Background In 1999 our laboratory led an international consortium in reporting that mutations in TNFRSF1A, the gene encoding the 55 kDa tumor necrosis factor (TNF) receptor, cause a dominantly inherited syndrome of fever and inflammation, and proposed the name """"""""TNF receptor-associated periodic syndrome"""""""" (TRAPS) for this condition. We demonstrated 6 mutations in the extracellular domain of the TNFRSF1A protein, 5 of which were missense substitutions at cysteine residues, resulting in the disruption of highly conserved disulfide bonds. The 7 original families included the Irish-Scottish family prototypic for """"""""familial Hibernian fever,"""""""" as well as families from other ethnic groups. In 3 C52F patients, we found a defect in the activation-induced shedding of the p55 (but not p75) TNF receptor, possibly leading to impaired homeostasis in the inflammatory response. During the 4 years leading to the present reporting period, we have identified new TRAPS mutations, studied genotype-phenotype correlations, and established p55 receptor shedding defects for most, but not all, mutations tested. We also conducted an open-label, dose-escalation study of the TNFR p75:Fc fusion protein, etanercept, in 15 TRAPS patients. There were highly significant dose-dependent improvements in a TRAPS attack score while on etanercept. Finally, we have developed TRAPS knockin mice for further mechanistic studies. During the immediately antecedent reporting period, we also identified de novo dominant mutations in a second gene, CIAS1 (also known as NALP3 or PYPAF1), that cause a distinct disorder known as neonatal onset multisystem inflammatory disease (NOMID) or chronic infantile neurologic cutaneous and arthropathy (CINCA) syndrome. Consistent with the recently discovered role of cryopyrin, the protein product of CIAS1, in IL-1 regulation, we found evidence for increased monocyte IL-1beta in a mutation-positive patient compared with normal controls. Results of the Last Year Clinical observations on TRAPS: During the last year we analyzed the biochemical markers of inflammation for the TRAPS-etanercept protocol, which had closed shortly before the beginning of the present reporting period. We measured the erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), and serum amyloid A (SAA) during 3-month baseline, low dose (2x/wk) and high dose (3x/wk) etanercept, and washout periods, both between and during any TRAPS attacks that the patients experienced. In general, etanercept was associated with significant reductions in all 3 acute phase reactants during symptomatic and asymptomatic periods. CRPs remained above the upper limit of normal even during the asymptomatic periods with both low- and high-dose etanercept administration. There were significant increases in acute phase reactants following discontinuation of etanercept. TRAPS knockin mice: Using standard techniques of homologous recombination in embryonic stem cells, followed by in vitro selection and blastocyst injection, we have developed two different types of Tnfrsf1a knockin mice. The first has the substitution of methionine for threonine at residue 50 (T50M), and represents one of the more common and severe mutations associated with human TRAPS. The second has the substitution of tyrosine for cysteine at residue 33 (C33Y), which is the mutation in the original Irish-Scottish """"""""familial Hibernian fever"""""""" family. Homozygous animals for T50M (derived from 2 different ES clones) and C33Y (1 clone) have been generated. These strains have subsequently been crossed with a Cre recombinase strain, permitting the excision of the neomycin resistance cassette flanked by lox p sites. These lines are currently being backcrossed onto the C57BL6 and 129Sv backgrounds. On the mixed C57BL6/129 background, we have confirmed the respective knockin mutations by DNA sequencing. RT-PCR from both T50M and C33Y mice demonstrated Tnfrsf1a transcripts of the appropriate size. Immunoprecipitation and Western blotting with anti-mouse Tnfrsf1a showed intact 55 kDa receptor in leukocytes from homozygous T50M and C33Y mice. More extensive phenotypic and functional studies have been carried out on the T50M mice. By flow cytometry, T50M homozygotes showed detectable, but reduced, levels of leukocyte cell surface Tnfrsf1a. T50M heterozygotes and homozygoes did not spontaneously develop fevers. Homozygotes showed resistance to the lethal effects of bacterial lipopolysaccharide (LPS) coinjected with D-galactosamine, similar to that seen with Tnfrsf1a knockout mice. The T50M knockin mice were also resistant to the thermal effects of low-dose LPS injection. In addition, homozygous T50M mice showed detectable but diminished levels of soluble serum Tnfrsf1a following low-dose LPS injection, with heterozygous littermates exhibiting an intermediate phenotype. These data have led us to hypothesize that the mutant T50M Tnfrsf1a receptor may not reach the cell surface as readily as wild-type receptor, and that, either because of this or as an additional defect, there is impaired TNF signaling through this receptor. Genetic studies in NOMID: We have analyzed an additional 8 patients with clinical NOMID for CIAS1 mutations. We found mutations in only 4, bringing our total to 10 mutation-positive patients out of 21, further strengthening the case for genetic heterogeneity. Two of the new mutations, F523C and G326E, are novel. Extending our previous finding of increased IL-1beta production in a patient with the D303N CIAS1 mutation, we found increased levels of IL-1beta precursor regardless of whether a CIAS1 mutation was present. Given the existence of 13 other human proteins with a domain structure similar to cryopyrin (sometimes denoted NALP or PYPAF proteins), we screened our mutation-negative patients for mutations in several related genes with similar tissue expression profiles. These included NALP1/DEFCAP, NALP2/PYPAF2, NALP4/PYPAF4, NALP6/PYPAF5, NALP12/PYPAF7. We also screened ASC, which encodes a protein that interacts with cryopyrin to regulate apoptosis and proinflammatory caspases, and the gene encoding the IL-1 receptor antagonist. We have identified several missense substitutions in these molecules, and we are currently screening controls and evaluating the functional consequences of these variants. Conclusions and Significance Taken together, our data confirm a major role for abnormalities in cytokine signaling in the dominantly inherited periodic fevers. TNF and IL-1 are the 2 major pyrogenic cytokines in man, and we have now found pathogenic mutations in both signaling pathways. The clinical importance of this insight is emphasized by the success of the TNF-inhibitor etanercept in controlling the symptoms and acute-phase response in TRAPS, although the optimal regimen required to prevent systemic amyloidosis remains to be established. Somewhat paradoxically, laboratory studies of the T50M TRAPS knockin mouse suggest that TNFRSF1A mutations may actually lead to decreased signaling through the p55 receptor, perhaps indicating a role for increased signaling through the p75 TNF receptor in TRAPS. The finding of mutations in cryopyrin, a newly recognized IL-1 regulator, in NOMID raises the hope that IL-1 inhibition may have an impact in this condition similar to the clinical benefit of etanercept in TRAPS. During the next year, our objectives will be: 1) to continue mutational studies of genes in the TNF and IL-1 pathways in patients with uncharacterized inflammatory disorders; 2) to continue our physiologic studies of TRAPS knockin mice; 3) in collaboration with Dr. Goldbach-Mansky, to conduct therapeutic studies of anakinra, the IL-1 receptor antagonist, in NOMID; and 4) to study gene expression profiles of patients with TRAPS and NOMID during and between attacks.
Showing the most recent 10 out of 28 publications
