In-Process and Final Product QC for Oligonucleotides

/Home/Biologics Delivery - Protein Synthesis/Oligonucleotides - Protein Synthesis

In-Process and Final Product QC for Oligonucleotides
2006
Trey Martin
GEN Updates in Biotechnology
Recently, in-process and final QC for oligonucleotides have become a selling point in the industry. Why all the buzz? Frankly, because there is a broad range of services available in the marketplace at similar prices.
For many people a survey article on oligo QC may be akin to watching paint dry; but, if tolerated, it may provide researchers with insights on which methods would be important or useful for custom nucleic acid products in their specific applications.
First, a quick primer on oligo synthesis. The addition of bases in traditional oligo synthesis happens in a controlled stepwise manner (Figure 1). Typically, the first nucleoside derivative (i.e., A,G,C,T,U) is attached to a solid support matrix. The next base is added?in a 3' to 5' direction?through a series of four basic chemical steps.

The first of these steps is called detritylation. The 5'?dimethoxytrityl (DMT) group is a chemical-protecting moiety covalently attached to the 5•OH of the growing chain. It prevents multiple polymeric additions of the same added base (or monomer). Before the next desired base can be added to the chain, the DMT group is removed with an acid solution to ready the 5•OH for the next monomer.1,2
Addition of the nucleotide is a condensation reaction typically referred to as coupling. This coupling to the 5'OH of the growing chain happens via an active intermediate state using nucleotide phosphoramidites. The monomers are selectively protected and purified preparations of each nucleotide base. Chemical protection sites include the phosphate, the 5'OH, and the primary nitrogen groups on cytosine, guanine, and adenine.
After the population of growing chains has undergone a coupling event, there will be a small percentage of oligos that did not react with the active intermediate. These oligos must be eliminated from participation in further coupling events, which would create deletion mutations. This process is called capping.
The final step of the process is the oxidation or stabilization of the growing oligo chain. The coupled amidite has a phosphite group in the unstable trivalent form. A simple oxidation reaction takes the phosphite to the stable pentavalent phosphate form.
Just as important as in-process QC is the concept of pre-process QC. Though intuitive to do so, it is typically not common practice for oligonucleotide suppliers to check all incoming reagents. The major enemy of the coupling reaction is water. DNA and RNA synthesis reagents must be as dry as possible to maximize the yield of the coupling step.
In an interesting recent study, several scientists at Isis Pharmaceuticals found over 150 different molecules in standard preparations of commercial monomers.3 After the final expiration of the extended patents covering beta-cyanoethylphosphoramidites in February 2005, new lower cost reagents have burst on the scene from inexperienced suppliers. While these providers stabilize their production procedures it is critically important to stringently QC the most vital reagent in oligo synthesis.
In detritylation, a handy benefit to the acid lability of the DMT group is that it also actively fluoresces in protonated form. By measuring the absorbance (ABS) at or around 498 nm one can quantify the amount of trityl released, giving a fairly good representation of the strength of the growing chain.
In-Process QC
If, for example, there was a poor coupling on the previous base the trityl ABS level might go sharply down from its steady state. That said, trityl monitoring is not a quantitative assay unless the entire volume of acid solution is removed for each base and measured volumetrically and spectroscopically. This is not practical on modern highly parallel oligo synthesizers, so an in-line or in-process (i.e., visual, CCD, fiber optic) approach is more commonly used.
Though not quantitative, this measurement taken for each coupling can show a rapid trend or catastrophic failure of synthesis. The methodology has been available on commercial synthesizers for over a decade (though sometimes using the less specific conductivity method).
Integrated DNA Technologies (IDT) has used trityl monitoring to identify mechanical or chemical synthesis problems since oligo production began in 1993. The effect of immediate failure discovery is an elimination of unnecessary work on unredeemable samples and a minimization of their effect on customers' turn-around times. Not all large suppliers with in-house equipment have elected to go this route due to the complexity of measurement, instead opting to wait until the quantification step to identify catastrophic failures.
After synthesis, oligos are typically cleaved and deprotected (chemical groups remain on the phosphate backbone and primary nitrogens of the bases after synthesis). The small-molecule byproducts of this deprotection can be removed though several chemical processing methods. For standard applications like PCR, oligos with this basic level of preparation are perfectly usable. Before they can be added to an experimental reaction however, the reaction yield must be quantified.
This process?Optical Density Measurement?gives a quantified yield for the custom reaction. The oligo preparation is typically measured for light absorbance at 260 nm. Using an extinction coefficient calculated through analysis of the sequence, one can take the OD260 number and calculate the synthesis yield in nanomoles and micrograms. A benefit to this process from a QC perspective is that it can provide another checkpoint for catastrophic failure.
Oligo yields will vary by sequence, prep, column, machine, etc., but should typically fall into a predictable range. Should a custom oligo yield fall well below this range it is an obvious clue that something may have gone wrong. Though the OD260 step is essentially required for use of the oligo, it can also be considered an in-process QC step for this reason.
However, if a synthesis (without trityl monitoring) has a catastrophic event at an inopportune time in the process it is possible to have a fully truncated oligo?say 21 out of 25 bases?or a majority population of deletion mutants that pass the standard OD260 parameters.
Post-Synthesis Quality Control
This argument alone is sufficient to drive the point on final product QC. There are two primary issues that can be addressed post-synthesis: product identity and product purity.
Identity can be assessed by measuring the predominant molecular weight of the population. The target sequence is known, so the calculated molecular weight of the bases themselves is the standard by which one compares the measured molecular weight to see if the desired compound was created.
By far the most pervasive identity QC in the oligo industry is MALDI-TOF mass spectroscopy (Figure 2). MALDI-TOF uses laser light in conjunction with a chemical matrix to impart a charge to the sample in question and repel it from the sample plate. The resulting ions travel through a flight tube to the detector, which measures particle counts as a function of time. The time-of-flight (TOF) is directly proportional to the mass of the molecule.
MALDI-TOF is a robust and incredibly high-throughput process for assessing molecular weight. Its only drawback is that the ionization efficiency (and therefore the resolution) of the procedure drops rapidly above ?45 bases or >13,000 Da. With the popularity of 70 mer arrays and long oligos for cloning and/or gene synthesis, another method is needed to assess these products. The method used at IDT for these oligos (above 13kDa) is electrospray ionization mass spectroscopy or ESI-MS (Figure 3).
ESI-MS ionizes the target molecules into multiple charge states. The readout of these charge states is a waveform that can be deconvoluted into parent peaks. The method uses a tight m/z window of 500?1,500, which gives it high mass accuracy. As only the charge state will vary for the ions, oligos with high molecular weights can be analyzed using this method.
For ESI-MS QC, IDT uses an allowable mass window of only +/- 0.02% of calculated molecular weight up to 150 bases. This work on highly mass accurate QC for long oligos has been published with our partners at Novatia.4
Astute readers will note that while a close comparison of calculated and measured molecular weight for a given sample will give high confidence that the correct oligo is in the tube, it is doing so with base composition as opposed to sequence. It is absolutely true that an oligo of ACGT will measure the same mass as TAGC.
In a high-throughput setting with barcodes and database tracking, confirmed MW provides a high confidence that the correct oligo is in the tube or plate. For absolute verification, there are a few methods of sequence validation for oligos but they are time consuming and expensive.
Purity Assessment
The coupling reaction is highly efficient but not quantitative. Imperfect reagents, processes, or machinery can affect the stepwise coupling yield of an oligo quite drastically. Note the effect on full-length product of subtle reductions in coupling efficiency in Figure 4.

Subtle differences in the coupling yield can have dramatic results on the purity of oligonucleotide samples, particularly for longer oligos. For example, a process with an overall stepwise coupling efficiency of 99% would produce a 20-mer oligo of 83% full-length product purity. A 70-base oligo with the same process would yield a 50% pure preparation. With a subtle drop in the coupling efficiency to 98.5%, the resulting full-length product purities are 75% for the 20mer and 35% for the 70mer.
While PCR itself is a relatively forgiving process, applications like clinical diagnostics, gene-expression microarrays, structural studies, gene synthesis, and highly multiplexed PCR are not. In these cases it is important to start with a sam