MS4, a seminaphthofluorescein-based chemosensor for the ratiometric detection of Hg(II)

Abstract

The synthesis and photophysical characterization of Mercury Sensor 4, MS4, an aniline-derivatized seminaphthofluorescein-based dye that contains a pyridyl-amine-thioether ligand analogous to that employed in the previously reported Zinspy (ZS) Zn(II) sensor family (Nolan and Lippard, Inorg. Chem. 2004, 43, 8310–8317) are reported. Sensor MS4 provides single-excitation, dual-emission ratiometric detection of Hg(II) in aqueous solution. A ∼4-fold ratiometric change (λ₆₂₄/λ₅₂₄) is observed upon introduction of Hg(II) to an aqueous chloride-containing solution of MS4 at pH 8. In this milieu, MS4 shows selectivity for Hg(II) over a background of environmentally relevant alkali and alkaline earth metals, a number of divalent first-row transition metals, and its Group 12 congeners Zn(II) and Cd(II).

---

Introduction

Heavy metal ion pollution is a serious global problem that adversely affects human and environmental health. Inorganic mercury is one contaminant of particular concern.¹ Sources of mercury in the environment vary and stem from both natural and anthropogenic origins. The Mediterranean Sea, for example, has vast cinnabar deposits, which leach mercury into its water.² In the United States, ∼87% of mercury emissions result from solid waste incineration and the combustion of fossil fuels.³ Gold mining, cement production, forest fires and volcanic activity are additional sources of mercury pollution.^3,4 Upon entering freshwater and marine ecosystems, bacteria convert inorganic mercury to methylmercury, which eventually reaches the top of the food chain and bioaccumulates in large predatory fish, such as tuna and swordfish, consumed by humans.^5–7 Methylmercury is lipophilic, readily absorbed through the GI tract, and a potent neurotoxin.^8,9 A number of neurological problems have been associated with methylmercury intoxication and are perhaps best exemplified by the various manifestations of Minamata Disease, the widespread poisoning of Japanese citizens following the release of methylmercury into the Agano River during the 1950s.^10,11

Given these environmental and toxicological concerns, providing new mercury detection strategies for use in aqueous solution and in biological samples is an important goal. Fluorescence offers one convenient route to metal ion sensing since it only requires simple instrumentation and provides high temporal resolution.¹² Some desirable features of a fluorescent metal ion probe for use in environmental and/or biological monitoring include a positive and selective fluorescence response for the analyte of interest, high selectivity and sensitivity for the analyte, reversibility and water-solubility. A significant body of research on fluorescence-based Hg(II) detection exists and this work includes a fiber optic,¹³ a functionalized lipid bilayer,¹⁴ and an array of fluoroionophores.^15–26 Some of these systems are incompatible with aqueous solution, have modest selectivity for Hg(II), are irreversible or give fluorescence turn-off upon Hg(II) coordination. A mercury sensor based on the 3-aminonaphthalimide chromophore was described that gives substantial and selective fluorescence turn-on upon Hg(II) coordination in aqueous solution.²⁶ We recently presented the fluorescein-based sensor MS1, which also affords a positive and highly selective fluorescence response to Hg(II) in neutral buffered solution (Fig. 1).²⁷ Fluorescence-based Hg(II) sensing strategies that utilize proteins,²⁸ oligonucleotides²⁹ and DNAzymes³⁰ have also been reported.

Fig. 1 Sensors MS1, ZS4 and ZNP1. Fluorescein-based MS1 and ZS4 are turn-on chemosensors for Hg(II) and Zn(II), respectively, that operate by photoinduced electron transfer (PET). Sensor ZNP1 utilizes a seminaphthofluorescein platform and affords single-excitation dual-emission ratiometric detection of Zn(II).

Most fluorescent Hg(II) ionophores are intensity based and rely either on fluorescence turn-off or turn-on following Hg(II) coordination. Ratiometric fluorescence detection, which compares fluorescence intensity ratios at two different wavelengths before and following analyte recognition, offers another means of Hg(II) sensing. It is arguably most useful in quantification and in applications involving inhomogeneous samples.³¹ In what follows, we present the preparation and photophysical characterization of the ratiometric Hg(II) sensor MS4. This chemosensor is based on an asymmetrical seminaphthofluorescein platform,³² gives a positive fluorescence response upon Hg(II) coordination and offers single-excitation dual-emission ratiometric detection of Hg(II) in aqueous solution.

Experimental

Reagents

Anhydrous methanol, chloroform, 1,2-dichloroethane (DCE) and sodium triacetoxyborohydride were purchased from Aldrich and used as received. The seminaphthofluorescein carboxaldehyde, 1, and aniline ligand 2 were synthesized according to previously published procedures.^32,33

Methods

Analytical thin-layer chromatography (TLC) was performed by using Merck F254 silica gel 60 plates (0.25 mm thickness) and read with UV illumination. Whatman silica gel-60 plates (1 mm thickness) were used as the solid phase for preparative TLC. NMR spectra were collected by using a Varian 300 MHz spectrophotometer operating at 283 K and the spectra were referenced to internal standards. IR spectra were collected on an Avatar FTIR instrument.

Preparation of MS4

2-{11-[(2-{[(2-Ethylsulfanylethyl)pyridin-2-ylmethylamino]methyl}phenylamino)methyl]-10-hydroxy-3-oxo-3H-benzo[c]xanthen-7-yl}benzoic acid (3, Mercury Sensor 4, MS4). To 4.5 mL of 7 : 2 CHCl₃–MeOH were added 1 (33 mg, 0.081 mmol) and 2 (24 mg, 0.081 mmol), and the resulting purple–brown solution was stirred at room temperature for 24 h. The reaction was diluted with 1 mL of DCE and NaB(OAc)₃H (27 mg, 0.13 mmol) was added; the reaction was stirred for an additional 24 h. The solvents were removed in vacuo and preparative TLC on silica gel (9 : 1 CHCl₃–MeOH) afforded pure MS4 as a deep purple solid (23 mg, 42%). TLC R_f = 0.55 (9 : 1 CHCl₃–MeOH); mp > 325 °C, dec. ¹H NMR (CD₃OD, 300 MHz) δ 0.84 (3H, t), 1.92 (2H, q), 2.22–2.46 (4H, m), 3.46–3.65 (4H, m), 4.70 (2H, q), 6.65 (1H, t), 6.73 (1H, d), 6.87 (1H, dd), 7.03–7.11 (3H, m), 7.17 (2H, t), 7.21–7.30 (5H, m), 7.38 (1H, t), 7.60–7.69 (2H, m), 8.15 (1H, dd), 8.21 (2H, m). FTIR (KBr, cm⁻¹) 3423 (s, br), 3050 (m), 2921 (m), 2849 (m), 1759 (w), 1639 (s), 1606 (m), 1587 (m), 1504 (w), 1469 (m), 1433 (m), 1374 (m), 1328 (w), 1297 (w), 1232 (w), 1148 (m), 1115 (m), 1071 (m), 1042 (m), 1000 (w), 878 (w), 846 (w), 830 (w), 791 (w), 748 (w), 697 (w), 672 (w), 607 (w), 523 (w).

Spectroscopic reagents and methods

Millipore water was used to prepare all aqueous solutions. Puratonic grade KCl was purchased from Calbiochem and buffers from either Calbiochem or Sigma. Mercury stock solutions (10 mM) were prepared from 99.999% anhydrous HgCl₂ (Aldrich) and water. DMSO stock solutions (1 mM) of MS4 were prepared, stored as aliquots at −25 °C, and thawed in the dark immediately before use. With the exception of the pH study, measurements were conducted in aqueous buffer at 100 mM ionic strength (100 mM KCl) with the pH maintained at 7 (PIPES buffer), 8 (HEPES buffer), 9 (CHES buffer), or 11 (CAPS buffer). Fluorescence spectra were collected with a Photon Technology International (Lawrenceville, NJ) Quanta Master 4L-format scanning spectrofluorimeter equipped with a LPS-220B 75 W xenon lamp and power supply, A-1010B lamp housing with integrated igniter, switchable 814 photon-counting/analog PMT detector, and a MD-5020 motor driver. Optical absorption spectroscopy was performed by using a Cary 1E double-beam scanning spectrophotometer. All samples were contained in either 1 or 3 mL quartz cuvettes (Starna) and maintained at 25 °C by means of a circulating water bath.

Spectroscopic measurements

With the exception of the pK_a titration, which was performed in duplicate, all measurements were repeated a minimum of three times. Quantum yields were determined relative to fluorescein in 0.1 N NaOH (Φ = 0.95)³⁴ and the reported values are the averages of four independent measurements. Extinction coefficients were determined over a concentration range of 10 to 1 µM and are the averages of four independent titrations. The effect of pH on the fluorescence emission of apo MS4 was determined by adding aliquots of 6, 2, 1 and 0.5 N HCl to a 5 µM solution of MS4 in 10 mM KOH, 100 mM KCl (pH ∼ 12.5). Experimental details for the metal ion selectivity and reversibility experiments are available elsewhere.²⁷

Results and discussion

Synthesis

Seminaphthofluorescein dye MS4 was synthesized according to methodology previously developed in our laboratory for the assembly of asymmetrical fluorescein-based sensors and for the seminaphthofluorescein-based ratiometric Zn(II) sensor ZNP1 (Fig. 1).^35,32 These synthetic routes involve Schiff base condensation of an asymmetric fluorophore containing an aldehyde functional group and an aniline-derivatized ligand moiety designed for metal ion complexation, followed by imine reduction under mild conditions. Scheme 1 illustrates the final reaction step that afforded sensor MS4 from previously reported compounds 1 and 2, although the full synthesis of this fluorophore involves nine steps starting from commercially available materials. Schiff base condensation of 1 and 2 followed by reduction of the intermediate imine using NaB(OAc)₃H afforded pure MS4 as a deep purple solid in moderate yield after preparative TLC purification on silica gel using CHCl₃–MeOH.

Scheme 1

Spectroscopic properties of MS4

The optical absorption spectrum of MS4 exhibits a maximum centered at 548 nm (ε = 9000 M⁻¹ cm⁻¹) with a shoulder at 508 nm (ε = 7000 M⁻¹ cm⁻¹) at pH 8 and 100 mM ionic strength (50 mM HEPES, 100 mM KCl). Introduction of excess Hg(II) to a solution of MS4 causes a slight 3 nm blue-shift of the 548 nm band and an increase in visible absorption with a maximum at 545 nm (ε = 11 200 M⁻¹ cm⁻¹) and a shoulder centered at 508 nm (ε = 8300 M⁻¹cm⁻¹). Upon excitation at 499 nm, the emission spectrum of apo MS4 shows two local maxima centered at 524 and 613 nm. In the presence of EDTA to complex any potentially interfering metal ions, free MS4 has a quantum yield of 0.05 (50 mM HEPES, 100 mM KCl, pH 8). Coordination of MS4 to Hg(II) causes the quantum yield to double to 0.10. Enhancement of the 613 nm emission band occurs upon Hg(II) coordination, whereas no discernible intensity change occurs at 524 nm (Fig. 2). This feature allows for single-excitation dual-emission ratiometric detection of Hg(II) by comparison of the intensity ratios at 624 and 524 nm (λ₆₂₄/λ₅₂₄) before and after Hg(II) binding. At pH 8 the ratio (λ₆₂₄/λ₅₂₄) increases ∼4-fold upon Hg(II) coordination. The extent of ratiometric enhancement shows some pH dependence. A ∼2-fold ratiometric change is observed at pH 7 (50 mM PIPES, 100 mM KCl) following Hg(II) coordination. The magnitude of the Hg(II) response at pH 9 (50 mM CHES, 100 mM KCl) is comparable to that observed at pH 8 and, in more alkaline solution, an ∼8-fold ratiometric change occurs (50 mM CAPS, 100 mM KCl, pH 11).

Fig. 2 Response of MS4 to Hg(II) at pH 8 (50 mM HEPES, 100 mM KCl). Dotted line: emission of free MS4. Solid line: emission change upon introduction of 10 equiv. Hg(II) to the solution of MS4. The concentration of MS4 was 5 µM and excitation was provided at 499 nm.

Effect of pH on free MS4 emission

Given the seminaphthofluorescein platform, which exhibits pH-dependent emission, and the presence of an aniline nitrogen atom that can function as a photoinduced electron transfer (PET) switch,¹² we anticipated that the emission of free MS4 would be pH-dependent. The effect of pH on the integrated emission of apo MS4 is shown in Fig. 3. A small change occurs between pH ∼12 to ∼6, the integrated emission rising by only ∼10% across this range. This fluorescence increase is significantly less than those observed for fluorescein-based sensors such as MS1 and the ZP and ZS dyes. One possible explanation for the relatively stable fluorescence of MS4 between pH 6–12 is that protonation of both the aniline nitrogen atom, which would enhance fluorescence due to quenching of PET,^33,35,36 and the seminaphthofluorescein platform to form the monoanion, which quenches its fluorescence, overlap in this regime and effectively cancel each other. The fluorescence of MS4 is quenched at low pH, the fluorescence decrease having a pK_a value of 4.7 (Fig. 3). This latter behavior is attributed to protonation of the seminaphthofluorescein platform and formation of a non-fluorescent species.

Fig. 3 Effect of pH on the integrated emission of apo MS4. A 5 µM solution of MS4 was prepared in 10 mM KOH, 100 mM KCl (pH ∼12.5) and the emission spectrum recorded. The pH was decreased in increments of ∼0.25 by addition of 6, 2, 1, or 0.5 N HCl and the emission spectrum collected at each point. Excitation was provided at 499 nm.

Fig. 4 depicts the effect of pH on the emission profile of MS4. From pH ∼12 to pH ∼10, emission is essentially insensitive to pH and exhibits two local maxima centered at 524 and 613 nm (spectra not shown). Upon further lowering of the pH to ∼6.5, the emission decreases and increases at 620 and 558 nm, respectively. Below pH ∼6.5, MS4 displays one emission maximum centered at 524 nm. Subsequent lowering of the pH causes fluorescence quenching with no shift in the wavelength of the maximum emission.

Fig. 4 Effect of pH on the emission spectrum of apo MS4. These spectra correspond to the experiment described in the caption of Fig. 3. Only negligible changes in the emission spectrum occur between pH 12.5 and 9.7 (data not shown).

MS4 fluorescence dependence on chloride ion

The fluorescence response of MS4 to Hg(II) depends on the presence of chloride ion. Fig. 5 depicts the effect of chloride ion on the emission of the MS4 : Hg(II) complex at pH 8. In the absence of chloride ion, a modest, ∼2-fold change in (λ₆₂₄/λ₅₂₄) occurs due to increased emission centered at 613 nm. Addition of aqueous KCl or NaCl to a solution of MS4 and Hg(II) causes the intensity of the 613 nm band to increase immediately following mixing. The presence of chloride ion has no effect on the emission of the free dye. Analogous behavior was previously observed for the fluorescein-based Hg(II) sensor MS1 (Fig. 1).²⁷ We tentatively propose that formation of a Hg–Cl bond may alter the relative disposition of molecular orbitals involved in the fluorescence response and thereby modulate the emission of the Hg(II)-fluorophore complex. Efforts to investigate this notion are in progress.

Fig. 5 Effect of chloride ion on the emission of the MS4 : Hg(II) complex. Top plot: The dotted line is the emission spectra of 10 µM MS4. The solid lines indicate the emission response of MS4 to 10 equiv. of Hg(II) in the absence and presence of chloride ion (50 mM HEPES, pH 8). Substantial fluorescence increase is observed upon addition of an aliquot of 1 M KCl (final concentration = 100 mM). Bottom plot: Addition of KCl to free MS4 in 50 mM HEPES at pH 8. The presence of chloride ion has no effect on the emission of the free dye. Excitation was provided at 499 nm.

Metal ion binding studies

The selectivity of MS4 for Hg(II) over a number of environmentally relevant alkali and alkaline earth metals, divalent transition metals and the Group 12 metals Zn(II) and Cd(II) was investigated at pH 8 and 100 mM ionic strength (50 mM HEPES, 100 mM KCl) and the results are depicted in Fig. 6. The ratiometric fluorescence response of MS4 for Hg(II) is unaffected by millimolar concentrations of Na(I), Li(I), Rb(I), Mg(II), Ca(II), Sr(II), and Ba(II). Introduction of 50 equiv. of the divalent first-row transition metals Mn(II), Fe(II), Co(II), Ni(II) and Cu(II) produces no significant change in the emission spectrum of MS4 and the chemosensor is selective for Hg(II) over the former three metal ions. MS4 preferably binds to Ni(II) and Cu(II), in agreement with our previous studies of the analogous ZS sensors, which indicated that Zn(II) cannot displace Ni(II) or Cu(II) from the ZS coordination sphere,³³ and in accord with earlier work on thioether-based ligands.³⁷ Modest ratiometric enhancement occurs upon Zn(II) and Cd(II) coordination to MS4 but, as anticipated from the previous ZS study, Hg(II) readily displaces Zn(II) and Cd(II) from the MS4 coordination sphere. The positive response of MS4 to Hg(II) also occurs in the presence of 20 equiv. of Cr(III) or Pb(II).

Fig. 6 Fluorescence response of MS4 to selected cations in aqueous solution at pH 8 (50 mM HEPES, 100 mM KCl). The response is normalized with respect to the emission of the free dye at 624 nm (F₀). Top plot: emission change at 624 nm upon addition of each cation: 1, Li(I); 2, Na(I); 3, Rb(I); 4, Mg(II); 5, Ca(II); 6, Sr(II); 7, Ba(II); 8, Cr(III); 9, Mn(II); 10, Fe(II); 11, Co(II); 12, Ni(II); 13, Cu(II); 14, Zn(II); 15, Cd(II); 16, Hg(II); 17, Pb(II). With the exception of Cr(III) and Pb(II) where 20 equiv. of cation were added, each solution contained 50 equiv. of the cation of interest. Bottom plot: response of MS4 to 50 equiv. Hg(II) in the presence of the selected cations. The grey bars are identical to those in the top plot and the black bars represent the change in emission at 624 nm that occurs option introduction of Hg(II) to the solutions containing MS4 and the selected cation. No discernable change in emission at 524 nm was observed. The response of MS4 is also unaffected by 100 mM Li(I), Na(I), Mg(II) and Ca(II), 10 mM Rb(I) and Sr(II) and 1 mM Ba(II). The concentration of MS4 was 5 µM and excitation was provided at 499 nm.

Binding of MS4 to Hg(II) is readily reversed by addition of the heavy metal ion chelator N′,N′,N″,N″-tetra(2-picolyl)ethylenediamine (TPEN), as illustrated in Fig. 7. Upon introduction of TPEN to a solution containing the MS4 : Hg(II) complex, the fluorescence emission decreases to background levels immediately following mixing. Subsequent addition of Hg(II) results in fluorescence enhancement and this off–on cycling can be repeated at least three times.

Fig. 7 Reversibility of MS4 binding to Hg(II) by the addition of TPEN. The top set of solid, dotted and dashed lines indicate the emission resulting from three additions of Hg(II) to a solution of MS4 at pH 8 (50 mM HEPES, 100 mM KCl). The bottom set of solid, dotted and dashed lines represent the emission of free MS4 before addition of Hg(II) and after successive additions of Hg(II) and TPEN. The concentration of MS4 was 5 µM and 10 equiv. of Hg(II) or TPEN was added. The increasing emission in the blue region of the spectrum is due to TPEN (control data not shown).

Comparison of MS4 to related group 12 metal ion sensors

As part of our research initiative to design and utilize water-soluble fluorescent Hg(II) sensors, we initially prepared and characterized mercury sensor MS1, shown in Fig. 1.²⁷ This fluorescein-based off–on Hg(II) switch operates via PET whereby the aniline nitrogen atom quenches fluorescein emission in the metal free form. Coordination of MS1 to Hg(II) alleviates PET quenching and results in ∼5-fold fluorescence enhancement in aqueous chloride-containing solutions at neutral pH. Sensor MS4 differs from MS1 in both the nature of the aniline-derivatized ligand and the fluorophore platform. Regarding the former, a comparison of the selectivity of MS1 and MS4 shows that substitution of a thioether moiety by a pyridyl group abrogates Hg(II) selectivity over Ni(II). We also prepared and characterized two fluorescein-based dyes containing pyridyl-amine-thiol ligands that give positive and selective fluorescence response to Hg(II) in aqueous media. These dyes also prefer Ni(II) over Hg(II) in neutral buffered solution.³⁸ These comparisons suggest that maintaining a soft, sulfur-rich coordination sphere is important for achieving selectivity for Hg(II) over Ni(II).

The second difference between MS1 and MS4 is the fluorophore platform, the choice of which is critical for obtaining desired photophysical properties. A potential advantage of the seminaphthofluorescein platform used in the assembly of MS4 is its ability to offer ratiometric metal ion detection, which is useful for studies involving quantification and inhomogeneous samples. MS1 and MS4 have similar quantum efficiencies in the metal-free and -bound forms. Because of the relatively high extinction coefficient of fluorescein, sensor MS1 exhibits considerably greater brightness (Φ × ε) than seminaphthofluorescein-based sensor MS4.

In a previous investigation of fluorescein-based dyes appended with pyridyl-amine-thioether ligands, one of which is shown in Fig. 1, selectivity for Hg(II) over Zn(II) was observed.³³ Fluorescence enhancement, however, did not occur upon Hg(II) coordination to ZS1–ZS3, and only a slight increase in emission was observed for ZS4. This behavior contrasts that of sensor MS4, which gives a positive response to Hg(II). These results indicate that proper matching of fluorophore and ligand moieties is necessary to achieve the desired fluorescence response.

Conclusion

The preparation, photophysical characterization and metal ion binding studies of seminaphthofluorescein-based Hg(II) sensor MS4 are described. This chemosensor contains an aniline-derivatized pyridyl-amine-thioether ligand and gives a ∼4-fold ratiometric (λ₆₂₄/λ₅₂₄) response upon Hg(II) coordination in aqueous chloride-containing media at pH 8. MS4 is selective for Hg(II) over a range of environmentally relevant alkali and alkaline earth metals, a number of divalent first-row transition metals and over Zn(II) and Cd(II).

Acknowledgements

This work was supported by Grant GM65519 from the National Institute of General Medical Sciences. Spectroscopic instrumentation at the MIT DCIF is maintained with funding from NIH Grant 1S10RR13886-01 and NSF Grants CH3-9808063, DBI9729592, and CHE-9808061. E.M.N. thanks NDSEG for a graduate fellowship.

References

1. D. W. Boening, Chemosphere, 2000, 40, 1335–1351.
2. A. Renzoni, F. Zino and E. Franchi, Environ. Res., 1998, 77, 68–72.
3. Mercury Update: Impact on Fish Advisories, EPA Fact Sheet EPA-823-F-01-011, Office of Water, Washington, DC, 2001.
4. O. Malm, Environ. Res., 1998, 77, 73–78.
5. M. Nendza, T. Herbst, C. Kussatz and A. Gies, Chemosphere, 1997, 35, 1875–1885.
6. H. H. Harris, I. J. Pickering and G. N. George, Science, 2003, 301, 1203.
7. S. Hardy and P. Jones, J. Chromatogr., A, 1997, 791, 333–338.
8. T. W. Clarkson, L. Magos and G. J. Myers, N. Engl. J. Med., 2003, 349, 1731–1737.
9. P. M. Bolger and B. A. Schwetz, N. Engl. J. Med., 2002, 347, 1735–1736.
10. T. Takeuchi, N. Morikawa, H. Matsumoto and Y. Shiraishi, Acta Neuropathol., 1962, 2, 40–57.
11. M. Harada, Crit. Rev. Toxicol., 1995, 25, 1–24.
12. A. P. de Silva, H. Q. N. Gunaratne, T. Gunnlaugsson, A. J. M. Huxley, C. P. McCoy, J. T. Rademacher and T. E. Rice, Chem. Rev., 1997, 97, 1515–1566.
13. X.-B. Zhang, C.-C. Guo, Z.-Z. Li, G-L. Shen and R.-Q. Yu, Anal. Chem., 2002, 74, 821–825.
14. D. Y. Sasaki and B. E. Padilla, Chem. Commun., 1998, 1581–1582.
15. A. W. Czarnik, Acc. Chem. Res., 1994, 27, 302–308.
16. M.-Y. Chae and A. W. Czarnik, J. Am. Chem. Soc., 1992, 114, 9704–9705.
17. J. Yoon, N. E. Ohler, D. H. Vance, W. D. Aumiller and A. W. Czarnik, Tetrahedron Lett., 1997, 38, 3845–3848.
18. J. D. Winkler, C. M. Bowen and V. Michelet, J. Am. Chem. Soc., 1998, 120, 3237–3242.
19. G. Hennrich, H. Sonnenschein and U. Resch-Genger, J. Am. Chem. Soc., 1999, 121, 5073–5074.
20. L. Prodi, C. Bargossi, M. Montalti, N. Zaccheroni, N. Su, J. S. Bradshaw, R. M. Izatt and P. B. Savage, J. Am. Chem. Soc., 2000, 122, 6769–6770.
21. H. Sakamoto, J. Ishikawa, S. Nakao and H. Wada, Chem. Commun., 2001, 2395–2396.
22. K. Rurack, M. Kollmannsberger, U. Resch-Genger and J. Daub, J. Am. Chem. Soc., 2000, 122, 968–969.
23. J. H. Kim, A.-R. Hwang and S.-K. Chang, Tetrahedron Lett., 2004, 45, 7557–7561.
24. R. Métivier, I. Leray and B. Valeur, Chem. Eur. J., 2004, 10, 4480–4490.
25. A. B. Descalzo, R. Martínez-Máñez, R. Radeglia, K. Rurack and J. Soto, J. Am. Chem. Soc., 2003, 125, 3418–3419.
26. X. Guo, X. Qian and L. Jia, J. Am. Chem. Soc., 2004, 126, 2272–2273.
27. E. M. Nolan and S. J. Lippard, J. Am. Chem. Soc., 2003, 125, 14270–14271.
28. P. Chen and C. He, J. Am. Chem. Soc., 2004, 126, 728–729.
29. A. Ono and H. Togashi, Angew. Chem., Int. Ed., 2004, 43, 4300–4302.
30. J. M. Thomas, R. Ting and D. M. Perrin, Org. Biomol. Chem., 2004, 2, 307–312.
31. R. Y. Tsien and M. Poenie, Trends Biochem. Sci., 1986, 11, 450–455.
32. C. J. Chang, J. Jaworski, E. M. Nolan, M. Sheng and S. J. Lippard, Proc. Natl. Acad. Sci. USA, 2004, 101, 1129–1134.
33. E. M. Nolan and S. J. Lippard, Inorg. Chem., 2004, 43, 8310–8317.
34. J. H. Brannon and D. Magde, J. Phys. Chem., 1978, 82, 705–709.
35. E. M. Nolan, S. C. Burdette, J. H. Harvey, S. A. Hilderbrand and S. J. Lippard, Inorg. Chem., 2004, 43, 2624–2635.
36. S. C. Burdette, C. J. Frederickson, W. Bu and S. J. Lippard, J. Am. Chem. Soc., 2003, 125, 1778–1787.
37. T. H. Cooper, M. J. Mayer, K.-H. Leung, L. A. Ochrymowycz and D. B. Rorabacher, Inorg. Chem., 1992, 31, 3796–3804.
38. E. M. Nolan and S. J. Lippard, unpublished results.

---