High resolution 4D HPCH experiment for sequential assignment of 13C-labeled RNAs via phosphodiester backbone
The three-dimensional structure determination of RNAs by NMR spectroscopy requires sequential resonance assignment, often hampered by assignment ambiguities and limited dispersion of 1H and 13C chemical shifts, especially of C4′/H4′. Here we present a novel through-bond 4D HPCH NMR experiment involving phosphate backbone where C4′–H4′ correlations are resolved along the 1H3′–31P spectral planes. The experiment provides high peak resolution and effectively removes ambiguities encountered during assignments. Enhanced peak dispersion is provided by the inclusion of additional 31P and 1H3′ dimensions and constant-time evolution of chemical shifts. High spectral resolution is obtained by using non-uniform sampling in three indirect dimensions. The experiment fully utilizes the isotopic 13C-labeling with evolution of C4′ carbons. Band selective 13C inversion pulses are used to achieve selectivity and prevent signal dephasing due to the C4′–C3′ and C4′–C5′ homonuclear couplings. Multiple quantum line narrowing is employed to minimize sensitivity loses. The 4D HPCH experiment is verified and successfully applied to a non-coding 34-nt RNA consisting typical structure elements and a 14-nt RNA hairpin capped by cUUCGg tetraloop.
KeywordsRNA resonance assignment HCP Selective pulses Four-dimensional NMR Non-uniform sampling
Recent advances in RNA research has led to the discovery of several new classes of non-coding RNAs (e.g. siRNA, miRNAs) many of whose functions remain largely unknown. The function of RNA during normal and diseased states has long been studied through its structure–function relationship (Briones et al. 2009; Mercer et al. 2009). Various NMR approaches (Varani et al. 1996; Wijmenga and van Buuren 1998; Furtig et al. 2003; Flinders and Dieckmann 2006) have successfully been utilized in this regard which expanded our knowledge about RNAs structure. However, precise three-dimensional structure determination of even moderately sized RNAs is still a bottleneck due to spectral overlap.
Sequential assignment of chemical shifts is one of the prerequisite for RNA structure determination by NMR spectroscopy. The assignment is usually based on through-space 2/3D NOE-type (Nikonowicz and Pardi 1993) and/or through-bond 3D HCP-type (Marino et al. 1994) experiments. However, most often the efficacy of both type of approaches is severely affected by low chemical shift dispersion of the involved nuclei, which results into spectral overlaps making the unambiguous resonance assignment a challenging task. In non-coding RNAs, similar chemical shifts in helical secondary structures and frequent lack of base stacking make the spectral crowding particularly demanding. Recently, an alternate automated approach involving no isotope labeling (Aeschbacher et al. 2013) for RNAs was proposed that requires peak lists from 2D TOCSY, 2D NOESY and natural abundance 1H–13C HSQC spectra. However, owing to irregular and very limited statistical data basis, difficulty may arise while assigning certain regions of RNAs. In addition, chemical shift degeneracy or low dispersion still remain a challenge for such approaches. These difficulties can be alleviated using similar high dimensional approaches (Kazimierczuk et al. 2013; Mobli and Hoch 2014b; Nowakowski et al. 2015) that have been successfully applied to proteins (Stanek et al. 2013a, c; Geist et al. 2013a, b; Bermel et al. 2013). A high-resolution 4D Caro,Cribo-NOESY experiment (Stanek et al. 2013b) was recently reported which provides intra and inter-nucleotide NOEs in RNA. Since NOE interactions have conformation dependencies, ambiguities and missing links may still arise during assignments via NOESY spectrum. One approach to find such missing links in NOESY spectrum is by retrieving complementary information from some through-bond correlations. To address this issue we recently reported a high-resolution 4D HC(P)CH experiment (Saxena et al. 2014) that provides sequential connectivities through linking the neighbouring H4′–C4′ planes. Although the experiment performs well for the regions having stem and bulges in non-coding RNAs, it faces difficulties with internal- and hairpin-loop regions. As the experiment involves two long transfer periods (C4′ → P and P → C4′), it significantly relies on C4′–P (3J) couplings, which are usually weaker for such regions (Schwalbe et al. 1994; Legault et al. 1995; Hu et al. 1999), an efficient coherence transfer could become an issue. Combined with long 13C–31P coupling-refocusing and coupling-evolution delays on phosphorous nuclei the NMR signal linking aforementioned regions could have low S/N ratio. It is evident that similar approaches involving more favourable couplings and shorter delays can be quite beneficial.
An alternative, but somewhat underutilized approach is to achieve sequential H3′i−1–Pi connectivities via H3′–P couplings (Sklenář et al. 1986; Kellogg 1992; Varani et al. 1995). The reasonably favourable 1H–31P (3J) couplings (~7 Hz) can be utilized to directly transfer the initial coherence to phosphorous nuclei and enable sequential assignment via backbone. With recent advancement in high dimensional NMR techniques (Kazimierczuk et al. 2010, 2013; Mobli and Hoch 2014a) this approach can be effectively extended to moderately sized RNAs.
Here we report a high resolution C4′/H4′ selective 4D HPCH experiment which includes the chemical shifts evolution of 1H3′, 31P, 13C4′ and 1H4′ of adjoining nucleotides thereby linking them in a single experiment providing higher peak resolution. The relatively better dispersed H3′ chemical shifts aided by additional P dimension and constant time evolution of all indirect dimensions collectively facilitated to resolve poorly overlapped C4′–H4′ region. Compared to 3D HCP and 4D HC(P)CH, which initially use two coherence transfer steps (H4′ → C4′ and then C4′ → P), 4D HPCH use only one step to transfer the coherence onto phosphorous nuclei (H3′ → P) using 1H–31P (3J) couplings. The proposed 4D HPCH experiment provides sequential connectivities via H3′i−1–Pi–C4′i–H4′i links. For the cases where H5′i–Pi couplings are not weak, an additional peak giving rise to H5′i–Pi–C4′i−1–H4′i−1 links further facilitates the sequential assignment. We implemented the non-uniform sampling (NUS) to achieve high resolution in all indirect dimensions. The experiment efficiently facilitates the assignment of sequential links in a non-coding RNA including the internal- and hairpin-looped regions.
Materials and methods
Two RNA samples are used to test and verify the efficacy of 4D HPCH experiment. The 14-nt cUUCGg-hairpin (5′-GGCAC(UUCG)GUGCC-3′; bold residues indicate the tetraloop-residues) was purchased from Silantes GmbH (Munich, Germany) as a uniformly 13C,15N-labeled RNA. The concentration of the NMR sample was 0.7 mM in 20 mM KH2PO4/K2HPO4, pH 6.4, 0.4 mM EDTA in D2O. Second RNA sample was 13C,15N-labeled 34-nt hairpin RNA LCS1co consisting of two A-RNA form stems, one adenine bulge, an asymmetric internal loop and a GAAA terminal loop (Cevec et al. 2010). RNA sample was prepared by in vitro transcription with T7 RNA polymerase (Promega), purified by denaturating PAGE, electroeluted and dialyzed against an NMR buffer. The concentration of RNA sample was 1.5 mM in 10 mM sodium phosphate buffer, pH 6.8, 20 mM NaCl in D2O.
In the consecutive block, the coherence is forward transferred to C4′i where its chemical shifts are indirectly recorded (t3). Since, as it was shown earlier for nucleic acids (Fiala et al. 1998; Fiala et al. 2000), the dominant 1H–13C dipolar relaxation mechanism is significantly attenuated for zero- and double-quantum coherences, MQ coherences are preserved during the frequency labeling of C4′ evolution period. During the same period (τc) refocusing of P–C4′ anti-phase coherence is also achieved by application of a 180° pulse on 31P in synchrony with the moving 180° pulse on 13C channel. Another 180° pulse is centrally placed on 1H channel to refocus its chemical shift evolution. During this period, the evolution due to homonuclear carbon coupling (C4′–C3′ and C4′–C5′) is refocused by two cosine modulated IBURP-2 (Geen and Freeman 1991) pulses (P in Fig. 2) which selectively and simultaneously invert the frequency bands of C3′ and C5′ ribose sugar carbons. Since C2′ carbons share the same spectral region as of C3′, inversion of later also inverts the C2′ carbons. Effectively, the use of inversion pulses leads to an indirect selection of C4′ and hence H4′ during this period. Finally, an in-phase coherence is generated on H4′ spins by refocused INEPT transfer during ∆ period.
The inversion profiles for shaped pulses are simulated and tested using Spinach library (Hogben et al. 2011) on MATLAB®. Gradients and phase cycling are employed to eliminate undesired coherences and improve the C4′/H4′ selectivity of experiment. After P → C4′ transfer period, spin-lock pulses (SLx, SLy) are employed (W in Fig. 2) to dephase any remaining transverse water magnetization. To achieve higher dimensionality with reasonable resolution in the indirectly detected dimensions, non-uniform sampling was employed. Using NUS we are able to acquire 4D HPCH experiment with high evolution times: 18 ms (t1), 18 ms (t2) and 14 ms (t3).
The processing of 4D NUS data was accomplished by the home-written software package SSA (Signal Separation Algorithm) (Stanek et al. 2012), which can be downloaded free of charge for non-commercial purposes from the website http://nmr.cent3.uw.edu.pl.
Results and discussion
The 4D HPCH spectrum can be easily analyzed with any NMR assignment programme. For this study we used SPARKY (Goddard and Kneller 2004) program and spectrum was analyzed by fixing two dimensions (H3′i−1 and Pi) of corresponding nucleotides (for illustration see Fig. 3a and b). The resulting 2D C4′–H4′ plane consists of intra- and inter-nucleotide peaks, i.e. to the (i − 1)th and ith residue. Consecutively, (i − l)th C4′/H4′ are fixed and H3′i−2–Pi−1 peak is assigned on H3′–P plane (which provides sequential link to previous nucleotide via H3′i−2 protons (see Fig. 3b). In earlier approaches, 3D HCP and 4D HC(P)CH, the coherence transfer efficiencies vary largely with C4′–P couplings and, in turn, with the β/ε torsional angles in different regions in RNA. For the regions where the C4′ → P couplings are weak, an efficient coherence transfer is difficult to achieve. In the presented method, however, coherence transfer is initiated by utilizing H3′–P couplings (3JC4′,P) which are comparatively large for aforementioned regions. This approach partially compensated the coherence transfer efficiency loss and assignment of correlation peaks belonging to such regions was achieved. For the cases which are not sensitivity limited, the proposed experiment can provide nearly complete assignment within the regions involving internal/hairpin loops in RNA. For 14-nt cUUCGg-tetraloop RNA used in this study assignment 11 out of 13 sequential peaks, including the connectivities present within cUUCGg hairpin loop, was achieved using 4D HPCH spectrum. The two missing correlations were due to small couplings (U6–U7) and heterogeneity at the 5′-end (G1–G2).
In summary, we have presented a novel NMR pulse sequence for sequential assignment in RNAs via highly-resolved C4′–H4′ correlation. The sequential peaks were effectively resolved along the intranucleotide H3′–P planes. The experiment ideally complement the 4D HC(P)CH experiment by relying on different coupling constants for magnetization transfer to generate uniformly intense cross peaks in different regions of the RNA spectrum. The attenuated relaxation due to MQ coherences and C4′/H4′ selectivity in the experiment partially compensated the sensitivity losses entailing the increased dimensionality. Band selective inversion pulses were used to prevent signal modulation due to C4′–C3′, C4′–C5′ couplings and to indirectly select the C4′/H4′ region. Employing high dimensionality and non-uniform sampling, the novel method allows for the assignment of most of the sequential links in highly overlapped NMR spectra of RNAs. Given the outstanding sensitivity and spectral resolution of the NMR approach, we envisage useful applications for the assignment of hairpin/internal loop and bulge regions in RNAs. The comparative study with other experiments suggests that the presented experiment can be especially beneficial for assigning such regions in RNA. Despite lower sensitivity, the proposed experiment clearly outperforms the conventional sequential correlation approaches for RNAs which suffer from critical spectra overlap. The experiment is proposed as a complementary tool to 3D HCP, 3D/4D NOESY experiments and augments the set of high dimensional experiments aimed at improving resolution and reducing ambiguities during resonance assignments in RNAs with poor chemical shift dispersion.
This work was supported by Polish National Science Centre, S.S. and W.K. thank for their financial support (NCN Grant No. 2013/11/N/ST4/01827). M.C. and J.P. thank Slovenian Research Agency (ARRS, P1-242 and J1-6733). The study was carried out at the Biological and Chemical Research Centre, University of Warsaw, established within the project co-financed by European Union from the European Regional Development Fund under the Operational Programme Innovative Economy.
- Aeschbacher T, Schmidt E, Blatter M, Maris C, Duss O, Allain FH-T, Güntert P, Schubert M (2013) Automated and assisted RNA resonance assignment using NMR chemical shift statistics. Nucleic Acids Res 41:e172Google Scholar
- Goddard T, Kneller D (2004) SPARKY 3, vol 14. University of California, San Francisco, p 15Google Scholar
- Marino JP, Schwalbe H, Anklin C, Bermel W, Crothers DM, Griesinger C (1994) Three-dimensional triple-resonance 1H, 13C, 31P experiment: sequential through-bond correlation of ribose protons and intervening phosphorus along the RNA oligonucleotide backbone. J Am Chem Soc 116:6472–6473CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.