DRUG DELIVERY STUDY OF SINGLE-WALL CARBON NANOTUBES COVALENT FUNCTIONALIZED WITH CISPLATIN
C. C. CIOBOTARUa*, C. M. DAMIANb, S. POLOSANa, M. PRODANAb, H. IOVUb
aNational Institute of Materials Physics, Atomistilor 105 bis, 077125, Magurele, Bucharest, Romania
bUniversity POLITEHNICA of Bucharest, Faculty of Applied Chemistry and Materials Science, Calea Victoriei 149, 010072, Bucharest, Romania
Carbon nanotubes are widely studied components for drug delivery systems due to their high surface area and low chemical reactivity. The research presented in this paper deals with the synthesis of drug delivery systems based on single walled carbon nanotubes (SWCNTs) and the well-known cancer treatment drug Cisplatin. The new nanomaterials obtained through covalent bonding between carboxyl groups from the SWCNTs surface and amino groups from the Cisplatin structure were characterized from structural point of view. To evaluate the content of drug released the Inductively Coupled Plasma Mass Spectrometry (ICP-MS) was employed. The releasing profile shows a slow rate in the beginning followed by a spectacular increase after 180 minutes which means that this type of system could be used for prolonged release.
(Received April 26, 2014; Accepted June 13, 2014)
Keywords: SWCNTs, functionalization, covalent activation, Cisplatin, XPS, ICP-MS
1. Introduction
Nanocarriers for low-molecular-weight drugs offer a promising strategy for improving body distribution and prolonging blood circulation. Recently, single-walled carbon nanotubes (SWCNTs) have been investigated as carriers in living systems [1-4] showing that these materials are very good for drug delivery systems.
Generally, drug delivery systems enter into the cells by endocytosis through the cell membrane. In order to deliver the drugs to the nucleus, it is important that the drug carrier escapes endosomal compartment and releases drug load in cytosolic compartments. One of the advantages of SWCNTs is the capability of delivering chemotherapeutic and imaging agents by overcoming these biological barriers and localizes the target tissue [5-6].
The loaded dose of the drug in direct bonding to CNT is quite limited. Therefore high concentrations of CNT are required for delivery of sufficient amount of drug or the functionalization of CNT. SWCNTs functionalization is important for dispersion and solubilisation [7-8] and to enable further modification [9].
Cis-diammineplatinum(II) dichloride (cisplatin, CDDP) a square planar Pt2+ complex was the first metal –based agent which enter into worldwide clinical use for the treatment of cancer.
Cisplatin is one of the most potent and widely-used anticancer drug and currently is used either by itself or in combination with other drugs for the treatment of a variety of solid tumors, including testicular, ovarian, bladder, cervical, head and neck, esophageal, colon, gastric, breast, melanoma, prostate cancer, and small-cell lung cancers [10-11].
C. Tripisciano et al. [12] described a method of embedding CDDP through SWCNTs internal diameter. They studied the release of the CDDP bonded noncovalent inside the SWCNTs by ICP showing that 68% of drug was released after 72h.
* Corresponding author: [email protected]
In another study, A. A. Bhirde et al. [13] reported a method of activation of carboxyl functionalized SWCNTs with CDDP by dispersing them in dimethyl sulfoxide at room temperature for 1 h. Using different methods they calculate the size of CDDP molecules bonded on the SWCNTs. Also they observed that one Pt atom is corresponding to 10 nm length of SWCNTs.
K. W. Ciecwierz et al. [14] developed a new method of introducing CDDP inside oxidized SWCNTs by dispersing SWCNTs in a solution of CDDP in dimethylformamide (DMF) or water through sonication followed by stirring for about 20 h at room temperature. After drug release studies they conclude that the best results were obtained using DMF as solvent.
A similar method for encapsulation of CDDP in MWCNT was presented by Li et al [15].
They mixed and ultrasonicated CDDP and MWCNT in ethyl acetate followed by stirring in the dark. They evaluated the loading of CDDP in MWCNTs with TGA and ICP-OES and the result was 0.621 mg of CDDP for each 1 mg of MWCNT-CDDP sample.
The aim of this paper is focused on developing of a new route for synthesis of nanocomposites based on covalent functionalization of SWCNTs with CDDP for drug delivery systems. Several experimental techniques were involved in order to investigate the formation of covalent bonds between SWNCTs and CDDP like Fourier Transform Infrared (FT-IR), Raman Spectroscopy measurements, X-ray Photoelectron Spectroscopy (XPS) and X-ray Diffraction (XRD) for structural characterization. Thermogravimetric Analysis (TGA) and Scanning Electron Microscopy (SEM) were employed to bring additional information about the newly synthesized system. Moreover, Inductively Coupled Plasma Mass Spectrometry (ICP-MS) was used to determine the concentration of drug released.
2. Materials and methods 2.1 Materials
Single-wall carbon nanotubes (SWCNTs) were purchased from Sigma Aldrich having chirality 6.5, more than 90% carbon basis, more than 77% carbon as SWCNTs, and diameter range between 0.7 and 0.9, produced by CoMoCAT® Catalytic Chemical Vapor Deposition (CVD) Method.
Activation process was done by using 1-(3-Dimethylaminopropyl)-3-ethylcarbodiimide (EDC), N-Hydroxysuccinimide (NHS) and Cisplatin purchased from Sigma Aldrich.
The reaction was made in potassium phosphate monobasic solution having pH=4.5.
Cellulose dialysis sacks having a molecular weight cutoff of 12000 (Aldrich) were used for the drug release study.
Phosphate buffer solution (PBS) having pH=7.4 was obtained by mixing 250 mL of potassium phosphate monobasic having concentration of 2M and 393.4 mL of sodium hydroxide having concentration of 0.1 M.
2.2 Methods
2.2.1. Activation of SWCNTs purified for 48h and oxidized (SWCNTs-p48h-COOH) with CDDP
SWCNTs as-received were previously purified and oxidized as already described [16].
Covalent functionalization of SWCNTs-p48h-COOH with CDDP was done by using EDC and NHS. Briefly 30 mg of EDC were dissolved in 3ml of phosphate monobasic solution. After that, 30 mg of SWCNTs-p48h-COOH were added. Then 90 mg of NHS were added and the mixture was sonicated 30 min at room temperature (~25°C) for activation of the carboxylic groups. The second step was the addition of 30 mg CDDP to obtain amide bonding to the SWCNTs surface, this step required sonication for another 90 min maintaining the mixture at room temperature.
When the sonication was finished the solution was filtered and washed several times with distilled water. The water used for washing was collected and analyzed by ICP-MS in order to calculate the amount of CDDP covalently bonded on the SWCNTs surface using equation (1).
100 ) (
* 100
%
CDDP
W feed
free CDDP W
feed CDDP W
efficiency Loading
CDDP
(1)where WfeedCDDP is the mass of CDDP that was introduced into the reaction and WfreeCDDP is the mass of unbound CDDP that was calculated using ICP-MS from washing solution.
2.2.2. In vitro CDDP release covalently bonded onto SWCNTs-p48h-COOH surface.
After activation the SWCNTs with CDDP was dispersed in 5 ml of PBS pH=7.4 for 2 min using ultrasounds for a better dispersion. Then the solution was transferred into a dialysis sack previously washed with PBS 7.4 to remove all the impurities.
The sack containing the solution of SWCNTs with CDDP and PBS was closed to both ends and immersed in a glass with 50 ml of PBS 7.4 and placed on a magnetic stirrer thermostated at 37 ○C and 300 rpm. From time to time 4 ml of dissolution medium were extracted and analyzed with ICP-MS. The sample volume was replaced with fresh PBS 7.4
Fig. 1. Scheme of purification, oxidation and activation process of SWCNTs with CDDP
2.3 Advanced characterization
Fourier transform infrared spectroscopy (FTIR) spectra of SWNCTs, purified SWCNTs oxidized SWCNTs and activated SWCNTs with CDDP were registered on a Bruker Vertex 70 equipment in 400 ÷ 4000 cm-1 range with 4 cm-1 resolution and 32 scans. The samples were analyzed in KBr pellets.
Raman spectra were recorded on a DXR Raman Microscope (Thermo Scientific) by 532 nm laser line. The 10x objective was used to focus the Raman microscope.
Thermogravimetry analysis (TGA) of the samples was done on Q500 TA equipment, using nitrogen atmosphere from 20 °C to 900 °C with 10 °C/min heating rate.
The X-ray photoelectron spectroscopy (XPS) spectra were recorded on Thermo Scientific K-Alpha equipment, fully integrated, with an aluminum anode monochromatic source. Survey scans (0-1350 eV) were performed to identify constitutive elements.
The X-Ray diffraction measurements have been performed on a BRUKER D8 ADVANCE type X-ray diffractometer. The working parameters are 40 kV and 40 mA. The 2θ scan range was set to 5–50° with a step size of 0.04° and a resolution of 0.01°.
Scanning electron microscopy (SEM) was done on a Quanta Inspect F, a FEI instrument, with a field emission electron gun, 1.2 nm resolution and X-ray energy dispersive spectrometer having an accelerating voltage of 30 kV.
Inductively coupled plasma mass spectrometry (ICP-MS) is an ELAN DRC-e Perkin Elmer SCIEX U.S.A equipment with detection limit of 0.001 μg/g.
method COOH p48h-C charac appear of two the sec
throug sp2 gra was ca
3. Resul 3.1 FT-IR The FT-IR d, thus KB H+CDDP. Fi COOH and cteristic peak
r at 1715 and According o new peaks, cond at 532 c
Fig. 2. FT
2.2 RAMA Raman sp gh characteris
aphitic struct alculated the
Fig. 3. R
lts and dis R Spectrosco
R spectrum Br pellets w
igure 2 pres SWNCTs-p ks correspon d 1635 cm-1 g to SWNCT , one at 694 cm -1 assigne
T-IR spectra fo
AN Spectros pectroscopy
stic D and G ture of the na
ID/IG ratio w
Raman spectra
scussion py
of SWNCT were obtain sents the FT p48h-COOH nding to C=
[17, 18] indi Ts-p48h-COO
cm-1 corresp ed to the HN-
for 1) SWCNTs 4) SWCNT
scopy was perform
peaks from anotubes wal which gave in
a for 1) SWCN 4) SWCN
Ts-p48h-COO ned having T-IR spectra H+CDDP. It
=O stretching icating that C OH+CDDP s
ponding to t -Pt stretching
Ts, 2) SWCNTs Ts-p48h-COO
med to obta CNT spectra ll. In order to nformation ab
NT, 2) SWCNT NT-p48h-COOH
OH+CDDP 0.5% conc for SWNCT t can be s g vibration b COOH functi structure, spe the deformat
g vibration [
s-p48h, 3) SW OH+CDDP
ain informati a by analyzin o verify the m
bout the func
T-p48h, 3) SW OH+CDDP
was obtain centration o Ts, SWNCT seen that af bonds from ionalization w ectrum 4 sho
ion vibration 19].
WCNTs-p48h-C
ion about th ng the defect modification ctionalization
WCNT-p48h-CO
ned using th of SWNCTs Ts-p48h, SW after oxidati
carboxylic was perform ows the appe n of NH bon
COOH and
he nanocom ts that can ap of CNT stru n reaction.
OOH and
he KBr s-p48h- WNCTs- on the
groups med.
earance nds and
mposites ppear in ucture it
was ob the SW possib SWCN by the noncov
Table
Sa SW SW SW SW
therma in nitro
presen weight dissoci 200○C third s COOH
sample which
Although p bserved that WNCTs surf le other stru NT-p48h-CO
enriched ele valently on t e 1. ID/IG ratio
ample WCNT WCNT-p48h WCNT-p48h WCNT-p48h
2.3 Therm Thermogra al stability an ogen atmosp
Fig. 4. Therm
Degradatio nts different t loss. Three iation and t , the second stage corresp H+CDDP is 5 2.4 XPS A The XPS a es. Thus from
increase dra
purification after oxidat face was co uctural defec OOH with CD ectronic dens the graphitic o for SWCNTs
h
h-COOH h-COOH+CD
mogravimetr avimetric an nd the weigh phere.
mogravimetric 4) SW
on of SWCN behavior co e main steps the decarbox d stage is ass ponds to the 56% compar Analysis
analyses wer m figure 5 it amatically af
of SWCNTs tion (spectrum onsiderably m cts, due to DDP, the ID/I
sity of CDDP wall of CNT s, SWCNTs-p4
ID 11 17 22 DDP 6.
ric analysis nalysis (TGA ht loss of nan
c curves for 1) WCNT-p48h-CO
NTs-p48h-C ompared wit s are noticed
xylation and signed to the e degradation ed with 42%
re employed can be obser fter oxidation
s did not mo um 3) this va modified by the extreme IG ratio show P which has Ts [12] betwe 48h, SWCNTs-
D XD, c
15.0 1308 74.2 1311 26.9 1349 13 1347
A) is a met nocomposite
) SWCNT, 2) OOH+CDDP
COOH+CDD th SWCNTs d, thus the f d dehydroge e degradatio n of SWCN
% for SWCNT
d to determin rved the pres n process fro
odified signi alue increase y the introdu e oxidation c ws a small de small molec een carboxyl -p48h-COOH,
cm-1 IG 8.3 155 1.0 167 9.6 284 7.8 8.28
thod that giv es by heating
SWCNT-p48h P and inset 5) C
P as it can -p48h-COOH first stage is enation proce on of CDDP Ts. The tota Ts-p48h-CO
ne the surface sence of oxy om 3.78 to 1
ficantly the es from 0.1 t uction of ca conditions. A ecrease whic cules and cou
l type defect H, SWCNTs-p4
XG, c 6.4 1581 9.1 1582 .7 1598 8 1574
ves the info g the samples
h, 3) SWCNT-p CDDP.
be seen in H curve in attributed to esses of –C
between 35 al mass loss
OH. [14, 15]
e elemental ygenated stru 18.97 At.%.
ID/IG ratio v to 0.8 meani arboxyl grou After activa ch can be ex uld be attach
sites.
48h-COOH+C
cm-1 ID/IG 1.2 0.07 2.9 0.10 8.4 0.79 4.4 0.76
ormation abo s from 20 to
p48h-COOH,
n figure 3 c terms of gr to the amino COOH group
50 – 500○C a of SWCNT ].
composition uctures throu This was ex
value, it ing that ups and ation of plained hed also
CDDP.
G
7 0 9 6
out the o 900°C
curve 4 adually o bonds ps near and the T-p48h-
n of our ugh O1s xpected
becaus of carb elemen
Tabl Samp At % C 1s O 1s Mo 3 N 1s Pt 4f
oxidati at % an on SW
se the ratio b boxylic grou nt which indi
Fig. 5
le 2. XPS data ple
%
3d f
The values ion and resp nd N 1s from WCNTs surfac
Fig. 6. X
between C at ups. Also th icates the bo
5. XPS spectra
a for SWCNTs SWCNT 96.43 3.25 0.32 0 0
s from table ectively CDD m amino and
ce.
500 1000 1500 2000 2500
Intensity(a.u.)
XPS spectra of
toms and O he activation onding of CD
a for 1) SWCN 4) SWCN , SWCNTs-p4 T SWCN
p48h 96.09 3.78 0.13 0 0
e 2 show the DP function d amide group
85 80
B
f Pt 4f element
atoms is mo n process wa DDP to the SW
NT, 2) SWCNT NT-p48h-COOH
48h, SWCNTs-
NT- SW
CO 81.0
18.9 0 0 0
e surface co alization. Th ups having 1.
75 Binding Energy (eV
4f 4f5/2
t from CDDP
odified durin as highlighte WCNTs surf
T-p48h, 3) SW OH+CDDP
-p48h-COOH, WCNT-p48h-
OOH 03 97
mposition o he appearanc 42 at % dem
70 6
V)
CDDP (x2.5) SWCNT-p48h-C
f7/2
P and SWCNTs
g oxidation ed by the ap face.
WCNT-p48h-C
, SWCNTs-p4 SW COO 83.4 14.1 0 1.42 1.02
f our sample ce of Pt 4f at monstrate the
65
COOH+CDDP
s-p48h-COOH
by the introd ppearance o
COOH,
48h-COOH+C WCNT-p48h- OH+CDDP 4
16 2 2
es before an t 72 eV havin
presence of
H+CDDP
duction f Pt 4f
CDDP
nd after ng 1.02 f CDDP
Deconvolution of Pt 4f element (Figure 6) from CDDP and SWCNTs-p48h- COOH+CDDP was performed in order to determine the bonding mode of CDDP to SWCNTs and the ratio between the absorption bands of Pt element was calculated.
For SWCNT-p48h-COOH+CDDP the maximum energy of Pt 4f5/2 shows a shifting from 74.57 to 76.48 eV and respectively Pt 4f7/2 band from 71.36 to 73.09 eV compared with CDDP. By calculating the difference, it results a shifting of 1.91 eV for Pt 4f5/2 and 1.73 eV for Pt 4f7/2. Through correlation with the data existing in the literature it was found that the bond between the carboxyl groups from SWNCTs and the amino groups from CDDP is amide type linkage [20].
2.5 X-ray Diffraction
X-ray diffraction was employed to calculate the dimensions of the nanotubes and nanocomposites through Debye-Scheerer ecuation (2)
) cos(
*
*
D K
(2)Where D is the dimension of the nanotubes and nanocomposites form 002 direction, λ=1.54 *10-10 m, β is half-width in radian calculated by peak aria from 002 direction and K is constant = 0.9.
Table 3. XRD data for SWCNTs, SWCNTs-p48h, SWCNTs-p48h-COOH, SWCNTs-p48h-COOH+CDDP.
Sample θ
(°)
θ/2 (°) Cos(θ) Half- width (°)
Half- width (rad)
Half-width
*cos(θ)
D*100 (nm)
SWCNT 25.71 12.85 0.97 2.20 38.37 37.22 3.72
SWCNT-p48h 25.74 12.87 0.97 2.33 40.64 39.42 3.51
SWCNT-p48h-
COOH 25.7 12.85 0.97 2.35 40.99 39.76 3.48
SWCNT-p48h-
COOH+CDDP 25.7 12.85 0.97 1.7 29.65 28.76 4.81
Structural characterization by XRD data from table 3 show that the dimension of nanotubes decreases after purification and oxidation process due to the removal of amorphous carbon and metal particles from catalyst used for the SWCNTs synthesis. This trend for decrease is correlated with XPS where Mo 3d content was 0 after oxidation process.
After activation process, the dimension of nanocomposites obtained by covalent functionalization of SWCNTs with CDDP increases from 3.48 to 4.81 nm. This increase can be explained by the CDDP bonding onto SWCNTs surface [21].
the cau SWCN loading molecu figure to the c
using t
Fi
2.6 Scanni Electronic use of their NTs have cle g occur, the ules hinder V
8b shows w carboxyl gro
2.7 Drug d In order to the data give
20
Intensity (a.u.)
ig. 7. XRD pat
ing electron microscopy r agglomerat ean-cut ends
bundles are Van der Wa white points oups from SW
Fig. 8.
delivery stud o evaluate the en by ICP-M
30
(002)
tterns for SWC SWCNT
n microscopy of SWCNTs ted appearan
due to highe e broken (no aals forces th
at the end of WCNTs .
SEM images
dy using Ind e drug releas
S.
4 2 SWCN SWCN SWCN SWCN
CNT, SWCNT T-p48h-COOH
y
s samples wa nce; also it c er density of oticeable eve hat manifest f the SWCN
for SWCNT-p
ductively co se from SWC
40
NT-p48h-COOH NT-p48h-COOH NT-p48h NT
T-p48h, SWCN H+CDDP
as performed can be notic f functional en in dry sta t usually betw NTs which ca
p48h-COOH+
upled plasm CNTs it was r
50 H+CDDP H
NT-p48h-COO
d on dry pow ced in Figur groups in th ate), due to t ween CNTs an be conside
+CDDP
ma mass spec realized the
OH,
wder which c re 8a that o hat area. Whe
the fact that . SEM imag ered CDDP
ctrometry drug release
ould be xidized en drug CDDP ge from bonded
profile
Fig. 9. Drug delivery profile of CDDP from SWCNT-p48h-COOH+CDDP
From figure 9 it can be observed that the CDDP begins to release after 150 minutes releasing 1.2% of drug. After 72 h (4320 minutes) the drug released was 21.3% which is in good agreement with the data reported in the literature [22, 23].
Calculations of the amount of CDDP attached on the SWCNTs surface showed that all the drug quantity was incorporated during nanocomposite synthesis, thus the initial stage of slow release could be explained by the fact that the drug was covalently bonded to the nanotubes and breaking these bonds takes some time given these conditions.
This low release rate can be explained by the impossibility of breaking all C-N bonds in PBS 7.4 formed during the activation reaction [24].
3. Conclusions
This study developed a new route for synthesis of nanocomposites based on covalent functionalization of SWCNTs with CDDP for drug delivery systems. Using XPS and FT-IR characterization it was established that the molecules of CDDP were covalently bonded on SWCNTs functionalized with carboxyl groups.
The XRD patterns proved that the surface of SWCNTs was modified through the increasing of the dimensions for nanocomposites samples. The SEM micrographs showed the presence of CDDP especially on the ends of SWCNTs due to the higher functionalization density in this area.
In vitro drug release from SWCNT-p48h-COOH+CDDP was performed using ICP-MS and it was observed that the content of drug released after 72h was 21%.
Acknowledgement
The work has been funded by the Sectoral Operational Programme Human Resources Development 2007-2013 of the Romanian Ministry of Labour, Family and Social Protection through the Financial Agreement POSDRU/107/1.5/S/76903.
References
[1] L. Meng, X. Zhang, Q. Lu, Z. Fei, P. J. Dyson, Biomaterials 33, 1689, (2012).
[2] Z. Liu, J. T. Robinson, S. M. Tabakman, K. Yang, H. Dai, Mater Today 14, 7-8, (2011).
[3] S. K. Vashist, D. Zheng, G. Pastorin, K. Al-Rubeaan, J. H.T. Luong, F.-S. Sheu, CARBON 49, 4077, (2011).
[4] S. Prakash, M. Malhotra, W. Shao, C. Tomaro-Duchesneau, S. Abbasi, Adv Drug Deliver Rev. 63, 14-15, (2011).
[5] X. Zhang, L. Meng, Q. Lu, Z. Fei, P. J. Dyson, Biomaterials 30, 6041 (2009).
[6] E. Heister, V. Neves, C. Tîlmaciu, K. Lipert, V. S. Beltran, H. M. Coley, S. Ravi P. Silva, J. McFadden, CARBON 47, 2152 (2009).
[7] J.L. Bahr, E.T. Mickelson, M.J. Bronikowski, R.E: Smalley, J.M. Tour, Chem Commun.
193, 4, (2001).
[8] M Lebron-Colon, M A Meador, D Lukco, F Sola, J Santos-Perez, L S McCorkle, Nanotechnology 22, 45, (2011).
[9] C. Nichita, I. Stamatin, Digest Journal of Nanomaterials and Biostructures, 8(1), 445, (2013).
[10] R.P. Feazell, N Nakayama-Ratchford, H. Dai, S.J. Lippard, J Am Chem Soc.
129, 8438, (2007).
[11] A. Bianco, K. Kastarelos, C.D: Partidos, M Prato, Chem Commun. 5, 571, (2005).
[12] C. Tripisciano, K. Kraemer, A. Taylor, E. Borowiak-Palen, CHEM PHYS LETT.
478, 200, (2009).
[13] A.A. Bhirde, A.A. Sousa, V. Patel, A.A. Azari, J.S. Gutkind, R.D. Leapman, J.F. Rusling, Nanomedicine-UK 4, 7, (2009).
[14] K. Werengowska-Ciecwierz, M. Wisniewski, A. P. Terzyk, N. Gurtowska, J. Olkowska, T. Kloskowski, T. A. Drewa, U. Kiełkowska, S. Druzynski, CARBON 70, 46, (2014).
[15] J. Li, S. Q. Yap, S. L. Yoong, T. R. Nayak, G. W. Chandra, W. H. Ang, T. Panczyk, S. Ramaprabhu, S. K. Vashist, F.-S. Sheu, A. Tan, G. Pastorin, CARBON 50, 1625 (2012).
[16] C. C. Ciobotaru, C. M. Damian, H. Iovu, U.P.B. Sci. Bull. B, 75, 2, (2013).
[17] M. M. Stylianakis, J. A. Mikroyannidis, E. Kymakis, SOL ENERG MAT SOL C, 94, 2, (2010).
[18] A. Li, L. Wang, M. Li, Z. Xu, P. Li, Z. Z. Zhang, Digest Journal of Nanomaterials and Biostructures, 9(2), 599, (2014).
[19] S. Likhitkar, A. K. Bajpai, Carbohyd Polym, 87, 1, (2012).
[20] A. Guven, I. A. Rusakova, M. T. Lewis, L. J. Wilson, Biomaterials 33, 1455, (2012).
[21] K. Ajima, M. Yudasaka, A. Maigne, J. Miyawaki, S. Iijima, J Phys Chem B 110, 5773, (2006).
[22] H. Xiao, R. Qi, S. Liu, X. Hua, T. Duan, Y. Zhenga, Y. Huang, X. Jing, Biomaterials 32, 7732, (2011).
[23] S. Bontha, A. V. Kabanov, T. K. Bronich, J Control Release 114, 163, (2006).
[24] K. Ajima, T. Murakami, Y. Mizoguchi, K. Tsuchida, T. Ichihashi, S. Iijima, M. Yudasaka, ACS NANO, 2, 10, (2008)