Recombinant DNA Fermenter, circa 1977

Fermenters like this one used genetically-manipulated bacteria to produce the first human insulin in 1977 and the first human growth factor in 1979.

© SSPL / Science Museum

In 1972, University of California, San Francisco, biochemist Herbert Boyer met Stanford University geneticist Stanley Norman Cohen at a meeting in Hawaii. The two then kicked off a collaboration that eventually led to the creation of the first recombinant DNA, a landmark that ushered in the era of modern biotechnology. By combining Cohen's expertise with bacterial plasmids and Boyer's know-how about restriction enzymes, the two found that they could use bacteria as tiny factories for producing many human proteins. Boyer went on to found Genentech in 1976.

In order to produce the proteins in mass quantities, the fledgling biotech company needed a way to grow transgenic bacteria on an industrial scale. To do that, they turned to the ancient art of fermentation. People had made wine, bread, and beer for thousands of years, yet it wasn't until World War I when tons of acetone and other explosive ingredients were needed that the process became industrialized, says Robert Bud, the principal curator of medicine at the Science Museum in London. Fermentation took another big leap in the 1950s and 1960s when scientists found new ways of growing large amounts of penicillin by continuously stirring air through the fermentation tank.

The 750-liter fermenter depicted here—which was painted by Alan Stones to mark the 1986 opening of the chemical industry gallery at the Science Museum—was one of the first used by Genentech. Finely-tuned valves controlled the flow of air and other nutrients through the 3-meter-tall tank, which was critical to growing large batches of insulin without contamination, Bud says. "By comparison to a chemical plant, this works at low temperatures and pressures," meaning it required less energy, he adds.

Recombinant DNA fermenters are still used today, though early fermenters like this one were gradually replaced by versions with greater sterility and more sophisticated controls.

Correction (posted April 15): The original version of this article misrepresented the expertise of Boyer and Cohen. The Scientist apologizes for the error, which has been corrected.