c-FLIP confers resistance to FAS-mediated apoptosis in anaplastic large-cell lymphoma

Anaplastic large-cell lymphoma (ALCL)¹  frequently carries the t(2;5)(p23;q35)²  or other 2p23 locus rearrangements³  that result in overexpression of anaplastic lymphoma kinase (ALK). The t(2;5), the most common abnormality, creates a novel fusion gene, npm-alk. Among other mechanisms, NPM-ALK is believed to mediate its oncogenic potential through activation of phosphatidylinositol 3-kinase (PI3K)/Akt and STAT3 signaling pathways resulting in downstream antiapoptotic signals.^4-8 

The extrinsic apoptotic pathway is initiated by death surface receptors, a subgroup of the tumor necrosis factor receptor (TNFR) superfamily, that have a characteristic intracytoplasmic domain, designated the death domain (DD). Upon ligation, CD95 (FAS), the best characterized death surface receptor, oligomerizes at the cell membrane and recruits the adapter protein FADD (FAS-associated death domain protein) through DD-DD association. FADD, in turn, recruits procaspase-8 by its death effector domain (DED), initiating the formation of the death-inducing signaling complex (DISC). In the DISC, procaspase-8 molecules are located in close proximity and are autocatalytically processed to active caspase-8. In “type I” cells, death receptors induce strong caspase-8 activation, directly activating effector caspases, including caspase-3, to ensure apoptosis. In “type II” cells, low amounts of caspase-8 are activated, depending on a mitochondrial amplification mechanism through the cleavage of BID.^9,10 

Death-receptor-induced apoptosis is inhibited by c-FLIP, a DED-containing protein with a nonfunctional caspaselike catalytic domain with antiapoptotic function in the DISC.^11,12  The death receptor system has been involved in normal and abnormal immune processes,^13,14  as well as in Hodgkin^15,16  and non-Hodgkin lymphomagenesis,¹⁷  and it is currently attracting interest as a possible therapeutic target.^18,19 

Here we show that overexpression of c-FLIP in ALCL may confer resistance to apoptosis induced by cell surface signals.

Two ALK⁺ (Karpas 299, SU-DHL1)²⁰  cell lines known to carry the t(2;5) were used. As controls, a CD30⁺ T-cell lymphoma cell line, Mac2a (a gift from Dr K. Elenitoba-Johnson, Salt Lake City, UT), a Hodgkin lymphoma cell line, L-1236 (purchased from DSMZ, Braunschweig, Germany), and a mantle cell lymphoma cell line, Mino (a gift from Dr R. Ford, Houston, TX) were also used. The cell lines were maintained according to methods reported.²⁰  Karpas 299 and SU-DHL1 cells were treated with the PI3K inhibitor LY294002, and the JAK3 inhibitors, WHI-P131 and WHI-P154, at the indicated concentrations. Whole-cell lysates were prepared 24 or 48 hours following treatment. Karpas 299 and SU-DHL1 cells also were infected with the HA-tagged, constitutively active, myrAkt adenovirus²¹  at a multiplicity of infection (MOI) of 20, previously shown to significantly increase ^Ser473p-AKT levels in vitro (not shown).

Each cell line (1 × 10⁶ cells/mL) was incubated with increasing concentrations of the anti-CD95/FAS agonistic CH11 antibody (Beckman Coulter, Fullerton, CA) for 24 hours in 6-well plates, and a dose-response curve of cell viability was generated. A 200-ng/mL concentration of CH-11 produced approximately 60% and 50% cell death 24 hours after treatment in control Mac2a and Mino cells, respectively (“Results”). Cell viability was evaluated using trypan blue staining in triplicate. Apoptosis was assessed with annexin-V staining (BD Biosciences Pharmingen, San Diego, CA) detected by flow cytometry (FACS Calibur; Becton Dickinson, San Jose, CA) and DAPI staining detected by immunofluorescence.⁸ 

Lysates prepared from cells in log-phase growth or after apoptosis induction were collected, washed twice in cold PBS, and lysed at 4°C in lysis buffer as previously described.²⁰  Western blot analysis was performed as reported elsewhere.²⁰  The primary antibodies used were as follows: CD95/FAS (sc-715, polyclonal), CD95L/FASL (N-20, polyclonal), and c-FLIP (H-202, polyclonal) (Santa Cruz Biotechnology, Santa Cruz, CA); procaspase-8 (Ab-3, monoclonal) and cleaved caspase-8 (Asp³⁸⁴, 11G10, monoclonal) (Oncogene, La Jolla, CA); and β-actin (Sigma, St Louis, MO).

Cleaved caspase-8 (dilution, 1:50; Oncogene) was also detected using an immunofluorescence method. Briefly, cytospin cell preparations were incubated with the primary antibody overnight. Detection was achieved by incubation with a fluorescent antimouse antibody (Alexa fluor 488; Molecular Probes, Eugene, OR) at a 1:200 dilution for 1 hour. 4′,6′-diamidino-2-phenylidole (DAPI) (Molecular Probes) was used as counterstain.

The sequences of c-FLIP small interference (si) RNA targeting the human c-FLIP gene product were purchased from Ambion (Austin, TX): GGGACCUUCUGGAUAUUUUtt (sense) and AAAAUAUCCAGAAGGUCCCtg (antisense). Transient transfection of Karpas 299 and SU-DHL1 cells was performed using the Nucleofector solution “T” (Amaxa Biosystems, Gaithersburg, MD). Approximately 2 × 10⁶ cells were transfected with 0, 5, and 10 μg c-FLIP siRNA using 100 μL T solution (Amaxa Biosystems) according to the manufacturer's protocol. Cells were harvested at 48 hours following transfection and whole-cell lysates were prepared for Western blot analysis. In addition, Karpas 299 and SU-DHL1 cells (each 1 × 10⁶ cells/mL) were preincubated with 10 μg c-FLIP siRNA for 24 hours in 6-well plates and subsequently with 200 ng/mL CH11 (Beckman Coulter). Control transfections with the siRNA of 2 housekeeping genes, GAPDH and 4EBP1, were also performed.

Twenty-six ALK⁺ ALCL tumors obtained prior to therapy at The University of Texas M. D. Anderson Cancer Center between 1984 and 2003 were analyzed using immunohistochemistry after antigen retrieval²⁰  and tissue microarrays as previously described.²²  The panel of antibodies used included CD95/FAS (C-20, polyclonal; dilution, 1:200), CD95L/FASL (N-20, polyclonal; dilution, 1:200), and c-FLIP (H-202, polyclonal; dilution, 1:50) (Santa Cruz Biotechnology). Lymph nodes with reactive follicular hyperplasia served as controls. A 10% cutoff was used to define positivity of all proteins studied.

Induction of the extrinsic apoptotic pathway using an agonistic anti-CD95/FAS antibody demonstrated no cell death in both ALK⁺ ALCL cell lines (Figure 1A). In contrast, substantial apoptotic cell death was observed after treatment of Mac2a and Mino cells with increasing concentrations of CH-11. Apoptotic death of Mac2a cells after CH-11 treatment was associated with cleavage of caspase-8 and down-regulation of c-FLIP (Figure S1, available on the Blood website; see the Supplemental Figures link at the top of the online article).

These results are consistent with previously published data by Dirks et al²³  who described resistance to CD95/FAS-induced apoptosis in ALK⁺ ALCL cells. They also assessed for FAS gene mutations as a possible underlying mechanism, but they did not identify mutations in 3 ALK⁺ ALCL cell lines, including Karpas 299. Another possibility would be inhibition of DISC function through overexpression of c-FLIP. In both ALK⁺ ALCL cell lines (Karpas 299, SU-DHL1), the long (c-FLIP_L) and short (c-FLIP_S) isoforms of c-FLIP were expressed (Figure S2). Of note, the short c-FLIP isoform, which has been shown to totally prevent caspase-8 cleavage,²⁴  was expressed at a relatively higher level in ALK⁺ ALCL cells than in control cell lines.

We show here that a major inhibitor of DISC function, c-FLIP, appears to be responsible for resistance of ALK⁺ ALCL cells to death-receptor-induced apoptosis. Endogenous c-FLIP expression was completely inhibited when 10 μg siRNA was used (Figure 1B). This was associated with a slight increase in the number of apoptotic cells (Figure 1C-D) that may be due to the endogenous FASL expression (Figure S2). However, preincubation of ALK⁺

ALCL cells with c-FLIP siRNA and treatment with CH-11 significantly induced apoptosis, associated with apoptotic morphology and cleavage of caspase-8 (Figure 1E-F). Therefore, inhibition of c-FLIP expression using c-FLIP siRNA sensitized ALK⁺ ALCL cells to FAS-induced apoptosis. In line with our data, 2 recent reports demonstrated sensitization to CD95/FAS-mediated cell death in Hodgkin lymphoma cell lines using a specific c-FLIP siRNA.^15,16  However, additional factors contributing to protection of ALK⁺ ALCL cells from extrinsic apoptotic signals cannot be excluded.

Using a 10% cutoff, CD95/FAS and c-FLIP were overexpressed in 23 (92%) of 25 and 21 (91%) of 23 tumors, respectively, (Figure 2A,C). By contrast, CD95L/FASL was expressed by the neoplastic cells in only 3 (12%) of 26 ALCL tumors (Figure 2B). However, FASL was always detected at high levels in the reactive small lymphocytes and endothelial cells, suggesting that the microenvironment of these tumors is capable of producing FASL. These results concur with previous findings reported by Sigel and Hsi.²⁶  In a series of 10 ALCL cases (including 2 ALK⁺), CD95/FAS was expressed in more than 90% of tumor cells in all cases, whereas CD95/FASL was expressed only in a subset of tumor cells.

The mechanisms underlying c-FLIP up-regulation in ALK⁺ ALCL cells most likely are complex. Although, it has been suggested that c-FLIP up-regulation depends on NF-κB activation in Hodgkin lymphoma,^15,16  recent evidence shows that NF-κB is not active in ALK⁺ ALCL.²⁷  Therefore, NF-κB-independent mechanisms may be involved. One possibility is that c-FLIP is up-regulated by the PI3K/Akt pathway,²⁸  as Akt can be phosphorylated/activated by NPM-ALK.^4,5,29  Here, we show that inhibition of PI3K/Akt down-regulates c-FLIP, and, inversely, forced expression of constitutively active Akt up-regulates c-FLIP_S splice variant (Figure 2D). Another possibility is that ALK or other kinases interacting with ALK, such as JAK3, up-regulate c-FLIP. Indeed, we have shown that inhibition of JAK3 kinase activity, which has been shown to inhibit ALK activity,²⁵  also down-regulates c-FLIP (Figure 2E).

Taken together, our results suggest that ALK⁺ ALCL cells are resistant to CD95/FAS-induced cell death due to c-FLIP overexpression. Up-regulation of c-FLIP may be a downstream target of oncogenic pathways involved in ALCL pathogenesis, thus providing a novel therapeutic target in ALCL patients.

We thank Dr Francois-Xavier Claret for providing us with the myrAkt adenoviral vector.