Specific recognition of cytosine by hypoxanthine in pyrrolidinyl peptide nucleic acid

Article abstract of DOI:10.1039/c3ob27129c

Hypoxanthine is an unnatural base that can potentially pair with all natural nucleobases. While hypoxanthine in DNA exhibits marginal preference for pairing with cytosine (C), little is known about its pairing behavior in other DNA analogues. In this study, we synthesized a hypoxanthine-containing monomer and incorporated it into pyrrolidinyl peptide nucleic acid with α/β-peptide backbone derived from d-prolyl-(1S,2S)-2- aminocyclopentanecarboxylic acid (acpcPNA). DNA binding studies clearly revealed that hypoxanthine in acpcPNA behaves like G-analogue because it can specifically form a stable base pair with dC in DNA. The ability to replace G by hypoxanthine without affecting the base pairing properties of acpcPNA can solve a number of problems associated with G-rich acpcPNA including difficult synthesis, formation of secondary structures and fluorescence quenching.

Full text of DOI:10.1039/c3ob27129c

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Specific recognition of cytosine by hypoxanthine in
pyrrolidinyl peptide nucleic acid†
Cite this: Org. Biomol. Chem., 2013, 11,
2310
Chotima Vilaivan,a Wimonmas Srinarang,a Nattawut Yotapan,a Woraluk Mansawat,a
Chalothorn Boonlua,a Junji Kawakami,b,c Yoshie Yamaguchi,b,c Yuko Tanakab,c and
Tirayut Vilaivan*a
Hypoxanthine is an unnatural base that can potentially pair with all natural nucleobases. While hypo-
xanthine in DNA exhibits marginal preference for pairing with cytosine (C), little is known about its
pairing behavior in other DNA analogues. In this study, we synthesized a hypoxanthine-containing
monomer and incorporated it into pyrrolidinyl peptide nucleic acid with α/β-peptide backbone derived
from D-prolyl-(1S,2S)-2-aminocyclopentanecarboxylic acid (acpcPNA). DNA binding studies clearly
revealed that hypoxanthine in acpcPNA behaves like G-analogue because it can specifically form a stable
base pair with dC in DNA. The ability to replace G by hypoxanthine without affecting the base pairing
properties of acpcPNA can solve a number of problems associated with G-rich acpcPNA including difficult
synthesis, formation of secondary structures and fluorescence quenching.
Received 31st October 2012,
Accepted 30th January 2013
DOI: 10.1039/c3ob27129c
www.rsc.org/obc
and peptide nucleic acid (PNA)6 – a DNA analogue in which
the deoxyribose phosphate backbone is completely replaced by
Introduction
Hypoxanthine is a purine base often found in the first anti- a peptide-like structure such as N-2-aminoethylglycine (also
codon position of tRNA.1 From its structure, there exist possibili- known as aegPNA)7 – had been investigated. While hypo-
ties of formation of reasonable hydrogen-bonded base pairs xanthine in LNA can be regarded as G analogue because it can
with all canonical nucleobases in DNA and RNA (Scheme 1), preferentially pair to C albeit with lower stability and selectivity
leading to a hypothesis that it can behave as universal base. than G,5 hypoxanthine in aegPNA may pair to either C or A,
Indeed, applications of oligonucleotides containing inosine depending on the identity of flanking nucleobases.6 Base
(the nucleotide form of hypoxanthine, often designated as I) pairing properties of aegPNA bearing other unnatural nucleo-
as degenerate probes and primers have been realized.2 The bases including xanthine,8 cyanuric acid9 as well as other
pairing behavior of hypoxanthine in DNA was by no means potential universal nucleobase systems10 have also been
universal, but was fairly selective for pairing with C. The base reported. In view of superior binding characteristic of our new
pairs containing I are generally destabilized in solution com- pyrrolidinyl PNA consisting of an α/β peptide backbone
pared to conventional Watson–Crick base pairs, with the I·C derived from D-proline/(1S,2S)-2-aminocyclopentanecarboxylic
pairs being the least destabilized and the order of stability of acid (acpcPNA) (Scheme 1)11 in terms of binding affinity as
the base pairs is I·C > I·A > I·G ∼ I·T.3 However, a recent stabi- well as base-pairing and directional specificity compared to
lity study suggested that I-containing base pairs are at least as aegPNA,12 we were interested in synthesizing and evaluating
stable as the A·T pairs in the gas phase.4 The base-pairing be- the DNA-pairing properties of acpcPNA carrying hypoxanthine
havior of hypoxanthine in other unnatural analogues of DNA as an unnatural base. The knowledge of pairing rules of hypo-
and RNA is less well characterized. The base-pairing properties xanthine (will be designated as I from now on) in acpcPNA
of hypoxanthine incorporated into locked nucleic acid (LNA)5 should be useful in expanding the scope of applications of this
acpcPNA system.
aOrganic Synthesis Research Unit, Department of Chemistry, Faculty of Science,
Chulalongkorn University, Phayathai Road, Patumwan, Bangkok 10330, Thailand.
E-mail: vtirayut@chula.ac.th; Fax: +66-2-2187598
bDepartment of Nanobiochemistry, FIRST, Konan University,
Results and discussion
Synthesis of hypoxanthine acpcPNA monomer
7-1-20 Minatojima-minamimachi, Chuo-ku, Kobe 650-0047, Japan
cFrontier Institute for Biomolecular Engineering Research (FIBER), Konan University,
7-1-20 Minatojima-minamimachi, Chuo-ku, Kobe 650-0047, Japan
Direct SN2 displacement of the tosylate group in the known
intermediate (1) with hypoxanthine under standard conditions
successfully employed for the synthesis of the protected
†Electronic supplementary information (ESI)
available. See DOI:
10.1039/c3ob27129c
2310 | Org. Biomol. Chem., 2013, 11, 2310–2317
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 1 Structures of aegPNA and acpcPNA (top) and hypothetical base pairing schemes between hypoxanthine and all four canonical nucleobases (bottom).
Scheme 2 Synthesis of hypoxanthine acpcPNA monomer (3).
adenine and cytosine monomers13 did not give the expected unprotected hypoxanthine monomer was accomplished by
substitution product. A related study had shown that regio- refluxing with 2 M aqueous HCl. Finally, the Fmoc group was
selectivity (most notably O-alkylation) was also problematic introduced using FmocOSu under aqueous basic conditions to
with direct alkylation of hypoxanthine by Mitsunobu reac- give the Fmoc-protected hypoxanthine acpcPNA monomer (3).
tion.14 An alternative method starting from 6-chloropurine as a The N9 substitution pattern in the final product was confirmed
masked hypoxanthine equivalent as shown in Scheme 2 was by 1H–13C HMBC experiment (Fig. S3†) whereby correlations
therefore employed.15 In the first step, 6-chloropurine was between H4′ (δH 5.11 ppm)–C(4) (δC 148.3 ppm) and between
reacted with 1 in the presence of K2CO3 to give the 6-chloro- H4′–C(8)H (δC 138.8 ppm), but not between H4′–C(5) (δC
purine derivative (2). The N9 regioselectivity was assumed 124.2 ppm), were observed.
based on literature precedent in a closely related system.16 The
Boc and diphenylmethyl (Dpm) groups were simultaneously
Synthesis of hypoxanthine-containing acpcPNA
removed by treatment with trifluoroacetic acid (TFA). Sub- The Fmoc-protected monomer (3) could be incorporated into
sequent hydrolysis of the 6-chloro group to give the any desired positions of acpcPNA sequences employing the
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 2310–2317 | 2311

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 1 Sequence and mass spectrometric characterization (MALDI-TOF) data
of the acpcPNA synthesized in this work
Entry
PNA
Sequence (N→C)a
m/z (calcd)
m/z (found)
1
2
3
4
5
6
7
8
PNA1
PNA2
PNA3
PNA4
PNA5
PNA6
PNA7
PNA8
PNA9
PNA10
PNA11
TTTTITTTT
3189.4
3821.1
3803.1
3773.1
3853.1
3836.2
3818.1
3860.2
3875.2
4178.4
4193.4
3190.9
3820.1
3801.0
3772.5
3853.9
3836.9
3816.0
3860.8
3873.0
4180.8
4196.5
CCTTA
CCTTT
CCTTC
CCTTG
CCTTA
CCTTT
AATTTICATCA
AATTTGCATCA
Flu-O-ICTGCTTCACTb
Flu-O-G
CTGCTTCACTb
̲
̲
̲
̲G̲A̲
9
10
11
̲
̲
a All acpcPNA were C-terminal capped with L-lysinamide and N-
acetylated. b The ACPC spacer at the N-terminus was omitted from
these two sequences. Flu
aminoethoxyethoxyacetyl linker.
=
5(6)-carboxyfluorescein;
O
=
standard protocol for Fmoc-based PNA synthesis using stan-
dard HATU/DIEA activation.11 The sequences synthesized
include a homothymine acpcPNA with the base T at the
middle position of the strand was replaced by I (PNA1), four
11mer mix-base acpcPNA with an I residue in the middle
flanked by each of the four different standard nucleobases
(PNA2–PNA5), two hypoxanthine-modified mix-base acpcPNA
(PNA8 and PNA10). The PNA10 containing hypoxanthine at the
N-terminal position was further modified at the N-terminus
with 5(6)-carboxyfluorescein. Selected acpcPNA sequences
having the base I replaced by G were also synthesized as con-
trols (PNA6, PNA7, PNA9 and PNA11). The identity of all
acpcPNA was confirmed by MALDI-TOF mass spectrometry
and the purity was determined to be >90% by reverse phase
HPLC. The sequences and characterization data of the
acpcPNA synthesized are summarized in Table 1.
Fig. 1 Melting curves of hybrids between PNA1 and dA4NA4 (N: –– C, –·–·– G,
– – T, ⋯·· A). Conditions: 1.0 μM PNA, 1.0 μM DNA, 10 mM sodium phosphate
buffer pH 7.0, 100 mM NaCl. The curves were smoothed by the boxcar algor-
ithm using the smooth macro in KaleidaGraph 4.1 employing a window size = 5.
with C in DNA is a general behavior or limited to certain
sequences. The PNA2–PNA5 with the same mix-base sequence
but carrying four different base triads (NIN; N = A, T, C, G) in
the middle position of the acpcPNA strand were synthesized
and their binding properties with DNA studied by thermal
denaturation experiments. These particular sequences were
chosen because the corresponding data for aegPNA system are
available for direct comparison.6 As shown in Table 2 and
Fig. S16,† the Tm data obtained from these mixed sequence
acpcPNA again confirmed the preferential pairing of pI with
dC. In all four cases the pI·dC pairs were the most stable (Tm
Hybridization properties of hypoxanthine-containing acpcPNA
We initially investigated the base pairing behavior of PNA1, with complementary DNA of PNA2: 70.2; PNA3: 66.1; PNA4:
which has a homothymine sequence with I in the middle posi- 48.0; PNA5: 60.8 °C). The unusually high thermal stability of
tion. We had earlier demonstrated that homothymine A+T rich acpcPNA hybrids has previously been observed in this
acpcPNA only form duplex with complementary DNA instead acpcPNA system.11 Thermodynamic parameters analyses indi-
of the triplex usually observed with aegPNA and other PNA cated that the formation of A·T base pairs in the acpcPNA·DNA
systems.11 It is therefore reasonable to assume that the PNA1 hybrids are enthalpically more favorable over G·C pairs.11b
should also form duplex with DNA. As shown in Fig. 1, the Tm Interestingly, the thermal stabilities of acpcPNA·DNA hybrids
of the hybrid of PNA1 with dA4CA4 (pI·dC pair) was 64.1 °C, carrying pI·dC were generally higher than the corresponding
which was much higher than the other three hybrids (dA4GA4: acpcPNA·DNA hybrids with pG·dC pairs (Fig. S17†). For
<20 °C; dA4TA4: 34.3 °C; dA9: 38.7 °C). The higher Tm of the example, the hybrid of PNA6, bearing the same sequence as
hybrid containing pI·dC pair by >25–30 °C clearly revealed that PNA2, but with I replaced by G, with its complementary DNA
the base hypoxanthine in PNA1 showed a strong preference in showed a Tm of 62.3 °C which was 7.9 °C lower than that of the
pairing with C. Hence, the base I behaves as a G analogue corresponding PNA2 hybrid. Likewise, the hybrid of PNA7 with
rather than as a universal base. This behavior is analogous to its complementary DNA showed a Tm of 64.4 °C which was
natural DNA3 and LNA,5 although the dI·dC pairing in 1.7 °C lower than that of the corresponding DNA hybrid of
acpcPNA described here is much more selective.
PNA3. Significant reduction in Tm values was observed in
As the stabilities of I-containing base pairs in DNA tend to acpcPNA·DNA hybrids carrying “mismatched” pI·dT, pI·dG and
vary depending on the sequence,17 it is important to deter- pI·dA pairs (ΔTm PNA2: −22.0 to −27.2; PNA3: −24.1 to −34.9;
mine whether the preference for pairing of base I in acpcPNA PNA4: −11.9 to −16.2; PNA5: −7.7 to −18.6 °C). The
2312 | Org. Biomol. Chem., 2013, 11, 2310–2317
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 2
T
m and thermodynamic parameters of hypoxanthine-containing PNA2–PNA5 and guanine-containing PNA6–PNA7a,b
Entry
PNA
DNAc
Tm (°C)
ΔH° (kcal mol−1
)
ΔS° (cal mol−1 K−1
)
ΔG°298 (kcal mol−1
)
1
2
3
4
5
6
7
8
PNA2
(…AIA…)
…TCT…
…TTT…
…TGT…
…TAT…
…TCT…
…TTT…
…TGT…
…TAT…
…ACA…
…ATA…
…AGA…
…AAA…
…ACA…
…ATA…
…AGA…
…AAA…
…GCG…
…GTG…
…GGG…
…GAG…
…CCC…
…CTC…
…CGC…
…CAC…
70.2
43.0
44.0
48.0
62.3
42.3
39.1
37.6
66.1
42.0
31.2
37.7
64.4
46.4
29.0
25.5
48.0
36.1
−82.0 0.5
−55.0 0.7
−57.0 0.8
−66.8 1.8
−74.6 0.8
−55.4 0.6
−48.2 0.2
−52.9 1.5
−69.1 0.5
−53.3 0.5
−44.0 0.5
−46.6 0.5
−68.6 0.6
−60.6 0.5
−41.1 0.7
−39.6 0.6
−50.4 0.5
−51.2 0.7
−210 1.5
−145 2.1
−151 2.6
−179 5.8
−194 2.4
−147 2.0
−125 0.7
−141 4.7
−175 1.5
−140 1.6
−116 1.7
−121 1.5
−174 1.5
−161 1.6
−107 2.1
−104 1.9
−128 1.4
−137 2.4
−19.4 0.7
−11.7 0.9
−12.0 1.1
−13.4 2.5
−16.9 1.1
−11.7 0.9
−10.8 0.3
−10.8 2.0
−17.0 0.7
−11.5 0.7
−9.5 0.7
−10.5 0.7
−16.6 0.8
−12.7 0.7
−9.2 0.9
−8.7 0.8
−12.2 0.6
−10.5 1.0
PNA6
(…AGA…)
9
PNA3
(…TIT…)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
PNA7
(…TGT…)
PNA4
(…CIC…)
Non-two-state melting curve
−134 2.1
31.8
60.8
47.7
42.2
53.1
−49.7 0.6
−64.0 0.6
−58.6 0.5
−49.8 0.5
−68.3 2.0
−9.7 0.9
−15.5 0.8
−12.8 0.6
−11.3 0.7
−14.5 2.7
PNA5
(…GIG…)
−163 1.9
−154 1.5
−129 1.6
−181 6.0
a The melting experiments were carried out in 10 mM sodium phosphate buffer pH 7.0, 100 mM NaCl, concentration of PNA and DNA = 1.0 μM,
monitored by UV-Vis spectrophotometry at 260 nm, heating rate 1 °C min−1 b Both Tm and thermodynamics parameters were determined from
.
thermal denaturation curves by nonlinear curve-fitting as described in the Experimental section. c The symbols …NNN… denote the three bases
in the middle position of the DNA sequences (5′→3′) e.g. …TCT… = GATGTCTAAGG; …AAA … = GATGAAAAAGG etc.
■
Fig. 2 Comparison of melting temperatures (Tm) of DNA hybrids of hypoxanthine-containing acpcPNA PNA2 (AIA: ) and aegPNA ( ) (left) and hypoxanthine-
■
containing acpcPNA PNA3 (TIT: ) and aegPNA ( ) (right). The symbols …NNN… in the x-axis denote the three base in the middle position of the DNA sequences.
Conditions: 1.0 μM PNA, 1.0 μM DNA, 10 mM sodium phosphate buffer pH 7.0, 100 mM NaCl. Tm data of aegPNA are taken from ref. 6.
corresponding ΔTm figures for PNA6 and PNA7 are −20.0 to other hand, when the flanking bases are A, I in aegPNA prefers
−24.7 and −18.0 to −38.9 °C, respectively. These results clearly pairing to C (Fig. 2). Based on a quantum chemical calcu-
indicated that the base I in acpcPNA could behave as a G ana- lation, it was suggested that the dA in the pI·dA pair adopted a
logue by pairing strongly and specifically with the base C in syn-conformation leading to the formation of Hoogsteen type
DNA regardless of the identity of flanking bases.
base pairing rather than the standard Watson–Crick base
The results are in sharp contrast to aegPNA whereby a pairing.6 In the case of acpcPNA, however, the base I with
similar study revealed that the pairing properties of the base I flanking A or T shows the same preference for pairing to C
were highly dependent on the flanking bases.6 I in aegPNA with almost the same level of mismatch discrimination
prefers pairing to A when the flanking bases are T. On the observed in the base G. The same preference for pairing to C
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 2310–2317 | 2313

View Article Online
Paper
Organic & Biomolecular Chemistry
is also observed in sequences having the base I with flanking
C or G (Table 2). Among mismatched pairs, the smallest
decrease of Tm was generally observed when the base opposite
to the base I was A. In the worst case (pI·dA mismatch in
PNA5), the ΔTm was only −7.7 °C compared to the pI·dC pair
derived from the same PNA. This ΔTm figure is rather small
for acpcPNA, but can still be considered relatively large com-
pared to DNA and other analogues.
Other thermodynamic data demonstrate more detailed cir-
cumstances. As shown in Table 2, the thermodynamic data of
pI·dN duplexes and pG·dN duplexes (N = C, T, G, and A) share
a feature with significant similarity. Especially when N = T, all
the parameters of PNA2 with pI·dT mismatch pair are in abso-
lute agreement with those of PNA6 with pG·dT pair. The rela-
tively stable pG·dT mismatches in DNA duplexes form a
wobble base pairs in which 2-amino group of guanine does
Table 3 Comparison of Tm of DNA hybrids of hypoxanthine-containing PNA8
and guanine-containing PNA9
a
b
PNA (N→C)
AATTTRCATCA
N in DNA (5′→3′)
TGATGNAAATT
Tm
ΔTm
(°C)
Entry
̲
̲
(°C)
1
2
3
4
5
6
7
8
PNA8
(R = I)
C
T
G
A
C
T
G
A
74.9
53.0
49.1
49.9
71.7
55.6
46.8
41.1
N/A
−21.9
−25.8
−25.0
N/A
−16.1
−24.9
−30.6
PNA9
(R = G)
a Tm were measured at 1.0 μM duplex concentration, 10 mM sodium
phosphate buffer pH 7.0, 100 mM NaCl. b ΔTm = Tm (pR·dN) − Tm
(pR·dC), when R = I or G and N is the DNA base under consideration.
not contribute, and therefore pI·dT pairs are likely to be in a strong as the corresponding DNA hybrid of PNA9 (with pG·dC
corresponding base geometry. On the other hand in the case pair). Furthermore, the specificity of the pI·dC pairing in PNA8
of N is a purine base (A or G) and a large purine–purine base remains very high, and is more uniform (ΔTm −21.9 to −25.8
pair arise, the difference of enthalpy change from the values of for mismatched pairs) compared to that of pG·dC pairing in
the most stable pI·dC (or pG·dC) pair (ΔΔH°) was almost the PNA9 (ΔTm −16.1 to −30.6 for mismatched pairs). Circular
same. The ΔΔH° of pI·dG or pG·dG pair of PNA2, PNA3, PNA6 dichroism (CD) spectra of DNA hybrids of PNA8 and PNA9
and PNA7 was 25.0, 25.1, 26.4, and 27.5 kcal mol−1, respecti- suggested that the overall helical conformations of the
vely. When N = A, the decrease of enthalpic binding energy duplexes were similar according to their virtually identical CD
was similar or slightly smaller than that of pI·dG (or pG·dG) spectra (Fig. 3).
pair. These energetic similarity, including the case of N = T,
From the results obtained above, we may confidently
strongly suggest that the binding mode and the structure of propose that hypoxanthine can be used as a perfect mimic of
pI·dN pairs do not have drastic change from those of pG·dN guanine in acpcPNA under a variety of sequence contexts. The
pairs. Despite the lack of a hydrogen bond between 2-amino ability to change the base from G to I without compromising
group of purine and carbonyl oxygen of cytosine, enthalpic the binding affinity and sequence specificity of acpcPNA can
interaction of pI·dC pair is rather stronger than the pG·dC be considered very significant because there are a number of
case. This finding suggests the situation in which the pI·dC problems associated with G-rich acpcPNA sequences. First of
pair stands at appropriate position for superior base stacking, all, G monomer of acpcPNA is difficult to synthesize, requiring
compensating for the absence of hydrogen bond. This idea is expensive starting materials and long, low-yielding synthetic
completely consistent with our previous study, i.e., the differ- sequences.13 Secondly, DNA as well as PNA with G-rich
ence of the enthalpy change between G·C and A·T pairs was sequences are known to form secondary structures such as
rather small and the result came from favorable stacking G-quadruplexes18 that can prevent them form binding with
rather than hydrogen bonding for acpcPNA·DNA duplexes.11b
complementary DNA targets by standard Watson–Crick base
pairing. Since hypoxanthine cannot form the hydrogen bond
network in the same way as guanine, replacement of some
Potential applications of acpcPNA carrying hypoxanthine base
Although it has been known for some time that hypoxanthine guanine residues in G-rich sequences acpcPNA should disfavor
behave rather like guanine in DNA,3 the stability of the dI·dC the formation of G-quadruplexes, and thus facilitate the DNA
pair is generally less than that of the canonical dG·dC pair and binding. Thirdly, the unique redox properties of guanine may
the difference between the stability of dI·dC pair over other complicate certain applications such as electrochemical
“mismatched” pairs is not large enough to be generally useful. assays.19 Hence, the ability to replace G with its less readily oxi-
The strong affinity and high specificity of pI·dC pairing over dized counterpart without changing the base pairing property
pI·dG, pI·dA and pI·dT in acpcPNA system is therefore quite is essential for development of PNA or DNA probes to be used
unexpected and could lead to several practical applications. In in these applications.20 Finally, guanine and G·C base pair are
order to evaluate the generality of the strong and selective known to quench several common fluorescent labels.21,22 The
pairing between pI and dC in acpcPNA, the PNA8 and PNA9 quenching effect by guanine has been utilized for certain
were synthesized and their hybridization with various DNA applications23 but needed to be avoided in other situations. To
targets were compared. The only difference between PNA8 and demonstrate this effect, the fluorescence spectra of the 5(6)-
PNA9 is the base at the position 6 (I in PNA8 and G in PNA9) carboxyfluorescein-labeled PNA10 and PNA11 were compared.
with T/C as neighboring bases. Results from thermal denatura- The PNA10, with hypoxanthine as the immediate base adjacent
tion experiments (Table 3) confirmed that the complementary to the fluorescein label showed a significantly stronger fluor-
DNA hybrid of PNA8 containing pI·dC pair was at least as escence emission (1.54 times) compared to the PNA11 which
2314 | Org. Biomol. Chem., 2013, 11, 2310–2317
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 3 CD spectra of PNA8 (⋯) (a) and PNA9 (⋯) (b) together with their complementary DNA hybrids (––). The CD spectrum of the single stranded complementary
DNA (– –) was also included for comparison. Conditions: 1.0 μM PNA, 1.0 μM DNA, 10 mM sodium phosphate buffer pH 7.0, 100 mM NaCl.
NMR spectra were recorded on a Bruker Avance 400 NMR spec-
trometer operating at 400 MHz for 1H and 100 MHz for 13C.
MALDI-TOF mass spectra were obtained on a Microflex MALDI-
TOF mass spectrometry (Bruker Daltonics) using α-cyano-4-
hydroxycinnamic acid (CCA) as the matrix. High resolution
mass spectra were recorded in positive ESI mode on a Bruker
Daltonics micrOTOF mass spectrometer. IR spectra were
recorded on a Nicolet iS10 FT-IR spectrometer.
Oligonucleotides were purchased from Biodesign
(Pathumthani, Thailand) and were used as received. The con-
centration of oligonucleotides and acpcPNAs was determined
from the absorbance at 260 nm using the reported molar
extinction coefficients at 260 nm (ε260) for the corresponding
DNA monomers.7c The molar extinction coefficient of I was
obtained as described for deoxyinosine-containing DNA.3b
N-tert-Butoxycarbonyl-(4′R)-(6-chloro-9H-purin-9-yl)-
(2′R)-proline diphenylmethyl ester (2)
A mixture of tosylate 113 (0.5863 g, 1.06 mmol), 6-chloropurine
Fig. 4 Comparison of steady state fluorescence spectra of fluorescein-labeled
(0.246 g, 1.60 mmol) and anhydrous K2CO3 (0.35 g,
PNA10 (⋯) and PNA11 (––). Conditions: 0.05 μM PNA, 10 mM sodium phos-
2.50 mmol) in anhydrous DMF (5 mL) was stirred under nitro-
phate buffer pH 7.0, 100 mM NaCl, λex = 485 nm (λem = 522 nm).
gen at 80 °C for 8 h. Water (30 mL) was added to the reaction
and the suspension was extracted with dichloromethane (3 ×
bears guanine at the same position under similar conditions
(Fig. 4).
30 mL). The organic phase was washed with water, dried over
MgSO4 and evaporated to give the crude product, which was
purified by column chromatography eluting with hexanes :
ethyl acetate 1 : 3 on silica gel to obtain compound 2 as a white
amorphous solid (0.1583 g, 27%). TLC Rf = 0.12 (EtOAc :
hexanes 1 : 1); [α]D25 = +7.86 (c 1.02 in CHCl3); νmax/cm−1 2977,
1748, 1704, 1590, 1561, 1400; δH (400 MHz, CDCl3) 1.32
Experimental
General
Chemicals and solvents were purchased from commercial sup- and 1.49 (9H, s, CH3 Boc), 2.53 and 2.95 [2H, m CH2(3′)],
1
pliers and were used without further purification. H and 13C 4.00–4.16 and 4.58–4.69 [2H, m, CH2(5′)], 4.17 [1H, m, CH(2′)],
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 2310–2317 | 2315

View Article Online
Paper
Organic & Biomolecular Chemistry
5.17 [1H, m, CH(4′)], 6.85 (1H, s, CHPh2), 7.19–7.29 (m, 10H, Fmoc-solid phase peptide synthesis on a Tentagel S-RAM resin
Ph), 8.12 and 8.19 [1H, m, purine C(2)H rotamers], 8.64 [1H, s, carrying the acid-labile Rink amide linker.11 The A, T and C
purine C(8)H]; δC (100 MHz, CDCl3) 28.0 and 28.3 (CH3 Boc monomers and the spacer were incorporated as Fmoc/penta-
rotamers), 34.6 and 35.8 [C(3′)H2 rotamers], 50.0 and 50.6 fluorophenyl ester whereas the G and I monomers were incor-
[C(5′)H2 rotamers], 52.7 and 53.3 [C(4′)H rotamers], 57.5 [C(2′)H], porated via HATU activation of the corresponding Fmoc acid.
77.7 and 78.1 (CHPh2 rotamers), 81.2 (CCH3 Boc), 127.1, L-Lysinamide was included at the C-termini to improve water
128.4 and 128.5 (Dpm Ar CH), 131.8 (purine C5), 139.2 (Dpm solubility of the PNA. The PNA was end-capped by acetylation
C), 143.5 [purine C(8)H], 151.5 (purine C4), 151.7 [purine C(2)H], (PNA1–PNA9) or by reaction with 5(6)-carboxyfluorescein
153.4 and 153.6 (CO Boc rotamers), 170.4 (CO proline); N-succinimidyl ester. After ether precipitation and HPLC
HRMS (ESI-TOF) m/z calcd for C28H29ClN5O4Na [M + Na]+ purification, the identity and purity were determined by MAL-
556.1727; found 556.1768.
DI-TOF mass spectrometry and by reverse phase HPLC using
the following conditions: C18 column 4.6 × 50 mm, particle
size 3 μm; mobile phase A = 0.1% TFA in water; B = 0.1% TFA
in MeCN; gradient = 10–90% B over 25 min with 5 min equili-
N-9H-Fluoren-9-ylmethoxycarbonyl-(4′R)-[6-oxo-1H-purin-9
(6H)-yl]-(2′R)-proline (3)
Compound 2 (0.1583 g, 0.29 mmol) was dissolved in trifluoroa- bration time at a flow rate of 0.5 mL min−1
.
cetic acid containing 10% anisole (5 mL) at room temperature.
After 1 h, the solvent was removed by a gentle stream of nitro-
Thermal denaturation experiments
gen. The residue was triturated and washed with diethyl ether. UV melting experiments were performed on a CARY 100 UV
The white solid formed was collected by filtration and heated Spectrophotometer (Varian Inc., Australia) equipped with a
with 2 M HCl (10 mL) under reflux for 2 h. The solvent was Peltier temperature controller and a thermal analysis software.
removed by rotary evaporation and the residue dissolved in The samples for thermal denaturation measurement were pre-
1 : 1 H2O : MeCN (2 mL) and treated with solid NaHCO3 until pared by mixing calculated amounts of stock oligonucleotide,
pH 8. FmocOSu (0.223 g, 0.6 mmol) was then added in small PNA, buffer solutions and deionized water to give the final
portions with stirring. After stirring at room temperature over- concentration of PNA = 1.0 μM, DNA = 1.0 μM in 10 mM
night, the solvent was removed by rotary evaporation. The sodium phosphate pH 7.0 and 100 mM sodium chloride at a
residue was diluted with water (10 mL) and extracted with total volume of 1.0 mL. The sample in a 10 mm quartz cell
diethyl ether (3 × 20 mL). After purging the extracted aqueous with Teflon stopper was equilibrated at 20 °C for 10 min. The
layer to remove the dissolved ether with a gentle stream of N2,
A260 was recorded in steps from 20 to 90 °C with a temperature
the pH was adjusted to 2 with conc. HCl. The white precipitate increment of 1.0 °C min−1. The recorded temperature was cor-
was collected by filtration, washed with water, diethyl ether rected by a linear equation obtained from the relationship
and dried under vacuum to afford 3 as a white solid (0.109 g, between the recorded temperature and actual temperature
77%). TLC Rf = 0.14 (EtOAc : MeOH 9 : 1 containing 0.1% TFA); measured by a built-in temperature probe. Thermodynamic
[α]D24 = −7.04 (c 0.50 in DMF); νmax/cm−1 1698, 1450, 1420, parameters were obtained from melting curves by nonlinear
1352 cm−1
;
δH (400 MHz, DMSO-d6) δH 2.55–2.68 and curve fitting.24 The data obtained are average values from at
2.85–3.02 [2H, 2 × m, CH2(3′) rotamers], 3.80–3.88 [1H, m, 1 × least three melting curves for each PNA sample at a concen-
CH2(5′)], 4.10–4.34 and 4.52–4.56 [5H, m, 1 × CH2(5′), CH(2′), tration of PNA = DNA = 1.0 μM (ct = 2.0 μM). The ΔG° at 298 K
Fmoc CH and CH2 rotamers], 5.11 [1H, m, CH(4′)], 7.35 (2H, and Tm were calculated from the ΔH° and ΔS° values.
m, Fmoc Ar CH), 7.43 (2H, m, Fmoc Ar CH), 7.69 (2H, d J =
Fluorescence experiments
7.3 Hz, Fmoc Ar CH), 7.91 (2H, d J = 7.4 Hz, Fmoc Ar CH), 8.10
[1H, br m, hypoxanthine C(2)H], 8.22 [1H, s, hypoxanthine Fluorescence experiments were performed on a CARY Eclipse
C(8)H], 12.40 (1H, s, NH), 12.80 (1H, br s, COOH); δC Spectrofluorometer (Varian Inc., Australia). The samples were
(100 MHz, DMSO-d6) 33.8 and 34.9 [CH2(3′) rotamers], 46.6 prepared by mixing calculated amounts of stock PNA, buffer
(Fmoc aliphatic CH), 49.8 and 50.3 [C(5′)H2 rotamers], 51.9 solutions and deionized water to give the final concentration
and 52.5 [C(4′)H rotamers] 57.1 and 57.4 [C(2′)H rotamers], of PNA = 0.05 μM in 10 mM sodium phosphate pH 7.0 and
66.9 and 67.3 (Fmoc CH2 rotamers), 120.1 (Fmoc Ar CH), 124.2 100 mM sodium chloride at a total volume of 1.0 mL. The
(hypoxanthine C5), 125.2 (Fmoc Ar CH), 127.2 (Fmoc Ar CH), fluorescence spectra were measured in a 10 mm quartz cell
127.7 (Fmoc Ar CH), 138.8 [hypoxanthine C(8)H], 140.7 (Fmoc with Teflon stopper at 20 °C using the following instrument
Ar C), 143.6 (Fmoc Ar C), 145.6 [hypoxanthine C(2)H], 148.3 parameters: Excitation slit = 5 nm, Emission slit = 5 nm, Exci-
[hypoxanthine C(4)], 153.8 (Fmoc CO), 156.5 [hypoxanthine tation filter = Auto, Emission filter = Auto, Smoothing =
C(6)O], 172.4 and 172.9 (proline CO rotamers); HRMS Moving average.
(ESI-TOF) m/z calcd for C25H22N5O5 [M + H]+ 472.1621; found
472.1592.
CD spectroscopy
Circular dichroism experiments were performed on a JASCO
J810 Spectropolarimeter (Jasco, Japan). The samples were pre-
PNA synthesis
All acpcPNAs were synthesized from the Fmoc-protected pared by mixing calculated amounts of stock oligonucleotide,
acpcPNA monomers (A, T, C, G, I) and ACPC spacer by manual PNA, buffer solutions and deionized water to give the final
2316 | Org. Biomol. Chem., 2013, 11, 2310–2317
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
concentration of PNA = 1.0 μM, DNA = 1.0 μM in 10 mM
sodium phosphate pH 7.0 and 100 mM sodium chloride at a
total volume of 1.0 mL. The sample in a 10 mm quartz cell
with Teflon stopper was equilibrated at 20 °C for 10 min. The
CD spectra were measured at 20 °C from 200 to 400 nm with
scanning speed 200 nm min−1 and averaged 4 times then sub-
tracted from a spectrum of the blank (10 mM sodium phos-
phate buffer pH 7.0, 100 mM NaCl) under the same
conditions.
6 Z. Timár, S. Bottka, L. Kovács and B. Penke, Nucleosides
Nucleotides, 1999, 18, 1131.
7 (a) P. E. Nielsen, M. Egholm, R. H. Berg and O. Buchardt,
Science, 1991, 254, 1497; (b) M. Egholm, O. Buchardt,
P. E. Nielsen and R. H. Berg, J. Am. Chem. Soc., 1992, 114,
1895; (c) M. Egholm, P. E. Nielsen, O. Buchardt and
R. H. Berg, J. Am. Chem. Soc., 1992, 114, 9677.
8 R. E. Hudson, M. Goncharenko, A. P. Wallman and
F. Wojciechowski, Synlett, 2005, 1442.
9 (a) G. J. Sanjayan, V. R. Pedireddi and K. N. Ganesh, Org.
Lett., 2000, 2, 2825; (b) R. Vysabhattar and K. N. Ganesh,
Tetrahedron Lett., 2008, 49, 1314.
Conclusion
10 (a) K. F. MacKinnon, D. F. Qualley and S. A. Woski, Tetra-
hedron Lett., 2007, 48, 8074; (b) H. Challa, M. L. Styers and
S. A. Woski, Org. Lett., 1999, 1, 1639; (c) H. Challa and
S. A. Woski, Tetrahedron Lett., 1999, 40, 8333.
11 (a) T. Vilaivan and C. Srisuwannaket, Org. Lett., 2006, 8,
1897; (b) C. Vilaivan, C. Srisuwannaket, C. Ananthanawat,
C. Suparpprom, J. Kawakami, Y. Yamaguchi, Y. Tanaka and
T. Vilaivan, Artif. DNA PNA XNA, 2011, 2, 50;
(c) C. Suparpprom, C. Srisuwannket, P. Sangvanich and
T. Vilaivan, Tetrahedron Lett., 2005, 46, 2833.
We have demonstrated that hypoxanthine in acpcPNA showed
a remarkable preference for pairing with C over other natural
DNA bases (A, G, T) under a variety of sequence contexts. The
study revealed that the base G in acpcPNA can be replaced
with hypoxanthine without affecting the binding affinity and
specificity of pairing to its DNA target. This suggests potential
use of this strategy to solve many problems associated with
G-rich acpcPNA such as synthetic difficulty, formation of
stable secondary structures and fluorescence quenching.
12 C. Ananthanawat, T. Vilaivan, V. P. Hoven and X. D. Su,
Biosens. Bioelectron., 2010, 25, 1064.
13 G. Lowe and T. Vilaivan, J. Chem. Soc., Perkin Trans. 1, 1997,
547.
Acknowledgements
Financial support to this work from the Thailand Research
Fund (RTA5280002), the Thai government stimulus package 2
(TKK2555, SP2) under the Project for Establishment of Com-
prehensive Center for Innovative Food, Health Products and
Agriculture, the National Research University of CHE and the
Ratchadaphiseksomphot Endowment Fund (AM1006A), the
Ratchadaphiseksomphot Endowment Fund from Chulalong-
korn University (postdoctoral fellowship to WM) and the Thai-
land Research Fund and Chulalongkorn University through
the Royal Golden Jubilee Ph.D. Program (Grant No. PHD/0316/
2550, to NY and TV).
14 V. Antle and C. A. Caperelli, Nucleosides Nucleotides, 1999,
18, 1911.
15 (a) S. Hikishima, M. Isobe, S. Koyanagi, S. Soeda,
H. Shimeno, S. Shibuya and T. Yokomatsu, Bioorg. Med.
Chem., 2006, 14, 1660; (b) R. A. Tromp, R. F. Spanjersberg,
J. K. von Frijtag Drabbe Knzel and A. P. IJzerman, J. Med.
Chem., 2005, 48, 321; (c) B. R. Baker and P. M. Tanna,
J. Org. Chem., 1965, 30, 2857.
16 M. L. Peterson and R. Vince, J. Med. Chem., 1991, 34, 2787.
17 L. R. Rutledge and S. D. Wetmore, Phys. Chem. Chem. Phys.,
2012, 14, 2743.
18 S. Burge, G. N. Parkinson, P. Hazel, A. K. Todd and
S. Neidle, Nucleic Acids Res., 2006, 34, 5402.
19 E. Paleček, Electroanalysis, 1996, 8, 7.
References
1 F. H. C. Crick, J. Mol. Biol., 1996, 19, 548.
2 (a) F. H. Martin, M. M. Castro, F. Aboul-ela and I. Tinoco Jr,
20 J. Wang, G. Rivas, J. R. Fernandes, J. L. L. Paz, M. Jiang and
R. Waymire, Anal. Chim. Acta, 1998, 375, 197.
Nucleic Acids Res., 1985, 13, 8927; (b) E. Ohtsuka, 21 C. A. M. Seidel, A. Schulz and M. H. M. Sauer, J. Phys.
S. Matsuki, M. Ikehara, Y. Takahashi and K. Matsubara,
J. Biol. Chem., 1985, 260, 2605.
3 (a) S. C. Case-Green and E. M. Southern, Nucleic Acids Res.,
Chem., 1996, 100, 5541.
22 I. Nazarenko, R. Pires, B. Lowe, M. Obaidy and
A. Rashtchian, Nucleic Acids Res., 2002, 30, 2089.
1994, 22, 131; (b) N. E. Watkins Jr and J. SantaLucia Jr, 23 A. O. Crockett and C. L. Wittwer, Anal. Biochem., 2001, 290,
Nucleic Acids Res., 2005, 33, 6258.
89.
4 X. Sun and J. K. Lee, J. Org. Chem., 2010, 75, 1848.
5 A. A. Koshkin, J. Org. Chem., 2004, 69, 3711.
24 J. Kawakami, Y. Tanaka and K. Kishimoto, Nucleic Acids
Symp. Ser., 2009, 53, 227.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 2310–2317 | 2317