5-(pyren-1-yl)uracil as a base-discriminating fluorescent nucleobase in pyrrolidinyl peptide nucleic acids

Article abstract of DOI:10.1002/asia.201100490

A pyrene-labeled uridine (U^Py) monomer for a pyrrolidinyl peptide nucleic acid with an alternating proline/2-aminocyclopentanecarboxylic acid backbone (acpcPNA) was synthesized and incorporated into the PNA. The U^Py base in acpcPNA c

Full text of DOI:10.1002/asia.201100490

DOI: 10.1002/asia.201100490
5-(Pyren-1-yl)uracil as a Base-Discriminating Fluorescent Nucleobase in
Pyrrolidinyl Peptide Nucleic Acids
Chalothorn Boonlua,[a] Chotima Vilaivan,[a] Hans-Achim Wagenknecht,*[b] and
Tirayut Vilaivan*[a]
Abstract:
A
pyrene-labeled uridine
ments. The fluorescence of the UPy-
containing single-stranded acpcPNA
was very weak in aqueous buffer. In
the presence of a complementary DNA
target, the fluorescence was enhanced
significantly (2.7–41.9 folds, depending
on sequences). The fluorescence en-
hancement was specific to the pairing
between UPy and dA, making the UPy-
modified acpcPNA useful as a hybridi-
zation-responsive fluorescence probe
for DNA-sequence determination.
(UPy) monomer for a pyrrolidinyl pep-
tide nucleic acid with an alternating
proline/2-aminocyclopentanecarboxylic
acid backbone (acpcPNA) was synthe-
sized and incorporated into the PNA.
The UPy base in acpcPNA could specif-
ically recognize the base A in its com-
plementary DNA strand as determined
by thermal denaturation (Tm) experi-
Keywords: electron transfer · fluo-
rescence · molecular beacons · oli-
gonucleotides
acids
·
peptide nucleic
Introduction
entirely re-designed nucleobase surrogate or a canonical nu-
cleobase that is modified with a fluorophore. The latter cate-
gory is more attractive since the pairing behavior of these
nucleobases is, at least in principle, more easily predicted.
The nature of the fluorophore itself as well as the linker
joining the fluorophore and the nucleobase is very impor-
tant in determining the fluorescence properties. In DNA,
nucleobases that are modified by a fluorophore, such as
pyrene through a flexible linker, are usually quenched upon
duplex formation as a result of intercalation.[4–7] On the
other hand, the fluorescence of the pyrene-modified nucleo-
base through a rigid linker often increases under similar sit-
uations, thus generating a more-desirable turn-on signal.[8–11]
In addition to pyrene, other fluorophores that have been
successfully incorporated into a DNA nucleobases and yield
turn-on signals upon duplex formation include perylene,[12]
anthracene,[13,14] and fluorene.[15,16] In all cases, there are one
or more of the following limitations: the fluorescence en-
hancement was modest, the fluorescence change was highly
dependent on the flanking bases, and the degree of mis-
match discrimination was far from perfect.
Fluorescent DNA-base analogues are potentially useful as
probes for DNA-sequence determination as well as physico-
chemical studies of DNA, such as electron transport. Al-
though a number of such fluorescent bases have been re-
ported,[1] not all of them show a significant fluorescence
change upon pairing with their complementary nucleobases.
The term base discriminating fluorescent (BDF) nucleo-
sides/nucleobases was introduced by Saito and co-workers
to describe such behavior.[2,3] BDF nucleobases can be an
[a] C. Boonlua, C. Vilaivan, Dr. T. Vilaivan
Organic Synthesis Research Unit
Department of Chemistry, Faculty of Science
Chulalongkorn University
Phayathai Road, Patumwan, Bangkok 10330 (Thailand)
Fax : (+662)2187598
E-mail: vtirayut@chula.ac.th
[b] Prof. Dr. H.-A. Wagenknecht
Institute for Organic Chemistry
Karlsruhe Institute of Technology
76131 Karlsruhe (Germany)
The optical properties of 5-(pyren-1-yl)uracil (UPy) deoxy-
riboside have been extensively studied, both as the mono-
mer[17–19] and when incorporated into DNA.[20,21] The indif-
ferences of the Tm and fluorescence among DNA duplexes
carrying different nucleobases opposite to the UPy suggest
that this strongly electronically coupled fluorophore-nucleo-
base does not form typical Watson–Crick-type base-pairing
Old address: Institute for Organic Chemistry
University of Regensburg
93040 Regensburg (Germany)
Fax : (+662)2187598
E-mail: hans-achim.wagenknecht@kit.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201100490.
Chem. Asian J. 2011, 6, 3251 – 3259
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3251

FULL PAPERS
Scheme 1. Synthesis of the Fmoc-protected UPy acpcPNA monomer (1). DMF=N,N-dimethylformamide, Fmoc=fluorenylmethyloxycarbonyl.
with adenine. Accordingly, its application as a BDF in DNA
system is limited. On the other hand, there is no report on
the pairing and optical properties of this particular UPy base
in other DNA analogues, including peptide nucleic acid
(PNA). We became interested in designing a quencher-free,
singly-labeled PNA beacon[22] using the BDF concept based
on our new pyrrolidinyl PNA with a proline/2-aminocyclo-
pentanecarboxylic acid backbone (acpcPNA).[23,24] Herein,
we report the synthesis of UPy-modified acpcPNA and dem-
onstrate that the UPy base in acpcPNA can specifically rec-
ognize adenine in DNA targets. Most importantly, the fluo-
rescence of the UPy-modified acpcPNA increased significant-
ly in response to the specific base pairing event between UPy
and dA.
Results and Discussion
Synthesis of the UPy Monomer and UPy-Containing PNA
The UPy acpcPNA monomer (1) was synthesized in an anal-
ogous way to the corresponding DNA nucleoside, as out-
lined in Scheme 1.[19,25,26] Suzuki–Miyaura cross-coupling be-
tween the commercially available pyren-1-boronic acid and
the protected iodouracil acpcPNA monomer (2) gave the
Boc-protected intermediate (4; Boc=tert-butoxycarbonyl),
which, after protecting-group exchange, afforded the desired
Fmoc-protected UPy acpcPNA monomer 1. Compound 2
was obtained from a Mitsunobu reaction between N3-benzo-
yl-5-iodouracil and the known Boc/diphenylmethyl-protect-
ed trans-hydroxy-d-proline intermediate.[27] The UPy acpcP-
NA monomer 1 was successfully incorporated into internal
positions of various acpcPNA sequences (PNA1–PNA8)
through standard O-(7-azabenzotriazol-1-yl)-N,N,N’,N’-tetra-
methyluronium hexafluorophosphate (HATU) coupling.[23]
The identity of all UPy-modified acpcPNA was confirmed by
MALDI-TOF mass spectrometry (see the Supporting Infor-
mation, Table S1).The sequences of all acpcPNA and DNA
used in this study are shown in Scheme 2.
Abstract in Thai:
Thermal Stabilities and Optical Properties of DNA Hybrids
with Homothymine acpcPNA Carrying UPy Modification
Initially, the ability of the UPy residue in acpcPNA to recog-
nize adenine was evaluated in homothymine acpcPNA. One
thymine residue in acpcPNA with a T9 sequence was re-
placed by a UPy base to give PNA1. Thermal stability and
fluorescence properties data of PNA1 and its DNA hybrids
are summarized in Table 1. Thermal denaturation experi-
ment of hybrids between PNA1 and DNA1a-DNA1d
(dA4XA4, X=A, T, C, and G, respectively) clearly indicated
that the UPy base could specifically recognize the adenine
base. The melting temperature (Tm) of the PNA1·DNA1a
3252
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 3251 – 3259

Base-Discriminating Fluorescent Nucleobase in Pyrrolidinyl Peptide Nucleic Acids
Scheme 2. Structure of acpcPNA and sequences of PNA/DNA used in this study.
Table 1. Melting temperatures, quantum yields, and fluorescence intensity
able difference in the thermal stabilities between the com-
ratios of PNA1–PNA5 and their hybrids with DNA.
plementary and mismatched hybrids of more than 358C.
The absorption spectrum of PNA1 (Figure 1) showed a
broad pyrene absorption band with a lmax of 345 nm in addi-
tion to the normal nucleobase absorption in the region of
260 nm. The pyrene absorption band showed no fine struc-
ture, except for a shoulder at ca. 340 nm which was consis-
tent in shape with the absorption spectra of DNA carrying a
UPy base reported previously.[17,18,21] The shape and magni-
tude of the pyrene absorption band changed noticeably
upon addition of the complementary DNA target (DNA1a).
The absorption maxima was broadened and shifted slightly
to a longer wavelength (350 nm). This shift was also accom-
panied by a small hyperchromism. No significant change in
Entry System[a]
Tm
[8C]
FF
I465
I465(ss)
I
(comp)[d]
I465
Notes
[b]
[c]
1
2
PNA1
–
0.006 1.0
41.9
1.0
ss, TUPy
T
ds, complemen-
tary
ds, pUPy·dT
mismatched
ds, pUPy·dC
mismatched
ds, pUPy·dG
mismatched
ss, AUPy
A
ds, complemen-
tary
ss, GUPy
G
ds, complemen-
tary
ss, CUPy
C
ds, complemen-
tary
ss, CUPy
C
PNA1+DNA1a 65.8 0.131 41.9
PNA1+DNA1b 25.7 0.022 5.3
PNA1+DNA1c 27.3 0.039 8.7
PNA1+DNA1d <20 0.013 3.0
3
4
5
14.0
6
7
PNA2
PNA2+DNA2
–
0.011 1.0
60.2 0.051 6.2
8
9
PNA3
PNA3+DNA3
–
0.006 1.0
54.1 0.032 8.3
10
11
PNA4
PNA4+DNA4
–
0.006 1.0
<20 0.010 2.0
12
13
PNA5
PNA5+DNA5
–
0.009 1.0
72.7 0.048 6.8
ds, complemen-
tary
[a] Conditions: [PNA]=2.5 mm; [DNA]=3.0 mm, 10 mm sodium phosphate
buffer, pH 7.0, lex 350 nm. [b] Measured using quinine sulfate as a standard.
[c] Ratio of fluorescence intensity at 465 nm to single-stranded PNA. [d] Ratio
of fluorescence intensity at 465 nm to complementary hybrids.
pair (pUPy·dA) was 65.88C. This value is considerably small-
er than the corresponding unmodified T9 acpcPNA·DNA1a
pair under identical conditions (Tm =82.08C), probably
owing to the unfavorable steric effects of the bulky pyrene
group. However, the remaining three PNA1·DNA pairs car-
rying a single base mismatch yielded either much-lower Tm
values (25.7 and 27.38C for pUPy·dT, pUPy·dC) or no sigmoi-
dal transition at all (pUPy·dG). This result suggests a remark-
Figure 1. UV/Vis spectra of PNA1 (2.5 mm)+DNA1a–DNA1d (3.0 mm) in
10 mm sodium phosphate buffer pH 7.0.
Chem. Asian J. 2011, 6, 3251 – 3259
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
3253

H. A. Wagenknecht, T. Vilaivan et al.
FULL PAPERS
the absorption spectra in the pyrene absorption region could
be observed with the three remaining mismatched hybrids.
The change in the absorption spectra indicated a ground-
state interaction between the UPy unit and the surrounding
nucleobases, which took place only when the perfect base-
pairing between the PNA and the DNA was formed. The
change in the absorption spectra of PNA1 upon hybridiza-
tion with DNA is fully reversible (Figure 2). At tempera-
tures above the Tm, the absorption spectra in the pyrene
chromophore region of PNA1·DNA1a was similar to that of
the single-stranded PNA1 and, after cooling, the absorption
spectra changed again.
Figure 3. Steady-state fluorescence spectra of PNA1 (2.5 mm)+DNA1a–
DNA1d (3.0 mm) in 10 mm sodium phosphate buffer pH 7.0, lex 350 nm.
difference between the fluorescence of the complementary
and non-complementary hybrids of PNA1, as well as the
single-stranded PNA1, could be visualized by the naked eye
under UV light irradiation (365 nm; Figure S17).
Effect of Neighboring Bases on the Fluorescence of DNA
Hybrids with Homothymine acpcPNA Carrying UPy
Modification
Figure 2. Temperature-dependent
UV/Vis
spectra
of
PNA1
(2.5 mm)+DNA1a (3.0 uM) in 10 mm sodium phosphate buffer pH 7.0,
from 25 to 958C (solid line) and back to 258C (dotted line), temperature
step=108C.
The fluorescence of BDF nucleobases is often strongly influ-
enced by neighboring bases. To investigate the effect of
neighboring bases on the fluorescence properties of acpcP-
NA carrying UPy modification, PNA2–PNA5 which had dif-
ferent bases adjacent to UPy were used as models. The data
on thermal stabilities and fluorescence properties are sum-
marized in Table 1 and Figure 4. As with PNA1, in all cases
the fluorescence of the single-stranded PNAs was very low
(FF =0.006–0.011). PNA2, which had a A/A neighboring
base, was the brightest, with a fluorescence approximately 2
fold stronger than PNA3 and PNA4. Not unexpectedly, the
fluorescence of the hybrids with corresponding complemen-
tary DNAs is quite sensitive to the nature of the neighbor-
ing bases. Nevertheless, the increase in fluorescence of the
UPy upon hybrid formation could be observed with A/A and
G/G as neighboring bases (PNA2 and PNA3). The seeming
absence of fluorescence increase with C/C neighboring bases
in PNA4 was attributed to the low stability of acpcPNA hy-
brids containing multiple TC/CT steps, as revealed by the
absence of a sigmoidal thermal denaturation curve (see the
Supporting Information, Figure S13). To circumvent this
problem, the fluorescence experiment was repeated with the
13 mer hybrid PNA5·DNA5 having two pT·dA pairs extend-
ed from both termini to increase the stability. This hybrid
had a Tm of 72.78C and displayed a 6.8-fold increase in fluo-
Most interestingly, the fluorescence of the UPy-containing
acpcPNA is highly sensitive to its hybridization state
(Figure 3). The single-stranded PNA1 exhibited a broad and
unstructured fluorescence peak owing to the UPy base that is
forming an exciplex state when excited at 350 nm, thereby
indicating strong electronic coupling between the pyrene
chromophore and the U base as observed in DNA cases.[21]
The maximum emission wavelength in aqueous phosphate
buffer appeared at 465 nm with a rather low quantum yield
(FF =0.006). When hybridized with DNA1a with a comple-
mentary sequence, the fluorescence intensity of the duplex
at 465 nm was enhanced by a factor of 42 relative to the
single-stranded PNA1, and the quantum yield was increased
to 0.131 without change in the maximum emission wave-
length. The increases in the fluorescence intensity were also
observed in other mismatched PNA·DNA hybrids, but to a
much-lesser degree ranging from 3.0 (in the PNA1·DNA1d
carrying a pUPy·dG mismatch) to 8.7 fold (in the PNA1·D-
NA1c carrying a pUPy·dC mismatch). This range translates
to a discrimination factor between complementary and
single-mismatched hybrids of between 4.8 and 14 fold. The
3254
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 3251 – 3259

Base-Discriminating Fluorescent Nucleobase in Pyrrolidinyl Peptide Nucleic Acids
Table 2. Melting temperatures, quantum yields, and fluorescence intensi-
ty ratios of PNA6–PNA8 and their hybrids with DNA.
[b]
[c]
Entry System[a] Tm
[8C]
FF
I465
I465(ss)
I
N
Notes
1
2
PNA6
PNA6+
DNA6a
PNA6+
DNA6b
PNA6+
DNA6c
PNA6+
DNA6d
PNA6+
DNA6e
PNA6+
DNA6 f
PNA7
–
0.009 1.0
6.5
1.0
ss, AUPy
C
66.3 0.052 6.5
54.3 0.018 1.1
54.2 0.025 1.6
54.5 0.014 1.0
46.1 0.029 3.1
40.0 0.011 0.8
ds, complemen-
tary
3
4
5
6
7
5.9
4.1
6.5
2.1
8.1
ds, pUPy·dC
mismatch
ds, pUPy·dG
mismatch
ds, pUPy·dT
mismatch
ds, indirect mis-
matched
ds, double mis-
matched
8
9
–
–
0.030 1.0
0.066 2.7
2.7
1.0
ss, AUPy
T
[e]
[e]
PNA7+
DNA7
PNA8
PNA8+
DNA8
ds, complemen-
tary
Figure 4. Steady-state fluorescence spectra of PNA1–PNA5 (2.5 mm) and
PNA1–PNA5+DNA1a–DNA5 (3.0 mm) in 10 mm sodium phosphate
buffer pH 7.0, lex 350 nm.
10
11
–
–
0.027 1.0
0.084 4.0
4.0
1.0
ss, TUPy
T
ds, complemen-
tary
[a] Conditions: [PNA]=2.5 mm; [DNA]=3.0 mm, 10 mm sodium phos-
phate buffer, pH 7.0, lex 350 nm. [b] Measured using quinine sulfate as a
standard. [c] Ratio of fluorescence intensity at 465 nm to single-stranded
PNA. [d] Ratio of fluorescence intensity at 465 nm to complementary hy-
brids. [e] non-two-state melting curve.
rescence relative to single-stranded PNA5. The order of the
fluorescence change as a function of neighboring base was
T/T>A/A~C/C>G/G. Despite these variations, there is a
general trend that the single-stranded UPy-labeled acpcPNA
is weakly fluorescent, and upon hybridization the fluores-
cence intensity at 465 nm increased consistently within the
range 6.2–41.9 fold. This result suggests that the UPy can
function as a base-discriminating fluorescent nucleobase in
acpcPNA system. The general increase in fluorescence
signal upon hybridization to complementary DNA regard-
less of the nature of neighboring bases suggests that the
UPy-modified acpcPNA should be useful as a quencher-free
beacon probe that yield turn-on signals in the presence of
the correct DNA target. Although many fluorophore-la-
beled nucleobases have been proposed as BDF in both
DNA and aegPNA systems, many of them do not give turn-
on signals.[3,7,28,29] In addition, the signal change is often
highly dependent on the nature of the flanking bases to the
extent that little or no discrimination could be observed for
some sequences.[16]
ingly, the Tm decrease in the mismatched hybrids is much
smaller than would be expected from acpcPNA (a Tm de-
crease of ca. 20–258C is typical for single mismatched
acpcPNA·DNA hybrids of this length). For example, Tm
values of the unmodified acpcPNA with the same sequence
as PNA6, but carrying T instead of UPy, with DNA6a (com-
plementary) and DNA6b (single mismatched) were 75.0
and 51.88C, respectively under similar conditions (Tm de-
crease=23.28C). Despite this, the ability of UPy in acpcPNA
to differentiate A from other bases is in sharp contrast to
the behavior of the same UPy base in DNA whereby there
was essentially no Tm difference in the duplexes having the
four bases located opposite to the UPy.[21] We propose that
the mismatched PNA·DNA and DNA·DNA hybrids could
be stabilized by rotation of the UPy base to allow “pairing”
between the pyrene and the mismatched base, which would
result in an intercalation of the pyrene unit between the
base stack.[30] This intercalation might also partially take
place in the complementary DNA·DNA hybrid with a rela-
tively weak pairing between UPy and T, but would be more
difficult in complementary PNA·DNA hybrids whereby the
pairing is much stronger. UV/Vis spectra support this hy-
pothesis as shown by the significant red-shifted absorption
maxima to ca. 360 nm with a pronounced broadening and
hypochromism of the pyrene absorption in all mismatched
hybrids of PNA6 compared to the single-stranded PNA6
(Figure 5). This spectral characteristic is remarkably similar
to that of complementary and mismatched DNA duplexes
carrying UPy reported in the literature.[21] In addition, CD
spectra of PNA6 with DNA6a (complementary) and
DNA6b–d (mismatched) (Figure S14) also support this in-
Thermal Stabilities and Optical Properties of DNA Hybrids
with Mixed Sequence acpcPNA Carrying UPy Modification
Next, the thermal stabilities and optical properties of UPy
containing a mixed-sequence acpcPNA PNA6 were investi-
gated. The data are summarized in Table 2. As shown by Tm
analysis, the hybrid between PNA6 and complementary
DNA (DNA6a) was more thermally stable than the mis-
matched hybrids. The Tm values of hybrids between PNA6
and DNA6a (complementary) and DNA6b–DNA6d (single
mismatched) were 66.3 and 54.2–54.58C, respectively. The
decrease in the Tm values of the single mismatched hybrid
from the complementary hybrid by ca. 128C suggested that
the UPy base in mixed sequence acpcPNA can still specifical-
ly recognize A in the complementary DNA strand. Interest-
Chem. Asian J. 2011, 6, 3251 – 3259
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
3255

H. A. Wagenknecht, T. Vilaivan et al.
FULL PAPERS
accommodate pyrene intercalation, which results in very un-
stable mismatched duplexes.
Fluorescence spectra of the single stranded PNA6 showed
a weak and broad emission band similar to PNA1 with a
fluorescence quantum yield of 0.009 (Figure 6). In the pres-
ence of the complementary DNA6a, the intensity of the
Figure 5. UV/Vis spectra of PNA6 (2.5 mm)+DNA6a–DNA6 f (3.0 mm) in
10 mm sodium phosphate buffer pH 7.0.
tercalation model. The difference in the pyrene absorption
region (ca. 360 nm) suggests that the pyrene chromophore is
differently organized in the complementary and mismatched
duplexes. Among all mismatched hybrids of PNA6 studied,
only the PNA6·DNA6e pair carrying a mismatch base not
directly opposite to the UPy base shows a similar absorption
spectrum to the complementary hybrid, thus indicating the
absence of intercalation of the pyrene moiety. Interestingly,
this particular indirect mismatch hybrid is also much-less
thermally stable (Tm =46.18C) than the other mismatch
pairs (Tm =54.3–54.58C). These results are completely con-
sistent with the proposed model above since in the indirect
mismatched hybrid PNA6·DNA6e, the UPy still forms hy-
drogen bonds with A; therefore, no intercalation is possible.
It is important to note that the absorption spectra of the
DNA hybrids of PNA6 are quite different from those of
PNA1. In the case of complementary hybrids, the change in
the UV absorption spectra of the pyrene chromophore of
PNA6 was much-smaller than that of PNA1, thereby sug-
gesting that the two duplexes might adopt different struc-
tures. The unusual structure of homothymine/homopyrimi-
dine sequence is well recognized in the DNA field, which is
also supported here by the unusually high thermal stability
and the very large increase in fluorescence change of the
PNA1·DNA1a hybrid compared to the PNA6·DNA6a
hybrid. In the case of mismatched duplexes of PNA6, signif-
icant bathochromic shifts were observed in the absorption
spectra. However, these shifts are not the same as that ob-
served in the complementary PNA1·DNA1a duplex because
the magnitude of the shifts was much-larger (15 nm vs 5 nm)
and the shape of the absorption spectra were different. We
attribute the spectral change in this case to the intercalation
of the pyrene into the base stack of the mismatched duplex.
No similar change was observed in the mismatched hybrids
of PNA1. This may be due to the poor ability of AT pairs to
Figure 6. Steady-state fluorescence spectra of PNA6 (2.5 mm)+DNA6a–
DNA6 f (3.0 mm) in 10 mm sodium phosphate buffer pH 7.0, lex 350 nm.
emission at 465 nm increased ca. 6.5 fold. This increase was
accompanied by an increase in the quantum yield to 0.052.
The presence of single mismatched targets DNA6b–
DNA6d resulted in little or almost no change in the fluores-
cence intensity, although the fluorescence quantum yields in-
creased a little because of the hypochromism of the pyrene
chromophore in the absorption spectra. The pUPy·dG mis-
match hybrid (PNA6·DNA6c) exhibited the highest fluores-
cence among all mismatched hybrids (1.6 fold enhancement
relative to the single-stranded PNA6). Nevertheless, the
signal is still much lower than that of the complementary
hybrid, allowing unambiguous differentiation between com-
plementary and single mismatched DNA targets. On the
other hand, when the mismatched base in the DNA strand
was located in a position that is not directly opposite to the
UPy as in DNA6e, a significant fluorescence enhancement
relative to the single-stranded PNA6 (3.1 folds) was ob-
served even though this single mismatched hybrid was less
thermally stable than the other mismatched hybrids studied
earlier (see the Tm figures in Table 2). When a second base
mismatch was introduced to the DNA strand at the position
opposite to the UPy base as in DNA6 f, the fluorescence be-
comes indifferent from the single-stranded PNA6.
These results indicate that the fluorescence enhancement
of the UPy is directly associated to the base pairing event be-
tween the UPy and A. The same fluorescence enhancement
upon hybridization was also observed in PNA7 and PNA8,
which carry different flanking bases (AUPyT and TUPyT).
3256
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 3251 – 3259

Base-Discriminating Fluorescent Nucleobase in Pyrrolidinyl Peptide Nucleic Acids
The fluorescence enhancements in these cases are consider-
ably smaller than in the case of PNA6 owing to a relatively
high fluorescence background of the single stranded PNA7
and PNA8. Nevertheless, the fluorescence enhancement still
reached 2.7 fold in the worst case (PNA7), which could still
be readily distinguished by the naked eye (Figure 7; also see
the Supporting Information, Figure S20). These results con-
PNA1 were almost identical. This is interpreted as an evi-
dence for the interaction of the UPy with the neighboring T
bases, which is disrupted in the presence of dimethyl sulfox-
ide. From the variation in the fluorescence quantum yield of
the single stranded PNA1–PNA8 (ranging from 0.006 in
PNA1 to 0.030 in PNA7), it is clear that the base sequence
is an important factor to determine the “darkness” of the
single-stranded probe. The homo-T sequence PNA1 was the
darkest, which seem to be reasonable based on the knowl-
edge that T is an effective quencher for pyrene.[31] In gener-
al, single-stranded T-rich sequences appear to be darker
than the mixed sequences, even when the flanking bases of
the UPy are not T. On the other hand, the mixed sequence
PNA7 and PNA8 carrying one and two flanking T bases, re-
spectively, are relatively bright in the single-stranded state.
These results suggest that the quenching effect must also, at
least additionally, take place by non-nearest neighboring
bases. This is not unexpected because the hydrophobic back-
bone of the single-stranded PNA can fold into a compact
shape in an aqueous environment,[32] thus leading to possible
interactions between the UPy and non-proximal nucleobases.
When the base pairing between the UPy and A takes place,
the rigidly-linked pyrene must be forced into a more pro-
nounced coplanar situation with the U moiety and to point
into the major groove direction, away from the helix center
and interacting with the aqueous environment.[8,22] The
change in the local environment of the UPy chromophore, as
well as the twisting between the pyrene and uracil aromatic
Figure 7. Steady-state fluorescence spectra of PNA7 and PNA8 (2.5 mm),
and with addition of DNA7 and DNA8 (3.0 mm), respectively. 10 mm
Sodium phosphate buffer pH 7.0, lex 350 nm.
ꢀ
planes along the connecting C C bond which controls the
charge-transfer contribution leads to the observed fluores-
cence enhancement.[33,34] The described differences in the
UV/Vis absorption spectra support this interpretation. Since
the pyrene unit is electronically coupled to the uracil base
and in the duplex state this base can only directly interact
with the two neighboring base pairs, the brightness of the
fluorescence should now be mainly determined by the
nature of the flanking bases. The observed brightness with
different neighboring bases is decreasing from T/T>A/A~
C/C>G/G. Nevertheless, in all cases the expected fluores-
cence enhancement was observed upon duplex formation,
indicating that the system should be widely applicable to a
variety of sequences. When hybridized to a mismatched
DNA target carrying a mismatched base opposite to the UPy,
the duplex is stabilized by intercalation of the pyrene sub-
stituent of the UPy into the duplex, as suggested by the red
shift of the UV spectra. Fortunately, this event does not lead
to a significant fluorescence change from the single strand
PNA as compared to the complementary hybrid, therefore
the UPy in acpcPNA can behave as a BDF which yields a
turn-on signal only when paired with adenine.
firms the above conclusion based on experiments with ho-
mothymine sequences that the UPy is a versatile BDF in
acpcPNA system that produce a turn-on signal upon pairing
with the base A, irrespective of the nature of the flanking
base sequences.
On the Origin of the Fluorescence Enhancement
In the single-stranded acpcPNA, the UPy base should be
quenched by photoinduced electron transfer because of in-
teractions with neighboring nucleobases. The fluorescence
spectra of the monomer (one acpcPNA dipeptide unit with
UPy as a base, capped with acetyl and lysinamide at the N-
and C-termini) showed an emission band at ca. 400 nm typi-
cal of free pyrene in aqueous buffer, possibly as a result of
the non-coplanarity between the pyrene and uracil chromo-
phores. In the same buffer, the PNA1 showed a weak emis-
sion band at 465 nm, which should arise from the internal
exciplex Py·+-U·ꢀ.[17] Addition of an organic co-solvent, es-
pecially dimethyl sulfoxide, which is known to disrupt base
stacking increased the emission intensities of the monomer
without changing the emission maxima (Figure S15). On the
other hand, a hypsochromic shift of the emission spectrum
with an increased in intensity was observed for PNA1 in the
presence of dimethyl sulfoxide (Figure S16). In pure dimeth-
yl sulfoxide, the emission maxima of the monomer and
Conclusions
In conclusion, we have demonstrated the suitability of 5-
(pyren-1-yl)uracil (UPy) as a base-discriminating fluorescent
(BDF) nucleobase in the conformationally rigid acpcPNA
Chem. Asian J. 2011, 6, 3251 – 3259
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
3257

H. A. Wagenknecht, T. Vilaivan et al.
FULL PAPERS
HRMS (ESI+): m/z calcd for C34H32O7N3I+Na+: 744.1183 [M+Na]+;
found: 744.1105.
system. In contrast to the analogous study in DNA, the UPy
base in acpcPNA exhibits a strict Watson–Crick base-pair-
ing-specificity towards A. Most importantly, the fluores-
cence of the UPy base is significantly enhanced when the
acpcPNA is hybridized to its DNA target, irrespective of the
nature of the flanking bases. The high specificity of the
acpcPNA system allows discrimination between the comple-
mentary and single-mismatched DNA targets. Experiments
with mismatched DNA targets revealed that the fluores-
cence enhancement was specific to the pairing between UPy
and dA. As a result, the UPy-modified acpcPNA is potential-
ly useful as a hybridization-responsive fluorescent probe for
DNA sequence determination.
N-tert-Butoxycarbonyl-(4’R)-[5-(pyren-1-yl)uracil-1-yl]-(2’R)-proline
diphenylmethyl ester (4)
A round-bottomed flask was charged with N-tert-butoxycarbonyl-(4’R)-
[3-benzoyl-5-iodouracil-1-yl]-(2’R)-proline diphenylmethyl ester (2;
0.3430 g, 0.475 mmol), pyren-1-boronic acid (0.1862 g, 0.755 mmol), and
tetrakis(triphenylphosphine)palladium
[Pd
G
(0.0200 g,
0.0165 mmol). After flushing the reaction vessel with nitrogen to remove
air, degassed CH3CN (3 mL) and triethylamine (0.3 mL, 0.500 mmol)
were added. The reaction mixture was heated to 808C with stirring under
nitrogen. When the starting material was completely consumed (40 h,
monitored by TLC), the solvent was removed under reduced pressure
and the residue was separated by column chromatography on silica gel
eluting with CH2Cl2/acetone (30:1). Without further characterization, the
protected intermediate N-tert-butoxycarbonyl-(4’R)-[3-benzoyl-5-(pyren-
1-yl)uracil-1-yl]-(2’R)-proline diphenylmethyl ester (3), was treated with
ethanolamine (0.2 mL) in DMF (1 mL) for 20 min. After the reaction
was finished, as indicated by TLC, the reaction mixture was diluted with
CH2Cl2 (20 mL). The organic layer was washed with 10% HCl (2ꢁ
15 mL), dried with Na2SO4, and the solvent was removed. The residue
was purified by column chromatography on silica gel eluting with hex-
anes/ethyl acetate (50:50) to obtain 4 as an off-white solid (0.2010 g,
60% yield from 2). Rf =0.73 (hexanes/EtOAc 50:50); ½aꢁ2D5 =+46.6 (c=
Experimental Section
General
Chemicals and solvents were purchased from commercial suppliers and
used without further purification. Oligonucleotides were purchased from
Biodesign (Bangkok, Thailand). RP-HPLC experiments were performed
on a Water Delta 600 HPLC system. 1H and 13C NMR spectra were re-
1
0.014m, CHCl3); H NMR (400 MHz, CDCl3): d=1.33 and 1.51 (2 s, 9H;
corded on
a Bruker Avance 400 NMR spectrometer operating at
CH3 Boc), 2.05 (m, 1H; 1ꢁCH2(3’)), 2.85 (m, 1H; 1ꢁCH2(3’)), 3.59 and
3.73 (m, 1H; 1ꢁCH2(5’) rotamers), 4.02 (m, 1H; 1ꢁCH2(5’)), 4.52 and
4.61 (m, 1H; CH(4’) rotamers), 5.19 (m, 1H; CH(2’)), 6.87 (s, 1H;
CHPh2), 7.25–7.51 (m, 10H; Ar Dpm), 7.42 and 7.45 (2 s, 1H; C(6)H UPy
rotamers), 7.86–8.24 (m, 9H; CH pyrene), 9.09 ppm (s, 1H; NH UPy);
13C NMR (100 MHz, CDCl3): d=28.0 and 28.3 (CH3 Boc rotamers), 34.2
and 35.5 (CH2(3’) rotamers), 49.0 and 49.4 (CH(5’) rotamers), 52.8 and
53.6 (CH(4’) rotamers), 57.3 and 57.5 (CH(2’) rotamers), 77.9 and 78.2
(CH Dpm rotamers), 81.3 (CCH3 Boc), 115.7 (C(5) UPy), 124.6–126.1
(CH pyrene), 126.8–128.6 (CH Ar Dpm, CH and C pyrene), 129.1 (C
pyrene), 130.8 (C pyrene), 131.2 (C pyrene), 131.7 (C pyrene), 139.2
(CH(6) UPy and C Dpm), 150.7 (C(2) UPy), 153.3 (CO Boc), 161.8 (C(4)
UPy), 170.9 ppm (CO proline); IR (ATR): n˜max =1670.6, 1452.4, 1390.5,
1358.2, 1280.1, 1177.7, 1150.8, 964.9 cmꢀ1; HRMS (ESI+): m/z calcd for
C43H37O6N3 +Na+: 714.2580 [M+Na]+; found: 714.2519.
400 MHz for 1H nuclei and 100 MHz for 13C nuclei. MALDI-TOF mass
spectra were obtained on a Microflex MALDI-TOF mass spectrometry
(Bruker Daltonics) using a-cyano-4-hydroxycinnamic acid (CCA) as the
matrix. High resolution mass spectra were recorded in positive ESI mode
on a Bruker Daltonics micrOTOF. IR spectra were recorded on Nicolet
iS10 FT-IR spectrometer. The acpcPNAs were synthesized according to
the previously published solid-phase peptide synthesis procedure[23] (0.5
mmol scale) and were purified by reverse-phase HPLC. Spectroscopic
measurements were carried out in sodium phosphate buffer solution
(10 mm, pH 7.0) using 10 mm quartz glass cuvettes. Absorption spectra
and the melting temperatures (2.5 mm PNA, 3.0 mm DNA, 20–908C,
1.08Cminꢀ1) were recorded on a Varian Cary 100 UV/Vis spectropho-
tometer. Fluorescence spectra were measured on a Varian Eclipse spec-
trofluorimeter.
N-tert-Butoxycarbonyl-(4’R)-[3-benzoyl-5-iodouracil-1-yl]-(2’R)-proline
diphenylmethyl ester (2)
N-Fluoren-9-ylmethoxycarbonyl-(4’R)-[5-(pyren-1-yl)uracil-1-yl]-(2’R)-
proline (1)
N-tert-Butoxycarbonyl-trans-4-hydroxy-d-proline diphenylmethyl ester[27]
(0.9938 g, 2.5 mmol), triphenylphosphine (0.7213 g, 2.75 mmol) and N3-
benzoyl-5-iodouracil (0.9407 g, 2.75 mmol) were dissolved in dry THF
(10 mL) with stirring under N2 and cooled down to 08C in an ice–salt
bath. To the reaction mixture was slowly added DIAD (0.54 mL,
2.75 mmol) via a syringe over a period of 15 min. The reaction was
slowly warmed to room temperature and the stirring was continued 24 h.
The solvent was removed and the residue was recrystallized from ethanol
to afford 2 as a white solid (0.5424 g, 47% yield). Rf =0.60 (hexanes/
EtOAc 75:25); m.p. 124–1268C; ½aꢁ2D5 =+50.2 (c=0.014m, CHCl3);
1H NMR (400 MHz, CDCl3): d=1.33 and 1.51 (2 s, 9H; CH3 Boc rotam-
ers), 2.05 (m, 1H; 1ꢁCH2(3’)), 2.85 (m, 1H; 1ꢁCH2(3’)), 3.59 and 3.73
(m, 1H; 1ꢁCH2(5’) rotamers), 4.02 (m, 1H; 1ꢁCH(5’) rotamers), 4.52
and 4.61 (m, 1H; CH(4’) rotamers) 5.19 (m, 1H; CH(2’)), 6.99 (s, 1H;
CHPh2), 7.29 (s, 1H; C(6)H UI), 7.37 (m, 10H; Dpm ArH), 7.51 (dd, J=
N-tert-Butoxycarbonyl-(4’R)-[5-(pyren-1-yl)uracil-1-yl]-(2’R)-proline di-
phenylmethyl ester (4) (0.1200 g, 0.170 mmol) was treated with para-tol-
uenesulfonic acid monohydrate (0.0950 g, 0.5 mmol) in MeCN (2 mL) at
308C for 3 h. After the Boc deprotection was complete (as determined
by TLC), diisopropylethylamine (DIEA) (150 mL, 0.85 mmol) was added
to the reaction mixture followed by 9-fluorenylmethoxycarbonyl chloride
(0.0660 g, 0.255 mmol). After the reaction was completed as indicated by
TLC (ca. 1 h), the solvent was removed and the residue was purified by
column chromatography on silica gel eluting with hexanes/ethyl acetate
(50:50) to give the intermediate N-fluoren-9-ylmethoxycarbonyl-(4’R)-[5-
(pyren-1-yl)uracil-1-yl]-(2’R)-proline diphenyl methyl ester. Without fur-
ther characterization, this intermediate was treated with trifluoroacetic
acid/CH2Cl2 (1:1) containing 10% triisopropylsilane (1 mL) and left at
308C for 30 min. After that, the volatiles were removed by a gentle
stream of nitrogen. The residue was triturated and washed with diethyl
ether to give a white solid (0.0600 g, 55% yield from 4). Rf =0.55 (hex-
anes/EtOAc 75:25+0.5% HOAc); ½aꢁ2D5 =ꢀ4.7 (c=0.015m, DMF);
1H NMR (400 MHz, CD3OD): d=2.34 and 2.39 (2 m, 1H; 1ꢁCH2(3’) ro-
tamers), 2.78 and 2.85 (2 m, 1H; 1ꢁCH2(3’) rotamers), 3.71 and 3.77 (m,
1H; 1ꢁCH2(5’) rotamers), 3.87 and 4.04 (m, 1H; 1ꢁCH2(5’) rotamers),
4.17 (m, 1H; CH Fmoc), 4.34 (m, 2H; CH2 Fmoc), 4.31 and 4.43 (2 m,
1H; CH(2’) rotamers), 5.11 and 5.20 (2 m, 1H; CH(4’) rotamers), 6.80–
7.80 (m, 8H; CH Fmoc), 7.92 (s, 1H; CH(6) UPy), 7.95–8.23 ppm (m, 9H;
CH pyrene); 13C NMR (100 MHz, CD3OD): d=34.2 and 34.9 (CH2(3’)
rotamers), 46.9 (CH Fmoc), 48.8 and 49.3 (CH(5’) rotamers), 53.2 and
7.0, 7.4 Hz, 2H; CH
ACHTUNGTRENNUNG
7.88 ppm (m, 2H; CH
ACHTUNGTRENNUNG
and 28.3 (CH3 Boc rotamers), 34.2 and 35.9 (CH2(3’) rotamers), 49.6 and
49.9 (CH(5’) rotamers), 53.3 and 53.7 (CH(4’) rotamers), 57.4 and 57.5
(CH(2’) rotamers), 68.4 (C(5) UI), 78.3 and 78.5 (CH Dpm rotamers),
81.6 (CCH3 Boc), 127.1–128.8 (CH Ar Dpm), 129.2 and 130.5 (CH(2)
and CH(3) Bz), 130.9 (C(1) Bz), 135.3 (C(4)H Bz), 139.2 (C Ar Dpm),
144.9 (CH(6) UI), 149.6 (C(2) UI), 153.4 (CO Boc), 158.5 (C(4) UI), 167.4
(CO Bz), 171.2 ppm (CO proline); IR (ATR): n˜max =1737.9, 1700.2,
1655.2, 1614.0, 1433.6, 1395.9, 1382.4, 1234.3, 1196.6, 1148.1, 983.8 cmꢀ1
;
3258
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 3251 – 3259

Base-Discriminating Fluorescent Nucleobase in Pyrrolidinyl Peptide Nucleic Acids
54.2 (CH(4’) rotamers), 58.6 and 58.9 (CH(2’) rotamers), 67.4 and 68.0
(CH2 Fmoc rotamers), 114.6 (CH(5) UPy), 119.4 (CH Ar Fmoc), 124.3–
125.1 (CH and C pyrene), 125.7 (CH Ar Fmoc), 126.8–128.6 (CH Ar
Fmoc, CH and C pyrene), 130.0 (C pyrene), 130.9 (C pyrene), 131.3 (C
pyrene), 131.4 (C pyrene), 141.1 (C Ar Fmoc), 141.9 (CH(6) UPy), 143.6–
144.1 (C Ar Fmoc), 151.5 (C(2) UPy), 154.9 and 155.2 (CO Fmoc), 163.7
(C(4) UPy), 175.2 and 175.9 ppm (CO proline rotamers); IR (ATR):
n˜max =1681.4, 1447.1, 1414.7, 1280.1, 1175.1, 1115.8 cmꢀ1; HRMS (ESI+):
m/z calcd for C40H29N3O6+Na+: 670.1954 [M+Na]+; found: 670.1923.
[12] S. S. Bag, Y. Saito, K. Hanawa, S. Kodate, I. Suzuka, I. Saito, Bioorg.
Med. Chem. Lett. 2006, 16, 6338–6341.
[13] Q. Xiao, R. T. Ranasinghe, A. M. P. Tang, T. Brown, Tetrahedron
2007, 63, 3483–3490.
[14] Y. Saito, K. Motegi, S. S. Bag, I. Saito, Bioorg. Med. Chem. 2008, 16,
107–113.
[15] G. T. Hwang, Y. J. Seo, B. H. Kim, J. Am. Chem. Soc. 2004, 126,
6528–6529.
[16] J. H. Ryu, Y. J. Seo, G. T. Hwang, J. Y. Lee, B. H. Kim, Tetrahedron
2007, 63, 3538–3547.
[17] T. Netzel, Tetrahedron 2007, 63, 3491–3514.
[18] T. L. Netzel, M. Zhao, K. Nafisi, J. Headrick, M. S. Sigman, B. E.
Eaton, J. Am. Chem. Soc. 1995, 117, 9119–9128.
[19] E. Mayer, L. Valis, R. Huber, N. Amann, H.-A. Wagenknecht, Syn-
thesis 2003, 2335–2340.
[20] E. Mayer-Enthart, H.-A. Wagenknecht, Angew. Chem. 2006, 118,
3451–3453; Angew. Chem. Int. Ed. 2006, 45, 3372–3375.
[21] C. Wanninger-Weiss, H.-A. Wagenknecht, Eur. J. Org. Chem. 2008,
64–71.
Acknowledgements
This work was supported by the Alexander von Humboldt Foundation
(to T.V.), the Development and Promotion of Science and Technology
Talents Project (to C.B.), the Thailand Research Fund (RTA5280002),
the Thai government stimulus package 2 (TKK2555, SP2) under the Proj-
ect for Establishment of Comprehensive Center for Innovative Food,
Health Products and Agriculture, the National Research University
Grant from the Commission of Higher Education, and the Ratchadaphi-
seksomphot Endowment (AM1006A).
[22] N. Venkatesan, Y. J. Seo, B. H. Kim, Chem. Soc. Rev. 2008, 37, 648–
663.
[23] a) T. Vilaivan, C. Srisuwannaket, Org. Lett. 2006, 8, 1897–1900;
b) C. Vilaivan, C. Srisuwannaket, C. Ananthanawat, C. Suparpprom,
J. Kawakami, Y. Yamaguchi, Y. Tanaka, T. Vilaivan, Artificial DNA:
PNA & XNA 2011, 2, 50–59; XNA 2011, 2, 50–59.
[24] C. Ananthanawat, T. Vilaivan, V. P. Hoven, X. D. Su, Biosens. Bio-
electron. 2010, 25, 1064–1069.
[1] R. W. Sinkeldam, N. J. Greco, Y. Tor, Chem. Rev. 2010, 110, 2579–
2619.
[2] A. Okamoto, K. Tainaka, I. Saito, J. Am. Chem. Soc. 2003, 125,
4972–4973.
[3] A. Okamoto, K. Tanaka, T. Fukuta, I. Saito, J. Am. Chem. Soc.
2003, 125, 9296–9297.
[4] K. Matsumoto, Y. Shinohara, S. S. Bag, Y. Takeuchi, T. Morii, Y.
Saito, I. Saito, Bioorg. Med. Chem. Lett. 2009, 19, 6392–6395.
[5] J. S. Mann, Y. Shibata, T. Meehan, Bioconjugate Chem. 1992, 3,
554–558.
[6] J. Telser, K. A. Cruickshank, L. E. Morrison, T. L. Netzel, J. Am.
Chem. Soc. 1989, 111, 6966–6976.
[25] T. Pesnot, G. K. Wagner, Org. Biomol. Chem. 2008, 6, 2884–2891.
[26] N. Amann, H.-A. Wagenknecht, Synlett 2002, 687–691.
[27] G. Lowe, T. Vilaivan, J. Chem. Soc. Perkin Trans. 1 1997, 547–554.
[28] N. J. Greco, Y. Tor, J. Am. Chem. Soc. 2005, 127, 10784–10785.
[29] A. Omumi, D. G. Beach, M. Baker, W. Gabryelski, R. A. Mander-
ville, J. Am. Chem. Soc. 2011, 133, 42–50.
[30] Pyrene has been shown to behave as a universal nucleobase in
aegPNA: T. Tedeschi, A. Tornelli, S. Sforza, R. Corradini, R. Mar-
chelli, Artificial DNA: PNA & XNA 2010, 1, 83–89; XNA 2010, 1,
83–89.
[7] Y. Saito, Y. Miyauchi, A. Okamoto, I. Saito, Chem. Commun. 2004,
1704–1705.
[31] C. A. M. Seidel, A. Schulz, M. H. M. Sauer, J. Phys. Chem. 1996,
100, 5541–5553.
[8] A. Okamoto, K. Kanatani, I. Saito, J. Am. Chem. Soc. 2004, 126,
4820–4827.
[32] O. Seitz, Angew. Chem. 2000, 112, 3389–3392; Angew. Chem. Int.
Ed. 2000, 39, 3249–3252.
[9] E. Mayer, L. Valis, C. Wagner, M. Rist, N. Amann, H.-A. Wagen-
knecht, ChemBiochem. 2004, 5, 865–868.
[33] Z. R. Grabowski, K. Rotkiewicz, W. Rettig, Chem. Rev. 2003, 103,
3899–4031.
[10] C. Wagner, M. Rist, E. Mayer-Enthart, H.-A. Wagenknecht, Org.
Biomol. Chem. 2005, 3, 2062–2063.
[34] A. Trifonov, I. Buchvarov, H.-A. Wagenknecht, T. Fiebig, Chem.
Phys. Lett. 2005, 409, 277–280.
[11] G. T. Hwang, Y. J. Seo, S. J. Kim, B. H. Kim, Tetrahedron Lett. 2004,
45, 3543–3546.
Received: May 27, 2011
Published online: September 26, 2011
Chem. Asian J. 2011, 6, 3251 – 3259
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
3259