Mechanistic and proteomic analyses of NF-kappaB-driven lymphomas
Immunology
Final Report Abstract
Upon antigen encounter, naive B cells form germinal centers (GC), where they terminally differentiate into antibody-secreting plasma cells. During GC reactions, B cells undergo somatic hypermutation and isotype class switch recombination of their B cell receptor to improve pathogen recognition and clearance. These processes essentially require DNA breaks and therefore bear the potential not only of self-recognition underlying autoimmunity but also of undesired somatic DNA alterations causing malignant transformation. NF-κB family transcription factors are centrally implicated in autoimmunity and lymphomagenesis. It is remarkable that c-Rel is the only of five NF-κB members that malignantly transforms lymphoid chicken cells in vitro. The REL gene locus is frequently amplified in human B cell lymphomas and a c-Rel splice variant lacking exon 10 (corresponding to exon 9 in mice) has been exclusively detected in diffuse large B cell lymphoma patients but not in healthy individuals. Besides these implications of c-Rel in human lymphomas, single nucleotide polymorphisms within the REL gene locus are associated with various human autoimmune diseases. Despite this evidence, no mouse model to investigate c-Rel gain-of-function in immune cells existed to date and the distinct role of c-Rel gain in B cells remained enigmatic. We generated the first conditional c-Rel transgenic mouse models with the aim of elucidating the precise in vivo consequences of c-Rel overexpression and aberrant splicing in B cells and GC B cells. These novel mouse models allow for Cre-inducible expression of transgenic c-Rel or a GFP-c-Rel fusion protein under control of a strong CAG promoter. In addition, flippase recombinase-mediated excision of exon 9 from this modified REL gene locus enables conditional expression of the c-Rel splice variant that has been identified in B cell lymphoma patients. Induction of c-Rel overexpression specifically in B cells (CD19Cre) or GC B cells (Cγ1Cre) causes a significant expansion of spontaneous GC B cells in lymphoid tissues of young mice, comprising spleen, lymph nodes and gut-associated mesenteric lymph nodes and Peyer’s patches. This dramatic phenotype is accompanied by an increase of specialized follicular helper T cells that provide crucial signals for GC B cells during GC reactions. Moreover, isotype class-switched antibody-secreting plasma cells significantly accumulate in spleen and bone marrow upon c-Rel overexpression leading to corresponding elevated serum antibody titers. Our analysis of mouse models with different c-Rel expression levels revealed that c-Rel expression strikingly correlated with GC B cell and plasma cell expansion. However, to date we could not identify a significant difference between the overexpression of c-Rel and of its exon 9 lacking splice variant. Although aged c-Rel transgenic mice do not spontaneously develop overt lymphoma, aged mice produce classswitched autoantibodies, indicating that the GC reactions are driven by self-antigens and contribute to autoimmunity. Cell cycle analysis suggests that higher proliferation of GC B cells could contribute to the observed GC B cell expansion. We strongly believe that our conditional c-Rel transgenic mouse models not only contributed to the knowledge of c-Rel gain of function in B cells and GC B cells, but that they can be prospectively applied to further advance the understanding of the particular role of c-Rel in immune cells and beyond. Large-scale dissection of lymphoma entities is almost exclusively performed at the RNA level. However, nearly all important cellular functions are carried out by proteins. The cellular protein content can deviate significantly from its corresponding mRNA levels due to posttranscriptional and post-translational modifications. Therefore, in collaboration with Matthias Mann, we set out to determine comparative lymphoma proteomes (choosing initially diffuse large B cell lymphomas characterized by NF-κB-dependent gene expression versus lymphomas without a NF-κB gene signature) in cell lines and in primary material of a small cohort of lymphoma patients. We identified separating protein sets in both lymphoma cell lines and primary patient material. In this analysis, we validated some known candidates from gene expression experiments, but we also identified novel candidates whose functions will be explored in future follow-up studies. Furthermore, we set the stage for large-scale proteomic profiling of different lymphoma entities.
Publications
-
(2014) N-linked Glycosylation Enrichment for In-depth Cell Surface Proteomics of Diffuse Large B-cell Lymphoma Subtypes. Mol Cell Proteomics 13, 240–251
Deeb SJ, Cox J, Schmidt-Supprian M & Mann M
-
(2015) An Oncogenic Role for Alternative NF-κB Signaling in DLBCL Revealed upon Deregulated BCL6 Expression. Cell Reports 11, 715–726
Zhang B, Calado DP, Wang Z, Fröhler S, Köchert K, Qian Y, Koralov SB, Schmidt-Supprian M, Sasaki Y, Unitt C, Rodig S, Chen W, Dalla-Favera R, Alt FW, Pasqualucci L & Rajewsky K
-
(2015) Machine Learning-based Classification of Diffuse Large B-cell Lymphoma Patients by Their Protein Expression Profiles. Mol Cell Proteomics 14, 2947–2960
Deeb SJ, Tyanova S, Hummel M, Schmidt-Supprian M, Cox J & Mann M
-
(2016) Canonical NF-κB signaling is uniquely required for the long-term persistence of functional mature B cells. Proc Natl Acad Sci USA 113, 5065–5070
Derudder E, Herzog S, Labi V, Yasuda T, Köchert K, Janz M, Villunger A, Schmidt-Supprian M & Rajewsky K
-
(2016) Exploring in vivo Consequences of c-Rel Overexpression in Terminal B Cell Differentiation. PhD Thesis
Kober M
