| Nucleic Acids Research | Pages |
N 2-Methylguanosine is iso-energetic with guanosine in RNA duplexes and GNRA tetraloops
Introduction
Materials And Methods
RNA synthesis and purification
Oligonucleotide melting studies and thermodynamics
NMR spectroscopy
Results And Discussion
Acknowledgements
References
N 2-Methylguanosine is iso-energetic with guanosine in RNA duplexes and GNRA tetraloops
ABSTRACT
INTRODUCTION
Post-transcriptional modifications increase the chemical diversity of the nucleotides found in RNAs. tRNAs are densest in modified bases, but rRNA, snRNA and mRNA are also significantly modified (1). The metabolic cost of modification is very high, yet, with few exceptions, the purposes served are unknown. Among more than 100 naturally occurring nucleotide modifications, the simplest and most common are methylations of nucleotide base or ribose moieties (1). One such example is methylation of the exocyclic amine of G which occurs within tRNA, rRNA and snRNA (1). N2-methylguanine (m2G) is found within both helical and looped regions of RNA secondary structure (2,3), and it can exist in either the s-cis or the s-trans rotamer (Fig. Helix 45 of bacterial 16S rRNA provides an excellent example of how post-synthetic methylation can alter RNA function (4). Helix 45, the last stem in 16S rRNA, contributes to the formation of the 30S initiation complex, and forms part of the 30S/50S subunit interface (4,5). The helix is capped by a highly conserved, hyper-methylated GGAA tetraloop, that is modified to yield the sequence m2G-G-m26A-m26A (where m26A is N6,N6-dimethyl-adenine) (4). Methylation of the first G is conserved in bacteria (6). Methylation of both A residues is also conserved in bacteria, though the loss of adenosine dimethylase activity responsible for methylating these residues confers resistance to the antibiotic kasugamycin (7-9). In contrast, the loss of adenosine dimethylase activity is lethal in yeast (10). These results suggest that base methylation within helix 45 plays an important role in ribosome function. The solution structure of the fully methylated tetraloop has been determined by NMR, and it is substantially different than the unmethylated GNRA (N = any base, R = A or G) tetraloops previously reported (11). One noticeable difference is the G@A sheared pair, which forms between the first G and the last A in standard GNRA tetraloops, is disrupted in the helix 45 tetraloop due to the dimethylation of the last A. The m2G-G-m26A-m26A structure does not appear to be directly affected by the m2G modification, though the structure of the m2G-G-A-A variant (kasugamycin resistant form) has not yet been determined. In an effort to extend our characterization of the effects of m2G substitution on RNA structure, we measured the thermodynamic consequences of the methylations in the helix 45 tetraloop from Bacillus stearothermophilis. Figure 1. The two possible rotamers of the methylamine of m2G. (a) s-cis m2G. (b) s-trans m2G. In addition to being of interest biologically, methylated nucleotides are also being employed in RNA structure/function analyses using a method termed Nucleotide Analog Interference Mapping (NAIM) (12-14). In this approach, the methylated nucleotide is tagged with an [alpha]-phosphorothioate linkage, incorporated into the RNA by in vitro transcription, and the active RNA molecules in the population separated from the inactive variants by a selection assay. Sites intolerant of methylation are identified by iodine cleavage of the phosphorothioate linkages in the active population followed by analysis of the cleavage products (15). While identification of positions intolerant of methylation is fairly straight forward, interpretation of interference patterns can be more difficult. It is important to understand the thermodynamic effects methylation has on duplex stability so that secondary and tertiary interference effects can be distinguished from one another. To help identify the role played by base methylation, we have determined the thermodynamic consequences of replacing G with m2G in two different RNA base pairs within model duplexes. We have also examined its effect on the stability of the GNRA tetraloop in helix 45. These measurements include the effect of substituting m2G in place of G in the context of three base pairing possibilities: an m2G@C base pair where the methyl group must be s-trans, an m2G@A sheared base pair where the methyl group must be s-cis, and an m2G@U base pair where the methyl group can be in either configuration. In all of these contexts m2G is essentially isoenergetic with G. We also re-investigated the effects of m26A substitution on the stability of the helix 45 tetraloop.
MATERIALS AND METHODS
RNA synthesis and purification
Previously synthesized N6,N6-dimethyladenosine and O6-NPE-N2-methylguanosine phosphoramidites (16) were incorporated into several RNAs using standard phosphoramidite chemistry. The RNA was cleaved from the solid support and the amino protecting groups were removed in a 3:1 mixture of ammonia (conc.) and ethanol (55°C overnight). To 1 µmol of synthetic RNA, 1.5 ml of neat triethylamine trihydrofluoride was added and agitated for 12-24 h. This mixture was diluted with 300 µl H2O and precipitated with 10 ml n-butanol (-20°C for 4 h). After centrifugation the supernatant was decanted and the RNA pellet dried under vacuum. For oligoribonucleotides containing NPE-O6 protected N2-methylguanosine, the NPE group was removed by dissolving the product in 1.8 ml of 1.0 M TBAF in THF and agitating for 24 h. The RNA was precipitated by adding 600 µl of 1 M NaCl, 1.8 ml H2O and 7.2 ml EtOH, and storing at -20°C for 4 h. After centrifugation the supernatant was decanted and the RNA pellet dried under vacuum. Fully deprotected RNAs were purified by denaturing (5 M urea) polyacrylamide (20%) gel electrophoresis. RNA was eluted from gel slices by repeated incubation (3 × 25 ml) in 1 M NH4OAc at 55°C for 3 h. The RNAs were desalted using C18 Sep-pak cartridges (Millipore) (17).
Oligonucleotide melting studies and thermodynamics
Oligonucleotide concentrations were calculated from their absorbance at 260 nm. For oligonucleotides containing only A, C, G and U, molar extinction coefficients were calculated from the nearest neighbor approximation (18). For oligonucleotidescontaining methylated bases, values of the nearest neighbor approximations were based on the observation that plots of the extinction coefficients of X versus XY (X,Y = A, C, G or U) are linear. Thus, by using the molar extinction coefficient of m26A ([epsis]260 = 11 000) and m2G ([epsis]260 = 13 000), the nearest neighbor contributions were estimated. Absorbance versus temperature curves were measured at 260 nm in 1.0 M NaCl, 1 mM EDTA and 10 mM sodium cacodylate (pH 7.0), for duplexes, and 50 mM NaCl, 5 mM cacodylate (pH 6.3), 1 mM EDTA, for hairpins, with a Cary 3E spectrophotometer using a heating rate of 1°C/min. Van't hoff enthalpies ([Delta]H°) and melting temperatures (Tm) were determined using the derivative method of Gralla and Crothers (19). The entropy of formation ([Delta]S°) was calculated using the following equations:
| 1/Tm = R/[Delta]H°lnCt + [Delta]S°/[Delta]H° |
for a self-complementary duplex, and
| 1/Tm = [Delta]S°/[Delta]H° |
for monomolecular hairpins. The reported thermodynamic values are the average of at least three trials, usually at different total strand concentrations (Ct).
NMR spectroscopy
The RNA was dissolved in 170 µl 90% D2O/10% H2O [50 mM NaCl, 5 mM cacodylate (pH 6.3) and 1 mM EDTA] to a strand concentration of 1.8-2.2 mM. Spectra at 5°C were collected using a water flip-back pulse sequence (20) for water suppression on either a Varian Unity (11.7 tesla) or a Varian UnityPlus (14 tesla) spectrometer. Stem imino proton resonances were assigned from sequential imino/imino NOEs using 2D NOESY experiments (data not shown).
RESULTS AND DISCUSSION
RNA containing the modified nucleotides m2G and m26A were prepared by solid phase synthesis (16). Terminal and internal m2G-C and m2G@U base pair stabilities were measured for the self-complementary duplexes 5[prime]-XCCGGY-3[prime] and 5[prime]-GYCUAGXG-3[prime] (where X is either G or m2G and Y is either C or U) (Table
Table 1.
| Pair (X, Y) | -[Delta]H° (kcal/mol)b |
-[Delta]S° (cal/mol/K)b |
Tm(°C)c | -[Delta]G°37 (kcal/mol) |
[Delta][Delta]Gd (m2G versus G) per base pair |
| Watson-Crick pairs | |||||
| internal: GYCUAGXC | |||||
| C-Ge | 81.63 (±1.88) | 218.4 (±6.0) | 71.9 | 13.9 (±0.1) | |
| C-m2G | 82.39 (±3.12) | 219.9 (±9.2) | 72.9 | 14.1 (±0.3) | -0.1 |
| terminal: XCCGGY | |||||
| C-Gf | 75.70 (±7.23) | 205.8 (±21.0) | 65.0 | 11.9 (±0.7) | |
| C-m2G | 74.50 (±2.61) | 202.5 (±8.4) | 64.4 | 11.9 (±0.1) | 0.0 |
| Wobble pairs | |||||
| internal: GYCUAGXC | |||||
| U[bull]Gg | 73.65 (±2.07) | 212.0 (±6.5) | 46.8 | 7.9 (±0.1) | |
| U[bull]m2G | 74.37 (±1.66) | 212.4 (±5.5) | 49.3 | 8.5 (±0.0) | -0.3 |
| terminal: XCCGGY | |||||
| U[bull]Gh | 64.15 (±1.34) | 176.7 (±3.9) | 56.0 | 9.4 (±0.2) | |
| U[bull]m2G | 64.21 (±3.03) | 176.7 (±9.9) | 56.2 | 9.4 (±0.2) | 0.0 |
| Tetraloop stabilities: | |||||
| XGGACCXG | |||||
| CCUGGYY | |||||
| -[Delta]G°70 i | |||||
| X = G; Y = A | 38.27 (±4.07) | 107.7 (±11.3) | 82.2 | 1.3 (±0.2) | |
| X = m2G; Y = A | 41.32 (±4.66) | 116.4 (±13.1) | 81.8 | 1.4 (±0.2) | -0.1 |
| X = m2G; Y = m26A | 35.17 (±5.70) | 100.0 (±16.4) | 78.8 | 0.9 (±0.1) | +0.4 |
The melting data indicate that replacement of G by m2G in G-C pairs has no effect on helix stability at either internal or terminal positions. This result is consistent with mononucleoside association studies performed in organic solvents, which showed that the equilibrium constants for G-C and m2G-C pairs are approximately equal (23). However, these data are in sharp contrast to a previous study done using poly(m2G)@poly(C) duplexes (24). If this were also true for mixed sequences, then we would have expected a substantial difference between the stability of an internal and a terminal m2G-C pair, because the methyl group would be less sterically constrained at the end of a helix. The fact that we do not observe such a difference suggests that the RNA minor groove can accommodate a methyl group on the s-trans face of the G exocyclic amine. The large disparity between our results and those obtained with poly(m2G) may simply be due to the difference in the levels of m2G incorporation. In the present study, the duplexes contain isolated m2G substitutions, which is typical in natural RNAs; whereas runs of m2G were present in the polynucleotide experiments. Thus, taken together, these two studies suggest that consecutive m2G substitutions have a destabilizing effect on the duplex, presumably due to steric clashes between neighboring methyl groups in the minor groove.
m2G@U pairs are slightly more stable (0.3 kcal/mol) than G@U pairs at internal positions within a helix (Table
We also characterized methylated versions of the GGAA tetraloop from 16S rRNA helix 45 to determine the effects of methylating the G in a sheared G@A pair (specifically the G6@A9 pair; see numbering in Fig. Figure 2. The three types of m2G pairings investigated. (a) In an m2G-C base pair the methyl group must adopt the s-trans rotamer, which places the methyl group into the minor groove of the double helix. (b) In an m2G@U wobble base pair, the methyl group can adopt either the s-cis or the s-trans configuration, because the exocyclic methylamine is not involved in hydrogen bonding. (c) In a sheared m2G@A base pair the methyl group must adopt the s-cis configuration. Figure 3. The imino regions from one dimensional 1H spectra collected at 5°C for three helix 45 tetraloop analogs variants. (a) The hairpin sequence and nucleotide numbering is shown and the tetraloop sequence for each sample (X6G7Y8Y9) is shown above each spectrum. (b) The imino region of the unmodified parental sequence (X = G; Y = A). The assignment of G6 was based on previous reports of other GNRA hairpins NMR assignments (31). (c) The imino region of the helix 45 tetraloop (X = m2G; Y = A). The imino resonance of m2G6 is absent and the intensity of the G10 imino resonance is reduced compared to the corresponding imino resonance in (b). (d) The imino region of the wild-type helix 45 tetraloop (X = m2G; Y = m26A). As with the spectrum in (c), the m2G6 imino resonance is absent; however, the intensity of the G10 imino resonance is restored to full strength. These results contradict an earlier study that suggested m2G destabilizes the helix 45 tetraloop by 1.3 kcal/mol (28). The earlier report was ambiguous, however, because the RNA molecules compared in that study differed in the length of their helical stems, and the authors had to include a correction factor to account for this difference. The hairpins used in the present study differ only in the methylation state of G6, and therefore no correction factor is necessary. In agreement with previous reports, m26A substitution of A8 and A9 had a modest destabilizing effect on the tetraloop (0.2 kcal/mol per substitution) (29). The primary cause of this loss of stability is likely to be the disruption of the sheared G6@A9 pair, which based upon the NMR structure of this tetraloop, does not form when A9 is methylated (11). The net iso-energetic effects observed for the G-C/m2G-C and G@A/m2G@A pairs are likely to be a balance between weak stabilization due to improved hydrophobic interactions and modest destabilization due to the loss of rotational entropy of the methyl group or weakened hydrogen bonding effects. The values of [Delta]S° and [Delta]H° do not change significantly in going from a G-C base pair to a m2G-C base pair and from a G@A base pair to a m2G@A base pair. To characterize the structural consequences of tetraloop methylation, we collected one-dimensional, imino proton NMR spectra for all of the variants of the helix 45 tetraloops studied thermodynamically (Fig. It is difficult to predict the effect of m2G on RNA secondary or tertiary structural stability because it can exist in either of two rotamers, s-cis or s-trans (Fig.
ACKNOWLEDGEMENTS
This work was supported by a grant from the NIH (GM41651) to P.B.M. and an NSF CAREER award (CHE-9701787) and a Beckman Young Investigator Award to S.A.S.
REFERENCES
This article has been cited by other articles:
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 30 Jul 1998
Copyright©Oxford University Press, 1998.
![]()
CiteULike
Connotea
Del.icio.us What's this?
![]()
![]()

![]()
![]()
![]()
P. S. Pallan, C. Kreutz, S. Bosio, R. Micura, and M. Egli
Effects of N2,N2 -dimethylguanosine on RNA structure and stability: Crystal structure of an RNA duplex with tandem m2 2G:A pairs
RNA,
October 1, 2008;
14(10):
2125 - 2135.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
M. Kwon and S. A. Strobel
Chemical basis of glycine riboswitch cooperativity
RNA,
January 1, 2008;
14(1):
25 - 34.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
J. Armengaud, J. Urbonavicius, B. Fernandez, G. Chaussinand, J. M. Bujnicki, and H. Grosjean
N2-Methylation of Guanosine at Position 10 in tRNA Is Catalyzed by a THUMP Domain-containing, S-Adenosylmethionine-dependent Methyltransferase, Conserved in Archaea and Eukaryota
J. Biol. Chem.,
August 27, 2004;
279(35):
37142 - 37152.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
J. Sarzynska, L. Nilsson, and T. Kulinski
Effects of Base Substitutions in an RNA Hairpin from Molecular Dynamics and Free Energy Simulations
Biophys. J.,
December 1, 2003;
85(6):
3445 - 3459.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
R. Micura, W. Pils, C. Hobartner, K. Grubmayr, M.-O. Ebert, and B. Jaun
Methylation of the nucleobases in RNA oligonucleotides mediates duplex-hairpin conversion
Nucleic Acids Res.,
October 1, 2001;
29(19):
3997 - 4005.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
M. Yasui, S. Matsui, M. Ihara, Y. R. S. Laxmi, S. Shibutani, and T. Matsuda
Translesional synthesis on a DNA template containing N2-methyl-2'-deoxyguanosine catalyzed by the Klenow fragment of Escherichia coli DNA polymerase I
Nucleic Acids Res.,
May 1, 2001;
29(9):
1994 - 2001.
[Abstract]
[Full Text]
[PDF]
![]()
This Article ![]()
![]()
Abstract
![]()
Print PDF (63K)
![]()
Alert me when this article is cited
![]()
Alert me if a correction is posted
![]()
Services ![]()
![]()
Email this article to a friend
![]()
Similar articles in this journal
![]()
Similar articles in ISI Web of Science
![]()
Similar articles in PubMed
![]()
Alert me to new issues of the journal
![]()
Add to My Personal Archive
![]()
Download to citation manager
![]()
Search for citing articles in:
ISI Web of Science (19)
![]()
Request Permissions ![]()
Commercial Re-use Guidelines
for Open Access NAR Content
![]()
Google Scholar ![]()
![]()
Articles by Rife, J. P.
![]()
Articles by Strobel, S. A.
![]()
Search for Related Content
![]()
PubMed ![]()
![]()
PubMed Citation
![]()
Articles by Rife, J. P.
![]()
Articles by Strobel, S. A.
![]()
Social Bookmarking ![]()
![]()
What's this?