A second-generation photocage for Zn2+ inspired by TPEN: Characterization and insight into the uncaging quantum yields of ZinCleav chelators

Article abstract of DOI:10.1002/chem.201001982

Photocages have been used to elucidate the biological functions of various small molecules and Ca²⁺; however, there are very few photocages available for other metal ions. ZinCleav-2 (1-(4,5-dimethoxy-2-nitrophenyl)-N,N, N′,N′-tetrakis-pyridin-2-ylmethyl-ethane-1,2-diamine) is a second-generation photocage for Zn²⁺ that releases the metal ion after a light-induced bifurcation of the chelating ligand. The structure of ZinCleav-2 was inspired by TPEN (N,N,N′,N′-tetrakis(2-pyridylmethyl) ethylenediamine), which is routinely used to sequester metal ions in cells owing to its high binding affinity. Inclusion of a 2-nitrobenzyl chromophore leads to the formation of two more weakly binding di-(2-picolyl)amine (DPA) fragments upon photolysis of the TPEN backbone. The desired ligand was prepared using a modified procedure used to access ZinCleav-1 (1-(4,5-dimethoxy-2-nitrophenyl)-N, N′-dimethyl-N,N′-bis-pyridin-2-ylmethyl-ethane-1,2-diamine). ZinCleav-2 has a conditional dissociation constant (K_d) of ～0.9 fM as measured by competitive titration with a quinoline-based fluorescent sensor for Zn²⁺. The K_d of the Zn²⁺ complex of the DPA photoproducts is ～158 nM; therefore, the ΔK_d for ZinCleav-2 photocage is ～10⁸. A large ΔK_d is required to significantly perturb free metal ion concentrations in biological assays. The quantum yield of photolysis of apo ZinCleav-2 and the [Zn(ZinCleav-2)]²⁺ complex are 4.7 and 2.3 %, respectively, as determined by HPLC analysis. Proof of concept Zn²⁺ release upon photolysis of [Zn(ZinCleav-2)]²⁺ was demonstrated using the fluorescent sensor Zinpyr-1, and the speciation of Zn²⁺ complexes was simulated using computational methods. The influence of benzylic substituents on the quantum yield of uncaging is also analyzed with the aim of tuning the photochemical properties caged complexes for in vivo experiments.

Full text of DOI:10.1002/chem.201001982

DOI: 10.1002/chem.201001982
A Second-Generation Photocage for Zn2+ Inspired by TPEN:
Characterization and Insight into the Uncaging Quantum Yields of ZinCleav
Chelators**
H. M. Dhammika Bandara, Timothy P. Walsh, and Shawn C. Burdette*[a]
Abstract: Photocages have been used
di-(2-picolyl)
G
tocage is ~108. A large DKd is required
to significantly perturb free metal ion
concentrations in biological assays. The
quantum yield of photolysis of apo
ZinCleav-2 and the [Zn(ZinCleav-2)]2+
complex are 4.7 and 2.3%, respectively,
as determined by HPLC analysis. Proof
of concept Zn2+ release upon photoly-
sis of [Zn(ZinCleav-2)]2+ was demon-
strated using the fluorescent sensor
Zinpyr-1, and the speciation of Zn2+
complexes was simulated using compu-
tational methods. The influence of ben-
zylic substituents on the quantum yield
of uncaging is also analyzed with the
aim of tuning the photochemical prop-
erties caged complexes for in vivo ex-
periments.
to elucidate the biological functions of
upon photolysis of the TPEN back-
bone. The desired ligand was prepared
using a modified procedure used to
access ZinCleav-1 (1-(4,5-dimethoxy-2-
nitrophenyl)-N,N’-dimethyl-N,N’-bis-
pyridin-2-ylmethyl-ethane-1,2-dia-
mine). ZinCleav-2 has a conditional
dissociation constant (Kd) of ~0.9 fm as
measured by competitive titration with
a quinoline-based fluorescent sensor
for Zn2+. The Kd of the Zn2+ complex
of the DPA photoproducts is ~158 nm;
therefore, the DKd for ZinCleav-2 pho-
various small molecules and Ca2+
;
however, there are very few photo-
ACHTUNGTRENNUNGcages available for other metal ions.
ZinCleav-2 (1-(4,5-dimethoxy-2-nitro-
phenyl)-N,N,N’,N’-tetrakis-pyridin-2-yl-
methyl-ethane-1,2-diamine)
is
a
second-generation photocage for Zn2+
that releases the metal ion after a light-
induced bifurcation of the chelating
ligand. The structure of ZinCleav-2
was inspired by TPEN (N,N,N’,N’-tet-
rakis(2-pyridylmethyl)ethylenedia-
mine), which is routinely used to se-
quester metal ions in cells owing to its
high binding affinity. Inclusion of a 2-
nitrobenzyl chromophore leads to the
formation of two more weakly binding
Keywords: cage compounds
gand effects · photolysis · photophy-
sics · zinc
Introduction
strated the feasibility of using photocaged compounds to
study the biology of complicated, dynamic living systems.
Photocages for a multitude of biomolecules including neuro-
transmitters,[8–10] inositols,[11,12] mRNA,[13] DNA,[14] pep-
tides[15,16] and enzymes[17–19] have been reported subsequent-
ly. Understanding the function of these analytes in numer-
ous processes has been facilitated through studies that can
only be conducted with photocages. Although alternative
photolabile groups, including ones that are amenable to 2-
photon excitation, are the subject of significant research ef-
forts, o-nitrobenzyl derived photocages remain the most
commonly encountered protecting group for caging.[20,21]
Nitr-5 and Nitr-7, two caged metal complexes designed to
release a metal cation upon irradiation, were reported by
Tsien et al. nearly a decade after the introduction of caged-
ATP.[22,23] The Nitr photocages consist of the Ca2+-specific
Photoactive protecting groups were developed nearly 50
years ago to take advantage of light-induced removal strat-
egies that are orthogonal to standard deprotection reac-
tions.[1–3] Subsequently this methodology was expanded to
engineer photocages that exploit the photolysis reaction to
liberate biologically active species.[4–6] The first photocage
consisted of an adenosine-5’-triphosphate (ATP) that was
rendered biologically inert by linking an o-nitrobenzyl
moiety to the terminal phosphate oxygen atom.[7] Irradiation
of the photocage liberates ATP, which can then participate
in normal physiological functions. This initial study demon-
[a] H. M. D. Bandara, T. P. Walsh, Prof. S. C. Burdette
Department of Chemistry, University of Connecticut
55 North Eagleville Road, Storrs, CT 06269-3060 (USA)
Fax : (+1)860-486-2981
chelator
BAPTA
(1,2-bis(o-aminophenoxy)ethane-
N,N,N’,N’-tetraacetic acid) covalently linked to a o-nitro-
benzhydrol group attached to the aminophenoxy ring para
to the amino group. Irradiation of the caged-Ca2+ complex
converts the benzhydrol into a nitrosobenzophenone. The
new resonance delocalization of the amino lone pair into
the benzophenone carbonyl lowers the affinity of the
BAPTA moiety for Ca2+, which induces a release of the
caged metal ion. This strategy has been exploited to gener-
E-mail: shawn.burdette@uconn.edu
[**] TPEN
Cleav-1
=
N,N,N’,N’-Tetrakis[2-pyridylmethyl]-ethylenediamine, Zin-
1-(4,5-dimethoxy-2-nitrophenyl)-N,N’-dimethyl-N,N’-bis-
pyridin-2-ylmethylethane-1,2-diamine and ZinCleav-2 = 1-(4,5-dime-
thoxy-2-nitrophenyl)-N,N,N’,N’-tetrakispyridin-2-ylmethylethane-1,2-
diamine.
=
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001982.
3932
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3932 – 3941

FULL PAPER
[25]
ate cages that release alkali earth metals,[24] Fe3+
,
and
Table 1. Quantum yields of photolysis (F) of apo cages and caged metal
complexes.[a]
Zn2+ [26] Unlike photocages for organic molecules, where un-
.
Apo cage
F [%]
Caged complex
F [%]
caging efficiency correlates primarily by the quantum yield
of photolysis (F), the release of metal ions from caged com-
plexes also depends on the difference in dissociation con-
stant between the caged and uncaged chelator (DKd =un-
ZinCleav-1[31],[b]
ZinCleav-2[a]
2[b]
2.2
4.7
10
6.6
12
20
[Zn(ZinCleav-1)]2+
[Zn(ZinCleav-2)]2+
[Zn(2)]2+
1.7
2.3
8.7
7.0
18
23
9.0
13
3[b]
[Ca(3)]2+
ACHTUNGTRENNUNGcaged Kd/caged Kd). Although CrownCast and ZinCast com-
DM-Nitrophen[45],[c]
NP-EGTA[27],[d]
DMNPE-4[57],[e]
DMNPE-5[57],[e]
4[b]
[Ca(DM-Nitrophen)]2ꢁ
pounds have greater DKd (~400-fold) than Nitr cages (~40-
fold), the quantum yields for all 2-nitrobenzhydrol derived
caged complexes are small (<5%), so the increase in free
metal ion concentration after flash photolysis remains
modest.
To address the shortcomings of these caged complexes,
Ellis-Davies and others developed an alternative uncaging
strategy in which chelators with a high affinity for Ca2+ un-
dergo bifurcation upon illumination.[27,28] The two resulting
photoproduct fragments possess significantly lower affinity
for Ca2+, which releases the caged metal ion. Investigations
of NP-EGTA and DM-nitrophen reveal that the reduction
of chelate effects provides photocages possessing DKd values
of 104–105-fold (Scheme 1). Photolysis of the complexes
[Ca
[Ca
(DMNPE-4)]2ꢁ
[Ca
(DMNPE-5)]2ꢁ
NA[h]
[Cu(CuCage)]
A
NA[g]
NA[g]
17.2
73
N
AHCTUNGTRENNUNG
NA
32
CuCage[30],[f]
ACHTUNGTRENNUNG
[a] Quantum yields were determined by HPLC methodologies with the
exception of the measurements of compounds 2, 3 and 4 that were mea-
sured using UV/Vis spectroscopy. [b] Aqueous buffer (40 mm HEPES,
100 mm KCl, pH 7); 350 nm, 150
[c] Aqueous buffer (40 mm HEPES, 100 mm KCl, pH 7.2); ~300–400 nm,
1000 filtered Hg acr lamp. [d] Aqueous buffer (10 mm HEPES,
W Xe lamp; HPLC analysis.
W
130 mm KCl, pH 7.1); Ruby laser, 347 nm or ~300–400 nm, 1000 W fil-
tered Hg acr lamp. [e] Aqueous buffer (40 mm HEPES, 100 mm KCl,
pH 7.2); ~300–400 nm, 500 W filtered Hg acr lamp. [f] Phosphate buffer
(NaH2PO4, pH 7.4); lmax =350 nm, 1200 W Rayonet-RPR-100 photoreac-
tor. [g] No value for the apo-photocage is reported. [h] 4 does not inter-
act strongly with Zn2+ under the experimental conditions.
cently reported.[30] The quan-
tum yield of photolysis and DKd
for this caged-Cu2+ complex
were 32% and 2ꢁ105-fold, re-
spectively.
Recently we demonstrated
the release of Zn2+ upon pho-
tolysis of the photocaged com-
[31]
plex
[Zn(ZinCleav-1)]2+
.
ZinCleav-1 combines the tetra-
dentate ligand EBAP (ethyl-
ene-bis-a,a’-(2-aminomethyl)-
pyridine) with a 4,5-dimethoxy-
2-nitrobenzyl
chromophore
(Scheme 1). The nomenclature
ZinCleav refers both to the an-
alyte of interest (zinc) and the
mechanism of metal ion release
(bond cleavage). Irradiation of
[Zn(ZinCleav-1)]2+
350 nm wavelength radiation
generates two weakly Zn2+
with
Scheme 1. Structures of metal ion photocages that fragment upon exposing the nitrobenzyl-based chelator to
light.
-
binding 2-(aminomethyl)pyri-
dine fragments, which releases
occurs with quantum yields of between 18–23% (Table 1),
which is significantly higher than those measured with o-ni-
trobenzhydrol derived cages. As a result, NP-EGTA and
DM-nitrophen produce significantly larger increases in free
Ca2+ concentration upon illumination; however, under in-
tracellular conditions, DM-nitrophen exists as the Mg2+
complex.[29] As with Nitr cages, replacement of the Ca2+-se-
lective binding moiety with new ligands allows caging of
other metals ions. A tetradentate chelator with amide and
pyridyl ligands for caging Cu2+, CuCage (Scheme 1) was re-
the bound metal ion. ZinCleav-1 possesses a high affinity
for Zn2+ (Kd =0.23 pm), whereas the affinities of the photo-
fragments are significantly lower (Kd =~40 mm), so the DKd
after uncaging is ~108. Despite binding Zn2+ tightly, Zn2+
-
binding proteins like metallothionein (Kd ꢀ 31 fm)[32] have a
higher affinity for the metal ion than the photocage. Since
these metalloproteins potentially can extract Zn2+ from Zin-
Cleav-1, a higher affinity caged complex would be desirable
for conducting some in vitro experiments. In addition to the
DKd, the photolysis quantum yield of [Zn(ZinCleav-1)]2+ of
Chem. Eur. J. 2011, 17, 3932 – 3941
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3933

S. C. Burdette et al.
0.55% measured by UV-visible
spectroscopy is lower than the
values reported for the structur-
ally similar Cu2+- and Ca2+
caged complexes.
-
To address these issues, we
sought to develop a second-
generation ZinCleav complex
with a higher affinity for Zn2+
and an improved quantum yield
of photolysis. The hexadentate
ligand TPEN (N,N,N’,N’-tetra-
kis(2-pyridylmethyl)-ethylene-
diamine) possesses a higher af-
finity for Zn2+ (Kd =0.4 fm)
than most metalloproteins.[33]
Owing to its high affinity for
Zn2+ and other transition metal
ions, investigators routinely use
Scheme 3. Synthesis of ZinCleav-2.
TPEN to scavenge unwanted metal ions in cells,[34–36] to
buffer intracellular free [Zn2+],[37] and as inspiration for the
receptor in fluorescent sensors for biological imaging.[38–41]
Although its high affinity makes TPEN a versatile chelator,
prolonged exposure of live cells to excess TPEN induces
apoptosis due to depletion of essential intracellular metal
ions.[42] Access to a ZinCleav derivative utilizing a TPEN
chelator would provide a caged complex that overcomes the
intrinsic competition between proteins and ZinCleav-1 for
Zn2+, and keep the cage metalated until exposed to light.
While the apo ligand would retain the toxicity of TPEN, the
metallation of a TPEN-inspired photocage with Zn2+ would
moderate the detrimental effects of introducing a high affin-
ity chelator to cells.
3,4-dimethoxybenzaldehyde, aldehyde 5 was synthesized as
described previously.[31] Reductive amination of 5 in the
presence of DPA provided the protected amine 6. Deprotec-
tion of 6 followed by reductive amination with two equiva-
lents of 2-pyridinecarboxaldehyde yielded the modified
TPEN ligand 8 with an overall yield of 28.5% over three
steps. Approximately 1 g of the TPEN ligand (unnitrated
ZinCleav-2) can be generated in three steps starting with 3 g
of 5. Attempts to recrystallize 8 as the hydrochloric acid salt
were unsuccessful due to decomposition of the ligand. Nitra-
tion of 8 was carried out at ꢁ408C with triflic acid
(CF3SO3H) and fuming nitric acid. The yield of the nitration
step was low (14%) compared to that of ZinCleav-1 (39%)
possibly due to the apparent acid sensitivity of 8. Although
the yield of the nitration reaction is low, the synthesis of
ZinCleav-2 further demonstrates the versatility of our new
methods for preparing nitrobenzyl derived metal ion chela-
tors.
Results and Discussion
We designed the TPEN-inspired photocage ZinCleav-2
based on the ZinCleav-1 scaffold. Similar to ZinCleav-1,
photolysis of ZinCleav-2 would generate two di-(2-picolyl)-
The apparent dissociation constant (Kd) of [Zn(ZinCleav-
1)]2+ was determined by competitive titration with 4-(2-pyr-
idyl-2-azo)resorcinol (PAR); however, this technique was
not sufficiently sensitive to measure the Kd of the [Zn(Zin-
Cleav-2)]2+ complex. Potentiometric titrations to determine
the Zn2+ binding constants and pKa values of ZinCleav-2
also were unsuccessful due to the acid sensitivity of the
ligand and its limited solubility in water at mm concentra-
tions. Alternatively we envisioned using a different chelator
with a fluorescent reporting group as a ligand for competi-
tive titrations. A quinoline-based Zn2+ sensor with a Kd =
0.45 fm was recently reported.[43] This sensor forms a 1:1
complex with Zn2+ with a concomitant increase in fluores-
cence emission at 438 nm. The sensor was synthesized as de-
scribed, and used in competitive titrations under simulated
physiological conditions (25 mm HEPES, 100 mm NaClO4,
pH 7.4, Scheme 4). The data were corrected for ligand pro-
tonation using pKa values reported for TPEN. Analysis in
triplicate provided an apparent Kd of 0.9ꢂ0.1 fm for Zin-
Cleav-2 at pH 7.4, which is in agreement with the parent
ACHTUNGTRENNUNG
amine fragments (DPA, Scheme 2). Since DPA binds Zn2+
more tightly than the photoproducts of ZinCleav-1, the final
resting free Zn2+ would be buffered at lower concentrations
after photolysis of ZinCleav-2. The desired caged-Zn2+ com-
plex, [Zn(ZinCleav-2)]2+ was accessed by modifying the
route used to prepare ZinCleav-1 (Scheme 3). Starting with
Scheme 2. Uncaging action of [Zn(ZinCleav-2)]2+
.
3934
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3932 – 3941

Cage Compounds
FULL PAPER
structurally similar caged complexes, the efficiency of the
photolysis reaction was interrogated further.
Repeating the quantum yield measurements using
HPLC to assess the progress of the photolysis provided
slightly different results for the efficiency of the uncaging.
While the quantum yields of photolysis measured for
apo-ZinCleav-2 by UV/Vis (5.0%) and HPLC (4.7%)
are within experimental error, the uncaging of [Zn(Zin-
Cleav-2)]2+ appears to be more efficient when using
HPLC to monitor the disappearance of the starting mate-
rial. Further analysis of the HPLC traces suggest that the
photoproducts of ZinCleav-2 undergo secondary reac-
tions in the presence of the uncaged metal ions, which
would lead to deviation in the quantum yields measured
spectrophotometrically. Similar observations have been
made with Ca2+ photocages.[45] In addition to the meas-
urements with ZinCleav-2, the quantum yields of photol-
ysis for ZinCleav-1 and [Zn(ZinCleav-2)]2+ were reas-
sessed by HPLC. The results with apo-ZinCleav-1 are
consistent with the values determined by UV/Vis spec-
troscopy; however, [Zn(ZinCleav-1)]2+ uncages more ef-
ficiently than previously determined (Table 1).
Release of Zn2+ upon photolysis of [Zn(ZinCleav-2)]2+
was demonstrated by uncaging the complex in the pres-
ence of the fluorescent sensor Zinpyr-1.[46] Zinpyr-1 has a
[47]
Kd of 0.04 pm for Zn2+
,
and when Zn2+ (5 mm) is added
to a solution of 30 mm Zinpyr-1 (40 mm HEPES, 100 mm
KCl, pH 7.4) an enhancement in emission is observed.
The fluorescence signal returns to near its initial intensity
upon the addition of 40 mm ZinCleav-2. Irradiation of the
solution with
a 900 W source causes photolysis of
Scheme 4. Competitive titration between ZinCleav-2 and the Zn2+ sensor 11 in
[Zn(ZinCleav-2)]2+ and release of Zn2+ resulting in com-
plete recovery of the fluorescence signal (Scheme 5). The
fluorescence assay is consistent with the fragmentation of
the ZinCleav-2 ligand, and Zinpyr-1 scavenging the Zn2+
from the more weakly binding DPA photoproducts.
the presence of Zn2+. A 10 mm solution of the Zn2+ sensor (25 mm HEPES,
100 mm NaClO4, pH 7.4) and 5 mm ZnACHTNUTRGNE(NUG ClO4)2 was titrated against a solution of
ZinCleav-2 and emission intensity was measured after each addition. The Kd
was calculated by monitoring the decrease in emission intensity.
Caged complexes derived from o-nitrobenzyl groups
have widely varying quantum yields of photolysis. Caged
complexes with high quantum yields are desirable; how-
ever, there are no clear rules for predicting the uncaging ef-
ficiency of nitrobenzyl derivatives. The photoreaction of o-
nitrobenzyl groups has been the subject of numerous mecha-
nistic studies. Uncaging requires the light-induced genera-
tion of an aci-nitro intermediate that undergoes rearrange-
ment and fragmentation to release the analyte (Scheme 6,
red).[48,49] We hypothesized that the low uncaging quantum
yields of the ZinCleav complexes might be explained by the
generation of an intermediate with decreased aci-nitro char-
acter. To test this hypothesis, we synthesized a series of o-ni-
trobenzyl derivatives and compared the quantum yields to
the ZinCleav complexes and the analogous photocages for
other metal ions. The most likely interference with the for-
mation of the aci-nitro intermediate would be a resonance
interaction with a heteroatom attached to the benzylic
carbon atom (Scheme 6, blue). In such a mechanism, o-ni-
trobenzyl cages with strong electron donors like nitrogen
atoms at the benzylic position would be expected to have
TPEN ligand under the same conditions (Kd =0.2 fm). The
Kd of the Zn2+ complex of DPA is 158 nm;[44] therefore, Zin-
Cleav-2 has a DKd of ~108. Introduction of excess Cu2+ to
[Zn(ZinCleav-2)]2+ induces a ligand exchange reaction as
predicted by the Irving–Williams series; however the low
concentration of free Cu2+ in cells would prevent Zn2+ dis-
placement. Ca2+ does not bind ZinCleav-2 or displace Zn2+
from [Zn(ZinCleav-2)]2+ under physiological conditions.
To determine the quantum yield of ZinCleav-2, changes
in the UV/Vis spectrum were measured during the course of
the photolysis reaction, and the number of photons was de-
termined using actinometry. Using optical methods, the
quantum yield of ZinCleav-2 and [Zn(ZinCleav-2)]2+ were
calculated to be 5.0 and 0.59%, respectively. Like the meas-
urements with ZinCleav-1, the apo-ligand uncaged with a
quantum yield of ~5%, which decreased by nearly an order
of magnitude in the presence of Zn2+. Since the quantum
yields deviated significantly from the values reported for
Chem. Eur. J. 2011, 17, 3932 – 3941
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3935

S. C. Burdette et al.
the lowest quantum yields of photolysis. Coordination of the
nitrogen atoms to a metal ion should mitigate the resonance
interaction and increase the quantum yield.
Initially model photocages 2 and 3 were prepared to de-
lineate the effect of ancillary ligands on the quantum yield
of cages containing benzylic amine groups (Scheme 1 and
Table 1). The modest difference in quantum yield suggests
that carboxylate and pyridine groups do not influence the
photochemistry. The presence of Ca2+ enhances the photoly-
sis efficiency of 3, while Zn2+ slightly decreases the quantum
yield of 2; however, quantum yields of both the apo ligands
and the complexes are the same within experimental error.
Quantum yields of the Ellis–Davies photocages containing
benzylic nitrogen ligands increase in the presence of Ca2+
.
This behavior is consistent with the proposed mechanism
since a lone pair involved in a bonding interaction would be
less available to engage in resonance interactions with the
aci-nitro intermediate. Unexpectedly, the quantum yields of
ZinCleav-1 and ZinCleav-2 decrease in the presence of
Zn2+. Since ZinCleav compounds do not bind Ca2+, addi-
tional quantum yield measurements with the Zn2+ com-
plexes of DM-Nitrophen and NP-EGTA could provide im-
portant insight into reasons for the deviation from expected
behavior. Additional support of the aci-nitro deactivation
mechanism can be garnered from the quantum yield mea-
sured for o-nitrobenzyl cages bearing benzylic amides.
Amide nitrogen atoms are weak electron donors owing to
the resonance delocalization of the nitrogen lone pair of
electrons onto the carbonyl oxygen. The quantum yields of
apo-CuCage[30] and compound 4 provide supportive evi-
dence that diminished resonance with the aci-nitro inter-
mediate increases the quantum yield, and is consistent with
cages for organic amides.[50]
Although the nature of the leaving group and other fac-
tors probably influence the uncaging efficiency, additional
time-resolved spectroscopic experiments will be required to
fully delineate the contribution of each variable. The effects
of variances in methodologies used to determine quantum
yields and different light sources also cannot be quantified
in this initial analysis that relies on previously published re-
ports. Preliminary observations with ZinCleav complexes as
well as compounds 2, 3 and 4 suggest a clean conversion to
the expected products when using monochromatic light, but
a different slate of products is formed when using a broad
spectrum light source. Future
Scheme 5. Fluorescence response of Zinpyr-1 upon uncaging of Zin-
Cleav-2 (top). The emission intensity of a solution of 5.0 mm Zinpyr-1
(40 mm HEPES, 100 mm KCl, pH 7.4) was recorded before and after the
addition of 30 mm ZnCl2. The emission was integrated between 500–
600 nm and normalized to the maximum response (inset). Subsequent ad-
dition of 40 mm ZinCleav-2 quenched the emission of the Zinpyr-1 com-
plex, which has a Kd of 0.10 nm for Zn2+. Irradiation of the solution in a
Rayonet RPR-100 photoreactor resulted in a complete photolysis of Zin-
Cleav-2 within 4 min. This behavior is consistent with Zinpyr-1 acquiring
Zn2+ from the photolyzed ZinCleav-2 (bottom).
investigations will focus on ex-
plaining this phenomenon as
well as designing photocages
that account for the photo-
chemistry of the aci-nitro inter-
mediate.
Zinc is essential for the
proper function of cells, but
since it is cytotoxic at high con-
centrations, intracellular Zn2+
must be tightly regulated. Met-
alloproteins and other chelators
Scheme 6. Uncaging mechanism in ZinCleav-2 and related chelators (top), and proposed pathway for the deac-
tivation of the aci-nitro intermediate by electron donation from benzylic heteroatoms (bottom).
3936
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3932 – 3941

Cage Compounds
FULL PAPER
tightly bind and sequester Zn2+; however, eukaryotic cells
may contain a small pool of weakly bound or chelatable
zinc. Measurements of free Zn2+ for different cell lines put
the amount at low levels between 5 pm and ~2 nm.[51–54] No
detectable free Zn2+ exists in prokaryotic cells, however,
values as low as 0.5 fm have been estimated.[55] The total
concentration of zinc binding chelators in cells and their
average affinity for Zn2+ are also known for different cell
types. With this information it is possible to calculate the
100 pm, the increase in Zn2+ following photolysis of
[Zn(ZinCleav-2)]2+ is lower than those simulated with Zin-
Cleav-1. The higher Zn2+ affinity of the tridentate DPA
photoproducts sequester uncaged Zn2+ more efficiently
than the bidentate aminomethylpyridine photoproducts
formed after ZinCleav-1 photolysis. Strategies for manipu-
lating ZinCleav-2 in cells and applying these photocages to
problems in Zn2+ biology are ongoing.
exogenous [ZinCleav]/ACHTNUTGRNEUNG
[Zn2+] ratio required to maintain the
intracellular free Zn2+ at physiologically normal levels. We
chose six hypothetical scenarios to simulate buffering the
resting free Zn2+ with ZinCleav caged complexes. Concen-
tration of 1 fm, 10, 100, 500 and 790 pm were selected based
on studies of estimated free Zn2+ concentrations in different
cell lines, and the absolute Zn2+ and zinc-binding chelator
concentrations were assumed to be 250 and 275 mm, respec-
tively. These values closely match the typical concentrations
in cells.[53]
Conclusion
The key intermediate aldehyde 5 enables a variety of zinc
cages with different binding as well as photochemical prop-
erties to be accessed. The affinity of ZinCleav-2 for Zn2+
was increased significantly over ZinCleav-1 by increasing
the number of chelating pyridyl groups to provide a TPEN-
like chelator. Computational simulations demonstrate that
ZinCleav-2 can be used to buffer Zn2+ at intracellular con-
centrations ranging from fm to sub-nm, which are typical
levels found in cells. The high fm affinity of ZinCleav-2 pre-
vents perturbation of Zn2+ homeostasis prior to uncaging,
which is more difficult to accomplish with the first genera-
tion photocage ZinCleav-1 that has a pm Kd for Zn2+. Al-
though we discovered that HPLC methodologies provide
more accurate measurements of the quantum yields of [Zn-
Fluctuations in the intracellular free Zn2+ concentration
following photolysis of the caged complex can be simulated
with HySS, a computer program for modeling speciation in
solution.[56] Of the two ZinCleav caged-zinc complexes, only
the TPEN-based photocage ZinCleav-2 has the capacity to
maintain free Zn2+ at low fm concentrations under resting
conditions and still generate a significant increase in free
Zn2+ after photolysis (Table 2). A [ZinCleav-2]/
N
(ZinCleav)]2+ complexes and that the quantum yields of
ACHTUNGTRENNUNG
ZinCleav caged complexes are
higher than originally believed,
ZinCleav cages have lower
quantum yields of photolysis
when compared with the typi-
cally low uncaging efficiency of
all o-nitrobenzyl compounds.
The low quantum yields are at-
tributed to the formation of an
intermediate that closely resem-
bles the aci-nitro one that is the
key precursor in the uncaging
mechanism by resonance inter-
actions from benzylic hetero-
Table 2. Free [Zn2+] buffering using ZinCleav-2.[a]
Resting
[Zn2+
Added
ZinCleav-2
Added
Zn2+
[Zn2+] with
10% uncaging
G
[Zn2+] with
ACHTUNGTRENNUNG
50% uncaging
G
]
1 fm
1 pm
10 pm
100 pm
500 pm
790 pm
50 mm
50 mm
50 mm
50 mm
50 mm
50 mm
40 mm
50 mm
50 mm
50 mm
50 mm
50 mm
1.2 fm
3.1 fm
1.2 pm
12.7 pm
127 pm
637 pm
1.0 nm
29.5 pm
935 pm
2.9 nm
6.7 nm
8.5 nm
[a] Calculations are based on a total cellular [Zn2+] of 250 mm and a cellular chelator concentration of 275 mm.
Speciation was simulated and concentrations were calculated with HySS. ZinCleav-1 photoproducts were as-
sumed to have logb=4.4.
of 1:1 maintains Zn2+ at pm concentrations, and a slight
excess of apo-chelator can buffer Zn2+ in the fm regime. In
contrast, a significant excess ZinCleav-1 is necessary to pre-
vent perturbing Zn2+ homeostasis when the resting concen-
tration is between 1–10 pm
ACHTUNGTRENaNGNU toms. The quantum yield of photolysis of the caged com-
plex can be improved by installing an amide substituent or
other functional group at the benzylic position that does not
donate electrons by resonance; however, these changes also
Table 3. Free [Zn2+] buffering using ZinCleav-1.[a]
(Table 3). In addition to requir-
ing excess apo-photocage which
may not have the same biodis-
tribution as the caged complex
in a dynamic cellular environ-
ment, the increase in free Zn2+
following photolysis of [Zn(Zin-
Cleav-1)]2+ is smaller because
the excess chelator buffers the
released Zn2+. At initial Zn2+
Resting
[Zn2+
Added
[ZinCleav-1]
[Zn2+] with
10% uncaging
A
[Zn2+] with
ACHTUNGTRENNUNG
A
]
25% uncaging
50% uncaging
1 fm
1 pm
10 pm
100 pm
500 pm
790 pm
50 mm
50 mm
50 mm
50 mm
50 mm
50 mm
1.0 fm
1.0 fm
1.1 fm
3.3 pm
280 pm
34.9 nm
78 nm
1.1 pm
12.7 pm
128 pm
639 pm
1.0 nm
1.5 pm
20.0 pm
211 pm
1.05 nm
1.67 nm
98 nm
[a] Calculations are based on a total cellular [Zn2+] of 250 mm and a cellular chelator concentration of 275 mm.
Speciation was simulated and concentrations were calculated with HySS. ZinCleav-2 photoproducts were as-
concentrations higher than sumed to have logb 6.8.
Chem. Eur. J. 2011, 17, 3932 – 3941
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3937

S. C. Burdette et al.
2919.1, 1729.4, 1740.3, 1576.6, 1514.9, 1470.5, 1436.8, 1369.4, 1324.9,
compromise the metal binding properties of the ligand. In
order to design caged complexes that uncage efficiently, ad-
ditional investigations into the mechanism of uncaging will
need to be conducted, and new strategies for ligand con-
struction will need to be devised. Other properties of the
cages will also need to be optimized to ensure that the pho-
tocages are appropriate for the desired biological applica-
tions. Synthesis and characterization of improved ZinCleav
derivatives are in progress.
1196.4, 1161.0, 1060.4, 1022.1, 982.9, 906.7, 847.7, 793.4, 756.8, 736.9 cmꢁ1
;
HRMS (ESI+): m/z: calcd for: 384.1533; found: 384.1547 [M+H+].
[Carboxymethyl-(4,5-dimethoxy-2-nitro-benzyl)-amino]-acetic acid (3):
Diester 9 (49 mg, 0.13 mmol) was dissolved in 1,4-dioxane (1.4 mL) and
treated with 1m KOH (630 mL). The resulting solution was stirred at
238C for 18 h. The pH of the solution was adjusted to 2 with concentrat-
ed HCl and the solution was left at ꢁ108C for 18 h. The desired product
precipitated as yellow crystals (7.0 mg, 15.5%). M.p. 210–2128C;
1H NMR (400 MHz, D2O): d = 7.55 (s, 1H, ArH), 7.43 (s, 1H, ArH),
3.99 (s, 2H, CH2), 3.99 (s, 3H, CH3), 3.89 (s, 3H, CH3), 3.10 ppm (s, 4H,
CH2); 13C NMR (100 MHz, D2O): d = 179.9, 153.2, 146.9, 141.2, 131.2,
113.2, 108.7, 58.6, 56.6, 56.3, 54.9 ppm; IR (thin film): n˜ = 3322.4, 3032.1,
2989.4, 2890.3, 1727.3, 1577.0, 1502.4, 1480.8, 1370.5, 1279.5, 1230.3,
1187.7, 1059.2, 981.3, 902.1, 796.4, 674.7 cmꢁ1; HRMS (ESI+): m/z: calcd
for: 328.0907; found: 328.0913 [M+H+].
Experimental Section
General synthetic procedures: All reagents were purchased from Acros
at the highest commercial quality and used without further purification
unless otherwise stated. [1-(3,4-dimethoxyphenyl)-2-oxo-ethyl]-carbamic
acid tert-butyl ester (5),[31] di-(2-picolyl)amine (DPA),[44] 9-(o-carboxy-
Pyridine-2-carboxylic acid [(4,5-dimethoxy-2-nitro-phenyl)-(methyl-pyri-
din-2-ylmethyl-amino)-methyl]-amide (4): Picolinic acid (122 mg,
0.99 mmol) and N-methylmorpholine (0.11 mL, 0.99 mmol) were com-
bined in CH2Cl2 (3.5 mL) and cooled to 08C. A solution of benzotriazoly-
phenyl)-2,7-dichloro-4,5-bisACTHNUGRTENUNG[bis(2-pyridylmethyl)-aminomethyl]-6-hy-
droxy-3-xanthanone (Zinpyr-1, 10),[46] and {2-[(bis-pyridin-2-ylmethyl-
amino)-methyl]-quinolin-8-yloxy}-acetic acid sodium salt (11)[43] were
synthesized according to literature procedures. Dichloromethane
(CH2Cl2) and tetrahydrofuran (THF) were sparged with argon and dried
by passage through a Seca Solvent Purification System. All chromatogra-
phy and TLC were performed on silica (230–400 mesh) from Silicycle
unless otherwise specified. TLCs were developed with mixtures of
CH2Cl2/CH3OH or CH2Cl2/CH3OH/NH4OH unless otherwise specified
and were visualized with 254 and 365 nm light. 1H and 13C NMR spectra
were recorded using a Bruker 400 MHz NMR instrument 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 sam-
ples were prepared as KBr pellets or thin films on KBr plates. High reso-
lution 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.
loxytris(dimethyl-amino)phosphoniumhexafluorophosphate
(BOP,
438 mg, 0.99 mmol) in CH2Cl2 (2.5 mL) was added and the resulting mix-
ture was stirred at 238C for 18 h. A solution of 1-(3,4-dimethoxy-phenyl)-
N’-methyl-N’-pyridin-2-ylmethyl-ethane-1,2-diamine (300 mg, 0.99 mmol)
in DMF (17 mL) was added and the resulting solution was stirred at
238C for18 h. The solvent was removed, saturated aqueous NaCl (15 mL)
was added and the product was extracted into CH2Cl2 (3ꢁ10 mL). The
organic layers were combined, dried over MgSO4, and the solvent was re-
moved. The crude product (264 mg, 0.48 mmol) and CuACTHNUTRGNE(UNG NO3)2·3H2O
were combined in Ac2O (1 mL) and stirred at 08C for 8 h. Water (5 mL)
was added and the product was extracted into CH2Cl2 (3ꢁ10 mL). The
organic layers were combined, washed with 1.0m solution of EDTA
(pH 11), dried over MgSO4 and the solvent was removed. Flash chroma-
tography on silica (91:8:1 CH2Cl2/CH3OH/NH4OH) yielded 4 as a yellow
oil (34.1 mg, 7.6%). 1H NMR (400 MHz, CDCl3): d = 8.73 (s, 1H, NH),
8.54–8.50 (m, 2H, ArH), 7.66 (dd, J=1.0, 6.7 Hz, 1H, ArH), 7.63–7.7.59
(m, 1H, ArH), 7.45 (d, J=7.7 Hz, 1H, ArH), 7.28–7.24 (m, 2H, ArH),
7.19–7.15 (m, 3H, ArH), 4.65 (dd, J=12.3, 5.1 Hz, 1H, CH), 3.86 (s, 3H,
CH3), 3.83 (s, 3H, CH3), 3.72 (d, J=4.0 Hz, 1H, CH2), 3.61 (d, J=4.0 Hz,
1H, CH2), 2.70 (dd, J=12.3, 16.1 Hz, 1H, CH2). 2.61 (dd, J=5.1, 12.3 Hz,
(4,5-Dimethoxy-2-nitro-benzyl)-pyridin-2-ylmethyl-pyridin-3-ylmethyl-
amine (2): 6-nitroveratraldehyde (113 mg, 0.54 mmol), DPA (107 mg,
0.54 mmol) and NaBHACHTUNGTRENNUNG(OAc)3 (172 mg, 0.81 mmol) were combined in
1,2-dichloroethane (10 mL). The resulting solution was stirred at 238C
for 18 h. Water (20 mL) was added and the product was extracted into
CH2Cl2 (3ꢁ10 mL). The organic layers were combined, dried over
MgSO4, and the solvent was removed. Flash chromatography on silica
(47:3 CH2Cl2/CH3OH) yielded 6 as a yellow oil (152 mg, 72.7%). TLC
Rf =0.76 (silica, 93:7:1 CH2Cl2/CH3OH/NH4OH); 1H NMR (400 MHz,
CDCl3): d = 8.47 (d, J=4.3 Hz, 2H, ArH), 7.57 (dd, J=11.5, 3.6 Hz, 3H,
ArH), 7.47 (s, 1H, ArH), 7.38 (d, J=7.8 Hz, 2H, ArH), 7.09–7.06 (t, J=
6.7 Hz, 2H, ArH), 4.09 (s, 2H, CH2), 3.95 (s, 4H, CH2), 3.84 (s, 3H,
CH3), 3.82 ppm (s, 3H, CH3); 13C NMR (100 MHz, CDCl3): d = 159.1,
153.1, 149.3, 147.5, 141.5, 136.5, 130.6, 123.2, 122.3, 112.7, 108.3, 60.6,
56.7, 56.5, 56 ppm; IR (thin film): n˜ = 3063.5, 3008.5, 2935.2, 2832.9,
1613.9, 1588.3, 1576.2, 1524.5, 1463.8, 1433.5, 1325.3, 1269.2, 1213.8,
1H, CH2), 2.1 ppm (s, 3H, CH3); 13C NMR (100 MHz, CDCl3): d
=
172.3, 159.1, 157.5, 149.5, 149.2, 149.1, 136.9, 136.7, 129.1, 123.1, 122.4,
122.3, 122.2, 122.0, 112.2, 111.2, 61.3, 57.0, 56.1, 44.7, 40.6 ppm; IR (thin
film): n˜ = 3012.1, 2924.5, 2853.4, 1652.9, 1521.4, 1506.3, 1457.1, 1436.5,
1374.4, 1339.0, 1277.6, 1220.2, 1174.9, 1094.6, 1056.4, 946.7, 842.9,
746.7 cmꢁ1; HRMS (ESI+): m/z: calcd for: 451.1856; found: 451.1861
[M+H+].
[2-(Bispyridin-2-ylmethylamino)-1-(3,4-dimethoxyphenyl)ethyl]-carbamic
acid tert-butyl ester (6): Aldehyde 5 (2.45 g, 8.30 mmol), DPA (1.65 g,
8.30 mmol), NaBHACTHNUTRGNEUNG(OAc)3 (2.78 g, 13.3 mmol) and AcOH (0.50 mL,
8.70 mmol) were combined in CH2Cl2 (50 mL). The reaction was stirred
at 238C for 24 h. Saturated Na2CO3 (30 mL) was added and the product
was extracted into CH2Cl2 (3ꢁ60 mL). The combined organic extracts
were dried over MgSO4 and the solvent was removed. Flash chromatog-
raphy on silica (24:1 CH2Cl2/CH3OH) yielded 6 as a brown oil (2.74 g,
69.1%). TLC Rf =0.67 (silica, 93:7:1 CH2Cl2/CH3OH/NH4OH); 1H NMR
(400 MHz, CDCl3): d = 8.46 (d, J=4.4 Hz, 2H, ArH), 7.47 (dt, J=7.7,
1.5 Hz, 2H, ArH), 7.22 (d, J=7.7 Hz, 2H, ArH), 7.03 (dd, J=5.1, 6.7 Hz,
2H, ArH), 6.88 (s, 1H, NH), 6.66 (dd, J=9.3, 5.1 Hz, 2H, ArH), 4.58 (s,
1H, CH), 3.83 (d, J=14.9 Hz, 2H, CH2), 3.73 (d, J=14.9 Hz, 2H, CH2),
3.79 (s, 3H, CH3), 3.75 (s, 3H, CH3), 2.84–2.75 (m, 2H, CH2), 1.42 ppm
1186.9, 1149.4, 1061.3, 1033.6, 993.2, 873.4, 794.5, 762.7, 730.3 cmꢁ1
;
HRMS (ESI+): m/z: calcd for: 394.1641; found: 394.1655 [M+H+].
[(4,5-Dimethoxy-2-nitro-benzyl)-ethoxycarbonylmethyl-amino]-acetic
acid ethyl ester (9): 6-Nitroveratraldehyde (602 mg, 2.85 mmol), iminodi-
acetic acid diethyl ester (500 mL, 2.85 mmol) and NaBH(OAc)3 (906 mg,
AHCTUNGTRENNUNG
4.27 mmol) were combined in 1,2-dichloroethane (50 mL). The resulting
solution was stirred at 238C for 18 h. Water (20 mL) was added and the
product was extracted into CH2Cl2 (3ꢁ10 mL). The organic layers were
combined, dried over MgSO4, and the solvent was removed. Flash chro-
matography on silica (24:1 CH2Cl2/CH3OH) yielded 9 as a yellow solid
(450 mg, 44.2%). TLC Rf =0.45 (1:1 CH2Cl2/hexanes); m.p. 62–648C;
1H NMR (400 MHz, CDCl3): d = 7.67 (s, 1H, ArH), 7.61 (s, 1H, ArH),
4.30 (s, 2H, CH2), 4.18 (q, J=7.1 Hz, 4H, CH2), 4.01 (s, 3H, CH3), 3.95
(s, 3H, CH3), 3.58 (s, 4H, CH2), 1.28 ppm (t, J=7.1 Hz, 6H, CH3);
13C NMR (100 MHz, D2O): d = 179.9, 153.2, 146.9, 141.2, 131.2, 113.2,
(s, 9H, CH3); 13C NMR (100 MHz, CDCl3): d
= 159.4, 156.2, 149.1,
149.0, 148.1, 136.5, 135.1, 123.1, 122.2, 118.5, 111.3, 109.8, 78.8, 60.5, 56.0,
55.8, 53.8, 28.6 ppm; IR (thin film): n˜ = 3357.2, 2972.9, 2931.7, 2833.9,
1703.3, 1590.9, 1513.3, 1464.3, 1364.2, 1252.7, 1163.8, 1144.5, 1025.6, 859.6,
760.4 cmꢁ1; HRMS (ESI+): m/z: calcd for: 478.2580; found: 478.2573
[M+H+].
108.7, 58.6, 56.6, 56.3, 54.9 ppm; IR (thin film): n˜
= 3066.6, 2978.4,
3938
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3932 – 3941

Cage Compounds
FULL PAPER
1-(3,4-Dimethoxyphenyl)-N’,N’-bispyridin-2-ylmethyl-ethane-1,2-diamine
(7): Trifluoroacetic acid (CF3CO2H, 8 mL) was added to a solution of 6
(2.00 g, 7.18 mmol) in CH2Cl2 (40 mL) and stirred at 238C for 24 h. After
removing the solvent, saturated aqueous K2CO3 (30 mL) was added to
the reaction mixture and the product was extracted into CH2Cl2 (3ꢁ
60 mL). The combined organic extracts were dried over MgSO4 and the
solvent was removed to yield 7 as a brown oil (2.18 g, 80.1%). TLC Rf =
0.36 (silica, 93:7:1 CH2Cl2/CH3OH/NH4OH); 1H NMR (400 MHz,
CDCl3): d = 9.42 (s, 2H, NH2), 8.46 (d, J=4.0 Hz, 2H, ArH), 7.64 (t, J=
7.4 Hz, 2H, ArH), 7.27 (d, J=7.4 Hz, 2H, ArH), 7.19 (t, J=6.0 Hz, 2H,
ArH), 7.05 (s, 1H, ArH), 6.89 (d, J=7.9 Hz, 1H, ArH), 6.70 (d, J=
8.3 Hz, 1H, ArH), 4.33 (d, J=8.8 Hz, 1H, CH), 4.07 (s, 4H, CH2), 3.75
(s, 3H, CH3), 3.69 (s, 3H, CH3), 3.46 (d, J=12.4 Hz, 1H, CH2), 3.12 ppm
(d, J=13.1 Hz, 1H, CH2); 13C NMR (100 MHz, CDCl3) 161.9, 161.5,
157.2, 149.5, 149.4, 147.0, 139.3, 127.7, 124.6, 123.4, 120.3, 118.4, 115.5,
111.4, 110.5, 60.2, 58.9, 56.0, 55.9, 54.3 ppm; IR (thin film): n˜ = 3057.0,
3009.0, 2939.3, 2841.0, 1671.1, 1595.7, 1519.7, 1466.3, 1426.1, 1265.0,
1197.5, 1175.1, 1125.3, 1023.9, 829.6, 798.1, 718.9 cmꢁ1; HRMS (ESI+):
m/z: calcd for: 378.2056; found: 378.2050 [M+H+].
film); n˜ = 3012.1, 2926.3, 2850.7, 1652.8, 1593.6, 1520.8, 1471.7, 1466.9,
1436.5, 1339.1, 1273.2, 1216.2, 1149.4, 1049.4, 795.1, 766.8 cmꢁ1; HRMS
(ESI+): m/z: calcd for: 605.2751; found: 605.2763 [M+H+].
General spectroscopic methods: All solutions were prepared with spec-
trophotometric grade solvents. HEPES (4-(2-hydroxyethyl)-1-piperazi-
neethanesulfonic acid) and KCl (99.5%) were purchased and used as re-
ceived. Zn2+ stock solutions were prepared from 99.999% pure ZnCl2.
All Zn2+ solutions were standardized by titrating with terpyridine and
measurement of the absorption spectra. ZinCleav-2 was introduced to
aqueous solution by addition of stock solution in DMSO (15.0 mm).
Graphs were manipulated and equations calculated by using Kaleida-
graph 4.0. The pH values of solutions were recorded with an Omega
PHB 212 glass electrode that was calibrated prior to each use. Absorp-
tion 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 258C, in
1 cm path length quartz cuvettes with a total volume of 3.0 mL. Fluores-
cence spectra were recorded on a Hitachi F-4500 spectrophotometer
under the control of a Pentium-IV PC running the FL Solutions 2.0 soft-
ware package. Excitation was provided by a 150 W Xe lamp (Ushio Inc.)
operating at a current of 5 A. Spectra were routinely acquired at 258C, in
1 cm quartz cuvette with a total volume of 3.0 mL using, unless otherwise
stated, 5 nm slit widths and a photomultiplier tube power of 700 V. Pho-
tolysis experiments were performed at 258C, in 1 cm path length quartz
cuvettes illuminated by a 150 W Xe lamp of a Hitachi F-4500 spectropho-
tometer with emission wavelength set to 350 nm or a Rayonet RPR-100
Photochemical Reactor containing 12 bulbs (900 W), each emitting at
3500 ꢂ. HPLC was performed on a Perkin–Elmer 250 binary pump
equipped with a diode array detector under the control of a Pentium-III
PC running the PeakSimple software package. An Ultra C18 column
(250ꢁ4.6 mm) was used for the separation of photoproducts and the
peaks were detected at 250 nm. A linear gradient from 50% A in B to
100% A was run at 0.8 mLminꢁ1 for 30 min, where A is 2% formic acid
in water and B is MeOH.
1-(3,4-Dimethoxyphenyl)-N,N,N’,N’-tetrakispyridin-2-ylmethyl-ethane-
1,2-diamine (8): Diamine 7 (2.00 g, 5.28 mmol) and 2-pyridinecarboxalde-
hyde (504 mL) were combined in EtOH (20 mL) and the solution was re-
fluxed at 788C for 1 h. The solution was allowed to cool to 238C and
NaBH4 (399 mg, 10.6 mmol) was added. The reaction mixture was stirred
at 238C for 24 h. After removing the solvent, saturated Na2CO3 (30 mL)
was added and the product was extracted in to CH2Cl2 (3ꢁ60 mL). The
combined organic extracts were dried over MgSO4 and the solvent was
removed. The crude product (2.24 g, 4.77 mmol) was dissolved in CH2Cl2
(40 mL) and treated with 2-pyridinecarboxaldehyde (455 mL, 4.77 mmol)
and NaBHACHTUNGTRENNUNG(OAc)3 (2.02 g, 9.54 mmol). The reaction was stirred at 238C
for 24 h. Saturated Na2CO3 (30 mL) was added and the product was ex-
tracted into CH2Cl2 (3ꢁ60 mL). The combined organic extracts were
dried over MgSO4 and the solvent was removed. Flash chromatography
on silica (97:2:1!93:6:1 CH2Cl2/CH3OH/NH4OH) yielded 8 as a brown
oil (1.52 g, 51.3%). TLC Rf =0.38 (silica, 93:7:1 CH2Cl2/CH3OH/
NH4OH); 1H NMR (400 MHz, CDCl3): d = 8.45–8.39 (m, 4H, ArH),
7.56–7.39 (m, 6H, ArH), 7.09–7.01 (m, 6H, ArH), 6.77–6.65 (m, 3H,
ArH), 3.97 (dd, J=8.7, 5.3 Hz, 1H, CH), 3.84 (s, 3H, CH3), 3.81 (d, J=
3.6 Hz, 2H, CH2), 3.76 (s, 2H, CH2), 3.74 (s, 3H, CH3), 3.66 (d, J=
4.6 Hz, 2H, CH2), 3.53 (d, J=14.9 Hz, 2H, CH2), 3.17 (dd, J=13.0,
5.2 Hz, 1H, CH2), 3.07 ppm (dd, J=13.0, 9.0 Hz, 1H, CH2); 13C NMR
(100 MHz, CDCl3): d = 160.5, 159.8, 159.6, 149.4, 149.2, 145.0, 148.9,
148.6, 148.3, 136.5, 136.4, 136.3, 131.0, 123.3, 123.2, 122.9, 122.5, 122.0,
121.9, 112.8, 111.1, 110.8, 110.6, 62.2, 61.1, 61.0, 57.2, 56.8, 56.1, 56.05,
Determination of Zn2+ binding constant
Calibration curve: A 10 mm solution of the fluorescent Zn2+ sensor 11
(25 mm HEPES, 100 mm NaClO4, pH 7.4) was prepared. A 3.0 mL ali-
quot of this solution was placed in a quartz cuvette and fluorescence was
measured. Ten 1.5 mL aliquots of a 1.0 mm ZnACHTNUTRGNE(UNG ClO4)2 stock solution was
added to the above solution and fluorescence was measured after each
addition. Fluorescence was corrected for dilution, normalized and plotted
against total Zn2+ concentration to generate a calibration curve.
Competitive titration with ZinCleav-2: To the above solution, 2.0 mL ali-
quots of a 15 mm ZinCleav-2 solution were added. Fluorescence of the
solution was measured after each addition. No further decrease in fluo-
rescence was observed after adding 30 equivalents (300 mm) of ZinCleav-
2. Fluorescence was corrected for dilution and normalized. The above
procedure was repeated three times. Under the experimental conditions
99.99% of Zn2+ is bound to the sensor at the beginning of the titration.
The binding equilibrium for this system may be expressed by Equa-
tion (1). The binding constant of the [Zn(ZinCleav-2)]2+ complex
(K’ZinCleav-2) is obtained by solving Equation (2), where K’ZnSensor-1 =2.2
56.0 ppm; IR (thin film): n˜
= 3060.6, 3006.3, 2932.9, 2832.8, 1589.2,
1560.7, 1513.3, 1471.4, 1432.4, 1364.0, 1258.1, 1234.3, 1143.9, 1026.7, 994.5,
809.3, 760.1, 730.6 cmꢁ1; HRMS (ESI+): m/z: calcd for: 560.2900; found:
560.2913 [M+H+].
1-(4,5-Dimethoxy-2-nitrophenyl)-N,N,N’,N’-tetrakispyridin-2-ylmethyl-
ethane-1,2-diamine (ZinCleav-2, 1): Trifluoromethanesulfonic acid
(457 mL, 5.18 mmol) and fuming nitric acid (91.5%, 118 mL, 2.59 mmol)
were combined in CH2Cl2 (10 mL) and stirred vigorously at 238C for
30 min. The mixture was cooled to ꢁ428C using a CH3CN/dry ice bath.
A solution of the diamine 8 (207 mg, 0.37 mmol) dissolved in CH2Cl2
(2 mL) was added to the above mixture in one portion. The reaction mix-
ture was stirred at ꢁ428C for 10 h. Saturated Na2CO3 (30 mL) was added
and the product was extracted in to CH2Cl2 (3ꢁ20 mL). The combined
organic extracts were dried over MgSO4 and the solvent was removed.
Flash chromatography on silica (91:8:1 CH2Cl2/CH3OH/NH4OH) yielded
ZinCleav-2 as a yellow oil (32 mg, 14.4%). TLC Rf =0.32 (silica, 24:1
CH2Cl2:CH3OH); 1H NMR (400 MHz, CDCl3): d = 8.50–8.345 (m, 4H,
ArH), 7.52–7.49 (m, 5H, ArH), 7.41–7.79 (m, 2H, ArH), 7.09–7.01 (m,
7H, ArH), 4.79 (d, J=9.0 Hz, 1H, CH), 3.84 (s, 3H. CH3), 3.81 (d, J=
3.6 Hz, 2H, CH2), 3.77 (s, 2H, CH2), 3.74 (s, 3H, CH3), 3.67 (d, J=
14.1 Hz, 2H, CH2), 3.53 (d, J=14.1 Hz, 2H, CH2), 3.18 (dd, J=13.0,
6.0 Hz, 1H, CH2), 3.06 ppm (dd, J=13.0, 9.0 Hz, 1H, CH2); 13C NMR
(100 MHz, CDCl3): d = 162.5, 159.8, 159.6, 150.4, 152.4, 149.2, 145.8,
149.9, 148.6, 148.3, 136.5, 136.4, 131.0, 123.9, 122.9, 122.0, 121.9, 113.8,
112.1, 110.9, 110.7, 62.2, 61.1, 62.0, 58.2, 56.8, 56.1, 56.0 ppm; IR (thin
ꢁ1015 mꢁ1
.
2þ
ZinCleav þ ½ZnðZnSensorÞꢃ ¼ ½ZnðZinCleavÞꢃ2þ þ ZnSensor
ð1Þ
ð2Þ
½ZnðZinCleavÞꢃ ½ZnSensorꢃ KZ0 inCleav
2þ
¼
KZ0 nSensor
2þ
½ZinCleavꢃ½ZnðZnSensorÞꢃ
Equations (3)–(5) were used to calculate the individual components of
Equation (2). [Zn(ZnSensor-1)] was determined from the calibration
curve.
2þ
ð3Þ
ð4Þ
ð5Þ
½ZnSensorꢃ ¼ ½ZnSensorꢃtotal ꢁ ½ZnðZnSensorÞꢃ
2þ
2þ
2þ
½ZnðZinCleavÞꢃ ¼ ½ZnðZnSensorÞꢃ ꢁ ½ZnðZnSensorÞꢃ
initial
½ZinCleavꢃ ¼ ½ZinCleavꢃtotal ꢁ ½ZnðZinCleavÞꢃ
Chem. Eur. J. 2011, 17, 3932 – 3941
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3939

S. C. Burdette et al.
Quantum efficiency of photolysis
[8] J. W. Walker, J. A. McCray, G. P. Hess, Biochemistry 1986, 25, 1799–
1805.
Calibration curve: A 1.2 mm [Zn(ZinCleav-2)]2+ stock solution was pre-
pared. The stock solution was diluted to make 10, 30, 40, 50 and 60 mm
[Zn(ZinCleav-2)]2+ solutions. HPLC traces of each solution were record-
ed and the area under the curve was determined by integration.
Photolysis of [Zn(ZinCleav-2)]2+: A 2.0 mL aliquot of a 60 mm solution
was irradiated at 350 nm using a 150 W Xe lamp. HPLC traces of the
above solution were recorded after irradiating for 60, 90 and 120 min.
The area of the [Zn(ZinCleav-2)]2+ peak was obtained by integration
and the concentration of remaining [Zn(ZinCleav-2)]2+ was determined
using the calibration curve. The quantum efficiency of photolysis of Zin-
Cleav-2 was obtained by solving equation 6 where NA is the Avogadroꢃs
number.
[9] T. Milburn, N. Matsubara, A. P. Billington, J. B. Udgaonkar, J. W.
Walker, B. K. Carpenter, W. W. Webb, J. Marque, W. Denk, J. A.
McCray, G. P. Hess, Biochemistry 1989, 28, 49–55.
[10] R. Wieboldt, K. R. Gee, L. Niu, D. Ramesh, B. K. Carpenter, G. P.
Hess, Proc. Natl. Acad. Sci. USA 1994, 91, 8752–8756.
[11] J. W. Walker, A. V. Somlyo, Y. E. Goldman, A. P. Somlyo, D. R.
Trentham, Nature 1987, 327, 249–252.
[12] J. W. Walker, J. Feeney, D. R. Trentham, Biochemistry 1989, 28,
3272–3280.
[13] H. Ando, T. Furuta, R. Y. Tsien, H. Okamoto, Nat. Genet. 2001, 28,
317–325.
[14] W. T. Monroe, M. M. McQuain, M. S. Chang, J. S. Alexander, F. R.
Haselton, J. Biol. Chem. 1999, 274, 20895–20900.
[15] J. W. Walker, S. H. Gilbert, R. M. Drummond, M. Yamada, R. Sree-
kumar, R. E. Carraway, M. Ikebe, F. S. Fay, Proc. Natl. Acad. Sci.
USA 1998, 95, 1568–1573.
[16] D. M. Rothman, E. J. Petersson, M. E. Vazquez, G. S. Brandt, D. A.
Dougherty, B. Imperiali, J. Am. Chem. Soc. 2005, 127, 846–847.
[17] G. Marriott, Biochemistry 1994, 33, 9092–9097.
[18] D. Mendel, J. A. Ellman, P. G. Schultz, J. Am. Chem. Soc. 1991, 113,
2758–2760.
[19] M. Ghosh, X. Song, G. Mouneimne, M. Sidani, D. S. Lawrence, J. S.
Condeelis, Science 2004, 304, 743–746.
[20] N. Gagey, P. Neveu, L. Jullien, Angew. Chem. 2007, 119, 2519–2521;
Angew. Chem. Int. Ed. 2007, 46, 2467–2469.
[21] M. J. Davis, C. H. Kragor, K. G. Reddie, H. C. Wilson, Y. Zhu, T. M.
Dore, J. Org. Chem. 2009, 74, 1721–1729.
[22] R. Y. Tsien, R. S. Zucker, Biophys. J. 1986, 50, 843–853.
[23] S. R. Adams, J. P. Y. Kao, G. Grynkiewicz, A. Minta, R. Y. Tsien, J.
Am. Chem. Soc. 1988, 110, 3212–3220.
[24] D. P. Kennedy, C. Gwizdala, S. C. Burdette, Org. Lett. 2009, 11,
2587–2590.
[25] D. P. Kennedy, C. D. Incarvito, S. C. Burdette, Inorg. Chem. 2010,
49, 916–923.
[26] C. Gwizdala, D. P. Kennedy, S. C. Burdette, Chem. Commun. 2009,
6967–6969.
[27] G. C. R. Ellis-Davies, J. H. Kaplan, Proc. Natl. Acad. Sci. USA 1994,
91, 187–191.
change in ½ZinCleav ꢁ 2ꢃ=irradiation time
ð6Þ
quantum efficiency ¼
NA
intensity of the source
The intensity of the source was determined as described previously.[31]
The quantum yield of photolysis of compounds 2, 3 and 4 were deter-
mined by UV-visible spectroscopy as described previously.[31]
Cu/Zn exchange: A 2.0 mL aliquot of a 60 mm solution of ZinCleav-2
(40 mm HEPES, 100 mm KCl, pH 7.4) was placed in a quartz cuvette. A
24 mL aliquot of a CuCl2 stock solution (50 mm) was added to give a final
Cu2+ concentration of 600 mm, and the absorption spectrum was record-
ed. Another 2.0 mL aliquot of a 60 mm solution of ZinCleav-2 (40 mm
HEPES, 100 mm KCl, pH 7.4) was placed in a quartz cuvette and a 12 mL
aliquot of a ZnCl2 stock solution (10 mm) was added to give final Zn2+
concentration of 60 mm.
A 24 mL aliquot of a CuCl2 stock solution
(50 mm) was added to give a final Cu2+ concentration of 600 mm and the
absorption spectrum was recorded. Absorption spectra were recorded at
5 min intervals for 30 min.
Ca/Zn exchange: A 2.0 mL aliquot of a 60 mm solution of ZinCleav-2
(40 mm HEPES, 100 mm KCl, pH 7.4) was placed in a quartz cuvette. A
24 mL aliquot of a CaCl2 stock solution (0.50m) was added to give a final
Ca2+ concentration of 6.0 mm, and the absorption spectrum was record-
ed. A 12 mL aliquot of a ZnCl2 stock solution (10 mm) was added to give
final Zn2+ concentration of 60 mm and the absorption spectrum was re-
corded.
Release of caged Zn2+: A 2.0 mL aliquot of a 5.0 mm ZinPyr-1 solution
(40 mm HEPES, 100 mm KCl, pH 7.4) was placed in a quartz cuvette and
the fluorescence spectrum was recorded. A 66 mL aliquot of a ZnCl2
stock solution (9.02 mm) was added to give a final concentration of 30 mm
and the fluorescence spectrum was recorded. A 40 mL aliquot of the Zin-
Cleav-2 stock solution (2.0 mm) was added to give a final concentration
of 40 mm and the fluorescence spectrum was recorded. The cuvette was ir-
radiated in a Rayonet RPR-100 photoreactor with 350 nm wavelength ra-
diation and the fluorescence spectra were recorded at 1 min intervals.
[28] G. C. R. Ellis-Davies, J. H. Kaplan, J. Org. Chem. 1988, 53, 1966–
1969.
[29] G. C. R. Ellis-Davies, Cell Calcium 2006, 39, 471–473.
[30] K. L. Ciesienski, K. L. Haas, M. G. Dickens, Y. T. Tesema, K. J.
Franz, J. Am. Chem. Soc. 2008, 130, 12246–12247.
[31] H. M. D. Bandara, D. P. Kennedy, E. Akin, C. D. Incarvito, S. C.
Burdette, Inorg. Chem. 2009, 48, 8445–8455.
[32] C. Jacob, W. Maret, B. L. Vallee, Proc. Natl. Acad. Sci. USA 1998,
95, 3489–3494.
[33] C. E. Outten, T. Y. OꢃHalloran, Science 2001, 292, 2488–2492.
[34] M. S. Nasir, C. J. Fahrni, D. A. Suhy, K. J. Kolodsick, C. P. Singer,
T. V. OꢃHalloran, J. Biol. Inorg. Chem. 1999, 4, 775–783.
[35] S. C. Paski, Z. Xu, J. Nutr. Biochem. 2001, 12, 655–661.
[36] M. St Croix Claudette, K. J. Wasserloos, K. E. Dineley, I. J. Rey-
nolds, E. S. Levitan, B. R. Pitt, Am. J. Physiol. 2002, 282, L185 L192.
[37] J. Cao, J. A. Bobo, J. P. Liuzzi, R. J. Cousins, J. Leukocyte Biol. 2001,
70, 559–566.
Acknowledgements
This work was supported by the University of Connecticut, the Universi-
ty of Connecticut Research Foundation grant PD09-0154 and by NSF
grant CHE-0955361.
[38] Y. Mikata, M. Wakamatsu, A. Kawamura, N. Yamanaka, S. Yano,
A. Odani, K. Morihiro, S. Tamotsu, Inorg. Chem. 2006, 45, 9262–
9268.
[39] Y. Mikata, M. Wakamatsu, S. Yano, Dalton Trans. 2005, 545–550.
[40] Y. Mikata, A. Yamanaka, A. Yamashita, S. Yano, Inorg. Chem.
2008, 47, 7295–7301.
[1] J. Hꢄbert, D. Gravel, Can. J. Chem. 1974, 52, 187–189.
[2] A. J. Kirby, G. A. Varvoglis, Chem. Commun. 1967, 406.
[3] A. Patchornik, B. Amit, R. B. Woodward, J. Am. Chem. Soc. 1970,
92, 6333–6335.
[41] K. Hanaoka, Y. Muramatsu, Y. Urano, T. Terai, T. Nagano, Chem.
Eur. J. 2010, 16, 568–572.
[42] J.-M. Lee, Y.-J. Kim, H. Ra, S.-J. Kang, S. Han, J.-Y. Koh, Y.-H.
Kim, FEBS Lett. 2008, 582, 1871–1876.
[4] B. Amit, U. Zehavi, A. Patchornik, J. Org. Chem. 1974, 39, 192–196.
[5] J. Engels, E. J. Schlaeger, J. Med. Chem. 1977, 20, 907–911.
[6] U. Zehavi, B. Amit, A. Patchornik, J. Org. Chem. 1972, 37, 2281–
2285.
[7] J. H. Kaplan, B. Forbush III, J. F. Hoffman, Biochemistry 1978, 17,
1929–1935.
[43] H.-H. Wang, Q. Gan, X.-J. Wang, L. Xue, S.-H. Liu, H. Jiang, Org.
Lett. 2007, 9, 4995–4998.
3940
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3932 – 3941

Cage Compounds
FULL PAPER
[44] D. W. Gruenwedel, Inorg. Chem. 1968, 7, 495–501.
[45] J. H. Kaplan, G. C. R. Ellis-Davies, Proc. Natl. Acad. Sci. USA 1988,
85, 6571–6575.
[46] S. C. Burdette, G. K. Walkup, B. Spingler, R. Y. Tsien, S. J. Lippard,
J. Am. Chem. Soc. 2001, 123, 7831–7841.
[51] R. A. Bozym, R. B. Thompson, A. K. Stoddard, C. A. Fierke, ACS
Chem. Biol. 2006, 1, 103–111.
[52] S. Overbeck, P. Uciechowski, M. L. Ackland, D. Ford, L. Rink, J.
Leukocyte Biol. 2007, 83, 368–380.
[53] A. Krezel, W. Maret, J. Biol. Inorg. Chem. 2006, 11, 1049–1062.
[54] T. J. B. Simons, J. Membr. Biol. 1991, 123, 63–71.
[55] L. A. Finney, T. V. OꢃHalloran, Science 2003, 300, 931–936.
[56] L. Alderighi, P. Gans, A. Ienco, D. Peters, A. Sabatini, A. Vacca,
Coord. Chem. Rev. 1999, 184, 311–318.
[47] B. A. Wong, S. Friedle, S. J. Lippard, J. Am. Chem. Soc. 2009, 131,
7142–7152.
[48] J. E. T. Corrie, A. Barth, V. R. N. Munasinghe, D. R. Trentham,
M. C. Hutter, J. Am. Chem. Soc. 2003, 125, 8546–8554.
[49] M. Gaplovsky, Y. V. Il’ichev, Y. Kamdzhilov, S. V. Kombarova, M.
Mac, M. A. Schworer, J. Wirz, Photochem. Photobiol. Sci. 2005, 4,
33–42.
[57] G. C. R. Ellis-Davies, R. J. Barsotti, Cell Calcium 2006, 39, 75–83.
Received: July 13, 2010
Revised: October 26, 2010
Published online: March 1, 2011
[50] B. E. Cohen, B. L. Stoddard, D. E. Koshland, Biochemistry 1997, 36,
9035–9044.
Chem. Eur. J. 2011, 17, 3932 – 3941
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3941