- Email: [email protected]
Simplifying DNA NMR spectroscopy by silencing GH8 and AH8 resonances
Accepted Manuscript Simplifying DNA NMR Spectroscopy by Silencing GH8 and AH8 Resonances
Sarah V. Nguyen, Ekaterina Stroeva, Markus W. Germann PII:
S0022-2860(18)30491-5
DOI:
10.1016/j.molstruc.2018.04.049
Reference:
MOLSTR 25122
To appear in:
Journal of Molecular Structure
Received Date:
26 January 2018
Revised Date:
11 April 2018
Accepted Date:
12 April 2018
Please cite this article as: Sarah V. Nguyen, Ekaterina Stroeva, Markus W. Germann, Simplifying DNA NMR Spectroscopy by Silencing GH8 and AH8 Resonances, Journal of Molecular Structure (2018), doi: 10.1016/j.molstruc.2018.04.049
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Simplifying DNA NMR Spectroscopy by Silencing GH8 and AH8 Resonances Sarah V. Nguyen‡, Ekaterina Stroeva‡, Markus W. Germann* ‡Department
of Chemistry, Georgia State University, Atlanta, Georgia, 30303 *Departments of Chemistry and Biology, Natural Sciences Center, 412, Georgia State University, Atlanta, Georgia, 30303. [email protected]. Tel. 001-404 413 5561 Keywords: DNA structure, NMR, deuteration
Abstract
NMR studies of DNA oligonucleotides are limited by extensive resonance overlap and the availability of isotopically labeled samples. We have explored a nondestructive and economical method to simplify DNA NMR spectroscopy by exchanging the hydrogens of GH8 and AH8 with deuterons. Exchanging these hydrogens with deuterons results in spectral simplification and alleviates spectral overlap particularly for A and G rich sequences and is therefore uniquely suited for G-quadruplex sequences. The silencing of the GH8 and AH8 signals facilitates structural and ligand DNA binding studies.
1. Introduction
NMR spectroscopy is used for the structural characterization of a large range of samples, including biologically relevant macromolecules. Large molecules, however, present a challenge due to large number of resonances, resulting in extensive peak overlap within a limited chemical shift window. Labeling strategies and NMR experiments have therefore been developed to combat these challenges. RNA is generally labeled by in vitro transcription reactions using isotopically labeled triphosphates [1-4]. Isotopic labeling of DNA is less common; for a recent comprehensive review see Nelissen et al [5]. Although enzymatic methods for DNA labeling have been described, they have not yet been widely used [6-9]. Most DNA labeling is instead carried out using solid phase synthesis with commercially available phosphoramidites [1,5]. The high cost of the isotopically labeled phosphoramidites in conjunction with the large excess of reactants needed for solid phase conditions, have curtailed their use.
ACCEPTED MANUSCRIPT
An alternative to simplify nucleic acid NMR is to selectively remove “unwanted” resonances by deuteration. This approach has been used for structural and dynamic studies of RNA, DNA and proteins [10-16]. Guo et al. developed an elegant chemical approach for deuteration of the H8 and H6 position in nucleosides and nucleotides [17]. Deuterated rNTPs can then be used to enzymatically produce oligoribonucleotides for structural studies [4,5,18]. In DNA applications, typically the sugar residues are deuterated while reports using deuterated bases are less common [4-5,9,17-18]. In either case, deuterated phosphoramidites are required for solid phases synthesis, which are not readily commercially available. Additionally, local synthesis resources are also required to produce oligonucleotides in large quantity for structural studies. Deuteration can also be accomplished by exposing DNA oligonucleotides to D2O solutions. It has long been recognized that certain base protons lose intensity through exchange with a deuteron upon extended storage in D2O solutions [17,19,20, 21]. This is generally troublesome as NOE intensities must be adjusted to account for this effect, especially for solvent exposed residues such as terminal or mismatched bases. Isotopic exchange of the H8 hydrogens (Figure 1) in purines has been explored previously by tritium exchange experiment [22]. This effect was also employed to characterize difference in the exchange kinetics of different DNA conformations and sequences [15-16]. We have taken advantage of this behavior to develop a robust, nondestructive, and economical method to simplify DNA NMR spectroscopy by deuterating H8 of adenine and guanine [Fig. 1A]. Our method allows for effective simplification of DNA NMR spectra, especially for sequences rich in adenine and guanine using common materials and inexpensive, commercial synthesized oligodeoxyribonucleotides. This approach is uniquely suited to study ligand binding and loop structures/dynamics of G-quadruplex by NMR. Here we present results of our method for both a self-complementary decamer duplex containing an EcoRI recognition sequence and a G-quadruplex (Pu-22, modified c-MYC oncogene sequence) [Fig. 1B,C].
2. Results and Discussion As a first step, the deuterium exchange properties of adenosine and guanosine monophosphates were determined to establish the timeframe for exchange over a range of pH conditions. The temperature
ACCEPTED MANUSCRIPT
was kept at 65°C to accelerate exchange while minimizing potential degradation. Analysis and comparison of guanine and adenine exchange rates are shown in Figure 2. Intensities of GMP and AMP H8 protons were measured relative to the non-exchanging AH2 control resonance. As anticipated, GMP H8 exchanges more rapidly than AMP H8. At pH 10.0 the half-life of GMP H8 is just a few hours [Fig 1]. For AMP H8, however, the exchange is essentially pH independent in the range tested in accordance to published reports [22]. This exchange is significantly slower, requiring more than 10 days to approach completion (>99% exchange). The GMP H8 NMR signals can therefore in principle be selectively silenced. As previously established by Benevides et al., the exchange process is slower for helical structures than for GMP and AMP [23-24]. The exchange time was therefore extended to 4 weeks for the EcoRI 10 mer duplex and Pu-22 quadruplex.
Figure 1. Adenosine and guanosine monophosphate 1H NMR peak intensities as a function of time and pH (filled circle – 7.4, open circle – 8.9, diamond – 10.4). Inset: half lifetime of GH8 and AH8 hydrogens. The commercially synthesized EcoRI oligodeoxyribonucleotide (Fig. 2C) was used as provided, without any additional purification, and exchanged for 4 weeks at 65 °C as described. Under these conditions, DNA secondary structures are destabilized and exchange is facilitated. Following exchange, it is important to neutralize the sample prior to desalting and lyophilize the sample immediately following desalting to minimize back exchange of GD8. If required, samples could be purified prior to the exchange
ACCEPTED MANUSCRIPT
reactions. A comparison of the exchanged to the original sample by denaturing gel electrophoresis showed no degradation (data not shown). Following reconstitution in the appropriate buffer, the sample was examined by NMR spectroscopy under native conditions. The aromatic region of the EcoRI duplex clearly shows a remarkable suppression of all H8 resonances while other base protons are not affected (Figure 3A). The fact that both exchanged and control DNA had the same spectra of the remaining non-exchangeable resonances indicates that the structure was not affected. This is further supported by the comparison of the 31P NMR spectrum of the exchanged and control sample which addresses the integrity of the DNA backbone
[Fig. 3C]. To assess the base pairing in these DNA duplexes, the samples were lyophilized and the dissolved in 90 % H2O/10 % D2O. The imino proton spectra of two samples were essentially identical as anticipated [Fig. 3B]. Application of the H/D exchange procedure for 4 weeks to Pu-22 also resulted in a remarkable simplification of the aromatic region [Fig. 3D]. As mentioned previously, the exchange was carried out at 65 °C and in absence
of
structure
stabilizing
counter ions (such as K+ or Na+). Following exchange, the sample was reconstituted in the appropriate buffer Figure 2. (A) Protons at position 8 of adenine and guanine are exchangeable with deuterium. (B) Hoogsteen base pairing in Gquadruplexes.
(C)
Sequences
of
oligonucleotides used in this study.
EcoRI
and
Pu-22
without any further purification. The spectral
simplification
was
more
dramatic than the for EcoRI duplex because exchange not only yielded fewer peaks, but also showed less
resonance overlap. Specifically, the H8 resonances of the guanine tetrads no longer interfere with resonances of important loop and flanking sequences which facilitates a more detailed and transparent analysis of these important residues.
The backbone of Pu-22, assessed from
31P
NMR, was left
uncompromised by the procedure [Fig. 3F]. Interestingly, the imino proton spectrum of the exchanged sample showed a noteworthy behavior [Fig. 3E]. While all imino proton chemical shifts matched the control, four of them, indicated with a star, have a reduced intensity. These correspond to the guanines of the center tetrad of the folded quadruplex. In the folded state of Pu-22, these imino protons are shielded by the top and bottom tetrad of the quadruplex. Therefore, when the fully exchanged sample was reconstituted
ACCEPTED MANUSCRIPT
in 90 % H2O/10 % D2O and measured immediately following sample preparation, these imino protons are not yet fully recovered. If the sample is measured a few hours later, all the imino proton intensities match that of the control.
Figure 3. Aromatic (A), imino proton- (B), and
31P
spectra (C) of 0.6 mM EcoRI measured at 298K.
Aromatic (D), imino proton (E), and 31P spectra (F) of 0.6 mM Pu-22 measured at 298K. Lower spectra represent control samples while upper spectra represent exchanged samples. Reduced peaks in the exchanged Pu-22 imino proton spectrum correspond to resonances from the middle tetrad, indicated by stars.
The reduction of the spectral complexity is also evident in 2D NMR spectra as shown for the base– H1’ region of the EcoRI NOESY spectrum. Clearly, all cross peaks involving H8 protons are suppressed which significantly simplifies the base to H1’ sugar NOE windows [Fig. 4A,B]. This also aids in investigating DNA drug interactions as it is much easier to follow drug aromatic signals if overlapping
ACCEPTED MANUSCRIPT
DNA signals are eliminated as. This is demonstrated for the binding of DAPI to control and exchanged EcoRI duplex DNA. Note, for example, the appearance of a DAPI-EcoRI complex peak at 8 ppm in both control and exchanged samples which can be much more easily followed in the exchanged sample [Fig. 4C,D].
Figure 4. NOESY base to H1’ section of control (A) and exchanged EcoRI -10mer (B) at 25 °C. Positions of the H8 and H1’ resonances of A and G residues are marked. Titration of control (C) and exchanged EcoRI-10mer (D) with DAPI at 25 °C and pH* 7.4. Ratios shown represent DNA:DAPI. Arrows indicate new resonances of the formed complex. 3. Materials and Methods Oligodeoxyribonucleotides were purchased from Integrated DNA Technologies (San Hose, CA). DNA concentrations were measured by UV spectroscopy using the extinction coefficient derived from the
ACCEPTED MANUSCRIPT
sum of mononucleotides at 260 nm absorbance at 85 oC (ε260= 93,600 and ε260 = 228,700 M-1cm-1, for EcoRI and Pu-22 respectively). 4',6-diamidino-2-phenylindole (DAPI) was purchased from Boehringer Mannheim Biochemical (Indianapolis, IN) and used without further purification. Stock concentrations were calculated using ε340 = 27,000 M-1cm-1. Adenine monophosphate and guanine monophosphate were purchased from Sigma-Aldrich (St. Illinois, MO) and used without further purification. 3.1 Exchange rate determination for monophosphates: Adenine monophosphate and guanine
monophosphate were dissolved in 50 mM Tris-HCl, 0.1 mM 4,4-dimethyl-4-silapentane-1sulfonic acid (DSS), and 99.8 % D2O to make 10 mM solutions. Sodium deuteroxide and deuterium chloride were used to adjust pH* (meter reading) to 7.4, 8.9, and 10.4. 1H NMR spectra were collected immediately before the samples were placed in a 65 oC water bath. 3.2 Deuterium exchange for oligodeoxyribonucleotides: To initiate proton-deuteron exchange, lyophilized DNA was dissolved in a solution of 50 mM Tris at pH* 9.0 in 99.96 % D2O. Tris base in 99.96 % D2O was used to adjust the pH* (meter reading) of exchanging samples to avoid introducing structure stabilizing cations. The solution was then allowed to undergo exchange in a 65 oC water bath for up to one month, with periodic monitoring of the H8 peak intensity via 1H NMR. The temperature and pH were chosen to accelerate exchange by destabilizing secondary structures in oligonucleotides while minimizing potential degradation. Upon completion of exchange, the samples were neutralized with DCl to pH* 7.3 and desalted with a HiTrap desalting column. Since desalting was carried out in H2O, the sample was lyophilized immediately following desalting to minimize back exchange of GD8/AD8 to GH8/AH8 in H2O solutions. 3.3 Binding studies: Both exchanged and non-exchanged samples were prepared in 10 mM sodium phosphate, 20 mM sodium chloride, 0.1 mM 4,4-dimethyl-4-silapentane 1-sulfonic acid (DSS) and 0.3 mM ethylenediaminetetraacetic acid (EDTA), at a pH* 7.3, in 99.96 % D2O. The same buffer in 90 %/10 % H2O/D2O was used to observe imino protons. Titration experiments consisted of adding concentrated volumes of DAPI solution in 99.96 % D2O to 0.3 mM DNA samples at the following DNA to DAPI ratios: 1:0, 1:0.25, 1:0.5, 1:0.75, 1:1. The complex formation was then monitored by NMR. 3.4 NMR spectroscopy: NMR spectra were obtained using Bruker Avance 600 and 500 MHz NMR systems equipped with 5-mm QXI and 5-mm TBI probes, respectively. All NMR experiments were carried out at 25 oC and referenced to internal DSS. Imino proton spectra were recorded with a 1-1 solvent suppression
ACCEPTED MANUSCRIPT
sequence19. NOESY spectra were acquired at 600 MHz using a mixing time of 400 ms. 31P NMR spectra were proton decoupled and referenced to 85% H3PO4 in a capillary.
4. Conclusion NMR studies of DNA are often hampered by resonance overlap and the expense and limited availability of isotopically labeled oligonucleotides. We have presented a robust, nondestructive and economical method to simplify DNA NMR spectroscopy by exchanging the hydrogens of GH8 and AH8 with deuterons. This procedure, using only resources that are easily available in any laboratory, provides an accessible means of reducing NMR resonance overlap without perturbing the native DNA structure as evidenced by NMR. The deletion of proton signals from guanine tetrads facilitates investigations of other structural aspects, namely the surrounding loops and flanking sequences which determine many of the G tetrad unique properties. With a significant number of resonances effectively silenced, drug binding studies are more straightforward. Aknowledgements We thank Drs. Wilson and Spring-Connell for suggestions. Part of this work was supported by the Georgia Cancer Coalition and NIH (GM55404-01A1). References 1. Nelissen, F. H., Gammeren, A. J., Tessari, M., Girard, F. C., Heus, H. A., & Wijmenga, S. S. Multiple segmental and selective isotope labeling of large RNA for NMR structural studies. Nucleic Acids Research 2008, 36(14). 2. Nelissen, F. H., Leunissen, E.H., Laar, L., Tessari, M., Heus, H.A., Wimenga, S. S. Fast production of homogeneous recombinant RNA – towards large-scale production of RNA. Nucleic Acids Research 2012, 40(13). 3. Nelissen, F. H., Girard, F. C., Tessari, M., Heus, H. A., Wijmenga, S. S. Preparation of selective and segmentally labeled single-stranded DNA for NMR by self-primed PCR and asymmetrical endonuclease double digestion. Nucleic Acids Research 2009, 37(17). 4. Cromsigt, J., Schleucher, J., Gustafsson, T., Kihlberg, J., Wijmenga, S. Preparation of partially 2H/13C-labelled RNA for NMR studies. Stereo-specific deuteration of the H5’’ in nucleotides. Nucleic Acids Research 2002, 30(7), 1639-1645. 5. Nelissen, F. H., Tessari, M., Wijmenga, S. S., Heus, H. A. Stable isotope labeling methods for DNA. Progression Nuclear Magnetic Resonance Spectroscopy 2016, 96, 89-103. 6. Khan, A. M., Mishra, S. H., & Germann, M. W. Cyclic enzymatic solid phase synthesis of isotopically labeled DNA oligonucleotides. Nucleosides, Nucleotides and Nucleic Acids 2009, 28(11-12), 1030-1041.
ACCEPTED MANUSCRIPT
7. Zimmer, D.P.; Crothers, D.M. NMR of enzymatically synthesized uniformly 13C, 15N-labeled DNA oligonucleotides. Proc. Natl. Acad. Sci. 1995, 92, 3091–3095. 8. Masse, J.E.; Bortmann, P.; Dieckmann, T.; Feigon, J. Simple, efficient protocol for enzymatic synthesis of uniformly 13C, 15N-labeled DNA for heteronuclear NMR studies. Nucleic Acids Res. 1998, 26, 2618–2620. 9. Macdonald, D., Lu, P. Determination of DNA structure in solution: enzymatic deuteration of ribose 2’ carbon. J. AM. CHEM. SOC. 2002, 124(33), 9722-9723. 10. Englander, S. W., Sosnick, T. R., Englander, J. J., & Mayne, L. (1996). Mechanisms and uses of hydrogen exchange. Current Opinion in Structural Biology 1996, 6(1), 18-23. 11. Printz, M. P., & Hippel, P. H. Hydrogen exchange studies of DNA structure. Proceedings of the National Academy of Sciences 1965, 53(2), 363-370. 12. Maltseva, T. V., Földesi, A., & Chattopadhyaya, J. T1 and T2 relaxations of the 13C nuclei of deuterium-labeled nucleosides. Magnetic Resonance in Chemistry 1998, 36(4), 227-239. 13. Dupureur, C. M., & Barton, J. K. Use of Selective deuteration and 1H NMR in demonstrating major groove binding of DELTA.-[Ru(phen)2dppz]2 to d(GTCGAC)2. Journal of the American Chemical Society 1994, 116(22), 10286-10287. 14. Lukin, M., & Santos, C. D. Stereoselective nucleoside deuteration for NMR studies of DNA. Nucleosides, Nucleotides and Nucleic Acids 2010, 29(7), 562-573. 15. Brandes, R., Ehrenberg, A. Kinetics of the proton-deuteron enchange at position H8 of adenine and guanine in DNA. Nucleic Acids Research 1986, 14(23), 9491-9508. 16. Reilly, K. E., Thomas, G. J. Hygrogen exchange dynamics of the P22 virion determined by timeresolved raman spectroscopy. J. Mol. Biol. 1994, 241, 68-82. 17. Huang, X. An efficient and economic site-specific deuteration strategy for NMR studies of homologous oligonucleotide repeat sequences. Nucleic Acids Research 1997, 25(23), 4758-4763. 18. Keane, S. C., Van, V., Frank, H. M., Sciandra, C. A., Mccowin, S., Santos, J., Heng, X., & Summers, M. F. NMR detection of intermolecular interaction sites in the dimeric 5′-leader of the HIV-1 genome. Proceedings of the National Academy of Sciences 2016, 113(46), 13033-13038. 19. Chirakul, P., Litzer, J., & Sigurdsson, S. Preparation of base-deuterated 2’-deoxyadenosine nucleosides and their site-specific incorporation into DNA. Nucleosides, Nucleotides & Nucleic Acids 2001, 20(12), 1903-1913. 20. Meyer, B., & Peters, T. NMR Spectroscopy techniques for screening and identifying ligand binding to protein receptors. Angew Chem Int Ed Engl. 2003, 42(8):864-90. 21. Yan, X., & Maier, C. S. Hydrogen/deuterium exchange mass spectrometry. Mass spectrometry of proteins and peptides: methods and protocols 2009, 15, 255-271. 22. Tomasz, M., Olson, J., Mercado, C. M. Mechanism of the isotopic exchange of the C-8 hydrogen of purines in nucleosides and in deoxyribonucleic acid. Biochemistry 1972, 11(7), 1235-1241. 23. Benevides, J.M., Lemeur, D., and Thomas, G.J. Molecular conformations and 8-CH exchange rates of purine ribo- and deoxyribonucleotides: investigation by raman spectroscopy. Biopolymers 1984, 23, 1011-1023; 24. Benevides, J.M., and Thomas, G.J. Dependence of purine 8C-H exchange on nucleic acid conformation and base-pairing geometry: a dynamic probe of DNA and RNA secondary structures. Biopolymers 1985, 24,667-682
ACCEPTED MANUSCRIPT
-
NMR studies of DNA oligonucleotides are limited by extensive resonance overlap and the availability of isotopically labeled samples Exchanging GH8 and AH8 hydrogens with deuterium greatly simplifies NMR spectra, particularly of quadruplex DNA. The method only uses regular/commercial synthesized oligodeoxyribonucleotides and common chemicals. This approach facilitates studies of loop residues, dynamics and drug binding.
Sarah V. Nguyen, Ekaterina Stroeva, Markus W. Germann PII:
S0022-2860(18)30491-5
DOI:
10.1016/j.molstruc.2018.04.049
Reference:
MOLSTR 25122
To appear in:
Journal of Molecular Structure
Received Date:
26 January 2018
Revised Date:
11 April 2018
Accepted Date:
12 April 2018
Please cite this article as: Sarah V. Nguyen, Ekaterina Stroeva, Markus W. Germann, Simplifying DNA NMR Spectroscopy by Silencing GH8 and AH8 Resonances, Journal of Molecular Structure (2018), doi: 10.1016/j.molstruc.2018.04.049
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Simplifying DNA NMR Spectroscopy by Silencing GH8 and AH8 Resonances Sarah V. Nguyen‡, Ekaterina Stroeva‡, Markus W. Germann* ‡Department
of Chemistry, Georgia State University, Atlanta, Georgia, 30303 *Departments of Chemistry and Biology, Natural Sciences Center, 412, Georgia State University, Atlanta, Georgia, 30303. [email protected]. Tel. 001-404 413 5561 Keywords: DNA structure, NMR, deuteration
Abstract
NMR studies of DNA oligonucleotides are limited by extensive resonance overlap and the availability of isotopically labeled samples. We have explored a nondestructive and economical method to simplify DNA NMR spectroscopy by exchanging the hydrogens of GH8 and AH8 with deuterons. Exchanging these hydrogens with deuterons results in spectral simplification and alleviates spectral overlap particularly for A and G rich sequences and is therefore uniquely suited for G-quadruplex sequences. The silencing of the GH8 and AH8 signals facilitates structural and ligand DNA binding studies.
1. Introduction
NMR spectroscopy is used for the structural characterization of a large range of samples, including biologically relevant macromolecules. Large molecules, however, present a challenge due to large number of resonances, resulting in extensive peak overlap within a limited chemical shift window. Labeling strategies and NMR experiments have therefore been developed to combat these challenges. RNA is generally labeled by in vitro transcription reactions using isotopically labeled triphosphates [1-4]. Isotopic labeling of DNA is less common; for a recent comprehensive review see Nelissen et al [5]. Although enzymatic methods for DNA labeling have been described, they have not yet been widely used [6-9]. Most DNA labeling is instead carried out using solid phase synthesis with commercially available phosphoramidites [1,5]. The high cost of the isotopically labeled phosphoramidites in conjunction with the large excess of reactants needed for solid phase conditions, have curtailed their use.
ACCEPTED MANUSCRIPT
An alternative to simplify nucleic acid NMR is to selectively remove “unwanted” resonances by deuteration. This approach has been used for structural and dynamic studies of RNA, DNA and proteins [10-16]. Guo et al. developed an elegant chemical approach for deuteration of the H8 and H6 position in nucleosides and nucleotides [17]. Deuterated rNTPs can then be used to enzymatically produce oligoribonucleotides for structural studies [4,5,18]. In DNA applications, typically the sugar residues are deuterated while reports using deuterated bases are less common [4-5,9,17-18]. In either case, deuterated phosphoramidites are required for solid phases synthesis, which are not readily commercially available. Additionally, local synthesis resources are also required to produce oligonucleotides in large quantity for structural studies. Deuteration can also be accomplished by exposing DNA oligonucleotides to D2O solutions. It has long been recognized that certain base protons lose intensity through exchange with a deuteron upon extended storage in D2O solutions [17,19,20, 21]. This is generally troublesome as NOE intensities must be adjusted to account for this effect, especially for solvent exposed residues such as terminal or mismatched bases. Isotopic exchange of the H8 hydrogens (Figure 1) in purines has been explored previously by tritium exchange experiment [22]. This effect was also employed to characterize difference in the exchange kinetics of different DNA conformations and sequences [15-16]. We have taken advantage of this behavior to develop a robust, nondestructive, and economical method to simplify DNA NMR spectroscopy by deuterating H8 of adenine and guanine [Fig. 1A]. Our method allows for effective simplification of DNA NMR spectra, especially for sequences rich in adenine and guanine using common materials and inexpensive, commercial synthesized oligodeoxyribonucleotides. This approach is uniquely suited to study ligand binding and loop structures/dynamics of G-quadruplex by NMR. Here we present results of our method for both a self-complementary decamer duplex containing an EcoRI recognition sequence and a G-quadruplex (Pu-22, modified c-MYC oncogene sequence) [Fig. 1B,C].
2. Results and Discussion As a first step, the deuterium exchange properties of adenosine and guanosine monophosphates were determined to establish the timeframe for exchange over a range of pH conditions. The temperature
ACCEPTED MANUSCRIPT
was kept at 65°C to accelerate exchange while minimizing potential degradation. Analysis and comparison of guanine and adenine exchange rates are shown in Figure 2. Intensities of GMP and AMP H8 protons were measured relative to the non-exchanging AH2 control resonance. As anticipated, GMP H8 exchanges more rapidly than AMP H8. At pH 10.0 the half-life of GMP H8 is just a few hours [Fig 1]. For AMP H8, however, the exchange is essentially pH independent in the range tested in accordance to published reports [22]. This exchange is significantly slower, requiring more than 10 days to approach completion (>99% exchange). The GMP H8 NMR signals can therefore in principle be selectively silenced. As previously established by Benevides et al., the exchange process is slower for helical structures than for GMP and AMP [23-24]. The exchange time was therefore extended to 4 weeks for the EcoRI 10 mer duplex and Pu-22 quadruplex.
Figure 1. Adenosine and guanosine monophosphate 1H NMR peak intensities as a function of time and pH (filled circle – 7.4, open circle – 8.9, diamond – 10.4). Inset: half lifetime of GH8 and AH8 hydrogens. The commercially synthesized EcoRI oligodeoxyribonucleotide (Fig. 2C) was used as provided, without any additional purification, and exchanged for 4 weeks at 65 °C as described. Under these conditions, DNA secondary structures are destabilized and exchange is facilitated. Following exchange, it is important to neutralize the sample prior to desalting and lyophilize the sample immediately following desalting to minimize back exchange of GD8. If required, samples could be purified prior to the exchange
ACCEPTED MANUSCRIPT
reactions. A comparison of the exchanged to the original sample by denaturing gel electrophoresis showed no degradation (data not shown). Following reconstitution in the appropriate buffer, the sample was examined by NMR spectroscopy under native conditions. The aromatic region of the EcoRI duplex clearly shows a remarkable suppression of all H8 resonances while other base protons are not affected (Figure 3A). The fact that both exchanged and control DNA had the same spectra of the remaining non-exchangeable resonances indicates that the structure was not affected. This is further supported by the comparison of the 31P NMR spectrum of the exchanged and control sample which addresses the integrity of the DNA backbone
[Fig. 3C]. To assess the base pairing in these DNA duplexes, the samples were lyophilized and the dissolved in 90 % H2O/10 % D2O. The imino proton spectra of two samples were essentially identical as anticipated [Fig. 3B]. Application of the H/D exchange procedure for 4 weeks to Pu-22 also resulted in a remarkable simplification of the aromatic region [Fig. 3D]. As mentioned previously, the exchange was carried out at 65 °C and in absence
of
structure
stabilizing
counter ions (such as K+ or Na+). Following exchange, the sample was reconstituted in the appropriate buffer Figure 2. (A) Protons at position 8 of adenine and guanine are exchangeable with deuterium. (B) Hoogsteen base pairing in Gquadruplexes.
(C)
Sequences
of
oligonucleotides used in this study.
EcoRI
and
Pu-22
without any further purification. The spectral
simplification
was
more
dramatic than the for EcoRI duplex because exchange not only yielded fewer peaks, but also showed less
resonance overlap. Specifically, the H8 resonances of the guanine tetrads no longer interfere with resonances of important loop and flanking sequences which facilitates a more detailed and transparent analysis of these important residues.
The backbone of Pu-22, assessed from
31P
NMR, was left
uncompromised by the procedure [Fig. 3F]. Interestingly, the imino proton spectrum of the exchanged sample showed a noteworthy behavior [Fig. 3E]. While all imino proton chemical shifts matched the control, four of them, indicated with a star, have a reduced intensity. These correspond to the guanines of the center tetrad of the folded quadruplex. In the folded state of Pu-22, these imino protons are shielded by the top and bottom tetrad of the quadruplex. Therefore, when the fully exchanged sample was reconstituted
ACCEPTED MANUSCRIPT
in 90 % H2O/10 % D2O and measured immediately following sample preparation, these imino protons are not yet fully recovered. If the sample is measured a few hours later, all the imino proton intensities match that of the control.
Figure 3. Aromatic (A), imino proton- (B), and
31P
spectra (C) of 0.6 mM EcoRI measured at 298K.
Aromatic (D), imino proton (E), and 31P spectra (F) of 0.6 mM Pu-22 measured at 298K. Lower spectra represent control samples while upper spectra represent exchanged samples. Reduced peaks in the exchanged Pu-22 imino proton spectrum correspond to resonances from the middle tetrad, indicated by stars.
The reduction of the spectral complexity is also evident in 2D NMR spectra as shown for the base– H1’ region of the EcoRI NOESY spectrum. Clearly, all cross peaks involving H8 protons are suppressed which significantly simplifies the base to H1’ sugar NOE windows [Fig. 4A,B]. This also aids in investigating DNA drug interactions as it is much easier to follow drug aromatic signals if overlapping
ACCEPTED MANUSCRIPT
DNA signals are eliminated as. This is demonstrated for the binding of DAPI to control and exchanged EcoRI duplex DNA. Note, for example, the appearance of a DAPI-EcoRI complex peak at 8 ppm in both control and exchanged samples which can be much more easily followed in the exchanged sample [Fig. 4C,D].
Figure 4. NOESY base to H1’ section of control (A) and exchanged EcoRI -10mer (B) at 25 °C. Positions of the H8 and H1’ resonances of A and G residues are marked. Titration of control (C) and exchanged EcoRI-10mer (D) with DAPI at 25 °C and pH* 7.4. Ratios shown represent DNA:DAPI. Arrows indicate new resonances of the formed complex. 3. Materials and Methods Oligodeoxyribonucleotides were purchased from Integrated DNA Technologies (San Hose, CA). DNA concentrations were measured by UV spectroscopy using the extinction coefficient derived from the
ACCEPTED MANUSCRIPT
sum of mononucleotides at 260 nm absorbance at 85 oC (ε260= 93,600 and ε260 = 228,700 M-1cm-1, for EcoRI and Pu-22 respectively). 4',6-diamidino-2-phenylindole (DAPI) was purchased from Boehringer Mannheim Biochemical (Indianapolis, IN) and used without further purification. Stock concentrations were calculated using ε340 = 27,000 M-1cm-1. Adenine monophosphate and guanine monophosphate were purchased from Sigma-Aldrich (St. Illinois, MO) and used without further purification. 3.1 Exchange rate determination for monophosphates: Adenine monophosphate and guanine
monophosphate were dissolved in 50 mM Tris-HCl, 0.1 mM 4,4-dimethyl-4-silapentane-1sulfonic acid (DSS), and 99.8 % D2O to make 10 mM solutions. Sodium deuteroxide and deuterium chloride were used to adjust pH* (meter reading) to 7.4, 8.9, and 10.4. 1H NMR spectra were collected immediately before the samples were placed in a 65 oC water bath. 3.2 Deuterium exchange for oligodeoxyribonucleotides: To initiate proton-deuteron exchange, lyophilized DNA was dissolved in a solution of 50 mM Tris at pH* 9.0 in 99.96 % D2O. Tris base in 99.96 % D2O was used to adjust the pH* (meter reading) of exchanging samples to avoid introducing structure stabilizing cations. The solution was then allowed to undergo exchange in a 65 oC water bath for up to one month, with periodic monitoring of the H8 peak intensity via 1H NMR. The temperature and pH were chosen to accelerate exchange by destabilizing secondary structures in oligonucleotides while minimizing potential degradation. Upon completion of exchange, the samples were neutralized with DCl to pH* 7.3 and desalted with a HiTrap desalting column. Since desalting was carried out in H2O, the sample was lyophilized immediately following desalting to minimize back exchange of GD8/AD8 to GH8/AH8 in H2O solutions. 3.3 Binding studies: Both exchanged and non-exchanged samples were prepared in 10 mM sodium phosphate, 20 mM sodium chloride, 0.1 mM 4,4-dimethyl-4-silapentane 1-sulfonic acid (DSS) and 0.3 mM ethylenediaminetetraacetic acid (EDTA), at a pH* 7.3, in 99.96 % D2O. The same buffer in 90 %/10 % H2O/D2O was used to observe imino protons. Titration experiments consisted of adding concentrated volumes of DAPI solution in 99.96 % D2O to 0.3 mM DNA samples at the following DNA to DAPI ratios: 1:0, 1:0.25, 1:0.5, 1:0.75, 1:1. The complex formation was then monitored by NMR. 3.4 NMR spectroscopy: NMR spectra were obtained using Bruker Avance 600 and 500 MHz NMR systems equipped with 5-mm QXI and 5-mm TBI probes, respectively. All NMR experiments were carried out at 25 oC and referenced to internal DSS. Imino proton spectra were recorded with a 1-1 solvent suppression
ACCEPTED MANUSCRIPT
sequence19. NOESY spectra were acquired at 600 MHz using a mixing time of 400 ms. 31P NMR spectra were proton decoupled and referenced to 85% H3PO4 in a capillary.
4. Conclusion NMR studies of DNA are often hampered by resonance overlap and the expense and limited availability of isotopically labeled oligonucleotides. We have presented a robust, nondestructive and economical method to simplify DNA NMR spectroscopy by exchanging the hydrogens of GH8 and AH8 with deuterons. This procedure, using only resources that are easily available in any laboratory, provides an accessible means of reducing NMR resonance overlap without perturbing the native DNA structure as evidenced by NMR. The deletion of proton signals from guanine tetrads facilitates investigations of other structural aspects, namely the surrounding loops and flanking sequences which determine many of the G tetrad unique properties. With a significant number of resonances effectively silenced, drug binding studies are more straightforward. Aknowledgements We thank Drs. Wilson and Spring-Connell for suggestions. Part of this work was supported by the Georgia Cancer Coalition and NIH (GM55404-01A1). References 1. Nelissen, F. H., Gammeren, A. J., Tessari, M., Girard, F. C., Heus, H. A., & Wijmenga, S. S. Multiple segmental and selective isotope labeling of large RNA for NMR structural studies. Nucleic Acids Research 2008, 36(14). 2. Nelissen, F. H., Leunissen, E.H., Laar, L., Tessari, M., Heus, H.A., Wimenga, S. S. Fast production of homogeneous recombinant RNA – towards large-scale production of RNA. Nucleic Acids Research 2012, 40(13). 3. Nelissen, F. H., Girard, F. C., Tessari, M., Heus, H. A., Wijmenga, S. S. Preparation of selective and segmentally labeled single-stranded DNA for NMR by self-primed PCR and asymmetrical endonuclease double digestion. Nucleic Acids Research 2009, 37(17). 4. Cromsigt, J., Schleucher, J., Gustafsson, T., Kihlberg, J., Wijmenga, S. Preparation of partially 2H/13C-labelled RNA for NMR studies. Stereo-specific deuteration of the H5’’ in nucleotides. Nucleic Acids Research 2002, 30(7), 1639-1645. 5. Nelissen, F. H., Tessari, M., Wijmenga, S. S., Heus, H. A. Stable isotope labeling methods for DNA. Progression Nuclear Magnetic Resonance Spectroscopy 2016, 96, 89-103. 6. Khan, A. M., Mishra, S. H., & Germann, M. W. Cyclic enzymatic solid phase synthesis of isotopically labeled DNA oligonucleotides. Nucleosides, Nucleotides and Nucleic Acids 2009, 28(11-12), 1030-1041.
ACCEPTED MANUSCRIPT
7. Zimmer, D.P.; Crothers, D.M. NMR of enzymatically synthesized uniformly 13C, 15N-labeled DNA oligonucleotides. Proc. Natl. Acad. Sci. 1995, 92, 3091–3095. 8. Masse, J.E.; Bortmann, P.; Dieckmann, T.; Feigon, J. Simple, efficient protocol for enzymatic synthesis of uniformly 13C, 15N-labeled DNA for heteronuclear NMR studies. Nucleic Acids Res. 1998, 26, 2618–2620. 9. Macdonald, D., Lu, P. Determination of DNA structure in solution: enzymatic deuteration of ribose 2’ carbon. J. AM. CHEM. SOC. 2002, 124(33), 9722-9723. 10. Englander, S. W., Sosnick, T. R., Englander, J. J., & Mayne, L. (1996). Mechanisms and uses of hydrogen exchange. Current Opinion in Structural Biology 1996, 6(1), 18-23. 11. Printz, M. P., & Hippel, P. H. Hydrogen exchange studies of DNA structure. Proceedings of the National Academy of Sciences 1965, 53(2), 363-370. 12. Maltseva, T. V., Földesi, A., & Chattopadhyaya, J. T1 and T2 relaxations of the 13C nuclei of deuterium-labeled nucleosides. Magnetic Resonance in Chemistry 1998, 36(4), 227-239. 13. Dupureur, C. M., & Barton, J. K. Use of Selective deuteration and 1H NMR in demonstrating major groove binding of DELTA.-[Ru(phen)2dppz]2 to d(GTCGAC)2. Journal of the American Chemical Society 1994, 116(22), 10286-10287. 14. Lukin, M., & Santos, C. D. Stereoselective nucleoside deuteration for NMR studies of DNA. Nucleosides, Nucleotides and Nucleic Acids 2010, 29(7), 562-573. 15. Brandes, R., Ehrenberg, A. Kinetics of the proton-deuteron enchange at position H8 of adenine and guanine in DNA. Nucleic Acids Research 1986, 14(23), 9491-9508. 16. Reilly, K. E., Thomas, G. J. Hygrogen exchange dynamics of the P22 virion determined by timeresolved raman spectroscopy. J. Mol. Biol. 1994, 241, 68-82. 17. Huang, X. An efficient and economic site-specific deuteration strategy for NMR studies of homologous oligonucleotide repeat sequences. Nucleic Acids Research 1997, 25(23), 4758-4763. 18. Keane, S. C., Van, V., Frank, H. M., Sciandra, C. A., Mccowin, S., Santos, J., Heng, X., & Summers, M. F. NMR detection of intermolecular interaction sites in the dimeric 5′-leader of the HIV-1 genome. Proceedings of the National Academy of Sciences 2016, 113(46), 13033-13038. 19. Chirakul, P., Litzer, J., & Sigurdsson, S. Preparation of base-deuterated 2’-deoxyadenosine nucleosides and their site-specific incorporation into DNA. Nucleosides, Nucleotides & Nucleic Acids 2001, 20(12), 1903-1913. 20. Meyer, B., & Peters, T. NMR Spectroscopy techniques for screening and identifying ligand binding to protein receptors. Angew Chem Int Ed Engl. 2003, 42(8):864-90. 21. Yan, X., & Maier, C. S. Hydrogen/deuterium exchange mass spectrometry. Mass spectrometry of proteins and peptides: methods and protocols 2009, 15, 255-271. 22. Tomasz, M., Olson, J., Mercado, C. M. Mechanism of the isotopic exchange of the C-8 hydrogen of purines in nucleosides and in deoxyribonucleic acid. Biochemistry 1972, 11(7), 1235-1241. 23. Benevides, J.M., Lemeur, D., and Thomas, G.J. Molecular conformations and 8-CH exchange rates of purine ribo- and deoxyribonucleotides: investigation by raman spectroscopy. Biopolymers 1984, 23, 1011-1023; 24. Benevides, J.M., and Thomas, G.J. Dependence of purine 8C-H exchange on nucleic acid conformation and base-pairing geometry: a dynamic probe of DNA and RNA secondary structures. Biopolymers 1985, 24,667-682
ACCEPTED MANUSCRIPT
-
NMR studies of DNA oligonucleotides are limited by extensive resonance overlap and the availability of isotopically labeled samples Exchanging GH8 and AH8 hydrogens with deuterium greatly simplifies NMR spectra, particularly of quadruplex DNA. The method only uses regular/commercial synthesized oligodeoxyribonucleotides and common chemicals. This approach facilitates studies of loop residues, dynamics and drug binding.