Abstract
The crystal structure of an alternating RNA heptamer r(GUAUACA) has been determined to 2.0 Å resolution and refined to an Rwork of 17.1% and Rfree of 18.5% using 2797 reflections. The heptamer crystallized in the space group C222 with a unit cell of a = 25.74, b = 106.58, c = 30.26 Å and two independent strands in the asymmetric unit. Each heptamer forms a duplex with its symmetry-related strand and each duplex contains six Watson–Crick base pairs and 3′-end adenosine overhangs. Therefore, two kinds of duplex (duplex 1 and duplex 2) are formed. Duplexes 1 stack on each other forming a pseudo-continuous column, which is typical of the RNA packing mode, while duplex 2 is typical of A-DNA packing with its termini in abutting interactions. Overhang adenine residues stack within the duplexes with C3′-endo sugar pucker and C2′-endo sugar pucker in duplexes 1 and 2, respectively. A Na+ ion in the crystal lattice is water bridged to two N1 atoms of symmetry-related A7 bases.
INTRODUCTION
RNA molecules with four basic building blocks are endowed with various biological functions, including storage of genetic information, translation and catalysis (1). In addition to the double helix structure, RNA molecules are also endowed with various structural motifs to perform their versatile functions (2). Unpaired bases are frequently observed in the biological RNA molecules. They exist in the form of bulges, loops and overhangs and offer the interaction sites for ligand–RNA recognition (3) and RNA folding. One good example is that transfer RNAs use the CCA overhangs as the universal 3′-termini. A strong preference towards the adenine base has been observed in rRNA among all the unpaired bases. Adenine bases account for 62% of the total unpaired bases in Escherichia coli 16S rRNA (4). The loop-out conformations of the bulge adenosines have been revealed by several X-ray crystallographic studies (5–8). Meanwhile, solution studies had shown that the adenine bulges adopt stacked-in conformations as an unpaired base in the RNA duplex (9) and pose little perturbation for the duplex conformation (10). Theoretical analysis of a single adenine bulge in RNA showed that a stacked-in conformation is the most energetically favorable form, especially when it has an upstream base C (11). In this paper, we present the crystal structure of the heptamer r(GUAUACA) with a six Watson–Crick base paired stem and a 3′ dangling adenine base.
MATERIALS AND METHODS
Synthesis, crystallization, data collection, structural solution and refinement
The RNA heptamer sequence, r(GUAUACA), was synthesized on the 1 µmol scale with our in-house Applied Biosystem 391 automatic DNA synthesizer using solid state phosphoramidite chemistry. The product was cleaved off the column by a 2:1 mixture of 37% ammonium hydroxide/absolute ethanol (v/v), followed by deprotection for 16 h in a hot water bath maintained at 55°C. The crude solution was then lyophilized to dryness. The resulting pellets were dissolved in 1 ml tetrabutylammonium fluoride (0.1 M in THF). The solution was kept at 55°C for 3–4 h in a dark place to deprotect the 2′-OH and the reaction was stopped by adding 200 µl of distilled water to the solution. The oligonucleotide products were recovered by butanol precipitation at –20°C and purified on FPLC. The hanging drop crystallization method was used at room temperature. Typical conditions were: 2 mM oligonucleotide (single-strand concentration), 40 mM sodium cacodylate buffer (pH 7.0), 24 mM CaCl2, 2% 2-methyl-2,4-dipentanediol (MPD) (v/v) against 40% MPD in the reservoir. Rod-shaped crystals grew to a size of 0.04 × 0.04 × 0.2 mm after 1 week. Intensity data were collected on our Rigaku imaging plate system, operated at 50 kV and 100 mA with Cu Kα radiation (λ = 1.5418 Å) by an oscillation method with 30 min exposure time for each frame, to 2.0 Å resolution. The data were reduced by DENZO software (12). A total of 22 206 reflections were collected for the crystal, of which 2840 (93.8% complete) reflections were unique. The space group is C222. The crystallographic data are summarized in Table 1.
Table 1. Crystal data and refinement statistics for the RNA heptamer.
| Crystal system | Orthorombic |
| Space group | C222 |
| Cell parameters | |
| a (Å) | 25.74 |
| b (Å) | 106.58 |
| c (Å) | 30.26 |
| Crystal dimensions (mm3) | 0.04 × 0.04 × 0.2 |
| Volume/base pair (Å3) | 1482 |
| Asymmetric unit | 2 strands |
| Rsym (%) based on I | 5.4 |
| Resolution (Å) | 2.0 |
| No. of unique reflections | 2840 |
| Data completeness (%) | 93.8 |
| Rwork/Rfree (%) | 17.1/18.5 |
| Last shell | |
| Completeness (%) | 77.1 |
| Rwork/Rfree (%) | 24.5/28.9 |
| Parameter file | dna-rna_rep.param |
| r.m.s. deviation from ideal geometry | |
| Bond lengths (Å) | 0.003 |
| Bond angles (°) | 0.8 |
| Torsion angles (°) | 10.5 |
| Improper angles (°) | 1.33 |
The molecular replacement program AmoRe was used to search the structural solution, r(GUAUACA). A single strand of r(GUAUAC) after removing rAdC from r(GUAUACA)dC (13), AR0002, was used as the search model. The structural solution shows there are two heptamer strands in an asymmetric unit. The CNS (14) program was used to refine the structure. Simulated annealing, heating to 3000 K and slow cooling to 300 K, was performed to minimize the bias from the initial models. A |Fo – Fc| map calculated at this stage shows the electron density of the overhang adenine (A7) in duplex 1. An adenine was introduced and the energy minimization reduced the Rwork and Rfree to 34.6 and 36.3%. The subsequently calculated electron density map clearly shows the anti conformation for the second overhang adenine with the ribose sugar in the C2′-endo conformation. Introducing the second adenine overhang and model building was done using CHAIN (15) by displaying electron densities. The final crystallographic Rwork and Rfree are 17.1 and 18.5% [10– 2.0 Å, F > 2σ(F)] using 2797 unique reflections. The final model for the structure consists of atoms for two strands, one sodium ion and 40 water molecules. The r.m.s. deviations in bond lengths and angles with other refinement information are summarized in Table 1. The coordinates and structure factors have been deposited with the NDB (16) with the accession no. AR0040.
RESULTS AND DISCUSSION
Overall structure
The heptamer RNA r(GUAUACA) crystallized in the space group C222 with two independent strands and a Na+ in the asymmetric unit. Each heptamer strand and its symmetry-related strand form a six Watson–Crick base paired stem with 3′-end dangling adenines. Therefore, there exist two different duplexes in the crystal lattice. The numbering scheme for the two duplexes and their relative orientations are shown in Figure 1. Duplex 1 molecules stack on each other forming pseudo-continuous columns, typical of RNA crystal packing. The terminal bases of duplex 2 stack onto the minor groove of duplex 1 (Fig. 1), typical of A-DNA crystal packing with abutting interactions. All the RNA duplexes known so far adopt the head-to-tail or head-to-head packing pattern forming pseudo-continuous columns. The A-form DNA duplex always exhibits the abutting interactions, with the terminal base pair interacting with the minor groove of the symmetry-related duplex (17). This is the first observation of an RNA duplex which illustrates the different packing modes in the same crystal.
Figure 1.
Diagram of arrangement of the heptamers in the crystal lattice. (A) Strand 1 and strand 2 in the asymmetric unit are designated in cyan (residues are numbered 1–7) and green (residues are numbered 8–14), respectively. All the symmetry-related strands are designated in black. The interaction sites between duplex 1 and duplex 2 are orange or yellow colored. (B) View of the crystal packing along the crystallographic a-axis. Duplexes 1 (cyan) stack on each other forming pseudo-continuous columns (perpendicular to the paper). Duplexes 2 (green) form bridges between the columns by positioning their terminal bases in the minor grooves of duplexes 1, which is the typical A-DNA packing mode. Duplex 1 has overhang adenine A·A pairs that are buried in the pseudo-continuous columns. Duplex 2 does not have an A·A pair, but the overhang adenines are buried.
The six base-paired stem of the duplex adopts a typical A-RNA (11-fold) conformation; with all the helical ribose sugars and bases in the C3′-endo and anti glycosyl conformations. The helical parameters were calculated using the program CURVES (18) and are listed in Table 2. Duplex 1 has a bending of 11° at the central step of its helical axis while duplex 2 is straight. Duplex 1 has a roll angle of 21° at the central step (average roll value 15.1°) while duplex 2 has a roll angle at the central step of 14° (average roll value 11.1°). Duplex 1 has an average rise of 2.3 Å and twist of 34.8° while duplex 2 has 2.5 Å and 33.3°, respectively. Comparison of the A-form structure of the DNA octamer d(CCCCGGGG) (19) and RNA octamer r(CCCCGGGG) (20) show similar trends but with more significant differences: the RNA duplexes have a shorter rise (average 2.5 Å) than the A-DNA duplexes (3.1 Å). The RNA duplexes display a relatively larger inclination (average 16.6°) than the A-DNA duplexes (average 8.0°). However, the torsion angles are similar for the duplex residues. The conformations of the overhang adenosines are different with the sugar pucker being C3′-endo in duplex 1 and C2′-endo in duplex 2.
Table 2. Selected helical parameters.
| Shift (Å) (Dx) | Slide (Å) (Dy) | Rise (Å) | Tilt (°) | Roll (°) | Twist (°) | |
|---|---|---|---|---|---|---|
| Strand 1 | ||||||
| G | –0.01 | –1.61 | 2.49 | –2.2 | 11.3 | 36.19 |
| U | –0.27 | –1.83 | 2.10 | 1.07 | 16 | 34.62 |
| A | 0 | –1.43 | 2.20 | 0 | 21.02 | 36.09 |
| U | 0.27 | –1.83 | 2.29 | –1.07 | 16 | 35.84 |
| A | 0.01 | –1.61 | 2.41 | 2.2 | 11.3 | 31.45 |
| Ave | 0 | –1.66 | 2.30 | 0 | 15.12 | 34.8 |
| Strand 2 | ||||||
| G | –0.03 | –1.8 | 2.69 | –3.31 | 8.54 | 34.20 |
| U | 0.13 | –1.87 | 2.38 | 0.1 | 12.34 | 36.74 |
| A | 0 | –1.58 | 2.48 | 0 | 14.05 | 34.74 |
| U | –0.13 | –1.87 | 2.41 | –0.1 | 12.34 | 30.63 |
| A | 0.03 | –1.8 | 2.53 | 3.31 | 8.54 | 30.48 |
| Ave |
0 |
–1.79 |
2.50 |
0 |
11.16 |
33.34 |
| Torsion |
α (°) P-O5′ |
β (°) O5′-C5′ |
γ (°) C5′-C4′ |
δ (°) C4′-C3′ |
χ (°) C1′-N |
|
| Ave (duplex 1) | –64.02 | 170.88 | 51.66 | 80.79 | –157.79 | |
| A7 | –61.21 | 170.33 | 53.45 | 74.43 | –150.95 | |
| Ave (duplex 2) | –62.68 | 172.11 | 51.40 | 81.35 | –158.50 | |
| A14 |
–63.79 |
178.41 |
62.03 |
133.72 |
–128.21 |
|
| |
v1 (C1′-C2′) |
v2 (C2′-C3′) |
v3 (C3′-C4′) |
v4 (C4′-O4′) |
v0 (O4′-C1′) |
|
| Ave (strand 1) | –26.34 | 39.87 | –39.87 | 24.09 | 1.66 | |
| A7 | –24.61 | 28.73 | –39.86 | 25.17 | –0.16 | |
| Ave (strand 2) | –26.16 | 39.30 | –39.30 | 23.73 | 1.70 | |
| A14 | 31.03 | –27.22 | 15.09 | 4.54 | –22.49 | |
Conformation of the overhang adenines
The overhang adenines from both duplexes stack within their respective duplexes with well-defined electron densities (Fig. 2). We have studied three other structures of RNA oligomers forming six Watson–Crick base-paired duplexes and two base overhangs, r(GUGUGUA)dC with 5′ overhangs (21), r(GUACACA)dC (22) and r(GUAUACA)dC (13) with 3′ overhangs. The overhang adenines from duplex 1 in the present structure can adopt several possible arrangements. The first possibility is that adenine swings out and forms a reverse Hoogsteen pair like the G·G pair in the Z-DNA overhang (23). The second possibility is that adenine lies in the minor groove of a symmetry-related duplex forming minor groove base triples (13,22). The structure shows that duplexes 1 stack on each other forming pseudo-continuous columns along the crystallographic a-axis. The overhang adenines from symmetry-related duplex 1 are closed together with the conformation resembling the A·I pair in r(CGCAIGCG)2 (24) and A·G in (CGCGAAUUAGCG)2 (25). However, there are no direct hydrogen bond interactions between the two adenines. Water molecules, including five water molecules coordinated to the Na+, form bridges between the two adenines (Fig. 3). The stacking between the A·A base pair and the adjacent G·C base pairs, with both twist angles nearly 0°, maximizes the stacking. The present A·A base pair shows the important role of water molecules, ion–water coordinates and base stacking in base pair formation.
Figure 2.
Electron density 2|Fo – Fc| maps for the two overhang adenines at 1σ. (A) A7 in strand 1 having C3′-endo sugar puckering and (B) A14 in strand 2 having a C2′-endo sugar puckering.
Figure 3.
Geometry of the adenine overhang at the junction of the duplexes and their interactions with water molecules. The original molecule (A7) is in green and the symmetry-related molecule (A7) is in blue. Na+ is in yellow. Water molecules are in red.
The deep groove of the RNA duplex is narrow, which limits the access of proteins or ligands to it, while the shallow groove is wide and is a better candidate. The adenine overhangs in duplex 2 do not form an A·A base pair as in duplex 1, but interact with duplex 1 on the minor groove side (Figs 1 and 4). The six member ring of A14 is in van der Waals distance of O4′ of U4 (Fig. 4A). A series of hydrogen bonding interactions are also formed between the terminal residues (G8 and A14) of duplex 2 and bases of duplex 1 (detailed in Fig. 4B). O2′ of G8* (* represents symmetry related) bridges O2′ and N3 of A5*. Both N1 and N6 of the overhang adenine of duplex 2 (A14) interact with O2′ of A3. A water molecule bridges N3 of A14 and O2 of U4*. O2, O2′ of U4, O2′ of A14 and a water molecule form a hydrogen bonding network. These interactions are formed only when the RNA molecules are close to each other and could mimic specific interactions between ligands or proteins and minor groove atoms of RNA molecules.
Figure 4.
(A) The terminal of duplex 2 (orange and green for each strand) stacks onto the minor groove of duplex 1 (blue and cyan for each strand). The interactions between the O4′ of U4 and the ring of the overhang adenine are shown as red dashed lines. (B) The hydrogen bonding interactions between the terminal residues of duplex 2 and the atoms of duplex 1, either directly or water molecule-mediated.
Hydration
A total of 40 water molecules are located in the asymmetric unit. Because of the different packing of duplex 1 and duplex 2, the two duplexes have quite different hydration patterns. Overall, duplex 1 is more hydrated than duplex 2. In duplex 1, 18 water molecules are located in the major groove. The minor groove of duplex 1 lacks hydration because it is occupied by the terminals of duplexes 2. Several water molecules are trapped between duplex 1 and duplex 2 and function as bridges between the two duplexes (Fig. 4B). The hydration pattern of the major groove of duplex 1 is shown in Figure 5A. The hydration sites in the major groove are N7, O(N)6 (O6 for guanine, N6 for adenine) of purine, O(N)4 (O4 for uracil, N4 for cytosine) of pyrimidine and phosphate oxygen atoms. In contrast, few water molecules have been located in the major groove of duplex 2 because the groove is open to the crystal environment. Two molecules of duplex 2 face each other on the minor groove sides with very few direct interactions between them. Six water molecules form an elaborate hydrogen bonding network bridging the two duplexes. The hydration pattern of duplex 2 in the minor groove is shown in Figure 5B. The hydration sites in the minor groove are N3 for purine, O2 for pyrimidine and O2′, O4′ for backbone. The pattern of one water molecule bridging O2′ and N3 (for purine) or O2 (for pyrimidine) is seen at most of the bases. One water molecule connecting the O2′ and O4′ atoms of the following residue is shown at U11.
Figure 5.
(A) Major groove hydration of duplex 1. (B) Minor groove hydration of duplex 2. The water molecules bridging two molecules of duplex 2 are in pink. Others are in red.
Acknowledgments
ACKNOWLEDGEMENTS
We gratefully thank the NIH for grant GM-17378 and the Board of Regents of Ohio for an Ohio Eminent Scholar Chair and Endowment to M.S. We also acknowledge the Hays Consortium Investment Fund by the Regions of Ohio for partial support for purchasing the Raxis IIc imaging plate system.
NDB no. AR0040
REFERENCES
- 1.Cech T.R. and Golden,B.L. (1999) Building a catalytic active site using only RNA. In Gesteland,R.F., Cech,T.R. and Atkins,J.F. (eds), The RNA World. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 321–350.
- 2.Tinoco I. and Bustamante,C. (1999) How RNA folds. J. Mol. Biol., 293, 278–281. [DOI] [PubMed] [Google Scholar]
- 3.Vicens Q. and Westhof,E. (2001) Crystal structure of paromomycin docked into the eubacterial ribosomal decoding a site. Structure, 9, 647–658. [DOI] [PubMed] [Google Scholar]
- 4.Gutell R.R., Weiser,B., Woese,C.R. and Noller,H.F. (1985). Comparative anatomy of 16-s-like ribosomal-RNA. Prog. Nucleic Acid Res. Mol. Biol., 32, 155–216. [DOI] [PubMed] [Google Scholar]
- 5.Cate J.H., Gooding,A.R., Podell,E., Zhou,K., Golden,B., Kundrot,C., Cech,T. and Doudna,J. (1996) Crystal structure of a group I ribozyme domain: principles of RNA packing. Science, 273, 1678–1685. [DOI] [PubMed] [Google Scholar]
- 6.Grahn E., Stonehouse,N.J., Murray,J.B., Vandenworm,S., Valegard,K., Fridborg,K., Stockley,P.G. and Liljas,L. (1999) Crystallographic studies of RNA hairpins in complexes with recombinant MS2 capsids: implications for binding requirements. RNA, 5, 131–138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ennifar E., Yusupov,M., Walter,P., Marquet,R., Ehresmann,C., Ehresmann,B. and Dumas,P. (1999) The crystal structure of the dimerization initiation site of genomic hiv-1 rna reveals an extended duplex with two adenine bulges. Structure, 7, 1439–1449. [DOI] [PubMed] [Google Scholar]
- 8.Sudarsanakumar C. Xiong,Y. and Sundaralingam,M. (2000) Crystal structure of an adenine bulge in the RNA chain of a DNA·RNA hybrid, (CTCCTCTTC)/r(gaagagagag). J. Mol. Biol., 299, 103–112. [DOI] [PubMed] [Google Scholar]
- 9.Borer P.N., Lin,Y., Wang,S., Roggenbuck,M.W., Gott,J.M., Uhlenbeck,O.C.and Pelczer,I. (1995) Proton NMR and structural features of a 24-nucleotide RNA hairpin. Biochemistry, 34, 6488–6503. [DOI] [PubMed] [Google Scholar]
- 10.Thiviyanathan V., Guliaev,A.B., Leontis,N.B. and Gorenstein,D.G. (2000) Solution conformation of a bulged adenosine base in an RNA duplex by relaxation matrix refinement. J. Mol. Biol., 300, 1143–1154. [DOI] [PubMed] [Google Scholar]
- 11.Zacharias M. and Sklenar,H. (1999) Conformational analysis of single-base bulges in A-form DNA and RNA using a hierarchical approach and energetic evaluation with a continuum solvent model. J. Mol. Biol., 289, 261–275. [DOI] [PubMed] [Google Scholar]
- 12.Otwinowski Z. and Minor,W. (1997) Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol., 276, 307–326. [DOI] [PubMed] [Google Scholar]
- 13.Shi K., Biswas,R., Mitra,S.N. and Sundaralingam,M. (2000) The crystal structure of the octamer [r(GUAUACA)dC]2 with six Watson-Crick base-pairs and two 3′ overhang residues. J. Mol. Biol., 299, 113–122. [DOI] [PubMed] [Google Scholar]
- 14.Brunger A.T., Adams,P.D., Clore,G.M., DeLano,W.L., Gros,P., Grosse-Kunstleve,R.W., Jiang,J.-S., Kuszewski,J., Nilges,N., Pannu,N.S. et al. (1998) Crystallography and NMR system (CNS): a new software system for macromolecular structure determination. Acta Crystallogr., D54, 905–921. [DOI] [PubMed] [Google Scholar]
- 15.Sack J. and Quiocho,F.A. (1997) Chain-crystallographic modelling program. Methods Enzymol., 277, 158–173. [DOI] [PubMed] [Google Scholar]
- 16.Berman H.M., Olson,W.K., Beveridge,D.L., Westbrook,J., Gelbin,A., Demeny,T., Hsieh,S.-H., Srinivasan,A.R. and Schneider,B. (1992) The Nucleic Acid Database: a comprehensive relational database of three-dimensional structures of nucleic acids. Biophys. J., 63, 751–759. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wahl M.C. and Sundaralingam,M. (1995) New crystal structures of nucleic acids and their complexes. Curr. Opin. Struct. Biol., 5, 282–295. [DOI] [PubMed] [Google Scholar]
- 18.Lavery R. and Sklenar,H. (1989) Defining the structure of irregular nucleic acids: principles and conventions. J. Biomol. Struct. Dyn., 4, 655–667. [DOI] [PubMed] [Google Scholar]
- 19.Eisenstein M. and Shakked Z. (1995) Hydration patterns and intermolecular interactions in A-DNA crystal structures. Implications for DNA recognition. J. Mol. Biol., 248, 662–678. [DOI] [PubMed] [Google Scholar]
- 20.Portmann S., Usman,N. and Egli,M. (1995) Crystal structure of r(CCCCGGGG) in two distinct lattices. Biochemistry, 34, 7569–7575. [DOI] [PubMed] [Google Scholar]
- 21.Biswas R. and Sundaralingam,M. (1997) Crystal structure of r(GUGUGUA)d(C) with tandem G-U/U-G wobble pairs with strand slippage. J. Mol. Biol., 270, 511–519. [DOI] [PubMed] [Google Scholar]
- 22.Mitra S.N. Biswas,R., Shi,K. and Sundralingam,M. (2000) Crystal structure of an octamer [r(GUGCACA)dC]2 with 3′-dinucleoside overhang forming a superhelix. J. Biomol. Struct. Dyn., 299, 113–122. [DOI] [PubMed] [Google Scholar]
- 23.Pan B., Ban,C., Wahl,M. and Sundaralingam,M. (1997) Crystal structure of d(GCGCGCG) with 5′-overhang G residues. Biophys. J., 73, 1553–1561. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Carter R.J., Baeyens,K.J., SantaLucia,J.,Jr, Turner,D.H. and Holbrook,S.R. (1997) The crystal structure of an RNA oligomer incorporating tandem adenosine-inosine mismatches. Nucleic Acids Res., 25, 4117–4122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Leonard G.A., McAuley-Hecht,K.E., Ebel,S., Lough,D.M., Brown,T. and Hunter,W.N. (1994) Crystal and molecular structure of r(CGCGAAUUAGCG): an RNA duplex containing two G(anti).A(anti) base-pairs. Structure, 2, 483–494. [DOI] [PubMed] [Google Scholar]





