Search

In medulloblastomas (MBs), the expression and activity of RE1-silencing transcription factor (REST) is increased in tumors driven by the sonic hedgehog (SHH) pathway, specifically the SHH-α (children 3 to 16 years) and SHH-β (infants) subgroups. Neuronal maturation is greater in SHH-β than SHH-α tumors, but both correlate with poor overall patient survival. We studied the contribution of REST to MB using a transgenic mouse model ( RESTTG ) wherein conditional NeuroD2 -controlled REST transgene expression in lineage-committed Ptch1 +/− cerebellar granule neuron progenitors (CGNPs) accelerated tumorigenesis and increased penetrance and infiltrative disease. This model revealed a neuronal maturation context–specific antagonistic interplay between the transcriptional repressor REST and the activator GLI1 at Ptch1 . Expression of Arrb1 , which encodes β-arrestin1 (a GLI1 inhibitor), was substantially reduced in proliferating and, to a lesser extent, lineage-committed RESTTG cells compared with wild-type proliferating CGNPs. Lineage-committed REST TG cells also had decreased GLI1 activity and increased histone H3K9 methylation at the Ptch1 locus, which correlated with premature silencing of Ptch1 . These cells also had decreased expression of Pten , which encodes a negative regulator of the kinase AKT. Expression of PTCH1 and GLI1 were less, and ARRB1 was somewhat greater, in patient SHH-β than SHH-α MBs, whereas that of PTEN was similarly lower in both subtypes than in others. Inhibition of histone modifiers or AKT reduced proliferation and induced apoptosis, respectively, in cultured REST-high MB cells. Our findings linking REST to differentiation-specific chromatin remodeling, PTCH1 silencing, and AKT activation in MB tissues reveal potential subgroup-specific therapeutic targets for MB patients.

Medulloblastoma (MB) is a malignant pediatric brain tumor characterized by poor neuronal differentiation. The RE1 Silencing Transcription Factor (REST), which negatively regulates neurogenesis by promoting repressive chromatin remodeling via epigenetic modifications, is overexpressed in human MBs. Analyses of a publically available gene expression array data set revealed association of elevated REST expression in patients with Sonic Hedgehog (SHH) medulloblastoma with poor prognosis. Upon closer evaluation, we identified a subset of highly undifferentiated SHHα tumors with elevated REST expression and activity as determined by expression of REST target genes. These patients had an overall survival rate of 50% at 5 years. To understand the underlying mechanisms, we generated a novel mouse model expressing human REST transgene in a conditional manner in granule neural progenitors (GNPs), the cells of origin of SHH driven MB. Mice with constitutive activation of Shh signaling and REST elevation (Ptch(+/-)/REST(TG)) in their GNPs developed tumors with a significant decrease in latency and increase in penetrance. We observed that 100% of Ptch(+/-)/REST(TG) animals developed heterogeneously proliferative and undifferentiated tumors compared to tumor-bearing Ptch(+/-) animals. Mechanistic studies revealed deregulation of the Shh signaling pathway by REST-dependent G9a activity, accompanied by loss of Ptch1 heterozygosity in Ptch(+/-)/REST(TG) tumors. Thus, our studies suggest that REST contributes to mis-regulation of SHH activity and drives an aggressive disease course by deregulating proliferation and differentiation. Ours is also the first study to implicate G9a, a repressive chromatin remodeler, in medulloblastoma pathogenesis and as a potential therapeutic target.

Defective Fas signaling leads to resistance to various anticancer therapies. Presence of potential inhibitors of Fas which could block Fas signaling can explain cancer cells resistance to apoptosis. We identified promyelocytic leukemia protein (PML) as a Fas-interacting protein using mass spectrometry analysis. The function of PML is blocked by its dominant-negative form PML–retinoic acid receptor α (PMLRARα). We found PMLRARα interaction with Fas in acute promyelocytic leukemia (APL)–derived cells and APL primary cells, and PML-Fas complexes in normal tissues. Binding of PMLRARα to Fas was mapped to the B-box domain of PML moiety and death domain of Fas. PMLRARα blockage of Fas apoptosis was demonstrated in U937/PR9 cells, human APL cells and transgenic mouse APL cells, in which PMLRARα recruited c-FLIPL/S and excluded procaspase 8 from Fas death signaling complex. PMLRARα expression in mice protected the mice against a lethal dose of agonistic anti-Fas antibody (P < .001) and the protected tissues contained Fas-PMLRARα-cFLIP complexes. Taken together, PMLRARα binds to Fas and blocks Fas-mediated apoptosis in APL by forming an apoptotic inhibitory complex with c-FLIP. The presence of PML-Fas complexes across different tissues implicates that PML functions in apoptosis regulation and tumor suppression are mediated by direct interaction with Fas.

We previously reported that testicular zinc finger protein (TZF) is a corepressor for androgen receptor (AR). The present study demonstrated that a central portion (amino acids 512–663) of TZF, TZF(512–663), is responsible for both binding to AR and repressing the transactivation. TZF recruited endogenous histone deacetylase 2 (HDAC2) and formed a complex with agonist-bound AR. Imaging analyses showed that TZF and TZF(512–663) were recruited by AR and simultaneously impaired distinct AR foci formation. Quantification of the foci number using a three-dimensional imaging method revealed that the number of intranuclear AR foci was related to its transactivation activity. Moreover, increased levels of TZF dissociated a coactivator, TIF2, from the AR foci and vice versa. These results indicate that the ligand-dependent transactivation function of AR is quantitatively related to its intranuclear foci formation, and suggest that corepressors, such as TZF, act on these intranuclear events competitively with coactivators.

Steroid hormones control the transcriptional activity of target genes mediated by intracellular nuclear receptors, and these transcriptional activities are modulated by the combination with coactivators and corepressors. We found in this study that testicular zinc finger protein (TZF) that was a nuclear protein with a zinc finger motif of the Cys2-His2 type was a novel corepressor of androgen receptor (AR). Fusion protein with green fluorescence protein GFP formed the specific foci in nuclei and TZF-dependent foci were located close to the splicing factor compartment. In addition, TZF was recruited into AR subnuclear foci after the treatment of dihydrotestosterone. Furthermore, we revealed that TZF bound to the activation function-1 (AF-1) domain (N-terminal transactivating domain) of AR protein. Transient over-expression of TZF in COS-7 cells or LNCaP human prostatic cancer cell resulted in decreased AR activity in a ligand-dependent fashion. Moreover, a transcriptional corepressor N-CoR additively decreased the transcriptional activity of AR with TZF. These findings suggest that TZF might be a novel corepressor of AR.

Resistance to Fas-mediated apoptosis is associated with poor cancer outcomes and chemoresistance. To elucidate potential mechanisms of defective Fas signaling, we screened primary lymphoma cell extracts for Fas-associated proteins that would have the potential to regulate Fas signaling. An activation-resistant Fas complex selectively included nucleolin. We confirmed the presence of nucleolin-Fas complexes in B-cell lymphoma cells and primary tissues, and the absence of such complexes in B-lymphocytes from healthy donors. RNA-binding domain 4 and the glycine/arginine-rich domain of nucleolin were essential for its association with Fas. Nucleolin colocalized with Fas on the surface of B-cell lymphoma cells. Nucleolin knockdown sensitized BJAB cells to Fas ligand (FasL)-induced and Fas agonistic antibody-induced apoptosis through enhanced binding, suggesting that nucleolin blocks the FasL–Fas interaction. Mice transfected with nucleolin were protected from the lethal effects of agonistic anti-mouse Fas antibody (Jo2) and had lower rates of hepatocyte apoptosis, compared with vector and a non-Fas-binding mutant of nucleolin. Our results show that cell surface nucleolin binds Fas, inhibits ligand binding, and thus prevents induction of Fas-mediated apoptosis in B-cell lymphomas and may serve as a new therapeutic target.

The epidermal growth factor receptor (EGFR) family of receptor tyrosine kinases, also known as ErbB or HER, plays crucial roles in the development of multicellular organisms. Mutations and over-expression of the ErbB receptors have been implicated in a variety of human cancers. It is widely thought that the ErbB receptors are located in the plasma membrane, and that ligand binding to the monomeric form of the receptors induces its dimeric form for activation. However, it still remains controversial whether prior to ligand binding the receptors exist as monomers or dimers on the cell surface. Using bimolecular fluorescence complementation (BiFC) assays in the present study, we demonstrate that in the absence of bound ligand, all the ErbB family members have preformed, yet inactive, homo- and heterodimers on the cell surface, except for ErbB3 homodimers and heterodimers with cleavable ErbB4, which exist primarily in the nucleus. BiFC assays of the dimerization have also suggested that the ligand-independent dimerization of the ErbB receptors occurs in the endoplasmic reticulum (ER) before newly synthesized receptor molecules reach the cell surface. Based on BiFC and mammalian two-hybrid assays, it is apparent that the intracellular domains of the receptors are responsible for the spontaneous dimer formation. These provide new insights into an understanding of transmembrane signal transduction mediated by the ErbB family members, and are relevant to the development of anti-cancer drugs.

We previously demonstrated that testicular zinc-finger protein (TZF) was a corepressor of the androgen receptor (AR). In the present study, we further showed that TZF-L, an alternative spliced variant of TZF, enhanced transactivation function of AR. Deletion analysis of TZF-L revealed that its N-terminus, which almost corresponded to that of TZF, but not its C-terminus was able to interact with AR. Additional analysis suggested that TZF and TZF-L were able to form both homodimers and heterodimers. TZF-L inhibited the homodimer formation of TZF and the intranuclear dot formation of TZF. We propose that in the unique regulation system of AR-mediated transactivation, two spliced isoforms of TZF act as coactivator and corepressor, respectively.

Abstract 2406 The death receptor Fas has a key role in mediating homeostasis, elimination of defective cells and more recently promotion of cancer. Many effective anti-cancer therapies depend on Fas-mediated apoptosis to eradicate tumor cells and ineffective Fas- apoptotic signaling is a basis for primary as well as acquired resistance to chemotherapy. We hypothesized that Fas is subjected to direct regulation and inhibition of Fas attained by cancer cells and may explain the emergence of chemoresistance. To screen for potential binding modulators of Fas, we analyzed lymphoma cells for Fas binding proteins. We first purified Fas associated proteins by using activating CH-11 antibody bound to intact BJAB cells. After immunoprecipitation, any remaining Fas, considered activation–resistant, was subjected to the second immunoprecipitation with Fas antibody B-10 followed by liquid chromatography and tandem mass spectroscopy. This purification scheme identified high scoring peptides derived from nucleolin, a nuclear protein known to be overexpressed in cancer. Nucleolin is selectively expressed on the surfaces of cancer cells and blood vessels undergoing angiogenesis. In a cell culture system, we confirmed binding of nucleolin to Fas and the presence of nucleolin-Fas complexes on the surface of lymphoma cells by surface biotin labeling. Using deletion mutants of nucleolin, we identified RNA binding domain 4 and glycine/arginine rich region to be required for the binding to Fas. BJAB cells with partially knockdown (KD) nucleolin showed significantly higher rates of apoptosis in response to stimulation with CH-11 and FasL when compared to nontarget KD controls. Importantly, the lower levels of nucleolin in knockdown cells did not affect total and surface Fas expression. Nucleolin present on the cell surface prevented binding of FasL and CH-11 to the receptor and thus provides a mechanism for blocking activation of Fas apoptosis. To examine the role of nucleon in vivo, we transfected mice with nucleolin-expressing plasmids using the hydrodynamic transfection method. The mice overexpressing nucleolin showed significantly higher survival rates than vector control transfected mice (P=.01) after a challenge with a lethal dose of agonistic anti-Fas antibody. We next examined the expression of nucleolin in human lymphomas. Cell lines derived from lymphomas of different histological types consistently expressed nucleolin protein. We found nucleolin expressed on the surface of cells in over 20 primary lymphoma isolates, whereas peripheral blood lymphocytes showed low or undetectable levels. Lymphoma tissue microarray staining showed a correlation between nucleolin and Ki-67 expression. Whether nucleolin expression also correlates with adverse clinical features in lymphoma is currently under evaluation. Taken together, we show here that the known cancer associated protein nucleolin is overexpressed on surface of lymphoma cells where it binds to Fas receptor and blocks Fas signaling and apoptosis. We expect that further analysis of nucleolin properties will reveal how Fas-nucleolin interaction can be targeted to enhance killing of cancer cells leading eventually to cell surface nucleolin targeting therapy. Disclosures: Fayad: Roche: Research Funding.

Abstract 1327 Objective: Chemotherapies and irradiation depend on an intact Fas signaling system to eradicate cancer cells. Defective Fas signaling is an important cause of acquired resistance to cancer therapy. If we were able to restore Fas apoptosis or sensitize cancer cells to Fas-mediated apoptosis, we could improve the efficacy of many current cancer therapies. To elucidate defects of Fas signaling in cancer cells, we sought to identify potential modulators of Fas selectively expressed in cancers cells and target them to sensitize cancer cells to Fas-mediated apoptosis as a component of chemotherapy. Methods: Liquid chromatography tandem mass spectroscopy was used to identify Fas-associated proteins. Co-immunoprecipitation (co-IP) and Western Blot (WB) were used to detect/confirm interactions of PML and PMLRARα with Fas, and components of death-inducing signaling complex (DISC) FADD, c-FLIP, and caspase-8 cleavage in tissues from PML wild-type (WT) and knock-out (KO) mice and acute promyelocytic leukemia (APL) cells. Deletional mutagenesis was used to map protein interacting domains. PML shRNA lentivirus and As2O3 were used to downregulate PML and PMLRARα. Flow cytometry analysis of propidium iodide- and Annexin-V-stained cells was used to detect apoptosis in response to Fas stimulation. Mice transfected with PMLRARα were monitored for survival after a lethal challenge with agonistic Fas antibody Jo2 and tissues were analyzed for apoptosis by staining for cleaved caspase-3 and TUNEL. Results: The promyelocytic leukemia protein (PML) was identified as a Fas-binding protein by mass spectroscopy analysis. Using co-IP/WB analysis of tissues from PML WT and KO mice, we found PML interaction with Fas and FADD in PML WT MEF cells and liver cells but absent in KO MEF and liver cells; PML-Fas complexes were exclusively present in the membrane/cytoplasmic extracts but not in the nuclear extracts. The B-box domain of PML was found to be required for Fas binding. Knockdown of PML was associated with suppressed rates of Fas-mediated apoptosis compared to non-targeted knockdown cells; PML KO cells reconstituted with cytoplasmic PML were sensitized to Fas apoptosis. Furthermore, we found that liver cells from PML KO mouse showed impaired assembly of the Fas death-inducing signaling complex (DISC) in response to Fas activation when compared to PML WT. PML functions are known to be blocked by its dominant-negative form PMLRARα. We found PMLRARα interaction with Fas in primary human and transgenic mouse APL cells blocked Fas-mediated apoptosis. Blockage of apoptosis was mediated through PMLRARα -dependent recruitment of c-FLIPL/Sto and exclusion of procaspase-8 from the DISC. PMLRARα effects were also observed in vivo, as expression of PMLRARα protected mice against a lethal effect of agonistic anti-Fas antibody (P Conclusions: PML binds to Fas and promotes Fas-mediated apoptosis through enhancing Fas DISC formation while binding of PMLRARα to Fas blocks Fas-mediated apoptosis in APL by forming an apoptotic inhibitory complex enriched in c-FLIP. Our data suggest that PML plays a critical role in initiation of Fas signaling; in contrast, the dominant-negative mutant PMLRARα is a confirmed cancer specific inhibitor of Fas-mediated apoptosis. Thus, deficiency of PML or expression of PMLRARα can contribute to cancer development and resistance to chemotherapy. The newly discovered PML-Fas and PMLRARα -Fas complexes can be sites for modulation of apoptosis. Thus, by neutralizing the inhibitory effect of Fas-binding proteins such as PMLRARα and/or promoting positive Fas modulators such as PML, we can improve responses to chemotherapy that depend on activation of death receptors for effective elimination of cancer cells. Disclosures: No relevant conflicts of interest to declare.

Abstract 2451 Background: Altered Fas signaling leads to resistance to various anticancer therapies. The presence of at least low level expression of Fas is required for the transformed phenotype of many cells. In order for Fas to offer this complex regulation and to prevent inadvertent Fas activation, we propose that Fas signaling is subjected to tight regulation and this regulation could produce resistance to apoptosis of cancer cells. Methods: Cell lines NB4 (PMLRARα positive), HL60, U937/PR9 (PMLRARα expression driven by metalothionine promoter), or transfected HEK293 cells were collected and cell extracts were prepared for immunoprecipitation and Western blot analysis with the indicated antibodies. Cellular fractionation was performed using the nuclear/cytosol fractionation kit (BioVision, Mountain View, CA), according to the manufacturer's instructions. We expressed PMLRARα in C57B6 female mice by transfection of PMLRARα plasmid or empty-vector plasmid (12 mice per group) by tail vein inoculation using a hydrodynamic transfection method. Twenty-four hours after transfection, mice were challenged intraperitoneally with 2 μg/g mouse weight of agonistic Fas antibody Jo2, and monitored for survival. Liver tissues were stained with TUNEL assay and antibodies targeting apoptosis proteins. Differences between two groups were assessed using the 2-tailed student's t test and a P value < 0.05 was considered to be statistically significant. Results: In this study, we identified, by mass spectrometry analysis of Fas-associated proteins, the proapoptotic promyelocytic leukemia protein (PML) as a Fas-interacting protein. PML is a ubiquitously expressed tumor suppressor, which is downregulated or mutated in over 50% of all cancers. PML promotes apoptosis especially under conditions of cell stress. The function of PML is blocked by its dominant-negative form promyelocytic leukemia - retinoic acid receptor α (PMLRARα). Our focus on PMLRARα showed that it interacts with Fas in acute promyelocytic leukemia-derived cell lines and APL primary cells. Because PMLRARα is considered to be a nuclear transcription factor, we fractionated cell compartment proteins and identified PMLRAR α-Fas complexes present only in the cytoplasmic fraction. Binding of PML to Fas was mapped to the B-box domain of PML and death domain of Fas. PMLRARα blockage of apoptosis was shown in U937/PR9 and NB4 cells, in which PMLRARα recruited c-FLIPL/S and excluded procaspase 8 from the Fas death-inducing signaling complex. PMLRARα expression in mice protected the mice against a lethal dose of agonistic anti-Fas antibody (P Conclusion: Together, PMLRARα binds to Fas and blocks Fas-mediated apoptosis in APL through the accumulation of c-FLIP- Fas complexes. Furthermore, the common expression of PML-Fas complexes suggests Fas signaling is routinely regulated in the cytoplasm by PML, but susceptible to the blocking effects of PMLRARα in cancer. Disclosures: Karp: Schering Merck: Research Funding.

Abstract 2965 Poster Board II-941 The long-term expression of human herpesvirus 8 (HHV-8) K1 produces hyperplasia of lymph nodes, splenomegaly, and lymphomas in mice. The mechanism of how K1 causes hyperplasia and lymphomas. The immunoreceptor tyrosine-based activation motif (ITAM) of K1 was shown previously to activate of Akt and nuclear factor kappa B (NF-kB). However, we have recently shown that K1 suppresses Fas-mediated apoptosis through its extracellular immunoglobulin-like domain and that K1-transfected mice survive a lethal dose of agonistic anti-Fas antibody (Jo2). We thus hypothesized that development of hyperplasia and lymphomas in K1-expressing mice is driven by altered Fas signaling. Examination of mice expressing K1 via a ubiquitous promoter showed that K1 transgenic mice (n=10) had 90% developed lymphoid hyperplasia (at least 3 lymph nodes >3 mm) and 60% developed lymphomas after 18 months, while all (26) control nontransgenic mice remained free of lymph node hyperplasia, splenomegaly, and lymphoma. Some K1 mice developed liver or mesenteric tumors (4 of 10 mice). The spleens of 78% of K1 mice were enlarged at 18 months and were on average 3.5 times heavier than spleens of non-K1 transgenic control mice. Hematoxylin and eosin staining of spleen sections showed lymphocyte expansion in the periarteriolar lymphocyte sheath with disruption of normal spleen architecture. Anti-kappa and anti-lambda light chain antibodies revealed the presence of monoclonal foci in 3 out of 3 K1 mice (average 6 foci per single section of spleen), but no foci were present in 4 control non-transgenic mice. Moreover, K1 protein was expressed in approximately 10% of splenic cells after staining with anti-K1 antibody 2H5. To test the hypothesis that expression of K1 protein confers resistant to Fas-mediated apoptosis, splenic cells of 6-month-old K1 mice (n=3) and matched controls (n=3) were isolated and incubated with 50 ng/mL of agonistic anti-Fas antibody Jo2. At 12 hours of treatment, only 4±1% of splenocytes from K1 mice versus 17±2% of control splenocytes showed morphology of apoptosis (P We mapped the region that K1 uses to bind to Fas to an immunoglobulin (Ig) chain-like domain by expressing deletion mutants of K1. Overexpression of an Ig domain-containing protein CD79b competed with K1-Fas binding in a dose-dependent manner. Two 20-amino acid peptides (N251, N253) representing the Ig domain of K1 competed with K1-Fas binding in immunoprecipitation/immunoblotting analysis. The N251 and N253 peptides (100 mM) enhanced anti-Fas antibody (CH-11, 50 ng/mL)-induced apoptosis of BJAB lymphoma cells that expressed K1 but not that of vector-transfected BJAB cells. K1 splenocytes incubated with N251 and N253 showed enhanced rates of Fas-mediated apoptosis over control peptide-treated K1, and peptides did not enhance the apoptosis rates of control non-K1 expressing splenocytes. This analysis indicates a key role of K1 in lymphoid hyperplasia and lymphoma and that K1 direct binding to Fas occurs in vivo, which is effectively targeted by competing peptides to restore apoptosis. Disclosures: No relevant conflicts of interest to declare.

Abstract 3770 Poster Board III-706 Hematopoietic cancers are commonly resistant to Fas-mediated apoptosis that contributes to their resistance to chemotherapy and reversing Fas resistance has become a primary interest in treating chemotherapy-resistant hematopoietic cancers. Wide expression of Fas receptor yet the limited extent of Fas-mediated apoptosis in tissues suggest a strict regulation of Fas signaling. A recently identified tissue-specific modulator of Fas-mediated apoptosis is hepatocyte growth factor (HGF) receptor HGFR/Met. HGFR/Met binds Fas and inhibits Fas signaling at an immediate early step. However, upon binding of ligand HGF to HGFR/Met, Fas is released and Fas signaling is restored. We hypothesized that hematopoietic cancers may express similar inhibitors of Fas, that can be targeted to resensitize cancer cells to Fas and thus also to therapy. We screen lymphoma cells for for such inhibitors. B cell lymphoma-derived BJAB cells and primary patient samples were incubated with Fas agonistic antibody. Upon removal of antibody excess, cell lysates were immuno-depleted of CH-11 antibody-bound and thus activated Fas. The remaining supernatant was then precipitated a second time with anti-Fas antibody B-10 that binds to the intracellular region of Fas. Protein bands present in B-10 precipitates and absent in CH-11 precipitates were excised from silver-stained gel and analyzed by mass-spectroscopy. CD74, also termed the invariant chain of MHC II and known to mediate pro-survival signaling as a receptor for macrophage migration inhibitory factor (MIF), was found to exclusively associate with activation-resistant Fas. Intriguingly, CD74 is known to be overexpressed in 80% to 95% of hematopoietic cancers and also in some solid tumors. Using RNA interference we showed that CD74 negative cells are more sensitive to Fas-mediated apoptosis and that CD74 overexpression confers resistance to Fas both in vitro and in mouse tissues. Incubation of cells with peptides or anti-CD74 antibody LL1 disrupts the CD74-Fas inhibitory complex and sensitizes CD74-positive cells to Fas-mediated apoptosis. Analysis at the molecular level revealed that CD74 interferes with the immediate early steps in Fas signaling – binding of agonistic CH-11 antibody and subsequent clustering of Fas receptor. We anticipate that specific disruption of the CD74-Fas interaction by CD74-derived peptides or anti-CD74 antibodies will sensitize cancer cells to Fas-mediated apoptosis. This approach represents a novel strategy for modulation of cell surface death receptors to produce effective treatments for CD74-positive cancers. Disclosures: No relevant conflicts of interest to declare.

Abstract 3976 Poster Board III-912 Fas plays a critical role in cell proliferation and in the selective killing of autoreactive lymphocytes and abnormal cells, including infected cells. To explain the common expression of Fas and the resistance to the Fas killing observed in some normal and cancer cells, we have screened cells for potential regulators of the Fas death receptor. By using mass spectroscopy analysis of Fas-associated proteins, we identified a group of peptides derived from promyelocytic leukemia (PML). PML enhances pro-apoptotic signaling, while the promyelocytic leukemia–retinoic acid receptor α (PMLRARα) activates pro-survival pathways. Given these opposing functions, we tested whether PMLRARα, which typically operates in a dominant-negative manner, blocks Fas-mediated apoptosis. Co-immunoprecipitation analysis demonstrated that PMLRARα interacts with Fas in acute promyelocytic leukemia (APL)-derived NB4 cells, U937-PR9 cells and in APL primary cells. The binding of PMLRARα to Fas was mapped to the B-box domain of PMLRARα. Flow cytometry analysis of propidium iodide-stained and Annexin-V-stained cells challenged with Fas ligand (FasL) or agonistic anti-Fas antibody (CH-11) indicated that the presence of PMLRARα was associated with blocked Fas-mediated apoptosis at early and late stages. The knockdown of PMLRARα with shRNA sensitized the NB4 cells to Fas-mediated apoptosis. Expression of PMLRARα in U937-PR9 cells prevented Fas-mediated cleavage of procaspase-8 and also prevented procaspase-8 from binding to the Fas complex upon stimulation with the agonistic anti-Fas antibody (CH-11). Further analysis indicated that PMLRARα bound to FLIPL/S and forms an apoptotic inhibitory complex with Fas, which prevents Fas activation. The data suggest that tissue-specific inhibitors of Fas such as PMLRARα block Fas-mediated apoptosis and thus can contribute to cancer development. Our results may provide an explanation for the long-known role of PMLRARα and PML in the regulation of Fas signaling, which we have shown to occur by direct regulation. We have identified an attractive potential target to the regulation of apoptosis at the PMLRARα-Fas and PML-Fas interfaces. By neutralizing the effect of death receptor inhibitors such as PMLRARα and other potential inhibitors, we can improve on the success of the many chemotherapeutic treatments that depend on activation of death receptors for effective elimination of cancer cells. Disclosures: No relevant conflicts of interest to declare.

Resistance to Fas-mediated apoptosis in hematopoietic cancers interferes with the efficacy of currently available chemotherapy. Our prior studies of human herpesvirus 8 oncoprotein K1 showed that K1 binds to Fas and interferes with activation of Fas-mediated apoptotic signaling (Wang, W, et al, Blood2007; 109:5455–62). Herpesvirus proteins often mimic host proteins or their functions, we thus searched for cellular proteins associated with inactive Fas in order to identify potential regulators of Fas signaling. We identified CD74, the invariant light chain of the major histocompatibility class II complex, associated exclusively with inactive/activation resistant Fas in B-cell lymphoma-derived BJAB cell line. Interestingly, overexpression of CD74 has been previously reported in ~ 90% of hematopoietic cancers and derived cell lines (Stein, R, et al., Ciln Cancer Res.2007; 13: 5556s–63s), as well as in ~ 80% of non-small-cell lung carcinoma (Ioachim, H.L., Am J Surg Pathol.1996; 20: 64–71). The role of CD74 in carcinogenesis was thus suspected. Through siRNA suppression of CD74 expression in BJAB cells we have determined that CD74 positive cells are more resistant to agonistic antibody CH-11-induced Fas-mediated apoptosis then the CD74 siRNA transfected cells (38 ± 7.8 % vs. 54 ± 9.8 %; P = 0.045). In addition, a challenge with agonistic anti-Fas antibody showed a significant survival advantage of the mice expressing CD74 in their livers over the vector-transfected mice (83% vs. 0 %; P < 0.016). We have mapped domain of CD74 required for its association with Fas and for significant protection of mice to a membrane-proximal extracellular region of CD74. Treatment of BJAB cells for 24 hours with humanized anti-CD74 antibody (hLL1 + crosslinking antibody), FasL-Fc, or anti-Fas antibody CH-11 induced apoptosis in 19%, 25% and 13% of cells, respectively. Combination of hLL1+crosslinking antibody with FasL-Fc or CH-11 increased apoptosis to ~ 45% of cells in both cases (P < 0.01), while combination of FasL-Fc or CH-11 with hLL1 + irrelevant antibody had no significant effect on the extent of apoptosis. Our results support the idea of an endogenous regulatory system of Fas-mediated apoptosis that utilizes transmembrane proteins interacting with Fas. We anticipate that specific blocking of the CD74-Fas interaction will sensitize CD74 overexpressing cancer cells to Fas-mediated apoptosis and thus will lead to a more effective chemotherapy for hematopoietic cancers.

Background: Expression of Human herpesvirus 8 (HHV-8) K1 causes hyperplasia of lymph nodes and lymphomas in mice. The exact mechanism of how K1 causes hyperplasia and lymphomas in K1 expressing mice is not known. The cytoplasmic immunoreceptor tyrosine-based activation motif (ITAM) of K1 was shown previously to be involved in activation of Nuclear Factor kappa B (NF-k-B). Moreover, we had shown recently that K1 suppresses Fas-mediated apoptosis through its extracellular immunoglobulin-like domain and that K1-transfected mice survived a lethal dose of agonistic anti-Fas antibody (Jo2). We thus hypothesized that development of hyperplasia and lymphomas in K1-expressing mice is driven by alterations of Fas signaling. Results: At 18 months, 10 K1 transgenic mice, 90% developed lymphoid hyperplasia (>3mm) and 60% developed lymphomas, while all (26) control mice remained hyperplasia and lymphoma free. In the extreme cases, K1 mice developed liver or mesenteric tumors (4 and 4 of 10 mice, respectively). Spleens of 78% of K1 mice were enlarged at 18 months and were on average 3.5 times heavier than spleens of non-expressing control mice (332 ± 200 mg vs. 94 ± 26 mg, P < 0.03). The H and E staining of spleen sections showed expansion to the periarteriolar lymphocyte sheath with disruption of normal follicular architecture. Staining of spleen sections with anti-kappa and anti-lambda light chain antibodies revealed presence of monoclonal foci in 3 out of 3 K1 mice (average 6 foci per single section of spleen), but none in the 4 control mice. Moreover, K1 protein was expressed in about 10% of splenic cells as judged from staining with anti-K1 antibody 2H5. To test the hypothesis that expression of K1 protein in spleens makes them resistant to Fas-mediated apoptosis, splenic cells of 6 month old K1 mice (n=3) and matched controls (n=3) were isolated and incubated with 50 ng/mL of agonistic anti-Fas antibody Jo2. At 12 hours of treatment, only 4 ± 1% of splenocytes from K1 mice versus 17 ± 2% of control splenocytes were undergoing apoptosis (P Conclusion: These results confirm that K1 is associated with lymphoid hyperplasia and lymphoma and provide plausible explanation. K1 blocks Fas-mediated apoptosis and competing peptides can reinstate apoptosis.

Background: Human herpesvirus 8 (HHV-8) infection is causally associated with the development of primary effusion lymphoma and Kaposi s sarcoma. A transmembrane protein of HHV-8, K1, is readily expressed in those tumors and the expression of K1 alone causes hyperplasia of lymph nodes and lymphomas in mice. The exact mechanism of how K1 causes hyperplasia and lymphomas in K1 expressing mice is not known. The cytoplasmic immunoreceptor tyrosine-based activation motif (ITAM) of K1 was shown previously to be involved in activation of Nuclear Factor kappa B (NF -B). Moreover, we had shown recently that K1 suppresses Fas-mediated apoptosis through its extracellular immunoglobulin-like domain and that K1-transfected mice survived a lethal dose of agonistic anti-Fas antibody (Jo2). We thus hypothesized that development of hyperplasia and lymphomas in K1-expressing mice is driven by alterations of Fas signaling. Results: Gross examination of thoracic and abdominal cavities of transgenic mice with K1 expression driven by a ubiquitous promoter were sacrificed at 18 months of age revealed enlarged cervical, mediastinal, renal, mesenteric lymph nodes and spleens. Peyer s patches were also enlarged and readily visible on the outer surface of the ileum. Of 10 K1 mice, 90% developed lymphoid hyperplasia (>3mm) and 60% developed lymphomas, while all (26) control mice remained hyperplasia and lymphoma free. In the extreme cases, K1 mice developed liver or mesenteric tumors (4 and 4 of 10 mice, respectively). Spleens of 78% of K1 mice were enlarged at 18 months and were on average 3.5 times heavier than spleens of non-expressing control mice (332 200 mg vs. 94 26 mg, P < 0.03). The H and E staining of spleen sections showed expansion to the periarteriolar lymphocyte sheath with disruption of normal follicular architecture. Staining of spleen sections with antikappa and anti-lambda light chain antibodies revealed presence of monoclonal foci in 3 out of 3 K1 mice (average 6 foci per single section of spleen), but none in the 4 control mice. Moreover, K1 protein was expressed in about 10% of splenic cells as judged from staining with anti-K1 antibody 2H5. To test the hypothesis that expression of K1 protein in spleens makes them resistant to Fas-mediated apoptosis, splenic cells of 6 month old K1 mice (n=3) and matched controls (n=3) were isolated and incubated with 50 ng/mL of agonistic anti-Fas antibody Jo2. At 12 hours of treatment, only 4 1% of splenocytes from K1 mice versus 17 2% of control splenocytes were undergoing apoptosis (P Conclusion: These results over all confirm that K1 is associated with lymphoid hyperplasia and lymphoma and provide plausible explanation of this phenomenon. Interference of K1 with Fas-mediated apoptosis disrupts the normal life cycle of lymphocytes.

Abstract 3148 Fas plays a critical role in cell proliferation and in the selective killing of autoreactive lymphocytes and abnormal cells, including infected cells. To explain the common expression of Fas and the resistance to the Fas-induced apoptosis observed in some normal and cancer cells, we screened cells for potential regulators of the Fas death receptor. By using mass spectroscopy analysis of Fas-associated proteins, we identified peptides derived from promyelocytic leukemia (PML). PML enhances pro-apoptotic signaling, while its dominant negative form, promyelocytic leukemia–retinoic acid receptor α (PMLRARα) fusion protein, activates pro-survival pathways. Given these opposing functions, we tested whether PMLRARα blocks Fas-mediated apoptosis. Co-immunoprecipitation analysis demonstrated that PMLRARα interacts with Fas in acute promyelocytic leukemia (APL)-derived NB4 cells, U937/PR9 cells and APL primary cells isolated from patients. The PMLRARα-Fas binding was mapped to the PML B-box domain of PMLRARα and death domain of Fas. Flow cytometry analysis of propidium iodide- and Annexin V-stained cells challenged with Fas ligand (FasL) or agonistic anti-Fas antibody CH-11 indicated that PMLRARα blocks Fas-mediated apoptosis at early and late stages. In line with this finding, knockdown of PMLRARα with shRNA sensitized the NB4 cells to Fas-mediated apoptosis. Detailed analysis showed that expression of PMLRARα prevents procaspase-8 from binding to the Fas complex upon stimulation with the agonistic anti-Fas antibody (CH-11) and thus, also prevents cleavage/activation of procaspase-8. Further analysis indicated that PMLRARα recruits caspase-8 inhibitor c-FLIPL/S to Fas to suppress Fas signaling. A significantly higher number of mice transfected with PMLRARα-expressing plasmid than mice transfected with empty vector survived the treatment with the mouse agonistic anti-Fas antibody Jo2 (11 of 12 vs. 0 of 12; P < 0.001). Livers from PMLRARα-transfected mice contained fewer cleaved caspase-3 positive/apoptotic cells when compared with vector-transfected mice. These data suggest that PMLRARα is a cancer specific Fas-binding inhibitor of Fas-mediated apoptosis and thus, can contribute to cancer development and resistance to therapy. Our results may provide an explanation for the long-known role of PMLRARα and PML in the regulation of Fas signaling, which, as we have shown, can occur by regulation via direct interaction. The newly-discovered PMLRARα-Fas and PML-Fas complexes can be sites for modulation of apoptossis. By neutralizing the effect of death receptor inhibitors such as PMLRARα, we can improve responses to many chemotherapeutic treatments that depend on activation of death receptors for effective elimination of cancer cells. Disclosures: No relevant conflicts of interest to declare.

Background Although significant progress has been made in the treatment of lymphomas, many lymphomas exhibit resistance to cell death, suggesting a defective Fas signaling, which remains poorly understood. We previously reported that cells expressing the K1 protein of human herpesvirus 8 (HHV-8) resist death through the complex formation of the Ig-like domain of K1 with Fas. Recently, we investigated whether peptides derived from the Ig-like domain of the K1 protein may affect cell death. Methods K1 positive and negative cell lines were incubated with the K1-derived peptides, and cell death (apoptotic and necrotic) was assessed by flow cytometry and LDH assay. Activation of caspases was assessed by fluorometric assay and flow cytometry. Fas receptor-independent, peptide-mediated cell killing was tested in the Fas-resistant Daudi cell line and Jurkat cell clones deficient in caspase-8 and FADD functionality. Activation of TNF receptors I and II was blocked by pre-incubation with corresponding blocking antibodies. The effect of the K1 peptide in vivo was tested in a mouse xenograft model. Results We observed that the peptide S20-3 enhanced cell death in K1-positive BJAB cells and HHV-8 positive primary effusion lymphoma (PEL) cell lines. Similar effects of this peptide were observed in B-cell lymphoma and T-lymphoblastic leukemia cells without K1 expression but not in normal human peripheral blood mononuclear cells. A single intratumoral injection of the S20-3 peptide decreased the growth of Jurkat xenografts in SCID mice. The mechanism of tumor cell death induced by the S20-3 peptide was associated with activation of caspases, but this activity was only partially inhibited by the pan-caspase inhibitor z-VAD. Furthermore, the K1 peptide also killed Fas-resistant Daudi cells, and this killing effect was inhibited by pre-incubation of cells with antibodies blocking TNFRI. Conclusion Taken together, these findings indicate that the S20-3 peptide can selectively induce the death of malignant hematological cell lines by Fas- and/or TNFRI-dependent mechanisms, suggesting the K1-derived peptide or peptidomimetic may have promising therapeutic potential for the treatment of hematological cancers.