Animal models are an important tool for dissection of pathways and mechanisms involved in inflammatory arthritis. Recently, significant interest has focused on the K/BxN mouse model, in which a spontaneously erosive arthritis with many similarities to human rheumatoid arthritis and other forms of inflammatory arthritides occurs. The disease in these mice is dependent on the major histocompatibility complex class II allele-specific recognition of a peptide derived from the ubiquitously expressed self-protein, glucose-6-phosphate isomerase (GPI), by T cells that escape negative selection. However, interestingly, the disease is caused by arthritogenic autoantibodies to GPI, and passive transfer of these antibodies to naı̈ve mice is sufficient to induce disease. Recently, careful dissection of the downstream pathways involved in arthritis in the K/BxN mouse has yielded significant insight into disease mechanisms that might be involved in human arthritis. These features will be discussed in this review, as well as the role of GPI as a potential autoantigen in human inflammatory arthritis.