The disease modifying anti-rheumatic drugs belong to a chemically heterogeneous group that includes inorganic salts (e.g. gold thiolates), plant derived products (e.g. chloroquine) and a drug metabolite (e.g. D-penicillamine). Understanding their mechanisms of action has long been clouded by numerous and conflicting theories. A wide variety of unpredictable adverse effects, extremely long half lives and lack of dose-response and concentration-response relationships have added to the confusion. Although they have no known high affinity receptors that might provide a clue to their mechanisms of action, the observation that gold(I) salts and penicillamine are reactive with thiols has suggested that there might be a mechanism in common between these two drugs. Recent advances in the function and chemistry of proteins involved in gene expression have indicated that thiol groups in the 'pro-inflammatory' transcription factors AP-1 and NF-κB are targets for at least some of the therapeutic effects of disease modifying anti-rheumatic drugs. Developments in understanding the transcriptional effects of glucocorticoid and retinoid receptors indicate that they too act, at least in part, via inhibition of AP-1 and/or NF-κB activities. It is possible to develop a unifying hypothesis for the actions of this seemingly diverse group of drugs and make testable predictions for the development of more specific, and less toxic anti-inflammatory therapies.