MCL-1 Promoter Insertions Dial-Up Aggressiveness of Chronic Leukemia

Chronic lymphocytic leukemia (CLL) is the most common form of leukemia in the western world, accounting for approximately 12 000 new cases annually, and a prevalence of about 50 000 to 60 000 patients in the United States alone. CLL is a quintessential example of a human malignancy caused predominantly by defective apoptosis, as opposed to accelerated cell division. The malignant lymphocytes that accumulate in patients with CLL are largely quiescent, noncycling cells that display prolonged lifespans due to failed programmed cell death (1). Normally, a delicate balance of anti-apoptotic and pro-apoptotic genes keeps cell lifespan in check, functioning analogous to an electrical rheostat to carefully adjust the body's cell flux dial that ensures the daily production of approximately 50–70 billion cells in humans that is offset by a commensurate amount of cell death. Imbalances in the activities of these apoptosis-regulatory gene networks create a fertile soil in which the seeds of carcinogenesis can take root. In this issue of the Journal, Moshynska et al. (2) describe somatic mutations in the promoter region of MCL-1, an anti-apoptotic gene. The findings provide new insights into the mechanisms responsible for dysregulation of apoptosis in CLL. MCL-1 is a member of the BCL-2 gene family, which includes 25 known members in humans [see (3) for review]. Bcl-2-family proteins include both anti-apoptotic and pro-apoptotic members. Imbalances in the levels or activities of these proteins are commonly associated with malignancy, including lymphocytic malignancies such as non-Hodgkin lymphoma, in which the founding member of the family, Bcl-2, was first discovered nearly two decades ago by virtue of the involvement of its encoding gene in chromosomal translocations (4). Analogously, transgenic mice overexpressing Mcl-1 protein in lymphocytes frequently develop lymphomas (5), attesting to the in vivo importance of this anti-apoptotic protein for B-cell neoplasia. The clinical course of CLL is highly heterogeneous, with time to progression varying from months to many years. Responses to chemotherapy also vary widely. This heterogeneity has prompted searches for biomarkers that might be predictive of prognosis. Previous studies documented high levels of anti-apoptotic Bcl-2 relative to pro-apoptotic Bax protein in CLLs, with higher Bcl-2:Bax ratios often correlating with aggressive disease or poor response to therapy [see for example (6–8)]. The mechanisms responsible for the high levels of Bcl-2 in CLL are poorly understood, but gene hypomethylation has been reported (9). Chromosomal translocations occasionally activate the BCL-2 gene in CLL, but these are relevant to only a few percent of cases (10). Thus, the reasons for elevated BCL-2 expression in CLL are incompletely resolved. Also, Bcl-2:Bax ratios are insufficient to explain differences among CLL patients with respect to progression and chemoresponse. Searches for other prognostic determinants of outcome and response have uncovered Mcl-1 as a candidate predictor of more aggressive disease. Roughly one-third of CLLs contain higher levels of Mcl-1 protein, as measured by immunoblotting using peripheral blood B-cells (7,8). Although cohort sizes are small, correlations have been noted for higher Mcl-1 and failure to achieve complete remission following single agent chemotherapy (with either fludarabine or chlorambucil). In fact, in two independent studies (of n = 42 and n = 37 patients, respectively), not a single patient with higher Mcl-1 protein achieved complete remission (7,8). Thus, although larger cohorts of patients must be analyzed before firm conclusions are reached, higher Mcl-1 protein may represent an indicator of adverse outcome for patients with CLL. Interestingly, comparisons of matched pairs of untreated and relapsed specimens from acute leukemia patients have revealed elevations in Mcl-1 during progression to chemorefractory disease, further supporting a role for this anti-apoptotic protein in chemoresistance (11). The article by Moshynska et al. (2) provides new insights into the mechanisms contributing to altered regulation of MCL-1 in CLL, showing the presence of nucleotide (nt) insertions in the promoter region of the MCL-1 gene. An insertion of either 6 or 18 nucleotides was found in 17 of 58 CLL specimens analyzed but in none of 18 normal specimens. Analysis of matched pairs of normal versus leukemia cell samples from two patients suggested that the MCL-1 promoter insertions represent somatic alterations and not hereditary polymorphisms. MCL-1 promoter insertions were associated with nonresponsiveness to chemotherapy (an insertion was present in seven of the 10 patients who had no response and in none of the 12 who had a complete or partial response), with rapid disease progression (the disease progressed rapidly in eight of the 17 patients with an insertion but only five of the 41 without an insertion), and with shorter overall survival, after correcting for differences in clinical stage, but results varied widely among patients. The presence of a promoter insertion was also associated ith higher MCL-1 mRNA and Mcl-1 protein levels, showing a tendency toward higher expression but certainly not a direct cause-and-effect relation. Thus, although possibly contributing, the presence of promoter insertions is evidently insufficient to reliably drive high levels of MCL-1 expression in CLL. Of note, no functional assays were performed, such as reporter gene assays, to prove that insertions actually confer increased transcriptional activity on the MCL-1 promoter. Also, although the region involved contains potential binding sites for several transcription factors (e.g., Stat, CREB, NFκB, Sp1, Ets), it is unclear precisely how the DNA sequences inserted would affect MCL-1 expression. Interestingly, the larger 18-nt insertion adds additional putative Sp1-binding sites (GGCCCC) to the promoter. If gene insertions are an inadequate explanation, then what might account for the dysregulation of MCL-1 expression in aggressive CLLs? First, exogenous signals provided by lymphokines, cytokines, chemokines, and cell adhesion molecules have all been reported to induce MCL-1 expression in either malignant or normal B cells [see, for example (12)]. Thus, signals provided by the microenvironment might combine with gene insertions to drive aberrant MCL-1 expression. Second, Mcl-1 protein levels are regulated by post-transcriptional mechanisms, particularly inducible protein degradation (13). In this regard, Mcl-1 is one of the most labile of the Bcl-2-family proteins, containing PEST sequences that target it for rapid degradation, but the mechanisms regulating its turnover are poorly understood. If Bcl-2 levels are high in most CLL, then why is Mcl-1 important? Presumably, Mcl-1 and Bcl-2 play nonredundant roles in cell survival. Hints of individualized roles for Mcl-1 and Bcl-2 are found in the following: (a) gene ablation studies in mice, showing striking differences in the in vivo requirements for Bcl-2 and Mcl-1 for survival of specific cell lineages (14,15); (b) protein interaction studies, showing preferences for different interaction partners (16); (c) differences in intracellular locations of Bcl-2 and Mcl-1 (17); and (d) gene knock-down studies using antisense oligonucleotides, demonstrating requirement for Mcl-1 for CLL survival in various in vitro contexts, even in the face of persistent Bcl-2 expression (18). The unique requirement for Mcl-1 for survival of lymphoid cells in vivo(15) suggests that it might make a good target for treatment of lymphoid malignancies such as CLL. Strategies for targeting Mcl-1 include reducing its expression by using antisense oligonucleotides that recognize its mRNA (19) or inhibiting its activity by using kinase inhibitors such as flavopiridol; therapeutic monoclonal antibodies such as anti-CD20 (Rituximab), which have been shown to interfere with signaling events necessary for maintaining expression of Mcl-1 protein in CLL (20,21); or small molecule compounds that mimic the inhibitory BH3 domain of endogenous Mcl-1 antagonists (22). However, conditional gene ablation studies in mice indicate that Mcl-1 is also required for survival of hematopoietic progenitor cells (15), raising the specter of toxicities that might limit utility. Nevertheless, the discovery of MCL-1 promoter insertions provides further evidence that this gene plays an important role in the biology of CLL and forecasts a day when improved knowledge of the molecular mechanisms responsible for aberrant control of apoptosis regulation in CLL may provide improved prognostic information that guides individualized therapeutic strategies for optimized medical management.