Chemical shift prediction of RNA imino groups: application toward characterizing RNA excited states

Abstract

NH groups in proteins or nucleic acids are the most challenging target for chemical shift prediction. Here we show that the RNA base pair triplet motif dictates imino chemical shifts in its central base pair. A lookup table is established that links each type of base pair triplet to experimental chemical shifts of the central base pair, and can be used to predict imino chemical shifts of RNAs to remarkable accuracy. Strikingly, the semiempirical method can well interpret the variations of chemical shifts for different base pair triplets, and is even applicable to non-canonical motifs. This finding opens an avenue for predicting chemical shifts of more complicated RNA motifs. Furthermore, we combine the imino chemical shift prediction with NMR relaxation dispersion experiments targeting both 15N and 1HN of the imino group, and verify a previously characterized excited state of P5abc subdomain including an earlier speculated non-native G•G mismatch.

Rapid measurement of Watson-Crick to Hoogsteen exchange in unlabeled DNA duplexes using high-power SELOPE imino <sup>1</sup>H CEST

“…It should be noted that to avoid any complexities due to NOE effects with water protons or hydrogen exchange, we restricted the offset to -6 ppm -6 ppm when analyzing and fitting the 1 H CEST profiles. This is common practice as relatively narrow offsets (< 4 ppm) were used in prior 1 H CEST studies of both nucleic acids (Dubini et al, 2020;Wang et al, 2021;Liu et al, 2020) 2017a; Yuwen et al, 2017b). While we did not observe a dip near the water chemical shift in the 1 H CEST profile for the internal residue T5-H3, a weak and broad dip near the water chemical shift was observed in the profile for the near terminal residue G2-H1 (Fig.…”

“…CC BY 4.0 License. Frank et al, 2013;Wang et al, 2021;Swails et al, 2015;Czernek et al, 2000;Lam and Chi, 2010).…”

“…The SELOPE experiment has already found several applications in studies of unlabeled nucleic acids, including in the characterization of fast (kex = k1 + k-1 > 1,000 s -1 ) RNA secondary structural rearrangements (Baronti et al, 2020) and DNA base opening (Furukawa et al, 2021), as well as slower (kex < 100 s -1 ) DNA hybridization kinetics (Dubini et al, 2020). Many 1 H relaxation dispersion (RD) studies have targeted exchangeable imino protons (Baronti et al, 2020;Furukawa et al, 2021), taking advantage of the well-known dependence of the imino 1 H chemical shifts on secondary structure (Wang et al, 2021;Lam and Chi, 2010).…”

“…An important consideration when performing 1 H CEST experiments are contributions due to 1 H-1 H cross-relaxation, which may give rise to extraneous NOE dips in the 1 H CEST profiles (Ishima et al, 1998;Lundstrom and Akke, 2005;Eichmuller and Skrynnikov, 2005;Bouvignies and Kay, 2012;Sekhar et al, 2016;Yuwen et al, 2017a;Yuwen et al, 2017b). These contributions have been suppressed in proteins through deuteration (Eichmuller and Skrynnikov, 2005;Lundstrom and Akke, 2005;Lundstrom et al, 2009;Otten et al, 2010;Hansen et al, 2012;Weininger et al, 2012), and in 15 N isotopically labelled proteins (Yuwen et al, 2017a;Yuwen et al, 2017b) and nucleic acids (Wang et al, 2021;Liu et al, 2020) using spin-state-selective magnetization transfer schemes, and through selective inversion of protons combined with use of short relaxation times (Lundstrom and Akke, 2005;Schlagnitweit et al, 2018). In the SELOPE experiment, imino protons are selectively excited and the magnetization belonging to non-imino protons is dephased prior to application of the B1 field.…”

“…Furthermore, many motifs of interest involve damaged or modified nucleotides, which are difficult to isotopically enrich with 13 C and 15 N nuclei. It is for this reason that we turned our The utility of protons as probes in CEST (Chen et al, 2016;Dubini et al, 2020;Wang et al, 2021;Liu et al, 2020), Carr-Purcell-Meiboom-Gill (CPMG) (Juen et al, 2016;Leblanc et al, 2018), and off-resonance R1ρ experiments (Wang and Ikuta, 1989;Lane et al, 1993;Steiner et al, 2016;Schlagnitweit et al, 2018;Baronti et al, 2020;Furukawa et al, 2021) to study conformational exchange in nucleic acids is now well-established. Many of these 1 H based approaches use experiments originally developed to study conformational exchange in proteins (Ishima et al, 1998;Eichmuller and Skrynnikov, 2005;Lundstrom and Akke, 2005;Lundstrom et al, 2009;Otten et al, 2010;Bouvignies and Kay, 2012;Hansen et al, 2012;Weininger et al, 2012;Weininger et al, 2013;Smith et al, 2015;Sekhar et al, 2016;Yuwen et al, 2017a;Yuwen et al, 2017b).…”