Abstract
In this Perspectives article, we outline a proposed model for understanding the specificity and function of anti-citrullinated protein antibodies (ACPAs). We suggest that ACPAs vary in specificity between two extremes: some are 'promiscuous' in that they are highly specific for the citrulline side chain, but cross-react with a range of citrullinated peptides, whereas others are 'private' in that their recognition of citrulline as well as proximal amino acid side chains enables protein-specific interactions. Promiscuous ACPAs tend to dominate in the sera both before and after the onset of rheumatoid arthritis, but their functional role has not been clarified. No firm evidence exists that these ACPAs are pathogenic. By contrast, private ACPAs encompass antibodies that specifically recognize citrullinated epitopes on joint proteins or that cross-react with joint proteins, thereby opening up the possibility that these private ACPAs are arthritogenic. These joint-reactive antibodies are more likely to target joints by binding to joint tissues and to promote the formation of local immune complexes leading to bone erosions, pain and arthritis.
