During the current reporting period we have completed 2 studies and published our findings: 1) Discovery of HA20, a Dominantly Inherited Disorder Caused by Haploinsufficiency of TNFAIP3, Encoding the A20 Protein The ubiquitin-editing enzyme A20 functions as a negative regulator of the NF-kappaB signaling pathway. Several regulatory mechanisms, including post-translational protein phosphorylation and ubiquitination of various signaling molecules, keep NF-kappaB signaling in balance. Germline inactivation of A20 or selective deletion of A20 in dendritic, myeloid, T, and B cells show strong evidence for inflammatory phenotypes with some features of autoimmunity. Human genetic studies have linked common variants in TNFAIP3 to several autoimmune diseases and somatic A20 mutations to malignant diseases. Through a combination of whole exome sequencing and candidate gene screening we identified 5 heterozygous germline mutations in TNFAIP3 that segregate with a dominantly inherited childhood-onset systemic inflammatory condition. Eleven patients from 5 unrelated families of diverse ancestry presented with arthralgia/arthritis, oral and/or genital ulcers, ocular and GI inflammation. The syndrome resembles Behcets disease (BD), a relatively common complex/polygenic disorder in several Middle Eastern and East Asian populations. Targeted sequencing of TNFAIP3 in 384 Turkish and 384 Japanese Behcets patients identified 1 additional Turkish family with a novel mutation, bringing the total number of rare disease-associated mutations in TNFAIP3 to 6. These high-penetrance rare variants segregate properly with the affection status in all 6 families and are not reported in any public database or in our database of more than 500 exomes of patients with autoinflammatory disorders. All 6 variants are either nonsense or frameshift mutations predicted to generate truncated proteins of similar length. We showed that primary patients cells, PBMCs, and fibroblasts have lower protein expression of wild type A20 protein relative to healthy controls, while the mutant truncated A20 proteins are not clearly detectable. Five out of 6 disease-associated TNFAIP3 variants are located in the amino-terminal ovarian tumor domain (OTU) of A20 while the sixth mutation resides in the ZnF4 domain. The OTU domain of A20 has a deubiquitinating activity by removing K63-linked polyubiquitin chains from various adaptor proteins such as RIP1, TRAF2, TRAF6, while the ZNF4 domain is known to mediate A20 ubiquitin ligase activity and binding to K63 linked ubiquitin chains. We overexpressed mutant A20 proteins and they fail to suppress TNF-induced NF-kappaB activity in a human T cell line and in 293T cells. Co-transfection of a mutant truncated A20 construct with WT A20 did not abrogate the suppressive activity of WT A20 protein, suggesting that the mutant protein does not have a dominant-negative effect. Stimulated cells from A20-deficient patients showed increased IKK-mediated phosphorylation of IkappaBalpha resulting in IkappaB degradation and nuclear translocation of p65. In addition, we observed an increase in MAP-kinase activity in mutant cells. These data provide strong evidence for enhanced NFkappaB signaling. Upon stimulation A20 is recruited to the TNFR complex to regulate downstream signaling pathways. Patients PBMCs showed reduced association of A20 with the TNFR complex and impaired deubiquitination of A20 target molecules including NEMO, RIP1, and TRAF6. This observation was confirmed by overexpression studies with transiently transfected mutant A20 proteins that failed to remove K63-ubiquitin chains from target proteins and failed to antagonize the deubiquitinase function of WT A20. Inefficient deubiquitination of A20 target proteins might explain a higher NFkappaB signaling activity in mutant cells. Active NFkappaB subunits promote the transcription of genes encoding pro-inflammatory cytokines such as interleukin IL-1beta, IL-6, and TNF. Levels of several proinflammatory cytokines, including IL-1, IL-6, IL-9, IL-17, IL-18, TNF, were substantially increased in patients serum and in the supernatants of stimulated PBMCs relative to healthy controls. Intracellular cytokine staining revealed increased polarization toward Th9 and Th17 CD4 T-effector cell lineages. Finally, the frequency of CD14+ inflammatory monocytes was significantly higher in patients than in healthy controls. A20 may function as a negative regulator of the Nlrp3 inflammasome independently of its role in NFkappaB regulation. We showed that patient cells have constitutive activation of the NLRP3 inflammasome as demonstrated by increased activation of caspase-1 and increased secretion of active IL-1beta and IL-18. This effect was attenuated by use of several known inhibitors for the NLRP3 inflammasome. Preliminary experience in a patient treated with anakinra, an IL-1 inhibitor, has been encouraging. A manuscript reporting these findings was published in Nature Genetics. 2) Discovery of a Dominantly Inherited Form of Vibratory Urticaria Caused by Mutations in ADGRE2 Vibratory urticaria (VU) is a rare condition in which sustained vibration against the skin induces a localized hive and facial flushing. We ascertained 2 large Lebanese kindreds in which VU segregates as an autosomal dominant trait. In affected family members, acute onset of symptoms, concurrent peripheral histamine release, and increased tryptase staining in post-vibration skin samples implicated mast cell degranulation. Through linkage analysis and exome sequencing we identified the missense mutation p.C492Y in ADGRE2 (formerly known as EMR2) as the only rare nonsynonymous or splice variant co-segregating with VU. The p.C492Y mutation was absent from variant databases and 1105 ancestry-matched controls. ADGRE2 encodes an adhesion G-protein coupled receptor that undergoes autocatalytic cleavage, producing an N-terminal extracellular alpha subunit that remains non-covalently bound to a C-terminal transmembrane beta subunit. ADGRE2 was highly expressed in human mast cells, and its alpha subunit binds to dermatan sulfate, which is abundant in skin. Patient-derived primary mast cells, when adhered with either dermatan sulfate or an antibody that ligates the ADGRE2 alpha subunit, degranulated when subjected to vibration, whereas mast cells from unaffected subjects showed no response. Likewise, human LAD2 mast cells expressing ADGRE2 with the p.C492Y mutation in the alpha subunit showed greater degranulation in response to vibration than control cells expressing non-mutant ADGRE2. This activity was cleavage-dependent, suggesting the subunit interaction must be non-covalent to permit vibration-induced degranulation. The p.C492Y mutation did not affect expression, trafficking, cellular adhesion, or cleavage of ADGRE2. However, during vibration of transfected cells, the mutant alpha subunit dissociated from the transmembrane beta subunit, whereas the wild type alpha subunit retained its co-localization with the beta subunit on the plasma membrane. Furthermore, LAD2 cells expressing an ADGRE2 truncation mutant encoding only the beta subunit showed constitutive degranulation, indicating the alpha subunit is likely auto-inhibitory. Our data suggest a pathogenic mechanism whereby the p.C492Y mutation destabilizes this inhibitory subunit interaction, sensitizing dermal mast cells to vibration-induced hyperactivation of beta subunit-mediated signaling. Degranulation induced by the p.C492Y mutation is likely an exaggeration of a normal response regulated by non-mutant ADGRE2, implicating this molecule as a mechanosensor and suggesting a dual role for mast cells in responding to both allergenic stimuli through the IgE receptor and physical stimuli through ADGRE2. We published these findings in the New England Journal of Medicine.
Showing the most recent 10 out of 36 publications
