All cell-based organisms we know of use deoxyribonucleic acid (DNA) to store their genetic information and to pass it on to the next generation. DNA molecules consist of polymeric double-strands built from only four recurring building blocks incorporating one of the nucleobases guanine, adenine, thymine or cytosine, respectively. The structural principle of DNA is thus quite straightforward. Complexity is introduced by the length of the DNA polymers. The genome of the human body, for example, is stored in DNA polymers consisting of around 3 billion building blocks. In our bodies, nearly every cell contains an identical copy of these data.

A perfectionised biochemical machinery is required to replicate, maintain and read-out the data stored in DNA in a fail proof fashion. It does not come as a surprise that the interest of scientists in understanding and mimicking the processes involved has always been extremely high. However, what all cellular forms of life accomplish routinely and apparently without effort can be arduous and in many cases practically impossible to do in a test tube. This extends to the chemical synthesis of DNA, for example. Even today, the de novo synthesis of DNA in dimensions corresponding to the human genome remains to be wishful thinking.

A person who knew the issues involved all too well was Har Gobind Khorana (1922-2011), author of the present talk. In the early 1960s, Khorana was one of the first to synthesize small snippets of artificial DNA, just a couple of building blocks long. Despite their tiny size, these snippets enabled a major breakthrough in DNA research. By systematically varying the sequence of nucleobases in the snippet it could be shown how the biochemical machinery translates the data stored in DNA into proteins. A more detailed discussion of this topic may be found in the Mediatheque’s Topic Cluster “The Life of Proteins”. Khorana’s work culminated in the 1968 Nobel Prize in Physiology or Medicine, which he received jointly with Robert Holley and Marshall Nirenberg "for their interpretation of the genetic code and its function in protein synthesis".

In the present talk, Khorana reviews the ongoing efforts of his group in the field of DNA synthesis with a focus on the work done from 1966 to 1972. He points out that the current chemical methods allow the synthesis of DNA snippets up to just about 20 nucleobase building blocks long. Classical test-tube chemistry would thus not be a promising route to making longer artificial DNA snippets, Khorana admits. Instead, he suggests using enzymes with the capability to join smaller DNA snippets, so-called DNA ligases, for the purpose of increasing the chain length of artificial DNA. Discussing first corresponding achievements of his group, Khorana points out another major issue hampering his work. At the time of his talk, the analysis of longer DNA sequences was not yet feasible and it was practically extremely difficult to determine whether the outcome of a synthetic experiment was actually the desired one. Only in the years following Khorana’s talk the development of methods for DNA sequencing gathered pace, eventually leading to the 1980 Nobel Prize in Chemistry, which was - in part - awarded to Walter Gilbert and Frederick Sanger "for their contributions concerning the determination of base sequences in nucleic acids".

The efficiency of the methods allowing for genome sequencing has rapidly improved ever since. While the cost of sequencing an entire human genome was still around 100 million USD in the year 2000 it is now just as low as a couple of thousand USD [1]. By making use of enzymatic methods, as suggested by Khorana, significant improvements were also made in the field of artificial DNA synthesis. Very recently, the synthesis of a eukaryotic chromosome, 272.871 building blocks long, has been reported [2].

David Siegel

[1] http://www.genome.gov/sequencingcosts/
[2] Annaluru et al., Science, 2014, Volume 344, Pages 55-58.