Mutations of multiple genes cause deregulation of NF-kappaB in diffuse large B-cell lymphoma

Abstract

Diffuse large B-cell lymphoma (DLBCL), the most common form of lymphoma in adulthood, comprises multiple biologically and clinically distinct subtypes including germinal centre B-cell-like (GCB) and activated B-cell-like (ABC) DLBCL. Gene expression profile studies have shown that its most aggressive subtype, ABC-DLBCL, is associated with constitutive activation of the NF-kappaB transcription complex. However, except for a small fraction of cases, it remains unclear whether NF-kappaB activation in these tumours represents an intrinsic program of the tumour cell of origin or a pathogenetic event. Here we show that >50% of ABC-DLBCL and a smaller fraction of GCB-DLBCL carry somatic mutations in multiple genes, including negative (TNFAIP3, also called A20) and positive (CARD11, TRAF2, TRAF5, MAP3K7 (TAK1) and TNFRSF11A (RANK)) regulators of NF-kappaB. Of these, the A20 gene, which encodes a ubiquitin-modifying enzyme involved in termination of NF-kappaB responses, is most commonly affected, with approximately 30% of patients displaying biallelic inactivation by mutations and/or deletions. When reintroduced in cell lines carrying biallelic inactivation of the gene, A20 induced apoptosis and cell growth arrest, indicating a tumour suppressor role. Less frequently, missense mutations of TRAF2 and CARD11 produce molecules with significantly enhanced ability to activate NF-kappaB. Thus, our results demonstrate that NF-kappaB activation in DLBCL is caused by genetic lesions affecting multiple genes, the loss or activation of which may promote lymphomagenesis by leading to abnormally prolonged NF-kappaB responses.

Figure 1. The NF-kB pathway is active in ABC-DLBCL and in a smaller fraction of GCB-DLBCL

a) Immunohistochemical staining of DLBCL biopsies with anti-NFKB1(p105/p50)(top) and anti-NFKB2(p100/p52)(bottom) antibodies. Nuclear localization of NF-kB denotes active signaling, as opposed to the inactive, cytoplasmic complex (200X). b) Prevalence of cases displaying constitutive NF-kB activation in DLBCL subgroups. Color codes indicate nuclear p50, p52 or both. c) Western blot analysis of DLBCL cell lines for processing of p100 to p52 (red, ABC-DLBCL; grey, GCB-DLBCL). The multiple myeloma cell line U266 and the epithelial cell line COS are used as positive and negative controls, respectively. Actin, loading control. d) GSEA enrichment score and distribution of NF-kB target genes along the rank of transcripts differentially expressed in ABC- vs GCB-DLBCL. e) Percentage of samples showing a transcriptional NF-kB signature by GSEA. f) Venn diagram illustrating the overlap between immunohistochemistry-defined and GSEA-defined NF-kB positive cases.

Figure 2. Mutations and deletions of the A20 gene in ABC-DLBCL

a) Percentage of A20 mutated cases in various DLBCL subtypes. The exact number over total analyzed is given on top. NC, samples co-expressing CD10 and IRF4 (see Methods). b) Schematic representation of the human A20 protein, with its key functional domains. Color-coded symbols depict distinct alterations leading to A20 inactivation. OTU, ovarian tumor domain; ZF, A-20 zinc-finger domains. c) Dual-color FISH analysis of representative DLBCLs, hybridized with A20-specific probes (red) and a chromosome 6 centromeric probe (green). Arrows point to representative cells displaying hemizygous (left panel) or homozygous (right panel) A20 deletions. d) Chromosome 6 ideogram; the region encompassing A20 on band 6q23 is enlarged in the bottom panel, where columns represent individual cases, and rows correspond to the two alleles. e) Overall frequency of A20 structural alterations (inactivating mutations and deletions, combined).

Figure 3. Reconstitution of A20-null cell lines causes apoptosis and cell growth arrest via inhibition of NF-kB

a, Flow cytometric analysis of AnnexinV-PE/7AAD staining in A20-null and wild-type cell lines, transduced with the indicated vectors; numbers indicate the percentage of single and double-positive cells (lower and upper right quadrants). The results of 3 independent experiments are summarized in panel b (mean ± SD). c, Reduced viability in SUDHL2 cells complemented with A20 expression-vectors (red bars), as compared to vector alone (black bars)(mean±SD, n=3). No effect was observed in two control cell lines. d, Analysis of GFP expression after transduction with pWPI and pWPI-HA-A20 documents the disappearance of the GFP+ population in SUDHL2^HA-A20 and RC-K8^HA-A20, but not in SUDHL2^pWPI and RC-K8^pWPI or in A20 wild-type cell lines (mean ± SD, n=2). e) Immunofluorescence analysis of p105/p50 in SUDHL2 cells transduced with pWPI and pWPI-HA-A20 (top). Nuclei are identified by DAPI (bottom).