Identification of LMO2 Transcriptome and Interactome in Diffuse Large B-Cell Lymphoma by Integrated Experimental and Computational Approach

Abstract 438

The LMO2 is a cysteine-rich protein containing two zinc binding LIM domains indirectly regulating gene expression by mediating protein-protein interactions with other transcriptional factors (TFs), facilitating the formation of multipartite DNA-binding complexes. It is mainly expressed in endothelial and hematopoietic cells forming cell type specific complexes containing LDB1, TAL1, E2A and GATA2 proteins in endothelial cells and hematopoietic stem cells and LDB1, TAL-1, GATA1 and E proteins in the erythroid lineage. In these cells LMO2 is involved in angiogenesis and erythroid hematopoiesis. In T cells where LMO2 is only expressed in immature CD4/CD8 double-negative thymocytes, the LMO2 complex consist of LDB1, TAL1 and E2A, but may also bind to GATA3. Aberrant expression of LMO2 in T cells induces leukemogenesis. In the B cells, LMO2 is specifically up regulated in Germinal Center (GC) B cells. LMO2 is also expressed in GC-derived non-Hodgkin's lymphomas and is one of the most powerful survival predictors in DLBCL patients. However, its function in GC B cells and DLBCL is currently unknown. In the present study we aimed to characterize the LMO2 transcriptome and interactome in DLBCL cells. Gene expression arrays were performed in Rck8 cells expressing low levels of endogenous LMO2, in which LMO2 was stably overexpressed to levels observed in GCB-like DLBCL. A total of 311 differentially expressed genes (DEGs) between control cell lines transfected with mock vector and LMO2 stably transfected samples were identified at FDR of 0.05. Sixty-four genes were down-regulated by LMO2 transfection, while 247 were up-regulated. Prominent amongst these were 27 genes encoding proteins of the histone cluster 1, colocalized on chromosome 6 and consistently higher expressed in LMO2 expressing cell lines. At the same time, multiple cell-cycle-related genes were also up-regulated by LMO2 transfection, including centromere proteins (CENPE, CENPI, and CENPN), the kinetochore associated protein NDC80 and the mitotic protein CDC25C. Expression of several randomly selected genes (SPIC, LAX1, DLEU2, DOCK3, CHND2 and TNFRSF9) was validated by real time PCR and confirmed the observed gene expression changes upon LMO2 overexpression. To obtain a higher-level view of expression changes, we compared the list of DEGs to Gene Ontology (GO) categories. This revealed significant induction of genes involved in chromosome, nucleosome, and chromatin and protein-DNA complex assembly. We screened our microarray data against the Broad Institute Molecular Signatures Database, to identify TFs whose target genes were significantly up- or down-regulated by stable LMO2 transfection. This candidate list was filtered to include only those TFs whose target genes significantly overlapped with genes which had previously identified binding motifs for LMO2 complexes in their promoter sequences. Our analysis identified Sp1, NFAT, Elk1, and LEF1 as potential novel LMO2 co-factors, but not classical TAL1 and GATA partner proteins. We examined the expression of previously reported and new potential interaction partners of LMO2 in DLBCL cell lines and GC lymphocytes by Western blotting and analyzed the interaction of the expressed proteins with LMO2 by co-immunprecipitation. We identified that in DLBCL the LMO2 complex contains some of the traditional partners such as LDB1, E2A, HEB, lyl1 and ETO2, but not TAL1 or GATA proteins. Furthermore, we demonstrated LMO2 interaction also with SP1, ELK1, NFATc1 and LEF1 proteins. All the identified LMO2 interacting partners are also expressed in GC B lymphocytes, suggesting the presence of a similar protein complex in both normal and malignant B cells. LMO2 interaction with LEF protein affected its transcriptional activity, as measured by reporter assays. A Chromatin immunoprecipitation assay using promoter region of the DLEU2 gene, whose expression was up-regulated by LMO2 expression, confirmed the direct and specific LMO2 binding to a region harboring E2A and a SP1 binding sites. Overall, our studies identified an LMO2 transcriptome in DLBCL as well as its interactome, which contains novel previously unknown interacting partners. These findings suggest that LMO2 may affect unique and novel cellular functions in GC lymphocytes which may potentially contribute to DLBCL pathogenesis and are currently being investigated in our laboratory.