SOLID-STATE NUCLEAR MAGNETIC RESONANCE (ssNMR) METHOD FOR DETECTING HYDROGEN BOND STRUCTURE

Abstract

An experimental technology for detecting a hydrogen bond based on an ssNMR technology includes: (1) exciting a .sup.1H nucleus of an RNA sample with a π/2 pulse; (2) applying two π pulses every half rotation period on an X_nucleus of the RNA sample; (3) applying a π pulse on the .sup.1H nucleus of the RNA sample; (4) applying two π pulses every half rotation period on the X nucleus of the RNA sample; (5) applying a 90° pulse on .sup.1H and X atoms of the RNA sample; (6) recording a chemical shift of X in indirect dimension; (7) applying the 90° pulse on the .sup.1H and X nuclei of the RNA sample; (8) repeating steps 2, 3, and 4; and (9) collecting the .sup.1H signal in direct dimension; where X is selected from the group consisting of .sup.15N and .sup.13C.

Claims

1. A solid-state nuclear magnetic resonance (ssNMR) method for detecting a hydrogen bond structure, comprising the following steps: (1) exciting a .sup.1H nucleus of an RNA sample with a π/2 pulse; (2) applying two π pulses every half rotation period on an X nucleus of the RNA sample; (3) applying a π pulse on the .sup.1H nucleus of the RNA sample; (4) applying two π pulses every half rotation period on the X nucleus of the RNA sample; (5) applying a 90° pulse on .sup.1H and X atoms of the RNA sample; (6) recording a chemical shift of X in indirect dimension; (7) applying the 90° pulse on the .sup.1H and X nuclei of the RNA sample; (8) repeating steps 2, 3, and 4; and (9) collecting the .sup.1H signal in direct dimension; wherein X is selected from the group consisting of .sup.15N and .sup.13C.

2. The ssNMR method for detecting a hydrogen bond structure according to claim 1, wherein the .sup.1H nucleus is applied with 1.9 .Math.s of the π/2 pulse and 3.8 .Math.s of the π pulse.

3. The ssNMR method for detecting a hydrogen bond structure according to claim 1, wherein the .sup.15N nucleus is applied with 9.2 .Math.s of the π pulse.

4. The ssNMR method for detecting a hydrogen bond structure according to claim 1, wherein the .sup.13C nucleus is applied with 7.0 .Math.s of the π pulse.

5. The ssNMR method for detecting a hydrogen bond structure according to claim 1, wherein the method further comprises the following steps between steps (1) and (2): applying.sup.1H-.sup.15N cross-polarization (CP); applying a shape pulse on the .sup.15N nucleus; suppressing a water signal with a MISSISIPPI pulse sequence; and applying.sup.55N-.sup.1H CP.

6. The ssNMR method for detecting a hydrogen bond structure according to claim 5, wherein .sup.1H-.sup.15N CP is applied at a contact time of 1 ms, a .sup.15N constant field lock at 50 kHz, and a .sup.1H constant field lock at 90 kHz.

7. The ssNMR method for detecting a hydrogen bond structure according to claim 5, wherein the shape pulse is applied on the .sup.15N nucleus at an excitation time of 625 .Math.s.

8. The ssNMR method for detecting a hydrogen bond structure according to claim 5, wherein .sup.15N-.sup.1H CP is applied at a contact time of 300 .Math.s, a .sup.15N constant field lock at 50 kHz, and a .sup.1H constant field lock at 90 kHz.

9. The ssNMR method for detecting a hydrogen bond structure according to claim 5, wherein the shape pulse is selected from the group consisting of Q3, GAUSSI, SQUARE, SINC, RSONB, REBurp, CRP, and REBURP.

10. The ssNMR method for detecting a hydrogen bond structure according to claim 2, wherein the method further comprises the following steps between steps (1) and (2): applying.sup.1H-.sup.15N CP; applying a shape pulse on the .sup.15N nucleus; suppressing a water signal with a MISSISIPPI pulse sequence; and applyíng.sup.15N-.sup.1H CP.

11. The ssNMR method for detecting a hydrogen bond structure according to claim 3, wherein the method further comprises the following steps between steps (1) and (2): applying.sup.1H-.sup.15N CP; applying a shape pulse on the .sup.15N nucleus; suppressing a water signal with a MISSISIPPI pulse sequence; and applying.sup.15N-.sup.1H CP.

12. The ssNMR method for detecting a hydrogen bond structure according to claim 4, wherein the method further comprises the following steps between steps (1) and (2): applying.sup.1H-.sup.15N CP; applying a shape pulse on the .sup.15N nucleus; suppressing a water signal with a MISSISIPPI pulse sequence; and applying.sup.15N-.sup.1H CP.

13. The ssNMR method for detecting a hydrogen bond structure according to claim 10, wherein .sup.1H-.sup.15N CP is applied at a contact time of 1 ms, a .sup.15N constant field lock at 50 kHz, and a .sup.1H constant field lock at 90 kHz.

14. The ssNMR method for detecting a hydrogen bond structure according to claim 11, wherein .sup.1H-.sup.15N CP is applied at a contact time of 1 ms, a .sup.15N constant field lock at 50 kHz, and a .sup.1H constant field lock at 90 kHz.

15. The ssNMR method for detecting a hydrogen bond structure according to claim 12, wherein .sup.1H-.sup.15N CP is applied at a contact time of 1 ms, a .sup.15N constant field lock at 50 kHz, and a .sup.1H constant field lock at 90 kHz.

16. The ssNMR method for detecting a hydrogen bond structure according to claim 10, wherein the shape pulse is applied on the .sup.15N nucleus at an excitation time of 625 .Math.s.

17. The ssNMR method for detecting a hydrogen bond structure according to claim 11, wherein the shape pulse is applied on the .sup.15N nucleus at an excitation time of 625 .Math.s.

18. The ssNMR method for detecting a hydrogen bond structure according to claim 12, wherein the shape pulse is applied on the .sup.15N nucleus at an excitation time of 625 .Math.s.

19. The ssNMR method for detecting a hydrogen bond structure according to claim 10, wherein .sup.15N-.sup.1H CP is applied at a contact time of 300 .Math.s, a .sup.15N constant field lock at 50 kHz, and a .sup.1H constant field lock at 90 kHz.

20. The ssNMR method for detecting a hydrogen bond structure according to claim 11, wherein .sup.15N-.sup.1H CP is applied at a contact time of 300 .Math.s, a .sup.15N constant field lock at 50 kHz, and a .sup.1H constant field lock at 90 kHz.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0015] FIG. 1A shows a schematic diagram of a two-dimensional .sup.1H-.sup.15N TEDOR pulse sequence; where Φ1 =x,x,x,x,-x,v,-x,-x; Φrec=x,-x,x, -x, -x, x, -x, x; rr/2 represents half a rotation period;

[0016] FIG. 1B shows a schematic diagram of a two-dimensional .sup.1H-.sup.13C TEDOR pulse sequence; where Φ1=x,x,x,x,-x, -x, -x, Φrec=x,-x,x, -x,-x,x,-x,x; rr/2 represents half a rotation period;

[0017] FIG. 1C shows a schematic diagram of a two-dimensional .sup.1H-.sup.15N-hNH(FSL)-[.sup.1H,.sup.15N]-TEDOR pulse sequence; where Φ1=x,x,-x,-x; Φ2=y,y,y,y, -y.-y,-y,-y; Φrec=y,-y,-y,y,-y,y,y,-y; τr/2 represents half a rotation period;

[0018] FIG. 1D shows a schematic diagram of a two-dimensional .sup.1H-.sup.13C-hNH(FSL)-[.sup.1H,.sup.13C]-TEDOR pulse sequence; where Φ1=x,x,-x,-x; Φ2=y,y,y,y, -y,-y,-y,-y; Φrec=y,-y,-y,y,-y,y,y,-y; rr/2 represents half a rotation period;

[0019] FIG. 2A shows a two-dimensional .sup.1H-.sup.15N TEDOR spectrum of a .sup.13C,.sup.15N isotopes-uniformly labeled (GGGGCC).sub.5RNA gel with a total evolution time of 100 .Math.s (n=1);

[0020] FIG. 2B shows a two-dimensional .sup.1H-.sup.15N TEDOR spectrum of a .sup.13C,.sup.15N isotopes-uniformly labeled (GGGGCC).sub.5RNA gel with a total evolution time of 400 .Math.s (n=4);

[0021] FIG. 2C shows a two-dimensional .sup.1H-.sup.15N TEDOR spectrum of a .sup.13C,.sup.15N isotopes-uniformly labeled (GGGGCC).sub.5RNA gel with a total evolution time of 600 .Math.s (n=6);

[0022] FIG. 3A shows a two-dimensional .sup.1H-.sup.13C TEDOR spectrum of a .sup.13C,.sup.15N isotopes-uniformly labeled (GGGGCC).sub.5RNA gel with a total evolution time of 200 .Math.s (n=2);

[0023] FIG. 3B shows a two-dimensional .sup.1H-.sup.13C TEDOR spectrum of a .sup.13C,.sup.15N isotopes-uniformly labeled (GGGGCC).sub.5RNA gel with a total evolution time of 600 .Math.s (n=6);

[0024] FIG. 3C shows a schematic diagram of a G-C paired structure;

[0025] FIG. 4A shows a two-dimensional .sup.1H-.sup.15N-hNH-(FSL)-[.sup.1H,.sup.15N]-TEDOR spectrum of a .sup.13C,.sup.15N isotopes-unifonnly labeled (GGGGCC).sub.5RNA gel with a total evolution time of 200 .Math.s (n=2);

[0026] FIG. 4B shows a two-dimensional .sup.1H-.sup.15N-hNH-(FSL)-[.sup.1H,.sup.15N]-TEDOR spectrum of a .sup.13C,.sup.15N isotopes-uniformly labeled (GGGGCC).sub.5RNA gel with a total evolution time of 600 .Math.s (n=6);

[0027] FIG. 4C shows a schematic diagram of a G-G paired structure;

[0028] FIG. 5A shows a two-dimensional .sup.1H-.sup.13C-hNH-(FSL)-[.sup.1H,.sup.13C]-TEDOR spectrum of a mixed-labeled (GGGGCC).sub.5RNA gel with a total evolution time of 600 .Math.s (n=6); and

[0029] FIG. 5B shows a schematic diagram of a G-G paired structure.

DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] The present disclosure is further described below with reference to examples, which are not intended to limit the present disclosure. Various changes and modifications can be made to the present disclosure by those skilled in the art. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present disclosure should be included within the protection scope of the claims of the present disclosure.

[0031] In order to realize the detection of intramolecular or intermolecular hydrogen bonds, the present disclosure provides a series of ssNMR methods for detecting N—H...N and N—H...O═C hydrogen bonds under a fast magic angle spinning (MAS). That is, two-dimensional (2D) proton detection .sup.1H-X(X=.sup.15N or .sup.13C) transferred-echo double resonance (TEOOR), which is also known as .sup.1H-.sup.15NTEDOR and .sup.1H-.sup.13C TEDOR, as well as TEDOR-related experiments incorporating a selective pulse strategy (namely .sup.1H-.sup.15N-hNH-(FSL)-[.sup.1H,.sup.15N]-TEDOR and .sup.1H-.sup.13C-hNH-(FSL)-[.sup.1H,.sup.13C]-TEDOR) are used.

[0032] A key design of the TEDOR experiment includes: through two rotational echo double resonance (REDOR) modules, the transfer of .sup.1H to .sup.15N or .sup.13C and the transfer of .sup.13N or .sup.13C to .sup.1H are achieved to establish a .sup.1H-.sup.15N or .sup.1H-.sup.13C two-dimensional chemical shift-correlated spectrum, so as to obtain correlation information between the H atoms and atoms on both sides of the hydrogen bond, thereby directly observing the hydrogen bond. REDOR is a pulse sequence of heteronuclear dipole recoupling under MAS conditions; under the MAS conditions, the dipole coupling effect of phase-coupled spin pairs is averaged by the MAS. In the evolution period of a detection nucleus (such as .sup.1H), a π pulse is applied to a channel of the coupling nucleus (X, X=.sup.15N or .sup.13C) every half of rotation period, such that the heteronuclear dipole coupling effect is not averaged by the MAS, and the single quantum coherence of .sup.1H (.sup.1Hx) evolves into a .sup.1H-X multi-quantum coherence (.sup.1HyXz). The optimal evolution time is related to a distance between the .sup.1H and X atoms. By adjusting the total evolution time, a hydrogen atom and the X atoms on both sides of the hydrogen atom (X=.sup.15N or .sup.13C) can be detected separately. After recording the chemical shift of X, .sup.1HyXz is transformed into .sup.1Hx by a REDOR module, and direct dimension sampling is conducted on .sup.1H.

[0033] In addition to 2D.sup.1H-X TEDOR, in the present disclosure, a selective pulse is combined with the TEDOR for the first time. Selective pulses can filter out specific signals through spectral editing. In the present disclosure, a spectral editing method includes a cross-polarization (CP) experiment selected based on a contact time in a CP process, and a pulse sequence selected based on the frequency.

[0034] By combining the CP pulse, frequency selective pulses (FSLs) and TEDOR block, .sup.1H-.sup.15N-hNH-(FSL)-[.sup.1H,.sup.15N]-TEDOR and .sup.1H-.sup.13C-hNH(FSL)-[.sup.1H,.sup.13C]-TEDOR pulses. The magnetic resonance signals are transferred by .sup.1H-.sup.15N CP in an hNH pulse; the frequency of .sup.15N is selected, and signals of a specific frequency are only retained, so as to select the .sup.15N signals with a specific chemical shift; furthermore, the .sup.15N signals with a specific chemical shift are transmitted to chemically-bonded .sup.1H signals through the .sup.1H-.sup.15N CP of short contact time. Therefore, only imino or amino signals of specific chemical shifts are retained by an hNH(FSL) pulsing module. Finally, the .sup.1R-.sup.15N or .sup.1H-.sup.13C two-dimensional spectrum is constructed by .sup.1H-X TEDOR to realize specific detection of the N—H ... N and N—H...O═C hydrogen bonds. Compared with .sup.1H-X TEDOR, these two pulses can realize the specific detection of hydrogen bonds involving certain imino and amino groups (such as the imino groups of guanine and uridine, and the amino groups of guanine, cytosine, and adenine), which can effectively avoid interference of other signals to further improve an accuracy of hydrogen bond detection.

[0035] In addition, in the present disclosure, by using the fast MAS condition (40 kHz), a transverse relaxation time of .sup.1H is significantly longer than that of medium- and low-speed MAS condition, which effectively reduces the signal loss during .sup.1H-X evolution, thereby improving a sensitivity of the experiment.

[0036] The present disclosure is achieved by the following technical solutions:

[0037] A ssNMR method for detecting a hydrogen bond structure includes the following steps:

[0038] (1) A sample to be tested is prepared, including a .sup.15N or .sup.13C label.

[0039] (2) A pulse sequence is set, including the following steps:

[0040] For the 2D.sup.1H-X TEDOR pulse (X=.sup.15N or .sup.13C), the following operations are conducted: a .sup.1H nucleus is excited with a π/2 pulse, and then a π pulse is applied to an X nucleus every half rotation period for a certain number of π pulses; the π pulse is applied in a .sup.1H channel to suppress the .sup.1H chemical shift anisotropy and the .sup.1H chemical shift evolution of .sup.1H during the .sup.1H-X evolution: the π pulse is repeatedly applied on the X channel every half rotation period, with the number equal to that of the π pulse on the X channel in front of the .sup.1H π pulse; after the above step, the .sup.1H channel is locked to suppress the solvent signal; a 90° pulse is applied on the .sup.1H and X atoms, respectively, and a chemical shift (t1) is recorded in an X dimension, and a 90° pulse is applied on the .sup.1H and X atoms, respectively, after the recording is completed. After completing the above operations for the two groups of REDOR pulse sequences, the .sup.1H signal is collected, and the .sup.1H-X two-dimensional spectrum is constructed.

[0041] For the .sup.1H-.sup.15N-hNH-(FSL)-[.sup.1H,.sup.15N]-TEDOR pulse and .sup.13H-.sup.13C-hNH-(FSL)-[.sup.1H,.sup.13C]-TEDOR pulse, the following operations are conducted: the .sup.1H nucleus is excited with the x/2 pulse, .sup.1H-.sup.15N CP realizes the polarization of .sup.15N nucleus with a low sensitivity, and a contact time is set to 1 ms to 10 ms; with a shape pulse, the .sup.15N off-set and a pulse length are adjusted, a .sup.15N signal with specific chemical shift is selected; the .sup.15N signal is transferred to a .sup.1H linked by a H—N chemical bond, and a contact time is set to 100 .Math.s to 1,000 .Math.s. The subsequent .sup.1H-X TEDOR module design is consistent with the other experiments described above.

[0042] (3) The sample is detected using an ssNMR spectrometer to obtain a two-dimensional spectrum of the sample.

[0043] (4) According to the two-dimensional spectrum of the sample, signals in the spectra at different mixing times are compared to determine whether there is a NH...N or N—H...O═C hydrogen bond.

[0044] The sample is a biological macromolecule or other substances containing hydrogen bonds, where the biological macromolecule includes DNAs, RNAs, and nucleic acid-protein complexes; and the other substances containing hydrogen bonds from N—H...N or N—H...O═C include, but are not limited to, N-containing soft substances, biological materials, and polymer materials.

[0045] The sample includes a solid sample, a gel sample or a sedimentation soluble sample.

[0046] Preparation methods of a RNA sample include in vitro transcription, solid-phase synthesis, or biological extraction.

[0047] When preparing the RNA sample, the RNA sample is purified by polyacrylamide gel electrophoresis.

[0048] A water signal is suppressed by pulsing on .sup.1H to suppress the solvent signals.

[0049] In .sup.1H-.sup.15N-hNH(FSL)-[.sup.1H,.sup.15N]-TEDOR and .sup.1H-.sup.13C-hNH-(FSL)-[.sup.1H,.sup.13C]-TEDOR pulses, the shape pulses used is selected from the group consisting of Q3, GAUSSI, SQUARE, SINC, RSONB, REBurp, CRP, and REBURP.

Example 1

[0050] In this example, the pulse sequence was tested using an RNA gel formed by (GGGGCC).sub.5 sequence as a model. The GGGGCC sequence was part of a C9ORF72 gene, with repeat expansion associated with various neurodegenerative diseases. Studies had shown that repeat-expanded GGGGCC RNA might form RNA gels under specific conditions, and its structure had not been determined. The RNA gel formed by (GGGGCC).sub.5sequence was analyzed for the first time using ssNMR methods.

1. Preparation of RNA Samples

(1) Construction of DNA Template

[0051] In this example, (GGGGCC).sub.5RNA was obtained by an in vitro transcription method, and a corresponding DNA template was designed according to the target RNA. The DNA template included two complementary DNA single strands, namely a forward strand (R strand) and a reverse strand (F strand), and sequences were as follows:

TABLE-US-00001  R strand: 5′-GGCCCCGGCCCCGGCCCCGGCCCCGGCCCCTATAGTGAGTCGTATTAA-3′

TABLE-US-00002  F strand: 5′-TTAATACGACTCACTATAGGGGCCGGGGCCGGGGCCGGGGCCGGGGCC-3′

[0052] The single-stranded DNA was centrifuged, dissolved, and annealed by heating to form a linear double-stranded DNA template (with a final concentration of 25 .Math.M).

(2) RNA Samples Preparation by in Vitro Transcription

[0053] In the present disclosure, two RNA samples were prepared, namely .sup.13C,.sup.15N isotopes-uniformly labeled (GGGGCC).sub.5RNA and mixed labeled (GGGGCC).sub.5RNA prepared by adding .sup.13C-labeled GTP and .sup.15N-labeled GTP separately.

[0054] .sup.13C,.sup.15N isotopes-uniformly labeled (GGGGCC).sub.5RNA samples were prepared by 5 mL of an in vitro transcription system: 2,600 .Math.L of DEPC-treated water, 665 .Math.L of 100 mM.sup.13C.sup.15N isotope-labeled GTP, 335 pC. of 100 mM.sup.13C.sup.15N isotope-labeled CTP, 500 .Math.L of Tris buffer (containing 400 mM Tris/HCl, pH 8.0, 10 mM Spermidine, 0.1% Triton X-100), 50 .Math.L of 1 M dithiothreitol (DTT), 200 .Math.L of 1 M MgCl.sub.2, 150 .Math.L of 25 .Math.M DNA template, and 500 .Math.L of 2 mg/mL T7 RNA polymerase were sequentially added to a 50 mL centrifuge tube. Incubation was conducted in a 37° C. water bath for 24 h to ensure complete reaction of the starting materials.

[0055] Mixed-labeled (GGGGCC).sub.5RNA was prepared by 5 mL of an in vitro transcription system: 2,600 .Math.L of DEPC-treated water, 330 .Math.L of 100 mM.sup.13C isotope-labeled GTP, 330 .Math.L of 100 mM .sup.15N isotope-labeled GTP, 340 .Math.L of non-isotopo-labeled CTP, 500 .Math.L of Tris buffer (containing 400 mM TrisiHCl, pH 8.0, 10 mM Spermidine, 0.1% Triton X-100), 50 .Math.L of 1 M dithiothreitol (DTT), 200 .Math.L of 1 M MgCb, 150 .Math.L of 25 uM DNA template, and 500 .Math.L of 2 mg/mL ‘17 RNA polymerase were sequentially added to a 50 mL centrifuge tube. Incubation was conducted in a 37° C. water bath for 24 h to ensure complete reaction of the starting materials. GTP in the resulting mixed-labeled (GGGGCC).sub.5RNA was either .sup.13C-labeled or .sup.15N-labeled only.

(3) RNA Purification

[0056] After the transcription reaction was completed, the RNA was purified by 12% polyacrylamide gel electrophoresis (PAGE), and a target band was obtained by cutting a gel under UV light after the electrophoresis. The band containing the target RNA was soaked in an extraction buffer (20 mM Tris-HCl, 300 mM sodium acetate, and 1 mM EDTA, pH 7.4) at 37° C. for about 12 h, and a solution was collected.

(4) Preparation of RNA Gel Samples

[0057] A (GGGGCC).sub.5RNA sample was dissolved in a solution containing 5 mM Tris (pH=7.0) and 10 mM MgCl.sub.2 by ultrafiltration, heated at 95° C. for 3 min, and then cooled down to 37° C. in a gradient manner at 3° C./min, to obtain the RNA gel samples.

2. Pulse Sequence Setup and Method for Detecting Hydrogen Bond

[0058] 2D.sup.1H-.sup.15N TEDOR, 2D.sup.1H-.sup.13C TEDOR, 2D.sup.1H-.sup.15N-hNH-(FSL)-TEDOR, and 2D.sup.1H-.sup.13C-hNH-(FSL)-[.sup.1H,.sup.13C]-TEDOR experiments each were completed at a MAS of 40 kHz, and one rotation period was 25 .Math.s, and half a rotation period was 12.5 .Math.s. In this example, π/2 had the same definition as 90°, and π had the same definition as 180°. In this example, an ssNMR instrument was used.

[0059] The 2D.sup.1H-.sup.15N TEDOR pulse sequence (FIG. 1A) was set up as follows: [0060] a.sup.1H nucleus of the RNA sample was excited with a π/2 pulse, and [0061] 2 π pulses were applied to a .sup.15N nucleus of the RNA sample every half rotation period; [0062] .sup.1H chemical shift anisotropy and .sup.1H chemical shift evolution of .sup.1H during .sup.1H-.sup.15N evolution were suppressed by applying a π pulse to the .sup.1H nucleus of the RNA sample; [0063] 2 π pulses were applied to the .sup.I5N nucleus of the RNA sample every half rotation period; [0064] a 90° pulse was separately applied on the .sup.1H and .sup.15N nuclei of the RNA sample,

[0065] The .sup.15N chemical shift was recorded in the indirect dimension

[0066] The 90° pulse was separately applied on the .sup.1H and .sup.15N nuclei of the RNA sample.

[0067] REDOR pulse sequence: a total of 2 π pulses were applied to the .sup.15N nucleus every half rotation period; the π pulse was applied on the .sup.1H nucleus to suppress .sup.1H chemical shift anisotropy and .sup.1H chemical shift evolution of .sup.1H during .sup.1H-.sup.15N evolution; a total of 2 π pulses were applied on the .sup.15N nucleus every half rotation period;

[0068] After completing the above operations for the two groups of REDOR pulse sequences, the .sup.1H signal was collected, and the .sup.1H-.sup.15N two-dimensional spectrum was constructed.

[0069] Referring to FIG. 1A, the number of π pulses was controlled by adjusting a value of n in the experiment. When n=1, a total of 8 π pulses were applied to the .sup.15N nucleus, with a total evolution time of 100 .Math.s. When n was an integer greater than 1, a total of 8n π pulses were applied to the .sup.15N nucleus, with a total evolution time of 100 n .Math.s.

[0070] In this experiment, the direct dimension (F1: .sup.1H) and the indirect dimension (F2: .sup.15N) took 958 points and 128 points in t1 evolution period and t2 acquisition period, respectively.

[0071] The π/2 pulse applied to the .sup.1H nucleus was 1.9 .Math.s, and the π pulse applied to the .sup.1H nucleus was 3.8 .Math.s, and

[0072] The π/2 pulse applied on the .sup.15N nucleus was 4.6 .Math.s, and the π pulse applied on the .sup.15N nucleus was 9.2 .Math.s.

[0073] A recycle waiting time was 2 s.

[0074] The 2D.sup.1H-.sup.13C TEDOR pulse sequence (FIG. 1B) was set up as follows: [0075] a.sup.1H nucleus of the RNA sample was excited with a π/2 pulse, and [0076] 2 π pulses were applied to a .sup.13C nucleus of the RNA sample every half rotation period; [0077] .sup.1H chemical shift anisotropy and .sup.1H chemical shift evolution of .sup.1H during .sup.1H-.sup.13C evolution were suppressed by applying a π pulse to the .sup.1H nucleus of the RNA sample; [0078] 2 π pulses were applied to the .sup.13C nucleus of the RNA sample every half rotation period; [0079] a 90° pulse was separately applied on the .sup.1H and .sup.13C nuclei of the RNA sample, [0080] a chemical shift t1 was recorded in a .sup.13C dimension, and [0081] the 90° pulse was separately applied on the .sup.1H and .sup.13C nuclei of the RNA sample.

[0082] REDOR pulse sequence: a total of 2 π pulses were applied to the .sup.13C nucleus of the RNA sample every half rotation period; the π pulse was applied on the .sup.1H nucleus of the RNA sample to suppress .sup.1H chemical shift anisotropy and .sup.1H chemical shift evolution of .sup.1H during .sup.1H-.sup.13C evolution; a total of 2 π pulses were applied on the .sup.13C nucleus every half rotation period;

[0083] After completing the above operations for the two groups of REDOR pulse sequences, the .sup.1H signal was collected, and the .sup.1H-.sup.13C two-dimensional spectrum was constructed.

[0084] Referring to FIG. 1B, the number of π pulses was controlled by adjusting a value of n in the disclosure experiment. When n=1, a total of 8 π pulses were applied to the .sup.13C nucleus, with a total evolution time of 100 .Math.s. When n was an integer greater than 1, a total of 8n π pulses were applied to the .sup.13C nucleus, with a total evolution time of 100 n .Math.s.

[0085] In this experiment, the direct dimension (F1: .sup.1H) and the indirect dimension (F2: .sup.13C) took 958 points and 128 points in t1 evolution period and t2 acquisition period, respectively.

[0086] The π/2 pulse applied to the .sup.1H nucleus was 1.9 .Math.s, and the π pulse applied to the .sup.1H nucleus was 3.8 .Math.s, and

[0087] The π/2 pulse applied on the .sup.13C nucleus was 3.5 .Math.s, and the .sub.π pulse applied on the .sup.13C nucleus was 7.0 .Math.s.

[0088] A recycle waiting time was 2 s.

[0089] The 2D.sup.1H-.sup.15N-hNH-(FSL)-[.sup.1H,.sup.15N]-TEDOR pulse sequence (FIG. 1C) was set up as follows: [0090] a.sup.1H nucleus of the RNA sample was excited with a π/2 pulse, and [0091] In a specific example, .sup.1H-.sup.15N CP was applied at a contact time of 1 ms, a .sup.15N constant field lock at 50 kHz, and a .sup.1H constant field lock at 90 kHz; [0092] a shape pulse was applied to the .sup.15N nucleus at an excitation center of 90 ppm and a excitation time of 625 .Math.s, and a .sup.15N signal at a specific chemical shift was selected; [0093] a water signal was suppressed with a MISSISIPPI pulse sequence; and [0094] .sup.15N-.sup.1H CP was applied at a contact time of 300 .Math.s, a .sup.15N constant field lock at 50 kHz, and a .sup.1H constant field lock at 90 kHz, and the .sup.15N signal was transferred to the .sup.15N-chemically bonded .sup.1H nucleus.

[0095] 2 π pulses were applied to a .sup.15N nucleus of the RNA sample every half rotation period; [0096] .sup.1H chemical shift anisotropy and .sup.1H chemical shift evolution of .sup.1H during .sup.1H-.sup.15N evolution were suppressed by applying a π pulse to the .sup.1H nucleus; [0097] 2 π pulses were applied to the .sup.15N nucleus every half rotation period; [0098] a 90° pulse was separately applied on the .sup.1H and .sup.15N nuclei, and the .sup.15N chemical shift was recorded in the indirect dimension, and [0099] the 90° pulse was separately applied on the .sup.1H and .sup.15N nuclei.

[0100] REDOR pulse sequence: a total of 2 π pulses were applied to the .sup.15N nucleus every half rotation period; the π pulse was applied on the .sup.1H nucleus to suppress .sup.1H chemical shift anisotropy and .sup.1H chemical shift evolution of .sup.1H during .sup.1H-.sup.15N evolution; a total of 2 π pulses were applied on the .sup.15N nucleus every half rotation period;

[0101] After completing the above operations for the two groups of REDOR pulse sequences, the .sup.1H signal was collected, and the .sup.1H-.sup.15N two-dimensional spectrum was constructed.

[0102] The number of π pulses was controlled by adjusting a value of n in the disclosure experiment. When n=1, a total of 8 π pulses were applied to the .sup.15N nucleus, with a total evolution time of 100 .Math.s. When n was an integer greater than 1, a total of 8n π pulses were applied to the .sup.15N nucleus, with a total evolution time of 100 n .Math.s.

[0103] In this experiment, the direct dimension (F1: .sup.1H) and the indirect dimension (F2: .sup.15N) took 958 points and 80 points in t1 evolution period and t2 acquisition period, respectively.

[0104] The π/2 pulse applied to the .sup.1H nucleus was 1.9 .Math.s, and the π pulse applied to the .sup.1H nucleus was 3.8 .Math.s, and

[0105] The π/2 pulse applied on the .sup.15N nucleus was 4.6 .Math.s, and the π pulse applied on the .sup.15N nucleus was 9.2 .Math.s.

[0106] A recycle waiting time was 2 s.

[0107] The shape pulse was Gaus1_l80i.1000.

[0108] The 2D.sup.1H-.sup.13C-hNH-(FSL)-(.sup.1H,.sup.13C]-TEDOR pulse sequence (FIG. 1D) was set up as follows: [0109] a.sup.1H nucleus of the RNA sample was excited with a π/2 pulse, and [0110] In a specific example, .sup.1H-.sup.15N CP was applied at a contact time of 1 ms, a .sup.15N constant field lock at 50 kHz, and a .sup.1H constant field lock at 90 kHz. [0111] a shape pulse was applied to the .sup.15N nucleus at an excitation center of 150 ppm and a excitation time of 625 .Math.s, and a .sup.15N signal at a specific chemical shift was selected; [0112] a water signal was suppressed with a MISSIS1PPI pulse sequence; and [0113] .sup.15N-.sup.1H CP was applied at a contact time of 300 .Math.s, a .sup.15N constant field lock at 50 kHz, and a .sup.1H constant field lock at 90 kHz, and the .sup.15N signal was transferred to the .sup.15N-chemically bonded .sup.1H nucleus.

[0114] 2 π pulses were applied to a .sup.13C nucleus of the RNA sample every half rotation period; [0115] .sup.1H chemical shift anisotropy and .sup.1H chemical shift evolution of .sup.1H during .sup.1H-.sup.13C evolution were suppressed by applying a π pulse to the .sup.1H nucleus of the RNA sample; [0116] 2 π pulses were applied to the .sup.13C nucleus of the RNA sample every half rotation period; [0117] After completion, the hydrogen was locked for 500 .Math.s. The .sup.1H constant field lock was 130 kHz, suppressing the solvent signal; [0118] a 90° pulse was separately applied on the .sup.1H and .sup.13C nuclei, [0119] The .sup.13C chemical shift was recorded in the indirect.sup.13C dimension, and [0120] the 90° pulse was separately applied on the .sup.1H and .sup.13C nuclei of the RNA sample. [0121] REDOR pulse sequence: a total of 2 π pulses were applied to the .sup.13C nucleus every half rotation period; [0122] .sup.1H chemical shift anisotropy and .sup.1H chemical shift evolution during.sup.13C evolution were suppressed by applying a π pulse to the .sup.1H nucleus; [0123] A total of 2 π pulses were applied to the .sup.13C nucleus every half rotation period; [0124] After completing the above operations for the two groups of REDOR pulse sequences, the .sup.1H signal was collected, and the .sup.1H-.sup.13C two-dimensional spectrum was constructed.

[0125] The number of π pulses was controlled by adjusting a value of n in the disclosure experiment. When n=1, a total of 8 π pulses were applied to the .sup.13C nucleus, with a total evolution time of 100 .Math.s. When n was an integer greater than l, a total of 8n π pulses were applied to the .sup.13C nucleus, with a total evolution time of 100 n .Math.s

[0126] In this experiment, the direct dimension (F1: .sup.1H) and the indirect dimension (F2: .sup.13C) took 958 points and 128 points in t1 evolution period and t2 acquisition period, respectively.

[0127] The π/2 pulse applied to the .sup.1H nucleus was 1.9 .Math.s, and the π pulse applied to the .sup.1H nucleus was 3.8 .Math.s, and

[0128] The π/2 pulse applied on the .sup.15N nucleus was 4.6 .Math.s, and the π pulse applied on the .sup.15N nucleus was 9.2 .Math.s.

[0129] The π/2 pulse applied on the .sup.13C nucleus was 3.5 .Math.s, and the π pulse applied on the .sup.13C nucleus was 7.0 .Math.s.

[0130] A recycle waiting time was 2 s.

[0131] The shape pulse was GausI_180i. 1000,

[0132] 3. Spectral collection and result analysis

[0133] In the present disclosure, two-dimensional .sup.1H-.sup.15N TEDOR spectra of .sup.13C,.sup.15N isotopes-uniformly labeled (GGGGCC).sub.5RNA gels were collected (FIG. 2A to FIG. 2C). The results were as follows: at a total evolution time of 100 .Math.s (short mixing time), only signal peaks with .sup.1H/.sup.15N chemical shifts of 13 ppm/147.6 ppm and 10.6 ppm/142.0 ppm were observed, representing the N1-H1 cross peak of a guanine base, namely the intra guanine base NH signal. When the total evolution time was 400 .Math.s, in addition to the internal cross peaks of the bases, the related information between the bases was observed, namely the signal peaks with .sup.1H/.sup.15N chemical shifts of 13 ppm/197 ppm, representing a cross peak of H1 in guanine base and the N3 in cytosine base, and proving the existence of N—H...N hydrogen bond between G-C base pair. In addition, when the total evolution time was extended to 600 .Math.s (short mixing time), under the influence of relaxation, the intra guanine base NH signal disappeared; while long-range signals (correlated signals between the guanine and the cytosine bases) could still be observed. Therefore, the NH...N hydrogen bond in RNA gel systems could be characterized using .sup.1H-.sup.15N TEDOR experiments.

[0134] In the present disclosure, two-dimensional .sup.1H-.sup.13C TEDOR spectra of .sup.13C,.sup.15N isotopes-uniformly labeled (GGGGCC).sub.5RNA gels were collected (FIG. 3A to FIG. 3B). The results were as follows: at a total evolution time of 300 .Math.s (short mixing time), only signal peaks with .sup.1H/.sup.13C chemical shifts of 13 ppm/156 ppm and 13 ppm/161 ppm were observed, representing the C2—H1 and C6-H 1 cross peaks, both being the intra guanine base CH signals. When the total evolution time was 700 .Math.s, in addition to the internal cross peaks of the bases, the related information between the bases was observed, namely the signal peaks with .sup.1H/.sup.13C chemical shifts of 13 ppm/168 ppm, representing a cross peak of H1 in guanine base and C4 in cytosine base, and proving the existence of N—H...N hydrogen bond between G-C (FIG. 3C). Therefore, the hydrogen bond in RNA gel systems could be characterized using .sup.1H-.sup.13C TEDOR experiments.

[0135] In addition, in the present disclosure, two-dimensional 2D.sup.1H-.sup.15N-hNH-(FSL)-[.sup.1H,.sup.15N]-TEDOR spectra of .sup.13C,.sup.15N isotopes-uniformly labeled (GGGGCC).sub.5RNA gels were collected (FIG. 4A to FIG. 4B). The results were as follows: at a total evolution time of 200 .Math.s (short mixing time), only signal peaks with .sup.1H/.sup.15N chemical shifts of 9 ppm/80 ppm and 8 ppm/80 ppm were observed, representing a N2-H2 cross peak, namely the intra guanine base NH.sub.2 signal. When the total evolution time was 600 .Math.s, in addition to the internal cross peaks of the bases, the related information between bases was observed, namely the signal peaks with .sup.1H/.sup.15N chemical shifts of 9 ppm/232.7 ppm, representing a cross peak of H2 in G base and N7 paired to the guanine base, and demonstrating the existence of the N—H...N hydrogen bond between the G-G Hoogesteen base pairings (FIG. 4C).

[0136] In the present disclosure, two-dimensional 2D.sup.1H-.sup.13C-hNH-(FSL)-[.sup.1H,.sup.13C]-TEDOR spectra of (GGGGCC).sub.5RNA gels were collected (FIG. 5A). The results were as follows: when the total evolution time was 600 .Math.s (long mixing time), signal peaks with .sup.1H/.sup.13C chemical shifts of 10.78 ppm/160.9 ppm were observed, representing a cross peak of C6 in guanine base and H1 paired to the guanine base, and demonstrating the existence of N—H ... O═C hydrogen bond between G-G pairings (FIG. 5B).

[0137] Compared with the prior art, the present disclosure has the following beneficial effects:

[0138] In the present disclosure, a .sup.1H-X TEDOR pulse sequence is designed by combining proton-detected technology with transfer echo double resonance pulse technology, and a ssNMR method is established for specific detection of N—H...N or N—H...O═C hydrogen bond. TEDOR achieves coherent transfer through a heteronuclear recoupling method based on dipole coupling, and establishes .sup.1H-X chemical shift correlations. In this experiment, in the .sup.1H-XTEDOR method, after the excitation of .sup.1H, π pulses are applied to the X channel every half rotation period. After several rotation cycles, dipole recoupling is achieved and the .sup.1H magnetization convert to multi-quantum .sup.1H-X coherence. The coupled .sup.1H-X coherence is converted into a detectable .sup.1H coherence through the second REDOR period, and the .sup.1H chemical shift is recorded to complete the two-dimensional experiment. In the TEDOR method, the cross-peak intensity is related to the distance between two nuclei and the total mixing time of REDOR period. Through the TEDOR pulse experiment, the chemical shift correlation can be realized between .sup.1H and the atoms on both sides in the N—H...N or N—H...O═C hydrogen bond, meanwhile, by adjusting the total time, the chemical shift correlation between the covalently bound N and H atoms, as well as the chemical shift correlation between X and H atoms connected by hydrogen bond can be identified.

[0139] In addition, in the present disclosure, the selective pulses, including the CP pulse and the frequency selective pulse, are combined with the TEDOR pulse for the first time, thus effectively reducing the spectral overlapping.

[0140] In the present disclosure, the method has the following advantages: (1) a proton detected technology is adopted with a high sensitivity; (2) a .sup.1H-.sup.15N two-dimensional spectrum is constructed to observe relevant signals with a high accuracy; (3) TEDOR pulse parameter setting is simple, without need to be optimized according to the angle and the bond length of the hydrogen bond; (4) acquisition of the .sup.1H atoms is related to a chemical shift of two nitrogen atoms in the hydrogen bond, and the hydrogen bond can be directly observed at atomic level; (5) through the combination of selective pulses and TEDOR pulses, the overlapping of spectra is effectively reduced; and (6) in the N—H...N hydrogen bond, N-H connected by chemical bonds has a distance of about 1.0 Å, and H and N atoms connected by hydrogen bonds has a distance of about 1.9 Å. Therefore, with the .sup.1H-.sup.15N TEDOR experiment, by setting different mixing times, N-H groups connected by chemical bonds, as well as H and N atoms connected by hydrogen bonds, are detected for structural identification and characterization.