NIH Public Access Author Manuscript Arthritis Rheum. Author manuscript; available in PMC 2010 January 1.
NIH-PA Author Manuscript
Published in final edited form as: Arthritis Rheum. 2009 January ; 60(1): 8–11. doi:10.1002/art.24145.
TNF Receptor-Associated Periodic Syndrome (TRAPS): Towards a Molecular Understanding of the Systemic Autoinflammatory Diseases John G Ryan, MB, MRCPI* and Ivona Aksentijevich, MD Genetics and Genomics Branch, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health, 9000 Rockville Pike, Bethesda, MD. 20892
NIH-PA Author Manuscript
Almost a decade has elapsed since the concept of autoinflammation was advanced to describe conditions characterized by ostensibly unprovoked inflammation, without the hightiter autoantibodies or antigen-specific T-cells found in known autoimmune diseases. This concept was introduced following the identification of mutations in the p55 tumor necrosis factor receptor (TNFR1) in patients with a dominantly inherited syndrome of systemic inflammation (TRAPS) (1). Systemic autoinflammatory diseases are caused by aberrant activation of the innate immune system. Some result in recurrent episodes of fever, localized serositis and skin rashes. Two hereditary recurrent fevers, familial Mediterranean fever (FMF) and the TNF receptor-associated periodic syndrome (TRAPS) were prototypes for this diagnostic category. In the past 10 years this concept has been extended to include a number of Mendelian disorders, including other hereditary recurrent fevers, as well as more common diseases with complex modes of inheritance including Behçet's disease and Crohn's disease (2).
NIH-PA Author Manuscript
When the molecular basis of TRAPS was discovered it was considered an exceedingly rare disease with only a few families described in the literature. Over the past decade, more than 50 disease-associated mutations have been identified in hundreds of patients from a variety of ethnic backgrounds (3). Genotype-phenotype studies showed that mutations at cysteine residues are associated with a more severe phenotype and higher incidence of amyloidosis (4). Although, most TRAPS-associated mutations are fully penetrant, two TNFR1 variants, P46L and R92Q, have been also identified in asymptomatic family members, and at a low frequency in healthy populations. Tumor necrosis factor (TNF) is a pleiotropic cytokine that mediates a wide range of cellular activities primarily through its interaction with TNF receptor 1 (TNFR1; TNFRSF1A; CD120a; p55/p60). Triggering the TNF signaling cascade results in an array of responses, including apoptosis, inflammation and modulation of the immune response. TNFR1 is constitutively expressed on the cell surface in most tissues. It is a large polypeptide with an extracellular domain, consisting of 4 cysteine-rich domains (CRD), a transmembrane domain, and an intracellular death domain (DD). Each CRD includes 3 pairs of cysteine residues that form intramolecular disulfide bonds important in maintaining the 3dimensional conformation of the extracellular part of the receptor. Thus, a substitution at a cysteine residue has the potential to disrupt the relevant disulfide bond and result in significant structural perturbation. The first CRD, sometimes called the pre-ligand association domain (PLAD), allows for TNF-independent interactions of TNFR1 molecules,
*Corresponding author. Phone +13014353042 Fax +1 3014802490 Email:
[email protected]
Ryan and Aksentijevich
Page 2
NIH-PA Author Manuscript
regulating cellular responsiveness to TNF. The second CRD contains the ligand binding sites for TNF. Most disease-associated TNFR1 mutations are missense nucleotide changes located in the first two cysteine-rich domains (CRD) of the extracellular region of TNFR1. Binding of trimeric TNF complexes induces trimerization of TNFR1 molecules and enables the DD to recruit downstream signaling molecules. Depending on cell type, the adaptor proteins engaged by the DD may lead to activation of pro-apoptotic, nuclear factor kappa B (NF-κB), or c-Jun N-terminal kinase (JNK) pathways. A series of complex feedback mechanisms may amplify or dampen these signals (5). One means of limiting TNFR1 signaling is by proteolytic cleavage of its extracellular domain by metalloproteinases, releasing a soluble fragment of TNFR1 (sTNFR1), which acts as a decoy receptor and neutralizes serum TNF.
NIH-PA Author Manuscript
Dissecting the pathophysiology of TRAPS has proved especially challenging. The initial report observed that serum sTNFR1 was decreased in patients compared to controls and that activated leukocytes from patients with the C52F mutation had increased TNFR1 cell surface expression and impaired cleavage of TNFR1 (1). The combination of decreased neutralizing sTNFR1, and the potential for prolonged pro-inflammatory signaling through retained active TNFR1 complexes was summarized as the ‘defective shedding’ hypothesis (1). This hypothesis presupposes effective interactions between TNF and mutant TNFR1 receptor. In an effort to restore the balance between TNF and decoy receptor, the use of etanercept, the p75 decoy receptor fusion protein, has proven beneficial in patients with TRAPS. However not all patients respond, perhaps suggesting that ‘defective shedding’ alone may not explain the pathogenesis of TRAPS. Indeed, defective shedding has not been observed in all patients with disease-associated variants and is not seen in all cell types (6-8). This, coupled with reports suggesting that anakinra, an interleukin-1 (IL-1) receptor antagonist, is effective in some patients with TRAPS, supports exploration of alternative means by which mutant TNFR1 may result in disease (9).
NIH-PA Author Manuscript
Subsequent theories regarding the pathogenesis of TRAPS have centered on observations that most TNFR1 mutants do not appear to bind TNF. Multiple cell line constructs and a mouse ‘knock-in’ model have shown decreased cell surface expression of mutant TNFR1 with impaired TNF binding (10-14), retention of mutant TNFR1 within the cytoplasm (12) and aggregate formation within the endoplasmic reticulum (ER) (13). The accumulation of a range of misfolded proteins within the ER is known to induce stress responses prompting the release of proinflammatory cytokines. This process is independent of TNF-TNFR1 interactions, and thus ‘ligand-independent.’ In addition mutant TNFR1 may accumulate within the cytoplasm forming oligomers triggering a range of DD-mediated processes. These processes, while ‘ligand-independent,’ require an intact DD to trigger an inflammatory response. In this issue of Arthritis & Rheumatism Rabelo et al. investigate the pathogenesis of TRAPS from a new perspective (15). Instead of focusing on just one or perhaps 2 aspects of the complex TNF signaling pathway, they have undertaken a study to determine how mutant TNFR1 affects global gene expression. A key benefit of gene expression studies is their unbiased nature, as multiple genes are simultaneously analyzed, with the potential to identify novel means by which mutant TNFR1 results in disease. They used endothelial cell lines, SK-Hep-1, to produce stable transfections with the wild-type (WT) and full length mutant TNFR1 (C33Y, C52F, T50M, R92Q). In an effort to determine the potential role of the ER stress response, which should not require the death domain (DD), they generated inactivated DD mutant TNFR1 constructs. Stringent statistical techniques were used to identify mutant-specific differentially expressed genes (DEGs). The majority of DEGs were found to be up-regulated with a high fold change.
Arthritis Rheum. Author manuscript; available in PMC 2010 January 1.
Ryan and Aksentijevich
Page 3
NIH-PA Author Manuscript
A significant number of DEGs were up-regulated in a DD-dependent mechanism, indicating that the ER stress response does not play a major role in mediating the pathogenesis of TRAPS in those with C33Y, C52F and T50M mutations. Some genes were uniquely induced by cysteine mutants suggesting that these structural mutations may activate specific signaling pathways. Given the reported efficacy of IL-1 blockade in patients with TRAPS it is perhaps surprising that genes encoding either IL-1β or other members of the IL-1 pathway did not feature prominently. Although the role of the R92Q variant is still under discussion, most R92Q-specific induced genes appeared to be up-regulated in a DD-independent manner suggesting that either the ER stress response or other nonspecific pathways are activated in patients with this variant. A significant strength of this paper is the use of an independent set of RNA samples to enable validation of selected genes. However, differences noted at a transcriptome level may not necessarily correlate with protein levels, and do not reflect post-translational modification, such as phosphorylation status. The authors addressed the former concern by quantifying protein levels from both cell lines and patients. However, all patient samples were from patients with C33Y mutations and given that the microarray study showed that altered gene expression was mutation dependent, the degree to which these findings can be generalized is unclear.
NIH-PA Author Manuscript
In ideal circumstances gene expression profiling in any disease would be performed on patient samples. However, controlling for the intrinsic variability in people requires large sample numbers, partially due to differing levels of disease activity and therapies employed. Transfection systems provide inherent stability, thus allowing easier evaluation of the effects of TRAPS-associated mutations. However, even in these stable systems conflicting results may arise as seen in this study with regard to PTX 3. This gene was identified as being induced in a DD-independent manner in gene expression studies, but assays of PTX3 protein levels in cell culture supernatants showed DD-dependent secretion. Over-expression systems introduce an inherent bias, with abnormally high levels of mutant protein and a relative lack of wild type TNFR1. The role that wild type TNFR1 plays in mediating TRAPS is of particular interest. A vicious cycle of pro-inflammatory cytokine production may occur whereby mutant TNFR1 initially triggers ligand independent DD-independent proinflammatory processes. TNF produced as part of this inflammatory response may then feedback via wild type TNFR1 leading to amplified inflammatory signaling. Unfortunately due to a relative lack of wild type TNFR1 expression in existing cell line constructs such complex biologic processes may not be adequately simulated. ‘Knock-in’ animals may however provide appropriate models for further investigation.
NIH-PA Author Manuscript
The insights provided by this report, in conjunction with the accumulated knowledge from other studies suggest that aggregation of mutant TNFR1 receptors in the cytosol results in constitutive activation of ligand-independent, DD-dependent pathways. Mutant cells may exist in a permanent state of cytokine network perturbation, resulting in inflammatory cytokine release. This is consistent with the observation that elevated acute phase reactants are detectable in TRAPS patients even between attacks. It remains possible that during acute TRAPS attacks, ligand-dependent (TNFR1-mediated) signaling becomes prominent leading to vastly amplified inflammatory responses, however data are lacking to support this hypothesis. The investigation of the relative roles of DD-dependent and DD-independent signaling will require detailed analyses of the numerous biochemical pathways activated by either means. Although our current, albeit limited, understanding of the pathogenesis of TRAPS has prompted the use of targeted therapies, we are nonetheless blind to its subtleties. Disease response has varied widely even with use of drugs from the same class of biologic therapies.
Arthritis Rheum. Author manuscript; available in PMC 2010 January 1.
Ryan and Aksentijevich
Page 4
NIH-PA Author Manuscript
While etanercept has been used with benefit, other anti-TNF agents have been reported to induce TRAPS attacks (16). Careful study of more physiologic models of TRAPS, such as animal ‘knock-ins,’ may shed more light on the pathogenesis of TRAPS. A decade of exciting, but conflicting, insights heralds the need for a collaborative project to enable adequately powered human studies to provide definitive answers to this fascinating disorder.
Acknowledgments This research was supported by the Intramural Research Program of the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health.
References
NIH-PA Author Manuscript NIH-PA Author Manuscript
1. McDermott MF, Aksentijevich I, Galon J, McDermott EM, Ogunkolade BW, Centola M, et al. Germline mutations in the extracellular domains of the 55 kDa TNF receptor, TNFR1, define a family of dominantly inherited autoinflammatory syndromes. Cell. 1999; 97(1):133–44. [PubMed: 10199409] 2. Galon J, Aksentijevich I, McDermott MF, O'Shea JJ, Kastner DL. TNFRSF1A mutations and autoinflammatory syndromes. Curr Opin Immunol. 2000; 12(4):479–86. [PubMed: 10899034] 3. Infevers: an online database for autoinflammatory mutations. Copyright. [cited 2008 08/08/2008]; Available from: http://fmf.igh.cnrs.fr/infevers/ 4. Hull KM, Drewe E, Aksentijevich I, Singh HK, Wong K, McDermott EM, et al. The TNF receptorassociated periodic syndrome (TRAPS): emerging concepts of an autoinflammatory disorder. Medicine (Baltimore). 2002; 81(5):349–68. [PubMed: 12352631] 5. Wajant H, Pfizenmaier K, Scheurich P. Tumor necrosis factor signaling. Cell Death Differ. 2003; 10(1):45–65. [PubMed: 12655295] 6. Aganna E, Hammond L, Hawkins PN, Aldea A, McKee SA, van Amstel HK, et al. Heterogeneity among patients with tumor necrosis factor receptor-associated periodic syndrome phenotypes. Arthritis Rheum. 2003; 48(9):2632–44. [PubMed: 13130484] 7. Huggins ML, Radford PM, McIntosh RS, Bainbridge SE, Dickinson P, Draper-Morgan KA, et al. Shedding of mutant tumor necrosis factor receptor superfamily 1A associated with tumor necrosis factor receptor-associated periodic syndrome: differences between cell types. Arthritis Rheum. 2004; 50(8):2651–9. [PubMed: 15334481] 8. Siebert S, Amos N, Fielding CA, Wang EC, Aksentijevich I, Williams BD, et al. Reduced tumor necrosis factor signaling in primary human fibroblasts containing a tumor necrosis factor receptor superfamily 1A mutant. Arthritis Rheum. 2005; 52(4):1287–92. [PubMed: 15818692] 9. Gattorno M, Pelagatti MA, Meini A, Obici L, Barcellona R, Federici S, et al. Persistent efficacy of anakinra in patients with tumor necrosis factor receptor-associated periodic syndrome. Arthritis Rheum. 2008; 58(5):1516–20. [PubMed: 18438813] 10. Todd I, Radford PM, Draper-Morgan KA, McIntosh R, Bainbridge S, Dickinson P, et al. Mutant forms of tumour necrosis factor receptor I that occur in TNF-receptor-associated periodic syndrome retain signalling functions but show abnormal behaviour. Immunology. 2004; 113(1): 65–79. [PubMed: 15312137] 11. Rebelo S, Bainbridge S, Amel-Kashipaz M, Radford P, Powell R, Todd I, et al. Modeling of tumor necrosis factor receptor superfamily 1A mutants associated with tumor necrosis factor receptorassociated periodic syndrome indicates misfolding consistent with abnormal function. Arthritis & Rheumatism. 2006; 54(8):2674–87. [PubMed: 16871532] 12. Siebert S, Fielding CA, Williams BD, Brennan P. Mutation of the extracellular domain of tumour necrosis factor receptor 1 causes reduced NF-kappaB activation due to decreased surface expression. FEBS Lett. 2005; 579(23):5193–8. [PubMed: 16162344] 13. Lobito AA, Kimberley FC, Muppidi JR, Komarow H, Jackson AJ, Hull KM, et al. Abnormal disulfide-linked oligomerization results in ER retention and altered signaling by TNFR1 mutants in TNFR1-associated periodic fever syndrome (TRAPS). Blood. 2006; 108(4):1320–7. [PubMed: 16684962]
Arthritis Rheum. Author manuscript; available in PMC 2010 January 1.
Ryan and Aksentijevich
Page 5
NIH-PA Author Manuscript
14. Todd I, Radford PM, Daffa N, Bainbridge SE, Powell RJ, Tighe PJ. Mutant tumor necrosis factor receptor associated with tumor necrosis factor receptor-associated periodic syndrome is altered antigenically and is retained within patients' leukocytes. Arthritis Rheum. 2007; 56(8):2765–73. [PubMed: 17665435] 15. Rebelo S, Amel-Kashipaz M, Radford PM, Bainbridge SE, Fiets R, Fang J, et al. Novel inflammatory markers identified in TRAPS by transcriptomic analysis of effects of TRAPSassociated TNFR1 mutations in an endothelial cell line. Arthritis & Rheumatism. X;X(X):X-X. 16. Drewe E, Powell R, McDermott E. Comment on: Failure of anti-TNF therapy in TNF Receptor 1Associated Periodic Syndrome (TRAPS). Rheumatology (Oxford). 2007; 46(12):1865–6. [PubMed: 17967816]
NIH-PA Author Manuscript NIH-PA Author Manuscript Arthritis Rheum. Author manuscript; available in PMC 2010 January 1.
