Ultraviolet Photodissociation and Activated Electron Photodetachment Mass Spectrometry for Top-Down Sequencing of Modified Oligoribonucleotides

Abstract

With the increased development of new RNA-based therapeutics, the need for robust analytical methods for confirming sequences and mapping modifications has accelerated. Characterizing modified ribonucleic acids using mass spectrometry is challenging because diagnostic fragmentation may be suppressed for modified nucleotides, thus hampering complete sequence coverage and the confident localization of modifications. Ultraviolet photodissociation (UVPD) has shown great potential for the characterization of nucleic acids due to extensive backbone fragmentation. Activated electron photodetachment dissociation (a-EPD) has also been used as an alternative to capitalize on the dominant charge-reduction pathway prevalent in UVPD, facilitate dissociation, and produce high abundances of fragment ions. Here, we compare higher-energy collisional activation (HCD), UVPD using 193 and 213 nm photons, and a-EPD for the top-down sequencing of modified nucleic acids, including methylated, phosphorothioate, and locked nucleic acid-modified DNA. The presence of these modifications alters the fragmentation pathways observed upon UVPD and a-EPD, and extensive backbone cleavage is observed that results in the production of fragment ions that retain the modifications and allow them to be pinpointed. LNA and 2′-O-methoxy phosphorothioate modifications caused a significant suppression of fragmentation for UVPD but not for a-EPD, whereas phosphorothioate bonds did not cause any significant suppression for either method. The incorporation of 2′-O-methyl modifications suppressed fragmentation of the antisense strand of patisiran, which resulted in some gaps in sequence coverage. However, UVPD provided the highest sequence coverage when compared to a-EPD.

Top-Down Mass Spectrometry of Synthetic Single Guide Ribonucleic Acids Enabled by Facile Sample Clean-Up

“…For deconvolution parameters, adduct element was H + , charge range was set to 1–50, analyzer type was OT, relative abundance was 0%, isotope table was nucleotide, negative charge was checked, and minimal number detected charge was 1. The mass list of deconvoluted MS/MS spectra was exported to an in-house software, which matches theoretical terminal fragment a , a - B , c , d , w , x , x - B , y , and z to experimental fragment ion mass with a user defined mass error . The mass error was set to 100 ppm initially for comparison evaluation (Figure S1) and then 10 ppm for sequence coverage mapping.…”

“…19,20 Another important factor to achieve good top-down sequencing is having effective fragmentation that can provide enough coverage. 21 Top-down workflows have been shown to nearly completely characterize entire sequences of RNA up to 76-nucleotide long using conventional collisionally activated dissociation (CAD). 13,14 More recently, alternative fragmentation strategies such as negative electron transfer dissociation (NETD), 22,23 electron detachment dissociation (EDD), 13,24 ultraviolet photodissociation (UVPD), 25,26 infrared multiple photon dissociation (IRMPD), 27,28 electron photodetachment dissociation (EPD), 26,29 radical transfer dissociation (RTD), 30 and some combinations of the above 13,21,31 have shown benefits to smaller oligonucleotides but have not yet been heavily explored in larger oligonucleotides.…”

“…The mass list of deconvoluted MS/MS spectra was exported to an in-house software, which matches theoretical terminal fragment a, a-B, c, d, w, x, x-B, y, and z to experimental fragment ion mass with a user defined mass error. 21 The mass error was set to 100 ppm initially for comparison evaluation (Figure S1) and then 10 ppm for sequence coverage mapping.…”

“…Because of this major obstacle, many others have used precipitation methods and molecular weight cutoff filters to desalt and buffer exchange oligonucleotides into desired solvent conditions. , We found these processes labor intensive with sub-optimal consistency. While an acidic wash was also found to be beneficial in removing salt adducts, periodic system conditioning was required. , Another important factor to achieve good top-down sequencing is having effective fragmentation that can provide enough coverage . Top-down workflows have been shown to nearly completely characterize entire sequences of RNA up to 76-nucleotide long using conventional collisionally activated dissociation (CAD). , More recently, alternative fragmentation strategies such as negative electron transfer dissociation (NETD), , electron detachment dissociation (EDD), , ultraviolet photodissociation (UVPD), , infrared multiple photon dissociation (IRMPD), , electron photodetachment dissociation (EPD), , radical transfer dissociation (RTD), and some combinations of the above ,, have shown benefits to smaller oligonucleotides but have not yet been heavily explored in larger oligonucleotides.…”

Spacer Fidelity Assessments of Guide RNA by Top-Down Mass Spectrometry

“…These methodologies also have challenges with assessing RNA with many chemically modified nucleotides. , Nanopore, on the other hand, while more direct has often been found to be inaccurate due to the lack of base calling routines that can account for modifications . Mass spectrometry (MS) has successfully been employed to characterize both RNA and DNA but has generally proved best suited for smaller oligonucleotides. − Recently, MS-based methodologies have been applied to larger RNA; however, such techniques have most often relied on digesting the RNA first into smaller pieces prior to mass spectrometry analysis, also known as bottom-up sequencing. − These workflows often lead to small fragments that lack sequence specificity and have the potential to introduce sample preparation-induced artifacts. , Top-down or direct sequencing of RNA by mass spectrometry has been applied to modified tRNA, , but this RNA type is considerably shorter and less modified than CRISPR gRNA. Furthermore, sequence fidelity assessed by highly sensitive and quantitative methodologies was not evaluated.…”