Oligonucleotides synthesis - a keystone of modern genomics

Ellen Prediger takes us on a historical journey exploring oligonucleotide synthesis

The influence of molecular techniques now impacts on every area of biology from drug discovery to conservation. The ability to synthesise DNA sequences has been a key driving force behind the uptake of these approaches, making it possible to engineer custom oligonucleotides (oligos) for use in PCR, DNA sequencing, cloning, mutagenesis and microarray production. Researchers now routinely purchase large numbers of oligos for their work, each synthesised with a specific sequence, giving them a uniqueness that is unmatched by any other experimental reagent. Although often taken for granted, the ease with which oligos can be quickly and easily manufactured has only been made possible by over sixty years of dedicated research into oligo synthesis.

In order to appreciate the synthesis and uses of DNA oligos, it is necessary to understand the structure of the DNA molecule. DNA carries the hereditary information present in each cell, which is encoded using a four letter alphabet of nucleotides arranged into a long, double-stranded polymer. Each nucleotide is connected to the next by a phosphate linker. Although most natural DNA molecules are usually very long, such as those that make up human chromosomes, oligos are usually less than 100 nucleotides in length and tend to be single-stranded rather than double-stranded. The latter point is an important consideration, as double-stranded DNA can be replicated using DNA polymerase via the construction of a new DNA strand based on an existing single-stranded template. This mechanism is not feasible for the synthesis of single-stranded oligos, as there is no initial template strand to base the new oligo sequence on. Instead, the addition of a new nucleotide to the growing oligo chain must be catalysed chemically, which also provides the opportunity to specify the precise nucleotide sequence of the molecule.

The first oligo to be chemically synthesised was a dithymidinyl nucleotide, created in 1955 by Michelson and Todd1 and composed of just two nucleosides linked together. A nucleoside is a simpler version of a nucleotide, in that it lacks any accompanying phosphate molecules, and is the building block from which oligonucleotides are formed. The reaction performed by Michelson and Todd was slow and the intermediates unstable, but the process illustrated that the in vitro synthesis of oligos was possible. During the late 1950s, the Nobel Prize winning scientist H. Gobind Khorana and his group made significant improvements to the process, using a new method termed phosphodiester synthesis, which employed the use of so called ‘protection steps’ 2,3.

Their technique allowed the production of tri- and tetra-nucleotide oligos, which could be later combined to produce even longer chains. The method relied on ‘protecting’ reactive parts of the nucleoside molecules until the polymerisation reaction was instigated, at which point the protective element was removed and the reaction could proceed (Figure 1). Reactive sites on the nucleoside were protected by three different groups. The integrity of the ring structure of the base was protected by a benzoyl or an isobutyryl group, the oxygen on the 3´ carbon of the sugar was prevented from forming a reactive hydroxyl until needed by the addition of an acetyl group, and the 5´ carbon was blocked by a trityl group. Subsequent couplings proceeded through the 3´ carbon after removal of the acetyl group. All nucleotides added from this point would have a reactive hydroxyl in place of the trityl group but would retain the other two protecting groups.

The continual cycling of reactive elements, from protected to unprotected, was key for the successful linear, step-wise production of an oligo molecule, and this fundamental theory is still part of today’s oligo synthesis methods. In this way, the construction of an oligo molecule can be carefully controlled, reducing the formation of impurities such as undesired reaction intermediates and products.

The work of Khorana was revolutionary at the time and was instrumental in deciphering the meaning of the genetic code. It also led to the first chemical synthesis of an intact gene, that encoding the 77 nucleotide-long yeast tRNAAla locus. However, the phosphodiester method of oligo synthesis was not without its disadvantages. Although more efficient than previous methods, the new approach still led to the production of large numbers of unwanted side chains on each molecule. These erroneous nucelotides had to be removed by time-consuming purification steps. In the 1960s, this problem was addressed by the laboratories of Letsinger and Reese. Building on the initial protection method of Khorana, both introduced additional protection steps into the synthesis reaction, creating a new method known as phosphotriester synthesis4. This approach significantly reduced the effects of side-chain reactions, increased the stability of reaction intermediates, and facilitated quicker reaction steps.

Although oligo synthesis had come a long way since the initial work of Michelson and Todd in 1955, the phosphotriester methods being used in the 1970s still had some weaknesses. For example, the method could never be optimised to maximal efficiency, and could only produce accurate oligos if they were less than 20 nucleotides in length, which limited their use in site-directed mutagenesis, as well as sophisticated PCR and cloning applications. To increase efficiency and yield, Letsinger and colleagues decided to replace the nucleoside building blocks with slightly modified versions, known as the nucleoside phosphochlorodites5,6.

These molecules are far more reactive than standard nucleosides, increasing the rate and efficiency of oligo polymerisation via a technique known as phosphate triester synthesis. This approach was taken a step further in the 1980s by the group of Caruthers, who had himself been a member of Letsinger’s lab. They replaced the phosphochlorodite building blocks with phosphoramidites7,8.

Although this involved only a subtle change – the swapping of a chloride group for an amine group – it completely altered the properties of the nucleoside monomers, allowing them to be synthesised and stored for later utilisation. In fact, phosphoramidites became so useful that they are still used in today’s oligo manufacturing processes.

Initial oligo synthesis reactions were carried out in solution, meaning that after each round of polymerisation, the reaction product had to be precipitated and cleaned to remove unwanted residual reagents. Therefore, as early as the 1950s, work was already underway to improve oligo synthesis by affixing the growing oligo polymers to a solid support structure, using methods similar to those that had been employed to optimise polypeptide synthesis. Khorana had been the first to experiment with this approach, performing oligo polymerisation using an initial nucleotide affixed to a soluble polystyrene bead. However, significant increases in yield were only achieved in the mid 1970s, when the use of solid supports was combined with phosphotriester chemistry. This approach made it easier to control the extension of the nucleotide chain, by favouring unidirectional growth and inhibiting the formation of complex tertiary structures. Caruthers’ group took this a step further in the early 1980s, performing phosphoramidite synthesis attached to a solid support, such as a silica gel or glass pore. It was also this combination that made the process amenable to automation, increasing efficiency and reducing the laborious nature of the synthesis process. For the first time, oligo synthesis chemistry was fast enough, stable enough, and simple enough for widespread adoption.

The approaches pioneered over the last fifty years have created an efficient mode of producing custom oligos cheaply, quickly and accurately. The current methodology is based on phosphoramidite synthesis and involves four main stages, namely deprotection, coupling, capping and stabilisation, which are repeated cyclically to add each new specific nucleoside to the growing oligo chain.
After synthesis, oligos are purified using one of several methods. Typically, polyacrylamide gel electrophoresis (PAGE) is used to separate the oligos by size, such that only those containing the correct number of nucleotides are selected for further use. In other cases, for example when synthesising oligos containing specially modified bases, high performance liquid chromatography (HPLC) is employed. This technique separates oligos based on charge and/or hydrophobicity via the movement of each molecule through the column. After the oligos have been purified their quality can be assessed using more accurate methods, such as mass spectrometry or capillary electrophoresis.

We’ve come a long way since two individual nucleosides were joined together to form the first oligo. It is now possible to produce oligos >200 nucleotides long, on a scale and level of purity that makes them a mainstay in molecular laboratories throughout the world. Oligos are also available with a range of nucleotide modifications that further expand their usefulness, including the incorporation of fluorescent dyes and specialised nucleotides optimised for specific applications. Many of the major molecular revolutions of the last few decades, such as the advent of PCR, DNA sequencing and microarray analysis, have hinged on our ability to synthesise custom oligos. As we enter the era of the personal genome, deep sequencing and high resolution microarrays, it is unlikely that our reliance on oligos will wane any time soon.