ALBERT

Description: Key Points Recurrent mutations in FOXO1 affect the DNA binding domain and the T24 phosphorylation site, which disrupt interactions with 14-3-3. Presence of FOXO1 mutations is associated with decreased OS, particularly in DLBCL patients of the low-risk R-IPI categories.

Description: Abstract 473 Introduction : Follicular lymphoma (FL) and diffuse large B cell lymphoma (DLBCL) are the most common types of B-cell derived non-Hodgkin lymphomas (NHL). A significant proportion of patients with FL develop transformed disease reminiscent of DLBCL, which arises from the cells of FL. A common feature of FL and a subset of DLBCL is the presence of a balanced cytogenetic translocation t(14;18)(q32;q21). In addition to chromosome rearrangements, inappropriate somatic hypermutation of proto-oncogenes, including MYC, PIM1, ARHH and PAX5, have been proposed to contribute to DLBCL, but not to FL. Evidence for other mutations contributing to the development of DLBCL and FL is lacking. Our research group had previously detected a residue of the SET domain of the methyltransferase EZH2 (Y641) that is frequently mutated both in FL and the GCB subtype of DLBCL (Morin, R et al. 2010 Nature Genetics 42 (2):181-5). It has also been recently reported that specific gene expression signatures can reveal unique diagnostic and prognostic information about FL and DLBCL. The presence of mutations in genes that contribute to chromatin modification (from transcriptionally active to silent, and vice versa) can have an effect on the regulation of gene expression, playing an important role in cancer. Methods: We have deep-sequenced the whole transcriptome of DLBCL samples using Illumina second-generation sequencing to detect candidate mutations in different genes that may contribute to lymphoma development and progression. Based on our observations of these data, we subsequently PCR amplified and sequenced the entire MLL2 locus from genomic DNA isolated from FL and DLBCL tumor samples, and from matched germline DNA where available. Using an alternative approach, consisting of a targeted hybridization capture of MEF2B exons from genomic DNA, we re-sequenced the exons of MEF2B in an extended cohort of FL and DLBCL samples. Results: Among many other mutated genes we have characterized the mutations found in two chromatin modifying genes, MLL2 and MEF2B. MLL2, a gene that encodes a histone methyltransferase, is somatically mutated in 89% of FL (n=35) and 32% of DLBCL (n=37) samples. The majority of these mutations were truncations and frame-shift insertions and deletions, likely inactivating MLL2. MEF2B, which encodes a MADS/MEF2 DNA binding protein involved in the regulation of gene expression, was found mutated in 12% of FL (n=274) and 9% of DLBCL (n=321) tumor samples. In contrast with the mutational pattern of MLL2, we did not detect non-sense mutations in MEF2B. All mutations in MEF2B affected a small number of residues. Several studies in MEF2 family members have shown the importance of some of these mutated residues for the correct function of the protein. Conclusions: After high-throughput sequencing of transcriptomes in a cohort of FL and DLBCL lymphoma samples, two targeted second generation re-sequencing approaches have enabled the screening of individual genes in a large cohort of samples. We chose two of the genes with strong evidence for recurrent somatic mutations and no previously known role in lymphoma, MLL2 and MEF2B, for detailed characterization. The high incidence of mutations in both of these genes suggests that these mutations might act as driver mutations of interest for further study. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 139 Background: Diffuse large B cell lymphoma (DLBCL) and follicular lymphoma (FL) constitute 70% of all non-Hodgkin lymphomas (NHL). Both malignancies derive from germinal center (GC) B-cells and are characterized by clinical and genomic heterogeneity. Chromosomal alterations that deregulate oncogenes as well as mutations in genes involved in cell proliferation and apoptosis have been described in DLBCL. This disease can also be divided into distinct molecular subtypes by gene expression profiling. Differences observed between the activated B-cell (ABC) and germinal center B cell (GCB) subtypes result in part from distinct genomic alterations (Lenz, PNAS 2008) . For example, recent targeted re-sequencing efforts have revealed mutations in various genes in the NFkB signaling pathway, which likely contribute mainly to the ABC subtype (Compagno, Nature 2009). Thus far, few GCB-specific mutations other than t(14;18) have been identified. Methods: Targeted re-sequencing studies are only able to reveal mutations in pre-selected candidate genes and so much of the genome has not yet been investigated in lymphomas. To obtain a more global view of the mutational landscape of NHL, we applied Illumina massively parallel second-generation sequencing to sequence the genomic DNA of a FL tumor sample. In parallel, we sequenced the transcriptomes (i.e. mRNA) of 31 DLBCL tumor samples using the same technology. Results: Genomic sequencing revealed a mutation altering tyrosine 641 (Y641) in the polycomb group oncogene Enhancer of Zeste Homolog 2 (EZH2), a gene responsible for trimethylating lysine 27 on histone H3 (H3K27). Mutations affecting the same codon were also observed in 4 DLBCL samples. These mutations were confirmed to be somatic in nature by Sanger sequencing exon 15 in tumor DNA and germline DNA derived from peripheral blood in these 5 patients. Sanger sequencing all coding exons in EZH2 from an additional 25 FL samples revealed that mutations were restricted to codon 641. These mutations could change the tyrosine residue to Asparagine, Serine, Phenylalanine or Histidine. The frequency of EZH2 Y641 mutations in GCB-derived lymphomas was determined by Sanger sequencing exon 15 in a total of 479 lymphoma samples. EZH2 (Y641) mutations occurred predominantly in lymphomas of the GCB type: 18/80 (22%) GCB-type de novo DLBCL, 6/52 (11.5)% of “transformed” DLBCL derived from FL and 16/225 (7.1%) of pre-treatment FL samples. No EZH2 mutations were found in 42 DLBCL of the ABC subype, 25 mantle cell lymphomas, 30 small lymphocytic lymphomas or 25 T cell lymphomas or 23 reactive tonsils. Introducing each of the four Y641 mutations into wild-type EZH2 resulted in 15-fold reduction in the ability to trimethylate lysine 27 on H3 peptides in-vitro. Discussion: Over-expression of EZH2 is thought to drive malignancy in a variety of solid tumors but mutations in this gene have never been described. The trimethylated H3K27 epigenetic mark is used to silence the transcription of genes involved in differentiation and hematopoiesis. In Drosophila, the phenotypic consequence of mutating the orthologous tyrosine residue in the E(z)(“Enhancer of Zeste”) gene indicates an apparent gain-of-function (Jones, Genetics 1990). Taken together, our mutation frequency and biochemical data are compatible with the notion that alteration of the ability of germinal center B-cells to trimethylate H3K27 may promote the development of FL and the GCB subtype of DLBCL. Disclosures: Zhu: BPS Biosciences: Employment. Kimbara:BPS Biosciences: Employment. Shashkin:BPS Biosciences: Employment. Charlot:BPS Biosciences: Employment. Tcherpakov:BPS Biosciences: Employment.

Description: Abstract 966 Chronic myeloid leukemia (CML) is a clonal multilineage myeloproliferative disorder characterized by the presence of the fusion gene BCR-ABL with increased tyrosine kinase activity. Imatinib mesylate (IM) and other BCR-ABL tyrosine kinase inhibitors (TKIs), including dasatinib (DA) and nilotinib (NL), have been introduced into clinical practice with remarkable effects on chronic phase CML. However, early relapses, acquired drug resistance, and persistence of leukemic stem cells remain problematic. Improved treatment approaches to target other key molecular elements active in CML stem/progenitor cells are needed. One candidate is AHI-1 (Abelson helper integration site 1), an oncogene that is highly deregulated in CML leukemic stem cells. It harbors two key domains, SH3 and WD40-repeat, which are known important mediators of protein-protein interactions. We recently demonstrated that AHI-1 physically interacts with BCR-ABL and JAK2 in CML cells and this interaction complex mediates transforming activity and TKI response/resistance of CML stem/progenitor cells. We have also shown that AHI-1 interacts independently with JAK2 and BCR-ABL via different binding sites to mediate their activities. In this study, we have characterized the biological and structural functions of the SH3 domain of AHI-1. To determine roles of the SH3 domain in regulation of cell proliferation and TKI response/resistance, several mutant forms, including SH3 domain deletion (SH3Δ), double WD40-repeat and SH3 domain deletion (SH3WD40Δ) and N-terminal deletion (N-terΔ, containing SH3 and WD40-repeat domains) were generated and stably transduced into BCR-ABL inducible BaF3 cells, in which the level of expression of BCR-ABL can be down-regulated by exposure to doxycycline. Overexpression of full-length Ahi-1 in BCR-ABL inducible cells resulted in fewer Annexin V+ apoptotic cells with doxycyclin (suppression of BCR-ABL) compared to BCR-ABL inducible cells (3 and 29% v.s.10 and 60% after 24 or 48 hours). Cells expressing the SH3Δ mutant and the SH3WD40D mutant displayed dramatically increased Annexin V+ cells (10, 77% and 34, 90% v.s.3 and 29%), while cells expressing the N-terΔ mutant had similar numbers of Annexin V+ cells compared to BCR-ABL inducible cells (6 and 41% v.s.10 and 60%). Similarly, BCR-ABL+ cells transduced with SH3Δ and SH3WD40D mutants displayed significantly increased apoptotic cells compared to cells transduced with full-length Ahi-1 in the presence of 2 μM IM (57, 87 vs. 26%), 2μM NL (65, 87 vs. 25%) and 150 nM DA (63, 96 vs. 34%) after 24 hour treatment. BCR-ABL+ cells transduced with the N-terδ mutant also showed more sensitivity to the drug treatments compared to the cells with the full-length Ahi-1(36% for IM, 40% for NL and 40% for DA), but with lower sensitivity than cells carrying the Ahi-1 SH3 domain deletion mutants, indicating that the SH3 domain of Ahi-1 plays a role in the mediation of TKI resistance. The crystal structure of the AHI-1 SH3 domain at 1.32-Å resolution revealed that the AHI-1SH3 domain adopts a canonical SH3 folding, but with an unusual C-terminal α helix. There are three large negatively charged patches, which are constructed by the n-Src loop, the end of the RT loop and the C-terminal helix, and this special feature may be involved in binding selectivity and specificity. PD1R peptide, known to interact with the PI3K SH3 domain, was used to model the binding pattern between AHI-1 SH3 domain and its ligands, and there may be formation of an “Arg-Arg-Trp” stack within the binding interface, which could be a targeting site for designing specific drugs. Moreover, using the AHI-1 SH3 domain as protein ‘bait' in immunoprecipitation/mass spectrometry, Dynamin-2 was identified as a potential interacting partner of AHI-1; both AHI-1 and Dynamin-2 are involved in trafficking and signaling processes. In conclusion, the investigation of the structure of AHI-1 SH3 domain and its interacting proteins will thus provide invaluable insight in identification of key interaction sites in regulation of drug resistance and may be utilized for development of small molecule inhibitors for CML. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 632 Introduction: Non-Hodgkin lymphomas (NHL) are the most common type of lymphoma and can be broadly classified as indolent (slow-growing) diseases, progressing over many years; and aggressive (fast-growing) diseases, which progress rapidly. The latter class includes diffuse large B-cell lymphoma (DLBCL), which accounts for approximately 30% of all NHL diagnoses. Three DLBCL subtypes have been identified based on gene expression profiling, namely: germinal center B-cell (GCB), activated B-cell (ABC) and primary mediastinal B-cell lymphoma (PMBCL). These subtypes show substantial differences in response to treatment and ultimate disease outcome, suggesting that molecular subtyping is an important prognostic indicator and that each subtype may benefit from a distinct treatment regimen. Despite recent advances in cancer genomics revealing molecular and mutational differences between these subtypes, further studies focused on the common NHL subtypes are required to identify critical players in the pathogenesis of DLBCL that may be targeted by pharmacological intervention to improve patient outcome. Methods: Using ultra-high throughput whole genome shotgun sequencing (WGSS) and whole transcriptome shotgun sequencing (WTSS/RNA-seq) we have discovered protein-coding mutations in NHL genomes. With a focus on recurrent and likely gain-of-function mutations we have established procedures to model the three-dimensional structures of mutant proteins and using a computational “molecular docking” pipeline have identified candidate molecules with specificity for the mutant protein. These small molecule compounds are acquired and tested in cell proliferation assays against a suite of DLBCL cell lines characterized for target mutations. Results: Mutations affecting a single key tyrosine in the catalytic site of enhancer of zeste, homolog 2 (EZH2), a member of the Polycomb-group family involved in transcriptional repression were identified (Morin, R. et al. 2010 Nature Genetics 42(2):181-5). This mutation, in a gene previously unknown to be mutated in cancer, is restricted to the GCB subtype of lymphomas and is highly prevalent in patient samples and DLBCL cell lines. Mutations have also been observed in other proteins involved in epigenetic regulation and thus afford potentially novel therapeutic targets. In proof-of-principle experiments small molecule inhibitors were identified using molecular docking approaches to target the effect of EZH2 mutations in both mutant and wild-type DLBCL cell lines. We identified and imported 96 compounds from the Developmental Therapeutic Program NCI/NIH repository. These compounds were tested in alamarBlue cell proliferation assays revealing three with activity at 10uM concentration in EZH2 mutant but not wild-type cells. Computational optimization of these compounds is underway to identify related compounds with improved activities at reduced concentrations. Conclusions: High-throughput sequencing platforms have enabled the identification of recurrent, non-synonymous protein mutations in tumor genomes and transcriptomes. Such a catalogue of mutations provides new avenues of exploration for targeted therapy including small molecule inhibitors. Despite intensive efforts launched in recent years to determine the crystal structure for every human protein, many (including EZH2) do not currently have three dimensional structures. This poses a challenge to novel drug discovery but can be overcome using homology modeling and/or targeting other members of a pathway. Our observations also demonstrate the importance of epigenetic regulation in NHL tumorigenesis and thus provide potential new therapeutic targets. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 804 Introduction: Follicular lymphoma (FL) and diffuse large B cell lymphoma (DLBCL) are the two most common types of non Hodgkin lymphoma (NHL). It is widely accepted that DLBCL can be divided into two major subtypes using gene expression profiling: germinal center B-cell (GCB) and activated B-cell (ABC). Both FL and the GCB subtype of DLBCL derive from germinal center B cells and have been found to share some common mutational events such as translocations leading to the deregulation of the BCL2 oncogene and mutations affecting a single tyrosine (Y641) in the histone methyltransferase EZH2. In contrast, ABC DLBCL tumors are characterized by mutations leading to the constitutive activity of NFkB. The clear differences in treatment response between subtypes allow this distinction to be used as a prognostic indicator and may ultimately lead to therapies that target individual features of each subtype. However, besides the gene expression and mutational signatures that differentiate the DLBCL subtypes, there is a paucity of molecular prognostic markers in these NHLs. Further, there is limited knowledge about the genetic events that drive the GCB subtype of DLBCL, which, if better understood, may enable the design of targeted therapeutics. Methods: To identify mutations driving lymphomagenesis and in particular, aggressive cases of NHL, we applied Illumina second-generation sequencing technology to the analysis of tumor genomes and constitutional DNAs from a FL and a DLBCL tumor and the exomes from two additional DLBCLs. In these “omes”, we identified somatic protein-altering point mutations in more than 250 genes including genes known to be involved in cancer, for example TP53, FAS and TNFAIP3 (A20). Many of these mutations may represent passenger rather than driver mutations, the latter of which are involved in disease progression. To identify the likely driver mutations, we sought to identify the genes that are recurrent targets of somatic mutation in these cancers. To this end, we further analyzed the transcriptome sequences we generated using RNA-seq from 95 primary DLBCLs,13 FL cases and 10 DLBCL-derived cell lines. Results: 105 of the genes found mutated in the FL and DLBCL genomes were observed to be recurrent targets of somatic mutation in these diseases. Some of these were known targets of aberrant somatic hypermutation (SHM) including BCL2, PIM1, and IRF4 and others have been previously identified as targets of recurrent mutation in lymphoma, such as EZH2, CD79B and CARD11. One of the most frequently mutated genes was MLL2, a histone methyltransferase never before implicated in lymphomagenesis. MLL2 showed a pattern of mutation characteristic of a dosage-sensitive tumor suppressor gene. Another frequently mutated gene was MEF2B, a calcium-regulated transcriptional co-activator/repressor that cooperates with histone modifying enzymes to epigenetically regulate the expression of genes. We found that mutations affecting MEF2B occur in 11.7% of FL and 9% of DLBCL, with the majority (73%) of these mutations affecting three amino acids (K4, Y69, and D83). Analysis of these 105 recurrently mutated genes for prognostic signatures is ongoing. Conclusions: High-throughput sequencing platforms have enabled the identification of recurrent targets of somatic mutations never suspected to be involved in lymphoma. Some of these mutated genes may have prognostic value while others may represent targets for the rational design of novel therapeutics. Disclosures: No relevant conflicts of interest to declare.