectants were selected in the presence of 0.5 mg/ml Geneticin (G418 sulphate) for 1 month and then further selected by single cell cloning in 96-well plates. A K/Bax transfectant which overexpresses the Bax- pRSC-Lac Z gene was determined by measuring the levels of active b-galacotosidase expression using the b-Gal Assay Kit.

Determination of DNA damage ± alkaline unwinding assay Cells in fresh culture medium were treated with 10 mg/ml etoposide. Cells were collected at each time point. The extent of bulk DNA breakage was assessed by the enhanced ¯uorescence alkaline unwinding method. The assay is based on the dierential binding and ¯uorescence of the indicator, Hoechst 33258 to single-stranded and double-stranded DNA after a ®xed period of alkaline denaturation. DNA ¯uorescence was determined by a TD-700 ¯uorometer (Turner Design, Sunnyvate, CA, USA) with excitation at 350 nm and emission detection at 450 nm. The degree of damaged DNA was expressed by the reduction in the ratio of duplex DNA to total DNA (Kanter and Schwartz 1982; Jia et al., 2001). Assessment of apoptosis by flow cytometry To induce apoptosis in intact cells, leukaemic cells (56105/ ml) were continually exposed to etoposide for up to 72 h. Cells were permeabilized with 70% ethanol and stained with 100 mg/ml PI. PI ¯uorescence of nuclei was measured by ¯ow cytometry (FACScan, Becton Dickinson, Oxford, UK) (Jia et al., 1997). DCm detection Cells were suspended in culture medium and incubated with 80 nM DiOC6(3) for 15 min at 378C followed by analysis on a FACScan ¯ow cytometer (Jia et al., 1999). Measurement of cytochrome c oxidase (COX or Complex IV) activity

Figure 7 Etoposide-induced apoptosis in Apaf-1 transfected K562 and K/Bax cells. K562 and K/Bax cells were transiently transfected with Apaf-1 plasmid DNA. Cells were washed and resuspended in fresh medium after transfection for 24 h. Cells were exposed to 10 mg/ml etoposide for up to 72 h. DNA contents were analysed using ¯ow cytometry. Data shown are means+s.d. from three independent experiments. Signi®cant dierence (*P50.0001; **P50.001) was analysed by ANOVA for non-transfected K562 cells

Oxygen consumption was used to determine the activity of COX. Cells (107) were suspended in a small volume of respiratory medium consisting of 250 mM sucrose, 0.1% BSA, 10 mM MgCl2, 20 mM HEPES, 5 mM KH2PO4 (pH 7.2) and 1 mM ADP and kept on ice. Oxygen consumption of K562 and K/Bax cells was measured with a Rank oxygen electrode (Rank Brothers, Cambridge, UK) in a thermojacketed sample chamber stirred with a magnetic ¯ea. Cells were suspended in 1 ml respiratory medium in the oxygen electrode and permeabilized by addition of 0.005% digitonin to permit free entry of mitochondrial inhibitor and

Figure 8 Etoposide-induced activation of caspases in Apaf-1 transfected cells. K562 and K/Bax cells were transfected with Apaf-1 plasmid DNA for 24 h and then treated with 10 mg/ml of etoposide. Cytosol was prepared at time 0, 24 and 48 h. Caspase-2 (a), caspase-3 (b) and caspase-9 (c) activities were assessed using ¯uorogenic substrates Oncogene


substrates. After addition of antimycin A (0.2 mM) for 2 min, the electron donors of COX, TMPD/ascorbate (0.4 mM/ 4 mM) were added to the chamber and the reading was continued for 3 min. The respiratory rate in the presence of electron donors TMPD/ascorbate was used as the activity of COX. Enzyme activity was expressed in ng atoms O2/min/107 cells (Jia et al., 1996). Preparation of cellular fractions Leukaemic cells (56107) were washed in Ca2+/Mg2+-free PBS and suspended in 1 ml of Buer A (250 mM sucrose, 10 mM HEPES ± KOH, pH 7.4, 10 mM KCl, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 0.1 mM PMSF, 20 mM cytochalasin B) and incubated for 20 min on ice. Cells were then broken with a glass Dounce homogenizer (Jencons, Leighton Buzzard, UK). Nuclei were separated by spinning at 790 g for 10 min at 48C, and the nuclear pellet was further washed with Buer A twice. The post-nuclear supernatant was further spun at 10 000 g for 10 min at 48C. The crude mitochondrial pellet was puri®ed by passing through a gradient sucrose (0.1 ± 0.3 M) cushion at 9000 g for 8 min to purify mitochondria (Jia et al., 1999). S-100 fraction was obtained by further ultracentrifugation at 100 000 g for 1 h. Protein concentration was measured using the Bradford reagent. Immunofluorescence analysis of the intracellular distribution of Bax To analyse the intracellular localization of Bax, intact cells were ®rst labelled with the mitochondrion-speci®c dye, MitoTrackerTM red CMXRos. Cells in culture medium were incubated with Mito-Tracker (100 nM) at 378C for 30 min. Cells were washed twice with Ca2+/Mg2+-free PBS and resuspended in 10% FCS containing culture medium. Fifty ml of cell suspension (106 cells/ml) was laid onto a microscope slide. Slides were air-dried, permeabilized and ®xed in acetone/methanol 1 : 1 (v/v) solution for 15 min. Cells were then washed twice in 0.1% Tween-20 containing PBS (PBST) and incubated in a blocking solution (1% BSA, 1% normal goat serum, and 0.1% Tween-20 in PBS) for 30 min. After washing once, cells were incubated with the monoclonal anti-Bax antibody (1 : 50 dilution) for 1 h at room temperature in a humidi®ed chamber. Cells were washed in PBST then incubated with FITC-conjugated antimouse secondary antibody (Sigma) at a 1 : 100 dilution in

blocking solution for 1 h in the dark. Cells were rinsed three times in PBST. Slides were air dried at 48C in the dark and viewed under a Zeiss Axioskop ¯uorescence microscope (Zeiss, Germany) attached to a CCD camera (Photometric Ltd. AZ, USA.) driven by IPLLabs Spectrum and SmartCapture (Cambridge, UK) software. The ®lter wheel was set at Texas red (excitation 540 ± 580 nm/emission 600 ± 660 nm), ¯uorescein (excitation 465 ± 495 nm/emission 515 ± 555 nm), and DAPI (excitation 310 ± 380 nm/emission 435 ± 485 nm).

4825

Measurement of the activity of caspases S-100 (50 mg protein) was diluted to 95 ml with Buer A. The reaction was initiated by the addition of 5 ml of 400 mM (®nal concentration was 20 mM) ¯uorescent substrates, Z-VDVADAFC for caspase-2, Ac-DEVD-AFC for caspase-3 and AcLEHD-AFC for caspase-9. After incubation at 308C for 15 min, the reaction was stopped by the addition of 50 ml of 1% sodium acetate trihydrate in 175 mM acetic acid and cooled on ice. After dilution to 1 ml with water, ¯uorescence at 400/505 nm for AFC release by activated caspases was measured with a TD-700 ¯uorometer. Measurements were calibrated against a standard linear regression curve of AFC. Caspase activity was de®ned as mM AFC release per mg protein per hour (mM/h/mg protein) (Jia et al., 2001). Western blottings Proteins from whole cells, puri®ed mitochondria or puri®ed nuclei were subjected to standard SDS ± PAGE at 20 ± 40 mA/gel and transfered onto PVDF membrane (Sigma) by semi-dry transfer at 15 V for 20 min. PVDF membrane was blocked with 5% milk in PBST for 1 h and probed for various proteins using monoclonal or polyclonal antibodies (as described individually in the ®gure legends). Bound antibodies were detected using appropriate HRP-conjugated secondary antibodies, followed by detection using SuperSignal ECL. The density of each band was analysed using an AlphalmagerTM 2000 Densitometer (Alpha Innotech Corp. San Jose, CA, USA).

Acknowledgments This project was supported by Leukaemia Research Fund (9946) to SM Kelsey and L Jia.

References Adams JM and Cory S. (1998). Science, 281, 1322 ± 1326. Amarante-Mendes GP, Naekyung Kim C, Liu L, Huang Y, Perkins CL, Green DR and Bhalla K. (1998). Blood, 91, 1700 ± 1705. Buckbinder L, Talbott R, Velasco-Miguel S, Takenaka I, Faha B, Seizinger BR and Kley N. (1995). Nature, 377, 646 ± 649. Deng Y and Wu X. (2000). Proc. Natl. Acad. Sci. USA, 97, 12050 ± 12055. Evan G and Littlewood T. (1998). Science, 281, 1317 ± 1322. Fang G, Kim CN, Perkins CL, Ramadevi N, Winton E, Wittmann S and Bhalla KN. (2000). Blood, 96, 2246 ± 2253. Fearnhead HO, Rodriguez J, Govek EE, Guo W, Kobayashi R, Hannon G and Lazebnik YA. (1998). Proc. Natl. Acad. Sci. USA, 95, 13664 ± 13669.

Fu WN, Kelsey SM, Newland AC and Jia L. (2001). Biochem. Biophys. Res. Commun., 282, 268 ± 272. Fukumi S, Horiguchi-Yamada J, Nakada S, Nagai M, Ohno T and Yamada H. (2000). Mol. Cell. Biochem., 206, 43 ± 50. Hsu YT, Wolter KG and Youle RJ. (1997). Proc. Natl. Acad. Sci. USA, 94, 3668 ± 3672. Jia L, Kelsey SM, Grahn MF and Newland AC. (1996). Blood, 87, 2401 ± 2410. Jia L, Allen PD, Macey MG, Grahn MF, Newland AC and Kelsey SM. (1997). Br. J. Haematol., 98, 686 ± 698. Jia L, Macey MG, Yin Y, Newland AC and Kelsey SM. (1999). Blood, 93, 2353 ± 2359. Jia L, Srinivasula MS, Liu FT, Newland AC, FernandesAlnemri T, Alnemri ES and Kelsey SM. (2001). Blood, 98, 414 ± 421. Oncogene


4826

Oncogene

Kanter PM and Schwartz HS. (1982). Mol. Pharmacol., 22, 145 ± 151. Kluck RM, Bossy-Wetzel E, Green DR and Newmeyer DD. (1997). Science, 275, 1132 ± 1136. Landowski TH, Shain KH, Oshiro MM, Buyuksal I, Painter JS and Dalton WS. (1999). Blood, 94, 265 ± 274. Li P, Nijhawan D, Budihardjo I, Srinvasula SM, Ahmad M, Alnemri ES and Wang X. (1997). Cell, 91, 479 ± 489. Liu X, Kim CN, Yang J, Jemmerson R and Wang X. (1996). Cell, 86, 147 ± 157. Lubbert M, Miller CW, Crawford L and Koeer HP. (1988). J. Exp. Med., 167, 873 ± 886. Martins LM, Mesner PW, Kottke TJ, Basi GS, Sinha S, Tung JS, Svinge PA, Madden BJ, Takahashi A, McCormick DJ, Earnshaw WC and Kaufmann SH. (1997). Blood, 90, 4283 ± 4296. Miyashita T and Reed JC. (1995). Cell, 80, 293 ± 299. Perkins C, Kim CN, Fang G and Bhalla KN. (2000). Blood, 95, 1014 ± 1022. Prokocimer M, Shaklai M, Bassat HB, Wolf D, Gold®nger N and Rotter V. (1986). Blood, 68, 113 ± 118.

Rosse T, Olivier R, Monney L, Rager M, Conus S, Fellay I, Jansen B and Borner C. (1998). Nature, 391, 496 ± 499. Schuler M, Bossy-Wetzel E, Goldstein JC, Fitzgerald P and Green DR. (2000). J. Biol. Chem., 275, 7337 ± 7342. Slee EA, Harte MT, Kluck RM, Wolf BB, Casiano CA, Newmeyer DD, Wang HG, Reed JC, Nicholson DW, Alnemri ES, Green DR and Martin SJ. (1999). J. Cell Biol., 144, 281 ± 292. Svingen PA, Karp JE, Krajewski S, Mesner PW, Gore SD, Burke PJ, Reed JC, Lazebnik YA and Kaufmann SH. (2000). Blood, 96, 3922 ± 3931. Wolter KG, Hsu Y-T, Smith CL, Nechushtan A, Xi XG and Youle RJ. (1997). J. Cell. Biol., 139, 1281 ± 1292. Yang J, Liu X, Bhalla K, Kim CN, Ibrado AM, Cai J, Peng TI, Jones DP and Wang X. (1997). Science, 275, 1129 ± 1132. Zhan Q, Fan S, Bae I, Guillouf C, Liebermann DA, O'Connor PM and Fornace AJ. (1994). Oncogene, 9, 3743 ± 3751.