GNA/aegPNA chimera loaded with RNA binding preference

Article abstract of DOI:10.1002/asia.201100003

Pure and simple: A simple chimeric PNA (named chiPNA) was observed to selectively bind RNA in a 1:1 mixture of RNA and DNA with identical base sequences. This binding was assessed by chip-based assay.

Full text of DOI:10.1002/asia.201100003

COMMUNICATION
DOI: 10.1002/asia.201100003
GNA/aegPNA Chimera Loaded with RNA Binding Preference
Taedong Ok,[a] Joohee Lee,[a] Cheulhee Jung,[b] Jisoo Lim,[a] Chul Min Park,[a]
Joon-Hwa Lee,[c] Hyun Gyu Park,[b] and Hee-Seung Lee*[a]
Dedicated to Professor Eun Lee on the occasion of his retirement and 65th birthday
Peptide nucleic acid (PNA) is
an oligonucleotide mimic, in
which
the
sugar-phosphate
backbone is replaced by
a
pseudo-peptide skeleton, com-
posed of N-(2-aminoethyl)-gly-
cine (aeg) units carrying nucleo-
bases via methylene carbonyl
linkages (Figure 1a).[1] Owing
to its neutral amide backbone,
aegPNA has outstanding chemi-
cal and enzymatic stability as
well as high binding affinities to
DNA and RNA, which gives it
great potential for development
as an antisense agent or diag-
nostic tools in
a chip-based
Figure 1. Structural relationship among aegPNA(a), GNA(b) and the chimeric PNA with an acyclic chiral mo-
nomer (c).
assay.[2]
However, drawbacks such as poor cellular uptake, an abil-
[a] T. Ok, J. Lee, J. Lim, Dr. C. M. Park,+ Prof. Dr. H.-S. Lee
Molecular-Level Interface Research Center
Department of Chemistry
ity to form hybrids with both parallel and antiparallel orien-
tations, and similar binding affinity to DNA/RNA are
known to be subjects for improvement. In particular, the
latter two drawbacks may lead to poor specificity toward
the target sequence, which give rise to undesirable effects
both in biological and diagnostic applications. Although
there are some examples that report excellent binding selec-
tivity between RNA and DNA by chemical modification of
the aegPNA monomer, those approaches generally require
tedious synthetic routes to secure the conformationally rigid
cyclic monomers.[3,4]
KAIST
Daejeon 305-701 (Korea)
Fax : (+82)42-350-2810
E-mail: hee-seung_lee@kaist.ac.kr
[b] C. Jung, Prof. Dr. H. G. Park
Department of Chemical and Biomolecular Engineering
(BK 21 Program)
KAIST
Daejeon 305-701 (Korea)
[c] Prof. Dr. J.-H. Lee
On the other hand, glycol nucleic acid (GNA) is a struc-
turally simplified oligonucleotide mimic, in which the nega-
tively charged phosphate backbone is intact, but the C5
cyclic sugar ring has been replaced by C3 acyclic chiral pro-
pylene glycol (Figure 1b).[5a] Homooligomers of GNA are
known to pair very well with themselves through Watson–
Crick base paring, but do not cross-pair with either RNA or
DNA.[5b,c]
Department of Chemistry and RINS
Gyeongsang National University,
Jinju, Gyeongnam 660-701 (Korea)
[+] Current adress:
Korea Research Institute of Chemical Technology
P.O. Box 107, Sinseongno 19, Yuseong-gu
Daejeon 305-600 (Korea)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201100003.
1996
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 1996 – 1999

Table 1. Tm (8C) of PNA:RNA and PNA:DNA Duplexes.[a]
We envisioned that the hy-
bridization of the aforemen-
tioned sugar-free oligonucleo-
tide mimics (aegPNA and
GNA) would generate a chi-
meric PNA (chiPNA) chain
with unique properties, such as
binding kinetics, selectivity, and
duplex stability. For example,
[d]
Entry
PNA
Sequences
RNA[b]
DNA[c]
DTm
1
2
3
4
5
6
7
8
P1(aegPNA)
G
H2N-GTAGATCACT-Lys-CONH2
H2N-GTAGATCACT-Lys-CONH2
H2N-GTAGAT*CACT-Lys-CONH2
H2N-GTAGATCACT-Lys-CONH2
H2N-Lys-AGGGTCGCTCGGTGT-CONH2
H2N-Lys-AGGGTCGCTCGGTGT-CONH2
H2N-Lys-AGGGTCGCTCGGTGT-CONH2
H2N-Lys-AGGGTCGCTCGGTGT-CONH2
58
46
33
58
42
24
+0
+4
+9
+6
+1
+4
+8
+8
(R)-P1a
(S)-P1a
P1b
43
37
P2
N
91 (77)
88 (70)
84 (59)
82 (62)
90 (77)
84 (65)
76 (54)
74 (nb)
P2a
P2b
P2c
the incorporation of
a few
[a] The UV melting curve was measured for a duplex (5 mm) in a sodium phosphate buffer (10 mm, pH 7.0)
containing 100 mm NaCl and 0.1 mm EDTA. [b] RNA1, 5’-AGUGAUCUAC-3’(ap); RNA2, 5’-ACACC-
GAGCGACCCU-3’(ap); RNA3, 5’-UCCCAGCGAGCCACA-3’(p). [c] DNA1, 5’-AGTGATCTAC-3’(ap);
DNA2, 5’-ACACCGAGCGACCCT-3’(ap); DNA3, 5’-TCCCAGCGAGCCACA-3’(p). [d] DTm is the differ-
ence in melting temperature between PNA:RNA and PNA:DNA. A, T, G, and C are aegPNA bases; T, (R)-
g3T; T*, (S)-g3T; Values in parentheses are Tm of parallel duplexes; nb, no binding.
GNA monomers into an
aegPNA backbone may induce
preorganized helical propensity,
which in turn may result in the
improvement of the antiparallel
binding preference of parent
aegPNA (Figure 1c). In order to test our hypothesis, the
synthesis of a new monomer (g3T), a chiral g-amino acid
with a thymine base that resembles the corresponding GNA
monomer, was required. Herein we report the streamlined
synthesis of the monomer and the unexpected binding selec-
tivity of the chiPNAs to the complementary RNA and
DNA.
The Fmoc-protected (R)-thymine-substituted monomer 5
was readily prepared from (R)-azidomethyl g-butyrolactone
1 in 5 steps and 56% overall yield (Scheme 1).[6] g-Lactone
1 was subjected to a ring-opening reaction with trimethylsi-
lylbromide (TMSBr) and ethanol to give the bromide 2. A
substitution reaction of 2 with N3-benzoyl thymine afforded
3. The benzoyl protecting group of 3 was removed during
the ensuing saponification reaction which gave the carboxyl-
ic acid 4. Reduction of the azide in 4 with PMe3/H2O fol-
lowed by Fmoc-protection produced the desired monomer
(R)-g3T (5) in an enantiomerically pure form.
backbone (Table 1).[7] The chiPNA oligomers were synthe-
sized by using automated solid-phase synthesis. The g3T
monomers were readily incorporated into aegPNA oligo-
mers without any further modification of typical solid-phase
synthesis method (see the Experimental Section and the
Supporting Information).
The duplex stability of chiPNA with their complementary
nucleic acids (RNA1-3 and DNA1-3) were measured by
temperature-dependent UV melting experiments at 260 nm
(Table 1). In most cases, the absolute Tm values of chiPNAs
(Table 1, entries 2–4, 6–8) were somewhat lower than those
of the parent aegPNAs (Table 1, entries 1 and 5). The re-
duced binding affinity might be attributed to the partial dis-
ruption of the g3T-A pair, but they still seem to interact with
each other through Watson–Crick base-paring as the mis-
matched cases showed much-lower Tm values (see the Sup-
porting Information, Table S2). A hydrogen-exchange NMR
spectroscopic study revealed that the g3T-A base pair was
much more unstable than the corresponding base pairs in
aegPNA:DNA and DNA:DNA pairs.[8] This study also
showed that the g3T-A base-pairing in the chiPNA:DNA
hybrid led to destabilization and fast opening of the neigh-
boring base pairs in an asymmetric manner compared to the
corresponding base pairs in the aegPNA:DNA hybrid.[8] It
should also be noted that Tm reduction in chiPNA is less sig-
nificant in the P2 series than in the P1 series. All g3T-A
A series of chiPNA chains were prepared in which the
g3T monomers were partially incorporated into the aegPNA
À
À
pairs were located between the A T and G C pairs in the
À
P1 series, and between both G C pairs in the P2 series
(Table 1, underlined), thus suggesting that the local stability
differences in the neighboring base pairs seem to be respon-
sible for the difference in the overall duplex stability. These
results are also consistent with the observation that the hy-
drogen-exchange study could not be carried out when the
3
À
g T-A base was in a neighboring position to A T base pairs
in the chiPNA:DNA hybrid. Besides, the structural pertur-
bation in the aegPNA skeleton by the insertion of g3T is ex-
pected to be more severe in the shorter oligomers (P1
series). Thus, both factors (oligomer length and the neigh-
boring base pairs around a g3T-A) should be considered
when designing the sequence of a chiPNA chain. The
duplex stabilities for parallel binding of chiPNA chains were
Scheme 1. Synthesis of (R)-g3T monomer. (a) TMS-Br, EtOH, CH2Cl2,
97% yield. (b) N3-Bz-Thymine, K2CO3, 18-crown-6, DMF, 90%
(c) LiOH·H2O, THF/MeOH/H2O (6:3:1), 92%, (d) PMe3, THF/H2O
(e) Fmoc-OSu, Na2CO3, acetone/H2O (2:1), 70% from 4. Bz=benzoyl,
Fmoc-OSu=N-(9-Fluorenylmethoxycarbonyloxy) succinimide, DMF=
N,N-dimethylformamide, THF=tetrahydrofuran.
Chem. Asian J. 2011, 6, 1996 – 1999
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1997

H.-S. Lee et al.
COMMUNICATION
also examined for the P2 series. Compared to those of anti-
parallel duplex, the Tm values of parallel binding were low-
ered more significantly in the chiPNAs than in the parent
aegPNA.
DTm) values obtained from the conventional UV-melting ex-
periment, which have been typically used to assess relative
thermodynamic stability for bimolecular association, may
not be directly correlated with actual binding selectivity. To
address this issue, a competitive-binding assay was per-
formed using a chip-based experiment to obtain the quanti-
tative binding selectivity of chiPNA chains (see the Support-
ing Information, Table S4). Surprisingly, P2c binds RNA 3.5
times more selectively than P2b, even though no difference
existed between their DTm values. This result was not ex-
pected from the conventional melting temperature measure-
Interestingly, a trend was observed from the UV-melting
experiments, in which the incorporation of g3T into an
aegPNA skeleton induced an improved binding preference
for RNA over DNA at the expense of the duplex stability
(see the DTm values in Table 1). The CD profiles of single-
stranded (R)-P1a and P1b containing (R)-g3T revealed that
even single- or double-mutation could induce the formation
of a typical RNA-type right-handed helical conformation,
whereas single-stranded (S)-P1a containing (S)-g3T gave a
left-handed helix (CD signal inverse of (R)-P1a, Figure 2a).
ments. Furthermore, chiPNA
ACHTUNGTRENNUNG
selectively than natural nucleic acidACHTUNGTRENNNUG
(30 and 20 times more, respectively), whereas Cy5-DNA2
binding of P2c was not detectable by the naked eye. Mean-
while, DNA4 and P2 did not discriminate between RNA
and DNA in the assay (see Figure 3a,b).
Figure 2. (a) CD spectra of single-stranded chiPNAs (5 mm each); (R)-
P1a and P1b are chiPNAs with (R)-g3T; (S)-P1a is a chiPNA with (S)-
g3T. (b) The selected CD spectra of chiPNA/RNA duplexes.
Figure 3. (a) The RNA binding selectivity of the capture probes are de-
noted on the bars and the values are the ratio of fluorescence intensity
between Cy3-RNA2 and Cy5-DNA2. (b) Fluorescence images of the
competition binding assay measured at 635 nm (A) and 532 nm (B);
DNA4, 5’-(H2N-(CH2)6)-AGGGTCGCTCGGTGT-3’.
The CD signals of a 1:1 mixture of chiPNA (P2a or P2b)
and RNA2 were similar to those of the aegPNA P2:RNA2
duplex (Figure 2b), thereby indicating that chiPNA forms a
duplex with RNA in a similar fashion. The results of both
the Tm and CD studies show that acyclic and structurally
simple chiral g3T monomers can induce RNA-like preorgan-
ized helical conformations, which gives the chiPNA chain
the ability to discriminate RNA from DNA as well as anti-
parallel from parallel binding.
Based on the results of these experiments, we reasoned
that the remarkable RNA binding selectivity of the chiPNAs
was a consequence of the adoption of RNA-like preorgan-
ized conformation, which greatly reduced the entropic cost
of binding to RNA compared to DNA. As a result, the asso-
ciation rate of chiPNA:RNA duplex formation became ac-
celerated and the dissociation rate of the duplex was retard-
ed, compared to chiPNA:DNA duplex.[9] This result also
confirms that the binding affinity of the 15-mer chiPNA
chain was high enough to bind and discriminate single-
Contrary to our expectations, P2b and P2c showed simi-
lar Tm and DTm values, irrespective of the number of muta-
tions. This finding gives rise to the possibility that the Tm (or
1998
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 1996 – 1999

GNA/aegPNA Chimera Loaded with RNA Binding Preference
buffer containing 0.01% Triton X-100 and 1X SSPE (150 mm NaCl,
10 mm NaH2PO4, 1 mm EDTA) to make final concentration of 10 nm
each. A 1:1 mixture of the probe solution/buffer was applied on a chip.
The hybridization reactions were performed at 378C for 4 h. After hy-
bridization, the array was rinsed with washing buffer containing 0.005%
Triton X-100 and 6X SSPE (0.9m NaCl, 60 mm NaH2PO4, 6 mm EDTA,
pH 7.4) for 10 min at room temperature. The binding selectivity was de-
termined from the ratio of the fluorescence intensity measured.
stranded nucleic acids, despite the reduced Tm compared to
the parent aegPNA.
In summary, we have found a very simple strategy to im-
prove binding preference of aegPNA to RNA/DNA. In-
spired by the structure of the GNA monomer, a highly
simple chiral PNA monomer was designed. By decorating
the achiral aegPNA backbone with a few chiral acyclic g-
amino acid molecules, we were able to make a chimeric
PNA with better RNA selectivity as well as antiparallel se-
lectivity. To the best of our knowledge, this is the simplest
monomeric unit that induces aegPNA to discriminate be-
tween RNA and DNA. Finally, we demonstrated that the
chip-based competition-binding assay is an alternative tool
to quantitatively analyze binding selectivities of modified
PNAs. We believe that chiPNA will expand the utility of
PNAs and find diverse applications related to RNA targets
in the future.[10]
Acknowledgements
This work was supported by a Korea Research Foundation Grant funded
by the Korea Government (MOEHRD, Basic Research Promotion Fund,
KRF-2005-205-C00043) and by the Basic Science Research Program
through the National Research Foundation of Korea (NRF-2010-
0001953; 2010-0014199; NRF-C1ABA001-2010-0020480), funded by the
Ministry of Education, Science and Technology.
Keywords: chimera · GNA · PNA · RNA · selectivity
Experimental Section
Temperature-Dependent UV-Melting Experiment
[1] a) P. E. Nielsen, M. Egholm, R. H. Berg, O. Buchardt, Science 1991,
254, 1497; b) P. E. Nielsen, Acc. Chem. Res. 1999, 32, 624.
[2] a) L. Good, P. E. Nielsen, Antisense Nucleic Acid Drug Dev. 1997, 7,
431; b) D. A. Braasch, D. R. Corey, Biochemistry 2002, 41, 4503.
[3] For recent reviews, see: a) V. A. Kumar, K. N. Ganesh, Curr. Top.
Med. Chem. 2007, 7, 715 and references therein; b) V. A. Kumar,
K. N. Ganesh, Acc. Chem. Res. 2005, 38, 404. For a recent example
of RNA selective binding oligonucleotide by triplex formation, see:
c) K. Gogoi, A. D. Gunjal, V. A. Kumar, Chem. Commun. 2006,
2373.
Tm measurements were performed on a CARY 300 UV Spectrophotome-
ter (Varian Ltd.) equipped with a thermal melt system. The samples were
prepared by mixing known amounts of stock oligonucleotide and PNA
solutions together to give final volumes, which were adjusted to 3 mL by
addition of distilled water. The samples were transferred to a 10 mm
quartz cell with Teflon stopper. The OD260 was recorded in steps from 5~
958C (block temperature) with a temperature increment of 0.58CminÀ1
.
The results were normalized by dividing the absorbance at each tempera-
ture by the initial absorbance. The Tm values were calculated from the
first derivative of the average heating and cooling curves.
[4] a) T. Govindaraju, V. A. Kumar, K. N. Ganesh, J. Am. Chem. Soc.
2005, 127, 4144; b) T. Govindaraju, V. A. Kumar, K. N. Ganesh, J.
Org. Chem. 2004, 69, 1858.
CD Analysis
CD spectra of the oligonucleotides were recorded on a JASCO-700 spec-
trometer. All samples were scanned over the wavelength range of 300 to
200 nm at a scan speed of 50 nmminÀ1 at room temperature using 0.1 cm
path length quartz cuvette, a resolution of 0.1 nm, a bandwidth of 1.0 nm,
sensitivity of 50 mdeg and a response of 1 s. Each measurement was re-
peated 5 times and the average taken.
[5] a) L. Zhang, A. Peritz, E. Meggers, J. Am. Chem. Soc. 2005, 127,
4174; b) M. K. Schlegel, A. E. Peritz, K. Kittigowittana, L. Zhang,
E. Meggers, ChemBioChem 2007, 8, 927; c) M. K. Schlegel, X. Xie,
L. Zhang, E. Meggers, Angew. Chem. 2009, 121, 978; M. K. Schlegel,
X. Xie, L. Zhang, E. Meggers, Angew. Chem. 2009, 121, 978; Angew.
Chem. Int. Ed. 2009, 48, 960.
[6] T. Ok, A. Jeon, J. Lee, J. H. Lim, C. S. Hong, H. S. Lee, J. Org.
Chem. 2007, 72, 7390.
Chip-Based Competition Binding Study
Four different capture probes (DNA4, P2, P2b, and P2c) were diluted
with a 0.1m carbonate buffer (pH 9.5) containing 10% (v/v) dimethyl
sulfoxide to a concentration of 25 mm each. The capture probes were
spotted on the same aldehyde-activated glass (CEL-Associates, Houston,
TX) using a VersArray ChipWriterTM Compact system (Bio-Rad Labora-
tories Inc., California, USA) according to the manufacturerꢁs instructions
and were allowed to bind covalently on the glass surface overnight in a
humid chamber. To verify the reproducibility, six spots were printed for
each type of capture probe. After incubation, the glass was washed twice
with 0.2% SDS (sodium dodecyl sulfate) and twice with distilled water,
submerged in water at 958C for 2 min, and transferred into a sodium bor-
ohydride solution (2.5 mg NaBH4 dissolved in 750 mL of phosphate-buf-
fered saline and 250 mL of 100% ethanol) for 5 min to reduce the Schiff
base and the free aldehydes. The slides were then rinsed three times with
0.2% SDS and twice with distilled water. Finally, the slides were air-
dried at room temperature and stored at 48C in a vacuum desiccator
until use.[11] To obtain the signals, the microarrays were scanned with an
Axon GenePix 4000B fluorescence reader (Axon, Union City, CA).
Laser power and photomultiplier tube (PMT) were set at 100 and 600, re-
spectively. The obtained 16 bit TIFF images were analyzed with GenePix
Pro 3.0 (Axon) software. The target probes Cy5-DNA2 (Genotech,
Korea) and Cy3-RNA2 (Bioneer, Korea) were diluted with hybridization
[7] We selected two mixed-base sequences with similar Tm between
PNA:RNA and PNA:DNA duplexes: the well-studied Nielsen se-
quence (P1) and a sequence known to inhibit Luciferase activity
(P2). For P1, see: a) B. Hyrup, M. Egholm, P. E. Nielsen, P. Wittung,
B. Norden, O. Buchardt, J. Am. Chem. Soc. 1994, 116, 7964; b) J. K.
Pokorski, M. A. Witschi, B. L. Purnell, D. H. Appella, J. Am. Chem.
Soc. 2004, 126, 15067. For P2, see: c) D. F. Doyle, D. A. Braasch,
C. S. Simmons, B. A. Janowski, D. R. Corey, Biochemistry 2001, 40,
53.
[8] Y.-J. Seo, J. Lim, E.-H. Lee, T. Ok, J. Yoon, J.-H. Lee, H.-S. Lee,
Nucl. Acids Res. 2011, submitted.
[9] a) T. Ratilainen, A. Holmen, B. Norden in Peptide Nucleic Acids:
Protocols and Applications, 2nd ed. (Ed.: P. E. Nielsen), Horizon Bio-
science, Norfolk, 2004, Chap. 4; b) K. K. Jensen, H. Orum, P. E.
Nielsen, B. Norden, Biochemistry 1997, 36, 5072.
[10] a) A. W. Wark, H. J. Lee, R. M. Corn, Angew. Chem. 2008, 120, 654;
Angew. Chem. Int. Ed. 2008, 47, 644; b) E. A. Hunt, A. N. Goulding,
S. K. Deo, Anal. Biochem. 2009, 387, 1.
[11] S. C. Yim, H. G. Park, H. N. Chang, D. Y. Cho, Anal. Biochem. 2005,
337, 332.
Received: January 4, 2011
Published online: April 5, 2011
Chem. Asian J. 2011, 6, 1996 – 1999
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1999