Are You Still Synthesizing Oligos? Can you do (fill in your special need here)?

It's 10:30 Thursday night. It's been a busy week and I really don't feel like going back up to the lab tonight. Nonetheless, I put on my raincoat and head for the car. It comes with the turf–if you want to offer DNA synthesis at an academic institution, you've got to find a competitive niche.


Commercial “oligo houses” offer oligos so much cheaper but they can’t get them to my clients the next morning. Only your “friendly neighborhood core facility” can do that. So, up the hill I go, one more time for the day. Speed is a niche we can fill. A client can submit an on-line work request for synthesis as late as 6:00 pm, just before I go home, and we’ll have it ready for them when they need it in the morning.

Quality is another niche for the small core facility. The inexpensive oligo vendors use highthroughput plate synthesizers. They're fast and stingy with reagents. But think about what you learned in Organic Chemistry — that’s not how you optimize yield. They can't get the coupling efficiency that our good old ABI 394s can get; and even a few tenths of a percentage point matter when you're doing solid-phase synthesis. Consider Table 1.

For primers, linkers, and other standard oligos of lengths less than 30 bases, the lower coupling efficiency of plate synthesizers may not matter for most molecular biology experiments. But for longer oligos and experiments where “signal-to-noise” matters, the quality difference becomes apparent. You do get what you pay for.

In addition, we do quite a bit of non-standard oligonucleotide synthesis. That is the synthesis of oligos with modified nucleotides, phosphorothioate instead of phosphate internucleotide linkages, or other non-nucleic acid molecules coupled to the oligo. The addition of biotin and fluorescence labels for various purification and detection approaches are the most common requests. But we also synthesize models of DNA damage and other unusual molecules, such as, most recently, siRNA oligonucleotides.

A brief history

It's curious, but current state-ofthe- art oligo synthesis uses machines and chemistries that have been around for quite a while; some of the most innovative aspects dating from the late 1950s. For example, there are several reactive sites on a nucleotide, so it’s crucial to control when and where reactions occur. This problem was solved with H.G. Khorana’s “on-off” protection scheme. The Khorana approach calls for two types of blocking groups for the reactive positions on a nucleic acid monomer. This synthesis scheme starts at the 3’ terminus and adds monomers in a reaction cycle until the final 5’-terminal nucleotide has been added. The initial monomer, at the 3' end of the requested sequence, is attached to a solid support with a linkage that is susceptible to a basic reagent, such as ammonium hydroxide, hence, the name “solid-phase synthesis.” The primary amines on the nucleobases are prevented from reacting with “base-labile” blocking groups. But the 5'-OH on the deoxyribose (or ribose for RNA synthesis) is blocked with an “acid-labile” blocking group. It is this blocking group that gets removed just before the next monomer is added. The solid-phase synthesis concepts grew out of earlier work on peptide synthesis by D. Letsinger. Solid-phase synthesis allows excess reagents and side products to be easily washed away without purification between the synthesis steps.

The actual coupling reaction occurs between the unblocked 5’-OH and an activated phosphorous moiety attached at the deoxyribose or ribose 3’ position. M. Caruthers with S. L. Beaucage and L. McBride improved the chemistry of Letsinger and devised the cyanoethylphosphoramidite cycle that is used today for coupling and then oxidation to the natural phosphate linkage.

After all the coupling reactions have been accomplished, the completed oligonucleotide is cleaved from the solid support and the remaining blocking groups are removed by the addition of a basic solution (e.g., aqueous ammonium hydroxide) to the column and incubation in this solution for several hours. There are a number of different blocking groups and column linkers available. These dictate the type of cleavage and deprotection reaction that must be performed to end up with the unblocked, “natural” single-stranded DNA molecule.

At this point, for most applications, all we have to do is remove the organic salts — the remains of the blocking groups — with a simple size-exclusion chromatography step. Voila! A single-stranded natural oligonucleotide. The desalted oligonucleotide is ready for use by the biochemist or molecular biologist for most applications. Occasionally, more extensive purification is needed and we do that upon request as well.

An excellent, brief history of the history of oligonucleotide synthesis can be found at Original references are cited there.

The basic chemistry cycle as implemented on the ABI 394 is shown in Figure 1.

Each complete cycle takes about six minutes, or about ten bases per hour. The final cleavage from the column takes another hour. The synthesis product is pushed to a glass vial for an additional period of time for removal of the remaining blocking groups.

Fill the niche needed by your clients

Besides the researcher who needs their primers or probes ASAP, we are able to synthesize modified and labeled oligos economically. Since we are just trying to breakeven and not make a profit, we only add the cost of the special reagent to our standard synthesis fee. So the modified or labeled oligos are actually quite competitive.

And, of course, we are there to help our grad students, post-docs, and faculty with experimental design and troubleshooting. We give a 100% guarantee with a rapid response to any problems.

RNA interference and synthetic siRNA Oligonucleotides

RNA interference, or post-transcriptional gene silencing, is an ancient, conserved cellular defense mechanism that evolved to help protect against microbial pathogens by targeting exogenous RNA for degradation. It is also involved in regulating the expression of protein-coding genes in many cell types of various organisms. An excellent review can be found at a NIH National Library of Medicine website.1 The salient feature of this relatively new area of research, for synthesis core facilities, is that short, double-stranded RNA molecules (siRNA) can serve as “guides” to the sequence-specific degradation of RNA. As a tool in functional genomic studies, the RNA that is targeted is the messenger RNA (mRNA) of the gene of interest, hence, the term “gene knockdown” for this experimental approach.

During the last several years, RNA interference with synthetic siRNA oligonucleotides has became widely used as a powerful tool for studying gene expression and we are now synthesizing a lot of siRNA oligonucleotides. RNA synthesis reagents have been around for a long time. The monomers are five to six times more expensive than those used in DNA synthesis. The extra hydroxyl on the ribose has to be blocked to prevent it from reacting during the coupling cycle (see Figure 1). The blocking group is a bit bulky, causing steric hindrance, and slows down the actual coupling reaction. We get decent yields by simply allowing a longer time for each cycle's coupling reaction. The “gentle” removal of this blocking group and the sensitivity of RNA to ubiquitous RNA degrading enzymes (RNases), even present on your skin, make processing the final product more difficult. But these difficulties even the playing field for university core facilities vs. commercial vendors and provide us with a whole new niche for our synthesis service.

What makes siRNA molecules cost-effective for the academic core facility to offer is the fact that they tolerate having a terminal dTdT. That is, we can synthesize a mixed oligo with two DNA bases at the 3’ terminus, and the rest of the bases composed of RNA. This makes siRNA synthesis easier and cheaper because we can use the so-called “Low Volume” (LV) columns. These columns contain a membrane instead of CPG 2 as the solid support. The polystyrene membrane is intrinsically drier and allows the use of cycles that are stingier with the expensive RNA reagents. Therefore, the university core facility can successfully compete in the siRNA synthesis area.

“Laissez le bons temps rouler!”3

Custom, synthetic oligonucleotides are used in a wide range of molecular biology experiments. They are used as “primers” for DNA sequencing and PCR (polymerase chain reaction) experiments. They are also used in oligo-directed mutagenesis and to identify specific locations on a genome via sequence specific hybridization. Perhaps because of their ubiquitous use in modern molecular research, they have become a commodity product, much like purified enzymes became a commodity back in the 1970s and 80s. University core facilities really can’t compete on price with commercial sources of primers — relatively short, unmodified oligos. We simply can’t achieve the economy of scale that a highthroughput commercial vendor can achieve, and we can’t use the loss leader technique of getting clients “in the door,” so to speak, with cheap primers in the hope of also selling them something else that makes money. NIH rules require university core facilities to use a straightforward, breakeven fee structure. For this reason, many universities have shut down their oligo synthesis facilities during the last decade.

We have found that university core facilities can continue serving their institution’s researchers with high-quality oligonucleotides when speed and/or modified oligos are needed. And the discovery of RNA interference with siRNA has really rejuvenated the usefulness of university oligo synthesis core facilities. We help with the design and experimental troubleshooting as well. We began synthesizing siRNA oligos just last year, and the demand is up 75% in 2006. It has stimulated interest in our other competitive niches as well. We expect this trend to continue. So hang in there core lab managers... just keep your raincoat handy.

  2. Controlled pore glass, aka CPG
  3. Let the good times roll.

Thomas J. Keller has a PhD from the University of California, San Francisco in Medicinal Chemistry. In 1995, Dr. Keller came to the Oregon Health & Science University to develop a small department-based facility into a campus-wide research core facility now known as the MMI Research Core Facility. His latest endeavor is to improve informatics and data processing in the lab using Perl, BioPerl and EMBOSS. He can be reached through

Categories: Laboratory Technology

Published In

Software Scientists Magazine Issue Cover
Software Scientists

Published: February 1, 2007

Cover Story

Software Scientists

A growing number of scientists are working for firms designing software that models chemical and physical processes.