Chem. Res. Toxicol. 2004, 17, 119-128
119
Depolarization of Mitochondria and Activation of Caspases Are Common Features of Arsenic(III)-Induced Apoptosis in Myelogenic and Lymphatic Cell Lines Markus T. Rojewski,*,† Sixten Ko¨rper,† Eckhard Thiel, and Hubert Schrezenmeier† Freie Universita¨ t Berlin, Universita¨ tsklinikum Benjamin Franklin, Medizinische Klinik III, Ha¨ matologie, Onkologie und Transfusionsmedizin, D-12203 Berlin, Germany Received May 27, 2003
The clinical efficacy of arsenic(III) oxide (As2O3) has been shown in patients with relapsed acute promyelocytic leukemia (APL). To identify potential common primary targets of action of As2O3 in myelogenic and lymphatic cell lines, we analyzed As2O3 effects on caspases and on the mitochondrial membrane potential (ΨM) under uniform conditions. As2O3 induced breakdown of ΨM and activated caspases in cell lines with different sensitivities for As2O3, including cell lines resistant to mitoxantron or camptothecin but sensitive to As2O3. Caspase inhibitors could not prevent breakdown of ΨM in lymphoid cell lines, whereas activation of caspases and apoptosis could be inhibited. Activation of caspases seems to be a downstream effect occurring after breakdown of ΨM. We could show that all of these effects are independent of MDR-1 expression. There was no difference in the mode of action of As2O3 in cell lines sensitive or resistant to camptothecin, mitoxantrone, or doxorubicin. As2O3 deserves further evaluation as an adjunct or alternative to other cytostatic drugs.
Introduction During the last 5 years, several groups demonstrated remission rates of >72% for arsenic(III) oxide (As2O3) treatment in relapsed or refractory acute promyelocytic leukemia (APL) (1-7). As2O3 was established as a potent alternative in therapy of all trans retinoic acid (ATRA) resistant APL and shows three common cellular effects with ATRA in APL cells: induction of differentiation and proliferation inhibition (8), induction of apoptosis (8-14), and degradation of the APL specific fusion protein PMLRARR by PIC-1/SUMO-1 (15-18). However, PML-RARRindependent effects on leukemia and lymphoma cell lines were also demonstrated (9, 13, 19-24). Recently, we could show that As2O3 exerts apoptosis in 21 and proliferation inhibition in 12 different myelogenic and lymphatic cell lines lacking the APL specific translocation (14). Obviously, there is no need for the PML-RARR fusion protein in As2O3-induced apoptosis, but a common mechanism for As2O3-mediated cellular effects seems likely, as the potency of As2O3 to induce programmed cell death could also be observed in cell lines derived from human multiple myeloma (25, 26), neuroblastoma (27, 28), colon carcinoma (29), hepatocarcinoma (30), kidney and bladder carcinoma (31), cervix carcinoma (32), and gastric cancer (33, 34) as well as in murine, bovine, swine, and rat primary cells or cell lines (12, 3539). Several different potential targets of As2O3 action have been described so far. No common pathway of * To whom correspondence should be addressed. Tel: +7 31-15 06 33. Fax: +7 31-15 05 00. E-mail:
[email protected]. † Present address: Universita ¨ tsklinikum Ulm, Abteilung Transfusionsmedizin und Institut fu¨r klinische Transfusionsmedizin und Immungenetik, D-89081 Ulm, Germany.
apoptosis induction could be proposed, as experiments were performed using different cell lines or tissues of different species. The potential key effects in As2O3induced apoptosis discussed so far for non-APL cell lines of the lymphohematopoietic system were activation of caspases (25); change in the equilibrium of Bcl-2, Bax, and Bad (23); Ca2+-dependent production of peroxide (19); and opening of the permeability transition pore (20, 21). However, the experimental approaches used for detection of involvement of mitochondria and activation of caspases differed (9-11, 19, 27, 29, 40-46). In this work, we address the question of whether a common primary target exists for As2O3-induced apoptosis in myelogenic and lymphatic cell lines derived from the lymphohematopoietic system. Therefore, caspase activation and breakdown of mitochondrial membrane potential ΨM were analyzed under unique conditions for a variety of different myelogenic and lymphatic cell lines whose sensitivity toward As2O3-induced apoptosis was characterized in detail in a previous report (14). We selected cell lines representing different sensitivity classes and cell lines resistant to cytostatic drugs, in order to identify potential differences in caspase activation and breakdown of ΨM.
Materials and Methods Cell Culture. Cell lines 697, CCRF-CEM, HL-60, Jurkat, K-562, KG-1a, and LOUCY were purchased from The German Collection of Microorganisms and Cell Cultures (DSMZ, Braunschweig, Germany); and CEM/C1, CEM/C2, HL-60/MX1, and HL-60/MX2 were purchased from the American Type Culture Collection (Manassas, U.S.A.). The doxorubicin resistant cell lines K-562(0.02) and K-562(0.1) were described previously (14). CEM/C1, CEM/C2, HL-60/MX1, HL-60-MX2, K-562(0.02), and K-562(0.1) are multiple drug resistant derivatives of the cell
10.1021/tx034104+ CCC: $27.50 © 2004 American Chemical Society Published on Web 12/24/2003
120
Chem. Res. Toxicol., Vol. 17, No. 1, 2004
lines CEM, HL-60, and K-562, respectively. Whereas KG-1a, K-562, and its derivatives show classical MDR-1 expression (data not shown), neither of the other cell lines analyzed expressed MDR-1. Cell lines CEM/C1 and CEM/C2 were selected for resistance to camptothecin and show 31- and 970-fold less sensitivity to this drug (as compared to their parental cell line), respectively (47, 48). Cross-resistance to topotecan, 9-aminocamptothecin, and 10,11-methylenedioxy-camptothecin was shown for CEM/C1; CEM/C2 additionally showed cross-resistance to etoposide, deactinomycin, bleomycin, mitoxantrone, 4′(9-acridinylamino)methanesulfon-m-ansisidide, daunorubicin, and doxorubicin but is still sensitive to vincristine. Cell lines HL-60/MX1 and HL-60/MX2 were selected for resistance toward mitoxantrone and originally tolerated doses of up to 39 and 190 nM, respectively (49). These derivatives show cross-resistance to etoposide, teniposide, bisantrene, dactinomycin, 4′-(9-acridinylamino)methanesulfon-m-ansisidide, daunarubicin, and doxorubicin but still exert sensitivity toward vincristine, vinblastine, melphalan, mitomycin C, and cis-platin. None of the K-562 derivatives was tested for cross-resistance to other drugs. The cell line LOUCY was discussed to be multiple drug resistant, as the cell line was isolated from a chemotherapy resistant T-ALL patient where remission never was achieved (50). Induction of Apoptosis with Arsenic(III) Trioxide (As2O3) and Cytostatic Drugs. Apoptosis was induced with an aqueous stock solution of 1 mM As2O3 (Sigma, Deisenhofen, Germany) in PBS without Ca2+/Mg2+ (Invitrogen/GIBCO Life Technologies, Karlsruhe, Germany) or a 50 mM stock solution of etoposide (Sigma) in DMSO (Sigma). Cells were seeded according to their optimal growth conditions indicated by the provider. Annexin V-FITC and 7-Amino-actinomycin D (7-AAD) Staining. As described previously (14), the percentage of apoptotic cells was determined by two-color fluorescence using an Annexin V-FITC (BD Pharmingen, Heidelberg, Germany) and 7-AAD system (Sigma). Whenever possible, fluorescence data of 50 000 cells were acquired using fluorescence channels FL-1 and FL-3 of a FACScan (Becton Dickinson, Heidelberg, Germany). MitoTrackerRed CMXRos Staining. MitoTrackerRed CMXRos (Molecular Probes, Leiden, The Netherlands) is a fluorescent dye that is enriched in mitochondria in a membrane potential-dependent manner and reacts with mitochondrial proteins (51-53). Therefore, MitoTrackerRed CMXRos can be used to detect cells with a loss of ΨM (54, 55). The CMX group binds to SH groups of mitochondrial proteins and is retained in living cells, while unbound dye can be washed out of the cells. Staining was performed in 1 mL of cell suspension (ca. 5 × 105 to 1 × 106 cells) in the presence of 200 nM MitoTrackerRed CMXRos. After 45 min, cells were washed twice in PBS without Ca2+/Mg2+ and the change of fluorescence intensity was measured by flow cytometry. To compare the percentage of apoptotic cells (as measured by 7-AAD and Annexin V positivity in flow cytometry) and the percentage of cells with reduced ΨΜ, the following presentation was chosen for MitoTrackerRed CMXRos stainings: the fluorescence intensity of untreated control cells was set to approximately 102 as described by the manufacturer. The percentage of cells with a reduced staining pattern for MitoTrackerRed CMXRos (as compared to the untreated controls) was referred to as MitoTrackerRed CMXRos negative. The percentage of the population with reduced fluorescence intensity (negative population) is shown in the diagrams, as these cells did not accumulate the dye in their mitochondria due to their reduced ΨΜ. Detection of Activated Caspases by Synthetic Substrates: FITC-VAD-Fmk, FITC-YVADAPK-Dnp, and PhiPhiLuxG2D2 Staining. Activation of caspases was determined by flow cytometry after staining cells with activated caspases using the CaspACE assay system (FITC-VAD-Fmk, from Promega, Mannheim, Germany) and the FITC-coupled synthetic peptide FITC-YVADAPK-Dnp (BACHEM Biochemica GmbH, Heidelberg, Germany). Staining with both fluorescent substrates was performed as described in the manual for
Rojewski et al. CaspACE. Both FITC-YVADAPK-Dnp and FITC-VAD-Fmk were used at a concentration of 10 µM. The caspase-3 specific substrate PhiPhiLuxG2D2 (peptide sequence GDEVDGI, OncoImmunin/MoBiTec, Go¨ttingen, Germany) is a nonfluorescent agent that shows fluorescence [λ(absorption) ) 580 nm, λmax(emission) ) 580 nm)] upon cleavage by caspase-3. The detection of cells with activated caspase-3 was performed as described in the manual. Incubation with Caspase Inhibitors. Cells were washed in PBS without Ca2+/Mg2+ and resuspended at a density of (24) × 105 cells/mL in serum free QBSF51 medium (Sigma). Caspase inhibitors Boc-Asp(OMe)-Fmk and Z-D(OMe)-E(OMe)VD(OMe)-Fmk (both from ALEXIS Biochemicals, Gru¨nberg, Germany) were dissolved in DMSO and used in a final concentration of 50 µM. CCRF-CEM and Jurkat cells were incubated (37 °C, 6% CO2, 98% humidity) in the presence of the inhibitors for 1 h prior to cultivation with As2O3. Detection of Cytoplasmatic Proteins by Antibodies. Cells (5 × 105 to 1 × 106) were washed with PBS without Ca2+/ Mg2+, fixed, and permeabilized for staining by using the IntrastainKit from DAKO (Hamburg, Germany). Detection of the cytoplasmatic protein was performed by flow cytometry analysis after incubation of fixed and permeabilized cells with directly labeled polyclonal rabbit anti-human activated caspase3-PE antibody (BD Pharmingen), polyclonal rabbit anti-human PARP cleavage site (214/215) (BIOSource, Ratingen, Germany), or the mAbs mouse IgG1,κ anti-human D4-GDI-FITC (clone 97A1015, IMGENEX/MoBiTec), or the corresponding isotype controls (BD Pharmingen).
Results Recently, we demonstrated induction of apoptosis in 22 different cell lines derived from the lymphohematopoietic system by low concentrations of As2O3 (0.1-5 µM) in long-term incubation experiments for up to 35 days (14). In the following experiments, we have chosen slightly higher concentrations of As2O3 (5 and 10 µM) since we were interested in early apoptotic events taking place at the membrane, mitochondria, and in the cytoplasm, which should be evaluated in short-term incubation experiments. As2O3 Induces Breakdown of Mitochondrial Membrane Potential (ΨM) and Apoptosis. To investigate the effects of As2O3 on cell lines, several parameters were analyzed. On the basis of the results of our previous report (14), we have chosen cell lines that represent different As2O3 sensitivity groups: LOUCY, 697, Jurkat, and KG-1a cells were incubated for 2, 6, 12, and 24 h with PBS, 10 µM As2O3 (5 µM As2O3 for cell line LOUCY), or 75 µM etoposide, respectively (see Figure 1 for 6 and 24 h of incubation time). The amount of apoptotic cells was determined by staining with the two markers for early apoptosis Annexin V-FITC and 7-AAD. In parallel, the presence of intact ΨM was detected by staining these cell lines with the dye MitoTrackerRed CMXRos. To achieve comparable data of apoptotic cells and cells with changes in ΨM, the amount of cells positive for Annexin V-FITC and 7-AAD staining but negative for MitoTrackerRed CMXRos staining are shown in the diagrams of Figure 1. All three methods showed comparable results. The amount of Annexin V-FITC positive, 7-AAD positive, and MitoTrackerRed CMXRos negative cells was similar for each assay. A population of MitoTrackerRed CMXRos negative cells appeared in LOUCY, 697, and Jurkat after 12 h of incubation and in KG-1a after 24 h of incubation with 10 µM As2O3. Activation of Caspases Is Involved during As2O3Induced Apoptosis. As an early event of apoptosis,
Effects of As2O3 in Lymphohematopoietic Cells
Figure 1. Treatment of cell lines with As2O3 results in a breakdown of ΨM. Cell lines were incubated with 5 µM As2O3 (LOUCY), 10 µM As2O3 (697, Jurkat, KG-1a), or 75 µM etoposide (black bars, control for apoptosis induction) for 6 and 24 h, respectively. The percentage of apoptotic cells was measured by flow cytometry after 7-AAD staining and binding of Annexin V-FITC. In addition, the percentage of cells with intact ΨM was determined by staining with the mitochondria specific, ΨMdependent dye MitoTrackerRed CMXRos. The graph shows the percentage of Annexin V-FITC (A) and 7-AAD (7) positive (i.e., apoptotic cells) and MitoTrackerRed CMXRos (M) negative cells (i.e., cells with breakdown of ΨM).
caspases are activated. We used several different assays to analyze activation of caspases. The detection of
Chem. Res. Toxicol., Vol. 17, No. 1, 2004 121
activated caspases was performed by addition of the pancaspase substrate FITC-VAD-Fmk (Figure 2). This fluorescent-modified peptide sequence binds irreversibly to most activated caspases. The activation of the central effector caspases caspase-1 and caspase-3 was analyzed using the specific modified peptides FITC-YVADAPKDnp and PhiPhiLuxG2D2, respectively. The fluorescent substrate FITC-YVADAPK-Dnp binds irreversibly to caspase-1 and to a lower extent to caspase-4, whereas PhiPhiLuxG2D2 is a nonfluorescent, caspase-3 specific substrate, which changes its characteristics of fluorescence after enzymatic cleavage by caspase-3. In addition to these functional assays for caspase activity, the activation of caspase-3 was analyzed by cleavage site specific antibodies for activated caspase-3, the highly caspase-3 specific substrate D4-GDI, and the caspase-3 specific substrate PARP (data not shown). Experiments were performed with LOUCY, 697, Jurkat, and KG-1a cells, incubated for 2, 6, 12, and 24 h with PBS, 10 µM As2O3 (5 µM As2O3 for cell line LOUCY), or 75 µM etoposide, respectively. Figure 2 summerizes the results for 6 and 24 h of incubation. After 6 h of incubation with As2O3 or etoposide, cleavage of the caspase-3 specific substrate PhiPhiLuxG2D2 and staining with FITC-VAD-Fmk were detectible in 697 cells; binding to FITC-YVADAPK-Dnp did not occur before 12 h of treatment. For cell lines Jurkat and LOUCY, PhiPhiLuxG2D2 cleavage and FITC-VAD-Fmk binding could be observed after 12 h for As2O3 and after 12 h for etoposide in LOUCY cells and 6 h for etoposide in Jurkat cells. FITC-YVADAPK-Dnp staining could not be shown before 12 h of incubation with As2O3 or etoposide in LOUCY cells and 24 h of incubation with As2O3 in Jurkat and KG-1a cells, respectively. In KG-1a cells, PhiPhiLuxG2D2 cleavage and FITC-VAD-Fmk binding could be detected after 24 h of incubation with As2O3 or etoposide in a low amount of cells (
