Oligonucleotides are an interesting class of therapeutic molecules in the bioanalytical world. They’re not as small as New Chemical Entities (NCEs) and not as large as New Biological Entities (NBEs) and there are no specific guidelines for their development. In our laboratory at Boehringer Ingelheim Pharmaceuticals, we’re tasked with providing bioanalytical support for the development of therapeutic oligonucleotides. While there are other assay platforms we could choose (e.g., RT-qPCR, ELISA), we found high resolution LC-MS using the SCIEX TripleTOF® 6600 system has some distinct advantages:
- Sensitivity < ng/mL (pg sensitivity for MRM mode)
- Dynamic range 104
- Can be used for regulatory bioanalysis
- Can be used for metabolite profiling/ID and quantitation
That last bullet regarding metabolite analysis is particularly exciting to us and is one of the most important future developments for LC-MS, especially since it can all be done in one injection.
I recently gave a webinar titled Oligonucleotide Quantitation and Characterization by High Resolution LC-MS. In the webinar, I demonstrated both high resolution LC-MS and MS/MS and how the same method can be used for different species (monkey, rat, and mouse) and matrices (plasma and liver tissue). Additionally, we were able to show how you can perform quantitation of a parent oligo and its metabolites without metabolite references. There were so many questions afterward, that we were unable to answer all of them during the live session, so I’ve provided more answers below. I hope you’ll see after listening to the webinar and reading the Q&A how powerful high resolution LC-MS can be for oligonucleotide analysis and the additional and exciting capabilities it can offer.
1. Do the ion-pairing reagents dirty the MS quickly? How often do you need to clean the ion source?
High molar ratios of ion-pairing reagents have the potential to dirty MS systems quickly. But ion source designs make a big difference. The orthogonal spray, actively vented source and ion optics (rather than an ion transfer tube) of SCIEX systems make them much less susceptible to ion-pairing reagent contamination.
2. How stable are oligo signals in the MS? Do you need to tune the MS more often?
The signals are very stable and there is no need to tune the MS that often, but you do need to clean the ion source more frequently.
3. Can you sum all ion charge states of an oligo analyte for quantitation?
Yes! Just make sure there are no interferences for each charge state.
4. Can you use columns other than reverse phase for LC/MS analysis of oligos?
Anion exchange columns with pH gradients would be next to try.
5. Is adding organic solvent in recon not enough to break the duplex?
We didn’t observe breakdown of double helix in the presence of 15% ACN.
6. How do you choose your ion-pairing reagents?
Do you change ion-pairing reagents significantly from oligo to oligo? Details for selection of ion-pairing reagents and optimization can be found in this reference: Journal of Pharmaceutical and Biomedical Analysis 138 (2017) 146–152.
7. Why do you think the dynamic range was narrower for the MRM vs HRMS methods?
I have not fully investigated this but most likely this is due to collision energy for that particular MRM pair.
8. For the HRMS you didn’t use any fragmentation?
No, not for these studies, but you definitely can. You can either specify the precursor for ion fragmentation or use SWATH® acquisition. In the MRM mode, we use high resolution for fragment ions.
9. Can you please re-explain the IP-HPLC conditions for analyzing your OGNs?
There is a good review paper on this topic: Journal of Pharmaceutical and Biomedical Analysis 138 (2017) 146–152.
10. The r=0.9945 equates to r2=0.989, any comments?
A value of r2 >0.98 is acceptable.
11. Can HRMS identify and characterize long term oligo repeats in disease states like Huntington’s (e.g., 36 or more CAG trinucleotide repeats)?
That is technically possible but has not been tested.
12. What software is used for the characterization of the oligonucleotides?
We used SCIEX PeakView® software. SCIEX data can also be analyzed in ProMass.
13. With your LLOQ for your study case, did you have some carryover issues? If yes, how did you get rid of that?
Please listen to my answer at 49:47 minutes.
14. If there were no reference standards for metabolites, how was the sensitivity evaluated against the parent?
Please listen to my answer at 50:38 minutes.
15. Do you have software to help with the metabolite ID?
Yes! for both metabolite ID and MS/MS fragmentation interpretation. The software package is ProMass for SCIEX.
16. Did you encounter any reproducibility issues due to non-specific binding?
Please listen to my answer at 51:46 minutes.
17 Have you compared the sensitivity between the TripleTOF® 5600 and 6600 systems?
Yes, in two different labs. The TripleTOF 6600 is better than the 5600.
18. Is adduct formation mostly from the HPLC system and the mobile phase? Does the MS system need to be cleaned before running oligo samples? Please listen to my answer at 55:17 minutes.
19. If MS response depends on the number of phosphate groups, how can you use the UV response for quantitation of truncated metabolites that have fewer phosphate groups?
Please listen to my answer at 52:55 minutes.
20. What is the guide strand and passenger strand?
For siRNA, the guide strand is the active strand which binds to target mRNA, whereas the passenger strand is not active and doesn’t interact with target mRNA.
21. Have you tried HILIC columns for this analysis?
This was addressed in the live portion. Please listen to my answer at 57:12 minutes.
22. Can you comment on the concentrations of HFIP and DIPA that you used? Did you need optimization for different oligonucleotides?
This was addressed in the live portion. Please listen to my answer at 53:54 minutes. Also please see this very good paper for more details: Journal of Pharmaceutical and Biomedical Analysis 138 (2017) 146–152.
23. What kind of containers can be used to store and prepare plasma samples?
We normally store plasma in freezers -20 or -70 in polypropylene.
24. How do you produce the higher charge states?
Charge state is usually a function of the length of the oligo. The longer the oligo, the higher the likelihood of getting another negative charge due to a phosphate group. However, charge state distributions can be tuned slightly with source voltages. Also, different ion sources will give different charge state distributions.
25. Is it necessary to prepare mobile phase fresh every day?
Yes, in our case we did. It is generally recommended to make fresh mobile phase every day.
26. Do you think that SFC (UPC2 with CO2) can be a nice alternative to avoid adducts prior to MS analysis?
Good idea! I am not aware that anyone has tried that.