ALBERT

Description: Abstract 2915 Bortezomib is a reversible first-generation proteasome inhibitor that inhibits the β5 and to a lesser extent the β1 catalytic site of the proteasome. However, bortezomib does not inhibit the β2 catalytic proteasomal site at clinically relevant concentrations, and bortezomib-resistance is accompanied by upregulation of the β2 subunit, suggesting that increased β2 activity may compensate for the loss of β1/ β5 activity during bortezomib-treatment. The second generation proteasome inhibitor carfilzomib, due to the chemistry of its epoxyketone warhead, has a higher substrate specificity and functions as an irreversible proteasome inhibitor, but is still a β1/ β5 inhibitor that does not affect the β2 active site. We investigated the effect of β2-specific proteasome inhibition on myeloma and acute myeloid leukemia (AML) cells and tested the hypothesis that β2-selective proteasome inhibition may overcome bortezomib-resistance. To this end we have developed a set of epoxyketone- and vinylsulfone-based, cell permeable proteasome inhibitors of which we selected the compounds PR523A and PR671A for further testing in cell-based assays. PR671A is a peptide-vinylsulfone that selectively inhibits the proteasome's β2/ β2i subunit in an irreversible fashion in human cell lines and primary cells at low micromolar concentrations without inhibition of other protease species. PR523A is a β5-selective peptide-epoxyketone with otherwise similar properties. Treatment of myeloma and AML cell lines (AMO-1, U-266, HL-60, THP-1) with PR523A induced ER-stress mediated apoptosis, very similar to bortezomib. The combination of bortezomib with PR523A led to additive, but not synergistic induction of apoptosis, as expected. Selective β2 inhibition by PR671A resulted in the induction of ER stress and the accumulation of poly-ubiquitinated protein, however, this was not effectively translated into apoptotic cell death. This indicates that selective inhibition of the β2 proteasome subunit alone has only a poor cytotoxic effect on myeloma and AML cell lines, suggesting that the function of β2 is largely redundant and can be compensated when the remaining proteasome catalytic subunits (β1 and β5) remain active. However, when the β2 inhibitor PR671A was combined with agents that target the proteasome's β5 active site (PR523A) or the β5 and the β1 site (bortezomib), the combination of either inhibitor with the β2 inhibitor PR671A was highly synergistic for both activation of ER stress and the induction of apoptotic death. Importantly, the bortezomib-resistance in bortezomib-adapted myeloma and AML cell lines could be overcome by combining PR671A with either bortezomib or PR523A, while β2 inhibition by PR671A alone had no effect on the viability of bortezomib-adapted cells. We conclude that PR671A is a β2 selective proteasome inhibitor. Selective Inhibition of the proteasome's β2 subunit has little effect on viability or ER stress both in normal and bortezomib-resistant myeloma and leukemia cells, suggesting that the function of the β2 catalytic site is largely redundant. However, when β1/ β5 proteasome activity is inhibited by drugs like bortezomib or carfilzomib, proper function of the β2 proteasome active site is crucial for cell survival, also in bortezomib-resistant myeloma cells. The use of specific β2 inhibitors like PR671A in combination with β1/ β5 inhibitors like bortezomib or carfilzomib is therefore a promising strategy to overcome resistance against β1/ β5-selective proteasome inhibitors. Disclosures: No relevant conflicts of interest to declare.

Description: Dendritic cells (DCs) play a key role in the pathogenesis of HIV infection. HIV interacts with these cells through 2 pathways in 2 temporal phases, initially via endocytosis and then via de novo replication. Here the transcriptional response of human DCs to HIV-1 was studied in these phases and at different stages of the virus replication cycle using purified HIV-1 envelope proteins, and inactivated and viable HIV-1. No differential gene expression was detected in response to envelope. However, more than 100 genes were differentially expressed in response to entry of viable and inactivated HIV-1 in the first phase. A completely different set of genes was differentially expressed in the second phase, predominantly in response to viable HIV-1, including up-regulation of immune regulation genes, whereas genes encoding lysosomal enzymes were down-regulated. Cathepsins B, C, S, and Z RNA and protein decreased, whereas cathepsin L was increased, probably reflecting a concomitant decrease in cystatin C. The net effect was markedly diminished cathepsin activity likely to result in enhanced HIV-1 survival and transfer to contacting T lymphocytes but decreased HIV-1 antigen processing and presentation to these T cells.

Description: Abstract 2169 Poster Board II-146 The tyrosine kinase inhibitors (TKIs) Imatinib mesylate (IM, Gleevec, Glivec) and nilotinib (Tasigna, AMN) are currently used in treatment of chronic myeloid leukaemia (CML). IM has been described to influence the function and differentiation of antigen presenting cells, to inhibit the effector function of T lymphocytes and to decrease the immunogenicity of CML cells by downregulation of tumor associated antigens. In the present study, we analyzed the effect of IM and AMN on proteasomal activity in IM-sensitive or IM/AMN- resistant CML cells as well as in patient samples using a biotinylated active site-directed probe, which, covalently binds and labels proteasomal subunits beta-1, beta-2 and beta-5 and their immunosubunit counterparts beta-1i, beta-2i and beta-5i, in an activity-dependent fashion. In addition, we analyzed the cleavage and processing of antigenic BCR-ABL derived peptides after degradation by the IM or AMN treated 20S immuno- and constitutive proteasome by massspectrometry. The analyzed epitopes KQSSKALQR and GFKQSSKAL are deduced from the fusion region of BCR-ABL and bind to HLA-A3/11 and HLA-B8, respectively. Both epitopes have been shown to be naturally presented by patient CML cells. So far, in vitro generation of these epitopes by proteasomal digestion experiments has been shown only for KQSSKLAQR. We found, that IM and AMN treatment in sensitive and resistant CML cells, as well as in primary patient samples and BCR-ABL-negative cells led to a concentration-dependent decrease of the MHC-class I expression in line with the decrease of proteasomal activity, indicating that the effects are BCR-ABL independent. This inhibitory effect was independent of the expression of proteasome subunits as analyzed by western blotting and it was not due to the induction of apoptosis. In vitro digestion experiments using purified proteasomes in the presence of AMN or IM showed little influence on the overall cleavage pattern observed. However, the generation of epitope-precursor peptides was significantly altered in the presence of AMN and IM for both peptides. Treatment of the immunoproteasome i20S with IM or AMN resulted in an almost complete reduction in the generation of the long precursor peptides for the HLA-A3/A11 and —B8 epitopes while the processing of the short peptide sequences significantly increased. In addition, we show for the first time that both epitopes KQSSKALQR and GFKQSSKAL can be generated as N-terminal elongated precursors in vitro by both constitutive and immuno-20S proteasomes. Interestingly, in all performed experiments AMN was more effective as compared to IM, while other TKIs such as Sunitinib, Sorafenib or the PI3K inhibitor LY294002 had no effect. Our results demonstrate that treatment with IM and AMN can affect the immunogenicity of malignant cells by affecting proteasomal degradation of cytosolic antigens, thereby modulating the repertoire of presented antigens. These strong effects of the TKIs IM and AMN on proteasomal activity might be a result of changes in the phosphorylation of proteasomal subunits akin to the recently showed endogenous phosphorylation sites of the mammalian 20S proteasome. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 4069 Background: HIV protease inhibitors (HIV-PI: Ritonavir, Lopinavir, Saquinavir, Nelfinavir, Amprenavir, Indinavir, Atazanavir, Tipranavir and Darunavir) are oral drugs approved for HIV treatment. Although designed to inhibit the HIV protease, HIV-PI likely have additional, yet unidentified, mammalian targets. Individual HIV-PI have been reported to inhibit the proteasome as well as AKT phosphorylation, induce ER stress and exert synergistic cytotoxicity with bortezomib in solid tumors, suggesting a therapeutic potential of HIV-PI especially in myeloma and AML. However, testing and comparing the effects of the different HIV-PI in AML or myloma cells, a prerequisite to identify the most promising HIV-PI to enter clinical testing, have not yet been performed. Methods: We systematically compared and analysed the effects of all currently approved HIV-PI on proteasome inhibition, cytotoxicity, induction or ER-stress, inhibition of p-AKT, and synergism with bortezomib in human myeloma and AML cell lines and primary cells in vitro. Results: HIV-PI can be divided into two subclasses, based on their activity on myeloma and AML cells: Lopinavir, Nelfinavir, Ritonavir and Saquinavir (LNRS-PI) showed biological and molecular activity at concentrations within or near to therapeutic drug levels (10-20 μ M). All remaining HIV-PI were inactive in this concentration range. The LNRS-PI induced ER stress in a very similar fashion (concentration-dependent increase in expression of the chaperones BIP and PDI at 10 μ M). This translated into a uniform pattern of ER stress-induced apoptosis, as deferred from increased expression of CHOP and cleaved PARP, leading to cell death. Likewise, the concentration-dependent inhibition of AKT-phosphorylation was similarly observed for all LNRS-PI, starting at 10 μ M. The LNRS-PI also showed a uniform synergistic cytotoxic activity with bortezomib at therapeutic drug levels, as observed with myeloma cell lines, as well as AML cell lines and primary cells in vitro. By contrast, all remaining HIV-PI did not show a synergistic effect with bortezomib. To assess the effects of the different HIV-PI on proteasome activity, we used activity-based proteasome-specific probes that visualize the activities of the proteasome β2 and β1/β5 active sites in intact, viable cells. Nelfinavir stood out as the only HIV-PI with proteasome-inhibiting activity at therapeutic drug levels. Nelfinavir led to a dose-dependent decrease in active proteasome β1/β5 as well as β2 species in the concentration range of 10–40 μ M. By contrast, low doses of Lopinavir, Ritonavir or Saquinavir induced proteasome β2 and β1/β5 activity, which then decreased to baseline activity levels upon higher drug concentrations (up to 80 μ M). All remaining HIV-PI had no effect on proteasome activity. When cells were pre-treated with bortezomib, which preferentially inhibits the proteasome β1/β5 subunits, and were then exposed to the different HIV-PI, only Nelfinavir led to an additional inhibition of the proteasomal β1/5 and also β2-activity, while all other HIV-PI did not have such an effect. Strikingly, primary myeloma cells from a patient who was refractory to bortezomib-lenalidomide combination therapy, and which were refractory to Bortezomib 10 nM also in vitro, showed robust (〉 90%) cytotoxicity when bortezomib was combined with therapeutic drug levels of Nelfinavir (10 μ M). Conclusions: Nelfinavir is a unique drug in the class of HIV-PI, which leads not only to the induction of ER stress and inhibition of AKT-phosphorylation, but also to proteasome inhibition of all active subunits in intact cells at therapeutic drug levels. It re-sensitizes bortezomib-refractory myeloma cells towards bortezomib treatment. Nelfinavir may therefore be an active drug warranting clinical testing in hematologic malignancies, such as myeloma, mantle cell lymphoma or AML. Disclosures: No relevant conflicts of interest to declare.

Description: Adaptive resistance of myeloma cells to proteasome inhibition is poorly understood. It is suggested to base on point mutations in PSMB5 and/or downmodulation of the activation state of the unfolded protein response (UPR) via reduced activity of its major regulatory axis IRE-1/XBP-1. We have generated subclones of the AMO-1 myeloma cell line resistant to bortezomib 〉 1000 nM (AMO-BTZ), or carfilzomib 〉 1000 nM (AMO-CFZ), that do or do not carry the PSMB5 A310G mutation in the β5 substrate pocket. We combine this model with a global quantitative proteomics approach, the analysis of the activation status of the IRE-1/XBP-1 pathway, and with an advanced set of proteasome activity-specific fluorescent affinity probes that allow direct, selective, simultaneous visualization of the activity of all six active β-subunits of the constitutive and the immunoproteasome. Our results demonstrate that the A310G mutation has a modest impact on β5c proteasome inhibition by bortezomib (increasing the IC50 from 25nM to 80 nM), and likewise by carfilzomib (IC50 increase from 10 nM to 50 nM). Strikingly, when AMO-CFZ or AMO-BTZ were exposed to the same functional level of proteasome inhibition (〉 90% inhibition of β5c/5i, 20% inhibition of β1/1i, β2/2i) that resulted in 70-90% cytotoxicity in AMO-1 cells, no cytotoxicity was observed in AMO-BTZ and AMO-CFZ cells. Likewise, AMO-BTZ and AMO-CFZ cells were resistant to the next generation proteasome inhibitors ixazomib, oprozomib and dalanzomib, irrespective of the presence or absence of the PSMB5 mutation. Analysis of the UPR and its major regulators on protein and mRNA levels revealed that all clones of AMO-BTZ and AMO-CFZ showed significantly lower expression of IRE-1 and its product, spliced XBP-1, compared to AMO-1 cells, in contrast to all other major regulators of the UPR (ATF6, PERK, elF2a). Proteasome inhibitor treatment induced phosphorylation of IRE-1 and the induction of sXBP1 similarly in AMO-1, AMO-BTZ and AMO-CFZ cells, however, the induction of downstream proteins of the UPR (ATF4, PDI) was exclusively found in AMO-1 cells. Mass spectrometry-based quantitative global proteomic analysis was performed to compare AMO-1 cells with AMO-BTZ and AMO-CFZ with a cut off of at least a 50% change in abundance of differentially expressed proteins in at least 2 out of triplicate experiments. This yielded 〉 3500 identified individual proteins in proteasome inhibitor adapted cells, of which 〉 600 were differentially expressed and subsequently subjected to a protein-protein interaction (PPI) search and a Gene Ontology (GO) analysis, resulting in an average of 30 GO terms for the overexpressed proteins and 10 for downregulated species in AMO-BTZ and AMO-CFZ. Manual grouping of GO into functionally related clusters resulted in 5-6 groups that were largely concordant between AMO-BTZ and AMO-CFZ. The clusters found overexpressed in AMO-BTZ and AMO-CFZ were proteins involved in protein catabolism, redox control and protein folding. Uniform downregulation was observed for protein clusters involved in transcription/translation, differentiation, apoptosis and structural/cytoskeletal functions. The quantitatively largest group of proteins with significantly altered expression levels in AMO-BTZ/AMO-CFZ vs. AMO-1 control cells consisted of proteins involved in metabolic regulation. This big cluster comprised close to 50 % of all polypeptides with significant quantitative changes, suggesting a key role for metabolic homeostasis. The quantitatively most significantly upregulated protein in both AMO-CFZ and AMO-BTZ was NADPH dehydrogenase, the most important reducing enzyme in eukaryotic cells (4-6 x upregulated). The top individual upregulated protein in AMO-CFZ was the p-glycoprotein 1 (Pgp, 12 x upregulated), while the transcription factor IKZF3 was among the top downregulated proteins in AMO-BTZ cells (0.2 x). Our data indicate that proteasome gene mutations are not required for proteasome inhibitor resistance of myeloma cells, that proteasome inhibitor adapted myeloma cells can compensate subtotal proteasome inhibition irrespective of the type of inhibitor used, and that they have undergone complex adaptive changes in particular in proteins that regulate metabolic functions. Thus we suggest that the metabolic machinery rather than the proteasome should be explored for drug targets in myeloma cells with acquired proteasome inhibitor resistance. Disclosures No relevant conflicts of interest to declare.

Description: Resistance towards proteasome inhibition by Bortezomib (Velcade®) represents a challenge for myeloma therapy. Its biology has not yet been characterized in detail. We have demonstrated that Bortezomib-sensitive malignant haematopoetic cells can acquire secondary resistance to Bortezomib in vitro. We here present the first analysis of proteasome biology and activity, alternative proteolytic pathways, ubiquitin-specific proteases (USP) and the ER stress response (unfolded protein response, UPR) upstream of the proteasome, as well as in vitro cytotoxicity of conventional cytotoxic drugs, alternative proteasome inhibitors and agents that target the UPR in Bortezomib-resistant (BR) cells, compared to wild type (WT) controls. BR cells had higher activities of all subunits of the constitutive and the immunoproteasome, as deferred from turnover of fluorogenic substrates as well as affinity-labelling of active proteasome subunits in intact cells. This was mirrored by increased levels of proteasomal β1 and β2, but especially β5 polypeptides, implicating a homeostatic system that senses and corrects low proteasome activity in cells chronically exposed to Bortezomib. While the vinylsulfone-type proteasome inhibitor NLVS abrogated detectable proteasome activity in both BR and WT cells, Bortezomib at therapeutic concentrations eliminated proteasomal β1 and β5-type activity only in WT cells, while BR cells retained residual activity. These changes in proteasome biology appear to be the molecular hallmark of required Bortezomib resistance, since no changes were observed between WT and BR cells in alternative cytosolic or lysosomal proteolytic pathways, UPR activity as well as the gross activity pattern of USP. As expected, this translated into sensitivity against cytotoxic drugs in vitro: BR cells were less sensitive towards alternative proteasome inhibitors. However, while the IC50 for pan-proteasome inhibitors was only roughly doubled in BR cells, it was nearly tenfold elevated for the β5-preferring vinylsulfone inhibitor NLVS. By contrast, sensitivity towards anthracyclines or cytotoxicity induced by ER stressors as well as the synergy between proteasome inhibitors and UPR-activators remained unaffected in BR cells. Based on our data, proteasome inhibitors with activity profiles different from that of Bortezomib, alone or in combination with induction of the UPR, may represent an appropriate concept to overcome secondary Bortezomib resistance.

Description: Background. EDO-S101 is a first-in-class alkylating, histone-deacetylase inhibitor (HDACi) fusion molecule with dual activity that is currently in Phase I. It structurally combines the strong DNA damaging effect of bendamustine with a fully functional pan-HDAC inhibitor, vorinostat. Bendamustine has substantial activity against B-cell malignancies; vorinostat sensitizes the same type of cancers against alkylators or proteasome inhibitors (PI). Bendamustine combined with the PI bortezomib (BTZ) is active against multiple myeloma (MM). Cytotoxicity of PI in MM relies on excess induction of proteotoxic stress and triggering of the unfolded protein response (UPR). Upon proteasome inhibition, HDACi synergize with PI by interfering with the a-tubulin-mediated transport of poly-ubiquitinated proteasome substrates to lysosomal destruction. Indeed, EDO-S101 has strong synergistic cytotoxicity with PI in vitro against hematological malignancies, including MM, mantle cell lymphoma and ABC type diffuse large B-cell lymphoma. The aim of this work is to characterize the molecular mechanism of action of the synergy of EDO-S101 with PI in comparison to its established structurally related drugs, bendamustine and vorinostat. Methods. The cytotoxic and molecular activity of EDO-S101 in combination with BTZ and other types of PI was assessed in vitro using the RPMI-8226 and several other MM cell lines. HDAC-inhibiting activity, accumulation of poly-ubiquitinated proteins and induction of ER stress, apoptotic signaling and autophagy induction were assessed by quantitative PCR and western blotting. Proteasome activity was measured with activity based probes (ABP). Apoptosis was assessed by AnnexinV/FITC staining with flow cytometry. Cell viability was evaluated by MTS assay. Results. EDO-S101 showed substantially stronger cytotoxicity in combination with PI than melphalan, bendamustine, cyclophosphamide or PI combined with equimolar vorinostat. EDO-S101 had higher HDACi-type of activity, compared to vorinostat, as demonstrated judged in particular by increased a-tubulin acetylation, providing a potential mechanistic basis for its superior synergy with PI. Consistent with this, EDO-S101 alone induced moderate cellular accumulation of poly-ubiquitinated proteins already in the absence of proteasome inhibition, which was potentiated when EDO-S101 was combined with BTZ. EDO-S101 induced activation of the UPR-regulators XBP1 and IRE1 known to control BTZ sensitivity of MM, in contrast to vorinostat or bendamustine alone. Co-treatment with BTZ and EDO-S101 or vorinostat resulted in highly synergistic triggering of the UPR (ATF4, CHOP, BIP). Interestingly, EDO-S101 in addition induced the pro-apoptotic machinery via upregulation of NOXA, downregulation of BCL2 and an increase of the BAX/BCL2 ratio, and also activated autophagy, as evidenced by upregulation of LC3A and LC3B. While this pro-apoptotic signaling of EDO-S101 was highly synergistic with BTZ-induced apoptotic signals, co-treatment with BTZ and vorinostat reduced apoptotic signaling compared to BTZ alone. EDO-S101 reduced c-Myc expression by 60%, while vorinostat had no effect on c-Myc levels. The combination BTZ+EDO-S101 decreased c-Myc levels by approx. 90%, while these levels remained unchanged during treatment with BTZ+vorinostat. Conclusion. EDO-S101 is a first-in-class, dual-mechanism, alkylator-HDAC-inhibitor fusion molecule that combines key structural features of bendamustine and vorinostat. The molecular mode of action of EDO-S101 differs from that of its structurally related drugs by a more effective interaction with a-tubulin, which may in part explain superior synergy with PI. Most importantly, EDO-S101 has a direct pro-apoptotic activity via downregulation of c-Myc and BCL2 while upregulating NOXA, features not observed with vorinostat. This results in highly synergistic signaling with the PI-induced pro-apoptotic effects. EDO-S101 is a promising advancement of bendamustine with molecular features clearly different from and superior to a combination of bendamustine with vorinostat. EDO-S101 should be explored in combination with proteasome inhibitors in particular in poor risk B cell neoplasms with c-Myc overexpression such as aggressive MM, Burkitt lymphoma or "double hit" aggressive B cell lymphoma. Disclosures Besse: Mundipharma-EDO: Other: travel support. Mehrling:Mundipharma-EDO: Employment. Driessen:Mundipharma-EDO: Honoraria, Membership on an entity's Board of Directors or advisory committees; celgene: Consultancy; janssen: Consultancy.

Description: Abstract 2956 Introduction: The HIV protease inhibitor nelfinavir has anti-myeloma activity in mice; it is approved at the 1250 mg bid dose for oral treatment of HIV. We performed a phase I dose escalation trial of nelfinavir in combination with bortezomib in patients with advanced hematologic malignancies. Methods: During cycle 1 (28 days), trial treatment consisted of 1 week nelfinavir monotherapy, followed by nelfinavir in combination with standard dose bortezomib (1.3 mg/m2i.v. day 8, 11, 15, 18), while cycles 2 and 3 (21 days each) consisted of 2 weeks nelfinavir in combination with bortezomib (day 1, 4, 8, 11). Non-progressing patients could continue therapy for up to 4 additional cycles with the same regimen as cycles 2 and 3. Nelfinavir dose was escalated in a 3+3 design over 3 dose levels (1250, 1875, 2500 mg bid). Dose limiting toxicity (DLT), the primary endpoint, was grade 3–4 non-hematological toxicity (excluding grade 3 bilirubin/alanine aminotransferase (ALT) or hyperlipidemia reversible within 2 weeks) or severe hematologic toxicity unrelated to the underlying disease during cycle 1. Secondary endpoints included pharmacodynamic and pharmacokinetic assessments during cycle 1 at baseline, nelfinavir monotherapy and after application of nelfinavir and bortezomib in combination, as well as signals for activity. Results: Twelve evaluable patients were registered (median age 58 years; 8 male; performance status 0–1 in 10/12 patients); 8 had multiple myeloma, 2 leukemia (1 acute myeloid, 1 acute lymphoblastic) and 2 lymphoma (1 diffuse large B-cell lymphoma, 1 mantle-cell (MCL)). All myeloma patients failed both prior bortezomib and lenalidomide-containing therapy; 7/8 had progressed under prior bortezomib. One patient (2500 mg bid dose) experienced a transient grade 4 elevated ALT, categorized as DLT, which resolved within 2 weeks. The patient continued the same regimen off study without recurrent hepatic toxicity. No further DLTs occurred, thus nelfinavir 2500 mg bid was established to be safe in combination with standard dose bortezomib. One patient with highly aggressive lymphoma died from cerebral vein thrombosis; a myeloma patient experienced a non-fatal pulmonary embolism. Elevated ALT (2 patients) was the only additional non-hematological toxicity grade 3/4 observed in 〉1 patient. Grade 3 febrile neutropenia and grade 4 thrombocytopenia were seen in 1 and 4 patients, respectively. Best treatment response was evaluated for 11 patients (1 not evaluable). Partial response was achieved in 3 patients (2 myeloma, 1 MCL) and stable disease for at least 2 cycles of therapy in 5 patients. Overall, 4/12 patients completed 〉=3 cycles of treatment. Assessment of proteasome activity in peripheral blood mononuclear cells (PBMC) from treated patients after 1 week nelfinavir monotherapy revealed inhibition of total proteasome activity in vivo by nelfinavir compared to baseline (mean inhibition, as determined by specific, quantitative intracellular affinity labeling of active proteasome subunits: 14.9 %, 95% confidence interval (CI): 8.8–23.5%, p=