CdTe magnetic--fluorescent composite

Home

Search

Collections

Journals

About

Contact us

My IOPscience

Magnetite/CdTe magnetic–fluorescent composite nanosystem for magnetic separation and bio-imaging

This article has been downloaded from IOPscience. Please scroll down to see the full text article. 2011 Nanotechnology 22 225101 (http://iopscience.iop.org/0957-4484/22/22/225101) View the table of contents for this issue, or go to the journal homepage for more

Download details: IP Address: 202.156.10.239 The article was downloaded on 26/12/2011 at 14:44

Please note that terms and conditions apply.

IOP PUBLISHING

NANOTECHNOLOGY

Nanotechnology 22 (2011) 225101 (12pp)

doi:10.1088/0957-4484/22/22/225101

Magnetite/CdTe magnetic–fluorescent composite nanosystem for magnetic separation and bio-imaging Anup Kale1 , Sonia Kale2 , Prasad Yadav1 , Haribhau Gholap1 , Renu Pasricha3 , J P Jog1 , Benoit Lefez4 , B´eatrice Hannoyer4 , Padma Shastry2,5 and Satishchandra Ogale1,5 1

National Chemical Laboratory, Council of Scientific and Industrial Research, Pune 411008, India 2 National Centre for Cell Science, Ganeshkhind, Pune 411007, India 3 National Physical Laboratory, Council of Scientific and Industrial Research, New Delhi 110012, India 4 Universit´e de Rouen, GPM UMR 6634 CNRS-BP 12, 76801 Etienne du Rouvray Cedex, France E-mail: [email protected] and [email protected]

Received 18 December 2010, in final form 13 March 2011 Published 4 April 2011 Online at stacks.iop.org/Nano/22/225101 Abstract A new synthesis protocol is described to obtain a CdTe decorated magnetite bifunctional nanosystem via dodecylamine (DDA) as cross linker. High resolution transmission electron microscopy (HRTEM), energy-dispersive x-ray spectroscopy (EDAX), vibrating sample magnetometry (VSM), Fourier transform infrared spectroscopy (FTIR), diffuse reflectance spectroscopy (DRS) and fluorescence microscopy are used to characterize the constitution, size, composition and physical properties of these superparamagnetic–fluorescent nanoparticles. These CdTe decorated magnetite nanoparticles were then functionalized with anti-epidermal growth factor receptor (EGFR) antibody to specifically target cells expressing this receptor. The EGFR is a transmembrane glycoprotein and is expressed on tumor cells from different tissue origins including human leukemic cell line Molt-4 cells. The magnetite–CdTe composite nanosystem is shown to perform excellently for specific selection, magnetic separation and fluorescent detection of EGFR positive Molt-4 cells from a mixed population. Flow cytometry and confocal laser scanning microscopy results show that this composite nanosystem has great potential in antibody functionalized magnetic separation and imaging of cells using cell surface receptor antibody. S Online supplementary data available from stacks.iop.org/Nano/22/225101/mmedia (Some figures in this article are in colour only in the electronic version)

nanoparticles are interesting because their movements can be externally manipulated and they exhibit relaxations under driving electromagnetic fields which lead to hyperthermia. Moreover, magnetic nanoparticles hold promise in bioseparation, magnetic resonance imaging (MRI) as contrast enhancement agents, and drug delivery [5, 6]. Quantum dots (QDs) are interesting because their luminescence properties can be tuned through the control of size, shape and

1. Introduction Magnetic and fluorescent nanoparticles (NPs) hold some of the most exciting application prospects of nanotechnology. They are also of great scientific interest to the fields of chemistry, biology, physics and engineering [1–4]. Magnetic 5 Authors to whom any correspondence should be addressed.

0957-4484/11/225101+12$33.00

1

© 2011 IOP Publishing Ltd Printed in the UK & the USA

Nanotechnology 22 (2011) 225101

A Kale et al

functionalization, and they are superior to dyes due to their photostability [7–9]. Combining magnetic and fluorescent properties in a single nanoparticle clearly opens up newer and broader avenues for their applications in the fields of biological imaging, cell tracing, magnetic bio-separation, bioand chemo-sensing, targeted drug delivery, MRI etc [10, 11]. The magnetic–fluorescent bifunctional nanoparticles have been prepared by various groups by different approaches. The magnetic materials used include Fe3 O4 , γ -Fe2 O3 , FePt, Co, Mn, and Ni, whereas fluorescent molecules include quantum dots like CdSe, CdSe/ZnS, CdTe, ZnS, CdO, ZnO etc and organic dyes like FITC, Cy5, rhodamine etc [10, 11]. One of the commonly used approaches is to encapsulate a fluorescent moiety in a shell of natural or synthetic polymers [12] or silica matrix [13]. Most of the composites synthesized by this route are micron sized. PEGylated hybrid magnetic and fluorescence micelles with 25 nm diameter have also been synthesized with controlled ratio of active groups [14]. Synthesis of amorphous silica coated magnetic γ -Fe2 O3 nanoparticles incorporated with commercial ferrofluid (EMG 304) has been reported by the sol–gel process [15]. Another strategy is based on coating of the magnetic core with materials like PEG, silica or dextran prior to reaction with the fluorescent entity [16, 17] in which the functional groups determine the linking between the coated material and the fluorophore. The other commonly explored strategy is the direct interaction of fluorescent dye or quantum dots like CdSe/ZnS, ZnO and CdTe with the magnetic core through covalent or electrostatic bonding [18–21]. This method leads to the formation of either core–shell nanocomposites or bifunctional hetero-nanostructures. The process steps of preparation of these nanocomposites are complex and are always at the risk of fluorescence quenching of the fluorophore shell by the magnetic core. Direct synthesis of ZnS, CdS and HgS quantum dots on the surface of γ -Fe2 O3 to form heteronanostructures has also been demonstrated [22–25]. In this work we report a new approach for the synthesis of superparamagnetic magnetite nanoparticles decorated with CdTe quantum dots (QDs) through dodecylamine (DDA) as the linker. It is shown that this linkage retains the fluorescent features of the QDs (disallowing charge transfer to the mangnetite) rendering them the desirable bifunctionality. Indeed, such bifunctional composite nanoparticles exhibit excellent potential in magnetic separation and fluorescent laser scanning imaging of positive leukemic cells (Molt-4) having epidermal growth factor receptor (EGFR) from a mixed population of leukemic cells. EGFR is a 170 kDa cell surface glycoprotein which belongs to the family of protein tyrosine kinases (PTK), which consists of an external EGF binding region required for cell proliferation. We selected two different human leukemic cells, Molt-4 (T-cell leukemia) and K562 (myelogenous leukemia), for our experiments. EGFR is highly expressed on Molt-4 cells, whereas K562 cells lack this receptor [26, 27]. The CdTe decorated magnetite nanoparticles were functionalized with anti-EGFR antibody and successfully used to specifically separate and image Molt-4 cells from a mixed cell suspension of Molt-4 and K562 cells.

2. Materials and methods Analytical grade iron (III) chloride (FeCl3 ), ferric sulfate (FeSO4 ), citric acid monohydrate (CA) and mercaptopropionic acid (MPA) (SRL Chemicals), tellurium powder (Te) and cadmium chloride (CdCl2 ) (SD Fine Chemicals), ammonium hydroxide NH4 OH (30% w/v) and sodium borohydride (NaBH4 ) (Merck Laboratories) were used as obtained. Deionized water was acquired from a Millipore milli- Q system with resistivity greater than 18 M cm. 2.1. Synthesis of magnetic water soluble magnetite, CdTe and CdTe decorated magnetite bifunctional nanoparticles Magnetite nanoparticles were synthesized by co-precipitation from a mixture of FeCl3 and FeSO4 (1:2 molar ratio) as described by Sahoo et al with some modifications [28]. In a typical reaction FeCl3 and FeSO4 were mixed in water and heated to 80 ◦ C under N2 atmosphere in three necked flask with continuous stirring. At this stage, (5 ml) NH4 OH was introduced by syringe and heating was continued for 30 min. After that citric acid was added to the reaction vessel and the temperature was raised to 95 ◦ C. The stirring continued for an additional 90 min at the same temperature. A small aliquot of the reaction mixture was withdrawn in a vial, diluted with water and then subjected to a static magnetic field of several hundred Gauss. The particles remained dispersed in the aqueous solvent for some time even in the presence of external magnetic field, indicating the formation of a stable suspension of magnetic nanoparticles. CdTe nanoparticles capped with mercaptopropionic acid (MPA) were synthesized by the organometallic route [42, 29]. NaHTe was used as the Te precursor and CdCl2 as the Cd precursor. NaHTe was prepared just prior to the start of CdTe synthesis by reacting NaBH4 and Te powder in molar ratio of 3:1. Te powder (3.0 mmol) was mixed with NaBH4 in a closed round bottom flask. After the reactor was filled with N2 , (0.5 ml) of N2 saturated water was added through a syringe. The reaction was operated under nitrogen flow at room temperature with magnetic stirring for about 25–30 min. The reaction was completed when the black Te powder disappeared and the solution turned completely transparent and pink in color. This freshly prepared NaHTe solution was added to the N2 saturated solution of CdCl2 (15 mM) and MPA (36 mM) at pH 9.0. The − molar ratio of Cd+ 2 /Te2 /MPA was 1:0.2:2.4. The solution was then heated and refluxed under nitrogen flow at 100 ◦ C for 2 h. CdTe decorated magnetite bifunctional nanoparticles (NPs) were prepared by controlled synthesis of CdTe onto preformed magnetite nanocrystals (NCs). Magnetite NPs were suspended in (50 ml) 50% ethanol in deionized water and (10 mmol) DDA (in 50% ethanol) was added to it. The reaction mixture was maintained at 100–120 ◦ C with constant stirring for 60 min. The DDA functionalized magnetite NCs were washed three or four times to remove unreacted DDA. For bifunctional nanoparticle preparation, DDA functionalized magnetite NCs were added to the solution of CdCl2 and MPA at pH 9.0. The solution was refluxed at 100 ◦ C for 30 min and the freshly prepared sodium hydrogen telluride (NaHTe) was added to it. The reaction mixture was further refluxed at 2


A Kale et al

100 ◦ C for 2 h. Then the heating was stopped and the reaction medium allowed to cool at room temperature. It was further kept stirring at room temperature for 24 h. The whole reaction was conducted in nitrogen atmosphere.

by placing the mixtures in a refrigerator at 4 ◦ C overnight. This allows the unreacted EDC to hydrolyze and lose its activity. The solution was further purified by centrifugation. The stock, ready-to-use solution was stored at 4 ◦ C. The luminescence intensity did not decrease after storing the samples at 4 ◦ C for over a month. The amount of antibody coupled to the nanoparticles was estimated by using a Bradford assay (Bradford protein assay kit, Bio-Rad Laboratories, Hercules, CA, USA).

2.2. Cell culture studies The human leukemic cell lines Molt-4 (T-cell leukemia) and K562 (myelogenous leukemia) were obtained from American Type Culture Collection (Rockville, MD, USA) and maintained in RPMI 1640 (Life Technologies, Rockville, MD, USA) supplemented with heat inactivated 10% fetal calf serum (Gibco BRL), 1 mM glutamine, 100 units ml−1 penicillin and 100 mg ml−1 streptomycin in a humidified atmosphere with 5% CO2 at 37 ◦ C.

2.5. Cytotoxicity assay The cytotoxicity of CdTe decorated magnetite nanoparticles on Molt-4 and K562 was assessed by MTT (3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay. Cells were seeded at a density of 5 × 104 /100 μl/well in a 96 well tissue culture plate (Falcon, USA) and treated with different concentrations of nanoparticles in the range of 0.5 ng ml−1 –1.5 μg ml−1 in fivefold increasing order. After 24 h incubation time, MTT, 10 μl (5 mg ml−1 in PBS)/well, was added and incubated for 3 h at 37 ◦ C in 5% CO2 . The formazan crystals ere solubilized by adding 10% SDS in 0.1 N HCl and incubated overnight. The optical absorbance was measured at 570 nm with the reference wavelength set at 650 nm by a microplate reader (Molecular Devices). The reading in untreated cells was considered as 100% viability.

2.3. Characterization of CdTe decorated magnetite bifunctional nanoparticles The samples were characterized for their phase purity and crystallinity by powder x-ray diffraction measurements (Panalytical Xpert Pro) with Cu KR radiation using an Ni filter. Particle sizes were investigated by transmission electron microscope (TEM), model JEOL 1200 EX, on a carbon coated TEM copper grid. HRTEM analyses were done on an FEI Technai 30 system operated at 300 kV. Thermogravimetric analysis (TGA) was performed on a TA (Thermal Analysis Instrument) SE Q-600. Vibrating sample magnetometry (VSM) was performed on a Lakeshore 7307. Optical properties of the nanoparticles were studied by UV–visible–NIR spectrophotometry in diffuse reflectance mode over the spectral range of 190–1400 nm. The measurements were carried out on a Jasco V-570 spectrophotometer. Photoluminescence measurements were performed on a Perkin-Elmer LS 55 spectrophotometer. FTIR measurements were performed on a Perkin-Elmer Spectrum One B spectrophotometer over the spectral range of 450– 4000 cm−1 . The ζ -potential as a function of the magnitude of the electric potential and the stability of the nanoparticles dispersed in water was measured using a Zetasizer, model 3000HS, Malvern, UK. DLS experiments were conducted on a Brookhaven Instrument (BIC—Brookhaven Instruments Corporation) 90 Plus particle size analyzer at a 90◦ angle using 100 μg ml−1 PBS-based particle suspensions.

2.6. Flow cytometry EGFR expression in Molt-4 and K562 cells was studied by flow cytometry analysis. Cells were harvested by centrifugation at 2000 rpm for 5 min and washed in ice cold 1× PBS (pH 7.4). Cells (105 /tube) were blocked with BSA (3% in PBS) for 1 h at 4 ◦ C. The cells were then washed in 1× PBS and incubated with EGFR mouse antibody (Santa Cruz Biotechnology) for 1 h at 4 ◦ C. Cells were then washed with PBS followed by staining with anti-mouse secondary antibody tagged with phycoerythrin (molecular probes, invitrogen) in the dark at 4 ◦ C. Finally, the cell pellet was thoroughly washed and resuspended in (500 μl) cold PBS. The negative control comprised of phycoerythrin labeled secondary antibody. Data were acquired for 10 000 events using an FACS Vantage (Becton Dickson) equipped with a 488 nm argon laser and analyzed with Cellquest Pro software. The percentage of positive cells and the mean fluorescence intensity were determined at a 1% positivity limit set for the negative control in both Molt-4 and K562 cells and a 0.30% positivity limit set for the negative control of a mixed population of Molt-4 and K562.

2.4. Conjugation of antibodies CdTe decorated magnetite bifunctional nanoparticles were conjugated to the primary antibody of EGFR mouse antibody (Santa Cruz Biotechnology) using 1-ethyl-3-(3dimethylaminopropyl)carbodiimide (EDC) as a coupling reagent [50]. The nanoparticle solution (10 μl, 5 mmol) was mixed with phosphate buffered saline (PBS) (90 μl), and EDC (1 mg) was added in PBS (100 μl) separately. Both the nanoparticle and EDC solutions were mixed and kept at room temperature with shaking for 25 min in the dark. EGFR primary Ab (20–25 μl) was added in PBS (200 μl). Both the solutions were mixed and kept at room temperature with shaking for 1.5 h in the dark. Finally, the reaction was blocked

2.7. Confocal laser scanning fluorescence microscopy Molt-4 cells–anti-EGFR–CdTe decorated magnetite conjugates after magnetic separation were harvested by centrifugation at 2000 rpm for 5 min and washed in ice cold 1× PBS. Cells were fixed with paraformaldehyde (2%) for 10 min on ice, washed thoroughly with PBS and then the nucleus was stained with DAPI for 20 min at room temperature. Cells were further washed with ice cold PBS and cytospined at 800 rpm 3


A Kale et al

for 2 min. Images were acquired with a Zeiss LSM 510 fluorescent microscope (Zeiss, Jena, Germany).

3. Results and discussion The nanocomposites of magnetic and semiconductor materials, due to combined magnetic and optical properties, are gaining increasing interest in the field of bio-applications. As mentioned earlier, various approaches have been used to fabricate magnetic–semiconductor magnetic–luminescent nanocomposites [10, 11], like embedding magnetic and semiconductor nanomaterials in nano- or micro-sized shells of natural polymers, organic polymers or silica based materials [12–14], use of linkers to connect magnetic and fluorescent moieties [18–21] or doped magnetic quantum dots [10, 11]. However, these involve multistep processes and often lead to composites with either large size or fluorescence quenching. The size of the nanocomposite is an important parameter for biological applications. The direct synthesis of fluorescent semiconductor material on the magnetic nanoparticles is the other choice for fabricating these bifunctional nanocomposites [20, 22, 23, 43, 44, 46–49]. Although the size of the composite can be restricted well under the desired nanometer range, it also may have the risk of fluorescence quenching. Many studies report the coating of magnetic nanomaterial with silica prior to the synthesis of the quantum dot to counter the quenching of the fluorescence, but use of the silica layer increases the mass loading of the magnetic–fluorescent composite there by decreasing its effective magnetization per gram. In this study we have presented the use of the organic molecule dodecylamine (DDA) by layering it on the magnetic nanoparticle first and then synthesizing the quantum dot on it. Table 1 (supplementary information SI-7 available at stacks.iop.org/ Nano/22/225101/mmedia) gives a comparison of the different properties studied in this work and some other studies where the approach of directly synthesizing semiconductor materials on a magnetic core has been used. DDA is a long carbon chain molecule so it can be used as a linker molecule between magnetite and CdTe nanoparticles by bilayer formation as explained earlier. The use of DDA, because of its organic nature and the mode of binding, does not increase mass loading significantly as compared to that of silica and also retains both magnetic and fluorescent properties in the composite construct. The approach used is facile and reproducible and can be flexible in terms of choice of both magnetic and semiconductor materials. In this work the integration of CdTe nanoparticles on the magnetite surface is accomplished by employing DDA via electrostatic interaction, where the role of surface charge is crucial. It is shown that such a linkage avoids the charge transfer from CdTe to magnetite, thereby retaining both the magnetic and the fluorescence properties. We used FTIR spectroscopy and ζ -potential measurements to elucidate the process steps. We synthesized the magnetite nanoparticles by the coprecipitation method wherein citric acid was used as a capping agent for the size control of these nanoparticles. The

Scheme 1. (a) Side chain CH2 –CH2 interaction (bilayer formation) between DDA molecules; (b) decoration of CdTe nanoparticles onto the surface of magnetite nanoparticles (gray, magnetite; yellow, CdTe).

ζ -potential of synthesized magnetite nanoparticles was found to be negative (−30 mV), which is consistent with the presence of –COO− groups protruding from the surface of the magnetite. These citric acid capped nanoparticles were then used for further functionalization with DDA. Towards this end, the nanoparticles were dispersed in water (75 mg/100 ml) and heated at 100 ◦ C in the presence of DDA for 1 h. This temperature is not sufficient for amide bond formation between the –COO− group protruding from the surface of the magnetite and the amine group of the DDA molecule. Instead, an ionic (electrostatic) interaction between the free –COO− end group and the –NH2 end group of the DDA molecule can result in the formation of ammonium salt as shown below.

The formation of amide requires strong heating of the ammonium ion with carboxylic acid and the absence of a distinct peak for the amide bond in FTIR at around 1700 cm−1 (discussed later) confirmed this. It is important to mention here that in the suggested process some fraction of particles was found to precipitate out (in water), indicating their outward non-polar CH3 termination, but a substantial fraction of particles remained dispersed in water, indicating their outward polar (NH+ 3 ) termination. Indeed, the ζ -potential measurements on these particles showed a positive value (+26 mV), confirming the presence of polar NH+ 3 termination. These dispersible particles, which exhibited an excellent magnetic property, were used for further CdTe attachment rendering concurrent fluorescent behavior. It is possible to envision the most likely scenario (please see schematic 1(a)) in which the polar NH+ 3 group can protrude outside after DDA functionalization, due to the side chain CH2 –CH2 interaction between two DDA molecules. Based on FTIR measurements and the work reported in the literature, we can infer that the side chain interaction is the distinctly favored possibility, as discussed later. We used these amine functionalized and water dispersible magnetite nanoparticles (with NH+ 3 groups protruding outside) 4


A Kale et al

3.1.2. Transmission electron microscopy (TEM). The TEM data for the magnetite/CdTe nanosystem are shown in figure 2. The image in figure 2(a) shows nanoparticles of two sizes, one typically in the range of ∼10–12 nm corresponding to magnetite nanoparticles, and another of a few nanometers corresponding to CdTe nanoparticles. Elemental scans shown in figure 2(b) clearly reveal the presence of Fe, O, Cd, and Te. More elemental scans with the line plot of elemental analysis and EDAX data are shown in the supplementary information (SI-1, SI-2, SI-3 available at stacks.iop.org/Nano/ 22/225101/mmedia) further confirm the presence of all these elements. From the elemental scans it can be noted that, although the Fe and O locations bear strict correlation as expected for the magnetite nanoparticle case, the Cd and Te locations do not have a strict correlation with Fe and O, although they have their internal correlation. This suggests that the CdTe nanoparticles are attached to functionalized magnetite and are not in core–shell form. The size of CdTe decorated iron oxide nanoparticles is found to be 11.2 ± 3.0 nm (from a Gaussian fit to the histogram in SI-4 available at stacks.iop.org/Nano/22/225101/mmedia). The EDAX analysis indicates that one Fe3 O4 particle of size 11 nm corresponds to three CdTe particles of size 3 nm in our sample. (Please refer to supplementary information SI-3 available at stacks.iop.org/ Nano/22/225101/mmedia.)

Figure 1. XRD patterns of magnetite (a), CdTe (b) and CdTe decorated magnetite (c) nanoparticles.

for further synthesis of CdTe decorated magnetite nanoparticles. The synthesis of CdTe nanoparticles was performed in situ using the corresponding precursors in the presence of the amine functionalized magnetite nanoparticles. In this process the CdTe nanoparticles are capped by mercaptopropionic acid (MPA), with the –COO− groups protruding outwards from the surface of CdTe nanoparticles. This was separately confirmed by the ζ -potential measurement of MPA functionalized CdTe nanoparticles. The negative charge on these MPA functionalized CdTe nanoparticles (ζ potential of −58 mV) facilitates binding to the DDA functionalized magnetite through electrostatic interaction to form CdTe decorated magnetite nanoparticles, as shown in schematic 1(b). The nanoparticle colloidal dispersions are moderately stable at each step of the reaction, with adequate electrical charge to resist the aggregation of the particles. The final nanosystem is found to be negatively charged (ζ potential of −31 eV), making it ideal for interaction with SH and NH2 end groups of biomolecules.

3.1.3. FTIR spectra. Figures 3(I)–(III) show the FTIR spectra of pure DDA (a), citric acid (CA) functionalized magnetite (b), DDA functionalized magnetite (c), CdTe decorated magnetite (d) and CA alone (e). The characteristic peak at 610 cm−1 corresponds to the metal oxide, i.e. Fe– O. FTIR spectroscopy further reveals that the surface functionalization of the magnetite particles occurs through the (–COO− ) groups, since the corresponding characteristic frequency (C=O stretch) is seen to shift from 1710 to 1630 cm−1 . In this bonding the carboxylate groups of CA undergo complexation with the Fe atoms on the magnetite surface, rendering a partial single bond character to the C=O bond. This weakens the double bond and shifts the stretching frequency to a lower value [28, 30]. As discussed earlier, when CA capped magnetite nanoparticles are heated at 100 ◦ C in the presence of DDA, there is an ionic interaction between the free –COO− end group protruding from the magnetite surface and the –NH2 end group of DDA molecules (ammonium salt), this temperature being insufficient for amide bond formation. This is borne out by the signatures at ca 1400 and 1615 cm−1 in the case of DDA functionalized magnetite, as shown in figure 3(II). These can be assigned to the asymmetric and symmetric stretching vibrations of –COO− groups, but as affected by the electrostatic interaction with the amine group of DDA molecule [31]. As stated earlier, the water dispersibility of the nanoparticles implies the presence of the amine groups outward. The DDA amine group interaction with citrate COO− would make the CH3 group protrude outside, and this cannot explain the water dispersibility. The observed water dispersibility necessarily needs protrusion of NH+ 3 outside, which must involve additional DDA molecules.

3.1. Physicochemical properties of the magnetite–CdTe nanosystem 3.1.1. X-ray diffraction. The x-ray diffraction patterns for the iron oxide, CdTe and CdTe decorated iron oxide nanoparticles are shown in figure 1. All the key plane systems in figures 1(a) and (c) can be identified with the magnetite (Fe3 O4 ) phase, although it has to be admitted that distinguishing between magnetite, non-stoichiometric magnetite, and maghemite cannot be easily done by routine XRD measurements. The line broadening at the full width at half maximum (FWHM) allows approximate estimation of the mean size of the crystallites from the Scherrer formula. The estimated size of magnetite NPs is ∼11 nm. In the case of CdTe, the diffraction peaks are broad and the three main peaks centered at about 2θ ∼ 25◦ (111), 40◦ (220) and 47◦ (311) correspond well to the three most intense peaks for bulk CdTe. The mean size of CdTe is found to be ∼3 nm by the Scherrer formula. 5


A Kale et al

Figure 2. (a) HRTEM image of CdTe decorated magnetite nanoparticles. (b) Selected area elemental scan of CdTe decorated magnetite nanoparticles. (Red—Fe, bluish green—O, green—Cd, blue—Te.)

CH2 side chain is significantly higher because of the higher number of the corresponding bonds in the molecule. Moreover the signatures for these bonds (CH3 –CH3 end groups at 2870 (symmetric stretch) and 2956 (antisymmetric stretch) cm−1 and CH2 –CH2 side chain at 2850 (symmetric stretch) and 2920 (antisymmetric stretch) cm−1 ) are separated in wavenumber to be able to identify them separately [36]. If only the end

If these molecules attach via CH3 –CH3 end groups or via the side chain CH2 –CH2 interaction (bilayer formation in schematic 1(a)) between DDA molecules, this can lead to water dispersibility [32–34]. It may be noted from the FTIR data for the DDA [35] molecule case shown in figure 3(III) that the intensity of C–H symmetric and antisymmetric stretching modes for the CH3 end group is very low, while that for the 6


A Kale et al

Figure 3. (I) FTIR spectra of (a) pure dodecylamine (DDA), (b) citric acid functionalized magnetite, (c) DDA functionalized magnetite, (d) CdTe decorated magnetite nanoparticles and (e) citric acid alone; (II) magnified view in the range of 1300–2000 cm−1 and (III) magnified view in the range of 2700–3000 cm−1 .

group interaction had occurred, the signatures corresponding to the side chain CH2 bonds would not have shown significant change. In the actual data however (figure 3(III)), the symmetric and antisymmetric stretching signatures for the CH2 side chain are seen to be shifted clearly and substantially. Thus, such interactions are unambiguously present in the system, leading to the outward protrusion of NH+ The 3. CH3 –CH3 end group interactions may also be present, which leads to the same outward protrusion of NH+ 3 and is not in conflict. The interaction between alkyl chains is of van der Waals type, due to which CH2 stretching vibrations shift to higher wavenumbers as compared to those in pure DDA. It is useful to point out further that, in a system of monolayers of alkyl chains with more than four methylene units, considerable end to end ordering between alkyl chains leads to vanishing of the resonant signal from the methylene groups and only bands due to terminal CH3 groups are present in the spectra [37]. However, if there is a lesser degree of order or more gauche defects [38, 39] the spectrum consists predominantly of CH2 vibrational modes. Hence the strong presence of CH2 symmetric and asymmetric bands in our case confirms the presence of a less ordered bilayer of DDA in the same. This could be understood for an object with positive curvature such as a nearly spherical nanoparticle.

Figure 4. TGA pattern of magnetite (a) and CdTe decorated magnetite (b) nanoparticles.

for the citric acid capped magnetite nanocrystals. On the other hand, in the case of CdTe decorated magnetite bifunctional NPs the TGA curve exhibits two weight loss steps: first above 210 ◦ C and second above 500 ◦ C. In the case of magnetite NPs the onset of weight loss temperature can be attributed to the enthalpy of adsorption of the citric acid molecules on the magnetite surface. The weight loss for magnetite nanoparticles is ∼5.5%. This weight loss is thus due to the removal of citric acid from the surface of magnetite nanoparticles. The TGA curve for the CdTe decorated magnetite indicates a weight loss

3.1.4. Thermogravimetric analysis (TGA). Figure 4 shows the thermogravimetric analysis (TGA) data for magnetite and CdTe decorated magnetite NPs recorded in nitrogen atmosphere. The weight loss onset temperature is above 185 ◦ C 7


A Kale et al

(a)

(b)

Figure 6. Photographic images of the CdTe decorated magnetite fluorescent bifunctional nanoparticles without UV irradiation and under UV illumination (λ = 365 nm) with and without an external magnetic field. Figure 5. (a) Hysteresis curves of magnetite nanoparticles (red curve) and CdTe decorated magnetite nanoparticles (black curve), at room temperature. (b) UV–visible and photoluminescence spectrum of CdTe decorated magnetite nanoparticles.

magnetization of CdTe decorated magnetite after normalization to emu per gram of magnetite content. The saturation magnetization was M s = 44 emu g−1 . It indicates that the nanocomposite retained the superparamagnetic property of the core magnetic component. The reduction in M s value for CdTe decorated magnetite can be attributed to the surface spin disorder due to capping. The absorbance and photoluminescence spectra of CdTe decorated magnetite in water are shown in figure 5(b). The excitonic emission peak is seen at 531 nm below the CdTe band edge, as expected. The fluorescence quantum yield (QY) of the CdTe decorated magnetite nanocomposite is found to be 18 ± 2% as measured against Rhodamine B (QY.0.95 in ethanol) as a standard. Earlier some novel methods were reported for the synthesis of magnetic–fluorescent (core) shell or hetero-dimer nanocomposites. But these nanocomposites showed lower QY, in the range of about 2–5%, for Fe3 O4 @– SiO2–CdSe/ZnS [28], Co/CdS [43] and FePt/CdS [44], 7.5– 10% for Co/CdSe [43] and 10–15% for Fe3 O4 /CdSe/ZnS [20] nanoparticles. Chang et al have reported a QY of up to 42% for Fe2 O3 –CdSe nanocomposites [18], but in an organic solvent. The emission quantum yield of CdTe decorated magnetite changed little over a period of one month, reflecting the resistance of nanocomposites to photobleaching. Figure 6 shows the photographic images of the CdTe decorated magnetite magnetic–fluorescent bifunctional nanoparticles without UV irradiation and under UV illumination (λ = 365 nm) with and without an external magnetic field. Clearly

of 20%. This weight loss can be understood from the loss of stabilizers or ligands as follows: % weight loss = 100 ×

(4πdp2 /a)(M/No ) [((1/6)πdp3 ρ) + (πdp2 /a)(M/No )]

where dp = average size of the nanoparticle of ∼11 nm (from ˚ 2 , ρ = density TEM analysis), a = area of monolayer ∼21 A of magnetite, N = Avogadro’s number, M = atomic weight of stabilizer/ligand = (192.12) amu for citric acid [40, 41]. The calculated percentage weight loss for magnetite nanoparticles is ∼5%, which is quite close to the experimentally observed weight loss of 5.5%. In the case of the CdTe decorated magnetite, the increase in the weight loss as compared to core magnetite NCs can be attributed to the loss of surfactant (DDA) and stabilizer (MPA) present in the bifunctional NPs [42]. 3.1.5. Magnetization and photoluminescence. Figure 5(a) shows the room-temperature magnetic hysteresis for magnetite and CdTe decorated magnetite NPs. It indicates that magnetite nanoparticles (black curve) are in the superparamagnetic regime at room temperature, reaching a saturation magnetization value ( M s) of ∼62 emu g−1 . The red curve shows the 8


A Kale et al

the nanosystem synthesized by us is an integrated one between magnetic and luminescent nanoparticles and all luminescent nanoparticles are attached to the magnetic ones. 3.1.6. Stability of CdTe decorated magnetite nanoparticles. The stability of the CdTe decorated magnetite nanoparticles was studied in phosphate buffer containing KH2 HPO4 , K2 HPO4 , NaCl and KCl in TST buffer containing Tris, NaCl and Tween 20. The MPA capped nanocomposites were diluted in the buffer at increasing values of pH 1, 3, 5, 7, 9 and 11. It was observed that the nanoparticle suspensions were most stable in the pH range of 5–9 with high luminescence (refer to the supporting information SI5, SI-6 available at stacks.iop.org/Nano/22/225101/mmedia). The suspensions were stable for more than 1.5 h in this range. The stability decreased beyond pH 3 and 9 with drastic reduction in luminescence intensity at pH 3 and 1. The emission wavelength at pH 5 was 531 nm. At pH 3 and 1 the nanoparticles showed a hypsochromic shift of about 3 nm, while a bathochromic shift of up to 4 nm was observed beyond pH 7. Hypsochromic and bathochromic shifts observed in our studies are similar to those reported by Algar and Krull [45] in their elegant study on luminescence and stability of thioalkyl acid capped QDs. MPA, being a smaller monodentate molecule, can pack densely on the surface of nanoparticles. The electrostatic repulsion of these thiol molecules is responsible for keeping nanocomposite dispersed in the aqueous solvent. The data for hydrodynamic size and size distribution of magnetite–CdTe nanoparticles are provided as the supplementary information in the revised manuscript. The diameter of the CdTe decorated iron oxide nanoparticles is found to be 11.2 ± 3.0 nm (from a Gaussian fit to the histogram in SI-4 available at stacks.iop.org/Nano/22/225101/ mmedia). The hydrodynamic size (diameter) of magnetite– CdTe magnetic–fluorescent composite nanoparticles is found to be 103 ± 5.5 nm. The hydrodynamic radius is rather higher compared to the actual size of the nanocomposites. This can be attributed to the aggregation of the nanocomposites due to the presence of a magnetic core in the composites and electrostatic interactions between the charged nanoparticles.

Figure 7. Cytotoxicity studies of CdTe decorated magnetite nanoparticles on Molt-4 and K562 cell lines by MTT assay. The cells were treated with serial doses of CdTe decorated magnetite nanoparticles for 24 h. The viability recorded in the control was considered as 100% and used to calculate the percentage viability in the treated sample. The data depict mean ± S.E. of three independent experiments.

even at the highest concentration of 1.5 μg ml−1 . Cell cytotoxicity assay after exposing both types of cell to nanocomposites for 24 h clearly showed no significant decrease in cell viability. Phase contrast microscopy observations of Molt-4 and K562 cells after treating with nanoparticles indicated that cells were quite healthy. Hence we decided to use these nanoparticles for cell surface imaging and magnetic separation by using live cell staining. 3.2.2. EGFR expression and magnetic separation of Molt4 and K562 cells. The expression of EGFR on Molt-4 and K562 cells was studied by flow cytometry. Figure 8 shows that 99.54% of the Molt-4 cells were positive for EGFR expression, with mean fluorescence intensity (MFI) of 58.57, as compared with control negative cells, which were set to 1% positivity and their MFI was 14.77. EGFR is highly expressed on the cell surface of Molt-4 cells which can be clearly seen in figures 8(a)–(c): 99.54% of the Molt-4 cells are positive for EGFR expression, which correlates with the up-regulation of EGFR in tumor progression in Molt-4 cells. In this paper we are using anti-EGFR antibody, which is recommended for detection of EGFR on cell surface specifically. Hence we can observe EGFR expression on the surface of Molt-4 cells. In contrast, only 0.91% of the K562 cells were positive for EGFR expression with MFI of 39.95, as compared with control negative cells, which were set to 1% positivity and their MFI was 38.05. This result clearly confirms that Molt-4 cells are positive for EGFR expression and K562 cells are negative for it. To demonstrate the bio-separation prospects of these magnetic–fluorescent nanoparticles, we first prepared a mixed suspension of Molt-4 and K562 cells in culture tubes. Anti-EGFR antibody conjugated CdTe decorated magnetite nanoparticles were added to the cell suspension and incubated for 1 h with slight shaking. It was expected that, during the incubation period, nanoparticles would specifically bind to Molt-4 cells. The culture tubes were kept near a strong magnet (of magnetic field strength 425 gauss) for an adequate time and we observed that some cells were attracted towards the

3.2. Bio-separation and imaging using bifunctional CdTe decorated magnetite NPs We tested our nanoparticles for specific selection, magnetic separation and imaging of EGFR expressing leukemic Molt4 cells from the K562 cells which lack this receptor. We first assessed the biocompatibility of these nanoparticles on Molt-4 and K562 cells. 3.2.1. Cell viability assay. CdTe decorated magnetite nanoparticles were tested for cell viability on Molt-4 cells (T-cell leukemic) and K562 cells (myelogenetic leukemia) by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide) assay. The nanoparticle concentrations were taken in the range of 0.5 ng ml–1.5 μg ml−1 in fivefold increasing order. From figure 7 it is clearly seen that CdTe decorated magnetite nanoparticles are not toxic to cells 9


A Kale et al

Figure 8. Flow cytometry analysis of expression of EGFR in human leukemic cell lines Molt-4 and K562. (a)–(c) Representative plots of EGFR expression in Molt-4 cells. (a) Scatter plot analysis of Molt-4 cells. (b) Histogram of negative control of Molt-4 cells consisting of only phycoerythrin labeled secondary antibody. (c) Histogram of EGFR positive Molt-4 cells. (d)–(f) Representative plots of EGFR expression in K562 cells. (d) Scatter plot analysis of K562 cells. (e) Histogram of negative control of K562 cells, consisting of only phycoerythrin labeled secondary antibody. (c) Histogram of K562 cells as positive control. (g)–(j) Expression of EGFR in a mixed population of Molt-4 and K562. (g) Scatter plot of a mixed population of Molt-4 and K562 cells. (h) Histogram of negative control of a mixed population of Molt-4 and K562 cells, consisting of only phycoerythrin labeled secondary antibody. (i) Histogram of positive control of a mixed population of Molt-4 and K562 cells. (j) Histogram analysis of a cell suspension left after immunomagnetic separation of Molt-4 cells by CdTe decorated magnetite nanoparticle–anti-EGFR antibody conjugate.

in the solution before conjugation and after conjugation were determined by the Bradford protein assay. The amount of immobilized EGFR antibody was about 300 μg Ab/mg of nanoparticles. Considering the size of the EGFR antibody (170 kDa) and the size of the Fe3 O4 –CdTe nanocomposite (11 ± 3 nm), the average number of antibodies bound to nanocomposites was estimated to be about three per nanoparticle.

magnet (schematic 2). The population of cells not attracted to the magnet was analyzed by flow cytometry. It showed that 0.92% of cells were only positive for EGFR expression with an MFI of 35.11, as compared with control negative cells, which were set to 0.30% positivity and 27.52 MFI. These results confirm the successful magnetic separation of Molt-4 cells from the mixed cell suspension by magnetite– CdTe nanoparticles functionalized with anti-EGFR antibody. It is noteworthy that both Molt-4 and K562 cells demonstrated comparable forward scatter and side scatter on dot plot analysis by flow cytometry, suggesting that they are similar in size with similar granularities. It is non-trivial to separate such cells. The advantage of this CdTe decorated magnetite mediated bioseparation process is that we are able to separate Molt-4 cells from K562 cells based on the expression of EGFR. The same procedure was repeated with a Molt-4 cell only suspension. The remaining suspension after magnetic separation of Molt4 cells was analyzed by flow cytometry. We found that the residual population had only 6% of the cells, clearly indicating that 94% of the Molt-4 cells were bound to the CdTe decorated magnetite–anti-EGFR conjugate and were magnetically separated.

3.2.4. Confocal laser scanning microscopy. CdTe decorated magnetite–anti-EGFR targeted Molt-4 cells were imaged by confocal laser scanning microscope after magnetic separation. Figure 9 clearly shows an outer fluorescent green ring on the plasma membrane of Molt-4 cells because of the binding of CdTe decorated magnetite–anti-EGFR antibody to EGFR present on the cells. The nucleus is stained blue with DAPI (41 ,6-diamidino-2-phenylindole, dihydrochloride). This confirms the localized binding of CdTe decorated magnetite nanoparticle–anti-EGFR conjugate to Molt-4 cells. The nanocomposites have thus been successfully used as an imaging tool for leukemic cells.

4. Conclusion 3.2.3. Antibody binding to nanocomposites. After the EGFR antibody was bound to nanocomposites, conjugates were precipitated by centrifugation. The amounts of antibody

In summary, we report a protocol for the synthesis of biocompatible bifunctional superparamagnetic CdTe decorated 10


A Kale et al

Scheme 2. Schematic illustration of immunomagnetic separation and fluorescence detection of EGFR+ Molt-4 cells by using CdTe decorated magnetite bifunctional nanoparticles.

Figure 9. Localization and expression of CdTe decorated magnetite nanoparticle–EGFR conjugate in Molt-4 cells. Molt-4 cells were incubated with CdTe decorated magnetite nanoparticle–EGFR conjugate for 1 h, washed with PBS, fixed with 2% paraformaldehyde, cytospined at 800 rpm for 2 min on slides and visualized under a confocal microscope. (a)–(d) The confocal image shows a nanoparticle–antibody conjugate ring (green) on the plasma membrane of the cell. It also shows the nucleus stained with DAPI. All images are captured at 60× magnification.

11


A Kale et al

magnetite nanoparticles. The strategy of forming CdTe decorated on the magnetite core with DDA as an intermediate linker is the key to this method. We demonstrate the use of these magnetic–fluorescent nanoparticles in effective bioseparation and fluorescence imaging of EGFR positive Tcell leukemic Molt-4 cells from a mixed suspension of cells with the same size and similar granule patterns, which is a non-trivial task. These bifunctional nano-tools hold great potential in current clinical diagnostics and therapeutics for bio-separation, bio-imaging and drug delivery. They can find use in several in vitro and possibly in vivo biological applications.

[18] Ang C Y, Giam L, Chan Z M, Lin A W H, Gu H, Devlin E, Papaefthymiou G C, Selvan S T and Ying J Y 2009 Adv. Mater. 21 869 [19] You X, He R, Gao F, Shao J, Pan B and Cui D 2007 Nanotechnology 18 035701 [20] Du G H, Liu Z, Lu Q, Xia X, Jia L, Yao K L, Chu Q and Zhang S M 2006 Nanotechnology 17 2850 [21] Turgeman R, Tirosh S and Gedanken A 2004 Chem.—Eur. J. 10 1845 [22] Kwon K and Shim M 2005 J. Am. Chem. Soc. 127 10269 [23] Gao J, Zhang W, Huang P, Zhang B, Zhang X and Xu B 2008 J. Am. Chem. Soc. 130 3710 [24] Chin S F, Iyer K S and Raston C L 2009 Cryst. Growth Des. 9 2685 [25] Xu Z, Hou Y and Sun S 2007 J. Am. Chem. Soc. 129 8698 [26] Allen H, Hsuan J, Clark S, Maziarz R, Waterfield M D, Flavell R A and Haley J 1990 Biochem. J. 271 785 [27] Kimura H, Sakai K, Arao T, Shimoyama T, Tamura T and Nishio K 2007 Cancer Sci. 98 1275 [28] Sahoo Y, Goodarzi A, Swihart M T, Ohulchanskyy T Y, Kaur K, Furlani E P and Prasad P N 2005 J. Phys. Chem. B 109 3879 [29] Kuo Y C, Wang Q, Ruengruglikit C, Yu H and huang Q 2008 J. Phys. Chem. C 112 4818 [30] Lan Q, Liu C, Yang F, Liu S, Xu J and Sun D 2007 J. Colloids Interface Sci. 310 260 [31] Zhang H, Sun P, Liu C, Gao H, Xu L, Fang J, Wang M, Liua J and Xu S 2009 Luminescence 2009 doi:10.1002/bio.1188 [32] Bromberg L, Chen L, Chang E P, Wang S and Hatton T A 2010 Chem. Mater. 22 5383 [33] Shen L, Stachowiak A, Fateen S E K, Laibinis P E and Hatton T A 2001 Langmuir 17 288 [34] Avdeev M V, Aksenov V L, Balasoiu M, Garamus V M, Schreyer A, Toeroek G, Rosta L, Bica D and Vekas L 2006 J. Colloids Interface Sci. 295 100 [35] Yang J, Sargent E H, Kelley S O and Ying J Y 2009 Nat. Mater. 8 683 [36] Wang X, Liu J, Du H and Miller J D 2010 Langmuir 26 3407 [37] Gragson D E, Mackarty B M and Richmond G L 1996 J. Phys. Chem. 100 14272 [38] Messmer M C, Conboy J C and Richmond G L 1995 J. Am. Chem. Soc. 117 8039 [39] Liang G L, Noid D W, Sumpter B G and Wunderlich B 1994 J. Phys. Chem. 98 11739 [40] Todorovsky D S, Dumanova D G, Todorovska R V and Getsova M M 2002 Croat. Chem. Acta 75 155 [41] Badia A, Cuccia L, Demers L, Morin F and Lennox R B 1997 J. Am. Chem. Soc. 119 2682 [42] Shen L, Laibinis P E and Hatton T A 1999 Langmuir 15 447 [43] Kim H, Achermann M, Balet L P, Hollingsworth J A and Klimov V I 2005 J. Am. Chem. Soc. 127 544 [44] Gu H, Zheng R, Zhang X and Xu B 2004 J. Am. Chem. Soc. 126 5664 [45] Algar W R and Krull U J 2007 ChemPhysChem 8 561 [46] Sun P, Zhang H, Liu C, Fang J, Wang M, Chen J, Zhang J, Mao C and Xu S 2010 Langmuir 26 1278 [47] Xu Y et al 2010 J. Phys. Chem. C 114 5020 [48] Selvan S T, Patra P K, Ang C Y and Ying J Y 2007 Angew. Chem. Int. Edn 46 2448 [49] Wang G, Xie P, Xiao C, Yuan P and Su X 2010 J. Fluoresc. 20 499 [50] Hermanson G T 2008 Bioconjugate Techniques (San Diego, CA: Academic) pp 216–219, 493–495, 598

Acknowledgments SO and BH acknowledge funding by CEFIPRA (Indo-French Center for Promotion of Advanced Research project code 3808-2). AK and SK are grateful to the Department of Biotechnology, India, for financial support. The authors would like to acknowledge help by Dr Kashinath Bogle and Professor N Valanoor for TEM and the elemental scan.

References [1] Lu A H, Salabas E L and Ferdi S 2007 Angew. Chem. Int. Edn 46 1222 [2] Park K, Lee S, Kang E, Kim K, Choi K and Kwon I C 2009 Adv. Funct. Mater. 19 1553 [3] McCarthy J R and Weissleder R 2008 Adv. Drug Deliv. Rev. 60 1241 [4] Kelsall R W, Hamley I W and Geoghegan M 2005 Nanoscale Science and Technology (West Sussex: Wiley) [5] Pankhurst Q A, Connolly J, Jones S K and Dobson J 2003 J. Phys. D: Appl. Phys. 36 R167 [6] Sun C, Lee J S H and Zhang M 2008 Adv. Drug Deliv. Rev. 60 1252 [7] Alivisatos A P, Gu W and Larabell C 2005 Annu. Rev. Biomed. Eng. 7 55 [8] Michalet X, Pinaud F F, Bentolila L A, Tsay J M, Doose S, Li J J, Sundaresan G, Wu A M, Gambhir S S and Weiss S 2005 Science 307 538 [9] Gao X, Yang L, Petros J A, Arshall F F, Simons J W and Nie S 2005 Curr. Opin. Biotechnol. 16 63 [10] Corr S A, Rakovich Y P and Gunko Y K 2008 Nanoscale Res. Lett. 3 87 [11] Quarta A, Corato R D, Manna L, Ragusa A and Pellegrino T 2007 IEEE Trans. Nanobiosci. 6 298 [12] Nagao D, Yokoyama M, Yamauchi N, Matsumoto H, Kobayashi Y and Konno M 2008 Langmuir 24 9804 [13] Tu C, Yang Y and Gao M 2008 Nanotechnology 19 105601 [14] Roullier V, Grasset F, Boulmedais F, Artzner F, Cador O and Marchi-Artzner V 2008 Chem. Mater. 20 6657 [15] Lu Y, Yin Y, Mayers B T and Xia Y 2002 Nano Lett. 2 183 [16] Nitin N, LaConte L E W, Zurkiya O, Hu X and Bao G 2004 J. Biol. Inorg. Chem. 9 706 [17] Shi D et al 2009 Adv. Mater. 21 2170

12