Multi-Scale Modeling Recapitulates the Effect of Genetic Alterations Associated With Diffuse Large B-Cell Lymphoma in the Germinal Center Dynamics

Abstract

Diffuse large B-cell lymphoma is the most common subtype of non-Hodgkin’s lymphoma. It is a germinal center (GC)–derived, aggressive, and heterogeneous disease. Several transcription factors and signaling pathways that play a central role in the progression of the GC reaction and B-cell differentiation have been shown to play an oncogenic role in diffuse large B-cell lymphoma. B-cell lymphoma 6 (BCL6) is a transcriptional repressor that induces the GC B-cell phenotype and blocks plasma cell (PC) differentiation, while interferon regulatory factor 4 (IRF4) and B lymphocyte-induced maturation protein 1 (BLIMP1), a transcriptional promoter, both mediate PC differentiation and exit from the GC (1). Computational models are useful alternatives to trial-and-error experimental investigation. Ordinary differential equation (ODE) models have been used to study different known mechanisms of lymphomagenesis and suggest candidate tumorigenic alterations (2). Furthermore, multi-scale models (MSMs) have been used to study the role of cellular and molecular mechanisms involved in tumor growth (3–6). In this study, we used an existing MSM of PC differentiation in the GC to simulate eight different models with several candidate genetic alterations of the BCL6-IRF4-BLIMP1 regulatory network that lead to transcription factor deregulation and could explain the onset of diffuse large B-cell lymphoma and recapitulate the GC dynamics observed in such conditions. We observed that models with loss of BLIMP1 function (BLIMPloss and BLIMPlossIRFinc) result in an accumulation of B cells in the GC and a block of PC differentiation and thus correctly recapitulate the observed GC and transcription factor dynamics. Models with constitutive activation of the nuclear factor kappa-light-chain-enhancer of activated B-cell (NF-kB) pathway alone and in codominance or co-expression with the enforced BCL6 expression (IRFinc and BCLincIRFinc) result in a decrease of GC B cells and unaltered PC production at early stages of the GC reaction, as observed experimentally. Interestingly, we also found that in IRFinc and BCLincIRFinc models, an increase in PC production could happen at later stages of the GC reaction. Nevertheless, models with enforced BCL6 expression (BCLauto and BCLinc) result in an expansion of GC B cell population and a block in the PC production that was not observed experimentally. Finally, models with loss of IRF4- and BLIMP1-mediated silencing of BCL6 (IRFsil and BLIMPsil) did not affect GC and transcription factor dynamics.