January 1, 1970 (Vol. , No. )
Julia Baek Ph.D. Research Scientist Thermofisher Scientific
Jim Thayer Ph.D. Staff Research Biochemist Thermo Fisher Scientific
Hongxia Wang Ph.D. Senior Marketing Specialist Thermo Fisher Scientific
Ilze Birznieks Global Product Manager Thermo Fisher Scientific
This is the final article in this series discussing HPLC oligonucleotide separations. This article will discuss LC/MS analysis of failure sequences, phosphorothioate modified siRNAs, CpG methylated oligonucleotide and 5’-phosphorylated oligonucleotide. In addition, the effect of high pH mobile phases on selectivity will be shown.
High performance liquid chromatography (HPLC) and mass spectrometry (MS) are valuable tools for purity assessment and characterization of oligonucleotide impurities. Anion-exchange (AEX) chromatography and ion-pair reversed phase chromatography (IP-RP) are the most commonly used techniques for LC/UV or LC/fluorescence analysis of oligonucleotides. However for LC/MS analysis, AEX chromatography is less popular due to the high salt concentration used in their mobile phases which requires a desalting step before MS analysis. Ion-pair reversed phase LC, with volatile mobile phase components, can be directly coupled to MS. High resolution mass spectrometers provide exact mass data for positive identification of oligonucleotide products and confident impurity characterization for these critical applications.
LC/MS Analysis of Failure Sequences
An LC/MS analysis of a 21-mer DNA sample using the Thermo Scientific™ DNAPac™ RP column coupled to the Thermo Scientific™ Q-Exactive™ Hybrid Quadrupole-Orbitrap™ mass spectrometer exemplifies analysis and characterization of the full-length product and its impurities (Figure 1). The full length product is separated from its n-1 failure sequence as shown in the UV trace. The MS data confirms the main peak as the desired product. At charge state -4, the monoisotopic m/z for the 21 mer DNA is at 1605.0160 and the mass accuracy is 1.87 ppm (Figure 1b). Since “n-1” failures sequences could arise at any step in the synthesis, the lost base could be from any of the four bases. In this example, the lost base in the n-1 peak was identified by MS (Figure 1c). In this peak, mass values corresponding to the loss of Guanine, Adenine, Cytosine and Thymine were all observed, indicating at least four independent coupling failures. This example demonstrates both the ability to resolve the failures on the DNAPac RP and the value of high resolution mass spectrometers, such as the Q Exactive for fast, high quality identification of target oligonucleotides and their failure sequences.
Separation of Phosphorothioate Modified siRNAs at Different pH values
A common stability-improving modification in DNA and RNA is incorporation of PS linkages. This linkage replaces a non-bridging oxygen with a sulfur atom in the phosphodiester linkage. It also adds a chiral center which, in addition to the chiral centers in the ribo-sugar produces diastereoisomer pairs at each PS linkage. Short interfering (si) RNA oligos harboring one or two of these linkages were prepared and analyzed using the DNAPac RP column. Since oligonucleotides may express different numbers of charges at different pH values, they often exhibit different selectivity at alternate pH values. Hence, separations at pH 7.9 and 9.9 were compared. Figure 2 shows the separation of a sense strand that has one PS substitution incorporated in the 14th linkage.
The two possible diastereoisomers co-eluted at pH 7.9, but were baseline resolved at pH 9.9. Both peaks exhibit identical high-resolution mass (data not shown), indicating the separated forms are diastereoisomers, and not other impurities. In Figure 3, the sense strand is modified with PS at the 19th and 20th linkages. As in Figure 2, the pH 7.9 separation does not resolve the diastereoisomers, but here a significant PS substitution failure (PS/PO) is separated. At pH 9.9, three of four possible diastereoisomers are resolved, and at least one PS/PO substitution failure is also apparent. The mass difference between the early eluting (PS/PO) peak and the target peak was 16 Da, corresponding to the mass difference between the oxygen and sulfur (Figure 3b).
LC/MS Analysis of CpG Methylation and 5’-Phosphorylation
Methylation of CpG dinucleotide islands is a major form of epigenetic modification in eukaryotic genome. This modification is involved in gene suppression, gene regulation and in some cases, development and progression of diseases including cancer. Therefore, detection of CpG methylation is important for epigenetics studies and cancer research. LC/MS analysis of CpG methylation was performed using the DNAPac RP column. In Figure 4, an unmodified oligonucleotide and a CpG methylated oligonucleotide of identical sequence are resolved. Figure 4b shows the -3 charge state of unmodified oligonucleotide at m/z 1517.919 and the -3 charge state of methylated oligonucleotide at m/z 1522.593.The mass difference between the methylated and unmodified peaks corresponds to one methyl group.
Figure 5. Analysis of 5’-phosphorylation
In addition to methylation, the DNAPac RP column effectively resolves 5’-phophorylated, from 5’ hydroxylated oligonucleotides (Figure 5). The MS result confirms the phosphate group difference in the two peaks.
Julia Baek, Ph.D. (email@example.com), is a research scientist, Jim Thayer, Ph.D. (Jim.firstname.lastname@example.org), is staff research biochemist, Hongxia Wang, Ph.D. (Hongxia.email@example.com), is senior marketing specialist, and Ilze Birznieks (Ilze.firstname.lastname@example.org) is the global product manager of the Bio LC column at at Thermo Fisher Scientific.