Highly stable triple helix formation by homopyrimidine (l)-acyclic threoninol nucleic acids with single stranded DNA and RNA

Article abstract of DOI:10.1039/c4ob02328e

Acyclic (l)-threoninol nucleic acid (aTNA) containing thymine, cytosine and adenine nucleobases were synthesized and shown to form surprisingly stable triplexes with complementary single stranded homopurine DNA or RNA targets. The triplex structures consist of two (l)-aTNA strands and one DNA or RNA, and these triplexes are significantly stronger than the corresponding DNA or RNA duplexes as shown in competition experiments. As a unique property the (l)-aTNAs exclusively form triplex structures with DNA and RNA and no duplex structures are observed by gel electrophoresis. The results were compared to the known enantiomer (d)-aTNA, which forms much weaker triplexes depending upon temperature and time. It was demonstrated that (l)-aTNA triplexes are able to stop primer extension on a DNA template, showing the potential of (l)-aTNA for antisense applications. This journal is

Full text of DOI:10.1039/c4ob02328e

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Highly stable triple helix formation by
homopyrimidine (L)-acyclic threoninol nucleic
acids with single stranded DNA and RNA†
Cite this: Org. Biomol. Chem., 2015,
13, 2366
Vipin Kumar,a,b Venkitasamy Kesavan*a and Kurt V. Gothelf*b
Acyclic (L)-threoninol nucleic acid (aTNA) containing thymine, cytosine and adenine nucleobases were
synthesized and shown to form surprisingly stable triplexes with complementary single stranded homo-
purine DNA or RNA targets. The triplex structures consist of two (L)-aTNA strands and one DNA or RNA,
and these triplexes are significantly stronger than the corresponding DNA or RNA duplexes as shown in
competition experiments. As a unique property the (L)-aTNAs exclusively form triplex structures with DNA
and RNA and no duplex structures are observed by gel electrophoresis. The results were compared to the
known enantiomer (D)-aTNA, which forms much weaker triplexes depending upon temperature and time.
It was demonstrated that (L)-aTNA triplexes are able to stop primer extension on a DNA template,
showing the potential of (L)-aTNA for antisense applications.
Received 3rd November 2014,
Accepted 17th December 2014
DOI: 10.1039/c4ob02328e
www.rsc.org/obc
did not exhibit hybridization with DNA, whereas based on the
CD studies, (D)-aTNA duplex was described as a right handed
Introduction
Fully modified acyclic nucleic acids are currently being inten- helical structure.1e
sely investigated owing to their potential for formation of
However, it is not straightforward to provide proof of the
highly stable duplex structures compared to DNA and RNA. helical handedness of artificial oligonucleotides based on the
Researchers have been synthesising various acyclic nucleic CD behaviour of the DNA or RNA.1c,1h,3 Therefore, we investi-
acids1 and have investigated their biophysical properties to gated the (L)-aTNA (Fig. 1a), because chirality plays a very
form homoduplexes and heteroduplexes with DNA or RNA. important role to induce the handedness, which can facilitate
These acyclic nucleic acids contain natural nucleobases and the binding of acyclic nucleic acids towards DNA or RNA.
acyclic scaffold units, which are interconnected by phospho-
Homopyrimidine peptide nucleic acids (PNAs) have been
diester linkages. The important goals for the construction of intensively investigated as triplex forming oligonucleotides
these acyclic oligonucleotides could be the development of (TFOs) with DNA duplex or single stranded DNA.4 However,
nanostructures,2a conformational switches,2b and bio-sensing acyclic nucleic acids containing phosphodiester linkages have
technology. Amino acid nucleic acid (AANA) was first reported an advantage over PNA with regard to solubility and binding
by Ramasamy et al. who showed that poly (T) of AANA specificity based on the chiral monomers.5
(synthesized from L-serine) did not display hybridization with
poly d(A).1a Later, Asanuma et al. reported an acyclic threoni-
nol nucleic acid (aTNA)1e and an acyclic serinol nucleic acid
(SNA)1f synthesized from D-threoninol and L-serinol respect-
ively, which exhibited highly stable homoduplexes. SNA has
been shown to make duplex with DNA and RNA while (D)-aTNA
aChemical Biology Laboratory, Department of Biotechnology, Indian Institute of
Technology, Madras (IITM), Chennai 600036, India. E-mail: vkesavan@iitm.ac.in;
Tel: +91-44-22574124
bDanish National Research Foundation Center for DNA Nanotechnology, iNANO and
Department of Chemistry, Aarhus University, 8000 Aarhus C, Denmark.
E-mail: kvg@chem.au.dk; Fax: +45 86196199
†Electronic supplementary information (ESI) available: Synthesis and character-
ization of compounds, NMR spectra, MALDI-TOF and additional figures are pro-
vided. See DOI: 10.1039/c4ob02328e
Fig. 1 (a) Chemical structures of (L)-aTNA and (D)-aTNA. (b) Schematic
model of the binding of homopyrimidine (L)-aTNA to the sequence
complementary single stranded homopurine DNA or RNA target.
2366 | Org. Biomol. Chem., 2015, 13, 2366–2374
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
So far, there has been no report on acyclic nucleic acids
that form a highly stable triple helix with DNA or RNA where
two acyclic nucleic acid strands are hybridized with single
stranded DNA or RNA. Here we present the synthesis and bio-
physical properties that show how homopyrimidine (L)-aTNA
forms a highly stable triplex helix with single stranded DNA or
RNA in a (L)-aTNA–DNA–(L)-aTNA and (L)-aTNA–RNA–(L)-aTNA
fashion (Fig. 1b).
Results and discussion
Synthesis of aTNA oligonucleotides
The (L)-aTNA phosphoramidites were synthesized starting from
commercially available (L)-threonine amino acid. Using this
synthetic methodology we obtained good yields compared to
the reported synthetic procedure1e (ESI† S1 and S2). The syn-
thesis of (L)-aTNA oligonucleotides followed the standard pro-
tocol of automatized oligonucleotide synthesis (ESI† S3 and
S4). In order to investigate the binding properties of (L)-aTNA
with DNA or RNA, various oligonucleotide strands were
synthesized (Table 1).
Biophysical properties of (L)-aTNA
First, we investigated the pairing properties of ON-1 and ON-2
with temperature-dependent UV spectroscopy at 260 nm in
10 mM phosphate buffer at pH 7 containing 100 mM sodium
chloride. The spectrum showed a highly stable duplex (Tm
=
72 °C) between ON-1 and ON-2 when compared to the analo-
gous DNA duplex (Tm = 46 °C, data not shown; Fig. 2a). The
CD profile of the ON-1/ON-2 duplex showed positive Cotton
effects at 257 and 218 nm and a negative Cotton effect at
241 nm (Fig. 2b). Furthermore single stranded ON-2 was
observed as a random coil while ON-1 displayed a preorga-
nised structure which showed CD signals at 267 nm and
218 nm as positive bands and negative bands at 247 nm and
203 nm (Fig. 2b). In addition UV melting of ON-1 showed a
Fig. 2 (a) Melting profile of (L)-aTNA duplex (ON-1/ON-2) measured by
UV at 260 nm. (b) CD profile of (L)-aTNA duplex (ON-1/ON-2), ON-1 and
ON-2 at 20 °C. Experimental conditions: 10 mM phosphate buffer con-
taining 100 mM sodium chloride at pH 7 with each oligo concentration
of 2 µM for UV-melting and 0.5 µM for CD studies.
very stable preorganized structure (ESI† S5, Fig. S1). Moreover,
poly (A) of acyclic butyl nucleic acid (BuNA) has been shown to
be a preorganized structure at neutral pH.2b It is important to
note that both stereoisomers of aTNA contain a natural DNA
nucleotide at 4′ end. However, the presence of a natural
nucleotide at the terminal position does not significantly
affect the biophysical properties of acyclic oligonucleotides.1h
Next, we investigated the duplex formation of the (D)-aTNA,
Table 1 Oligonucleotides used in the study
Entry
Oligonucleotide sequencesa
Type
ON-1
ON-2
ON-3
ON-4
ON-5
ON-6
ON-7
ON-8
ON-9
ON-10
ON-11
ON-12
ON-13
ON-14
ON-15
5′-aaaaaaaaaaaaaa-4′-dA-3′
5′-tttttttttttttt-4′-dT-3′
5′-aaaaaaaaaaaaaa-4-’dA-3′
5′-tttttttttttttt-4′-dT-3′
5′-GATCCTTTTTTTTTTTTTTTG-3′
5′-GATCCAAAAAAAAAAAAAAAG-3′
5′-tttttctctctctc-4′-dT-3′
5′-tctctctctctttt-4′-dT-3′
5′-AGAGAGAGAGAAAAA-3′
5′-tttttttt-4′
(L)-aTNA
(L)-aTNA
(D)-aTNA
(D)-aTNA
DNA
ON-3 and ON-4, which revealed a similar duplex stability (Tm
=
71 °C) with the inverse CD pattern (ESI† S5, Fig. S2). These
studies confirm that the inversion of the stereogenic center
does not affect the duplex stability; rather the opposite helical
structure (right handed or left handed) was obtained based on
the stereochemistry.
DNA
(L)-aTNA
(L)-aTNA
DNA
(L)-aTNA
(L)-aTNA
(L)-aTNA
(L)-aTNA
RNA
Triple helix properties of (L)-aTNA oligonucleotides with DNA
5′-cccccccc-4′
5′-aaaatttatattatt-4′-dA-3′
5′-taataatataaatt-4′-dT-3′
5′-AAAAAAAAAAAAAAA-3
5′-AGAGAGAGAGAAAAA-3
Our aim in the construction of the artificial nucleic acid was to
target natural nucleic acids; therefore we investigated the
hybridisation properties of aTNA with DNA and RNA. Interest-
ingly, we observed very strong interactions when (L)-aTNA
(ON-2) was mixed with DNA (ON-6), but in a 2 : 1 stoichio-
metric ratio, where the melting temperature was determined to
RNA
a Small letters indicate acyclic nucleotides. The dA and dT indicate
standard DNA nucleotides. The aTNA oligonucleotides are shown from
the 5′ to 4′ direction.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 2366–2374 | 2367

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 4 Melting profile of (D)-aTNA triplex (ON-4/ON-6/ON-4) at
260 nm in 10 mM phosphate buffer with 100 mM sodium chloride at pH
7 with each oligo concentration of 2 µM for (D)-aTNA and 1 µM for DNA.
The sample was incubated overnight at 3 °C before experiments.
Table 2 Thermal stabilities of duplex and triplexes measured by UV at
260 nma
Oligonucleotides
Tma (°C)
ON-5/ON-6
ON-1/ON-2
ON-2/ON-6/ON-2
ON-4/ON-6/ON-4
(DNA duplex)
(L-aTNA duplex)
(L-aTNA–DNA triplex)
(D-aTNA–DNA triplex)
46
72
82
47
a Experimental conditions: each oligo with 1 µM for triplexes and 2 µM
for duplexes in 10 mM phosphate buffer containing 100 mM sodium
chloride at pH 7.
Fig. 3 (a) Melting profile of triplex (ON-2/ON-6/ON-2) measured by UV
at 260 nm. (b) CD profile of triplex formation of ON-2/ON-6/ON-2 and
ON-4/ON-6/ON-4 (sample was incubated for 48 h at 3 °C) at 20 °C.
Experimental conditions: 10 mM phosphate buffer containing 100 mM
sodium chloride at pH 7 with oligo concentration for (L)-aTNA (10 µM
for CD studies and 2 µM for UV-melting) and for DNA (5 µM for CD
studies and 1 µM for melting experiment).
(L)-aTNA scaffold. These data indicate that the triple helix
structure made by (L)-aTNA is stronger than aTNA–aTNA or
DNA–DNA homoduplexes (Table 2).
Since the melting transition of the (L)-aTNA/DNA complex
measured by UV showed major hysteresis, it was important to
investigate the nature of the complex formation. In the litera-
ture, similar results have been reported for the PNA–DNA–PNA
triplex structure.4b Therefore, we investigated the stoichio-
metric ratio between (L)-aTNA and DNA by CD spectroscopy
and native gel electrophoresis. Moreover, we did not consider
(D)-aTNA for further studies due to the weak interaction with
DNA. Fig. 5a shows the CD spectrum of DNA (ON-6) at a fixed
concentration, in the presence of different concentrations of
ON-2 in 10 mM phosphate buffer with 100 mM sodium chlor-
ide at pH 7. CD titration clearly demonstrates that 5 µM of
DNA is saturated by 10 µM of ON-2 (Fig. 5a). Further addition
of ON-2 decreases the CD signal at 272 nm which is the charac-
teristic negative band for this triplex structure. This study indi-
cates that ON-2 is hybridized with ON-6 in an (L)-aTNA–DNA–
(L)-aTNA fashion. Moreover the potential formation of a triple
helix structure like DNA–DNA–(L)-aTNA was analysed by the gel
mobility shift assay (Fig. 5b). ON-2 and ON-5 homopyrimidine
strands were added together to the ON-6 and incubated for
12 hours at room temperature (25 °C). Surprisingly, we did not
be 82 °C (Fig. 3a). The CD analysis revealed positive bands at
255 nm with two shoulder peaks at 214 nm and 222 nm and
negative bands at 272 nm and 240 nm with a shoulder peak at
283 nm indicating the formation of triplex structure (Fig. 3b).
The CD obtained for (D)-aTNA (ON-4) with ON-6 demon-
strated a similar triple helical structure with opposite handed-
ness (Fig. 3b). However in this case it was required that the
mixture was incubated at 3 °C for 48 hours prior to the CD
experiments in order to obtain a strong CD signal. Moreover,
the Tm profile showed weak interaction of ON-4 with the DNA
strand (ON-6) with a melting temperature of 47 °C, where
interactions were time dependent (Fig. 4). This study shows
that the hybridization behaviour is dependent on time and
temperature. It was not possible to assign the handedness to
the resulting (L)-aTNA–DNA–(L)-aTNA complex based on the
CD studies. However, based on the stability of the (L)-aTNA/
DNA triplex as compared to the (D)-aTNA/DNA triplex it can be
speculated that the highly stable triplex formation by (L)-aTNA
with DNA could be due to the right handedness of the
2368 | Org. Biomol. Chem., 2015, 13, 2366–2374
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
(L)-aTNA does not form a duplex with DNA while it exists only
in the triplex structure with a single stranded DNA target in an
(L)-aTNA–DNA–(L)-aTNA fashion. This triplex structure consists
of Watson–Crick (2 hydrogen bonds) and Hoogsteen (2 hydro-
gen bonds) nucleobase pairing and therefore it was expected
to show a single very sharp melting transition where three
strands disrupt simultaneously (Fig. 3a).
Because two (L)-aTNA strands are involved in the triplex
structure, it is important to investigate the polarity7 of the
(L)-aTNA strands in the duplex structure. Moreover, the pres-
ence of cytosines is also important in order to target the
homopurine rich tracts of DNA or RNA strands. For the for-
mation of Hoogsteen bonds, homopyrimidine (L)-aTNA con-
taining cytosines require protonation. Therefore, the pH of the
buffer plays a very important role for the stability of the triple
helix. To test this hypothesis we constructed two asymmetric
(L)-aTNA strands (ON-7 and ON-8) containing thymine and
cytosine nucleobases. Melting temperatures were measured in
10 mM phosphate buffer with 100 mM sodium chloride at pH
5.8 and 7. It was evident from Fig. 6a that at pH 5.8, ON-7,
which is oriented in an antiparallel direction relative to the
Fig. 5 Titration of DNA by (L)-aTNA. (a) CD titration for a fixed concen-
tration of ON-6 (5 µM) with increasing concentration of ON-2 at 20 °C.
(b) Non-denatured gel electrophoresis (each DNA strand with 10 pmol).
Lane 1 = DNA duplex (ON-5/ON-6), lane 2 = ON-5/ON-6/ON-2 (50
pmol), lane 3 = ON-5/ON-6/ON-2 (100 pmol), lane 4 = ON-6/ON-2 (50
pmol) + ON-5, and lane 5 = ON-6/ON-2 (100 pmol) + ON-5. (c) Titra-
tion of ON-6 (7 pmol) with increasing amount of ON-2 (0, 1.4, 2.8, 4.2,
5.6, 7, 8.4, 9.8, 11.2, 12.6, 14 pmol). “+” sign in the figure caption indi-
cates that (L)-aTNA or DNA was added after 1 hour of incubation (10 mM
phosphate buffer with 100 mM sodium chloride at pH 7) of the first two
strands at room temperature. Conditions: TBM running buffer, 12% non-
denatured PAGE at a constant power of 3 watts.
observe any band that corresponds to a DNA–DNA–(L)-aTNA
triplex. Moreover, only two bands were observed, which indi-
cated only DNA duplex and (L)-aTNA–DNA–(L)-aTNA triplex
(Fig. 5b, lanes 2 and 3). These results clearly suggest that
(L)-aTNA was unable to hybridize with DNA duplex in order to
cause conventional triple helix formation as observed with
natural oligonucleotides.6 In addition, if (ON-2) was added to
the DNA duplex (ON-5/ON-6), then ON-2 was unable to dis-
place the homopyrimidine DNA strand (data not shown). Simi-
larly, the homopyrimidine DNA strand (ON-5) was unable to
displace the ON-2 from the triple helix (Fig. 5b, lanes 4 and 5).
These data revealed that the homopyridine DNA strand cannot
break the triple helix structure; on the other hand the homo-
pyrimidine aTNA strand is unable to invade the DNA duplex.
Furthermore, the stoichiometry was confirmed by native gel
electrophoresis. A fixed amount (7 pmol) of ON-6 was titrated
with increasing amounts of ON-2. Fig. 5c (lanes 2–11) demon-
strated that all the stoichiometric ratios of ON-2/ON-6 dis-
played only one band, which corresponds to the triplex
structure. Addition of more than two fold of ON-2 revealed
exclusively a triple helix structure (data not shown). Collec-
tively, all these studies clearly indicate that homopyrimidine of
Fig. 6 (a) Melting profiles of the triple helix at 260 nm. (b) CD spectra at
20 °C. Experimental conditions: 10 mM phosphate buffer containing
100 mM sodium chloride at pH 5.8 with oligo concentration for (L)-aTNA
(10 µM for CD studies and 2 µM or 1 µM for UV-melting) and for DNA (5
µM for CD studies and 1 µM for UV studies).
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 2366–2374 | 2369

View Article Online
Paper
Organic & Biomolecular Chemistry
DNA or RNA containing guanine and adenine nucleotides for
excellent specificity.
Table 3 Thermal stabilities of (L)-aTNA/DNA triplexes measured by UV
at 260 nma
In order to investigate the possibility of forming a duplex
between (L)-aTNA and DNA a study of mixed sequences of
(L)-aTNA nucleotides was carried out. We therefore synthesised
ON-12 and ON-13 which are complementary to each other.
Duplex stability (ON-12/ON-13) by UV was determined to be
70 °C (ESI† S5, Fig. S5). The CD profile showed positive Cotton
effects at 253 nm, 236 nm and 217 nm with a negative Cotton
effect at 271 nm (ESI† S5, Fig. S5), which is inverse compared
to the reported (D)-aTNA.1d These data suggest that the CD
profile obtained by duplex of poly (L)-aTNA (A) and poly
(L)-aTNA (T) is completely different from the mixed sequence
of (L)-aTNA, which reveals the different arrangement of nucleo-
bases in their respective duplexes. On the other hand when
ON-12 and ON-13 were mixed with their complementary DNA
strands in 10 mM phosphate buffer containing 100 mM
sodium chloride at pH 7 no hybridization was observed (data
not shown). The data strongly support the interpretation that
(L)-aTNA can interact with DNA or RNA only in the form of a
triplex structure.
Triplexes
Tmb (°C)
ON-7/ON-9/ON-7
ON-8/ON-9/ON-8
ON-7/ON-8/ON-9
65, 81
66, 84
76, >83
a Experimental conditions: each oligo with 1 µM concentration in
10 mM phosphate buffer with 100 mM sodium chloride at pH 5.8.
b Two melting transitions were observed because Watson–Crick
interactions are stronger than Hoogsteen bonding in the case of G–C
pairing.
Table 4 Thermal stabilities of triplexes measured by UV at 260 nm
Triplexes
Tma (°C)
ON-10/d(A)8/ON-10
ON-11/d(G)8/ON-11
44
22, 46
a Experimental conditions: each oligo with 1 µM concentration in
10 mM phosphate buffer with 100 mM sodium chloride at pH 7 for
ON-10 and at pH 5.8 for ON-11.
Triple helix properties of (L)-aTNA oligonucleotides with RNA
complementary DNA strand (ON-9), resulted in a similar
melting temperature to that obtained when ON-8, which is
oriented in a parallel direction, was used. These results clearly
indicate that the stability of the triple helix is independent of
the polarity of the (L)-aTNA strands where two strands of
(L)-aTNA are hybridized with one DNA strand by Watson–Crick
and Hoogsteen hydrogen bonds either in a parallel or an anti-
parallel direction.7a
Next, we investigated the triple helix stability obtained by
mixing both (L)-aTNA strands (ON-7, ON-8) with ON-9 at
pH 5.8. This resulted in improved stability compared to the
triplexes formed with two of the same (L)-aTNA strand (Fig. 6a
and Table 3). This further demonstrates that a highly stable
triplex can be obtained when ON-7 is in the antiparallel direc-
tion and ON-8 strand is in the parallel direction. Furthermore,
we also investigated the triplex structure stability under
neutral conditions (ESI† S5, Fig. S3). These investigations
revealed that low pH stabilizes the (L)-aTNA–DNA–(L)-aTNA
structure to a higher extent than neutral conditions. Next, we
investigated the conformational properties by CD spectroscopy
and the CD analysis showed positive peaks at 258 nm and
221 nm and negative peaks at 275 nm and 245 with a shoulder
peak at 283 nm (Fig. 6b).
We also investigated the short sequences (8 nucleotides) of
(L)-aTNA for triplex formation with DNA. The melting tempera-
ture of (L)-aTNA (t)8 with poly d(A)8 was determined to be 44 °C
(ESI† S5, Fig. S4 and Table 4). In addition, (L)-aTNA (c)8
showed the two melting transitions at 22 °C and 46 °C (ESI†
S5, Fig. S4).
Taken together, these results suggest that for the design of
homopyrimidine (L)-aTNA in order to capture single stranded
homopurine DNA or RNA, a range of 8 to 15 nucleotides would
be the optimum length which can be further anchored with
Controlling and detecting the expression of single stranded
RNA have wide applications in therapeutics and diagnostics.8a,b
Locked nucleic acid (LNA) has been utilized for the develop-
ment of antisense against apolipoprotein-B mRNA.8c Thus
RNAs are important target molecules and it is important to
investigate the binding properties of (L)-aTNA with RNA targets.
Interestingly, when ON-2 was mixed with poly r(A)15 (ON-14) a
stable triple helix was observed by temperature dependent UV
spectroscopy at 260 nm (Tm = 74 °C) in 10 mM phosphate
buffer with 100 mM sodium chloride at pH 7 (Fig. 7a). The CD
spectrum displayed positive bands at 258 nm, 219 nm and
211 nm with negative bands at 277 nm and 243 nm (Fig. 7b).
In addition, we investigated the triplex formation by ON-7 and
ON-8 to its complementary strand of RNA target (ON-15).
Temperature dependent UV spectroscopy at 260 nm of the
triplexes ON-7/ON-15/ON-7 revealed two melting transitions at
58 °C and 66 °C in 10 mM phosphate buffer containing
100 mM sodium chloride at pH 5.8 (Fig. 7a and Table 5).
In addition the ON-8/ON-15/ON-8 triplex stability was deter-
mined to be 66 °C indicating that either parallel or antiparallel
orientations of both homopyrimidine (L)-aTNA strands towards
the homopurine RNA target do not affect the triplex stability
as we observed for the (L)-aTNA/DNA triplex. Moreover, an
increase in melting temperature (Tm = 66 °C and 79 °C) was
observed when mixing ON-7, ON-8 and ON-15 in equal ratio at
pH 5.8. This further proves that in order to obtain highly
stable triplex structures one (L)-aTNA strand is hybridized in
an antiparallel direction (Watson–Crick base pairing) and
another (L)-aTNA strand binds in the parallel direction
(Hoogsteen base pairing) with the RNA target. The CD profile
at pH 5.8 shows positive Cotton effects at 258 nm and 224 nm
and 215 nm with negative Cotton effects at 277 nm and
244 nm (Fig. 7b). These results indicate that homopyrimidine
2370 | Org. Biomol. Chem., 2015, 13, 2366–2374
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 6 Thermal melting of triple helixes of ON-2 with mismatched
DNA strands measured by UV at 260 nma
Entry
DNA strands
Tm (°C)
ΔTm (°C)
ON-6
5′-GATCCAAAAAAAAAAAAAAAG-3′
5′-GATCCAAAAAAATAAAAAAAG-3′
5′-GATCCAAATAAAAAAATAAAG-3′
5′-GATCCAAATAAATAAATAAAG-3′
5′-GATCCAAAAAAACAAAAAAAG-3′
5′-GATCCAAACAAAAAAACAAAG-3′
5′-GATCCAAACAAACAAACAAAG-3′
5′-GATCCAAAAAAAGAAAAAAAG-3′
5′-GATCCAAAGAAAAAAAGAAAG-3′
82
70
60
49
70
60
49
70
59
0
ON-6a
ON-6b
ON-6c
ON-6d
ON-6e
ON-6f
ON-6g
ON-6h
−12
−22
−33
−12
−22
−33
−12
−33
a Experimental conditions: ON-2 with 4 µM and DNA strands with 2
µM concentration in 10 mM phosphate buffer containing 100 mM
sodium chloride at pH 7. Bold and italic letters indicate mismatched
nucleotides.
spectroscopy studies of ON-2 with various DNA strands
(Table 6 and ESI† S5, Fig. S6). Incorporation of one thymine
mismatch (T:T:T) into the DNA strand resulted in a strong
decrease in triplex stability (ΔTm = −12 °C), while two mis-
matches destabilized the triplex stability to 60 °C (ΔTm
=
−22 °C) as compared to the fully complementary DNA strand
(ON-6). Further, three mismatches displayed the triplex stabi-
lity with 49 °C (ΔTm = −33 °C). Similar results were obtained
when cytosine mismatches (T:C:T) were introduced into the
DNA strands (Table 6, ON-6d to ON-6f). Incorporation of three
cytosine mismatches revealed a less cooperative melting tran-
sition as compared to three thymine nucleotide mismatches
(ESI† S5, Fig. S6). Furthermore, one and two guanine nucleo-
tide mismatches (T:G:T) also decreased the melting tempera-
ture with ΔTm = −12 °C and ΔTm = −23 °C respectively
(Table 6, ON-6g and ON-6h). These data indicate that the
incorporation of one pyrimidine or purine mismatch destabi-
lizes the triple helix stability in a similar fashion, which is
independent of the incorporation site in the DNA strand.
These results showed the sequence specificity of complemen-
tary homopyrimidine (L)-aTNA towards homopurine DNA
strands.
Fig. 7 (a) Melting profiles at 260 nm. (b) CD spectra at 20 °C. Experi-
mental conditions: 10 mM phosphate buffer containing 100 mM sodium
chloride at pH 7 (ON-2) and pH 5.8 (ON-7 and ON-8) with oligo concen-
tration for (L)-aTNA (10 µM or 5 µM for CD studies and 2 µM or 1 µM for
UV-melting) and for RNA (5 µM for CD studies and 1 µM for UV studies).
Table 5 Thermal stabilities of (L)-aTNA/RNA triplexes measured by UV
at 260 nma
Triplexes
Tm (°C)
ON-2/ON-14/ON-2
ON-7/ON-15/ON-7
ON-8/ON-15/ON-8
ON-7/ON-15/ON-8
74
58, 66
66
Quaternary complex formation of (L)-aTNA with DNA
Homopyrimidine PNA binds to double-stranded DNA contain-
ing a sequence complementary purine target which results in
‘P-loop’ triplex invasion structure.9 In order to investigate the
strand invasion properties of (L)-aTNA, DNA strands of 55
nucleotides (ON-16 and ON-17) were synthesized containing
d(A)15 nucleotides in the middle. ON-2 was added to the DNA
duplex in 10 mM Tris buffer containing 100 mM sodium chlor-
ide and 10 mM MgCl2 at pH 8.2 and incubated overnight at
25 °C. However, no invasion product was observed which was
indicated by the non-denatured gel electrophoresis (Fig. 8a,
66, 79
a Experimental conditions: each oligo with 1 µM concentration in
10 mM phosphate buffer with 100 mM sodium chloride at pH 7 or pH
5.8.
(L)-aTNA also forms a stable triplex with a complementary
homopurine RNA strand.
Mismatch study
We have shown that homopurine (L)-aTNA was capable of lanes 5 and 6). On the other hand, when ON-2 was mixed with
making a stable triple helix with complementary single the DNA strand containing d(A)15 and incubated for 2 hours
stranded DNA or RNA. Thus, it was important to investigate followed by the addition of complementary DNA strand, a qua-
the binding properties of mismatched DNA strands with ternary product was observed (Fig. 8a, lanes 7 and 8). More-
(L)-aTNA. Therefore, we carried out temperature dependent UV over, when all three strands were mixed at the same time only
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 2366–2374 | 2371

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 8 Quaternary complex of (L)-aTNA with a DNA duplex. (a) The
experiment was carried out using 10 pmol of each DNA strand, lane 1 =
ON-16, lane 2 = ON-17, lane 3 = ON-16/ON-17, lane 4 = ON-17/ON-2
(50 pmol), lane 5 = ON-16/ON-17 + ON-2 (50 pmol), lane 6 = ON-16/
ON-17 + ON-2 (100 pmol), lane 7 = ON-17/ON-2 (50 pmol) + ON-16,
lane 8 = ON-17/ON-2 (100 pmol) + ON-16, and lane 9 = ON-16/ON-17/
ON-2 (100 pmol). (b) Each lane containing 10 pmol of ON-17 and 50
pmol of ON-2, lane 1 = ON-16, lane 2 = ON-17, lane 3 = ON-16/ON-17,
lane 4 = ON-17/ON-2, lane 5 = ON-17/ON-2 + ON-16 (5 pmol), lane 6 =
ON-17/ON-2 + ON-16 (10 pmol), lane 7 = ON-17/ON-2 + ON-16 (15
pmol), lane 8 = ON-17/ON-2 + ON-16 (20 pmol), and lane 9 = ON-17/
ON-2 + ON-16 (30 pmol). “+” sign in the figure caption indicates that
(L)-aTNA or DNA was added after 1 hour of incubation (10 mM Tris buffer
with 100 mM NaCl, 10 mM MgCl2 at pH 8.2) of duplex or triplex. Con-
ditions: TBM running buffer, 10% non-denatured PAGE at a constant
power of 3 watts.
Fig. 9 DNA polymerase stop assay. Primer extension studies were
carried out using 5 pmol of primer (ON-18) and 5 pmol of DNA template
(ON-17). (a) Lane 1 = ON-17, lane 2 = ON-16/ON-17, lane 3 = ON-17/
ON-18 + Bsu polymerase + dNTPs, lane 4 = lane 3 + ON-2 (5 pmol),
lane 5 = lane 3 + ON-2 (10 pmol), lane 6 = lane 3 + ON-2 (20 pmol),
lane 7 = lane 3 + ON-2 (40 pmol), lane 8 = lane 3 + ON-2 (60 pmol),
lane 9 = lane 3 + ON-2 (80 pmol), and lane 10 = ON-17/ON-18. (b) Lane
1 = ON-17, lane 2 = ON-17/ON-18, lane 3 = ON-16/ON-17, lane 4 =
ON-17/ON-18 + Bsu polymerase + dNTPs, lane 5 = lane 3 + ON-7 (5
pmol), lane 6 = lane 3 + ON-7 (10 pmol), lane 7 = lane 3 + ON-7 (20
pmol), lane 8 = lane 3 + ON-7 (40 pmol), lane 9 = lane 3 + ON-7 (60
pmol), and lane 10 = lane 3 + ON-7 (80 pmol). “+” sign in the figure
caption indicates that (L)-aTNA was added after 1 hour of incubation of
the primer and the DNA template. Conditions: TBE running buffer, 10%
non-denatured PAGE at a constant power of 3 watts.
a small band of quaternary product was observed (Fig. 8a, lane
9). For further confirmation we carried out titration of the
triplex product with increasing amounts of ON-16.
(L)-aTNA towards its target. In future (L)-aTNA could provide a
platform in order to stop the translation of unwanted protein
biosynthesis.
Fig. 8b (lanes 5–9) suggests the formation of the quaternary
product preferably, which rules out the formation of a more
complex product, where two strands of ON-16 can bind with
the triplex (ON-2/ON-17/ON-2). These results indicate that (L)-
aTNA is unable to invade the DNA duplex, which might be due
to the repulsion of the negatively charged phosphate back-
bone. On the other hand due to the high stability of the triplex
(L)-aTNA–DNA–(L)-aTNA, once formed, the complementary strand
of DNA was unable to displace the (L)-aTNA strands. These
data indicate very strong interaction of (L)-aTNA towards DNA.
Conclusions
We have synthesised (L)-aTNA from (L)-threonine amino acid.
Temperature dependent UV studies, CD spectroscopy and gel
electrophoresis demonstrated that homopurine (L)-aTNA pre-
ferably binds with homopurine DNA or RNA targets in the
form of triple helical structure. The most stable (L)-aTNA–
DNA–(L)-aTNA or (L)-aTNA–RNA–(L)-aTNA triplexes were
obtained when one (L)-aTNA strand is associated in antiparal-
lel orientation and the other (L)-aTNA strand is in the parallel
orientation relative to the DNA or RNA strand. Furthermore,
we demonstrated that (L)-aTNA was capable of blocking the
DNA polymerase by making a triplex with the DNA template.
These novel findings reveal the homopyrimidine (L)-aTNA as
an efficient artificial nucleic acid for targeting single stranded
homopurine DNA or RNA. In turn these properties may have
potential for the development of biosensors and therapeutic
molecules and may also find application in DNA nanotechnol-
ogy. The incorporation of pseudoisocytosine11 into the
(L)-aTNA scaffold and construction of bis-(L)-aTNA is in pro-
gress to abolish the pH-sensitivity and improve the kinetics
respectively.
DNA polymerase stop assay
Acyclic oligonucleotides are very stable against nucleases.10
The construction of such a type of fully modified nucleic acids
could provide a tool for the development of antisense mole-
cules. Recently, SNA nucleotides have been incorporated into
siRNA which showed improved activity and stability. However
fully modified SNA did not show any activity towards its
target.10 Since (L)-aTNA forms a highly stable triple helix with
DNA or RNA targets, we set out to investigate the DNA poly-
merase stop assay on a DNA template in the presence of (L)-
aTNA. We chose a DNA strand of 55 nucleotides (ON-17) which
is containing d(A)15 in the middle of the strand. Interestingly,
ON-2 was capable of stopping the primer (ON-18) extension by
Bsu DNA polymerase (Fig. 9a, lanes 4–10). In order to investi-
gate the specificity of the (L)-aTNA oligomers towards its DNA
target we analysed the primer extension stop assay in the pres-
ence of ON-7 which was not fully complementary to its DNA
Experimental section
Circular dichroism (CD)
target and as a result was unable to stop primer extension CD experiments were performed in Milli-Q water using a Jasco
(Fig. 9b, lanes 4–10). This clearly shows the specificity of model J-810 instrument. Quartz cuvettes (Hellma) were used
2372 | Org. Biomol. Chem., 2015, 13, 2366–2374
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
for scanning the samples at path lengths of 1 mm or 10 mm at their kind support of the CD instrument. Vipin Kumar thanks
20 °C. The concentration of each oligonucleotide was 5 µM Novozymes and Holck-Larsen Foundation Biotechnology
(for a 1 mm path length cuvette) or 0.5 µM (for a 10 mm path Mobility Scholarships Denmark–India. Vipin Kumar thanks
length cuvette) in 10 mm phosphate buffer containing Grit Jørgensen and Michael Kjelstrup for help.
100 mM sodium chloride at pH 7 or pH 5.8. Samples were
scanned at 100 nm min−1 with data pitch 0.2 nm and each CD
spectrum was obtained by the average of five scans. CD-titra-
tion (Fig. 5a) was carried out by preparing all the samples
Notes and references
separately with different concentration ratios and incubating
them overnight at 25 °C. CD spectra were baseline subtracted
and smoothed. RNAs were purchased from Sigma-Aldrich.
1 (a) K. S. Ramasamy and W. Seifert, Bioorg. Med. Chem. Lett.,
1996, 6, 1799–1804; (b) A. K. Sharma, P. Kumar and
K. C. Gupta, Helv. Chim. Acta, 2001, 84, 3643–3649;
(c) L. Zhang, A. Peritz and E. Meggers, J. Am. Chem. Soc.,
2005, 127, 4174–4175; (d) J. J. Chen, X. Cai and
J. W. Szostak, J. Am. Chem. Soc., 2009, 131, 2119–2121;
(e) H. Asanuma, T. Toda, K. Murayama, X. Liang and
H. Kashida, J. Am. Chem. Soc., 2010, 132, 14702–14703;
(f) H. Kashida, K. Murayama, T. Toda and H. Asanuma,
Angew. Chem., Int. Ed., 2011, 50, 1285–1288; (g) P. Karri,
V. Punna, K. Kim and R. Krishnamurthy, Angew. Chem.,
Int. Ed., 2013, 52, 5840–5844; (h) V. Kumar, K. R. Gore,
P. I. Pradeepkumar and V. Kesavan, Org. Biomol.
Chem., 2013, 11, 5853–5865; (i) P. Li, J. Sun, M. Su,
X. Yang and X. Tang, Org. Biomol. Chem., 2014, 12, 2263–
2272.
2 (a) R. S. Zhang, E. O. McCullum and J. C. Chaput, J. Am.
Chem. Soc., 2008, 130, 5846–5847; (b) V. Kumar and
V. Kesavan, RSC Adv., 2013, 3, 19330–19340.
3 (a) S. Sforza, G. Haaima, R. Marchelli and P. E. Nielsen,
Eur. J. Org. Chem., 1999, 197–204; (b) T. Tedeschi, S. Sforza,
A. Dossena, R. Corradini and R. Marchelli, Chirality, 2005,
17, S196–S204.
Ultraviolet-melting
Temperature dependent UV spectroscopy experiments were
carried out on a Varian Cary 300 BIO instrument. Quartz cuv-
ettes (Hellma) were used for scanning the samples at a path
length of 10 mm in 10 mM phosphate buffer with 100 mM
sodium chloride at pH 7 or pH 5.8. Samples were scanned
(heating and cooling ramps) at a rate of 1 °C min−1 at 260 nm.
The melting temperature (Tm) was determined by the first
derivative of the melting curve.
Electromobility shift assay (EMSA)
A known amount of (L)-aTNA was mixed with complementary
single stranded DNA in 10 mM phosphate buffer containing
100 mM sodium chloride at pH 7 or 10 mM Tris buffer con-
taining 100 mM sodium chloride and 10 mM magnesium
chloride at pH 8.2. Samples were incubated at room tempera-
ture (∼25 °C). Gel loading dye Blue (6×) was added to the oligo-
nucleotide samples to final 1× concentration. The complexes
were separated on 10 or 12% non-denaturing gel (polyacryl-
amide) at low temperature (3 °C) with 1× TBE (89 mM Tris-
borate and 2 mM EDTA, pH 8.3) or 1× TBM (100 mM Tris-
borate and 5 mM MgCl2, pH 8.2) running buffer at constant
power of 3 watts and visualized using a fluorescent intercalat-
ing dye.
4 (a) P. E. Nielsen, M. Egholm, R. H. Berg and O. Buchardt,
Science, 1991, 254, 1497–1500; (b) M. C. Griffith,
L. M. Risen, M. J. Greig, E. A. Lesnik, K. G. Sprankle,
R. H. Griffey, J. S. Kiely and S. M. Freier, J. Am. Chem. Soc.,
1995, 117, 831–832; (c) H. Kuhn, V. V. Demidov,
P. E. Nielsen and M. D. Frank-Kamenetskii, J. Mol. Biol.,
1999, 286, 1337–1345.
Primer extension studies
5 (a) P. E. Nielsen, G. Haaima, A. Lohse and O. Buchardt,
Angew. Chem., Int. Ed. Engl., 1996, 35, 1939–1942;
(b) A. Dragulescu-Andrasi, S. Rapireddy, B. M. Frezza,
C. Gayathri, R. R. Gil and D. H. Ly, J. Am. Chem. Soc., 2006,
128, 10258–10267.
6 M. Duca, P. Vekhoff, K. Oussedik, L. Halby and
P. B. Arimondo, Nucleic Acids Res., 2008, 36, 5123–5138.
7 (a) N. J. Peffer, J. C. Hanvey, J. E. Bisi, S. A. Thomson,
C. F. Hassman, S. A. Noble and L. E. Babiss, Proc. Natl.
Acad. Sci., U. S. A., 1993, 90, 10648–10652;
(b) O. V. Krupnik, Y. A. Guscho, K. A. Sluchanko,
P. E. Nielsen and Y. S. Lazurkin, J. Biomol. Struct. Dyn.,
2001, 19, 535–542.
The inhibition of primer (5 pmol) extension on the DNA tem-
plate (5 pmol) by Bsu DNA polymerase (2 units) was carried
out with dNTPs (200 µM) in 10 µl buffer. Increasing amounts
of (L)-aTNA were added in order to stop the DNA polymerase.
The reaction mixture was incubated overnight at 25 °C. The
reaction mixture was analyzed by non-denature gel electro-
phoresis in TBE buffer. DNA polymerase and dNTPs were
purchased from New England Biolabs.
Acknowledgements
The authors acknowledge financial support from the Danish
National Research Foundation (DNRF 81) and Department of
Biotechnology (BT/PR/10064/AGR/36/30/07), Govt. of India.
The authors gratefully thank Prof. Daniel Otzen from Center
for Insoluble Protein Structures (inSPIN), iNANO and Depart-
ment of Molecular Biology and Genetics, Aarhus University for
8 (a) F. Boutimah-Hamoudi, E. Leforestier, C. Sénamaud-
Beaufort, P. E. Nielsen, C. Giovannangeli and T. E. Saison-
Behmoaras, Nucleic Acids Res., 2007, 35, 3907–3917;
(b) T. Koch, Nucleic Acids Symp. Ser., 2008, 52, 45;
(c) E. M. Straarup, N. Fisker, M. Hedtjärn, M. W. Lindholm,
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 2366–2374 | 2373

View Article Online
Paper
Organic & Biomolecular Chemistry
C. Rosenbohm, V. Aarup, H. F. Hansen, H. Ørum, 10 Y. Kamiya, J. Takai, H. Ito, K. Murayama, H. Kashida
J. B. R. Hansen and T. Koch, Nucleic Acids Res., 2010, 38,
7100–7111.
and H. Asanuma, ChemBioChem, 2014, 15, 2549–
2555.
9 (a) T. Bentin, G. I. Hansen and P. E. Nielsen, Nucleic Acids 11 M. Egholm, L. Christensen, K. L. Deuholm, O. Buchardt,
Res., 2006, 34, 5790–5799; (b) M. E. Hansen, T. Bentin and
P. E. Nielsen, Nucleic Acids Res., 2009, 37, 4498–4507.
J. Coull and P. E. Nielsen, Nucleic Acids Res., 1995, 23, 217–
222.
2374 | Org. Biomol. Chem., 2015, 13, 2366–2374
This journal is © The Royal Society of Chemistry 2015