Transglutaminase-mediated synthesis of a DNA-(Enzyme)n probe for highly sensitive DNA detection

Article abstract of DOI:10.1002/chem.201003744

A new synthetic strategy for DNA-enzyme conjugates with a novel architecture was explored using a natural cross-linking catalyst, microbial transglutaminase (MTG). A glutamine-donor substrate peptide of MTG was introduced at the 5-position on the pyrimidine of deoxyuridine triphosphate to prepare a DNA strand with multiple glutamine-donor sites by polymerase chain reaction (PCR). A substrate peptide that contained an MTG-reactive lysine residue was fused to the N terminus of a thermostable alkaline phoshatase from Pyrococcus furiosus (PfuAP) by genetic engineering. By combining enzymatically the substrate moieties of MTG introduced to the DNA template and the recombinant enzyme, a DNA-(enzyme)_n conjugate with 1:n stoichiometry was successfully obtained. The enzyme/DNA ratio of the conjugate increased as the benzyloxycarbonyl-L-glutaminylglycine (Z-QG) moiety increased in the DNA template. The potential utility of the new conjugate decorated with signaling enzymes was validated in a dot blot hybridization assay. The DNA-(enzyme) _n probe could clearly detect 10⁴ copies of the target nucleic acid with the complementary sequence under harsh hybridization conditions, thereby enabling a simple detection procedure without cumbersome bound/free processes associated with a conventional hapten-antibody reaction-based DNA-detection system. DNA detector: A glutamine-modified DNA probe was synthesized by the polymerase chain reaction by using a glutamine-modified 2′-deoxyuridine 5′-triphosphate analogue as a substrate of DNA polymerase. The DNA-(alkaline phosphatase)_n conjugate probe was highly sensitive and could directly visualize the target DNA bound on a membrane immediately after hybridization (see graphic).

Full text of DOI:10.1002/chem.201003744

FULL PAPER
DOI: 10.1002/chem.201003744
Transglutaminase-Mediated Synthesis of a DNA–(Enzyme)n Probe for
Highly Sensitive DNA Detection
Momoko Kitaoka,[a] Yukito Tsuruda,[a] Yukari Tanaka,[a] Masahiro Goto,[a, b]
Masayuki Mitsumori,[c] Kounosuke Hayashi,[c] Yoshiyuki Hiraishi,[c]
Katsuyuki Miyawaki,[d] Sumihare Noji,[d] and Noriho Kamiya*[a, b]
Abstract: A new synthetic strategy for
DNA–enzyme conjugates with a novel
architecture was explored using a natu-
ral cross-linking catalyst, microbial
transglutaminase (MTG). A glutamine-
donor substrate peptide of MTG was
introduced at the 5-position on the pyr-
imidine of deoxyuridine triphosphate
to prepare a DNA strand with multiple
glutamine-donor sites by polymerase
chain reaction (PCR). A substrate pep-
tide that contained an MTG-reactive
lysine residue was fused to the
N terminus of a thermostable alkaline
phoshatase from Pyrococcus furiosus
(PfuAP) by genetic engineering. By
combining enzymatically the substrate
moieties of MTG introduced to the
DNA template and the recombinant
enzyme, a DNA–(enzyme)n conjugate
with 1:n stoichiometry was successfully
obtained. The enzyme/DNA ratio of
the conjugate increased as the benzy-
loxycarbonyl-l-glutaminylglycine (Z-
QG) moiety increased in the DNA
template. The potential utility of the
new conjugate decorated with signaling
enzymes was validated in a dot blot hy-
bridization assay. The DNA–(enzyme)n
probe could clearly detect 104 copies of
the target nucleic acid with the comple-
mentary sequence under harsh hybridi-
zation conditions, thereby enabling a
simple detection procedure without
cumbersome bound/free processes as-
sociated with a conventional hapten–
antibody reaction-based DNA-detec-
tion system.
Keywords: conjugation · DNA · fil-
ter hybridization · protein engineer-
ing · transglutaminase
Introduction
have been developed.[3–6] In particular, Carell et al. and
Hocek et al. reported highly sensitive detection and func-
tionalization of DNA, respectively, by site-specific and cova-
lent modification with properly designed chemicals.[7,8] Cur-
rently, among nonradioactive probe variants, a digoxigenin
(DIG)-labeled probe is widely used. The DIG probe is pre-
pared by the incorporation of DIG-labeled uridine-5’-tri-
phosphate into the DNA probe, and is detected using an
anti-DIG antibody–CIAP (calf intestinal alkaline phospha-
tase) conjugate. The DIG probe achieves comparable sensi-
tivity to the RI probe.[9,10] However, antibodies show insta-
bility toward heat and denaturants that are often required
during the hybridization processes employed in filter and in
situ hybridization methods. Therefore, binding of an
Filter and in situ hybridization techniques are fundamental
approaches for detecting target nucleic acid sequences. Pio-
neering research of sequence-specific nucleic acid detection
utilized radioisotope (RI)-labeled probes.[1,2] These probes
were found to show high sensitivity; however, stability and
safety issues remained. Therefore, probes with nonradioac-
tive labels, such as chemicals, photochemicals, and enzymes
coupled with chromogenic or chemiluminescent substrates
[a] Dr. M. Kitaoka, Y. Tsuruda, Y. Tanaka, Prof. M. Goto,
Prof. N. Kamiya
Department of Applied Chemistry
Graduate School of Engineering
Kyushu University, 744 Motooka, Nishi-ku
Fukuoka 819-0395 (Japan)
enzyme–antibody conjugate to
a hapten-labeled probe
should always be performed after probe hybridization.
Some enzyme-labeled probes are easy to handle and widely
used; however, the enzyme-labeled probes are generally less
sensitive because of the reduced enzymatic activities upon
labeling and/or the declined hybridization efficiency of the
nucleic acid probes caused by the random cross-linking be-
tween enzymes and nucleic acids.
Fax : (+81)92-2810
E-mail: nori_kamiya@mail.cstm.kyushu-u.ac.jp
[b] Prof. M. Goto, Prof. N. Kamiya
Center for Future Chemistry, Kyushu University (Japan)
[c] Dr. M. Mitsumori, K. Hayashi, Dr. Y. Hiraishi
Aloka Co., Ltd., 3-7-19 Imai, Ome-shi, Tokyo (Japan)
[d] Dr. K. Miyawaki, Prof. S. Noji
Department of Life Systems
Here we propose a transglutaminase (TGase)-mediated
site-specific prelabeling system of signaling enzymes on nu-
cleic acid probes as an approach that skips a series of post-
hybridization processes (i.e., blocking, hapten–antibody re-
action, and washing), thereby enabling direct hybridization
and detection of target sequences (Figure S1 in the Support-
Institute of Technology and Science
The University of Tokushima, 2–1 Minamijosanjima-cho
Tokushima (Japan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201003744.
Chem. Eur. J. 2011, 17, 5387 – 5392
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5387

ing Information). A variety of synthetic methodologies have
been proposed for the preparation of semisynthetic DNA–
protein conjugates.[11] Since enzymatic labeling of DNA with
target proteins has clear advantages with respect to site-spe-
cificity and protein-friendly reaction conditions, several en-
zymes that post-translationally modify proteins have been
exploited.[12–14] However, the majority of previous research,
including our efforts,[12] have focused on the preparation of
DNA–protein conjugates with a 1:1 stoichiometry. The pres-
ent study has focused on a DNA–(protein)n conjugate with
1:n stoichiometry. More specifically, we explored the crea-
tion of a novel nucleic acid probe decorated with a site-spe-
cifically linked enzyme for the highly sensitive detection of
a target nucleic acid with the complementary sequence of a
probe (Scheme 1).
ent TGase from Streptomyces mobaraensis (microbial
TGase, MTG)[17,18] effectively catalyzes the 1:1 cross-linking
reaction of a K-tagged enzyme to a 5’-end Q-labeled oligo-
nucleotide.[12] In the present research, an acyl-donor sub-
strate of MTG, benzyloxycarbonyl-l-glutaminylglycine
AHCTUNGTRENNUNG
(Z-QG)[19–21] was coupled to the primary amino group of 5-
(3-aminoallyl)-2’-deoxyuridine-5’-triphosphates (AA-dUTP)
by an amide linkage (Figure S2 in the Supporting Informa-
tion) to obtain 5-[3-(Z-Gln-Gly-amido)-1-(E)-propenyl]-2’-
deoxyuridine 5’-triphosphate (Z-QG-dUTP, Scheme 1), and
to subsequently prepare Z-QG-modified DNA by using the
polymerase chain reaction (PCR). A UV-absorbance spec-
trum of Z-QG-dUTP showed absorbance maxima at 240
and 296 nm, which are representative of a pyrimidine ring
modified with an exocylic double bond.[22,23] The concentra-
tion of Z-QG-dUTP was determined by the absorbance at
296 nm, and a series of concentrations of dTTP (0–80%)
was substituted with Z-QG-dUTP in the PCR mixture. The
obtained Z-QG-modified DNAs were designated as Z-QG-
ACHUTNGERN(NUG 0–80%)–DNA (e.g., Z-QGACHUTNGTREN(NGUN 40%)–DNA indicates that this
probe was prepared with 40% Z-QG-dUTP and 60%
dTTP). We examined three DNA polymerases for the incor-
poration of the dUTP analogue into DNA: KOD FX, KOD
Dash (a mixture of KOD and KOD exoÀ), and Ex Taq.
DNA that contained a fragment of mouse sonic hedgehog
(Shh) cDNA (300 bp) was amplified and analyzed by using
agarose gel electrophoresis. The result demonstrated that a
uniform mobility of Z-QG-DNA was obtained in each PCR
solution (Figure 1). The mobility of the Z-QG-DNA de-
creased according to the percentage content of the Z-QG-
dUTP in the PCR mixtures. Mobility shifts were smaller in
the Z-QG-DNAs amplified by Ex Taq DNA polymerase (a
family A-type DNA polymerase). Moreover, when Ex Taq
DNA polymerase was used, the 80% substitution of dTTP
with Z-QG-dUTP in the PCR reaction mixture led to a de-
crease in the yield to (28Æ8)% compared with the PCR re-
action with 0% dTTP substitution. In contrast, relative
Scheme 1. Microbial transglutaminase (MTG)-mediated construction of a
DNA–(enzyme)n probe. An MTG substrate (benzyloxycarbonyl-l-gluta-
minylglycine, Z-QG) was incorporated into a probe DNA during PCR
with Z-QG-modified dUTP. Glutamine (Q) on the Z-QG-labeled DNA
was site-specifically combined with MTG-reactive lysine (K)-tagged Py-
rococcus furiosus alkaline phosphatase (PfuAP). The resulting DNA
probe (i.e., DNA–(PfuAP)n conjugate) was employed to detect a target
DNA immobilized on a membrane. In the color-development reaction,
PfuAP catalyzed dephosphorylation of 5-bromo-4-chloro-3-indolyl phos-
phate (BCIP), thereby priming the membrane for blue precipitates to
appear on the membrane following the reduction of p-nitroblue tetrazoli-
um chloride (NBT).
Results and Discussion
Figure 1. Agarose gel electrophoresis analysis of Z-QG-DNA synthesized
by family B- (KOD FX and KOD Dash) and family A-type (Ex Taq)
DNA polymerases. M: DNA Marker. 1) 0%, 2) 20%, 3) 40%, 4) 60%,
and 5) 80% of dTTP was substituted by Z-QG-dUTP in the PCR mix-
tures.
TGase (EC 2.3.2.13) catalyses an acyl-transfer reaction be-
tween the e-amino group in lysine (K) and the g-carboxa-
mide group in a peptide-bound glutamine (Q).[15,16] Our pre-
vious research revealed that the stable and Ca2+-independ-
5388
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5387 – 5392

DNA–(Enzyme)n Probe
FULL PAPER
yields of (55Æ6)% (KOD FX) and (65Æ8)% (KOD Dash)
were obtained when family B-type DNA polymerases were
tested in the same manner, thus indicating that the family B-
type enzymes accepted the Z-QG-dUTP as a substrate of
the DNA extension reaction, in addition to the utilization of
the Z-QG-modified DNA as a template. The results are sim-
ilar to previous reports that showed that the family B-type
DNA polymerases display superior incorporation efficien-
cies of a wide variety of dNTP analogues.[24–29]
An incorporation ratio of Z-QG-deoxyuridine (Z-QG-
dU) in each Z-QG-DNA was quantified by using reversed-
phase HPLC after fragmentation of the Z-QG-DNA pre-
pared with KOD FX to nucleosides (Figure S3 in the Sup-
porting Information). The incorporation ratio increased as
the percentage of Z-QG-dUTP increased in a PCR mixture
(Figure 2). However, actual substitution ratios of Z-QG-dU
Z-QG-DNA probes using MTG, an acyl-acceptor tag se-
quence, MKHKGGGSGGGSGS (the underlined K is a
cross-linkable site recognized by MTG) was fused to the N
terminus of PfuAP (NK14-PfuAP, Figure S5 in the Support-
ing Information).[30] In a previous study, bacterial alkaline
phosphatase (BAP) with an MKHKGS tag at the N and C
termini were successfully linked to a 5’-end, Z-QG-labeled
oligonucleotide by MTG at pH 8.0,[12] and the DNA–BAP
conjugate retained both oligonucleotide and enzymatic func-
tion following MTG treatment. In this study, a Z-QG-DNA
probe contains a number of Z-QG moieties on a strand.
Thus, we examined the coupling efficiency of a series of
molar ratios of NK14-PfuAP to the Z-QG moiety by the
MTG reaction. Following heat denaturation, Z-QG-DNA
was incubated with NK14-PfuAP and MTG at pH 8.0 for
30 min. Subsequently, the mobility shifts of the DNA bands
were examined in a 1.5% agarose gel. The mobility of
Z-QG-DNA was gradually retarded according to the incre-
ment of NK14-PfuAP in a MTG reaction mixture. The shift
finally reached a plateau at a band size of around 2 kb when
the molar ratio of protein/ssDNA was 600 and 420 for
Z-QGACHTNUTRGENNUG(40%)-DNA and Z-QGHCATUNGTRNE(NUGN 80%)-DNA, respectively
(Figure S6 in the Supporting Information).
The NK14-PfuAP was more efficiently cross-linked on
Z-QG-DNA at pH 6.0 for 30 min with a lower protein/DNA
ratio ([protein]/
ACHTUNGTRENUN[NG ssDNA]=120) (Figure 3). Since the single-
stranded Z-QG(80%)-DNA contains 58 Z-QG moieties on
AHCTUNGTRENNUNG
Figure 2. Fragmentation analysis of the incorporation ratio of Z-QG-
dUTP in Z-QG-DNA synthesized using KOD FX DNA polymerase. Per-
centages of nucleosides determined using reversed-phase HPLC analyses
are presented from the left: deoxyadenosine (dA), deoxycytidine (dC),
deoxyguanosine (dG), deoxythymidine (dT), and Z-QG-deoxyuridine
ACHTUNGTRENNUNG(Z-QG-dU). Data represent averages of three repeats, and standard devi-
ations were within Æ1.1%.
to the total deoxythymidine in the DNA strands were slight-
ly smaller than the substitution ratios of Z-QG-dUTP to
dTTP in the PCR mixture. From the fragmentation analysis,
Figure 3. Agarose gel electrophoresis of DNA–(PfuAP)n conjugates pre-
pared in a phosphate buffer (pH 6.0). a) DNA was stained with SYBR
Green II. b) NK14-PfuAP was detected by the PfuAP-catalyzed NBT/
BCIP coloring reaction. Z-QG-DNAs were synthesized using KOD FX
it was estimated that single-stranded Z-QG
ACHTUNGTRENN(UNG 40%)- and
Z-QG(80%)-DNAs contained 20 and 58 Z-QG moieties on
ACHTUNGTRENNUNG
average, respectively. Melting temperature (Tm) analysis
using SYBR Green I revealed that the Z-QG modification
lowered the Tm value of Z-QG-DNA, and the Tm values of
DNA polymerase. In the MTG reaction mixtures, 1) Z-QGACHTUNGTRENNUNG
2) Z-QG
ACHTUNGERTN(UNGN 20%)-DNA, 3) Z-QGACHTUNGTREG(NNUN 40%)-DNA, 4) Z-QGACTHUNTGRENNUGN
5) Z-QGAHCNUTGTRENNUNG
of [protein]/CAHTUNGTRENNNUG
Z-QGACHTUNGTRENNUNG(0%)-, Z-QGACHTUNTGRENNUG(40%)-, and Z-QGACTHNUGTREN(NNGU 80%)-DNA in 2ꢁ
SSC (0.3m NaCl, 0.03m sodium citrate, pH 7.0) buffer were
97, 94, and 918C, respectively (Figure S4 in the Supporting
Information).
average, the cross-linking of PfuAP and Z-QG was conduct-
ed at the molar ratio of [protein]/[Z-QG]=2. The calculated
Then we examined the labeling of a signaling enzyme, al-
kaline phosphatase, on a Z-QG-DNA. An alkaline phospha-
tase from Pyrococcus furiosus (PfuAP) was employed for
the signal amplification in the dot blot detection of the
target DNA, because this enzyme is from a hyperthermo-
phile and was expected to be stable against heat or chemical
denaturants applied during the hybridization and washing
processes. For the site-specific cross-linking of PfuAP and
ACHTUNGTRENNUNG
isoelectric point (pI) of NK14-PfuAP is 5.3. Lowering the
pH value of the reaction mixture may reduce the electro-
static repulsion between DNA and the protein, thereby ena-
bling the coupling of multiple NK14-PfuAP to the dense
spot. Under the reaction conditions, the number of PfuAP
molecules per DNA is likely to depend on the increment of
the Z-QG moiety in the DNA.
Chem. Eur. J. 2011, 17, 5387 – 5392
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5389

N. Kamiya et al.
A coupling efficiency of protein on DNA was examined
using Cy5-labeled NK14-PfuAP and 5’-fluorescein-labeled
Z-QG-DNA. The protein/DNA ratio of the conjugate in-
creased as the Z-QG moiety increased in the DNA. Averag-
es of 14 and 21 proteins were cross-linked on single-stranded
formed at 508C immediately after washing of the unhybri-
dized probes. P. furiosus is a hyperthermophile, and the opti-
mum temperature of NK14-PfuAP activity was around
708C (Figure S7 in the Supporting Information).[31] Howev-
er, when we performed the color development reaction at
room temperature (ꢀ20–258C), 50, and 608C, the strongest
signal was observed at 508C, thus implying that the double
strands dissociated at the higher temperature. The DNA–
Z-QG
N
and
Z-QG
G
respectively
(Figure 4). The amount of labeled proteins did not increase
(PfuAP)n conjugate obtained from Z-QGACTHNURGTNE(UNG 40%)-DNA
showed the highest sensitivity and specificity, and we could
clearly detect at least 100 fg (5ꢁ10À19 mol=104 copies) of
the target DNA (Figure 5b). The probe was highly specific
and signals were not observed from the noncomplementary
protamine DNA. The results indicate that the present
DNA–(PfuAP)n conjugate is a robust probe applicable to
DNA detection under harsh hybridization conditions.
In a previous study, horseradish peroxidase was randomly
prelabeled to DNA probes by using p-benzoquinone, poly-
ethyleneimine, and glutaraldehyde as cross-linkers, and the
conjugate probes could detect 4.1 pg of 2.06 kb plasmid frag-
ments, with a hybridization buffer that contained 6% poly-
ethylene glycol as a hybridization enhancer.[32] For compari-
son of our enzymatic strategy with a chemical approach, an
alkaline phosphatase-labeled DNA probe was prepared by
chemical cross-linking using a commercial kit. Hybridization
was performed in the hybridization buffer included in the
kit. The probe showed the detection limit of 10–100 pg
when NBT and BCIP were used for coloring reaction in-
stead of a chemiluminescent substrate recommended by the
manufacturer (i.e., CDP-Star) (Figure 5a). The higher sensi-
tivity with the present new probe can be attributed to the
superior DNA hybridization ability due to the homogeneous
cross-linking sites between the DNA template and signaling
enzymes based on enzymatic site-specific conjugation. Cur-
rently, the most widely used probe is the DIG-labeled
probe. Previously, postlabeling of enzymes to probes that re-
cruit an anti-DIG-antibody–alkaline phosphatase conjugate
for signal generation could detect 0.08 pg of a 1.2 kb DNA
target.[10] Our experiment revealed that the DNA–(PfuAP)n
probe exhibited comparable performance to the DIG probe
system that could detect a minimum of 10–100 fg of the
target DNA (Figure 5c).
Figure 4. Coupling efficiency of Cy5-NK14-PfuAP on denatured fluores-
cein-Z-QG-DNAs. a) Concentrations of fluorescein-Z-QG-DNAs and
Cy5-NK14-PfuAP after gel filtration using sephacryl microspin columns,
then b) molar ratios of protein to ssDNA were calculated. The green
dotted line in Figure 4a implies the background level of free Cy5-NK14-
PfuAP eluted when the MTG reaction mixture incubated with Z-QG-
For in situ hybridization in cells and tissues, fluorescent
dyes or signaling enzyme in combination with precipitating
chromophores are often used to visualize the target mRNA.
The novel DNA–(PfuAP)n conjugate probe detected a com-
parable level of the target DNA to the DIG probe using
precipitating chromophores. Thus, future application of this
novel conjugate probe to in situ hybridization would be ex-
pected. In addition, since the antibody-free DNA-detection
protocol was simple and cost-effective, the protocol could
be widely accepted in the field of molecular biology and
medical diagnostics.
ACHTUNGTRENNUNG(0%)-DNA was applied. Data represent averages of three repeats, and
bars show standard deviations.
by the further incubation with MTG. The amount of protein
cross-linked on DNA was smaller than that of the Z-QG
moieties determined by HPLC analysis, presumably because
of steric hindrance upon the enzymatic labeling of DNA
with a relatively large protein. The insertion of a spacer that
separates the reactive sites (i.e., Z-QG moieties) from the
DNA strand should represent an effective approach to fur-
ther control the enzymatic DNA–(protein)n conjugation.
The DNA–(protein)n conjugate probes were then subject-
ed to dot blot hybridization without the removal of unconju-
gated PfuAP. The color development reaction was per-
5390
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5387 – 5392

DNA–(Enzyme)n Probe
FULL PAPER
Figure 5. Detection of Shh DNA with a) an alkaline phosphatase-labeled DNA probe prepared using a commercial kit, b) Z-QGACTHNURGTNE(GNU 40%)-DNA–(PfuAP)n
conjugate probe, and c) the DIG system. A series of dilutions (100 pg–1 fg) of Shh DNAs (upper) and prm DNAs (lower) were spotted onto Hybond-
N+ membranes. Intensity profiles of signals were obtained using ImageJ software. Details on the experimental procedures used are provided in the Sup-
porting Information.
in acetonitrile/ultrapure water/trifluoroacetic acid (TFA) (50:50:0.1 v/v),
then a MALDI-TOF mass spectrum was recorded using a Autoflex-III
Conclusion
(Bruker Daltonics; Billerica, MA) (Figure S2b). HRMS: m/z: calcd. for
À
C27H36N6O19P3 [MÀH]À: 841.13; found: 840.87. A UV-absorption spec-
A method to prepare a DNA–(enzyme)n conjugate for hy-
bridization was developed. Substrate peptides of MTG that
contained Lys and Gln were introduced at the N terminus
of PfuAP and at the 5-position on the pyrimidine of dUTP,
trum was recorded using a V-570 VU/VIS/NIR spectrometer (JASCO) in
a 500 mL cuvette (Figure S2c). The concentration of Z-QG-dUTP was de-
termined by the absorbance at 296 nm using the molar extinction coeffi-
cient of 7.1ꢁ103 mÀ1 cmÀ1
[22,23]
.
respectively. The DNA–(enzyme)n conjugate probe exhibit-
ed comparable sensitivity to a DIG probe without cumber-
some bound/free processes associated with an hapten–anti-
body reaction prior to color development. The new synthetic
strategy for DNA–enzyme conjugates with a novel architec-
ture will lead the way to easy, sensitive, and cost-effective
filter hybridization assays. Further studies on the design and
the validation of RNA–enzyme conjugates for in situ hybrid-
ization are in progress in our laboratory.
Determination of the incorporation ratio of Z-QG-deoxyuridine: Purified
Z-QG-DNA from a PCR reaction mixture (100 mL, 5–10 mg) was enzy-
matically digested to deoxynucleosides. The Z-QG-0DNA was incubated
with nuclease P1 (2 U) from Penicillium citrinum in sodium acetate
(10 mL, 30 mm, pH 4.5) and zinc chloride (5 mm) for 3 h at 378C. The so-
lution was further incubated with phosphodiesterase I (0.002 U) from
Crotalus adamanteus venom and with alkaline phosphatase (20 U) from
calf intestine in Tris-HCl (20 mL, 100 mm, pH 9.5) and MgCl2 (10 mm) for
3 h at 378C.[33,34] After centrifugation, the supernatant was analyzed by
reversed-phase HPLC coupled with a photodiode array (PDA) detector,
using a 4.6ꢁ250 mm COSMOSIL 5C18-AR column (Nacalai tesque) and
a mobile phase of triethylammonium acetate (0.1m, pH 7.0) and acetoni-
trile. The HPLC gradient conditions employed were as follows: 0% ace-
tonitrile (10 min) followed by a linear gradient of 0 to 30% acetonitrile
(30 min) at a flow rate of 1 mLminÀ1. The effluent was monitored by ab-
sorbance at 300 nm for Z-QG-deoxyuridine and at 260 nm for the other
deoxynucleosides (Figure S3 in the Supporting Information). The relative
concentration of each deoxynucleoside was determined using the peak
area of each reference standard.
Experimental Section
Synthesis of Z-QG-dUTP: Z-QG and AA-dUTP were obtained from
Sigma–Aldrich (St. Louis, MO). Z-QG (50 mm) was incubated for 20 h in
DMF (160 mL) that contained N-hydroxysuccinimide (100 mm) and N,N’-
diisopropylcarbodiimide (100 mm). Subsequently, AA-UTP (160 mL,
5 mm) in borate buffer (100 mm, pH 8.8) was added, and the mixture was
incubated for 12 h to obtain Z-QG-dUTP. For the purification of Z-QG-
dUTP using reversed-phase HPLC, a 10 mmꢁ250 mm Inertsil ODS-3
column (GL Sciences, Tokyo, Japan) and a mobile phase of triethylam-
monium acetate (0.1m, pH 7.0) and acetonitrile were used. The HPLC
gradient conditions employed were a linear gradient of 2 to 12% acetoni-
trile (5 min), followed by a linear gradient of 12 to 27% acetonitrile
(30 min). A flow rate of 5 mLminÀ1 was used. The effluent was moni-
tored by absorbance at 260 nm. The effluent at 18 min was isolated and
dried by lyophilization (Figure S2a in the Supporting Information).
Z-QG-dUTP was dissolved in Tris-HCl (10 mm, pH 7.5; Tris=tris(hy-
droxymethyl)aminomethane). The Z-QG-dUTP solution (1 mL) was di-
Determination of the protein/ssDNA ratio of the conjugates: PCR pri-
mers labeled with fluorescein at the 5’-ends were synthesized at Tsukuba
Oligo Service (Ibaraki, Japan). After PCR using KOD FX DNA poly-
merase, fluorescein-Z-QG-DNAs were purified using a QIAquick PCR
purification kit (Qiagen). Amino groups in a NK14-PfuAP were random-
ly labeled with Cy5 dyes by using Cy5-succinimidyl esters from GE
Healthcare (Amersham Cy5 Mono-Reactive Dye Pack), according to the
manufacturerꢂs instructions. Unreacted dyes were separated using PD-10
columns (GE Healthcare) with the elution buffer of Tris-HCl (10 mm,
pH 8.0). An absorption spectrum of the Cy5-NI14-PfuAP was recorded
using a NanoDrop ND-1000 device. The labeling efficiency was calculat-
ed using the molar extinction coefficients of 29000mÀ1 cmÀ1 at 280 nm for
luted with a-cyano-4-hydroxycinnamic acid (CHCA) (9 mL, 10 mgmLÀ1
)
Chem. Eur. J. 2011, 17, 5387 – 5392
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5391

N. Kamiya et al.
NK14-PfuAP and 250000mÀ1 cmÀ1 at 650 nm for the Cy5 dye. An esti-
mated dye/protein ratio was 1.1:1.
[12] J. Tominaga, Y. Kemori, Y. Tanaka, T. Maruyama, N. Kamiya, M.
Goto, Chem. Commun. 2007, 401–403.
[13] B. P. Duckworth, Y. Chen, J. W. Wollack, Y. Sham, J. D. Mueller,
T. A. Taton, M. D. Distefano, Angew. Chem. 2007, 119, 8975–8978;
Angew. Chem. Int. Ed. 2007, 46, 8819–8822.
[14] M. A. Jongsma, R. H. G. M. Lithens, Proteomics 2006, 6, 2650–2655.
[15] M. Griffin, R. Casadio, C. M. Bergamini, Biochem. J. 2002, 368,
377–396.
[16] L. Lorand, R. M. Graham, Nat. Rev. Mol. Cell Biol. 2003, 4, 140–
156.
[17] K. Yokoyama, N. Nio and Y. Kikuchi, Appl. Microbiol. Biotechnol.
2004, 64, 447–454.
Cy5-NK14-PfuAP (12 mm) and 100 nm of denatured fluorescein-Z-QG-
DNA ([protein]/ACHTUNGTRENNUNG[ssDNA]=120) were subjected to the cross-linking reac-
tion in a phosphate buffer (50 mL, pH 6.0). After incubation at 408C for
30 min, unconjugated protein was separated using a Microspin S-400 HR
column (GE Healthcare). Fluorescence spectra of the effluents were
measured using an LS-55 Fluorescence Spectrometer (PerkinElmer).
Molar ratios of Cy5-NK14-PfuAP to fluorescein-Z-QG-DNA were calcu-
lated using the fluorescence intensity at 520 nm for fluorescein (ex/
480 nm) and at 670 nm for Cy5 (ex/640 nm). Free Cy5-NK14-PfuAP
((525Æ83) nm) was eluted when the MTG reaction mixture incubated
[18] Y. Zhu, J. Tramper, Trends Biotechnol. 2008, 26, 559–565.
[19] H. Ando, M. Adachi, K. Umeda, A. Matsuura, M. Nonaka, R.
Uchio, H. Tanaka, M. Motoki, Agric. Biol. Chem. 1989, 53, 2613–
2617.
with Z-QGACHTUNGTRENNUNG(0%)-DNA was applied to gel filtration. Thus, the background
data was subtracted from the concentration of the cross-linked Cy5-
NK14-PfuAPs; then the protein/ssDNA ratios were calculated.
[20] R. Pasternack, H.-P. Laurent, T. Rꢄth, A. Kaiser, N. Schçn, H.-L.
Fuchsbauer, Anal. Biochem. 1997, 249, 54–60.
[21] X. Yang, X.-W. Shi, Y. Liu, W. E. Bentley, G. F. Payne, Langmuir
2009, 25, 338–344.
Acknowledgements
[22] P. R. Langer, A. A. Waldrop, D. C. Ward, Proc. Natl. Acad. Sci.
USA 1981, 78, 6633–6637.
[23] T. M. Herman, B. J. Fenn, Methods Enzymol. 1990, 184, 584–588.
[24] S. H. Weisbrod, A. Marx, Chem. Commun. 2008, 5675–5685.
[25] Z. Zhu, J. Chao, H. Yu, A. A. Waggoner, Nucleic Acids Res. 1994,
22, 3418–3422.
This research was supported mainly by the Science and Technology Incu-
bation Program in Advanced Regions from the Japan Science and Tech-
nology Agency (JST) (to N. K.) and in part by a Grant-in-Aid for the
Global COE program, “Science for Future Molecular Systems” from the
Ministry of Education, Culture, Sports, Science and Technology of Japan.
[26] a) G. Giller, T. Tasara, B. Angerer, K. Mꢄhlegger, M. Amacker, H.
Winter, Nucleic Acids Res. 2003, 31, 2630–2635; <lit b>T. Tasara,
B. Angerer, M. Damond, H. Winter, S. Dçrhçfer, U. Hꢄbscher, M.
Amacker, Nucleic Acids Res. 2003, 31, 2636–2646.
[27] D. Cherkasov, T. Biet, E. Bꢅuml, W. Traut, M. Lohoff, Bioconjugate
Chem. 2010, 21, 122–129.
[28] P. M. E. Gramlich, C. T. Wirges, A. Manetto, T. Carell, Angew.
Chem. 2008, 120, 8478–8487; Angew. Chem. Int. Ed. 2008, 47, 8350–
8358.
[29] M. Kuwahara, K. Hanawa, K. Ohsawa, R. Kitagawa, H. Ozaki, H.
Sawai, Bioorg. Med. Chem. 2006, 14, 2518–2526.
[30] J. Tominaga, N. Kamiya, S. Doi, H. Ichinose, T. Maruyama and M.
Goto, Biomacromolecules 2005, 6, 2299–2304.
[31] a) S. Zappa, J. Boudrant, E. R. Kantrowitz, J. Inorg. Biochem. 2004,
98, 575–581; <lit b>S. Zappa, J.-L. Rolland, D. Flament, Y.
Geguen, J. Boudrant, J. Dietrich, Appl. Environ. Microbiol. 2001,
67, 4504–4511.
[32] M. Renz, C. Kurz, Nucleic Acids Res. 1984, 12, 3435–3444.
[33] A. Chango, A. M. Adbel Nour, C. Niquet, F. J. Tessier, Med. Princ.
Pract. 2009, 18, 81–84.
[1] G. A. Ricca, J. Taylor, J. E. Kalinyak, Proc. Natl. Acad. Sci. USA
1982, 79, 724–728.
[2] J. Messing, R. Crea, P. H. Seeburg, Nucleic Acids Res. 1981, 9, 309–
321.
[3] M. S. Urdea, B. D. Warner, J. A. Running, M. Stempien, J. Clyne, T.
Horn, Nucleic Acids Res. 1988, 16, 4937–4956.
[4] J. J. Leary, D. Brigati, D. C. Ward, Proc. Natl. Acad. Sci. USA 1983,
80, 4045–4049.
[5] A. H. N. Hopman, J. Wiegant, G. I. Tesser, P. Van Duijn, Nucleic
Acids Res. 1986, 14, 6471–6488.
[6] R. Urrutia, M. A. McNiven, B. Kachar, J. Biochem. Biophys. Meth-
ods 1993, 26, 113–120.
[7] D. M. Hammond, A. Manetto, J. Gierlich, V. A. Azov, P. M. E.
Gramlich, G. A. Burley, M. Maul, T. Carell, Angew. Chem. 2007,
119, 4262–4265; Angew. Chem. Int. Ed. 2007, 46, 4184–4187.
ˇ
[8] V. Raindlovꢃ, R. Pohl, M. Sanda, M. Hocek, Angew. Chem. 2010,
122, 1082–1084; Angew. Chem. Int. Ed. 2010, 49, 1064–1066.
[9] S. A. McCarthy, A. DePaola, D. W. Cook, C. A. Kaysner, W. E. Hill,
Lett. Appl. Microbiol. 1999, 28, 66–70.
[10] C. Kessler, H.-J. Hçltke, R. Seibl, J. Burg, K. Mꢄhlegger, Biol.
Chem. Hoppe-Seyler 1990, 371, 917–927.
[34] J. Gierlich, K. Gutsmiedl, P. M. E. Gramlich, A. Schmidt, G. A.
Burley, T. Carell, Chem. Eur. J. 2007, 13, 9486–9494.
[11] a) C. M. Niemeyer, Angew. Chem. 2010, 122, 1220–1238; Angew.
Chem. Int. Ed. 2010, 49, 1200–1216; b) W. P. Heal, E. W. Tale, Org.
Biomol. Chem. 2010, 8, 731–738.
Received: December 28, 2010
Published online: April 5, 2011
5392
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5387 – 5392