Insulin-related material in microbes: similarities and differences from mammalian insulins.

Abstract

We have reported that extracts of Escherichia coli and Tetrahymena grown in synthetic media contained material that reacted specifically in the immunoassay and bioassay for insulin. One additional strain of Tetrahymena and four of E. coli yielded amounts of material similar to those reported previously. In addition to their behavior on Sephadex G-50, the immunoactive insulin-related materials from the microbial sources behaved like authentic vertebrate insulins in their ability to be adsorbed to and eluted from disposable octadecasilylsilica cartridges, DEAE-Sephadex, DEAE-cellulose, and one system of high-pressure liquid chromatography (HPLC). As with less purified microbial material, the "insulin" that had been purified on DEAE and HPLC, when tested for its bioactivity, had an immunoactivity:bioactivity ratio of approximately unity and the bioactivity was largely neutralized by anti-insulin antibody. Because the material from the microbes was so similar to authentic insulins, studies were undertaken to demonstrate that inadvertent contamination with vertebrate insulins was highly unlikely. Blanks carried through the entire procedure were always negative. Tetrahymena grown and extracted in another laboratory gave the same results. Tetrahymena that had been grown but then allowed to stand in the fermenter under adverse conditions and then carried through the entire procedure were devoid of insulin. Tetrahymena that were homogenized and subjected to acid hydrolysis were devoid of insulin. Further substantiation that exogenous contamination was highly unlikely was provided by two other types of experiments. In one of these, it was shown that the subcellular distribution of exogenously added porcine insulin or porcine 125I-labeled insulin was different from the distribution of endogenous insulin. In the second type of experiment, it was shown that during the log phase of growth of Tetrahymena or of E. coli the insulin content of the system increased multifold in a fashion that might be expected for living organisms but quite unexpected for exogenous contamination. (Interestingly, the insulin content of the E. coli medium far exceeded that which might be contributed by death of cells, estimated by the content in the medium of an intracellular enzyme.) When E. coli was grown and processed in four other laboratories having no contact with our own, similar levels of insulin-related material were recovered.