1′,5′-Anhydro-L-ribo-hexitol Adenine Nucleic Acids (α-L-HNA-A): Synthesis and Chiral Selection Properties in the Mirror Image World

Article abstract of DOI:10.1021/acs.joc.5b00406

The synthesis and a preliminary investigation of the base pairing properties of (6′ → 4′)-linked 1′,5′-anhydro-L-ribo-hexitol nucleic acids (α-L-HNA) have herein been reported through the study of a model oligoadenylate system in the mirror image world. D

Full text of DOI:10.1021/acs.joc.5b00406

Article
pubs.acs.org/joc
1′,5′-Anhydro‑L-ribo-hexitol Adenine Nucleic Acids (α‑L‑HNA-A):
Synthesis and Chiral Selection Properties in the Mirror Image World
Daniele D’Alonzo,*,† Mathy Froeyen,‡ Guy Schepers,‡ Giovanni Di Fabio,† Arthur Van Aerschot,‡
Piet Herdewijn,‡ Giovanni Palumbo,† and Annalisa Guaragna†
†Dipartimento di Scienze Chimiche, Universita
̀
degli Studi di Napoli Federico II, via Cintia 21, 80126 Napoli, Italy
‡Laboratory for Medicinal Chemistry, Rega Institute for Medical Research, KU Leuven, Minderbroedersstraat 10, 3000 Leuven,
Belgium
S
* Supporting Information
ABSTRACT: The synthesis and a preliminary investigation of the
base pairing properties of (6′ → 4′)-linked 1′,5′-anhydro-L-ribo-
hexitol nucleic acids (α-L-HNA) have herein been reported through
the study of a model oligoadenylate system in the mirror image
world. Despite its considerable preorganization due to the rigidity of
the “all equatorial” pyranyl sugar backbone, α-L-HNA represents a
versatile informational biopolymer, in view of its capability to cross-
communicate with natural and unnatural complements in both
enantiomeric forms. This seems the result of an inherent flexibility
of the oligonucleotide system, as witnessed by the singular
formation of iso- and heterochiral associations composed of regular,
enantiomorphic helical structures. The peculiar properties of α-L-
HNA (and most generally of the α-HNA system) provide new
elements in our understanding of the structural prerequisites ruling the stereoselectivity of the hybridization processes of nucleic
acids.
INTRODUCTION
■
Over the last few decades, the propensity for base pairing
through Watson−Crick or any other recognition modehas
been established not to be restricted only to natural nucleic
acids, but has been found to be rather widespread among
alternative genetic polymers recruited from the structural
neighborhood of DNA and RNA.1,2 Particularly, oligonucleo-
tide systems involving replacement of the native (deoxy)-
ribofuranose core with either preorganized3,4 or flexible5,6 sugar
units have often been demonstrated to possess superior self-
and cross-pairing properties, thereby representing excellent
candidates for biomedical purposes,3,5 inspiring etiological
investigations on life’s origin6,7 and opening up a new world
for information storage, propagation, and evolution.8 In this
area, the family of stereoisomeric six-membered oligonucleo-
tides composed of (6′ → 4′)-linked 1′,5′-anhydrohexitol
Figure 1. Stereoisomeric oligonucleotides belonging to the HNA
family.
nucleotides (HNA, Figure 1) has been identified as one of
the most powerful and versatile classes of synthetic informa-
tional polymers discovered to date.4 The earliest and most
prominent member of this family is the oligonucleotide system
with D-arabino configuration (β-D-HNA).9,10 β-D-HNA has
been demonstrated to possess a remarkable hybridization
aptitude toward natural complements, acting as an A-type
(RNA) mimic2b (Figure 1). Besides its manifest potential as a
therapeutic agent (e.g., as antisense,11 siRNA,12 and aptamer
probes13) and as a diagnostic tool,14 the β-D-HNA system was
well suited for templated synthesis, under both chemical15 and
enzymatic16 conditions, even in vivo,17 acting both as substrate
for oligonucleotide synthesis and as template for reverse
transcription.18 Very recently, HNA-based enzymes (HNA-
zymes) have been synthesized and found capable to display
RNA endonuclease activity.19
Received: February 20, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.5b00406
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Among β-D-HNA stereoisomers, its C2′ epimer (i.e., the
hexitol nucleic acid having D-ribo configuration: α-D-HNA) also
was demonstrated to form stable complexes with RNA, with its
conformationally constrained analogues (β-HNA, CNA) and
other alternative oligonucleotide systems (e.g., β-homoDNA)
having the same sugar chirality.20 Different from β-D-HNA, its
optical antipode (1′,5′-anhydro-L-arabino-hexitol nucleic acids:
β-L-HNA), designed to potentially act as a conformationally
constrained oligonucleotide spiegelmer,21 was not able to pair
with natural complements.22 On the other hand, it exhibited
the singular property of pairing with β-HNA complements
(among other artificial oligonucleotide systems) belonging to
both (same and opposite) senses of chirality.22 This peculiarity,
so far restricted only to a limited number of six-membered
nucleic acids,23−25 has inspired etiology-oriented studies aimed
at elucidating the existing relationship between sugar structure
and chiral selection properties of nucleic acids.25−27 Intrigued
by the remarkable hybridization potential of the HNA family,
we have herein widened the repertoire of HNA stereoisomers
by exploring the base pairing properties of a novel
oligonucleotide system composed of (6′ → 4′)-linked 1′,5′-
anhydro-L-ribo-hexitol nucleotides (α-L-HNA; Figure 1). Our
interest in α-L-HNA comes from its potential pairing versatility,
lying in its capability to reproduce the structure of two
oligonucleotide systems having opposite hybridization profiles.
Indeed, because of the “all-equatorial” pyranyl sugar unit in the
backbone, α-L-HNA is expected to hold structural and
conformational similarities with the six-membered oligonucleo-
tide system composed of (6′ → 4′)-linked β-L-erythro-
hexopyranosyl nucleotides (β-L-homoDNA; Figure 2a). The
structures of the corresponding adenine nucleosides, aligned
to show a fairly close three-dimensional fit (Figure 2c).
The synthesis and an early evaluation of the hybridization
properties of α-L-HNA [(oligo)adenylates being preliminarily
considered] to natural and unnatural complements were carried
out through studies in the mirror image world.29 Accordingly,
in place of a direct evaluation of the hybridization potential of
α-L-HNA by annealing experiments with D-oligonucleotides
(e.g., D-DNA and D-RNA), we instead conceived to perform
hybridization studies of its mirror image (α-D-HNA) with
complementary sequences of L-oligonucleotides (e.g., commer-
cially available L-DNA and L-RNA). Exploiting the principles of
chirality and enantiomeric recognition of nucleic acids, this
evaluation was expected to indirectly provide information on
the hybridization properties of our target α-L-HNA system.
RESULTS AND DISCUSSION
■
Nucleoside synthesis. An expeditious route to the 1′,5′-
anhydro-D-ribo-hexitol nucleoside 13 to be used as building
block for oligonucleotide synthesis was carried out as depicted
in Schemes 1 and 2. Compared to previous synthetic
Scheme 1. Synthesis of D-ribo-Hexitol Nucleosides. Part I:
Preparation of Alcohol Intermediate 5
approaches devised to the same end,20,30 an alternative path
was herein envisioned starting from commercially available
3,4,6-tri-O-acetyl-D-glucal (1). Rearrangement of 1 under
modified31 Ferrier conditions (Et3SiH/I2) yielded the corre-
sponding 2,3-unsaturated sugar 2 (90%). Then, after protective
group replacement [MeONa, then NaH/BnBr (90% over two
steps) or PhCHO/PTSA (88% over two steps)], regio- and
stereoselective double bond hydration of 4a−b was achieved
using classical oxymercuriation/demercuriation conditions
[Hg(OAc)2, then H2O/NaBH4] (Scheme 1). In line with
previous data,32 treatment of 4a with Hg(OAc)2 in a H2O/
THF mixture followed by one-pot addition of NaBH4 to the
reaction mixture yielded alcohol 5a as the only isomer in a good
83% yield. Under the same conditions, use of bis-benzyl ether
4b as starting material provided 5b with an even better yield
(92%), although traces of the regioisomer 6b were also
detected (5%).33
Figure 2. (a and b) Backbone similarities among α-L-HNA, β-D-
erythro-hexopyranosyl nucleic acids (β-D-homoDNA), and α-L-ribo-
pentofuranosyl locked nucleic acids (α-L-HNA). (c) Superimposition
of the energy-minimized structures (Hyperchem 8.0, MP3 algorithm)
of adenine nucleosides of α-L-HNA and α-L-LNA.
latter and its more prominent D-enantiomer (β-D-homoDNA)7
belong to a class of preorganized pairing systems (pyranosyl
nucleic acids) characterized by excellent self- and cross-
communication properties, albeit unable to transfer information
to natural nucleic acids. On the other hand, α-L-HNA is
conceived to act as a mimic of the locked nucleic acid with α-L-
ribo configuration, i.e. α-L-LNA (Figure 2b). The latter is a
highly efficient DNA- and RNA-targeting pairing system, in
turn acting as a DNA mimic because of the close resemblance
of its dioxabicyclo[2.2.1]heptane skeleton to the bioactive C2′-
endo sugar ring pucker of natural deoxynucleotides.28 The
structural analogy between the sugar units of α-L-LNA and α-L-
HNA is preliminarily suggested by the energy-minimized
Tosylation (TsCl/Py) of 5a−b (Scheme 2) led to tosyl
esters 7a−b (88−90%), which were then subjected to coupling
reaction with unprotected adenine (A/NaH). Disappointingly,
deoxyadenosine analogue 8a was obtained only in 25% yield,
while undesired olefins 4a and 9a were recovered as the main
B
DOI: 10.1021/acs.joc.5b00406
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 2. Synthesis of D-ribo-Hexitol Nucleosides. Part II:
Preparation of Amidite 13
Table 1. Thermal Stability Studies of ds-L-DNA, ds-L-RNA,
and L-DNA:L-RNA Duplexes Containing One or More D-
a
ribo-Hexitol Nucleotides
5′-L-d(T6XT6)-3′
Entry
X
Complement
Tm [°C]
ΔTm [°C]
−13.9
b
1
2
L-dA
D-αAH
L-DNA
29.6
b
L-DNA
15.7
5′-L-d(T5X3T5)-3′
Entry
X
Complement
Tm [°C]
ΔTm [°C]
b
3
4
L-dA
D-αAH
L-DNA
L-DNA
30.3
b,c
ND
5′-L-d(X3YT3ATZT2)-3′
Entry
X
Y
Z
Complement Tm [°C] ΔTm [°C]
5
L-dA
L-dA
D-αAH
L-dA
L-dA
D-αAH
L-dA
L-dA
D-αAH
L-dA
L-dA
D-αAH
L-dA
L-DNA
L-DNA
L-DNA
L-RNA
L-RNA
L-RNA
38.0
14.3
17.2
22.5
6
D-αAH
D-αAH
L-dA
−23.7
−20.8
7
8
c
9
D-αAH
D-αAH
ND
10
21.8
−0.7
5′-L-(X3YU3AUZU2)-3′
Entry
X
Y
Z
Complement Tm [°C] ΔTm [°C]
11
12
13
L-rA
L-rA
L-rA
L-RNA
L-RNA
L-RNA
34.7
10.0
26.5
L-rA
D-αAH
D-αAH
D-αAH
D-αAH
−24.7
−8.2
products. The same reaction, conducted starting from 7b,
provided nucleoside 8b with a slightly better yield (37%).
Standard nucleoside chemistry involving benzyl groups removal
of 8b (Pd/C, H2, 91%), chemoselective N-benzoylation of 10
(TMSCl/BzCl), regioselective 6′-O-monomethoxytritylation of
11 (MMTCl, 85%), and 4′-O-phosphitylation of 12 [(i-
Pr)2NP(Cl)OCH2CH2CN/DIPEA, 90%] eventually gave
phosphoramidite monomer 13 to be used for incorporation
studies.
Hybridization studies in the mirror image world.
Assembly of oligonucleotides went flawless, and analogous
yields were obtained in comparison to the assembly of natural
oligonucleotides and to fully modified L-DNA and L-RNA
sequences. Typically, yields of minimal 15 up to 50 OD260 were
obtained for all 1 to 1.5 μmol scale reactions (hand-filled
columns) following anion exchange HPLC purification and
desalting.
To preliminarily assess the pairing potential of α-L-HNA
toward natural complements, thermal denaturation studies in
the mirror image world between α-D-HNA-(D-αAH)-containing
oligomers and L-DNA/L-RNA complements were carried out
(Tables 1 and 2). In all cases, compared to unmodified ds-L-
DNA, ds-L-RNA, or L-DNA:L-RNA duplexes, incorporation of
D-hexitol nucleotides led to a reduced thermal stability of the
corresponding complexes (Table 1). Duplex destabilization was
especially apparent with DNA: incorporation of a single D-
hexitol nucleotide into a ds-L-DNA duplex led to a drop in the
Tm value of the corresponding complex of about 14 °C (Table
1, entries 1−2). Further insertions of contiguous D-hexitol
nucleotides caused a similar or even greater destabilization of
the system (entries 3−7). Internal incorporations of D-hexitol
nucleotides into L-DNA:L-RNA and ds-L-RNA duplexes
analogously produced highly unstable complexes (entries 9
and 12). On the other hand, incorporations at the side
positions provided a minor destabilization of the duplexes
(entries 10 and 13). As an example, four contiguous α-D-HNA
building blocks inserted into a L-DNA:L-RNA duplex did not
alter the thermodynamic stability of the latter (ΔTm: −0.7 °C)
(entries 8 and 10). A similar incorporation into a ds-L-RNA
L-rA
a
Melting points were determined at 260 nm in 1 M NaCl, 20 mM
KH2PO4 (pH 7.5), 0.1 mM EDTA (unless otherwise specified).
Experiments conducted in 0.1 M NaCl buffer. ND: no regular
b
c
transition detected.
Table 2. Thermal Stability Studies of Iso- and Heterochiral
Complexes Containing a Fully Modified α-D-HNA Strand [D-
a
(αAH)13]
b
Entry
Oligonucleotide
Complement
Tm [°C]
1
α-D-HNA
α-D-HNA
α-D-HNA
L-DNA
none
41.7
c
c
,
,
d
d
2
L-DNA
41.7
41.8
47.0
29.0
40.5
35.0
20.0
40.3
67.0
69.0
50.1
58.9
3
D-DNA
4
L-DNA
5
L-DNA
L-RNA
6
L-RNA
L-DNA
7
L-RNA
L-RNA
8
α-D-HNA
α-D-HNA
α-D-HNA
α-D-HNA
α-D-HNA
α-D-HNA
L-RNA
e
9
D-RNA
c
f
f
f
,
f
10
11
12
13
β-D-HNA
β-L-HNA
β-D-homoDNA
β-L-homoDNA
a
Melting points were determined at 260 nm in 1 M NaCl, 20 mM
b
KH2PO4 (pH 7.5), 0.1 mM EDTA (unless otherwise specified). 13-
c
mer oligoadenylates were used in all cases. Referred to an AH:αAH
α
d
association. The formation of hairpin structures was excluded by
concentration-dependent thermal denaturation studies (see the
Supporting Information, Figure S7). Taken from ref 20. Experiments
conducted in 0.1 M NaCl buffer.
e
f
duplex instead produced a less stable hybrid (ΔTm: −8.2 °C;
entries 11 and 13), although the latter still displayed a regular
sigmoidal denaturation curve. Considering the data in the
mirror image world, the apparent discrepancy of the data of
Table 1 could be likely overcome hypothesizing that α-L-HNA
holds a propensity for pairing with D-RNA, even though not
acting as a D-DNA mimic. Accordingly, internal incorporations
C
DOI: 10.1021/acs.joc.5b00406
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
of single modified nucleotides would strongly perturb the
regularity of the natural A/B-type duplexes, while side
incorporations of even several hexitol nucleotides would be
better tolerated.
experiments between α-D-HNA and D- and L-RNA, the stability
of the isochiral complex was higher than that of the
corresponding heterochiral association; on the other hand, α-
D-HNA formed complexes of about the same stability with the
more preorganized β-D-HNA (Tm 67 °C; entry 9) and β-L-
HNA (Tm 69 °C; entry 10); eventually, the stability of the
complex formed between α-D-HNA and the most rigid β-L-
homoDNA (Tm 58.9 °C, entry 12) was markedly higher than
that of the corresponding isochiral association (Tm 50.1 °C;
entry 11).
The data reported in Table 2 reveal a certain versatility in the
hybridization profile of the α-HNA system. Clues on the
geometry of α-D-HNA-containing associations (and reciprocally
likewise on that of the corresponding α-L-HNA associations)
were herein preliminarily provided by further studies of the
complexes with β-D- and β-L-HNA (Figures 4−6). CD analysis
Along this line, annealing experiments of the fully modified
α-D-HNA sequence D-(αAH)13 with DNA/RNA complements
in both enantiomeric forms were performed (Table 2). α-D-
HNA did not pair with D- or L-DNA, both at 0.1 and 1 M NaCl;
α
the formation of an intermolecular AH:αAH association (Tm
41.7 °C) was instead detected in both cases (entries 1−3). On
the other hand, regular sigmoidal curves were clearly observed
after annealing of D-(αAH)13 with D- and L-RNA complements
(entries 8−9). Looking at the mirror image world, it is worth
mentioning that, to the best of our knowledge, this represents
the first case of an oligonucleotide system with an L-hexose
derivative in the backbone displaying hybridization with natural
complements. However, the thermodynamic stability of the
complex was very low (Tm 20 °C) and lower than that of the
corresponding unmodified DNA:RNA duplex (Tm 29 °C; entry
5). Unexpected but not unprecedented,25 no trace of the
thermodynamically more stable αAH:αAH complex (Tm 41.8 °C)
was detected in this case.34
A comparative CD analysis of iso- and heterochiral hybrids
was also carried out (Figure 3). The differences between the
Figure 4. CD spectra of α-D-HNA + β-D-HNA [D-(αAH)13
+ D-
(βTH)13] () and α-D-HNA + β-L-HNA [D-(αAH)13 + L-(βTH)13] (−
−). All measurements were taken at 0 °C in 0.1 M NaCl, 20 mM
KH2PO4, 0.1 mM EDTA (pH 7.5).
surprisingly displayed that, in spite of their diastereoisomeric
relationship, α-D-HNA:β-D-HNA and α-D-HNA:β-L-HNA
exhibited nearly perfect mirror image profiles (Figure 4).
Indeed, the CD spectrum of the α-D-HNA:β-D-HNA complex
[D-(αAH)13:D-(βTH)13] had a positive Cotton effect with a
maximum of absorption at 275 nm and a minimum at 253 nm;
conversely, the negative Cotton effect of the CD spectrum of
the α-D-HNA:β-L-HNA complex [D-(αAH)13:L-(βTH)13] in-
volved a minimum of absorption at 269 nm and a maximum
at 251 nm.
MD studies involving the construction of duplex models of
D-(αAH)13:D-(βTH)13 and D-(αAH)13:L-(βTH)13 were then con-
ducted. Early simulations aimed at the development of either
quasi-linear or helical models through standard Watson−Crick
base pairing gave weak and irregular structures; on the other
hand, the α-D-HNA strand was found to form stable
associations with β-D-HNA and β-L-HNA complements
through reverse-Hoogsteen base pairing,36 respectively appear-
ing as smooth right- and left-handed double helices (Figure 5).
Considering the high rigidity of the pyranyl sugar backbone, the
formation of helical structures in both iso- and heterochiral
associations is an unexpected result which has been
documented only seldom before.24 Both duplexes displayed
antiparallel orientation, with a predominance for interstrand
over intrastrand base stacking. As a consequence of the reverse-
Hoogsteen base pairing, in both duplexes one of the grooves is
Figure 3. CD spectra of ds-α-D-HNA [ds-D-(αAH)13] (), α-D-HNA +
L-RNA [D-(αAH)13 + L-(rU)13] (···), and L-DNA + L-RNA [L-(dA)13
+
L-(rU)13] (−··−). All measurements were taken at 0 °C in 1 M NaCl,
20 mM KH2PO4, 0.1 mM EDTA (pH 7.5).
CD profiles of ds-α-D-HNA [ds-D-(αAH)13] and α-D-HNA:L-
RNA [D-(αAH)13:L-rU13] confirmed the occurred formation of
an heterochiral interaction. Likewise, the even larger differences
between the latter and the “natural” L-DNA:L-RNA duplex (L-
dA13:L-rU13) (including opposite handednesses) clearly in-
dicated the deep structural diversity between the two structures,
and therefore, differently from α-L-LNA, the limited ability by
the α-L-HNA system to act as a D-DNA mimic.35
In a second set of experiments, the stereoselectivity of the
hybridization processes of α-L-HNA was studied through
annealing experiments in the mirror image world involving D-
(αAH)13 and unnatural D- and L-configured complements (Table
2). As already found for similar substrates,22,25 α-D-HNA
displayed excellent hybridization capacity with β-HNA (Figure
1) and β-homoDNA (Figure 2) complements in both
enantiomeric forms. Clearer than in previous studies,25 the
thermodynamic preference for the heterochiral complexes
increased gradually with the sugar backbone rigidity of the
oligonucleotide partners. Indeed, while in the annealing
D
DOI: 10.1021/acs.joc.5b00406
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(average χ = −38°), while in the heterochiral α-D-HNA:β-L-
HNA complex [D-(αAH)13:L-(βTH)13] the same nucleotides
adopted standard anti conformations around the glycosidic
bonds (average χ = 117°). This represented the most striking
difference in the conformational profile of D-(αAH)13 when
involved in right- and left-handed duplexes, although other
variations, especially in the backbone torsion angles α, β, and ε,
also occurred (Table 3).
Larger differences in the backbone angles (especially γ, δ, ε,
and ζ) were instead detected between enantiomeric β-HNA
nucleotides. This was mainly the result of a sugar chair
inversion24 (1C4 → C1) of the L-arabino-hexitol nucleotides
4
(average δ = −159°) (Figure 6b); conversely, most D-
Figure 5. Left: side view of the simulated α-D-HNA:β-D-HNA duplex
[D-(αAH)13:D-(βTH)13]. Right: side view of the simulated α-D-HNA:β-L-
HNA duplex [D-(αAH)13:L-(βTH)13]. Yellow: carbon atoms in the D-
(αAH)13 strand; green: carbon atoms in the D-(βTH)13 and L-(βTH)13
strands. Both duplexes involve reverse-Hoogsteen base pairing. In both
duplexes, one of the terminal Hoogsteen base pairings is broken.
Images are generated using Chimera.
Figure 6. (a) Close-up view of the reverse-Hoogsteen base pair in D-
(αAH)13:D-(βTH)13. The D-ribo-hexitol nucleotides (yellow) adopt a 4C1
chair and a syn conformation around the glycosidic bond (average χ =
4
−38°), while the D-arabino-hexitol nucleotides (green) adopt a C1
chair (axially oriented thymine; excluded T8, adopting a 1C4 chair) and
an anti conformation around the glycosidic bond (average χ = −139°).
(b) Close-up view of the reverse-Hoogsteen base pair in D-(αAH)13:L-
(βTH)13. The D-ribo-hexitol nucleotides (yellow) adopt a 4C1 chair and
an anti conformation around the glycosidic bond (average χ = 117°),
while the L-arabino-hexitol nucleotides (green) adopt a 4C1 chair
(equatorially oriented thymine) and an anti conformation around the
glycosidic bond (average χ = 134).
hydrophobic (C8 atom of A and C5−C7 atoms in T) while the
other groove (exposing N6 of A and O2 in T) is more polar.
Remarkable observations were also drawn by the analysis of
the backbone torsion angles (Table 3). Both α-D-HNA and β-
HNA strands exhibited α−ζ values only faintly similar to A-
DNA, while they largely differed from those of B-DNA. In
addition, α-D-HNA and β-D- and β-L-HNA strands displayed a
much wider variability in the torsion angles than those of (A-
and B-)DNA. This feature, required to enable suitable
accommodation of the complementary strands regardless of
their sugar chirality, is a strong indication of a certain flexibility
of hexitol nucleotides within the oligonucleotide strands. A
comparative analysis of the torsions in the two complexes
highlighted that in the isochiral α-D-HNA:β-D-HNA duplex [D-
(αAH)13:D-(βTH)13] the adenine moieties of the D-ribo-hexitol
nucleotides were arranged into unexpected syn conformations
4
enantiomers were in the expected C1 form (Figure 6a), with
the exception being T8, which adopts a 1C4 sugar chair (e.g., δ
= 154°, very close to the value observed in the α-
homoDNA:RNA duplex38) (Table 3). Substantially in agree-
ment with CD analysis, despite such deep conformational
differences, the two left- and right-handed duplexes overall
exhibited mirror-image shapes (Figure 5), with the twist values
being −18° and 22° (see the Supporting Information, Table
S3).
Table 3. Average Torsion Angles for Natural DNA Strands in A- and B-Form ds-DNA as Well as for α-D-HNA, β-D-HNA, and β-
a
L-HNA Strands in α-D-HNA:β-D-HNA [D-(αAH)13:D-(βTH)13] and α-D-HNA:β-L-HNA [D-(αAH)13:L-(βTH)13] Duplexes
α-D-HNA:β-D-HNA
α-D-HNA β-D-HNA
α-D-HNA:β-L-HNA
α-D-HNA β-L-HNA
−sc
+sc,+ac/+ap
b
b
DNA (A-form)
DNA (B-form)
α
β
γ
−sc
+ap
−sc/−sp
+ac
−sc,+sc
−sc
−sc,+sc
c
+ap/−ap
+sc,+ap
+ap/−ap
+sc
+sc,+ap/−ap
+ap/−ap
−ap
+sc
+sc
+sc,+ap/−ap
+sc
c
δ
ε
ζ
χ
+sc
+ac
+sc
+sc
c
−ac/−ap
−sc
anti
−ac
−ap
anti
+sc,−ap
−ap
−sc
−ap
−sc
anti
+sc/+ac
+sc,+ap/−ap
anti
c
+sc,−sc
syn
anti
a
The forward slash is used to indicate a continuous range in which the backbone angle values are included; the comma is used to indicate that the
b
c
backbone angle values are clustered in two nonadjacent ranges. Taken from ref 37. A large deviation from this value was observed in T8 due to
sugar chair inversion (4C1 → C4).
1
E
DOI: 10.1021/acs.joc.5b00406
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
under reduced pressure. NaH (0.88 g, 0.02 mol, 60% dispersion in
mineral oil) was added to a solution of crude 3 in anhydrous DMF (15
mL) at 0 °C. After 0.5h, BnBr (5.83 mL, 0.05 mol) was added in one
portion to the reaction mixture. The resulting solution was warmed to
rt and stirred at the same temperature for 16h. The reaction was then
quenched with MeOH (3 mL), the solution was washed with aqueous
NH4Cl and extracted with AcOEt. The combined organic phases were
dried (Na2SO4) and the volatiles removed under reduced pressure.
Chromatography of the crude residue over silica gel (hexane:ethyl
acetate, 100:0 to 80:20) provided the pure benzyl ether 4b (3.08 g,
CONCLUSIONS
■
The synthesis and a preliminary analysis of the pairing
properties of a novel member of the HNA family composed
of (6′ → 4′)-linked 1′,5′-anhydro-L-ribo-hexitol nucleotides (α-
L-HNA) have been reported through studies of a model
oligoadenylate system in the mirror image world. Despite the
considerable structural preorganization provided by the rigid
“all equatorial” pyranyl sugar backbone, α-L-HNA (and most
generally the α-HNA system) represents a truly versatile
informational biopolymer, given its capability to pair with either
natural or unnatural complements, regardless of the structure,
conformational features, and sugar chirality of the oligonucleo-
tide partners. This is unexpectedly the result, inter alia, of a
certain degree of flexibility displayed by the nucleotide units in
the oligonucleotide strand. As a case study, the iso- and
heterochiral complexes deriving from the hybridization of a α-
D-HNA oligoadenylate with β-HNA oligothymidylate enan-
tiomers remarkably adopted regular, enantiomorphic double
helical structures. As suggested by MD studies, the suitable
accommodation of the complementary α-D-HNA and β-(D- or
L-)HNA strands involved cooperative structural and conforma-
tional effects, including the unexpected formation of reverse-
Hoogeteen base pairing, the syn nucleobase arrangement in α-
D-HNA nucleotides, and the conformational change (1C4 →
4C1) of the β-L-HNA sugar units. The peculiar hybridization
profile of α-(D- and L-)HNA has main relevance in the search
for the structural prerequisites ruling the stereoselectivity of the
hybridization processes. Herein, the observation on which the
increase of the sugar core rigidity of nucleic acids (obtained
replacing the natural furanose with six-membered rings) leads
to a reduction in the capacity to discriminate between
enantiomeric oligonucleotide partners finds in α-HNA an
illustrative example. In-depth and more comprehensive studies
aimed at identifying the structural factors and geometrical
parameters affecting the chiral selection properties of nucleic
acids are currently ongoing and will be published in due course.
1
90% yield) as a colorless oil. H NMR (CDCl3, 500 MHz): δ 3.59
(ddd, J = 2.1, 5.5, 8.2, 1H), 3.63 (dd, J = 5.3, 10.3, 1H), 3.71 (dd, J =
2.0, 10.3, 1H), 4.04 (bd, J = 8.2, 1H), 4.14−4.24 (m, 2H), 4.44 (d, J =
11.4, 1H), 4.54 (d, J = 12.3, 1H), 4.59 (d, J = 11.4, 1H), 4.60 (d, J =
12.3, 1H), 5.85 (bd, J = 10.5, 1H), 5.92 (bd, J = 10.5, 1H), 7.20−7.38
(m, 10H). 13C NMR (CDCl3, 125 MHz): ppm 65.6, 69.5, 70.4, 71.1,
73.5, 76.3, 125.5, 127.6, 127.7, 127.9, 128.2, 128.3, 128.4, 138.1, 138.2.
Elemental analysis calcd (%) for C20H22O3: C 77.39, H 7.14; found: C
77.46, H 7.12.
4,6-Di-O-benzyl-3-deoxy-1,5-anhydro-D-arabino-hexitol (5b).
THF (51 mL) was added to a stirred solution of Hg(OAc)2 (4.09 g,
9.6 mmol) in H2O (96 mL). After 15 min, a solution of olefin 4b (3.0
g, 9.6 mmol) in THF (45 mL) was added to the reaction mixture. The
resulting yellow solution was stirred for 16h at rt. Afterward, aqueous
NaBH4 was carefully added at 0 °C. Temperature was then raised
again to rt, and the resulting gray suspension was stirred at the same
temperature for 2h. The mixture was then washed with brine and
extracted with AcOEt. The organic phase was dried (Na2SO4) and
evaporated under reduced pressure. Chromatography of the crude
residue over silica gel (hexane:ethyl acetate, 9:1 to 6:4) provided the
pure alcohol 5b (2.92 g, 92% yield) as a colorless oil. 1H NMR
(CDCl3, 500 MHz): δ 1.55 (ddd, J = 2.8, 11.3, 13.2, 1H), 2.08 (bs, 1H,
OH), 2.45 (dddd, J = 2.9, 3.0, 5.1, 13.2, 1H), 3.40, (ddd, J = 2.0, 5.4,
9.5, 1H), 3.57 (d, J = 12.2, 1H), 3.66 (dd, J = 10.5, 5.4, 1H), 3.71−3.79
(m, 2H), 3.89 (d, J = 12.2, 1H), 3.98 (bs, 1H), 4.38 (d, J = 11.4, 1H),
4.54 (d, J = 12.1, 1H), 4.55 (d, J = 11.4, 1H), 4.60 (d, J = 12.1, 1H),
7.18−7.36 (m, 10H). 13C NMR (CDCl3, 125 MHz): ppm 36.1, 66.8,
69.6, 69.9, 71.1, 72.2, 73.6, 80.7, 127.6, 127.7, 127.9, 128.3, 128.4,
138.1. Elemental analysis calcd (%) for C20H24O4: C 73.15, H 7.37;
found: C 73.09, H 7.34.
4,6-Di-O-benzyl-3-deoxy-2-O-p-toluenesulfonyl-1,5-anhydro-D-
arabino-hexitol (7b). A solution of alcohol 5b (2.4 g, 7.70 mmol) in
anhydrous pyridine (65 mL) was treated with freshly purified TsCl
(2.07 g, 11.0 mmol). The resulting solution was stirred at rt for 24h.
The mixture was then washed with brine and extracted with AcOEt.
The organic phase was dried (Na2SO4) and evaporated under reduced
pressure. Chromatography of the crude residue over silica gel
(hexane:ethyl acetate, 100:0 to 90:10) gave the pure 7b (3.17 g,
EXPERIMENTAL SECTION
■
Nucleoside synthesis. General. All moisture-sensitive reactions
were performed under nitrogen atmosphere by using oven-dried
glassware. Solvents were dried over standard drying agents and freshly
distilled prior to use. Reactions were monitored by TLC (precoated
silica gel plate F254). Column chromatography: Kieselgel 60 (70−230
1
1
mesh); flash chromatography: Kieselgel 60 (230−400 mesh). H and
90% yield). H NMR (CDCl3, 500 MHz): δ 1.63 (ddd, J = 3.0, 10.8,
13C NMR spectra were recorded on NMR spectrometers operating at
200, 300, 400, or 500 MHz and 50, 75, 100, or 125 MHz, respectively.
High-resolution MS analysis was performed using a quadrupole/
orthogonal acceleration time-of-flight tandem mass spectrometer
(qTOF2) fitted with a standard electrospray ionization (ESI) interface.
Combustion analyses were performed by using a CHNS analyzer.
4,6-Di-O-acetyl-1,5-anhydro-D-erythro-hex-2-enitol (2). Iodine
(69 mg, 0.27 mmol) was added to a mixture of 3,4,6-tri-O-acetyl-D-
glucal (1) (3.0 g, 11.01 mmol) and Et3SiH (5.28 mL, 33.0 mmol) in
anhydrous dichloromethane (45 mL) at rt. The reaction mixture was
stirred at the same temperature for 2h; then it was diluted with water
and extracted with dichloromethane. The organic layers were dried
(Na2SO4) and evaporated under reduced pressure. Chromatography
of the crude residue (hexane:ethyl acetate, 100:0 to 80:20) provided
13.8, 1H), 2.44 (s, 3H), 2.51 (bd, J = 13.8, 1H), 3.40 (ddd, J = 1.5, 4.4,
10.3, 1H), 3.47 (d, J = 13.0, 1H), 3.62 (dd, J = 10.7, 6.0, 1H), 3.67
(ddd, J = 4.4, 10.8, 10.0, 1H), 3.76 (dd, J = 1.5, 10.7, 1H), 3.91 (bd, J =
13.0, 1H), 4.33 (d, J = 11.3, 1H), 4.46 (d, J = 11.3, 1H), 4.54 (d, J =
12.1, 1H), 4.59 (d, J = 12.1, 1H), 4.76 (bs, 1H), 7.18−7.36 (m, 12H),
7.78 (d, J = 8.2, 2H). 13C NMR (CDCl3, 125 MHz): ppm 21.6, 34.2,
69.0, 69.5, 69.6, 71.4, 73.5, 77.2, 80.2, 127.6, 127.7, 127.8, 127.9, 128.3,
128.4, 129.9, 134.5, 137.8, 138.1, 144.8. Elemental analysis calcd (%)
for C27H30O6S: C 67.20, H 6.27, S 6.64; found: C 67.31, H 6.25, S
6.62.
4′,6′-Di-O-benzyl-3′-deoxy-2′-(adenin-9-yl)-1′,5′-anhydro-D-ribo-
hexitol (8b). NaH (0.26 g, 66 mmol) was added to a stirring solution
of adenine (1.77 g, 13.0 mmol) in anhydrous DMF (126 mL). The
resulting suspension was warmed to 100 °C until a clear solution was
obtained (1h). The mixture was cooled to 25 °C, then a solution of
tosyl ester 7b (3.17 g, 66 mmol) in anhydrous DMF (20 mL) was
added. The solution was again warmed to 100 °C and stirred at the
same temperature for 16h. The mixture was then washed with brine
and extracted with AcOEt. The organic phase was dried (Na2SO4) and
evaporated under reduced pressure. Chromatography of the crude
residue over silica gel (CH2Cl2:MeOH, 100:0 to 95:5) provided the
1
the pure diacetate 2 (2.31 g, 90%). H and 13C NMR data are fully in
agreement with those already reported elsewhere.31
4,6-Di-O-benzyl-1,5-anhydro-D-erythro-hex-2-enitol (4b). Diace-
tate 2 (2.3 g, 0.01 mol) was treated with a 0.1 M NaOMe solution in
anhydrous MeOH (15 mL) at 0 °C. The reaction mixture was warmed
to rt and stirred for 16h at the same temperature. A few drops of acetic
acid were then added until neutrality, and the solvent was removed
F
DOI: 10.1021/acs.joc.5b00406
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
pure nucleoside 8b (1.08 g, 37% yield) as a white solid. H and 13C
NMR data are identical to those already reported elsewhere.30
3′-Deoxy-2′-(N6-benzoyl-adenin-9-yl)-1′,5′-anhydro-D-ribo-hexi-
tol (11). A solution of benzyl ether 8b (0.72 g, 1.62 mmol) in MeOH
(20 mL) was purged of oxygen by bubbling nitrogen for 15 min. 10%
Pd/C (0.1 g) was added and the mixture was exposed to H2 (20 psi)
while stirring for 20h. The suspension was filtered off and the filter
cake was rinsed with further MeOH (4 × 25 mL). The combined
filtrate was then concentrated under reduced pressure. The crude
residue was dissolved in anhydrous pyridine (20 mL), the mixture was
cooled to 0 °C and TMSCl (2.06 mL, 16.2 mmol) was added. After
0.5h, BzCl (0.56 mL, 4.86 mmol) was added at the same temperature.
The resulting reaction mixture was warmed to room temperature and
further stirred for 16h. The solution was cooled again to 0 °C, and
H2O (6 mL) and then NH4OH (6 mL) were sequentially added. The
reaction mixture was stirred at the same temperature for 1h; afterward
the solvent was removed under reduced pressure. Chromatography of
the crude residue (CH2Cl2:MeOH=90:10) afforded the pure N-
benzoyl nucleoside 11 (0.49 g, 82% overall yield) as a white foam. 1H
NMR (DMSO-d6, 400 MHz): δ 3.44−3.56 (m, 2H), 3.68−3.80 (m,
2H), 4.04 (dd, J = 3.1, 10.8 Hz, 1H), 4.56 (t, J = 5.7 Hz, 1H), 5.15 (d,
J = 6.6 Hz, 1H), 7.53 (t, J = 7.3 Hz, 2H), 7.63 (t, J = 7.3 Hz, 1H), 8.02
(d, J = 7.3 Hz, 2H), 8.56 (s, 1H), 8.73 (s, 1H). 13C NMR (DMSO-d6,
100 MHz): ppm 38.0, 50.9, 61.6, 65.2, 68.7, 83.6, 126.1, 128.9, 132.9,
133.1, 133.9, 143.7, 150.8, 152.0, 152.5, 166.0. Elemental analysis calcd
(%) for C18H19N5O4: C 58.53, H 5.18, N 18.96; found: C 58.40, H
5.20, N 19.02.
system: A=10 mM NaOH, pH 12.0, 0.1 M NaCl; B=10 mM NaOH,
pH 12.0, 0.9 M NaCl. The low-pressure liquid chromatography system
consisted of a L-6200A intelligent pump, a Mono-Q HR 10/10
column, a Uvicord SII 2138 UV detector and a recorder. The product-
containing fraction was desalted on a NAP-10 column and lyophilized.
Oligonucleotides were purified by RPHPLC on a C-18 column prior
to mass spectrometric analysis. A linear gradient of A: ammonium
bicarbonate (25 mM in H2O, pH 7.0), and B: acetonitrile (80% in
H2O) was applied.
Hybridization studies. LC-MS analysis. Oligonucleotides were
dissolved at a concentration of 100 μM in H2O. Samples (500 nL)
were injected on a reverse phase column (C18 PepMap 0.5 × 15 mm)
and eluted with a N,N-dimethylaminobutane/1,1,3,3,3-hexafluoro-2-
propanol and acetonitrile system at a flow rate of 12 μL/min. Spectra
were acquired using an orthogonal acceleration/time-of-flight mass
spectrometer (Q-Tof-2) in negative ion mode and subsequently
deconvoluted using the MaxEnt algorithm (MassLynx 3.4).
UV-melting experiments. Oligomers were dissolved in a buffer
solution containing NaCl (0.1 or 1 M), potassium phosphate (0.02 M,
pH 7.5), EDTA (0.1 mM). The concentration was determined by
measuring the absorbance in MilliQ water at 260 nm at 80 °C and by
assuming that hexitol nucleosides have the same extinction coefficients
per base moiety in the denatured state as the natural nucleosides (αAH,
e = 15000). The concentration for each strand was 4 μM in all
experiments. Melting curves were determined with a spectropho-
tometer. Cuvettes were maintained at constant temperature by water
circulation through the cuvette holder. The temperature of the
solution was measured with a thermistor that was directly immersed in
the cuvette. A quick heating and cooling cycle was carried out to allow
proper annealing of both strands. The samples were then heated from
10 °C to 80 °C at a rate of 0.2 °C min−1, and were cooled again at the
same speed. Melting temperatures were determined by plotting the
first derivative of the absorbance as a function of temperature; data
plotted were the average of two runs. Up and down curves in general
showed identical Tm values.
Circular dichroism measurements. CD spectra were measured at 5
°C with a J-715 spectropolarimeter equipped with a Peltier-type
temperature control system (model PTC-348WI) in thermostatically
controlled 0.1 cm cuvettes. The oligomers were dissolved and analyzed
in buffer containing NaCl (0.1 or 1 M), potassium phosphate (0.2 M,
pH 7.5) and EDTA (0.1 mM) and at a concentration of 8 μm of each
strand.
Computational methods. Model building and stability check of
α-D-HNA-containing duplexes via molecular dynamics simulations.
MD simulations were performed for a period of 20 ns using initially
AMBER 12 and later AMBER 14.39 Atomic electrostatic charges of the
HNA molecules, to be used in the Amber software package, were
calculated from the electrostatic potential at the 6-31G* level using the
package Gamess40 and a RESP fitting procedure41 (see the Supporting
Information, Table S2 for the assigned atomic charges). The force field
parameters used in the Amber simulations are those from the ff99bsc0
data set.42
Model building of duplexes. The PDB structure 2BJ6, a decameric
antiparallel β-D-HNA:RNA hybrid duplex with sequence CGCG-
AATTCGCG:CGCGAATTCGCG,43 was used as a template to
construct the right turning antiparallel β-D-HNA:α-D-HNA duplex
[D-(αAH)13:D-(βTH)13]. To “mutate” bases into Thy in the β-D-HNA
strand the quatfit software from the CCL software archives was used to
get the Thy base (with HNA sugar) in the right position in the
duplexes (keeping sugar conformations as 4C1, with Thy axially
oriented). The same tool was used to add three additional Thy
nucleotides to the strand to construct a 13mer strand. This single
strand D-(βTH)13 was then expanded by complementary adenine bases
in a Watson−Crick configuration connected to a α-D-HNA [D-
(αAH)13] sugar using the same tool. Those sugars have a 4C1
conformation with the Ade base in an equatorial orientation.
A similar method was then tried to build the α-D-HNA:β-L-HNA
duplex [D-(αAH)13:L-(βTH)13]. First, L-(βTH)13 was constructed from D-
(βTH)13 by mirroring the 3D coordinates. Starting from the right
turning D-(αAH)13:D-(βTH)13 duplex described above, the comple-
1
1′,5′-Anhydro-2′-(N6-benzoyl-adenin-9-yl)-2′,3′-dideoxy-6′-O-
monomethoxytrityl-D-ribo-hexitol (12). Monomethoxytrityl chloride
(0.43 g, 1.40 mmol) was added at rt to a solution of diol 11 (0.43 g,
1.16 mmol) in anhydrous pyridine (18 mL) and under nitrogen
atmosphere. After being stirred at the same temperature for 3h,
saturated NaHCO3 solution (2 mL) was added, the reaction solvent
was evaporated and the resulting residue was diluted with ethyl acetate
(100 mL) and washed with brine (3 × 50 mL). The organic layer was
dried (Na2SO4), evaporated under reduced pressure and the crude
residue was purified by flash chromatography (ethyl acetate) to afford
the pure monomethoxytrityl ether 12 (0.63 g, 85% yield) as a white
foam. H and 13C NMR data are in agreement with those already
1
reported elsewhere.20
1′,5′-Anhydro-2′-(N6-benzoyl-adenin-9-yl)-2′,3′-dideoxy-6′-O-
monomethoxytrityl-4′-O-[N,N-diisopropyl(2-cyanoethyl) phosphor-
amidite]-D-ribo-hexitol (13). To a solution of monomethoxytrityl
nucleoside 12 (0.63 g, 0.98 mmol) in anhydrous CH2Cl2 (6 mL), kept
at 0 °C under argon atmosphere, freshly dried diisopropylethylamine
(0.51 mL, 2.94 mmol) and 2-cyanoethyl-N,N-diisopropyl chlorophos-
phoramidite (0.33 mL, 1.47 mmol) were added. The reaction mixture
was stirred at the same temperature for 2h; then saturated aqueous
NaHCO3 (2 mL) was added. The solution was stirred for another 10
min and partitioned between CH2Cl2 (50 mL) and aqueous NaHCO3
(30 mL). The organic layer was washed with brine (3 × 30 mL) and
the aqueous phases were back extracted with CH2Cl2 (30 mL). The
collected organic layers were dried (Na2SO4) and the solvent removed
under reduced pressure. Flash chromatography of the crude residue
(hexane/acetone/TEA = 69/30/1) gave a foam which was dissolved in
CH2Cl2 (3 mL) and precipitated in cold hexane (200 mL, −30 °C)
containing 2% of diisopropylether, to afford the desired phosphor-
amidite 13 (0.74 g, 90% yield) as a white powder. The obtained
product was dried under vacuum and stored overnight under nitrogen
at −20 °C. Data for compound 13: exact mass calcd for C47H53N7O6P
[M + H]+: 842.3795, found 842.3772; 31P NMR: δ 147.60, 149.01.
Oligonucleotide synthesis. Oligonucleotide assembly was
performed with an Expedite DNA synthesizer by using the
phosphoramidite approach. The oligomers were deprotected and
cleaved from the solid support by treatment with methylamine (40%
in water) and concentrated aqueous ammonia (1:1, 30 °C). After gel
filtration on a NAP-10 column (Sephadex G25-DNA grade) with
water as eluent, the crude mixture was analyzed by using a Mono-Q
HR 5/5 anion exchange column, after which purification was achieved
by using a Mono-Q HR 10/10 column with the following gradient
G
DOI: 10.1021/acs.joc.5b00406
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
mentary D-(βTH)13 strand was replaced by a L-(βTH)13, giving us an
initial antiparallel right turning heterochiral D-(αAH)13:L-(βTH)13
duplex. However, preliminary force field-based calculations using
Amber did not result in a stable duplex structure. Then we tried an
antiparallel quasi linear duplex starting from the D-(αAH)13 single
strand with angles ε = 60, ξ = −60, α = −60, β = 180, γ = 60, δ = 60
REFERENCES
■
(1) Chemistry and Biology of Artificial Nucleic Acids; Egli, M.,
Herdewijn, P., Eds.; Wiley-VCH: Zurich, 2012.
̈
(2) (a) Nielsen, P. E. Acc. Chem. Res. 1999, 32, 624. (b) Lescrinier,
E.; Froeyen, M.; Herdewijn, P. Nucleic Acids Res. 2003, 31, 2975.
(c) Benner, S. A.; Sismour, A. M. Nat. Rev. Genet. 2005, 6, 533.
(d) Hirao, I.; Kimoto, M.; Yamashige, R. Acc. Chem. Res. 2012, 45,
2055.
(3) Kaur, H.; Babu, B. R.; Maiti, S. Chem. Rev. 2007, 107, 4672.
(4) Herdewijn, P. Chem. Biodivers. 2010, 7, 1.
(5) Nielsen, P. E. Chem. Biodivers. 2010, 7, 786.
4
and χ (C6′C1′N9C4) = 161°. Sugars have a C1 conformation, Ade
bases being equatorially connected to the sugar. Complementary Thy
1
bases connected to β-L-HNA sugars [χ (C6′C1′N1C2) = 112°; C4
conformation, with the Thy bases being axially connected] were added
and the final structure was fed to the leap program in the Amber suite
software for energy minimization and molecular dynamics simulations.
MD simulations. Solvated molecular dynamics was used to verify
the stability of the two duplexes. The structures were solvated in a
truncated octahedron TIP3P water box.44 Na+ counterions were added
to get electrostatic neutral systems. The water molecules and
counterions were then allowed to relax their positions while keeping
the solute fixed. Initially, for 20 ps, the systems were heated up to 300
K with constant-T, constant-V conditions while constraining the
position of the solute and using a Langevin temperature equilibration
scheme. MD simulations at 300 K were initiated with periodic
boundary conditions, using a cutoff distance of 10 Å for the
nonbonded interactions and the particle-mesh-Ewald method for the
summation of the Coulombic interactions,45 MD time step = 0.002 ps.
Total production simulation times per system were 10 ns, after an
equilibration of 500 ps. Initial molecular dynamics simulations resulted
in quasi-linear structures, however not very stable, resulting in
irregularities in the Watson−Crick base pairing and sugar
conformations. Some base pairs took a reverse Hoogsteen
configuration which prompted us to force by restraints all base pairs
to take this configuration. After the conformational change, the
restraints were removed and the simulations were continued for 20 ns.
Analysis of the simulated structures. Clustering. The rmsd plots
of the trajectories (see the Supporting Information) showed that the
duplexes were very flexible. Nevertheless, the last 5 ns of the 20 ns MD
trajectories were clustered into 5 groups using the average linkage
algorithm implemented in the cpptraj program of the Amber software
to extract a representative structure. For the different simulations, a
representative structure from the cluster with the highest member
count was selected.
(6) Meggers, E.; Zhang, L. Acc. Chem. Res. 2010, 43, 1092.
(7) Eschenmoser, A. Angew. Chem., Int. Ed. 2011, 50, 12412.
(8) (a) Herdewijn, P.; Marliere, P. Chem. Biodivers. 2009, 6, 791.
̀
(b) Pinheiro, V. B.; Holliger, P. Curr. Opin. Chem. Biol. 2012, 16, 245.
(9) Hendrix, C.; Rosemeyer, H.; Verheggen, I.; Seela, F.; Van
Aerschot, A.; Herdewijn, P. Chem.Eur. J. 1997, 3, 110.
(10) As already reported in previous papers, the symbols “β” and “α”
were used to rapidly distinguish hexitol nucleotides bearing the C2′
and C5′ substituents in a cis configuration (“β-HNA”) from those
displaying a trans relationship between the same substituents (“α-
HNA”).
(11) Kang, H.; Fisher, M. H.; Xu, D.; Miyamoto, Y. J.; Marchand, A.;
Van Aerschot, A.; Herdewijn, P.; Juliano, R. L. Nucleic Acids Res. 2004,
32, 4411.
(12) Fisher, M.; Abramov, M.; Van Aerschot, A.; Rozenski, J.; Dixit,
V.; Juliano, R. L.; Herdewijn, P. Eur. J. Pharmacol. 2009, 606, 38.
(13) Kolb, G.; Reigadas, S.; Boiziau, C.; van Aerschot, A.;
Arzumanov, A.; Gait, M. J.; Herdewijn, P.; Toulme,
2005, 44, 2926.
(14) Abramov, M.; Schepers, G.; Van Aerschot, A.; Van Hummelen,
P.; Herdewijn, P. Biosensors Bioelectron. 2008, 23, 1728.
(15) Kozlov, I. A.; Politis, P. K.; Pitsch, S.; Herdewijn, P.; Orgel, L. E.
J. Am. Chem. Soc. 1999, 121, 1108.
́
J.-J. Biochemistry
(16) Vastmans, K.; Froeyen, M.; Kerremans, L.; Pochet, S.;
Herdewijn, P. Nucleic Acids Res. 2001, 29, 3154.
(17) Pezo, V.; Liu, F. W.; Abramov, M.; Froeyen, M.; Herdewijn, P.;
̀
Marliere, P. Angew. Chem., Int. Ed. 2013, 52, 8139.
(18) Pinheiro, V. B.; Taylor, A. I.; Cozens, C.; Abramov, M.; Renders,
M.; Zhang, S.; Chaput, J. C.; Wengel, J.; Peak-Chew, S.-Y.;
McLaughlin, S. H.; Herdewijn, P.; Holliger, P. Science 2012, 336, 341.
(19) Taylor, A. I.; Pinheiro, V. B.; Smola, M. J.; Morgunov, A. S.;
Peak-Chew, S.; Cozens, C.; Weeks, K. M.; Herdewijn, P.; Holliger, P.
Nature 2015, 518, 427.
(20) Kerremans, L.; Schepers, G.; Rozenski, J.; Busson, R.; Van
Aerschot, A.; Herdewijn, P. Org. Lett. 2001, 3, 4129.
(21) D’Alonzo, D.; Guaragna, A.; Van Aerschot, A.; Herdewijn, P.;
Palumbo, G. Tetrahedron Lett. 2008, 49, 6068.
(22) D’Alonzo, D.; Van Aerschot, A.; Guaragna, A.; Palumbo, G.;
Schepers, G.; Capone, S.; Rozenski, J.; Herdewijn, P. Chem.Eur. J.
2009, 15, 10121.
Helical analysis. The representative structure was analyzed by a
visual inspection in chimera and a curvature calculation with CURVES
5.3.46−48 In the calculations, linearity was imposed on the axis. Because
of Curves only handles Watson−Crick base pairing correctly, some of
the helical parameters could not be calculated. Rise and twist were
obtained from local interbase pairs and averaged over all nucleotide
pairs (Supporting Information, Table S3). Images were made using the
Chimera software46 (Supporting Information, Figures S10−11).
ASSOCIATED CONTENT
* Supporting Information
■
S
Copies of NMR spectra, MS data, UV experiments, and MD
simulations. This material is available free of charge via the
Internet at http://pubs.acs.org/.
(23) Krishnamurthy, R.; Pitsch, S.; Minton, M.; Miculka, C.;
Windhab, N.; Eschenmoser, A. Angew. Chem., Int. Ed. Engl. 1996,
35, 1537.
(24) Froeyen, M.; Morvan, F.; Vasseur, J.-J.; Nielsen, P.; Van
Aerschot, A.; Rosemeyer, H.; Herdewijn, P. Chem. Biodivers. 2007, 4,
803.
(25) D’Alonzo, D.; Amato, J.; Schepers, G.; Froeyen, M.; Van
Aerschot, A.; Herdewijn, P.; Guaragna, A. Angew. Chem., Int. Ed. 2013,
52, 6662.
AUTHOR INFORMATION
Corresponding Author
*E-mail: dandalonzo@unina.it.
Notes
■
The authors declare no competing financial interest.
(26) D’Alonzo, D.; Guaragna, A.; Palumbo, G. Chem. Biodivers. 2011,
8, 373.
ACKNOWLEDGMENTS
■
(27) Eschenmoser, A.; Loewenthal, E. Chem. Soc. Rev. 1992, 21, 1.
(28) Sørensen, M. D.; Kværnø, L.; Bryld, T.; Hakansson, A. E.;
Verbeure, B.; Gaubert, G.; Herdewijn, P.; Wengel, J. J. Am. Chem. Soc.
2002, 124, 2164.
(29) Christensen, N. K.; Bryld, T.; Sørensen, M. D.; Arar, K.;
This work was supported by a FWO (Flemish Scientific
Research) grant G.0784.11 and KU Leuven financial support
(GOA/10/13). Mass spectrometry was made possible by the
support of the Hercules Foundation of the Flemish Govern-
ment (grant 20100225-7).
̊
Wengel, J.; Nielsen, P. Chem. Commun. 2004, 282.
H
DOI: 10.1021/acs.joc.5b00406
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(30) Hossain, N.; Rozenski, J.; De Clercq, E.; Herdewijn, P. J. Org.
Chem. 1997, 62, 2442.
(31) Yadav, J. S.; Reddy, B. V. S.; Premalatha, K.; Swamy, T.
Tetrahedron Lett. 2005, 46, 2687.
(32) Bruno, T.; Borcherng, H.; Wei-Hung, C.; Yea-Fen, J.; Shuo-
Cang, Z.; Subramanian, M.; Sepehr, S. Pct. Int. Appl. WO2005/049582
(2005).
(33) The observed regioselectivity was apparently the result of the
stereoselective formation of the mercurinium ion II (trans with respect
to the C5 substituent and cis with respect to the one at C4) rather
than that of ion I (cis to C5 and trans to C4). Accordingly, it can be
argued that the unfavorable interactions between the acetoxymercuric
moiety and the endocyclic oxygen in I strongly influence the
stereochemical outcome of the process; compare: (a) Paterson, I.;
Smith, J. D.; Ward, R. A. Tetrahedron 1995, 51, 9413. (b) Larrosa, I.;
Romea, P.; Urpi, F.; Balsells, D.; Vilarrasa, J.; Font-Bardia, M.; Solans,
X. Org. Lett. 2002, 4, 4651.
(34) As already suggested [ref 25], an uncommon kinetic control
ruling the pairing process has been hypothesized to explain this
phenomenon. Studies aimed to understand it are, however, underway.
(35) This last observation is not in contradiction with the overlay of
Figure 2. Indeed, it is clearly established that, despite close backbone
similarities of two nucleosides/nucleotides, large variations in various
geometrical parameters (e.g., backbone torsion angles, helical
parameters, etc.) may occur when they are incorporated into
oligonucleotide strands.
(36) As widely known, a reverse-Hoogsteen A:T base pair involves
hydrogen bonding between atoms N6 (A) and O2 (T) and between
atoms N7 (A) and N3 (T):
(37) Neidle, S. Oxford Handbook of Nucleic Acid Structure; Oxford
University Press: 1999.
(38) Nauwelaerts, K.; Lescrinier, E.; Herdewijn, P. Chem.Eur. J.
2007, 13, 90.
(39) Salomon-Ferrer, R.; Case, D. A.; Walker, R. C. WIREs Comput.
Mol. Sci. 2013, 3, 198.
(40) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.;
Gordon, M. S.; Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.;
Su, S.; Windus, T. L.; Dupuis, M.; Montgomery, J. A., Jr. J. Comput.
Chem. 1993, 14, 1347.
(41) Bayly, C. I.; Cieplak, P.; Cornell, W. D.; Kollman, P. A. J. Phys.
Chem. 1993, 97, 10269.
(42) Perez, A.; Marchan, I.; Svozil, D.; Sponer, J.; Cheatham, T. E.;
Laughton, C. A.; Orozco, M. Biophys. J. 2007, 92, 3817.
(43) Maier, T.; Przylas, I.; Strater, N.; Herdewijn, P.; Saenger, W. J.
Am. Chem. Soc. 2005, 127, 2937.
(44) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J.; Klein, M. L. J.
Chem. Phys. 1983, 79, 926.
(45) Cheatham, T. E., III; Miller, J. L.; Fox, T.; Darden, T. A.;
Kollman, P. A. J. Am. Chem. Soc. 1995, 117, 4193.
(46) Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.;
Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. J. Comput. Chem. 2004,
25, 1605.
(47) Lavery, R.; Sklenar, H. J. Biomol. Struct. Dyn. 1988, 6, 63.
(48) Lavery, R.; Sklenar, H. J. Biomol. Struct. Dyn. 1989, 6, 655.
I
DOI: 10.1021/acs.joc.5b00406
J. Org. Chem. XXXX, XXX, XXX−XXX