Two-photon fluorescent probes for acidic vesicles in live cells and tissue

Article abstract of DOI:10.1002/anie.200704586

(Figure Presented) Move with the groove: Two-photon fluorescent pH probes and a two-photon lysotracker (AL1) can visualize acidic vesicles in live cells and living tissue for a long period of time without mistargeting and photobleaching problems. Using AL1, vesicles can be tracked in real time (see picture).

Full text of DOI:10.1002/anie.200704586

Angewandte
Chemie
DOI: 10.1002/anie.200704586
Fluorescent Probes
Two-Photon Fluorescent Probes for Acidic Vesicles in Live Cells and
Tissue**
Hwan Myung Kim, Myoung Jin An, Jin Hee Hong, Byeong Ha Jeong, Ohyun Kwon, Ju-
Yong Hyon, Seok-Cheol Hong, Kyoung J. Lee, and Bong Rae Cho*
We report two-photon (TP) pH probes (AH1, AH2) and a TP
lysotracker (AL1, Scheme 1) derived from 2-acetyl-6-(dime-
thylamino)naphthalene (acedan). These probes can visualize
the acidic vesicles in live cells and tissues. Lysosomes and
lysosome-related organelles constitute a system of acidic
compartments (pH 4.0–5.0) that contain a large number of
of advantages over OPM, including increased penetration
depth (greater than 500 mm), localized excitation, and pro-
longed observation time.[7] The extra penetration depth that
TPM affords is of particular interest in tissue imaging,
because surface preparation artifacts such as damaged cells
extend over 70 mm into the tissue interior. However, most of
the OP fluorescent probes presently used for TPM have small
TP action cross sections (Fd) that limit their usage.[8]
Although a TP pH probe with appreciable Fd (ca. 42 GM)
has been reported,[4a] the utility of this probe in TPM imaging
has not been verified. Therefore, there is a need to develop an
efficient TP probe that can visualize acidic vesicles deep
inside tissue for a long period of time.
Scheme 1. The structures of AH1, AH2, and AL1.
To design an efficient TP probe for acidic vesicles, we
chose acedan as the TP fluorophore, because acedan-derived
TP probes for Mg2+ (AMg1)[9] and Ca2+ (ACa1)[10] exhibited
high photostability as well as significant TP action cross
sections for the bright TPM image at low probe concentra-
tion, thus allowing the detection of the metal ions deep inside
live tissues for over 1100 s. We have introduced an aniline,
o-methoxy aniline (pKa(BH+) ꢀ 4), or tertiary amine
(pKa(BH+) ꢀ 10) substituent as the proton-binding site
through an amide linkage to the fluorophore. It is expected
that AH1 and AH2 would emit TP-excited fluorescence
(TPEF) upon protonation at pH < 4, whereas AL1 would
emit TPEF in the acidic vesicles, where it should accumulate
as the protonated form.[5] Herein, we report that these probes
are capable of imaging the acidic vesicles in live cells and
living tissues at greater than 100 mm depth without mistarget-
ing[9–11] and photobleaching problems. Moreover, AL1 can
visualize the transportation of the acidic vesicles in the
hippocampal cornu ammonis CA3 region for a long period of
time with the use of TPM.
enzymes and secretory proteins exhibiting a variety of
functions.[1,2] To determine their functions, a variety of
membrane-permeable fluorescent pH and lysosomal probes
have been developed, some of which are commercially
available.[3–5] However, use of these probes with one-photon
microscopy (OPM) requires excitation with short-wavelength
light (ca. 350–550 nm) that limits their application in deep-
tissue imaging, owing to the shallow penetration depth (less
than 80 mm) as well as to photobleaching, photodamage, and
cellular autofluorescence.[6] To overcome these problems, it is
crucial to use two-photon microscopy (TPM). TPM employs
two near-infrared photons for excitation and offers a number
[*] H. M. Kim, M. J. An, Prof. Dr. B. R. Cho
Department of Chemistry, Korea University
1-Anamdong, Seoul, 136-701 (Korea)
Fax: (+82)2-3290-3544
E-mail: chobr@korea.ac.kr
AH1, AH2, and AL1 were prepared in 47–77% yields by
reactions of 6-acyl-2-[N-methyl-N-(carboxymethyl)amino]-
naphthalene and a p-phenylenediamine derivative or N,N-
dimethylethylenediamine (see the Supporting Information).
The solubilities of AH1, AH2, and AL1 in water are in the
range of 5.0–9.0 mm, which are sufficient to stain the cells
(Figure S2 in the Supporting Information). The fluorescence
spectra of AH1, AH2, and AL1 show gradual bathochromic
shifts with solvent polarity (ENT) in the order 1,4-dioxane <
DMF < EtOH < H2O (Figure S1 and Table S1 in the Support-
ing Information). The large bathochromic shifts with increas-
ing solvent polarity indicate the utility of these molecules as
polarity probes.
Dr. J. H. Hong, B. H. Jeong, Prof. Dr. S.-C. Hong, Prof. Dr. K. J. Lee
Department of Physics, Korea University (Korea)
Dr. J. H. Hong, B. H. Jeong, Prof. Dr. K. J. Lee
National Creative Research Initiative Center for Neurodynamics
Korea University (Korea)
J.-Y. Hyon
Biomicrosystems Technology Program, Korea University (Korea)
Dr. O. Kwon
Samsung Advanced Institute of Technology (Korea)
[**] This work was supported by a KOSEF grant funded by the Korea
Ministry of Science and Technology (MOST) (No. R0A-2007-000-
20027-0). J.H.H., B.H.J., and K.J.L. were supported by Creative
Research Initiatives of the MOST. S.C.H. and J.-Y.H. were supported
by the Seoul R&D program.
TP action cross section was determined by investigating
the TPEF of the probes using fluorescein as the reference (see
the Supporting Information).[8] The TP action spectra of AH1,
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2008, 47, 2231 –2234
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2231

Communications
Table 1: Photophysical data for AH1, AH2, AL1, and LTR.
AH2, and AL1 at pH 3.2 indicated a Fd value of approx-
fl [b]
max
[f]
Compound Buffer[a] lmð1aÞx/l
F[c] FEF[d] pKOa P
/
Fdmax
imately 90 GM at 780 nm, ninefold larger than that of
LysoTracker Red (LTR; Figure 1a and Table 1). Thus, TPM
images of the samples stained with these probes would be
much brighter than those stained with LTR.
When the pH value of the solutions containing AH1 or
AH2 was gradually decreased in universal buffer solution
(0.1m citric acid, 0.1m KH2PO4, 0.1m Na2B4O7, 0.1m Tris, 0.1m
KCl), the fluorescence intensity increased dramatically with-
out changing the absorption spectra, probably as a result of
protonation blocking the photoinduced electron-transfer
[nm]
pKTaP[e]
pH 3.2 364/498
pH 7.0 365/498
pH 3.2 365/496
pH 7.0 365/496
pH 3.2 364/496
pH 7.0 364/496
pH 3.2 575/593
0.60 22
0.03 (26)
0.64 64
0.01 (68)
0.72
86
AH1
AH2
4.4/4.5
4.5/4.5
nd[g]
88
nd[g]
92
AL1
LTR
NA[h] NA[h]
0.76
0.08 NA
93
10
[h]
NA[h]
[a] All data were measured in universal buffer (0.1m citric acid, 0.1m
KH2PO4, 0.1m Na2B4O7, 0.1m tris(hydroxymethyl)aminomethane (Tris),
0.1m KCl). [b] lmax of the one-photon absorption and emission spectra.
[c] Fluorescence quantum yield ꢂ10%. [d] Fluorescence enhancement
factor. The numbers in parentheses are FEF values measured by two-
photon processes. [e] pKa values measured by one- (pKOa P) and two-
photon (pKTaP) processes. [f] Two-photon action cross section. [g] nd=
not determined. The two-photon excited fluorescence intensity was too
weak to measure the cross section accurately. [h] Not available.
(PET) process (Figure 1b,c and Figure S3 in the Supporting
Information).[3a] The fluorescence enhancement factors
ꢁ1
(FꢁFmin)Fmin of AH1/AH2 determined by one- and two-
photon processes were 22/64 and 26/68, respectively (Table 1).
There is a threefold decrease in the fluorescence quantum
yield from AH1 (F = 0.03) to AH2 (F = 0.01) at pH 7.0
owing to the o-OMe group. Calculations at the B3LYP/6-
31G** level revealed that the energies of the highest occupied
molecular orbitals (HOMOs) of 4-acetamidoaniline (R1), 4-
acetamido-2-methoxyaniline (R2), and acedan are ꢁ5.044,
ꢁ4.827, and ꢁ5.164 eV, respectively (Figure S6 in the
Supporting Information).[12] Hence, PET from R2 to acedan
should occur more efficiently because it is more exothermic.
This would predict a smaller fluorescence quantum yield for
AH2, as observed. Moreover, the pKa values of the conjugate
acids of AH1 and AH2 estimated from the fluorescence
titration curves are 4.42 ꢂ 0.03 and 4.48 ꢂ 0.02, respectively,
indicating that they are suitable for detecting acidic vesicles
with pH < 4 (Figure 1c and Table 1). On the other hand, there
was little change in the fluorescence spectra of AL1 with the
pH change (Figure S5 in the Supporting Information).
When the macrophages labeled with AH2 and AL1 were
repeatedly washed with phenol-red-free Dulbeccoꢀs modified
Eagleꢀs medium (DMEM), the TPEF intensity decreased
gradually, and the change was significantly smaller with AL1
(Figure S8 in the Supporting Information). This result indi-
cates that AH2 is distributed anywhere within the membrane,
whereas AL1 exists mostly in the acidic vesicles as AL1H+,
which does not generally leak out. Thus, AH2 acts as a TP pH
probe and AL1 as a TP lysotracker.
To demonstrate the utility of these probes, TPM images
were obtained of individual macrophages labeled with 2 mm
AH1, AH2, and AL1 (Figure 2). It was previously reported
that the TPM images of AMg1- and ACa1-labeled cells
emitted TPEF in the 500–620 and 360–460 nm regions, which
had been attributed to probes associated with cytosol and
membrane, respectively.[9,10] In contrast, the TPM images of
AH2- and AL1-labeled macrophages emitted TPEF only at
500–620 nm (Figure 2b,f) and not at 360–460 nm (Fig-
~
&
*
Figure 1. a) Two-photon action spectra of AH1 ( ), AH2 ( ), AL1 ( ),
!
and LTR ( ) in universal buffer (pH 3.2). b) Two-photon emission
spectra of AH2 as a function of pH (3.2–9.5) in universal buffer.
!
!
&
c) One- ( , ) and two-photon (&, ) fluorescence titration curves of
!
!
&
AH1 (
,
) and AH2 ( , &) as a function of pH (3.2–8.0). The
fluorescence intensity was integrated over the entire wavelength range.
The wavelengths for one- and two-photon excitation were 365 and
780 nm, respectively.
2232
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 2231 –2234

Angewandte
Chemie
axon terminal along the axon
(Figure 3c,d and movie S1 in
the Supporting Informa-
tion).[13,14] Furthermore, the
activity of the acidic vesicles
could be visualized for more
than 1100 s without appreci-
able decay (Figure S14 in the
Supporting
Information).
This finding underlines the
high photostability of this
probe in addition to its capa-
bility for deep-tissue imag-
ing. Finally, we note that
AL1 is superior to AH2 in
monitoring vesicle transpor-
tation, although both are
capable of detecting acidic
vesicles deep inside live tis-
sues.
Figure 2. Pseudo-colored TPM images of a,b) AH2- and e,f) AL1-labeled (2 mm) macrophages collected at
360–460 nm (a, e), and 500–620 nm (b, f). c,g) OPM images of LTR-labeled macrophages and d,h) co-
localized images. The wavelengths for one- and two-photon excitation were 543 and 780 nm, respectively.
Cells shown are representative images from replicate experiments (n=5). Scale bars in (c) and (g): 30 mm.
ure 2a,e). Moreover, the TPEF spectra of the intense and
bright spots in the TPM images of the AH2-labeled cells are
fl
max
almost the same as that of AH2 in the buffer solution with l
ꢀ 500 nm (Figure S10 in the Supporting Information). A
nearly identical result was observed in the TPM images of
AH2- and AL1-labeled tissues (Figure S13 in the Supporting
Information). These results suggest that the probes are
predominantly located in the hydrophilic regions in the cell,
presumably because of their low molecular weight. Thus, the
acidic vesicles can be detected without interference from the
membrane-bound probes.
To unambiguously determine whether the intense (red)
spots in Figure 2b are indeed the acidic vesicles, the macro-
phages were co-stained with AH2 and LTR, a well-known
one-photon fluorescent probe for acidic vesicles,[5] and the
TPM image was co-localized with the OPM image. The two
images corresponded well (Figure 2d), thus confirming that
Figure 3. Images of an acute rat hippocampal slice stained with 10 mm
AL1. a) Bright-field image showing the CA1 and CA3 regions as well as
the dentate gyrus (DG) upon 10” magnification. White dotted lines
indicate the pyramidal neuron layer. b) 40 TPM images accumulated
along the z direction at a depth of approximately 100–250 mm with
10” magnification to visualize the average distribution of the acidic
vesicles in the same regions. c) Image at 100” magnification in CA3
regions at a depth of approximately 120 mm. d) Enlargement of a red
box in (c) showing the transport of acidic vesicles along the axon.
Scale bars: 300 (b) and 30 mm (c).
the bright regions reflect acidic vesicles. Similar results were
obtained in the co-localization experiments of AL1 (Fig-
ure 2h) and AH1 (Figure S9d in the Supporting Information)
with LTR. Hence, these probes are clearly capable of imaging
the acidic vesicles in live cells by TPM.
We further investigated the utility of AL1 and AH2 in
tissue imaging. The bright-field image of a part of an acute rat
hippocampal slice from a 2-day postnatal rat incubated with
10 mm AL1 or AH2 for 30 min at 378C reveals the CA1 and
CA3 regions as well as the dentate gyrus (DG, Figure 3a). The
TPM images revealed the acidic vesicles in the same region at
100–250 mm depth (Figure S11 in the Supporting
Information), but each image represents the distribution
exclusively in the given plane. Since the structure of the brain
tissue is known to be inhomogeneous in its entire depth, we
have accumulated 40 TPM images at different depths in the
range of 100–250 mm to visualize the distribution of acidic
vesicles. The accumulated TPM images show that the acidic
vesicles are more abundant in CA3 and DG than in the CA1
region (Figure 3b and Figure S12 in the Supporting
Information). Moreover, the real-time images reveal rapid
transportation of the acidic vesicles between cell body and
To conclude, we have developed TP pH probes (AH1,
AH2) and a TP lysotracker (AL1) that can be excited by 780-
nm femtosecond laser pulses, easily loaded into the cell and
tissue, and which can visualize the acidic vesicles in live cell
and tissue by TPM without the artifacts of photobleaching
and mistargeting problems. AH1 and AH2 show pKa(BH+)
values of approximately 4.5 and TP fluorescence enhance-
ment factors of 26–68 upon binding with a proton and are
suitable for detecting acidic vesicles with pH < 4 at approx-
imately 100–250 mm depth in live tissue. Moreover, AL1 can
visualize the transportation of acidic vesicles along the axon
Angew. Chem. Int. Ed. 2008, 47, 2231 –2234
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2233

Communications
[5] A Guide to Fluorescent Probes and Labeling Technologies, 10th
ed. (Ed.: R. P. Haugland), Molecular Probes, Eugene, OR, 2005.
[6] a) D. Thomas, S. C. Tovey, T. J. Collins, M. D. Bootman, M. J.
Berridge, P. Lipp, Cell Calcium 2000, 28, 213 – 223; b) E.
Grapengiesser, Cell Struct. Funct. 1993, 18, 3 – 17.
[7] a) W. R. Zipfel, R. M. Williams, W. W. Webb, Nat. Biotechnol.
2003, 21, 1369 – 1377; b) F. Helmchen, W. Denk, Nat. Methods
2005, 2, 932 – 940.
[8] C. Xu, W. W. Webb, J. Opt. Soc. Am. B 1996, 13, 481 – 491.
[9] H. M. Kim, C. Jung, B. R. Kim, S.-Y. Jung, J. H. Hong, Y.-G. Ko,
K. J. Lee, B. R. Cho, Angew. Chem. 2007, 119, 3530 – 3533;
Angew. Chem. Int. Ed. 2007, 46, 3460 – 3463.
at approximately 120 mm depth for more than 1100 s in live
tissue by TPM.
Received: October 4, 2007
Revised: December 27, 2007
Published online: February 13, 2008
Keywords: acidity · fluorescent probes · live tissue ·
.
two-photon microscopy · vesicles
[1] S. Kornfeld, I. Mellman, Annu. Rev. Cell Biol. 1989, 5, 483 – 525.
[2] a) E. J. Blott, G. M. Griffiths, Nat. Rev. Mol. Cell Biol. 2002, 3,
122 – 131; b) J. Stinchcombe, G. Bossi, G. M. Griffiths, Science
2004, 305, 55 – 59.
[3] a) A. P. de Silva, H. Q. N. Gunaratne, T. Gunnlaugsson, A. J. M.
Huxley, C. P. McCoy, J. T. Rademacher, T. E. Rice, Chem. Rev.
1997, 97, 1515 – 1566; b) G. Greiner, I. Maier, J. Chem. Soc.
Perkin Trans. 2 2002, 1005 – 1011; c) D. Cui, X. Qian, F. Liu, R.
Zhang, Org. Lett. 2004, 6, 2757 – 2760.
[4] a) S. Charier, O. Ruel, J.-B. Baudin, D. Alcor, J.-F. Allemand, A.
Meglio, L. Jullien, Angew. Chem. 2004, 116, 4889 – 4892; Angew.
Chem. Int. Ed. 2004, 43, 4785 – 4788; b) F. Galindo, M. I.
Burguete, L. Vigara, S. V. Luis, N. Kabir, J Gavrilovic, D. A.
Russell, Angew. Chem. 2005, 117, 6662 – 6666; Angew. Chem. Int.
Ed. 2005, 44, 6504 – 6508; c) K. M. Sun, C. K. McLaughlin, D. R.
Lantero, R. A. Manderville, J. Am. Chem. Soc. 2007, 129, 1894 –
1895.
[10] H. M. Kim, B. R. Kim, J. H. Hong, J.-S. Park, K. J. Lee, B. R.
Cho, Angew. Chem. 2007, 119, 7589 – 7592; Angew. Chem. Int.
Ed. 2007, 46, 7445 – 7448.
[11] C. L. Slayman, V. V. Moussatos, W. W. Webb, J. Exp. Biol. 1994,
196, 419 – 438.
[12] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648 – 5692; b) M. J.
Frisch et al., Gaussian98 (RevisionA.7), Gaussian, Inc., Pitts-
burgh, PA, 1998.
[13] a) K. N. Hartman, S. K. Pal, J. Burrone, V. N. Murthy, Nat.
Neurosci. 2006, 9, 642 – 649; b) J. Burrone, Z. Li, V. N. Murthy,
Nat. Protoc. 2006, 1, 2970 – 2978.
[14] a) A. L. Scotti, H. Reuter, Proc. Natl. Acad. Sci. USA 2001, 98,
3489 – 3494; b) T. C. Sꢁdhof, Annu. Rev. Neurosci. 2004, 27, 509 –
547.
2234
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 2231 –2234