A photoactivable fluorophore based on thiadiazolidinedione as caging group

Article abstract of DOI:10.1021/ol9906965

(Matrix presented) Photoactivable ("caged*) fluorescent dyes and probes are crucial for temporally and spatially resolved tracer experiments, e.g., in cell biology or fluid physics. The thiadiazolidinedione 1 represents a new class of caged fluorophore. Upon UV-irradiation it releases in a rapid photoreaction with high quantum yield the azoalkane 2. Longer wavelength excitation of 2 to the singlet excited state results in strong and long-lived fluorescence with maximum intensity at 425 nm. It has been demonstrated that one single uncaging laser pulse suffices for time-resolved or steady-state detection of the fluorescence.

Full text of DOI:10.1021/ol9906965

ORGANIC
LETTERS
1999
Vol. 1, No. 4
603-605
A Photoactivable Fluorophore Based on
Thiadiazolidinedione as Caging Group
Gabriela Gramlich and Werner M. Nau*
Departement Chemie der UniVersita¨t Basel, Klingelbergstrasse 80,
CH-4056 Basel, Switzerland
nau@ubaclu.unibas.ch
Received May 28, 1999
ABSTRACT
Photoactivable (“caged”) fluorescent dyes and probes are crucial for temporally and spatially resolved tracer experiments, e.g., in cell biology
or fluid physics. The thiadiazolidinedione 1 represents a new class of caged fluorophore. Upon UV-irradiation it releases in a rapid photoreaction
with high quantum yield the azoalkane 2. Longer wavelength excitation of 2 to the singlet excited state results in strong and long-lived
fluorescence with maximum intensity at 425 nm. It has been demonstrated that one single uncaging laser pulse suffices for time-resolved or
steady-state detection of the fluorescence.
Photoactivable or “caged“ fluorophores are nonfluorescent
molecules that can be converted to a fluorescent form by a
photoinduced reaction. They are powerful tools to investigate
fluid dynamics in rheology and cell biology. Since their
introduction,1 they have been used in the study of cyto-
skeleton dynamics, e.g., that of actin microfilament with
caged resorufin2 and that of tubulin with caged fluorescein,3
or as chemical actinometers for flux determination in
biological tissue samples.4 In rheology they are invaluable
for the study of turbulent and laminar hydrodynamic flows
and scalar mixing studies.5
The design and optimization of appropriate caged fluo-
rescent probes and dyes presents a synthetic and mechanistic
challenge. The caged fluorophore should meet a number of
criteria: thermal stability, water solubility, ease of synthetic
accessibility, no fluorescence, UV-only absorption, and an
efficient as well as rapid photoactivation. The uncaged
fluorophore itself should be strongly fluorescent upon long-
wavelength excitation. In addition, the application for cellular
studies requires the caged fluorophore to be biostable and
biocompatible, the photoproducts to be nontoxic, and the
uncaged fluorophore to be photostable since the measure-
ments may extend over longer periods of time. In contrast,
for rheological applications the main emphasis lies on fast
uncaging rates.
To date, most of the established caged fluorophores are
fluorescein, rhodamine, and resorufin derivatives which
employ variants of the o-nitrobenzyl caging group.6,7 Long-
wavelength absorption (>360 nm), poor water solubility, and
low photolysis quantum yields may limit the practical use
of nitrobenzyl caging groups. The slow uncaging rates in
the µs to ms region present another drawback. On the other
hand, some fluorescent dyes suffer rapid photobleaching
(fluorescein, resorufin) or require more complex syntheses
(Q-rhodamines).6,7 Others like caged resorufins have lifetimes
of less than 1 h in cells or already display some fluorescence
before their photoactivation.6-8
(1) (a) Zweig, A. Pure Appl. Chem. 1973, 33, 389-410. (b) Krafft, G.
A.; Sutton, W. R.; Cummings, R. T. J. Am. Chem. Soc. 1988, 110, 301-
303.
(2) Theriot, J. A.; Mitchison, T. J. Nature 1991, 352, 126-131.
(3) Mitchison, T. J. J. Cell. Biol. 1989, 109, 637-652.
(4) Lilge, L.; Flotte, T. J.; Kochevar, I. E.; Jacques, S. L.; Hillenkamp,
F. Photochem. Photobiol. 1993, 58, 37-44.
(5) (a) Lempert, W. R.; Magee, K.; Ronney, P.; Gee, K. R.; Haugland,
R. P. Exp. Fluids 1995, 18, 249-257. (b) Guilkey, J. E.; Gee, K. R.;
McMurtry, P. A.; Klewicki, J. C. Exp. Fluids 1996, 21, 237-242.
(6) Haugland, R. P. Handbook of Fluorescent Probes and Research
Chemicals; Molecular Probes: Eugene, OR, 1996; pp 447-455.
(7) Mitchison, T. J.; Sawin, K. E.; Theriot, J. A.; Gee, K.; Mallavarapu,
A. Caged Compounds; Marriott, G., Ed.; Academic Press: New York, 1998;
Vol. 291, pp 63-78.
10.1021/ol9906965 CCC: $18.00 © 1999 American Chemical Society
Published on Web 07/16/1999

We report in this Letter the suitability of the thiadiazo-
lidinedione derivative 19-11 as a prototype for a new caged
fluorophore composed of a new caging group as well as a
new fluorescent probe, the 2,3-diazabicyclo[2.2.2]oct-2-ene
(2).12,13 3,4-Dialkyl-1-thia-3,4-diazolidine-2,5-diones are readily
accessible by Diels-Alder addition,14 and their use as
precursors for azoalkanes has long been recognized.9,10,15
Nonetheless, their potential to serve as caged fluorophores
has not been addressed. The reaction sequence of the
wavelength-selective release of the fluorescent dye and its
subsequent excitation to the fluorescent singlet excited state
is shown in Scheme 1.
extinction coefficient in the UV region (ꢀ ca. 5000 M-1 cm-1,
Table 1) it does not absorb above 270 nm. Irradiation of
precursor 1 with UV light (λ e 260 nm) releases quantita-
tively Fluorazophore-P (2)17 with medium to high quantum
yields, depending on solvent (Table 1). The absorption
spectrum of the resulting azoalkane 2 shows a UV window
(260-320 nm) and a comparatively weak absorption band
around 380 nm. The absorption spectra of 1 and its
photoproduct 2 are shown in Figure 1.
Scheme 1
Of special interest is the released fluorophore 2, which
exhibits an exceedingly long-lived fluorescence (up to 1 µs)16
and serves as the novel fluorescent probe Fluorazophore-P
for monitoring antioxidant activity in biological systems12
and for investigating supramolecular association kinetics.13
Accordingly, we refer to 1 as caged Fluorazophore-P.
The photophysical properties of 1 and 2 in different
solvents are shown in Table 1. These data are essential to
assess the practical suitability of 1 as a caged fluorophore.
Figure 1. Absorption spectra of the caged Fluorazophore-P 1
(dashed line, ꢀ scale) and the uncaged Fluorazophore-P 2 (solid
line, ꢀ′ scale) after 100% conversion of a 0.3 mM solution of 1 in
n-hexane.
The use of the high-energy UV light for uncaging may
present an obstacle for biological but not necessarily for
rheological studies. The small extinction coefficient of 2
presents another drawback which, however, is effectively
balanced by its high fluorescence quantum yield (φf up to
0.5, Table 1).16,18 As illustrated in Figure 2, a single uncaging
laser pulse (λuncage ) 248 nm)19 was sufficient to detect the
characteristic broad, unstructured fluorescence of 2 by steady-
state fluorimetry (λexc ) 350 nm). In contrast, the nonirra-
diated solution of 1 showed no fluorescence.
Table 1. Photophysical Data for the Caged (1) and Uncaged
(2) Fluorazophore-P
ꢀ(1)/M-1
cm-1
ꢀ(2)/M-1
cm-1
solvent
(λmax/nm)
Φ
uncage(1)a (λmax/nm) Φf(2)b Φd(2)c
n-hexane 4640 (217)
0.27d
0.15
0.05
193 (377)e 0.15
0.03f
0.01
CH3CN
water
5310 (217)
5750 (216)
114 (378)
48 (364)
0.41
0.20g 0.001
a Uncaging quantum yield (λuncage ) 254 nm) determined with preirra-
diated azobenzene as actinometer, cf. Gauglitz, G.; Hubig, S. J. Photochem.
1985, 30, 121-125. b Fluorescence quantum yield calculated by assuming
a natural fluorescence lifetime of 1700 ns according to ref 16. c Decom-
position quantum yield; Feth, M. P.; Greiner, G.; Rau, H.; Nau, W. M.,
unpublished results. d Independently determined by laser flash actinometry.
e See ref 18. f Value for n-heptane. g See ref 16.
The thiadiazolidinedione 111 is a colorless compound with
thermal stability at ambient temperature and, like 2, dissolves
both in organic solvents and in water. While 1 has a large
(8) Bendig, J.; Helm, S.; Hagen, V. J. Fluoresc. 1997, 7, 357-361.
(9) Corey, E. J.; Snider, B. B. J. Org. Chem. 1973, 38, 3632-3633.
(10) Moje´, S. W.; Beak, P. J. Org. Chem. 1974, 39, 2951-2956.
(11) Beak’s method (ref 10) was used for the synthesis of 1, cf.
Supporting Information.
(12) Nau, W. M. J. Am. Chem. Soc. 1998, 120, 12614-12618.
(13) Nau, W. M. Chimia 1999, 53, 217.
Figure 2. Fluorescence spectrum of Fluorazophore-P 2 upon
photolysis of a solution of 1 in n-hexane (0.3 mM).
604
Org. Lett., Vol. 1, No. 4, 1999

Time-resolved transient absorption spectroscopy revealed
that the uncaging process, monitored through the absorption
of 2 (λobs ) 375 nm), occurs faster than 1 µs, e.g., 250 ns in
acetonitrile. The photolysis yields CO and COS as the
presumed side products, in analogy to the parent compound
1-thia-3,4-diazoline-2,5-dione that decomposes thermally to
N2, CO, and COS.10
a nonfluorescent precursor, ease of synthesis, low molecular
weight, favorable solubility, the long-wavelength absorption
of the photoproduct, and its strong fluorescence. Most
importantly, the photolytic release is efficient and fast (<1
µs). The uncaged fluorophore 2 exhibits additional interesting
properties which include the transparency in the UV region,
good water solubility, and exceedingly long fluorescence
lifetime.16 The high photostability of 2, indicated by the low
decomposition quantum yields (φd in Table 1),16,18 should
prevent unwanted photobleaching in long-time measure-
ments, e.g., for biological studies.
For further application several DBO derivatives,21 e.g., the
hydroxymethyl or carboxyl substituted ones 3 and 4, could
be employed in their caged form. These substitution patterns
allow covalent attachment to molecules of interest, e.g.,
peptides.
The suitability of the photoactivable probe 1 for temporally
and spatially resolved investigations could be demonstrated
by a two-photon two-color flash photolysis experiment. The
fluorophore 2 was generated with one single laser pulse19 at
λuncage ) 248 nm from a thiadiazolidinedione 1 solution and
subsequently excited with a second laser pulse20 at λprobe
)
355 nm. While the uncaging laser flash results in an
inevitable autoluminescence, the probing pulse gives fluo-
rescence with the typical long fluorescence lifetime of 2
(τ ) 690 ns)16 which can be readily differentiated as
illustrated in Figure 3.
Acknowledgment. This work was supported by the Swiss
National Science Foundation. We thank Dr. T. Winkler
(Novartis Agro, Basel) for a sample of 1-thia-3,4-diazolidine-
2,5-dione and Prof. J. Wirz for helpful comments.
Supporting Information Available: Synthesis and char-
acterization for compound 1. This material is available free
of charge via the Internet at http://pubs.acs.org.
OL9906965
(14) The Diels-Alder reaction is carried out by addition of the
appropriate diene to the in situ generated 1-thia-3,4-diazoline-2,5-dione.
Subsequent hydrogenation yields the thiadiazolidinedione product.
(15) Squillacote, M.; De Felippis, J. J. Org. Chem. 1994, 59, 3564-
3571.
(16) Nau, W. M.; Greiner, G.; Rau, H.; Wall, J.; Olivucci, M.; Scaiano,
J. C. J. Phys. Chem. A 1999, 103, 1579-1584.
(17) This work (by UV) and ref 15 (by NMR).
Figure 3. Two-photon two-color flash photolysis experiment. A
solution of 1 in acetonitrile (0.2 mM) was irradiated with a single
uncaging pulse and subsequently excited with a second laser pulse.
The emission was detected at λobs ) 425 nm (arbitrary time delay).
(18) Mirbach, M. J.; Liu, K.-C.; Mirbach, M. F.; Cherry, W. R.; Turro,
N. J.; Engel, P. S. J. Am. Chem. Soc. 1978, 100, 5122-5129.
(19) COMPex 205 laser, fwhm ca. 20 ns, pulse energy ca. 207 mJ.
(20) Nd:YAG laser (Continuum Surelite II), fwhm ca. 5 ns, pulse energy
ca. 10 mJ.
(21) Engel, P. S.; Horsey, D. W.; Scholz, J. N.; Karatsu, T.; Kitamura,
A. J. Phys. Chem. 1992, 96, 7524-7535.
In conclusion our investigations show that the caged
fluorescent probe 1 fulfills the desirable features for photo-
activable fluorophores, namely the UV-induced release from
Org. Lett., Vol. 1, No. 4, 1999
605