MYD88 is an adaptor protein through which most of the Toll-like receptors (TLRs) and receptors for interleukin (IL)-1 and IL-18 cytokines signal.1 MYD88 transduces signals to the nuclear factor κB (NFκB) transcription factors that, in turn, regulate the production of cytokines and anti-apoptotic molecules. In mice, MYD88 deficiency impairs immunity to a range of microorganisms, and in humans absence of MYD88 results in life-threatening recurrent pyogenic bacterial infections.2,3 Given the central role of MYD88 and TLRs in immunity, overactivity of this pathway has been associated with autoimmune diseases and now, for the first time, TLR signalling is implicated in lymphomagenesis. In a recent Nature letter, Ngo et al.4 report that gain-of-function mutations in MYD88 drive a substantial percentage of B-cell lymphomas. This study demonstrates the need for developing pharmacological inhibitors targeting MYD88 and other components of the Toll pathway in lymphomas. Diffuse large B-cell lymphoma (DLBCL) is the most common type of non-Hodgkin's lymphoma, comprising about 40% of cases. The DLBCL category encompasses a heterogenous mixture of clinically distinct lymphomas defined by gene expression signatures that reflect their origin from various stages of normal B-cell differentiation.5 It is emerging that these tumour subtypes use different oncogenic pathways, which may explain their variable response to standard treatment regimes. The 'activated B cell' (ABC) lymphoma subtype, at present the least curable with about 40% success rate, is likely to originate from the early stages of a germinal centre reaction, as its gene profile is indicative of an in vitro activated B cell.6 A hallmark of ABC DLBCL is the constitutive activation of the NFκB pathway, on which these tumours rely for survival and proliferation.7 In recent years, much progress has been made in uncovering the root causes of NFκB overactivity in ABC lymphomas. Using 'Achilles heel' screens that involved the use of RNA interference to identify genes required for the survival of cancer cells, Staudt and colleagues8 previously discovered that ABC DLBCL depend on a complex composed of CARD11, BCL10 and MALT1 to activate NFκB. Further analysis of lymphoma cell lines and patient tumour samples revealed oncogenic driver mutations affecting CARD11 (∼10% cases)9 and CD79A/B (∼20% cases)10 that constitutively amplified signals in the B-cell receptor–NFκB signalling axis. Other groups also uncovered frequent mutations inactivating the NFκB-negative regulator A20 in ∼30% cases of ABC lymphomas11,12 (Figure 1). Chromosomal translocations affecting the germinal centre regulator BCL6 and inactivating mutations in the plasma cell differentiation regulator PRDM1 have also been found in human ABC lymphomas, and these disruptions are thought to promote malignant progression by allowing escape from terminal differentiation and cell cycle arrest.13,14 What drives constitutive NFκB signalling in the remaining ABC DLBCL cases? Ngo et al.4 shed light on this question by discovering MYD88 mutations in 39% of cases, implicating deregulation of TLR signalling in pathogenesis of ABC lymphoma. First, MYD88 was found to be essential for the survival of ABC cell lines in 'Achilles heel' screens. Then, Ngo et al.4 used a powerful combination of high throughput mRNA resequencing of ABC cell lines and sequence analysis of the MYD88 coding region in 382 biopsied tumour samples. Remarkably, a single amino acid substitution (L265P) occurred in 29% of the biopsies, whereas other mutations were found at much lower frequency. The L265P mutation was absent in other DLBCL subtypes, and copy number variation analysis among the ABC lymphoma samples suggested that the malignant cells preferentially selected this mutant allele. This intriguing L265P mutation falls in the MYD88 Toll/IL-1 receptor domain. In normal cells, the Toll/IL-1 receptor domain is involved in homotypic interactions, recruiting MYD88 to the activated receptor to form a complex consisting of the kinases IRAK1 and IRAK4.1 The L265P mutant MYD88 associated with phosphorylated IRAK1, which was abundant in ABC DLBCL cell lines. Knockdown of IRAK4 compromised the survival of ABC lines specifically, and this was reversible by introduction of wild-type but not kinase-inactivated IRAK4, suggesting that the tumour cells relied on the kinase activity of IRAK4. This was supported by the finding that an inhibitor of the kinase activity of IRAK1 and IRAK4 killed ABC cell lines, but not GC or myeloma cell lines. These findings hold considerable promise for the pharmacological inhibition of IRAK4 in lymphomas bearing oncogenic MYD88 mutations or other mutations enhancing TLR–NFκB signalling. Downstream functional effects of the L265P mutation in MYD88 appear to include increases in NFκB activity and, interestingly, enhanced JAK–STAT3 (Janus kinase–signal transducer and activator of transcription 3) signalling and interferon-β production. Specifically, Ngo et al.4 knocked down MYD88 and inhibited IRAK1/4 in ABC cell lines and observed decreases in STAT3 signalling and production of IL-6, IL-10 and interferon-β. These findings were reminiscent of previously determined ABC DLBCL subsets in which IL-6 and IL-10 induced JAK–STAT3 signalling as part of an autocrine loop that contributed to their survival.15,16 It will be important to dissect the contributions of these cytokines in the malignant progression of ABC lymphomas. The detection of IL-6, IL-10 and interferon-β in patients could potentially be used to improve existing diagnosis protocols, and inhibition of these cytokines may strengthen treatment regimes against certain subtypes of lymphomas. The activation of several signalling pathways that may potentially cooperate to promote malignancy demonstrates the potency of this single L265P amino acid substitution in MYD88. This oncogenic version of MYD88 was also found in ABC lymphomas bearing mutations in CARD11, CD79A/B or A20, suggesting that, together, these lesions may confer non-redundant attributes that enhance the malignant potential of a B-cell clone. These observations could be important from a therapeutic standpoint, as they indicate that a multipronged approach targeting several key drivers simultaneously may be necessary to improve on traditional treatment regimes or single-target approaches. Worth noting here is the striking overlap between the signalling pathways that are commonly disrupted in autoimmunity and cancer. In 2006, Pisitkun et al.17 showed that overactive Toll signalling is sufficient to drive autoimmunity, which was previously linked by Burnet18 to a 'conditioned malignancy' of forbidden autoreactive lymphocytes. Although indirect, this study by Ngo et al.4 provides further evidence to the 'multistep hypothesis of autoimmune disease pathogenesis',19 as it uncovers an extra capacity of overactive Toll signalling to drive malignant lymphomas. Oncogenic lesions in activated B-cell-like lymphoma. Under normal circumstances, an activated B-cell clone receives T-cell help and expands to give rise to antibody producing cells in a germinal centre reaction. However, genetic lesions targeting genes such as MYD88, CARD11, CD79, PRDM1, BCL6 and TNFAIP3 (A20) may occur and give rise to oncogenic alleles. This results in increased signalling through the NFκB pathway, and the production of antiapoptotic molecules and cytokines (IL-6 and IL-10). These cytokines form part of an autocrine loop that potentially enhance the survival of the B-cell clone through increased STAT3 signalling. Aberrant activity of NFκB and STAT3 pathways could lead to aggressive expansion of the B cell, during which some clones acquire more growth-promoting lesions. These malignant clones expand and select for mutations to further enhance their survival and ability to escape tumour surveillance, eventually producing lymphomas.
Read more
1
