Coumarinylmethyl caging groups with redshifted absorption

Article abstract of DOI:10.1002/chem.201302630

The small and synthetically easily accessible coumarinylmethyl backbone has been modified to generate a family of photolabile protecting groups with redshifted absorption. We relied on introducing electron-donating groups in the 7 position and electron-withdrawing groups in the 2-, and 2- and 3 positions. In particular, we showed that the diethylamino-thiocoumarylmethyl and the diethylamino-coumarylidenemalononitrilemethyl are relevant for uncaging with cyan light. They both exhibit a significant action cross section for uncaging in the 470-500 nm wavelength range and a low light absorption between 350 and 400 nm. These attractive features are favorable to perform chromatic orthogonal photoactivation with UV and blue-cyan light sources, respectively. Revealing protecting groups: The coumarinylmethyl backbone was used to generate a family of photolabile protecting groups with redshifted absorption (see scheme). Several members exhibit a significant action cross section for uncaging in the 470-500 nm wavelength range and a low light absorption between 350 and 400 nm. These features are favorable for chromatic orthogonal photoactivation with UV and blue-cyan light sources, respectively.

Full text of DOI:10.1002/chem.201302630

DOI: 10.1002/chem.201302630
Coumarinylmethyl Caging Groups with Redshifted Absorption
Ludovic Fournier,[a] Isabelle Aujard,[a] Thomas Le Saux,[a, b] Sylvie Maurin,[a]
Sandra Beaupierre,[a] Jean-Bernard Baudin,[a] and Ludovic Jullien*[a, b]
Abstract: The small and synthetically
easily accessible coumarinylmethyl
backbone has been modified to gener-
ate a family of photolabile protecting
groups with redshifted absorption. We
relied on introducing electron-donating
groups in the 7 position and electron-
withdrawing groups in the 2-, and 2-
showed that the diethylamino-thiocou-
marylmethyl and the diethylamino-cou-
marylidenemalononitrilemethyl are rel-
evant for uncaging with cyan light.
They both exhibit a significant action
cross section for uncaging in the 470–
500 nm wavelength range and a low
light absorption between 350 and
400 nm. These attractive features are
favorable to perform chromatic orthog-
onal photoactivation with UV and
blue-cyan light sources, respectively.
Keywords: caged compounds · pho-
tochemistry · photolysis · protecting
groups · UV/Vis spectroscopy
and
3
positions. In particular, we
Introduction
mitogen-activated protein kinases (MAPK) such as p38 and
c-Jun-N-terminal kinase (JNK).[15,16] Carotenoids have been
also incriminated since they are prone to UVA-induced pho-
todegradation and are involved in the control of gene ex-
pression, either directly or through their 1O2 quenching
action.[17]
Photolabile protecting (caging) groups that can be cleaved
after light absorption have recently proved attractive in vari-
ous fields of Chemistry and Biology.[1–13] In particular, caged
molecules have been already used to study the dynamics of
several biological processes with high spatio-temporal reso-
lution.[4–12] To be relevant in such experiments, the caging
group must fulfill various requirements dealing with non-
photophysical issues (access of the caged molecules to the
biological targets, absence of any “dark” biological activity,
etc.) but also depends on photophysics and on photochemis-
try (no biological effect of the illumination only, etc.).
Hence, the large number of engineering constraints has led
to the continuous introduction of new types of photolabile
protecting groups. However, the most available caging
groups still require ultraviolet (UV) light for photoactiva-
tion with one-photon excitation. This constraint may be sig-
nificant to specifically analyze the effect of a photoreleased
substrate at the level of the gene expression profile. Indeed,
many endogenous biologically active molecules significantly
absorb UV light, even in the Ultraviolet A range (UVA,
320–400 nm). Hence, illumination of riboflavin or flavin mo-
nonucleotide (FMN) can photosensitize triplet dioxygen[14]
The first strategy to overcome the limit of a maximum of
wavelength absorption lying below the 450 nm range re-
quires the tailoring of new caging groups based on chromo-
phores that absorb in the visible range. This approach has
been implemented by Chen and Steinmetz, who adapted the
rearrangement of an amino-substituted 1,4-benzoquinone to
photorelease carboxylic acids and phenols upon illuminating
at 542 nm.[18,19] Wirz and Klan have explored the xanthene
chromophore with absorption maxima in the 520 nm
range.[20] Ruthenium complexes have been used by Etcheni-
que et al. to photorelease complexed amines upon illuminat-
ing in the 400–500 nm wavelength range.[21–25] It is here also
relevant to mention the intramolecular multichromophoric
approach relying on an electron transfer from a photosensi-
tizer molecule to a reducible protecting group, which is ex-
plored by Lee and Falvey,[26] who showed it to be compatible
with photoactivation in the visible range.[27,28] The alterna-
tive strategy retains the established photochemistry with ex-
tension of the conjugation of the absorbing moiety. Depend-
ing on the investigated series, this approach has known suc-
cesses[29–31] but also limitations.[32] Hence, we previously
showed that the redshift absorption in the ortho-nitrobenzyl
series was accompanied by a drop in the uncaging efficien-
cy.[32] In contrast, efficient caging groups with maxima up to
415 and to 450 nm, respectively, have been reported by
Goeldner et al. in the nitrophenethyl series[29,30] and by
Ellis-Davies et al. in the coumarinylmethyl series.[31]
1
and the resulting singlet state O2 can subsequently induce
[a] Dr. L. Fournier, Dr. I. Aujard, Dr. T. Le Saux, S. Maurin,
S. Beaupierre, Prof. Dr. J.-B. Baudin, Prof. Dr. L. Jullien
Ecole Normale Supꢀrieure, Dꢀpartement de Chimie
UMR CNRS-ENS-UPMC, Paris 06 8640 PASTEUR
24, rue Lhomond, 75231 Paris Cedex 05 (France)
E-mail: Ludovic.Jullien@ens.fr
[b] Dr. T. Le Saux, Prof. Dr. L. Jullien
UPMC Paris 06, 4, Place Jussieu
75232 Paris Cedex 05 (France)
Among available photoremovable protecting groups, the
coumarinylmethyl moiety has gained wide acceptance.[31,33–36]
It is compatible with caging multiple organic functions[37–42]
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201302630.
17494
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17494 – 17507

FULL PAPER
and exhibits large uncaging cross sections with two-photon
excitation.[37] Thus, many caged biologically-active com-
pounds have been synthesized in this series.[31,34,35,37,38,40,43,44]
Eventually, the mechanism leading to the photolytic release
of the protected substrate has been extensively covered.[45–47]
In the context of caging groups, the coumarinylmethyl back-
bone has been already modified to shift its absorption spec-
trum up to the limit between the UV and visible ranges
a redshift the absorption spectrum; we chose the methoxy
and the diethylamino substituents. Moreover, to further red-
shift the light absorption, we considered increasing the pho-
toinduced intramolecular charge-transfer by introducing
electron-withdrawing groups on the 2- and 3 positions of the
coumarin backbone. We first targeted replacement of the
oxygen atom at the 2 position by various atoms and groups.
Hence, we retained 1) the nitrogen atom, which becomes
A
strongly electron-withdrawing when it is protonaACTHNGUTRENNUtG ed and cor-
7 position[37,38,48–51] or after coumarin thionation.[52,53] Howev-
er, literature data from various series of fluorophores have
suggested that the coumarin platform could exhibit even
more pronounced redshifted absorptions. In particular, up
to the 600 nm range has been reached upon appropriately
substituting the 2- and 3-coumarin positions.[54,55,56]
respondingly gives rise to redshifted absorption;[57] 2) the
sulfur atom, in which 3d empty atomic orbitals can be in-
volved in the intramolecular charge-transfer; 3) the conju-
gated vinylene malonitrile motif, which is strongly electron-
withdrawing. We also considered introducing an additional
electron-withdrawing group at the 3 position. We have re-
tained the cyano group, which can be introduced by various
synthetic pathways and we did not expect this to significant-
ly interfere with the coumarinylmethyl photochemistry
except for a redshift in light absorption. The design of the
putative caged model substrates was additionally governed
by the nature of the groups to be protected. Since the cou-
marinylmethyl group is appropriate to cage acidic substrates
(carboxylic acids, carbamic acids, etc.), we chose to cage
a carboxylic acid with absorption properties that were not
expected to alter the photochemical behavior of the coumar-
inylmethyl moiety: we adopted benzoic acid. Thus, we tar-
geted the model compounds displayed below.
In the present paper, we further explored the modification
of the coumarin backbone to obtain new photolabile pro-
tecting groups with enhanced redshifted absorption. More
specifically, we synthesized new coumarinylmethyl deriva-
tives substituted in conjugated positions and studied their
putative uncaging properties in vitro. The first section deals
with the syntheses of two series of putative caged model
compounds containing electron-donating groups in the 7 po-
sition and electron-withdrawing groups in the 2- and the 2-
and 3 positions, respectively. The second section reports on
the absorption and emission properties of the synthesized
species. In the third section, we measure uncaging action
cross sections for one-photon excitation. The final sections
are devoted to the discussion and the concluding remarks.
Their names of the caged benzoic acid model compounds
originate from applying the following rules: 1) the prefixes
OMe and NdiEt refer to the methoxy and diethylamino
borne on the 7 position; 2) 3-CN indicates the presence of
a cyano group at the 3 position; 3) the letters c (O), ic
(NH), oc (N(OH)), tc (S), and mc (C(CN)2) precise the
nature of the substituent at the 2 position; BA refers to ben-
zoic acid
Results
Design and syntheses
Design: In view of the literature results, we have first re-
tained 1) the 4 position of the coumarin backbone to intro-
duce the methylene for grafting the caged substrate and
2) an electron-donating group on the 7 position to cause
Syntheses: The syntheses of the 7-methoxy 3-non-substituted
coumarins are displayed in Scheme 1. The OMe-cBA com-
pound had been already obtained by esterification[58] from
Chem. Eur. J. 2013, 19, 17494 – 17507
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
17495

L. Jullien et al.
commercially available 7-diethylamino-4-methylcoumarin,
we first prepared the 7-diethylamino-4-hydroxymethylcou-
marin NdiEt-cOH upon adapting the reported two-step pro-
cedures (SeO2, dioxane/water, heat at reflux, 14 days, 58%;
NaBH4, ethanol, RT, 12 h, 72%).[38,68] Compound NdiEt-
cOH was then esterified with benzoic acid by using the cou-
pling reagent N,N-dicyclohexylcarbodiimide (DCC) to yield
the ester NdiEt-cBA[69] in 75% yield. This ester was subse-
quently converted into the corresponding thiocoumarin
NdiEt-tcBA (by using Lawessonꢂs reagent, toluene, heat to
reflux, 24 h) in 84% yield. In view of our poor results in the
methoxy series, we did not attempt to synthesize the imino-
coumarin NdiEt-icBA. In contrast, we prepared the dicya-
nocoumarin NdiEt-mcBA in 56% yield by condensing malo-
nonitrile onto the thiocoumarin NdiEt-tcBA in the presence
of triethylamine[70] and yellow lead oxide.[71]
Scheme 1. Syntheses of the caged benzoic acid compounds in the 7-MeO-
3-H series. a) Benzoic acid, KF, DMF, room temperature, 12 h, 71%;
b) Lawessonꢂs reagent, toluene, heated at reflux, 5 h, 78%; c) aq. NH3,
EtOH, 608C, 2 h, 3%; d) Hydroxylamine hydrochloride, sodium acetate,
EtOH, RT, 3 h, 70%.
The syntheses of the 3-cyano-7-methoxy coumarins are
displayed in Scheme 3. We aimed to introduce the cyano
group in a coupling from the corresponding bromo deriva-
benzoic acid and 4-bromomethyl-7-methoxy-coumarin.[59]
Starting from 4-chloromethyl-7-methoxy-coumarin,[60] we
correspondingly used the same procedure to access OMe-
cBA in 71% yield. Thiocoumarins (or coumarine thiones)
can be obtained in good yields in one step from the corre-
sponding coumarins by using Lawessonꢂs reagent.[61–64] Addi-
tional carboxylic esters are not affected under such condi-
tions.[62] We correspondingly converted OMe-cBA into
OMe-tcBA by using Lawessonꢂs reagent in toluene heated
at reflux in 78% yield. Inspired by works in the thiocarba-
mide[65] and thioester[66] series, we used ammonia to trans-
form OMe-tcBA into OMe-icBA. The resulting product de-
composed on silica gel and it was purified by basic washing
and extensive digestion with methanol to give the imino
coumarin OMe-icBA with 3% yield. In contrast, the oximo-
coumarin OMe-ocBA was easily prepared in 70% yield
from condensing hydroxylamine chlorhydrate in the pres-
ence of sodium acetate[63,67] onto OMe-tcBA.
Scheme 3. Syntheses of the caged benzoic acid compounds in the 7-MeO-
3-CN series. a) NBS, CHCl3, RT, 12 h, quantitative; b) NBS, CCl4, RT,
12 h, quantitative; c) Benzoic acid, KF, DMF, RT, 12 h, 90%; d) CuCN,
NMP, 1858C, 1 h, 53%; e) Lawessonꢂs reagent, toluene, heat at reflux,
5 h, 70%.
The syntheses of the 7-diethylamino 3-non-substituted
coumarins are displayed in Scheme 2. Starting from the
tive.[72] Since we failed to directly introduce a bromine atom
in the 3 position of OMe-cBA, we started from the 7-me-
thoxy-4-methyl-coumarin 2, which was selectively dibromi-
nated at the 3 position and on the methyl group using
a two-step procedure adapted from the literature to succes-
sively afford 3 and 4.[73] Benzoic acid was then esterified
with the benzyl bromide 4 to give the benzoic ester 5 in
90% yield. The cyano group was eventually introduced in
the 3 position by treatment of 5 with cuprous cyanide in N-
methylpyrrolidone at 1858C to provide OMe-3CN-cBA in
53% yield. Compound OMe-3CN-cBA was further convert-
ed with the Lawessonꢂs reagent in toluene into the corre-
sponding thiocoumarin OMe-3CN-tcBA in 70% yield.
The syntheses of the 3-cyano-7-diethylamino coumarins
are displayed in Scheme 4. In contrast to the 7-methoxy
series, we adapted the experimental conditions (N-bromo-
succinimide (NBS) in acetonitrile) to selectively introduce
Scheme 2. Syntheses of the caged benzoic acid compounds in the 7-
NdiEt-3-H series. a) Benzoic acid, DCC, 4-dimethylaminopyridine
(DMAP), CH2Cl2, RT, 12 h, 75%; b) Lawessonꢂs reagent, toluene, heat at
reflux, 6 h, 84%; c) Malonitrile, PbO, Triethylamine, toluene, 908C, 24 h,
56%.
a bromine atom in the 3 position of the NdiEt-cBA cou
ACHTUNGTRENNUNGma-
AHCTUNGTREGrNNUN in and obtain 6 in 78% yield. Then, we substituted the
17496
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17494 – 17507

Coumarinylmethyl Caging Groups
FULL PAPER
Absorption and emission properties: In view of possible bio-
logical applications, the following in vitro physicochemical
experiments were performed in a mixture of water and an
organic solvent (for substrate dissolution). We adopted
20 mm Tris pH 7.5/acetonitrile (1:1, v/v).
In a first step, we analyzed the absorption properties of
the putative caged model compounds and of the correspond-
ing alcohols. We also examined the luminescence emission
of the investigated compounds, which may be significant for
biological applications. Indeed, it may be essential to avoid
any interference between the emission from the starting
caged molecule and from the released substrate (tracer ap-
plications) or from any species produced after substrate re-
lease (detection of a biological response to a stimulus). The
main features of the absorption and emission properties of
the putative caged model compounds and the corresponding
alcohols are given in Table 1.
Scheme 4. Syntheses of the caged benzoic acid compounds in the 7-
NdiEt-3-CN series. a) NBS, CH3CN, RT, 2 h, 78%; b) CuCN, NMP,
1808C, 1 h, 81%.
bromo for the cyano group by using CuCN in N-methyl-2-
pyrrolidone (NMP) to give NdiEt-3CN-cBA in 81% yield.
In view of the promising results obtained during the pre-
liminary uncaging experiments with NdiEt-tcBA and NdiEt-
mcBA (see below), we eventually decided to synthesize the
corresponding benzylic alcohols, NdiEt-tcOH and NdiEt-
mcOH, to have reference compounds to analyze the forth-
coming uncaging experiments, as well as to get starting ma-
terials appropriate to cage other substrates than benzoic
acid.
We first considered relying on available synthons to
access NdiEt-tcOH and NdiEt-mcOH. Thus, we attempted
to treat NdiEt-cOH with the Lawessonꢂs reagent in toluene
but failed to obtain the desired thiocoumarin NdiEt-tcOH.
We were also unable to cleanly hydrolyze the ester NdiEt-
tcBA either under acidic or basic conditions. Therefore, we
decided to synthesize a more labile ester than NdiEt-tcBA,
which could react with malonitrile and be hydrolyzed to
yield both NdiEt-tcOH and NdiEt-mcOH (Scheme 5). Thus,
the alcohol NdiEt-cOH was esterified with acetic acid by
using DCC to provide the ester NdiEt-cAA in 85% yield.
This ester was subsequently transformed into the corre-
sponding thiocoumarin NdiEt-tcAA (Lawessonꢂs reagent,
toluene, heated at reflux, 6 h) in 92% yield. The latter thio-
coumarin was subsequently hydrolyzed (1.25m HCl in etha-
nol, heated at reflux, 15 h) to provide the first targeted alco-
hol NdiEt-tcOH in 75% yield. NdiEt-tcAA was also treated
with malonitrile in the presence of 4-dimethylaminopyridine
and yellow lead oxide to afford the ester NdiEt-mcAA with
85% yield. The second targeted alcohol NdiEt-mcOH was
eventually obtained in 97% yield upon hydrolyzing NdiEt-
mcAA under acidic conditions.
The absorption maxima are within the 320–430 and 380–
490 nm wavelength ranges in the 7-methoxy and 7-diethyl-
AHCTUNGERTGaNNUN mino series, respectively, with molar absorption coefficients
between 7000 and 35000mꢁ1 cmꢁ1. The molar absorption co-
efficients of the most redshifted species (NdiEt-3CN-cBA,
NdiEt-tcBA, and NdiEt-mcBA) are also interestingly ten
times lower around 365 nm than at their maximal absorption
wavelength.
Our results satisfactorily compare with the absorption
properties of similar chromophores. For instance, OMe-cBA
was reported to exhibit lmax =324 nm and e=13000mꢁ1 cmꢁ1
in a similar solvent.[46] Moreover, the absorption maxima
and molar absorption coefficients of 7-diethylamino-4-
methyl-thiocoumarin (lmax =460 nm and e=28900mꢁ1 cmꢁ1)
and
7-diethylamino-4-methyl-coumarylidenemalononitrile
(lmax =475 nm and e=31600mꢁ1 cmꢁ1) reported in ethanol[55]
are in agreement with the results of NdiEt-tcOH and NdiEt-
mcOH.
All other things being equal, the absorption maxima in
the diethylamino series are redshifted with respect to the
methoxy one. This observation was anticipated from the
lower electronegativity of the nitrogen atom leading to
a larger charge-transfer in the excited state. One can similar-
ly observe that the absorption maximum shifts to larger
values when the withdrawing electronic-demand of the
atom/group in the 2- (compare O, NH, NH2+, S, C(CN)2)
and 3 positions (compare -H
and -CN) of the coumarin back-
bone is increased. Eventually,
one notices that the benzoic
esters NdiEt-tcBA and NdiEt-
mcBA exhibit larger values of
their absorption maximum and
molar absorption coefficient
than their benzyl alcohol coun-
terparts
NdiEt-tcOH
and
NdiEt-mcOH.
Similar trends were observed
with the emissive properties
(see Table 1). Our observations
Scheme 5. Syntheses of the caging alcohols in the 7-NdiEt-3-H series. a) Acetic acid, DCC, DMAP, CH2Cl2,
RT, 12 h, 85%; b) Lawessonꢂs reagent, toluene, heat at reflux, 6 h, 92%; c) Malonitrile, PbO, DMAP, toluene,
908C, 24 h, 85%; d) and e) 1.25m HCl in ethanol, heat at reflux, 15 h, 75 and 97%, respectively.
Chem. Eur. J. 2013, 19, 17494 – 17507
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
17497

L. Jullien et al.
Tkach.[55] This different behavior may originate from the
nature of luminescence in the thiocoumarin series. Indeed,
phosphorescence instead of fluorescence has been invoked
to account for emission of thiones.[74–76]
Table 1. Absorption and emission properties of the putative caged model
compounds and of the corresponding alcohols.[a]
ð1Þ
ð1Þ
Compound
l
e(l ), e(365)
lðe1mÞ [nm],
(FðL1Þ[%])
max
max
[nm]
[mmꢁ1 cmꢁ1
]
OMe-cBA
323
325
330
330
398
360
427
385
379
443
472
467
487
478
13.5, 0.4
7, 3
6.5, 5
396, (14)
–
–
OMe-icBA (pH 8.5)[b]
OMe-icBA (pH 5.0)[b]
OMe-ocBA
Preliminary screening of the photochemical properties: In
a first step we screened the putative caged model com-
pounds for their thermal stability as well as for their photol-
ysis after one-photon absorption upon illuminating at
a common absorption wavelength (365 nm), which is repre-
sentative of the absorption range of most present caging
groups.
To evaluate the thermal stability, we recorded the tempo-
ral evolution of the absorption spectrum of 25 mm solutions
in 20 mm pH 7.5 Tris buffer/acetonitrile 1:1 (v/v) in the dark
at room temperature. Two compounds (OMe-icBA and
OMe-3CN-tcBA) exhibited an absorbance change at the 3 h
timescale; those compounds were discarded for the subse-
quent experiments. In contrast, the eight remaining model
compounds proved stable at the 3 h timescale and their pho-
tochemical features were examined.[77]
We submitted the remaining 25 mm solutions in 20 mm
pH 7.5 Tris buffer/acetonitrile (v/v) to the illumination of
a benchtop UV lamp mostly emitting at 365 nm. In this
series of experiments, we mainly focused on evidencing any
uncaging process leading to the photorelease of benzoic
acid. Aliquots were extracted after increasing illumination
durations. We analyzed the temporal dependence of the ab-
sorption spectrum and of the concentration in benzoic acid
measured by HPLC (see Figure 11S, the Supporting Infor-
mation). Except in the case of OMe-ocBA,[78] UV illumina-
tion altered the absorption spectrum and simultaneously led
to the photorelease of benzoic acid. Figure 1 displays the
typical results (see also Figures 12S–16S, the Supporting In-
formation). In the investigated range of illumination dura-
tions, two different behaviors have been encountered. Fig-
ure 1a and b display the behavior associated to a “fast” pho-
toinduced release of benzoic acid. The curvature observed
in the temporal dependence of both the normalized absorb-
ance and of the concentration in photoreleased benzoic acid
enabled the extraction of the action uncaging cross section,
the uncaging quantum yield, the molar absorption coeffi-
cient of the photoproducts, and the final concentration in
photoreleased benzoic acid (see the Experimental Section).
In contrast, Figure 1c and d show the behavior correspond-
ing to the regime of “slow” photoinduced release of the
caged substrate. The observed temporal dependences are
now linear. Since we did not investigate the nature of the
photoproducts at that step, their photophysical properties
have remained unknown so as to restrict the extraction of
the photochemical parameters associated to uncaging from
the temporal dependence of the concentration in photore-
leased benzoic acid (see the Experimental Section). The
overall results are given in Table 2.
14, 3.5
17, 10
20.5, 20
18.5, 5
24, 17.5
17, 14.5
26, 2
31, 2.5
23.5, 1
33, 3.5
32.5, 3
394, (1)
[c]
OMe-tcBA
–
OMe-3CN-cBA
OMe-3CN-tcBA
NdiEt-cBA
NdiEt-cOH
NdiEt-3CN-cBA
NdiEt-tcBA
NdiEt-tcOH
NdiEt-mcBA
NdiEt-mcOH
422, (64)
[c]
–
493, (11)
478, (18)
503, (1)
578, (2)
551, (1)
561, (17)
552, (19)
ð1Þ
[a] Maxima of single-photon absorption l , and of steady-state lumines-
max
cence emission after one-photon excitation lðe1mÞ , molar absorption coeffi-
cients for single-photon absorption at lð1Þ, eðlð1ÞÞ (ꢂ5%), quantum yields
of luminescence after one-photon excitation FðL1Þ (ꢂ10%). Solvent: ace-
tonitrile/Tris pH 7.5 buffer (20 mm) 1:1 (v/v) except for OMe-icBA.
[b] The pH was fixed at 8.5 in the 20 mm Tris buffer for the pH 8.5 ex-
periment, whereas we used a 20 mm pH 5.0 acetate buffer instead of the
Tris buffer for the pH 5.0 experiment. A pK in the 7.5 range has been re-
ported for iminocoumarins.[57] [c] Too weakly luminescent to extract relia-
ble values.
are in line with literature results when taking into account
solvatochromic effects. Hence, the emission wavelength of
NdiEt-cBA satisfactorily compares to the one of a reported
phosphoric ester in a similar solvent.[38] In contrast, the
emission wavelengths observed for NdiEt-tcOH and NdiEt-
mcOH are significantly redshifted with respect to the re-
ported values for 7-diethyl
ACHTUNGTNERaNUNG mino-4-methyl-thiocoumarin
(lmax =515 nm) and 7-diethylamino-4-methyl-coumarylidene-
AHCTUNGTRENNUNG
malononitrile (lmax =522 nm) in ethanol.[55] In fact, this dis-
crepancy originates from the positive solvatochromism of
the thiocoumarin and of the coumarylidenemalononitrile
(see the Supporting Information): we recorded consistent
values in ethanol.
Substituent effects are close to the ones observed in ab-
sorption and for the same reasons: All other things being
equal, emission is more redshifted 1) in the diethylamino
series than in the methoxy one, 2) when the withdrawing
effect of the atom/group in the 2- and 3 positions of the cou-
marin backbone is increased, 3) in the benzoic esters NdiEt-
tcBA and NdiEt-mcBA than in the corresponding benzyl al-
cohols NdiEt-tcOH and NdiEt-mcOH.
Eventually, the observed quantum yields of luminescence
cover a large range (from vanishing to large values) and
they exhibit contrasting trends. In the coumarin series,
which is the most fluorescent, we observed that the methoxy
series is more emissive than the diethylamino series (com-
pare OMe-cBA and NdiEt-cBA, and OMe-3CN-cBA and
NdiEt-3CN-cBA). This behavior is in agreement with previ-
ous observations.[45] We observed an opposite trend in the
thiocoumarin series (compare OMe-tcBA and NdiEt-tcBA),
which is overall poorly emissive as already observed by
In view of the wide acceptance of the diethylamino cou-
marinylmethyl moiety as a caging group, we used the results
obtained with NdiEt-cBA to define a threshold to further
17498
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17494 – 17507

Coumarinylmethyl Caging Groups
FULL PAPER
Figure 1. Evolution of solutions (25 mm) of two caged-model compounds (NdiEt-tcBA in a and b, and NdiEt-3CN-cBA in c and d) upon illuminating at
365 nm for various durations t(I0(365)=5.8ꢃ10ꢁ5 Einminꢁ1; t(min)=0, 0.3, 1, 2, 4, 7, and 12 in a and b, and 0, 5, 20, 30, 45, 60, 75 in c and d. a and
c) Temporal evolution of the solution absorbance divided by its concentration to afford its molar absorption coefficient eðlð1ÞÞ (the initial absorption
ð1Þ
~
spectrum is drawn with a dotted line); b and d) Temporal dependence of the normalized absorption coefficient at the absorption maximum eðlmaxÞ ( ).
The associated exponential (with Eq. (9), see the Supporting Information, in b) or linear (with Eq. (10), see the Supporting Information, in d) fit is
shown as a dotted line. For NdiEt-tcBA, we extracted kuð1Þ =(0.8ꢂ0.3) minꢁ1 and e(472,1)=(27000ꢂ1000)mꢁ1 sꢁ1 from the exponential fit; b and
*
d) Temporal evolution of the concentration [BA] of benzoic acid extracted from the HPLC chromatogram ( ). The exponential (with Eq. (6), see the
Supporting Information, in b) or linear (with Eq. (7), see the Supporting Information, in d) fit is shown as a solid line. For NdiEt-tcBA and NdiEt-3CN-
cBA, we respectively extracted kðu1Þ =(1.0ꢂ0.4) minꢁ1 and [BA]1 =(26ꢂ6) mm, and kðu1Þ =(0.001ꢂ0.0004) minꢁ1 using 25 mm for the initial concentration
in caged compound. Solvent: 20 mm pH 7.5 Tris buffer/acetonitrile (v/v). T=293 K.
screen the uncaging properties. In fact, our present 1–3ꢃ
10ꢁ3 range for Fðu1Þ(365) fairly compares with results report-
20 mm pH 7.5 Tris buffer/acetonitrile (v/v) to the illumina-
tion of the 75 W xenon lamp of a spectrofluorometer at
385 nm and we recorded the absorption spectrum of the so-
lution after various illumination times. Upon increasing the
illumination duration, Figure 17Sa (the Supporting Informa-
tion) shows that the absorbance 1) blueshifts from 385 to
378 nm and 2) drops (the molar absorption coefficient at
385 nm decreases from 24000 to 17000mꢁ1 cmꢁ1). Since the
photoreleased benzoate anion did not contribute to the ab-
sorbance in the investigated wavelength range, it was notice-
able that the absorbance of the illuminated NdiEt-cBA solu-
tion asymptotically tends to the absorbance of the reference
alcohol NdiEt-cOH (see Table 1). From the exponential
decay of the normalized absorbance on time (Figure 17Sb,
the Supporting Information), we derived the NdiEt-cBA
photo-consumption rate constant kðu1Þ =(0.065ꢂ0.002) minꢁ1.
After calibration of the photon flux at the sample, we ex-
ACHTUNGTRENNUNG
ed for a similar caged group (a carbonate) in a related sol-
vent mixture (potassium 3-(N-morpholino)propanesulfonic
acid (KMOPS), pH 7.2 buffer/acetonitrile, 3:1 v/v).[40]
Hence, three caged-model compounds (OMe-cBA, OMe-
tcBA, and OMe-3CN-cBA) were estimated to exhibit too
poor photochemical properties and were not further exam-
ined. In contrast, we eventually retained three caged-model
compounds (NdiEt-3CN-cBA, NdiEt-tcBA, and NdiEt-
mcBA), which were shown to quantitatively photorelease
benzoic acid at rates similar or larger than the ones of
NdiEt-cBA. They were subsequently examined in more
detail upon illuminating at their absorption maximum.
In a first step we relied on the reference NdiEt-cBA to
set up our spectrometric protocol for analyzing the uncaging
process. We submitted a solution of NdiEt-cBA (5 mm) in
Chem. Eur. J. 2013, 19, 17494 – 17507
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
17499

L. Jullien et al.
ð1Þ
Table 2. Action uncaging cross section of the caged model compounds for one-photon excitation at l
=
calibration of the photon flux at
the sample, we extracted from
the average NdiEt-3CN-cBA
photo-consumption rate con-
exc
365 nm, eð365ÞFðu1Þ(365) (in mꢁ1 cmꢁ1), and quantum yield of uncaging after one-photon excitation Fðu1Þ(365), as
measured from the temporal dependence of the normalized absorbance or of the concentration in photore-
leased benzoic acid measured by HPLC.[a]
ð1Þ
[c]
[d]
Compound Thermal eu(365)Fðu1Þ(365) 102Fðu1Þ(365) eðl 1Þ
eu(365)Fðu1Þ(365) 102Fuð1Þ(365) [BA]1
stant
and
kuð1Þ :Fuð1Þ =(4ꢂ1ꢃ10ꢁ4)
max
stability[b] [mꢁ1 cmꢁ1
]
[mmꢁ1 cmꢁ1
]
[mꢁ1 cmꢁ1
]
[mm]
eu
G
OMe-cBA Yes
OMe-icBA No
(pH 8.5)
–
–
–
–
–
–
0.16ꢂ0.04
0.04ꢂ0.01
–
–
cmꢁ1, respectively, for the quan-
tum yield of uncaging and for
the uncaging action cross sec-
–
–
OMe-icBA No
(pH 5.0)
–
–
–
–
–
–
tion with one-photon excitation
OMe-ocBA Yes
OMe-tcBA Yes
OMe-3CN- Yes
cBA
–
–
–
–
–
–
–
–
–
–
–
–
–
–
ð1Þ
at l =443 nm (see Table 3).
max
0.4ꢂ0.2
0.002ꢂ0.001
Figure 3a shows that the tem-
poral evolution of the absorb-
ance of the NdiEt-tcBA solu-
tion was more complex. In
a first kinetic regime, the ab-
sorbance slightly blueshifts
from 472 to 467 nm and drops
1.0ꢂ0.2
0.005ꢂ0.001
OMe-3CN- No
tcBA
–
–
–
–
–
–
NdiEt-cBA Yes
52 ꢂ35
0.3ꢂ0.2
19ꢂ1
17ꢂ2
0.1ꢂ0.01
29ꢂ2
NdiEt-
3CN-cBA
NdiEt-
tcBA
Yes
Yes
Yes
–
–
–
0.9ꢂ0.2
0.05ꢂ0.01
–
240ꢂ80
4.3ꢂ2.3
9ꢂ3
27ꢂ1
0ꢂ2
320ꢂ130
2.3ꢂ1.3
12ꢂ5
23ꢂ6
16ꢂ8
from e
N
NdiEt-
mcBA
0.13ꢂ0.08
0.07ꢂ0.04
mꢁ1 cmꢁ1 to e
A
1000)mꢁ1 cmꢁ1, yielding an ab-
sorption spectrum that interest-
ingly compared with the one of
the NdiEt-tcOH alcohol. Then,
we observed the gradual disap-
pearance of the absorption
band at 467 nm simultaneously
occurring with an absorption
[a] When an exponential fit was used to analyze the data, the asymptotic normalized absorbance at the absorp-
ð1Þ
tion maximum, eðl 1Þ, or concentration in benzoic acid, [BA]1, is given in the Table. Initial concentration
max
in caged model compound: 25 mm; Solvent: acetonitrile/Tris pH 7.5 buffer (20 mm). See the main text and the
Experimental Section. [b] Yes/No indicates whether any significant change of the absorption spectrum was ob-
served at the 3 h timescale. [c] Measured from the temporal dependence of the normalized absorbance.
[d] Measured from the temporal dependence of the concentration in photoreleased benzoic acid measured by
HPLC.
tracted from the rate constant Fðu1Þ =(2.1ꢂ0.2)ꢃ10ꢁ3 and
band growth at 365 nm. As displayed in Figure 3b, such a be-
havior could be satisfactorily fitted to a biexponential asso-
ciated to rate constants of (9ꢂ1) minꢁ1 and (0.016ꢂ
0.001) minꢁ1 with a vanishing asymptotic value of the molar
absorption coefficient of the solution. In the meantime, the
concentration in benzoic acid exponentially increases with
a rate constant of (9ꢂ1) minꢁ1 and converges towards 22 mm
(Figure 3c). The similarity between the largest rate con-
stants observed by UV absorption and HPLC suggested that
euACHTUNGTRENNUNG
yield of uncaging and for the uncaging action cross section
ð1Þ
with one-photon excitation at l =385 nm. These results
max
were in fair agreement with our observations at 365 nm (see
Table 2). Therefore, we considered that this experiment vali-
dated our protocol for measuring action uncaging cross sec-
tions and uncaging quantum yields. Thus, we submitted solu-
tions of NdiEt-3CN-cBA, NdiEt-tcBA, and NdiEt-mcBA in
20 mm pH 7.5 Tris buffer/acetonitrile (v/v) to the illumina-
tion of the 75 W xenon lamp of a spectrofluorometer at 443,
472, and 487 nm, respectively, and analyzed the temporal de-
pendence of their absorption spectrum. Furthermore adopt-
ing a concentration of 25 mm allowed us to measure the con-
centration in photoreleased benzoic acid by using HPLC.
The results are displayed in Figures 2–4.
the first kinetic regime was associaACHTUNTRGNEUNGted with the quantitative
uncaging of NdiEt-tcBA. After calibration of the photon
flux at the sample, from NdiEt-tcBA we extracted the
photo-consumption rate constant kðu1Þ : Fðu1Þ =(18ꢂ2)ꢃ10ꢁ2
and eu ACHTNUGTRENNUNG
yield of uncaging and for the uncaging action cross section
with one-photon excitation at l =472 nm (see Table 3).
ð1Þ
max
Upon increasing the illumination duration, Figure 2a
shows that the absorbance of the NdiEt-3CN-cBA solution
1) blueshifts from 443 to 430 nm and 2) drops (the molar ab-
sorption coefficient at 443 nm decreases from 25900 to
16500mꢁ1 cmꢁ1). The normalized absorbance exponentially
decays with time (Figure 2b) with an associated rate con-
stant equal to (0.015ꢂ0.003) minꢁ1 and 15000mꢁ1 cmꢁ1 for
the asymptotic value of the molar absorption coefficient of
the solution. In the meantime, the concentration in benzoic
acid exponentially increases with a rate constant of (0.014ꢂ
0.002) minꢁ1 and converges towards 15 mm. The latter result
evidenced that the uncaging of NdiEt-3CN-cBA occurred in
one step at the considered timescale in good yield. After
In view of the existence of the second kinetic regime, the
photoconversion of the NdiEt-tcBA solution was further in-
vestigated by analyzing the temporal evolution of the con-
centrations of the NdiEt-tcBA reactant and of the putative
photoproducts by using HPLC (Figure 3d). The decay rate
of (9ꢂ3) minꢁ1 associated with the total NdiEt-tcBA photo-
conversion was in line with the preceding uncaging interpre-
tation. Moreover, HPLC confirmed that NdiEt-tcOH alco-
hol formation at (7ꢂ2) minꢁ1 quantitatively resulted from
NdiEt-tcBA photoconversion. Furthermore, NdiEt-tcOH
was suggested to exponentially degrade at longer times with
a rate constant of (0.008ꢂ0.007) minꢁ1, which corresponded
to Fðd1Þ =(2.5ꢂ0.5)ꢃ10ꢁ4 and eðd1Þ(lð1Þ )(472)Fdð1Þ =(5.6ꢂ
max
17500
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17494 – 17507

Coumarinylmethyl Caging Groups
FULL PAPER
1.1)mꢁ1 cmꢁ1 for the quantum yield of degradation and for
the degradation action cross section with one-photon excita-
ð1Þ
tion at l =472 nm (see Table 4). This observation was
max
confirmed by an independent NdiEt-tcOH illumination ex-
periment (Figure 21S, the Supporting Information), which
similarly provided Fdð1Þ =(2.3ꢂ0.1)ꢃ10ꢁ4 for the quantum
ð1Þ
yield of degradation with one-photon excitation at l
=
max
472 nm.[79]
The degradation rate constant observed by HPLC fairly
compared with the second rate constant observed by UV/
Vis absorption. This agreement led us to assume that the
thiocoumarine chromophore disappeared during the degra-
dation step. Hence, upon considering the rise of an absorp-
tion band in the range expected for the coumarin chromo-
phore, we examined whether the formation of NdiEt-cOH
occurred during NdiEt-tcOH degradation. Indeed, we ob-
served the exponential growth of a peak associated with the
formation of
a photoproduct exhibiting the molecular
weight of NdiEt-cOH with a rate constant of (0.018ꢂ
0.003) minꢁ1, in line with the degradation rate. By using
a sample of NdiEt-cOH for calibration, our analysis suggest-
ed that the photoconversion of NdiEt-tcOH into NdiEt-
cOH was not quantitative (typically in the 35% yield)
making the formation of other photoproducts likely.
Figure 4a displays the temporal evolution of the absorb-
ance of the NdiEt-mcBA solution. We observed that the ab-
sorbance slightly blueshifts from 487 to 482 nm and drops
from e(487,1)=(33000ꢂ1000)mꢁ1 cmꢁ1 to e(482,1)=
(17500ꢂ1000)mꢁ1 cmꢁ1. Whereas the range of absorption
wavelength is in the range expected for the NdiEt-mcOH al-
cohol (see Table 1), the asymptotic molar absorption coeffi-
cient would be notably low for the diethylamino-coumaryli-
denemalononitrile chromophore. As displayed in Figure 4b,
the absorbance drop could be satisfactorily fitted to a mono-
exponential associated to
a
rate constant of (0.018ꢂ
0.001) minꢁ1 with the asymptotic value of the molar absorp-
tion coefficient of the solution being e(487,1)=(17000ꢂ
1000)mꢁ1 cmꢁ1. In the meantime, the concentration in benzo-
ic acid exponentially increases with a rate constant of
(0.016ꢂ0.001) minꢁ1 and converges towards 26 mm, which
suggested that the uncaging was quantitative. After calibra-
tion of the photon flux at the sample, we extracted from the
NdiEt-mcBA photo-consumption rate constant kðu1Þ : Fuð1Þ
=
(3.4ꢂ0.4)ꢃ10ꢁ4 and euð1Þ(487)Fuð1Þ =(11ꢂ1)mꢁ1 cmꢁ1 for the
quantum yield of uncaging and for the uncaging action cross
ð1Þ
section with one-photon excitation at l =487 nm (see
max
Figure 2. Evolution of a solution of NdiEt-3CN-cBA (25 mm) upon illumi-
Table 3).
nating at 443 nm for various durations t(I0(443))=3.15ꢃ10ꢁ6 Einminꢁ1
;
Beyond the desired observation of NdiEt-mcBA uncaging
liberating benzoic acid, the unexpected absorbance behavior
motivated us to further analyze the nature of the co-product
resulting from NdiEt-mcBA illumination. In analogy to
NdiEt-tcOH, we first hypothesized that the NdiEt-mcOH al-
cohol could be formed in a first step and then reacts photo-
chemically. After checking for the thermal stability of
NdiEt-mcOH, we correspondingly exposed a solution of
NdiEt-mcOH (25 mm) at 487 nm under the illumination con-
ditions, which had been used with NdiEt-mcBA (see Fig-
t(min)=0, 1, 3, 7, 15, 23, 77, and 120). a) Temporal evolution of the solu-
tion absorbance divided by its concentration to afford its molar absorp-
tion coefficient eðlð1ÞÞ (the initial absorption spectrum is drawn with
a dotted line); b) Temporal dependence of the normalized absorption co-
~
efficient at the absorption maximum e(443) ( ). The associated exponen-
tial fit with Eq. (9) (see the Supporting Information) is shown as a solid
line. We extracted kðu1Þ =(0.015ꢂ0.002) minꢁ1 and e(443,1)=(15000ꢂ
1000)mꢁ1 cmꢁ1 from the exponential fit; c) Temporal evolution of the
concentration [BA] of benzoic acid extracted from the HPLC chromato-
~
gram ( ). The exponential fit with Eq. (6) (see the Supporting Informa-
tion) shown as a solid line yields kðu1Þ =(0.014ꢂ0.002) minꢁ1 and [BA]1
=
(15ꢂ3) mm. Solvent: 20 mm pH 7.5 Tris buffer/acetonitrile (v/v). T=
293 K.
Chem. Eur. J. 2013, 19, 17494 – 17507
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
17501

L. Jullien et al.
Table 3. Action uncaging cross section of the caged model compounds NdiEt-cBA,
higher concentrations in Figures 2–4. In particular,
the variation of the absorbance at the maximum of
absorption could be also satisfactorily fitted with
mono- (for NdiEt-3CN-cBA and NdiEt-mcBA) and
bi- (for NdiEt-tcBA) exponential laws. After cali-
ð1Þ
NdiEt-3CN-cBA, NdiEt-tcBA, and NdiEt-mcBA for one-photon excitation at l
,
max
ð1Þ
eu(lmð1aÞx)Fuð1Þ(l ) (in mꢁ1 cmꢁ1) and the quantum yield of uncaging after one-photon
max
ð1Þ
excitation Fðu1Þ(l ), as extracted from the temporal dependence of the normalized
max
absorbance and the concentration in photoreleased benzoic acid as measured by
HPLC.[a]
bration of the light source, we extracted Fuð1Þ
=
ð1Þ
ð1Þ [b]
ð1Þ [b]
[c]
Compound [cBA]0
[mm]
l
eu(lðm1aÞx)Fuð1Þ(l
)
102Fðu1Þ(l
)
[BA]1 eu(lðm1aÞx1)[d]
max
max
max
(1.0ꢂ0.2)ꢃ10ꢁ4 and Fuð1Þ =(2.4ꢂ0.2)ꢃ10ꢁ4 for the
[nm] [mꢁ1 cmꢁ1
]
[mm]
[mmꢁ1 cmꢁ1
15ꢂ1
]
quantum yield of uncaging of NdiEt-3CN-cBA and
NdiEt-
3CN-cBA
NdiEt-
3CN-cBA
NdiEt-
tcBA
25
443 9ꢂ2
0.04ꢂ0.01
0.010ꢂ0.002
18ꢂ2
15ꢂ1
ð1Þ
NdiEt-mcBA with one-photon excitation at l
=
max
4
443 3ꢂ1
–
11ꢂ1
23ꢂ1
24ꢂ1
17ꢂ1
14ꢂ1
17ꢂ1
443 and 487 nm, respectively (see Table 3). Such
values did not significantly depart from the corre-
sponding values measured with 25 mm solutions. In
25
4
472 5600ꢂ600
472 160ꢂ90
487 11ꢂ1
487 8ꢂ1
22ꢂ1
NdiEt-
tcBA
0.5ꢂ0.3
–
contrast, we extracted Fðu1Þ =(5ꢂ3)ꢃ10ꢁ3 for the
ð1Þ
quantum yield of uncaging of NdiEt-tcBA at l
=
max
NdiEt-
mcBA
NdiEt-
mcBA
NdiEt-
cBA
25
4
0.034ꢂ0.004 26ꢂ1
472 nm (see Table 3). Whereas the latter value was
in excellent agreement with our result for the thio-
coumarin-caged inducer illuminated at 5 mm,[53] it
was about 40 times lower than the quantum yield of
uncaging of NdiEt-tcBA measured at 25 mm. How-
ever, we observed that the quantum yields of degra-
dation of NdiEt-tcOH were essentially similar at
concentrations of 4 and 25 mm (see Table 4).
0.024ꢂ0.002
0.21ꢂ0.02
–
–
5
385 51ꢂ5
[a] The asymptotic normalized absorbance at the absorption maximum after uncaging,
eu(lðm1aÞx1), and concentration in benzoic acid, [BA]1, extracted from the fits are
given in the Table. Initial concentration [cBA]0 in caged model compound cBA; Sol-
vent: acetonitrile/Tris pH 7.5 buffer (20 mm). See the main text and the Experimental
Section. [b] Average obtained from the temporal dependence of the concentration in
photoreleased benzoic acid measured by HPLC and the normalized absorbance.
[c] Extracted from the temporal dependence of the concentration in photoreleased
benzoic acid measured by HPLC. [d] Extracted from the temporal dependence of the
normalized absorbance.
Discussion
In this work, our goal was primarily to identify new
coumaACTHNURGTNErUNG inylmethyl moieties exhibiting redshifted
ure 22S, the Supporting Information). We did not observe
any temporal evolution of the solution absorbance after 2 h,
which suggested that NdiEt-mcOH was photostable. In fact,
during HPLC analysis of NdiEt-mcBA uncaging, we ob-
served the formation in significant amounts of a photoprod-
uct exhibiting the same retention time and molecular weight
as NdiEt-mcOH. In particular, the associated rate constant
of (0.020ꢂ0.001) minꢁ1 was in fair agreement with the
photo-consumption rate constant of NdiEt-mcBA. We thus
concluded that NdiEt-mcBA uncaging resulted in the libera-
tion of benzoic acid and a NdiEt-mcOH isomer, which we
could not identify by further NMR experiments (data not
shown).
The series of in vitro uncaging experiments was eventually
completed by reproducing the preceding series of illumina-
tion experiments on NdiEt-3CN-cBA, NdiEt-tcBA, and
NdiEt-mcBA at lower concentrations than 25 mm. Indeed,
we were surprised by the significant discrepancy that we ob-
served for the quantum yield of uncaging after one-photon
excitation for NdiEt-tcBA and a thiocoumarin-caged induc-
er, which involved a similar leaving group (a carbamate) at
the 4-coumarin benzylic position: whereas Fðu1Þ(472) was
about 0.2 for NdiEt-tcBA at 25 mm, we found Fðu1Þ(470) to
be about 5ꢃ10ꢁ3 for the thiocoumarin-caged inducer at
5 mm.[53] Figures 18S–20S (the Supporting Information) dis-
play the temporal evolution of the absorption spectra of
4 mm solutions in 20 mm pH 7.5 Tris buffer/acetonitrile (v/v)
upon illuminating at 443, 472, and 487 nm, respectively. We
observed a qualitatively similar behavior to that displayed at
absorption together with satisfactory photochemical proper-
ties to be used as caging groups. After the present study,
three caging moieties among the ten investigated can be
considered as relevant for that purpose.
The NdiEt-3CN-c caging moiety exhibits favorable ab-
sorption, emission, and photochemical properties. It signifi-
cantly absorbs at the border between UV and the visible
range. It is weakly fluorescent. In addition, its uncaging
quantum yield is about five times smaller than the one of
the reference NdiEt-c coumarinylmethyl caging moiety.
However, this uncaging quantum yield is notably weaker
than for the recently reported DEAC450 caging group,
which exhibits a similar substitution pattern to the NdiEt-
3CN-c one.[31] The NdiEt-tc caging moiety also exhibits at-
tractive absorption, emission, and photochemical properties.
It now satisfactorily absorbs in the blue-wavelength range. It
is also weakly fluorescent. In the investigated range of con-
centration, its uncaging quantum yield is always larger than
the one of the reference NdiEt-c coumarinylmethyl caging
moiety. Nevertheless, this uncaging quantum yield signifi-
cantly depends on concentration in the micromolar range.
This behavior could result from the involvement of a triplet
state, which is frequently observed in thiones.[74–76] Despite
our recent implementation in vivo,[53] this feature somehow
questions the robustness of the NdiEt-tc caging behavior in
unknown media. Eventually, the NdiEt-mc moiety can also
be considered for uncaging with cyan light; it exhibits the
most redshifted absorption in our series. In contrast, as
NdiEt-c, it is quite fluorescent, which could be a drawback
17502
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17494 – 17507

Coumarinylmethyl Caging Groups
FULL PAPER
cently attracted an increasing
interest both for chemical and
biological
applications.[41,80–86]
From the latter point of view,
the redshifted absorption of the
NdiEt-3CN-c, NdiEt-tc, and
NdiEt-mc
caging
moieties
would easily permit to selec-
tively achieve their photoactiva-
tion with respect to most classi-
cal caging groups absorbing in
the UV range upon adopting
a large enough excitation wave-
length. Moreover, the observa-
tion that NdiEt-3CN-c, NdiEt-
tc, and NdiEt-mc absorb light
about ten times less around
365 nm than at their maximal
absorption wavelength, would
be also appropriate to selective-
ly achieve the photoactivation
of most UV-absorbing caging
groups without significantly
photoactivating NdiEt-3CN-c,
NdiEt-tc, or NdiEt-mc. Indeed,
the action uncaging cross sec-
tion of the latter caging groups
around 365 nm is typically two
orders of magnitude smaller (1–
10mꢁ1 cmꢁ1 range) than the one
of most UV-absorbing caging
groups, which exhibit a value of
102–103 mꢁ1 cmꢁ1 at that wave-
length.[6] Hence, one could
easily tune either the power or
the illumination duration of
a 365 nm light source to selec-
tively uncage the caging groups
with maximal absorption in the
UV. In fact, we already exploit-
ed the favorable features of the
NdiEt-tc caging group in living
zebrafish embryos to precisely
perform chromatic orthogonal
Figure 3. Evolution of a solution of NdiEt-tcBA (25 mm) upon illuminating at 472 nm for various durations
t(I0(472))=6.1ꢃ10ꢁ7 Einminꢁ1; t(min)=0, 0.05, 0.17, 0.5, 1, 3, 8, 20, 35, 60, and 95). a) Temporal evolution of
the solution absorbance divided by its concentration to afford its molar absorption coefficient eðlð1ÞÞ (the ini-
tial absorption spectrum is drawn with a dotted line); b) Temporal dependence of the normalized absorption
~
coefficient at the absorption maximum e(472) ( ). The associated biexponential fit with Eq. (16) (see the Sup-
porting Information) is shown as a solid line. We extracted kðu1Þ =(9ꢂ1) minꢁ1 and kdð1Þ =(0.016ꢂ0.001) minꢁ1
,
as the fastest and slowest components of the biexponential fit, respectively; c) Temporal evolution of the con-
~
centration [BA] of benzoic acid extracted from the HPLC chromatogram ( ). The exponential fit with Eq. (6)
(see Supporting Information) shown as a solid line yields kðu1Þ =(9ꢂ1) minꢁ1 and [BA]1 =(22ꢂ1) mm; d) Tem-
poral evolution of the concentrations of the reactant [NdiEt-tcBA] (&) and the photoproducts [NdiEt-tcOH]
*
(
), and [NdiEt-cOH] (+) extracted from the HPLC chromatogram. The global biexponential fit with
Eqs. (13–15) (see the Supporting Information) shown as solid lines yields kðu1Þ =(9ꢂ3 min)ꢁ1 and kdð1Þ =(0.013ꢂ
0.005) minꢁ1 as the fastest and slowest components of the biexponential fit, respectively. Solvent: 20 mm
pH 7.5 Tris buffer/acetonitrile (v/v). T=293 K.
for several applications. Its uncaging quantum yield can be
compared to the NdiEt-3CN-c one; it is typically ten times
less efficient for uncaging than the NdiEt-c reference.
Hence, rather large light photon fluxes should be required
for uncaging. In contrast, such a low uncaging quantum
yield could be advantageous to perform chromatically or-
thogonal photoactivations (see below).
photoactivation of two biologically active species controlling
biological development with UV and blue-cyan light sources,
respectively.[53]
Conclusion
Beyond the evaluation of the preceding caging groups in
absolute terms, it is also appropriate to estimate their rele-
vance in the context of combinations with other caging
groups (or photoactive substrates) absorbing light at a differ-
ent wavelength. Indeed, the orthogonal photocontrol of two
different substrates with two different wavelengths has re-
Out of ten investigated coumarinylmethyl moieties, we have
identified three new caging groups that exhibit redshifted
absorption together with satisfactory photochemical proper-
ties with respect to the reference 7-diethylamino coumari-
nylmethyl caging group. The NdiEt-3CN-c, NdiEt-tc, and
NdiEt-mc caging moieties are synthetically easily accessible
Chem. Eur. J. 2013, 19, 17494 – 17507
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
17503

L. Jullien et al.
Table 4. Action degradation cross section of the intermediate resulting
from NdiEt-tcBA uncaging and of the alcohol NdiEt-tcOH for one-
photon excitation at lð1Þ , ed(lð1Þ )Fðd1Þ(lð1Þ ) (in mꢁ1 cmꢁ1) and quantum
and exhibit a significant action cross section for uncaging
with blue-cyan light. Moreover their low action cross section
for uncaging in the 350–400 nm wavelength range should
make them attractive to perform chromatic orthogonal un-
caging in combination with many presently available UV-ab-
sorbing caging groups, which do not exhibit any significant
absorption beyond 450 nm. Indeed, their uncaging efficiency
in this spectral domain remains significantly lower than the
action cross section for uncaging of even modestly efficient
UV-absorbing caging groups so as to avoid their photoacti-
vation when a properly tuned UV illumination is applied.
max
max
max
yield of degradation after one-photon excitation Fðd1Þ(lð1Þ ), as measured
max
from the temporal dependence of the normalized absorbance and of the
concentrations in reactants and photoproducts.[a]
ð1Þ
ð1Þ [b]
Compound
[tcS] ed(lð1Þ )Fdð1Þ(lð1Þ [b]
)
102Fd (l
)
Fðd1Þ(lð1Þ 1)[c]
max
max
max
[mm] [mꢁ1 cmꢁ1
]
[mMꢁm1 caxmꢁ1
]
NdiEt-tcBA 25
NdiEt-tcBA
NdiEt-tcOH 25
5.6ꢂ1.1
8ꢂ1
0.025ꢂ0.005
0.036ꢂ0.005
0.022ꢂ0.001
0ꢂ1
0ꢂ1
0ꢂ1
4
5.0ꢂ0.2
[a] The asymptotic normalized absorbance at the absorption maximum
after degradation, Fðd1Þ(lð1Þ 1), extracted from the fits is given in the
max
Table. Initial concentration [tcS] in substrate tcS (NdiEt-tcBA or NdiEt-
tcOH): 4 or 25 mm; Solvent: acetonitrile/Tris pH 7.5 buffer (20 mm). See
Text and Experimental Section. [b] Average from the values extracted
from the temporal dependence of the concentration in photoreleased
NdiEt-tcOH and NdiEt-cOH measured by HPLC, and of the normalized
absorbance. [c] Extracted from the temporal dependence of the normal-
ized absorbance.
Experimental Section
Reagents and solutions: Tris base (2-amino-2-(hydroxymethyl)-1,3-pro-
panediol 99.8+% A.C.S. reagent) and acetonitrile (spectrophotometric
grade) were from Sigma–Aldrich. All solutions were prepared using
water purified through a Direct-Q 5 (Millipore, Billerica, MA). All the
Figure 4. Evolution of a solution of NdiEt-mcBA (25 mm) upon illuminating at 487 nm for various durations t(I0(472))=6.1ꢃ10ꢁ7 Einminꢁ1; t(min)=0, 1,
10, 36, 60, 102, and 180). a) Temporal evolution of the solution absorbance divided by its concentration to afford its molar absorption coefficient eðlð1ÞÞ
(the initial absorption spectrum is drawn with a dotted line); b) Temporal dependence of the normalized absorption coefficient at the absorption maxi-
mum e(487) ( ). The associated exponential fit with Eq. (9) (see the Supporting Information) is shown as a solid line. We extracted kðu1Þ =(0.018ꢂ
~
0.001) minꢁ1 from the exponential fit; c) Temporal evolution of the concentration [BA] of benzoic acid extracted from the HPLC chromatogram ( ).
~
The exponential fit with Eq. (6) (see the Supporting Information) shown as a solid line yields kðu1Þ =(0.016ꢂ0.002) minꢁ1 and [BA]1 =(26ꢂ1) mm;
~
d) Temporal evolution of the concentration [mcP] of the photoproduct extracted from the HPLC chromatogram ( ). The exponential fit with Eq. (6)
(see the Supporting Information) shown as a solid line yields kðu1Þ =(0.02ꢂ0.01) minꢁ1. Solvent: 20 mm pH 7.5 Tris buffer/acetonitrile (v/v). T=293 K.
17504
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17494 – 17507

Coumarinylmethyl Caging Groups
FULL PAPER
experiments were performed at 293 K in a buffered solution obtained by
mixing in equal volumes acetonitrile and a 20 mm Tris base aqueous solu-
tion acidified to 7.5 (evaluated from glass electrode reading with a pH-
meter PHM 210; Radiometer Analytical calibrated at pH 4 and 7) with
1m chlorhydric acid.
Irradiation experiments with one-photon excitation: Two different proto-
cols have been used to perform the irradiations relying on one-photon
excitation.
Preliminary illuminations at 365 nm: One-photon illumination experi-
ments were performed at 208C, with a benchtop UV lamp[88] (365 nm, es-
sentially a 40 nm-wide, at half height, Gaussian distribution centered at
350 nm; Fisher Bioblock) delivering typical 1.0ꢃ10ꢁ2 Einsteinminꢁ1 mꢁ2
surfacic photon fluxes (measured using the reference PheP reported in
reference [88]) in the illuminated 25 mm sample (20 mL) contained in a cy-
lindrical glass dish: diameter 8.5 cm; height: 2.0 cm. Aliquots (3 mL)
were withdrawn after given durations of illumination for recording a UV/
Vis absorption spectrum and put back into the glass dish for further irra-
diation. Aliquots (200 mL) were independently extracted after various il-
lumination times and kept at ꢁ188C for HPLC analysis.
Illumination at the absorption maxima: The illumination experiments per-
formed at the maximum of absorption of the caged substrates selected
after the preliminary screening at 365 nm were performed with the 75 W
xenon lamp of the spectrofluorometer. We also notably used this configu-
ration with 365 nm excitation to perform control experiments.
Photophysical properties: The UV/Vis absorption spectra were recorded
on a Kontron Uvikon-940 spectrophotometer at 293 K. Molar absorption
coefficients were extracted while checking the validity of the Beer–Lam-
bert law.
Corrected emission spectra upon one-photon excitation were obtained
from a Photon Technology International QuantaMaster spectrofluorime-
ter. Solutions for emission measurements were adjusted to concentrations
such that the absorption maximum was around 0.15 at the excitation
wavelength. The overall emission quantum yields after one-photon exci-
ð1Þ
tation F values were calculated from the following relation:
em
ðlexe
Þ
ꢀ
ꢁ
2
1 ꢁ 10ꢁA
D
n
nref
ref
F
¼ Fð1Þ
ð1Þ
ref 1 ꢁ 10ꢁAðlexe
Þ
em
D
ref
We made two series of illumination experiments associated to different
concentrations of caged substrate:
in which the subscript “ref” stands for standard samples, Aref(lexe) is the
absorbance at the excitation wavelength, D is the integrated emission
Concentrated solutions (typically 25 mm as 0.4 mL samples in 1ꢃ0.2 cm2
quartz cuvettes under constant stirring; 1 cm illumination pathway) were
used to facilitate the HPLC analysis leading to identify and quantify the
photoproducts. One sample (0.4 mL) was used for each illumination
delay. After illumination, a sample (200 mL) was removed and kept at
ꢁ188C before performing the concentration-demanding analysis of ben-
zoic acid by HPLC with UV detection. The remaining mother solution
was diluted by a factor five and the resulting daughter solution was sub-
sequently used for analysis with HPLC/Mass spectrometry and UV/Vis
absorption;
spectrum, and n is the refractive index for the solvent. The uncertainty in
ð1Þ
the experimental value of F was estimated to be ꢂ10%. The standard
em
fluorophore for the quantum yields measurements was fluorescein in
ð1Þ
sodium hydroxide (0.1m) with F =0.92.[87]
ref
The quartz cuvettes (Hellma) were either 1ꢃ1 cm (3 mL samples) or 1ꢃ
0.2 cm (0.4 mL samples). The holders were thermostated using circulating
baths (Polystat 34-R2, Fisher Bioblock Scientific, Illkirch, France) and
the temperature was directly measured in the cuvettes by using a type K
thermocouple connected to a ST-610B digital pyrometer (Stafford Instru-
ments, Stafford, UK).
More diluted solutions (typically 5 mm as 3 mL samples in 1ꢃ1 cm2 quartz
cuvettes under constant stirring; absorbance at the excitation wavelength
lower than 0.15) were used to monitor the absorbance of the sample
HPLC coupled to mass spectrometry: High-pressure liquid chromatogra-
phy was carried out by using an Accela System liquid chromatograph
(Thermo Finnigan, Les Ulis, France) equipped with a Hypersil Gold
during illumination.
ð1Þ
Light intensity at the excitation wavelength F was systematically cali-
exc
column (1.9 mmꢃ2.1ꢃ50 mm) connected to
a
Thermo-Finnigan TSQ
brated before each experiment using a procedure relying on analyzing
the intensity IDCM of fluorescence emission of a 4-dicyanomethylene-2-
methyl-6-p-dimethylamino-styryl-4H-pyrane (DCM) solution in absolute
Quantum Discovery Max triple quadrupole mass spectrometer. The illu-
minated sample solution (5 mL) was injected in the chromatographic
column. The analyte eluted from the column with a mobile phase com-
posed of two solvents A (water containing formic acid at 0.05% v/v) and
B (acetonitrile). A gradient was used to optimize the separation of the
analytes. Initially, the column was equilibrated at 200 mLminꢁ1 with
a mobile phase consisting of 90% A and 10% B. One minute after the
injection, the proportion of B was linearly increased to 85% within 4 min
and at that position for two additional minutes. After this step, composi-
tion of the mobile phase was set to initial and the column was equilibrat-
ed for 3 min prior to next injection. After separation, the analytes
(except benzoic acid) were introduced in the mass spectrometer through
a heated electrospray ionization source (508C, 4000 V) operating in the
positive mode. The temperature of the capillary transfer was set at
2708C. Nitrogen was employed as nebulizing (35 psi) and auxiliary gas
(30 arbitrary units). Argon was used as collision gas (1.0 milliTorr in Q2).
Again peak areas were converted into concentrations after preliminary
calibrations. Instrument control and data collection were handled by
a computer equipped with Xcalibur software (version 2.0).
ethanol upon exciting at two different wavelengths. The fluorescence
ð1Þ
emission IDCM(Fðe1xÞc) associated to I0(F ) was first recorded upon excit-
exc
ing at Fðe1xÞc, in which DCM absorbs light with a molar absorption coeffi-
ð1Þ
cient eDCM(F ). Then, the corresponding fluorescence emission
exc
IDCM(365) was recorded upon exciting at 365 nm with light intensity
I0(365) (molar absorption coefficient eDCM(365)). I0(365) was subsequent-
ly extracted from determining the kinetics of photoconversion at 365 nm
of the a-(4-dimethylaminophenyl)-N-phenylnitrone into 3-(4-dimethyla-
minophenyl)-2-phenyloxaziridine in absolute ethanol as described in ref-
ð1Þ
erence [89]. Eventually, the light intensity I0(F ) was computed from
exc
ð1Þ
eDCM ð365ÞI
l
DCM
exc
I0ð365Þ. During the present series of experiments, typical
eDCM ðlð1Þ ÞIDCM ð365Þ
exc
overall photon fluxes at the sample were in the 10ꢁ6–10ꢁ7 Einsꢁ1 range.
Acknowledgements
In the case of benzoic acid, the HPLC system consisted of a Dionex
This work has been supported by the ANR (PCV 2008, Proteophane)
and the Ministꢄre de la Recherche (for a fellowship to L.F.). The authors
thank Dr. V. Jullien and Prof. G. Pons at Service de Pharmacologie Clini-
que, Saint Vincent de Paul Hospital Paris, for access to their HPLC-MS
instrument. The authors declare no competing financial interests.
HPLC P680A LPG-4 pump,
a Dionex ASI-100 autosampler, and
a Dionex UVD 340U UV detector or an Agilent 1260 Chromatographic
system. Instrument control and data collection were handled by a comput-
er equipped with Chromeleon software. Samples (10 or 20 mL) were in-
jected on Kinetex C18 columns (50ꢃ2.1 mm or 4.6ꢃ100 mm, 2.6 mm, Phe-
nomenex, Torrance, CA), and benzoic acid eluted out of the column in
isocratic mode with a mixture of water containing 0.1% formic acid and
acetonitrile (65–35 v/v) at a flow rate of 0.5 or 1 mLminꢁ1. Benzoic acid
was detected by UV absorption at 227 nm and the conversion of the peak
area to the solution concentration was achieved from a preliminary cali-
bration.
[1] V. N. Rajasekharan Pillai, Synthesis 1980, 1–26.
[2] C. G. Bochet, J. Chem. Soc. Perkin Trans. 1 2002, 125–142.
[3] A. P. Pelliccioli, J. Wirz, Photochem. Photobiol. Sci. 2002, 1, 441–
458.
Chem. Eur. J. 2013, 19, 17494 – 17507
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
17505

L. Jullien et al.
[4] Dynamic Studies in Biology (Eds.: M. Goeldner, R. Givens), Wiley-
VCH, Weinheim, 2005.
[5] G. Mayer, A. Heckel, Angew. Chem. 2006, 118, 5020–5042; Angew.
Chem. Int. Ed. 2006, 45, 4900–4921.
[6] G. C. R. Ellis-Davies, Nat. Methods 2007, 4, 619–628.
[7] D. D. Young, A. Deiters, Org. Biomol. Chem. 2007, 5, 999–1005.
[8] H.-M. Lee, D. R. Larson, D. S. Lawrence, ACS Chem. Biol. 2009, 4,
409–427.
[9] A. Specht, F. Bolze, Z. Omran, J.-F. Nicoud, M. Goeldner, HFSP J.
2009, 3, 255–264.
[10] H. Yu, J. Li, D. Wu, Z. Qiu, Y. Zhang, Chem. Soc. Rev. 2010, 39,
464–473.
[11] A. Deiters, ChemBioChem 2009, 11, 47–53.
[12] C. Brieke, F. Rohrbach, A. Gottschalk, G. Mayer, A. Heckel,
Angew. Chem. Int. Ed. 2012, 51, 8446–8476.
[13] P. Klꢅn, T. Solomek, C. G. Bochet, A. Blanc, R. Givens, M. Rubina,
V. Popik, A. Kostikov, J. Wirz, Chem. Rev. 2013, 113, 119–191.
[14] J. Baier, T. Maisch, M. Maier, E. Engel, M. Landthaler, W. Baꢆmler,
Biophys. J. 2006, 91, 1452–1459.
[40] A. Z. Suzuki, T. Watanabe, M. Kawamoto, K. Nishiyama, H. Yama-
shita, M. Ishii, M. Iwamura, T. Furuta, Org. Lett. 2003, 5, 4867–
4870.
[41] N. Kotzur, B. Briand, M. Beyermann, V. Hagen, J. Am. Chem. Soc.
2009, 131, 16927–16931.
[42] V. Hagen, F. Kilic, J. Schaal, B. Dekowski, R. Schmidt, N. Kotzur, J.
Org. Chem. 2010, 75, 2790–2792.
[43] V. Hagen, J. Bendig, S. Frings, B. Wiesner, B. Schade, S. Helm, D.
Lorenz, U. B. Kaupp, J. Photochem. Photobiol. B 1999, 53, 91–102.
[44] M. Skwarczynski, M. Noguchi, S. Hirota, Y. Sohma, T. Kimura, Y.
Hayashi, Y. Kiso, Bioorg. Med. Chem. Lett. 2006, 16, 4492–4496.
[45] T. Eckardt, V. Hagen, B. Schade, R. Schmidt, C. Schweitzer, J.
Bendig, J. Org. Chem. 2002, 67, 703–710.
[46] B. Schade, V. Hagen, R. Schmidt, R. Herbrich, E. Krause, T. Eck-
ardt, J. Bendig, J. Org. Chem. 2002, 67, 703–710.
[47] R. Schmidt, D. Geissler, V. Hagen, J. Bendig, J. Phys. Chem. A 2005,
109, 5000–5004.
[48] V. Hagen, S. Frings, B. Wiesner, S. Helm, U. B. Kaupp, J. Bendig,
ChemBioChem 2003, 4, 434–442.
[49] T. Furuta, H. Takeuchi, M. Isozaki, Y. Takahashi, M. Kanehara, M.
Sugimoto, T. Watanabe, K. Noguchi, T. M. Dore, T. Kurahashi, M.
Iwamura, R. Y. Tsien, ChemBioChem 2004, 5, 1119–1128.
[50] V. R. Shembekar, Y. Chen, B. K. Carpenter, G. P. Hess, Biochemistry
2005, 44, 7107–7114.
[51] C. Bao, G. Fan, Q. Lin, B. Li, S. Cheng, Q. Huang, L. Zhu, Org.
Lett. 2012, 14, 572–575.
[52] A. S. C. Fonseca, A. M. S. Soares, M. S. T. Goncalves, S. P. G. Costa,
Tetrahedron 2012, 68, 7892–7900.
[53] L. Fournier, C. Gauron, L. Xu, I. Aujard, T. Le Saux, N. Gagey-Eil-
stein, S. Maurin, S. Dubruille, J.-B. Baudin, D. Bensimon, M. Volo-
vitch, S. Vriz, L. Jullien, ACS Chem. Biol., 2013, 8, 1528–1536.
[54] M. A. Kirpichꢈnok, V. M. Baukulev, L. A. Karandashova, Chem.
Heterocycl. Compd. 1991, 27, 1193–1199.
[55] I. I. Tkach, A. V. Reznichenko, E. A. Lukyanets, Chem. Heterocycl.
Compd. 1992, 28, 872–880.
[56] J. Yu, Y. Shirota, Chem. Lett. 2002, 31, 984–985.
[57] K. Komatsu, Y. Urano, H. Kojima, T. Nagano, J. Am. Chem. Soc.
2007, 129, 13447–13454.
[58] M. Wilcox, R. W. Viola, K. W. Johnson, A. P. Billington, B. K. Car-
penter, J. A. McCray, A. P. Guzikowski, G. P. Hess, J. Org. Chem.
1990, 55, 1585–1589.
[59] B. Schade, V. Hagen, R. Schmidt. R. Herbrich, J. Bendig, J. Org.
Chem. 1999, 64, 9109–9117.
[60] A. M. Piloto, A. S. C. Fonseca, S. P. G. Costa, M. S. T. GonÅalves,
Tetrahedron 2006, 62, 9258–9276.
[61] S. Scheibye, J. Kristensen, S.-O. Lawesson, Tetrahedron 1979, 35,
1339–1343.
[62] J. N. Gadre, A. A. Audi, N. P. Karambelkar, Ind. J. Chem. 1996, 35B,
60–62.
[63] M. Yokoyama, M. Irie, K. Sujino, T. Kagemoto, H. Togo, M. Funa-
bashi, J. Chem. Soc. Perkin Trans. 1 1992, 2127–2134.
[15] G. Kick, G. Messer, G. Plewig, P. Kind, A. E. Goetz, Br. J. Cancer
1996, 74, 30–36.
[16] L.-O. Klotz, C. Pellieux, K. Briviba, C. Pierlot, J.-M. Aubry, H. Sies,
Eur. J. Biochem. 1999, 260, 917–922.
[17] K. Wertz, N. Seifert, P. Buchwald Hunziker, G. Riss, A. Wyss, W.
Hunziker, R. Goralczyk, Pure Appl. Chem. 2006, 78, 1539–1550.
[18] Y. Chen, M. G. Steinmetz, Org. Lett. 2005, 7, 3729–3732.
[19] Y. Chen, M. G. Steinmetz, J. Org. Chem. 2006, 71, 6053–6060.
[20] P. Sebej, J. Wintner, P. Mꢆller, T. Slanina, J. A. Anshori, L. A. P.
Antony, P. Klan, J. Wirz, J. Org. Chem. 2013, 78, 1833–1843.
[21] L. Zayat, C. Calero, P. Albores, L. Baraldo, R. Etchenique, J. Am.
Chem. Soc. 2003, 125, 882–883.
[22] L. Zayat, M. Salierno, R. Etchenique, Inorg. Chem. 2006, 45, 1728–
1731.
[23] L. Zayat, M. G. Noval, J. Campi, C. I. Calero, D. J. Calvo, R. Etche-
nique, ChemBioChem 2007, 8, 2035–2038.
[24] M. Salierno, C. Fameli, R. Etchenique, Eur. J. Inorg. Chem. 2008,
1125–1128.
[25] E. M. Rial Verde, L. Zayat, R. Etchenique, R. Yuste, Front. Neural
Circuits 2008, 2, 1–7.
[26] K. Lee, D. E. Falvey, J. Am. Chem. Soc. 2000, 122, 9361–9366.
[27] C. Sundararajan, D. E. Falvey, J. Am. Chem. Soc. 2005, 127, 8000–
8001.
[28] J. B. Borak, S. Lꢇpez-Sola, D. E. Falvey, Org. Lett. 2008, 10, 457–
460.
[29] S. Gug, F. Bolze, A. Specht, C. Bourgogne, M. Goeldner, J.-F.
Nicoud, Angew. Chem. 2008, 120, 9667–9671; Angew. Chem. Int.
Ed. 2008, 47, 9525–9529.
[30] L. Donato, A. Mourot, C. M. Davenport, C. Herbivo, D. Warther, J.
Lꢀonard, F. Bolze, J.-F. Nicoud, R. H. Kramer, M. Goeldner, A.
Specht, Angew. Chem. Int. Ed. 2012, 51, 1840–1843.
[31] J. P. Olson, H.-B. Kwon, K. T. Takasaki, Q. Chiayu, C. Q. Chiu, M. J.
Higley, B. L. Sabatini, G. C. R. Ellis-Davies, J. Am. Chem. Soc. 2013,
135, 5954–5957.
´
[64] W. Przychodzen, Eur. J. Org. Chem. 2005, 2002–2014. W. Przychod-
´
zen, J. Chem. Soc. Perkin Trans. 1 1992, 2127–2134.
[65] P. N. Bhargava, S. Prakash, J. Indian Chem. Soc. 1979, 56, 937–938.
[66] P. Reynaud, J.-D. Brion, G. Mꢀnard, Bull. Soc. Chim. Fr. 1976, 1–2,
301–306.
[67] Ya. L. Garazd, M. M. Garazd, V. A. Turov, V. P. Khilya, Chem. Nat.
Compd. 2006, 42, 652–655.
[32] I. Aujard, C. Benbrahim, M. Gouget, O. Ruel, J. B. Baudin, P.
Neveu, L. Jullien, Chem. Eur. J. 2006, 12, 6865–6879.
[33] R. S. Givens, B. Matuszewski, J. Am. Chem. Soc. 1984, 106, 6860–
6861.
[34] T. Furuta, H. Torigai, T. Osawa, M. Iwamura, J. Org. Chem. 1995,
60, 3953–3956.
[68] P. Stegmaier, J. M. Alonso, A. del Campo, Langmuir 2008, 24,
11872–11879.
[35] T. Furuta, M. Iwamura, Methods Enzymol. 1998, 291, 50–63.
[36] R. S. Givens, M. Rubina, J. Wirz, Photochem. Photobiol. Sci. 2012,
11, 472–488.
[69] K. Ito, J. Maruyama, Chem. Pharm. Bull. 1983, 31, 3014–3023.
[70] H. Abdel-Ghany, H. M. Moustafa, A. Khodairy, Synth. Commun.
1998, 28, 3431–3441.
[37] T. Furuta, S. S.-H. Wang, J. L. Dantzker, T. M. Dore, W. J. Bybee,
E. M. Callaway, W. Denk, R. Y. Tsien, Proc. Natl. Acad. Sci. USA
1999, 96, 1193–1200.
[38] R. O. Schçnleber, J. Bendig, V. Hagen, B. Giese, Bioorg. Med.
Chem. 2002, 10, 97–101.
[71] N. Latif, I. Fathy, Can. J. Chem. 1966, 44, 1075–1081.
[72] S. S. Lele, M. G. Patel, S. Sethna, J. Org. Chem. 1962, 27, 637–639.
[73] K. H. Al-Zghoul, K. S. M. Salih, M. T. Ayoub, M. S. Mubarak, Het-
erocycles 2005, 65, 2937–2947.
[74] R. P. Steer, V. Ramamurthy, Acc. Chem. Res. 1988, 21, 380–386.
[75] A. Maciejewski, R. P. Steer, Chem. Rev. 1993, 93, 67–98.
[39] W. Lin, D. S. Lawrence, J. Org. Chem. 2002, 67, 2723–2726.
17506
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17494 – 17507

Coumarinylmethyl Caging Groups
FULL PAPER
[76] a) K. Bhattacharyya, P. K. Das, V. Ramamurthy, V. P. Rao, J. Chem.
Soc. Faraday Trans. 2 1986, 82, 135–147; b) S. Devanathan, V. Ram-
amurthy, J. Org. Chem. 1988, 53, 741–744; c) R. S. Becker, S. Chak-
ravorti, C. A. Gartner, M. de Graca Miguel, J. Chem. Soc. Faraday
Trans. 1993, 89, 1007–1019; d) G. Burdzinskia, G. Buntinx, O.
Poizat, C. Lapouge, J. Mol. Struct. 2005, 735–736, 115–122.
[77] We cannot exclude that thermal stability could be substrate depen-
dent. Hence, Ellis–Davies caged glutamate with the NdiEt-tc caging
moiety to yield NdiEt-tc-Glu (personal communication). In contrast
to our observations with NdiEt-tc-BA, as well as with a NdiEt-tc-
caged inducer reported in Reference [53], he observed that NdiEt-
tc-Glu was rapidly converted in 1 h to NdiEt-c-Glu in methanol
(MeOD) or 1% methanol in HEPES pH 7.4.
[78] In fact, we observed that UV illumination caused a slow increase of
absorbance. However, the corresponding photochemical reaction
did not lead to the release of benzoic acid.
[79] In the absence of any illumination, we checked that the NdiEt-tcOH
solution was thermally stable at the timescale of the illumination ex-
periment (typically, 3 h).
[83] M. Priestman, L. Sun, D. S. Lawrence, ACS Chem. Biol. 2011, 6,
377–384.
[84] V. San Miguel, C. G. Bochet, A. del Campo, J. Am. Chem. Soc. 2011,
133, 5380–5388.
[85] C. Menge, A. Heckel, Org. Lett. 2011, 13, 4620–4623.
[86] M. N. Stanton-Humphreys, R. D. T. Taylor, C. McDougall, M. L.
Hart, C. Tom, A. Brown, N. J. Emptage, S. J. Conway, Chem.
Commun. 2012, 48, 657–659.
[87] J. N. Demꢅs, G. A. Crosby, J. Phys. Chem. 1971, 75, 991–1024.
[88] a) P. Neveu, I. Aujard, C. Benbrahim, T. Le Saux, J.-F. Allemand, S.
Vriz, D. Bensimon, L. Jullien, Angew. Chem. 2008, 120, 3804–3806;
Angew. Chem. Int. Ed. 2008, 47, 3744–3746; b) P. Neveu, D. Sinha,
P. Kettunen, S. Vriz, L. Jullien, D. Bensimon in Nobel Volume on
Single Molecule Spectroscopy in Chemistry, Physics and Biology,
Springer, Heidelberg, 2009; c) D. K. Sinha, P. Neveu, N. Gagey, I.
Aujard, C. Benbrahim-Bouzidi, T. Le Saux, C. Rampon, C. Gauron,
B. Goetz, S. Dubruille, M. Baaden, M. Volovitch, D. Bensimon, S.
Vriz, L. Jullien, ChemBioChem 2010, 11, 653–663; d) D. K. Sinha, P.
Neveu, N. Gagey, I. Aujard, T. Le Saux, C. Rampon, C. Gauron, K.
Kawakami, C. Leucht, L. Bally-Cuif, M. Volovitch, D. Bensimon, L.
Jullien, S. Vriz, Zebrafish 2010, 7, 199–204.
[80] C. G. Bochet, Angew. Chem. 2001, 113, 2140–2142; Angew. Chem.
Int. Ed. 2001, 40, 2071–2073.
[81] S. Kantevari, M. Matsuzaki, Y. Kanemoto, H. Kasai, G. C. R. Ellis-
Davies, Nat. Methods 2010, 7, 123–125.
[82] D. Goguen, A. Aemissegger, B. Imperiali, J. Am. Chem. Soc. 2011,
133, 11038–11041.
[89] P. F. Wang, L. Jullien, B. Valeur, J.-S. Filhol, J. Canceill, J.-M. Lehn,
New J. Chem. 1996, 20, 895–907.
Received: July 6, 2013
Revised: August 16, 2013
Published online: November 11, 2013
Chem. Eur. J. 2013, 19, 17494 – 17507
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
17507