Design and synthesis of a fluorescent probe for Zn2+, 5,7-Bis(N,N-dimethylaminosulfonyl)-8-hydroxyquinoline-Pendant 1,4,7,10-Tetraazacyclododecane and Zn2+-Dependent Hydrolytic and Zn2+-independent photochemical reactivation of its benzenesulfonyl-caged derivative

Article abstract of DOI:10.1021/ic901279t

We previously reported on a 8-quinolinol-pendant cyclen (L⁵) as a Zn²⁺ fluorophore (cyclen = 1,4,7,10-tetraazacyclododecane) and its caged derivative, 8-(benzenesulfonyloxy)-5-(N,N-dimethylaminosulfonyl)quinolin- 2ylmethyl-pendant cyclen (BS-caged-L⁵), which can be reactivated by hydrolysis of benzenesulfonyl group upon complexation with Zn²⁺ at neutral pH to give a 1:1 Zn²⁺-L⁵ complex (Zn(H _-1L⁵)). We report herein on the synthesis of 5,7-bis(N,N-dimethylaminosulfonyl)-8-hydroxyquinolin-2-ylmethyl-pendant cyclen (L⁶) and its caged derivative (BS-caged-L⁶) for more sensitive and more efficient cell-membrane permeability than those of L ⁵ and BS-caged-L⁵. By Potentiometrie pH, ¹H NMR, and UV-vis spectroscopic titrations, the deprotonation constants pK _a1-pK_aB of H₅L⁶ were determined to be <2, <2, <2,2.5 ± 0.1 (for the 8-OH group of the quinoline moiety), 9.7 ± 0.1, and 10.8 ± 0.1 at 25 °C with l=0.1 (NaNO₃). The results of ¹H NMR, Potentiometrie pH, UV-vis, and fluorescent titrations showed that L⁶ rapidly forms a 1:1 complex with Zn²⁺ (Zn(H_-1L⁶)), the dissociation constant of which is 50 fM at pH 7.4. The fluorescent emission of Zn(H _-1L⁶) at 478 nm is 32 times as large as that of L6 (excitation at 370 nm), and the fluorescent quantum yield of Zn(H _-1L⁶) (_F = 0.41) is much greater than that of Zn(H_-1L⁵) (4>F = 0.044). The BScaged-L ⁶ was reactivated by hydrolysis of the benzenesulfonyl moiety more rapidly (completes in 30 min at pH 7.4 at 37 °C) than BS-caged-L ⁵, presumably enabling the practical detection of Zn²⁺ in sample solutions and living cells. The photochemical deprotection of BS-caged-L⁶ and the cell membrane permeability of L⁶ and BS-caged-L⁶ are also described.

Full text of DOI:10.1021/ic901279t

888 Inorg. Chem. 2010, 49, 888–899
DOI: 10.1021/ic901279t
Design and Synthesis of a Fluorescent Probe for Zn2þ,
5,7-Bis(N,N-dimethylaminosulfonyl)-8-hydroxyquinoline-Pendant
1,4,7,10-Tetraazacyclododecane and Zn2þ-Dependent Hydrolytic and
Zn2þ-Independent Photochemical Reactivation of Its Benzenesulfonyl-Caged
Derivative
Ryosuke Ohshima,† Masanori Kitamura,†,‡ Akinori Morita,§ Motoo Shiro, Yasuyuki Yamada,† Masahiko Ikekita,§
Eiichi Kimura,^ and Shin Aoki*,†,‡
†Faculty of Pharmaceutical Sciences, Tokyo University of Science, 2641 Yamazaki, Noda 278-8510, Japan,
‡Center for Technologies against Cancer, Tokyo University of Science, 2641 Yamazaki, Noda 278-8510, Japan,
§Faculty of Science and Technology, Tokyo University of Science, 2641 Yamazaki, Noda 278-8510, Japan,
Rigaku Corporation X-ray Research Laboratory, 3-9-12 Matsubaracho, Akishima, Tokyo 196-8666, Japan,
and ^Faculty of Science, Shizuoka University, 836 Ohya, Suruga-ku, Shizuoka 422-8529, Japan
Received July 15, 2009
We previously reported on a 8-quinolinol-pendant cyclen (L5) as a Zn2þ fluorophore (cyclen = 1,4,7,10-tetra-
azacyclododecane) and its caged derivative, 8-(benzenesulfonyloxy)-5-(N,N-dimethylaminosulfonyl)quinolin-2-
ylmethyl-pendant cyclen (BS-caged-L5), which can be reactivated by hydrolysis of benzenesulfonyl group upon
complexation with Zn2þ at neutral pH to give a 1:1 Zn2þ-L5 complex (Zn(H-1L5)). We report herein on the synthesis
of 5,7-bis(N,N-dimethylaminosulfonyl)-8-hydroxyquinolin-2-ylmethyl-pendant cyclen (L6) and its caged derivative
(BS-caged-L6) for more sensitive and more efficient cell-membrane permeability than those of L5 and BS-caged-L5. By
potentiometric pH, 1H NMR, and UV-vis spectroscopic titrations, the deprotonation constants pKa1-pKa6 of H5L6
were determined to be <2, <2, <2, 2.5 ( 0.1 (for the 8-OH group of the quinoline moiety), 9.7 ( 0.1, and 10.8 ( 0.1 at
25 ꢀC with I = 0.1 (NaNO3). The results of 1H NMR, potentiometric pH, UV-vis, and fluorescent titrations showed that
L6 rapidly forms a 1:1 complex with Zn2þ (Zn(H-1L6)), the dissociation constant of which is 50 fM at pH 7.4. The
fluorescent emission of Zn(H-1L6) at 478 nm is 32 times as large as that of L6 (excitation at 370 nm), and the
fluorescent quantum yield of Zn(H-1L6) (ΦF = 0.41) is much greater than that of Zn(H-1L5) (ΦF = 0.044). The BS-
caged-L6 was reactivated by hydrolysis of the benzenesulfonyl moiety more rapidly (completes in 30 min at pH 7.4 at
37 ꢀC) than BS-caged-L5, presumably enabling the practical detection of Zn2þ in sample solutions and living cells. The
photochemical deprotection of BS-caged-L6 and the cell membrane permeability of L6 and BS-caged-L6 are also
described.
Introduction
extensively studied.1 In recent years, a variety of additional
functions of free Zn2þ as a neural signal transmitter2 and a
key intracellular regulator of apoptosis3 have been revealed.
Therefore, the development of sensitive Zn2þ sensors based
Zn2þ is one of the indispensable metal ions in living
systems, and its biological roles in Zn2þ enzymes, DNA
and RNA synthesis, gene expression, and so on have been
(2) (a) Cuajungco, M. P.; Lees, G. J. Neurobiol. Dis. 1997, 4, 137–169.
(b) Choi, D. W.; Koh, J. Y. Annu. Rev. Neurosci. 1998, 21, 347–375.
(c) Frederickson, C. J.; Bush, A. I. BioMetals 2001, 14, 353–366. (d) Takeda,
A. BioMetals 2001, 14, 343–351. (e) Burdette, S. C.; Lippard, S. J. Coord. Chem.
Rev. 2001, 216-217, 333–361. (f) Burdette, S. C.; Lippard, S. J. Proc. Natl.
Acad. Sci. U.S.A. 2003, 100, 3605–3610.
(3) (a) Treves, S.; Trentini, P. L.; Ascannelli, M.; Bucci, G.; Di Virgilio, F.
Exp. Cell Res. 1994, 211, 339–343. (b) Perry, D. K.; Smyth, M. J.; Stennicke, H.
R.; Salvesen, G. S.; Duriez, P.; Poirier, G. G.; Hannun, Y. A. J. Biol. Chem. 1997,
272, 18530–18533. (c) Chai, F.; Truong-Tran, A. Q.; Ho, L. H.; Zalewski, P. D.
Immuno. Cell. Biol. 1999, 77, 272–278. (d) Truong-Tran, A. Q.; Carter, J.;
Ruffin, R. E.; Zalewski, P. D. BioMetals 2001, 14, 315–330.
*To whom correspondence should be addressed. E-mail: shinaoki@
rs.noda.tus.ac.jp.
(1) (a) Auld, D. A. In Handbook of Metalloprotein; Bertini, I., Sigel, A.,
Sigel, H., Eds.; Marcel Dekker: New York, 2001; pp 881-959. (b) Auld, D. S.
BioMetals 2001, 14, 271–313. (c) Pabo, C. O.; Peisach, E.; Grant, R. A. Annu.
Rev. Biochem. 2001, 70, 313–340. (d) Aoki, S.; Kimura, E. In Comprehensive
Coordination Chemistry II; Que, L., Jr., Tolman, W. B., Eds.; Elsevier:
Amsterdam, 2004; Vol. 8, pp 601-640. (e) Maret, W. J. Trace Elem. Med.
Biol. 2005, 19, 7–12. (f) Papworth, M.; Kolasinska, P.; Minczuk, M. Gene 2005,
366, 27–38. (g) Ye, B.; Maret, W.; Vallee, B. L. Proc. Natl. Acad. Sci. U.S.A.
2005, 98, 2317–2322.
r
pubs.acs.org/IC
Published on Web 12/29/2009
2009 American Chemical Society

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 889
Scheme 1
Scheme 2
on chelation-enhanced fluorescent mechanisms is currently
of great interest.4,5
(Kd = 8 fM at pH 7.4) and that its Zn2þ complexation is
complete in the order of milliseconds. A “caged (chemically
protected)” derivative of 5, 7 (BS-caged-L5), was recently
synthesized, in attempts to improve Zn2þ/Cd2þ selectivity
and cell-membrane permeability (Scheme 2).11 It was found
that 7 forms a Zn2þ complex 8 Zn2þ-BS-caged-L5) and that
the Zn2þ-bound H2O in 8a was deprotonated to afford
Zn2þ-HO- species (8b), which promoted the hydrolysis of
the benzenesulfonyl group, resulting in the formation of 6
(Zn(H-1L5)) (Zn2þ-dependent hydrolytic uncaging). While
Zn2þ/Cd2þ selectivity was not improved, considerable en-
hancement in cellular uptake was achieved. In addition, we
happened to find that the photoirradiation of 7 promotes the
photochemical hydrolysis of its sulfonate moiety to yield 5
and benzenesulfonate (PhSO3-) in the absence and presence
of Zn2þ (Zn2þ-independent photolytic uncaging).
In this work, we wish to report on the synthesis of 5,7-
bis(N,N-dimethylaminosulfonyl)-8-hydroxyquinolin-2-yl-
methyl-pendant cyclen 10 (L6) (Scheme 3). It had previously
been reported that 5,7-bis(N,N-dimethylaminosulfonyl)-2-
methyl-8-quinolinol 9b has a much greater quantum yield
for emission (ΦF = 0.70) than that (ΦF = 0.24) of 9a.12
Therefore, it was expected that 10 (L6) would be a more
sensitive Zn2þ probe than 5 (L5). In this manuscript, we
describe the chemical and photochemical behaviors of 10
and the Zn2þ-dependent hydrolytic and Zn2þ-independent
photochemical reactions of the caged derivative 12 (BS-
caged-L6). Data on the intracellular concentrations of 10
and 12 and the staining of Zn2þ-loaded living cells with these
probes are also presented.
It has been established that 1,4,7,10-tetraazacyclodode-
cane (cyclen, L1) forms a very stable Zn2þ complex 1 (ZnL1)6
in aqueous solution at neutral pH (Scheme 1) and is a
potential platform for Zn2þ-selective fluorophores4a,c such
as 2 (L2),7 3 (L3),8 4 (L4).9
We recently reported on a Zn2þ fluorophore 5 (L5)10
bearing 5-N,N-dimethylamino-8-hydroxyquinoline and cy-
clen moieties (Scheme 2). The fluorescent emission of 5
increased by a factor of 22 upon complexation with Zn2þ
(quantum yields (ΦF) for the emission of 5 and 6 are 2.0 ꢀ
10-3 and 4.4 ꢀ 10-2). In addition, it was found that 5 forms a
very stable Zn2þ complex 6 (Zn(H-1L5)) in aqueous solution
(4) Reviews: (a) Kimura, E.; Koike, T. Chem. Soc. Rev. 1998, 27, 179–184.
(b) Prodi, L.; Bolletta, F.; Montalti, M.; Zaccheroni, N. Coord. Chem. Rev. 2000,
205, 59–83. (c) Kimura, E.; Aoki, S. BioMetals. 2001, 14, 191–204. (d) Kikuchi,
K.; Komatsu, K.; Nagano, T. Curr. Opin. Chem. Biol. 2004, 8, 182–191.
(e) Jiang, P.; Guo, Z. Coord. Chem. Rev. 2004, 248, 205–229. (f) Dai, Z.;
Canary, J. W. New J. Chem. 2007, 31, 1708–1718. (g) Nolan, E. M.; Lippard, S. J.
Acc. Chem. Res. 2009, 42, 193–203.
(5) For recent examples, see: (a) Burdette, S. C.; Lippard, S. J. Proc. Natl.
Acad. Sci. U.S.A. 2003, 100, 3605–3610. (b) Woodroofe, C. C.; Masalha, R.;
Barnes, K. R.; Frederickson, C. J.; Lippard, S. J. Chem. Biol. 2004, 11, 1659–
1666. (c) Kiyose, K.; Kojima, H.; Urano, Y.; Nagano, T. J. Am. Chem. Soc. 2006,
128, 6548–6549. (d) Harriman, A.; Mallon, L. J.; Stewart, B.; Ulrich, G.; Ziessel,
R. Eur. J. Org. Chem. 2007, 3191–3198. (e) Parkesh, R.; Lee, T. C.;
Gunnlaugsson, T. Org. Biomol. Chem. 2007, 5, 310–317. (f) Atilgan, S.;
Ozdemir, T.; Akkaya, E. U. Org. Lett. 2008, 10, 4065–4067. (g) Zhang, Y.;
ꢀ
Guo, X.; Si, W.; Jia, L.; Qian, X. Org. Lett. 2008, 10, 473–476. (h) Alarcon, J.;
Albelda, M. T.; Belda, R.; Clares, M. P.; Delgado-Pinar, E.; Frías, J. C.; García-
~
ꢀ
Espana, E.; Gonzalez, J.; Soriano, C. Dalton Trans. 2008, 6530–6538. (i) Qian,
F.; Zhang, C.; Zhang, Y.; He, W.; Gao, X.; Hu, P.; Guo, Z. J. Am. Chem. Soc.
2009, 131, 1460–1468. (j) Tamanini, E.; Katewa, A.; Sedger, L. M.; Todd, M. H.;
Watkinson, M. Inorg. Chem. 2009, 48, 319–324.
Experimental Section
(6) (a) Kimura, E.; Shiota, T.; Koike, T.; Shiro, M.; Kodama, M. J. Am.
Chem. Soc. 1990, 112, 5805–5811. (b) Kimura, E.; Shionoya, M.; Hoshino, A.;
Ikeda, T.; Yamada, Y. J. Am. Chem. Soc. 1992, 114, 10134–10137.
(7) (a) Koike, T.; Watanabe, T.; Aoki, S.; Kimura, E.; Shiro, M. J. Am.
Chem. Soc. 1996, 118, 12696–12703. (b) Kimura, E.; Koike, T. Chem. Soc. Rev.
1998, 27, 179–184. (c) Kimura, E.; Aoki, S.; Kikuta, E.; Koike, T. Proc. Natl.
Acad. Sci. U.S.A. 2003, 100, 3731–3736. (d) Kimura, E.; Takasawa, R.; Tanuma,
S.; Aoki, S. Science STKE 2004, 223, 7.
General Information. Reagents and solvents were purchased
at the highest commercial quality and were used without further
purification. N-bromosuccinimide (NBS) was recrystallized
from water before use. ZnSO4 7H2O, Zn(NO3)2 6H2O,
3
3
3CdSO4 8H2O, CuSO4 5H2O, CoSO4 7H2O, FeSO4 7H2O,
3
3
3
3
AgNO3, FeCl3 6H2O, KCl, MgSO4, and CaSO4 2H2O were
3
3
(8) Aoki, S.; Kaido, S.; Fujioka, H.; Kimura, E. Inorg. Chem. 2003, 42,
1023–1030.
(9) Aoki, S.; Kagata, D.; Shiro, M.; Takeda, K.; Kimura, E. J. Am. Chem.
Soc. 2004, 126, 13377–13390.
(10) Aoki, S.; Sakurama, K.; Matsuo, N.; Yamada, Y.; Takasawa, R.;
Tanuma, S.; Shiro, M.; Takeda, K.; Kimura, E. Chem.;Eur. J. 2006, 12,
9066–9080.
(11) Aoki, S.; Sakurama, K.; Ohshima, R.; Matsuo, N.; Yamada, Y.;
Takasawa, R.; Tanuma, S.; Takeda, K.; Kimura, E. Inorg. Chem. 2008, 47,
2747–2754.
(12) Pearce, D. A.; Jotterand, N.; Carrico, I. S.; Imperiali, B. J. Am.
Chem. Soc. 2001, 123, 5160–5161.

890 Inorganic Chemistry, Vol. 49, No. 3, 2010
Ohshima et al.
Scheme 3
MS measurements were performed on a JEOL JMS-SX-102A.
Elemental analyses were performed on a Perkin-Elmer CHN
2400 analyzer. Thin-layer (TLC) and silica gel column chroma-
tographies were performed using a Merck Silica gel 60 F254 TLC
plate and Fuji Silysia Chemical FL-100D, respectively.
8-(Benzenesulfonyloxy)-5,7-bis(N,N-dimethylaminosulfonyl)-
quinaldine (13). Benzenesulfonyl chloride (1.2 g, 6.9 mmol) and
4-dimethylaminopyridine (126 mg, 1.0 mmol) were added to a
solution of 5,7-bis(N,N-dimethylaminosulfonyl)-8-hydroxyqui-
naldine 9b12 (1.3 g, 3.4 mmol) and Et3N (630 μL, 4.5 mmol) in
CH2Cl2 (30 mL) at 0 ꢀC, and the resulting solution was stirred at
reflux temperature for 1 day. The reaction mixture was concen-
tratedunderreducedpressure, andtheresulting residuewas purified
by silica gel column chromatography (hexane/AcOEt = 3:1);
recrystallization from AcOEt afforded 13 as colorless prisms
(1.3 g, 74% yield). The starting material 9b was recovered
1
(199 mg, 16% yield). mp. 176-178 ꢀC. H NMR (300 MHz,
CDCl3/TMS): δ 2.40 (s, 3H), 2.82 (s, 6H), 2.91 (s, 6H), 7.49
(d, J = 9.0 Hz, 1H), 7.62 (t, J = 8.4 Hz, 2H), 7.74 (t, J = 8.4 Hz,
1H), 8.06 (d, J = 8.4 Hz, 2H), 8.53 (s, 1H), 9.00 ppm (d, J = 9.0
Hz, 1H). 13C NMR (75 MHz, CDCl3/TMS): δ 24.8, 37.6, 125.9,
126.6, 128.1, 128.4, 128.8, 130.1, 132.2, 133.6, 133.8, 139.0,
142.2, 148.2, 161.5 ppm. IR (KBr pellet): 3098, 3065, 2975,
2923, 1609, 1586, 1548, 1495, 1450, 1372, 1345, 1284, 1245, 1181,
1165, 1144, 1075, 1033, 968, 835, 797, 764, 718, 689, 654, 625,
607, 587, 557, 509, 490, 438 cm-1. Elemental analysis: calcd for
C20H23N3O7S3 (513.61): C, 46.77; H, 4.51; N, 8.18; found: C,
46.77; H, 4.18; N, 8.16.
8-(Benzenesulfonyloxy)-5,7-bis(N,N-dimethylaminosulfonyl)-
2-(bromomethyl)quinoline (14). A mixture of 13 (300 mg, 0.6
mmol), N-bromosuccinimide (84 mg, 0.5 mmol) and AIBN
(20 mg, 0.1 mmol) in distilled CCl4 (20 mL) was stirred at reflux
temperature for 3 h (the N-bromosuccinimide was added
portionwise), and the reaction mixture was concentrated under
reduced pressure. The residue was dissolved in CH2Cl2 (5 mL)
and washed with a saturated solution of Na2CO3 and Na2S2O4.
The organic layer was dried over anhydrous Na2SO4, filtered,
and concentrated under reduced pressure. The resulting residue
was purified by silica gel column chromatography (hexane/
AcOEt = 2:1) to afford 14 as a colorless amorphous solid (52
mg, 15% yield). The starting material 13 was recovered in 58%
purchased from Kanto Chemical Co. Ltd.; HgCl was purchased
from Yoneyama Kogyo Co.; Al(NO3)3 9H2O and MnSO4
3
3
H2O were purchased from Sigma-Aldrich Chemical Co.; Eu-
(NO3)3 6H2O and Y(NO3)3 6H2O were purchased from
3
3
Soekawa Co. All aqueous solutions were prepared using deio-
nized and distilled water. Buffer solutions (CAPS, pH 13.0, 12.0,
11.0 and 10.0; CHES, pH 9.0; TAPS, pH 8.0; HEPES, pH 7.4,
and 7.0; MES, pH 6.0; acetic acid/sodium acetate, pH 5.0 and
4.0; Gly/HCl, pH 3.0; KCl/HCl, pH 2.0 and 1.0) were used, and
the ionic strengths were adjusted with NaNO3. The Good’s buf-
fer reagents (Dojindo) were obtained from commercial sources:
MES (2-morpholinoethanesulfonic acid, pKa = 4.8), HEPES
1
yield (175 mg). H NMR (300 MHz, CDCl3/TMS): δ 2.87 (s,
(N-(2-hydroxyethyl)piperazine-N0-2-ethanesulfonic
acid,
6H), 2.92 (s, 6H), 4.08 (s, 2H), 7.66 (t, J = 7.8 Hz, 2H), 7.78 (t,
J = 7.8 Hz, 1H), 7.80 (d, J = 9.0 Hz, 1H), 8.06 (d, J = 7.5 Hz,
2H), 8.61 (s, 1H), 9.13 ppm (d, J = 9.0 Hz, 1H). 13C NMR (75
MHz, CDCl3/TMS): δ 32.3, 37.6, 125.2, 127.2, 128.3, 129.0,
129.4, 131.1, 132.5, 133.9, 135.1, 139.0, 141.7, 148.4, 159.0 ppm.
IR (KBr pellet): 3098, 2967, 2921, 1542, 1495, 1450, 1384, 1495,
1349, 1290, 1196, 1177, 1146, 1069, 961, 798, 769, 725, 688, 669,
607, 583, 551 cm-1. HRMS (FABþ): calcd for [MþH]þ,
591.9881; found, 591.9879.
pKa = 7.5), TAPS (N-[tris(hydroxymethyl)methylamino]-3-
propanesulfonic acid, pKa = 8.4), CHES (2-(cyclohexylamino)-
ethanesulfonic acid, pKa = 9.5), CAPS (3-(cyclohexylamino)-
propanesulfonic acid, pKa = 10.4). Melting points were
€
measured on a Buchi 510 Melting Point Apparatus and are
uncorrected. UV-vis spectra were recorded on a JASCO UV/
vis spectrophotometer V-550, and fluorescence (excitation and
emission) spectra were recorded on JASCO FP-6200 and FP-
6500 spectrofluorometers at 25.0 ( 0.1 ꢀC. The quantum yields
(ΦF) for fluorescence emission were determined by comparison
with the integrated corrected emission spectrum of quinine
sulfate (a standard), whose quantum yield in 0.1 M H2SO4
was assumed to be 0.55 (excitation at 366 nm).13 IR spectra were
recorded on a JASCO FTIR-410 spectrophotometer at room
1-[8-(Benzenesulfonyloxy)-5,7-bis(N,N-dimethylaminosulfonyl)-
quinolin-2-ylmethyl]-4,7,10-tris(tert-butyloxycarbonyl)-1,4,7,10-
tetraazacyclododecane (16). A mixture of 3Boc-cyclen 1514 (131
mg, 0.28 mmol), 14 (180 mg, 0.30 mmol) and Na2CO3 (50 mg,
0.47 mmol) in CH3CN (20 mL) was stirred at 60 ꢀC overnight
under an argon atmosphere. After removing the insoluble
inorganic salts by filtration, the filtrate was concentrated under
reduced pressure. The resulting residue was purified by silica gel
column chromatography (CH2Cl2/MeOH = 100:1) to afford 16
as a pale yellow amorphous solid (246 mg, 90% yield). 1H NMR
(300 MHz, CDCl3/TMS): δ 1.43-1.48 (m, 27H), 2.73 (s, 6H),
2.90 (s, 6H), 2.73-3.57 (br, 16H), 3.86 (s, 2H), 7.61 (t, J = 7.5
Hz, 2H), 7.74 (t, J = 7.5 Hz, 1H), 7.81 (d, J = 9.0 Hz, 1H), 8.04
(d, J = 7.5 Hz, 2H), 8.55 (s, 1H), 9.02 ppm (d, J = 9.0 Hz, 1H).
13C NMR (75 MHz, CDCl3/TMS): δ 19.2, 27.1, 28.3, 28.5, 35.1,
1
temperature. H (300 MHz) and 13C (75 MHz) NMR spectra
were recorded on a JEOL Always 300 spectrometer. Tetra-
methylsilane was used as an internal reference for 1H and CDCl3
for 13C NMR measurements in CDCl3. 3-(Trimethylsilyl)-
propionic-2,2,3,3-d4 acid (TSP) sodium salt was used as an
1
external reference for H and 1,4-dioxane for 13C NMR mea-
surements in D2O. The pD values in D2O were corrected for a
deuterium isotope effect using pD = (pH-meter reading) þ 0.40.
(13) (a) Turro, N. J. Modern Molecular Photochemistry; University Science
Books: Sausalito, CA, 1991. (b) Murov, S. L.; Carmichael, I.; Hug, G. L.
Handbook of Photochemistry, 2nd ed.; Marcel Dekker: New York, 1993.
(14) Kimura, E.; Aoki, S.; Koike, T.; Shiro, M. J. Am. Chem. Soc. 1997,
119, 3068–3076.

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 891
37.4, 46.1, 47.7, 49.9, 50.6, 54.3, 55.2, 58.1, 79.5, 125.9, 127.2,
128.5, 128.6, 128.8, 129.2, 132.0, 133.3, 134.0, 138.3, 142.0,
148.5, 155.7, 176.4 ppm. IR (KBr pellet): 2974, 2931, 2815,
1689, 1597, 1560, 1458, 1415, 1365, 1342, 1250, 1152, 1049, 962,
859, 774, 725, 622, 556 cm-1. HRMS (FABþ): calcd for
[MþH]þ, 984.3881; found, 984.3881.
36.8, 41.2, 41.6, 44.2, 47.7, 57.8, 126.8, 127.3, 127.7, 128.3, 129.1,
131.0, 134.5, 135.2, 135.4, 142.2, 147.6, 149.9, 159.8, 162.2, 162.6
ppm. IR (KBr pellet): 3411, 2923, 2851, 1682, 1550, 1454, 1347,
1202, 1142, 1071, 962, 835, 799, 722, 688, 607, 585, 556 cm-1
.
Elemental analysis: calcd for C34H45F9N7O13.5S3 (1034.94): C,
39.46; H, 4.38; N, 9.47; found: C, 39.67; H, 4.63; N, 9.67.
1-[5,7-Bis(N,N-dimethylaminosulfonyl)-8-hydroxyquinolin-2-
ylmethyl]-1,4,7,10-tetraazacyclododecane Zn(NO3) Complex
(11 NO3 1.3H2O). A solution of Zn(NO3)2 6H2O (15 mg,
1-[5,7-Bis(N,N-dimethylaminosulfonyl)-8-hydroxyquinolin-2-
ylmethyl]-4,7,10-tris(tert-butyloxycarbonyl)-1,4,7,10-tetraaza-
cyclododecane (17). One N aq. NaOH (1 mL) was added to a
solution of 16 (80 mg, 0.081 mmol) in MeOH (5 mL) at room
temperature, and the mixture was stirred at reflux temperature
for 3 h. After cooling, the reaction mixture was concentrated
under reduced pressure and purified by silica gel column chro-
matography (CH2Cl2/MeOH = 100:1) to afford 17 as a color-
less powder (64 mg, 93% yield). mp. 95-98 ꢀC. 1H NMR (300
MHz, CDCl3/TMS): δ 1.42-1.49 (m, 27H), 2.81 (s, 6H), 2.95 (s,
6H), 2.81-3.61 (m, 16H), 4.14 (s, 2H), 7.83 (d, J = 9.0 Hz, 1H),
8.50 (s, 1H), 9.02 ppm (d, J = 9.0 Hz, 1H). 13C NMR (75 MHz,
CDCl3/TMS): δ 20.6, 22.6, 25.2, 28.4, 28.6, 31.5, 34.6, 37.4, 37.7,
47.4, 49.9, 54.5, 57.6, 79.7, 117.7, 122.9, 125.9, 126.3, 130.9,
134.8, 137.7, 154.6 ppm. IR (KBr pellet): 3447, 2974, 2931, 2815,
1689, 1597, 1560, 1458, 1415, 1365, 1250, 1152, 962, 859, 774,
725, 622, 556 cm-1. HRMS (FAB-): calcd for [M-H]-,
842.3792; found, 842.3793.
3
3
3
0.05 mmol) in water (1 mL) was added into a solution of
10 3HCl H2O (34 mg, 0.05 mmol) in H2O (3 mL), and the pH
3
3
of a reaction mixture adjusted to 10.0 by adding aq. NaOH.
After the insoluble compounds were removed by filtration, the
filtrate was slowly concentrated at atmospheric pressure to
obtain 11 NO3 1.3H2O as colorless prisms (8 mg, 23% yield).
3
3
mp. > 250 ꢀC. 1H NMR (300 MHz, D2O/external TSP): δ 2.74
(s, 6H), 2.86 (s, 6H), 2.91-3.20 (m, 16H), 4.34 (s, 2H), 7.74 (d,
J = 8.7 Hz, 1H), 8.27 (s, 1H), 8.94 ppm (d, J = 8.7 Hz, 1H). 13
C
NMR (75 MHz, DMSO/TMS): δ 37.0, 38.0, 43.6, 44.2, 45.0,
52.5, 57.5, 108.6, 116.0, 123.9, 126.7, 134.9, 136.8, 140.3, 155.4,
166.1 ppm. IR (KBr pellet): 3570, 3448, 3332, 3273, 2928, 2879,
1649, 1548, 1508, 1488, 1385, 1316, 1283, 1259, 1198, 1167, 1140,
1119, 1092, 1042, 993, 950, 930, 853, 796, 750, 720, 688, 596, 560
cm-1
. Elemental analysis: calcd for C22H36.6N8O9.3S2Zn
(691.49): C, 38.21; H, 5.33; N, 16.20; found: C, 37.83; H, 5.21;
N, 16.00.
1-[5,7-Bis(N,N-dimethylaminosulfonyl)-8-hydroxyquinolin-2-
ylmethyl]-1,4,7,10-tetraazacyclododecane Trihydrochloride Salt
(10 3HCl H2O). Conc. HCl (1 mL) was added dropwise to a
solution of 17 (290 mg, 0.34 mmol) in MeOH (5 mL), and the
resulting solution was stirred at 70 ꢀC for 10 min. The reaction
mixture was concentrated under reduced pressure, azeotroped
with CHCl3, and the resulting solid was recrystallized from
Potentiometric pH Titrations. The preparation of the test
solutions and the calibration method of the electrode system
(Potentiometric Automatic Titrator AT-400 and Auto Piston
Buret APB-410, Kyoto Electronics Manufacturing, Co. Ltd.
with a Kyoto Electronics Manufacturing Co. Combination
pH Electrode 98100C171) have been described in earlier re-
ports.10,11 All test solutions (50 mL) were stored under an argon
atmosphere. Potentiometric pH titrations were performed with
3
3
H2O/EtOH to afford 10 3HCl salt (characterized by elemental
3
analysis and potentiometric pH titration) as a colorless powder
(170 mg, 74% yield). After concentrating the filtrate under
reduced pressure, a green powder was obtained, which was
I = 0.1 (NaNO3) at 25 ( 0.1 ꢀC. The deprotonation constants
0
characterized as the 10 4HCl salt (26 mg, 11% yield) by
elemental analysis.15
of Zn2þ-bound water K2 (= [HO--bound species][Hþ]/[H2O-
3
bound species]) were determined by means of the “BEST”
0
1
For the 10 3HCl salt: mp. > 250 ꢀC. H NMR (300 MHz,
software program.16 The Kw (equivalent to aHþaOH-), Kw
(equivalent to [Hþ][HO-]), fHþ values used at 25 ꢀC were
10-14.00, 10-13.79, and 0.825, respectively. The corresponding
mixed constants K2 (= [HO--bound species]aHþ/[H2O-bound
species]), were derived by using [Hþ] = aHþ/fHþ. The percentage
species distribution values against pH (= -log [Hþ] þ 0.084)
were obtained using the SPE software program.16
3
D2O/external TSP): δ 2.81 (s, 6H), 2.89 (s, 6H), 3.05-3.25 (m,
16H), 4.22 (s, 2H), 7.75 (d, J = 9.0 Hz, 1H), 8.34 (s, 1H), 8.94
ppm (d, J = 9.0 Hz, 1H). 13C NMR (75 MHz, D2O/external
1,4-dioxane): δ 36.8, 37.2, 41.7, 42.5, 42.9, 48.6, 57.0, 115.4,
119.9, 126.5, 130.6, 135.4, 138.3, 156.5, 156.7 ppm. IR (KBr
pellet): 3423, 2961, 1595, 1560, 1499, 1456, 1336, 1145, 1099,
1072, 955, 848, 787, 727, 622, 557 cm-1. Elemental analysis:
calcd for C22H42Cl3N7O6S2 (671.10): C, 39.37; H, 6.31; N, 14.61;
found: C, 39.43; H, 6.39; N, 14.42.
Crystallographic Study of 10 (H2(H-1L6) (= HL6)). 10
(H2(H-1L6) (= HL6)) was recrystallized from an aqueous
solution of 10 at pH 5 and 4 ꢀC. These crystals, which were
filtered and dried, were determined to be [HL6]þ Cl- 4.2H2O
For the 10 4HCl salt: mp. 248-251 ꢀC (decomp.). Elemental
3
3
3
analysis: calcd for C22H41Cl4N7O5S2 (689.55): C, 38.32; H, 5.99;
N, 14.22; found: C, 38.47; H, 6.28; N, 13.99.
by elemental analysis (Anal. Calcd for C22H38Cl1N7O5S2
3
4.2H2O: C 40.29, H 7.13, N 14.95. Found: C 39.85, H 6.68, N
14.61). All measurements were made on a Rigaku RAXIS-
RAPID instrument with graphite monochromated Cu KR
radiation at 93 K. The structure was solved by direct methods17
and refined by full-matrix least-squares techniques. All calcula-
tions were performed using the CrystalStructure crystallo-
graphic software package except for refinements, which were
performed with SHELXL-97.18 C22H50.4Cl1N7O11.2S2, Mr =
691.85, a colorless block crystal, crystal size 0.15 ꢀ 0.14 ꢀ 0.14
mm, monoclinic, space group C2/c (#15), a = 35.8779(7), b =
1-[8-(Benzenesulfonyloxy)-5,7-bis(N,N-dimethylaminosulfo-
nyl)quinolin-2-ylmethyl]-1,4,7,10-tetraazacyclododecane 3TFA
Salt (12 3TFA 0.5H2O). A solution of trifluoroacetic acid
3
3
(1.5 mL, 5.2 mmol) in CH2Cl2 (2.5 mL) was added to a solution
of 16 (200 mg, 0.20 mmol) in CH2Cl2 (2.5 mL), and the resulting
solution was stirred at 60 ꢀC for 0.5 h. The reaction mixture was
concentrated under reduced pressure, azeotroped with CHCl3,
and the resulting residue was recrystallized from Et2O/EtOH to
afford a 12 3TFA salt (12 3TFA 0.5H2O) as a colorless pow-
3
3
3
1
˚
der (177 mg, 84% yield). mp. > 250 ꢀC. H NMR (300 MHz,
D2O/external TSP): δ 2.73 (s, 6H), 2.88 (s, 6H), 2.98-3.30 (m,
16H), 4.13 (s, 2H), 7.70 (t, J = 7.8 Hz, 2H), 7.89 (t, J = 7.8 Hz,
1H), 7.94-8.01 (m, 3H), 8.42 (s, 1H), 9.21 ppm (d, J = 9.0 Hz,
1H). 13C NMR (75 MHz, D2O/external 1,4-dioxane): δ 36.6,
19.7401(4), c = 21.4880(7) A, β = 118.8100(7)ꢀ, V = 13334.9(6)
A , Z = 16, Dcalc = 1.378 g cm-3, 91503 measured reflections,
3
˚
3
11999 unique reflections, 2θmax = 68.3ꢀ, R1 (wR2) = 0.0685
(16) Martell, A. E.; Motekaitis, R. J. Determination and Use of Stability
Constants, 2nd ed.; VCH: New York, 1992.
(17) Burla, M. C.; Caliandro, R.; Camalli, M.; Carrozzini, B.; Cascarano,
G. L.; De Caro, L.; Giacovazzo, C.; Polidori, G.; Spagna, R. J. Appl.
Crystallogr. 2005, 38, 381–388.
(15) As described in the Experimental Section, 3HCl and 4HCl salt of 10
were obtained as colorless and green powders, respectively. As pointed out
by the reviewer, this may be related to the protonation status of both salts in
the solid state, although we do not have experimental evidence.
(18) Sheldrick, G. M. SHELX-97, Program for the Refinement of Crystal
€
€
Structures; University of Gottingen: Gottingen, Germany, 1997.

892 Inorganic Chemistry, Vol. 49, No. 3, 2010
Ohshima et al.
Scheme 4
(0.1966), GOF = 1.033. Full details of the crystallographic
analysis of 10 are given in the Supporting Information.
Crystallographic Study of 11 NO3 (Zn(H-1L6) NO3). 11
Scheme 4. The reaction of 9b12 with benzenesulfonyl
chloride in the presence of 4-dimethylaminopyridine
(DMAP) and Et3N in CH2Cl2 gave 13, the 2-methyl
group of which was brominated with NBS and AIBN in
CCl4 to give 14. The reaction of 14 with 3Boc-cyclen 1514
gave 16, the three Boc groups of which were deprotected
with trifluoroacetic acid (TFA) in CH2Cl2 afforded 12
(BS-caged-L6) as the 3TFA salt. The PhSO2 group of 16
was removed by treatment with aqueous NaOH to give
17, the Boc groups of which were deprotected by treat-
ment with HCl in MeOH afforded 10 (L6). It should be
noted that 10 was obtained as a colorless powder,
and concentration of its filtrate gave a green powder.
These two powders were determined to be 3HCl salt
(10 3HCl H2O) and 4HCl salt (10 4HCl), respectively,
3
3
3
NO3 (Zn(H-1L6) NO3) was recrystallized by the slow evapora-
3
tion of an aqueous solution. All measurements were made on a
Rigaku Saturn CCD area detector with graphite monochro-
mated Mo KR radiation at 133 K. C22H36N8O9S2Zn, Mr =
686.07, a colorless block crystal, crystal size 0.25 ꢀ 0.20 ꢀ 0.15
mm, monoclinic, space group P21/n (#14), a = 11.602(5), b =
3
˚
˚
13.792(6), c = 17.941(8) A, β = 93.923(3)ꢀ, V = 2864(2) A ,
Z = 4, Dcalc = 1.591 g cm-3, 22048 measured reflections, 7339
unique reflections, 2θmax = 57.4ꢀ, R1 (wR2) = 0.0646 (0.1901),
GOF = 0.999. Full details of crystallographic analysis of
11 NO3 are given in the Supporting Information.
3
Photoreaction. Sample solutions were prepared in quartz
cuvettes (GL Science Inc. Japan, 10 mm light path) and irra-
diated at a wavelength 328 or 330 nm on JASCO FP-6200 or
those of FP-6500 spectrofluorometer. The photoreactions were
followed by UV-vis (50 μM) spectra or 1H NMR spectra. The
averaged light intensities of the JASCO FP-6200 spectrofluo-
3
3
3
by elemental analysis and potentiometric pH titrations.15
Deprotonation Constants for 10 (L6) Determined by
Potentiometric pH Titration, 1H NMR, and UV-vis
Spectra. A typical potentiometric pH titration curve for
1 mM 10 3HCl H2O (H3L6) þ 2 mM HNO3 against
rometer at 330 nm were determined to be 1.4 ꢀ 10-5 mol sec-1
3
(slit width for excitation = 20 nm), and those of the JASCO FP-
6500 spectrofluorometer at 328 nm were determined to be 5.3 ꢀ
3
3
aqueous 0.1 M NaOH with I = 0.1 (NaNO3) at 25 ꢀC
(dashed curve (a) in Figure 1) was analyzed for acid-base
equilibrium (1). The deprotonation constants pKai (i =
1-6) of H5L6 were determined to be <2 (pKa1∼pKa3), 2.5
( 0.1 (pKa4), 9.7 ( 0.1 (pKa5), and 10.8 ( 0.1 (pKa6)
(Table 1) by using the “BEST” software program.16 The
pKa1, pKa2, pKa5, and pKa6 values were assigned to the
pKa values for the four amines in the cyclen ring by
comparison with the pKa values for cyclen (L1) and 5
(L5),10 as summarized in Table 1.
10-6 mol sec-1 (slit width for excitation = 20 nm) relative to the
3
potassium ferioxalate actinometer.13 The reactions were re-
peated two or three times, and the averaged values were
calculated. Experimental fluctuations were ( 5%.
Treatment of HeLa Cells with Zn2þ Probes and Fluorescence
Microscopy. HeLa cells were seeded onto 35 mm glass-bottom
dishes. The cells were incubated with Zn2þ fluorophores 10 and
12 (50 μM) in culture medium for 0.5 h in a humid atmosphere of
5% CO2 at 37 ꢀC, then washed twice with PBS, and the culture
medium (2 mL) was replaced. The cells were treated with 25 μM
ZnSO4 7H2O and 20 μM pyrithione for 10 min, washed with
3
L Hþ a L þ Hþ
Ka ¼ ½LꢁaH þ =½L Hþꢁ
ð1Þ
PBS and observed by phase contrast and fluorescence micro-
scopy (KEYENCE fluorescent microscope, BZ-9000; excitation
at 360/40 nm, emission 460/50 nm).
3
3
The deprotonation behavior of 10 is shown in Scheme
5. Changes in the UV-vis absorption of 10 at 265 nm in
the pH range 1-4 (Figure S1a in the Supporting In-
formation), gave a sigmoidal curve (a) with closed squares
in Figure 2a, which gave a pKa value of 2.7 ( 0.2,
suggesting that the pKa2 value of 2.5 determined by
Results and Discussion
Synthesis of Zn2þ Fluorophores 10 and 12 (L6 and BS-
caged-L6). A ligand 10 (L6) and its benzenesulfonyl-caged
derivative 12 (BS-caged-L6) were synthesized as shown in

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 893
also Figure S1b in the Supporting Information). This pKa
value obtained from emission spectra is much greater
than the pKa value of 2.5-2.7 in the ground state deter-
mined by potentiometric pH and UV-vis titrations. For
comparison, the emission intensity of 9b is very small in
the pH range of 2-13, as indicated by the open circles
(c) in Figure 2. Therefore, this unique emission behavior
of 10 can be attributed to the effect of its cyclen ring.
Moreover, UV-vis absorption spectra of 10 at pH 10-12
showed negligible change, as displayed in Figure 2 and
Figure S1 in the Supporting Information, supporting that
the pKa value of around 10-11 obtained from pH-emis-
sion profile (Figure 2) corresponds to the pKa value of
quinolinol group in 10 influenced by the cyclen ring in the
excited state. We previously observed a similar phenom-
enon for 5 (L5) and concluded that fluorescent quenching
of quinolinols, which resulted in the higher pKa value
obtained from emission spectra than that in the ground
state, was due to the excited-state proton transfer (ESPT)
from the cyclen ring.20 Analogously, in the case of the
anionic quinolinolate of the H2(H-1L6) (= HL6) form or
the H(H-1L6) (= L6) form, we assume that protons
would be extracted from the protonated cyclen ring (its
higher two pKa values, pKa5 and pKa6, are 9.7 and 10.8,
respectively) in the exited states (Scheme 5), resulting in
the quenching of its fluorescent emission.21
Figure 1. Typical potentiometric pH titration curves for (a) 1 mM H3L6
þ 2 mM HNO3 and (b) 1 mM H3L6 þ 2 mM HNO3 þ 1 mM Zn2þ with
I = 0.1 (NaNO3) at 25 ꢀC. Eq (HO-) is the number of equivalents of base
(NaOH) added.
Table 1. Deprotonation Constants (pKai) and Complexation Constants for
Cyclen (L1), 5 (L5), and 10 (L6) with Zn2þ in Aqueous Solution with I = 0.1
(NaNO3) at 25 ꢀC
cyclen (L1)
5 (L5)a
10 (L6)
pKa1
pKa2
pKa3
pKa4
<2
<2
9.9
<2
<2
<2
7.2b
<2
<2
<2
11.0
2.51 ( 0.02b
2.7 ( 0.2b,c
9.69 ( 0.02
10.84 ( 0.02
19.5
UV-vis and Emission Titrations of 10 (L6) with Zn2þ
.
pKa5
pKa6
10.1
11.5
22.4
14.1
10.8
Figure 4a shows the results of the UV-vis absorption titra-
tion of 50 μM 10 (L6) with Zn2þ at pH 7.4 (10 mM HEPES
with I = 0.1 (NaNO3)) and 25 ꢀC. The absorption max-
imum shifted from 265 to 269 nm upon addition of the Zn2þ
(0-1.0 equiv) with isosbestic points at 284 and 370 nm.
As shown in Figure 4b, the emission of 10 (5 μM) was
very weak at pH 7.4 and quantitatively increased with
increasing Zn2þ concentrations (excitation at 370 nm,
which is an isosbestic point found in Figure 4a), reaching
a plateau at 1.0 equiv with 32-fold increase in emission
intensity at 478 nm (the inset of Figure 2b), which strongly
indicated a 1:1 complexation of 10 with Zn2þ, considering
together with the results of UV-vis titration.
Figure 5 shows a comparison of the pH-dependent
changes in emission intensity of the two Zn2þ complexes,
6 and 11 (5 μM). It is noteworthy that the fluorescent
emission of 11 (Zn(H-1L6), ΦF = 0.41 at pH 7.4) is much
greater than that of 6 (Zn(H-1L5), ΦF = 0.044 at pH 7.4),
indicating that 10 is a more sensitive Zn2þ probe than L5.
The facts that emission of 11 is almost constant at pH
4-11 and that the emission of 10 is very weak below pH 9
log Ks(ZnL)
log Kapp(ZnL) at pH 7.4
log Kapp(ZnL) at pH 5.0
16.2
10.6
5.5
13.3
9.0
a From ref 10. b The pKa values of 8-OH groups of quinolinol
moieties. c The pKa value determined by pH-UV absorption profile.
potentiometric pH titration corresponds to the pKa value
of the 8-OH group of 10 (H2L6 a H2(H-1L6) in
Scheme 5). Interestingly, this value is much smaller than
those for 5 (L5) (7.0) and 9b (6.1) (See also Scheme 3).
Figure 3a shows an ORTEP drawing for the single-
crystal X-ray structure of H2(H-1L6) (= HL6), in which
the 8-O- group of the quinolinolate moiety is hydrogen
bonded to the ammonium cation of the cyclen ring.
Representative crystallographic data for 10 (HL6) and
its Zn2þ complex 11 (Zn(H-1L6) displayed below are
listed in Table 2. A space-filling model of the same crystal
(Figure 3b) suggests that the 8-O- group of quinolinolate
(O1 in Figure 3b) is surrounded by the quinoline ring, the
Me2NSO2 group at the 7-position of 8-quinolinolate, and
the cyclen ring. We, therefore, consider that the ion pair of
Ar-O- (quinolinolate) and the ammonium group (cyclen
ring) is stabilized in this hydrophobic space, resulting in
the extremely lower pKa value of 2.5 in the ground state,
compared to that for 9b (6.1).19
(20) (a) Bardez, E.; Chatelain, A.; Larrey, B.; Valeur, B. J. Phys. Chem.
1994, 98, 2357–2366. (b) Bardez, E.; Devol, I.; Larrey, B.; Valeur, B. J. Phys.
Chem. B 1997, 101, 7786–7793. (c) Cheatum, C. M.; Heckscher, M. M.; Crim, F.
F. Chem. Phys. Lett. 2001, 349, 37–42. (d) Li, Q.-S.; Fang, W.-H. Chem. Phys.
Lett. 2003, 367, 637–644.
(21) Prodi, Izatt, and Savage et al. reported that the fluorescent emission
intensity of 7-methyl-8-quinolinol was much greater than that of 2-methyl-8-
quinolinol, suggesting that a functional group at the 7-position affects the
photochemical properties of 8-quinolinols. Bronson, R. T.; Montalti, M.;
Prodi, L.; Zaccheroni, N.; Lamb, R. D.; Dalley, N. K.; Izatt, R. M.;
Bradshaw, J. S.; Savage, P. B. Tetrahedron 2004, 60, 11139–11144.
pH-Dependent Fluorescent Behaviors of 10 (L6). The
pH-emission profile of 10 (excitation at 370 nm and
emission at 478 nm) in Figure 2 (closed circles) shows a
sigmoidal curve giving a pKa value of around 10-11 (See
(19) As shown in Figure S2 in the Supporting Information, H2(H-1L6)
was isolated as a dimeric structure. We cannot exclude the possibility that
π-π interactions between two quinolinol moieties from two separate
molecules of L6 could facilitate the deprotonation of the 8-OH groups.

894 Inorganic Chemistry, Vol. 49, No. 3, 2010
Ohshima et al.
(Figures 2 and 4b) would permit a quantitative measure-
Complexation Behavior of 10 (L6) with Zn2þ As Studied
by Potentiometric pH Titrations. Analysis of a potentio-
metric pH titration curve for a mixture of 1 mM H3L6,
2 mM HNO3, and 1 mM ZnSO4 (curve b in Figure 1) with
the “BEST” software program16 gave a complexation con-
stant defined by eq 2, log Ks(Zn(H-1L6)), of 19.5, from
which the apparent complexation constants, log Kapp(Zn-
(H-1L6)), were calculated to be 13.3 at pH 7.4 (dissociation
constant, Kd, is 50 fM) and 9.0 at pH 5.0, respectively (See
also Table 1). The Ks(Zn(H-1L6)) value for 11 is slightly
smaller than that for 6, possibly because of the electron-with-
drawing effect of the SO2NMe2 at the 7-position. Figure 6
shows a distribution diagram for a mixture of 5 μM 10 (L6)
and 5 μM Zn2þ obtained by calculation using the software
program “SPE”,16 indicating that Zn(H-1L6) is quantita-
tively formed at pH >5 in micromolar concentrations.
ment of Zn2þ to be made at pH 4-9.
KsðZnðH-1LÞÞ ¼ ½ZnðH-1LÞꢁ=½H-1Lꢁ½Zn2þꢁ
ð2Þ
ð3Þ
Figure 2. pH-dependent change in ε265 (a) (closed squares) and
emission intensity at 478 nm (b) (closed circles) for 10 (L6) (excita-
tion at 370 nm). [10] = 50 μM for UV-vis spectra and 5 μM for
emission spectra in comparison with that of 5 μM 9b at 478 nm
(c, open circles).
2þ
KappðZnðH-1LÞÞ ¼ ½ZnðH-1LÞꢁ=½Lꢁfree½Zn
ꢁ
free
X
½Lꢁfree
¼
½HnLꢁfree at designated pH ðn ¼ ð -1Þ∼5Þ ð4Þ
Scheme 5

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 895
Figure 4. (a) Change in UV-vis spectra of 50 μM 10 (L6) with Zn2þ
(0-2.0 equiv), pH 7.4 (10 mM HEPES) with I = 0.1 (NaNO3) at 25 ꢀC.
(b) Change in the emission spectra of 5 μM 10 (L6) with Zn2þ (0-2.0
equiv), pH 7.4 with I = 0.1 (NaNO3) at 25 ꢀC (excitation at 370 nm). The
inset shows the increase in the relative emission intensity of 10 (L6) at
Figure 3. (a) ORTEP drawing (50% probability ellipsoids) of 10 H2-
478 nm on the addition of Zn2þ
.
6
6
˚
(H-1L ) (= HL ). Selected bond lengths [A]: O(1)-C(15) 1.272(4),
S(2)-C(16) 1.756(3), S(2)-N(7) 1.629(3), N(7)-C(22) 1.476(5), N(7)-
C(21) 1.472(5). Water, hydrogen chloride and hydrogen atoms have been
omitted for clarity. (b) Space-fillingmodel of10 H2(H-1L6) (= HL6). The
crystallographic parameters are given in Table 2.
Table 2. Crystallographic Data for 10 (H2(H-1L6) (= HL6)) and 11 (Zn(H-1L6))
10(H2(H-1L6) (= HL6))
22H50.4Cl1N7O11.2S2
691.85
C2/c (#15)
35.8779(7)
19.7401(4)
21.4880(7)
118.8100(7)
13334.9(6)
16
11 (Zn(H-1L6))
chemical formula
formula weight
space group
C
C22H36N8O9S2Zn1
686.07
P21/n (#14)
11.602(5)
13.792(6)
17.941(8)
93.923(3)
2864(2)
˚
a (A)
˚
b (A)
˚
c (A)
β (deg)
3
V (A )
Figure 5. pH-Dependent change in the emission intensity of
6
˚
(Zn(H-1L5)) (open circles) and 11 (Zn(H-1L6)) (closed circles) with I =
0.1 (NaNO3) at 25 ꢀC (excitation at 338 nm for 6 and at 370 nm for 11,
[6] = [11] = 5 μM). Arbitrary unit is a.u.
Z
4
T (ꢀC)
-180
-140
˚
λ (A)
1.54187
1.378
0.71070
1.591
Dcalc (g cm-3
μ (Cu KR for 10 and
Mo KR for 11) (cm-1
R1 (I > 2σ(I))
)
3
˚
are 2.10-2.13 A, which are shorter than Zn(1)-N(2)
)
27.38
0.0685
0.1966
10.68
0.0646
0.1901
and Zn(1)-N(4). Four N atoms of the cyclen ring and
the Zn2þ ion form a tetragonal-pyramidal structure. Re-
presentative crystallographic parameters of 11 are listed in
Table 2.
wR2 (all reflections)
X-ray Crystal Structure of 11 (Zn(H-1L6) NO3 H2O).
3
3
Fluorescent Response of 10 to Various Metal Ions. We
examined the fluorescent response of 10 (5 μM) to various
metal ions in aqueous solutions at pH 7.4. As shown in
Figure 8, 10 shows selectivity of fluorescent response
to Zn2þ (32 times emission enhancement) and Cd2þ (41
times enhancement). When Cu2þ, Co2þ, Mn2þ, or Fe2þ
were added to 10 prior to the addition of Zn2þ, the
increase in emission was inhibited. The results of UV-vis
Colorless crystals were obtained from a 1:1 mixture of 10
(L6) and Zn2þ in aqueous solution at pH 10. A single-
crystal X-ray structure analysis disclosed that the Zn2þ in
11 (Figure 7) is six-coodinated by the deprotonated O(1) at
the 8-position of the quinoline ring, N(5) of the quinoline
ring, and N(1)-N(4) of the cyclen ring, similar to the pre-
viously reported Zn2þ complex 6.10 The bond lengths of
Zn(1)-O(1), Zn(1)-N(1), Zn(1)-N(3), and Zn(1)-N(5)

896 Inorganic Chemistry, Vol. 49, No. 3, 2010
Ohshima et al.
Zn2þ-Dependent Hydrolytic Uncaging of 12. As de-
scribed in the Introduction, we reported on the synthesis
of a benzenesulfonyl-caged (BS-caged) derivative of L5
(7) in our previous report.11 We then synthesized 12 in this
study (Scheme 3) and tested Zn2þ-promoted hydrolytic
uncaging, as evidenced by changes in UV-vis absorption
spectra (Figure 9). While 12 was resistant to hydrolysis in
the absence of Zn2þ at pH 7.4, the addition of Zn2þ
triggered the increase in UV absorption at 269 nm with
isosbestic points at 249, 281, and 329 nm, strongly
suggesting the hydrolytic removal of the benzenesulfonyl
moiety of 12 by Zn-bound HO- to give 11 (Zn(H-1L6),
as shown in Scheme 3. The Zn2þ-dependent hydrolysis
of 12 was also confirmed by 1H NMR and UV-vis
spectra measurements (Figures S5 and S6 in the Support-
ing Information).
Figure 6. Distribution diagram for a mixture of 5 μM 10 (L6) þ 5 μM
Zn2þ with I = 0.1 (NaNO3) at 25 ꢀC.
Supporting Information, Figure S7 shows the time
course for the hydrolysis of 12 at pH 5-9, from
which the linear rate constants, k1 (sec-1), at pH 5.0,
6.0, 7.0, 7.4, 8.0, and 9.0 were determined and plotted in
Figure 10a. The k1-pH profiles for 12 gave a kinetic pKa
value of 7.7 ( 0.2, which is in good agreement with the
kinetic pKa values of 7.7 for 7,11 suggesting that the
hydrolysis of 12 is promoted by Zn-bound HO-. The
data shown in figure 10a also indicates that the hydrolysis
rate of 12 was much greater than that of 7 because of the
electron-withdrawing effect of the SO2NMe2 group at
the 7-position.
From an Eyring plot (the 1/T-ln(k/T) profile in
Figure 10b) for the hydrolytic uncaging of 12 (BS-
caged-L6) promoted by Zn2þ or Cd2þ-bound HO-, the
Gibbs activation energy, ΔGq, the enthalpy of activa-
tion, ΔHq, and the entropy of activation, ΔSq, were
calculated and the results were summarized in Table 3.
The (-TΔSq) values for the hydrolysis of 12 with Zn2þ
and Cd2þ are nearly zero. We presume that hydropho-
bic and steric effect of the Me2NSO2 group at the
7-position of the quinoline moiety fix the conformation
of the PhSO2 group and hence restrict the transition
states of the Zn2þ and Cd2þ-promoted hydrolysis,
resulting in negligible change in the degree of freedom
(18a and 18b in Scheme 6). The smaller ΔHq value for
the Cd2þ-promoted hydrolysis of 12 than that for the
Zn2þ-promoted hydrolysis indicates the possibility of a
different mechanism operating in these two reactions.
The Cd2þ ion, which has a larger ionic radius than that of
Zn2þ, might be coordinated by the PhSO2 moiety,
resulting in a higher reactivity of the PhSO2 group in
18 (Scheme 6).
Figure 7. ORTEP drawing (50% probability ellipsoids) of 11
˚
6
(Zn(H-1L )). Selected bond lengths [A]: Zn(1)-O(1) 2.123(2), Zn(1)-
N(1) 2.129(3), Zn(1)-N(2) 2.229(3), Zn(1)-N(3) 2.126(3), Zn(1)-N(4)
2.427, Zn(1)-N(5) 2.103(3). Water and hydrogen atoms have been
omitted for clarity.
Measurement of Zn2þ Concentrations in Sample Solu-
tions by 12 (BS-caged-L6). We attempted to quantitatively
measure [Zn2þ] in some sample solutions utilizing 12.
Essentially no change in the emission of 12 was observed
in the absence of Zn2þ, as mentioned above, and the
addition of given amounts of Zn2þ (0-5 μM) followed by
incubation at 25 ꢀC for 30 min induced a nearly linear
increase in emission, as shown in Figure S8 in the Sup-
porting Information and Figure 11. In addition, it was
found that this hydrolysis reaction of 12 with Zn2þ
completes within 10 min at 50 ꢀC (Figure S6 in the
Supporting Information), indicating that 12 has consid-
erable potential for use in the quantitative analysis of
Zn2þ in solutions.
titrations of 10 with Cu2þ, Mn2þ, Co2þ, and Fe2þ sug-
gested that 10 forms 1:1 complexes with these metal
cations as shown in Figure S3 and S4 in the Supporting
Information (the log Kapp values were estimated to be
>108 for Cu2þ, Co2þ, and Mn2þ, and >106 for Fe2þ).22
Therefore, the inhibition of the emission by these metal
cataions could be attributed to competitive binding with
Zn2þ for 10 and/or quenching of the emission by para-
magnetic metal cations.23
(22) The apparent complexation of 10 with Fe2þ may be underestimated
because of the UV-vis absorption of Fe2þ itself at 250-600 nm.
(23) Lu, C.; Xu, Z.; Cui, J.; Zhang, R.; Qian, X. J. Org. Chem. 2007, 72,
3554–3557.

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 897
Figure 8. Fluorescent response of 5 μM 10 (L6) at 478 nm to 1.0 equiv of various metal cations (shaded bars) and to 1.0 equiv of Zn2þ added after 1.0 equiv
of various metals (open bars) at pH 7.4 (10 mM HEPES) with I = 0.1 (NaNO3) at 25 ꢀC.
Figure 9. Change in UV-vis spectra as the result of the Zn2þ-promoted
hydrolysis of 50 μM 12 (BS-caged-L6) in the presence of 50 μM Zn2þ at
pH 7.4 (10 mM HEPES) with I = 0.1 (NaNO3) at 37 ꢀC.
Metal Selectivity in the Hydrolytic Uncaging of 12
(BS-caged-L6). We examined the hydrolytic uncaging of
12 (50 μM) in the presence of Cd2þ, Cu2þ, Mn2þ, Co2þ
,
Fe2þ, Fe3þ, Ca2þ, and Mg2þ (50 μM each) at pH 7.4 at
25 ꢀC, as evidenced by changes in UV-vis absorption
spectra like Figure 9. We previously reported that the
Cd2þ-catalyzed hydrolysis of 7 proceeds at almost same
speed as Zn2þ-catalyzed hydrolysis.11 Concerning 12,
hydrolysis by Cd2þ was faster than that by Zn2þ. The
Figure 10. (a) pH-dependent plot of linear rate constants for the
hydrolytic uncaging of 7 (BS-caged-L5) and 12 (BS-caged-L6) in the
presence of Zn2þ at 25 ꢀC. (b) 1/T-ln(k/T) profile for the hydrolytic
uncaging of 12 (BS-caged-L6) promoted by Zn2þ or Cd2þ-bound HO-
([12] = [Zn2þ] =[Cd2þ] = 50 μMin10mMHEPES (pH 7.4) with I = 0.1
(NaNO3)).
ε269 values increased in the presence of Cu2þ, Mn2þ
,
Co2þ, and Fe2þ, indicating that these metals promoted
a partial hydrolysis of 12 (data not shown). The ε269 value
decreased with Fe3þ, possibly because of the complexa-
tion of 12 with Fe3þ without hydrolysis.
Photochemical Uncaging of 12 (BS-caged-L6). In our
previous paper,11 we reported that the benzenesulfonyl
moiety of 7 can be removed by photoirradiation in
aqueous solution to give 5 and benzenesulfonate
(Scheme 2). Thus, we examined the photoreaction of 12,
as followed by UV-vis (Figure 12) and 1H NMR (Figure
S9 in the Supporting Information) spectra. The UV-vis
absorption curve for 12 (50 μM, curve (a) in Figure 12)
changed to curve (b) after photoirradiation at 330 nm for
10 min in the absence of Zn2þ. Photoirradiation for
30 min gave curve (c), which agreed well with that of 10
(not shown). The fact that curves (a), (b), and (c) have an
isosbestic point at 322 nm suggests that photolysis of 12
affords two main products 10 (L6) and benzenesulfonate
(PhSO3-), as proven by 1H NMR experiments (described
below). Another isosbestic point at about 280 nm is not so
clear, possibly because of the formation of PhSO3- that

898 Inorganic Chemistry, Vol. 49, No. 3, 2010
Ohshima et al.
Scheme 6
Table 3. Comparison of Activation Parameters for Hydrolysis of 7 and 12
Promoted by Zn2þ or Cd2þ-bound HO- (pH 7.4 with I = 0.1 (NaNO3))a
ligands metals ΔGq (kcal/mol) ΔHq (kcal/mol) -TΔSq (kcal/mol)b
7
7
12
12
Zn2þ 24c
Cd2þ 19c
Zn2þ 21
Cd2þ 16
19c
20c
21
4.6c
-0.6c
0.2
16
0.0
a [7] = [12] = [Zn2þ or Cd2þ] = 50 μM. b T = 298 K. c From ref 11.
Figure 12. Photolysis of 50 μM 12 in pH 7.4 (10 mM HEPES with I =
0.1 (NaNO3)) at 25 ꢀC (photoirradiation at 330 nm) followed by UV-vis
absorption spectra. After 30 min of irradiation, 1.0 equiv of Zn2þ was
added into the reaction mixture.
recent study on the reaction mechanism involved in the
photolysis of 7 (BS-caged-L5), 12 (BS-caged-L6), and
some 8-quinolinyl sulfonate derivatives that have no
cyclen unit suggested that this photolysis proceeds via
homolytic S-O bond fission in the excited triplet state.24
Comparison of Intracellular Concentrations of Zn2þ
Probes. Intracellular concentrations of 2 (L2), 5 (L5), 10
(L6), and 12 (BS-caged-L6) in HeLa cells were estimated
using our previously developed methods (concentrations
of these probes: 50 μM, culture time: 0.5-1.0 h at
37 ꢀC).11 As summarized in Table 4, the intracellular
concentration of 12 (10 ( 1 μM) was higher than 10 (2 (
1 μM), possibly because of the presence of the two
Me2NSO2 groups. Negligible cell damage was observed
under these culture conditions.
Figure 11. Quantitative increase in the fluorescent intensity of 5 μM 12
(BS-caged-L6) by Zn2þ-dependent hydrolysis ([Zn2þ] = 0-6 μM)
(emission was observed 478 nm with excitation at 370 nm) at pH 7.4
(10 mM HEPES with I = 0.1 (NaNO3)) and 37 ꢀC (see also Figure S8 in
the Supporting Information). Emission spectra were obtained by rapidly
scanning the emission wavelength, to minimize photochemical uncaging
(See below).
has small absorption at 240-290 nm (ε260 ≈ 0.6 ꢀ
104 M-1 cm-1). The addition of 1.0 equiv of Zn2þ to
3
curve (c) gave curve (d) with an absorption maximum
at 269 nm, strongly suggesting the formation of 10
(Zn(H-1L6)). The quantum yield for the photolysis of
12 (500 μM) was calculated to be 7.2 ꢀ 10-5, which was
greater than that of 7 (2.3 ꢀ 10-5).11
Fluorescent Staining of Zn2þ-loaded HeLa Cells. Cell
staining experiments of Zn2þ in HeLa cells were per-
formed using 10 and 12. Figures 13a and 13b show phase
contrast and fluorescent images of HeLa cells treated with
50 μM 10. Figures 13c and 13d show data treated with
50 μM 10 for 0.5 h and then 25 μM ZnSO4 and 20 μM
Zn2þ pyrithione (Zn2þ carriers). Figures 13e and 13f show
images of cells that had been treated with 50 μM 12 for
0.5 h and then 25 μM ZnSO4 and 20 μM Zn2þ pyrithione.
These results indicate that 10 and 12 would be useful for
The photolysis of 12 (1 mM) in D2O at pD 7.4 was also
1
confirmed by H NMR experiments. After photoreac-
tions for 30 and 90 min (irradiation at 330 nm), the spectra
changed to those shown in Figures S9b and S9c in the
Supporting Information, in which new H signals ap-
1
peared, showing good coincidence with those of 10 (L6)
-
(Figure S9d in the Supporting Information) and PhSO3
.
Upon the addition of Zn2þ to the solution shown in
Figure S9c, Figure S9e was obtained, indicating the
formation of 11 (Figure S9f). It is also noteworthy that
the photolysis proceeded quantitatively with negligible
byproducts being produced in aqueous solution. Our
detecting levels of intracellular Zn2þ
.
(24) Kageyama, Y.; Oshima, R.; Sakurama, K.; Fujiwara, Y.; Tanimoto,
Y.; Yamada, Y.; Aoki, S. Chem. Pharm. Bull. 2009, 57, 1257–1266.

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 899
Table 4. Estimated Intracellular Concentrations of Zn2þ Fluorophores in HeLa Cells
2 (L2)
8 ( 1a
5 (L5)
7 (BS-caged-L5)
12 ( 3a
10 (L6)
2 ( 1
12 (BS-caged-L6)
10 ( 2
intracellular concentrations (μM)
0a
a From ref 11.
Figure 13. Microscopic images (ꢀ60 with digital zoom (ꢀ3.0)) of HeLa cells stained with 50 μM 10 without Zn2þ-pyrithione (a: phase contrast,
b: fluorescent) and 10 (c: phase contrast, d: fluorescent) and 12 (e: phase contrast, f: fluorescent) with 25 μM ZnSO4 and 20 μM pyrithione.
Conclusion
Acknowledgment. This work was supported by grants-
in-aid from the Ministry of Education, Culture, Sports,
Science and Technology (MEXT) of Japan (No.
18390009 and 19659026) and “Academic Frontier” pro-
ject for private universities: matching fund subsidy
from MEXT, 2009-2013. S.A. is grateful for grants
from Mitsubishi Chemical Corporation Fund (Tokyo),
the Terumo Life Pharmaceutical Research (Tokyo),
the Novartis Foundation (Tokyo, Japan), the Naito
Foundation (Tokyo), and the Tokuyama Science
Foundation (Tokyo). M.K. is thankful for a grant-
in-aid from MEXT of Japan (No. 20750081) and a
Sasakawa Scientific Research Grant from the Japan
Science Society.
A new Zn2þ fluorophore 10 (L6) was synthesized that has
advantages over the previously described Zn2þ probe 5 (L5).
The binding properties of 10 with Zn2þ were comparable to
that for 5, while Zn2þ sensitivity was improved considerably.
We should note that the pKa value (pKa = ca. 10-11) of
8-OH group of quinolinol moiety of 10 in the exited state is
much higher than the corresponding pKa value in the ground
state (pKa = 2.5), which is much smaller than that of 9b
(pKa = 6.1) having no cyclen unit, permittng a quantitative
measurement of Zn2þ concentrations at neutral pH. A caged
derivative of 12 (BS-caged-L6) was also synthesized, and its
uncaging properties were examined. The rapid uncaging of12
by Zn2þ-dependent hydrolysis and Zn2þ-independent photo-
lysis would be useful for Zn2þ sensing. Even more efficient
hydrolytic uncaging may be possible by introducing addi-
tional electron-withdrawing groups on the quinoline ring.24
In addition, the staining of HeLa cells containing 10 and 12
was successful. Attempts to sense Zn2þ in apoptotic cells
are currently underway.
Supporting Information Available: Figures S1-S9, tables,
and CIFs for the X-ray crystal structure analysis of 10
(H2(H-1L6) (= HL6)) and 11 (Zn(H-1L6)). This material
is available free of charge via the Internet at http://
pubs.acs.org.