Dr. Wallace Rowe 
Page 3 
Injection of rats with acetylocholine receptor prepared either from 
the electric organ of the eel or from rat muscle in a dose of approxi- 
mately 15 yg emulsified in complete Freund's adjuvant and administered 
together with killed B pertussis organisms will cause the production 
of antibodies capable of binding rat acetylcholine receptor and will 
lead, in a variable number of recipients, to the development of 
experimental autoimmune myasthenia gravis (EAMG) , which in some 
instances may lead to the death of the animal (4) . Serum from rats 
with EAMG can cause development of acute symptoms of EAMG in recipients, 
suggesting a critical role for antibodies in the pathogenesis of the 
disorder. Such recipients do not develop a self-sustaining autoimmune 
disease. 
No good evidence on the gut as a route of induction of autoantibodies 
is, to our knowledge, available. In general, the enteric route of 
immunization is quite poor when compared to parenteral routes. 
The theoretical basis for autoantibody production in the situations 
described is that in many cases, antigen-reactive B lymphocytes exist 
in individuals who are "tolerant" to these antigens. These cells are 
normally not activated either because helper T lymphocytes specific 
for the antigen do not exist or because antigen-specific suppressor 
cells actively inhibit the response. Such B lymphocytes could be 
stimulated to autoantibody production by the association of "auto- 
antigens" with new antigenic determinants for which antigen-reactive 
T lymphocytes exist; by the association of the antigenic determinant 
with a polyclonal B lymphocyte activator, such as lipopolysaccharide; 
or possibly through the deletion of suppressor T lymphocytes. The 
results obtained from the experimental systems described above suggest 
that microgram quantities of antigen are required for the initiation 
of autoantibody production. 
It is possible then, that acute exposure to a mass of bacteria, even 
dead, which is large enough to provide approximately several yg of human 
or cross-reactive protein could lead to autoantibody production. However, 
if the organisms are ingested rather than injected parenterally , the 
possibility of autoantibody production is likely to be substantially 
diminished. A small number of bacteria capable of growing in the 
host, particularly if they could penetrate the GI tract, or capable of 
transferring their genetic information to a more pathogenic organism 
could cause more serious effects. However, if the microorganism used 
is enfeebled to the extent that it can neither grow in the host nor 
transmit its genetic information, then the risk of autoantibody production 
would largely be that of ingesting, injecting, or inhaling a relatively 
large number of organisms. It seems appropriate to point out that this 
risk is almost certainly less than that routinely faced by immunologists 
who prepare complete Freund's adjuvant emulsions of human and cross- 
reactive proteins and inject these into animals. Accidental injection 
[ 272 ] 
