Oncogene (2001) 20, 4629 ± 4639 ã 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00 www.nature.com/onc Constitutive expression of ectopic c-Myc delays glucocorticoid-evoked apoptosis of human leukemic CEM-C7 cells Rheem D Medh*,1, Aixia Wang1, Feng Zhou2 and E Brad Thompson*,1 1Department of Human Biological Chemistry and Genetics, The University of Texas Medical Branch, Galveston, Texas, TX-77555-0645, USA Sensitivity to glucocorticoid (GC)-evoked apoptosis in roles in normal thymocyte selection and in the therapy lymphoid cell lines correlates closely with GC-mediated of lymphoid malignancies (Gaynon and Carrel, 1999; suppression of c-Myc expression. To establish a Leung and Munck, 1975). GC-evoked lympholytic functional role for c-Myc in GC-mediated apoptosis, responses are mediated via the GC receptor (GR), we have stably expressed MycERTM, the human c-Myc and can be blocked by the GR antagoinst RU 38486 protein fused to the modi®ed ligand-binding domain of (Homo-Delarche, 1984; Thompson et al., 1995). the murine estrogen receptor a, in GC-sensitive CEM- Ligand-activated GR is known to function as a C7-14 cells. In CEM-C7-14 cells, MycERTM constitu- transcription factor for numerous genes; hence the tively imparts c-Myc functions. Cells expressing My- current hypothesis for GC-evoked apoptosis favors cERTM (C7-MycERTM) exhibited a marked reduction in receptor mediated alteration in expression of key cell death after 72 h in 100 nM dexamethasone (Dex), vitality and/or death genes (Medh and Thompson, with 10 ± 20-fold more viable cells when compared to the 2000; Montague and Cidlowski, 1995; Thompson, parental CEM-C7-14 clone. General GC responsiveness 1999). GR-mediated transcriptional activation (Dowd was not compromised, as evidenced by Dex-mediated et al., 1991; Zhou and Thompson, 1996) and/or suppression of endogenous c-Myc and cyclin D3, and repression (Eastman-Reks and Vedeckis, 1986; Helm- induction of c-Jun and the glucocorticoid receptor. berg et al., 1995; Yuh and Thompson, 1989), as well as MycERTM also blunted Dex-mediated upregulation of post-transcriptional regulation may play a role in GC- p27kipI and suppression of the Myc target p53. In evoked apoptosis of thymocytes and leukemic cells. comparison to parental CEM-C7-14 cells, Dex-evoked The nuclear phosphoprotein c-Myc has been im- DNA strand breaks were negligible and caspase plicated as an important mediator of GC-evoked activation was delayed, but the extent of G1 cell cycle lympholytic responses. Cell cycle progression through arrest was similar in C7-MycERTM cells. Myc-ERTM did the G1 - S boundary requires c-Myc, and deregulated not result in permanent, complete resistance to GC expression or activation of c-Myc by chromosomal however, and the GC-treated cells eventually died, translocation or gene ampli®cation is known to be a indicative of redundant or interactive mechanisms in major causative factor for several human cancers, the GC-evoked lytic response of lymphoid cells. Our including lymphomas, lung carcinomas and neuroblas- results emphasize the importance of c-Myc suppression tomas (Bishop, 1983; Spencer and Groudine, 1991). in GC-evoked apoptosis of CEM-C7-14 cells. Oncogene Both the c-Myc protein and c-myc mRNA are highly (2001) 20, 4629 ± 4639. unstable, each with a half-life of approximately 30 min (Rabbitts et al., 1985), facilitating their tight regulation Keywords: cell cycle; caspase 3; TUNEL; Max; p53; and re¯ecting c-Myc's importance in controlling the p27 proliferative state of the cell. In performance of many of its actions, c-Myc heterodimerizes with a homo- logous protein, Max, and the heterodimer activates Introduction transcription of genes containing the E-box consensus sequence CAC G/A TG (Blackwood and Eisenman, Glucocorticoids (GCs) kill immature thymocytes and 1991). Max-independent actions of c-Myc may involve sensitive malignant cells of lymphoid origin via complex interactions with other coregulatory proteins classical apoptosis; consequently GCs have important (Peukert et al., 1997; Philipp et al., 1994; Sakamuro and Prendergast, 1999). Some of the growth promoting genes that are positively regulated by Myc-Max *Correspondence: RD Medh or EB Thompson; heterodimers include ornithine decarboxylase (odc; E-mail: [email protected] or [email protected] (Tobias et al., 1995), cdc25A, the cdk 2 and 4 2Current address: National Flow Cytometry Resource, Life Science activating cell cycle phosphatase gene (Galaktionov et Division, Los Alamos National Laboratory, Los Alamos, New al., 1996), the translation initiation factor eIF-4E Mexico 87545, USA Received 12 January 2001; revised 23 May 2001; accepted 31 May (Jones et al., 1996) and the transcription factor p53 2001 (Facchini and Penn, 1998; Reisman et al., 1993). c-Myc delays glucocorticoid-evoked apoptosis RD Medh et al 4630 Several reports over the past few years have Jun, repression of cyclin D3, and G1 growth arrest are demonstrated the dual role of c-Myc in apoptosis not aected. Our data con®rm the hypothesis that c- (reviewed in (Thompson, 1998)). Depending on the cell Myc suppression is important for GC-evoked apoptosis type and conditions being studied, either deregulated of CEM cells, and further assess the limits of c-Myc's overexpression or suppression of c-Myc can lead to role in the process. apoptosis (Shi et al., 1992; Thompson, 1998). Based on reports where growth factor deprivation induced apoptosis is potentiated by enforced ectopic c-Myc Results expression (Askew et al., 1991; Evan et al., 1992), it has been proposed that c-Myc has a dual function of We have been studying the mechanism of GC-evoked inducing cell proliferation as well as triggering leukemic cell apoptosis using clones derived from the apoptosis. In contrast to cell death provoked by human acute lymphoblastic leukemic cell line CCRF- overexpression of c-Myc, in several lymphoid cell CEM. When GC-sensitive clonal CEM-C7 cells are systems, apoptosis is preceded by a severe down- treated with 100 nM dexamethasone (Dex), approxi- regulation of c-myc message and c-Myc protein mately 90% cells are killed by 72 h and by 96 h expression (Thompson, 1998). In normal mouse essentially all cells are dead (Medh et al., 1998). In the thymocytes, mouse S49 and human CEM and Jurkat studies presented here, to ensure karyotypic and cells, GC-evoked, GR-dependent apoptosis is asso- phenotypic purity, we have used a fresh subclone, ciated with c-myc downregulation (Eastman-Reks and CEM-C7-14, that replicates the GC response of the Vedeckis, 1986; Helmberg et al., 1995; Martins and original CEM-C7 clone. Aguas, 1998; Yuh and Thompson, 1989). Only those CEM clones that apoptosed in response to GC Establishment of CEM-C7-14 cells stably expressing exhibited such downregulation (Thulasi et al., 1993; ectopic c-Myc Yuh and Thompson, 1989), further showing the relevance of this repression. GC resistance of one CEM-C7-14 cells were transfected with the construct CEM clone can be overcome by simultaneous activa- pBpuroMycERTM which contains the human c-Myc tion of the cAMP and GC pathways, with a protein fused to the modi®ed ligand binding domain concomitant restoration of c-Myc downregulation (LBD) of the murine estrogen receptor a (ERa) (Medh et al., 1998). In contrast to normal mouse (Littlewood et al., 1995). This modi®ed LBD contains thymocytes, those from non-obese diabetic (NOD) a point mutation (G525R) that is theoretically modi®ed mice that are resistant to GC-induced apoptosis exhibit to render it incapable of binding to estrogen, while an increase in c-myc mRNA and c-Myc protein levels retaining normal anity for the synthetic ligand 4- in response to GC treatment (Casteels et al., 1998). hydroxy tamoxifen (4HT). After selection in puromy- Transient expression of transfected c-myc in GC- cin, the cells were termed C7-MycERTM. These were sensitive CEM cells appeared to protect them against tested for expression of the fusion protein. Parallel steroid-evoked death (Thulasi et al., 1993). immunoblots probed with the c-Myc monoclonal The above results have established a strong correla- antibody Myc1-9E-10.2 and sc-544, a monoclonal tion between GC-evoked apoptosis and suppression of antibody raised against a C-terminal portion of ERa, c-Myc in CEM leukemic lymphoblasts; however more recognized this protein in mass cultures of C7- direct evidence has not been forthcoming. Further- MycERTM cells (Figure 1a). Its larger size makes it more, a recent report suggests that GC-evoked clearly distinguishable from endogenous c-Myc. Several apoptosis of a subclone of GC-sensitive CEM-C7 cells clones of C7-MycERTM cells expressing moderate levels is not aected by expression of a tetracycline of MycERTM were isolated (Figure 1b). After a 24 h regulatable c-Myc construct (Loer et al., 1999), treatment with Dex, there is a considerable reduction contradicting several earlier observations. In murine of endogenous c-Myc in both CEM-C7-14 and C7- P1798 cells, overexpression of c-Myc alone is not MycERTM cells, but not of MycERTM (Figure 1c). The sucient to protect from apoptosis, but coexpression presence of MycERTM protein despite Dex treatment of c-Myc and cyclin D3 complement each other and suggested that there might be sustained c-Myc activity protect (Rhee et al., 1995). To better demonstrate the in Dex-treated cells. In correlation with their lytic role of c-Myc in GC-evoked apoptosis of CEM cells, response, CEM-C7-14 cells treated for 24 h with Dex and to gain further insight into the mechanisms exhibit approximately 80% downregulation of c-Myc controlling apoptosis that are modulated by c-Myc, immunoreactive protein (Figure 1c). we have stably expressed a chimeric protein, MycERTM, in CEM-C7 cells. We demonstrate here that MycERTM MycERTM expression protects CEM-C7-14 cells from imparts c-Myc functionality and conveys signi®cant Dex-evoked death protection from GC-evoked apoptosis, delaying cell death by 24 ± 48 h.