The effects of the increasing number of the same chromophore on photosensitization of water-soluble cyclen-based europium complexes with potential for biological applications

Abstract

Five water-soluble cyclen-based europium complexes incorporated with the pendant chelating chromophore, 2-methyl-4-(2-(4-propoxyphenyl)ethynyl)pyridine (L), as the antenna in different ratios have been synthesized and full structural characterization, photophysical measurement and in vitro biological studies were carried out. The antenna-to-metal ratios were 4 (Eu-4L), 3 (Eu-3L), 2 (1,4-disubstituted Eu-o-2L and 1,7-disubstituted Eu-p-2L) and 1 (Eu-1L). It was found that the definitive distinction between Eu-o-2L and Eu-p-2L could be interpreted from the emission spectra of Eu³⁺ at room and low temperatures. With all the results from subcellular stability testing (via titrations), in vitro imaging, cytotoxicity assays and cellular uptake profiles, Eu-4L with an overall +3 charge demonstrated subcellular localization towards lysosomes and was a potential compound for a selective time-resolved bio-labelling probe.

---

Introduction

The lanthanide series, otherwise known as rare earth elements, are the first row f-block metals and are well-known for their unique photophysical and chemical properties.¹ The factors to be considered when selecting the chelating systems for long-lived, hypersensitive, and sharply luminescent lanthanide materials are generally: (i) energy transfer efficiency, (ii) quantum yield efficiency, and (iii) the nature of the metal complexation.^2,3 In reports in the literature, the most common ligands in use are nitrogen-based macrocyclic ligands, such as cyclen, triazane, triazine, and porphyrin which act as the antenna for transferring energy to the lanthanide ions and thus overcome the Laporte forbidden lanthanide 4fN–4fN transition.^4–6 In bioimaging, all the above lanthanide complexes formed have significantly advanced the field (as time-resolved imaging agents), however, together with the development of lanthanide bioprobes, systematic studies of using them simultaneously in aqueous and in vitro studies are still very rare. Most of the studies in the literature focus on the relationship of their skeleton structure–photophysical properties and then only with the use of the same antenna.^7,8 Data about the effects because of the different number of or the different spatial arrangement of the antenna are, therefore, still lacking. In 2010, Walton et al. reported a series of europium(III) complexes bearing one to four coordinated azaxanthone groups.⁹ Recently, the control of the number of terdentate N-donor pyridine ligands to improve the antenna effect of lanthanide complexes has also been reported and their quantum efficiencies and kinetic data have been compared between trivalent lanthanide cations and the number of N-ligands under different experimental conditions.¹⁰ However, to develop lanthanide complexes for practical biological applications, full characterization of them needs to be done ambiguously in aqueous solution, and this is particularly difficult. The use of proton nuclear magnetic resonance spectroscopy (¹H-NMR) or high resolution mass spectra (HRMS) is more challengingly, and these methods do not provide a definitive distinction for the isomerism between the two antennae in a cyclen-based europium system. Because of the room and low temperature Eu³⁺ luminescence, it has been possible in this work to give a clear identification of them. In this paper is reported the synthesis of five new europium complexes based on the cyclen skeleton with different numbers of 2-methyl-4-(2-(4-propoxyphenyl)ethynyl)pyridine (L) antennae attached (Fig. 1). The room and low temperature photophysical measurements were used in solution to gain insights from the coordination modes against the hypersensitive europium that can provide essential information for the further development of lanthanide complexes as luminescent bioprobes.

Fig. 1 The molecular structures of the ligand L: 2-methyl-4-(2-(4-propoxyphenyl)ethynyl)pyridine and their cyclen-based europium complexes.

Results and discussion

Synthesis of cyclen-based europium complexes with various numbers of the same antenna

Europium complexes Eu-1L, Eu-o-2L and Eu-p-2L were prepared using a similar synthetic strategy. As shown in Scheme 1, alcohol 1 was converted to bromide 2 with a yield of 72% via the mesylation/bromination protocol.¹⁰ Alkylation of bromide 2 with cyclen derivatives¹¹ 3, 4 and 5 gave ligands 1L, o-2L and p-2L, respectively, with yields of 62–75%. After acid hydrolysis of the t-butyl esters with trifluoroacetic acid (TFA), the Eu complexes were prepared by treatment of europium(III) chloride hexahydrate (EuCl₃·(H₂O)₆) at a neutral pH. Recrystallization of the crude Eu complexes with diethyl ether gave Eu-1L with a yield of 75%, Eu-o-2L with a yield of 73% and Eu-p-2L with a yield of 73%. Synthesis of Eu complexes Eu-3L and Eu-4L are shown in Scheme 2. Direct alkylation of cyclen with 3.6 equivalents of bromide 2 generated the tri-substituted cyclen derivative (6) and ligand 4L with yields of 30% and 45%, respectively. Eu complex Eu-4L was prepared by the treatment of ligand 4L with europium chloride salt in acetonitrile (MeCN). Ligand 3L was obtained via alkylation of 6 with t-butyl 2-bromoacetate. Deprotection of the t-butyl ester group followed by complexation with the europium chloride salt gave Eu-3L with a yield of 63%. All the ligands (1L, o-2L, p-2L, 3L and 4L) were characterized unambiguously using ¹H-NMR, ¹³C-NMR and HRMS and the corresponding Eu complexes were characterized unambiguously using HRMS and high-performance liquid chromatography (HPLC).

Scheme 1 Synthesis of Eu-1L, Eu-o-2L and Eu-p-2L.

Scheme 2 Synthesis of Eu-3L and Eu-4L.

The reverse-phase HPLC analysis of these five europium complexes was carried out at room temperature using an Agilent ZORBAX SB-C₁₈ Stable Bond Analytical 4.6 × 150 mm 5 μm column. The mobile phase was 0.1% formic acid in Milli-Q (mQ; Merck Millipore) water and 0.1% formic acid in MeCN solvent system, and the flow rate was 1.0 mL min⁻¹. The solvent gradient program is listed in Table S1 [see ESI†]. Under the experimental conditions given previously, the retention times of the five europium complexes were: Eu-1L in 10.48 min, Eu-p-2L in 10.93 min, Eu-o-2L in 11.04 min, Eu-3L in 10.62 min and Eu-4L in 11.33 min.

The ligands and the europium complexes were analyzed using ¹H-NMR spectroscopy. In deuterated methanol (CD₃OD) at room temperature, the ¹H-NMR spectrum of ligand 4L in CD₃OD displays broad signals for both the azamacrocyclic protons and the pridinyl protons. Furthermore, the Eu-4L exhibits a well-resolved ¹H-NMR spectrum, ranging over 35 ppm because of the paramagnetism of the europium ion (Fig. 2). Only one set of the paramagnetically shifted resonance signals was observed, which is in agreement with local C₄ symmetry of Eu-4L. The pendant arms (ranging from 2.5 to 3.6 ppm) at room temperature became well-resolved upon complexation because of the rigidification of the macrocyclic ligand. As a result, the azamacrocyclic protons and pyridinyl protons of the pendant arms, which are in the close vicinity of the metal, are strongly field-shifted (at 15.0 ppm, −0.3 ppm, −6.5 ppm, −7.8 ppm, −8.2 ppm, −13.3 ppm, −14.0 ppm and −16.5 ppm) and show large J coupling constants. In addition, directly coordinating to the europium ion via the available donor nitrogen in the π system results in high-field shift of the protons in the pyridine moiety (in the 5.5–6.4 ppm range). The mono-cationic complex, Eu-p-2L, gives two sets of paramagnetically shifted resonance signals in CD₃OD at room temperature (Fig. 3), which is consistent with the presence of two diastereoisomeric species with C₂ symmetry, and the shifts showed a similar correspondence with those observed for Eu-4L. The CH₂ protons at the α position to the carboxylates in Eu-p-2L occupy the positions at 24.5 ppm, 23.60 ppm, −21.1 ppm and −25.7 ppm. For the complexes Eu-1L, Eu-o-2L and Eu-3L, the ¹H-NMR spectral width corresponds well with that of Eu-p-2L. But the spectrum is rather broad and not well-resolved, because of the presence of chemical exchange among the various isomers (see ESI†).¹²

Fig. 2 ¹H-NMR (500 MHz, 298 K) spectrum of Eu-4L in CD₃OD.

Fig. 3 ¹H-NMR (500 MHz, 298 K) spectrum of Eu-p-2L in CD₃OD.

Photophysical properties of the five water-soluble europium complexes

The choice of ligand and number of ligands for complexation plays a key role in constructing efficient luminescent lanthanide complexes, especially for biological applications (i.e., overall charge for cellular uptake, solubility for biological media). This is a very interesting and important area which needs to be explored, but there are very limited studies which have all five complexes soluble in water with the change of number and position of ligand in complexes. Therefore, it gave us the opportunity to run comprehensive studies to evaluate the effect of increasing the number of the same antenna on the cyclen-based skeleton as well as to elucidate the spatial arrangement in aqueous solution and even in vitro. The electronic absorption, emission and excitation spectra were recorded for the five Eu³⁺ complexes at 298 K (Eu-o-2L and Eu-p-2L are also recorded at 77 K) in the aqueous solution. The aim was to understand the effect on the triplet state of the increased ligand(s) number in these five complexes. The gadolinium (Gd) motif complexes of Eu-1L, Eu-o-2L, Eu-p-2L, Eu-3L and Eu-4L were synthesized and their phosphoresence bands at 77 K were monitored.

Absorption and excitation spectra of the five water-soluble europium complexes

The ultraviolet (UV) absorption spectra of the five complexes are shown in Fig. 4 and their absorption coefficients are listed in Table 2. The absorption band(s) of these complexes originated from the antenna. All five complexes showed bands at 481 cm⁻¹ red shifted after europium complexation. Eu-4L with four antennae had, as expected, demonstrated the molar extinction coefficient at 325 nm (Eu-1L: 4318.6 M⁻¹ cm⁻¹; Eu-o-2L: 11 413.5 M⁻¹ cm⁻¹; Eu-p-2L: 1063.37 M⁻¹ cm⁻¹; Eu-3L: 14 824.3 M⁻¹ cm⁻¹; Eu-4L: 20 454.6 M⁻¹ cm⁻¹). For Eu-o-2L and Eu-p-2L, two antennas were conjugated at the 1,4 and 1,7 positions, respectively. Eu-o-2L has shown a slightly larger antenna coefficient (energy transfer coefficient) as the two antenna in the 1,4 position are closer. In addition, the 325 nm absorption bands in Eu-3L and Eu-4L are very strong because of the antenna. The excitation spectra of all five complexes under the same experimental conditions (concentration = 1 μM in water) were also recorded and have showed a broad excitation band at 340 nm, which is similar to the absorption band.

Fig. 4 Excitation (upper) and UV absorption (lower) spectra of the five complexes in aqueous solution (10 μM in water).

Room temperature emission spectra of the five water-soluble europium complexes (Eu-1L, Eu-o-2L, Eu-p-2L, Eu-3L and Eu-4L)

There are limited examples for the comparison of the photophysical properties of cyclen-based lanthanide complexes with various numbers of the same antenna in the same solution medium, especially in aqueous solution. The emission spectra of the five europium complexes were measured with excitation at 330 nm in an aqueous solution and are shown in Fig. 5 which has been arbitrarily scaled. All the ⁵D₀ → ⁷F_J transitions (J = 0 to 4) of europium in the five complexes were detected and marked in Fig. 5. The increases in the ⁷F₁ (magnetic dipole) to ⁷F₂ (electric dipole) ratio are listed in the Table 1, indicating the change from unsymmetrical Eu-1L to symmetrical Eu-4L. The ⁵D₀ to ⁷F₀ transition at 580 nm diminished in relative intensity over the series and is weakest for Eu-4L that is eight coordinated in solution and in C₄ symmetry. This is largely correlated to the ¹H-NMR spectrum shown in Fig. 2. Furthermore, this has been supported by the q-value of Eu-4L which is determined to be zero (Table 2). The result is overall correlated to the findings by Walton et al. as well in C₄ symmetric cyclen-based europium with three identical antennae and the ratio of ⁵D₀ to ⁷F₁ to ⁷F₂ in Eu-1L, Eu-p-2L, and Eu-3L are almost constant.⁹ Walton et al. did not report the two antennae at the adjacent position in cyclen skeleton. Our measurements have found that the ⁵D₀ to ⁷F₁ to ⁷F₂ in Eu-o-2L are similar to the other three, thereby completing the missing data in the literature. The emission lifetime and quantum yield of the five complexes are recorded in aqueous medium and the increase in the number of antenna also increased the quantum yield, and also the overall absorption coefficients (Table 2).

Fig. 5 Emission spectra of five europium complexes of Eu-1L, Eu-p-2L, Eu-o-2L, Eu-3L and Eu-4L in aqueous solution, 1 μM (λ_ex = 330 nm).

Table 1 The ratios of ⁵D₀ → ⁷F_J (J = 0–4) emission bands of five water-soluble europium complexes (Eu-1L, Eu-o-2L, Eu-p-2L, Eu-3L and Eu-4L) in as shown in Fig. 5

^5D0 →	^7F0	^7F1	^7F2	^7F3	^7F4
Eu-1L	0.64	1	1.56	0.40	4.51
Eu-o-2L	0.39	1	1.89	0.29	2.92
Eu-p-2L	0.54	1	1.15	1.70	1.73
Eu-3L	0.51	1	2.92	0.16	1.70
Eu-4L	0.53	1	3.85	0.12	2.15

---

Table 2 The photophysical and kinetic data of five water soluble europium complexes (Eu-1L, Eu-o-2L, Eu-p-2L, Eu-3L and Eu-4L) as shown in Fig. 5^a

	Eu-1L	Eu-o-2L	Eu-p-2L	Eu-3L	Eu-4L
a ε: the molar extinction coefficient, τH2O: luminescence lifetimes in H2O, τD2O: luminescence lifetimes in deuterium oxide (for europium, the lifetime of the excited state ^5D0 to ^7F2), q: the number of coordinated water molecules, τrad: the radiative (or natural) lifetimes of the ^5D0 state, Q^LEu: the overall quantum yield upon ligand excitation, ηsen: the sensitization efficiencies of the Eu^3+ 5D0 luminescent state.^13,14b Values of the number of coordinated water molecules, q, (±20%), were determined according to methods given in references. ^15,16 The non-integral q value of Eu-o-2L could be because of: (i) the equilibrium distribution; (ii) the hydration impact of a relatively distant, loosely bound water molecule in the second-coordination sphere that exchanges in an all-or-nothing manner.^15,16c The quantum yield was measured using an integrated sphere in aqueous solution.
ε325 nm M^−1 cm^−1	4318.6	11413.5	10633.7	14824.3	20454.6
τH2O ms^−1	0.33	0.72	0.52	0.36	0.57
τD2O ms^−1	1.03	1.48	1.22	1.1	0.58
q^b	1.0	0.56	1.0	1.1	0
τrad^−1	0.429	0.334	0.45	0.33	0.406
Q^LEu^c	1.6	4.5	3.3	13.3	14.5
ηsen	0.073	0.082	0.47	0.98	0.69
pKa	5.1	4.9	4.6	5.4	5.3

---

Low temperature (77 K) emission spectra of water-soluble gadolinium and europium complexes (Ln-1L, Ln-o-2L, Ln-p-2L, Ln-3L and Ln-4L, Ln = Gd or Eu)

The energy difference in the levels between the triplet state of the organic antenna and the excited state of the lanthanide ion is critical for the emission efficiency. There are lots of studies reported in the literature showing that the optimum energy gap should be between 2000 and 5000 cm⁻¹.¹⁷ However, there are limited studies on the increased number of the antenna in the same coordinated skeleton to the lanthanide. Five Gd motif complexes were synthesized and their triplet state of antenna had been determined using their phosphorescence spectra at 77 K. The lowest energy level of Gd³⁺ (⁶P_7/2) is much higher than the energy levels of the ligands, therefore, it is assumed that there is no energy transition between the ligand and metal.¹⁸ The phosphorescence of these Gd complexes was measured under cold conditions (77 K) in a mixture of H₂O–glycerol (Fig. 6). The luminescence lifetimes were recorded to verify the fluorescence and phosphorescence at 77 K (Table 3). The emission lifetimes at a nanosecond scale (S₁ → S₀) were measured at approximately 430 nm for ligands in five complexes, and at a microsecond scale (T₁ → S₀) at approximately 460 to 470 nm for ligands in five complexes. All the triplet states of ligands in five complexes were within 2000–5000 cm⁻¹ compared with the first excited state of ⁵D₀ in europium, and are appropriate for the energy transfer to europium from the ligand for f–f emission. It is impossible to identify the o- and p- isomerism of europium complexes with two antenna by mass spectroscopy (MS) and ultraviolet-visible (UV-Vis) absorption spectroscopy. The ¹H-NMR spectrum is only available for the symmetrical p-isomer (Fig. 3). The hypersensitive and fingerprint emission band of europium in this case can help to identify the o- and p-isomers. In the room temperate emission spectra of Eu-o-2L, the ratio of ⁵D₀ to ⁷F_J, (J = 1, 2 and 4) is 1 : 1.89 : 2.92 but for Eu-p-2L, the ratio is 1 : 1.15 : 1.73 (Fig. 5 and Table 1). The spectral features are only partially resolved at room temperature. In order to gain a clearer distinction between Eu-o-2L and Eu-p-2L, the high resolution low temperature (77 K) spectra of these two complexes were obtained. As shown in Fig. 7, all the electric dipole and the magnetic transition with crystal splitting are well defined. The major difference in the 77 K spectra compared with the room temperature spectra is the smaller relative intensity of the ⁵D₀ → ⁷F₀ transition for both Eu-o-2L and Eu-p-2L, which may indicate changes in the Eu³⁺ site symmetry or less thermal disorder. A greater noise level can be found in Eu-p-2L and the spectral feature of Eu-p-2L is a slight blue shift relative to those in Eu-o-2L (i.e., ⁵D₀ → ⁷F₀ transition for Eu-o-2L is 17 218 cm⁻¹ and in Eu-p-2L is 17 226 cm⁻¹). The most prominent ⁵D₀ → ⁷F₁ transition bands are at 591.2 and 594.9 nm for Eu-o-2L, whereas for Eu-p-2L, there are four emission bands located at 590, 592.4, 597.6 and 600 nm. For Eu-o-2L, two bands are observed for the ⁵D₀ → ⁷F₁ transition region. This could be a consequence of the resonance coupling of the excited state electronic crystal field levels with low energy vibrational excitations of the electronic ground state. However, for Eu-p-2L, it seems that there are two set of ⁵D₀ → ⁷F₁ transition bands and similar observations have been found in ⁵D₀ to ⁷F₄ transitions for Eu-p-2L. This should be induced by the presence in solution of a mixture of two distinct coordination environments associated with the square antiprismic and twisted square antiprismic isomers.¹⁹ From Fig. 7, a clear distinction of isomers Eu-o-2L and Eu-p-2L from their emission spectra could be achieved. In view of the presence of more than 2J + 1 bands for the ⁵D₀ → ⁷F_1,2 transitions for both Eu-o-2L and Eu-p-2L, especially in Eu-p-2L, the spectra illustrate the possibility of square antiprismic and twisted square antiprismic isomers; however, to carry out a crystal field analysis of the energy levels here is far beyond the scope of this study and thus was not attempted in this research.

Fig. 6 Fluorescence spectra (upper, 298 K) and phosphorescence (lower, 77 K) spectra of antenna in five water-soluble Gd complexes. (Motif structures of the five europium complexes are given in Fig. 1.)

Table 3 The emission bands and lifetimes (in brackets, μs) of five Gd complexes (Gd-1L, Gd-o-2L, Gd-p-2L, Gd-3L and Gd-4L) at 298 K and 77 K (λ_ex = 330 nm)

Temp.	Gd-1L	Gd-o-2L	Gd-p-2L	Gd-3L	Gd-4L
298 K	23 365 (0.017)	23 041 (0.021)	23 474 (0.018)	22 989 (0.025)	23 095 (0.026)
77 K	20 619 (0.101)	20 450 (0.110)	21 097 (0.103)	21 231 (0.116)	21 368 (0.125)

---

Fig. 7 Emission spectra at 77 K of Eu-p-2L and Eu-o-2L in aqueous solution (1 μM; λ_ex = 330 nm).

In vitro studies of the five water-soluble europium complexes

With the aim of further developing time-resolved imaging probes using the five complexes, the subcellular uptake profile (amounts and localization) and their cytotoxicity had to be evaluated using inductively coupled plasma – mass spectroscopy (ICP-MS), 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay and in vitro imaging (with a co-staining commercially available marker). From the cellular uptake profile, it can be seen that Eu-4L (+3) and Eu-3L (+2) have a higher uptake amount than any of the other three in HeLa cells (human cervical carcinoma cell line) by checking the europium content inside the cells using ICP-MS (Fig. 8). For Eu-o-2L and Eu-p-2L, the overall charge is the same, in addition to the molecular weight and the antenna, but the uptake profiles are different as shown in the emission spectra in Fig. 9. As a potential time resolved imaging marker, the dark toxicity of each also had to be evaluated. The MTT assays of five water-soluble europium complexes had been done using HeLa cells. The half maximal inhibitory concentration (IC₅₀) of these complexes in HeLa cells were around 0.5 mM for Eu-1L, Eu-p-2L, Eu-4L, and Eu-o-2L, whereas for Eu-3L it is ∼0.25 mM. The uptake amount of Eu-p-2L is greater than Eu-o-2L. Less steric hindrance may improve the cellular uptake rate in vitro. Even the uptake of Eu-p-2L is much greater than Eu-o-2L, and Eu-p-2L has shown a much higher toxicity than Eu-o-2L (Fig. 8c). The IC₅₀ value of Eu-o-2L in HeLa cells is 0.5 mM, but for Eu-p-2L it is ∼0.23 mM. All five europium complexes were soluble in water and were localized in the cytoplasm, except for Eu-4L. The pale red emission can be observed for Eu-1L, Eu-o-2L, and Eu-p-2L in HeLa cells (Fig. 9). Eu-3L and Eu-4L had a strong emission in water (Fig. 9d and e). After a three hour incubation of 1 μM Eu-4L in HeLa cells, the red emission of Eu-4L can be observed in the lysosome of HeLa cells. This subcellular localization was also confirmed using co-staining with the Lyso Tracker Green (Life Technologies; Fig. 9i).

Fig. 8 (a and b) Results of the subcellular uptake evaluation using ICP-MS and (c) the dark cytotoxicity of five water-soluble europium complexes in HeLa cells with using various concentrations.

Fig. 9 In vitro imaging of five water-soluble europium complexes in HeLa cells with a three hour incubation (λ_ex = 405 nm). (a) Eu-1L (24 hours); (b) Eu-o-2L (24 hours); (c) Eu-p-2L (24 hours); (d) Eu-3L (24 hours); (e) Eu-4L (3 hours); (f) Eu-4L (6 hours); (g) Eu-4L (24 hours); (h) Lyso Tracker; (i) Eu-4L + Lyso Tracker.

Conclusions

In conclusion, five water-soluble, cyclen-based europium complexes with various numbers of the same antenna, 2-methyl-4-(2-(4-propoxyphenyl)-ethynyl)pyridine (L), were synthesized successfully. Full characterization of them was achieved using various spectroscopic techniques and from these Eu-3L with an overall positively doubled charge was found to be the best energy-transfer sensitization system, while Eu-4L exhibited the most impressive europium emission quantum yield in aqueous solution (∼16%). After a series of in vitro studies, it is Eu-4L which was non-cytotoxic, cell permeable, and in particular lysosome selective that holds great promise for further development.

Experimental section

Synthesis

General information. All air and water sensitive reactions were carried out under a nitrogen atmosphere with dry solvents under anhydrous conditions unless otherwise noted. All the chemicals were purchased commercially and used without further purification. Anhydrous tetrahydrofuran and toluene were distilled from sodium benzophenone, and dichloromethane (CH₂Cl₂) was distilled from calcium hydride. Yields refer to chromatographically treated quantities, unless otherwise stated. Reactions were monitored by thin-layer chromatography carried out on 0.25 mm silica gel plates (60 F254) that were analyzed by staining with potassium permanganate (KMnO₄; 200 mL of H₂O of 1.5 g of KMnO₄, 10 g potassium carbonate (K₂CO₃) and 1.25 mL of 10% aqueous sodium hydroxide (NaOH)). Fluorescence was triggered upon irradiation at 254 nm or by staining with anisaldehyde (450 mL of 95% ethanol, 25 mL of concentrated sulfuric acid (H₂SO₄). 15 mL of acetic acid, and 25 mL of anisaldehyde). Silica gel 60 (particle size 0.040–0.063 mm) was used for flash chromatography. Infrared (IR) spectra were obtained using a Fourier-transform – infrared (FT-IR) spectrometer, whereas NMR spectra were recorded on either a 400 (¹H: 400 MHz, ¹³C: 100 MHz), or 500 (¹H: 500 MHz, ¹³C: 125 MHz). The following abbreviations are used to explain the multiplicities: s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, b = broad. HRMS were obtained using a matrix-assisted laser desorption/ionisation – time-of-flight (MALDI-TOF) mass spectrometer. Melting points were uncorrected and determined using a micro-melting point meter.

Compound 2. To a stirred solution of 1 (800 mg, 3.0 mmol) and triethylamine (910 mg, 9.0 mmol) in CH₂Cl₂ (30 mL) at room temperature, methanesulfonyl chloride (MsCl; 517 mg, 4.5 mmol) was added dropwise. After stirring at room temperature for 20 minutes, the reaction was quenched by the addition of a saturated sodium bicarbonate solution (30 mL). The organic phase was separated, dried over magnesium sulfate (MgSO₄), filtered, and then concentrated. To a stirred solution of the crude product in acetone (30 mL) at room temperature lithium bromide (LiBr; 6.0 mmol) was added. The resulting mixture was stirred at 60 °C for two hours and then concentrated. The residue was dissolved in water (30 mL) and extracted with CH₂Cl₂ (3 × 30 mL). The combined organic layers were dried over MgSO₄, filtered, and then concentrated. Silica gel flash column chromatography (hexane/ethyl acetate; 15/1 to 10/1 to 8/1) of the residue gave a white solid (710 mg, 72%) as the product. 2: R_f = 0.75 (silica gel, hexane/ethyl acetate = 8 : 1); ¹H-NMR [400 MHz, deuterated chloroform (CDCl₃)] δ 8.55 (d, J = 7 Hz, 1H), 7.48 (s, 1H), 7.50 (d, J = 7 Hz, 2H), 7.28 (d, J = 7 Hz, 1H), 6.91 (d, J = 7 Hz, 1H), 4.67 (s, 2H), 3.96 (t, J = 7 Hz, 2H), 1.88–1.78 (m, 2H), 1.05 (t, J = 7 Hz, 3H); ¹³C-NMR (100 MHz, CDCl₃) δ 160.1, 156.6, 149.3, 133.5, 133.2, 124.6, 124.5, 114.7, 113.7, 95.11, 85.4, 69.6, 46.4, 22.5, 10.5; HRMS [electrospray ionization (ESI)] m/z calcd for C₁₇H₁₇BrNO (M + H)⁺: 330.0494, found: 330.0451.

Compound 1L. To a degassed solution of 2 (80 mg, 0.24 mmol) in MeCN (15 mL) at room temperature, K₂CO₃ (92 mg, 1.21 mmol) and DO₃A(O^tBu)₃ (123 mg, 0.24 mmol) were added. The resulting mixture was stirred at 80 °C for six hours and then quenched by the addition of water (15 mL). The aqueous solution was extracted with CH₂Cl₂ (3 × 15 mL). The combined organic layers were dried over MgSO₄, filtered, and then concentrated. Silica gel flash column chromatography (CH₂Cl₂/methanol (MeOH); 25/1 to 20/1 to 10/1) of the residue gave an off-white oil (124 mg, 68%) as the product. 1L: R_f = 0.42 (silica gel, CH₂Cl₂/MeOH = 10 : 1); ¹H-NMR (500 MHz, CDCl₃) δ 8.20 (d, J = 7 Hz, 1H), 7.44 (d, J = 7 Hz, 2H), 7.24 (s, 1H), 7.17 (d, J = 7 Hz, 1H), 6.86 (d, J = 7 Hz, 2H), 3.93 (t, J = 7 Hz, 2H), 3.57 (s, 2H), 3.21–2.30 (m, 22H), 1.80–1.76 (m, 2H), 1.46 (s, 18H), 1.38 (s, 9H), 1.05 (t, J = 7 Hz, 3H); ¹³C-NMR (125 MHz, CDCl₃) δ 172.7, 172.4, 160.1, 158.8, 148.6, 133.4, 132.8, 124.9, 123.6, 114.7, 113.5, 95.1, 85.2, 81.9, 81.8, 69.6, 58.7, 56.3, 55.5, 53.3, 50.3, 27.9, 22.4, 10.3; HRMS (ESI) m/z calcd for C₄₃H₆₆N₅O₇ (M + H)⁺: 764.4962, found: 764.4961.

Compound o-2L. To a degassed solution of 2 (80 mg, 0.24 mmol) in MeCN (15 mL) at room temperature, K₂CO₃ (166 mg, 1.21 mmol) and DO₂A-o-(O^tBu)₂ (48 mg, 0.12 mmol) were added. The resulting mixture was stirred at 80 °C for six hours and then quenched by the addition of water (15 mL). The aqueous solution was extracted with CH₂Cl₂ (3 × 15 mL). The combined organic layers were dried over MgSO₄, filtered, and concentrated. Silica gel flash column chromatography (CH₂Cl₂/MeOH; 25/1 to 20/1 to 10/1) of the residue gave an off-white oil (67 mg, 62%) as the product. o-2L: R_f = 0.45 (silica gel, CH₂Cl₂/MeOH = 10 : 1); ¹H-NMR (400 MHz, CDCl₃) δ 7.53 (d, J = 7 Hz, 2H), 7.43 (d, J = 7 Hz, 4H), 7.22 (s, 2H), 7.11 (s, 2H), 6.86 (d, J = 7 Hz, 4H), 3.93 (t, J = 7 Hz, 4H), 3.50 (s, 4H), 3.21–2.30 (m, 20H), 1.80–1.74 (m, 4H), 1.42 (s, 18H), 1.02 (t, J = 7 Hz, 6H); ¹³C-NMR (75 MHz, CD₃OD) δ 172.9, 160.7, 160.3, 159.1, 159.0, 148.8, 133.2, 132.8, 125.0, 123.4, 114.5, 113.4, 94.5, 84.8, 81.8, 69.3, 58.5, 55.8, 55.2, 50.3, 49.9, 27.1, 22.1, 9.4; HRMS (ESI) m/z calcd for C₅₄H₇₁N₆O₆ (M + H)⁺: 899.5435, found: 899.5432.

Compound p-2L. To a degassed solution of 2 (80 mg, 0.24 mmol) in MeCN (15 mL) at room temperature, K₂CO₃ (166 mg, 1.21 mmol) and DO₂A-p-(O^tBu)₂ (48 mg, 0.12 mmol) were added. The resulting mixture was stirred at 80 °C for six hours and then quenched by the addition of water (15 mL). The aqueous solution was extracted with CH₂Cl₂ (3 × 15 mL). The combined organic layers were dried over MgSO₄, filtered, and then concentrated. Silica gel flash column chromatography (CH₂Cl₂/MeOH; 25/1 to 20/1 to 10/1) of the residue gave an off-white oil (80 mg, 75%) as the product. p-2L: R_f = 0.45 (silica gel, CH₂Cl₂/MeOH = 10 : 1); ¹H-NMR (400 MHz, CDCl₃) δ 8.34 (d, J = 7 Hz, 2H), 7.48 (d, J = 7 Hz, 4H), 7.29 (s, 2H), 7.21 (d, J = 7 Hz, 2H), 6.90 (d, J = 7 Hz, 4H), 3.96 (t, J = 7 Hz, 4H), 3.52 (s, 4H), 3.21–2.30 (m, 20H), 1.84–1.79 (m, 4H), 1.33 (s, 18H), 1.05 (t, J = 7 Hz, 6H); ¹³C-NMR (100 MHz, CDCl₃) δ 171.9, 160.1, 158.6, 149.1, 133.5, 132.8, 125.0, 123.7, 114.7, 113.5, 95.2, 85.3, 81.8, 69.7, 58.7, 56.9, 50.3, 28.0, 22.5, 10.5; HRMS (ESI) m/z calcd for C₅₄H₇₁N₆O₆ (M + H)⁺: 899.5435, found: 899.5427.

Compound 4L and 6. To a degassed solution of 2 (240 mg, 0.72 mmol) in MeCN (25 mL) at room temperature, K₂CO₃ (500 mg, 3.63 mmol) and cyclen (35 mg, 0.20 mmol) were added. The resulting mixture was stirred at 80 °C for six hours and then quenched by the addition of water (25 mL). The aqueous solution was extracted with CH₂Cl₂ (3 × 25 mL). The combined organic layers were dried over MgSO₄, filtered, and then concentrated. Silica gel flash column chromatography (CH₂Cl₂/MeOH; 25/1 to 20/1 to 10/1) of the residue gave two off-white oils 4L (105 mg, 45%) and 6 (55 mg, 30%) as the major products. 4L: R_f = 0.71 (silica gel, CH₂Cl₂/MeOH = 10 : 1); ¹H-NMR (400 MHz, CDCl₃) δ 7.53 (d, J = 7 Hz, 4H), 7.47 (d, J = 7 Hz, 8H), 7.20 (s, 4H), 7.08 (d, J = 7 Hz, 4H), 6.90 (d, J = 7 Hz, 8H), 3.96 (t, J = 7 Hz, 8H), 3.41–2.30 (m, 24H), 1.86–1.79 (m, 8H), 1.06 (t, J = 7 Hz, 12H); ¹³C-NMR (125 MHz, CDCl₃) δ 160.2, 158.9, 148.8, 133.5, 132.8, 125.4, 123.9, 114.8, 113.5, 95.4, 85.1, 69.7, 59.0, 50.4, 22.4, 10.3; HRMS (ESI) m/z calcd for C₇₆H₈₁N₈O₄ (M + H)⁺: 1169.6383, found: 1169.6373. 6: R_f = 0.50 (silica gel, CH₂Cl₂/MeOH = 10 : 1); ¹H-NMR (400 MHz, CDCl₃) δ 7.53 (d, J = 7 Hz, 3H), 7.48 (d, J = 7 Hz, 6H), 7.19 (s, 3H), 7.08 (d, J = 7 Hz, 3H), 6.90 (d, J = 7 Hz, 6H), 3.96 (t, J = 7 Hz, 6H), 3.41–2.30 (m, 22H), 1.86–1.79 (m, 6H), 1.06 (t, J = 7 Hz, 9H); ¹³C-NMR (125 MHz, CDCl₃) δ 160.3, 158.7, 148.9, 133.5, 132.8, 125.4, 123.9, 114.8, 113.5, 95.4, 85.1, 69.6, 58.7, 50.4, 48.5, 47.6, 22.4, 10.3; HRMS (ESI) m/z calcd for C₅₉H₆₆N₇O₃ (M + H)⁺: 920.5229, found: 920.5212.

Compound 3L. To a degassed solution of 6 (30 mg, 0.03 mmol) in MeCN (10 mL) at room temperature, K₂CO₃ (69 mg, 0.5 mmol) and t-butyl 2-bromoacetate (19 mg, 0.10 mmol) were added. The resulting mixture was stirred at 80 °C for six hours and then quenched by adding water (10 mL). The aqueous solution was extracted with CH₂Cl₂ (3 × 10 mL). The combined organic layers were dried over MgSO₄, filtered, and then concentrated. Silica gel flash column chromatography (CH₂Cl₂/MeOH; 25/1 to 20/1 to 10/1) of the residue gave an-off white oil (28 mg, 89%) as the product. 3L: R_f = 0.62 (silica gel, CH₂Cl₂/MeOH = 10 : 1); ¹H-NMR (500 MHz, CDCl₃) δ 7.53 (d, J = 7 Hz, 3H), 7.48 (d, J = 7 Hz, 6H), 7.26 (d, J = 7 Hz, 3H), 7.05 (d, J = 7 Hz, 3H), 6.90 (d, J = 7 Hz, 6H), 3.96 (t, J = 7 Hz, 6H), 3.46, (s, 6H), 3.41–2.30 (m, 18H), 1.86–1.79 (m, 6H), 1.38 (s, 9H), 1.06 (t, J = 7 Hz, 9H); ¹³C-NMR (125 MHz, CDCl₃) δ 160.2, 158.7, 148.9, 148.7, 133.5, 132.8, 125.4, 123.9, 114.8, 113.5, 95.4, 85.2, 85.1, 81.8, 69.6, 59.2, 58.8, 56.9, 50.3, 28.0, 22.4, 10.4; HRMS (ESI) m/z calcd. for C₆₅H₇₆N₇O₅ (M + H)⁺: 1034.5910, found: 1034.5906.

General procedures for Ln (Eu, Gd) complexes Ln-nL (n = 1, o-2, p-2, 3). Ligand Ln (n = 1, o-2, p-2, 3) (0.020 mmol) were dissolved in TFA (1 mL) and the mixture was stirred at room temperature under argon for 12 hours. After removal of TFA under reduced pressure, the pale yellow residue was dissolved in CH₂Cl₂, and concentrated to ensure that all the TFA had been removed. Deprotection of the ^tBu groups was confirmed using ¹H-NMR of the crude product. The residue was then dissolved in H₂O/MeOH (2 : 1 v/v, 3 mL) and treated with EuCl₃·(H₂O)₆ or gadolinium chloride (GdCl₃; 0.020 mmol). After adjusting the pH to 7.0 with a 1 N aqueous NaOH solution, the mixture was stirred at 70 °C for 24 hours. Concentration and recrystallization of the crude product from MeCN gave an off-white solid as the product. Eu-1L: 75% yield; ¹H-NMR (500 MHz, CD₃OD) δ 28.21, 27.50, 24.27, 13.01, 11.78, 10.09, 7.01–9.27, 4.02, 1.81, 1.09, −1.71, −3.78, −5.12, −7.50, −8.10, −8.72, −10.10, −11.21, −12.02, −12.60, −15.01, −16.28, −17.50, −21.02, −22.47; HRMS (ESI) m/z calcd for C₃₁H₄₁EuN₅O₈ (M + H₂O + H)⁺: 764.2167, found: 764.2221, (M + H)^+: 746.2062, found: 746.2051; t_R (retention time for HPLC) = 10.48 min; IR (neat, 298 K, cm⁻¹) 3223.20, 2877.79, 1585.49, 1411.89, 1321.24, 1253.73, 1176.58, 1082.07, 1006.84, 947.05, 839.03, 798.53; melting point (mp) > 320 °C; [Eu-p-2L]Cl: 73% yield; ¹H-NMR (500 MHz, CD₃OD) δ 24.50, 23.62, 13.21, 14.09, 9.51, 9.27, 9.10, 8.17, 7.62, 7.40, 7.12, 4.37, 4.02, 1.81, 1.02, −2.21, −3.75, −5.91, −7.50, −11.72, −12.52, −16.07, −18.23, −21.10, −25.71; HRMS (ESI) m/z calcd for C₄₆H₅₃ClEuN₆O₆ (M + H₂O + H)⁺: 991.3035, found: 991.2779, (M + H)⁺: 973.2927, found: 973.3058; t_R = 10.93 min; IR (neat, 298 K, cm⁻¹) 3176.67, 3099.61, 2879.72, 1639.49, 1371.38, 1259.52, 1172.72, 1082.07, 999.13, 974.05, 833.25, 567.07; mp > 320 °C; [Eu-o-2L]Cl: 73% yield; ¹H-NMR (500 MHz, CD₃OD) δ 18.60, 14.79, 15.02, 9.57, 9.21, 9.07, 8.02, 7.57, 7.10, 4.02, 1.82, 1.07, −2.25, −3.90, −5.01, −7.50, −10.01, −11.70, −12.54, −14.22, −15.09, −18.29; HRMS (ESI) m/z calcd for C₄₆H₅₃ClEuN₆O₆ (M + H₂O + H)⁺: 991.3035, found: 991.2933, (M + H)⁺: 973.2927, found: 973.2929; t_R = 11.04 min; IR (neat, 298 K, cm⁻¹) 3334.92, 3174,83, 2943.37, 2879.72, 2362.80, 2227.78, 1417.68, 1327.03, 1207.44, 1143.79, 1080.14, 974.05, 796.60, 605.65; mp > 320 °C; [Eu-3L]Cl₂: 80% yield; ¹H-NMR (500 MHz, CD₃OD) δ 31.07, 25.23, 24.09, 13.12, 11.03, 10.18, 9.57, 9.10, 8.59, 7.52, 7.21, 5.13, 4.01, 1.82, 1.07, −3.09, −5.10, −6.92, −7.83, −8.90, −10.11, −11.70, −15.28, −20.13, −21.95; HRMS (ESI) m/z calcd for C₆₂H₆₇Cl₂EuN₇O₅ (M + H₂O + H)⁺: 1218.3899, found: 1218.4003, (M + H)⁺: 1200.3793, found: 1200.3915; t_R = 10.62 min; IR (neat, 298 K, cm⁻¹) 3336.85, 2943.37, 2227.78, 1624.06, 1207.44, 1145.72, 1010.70, 796.60, 601.79, 487.99; mp > 320 °C; Gd-1L: 70% yield; HRMS (ESI) m/z calcd for C₃₁H₄₁GdN₅O₈ (M + H₂O + H)⁺: 769.2196, found: 769.2229, (M + H2O + Na)⁺: 791.2016, found: 791.2047. [Gd-p-2L]Cl: 78% yield; HRMS (ESI) m/z calcd for C₄₆H₅₃ClGdN₆O₆ (M + H)⁺: 978.2956, found: 978.3188. [Gd-o-2L]Cl: 72% yield; HRMS (ESI) m/z calcd for C₄₆H₅₃ClGdN₆O₆ (M + H)⁺: 978.2956, found: 978.3176. [Gd-3L]Cl₂: 81% yield; HRMS (ESI) m/z calcd for C₆₂H₆₇Cl₂EuN₇O₅ (M + H)⁺:1205.3822, found: 1205.4015.

Ln complex Ln-4L (Ln = Eu, Gd). Ligand 4L (0.040 mmol) was dissolved in MeCN and treated with EuCl₃·(H₂O)₆ or GdCl₃ (0.040 mmol). The mixture was stirred at 70 °C for 24 hours. Concentration and recrystallization of the crude product from MeCN gave an off-white solid as the product. [Eu-4L]Cl₃: 88% yield; ¹H-NMR (500 MHz, CD₃OD) δ 15.02, 3.99, 1.82, 1.06, −0.33, −6.49, −7.81, −8.22, −13.31, −14.06, −16.47; HRMS (ESI) m/z calcd for C₇₆H₈₀Cl₂EuN₈O₄ (M − Cl)⁺: 1391.4892, found: 1391.4657; C₇₆H₈₀ClEuN₈O₄ (M − 2Cl)²⁺: 678.2569, found: 678.2497, (M − 3Cl)³⁺: 440.5172, found: 440.5215; t_R = 11.33 min; IR (neat, 298 K, cm⁻¹) 3381.21, 2935.66, 2875.86, 2204.64, 1514.12, 1249.87, 1172.72, 1004.91, 933.55, 831.32, 553.57, 540.07; mp > 320 °C; [Gd-4L]Cl₃: 85% yield; HRMS (ESI) m/z calcd for C₇₆H₈₀Cl₂GdN₈O₄ (M − Cl)⁺: 1396.4921, found: 1396.4465; C₇₆H₈₀ClGdN₈O₄ (M − 2Cl)²⁺: 680.7611, found: 680.7364.

Photophysical properties of the lanthanide complexes

UV-Vis absorption spectra in the spectral range 200 to 1100 nm were recorded using an HP UV-8453 (Agilent) spectrophotometer. Single photon luminescence spectra were recorded using a FLS920 Combined Fluorescence Lifetime and Steady State spectrophotometer (Edinburgh Instruments) that was equipped with a red sensitive single photon counting photomultiplier in a Peltier Cooled Housing. The spectra were corrected for detector response and stray background light phosphorescence. The quantum yields of the compounds were measured using a demountable 142 mm (inner) diameter barium sulfide coated integrating sphere supplied with two access ports. Low temperature (77 K) phosphorescence spectra were obtained by exciting the samples with a xenon lamp. The transparent glassy materials were formed by mixing samples with 2-methyltetrahydrofuran. The samples were placed in a tailor-made quartz tube housed in a liquid nitrogen cryostat (Oxford Instruments).

The sensitization efficiencies of the Eu^{3+ 5}D₀ luminescent state, η_sens, in five europium complexes are related to the overall quantum yield upon ligand excitation, Q^L_Eu, which were determined using the integrating sphere method and by the following equation:^12,14,15

Q^L_Eu = η_sensQ^Eu_Eu = η_sensτ_obs/τ_rad

where Q^Eu_Eu represents the intrinsic quantum yield (i.e., employing direct excitation into Eu³⁺ energy levels); and τ_obs and τ_rad are the observed and radiative (or natural) lifetimes, respectively, of the ⁵D₀ state. The latter can be estimated if the spontaneous emission probability in vacuo of the magnetic dipole allowed ⁵D₀ → ⁷F₁ transition is assumed to be constant (i.e., 14.65 s⁻¹), with constant energy for the transition. This value needs to be corrected by the cube of the refractive index of the water and taking the value 1.333 gives the radiative lifetime of the ⁵D₀ → ⁷F₁ transition as 19.51 ms. The radiative lifetime of the ⁵D₀ state (in ms) may then be estimated by the comparison of integrated areas, I, as follows:^12,14,15
where I_TOT denotes the entire area of the ⁵D₀ → ⁷F_J transitions. Thus, from measurements of ⁵D₀ lifetime (τ_obs), overall quantum efficiency, Q^L_Eu, and integrated spectral areas, it is possible to estimate the sensitization efficiency (η_sens) of Eu^{3+ 5}D₀ luminescence by the coordinated ligands.

pK_a values²⁰

The pH of the solution was carefully controlled to the designated pH according to the usual Tris buffer (pH 2–12) protocol. The pH was monitored using a DT120 glass electrode (Mettler Toledo) equipped with a PT100, class B temperature sensor. This electrode was operated using a MP120 pH meter (Mettler Toledo), which was calibrated using standard buffer solutions of pH = 7.0. The pH dependent luminescent intensity profile of the europium complexes (10 μM) was obtained by monitoring the emission intensity at 615 nm at each point (1.0 pH value per titration interval from 2.0–14.0, see Fig. S8 in the ESI†). The apparent protonation constants were calculated from the data obtained following these luminescent titrations. The equation shown next was fitted to the data, using a nonlinear least squares fitting algorithm:
where
[X] is the concentration of the anion or selected added species in solution, K is the apparent protonation constant, F is the ratio or intensity of the selected peak, F₀ is the ratio or intensity of the selected peak at the beginning of the titration, F₁ is the final ratio or intensity of the selected peak, and [LnL] is the total concentration of the Ln(III) complex in solution.

Tris buffer (pH 2–12) protocol. 1. Base solution: 0.1 M citric acid (21.01 g L⁻¹), 0.1 M potassium phosphate (13.61 g L⁻¹), 0.1 M sodium tetraborate (19.07 g L⁻¹), 0.1 M Tris (12.11 g L⁻¹), 0.1 M potassium chloride (7.46 g L⁻¹) in 50 mL distilled water.

2. Add x mL of 0.4 M hydrochloric acid (HCl) or 0.4 M NaOH to adjust to the desired pH value.

pH (25 °C)	x	Additive
2	34.8	0.4 M HCl
3	19.6
4	10.0
5	0.4	0.4 M NaOH
6	11.4
7	22.4
8	33.2
9	46.2
10	59.0
11	65.6
12	77.2

3. Dilute to 200 mL with mQ water.

HPLC analysis

The reverse-phase HPLC analysis of these five europium complexes was carried out at room temperature by using an Agilent ZORBAX SB-C₁₈ Stable Bond analytical, 4.6 × 150 mm 5 μm column. The mobile phase was 0.1% formic acid in mQ water and 0.1% formic acid in MeCN solvent system, and the flow rate was 1.0 mL min⁻¹. The solvent gradient program is listed in the next table.

Time (min)	0.1% CHOOH in mQ water (%)	0.1% CHOOH in MeCN (%)
0.0	95	5
14.0	50	50
15.0	50	50
20.0	0	100

In vitro studies

Cell culture. Human cervical carcinoma (HeLa) cells were purchased from the American Type Culture Collection (ATCC) (#CCL-2, ATCC, Manassas, VA, USA). The HeLa cells were grown in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum, 1% penicillin and streptomycin at 37 °C, and 5% carbon dioxide (CO₂). To apply europium complexes for fluorescence imaging, HeLa cells were incubated in DMEM containing 1 to 5 L complexes at 37 °C for 24 hours under 5% CO₂, and then washed with phosphate-buffered saline (PBS) to remove excess complexes completely.

In vitro imaging. To test the suitability of the five water-soluble europium complexes as bioprobes, in vitro imaging of HeLa cells incubated with five complexes was performed on a TCS SP5 confocal laser scanning microscope (Leica), equipped with a Ti:Sapphire laser (Libra II, Coherent). All the samples were excited using a 980 nm wavelength laser.

MTT cell viability assay. Cells treated with five water-soluble europium complexes for 24 hours were further incubated with MTT (0.5 mg mL⁻¹) for four hours, to produce formazan during cell metabolism. Then, the formazan was thoroughly dissolved using dimethyl sulfoxide and the absorbances of solutions were measured using an iMark microplate absorbance reader (Bio-Rad; 490 nm). Quadruplicate measurements were performed. Data analysis and plotting were made using the GraphPad Prism 5 software.

ICP-MS studies. To measure the intracellular concentration of the complexes, HeLa cells were plated in each well of a microplate and incubated with the complexes with the concentration from 0.05 to 0.25 mM, which was the same as in the in vitro imaging. After co-incubation, the cell culture medium containing complexes was removed and the exposed cells were further washed three times with 1 mL of PBS to remove the complexes adhering to the outer cell membrane. Then the cells were trypsinized and dispersed into 1 mL of culture medium. The exposed cells were collected by centrifugation at 1000 rpm for 10 minutes and the cell pellet was digested in 100 μL of concentrated nitric acid at 70 °C for three hours. The cellular uptake of complexes was determined using an Agilent 7500 series inductively coupled plasma – mass spectrometer (ICP-MS). All ICP experiments were performed in triplicate and the values obtained were averaged. The concentration of Eu/Gd per cell was calculated by determining the concentration of Eu/Gd in the cell lysate by ICP-MS and then dividing it by the number of cells, which were counted using a haematocytometer.

Acknowledgements

This work was funded by Peking University Shenzhen Graduate School (Key State Laboratory of Chemical Genomics open-project fellowship program), and the grants from The Hong Kong Research Grants Council (HKBU 203013, HKPolyU 5032/11P), Hong Kong Baptist University (FRG 2/12-13/002), and Shenzhen Science and Technology Innovation Committee (KQTD201103).

Notes and references

1. P. A. Tanner, Chem. Soc. Rev., 2013, 42, 5090–5101.
2. G. Mancino, A. J. Ferguson, A. Beeby, N. J. Long and T. S. Jones, J. Am. Chem. Soc., 2005, 127, 524–525.
3. S. V. Eliseeva and J. C. Bünzli, Chem. Soc. Rev., 2010, 39, 189–227.
4. R. Carr, N. H. Evan and D. Parker, Chem. Soc. Rev., 2012, 41, 7273–7686.
5. W. S. Lo, W. M. Kwok, G. L. Law, C. T. Yeung, C. T. L. Chan, H. L. Yeung, H. K. Kong, C. H. Chen, M. Murphy, K. L. Wong and W. T. Wong, Inorg. Chem., 2011, 50, 5309–5311.
6. T. Zhang, X. Zhu, W. M. Kwok, C. T. L. Chan, H. L. Tam, W. K. Wong and K. L. Wong, J. Am. Chem. Soc., 2011, 50, 20120–20122.
7. S. J. Butler, L. Lamarque, R. Pal and D. Parker, Chem. Sci., 2014, 5, 1750–1756.
8. E. G. Moore, A. P. S. Samuel and K. N. Raymond, Acc. Chem. Res., 2009, 42, 542–552.
9. D. Parker, J. W. Walton, L. Lamarque and J. M. Zwier, Eur. J. Inorg. Chem., 2010, 25, 3961–3966.
10. Y. W. Yip, H. Wen, W. T. Wong, P. A. Tanner and K. L. Wong, Inorg. Chem., 2012, 51, 7013–7015.
11. C. Li and W. T. Wong, Tetrahedron, 2004, 60, 5595–5601.
12. P. Caravan, J. J. Ellison, T. J. McMurry and R. B. Lauffer, Chem. Rev., 1999, 99, 2293–2352.
13. M. C. Heffern, L. M. Matosziuk and T. J. Meade, Chem. Rev., 2014, 114, 4496–4539.
14. J. G. G. Bunzli, Chem. Rev., 2010, 110, 2729–2755.
15. S. Aime, M. Botta, M. Fasano, M. P. M. Marques, C. F. G. C. Geraldes, D. Pubanz and A. E. Merbach, Inorg. Chem., 1997, 36, 2059–2068.
16. A. Beeby, I. M. Clarkson, R. S. Dickins, S. Faulkner, D. Parker, M. Woods, L. Royles, A. S. De Sousa, J. A. G. William and M. Woods, J. Chem. Soc., Perkin Trans. 2, 1999, 493–504.
17. M. H. V. Werts, R. T. F. Jukes and J. W. Verhoeven, Phys. Chem. Chem. Phys., 2002, 4, 1542–1548.
18. G. L. Law, K. L. Wong, K. K. Lau, H. L. Yam, K. W. Cheah and W. T. Wong, Eur. J. Inorg. Chem., 2007, 34, 5419–5425.
19. H. Wen, C. K. Duan, G. Jia, P. A. Tanner and M. G. Brik, J. Appl. Phys., 2011, 110, 033536–033538.
20. D. G. Smith, B. K. McMahon, R. Pal and D. Parker, Chem. Commun., 2012, 48, 8520–8852.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra16536e

---