Understanding the relationship between photolysis efficiency and metal binding using argencast photocages

Article abstract of DOI:10.1111/j.1751-1097.2012.01136.x

ArgenCast-1 (1), a photocage for silver utilizing acyclic polythioether 3,6,12,15-tetrathia-9-azaheptadecane receptor and 4,5-dimethoxy-2-nitrobenzyl (DMNB) chromophore has been prepared using trimethylsilyl trifluoromethanesulfonate-assisted electrophilic aromatic substitution. Metal binding studies with ArgenCast-1 reveal interactions with Ag⁺, Hg²⁺ and Cu⁺, but only Ag⁺ coordinates in both aqueous and organic solvents. The uncaging mechanism of ArgenCast-1 metal complex involves a photoreaction that converts the nitrobenzydrol into the electron withdrawing nitrosobenzophenone that participates in a resonance interaction with a metal-bound aniline nitrogen atom. The structural change following photolysis decreases availability of the nitrogen lone pair for Ag⁺ coordination, but strong interactions between Ag⁺ and the thioether ligands mitigates metal ion release. A resonance interaction with a key aci-nitro intermediate reduces the photolysis quantum yield of ArgenCast-1, so several naphthyl-based nitrobenzyl groups were screened as alternatives to DMNB. The naphthyl Ag⁺ photocages, ArgenCast-2 and -3, exhibit nearly identical quantum yield to ArgenCast-1 owing to the dominance of resonance between aci-nitro intermediate and the aniline nitrogen atom.

Full text of DOI:10.1111/j.1751-1097.2012.01136.x

Photochemistry and Photobiology, 2012, 88: 844–850
Understanding the Relationship Between Photolysis Efficiency and Metal
Binding Using ArgenCast Photocages
Hannah W. Mbatia1, Daniel P. Kennedy1 and Shawn C. Burdette*2
1Department of Chemistry, University of Connecticut, Storrs, CT
2Department of Chemistry and Biochemistry, Worcester Polytechnic Institute, Worcester, MA
Received 16 December 2011, accepted 2 March 2012, DOI: 10.1111/j.1751-1097.2012.01136.x
developed several tools for controlling the release of Mg2+ (4),
Fe3+ (5,6), Hg2+ (7) and Zn2+ (8) that utilize analogous
ABSTRACT
ArgenCast-1 (1), a photocage for silver utilizing acyclic poly-
thioether 3,6,12,15-tetrathia-9-azaheptadecane receptor and
4,5-dimethoxy-2-nitrobenzyl (DMNB) chromophore has been
prepared using trimethylsilyl trifluoromethanesulfonate-assisted
electrophilic aromatic substitution. Metal binding studies with
ArgenCast-1 reveal interactions with Ag+, Hg2+ and Cu+, but
only Ag+ coordinates in both aqueous and organic solvents. The
uncaging mechanism of ArgenCast-1 metal complex involves a
photoreaction that converts the nitrobenzydrol into the electron
withdrawing nitrosobenzophenone that participates in a reso-
nance interaction with a metal-bound aniline nitrogen atom. The
structural change following photolysis decreases availability of
the nitrogen lone pair for Ag+ coordination, but strong
interactions between Ag+ and the thioether ligands mitigates
metal ion release. A resonance interaction with a key aci-nitro
intermediate reduces the photolysis quantum yield of ArgenCast-
1, so several naphthyl-based nitrobenzyl groups were screened as
alternatives to DMNB. The naphthyl Ag+ photocages, Argen-
Cast-2 and -3, exhibit nearly identical quantum yield to
ArgenCast-1 owing to the dominance of resonance between
aci-nitro intermediate and the aniline nitrogen atom.
uncaging strategies. We have dubbed the class of photocaged
complexes based on the nitrobenzhydrol to benzophenone
conversion Cast complexes, because the metal ion is cast off
after photolysis. To date, adapting the Cast uncaging strategy
for use with soft metal ions has been problematic, so we were
interested in further elucidating the scope of applications for
this family of photocaged complexes. Soft metal ions are
involved in neurotransmission like Cu+ (9), neural toxicity like
Hg2+ (10) and therapeutics like Ag+ (11), so there is a wide
scope of possible uses for the resulting photocaged complexes.
MATERIALS AND METHODS
General synthetic procedures. All the materials listed below were of a
research grade or a spectro-grade in the highest purity commercially
available from Acros Organics or TCI America. Dichloromethane
(CH2Cl2), toluene (C6H5CH3) and tetrahydrofuran (THF) were
sparged with argon and dried by passage through a Seca Solvent
Purification System. All chromatography and TLC were performed on
silica (230–400 mesh) from Silicycle unless otherwise specified. Acti-
vated basic alumina (ca 150 mesh) was obtained from Acros Organics.
TLCs were developed with mixtures of ethyl acetate (EtOAc) ⁄ hexanes
or CH2Cl2 ⁄ methanol (CH3OH) unless otherwise specified and were
visualized with 254 and 365 nm light and I2 or Br2 vapor. 1H and 13C
NMR spectra were recorded using a Bruker 400 MHz NMR instru-
ment and chemical shifts are reported in ppm on the d scale relative to
tetramethylsilane. IR spectra were recorded on a Nicolet 205 FT-IR
instrument and the samples were prepared as KBr pellets. High-
resolution mass spectra were recorded at the University of Connecticut
mass spectrometry facility using a micromass Q-TOF-2TM mass
spectrometer operating in positive ion mode. The instrument was
INTRODUCTION
Initially, photocages were developed to interrogate the biolog-
ical activity of adenosine-5¢-triphosphate (ATP) using light (1).
Photocages allow spatial- and temporal-controlled introduction
of analytes by blocking biological function until a photoreaction
initiates activator release (2). The first photocaged complexes
were developed to study Ca2+-based signal transduction path-
ways by Tsien and Zucker (3). The original Ca2+ photocaged
complexes integrate a nitrobenzhydrol photoactive group into
the aniline-derived chelator 1,2-bis(o-aminophenoxy)ethane-N,
N,N¢,N¢-tetraacetic acid (BAPTA). When irradiated, the benz-
hydrol is converted into a benzophenone. The resulting reso-
nance interaction between the Ca2+-bound aniline lone pair
and the benzophenone carbonyl decreases the chelators’ affin-
ity for the metal ion, which shifts the equilibrium toward
calibrated with Glu-fibrinopeptide
B
10 pmol lL)1 using 50:50
CH3CN ⁄ H2O with 0.1% acetic acid (AcOH). N,N-bis(2-((2-(ethyl-
thio)ethyl)thio)ethyl)aniline (1) (12) and 3-nitro-2-naphthaldehyde (13)
were synthesized as described.
(4-(Bis(2-((2-(Ethylthio)ethyl)thio)ethyl)amino)phenyl)(4,5-di-
methoxy-2-nitrophenyl)methanol (2, ArgenCast-1). To a solution of
1
(390 mg, 0.65 mmol) and ortho-nitroveratraldehyde (161 mg,
0.650 mmol) in CH2Cl2 (3.0 mL) was added 2,6-dimethypyridine
(0.316 mg, 1.51 mmol) and trimethylsilyl trifluoromethanesulfonate
(TMSOTf, 0.289 g, 1.30 mmol). The resulting dark blue ⁄ green colored
solution was stirred for 16 h at 23ꢀC. The solution was diluted with
50 mL of CH2Cl2, washed with brine (50 mL), dried over MgSO4,
filtered and evaporated to dryness. Flash chromatography on silica
with a solvent gradient (20 fi 30% EtOAc in petroleum ether)
afforded 0.109 g of TMS-ether protected ArgenCast as a yellow oil
(Rf = 0.39. 3:1, EtOAc ⁄ petroleum ether). The TMS-ether was dis-
solved in CH2Cl2 (3.0 mL) and 1 M TBAF in THF was added
(1.2 mL). After 5 min of stirring, the solvent was removed. Flash
chromatography (silica, 1:1 EtOAc ⁄ petroleum ether) furnished 2 as
unbound Ca2+
.
In addition to the various photocaged complexes that have
elucidated the roles of Ca2+ in neurotransmission, we have
*Corresponding author email: scburdette@wpi.edu (Shawn C. Burdette)
ꢁ 2012 Wiley Periodicals, Inc.
Photochemistry and Photobiology ꢁ 2012 The American Society of Photobiology 0031-8655/12
844

Photochemistry and Photobiology, 2012, 88 845
yellow oil (0.096 g, 16%). Rf = 0.47 (1:1 EtOAc ⁄ petroleum ether).
1H-NMR (400 MHz, CDCl3) d 7.63 (s, 1 H,), 7.45 (s, 1 H), 7.19 (dd,
J = 8 Hz, 2 H), 6.60 (dd, J = 8 Hz, 2 H), 6.48 (s, 1 H), 4.01 (s, 3 H),
3.96 (s, 3 H), 3.53 (m, 4 H), 2.47 (m, 12 H), 2.60 (m, 4 H), 1.28 (m, 6
H). 13C-NMR (100 MHz) d 153.59, 147.93, 146.49, 140.11, 134.87,
130.48, 128.81, 111.87, 110.26, 108.34, 71.53, 56.68, 56.58, 51.80, 32.71,
32.06, 29.65, 26.33, 15.03 ppm. IR (neat) 3434.6, 2963.0, 2925.4,
1611.2, 1517.7, 1330.6, 1271.8, 1062.5. HRMS (+ES1), calculated for
MH+ 601.1898, observed 601.1824.
(4-(Bis(2-((2-(ethylthio)ethyl)thio)ethyl)amino)phenyl)(4,5-
dimethoxy-2-nitrosophenyl)methanone (3, ArgenUnc-1). A solution of
2 (0.076 g, 0.127 mmol) in CH3CN (4.0 mL) sealed in a quartz cuvette
was irradiated with a 1000 W source for 5 h. The solvent was removed
under reduced pressure, and flash chromatography (silica, 1:9
perchlorate salts. Cu+ solutions was introduced stabilized in thiourea
(0.2 mM). Absorption spectra were recorded on a Cary 50 UV–visible
spectrophotometer under the control of a Pentium IV-based PC
running the manufacturer supplied software package. Spectra were
routinely acquired at 25ꢀC, in 1-cm path length quartz cuvettes with a
total volume of 3.0 mL. Analytical photolysis was performed using a
8 W photoreactor at 350 nm. Photolysis reactions were conducted at
25ꢀC, in 1 cm quartz cuvette with
a total volume of 3.0 mL.
Preparative scale photolysis was carried out using a 1000 W Xe small
arc lamp. Solutions of 2 in CH3CN (3.0 mL) were irradiated for 5 h in
a 1 cm quartz cuvette and the conversion checked periodically with
TLC. Quantum yields of photolysis were measured using reported
procedures at equimolar concentrations of metal ions (7).
Binding constants. Stock solutions of ArgenCast-1 and -1, and the
metalions of Cu+, Hg2+ and Ag+ werepreparedin mM concentration in
spectrophotometric grade CH3CN. A 25 lM solution of the ligand was
prepared in 3000 lL CH3CN and titrated in triplicate with each of the
metal stock solutions. Absorbance spectra were corrected for dilution by
multiplying the measure absorbance by the following formula:
EtOAc ⁄ CH2Cl2) furnished
3 (0.039 g, 54.0%) as a yellow oil.
Rf = 0.58 (1:1 EtOAc ⁄ petroleum ether). 1H-NMR (400 MHz, CDCl3)
d 7.78 (dd, J = 8 Hz, 2 H), 7.14 (s, 1 H), 6.61 (dd, J = 8 Hz, 2 H),
6.36 (s, 1 H), 4.05 (s, 3 H), 3.93 (s, 3 H), 3.63 (m, 4 H), 2.79–2.73 (m, 12
H), 2.60–2.54 (m, 4 H), 1.28–1.25 (m, 6 H). 13C-NMR (100 MHz) d
193.9, 160.0, 156.0, 151.2, 139.9, 133.0, 127.7, 110.8, 109.8, 92.0, 57.0,
56.4, 51.7, 32.8, 32.1, 29.6, 26.4, 15.0 ppm. IR (neat) 2913, 2848, 1632,
1562, 1342, 1257, 1043 cm)1. HRMS (+ESI) calculated for MH+
583.1793, observed 583.1764.
V0 ꢀ Vadd
ð1Þ
V0
where V0 is the initial volume and Vadd is the added volume of the
titrant. The conditional dissociation constant (Kd’) was calculated by
fitting data at ca 270 and 350 nm using the XLfit (14) Richards model.
The three values of Kd’ found for each cation were averaged and
standard deviation was calculated. Using 25 lM of ArgenUncast-1, the
procedure was repeated and the conditional dissociation constant was
calculated as described elsewhere at ca 350 nm.
(4-(Bis(2-((2-(Ethylthio)ethyl)thio)ethyl)amino)phenyl)(3-
nitronaphthalen-2-yl)methanol(5,ArgenCast-2). Toasolutionof1(0.186 g,
0.479 mmol) and 3-nitro-2-naphthaldehyde (4, 0.106 g, 0.527 mmol) in
CH2Cl2 (3.0 mL) was added lutidine (0.077 g, 0.718 mmol) and
TMSOTf (0.138 g, 0.623 mmol). The resulting dark blue ⁄ green colored
solutionwasstirredfor16 hat23ꢀC.Thesolutionwasdilutedwith50 mL
of CH2Cl2, washed with brine (50 mL), dried over MgSO4, filtered and
the solvent was removed. The resulting crude TMS-ether was dissolved
in CH3CN (3.0 mL) to which 18-crown-6 (0.379 g, 1.43 mmol) and
KF (0.278 g, 4.79 mmol), dissolved in minimal amount of water, was
added. The resulting reaction mixture was stirred for 12 h at 23ꢀC. The
reaction mixture was diluted with water (50 mL) and extracted with
CH2Cl2. The organic extract was washed with aqueous KCl and dried
over MgSO4. The solvent was removed under vacuum and flash
chromatography (silica, 9:1 CH2Cl2 ⁄ EtOAc) furnished 5 an orange oil
(0.164 g, 98%). Rf = 0.40 (9:1 CH2Cl2 ⁄ EtOAc). 1H-NMR (400 MHz,
CDCl3) d 8.49 (s, 1 H,), 8.39 (s, 1 H), 7.98 (m, 2 H), 7.71–7.68 (m, 2 H),
7.65–7.61 (m, 2 H), 7.18 (m, 2 H), 6.61 (m, 2 H), 6.53 (m, 1 H), 3.56–3.52
(m, 4 H), 2.77–2.74 (m, 12 H), 2.60–2.54 (m, 4 H), 1.26 (m, 6). 13C-NMR
(100 MHz) d 134.9,134.8, 131.4, 130.3, 129.8, 129.2, 129.1, 128.6, 128.5,
128.1,125.9,72.1,51.8,32.7,32.1,29.7,26.3,15.01 ppm.IR(neat)3415.3,
2964.0,2910.0, 1610.2,1519.6,1349.9, 1176.3.HRMS(+ESI)calculated
for MH+ 591.1844, observed 591.1802.
RESULTS AND DISCUSSION
Several fluorescent sensors utilize metal ion chelators contain-
ing thioether ligands to provide selectivity for Ni2+ (15), Pd2+
(16), Cu+ (17,18), Ag+ (19), Cd2+ (20) and Hg2+ (21,22).
Binding selectivity primarily stems from favorable soft–soft
acid–base interactions. CrownCast-3, which utilizes a 10-
phenyl-1,4-dioxa-7,13-dithia-10-azacyclopentadecane (AT215
C5) macrocyle, exhibits high affinity for Hg2+ similar to
analogous fluorescent sensors (23), but only a modest decrease
in affinity upon uncaging (7). Strong Hg2+-thioether coordi-
nation coupled with the structural restrictions of the macro-
cycle force a Hg2+-anilino interaction that inhibits metal ion
release. By eliminating macrocyclic effects, we reasoned that a
photocage could utilize thioether-heavy metal ion selectivity
without compromising uncaging properties.
Many Cu+ and Ag+ fluorescent sensors derive binding
specificity from the acyclic polythioether ligand 3,6,12,15-
tetrathia-9-azaheptadecane (TTAHD; 17,19,24–30). We envi-
sioned using this chelator to make a photocaged complex for
soft, monovalent metal ions using the Cast photocage frame-
work. With a phenyl-derived TTAHD ligand, a nitrobenzhy-
drol caging group would provide a means to attenuate the
electron-donating ability of the aniline lone pair similar to the
Tsien Ca2+ photocages (Fig. 1).
(4-(Bis(2-((2-(ethylthio)ethyl)thio)ethyl)amino)phenyl)(1-nitro
naphthalen-2-yl)methanol (7, ArgenCast-3). To a solution of 1
(0.204 g, 0.525 mmol) and 1-nitro-2-naphthaldehyde (6, 0.116 g,
0.577 mmol) in CH2Cl2 (3.0 mL) was added lutidine (0.084 g,
0.787 mmol) and TMSOTf (0.151 g, 0.682 mmol). The resulting dark
green colored solution was stirred for 16 h at 23ꢀC. The solution was
diluted with 50 mL of CH2Cl2, washed with brine (50 mL), dried over
MgSO4, filtered and evaporated to dryness. The resulting crude TMS-
ether was dissolved in CH3CN (3.0 mL) to which 18-crown-6 (0.416 g,
1.57 mmol) and KF (0.305 g, 5.24 mmol) dissolved in minimal amount
of water, was added. The resulting reaction mixture was stirred for
12 h at 23ꢀC. The reaction mixture was diluted with water (50 mL),
extracted into CH2Cl2. The organic extract was washed with aqueous
KCl, dried over MgSO4, and the solvent was removed. Flash
chromatography on silica with CH2Cl2 furnished 7 a yellow oil
(0.225 g, 72%). Rf = 0.34 (CH2Cl2). 1H-NMR (400 MHz, CDCl3) d
7.98 (d, 1 H,), 7.86 (d, 1 H), 7.71 (t, 2 H), 7.63–7.55 (m, 2 H), 7.24 (d, 2
H), 6.60 (d, 2 H), 6.03 (s, 1 H), 3.55–3.51 (m, 4 H), 2.75–2.73 (m, 12 H),
2.56–2.54 (m, 4 H), 1.27 (m, 6). 13C-NMR (100 MHz) d 146.6,146.4,
133.4, 133.2, 131.1, 130.0, 128.8, 128.2, 127.6, 124.5, 124.4, 121.9,112.0,
71.2,51.8, 32.7, 32.0, 29.64, 26.3, 15.0 ppm. IR (neat) 3407.6, 2964.0,
2910.0, 1610.2, 1519.6, 1357.6, 1267.0, 1176.3. HRMS (+ESI)
calculated for MH+ 591.1844, observed 591.1802.
The desired 9-phenyl-TTAHD (1) was prepared as previously
described (12), and the corresponding photocage was synthe-
sized by an electrophilic aromatic substitution reaction with
ortho-nitroveratraldehyde promoted by TMSOTf (Scheme 1).
The photocage ArgenCast-1 was isolated in 63% yield after
removal of the silyl group and purification by flash chromatog-
raphy. The photoproduct, ArgenUnc-1, also was prepared by
bulk photolysis for metal binding studies.
General spectroscopic methods. All solutions were prepared with
spectrophotometric grade solvents. HEPES (4-(2-hydroxyethyl)-1-
piperazineethanesulfonic acid) and KCl (99.5%) were purchased and
used as received. Cu+ stock solutions were prepared from 99.9% pure
(CH3CN)4CuPF6, whereas Ag+ and Hg2+ were prepared from the
The interaction between the photocages and metal ions was
assessed spectrophotometrically. Monitoring the disappear-

846 Hannah W. Mbatia et al.
S
S
Mn+
S
S
S
Mn+
S
Mn+
S
S
S
S
S
S
N
N
N
h
Mn+ release
H2O
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
HO
O2N
O
O
ON
ON
ArgenCast-1 (2)
ArgenUnc-1 (3)
Figure 1. Expected uncaging action of ArgenCast-1 photocage. The conversion of the nitrobenzhydrol into a benzophenone creates a new
resonance interaction between the aniline lone pair and the carbonyl oxygen. The decreased electron density on the aniline nitrogen atom inhibits its
ability to engage in bonding; in addition, the introduction of partial positive charge on the aniline helps to repel the positively charged guest metal
ion. With thiophilic metal ions, the release is inhibited by strong sulfur-metal interactions, which significantly stabilize the photocage-metal
complex.
S
S
S
S
S
S
O
S
S
S
S
1. H3CO
H3CO
N
N
H
S
S
NO2
N
h
TMSOTf, 2,6-lutidine
2. TBAF
HO
O2N
O
H2O
OCH3
OCH3
OCH3
OCH3
ON
ArgenUnc-1 ( )
1
3
2
ArgenCast-1 ( )
Scheme 1. Synthesis of ArgenCast-1 and ArgenUnc-1.
0.6
0.5
0.4
0.3
0.2
0.1
0
0.25
ArgenCast-1
ArgenUnc-1
2 μM Ag+
A
B
0.25
0.6
2.5 μM Ag+
0.2
0.15
0.1
0.2
0.15
0.1
5 μM Ag+
4 μM Ag+
0.5
7.5 μM Ag+
6 μM Ag+
0.05
0
10 μM Ag+
8 μM Ag+
0.4
0.05
0
0
50
100
150
200
250
12.5 μM Ag+
10 μM Ag+
12 μM Ag+
14 μM Ag+
16 μM Ag+
20 μM Ag+
28 μM Ag+
44 μM Ag+
76 μM Ag+
140 μM Ag+
268 μM Ag+
[Ag] μM
0
5
10 15 20 25 30 35
0.3
0.2
0.1
0
15 μM Ag+
17.5 μM Ag+
20 μM Ag+
22.5 μM Ag+
25 μM Ag+
[Ag] μΜ
+
250
300
350
400
450
27.5 μM Ag
320
360
400
440
Wavelength (nm)
30 μM Ag+
32.5 μM Ag+
35 μM Ag+
Wavelength (nm)
Figure 2. (A) Titration of 25 lM ArgenCast-1 with AgClO4 in CH3CN. The absorbance changes were fit to a 1:1 binding isotherm (inset). The error
bars represent the variance in the measurements over three trials. (B) Titration of 25 lM ArgenUnc-1 with AgClO4 in CH3CN. The absorbance
changes were fit to a 1:1 binding isotherm (inset). The error bars represent the variance in the measurements over three trials.
ance of the aniline-derived absorption band at ca 270 nm upon
metal ion addition allows the metal ion affinity of the
photocage to be quantified (Fig. 2A). In an analogous manner,
the metal ion binding-induced erosion of the charge-transfer
band at ca 350 nm for the uncaged ligand allows the binding
strength to be addressed (Fig. 2B). As predicted, the soft metal
ions Ag+, Hg2+ and Cu+ interacted with ArgemCast-1
because of the thioether ligands (Table 1). The binding

Photochemistry and Photobiology, 2012, 88 847
Table 1. Dissociation constants of ArgenCast-1 and ArgenUnc-1 with metal ions in different solvent systems.
CH3CN (lM) CH3OH (lM)
ArgenUnc-1 ArgenUnc-1
Buffer (lM)
M+
ArgenCast-1
DKd
ArgenCast-1
DKd
ArgenCast-1
ArgenUnc-1
DKd
Ag+
Cu+
Hg2+
11.4
NB†
7.2
0.4
0.3
10.2
NB†
6.9
0.1
0.1
0.89
NA‡
0.95
11.9
10.2
NB†
0.1
0.2
7.2
7.7
0.4
0.6
0.61
0.76
NA‡
9.8
0.2
5.3
0.3
0.54
NA‡
NA‡
NA†
NB†
NA‡
NB†
NB†
The 1:1 binding constants were calculated with XLfit; †No binding observed (NB): metal binding was not observed when the ligand was exposed to
>5000 equiv of Mn+; ‡Not applicable (NA): the value for DKd could not be calculated, because no Kd value could be measured. Buffer: 20 mM
HEPES ⁄ dioxane (1:1), pH 7.0; see supporting information for titration data.
chemistry of ArgenCast-1 with Cd2+, Zn2+ and Cu2+ also
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 min
was screened, but no interactions were observed when the
ligand were exposed to >3000 equivalents of these metal ions.
The binding properties of ArgenCast-1 were investigated in
different solvent systems. In CH3CN, binding with both Ag+
and Hg2+ were observed with Kd values of 11.4 and 7.2 lM
respectively, but the ligand could not compete effectively with
CH3CN for Cu+ binding. In CH3OH, Cu+ and Ag+
interacted with ArgenCast-1 with Kd values of 10.2 and
5 min
10 min
15 min
20 min
11.9 lM respectively, but no binding was observed with Hg2+
.
As the spectrophotometric assay relies on interaction between
the aniline nitrogen atom and the metal ion, a thiophilic metal
ion like Hg2+ can bind tightly to the ligand without signif-
icantly perturbing the ligand absorption bands. Binding also
was investigated under aqueous conditions (20 mM
HEPES ⁄ dioxane 1:1, pH 7.0) where dioxane was used to
enhance solubility of the photocage. Only Ag+ binding could
be assessed under these conditions. As Ag+ was the only metal
ion that consistently interacted with the photocage, we
adopted the name ArgenCast-1 and ArgenUnc-1 for the
photocage 2 and photoproduct 3 respectively. The components
of the name reflect the class of photocages (Cast) and the Latin
word for silver, Argentum (Argen).
To function as an effective photocage, the chelator should
exhibit a strong metal ion affinity that drops significantly upon
photolysis. The DKd, which is the ratio of the disassociation
constant of the uncaged ligand to the original photocage,
ideally should be >10, and larger if possible; however, the
DKd < 1 suggests that ArgenUnc-1 possesses a higher binding
affinity than ArgenCast-1. This anomalous behavior results
from the nature of the absorption bands. As demonstrated
previously, the ca 350 nm absorption feature of the uncaged
ligand is a well-behaved benzophenone charge-transfer band.
The absorption band at ca 270 nm of the unphotolyzed
photocage appears to be a complicated combination of
transitions that do not change linearly with metal ion binding
(7). The DKd < 1 measured for ArgenCast instead suggests
that there is only a minimal change in binding affinity when
uncaged, similar to CrownCast-3. The minimal changes
compromise the ability of ArgenCast-1 to function as an
effective Ag+ photocage. Furthermore, the results suggest that
the thioether-metal ion interaction dominates the observed
affinity even when not incorporated into a macrocycle.
250
300
350
400
450
500
Wavelength (nm)
Figure 3. Photolysis of 25 lM ArgenCast-1 in CH3CN using an 8 W
source. The solution was photolyzed at k = 350 nm.
measured in the three different solvent systems (Table 2). The
F of apo-ArgenCast-1 was 0.8% in CH3CN, 1.4% in CH3OH
and 1.4% in buffer ⁄ dioxane. The addition of metal ions did
not have a statistically significant effect on the F, which
provides further evidence that the aniline nitrogen atom does
not contribute significantly to metal complex stability in
ArgenCast-1. In analogous Cast photocages, metal ion binding
appears to inhibit an unwanted resonance interaction between
the key aci-nitro intermediate and the aniline lone pair, which
enhances the quantum yield (6). The formation of a stable
thioether-metal ion complex would result in a weaker interac-
tion between the aniline nitrogen atom and metal ion that
would not prevent the resonance interaction that lowers the F
(Fig. 4).
Previously, dimethoxylnitrobenzyl (DMNB) group has been
replaced with other chromophores in photocages to enhance F
(31,32). The typical strategy for enhancing photolysis effi-
ciency is to increase the absorption cross section of the
chromophore, which means more of the photons introduced
into the system instigate photochemical reactions. The 3-nitro-
2-naphthyl–substituted photocages (3NN) have F as high as
60% (33–35), and we have investigated the 1-nitro-2-naphthyl
(1NN) as an alternative (6). Although 3NN requires a
multistep synthesis, the aldehyde precursor to make 1NN-
derived Cast photocages is commercially available. We were
interested in comparing the two nitronaphthyl chromophores
to determine if there are any quantitative differences in the
resulting photocages. ArgenCast-3 (5) was prepared with the
TMSOTf promoted reaction between 1 and 3-nitro-2-naph-
The conversion of ArgenCast-1 into ArgenUnc can be
quantified by simultaneously monitoring the disappearance of
the absorption band at k = ꢁ270 nm and the appearance of
the absorption at k = ꢁ350 nm (Fig. 3). Similar to the
binding studies, the quantum yield of photolysis (F) was

848 Hannah W. Mbatia et al.
Table 2. Fphotolysis for ArgenCast ligands and their corresponding metal complexes in various solvents.
e (cm)1
M
)1, k = 350 nm)
Fphotolysis (%)
CH3CN
CH3OH
Buffer
CH3CN
CH3OH
Buffer
1.3
1.2
0.39
1.4
ArgenCast-1
ArgenCast-2
ArgenCast-3
6300
2600
2200
6000
NA
2400
NA
1300
NA
5300
2400
2300
5700
5600
2500
2200
1500
1600
6600
2600
2700
6300
NA
2400
NA
NA
NA
0.77
0.2
0.1
0.3
0.3
1.4
1.3
1.3
1.5
1.5
1.4
1.4
1.3
1.3
0.3
0.1
0.2
0.5
0.2
0.3
0.4
0.3
0.1
0.2
0.3
0.1
0.4
0.81
0.69
0.92
NA
0.98
NA
0.74
NA
[Ag(ArgenCast-1)]+1
[Cu(ArgenCast-1)]+1
[Ag(ArgenCast-2)]+1
[Cu(ArgenCast-2)]+1
[Ag(ArgenCast-3)]+1
[Cu(ArgenCast-3)]+1
NA
1.3
0.1
0.4
0.2
NA
NA
NA
NA, metal binding not observed under these conditions; photolysis for 25 lM photocage using an 8 W reactor with k = 350 nm. Buffer: 20 mM
HEPES ⁄ dioxane (1:1), pH 7.0 (lM). Complex data was obtained at 25 lM of Mn+; see Supporting Information for photolysis data.
resonance delocalization of the aniline electrons into aci-nitro intermediate
S
S
S
S
S
S
Ag+
Ag+
Ag+
S
S
S
S
S
S
N
N
N
H3CO
H3CO
H3CO
H3CO
H3CO
H3CO
reduced uncaging due
to decreased aci-nitro
character in the intermediate
h
OH
NO2
OH
O
OH
O
N
N
OH
aci-nitro intermediate
OH
Figure 4. Resonance between the aci-nitro intermediate and the aniline nitrogen atom during photolysis of ArgenCast-1. In metal complexes with a
strong nitrogen-metal bond, the electron pair less readily participates in resonance with the aci-nitro intermediate, which increases the photolysis
quantum yield. In ArgenCast, the Ag+-thioether interaction appears to dominate complex stability with the aniline lone pair interacting minimally
with the metal, which leads to increased participation in resonance interactions and lower quantum yields.
O
OH
3
1. , TMSOTf
H
2,6-lutidine
S
NO2
NO2
N
S
2. KF, 18-crown-6
S
S
4
ArgenCast-2 (5)
NO2 OH
NO2
O
3
1. , TMSOTf
2,6-lutidine
H
2. KF, 18-crown-6
S
N
S
S
S
6
7
ArgenCast-3 ( )
Scheme 2. Synthesis of ArgenCast-2 and -3.
thaldehyde (13), which was prepared as previously described.
ArgenCast-4 (7) was synthesized in an analogous manner using
1-nitro-2-naphthaldehyde. The yield of ArgenCast-2 was 58%
and ArgenCast-3 was prepared in 72% yield (Scheme 2).
Upon investigation of the photochemistry, no significant
variation in F was observed for either the apo ligand or the
corresponding metal complexes. Moreover, there is no signif-
icant difference between the DMNB caging group and the two
naphthyl derivatives, which suggests that the resonance in the
Cast photocage framework precludes realizing the advantages
of a more strongly absorbing chromophore. Determination of
whether 1NN has any advantages or disadvantages over 3NN

Photochemistry and Photobiology, 2012, 88 849
will require incorporation and analysis in photocages con-
structed to avoid resonance between the aci-nitro intermediate
and the receptor.
isotherm. Error bars represent the variance in the measure-
ments over three trials.
Figure S10. (A) Titration of 25 lM ArgenUnc-1 with
AgClO4 in buffer. (B) Difference curve. (C) Binding isotherm.
Error bars represent the variance in the measurements over
three trials.
Figure S11. Photolysis of 25 lM ArgenCast-1 in MeCN
using a 8 W lamps photo reactor source at 350 nm.
Figure S12. Photolysis of 25 lM ArgenCast-1 in MeOH
using a 8 W lamps photo reactor source at 350 nm.
Figure S13. Photolysis of 25 lM ArgenCast-1 in buffer using
a 8 W lamps photo reactor source at 350 nm.
Figure S14. Photolysis of 25 lM ArgenCast-2 in MeCN
using a 8 W lamps photo reactor source at 350 nm.
Figure S15. Photolysis of 25 lM ArgenCast-2 in MeOH
using a 8 W lamps photoreactor source at 350 nm.
Figure S16. Photolysis of 25 lM ArgenCast-2 in buffer using
a 8 W lamps photoreactor source at 350 nm.
In summary, three ArgenCast compounds have successively
been synthesized and characterized. The thioether-based
receptor of the photocages has affinity for Ag+, Hg2+ and
Cu+ as predicted by the hard–soft acid–base theory. Argen-
Cast-1 interacts with Ag+ in all solvents tested with a Kd
values ranging 9–11 lM, which demonstrates that solvents had
limited effects on the binding affinity. In principle, ArgenUnc-1
has a decreased electron density on the aniline nitrogen atom,
and hence a lower affinity for guest cations; however, it retains
a relatively high affinity for Ag+. These results suggest that
while thioether ligands can impart the desired soft metal ion
binding, they also compromise metal ion release upon photol-
ysis. Devising strategies to enhance metal ion release and
increasing quantum yields in Cast systems are the subject of
ongoing efforts to design effective photocages for Ag+ and
other soft metal ions, such as Pd2+, Pt2+ and Au+ that
present similar challenges.
Figure S17. Photolysis of 25 lM ArgenCast-3 in MeCN
using a 8 W lamps photoreactor source at 350 nm.
Figure S18. Photolysis of 25 lM ArgenCast-3 in MeOH
using a 8 W lamps photoreactor source at 350 nm.
Figure S19. Photolysis of 25 lM ArgenCast-3 in buffer using
a 8 W lamps photoreactor source at 350 nm.
Acknowledgements—This work was supported by NSF grant CHE-
0955361.
Figure S20. Photolysis of 25 lM ArgenCast-1 in the presence
of 25 lM AgClO4 (1 Eq) in MeCN using a 8 W lamps
photoreactor source at 350 nm.
SUPPORTING INFORMATION
Additional Supporting Information may be found in the online
version of this article:
Figure S1. (A). Titration of 25 lM ArgenCast-1 with
AgClO4 in CH3CN. (B) Difference curve (C) Binding iso-
therm. Error bars represent the variance in the measurements
over three trials.
Figure S21. Photolysis of 25 lM ArgenCast-1 in the presence
of 25 lM AgClO4 (1 Eq) in MeOH using a 8 W lamps
photoreactor source at 350 nm.
Figure S22. Photolysis of 25 lM ArgenCast-1 in the presence
of 25 lM AgClO4 (1 Eq) in buffer using a 8 W lamps
photoreactor source at 350 nm.
Figure S2. (A) Titration of 25 lM ArgenCast-1 with
Hg2ClO4 in CH3CN. (B) Difference curve. (C) Binding
isotherm. Error bars represent the variance in the measure-
ments over three trials.
Figure S3. (A) Titration of 25 lM ArgenCast-1 with AgClO4
in MeOH. (B) Difference curve. (C) Binding isotherm. Error
bars represent the variance in the measurements over three trials.
Figure S4. (A) Titration of 25 lM ArgenCast-1 with
(CH3CN)4CuPF6 in MeOH. (B) Difference curve. (C) Binding
isotherm. Error bars represent the variance in the measure-
ments over three trials.
Figure S23. Photolysis of 25 lM ArgenCast-1 in the presence
of 25 lM (CH3CN)4CuPF6 (1 Eq) in MeOH using a 8 W lamps
photoreactor source at 350 nm.
Figure S24. Photolysis of 25 lM ArgenCast-2 in the presence
of 25 lM AgClO4 (1 Eq) in MeCN using a 8 W lamps
photoreactor source at 350 nm.
Figure S25. Photolysis of 25 lM ArgenCast-2 in the presence
of 25 lM AgClO4 (1 Eq) in MeOH using a 8 W lamps
photoreactor source at 350 nm.
Figure S26. Photolysis of 25 lM ArgenCast-2 in the presence
of 25 lM AgClO4 (1 Eq) in buffer using a 8 W lamps
photoreactor source at 350 nm.
Figure S5. (A) Titration of 25 lM ArgenCast-1 with AgClO4
in buffer. (B) Difference curve. (C) Binding isotherm. Error
bars represent the variance in the measurements over three
trials.
Figure S27. Photolysis of 25 lM ArgenCast-2 in the presence
of 25 lM (CH3CN)4CuPF6 (1 Eq) in MeOH using a 8 W lamps
photoreactor source at 350 nm.
Figure S6. (A) Titration of 25 lM ArgenUnc-1 with AgClO4
in CH3CN. (B) Difference curve. (C) Binding isotherm. Error
bars represent the variance in the measurements over three
trials.
Figure S7. (A) Titration of 25 lM ArgenUnc-1 with
Hg2ClO4 in CH3CN. (B) Difference curve. (C) Binding
isotherm. Error bars represent the variance in the measure-
ments over three trials.
Figure S8. (A) Titration of 25 lM ArgenUnc-1 with AgClO4
in MeOH. (B) Difference curve. (C) Binding isotherm. Error
bars represent the variance in the measurements over three trials.
Figure S9. (A) Titration of 25 lM ArgenUnc-1 with
(CH3CN)4CuPF6 in MeOH. (B) Difference curve. (C) Binding
Figure S28. Photolysis of 25 lM ArgenCast-3 in the presence
of 25 lM AgClO4 (1 Eq) in MeCN using a 8 W lamps
photoreactor source at 350 nm.
Figure S29. Photolysis of 25 lM ArgenCast-3 in the presence
of 25 lM AgClO4 (1 Eq) in MeOH using a 8 W lamps
photoreactor source at 350 nm.
Figure S30. Photolysis of 25 lM ArgenCast-3 in the presence
of 25 lM (CH3CN)4CuPF6 (1 Eq) in MeOH using a 8 W lamps
photoreactor source at 350 nm.
1
Figure S31. H and 13C NMR for ArgenCast-1.
1
Figure S32. H and 13C NMR for ArgenUNC-1.
1
Figure S33. H and 13C NMR for ArgenCast-2.
1
Figure S34. H and 13C NMR for ArgenCast-3.

850 Hannah W. Mbatia et al.
18. Yang, L., R. McRae, M. M. Henary, R. Patel, B. Lai, S. Vogt and
C. J. Fahrni (2005) Imaging of the intracellular topography of
copper with a fluorescent sensor and by synchrotron X-ray fluo-
rescence microscopy. Proc. Natl Acad. Sci. USA 102, 11179–
11184.
Please note: Wiley-Blackwell is not responsible for the
content or functionality of any supporting information sup-
plied by the authors. Any queries (other than missing material)
should be directed to the corresponding author for the article.
19. Ishikawa, J., H. Sakamoto, S. Nakao and H. Wada (1999) Silver
ion selective fluoroionophores based on anthracene-linked poly-
thiazaalkane or polythiaalkane derivatives. J. Org. Chem. 64,
1913–1921.
REFERENCES
1. Kaplan, J., B. Forbush and J. Hoffman (1978) Rapid photolytic
release of adenosine 5¢-triphosphate from a protected analog:
utilization by the sodium:potassium pump of human red blood cell
ghosts. Biochemistry 17, 1929–1935.
2. Lee, H., D. R. Larson and D. S. Lawrence (2009) Illuminating the
chemistry of life: design, synthesis, and applications of ‘‘caged’’
and related photoresponsive compounds. ACS Chem. Biol. 4, 409–
427.
3. Tsien, R. Y. and R. S. Zucker (1986) Control of cytoplasmic
calcium with photolabile tetracarboxylate 2-nitrobenzhydrol che-
lators. Biophys. J . 50, 843–853.
4. Kennedy, D. P., C. Gwizdala and S. C. Burdette (2009) Methods
for preparing metal ion photocages: application to the synthesis of
crowncast. Org. Lett. 11, 2587–2590.
5. Kenndy, D. P., C. D. Incarvilo and S. C. Burdette (2010) Ferri-
Cast: a macrocyclic photocage for Fe3+. Inorg. Chem. 49, 916–
923.
6. Kennedy, D. P., D. C. Brown and S. C. Burdette (2010) Probing
nitrobenzhydrol uncaging mechanisms using FerriCast. Org. Lett.
12, 4486–4489.
20. Mameli, M., M. C. Aragoni, M. Arca, C. Caltagirone,
F. Demartin, G. Farruggia, G. De Filippo, F. A. Devillanova,
A. Garau, F. Isaia, V. Lippolis, S. Murgia, L. Prodi, A. Pintus and
N. Zaccheroni (2010) A selective, nontoxic, OFF-ON fluorescent
molecular sensor based on 8-hydroxyquinoline for probing Cd2+
in living cells. Chem. Eur. J. 16, 919–930.
21. Chen, T., W. Zhu, Y. Xu, S. Zhang, X. Zhang and X. Qian (2010)
A thioether-rich crown-based highly selective fluorescent sensor
for Hg2+ and Ag+ in aqueous solution. Dalton Trans. 39, 1316–
1320.
22. Yoon, S., E. W. Miller, Q. He, P. H. Do and C. J. Chang (2007)
A bright and specific fluorescent sensor for mercury in water, cells,
and tissue. Angew. Chem. Int. Ed. 46, 6658–6661.
23. Yuan, M., Y. Li, J. Li, C. Li, X. Liu, J. Lv, J. Xu, H. Liu, S. Wang
and D. Zhu (2007) A colorimetric and fluorometric dual-modal
assay for mercury ion by a molecule. Org. Lett. 9, 2313–2316.
24. Coskun, A. and E. Akkaya (2005) Ion sensing coupled to reso-
nance energy transfer: a highly selective and sensitive ratiometric
fluorescent chemosensor for Ag(I) by a modular approach. J. Am.
Chem. Soc. 127, 10464–10465.
7. Mbatia, H. W., D. P. Kennedy, C. E. Camire, C. D. Incarvito and
S. C. Burdette (2010) Buffering heavy metal ions with photoactive
crowncast cages. Eur. J. Inorg. Chem. 2010, 5069–5078.
8. Gwizdala, C., D. P. Kennedy and S. C. Burdette (2009) ZinCast-1:
a photochemically active chelator for Zn2+. Chem. Commun.,
6967–6969.
9. Peters, C., B. Munoz, F. J. Sepulveda, J. Urrutia, M. Quiroz, S.
Luza, G. V. De Ferrari, L. G. Aguayo and C. Opazo (2011)
Biphasic effects of copper on neurotransmission in rat hippo-
campal neurons. J. Neurochem. 119, 78–88.
25. Wang, H., L. Xue and H. Jiang (2011) Ratiometric fluorescent
sensor for silver ion and its resultant complex for iodide anion in
aqueous solution. Org. Lett. 13, 3844–3847.
26. Miller, E. W., L. Zeng, D. W. Domaille and C. J. Chang (2006)
Preparation and use of Coppersensor-1, a synthetic fluorophore
for live-cell copper imaging. Nat. Protoc. 1, 824–827.
27. Domaille, D. W., E. L. Que and C. J. Chang (2008) Synthetic
fluorescent sensors for studying the cell biology of metals. Nat.
Chem. Biol. 4, 168–175.
28. Dodani, S. C., S. C. Leary, P. A. Cobine, D. R. Winge and C. J.
Chang (2011) A targetable fluorescent sensor reveals that copper-
deficient SCO1 and SCO2 patient cells prioritize mitochondrial
copper homeostasis. J. Am. Chem. Soc. 133, 8606–8616.
29. Dodani, S. C., D. W. Domaille, C. I. Nam, E. W. Miller, L. A.
Finney, S. Vogt and C. J. Chang (2011) Calcium-dependent cop-
per redistributions in neuronal cells revealed by a fluorescent
copper sensor and X-ray fluorescence microscopy. Proc. Natl
Acad. Sci. USA 108, 5980–5985.
30. Domaille, D. W., L. Zeng and C. J. Chang (2010) Visualizing
ascorbate-triggered release of labile copper within living cells using
a ratiometric fluorescent sensor. J. Am. Chem. Soc. 132, 1194–
1195.
10. Florea, A. and D. Buesselberg (2006) Occurrence, use and
potential toxic effects of metals and metal compounds. Biometals
19, 419–427.
11. Furno, F., K. Morley, B. Wong, B. Sharp, P. Arnold, S. Howdle,
R. Bayston, P. Brown, P. Winship and H. Reid (2004) Silver
nanoparticles and polymeric medical devices: a new approach to
prevention of infection? J. Antimicrob. Chemother. 54, 1019–1024.
12. Ishikawa, J., H. Sakamoto, T. Mizuno and M. Otomo (1995)
Hydrazones derived from dithiamonoaza and tetrathiamonoaza
analogs of polyethers as silver ion selective ionophores: syntheses,
proton-dissociation behaviors, and metal ion complexing proper-
ties in 1,4-dioxane-water acidic solution. Bull. Chem. Soc. Jpn. 68,
3071–3076.
31. Momotake, A., N. Lindegger, E. Niggli, R. Barsotti and G. Ellis-
Davies (2006) The nitrodibenzofuran chromophore: a new caging
group for ultra-efficient photolysis in living cells. Nat. Methods 3,
35–40.
13. Kienzle, F. (1980) The reaction of phthalaldehydes with 3-nitro-
propionates—a simple route to 3-nitro-2-naphthoic acids. Helv.
Chim. Acta 63, 2364–2369.
14. ID Business Solutions Limited (2009) XLfit, 5.1.0.0. ID Business
Solutions Limited, Guildford.
15. Dodani, S. C., Q. He and C. J. Chang (2009) A turn-on fluorescent
sensor for detecting nickel in living cells. J. Am. Chem. Soc. 131,
18020–18021.
16. Schwarze, T., H. Mueller, C. Dosche, T. Klamroth, W. Mickler,
A. Kelling, H. Loehmannsroeben, P. Saalfrank and H. Holdt
(2007) Luminescence detection of open-shell transition-metal ions
by photoinduced electron transfer controlled by internal charge
transfer of a receptor. Angew. Chem. Int. Ed. 46, 1671–1674.
17. Zeng, L., E. W. Miller, A. Pralle, E. Y. Isacoff and C. J. Chang
(2006) A selective turn-on fluorescent sensor for imaging copper in
living cells. J. Am. Chem. Soc. 128, 10–11.
32. Lusic, H., R. Uprety and A. Deiters (2010) Improved synthesis of
the two-photon caging group 3-nitro-2-ethyldibenzofuran and its
application to a caged thymidine phosphoramidite. Org. Lett. 12,
916–919.
33. Singh, A. K. and P. K. Khade (2002) Synthesis and photochemical
properties of nitro-naphthyl chromophore and the corresponding
immunoglobulin bioconjugate. Bioconjug. Chem. 13, 1286–1291.
34. Singh, A. and P. Khade (2005) 3-Nitro-2-naphthalenemethanol: a
photocleavable protecting group for carboxylic acids. Tetrahedron
61, 10007–10012.
35. Singh, A. K. and P. K. Khade (2011) 7-Methoxy-3-nitro-2-naph-
thalenemethanol-a new phototrigger for caging applications.
Tetrahedron Lett. 52, 4899–4902.