Abstract Diffuse large B-cell lymphoma of activated B-cell type (ABC-DLBCL) is characterized by constitutive NF-kB activation through activation of the B-cell receptor (BCR) pathway. Individual BCR complexes induce autonomous, i. e. antigen-independent, signaling in the majority of ABC-DLBCL (Eken et al. , J Exp Med 2024). Besides autonomous BCR signaling, recurrent activating mutations of various members of the BCR signaling cascade, in particular CD79 and CARD11, and in the TLR pathway member MYD88 contribute to the ABC phenotype. Reciprocal genetic engineering of the ABC-DLBCL cell lines TMD8 (CARD11wt/wt) and OCI-Ly3 (CARD11L251P/L251P/L251P) has demonstrated that the autonomously signaling TMD8 BCR and the oncogenic CARD11 L251P mutation act as functionally near-equivalent and alternative oncogenic drivers (Eken, ASH 2024). ABC-DLBCL cells dependent on autonomous BCR signaling are highly sensitive to BTK inhibition with acalabrutinib and acquire complete acalabrutinib resistance by the CARD11L251P variant. To identify possible alterations of gene expression profiles after reciprocal exchange of these oncogenic drivers by gene editing and retroviral transduction, we established viable TMD8 and OCI-Ly3 clones with alternative driver combinations, specifically for TMD8 (a) mock gene-edited BCR+ CARD11wt/wt, (b) BCR+CARD11L251P/-, and (c) BCRko CARD11L251P/-, and for OCI-Ly3 (a) BCRko CARD11L251P/L251P/L251P, (b) BCRko tgBCRTMD8 CARD11L251P/L251P/L251P, and (c) BCRko tgBCRTMD8 CARD11-/-/-tgCARD11wt. Three independent clones of each genotype were analyzed by RNA sequencing. Genetically modified and unmodified clones clustered with their respective original cell line and at a large distance from the GCB-DLBCL cell line Karpas 422. Tight co-clustering of each group of three clones with bona fide identical genotypes indicated the absence of potentially perturbing founder effects. In order to search for subtle driver-dependent transcriptional changes, we performed differential gene expression analyses. In gene-modified TMD8 clones, 3904 genes showed significant differential expressed after exchange of the autonomous BCR signal with the oncogenic CARD11 allele. In OCI-Ly3 clones, alternative expression of these oncogenic drivers differentially regulated 1081 genes. Only 216 genes were differentially expressed in both ABC-DLBCL cell line systems. One hundred genes were concordantly upregulated in TMD8 and OCI-Ly3 derivative clones that were driven by autonomous BCR signaling; higher expression of 68 genes was concordantly associated with dependence on the CARD11L251P variant. Among the upregulated genes associated with autonomous BCR signaling in both ABC-DLBCL cell lines were CD69, an early activation marker for T and B cells, and the chemokine receptor CXCR4. CARD11L251P induced upregulation of CCND2 in both cell line systems. Gene set enrichment analysis (GSEA) of all differentially expressed genes among the range of clone sets based on gene expression signatures (SignatureDB) and pathway enrichment (MSigDB) databases revealed only marginal overlap between the two cell line systems. Increased activation of IRF4 targets by the CARD11L251P variant was suggested by enrichment in the SignatureDB TF target gene sets ‘IRF4ABCᵣepressed’ and ‘IRF4SPIBABCᵣepressedₐll’ in both ABC-DLBCL cell line systems. Furthermore, concordant enrichment in the MSigDB Module 75 – Immune response, encompassing enriched clinical annotations of DLBCL – was noted for CARD11L251P-driven cells. Gene expression profiling therefore further demonstrates the functional equivalency of autonomous BCR signaling and an oncogenic CARD11 variant as alternative pathogenic drivers in ABC-DLBCL at the transcriptomic level. Upon mutual exchange of the TMD8 BCR and an activating CARD11 mutation, the respective parental cellular identities were tightly preserved despite several subsequent genetic alterations and lacked major transcriptional reprogramming or appreciable alterations of the original ABC-DLBCL phenotype. Very subtly, cells driven by an autonomous BCR signal as a “paraphysiological” but permanent driver may retain more resemblance to immunologic reactions than cells driven by downstream BCR pathway activation. Gene expression profiling does not permit to predict sensitivity to BTK inhibition, and autonomous BCR signaling is an purely immunologic yet oncogenic driver that is not captured by proposed subclassification of DLBCL based on genetic clustering.
Bergen et al. (Mon,) studied this question.