When Nathans returned from Israel in the summer of 1969, he brought some SV40 to begin testing his ideas about restriction enzymes. At the time, only four such enzymes were known, including the one recently found in Haemophilus influenzae bacteria by his JHU colleague Hamilton Smith. The phenomenon of restriction (or host-controlled modification)--where phages (bacterial viruses) grown in one strain of bacteria lose their ability to grow on other strains--had been noted in the early 1950s by Salvador Luria and others, but its mechanism was unknown. In 1962, Werner Arber's studies showed that restriction involved changes in the DNA of the phage and was accompanied by degradation of the DNA. It appeared that bacteria could somehow limit phage infection by breaking down the virus DNA, although certain phage genes were still expressed in the bacteria. Several years later, Arber's lab confirmed that the bacterial host was able to modify or restrict viral DNA, as well as foreign bacterial DNA. Arber theorized that each particular bacterial strain generated an endonuclease (enzyme) that could recognize specific sites--sequences of nucleotides--in foreign DNAs, and cut them there. Each bacterium would protect its own DNA from the enzyme by adding methyl groups to the specific sites, so the enzyme would not recognize them. By 1968, two such enzymes had been identified in strains of E. coli, generating much excitement among biochemists. However, it turned out that while they recognized specific DNA sequences, they did not always cut the DNA close to those sites. Since the fragments produced were of random size and character, the enzymes (later known as Type I restriction enzymes) were unsuitable for mapping or sequencing.
Smith's Haemophilus influenzae enzyme--the first of the Type II restriction enzymes--was soon shown to cleave several bacterial virus DNAs at specific nucleotide sequences. While Nathans was setting up his cell culture lab to grow SV40 in quantity, he had his student Stuart Adler test the ability of all four known endonucleases to cut SV40. Two of the enzymes cut it only once, one didn't cut it at all, but Smith's enzyme cut it into eleven specific fragments. As Nathans had hoped, this broke the DNA into consistent, manageable pieces, onto which individual genetic activities could be mapped. Using polyacrylamide gel electrophoresis, Nathans and graduate student Kathleen Danna fully separated and analyzed the structure of the SV40 fragments. (They also soon determined that Smith's enzyme was actually two enzymes.) New Type II enzymes were rapidly discovered, and during the next several years, Nathans' group (including Danna, Adler, George Sack, Ching-Ju Lai, William Brockman, Mary Gutai, and David Shortle) used them to delineate the SV40 genome. They deduced the size and physical order of the fragments in the genome, and created the first cleavage maps of a viral DNA, showing where each restriction enzyme cut it. Using these maps, they were also able to identify the origin and terminus of DNA replication in the circular SV40. They worked with George Khoury and Malcolm Martin at NIH to map the messenger RNAs in SV40, and also located some of the virus's structural genes, using mutant strains of the virus. Later, they found they could actually create mutants by modifying the DNA at the known restriction sites. These constructed "deletion mutants" proved useful for identifying structural gene products, locating non-essential parts of the DNA, and defining those regions required for cell transformation.
When he began working on restriction enzymes, Nathans didn't foresee their use for the in vitro manipulation of DNA in genetic engineering, but he soon recognized the uses and potential risks of recombinant technology. He was among the scientists who signed the 1974 letter calling for a moratorium on recombinant DNA (rDNA) experiments, and he participated in the 1975 Asilomar Conference and the NIH council that formulated the initial guidelines for rDNA research. Later he was an advisor to the Human Genome Project.
The enzyme found by Hamilton Smith and the techniques that Nathans pioneered for analyzing DNA fragments and mapping genomes transformed molecular biology. Within a few years of their initial experiments, dozens more Type II restriction enzymes had been identified, researchers such as Fred Sanger, Walter Gilbert, and Allan Maxam had developed rapid gene sequencing techniques, and Paul Berg, Herbert Boyer, Stanley N. Cohen, and Ronald Davis had introduced recombinant DNA techniques. The "new genetics" would allow researchers to analyze even very large genomes at the molecular level. In 1978, the Nobel Prize in Physiology or Medicine was awarded jointly to Werner Arber, who predicted the existence of restriction enzymes, Hamilton Smith, who discovered the first Type II restriction enzyme, and Nathans, who demonstrated how to use the restriction enzymes to analyze viral DNA.