Atom arrangement strategy for designing a turn-on 1H magnetic resonance probe: A dual activatable probe for multimodal detection of hypochlorite

Article abstract of DOI:10.1039/c1cc12044a

The first dual activatable hypochlorite (^-OCl)-sensing probe was developed, based on a new proof-of-concept design involving signal-activatable ¹H chemical probes using the triple-resonance NMR technique. The probe enabled fluorescen

Full text of DOI:10.1039/c1cc12044a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 1565–1567
www.rsc.org/chemcomm
COMMUNICATION
1
Atom arrangement strategy for designing a turn-on H magnetic
resonance probe: a dual activatable probe for multimodal detection
of hypochloritewz
Tomohiro Doura,a Hiroshi Nonakaa and Shinsuke Sando*ab
Received 11th April 2011, Accepted 27th April 2011
DOI: 10.1039/c1cc12044a
The first dual activatable hypochlorite (ꢀOCl)-sensing probe was
developed, based on a new proof-of-concept design involving
The first challenge in developing a dual-modal probe for
turn-on sensing of ROS is to design a signal-activatable MR
probe. Particularly, an activatable 1H-MR probe is attractive for
MRS or MRI because 1H has the highest sensitivity of available
1
signal-activatable H chemical probes using the triple-resonance
NMR technique. The probe enabled fluorescence–1H MR dual
turn-on detection of ꢀOCl in solution and in crude tissue extracts.
1
nuclei. However, H detection typically suffers from nonnegli-
gible interference from the high background emanating from a
1
variety of endogenous H-containing biological compounds.
Reactive oxygen species (ROS) play important roles in physio-
logical and pathological events in the cell. Consequently, the
demand for ROS-sensing and -imaging probes has increased.1
These probes are powerful tools for elucidating the precise
roles of ROS and may also be useful for medical diagnosis.
Most ROS bioimaging tools involve fluorescent probes
because they can be used in conjunction with fluorescence
microscopy. Activatable fluorescent probes,2 which are otherwise
nonfluorescent but fluoresce upon reaction with the target
species, have been developed for direct monitoring of ROS in
live cells.1 Although fluorescent probes have superior sensitivity
and spatiotemporal resolution, their use for ex vivo and in vivo
applications is limited by low penetration of excitation or
emission light, with the exception of potent near-infrared
fluorescence3 and bioluminescence probes.4 A magnetic resonance
(MR)-based technique such as MR spectroscopy (MRS) or MR
imaging (MRI) is appropriate for this purpose because a MR
probe could be detected even if it were situated deep within the
body.5 However, the disadvantages of MR are that it has low
spatiotemporal resolution compared with the fluorescence-based
approach. In this context, ideal chemical probes are those that can
be detected in a fluorescence–MR dual-modal fashion so that an
appropriate detection mode can be selected depending on the
situation, e.g., MR could be used to obtain anatomic information
and fluorescence could be used when high spatiotemporal
resolution is required on the targeted region.6,7 Herein, we present
the first description of a ROS-sensing probe for fluorescence–
1H-MR dual-modal detection. The probe achieved dual turn-on
sensing of hypochlorite (ꢀOCl) with high specificity.
To overcome these problems, we focused on a one-dimensional
triple-resonance 1H NMR technique. Triple resonance is used
to correlate three NMR-active nuclei with different Larmor
frequencies8 and has been applied mostly to structural analysis
of biopolymers, except for precedents,9,10 including studies
from our laboratory.10 For example, when the pulse scheme
allows the magnetic coherence of 1H to transfer to two
successive 13C nuclei with different Larmor frequencies
13
(13C and C0) through scalar couplings, only the 1H in the
specific sequence 1H–13C–13C0 is detectable. Therefore, we hypo-
thesized that if we could design a 13C-labelled chemical probe,
which initially has no detectable 1H–13C–13C0 atom sequence but
1
develops an active H–13C–13C0 sequence upon reaction with
ROS, it should function as an off-to-on-type 1H probe under
triple resonance conditions (Fig. 1). This is a completely new
1
mode for designing signal-activatable H MR probes.
In addition to signal activation, the triple resonance technique
affords selectivity. Because of the low natural abundance of 13
C
(1.1%), the probability of a naturally occurring H–13C–13C0
sequence is o0.01%. Therefore, this technique could markedly
improve the selectivity of the target 1H connecting with 13C–13C0
1
1
because of low background H signals.
Along this concept, we designed 13C-labelled 4-methyl-3,4-
dihydrocoumarin (HCm in Fig. 2a) as a probe scaffold. Under
1
the NMR pulse sequence of H–{13C–13C0(sp2)} for detection
of an 1H bound to an aliphatic 13C, which in turn is bound to
13 (sp2)
sp2 C0
on a conjugated double bond, HCm has no
1
detectable 1H because of the lack of a H–13C–13C0(sp2)
sequence. On the other hand, if the HCm structure is oxidized
a INAMORI Frontier Research Center, Kyushu University,
744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan.
E-mail: ssando@ifrc.kyushu-u.ac.jp
b PRESTO, Japan Science and Technology Agency, 4-1-8 Honcho,
Kawaguchi, Saitama 332-0012, Japan
w This article is part of the ‘Emerging Investigators’ themed issue for
ChemComm.
z Electronic supplementary information (ESI) available: Fig. S1–S5
1
Fig. 1 A strategy for designing a signal activatable H MR probe.
and methods. See DOI: 10.1039/c1cc12044a
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 1565–1567 1565

View Article Online
473 nm = 19).11 The reaction of HCm3 with ꢀOCl was clearly
dose-dependent (Fig. 2b) and fast (Fig. 2c). Judging from a
comparison with the fluorescence intensity of authentic Cm3, it
was revealed that approximately 6.1% of HCm3 (10 mM) was
converted to a fluorescent coumarin scaffold by reaction with
3 equivalents of ꢀOCl. In addition, a detailed product analysis of
the reaction solution (HCm3 10 mM,ꢀOCl 30 mM) using
HPLC-fluorescence suggested that the reacted solution contained
several fluorescent products; one was Cm3 and others were
coumarin derivatives produced probably by further reaction of
the Cm3 with excess ꢀOCl (Fig. S2a, ESIz). Although the
conversion yield was not so high, the probe allowed a practical
ꢀOCl detection because of high sensitivity of produced fluorescent
coumarins. In fact, the HCm3 enabled clear off-to-on type
visualization of ꢀOCl under UV light (Fig. 2d) and reached the
detection limit of 0.2 mM under our experimental conditions. In
addition, the HCm3 showed fluorescence enhancement specifically
to ꢀOCl compared with other biologically important ROS
(H2O2, ROOꢁ, O2 ꢀ, and ꢁOH) and reactive nitrogen species
ꢁ
(RNS; NO and ONOOꢀ) (Fig. 2e). When the fluorescence
intensity of HCm3 in the presence of ꢀOCl was expressed as
100%, the fluorescence of HCm3 in the presence of all other
species was less than 2% (reaction of 10 mM of HCm3 with 30 mM
of ROS or RNS). Product analyses by HPLC-UV revealed that
HCm3 reacted with most of the ROS tested and produced a
variety of products (Fig. S2b, ESIz). In addition, it was found that
the HCm3 probe was not stable under physiological conditions to
give a hydrolysed product (Fig. S2b, ESIz). However, such
products were non-fluorescent except for that produced by
reaction with ꢀOCl, thus resulting in high specificity in fluores-
cence analysis (Fig. 2e and Fig. S3, ESIz).
Fig. 2 (a) Chemical structures of 4-methyl-3,4-dihydrocoumarin
derivatives (HCm1–4) and a hypothesized reaction scheme showing
how the dihydrocoumarin probe functions as a dual turn-on probe
against ROS. (b) Fluorescence spectra (excitation at 342 nm) of HCm3
(10 mM) after incubation with ꢀOCl (0, 1.5, 3, 4.5, 6, 7.5, 9, 10.5 mM
from bottom to top) in phosphate buffer (pH 7.4, 100 mM) containing
150 mM NaCl and 0.1% DMF at 37 1C. Inset: ꢀOCl-dependent linear
increase of integral fluorescence. (c) Time course of fluorescence
intensity of HCm3 (10 mM, excitation at 342 nm and emission at
442 nm) upon addition of one equivalent of ꢀOCl. The arrow indicates
the addition of ꢀOCl. (d) Fluorescent images of HCm3 (10 mM) in the
absence or presence of three equivalents of ꢀOCl. (e) Relative fluorescence
intensity (%) of HCm3 (10 mM) (excitation at 342 nm and emission at
442 nm) incubated with various ROS (3 equivalents) and reactive nitrogen
species (RNS, 3 equivalents); fluorescence intensity in the presence of
ꢀOCl was set at 100%. Typical fluorescence spectra of HCm3 with various
ROS and RNS are shown in Fig. S3 (ESIz).
Having developed an ꢀOCl-specific fluorescent HCm3
1
probe, we tried to add an H-activatable moiety to the probe.
As 3,4-13C2-labelled ethyl acetoacetate is commercially available,
13C-labelled HCm3 (HCm3-13C2) and Cm3 (Cm3-13C2) can be
synthesized according to the procedure in Scheme 1.
First, we measured the NMR spectra of Cm3-13C2. The
conventional 1H NMR spectrum of Cm3-13C2 showed aromatic,
amino, and aryl 1H (left in Fig. 3a, CDCl3). In contrast, only aryl
1H was detected in the 1H–{13C–13C0(sp2)} triple resonance experi-
ment (right in Fig. 3a, CDCl3); the detection limit was 1 mM in
D2O under our experimental conditions. We then examined ꢀOCl
detection using HCm3-13C2 as a probe. Again, our purpose was
to develop an off-to-on-type fluorescence–1H MR dual-modal
probe that would ideally be undetected until reaction with the
target ROS. The HCm3-13C2 construct is ideal in this respect.
by ROS to produce an aromatized coumarin structure
1
(Cm in Fig. 2a), the detectable H–13C–13C0(sp2) sequence is
formed (red atoms in Fig. 2a). Because the coumarin scaffold
is a fluorescent chromophore, this compound functions as a
dual activatable fluorescence–1H-MR probe against ROS.
First, we investigated whether HCm would react with ROS to
produce aromatized Cm as proposed in Fig. 2a. HCm derivatives,
where R is OH, OCH3, NH2, or N(CH3)2, were synthesized by
hydrogenation of Cm derivatives, which were prepared by simple
condensation of the corresponding phenols with ethyl acetoace-
tate. Fluorescence spectra for the otherwise very weakly fluores-
cent HCm1–4 were measured after incubation with one equivalent
of hypochlorite (ꢀOCl) (Fig. S1, ESIz). HCm3 afforded the highest
fluorescence enhancement (Ion/Ioff (HCm3) at 442 nm = 45)
1
HCm3-13C2 was silent under the triple resonance H conditions
(upper spectrum in Fig. 3b). In contrast, a single signal was
produced after incubation with ꢀOCl (lower spectrum in Fig. 3b)
with concomitant fluorescence enhancement (Fig. S4, ESIz).
Scheme 1 Synthesis of HCm3-13C2. (a) Ethyl chloroformate, Et2O,
rt (48%); (b) (i) ethyl acetoacetate-3,4-13C2, H2SO4, rt, (ii) glacial
acetic acid, 100 1C (25% in two steps); (c) H2, Pd–C, EtOH, rt (50%).
among HCm1–4 (Ion/Ioff (HCm1) at 449 nm = 5.5,
Ion/Ioff (HCm2) at 383 nm = 1.2, and Ion/Ioff (HCm4) at
c
1566 Chem. Commun., 2012, 48, 1565–1567
This journal is The Royal Society of Chemistry 2012

View Article Online
achieved using a new proof-of-concept with the triple resonance
NMR technique. There are at least three advantages to our
method. The first is its selectivity. Because of the low natural
13
1
abundance of C, the H–13C–13C0 moiety renders the probe
extremely selective. The second advantage is that our probe is
signal activatable. Thus, our ‘‘switch-on’’ 1H MR probe is
otherwise silent and is rendered active only when a detectable
1H–13C–13C0 sequence is produced upon reacting with the
target. The third is versatility. The technique is not limited to
(sp2)
but can be applied to a variety of atom sequences
H–C–C0
such as H–C–N. This strategy of ‘‘atom arrangement’’ can be
used by chemists who can design and synthesize chemical
probes with a variety of atom sequences with different targets.
We demonstrated a new proof-of-concept for designing
activatable NMR probes. However, the present probe has
room to be improved further, especially in terms of reactivity,
conversion yield, and stability under physiological conditions.
Further challenge is now in progress in our laboratory.
This work was supported by NEXT program and Grant-
in-Aid Number 22685018 from JSPS and performed under the
Cooperative Research Program (IMCE, Kyushu University).
Notes and references
Fig. 3 (a) 1H (left) and 1H–{13C–13C0(sp2)} (right) spectra of Cm3-13C2
(1 mM) in CDCl3. (b) Triple resonance 1H NMR spectra of
HCm3-13C2 (100 mM) incubated with (lower) or without (upper) ꢀOCl
(180 mM) in phosphate buffer (pH 7.4, 1 mM) containing 0.1% DMF.
(c) Single (1H, upper) and triple resonance (lower) 1H NMR spectra of
HCm3-13C2 (100 mM) incubated with ꢀOCl (800 mM) in the presence
of pork muscle extract. (d) Fluorescence (left, excitation at 342 nm)
and triple resonance 1H NMR (right) spectra of HCm3-13C2 (100 mM)
after incubation with MPO/H2O2/Clꢀ or H2O2/Clꢀ.
1 For a recent review on ROS/RNS imaging probes, see: T. Nagano,
J. Clin. Biochem. Nutr., 2009, 45, 111–124 and references therein.
2 For a recent review on activatable probes, see: H. Kobayashi,
M. Ogawa, R. Alford, P. L. Choyke and Y. Urano, Chem. Rev.,
2010, 110, 2620–2640.
3 (a) K. Kundu, S. F. Knight, N. Willett, S. Lee, W. R. Taylor and
N. Murthy, Angew. Chem., Int. Ed., 2009, 48, 299–303;
(b) P. Panizzi, M. Nahrendorf, M. Wildgruber, P. Waterman,
J.-L. Figueiredo, E. Aikawa, J. McCarthy, R. Weissleder and
S. A. Hilderbrand, J. Am. Chem. Soc., 2009, 131, 15739–15744;
(c) D. Oushiki, H. Kojima, T. Terai, M. Arita, K. Hanaoka,
Y. Urano and T. Nagano, J. Am. Chem. Soc., 2010, 132, 2795–2801.
4 S. Gross, S. T. Gammon, B. L. Moss, D. Rauch, J. Harding,
J. W. Heinecke, L. Ratner and D. Piwnica-Worms, Nat. Med.
(N. Y.), 2009, 15, 455–461.
The signal was confirmed as aryl 1H of the Cm3-13C2 scaffold by
2D 1H–13C HSQC analysis (Fig. S5, ESIz), indicating actual
production of Cm3-13C2 or derivatives from HCm3-13C2 as
initially designed. Since the triple resonance technique greatly
improves selectivity, ꢀOCl is sensed by HCm3-13C2 even in crude
tissue extracts (lower spectrum in Fig. 3c). On the other hand,
detection of generated Cm3-13C2 or derivatives by conventional
1H NMR was unsuccessful because of spectral overlaps with a
5 For ROS-sensitive MR probes, see: (a) J. W. Chen, W. Pham,
R. Weissleder and A. Bogdanov, Magn. Reson. Med., 2004, 52,
1021–1028; (b) H. Utsumi, K. Yamada, K. Ichikawa, K. Sakai,
Y. Kinoshita, S. Matsumoto and M. Nagai, Proc. Natl. Acad. Sci.
U. S. A., 2006, 103, 1463–1468; (c) E. Rodrıguez, M. Nilges,
´
R. Weissleder and J. W. Chen, J. Am. Chem. Soc., 2010, 132, 168–177.
6 For a recent review on dual-modal probes, see: L. E. Jennings and
N. Long, Chem. Commun., 2009, 3511–3524 and references therein.
7 Recent examples on fluorescence-MR dual-modal probes, see:
(a) S. Mizukami, R. Takikawa, F. Sugihara, M. Shirakawa and
K. Kikuchi, Angew. Chem., Int. Ed., 2009, 48, 3641–3643; (b) Y. You,
E. Tomat, K. Hwang, T. Atanasijevic, W. Nam, A. P. Jasanoff and
S. J. Lippard, Chem. Commun., 2010, 46, 4139–4141.
1
variety of H signals derived from coexisting molecules (upper
spectrum in Fig. 3c). These results clearly indicate that
HCm3-13C2 functions as an off-to-on-switching 1H MR probe
for specific detection of ꢀOCl.
Finally, we applied the HCm3-13C2 probe to dual-mode
detection of myeloperoxidase (MPO) activity. MPO is a heme-
containing enzyme present mainly in neutrophils and mediates
the production of ꢀOCl from Clꢀ and H2O2 in the killing of
bacteria. As shown in Fig. 3d, HCm3-13C2 turns on its
fluorescence and 1H signal upon incubation with MPO/
H2O2/Clꢀ in an off-to-on manner. Although the possibility
that a fluorescent Cm3 scaffold was produced by direct
oxidation with MPO or reaction with 1O2 generated from
MPO cannot be excluded at this stage, this result clearly
indicated that our dual turn-on probe is also applicable to
sensing of the enzymatic MPO system.
8 L. E. Kay, M. Ikura, R. Tschudin and A. Bax, J. Magn. Reson.,
1990, 89, 496–514.
9 (a) J. K. Gard, P. C. C. Feng and W. C. Hutton, Xenobiotica, 1997,
27, 633–644; (b) W. C. Hutton, J. J. Likos, J. K. Gard and
J. R. Garbow, J. Labelled Compd. Radiopharm., 1998, 41, 87–95;
(c) J. K. Gard, W. C. Hutton, J. A. Baker, R. K. Singh and
P. C. C. Feng, Pestic. Sci., 1999, 55, 215–218.
10 K. Mizusawa, R. Igarashi, K. Uehira, Y. Takafuji, Y. Tabata,
H. Tochio, M. Shirakawa, S. Sando and Y. Aoyama, Chem. Lett.,
2010, 39, 926–928.
11 For ꢀOCl-selective fluorescence probes, see for example:
(a) K.-I. Setsukinai, Y. Urano, K. Kakinuma, H. J. Majima and
T. Nagano, J. Biol. Chem., 2003, 278, 3170–3175; (b) J. Shepherd,
S. A. Hilderbrand, P. Waterman, J. W. Heinecke, R. Weissleder
and P. Libby, Chem. Biol., 2007, 14, 1221–1231; (c) S. Kenmoku,
Y. Urano, H. Kojima and T. Nagano, J. Am. Chem. Soc., 2007,
129, 7313–7318; (d) ref. 3b.
In conclusion, our probe is the first dual activatable
fluorescence–1H MR probe for sensing of ꢀOCl. Especially, it
should be noted that the 1H MR-based turn-on sensing was
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 1565–1567 1567