Methods for preparing metal ion photocages: Application to the synthesis of crowncast

Article abstract of DOI:10.1021/ol900902z

Three different synthetic strategies were utilized in the construction of a novel class of macrocyclic containing o-nitrobenzhydrol group II cation cages. The synthetic methodology presented herein is unparalleled in scope toward the preparation of caged

Full text of DOI:10.1021/ol900902z

ORGANIC
LETTERS
2009
Vol. 11, No. 12
2587-2590
Methods for Preparing Metal Ion
Photocages: Application to the
Synthesis of CrownCast
Daniel P. Kennedy, Celina Gwizdala, and Shawn C. Burdette*
Department of Chemisty, UniVersity of Connecticut, 55 North EagleVille Road U-3060,
Storrs, Connecticut 06269
shawn.burdette@uconn.edu
Received April 24, 2009
ABSTRACT
Three different synthetic strategies were utilized in the construction of a novel class of macrocyclic containing o-nitrobenzhydrol group II
cation cages. The synthetic methodology presented herein is unparalleled in scope toward the preparation of caged complexes for various
main group and transition block cations.
Caged molecules have been used to study the cellular
signaling pathways of different molecular or ionic species.1
In caged molecules, the biological activity of the analyte is
masked by a photochemically labile bond.2 Upon irradiation,
the caged molecule or metal ion is released. Only a few cages
for divalent metal ions including Ca,3-5 Mg,6 and Cu7 have
been reported to date.
bound cation after photolysis. Upon irradiation, CrownCast
undergoes a Norrish-type II photochemical reaction to afford
the corresponding nitrosobenzophenone.8 The resulting pho-
toproduct possesses a reduced binding affinity for the cation
because of delocalization of the anilino lone pair onto the
benzophenone carbonyl oxygen atom.
Initially, the CrownCast analogue 3 was prepared by the
strategy that was utilized by Tsien et al. in the construction
of Ca cages (Scheme 1, Method A).3 The macrocycle
N-phenyl-1-aza-15-crown-5 (1) was prepared as reported9
and reacted with trimethylsilyl triflate (TMSOTf) and 2
affording the desired benzhydrol as the trimethylsilyl (TMS)
ether. The TMS ether was subsequently reacted in situ with
tetrabutylammonium fluoride (TBAF) to afford 3 in 33%
yield. Since the isolated yields of the desired compound 3
was poor and TMSOTf has potential incompatibility with
other functional groups, alternative methods for the construc-
tion of CrownCast ligands were explored to facilitate future
syntheses.
To facilitate the development of caged metal complexes,
we have developed two new synthetic approaches for
accessing nitrobenzhydrol derived cages. We have demon-
strated the utility of each method by preparing CrownCast,
a cage that utilizes a macrocyclic NO4 ligand as the cation
receptor. The name CrownCast is derived from the crown
ether receptor and the ability of the compound to cast off a
(1) Kaplan, J. H.; Forbush, B., III; Hoffman, J. F. Biochemistry 1978,
17, 1929–1935.
(2) Kaplan, J. H.; Somlyo, A. P. Trends NeuroSci. 1989, 12, 54–59.
(3) Adams, S. R.; Kao, J. P. Y.; Grynkiewicz, G.; Minta, A.; Tsien,
R. Y. J. Am. Chem. Soc. 1988, 110, 3212–3220
.
(4) Ellis-Davies, G. C. R.; Barsotti, R. J. Cell Calcium 2006, 39, 75–
83.
(5) Ellis-Davies, G. C. R. Chem. ReV. 2008, 108, 1603–1613
(6) Ellis-Davies, G. C. R. Cell Calcium 2006, 39, 471–473.
(7) Ciesienski, K. L.; Haas, K. L.; Dickens, M. G.; Tesema, Y. T.; Franz,
K. J. J. Am. Chem. Soc. 2008, 130, 12246–12247.
.
(8) Bochet, C. G. J. Chem. Soc., Perkin Trans. 1 2002, 125–142.
(9) Lu, X.-X.; Qin, S.-Y.; Zhou, Z.-Y.; Yam, V. W.-W. Inorg. Chim.
Acta 2003, 346, 49–56.
10.1021/ol900902z CCC: $40.75
Published on Web 05/19/2009
 2009 American Chemical Society

Scheme 1. Synthesis of CrownCast
In the first alternative route, a Pd-catalyzed cross coupling
between 5 and 1-aza-15-crown-5 was utilized to assemble
the aryl macrocycle (Scheme 1, method B). Compound 4,
which was prepared by reported methods,10 underwent a
regioselective nitration to afford 5 in 61% yield. The
macrocycle 1-aza-15-crown-5 was incorporated utilizing a
Buchwald-Hartwig coupling using catalytic Pd(OAc)2 and
triphenylphosphine (PPh3).11,12 The resulting benzophenone
6 was reduced with NaBH4 to afford CrownCast (7) in 12%
yield. Alternative methods for the reduction of compound 6
to CrownCast resulted in the formation of several byproducts
as evidenced by TLC analysis. While borohydride reduction
is not ideal for CrownCast, we have successfully applied
this method to other targets.
The highest isolated yields of CrownCast were obtained
using method C (Scheme 1). Macrocycle 1 was brominated
with the unique reagent {[K(18-crown-6)]Br3}n,13 which
afforded 8 in nearly quantitative yield. Compound 8 was
converted into the corresponding aryl iodide, which is a better
substrate in the subsequent palladium-mediated borylation
chemistry.14 Palladium-catalyzed borylation with bis(pina-
colato)diboron provided 9 in 72% yield. The subsequent
palladium-catalyzed reaction between 9 and 6-nitroveratral-
dehyde afforded CrownCast in 71% yield. Rigorous exclu-
sion of O2 and H2O from the reaction mixture was not
necessary. Generally, palladium-catalyzed 1,2-additions of
arylboronic acids into aldehydes results in poor isolated
yields;15 however, tris(1-naphthyl)phosphine (P(1-NAP)3)
facilitates the PdCl2-catalyzed 1,2-addition of phenylboronic
acid to aldehyde substrates.16 Although the original report
focused on boronic acids as the reactive substrates, the
reaction proceeds cleanly with the borate ester. This appears
to be the first example where a pinacolato borate is used as
the substrate instead of the boronic acid, which is noteworthy
since the steric bulk of the reactive components does not
inhibit the formation of the desired product. This step not
only eliminates the pinacolato borate cleavage step that
requires harsh conditions17 but also provides a reagent that
is easier to handle and purify than boronic acids.
The 1:1 binding of CrownCast to several alkaline earth
metals Zn2+ and Cd2+ was evaluated in MeCN using
absorbance spectroscopy.18,19 Upon metal binding, the
intensity of the charge-transfer transition of apo-CrownCast
(15) Yamamoto, T.; Ohta, T.; Ito, Y. Org. Lett. 2005, 7, 4153–4155.
(16) Qin, C.; Wu, H.; Cheng, J.; Chen, X. a.; Liu, M.; Zhang, W.; Su,
W.; Ding, J. J. Org. Chem. 2007, 72, 4102–4107.
(10) Kim, B.-K.; Deng, J.; Xiao, W.; Van Gemert, B.; Chopra, A.;
Molock, F.; Mahadevan, S. Preparation of photochromic materials having
extended π-conjugated systems for ophthalmic devices: 2005102047,
2006226402, 20050408, 2006.
(17) Coutts, S. J.; Adams, J.; Krolikowski, D.; Snow, R. J. Tetrahedron
Lett. 1994, 35, 5109–5112.
(18) Jimenez, D.; Martinez-Manez, R.; Sancenon, F.; Ros-Lis, J. V.;
Soto, J.; Benito, A.; Garcia-Breijo, E. Eur. J. Inorg. Chem. 2005, 239, 3–
(11) Witulski, B. Synlett 1999, 1223–1226
.
(12) Schlummer, B.; Scholz, U. AdV. Synth. Catal. 2004, 346, 1599–
2403.
1626
.
(19) Bourson, J.; Pouget, J.; Valeur, B. J. Phys. Chem. 1993, 97, 4552–
(13) Zolfigol, M. A.; Chehardoli, G.; Salehzadeh, S.; Adams, H.; Ward,
M. D. Tetrahedron Lett. 2007, 48, 7969–7973.
(14) Ishiyama, T.; Murata, M.; Miyaura, N. J. Org. Chem. 1995, 60,
7508–7510.
4557
.
(20) Qin, W.; Baruah, M.; Sliwa, M.; Van der Auweraer, M.; De
Borggraeve, W. M.; Beljonne, D.; Van Averbeke, B.; Boens, N. J. Phys.
Chem. A 2008, 112, 6104–6114
.
2588
Org. Lett., Vol. 11, No. 12, 2009

Table 1. Binding Properties of CrownCast and UNC in MeCNa
cation
ionic radius (Å)
Kd′ CrownCast/µM
∆λmax CrownCast (nm)
Kd′ UNC/mM
∆λmax UNC (nm)
∆logKf′
Mg2+
Ca2+
Sr2+
Ba2+
Zn2+
Cd2+
0.86
1.14
1.32
1.49
0.88
1.09
64 ( 7
14 ( 2
48 ( 1
79 ( 3
161 ( 8
68 ( 1
-40
-32
-28
-25
-42
-42
13.1 ( 0.7
0.6 ( 0.1
2.2 ( 0.2
3.1 ( 0.1
NAb
-75
-61
-56
-50
-2.35
-1.64
-1.66
-1.60
NAb
a Binding assays were carried out between the respective metal perchlorate and CrownCast (25 µM) or UNC (20 µM) in MeCN. b Minimal changes were
observed in the absorbance spectra (<5% at 347 nm) when UNC was saturated with >5000 equivs of Zn2+ or Cd2+. The ∆log Kf′ value represents the change
in binding affinity in going from CrownCast to UNC.
that occurs at 269 nm was attenuated with the concomitant
development of a blue-shifted band at ∼220-245 nm (Table
1, Figure 1). CrownCast has a slight binding preference for
the metal complex as compared to Ca2+. Overall, the weakest
binding was observed with Zn2+, where a combination of
poor size match and employment of nonideal oxygen donors
likely accounts for the weaker binding.
After 4 h of irradiation with a high intensity source (1000
W Xe lamp), the photoproduct UNC was produced on a
preparative scale (Scheme 2). The binding constants of UNC
Scheme 2. Uncaging of Apo-CrownCast
Figure 1. Titration of 25 µM CrownCast with Mg(ClO4)2 in MeCN.
The inset describes the 1:1 binding isotherm with error bars
calculated from three trials.
Ca2+ over the other divalent cations tested, which is
consistent with log Kf′ values measured for other 15-crown-5
derivatives.20 The strength of the binding of each metal ion
appears to be determined by a combination of three factors:
the strength of the aniline-metal bond, the fit between the
macrocyclic cavity and the cation, and hard-soft acid-base
(HSAB) interactions.
The strength of the resulting N-M bond is proportional
to the energy of the blue-shifted CT band that occurs upon
metal binding. Mg2+, Zn2+, and Cd2+ cause the greatest
shifting in the CT band (Table 1). Since Zn2+ and Cd2+ are
moderate Lewis acids, interactions with the borderline aniline
base is predicted to be stronger than with alkali earth metals.
Despite being a hard Lewis acid, Mg2+ induces a large blue
shift in the CT transition, which suggests that electrostatic
interactions factor into its binding. The ionic potential of
Mg2+ is the highest of the alkaline earth metals tested since
the charge to radius ratio (q/r) decreases down the group in
the periodic table. In addition to HSAB factors, metal ion
binding with CrownCast becomes weaker as the size of the
cation increases. Poor size match between the macrocycle
and Mg2+ likely accounts for the slight destabilization of
were determined without interference from CrownCast. The
trend in binding selectivity was maintained, but the overall
strength of cation binding was attenuated ∼40-200 times.
The largest changes in log Kf′ were seen for Mg2+, Zn2+,
and Cd2+, where conjugation of the aniline lone pair into
the resulting benzophenone fragment substantially weakens
the N-M interaction in the metal complex. The binding of
UNC to Zn2+ and Cd2+ is too weak to accurately determine
the Kd′.
The quantum yield of photolysis was measured by pho-
tolyzing MeCN solutions of CrownCast and its Ca2+ and
Zn2+ complexes at 350 nm with a 150 W source (Figure 2,
Table 2). The measured quantum yields (ca. 1%) were in
accord with literature values for other nitrobenzhydrol Ca2+
cages.5 A slight increase in the quantum yield was observed
for the Ca2+ and Zn2+ complexes, likewise in agreement with
literature reports of the Ca2+ cage Nitr-5.3 Photolysis was
carried out using concentrations of M2+ where binding to
the photoproduct UNC was known from the titration data to
be negligible. This allowed for direct spectrophotometric
measurements of UNC concentrations over the course of the
photolysis.
Org. Lett., Vol. 11, No. 12, 2009
2589

Table 2. Photophysical Data for Ligands and Complexesa
species
λmax/nm (ε cm-1 M-1
)
Φphotolysis
CrownCast
269 (26300), 347 (5300)
347 (46000)
244 (17100), 347 (5300)
245 (14500), 347 (5300)
0.005 ( 0.002
UNC
[Ca(CrownCast)]2+
[Zn(CrownCast)]2+
0.007 ( 0.002
0.016 ( 0.006
a CrownCast (25 µM) was photolyzed in MeCN with a 150 W Xe source
at 350 nm as the apo-cage and in the presence of Ca(ClO4)2 (125 µM, 5.0
equiv) and Zn(ClO4)2 (1.25 mM, 50 equiv), where CrownCast was saturated
but binding to UNC is negligible.
preparative strategy, the observed trends in metal ion affinity
between the caged and uncaged forms of CrownCast will
facilitate the selection and rational design of appropriate
ligands for caging specific metal analytes.
Figure 2. Photolysis of 25 µM CrownCast in MeCN at 350 nm
using 150 W source.
In conclusion, we have demonstrated a modular approach
toward the construction of nitrobenzhydrol derived cages that
utilize a NO4 macrocyclic receptor. In particular, the Pd
coupling between the pinacolato borate and a nitroaldehyde
provided CrownCast in yields that were in excess of twice
the traditional TMSOTf methods. We also have developed
a third route based on Pd-catalyzed aryl amination. Although
CrownCast does not interact strongly with metal ions in
aqueous solution, our new synthetic methodology provides
a roadmap for the future construction of metal ion photocages
that will be employed in biological studies. In addition to
Acknowledgment. We thank Dr. D. Niedzwiedzki and
Professor H. A. Frank of the University of Connecticut for
help with the 1000 W source. This work was funded by the
University of Connecticut.
Supporting Information Available: Detailed synthetic
procedures and 1H and 13C NMR spectra for all compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL900902Z
2590
Org. Lett., Vol. 11, No. 12, 2009