Chemical modification of multiwalled carbon nanotube with a bifunctional caged ligand for radioactive labelling

Available online at www.sciencedirect.com

ScienceDirect Acta Materialia 64 (2014) 54–61 www.elsevier.com/locate/actamat

Chemical modification of multiwalled carbon nanotube with a bifunctional caged ligand for radioactive labelling Ross W. Ormsby a, Tony McNally b, Christina A. Mitchell c, Anthony Musumeci d, Tara Schiller e, Peter Halley f, Lawrence Gahan f, Darren Martin f, Suzanne V. Smith g, Nicholas J. Dunne a,⇑ a

School of Mechanical and Aerospace Engineering, Queen’s University of Belfast, Ashby Building, Stranmillis Road, Belfast BT9 5AH, UK b International Manufacturing Centre, University of Warwick, Coventry CV4 7AL, UK c School of Medicine, Dentistry and Biomedical Sciences, Queen’s University of Belfast, Grosvenor Road, Belfast BT12 6BP, UK d Qutbluebox Ltd., ABN, Musk Avenue, Kelvin Grove, Queensland 4059, Australia e Materials Research, Clayton Campus, Monash University, Clayton, Victoria 3800, Australia f The Australian Institute of Bioengineering and Nanotechnology, St. Lucia, The University of Queensland, Queensland 4072, Australia g Collider Accelerator Department, Brookhaven National Laboratory, Upton, NY 11973, USA Received 23 July 2013; received in revised form 22 October 2013; accepted 23 October 2013 Available online 18 December 2013

Abstract Carboxyl-functionalized multiwalled carbon nanotubes (MWCNTs) have been successfully radiolabelled with cobalt-57 (57Co) (T = 270 days) via the attachment of the bifunctional caged ligand MeAMN3S3sar. In this study MeAMN3S3sar has been synthesized and coupled to MWCNTs to form the conjugate MWCNT–MeAMN3S3sar. Synthesis was confirmed with nuclear magnetic resonance. X-ray photoelectron spectroscopy (XPS) confirmed the conjugation. Non-radioactive labelling of this conjugate was completed with Cu(II) ions to confirm the stability of the MeAMN3S3sar after coupling with the MWCNTs. The complexation of the Cu(II) was also confirmed with XPS. Transmission electron microscopy was used to demonstrate that the coupling reaction had a negligible effect on the size and shape of the MWCNTs. Radiolabelling of the MWCNT–MeAMN3S3sar conjugate and pristine (untreated) MWCNTs (nonspecific) with the gamma-emitting radioactive isotope 57Co were compared. The radiolabelling efficiency of the MWCNT–MeAMN3S3sar conjugate was significantly higher (95% vs. 0.1%) (P 6 0.001) than for the unconjugated pristine MWCNTs. This will allow for the potential tracking of nanoparticle movement in vitro and in vivo. Crown Copyright Ó 2013 Published by Elsevier Ltd. on behalf of Acta Materialia Inc. All rights reserved. 1/2

Keywords: Multi-walled carbon nanotube; Radioactive labelling; Bifunctional caged ligand; Nanoparticle tracking

1. Introduction In recent years there have been increased efforts to determine how artificially introduced nanoparticles interact with human tissues and systems in vivo, with an emphasis on tracking their movement. Although nanoparticles are used in a number of consumer products, including photocatalysis [1], cancer therapy [2–6] and composites [7–20], not ⇑ Corresponding author. Tel.: +44 28 90934122; fax: +44 28 90661729.

E-mail address: [email protected] (N.J. Dunne).

enough is understood about their biological interactions and toxicity. Such biological responses may be dependent on the surface chemistry of a nanoparticle, in addition to their size, shape and chemical composition. The ability to predict the relationship between physicochemical parameters and biological systems has proven to be a very complicated and challenging assignment to date. To further the understanding of nanoparticle interactions with biological systems, it is important to develop highly sensitive, reliable and robust methodologies to label such nanoparticle systems.

1359-6454/$36.00 Crown Copyright Ó 2013 Published by Elsevier Ltd. on behalf of Acta Materialia Inc. All rights reserved. http://dx.doi.org/10.1016/j.actamat.2013.10.047

R.W. Ormsby et al. / Acta Materialia 64 (2014) 54–61

Radioisotopic (active) labelling is one such method offering unparalleled detection sensitivity and compatibility with nanoparticle synthesis methodologies. The attachment of radioactive metal ions to multiwalled carbon nanotubes (MWCNTs) could potentially allow for labelling and tracking. One type of functionalized carbon nanotube (CNT) previously explored in biomedical applications is based on the covalent surface modification developed by Kostarelos et al. [21]. These CNTs have been studied for various applications, including imaging using various radionuclides (111In, 86Y) [22–24]. Another type of functionalized CNT that has been explored in vivo is based on the chemical modification of carboxylic acid groups after strong acid treatment. These CNTs have also been studied using tracing radionuclides (125I, 14C) [25,26]. However, in order to track CNT movement within a polymer matrix it is essential that in designing these labelling techniques, it can be demonstrated that the label is stable and that the native physical characteristics of the nanoparticles are not altered. Furthermore, these labelling techniques may then be used to monitor nanoparticle transport and interactions in typical, but realistic environments. The attachment of a radioactive metal ion to nanoparticles can be achieved by either direct labelling (where a radioemitting element is inserted into the MWCNT structure) [27], or with the use of a bifunctional ligand. Direct labelling is difficult to control and may lead to unplanned changes to the geometry and behaviour of the nanoparticles. A more suitable technique (in theory) for attachment of radioactive metal ions to MWCNTs is via a bifunctional ligand. Such ligands offer a highly favourable three-dimensional, cage-like structure capable of the stable encapsulation of metal ions. In addition, the ligands are also able to readily bind to the target nanoparticles through a favourable functional group [27,28]. For example, the bifunctional caged ligand MeAMN3S3sar has an aromatic amine (ANH2) functional group attached, which would allow for coupling with nanoparticles that have a carboxyl (ACOOH) group on the surface (Fig. 1). To achieve a high conjugation efficiency of the bifunctional ligand to a MWCNT, the ligand has to be free from any co-products formed during the reaction [29].

Fig. 1. Schematic diagram of the MeAMN3S3sar bifunctional cage ligand [27].

55

The potential to radioactively label and track MWCNTs is of particular interest. Previously, we have demonstrated that the incorporation of MWCNTs of varied chemical functionality to polymethyl methacrylate (PMMA)-based bone cements can significantly augment the mechanical and thermal properties of the resultant composite cement [30–32]. Although our previous work has shown that the MWCNT–PMMA bone cements did not invoke a cytotoxic or negative response when in contact with osteoblast-like cells [33], it is still crucial to understand the potential distribution of these MWCNTs within the human body. Therefore, the aims of this study were twofold: (i) to radioactively label MWCNTs with cobalt-57 (57Co) isotopes via coupling the MWCNTs with a bifunctional caged ligand (MeAMN3S3sar); and (ii) to assess the radioactive retention of the radioactively labelled MWCNTs. 2. Experimental procedures 2.1. Materials preparation 2.1.1. MWCNTs The MWCNTs used in this study were carboxyl-functionalized (MWCNT-COOH) (4 wt.% COOH concentration) MWCNTs (Nanocyl SA, Belgium). These MWCNTs were all grown using chemical vapour deposition and had average diameters of 9.5 nm and average lengths of 61 lm [34]. 2.1.2. Synthesis of the bifunctional caged ligand MeAMN3S3sar The Co(III) complex [Co(MeNH3N3S3Sar)]Cl42H2O, and the free ligand MeAMN3S3sar were prepared as described previously [27–29]. The desired bifunctional caged ligand MeAMN3S3sar was isolated as a white solid. 13C nuclear magnetic resonance (NMR) (D2O, 1,4-dioxane): d = 34.67, 40.39, 49.23, 54.91 (ACH2A), 42.64, 56.91 (Cq), 28.34 ppm (CH3); mass spectroscopy (MS) (electron impact (EI), theoretical): m/z: 364.18; found: 365.23 [MeAMN3S3sar]+; elemental analysis calculated (%) for C15H32N4S30.5H2O: C48.22, H8.90, N14.99, S25.75%; found: C48.23, H8.97, N14.82, S25.95. 2.1.3. Coupling MWCNTs with the bifunctional caged ligand MeAMN3S3sar The COOH functional groups on the surface of the MWCNTs were coupled with the NH2 functional group on the caged ligand to form a stable covalent bond. This was completed by an esterfication reaction using the coupling agents N-(3-dimethylaminopropyl)-N0 -ethylcarbodiimide (EDC) and N-hydroxysuccinimide (NHS) (all chemicals supplied by Sigma–Aldrich, UK unless otherwise stated) to form a stable active ester bond (Fig. 2). The methodology described is a modified version of the technique used by Jiang et al. [35] for coupling proteins to MWCNT-COOH. This was achieved by suspending 20 mg MWCNTs in 60 ml 0.1 M MES-buffered solution (pH 6.2) via sonication at an amplitude of 10 mA for 4 min. This

56

R.W. Ormsby et al. / Acta Materialia 64 (2014) 54–61

Fig. 2. Schematic representation of the synthetic reaction of MWCNT-COOH covalently linked to NH2 group of the caged ligand through the EDC/NHS conjugation reaction.

solution contained 1.788  105 moles of COOH groups. As there are 4.0 wt.% COOH groups on the MWCNTs, this equates to 0.8 mg of the 20 mg of MWCNTs. A 10-fold excess of the coupling agents (NHS and EDC) was added in relation to the available COOH groups. NHS solution (0.4 M, 23 ml) was added to 20 mg MWCNTs in 60 ml 0.1 M MES-buffered solution and stirred using a magnetic stirrer for 5 min. EDC solution (0.1 M, 12 ml) was then added to the mixture and stirred for a further 30 min. The solution was then filtered under vacuum through a 0.2 lm hydrophilized PTFE membrane (PALL Life Sciences, HT-200 Tuffryn Membrane Filters, 47 mm diameter). The retained MWCNTs were rinsed thoroughly using 1 l of deionized water and 1 l of MES-buffered solution to remove excess NHS/EDC. The MWCNTs were collected and dried for 40 min in an oven at 60 °C. The dried MWCNT precipitate was redispersed in 90 ml 0.1 M MESbuffered solution (pH 6.2) by ultrasonic dispersion. A 10 ml solution of 3.6 mM MeAMN3S3sar caged compound was then added and continually stirred for 1 h at room temperature (this concentration was produced by adding 12.9 mg of caged compound to 10 ml of deionized water). Based on previous calculations for the number of moles of COOH groups, a 2-fold concentration of MeAMN3S3sar caged compound would require 3.556  105 moles of caged molecule. This solution was then filtered under vacuum as per the method described above. The filtered MWCNTs were collected and dried in an oven at 60 °C for 12 h. 2.1.4. Non-radioactive analogue labelling of MWCNTs Non-radioactive analogues of the MWCNT–MeAMN3S3sar were labelled with natural copper, Cu(II), and assessed using X-ray photoelectron spectroscopy (XPS). This was completed to confirm the conjugation of the caged ligand to the MWCNTs via the coupling reaction described above.

MWCNT–MeAMN3S3sar (10 mg) was ultrasonically dispersed in 500 ml of deionized water at amplitude of 10 mA for 5 min. Then 6.59 mg of copper perchlorate (Cu(ClO4)26H2O) was added to the solution and stirred for 40 min at 40 °C. This mass equated to a Cu concentration of 3.556 mM, which was equal to the concentration of MeAMN3S3sar initially added. In parallel, 10 mg of pristine MWCNTs (without the caged ligand MeAMN3S3sar) was also dispersed in 500 ml of deionized water in the same manner as previously described. Similarly 6.59 mg of Cu(ClO4)26H2O was added to this solution and stirred for 40 min at 40 °C. These solutions were then filtered under vacuum to isolate any free Cu2+ species from the MWCNTs. The precipitates were subsequently washed thoroughly with an excess of methanol and then dried for 12 h in an oven at 40 °C. XPS was then used to confirm the presence of the Cu(II) ions, which would quantify the association of caged ligands to the MWCNTs following the coupling reaction. 2.1.5. Radioactive labelling of MWCNT–MeAMN3S3sar Radioactive labelling of MWCNT–MeAMN3S3sar with the gamma-emitting radioactive isotope 57Co (T1/2 = 270 days) was explored. To achieve this 3 mg of the MWCNT–MeAMN3S3sar was ultrasonically dispersed in 13.35 ml of 0.1 M Na2PO4-buffered solution (pH 7.0) at an amplitude of 15 mA for 10 min. The MeAMN3S3sar were present at 2  104 M concentration, assuming 100% binding efficiency in Section 2.1.3. An equivalent molar amount of the 57Co to ligand reactant was added. The 57Co was added to the solution in the presence of a non-radioactive carrier species. In this instance, the carrier molecule was cobalt chloride (CoCl2). 51 ll of 2.94  103 M CoCl2 (in 0.1 M Na2PO4 buffered solution, pH 7.0) was introduced using a pipette into a Kimble tube along with 9 ll of 57Co to yield a 2.5  103 M 57/natCoCl2

R.W. Ormsby et al. / Acta Materialia 64 (2014) 54–61 57/nat

solution. CoCl2 solution (20 ll) was added to 20 ll of MWCNT–MeAMN3S3sar to determine non-specific binding of 57Co. For comparative purposes, 20 ll of pristine MWCNTs was exposed to the same solution as the control. The solutions were placed on a MaxQ benchtop Orbital Shaker (Thermo Scientific, UK) and allowed to incubate for 8 h. Following incubation, the Kimble tubes were centrifuged (at 14100g for 12 min) and the retentate collected to analyse the amount of bound (i.e. 57 Co–MWCNT–MeAMN3S3sar) and free 57Co species using instant thin-layer chromatography (ITLC). 2.2. Characterization 2.2.1. X-ray photoelectron spectroscopy The XPS system used was a Kratos Axis Ultra DLD spectrometer (Kratos, Manchester, UK) equipped with an aluminium X-ray source (monochromated) set at 5 mA and 15 kV (75 W). A wide energy survey scan (WESS) was performed from 1300–0 eV along with high-energy resolution regions of C (1s), O (1s), S (2s), N (1s) and Cu (2s). A WESS step size of 0.5 eV with a dwell time of 100 ms and five sweeps per scan was used. High-resolution scans were acquired at a step size of 0.025 eV with a dwell time of 200 ms and 10 sweeps per scan. Charge neutralisation was performed using the instrument’s charge neutraliser. All peak positions on the spectra were calibrated to the 285 eV (the peak at 285 eV is due to the presence of adventitious carbon contamination). Linear background removal was also performed on all scans. The XPS binding energies of elements of interest are outlined in Table 1. 2.2.2. Carbon-13 nuclear magnetic resonance 13 C NMR spectra (d) were recorded with a JEOL FX 60 spectrometer (JEOL Ltd., USA) using an external lock (D2O) and 1,4-dioxane as the internal reference (d +66.6 downfield from Me4Si). Visible spectra (e in M1 cm1) were recorded with a Cary 118C spectrophotometer (Olis Inc., USA), and infrared spectra were recorded with an IR spectrophotometer with KBr plates (Perkin Elmer, Model 457, USA). Rotatory dispersion and circular dichroism spectra were recorded with a P22 spectropolarimeter (Perkin Elmer, USA) and a Jasco Model ORD/ UV-5 with a Sproul Scientific SS 20 CD modification, respectively. All evaporations were conducted in a Buchi rotary evaporator under reduced pressure (20 mmHg) so that the temperature of the solution did not exceed

57

25 °C. 1,1,1-Tris(mercaptomethy1)ethane was prepared from 1,1,1-tris-(bromomethy1)ethane. 2.2.3. Instant thin layer chromatography and gamma signal counting ITLC was performed in duplicate. The 57Co-labelled MWCNT–MeAMN3S3sar solutions were spotted (5 ll) onto ITLC strips. The ITLC strips were then developed in 9:1 buffered solution (pH 4.5)/EtOH mobile phase for 20 min. Essentially, the MWCNTs will remain stationary (and any gamma activity/radioisotope associated with the MWCNTs) for a retention factor (Rf) of 0.2, and any of the unbound 57Co will ascend to the top of the ITLC strip (i.e. Rf = 1.0). The strips were subsequently dried and cut into 10 cm  1 cm pieces and placed in Kimble tubes. Each ITLC specimen was monitored for radioactivity for 10 s in a gamma counter (Wallace Wizard 1480, Perkin Elmer, USA). A representative gamma emission profile for 57Co is shown in Fig. 3. 2.2.4. Transmission electron microscopy of MWCNT coupled with MeAMN3S3sar As stated previously, it is essential that radioactively labelling MWCNT does not alter its native physical characteristics. Transmission electron microscopy (TEM) images were taken using a JEOL 1010 microscope fitted with a SIS Megaview Slowscan camera at 100 keV (JEOL Ltd., USA). Samples were prepared for TEM analysis by dip-coating 300 mesh formavar-coated Cu TEM grids into aqueous MWCNT dispersions and subsequently dried for 8 h at 22 °C. 3. Results 3.1. MeAMN3S3sar synthesis and coupling to MWCNTs XPS and 13C NMR were used to confirm successful synthesis of the MeAMN3S3sar. The presence of nitrogen (N), carbon (C) and sulphur (S) species, which are all present in the molecular structure of the bifunctional caged ligand were clearly observed from the XPS spectra of the MeAMN3S3sar (Fig. 4). There was a clear absence of any

Table 1 XPS binding energies of elements of interest. Element

XPS binding energy (eV)

C (1s) O (1s) S (2s) N (1s) Cu (2s)

284.8 530.8 163.9 and 228 398.5 935.4

Fig. 3. Major gamma-decay energies of

57

Co radioisotopes.

58

R.W. Ormsby et al. / Acta Materialia 64 (2014) 54–61

Following synthesis of the MeAMN3S3sar bifunctional caged ligand, the next step was to couple it with the COOH-MWCNT. XPS was again used to confirm the presence of a reactive intermediate provided by an NHS/EDC reaction (Fig. 5a). After confirming the presence of the reactive intermediate, attachment of the MeAMN3S3sar bifunctional caged ligand was attempted. XPS analysis was used to confirm the successful conjugation of the MeAMN3S3sar bifunctional caged ligand (through the presence of enhanced N and S peaks) with the MWCNT (Fig. 5b). 13C NMR (D2O, 1,4-dioxane): d = 34.67, 40.39, 49.23, 54.91 (ACH2A), 42.64, 56.91 (Cq), 28.34 ppm (CH3); 13C NMR (CDCl3, TMS): d = 35.28, 39.14, 50.07, 53.32 (ACH2A), 43.82, 61.34 (Cq), 25.03 ppm (CH3); 13C NMR (DMSO): d = 35.49, 39.44, 50.48, 54.07 (ACH2A), 43.61, 62.01 (Cq), 25.64 ppm (CH3); MS (EI, theoretical): m/z: 364.18; found: 365.23 [MeAMN3S3sar]+; elemental analysis calculated (%) for C15H32N4S30.5H2O: C48.22, H8.90, N14.99, S25.75%; found: C48.23, H8.97, N14.82, S25.95. 3.2. Transmission electron microscopy of MWCNTs coupled with MeAMN3S3sar Fig. 4. Representative XPS spectrum of successfully synthesized MeAMN3S3sar bifunctional cage ligand; all expected elements are present.

other atomic species in the XPS spectra, suggesting the caged ligand was isolated during synthesis.

Fig. 6 shows that the coupling reaction of the MeAMN3S3sar bifunctional caged ligand to the MWCNTs had no effect on its morphology or dimensions. The average diameter and length of the MWCNT before and after the coupling reaction were 10.23 ± 2.3 nm/35.14 ± 13.6 lm (Fig. 6a) and 10.79 ± 4.1 nm/37.09 ± 15.04 lm (Fig. 6b), respectively.

Fig. 5. (a) XPS spectrum of MWCNT-COOH after the addition of the reactive intermediate via reaction with NHS/EDC; this is highlighted by the N peak. The successful reaction is confirmed by the presence of the N peak. (b) XPS spectrum of MWCNT–MeAMN3S3sar; the successful reaction is confirmed by the presence of the N and S peaks.

R.W. Ormsby et al. / Acta Materialia 64 (2014) 54–61

59

Fig. 7. XPS spectrum of MWCNT–MeAMN3S3sar with copper (II) ions; the successful reaction is confirmed by the presence of the copper (II) peak.

Fig. 6. TEM images showing (a) MWCNTs before coupling with MeAMN3S3sar and (b) MWCNTs after coupling with MeAMN3S3sar.

3.3. Non-radioactive analogue labelling of MWCNTs Following the successful coupling of the MWCNTs with MeAMN3S3sar, non-radioactive labelling was completed with Cu2+ ions to confirm the presence of MeAMN3S3sar after the coupling reaction. XPS analysis demonstrated the Cu2+ ions were successfully associated with the MWCNT–MeAMN3S3sar complex (Fig. 7).

Fig. 8. Mean percentage (±SD) distribution of the radioactivity on the ITLC strip for free 57Co species.

3.4. Radioactive analogue labelling of MWCNT– MeAMN3S3sar via instant thin-layer chromatography The ITLC profiles of free (unbound) 57Co radioactive isotopes were obtained (Fig. 8). Various mobile phases were tested, and the optimum mobile phase used to separate free 57Co (Rf P 0.9) from radiolabelled MWCNT–MeAMN3S3sar (Rf P 0.1) was determined to be a 9:1 mixture of 0.1 M sodium acetate buffered solution, and 0.1 M ethylenediaminetetraacetic acid (EDTA) solution. EDTA was used as a complexing ligand to scavenge free 57Co species in solution. From Fig. 8. it can be observed that efficient mobilization of free 57Co species was achieved, as 92% of the 57Co species are located at

Fig. 9. Mean percentage distribution (±SD) of the radioactivity on the ITLC strip for the pristine MWCNT after 0 and 2 h.

60

R.W. Ormsby et al. / Acta Materialia 64 (2014) 54–61

Fig. 10. Mean percentage distribution (±SD) of radioactivity on the ITLC strip for the MWCNT–MeAMN3S3sar after 0 and 2 h.

an Rf of 1. Figs. 9 and 10 demonstrate that washing of the MWCNT effectively removed free and loosely bound radioisotopes. It can therefore be inferred that all remaining radioactivity was bound to the surface of the MWCNTs, and consequently remained at the origin (Rf = 0.1). Figs. 9 and 10 show the simultaneous gamma-activity profiles of 57Co-labelled pristine MWCNT and the MWCNT–MeAMN3S3sar conjugate. It can be observed in these figures that the addition of 57Co to the pristine MWCNTs showed negligible binding with 97% of the 57 Co moving with the mobile front (Rf = 1). In contrast, 95% of the 57Co was found to remain associated with the MWCNT–MeAMN3S3sar conjugate after 2 h (Fig. 10). Statistical analysis using a one-way analysis of variance (ANOVA) with Gabriel’s pairwise comparisons post hoc test analysis (PASW Statistics 18, USA) showed that this successful binding of the 57Co to the MWCNT–MeAMN3S3sar was significant (P 6 0.001) when compared to the binding efficiency of the pristine MWCNTs. 4. Discussion Within the current literature, there are few publications regarding the successful radioactive labelling of nanoparticles with bifunctional caged ligands. Musumeci et al. [36] used a series of bifunctional caged ligands (including the MeAMN3S3sar used in this study) to radiolabel a range (i.e. shapes and sizes) of titanium dioxide (TiO2) nanoparticles. These bifunctional cages were covalently attached to the surface of the particles via the use of a dopac derivative (dopamine) and then radiolabelled with a gamma-emitting radioisotope (also 57Co). The final radiolabelled nanoparticles were easily prepared and proved to be stable in solution. The application of a gamma-emitting isotope offers the potential for the radiolabelled particles to be tracked in vitro and in vivo. Hong et al. [37] recently showed that the covalent functionalization of radioactive metal ion-filled, single-walled carbon nanotubes (SWNCTs) can be used as radioprobes.

They sealed radioactive metal ions, including radioactive iodine (125I), inside SWCNTs to create high-density radio-emitting crystals. The surfaces of these filled and sealed SWCNTs were covalently modified with carbohydrates, as a means of improving biocompatibility. Intravenous administration of 125I-filled SWCNTs in mice was tracked in vivo using single-photon emission computed tomography. Specific tissue accumulation within the lung, coupled with high in vivo stability of the radio-emitting SWCNTs, prevented leakage of the radioactive metal ion to neighbouring organs (i.e. thyroid/stomach). This resulted in ultrasensitive imaging. Hong et al. [37] concluded that surface functionalization of 125I-filled SWCNTs offers a versatile means of modulating the biodistribution of these radio-emitting crystals in a manner determined by the capsule that delivers them, i.e. the SWCNTs. Deng et al. [38] developed a radiolabelling and tracing method for a biodistribution study of functionalized MWCNT in vivo. Taurine covalently functionalized MWCNTs (tau-MWCNTs) and Tween-80 wrapped MWCNTs (Tween-MWCNTs) were labelled with 125I. These radiolabelled MWCNTs where intravenously administered to the mice through a tail vein. The resulting distribution of 125I-tau-MWCNTs was consistent with that previously reported in a study by Deng et al. [25], who used carbon-14 (14C) as the radioactive isotope. Deng et al. [27] claimed that the method used for the 125I labelling was reliable and effective. However, this method was only suitable for radioisotopes with relatively short half-lives (60 days), which could limit the length of time that nanoparticles could be potentially tracked. It is also noteworthy that the methods used by Deng et al. [38,25] are limited in application as the chemical bonding of the radioactive isotope and the MWCNTs is not wholly stable. Therefore, the application of using a bifunctional caged ligand as a more stable means of radiolabelling MWCNT represents an attractive proposition. It is therefore postulated that the findings reported within this study will allow for the monitoring of MWCNTs when incorporated into PMMA bone cement in vitro, and potentially in vivo. The protocol developed and used within this study may be adopted for other radioactive isotopes and other polymer matrices. Furthermore, the methodologies presented within the current study may be adapted without significant modification for use in other nanoparticulate systems (e.g. much finer-scale SWCNTs and double-wall CNT powders) and allow for non-invasive imaging and monitoring of the fate of nanoparticles. Careful selection of the appropriate radioisotopes can provide the ability to track not only the nanoparticle but also its dissolution or breakdown by capitalising on their different biological pathways in animal models. Further investigations into the effect of these radiolabelled MWCNTs in biological systems will be undertaken, employing the tailored synthesis and labelling protocols

R.W. Ormsby et al. / Acta Materialia 64 (2014) 54–61

described herein. Non-invasive imaging and monitoring of radiolabelled MWCNTs is ongoing. 5. Conclusions This study investigated the potential of radiolabelling of MWCNTs using a bifunctional caged ligand. The bifunctional caged ligand, MeAMN3S3sar, was successfully synthesized and coupled to COOH-MWCNTs using an esterification reaction with the coupling agents NHS and EDC. The stability of the bifunctional caged ligand after the coupling reaction with MWCNTs was confirmed using XPS, and radioactive 57Co was successfully added to this ligand. ITLC demonstated that 95% of the 57Co was found to remain associated with the MWCNT–MeAMN3S3sar conjugate after 2 h; conversely, addition of 57Co to pristine MWCNTs showed negligible binding. Thus, the potential for radioactive labelling of MWCNT was successfully demonstrated. The use of the radioisotope 57Co for labelling MWCNTs may provide an alternative labelling methodology for following the biological fate of nanoparticles both in vitro and in vivo. Acknowledgments The authors thank Nanocyl SA, Belgium for kindly supplying the MWCNT powders for this study. This research was financially supported by the Department of Education and Learning, Northern Ireland. The authors also acknowledge the formal exchange agreement between the University of Queensland and Queen’s University of Belfast that facilitates ongoing academic and research collaborations. References [1] Woan K, Pyrgiotakis G, Sigmund W. Adv Mater 2009;21:2233. [2] Bianco A, Kostarelos K, Prato M. Expert Opin Drug Deliv 2008;5:331. [3] Bhirde AA, Patel V, Gavard J, Zhang G, Sousa AA, Masedunskas A, et al. ACS Nano 2009;3:307. [4] Ali-Boucettaa H, Al-Jamala KT, McCarthya D, Pratob M, Biancoc A, Kostarelos K. Chem Commun 2008:459. [5] Scheinberg DA, Villa CH, Escorcia FE, McDevitt MR. Nat Rev Clin Oncol 2010;7:266. [6] Nie S, Xing Y, Kim GJ, Simons JW. Annu Rev Biomed Eng 2007;9:257. [7] Andrews R, Jacques D, Qian DL, Rantell T. Acc Chem Res 2002;35:1008. [8] Andrews R, Weisenberger MC. Curr Opin Solid St M 2004;8:31.

61

[9] Bellare A, Fitz W, Gomoll A, Turell MB, Scott R, Thornhill T. The Orthopaedic J Harv Med School 2002;4:93. [10] Bonnet P, Sireude D, Garnier B, Chauvet O. App Phys Lett 2007;91:201910. [11] Chao G, Hailin Z, Yanping W, Watts PCP, Hao K, Xiaowen C, et al. Polymer 2006;47:113. [12] Ruan SL, Gao P, Yang XG, Yu TX. Polymer 2003;44:5643. [13] Ryan KP, Cadek M, Nicolosi V, Walker S, Ruether M, Fonseca A, et al. Synthetic Met 2006;156:332. [14] Sinha N, Yeow JTW. IEEE Trans Nanobiosci 2005;4:180. [15] Thostenson ET, Zhifeng R, Tsu-Wei C. Compos Sci Technol 2001;61:1899. [16] Webster TJ, Waid MC, McKenzie JL, Price RL, Ejiofor JU. Nanotechnology 2004;15:48. [17] Wu TM, Lin YW, Liao CS. Carbon 2005;43:734. [18] Xie XL, Mai YM, Zhou XP. Mater Sci Eng R 2005;49:89. [19] Zhang D, Kandadai MA, Cech J, Roth S, Curran SA. J Phys Chem B 2006;110:12910. [20] Zhang Y, Gu H, Iijima S. App Phys Lett 1998;73:3827. [21] Kostarelos K, Bianco A, Prato M. Nat Nanotech 2009;4: 627. [22] Lacerda L, Soundararajan A, Singh R, Pastorin G, Al-Jamal KT, Turton J, et al. Adv Mater 2008;20:225. [23] McDevitt MR, Chattopadhyay D, Jaggi JS, Finn RD, Zanzonico PB, Villa C, et al. PLoS ONE 2007;2:907. [24] Singh R, Pantarotto D, Lacerda L, Pastorin G, Klumpp C, Prato M, et al. Proc Natl Acad Sci 2006;103:3357. [25] Deng X, Jia G, Wang H, Sun H, Wang X, Yang S, et al. Carbon 2007;45:1419. [26] Wang H, Wang J, Deng X, Sun H, Shi Z, Gu Z, et al. Nanosci Nanotechnol 2004;4:1019. [27] Gahan LR, Hambley TW, Sargeson AM, Snow MR. Inorg Chem 1982;21:2699. [28] Bell CA, Bernhardt PV, Gahan LR, Martinez M, Monteiro MJ, Rodriguez C, et al. Chem Eur J 2010;16:3166. [29] Donlevy TM, Gahan LR, Hambley TW, Stranger R. Inorg Chem 1992;31:4376. [30] Ormsby RW, McNally T, Mitchell CA, Dunne NJ. J Mech Behav Biomed Mater 2010;3:136. [31] Ormsby RW, McNally T, Mitchell CA, Dunne NJ. J Mater Sci Mater Med 2010;21:2287. [32] Ormsby RW, McNally T, Mitchell CA, Halley P, Martin D, Dunne NJ. Carbon 2011;45:2893. [33] Ormsby RW, McNally T, O’Hare P, Burke G, Mitchell CA, Dunne NJ. Acta Biomater 2012;8:1201. [34] Nanocyl data sheet. Nanocyl, S.A., Belgium . [35] Jiang K, Schadler LS, Siegel RW, Zhang X, Zhang H, Terrones MJ. Mater Chem 2003;14:37. [36] Musumeci A. Ph.D. thesis. University of Queensland, Australia; 2010. [37] Hong SY, Tobias G, Al-Jamal KT, Ballesteros B, Ali-Boucetta H, Lozano-Perez S, et al. Nat Mater 2010;9:485. [38] Deng X, Yang S, Nie H, Wang H, Liu Y. Nanotechnology 2008;19:75101.