The human leukocyte antigen HLA-B27 is strongly associated with development of a group of inflammatory arthritides collectively known as the spondyloarthritides. We have set out to define the natural immunological function of HLA-B27, and then to apply this knowledge to understand its pathogenic role. Human leukocyte antigen class 1 molecules bind antigenic peptides for cell surface presentation to cytotoxic T lymphocytes. HLA-B27 binds and presents peptides from influenza, HIV, Epstein-Barr virus, and other viruses. This leads to vigorous and specific cytotoxic T lymphocyte responses, which play an important role in the body's immune response to these viruses. HLA-B27 thus carries out its natural function highly effectively. Although many theories have been proposed to explain the role of HLA-B27 in the pathogenesis of spondyloarthropathy, we favour those postulating that the pathogenic role of HLA-B27 stems from its natural function. For example, the 'arthritogenic' peptide hypothesis suggests that disease results from the ability of HLA-B27 to bind a unique peptide or a set of antigenic peptides. Additionally, a number of lines of evidence from our laboratory and other laboratories have suggested that HLA-B27 has unusual cell biology. We have recently demonstrated that HLA-B27 is capable of forming disulfide-bonded homodimers. These homodimers are expressed on the cell surface and are ligands for a number of natural killer and related immunoreceptors, expressed on a variety of cell types including natural killer cells, T lymphocytes and B lymphocytes, and members of the monocyte/macrophage lineage. We are currently investigating the possibility that such interactions could be involved in disease pathogenesis.