The abnormality in myasthenia gravis (MG) is a deficiency of acetylcholine receptors (AChRs) at neuromuscular junctions due to an antibody-mediated autoimmune attack. Although immunosuppressive drugs are usually beneficial in MG, they produce generalized suppression of the immune system. Treatment should specifically inhibit the immune response to AChR. Oral administration of an antigen may induce specific tolerance and has recently been tested for treatment of several cell-mediated experimental and human autoimmune diseases. In this study, we investigated the effects of oral administration of AChR in an experimental rat model of MG (EAMG), which is antibody mediated. Lewis rats were fed various doses of purified or unpurified Torpedo AChR, or a control antigen, ovalbumin (OVA). They were then immunized with AChR or OVA. We measured antibody responses to Torpedo AChR or OVA, autoantibody responses to rat AChR, cellular responses, cellular suppressive effects, and clinical status. Our results showed that AChR feeding prevented clinical signs of EAMG. It initially primed, then inhibited, antibody responses to foreign (Torpedo) AChR and self (rat) AChR, with a delayed onset. Cellular responses to AChR, measured by lymphocyte proliferation and interleukin-2 production, were markedly inhibited. The effects were dose dependent. Unpurified AChR given in comparable amounts was far less effective than pure AChR. OVA feeding had similar, but even more potent effects on humoral and cellular immune responses to OVA, but did not inhibit clinical EAMG or AChR responses. Moderate nonspecific suppression by splenic T cells from orally treated animals was demonstrated in vitro. We conclude that oral therapy is beneficial in EAMG and may prove effective in MG patients. Early priming and delayed inhibition suggest that a molecule with less immunogenic potential than intact AChR might be more effective as a therapeutic agent.