A near-infrared fluorescent calcium probe: A new tool for intracellular multicolour Ca2+ imaging

Article abstract of DOI:10.1039/c1cc14045k

We report a novel near-infrared fluorescent calcium probe (KFCA), which has good optical properties such as intense NIR fluorescence emission (670 nm, QY: 0.24), excellent ON/OFF ratio (120-fold), and good wavelength-compatibility with visible-light-emissive fluorophores (Fluo-4, DsRed2), and which is applicable for real-time dual-colour intracellular Ca²⁺ imaging.

Full text of DOI:10.1039/c1cc14045k

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 10407–10409
www.rsc.org/chemcomm
COMMUNICATION
A near-infrared fluorescent calcium probe: a new tool for intracellular
multicolour Ca2+ imagingw
Akihiro Matsui,a Keitaro Umezawa,b Yutaka Shindo,b Tomohiko Fujii,b Daniel Citterio,a
Kotaro Oka*b and Koji Suzuki*a
Received 6th July 2011, Accepted 27th July 2011
DOI: 10.1039/c1cc14045k
We report a novel near-infrared fluorescent calcium probe
(KFCA), which has good optical properties such as intense NIR
fluorescence emission (670 nm, QY: 0.24), excellent ON/OFF
ratio (120-fold), and good wavelength-compatibility with
visible-light-emissive fluorophores (Fluo-4, DsRed2), and which
is applicable for real-time dual-colour intracellular Ca2+ imaging.
Since the first establishment of fluorescent indicators for
Fig. 1 (a) Chemical structure of KFCA. (b) Normalized absorption
calcium ions (Ca2+) in the 1980’s1 that had a great impact on
(blue) and fluorescence emission (red) spectra of KFCA in the absence
physiological Ca2+ imaging, several new types of revolutionary
fluorescent calcium probes (e.g. genetically encoded Ca2+
sensors,2 two-photon Ca2+ probes3) have been developed in
order to satisfy the increasing demands and interest in calcium
imaging, as well as for the development of fluorescence
imaging technology. However, one type of fluorescent probe,
which is required for one of the cutting-edge topics in bioimaging,
has not been well established, yet. That is, a near-infrared
(NIR: 650–900 nm) fluorescent calcium probe. NIR probes
are expected to enable significant improvements in simultaneous
imaging of multiple targets in combination with typical
visible-light-emitting fluorophores for interaction studies, which
is one of the hottest recent research fields.4–6 Furthermore, they
are also applicable to in vivo fluorescence imaging of tissues,
organs, and living animals.7 Despite this high potential, there
are presently only few NIR calcium probes reported based on
cyanine8 or squaraine9 chromophores, with limited optical
characteristics such as low quantum yield (f) or poor ON/
OFF fluorescence signal contrast. To the best of our knowledge,
no application data for physiological Ca2+ imaging using these
probes have been reported to date.
(0 mM, dotted line) and presence (39 mM, solid line) of Ca2+
;
excitation: 635 nm. (c) Ca2+-concentration-dependent response curve
of KFCA in buffer solution (see the ESIw). Circle plots represent the
fluorescence intensity of KFCA at varying Ca2+ concentrations
(0–39 mM). The solid line is based on a curve fitting with a theoretically
calculated equilibrium constant assuming a 1 : 1 complex formation.
The inset pictures show the emission of KFCA in the absence
(0 mM, left) and presence (39 mM, right) of Ca2+ observed under
365 nm UV irradiation.
BODIPY-based KFL-fluorophores previously developed in
our laboratory was chosen, due to the good optical properties
such as sharp fluorescence spectra with high extinction coefficients
and high quantum yields, as well as the fine-tuning ability of the
absorption/emission spectral bands in the visible-NIR
region.10,11 On the basis of these findings, KFL-5 was selected
from the KFL library.11 Its absorption/emission spectral
peaks (650/660 nm) are almost identical to those of Cy5
(649/670 nm), which is expected to result in good compatibility
with presently available imaging instrumentation using standard
optical equipment (laser, filter, dichroic mirror) for Cy5.
For selective Ca2+ recognition and binding, the most
well-known Ca2+ chelating moiety, BAPTA (O,O0-bis(2-amino-
phenyl)ethyleneglycol-N,N,N0,N0-tetra acetic acid), was selected
owing to its excellent selectivity for Ca2+ over other ions.12
Grafting BAPTA on KFL-5 is expected to invoke a photoinduced
electron transfer (PET) from the electron-rich ion chelating
moiety to the fluorescent dye moiety.13 In the absence of
Ca2+, the fluorescence of KFL is likely quenched due to this
PET process, and emission is expected to recover upon binding
to Ca2+, because of the weakening of the electron-donating
ability of the BAPTA moiety. To enhance the water solubility,
two additional carboxyl moieties were introduced into the
KFL core via methylene chains.
In this study, a novel NIR fluorescent Ca2+ probe (KFCA,
see Fig. 1a and Scheme S1 in the ESIw) is described. This probe
is composed of two moieties, a fluorescent moiety and a Ca2+
binding moiety. As the fluorescent moiety, one of the new
a Department of Applied Chemistry, Faculty of Science and
Technology, Keio University, 3-14-1, Hiyoshi, Kohoku-ku,
Yokohama 223-8522, Japan. E-mail: suzuki@applc.keio.ac.jp;
Fax: +81 45-566-1566
b Department of Biosciences and Informatics, Faculty of Science and
Technology, Keio University, 3-14-1, Hiyoshi, Kohoku-ku,
Yokohama 223-8522, Japan. E-mail: oka@bio.keio.ac.jp
w Electronic supplementary information (ESI) available: Further
experimental details and data. See DOI: 10.1039/c1cc14045k
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 10407–10409 10407

View Online
As shown in Fig. 1b, the absorption band of KFCA is
located in the far-red to NIR region (lmax = 655 nm) and
not essentially influenced by the presence or absence of Ca2+
(0 or 39 mM Ca2+). Bright NIR emission (lmax: 670 nm, f = 0.24)
was observed with a sharp fluorescence spectral band (full width at
half-maximum height: Dl1/2 = 33 nm) in the presence of
Ca2+ (39 mM), while basically no fluorescence was measurable
(f o 0.002) in the absence of Ca2+. The ON/OFF fluores-
cence signal contrast (120-fold) is extraordinarily high compared
to those of cyanine-based (o4-fold) and squaraine-based
Ca2+ probes (B3 fold). It reaches almost the same level as
Fluo-4, which is one of the most widely used Ca2+ probes
(Table 1).
As expected from the optical properties of the KFL-5
fluorophore, the absorption/fluorescence emission wave-
lengths are almost identical to the commercially available
Cy5. The fluorescence emission intensity Ca2+-titration plot
of KFCA in buffered aqueous solution well fitted the theoretical
sigmoidal response curve calculated based on a 1 : 1 complex
formation (Fig. 1c). The calculated Kd value (500 nM, Fig. S1
in the ESIw) is almost the same as that of a typical methyl-
modified BAPTA compound like Fluo-4,12 indicating that the
binding affinity and characteristics of BAPTA are retained on
conjugation to the KFL fluorophore. Finally, KFCA has a high
molar extinction coefficient (e = 180 000 Mꢀ1 cmꢀ1). Although
BODIPY-based probes are sometimes less soluble in water
owing to their hydrophobic BODIPY core, KFCA has
relatively high water solubility (41 mM without the use of
any cosolvent).
Fig. 2 Signal crosstalk of KFCA and other fluorophores:
(a) fluorescence images of HeLa cells loaded with Fluo-4 or KFCA;
scale bar, 10 mm. (b) Fluorescence images of HeLa cells expressing
DsRed2-actin or loaded with KFCA; scale bar, 20 mm. (c) Fluorescence
images of HeLa cells loaded with both Fluo-4 and KFCA; scale bar,
20 mm. (d) Fluorescence images of HeLa cells expressing DsRed2-actin
and loaded with KFCA; scale bar, 20 mm.
Encouraged by these particular optical properties, especially
the brightness and the excellent ON/OFF ratio, the usefulness
of KFCA for biological application was investigated, focusing
on intracellular Ca2+ imaging. KFCA (1 mM) was introduced
into HeLa cells by the beads-loading method.14,15 After a short
period of incubation (15 min), KFCA was well accumulated in
the cytoplasm, and its fluorescence signal could be detected by
using a typical excitation laser and filter set for Cy5 (see the
ESIw).
In order to reveal the applicability of KFCA for dual
imaging in combination with typical visible-light-emitting
fluorescent probes, the signal crosstalk between KFCA and
other fluorescent probes or proteins (Fluo-4 and DsRed2) was
evaluated. First, HeLa cells were stained by either Fluo-4 or
KFCA, respectively, and their fluorescence was observed. As
shown in Fig. 2a, fluorescence of Fluo-4 (ex. 488 nm) was
observed only in the fluorescein channel (500–545 nm), while
no signal was captured in the Cy5 channel (655–755 nm). On
the other hand, fluorescence of KFCA (ex. 635 nm) was only
captured in the Cy5 channel, showing the possibility of
complete signal separation.
Similar, and remarkable results were also obtained in terms
of the red fluorescent protein DsRed2 (ex./em. = 558/583 nm)
in Fig. 2b. Fluorescence of the DsRed2 protein (ex. 559 nm)
expressed on the actin filament in the HeLa cells was observed
in the DsRed2 channel (575–620 nm), whereas no fluorescence
signal was captured in the Cy5 channel (655–755 nm). On the
other hand, KFCA in HeLa cells without DsRed2 protein
could be excited by 635 nm lasers, and its fluorescence was
detected only in the Cy5 channel. DsRed2 is widely used not
only as a fluorescent marker for specific organelles or proteins,
but also as a suitable resonance energy transfer (RET)-
acceptor in combination with many green or yellow
fluorescent proteins due to its long-wavelength emission
property. Hence, the fact that KFCA shows no crosstalk with
DsRed2 indicates that it can be used in combination with most
fluorescent proteins for various imaging purposes. Furthermore,
these results prove the usefulness of KFCA as an orthogonal
Table 1 Kd values and optical properties of KFCA and some reported fluorescent probes
Kd/mM
ffree
fbound
ON/OFF contrast (Fbound/Ffree
)
Reference
KFCA
0.50
0.24
3.6
0.002a
0.05
0.24
0.12
n.d.
0.14
120
o4
3
—
8
9
14
Cyanine
Squaraine
Fluo-4
n.d.b
n.d.
0.35
120
a
b
Estimated from the following calculation: ffree = fbound ꢁ (Ffree/Fbound), F denotes the fluorescence integral. No data.
c
10408 Chem. Commun., 2011, 47, 10407–10409
This journal is The Royal Society of Chemistry 2011

View Online
(corresponding to Ca2+ oscillation) were observed from
KFCA and Fluo-4 (Fig. 3c and d and Video S5 in the ESIw).
These results reveal not only that the fluorescent signal of
KFCA clearly reflects the intracellular calcium behaviour, but
also that KFCA is fully compatible for real-time multiplexed
imaging in combination with typical visible-light-emitting
probes.
For these reasons, the present results make KFCA a
promising new tool for various advanced Ca2+ imaging
applications (e.g. simultaneous real-time bioimaging of
multiple targets using numerous fluorescent sensors and
fluorescent protein sensors, in vivo calcium imaging), which
are difficult or impossible to be performed using existing
fluorescent Ca2+ probes. Additionally, KFCA as the first
example of a fluorescent probe derived from the KFL
fluorophores reveals their potential in the development of
fluorescent sensors. Therefore, the design of NIR fluorescent
probes for other target ions or for reactive oxygen species
(ROS) is possible based on this study.
Fig. 3 Real-time single- or dual-colour Ca2+ imaging: (a) pseudo-
coloured images of KFCA-loaded HeLa cells with ATP stimulation
(100 mM) at 30 s. Images were captured at 15 s, 45 s, and 120 s; scale
bar, 20 mm. (b) Time course of fluorescence change of KFCA. The
arrowhead indicates the timing of ATP addition. (c) Pseudocoloured
images of HeLa cells loaded with both Fluo-4 and KFCA with ATP
stimulation (100 mM) at 30 s. Upper and lower figures represent the
imaging data of Fluo-4 (laser: 488 nm, filter: 500–545 nm) and KFCA
(laser: 635 nm, filter: 655–755 nm), respectively. Images were captured
at 15 s, 45 s, and 120 s, respectively; scale bar, 20 mm. (d) Time course
of fluorescence change of Fluo-4 (blue) and KFCA (red). The arrow-
head indicates the timing of ATP addition.
This work was supported by a JSPS stipend to K.U. and a
Grant-in-Aid for Scientific Research (A) from the Ministry of
Education, Culture, Sports, Science and Technology, Japan
(No. 20245019). We acknowledge T. Nagai (Hokkaido Univ.)
and M. Kamiya (Univ. Tokyo) for supporting us with the
beads-loading method, and Y. Hiruta (Keio Univ.) for his help
with NMR data acquisition.
fluorescent probe to most of the presently available visible-light-
emissive probes including orange or red fluorophores. On the
basis of these results, co-staining of HeLa cells with Fluo-4 and
KFCA (Fig. 2c), and also staining of DsRed2-actin-expressed
HeLa cells with KFCA (Fig. 2d) were performed. The
fluorescent signals of each fluorophore could be distinguished
in the different channels using different sets of excitation lasers
and emission filters without any signal crosstalk.
Notes and references
1 R. Y. Tsien, Biochemistry, 1980, 19, 2396–2404.
2 A. Miyawaki, J. Llopis, R. Heim, J. M. McCaffery, J. A. Adams,
M. Ikura and R. Y. Tsien, Nature, 1997, 388, 882–887.
3 H. M. Kim, B. R. Kim, M. J. An, J. H. Hong, K. J. Lee and
B. R. Cho, Chem.–Eur. J., 2008, 14, 2075–2083.
4 H. W. Ai, K. L. Hazelwood, M. W. Davidson and R. E. Campbell,
Nat. Methods, 2008, 5, 401–403.
5 W. Tomosugi, T. Matsuda, T. Tani, T. Nemoto, I. Kotera,
K. Saito, K. Horikawa and T. Nagai, Nat. Methods, 2009, 6,
351–353.
6 Y. Niino, K. Hotta and K. Oka, PLoS One, 2009, 4, e6036.
7 R. Weissleder, Nat. Biotechnol., 2001, 19, 316–317.
8 B. Ozmen and E. U. Akkaya, Tetrahedron Lett., 2000, 41,
9185–9188.
9 E. U. Akkaya and S. Turkyilmaz, Tetrahedron Lett., 1997, 38,
4513–4516.
10 K. Umezawa, Y. Nakamura, H. Makino, D. Citterio and
K. Suzuki, J. Am. Chem. Soc., 2008, 130, 1550–1551.
11 K. Umezawa, A. Matsui, Y. Nakamura, D. Citterio and
K. Suzuki, Chem.–Eur. J., 2009, 15, 1096–1106.
12 A. Takahashi, P. Camacho, J. D. Lechleiter and B. Herman,
Physiol. Rev., 1999, 79, 1089–1125.
Next, the usefulness of KFCA for monitoring intracellular
calcium concentrations [Ca2+]i was evaluated. As shown in Fig. 3a
and b and in Video S1 in the ESIw, the addition of 100 mM
ATP (a reagent leading to an increase in [Ca2+]i) caused a
significant increase in the fluorescence emission (F/F0, up to
250%) with time-related changes (including oscillation).16
Stimulation by 5 mM ionomycin (a reagent causing a cytosolic
Ca2+ increase) also resulted in increasing fluorescence
emission (Fig. S2 and Video S2 in the ESIw). These fluorescence
emission intensity changes observed with KFCA well agreed
with the data obtained by using Fluo-4 (Fig. S3 and S4 and
Videos S3 and S4 in the ESIw), emphasising that the emission
intensity changes of KFCA correspond to the changes of the
Ca2+ concentrations. According to the literature, the ON/OFF
ratio (dynamic range) of typical Ca2+ indicators (e.g. Fluo-4,
Calcium Green, Calcium Orange, Oregon Green 488 BAPTA-1)
in cells tends to be a few times (sometimes over one digit) lower
than those in vitro.17 For this reason, the observed F/F0 value is
considered to be consistent.
13 A. P. deSilva, H. Q. N. Gunaratne, T. Gunnlaugsson, A. J.
M. Huxley, C. P. McCoy, J. T. Rademacher and T. E. Rice, Chem.
Rev., 1997, 97, 1515–1566.
14 M. Kamiya and K. Johnsson, Anal. Chem., 2010, 82, 6472–6479.
15 P. L. McNeil and E. Warder, J. Cell Sci., 1987, 88, 669–678.
16 M. J. Berridge, M. D. Bootman and P. Lipp, Nature, 1998, 395,
645–648.
Finally, dual staining of HeLa cells by both KFCA and
Fluo-4 and dual-colour real-time [Ca2+]i monitoring were
performed. After stimulation by ATP (100 mM), completely
identical fluorescence emission patterns with oscillation
17 D. Thomas, S. C. Tovey, T. J. Collins, M. D. Bootman,
M. J. Berridge and P. Lipp, Cell Calcium, 2000, 28, 213–223.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 10407–10409 10409