Overexpression of Mcl-1 in Anaplastic Large Cell Lymphoma Cell Lines and Tumors

To the Editor-in-Chief:

Systemic anaplastic large cell lymphoma (ALCL) as outlined in the recent World Health Organization classification 1 includes a subset of tumors that carry t(2;5)(p23;q35) resulting in overexpression of anaplastic lymphoma kinase (ALK). Although the pathogenesis of ALCL is unknown, recent in vitro studies provide evidence that several mechanisms including the ras, PLC-γ, and PI-3-kinase pathways may possibly be involved in deregulation of cell proliferation and apoptosis. 2

In an article published previously in The American Journal of Pathology, we showed that BCL-2 expression, as detected immunohistochemically, is absent in ALK+ ALCL and correlates with a relatively higher apoptotic rate in these tumors, thus providing a possible biological explanation for the superior prognosis reported for patients with ALK+ ALCL. 3 In addition, we reported that other members of the BCL-2 family are differentially expressed in ALK+ and ALK− ALCL. The pro-apoptotic proteins BAX and BCL-XS are present at relatively higher levels and the anti-apoptotic protein BCL-XL is present at relatively lower levels in the ALK+ ALCL group. 3 However, other mechanisms may promote tumor cell survival in ALCL.

The myeloid cell leukemia-1 (Mcl-1) protein is an antiapoptotic member of the BCL-2 family. 4,5 Recently, it has been shown that approximately 85% of mice expressing a Mcl-1 transgene develop a variety of B cell lymphomas after a relatively long follow-up of 2 years. 6 In view of its implication in lymphomagenesis, we investigated Mcl-1 expression in systemic ALCL and correlated our findings with apoptotic rate, ALK expression, clinical and laboratory features at time of diagnosis, and outcome in previously untreated patients.

We initially analyzed Mcl-1 expression in five ALK+ ALCL cell lines using Western blotting methods and a polyclonal antibody (DAKO, Carpinteria, CA). The panel of ALK+ ALCL cell lines included Karpas 299 (a gift from Dr. M. Kadin, Boston, MA), SR-786 and SU-DHL-1 (both from DSMZ, Braunschweig, Germany), and JB-6 and TS-G1 (gifts from Dr. D. Jones, Houston, TX). All five cell lines have been shown to carry t(2;5) and overexpress ALK. Western blot analysis detected a 40/42-kd protein product corresponding to Mcl-1 in all ALCL cell lines. JB-6 and TS-G1 cells expressed Mcl-1 at relatively lower levels compared with the other three cell lines.

We also evaluated Mcl-1 expression using immunohistochemical methods in a series of previously untreated patients with systemic ALCL diagnosed at the University of Texas M.D. Anderson Cancer Center between 1984 and 2000. For this purpose, we used a previously constructed tissue microarray that included triplicate tumor cores from 50 lymphomas including 12 ALK+ ALCL, 26 ALK− ALCL, 7 primary cutaneous ALCL and 5 diffuse large B cell lymphomas with anaplastic cytologic features, as well as 2 reactive lymph nodes. For the detection of Mcl-1, we used immunohistochemical methods and a polyclonal antibody (DAKO). The terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end labeling (TUNEL) assay was applied to full tissue sections for the evaluation of apoptotic rate as described previously. 3

As described by others, Mcl-1 expression in reactive lymph nodes was restricted to germinal center cells, which served as a positive control and also allowed for comparison of staining intensity. 7 Expression of Mcl-1 is reciprocal to that of BCL-2 in normal lymph nodes. The latter is expressed in mantle zone cells and is negative in germinal center cells. Using a 10% cutoff, Mcl-1 was detected in all 12 (100%) ALK+ ALCL and in 16 of 26 (61.5%) ALK− ALCL with a granular cytoplasmic staining pattern (Figure 1a) ▶ and this difference was statistically significant (P = 0.016, Fisher’s exact test). The percentage of Mcl-1-positive tumor cells also differed significantly between ALK+ and ALK− ALCL. Most tumor cells were uniformly and strongly positive in all ALK+ ALCL. By contrast, only a subset of tumor cells in ALK− ALCL were positive for Mcl-1 (Figure 1b) ▶ and 11 (42%) cases of ALK− ALCL showed weak staining intensity. Mcl-1 expression was inversely correlated with BCL-2 expression in 34 tumors with BCL-2 results. Fifteen of 18 (83.3%) BCL-2-negative ALCL were Mcl-1-positive whereas 8 of 16 (50%) BCL-2-positive tumors were Mcl-1-positive (P = 0.04, Fisher’s exact test). However, apoptotic rate did not differ significantly between Mcl-1-positive and Mcl-1-negative ALCL. Mcl-1 was weakly positive in 3 of 7 (43%) cutaneous ALCL, in a subset of tumor cells (generally <25%), and in 2 of 5 (40%) diffuse large B cell lymphomas with anaplastic cytologic features.

Figure 1.

a: A representative case of ALK+ ALCL with strong Mcl-1 expression in most tumor cells. (immunoperoxidase with hematoxylin counterstain, ×400). b: Double histogram demonstrating the distribution of ALK+ (black) and ALK− (gray) systemic ALCL tumors according to the percentage of Mcl-1-positive tumor cells. The x axis represents the percentage of Mcl-1-positive tumor cells and the y axis shows the percentage of tumors.

Statistical analysis did not reveal any significant association between Mcl-1 expression and clinical or laboratory features at the time of diagnosis in patients with systemic ALCL. Mcl-1 expression was not associated with progression-free survival (PFS) in this group of patients (5 year PFS for Mcl-1-positive versus -negative tumors; 72% versus 68%, P = 0.3 by log rank), although the relatively small number of patients precludes definite conclusions regarding clinical outcome.

Overexpression of the antiapoptotic protein Mcl-1 may overcome the absence of immunohistochemically detectable BCL-2 expression in ALK+ ALCL thus resulting in prolonged tumor cell survival. The mechanisms of Mcl-1 up-regulation in ALCL tumors are unknown. One signaling pathway that has been shown to be important in up-regulating Mcl-1 is the Jak/STAT pathway. 8 In a preliminary study we have shown that STAT3 is consistently activated in all ALK+ ALCL cell lines we have analyzed (unpublished data). Recently, Zamo et al 9 also demonstrated constitutive activation of STAT-3 by NPM-ALK.

We conclude that the antiapoptotic protein Mcl-1 is consistently overexpressed in ALK+ ALCL cell lines and tumors, in contrast with BCL-2, which may protect cells from apoptosis and confer survival advantages to these tumors. Further studies are needed to explore the mechanisms of Mcl-1 involvement in both apoptosis and proliferation and determine whether Mcl-1 is a potential target for therapeutic intervention in systemic ALCL.