As scientists, we use oligos everyday – primers for PCR, labelled hybridization probes in rtqPCR detection or expression microarrays, biotinylated pulldown probes, sequencing adaptors etc. Oligos, defined as short stretches of chemically-synthesized nucleic acids, can be chemically made, amplified, assembled and ligated together to form synthetic genes. They can be put into cells and animals (including humans) to perform all sorts of functions, not just via their translation into proteins, but also by interacting with the cell’s molecular machinery to block the expression of RNA, trigger immune responses, mediate gene editing and so on. So how much do you know about how they are actually made?
The history of oligo synthesis goes back a long way. The discoveries made in the 1940s and 1950s on the structure and composition of DNA and how they make up the code for life triggered attempts to chemically synthesize oligos. The first report of oligo synthesis was in 1955 by Todd and Michelson in Cambridge who linked two thymine nucleotides together using phenylphosphoryl dichloride. The reaction however worked slowly and was compromised by the instability of the phosphoryl chloride intermediate (leftmost molecule) which was susceptible to hydrolysis.
Then came along the exceptional scientist, Har Gobind Khorana, born in Raipur and based at the University of Wisconsin-Madison, who transformed the way oligos were made by making two significant developments in the 1950s-60s:
- Instead of using phenylphosphoryl dichloride to activate the 3′ phosphate, he used condensing agents – sulphonic acid chlorides or carbodi-imides – that allowed the reaction to proceed faster and more efficiently.
- He found suitable protective groups to protect the 5′ hydroxyl, 3′ hydroxyl, amino group on the heterocyclic ring, and the phosphate group itself to prevent any non-specific reactions and ensure the correct product was made. These groups could be selectively removed (e.g. with mild acids) at the appropriate time to allow the oligo strand to grow. These protective groups are still used till today.
- For the 5′ hydroxyl (OH) group: dimethoxytrityl (DMT)
- For the 3′ hydroxyl (OH) group: acetyl group which was later replaced by diisopropylamino with a 2-cyanoethyl to block branching at the phosphate
- The amine group: a benzoyl or isobutyryl group
A phosphoramidite – common building block for oligo synthesis used today
Khorana’s phosphodiester method still had limitations though, producing various by-products that slowed the reaction down and required further purification steps to isolate the target oligo.
Subsequently, an American biochemist, Robert Letsinger protected the 3′ phosphate with a 2-cyanoethyl group which prevented these by-products from forming and significantly sped up the reaction. Letsinger also started using Phosphorous (III) rather than Phosphorous (V), which increased the reactivity of the nucleosides, increasing nucleotide coupling efficiency. This was termed the phosphite-triester approach. Letsinger with the help of Khorana’s former student, Marvin Caruthers, were also the first to initiate synthesis of oligos on solid supports. The link to a solid support or resins, usually made up of polystyrene or controlled pore glass (CPG), is made via the 3′ hydroxyl group thereby initiating chemical oligo synthesis from the 3′ to 5′ direction.
Caruthers continued to further improve oligo synthesis and by substituting the chloride group of the highly unstable phosphomonochloridite intermediate with an amide, he created a much more stable intermediate – phosphoramidite – that could be made and stored and activated with tetrazole when required. So here’s the whole process of oligo synthesis:
- The protective DMT group is removed from the 5′ hydroxyl in the detritylation step with mild acid. The solid support is then washed with acetonitrile to remove the acids and dehydrate the reaction.
- The two nucleosides are coupled with addition only taking place on the 5′ of the unprotected nucleoside, activated by tetrazole.
- The oxidation step is required to convert the phosphite to phosphate, which is more stable and also represents the natural form seen in DNA.
- Capping is performed to block nucleotides that were unsuccessfully coupled. An acetyl group is generally added to prevent the hydroxyl group from reacting.
The oligos then have to be cleaved from the solid support and eluted. These oligos grow in pores of the solid supports that are typically on columns or in wells of a plate. It is important to note that as the oligo strand grows, the synthesis yield drops as the growing oligo blocks the pore and reduces diffusion of the reagents through the matrix. Larger pores (1000 angstroms) are thus used for longer oligos but can only support oligo lengths up to 100 bases. Added to these physical constraints, varying inefficiencies in the individual reaction steps ensure coupling efficiency is never 100%. Furthermore, depurination can sometimes occur in the detritylation and capping step, where the glycosidic bond between the purine and sugar is cleaved, producing more truncated products. Hence, the yield of full-length product drops as length of oligo increases and truncated products not of the correct length are often present in most ordered oligos unless HPLC purification is employed.
In recent years however, modern advances have seen oligo synthesis taking place on 2D planar surfaces with higher efficiency and throughput. Agilent for example, employs inkjet printing that takes place in an anhydrous chamber to deposit and couple phosphoramidites on a 1 x 3 inch microarray chip that can hold up to a million growing oligo spots. Oxidation and detritylation are carried out with a flowcell that floods all the oligo features with the necessary reagents. The highly controlled fluidics of the flowcell and the anhydrous property of the printing chamber provides a better coupling efficiency and better control of depurination, allowing oligo lengths of up to 150 nucleotides to be achieved.
Twist Bioscience have scaled it up a notch employing silicon wafers containing 9600 wells, each well being capable of synthesizing 121 different oligos. Furthermore, these oligos can be assembled into longer sequences (synthetic genes) within the well itself, avoiding the fuss that comes with handling small volumes of liquid.
Oligo synthesis has come a long way since the 50s indeed, we now can make a lot of oligos in parallel and can even stitch them together in the process. Ensuring they are error-free and pure however is another problem, but it can only get better from here.
Khorana, H. G. (1968). Synthesis in the study of nucleic acids. The Fourth Jubilee Lecture. Biochemical Journal, 109(5), 709–725. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1187021/
Some images adapted and modified from A Short History of Oligonucleotide Synthesis
By Richard Hogrefe, Ph.D.; TriLink BioTechnologies (contains more detailed description of the chemical processes, recommended reading)
Image of % oligo synthesis yield from FAQs > Oligo Synthesis Resource | W.M. Keck Foundation | Yale School of Medicine
LeProust, E. M., Peck, B. J., Spirin, K., McCuen, H. B., Moore, B., Namsaraev, E., & Caruthers, M. H. (2010). Synthesis of high-quality libraries of long (150mer) oligonucleotides by a novel depurination controlled process. Nucleic Acids Research, 38(8), 2522–2540. http://doi.org/10.1093/nar/gkq163
Banner image “Digital DNA” by Wonderlane