- Email: [email protected]
A novel biomimetic polymer as amphiphilic surfactant for soluble and biocompatible carbon nanotubes (CNTs)
Colloids and Surfaces B: Biointerfaces 67 (2008) 67–72
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
A novel biomimetic polymer as amphiphilic surfactant for soluble and biocompatible carbon nanotubes (CNTs) Fang-Ming Xu, Jian-Ping Xu ∗ , Jian Ji ∗ , Jia-Cong Shen Department of Polymer Science, Key Laboratory of Macromolecule Synthesis and Functionalization of Minster of Education, Zhejiang University, Hangzhou 310027, PR China
a r t i c l e
i n f o
Article history: Received 11 March 2008 Received in revised form 30 May 2008 Accepted 29 July 2008 Available online 6 August 2008 Keywords: Carbon nanotubes Phosphorylcholine Cholesterol Biomimetic Biocompatibility
a b s t r a c t Novel amphiphilic diblock copolymer, cholesterol-end-capped poly(2-methacryloyloxyethyl phosphorylcholine) (CPMPC), which has poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC) as hydrophilic segment and cholesterol as hydrophobic segment, was specially designed as amphiphilic surfactant to achieve water-soluble and biocompatible carbon nanotubes (CNTs). The pristine CNTs were facilely dispersed via non-covalently binding the zwitterionic phosphorylcholine-based amphiphile onto the surfaces of the CNTs. It is interesting to find that CPMPC shows better CNTs solubilizing ability compared with the surfactant of pyrene-end-capped poly(2-methacryloyloxyethyl phosphorylcholine) (PPMPC). The biocompatibility of the CPMPC stabilized CNTs was evaluated using cholesterol-end-capped poly(2(dimethylamino) ethyl methacrylate) (CPDMAEMA), cholesterol-end-capped poly(acrylic acid) (CPAA) and cholesterol-end-capped poly(ethylene oxide) (CPEG) as surfactants for CNTs as controls. While CPDMAEMA stabilized CNTs and CPAA stabilized CNTs showed obvious cytotoxicity, cytotoxicity of this novel zwitterionic phosphorylcholine-based amphiphile stabilized CNTs was not observed as indicated by cell culture. The biocompatible CNTs represent an excellent nano-object for potential biomedical applications. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Surfactant molecules composed of a hydrophobic tail and a hydrophilic segment have been shown to generate ordered supramolecular micellar structure in water when the surfactant concentration is higher than the critical micelle concentration (CMC) [1]. Potential applications of this nano-organized object including the microreservoir for the incorporation of lipophilic drug [2] and catalyst [3], surfactants for emulsions [4] and materials for chemical separation [5] have been well documented. In recent years, the interest in micelles as potential solubilizing and stabilizing agents for nano-objects including metallic nanoparticles [6], quantum dots (QD) [7], and magnetic nanoparticles [8] has grown enormously, due to the fact that they can not only stabilize hydrophobic nano-objects with otherwise limited water solubility and stability but also decrease their eventual high toxicity to healthy cells. For example, water-soluble, biocompatible semiconductor QD micelles and gold nanocrystal micelles in aqueous media have been obtained by incorporation of the
∗ Corresponding authors. Tel.: +86 571 87953729; fax: +86 571 87953729. E-mail addresses: jianping [email protected] (J.-P. Xu), [email protected] (J. Ji). 0927-7765/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2008.07.016
hydrophobic semiconductor QD [9] or gold nanocrystal [10] into the hydrophobic interiors of surfactant micelles. Also, multifunctional polymeric micelles with controlled drug delivery and efficient magnetic resonance imaging contrast characteristics were developed by simultaneously incorporating chemotherapeutic agent doxorubicin and superparamagnetic iron oxide nanoparticles inside the hydrophobic core of micelles [11]. Additionally, CNTs represent one of the most widely used nanoobjects for nanomedicine. Potential applications of CNTs have currently been explored in many different areas including sensitive diagnostic assays, thermal therapy [12], molecular recognition, drug and gene delivery, as well as sensor science [13,14]. While CNTs are usually insoluble in water due to the large intertube attraction energy, most biological or biomedical applications require that the CNTs readily dissolve in aqueous media. Among the various schemes to overcome the hydrophobicity of CNTs, solubilizing CNTs with amphiphilic surfactants was suggested to minimize alterations to the CNTs properties and importantly, the electronic structure of the dispersed tubes and their physical properties are believed to be conserved during the stabilizing process [15–17]. In this case, the hydrophobic and the hydrophilic moieties of the attached amphiphilic molecules interact with the tube surfaces and the solution, respectively. For the biomedical applications of
68
F.-M. Xu et al. / Colloids and Surfaces B: Biointerfaces 67 (2008) 67–72
Table 1 Preparation parameters and properties of the amphiphilic surfactants
CPMPC CPEG CPDMAEMA CP(tBA)c CPAA PPMPC a b c
Initiator (g)
Monomer (g)
Solvent (ml)
Mna
CMCb (mg/ml)
CholBr 0.1588 – CholBr 0.1697 CholBr 0.3756 – PyBr 0.05470
MPC 2.638 – DMAEMA 4.852 tBA 8.969 – MPC 2.120
Ethanol/2-propanol 25/25 – Anisole 20 Toluene 40 – Ethanol/2-propanol 10/10
7917 2300 5323 6687 3706 9478
23.18 × 10−3 3.240 × 10−3 43.45 × 10−3 – 78.16 × 10−3 6.026 × 10−5
Molecular weight from the 1 H NMR results. Critical micelle concentration calculated from the excitation spectra of pyrene as a function of polymer concentrations in water. The ligand for the synthesis was N,N,N ,N ,N -pentamethyldiethylenetriamine.
the CNT, the amphiphilic surfactant-based dispersing route might be, however, problematic due to the potential toxicity of the surfactant [18–20]. We conceived that the design and nature of both the hydrophobic part and the hydrophilic part of the amphiphilic molecules are extremely important for the solubility and biocompatibility of the dispersed CNTs. In our previous studies, novel biomimetic block copolymers, cholesterol-end-capped poly(2-methacryloyloxyethyl phosphorylcholine) (CPMPC), which have poly(MPC) as hydrophilic segment and cholesterol as hydrophobic segment were specially designed and used as drug delivery system [21,22]. Micelles based on these biomimetic polymers were found to show better biocompatibility compared with the micelles prepared from the cholesterol-endcapped poly(ethylene oxide), which showed obvious toxicity. Furthermore, anti-cancer drug adriamycin was successfully incorporated into the hydrophobic cholesterol core of the micelles. Here we intent to study the applicability of the novel biomimetic block copolymer to solubilize and stabilize the CNTs, to investigate the effect of polymer concentration on the formation of dispersions, and to examine the effect of the nature of the surfactants on the solubility and biocompatibility of dispersed CNTs.
was purified by washing with glacial acetic acid, followed by absolute ethanol and ethyl ether, and then dried under vacuum. 2,2 -Bipyridine (bpy) (AR, Hangzhou Chemical Reagent Factory) and 1-pyrenemethanol (Aldrich Co) were used as received. CNTs purchased from Shenzhen Nanotech Port Co. Ltd. (multi-walled carbon nanotubes, diameter 10–20 nm, purity >95%) were used as received. Other reagents were purified by conventional methods. All the water used in this work is distilled and deioned (DDI) water. The conversion of cholesterol or 1-pyrenemethanol into the ATRP initiator (CholBr or PyBr) by reaction with 2-bromoisobutyryl bromide was performed as follows. In a 250 ml three-neck flask, cholesterol (4.988 g, 12.9 mmol) or 1-pyrenemethanol (3.0 g, 12.9 mmol) and triethylamine (5.4 ml, 38.7 mmol) were dissolved in dry CH2 Cl2 (40 ml). Then 2-bromoisobutyryl bromide (4.8 ml, 38.7 mmol) dissolved in dry CH2 Cl2 (20 ml) was added dropwise into the flask under stirring at 0 ◦ C. Then the reaction solution was allowed to warm to room temperature, and the reaction mixture was stirred for another 12 h. The reaction mixture was then filtered, and the CH2 Cl2 was removed using the rotary evaporator. The resulting yellow crude product was precipitated into cold ethanol. The precipitation was filtered and dried under vacuum [22,26].
2. Experimental 2.2. Preparation of the amphiphilic polymers 2.1. Materials The 2-methacryloyloxyethyl phosphorylcholine (MPC) monomer and poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC) were synthesized according to the literatures [22–25]. 2(Dimethylamino) ethyl methacrylate (DMAEMA), t-butyl acrylate (tBA) and 2-bromoisobutyryl bromide were purchased from Aldrich and used as received. Cholesterol-end-capped poly(ethylene oxide) (CPEG) was purchased from Nihon Emulsion Co. Ltd. and used as received. The CuBr (AR, Shanghai No. 1 Chemical Reagent Factory)
Typical protocols for the polymerization of MPC, t-butyl acrylate and DMAEMA using the CholBr as initiator or the polymerization of MPC using PyBr as initiator were carried out as follows. The monomers and corresponding initiators were dissolved in solvents. The flask, containing a stir bar, was degassed by three freezepump-thaw cycles with argon. The Cu(I)Br catalyst and bpy ligand were added to the stirred solution. The flask was then placed into an oil bath and thermostated at reaction temperature. After 20 h, the reaction mixture was cooled down and the flask was opened.
Fig. 1. TEM images for (A) the pristine CNTs, (B) the CPMPC coated CNTs, and (C) chemically oxidized CNTs-COOH.
F.-M. Xu et al. / Colloids and Surfaces B: Biointerfaces 67 (2008) 67–72
The resulting polymer mixtures were purified and characterized according to literatures [22,26–29]. The cholesterol-end-capped poly(t-butyl acrylate) was subjected to acidic hydrolysis in 1,4-dioxane (3 wt.% concentration solution, with the addition of a five-fold excess HCl 1 M, at 80 ◦ C for more than 12 h to afford cholesterol-end-capped poly(acrylic acid) [30]. The preparation of amphiphilic polymers were followed by FTIR and NMR and the preparation parameters and properties of the amphiphilic polymers are summarized in Table 1. 2.3. Dispersion of CNTs Cholesterol-end-capped poly(2-methacryloyloxyethyl phosphorylcholine) aqueous solution of different concentrations were prepared by dissolving weighted amount of polymer powder in DDI water and diluted to different concentrations by pipetted volume of DDI water. 3 ml polymer solution of each concentration was added into a vial. 20 mg of CNTs were added to each sample and the mixture was sonicated for 30 s in the low-power ultrasonic cleaner KQ2200 (Kunshan Ultrasonic Instruments Co. Ltd., 100 W, 40 kHz). The mixture was left for a certain time (1.5 h for UV absorbance test and 24 h for cell culture) without any disturbance to precipitate, then the upper 2 ml homogeneous solution of each vial was carefully transferred into another vial and preserved for other tests. To prepare the CNT suspensions using other type amphiphilic polymers as surfactants, the same procedure mentioned above was used. For stability assay, 3 ml aqueous solution of different polymer solutions (2 mg/ml) were added into the vials together with 15 mg CNTs. Then the samples were prepared by hand shaking. The transmittance at 600 nm of the CNTs suspension was recorded by UV spectrometer for 6 h. In cell culture experiments, CNTs were autoclaved under 121 ◦ C for 20 min and all the polymer solutions were sterilized by ultrafiltration (Millipore, 220 nm). 2.4. Instruments and measurements The Raman characterization was carried out on a Jobin Yvon microRaman system (Ramanor U1000, Instruments SA, USA) using a spectra-physics Ar ion laser at an excitation wavelength of 514.5 nm (2.41 eV). Their film samples were formed by drop coating on a clean silicon wafer with CNT suspension and drying at 80 ◦ C. The backscattered data were analyzed using a double-gating spectrometer and collected using a Hamamatsu photomultiplier (R 943-02, Hamamatsu, USA). For every Raman spectrum taken, the position of the peaks was verified by calibrating the spectral positions in respect to silicon substrate peak seen at 521 cm−1 [31]. Transmission Electron Microscopy (TEM) analysis was performed on a JEM-1200EX TEM operating at 200 kV in bright field mode. CNTs samples were prepared by placing a drop of the solutions (the concentration of CNTs is 1 mg/ml) on a 400-mesh carboncoated copper grid and air-drying the grid at 25 ◦ C. UV–vis spectra were carried out with a UV–vis Shimadzu UV-2055 spectrometer using quartz cuvettes. Spectra were collected within a range of 200–800 nm and the absorbance value at 600 nm was taken to compare CNTs solubility, since there was no absorbance value of each kind of polymer solution at 600 nm including PPMPC. 2.5. Cell viability Human umbilical vein endothelial cell line was chosen for cell culture studies. The cells were seeded into a 96 well cell culture plate at a density of 8000 cells per well for 24 h, then the growth medium was removed and replaced with the medium containing surfactant coated CNTs. The CNTs concentration in different sam-
69
ples was adjusted to the same level by UV absorbance at 600 nm before they were added into the culture medium, the concentration of CNTs in the culture medium of each well was 30 g/ml. The medium used was RPMI-1640 medium containing 10% heat inactivated fetal calf serum, 100 unit/ml penicillin and 0.1 mg/ml streptomycin. The cells were incubated at 37 ◦ C with 100% humidity in a 5% CO2 atmosphere for 96 h. To determine cell viability, 20 l MTT solution (5 mg/ml in PBS) was added to each sample, and incubated at 37 ◦ C for 4 h for MTT formazan formation. After the medium and MTT were replaced by 150 l DMSO, the samples were incubated at 37 ◦ C for 20 min to dissolve the MTT formazan. In order to avoid the effect of CNTs on the absorbance, all the DMSO solution in each well was pipetted into the same well of another new TCPS cell culture plate before the absorbance values were measured by using microplate reader (BIO-RAD, model 550) at wavelength 570 nm. Five replicates were read for each sample, the mean value of the five was used as the final result. Cells which were incubated in culture medium without any CNTs were used as the control. The cell viability was calculated as the ratio of the absorbency value of samples incubated with polymer coated CNTs to that of the control [21,32]. 3. Results and discussion CNTs have an exceptionally strong tendency to aggregate and it is difficult to suspend them in aqueous solutions. Surfactants showed the ability to encapsulate and separate CNTs and achieve CNTs colloidal stability. Traditional surfactants, such as sodium dodecyl sulfate, provide stable CNTs dispersions only at very high concentrations [15]. Furthermore, the detergent effect of the traditional surfactants induced potential toxicity. It is suggested that significant improvement of the dispersion efficiency can be obtained with polymeric agents [33,34]. While, there are a lot of researches published on suspending CNTs by the amphiphilic polymers and especially on the binding mode of the polymer onto the CNTs, the reports on the potential toxicity of the non-covalently dispersed CNTs via amphiphilic surfactant-based route are, however, scarce. The trial to resolve the toxicity of the amphiphilic surfactant dispersed CNTs is even much fewer. Furthermore, for the biomedical applications, the MWNTs with lower toxicity and larger loading capacity seem to be a better choice than SWNTs. Here we intent to study the ability of a novel biomimetic block copolymer, cholesterol-end-capped poly(2-methacryloyloxyethyl phosphorylcholine), for the solubilizing and stabilizing the CNTs (MWNTs). We showed in our previous studies that this biomimetic block copolymers combine both advantages of liposome and polymeric micelles and resulted in biocompatible and biological stable drug delivery systems [21]. Here, the cholesterol hydrophobic segment is believed to provide strong interactions with the CNTs surface and the poly(2methacryloyloxyethyl phosphorylcholine) hydrophilic segment is suggested to provide stability and biocompatibility of the system. 3.1. Dispersion of CNTs in aqueous medium by CPMPC The effectiveness of using CPMPC to disperse the CNTs was first followed by TEM as shown in Fig. 1. Although the pristine CNTs formed large bundles, Fig. 1(A), the addition of CPMPC into the pristine CNTs followed by 30 s sonication produced well dispersed CNTs, Fig. 1(B). This non-covalent dispersion process did not change the diameter of the MWNTs as compared with the pristine MWNTs, which showed an average diameter of 20 nm. It is notably to observe that while the pristine CNTs of 1–2 m were cut into short pieces of 100–500 nm during the carboxylation (Fig. 1(C)), which is a mandatory process in the chemically
70
F.-M. Xu et al. / Colloids and Surfaces B: Biointerfaces 67 (2008) 67–72
Fig. 2. UV/vis absorption spectrum of dispersed CNTs in CPMPC aqueous solution for different relative concentration (A, C0 ; B, (5/6)C0 ; C, (2/3)C0 ; D, (1/2)C0 ; E, (1/3)C0 ; F, (1/6)C0 ). C0 denote the saturated concentration of dispersed MWNTs in 1 mg/ml CPMPC aqueous solution. The inset is Lambert–Beer’s plot for the absorption at 600 nm.
modification of CNTs, the non-covalent dispersion process did not change the length of CNTs. The conservation of the diameter and the length suggests that the non-covalent dispersion process did not influence the physical structures of the CNTs. UV/vis absorption spectra of the CNTs dissolved in 1 mg/ml CPMPC aqueous solution at different concentrations were collected, Fig. 2. The intensities of absorption spectra had a linear relation with the concentration of the CNTs [35], Fig. 2, inset, and followed Lambert–Beer’s law, indicating the homogenous dispersion of the CNTs in solutions. The solubility of CNTs in 1 mg/ml CPMPC aqueous solution was calculated to be 1.705 mg/ml. Raman spectra provide an excellent method to examine the adherence of the amphiphilic polymers onto the surface of CNTs. The band in the 1500–1600 cm−1 region is the so-called G-band. An evident shift in the spectral position of the G-band before and after the modification with polymer can be noticed, where the G-band shifts by 10 cm−1 from 1572 to 1582 cm−1 (Fig. 3A). This change can be explained by considering wrapping of the polymers around CNTs. The hydrophobic and van der Waals attraction forces between the polymer and the graphite sheet increase the energy necessary for vibrations to occur, which resulted in the higher frequency of Raman peaks. The disorder peak, known as the D-band, can be found in the 1300–1400 cm−1 region. The position of the Dband typically changes when covalent modification of the graphene sheet occurs [31]. It was noticed that some insignificant changes in
Fig. 3. Raman spectra of (A) the pristine CNTs and (B) the CPMPC coated CNTs.
Fig. 4. (a) Comparison of CNTs solubility in DDI water with different concentrations of (A) CPMPC, (B) PMPC, and (C) PPMPC. (b) Comparison of CNTs solubility in DDI water with different concentrations of (A) CPMPC, (B) CPEG, (C) CPAA, and (D) CPDMAEMA.
the position of this D-band appeared, which might be attributed to the formation of polymer coating on carbon particles instead of CNTs. The Raman spectra data indicate that the polymers attach onto the CNTs non-covalently and the integrity of physical structure of the CNTs preserved. 3.2. The effect of the nature of the surfactants on the solubility of dispersed CNTs The role of the cholesterol hydrophobic segment of the CPMPC in stabilizing the CNTs was inspected. This was carried out by checking the solubility of the CNTs against concentrations of the polymers by using PPMPC and PMPC as controls as shown in Fig. 4(a). The increase of the concentration of CPMPC resulted in the increased solubility of CNTs. The solubility of CNTs reached a plateau when the concentration of CPMPC exceeded 3 mg/ml and the saturated concentration of CNTs reached 3.307 mg/ml. The saturated concentrations of CNTs reached 2.237 mg/ml in 3 mg/ml PMPC aqueous solution and 1.906 mg/ml in 3 mg/ml PPMPC aqueous solution, respectively, were much lower than that when the CPMPC was used. This result indicated that the presence of cholesterol hydrophobic segment is crucial for the solubilising ability of the CPMPC. While pyrene-based surfactants are believed to disperse the CNTs through – stacking between the pyrene groups and CNTs, the CNTs can also be dispersed through van der Waals force between the hydrophobic part and the CNTs. Several traditional surfactants [15], for example Phospholipids [17], have been showed to efficiently
F.-M. Xu et al. / Colloids and Surfaces B: Biointerfaces 67 (2008) 67–72
71
Fig. 7. The cell viability of cells cultured with CPMPC, PPMPC, CPEG, CPAA and CPDMAEMA-coated CNTs after 96 h incubation time. In control experiment, the cells were cultured in culture medium without CNTs. Fig. 5. Stability comparison of CNTs stabilized by (A) CPMPC, (B) PMPC and (C) PPMPC.
disperse and stabilize the CNTs. It was important to note here that the CPMPC with cholesterol hydrophobic segment showed better CNT-solubilising ability than PPMPC with pyrene hydrophobic segment, which is a traditional segment for CNT dispersing. It is well accepted that the attachment of hydrophilic and water-soluble groups onto the surface of CNTs is crucial for the biomedical applications. For the biocompatibility and colloidal stability reasons, a wide range of polymer-functionalized CNTs have been synthesized, and in particular, PAA [36,37], PDMAEMA [34], and PEG-functionalized CNTs [38] are the most widely studied. Here, we are interested in using the phosphorylcholine-type surfactants to generate stable and biocompatible CNTs. The effect of the hydrophilic segment composition of the surfactants on stabilizing the CNTs was inspected by checking the solubility of the CNTs as shown in Fig. 4(b). The solubility of CNTs reached a value of 1.705 mg/ml in 1 mg/ml CPMPC aqueous solution. When 1 mg/ml of CPAA, CPDMAEMA or CPEG aqueous solution was used, the solubility of CNTs was, however, only 0.7875 mg/ml, 0.5626 mg/ml and 1.468 mg/ml, respectively. Moreover, it was interesting to find that the solubility of CNTs decreased to 1.238 mg/ml, 0.3952 mg/ml, and 0.4669 mg/ml in 1 mg/ml CPMPC, CPDMAEMA, CPAA when PBS was used as buffer solution. The effect of the hydrophobic segment of the polymers on the stability of the CNTs was followed (Fig. 5). While the transmittance of the PMPC dispersed CNTs and the PPMPC dispersed CNTs increased rapidly within 6 h to 5.8% and 3.1%, respectively (Fig. 5, curve (B) and (C)), the transmittance of CPMPC did not show obvious increase (Fig. 5, curve (A)). This result indicated that the hydrophobic segment of the polymers showed great influence on the stability of the CNTs and the cholesterol-end-capped CPMPC showed excellent stabilizing ability for CNTs.
Furthermore, the stabilizing ability of different surfactants against pH changes was followed, Fig. 6. Although the CPDMAEMA stabilized CNTs was stable at acidic pH, Fig. 6(C), the change of pH to alkali value resulted the precipitation of the CNTs, Fig. 6(I). Similarly, the change of pH from alkali value to acidic value resulted in the transformation from dispersion state (Fig. 6(G)) to precipitation state of the CPAA stabilized CNTs (Fig. 6(A)). It was interesting to find that the CPMPC stabilized CNTs showed dispersion stability at both alkali (Fig. 6(H)) and acidic values (Fig. 6(B)). The solubility data indicate that the phosphorylcholine-type surfactants showed excellent solubilising ability for CNTs. 3.3. In vitro cytotoxicity measurements It is suggested that to achieve water-soluble CNTs for the biomedical application, physical adsorption of surfactants should be avoided since the surfactants used widely for solubilization of CNTs are always toxic by themselves [18–20]. A novel biomimetic surfactant with excellent biocompatibility, cholesterol-end-capped poly(2-methacryloyloxyethyl phosphorylcholine), was specially designed and synthesized. We were therefore interested in evaluating the biocompatibility of the CNTs solubilised by CPMPC. Human umbilical vein endothelial cells were cultured with CPMPC, CPAA, CPDMAEMA, CPEG or PPMPC-coated CNTs for 4 days. In control experiments, the cells were cultured with media alone. Cell viability was evaluated with MTT assay as shown in Fig. 7. The viabilities of the cells cultured with the CPEG, CPAA and CPDMAEMA coated CNTs reached only 8%, 70% and 12%, respectively. Presumably, the CPEG, CPAA or CPDMAEMA coated CNTs either inhibited cell growth or induced cell death. By contrast, cells cultured with CPMPC coated CNTs were able to growth very well during the course of the experiment and cell viability as high
Fig. 6. Dispersion stability tests of CNTs in DDI water with different surfactants at varied pH values: (A) CPAA at pH 2, (B) CPMPC at pH 2, (C) CPDMAEMA at pH 2, (D) CPAA at pH 7, (E) CPMPC at pH 7, (F) CPDMAEMA at pH 7, (G) CPAA at pH 13, (H) CPMPC at pH 13, and (I) CPDMAEMA at pH 13.
72
F.-M. Xu et al. / Colloids and Surfaces B: Biointerfaces 67 (2008) 67–72
as 92% was achieved. Similar results were obtained with PPMPC coated CNTs and cell viability of 95% was achieved. Thus, the phosphorylcholine-polymer coating renders the CNTs non-toxic. 4. Conclusions In conclusion, the present study has demonstrated the success of using the novel biomimetic copolymer CPMPC as amphiphilic surfactant to achieve water-soluble and biocompatible CNTs. The use of cholesterol group as hydrophobic segment is crucial for the solubilising ability and solubility of pristine CNTs as high as 3.307 mg/ml was achieved. The zwitterionic PMPC design afforded better solubilising ability than the negative PAA or positive PDMAEMA. We demonstrated here also that excellent biocompatibility of the zwitterionic engineered CNTs by checking the cell viability. The biocompatible CPMPC coated CNTs represent an excellent nano-object for the biomedical applications in nanomedicine. Acknowledgements This research was financially supported by Natural Science Foundation of China (NSFC-20774082, 50703036), 863 National High-Tech R&D Program (2006AA03Z329, 2006AA03Z444), Program for New Century Excellent Talents in University (NCET-050527) and High Technology Research and Development Program of Zhejiang Province (2007C24008). References [1] L. Leibler, H. Orland, J.C. Wheeler, J. Chem. Phys. 79 (1983) 3550. [2] P. Blasi, S. Giovagnoli, A. Schoubben, M. Ricci, C. Rossi, Adv. Drug Deliv. Rev. 59 (2007) 454. [3] A. Takasu, A. Takemoto, T. Hirabayashi, Biomacromolecules 7 (2006) 6. [4] D. Cochin, A. Laschewsky, F. Nallet, Macromolecules 30 (1997) 2278. [5] R.O. Dunn Jr., J.F. Scamehorn, S.D. Christian, Sep. Sci. Technol. 20 (1985) 257. [6] T. Ishii, H. Otsuka, K. Kataoka, Y. Nagasaki, Langmuir 20 (2004) 561. [7] B.A. Korgel, H.G. Monbouquette, Langmuir 16 (2000) 3588. [8] M.-S. Martina, J.-P. Fortin, C. Ménager, O. Clément, G. Barratt, C. GrabielleMadelmont, F. Gazeau, V. Cabuil, S. Lesieur, J. Am. Chem. Soc. 127 (2005) 10676.
[9] H.Y. Fan, E.W. Leve, C. Scullin, J. Gabaldon, D. Tallant, S. Bunge, T. Boyle, M.C. Wilson, C.J. Brinker, Nano Lett. 5 (2005) 645. [10] P. He, M.W. Urban, Biomacromolecules 6 (2005) 1224. [11] N. Nasongkla, E. Bey, J. Ren, H. Ai, C. Khemtong, J.S. Guthi, S.-F. Chin, A.D. Sherry, D.A. Boothman, J.M. Gao, Nano Lett. 6 (2006) 2427. [12] J.-W. Kim, E.V. Shashkov, E.I. Galanzha, N. Kotagiri, V.P. Zharov, Lasers Surg. Med. 39 (2007) 622. [13] L. Lacerda, A. Bianco, M. Prato, K. Kostarelos, Adv. Drug Deliv. Rev. 58 (2006) 1460. [14] Y. Lin, S. Taylor, H.P. Li, K.A.S. Fernando, L.W. Qu, W. Wang, L.R. Gu, B. Zhou, Y.P. Sun, J. Mater. Chem. 14 (2004) 527. [15] M.F. Islam, E. Rojas, D.M. Bergey, A.T. Johnson, A.G. Yodh, Nano Lett. 3 (2003) 269. [16] N. Nakashima, Y. Tomonari, H. Murakamiy, Chem. Lett. 31 (2002) 638. [17] Y. Wu, J.S. Hudson, Q. Lu, J.M. Moore, A.S. Mount, A.M. Rao, E. Alexov, P.C. Ke, J. Phys. Chem. B 110 (2006) 2475. [18] H. Isobe, T. Tanaka, R. Maeda, E. Noiri, N. Solin, M. Yudasaka, S. Iijima, E. Nakamura, Angew. Chem. Int. Ed. 45 (2006) 6676. [19] M. Zhang, M. Yudasaka, J. Miyawaki, J. Fan, S. Iijima, J. Phys. Chem. B 109 (2005) 22201. [20] M.J. O’Connell, S.M. Bachilo, C.B. Huffman, V.C. Moore, M.S. Strano, E.H. Haroz, K.L. Rialon, P.J. Boul, W.H. Noon, C. Kittrell, J.P. Ma, R.H. Hauge, R.B. Weisman, R.E. Smalley, Science 297 (2002) 593. [21] J.P. Xu, J. Ji, W.D. Chen, J.C. Shen, J. Control. Release 107 (2005) 502. [22] J.P. Xu, J. Ji, W.D. Chen, J.C. Shen, Macromol. Biosci. 5 (2005) 164. [23] K. Ishihara, T. Ueda, N. Nakabayashi, Polym. J. 22 (1990) 355. [24] H. Kitano, K. Sudo, K. Ichikawa, M. Ide, K. Ishihara, J. Phys. Chem. B 104 (2000) 11425. [25] J.P. Xu, J. Ji, W.D. Chen, D.Z. Fan, Y.F. Sun, J.C. Shen, Eur. Polym. J. 40 (2004) 291. [26] M.F. Wang, T.E. Dykstra, X.D. Lou, M.R. Salvador, G.D. Scholes, M.A. Winnik, Angew. Chem. 118 (2006) 2279. [27] I.Y. Ma, E.J. Lobb, N.C. Billingham, S.P. Armes, A.L. Lewis, A.W. Lloyd, J. Salvage, Macromolecules 35 (2002) 9306. [28] X. Zhang, J.H. Xia, K. Matyjaszewski, Macromolecules 31 (1998) 5167. [29] J.Y. Huang, T. Pintauer, K. Matyjaszewski, J. Polym. Sci. Pol. Chem. 42 (2004) 3285. [30] S.A. Angelopoulos, C. Tsitsilianis, Macromol. Chem. Phys. 207 (2006) 2188. [31] V.A. Sinani, M.K. Gheith, A.A. Yaroslavov, A.A. Rakhnyanskaya, K. Sun, A.A. Mamedov, J.P. Wicksted, N.A. Kotov, J. Am. Chem. Soc. 127 (2005) 3463. [32] L.Z. Gao, L. Nie, T.H. Wang, Y.J. Qin, Z.X. Guo, D.L. Yang, X.Y. Yan, Chem. Bio. Chem. 7 (2006) 239. [33] D. Wang, W.X. Ji, Z.C. Li, L.W. Chen, J. Am. Chem. Soc. 128 (2006) 6556. [34] P. Petrov, F. Stassin, C. Pagnoullea, R. Jérôme, Chem. Commun. 3 (2003) 2904. [35] Y.P. Sun, W. Huang, Y. Lin, K. Fu, A. Kitaygorodskiy, L.A. Riddle, Y.J. Yu, D.L. Carroll, Chem. Mater. 13 (2001) 2864. [36] Y. Kang, T.A. Taton, J. Am. Chem. Soc. 125 (2003) 5650. [37] A.H. Liu, I. Honma, M. Ichihara, H.S. Zhou, Nanotechnology 17 (2006) 2845. [38] R.J. Chen, S. Bangsaruntip, K.A. Drouvalakis, N.W.S. Kam, M. Shim, Y.M. Li, W. Kim, P.J. Utz, H.J. Dai, Proc. Natl. Acad. Sci. U.S.A. 100 (2003) 4984.
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
A novel biomimetic polymer as amphiphilic surfactant for soluble and biocompatible carbon nanotubes (CNTs) Fang-Ming Xu, Jian-Ping Xu ∗ , Jian Ji ∗ , Jia-Cong Shen Department of Polymer Science, Key Laboratory of Macromolecule Synthesis and Functionalization of Minster of Education, Zhejiang University, Hangzhou 310027, PR China
a r t i c l e
i n f o
Article history: Received 11 March 2008 Received in revised form 30 May 2008 Accepted 29 July 2008 Available online 6 August 2008 Keywords: Carbon nanotubes Phosphorylcholine Cholesterol Biomimetic Biocompatibility
a b s t r a c t Novel amphiphilic diblock copolymer, cholesterol-end-capped poly(2-methacryloyloxyethyl phosphorylcholine) (CPMPC), which has poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC) as hydrophilic segment and cholesterol as hydrophobic segment, was specially designed as amphiphilic surfactant to achieve water-soluble and biocompatible carbon nanotubes (CNTs). The pristine CNTs were facilely dispersed via non-covalently binding the zwitterionic phosphorylcholine-based amphiphile onto the surfaces of the CNTs. It is interesting to find that CPMPC shows better CNTs solubilizing ability compared with the surfactant of pyrene-end-capped poly(2-methacryloyloxyethyl phosphorylcholine) (PPMPC). The biocompatibility of the CPMPC stabilized CNTs was evaluated using cholesterol-end-capped poly(2(dimethylamino) ethyl methacrylate) (CPDMAEMA), cholesterol-end-capped poly(acrylic acid) (CPAA) and cholesterol-end-capped poly(ethylene oxide) (CPEG) as surfactants for CNTs as controls. While CPDMAEMA stabilized CNTs and CPAA stabilized CNTs showed obvious cytotoxicity, cytotoxicity of this novel zwitterionic phosphorylcholine-based amphiphile stabilized CNTs was not observed as indicated by cell culture. The biocompatible CNTs represent an excellent nano-object for potential biomedical applications. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Surfactant molecules composed of a hydrophobic tail and a hydrophilic segment have been shown to generate ordered supramolecular micellar structure in water when the surfactant concentration is higher than the critical micelle concentration (CMC) [1]. Potential applications of this nano-organized object including the microreservoir for the incorporation of lipophilic drug [2] and catalyst [3], surfactants for emulsions [4] and materials for chemical separation [5] have been well documented. In recent years, the interest in micelles as potential solubilizing and stabilizing agents for nano-objects including metallic nanoparticles [6], quantum dots (QD) [7], and magnetic nanoparticles [8] has grown enormously, due to the fact that they can not only stabilize hydrophobic nano-objects with otherwise limited water solubility and stability but also decrease their eventual high toxicity to healthy cells. For example, water-soluble, biocompatible semiconductor QD micelles and gold nanocrystal micelles in aqueous media have been obtained by incorporation of the
∗ Corresponding authors. Tel.: +86 571 87953729; fax: +86 571 87953729. E-mail addresses: jianping [email protected] (J.-P. Xu), [email protected] (J. Ji). 0927-7765/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2008.07.016
hydrophobic semiconductor QD [9] or gold nanocrystal [10] into the hydrophobic interiors of surfactant micelles. Also, multifunctional polymeric micelles with controlled drug delivery and efficient magnetic resonance imaging contrast characteristics were developed by simultaneously incorporating chemotherapeutic agent doxorubicin and superparamagnetic iron oxide nanoparticles inside the hydrophobic core of micelles [11]. Additionally, CNTs represent one of the most widely used nanoobjects for nanomedicine. Potential applications of CNTs have currently been explored in many different areas including sensitive diagnostic assays, thermal therapy [12], molecular recognition, drug and gene delivery, as well as sensor science [13,14]. While CNTs are usually insoluble in water due to the large intertube attraction energy, most biological or biomedical applications require that the CNTs readily dissolve in aqueous media. Among the various schemes to overcome the hydrophobicity of CNTs, solubilizing CNTs with amphiphilic surfactants was suggested to minimize alterations to the CNTs properties and importantly, the electronic structure of the dispersed tubes and their physical properties are believed to be conserved during the stabilizing process [15–17]. In this case, the hydrophobic and the hydrophilic moieties of the attached amphiphilic molecules interact with the tube surfaces and the solution, respectively. For the biomedical applications of
68
F.-M. Xu et al. / Colloids and Surfaces B: Biointerfaces 67 (2008) 67–72
Table 1 Preparation parameters and properties of the amphiphilic surfactants
CPMPC CPEG CPDMAEMA CP(tBA)c CPAA PPMPC a b c
Initiator (g)
Monomer (g)
Solvent (ml)
Mna
CMCb (mg/ml)
CholBr 0.1588 – CholBr 0.1697 CholBr 0.3756 – PyBr 0.05470
MPC 2.638 – DMAEMA 4.852 tBA 8.969 – MPC 2.120
Ethanol/2-propanol 25/25 – Anisole 20 Toluene 40 – Ethanol/2-propanol 10/10
7917 2300 5323 6687 3706 9478
23.18 × 10−3 3.240 × 10−3 43.45 × 10−3 – 78.16 × 10−3 6.026 × 10−5
Molecular weight from the 1 H NMR results. Critical micelle concentration calculated from the excitation spectra of pyrene as a function of polymer concentrations in water. The ligand for the synthesis was N,N,N ,N ,N -pentamethyldiethylenetriamine.
the CNT, the amphiphilic surfactant-based dispersing route might be, however, problematic due to the potential toxicity of the surfactant [18–20]. We conceived that the design and nature of both the hydrophobic part and the hydrophilic part of the amphiphilic molecules are extremely important for the solubility and biocompatibility of the dispersed CNTs. In our previous studies, novel biomimetic block copolymers, cholesterol-end-capped poly(2-methacryloyloxyethyl phosphorylcholine) (CPMPC), which have poly(MPC) as hydrophilic segment and cholesterol as hydrophobic segment were specially designed and used as drug delivery system [21,22]. Micelles based on these biomimetic polymers were found to show better biocompatibility compared with the micelles prepared from the cholesterol-endcapped poly(ethylene oxide), which showed obvious toxicity. Furthermore, anti-cancer drug adriamycin was successfully incorporated into the hydrophobic cholesterol core of the micelles. Here we intent to study the applicability of the novel biomimetic block copolymer to solubilize and stabilize the CNTs, to investigate the effect of polymer concentration on the formation of dispersions, and to examine the effect of the nature of the surfactants on the solubility and biocompatibility of dispersed CNTs.
was purified by washing with glacial acetic acid, followed by absolute ethanol and ethyl ether, and then dried under vacuum. 2,2 -Bipyridine (bpy) (AR, Hangzhou Chemical Reagent Factory) and 1-pyrenemethanol (Aldrich Co) were used as received. CNTs purchased from Shenzhen Nanotech Port Co. Ltd. (multi-walled carbon nanotubes, diameter 10–20 nm, purity >95%) were used as received. Other reagents were purified by conventional methods. All the water used in this work is distilled and deioned (DDI) water. The conversion of cholesterol or 1-pyrenemethanol into the ATRP initiator (CholBr or PyBr) by reaction with 2-bromoisobutyryl bromide was performed as follows. In a 250 ml three-neck flask, cholesterol (4.988 g, 12.9 mmol) or 1-pyrenemethanol (3.0 g, 12.9 mmol) and triethylamine (5.4 ml, 38.7 mmol) were dissolved in dry CH2 Cl2 (40 ml). Then 2-bromoisobutyryl bromide (4.8 ml, 38.7 mmol) dissolved in dry CH2 Cl2 (20 ml) was added dropwise into the flask under stirring at 0 ◦ C. Then the reaction solution was allowed to warm to room temperature, and the reaction mixture was stirred for another 12 h. The reaction mixture was then filtered, and the CH2 Cl2 was removed using the rotary evaporator. The resulting yellow crude product was precipitated into cold ethanol. The precipitation was filtered and dried under vacuum [22,26].
2. Experimental 2.2. Preparation of the amphiphilic polymers 2.1. Materials The 2-methacryloyloxyethyl phosphorylcholine (MPC) monomer and poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC) were synthesized according to the literatures [22–25]. 2(Dimethylamino) ethyl methacrylate (DMAEMA), t-butyl acrylate (tBA) and 2-bromoisobutyryl bromide were purchased from Aldrich and used as received. Cholesterol-end-capped poly(ethylene oxide) (CPEG) was purchased from Nihon Emulsion Co. Ltd. and used as received. The CuBr (AR, Shanghai No. 1 Chemical Reagent Factory)
Typical protocols for the polymerization of MPC, t-butyl acrylate and DMAEMA using the CholBr as initiator or the polymerization of MPC using PyBr as initiator were carried out as follows. The monomers and corresponding initiators were dissolved in solvents. The flask, containing a stir bar, was degassed by three freezepump-thaw cycles with argon. The Cu(I)Br catalyst and bpy ligand were added to the stirred solution. The flask was then placed into an oil bath and thermostated at reaction temperature. After 20 h, the reaction mixture was cooled down and the flask was opened.
Fig. 1. TEM images for (A) the pristine CNTs, (B) the CPMPC coated CNTs, and (C) chemically oxidized CNTs-COOH.
F.-M. Xu et al. / Colloids and Surfaces B: Biointerfaces 67 (2008) 67–72
The resulting polymer mixtures were purified and characterized according to literatures [22,26–29]. The cholesterol-end-capped poly(t-butyl acrylate) was subjected to acidic hydrolysis in 1,4-dioxane (3 wt.% concentration solution, with the addition of a five-fold excess HCl 1 M, at 80 ◦ C for more than 12 h to afford cholesterol-end-capped poly(acrylic acid) [30]. The preparation of amphiphilic polymers were followed by FTIR and NMR and the preparation parameters and properties of the amphiphilic polymers are summarized in Table 1. 2.3. Dispersion of CNTs Cholesterol-end-capped poly(2-methacryloyloxyethyl phosphorylcholine) aqueous solution of different concentrations were prepared by dissolving weighted amount of polymer powder in DDI water and diluted to different concentrations by pipetted volume of DDI water. 3 ml polymer solution of each concentration was added into a vial. 20 mg of CNTs were added to each sample and the mixture was sonicated for 30 s in the low-power ultrasonic cleaner KQ2200 (Kunshan Ultrasonic Instruments Co. Ltd., 100 W, 40 kHz). The mixture was left for a certain time (1.5 h for UV absorbance test and 24 h for cell culture) without any disturbance to precipitate, then the upper 2 ml homogeneous solution of each vial was carefully transferred into another vial and preserved for other tests. To prepare the CNT suspensions using other type amphiphilic polymers as surfactants, the same procedure mentioned above was used. For stability assay, 3 ml aqueous solution of different polymer solutions (2 mg/ml) were added into the vials together with 15 mg CNTs. Then the samples were prepared by hand shaking. The transmittance at 600 nm of the CNTs suspension was recorded by UV spectrometer for 6 h. In cell culture experiments, CNTs were autoclaved under 121 ◦ C for 20 min and all the polymer solutions were sterilized by ultrafiltration (Millipore, 220 nm). 2.4. Instruments and measurements The Raman characterization was carried out on a Jobin Yvon microRaman system (Ramanor U1000, Instruments SA, USA) using a spectra-physics Ar ion laser at an excitation wavelength of 514.5 nm (2.41 eV). Their film samples were formed by drop coating on a clean silicon wafer with CNT suspension and drying at 80 ◦ C. The backscattered data were analyzed using a double-gating spectrometer and collected using a Hamamatsu photomultiplier (R 943-02, Hamamatsu, USA). For every Raman spectrum taken, the position of the peaks was verified by calibrating the spectral positions in respect to silicon substrate peak seen at 521 cm−1 [31]. Transmission Electron Microscopy (TEM) analysis was performed on a JEM-1200EX TEM operating at 200 kV in bright field mode. CNTs samples were prepared by placing a drop of the solutions (the concentration of CNTs is 1 mg/ml) on a 400-mesh carboncoated copper grid and air-drying the grid at 25 ◦ C. UV–vis spectra were carried out with a UV–vis Shimadzu UV-2055 spectrometer using quartz cuvettes. Spectra were collected within a range of 200–800 nm and the absorbance value at 600 nm was taken to compare CNTs solubility, since there was no absorbance value of each kind of polymer solution at 600 nm including PPMPC. 2.5. Cell viability Human umbilical vein endothelial cell line was chosen for cell culture studies. The cells were seeded into a 96 well cell culture plate at a density of 8000 cells per well for 24 h, then the growth medium was removed and replaced with the medium containing surfactant coated CNTs. The CNTs concentration in different sam-
69
ples was adjusted to the same level by UV absorbance at 600 nm before they were added into the culture medium, the concentration of CNTs in the culture medium of each well was 30 g/ml. The medium used was RPMI-1640 medium containing 10% heat inactivated fetal calf serum, 100 unit/ml penicillin and 0.1 mg/ml streptomycin. The cells were incubated at 37 ◦ C with 100% humidity in a 5% CO2 atmosphere for 96 h. To determine cell viability, 20 l MTT solution (5 mg/ml in PBS) was added to each sample, and incubated at 37 ◦ C for 4 h for MTT formazan formation. After the medium and MTT were replaced by 150 l DMSO, the samples were incubated at 37 ◦ C for 20 min to dissolve the MTT formazan. In order to avoid the effect of CNTs on the absorbance, all the DMSO solution in each well was pipetted into the same well of another new TCPS cell culture plate before the absorbance values were measured by using microplate reader (BIO-RAD, model 550) at wavelength 570 nm. Five replicates were read for each sample, the mean value of the five was used as the final result. Cells which were incubated in culture medium without any CNTs were used as the control. The cell viability was calculated as the ratio of the absorbency value of samples incubated with polymer coated CNTs to that of the control [21,32]. 3. Results and discussion CNTs have an exceptionally strong tendency to aggregate and it is difficult to suspend them in aqueous solutions. Surfactants showed the ability to encapsulate and separate CNTs and achieve CNTs colloidal stability. Traditional surfactants, such as sodium dodecyl sulfate, provide stable CNTs dispersions only at very high concentrations [15]. Furthermore, the detergent effect of the traditional surfactants induced potential toxicity. It is suggested that significant improvement of the dispersion efficiency can be obtained with polymeric agents [33,34]. While, there are a lot of researches published on suspending CNTs by the amphiphilic polymers and especially on the binding mode of the polymer onto the CNTs, the reports on the potential toxicity of the non-covalently dispersed CNTs via amphiphilic surfactant-based route are, however, scarce. The trial to resolve the toxicity of the amphiphilic surfactant dispersed CNTs is even much fewer. Furthermore, for the biomedical applications, the MWNTs with lower toxicity and larger loading capacity seem to be a better choice than SWNTs. Here we intent to study the ability of a novel biomimetic block copolymer, cholesterol-end-capped poly(2-methacryloyloxyethyl phosphorylcholine), for the solubilizing and stabilizing the CNTs (MWNTs). We showed in our previous studies that this biomimetic block copolymers combine both advantages of liposome and polymeric micelles and resulted in biocompatible and biological stable drug delivery systems [21]. Here, the cholesterol hydrophobic segment is believed to provide strong interactions with the CNTs surface and the poly(2methacryloyloxyethyl phosphorylcholine) hydrophilic segment is suggested to provide stability and biocompatibility of the system. 3.1. Dispersion of CNTs in aqueous medium by CPMPC The effectiveness of using CPMPC to disperse the CNTs was first followed by TEM as shown in Fig. 1. Although the pristine CNTs formed large bundles, Fig. 1(A), the addition of CPMPC into the pristine CNTs followed by 30 s sonication produced well dispersed CNTs, Fig. 1(B). This non-covalent dispersion process did not change the diameter of the MWNTs as compared with the pristine MWNTs, which showed an average diameter of 20 nm. It is notably to observe that while the pristine CNTs of 1–2 m were cut into short pieces of 100–500 nm during the carboxylation (Fig. 1(C)), which is a mandatory process in the chemically
70
F.-M. Xu et al. / Colloids and Surfaces B: Biointerfaces 67 (2008) 67–72
Fig. 2. UV/vis absorption spectrum of dispersed CNTs in CPMPC aqueous solution for different relative concentration (A, C0 ; B, (5/6)C0 ; C, (2/3)C0 ; D, (1/2)C0 ; E, (1/3)C0 ; F, (1/6)C0 ). C0 denote the saturated concentration of dispersed MWNTs in 1 mg/ml CPMPC aqueous solution. The inset is Lambert–Beer’s plot for the absorption at 600 nm.
modification of CNTs, the non-covalent dispersion process did not change the length of CNTs. The conservation of the diameter and the length suggests that the non-covalent dispersion process did not influence the physical structures of the CNTs. UV/vis absorption spectra of the CNTs dissolved in 1 mg/ml CPMPC aqueous solution at different concentrations were collected, Fig. 2. The intensities of absorption spectra had a linear relation with the concentration of the CNTs [35], Fig. 2, inset, and followed Lambert–Beer’s law, indicating the homogenous dispersion of the CNTs in solutions. The solubility of CNTs in 1 mg/ml CPMPC aqueous solution was calculated to be 1.705 mg/ml. Raman spectra provide an excellent method to examine the adherence of the amphiphilic polymers onto the surface of CNTs. The band in the 1500–1600 cm−1 region is the so-called G-band. An evident shift in the spectral position of the G-band before and after the modification with polymer can be noticed, where the G-band shifts by 10 cm−1 from 1572 to 1582 cm−1 (Fig. 3A). This change can be explained by considering wrapping of the polymers around CNTs. The hydrophobic and van der Waals attraction forces between the polymer and the graphite sheet increase the energy necessary for vibrations to occur, which resulted in the higher frequency of Raman peaks. The disorder peak, known as the D-band, can be found in the 1300–1400 cm−1 region. The position of the Dband typically changes when covalent modification of the graphene sheet occurs [31]. It was noticed that some insignificant changes in
Fig. 3. Raman spectra of (A) the pristine CNTs and (B) the CPMPC coated CNTs.
Fig. 4. (a) Comparison of CNTs solubility in DDI water with different concentrations of (A) CPMPC, (B) PMPC, and (C) PPMPC. (b) Comparison of CNTs solubility in DDI water with different concentrations of (A) CPMPC, (B) CPEG, (C) CPAA, and (D) CPDMAEMA.
the position of this D-band appeared, which might be attributed to the formation of polymer coating on carbon particles instead of CNTs. The Raman spectra data indicate that the polymers attach onto the CNTs non-covalently and the integrity of physical structure of the CNTs preserved. 3.2. The effect of the nature of the surfactants on the solubility of dispersed CNTs The role of the cholesterol hydrophobic segment of the CPMPC in stabilizing the CNTs was inspected. This was carried out by checking the solubility of the CNTs against concentrations of the polymers by using PPMPC and PMPC as controls as shown in Fig. 4(a). The increase of the concentration of CPMPC resulted in the increased solubility of CNTs. The solubility of CNTs reached a plateau when the concentration of CPMPC exceeded 3 mg/ml and the saturated concentration of CNTs reached 3.307 mg/ml. The saturated concentrations of CNTs reached 2.237 mg/ml in 3 mg/ml PMPC aqueous solution and 1.906 mg/ml in 3 mg/ml PPMPC aqueous solution, respectively, were much lower than that when the CPMPC was used. This result indicated that the presence of cholesterol hydrophobic segment is crucial for the solubilising ability of the CPMPC. While pyrene-based surfactants are believed to disperse the CNTs through – stacking between the pyrene groups and CNTs, the CNTs can also be dispersed through van der Waals force between the hydrophobic part and the CNTs. Several traditional surfactants [15], for example Phospholipids [17], have been showed to efficiently
F.-M. Xu et al. / Colloids and Surfaces B: Biointerfaces 67 (2008) 67–72
71
Fig. 7. The cell viability of cells cultured with CPMPC, PPMPC, CPEG, CPAA and CPDMAEMA-coated CNTs after 96 h incubation time. In control experiment, the cells were cultured in culture medium without CNTs. Fig. 5. Stability comparison of CNTs stabilized by (A) CPMPC, (B) PMPC and (C) PPMPC.
disperse and stabilize the CNTs. It was important to note here that the CPMPC with cholesterol hydrophobic segment showed better CNT-solubilising ability than PPMPC with pyrene hydrophobic segment, which is a traditional segment for CNT dispersing. It is well accepted that the attachment of hydrophilic and water-soluble groups onto the surface of CNTs is crucial for the biomedical applications. For the biocompatibility and colloidal stability reasons, a wide range of polymer-functionalized CNTs have been synthesized, and in particular, PAA [36,37], PDMAEMA [34], and PEG-functionalized CNTs [38] are the most widely studied. Here, we are interested in using the phosphorylcholine-type surfactants to generate stable and biocompatible CNTs. The effect of the hydrophilic segment composition of the surfactants on stabilizing the CNTs was inspected by checking the solubility of the CNTs as shown in Fig. 4(b). The solubility of CNTs reached a value of 1.705 mg/ml in 1 mg/ml CPMPC aqueous solution. When 1 mg/ml of CPAA, CPDMAEMA or CPEG aqueous solution was used, the solubility of CNTs was, however, only 0.7875 mg/ml, 0.5626 mg/ml and 1.468 mg/ml, respectively. Moreover, it was interesting to find that the solubility of CNTs decreased to 1.238 mg/ml, 0.3952 mg/ml, and 0.4669 mg/ml in 1 mg/ml CPMPC, CPDMAEMA, CPAA when PBS was used as buffer solution. The effect of the hydrophobic segment of the polymers on the stability of the CNTs was followed (Fig. 5). While the transmittance of the PMPC dispersed CNTs and the PPMPC dispersed CNTs increased rapidly within 6 h to 5.8% and 3.1%, respectively (Fig. 5, curve (B) and (C)), the transmittance of CPMPC did not show obvious increase (Fig. 5, curve (A)). This result indicated that the hydrophobic segment of the polymers showed great influence on the stability of the CNTs and the cholesterol-end-capped CPMPC showed excellent stabilizing ability for CNTs.
Furthermore, the stabilizing ability of different surfactants against pH changes was followed, Fig. 6. Although the CPDMAEMA stabilized CNTs was stable at acidic pH, Fig. 6(C), the change of pH to alkali value resulted the precipitation of the CNTs, Fig. 6(I). Similarly, the change of pH from alkali value to acidic value resulted in the transformation from dispersion state (Fig. 6(G)) to precipitation state of the CPAA stabilized CNTs (Fig. 6(A)). It was interesting to find that the CPMPC stabilized CNTs showed dispersion stability at both alkali (Fig. 6(H)) and acidic values (Fig. 6(B)). The solubility data indicate that the phosphorylcholine-type surfactants showed excellent solubilising ability for CNTs. 3.3. In vitro cytotoxicity measurements It is suggested that to achieve water-soluble CNTs for the biomedical application, physical adsorption of surfactants should be avoided since the surfactants used widely for solubilization of CNTs are always toxic by themselves [18–20]. A novel biomimetic surfactant with excellent biocompatibility, cholesterol-end-capped poly(2-methacryloyloxyethyl phosphorylcholine), was specially designed and synthesized. We were therefore interested in evaluating the biocompatibility of the CNTs solubilised by CPMPC. Human umbilical vein endothelial cells were cultured with CPMPC, CPAA, CPDMAEMA, CPEG or PPMPC-coated CNTs for 4 days. In control experiments, the cells were cultured with media alone. Cell viability was evaluated with MTT assay as shown in Fig. 7. The viabilities of the cells cultured with the CPEG, CPAA and CPDMAEMA coated CNTs reached only 8%, 70% and 12%, respectively. Presumably, the CPEG, CPAA or CPDMAEMA coated CNTs either inhibited cell growth or induced cell death. By contrast, cells cultured with CPMPC coated CNTs were able to growth very well during the course of the experiment and cell viability as high
Fig. 6. Dispersion stability tests of CNTs in DDI water with different surfactants at varied pH values: (A) CPAA at pH 2, (B) CPMPC at pH 2, (C) CPDMAEMA at pH 2, (D) CPAA at pH 7, (E) CPMPC at pH 7, (F) CPDMAEMA at pH 7, (G) CPAA at pH 13, (H) CPMPC at pH 13, and (I) CPDMAEMA at pH 13.
72
F.-M. Xu et al. / Colloids and Surfaces B: Biointerfaces 67 (2008) 67–72
as 92% was achieved. Similar results were obtained with PPMPC coated CNTs and cell viability of 95% was achieved. Thus, the phosphorylcholine-polymer coating renders the CNTs non-toxic. 4. Conclusions In conclusion, the present study has demonstrated the success of using the novel biomimetic copolymer CPMPC as amphiphilic surfactant to achieve water-soluble and biocompatible CNTs. The use of cholesterol group as hydrophobic segment is crucial for the solubilising ability and solubility of pristine CNTs as high as 3.307 mg/ml was achieved. The zwitterionic PMPC design afforded better solubilising ability than the negative PAA or positive PDMAEMA. We demonstrated here also that excellent biocompatibility of the zwitterionic engineered CNTs by checking the cell viability. The biocompatible CPMPC coated CNTs represent an excellent nano-object for the biomedical applications in nanomedicine. Acknowledgements This research was financially supported by Natural Science Foundation of China (NSFC-20774082, 50703036), 863 National High-Tech R&D Program (2006AA03Z329, 2006AA03Z444), Program for New Century Excellent Talents in University (NCET-050527) and High Technology Research and Development Program of Zhejiang Province (2007C24008). References [1] L. Leibler, H. Orland, J.C. Wheeler, J. Chem. Phys. 79 (1983) 3550. [2] P. Blasi, S. Giovagnoli, A. Schoubben, M. Ricci, C. Rossi, Adv. Drug Deliv. Rev. 59 (2007) 454. [3] A. Takasu, A. Takemoto, T. Hirabayashi, Biomacromolecules 7 (2006) 6. [4] D. Cochin, A. Laschewsky, F. Nallet, Macromolecules 30 (1997) 2278. [5] R.O. Dunn Jr., J.F. Scamehorn, S.D. Christian, Sep. Sci. Technol. 20 (1985) 257. [6] T. Ishii, H. Otsuka, K. Kataoka, Y. Nagasaki, Langmuir 20 (2004) 561. [7] B.A. Korgel, H.G. Monbouquette, Langmuir 16 (2000) 3588. [8] M.-S. Martina, J.-P. Fortin, C. Ménager, O. Clément, G. Barratt, C. GrabielleMadelmont, F. Gazeau, V. Cabuil, S. Lesieur, J. Am. Chem. Soc. 127 (2005) 10676.
[9] H.Y. Fan, E.W. Leve, C. Scullin, J. Gabaldon, D. Tallant, S. Bunge, T. Boyle, M.C. Wilson, C.J. Brinker, Nano Lett. 5 (2005) 645. [10] P. He, M.W. Urban, Biomacromolecules 6 (2005) 1224. [11] N. Nasongkla, E. Bey, J. Ren, H. Ai, C. Khemtong, J.S. Guthi, S.-F. Chin, A.D. Sherry, D.A. Boothman, J.M. Gao, Nano Lett. 6 (2006) 2427. [12] J.-W. Kim, E.V. Shashkov, E.I. Galanzha, N. Kotagiri, V.P. Zharov, Lasers Surg. Med. 39 (2007) 622. [13] L. Lacerda, A. Bianco, M. Prato, K. Kostarelos, Adv. Drug Deliv. Rev. 58 (2006) 1460. [14] Y. Lin, S. Taylor, H.P. Li, K.A.S. Fernando, L.W. Qu, W. Wang, L.R. Gu, B. Zhou, Y.P. Sun, J. Mater. Chem. 14 (2004) 527. [15] M.F. Islam, E. Rojas, D.M. Bergey, A.T. Johnson, A.G. Yodh, Nano Lett. 3 (2003) 269. [16] N. Nakashima, Y. Tomonari, H. Murakamiy, Chem. Lett. 31 (2002) 638. [17] Y. Wu, J.S. Hudson, Q. Lu, J.M. Moore, A.S. Mount, A.M. Rao, E. Alexov, P.C. Ke, J. Phys. Chem. B 110 (2006) 2475. [18] H. Isobe, T. Tanaka, R. Maeda, E. Noiri, N. Solin, M. Yudasaka, S. Iijima, E. Nakamura, Angew. Chem. Int. Ed. 45 (2006) 6676. [19] M. Zhang, M. Yudasaka, J. Miyawaki, J. Fan, S. Iijima, J. Phys. Chem. B 109 (2005) 22201. [20] M.J. O’Connell, S.M. Bachilo, C.B. Huffman, V.C. Moore, M.S. Strano, E.H. Haroz, K.L. Rialon, P.J. Boul, W.H. Noon, C. Kittrell, J.P. Ma, R.H. Hauge, R.B. Weisman, R.E. Smalley, Science 297 (2002) 593. [21] J.P. Xu, J. Ji, W.D. Chen, J.C. Shen, J. Control. Release 107 (2005) 502. [22] J.P. Xu, J. Ji, W.D. Chen, J.C. Shen, Macromol. Biosci. 5 (2005) 164. [23] K. Ishihara, T. Ueda, N. Nakabayashi, Polym. J. 22 (1990) 355. [24] H. Kitano, K. Sudo, K. Ichikawa, M. Ide, K. Ishihara, J. Phys. Chem. B 104 (2000) 11425. [25] J.P. Xu, J. Ji, W.D. Chen, D.Z. Fan, Y.F. Sun, J.C. Shen, Eur. Polym. J. 40 (2004) 291. [26] M.F. Wang, T.E. Dykstra, X.D. Lou, M.R. Salvador, G.D. Scholes, M.A. Winnik, Angew. Chem. 118 (2006) 2279. [27] I.Y. Ma, E.J. Lobb, N.C. Billingham, S.P. Armes, A.L. Lewis, A.W. Lloyd, J. Salvage, Macromolecules 35 (2002) 9306. [28] X. Zhang, J.H. Xia, K. Matyjaszewski, Macromolecules 31 (1998) 5167. [29] J.Y. Huang, T. Pintauer, K. Matyjaszewski, J. Polym. Sci. Pol. Chem. 42 (2004) 3285. [30] S.A. Angelopoulos, C. Tsitsilianis, Macromol. Chem. Phys. 207 (2006) 2188. [31] V.A. Sinani, M.K. Gheith, A.A. Yaroslavov, A.A. Rakhnyanskaya, K. Sun, A.A. Mamedov, J.P. Wicksted, N.A. Kotov, J. Am. Chem. Soc. 127 (2005) 3463. [32] L.Z. Gao, L. Nie, T.H. Wang, Y.J. Qin, Z.X. Guo, D.L. Yang, X.Y. Yan, Chem. Bio. Chem. 7 (2006) 239. [33] D. Wang, W.X. Ji, Z.C. Li, L.W. Chen, J. Am. Chem. Soc. 128 (2006) 6556. [34] P. Petrov, F. Stassin, C. Pagnoullea, R. Jérôme, Chem. Commun. 3 (2003) 2904. [35] Y.P. Sun, W. Huang, Y. Lin, K. Fu, A. Kitaygorodskiy, L.A. Riddle, Y.J. Yu, D.L. Carroll, Chem. Mater. 13 (2001) 2864. [36] Y. Kang, T.A. Taton, J. Am. Chem. Soc. 125 (2003) 5650. [37] A.H. Liu, I. Honma, M. Ichihara, H.S. Zhou, Nanotechnology 17 (2006) 2845. [38] R.J. Chen, S. Bangsaruntip, K.A. Drouvalakis, N.W.S. Kam, M. Shim, Y.M. Li, W. Kim, P.J. Utz, H.J. Dai, Proc. Natl. Acad. Sci. U.S.A. 100 (2003) 4984.