Stanford University San Francisco, CA, United States
Alejandro Gomez1, Camille Brewer1, Jae-Seung Moon2, Suman Acharya1, Tobias V. Lanz1, Qian Wang1, Gundula Min-Oo3, Anita Niedziela-Majka3 and William Robinson4, 1Stanford University, Stanford, CA, 2Stanford University, Palo Alto, CA, 3Gilead Sciences, Foster City, CA, 4Stanford University School of Medicine, Palo Alto, CA
Background/Purpose: ACPAs are present in two-thirds of patients with rheumatoid arthritis (RA) and are associated with higher risks for severe bone erosions. The pathogenic role of ACPAs in joint inflammation has been extensively studied, primarily in animal models, with mixed results. The deposition of immune complexes (ICs) of ACPAs and citrullinated antigens (citAgs) in affected joints is largely considered as pro-inflammatory. One of the primary sources of citAgs in inflamed joints are neutrophil extracellular traps (NETs) released by activated neutrophils. Binding of ACPAs to citAgs exposed in NETs could generate ICs that accumulate in joints and activate infiltrating immune cells. In this study, we investigated how recombinant ACPAs derived from RA patients affected the course and severity of CAIA, an acute-phase murine model of RA that engages primarily the innate immune system and relies heavily on neutrophil activation.
Methods: We administered "mouse-ized" versions of patient-derived ACPAs to CAIA mice at 4 different stages of the model: 1, Induction (Day 0); 2, Acute phase (Days 3 & 7); 3, Maximum symptoms (Day 10); and 4, Resolution (Day 13). We also conducted in vivo imaging experiments to assess biodistribution and clearance of ACPAs. CAIA mice were injected with fluorescently-labeled antibodies and a fluorescent probe specific for neutrophil elastase (NE), a proteolytic enzyme released by activated neutrophils that co-localizes with NETs, and were imaged for up to 5 days.
Results: Co-injection of ACPAs alongside the antibody cocktail that induces CAIA fully prevented the development of paw inflammation. Furthermore, ACPAs injected at early stages of the model (d3 & d7) significantly ameliorated clinical symptoms compared to control groups. Paw thickness in ACPA-injected groups was reduced to wild-type control levels at endpoint (d14). Similarly, ACPA injections at d10 (max symptoms) significantly ameliorated paw inflammation compared to controls, though to a lesser extent than when ACPAS were injected earlier in the model. Finally, ACPAs injected during the resolution phase of CAIA (d13) did not affect either the speed or the degree of recovery from inflammation. Biodistribution experiments with fluorescently-labeled antibodies showed that both ACPAs and isotype control accumulated preferentially in inflamed paws, and that NE followed the same pattern. ACPAs, however, were cleared faster from paws than isotype controls, and endpoint levels of NE in paws were significantly reduced only in ACPA-injected animals. Additionally, systemic clearance of NE, particularly by the liver, was significantly enhanced only in ACPA-injected mice.
Conclusion: Contrary to our expectations, injection of recombinant ACPAs at early stages of CAIA both prevented and ameliorated paw inflammation. ACPAs, however, did not affect the normal course of CAIA when injected at later stages. In vivo imaging studies with labeled antibodies and a probe for NE (marker for NETs) showed enhanced clearance of NE only in ACPA-injected CAIA mice. These findings suggest that clearance of NETs from inflamed joints, potentially by ICs formation between ACPAs and NET-bound citAgs, might contribute to the amelioration of paw inflammation.
Disclosures: A. Gomez, None; C. Brewer, None; J. Moon, None; S. Acharya, None; T. Lanz, None; Q. Wang, None; G. Min-Oo, Gilead; A. Niedziela-Majka, Gilead; W. Robinson, None.
