SMARCA4 encodes one of two mutually exclusive ATPases present in BAF (aka SWI/SNF) chromatin remodeling complexes. SMARCA4 is dynamically expressed during normal B-cell development, with high expression in germinal center B (GCB)-cells, and is targeted by heterozygous inactivating mutations in GCB-derived lymphomas including 34% of pediatric Burkitt lymphoma (BL), 17% of adult BL, 4% of GCB-DLBCL. However, the functional role of SMARCA4 in GCB cell development and malignant transformation has not been thoroughly investigated.

We generated a novel floxed Smarca4 transgenic mouse model and crossed to the Aid-Cre line for GC-specific conditional knock-out (cKO). Following immunization with sheep red blood cells (SRBCs) to induce a GC reaction, Smarca4+/- mice had a significant increase in GCB-cell abundance compared to Smarca4+/+ littermates (p<0.001) due to both an increased average GC size and number. Notably, Smarca4-/- mice displayed a significantly reduced frequency of GC B-cells (p<0.001) and sorting of residual cells revealed retention of 1 or more copies of the floxed allele, suggesting that biallelic cKO of Smarca4 is lethal to GCB cells. We therefore focused subsequent studies on comparisons between Smarca4+/- and Smarca4+/+ mice. Further analysis revealed significant polarization towards a centroblastic phenotype in Smarca4+/- mice (p<0.001), in line with the centroblastic morphology of human BL. Analysis of competitive advantage with the same GC using bone marrow chimera models showed a significant advantage of Smarca4+/- GCB cells compared to Smarca4+/+, associated with a significant increase in proliferation measured by Edu incorporation and reduction of apoptosis measured by Annexin-B. We extended upon this using single cell RNA-seq (scRNA-seq) of GCB cells from Smarca4+/- and Smarca4+/+ mice, which confirmed an increased centroblastic polarization of Smarca4+/- GCB cells and identified a significant increase in light zone to dark zone recycling cells that may underly this polarization. We validated this using the Myc-GFP allele, which confirmed increased Myc+ recycling cells in GCs from Smarca4+/- mice compared to Smarca4-/- mice.

BAF complexes drive ATP-dependent mobilization and eviction of nucleosomes to convert chromatin from inactive to active states. Brg1 deficiency may therefore alter distinct BAF complexes and transcription factor programs. We therefore explored the role of Smarca4 in regulating chromatin accessibility and gene expression using ATAC-seq and RNA-seq in Smarca4+/- and Smarca4+/+ murine GCB-cells. Supervised analysis revealed predominant loss of chromatin accessibility associated with repression of nearby genes. Smarca4+/- cells showed significant loss of accessibility of genes upregulated in centrocytes, and a preferential loss of NF-ĸB target genes (GSEA FDR<0.01).

MYC translocations are pathognomonic for BL, in which SMARCA4 mutations are most frequent. We therefore crossed our Smarca4 cKO mice with a cre-inducible Myc knock-in (Myc-KI) allele to mimic combined Smarca4 loss-of-function and Myc activation. The addition of Myc-KI did not lead to a significant increase in GCB cell frequency following a single SRBC immunization. However, serial re-challenge of mice with SRBCs led to lymphadenopathy, splenomegaly and infiltration of highly proliferative B-cells into the major organs (lung, liver, and kidney) in Smarca4+/- + Myc-KI mice following 3 immunizations that was not observed with Smarca4+/- or Myc-KI alleles alone. Furthermore, long term follow-up of serially-challenged mice showed a significant reduction in overall survival in Smarca4+/- + Myc-KI mice compared to Smarca4+/- or Myc-KI mice (p=0.010).

Collectively, these data demonstrate a competitive advantage of GCB cells with monoallelic Smarca4 cKO, and a centroblastic GC polarization driven in part by increased recycling. Mechanistically, this is associated with loss of chromatin accessibility and gene expression of genes that are expressed in centrocytes, including NFkB target genes. Monoallelic Smarca4 cKO cooperates with Myc over-expression to drive expansion and extranodal dissemination of GCB cells following serial re-challenge, leading to reduced survival. These data provide the first functional and mechanistic insight into the role of SMARCA4 in GCB cell development and lymphomagenesis.

Vega:Geron Corporation: Research Funding; Allogene: Research Funding; Crisp Therapeutics: Research Funding. Mason:Tempus Labs: Membership on an entity's Board of Directors or advisory committees. Green:Abbvie: Research Funding; Daiichi Sankyo: Consultancy, Honoraria; Kite/Gilead: Research Funding; Tessa Therapeutics: Honoraria; Monte Rosa Therapeutics: Honoraria; Allogene: Research Funding; KDAc Therapeutics: Current holder of stock options in a privately-held company; Sanofi: Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution