Background Familial Mediterranean fever (FMF) is an inherited disorder of inflammation characterized by self-limited attacks of fever with serosal, synovial, or cutaneous inflammation, sometimes complicated by systemic amyloidosis. In 1992 our laboratory mapped the FMF locus to chromosome 16p13.3, and in 1997 we isolated the underlying gene, MEFV, and demonstrated that it is highly expressed in granulocytes. At present there are over 50 known disease-associated mutations in MEFV. Although FMF is usually regarded as a recessively inherited disease, we and others have shown that a single mutation in MEFV may, in some circumstances, be sufficient for clinical FMF. We have also studied FMF population genetics, the regulation of FMF gene expression in leukocyte populations, the biochemistry and cell biology of pyrin (the FMF protein), and have developed animal models for FMF. The N-terminal 90 amino acids of pyrin comprise a motif, commonly called the PYRIN domain, found in approximately 20 human proteins. The cognate interaction of the PYRIN domain of pyrin with the homologous domain of the apoptosis-associated specklike protein with a CARD (ASC) places pyrin upstream in the regulation of caspase-1-mediated interleukin-1 (IL-1) beta activation. In studies of peritoneal macrophages from a mouse strain expressing a truncated form of pyrin, we found increased caspase-1 activation and IL-1 beta processing, and impaired apoptosis through a caspase-8-dependent, IL-1 beta-independent pathway. By yeast two-hybrid studies, we demonstrated that pyrin interacts with the cytoskeletal protein PSTPIP1, and that PSTPIP1 mutations associated with the syndrome of pyogenic arthritis with pyoderma gangrenosum and acne (PAPA) lead to markedly increased pyrin-binding and IL-1 beta activation. Results of the Last Year Studies of knockin (KI) models of FMF: During the last year we have completed initial studies of KI mouse models of FMF, and currently have a manuscript under revision describing our findings. To study the role of FMF-associated mutations in the C-terminal B30.2 domain of human pyrin in the molecular pathogenesis of disease, we generated KI mouse models by inserting the B30.2 domain of wild-type (WT) or FMF-associated M680I, M694V, and V726A mutant pyrin into mouse pyrin, which ordinarily does not include a B30.2 domain (although there is strong homology with human pyrin at the nucleotide level at the 3 end of the gene). While the WT human B30.2 was embryonic lethal, the mice homozygous for human FMF mutations exhibited a phenotype similar to human disease, but that was chronic rather than episodic, and generally more severe. KI mice exhibited growth retardation, spontaneous dermatitis and arthritis, and increased CD11b+ cells (especially Ly-6G+ neutrophils) in the blood. Somewhat surprisingly, the phenotypic severity in the mouse was inversely proportionate to the severity of the mutations in humans, with M694V the most severe mutation in humans but the least severe in the mouse, and V726A the least severe mutation in humans but the most severe in the mouse. The inflammatory phenotype was observed in homozygous KI mice, but not heterozygous or hemizygous mice, suggesting that FMF-associated mutations are gain-of-function, with a gene-dosage effect. Bone marrow (BM) cells of KI mice transferred the phenotype into WT mice, and WT BM cells rescued the diseased KI mice, suggesting that BM-derived cells are necessary and sufficient for the disease. Lymphocytes are not required for inflammation since Rag-1 deficient KI mice showed phenotypes similar to Rag-1 sufficient KI mice. In CD45 congenic mixed bone marrow chimeras we found evidence that KI hematopoietic cells could induce WT CD11b expansion. Proinflammatory cytokines, such as IL-1beta, were significantly increased in the KI mouse sera. Caspase-1 was constitutively activated in KI macrophages and active IL-1 beta was secreted when stimulated with LPS alone, without ATP, which is also observed in FMF patients. The inflammatory phenotype of KI mice was completely ablated by crossing with IL-1 receptor KO or Asc KO mice, but not Nlrp3 KO mice. Taken together, these data provide evidence for a heretofore unrecognized ASC-dependent NLRP3-independent inflammasome in which gain-of-function mutations cause autoinflammatory disease. Systems biology analysis of the syndrome of periodic fever with aphthous stomatitis, pharyngitis, and/or cervical adenitis (PFAPA): During the last year we have completed a comprehensive study of patients with PFAPA during and between febrile episodes, and are revising a manuscript describing our findings. PFAPA is perhaps the most common recurrent fever syndrome of childhood. Although we have observed that PFAPA patients frequently have relatives who experienced one or more features of PFAPA in childhood, there is not a Mendelian pattern of inheritance, and the pathogenesis is unknown. Using a systems biology approach we analyzed blood samples from PFAPA patients after genetic testing to exclude other known hereditary periodic fevers (HPFs), from healthy children, and from pediatric HPF patients. Gene expression profiling clearly distinguished symptomatic and asymptomatic periods in PFAPA patients and symptomatic periods in children with HPFs. During symptomatic PFAPA episodes, complement (C1QB, C2, SERPING1), interleukin (IL)-1-related (IL1B, IL1RN, CASP1, IL18RAP), and interferon-induced (AIM2, IP10/CXCL10) genes were significantly overexpressed, while T cell-associated transcripts (CD3, CD8B) were downregulated. At the protein level, symptomatic PFAPA episodes manifested significantly increased serum levels of granulocyte colony-stimulating factor, proinflammatory cytokines (IL-18, IL-6), and chemokines for activated T lymphocytes (IP-10/CXCL10, MIG/CXCL9). A relative lymphopenia was also noted during PFAPA attacks. Activated T lymphocyte counts correlated negatively with serum concentrations of IP-10/CXCL10, but positively with those of IL-1 receptor antagonist. Five PFAPA patients were treated with a recombinant IL-1 receptor antagonist and all five demonstrated a prompt clinical and IP-10/CXCL10 response. These data suggest a host-derived activation of IL-1beta and IL-18 during symptomatic PFAPA episodes, with induction of Th1-chemokines and subsequent retention of activated T cells in peripheral tissues. IL-1 inhibition may thus be beneficial for the treatment of PFAPA attacks, with IP-10/CXCL10 serving as a potential new biomarker. Whole-exome sequencing of selected families with unexplained inflammatory or rheumatic phenotypes: With the availability of whole-exome and whole-genome sequencing technologies, we are now intensively studying several families with unexplained phenotypes. These include a family with three children manifesting a PFAPA-like illness, a large family with what appears to be a dominantly inherited cold urticaria, and a third large family with a dominantly-inherited inclusion-body myositis. During the next year we will attempt to filter the data appropriately to identify a tractable number of candidate loci.
Showing the most recent 10 out of 11 publications
