Atomic resolution duplex structure of the simplified nucleic acid GNA

Article abstract of DOI:10.1039/b916851f

Double helix variations of glycol nucleic acids (GNA) are revealed by the atomic resolution crystal structure of a 6mer GNA duplex containing solely Watson-Crick type hydrogen-bonded base pairs.

Full text of DOI:10.1039/b916851f

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Atomic resolution duplex structure of the simplified nucleic acid GNAw
Mark K. Schlegel, Lars-Oliver Essen* and Eric Meggers*
Received (in Cambridge, UK) 14th August 2009, Accepted 25th November 2009
First published as an Advance Article on the web 21st December 2009
DOI: 10.1039/b916851f
Double helix variations of glycol nucleic acids (GNA) are
revealed by the atomic resolution crystal structure of a 6mer
GNA duplex containing solely Watson–Crick type hydrogen-
bonded base pairs.
hydrogen bonding. We have determined the atomic resolution
structure of a (S)-GNA duplex formed from the self-
complementary strand 30-GBrCGCGC-20 by X-ray crystallo-
graphy. The GNA nucleotide BrC is composed of the
5-bromocytosine nucleobase which establishes a hydrogen-
bonded base pair with guanine in the same fashion as a
cytosine nucleotide with little or no structural pertubation.6
The phosphoramidite 1 was synthesized from 5-bromo-
cytosine in just three steps with an overall yield of 45%
(Scheme 1). For this, 5-bromocytosine was reacted with
(S)-glycidyl 4,40-dimethoxytrityl ether 2 in the presence of
0.2 equivalents of NaH in DMF to afford the nucleoside 3
in yields of 60%. The exocyclic amino group was subsequently
protected as a dimethylformamidine group by reacting 3 with
dimethylformamide dimethylacetal in MeOH at 50 1C to
afford 4 in nearly quantitative yield. Conversion to the
phosphoramidite 1 was then accomplished in 77% yield by
We recently discovered the remarkable duplex formation
properties of glycol nucleic acids (GNA).1–5 Bearing an acyclic
backbone of propylene glycol nucleosides that are connected
by phosphodiester bonds, GNA constitutes the most simplified
solution for a phosphodiester-bond-containing nucleic acid
(Fig. 1). Nevertheless, the thermal stabilities of GNA duplexes
significantly surpass the stabilities of analogous duplexes of
DNA and RNA. The ability to form antiparallel
Watson–Crick type double helices was confirmed by an
X-ray crystal structure of a (S)-GNA duplex from a self-
complementary strand 30-CGHATHCG-20, in which H denotes
an artificial hydroxypyridone nucleobase which forms
a
reaction
with
2-cyanoethyl
N,N,N0,N0-tetraisopropyl
copper(II)-mediated homobase pair.3,4 This metallo-base pair
was utilized as a necessary handle to site-selectively introduce
two heavy atoms per duplex for phasing the crystallographic
data. However, it is uncertain how representative such a metallo-
GNA duplex is, since it is unknown to what extent the artificial
metallo-base pair influences the global GNA duplex structure.
Here we report the first crystal structure of a (S)-GNA
duplex composed entirely of base pairs with Watson–Crick
phosphordiamidite and substoichiometric amounts of
4,5-dicyanoimidazole in CH2Cl2.7 The self-complementary
6mer (S)-GNA strand 30-GBrCGCGC-20 was synthesized
and purified according to published standard procedures.2,7
Crystals diffracting up to 0.965 A resolution were obtained
from a sample of the purified oligonucleotide using the sitting
drop vapor diffusion method and 2-methyl-2,4-pentanediol as
precipitant. The asymmetric unit of the refined structure
30-GBrCGCGC-20 contains one single strand, 58 H2O, one
Mg2+, and two [Co(NH3)6]3+ cations. The atomic resolution
of this structure allows us for the first time to determine
Fig. 1 Constitution of (S)-GNA oligonucleotides and the base pairs
used in this study.
Fachbereich Chemie, Philipps-Universitat Marburg,
¨
Hans-Meerwein-Strasse, 35032 Marburg, Germany.
E-mail: essen@chemie.uni-marburg.de, meggers@chemie.uni-marburg.de
w Electronic supplementary information (ESI) available: Synthetic
procedures with analytical data, experimental details regarding
GNA crystallography, backbone torsional angles and phosphate bond
angles, analysis of GNA duplex parameters. See DOI: 10.1039/
b916851f
Scheme 1 Synthesis of the phosphoramidite 1 for automated solid
phase synthesis of BrC-containing GNA strands. DCI = 4,5-dicyano-
imidazole.
ꢀc
This journal is The Royal Society of Chemistry 2010
1094 | Chem. Commun., 2010, 46, 1094–1096

View Article Online
Table 1 Crystallographic data and refinement statistics
accurate bond lengths, torsional angles, and phosphate bond
angles (see ESIw). Crystallographic data and refinement
statistics are given in Table 1. Fig. 2 visualizes the atomic
resolution with the electron density of a dinucleotide unit
contoured at 1.5s.
Data collection statistics
Wavelength/A
Resolution/A
Space group
Completeness
0.9183
21.0–0.965
C2221
0.982 (0.914)
4.0
0.059 (0.132)
17.7 (7.2)
4.6
The self-complementary (S)-GNA strand forms an anti-
parallel duplex with Watson–Crick base pairing and coaxial
stacking of individual duplexes, resulting in a quasi-continuous
helix within the crystal lattice (Fig. 3, see also ESIw). All base
pairs form standard Watson–Crick hydrogen bonding
patterns with the 5-bromocytosine nucleotide appearing to
have little, if any, distorting effect (Fig. 4).8 The distances
between the C10 carbons of opposing nucleotides range from
10.71 to 10.85 A which is in agreement with average values
found in DNA duplexes (10.85 A). A striking common feature
of (S)-GNA and the previous metallo-(S)-GNA duplex is the
large average slide of around 3.4 A between neighboring base
pairs which is due to a large backbone–base inclination
ranging in this new duplex from ꢁ461 to ꢁ531 (Table 2).9
This results in extensive interstrand and at the same time
reduced intrastrand base–base stacking interactions. The latter
are apparently compensated by intrastrand backbone–base
hydrophobic interactions, in particular the stacking of the
C10-methylene groups of the propylene glycol backbone
against neighboring nucleobases. Another common aspect
with the metallo-(S)-GNA is the formation of a helical ribbon
by the (S)-GNA that is loosely wrapped around the central
helix axis, displaying just one large groove (Fig. 3 and 4).
Unlike the previously reported metallo-(S)-GNA double
helix (from now on referred to as Type M),3 the structure of
the continuous right-handed (S)-GNA double helix (referred
to as Type N), that can be derived from its 6mer structure by
repetitive superimposition, shows two significant differences:
first of all, a comparison of the two (S)-GNA helices in
Fig. 3 demonstrates that the new Type N helix is considerably
compressed along the z-axis relative to the Type M helix.
Although this compression is accompanied by very little
change in the helix diameter, it causes the Type N helix to
adopt a significantly shorter helical pitch of 26 A with
10 residues per turn, in comparison to 60 A with 16 residues
Redundancy
Rmerge
a,b
I/s(I)a,c
Bwilson/A2
Refinement statistics
Resolution/A
8.00–0.965
0.105/0.127
6.69
58
0.006
1.622
d
R-factor/Rfree
Average B-factor/A2
Water molecules
Rmsd bonds/A
Rmsd angles/A
a
Values in parentheses correspond to the highest resolution shell.
|Ij(h) ꢁ hI(h)i|/
PP
PP
b
c
Ij(h). As calculated with the
Rmerge
program SCALA. R =
=
P
P
d
JFo| ꢁ k|FcJ/ |Fo| with k as scaling
factor; Rfree calculated with test set (7% of all data). The coordinates
and structure factors are deposited in the RCSB under accession
code 2WNA.
Fig. 2 SIGMAA-weighted 2Fobs ꢁ Fcalc electron density of the
30-terminal glycol nucleotides contoured at 1.5s.
Fig. 3 Comparison of the (S)-GNA overall duplex structures of 30-GBrCGCGC-20 (Type N) and the metallo-GNA duplex 30-CGHATHCG-20
(Type M). The continuous model of the Type N helix was derived by repetitively superimposing the ends of individual 6mer GNA duplexes onto
each other and it differs slightly from the coaxially stacked 6mer duplexes in the crystal lattice (see ESIw for more details). Single duplexes are
shown in red sticks. Right: view along the helix axis. Images generated using PyMOL.
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 1094–1096 | 1095

View Article Online
Interestingly, the crystal structure of a DNA-fragment with
the same sequence, 50-CGCGCG-30, demonstrated the
existence of the Z-DNA.13 Like the latter, the repeat unit of
Type N GNA-helices also comprises two nucleotides with
alternating backbone conformations suggesting an influence
of GC-rich sequences on GNA backbone conformations.
However, unlike the Z-DNA, the interconversion between
Type M and N helices of GNA should not require extensive
conformational changes of the Watson–Crick-like base pairs.
In conclusion, we have presented the first atomic resolution
crystal structure of a (S)-GNA duplex containing solely
Watson–Crick type hydrogen-bonded base pairs. This new
GNA double helix structure reveals common features but also
significant differences compared to the previously reported
metallo-GNA duplex structure and gives insight into the
degree of variation within the family of GNA duplex
structures. The main contributor of this variation is the ability
of the propylene glycol nucleotides to choose between a gauche
or anti conformation around the vicinal C–O bonds and it can
be hypothesized that this may vary with the duplex sequence.
The atomic resolution of this structure allows us to accurately
determine bond and angle parameters and will serve as the
basis for solving and refining additional GNA duplex
structures of different sequences and length in order to gain
further insight into the exceptional duplex formation abilities
of this minimal nucleic acid backbone.
Fig. 4 Structure of a single (S)-GNA duplex from 30-GBrCGCGC-20.
Gauche and anti refer to the torsional angles between the vicinal
C–O bonds C20–O and C30–O. Generated using PyMOL.
Table 2 Comparison of average helical parameters for (S)-GNA,
B-DNA, and A-DNA
(S)-GNA (S)-GNA
Type Ma Type Na
B-DNAa
A-DNAa
Helical sense
Residues per turn
Helical pitch/A
Helical rise/A
x-Displacement/A
Tiltb/1
Right
16
60
Right
10
26
Right
10
34
3.4
0.1
0.1
0.6
36.0
0.2
7.0
Right
12
34
Notes and references
3.8
ꢁ7.0
2.6
ꢁ6.0
2.9
ꢁ4.2
ꢁ0.1
8.0
31.0
ꢁ1.5
5.9
1 L. Zhang, A. Peritz and E. Meggers, J. Am. Chem. Soc., 2005, 127,
4174.
2 M. K. Schlegel, A. E. Peritz, K. Kittigowittana, L. Zhang and
E. Meggers, ChemBioChem, 2007, 8, 927.
3 M. K. Schlegel, L.-O. Essen and E. Meggers, J. Am. Chem. Soc.,
2008, 130, 8158.
0.0
0.5
6.4
Rollb/1
ꢁ2.7
23.5
ꢁ3.5
5.4
Twistb/1
35.7
ꢁ3.4
5.4
Slideb/1
P–P distancec/A
a
4 M. K. Schlegel, L. Zhang, N. Pagano and E. Meggers, Org.
Biomol. Chem., 2009, 7, 476.
5 M. Schlegel, X. Xie, L. Zhang and E. Meggers, Angew. Chem., Int.
Ed., 2009, 48, 960.
Data for GNA were calculated using the program CURVES
(ref. 12). Data for B-DNA and A-DNA were taken from ref. 10 and
b
11. Local base pair step parameters. Intrastrand P–P distances.
c
6 For the use of 5-bromocytosine in nucleic acid crystallography, see
for example: D. B. Tippin and M. Sundaralingam, J. Mol. Biol.,
1997, 267, 1171; M. Ortiz-Lombardia, A. Gonzalez, R. Eritja,
J. Aymami, F. Azorin and M. Coll, Nat. Struct. Biol., 1999, 6, 913.
7 For DMF-protected GNA phosphoramidites, see also:
M. K. Schlegel and E. Meggers, J. Org. Chem., 2009, 74, 4615.
8 Short intermolecular Brꢂ ꢂ ꢂBr contacts at 3.96 A suggest that the
bromine substituents appear to have an effect on the packing of the
individual duplexes in the crystal. See ESIw.
9 Such a strong backbone–base inclination has also been observed
for unnatural hexose nucleic acids. See: R. Declercq, A.
Van Aerschot, R. J. Read, P. Herdewijn and L. Van Meervelt,
J. Am. Chem. Soc., 2002, 124, 928; M. Egli, P. S. Pallan, R.
Pattanayek, C. J. Wilds, P. Lubini, G. Minasov, M. Dobler, C. J.
Leumann and A. Eschenmoser, J. Am. Chem. Soc., 2006, 128, 10847.
10 R. E. Dickerson, Methods Enzymol., 1992, 211, 67.
11 W. K. Olson, M. Bansal, S. K. Burley, R. E. Dickerson,
M. Gerstein, S. C. Harvey, U. Heinemann, X.-J. Lu, S. Neidle,
Z. Shakked, H. Sklenar, M. Suzuki, C.-S. Tung, E. Westhof,
C. Wolberger and H. M. Berman, J. Mol. Biol., 2001, 313, 229.
12 R. Lavery and H. Sklenar, J. Biomol. Struct. Dyn., 1988, 6, 63;
R. Lavery and H. Sklenar, J. Biomol. Struct. Dyn., 1989, 6, 655.
13 A. H.-J. Wang, G. J. Quigley, F. J. Kolpak, J. L. Crawford,
J. H. van Boom, G. van der Marel and A. Rich, Nature, 1979,
282, 680.
per turn for the Type M helix (Table 2).10–12 This shorter
helical pitch results from the much stronger twist in the new
Type N helix which subsequently brings the phosphate groups
of opposing strands in closer contact. The base pairs of the
Type N helix are displaced from the helix axis (x-displacement)
by 5.4 to 6.8 A resulting in a large hollow core, similar to the
previous metallo-GNA duplex (Type M). However, the hollow
core is circular rather than elliptical as observed in the Type M
helix for which there is a greater variance in x-displacement
values from the central helix axis (5.1–8.6 A).
The second difference compared to Type M helices concerns
the conformation of the propylene glycol backbone, especially
the torsional angles between the vicinal C30–O and C20–O bonds.
In the new Type N structure the nucleotides adopt strictly
alternating gauche and anti conformations with average torsional
angles g of ꢁ661 and ꢁ1741, respectively (Fig. 4). Apparently, the
acyclic backbone has the flexibility to adopt these two distinct
conformations and can vary them within a single duplex.
ꢀc
This journal is The Royal Society of Chemistry 2010
1096 | Chem. Commun., 2010, 46, 1094–1096