Ambient temperature synthesis of citrate stabilized and biofunctionalized, fluorescent calcium fluoride nanocrystals for targeted labeling of cancer cells

Abstract

Targeted biological contrast agents are emerging as promising candidates in the field of cancer theragnostics. Herein, we report an ambient temperature synthesis of a nanosized, antibody functionalized lanthanide doped CaF₂ biolabel and demonstrate in vitro its potential for cancer cell targeting efficacy and specificity. Monodispersed citrate stabilized lanthanide (Eu³⁺) doped CaF₂ nanoparticles with size ∼25 nm, exhibiting strong fluorescent emission at 612 nm, were prepared using an aqueous wet chemical route at room temperature. Biofunctionalization of the fluorescent nanoparticles using an anti-EGFR antibody through EDC–NHS coupling chemistry enabled targeting of EGFR over-expressing cells. The nanobioconjugates showed preferential binding to EGFR^+ve oral epithelial carcinoma cells (KB) and human epidermoid carcinoma cells (A431) with no accumulation onto EGFR^−ve non-cancerous NIH 3T3 cells. The fluorescence was maintained after the bioconjugation as well as after attachment to the cancer cells, demonstrating their potential as targeted biolabels. Cytotoxicity evaluation with several cancerous (A431, KB) and non-cancerous (NIH 3T3, L929) cell lines revealed no toxicity at concentrations up to 1 mM. Thus, the fluorescence characteristics and biocompatibility, coupled with the molecular receptor targeting capability, suggest the potential use of CaF₂ in the field of bioimaging.

---

1. Introduction

Advances in bioimaging enabled by nanotechnology are driven primarily by the emergence of new tools that assist manipulations at an atomic scale, as well as new materials with tailorable size, surface chemistry, improved solubility, and multifunctionality.¹ The use of nanosized fluorescent biolabels for cancer diagnosis is by now well recognized as having diagnostic value both ex vivo^2,3 and in vivo.^4,5 Such nanomaterials should necessarily possess good photostability and high quantum yield, as well as non-toxicity, to serve as diagnostic agents ex vivo or in vivo. Conventionally, organic fluorophores, fluorescent silica beads, or quantum dots have been used as fluorescent bio-labeling materials.^6,7 The first two suffer from problems associated with photobleaching and small Stokes shift, whereas quantum dots pose toxicity concerns in vivo.^8,9 This has stimulated interest in developing biocompatible and highly efficient fluorescent nanomaterials for use as biological labels. Our group has developed a host of inorganic nanoparticles based on ZnS,^10,11 Y₂O₃,¹² the biomineral hydroxyapatite,¹³ and organic protein containing gold molecular clusters^2,14,15 as novel contrast agents, and demonstrated success in the targeted labeling of cancer cells.

Inorganic luminescent materials based on rare earth ions have gained considerable attention and have become promising alternative biolabels owing to their superior optical properties, such as high quantum yield, narrow bandwidth, long-lived emission, large Stokes shift and ligand-dependent luminescence sensitization, all of which meet the stringent requirements of a biological label.^12,13,16 The fluorescence emission from lanthanide ions has been made use of in various optoelectronic applications such as displays, lasers, etc.^17,18 However, non-radiative relaxation in the luminescence of Ln³⁺ ions necessitates an appropriate lattice with a lower phonon energy to host the lanthanide ions.¹⁹ Among the various inorganic matrices, fluorides, owing to their high transparency arising from low phonon energy, serve as perfect host materials and have been widely established as a laser material for several years.^20–22 Additionally, the optically isotropic structure, high stability, non-hygroscopic behavior, non-toxicity and biocompatibility of CaF₂ among all the other fluorides make it an ideal host matrix for biological fluorescent contrast material.²³

In this study, we have investigated a CaF₂ matrix doped with lanthanide ions as a suitable biolabel. Various researchers have developed this promising system as a good host matrix for lanthanide ions, through several approaches such as hydrothermal, reverse micelle, chemical co-precipitation, etc.;^24–30 however, two new objectives are explored here: (a) an ambient temperature processing route for the biolabel, which is expected to provide a water dispersible product and also better retain the luminescence intensity of lanthanide ions, and (b) antibody functionalization of the label with the ability to specifically target cancer cells and retain luminescence when functionalized, as well as when in contact with the target cells. Several investigators have attempted to render functional groups on doped CaF₂. Bioconjugation of chitosan coated CaF₂:Eu nanoparticles synthesized via the microemulsion method with bovine serum albumin has been previously attempted.³¹ Oleic acid modified CaF₂:Eu nanocrystals well dispersed in chloroform were synthesized by Song et al.³² Likewise, using a single step hydrothermal method, oleate capped lanthanide doped upconverting colloidal CaF₂ nanoparticles have been prepared.³³ Citrate, a calcium complexing agent, is reported to tailor the size and control the morphology of apatite crystals upon binding to its surface.^34,35 This strategy of size and morphology control by citrate, inspired by nature, is exploited for the synthesis of nanomaterials such as lanthanide fluorides³⁶ and apatite nanocrystals.^37–40 Citrate ions are reported to have the affinity to inhibit CaF₂ crystal growth⁴¹ and recently citrate stabilized nanoparticles of Tm³⁺,Yb³⁺ and Er³⁺,Yb³⁺ doped CaF₂ have been synthesized by hydrothermal processing at 190 °C.⁴² However, the combination of water dispersibility and conjugation with biomolecules has not so far been demonstrated. This is challenging, owing to the extreme reaction conditions such as high temperature and pressure used for the synthesis, and hence the need for an alternative strategy.

We report for the first time the synthesis of citrate stabilized lanthanide doped CaF₂ nanoparticles (Cit-CaF₂:Ln) via a simple wet chemical route at room temperature, and demonstrate its application as a biolabel. Cit-CaF₂:Eu nanoparticles synthesized via aqueous chemistry in the present work were bioconjugated with an epidermal growth factor receptor (EGFR) specific antibody and the target specific imaging has been demonstrated using EGFR^+ve cells. The high fluorescence emission, biocompatibility and receptor specific targeting of bioconjugated CaF₂ nanoparticles suggest its potential use as an inorganic contrast agent for molecular imaging of cancer.

2. Results and discussion

2.1 Synthesis of bare and citrate stabilized CaF₂:Ln nanoparticles

In the synthesis of undoped calcium fluoride particles, although the precursors, viz. CaCl₂·2H₂O and NH₄F, are readily water soluble, the formed product (CaF₂), being insoluble in water, precipitates out from aqueous solution. The appearance of turbidity during synthesis indicated the formation of CaF₂.⁴⁴ However, it was found very difficult to control the particle size of CaF₂ during growth, owing to the high concentration of Ca²⁺ and F⁻ ions, leading to enhanced reaction rates. Hence, to limit the availability of ions, in the present synthesis route the precursors were added drop-wise into the reaction mixture.

Trivalent lanthanide ions (Tb³⁺, Eu³⁺) when doped into the calcium fluoride matrix incorporate into the crystal lattice by substituting divalent calcium ions, resulting in the formation of CaF₂:Ln particles. The excess positive charge resulting from this doping is generally compensated for by the incorporation of fluoride ions at interstitial sites, resulting in structural defects in the lattice.^24,45 A typical reaction scheme for the preparation of lanthanide doped CaF₂ is represented below, with the schematic of the synthesis procedure depicted in Fig. 1a.

CaCl₂·2H₂O + NH₄F + LnCl₃(Ln:Tb, Eu) ⇒ Ca_1−xF₂:Ln_x + NH₄Cl + 2H₂O

Fig. 1 Schematic reaction mechanism depicting (a) formation of lanthanide doped CaF₂ particles in the absence of citrate; (b) pre-complexation of citrate ions by Ca²⁺ followed by formation of Cit-CaF₂:Eu nanoparticles in the presence of F⁻ ions at pH 5.5; and (c) antibody conjugation of citrate stabilized CaF₂:Eu nanoparticles.

In this reaction, only the product Ca_1−xF₂:Ln_x is the insoluble component, whereas all other reactants and products are water soluble, and hence can be easily removed by centrifugation. Optimization of the sequence of addition of reactants in the above reaction aided efficient incorporation of lanthanide ions into the CaF₂ matrix. The different reaction sequences used in preparing highly fluorescent CaF₂ particles were: (i) mixing dopant ions with Ca²⁺ precursors to which the F⁻ precursor (NH₄F) was added drop-wise; (ii) mixing the dopant ions with NH₄F and adding them drop-wise to the Ca²⁺ precursor and (iii) adding dopant ions separately while Ca²⁺ and F⁻ ions react to nucleate CaF₂.

It was observed that case (i) and case (iii) resulted in the formation of CaF₂ with efficiently incorporated lanthanide ions, as suggested by its high fluorescence emission. However, in case (ii), the dopant ions, upon reacting with NH₄F, were found to form white precipitates of lanthanide fluorides (EuF₃ or TbF₃). Hence, for all further experiments, dopant ions were mixed with Ca²⁺ precursors and the F⁻ precursor (NH₄F) was added drop-wise to this mixture [case (i)].

Additionally, the pH of the solution was found to play a critical role in the formation of CaF₂, with a range of 5–6 facilitating a maximum yield. Upon addition of lanthanide chlorides, the pH of the solution decreased below the limit that yielded CaF₂ particles. Hence, several trials were made, with variations in pH facilitated by the addition of liquor ammonia. Addition of NaOH to the reaction mixture to maintain the pH can result in the formation of NaF, which decreases the yield of CaF₂, and hence was not used in our experiments. Another important observation was that the synthesized lanthanide doped CaF₂ nanoparticles were found to agglomerate with time, since no stabilizers were used initially. Use of citrates was additionally able to provide a stable colloid that could be further functionalized with the required biomolecules.

In our experiments, we used trisodium citrate (Na₃C₆H₅O₇) as the stabilizer, by drop-wise addition separately with NH₄F. Trisodium citrate did not hinder CaF₂ formation, as long as it was not mixed with the initial precursors of CaCl₂ and NH₄F. Further, it was capable of not only providing citrate ions for electrostatic stabilization, but also maintained the required pH for the formation of CaF₂, resulting in stable CaF₂:Ln nanoparticles. It was noted that a pH of 5.5 yielded nanoparticles with maximum stability.

We hypothesize the mechanism of citrate ion stabilization as follows. Ca²⁺ ions first form complexes with citrate anions, as has been proposed in the case of oleic acid stabilization.³² Subsequently, citrate complexed metal ions react with F⁻ to form crystalline nuclei, whose growth is terminated in the nanosize range by the association of citrate ions on the growing particle interfaces.⁴⁶ This proposed mechanism of citrate stabilization of doped CaF₂ nanoparticles is illustrated in Fig. 1b. Citrate stabilization of CaF₂ nanoparticles also provides the necessary carboxyl functional groups, which readily permit further conjugation with the antibody, using EDC–NHS coupling chemistry.⁴³

Over-expression of EGFR has been observed in human cancers such as head and neck, ovarian, colon, bladder and breast cancer.^47,48 Hence, in this study, Epidermal Growth Factor Receptor (EGFR), a transmembrane glycoprotein and a member of the erbB receptor tyrosine kinase family, was chosen as the primary target. Activation of carboxyl groups of citrate molecules by EDC is known to occur efficiently at acidic pH (4.5–6). This reaction results in the formation of an amine reactive O-acylisourea intermediate, which is susceptible to fast hydrolysis in water. Hence this intermediate is stabilized to form amine reactive NHS by the addition of S-NHS. The amine reactive Cit-CaF₂:Eu was coupled with the primary amine of the anti-EGFR antibody, resulting in covalent attachment of the antibody to the nanoparticle surface, as schematically represented in Fig. 1c.

2.2 Crystal structure, crystallinity, size, shape and composition

Fig. 2a and 2b depict the SEM images of bare and citrate stabilized CaF₂:Eu nanoparticles. The images clearly reveal that citrate stabilization has resulted in well stabilized and monodispersed nanoparticles of size ∼25 nm, in contrast to the agglomerated bare CaF₂. The DLS data of citrate stabilized nanoparticles shown in the inset of Fig. 2b confirm this observation, showing a particle size of 20–25 nm. The lack of colloidal stability of bare CaF₂:Eu hindered its DLS analysis. The TEM images of Cit-CaF₂:Eu nanoparticles shown in Fig. 2c indicate well dispersed spherical nanoparticles with a particle size of ∼25 nm. The HRTEM images shown in Fig. 2d clearly show the ordered crystalline lattice planes of CaF₂, with the measured lattice fringe spacing of 0.193 nm agreeing well with the lattice spacing of (220) of cubic CaF₂.⁴⁹ Additionally, few localized defect atomic clusters were seen as dark spots in the HRTEM image. The selected area electron diffraction (SAED) pattern of CaF₂ nanocrystallites shown in the inset of Fig. 2d corresponds to the crystalline lattice planes (111), (220), (311) and (400) of CaF₂. The virtual circular pattern observed in the diffraction pattern may be due to the presence of several crystallites of CaF₂, as in polycrystalline materials.

Fig. 2 Representative SEM images of (a) bare CaF₂:Eu particles; (b) Cit-CaF₂:Eu nanoparticles (inset – particle size distribution by DLS); (c) TEM image and (d) HRTEM image (inset: SAED pattern) of Cit-CaF₂:Eu nanoparticles.

Citrate stabilized doped CaF₂ nanoparticles possessed a moderately high zeta potential value of −22 ± 2.5 mV, with the negative charge due to the carboxyl groups of citrate on the nanoparticle surface. The electrostatic repulsion rendered by the carboxyl groups accounts for its stability in suspension. It was observed that Cit-CaF₂:Ln nanoparticles prepared by altering the concentration of citrate ions, thereby changing the solution pH, showed remarkably different colloidal stability. Fig. 3a depicts the changes in zeta potential for samples prepared with varying pH via the addition of different amounts of trisodium citrate. As evident from Fig. 3b, a maximum colloidal stability was obtained at a pH of 5.5.

Fig. 3 (a) Relative frequency shift and (b) zeta potential of Cit-CaF₂:Eu nanoparticles synthesized at different pH.

Crystallinity as well as phase purity of the synthesized doped and undoped CaF₂ nanoparticles were analyzed using XRD. Fig. 4a depicts the XRD patterns of CaF₂, CaF₂:Eu and europium citrate, in comparison with the theoretical JCPDS data of CaF₂ (line spectrum). Fig. 4b depicts the XRD of Cit-CaF₂:Eu nanoparticles with varying dopant concentrations of Eu. The diffraction patterns of CaF₂, CaF₂:Eu and Cit-CaF₂:Eu can be indexed to the cubic phase of the fluorite-type CaF₂ structure (space group: Fm3m). These spectra exhibit the prominent reflections corresponding to (111), (220), (311), (400) and (311), well in accordance with the JCPDS data base (35-0816) of the CaF₂ crystal.²⁴ The absence of any characteristic peak of europium citrate in citrate stabilized CaF₂:Eu (Fig. 4a) rules out the inclusion of europium citrate in the lattice, thereby confirming the phase purity of the sample. It can be noted from Fig. 4b that with increasing doping concentration of Eu³⁺ ions, diffraction peaks showed a marked decrease in intensity, with increased full width at half maximum (FWHM), indicating decreased crystallinity with increased disorder. This can be attributed to the structural defects that arise in CaF₂ due to the substitution of Ca²⁺ ions with Eu³⁺, coupled with the charge compensatory incorporation of the F⁻ ion.^18,24 The presence of such lattice defects was evident from the HRTEM image (Fig. 2d). The results collectively suggest that the synthesized Cit-CaF₂:Eu nanoparticles are isostructural with the CaF₂ crystal, with some disorder and decreased crystallinity.

Fig. 4 X-ray diffraction spectrum of (a) CaF₂ – 35-0816, CaF₂, CaF₂:Eu, and europium citrate (C₆H₅EuO₇) (b) Cit-CaF₂:Eu nanoparticles with varying dopant concentration.

The elemental composition of the synthesized nanoparticles was confirmed through EDX analysis. Fig. 5a shows a representative EDX spectrum of 15% Eu doped Cit-CaF₂:Eu, displaying elemental peaks relating to Ca, F and Eu, confirming detectable levels of dopant ions within the CaF₂ matrix. A linear increase in the atomic weight percentage of Eu³⁺ with respect to Ca²⁺ was noted for increasing Eu content used for synthesis, as depicted in tabular form in Fig. 5b. The values correspond to the experimentally used dopant concentrations, and it was observed that for higher Eu doping concentration, the estimated Eu content was lower, perhaps also due to the precipitation of europium citrate. However, the washed final product did not reveal any peaks of europium citrate, as evident from its XRD pattern (Fig. 4a and 4b).

Fig. 5 (a) Energy dispersive X-ray analysis spectrum of 15% Eu doped Cit-CaF₂:Eu NP, (b) atomic weight percentage of europium with respect to calcium, as determined by EDAX analysis of Cit-CaF₂:Eu nanoparticles synthesized with varying dopant concentration, (c) C, H, and N composition by CHN analysis of bare, Cit-CaF₂:Eu nanoparticles and trisodium citrate, (d) footprint of citrate ligand and number of citrate ligands per CaF₂ nanoparticle for its attachment through 1, 2 and 3 carboxyl groups.

CHN analysis was performed to assess the elemental C, H and N composition of the samples and thereby evaluate the amount of citrate present on the particles. Fig. 5c depicts the tabulated results of elemental analysis of dried Cit-CaF₂:Eu (15%), bare CaF₂:Eu (15%) and trisodium citrate (Na₃C₆H₅O₇). These results are consistent with a mass content of 19.5% of citrate and 80.5% of Eu doped CaF₂ in Cit-CaF₂:Eu nanoparticles.

In the present study, we have used CHN analysis to assess the mode of attachment of citrate molecules to the surface of nanoparticles, i.e., whether the attachment is via one, two or three of its carboxyl groups. The number of citrate ligands that can attach to spherical CaF₂ nanoparticles of size ∼25 nm was deduced by assuming a ligand footprint (corresponding to the area occupied by an anchoring group on the surface) of 0.514, 0.195, 0.311 nm² for citrate molecule attachment through one, two and three of its carboxyl groups, respectively.⁵⁰Fig. 5d, in tabulated form, depicts the number of citrate ligands per nanoparticle via one, two and three carboxyl attachments. It was observed that the number of citrate ligands on each nanoparticle (calculated from C % of CHN analysis) coincided with that calculated using a 0.195 nm² footprint of the citrate molecule, implying that citrate molecules attached to CaF₂ nanoparticles via two carboxyl groups. This renders one of the three carboxyl groups of citrate free for further bioconjugation, which is confirmed in the ensuing sections.

2.3 FTIR analysis

Fig. 6 shows the FTIR spectra of trisodium citrate, bare CaF₂:Eu, Cit-CaF₂:Eu and antibody conjugated nanoparticles. The broad absorption band occurring around 3440 cm⁻¹ is characteristic of H–O–H bending of the water molecules, revealing the presence of hydroxyl groups in all the samples. For CaF₂:Eu, the prominent absorbance noted at 1638 cm⁻¹ can be assigned to the O–H bending vibrations of the residual water that is physically adsorbed onto the surface of the particles.^31,51

Fig. 6 FTIR spectra of CaF₂:Eu, trisodium citrate, Cit-CaF₂:Eu and antibody conjugated Cit-CaF₂:Eu nanoparticles, showing successful citrate functionalization and antibody conjugation on the nanoparticles.

The absorption band at 2360 cm⁻¹ is from the KBr pellets used for recording the FTIR spectrum. The spectra of Cit-CaF₂:Eu have absorption bands that correspond to those of trisodium citrate, revealing the presence of the organic functional groups, i.e. citrate anions. Two strong absorption peaks at 1585 and 1394 cm⁻¹ of trisodium citrate, characteristic of the RCO₂ symmetric and asymmetric stretches, appear shifted to 1614 and 1434 cm⁻¹, respectively, implying the conjugation of carboxyl groups to the nanoparticle surface (Ca). The very broad absorption band in the region 2500–3500 cm⁻¹ for trisodium citrate and Cit-CaF₂:Eu, corresponding to the hydrogen bonded OH stretch of the carboxyl group, obscures other peaks in this region. The typical peaks at 620 cm⁻¹ prove the existence of alkyl groups of citrate.

The conjugation of the antibody to the nanoparticles is confirmed by the emergence of two new bands in the spectrum at 1654 and 1534 cm⁻¹, which are characteristic of amide I and amide II bonds. The absence of carboxyl peaks and the emergence of these two new amide peaks are typical of antibody conjugation.⁵² Thus, FTIR analysis clearly confirms the presence of citrate ions on the surface of CaF₂:Eu, providing it with adequate stabilization and surface functionalization with carboxyl groups, and amide bond formation in the case of antibody conjugated nanoparticles.

2.4 Fluorescence characteristics of doped CaF₂

2.4.1 Terbium doped calcium fluoride (CaF₂:Tb). To assess the applicability of lanthanide doped CaF₂ nanoparticles as contrast agents for fluorescence imaging, the synthesized samples were subjected to a detailed spectrofluorimetric analysis. Unlike intrinsic QDs such as CdSe, CdTe, CdS, ZnSe or ZnTe, which show a size dependent luminescence, fluorescence emission from lanthanide doped nanomaterials, such as that of CaF₂, arises from the intra orbital electronic transitions within dopant ions (Fig. 8d). As mentioned earlier, two lanthanides, viz. Eu³⁺ and Tb³⁺, were doped into the calcium fluoride lattice, yielding varying fluorescence emission wavelengths. Undoped CaF₂ nanoparticles did not show any fluorescence emission in the same wavelength range.

For terbium doped CaF₂ nanoparticles, a characteristic green emission with varying intensity was observed when tested initially using a handheld UV lamp at 254 and 365 nm. From spectrofluorimetric analysis, it was deduced that Tb doped CaF₂ nanoparticles exhibited characteristic emission peaks at 488, 542, 583 and 620 nm, which correspond to the ⁵D₄ → ⁷F₆, ⁵D₄ → ⁷F₅, ⁵D₄ → ⁷F₄ and ⁵D₄ → ⁷F₃ transitions, respectively, of Tb³⁺ ions in the host lattice.²⁶ Amongst them, the most intense emission peak lies in the green region at λ = 542 nm, which corresponds to the ⁵D₄ → ⁷F₅ transition.^26,27 The excitation peaks for this green emission at 542 nm are characteristic of ⁷F₆ → ⁵D₃, ⁷F₆ → ⁵G_J and ⁷F₆ → ⁵L₆ transitions of Tb³⁺ ions for 257, 350 and 368 nm, respectively.²⁶

The influence of doping content on these spectral excitation and emission peaks was also assessed. Fig. 7a and 7b depict a comparison of the excitation and emission spectra of CaF₂:Tb for different dopant percentages of Tb (2, 5 and 10%) with respect to atomic weight percentage of calcium. Remarkable differences were noted when the dopant content in the crystalline lattice was varied. The predominant excitation peak for a dopant concentration of 2 and 5% Tb was 257 nm. On the other hand, upon increasing the dopant content to 10%, the excitation spectrum displayed two major peaks at 368 and 378 nm, instead of the prominent peak at 257 nm for low doping concentrations. Such an observation has not been reported earlier on Tb doped CaF₂ nanoparticles. However, for 5% Tb doped nanoparticles prepared in anhydrous methanol and re-dispersed in water, Wang et al.²⁶ have observed a green emission at 542 nm for an excitation at 375 nm. On the contrary, nanoparticles co-doped with Ce and Tb (5, 10 mol%) prepared by the polyol-mediated synthesis route, displayed emission at 534 nm for an excitation at 254 nm.²⁷ This implies that electronic transitions are sensitive to dopant ion positions, as well as to their nearest neighbor interactions in the crystal lattice, aside from the crystal symmetry, all of which are influenced by the synthesis route and the dopants used.

Fig. 7 Fluorescence (a) excitation and (b) emission spectra of CaF₂:Tb. Inset shows the picture of CaF₂:Tb powder when viewed under a fluorescence microscope.

For the highest doping content of 10% Tb used in this study, a marked enhancement in fluorescence emission at 542 nm at a red-shifted excitation wavelength of 378 nm was noticed, which would enable bioimaging applications with better spectral resolution. This implies that an increase in doping concentration enabled better incorporation of Tb³⁺ into the CaF₂ lattice. Thus, from these studies, the optimal dopant concentration of Tb into the CaF₂ lattice was found to be 10%, which resulted in high greenish luminescence emission.

2.4.2 Europium doped calcium fluoride (CaF₂:Eu). Eu doped CaF₂ nanoparticles showed a characteristic red emission when excited by a handheld UV lamp at 365 nm, which was further confirmed spectrofluorimetrically. The excitation (λ_ex = 394 nm) and emission (λ_em = 590 nm) spectra of CaF₂:Eu nanoparticles are given in Fig. 8a and 8b, respectively. The emission spectrum exhibits the characteristic transition ⁵D₀ → ⁷F_J (J = 1, 2, 3, 4 for 590, 612, 649, 691 nm, respectively) of Eu³⁺ ions (Fig. 8d). As evident, the spectrum is dominated by the peaks at 590 nm due to ⁵D₀ → ⁷F₁ and at 612 nm due to the ⁵D₀ → ⁷F₂ transitions in Eu³⁺. It is well established that the luminescence properties of Eu³⁺ are significantly influenced by the symmetry of the host lattice.^32,53 Typically, the symmetry of the crystal site in which the Eu³⁺ ions are located decides the ratio of the ⁵D₀ → ⁷F₁ and ⁵D₀ → ⁷F₂ transitions. The ⁵D₀ → ⁷F₁ emission is allowed by magnetic dipole considerations (insensitive to local symmetry) and ⁵D₀ → ⁷F₂ by electric dipole interactions (very sensitive to local symmetry).⁵³ Thus, in a site with inversion symmetry, ⁵D₀ → ⁷F₁ would dominate and ⁵D₀ → ⁷F₂ would dominate in sites without inversion symmetry. Such variations in the electronic transitions of Eu doped CaF₂ nanoparticles and the resulting changes in their emission peaks were earlier reported by Song et al. for particles prepared with oleic acid as the stabilizing agent.³² In our experiments, as evident from the spectrum of CaF₂:Eu nanoparticles, the transition ⁵D₀ → ⁷F₁ corresponding to the 590 nm emission peak dominates, which clearly implies the significant localization of Eu³⁺ ions in sites with inversion symmetry. Fig. 8b depicts this variation in fluorescence emission for an excitation of 394 nm, when the dopant ion concentration was varied from 2 to 20%. As expected, the fluorescence emission intensity was remarkably enhanced with increased doping, clearly indicating a higher incorporation of Eu³⁺ into the nanocrystal lattice. However, fluorescence was found to be saturated at 15 wt% of Eu³⁺. Similar effects in fluorescence intensity variations due to Eu doping have been reported for CaF₂ nanocrystals prepared using the chemical co-precipitation method in ethanol solution.²⁴ An explanation for the saturation effects observed for high doping concentrations is as follows. In heavily doped samples, the increased interaction of dopant ions with the nearest neighbour host atoms in the crystal lattice may lead to the precipitation of atomic defect clusters within the crystal structure, as revealed from the dark contrast in the HRTEM of doped CaF₂ nanocrystals (Fig. 2d). Hence, the optimum dopant percentage necessary to yield maximum emission was set at 15 at%, and all further studies were carried out with this dopant percentage of Eu³⁺.

Fig. 8 Spectrofluorimetric analysis of CaF₂:Eu at varying dopant percentages (a) excitation and (b) emission spectra. Inset of (b) shows a graphical representation of emission (590 nm λ_ex = 394 nm) intensity vs. dopant percentage of CaF₂:Eu with saturation at 15%, and a picture of CaF₂:Eu powder when viewed under a fluorescence microscope. (c) Emission spectra (λ_ex = 394 nm) of Cit-CaF₂:Eu with varying dopant percentages. (d) Intra orbital electronic transitions of Eu³⁺ ions. (e) Comparative photoluminescence spectra of bare, citrate stabilized and antibody conjugated CaF₂:Eu. (f) Emission spectra (λ_ex = 394 nm) of CaF₂:Eu synthesized with liquor ammonia, maintaining a pH of 5.5.

As discussed earlier, citrate stabilization of CaF₂ yielded nanoparticles with high colloidal stability and monodispersed size. Cit-CaF₂:Eu containing Eu at different dopant concentrations (2, 5, 10, and 15 at%) also showed a similar variation in its emission spectra (λ_ex = 394 nm) (Fig. 8c). Similar to the effects observed earlier, increasing concentration of Eu³⁺ resulted in an enhancement in the emission, indicating higher doping of Eu³⁺ into the nanocrystal lattice. A doping concentration of 15 at% yielded the maximum fluorescence emission as in bare nanoparticles.

To understand if citrate capping or antibody conjugation in any way altered the pattern of fluorescence emission from Eu doped CaF₂, a comparison of the spectra of bare CaF₂:Eu and Cit-CaF₂:Eu and Ab-Cit-CaF₂:Eu was made, as shown in Fig. 8e. Surprisingly, although the excitation spectra of bare and citrate stabilized particles were similar, their emission spectra (λ_ex = 394 nm) appeared distinctly different with regard to the ratios of intensity of the prominent emission transitions, viz.⁵D₀ → ⁷F₁ (590 nm) and ⁵D₀ → ⁷F₂ (612 nm). It is established that the ⁵D₀ → ⁷F₂ electric dipole transition is very sensitive to relatively small changes in the chemical surroundings of Eu³⁺ ions.^32,54,55 The intensity of this transition is markedly affected by the degree of symmetry in the environments around Eu³⁺ ions. During the preparation of Cit-CaF₂:Eu, addition of trisodium citrate alters the pH of the solution. Hence, we hypothesize that this change in pH to 5.5 might have caused variations in the chemical environment of the Eu³⁺ ions. This in turn might have resulted in the predominance of the ⁵D₀ → ⁷F₂ transition peak (612 nm) in Cit-CaF₂:Eu nanoparticles, as compared to ⁵D₀ → ⁷F₁ (590 nm) in bare CaF₂:Eu. This influence of pH in regulating the fluorescence emission was further confirmed by the use of liquor ammonia to maintain the pH at 5.5 during the synthesis of CaF₂:Eu nanoparticles. The result of this study shown in Fig. 8f depicts the enhancement in 612 nm emission peak for an excitation at 394 nm, in comparison to bare nanoparticles, which always display a predominant 590 nm peak (Fig. 8b). This helps us to conclude that pH has a significant role in influencing the fluorescence emission from CaF₂ nanoparticles.

Citrate stabilized CaF₂:Eu nanoparticles obtained with citric acid as the precursor for citrate ions and liquor ammonia used to maintain the pH at 5.5 also displayed similar emission/excitation spectra as for nanoparticles prepared using trisodium citrate (data not shown). This implies that the source of citrate does not alter the fluorescence characteristics of CaF₂:Eu.

Additionally, as evident from Fig. 8e, citrate stabilized CaF₂:Eu nanoparticles, when further conjugated with the antibody, also displayed brilliant fluorescence, although with a slight reduction in intensity. However, the fluorescence intensity measured for Ab-Cit-CaF₂:Eu was high enough to enable fluorescent labeling of cells.

2.5 Biological studies

2.5.1 Cell morphology and cell viability assay. The primary focus of this investigation being the development of a biocompatible contrast agent with desirable optical properties for cellular imaging, the toxicity of the material was tested on both normal and cancer cells. Fig. 9 (top panel) depicts the microscopic images of representative KB cells treated with the highest concentration (1 mM) of CaF₂:Eu nanoparticles for 24 h with reference to cells alone (negative control). The bottom panel in Fig. 9 gives the graphical representation of cell viability MTT tests carried out on NIH 3T3, L929, KB and A431 cells for nanoparticle treatment at varying concentrations (0.01, 0.05, 0.1, 0.5 and 1 mM) and 24 hours of incubation.

Fig. 9 Top panel: representative microscopic images of KB cells incubated with bare CaF₂:Eu, Cit-CaF₂:Eu, Ab conjugated Cit-CaF₂:Eu nanoparticles and negative control for 24 h (magnification: 40×); bottom panel: graphical representation of cell viability tests on L929, NIH 3T3, KB and A431 cells treated with CaF₂:Eu nanoparticles (bare, citrate capped and antibody conjugated) at varying concentrations for 24 h.

It is clearly evident from Fig. 9 (top panel) that the cells treated with CaF₂ nanoparticles at the highest concentration for 24 hours remained morphologically intact and proliferated well in the culture medium in well plates. The nanomaterial treated cells did not reveal any characteristic features of apoptosis, proving its cytofriendly nature, which was further confirmed by the cytotoxicity studies. In comparison to the negative control, MTT analysis (Fig. 9 bottom panel) showed that more than 80% of cells were viable and metabolically active when tested on four different cell lines using bare, citrate stabilized and antibody conjugated CaF₂:Eu nanoparticles at varying doses. These studies confirm that the fluorescent nanobioprobes are cytocompatible, with no adverse effects on their cellular functions, which is an essential requisite for contrast agents. Similar observations on the non-toxic effects of CaF₂ upconversion nanophosphors prepared hydrothermally have been reported on HeLa cells for 18 hours of incubation.⁴² Fluorides are conventionally toxic in vivo.^56,57 However, fluoride toxicity is not a concern in the case of CaF₂, due to its extreme insolubility, unlike other fluorides such as NaF, BaF₂, etc.⁵⁸ This makes CaF₂ an important material for various human applications such as in dentistry, food supplements, etc. Hence, lanthanide doped fluorescent CaF₂ nanoparticles, if proposed for in vivo applications, would offer ideal contrast agents for bioimaging, owing to their in vivo biocompatibility.

2.5.2 Targeted cell imaging studies using fluorescence microscopy. Having developed a non-toxic, fluorescent nanomaterial, we now demonstrate the molecular receptor targeted delivery and fluorescence imaging of EGFR over-expressing cancer cells with high specificity, using the antibody conjugated CaF₂:Eu nanoparticles. In this experiment, both EGFR over-expressing cells such as the human epidermoid carcinoma A431 cells^59,60 and oral epithelial carcinoma KB cells, as well as the EGFR^−ve NIH3T3 cells, have been used.⁶¹Fig. 10 shows the fluorescence microscopic images of EGFR^−ve NIH3T3 cells treated with bare and antibody conjugated nanoparticles at a concentration of 0.5 mM for 2 and 4 hours, respectively. It is evident that even after 4 hours of incubation with bare and antibody conjugated red emitting nanoparticles (Fig. 10 (i, ii, iii)), there was no specific interaction with the NIH3T3 cells. The particles appear to be randomly distributed, with no explicit attachment to EGFR^−ve cells. Likewise, bare fluorescent nanoparticles incubated for 4 hours with EGFR over-expressing A431 and KB cells also displayed only a scattered distribution, with no definite interaction with their cell membranes (Fig. 10 (iv) and 11 (i and ii)). On the contrary, EGFR^+ve cells treated with antibody conjugated Cit-CaF₂:Eu nanoparticles for 2 and 4 hours at 0.5 mM concentration showed a drastic difference in their attachment, with very specific accumulation on their cell membrane (Fig. 10 (v, vi) and 11 (iii, iv, v, vi)). It can be seen that a large number of fluorescent nanoparticles were found specifically attached to the cell membranes after as little as 2 hours of incubation on EGFR over-expressing cells (Fig. 10 (v) and 11 (iii, v)).

Fig. 10 Fluorescent microscopic images showing the interaction of bare CaF₂:Eu and antibody conjugated Cit-CaF₂:Eu nanoparticles (Ab-Cit-CaF₂:Eu) with EGFR^−ve NIH 3T3 (top panel) and EGFR^+ve KB cells (bottom panel) for different incubation periods.

Fig. 11 Fluorescent microscopic images showing the interaction of bare CaF₂:Eu and antibody conjugated Cit-CaF₂:Eu nanoparticles (Ab-Cit-CaF₂:Eu) with EGFR over-expressing, human epidermoid carcinoma A431 cells for 2 and 4 hours of incubation, observed at different magnifications.

Similar to the microscopic images shown in Fig. 9, the cells (KB, A431), presenting an enhanced accumulation of nanoparticles on the cell membrane, retained their characteristic morphology and showed good spreading with time, implying that the material does not induce any toxicity to cells. This study clearly indicates the enhanced affinity and specificity of antibody conjugated nanoparticles to EGFR receptors on cancer cells, with minimal attachment to the normal cells, which are essential attributes to any contrast agents suggested for targeted imaging applications.

3. Conclusion

In summary, a biological fluorescent label based on lanthanide doped calcium fluoride on the nanosize scale that can target the EGFR over-expressing cells has been developed in this study. A simple aqueous wet chemical route of synthesis using citrate ions as stabilizers resulted in highly fluorescent lanthanide (europium) doped CaF₂ nanoparticles, with a characteristic emission at 612 nm. These nanoparticles of size ∼25 nm and high colloidal stability were conjugated with anti-EGFR antibodies, demonstrating specific targeting of EGFR over-expressing cells and no interaction with normal cells. Cell viability studies on four different cell lines demonstrated its non-toxic nature, even up to a relatively high concentration of 1 mM. Thus a detailed investigation on the synthesis, and characteristic physical, chemical and biological features of citrate stabilized Eu³⁺ doped calcium fluoride nanoparticles were carried out. Terbium doped CaF₂, also developed using similar aqueous chemistry routes, and exhibiting a strong green fluorescence at 542 nm, can be likewise bioconjugated for imaging applications. Hence this study presents the application of biocompatible fluorescent calcium fluoride nanoparticles for targeted imaging of cancer cells. This work opens up the possibilities for extending this strategy to up-conversion nanophosphors in vivo or for multiplexed cancer diagnosis on clinical tissues.

4. Materials and methods

4.1 Materials

The starting materials used for the synthesis of doped calcium fluoride nanoparticles include: calcium chloride dihydrate (CaCl₂·2H₂O) [≥97%, Merck], ammonium fluoride (NH₄F) [≥95%, Merck, India], terbium chloride hexahydrate (TbCl₃·6H₂O) [99.999%, Sigma Aldrich, USA], europium chloride hexahydrate (EuCl₃·6H₂O) (99.9%, Sigma Aldrich, USA), trisodium citrate dihydrate (99%, Merck, India), citric acid anhydrous (99%, SD Fine Chem Ltd, India) and liquor ammonia (25% NH₃, Fischer Scientific, India). EDC [1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride] (Thermo Scientific, USA) and S-NHS [sulfo-N-hydroxysulfosuccinimide] (Thermo Scientific, USA) were used for bioconjugation of the nanoparticles with Anti EGFR mouse monoclonal antibody (1 mg ml⁻¹ in PBS, Abcam, UK). All the chemicals were used as received without any further purification.

4.2 Synthesis of lanthanide (Ln³⁺-Tb³⁺, Eu³⁺) doped calcium fluoride (CaF₂) particles [CaF₂:Ln]

An aqueous wet chemical route at room temperature was adopted for the synthesis of bare fluorescent CaF₂:Ln particles. In a typical experiment, 5 ml of 0.1 M CaCl₂·2H₂O was mixed thoroughly with a 0.1 M concentration of the dopant, viz. lanthanide chloride hexahydrate (TbCl₃·6H₂O or EuCl₃·6H₂O), at room temperature. To this mixture, 5 ml of 0.2 M NH₄F as the precursor for fluorine was added drop-wise and stirred continuously for 2 h, yielding CaF₂:Ln particles. To optimize the reaction conditions resulting in the efficient incorporation of lanthanide ions into the CaF₂ matrix, the lanthanide precursor was added in three different ways during the synthesis, viz., mixed with the initial precursors CaCl₂ or NH₄F, or by drop-wise addition separately with NH₄F. Highly fluorescent CaF₂ particles were obtained under favourable conditions.

To obtain different doping concentrations with respect to atomic weight percentage of calcium, varying amounts of dopant ions (0–15 at% for Tb and 0–20 at% for Eu) were employed. The resultant mixture was then centrifuged (24 780g) and washed thrice with MilliQ water to remove the reaction by-products and any un-reacted precursors. To visualize the formation of lanthanide doped CaF₂, a handheld UV lamp with excitation at 365 nm was used, exhibiting a pale red or green emission for Eu and Tb dopants, respectively. The final product was dispersed in MilliQ water and stored. The resultant colloidal suspension was not stable on storage. The precipitate was also dried at 60 °C for 24 h for further characterization.

To evaluate the influence of pH on the fluorescence emission as well as the yield, CaF₂:Ln was also synthesized in a similar manner as above by varying the solution pH using liquor ammonia.

4.3 Synthesis of water dispersible citrate stabilized Eu doped calcium fluoride nanoparticles (Cit-CaF₂:Eu)

To prepare well stabilized lanthanide doped nanoparticles of CaF₂, citrate stabilization was employed. Trisodium citrate dihydrate was used as a stabilizer and pH modifier to prepare stabilized doped CaF₂ nanoparticles (Cit-CaF₂:Eu) via the same wet chemical route as described earlier. To assess which method would yield the most stable colloidal solution of doped CaF₂, citrate was added in three different ways during the synthesis, viz., mixed with the initial precursors CaCl₂ or NH₄F, or separately with NH₄F. Under optimal conditions, a white colloid of well stabilized doped CaF₂ nanoparticles was formed, which exhibited brilliant fluorescence under UV excitation. This colloid was then centrifuged and dispersed in water for further characterization. The same protocol can be extended to prepare citrate stabilized Tb doped CaF₂ nanoparticles as well.

Additionally, a few trials with citric acid instead of trisodium citrate as the precursor for citrate ions were also performed, with liquor ammonia maintaining the desired pH of 5.5.

4.4 Bio-conjugation of Cit-CaF₂:Eu with anti-EGFR antibody (Ab) using EDC–NHS coupling chemistry

To obtain fluorescent nanoparticles for bioimaging applications, an appropriate ligand is required to target specific receptors on cancer cells. Here, bio-conjugation of doped Cit-CaF₂ nanoparticles was carried out with the anti-EGFR antibody using the conventional EDC–NHS conjugation chemistry to obtain a covalent attachment of the anti-EGFR antibody to the surface of nanoparticles.⁴³ Citrate stabilized nanoparticles bearing free carboxyl groups on their surface were activated by the addition of the zero-length cross-linker EDC and further linked to the amine groups present on the antibody. Briefly, 1 mg Cit-CaF₂:Eu was washed and re-dispersed with 10 mM MES buffer and incubated with 5 μM EDC and 7.5 μM S-NHS for 20 min, in the dark, at room temperature. They were then incubated with 50 μl anti-EGFR antibody (1 mg ml⁻¹ in PBS, mouse monoclonal antibody with reactivity towards mouse and human) for 2 h, resulting in the formation of stable antibody conjugated CaF₂ nanoparticles (Ab-Cit-CaF₂:Eu). Antibody conjugated nanoparticles retained the bright fluorescence emission and were well stabilized.

4.5 Particle characterization

Particle size and size distribution of the nanoparticles were measured by the dynamic light scattering technique using NICOMP^TM380 Particle size analyzer (PSS, USA). The surface charge and thereby stability of the particles were assessed from the zeta potential measurements using a NICOMP^TM380 ZLS zeta potential analyzer. Nanoparticles prepared in aqueous dispersions were used for the measurements in triplicate. Size and surface morphology of bare and citrate stabilized nanoparticles were evaluated by scanning electron microscopy (SEM) (JSM-6490LA, JEOL, Japan). A diluted sample drop-cast on an aluminium stub was sputter-coated with gold before imaging. Transmission electron microscopy (TEM) (JEM-200CX, JEOL, Japan) was employed for deciphering the shape, size and lattice structure of the nanoparticles. An aqueous dispersion of the nanoparticles at appropriate dilution was drop-cast onto a carbon-coated copper grid and air-dried at room temperature before TEM imaging.

The phase purity and crystallinity of bulk CaF₂, CaF₂:Eu and Cit-CaF₂:Eu nanoparticles and europium citrate (C₆H₅EuO₇) were studied from X-ray diffraction (XRD) measurements using a Philips X'PERT PRO powder diffractometer fitted with a Cu-Kα source (λ = 1.54056 Å). The spectrum was recorded in the range 10°–80° in steps of 0.02°. Phase identification was done using the standard JCPDS database.

Composition of the nanoparticles, particularly the presence of dopants, was analyzed by Energy Dispersive X-Ray (EDX) spectroscopy (JSM-6490LA, JEOL, Japan). To quantify the amount of citrate molecules, as well as to decipher their mode of attachment, CHN analysis was performed on the samples using a CHN Analyzer (Perkin Elmer 2400, USA) to determine the percentage of carbon (C), hydrogen (H) and nitrogen (N) present in each material.

Spectrofluorimetric measurements of doped CaF₂ colloids were recorded using a HORIBA, JOBIN YVON spectrofluorimeter (Model Fluoro Max 4). Fourier transform infrared (FT-IR) spectra of the samples supported on KBr pellets were recorded in the frequency range 4000–400 cm⁻¹ at a resolution of 4 cm⁻¹ using a Spectrum RX1 FT-IR Spectrometer, Perkin Elmer, USA.

4.6 Cell culture

Oral epithelial carcinoma cells (KB) and human epidermoid carcinoma cells (A431), which both over-express EGFR, and normal mouse fibroblast cells (L929) and normal mouse embryonic fibroblast cells (NIH 3T3) that do not have EGFR expression, used in the present study, were purchased from the National Center for Cell Science, Pune, India. L929 and KB cell lines were cultured in a T25 flask in Minimum Essential Medium (MEM) (Sigma-Aldrich, USA) and A431 cell line in Dulbecco's modified Eagles Medium (DMEM) (Sigma Aldrich, USA) supplemented with 10% fetal bovine serum (FBS, Invitrogen, CA, USA), 50 IU ml⁻¹ penicillin and 50 μg ml⁻¹ streptomycin. The cells were maintained in the medium at 37 °C in a humidified environment of 5% CO₂.

4.7 In vitro cell viability assay

To determine the cell viability upon interaction with calcium fluoride nanoparticles, L929, NIH3T3, A431 and KB cells in the log phase were trypsinized and seeded at a density of 7 × 10³ cells per well into 96 well tissue culture plates. MTT assay, a colorimetric assay, based on the ability of viable cells to reduce a soluble yellow tetrazolium salt [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)] to purple formazan crystals, was used in the present study. Cells were incubated for 24 h in a humidified 5% CO₂ incubator at 37 °C. After 24 h, the culture medium was replaced with media containing five different concentrations (0.01, 0.05, 0.1, 0.5 and 1 mM) of the nanoparticles (bare CaF₂:Eu, Cit-CaF₂:Eu and antibody conjugated Cit-CaF₂:Eu). Cells with Triton X100 (1%) were used as the positive control, while cells maintained in fresh 10% FBS media served as the negative control, all in triplicate. After 24 h of incubation, the wells were washed using PBS and 10 μl of the MTT stock solution supplemented with 100 μl of the medium was added into each well. The cells were incubated further for 4 h, after which 110 μl of the solubilization buffer was added into each well and incubated for 1 h. Optical density of the solution was measured using a microplate reader (BioTek Power Wave XS, USA) at a wavelength of 570 nm and the percentage cell viability was assessed. Cell viability was calculated using the equation:
where Intensity_sample and Intensity_control represent the OD values of cells incubated the with sample and culture medium, respectively.

4.8 Cell morphology analysis

The effect of nanoparticle treatment on the cell morphology of normal and cancer cells (L929, NIH 3T3, A431 and KB) was visualized using an inverted microscope (Leica Microsystems, USA). L929, NIH 3T3, and KB cells seeded in MEM medium and A431 cells in DMEM medium were treated with the highest concentration of 1 mM for 24 h and morphological analysis was made by recording their digital live images microscopically.

4.9 Targeted cell imaging using fluorescence microscopy

EGFR over-expressing, human epidermoid carcinoma (A431) and oral epithelial carcinoma (KB) cell lines were used to test the cancer specific targeting of antibody conjugated Cit-CaF₂:Eu nanoparticles. The EGFR^−ve normal mouse embryonic fibroblast NIH 3T3 cell line was kept as the control. The cultured cells were trypsinized and seeded at a density of 5000 cells onto acid etched coverslips placed in 24 well tissue culture plates. After 24 h, adherent cells were washed carefully with PBS and replaced with fresh media containing bare and antibody conjugated nanoparticles of varying concentrations (100 μM, 500 μM, 1 mM). Cells were incubated for two different time durations (2 and 4 h) at 37 °C and further washed twice with PBS, fixed with 2% paraformaldehyde, washed and mounted with DPX mounting medium. Imaging was carried out using an Olympus Fluorescent Microscope (Model BX51) equipped with a CCD camera (Model DP71), using oil immersion objectives. Fluorescence was detected using band pass excitation and emission filters (370–410 nm, >485 nm) and a 400 nm dichroic mirror.

Acknowledgements

Authors gratefully acknowledge Department of Science & Technology, Government of India for financial assistance under the projects “Theragnostics, Regenerative Medicine and Stem Cell Research using Nanomaterials” and “PG Training Programs (M. Tech Nanomedical Science) SR/NM/PG-02/08. The authors also thank Mr Sajin P. Ravi for SEM analysis and Amrita Vishwa Vidyapeetham for providing all infrastructural facilities.

References

1. F. Wang, W. B. Tan, Y. Zhang, X. Fan and M. Wang, Nanotechnology, 2006, 17, R1–R13.
2. V. M. Peterson, C. M. Castro, H. Lee and R. Weissleder, ACS Nano, 2012, 6, 3506–3513.
3. A. P. Retnakumari, P. L. Hanumanthu, G. L. Malarvizhi, R. Prabhu, N. Sidharthan, M. V. Thampi, D. Menon, U. Mony, K. Menon, P. Keechilat, S. Nair and M. Koyakutty, Mol. Pharmaceutics, 2012 DOI:10.1021/mp300172e.
4. X. He, K. Wang and Z. Cheng, Wiley Interdiscip. Rev.: Nanomed. Nanobiotechnol., 2010, 2, 349–366.
5. S. Santra and A. Malhotra, Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol., 2011, 3, 501–510.
6. P. Sharma, S. Brown, G. Walter, S. Santra and B. Moudgil, Adv. Colloid Interface Sci., 2006, 123–126(SPEC. ISS.), 471–485.
7. W. J. Parak, T. Pellegrino and C. Plank, Nanotechnology, 2005, 16, R9–R25.
8. D. Beer and J. Weber, Opt. Commun., 1972, 5, 307–309.
9. C. Kirchner, T. Liedl, S. Kudera, T. Pellegrino, A. M. Javier, H. E. Gaub, S. Stolzle, N. Fertig and W. J. Parak, Nano Lett., 2005, 5, 331–338.
10. J. Aswathy, S. Sajith, M. Koyakutty, S. Nair and D. Menon, Carbohydr. Polym., 2011, 85, 37–43.
11. J. Aswathy, S. Jahnavi, R. Krishna, M. Koyakutty, S. Nair and D. Menon, J. Nanosci. Nanotechnol., 2011, 11, 7611–7620.
12. S. Setua, D. Menon, A. Asok, S. Nair and M. Koyakutty, Biomaterials, 2010, 31, 714–729.
13. A. Ashokan, D. Menon, S. Nair and M. Koyakutty, Biomaterials, 2010, 1–11.
14. A. Retnakumari, S. Setua, D. Menon, P. Ravindran, H. Muhammed, T. Pradeep, S. Nair and M. Koyakutty, Nanotechnology, 2010, 21, 055103 (12pp).
15. A. Retnakumari, J. Jayasimhan, P. Chandran, D. Menon, S. Nair, U. Mony and M. Koyakutty, Nanotechnology, 2011, 22, 285102 (11pp).
16. C. Bouzigues, T. Gacoin and A. Alexandrou, ACS Nano, 2011, 5, 8488–8505.
17. S. Cotton, Lanthanides and Actinides, Macmillan Education, Basingstoke, 1991.
18. C. Feldmann, T. Jüstel, C. R. Ronda and P. J. Schmidt, Adv. Funct. Mater., 2003, 13, 511–516.
19. Yu. E. Perlin and A. A. Kaminskii, Phys. Status Solidi B, 1985, 132, 11–40.
20. D. C. Tran, G. H. Sigel Jr. and B. Bendow, J. Lightwave Technol., 1984, 2, 566–586.
21. F. Auzel, Mater. Sci. Forum, 1991, 67–68, 489–502.
22. C. A. Klein, J. Appl. Phys., 2006, 100, 083101 (1–13).
23. P. Maushake, Optik & Photonik, 2008, 2, 46–47.
24. F. Wang, X. Fan, D. Pi and M. Wang, Solid State Commun., 2005, 133, 775–779.
25. L. G. Jacobsohn, C. J. Kucera, T. L. James, K. B. Sprinkle, J. R. DiMaio, B. Kokuoz, B. Y. Kukouz, T. A. DeVol and J. Ballato, Materials, 2010, 3, 2053–2068.
26. J. Wang, W. Miao, Y. Li, H. Yao and Z. Li, Mater. Lett., 2009, 63, 1794–1796.
27. C. Feldmann, M. Roming and K. Trampert, Small, 2006, 2, 1248–1250.
28. S. Rungrodnimitchai, Jpn. J. Appl. Phys., 2011, 50, 01BJ17 (1–4).
29. C. Pandurangappa, B. N. Lakshminarasappa and B. M. Nagabhushana, J. Alloys Compd., 2010, 489, 592–595.
30. Z. Quan, D. Yang, P. Yang, X. Zhang, H. Lian, X. Liu and J. Lin, Inorg. Chem., 2008, 47, 9509–9517.
31. Y. Tian, J. Yu, X. Qi, X. Wu, R. Hua and S. Fan, J. Mater. Sci.: Mater. Electron., 2009, 20, 439–444.
32. L. Song, J. Gao and R. Song, J. Lumin., 2010, 130, 1179–1182.
33. M. Pedroni, F. Piccinelli, T. Passuello, M. Giarola, G. Mariotto, S. Polizzi, M. Bettinellia and A. Speghini, Nanoscale, 2011, 3, 1456–1460.
34. Y.-Y. Hu, A. Rawal and K. Schmidt-Rohr, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 22425–22429.
35. B. Xie and G. H. Nancollas, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 22369–22370.
36. A. M. Cross, P. S. May, F. C. J. M. van Veggel and M. T. Berry, J. Phys. Chem. C, 2010, 114, 14740–14747.
37. A. López-Macipe, J. Gómez-Morales and R. Rodríguez-Clemente, Adv. Mater., 1998, 10, 49–53.
38. A. Wang, H. Yin, D. Liu, H. Wu, M. Ren, T. Jiang, X. Cheng and Y. Xu, Mater. Lett., 2007, 61, 2084–2088.
39. J. M. Delgado-López, M. Iafisco, I. Rodríguez, A. Tampieri, M. Prat and J. Gómez-Morales, Acta Biomater., 2012, 8, 3491–3499.
40. Y. Wu, Y. Tseng and J. C. C. Chan, Cryst. Growth Des., 2010, 10, 4240–4242.
41. Z. Amjad, Langmuir, 1993, 9, 597–400.
42. N. N. Dong, M. Pedroni, F. Piccinelli, G. Conti, A. Sbarbati, J. E. R. Hernández, L. M. Maestro, M. C. I. Cruz, F. S. Rodriguez, A. Juarranz, F. Chen, F. Vetrone, J. A. Capobianco, J. G. Sole, M. Bettinelli, D. Jaque and A. Speghini, ACS Nano, 2011, 5, 8665–8671.
43. G. T. Hermanson, Bioconjugate Techniques, Academic Press Inc., 2nd edn, 2008, ch. 3, 598–599.
44. M. Markovic, S. Takagi, L. C. Chow and S. Frukhtbeyn, J. Res. Natl. Inst. Stand. Technol., 2009, 114, 293–301.
45. A. S. Shcheulin, T. S. Semenova, L. F. Koryakina, M. A. Petrova, A. E. Angervaks and A. I. Ryskin, Opt. Spectrosc., 2011, 110, 617–623.
46. J. Turkevich, P. C. Stevenson and J. Hillier, Discuss. Faraday Soc., 1951, 11, 55–75.
47. R. S. Herbst, Int. J. Radiat. Oncol., Biol., Phys., 2004, 59, S21–S26.
48. D. S. Salomon, R. Brandt, F. Ciardiello and N. Normanno, Crit. Rev. Oncol. Hematol., 1995, 19, 183–232.
49. S. Hou, Y. Zou, X. Liu, X. Yu, B. Liu, X. Sun and Y. Xing, Cryst. Eng. Commun., 2011, 13, 835–840.
50. T. Huang, P. D. Nallathamby and X. N. Xu, J. Am. Chem. Soc., 2008, 130, 17095–17105 Supporting information.
51. M. Falk, J. Raman Spectrosc., 1990, 21, 563–567.
52. J. Bandekar, Vib. Spectrosc., 1993, 5, 143–173.
53. A. F. Kirby and F. S. Richardson, J. Phys. Chem., 1983, 87, 2544–2556.
54. X. Fan, M. Wang and G. Xiong, J. Mater. Sci. Lett., 1993, 12, 1552–1554.
55. P. P. Gawryszewska, M. Pietraszkiewicz, J. P. Riehl and J. Legendziewicz, J. Alloys Compd., 2000, 300–301, 283–288.
56. B. D. Gessner, M. Beller, J. P. Middaugh and G. M. Whitford, N. Engl. J. Med., 1994, 330, 95–99.
57. National Research Council, Fluoride in Drinking Water: A Scientific Review of EPA's Standards, National Academies Press, Washington, DC, 2006.
58. Scientific Opinion of the Panel on Food Additives and Nutrient Sources added to Food (F. Aguilar, U. R. Charrondiere, B. Dusemand, P. Galtier, J. Gilbert, D. M. Gott, S. Grilli, R. Guertler, G. E. N. Kass, J. Koenig, C. Lambré, J.-C. Larsen, J.-C. Leblanc, A. Mortensen, D. Parent-Massin, I. Pratt, I. Rietjens, I. Stankovic, P. Tobback, T. Verguieva and R. Woutersen), Calcium fluoride as a source of fluoride added for nutritional purposes to food supplements, EFSA J., 2008, 882, 1–15.
59. C. M. Stoscheck and G. Carpenter, J. Cell. Physiol., 1984, 120, 296–302.
60. H. Miyagi and I. N. Maruyama, Open Spectrosc. J., 2010, 4, 28–31.
61. M. C. Willingham, H. T. Haigler, D. J. P. Fitzgerald, M. G. Gallo, A. V. Rutherford and I. H. Pastan, Exp. Cell Res., 1983, 146, 163–175.

---