Pathogenic Mechanisms in Spondyloarthritis
National Institute Of Arthritis And Musculoskeletal And Skin Diseases
Investigators
Linked publications & trials
Abstract
Objective 1: To define clinical, laboratory (including biomarkers) and imaging features that predict axial involvement in children and young adults with SpA. In initial studies we have analyzed clinical and laboratory features in an inception cohort of ERA patients compared with retrospective cohorts and patients with other subtypes of JIA. Thirty-nine patients satisfying the ILAR criteria for ERA from 5 pediatric rheumatology centers were studied at presentation. Median time from disease onset was 7.3 months;patients were predominantly male (79%) with a median age of 12 years. HLA-B27 was present in 59%, ANA in 41%. Markers of inflammation (ESR/CRP) were not elevated. The median active joint count was 3.5 and active enthesis count was 3.8. Only 1 patient had x-ray evidence of sacroiliitis and 1 had acute anterior uveitis. Interestingly, serum TGF-βlevels were elevated (p<0.01), while IL-22 and GM-CSF were decreased (p<0.04), relative to healthy controls. Other Th17 and Th1 cytokines were unchanged. TGF-βlevels were higher in patients with symptoms of gastrointestinal distress. Consistent with TGF-βelevation, our previous studies have shown evidence of a TGF-βgene expression signature in the peripheral blood of these subjects. We are currently analyzing genes whose differential expression correlates the strongest with TGF-βlevels and thus best represent this signature, so that a more refined signature can be followed in a longitudinal analysis. In collaboration with Pam Weiss (CHOP), we have found that being classified with ERA, and/or having enthesitis, is a consistent predictor for high CHAQ (Childhood Health Assessment Questionnaire), HRQOL (Health Related Quality of Life), and VAS pain scores in a large cohort of JIA patients (2,571) enrolled in the CARRANet Registry, reflecting the health burden of this disease. In collaboration with Lianne Gensler (UCSF) we are analyzing biomarkers associated with axial SpA in adults. We plan to apply what we have learned from these studies to early disease in children. In a study of 160 subjects, we found that matrix metalloproteinase-3 (MMP-3) is elevated in AS (n=110) compared to control groups, including individuals with undifferentiated SpA (n=37;p<0.05) and mechanical back pain (n=13;p<0.02), even when adjusted for age, gender, ethnicity, disease duration, HLA-B27 status, and CRP levels. Interestingly, analyzing the 147 SpA patients we found that bone alkaline phosphatase (BAP), a marker of bone formation, is significantly higher in individuals with grade IV sacroiliitis (vs. <IV) (p<0.03, CI=1.01-1.1), which represents complete fusion of the sacroiliac joints. These results suggest that MMP-3 and BAP may be useful markers of severity and bone formation in SpA. Objective 2: To determine the role of HLA-B27 in the development of SpA. Previous work has shown that HLA-B27 misfolding can activate the unfolded protein response (UPR) in cells from HLA-B27 and human β2m transgenic rats (HLA-B27 transgenic rats). These animals develop spondyloarthritis-like disease and thus serve as an experimental model to explore the role of HLA-B27 in pathogenesis. We have shown that the IL-23/IL-17 axis is strongly activated in the colon lamina propria of HLA-B27 transgenic rats, and that HLA-B27 misfolding and UPR activation can result in enhanced production of the IL-17-inducing cytokine, IL-23, suggesting a link between HLA-B27 misfolding and immune dysregulation in this model. The discovery that IL23R polymorphisms are associated with predisposition to ankylosing spondylitis and inflammatory bowel disease in genome-wide association studies, suggests potential mechanistic links between HLA-B27 and IL-17 production. Since moving our HLA-B27 transgenic rat colony to the NIH we have discovered an interesting diet-related difference in colitis in these animals. The 'NIH diet'seems to delay and/or reduce the severity of colitis compared to the standard Purina diet used previously. We are currently documenting and quantitating these differences, and will be exploring whether they are associated with differences in the microbiome and Th17 development and activation. In addition, we have developed siRNA that reduces HLA-B27 expression and are planning to use this to explore the extent to which knocking down HLA-B27 expression before, during or after the inflammatory phenotype develops can ameliorate UPR activation and its sequelae and disease. Objective 3: To explore the effects of ERAP1 on the immunobiology of HLA-B27. ERAP1 encodes an ER aminopeptidase that is involved in trimming peptides that are destined for presentation by MHC class I molecules. Recently, a gene-gene interaction has been demonstrated for ERAP1 in susceptibility to AS, where ERAP1 is a risk gene only in HLA-B27 positive individuals. It appears that disease-related polymorphisms may result in loss of function, however, their role in pathogenesis and specifically their effect on HLA-B27 remain unexplored. We have initiated studies to examine the role of ERAP1 on the immunobiology of HLA-B27 with basic loss-of-function/gain-of-function studies. Since we are able to study the role of HLA-B27 in the transgenic rat model, we are screening rat ERAP1 knockdown constructs (shRNA). We have successfully knocked down ERAP1 expression in Rat1 fibroblasts, and are currently engineering our shRNA sequences into a lentiviral construct in order to perform studies in primary cells from HLA-B27 transgenic rats. Effects of ERAP1 under and overexpression on the folding/misfolding of HLA-B27, and how this effects UPR activation, will be examined in detail once these constructs are available. Parallel studies will be performed in human cells. Objective 4: To determine the effects of susceptibility genes on osteoclast and osteoblast development and function. One of the most poorly understood aspects of spondyloarthritis pathogenesis is bone loss juxtaposed with abnormal bone formation, which results in fusion of scaroiliac and facet joints, and the formation of syndesmophytes bridging adjacent vertebral bodies. To determine the effects of susceptibility genes on osteoclast (OC) and osteoblast (OB) development and function we have begun examining these cells differentiated from bone marrow precursors in HLA-B27 transgenic rats. The discovery that HLA-B27 misfolding in monocytes/macrophages can activate the unfolded protein response (UPR) suggests that this protein may affect cellular function beyond its canonical role as an antigen-presenting molecule. Indeed, under the influence of TNF-α, HLA-B27 expression promotes OC development from bone marrow monocytes 2-3 fold (p<0.05;CI=2.16-2.94) compared to wild type (WT) rats, whereas no effect is seen with RANKL alone. TNF-αtreatment upregulates HLA-B27 expression, exacerbates misfolding, and activates the UPR, whereas RANKL has no such effect. Key roles for IL-1αand IFN-βin mediating the HLA-B27 effect on OC formation in TNF-α-treated cells have been established with neutralizing antibodies. Interestingly, neutralization of IL-1αcompletely abrogates the increased OC formation, whereas neutralization of IFN-βfurther enhances OC formation. The net effect under the conditions studied is pro-osteoclastogenic, and may contribute to the bone loss seen in these rats. However, other factors that induce IFN-βcould shift this balance toward a net anti-osteoclastogenic effect, which is of interest given the pro-osteogenic phenotype seen in certain anatomic sites of individuals with AS. Effects of HLA-B27 on OB formation are currently being examined.
View original record on NIH RePORTER →