JOSLIN DIABETES FOUNDATION, INC. 
ONE JOSLIN PLACE, BOSTON, MASSACHUSETTS 02215 (617) 732-2400 
732-2540 
April 30, 1979 
Dr. Wallace P. Rowe 
Chief, Laboratory of Viral Diseases 
National Institute of Allergy and 
Infectious Diseases 
National Institutes of Health 
Bethesda, Maryland 20205 
Dear Dr. Rowe: 
In reply to your letter of April 20, it is my guess that insulin-producing 
strains of E.coli pose a minimal threat to human health. It is true that small 
amounts of insulin can enter into the circulation via the upper intestinal 
tract and there is now some recent evidence that specially prepared insulin 
contained in liposomes is a potential mechanism whereby insulin can be given 
therapeutically. However the upper part of the intestinal tract is effectively 
sterile under usual circumstances. Concerning the lower intestinal tract, there 
is very little evidence that any proteins are absorbed since not only is there no 
physiological mechanism for their absorption but the overwhelming growth of the 
enteric flora utilizes what proteins are synthesized by the bacteria and thus we 
are left with ammonia and carbon dioxide and traces of structural protein plus 
some structural carbohydrate. The fact that clostridia inhabits the bowel of 
many normal animals, including man, documents to me the benign nature of potential 
hazardous proteins in the large bowel. 
When we get to extra intestinal infections, I suppose that an insulin-producing 
strain of E.Coli could pose a threat, particularly if there is an overwhelming 
septicemia. But here I feel the insulin byproducts should be readily handled with 
glucose administration and 99% of the hazard would be the organism itself and its 
other toxins. 
Finally, there is the possibility that should a few of these organisms be either 
in or outside of the gut and produce small amounts of insulin, might this initiate 
an immune response to the insulin which then would turn upon the host’s own insulin 
or else his or her insulin-producing Beta cells. This might be true were the 
insulin to be a component of a number of other proteins, thus working as a Freund's 
mixture which could so heighten the immune response that antibodies against one's 
own insulin could be produced, as was done by Drs. Renold and LeCompte in cattle 
Cont'd . . 
DEDICATED :c DIABETES RESEARCH. EDUCATION. PATIENT CARE and YOuTH PROGRAMS. 
[ 248 ] 
