Summary of the impact

This case study describes both economic and healthcare benefits that have resulted from a new DNA (gene) sequencing technique known as SOLiD sequencing. Through the 1990s until the present, Cosstick (University of Liverpool since 1984) has both developed the synthesis and studied the properties of chemically modified DNA in which a single oxygen atom is replaced by sulfur; we have termed this a 3'-phosphorothiolate (3'-sp) modification. Chemically prepared DNA containing the 3'-sp modification is a key enabling component of the Applied Biosystems SOLiD DNA sequencing instrument which is able to produce extremely rapid, cost-effective and exceptionally accurate DNA sequence information. The impact of this very powerful sequencing technology extends beyond economic benefits as it has many healthcare applications which have impacted medical practice.

Underpinning research

The chemical modification to DNA that is required for the SOLiD sequencing technology involves replacing a single oxygen atom in the phosphodiester linkage of the DNA with sulfur to produce the 3'-phosphorothiolate (3'-sp) linkage (Figure). This modification was originally designed by Cosstick (then lecturer in Chemistry, University of Liverpool) as a DNA analogue that was expected to be resistant to cleavage by enzymes that process and manipulate DNA, but susceptible to cleavage by mild chemical reagents. The motivation for this research was to provide new approaches to manipulate DNA that might find applications in biotechnology. The chemical synthesis of oligonucleotides (short pieces of DNA) containing this modification was first reported by the Cosstick Group. An important step towards realising the potential application of the 3'-sp modification was the subsequent report by Cosstick and co-workers that a DNA molecule with 7250 natural phosphate linkages could be specifically cleaved at the site of a single 3'-sp linkage by aqueous solutions of silver ions.¹ It was apparent at this time that in order to fully explore the potential applications of the 3'-sp modification greatly improved methods were required for its preparation and thus through the 1990s and 2000s the Cosstick group reported a number of methods for the chemical synthesis of DNA sequences containing the 3'-sp modification. These studies culminated in the key report (2004) of a preparative method based on the use of 3'-S-phosphorothioamidites which was compatible with standard automated procedures for the chemical synthesis of DNA. This method therefore made oligonucleotides containing the 3'-sp linkage much more readily accessible by routine DNA synthesis procedures.²

The resulting improved access to these 3'-sp DNA analogues made it possible to investigate their interactions with complementary sequences of DNA or RNA in much greater detail. It was shown that a DNA strand containing this modification could form duplex structures with either complementary DNA or RNA strands and indeed the 3'-sp modification actually stabilised some of these structures.^3,4 Significantly, it was established that the effect of the 3'-sp modification had a predictable effect on the thermal melting temperature of complementary oligonucleotides. This was important in establishing that DNA containing the 3'-sp modification would be compatible with the hybridisation steps in the SOLiD sequencing technique.

In summary, the work conducted by the Cosstick group at Liverpool has covered the synthesis of the 3'-sp DNA analogues, the effect that this modification has on duplex and higher order DNA/RNA structures and their mild chemical and enzymatic cleavage. This work established information on all the important chemical properties of this modification and thus provided the necessary background from which the chemical/biochemical aspects of the sequencing method could be developed. Much of this work was performed though national and international collaborations. The work has been supported by both the EPSRC⁵ and the BBSRC.⁵

References to the research

* = Three references to indicate the quality of the research.

5. Cosstick PI on all grants. EPSRC: EP/F011938/1, Increasing the potency of RNA interference using RNA mimetics, 2007-10, £299,576; BBSRC: B05097, Synthesis and application of nucleic acid analogues containing 3'-S-phosphorothiolate linkages, 1995-98, £136,725; BBSRC: B18146, Oligonucleotides containing phosphorothiolate linkages, 2003-06, £192,736.

Details of the impact

The body of research published by the Cosstick group from the 1990's through to the present, has described a wealth of information relating to the chemistry and biochemistry of the 3'-sp linkages. Oligonucleotides containing this 3'-sp linkage are an essential component in a DNA sequencing method known as SOLiD™ (Sequencing by Oligonucleotide Ligation and Detection). SOLiD sequencing was launched by Applied Biosystems Inc (ABI, now incorporated into Life Technologies as of 2008) in 2007 and the SOLiD sequencing instruments became commercially available in 2008.^6,7 The SOLiD method is based on sequencing by ligation and uses universal sequencing primers (essentially short oligonucleotides containing the 3-sp linkage) to interrogate the DNA template to be sequenced. The sequence is read through rounds of hybridisation, ligation, detection and cleavage. The cleavage step removes the fluorescent label (required for the detection step) from the 5'-end of the oligonucleotide and resets the system for the next round of hybridisation and ligation, so that the next nucleotide can be sequenced. The specific cleavage step of the 3'-sp linkage, which is conducted under mild conditions compatible with the requirements of the SOLiD method using aqueous silver ions as demonstrated by Cosstick,¹ is crucial: at the time when the method was developed, no other of DNA modification was known to work in SOLiD sequencing. Details of the sequencing method and the role of the 3'-sp linkage are clearly evident from the patents of McKernan⁸ and from a review article on sulfur analogues of nucleic acids by Zon.⁹ Cosstick's chemistry was thus decisive in enabling implementation of SOLiD sequencing. It was subsequently shown that SOLiD sequencing can also be performed in what is known as "the reverse direction" using the isomeric 5'-sp linkages,^8,9 but in this case an additional step is necessary to remove the 3'-phosphate prior to ligation.

In 2006, ABI were working on the development of their SOLiD sequencing method and needed to produce oligonucleotides containing the 3'-sp linkages on a large-scale, as these were essential reagents for the SOLiD instruments. Their synthetic approach to the large-scale synthesis of these oligonucleotides was principally based on scaling-up the synthetic procedures published by Cosstick in 2004.² On invitation from Dr Gerry Zon (then director of sequencing chemistry), Cosstick visited the ABI chemistry labs (Foster City, CA, USA) in October 2006 for detailed discussions relating to some of the difficulties the chemistry sequencing group had encountered with the synthesis of 3'-sp modified oligonucleotides. At that time there was considerable urgency to establish an efficient method for the production of the 3'-sp modified oligonucleotides in order to meet the demand for these reagents when the instruments became commercially available.

In terms of their performance, the latest SOLiD instruments (5500 series genetic analysers) deliver greater than 90 giga base-pairs of sequence information in one day and a 7 day run can complete the sequence of the human genome. The two-base encoding, which is unique to the SOLiD method, provides exceptional sequencing accuracy at >99.94%, (for further information see: http://tools.lifetechnologies.com/content/sfs/brochures/cms_057511.pdf),¹⁰ as each nucleotide in the sequence is essentially read twice. The SOLiD technology is one of several so-called second/next-generation sequencing systems (together with instruments from Illumina, Roche and to a lesser extent Helicos) which were available as of 2008. Each of these sequencing systems seems to have its own advantages and limitations, although in 2011 the SOLiD system was reported to have the lowest reagent cost needed to reassemble a human genome¹¹ and came top in the J.P. Morgan Next Generation Sequencing Survey¹² (published in 2010) for accuracy. Accuracy was also shown to be the most important attribute when choosing a DNA sequencing system.¹²

The estimated sales value of SOLiD sequencing systems to Life Technolgies is also presented in the J.P. Morgan survey¹² and rose from zero in 2007 to $68 million in 2008, $100million in 2009 and predicted sales of $136.9 million in 2010, $178 million in 2011 and $222.5 million in 2012 (figures based on J.P. Morgan estimates and company reports¹²). In 2010, only two years after becoming commercially available, the SOLiD instrument was estimated to have about 20% market share for second/next-generation sequencing and was predicted to rise to 22.6% (second largest market share after Illumina) by 2012.¹² New jobs created by commercialization of SOLiD sequencing include USA-based manufacturing of consumable reagents, world-wide hiring of technical specialists to support customers, and (especially) world-wide sample preparation, sequencing and bioinformatics analysis at service-provider facilities.

The benefit of the SOLiD sequencing system goes beyond that of generating economic impact, as the technology is now enabling patients to benefit from personalised medicine derived from DNA sequence information. For example, whole genome sequencing (conducted with the SOLiD 4 instrument) of twins with dopamine-responsive dystonia, a clinically complex neurological movement disorder, that is normally treated with L-dopa, identified unexpected mutations in the gene encoding sepiapterin reductase. This enzyme is responsible for the synthesis of co-factors required for the synthesis of dopamine and seratonin. Supplement of the L-dopa therapy with a serotonin precursor resulted in considerable clinical improvements in both twins.¹³

The exceptional accuracy of the 2-base encoding means that SOLiD sequencing is ideally suited to cancer research/diagnosis, because tumors contain mixed sub-populations with different mutations. Deep-sequencing on the SOLiD platform has revealed differential expression of microRNAs in favourable versus unfavourable neuroblastoma (the most common extracranial solid tumor of childhood) and provides a reliable method to assess the aggressiveness and hence prognosis of the tumor.¹⁴ Whilst just two specific examples are presented here, they indicate the tremendous scope of the SOLiD sequencing technique to aid diagnosis and treatment of a wide range of diseases.

There is a list of over 100 publications resulting from SOLiD sequencing at the Life Technologies website which cover applications on: sequencing accuracy, bioinformatics, de novo sequencing, targeted resequencing and transcriptome analysis (further details available at: http://www.invitrogen.com/site/us/en/home/Products-and-Services/Applications/Sequencing/Next-Generation-Sequencing/Publications-Literature.html )¹⁰

The wealth of research that was conducted by the Cosstick group on DNA containing the 3'-S-phosphorothiolate linkage has been absolutely crucial to the development of SOLiD sequencing. The contribution that the work in Liverpool made to the SOLiD programme has been corroborated by Dr Gerry Zon¹⁵ (letter available) who was directing the ABI chemistry sequencing group in Foster City (USA) at the time SOLiD sequencing was being developed. To quote from Dr Zon's letter, "the scale-up of the synthesis of the phosphorothiolate-containing DNA reagents had to be done in a very short period of time, because of competing sequencing techniques which were due to come on the market. It was an enormous benefit therefore to be able to use and adapt your published procedures, particularly the automated synthesis of the phosphorothiolate oligomers you published in 2004" (reference 2). To quote further, "from the work you had published (reference 3 is quoted in the letter, it was apparent that) the phosphorothiolate reagents would be fully compatible with all aspects of the sequencing technique (and) gave ABI confidence to commit to the technique".

Sources to corroborate the impact

6. http://en.wikipedia.org/wiki/ABI_Solid_Sequencing¹⁰
7. Metzker, M. Application of next-generation sequencing. Nature Reviews Genetics 11, 31-46 (2010). DOI: 10.1038/nrg2626
8. McKernan, K., Blanchard, A., Kotler, L. & Costa, G. Reagents, methods, and libraries for bead-based sequencing. US Patent Application 11/345,979 (2005).
9. Zon, G. Automated synthesis of phosphorus-sulfur analogs of nucleic acids-25 years on. New J. Chem., 34, 795-804 (2010). DOI: 10.1039/b9nj00577c
10. Web references can alternatively be found at: http://tinyurl.com/livchemref
11. Niedringhaus, T. P. et al. Landscape of next generation sequencing methodologies. Anal. Chem. 83, 4327-4341 (2011). DOI: 10.1021/ac2010857
12. Peterson T.W., Nam, S.J., Darby, A. J.P. Morgan Next Generation Sequencing Survey, 12 May 2010. Available at: http://www.genomicslawreport.com/wp-content/uploads/2011/04/JP-Morgan-NGS-Report.pdf ¹⁰. Sales figures taken from Appendix 1 of this survey.
13. Genome study solves twins' mystery condition, Nature News, 15 June 2011 DOI: 10.1038/news.2011.368. (This story has been featured on Good Morning America, the Today show, the New York Times and about 240 other publications worldwide.)
14. Schulte, J.H. Deep sequencing reveals differential expression of microRNAs of favorable versus unfavorable neuroblastoma. Nucl. Acids Res., 38, 5919-5928 (2010). DOI:10.1093/NAR/gkq342
15. Letter of corroboration available from the Director of Business Development now at TriLink Biotech (details at: http://zon.trilinkbiotech.com/about-jerry-zon/)¹⁰. Between 1999 and 2011 he was at Applied Biosystems (subsequently Life Technologies) as detailed in reference 9 above.