NLRP1 Haplotypes Associated with Vitiligo and Autoimmunity Increase Interleukin-1Β Processing Via the NLRP1 Inflammasome
Total Page:16
File Type:pdf, Size:1020Kb
NLRP1 haplotypes associated with vitiligo and autoimmunity increase interleukin-1β processing via the NLRP1 inflammasome Cecilia B. Levandowskia, Christina M. Maillouxa, Tracey M. Ferraraa, Katherine Gowana, Songtao Bena, Ying Jina,b, Kimberly K. McFannc, Paulene J. Hollanda, Pamela R. Faina,b,d, Charles A. Dinarelloe,1, and Richard A. Spritza,b,1 aHuman Medical Genetics and Genomics Program and bDepartment of Pediatrics, University of Colorado School of Medicine, Aurora, CO 80045; cColorado Biostatistics Consortium, Colorado School of Public Health, Aurora, CO 80045; and dBarbara Davis Center for Childhood Diabetes, and eDepartment of Medicine, University of Colorado School of Medicine, Aurora, CO 80045 Contributed by Charles A. Dinarello, January 8, 2013 (sent for review December 26, 2012) Nuclear localization leucine-rich-repeat protein 1 (NLRP1) is a key associated molecular patterns stimulate surface Toll-like recep- regulator of the innate immune system, particularly in the skin tors (TLRs) with subsequent assembly of the NLRP1 inflam- where, in response to molecular triggers such as pathogen- masome and activation of caspase-1. Active caspase-1 then associated or damage-associated molecular patterns, the NLRP1 cleaves the inactive IL-1β precursor (pro–IL-1β) to the mature inflammasome promotes caspase-1–dependent processing of bio- bioactive IL-1β (23), thereby stimulating downstream inflam- active interleukin-1β (IL-1β), resulting in IL-1β secretion and down- matory responses (23, 24). Because of its highly inflammatory stream inflammatory responses. NLRP1 is genetically associated and immunostimulating properties, the processing and release of with risk of several autoimmune diseases including generalized bioactive IL-1β are tightly controlled. However, in cryopyrin- vitiligo, Addison disease, type 1 diabetes, rheumatoid arthritis, associated periodic syndromes (CAPSs), which include familial and others. Here we identify a repertoire of variation in NLRP1 cold autoinflammatory disease, Muckle–Wells syndrome, and by deep DNA resequencing. Predicted functional variations in neonatal onset multi-inflammatory disease, amino acid sub- NLRP1 reside in several common high-risk haplotypes that differ stitutions in the NACHT domain of NLRP3 result in constitutive from the reference by multiple nonsynonymous substitutions. The activation of caspase-1 and processing and secretion of bioactive haplotypes that are high risk for disease share two substitutions, IL-1β, accompanied by marked systemic and local inflammation L155H and M1184V, and are inherited largely intact due to exten- (21, 25–28). Blood monocytes from affected patients release sive linkage disequilibrium across the region. Functionally, we significantly more mature bioactive IL-1β than cells from healthy found that peripheral blood monocytes from healthy subjects ho- individuals (28). Similarly, a dominant activating mutation in mozygous for the predominant high-risk haplotype 2A processed mouse Nlrp1a has recently been shown to result in a severe sys- significantly greater (P < 0.0001) amounts of the IL-1β precursor temic inflammatory phenotype associated with greatly elevated to mature bioactive IL-1β under basal (resting) conditions and in release of active IL-1β (29). response to Toll-like receptor (TLR) agonists (TLR2 and TLR4) com- In the present study we have investigated the hypothesis that pared with monocytes from subjects homozygous for the refer- common inherited variation in human NLRP1, which is associ- ence haplotype 1. The increase in basal release was 1.8-fold ated with genetic risk of autoimmune disease, similarly results in greater in haplotype 2A monocytes, and these differences be- constitutively increased processing and release of bioactive IL-1β. tween the two haplotypes were consistently observed three times By next-generation DNA sequencing, we identified common over a 3-mo period; no differences were observed for IL-1α or NLRP1 haplotypes containing multiple predicted functional var- TNFα. NLRP1 RNA and protein levels were not altered by the pre- iants that are associated with disease. We next studied the in vitro dominant high-risk haplotype, indicating that altered function of production of cytokines from monocytes of healthy individuals the corresponding multivariant NLRP1 polypeptide predisposes to homozygous for the most prevalent high-risk NLRP1 haplotype autoimmune diseases by activation of the NLRP1 inflammasome. and compared the responses to monocytes from subjects with the low-risk reference haplotype. Our findings indicate that these cytokine | DNA sequencing | interleukin-1beta multivariant NLRP1 haplotypes predispose to common autoim- mune diseases by activation of the NLRP1 inflammasome. utoimmune diseases are a diverse group of more than 80 Achronic disorders in which the immune system attacks “self” Results tissues and cells (1), altogether affecting 3–5% of the U.S. pop- NLRP1 Sequencing Identifies Conserved Disease-Associated Haplotypes ulation (2). Concomitant occurrence of multiple autoimmune That Differ by Multiple Nonsynonymous Substitutions. To identify diseases in some patients is well known (3–5) and results from genetic variation in NLRP1 that might contribute to autoimmune shared genetic and perhaps environmental risk factors among susceptibility, we resequenced the complete NLRP1 gene region these different diseases (6–8). Polymorphisms of the nuclear lo- in probands of the 114 families with multiple cases of vitiligo and calization leucine-rich-repeat protein 1 (NLRP1) gene encoding other autoimmune diseases in which we originally established the NACHT (NAIP, CIITA, HET-E, TP1) domain of NLRP1 genetic linkage and association with NLRP1 (9, 10). The 161.5 kb have been associated with a number of autoimmune diseases, including vitiligo (9, 10), Addison disease (11, 12), type 1 di- abetes (11), celiac disease (13), systemic lupus erythematosus Author contributions: P.R.F., C.A.D., and R.A.S. designed research; C.B.L., C.M.M., T.M.F., (14), rheumatoid arthritis (15), systemic sclerosis (16), Kawasaki S.B., and Y.J. performed research; C.M.M., T.M.F., and P.J.H. contributed new reagents/ fl analytic tools; C.B.L., C.M.M., T.M.F., K.G., S.B., Y.J., K.K.M., P.R.F., C.A.D., and R.A.S. disease (17), as well as steroid responsiveness in in ammatory analyzed data; and C.B.L., K.K.M., C.A.D., and R.A.S. wrote the paper. fi bowel disease (18), although the speci c biological mechanism The authors declare no conflict of interest. underlying these genetic associations remains unknown. 1To whom correspondence may be addressed. E-mail: [email protected] or richard. NLRP1 is a regulator of the innate immune response (19–21) [email protected]. and is expressed in many immunocompetent cell types, particu- This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. larly the Langerhans cells of the skin (22). Pathogen- or damage- 1073/pnas.1222808110/-/DCSupplemental. 2952–2956 | PNAS | February 19, 2013 | vol. 110 | no. 8 www.pnas.org/cgi/doi/10.1073/pnas.1222808110 Downloaded by guest on September 23, 2021 region sequenced (chr17:5386564–5548062) included the entire otherwise specific to haplotype 3. In addition to these major 70.4 kb NLRP1 structural gene, as well as 60.2 kb upstream and multivariant haplotypes, we also observed four singleton non- 30.9 kb downstream. We observed a total of 656 sequence variants synonymous substitutions, G354D, E869K, T1113M, and R1289H, (Dataset S1), including 261 not annotated in the Short Genetic of which only T1113M cosegregated with autoimmunity in the Variations database (dbSNP), Build 135. Within the exons, these corresponding multiplex family. Of the 16 observed non- included 16 nonsynonymous variants, with no observed nonsense synonymous substitutions, bioinformatic analyses predict that variants, splice junction variants, or coding region indels. We six are probably or possibly deleterious (Fig. 1), although in- verified all previously unannotated nonsynonymous variants by dividual in silico functional predictions are of uncertain val- Sanger sequencing in the corresponding patients, and we in- idityinthecontextofsuchmultivariant haplotypes. Of these, tegrated the observed sequence variants with high-density NLRP1 L155H is the only predicted deleterious substitution shared by region SNP data in these same families (9). both high-risk haplotypes 2A and 3, which additionally share The NLRP1 region consists of three main haplotype blocks, the predicted benign variant M1184V. comprising the NLRP1 structural gene, the 5′-untranslated se- quence and extended promoter region, and the 3′-untranslated Increased Processing of Mature IL-1β from Monocytes of Healthy and 3′- flanking region. Of the 16 nonsynonymous variants, 12 Subjects with the Prevalent High-Risk NLRP1 Haplotype. To de- were genotyped directly or were sufficiently frequent to permit termine whether the multivariant NLRP1 polypeptides encoded accurate genotype imputation (r2 > 0.3) in the corresponding by disease-associated NLRP1 haplotypes differentially affect families. Previous family-based association analyses showed that NLRP1 function, we screened healthy individuals to identify rs12150220 (L155H) was the single variant most significantly subjects alternatively homozygous for the NLRP1 reference associated with vitiligo (9). However, the L155H variant occurs haplotype 1 or for the prevalent high-risk NLRP1 haplotype 2A. on three main haplotypes, each differing by multiple additional We genotyped a total of 115