Recombinant Human Insulin (synthesis of the gene)

By the mid-1970s, the global supply of therapeutic insulin depended entirely on the pancreases of slaughtered pigs and cattle. The insulin itself was biologically active in humans, but it differed from human insulin at one or three amino acid positions, and a small but clinically significant fraction of patients developed antibody responses that blunted its effect or caused local reactions. Insulin made from a synthetic human gene would carry the human amino acid sequence exactly.

The solution came from the newly formed company Genentech and its collaborators at City of Hope National Medical Center in Duarte, California. In 1977 and 1978, David Goeddel and a small team at Genentech, working with Arthur Riggs and Keiichi Itakura at City of Hope, synthesized the DNA sequences encoding insulin's A chain and B chain separately, inserting each into E. coli expression vectors. The bacteria produced each chain as a fusion protein that could be cleaved and purified. The two chains were then combined chemically under oxidizing conditions to form active insulin through correct disulfide-bond pairing.

Herbert Boyer, Genentech's co-founder, had developed the cloning methods the project relied on, in work with Stanley Cohen begun in 1973 on putting recombinant plasmids into bacteria. The results were published in the Proceedings of the National Academy of Sciences in January 1979. The A chain was purified completely and the B chain partly, and after the chains were mixed, reduced and reoxidized, insulin was detected by radioimmunoassay.

Eli Lilly licensed the technology from Genentech and invested in scaling the fermentation and purification processes to pharmaceutical-grade manufacturing. The FDA approved Humulin, Lilly's recombinant human insulin, in 1982; Lilly's Irving Johnson described it as the first commercial health care product derived from recombinant DNA technology. The approval came about four years after the gene synthesis, for work that had begun before there were federal guidelines for large-scale recombinant DNA production.

The insulin project showed that bacteria could be engineered to make a human protein pure enough for patients. Later recombinant drugs, including growth hormone, erythropoietin and clotting factors, used the same steps: identify the gene, express it in a host cell, purify the product and show that it matches the natural protein. Johnson's 1983 account in Science described the steps taken to prove the product's identity and safety, and predicted that commercial uses of the technology would depend on economic conditions and investment capital.