Horror autoinflammaticus: the molecular pathophysiology of autoinflammatory disease (*)

Abstract

The autoinflammatory diseases are characterized by seemingly unprovoked episodes of inflammation, without high-titer autoantibodies or antigen-specific T cells. The concept was proposed ten years ago with the identification of the genes underlying hereditary periodic fever syndromes. This nosology has taken root because of the dramatic advances in our knowledge of the genetic basis of both mendelian and complex autoinflammatory diseases, and with the recognition that these illnesses derive from genetic variants of the innate immune system. Herein we propose an updated classification scheme based on the molecular insights garnered over the past decade, supplanting a clinical classification that has served well but is opaque to the genetic, immunologic, and therapeutic interrelationships now before us. We define six categories of autoinflammatory disease: IL-1beta activation disorders (inflammasomopathies), NF-kappaB activation syndromes, protein misfolding disorders, complement regulatory diseases, disturbances in cytokine signaling, and macrophage activation syndromes. A system based on molecular pathophysiology will bring greater clarity to our discourse while catalyzing new hypotheses both at the bench and at the bedside.

Figure 1

Schematic representation of mutations in five proteins that cause autoinflammatory diseases. (a) For pyrin, mutations that are most frequently found to cause FMF are presented in red, while those in black are less common disease-causing variants. Residues in purple are found at approximately 1% allele frequency in the general population and may therefore represent functional polymorphisms. (b) Cryopyrin mutations cause a spectrum of disease states that range in severity from severe (NOMID/CINCA, red ), to intermediate (MWS, blue), to mild (FCAS, purple). In some instances, it is difficult to distinguish between NOMID/CINCA and MWS. Residues for which disease presentation overlaps are depicted in black. (c) NOD2/CARD15 mutations can cause Blau syndrome/early-onset sarcoidosis (blue), which cluster within the NACHT/NBD and NAD domains. Other variants that are spread throughout this protein are associated with Crohn’s disease (red ). (d ) Mevalonate kinase mutations can cause the severe metabolic disease mevalonic aciduria (red ), or the less severe autoinflammatory disease HIDS (blue). HIDS mutations are recessively inherited and often include one mild mutation coupled with a severe mutation; thus, the severe mutations can be found in both HIDS and mevalonic aciduria (black). (e) Dominantly inherited missense mutations in TNFR1 that cause TRAPS are now known to affect almost every cysteine residue within the first two cysteine-rich domains of the extracellular region of the protein (red ). These mutations appear to affect the protein folding, whereas at least two mutations, I170N and V173D, can affect ectodomain cleavage that generates the soluble form of the receptor. P46L and R92Q ( purple) are probably functional polymorphisms also present in unaffected individuals. The numbering system for TNFR1 used here begins at residue 30, which is at the N terminus after removal of the 29-residue signal peptide, as per common convention. Abbreviations used: PYD, pyrin domain; NBD, nucleotide-binding domain; NAD, NACHT-associated domain; LRR, leucine-rich repeats; CRD, cysteine-rich domain; TM, transmembrane domain.

Figure 1

Schematic representation of mutations in five proteins that cause autoinflammatory diseases. (a) For pyrin, mutations that are most frequently found to cause FMF are presented in red, while those in black are less common disease-causing variants. Residues in purple are found at approximately 1% allele frequency in the general population and may therefore represent functional polymorphisms. (b) Cryopyrin mutations cause a spectrum of disease states that range in severity from severe (NOMID/CINCA, red ), to intermediate (MWS, blue), to mild (FCAS, purple). In some instances, it is difficult to distinguish between NOMID/CINCA and MWS. Residues for which disease presentation overlaps are depicted in black. (c) NOD2/CARD15 mutations can cause Blau syndrome/early-onset sarcoidosis (blue), which cluster within the NACHT/NBD and NAD domains. Other variants that are spread throughout this protein are associated with Crohn’s disease (red ). (d ) Mevalonate kinase mutations can cause the severe metabolic disease mevalonic aciduria (red ), or the less severe autoinflammatory disease HIDS (blue). HIDS mutations are recessively inherited and often include one mild mutation coupled with a severe mutation; thus, the severe mutations can be found in both HIDS and mevalonic aciduria (black). (e) Dominantly inherited missense mutations in TNFR1 that cause TRAPS are now known to affect almost every cysteine residue within the first two cysteine-rich domains of the extracellular region of the protein (red ). These mutations appear to affect the protein folding, whereas at least two mutations, I170N and V173D, can affect ectodomain cleavage that generates the soluble form of the receptor. P46L and R92Q ( purple) are probably functional polymorphisms also present in unaffected individuals. The numbering system for TNFR1 used here begins at residue 30, which is at the N terminus after removal of the 29-residue signal peptide, as per common convention. Abbreviations used: PYD, pyrin domain; NBD, nucleotide-binding domain; NAD, NACHT-associated domain; LRR, leucine-rich repeats; CRD, cysteine-rich domain; TM, transmembrane domain.

Figure 2

Mechanisms of autoinflammatory disease regulated by IL-1β (inflammasomopathies) and NF-κB. The NLRP3 protein interacts with ASC and caspase-1 to form a complex termed the inflammasome, which is a macromolecular complex that processes IL-1β into its active form. Mutations (denoted by asterisks) in proteins that affect the function of this complex such as pyrin, PSTPIP1, mevalonate kinase (MK), and NLRP7 thus represent extrinsic inflammasomopathies. Mutations in NLRP3 clearly activate the molecule; however, the precise mechanism by which pyrin and PSTPIP1 mutations regulate inflammasome activity has not been determined. As NLRP7 is a negative regulator of IL-1β production, these mutations are likely to inactivate the protein, and it has been shown that MK mutations are also loss of function, although the pathway that leads from these mutations to Rac1/PI3K/PKB activation is not yet formally described. Preliminary data from our laboratory suggest that mutations in the IL-1 receptor antagonist (IL-1Ra) cause another extrinsic inflammasomopathy. For NF-κB activation disorders, NOD2 could become activated by mutations that may relieve autorepression of the molecule; however, there are also persuasive data showing that mutations inactivate the protein from tolerizing the host to the bacterial cell wall component muramyl dipeptide (MDP). NLRP12 is a negative regulator of NF-κB, and thus a proinflammatory disease would be caused by what appear to be loss-of-function mutations. Abbreviations used: CPPD, calcium pyrophosphate dihydrate; Aβ, amyloid-β; ROS, reactive oxygen species; PKB, protein kinase B.

Figure 3

The evolution and structural ramifications of ancestral mutations in pyrin. (a) This figure highlights both that pyrin mutations are very common in several Mediterranean populations and that it is not the same mutation that is expanded in each population. Blue and red lines indicate the migration of populations carrying the M649V and V726A mutations, respectively, while the E148Q mutation ( purple) is common to countries along the medieval trade route (the Silk Road) that extended from the Middle East through to Japan. Pie diagrams indicate the carrier frequencies of selected mutations in Ashkenazi Jewish, Sephardi Jewish, and Japanese populations. (b) A model for the structure of the B30.2/SPRY domain of pyrin (based on that of TRIM21) indicates the preponderance of FMF mutations (cyan) affecting one side of the molecule that is thought to present a protein-binding pocket. A phenotype-genotype correlation is proposed that links the more severe disease-causing mutations (M694V blue, M680I green) to the putative binding site, whereas a less severe mutation (V726A red ) is located further away from this region.

Figure 4

Mechanisms of autoinflammatory protein folding disorders. Left: TNF-receptor associated periodic syndrome (TRAPS). TNF-receptor type 1 (TNFR1) accumulates in the endoplasmic reticulum (ER) when mutated (indicated by asterisk). This accumulation leads to increased reactive oxygen species (ROS) activation, and subsequent MAPK phosphorylation ( JNK and p38), which makes the cells more susceptible to inflammatory stimuli. TRAPS patients, who are heterozygous for TNFR1 mutations, still carry the wild-type TNFR1, which is thought to play a role in propagating the inflammatory cascade. Right: spondyloarthropathies, ankylosing spondylitis. An HLA-B27 variant is strongly associated with ankylosing spondylitis and may accumulate in the ER, leading to an unfolded protein response. Weaker gene associations include ARTS1 (ER-associated aminopeptidase 1, ERAP1), which could affect the pathogenic presentation of antigenic peptide fragments or the cleavage of membrane bound receptors, and IL-23R/IL-1α, which argue for the role of proinflammatory cytokines in this disease.

Figure 5

Pathogenesis of the complement-mediated autoinflammatory disease atypical hemolytic uremic syndrome. Activating gain-of-function mutations in CFB increase C3bBb convertase stability and lead to permanent activation of the alternative pathway of complement. Autoantibodies against, and inactivating mutations within negative regulators of this process have also been described, such as for complement factor H (CFH) and membrane cofactor protein (MCP), which are cofactors for the inactivation of C3b by complement factor I (CFI). Asterisks denote proteins that carry mutations with known disease associations.

Figure 6

Autoinflammatory disease caused by aberrant cytokine signaling. Osteoclast and macrophage differentiation is controlled by the cytokines M-CSF and RANKL. Dominantly inherited mutations in SH3BP2 alter the signals that normally come from these cytokines, which results in the differentiation of hyperactive osteoclasts and macrophages that can cause the autoinflammatory disease cherubism. Excess TNF-α produced by these hyperactive macrophages could also feed back through the stroma to increase M-CSF and RANKL levels, although this is still hypothetical.

Figure 7

Molecular lesions that affect cytotoxic adaptive immune system cells lead to unregulated macrophage activation. Familial hemophagocytic lymphohistiocytosis (HLH) is caused by mutations in genes encoding proteins associated with vesicular transport and granule shedding, such as Munc13-4 (UNC13D), Perforin 1 (PERF1), and Syntaxin11 (STX11). The functions of NK and cytotoxic T cells are also controlled by several other genes that cause individual monogenic diseases when mutated (LYST, RAB27A, SH2D1A, AP3B1). Impairment of the normal efficacy of cytotoxic T cells for virally infected targets and upregulation of IFN-κ production leads to the compensatory development of a subset of scavenger macrophages that mediate the macrophage activation syndrome (MAS). They secrete proinflammatory cytokines, phagocytose erythrocytes, express the CD163 receptor for haptoglobin-hemoglobin complexes, and produce ferritin. Their production of IL-18 in particular could amplify the production of IFN-κ in a positive feedback loop. Secondary HLH is most commonly observed as a sequella of systemic-onset juvenile arthritis, for which variants of IL-6 and MIF are risk factors (double asterisks). A clinical hallmark of MAS is profound hyperferritinemia.