FMF = familial Mediterranean fever; TRAPS = tumor necrosis factor receptor (TNFR)–associated periodic syndrome; HIDS = hyperimmunoglobulinemia D with periodic fever syndrome; FCAS = familial cold autoinflammatory syndrome; MWS = Muckle-Wells syndrome; NOMID = neonatal-onset multisystem inflammatory disease.
Genomic-based therapy: Targeting interleukin-1 for autoinflammatory diseases
Version of Record online: 5 FEB 2004
Copyright © 2004 by the American College of Rheumatology
Arthritis & Rheumatism
Volume 50, Issue 2, pages 345–349, February 2004
How to Cite
Hoffman, H. M. and Patel, D. D. (2004), Genomic-based therapy: Targeting interleukin-1 for autoinflammatory diseases. Arthritis & Rheumatism, 50: 345–349. doi: 10.1002/art.20032
- Issue online: 5 FEB 2004
- Version of Record online: 5 FEB 2004
- Manuscript Accepted: 21 OCT 2003
- Manuscript Received: 10 AUG 2003
The Human Genome Project has certainly facilitated a greater understanding of the genetic basis of disease and normal genetic variation, but only recently have the initial promises of improved therapy based on genomic knowledge been realized. In parallel with the progress in human genetics, improved understanding of the immunologic mechanisms underlying inflammatory diseases has led to remarkable advances in treatment. Biologic agents directed toward specific inflammatory mediators, such as tumor necrosis factor α (TNFα) and interleukin-1 (IL-1), have been shown to have significant disease-modifying effects, especially in the treatment of arthritis. In this editorial, we explore the convergence of human genetics and targeted therapeutic immunology.
Although most diseases with arthritis have complex etiologies based on multiple genes and multiple environmental influences, several distinct diseases with joint symptoms are attributable to single-gene defects. These include familial Mediterranean fever (FMF), TNF receptor–associated periodic syndrome (TRAPS), hyperimmunoglobulinemia D with periodic fever syndrome (HIDS), Blau syndrome, familial cold autoinflammatory syndrome (FCAS), Muckle-Wells syndrome (MWS), and neonatal-onset multisystem inflammatory disease (NOMID). These disorders, all of which are within the family of hereditary periodic fever disorders, are characterized by recurrent bouts of systemic inflammation involving several tissues, including joints and skin. Because they are not associated with high-titer autoantibodies or antigen-specific T cells, these disorders are not classic autoimmune diseases and are referred to as autoinflammatory diseases (1). Table 1 lists these diseases, their clinical presentation, inheritance pattern, and underlying genes. Identification of the causative genetic defects is beginning to lead to targeted therapeutic approaches in autoinflammatory syndromes. Directed therapies have now been successfully used in TRAPS, HIDS, and MWS, suggesting the possibility of a similar role for these therapies in FCAS, NOMID, Blau syndrome, and FMF.
|FMF||2–3-day episodes of fever, monarthritis, erythematous rash, and abdominal pain||Autosomal recessive||MEFV/pyrin|
|TRAPS||>7-day episodes of fever, polyarthralgia, migratory rash, abdominal pain, and periorbital edema||Autosomal dominant||TNFRSFIA/TNFRI (p55)|
|HIDS||3–7-day episodes of fever, polyarthralgia, maculopapular rash, abdominal pain, and lymphadenopathy||Autosomal recessive||MVK/mevalonate kinase|
|Blau syndrome||Chronic granulomatous arthritis, erythematous rash, and uveitis||Autosomal dominant||NOD2|
|FCAS||1–2-day episodes of fever, arthralgia, urticaria-like rash, conjunctivitis (predominantly after cold exposure)||Autosomal dominant||CIAS1/cryopyrin or NALP3|
|MWS||1–2-day episodes of fever, arthralgia, myalgia, urticaria-like rash, sensorineural hearing loss||Autosomal dominant||CIAS1/cryopyrin or NALP3|
|NOMID||Chronic arthropathy, urticaria-like rash, organomegaly, aseptic meningitis, papilledema, sensorineural hearing loss||Sporadic/autosomal dominant||CIAS1/cryopyrin or NALP3|
The most well known and best characterized of the periodic fever disorders is FMF, an ancient systemic inflammatory disease initially described in Jewish, Armenian, Arab, and Turkish families. The primary morbidity/mortality of FMF is the development of end-stage renal disease due to systemic amyloidosis. Although the mechanism of action is unclear, colchicine has been extremely effective in reducing the frequency and severity of attacks as well as preventing amyloidosis (2). The gene mutated in FMF was identified in 1997, using positional cloning and codes for a protein called pyrin that is expressed in neutrophils, eosinophils, and monocytes (3, 4). Although the symptoms in FMF and leukocyte expression of pyrin suggest a role for this protein in the regulation of inflammation, the normal function of pyrin has been elusive.
Familial Hibernian fever was initially described in an Irish kindred and has been shown to be partially responsive to corticosteroids. In 1999, mutations in the p55 TNF receptor I (TNFRI) gene were identified in patients with this disease, prompting a new name, TRAPS. McDermott et al proposed an elegant model of impaired shedding of membrane-associated TNFRI that effectively results in excess numbers of membrane-associated receptors and decreased numbers of soluble receptors, leading to increased TNF-mediated inflammation (5). This theory is supported by preliminary studies demonstrating that etanercept, a soluble TNFRI fusion protein that is an effective TNFα antagonist, successfully decreases the severity, duration, and frequency of symptoms in TRAPS (6).
HIDS was initially described in several Dutch patients with recurrent episodes of inflammation and an elevated serum IgD level. In 1999, after a whole genome screen and the finding of increased concentrations of mevalonic acid during severe episodes of fever, HIDS was determined to be attributable to mutations in the mevalonate kinase gene (MVK) (7, 8). Mevalonate kinase (MK) is an enzyme involved in cholesterol biosynthesis, and mutations in this gene are also associated with the more severe metabolic disease, mevalonic aciduria. Although the inflammatory mechanisms involved with MK deficiency are still unclear, the lack of isoprenoid products of MK results in increased IL-1β production (9). HIDS has been resistant to treatment with most antiinflammatory medications, but successful treatment with etanercept was recently reported (10).
Blau syndrome, a granulomatous disease that usually develops in childhood, is corticosteroid responsive. The disease was linked to a large region on chromosome 16, but the gene was not identified until mutation screening was performed on NOD2 (11), a gene recently found also to be associated with susceptibility to Crohn's disease (12, 13). NOD2 is expressed primarily in monocytes and has been postulated to have a role in innate immunity as a detector of intracellular pathogens and an activator of NF-κB, leading to inflammation (14). No directed therapy for Blau syndrome has yet been reported.
Symptoms of MWS and FCAS (commonly known as familial cold urticaria) are similar, and the disorders are distinguished primarily by progressive sensorineural hearing loss and systemic amyloidosis leading to renal failure in patients with MWS, and precipitation of episodes by generalized cold exposure in FCAS patients (15). To date, both conditions have been relatively refractory to treatment. The phenotypic similarities between these 2 disorders had prompted some physicians to suggest that they were genetically related. This suggestion was supported by the finding that both diseases were linked to chromosome 1q44 (16–18) and was later confirmed by the identification of mutations in a single gene that is responsible for both diseases (19–21). This gene, CIAS1, codes for a protein called cryopyrin (also known as NALP3 or PYPAF1). Cryopyrin, like pyrin, is expressed primarily in monocytes and neutrophils (22). A phenotypic overlap was also noted between MWS and NOMID (also known as chronic infantile neurologic, cutaneous, articular syndrome), although symptoms of NOMID are typically more severe and include central nervous system involvement. Because of these similarities, patients with NOMID were screened and also found to have CIAS1 mutations (23, 24). Thus, what were previously thought to be 3 distinct diseases are actually a spectrum of clinical syndromes.
While investigators were identifying the genetic etiologies of the specific inflammatory disorders, several other groups of researchers were scanning the genome for gene families with structural domains similar to known genes involved in apoptosis and inflammation. Two of these domains, the caspase activation recruitment domain (CARD) and the PYRIN domain (named after the protein pyrin, where it was initially identified), are related to death domains seen in important apoptosis proteins such as Apaf-1.
More than 20 genes have been identified that contain these domains in combination with a number of other common domains (14, 25–27). The most well known is NOD2, the gene responsible for Blau syndrome, but a great deal of attention has also been given to another novel protein called ASC, which possesses a PYRIN domain in addition to a CARD domain (28). The ASC protein (also known as PYCARD, CARD5, and TMS1) received special notice when a two-hybrid screen (29) found that it interacts with the pyrin protein. It was later found that ASC also interacts with cryopyrin through PYRIN–PYRIN domain interactions on each protein. Although the exact mechanisms are still unclear, several groups of investigators have shown that when cryopyrin binds to ASC it can result in NF-κB activation, caspase 1 activation, and apoptosis (22, 30–32). Figure 1 summarizes some of these mechanisms. It has also been shown that pyrin disrupts the interaction between ASC and cryopyrin, thereby acting as a negative regulator of cryopyrin (33). This role for pyrin is supported by the finding that in pyrin-deficient mice, caspase 1 activation, and hence IL-1 release, are increased (34). Thus, the pyrin family of proteins regulates the proinflammatory cytokine IL-1.
The IL-1 family of proteins, including IL-1α, IL-1β, and IL-1 receptor antagonist (IL1-Ra), has long been known to have a central role in fever, systemic inflammation, and the pathogenesis of joint disease. IL-1α is thought to be primarily cell-associated, whereas IL-1β is the secreted form. IL-1β is expressed in an inactive cytoplasmic form (pro–IL-1β) and requires cleavage by caspase 1 to create its active form. IL-1Ra is a naturally occurring inhibitor of IL-1 by binding to the IL-1 receptor without inducing an intracellular signal. A recombinant form of IL-1Ra called anakinra has been shown to be efficacious in the treatment of rheumatoid arthritis (RA), presumably by competing for the IL-1 receptor (35). Several studies support involvement of IL-1 in the periodic fever disorders (36, 37). However, its role in the cryopyrin-associated diseases was not suspected until IL-1β protein levels in monocyte lysates and IL-1Ra levels in serum were found to be increased constitutively in a patient with NOMID (24).
In this issue of Arthritis & Rheumatism, less than 2 years after identification of the CIAS1 gene, Hawkins and colleagues report on the use of anakinra to successfully treat 3 patients with MWS, confirming the central role of IL-1 in this disease and providing a potentially life-saving therapy for this group of patients (38). Hawkins et al observed significant improvement in clinical symptoms within hours of the first injection of anakinra, with continued efficacy of daily therapy for up to 3 months. Additionally, they demonstrated a marked and prolonged reduction in serum amyloid A and C-reactive protein levels during treatment; such a reduction has the potential to prevent progression of AA amyloidosis, which commonly leads to renal disease and early death in a significant number of patients with MWS. The fact that anakinra was equally effective at 50% of the daily dose recommended for arthritis suggests that MWS is particularly sensitive to IL-1 blockade and indicates a need for further studies to identify the appropriate dosage for patients with this disorder. The clinical efficacy of anakinra has been modest in RA, which may result from the short half-life of anakinra and the requirement for 10–100-fold excess levels to effectively block the IL-1 receptor. It is also possible that in RA, targeting only IL-1 is not appropriate. In this study, the remarkable efficacy demonstrated in MWS patients may be attributable to the relatively lower levels of IL-1 in this disease, or may have occurred because IL-1 is the key regulator of MWS pathogenesis.
In addition to reporting successful therapy, Hawkins et al also describe several interesting clinical features of this family (38). Although these patients had symptoms of classic MWS, including rash, arthralgia, fever, and hearing loss, they also reported exacerbation of symptoms after cold exposure, a trait commonly seen in FCAS, and they had neurologic and ocular findings characteristic of NOMID. This report and others (39, 40) are making it increasingly clear that the disorders caused by CIAS1 mutations are not distinct diseases, but actually represent a clinical continuum of sub-phenotypes, with FCAS being the mildest, MWS being of intermediate severity, and NOMID being the most severe. It is still not understood how these patients can have such different clinical presentations but can all possess mutations in CIAS1. Interestingly, the same nucleotide substitution in CIAS1 in different patients may be associated with different clinical subtypes (20). This suggests the involvement of additional genes in the clinical phenotype and hence challenges the concept of these conditions as single-gene disorders. To complicate matters more, several patients with classic FCAS or MWS do not have coding mutations in CIAS1, and approximately half of all patients with NOMID do not have obvious CIAS1 mutations, which suggests involvement of additional mutated genes (23, 24). The genetics of these diseases is certainly not as complex as that of the more common rheumatic diseases, but perhaps it is appropriate to describe these diseases as pauci-gene disorders.
All of the genes responsible for these inflammatory disorders are related by their role in innate immunity. TNFα and IL-1 are a few of the characteristic proinflammatory mediators released immediately by monocytes and macrophages in response to pathogen-associated molecular patterns. Toll-like receptors detect extracellular pathogens using their leucine-rich repeat (LRR) domain, whereas the Nod proteins, such as Nod2 and cryopyrin, located in the cytoplasm of cells, are postulated to recognize intracellular pathogens with their LRRs. The involvement of neutrophils, the primary effector cells of innate immunity, also lends support to the theory of a central role for this ancient portion of the immune system in these and other inflammatory diseases. An understanding of the autoinflammatory disorders associated with joint pathology may provide clues to the genes involved in more common and complex autoimmune diseases such as RA. However, the presence of autoantibodies in autoimmune diseases suggests the requirement of additional genes or perhaps environmental exposures in addition to the simple proinflammatory phenotype seen in the autoinflammatory syndromes.
Several questions remain to be answered. What are the genetic or environmental influences that determine the clinical presentation in patients with CIAS1 mutations? Are there additional genes that when mutated result in these disorders? Are there additional cytokines involved that may be reasonable targets for therapy? What is the normal function of cryopyrin and pyrin in inflammation, and how do single-base substitution mutations result in aberrant function? How does environmental exposure to cold in some of these patients lead to inflammation? Is cryopyrin involved in other more common inflammatory diseases? It might be expected that the answers to these questions will not be unraveled by investigators for many years. However, we hope that the rapid progression from gene identification to effective therapy in such disorders is an indication that these and other questions can be answered sooner rather than later.
- 16A locus for familial cold urticaria maps to distal chromosome 1q: familial cold urticaria and Muckle-Wells syndrome are probably allelic [abstract]. Am J Hum Genet 1996; 59: A223., , , , , .
- 21Association of mutations in the NALP3/CIAS1/PYPAF1 gene with a broad phenotype including recurrent fever, cold sensitivity, sensorineural deafness, and AA amyloidosis. Arthritis Rheum 2002; 46: 2445–52., , , , , , et al.
- 24De novo CIAS1 mutations, cytokine activation, and evidence for genetic heterogeneity in patients with neonatal-onset multisystem inflammatory disease (NOMID): a new member of the expanding family of pyrin-associated autoinflammatory diseases. Arthritis Rheum 2002; 46: 3340–8., , , , , , et al.
- 38Spectrum of clinical features in Muckle-Wells syndrome and response to anakinra. Arthritis Rheum 2003; 48: 607–12., , , .