Enhanced ratiometric fluorescent indicators for magnesium based on azoles of the heavier chalcogens

Article abstract of DOI:10.1039/c4cc04460f

Red-shifted fluorescent indicators for magnesium were developed by incorporation of sulfur or selenium in the azole moiety of 'fura' fluorophores. Single atom replacement in the acceptor of these ITC probes affords longer excitation and emission wavelengths as well as greater separation between excitation bands, valuable for ratiometric intracellular Mg²⁺ imaging. the Partner Organisations 2014.

Full text of DOI:10.1039/c4cc04460f

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Enhanced ratiometric fluorescent indicators for
magnesium based on azoles of the heavier
chalcogens†
Cite this: Chem. Commun., 2014,
0, 11358
5
Received 12th June 2014,
Accepted 23rd July 2014
Mohammad S. Afzal, Jean-Philippe Pitteloud and Daniela Buccella*
DOI: 10.1039/c4cc04460f
www.rsc.org/chemcomm
Red-shifted fluorescent indicators for magnesium were developed significant progress has been made in the design of new gen-
9
by incorporation of sulfur or selenium in the azole moiety of ‘fura’ erations of improved sensors. Ratiometric indicators, however,
fluorophores. Single atom replacement in the acceptor of these ITC remain scarce.
probes affords longer excitation and emission wavelengths as well
For live cell imaging, nontoxic dyes with high brightness and
as greater separation between excitation bands, valuable for ratio- excitation and emission wavelengths in the visible range are desir-
2+
metric intracellular Mg imaging.
able. Lower energy light minimizes sample damage, scattering, and
background caused by autofluorescence. Tuning the excitation or
1,2
Fluorescent indicators are some of the most versatile tools for the emission wavelength of indicators for fluorescence microscopy
study of the cell biology of metals, allowing for the rapid detection of typically involves introducing significant changes in the structure
ions with remarkable selectivity, sensitivity, and spatial resolution of the fluorophores, which, for small-molecule indicators, often
1,2
via fluorescence microscopy. Wavelength ratiometric indicators results in changes in their cellular uptake and localization pro-
are a class of probes that exhibit a change in excitation or emission perties. The ability to tune the photophysical features of a probe
wavelength upon interaction with an analyte, thus enabling the use with minimal change of its structural and chemical properties is
3
of a ratio of intensities at two wavelengths as the readout. In the highly desirable. Sulfur and, to a lesser extent, selenium and
context of biological imaging of metals, wavelength ratiometric tellurium have been exploited for the improvement of common
10
indicators are particularly useful, as they provide quantitative organic luminophores, affording compounds with longer absorp-
information in a wide range of ion concentrations and sample tion and emission wavelengths than their oxygen-containing coun-
matrices, with fluorescence outputs that are insensitive to uneven terparts. Furthermore, Moerner and coworkers recently reported the
sensor concentration, photobleaching, and fluctuations in excita- incorporation of selenium into an amino-D-luciferin scaffold, which
3,4
tion intensity, among other external factors. A number of ratio- led to a noticeable red shift in the bioluminescence of the com-
11
metric indicators display conjugated donor–acceptor structures that pound with respect to the sulfur-based parent. In our efforts to
undergo photoinduced intramolecular charge transfer (ICT). Inter- develop improved chemical tools for the study of biological magne-
action of the metal cation with either donor or acceptor moieties sium, we surmised that the incorporation of heavier chalcogens into
may be employed to stabilize or destabilize the charge transfer state, the structure of the Mag-fura-2 indicator would aid in shifting its
5
thus affecting the excitation and emission wavelength. The fura excitation and emission to longer wavelengths, more amenable for
6
,7
2+
family of indicators is a notable example of such a design applied live cell fluorescence imaging of Mg . We demonstrate herein that
6
to the detection of metal ions in biological samples. Fura-2 and its strategic single atom replacement of the chalcogen in the acceptor
8
congener Furaptra (commercially available as Mag-Fura-2), have oxazole moiety of this magnesium-responsive ICT indicator induces
been instrumental in the elucidation of numerous physiological a large shift in emission wavelength and an increase in Stokes shift,
and pathological roles of calcium and magnesium, respectively, without affecting significantly the essential chemical features of
despite various limitations associated with their short excitation the compound. Most significantly, we show that the sulfur- and
and emission wavelengths. In the context of magnesium detection, selenium-based dyes display greater separation between excitation
maxima associated with their metal-free and metal-bound forms,
Department of Chemistry, New York University, New York, NY 10003, USA.
a feature that becomes advantageous in ratiometric imaging.
For the incorporation of sulfur or selenium into the Mag-fura-2
structure, we prepared 2-methyl azoles 1a and 1b (Scheme 1) by
E-mail: dbuccella@nyu.edu
†
Electronic supplementary information (ESI) available: Supporting figures, full
experimental details and crystallographic information files. CCDC 1007936 and
007937. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c4cc04460f
cyclization of ethyl 2-chloro-3-oxopropanoate and either thioacet-
1
1
2
amide or freshly prepared selenoacetamide, respectively.
1
1358 | Chem. Commun., 2014, 50, 11358--11361
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
Scheme 1 Synthesis of indicators Mag-S and Mag-Se. Conditions: (a)
MeCN, reflux; (b) 1,3-dibromo-5,5-dimethylhydantoin, AIBN, AcOH(cat)
,
i
C
6
H
5
Cl–CCl
4
, 75 1C; (c) HPO(OEt)
2 2 2 3
, Pr EtN, THF, 0 1C; (d) K CO , DMF,
1
00 1C; (e) KOH, MeOH–H
2
O; (f) HCl or buffer pH 7.
Subsequent unselective radical bromination led to mixtures of Fig. 1 Fluorescence excitation spectra of 2 mM solutions of Mag-S (A)
2
-bromomethylazole and 2-dibromomethylazole products. Treatment and Mag-Se (C) with increasing concentrations of MgCl
2
(50 mM PIPES,
1
00 mM KCl, pH 7.0, 25 1C). Double reciprocal plots (B) and (D) show the
of the mixtures with diethylphosphite and H u¨ nig’s base enabled
2+
dependence of the emission ratio on Mg concentration for Mag-S and
Mag-Se, respectively.
conversion of the undesired gem-dibromo byproducts to afford pure
13
2
-bromomethylazoles 2a,b. These precursors were employed for the
final assembly of the benzofuran fluorophores via condensation with
8
salicylaldehyde 6, bearing the protected magnesium-binding moiety. of Mag-S and Mag-Se, the Stokes shifts are very large, reaching
The solid-state structure of the resulting dyes in their ester form was B200 nm in the metal-saturated state and in general B30 to 40 nm
determined by single crystal X-ray diffraction, as shown in Fig. S12 larger than those of Mag-fura-2.
2+
(ESI†). Metal responsive sensors Mag-S and Mag-Se in the free acid
Both Mag-S and Mag-Se compounds respond to Mg showing a
form may be obtained via base-catalyzed hydrolysis of the esters. The significant blue shift in the fluorescence excitation and emission
deprotected o-aminophenol-N,N-triacetic acid (APTRA) group, how- spectra (Fig. 1), consistent with a destabilizing effect of the cation
ever, is susceptible to degradation in acidic solutions. Accordingly, we on an excited state characterized by a large dipole moment. The
chose to store the sensors in their more stable ester forms; stock appearance of two distinguishable excitation bands makes the
solutions of the free acids for photophysical characterization were indicators suitable for excitation-ratiometric determination of mag-
prepared by quantitative hydrolysis followed by neutralization via nesium in solution. The dissociation constants for 1 : 1 indicator-to-
dilution into aqueous buffer right before use.
metal stoichiometry complexes at 25 and 37 1C were determined
Photophysical properties of the new indicators were studied in from linear fits of the intensity ratios, yielding low millimolar
aqueous buffer at pH 7.0, as summarized in Table 1 and Fig. S1 values (Mag-S Kd,Mg2+ = 3.2 Æ 0.1 mM; Mag-Se Kd,Mg2+ = 3.3 Æ 0.3
(
ESI†). The thiazole derivative, Mag-S, displays an excitation maxi- at 25 1C, Table 1) close to those of the parent Mag-fura-2 and well in
16,17
mum at 392 nm, with emission at 572 nm (F392 = 0.17 Æ 0.01) in the tune with typical intracellular free magnesium concentrations.
metal-free form. The selenazole analogue Mag-Se shows an excita- Remarkably, the separation between excitation bands for the metal-
tion maximum at 410 nm, with emission maximum at 584 (F410 free and -bound sensors based on the electron poorer thiazole and
=
0
.09 Æ 0.02) in the free state. Thus, a single replacement of O by Se selenazole is larger than that of the oxazole derivative. This
in the structure of the fluorophore affords a greater than 70 nm observation is indicative of the heavier chalcogen exerting a greater
bathochromic shift in the emission of the indicator. The lower influence in stabilizing the excited state of the metal-free dye than
fluorescence quantum yield of Mag-Se compared to that of the the locally excited state of the metal-bound form. This feature is
lighter chalcogen counterparts is likely due to a greater rate of desirable for ratiometric imaging; as the separation between bands
14,15
intersystem crossing to a triplet state.
Large Stokes shifts in increases and their overlap is reduced, a greater portion of each
polar solvents are typical of fluorophores based on ICT. In the case band may be excited using wider excitation filters, thus increasing
a
Table 1 Photophysical properties of new sensors Mag-S and Mag-Se, compared to oxazole analogue Mag-fura-2
3
À1
À1
b
Absorption lmax (nm), e  10 (M cm
Unbound
)
Excitation lmax (nm)
Emission lmax (nm), F
K
d,Mg2+ (mM)
Kd,Ca2+ (mM)
2+
2+
2+
Mg -saturated
Unbound Mg -saturated Unbound
Mg -saturated 25 1C 37 1C 25 1C 37 1C
c
c
d
d
c
d
c
d
c
d
c
Mag-fura-2 369, 22
Mag-S
Mag-Se
330, 24
370
392
410
335
347
361
511, 0.24
491, 0.30
1.9
1.5
25
—
396, 29.1(7)
412, 26.6(8)
350, 30(1)
360, 28.3(9)
572, 0.17(1) 547, 0.30(1)
584, 0.09(2) 562, 0.18(1)
3.2(1) 1.97(7) 38(1) 35.3(7)
3.3(3) 2.3(1) 41(2) 37.0(5)
a
b
2 4
Measurements performed in 50 mM PIPES, 100 mM KCl, pH 7.0 at 25 1C unless otherwise indicated. Quinine sulfate in 0.5 M H SO
18
c
d
(F347 = 0.546) was employed as a fluorescence standard. Properties reported at 22 1C, from ref. 7. From ref. 8.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 11358--11361 | 11359

View Article Online
ChemComm
Communication
the overall signal without compromising the selective excitation of fluorescence emission above 500 nm (Fig. 2). The absolute intensity
the free and metal-saturated forms of the dye (i.e. minimal cross- images captured for each excitation channel (Fig. S13, ESI†) reveal a
talk between excitation channels).
relatively even staining of the entire cell, with little evidence for
We evaluated the fluorescence emission of the new com- dye compartmentalization. The ratio image (Fig. 2B), more clearly
pounds in response to various other biologically relevant metal shows the differences in magnesium levels in various intra-
2
+
2+
ions including Ca , Zn and the first row transition metals cellular compartments, with lower levels in the nucleus compared
2
+
2+
from Mn to Cu (Fig. S2 and S3, ESI†). The metal selectivity to surrounding organelles. To confirm the responsiveness of the
2+
of the Mag-S and Mag-Se probes is comparable to that of other indicator to Mg , the imaging medium was supplemented with
7
reported APTRA-based metal ion indicators. The fluorescence non-fluorescent ionophore 4-bromo-A-23187 (Molecular Probes)
ratio is insensitive to low micromolar concentrations of most and 30 mM of MgCl
2
for 60 min before capturing a second set of
2+
ions tested, with the exception of Zn (Mag-S K
Mag-Se Kd,Zn2+ = 60 Æ 10 nM at 25 1C). Absolute fluorescence per cell was observed (Fig. 2D and E), in response to the increase
d,Zn2+
= 62 Æ 3 nM; images. A clear increase (B17%) in the average fluorescence ratio
19
2+
intensity values, however, show a decrease in the presence of the in intracellular free Mg mediated by the ionophore.
paramagnetic transition metals, thus implying some degree of In summary, we have synthesized and characterized a new set
binding that leads to quenching of the fluorescence. Despite this of red-shifted ratiometric indicators for biological magnesium
effect, the typical low concentrations of these trace metals in based on modified ‘fura’ fluorophores. The new compounds
biological samples, and the negligible impact of their presence on illustrate how the heavier chalcogens sulfur and selenium may
the fluorescence ratio, suggests they would cause little inter- be exploited to enhance the properties of fluorophores beyond the
ference in magnesium detection. In the case of calcium, a change realm of fluorescence tags and labels, but also to enhance the
in the fluorescence ratio is observed in the mid-micromolar concen- properties of ion responsive fluorescent indicators. The replace-
tration range. The new sensors may be, therefore, applied as low ment of the chalcogen in the azole acceptor group of the fura ICT
affinity calcium indicators (Mag-S Kd,Ca2+ = 38 Æ 1 mM and Mag-Se dye, remote from the metal binding moiety, induces significant
2+
K
d,Ca2+ = 41 Æ 2 mM at 25 1C, Table 1) for the detection of high Ca
changes in the photophysical properties of the indicators with little
impact in metal selectivity, dissociation constants, and cellular
concentrations such as those associated with calcium fluxes.
Taking into consideration the superior brightness of the uptake properties that are essential for intracellular ion detection.
sulfur-containing indicator over the selenium analogue, we Significantly, the 2-oxazolyl substituted 6-aminobenzofuran is the
chose to test the response of Mag-S for the ratiometric detection fluorescent core for various ion-selective fluorescent probes, not
2
+
2+
2+
of Mg in live cells. To facilitate cellular uptake, the negatively only for detection of Mg and of Ca , but also functionalized with
2
0
2+
21
22
charged carboxylates were masked as acetoxymethyl esters.
metal-chelating groups specific for Zn (FuraZin and ZnAF-R2 ),
+
23
24
HeLa cells were loaded with 2 mM of the membrane-permeable and for Na (SBFO and related crown ether-based indicators ).
dye, followed by washing and an additional 25 min incubation The chalcogen replacement strategy described herein may, there-
period to allow for the full de-esterification of the probe by fore, open the doors to enhanced red-shifted ratiometric detection
intracellular esterases. Live cells were then imaged using a for other biologically relevant ions.
custom ratiometric filter set (see ESI† for details), capturing
This work was supported by start-up funds from NYU to
D. B. The authors thank Mr. Guangqian Zhang for fluorophore
precursors. The Bruker AXS SMART APEXII Diffractometer was
acquired through the support of the Molecular Design Institute at
NYU. Bruker Avance-400 and 500 NMR Spectrometers were acquired
with support of the NSF under award number CHE-01162222.
Notes and references
1
2
3
K. P. Carter, A. M. Young and A. E. Palmer, Chem. Rev., 2014, 114,
564–4601.
D. W. Domaille, E. L. Que and C. J. Chang, Nat. Chem. Biol., 2008, 4,
68–175.
J. R. Lakowicz, Principles of Fluorescence Spectroscopy, Kluwer Academics/
Plenum, New York, 2nd edn, 1999.
4
1
4
5
6
A. Demchenko, J. Fluoresc., 2010, 20, 1099–1128.
B. Valeur and I. Leray, Coord. Chem. Rev., 2000, 205, 3–40.
G. Grynkiewicz, M. Poenie and R. Y. Tsien, J. Biol. Chem., 1985, 260,
3440–3450.
7
8
9
R. P. Haugland, Handbook of Fluorescent Probes and Research
Products, Molecular Probes Inc., Eugene, Oregon, 9th edn, 2002.
B. Raju, E. Murphy, L. A. Levy, R. D. Hall and R. E. London, Am.
Fig. 2 Fluorescence imaging of intracellular free magnesium in live HeLa
cells treated with 2 mM Mag-S acetoxymethyl ester before (A, B) and after
(C, D) 60 min incubation with 2.5 mM magnesium ionophore 4-bromo-A-
J. Physiol.: Cell Physiol., 1989, 256, C540–C548.
For recent examples of indicators for fluorescence imaging of Mg
2
3187 and 30 mM exogenous MgCl . (A, C) DIC images; (B, D) fluorescence
2
2+
,
ratio images. Scale bars represent fluorescence ratio. (E) Change in the
average fluorescence ratio per cell, reflecting an increase in intracellular
see: (a) T. Fujii, Y. Shindo, K. Hotta, D. Citterio, S. Nishiyama,
K. Suzuki and K. Oka, J. Am. Chem. Soc., 2014, 136, 2374–2381;
(b) C. Marraccini, G. Farruggia, M. Lombardo, L. Prodi, M. Sgarzi,
V. Trapani, C. Trombini, F. I. Wolf, N. Zaccheroni and S. Iotti, Chem.
2+
magnesium upon treatment with exogenous Mg and ionophore. Error
bars = standard error of the mean, N = 11.
11360 | Chem. Commun., 2014, 50, 11358--11361
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
Sci., 2012, 3, 727–734; (c) L. H. Lindenburg, J. L. Vinkenborg, 12 K. Geisler, W.-D. Pfeiffer, A. K u¨ nzler, H. Below, E. Bulka and
J. Oortwijn, S. J. A. Aper and M. Merkx, PLoS One, 2013, 8, e82009; P. Langer, Synthesis, 2004, 875–884.
d) X. Dong, J. H. Han, C. H. Heo, H. M. Kim, Z. Liu and B. R. Cho, 13 P. Liu, Y. Chen, J. Deng and Y. Tu, Synthesis, 2001, 2078–2080.
(
Anal. Chem., 2012, 84, 8110–8113; (e) Y. Zhao, B. Zheng, J. Du, 14 M. R. Detty and P. B. Merkel, J. Am. Chem. Soc., 1990, 112, 3845–3855.
D. Xiao and L. Yang, Talanta, 2011, 85, 2194–2201; ( f ) Y. Shindo, 15 T. Y. Ohulchanskyy, D. J. Donnelly, M. R. Detty and P. N. Prasad,
T. Fujii, H. Komatsu, D. Citterio, K. Hotta, K. Suzuki and K. Oka,
J. Phys. Chem. B, 2004, 108, 8668–8672.
PLoS One, 2011, 6, e23684; (g) V. Trapani, G. Farruggia, 16 A. M. P. Romani, Arch. Biochem. Biophys., 2011, 512, 1–23.
C. Marraccini, S. Iotti, A. Cittadini and F. I. Wolf, Analyst, 2010, 17 R. D. Grubbs, BioMetals, 2002, 15, 251–259.
1
35, 1855–1866.
d
0 For examples, see: (a) H. Takechi, Y. Oda, N. Nishizono, K. Oda and 19 The oxazole analogue, Mag-fura-2, binds Zn with apparent K of
2+
18 A. M. Brouwer, Pure Appl. Chem., 2011, 83, 2213–2228.
2+
1
M. Machida, Chem. Pharm. Bull., 2000, 48, 1702–1710; (b) M. R.
Detty, P. N. Prasad, D. J. Donnelly, T. Ohulchanskyy, S. L. Gibson
and R. Hilf, Bioorg. Med. Chem., 2004, 12, 2537–2544; (c) B. Calitree,
20 nM and has been employed for ratiometric measurement of Zn
in the mid to high nanomolar range. See: T. J. B. Simons, J. Biochem.
Biophys. Methods, 1993, 27, 25–37.
D. J. Donnelly, J. J. Holt, M. K. Gannon, C. L. Nygren, D. K. 20 R. Y. Tsien, Nature, 1981, 290, 527–528.
Sukumaran, J. Autschbach and M. R. Detty, Organometallics, 2007, 21 K. R. Gee, Z. L. Zhou, D. Ton-That, S. L. Sensi and J. H. Weiss,
2
6, 6248–6257; (d) M. J. Rose and P. K. Mascharak, Inorg. Chem.,
Cell Calcium, 2002, 31, 245–251.
2
009, 48, 6904–6917; (e) M. W. Kryman, G. A. Schamerhorn, J. E. 22 S. Maruyama, K. Kikuchi, T. Hirano, Y. Urano and T. Nagano,
Hill, B. D. Calitree, K. S. Davies, M. K. Linder, T. Y. Ohulchanskyy
and M. R. Detty, Organometallics, 2014, 33, 2628–2640.
J. Am. Chem. Soc., 2002, 124, 10650–10651.
23 A. Minta and R. Y. Tsien, J. Biol. Chem., 1989, 264, 19449–19457.
1
1 N. R. Conley, A. Dragulescu-Andrasi, J. Rao and W. E. Moerner, 24 V. V. Martin, A. Rothe, Z. Diwu and K. R. Gee, Bioorg. Med. Chem.
Angew. Chem., Int. Ed., 2012, 51, 3350–3353. Lett., 2004, 14, 5313–5316.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 11358--11361 | 11361