In most patients, antibiotic treatment of Lyme arthritis (LA) combined with host immunity results in spirochetal elimination, tissue repair and resolution of arthritis, called antibiotic-responsive LA. However, despite spirochetal killing, some patients develop an inflammatory, proliferative synovitis lasting months to several years after antibiotic therapy, called post-infectious (antibiotic- refractory) LA. The synovial lesion in these patients is similar with that in other forms of chronic inflammatory arthritis, including rheumatoid arthritis. The central feature in setting the stage for post-infectious LA seems to be a sustained, excessive pro-inflammatory response, with exceptionally high levels of IFN-, which overwhelms regulatory control mechanisms. In these patients, persistent inflammation leads to further vascular damage, fibrosis, and autoimmune phenomena in synovia, as seen in other forms of chronic inflammatory arthritis. We have shown that this response may be accompanied by Lyme disease (LD)-associated autoantibodies, a response that becomes T cell dependent and potentially pathogenic. In these patients, IgG4 autoantibodies in synovial fluid, but not antibodies of other subclasses, correlate with specific synovial pathology – obliterative microvascular lesions and fibrosis. Moreover, obliteration of blood vessels suggests cytotoxic immune responses, and our recent transcriptomic studies of post-infectious LA synovial tissue show a strong cytotoxic gene signature, similar with that in RA. In this grant, we will use a “systems serology” approach to understand functions of Borrelia burgdorferi (Bb) antibodies and LD autoantibodies by delineating Fc receptor binding, binding affinity, Fc glycans, antibody-dependent phagocytosis and antibody-dependent cytotoxicity in antibiotic-responsive vs. post-infectious LA patients. In addition, we will define the types and frequencies of lymphocytes with cytotoxic and inflammatory potential in peripheral blood and synovial fluid in these two patient groups using flow cytometry, and will further delineate the complete phenotype of implicated cytotoxic cells by single-cell RNA-sequencing. We will correlate these data with determinations of cytokine, granzyme, and perforin levels in vivo in serum and synovial fluid of these patients, as measured by Luminex. Finally, we will examine the cytotoxic potential of implicated T cells in directly killing target cells, and indirectly by asessing the effects of extracellular cytotoxic effector molecules on cells. This research has significant implications for understanding pathogenesis and aiding in diagnosis and treatment of LA, and more generally, as a human model of infection-induced chronic inflammatory arthritis.