spent much of his academic life studying the synthesis of proteins—but the La Jolla sabbatical had given him a chance to think about new themes. Perched high on a mesa above the Pacific, often closed in by a dense wall of morning fog, the Salk was like an open-air monk’s chamber. Working with Renato Dulbecco, the virologist, Berg had focused on studying animal viruses. He had spent his sabbatical thinking about genes, viruses, and the transmission of hereditary information.
One particular virus intrigued Berg: Simian virus 40, or SV40 for short—“simian” because it infects monkey and human cells. In a conceptual sense, every virus is a professional gene carrier. Viruses have a simple structure: they are often no more than a set of genes wrapped inside a coat—a “piece of bad news wrapped in a protein coat,” as Peter Medawar, the immunologist, had described them. When a virus enters a cell, it sheds its coat, and begins to use the cell as a factory to copy its genes, and manufacture new coats, resulting in millions of new viruses budding out of the cell. Viruses have thus distilled their life cycle to its bare essentials. They live to infect and reproduce; they infect and reproduce to live.
Even in a world of distilled essentials, SV40 is a virus distilled to the extreme essence. Its genome is no more than a scrap of DNA—six hundred thousand times shorter than the human genome, with merely seven genes to the human genome’s 21,000. Unlike many viruses, Berg learned, SV40 could coexist quite peaceably with certain kinds of infected cells. Rather than producing millions of new virions after infection—and often killing the host cell as a result, as other viruses do—SV40 could insert its DNA into the host cell’s chromosome, and then lapse into a reproductive lull, until activated by specific cues.
The compactness of the SV40 genome, and the efficiency with which it could be delivered into cells, made it an ideal vehicle to carry genes into human cells. Berg was gripped by the idea: if he could equip SV40 with a decoy “foreign” gene (foreign to the virus, at least), the viral genome would smuggle that gene into a human cell, thereby altering a cell’s hereditary information—a feat that would open novel frontiers for genetics. But before he could envision modifying the human genome, Berg had to confront a technical challenge: he needed a method to insert a foreign gene into a viral genome. He would have to artificially engineer a genetic “chimera”—a hybrid between a virus’s genes and a foreign gene.
Unlike human genes that are strung along chromosomes, like beads on open-ended strings, SV40 genes are strung into a circle of DNA. The genome resembles a molecular necklace. When the virus infects the cell and inserts its genes into chromosomes, the necklace unclasps, becomes linearized, and attaches itself to the middle of a chromosome. To add a foreign gene into the SV40 genome, Berg would need to forcibly open the clasp, insert the gene into the open circle, and seal the ends to close it again. The viral genome would do all the rest: it would carry the gene into a human cell, and insert it into a human chromosome.I
Berg was not the only biologist thinking about unclasping and clasping viral DNA to insert foreign genes. In 1969, a graduate student working in a laboratory down the hall from Berg’s lab at Stanford, Peter Lobban, had written a thesis for his third qualifying exam in which he had proposed performing a similar kind of genetic manipulation on a different virus. Lobban had come to Stanford from MIT, where he had been an undergraduate. He was an engineer by training—or, perhaps more accurately, an engineer by feeling. Genes, Lobban argued in his proposal, were no different from steel girders; they could also be retooled, altered, shaped to human specifications, and put to use. The secret was in finding the right tool kit for the right job. Working with his thesis adviser, Dale Kaiser, Lobban had even launched preliminary experiments using standard enzymes found in biochemistry to shuttle genes from one molecule of DNA to another.
In fact, the real secret, as Berg and Lobban had independently figured out, was to forget that SV40 was a virus at all, and treat its genome as if it were a chemical. Genes may have been “inaccessible” in 1971—but DNA was perfectly accessible. Avery, after all, had boiled it in solution as a naked chemical, and it had still transmitted information