(T1130-02-09) Considerations for Probe Selection for Oligonucleotide Hybridization LC/MS Workflows

Purpose: LC/MS Bioanalysis of oligonucleotides has had its historical challenges in all areas of the workflow including the extraction, liquid chromatography, and mass spectrometry detection. One of the main pain points in the extraction of oligonucleotides is poor recovery from nonspecific binding or poor extraction efficiency using either the common extraction approachs. Recent publications of a more specific biotinylated probe hybridization approach has addressed both challenges of nonspecific binding and extraction efficiencies, but there hasn’t been much discussion on the specific advantages for the types of probes that can be used for this hybridization work. Although there has been a focus on DNA, LNA, and PNA probe design with research to demonstrate the specific attributes each offer, there has been limited discussions of the overall impact for recovery and interferences. What has been detailed in the design of the biotinylated probe is a focus to limit self hybridization of the probe while maintain a complimentary sequence with a high enough score to out compete any interferences in matrix for ASO’s or from the sense strand in modalities like siRNA while keeping the Tm of the hybridized duplex low enough to recovery from streptavidin bead. Recently, while developing an assay for a siRNA complex using the LNA approach we observed interference from the probe to the antisense strand when it was analyzed on the mass spectrometer. We modified the melting temperature to release the streptavidin/biotinylated hybridized to avoid as much of the biotinylated probe/antisense complex, but limited release still had enough of the probe in the final extracts to afford interference to the LLOQ samples. Modifying LC conditions also did not resolve the interference, but that was not successful due to peak shape issues with the internal standard. Going back to the design of the probe we looked to use a PNA probe with the intention that any residual PNA probe would be easy to resolve from a chromatographic approach.

Methods: A 25.0uL aliquot of rat plasma fortified with the siRNA was placed into a 96 well Kingfisher plate followed by 100 uL Loading Buffer, Tris digestion buffer and TCEP. Each sample was then fortified with the PNA probe and Proteinase K. The plate was incubated at 65°C for approximately one hour and then cooled to hybridize the PNA probe to the antisense strand. The plate was then processed on the Kingfisher and the antisense strand was then melted off of the streptavidin beads to clean 96 well plate containing 100 µL of final solution reagents in water and 10.0 uL of working IS and stored at 5°C until analysis. The extracts were then analyzed on a Sciex API 7500 with an ESI (-) source using a Thermo, DNApac, 2.1 x 50mm, 4.0 um column.

Results: Initial method development focused on the production of LNA probes using the workflow designated by the paper authored by Karan Agrawal (Bioanalysis-2023-0079) paying close attention self hybridization score and maximizing the melting temperatures of the LNA-AS complex to be above 80°C but below 90°C. To determine final methodology for extraction of the AS strand while keeping the aliquot size at 25.0 uL, we ran experiments to determine conditions for the LNA capture probe, the amount of Proteinase K, and the digestion time and time required to anneal the capture probe to the antisense strand. After working out the initial extraction conditions it was noted that we were seeing interference in the IS only samples which was determined to be due to free LNA-probe being in the final extracts. We worked on optimizing the conditions of the “melt” portion on the Kingfisher but we were not able to completely eliminate the interference. We focused on LC conditions to minimize the interference but once again ran into challenges of getting optimal peak shape for both the AS and the IS while removing the interference. We then looked to the probe type with the theory that if the probe were a PNA instead of a LNA we could easily eliminate any interference with LC conditions common to oligonucleotides since the backbone of the PNA is positive instead of negative. After redesign of the capture probe going from a LNA to a PNA we were able to afford extraction conditions that had high recovery and not interference from the capture probe.

Conclusion: The use of biotinylated capture probes for the extraction of oligonucleotides is gaining a lot interest in the bioanalytical community since it affords very clean extracts, high recovery, and low LLOQs. But there is still a lot of research to be done in order to make the design of the capture probes more efficient which then reduces the overall time required to develop a method. Here we outline another approach that can be used in cases where the LNA capture probe may produce interference to the oligonucleotide being analyzed in the method.

References: Karan Agrawal and et al, Bioanalysis-2023-0079