Synthetic macrocyclic peptidomimetics as tunable pH probes for the fluorescence imaging of acidic organelles in live cells

Article abstract of DOI:10.1002/anie.200501920

(Chemical Equation Presented) Acid test: Macrocyclic compounds that comprise a 9,10-anthracene subunit linked by a C₂-symmetric peptidomimetic chain (see formula) are useful as fluorescent probes in a pH range of biomedical interest. Changes in

Full text of DOI:10.1002/anie.200501920

Communications
Fluorescent Probes
DOI: 10.1002/anie.200501920
Synthetic Macrocyclic Peptidomimetics as
Tunable pH Probes for the Fluorescence Imaging
of Acidic Organelles in Live Cells**
Francisco Galindo,* M. Isabel Burguete, Laura Vigara,
Santiago V. Luis,* Nurul Kabir, Jelena Gavrilovic, and
David A. Russell*
Synthetic fluorescent probes have found widespread applica-
tions in cell biology for the intracellular measurement of
several species, from the zinc(ii) cation[1] and citrate anion[2] to
singlet oxygen[3] and nitric oxide.[4] In recent years, several
reviews have appeared that deal with synthetic fluorescent
chemosensors.[5,6] Protons are one of the most important
targets among the intracellular species of interest, as it is well-
known that pH plays a central role in many cellular events.[7]
In general terms, two broad ranges of pH values are found in
cells. Consequently, two types of synthetic probes have been
developed,[8] namely, probes for cytosol that work at a pH of
about 6.8–7.4, and probes for the so-called acidic organelles
(for example, lysosomes) that function over the pH range of
about 4.5–6.0. It is possible to find a large array of commercial
probes for the former pH range; however, this is not the case
for acidic organelles. Many reports have appeared that
connect some cellular dysfunction with abnormal pH values
in acidic organelles. For instance, defective pH regulation of
acidic compartments has been observed in human breast
cancer cells.[9] Another illustrative example is the finding of
elevated lysosomal pH in neural ceroid lipofuscinoses, a
common inherited neurodegenerative disorder that affects
[*] Dr. F. Galindo, Dr. M. I. Burguete, L. Vigara, Prof. Dr. S. V. Luis
Departamento de Química Inorgµnica y Orgµnica
Universitat Jaume I
Avda. Sos Baynat, s/n, 12071 Castellón (Spain)
Fax: (+34)964-72-8214
E-mail: francisco.galindo@qio.uji.es
luiss@qio.uji.es
Dr. N. Kabir, Prof. Dr. D. A. Russell
School of Chemical Sciences and Pharmacy
University of East Anglia
Norwich, Norfolk NR47TJ (UK)
Fax: (+44)1603-593-012
E-mail: d.russell@uea.ac.uk
Dr. N. Kabir, Prof. Dr. J. Gavrilovic
School of Biological Sciences
University of East Anglia
Norwich, Norfolk NR47TJ (UK)
[**] This work was supported by the Spanish Ministerio de Ciencia y
Tecnología (Project BQU-2003-0915-C03-02), Fundació Caixa Cas-
telló-Bancaixa and Universitat Jaume I (Project 04I007.25/1 to F.G.),
and The Wellcome Trust (Showcase award to D.A.R. and J.G.). F.G.
thanks the Spanish MEC for financial support (Ramón y Cajal
Program).
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
6504
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6504 –6508

Angewandte
Chemie
children.[10] It would, therefore, be extremely desirable to
have access to a broad variety of “acidic” probes to study
these and other conditions. Unfortunately, as a consequence
of the current unavailability of probes for the acidic pH range,
there is a research bottleneck as regards progress in some
areas of cell biology or medicine. Additionally, in the light of
the above discussion, a further fundamental criterion emerges
for the design of new probes. This criterion is the possibility of
chemical control of the pH reporter probe under extracellular
conditions, that is, the potential for precise control (and
prediction) of the pKa value of the probe by adjusting the
appropriate elements in the molecular architecture.
cycles in moderate to high yields without the need for high-
dilution conditions. In the case of the FG-H series shown in
Scheme 1, two units of l-valine were selected because of their
reported ability to induce the formation of b sheets.[14]
However, it should be possible to introduce virtually any a-
amino acid with an appropriate protecting group into the
general synthetic procedure (Scheme 2) to obtain a new
Herein, we present a new family of fluorescent macro-
cyclic probes (FG-H series)[11] that consist of a 9,10-anthra-
cene subunit linked by a C2-symmetric peptidomimetic chain
(Scheme 1). The probes differ only in the size of the chain
Scheme 2. a) Acid activation; b) coupling with 1,n-diaminoalkane;
c) deprotection; d) ring-closing with 9,10-bis(bromomethyl)anthracene.
Cbz=carbobenzyloxy.
family of peptidomimetics with different properties as pH
probes. The 9,10-dimethylanthracene moiety was chosen as a
fluorophore as it is known to be photostable and also to allow
comparison with the widely used pH probe for acidic
organelles, LysoSensor Blue DND-167 (Scheme 1). DND-
167, which was originally studied by de Silva et al.,[15] is also
an anthracene derivative that shares with the compounds in
the FG-H series the presence of two amine groups decoupled
electronically from the fluorophore by both methylene
spacers. However, in the case of DND-167, there is the
possibility that more flexible configurations could be adopted.
For comparison a second commercial probe, LysoSensor
Green DND-189, which has a different photochemical nature,
was also selected. LysoTracker Green DND-26 (see
Scheme 1)[16] was also used for fluorescence imaging experi-
ments in vivo. This latter compound was employed to certify
the acidic nature of the organelles, as it is known to
accumulate in lysosomes. (Note, however, that LysoTracker
Green DND-26 is not a pH “reporter”, as its fluorescence
Scheme 1. Structures of the fluorescent probes studied.
emission does not vary with changing pH.[8a]
)
Standard fluorescence pH titrations were performed for
all the compounds in Scheme 1 in buffered aqueous media
(containing 0.2% DMSO from the initial stock solution) at a
probe concentration of 2 mm, with measurement at pH
intervals narrow enough to allow a precise determination of
the apparent pKa value. Figure 1 shows the absorption
spectrum of FG-H503 (Figure 1a) along with the fluorescence
emission spectra at different pH values (Figure 1b–g). The
fluorescence changes in systems composed of fluorophore–
CH2–amine moieties, such as the FG-H series, can be
rationalized according to the photoinduced electron-transfer
(PET) scheme developed by the research groups of de Silva
and Czarnik:[6] Under basic conditions the amino groups are
not protonated, and hence the PET process takes place (if
thermodynamically allowed by the Rehm–Weller formal-
ism[17]) which leads to fluorescence quenching. In acidic media
the electron pair in each amine is protonated, thus preventing
(n = 3,4,6,8), but this structural variation produces a smooth
shift in the pKa values, which is an essential feature that
enables a range of pH “windows”. The synthesis of the FG-H
family members is facile, and the compounds exhibit good
photostability and present excellent cellular uptake without
the need for derivatization to improve crossing of the cellular
membrane. Moreover, the series extends the range of pH
utility by about 0.5 pH units (toward a lower pH) over that of
currently available probes.
Compounds FG-H503, FG-H504, FG-H506, and FG-
H508 were synthesized according to a procedure developed
for the preparation of peptidomimetic macrocycles in high
yields.[12,13] It has been previously demonstrated that the b-
turn conformation adopted by the synthetic intermediates
(arising from the presence of two a-amino acids as building
blocks) allows the synthesis of the peptidomimetic macro-
Angew. Chem. Int. Ed. 2005, 44, 6504 –6508
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6505

Communications
Table 1: Spectral characteristics and pKa values of the studied systems.
Entry
Compound[a]
lex [nm]
lem [nm]
pKa
1
2
3
4
5
6
7
8
9
FG-H503
FG-H504
FG-H506
FG-H508
DND-167
DND-189
377
377
377
377
374
443
377
374
443
377
374
443
430
430
430
430
427
505
430
427
505
430
427
505
5.06Æ0.02
5.24Æ0.02
5.41Æ0.01
5.43Æ0.01
5.61Æ0.01
6.06Æ0.06
5.02Æ0.02
5.57Æ0.01
6.01Æ0.03
4.95Æ0.03
5.52Æ0.01
5.70Æ0.03
FG-H503+CuII
DND-167+CuII
DND-189+CuII
FG-H503+ZnII
DND-167+ZnII
DND-189+ZnII
10
11
12
Figure 1. FG-H503 (2 mm) in aqueous solution (0.2% DMSO). a) Nor-
malized absorption spectrum at pH 1.02, and fluorescence spectra
(lex =377 nm) at b) pH 2.55, c) 4.70, d) 5.10, e) 5.41, f) 5.83, and
g) 6.90.
[a] Probe concentration: 2 mm in aqueous solution (0.2% DMSO); metal
concentration: 200 mm.
PET from occurring, and hence emission from the fluoro-
phore takes place.
In the case of the FG-H series, a minimal variation in the
chemical structure of the probe causes an appreciable change
in the titration curve (see Figure 2 for some of the titration
Figure 3. Schematic representation of the pH intervals where the fluo-
rescence emission of each probe experiences a change between 20
and 80% of its maximum intensity. Probe concentration: 2 mm in aque-
ous solution (0.2% DMSO); excitation and emission wavelengths as
indicated in Table 1.
DND-167, and at pH 4 for DND-189). Thus, it can be seen
that the peptidomimetic macrocycles are complementary to
the DND probes, as they extend the pH range of measure-
ment toward the acidic side.
Figure 2. Fluorescence versus pH titration curves for the probes under
study. Data are normalized at the maximum emission (Irel). Probe con-
centration: 2 mm in aqueous solution (0.2% DMSO); excitation wave-
On account of the complexity of the intracellular environ-
ment, an additional examination of the probes was performed
to determine whether other ions were potential interferents.
For example, it is well-known that amines can bind many
metal cations in solution.[6] In fact, many chemosensors for
metal ions use the ability of amines to coordinate the vacant
orbitals of such analytes. Among them, zinc(ii) and copper(ii)
are the types of species likely to be targeted by free amines in
solution. Consequently, control experiments with FG-H503,
DND-167, and DND-189 in the presence of an excess of ZnII
and CuII ions were performed to assess such possible
coordination. As can be seen in Table 1, the effect of such
metals on the peptidomimetic macrocycle is negligible,
particularly when considering that the concentrations used
for the experiment were significantly higher than those
present in the intracellular environment.
~
~
lengths as indicated in Table 1. FG-H503 ( ), FG-H504 ( ), FG-H508
&
*
( ), DND-167 (&), DND-189 ( ).
curves; complete spectra and titration results can be found in
the Supporting Information).[18,19] The pKa values are derived
from the nonlinear curve-fitting of the data and are shown in
Table 1 along with the excitation and emission wavelengths.
It can be seen in Figure 2 that the DND-189 probe
exhibits a fluorescence intensity versus pH profile that is not
the same as the other profiles, which reflects its different
chemical nature with respect to the anthracene framework.
Additionally, a sharp decrease in the fluorescence intensity is
measured with DND-189 at acidic pH values.
To examine the useful pH range for each probe, the pH
interval for which the tested compound changes from 20 to
80% of its maximum fluorescence was determined (Figure 3;
the maximum is measured at pH 2 for the FG-H series and
FG-H503 was investigated as a fluorescent probe in
cultured cells and was compared to DND-189. The probe
6506 www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6504 –6508

Angewandte
Chemie
DND-167 did not produce images with sufficient resolution;
therefore, the additional probe LysoTracker Green DND-
26[16] was also used which is known to be retained specifically
in acidic organelles. Two series of experiments were per-
formed: Raw 264.7, a mouse macrophage cell line, was loaded
with FG-H503 and with either DND-189 or DND-26. As
shown in Figure 4b,g, FG-H503 exhibited a vesicular distri-
Experimental Section
Synthesis of the peptidomimetic macrocycles: The appropriate 1,n-
diaminoalkane was reacted with N-protected (Cbz) l-valine through
the formation of an N-hydroxysuccinimide ester.[12] Yields of these
intermediate peptidomimetic compounds were about 70%. The
cyclization step was performed by treating 9,10-bis-bromomethylan-
thracene with the diamine formed by N-deprotection of the
intermediates
with
HBr/
AcOH (deprotection yields
over 90%). The products of
the cyclization step (at refluxin
acetonitrile as the solvent, and
anhydrous K2CO3 as the base)
were purified by column chro-
matography to afford the final
products in about 30% yield
(see Supporting Information
for chemical characterization).
Fluorescence
measure-
ments: The probes were dis-
solved in dimethyl sulfoxide
(DMSO) to obtain 1 mm stock
solutions, and then aliquots
were diluted to 2 mm with
water containing a mixture of
several buffers to facilitate
titrations between pH 9 and 2
(40 mm of each sodium salt:
acetate, phosphate, borate, and
carbonate). All the measure-
ments were performed in the
presence of 100 mm NaCl to
Figure 4. Distribution and co-localization of FG-H503 with lysosomal probes DND-189 (a–e) and DND-26
(f–j) in Raw 264.7 cells. Differential interference contrast (DIC; a,f) and fluorescence images of FG-
H503 (b,g) and DND-189 (c) or DND-26 (h) were collected with a confocal laser scanning microscope in the
DIC, DAPI, and FITC modes. d),i) Merged images of red and green channels; e),j) composite images of red,
green, and DIC channels. DND-189 (a–e) and DND-26 (f–j) are from a different set of experiments.
maintain a constant ionic strength. Slight variations in the pH of the
solutions were achieved by adding the minimum volumes (typically
10 mL added to 10 mL) of 0.1–1.0m NaOH or 0.1–1.0m HCl, in such a
way that dilution effects were negligible. Fluorescence spectra were
recorded (Spex Fluorolog Max-2) with excitation at the wavelengths
indicated in Table 1. Plots of the fluorescence intensity versus pH
were fitted by using a nonlinear curve-fitting technique (Microcal
Origin 6.0). All the measurements were carried out at ambient
temperature and in air-equilibrated solutions. Copper(ii) and zinc(ii)
were added as the chloride and nitrate salts, respectively (200 mm).
Confocal fluorescence imaging: Raw 264.7, a mouse macrophage
cell line, was cultured on 42-mm-diameter round glass coverslips in
Dulbeccoꢀs modified Eagleꢀs medium (MEM) containing fetal bovine
serum (10%), l-glutamine (2 mgmLÀ1), and antibiotics (penicillin/
streptomycin). The cells were stimulated with interferon-g (IFN-g)
and lipopolysaccharide (LPS) for 16 h, loaded with the dyes for
15 min in serum-free media, and then mounted on a stage at 378C
after changing to phenol-red-free L15 medium containing the above
supplements. The cells were observed with a confocal laser scanning
microscope (Carl Zeiss, LSM 510 meta) with 350- and 488-nm laser
excitation for the 4,6-diamidino-2-phenylindole (DAPI) and fluores-
cein isothiocyanate (FITC) channels, respectively. An appropriate
emission band was selected for the two fluorescent channels. A 63 ” ,
1.4-NA objective was used to ensure high-resolution images. DIC
images were collected simultaneously with transmitted light by
excitation at 488 nm. Sequential rather than simultaneous acquisition
was used to avoid bleed-through between the two fluorescent
channels. Images were processed, and double- and composite-
merged images were made with Adobe Photoshop.
bution. The probe predominantly co-localized with both
DND-189 and DND-26, which indicates that the probe is
taken up by the cells and that it localizes within acidic
organelles. Note that the different excitation and emission
wavelengths of FG-H503, as compared to those of the
commercial probes, allow simultaneous visualization of both
probes (FG-H503 and either DND-189 or DND-26) from the
same intracellular compartment. Interestingly, the fluores-
cence intensity of FG-H503 appears to match that of DND-
189: areas of low and high fluorescence (light and dark blue
arrows, respectively, in Figure 4b–d) of the probe match those
of DND-189, which indicates that the FG-H probe can
distinguish between different pH values in the cell in a
manner similar to the commercial probe. Analogous results
are obtained from the joint use of FG-H503 and DND-26.
Composite images of the fluorescence and differential
interference contrast (Figure 4e, j) confirm that FG-H503 is
located within acidic compartments of this macrophage cell
line.
In summary, a new family of fluorescent macrocyclic
peptidomimetic compounds has been synthesized, and their
abilities as pH probes have been tested both in solution and in
live cells with positive results. The synthetic modular scheme
is an excellent approach for the development of new tailor-
made probes with targeted pH “windows” ideal for measure-
ment of the changing acidic environment of intracellular
organelles. Current efforts involve the expansion of the FG-H
family with other a-amino acids, spacers, and fluorophores,
and determination of the extent by which the peptidomimetic
nature of the new probes favors their performance in
biological media.
Received: June 2, 2005
Published online: September 15, 2005
Angew. Chem. Int. Ed. 2005, 44, 6504 –6508
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6507

Communications
bis(aminomethyl)anthracene derivative is negatively influenced
by the first protonated amine, as a result of the repulsion existing
between its ammonium group and the second incoming proton.
In this way, larger macrocycles like those presented here would
protonate the second amine (and hence lead to restoration of the
fluorescence) at a lower concentration of protons (higher pKa)
than the smaller macrocycles, as the former macrocycles would
accommodate the second positive charge more easily than the
latter. However, other contributions should not be disregarded,
in particular solvation effects in protonated species. A more
extensive study of a larger family of peptidomimetic molecules is
in progress.
Keywords: biosensors · fluorescent probes · macrocycles ·
peptidomimetics · protonation
.
[1] T. Hirano, K. Kikuchi, Y. Urano, T. Higuchi, T. Nagano, Angew.
Chem. 2000, 112, 1094 – 1096; Angew. Chem. Int. Ed. 2000, 39,
1052 – 1054.
[2] Z. H. Lin, M. Wu , M. Schaferling, O. S. Wolfbeis, Angew. Chem.
2004, 116, 1767 – 1770; Angew. Chem. Int. Ed. 2004, 43, 1735 –
1738.
[3] N. Umezawa, K. Tanaka, Y. Urano, K. Kikuchi, T. Higuchi, T.
Nagano, Angew. Chem. 1999, 111, 3076 – 3079; Angew. Chem.
Int. Ed. 1999, 38, 2899 – 2901.
[4] H. Kojima, Y. Urano, K. Kikuchi, T. Higuchi, Y. Hirata, T.
Nagano, Angew. Chem. 1999, 111, 3419 – 3422; Angew. Chem.
Int. Ed. 1999, 38, 3209 – 3212.
[19] For a recent and comprehensive study of the effect of variations
of the chemical structure on the photophysical properties of pH
probes useful in vivo, see: C. J. Fahrni, L. Yang, D. G. VanDerv-
eer, J. Am. Chem. Soc. 2003, 125, 3799 – 3812.
[5] a) P. D. Beer, P. A. Gale, Angew. Chem. 2001, 113, 502 – 532;
Angew. Chem. Int. Ed. 2001, 40, 486 – 516; b) R. Martínez-
Mµæez, F. Sancenón, Chem. Rev. 2003, 103, 4419 – 4476.
[6] 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) A. W. Czarnik, Acc. Chem. Res. 1994,
27, 302 – 308; c) Fluorescent Chemosensors for Ion and Molecule
Recognition, ACS Symp. Ser. 1993, 538.
[7] a) H. Izumi, T. Torigoe, H. Ishiguchi, H. Uramoto, Y. Yoshida,
M. Tanabe, T. Ise, T. Murakami, T. Yoshida, M. Nomoto, K.
Kohno, Cancer Treat. Rev. 2003, 29, 541 – 549; b) M. Chesler,
Physiol. Rev. 2003, 83, 1183 – 1221; c) M. M. Wu, J. Llopis, S.
Adams, J. M. McCaffery, M. S. Kulomaa, T. E. Machen, H.-P. H.
Moore, R. Y. Tsien, Chem. Biol. 2000, 7, 197 – 209; d) A. M.
Paradiso, R. Y. Tsien, T. E. Machen, Nature 1987, 325, 447 – 450.
[8] a) R. P. Haugland,
Handbook of Fluorescent Probes and
Research Products, 9th ed., Molecular Probes, Eugene, OR,
2002; b) Z. J. Diwu, C.-S. Chen, C. Zhang, D. H. Klaubert, R. P.
Haugland, Chem. Biol. 1999, 6, 411 – 418; c) J. Liu, Z. J. Diwu,
W.-Y. Leung, Bioorg. Med. Chem. Lett. 2001, 11, 2903 – 2905;
d) H.-J. Lin, P. Herman, J. S. Kang, J. R. Lakowicz, Anal.
Biochem. 2001, 294, 118 – 125; e) J. Liu, Z. Diwu, D. H. Klaubert,
Bioorg. Med. Chem. Lett. 1997, 7, 3069 – 3072.
[9] M. Schindler, S. Grabski, E. Hoff, S. M. Simon, Biochemistry
1996, 35, 2811 – 2817.
[10] J. M. Holopainen, J. Saarikoski, P. K. J. Kinnunen, I. Jarvela, Eur.
J. Biochem. 2001, 268, 5851 – 5856.
[11] Patent pending. Application number GB 0509245.7
[12] J. Becerril, M. Bolte, M. I. Burguete, F. Galindo, E. García-
Espaæa, S. V. Luis, J. F. Miravet, J. Am. Chem. Soc. 2003, 125,
6677 – 6686.
[13] a) F. Galindo, M. I. Burguete, S. V. Luis, Chem. Phys. 2004, 302,
287 – 294; b) F. Galindo, J. Becerril, M. I. Burguete, S. V. Luis, L.
Vigara, Tetrahedron Lett. 2004, 45, 1659 – 1662; c) B. Escuder, J.
Becerril, M. I. Burguete, F. Galindo, R. Gavara, J. F. Miravet,
S. V. Luis, G. Peris, Chem. Eur. J. 2004, 10, 3879 – 3890.
[14] J. S. Nowick, S. Insaf, J. Am. Chem. Soc. 1997, 119, 10903 –
10908.
[15] a) A. P. de Silva, R. A. D. D. Rupasinghe, J. Chem. Soc. Chem.
Commun. 1985, 1669 – 1670; b) R. A. Bissell, E. Calle, A. P.
de Silva, S. A. de Silva, H. Q. N. Gunaratne, J.-L. Habib-Jiwan,
S. L. A. Peiris, R. A. D. D. Rupasinghe, T. K. S. D. Samara-
singhe, K. R. A. S. Sandanayake, J.-P. Soumillion, J. Chem. Soc.
Perkin Trans. 2 1992, 1559 – 1564.
[16] LysoSensor and LysoTracker probes are available from Molec-
ular Probes, Eugene, OR, USA.
[17] D. Rehm, A. Weller, Isr. J. Chem. 1970, 8, 259 – 271.
[18] Such
a different protonation behavior must occur for a
combination of reasons. To partially explain the phenomenon,
the observation made years ago for related systems (ref. [15b]) is
relevant: The protonation of the second amine in a 9,10-
6508 www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6504 –6508