Biotinylated amphiphile-single walled carbon nanotube conjugate for target-specific delivery to cancer cells

Article abstract of DOI:10.1039/c3tb21334j

The present work reports the specific targeting of cancerous cells using a non-covalently water dispersed nanoconjugate of biotinylated amphiphile-single walled carbon nanotube (SWNT). The fundamental approach involves incorporation of the biotin into the architecture of the carbon nanotube (CNT) dispersing agent to develop a multifaceted delivery vehicle having a high colloidal stability, substantial cell viability and targeted specificity towards cancer cells. A three way functionalization strategy was employed to introduce a C-16 hydrophobic segment, polyethylene glycol hydrophilic fragment and biotin as the target-specific unit at the -OH, -COOH and -NH₂ terminals of l-tyrosine, respectively. The newly developed neutral amphiphile exhibited an efficient SWNT dispersion (72%) in water, significant viability of different mammalian cells (Hela, HepG2, CHO and HEK-293) up to 48 h and also media stability. Most importantly, the biotinylated amphiphile-SWNT dispersion successfully transported the fluorescently labelled Cy3-oligoneucleotide (loaded on the surface of CNT) inside the cancerous Hela, HepG2 cells after 3 h of incubation, in contrast to CHO and HEK-293 cells (devoid of overexpressed biotin receptors). The presence of the biotin moiety in the cellular transporters facilitated the internalization of cargo due to the overexpressed biotin receptors in the cancer cells. Importantly, this nanohybrid was also capable of specifically transporting the anticancer drug doxorubicin to cancer cells, which led to the significant killing of Hela cells compared to the normal CHO cells. Thus, the receptor-mediated specific transportation of cargo into cancer cells was possible only due to the biotinylated CNT dispersing agent. To the best of our knowledge this is the first reported amino acid based biotinylated small amphiphilic molecule that non-covalently dispersed SWNTs and the corresponding nanoconjugate showed excellent cell viability, antibiofouling properties and the desired target-specific drug delivery.

Full text of DOI:10.1039/c3tb21334j

View Article Online
View Journal
JMouarntaelroiaf lsChemistryB
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: S. Brahmachari, M. Ghosh, S. Dutta and
P. K. Das, J. Mater. Chem. B, 2013, DOI: 10.1039/C3TB21334J.
This is an Accepted Manuscript, which has been through the RSC Publishing peer
Journalof
review process and has been accepted for publication.
Materials Chemistry B
Accepted Manuscripts are published online shortly after acceptance, which is prior
Materials for biology and medicine
www.rsc.org/MaterialsB
Volume 1 | Number 1 | January 2013 | Pages 0000–0000
to technical editing, formatting and proof reading. This free service from RSC
Publishing allows authors to make their results available to the community, in
citable form, before publication of the edited article. This Accepted Manuscript will
be replaced by the edited and formatted Advance Article as soon as this is available.
To cite this manuscript please use its permanent Digital Object Identifier (DOI®),
which is identical for all formats of publication.
More information about Accepted Manuscripts can be found in the
Information for Authors.
Please note that technical editing may introduce minor changes to the text and/or
graphics contained in the manuscript submitted by the author(s) which may alter
content, and that the standard Terms & Conditions and the ethical guidelines
that apply to the journal are still applicable. In no event shall the RSC be held
responsible for any errors or omissions in these Accepted Manuscript manuscripts or
any consequences arising from the use of any information contained in them.
ISSN 2050-750X
XXXXXXXXXXXX
XXXXXXXXXXXX
Xxxx Xxxxx xxxXxxx Xxxxx
Xxxx Xxxxx xxxXxxx Xxxxx
Xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx Xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
xxxxxxxxxxxxx
xxxxxxxxxxxxx
0959-9428(2010)20:1;1-A
www.rsc.org/materialsB
Registered Charity Number 207890

Page 1 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
Biotinylated amphiphile-single walled carbon nanotube conjugate
for target-specific delivery to cancer cells
Sayanti Brahmachari, Moumita Ghosh, Sounak Dutta, and Prasanta Kumar Das*
Department of Biological Chemistry, Indian Association for the Cultivation of Science,
Jadavpur, Kolkata – 700 032, India
*To whom correspondence should be addressed. Fax: +(91)ꢀ33ꢀ24732805, Eꢀmail:
bcpkd@iacs.res.in
1

Journal of Materials Chemistry B
Page 2 of 41
View Article Online
DOI: 10.1039/C3TB21334J
Abstract
The present work reports specific targeting of cancerous cells using nonꢀcovalently
water dispersed nanoconjugate of biotinylated amphiphileꢀsingle walled carbon nanotube
(SWNT). The fundamental approach involves incorporation of the biotin into the architecture
of the carbon nanotube (CNT) dispersing agent to develop a multifaceted delivery vehicle
having high colloidal stability, substantial cell viability and target specificity towards cancer
cells. A three way functionalization strategy has been employed to introduce a Cꢀ16
hydrophobic segment, polyethylene glycol hydrophilic fragment and biotin as the targetꢀ
specific unit at the ꢀOH, ꢀCOOH and ꢀNH2 terminals of Lꢀtyrosine, respectively. The newly
developed neutral amphiphile exhibited efficient SWNT dispersion (72%) in water,
significant viability of different mammalian cells (Hela, HepG2, CHO and HEKꢀ293) up to
48 h and also media stability. Most importantly, the biotinylated amphiphileꢀSWNT
dispersion successfully transported the fluorescently labelled Cy3ꢀoligoneucleotide (loaded
on the surface of CNT) inside the cancerous Hela, HepG2 cells after 3 h of incubation in
contrast to CHO and HEKꢀ293 cells (devoid of overexpressed biotin receptors). The presence
of biotin moiety in the cellular transporters facilitated the internalization of cargo due to the
overexpressed biotin receptors in cancer cells. Importantly, this nanohybrid was also capable
of specifically transporting anticancer drug doxorubicin to cancer cells that led to the
significant killing of Hela cells compared to the normal CHO cells. Thus, receptorꢀmediated
specific transportation of cargo into cancer cells was possible only due to the biotinylated
CNT dispersing agent. To the best of our knowledge this is the first reported amino acid
based biotinylated small amphiphilic molecule that nonꢀcovalently dispersed SWNT and the
corresponding nanoconjugate showed excellent cell viability, antibiofouling property and
desired targetꢀspecific drug delivery.
2

Page 3 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
Introduction
In the last decade, the uses of carbon allotropes in several frontiers of research have gained
significant importance owing to its unique physical, electronic and optical properties.1
Amongst others, SWNT being endowed with cylindrical structure and high aspect ratio have
emerged with enormous promise in nanobiotechnology.2 Of late substantial efforts have been
made for the engineering of multifaceted nanosystems capable of serving as efficient
diagnostic and therapeutic tools.3 In fact the ability of SWNT to translocate through cell
membrane and its chemical inertness make this carbon nanomaterial highly promising in the
quest of developing cutting edge drug delivery vehicle. The huge surface area of SWNT is
considered to be ideal for uploading cargo at higher concentration, which can assist in
improving the bioavailability and effectiveness of relevant molecules.4 However, the first
step to utilize these nanomaterials in biomedicinal applications involves its solubilization in
the biological milieu. In order to solubilize SWNTs, exogenous capping units have been
attached to the sideꢀwalls and these functional units primarily govern the properties of the
dispersions.5 Hence, proper exploitation of the SWNTs in the biomedicinal arena lies in the
design and functionality of the inducted motifs within dispersing agents.
The designed nanocarriers must satisfy several prerequisites for their effective
utilization as cellular transporters. Firstly (i) the SWNT needs to be sufficiently dispersed in
water, (ii) the aqueous SWNT dispersion must be satisfactorily viable to mammalian cells,
(iii) preferably it should exhibit stealth properties to escape the immune system, and (iv)
suitable ligands need to be fringed in order to achieve the “active targeting” to specific cells
or tissues. Till date SWNT dispersion has been achieved using various small molecules like
surfactants as well as by macromolecules like polymers and proteins.6 To this end, recently
we reported the efficient SWNT dispersion and successful delivery of proteins inside
3

Journal of Materials Chemistry B
Page 4 of 41
View Article Online
DOI: 10.1039/C3TB21334J
mammalian cells using SWNTꢀamphiphile (comprising of Lꢀalanine and polyethylene glycol
(PEG), Fig. S1, Electronic Supplementary Information (ESI)) composite as vectors.7 PEG
moiety is normally integrated within the dispersing agent or at the side wall of nanotube with
the aim to improve the cell viability of SWNT dispersion as well as to impart stealth
properties within the developed hybrids.8
The other challenging hurdle for any delivery system is targetꢀspecific cellular
transportation. Till date quite a few tumour targeting ligands are known that are integrated
within the designed delivery systems.9 Delivery vehicle equipped with tumorꢀtargeting
ligands recognize the overexpressed cancerꢀspecific receptors on the cell surface and thereby
specifically internalized within cancerous cells. Some typical tumorꢀtargeting motifs include
polysaccharides like hyaluronic acid, monoclonal antibodies, polyunsaturated fatty acids,
EGF (epidermal growth factor), peptide like RGD and vitamins like folic acid and biotin.9a
The CNTꢀbased delivery systems are gaining more significance in this line of research by the
inclusion of targetꢀspecific ligand on its surface. Among all the known targeting markers,
exploration on the specificity of biotin has received comparatively little attention.10 Biotin is
an essential vitamin for the growth of all cells but the demand for this vitamin is considerably
higher in the rapidly dividing cells and tissues. Biotin receptors are even more overexpressed
than the folate receptors in many cancer cells.11 In this regard, Ojima and coꢀworkers reported
the specific delivery of fluorophore tagged SWNT covalently functionalized with biotin to
L1210FR cells.9a Thus, designing of dispersing agents, which are cell viable, antibiofouling
as well as targetꢀspecific towards cancerous cells deserve considerable attention as cellular
transporters.
In the present manuscript we report efficient aqueous dispersion of SWNT using
newly designed biotinylated amphiphile comprising of tyrosine, PEG moiety and Cꢀ16 alkyl
chain (1, Scheme 1). Biotin was integrated in the architecture of the amphiphilic dispersing
4

Page 5 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
agent to achieve specificity towards cancerous cells (Fig. 1). The well dispersed nanotubes
were characterized using spectroscopic and microscopic techniques. Subsequently the cell
viability and media stability in the biological milieu was ensured. The developed biotinylated
amphiphileꢀSWNT dispersions facilitated the specific transportation of fluorescently labelled
oligoneucleotide into different cancerous cells (cervical cancer derived Hela cells and Human
hepatic cancer derived HepG2) in contrast to nonꢀcancerous Chinese Hamster Ovarian
(CHO) and Human embryonic Kidney (HEKꢀ293) cells.12 Most interestingly, targeted killing
of cancer cell was achieved by the specific delivery of doxorubicin loaded on this newly
developed nanoconjugate into Hela cells.
Results and discussion
Synthesis of SWNT-dispersing agent
A common approach to disperse CNTs in water involves nonꢀcovalent functionalization of
CNTs with small amphiphilic molecules.5 In general the hydrophobic end of molecule binds
to the SWNT surface while the hydrophilic part remains oriented towards the aqueous
domain. To this end we have recently reported the synthesis of amphiphilic dispersing agents,
comprising of Lꢀalanine or Lꢀphenyl alanine linked to a Cꢀ16 alkyl chain and a PEG unit at
the hydrophilic terminal (Fig. S1, ESI).7 The SWNT debundling efficiency of those PEG
containing amphiphiles was about 85%. However these nanoconjugates transported
fluorescently labelled proteins into both HepG2 cells and CHO cells with equal efficiency.
Therefore, despite being an efficient delivery system, these nanohybrids lack specificity
between cancerous and nonꢀcancerous cells.
Hence the primary aim of this work is to synthesize a SWNT dispersing agent which
can exhibit significant stability in biological media, high cell viability and target specificity
towards cancer cells. Thus inclusion of an additional targetꢀspecific moiety in the structure of
5

Journal of Materials Chemistry B
Page 6 of 41
View Article Online
DOI: 10.1039/C3TB21334J
dispersing agent is required along with the presence of hydrophobic and hydrophilic
segments. With this aim we choose naturally occurring Lꢀtyrosine as the molecular core
where three way functionalizations can be carried out to synthesize desired dispersing agent,
1. Firstly the Cꢀ16 alkyl chain (hydrophobic/lipophilic segment) was incorporated at the
phenolic –OH terminal by the substitution reaction using corresponding alkyl bromide
(Scheme 1). Followed by the COOH terminal of the amino acid was coupled with the PEG
(550) unit through DCC coupling. The targeting biomarker, biotin was converted to the
corresponding active ester using Nꢀhydroxy succinimide (NHS). This activated intermediate
(NHSꢀbiotin) subsequently reacted with the –NH2 end of the amino acid in presence of dry
pyridine to synthesize the amphiphilic SWNT dispersing agent, 1.
SWNT dispersion and characterization
The SWNT dispersing ability of 1 was tested following the previously reported protocol.6h
About 1 mg of pristine SWNT was taken in 4 mL aqueous solution of 1 (2.5 mg/mL) and
sonicated by both probe and bath sonicator. It was subsequently centrifuged at 2500 g for 90
min to remove the heavier bundled CNTs. The supernatant solution contained dispersed
nanotubes and this SWNTꢀ1 dispersion remained stable for months. The amount of SWNT in
the dispersed state was determined from the calibration plot previously prepared using
SDBS.6h,13 The amphiphile 1 efficiently dispersed SWNT (72%) in water. The van der Waals
interaction between the long alkyl chain of 1 and the SWNT possibly facilitated the
anchoring of former onto the nanotube walls while the hydrophilic PEG unit oriented towards
the hydrophilic domain helped in exfoliation of the CNTs by steric repulsion.14 The
hydrogenꢀbonded water molecules also hinder the reꢀaggregation and precipitation of the
nanotubes. In order to determine the percentage of amphiphile interacting with the nanotubes
and involved in the process of dispersion, the obtained supernatant was centrifuged at 375000
g. The supernatant was discarded and the palette was subjected to Thermogravimetric
6

Page 7 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
Analysis (TGA) (Fig. S2, ESI). The sharp decrease in weight between 200ꢀ400 °C
corresponds to the decomposition of the amphiphile. About 25% residue remains after the
decomposition indicating the amount of SWNT present in the sample. Hence ~ 75% of the
used amphiphilic dispersing agent gets attached to the walls of SWNT.
The stability of any colloidal system is primarily governed byꢀ i) the repulsive force
acting between the particles and ii) the attractive van der Waals attraction that leads to
coagulation and precipitation. The zeta (ζ)ꢀpotential study of a colloidal system has been
extensively used in the literature to determine the stability of such colloidal systems. The ζꢀ
potential provides a good insight into the interplay of the various forces and the overall
stability of the dispersed nanohybrids in solution.6h,14 The ζꢀpotential of SWNTꢀ1 was found
to be ꢀ18.9 mV. In general colloidal solutions having ζꢀpotential values above and below
±15.5 mV is considered to be a stable solution.6h,14 High ζꢀpotential values for SWNT
dispersions are mostly observed where electrostatic interactions are playing a dominating role
on the stability of colloids. Nonetheless, even in the absence of any such charge bearing unit,
the high ζꢀpotential value for SWNTꢀ1 indicates high stability of the debundled nanotubes
possibly due to the repulsive forces between individual nanotubes that was originated from
the steric bulk of the PEG unit.14 Also the hydrogen bonding of oxygenꢀcontaining glycol
chain with water molecules might have prevented the coagulation of the SWNTꢀamphiphile
nanohybrids and facilitated its homogenous dispersion in water.
Characterization of the aqueous dispersion of SWNTꢀamphiphiles was further done
by UVꢀvisꢀNIR spectroscopy. Discernable peaks were observed both in the range of 550ꢀ900
nm (S22 transition) and 800ꢀ1600 nm (S11 transition) for the aqueous dispersion of SWNT-1
(Fig. 2a).5d,m,6h,15 The distinct peaks due to the interband transitions (S22 and S11) between the
mirror image spikes in the density of states of the SWNTs indicate well exfoliation of
nanotubes in water and the unperturbed electronic structure of SWNT. The SWNT dispersion
7

Journal of Materials Chemistry B
Page 8 of 41
View Article Online
DOI: 10.1039/C3TB21334J
was also examined by transmission electron microscopy (TEM). The TEM images of SWNTꢀ
1 revealed highly dispersed nanotubes having an average diameter of 5 nm (Fig. 2b).
Additionally the quality of dispersion was studied using atomic force microscopy (AFM)
(Fig. 2c) and histograms were prepared to estimate the dimension of the dispersed
nanotubes.5m,7 The average diameter of the dispersed nanotubes was between 5ꢀ6 nm while
the average length was between 400ꢀ500 nm (Fig. S3, S4, ESI). Both spectroscopic and
microscopic investigations confirmed the efficient debundling of SWNT in water by the
newly designed amphiphile 1.
Cell viability and stability
The issue of cell viability is an extremely basic prerequisite that needs to be confirmed while
designing any kind of delivery system. Having ensured the quality and quantity of SWNTꢀ1
dispersion, the next important issue was to examine the cell viability of this nanohybrid. The
cell viability of SWNTꢀ1 was tested against HEKꢀ293, CHO, Hela and HepG2 cells using
MTT assay. The different cells were incubated for 48 h with varying concentration of
SWNTꢀ1 and interestingly almost 85% cells were viable after 48 h of incubation up to 25
ꢀg/mL of SWNTꢀ1 (Fig. 3a). Encouragingly, when the concentration of the nanohybrid
increased to 50 ꢀg/mL, the cell viability was still considerably high, 81% in case of Hela,
85% in HepG2, 81% in HEKꢀ293 and to 80% in case of CHO cells. Hence the newly
developed nanohybrid is inferred to be satisfactorily cell viable.
In addition to the cell viability, the SWNT based biological transporter must have
substantial stability in the biological milieu particularly in presence of high salt and protein
content.8 The stability of colloidal suspension depends on the balance between the attractive
van der Waals force and the repulsive electrostatic force.14 Aggregation induced coagulation
and precipitation of the nanomaterials in the biological fluid is a common occurrence and that
often leads to the blocking of capillaries. The instability of colloidal nanoparticles arises due
8

Page 9 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
to nonspecific adsorption of proteins and consequently the replacement of the dispersing
agent resulted in the precipitation of nanomaterials.4b,8a High salt and protein content of the
blood plasma may become a major obstacle towards actual implementation of the colloidal
delivery system.1a,8,16 The stability of the newly developed nanohybrid was visually assessed
in phosphate buffer saline (PBS) with increasing ionic concentrations (0ꢀ150 mM) for 48 h
using 25 ꢀg/mL of SWNTꢀ1. Notably, the nanoconjugate was found to be highly stable (Fig.
3b) even at 150 mM concentration of PBS. The suspension stability index (SSI) (details given
in ESI) of all the incubations was measured and no decrease in the value was observed after
48 h of incubation (Fig. S5, ESI). The uncharged hydrophilic PEG unit with its steric bulk
presumably hinders the formation of electrical doubleꢀlayer on the surface of the colloidal
particles that is known to reduce the repulsive forces.14
Furthermore we tested the stability of SWNT dispersions in serum mimicking
compositions having cell culture media (DMEM) and varying concentration of fetal bovine
serum (FBS) protein. It is known that nonꢀspecific adsorption of biological components of
plasma on nanotube surface particularly in case of nonꢀcovalent CNTꢀdispersion, could result
in the uptake of the foreign materials by the macrophages of the reticuloendothelial system
(RES).1a,8,16 Thus, SWNTꢀ1 (25 ꢀg/mL) was incubated in DMEM media containing FBS (0ꢀ
100%) protein for 48 h under sterile condition. Encouragingly, no precipitation of the
nanomaterial was observed even when the protein concentration was increased to 100% FBS
(Fig. 3c). Additionally the SSI of the solutions was found to be above 90% for all the
solutions after 48 h (Fig S6, ESI). Again the neutral PEG unit possibly acts as a protection
layer around the colloidal entities because of having hydration around itself (stealth
character) and resists direct interaction of the proteins and ions with the nanomaterial in the
biological system.4b,8a,e,f These observations also indicate that the attachment of the
amphiphilic dispersing agent (1) on SWNT surface is highly stable and chances of surfactant
9

Journal of Materials Chemistry B
Page 10 of 41
View Article Online
DOI: 10.1039/C3TB21334J
exchange in the blood plasma is very poor. Furthermore circulation of nanoconjugate within
body fluids is also essential for the sustained delivery of cargo inside the cells. To this end,
stability of SWNTꢀ1 (25 ꢀg/mL) was also investigated for ten days in 10% FBSꢀDMEM
culture media (condition normally used for the proliferation of cells). SWNTꢀ1 dispersion
was found to be stable for the entire period of time and SSI was above 95% at the end of ten
days (Fig. S7, ESI). This highly dispersed SWNTꢀ1 nanoconjugate along with its
considerable cell viability and remarkable stability in biological milieu meets all the essential
criteria that required for a cellular transporters.
Oligonucleotide loading on dispersed SWNT
The loading capacity of the nanohybrid was investigated prior to its utilization as cellular
transporters. 15ꢀmer oligonucleotide = ACCTGTGAGCCTTAC (Olg) was chosen as the
cargo and the percentage of oligonucleotide loaded on to SWNTꢀ1 surface was calculated
using agarose gel electrophoretic study. The complexation of oligonucleotide with SWNT
inhibits the intercalation of ethidium bromide (EtBr) with the oligonucleotide thereby EtBr
cannot exhibit any fluorescence emission. Hence it is not possible to visualise the complexed
oligonucleotide with SWNT, instead the free oligonucleotide was estimated to calculate the
loading capacity of the nanoconjugates.17 The migration behaviour of the different SWNTꢀ1-
Olg composites (having varied concentration ratio of SWNTꢀ1 to oligonucleotide) was
studied against uncomplexed oligonucleotide (30 ꢁM) (Fig. 4a,b). The observed fluorescent
band in lane 1 originated due to the intercalation of EtBr with the free oligonucleotide and
other three lanes (2 to 4) were loaded with same amount of oligonucleotide premixed with 25,
50 and 100 ꢁg/mL of SWNTꢀ1, respectively (Fig. 4a). In order to quantify the fluorescence
intensity of EtBr intercalated with unbound oligonucleotide, densitometric analysis was
carried for the luminescent zone (black region just above the luminescent zone in each lane
was due to the presence of bromophenol blue). Sequential decrease in the fluorescence
10

Page 11 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
intensity from lane 2 to 4 indicates a decrease in the amount of available free oligonucleotide
or increase in the complexed oligonucleotide with increase in the concentration of
nanoconjugate (Fig. 4a,b). It was found almost 75% oligonucleotide remained in the unbound
state when 25 ꢀg/mL SWNTꢀ1 was incubated with 30 ꢀM oligonucleotide i.e. 25% of the
cargo (7.5 ꢁM) was loaded on to the nanotube surface. The percentage of loaded
oligonucleotide sequentially increased to 31 and 45% when the concentration of SWNTꢀ1
was increased to 50 and 100 ꢁg/mL, respectively.
The loading of oligonucleotide onto the nanotubes was also studied using
fluorescence spectroscopy. For this purpose a stock solution (1.25 ꢁM) of fluorescently
labelled oligonucleotide (Cy3ꢀACCTGTGAGCCTTAC = Cy3ꢀOlg) was prepared in water
and the fluorescence spectrum of this solution exhibited maximum emission at 565 nm (Fig.
4c) upon exciting the probe at 550 nm. Hereafter SWNTꢀ1 (50 ꢁg/mL) was mixed with this
Cy3ꢀOlg solution for 4 h to load the oligonucleotide on the surface of dispersed nanotubes.
The aqueous dispersion was centrifuged at 175000 g to precipitate out the Cy3ꢀOlg loaded
SWNTꢀ1. The fluorescence spectrum of the supernatant was then recorded which showed
significant drop in the emission intensity compared to that of native Cy3ꢀOlg (Fig. 4c). This
dramatic decrease in the emission intensity clearly suggests the absence of free Cy3ꢀOlg in
the supernatant solution. Therefore almost complete loading of Cy3ꢀOlg has taken place on
the surface of nanotube leading to the formation of SWNTꢀ1ꢀCy3ꢀOlg nanobioconjugate. We
also estimated the amount of loaded cargo on the nanovectors by using the UVꢀvis absorption
of Cy3ꢀOlg at 540 nm. A calibration plot of Cy3ꢀOlg was prepared by sequential dilution of
its stock solution (Fig S8, ESI). The conjugate was then prepared by incubating 50 ꢀg/mL
SWNTꢀ1 with1.25 ꢀM Cy3ꢀOlg for 4 h. The hybrid was centrifuged and washed twice at
175000 g till no unbound oligoneucleotide was present in the supernatant. The obtained
palette was reꢀdispersed in 1 wt % (w/v) of SDS solution and sonicated. The dispersed
11

Journal of Materials Chemistry B
Page 12 of 41
View Article Online
DOI: 10.1039/C3TB21334J
solution was again centrifuged and the supernatant was collected. The absorbance was
measured and the concentration of the Cy3ꢀOlg in the supernatant was found to be 1.12 ꢀM .
The addition of the anionic surfactant SDS led to the removal of the bound oligoneucleotide
from the surface of the nanotubes which resulted in the increased absorbance of the
supernatant.17b Thus the experiment further showed almost complete loading of the
oligoneuclotide on the nanotube surface.
Target-specific cellular internalization
After ensuring the cell viability, media stability and loading capacity of the SWNTꢀ1
nanohybrid, the vector was used for transporting the fluorescent probe labelled Cy3ꢀOlg
inside the mammalian cells. Cy3ꢀOlg was nonꢀcovalently loaded onto the surface of
dispersed nanotubes by stirring the oligonucleotide in the aqueous dispersion of SWNTꢀ1 in
dark for 4 h (see experimental section for details). The cargo loaded nancomposites were then
incubated with cancerous Hela and HepG2 cells for 3 h at 37 °C in 5% CO2 atmosphere. Both
cells have overexpressed biotin receptors on their surface.3a,9a,10,11b,c The typical concentration
of the delivering unit (SWNTꢀ1) was 10 ꢀg/mL and that of the cargo (Cy3ꢀOlg) was 0.25
ꢀM. In accordance with the binding affinity pattern of oligonucleotide towards the dispersed
SWNTꢀ1 (Fig. 4), it is expected that almost complete loading of oligonucleotide has taken
place on the surface of SWNT. The incubation was followed by washing of the cells with
PBS buffer in order to remove the excess conjugate from the medium. Encouragingly,
successful internalization of oligonucleotide inside the cancerous cells by SWNTꢀ1 was
evident from the highly intense red fluorescence microscopic images (Fig. 5aꢀb, dꢀe) of Hela
and HepG2 cells. Also the corresponding flow cytometric analysis of the cancerous cells
further confirmed the successful internalization of Cy3ꢀOlg inside cancerous cell (Fig. 5c,f).
The healthy nature of the cells was ascertained from their spindle shape in the corresponding
bright field images. The viability of the cells was confirmed by incubating both Hela and
12

Page 13 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
HepG2 cells with 10 ꢀg/mL SWNTꢀ1 for 3 h followed by treating with the LIVE/DEAD
viability kit (Fig. S9 a,b,d,e, ESI). The predominant presence of green cells and absence of
red cells in fluorescence images further validated the existence of living cells as well as
significant cell viability of the nanohybrid. The cell viability was also studied using flow
cytometry (Fig S9 c,f) where the significant presence of green cells and clear absence of red
stained cells indicate the healthy nature of the cells. The selfꢀinternalization ability of Cy3ꢀ
Olg (0.25 ꢁM) alone within Hela and HepG2 cells in absence of SWNTꢀ1 was studied
keeping other experimental conditions same (details given in ESI). Absence of almost any red
fluorescent cells confirmed the inability of selfꢀinternalization of Cy3ꢀOlg without the
assistance of nanoconjugate (Fig. S10, ESI). The flow cytometric analysis also confirmed the
poor self internalisation ability of the cargo (Fig. S11, ESI). The transportation ability of the
amphiphilic dispersing agent (1) through formation of any supramolecular aggregate was also
investigated. For this purpose the critical micellar concentration (CMC) of 1 was measured
tensiometrically and it was found to be 14.3 ꢁg/mL (Fig. S12, ESI). The solutions of 1 (10
ꢁg/mL) and (50 ꢀg/mL) was mixed with Cy3ꢀOlg (0.25 ꢁM) and incubated with Hela and
HepG2 cells under identical experimental conditions for 3h at 37 °C. The amphiphile
concentration was kept well above and below the CMC to ascertain the presence and absence
of its selfꢀaggregated structure during the experiment. However, no significant internalization
of the fluorescent labelled oligonucleotide was observed (Fig. S13, ESI). The above
observation was also confirmed using flow cytometric experiments and the low mean
fluorescence intensity indicated the poor internalisation ability of the cargo by formation of
other supramolecular aggregates (Fig S11, ESI). Hence, SWNTꢀ1 nanoconjugate exclusively
transported the cargo inside the cancer cells.
At this point, the most crucial issue is to investigate the specificity of SWNTꢀ1
towards cancerous cells. The specificity of the developed nanohybrid was evaluated by
13

Journal of Materials Chemistry B
Page 14 of 41
View Article Online
DOI: 10.1039/C3TB21334J
investigating the cargo delivery ability of SWNTꢀ1 into CHO cells and HEKꢀ293 cells,
devoid of overexpressed biotin receptors. Similar to the preceding experiments for cancerous
cells, here also Cy3ꢀOlg (0.25 ꢀM) loaded SWNTꢀ1 (10 ꢁg/mL) was incubated with CHO
cells and HEKꢀ293 cells for 3 h at 37 °C. Almost complete absence of red fluorescent cells in
the corresponding microscopic image (Fig. 5g,h,j,k) confirmed the nonꢀdelivery of Cy3ꢀOlg
into these cells. Similarly, the corresponding flow cytometry data showed poor fluorescent
intensity (Fig. 5i,l). Thus the extent of delivery of the oligoneucleotide in normal cells was
significantly lower than that was observed for Hela and HepG2 cells. Moreover, the receptorꢀ
mediated specific transportation of cargo in cancerous cells was examined by preꢀtreating the
cells with ligand. The Hela cells were preꢀincubated with biotin solution (2 mM) for 1 h and
these treated cells were incubated with the Cy3ꢀOlg loaded SWNTꢀ1 for 3 h at 37 °C.9a The
extent of cargo internalization was notably lower as observed in the microscopic images (Fig.
5,m,n) compared to that was observed for biotinꢀuntreated Hela cells. Corresponding flow
cytometry data affirmed the decreased cargo internalisation in the treated cells (Fig 5o). The
preꢀtreatment with the ligand probably led to the blocking of the overexpressed receptors in
the cancer cells resulting in the reduced internalization. Also the extent of delivery was
checked in Hela cells using the Cy3ꢀOlg loaded SWNT dispersed by previously reported
PEGylatedꢀamino acid based amphiphile (Fig S1, ESI) devoid of biotin. Interestingly the
corresponding fluorescence image and the flow cytometry data revealed significantly lower
transportation after 3 h of incubation (Fig 5 pꢀr). The targetꢀspecific transportation takes place
due to the presence of overexpressed biotin receptors in the rapidly dividing cancerous cells
and tissues.3a,9a,10,11 Thus, the overexpression of biotin receptors on Hela and HepG2 cells
surface and lack of the same on CHO cell and HEKꢀ293 cell made the biotinylated
amphiphileꢀSWNT nanohybrid targetꢀspecific toward cancerous cells.
14

Page 15 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
The cargo internalization experiments for Cy3ꢀOlg loaded SWNTꢀ1 were again
carried out in CHO cells and HEKꢀ293 cells at a longer time scale (from 6 to 48h). However
after 48 h of incubation, detectable delivery of cargo was observed into CHO cells and HEKꢀ
293 cells (Fig. S14,a,b,d,e, ESI). The similar pattern of uptake is also evident from the
enhanced fluorescence intensity in the flow cytometry data of CHO and HEKꢀ293 cells after
48 h of incubation (Fig S14, c,f). Hence, SWNTꢀ1 hybrid has at least 48 h specificity towards
Hela and HepG2 cancerous cells compared to that of CHO cells and HEKꢀ293 cells. The
viability of the CHO and HEKꢀ293 cells were also examined using LIVE/DEAD viability kit
after 48 h of incubation with SWNTꢀ1. Here too, the absence of red cells and presence of
green cells affirmed the substantial viability of the mammalian cells (Fig. S15,a,b,d,e, ESI).
Almost complete absence of red cells was also confirmed from the flow cytometry (Fig S15
c,f, ESI). Thus by way of rational designing we have been able to develop simple surfactant
based efficient SWNT dispersing agent, which is significantly biocompatible and has
achieved targetꢀspecific transportation of biomolecules into cancerous cells.
Internalization pathway mechanism
In view of the successful targetꢀspecific transportation of biomolecule into cancerous cells by
biotinylated amphiphileꢀSWNT dispersion we were curious to investigate the mechanism of
internalization. The exact mechanism of cellular internalization of SWNT is still a matter of
serious debate.9a The two mostly discussed mechanisms of cellular uptake are i) CNT act as
nanoneedles and pierce through the cell membrane and (ii) the other mechanism involves
endocytotic pathway. Generally, the former is an energy independent process while the latter
is energy dependent process, which is hindered at low temperature or in the presence of a
metabolic inhibitor, such as NaN3. To probe the mechanism of uptake, cellular transportation
was carried out under energy depleted condition keeping all other experimental conditions
identical (details given in SI). Hela cells were preꢀtreated with 10 mM solution of NaN3 for
15

Journal of Materials Chemistry B
Page 16 of 41
View Article Online
DOI: 10.1039/C3TB21334J
30 min to ensure energy depleted condition. This was followed by incubation of those cells
with 10 ꢀg/mL SWNTꢀ1ꢀCy3ꢀOlg (0.25 ꢀM) for 3 h at 37 °C. Microscopic imaging of the
cells revealed significant decrease in the nanoconjugates uptake (Fig. S16 a,b, ESI) since no
detectable red fluorescent cell was observed. Similarly, Hela cells were incubated with the
same SWNTꢀ1ꢀOlg for 3 h at 4 °C instead of 37 °C and the very poor intensity of red
fluorescence in microscopic images delineated substantial drop in the internalization of
nanohybrid (Fig. S16,d,e, ESI). The flow cytometric analysis also indicated the poor
transportation of the oligoneucleotide under energy depleted conditions (Fig S16, c,f). The
viability of the cells after treatment with NaN3 was checked using LIVE/DEAD assay and the
predominance of green cells indicated the presence of live cells (Fig. S17,a,b ESI). The flow
cytometric analysis also confirmed the viability of the azide treated cells (Fig S17 c, ESI).
Hence the biological transportation by SWNTꢀ1 predominantly followed the energy
dependent receptorꢀmediated endocytotic pathway.
This biological transportation of oligonucleotide inside cancer cell by SWNTꢀ1 was
further studied by nucleus staining experiments. This was done to counter stain the nucleus
and clearly demarcate the cytoplasm from the nucleus of cell. The Cy3ꢀOlg internalized Hela
cells were subsequently incubated with nucleus staining agent Hoechst dye (2.5 ꢀg/mL) for
30 min. The blue colour originates from the intercalation of Hoechst dye into the nuclear
oligonucleotide (Fig. 6a) and the red coloured region corresponds to the already internalized
Cy3ꢀOlg (Fig. 6b). The dual colour in the superimposed image (Fig. 6c) clearly showed the
accumulation of red Cy3ꢀOlg in the cytoplasmic region and blue Hoechst dye in the nucleus
with little coꢀlocalization of Cy3ꢀOlg. However after 24 h incubation of SWNTꢀ1ꢀCy3ꢀOlg
with Hela cells, microscopic images clearly revealed significant localization of the cargo
inside the nucleus (Fig. S18, ESI), which was also supplemented by nucleus staining
16

Page 17 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
experiments. Thus, this newly developed nanovector can also be utilized for targeted
gene/drug delivery in biotin overexpressed cancer cells.
Target-specific drug delivery
Receptorꢀmediated successful delivery of oligonucleotide into cancerous cells prompted us to
utilize this newly developed nanovector for clinically important targeted drug delivery to
cancer cells. For this purpose a well known anticancer drug, doxorubicin was chosen. The
poor solubility and lack of specificity are some of the primary problems associated with the
drug administration. Hence with the aim of targetꢀspecific delivery of doxorubicin to cancer
cells, the drug was hydrophobically loaded onto the walls of the nanohybrid (SWNTꢀ1). This
was done by overnight incubation of the drug with 100 ꢀg/mL SWNTꢀ1 in pH 8.0 (PBS
buffer) to maintain an optimal weight ratio of 2:1 (SWNT:doxorubicin).18 The loading of the
drug was ensured from the UVꢀvis absorbance spectra where doxorubicin itself exhibits a
UVꢀVis absorbance maxima at 492 nm (Fig. 7a). Upon loading of this chromophoric drug on
SWNT surface, a new peak was observed for the SWNTꢀ1ꢀDOX hybrid having an
absorbance maxima at 502 nm, which was absent in the spectra of the SWNT alone.18 This
red shifted peak of doxorubicin in the hybrid material clearly indicates its interaction with the
walls of the nanotube and loading of the same.18b,6j To further ensure the loading of
doxorubicin on the surface of nanohybrid, the photoluminescence spectra of the drug were
recorded (Fig. 7b). The spectrum of free doxorubicin showed an emission maximum around
600 nm upon excitation at 490 nm. Interestingly, almost complete quenching of this peak was
noted for the drug loaded SWNTꢀ1 composite indicating the absence of free doxorubicin in
the sample.18a Thus the SWNT backbone of the nanovector allowed the binding of the drug
possibly through interaction between aromatic groups and efficient loading of the
doxorubicin took place leading to the formation of the targetꢀspecific vector, SWNTꢀ1ꢀDOX.
17

Journal of Materials Chemistry B
Page 18 of 41
View Article Online
DOI: 10.1039/C3TB21334J
The loading of doxorubicin was further confirmed by measuring the amount of
released doxorubicin at low pH. The drug is known to have higher hydrophilicity at lower pH
due to the protonation of the ꢀNH2 leading to lower affinity towards hydrophobic SWNT
surface.6j The chromophoric drug exhibits a maxima in its UVꢀvis spectra at 490 nm hence a
calibration plot was prepared by sequential dilution at 470 nm (Fig S19, ESI). Doxorubicin
(50 ꢀg/mL) was loaded onto the surface of the nanotube by incubating with 100 ꢀg/mL
SWNTꢀ1 at pH 8.0 buffer. The conjugate was centrifuged at 175000 g to remove the unbound
doxorubicin and the obtained palette was reꢀdispersed in 1 N HCl to detach the loaded
doxorubicin. The solution was again centrifuged and the amount of doxorubicin released in
the supernatant was found to be 46 ꢀg/mL. Hence the study vividly showed almost complete
the loading of the drug on the surface of dispersed SWNT.
The targeted drug delivery of doxorubicin by this nanohybrid (SWNTꢀ1) was then
examined by the incubation of cancerous Hela cells and nonꢀcancerous CHO cells with the
SWNTꢀ1ꢀDOX for 12 h. The concentration of SWNTꢀ1ꢀDOX was varied over a range and
the maximum concentration of the drug was kept at 10 ꢀg/mL and that of the vector SWNTꢀ1
was at 20 ꢀg/mL. The percentage of live cells after the incubation was determined using
MTT assay and encouragingly a marked difference in the killing efficiency was observed. At
2.5 ꢀg/mL of doxorubicin and 5 ꢀg/mL of SWNTꢀ1, 92% CHO cells were alive while only
46% Hela cells were found to be dead (Fig. 7c). Importantly, when the concentration of the
hybrid was increased to 10 ꢀg/mL of doxorubicin and 20 ꢀg/mL of SWNTꢀ1, more than 60%
Hela cells were killed in contrast to that of only 15% for CHO cells. Alongside, the efficiency
of native doxorubicin was tested and it was observed that the drug alone had very poor killing
efficiency. Almost 80% Hela cells and 85% CHO cells were alive when 10 ꢀg/mL drug was
incubated for 12 h and the IC50 (Half Inhibitory Concentration) was >10ꢀg/mL (Fig S20,
ESI). Most interestingly this IC50 value decreased to 7.5 ꢀg/mL when the drug was used in
18

Page 19 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
combination with SWNTꢀ1. Also the marked difference in the killing ability of doxorubicin
between Hela and CHO cells can only be attributed to the targetꢀspecific drug delivery by
SWNTꢀ1 induced by the tumour targeting agent, biotin.
Conclusion
In summary, the present work described a simple approach to develop a SWNT based
cellular delivery vector in combination with amino acid based biotinylated amphiphile. The
spectroscopic and microscopic studies indicate the superior dispersing ability of the newly
developed amphiphilic dispersing agent. This amino acid based amphiphile comprising of a
PEG unit ensured the low cytotoxicity and significant stability in media containing varying
concentration of salt and protein. Most encouragingly the incorporation of biotin in the
amphiphile’s structure facilitated the specific delivery of Cy3ꢀOlg and doxorubicin loaded
SWNTꢀ1 conjugate into cancerous cells due to the overexpressed biotin receptors in these
rapidly dividing cells. Thus by way of rational designing of amphiphilic molecule we have
achieved targetꢀspecific delivery of biomolecule and drug into cancerous cells using nonꢀ
covalently modified SWNT. Such ‘specificꢀtargeting’ model can be used to develop effective
delivery platforms in biomedicine.
Materials and Methods
Materials
Silica gel of 60ꢀ120 mesh, Lꢀtyrosine, cetyl bromide, N,Nꢀdicyclohexylcarbodiimide (DCC),
4ꢀN, Nꢀ(dimethylamino)pyridine (DMAP), 1ꢀhydroxybenzotriazole (HOBT), Dꢀbiotin,
sodium dodecylbenzene sulfonate, Nꢀhydroxysuccinimde (NHS), pyridine, dimethyl
formamide (DMF), agarose for gel electrophoresis, ethidium bromide, solvents and all other
reagents were procured from SRL, India. thionyl chloride, and sodium hydroxide were
purchased from Spectrochem, India. Glycerol was obtained from Merck. Water used
19

Journal of Materials Chemistry B
Page 20 of 41
View Article Online
DOI: 10.1039/C3TB21334J
throughout the study was MilliꢀQ water. Single walled carbon nanotubes (SWNT, 1.2ꢀ1.5 nm
diameter), CDCl3 for NMR experiments, polyethylene glycol (PEG, Mn = 550) and
bromophenol blue, sodium azide were obtained from Aldrich Chemical Co. The
LIVE/DEAD® viability kit (Lꢀ3224) for eukaryotic cells, Dulbecco’s Modified Eagles’
Medium (DMEM), heat inactivated fetal bovine serum (FBS), Hoechst dye, trypsin from
porcine pancreas and penicillinꢀstreptomycin was procured from Molecular Probes,
Invitrogen Chemical Company. Thiazolyl blue tetrazolium bromide (MTT), Cy3 labelled
oligonucleotide and oligonucleotide were obtained from Sigma Aldrich Chemical Company.
Perkin Elmer Lambda 25 spectrophotometer was used to record the UVꢀVIS spectra while
1
the UVꢀVisꢀNIR spectra were monitored using Varian Cary 5000 spectrophotometer. H
NMR spectra were recorded in an AVANCE 500 MHz (Bruker) spectrometer. Fluorescence
emission spectra were recorded in Varian Cary Eclipse Luminescence spectrometer at
excitation and emission slit widths of 2.5 nm. Transmission Electron Microscopy (TEM)
measurements were performed on JEOL JEM 2010 microscope. Atomic force microscopy
(AFM) was performed on Veeco, model AP0100 microscope in nonꢀcontact mode. Probe
sonication was done using Omni Sonic Ruptor 250. Bath sonication was performed with a
Telsonic Ultrasonics bath sonicator. Sorvall RC 6 and Sorvall RC 90 were used for
centrifugation and ultracentrifugation, respectively. Bio Rad Gel Doc XR was used for the
imaging of agarose gels. Olympus IX51 inverted microscope was used for imaging of the
cells. Cellꢀassociated fluorescence was monitored on fluorescence activated cell sorter
(FACS) Calibur flow cytometer (Becton Dickinson, San Jose, CA) with the excitation
wavelength at 560 nm.
Synthesis of 1
Complete synthetic procedure and reagents are delineated in Scheme 1. At first the ꢀCOOH
terminal of Lꢀtyrosine was esterified using thionyl chloride in methanol. The methyl ester
20

Page 21 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
protected amino acid was then taken in DCM and drop wise BOCꢀanhydride was added in the
presence of triethylamine. The reaction mixture was then taken in ethyl acetate (EtOAc) and
washed well with 1 N HCl to remove the free amine. The organic layer was then washed to
neutrality using brine and evaporated to get the Bocꢀprotected methyl ester of Lꢀtyrosine (I).
The intermediate, I was then taken in dry DMF to which cetyl bromide (1.5 equivalent) and
equivalent amount of K2CO3 were added. The solution was stirred at 80 °C for 24 h in an oil
bath. EtOAC was added to the solution and washed with water to remove DMF. The organic
layer was dried over sodium sulphate and column chromatography was performed in 60ꢀ120
mesh silica gel using chloroform/methanol as the eluent to obtain Cꢀ16 alkylated Lꢀtyrosine
at the –OH terminal (II).
Compound II was then dissolved in 3:1 methanol/chloroform (v/v) solution and
stirred overnight with 1.1 equivalent of NaOH to hydrolyse the ester linkage. The organic
solvents were removed and the reaction mixture taken in EtOAc was washed with excess
dilute HCl to ensure conversion of the sodium salt of III to the neutral carboxylic acid. The
organic layer was dried over sodium sulphate and evaporated in a rotary evaporator. The
white solid residue was then purified by column chromatography using 60ꢀ120 mesh silica
gel and chloroform/methanol as the eluent to obtain free –COOH terminal of Oꢀalkylated and
NꢀBoc protected tyrosine (III).
Acid terminal of intermediate III was then coupled with PEG using DCC (1 equiv)
and a catalytic amount of DMAP in the presence of 1 equiv of HOBT in dry DCM. The
amide was then purified through column chromatography using 60ꢀ120 mesh silica gel and
chloroform/methanol as the eluent to obtain PEG linked IV. Intermediate IV was then
subjected to deprotection by trifluoroacetic acid (1.5 equivalent) in dry DCM. After being
stirred for 2 h, solvents were removed on a rotary evaporator and material was dried in a
21

Journal of Materials Chemistry B
Page 22 of 41
View Article Online
DOI: 10.1039/C3TB21334J
vacuum pump. It was then dissolved in MeOH and treated with solid sodium carbonate. The
organic part was filtered and concentrated to get the corresponding free amine (V).
NHS linked biotin was prepared by taking Dꢀbiotin (500 mg), DCC (1.5 equivalent)
and NHS (1.1 equivalent) in dry DMF (8 mL) and stirred overnight under N2 atmosphere. To
this activated biotin (NHSꢀbiotin) mixture, compound V and dry pyridine was added. The
solution was stirred overnight and the DMF was distilled out under vacuum. The residue
mixture was then purified using 100ꢀ200 mesh silica gel and chloroform/methanol as the
eluent to obtain pure 1.
Characterization of dispersing agents
1H NMR of 1 (500 MHz, CDCl3, Me4Si, 25 °C)
δ = 0.82ꢀ0.85 (m, 3H), 1.22ꢀ1.40 (m, 28H),
1.59ꢀ1.73 (m, 6H), 2.16ꢀ2.18 (m, 2H), 2.85ꢀ3.10 (m, 5H), 3.34 (s, 3H), 3.50ꢀ3.69 (m, 46H),
3.85ꢀ3.88(m, 2H), 4.23ꢀ4.24 (m, 2H), 4.29ꢀ4.31 (m, 1H), 4.49ꢀ4.50 (m, 1H), 4.78ꢀ4.80 (m,
1H), 6.75ꢀ6.77 (d, 2H), 7.01ꢀ7.02 (d, 2H). MS (ESI): m/z calculated for C60H107O17N3S:
1173.71; found: 1174.7133 [M + H+].
Preparation of SWNT-amphiphile dispersion
To an aqueous solution (4 mL) of 1 (2.5 mg/mL) SWNT (1 mg) was added. The solution was
tipꢀsonicated for 30 min and bath sonicated for 2 h and again tip sonicated for 30 min. To
remove the heavy bundles the solution was then centrifuged at 2500 g for 90 min. The
amount of the dispersed SWNTs in the supernatant was calculated from the observed
absorbance value at 550 nm that was derived from the previously reported calibration plot of
absorbance versus concentration using sodium dodecylbenzene sulfonate.6h The percentage
dispersion was calculated as the ratio of the amount of SWNTs in the dispersion to the
amount of SWNTs initially added.
22

Page 23 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
TGA analysis:
The nonꢀadsorbed amphiphile in the SWNT dispersion was removed by centrifugation. The
obtained palette was dried in a vaccum oven. TGA thermogram of SWNTꢀ1 composite was
recorded by using a TA SDT Q600 instrument at a heating rate of 20 °C min under a N2
atmosphere.
Sample preparation for ξ-potential, UV-vis-NIR, TEM and AFM
The aqueous SWNT suspension obtained after centrifugation at 2500 g was ultracentrifuged
at 375000 g for 15 min. The supernatant was discarded to remove excess amphiphile 1 and
the palette was redispersed in water (SWNTꢀ1). The aqueous dispersion obtained was used
for ζ ꢀpotential, UVꢀvisꢀNIR, TEM and AFM experiments. A drop of the SWNT suspension
was placed on a 300ꢀmesh Cuꢀcoated TEM grid and dried under vacuum for 4 h before taking
the image. Similarly in case of AFM studies, a drop of the dispersion was cast on a freshly
cleaved mica surface and the samples were airꢀdried overnight before imaging. The bundle
diameter was calculated from the height profile of the nanotubes. From 20 AFM images of
SWNTꢀ1 a statistical analysis of the bundle diameter and nanotue length was done by plotting
histogram.
Cell culture
HepG2 and Hela cells were procured from National Center for Cell Science (NCCS), Pune
(India), and HEKꢀ293 and CHO cells were procured from ATCC and cultured in 10% FBS
DMEM medium in presence of 100 mg/L streptomycin and 100 IU/mL penicillin. The cells
were maintained in a 25 mL cell culture flask and kept at 37 °C in a humidified atmosphere
of 5% CO2 to about 70ꢀ80% confluence. Subculture was performed in every 2ꢀ3 days. The
media was changed after 48ꢀ72 h to remove the dead cells. Subsequently the adherent cells
23

Journal of Materials Chemistry B
Page 24 of 41
View Article Online
DOI: 10.1039/C3TB21334J
were detached from the surface of the culture flask by trypsinization. These cells were then
used for cytotoxicity study as well as for internalization experiments of the SWNTꢀ1
conjugates.
MTT Assay
Cell viability of the nanoconstructs was assessed by the microculture MTT reduction assay as
previously reported.19 The assay involves the conversion of a soluble tetrazolium salt by
mitochondrial dehydrogenase present in live cells to an insoluble colored product, formazan.
The amount of formazan formed is estimated spectrophotometrically after dissolution of the
product in DMSO. The amount of the formazan produced is proportional to the number of
alive cells. The mammalian cells were seeded at a density of 15,000 cells per well in a 96ꢀ
well microtiter plate18ꢀ24 h before the assay. A stock solution of the SWNTꢀamphiphile
dispersion was prepared in sterile water. Concentration of the nanohybrid in the microtiter
plate was varied from 5 to 50 ꢂg/mL. The cells were incubated with the nanoconstructs for 48
h at 37 °C under 5% CO2. The cells were further incubated for another 4 h in 15 ꢂL MTT
stock solution (5 mg/mL). The produced formazan was dissolved in DMSO and absorbance
at 570 nm was measured using BioTek® Elisa Reader. The number of surviving cells were
expressed as percent viability = (A570(treated cells)ꢀbackground)/ A570(untreated cells)ꢀ
background) × 100.
Gel Electrophoresis
The gel for the electrophoresis was prepared by dissolving agarose (600 mg) in 20 mL of 1X
Trisꢀboric acidꢀEDTA buffer. 2 ꢁL of ethidium bromide (1 mg/mL) was added in that
agarose solution during the preparation of the gel. A stock solution of SWNTꢀ1 was prepared
as mentioned above and that was diluted to prepare varying concentrations (25 ꢁg/mL, 50
ꢁg/mL and 100 ꢁg/mL) of solutions of (100 ꢁL) of the nanoconjugate. The concentration was
24

Page 25 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
determined using calibration plot. These solutions having varying concentration of SWNTꢀ1
were separately incubated with 30 ꢁM oligonucleotide solution (in TrisꢀEDTA pH=8 buffer)
bearing the sequence ACCTGTGAGCCTTAC for 4 h at room temperature. Next the samples
were mixed with 1 ꢁL bromophenol blue (to track the oligonucleotide) taken in glycerol and
loaded on to the above prepared agarose gel. Only oligonucleotide (30 ꢁM) was also loaded
onto the same gel as a standard. The gel was run at E = 85 Vcmꢀ1 for 30 min and
photographed by irradiating UV light. The band intensities were calculated with “quantity
one” 1ꢀD analysis software (BioꢀRad, Hercules, CA, USA).
Fluorescence Study
SWNTꢀ1 (50 ꢁg/mL) was mixed with fluorophore labeled Cy3ꢀoligonucleotide (1.25 ꢁM) for
4 h. At first the fluorescence intensity of only Cy3ꢀOlg (1.25 ꢁM) was measured by
excitation at 550 nm and recording the 555ꢀ700 nm. Then the solution containing SWNTꢀ1ꢀ
Cy3ꢀOlg was centrifuged at 175000 g to precipitate out SWNTꢀ1ꢀOlg conjugate as palette.
Next the fluorescence intensity of the supernatant was recorded by exciting at same
wavelength.
Preparation SWNT-1-Olg Conjugates
SWNTꢀ1 (50 ꢂg/mL) was mixed with the Cy3 tagged Olg (1.25 ꢂM) and stirred for 4 h at
room temperature to achieve the loading of the Cy3ꢀOlg onto the SWNT surface. This
SWNTꢀ1ꢀCy3ꢀOlg nanobioconjugate was diluted according to the requirement before
incubation with different cells.
Incubation, Delivery and Imaging of Cells
Cy3ꢀOlg loaded SWNTꢀ1 suspensions were added into 24 well chambered plates containing
confluent Hela, HepG2, HEK 293 and CHO cells in 10% FBSꢀDMEM culture media. The
25

Journal of Materials Chemistry B
Page 26 of 41
View Article Online
DOI: 10.1039/C3TB21334J
concentration of SWNT in each well was 10 ꢂg/mL and the Cy3ꢀOlg concentration was fixed
at 0.25 ꢂM. The final volume of the solution was 250 ꢂL in each well. The
nanobioconjugates were incubated with these cells for 3 h at 37°C in 5% CO2 atmosphere.
The cells were repeatedly washed with PBS buffer to ensure complete removal of the
extracellular hybrids. The cells were then observed under the Olympus IX51 inverted
microscope using an excitation filter BP530ꢀ550 and a band absorbance filter covering
wavelength below 570 nm. Bright red images were observed when SWNTꢀ1ꢀOlg was
internalized inside the cells. Similar cargo delivery experiments were carried out for HEKꢀ
293 and CHO cells in longer time scale (6ꢀ24 h) keeping all other conditions identical. The
images were taken at 10x magnification.
Flow Cytometry
Required treated cells were suspended in 0.3 mL of PBS. Cells were analyzed using a flow
cytometer, FACS Aria, operating at 560 nm excitation wavelength and at 530±30 nm
detecting emission wavelength using a 500ꢀ570 nm bandpass filter for Cy3ꢀOlg. The data was
acquired and analyzed by ‘Cell Quest’ proꢀsoftware (Becton Dickinson, USA). Initially
million cells were taken and the data was acquired from 10,000 cells. In case of the cell
viability experiments the cells after treatment were stained with the LIVE/DEAD kit (Calcein
and Ethidium Homodimer) for 30 min. The data was acquired using 488 nm (for live cells)
and 585 nm (for dead cells).
Extraction of Doxorubicin from Doxorubicin Hydrochloride
Doxorubicin was exracted from a doxorubicin hydrochloride solution from a mixture of
chloroform and pH 8.0 Tris EDTA buffer. Initially doxorubicin hydrochloride was dissolved
in 100mL pH 8.0 Tris EDTA buffer having a composition of 0.1 M KCl, 40 mM
tris(hydroxymethyl)aminomethane (Tris), 100 mM EDTA. Then 100 mL choloroform was
26

Page 27 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
added to it and the biphasic mixture was vigorously stirred for 30 mins. The chloroform layer
was separated and evaporated in vacuum to get pure doxorubicin.
Doxorubicin Loading
Doxorubicin was loaded onto SWNTꢀ1 following a conventional protocol. To a dispersed
solution of SWNTꢀ1 in pH 8.0 PBS buffer having 100 ꢂg/mL SWNT, aqueous solution of
doxorubicin was added such that the final concentration of doxorubicin in the conjugate is 50
ꢂg/mL and the mixture was kept overnight under stirring condition. UVꢀvis and fluorescence
spectra of this hybrid material were checked with respect to control solutions having 100
ꢂg/mL SWNT and 50 ꢂg/mL doxorubicin concentration.
Targeted Delivery of Doxorubicin
The nanoconjugate SWNTꢀ1ꢀDOX was incubated with Hela and CHO cells for 12 h in cell
seeded 96ꢀwell plates. The concentration of SWNTꢀ1 was varied from 5 ꢀg/mL to 20 ꢀg/mL
while the doxorubicin concentration was varied from 2.5 ꢀg/mL to 10 ꢀg/mL. The
concentration ratio of the nanovector : drug was maintained at 2:1 (w/w) throughout the
experiment. The percentage of viable cells was determined using MTT assay as reported
previously.
Acknowledgements
P.K.D. is thankful to CSIR, India (ADD, CSC0302) and DST, India (SR/S1/OCꢀ25/2011) for
financial assistance. S.B, M.G. and S.D. acknowledge CSIR, India for Research fellowships.
Electronic Supplementary Information (ESI)
TGA plot, suspension stability experiment, LIVE/DEAD images of Hela, HepG2 and CHO
cells, structure reported amphiphile, control experiments with Cy3ꢀOlg, amphiphile Cy3ꢀOlg,
27

Journal of Materials Chemistry B
Page 28 of 41
View Article Online
DOI: 10.1039/C3TB21334J
plot of surface tension versus concentration of amphiphile, energy depletion experiment,
nuclear internalization experiment.
References.
1. (a) Z. Liu, K. Yang and S.T. Lee, J. Mater. Chem. 2011, 21, 586; (b) F. Lu, L. Gu, M. J.
Meziani, X. Wang, P. G. Luo, L. M. Veca, L. Cao and Y. P. Sun, Adv. Mater. 2009, 21, 139;
(c) Y. L. Wu, N. Putcha, K. W. Ng, D. T. Leong, C. T. Lim, S. C. J. Loo and X. Chen, Acc.
Chem. Res. 2013, 46, 782; (d) B. Wang, X. He, Z. Zhang, Y. Zhao and W. Feng, Acc. Chem.
Res. 2013, 46, 761.
2. (a) M. Prato, K. Kostarelos and A. Bianco, Acc. Chem. Res. 2008, 41, 60; (b) S. Lee , M.
G. Hahm , R. Vajtai , D. P. Hashim , T. Thurakitseree , A. C. Chipara , P. M. Ajayan and J.
H. Hafn, Adv.Mater. 2012, 24, 5261; (c) M. G. Hahm, A. L. M. Reddy, D. P. Cole, M.
Rivera, J. A. Vento, J. Nam, H. Y. Jung, Y. L. Kim, N. T. Narayanan, D. P. Hashim, C.
Galande, Y. J. Jung, M. Bundy, S. Karna, P. M. Ajayan and R. Vajtai, Nano Lett. 2012, 12,
5616; (d) P. Chakravarty, R. Marches, N. S. Zimmerman, A. D. E. Swafford, P. Bajaj, I. H.
Musselman, P. Pantano, R. K. Draper and E. S. Vitetta, Proc. Natl. Acad. Sci. 2008, 105,
8697; (e) M. Ashokkumar, K. M. Sumukh, R. Murali, N. T. Narayanan, P. M. Ajayan and P.
Thanikaivelan, Carbon 2012, 50, 5572.
3.(a) B. L. Droumaguet, J. Nicolas, D. Brambilla, S. Mura, A. Maksimenko, L. D. Kimpe, E.
Salvati, C. Zona, C. Airoldi, M. Canovi, M. Gobbi, M. Noiray, B. L. Ferla, F. Nicotra, W.
Scheper, O. Flores, M. Masserini, K. Andrieux and P. Couvreur, ACS Nano 2012, 6, 5866;
(b) C. Kaewsaneha, P. Tangboriboonrat, D. Polpanich, M. Eissa and A. Elaissari, ACS Appl.
Mater. Interfaces 2013, 5, 1857; (c) Z. Su, S. Zhu, A. D. Donkor, C. Tzoganakis and J. F.
Honek, J. Am. Chem. Soc. 2011, 133, 6874; (d) S. T. Kim, K. Saha, C. Kim and V. M.
Rotello, Acc. Chem. Res. 2013, 46, 681; (e) N. W. S. Kam, M. O. Connell, J. A. Wisdom and
28

Page 29 of 41
Journal of Materials Chemistry B
View Article Online
DOI: 10.1039/C3TB21334J
H. Dai, Proc. Natl. Acad. Sci. 2005, 102, 11600; (f) C. Bussy, H. A.; Boucetta and K.
Kostarelos, Acc. Chem. Res. 2013, 46, 692.
4. (a) D. Pantarotto, R. Singh, D. M. Carthy, M. Erhardt, J. P. Briand, M. Prato, K. Kostarelos
and A. Bianco, Angew. Chem. Int. Ed. 2004, 43, 5242; (b) A. P. Goodwin, S. M. Tabakman,
K. Welsher, S. P. Sherlock, G. Prencipe and H. Dai, J. Am. Chem. Soc. 2009, 131, 289; (c) R.
Singh, D. Pantarotto, L. Lacerda, G. Pastorin, C. Klumpp, M. Prato, A. Bianco and K.
Kostarelos, Proc. Natl. Acad. Sci. 2006, 103, 3357; (d) N. Nakayamaꢀ Ratchford, S.
Bangsaruntip, X. Sun, K. Welsher and H. Dai, J. Am. Chem. Soc. 2007, 129, 2448; (e) M.
Benincasa, S. Pacor, W. Wu, M. Prato, A. Bianco and R. Gennaro, ACS Nano 2011, 5, 199.
5. (a) D. Das and P. K. Das, Langmuir 2009, 25, 4421ꢀ4428; (b) H. Xie, A. O. Acevedo, V.
Zorbas, R. H. Baughman, R. K. Draper, I. H. Musselman, A. B. Dalton and G. R.
Dieckmann, J. Mater. Chem., 2005, 15, 1734; (c) C. G. Salzmann, M. A. H. Ward, R. M. J.
Jacobs, G. Tobias and M. L. H. Green, J. Phys. Chem. C 2007, 111, 18520; (d) E. J. Becraft,
A. S. Klimenko and G. R. Dieckmann, Peptide Sci. 2009, 92, 212; (e) C. Backes, C. D.
Schmidt, K. Rosenlehner, F. Hauke, J. N. Coleman and A. Hirsch, Adv. Mater. 2010, 22, 788;
(f) S. Dutta, T. Kar, S. Brahmachari and P. K. Das, J. Mater. Chem. 2012, 22, 6623; (g) V.
Georgakilas, K. Kordatos, M. Prato, D. M. Guldi, M. Holzinger and A. Hirsch, J. Am. Chem.
Soc. 2002, 124, 760; (h) B. G. Cousins, A. K. Das, R. Sharma, Y. Li, J. P. McNamara, I. H.
Hillier, I.A. Kinloch and R. V. Ulijn, Small 2009, 5, 587; (i) E. Schopf, R. Broyer, L. Tao, Y.
Chen and H. D. Maynard, Chem. Commun. 2009, 4818; (j) L. Shao, G. Tobias, C. G.
Salzmann, B. Ballesteros, S. Y. Hong, A. Crossley, B. G. Davis and M. L. H. Green, Chem.
Commun. 2007, 5090; (k) A. D. Crescenzo, D. Demurtas, A. Renzetti, G. Siani, P. D. Maria,
M. Meneghetti, M. Prato and A. Fontana, Soft Matter 2009, 5, 62; (l) C. Backes, C. D.
Schmidt, F. Hauke, C. Bottcher and A. Hirsch, J. Am. Chem. Soc. 2009, 131, 2172; (m) Y. Li,
B. G. Cousins, R. V. Ulijn and I. A. Kinloch, Langmuir 2009, 25, 11760.
29