William Summers looks at the breakthroughs that flowed from knowledge of the double helix

Scientists, for the most part, come in one of two persuasions.

In the first group are the puzzle-solvers, who ask of nature: "How does it work?" In the second group are those who ask of themselves: "What can I do with the answers found by the first group?"

These two approaches are not always as tightly coupled as we might like to believe. It is often years before fundamental understanding of nature's secrets yields scientific solutions to real-world problems. This slippage is nowhere more evident than in medicine. It took centuries for Renaissance anatomical knowledge to find any real applications in medical diagnosis and treatment. Some 19th-century germ theories took decades to have any impact on hygiene, public health and clinical medicine. The so-called translation of science from the laboratory to the bedside is a hot topic among research managers and health-policy gurus.

ÖйúAƬ

The history of the impact of the discovery of the three-dimensional macro-molecular structure of the DNA molecule is similar. Although laboratory scientists immediately hailed the 1953 discovery of the double-helical structure of DNA as groundbreaking, physicians paid no attention at all. This beautiful structure had magnificent explanatory power: it showed in graphic detail how like begets like with simple chemical principles involving stereochemical fits and base-pairing rules.

Very nice, but what can we do with it? Not much, it turned out.

ÖйúAƬ

For about three decades, it had no discernable impact on anything remotely relevant to the practice of medicine. True, anti-cancer drugs aimed at blocking DNA metabolism were being developed, and there was an occasional mention of DNA in the medical literature related to anti-DNA antibodies in autoimmune disease, but none of these medical applications required an understanding of the structure of DNA. The medicine of the 1950s, 1960s and 1970s used knowledge of DNA that was well in hand before 1953.

But in the early 1970s, several methodological advances came together to make possible the medical promises of DNA. First, from esoteric research on the odd phenomenon of how bacteria protect themselves from the invasion of foreign genes came the discovery of enzymes that cut the DNA strands at specific sites in the DNA depending on the sequence of nucleotide sub-units in the DNA structure.

This discovery gave scientists a specific tool to compare genes on the basis of their cutting patterns - a procedure that is the same as comparing two pages of text by determining the lengths of all the sentences in each text. If the two texts have the same number of sentences of the same lengths, it is highly likely that the texts are the same, even though we do not know the texts themselves.

This technique of gene comparisons allowed detailed comparisons, first of viruses and bacteria and later of human beings. The comparisons revealed the degree of genetic similarity and diversity between individuals and in populations. This approach, called restriction nuclease fragment length polymorphism (RFLP) analysis, provided the clinician with useful knowledge of the source and transmission of outbreaks of infections and mini-epidemics by allowing the identification of genetically related strains of germs.

ÖйúAƬ

A second set of techniques was soon developed to improve and refine RFLP analysis, so that it became possible to focus on just one part of the massive human genome. Initially, simple viruses and the accessory bacterial DNA of plasmids were the only DNA molecules that could be studied by RFLP analysis because they yielded a manageable number of fragments. When it became possible to prepare radioactively labelled DNA molecules that were complementary to a gene of interest, scientists could focus their study on a tiny region of particular interest. Methods to separate and detect DNA fragments from human DNA finally opened the path to the clinical promises of the earlier discovery of its macro-molecular structure. RFLP analysis was first applied to improve prenatal diagnosis of hereditary anaemias such as sickle-cell disease. Applications to other hereditary conditions soon followed, and RFLP analysis has become a routine tool in genetic counselling. Studies on human populations and refinements of these methodological approaches have made possible the almost unique identification of an individual by RFLP analysis of her or his DNA.

The Human Genome Project was the logical extension of RFLP analysis. Methods for determining the sequence of DNA molecules are on the one hand very clever and on the other quite primitive. They work very well for small chunks of DNA but fail miserably for DNA molecules the size of a human chromosome. The Human Genome Project was such a tour de force precisely because it approached the problem of determining the sequence of the entire linear DNA by breaking the DNA into manageable fragments that could be analysed by existing methods that work for little pieces of DNA. The study and assembly of millions of "sentences" into the final "text" was a massive undertaking. In some ways, its accomplishment marks the completion of the work begun in 1953 to provide a full description of the structure of DNA.

While study of fragmentation patterns of DNA can be useful for comparisons, the sequence of DNA bases is more informative of the functional state of a gene in a particular individual patient. As we learn more about the relationships between disease manifestations, disease susceptibility and gene sequences, the clinician will rely more and more on the study of her patient's DNA to guide diagnostic decisions and therapeutic recommendations.

As in the case of many brilliant advances in conceptual understanding, the application of that knowledge was not immediate in the case of the discovery of the macro-molecular structure of DNA. But the knowledge of the particulars of DNA structure, more specifically the sequences that give us our chemical individuality, is at last allowing us to relate facts about DNA to health and sickness, to diagnosis and prognosis, and to therapy.

ÖйúAƬ

William C. Summers is a professor in the Yale University School of Medicine. He will speak at an AAAS symposium on "The double helix at 50" on February 15 at 8.30am.