October 27, 1947

Dr. E. C. Kleiderer

Lilly Research Laboratories

Eli Lilly and Company

Indianapolis 6, Indiana

Dear Dr. Kleiderer:

I have been thinking over the problem of increasing the time during which penicillin is retained in the blood stream in large amounts after an injection, and am writing to make some suggestions about a possible method of achieving this result.

The general method that I propose is that of increasing the size of the molecules, so that there will be a diminished leakage from the blood stream through the glomerular filter and also through other physiological membranes, or into the tissues. I suggest that this result might be achieved by coupling penicillin molecules together into a large aggregate, perhaps with use also of other molecules. Thus it might be possible to diazotize para-amino-phenyl penicillin and cause it to couple with para-hydroxyphenyl penicillin. Similarly diazotized para-aminophenyl penicillin could be coupled to resorcinol or phloroglucinol or some similar molecule. I think that if three penicillin molecules were thus coupled to resorcinol or phloroglucinol the resulting molecule might well be retained for a much longer time in the blood stream than otherwise. It seems to me not unlikely that these very large molecules would still be physiologically active, but of course that point needs to be tested. There is also the possibility that large molecules could be made that would decompose at a steady rate to produce a supply of penicillin.

Still another possibility would be to couple penicillin to gelatin, and inject this penicillin azo gelatin into the blood stream.

This suggestion is, of course, closely related to the suggestion that I made during my visit (on October 9) that insulin might be "polymerised" by treatment with bis-diazotized benzidine, to produce aggregates that might retain insulin activity, and remain for a longer time in the blood stream.

I shall be interested to have your opinion about these suggestions.

Sincerely yours,

Linus Pauling:par