In 1959, Arthur Kornberg left the Department of Microbiology at Washington University School of Medicine to head a new biochemistry department at the Stanford University School of Medicine. Most of his WU faculty and staff, including Berg, made the move with him. At Stanford, Berg's research continued to focus on the assembly of amino acids into proteins, specifically the activation step. It was a complex problem, for he found that for every amino acid, there were three or four different tRNAs that might transfer it for protein assembly. The enzymes that bound the amino acids to the appropriate tRNAs were also highly specific. It was becoming clear that in protein synthesis, the accuracy of amino acid attachment to a specific tRNA was crucial for the correct transcription of genetic information, and Berg's team devoted many years to understanding the structure and specificity of those individual enzymes. By 1967 they were able to show that genetically altered tRNA molecules can cause misreading of the genetic code at the ribosomes. Studies with such altered RNAs provided greater insights into the specificity of amino acyl tRNA synthesis and utilization in protein biosynthesis. During these years Berg also investigated the mechanisms of transcription from DNA to RNA, and isolated a RNA polymerase in E. coli that synthesized RNA from a DNA template.
Investigations by Berg and others into DNA-RNA transcription and the translation of messenger RNA to proteins (gene expression) helped illuminate the related process of gene regulation, i.e., to what degree and under what circumstances a given gene or group of genes is actually expressed. It had been clear for many years that some bacterial genes were expressed only under certain conditions; for example, a bacterial culture's production of an enzyme could be switched on or off by changing its food, oxygen, or other variables. Understanding of such switching mechanisms in bacteria was expanding rapidly during the early 1960s, led by the research of Jacques Monod and François Jacob at the Pasteur Institute. The French team had identified in their E. coli cultures both the regions of the bacterial DNA that control expression of certain functions (e.g., synthesis of a needed enzyme), which they called "operators," and groups of genes that were regulated together ("operons"). Near the operators, there are often genes that produce repressor proteins, which prevent the operators from expressing. Monod and Jacob showed that enzyme induction worked by binding repressor genes of a given region of DNA, allowing the hitherto repressed gene to express. Several members of Berg's Stanford department, including Dale Kaiser, had trained at the Pasteur Institute and followed the research there closely. Kaiser had done much work on Phage lambda, a lysogenic virus that infects E. coli bacteria. Lysogenic phages had attracted scientists studying gene regulation because, instead of killing the host cell, such viruses stay dormant, integrate into the host DNA, and multiply with it. The phage genes stay repressed, until some stimulus activates them; at that point, they multiply and kill the host cell like a normal virus. Berg was increasingly drawn to the challenges posed by gene regulation research and began to shift his research focus in that direction.
In a 1965 presentation, Kaiser suggested that the mechanisms at work in the bacteriophage lambda might find analogues in mammalian tumor viruses such as simian virus 40 (SV40) and polyoma which cause tumors in monkeys and mice, respectively. The parallel was not exact (tumor virus genes integrate but are not repressed in the host cell, and continue to express genes that convert the cell to a tumor cell), but Berg was fascinated by the implications: could viruses be used to study gene regulation in mammalian cells, as they had been in bacterial cells? He felt that scientific understanding of gene expression in bacteria was becoming very clear, but there was still little known about gene expression in higher life forms. Eager to explore fresh territory, Berg decided to stop working on bacterial systems for protein synthesis, learn how to culture mammalian cells, and use tumor viruses as models for studying gene expression in mammals. He spent his 1967-68 sabbatical year at the Salk Institute in San Diego, learning cell culture techniques with Renato Dulbecco.
When he returned to Stanford, he built a new laboratory suitable for working with mammalian viruses and spent several years investigating mutations of SV40, characterizing its genome, and determining which regions of its DNA specified which gene. The next phase of SV40 research, which would ultimately lead to the development of recombinant DNA technologies, was inspired by work Berg had done earlier with Charles Yanofsky. That work focused on the mutations in tRNAs that change the reading of the genetic code, using lambda phage as a transducing agent with bacteria--bringing along small pieces of bacterial DNA from one host to another. The genes to be carried could be specified because some phages integrate in only one place on the bacterial genome and carry out the genes nearby. (Other phages fragment the bacterial DNA and pick up pieces at random, so each new phage carries a different piece of the host DNA.)
Berg wanted to know whether mammalian tumor viruses are capable of picking up genes and bringing them to new cells in this manner. One immediate problem was that the obvious vectors--the SV40 and polyoma viruses--were very small. When phages pick up and transfer genes, they leave a few of their own behind; with a length of about 50,000 base pairs, they can accommodate such exchanges without compromising their genomic integrity. But SV40, just 5,000 base pairs long and encoding only 5 genes, lacked that flexibility, and was unlikely to spontaneously pick up genes from other sources. It wouldn't be able to carry even a single gene along to a host cell. To overcome this limitation, Berg and his team set out to combine SV40 DNA with a plasmid (a DNA fragment that can replicate independently of a cell's genome) found in one strain of E. coli. The plasmid consisted of lambda phage DNA together with three genes from E. coli. By splicing the plasmid into the SV40 DNA, they could create a recombinant DNA that would have both the ability to enter a mammalian cell and the ability to carry new genes from other sources when it did so.
The technique that Berg and his colleagues developed for splicing the two DNA molecules together utilized an array of enzymes isolated by Kornberg and others. First, they cut open the circular SV40 and plasmid DNAs. Working from the knowledge that lambda phage DNA strands have cohesive or "sticky" ends that allow complimentary base pair bonding between single DNA strands (much like Velcro fabric closures) into long chains or circles, they created their own sticky ends, by adding strings of thymine (T) or adenine (A) nucleotides (which are complimentary) to one end of each DNA, using another enzyme. Then finally, the A and T strands of the two DNAs were annealed, using DNA polymerase, ligase, and other enzymes. This complex procedure, described by Berg, David Jackson, and Robert Symons in a 1972 article in Proceedings of the National Academy of Sciences, resulted in the first recombinant DNA molecule. Within a year, Stanley N. Cohen, Herbert Boyer, and others had discovered that a restriction enzyme, EcoR1, would create the necessary sticky ends on almost any DNA molecule, which greatly simplified the process and made it possible, at least theoretically, for researchers to splice together any sort of DNA from any sources they liked. In 1980, Berg received both the Nobel Prize in Chemistry and the Lasker Award for this groundbreaking work. The recombinant technologies held immense promise for scientific understanding of genetics and of fundamental life processes, and for genetic therapeutics. And they have indeed revolutionized research not only in biology but in fields such as anthropology, medicine, and forensics. They have also made it possible to create specialized strains of organisms for various purposes, from genetically engineered food plants to bacteria that produce insulin and human growth hormone.
In the early 1970s, however, the revolutionary techniques also posed the possibility of creating unpredictable occupational or environmental hazards, e.g., common bacteria such as E. coli which carried tumor genes. This and other scenarios raised much concern among the public and among scientists, and Berg would soon take a leading role in organizing the scientific community to address those issues.