Carbon Nanotechnologies for Drug Delivery · the existing literature on CNT application for drug...
Transcript of Carbon Nanotechnologies for Drug Delivery · the existing literature on CNT application for drug...
Università degli Studi di Trieste
XXIII ciclo del
Dottorato di Ricerca in Scienze e Tecnologie Chimiche e Farmaceutiche
PhD in Chemical and Pharmaceutical Sciences and Technologies
Carbon Nanotechnologies for Drug
Delivery
Settore scientifico-disciplinare CHIM/06
PhD student PhD School Director
Chiara Fabbro Prof. Enzo Alessio
Advisor
Prof. Maurizio Prato
Università degli Studi di Trieste
Co-Advisor
Dr. Tatiana Da Ros
Università degli Studi di Trieste
academic year 2009/2010
Table of Contents
Abbreviations I
Abstract III
Riassunto V
1 INTRODUCTION 1
1.1 Carbon nanotubes 1
1.2 CNT functionalization 3
1.2.1 Non-covalent functionalization 4
1.2.2 Covalent functionalization 4
1.2.2.1 Oxidation and defect site chemistry 6
1.2.2.2 1,3-dipolar cycloaddition of azomethine ylides 7
1.3 CNT characterization 7
1.3.1 UV-vis-NIR spectroscopy 8
1.3.2 Resonant Raman spectroscopy 9
1.3.3 Thermogravimetric analysis (TGA) 11
1.3.4 TGA-coupled mass spectrometry (TGA-MS) 11
1.3.5 Transmission electron microscopy (TEM) 12
1.3.6 Atomic force microscopy (AFM) 12
1.4 CNT Toxicity 13
1.5 CNT as delivery system for cancer 16
1.5.1 Uptake Mechanism 18
1.5.2 Delivery of antineoplastic chemotherapeutic drugs 24
1.5.3 Delivery of nucleic acids 37
1.6 Alternative CNT-based anticancer strategies 39
1.6.1 Thermal ablation 40
1.6.2 Radiotherapy & BNCT 42
1.7 References 43
2 SHORTENING OF SWCNTs 53
Table of Contents
2.1 Oxidative shortening of SWCNTs 53
2.1.1 Re-pristinization of the oxidized SWCNTs 57
2.1.2 Sodium Hydroxide washing 60
2.2 Mechanochemical shortening of SWCNTs 63
2.3 Acknowledgments 71
2.4 References 71
3 CONJUGATION OF ANTIBODIES TO CNTs 73
3.1 IgG-DWCNT 74
3.1.1 Preparation 74
3.1.2 Characterization 77
3.2 Fab’-MWCNT and scFv-MWCNT 81
3.2.1 Preparation 82
3.2.2 Characterization 87
3.2.3 Preliminary biological results 91
3.3 Acknowledgments 95
3.4 References 95
4 CONJUGATION OF DOXORUBICIN TO CNSs 97
4.1 Fullerene derivatives 98
4.1.1 Synthesis 98
4.1.2 Spectroscopic characterization 101
4.2 CNT derivatives: preparation and characterization 102
4.3 Preliminary biological results with fullerene derivatives 109
4.4 Acknowledgments 112
4.5 References 113
5 EXPERIMENTAL PART 115
5.1 Materials and Methods 115
5.2 Instrumentation 116
5.3 Experimental Procedures 118
5.3.1 Organic compounds 118
5.3.2 Fullerene derivatives 122
5.3.3 Antibodies 126
Table of Contents
5.3.4 CNT derivatives 127
5.4 References 138
I
Abbreviations
°C degree Celsius µm micrometre 111In Indium 111 2-IT 2-iminothiolane Å Angstrom a.a. aminoacid Ab antibody AcOEt ethyl acetate AFM atomic force microscopy Alloc allyloxycarbonyl aq. aqueous Ar Argon Boc tert-butoxycarbonyl C carbon CH2Cl2 dichloromethane CNS carbon nanostructure CNT carbon nanotube CO carbon monoxide DIEA N,N-diisopropylethylamine DiOC6 3,3′-dihexyloxacarbocyanine iodide DMF N,N-dimethylformamide DMSO dimethylsulfoxide DOS density of states DTNB 5,5'-dithio-bis-(2-nitrobenzoic acid) DTPA diethylenetriaminepentaacetic acid or diethylenetriaminepentaacetic dianhydride DWCNT double-walled carbon nanotube EDC·HCl N-ethyl-N′-(3-dimethylaminopropyl)carbodiimide hydrochloride EDTA ethylenediaminetetraacetic acid ESI-MS electrospray ionization mass spectrometry Et2O diethyl ether Fab’ antigen-binding fragment FITC fluorescein isothiocyanate GFLG glycine-phenylalanine-leucine-glycine h hour/hours H2 hydrogen (molecular) H2O water H2SO4 sulphuric acid HCl hydrochloric acid He Helium HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid HNO3 nitric acid HOBt 1-hydroxybenzotriazole
IC50 half maximal inhibitory concentration
Abbreviations
II
ID injected dose IgG immunoglobulin G IgG-SH thiol-bearing immunoglobulin G KMnO4 potassium permanganate MALDI-TOF-MS matrix-assisted laser desorption/ionization time-of-flight mass spectrometry MeOH methanol MFI mean fluorescence intensity min minute/minutes mm millimetre MW microwave or molecular weight MWCNT multi-walled carbon nanotube N2 Nitrogen (molecular) NaHCO3 sodium bicarbonate (or sodium hydrogen carbonate) NaOH sodium hydroxide NHS N-hydroxysuccinimide nm nanometre NMR Nuclear magnetic resonance O2 Oxygen (molecular) PBS phosphate buffered saline Pd(PPh3)4 palladium tetrakis p-HCHO paraformaldehyde Pht phthalimide PI propidium iodide PyAOP (7-azabenzotriazol-1-yloxy)tripyrrolidinophosphonium hexafluorophosphate r.t. room temperature RBM radial breathing mode RPM revolutions per minute scFv single-chain variable fragment SDBS sodium dodecylbenzenesulfonate s-DWCNT shortened double-walled carbon nanotube s-MWCNT shortened multi-walled carbon nanotube SO3 sulphuric anhydride SPR surface plasmon resonance s-SWCNT shortened single-walled carbon nanotube SWCNT single-walled carbon nanotube TB trypan blue Tc Technetium TEA triethylamine TEM transmission electron microscopy TFA trifluoroacetic acid TGA thermogravimetric analysis TGA-MS TGA-coupled mass spectrometry THF tetrahydrofuran TLC thin layer chromatography TM-AFM tapping mode-AFM UV-vis-NIR ultraviolet-visible-near infrared
III
Abstract
Nanotechnology underwent a very rapid development in the last decades, thanks to the
invention of different techniques that allow reaching the nanoscale. The great interest in this
area arises from the variety of possible applications in different fields, such as electronics, where
the miniaturization of components is a key factor, but also medicine. The creation of smart
systems able to carry out a specific task in the body in a controlled way, either in diagnosis or in
therapy, is the ultimate goal of a newborn area of research, called nanomedicine. In fact, to
reach such an outstanding objective, a nanometre‐sized vector is needed and carbon nanotubes
(CNTs) are among the most promising candidates.
The aim of this thesis was to study this opportunity, and in particular, the possible
application of carbon nanostructures for drug delivery.
In the Introduction (Chapter 1), an overview on CNTs is given, explaining their main
features and the key issues associated with their manipulation. The different existing
possibilities for CNT functionalization are described, focusing the attention on the covalent
approaches exploited in this thesis, namely oxidation and subsequent amidation of carboxyl
groups, and 1,3‐dipolar cycloaddition of azomethine ylides. The main characterization
techniques used are then described, illustrating their advantages and their limitations.
Subsequently, the very controversial issue of CNT toxicity is analysed, in terms of possible
consequences to CNT exposure and risk for the workers in the field. Finally, a detailed account of
the existing literature on CNT application for drug delivery is given, underlining the drawback of
non‐covalent approaches, and then focusing on the different targeted strategies studied so far in
CNT‐based nanomedicine.
In Chapter 2, a study on shortening of single‐walled carbon nanotubes (SWCNTs) is
presented, as a strategy to obtain a more soluble material, quite important for any medical
application. Two different options are explored. In the first part, an oxidative treatment of
SWCNTs, with a mixture of oleum and nitric acid, is described, reporting a detailed
characterization, for the production of shortened and disentangled CNTs, bearing carboxyl
groups, with a big improvement in dispersibility. Subsequently, an alkaline treatment, for the
removal of the amorphous carbonaceous material introduced with the oxidation, is shown. In
addition, a laser‐mediated re‐pristinization of the oxidized SWCNTs is reported for the first time.
Abstract
IV
In the second part of Chapter 2, a mechanochemical shortening of SWCNTs is presented,
as an alternative to the wet chemistry approach. Different complementary characterizations of
the SWCNTs obtained are shown, to highlight the very interesting possibility to control the
quality of the final product, by tuning experimental conditions.
In Chapter 3, the covalent conjugation of CNTs and antibodies (Abs) is presented, with
the aim of studying two different opportunities. Abs could mediate the specific targeting of a
CNT‐based nanovector with possible applications in both therapy and diagnosis. On the other
hand, CNTs could mediate the cellular uptake of Abs, thus paving the way to a variety of
unexplored intracellular, Ab‐based, therapeutic possibilities. The preparation of different kinds
of covalent Ab‐CNT conjugates is described, with both double‐walled carbon nanotubes
(DWCNTs) and multi‐walled carbon nanotubes (MWCNTs), using either the whole Ab or smaller
fragments. Moreover, a double functionalization strategy is reported, for the simultaneous
introduction of a probe, required for biological studies. Different complementary
characterizations of the conjugates are presented, to prove the effectiveness of the covalent
strategy and the preserved capability of the Ab, bound to CNTs, to specifically recognize its
antigen. Also, some preliminary biological data are shown.
In Chapter 4, the functionalization of both fullerene C60 and CNTs, either pristine or
oxidized, is presented, for studying the delivery of the antineoplastic drug doxorubicin. The drug
is attached either directly or introducing a cleavable peptidic sequence to trigger its release
inside cells. A detailed account on the adopted synthetic procedure is reported, together with
the spectroscopic characterization of all the constructs. Furthermore, preliminary biological data
on the fullerene derivatives are presented.
In conclusion, the present thesis reports the covalent modification of different kinds of
CNTs, described in a critical way, with the aid of many characterization techniques. All the
different constructs prepared could find an application in nanomedicine, giving a contribution to
the understanding and the development of this fascinating and promising field.
V
Riassunto
Le nanotecnologie si sono sviluppate molto rapidamente negli ultimi decenni, grazie
all’invenzione di un crescente numero di tecniche che permettono di lavorare su scala
nanometrica. Il grande interesse in quest’area trae origine dalle sue svariate possibili
applicazioni in diversi campi, quali l’elettronica, dove la miniaturizzazione ha un ruolo
fondamentale, ma la medicina. Infatti, lo sviluppo di un sistema ‘intelligente’ capace di eseguire
un compito specifico all’interno del nostro organismo in modo controllato, sia in campo
diagnostico che terapeutico, è lo scopo di una nuova branca della ricerca, la nanomedicina. Per
raggiungere un così notevole obiettivo, è necessario un vettore di dimensioni nanometriche, e i
nanotubi di carbonio (CNTs) rappresentano uno dei migliori candidati.
Lo scopo di questo lavoro di tesi era lo studio di questa opportunità e, in particolare,
della possibile applicazione di nanostrutture di carbonio per la veicolazione di farmaci.
Nell’Introduzione (Capitolo 1), viene fatta una breve panoramica sui CNTs, spiegando le
loro più importanti caratteristiche e le questioni fondamentali associate alla loro manipolazione.
Vengono descritte le diverse possibilità esistenti per la funzionalizzazione dei CNTs,
concentrandosi soprattutto sugli approcci covalenti sfruttati in questa tesi, ossia l’ossidazione e
la conseguente amidazione dei gruppi carbossilici così introdotti, e la cicloaddizione 1,3‐dipolare
di ilidi azometiniche. Sono poi trattate le principali tecniche di caratterizzazione impiegate,
illustrandone i vantaggi e le limitazioni. In seguito viene affrontato il tema molto controverso
della tossicità dei CNTs, in termini di possibili conseguenze a una loro esposizione e di rischio per
chi lavora in questo campo. Infine viene fatto un dettagliato resoconto della letteratura esistente
sull’applicazione dei CNTs per la veicolazione di farmaci, ponendo l’accento in particolare sugli
svantaggi connessi agli approcci non‐covalenti, e focalizzando poi l’attenzione sulle diverse
strategie di indirizzamento specifico studiate a oggi nella nanomedicina con i CNTs.
Nel Capitolo 2, viene presentato uno studio sull’accorciamento di nanotubi di carbonio a
parete singola (SWCNTs), come strategia per ottenere un materiale con migliori caratteristiche di
solubilità, estremamente importante per qualsiasi applicazione biomedica. Vengono esplorate
due diverse possibilità. Nella prima parte, si descrive il trattamento ossidativo di SWCNTs con
una miscela di oleum e acido nitrico, e la loro completa caratterizzazione, per la produzione di
nanotubi accorciati e disaggregati, recanti funzioni carbossiliche, con un grande miglioramento
Riassunto
VI
nella dispersibilità. In seguito viene descritto un trattamento basico per la rimozione delle
impurezze carboniose amorfe introdotte dall’ossidazione. In aggiunta, viene riportato per la
prima volta un fenomeno di ri‐pristinizzazione dei SWCNTs ossidati, mediante trattamento laser.
La seconda parte del Capitolo 2 riguarda l’accorciamento meccano‐chimico di SWCNTs,
come alternativa all’approccio in soluzione. Vengono mostrate diverse caratterizzazioni
complementari dei SWCNTs ottenuti, sottolineando la possibilità, molto interessante, di
controllare la qualità del prodotto finale variando le condizioni sperimentali.
Il Capitolo 3 tratta la coniugazione covalente di CNTs e anticorpi (Abs), con lo scopo di
valutare due diverse opportunità. Gli anticorpi, infatti, possono mediare l’indirizzamento
specifico di un nanovettore a base di CNTs, con possibili applicazioni terapeutiche e
diagnostiche. Viceversa, i CNTs possono mediare l’internalizzazione cellulare degli Abs, aprendo
così la strada a svariate possibilità terapeutiche a livello intracellulare, a oggi inesplorate. La
preparazione di diversi coniugati covalenti Ab‐CNT viene descritta, utilizzando nanotubi a parete
doppia (DWCNTs) o multipla (MWCNTs), e sia Abs interi, sia frammenti degli stessi. Inoltre viene
descritta una doppia funzionalizzazione dei CNTs, volta a introdurre simultaneamente una sonda
necessaria per gli studi biologici. Vengono poi presentate diverse caratterizzazioni
complementari dei coniugati, per dimostrare l’efficacia della strategia covalente adottata, e la
preservata capacità dell’Ab legato ai CNTs di riconoscere il suo specifico antigene. Infine,
vengono presentati alcuni risultati biologici preliminari.
Nel Capitolo 4, viene descritta la funzionalizzazione del fullerene C60 e dei CNTs, sia
pristine che ossidati, per lo studio della veicolazione della doxorubicina, un farmaco
antineoplastico. Il farmaco viene legato sia in maniera diretta che indiretta, mediante
l’introduzione di una catena peptidica scindibile a livello intracellulare, al fine di mediarne il
rilascio. Viene data una dettagliata descrizione delle procedure sintetiche adottate, oltre alla
caratterizzazione spettroscopica di tutti i composti preparati. Infine vengono illustrati i primi
risultati biologici ottenuti sui derivati fullerenici.
In conclusione, il presente lavoro di tesi descrive la funzionalizzazione di diversi tipi di
CNTs mediante un approccio covalente, con un’analisi critica dei risultati, tramite l’utilizzo di
svariate tecniche di caratterizzazione. Tutti i derivati preparati trovano potenziale applicazione
nel campo della nanomedicina, contribuendo alla comprensione e allo sviluppo di questa scienza
così affascinante e promettente.
1.1
field
com
atte
the
arom
ther
Cha
givin
(DW
of ca
the
CNT
Figu
of
Carbon nan
The rapi
ds rendered
mmunity. Thi
ntion on the
present thes
matic conjug
refore subjec
pter 1.2.2). T
ng raise to d
WCNTs) up to
a. 0.35 nm.
The coili
angle betwe
T will depend
ure 1. Graph
rolling up of
notubes
id developm
this mater
s paragraph
e key issues r
sis work. CN
gated netwo
cted to a cur
The number
ifferent kind
o generic mu
ing up direct
een C‐C bond
d on this ang
ical explanat
f a CNT. Thus
Intr
ent of carbo
rial not as
gives a brie
related to CN
Ts are tubul
ork as they a
rvature, with
of cylinders
s of CNTs, na
lti‐walled CN
tion of the g
ds and the a
le.
tion of the ch
s, each CNT c
two
roduc
on nanotube
much unkn
ef overview
NTs, necessa
ar structure
are in graphi
h important
that constit
amely single
NTs (MWCNT
graphene she
axis of the tu
hiral vector θ
can be descr
o examples d
ction
s (CNTs)‐bas
nown as fe
to remind t
ry as a backg
s made of o
ite, but rolle
consequenc
tute a CNT ca
e‐walled CNT
Ts), where th
eet determin
ube can vary
θ of a CNT. T
ibed by the c
depicted.
sed technolo
w years ag
heir main fe
ground for th
nly carbon a
ed up in a cy
es in the che
an vary from
Ts (SWCNTs),
he distance b
ne different k
y, and the so
The vector re
couple of val
ogy in many
go for the
eatures and
he comprehe
atoms arrang
ylindrical sh
emical react
m only one to
, double‐wal
between two
kinds of CNT
o‐called chir
efers to the d
lues (n,m), a
1
different
scientific
to focus
ension of
ged in an
ape, and
ivity (see
o several,
led CNTs
o walls is
T. In fact,
ality of a
direction
s in the
Chapter
2
T
Figure 1
former h
with 0° <
tube inc
a comm
chiral ve
or chira
example
exist.1
being sc
ballistic
stability,
addresse
irreprod
mean ge
of a CN
syntheti
content
structur
pentago
other th
Figure 2
T
pristine
1
Thus, differe
1. The two ex
having θ = 0
< θ < 30° are
creases. The
mon vocabula
ector. In fact,
al CNTs are
e of small dia
Additionally,
cattered. Th
transport an
Besides the
, together w
Nevertheles
ed by the re
ducibility. In f
etting the sa
NT sample,
c process. T
(i.e. metal a
es are vac
ons/heptagon
han carbon.2
2. Example o
strain. Note
The other b
CNTs are i
ent kinds of
xtremes of th
0° and m=0,
e defined as c
description o
ary, but ma
, infinite‐len
semi‐condu
ameter tube
, as a result
he absence
nd it allows t
se unique e
with extremel
s they still
esearch in th
fact, buying
ame product
that make
This variabili
and carbona
ancies, i.e.
ns interrupti
of pentagons
the change
ig limitation
nsoluble in
CNTs are de
he huge vari
and the latt
chiral. Of cou
of this featur
inly because
gth armchair
cting, even
s, where the
of the 1‐D n
of scattering
he nanotube
electronic pr
ly high tensil
present to
he field. One
the same kin
t. This lack o
it subjected
ity is associa
ceous impur
incomplete
ing the hexa
s/heptagons
in tube chira
associated w
water and
escribed acco
iety of possib
ter with θ =
urse, with in
re is importa
e electronic
r SWCNTs ar
though the
e curvature p
nature of CN
g of the ele
e to conduct
roperties, id
le strength a
date some
e of the majo
nd of CNTs b
of reliability i
d to too ma
ated with C
rities) and pr
e bonding
agonal netwo
defect in the
ality between
with CNT re
in any orga
ording to a c
bilities are zi
30° and n =
creasing n a
ant not only t
properties v
re metallic, w
ese are not
plays an imp
NTs, electron
ectrons duri
t without diss
eal CNTs ex
nd elasticity
important d
or problems
by the same s
is probably d
any unpredi
CNT length,
resence of d
defects, top
ork (Figure 2
e hexagonal
n left and rig
search is the
anic solvent
chiral vector
gzag and arm
m ≠ 0. In‐be
nd/or m, the
to define dif
vary a lot de
whereas infin
strict rules.
ortant role,
ns can be co
ing conducti
sipating ene
xhibit chemi
.
drawbacks t
is related to
supplier doe
due to the b
ictable varia
diameter, ch
efect. The p
pological ch
2), and dopin
CNT networ
ght side of th
eir poor disp
t, and a fun
r, as explaine
mchair CNTs
etween, all C
e diameter o
fferent CNTs
epending on
nite‐length z
In the case
some except
nducted wit
ion is know
rgy as heat.
ical and the
that need t
o batch‐to‐b
es not necess
big heteroge
ables during
hirality, imp
possible defe
hanges, suc
ng with elem
rk, inducing l
he defect.
persibility. In
nctionalizatio
ed in
s, the
CNTs
of the
with
n the
igzag
e for
tions
thout
wn as
ermal
o be
batch
sarily
neity
g the
purity
ctive
h as
ments
ocal
n fact
on is
requ
1.2)
deve
The
and
gas
prop
SWC
carb
Amo
CNT
proc
carb
tech
prec
and
of SW
1.2
hand
For
of fu
can
furth
uired in orde
.
Differen
eloped for fu
first two me
form CNTs a
phase and it
per metallic
CNTs. Usual
bonaceous m
ong the thre
Ts and can
cesses have
bon monoxi
hnique explo
cursor. The p
C for the CN
WCNTs with
CNT functio
As alrea
dling. In fact
this reason,
unctionalizat
be introduc
her modifica
Figure 3
er to improve
t synthetic t
ullerene synt
ethodologies
at high temp
t takes place
catalyst ca
impurities d
material othe
ee different s
be scaled u
been devel
de dispropo
oits carbon
process take
NT growth o
diameters r
onalization
ady mention
t, as produce
in most case
tion, the str
ed to increa
ations.
. Non‐covale
e dispersibilit
techniques e
thesis, laser
s involve high
erature (tho
at relatively
n switch fro
deriving from
er than the d
synthetic me
up to indust
oped, amon
ortionation
monoxide (C
es place at 8
n the iron na
anging from
ned, CNT us
ed, they are
es, a chemic
ong inter‐tu
ase solubility
ent (a) and co
ty for a bette
exist for CNTs
ablation and
h‐power ene
ousands of Ce
y low temper
om the pro
m the synth
desired CNTs
ethods, CCV
try level be
ng which on
process (Hi
CO) as the
800‐1000°C t
anoparticles
0.7 to 1.4 n
efulness is
not soluble i
cal modificat
ube interacti
y in certain s
ovalent (b) a
er manipulat
s. The main
d catalytic ca
ergy to vapo
elsius), while
ratures (500
oduction of
hetic process
s, represente
D provides a
ecause of th
ne of the mo
iPCO), used
carbon sour
through the
. Optimized
m, with a go
strongly hin
in water and
tion is neede
ons are red
solvents, and
approaches f
tion of the m
ones are arc
rbon vapour
rize solid‐sta
e CCVD uses
‐1000 °C). In
MWCNTs to
s are metall
ed mainly by
a better cont
he lower co
ost famous
for produ
rce, and Fe(
disproportio
conditions le
ood level of p
ndered by th
d in all comm
ed prior to ap
uced. Moreo
d to provide
for CNT funct
Intr
material (see
c discharge, o
r deposition
ate carbon (g
carbon prec
n general the
o the produ
ic nanoparti
y amorphous
trol in the g
osts. Differe
is the high‐
cing SWCNT
(CO)5 as the
onation of C
ed to the pr
purity.
he difficulty
mon organic
pplication. B
over, specifi
e an anchor
tionalization
oduction
3
e Chapter
originally
(CCVD).3
graphite)
cursors in
e use of a
uction of
icles and
s carbon.
rowth of
nt CCVD
‐pressure
Ts.4 This
e catalyst
O to CO2
oduction
in their
solvents.
By means
c groups
point for
n.
Chapter 1
4
The main distinction to be underlined when speaking of CNT functionalization is
between non‐covalent and covalent approaches (Figure 3). They both present pros and cons,
which should be carefully taken into account when deciding which is the best route to follow,
according to the desired application.
1.2.1 Non‐covalent functionalization
Among non‐covalent CNT functionalization strategies, one possibility is based on small
aromatic molecules, bound to the tubes by means of π‐π stacking,5 while another one is based
on the wrapping of polymeric molecules around the tubes. The latter method can involve both
biological macromolecules, such as nucleic acids,6 lipids7 or peptides8 and synthetic polymers,9
and it occurs via π‐π stacking and/or van der Waals interactions. Another, quite particular,
approach for the non‐covalent modification of CNTs is the filling of their inner cavity.10 The main
advantage of the non‐covalent modification of CNTs is that the electronic and mechanical
properties of the tubes are preserved. Therefore this should be the first choice when CNTs are
exploited in fields such as molecular electronics. On the other hand, the drawback of a non‐
covalent approach is the possible reversibility of the bonds involved, which can result in the loss
of the functionalization. This risk needs to be considered for example in the case of most
biomedical applications, since it is difficult to exactly foresee the fate of a non‐covalently bound
molecule when the functionalized CNTs are administered in vivo. For this reason, a covalent
approach should be preferred if CNTs will serve as drug delivery system, as in the work
presented in this thesis.
1.2.2 Covalent functionalization
The reactivity of CNTs in terms of covalent chemistry on the C backbone is due to local
strain, which is caused by two main reasons: the first one is the curvature‐induced
pyramidalization of the conjugated carbon atoms, and the second is the π orbital misalignment
between adjacent pairs of conjugated carbon atoms.11 Since pyramidalization angle gives a good
measure of the local weakening of π‐conjugation and of the strain energy of pyramidalization, it
appears clear why this effect is quite important for fullerenes and for CNTs caps, that are like an
hemi‐fullerene, while it carries less weight in the case of the flatter CNT sidewalls (Figure 4,
upper part).
On the other hand, π orbital misalignment accounts for strain in CNT sidewall much
more than in fullerenes, where it is very little (in fullerene C60 π orbitals alignment is even
perfect) (Figure 4, lower part).
Fig
geom
angl
tube
stra
the
hexa
nano
As a
and
CNT
mod
usua
first
func
Afte
com
nucl
gure 4. Repre
Therefor
metry‐depen
les and the
es, a higher
in‐induced r
reactivity of
agons netw
ostructures,
already antic
of the rem
Ts, which dis
dification of
ally harsh co
covalent r
ctionalization
erwards CNT
mpounds.13 O
leophilic16 or
esentation o
re C60 and
ndent strain,
π‐orbital mi
reactivity is
reactivity. Ne
f CNT sidewa
ork, that re
and can cau
cipated, cova
arkable mec
rupt the ban
the unsatur
onditions, gr
eactions eve
n could reac
Ts can be f
Other existin
r electrophili
of pyramidali
for fulleren
CNT reacti
, but for diff
salignment a
expected fo
evertheless,
alls: the pres
esult in a
use an unpre
alent sidewa
chanical pro
nd‐to‐band t
rated carbon
ew up a lot
er performe
h such high
further mod
ng possibiliti
ic17 additions
zation angle
e C60 and for
ivity toward
ferent reaso
angles of CN
or smaller tu
there is ano
sence of defe
locally enha
dictable che
all functional
operties of C
transitions o
n network o
in the last
ed on SWCN
levels that i
dified using
ies include a
s and cycload
e (θ – 90°) an
r a (5,5) SWC
ds addition
ons. Furtherm
NTs scale inv
ubes than fo
other param
ects, such as
anced chem
mical behav
lization caus
CNTs, since i
f π electrons
of CNTs, tho
years, starti
NTs in 1998
in the end th
Grignard r
addition of
dditions.18
d of π orbita
CNT.
reactions a
more, since t
versely with
or larger one
eter that co
s pentagon‐h
mical reactivi
iour of differ
e a loss of t
t generates
s. The possib
ugh being li
ng from fluo
8.12 In this
he tubes can
eagents or
both alkyl14
Intr
al misalignm
are both d
the pyramid
the diamete
es, if conside
ould deeply i
heptagon pa
ity of the
rent diamete
the high con
sp3 carbon
bilities for a
imited and r
orination, on
case the d
n become in
with organ
4 and aryl15
oduction
5
ent (Φ)
riven by
dalization
er of the
ering just
influence
irs in the
graphitic
ers CNTs.
nductivity
sites on
covalent
requiring
ne of the
egree of
nsulators.
nolithium
radicals,
Chapter 1
6
Herein the attention will be focused mainly on oxidation of CNTs and the subsequent
defect site chemistry, and on 1,3‐dipolar cycloaddition of azomethine ylides, widely exploited in
this thesis.
1.2.2.1 Oxidation and defect site chemistry
The most common way to oxidize CNTs involves acid treatments, with sonication of CNTs
in concentrated acid mixtures, where nitric acid or hydrogen peroxide plays the role of the
oxidative agent.19 Also other oxidative agents have been used, such as phosphomolybdic acid,20
potassium permanganate,21 or oxygen O2, by heat treatment of the CNTs in O2 atmospheres.22
CNTs are oxidized with multiple aims, mainly purification and shortening. In fact, CNTs
are typically synthesized with poly‐disperse micrometre lengths, bound together into
macroscopic entangled ropes, and they contain metallic impurities, deriving from the catalyst
used during their synthesis. Many applications, however, require individual short CNTs, void of
metallic impurities. Oxidative treatments reduce metal content due to the removal of the
amorphous carbon usually covering metallic particles, or the etching of CNTs caps when the
particles are still inside them, and the subsequent oxidation of the metal to soluble species,
which are then easily removed. Moreover, as a result of chemical oxidation, the ends and the
sidewalls of CNTs are covered with oxygen containing groups, mainly carboxylic groups, useful
for further derivatization.23 The process of the oxidation of CNTs can be described as step‐wise
with (i) an initial attack on the originally existing reactive sites, such the terminal fullerene‐like
caps, and the heptagon‐pentagon defects, which carry the higher strain, (ii) a defect‐generation
step, when additions to hexagon π bonds take place, generating new defects, and introducing
hydroxyl groups, then further oxidized to carboxyl groups, and (iii) a defect‐consuming step,
when the graphene structure around the defect sites is broken, and the tube is cut. Depending
on the strength of the conditions, oxidation can stop at the first step, thus attacking only already
existing defects, or proceed with the generation of new defects and finally with the cutting of
the nanotubes. A detailed study of diameter‐dependent oxidative stability confirmed a direct
relationship between diameter and reactivity, as expected on the basis of what has been
previously explained in this chapter. Studying the resonant Raman radial breathing mode (RBM),
the authors clearly showed that smaller diameter tubes are more easily air‐oxidized than larger
diameter ones.24
As already anticipated, the introduction of carboxylic groups allows a further
derivatization of CNTs. Esterification and amidation reactions are in fact widely exploited
possibilities for the covalent modification of oxidized CNTs, since they usually permit to easily
obta
envi
1.2.2
func
from
addi
reac
bon
are
equi
func
whe
Sinc
diam
diffe
diam
the
bein
grou
grou
1.3
appr
synt
ain highly fu
isaged applic
2.2 1,3‐dipo
The 1,3
ctionalization
m 2002.27 Th
ition of a d
ction of an a
d on the full
Scheme 1.
In fullere
all equivalen
ivalent. In
ctionalization
ereas zigzag
ce real samp
meter, length
erent CNTs.
meter, and o
π‐orbital mi
ng the dipole
ups in the d
ups introduc
CNT charac
A great
roach is cho
thesis cannot
unctionalized
cation.25
lar cycloadd
3‐dipolar cyc
n since 1993
his very ver
ipole to a d
aldehyde (or
erene cage o
General mec
ene C60, the
nt. Instead, C
fact, it w
n at C‐C bo
CNTs prefer
ples of CNTs
h, chiralities
Theoretica
only weakly o
salignment
e a nucleoph
ipole, and e
ed on the tu
cterization
difficulty a
osen, is their
t be translat
d material, b
dition of azom
cloaddition
3,26 was tran
satile reacti
dipolarophile
r a ketone) w
or on the CN
chanism of 1
first addition
CNTs sidewal
was found
onds tilted w
r attachmen
s are norma
and defects
al calculatio
on the chiral
angles.29 Mo
hile, is in pri
lectron‐with
bes by an ox
ssociated w
r characteriz
ed to this m
bearing the
methine ylid
of azometh
nslated to ca
on consists
e, where the
with an α‐am
T (Scheme 1
1,3‐dipolar cy
n takes place
ll is made on
by theoret
with respect
nt of the ylid
ally made o
s, we should
ns showed
lity, as expec
oreover, this
nciple accele
hdrawing gro
xidation step
with CNT fun
zation. In fac
aterial. The
desired mo
des
hine ylides,
arbon nanot
in the conc
e dipole is a
mino acid, a
1).
ycloaddition
e in any of th
nly of [6,6] d
ical studies
t to tube ax
des at segm
of a mixture
ask ourselv
that react
cted conside
s particular k
erated by th
oups in the d
p.29,30
nctionalizatio
ct all the no
main reason
olecule accor
widely exp
tubes chemis
certed [π4 +
an azomethi
nd the dipo
of azomethi
he [6,6] doub
ouble bonds
s that arm
xis, (i.e. bon
ments paralle
of differen
es which is t
ivity depen
ring pyramid
kind of 1,3‐d
e presence
dipolarophile
on, especial
rmal techniq
n is that the a
Intr
rding to the
ploited in f
stry as well,
+ π2] pericy
ine ylide, fo
larophile is
ine ylides to
ble bonds, si
s, but they a
chair tubes
nd A‐C in F
el to the tub
nt tubes in t
the behavio
ds mainly
dalization an
dipolar cyclo
of electron‐
e, such as c
lly when a
ques used in
amount of fu
oduction
7
e specific
fullerene
, starting
yclic 1,3‐
ormed by
a double
C60.
ince they
re not all
s favour
igure 4),
be axis.28
terms of
ur of the
on tube
ngles and
addition,
donating
arboxylic
covalent
n organic
unctional
Chapter
8
groups i
Also, for
function
what co
paramag
1.3.1 UV
T
continuo
singular
observe
apprecia
interacti
commer
singular
semicon
transitio
Figure
In fact,
structur
1
is usually to
r techniques
nalized CNT,
oncerns NM
gnetic metal
Herein some
V‐vis‐NIR spe
The electron
ous function
ities (Figure
d in the UV
able due to
ions betwee
rcial HiPCO S
ities in meta
nducting tub
on is inversel
5. a) Density
UV‐vis‐NIR s
covalent m
e, and hence
o low, with
s performed
and the req
MR, further
lic particles
e of the main
ectroscopy
nic density
n of energy
e 5a). Optic
V‐vis‐NIR re
the overlap
en nanotube
SWCNTs, the
allic tubes (M
es (S11 and S
y proportion
y of states di
spect
spectroscopy
modifications
e perturb the
respect to t
in solution,
uired concen
problems
in CNT samp
n characteriz
of states of
y, but it p
ally allowed
gion as sha
p of similar
es in a bun
e main featu
M11) and betw
S22), as indica
nal to tube d
iagrams for m
rum of pristi
y could prov
can disrupt
eir electronic
the CNT com
the choice
ntrations are
are given
ples, due to t
zations explo
f SWCNTs, b
ossesses sp
d transitions
arp peaks, e
transitions
ndle broade
ures represe
ween the firs
ated in Figur
iameter.31
metallic and
ine HiPCO SW
ide informat
t the period
c transitions,
mponent of
of the solve
e usually far
by the pre
the magnetic
oitable for CN
being one d
pike‐like fea
s between t
even though
given by di
n optical lin
ent transitio
st and the se
e 5b. Moreo
semiconduc
WCNTs (DMF
tion on the f
dicity of SW
, resulting in
the sample,
nt is highly
too high to
sence of fe
c field.
NT will be pre
imensional o
tures know
these featur
h the sharpn
fferent SWC
nes. In the
ns between
cond pairs o
over, the ene
cting SWCNT
F).
functionaliza
WCNTs conju
a loss of van
, to be dete
limited, eve
be achieved
erromagneti
esented.
objects, is n
wn as van H
res are typ
ness is not
CNTs. Moreo
specific cas
the first pa
of singularitie
ergy of elect
Ts; b) UV‐vis‐
ation of SWC
ugated elect
n Hove featu
cted.
n for
d. For
c or
not a
Hove
ically
fully
over,
se of
air of
es for
ronic
NIR
CNTs.
ronic
ures.
1.3.2
and
than
coul
wea
i.e. u
mat
reso
excit
of tu
typic
Fi
exac
inve
is pa
also
band
ones
due
2 Resonant
When a
a small frac
n the incide
ld result in a
ak. However,
using a laser
erial studied
SWCNTs
onance Rama
ting laser en
ubes in a he
cal SWCNT R
gure 6. a) Ra
Radial b
ctly as if the
ersely propor
articularly hig
G‐mode
in graphite,
d (1580 cm−
s are normal
to vibration
Raman spec
beam of ligh
ction of this
nt light. In f
an energy ex
, its intensity
r energy that
d.
s, due to th
an response,
nergy are pro
eterogeneou
Raman spect
aman spectr
breathing mo
tube was br
rtional to tub
gh, weak ove
correspond
, from which
−1) is split int
lly considere
ns in the cir
troscopy
ht pass acros
small fractio
fact, the int
change, nam
y can be stro
t matches th
heir broad
and conseq
obed. Theref
us SWCNT sa
rum are dep
a of pristine
SWC
ode (RBM) c
eathing. In f
be diameter
ertones can
ds to tangen
h the name
to up to six
ed, i.e. the G+
rcumferentia
ss a transpar
on is scatter
teraction bet
mely the Ram
ongly enhan
e energy of
range of o
quently only t
fore, using d
ample are a
picted in Figu
HiPCO SWC
CNT Raman f
corresponds
fact, it depen
. RBM range
be observed
tial planar v
derives. Ho
peaks, due
+, which is gi
al direction
ent sample,
red inelastic
tween the i
man effect. U
ced working
optically allo
ptically allo
tubes with o
different lase
nalysed (Fig
ure 6b.
NTs obtained
features.
to radial ex
nds on nanot
e is typically
d at double fr
vibrations of
wever, diffe
to the loss
ven by vibra
(Figure 6b).
a small fract
ally, i.e. at d
ncident pho
Usually Rama
g in resonanc
owed electro
wed transit
one of the ba
er energies, d
ure 6a). The
d with differ
pansion‐con
tube diamete
100‐350 cm‐
requency.
f carbon ato
rently from
of symmetry
ations along t
In this latte
Intr
tion of it is sc
different fre
oton and the
an scattering
ce Raman sc
onic transitio
tions, alway
and‐gaps equ
different pop
e main featu
rent lasers; b
ntraction of S
er: RBM freq
‐1, and, if its
oms and it is
graphite, SW
y. Just the t
tube axis, an
er band, the
oduction
9
cattered,
quencies
e sample
g signal is
cattering,
ons in the
s give a
ual to the
pulations
ures of a
b) main
SWCNTs,
quency is
intensity
s present
WCNT G‐
wo main
nd the G‐,
ere is an
Chapter 1
10
evident difference in the line‐shape between semiconducting SWCNTs, which give a Lorentzian‐
like peak (as G+) and metallic ones, where the peak is broadened, due to the presence of free
electrons.32 This difference can be clearly appreciated comparing the spectra depicted in Figure
6a: when using 532 nm laser, mainly metallic tubes are in resonance, due to their M11 transition
(see Figure 5 in Chapter 1.3.1), and in fact the G‐ line‐shape is much broader than in the case of
the 785 nm laser, which is instead in the semiconductor S22 transition region.
Another important Raman feature is the D band, whose position is at around 1300 cm‐1,
and it shifts to higher wavenumbers as the laser excitation energy is increased. This mode is
associated with disorder, i.e. with the sp3‐hybridized carbon atoms present in the tube or in the
carbonaceous impurities. Therefore, it is often used to evaluate the structural quality of
SWCNTs, expressed usually as D/G ratio. Moreover, the full‐width‐at‐half‐maximum (FWHM)
intensity of the D‐band of the various carbon impurities is generally much broader than that of
CNT D‐band, and thus D‐band line‐width could give an indication of the SWNCT purity level.33
Other less significant bands are typically present in a SWCNT Raman spectrum, due to
overtone modes of single bands or combinations of different ones. Among them the strongest
one is usually the G’ band. The name of this band is misleading, since it derives from graphite,
where this mode is usually the second strongest after the G mode, but it is actually the overtone
of the defect‐induced D mode, even though being present even in defect‐free nanotubes.
The main changes that could be appreciated in a SWCNT Raman spectrum upon
chemical modification are the following:
‐ loss or decrease of some RBM bands, due to destruction or extensive modification of the
corresponding kind of CNTs;
‐ change in the D/G ratio, due to modification of sp3/sp2 carbon atoms ratio (an increased D/G
ratio could give a quali‐quantitative information on the degree of covalent functionalization, but
it could also be due to increased amorphous carbon content);
‐ change in the FWHM intensity of the D‐band, according to the amorphous carbon content of
the sample, as already explained;
‐ change in the baseline height (a great content of amorphous carbon can give a strong
fluorescence, perturbing the Raman spectrum).
A big drawback of the two characterizations just described is that their usefulness is
usually limited to SWCNTs.
In fact, MWCNT UV‐vis‐NIR absorption, deriving from the overlap of many different
contributions in each tube, does not result in the typical well‐defined van Hove features of a
SWCNT spectrum, and thus it cannot give information on CNT covalent functionalization.
Introduction
11
Concerning Raman, MWCNTs do not present defined RBM, being each tube made of
several concentric walls. Moreover, the D band is quite intense, often more than the G‐band,
because of the high extent of defects usually present in MWCNTs. Consequently, it is difficult to
appreciate an increase in the D/G ratio due to functionalization. For DWCNTs the situation is
intermediate between single and multi‐walled CNTs and it is normally possible to appreciate
RBM and D/G ratio variations.
1.3.3 Thermogravimetric analysis (TGA)
TGA is an analytical technique for determination of the thermal stability of a material.
The weight of the sample is recorded, while heating it in a furnace under a controlled
atmosphere, which can be either inert (using gases as N2 or He), or oxidative (using air or pure
O2).
In the latter case CNT combustion takes place usually between 400 and 600°C,
depending on different factors, and resulting in the formation of CO2. CNT combustion
temperature, easily determined as the maximum in the first derivative (see Figure 4 in Chapter
2.1), depends on different factors. For example, smaller diameter nanotubes are believed to
oxidize at lower temperature due to higher curvature strain. The same effect could be given by
defects in the CNT structure. Also the presence of metal particles could lower combustion
temperature, since they can catalyse carbon oxidation.34 Furthermore, it is possible to use air
TGA to determine the metal content of a sample, since the residual weight after completeness
of CNT combustion is ascribable to metals in their oxidized form.
On the other hand, in the case of TGA run under inert atmosphere, CNTs are stable up to
ca. 800°C, and therefore it is possible to ascribe the weight loss at lower temperatures to the
organic material present in the sample, thus estimating the degree of functionalization after a
chemical modification of CNTs. Nevertheless, it is important to underline that also amorphous
carbonaceous species possibly present in a CNT sample could burn in the same temperature
range, and this fact should be taken into account when using TGA measurements for
quantitative characterization.
1.3.4 TGA‐coupled mass spectrometry (TGA‐MS)
The possibility offered by MS‐TGA is quite interesting: a mass spectrometer has been
coupled to the vent‐hole of the TGA furnace, in order to analyse the gaseous material evolving
during the experiment, thus giving precious information on what is going on. In this case, the
TGA analysis should be performed in He, thanks to its low molecular weight (4), so that the mass
Chapter 1
12
spectrometer could screen all the masses with molecular weights bigger than 4. Otherwise the
signal given by the gas would completely saturate the instrument. It is important to point out
that the mass spectra obtained are not like normal ones. In fact a massive fragmentation occurs
during TGA, and so very small fragments are produced. In a typical analysis a big CO2 peak
appears in correspondence of every combustion taking place, associated with a decrease in the
O2 signal (inert gas cylinders always contain some), which is being consumed by the combustion
itself. However, also higher molecular weight peaks appear, making it possible to determine the
presence of certain molecules if very characteristic fragments are produced (see Figure 5b in
Chapter 2.1) or by comparison of the pattern obtained with known samples (see Figure 5 in
Chapter 4.2).
1.3.5 Transmission electron microscopy (TEM)
TEM is a major tool for the morphological characterization of CNTs. A TEM image derives
from the interference of the sample, deposited onto a grid, with an electron beam transmitted
through it. Thus, metal particles interfere more, producing spots with higher contrast. On the
other hand, individual SWCNTs are usually very difficult to visualize (with a non‐high resolution
TEM), due to the low contrast with the grid surface, which is usually made of carbon. Bundles of
SWCNTs instead, as well as DWCNTs or MWCNTs, can be easily detected providing different
useful information, such as CNT length and diameter, together with a rough assessment of their
purity, in terms of metals and amorphous material content. Regarding the evaluation of the
dispersibility of a sample, it should be underlined that, during the drying procedure that follows
the deposition of the CNT dispersion on the grid, CNTs can easily re‐aggregate. Therefore the
TEM analysis do not provide reliable information on their original aggregation state.
1.3.6 Atomic force microscopy (AFM)
AFM is a particular kind of Scanning Probe Microscopy, which consists in scanning a
sharp tip (radius of curvature = 3‐50 nm) over a surface and detect one or more probe/surface
interactions. In AFM the result is a topographical image of the surface. The tip is mounted on a
flexible cantilever, allowing it to follow the surface profile, and its deflection is measured using a
laser spot reflected from the top surface of the cantilever into an array of photodiodes. AFM
could be performed in Contact Mode, when the tip actually touches the surface, thus possibly
moving objects or damaging soft samples, in Non‐Contact Mode, when the tip scans the surface
at a certain distance, with a consequent lower resolution, and in Tapping Mode (TM), when the
tip is oscillated rapidly against the surface during scanning, and the microscope extrapolates the
surf
inte
imag
acro
this
shad
How
the
depo
prec
diam
inte
obje
16c
1.4
dive
the
synt
chem
appl
ace topogra
ractions (Fig
Typically
ge, while th
oss the surfa
image show
dows on sha
wever, shado
set value, an
TM‐AFM
osited on a
cise enough
meter of a
ractions. Ne
ects on the s
in Chapter 2
CNT Toxicit
One of t
ersity of this
modified on
thetic metho
mical modif
lications, thu
aphy by th
gure 7).
F
y, when scan
e other sho
ce, being a s
ws where a
arp feature
ows are actu
nd the real to
M analysis co
surface, suc
(being in t
SWCNT, du
vertheless, i
surface, sinc
2.2).
ty
the big conc
material sh
nes. In fact
odologies an
ication (cov
us leading to
e change i
Figure 7. Can
nning a surf
ows the amp
sort of “erro
and how mu
boundaries,
ally areas w
opographical
ould be explo
ch as length
the range 1
e to the de
t is possible
e vertical re
cerns regard
ould be take
not only co
nd the suppl
alent or no
o changes in
n the oscil
ntilever tappi
face in TM, t
plitude of th
r” image (se
uch the am
, which ma
where the am
l image is the
oited to gat
h distribution
1‐5 nm) to e
elay in the
to obtain th
esolution doe
ing CNTs is,
en into acco
ommercial p
ier, but, mo
n‐covalent)
their water
lation ampl
ing on a surf
two images
he tapping o
e Figure 16a
mplitude cha
kes it look
mplitude had
e first one.36
her structur
n. However,
evaluate ve
response o
his informati
es not suffer
of course, t
ount, both fo
pristine samp
ore importan
when inten
dispersibilit
itude, trigge
ace.35
are obtaine
oscillation as
a and 16b in
nged, giving
like a snaps
d not yet bee
al informatio
lateral reso
ry short dis
of the canti
on from the
r from this li
their toxicity
or the as‐pro
ples can diff
ntly, CNTs al
ded to be
y and in tox
Intr
ered by tip
ed: one is th
s the probe
Chapter 2.2
g the impre
shot of the
en corrected
on on a CNT
olution is us
stances, such
ilever to tip
height prof
imitation (se
y. First of all
oduced CNTs
fer accordin
lways under
used for bio
xicological be
oduction
13
p/surface
he height
scanned
). In fact,
ession of
surface.
d back to
T sample
ually not
h as the
p/surface
ile of the
ee Figure
l, the big
s and for
g to the
rgo some
omedical
ehaviour.
Chapter 1
14
Therefore results should always be considered as referred to the particular kind of CNTs they
have been obtained with, and could not be generalized.
When considering CNT toxicity it is important to distinguish among (i) biocompatibility
related to possible biomedical application of CNT‐based systems, and (ii) safety related to CNT
manipulation by all the workers in the field. In the first case it will be referred to processed CNTs,
likely administered in solution via injection (this topic will be reviewed later, see Chapter 1.5),
while in the second case the main risk is related to inhalation, and it refers not only to processed
CNTs but, primarily, to pristine ones.
Pristine MWCNTs have been compared to asbestos, since they tend to form fibrous
aggregates whose dimensions (in the micrometre range) and shape are similar to those reported
to be carcinogenic for asbestos. Once the latter non‐degradable and bio‐persistent particles are
inside the body, they deposit. Thus, unsuccessful inflammation and scavenging occur towards
them (frustrated phagocytosis), leading to continuous oxidative stress. Furthermore, there is a
contribution, given by iron impurities in the samples, that accelerates the generation of reactive
oxygen species (ROS). This being true for asbestos, the possibility for CNTs to share the same
carcinogenic mechanism was investigated.37 Experiments on p53 heterozygous mice (reported to
be sensitive to asbestos and to fast develop mesothelioma), intraperitoneally administered with
MWCNTs, showed the induction of mesothelioma, leading to death of the animals within 25
weeks. Nevertheless, it is necessary to underline that these experiments were carried out with
micrometre sized MWCNTs particles (10‐20 µm) and there is no evidence, as the authors pointed
out, that the same effect would occur using pure nanometre‐sized CNTs. In a similar study CNTs,
introduced into the abdominal cavity of mice, were proved to cause asbestos‐like pathogenicity
(inflammation and granulomas formation) in a length‐dependent manner, namely only fibers
long enough (> 20 micrometre) were able to cause pathogenic response during the 7 days
experiment.38
Nonetheless, being based on intraperitoneal injection, none of these studies addresses
the question if and how much CNTs accumulate in the mesothelium after inhalation. To answer
this question other studies on inhaled CNT effects should be considered. Mitchell et al. exposed
male mice to MWCNTs (5‐15 µm long) aerosol in a whole body inhalation chamber.39 The aerosol
was generated by a jet mill and particles bigger than 3 µm were removed in order to work within
a range respirable by a rodent. It is necessary to underline the difference between “inhalable”,
which refers to the fraction of airborne material which enters nose and mouth during breathing
and is therefore available for deposition in the respiratory tract, and “respirable”, referring to
the fraction able to penetrate to the gas exchange region of the lung. The authors observed the
pres
dam
Nev
afte
MW
5‐15
mg/
fibro
asbe
Baro
typic
proc
part
pred
gene
volu
part
was
from
mon
F
hand
duri
sence of bla
mage were
ertheless th
r inhalation
WCNTs in sub
5 µm) and th
/m3 for 6 h w
osis, which
estos causes
It is imp
on et al. stud
cal workplac
cess) led to
ticles smaller
dominantly o
eration rates
ume of fume
ticles. A field
also conduc
m one contai
nitoring of ae
Figure 8. Pou
There w
dling of unr
ng material
ack particles
evidenced w
he treatment
n of environ
b‐pleural mac
he treatmen
with a nose‐o
could be c
pleural infla
portant to a
died, in 2003
ce exposure
the genera
r than 100 n
of nanotube
s were appr
ed alumina, a
evaluation o
cted. The bul
ner into ano
erosol numb
uring HiPCO S
was no clea
refined nano
handling we
inside alve
with exposu
t caused sys
nmental poll
crophages.40
t consisted
only inhalatio
onsistent w
ammation an
nalyze the r
3, the tende
to this mate
ation of par
nm, even if i
es, catalyst p
oximately tw
another low‐
of the airbor
lk material w
other one) in
er and mass
SWCNTs for
r evidence
otube mater
ere estimated
olar macrop
ure to up t
stemic imm
lutants. Oth
0 In this case
in a single in
on chamber.
with the ind
nd diffuse ple
real risk for
ency of SWC
erial.41 Vigor
rticles below
t was unclea
particles, or
wo orders of
‐density mat
rne contamin
was removed
n a closed are
s concentrati
airborne con
of increase
rial in the f
d to be lowe
phages, but
to 5 mg/m
unosuppress
her authors
e CNTs were
nhalation ex
. The author
uction of m
eural fibrosis
people wor
NTs to form
ously agitati
w 10 μm ae
ar whether t
compact ca
f magnitude
terial compr
nation, while
d from the re
ea with inlet
on (Figure 8
ntamination
d aerosol m
field. The ae
er than 53 μg
no lung inf
3 for a max
sion, respon
also report
longer (0.5‐
xposure at a
s also observ
mesotheliom
s.
rking with C
aerosols, in
ng SWCNTs
erodynamic
these smalle
arbonaceous
lower than
ised of nano
e manipulati
eactors and m
of filtered a
).
study (adapt
mass concen
erosol nano
g/m3, a value
Intr
flammation o
ximum of
nse already
ted accumu
‐50 µm com
concentrati
ved focal su
ma, even if
CNTs to inha
n order to ev
(produced b
diameter,
er particles c
s particles. H
those from
ometre‐sized
ng the bulk S
manipulated
air, with a co
ted from Ref
ntrations du
otube concen
e that is belo
oduction
15
or tissue
14 days.
reported
lation of
pared to
on of 30
b‐pleural
normally
ale them.
valuate a
by HiPCO
and also
consisted
However,
a similar
d primary
SWCNTs,
d (poured
ntinuous
f. 41).
uring the
ntrations
ow those
Chapter 1
16
associated with re‐suspension of ambient dust due to personal movement, and cleaning
operations following handling of material. Importantly, these experiments could therefore lead
to the conclusion that a typical aerosol concentration of CNTs during handling of the material is
far below the ones used in the in vivo inhalation studies conducted so far, thus making those
results not relevant. Nevertheless, the CNTs are not the same, so it is necessary to be careful in
making comparison.
It is still not possible to reach a clear, unambiguous conclusion on CNTs inhalation
toxicity, and therefore, precautions should be taken to prevent any unnecessary release of
respirable particles, when handling CNTs, especially the pristine ones, but without getting into a
panic. Also, to avoid exposure and guarantee respiratory protection, an adequate personal
protective equipment should be used.
1.5 CNT as delivery system for cancer
Anticancer therapy, though being often very effective, still suffers of some severe
drawbacks, which render vital and urgent the need for further research. One of the main
limitations of almost every cancer treatment so far is the lack of selectivity for tumour tissues. In
fact cytotoxic drugs (or other therapies) do not exert their action solely on cancer cells, but also
on healthy organs, resulting in severe side effects for patients, therefore limiting their
compliance and the maximum administrable dose. Another major problem related to anti‐
neoplastic chemotherapy is Multi‐Drug Resistance (MDR).42 This phenomenon can be often
ascribed to the activity of an efflux pump, namely P‐glycoprotein (P‐gp), able to recognize the
drug and transport it out of the cell once it has been internalized, thus preventing it from
exerting its cytotoxic action.43 Importantly, the P‐gp mediated MDR often characterizes residual
tumour cells after chemotherapy and tumour stem cells. Once this capability is acquired, it can
be directed towards many different drugs to which the tumour had never been exposed before,
thus heavily hampering chemotherapy’s efficacy.
Nanotechnology can help in overcoming these limitations; in fact a nanovector could be
conceived to be targeted towards cancer, either only in a non‐specific way, or with a specific
targeting agent attached to the carrier itself. The non‐specific targeting is based on the
Enhanced Permeability and Retention (EPR) effect (Figure 9).44 Cancer tissue is usually
characterized by a rapid and defective angiogenesis, resulting in leaky blood vessels with large
fenestrations, due to endothelial cell disorganization. Moreover, the smooth‐muscle layer in the
vascular wall is frequently absent or abnormal, leading to passive dilatation of vessels. The
consequence of these characteristics is an enhanced extravasation of macromolecules in the
tum
and
rete
on N
drug
bind
spec
diffe
F
curr
part
whe
lipos
or si
its t
plet
func
solu
pres
as g
betw
our tissue, w
undergo ren
ention of acc
Nano‐sized s
g (small mo
ding to the n
cific tumour
erent metabo
Figure 9. Enh
Carbon
rently under
ticles coming
en a biologi
somes or po
ilica ones.
When a
oxicity profil
hora of vari
ctionalized C
uble function
serving their
good dispers
ween a high
whereas low
nal clearance
umulated m
systems (ma
lecule). Mor
nanocarrier t
markers. At
olic and cellu
hanced Perm
nanotubes
study. They
g from the sy
cal applicat
olymeric nan
material is
le and its in
iables such
CNTs are no
nalized CNT
activity,46 a
ions.47 Cohe
er degree of
w molecular
e. Moreover,
acromolecul
cromolecule
reover, this
targeting age
t the same t
ular pathway
eability and
(CNTs) are
y are made
ynthetic proc
ion is inten
oparticles, w
proposed fo
vivo fate. Ev
as the diffe
on‐cytotoxic.
s are uptak
nd it has bee
rently with t
f functionali
weight‐mol
, slow venou
les in the tum
es) would ac
effect can
ents as antib
time, the de
ys, thus givin
Retention e
system.
among the
almost entir
cedure, whic
nded. They
with the stab
or biomedica
ven though t
rent CNT fu
In fact, it
ken by imm
en found tha
these obser
zation for SW
ecules can r
us return and
mour. For th
cumulate m
be further e
bodies or ot
elivery of a c
ng the possib
ffect for a m
.
most prom
rely of carbo
ch are usuall
combine th
bility of inorg
al application
the question
nctionalizati
has been d
mune system
at CNTs are
vations, oth
WCNTs and
rapidly diffu
d poor lymph
ese reasons,
ore in the tu
enhanced in
ther molecul
carrier driven
bility to elude
model CNT‐ba
ising possib
on, a part fr
y removed b
erefore the
ganic nanopa
ns, it is fund
is still unde
ons, it appe
emonstrated
m cells witho
more toxic a
er authors r
a lower tox
Intr
use in the ci
hatic drainag
, drug delive
umour than
n a specific
les able to r
n drug could
e MDR.45
ased drug de
le drug nan
rom residual
by a purifica
e biocompat
articles, such
damental to
er debate, du
ears that ad
d that highl
out being t
as agglomera
reported cor
xicity toward
oduction
17
rculation
ge lead to
ery based
the free
way, i.e.
recognize
d involve
elivery
novectors
metallic
tion step
ibility of
h as gold
establish
ue to the
equately
y water‐
oxic and
ates than
rrelations
s human
Chapter 1
18
dermal fibroblast. Interestingly, covalently modified SWCNTs appeared to be less cytotoxic than
the ones stabilized trough surfactants.48 Importantly, numerous studies by different groups have
proven so far how surface‐modified CNTs are well tolerated in vivo.49 In general they have a
blood clearance half‐life in the order of hours. Moreover, tissue biodistribution studies showed
that they are eliminated into urine, via glomerular filtration, or into faeces with low residual
amounts in the body. All these results indicate how the in vivo behaviour of these materials
could be modulated by the degree and kind of functionalization, two critical aspects that need to
be accurately controlled.
Furthermore, some recent works showed in vitro enzymatic degradation of SWCNTs,
thus presenting another possibility for the elimination of this carrier from the body once the
therapeutic function has been exerted.50 Importantly, it has been shown that MWCNTs take
more time to be degraded than SWCNTs, and the process seems to occur gradually from the
external walls towards the inner ones. Moreover oxidized MWCNTs were more easily degraded
than pristine tubes, indicating that defects introduced through oxidation probably facilitate the
attack from the enzyme.51 Anyway further studies, especially in vivo, are necessary to assess if
CNT metabolic digestion is actually possible.
1.5.1 Uptake Mechanism
Carbon nanotubes are able to enter cells, as recognized by the scientific community, but
the way it happens is still controversial. In fact, although many studies have been carried out so
far to understand which is the uptake mechanism, still there is not a unique answer to the
question. The type of cell should be taken into account when comparing different results and,
more importantly, the type of CNTs could play a crucial role. In fact, the great heterogeneity of
this material implies the possibility of heterogeneous behaviour in a biological environment.
CNTs can differ a lot in size and functionalization, thus giving a plethora of derivatives, from
almost naked to differently coated ones.
Among the existing hypotheses, one is phagocytosis, a process employed by specialized
cells, in fact different authors observed engulfment of CNTs by phagocytes. Cherukuri et al.
treated mouse peritoneal macrophage‐like cells with SWCNTs and visualized the nanotubes
inside the cells exploiting their spontaneous NIR fluorescence.52 They observed the tubes
confined in small vesicles, probably phagosomes, derived from an active ingestion process,
which is consistent with their observation of a temperature‐dependence of the uptake. Porter et
al. studied the uptake of SWCNTs by human monocyte derived macrophage cells and found
them inside lysosomes and phagosomes,53 while the treatment of the same cell line with
MW
size
µm)
were
cons
in th
obse
did
poss
whic
show
mem
of C
in 2
biot
over
cons
the
sugg
WCNTs led to
of the nano
.54 Four wee
e found insid
sequent cell
his case from
erved their i
not specifi
sibilities: pha
ch are phag
wing single n
The nee
mbranes, and
NTs by mam
Figu
Kam et
004, treatin
in and fluore
rlap the sig
sistently with
lack of int
gested an en
o frustrated
otubes used
eks after the
de lysosome
ular stress w
m few to seve
internalizatio
ically invest
agocytosis, s
gocytic cells,
nanotubes pi
edle‐like pe
d the endocy
mmalian cells
ure 10. Nano
al. proposed
g human le
esceinated s
gnal of an
h energy‐dep
tracellular fl
ndocytotic pr
phagocytosi
(diameter o
e subcutaneo
s of macroph
was observed
eral µm).56 O
on by mouse
tigate the
supported by
and a nee
iercing the ce
enetration, c
ytosis repres
(Figure 10).
o‐needle (a) a
d for the firs
ukaemia cel
streptavidin.
endosome
pendent end
uorescence
rocess for SW
s, an incom
of about 70
ous implanta
hages.55 On
d for human
Other author
e microglia c
uptake mec
y the higher
dle‐like pen
ell membran
consisting i
sent the two
and endocyt
st time the e
lls and hum
The intrace
marker. Mo
docytosis.58 S
at 4°C and
WCNTs funct
plete phago
nm and len
ation of 200
the other ha
monocytes
rs also used
cells and mu
chanism, th
r internalizat
etration, su
ne.
n a direct
o more impo
tosis (b) pene
endocytosis i
an T cells w
llular fluores
oreover, the
Similarly, in o
d the co‐sta
tionalized w
cytosis, prob
gth ranging
0 nm long M
and, frustrate
treated with
long MWCNT
urine glioma
heir results
tion observe
pported by
insertion o
rtant mecha
etration mec
nternalizatio
with SWCNTs
scence distri
e uptake w
other works
aining with
ith fluoresce
Intr
bably due to
from few to
MWCNTs in r
ed phagocyt
h long MWCN
Ts (200‐400
a cells.57 Eve
suggest tw
ed for microg
TEM image
of CNTs ac
nisms for th
chanism.
on pathway
s functionali
ibution was
was blocked
by the same
endosome
ently labelled
oduction
19
o the big
o several
ats, they
osis with
NTs (also
µm) and
n if they
wo main
glia cells,
es clearly
ross cell
e uptake
for CNTs
zed with
found to
at 4°C,
e authors
markers
d DNA or
Chapter 1
20
proteins (MW < 80kDa).59 The authors speculated about a hydrophobic interaction between the
naked SWCNTs surface and the cell membrane as the driving force for the internalization. In a
later work, they expanded the studies to precisely understand the uptake mechanism.60 The
internalization inhibition was observed with NaN3 pre‐incubation of the cells, confirming
endocytosis, since ATP‐production and therefore energy‐dependent activities are strongly
reduced by this chemical agent. Moreover, with a well‐designed experimental set endocytosis
was found to happen through clathrin‐coated pits, rather than through caveolae or lipid‐raft
pathway. Endocytosis has been suggested as the uptake pathway also by other authors. DNA‐
wrapped SWCNTs were tracked during in vitro trafficking exploiting their intrinsic NIR
fluorescence.61 Each trajectory was then associated with a specific kind of motion, recognizing
that SWCNTs were adsorbed onto the membrane and internalized through endocytosis. Once
inside the cell they diffused consistently with confinement inside vesicles, and they finally
underwent exocytosis. Moreover, co‐localization experiments showed an overlap between
SWCNTs and lysosomes inside the cells, confirming again the endocytotic uptake. The authors
proposed this mechanism as a result of the SWCNTs coverage by the proteins present in the
medium. This coating should lead to a subsequent clustering in aggregates, big enough to reach
the radius cutoff thermodynamically required for being endocytosed.62 Wang et al.
functionalized MWCNTs with the enzymatic toxic domain of ricin (RTA) in a non‐covalent way,
and studied its uptake mechanism by HeLa cells, observing clathrin‐mediated endocytosis.63 The
internalization of SWCNTs‐QD‐streptavidin by CD3 positive‐leukaemia cells was also studied,
monitoring the intracellular QDs fluorescence.64 In this case the system was rather complex, in
fact the uptake was mediated by the recognition of three components. First, a biotinylated anti
CD3‐antibody bound CD3 receptor on cell membrane and subsequently the streptavidin‐bearing
SWCNTs attached the biotinylated antibody, showing high internalization. On the contrary, poor
internalization was observed when the antibody was absent, with the use of non‐biotinylated
antibody, when experiments were carried out at 4°C and when the cells did not express CD3‐
receptor on the membrane (CD3 negative cell line). All these data suggest a receptor‐mediated
endocytosis for this construct.
To target cancer cells, Bhirde et al. described the use of a drug delivery system based on
a cisplatin‐SWCNT conjugate, exploiting the interaction of the epidermal growth factor (EGF)
with its receptor (EGFR), over‐expressed on the surface of many cancer cells.65 The authors
demonstrated the efficient internalization of the SWCNT‐EGF conjugate in vitro. Moreover they
proved that it is actually mediated by the ligand‐receptor interaction, since the uptake was
drastically reduced knocking down the receptor expression. These findings are consistent with
Introduction
21
receptor‐mediated endocytosis. Nevertheless, a certain degree of internalization observed even
without receptor let consider other possibilities, as stated by the authors themselves. Another
paper conceiving two options for the uptake mechanism has been published in 2008.66 The
authors prepared SWCNTs bearing taxoids and biotin molecules, as a drug delivery system
targeted towards cancer cells over‐expressing the biotin receptor, demonstrating the actual
internalization of the whole conjugate, hindered by both incubation at 4°C and presence of
NaN3, and strongly reduced by saturation of accessible receptors via biotin pre‐treatment, in
accordance with receptor‐mediated endocytosis. On the other hand, they observed a strong
decrease in the uptake of SWCNT derivatives without biotin at low temperatures (4°C), but no
changes when NaN3 was added. This behaviour is consistent with a non‐endocytotic process, as
the needle‐like diffusion through membranes, which is temperature‐dependent but energy‐
independent.
It is necessary to consider that temperature can affect not only endocytosis but also
membrane piercing, in fact as simple diffusion of a molecule in a liquid is influenced by
temperature, the same could happen to CNTs in biological media. More importantly, the fluidity
of cellular membranes, which are made of phospholipidic bilayers, is strongly dependent on
temperature. As a consequence, the mobility of an object in the bilayer will increase with
temperature. Thus, it is not surprising that a decrease in temperature has been sometimes
correlated with a decrease in cellular uptake of CNTs, and this experimental observation alone
should not necessarily mean endocytosis, unless it is corroborated by other evidences. Therefore
some cases of CNTs internalization ascribed to endocytosis by the authors could be also
differently interpreted.67 The same observation can be applied to the work published by Cheng
et al.68 In this well‐conceived study of cellular internalization of fluorescently labelled SWCNTs
derivative by several mammalian cell lines, the authors compared the results obtained in normal
conditions with the ones obtained at 4°C and in the presence of sodium azide. While for the
former the uptake was almost completely blocked, the latter showed an intracellular fluorescent
signal weaker but comparable with the normal conditions, differently from what stated by the
authors (for a better understanding see Fig. 2B of Ref. 68). Therefore it is possible to conclude
that both an energy‐dependent and an energy‐independent mechanism are contributing to the
CNTs uptake by cells in this case. Moreover the author co‐incubated cells with labelled Dextran,
known to be uptaken by endocytosis, and found it in the endosomes, while the fluorescence
from the SWCNTs was in the nucleus, indicating that either the retention time inside the
endosomes or the entire uptake pathway was not the same for the two compounds.
Chapter 1
22
The nano‐needle penetration mechanism for CNTs was proposed in 2004. Internalization
of functionalized CNTs was found even in energy‐depleting conditions, and ultrathin transverse
sections of HeLa cells incubated with the CNTs were analyzed by TEM, observing MWCNTs inside
cells and in the act of crossing the membrane.69 Therefore the authors hypothesized that the
cylindrical shape of CNTs let them directly insert across biological membranes, like a Nano‐
needle. This description is in accordance with what observed for some peptides and proteins
exhibiting non‐classical transport activity70 and also with what has been predicted for nanotube‐
shaped objects.71 Another theoretical study, instead, reported that the energy cost for a SWCNT
to insert into a model phospholipid bilayer is too high to occur only because of thermal motion
energy.72 Nevertheless further studies on the penetration of a patterned CNT were performed
by the same authors. Since probably CNTs in a biological environment will not have a
homogeneous surface, but will be covered by biomolecules, they studied a CNT covered by
alternating bands of hydrophilic and hydrophobic regions. This model can be also representative
of a CNT functionalized with hydrophilic groups, since usually functionalization do not cover the
whole CNT surface. In this case the energy required by CNTs to traverse the bilayer was much
lower, thus resulting in enhanced penetration.73 The same internalization pathway was proposed
for a huge selection of cell types, since a comprehensive study revealed effective CNTs
internalization even by cells unable to carry out endocytosis or under conditions that prevent
energy‐dependent processes.74 The cells were treated with differently functionalized SWCNTs
and MWCNTs, which were not coated by macromolecules. As speculated by the authors, the
discrepancy between these results and the endocytotic hypothesis can lie in the different kind of
CNTs derivatives used.
Zeineldin et al. wrapped PL‐PEG2000 onto SWCNTs and found that when the PEG
integrity was preserved, the coating prevented the non‐specific uptake by cells (different ovarian
cancer cell lines), while the PEG fragmentation (by sonication for a prolonged period, i.e. 1 hour)
allowed CNTs internalization by cells.75 A further functionalization of the PEGylated, not
sonicated, SWCNTs with a specific targeting molecule led to selective uptake by cancer cells
over‐expressing the receptor for that molecule. Although the authors did not speculate on
possible uptake mechanisms, these findings underline the importance of CNTs coatings in
determining their biological fate.
CNT size seems to have a role as well. A recent work compares the internalization by human
hepatocellular carcinoma cells of different CNTs: namely MWCNTs with a diameter of 10‐30 nm
(long and short, i.e. 1‐2 µm or 0.5‐1 µm long) and SWCNTs with a diameter of 1‐3 nm (long and
short, i.e. 100‐200 nm or 50‐100 nm long).76.All the CNTs were non‐covalently functionalized
Introduction
23
with chitosan. The authors reported internalization only for SWCNTs, and with different
mechanism according to size: for long SWCNTs the mechanism observed was mainly clathrin‐
mediated endocytosis, while for short SWCNTs they describe also endocytosis via caveolae and
direct insertion across cell membrane.
Evidences supporting the piercing hypothesis were reported elsewhere. SWCNTs were
observed translocating across the lipid bilayer and entering the cytoplasm.53,77 SWCNTs and
MWCNTs were found across the plasma, lysosomal, and nuclear membranes,78 and, by TEM,
MWCNTs were observed both inside cytoplasmic vacuoles of epidermal keratinocytes
(consistently with endocytosis) and piercing the nuclear membrane.79 Similarly, as
aforementioned, Kateb et al. presented TEM visual confirmation of membrane insertion of
MWCNTs.57
A very interesting study, published in 2008, confirmed the ability of CNTs to cross
biological membranes, considering the question from a rather different point of view. MWCNTs
were visualized while traversing the fenestrated endothelium in the glomerular filtration system
of mice and crossing the renal filtration membrane. In order to do so the tubes had to adopt an
adequate spatial conformation, facing the filtration barrier perpendicularly, since the
fenestration size is compatible only with the traverse dimension of these MWCNTs, to let them
pass and pierce the underlying basal membrane.49c
Further hint for understanding CNT ability to cross cell membranes could be given by
another paper.80 During an in vitro experiment MWCNTs were guided through a magnetic field
towards mammalian cells and they cross the membrane spearing it, and thus translocating inside
cells their cargo, i.e. a vector containing the sequence for a fluorescent protein, that allowed the
recognition of cells internalizing it. SEM pictures of cell surface confirmed the presence of the
CNTs that actually look like spearing it. This work aimed to prove that actively guiding CNTs
towards cells, it is possible to induce a temporary permeabilization of the membrane, as effect
of its Nano‐mechanical penetration. The system was thought to be exploited for delivery
purposes, since, despite the great efficiency of transfection, it leads to minor perturbation of
cells. A recently published work confirmed this property, suggesting the possible use of CNTs as
cellular endoscopes.81 The authors of this paper prepared an endoscope ending with a MWCNT
(with a diameter in the range of 50‐200 nm or more and a length of 50‐60 µm) and found that,
thanks to its narrow cylindrical shape, it produced no cell damage, and it resulted in less
mechanical stress for the cell compared to common conical glass pipettes, even for prolonged
periods.
Chapter 1
24
In conclusion, many different studies aimed at understanding CNT uptake mechanism by
cells, considering the question from different points of view. The discrepancy between the
results present in literature can lie in the different kind of CNTs derivatives used. Considering all
the works published so far, we could conclude that both endocytosis and needle‐like piercing of
the membrane exist and can contribute to internalization of CNTs, with a predominance of
needle penetration when internalization is driven by the CNTs, i.e. when CNTs are functionalized
with non‐bulky molecules. Whereas, when CNTs are coated with big molecules or with
molecules that trigger a specific receptor‐mediated endocytosis, this mechanism prevails.
1.5.2 Delivery of antineoplastic chemotherapeutic drugs
The most studied possibility in CNT‐based cancer therapy is drug delivery, in which CNTs
are used as vehicles for a chemotherapeutic agent, in order to improve its delivery to the target.
Doxorubicin is a drug widely used in clinical cancer treatment and it is very interesting to
study its delivery through nanocarriers. So far, a number of works in which carbon nanotubes
were used for this purpose has been published. In one of the first papers doxorubicin was bound
to copolymer‐coated MWCNTs forming a supramolecular complex based on π‐π stacking.82 The
cytotoxicity of this derivative was tested in vitro on human breast cancer cells showing an
increased mortality rate in comparison with doxorubicin alone. This drug was also bound,
through supramolecular chemistry, to two different water‐soluble SWCNTs: non‐covalently
functionalized with a phospholipidic‐PEG surfactant and oxidized SWCNTs covalently PEGylated
through amidation of the carboxylic moieties.83 The authors suggested π‐π stacking and
hydrophobic interactions as the binding forces between doxorubicin and nanotubes. Moreover
they report a pH‐dependent interaction: the loading was made at pH 9, condition at which
doxorubicin is deprotonated and so has low water solubility, reaching a doxorubicin/nanotubes
weight ratio of ~4:1; a decrease in the pH led to the release of the drug from the SWCNT carrier
due to the its consequent higher hydrophilicity. This mechanism could be very useful in cancer
therapy due to the acidic environment of extracellular tissues in tumour, potentially leading to
selective drug release in vivo. The cytotoxicity of the phospholipidic‐PEG wrapped SWCNT‐
doxorubicin construct was tested on human glioblastoma cancer cells. The derivative induced
cell death similarly to free doxorubicin at a concentration of 10 µM, although the observed IC50
value was higher (~8 µM for the carbon nanotube conjugate compared to ~2 µM for free
doxorubicin). Furthermore, a targeted doxorubicin delivery was tested using a cyclic arginine–
glycine–aspartic acid (RGD) peptide, which acts as recognition moiety for integrin αVβ3 receptors,
over‐expressed in a wide range of solid tumours. The targeting agent was bound on the PEG
Introduction
25
chain of the non‐covalent SWCNT derivative, and then doxorubicin was loaded. This conjugate,
tested on integrin αVβ3‐positive cells, showed enhanced drug delivery compared to the
derivative without RGD, according to the degree of brightness in confocal fluorescence
experiments and to the IC50 value, which was smaller (~3 µM) for the RGD‐bearing derivative
than for the one void of any targeting agent. The behaviour of phospholipidic‐PEG wrapped
SWCNT‐doxorubicin complex was studied also in vivo into SCID mice bearing Raji lymphoma
xenografts.84 The authors observed higher tumour uptake, probably due to the prolonged
circulation half‐life of the construct, and greater inhibition of tumour growth with respect to the
free drug but the therapeutic efficacy was not better than DOXIL (liposomal Doxorubicin). On the
other hand, differently from DOXIL, SWCNT‐doxorubicin caused neither significant toxicity nor
mortality, allowing the use of higher doses. Treatments with 10 mg/kg (instead of 5 mg/kg) of
SWCNTs‐doxorubicin (dose normalized on doxorubicin) led to improved efficacy, still without
causing any severe toxicity, whereas 5 mg/kg of free doxorubicin or DOXIL resulted in strong
decrease in the weight of the animals and mortality rates of 20% and 40%, respectively. The
possible reasons for this behaviour involve different aspects: the bigger size of the SWCNT
construct with respect to free doxorubicin probably slowed down glomerular filtration, and,
moreover, the PEG coating could have hidden doxorubicin from macrophages. Both these
effects could determine prolonged blood circulation, thus enhancing tumour accumulation
because of EPR effect. Moreover the slightly acidic microenvironment of tumours should
facilitate doxorubicin detachment from the SWCNTs, as previously reported,83 therefore leading
to selective drug release. Nevertheless the same authors observed a slow in vivo dissociation of
doxorubicin from the SWCNTs after administration, and this fact can suggest that non‐covalent
constructs cannot guarantee proper in vivo stability. An in vivo targeting study was also
performed with PEGylated SWCNTs functionalized with RGD peptide.85 After tail vein injection
into U87MG tumour‐bearing mice, the nanotubes accumulated in tumour (13% ID/g in 24 h),
with no obvious toxicity or negative health effects for the animals. Zhang et al. prepared a
different kind of SWCNT‐based vector for doxorubicin.86 First CNTs were oxidized, thus
introducing carboxylic groups, and then derivatized via wrapping with two different
polysaccharides, namely sodium alginate, which is anionic, and chitosan, which is cationic.
Doxorubicin was then loaded on the modified SWCNTs and the interaction was mediated
probably by electrostatic interactions and π‐π stacking. The release of doxorubicin in acidic
condition (pH 5.5) was assessed, in order to mimic the typical environment of lysosomes or
cancerous tissues, finding a release of almost 40% in 72 h, while the construct was stable at pH
7.4. Finally, an identical conjugate bearing folic acid (linked to chitosan) was prepared, to target
Chapter 1
26
tumours overexpressing the folate‐receptor. In vitro tests on HeLa cells were performed to
evaluate cellular uptake, by fluorescence microscopy, and cell viability. The compound led to
stronger reduction in cell viability then the drug itself, proving its efficacy as targeted delivery
system. A triple functionalized SWCNT derivative was prepared, bearing doxorubicin, fluorescein
and a targeting antibody that recognizes carcinoembryonic antigen (CEA), a tumour marker
relevant for a variety of adenocarcinomas.87 Doxorubicin was non‐covalently loaded on oxidized
SWCNTs by π‐π stacking. Afterwards a fluorescein‐labelled BSA was covalently attached through
its amines to the carboxylic groups present on the oxidized SWCNTs, via amidation reaction.
Finally the antibody was tethered through its amines to the BSA carboxylic groups again with an
amidation step. In this way the authors designed a construct that should target CEA‐
overexpressing cancer cells, be internalized by endocytosis, and release doxorubicin at lysosomal
acidic pH. The fluorescein was introduced to localize the SWCNT construct by the fluorescent
signal after doxorubicin release. The internalization by WiDr human colon cancer cells was
studied by confocal microscopy: after a 4 h treatment, the fluorescently labelled SWCNTs were
found in the cytoplasm, while doxorubicin was localized in the nucleus, where the drug exerts its
activity. As a control, the fluorescein‐labelled BSA did not show internalization to the same high
extent, indicating the good ability of CNTs to act as vectors. Nevertheless no information on the
cytotoxicity of the complex is reported, nor any stability study to evaluate the release of
doxorubicin from CNTs. With a different approach, doxorubicin was linked to pyrene and this
system was non‐covalently bound to SWCNTs via π‐π bonding driven by the polyaromatic unit.88
The linker between doxorubicin and pyrene was a short chain with a carbamate group, designed
to be enzymatically cleaved inside lysosomes. In vitro studies on a mouse melanoma cell line
were performed to assess the internalization of the construct and its cellular toxicity. The results
showed accumulation in the lysosomes and cell growth inhibition (via apoptosis) in a time‐ and
dose‐dependent manner, approaching the doxorubicin profile for the 72 h treatment.
Furthermore, the therapeutic efficacy was evaluated in vivo on melanoma‐bearing mice. Free
doxorubicin and SWCNT‐doxorubicin induced similar reduction of tumour volume, with a
significant decrease of systemic toxicity for the CNT construct, according to animal body weight,
organs weight and histopathological analyses. Also in this case the release of the drug from the
CNTs was studied, comparing the effect of incubation with cell lysate to the use of normal
buffer. Though the former system was significantly more efficient, proving that intracellular
enzymes play a fundamental role, a release was observed also in the latter case, demonstrating
how a non‐covalent construct is not completely stable in biological environment. With the
double aim of obtaining a targeted drug delivery system and of overcoming MDR, SWCNTs were
Introduction
27
functionalized via amidation with an antibody directed towards P‐glycoprotein, one of the main
actors of MDR.89 Subsequently the Ab‐SWCNTs construct was non‐covalently loaded with
doxorubicin, via π‐π stacking. This construct was tested in vitro on human leukaemia cells,
overexpressing P‐gp on their surface and, therefore, resistant to doxorubicin. Cells were exposed
to NIR radiation to trigger doxorubicin release, as a consequence of NIR absorption by SWCNTs.
This system was indeed able to triplicate the amount of doxorubicin released in 24 h, but it
should be underlined that a partial release was observed without NIR radiation as well. The Ab‐
CNT construct resulted to be bound to the cell membrane and doxorubicin was localized inside
cells, in much higher amount with respect to treatments with free doxorubicin. Doxorubicin
alone or together with anti P‐gp antibody (at a concentration of 5 µg/mL) could partially inhibit
cell growth, but, increasing the incubation time after treatment (from 24 h, to 48 h, to 72 h),
cells gradually recovered and started growing again. Instead, very interestingly, the SWCNT
construct, used at an equivalent concentration of drug, inhibited cell growth more efficiently
than the free drug and in a time‐dependent way, probably due to a slow drug release from the
vector surface.
Methotrexate is another well‐known anticancer drug, which suffers, as many others,
from low bioavailability and toxic side effects. Preliminary studies of CNT‐based drug delivery
with this drug were performed by Pastorin et al.90 MWCNTs were covalently functionalized with
methotrexate and fluorescein at the same time and results obtained on human Jurkat T
lymphocytes showed a rapid internalization of the compound, even though the efficacy of the
construct was lower than the free drug. In a more recent work the same drug was bound to
MWCNTs through cleavable linkers. The use of a peptide recognized by intracellular proteases
led to a significant decrease in MCF‐7 breast cancer cell viability if compared to free
methotrexate.91 Therefore it is possible to conclude that the lower efficacy in the first approach
was probably due to the lack of drug release once the construct was internalized.
Taxanes represent another class of antineoplastic drugs that, to date, have already been
conjugated to CNTs for new drug delivery approaches. Paclitaxel (PTX), which belongs to this
class, was covalently bound through an ester bond to the PEG of non‐covalently PEGylated
SWCNTs and i.v. injected into xenograft tumour‐mice (PTX resistant 4T1 murine breast cancer
mice model) to test the in vivo efficacy.92 Results were very promising, showing a tumour growth
inhibition value of almost 60% for the SWCNT‐PTX derivative, much higher than the one
reported for Taxol and PEG‐PTX (28% and 21% respectively). Higher apoptosis level and lower
proliferation active cells level were observed after CNT derivative treatment, when compared to
the drug alone. It is important to underline the lack of toxicity in mice for these PEGylated
Chapter 1
28
SWCNTs themselves, demonstrated by the authors in another work.49e The evaluation of the
pharmacokinetics showed a longer plasmatic half‐life for SWCNT‐PTX, coherently with the
enhanced hydrophilicity of the conjugate with respect to the drug itself, and much higher PTX
presence in RES organs (liver/spleen) and intestine 2 h after injection. This is a predictable
behaviour for nanomaterials in general and could give concern for the toxicity towards these
organs. Nevertheless the authors reported differences between the biodistribution of SWCNT
and PTX, indicating a rapid release of the drug from the conjugate probably due to ester
cleavage by carboxylesterases. As a consequence the drug seemed to be rapidly excreted,
lowering its toxicity. Importantly, tumour PTX levels were 10 times higher for SWCNT‐PTX
derivative than for free taxol 2 h after injection and tumour‐to‐normal organ/tissue PTX uptake
ratio was bigger, thus indicating a better selectivity of the CNTs‐delivered drug. Chen et al
performed another study with the same class of drugs.66 They prepared a covalent derivative of
SWCNTs bearing a taxoid molecule and biotin as targeting unit towards cancer cells
overexpressing its receptor, thus leading to receptor‐mediated endocytosis. This construct was
intended to be a prodrug, which frees its cargo, once inside the cell, upon reduction of the
disulphide bond by endogenous thiols such as glutathione (GSH), whose concentrations are
typically more than 103 times higher in tumour tissues than in blood plasma. Therefore, the
system should act specifically on cancer cells, leading the taxoid to carry out its mitosis inhibition
action only in the desired sites. The authors demonstrated the actual internalization of the
whole conjugate in leukaemia L1210FR cells, by means of confocal fluorescence microscopy,
thanks to the presence of a fluorescent tag covalently bound to the taxoid molecule. Moreover,
in the case of a control SWCNT derivative without biotin, they observed a temperature‐
dependent but energy‐independent internalization. This is coherent with a non‐endocytotic
mechanism, in accordance with the hypothesis of a needle‐like diffusion of carbon nanotubes
through cell membranes. On the other hand, for the biotin‐conjugate, the mechanism observed
was endocytosis, and the internalization degree was far higher if compared to the one for cells
not expressing the biotin receptor. Furthermore the authors proved the efficient release of the
drug by GSH and the binding of the conjugate to the microtubule network, where it exerts its
cytotoxic action, with an IC50 value smaller than the one for the drug alone, probably due to the
increase in the drug uptake.
Also platinum analogues have been investigated in CNT‐based drug delivery for cancer
treatment. In a first work, Lippard and co‐workers reported the preparation of a Pt(IV)‐based
SWCNTs prodrug able to deliver and release the drug inside cells, leading to intracellular
concentrations 6 times higher than the ones reached treating the cells with the free drug.93
Introduction
29
More recently the same group prepared another Pt‐based targeted prodrug using SWCNTs as a
longboat.94 The derivative was prepared binding folic acid, whose receptors are overexpressed in
many cancer cells, to a Pt(IV) compound, and tethering this conjugate to CNTs through an amidic
coupling with terminal amines of wrapped phospholipid‐PEG chains. The Pt(IV), once inside cells,
can be reduced thanks to acidic endosomal pH, therefore losing the two axial ligands and leading
to the active Pt(II) compound. The authors demonstrated internalization via folate receptor‐
mediated endocytosis (FRME), and the to kill cancer cells (human choriocarcinoma and human
nasopharyngeal carcinoma) with IC50 values more than 8 times lower than cisplatin alone.
Moreover the formation of the major reaction product of cisplatin with DNA was detected, using
a specific monoclonal antibody, proving the actual ability of the system to act as a prodrug,
generating the cytotoxic derivative once internalized and thus killing in a selective way folate
receptor overexpressing cells. Also cisplatin‐delivery systems based on drug‐SWCNT covalent
bioconjugates were investigated.65 The interaction of the epidermal growth factor (EGF) with its
receptor (EGFR), overexpressed on the cell surface of a big variety of cancers, was exploited to
target CNTs both in vitro and in vivo. The efficient in vitro internalization in head and neck
squamous carcinoma cells of the SWCNTs‐EGF conjugate was proved and resulted to be
mediated by the EGF‐EGFR ligand‐receptor interaction. Furthermore quantum dot (Qdot)‐
functionalized SWCNTs were administrated to a tumour‐bearing athymic mice to study the
short‐term biodistribution. The results showed a much higher accumulation within the tumour
mass of the EGF‐conjugate compared to the control without EGF. Small amounts of CNTs were
found within spleen, lungs, liver, kidneys and heart, regardless of the presence of EGF. Finally,
the animals were treated with SWCNT‐cisplatin‐EGF showing a decrease in the tumour growth in
comparison with the untargeted SWCNTs‐cisplatin conjugate, demonstrating that SWCNTs can
selectively deliver cisplatin in vivo towards EGFR overexpressing cancer.
Another interesting approach for the functionalization of CNTs, exploited also in
oncology studies, is the filling of the nanotubes. This method was first investigated by Green’s
group in 1994 and involves the opening of MWCNTs’ caps by a nitric acid treatment and the
filling of the inner cavity through a wet chemistry approach.95 It was applied to fill the tubes with
different materials, among which carboplatin, a widely used chemotherapeutic agent for cancer
treatment.96 MWCNTs were opened with nitric acid and subsequently filled sonicating them in a
carboplatin solution at different temperatures. Finally, they were cleaned from particles
deposited on the outer surface, washing them in water and in ethanol. The derivatives were
characterized trough different techniques to confirm the presence of the drug inside the tubes,
and it was shown that the best temperature for the filling protocol was 90°C, leading to a drug
Chapter 1
30
loading of 30 % in weight. This derivative was then tested on human bladder cancer cells to
evaluate its cytotoxicity and the test showed reduced cell viability compared to the drug alone.
More recently also cisplatin was encapsulated into CNTs, in this case single‐walled ones, the
diameter of which is big enough to host the drug.97 The SWCNTs were opened by concentrated
acid treatment, and subsequently annealed at 850°C under high vacuum in order to remove
oxygen functional groups introduced in the first step. The filling with cisplatin was achieved by
stirring at 40°C the SWCNTs in a DMF solution of the drug for 48 h. The release of the drug in
physiological solution was studied, finding that it started after 24 h and continued up to 72 h,
and the inhibition of cell growth on two prostate cell lines (DU145 and PC3) obtained with the
construct was comparable to the one caused by the free drug.
Another antineoplastic drug, namely 10‐hydroxycamptothecin (HCPT), was covalently
attached to MWCNTs to study the influence of CNTs on drug efficacy.98 MWCNTs were first
oxidized with a 36 h acid treatment, and the carboxylic groups thus introduced underwent an
amidation step to introduce a spacer bearing a terminal protected amine. After deprotection of
the amino group, an ester derivative of HCPT was bound to the construct via an amide bond. The
ester linkage, hydrolytically unstable, was introduced to subsequently trigger the release of the
drug. The authors observed a small drug release (< 15% after 128 h) under both acidic or basic
conditions, and a much higher release (80% after 128 h), when incubating the construct with
fetal bovine serum, which contains esterases that could catalyse the hydrolysis. Furthermore a
fluorescein molecule was bound to the unreacted amine after HCPT conjugation, for in vitro
tests on human gastric carcinoma cells. Fluorescence confocal microscopy permitted to localize
intracellularly the constructs, with a uniform distribution in the cytoplasm. WST‐1 assay revealed
a significantly improved cytotoxicity with respect to equal concentrations of HCPT, whereas
CNTs alone did not cause any decrease in cell viability. Finally, a different construct bearing DTPA
was prepared, in order to chelate technetium (Tc), a radioactive nuclide, for in vivo
biodistribution studies in tumour‐bearing mice. High uptake was found in the liver, spleen, lung,
kidney, stomach, femur and tumour. In the latter the maximum uptake level (3.6 % ID/g) was
reached within 4 h post injection. The observed blood circulation half‐life was of 3.6 h, versus
the 30 min reported for HCPT. Since the in vitro drug release tests showed that 4 h after
incubation with serum only 12% of the HCPT was released, the authors considered that the
majority of the conjugate could reach the different organs (and the tumour) without releasing a
big amount of drug in the blood stream. The in vivo antitumor performance of the HCPT‐bearing
MWCNTs was also studied, finding a tumour growth inhibition much more efficient than HCPT
alone, without causing any severe toxicity to the animals.
Introduction
31
Yang et al. delivered gemcitabine‐MWCNTs construct to lymphatic vessels, exploiting
EPR effect and the specific guidance of an external magnetic field.99 SWCNTs were functionalized
with poly(acrylic acid) and then decorated with magnetite nanoparticles (FeO∙Fe2O3) by a co‐
precipitation step with Fe2+ and Fe3+. Three hours after subcutaneous injection in SD rats, the
construct was able to reach popliteal lymph nodes without accumulation in the major organs as
liver, spleen, kidney, heart and lung, simply by EPR effect. The same CNTs were loaded with the
antineoplastic drug gemcitabine by physical adsorption. The system was guided in vivo applying
a permanent magnet on the projection surface of one popliteal lymph node and very high
accumulation of gemcitabine was detected in the lymph node, with a maximum after 24 h. At
the same time, the blood plasma concentration was lower if compared to gemcitabine alone, to
the treatment without the magnetic field, or to a control represented by nanosized activated
carbon decorated with magnetic nanoparticles.
CNTs‐based antitumor immunotherapy was explored by Meng et al.100 This technique
employs tumour cell vaccines (TCV), which are made of inactivated cancer cells or dendritic cells
presenting tumour antigens, in order to trigger the immune response of the patient against the
tumour itself. With the aim of improving the efficacy of TCV, tumour lysate proteins were
covalently coupled to oxidized MWCNTs via an amidic bond. The conjugate was injected
subcutaneously to H22 hepatoma‐bearing mice treated with TCV (controls were performed with
TCV + CNTs only and TCV + tumour lysate proteins only). The cure rate was significantly
increased with respect to animals treated only with TCV or with TCV + tumour lysate proteins,
even if a partial effect was exerted by CNTs alone as well. To assess whether or not the immunity
was tumour‐specific, the animals survived after treatment with tumour lysate proteins‐bearing
MWCNTs were challenged again with subcutaneous injection of tumour cells. In the case of H22
cells they rejected the tumour, while they did not reject mouse breast cancer, proving how the
therapeutic system made them develop a specific immunity.
C
3
T
Chapter 1
32
Table 1
Commpound
CNT type
MWCNTs D
SWCNTs D
SWCNTs D
SWCNTs D
Drug cov
Doxorubicin
Doxorubicin
Doxorubicin
Doxorubicin
Bond drug‐CNT
valent non‐
covalent
Specific release
mechanism (if any)
‐
acidic tumour or lysosomal
pH
R
acidic tumour or lysosomal
pH
acidic tumour or lysosomal
pH
Specific targeting (if any) I
‐ bre
RDG peptide glioca
‐
‐ cca
Biological studie
In vitro In
human ast cancer cells
human oblastoma ncer cells
‐ lympbeari
human cervical ncer cells
es
Ref. vivo
‐ 82
‐ 83
phoma‐ng mice
84
‐ 86
SWCNTs D
SWCNTs D
SWCNTs D
MWCNTs M
Doxorubicin
Doxorubicin
Doxorubicin
Methotrexate
acidic tumour or lysosomal
pH
c
enzymatic cleavage of carbamate
NIR radiation
g
‐
anti‐carcinoembr
yonic antigen antibody
humca
‐ m
anti‐P‐glycoprotein antibody
le
‐ h
lym
man colon ancer cell line
murine elanoma cell
melabeari
human eukemia cells
human T mphocytes
Introductio
3
‐ 87
anoma‐ng mice
88
‐ 89
‐ 90
on
33
C
3
Chapter 1
34
MWCNTs M
SWCNTs
SWCNTs
Methotrexate
Paclitaxel
Taxoid
enzymatic cleavage of peptidic linker
enzymatic cleavage of ester (by
carboxylesterases)
reduction of the
disulphide bond (by
endogenous thiols)
‐ bre
‐ bre
biotin le
human ast cancer cells
murine ast cancer cell
mubrca
beari
murine eukemia cells
‐ 91
urine reast ncer‐ng mice
92
‐ 66
SWCNTs
SWCNTs
SWCNTs
MWCNTs C
Cisplatin
Cisplatin
Cisplatin
Carboplatin
reduction from Pt(IV) to Pt(II) due
to endosomal acidic pH
reduction from Pt(IV) to Pt(II) due
to endosomal acidic pH
ligand exchange
(hypothesized)
‐
‐ teca
folic acid
cho
nas
caandca
epidermal growth factor
humasqca
‐ bca
human esticular arcinoma cells
human oriocarcino
ma, sopharyngeal
arcinoma d testicular arcinoma cells
man head nd neck quamous arcinoma cells
humaandsquacarcbeari
human bladder ncer cells
Introductio
3
‐ 93
‐ 94
an head d neck amous inoma‐ng mice
65
‐ 96
on
35
C
3
Chapter 1
36
SWCNTs
MWCNTs
SWCNTs G
MWCNTs Tp
Cisplatin
HCPT
Gemcitabine
Tumor lysate proteins (for
TCV)
‐
enzymatic cleavage of ester (by esterases)
‐
n
‐
‐ pro
‐ ca
magnetite nanoparticles guided with an external magnetic field
‐
human ostate cell
human gastric arcinoma cells
hetu
beari
‐ SprDaw
‐ hepabeari
‐ 97
epatic mor‐ng mice
98
ague–ley rats
99
atoma‐ng mice
100
Introduction
37
1.5.3 Delivery of nucleic acids
The way CNTs can interact with nucleic acids has been extensively studied because it
could have very interesting applications. Both antisense oligonucleotides and small interfering
RNA (siRNA) are very promising techniques for gene silencing, applicable for the treatment of
many diseases. In fact, they can inhibit protein expression, potentially blocking many cellular
pathways. Cancer therapy is one of the possible applications, when the targets are oncogenes or
genes involved for instance in angiogenesis or chemotherapy resistance.
One of the first studies in this field was performed by Zhang et al.101 SWCNTs, covalently
functionalized via amidation and bearing a terminal ammonium group, were used to complex a
siRNA able to silence the expression of telomerase reverse transcriptase and thus to inhibit cell
growth. This activity was proved in vitro, on different cell lines, both murine and human, and in
vivo, after intra‐tumour injection in xenografted mice. The in vivo antitumor activity of a CNT‐
based siRNA delivery system was also assessed by Podesta et al, using a different kind of
construct. MWCNTs, covalently functionalized via 1,3‐dipolar cycloaddition and bearing terminal
amino groups, were used to load the proprietary siRNA sequence siTOX.102 The system was
injected within the tumour mass on human lung carcinoma (Calu 6) xenografted mice and
compared with MWCNTs with a non‐coding sequence (siNEG), functionalized MWCNTs alone,
and both siTOX and siNEG delivered with cationic liposomes. MWCNTs‐siTOX significantly
inhibited tumour growth in comparisons to the controls, while cationic liposomes based systems
did not affect tumour growth nor increase animal survival. Moreover, tumour tissue analyses
revealed extended necrosis in the regions where CNTs accumulated. In the context of targeted
delivery, Yang et al. reported the preparation of a folate targeted DNA transporter with CNTs, in
principle exploitable to deliver siRNA to cancer cells as well.103 In fact also in this case the
covalently functionalized SWCNTs presented positive charges to form electrostatic interactions
with nucleic acids. They bound fluorescently labelled‐dsDNA to this derivative, proving by UV a
strong enhancement in the loading of ds‐DNA for the positively charged SWCNTs in comparison
to non‐charged SWCNTs. They further functionalized the derivative by wrapping a folic acid
modified phospholipid. The complex was tested on mouse ovarian epithelial cells, showing an
increased uptake for the derivative with the folic acid, compared to the one without the
targeting unit. Furthermore, the fluorescently labelled dsDNA alone was internalized only at a
very poor level, thus demonstrating the carrier role of CNTs. In a final experiment, HeLa cells
were induced to overexpress folate receptor, culturing them with folic acid free medium, and
these cells showed a much higher internalization of the derivative than normal HeLa cells,
Chapter 1
38
confirming the efficacy of the delivery system. In another study, oxidized MWCNTs were
complexed to polyethylenimine, to which an antisense sequence was bound by means of
electrostatic interactions.67a The oligonucleotide was coupled to fluorescent cadmium telluride
quantum dots in order to follow the cellular trafficking of the complex. The authors
demonstrated for the system an efficient uptake and the expected apoptotic activity.
Dendrimer‐CNT constructs were also exploited for gene delivery. The anti‐survivin
oligonucleotide was anchored onto the polyamidoamine (PAMAM) dendrimer covalently
attached to the CNTs and transfected into MCF‐7 cells.104 The conjugate distributed mainly in the
cytoplasm, endosomes and lysosomes of the cells and was able to release the antisense
oligonucleotide, which then exerted its apoptotic activity. The PAMAM dendrimer was also
grown directly on MWCNT surface and used as anchoring point for siRNA, after introduction of a
trimethylammonium unit on each branch termination. Different dendrimer generations were
prepared and their efficacy in delivering a fluorescent oligonucleotide was compared, obtaining
the best results for the second generation.105
A unique approach conjugated the use of CNTs as nucleic acids delivery system with
photodynamic therapy.106 This therapy represents a very interesting option for cancer
treatment, and it is based on the delivery of a photosensitizer, which, upon activation by an
appropriate light source, transfers the light energy to tissue oxygen thus generating singlet
oxygen, which can react rapidly with cellular molecules and trigger a cellular damage. In this
study an aptamer, a synthetic DNA/RNA probe able to recognize and bind a specific target, was
covalently bound to chlorine e6 (Ce6), a well‐known photosensitizer. Then the aptamer was
wrapped on SWCNTs, and they were able to quench 98% of the singlet oxygen generation (SOG)
normally occurring upon excitation of Ce6. When the aptamer was involved in the binding to its
target, in this case human α‐thrombin, it was released from the CNTs, and therefore SOG was
not quenched anymore. Phototoxicity on human lymphoma cells treated with the construct was
studied, showing a reduction in cell viability comparable to treatment with Ce6 alone, when
thrombin was added, indicating the feasibility of a targeted SWCNTs‐based system for
photodynamic therapy.
All the examples reported demonstrate that CNTs, both single‐ and multi‐walled, are
good vectors for nucleic acid delivery and that the transported sequences retain their activity.
These results are even more attractive considering that CNTs exert also an important activity
against enzymatic digestion.107 In fact oligonucleotides, when wrapped on SWCNTs, are
protected from enzymatic cleavage and interference from nucleic acid binding proteins,
increasing their stability in cells.
Introduction
39
Furthermore, Li et al. reported that oxidized SWCNTs induced a stabilization of the
human telomeric i‐motif of DNA. Even though the question still needs to be studied, this very
particular possibility could be exploited in anti‐neoplastic systems, since the stabilization of
telomeric quadruples inhibits telomerase, which is an essential enzyme for the proliferation of
cancer cells.108
1.6 Alternative CNT‐based anticancer strategies
Cancer treatments can be exerted not only by the administration of drugs, and CNT‐
based systems have been explored also with different approaches.
Some of these works focus mainly on the targeting issue, which is of course of great
importance in cancer therapy, without studying a specific therapeutic possibility. A specific
recognition of membrane receptor was achieved with Ab‐functionalized SWCNTs.109 They were
functionalized with a phospholipid‐PEG, in order to increase solubility and to prevent non‐
specific binding in the biological environment, and subsequently the terminal amines of the PEG
were covalently functionalized with two different antibodies (Rituxan and Herceptin). The
binding to corresponding membrane receptors (CD20 on B‐cells, and HER2/neu on breast cancer
cells, respectively) was proved in vitro exploiting the intrinsic near‐infrared fluorescence of CNTs,
thus demonstrating the feasibility of using antibodies as targeting agent in applications such as
imaging or drug delivery. Raman spectroscopy was also used as an imaging technique for
CNTs.110 Also in this case a phospholipid‐PEG‐CNT construct was used. Derivatives bound to
different antibodies or to RGD peptide were prepared using SWCNTs with different isotope
composition, that display well‐shifted Raman G‐band peaks. In this way, it was possible to
distinguish the different CNT constructs in mixed cellular populations, proving that they were
able to selectively recognize their specific target. Raman imaging was used also for both in vitro
and direct in vivo studies by Zavaleta et al., with PEGylated SWCNTs functionalized with RGD
peptide (same construct of ref. 85).111 The in vitro data showed that αVβ3 integrin‐positive
tumour cells (U87MG) internalized the construct much better than SWCNTs without RGD and
more than αVβ3 integrin‐negative cells (HT29). Direct in vivo data showed selective prolonged
accumulation of the RGD‐bearing SWCNTs in the tumour, while the non‐targeted SWCNTs gave
an initial tumour accumulation, followed by a rapid decrease. With an alternative approach, an
antibody was bound to SWCNTs exploiting the high affinity interaction between protein A and
the Fc portion of the antibody itself.112 SWCNTs wrapped with phospholipid‐PEG‐COOH were
bound to protein A by amidation. Then, an anti‐integrin antibody, previously marked with
fluorescein, was linked to the system. This antibody recognizes a specific integrin overexpressed
Chapter 1
40
on various cancer cells, rendering the construct suitable for cancer targeting. After
demonstrating that the construct was not cytotoxic, its internalization was studied via
fluorescence confocal microscopy both in integrin‐positive (U87MG human glioblastoma cancer
cells) and integrin‐negative (MCF‐7 human breast cancer cells) cell lines, proving the specific
targeting. In fact, only U87MG cells internalized the SWCNT construct, while no intracellular
fluorescence was detected for MCF‐7 cells or for U87MG cells after a pre‐treatment with the Ab
alone. The authors therefore hypothesized an integrin‐mediated endocytosis as uptake
mechanism. Bottini et al. studied the internalization of SWCNTs‐Quantum Dot (QD)‐streptavidin
complexes by CD3 positive‐leukaemia cells, monitoring the intracellular QD fluorescence.64 In
this case the process was rather complex and the uptake was mediated by a multi‐component
recognition. First, a biotinylated anti CD3‐antibody bound the CD3 receptor on cell membrane,
then the streptavidin‐bearing SWCNTs attached the biotinylated antibody, showing high
internalization. On the contrary, poor uptake was observed when: i) the antibody was absent, ii)
a non‐biotinylated antibody was used, iii) the experiments were carried out at 4°C and iv) the
cells did not express CD3‐receptor on the membrane (CD3 negative cell line). All these data
suggest a receptor‐mediated endocytosis for this construct.
1.6.1 Thermal ablation
A very interesting application of carbon nanotubes in cancer therapy could arise from
their intrinsic optical properties, which can be exploited to kill cancer cells by photo‐thermal
destruction. In fact the optical absorbance of this material is very high in the NIR (near‐infrared)
region, 700‐1100 nm, while biological systems are transparent to these wavelength lights.
It is worth notice that, to our best knowledge, gold nanoshells and gold nanoparticles
represent the only other materials which have given good results in photothermal cancer
treatment with NIR radiation.113 In this case the laser intensity and the radiation time used are
often higher than the ones needed to kill the cells with CNTs. This observation make CNTs an
even more promising candidate in the field, opening the way to further exploration, to reach the
laser energy level of 35‐45 mJ/cm2, established as the safety standard for medical lasers.114
Among the studies carried out, it is important to distinguish the non‐targeted ones, both
in vitro,115 and in vivo, when CNTs were directly injected in the tumour,116 and the ones where
CNTs were specifically targeted to tumour cells.
The first paper reporting this possibility was published in 2005 by Dai and co‐workers.59a
SWCNTs were non‐covalently functionalized with phospholipid‐PEG chains bearing a fluorescent
tag or a folic acid molecule. The complex was then administered to HeLa cells overexpressing the
folate receptor (FR+ cells) and to normal HeLa cells as a control. The FR+ cells showed a high
Introduction
41
internalization of the folic acid‐SWCNTs derivative, imaged by fluorescence microscopy, while
the normal cells showed poor uptake. Cells were then radiated by an 808 nm laser (1.4 W/cm2)
for 2 minutes, this treatment resulting in extensive FR+ cell death and in normal proliferation
behaviour for cells that did not internalize the carbon nanotubes. A multi‐component targeting
system was created, binding to SWCNTs two different monoclonal antibodies, specific for breast
cancer cells antigens (IGF1R and HER2).117 The double targeting should ensure high efficacy and
selectivity. The derivatives were prepared functionalizing CNTs with 1‐pyrenebutanoyl
succinimmide, able to stick to the aromatic surface of the nanotubes by π‐π interaction, and
bearing an anchoring group for linking the antibodies. Cells were incubated with different
conjugates and they were then excited by 808 nm photons at 800 mW/cm2 for 3 minutes. As a
consequence, the nanotubes were heated and the increased temperature destroyed the cells
that internalized them. After this treatment, all the cells incubated with the IGF1R and HER2
antibody/SWCNTs hybrids were destroyed, while the 80% of the cells incubated with non‐
specific antibody/SWCNTs hybrids survived, indicating the specific internalization of the
nanotubes driven by the presence of specific receptors on cancer cells. In another work, non‐
covalent SWCNTs‐Ab derivatives were prepared exploiting the strong binding between
neutravidin attached to the Ab and biotin present on a polymer coating of SWCNTs.118 They were
targeted to human cells presenting the specific antigen for the Ab and subsequent treatment
with a NIR laser (808 nm, 5000 mW/cm2 for 7 min) resulted in a significant decrease in cells
viability. Since the disadvantage of non‐covalent constructs is, as already mentioned, the
possible dissociation in biological fluids, the same authors prepared a covalent SWCNTs‐Ab
construct by forming amidic bonds between carboxylic groups of oxidized SWCNTs and amines
of two different Ab (anti‐CD22 or anti‐CD25).119 To assess the specific binding to human Burkitt’s
lymphoma cells (CD22+CD25‐) and phytohemagglutinin‐activated normal human peripheral
blood mononuclear cells (CD22‐CD25+), the cells were treated with the constructs and with
fluorescein‐conjugated goat anti‐mouse immunoglobulin (FITC‐GAMIg) and analysed through
flow cytometry. The results clearly showed specific recognition of the correspondent cell
receptor by the two Ab‐SWCNTs conjugates. Finally the thermal ablation of the specifically
targeted cells was achieved after exposure to NIR radiation (808 nm laser, 9.5 W/cm2, for 4 min).
Among cancer thermal ablation studies involving CNTs, an interesting option was
recently described by Peng and co‐workers.120 Targeting cancer could be translated more
specifically into targeting cancer stem cells. These cells are responsible for formation of
metastases, resistance to therapies and restoration of tumours, and therefore they represent a
very important target to study and more efforts should be made in this direction. The chosen
Chapter 1
42
targets were CD133+ glioblastoma cells, presenting high tumorigenicity and cancer stem‐like
properties. SWCNTs were wrapped with chitosan, which was then covalently coupled to CD133
monoclonal antibody. These CNTs were tested in vitro with both CD133+ and CD133‐
glioblastoma cells, and internalization was observed only by the CD133+ cells, which were
subsequently killed upon exposure to NIR laser radiation (808 nm, 2 W/cm2, for 5 min).
Furthermore the treatment with CNTs followed by NIR radiation inhibited spheroid body
formation, which represents an index of cells self‐renewal. To partially translate the study in
vivo, CNTs‐treated or untreated cells were used to induce tumour formation in nude mice. Two
days after the subcutaneous injection the mice were subjected to NIR laser radiation, resulting in
significant inhibition of tumour growth when CNTs were used.
All these reports are a further confirmation of the ability of carbon nanotubes to serve as
a targeted nanovector for cancer, exploiting in this case a NIR‐induced thermal ablation.
1.6.2 Radiotherapy & BNCT
As already mentioned, the main purpose of attaching radioisotopes to CNTs is
represented by imaging studies, to determine their biodistribution following the radioisotope
traces, but also a therapeutic application can be envisaged. McDevitt et al. reported the
preparation of a SWCNT derivative bearing an 111In (Indium 111) chelate and they have studied
its biodistribution, finding accumulation mainly in kidney, spleen and liver.121 The same
derivative was then further functionalized with a tumour specific monoclonal antibody, and it
was delivered in vivo in a murine model of disseminated human lymphoma, showing a selective
tumour targeting. Therefore, the replacement of indium with a proper radionuclide would
render these constructs ideal for their application in radiotherapy. In fact, the preparation of
covalent CNTs‐Ab constructs intended to target tumour neo‐vasculature for radiotherapy was
reported recently by same authors.122 Angiogenic endothelial cells express on their surface a
monomeric cadherin that, forming dimers with cadherins on close cells, constitutes the tight
junctions of normal vascular endothelium. In tumours, these cells are poorly connected (that is
one of the reasons for EPR effect) and the cadherin are in the monomeric form, which is the only
one recognized by antibody E4G10. Doubly functionalized SWCNTs were prepared, binding to
the CNTs the specific antibody and the alpha particle‐emitting 225Ac (actinium 225) radionuclide
generator, able to kill the targeted cell and the tissue in its proximity. The compound was tested
in vivo via intravenous injection on xenografted tumour‐mice: it reduced tumour growth,
improving mice survival in comparison to controls bearing different antibody. The success of this
approach is very promising because, in principle, the targeting of tumour vasculature renders
these kinds of constructs suitable for any kind of solid tumour.
Introduction
43
The interesting possibility of employing CNTs in boron neutron capture therapy (BNCT)
was explored.18a BNCT is based on the reaction between 10B, after it reached the target cells, and
a neutron beam, that generates 11B, which subsequently ejects an energetic short‐range alpha
particle and a lithium ion. These species deposit most of their energy within the cell, seriously
damaging DNA. In order to be effective in cancer therapy, the boron needs to reach tumour cells
with an adequate concentration and CNTs could be the right candidate, considering their
capacity to cross biological membranes. Carborane were derivatized with azides and allowed to
react with the double bonds of SWCNTs, leading to the formation of aziridine rings,
subsequently opened to give a water‐soluble ethoxo‐derivative suitable for in vivo tests. The
derivatives were used in tissue distribution experiments on mice transplanted with EMT6
mammary cancer cells, achieving a maximum boron concentration in tumour tissue (27.9 μg
boron/g tissue) 16 hours after administration. The tumour‐to‐blood boron ratio was favourable,
with a much higher concentration in tumour cells than in the bloodstream, probably due to EPR
effect. These results permit to seriously consider boron‐CNT derivatives suitable for BNCT, even
though, to the best of our knowledge, no further toxicity studies on these derivatives have been
reported so far.
1.7 References
1. X. Lu and Z. Chen, Chem. Rev., 2005, 105, 3643‐3696.
2. J.‐C. Charlier, Acc. Chem. Res., 2002, 35, 1063‐1069.
3. a) A. Govindaraj and C. N. R. Rao, Chapter 2 in Carbon Nanotechnology, edited by Liming
Dai, Elsevier B.V., 2006, 15‐51; b) Q. Fu and J. Liu, Chapter 4 in Carbon Nanotechnology,
edited by Liming Dai, Elsevier B.V., 2006, 81‐107.
4. P. Nikolaev, M. J. Bronikowski, R. K. Bradley, F. Rohmund, D. T. Colbert, K. A. Smith, and
R. E. Smalley, Chem. Phys. Lett., 1999, 313, 91‐97.
5. a) C. Ehli, G. M. A. Rahman, N. Jux, D. Balbinot, D. M. Guldi, F. Paolucci, M. Marcaccio, D.
Paolucci, M. Melle‐Franco, F. Zerbetto, S. Campidelli, and M. Prato, J. Am. Chem. Soc.,
2006, 128, 11222‐11231; b) R. J. Chen, Y. Zhang, D. Wang, and H. Dai, J. Am. Chem. Soc.,
2001, 123, 3838‐3839.
6. M. Zheng, A. Jagota, E. D. Semke, B. A. Diner, R. S. McLean, S. R. Lustig, R. E. Richardson,
and N. G. Tassi, Nat. Mater., 2003, 2, 338‐342.
7. a) R. Qiao and P. C. Ke, J. Am. Chem. Soc., 2006, 128, 13656‐13657; b) C. Richard, F.
Balavoine, P. Schultz, T. W. Ebbesen, and C. Mioskowski, Science, 2003, 300, 775‐778.
Chapter 1
44
8. a) X. Li, W. Chen, Q. Zhan, L. Dai, L. Sowards, M. Pender, and R. R. Naik, J. Phys. Chem. B,
2006, 110, 12621‐12625; b) V. Zorbas, A. L. Smith, H. Xie, A. Ortiz‐Acevedo, A. B. Dalton,
G. R. Dieckmann, R. K. Draper, R. H. Baughman, and I. H. Musselman, J. Am. Chem. Soc.,
2005, 127, 12323‐12328.
9. A. Star, J. F. Stoddart, D. Steuerman, M. Diehl, A. Boukai, E. W. Wong, X. Yang, S.‐W.
Chung, H. Choi, and J. R. Heath, Angew. Chem. Int. Ed., 2001, 40, 1721‐1725.
10. a) E. Dujardin, T. W. Ebbesen, A. Krishnan, and M. M. J. Treacy, Adv. Mater., 1998, 10,
1472‐1475; b) B. W. Smith, M. Monthioux, and D. E. Luzzi, Chem. Phys. Lett., 1999, 315,
31‐36; c) M. Wilson and P. A. Madden, J. Am. Chem. Soc., 2001, 123, 2101‐2102.
11. S. Niyogi, M. A. Hamon, H. Hu, B. Zhao, P. Bhowmik, R. Sen, M. E. Itkis, and R. C. Haddon,
Acc. Chem. Res., 2002, 35, 1105‐1113.
12. E. T. Mickelson, C. B. Huffman, A. G. Rinzler, R. E. Smalley, R. H. Hauge, and J. L.
Margrave, Chem. Phys. Lett., 1998, 296, 188‐194.
13. a) R. K. Saini, I. W. Chiang, H. Peng, R. E. Smalley, W. E. Billups, R. H. Hauge, and J. L.
Margrave, J. Am. Chem. Soc., 2003, 125, 3617‐3621; b) P. Boul, Chem. Phys. Lett., 1999,
310, 367‐372.
14. Y. Ying, R. K. Saini, F. Liang, A. K. Sadana, and W. E. Billups, Org. Lett., 2003, 5, 1471‐
1473.
15. a) J. L. Bahr, J. Yang, D. V. Kosynkin, M. J. Bronikowski, R. E. Smalley, and J. M. Tour, J.
Am. Chem. Soc., 2001, 123, 6536‐6542; b) J. L. Hudson, M. J. Casavant, and J. M. Tour, J.
Am. Chem. Soc., 2004, 126, 11158‐11159.
16. a) M. Holzinger, O. Vostrowsky, A. Hirsch, F. Hennrich, M. Kappes, R. Weiss, and F. Jellen,
Angew. Chem. Int. Ed., 2001, 40, 4002‐4005; b) R. Graupner, J. Abraham, D. Wunderlich,
A. Vencelová, P. Lauffer, J. Röhrl, M. Hundhausen, L. Ley, and A. Hirsch, J. Am. Chem.
Soc., 2006, 128, 6683‐9; c) Z. Syrgiannis, F. Hauke, J. Röhrl, M. Hundhausen, R. Graupner,
Y. Elemes, and A. Hirsch, Eur. J. Org. Chem., 2008, 2008, 2544‐2550.
17. J.‐P. Tessonnier, A. Villa, O. Majoulet, D. S. Su, and R. Schlögl, Angew. Chem. Int. Ed.,
2009, 48, 6543‐6546.
18. a) Z. Yinghuai, A. T. Peng, K. Carpenter, J. A. Maguire, N. S. Hosmane, and M. Takagaki, J.
Am. Chem. Soc., 2005, 127, 9875‐98801; b) C. Ménard‐Moyon, F. Dumas, E. Doris, and C.
Mioskowski, J. Am. Chem. Soc., 2006, 128, 14764‐14765; c) D. Ogrin, J. Chattopadhyay, A.
K. Sadana, W. E. Billups, and A. R. Barron, J. Am. Chem. Soc., 2006, 128, 11322‐11323.
19. a) J. Liu , A. G. Rinzler , H. J. Dai , J. H. Hafner , R. K. Bradley , P. J. Boul , A. Lu , T. Iverson ,
K. Shelimov , C. B. Huffman , F. Rodriguez‐Macias , Y. S. Shon , T. R. Lee , D. T. Colbert , R.
Introduction
45
E. Smalley , Science 1998 , 280 , 1253; b) K. J. Ziegler, Z. Gu, H. Peng, E. L. Flor, R. H.
Hauge, and R. E. Smalley, J. Am. Chem. Soc., 2005, 127, 1541‐1547.
20. C. Warakulwit, J. Majimel, M.‐H. Delville, P. Garrigue, J. Limtrakul, and A. Kuhn, J. Mater.
Chem., 2008, 18, 4056‐4061.
21. T. J. Aitchison, M. Ginic‐Markovic, J. G. Matisons, G. P. Simon, and P. M. Fredericks, J.
Phys. Chem. C, 2007, 111, 2440‐2446.
22. a) I. W. Chiang, B. E. Brinson, A. Y. Huang, P. A. Willis, M. J. Bronikowski, J. L. Margrave,
R. E. Smalley, and R. H. Hauge, J. Phys. Chem. B, 2001, 105, 8297‐8301; b) K. L. Strong, D.
P. Anderson, K. Lafdi, and J. N. Kuhn, Carbon, 2003, 41, 1477‐1488.
23. J. Zhang, H. Zou, Q. Qing, Y. Yang, Q. Li, Z. Liu, X. Guo, and Z. Du, J. Phys. Chem. B, 2003,
107, 3712‐3718.
24. W. Zhou, Y. H. Ooi, R. Russo, P. Papanek, D. E. Luzzi, J. E. Fischer, M. J. Bronikowski, P. a
Willis, and R. E. Smalley, Chem. Phys. Lett., 2001, 350, 6‐14.
25. a) J. Chen, M. A. Hamon, H. Hu, Y. Chen, A. M. Rao, P. C. Eklund, and R. C. Haddon,
Science, 1998, 282, 95‐98; b) M. A. Hamon, J. Chen, H. Hu, Y. Chen, M. E. Itkis, A. M. Rao,
P. C. Eklund, and R. C. Haddon, Adv. Mater., 1999, 11, 834‐840; c) E. Del Canto, K. Flavin,
D. Movia, C. Navio, C. Bittencourt, and S. Giordani, Chem. Mater., 2011, 23, 67‐74.
26. M. Maggini, G. Scorrano, and M. Prato, J. Am. Chem. Soc., 1993, 115, 9798‐9799.
27. V. Georgakilas, K. Kordatos, M. Prato, D. M. Guldi, M. Holzinger, and A. Hirsch, J. Am.
Chem. Soc., 2002, 124, 760‐761.
28. E. Cho, S. Shin, and Y.‐G. Yoon, J. Phys. Chem. C, 2008, 112, 11667‐11672.
29. X. Lu, F. Tian, X. Xu, N. Wang, and Q. Zhang, J. Am. Chem. Soc., 2003, 125, 10459‐10464.
30. a) F. A. Carey and R. J. Sundberg, Chapter 10 in Advanced Organic Chemistry, 5th ed.
2007, ISBN: 978‐0‐387‐68346‐1; b) D. H. Ess and K. N. Houk, J. Am. Chem. Soc., 2008,
130, 10187‐10198.
31. H. Kataura, Y. Kumazawa, Y. Maniwa, I. Umezu, S. Suzuki, Y. Ohtsuka, and Y. Achiba,
Synth. Met., 1999, 103, 2555‐2558.
32. A. Jorio, A. G. Souza Filho, G. Dresselhaus, M. S. Dresselhaus, A. K. Swan, M. S. Unlu, B. B.
Goldberg, M. A. Pimenta, J. H. Hafner, C. M. Lieber, and R. Saito, Phys. Rev. B, 2002, 65,
155412(9).
33. A. C. Dillon, M. Yudasaka, and M. S. Dresselhaus, J. Nanosci. Nanotechnol., 2004, 4, 691‐
703.
34. S. Arepalli, P. Nikolaev, O. Gorelik, V. G. Hadjiev, W. Holmes, B. Files, and L. Yowell,
Carbon, 2004, 42, 1783‐1791.
Chapter 1
46
35. MultiMode SPM Instruction Manual, Copyright © 1996, 1997, 2004 Veeco Instruments
Inc.
36. SPM Training Notebook Copyright © 2003 Veeco Instruments Inc.
37. A. Takagi, A. Hirose, T. Nishimura, N. Fukumori, A. Ogata, N. Ohashi, S. Kitajima, and J.
Kanno, J. Toxicol. Sci., 2008, 33, 105‐116.
38. C. A. Poland, R. Duffin, I. Kinloch, A. Maynard, W. A. H. Wallace, A. Seaton, V. Stone, S.
Brown, W. MacNee, and K. Donaldson, Nat. Nanotechnol., 2008, 3, 423‐428.
39. L. A. Mitchell, J. Gao, R. V. Wal, A. Gigliotti, S. W. Burchiel, and J. D. McDonald, Toxicol.
Sci., 2007, 100, 203‐214.
40. J. P. Ryman‐Rasmussen, M. F. Cesta, A. R. Brody, J. K. Shipley‐Phillips, J. I. Everitt, E. W.
Tewksbury, O. R. Moss, B. A. Wong, D. E. Dodd, M. E. Andersen, and J. C. Bonner, Nat.
Nanotechnol., 2009, 4, 747‐751.
41. Baron, P.A., Maynard, A.D., Foley, M., NIOSH DART‐02‐191 Rev. 1.1, 2003.
42. D. S.‐W. Tan, M. Gerlinger, B.‐T. Teh, and C. Swanton, Eur. J. Cancer, 2010, 46, 2166‐
2177.
43. S. V. Ambudkar, S. Dey, C. A. Hrycyna, M. Ramachandra, I. Pastan, and M. M. Gottesman,
Ann. Rev. Pharmacol. Toxicol., 1999, 39, 361‐398.
44. Y. Matsumura and H. Maeda, Cancer Res., 1986, 46, 6387‐6392., A. K. Iyer, G. Khaled, J.
Fang, and H. Maeda, Drug Discov. Today, 2006, 11, 812‐818.
45. a) K. T. Oh, H. J. Baik, a H. Lee, Y. T. Oh, Y. S. Youn, and E. S. Lee, Int. J. Mo.l Sci., 2009, 10,
3776‐3792; b) L. S. Jabr‐Milane, L. E. van Vlerken, S. Yadav, and M. M. Amiji, Cancer
Treat. Rev., 2008, 34, 592‐602.
46. H. Dumortier, S. Lacotte, G. Pastorin, R. Marega, W. Wu, D. Bonifazi, J.‐P. Briand, M.
Prato, S. Muller, and A. Bianco, Nano Lett., 2006, 6, 1522‐1528.
47. P. Wick, P. Manser, L. K. Limbach, U. Dettlaff‐Weglikowska, F. Krumeich, S. Roth, W. J.
Stark, and A. Bruinink, Toxicol. Lett., 2007, 168, 121‐131.
48. C. M. Sayes, F. Liang, J. L. Hudson, J. Mendez, W. Guo, J. M. Beach, V. C. Moore, C. D.
Doyle, J. L. West, W. E. Billups, K. D. Ausman, and V. L. Colvin, Toxicol. Lett., 2006, 161,
135‐142.
49. a) R. Singh, D. Pantarotto, L. Lacerda, G. Pastorin, C. Klumpp, M. Prato, A. Bianco, and K.
Kostarelos, Proc. Natl. Acad. Sci. USA, 2006, 103, 3357‐3362; b) L. Lacerda, H. Ali‐
Boucetta, M. A. Herrero, G. Pastorin, A. Bianco, M. Prato, and K. Kostarelos,
Nanomedicine, 2008, 3, 149‐161; c) L. Lacerda, M. A. Herrero, K. Venner, A. Bianco, M.
Prato, and K. Kostarelos, Small, 2008, 4, 1130‐1132; d) M. R. McDevitt, D.
Introduction
47
Chattopadhyay, J. S. Jaggi, R. D. Finn, P. B. Zanzonico, C. Villa, D. Rey, J. Mendenhall, C. A.
Batt, J. T. Njardarson, and D. A. Scheinberg, PloS ONE, 2007, 2, e907; e).Z. Liu, C. Davis,
W. Cai, L. He, X. Chen, and H. Dai, Proc. Natl. Acad. Sci. USA, 2008, 105, 1410‐1415; f) B.
Kang, D. Yu, Y. Dai, S. Chang, D. Chen, and Y. Ding, Carbon, 2009, 47, 1189‐1192; g) A.
Ruggiero, C. H. Villa, E. Bander, D. A. Rey, M. Bergkvist, C. A. Batt, K. Manova‐Todorova,
W. M. Deen, D. A. Scheinberg, and M. R. McDevitt, Proc. Natl. Acad. Sci. USA, 2010, 107,
12369‐12374.
50. a) B. L. Allen, P. D. Kichambare, P. Gou, I. I. Vlasova, A. A. Kapralov, N. Konduru, V. E.
Kagan, and A. Star, Nano Lett., 2008, 8, 3899‐3903; b) B. L. Allen, G. P. Kotchey, Y. Chen,
N. V. K. Yanamala, J. Klein‐Seetharaman, V. E. Kagan, and A. Star, J. Am. Chem. Soc.,
2009, 131, 17194‐205; c) V. E. Kagan, N. V. Konduru, W. Feng, B. L. Allen, J. Conroy, Y.
Volkov, I. I. Vlasova, N. A. Belikova, N. Yanamala, A. Kapralov, Y. Y. Tyurina, J. Shi, E. R.
Kisin, A. R. Murray, J. Franks, D. Stolz, P. Gou, J. Klein‐seetharaman, B. Fadeel, A. Star,
and A. A. Shvedova, Nat. Nanotechnol., 2010, 5, 354‐359.
51. Y. Zhao, B. L. Allen, and A. Star, J. Phys. Chem. A, 2011, doi/10.1021/jp112324d.
52. P. Cherukuri, S. M. Bachilo, S. H. Litovsky, and R. B. Weisman, J. Am. Chem. Soc., 2004,
126, 15638‐15639.
53. A. E. Porter, M. Gass, K. Muller, J. N. Skepper, P. A. Midgley, and M. Welland, Nat.
Nanotechnol., 2007, 2, 713‐717.
54. C. Cheng, K. H. Müller, K. K. K. Koziol, J. N. Skepper, P. A. Midgley, M. E. Welland, and A.
E. Porter, Biomaterials, 2009, 30, 4152‐4160.
55. Y. Sato, A. Yokoyama, K.‐I. Shibata, Y. Akimoto, S.‐I. Ogino, Y. Nodasaka, T. Kohgo, K.
Tamura, T. Akasaka, M. Uo, K. Motomiya, B. Jeyadevan, M. Ishiguro, R. Hatakeyama, F.
Watari, and K. Tohji, Mol. BioSyst., 2005, 1, 176‐182.
56. D. M. Brown, I. A. Kinloch, U. Bangert, A. H. Windle, D. M. Walter, G. S. Walker, C. A.
Scotchford, K. Donaldson, and V. Stone, Carbon, 2007, 45, 1743‐1756.
57. B. Kateb, M. Van Handel, L. Zhang, M. J. Bronikowski, H. Manohara, and B. Badie,
NeuroImage, 2007, 37, S9‐S17.
58. N. W. S. Kam, T. C. Jessop, P. A. Wender, and H. Dai, J. Am. Chem. Soc., 2004, 126, 6850‐
6851.
59. a) N. W. S. Kam, M. O. Connell, J. A. Wisdom, and H. Dai, Proc. Natl. Acad. Sci. USA, 2005,
102, 11600‐11605; b) N. W. S. Kam and H. Dai, J. Am. Chem. Soc., 2005, 127, 6021‐6026.
60. N. W. S. Kam, Z. Liu, and H. Dai, Angew. Chem. Int. Ed., 2006, 45, 577‐581.
61. H. Jin, D. A. Heller, and M. S. Strano, Nano Lett., 2008, 8, 1577‐1585.
Chapter 1
48
62. H. Jin, D. A. Heller, R. Sharma, and M. S. Strano, ACS Nano, 2009, 3, 149‐158.
63. M. Wang, S. Yu, C. Wang, and J. Kong, ACS Nano, 2010, 4, 6483‐6490.
64. M. Bottini, F. Cerignoli, M. I. Dawson, A. Magrini, N. Rosato, and T. Mustelin,
Biomacromolecules, 2006, 7, 2259‐2263.
65. A. A. Bhirde, V. Patel, J. Gavard, G. Zhang, A. A. Sousa, A. Masedunskas, R. D. Leapman,
R. Weigert, J. S. Gutkind, and J. F. Rusling, ACS Nano, 2009, 3, 307‐316.
66. J. Chen, S. Chen, X. Zhao, L. V. Kuznetsova, S. S. Wong, and I. Ojima, J. Am. Chem. Soc.,
2008, 130, 16778‐16785.
67. a) N. Jia, Q. Lian, H. Shen, C. Wang, X. Li, and Z. Yang, Nano Lett., 2007, 7, 2976‐2980; b)
Y. Liu, W. De‐Cheng, W.‐D. Zhang, X. Jiang, C.‐bin He, T. S. Chung, S. H. Goh, and K. W.
Leong, Angew. Chem. Int. Ed., 2005, 44, 4782‐4785.
68. J. Cheng, K. A. S. Fernando, L. M. Veca, Y.‐P. Sun, A. I. Lamond, Y. W. Lam, and S. H.
Cheng, ACS Nano, 2008, 2, 2085‐2094.
69. a) D. Pantarotto, J.‐Paul Briand, M. Prato, and A. Bianco, Chem. Commun., 2004, 16‐17;
b) D. Pantarotto, R. Singh, D. McCarthy, M. Erhardt, J.‐P. Briand, M. Prato, K. Kostarelos,
and A. Bianco, Angew. Chem. Int. Ed., 2004, 43, 5242‐5246.
70. L. A. Kueltzo and C. R. Middaugh, J. Pharm. Sci., 2003, 92, 1754‐1772.
71. C. F. Lopez, S. O. Nielsen, P. B. Moore, and M. L. Klein, Proc. Natl. Acad. Sci. USA, 2004,
101, 4431‐4434.
72. S. Pogodin and V. A. Baulin, ACS Nano, 2010, 4, 5293‐300.
73. S. Pogodin, N. K. H. Slater, and V. A. Baulin, ACS Nano, 2011, 5, 1141–1146.
74. K. Kostarelos, L. Lacerda, G. Pastorin, W. Wu, S. Wieckowski, J. Luangsivilay, S. Godefroy,
D. Pantarotto, J.‐Paul Briand, S. Muller, M. Prato, and A. Bianco, Nat. Nanotechnol.,
2007, 2, 108‐113.
75. R. Zeineldin, M. Al‐Haik, and L. G. Hudson, Nano Lett., 2009, 9, 751‐757.
76. B. Kang, S. Chang, Y. Dai, D. Yu, and D. Chen, Small, 2010, 6, 2362‐2366.
77. A. E. Porter, M. Gass, J. S. Bendall, K. Muller, A. Goode, J. N. Skepper, P. A. Midgley, and
M. Welland, ACS Nano, 2009, 3, 1485‐1492.
78. C. Cheng, A. E. Porter, K. Muller, K. Koziol, J. N. Skepper, P. Midgley, and M. Welland, J.
Phys. Conf., 2009, 151, 1‐15.
79. N. A. Monteiro‐Riviere, R. J. Nemanich, A. O. Inman, Y. Y. Wang, and J. E. Riviere, Toxicol.
Lett., 2005, 155, 377‐384.
80. D. Cai, J. M. Mataraza, Z.‐hong Qin, Z. Huang, J. Huang, T. C. Chiles, D. Carnahan, K.
Kempa, and Z. Ren, Nat. Methods, 2005, 2, 449‐454.
Introduction
49
81. R. Singhal, Z. Orynbayeva, R. V. K. Sundaram, J. J. Niu, S. Bhattacharyya, E. A. Vitol, M. G.
Schrlau, E. S. Papazoglou, G. Friedman, and Y. Gogotsi, Nat. Nanotechnol., 2011, 6, 57‐
64.
82. H. Ali‐Boucetta, K. T. Al‐Jamal, D. McCarthy, M. Prato, A. Bianco, and K. Kostarelos,
Chem. Commun., 2008, 459‐461.
83. Liu, X. Sun, N. Nakayama‐Ratchford, and H. Dai, ACS Nano, 2007, 1, 50‐56.
84. Z. Liu, A. C. Fan, K. Rakhra, S. Sherlock, A. Goodwin, X. Chen, Q. Yang, D. W. Felsher, and
H. Dai, Angew. Chem. Int. Ed., 2009, 48, 7668‐7672.
85. Z. Liu, W. Cai, L. He, N. Nakayama, K. Chen, X. Sun, X. Chen, and H. Dai, Nat.
Nanotechnol., 2007, 2, 47‐52.
86. X. Zhang, L. Meng, Q. Lu, Z. Fei, and P. J. Dyson, Biomaterials, 2009, 30, 6041‐6047.
87. E. Heister, V. Neves, C. Tîlmaciu, K. Lipert, V. S. Beltran, H. M. Coley, S. R. P. Silva, and J.
McFadden, Carbon, 2009, 47, 2152‐2160.
88. P. Chaudhuri, S. Soni, and S. Sengupta, Nanotechnology, 2010, 21, 025102 (11pp).
89. R. Li, R. Wu, L. Zhao, M. Wu, L. Yang, and H. Zou, ACS Nano, 2010, 4, 1399‐1408.
90. G. Pastorin, W. Wu, S. Wieckowski, J.‐Paul Briand, K. Kostarelos, M. Prato, and A. Bianco,
Chem. Commun., 2006, 1182‐1184.
91. C. Samorì, H. Ali‐Boucetta, R. Sainz, C. Guo, M. F. Toma, C. Fabbro, T. Da Ros, M. Prato, K.
Kostarelos, and A. Bianco, Chem. Commun., 2010, 46, 1494‐1496.
92. Z. Liu, K. Chen, C. Davis, S. Sherlock, Q. Cao, X. Chen, and H. Dai, Cancer Res., 2008, 68,
6652‐6660.
93. R. P. Feazell, N. Nakayama‐Ratchford, H. Dai, and S. J. Lippard, J. Am. Chem. Soc., 2007,
129, 8438‐8439.
94. S.Dhar, Z. Liu, J. Thomale, H. Dai, and S. J. Lippard, J. Am. Chem. Soc., 2008, 130, 11467‐
11476.
95. S. C. Tsang, Y. K. Chen, P. J. F. Harris and M. L. H. Green, Nature, 1994, 372, 159‐162.
96. S. Hampel, D. Kunze, D. Haase, K. Krämer, M. Rauschenbach, M. Ritschel, A. Leonhardt, J.
Thomas, S. Oswald, V. Hoffmann, and B. Büchner, Nanomedicine, 2008, 3, 175‐182.
97. C. Tripisciano, K. Kraemer, a Taylor, and E. Borowiak‐Palen, Chem. Phys. Lett., 2009, 478,
200‐205.
98. W. Wu, R. Li, X. Bian, Z. Zhu, D. Ding, X. Li, Z. Jia, X. Jiang, and Y. Hu, ACS Nano, 2009, 3,
2740‐2750.
99. D. Yang, F. Yang, J. Hu, J. Long, C. Wang, D. Fu, and Q. Ni, Chem. Commun., 2009, 4447‐
4449.
Chapter 1
50
100. J. Meng, J. Meng, J. Duan, H. Kong, L. Li, C. Wang, S. Xie, S. Chen, N. Gu, H. Xu, and X.‐da
Yang, Small, 2008, 4, 1364‐1370.
101. Z. Zhang, Y. Zhang, B. Zeng, S. Wang, T. Zhu, R. B. S. Roden, Y. Chen, and R. Yang, Clin.
Cancer Res., 2006, 12, 4933‐4939.
102. J. E. Podesta, K. T. Al‐Jamal, M. A. Herrero, B. Tian, H. Ali‐boucetta, V. Hegde, A. Bianco,
M. Prato, and K. Kostarelos, Small, 2009, 5, 1176‐1185.
103. X. Yang, Z. Zhang, Z. Liu, Y. Ma, R. Yang, and Y. Chen, J. Nanopart. Res., 2008, 10, 815‐
822.
104. B. Pan, F. Liu, Q. Li, T. Huang, X. You, J. Shao, C. Bao, D. Cui, P. Xu, H. Chen, F. Gao, R. He,
M. Shu, and Y. Ma, Chin. J. Cancer Res., 2007, 19, 1‐6.
105. M.A. Herrero, F.M. Toma, K.T. Al‐Jamal, K. Kostarelos, A. Bianco, T. Da Ros, F. Bano, L.
Casalis, G. Scoles, M. Prato, J. Am. Chem. Soc., 2009, 131, 9843–9848.
106. Z. Zhu, Z. Tang, J. A. Phillips, R. Yang, H. Wang, and W. Tan, J. Am. Chem. Soc., 2008, 130,
10856‐10857.
107. Y. Wu, J. A. Phillips, H. Liu, R. Yang, and W. Tan, ACS Nano, 2008, 2, 2023‐2028.
108. X. Li, Y. Peng, J. Ren, and X. Qu, Proc. Natl. Acad. Sci. USA, 2006, 103, 19658‐19663.
109. K. Welsher, Z. Liu, D. Daranciang, and H. Dai, Nano Lett., 2008, 8, 586‐590.
110. Z. Liu, X. Li, S. M. Tabakman, K. Jiang, S. Fan, and H. Dai, J. Am. Chem. Soc., 2008, 130,
13540‐13541.
111. C. Zavaleta, A. de La Zerda, Z. Liu, S. Keren, Z. Cheng, M. Schipper, X. Chen, H. Dai, and S.
S. Gambhir, Nano Lett., 2008, 8, 2800‐2805.
112. Z. Ou, B. Wu, D. Xing, F. Zhou, H. Wang, and Y. Tang, Nanotechnology, 2009, 20, 105102
(7pp).
113. a) L. R. Hirsch, R. J. Stafford, J. A. Bankson, S. R. Sershen, B. Rivera, R. E. Price, J. D. Hazle,
N. J. Halas, and J. L. West, Proc. Natl. Acad. Sci. USA, 2003, 100, 13549‐13554; b) X.
Huang, I. H. El‐Sayed, W. Qian, and M. A. El‐Sayed, J. Am. Chem. Soc., 2006, 128, 2115‐
2120; c) D. P. O’Neal, L. R. Hirsch, N. J. Halas, J. D. Payne, and J. L. West, Cancer Lett.,
2004, 209, 171‐176.
114. American National Standard for Safe Use of Lasers. ANSI Z136.1, 2000
115. a) S. V. Torti, F. Byrne, O. Whelan, N. Levi, B. Ucer, M. Schmid, F. M. Torti, S. Akman, J.
Liu, P. M. Ajayan, O. Nalamasu, and D. L. Carroll, Int. J. Nanomedicine, 2007, 2, 707‐714;
b) L. Gomez‐De Arco, M.‐T. Chen, W. Wang, T. Vernier, P. Pagnini, T. Chen, M.
Gundersen, and C. Zhou, Mater. Res. Soc. Symp. Proc., 2008, 1065, QQ04‐07.
Introduction
51
116. a) H. K. Moon, S. H. Lee, and H. C. Choi, ACS Nano, 2009, 3, 3707‐3713; b) S. Ghosh, S.
Dutta, E. Gomes, D. Carroll, R. D’Agostino, J. Olson, M. Guthold, and W. H. Gmeiner, ACS
Nano, 2009, 3, 2667‐2673.
117. N. Shao, S. Lu, E. Wickstrom, and B. Panchapakesan, Nanotechnology, 2007, 18, 315101
(9pp).
118. 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. USA, 2008,
105, 8697‐8702.
119. R. Marches, P. Chakravarty, I. H. Musselman, P. Bajaj, R. N. Azad, P. Pantano, R. K.
Draper, and E. S. Vitetta, Int. J. Cancer, 2009, 125, 2970‐2977.
120. C.‐H. Wang, S.‐H. Chiou, C.‐P. Chou, Y.‐C. Chen, Y.‐J. Huang, and C.‐A. Peng,
Nanomedicine: NBM, 2011, 7, 69‐79.
121. M. R. McDevitt, D. Chattopadhyay, B. J. Kappel, J. S. Jaggi, S. R. Schiffman, C. Antczak, J.
T. Njardarson, R. Brentjens, and D. A. Scheinberg, J. Nuclear Medicine, 2007, 48, 1180‐
1189.
122. A. Ruggiero, C. H. Villa, J. P. Holland, S. R. Sprinkle, C. May, J. S. Lewis, D. A. Scheinberg,
and M. R. McDevitt, Int. J. Nanomedicine, 2010, 5, 783‐802.
52
men
intro
whic
cova
nam
with
2.1
and
an in
of C
help
SWC
orde
supe
whic
anio
Figu
This firs
ntioned, the
oduction of
ch not only
alent derivat
mely a strong
h a ball‐mill.
Oxidative s
The acid
co‐workers
ntercalation
CNTs in solut
p intercalatio
CNTs ropes,
ered layers
eracids, whe
ch disrupts v
ons.3
ure 1. A mod
Sho
st part of t
shortening o
oxidized fun
contribute t
tization of th
g oxidative t
shortening
d oxidation o
in 2006.1 Nit
step with ol
ion. Oleum i
on inside SW
making the
of acid (Fig
ere no compe
van der Waa
del illustratin
orten
the work is
of the CNTs g
nctions, mai
o solubility i
he material. T
reatment, w
of SWCNTs
of SWCNTs w
tric acid was
leum was pe
s a solution
WCNTs ropes
em swell and
gure 1).2 The
eting base is
als inter‐tube
g the swellin
ing o
s focused o
goes with an
nly carboxyl
improvemen
To reach this
with the use
s
was perform
employed a
erformed, in
of sulphuric
s. In fact, it w
d resulting
e mechanis
present, sin
e interaction
ng of a SWCN
2).
of SW
on shortenin
n improveme
ic groups, a
nt, but offer
s objective w
of acids, an
ed exploiting
as the oxidizi
order to ob
c anhydride i
was observe
in individua
m for this i
ce it consists
ns and trigge
NTs rope in s
CNTs
ng HiPCO S
nt in their so
t their tips a
also an anch
we explored
d a mechano
g a procedur
ng agent, bu
tain a homo
n sulphuric a
ed that oleum
l nanotubes
intercalation
s in the proto
ers surround
sulphuric acid
s
SWCNTs. As
olubility and
and on the
hor point fo
two distinct
ochemical a
re described
ut, ahead of t
ogeneous dis
acid, whose
m intercalat
surrounded
n is possible
onation of th
ing by sulph
d (adapted f
53
already
with the
sidewall,
or further
options,
pproach,
d by Tour
this step,
tribution
aim is to
es inside
d by few
e only in
he tubes,
huric acid
rom Ref.
Chapter
54
atmosph
final rat
the reac
thoroug
vacuum
T
differen
increase
T
with res
certain
aggressi
tubular
content
probably
with the
sample.
2
We treated
here, and we
io oleum/nit
ction mixtur
ghly washed
.
The shorten
t compleme
ed with respe
Figur
TEM images
spect to the
amount of
ve acid trea
structure an
of oxidized
y removed d
e observed o
From image
d HiPCO SW
e subsequen
tric acid of 3
e was dilute
the nanotu
ned SWCNT
ntary techni
ect to the pr
re 2. DMF dis
clearly show
e pristine on
amorphous
tment oxidiz
nd underwe
d groups. A
during the f
overall weigh
es, it is also
WCNTs with
tly added a
/1, exposing
ed in water
bes with H2O
S
Ts thus obt
ques. The di
istine SWCN
spersions of
w how the tr
es (Figures
carbonaceo
zed the tube
nt fragment
Among these
iltration pro
ht yield of 77
possible to
oleum for
mixture of o
g the mixture
and filtered
O, MeOH an
Scheme 1
tained, s‐SW
ispersibility i
Ts (Figure 2)
pristine SWC
reatment pro
3a and 3b).
ous material
es at a high
tation in sma
e highly da
ocedure carr
7%, but a gr
appreciate
r 12 h stirr
oleum and aq
e to air at 65
to remove
nd Et2O, and
WCNT‐1, we
n both organ
).
CNTs (left) a
oduced short
Neverthele
was presen
extent, and
all pieces, li
maged piec
ied out afte
reat part of t
the decreas
ring the mi
queous nitric
5°C for 2 h (S
the acid sol
d we dried t
re fully cha
nic solvents
nd s‐SWCNT
ter and less a
ss, they also
nt in our sa
part of them
kely charact
ces, the sma
r the oxidat
them was st
e in the iron
ixture unde
c acid (65%),
Scheme 1). T
vent. Finally
them under
aracterized
and water h
T‐1 (right).
aggregated C
o indicate th
ample. The q
m finally lost
terized by a
aller ones w
tion, consiste
till present in
n content of
er Ar
, to a
Then,
y, we
high
with
ighly
CNTs
hat a
quite
t the
high
were
ently
n the
f the
sam
ope
reac
off d
inte
and
Fi
Lo
in ai
and
deri
incre
mple, even th
ning of CNT
ction depicte
during the w
Metal p
rfere more
d (Figure 3)
igure 3. Repr
ower panels
More in
ir (Figure 4).
oxygenated
vative of the
ease of 35°C
hough this o
caps and the
ed in Scheme
work‐up.
particles app
with the ele
clearly show
resentative T
show enlarg
formation co
In the lowe
d groups tak
e curve indi
C in this valu
bservation i
e oxidation o
e 2: an iron
pear as spot
ectron‐beam
ws how the tr
TEM images
gements of a
poin
ould be obta
r temperatu
es place. At
cates the pr
ue is consiste
s only qualit
of the metal
water‐solub
Scheme
ts with high
of the micr
reatment str
of pristine S
area indicate
nt at metal p
ained from t
ure part of th
temperatur
recise tempe
ent with the
tative. This p
which was i
ble salt is for
2
er contrast
roscope. The
rongly reduc
WCNTs (a an
ed by the squ
particles.
he thermogr
he curve, com
res above 30
erature at w
e decrease o
Sh
phenomenon
nside the tub
rmed that co
in TEM ima
e compariso
ed their pres
nd c) and s‐S
uare in the up
ravimetric pr
mbustion of
00‐400°C CN
which this ev
of iron conte
hortening of
n is triggere
bes, accordin
ould then be
ages, since t
on between
sence.
SWCNT‐1 (b a
pper ones. A
rofile of the
amorphous
NTs burn, an
vent takes pl
ent in the sa
SWCNTs
55
ed by the
ng to the
e washed
they can
images c
and d).
Arrows
material
material
d the 1st
lace. The
ample, as
Chapter
56
observe
since it
tempera
which is
be used
SWCNTs
content
iron. Ho
groups a
CNTs wo
(Figure 5
40%), a
pictures
fact, TG
thermog
correspo
which is
2
d at the TEM
creates hot
ature of CNT
Furthermore
s oxidized du
d to obtain
s the residue
of 19.8%. F
owever, we n
and amorph
ould be highe
Comparing s
5a), a big inc
scribable to
s, to the carb
GA‐MS of s‐
gravimetric
ond to SO+
in turn gene
M. In fact, the
spots in the
T oxidation.4
e, the residu
uring this ana
a quantitati
e is of 26.1%
For s‐SWCNT
need to cons
hous carbon,
er.
Figure 4. TG
s‐SWCNT‐1 T
crease in the
o different m
boxylic acids
‐SWCNT‐1 le
experiment
and SO2+ re
erated by su
e presence o
sample dur
ual weight o
alysis, resulti
ve data for
(in the form
T‐1 the Fe2O
sider that a b
and therefo
GA (air) of pr
TGA under N
e weight loss
material, na
introduced
et us analys
and peaks
spectively, d
lfonic acid, in
of metal part
ring the ther
of air TGA is
ing in iron(II
r the iron co
of Fe2O3). It
3 residue is
big part of th
ore the perc
ristine SWCN
N2 atmosphe
s was observ
mely to the
with the cu
se the gas
of mass 48
deriving from
n the same w
ticles decrea
mal experim
due to the
I) oxide (Fe2
ontent of th
t is therefore
of 4.7%, wh
his sample is
centage of ir
NTs and s‐SW
ere with the
ed at tempe
e amorphou
tting, but al
evolving fr
8 and 64 we
m decompos
way CO2 is ge
ses thermal
ment, and thu
iron presen
O3). This val
he CNT sam
e possible to
hich gives a v
s represente
ron with res
WCNT‐1.
one of the
eratures belo
us carbon se
so to some
om the sam
ere found (F
sition and io
enerated by
stability of C
us it reduces
nt in the sam
ue can there
mple. For pri
calculate an
value of 3.3
ed by oxygen
spect to the
pristine SWC
ow 800°C (al
een in the
sulfonic acid
mple during
Figure 5b).
onization of
carboxylic ac
CNTs,
s the
mple,
efore
stine
n iron
3% of
nated
only
CNTs
most
TEM
ds. In
g the
They
SO2,
cid.
Figu
and
chan
sing
sem
a bi
toge
sam
tube
Figu
2.1.
in b
ure 5. a) TGA
Other in
Raman spe
nges in both
ularities wa
miconducting
ig increase i
ether with a
me time, a los
es were dest
ure 6. a) UV‐v
nm) of pristi
1 Re‐pristini
During t
etter unders
A (N2) of prist
nformation r
ctroscopies,
h kind of spe
as found, es
one. Regard
in the D/G
background
ss of most o
troyed by the
vis‐NIR spect
ne SWCNTs
ization of the
the Raman a
standing the
tine SWCNTs
regarding the
since disrup
ectra, as can
specially in
ding the Ram
ratio, becau
d fluorescen
f the RBM b
e harsh cond
tra of pristin
and s‐SWCN
e oxidized SW
analyses, we
e nature of t
s and s‐SWCN
e modificatio
ption of the
be observe
the metallic
man spectra,
use of the i
ce, due to t
bands was ob
ditions of the
e SWCNTs a
NT‐1. In the in
shown.
WCNTs
observed a
he product o
NT‐1; b) TGA
on of CNT st
sp2 networ
ed in Figure 6
c region of
the oxidativ
ntroduction
the presence
bserved, sinc
e treatment.
nd s‐SWCNT
nset, the enl
very particu
obtained. Qu
Sh
A‐MS analysis
tructure are
k of the CN
6. A partial l
the spectru
ve treatment
of many d
e of amorph
ce the smalle
T‐1 (DMF); b)
argement of
ular behaviou
uite surprisin
hortening of
s (He) of s‐SW
e given by UV
Ts resulted
loss of the v
um, but als
t led, as exp
efects on th
hous materia
er and more
) Raman spec
f the RBM zo
ur, which m
ngly, increas
SWCNTs
57
WCNT‐1.
V‐vis‐NIR
in major
van Hove
o in the
ected, to
he CNTs,
al. At the
e reactive
ctra (633
one is
ight help
sing laser
Chapter 2
58
power a sort of re‐pristinization of the sample occurred, as can be clearly observed in Figure 7.
We analysed the Raman bands of the sample using not only the 633 nm laser, but also the 785
nm and the 532 nm ones and we obtained spectra similar to the one at 633 nm, in terms of D‐
band and loss of RBM. The slight differences are, of course, due to the fact that different tubes
are in resonance at the energy of the specific laser used.
Considering the 785 nm laser, the starting laser power used was 0.57 mW/µm2. Laser
power density was then gradually increased. At 2.86 mW/µm2 we started to observe a decrease
in the D/G ratio, a lower baseline and a better definition of RBM bands. Afterwards we exposed
the sample to a laser power density of 5.73 mW/µm2, which gave a too high response,
corresponding to a saturated spectrum (not reported), but resulted in a complete re‐
pristinization of the sample, appreciable when the analysis was repeated again at lower laser
power (Figure 7a). By comparison of this spectrum with the one of the pristine SWCNTs obtained
using the same laser wavelength, we can clearly notice that the two profiles are pretty similar,
apart from the RBM zone, where an enrichment of certain tubes with respect to others is
evident (Figure 7b). Also, an overall increase in the Raman cross section was observed (note that
all the spectra are normalized to the G band maximum), in fact we needed to lower the laser
power density to collect the spectrum after the re‐pristinization took place (Figure 7a).
The same phenomenon was observed using the 532 nm and 633 nm lasers, with laser
power of 2.07 mW/µm2 and 2.55 mW/µm2 respectively. In both cases, the main difference
between the pristine SWCNTs and the re‐pristinized ones were in the RBM zone and in the shape
of the G‐band, both correlated with different composition of the SWCNT sample (Figures 7c and
7d).
We hypothesize that this phenomenon is due to thermal effects. In fact, treating CNTs
with a high power laser resulted in a local increase of temperature, to which we ascribe an
annealing of amorphous carbon and structural defects, and, consequently a re‐establishment of
the sp2 network. Of course, the tubes that where completely destroyed during the oxidation
were not present anymore, therefore leading to an enrichment of the sample with remaining
tubes, as confirmed by the changes in the RBM and G‐band patterns. In particular, the loss of the
higher wavenumber RBM bands, observed with all the lasers, confirmed the selective
destruction of smaller diameter tubes by the aggressive acid treatment.
A confirmation for this thermal effect could arise from the reversible shift in the G‐band
to lower wavenumbers (from approximately 1593 nm to 1584 nm) upon increasing of laser
power density (Figure 7a), since this downshift has been correlated with a lengthening of the C‐C
distances as the nanotube undergoes thermal expansion.5
Fig
en
lase
refe
lase
a rev
with
repo
high
anne
lase
out
betw
relia
Ano
lowe
gure 7. a) Ra
largement o
ers, before a
Few pap
er to pristine
r power den
versible shift
h an increas
orted, explai
her Raman
ealing of the
r power den
that this va
ween Stoke
able method
ther report
ering of the
aman re‐prist
of the G‐band
nd after re‐p
In the in
pers appeare
e CNTs. In tw
nsities of 1 or
t of the G pe
e in their in
ined as a re
response. B
e CNTs. They
nsity used, w
alue should
and anti‐Sto
. In fact diffe
appeared,
D/G ratio, fu
tinization pro
d zone is sho
pristinization
sets, the enl
ed in the lit
wo cases, pr
r 0.1 mW/µm
eak to lower
ntensities. A
esult of the
Both authors
y calculated
which was in
be consider
oke intensit
erent kind of
regarding p
urther confirm
ocess of s‐SW
own; b, c, d)
n of the samp
largement of
terature so
ristine SWCN
m2, and the a
wavenumbe
Also a genera
higher purit
s explained
the temper
the range 30
red just as i
ies, which,
f CNTs are re
pristine MWC
ming a sort o
WCNT‐1 with
Raman spect
ple, in compa
f the RBM zo
far, reportin
NTs were tre
authors obse
ers, and a ch
al increase i
ty of the sam
the fact as
ature reache
00‐450°C. Ne
ndicative, si
in resonance
esonant in th
CNTs, where
of annealing
Sh
h 785 nm lase
tra of s‐SWC
arison with t
one is shown
ng a similar
eated with t
erved a decre
ange in the R
in the Rama
mple, since
s a thermal
ed in their s
evertheless,
nce it is cal
e Raman sp
he Stoke and
e a laser tre
and purifica
hortening of
er. In the ins
CNT‐1 with d
the pristine S
n.
behaviour,
the 633 nm
ease in the D
RBM peaks,
an cross sec
more CNTs
effect, res
ample, at th
it should be
culated by t
pectroscopy,
d anti‐Stoke
eatment res
ation effect.7
SWCNTs
59
set, the
ifferent
SWCNTs.
but they
laser, at
D/G ratio,
together
tion was
produce
ulting in
he higher
e pointed
the ratio
is not a
spectra.6
sulted in
Chapter
60
of our k
mW/µm
authors
defects,
zone.8 A
contrary
the D/G
reported
method
apprecia
harsh tre
2.1.2 So
washing
amorph
highly o
that wou
We trea
Afterwa
and thor
T
successf
content
2
Regarding o
knowledge. A
m2) of oxidize
ascribed to
since they
Another repo
y to our find
ratio, and a
In conclusio
d for oxidize
for the oxi
ate their Ram
eatment.
odium Hydro
In the atte
gs with sodiu
ous impuriti
xidized carb
uld bring the
ated s‐SWC
rds, purified
roughly wash
The purifie
fulness of th
of amorpho
xidized SWC
A paper pub
ed SWCNTs,
o degassing
did not obse
ort on 633 n
ings, the sele
decrease in
on, this is th
ed CNTs. Th
dized SWCN
man features
oxide washin
mpt to pur
um hydroxid
es deriving f
onaceous m
em in solutio
NTs‐1 in aq
d CNTs were
hed with H2O
d s‐SWCNT
e procedure
ous materia
CNTs, no sim
lished in 20
resulting in
of the samp
erve any dec
nm laser trea
ective destru
the Raman c
e first time
his effect co
NTs, and, qu
s, many of th
ng
rify the oxid
de (NaOH).
from the stro
material, and
on, allowing t
queous NaO
separated f
O, MeOH and
S
Ts‐2 and th
. TEM image
l, found ins
milar results a
00 described
a change of
ple due to h
crease in the
atment of pr
uction of me
cross section
that such a
ould in princ
uite importa
he CNTs are
dized s‐SWC
In fact, as d
ong oxidative
could there
their remova
OH 8M at 1
from the dis
d Et2O to aff
Scheme 3
he filtrate
es clearly sho
stead in the
are present
d a similar la
f the line sha
heating effec
e D/G ratio,
ristine SWCN
etallic tubes,
n.9
a laser‐induc
ciple be exp
ntly, it tells
still present
CNT‐1 we d
described by
e treatment
efore be rem
al through fil
100°C for 4
solved amor
ord s‐SWCNT
were char
owed much c
filtrate whe
in the literat
aser treatme
ape of the G
cts, but not
nor any cha
NTs at 1 mW
together wi
ced re‐pristin
loited to fin
us that, ev
in the samp
ecided to p
y several rep
are probably
oved by an
tration.10
48h, under
rphous mate
T‐2 (Scheme
acterized to
cleaner CNTs
ere no CNTs
ture, to the
ent (633 nm
G bands, tha
to annealin
ange in the
W/µm2 descr
ith an increa
nization effe
nd a purifica
ven if we ca
ple even afte
perform alk
ports, the m
y represente
alkaline was
Ar atmosph
erial by filtra
e 3).
o evaluate
s, with a red
s were dete
best
m, 2.2
t the
ng of
RBM
ribes,
se in
ect is
ation
nnot
r the
aline
major
ed by
shing
here.
ation,
the
uced
ected
(Figu
SWC
this
Figu
for
sam
Figu
ure 8). Neve
CNTs‐1, and T
product, diff
ure 8. Repres
Coheren
N2 and air
mple the amo
ure 9. TGA of
ertheless, the
TEM images
ferently from
sentative TE
ntly with the
measureme
orphous mate
f pristine SW
e dispersibil
confirmed i
m s‐SWCNT‐1
M images of
se observati
nts, confirm
erial, which b
WCNTs, s‐SWC
ity of s‐SWC
t, since man
1.
f s‐SWCNT‐2
the filtrate
ions, TGA sh
ming that ou
burns at low
CNT‐1 and s‐
CNTs‐2 stron
ny bundles an
at different
(d).
owed a big d
ur treatmen
w temperatur
‐SWCNT‐2, p
Sh
gly decrease
nd big aggreg
magnificatio
decrease in t
t effectively
res (Figure 9)
performed in
hortening of
ed with resp
gates were f
on (a, b and c
the weight lo
y removed f
).
n air (a) and i
SWCNTs
61
pect to s‐
found for
c) and of
oss, both
from the
n N2 (b).
Chapter
62
T
fact, as c
and the
fluoresc
Figure
T
pristine
effective
obtained
that not
introduc
Figure 1
nm (
obtained
2
The Raman a
can be seen
Raman spe
ence backgr
10. Raman s
The Raman
SWCNTs, a
ely cleaned
d for s‐SWC
t all the nano
ced on the st
11. Raman sp
(a) and the 7
Finally, we
d exactly th
analyses sho
in Figure 10
ectrum of th
ound, did no
spectra (633
In the inset,
spectra obta
as shown in
the sample
NT‐2 were d
otubes in th
tructure, by t
pectra of pris
85 nm laser
performed
e same pro
owed that th
, the profile
he filtrate, b
ot present RB
nm) of prist
, the enlarge
ained at 532
Figures 11
from carbon
different acc
e sample we
the oxidative
stine SWCNT
(b). In the in
the re‐prist
file previous
he NaOH trea
became mo
besides havin
BM bands.
ine SWCNTs
ement of the
2 nm and 78
a and 11b,
naceous imp
cording to th
ere modified
e treatment.
Ts, s‐SWCNT‐
nsets, the en
tinization ex
sly reported
atment deep
re similar to
ng a very hi
, s‐SWCNT‐1
RBM zone is
85 nm resem
indicating t
purities. Mo
he laser wav
d at the sam
.
‐1 and s‐SWC
largement o
xperiment al
for re‐prist
ply cleaned th
the one of p
gh D‐band a
, s‐SWCNT‐2
s shown.
mbled even m
that the alk
reover, since
velength, we
e extent, in
CNT‐2, obtain
f the RBM zo
lso for s‐SW
tinized s‐SW
he nanotube
pristine SWC
and a signif
2 and its filtra
more the on
kaline treatm
e the D/G r
e could conc
terms of de
ned with the
one is shown
WCNT‐2, and
WCNT‐1 (data
es. In
CNTs,
icant
ate.
es of
ment
atios
clude
fects
e 532
n.
d we
a not
Shortening of SWCNTs
63
shown). This is a further confirmation that the phenomenon we observed consisted not only in a
purification of the sample from the amorphous carbon content, but also in a structural recovery.
In conclusion, oleum/nitric acid treatment of SWCNTs resulted in the production of
shortened and disentangled CNTs, bearing carboxyl and sulfonic groups, with a big improvement
in dispersibility. On the other hand, the treatment produced a lot of amorphous carbonaceous
material, which was successfully removed by a subsequent NaOH washing treatment. In
addition, we observed for the first time a laser‐mediated re‐pristinization of the SWCNT tubular
structure, that effectively mended the defects introduced by the oxidation.
2.2 Mechanochemical shortening of SWCNTs
One of the main limitations of a wet approach for SWCNT shortening is their lack of
solubility as pristine material that can determine a non‐homogenous cutting, since the material
is in form of bundles and therefore the oxidation will not evenly affect all the nanotubes in each
bundle. With the oleum‐based treatment, as already explained, the intercalation step was
intended for a better homogeneity of the dispersion. In the case of the ball milling shortening,
the problem of the lack of solubility has been directly bypassed, since we worked in the absence
of any solvent. At the same time a dry approach has the advantage of reduced pollution, lower
costs, and ease of scale‐up. Mechanochemistry involves the transformation of mechanical
energy into the driving force for chemical modifications, that can be seen as an alternative way
to overcome thermodynamic barriers with respect to more classical systems, such as heating or
enzymatic catalysis, and it is generally performed in the absence of any solvent. Few examples
have been reported so far in this field: e.g. stress‐induced homolytic bond cleavage for polymers
has been extensively studied.11
In the case of CNTs, the mechanical stress should break them in correspondence of pre‐
existing defects, where local strain is higher, but could also induce local deformations, such as
kinking or twisting, that cause strain energy to concentrate in some points, where breaking will
occur. Some studies have already been published where ball milling was used to cut MWCNTs,
leading to the formation of carbon nanoparticles similar to carbon onions (spherical form of
carbon made of concentric fullerenes),12 or to shortened MWCNTs.13 Also a study on the
shortening of SWCNTs has been published, but in this case the authors observed an aggregation
effect of the ball milling treatment on the CNTs.14
We have studied the possibility of using mechanical activation through a planetary ball
mill, for obtaining large‐scale quantities of shortened SWCNTs (s‐SWCNT), with a homogeneous
distribution of length. The instrument is made of a steel grinding jar containing steel grinding
Chapter
64
balls, wh
in one d
our instr
with the
of the m
Fig
time of
HCl in or
milling,
or D ac
calculate
the tech
structur
milling t
formatio
eliminat
with exp
indeed t
2
here the ma
direction, and
rument), thu
e inner walls
material.
gure 12. Plan
HiPCO SWCN
milling were
rder to remo
and the air o
cording to t
ed and the m
hnique succ
es, the weig
times led to
on of volatil
ted during th
periments pe
that the cor
terial is plac
d the wheel
us subjecting
of the jar ca
netary ball m
NTs were ba
e varied (Sch
ove any Fe2O
oxidation of
the differen
material was
cessfully sho
ght yields w
o almost com
le by‐produc
he washing t
erformed in
rect choice
ced (Figure 1
in the oppos
g the balls to
auses frictio
mill apparatus
all‐milled in a
eme 4). Foll
O3 possibly d
the iron pre
t speed and
s fully charac
ortened the
were not sat
mplete destr
cts and high
treatment. F
N2 atmosph
of the exper
12). Once ins
site one (up
o superimpos
nal and imp
s. Red arrow
a set of expe
lowing the tr
eriving from
esent inside,
d time of tr
cterized. Eve
CNTs with
tisfactory. In
ruction of SW
hly oxidized
For this reaso
here, giving
rimental con
side the plan
to a maximu
sed rotationa
act forces, t
ws indicate th
eriments in
reatment, th
m the opening
and MeOH,
eatment). F
en though ch
out causing
n fact, high
WCNTs, that
material w
on, we decid
s‐SWVNT‐4
nditions cou
netary mill, t
um of 650 rp
al movement
hus leading t
he direction o
which rotati
he material w
g of the tube
to afford s‐S
inally, the w
haracterizatio
g severe alt
rotational s
t could be e
hich could
ded to compa
(B, C, E or F
ld lead to re
the jar is rota
pm in the ca
ts. Their coll
to size reduc
of rotation.
ional speeds
was washed
es during the
SWCNT‐3 (A,
weight yield
ons showed
teration of
speeds and
explained by
be subseque
are these re
F). We found
easonable yi
ating
se of
ision
ction
s and
with
e ball
, B, C
was
that
their
long
y the
ently
esults
d out
ields,
whil
weig
our
prod
SWC
prob
the
dich
perm
pres
info
char
the
the
of th
le retaining
ght yields are
s‐SWCNT‐3
s‐SWCNT‐3
s‐SWCNT‐4
s‐SWCNT‐3
s‐SWCNT‐4
s‐SWCNT‐3
s‐SWCNT‐4
s‐SWCNT‐4
Tab
The disp
results, the
ducing in gen
CNT‐3, obtai
bably due to
other hand
hloromethan
mitted to b
sented good
The UV‐
rmation on
racteristic va
Raman spec
one of the p
he nanotube
the structur
e summarize
Atm
3A
3B
4B
3C
4C
3D
4E
4F
le 1. Differen
persibility of
milling treat
neral better
ned in air, p
o the oxidativ
d these tub
e. Different
boost more
dispersibility
‐vis‐NIR and
n SWCNTs.
an Hove sing
ctra of s‐SWC
pristine SWC
es remained
ral integrity
ed in Table 1
mosphere
air
air
N2
air
N2
air
N2
N2
nt experimen
s‐SWCNT de
tment decrea
dispersions.
resented en
ve defects cr
es did not
ly, the use
experimen
y both in me
the Raman
The UV‐vis
gularities, as
CNT obtained
CNTs (Figure
largely intact
of SWCNTs.
.
Scheme
Speed (rp
100
250
250
500
500
650
650
650
ntal conditio
epended crit
ases the am
But the atm
hanced disp
reated along
give good
of a nitrog
ntal conditio
ethanol and i
spectra, as a
s‐NIR spectr
s the examp
d with mech
13b). These
t after ball m
The differe
4
m) Tim
ons for the ba
tically on the
ount of bund
mosphere cho
persibility in
g the tube du
dispersions
gen atmosph
ons, thus o
in dichlorom
already expla
rum display
le reported
hanochemica
e findings ind
milling.
Sh
nt condition
me (min)
30
120
120
30
30
30
120
360
all milling tre
e milling con
dles and the
osen played
polar solven
uring the me
in less pol
here for the
obtaining sm
methane.
ained, provid
ed for all
in Figure 13
al treatment
dicate that th
hortening of
ns used and
Yield (wt%
39
43
70
35
74
9
71
48
eatment.
ditions. Acco
e length of th
an importan
nts such as m
echanical pro
ar solvents
e milling tre
maller tubes
de reliable s
the s‐SWC
3a shows. M
were very s
he intrinsic s
SWCNTs
65
the final
%)
ording to
he tubes,
nt role: s‐
methanol,
ocess. On
such as
eatment,
s, which
structural
CNTs the
Moreover,
similar to
structure
Chapter
66
Figure
rpm, 1
120
SWCNTs
D/G rati
SWCNT‐
nm indic
contrary
mechan
Figure 1
T
organic
a higher
tubes by
2
e 13. a) UV‐vi
120 min, N2);
0 min, air), 4
By comparis
s (see Figure
io of s‐SWC
‐1, all the RB
cates that b
y to the be
isms were in
14. TGA (N2)
min
TGA data ca
moieties att
r weight loss
y the milling
is‐NIR spectr
b) Raman sp
B (250 rpm,
enla
son of these
6b in Chapt
NT‐3 and 4
BM peaks we
igger diamet
ehaviour ob
nvolved in th
of pristine S
, air), 3D (65
an be used
tached to CN
s (Figure 14)
g treatment,
ra of DMF dis
pectra (633 n
120 min, N2)
argement of
spectra wit
er 2.1), rema
did not pre
ere preserve
ter tubes we
bserved for
e two cases.
SWCNTs and
50 rpm, 30 m
to estimate
NT sidewalls.
that may ar
, but could
spersions of
nm) of pristi
) and 4E (650
f the RBM zo
h the one o
arkable diffe
sent a dete
ed. Interestin
ere the mos
the oxidati
.
s‐SWCNT‐3A
min, air) and 4
e CNT purit
. The thermo
ise from dec
also result f
pristine SWC
ne SWCNTs
0 rpm, 120 m
ne is shown.
btained with
erences could
ctable increa
ngly, the low
t affected by
ve shorteni
A (100 rpm, 3
4C (500 rpm,
y and the p
ogravimetric
composition
from adsorb
CNTs and s‐S
and s‐SWCN
min, N2). In th
.
h the oxidati
d be appreci
ase, and, dif
decrease in
y the mecha
ng, meaning
30 min, air),
, 30 min, N2)
presence an
profiles of s
of defects c
ed molecule
SWCNT‐4E (6
NT‐3B (250 rp
he inset, the
ive shortenin
iated. In fact
fferently fro
n the peak at
anical treatm
g that diffe
3C (500 rpm
.
nd abundanc
s‐SWNTs sho
created along
es. In fact it
650
pm,
ng of
t, the
om s‐
t 193
ment,
erent
m, 30
ce of
owed
g the
was
repo
resu
resp
stru
amo
12‐1
basi
milli
subs
auth
carb
aggr
disp
Mor
trea
Fig
orted that C
ults suggeste
pect to s‐SW
cture occurr
Accordin
ount of iron
14% for the
s of a decom
ing treatme
sequently re
Figure 1
hors have p
bon, due to
regates of e
persed tubes
reover, as ex
tment.
gure 15. Rep
120 min
CNTs treated
ed the prese
WCNT‐3, cor
red thanks to
ng to the re
present in th
milling trea
mposition of
ent. Thus, i
moved with
15 shows so
reviously de
o the mech
entangled S
s, and to exf
xpected, tub
resentative T
n, air) (b), 4B
with ball m
nce of much
roborating t
o the inert at
esidual mas
he samples f
ted ones. Th
the carbon s
n the prese
the acid was
me typical T
escribed the
anochemica
WCNTs.14 N
foliate the o
es shorten w
TEM images
(250 rpm, 1
mill increase
h less impuri
the hypothe
tmosphere.
s in the TG
falls from 19
his decrease
shells that su
ence of O2
shings.
TEM images
e undesired
l treatment
Nevertheless,
original bund
with the incr
of s‐SWCNT
20 min, N2) (
their adsor
ties on s‐SW
esis that less
GA performe
9.8% for pris
e could be e
urround the
, Fe is oxid
s of the tub
formation o
t of SWCNT
, our treatm
dles, as can
rease in the
T‐3A (100 rpm
(c) and 4E (6
Sh
ptive capaci
WCNT‐4, prep
s oxidative
ed in air (da
tine tubes to
explained, as
metal partic
dized to Fe
es obtained
of large amo
Ts, and the
ment seems
be apprecia
rotational sp
m, 30 min, ai
50 rpm, 120
hortening of
ity.15 Moreo
pared under
damage of
ata not sho
o values in t
s anticipated
les, promote
e2O3, which
after millin
ounts of am
formation
s to give cl
ated by TEM
peed and the
ir) (a), 3B (25
0 min, N2) (d)
SWCNTs
67
ver, TGA
N2, with
the CNT
wn), the
he range
d, on the
ed by the
can be
ng. Other
morphous
of large
ean and
M images.
e time of
50 rpm,
).
Chapter
68
we dec
meaning
individu
effective
water di
surfaces
an oven
reports
the tub
bundles
good ag
vertical
Figure
rp
the leng
between
conditio
2
In order to e
ided to pe
gful evaluatio
ally disperse
e in helping
ispersions of
s by spin‐coa
.16 Finally, th
a representa
ular objects
. The height
greement wit
sample com
e 16. Repres
pm, 120 min
Samples s‐SW
gth distributi
n the milling
ons led to a
evaluate the
rform statis
on, however
ed CNTs wer
CNTs dispe
f the selecte
ating, and we
he surfaces w
ative TM‐AFM
s observed
profile show
th the typica
pression of C
entative TM
n, N2); c) heig
WCNT‐3B an
ion of the tu
g treatment
narrow dist
e length dist
stical analys
r, we could n
e needed. T
ersion, i.e. so
d samples. W
e thermally r
were analyze
M image of s
at the AFM
wn in Figure
al HiPCO SWC
CNTs when t
‐AFM height
ght profile of
nd 4B, and s
ubular object
at 250 rpm
ribution of s
ribution of t
ses using a
not settle for
herefore we
odium dode
We deposited
removed the
ed through t
s‐SWCNT‐4B
M let us dis
16b, shows a
CNT diamete
they are touc
t (a) and amp
f the line ind
s‐SWCNT‐4E
ts observed
m for 120 m
shorter CNT
the s‐SWCNT
tomic force
r the thin bun
e used a surf
ecylbenzenes
d the dispers
e surfactant,
tapping mod
B, showing in
tinguish bet
a vertical dis
ers, consider
ched by the A
plitude (b) im
icated in the
were chosen
in the TM‐A
min in air or
s (Figure 17
Ts obtained w
e microscop
ndles observ
factant know
sulfonate (SD
sions on fres
heating the
de‐AFM (TM‐
dividual tub
tween indiv
stance of 0.7
ring the well
AFM tip.17
mages of s‐SW
e topographic
n as represe
AFM images.
N2 revealed
a). The mea
with ball mi
py (AFM). F
ved at TEM, s
wn as being q
DBS), to pre
shly cleaved
mica surfac
‐AFM). Figur
es. The heig
vidual tubes
71 nm, which
l‐known effe
WCNT‐4B (2
c image a.
entative, to s
The compa
d that the l
an lengths of
lling,
or a
since
quite
epare
mica
ces in
re 16
ht of
and
h is in
ect of
50
study
rison
atter
f the
sam
N2
dest
our
ther
the
and
atm
redu
conf
leng
Fi
SW
mille
N2, s
of th
SWC
to t
pow
decr
micr
cons
mples were 0
(s‐SWCNT‐4B
truction of S
results we ca
refore leadin
lower yields
4B respect
osphere is u
uced the me
firming our h
gth distributi
igure 17. a) L
WCNT‐4B (250
120 min,
In order
ed SWCNTs
strongly low
he tubes to
CNT‐4F (Figu
We tried
his sample,
wer laser tr
reased, and
roscope aft
stituted by v
.74 ± 0.51 µ
B) respectiv
SWCNTs with
an hypothes
ng to the two
s relative to
tively). Ther
used. Under
ean length t
hypothesis th
on of the s‐S
Length distri
0 rpm, 120 m
N2, count =
r to determin
at 650 rpm,
wered the we
quite a cons
re 18).
d to apply th
but it was u
eatment, b
the focal
er the trea
very short tub
µm and 0.46
vely. In fact
h respect to
size that shor
o different le
s‐SWCNT‐3 w
refore the
N2 instead,
to 0.36 ± 0.
hat, tuning t
SWCNT.
bution of s‐S
min, N2, coun
315). Vertica
ne if there w
for much lo
eight yield (4
siderable ext
he laser re‐p
unsuccessful
ut the opp
area appea
atment. The
bes or by am
± 0.35 µm f
t, as previo
N2, speed a
rter tubes, o
ength distrib
with respect
production
the increase
.16 µm, with
he condition
SWCNT‐3B (2
nt = 775); b) l
al lines indica
was a limit in
onger time,
48 %). Moreo
tent, as can
ristinization
l. In fact, tu
posite effect
ared emptie
erefore, we
morphous ma
for the treat
ously discus
nd time bei
nce formed,
bution profile
t to s‐SWCNT
of short SW
e in the rota
h no tubes
ns, it would h
250 rpm, 120
length distrib
ate the avera
n boosting e
namely 6 h.
over, these c
be clearly se
observed fo
bes not only
t occurred:
ed when lo
could con
aterial, was b
Sh
ment in air (
ssed, air co
ng equal. M
are more ea
es. This fact
T‐4 (43% vs.
WCNTs is p
tional speed
longer than
have been po
0 min, air, co
bution of s‐S
age length of
experimental
The treatm
conditions tr
een by the Ra
or s‐SWCNT‐1
y did not re
Raman cro
oking at th
clude that
burned by th
hortening of
(s‐SWCNT‐3B
ould lead to
oreover, con
asily destroy
can also acc
70% for sam
prevented w
d to 650 rpm
n 1 μm (Figu
ossible to co
ount = 1288)
SWCNT‐4E (6
f the sample
l conditions,
ent, though
riggered the
aman spectr
1 (see Chapt
ecover after
oss section
he sample w
the sample
he laser.
SWCNTs
69
B) and in
o higher
nsidering
yed in air,
count for
mples 3B
when air
m further
ure 17b),
ntrol the
and s‐
650 rpm,
es.
we ball‐
being in
e damage
rum of s‐
ter 2.1.1)
the high
strongly
with the
e, either
Chapter
70
Figure 1
objects
average
(such as
individu
of shorte
Figure
r
shortene
these tu
the poss
it a good
2
18. Raman sp
Nevertheles
that appear
length was
s HR‐TEM) w
ally disperse
er treatment
e 19. Represe
report the sa
In conclusio
ed SWCNTs
ubes showed
sibility to co
d alternative
pectra (633 n
s, we could
red as 3 D w
measured t
would be nee
ed without th
t time is reco
entative TEM
ame image w
on, we repo
with a pla
d that the tre
ntrol the qu
e to wet chem
nm) of pristin
visualize the
when adjust
to be 0.15 ±
ded to asses
he need of a
ommended.
M image of s‐
with white lin
orted an ea
netary mill.
eatment did
ality of the f
mistry appro
ne SWCNTs a
e material ob
ting the foc
± 0.09 µm (
ss it doubtle
any surfactan
‐SWCNT‐4F (
nes drawn to
sy procedur
The differe
d not compro
final product
aches.
and s‐SWCN
btained at T
us of the m
count = 84)
ss, these ob
nt. However
(650 rpm, 36
help the eye
re for prep
ent complem
omise their t
ts tuning exp
T‐4F (650 rp
EM, finding
microscope (F
. Even if dee
jects seemed
, given these
60 min, N2). T
e in recogniz
aring scalab
mentary cha
tubular struc
perimental c
pm, 360 min,
very short li
Figure 19). T
eper evaluat
d to be SWC
e results, the
The right pan
zing CNTs.
ble quantitie
racterization
cture. Moreo
conditions re
N2).
inear
Their
tions
CNTs,
e use
nel
es of
ns of
over,
ender
Shortening of SWCNTs
71
2.3 Acknowledgments
The study on mechanochemical shortening of SWCNTs was carried out in collaboration
with Noelia Rubio, Prof. M. Antonia Herrero and Prof. Ester Vázquez (Departamento de Química
Orgánica, Facultad de Químicas‐IRICA, Universidad de Castilla‐La Mancha in Ciudad Real, Spain).
In particular, I wish to thank Noelia Rubio for some of the milling treatment and TGA.
2.4 References
1. Z. Chen, K. Kobashi, U. Rauwald, R. Booker, H. Fan, W.‐F. Hwang, and J. M. Tour, J. Am.
Chem. Soc., 2006, 128, 10568‐10571.
2. L. M. Ericson, H. Fan, H. Peng, V. A. Davis, W. Zhou, J. Sulpizio, Y. Wang, R. Booker, J.
Vavro, C. Guthy, A. N. G. Parra‐Vasquez, M. J. Kim, S. Ramesh, R. K. Saini, C. Kittrell, G.
Lavin, H. Schmidt, W. W. Adams, W. E. Billups, M. Pasquali, W.‐F. Hwang, R. H. Hauge, J.
E. Fischer, and R. E. Smalley, Science, 2004, 305, 1447‐1450.
3. S. Ramesh, L. M. Ericson, V. A. Davis, R. K. Saini, C. Kittrell, M. Pasquali, W. E. Billups, W.
W. Adams, R. H. Hauge, and R. E. Smalley, J. Phys. Chem. B, 2004, 108, 8794‐8798.
4. S. Arepalli, P. Nikolaev, O. Gorelik, V. G. Hadjiev, W. Holmes, B. Files, and L. Yowell,
Carbon, 2004, 42, 1783‐1791.
5. P.V. Huong, R. Carvagnat, P.M. Ajayan, O. Stephan, Phys. Rev. B 51 (1995) 10048.
6. a) P. Corio, P. S. Santos, M. a Pimenta, and M. S. Dresselhaus, Chem. Phys. Lett., 2002,
360, 557‐564; b) L. Zhang, H. Li, K.‐T. Yue, S.‐L. Zhang, X. Wu, J. Zi, Z. Shi, and Z. Gu, Phys.
Rev. B, 2002, 65, 07340 (4).
7. J. S. Kim, K. S. Ahn, C. O. Kim, and J. P. Hong, Appl. Phys. Lett., 2003, 82, 1607.
8. Z. Yu and L. E. Brus, J. Phys. Chem. A, 2000, 104, 10995‐10999.
9. H. Huang, R. Maruyama, K. Noda, H. Kajiura, and K. Kadono, J. Phys. Chem. B, 2006, 110,
7316‐7320.
10. a) E. Heister, C. Lamprecht, V. Neves, C. Tilmaciu, L. Datas, E. Flahaut, B. Soula, P.
Hinterdorfer, H. M. Coley, S. R. P. Silva, and J. McFadden, ACS Nano, 2010, 4, 2615‐2626;
b) B. C. G. Salzmann, S. A. Llewellyn, G. Tobias, M. A. H. Ward, Y. Huh, and M. L. H.
Green, Adv. Mater., 2007, 19, 883‐887.
11. M. K. Beyer and H. Clausen‐Schaumann, Chem. Rev., 2005, 105, 2921‐2948.
12. Y. B. Li, B. Q. Wei, J. Liang, Q. Yu, and D. H. Wu, Carbon, 1999, 37, 493‐497.
13. a) N. Pierard, A. Fonseca, Z. Zonya, I. Willems, G. Van Tendeloo, and J. B. Nagy, Chem.
Phys. Lett., 2001, 335, 1‐8; b) Z. Kónya, J. Zhu, K. Niesz, D. Mehn, and I. Kiricsi, Carbon,
2004, 42, 2001‐2008.
Chapter 2
72
14. N. Pierard, A. Fonseca, J.‐F. Colomer, C. Bossuot, J.‐M. Benoit, G. Van Tendeloo, J.‐P.
Pirard, and J. B. Nagy, Carbon, 2004, 42, 1691‐1697.
15. Y. F. Sun, A. M. Zhang, Y. Yin, Y. M. Dong, Y. C. Cui, X. Zhang, and J. M. Hong, Mater.
Chem. Phys., 2007, 101, 30‐34.
16. M. F. Islam, E. Rojas, D. M. Bergey, A. T. Johnson, and A. G. Yodh, Nano Lett., 2003, 3,
269‐273.
17. a) T. DeBorde, J. C. Joiner, M. R. Leyden, and E. D. Minot, Nano Lett., 2008, 8, 3568‐3571;
b) V. Zorbas, A. Ortiz‐Acevedo, A. B. Dalton, M. M. Yoshida, G. R. Dieckmann, R. K.
Draper, R. H. Baughman, M. Jose‐Yacaman, and I. H. Musselman, J. Am. Chem. Soc.,
2004, 126, 7222‐7227.
Anti
play
they
outs
canc
read
an i
canc
diffe
shou
cons
stab
disp
CNT
poly
tum
deri
one,
Conj
The pos
ibodies, or im
y a crucial ro
y have rema
standing ant
cer. Neverth
dily cross cel
intracellular
cer therapy,
erent approa
uld specifica
With the
structs. To d
bility to the
placement by
Ts using diffe
The ant
ymorphic ep
ours.1 The t
ving from a
, in order to
jugat
sibility to co
mmunoglobu
ole in the rec
rkable select
tineoplastic
heless, to dat
ll membrane
environmen
and paving
ach, Abs can
lly direct the
e aim of stu
do so, we d
final constru
y other biolo
rent covalen
Figure 1. A
tibody hCTM
pithelial muc
term “human
non‐human
avoid recogn
ion o
ombine the t
ulins, are a f
cognition and
tivity and pi
agent, since
te, they hav
es. The conju
nt, addressin
the way to
n also serve a
e system and
udying these
decided to u
ucts, avoidin
ogical macro
nt approache
A typical IgG
M01 IgG is
cins (PEM),
nized” is use
n immunoglo
nition of the
of ant
technologies
family of pro
d neutralizat
comolar pot
e they can ta
ve not been
ugation of an
ng one of t
a variety of
as targeting
d thus they w
different o
use a coval
ng an in vivo
omolecules.
es.
(lysine resid
s a human
the overexp
ed to descri
obulin, and
e antibody fro
tibod
s of CNTs an
oteins produ
tion of foreig
tency, and th
arget key re
used for int
ntibodies to
he importan
unexplored
agents. If w
will enable a t
pportunities
ent approac
o dissociatio
As a model,
dues are high
ized monoc
pression of
be an antibo
the remaini
om the body
ies to
nd antibodie
ced by the i
gn objects in
his quality co
egulators in
racellular ta
CNTs should
nt limitation
therapeutic
we conjugate
targeted dru
we prepare
ch, that sho
on of the Ab
, hCTM01 Ab
hlighted in re
clonal antib
which is as
ody with an
ing part der
y as “foreigne
o CNT
es (Abs) was
mmune syst
n the body. T
ould render
the develop
rgets as the
d enable the
ns of antibod
c possibilities
e them to CN
ug delivery.
ed different
ould guarant
b due to its
b was conju
ed).
body that b
ssociated wi
antigen‐bin
riving from a
er”.
73
Ts
studied.
tem, that
To do so,
them an
pment of
ey do not
eir use in
dy‐based
s. With a
NTs, they
CNTs‐Ab
tee good
possible
ugated to
binds to
th many
ding site
a human
Chapter
74
3.1 IgG‐
3.1.1 Pre
T
mainly a
reaction
solubiliz
azometh
shortene
diamete
TEM let
of 400 ±
Figu
correspo
carboxy
Ab. As i
carboxy
purificat
lysine sid
3
‐DWCNT
eparation
The first syn
at the tips o
n to tether t
zing chain v
hine ylides o
ed and oxid
er of 3.5 nm,
us evaluate
± 368 nm.
re 2. Length
In the firs
ondence of t
lic functions
illustrated in
lic groups o
tion of this
de chains of
nthetic strate
of oxidized C
the antibody
ia amidation
on the CNT s
dized and th
, according t
e the length
distribution
indica
t conjugatio
the sidewall
of s‐DWCNT
n Scheme 1,
n the s‐DWC
construct fr
the Ab, with
Schem
egy exploite
CNTs. An alt
y on amino‐f
n of the ca
sidewall. We
hus bearing
to the suppl
distribution
of s‐DWCNT
ates the aver
on strategy
defect sites
T and the am
, a two‐step
CNT to their
om any exce
hout using ad
me 1. (Ab and
ed an amidat
ternative ap
functionalize
rboxylic gro
e used DWCN
carboxylic a
ier, and con
of the samp
T analysed b
rage length o
y, the Ab
of CNTs, per
mine groups
p amidation
r correspond
ess reagent,
dditional cou
d CNT are no
tion process
proach was
ed CNTs, foll
oups or via
NTs, provide
acids (s‐DW
nfirmed by o
ple (Figure 2
y TEM (coun
of the sampl
was introd
rforming an a
of the lysine
was perform
ding NHS‐es
, we made i
upling agents
ot in size sca
s to immobil
based on a
lowing the i
1,3‐dipolar
ed by the su
CNT). They
our TEM ana
), whose ave
nt = 127). The
e.
duced at th
amidic coupl
e side chains
med. First, w
ter, and sub
it statisticall
s, to obtain s
le)
lize the anti
a chemosele
ntroduction
cycloadditio
pplier as alr
had an ave
lyses. Moreo
erage length
e vertical line
he tips an
ling between
s available on
we activated
bsequently,
y react with
s‐DWCNT‐1.
body
ctive
of a
on of
eady
erage
over,
h was
e
d in
n the
n the
d the
after
h the
conj
dens
coup
betw
this
afte
s‐DW
Thes
bear
was
to f
invo
trea
lysin
dete
sulfh
lysin
This stra
jugation of
sity of both c
As an al
pling of the
ween a solub
case, we fir
r cleavage o
WCNT‐2. The
se amines w
ring a malei
evaluated b
The mal
form stable
olved in disu
t the Ab wi
ne residues i
ermined by
hydryl group
ne amines (m
ategy was re
Ab would h
carboxyl and
ternative to
Ab in the s
bilizing chain
rst modified
of the Boc pr
e amount of
were subseq
mide group,
by the differe
Sc
eimide grou
thioether b
ulphide bond
th 2‐iminoth
ntroducing s
Ellman’s ass
ps per Ab, a
more or less 2
equired to av
have been p
d amine grou
the direct c
same positio
, which serv
the carboxy
otecting gro
f amine gro
quently mod
, to give s‐D
ence in the K
cheme 2. (Ab
p allows the
bonds, but a
ds to form it
hiolane (Tra
sulfhydryl gro
say.4 In a ty
ratio that is
20 in a IgG) c
void any und
possible in a
ups on the pr
conjugation t
on (i.e. main
ved also as a
ylic groups b
oup with HCl
ups was det
ified by rea
DWCNT‐3 (Sc
Kaiser test to
b and CNT ar
e covalent co
all the cyste
ts quaternar
ut’s reagent
oups (Schem
ypical exper
s considered
could advers
Scheme
C
desirable side
a one‐step a
rotein.
to the carbo
nly the tips o
spacer betw
by an amidat
(4 M) in dio
termined by
ction with a
cheme 2), w
be 60 μmol/
re not in size
onjugation of
eine residue
ry structure
t),3 which re
me 3), the am
iment we in
d reasonable
sely affect Ab
3
onjugation o
e reactions,
amidation, b
oxylic groups
of the CNTs
ween the nan
tion step wit
oxane, ammo
y Kaiser test
an NHS‐activ
hose degree
/g.
e scale)
f sulfhydryl‐c
es within th
. Therefore,
eact with the
mount of whi
ntroduced a
e, since react
b function.
of antibodies
since interm
because of
s we studied
), but introd
notube and t
th amine 1,
onium‐funct
t2 to be 120
vated propio
e of function
containing m
e IgG seque
it was nece
e primary a
ich was subs
n average o
cting all the
s to CNTs
75
molecular
the high
also the
ducing in
he Ab. In
yielding,
ionalized
μmol/g.
onic acid
nalization
molecules
ence are
essary to
mines of
equently
of 5 free
available
Chapter
76
T
(Scheme
at highe
through
perform
azometh
cleavage
group a
maleimi
bearing
s‐DWCN
T
spectros
high spe
superna
280 nm
superna
Followin
tubes w
fresh PB
3
The SH‐bea
e 2). Maleim
er pH the ma
Traut’s reag
med at pH 6.6
With a third
hine ylides w
e of the Boc
at the end
de group, as
antibody pre
NT‐7.
The three d
scopy. At ce
eed (16000
tant) from C
(due to the
tant of the
ng the coupli
as eventuall
BS.
aring antibo
mide reaction
aleimide gro
gent could u
6.
d synthetic
was performe
protecting g
of the triet
s we did for
epared as pr
Schem
different cou
ertain time in
g) to sepa
CNTs and Ab
e side chain
solution de
ing, the rem
y eliminated
ody was the
with sulfhy
oup could hy
undergo cycl
approach, il
ed on s‐DWC
group (amin
hylene glyco
s‐DWCNT‐3,
reviously des
me 4. (Ab and
upling reacti
ntervals, an
arate the so
‐CNT conjug
s of the aro
ecreased as
aining free A
d by cycles of
en conjugat
dryl groups
ydrolyse. Mo
lization at p
lustrated in
CNT, using a
ne content b
ol chain wa
, obtaining a
scribed was t
d CNT are no
ons betwee
aliquot of t
olution cont
gates precipi
omatic amin
the antibo
Ab or the Ab
f centrifugat
ted to s‐DW
should be pe
oreover, the
H ≥ 7.5 Ther
Scheme 4,
minoacid 3 a
y Kaiser test
as further d
a functionaliz
then conjuga
ot in size sca
n CNTs and
the reaction
taining the
tating to the
no acids pres
ody bound t
b simply adso
tion and rem
WCNT‐3, giv
erformed at
chain introd
refore coupl
1,3‐dipolar
and paraform
t was 80 μm
erivatized t
zation of 30
ated to s‐DW
le)
Ab were m
mixture wa
unreacted a
e bottom. Th
sent in the
to the nano
orbed on the
oval of the s
ving s‐DWC
pH 6.5‐7.5 s
duced on th
ing reaction
cycloadditio
maldehyde. A
mol/g), the a
to introduce
μmol/g. The
WCNT‐6, to af
monitored by
as centrifuge
antibody (in
he absorban
antibody) in
otube (Figure
e sidewalls o
supernatant
NT‐4
since
e Ab
was
on of
After
mine
e the
e SH‐
fford
y UV
ed at
n the
ce at
n the
e 3).
f the
with
3.1.2
com
stru
(Figu
Fig
abse
elec
thro
Figure 3. U
s‐DWCN
2 Characteri
To prov
mplementary
cture of the
ure 4).
gure 4. Repr
Gel elec
ence of any
ctrophoretic
ough the gel,
UV‐Vis spect
NT‐3 and the
ization
ve the succe
techniques.
e nanotubes
esentative T
ctrophoresis
unbound A
apparatus, t
and only un
tra of the sup
Ab, showing
essful immo
TEM image
was not aff
TEM images o
DWCNT‐
analysis u
Ab. In fact, a
the nanotube
nbound mate
pernatant fo
g the decrea
bilization of
s of the Ab‐C
fected by th
of pristine DW
‐4 (d) and s‐D
nder non‐re
as Figure 5
es are stayin
erial is movin
C
ollowing the c
se of the abs
f the antibo
CNT conjuga
he conditions
WCNTs (a), s
DWCNT‐7 (e)
educing con
shows, whe
ng in the load
ng.
onjugation o
coupling rea
sorption pea
dy on CNTs
tes show tha
s used for th
s‐DWCNT (b)
).
ditions was
en loading a
ding well, sin
of antibodies
ction betwe
ak at 280 nm
s, we used
at the morp
he coupling
), s‐DWCNT‐
s used to a
a CNT samp
nce they can
s to CNTs
77
en
.
different
hological
reaction
1 (c), s‐
assess to
le in the
nnot pass
Chapter
78
Figure
the sam
CNT usi
molecul
the effec
Figu
DWCNT
of the A
DWCNT
3
e 5. Electrop
For derivativ
me did not ap
ng a memb
ar weight is
ctiveness of
re 6. Electro
‐4 (lane 2), s
A possible e
Ab to naked
previously f
horetic appa
ves s‐DWCNT
pply to s‐DW
brane with a
of 150 kDa)
this step (lan
ophoretic gel
s‐DWCNT‐7 (
explanation f
s‐DWCNT, w
functionalize
aratus used f
T‐4 and 7 we
WCNT‐1 (Figu
a 300 kDa c
adsorbed on
ne 5 vs. lane
s under non‐
lane 3) and s
for this even
while in the
ed with a cha
for the analy
wells.
e did not ob
ure 6). There
cutoff, in ord
n CNT surfac
e 4).
‐reducing or
s‐DWCNT‐1 b
nt is that s‐D
case of s‐D
ain. As a con
ysis with an e
serve any un
efore we per
der to remo
ce, and the e
r reducing co
before (lane
WCNT‐1 der
DWCNT‐4 an
nsequence, it
enlargement
nbound prot
rformed a di
ove any resi
electrophoret
nditions of A
4) and after
rived from th
d 7 the Ab w
t was probab
t of the loadi
teic material
ialysis of the
idual Ab (w
tic gel confir
Ab (lane 1), s
r (lane 5) dia
he direct bin
was bound
bly easier fo
ng
, but
e Ab‐
hose
rmed
s‐
lysis.
nding
to s‐
r the
Ab t
by t
port
redu
the
to th
Ab‐C
DWC
the
func
nano
of p
prec
com
to adsorb on
he functiona
Furtherm
tions of the
uced. As exp
light chains
he CNT, but t
Thermog
CNT conjuga
CNTs were s
oxidized gro
ctionalized s
otube could
proteic mate
cursors at th
mplete (Figure
n CNT surfac
alization, and
more, electr
Ab, since in
pected, a pat
(25 kDa) of
they also sug
gravimetric a
ates. Figure
table up to 8
oups introduc
‐DWCNT‐2,
burn. For th
rial bound t
he temperat
e 7).
F
e in the first
d therefore n
rophoresis
n this case th
ttern of two
the Ab. Thes
ggest that it
analysis was
7 shows t
800°C, while
ced on their
3, 5 and 6,
he final Ab‐C
to CNTs, by
ure at which
Figure 7. TGA
t case, while
no longer ava
under redu
he disulphur
o bands appe
se results no
was intact a
s used to eva
the weight
e s‐DWCNT s
structure. T
in which als
CNTs conjug
comparison
h, according
A (N2) of DW
C
e in the seco
ailable to the
cing condit
ric bonds be
eared, due t
ot only confi
fter the conj
aluate the d
loss under
howed an in
This thermal
so the molec
ates TGA res
with the th
g to the first
WCNTs deriva
onjugation o
nd case the
e Ab.
ions can be
etween the c
o the heavy
rm that the
jugation.
egree of fun
inert atmos
ncrease in th
weight loss f
cules covale
sults let us e
hermogravim
derivative,
tives.
of antibodies
surface was
e used to
chains of th
chains (50 k
Ab was real
nctionalizatio
sphere. The
e weight los
further incre
ently attache
estimate the
metric profile
the weight
s to CNTs
79
s covered
visualize
e Ab are
kDa) and
ly bound
on of the
pristine
s, due to
eased for
ed to the
e amount
e of their
loss was
Chapter 3
80
We could calculate a loading of 24% of Ab with respect to the whole conjugate for s‐
DWCNT‐1 (in terms of weight). Similarly, loading values of 25% and 28% were obtained for s‐
DWCNT‐4 and 7, respectively. These values are in good agreement with the decrease of Ab
concentration in the supernatants of the coupling reactions (Figure 3), considering that part of
the antibody disappearing from the supernatant was simply adsorbed on CNT and therefore it
was washed off after the reaction. The loading calculated by TGA, if expressed in molarity,
correspond to concentrations of 2.1, 2.2 and 2.6 µmol of Ab per gram of CNT for the three
conjugates respectively, meaning that only a small fraction (less than 10%) of the maleimide
groups reacted with the antibody, probably due to steric hindrance.
Finally, the functional activity of the Ab coupled to carbon nanotubes was studied. This is
the first fundamental step to prove that the Ab is still able to exert its activity. In this context, we
assessed by surface plasmon resonance (SPR) the recognition capability of the CNT‐coupled
antibody towards its antigen.6 This technique allows the measurement of the specific
interactions between the antibody and its antigen in real time. For this purpose, the antigen is
immobilized on a sensor chip surface while the antibody is allowed to continuously flow over it,
in order to let it bind to the surface if the recognition takes place. The interaction is then
detected as a change in the refractive index at the interface between the chip surface and the
aqueous media above it, and it is presented as a sensorgram. The registered sensorgram offers
the possibility to calculate the kinetic parameters of the interaction (i.e. the association and
dissociation rate constants). First of all, the antigen sequence 296HGVTSAPDTRPAPGSTAPPA315,
which belongs to the antibody‐recognized domain of PEM,7 was synthesized, together with a
control peptide (called “scrambled” antigen, APHADPSTPGAPSVTPRTAG) with the same number
and type of amino acids, but placed at different positions in the sequence. This control enabled
us to evaluate the non‐specific component of the binding between the antibody and the antigen.
Each sensorgram was obtained measuring simultaneously (in two different channels) the
response to the analyte with the real antigen and with the scrambled one, in order to subtract
the non‐specific component of the binding. Indeed, the antibody recognized only the antigen but
not the “scrambled” antigen, thus confirming its high selectivity. We also performed control
experiments using s‐DWCNT‐2 to evaluate the possibility of non‐specific binding between the
antigen and the nanotubes and we found a response close to zero (data not shown). Finally, we
assessed the interaction of the three Ab‐CNT constructs obtained. All conjugates were able to
recognize the antigen on the sensor chip, giving a positive response (Figure 8).
qua
weig
and
conj
bind
atta
feas
3.2
the
to m
orde
to g
bett
to p
intro
func
we
Figure
Howeve
ntitative bin
ght for CNTs
dissociation
jugates are l
ding capabil
chment of
sibility of the
Fab’‐MWC
Exploitin
effort to pre
move to a di
er to assess
get an impro
ter dispersio
prepare a d
oducing two
ctionalization
choose to
e 8. SPR anal
er, it is worth
nding inform
, being an he
n rate const
ikely distribu
ity. Neverth
Ab on CNT
e system.
NT and scFv
ng the exper
epare covale
ifferent kind
the reprodu
ovement in
ns in both or
oubly funct
o chains bea
n strategies
o place it
lyses of hCTM
h pointing ou
mation from
eterogeneou
tants could
uted with dif
heless, these
Ts does not
v‐MWCNT
rience gaine
ent CNTs‐Ab
of CNTs, i.e
cibility of ou
the solubilit
rganic solven
ionalized de
aring termin
studied so f
on the ch
M01 IgG (a) a
ut that it is n
the SPR, fo
us material, c
not be obt
fferent orien
e results hig
t affect the
d in the con
conjugate. C
e. MWCNTs,
ur functional
ty of the de
nts and wate
erivative, us
nal amines o
far for the co
hain introd
C
and s‐DWCN
ot possible,
r two main
cannot be de
ained from
ntations, and
ghlight very
ir recognitio
njugation of
Considering
with a com
lization strat
erivatives. M
er, with respe
ing both am
orthogonally
oupling of th
duced throu
onjugation o
NT‐1 (b), 4 (c)
with an Ab‐C
reasons: 1)
etermined an
the sensorg
this fact ma
importantly
on capability
IgG to DWC
our previou
mpletely diffe
tegy with dif
MWCNTs are
ect to DWCN
midation and
y protected.
he Ab prove
ugh the cy
of antibodies
) and 7 (d).
CNT constru
a precise m
nd hence ass
grams; 2) A
ay affect the
y how the
y, thus pro
CNTs, we we
s results we
erent aspect
fferent mate
in fact able
NTs. Also, we
d cycloaddit
Since the
ed all to be e
ycloaddition,
s to CNTs
81
ct, to get
molecular
sociation
Ab in the
ir overall
covalent
ving the
ent on in
e decided
ratio, in
erials and
e to give
e decided
tion, and
different
effective,
, whose
Chapter
82
function
to leave
biologica
the anti
specifica
the conj
3.2.1 Pre
sulphuri
CNTs, a
shorteni
their str
shortene
the cutt
Figure
length d
3
nalization in t
e the chain in
al studies. Fu
gen‐binding
ally recogniz
jugation with
Figu
eparation
First of all,
ic and nitric
s we alread
ing the nano
ructure. The
ed MWCNTs
ing did not a
e 10. a) Repre
distributions
terms of ma
ntroduced w
urthermore,
fragment (F
ze the antige
h the CNTs, f
re 9. Schema
MWCNTs un
acid and son
dy explained
otubes, imp
ese carboxyl
s (s‐MWCNT
affect their st
esentative T
of pristine M
in
aleimide grou
with amidatio
we moved f
Fab’) and th
en (Figure 9)
for reasons t
atic represen
nderwent an
nicating them
in Chapter
roving their
ic groups ar
T) were analy
tructure (Fig
EM images o
MWCNTs (cou
ndicate the a
ups was by f
on for the a
from the wh
he single‐cha
), which in p
hat will be d
ntation of a f
n oxidative c
m for 24 h in
r 2.1 in rega
solubility, a
re necessary
ysed by TEM
gure 10a).
of pristine M
unt = 125) an
average leng
far high enou
ttachment o
ole Ab to sm
ain variable f
principle sho
discussed in d
full IgG, a Fa
cutting step,
a sonic bath
ard to SWCN
and of intro
y for the fo
M, to check
MWCNTs (left
nd s‐MWCNT
gth values.
ugh for the p
of a probe, n
maller fragme
fragment (sc
uld offer a b
details later.
b’ and a scFv
using a mix
h. An oxidati
NTs, has the
ducing carbo
llowing amid
that the har
) and s‐MWC
T (count = 15
purpose, in o
necessary fo
ents of it, na
cFv), still ab
better contr
v.
xture of aqu
ive treatmen
e double aim
oxylic group
dation step.
rsh condition
CNTs (right);
56). Vertical
order
r the
mely
le to
rol of
eous
nt for
m of
ps on
The
ns of
; b)
lines
distr
occu
one
Inde
268
app
two
sele
isoth
Sche
The ave
ribution of p
urred upon t
Moreove
of pristine
eed, the MW
nm ± 174 nm
s‐MWCN
roaches, nam
chains with
After th
ctively depr
hiocyanate (
emes 6 and 7
rage diamet
pristine and
he oxidation
er TEM let u
MWCNTs, i
WCNTs that w
m after oxida
NT were fu
mely amidat
orthogonall
ese steps, th
rotected wit
(FITC) or wi
7.
ter of pristin
shortened
n, and indeed
s study the l
n order to v
we used had a
ation.
nctionalized
tion and 1,3‐
y protected
he phthalim
th hydrazine
th diethylen
ne MWCNTs
MWCNTs to
d it remained
length distrib
verify the a
an average l
d in two co
‐dipolar cycl
amino group
Scheme
ide‐protecte
e, in order
netriaminepe
Scheme
C
being 34 ±
o further con
d unchanged
bution of s‐M
ctual shorte
ength of 428
onsecutive
loaddition o
ps (Scheme 5
5
ed amine int
to make i
entaacetic d
6
onjugation o
10 nm, we
nfirm that n
d (data not sh
MWCNT, and
ening of the
8 nm ± 354 n
steps, explo
f azomethin
5).
roduced thr
t react eith
dianhydride
of antibodies
compared d
no extensive
hown).
d compare it
e tubes (Figu
nm as pristin
oiting two
ne ylides, int
rough amida
her with flu
(DTPA) as s
s to CNTs
83
diameter
damage
with the
ure 10b).
e, and of
different
roducing
tion was
uorescein
shown in
Chapter
84
fluoroph
which w
the seco
with a m
allow th
made of
Kaiser te
degree o
170 µm
while th
T
spectra
intensity
3
Fluorescein
hore, will all
will be used t
ond amine w
maleimide gr
e coupling w
In this way,
f s‐MWCNT‐
est values fo
of functiona
ol/g, where
e DTPA load
Fig
The effectiv
of s‐MWCNT
y.
and DTPA a
ow to ident
to trap a rad
was deprote
roup (Schem
with free thio
two distinct
2A, 3A and 4
or the two se
alization obta
as cycloadd
ding was 50 µ
gure 11. Kais
veness of FI
T‐2A (Figure
S
re both need
tify CNTs in
dionuclide fo
ected from t
mes 6 and 7),
ols in the Ab
t series of co
4A, and the
eries of comp
ained in the
ition gave 3
µmol/g. The
ser test trend
TC coupling
12), though
Scheme 7
ded for biolo
in vitro test
or in vivo lo
the Boc grou
, as we did f
fragments.
ompounds w
DTPA series
pounds, as d
e different st
35 µmol/g. T
maleimide f
d for the two
g was also
quenching b
ogical assays
s, while the
calization of
up and both
for preparing
were basical
s, made of s‐
depicted in F
teps. The on
The fluoresc
unctionalizat
o series of co
confirmed b
by CNTs stro
s: the first m
second is a
f the constru
h MWCNTs w
g Ab‐DWCNT
ly prepared:
‐MWCNT‐2B
Figure 11, let
ne relative to
ein loading
tion was 35 µ
ompounds.
by absorptio
ongly decreas
molecule, be
a chelating a
ucts. Afterw
were deriva
Ts conjugate
: the FITC se
B, 3B and 4B
t us calculate
o amidation
was 60 µm
µmol/g.
on and emis
sed fluoresc
ing a
agent
ards,
tized
es, to
eries,
. The
e the
was
mol/g,
ssion
ence
F
f
conj
with
mol
whe
as d
free
glut
cyst
redu
conj
befo
the
exist
Ab f
high
with
Figure 12. a)
fluorescence
Once ob
jugation wit
hout further
ecule. In the
ereas on the
dimers, due t
e thiols in tw
athione, der
eine thiol. T
uced in orde
As alrea
jugation with
ore, and to d
exact positio
ting thiol wa
fragments, t
her and a po
h CNTs. With
Absorption
e spectra of F
btained the
h thiol grou
r modificatio
e Fab’ the th
scFv it was i
to the spont
wo different m
riving from t
Therefore, p
r to restore t
dy anticipate
h the CNTs.
do so some
on of the thi
as exploited,
here was on
ossible undes
Fab’ and scF
spectra of FI
FITC and s‐M
recorded u
maleimide‐f
ps. Fab’ and
on of the pr
hiol derives
introduced o
aneous form
monomers. M
their produc
prior to con
the free thio
ed, the use o
With the ful
of the amin
iols was unk
thus knowi
nly one thiol
sired conseq
Fv we did no
ITC and s‐MW
MWCNT‐2A (1
upon excitat
functionalize
d scFv offere
roteins, sinc
from a disu
on purpose. A
mation, upon
Moreover, F
ction, and so
njugation wi
ols.
of these two
ll IgG we had
es were mo
nown. Instea
ng precisely
per molecu
quence were
ot have this r
Scheme
C
WCNT‐2A (10
10‐6 M in DM
tion at 488 n
ed CNTs 4A
ed us the op
ce they both
lphide bond
As a consequ
n storage, of
ab’ and scFv
o they could
ith CNTs, th
fragments s
d to introdu
dified using
ad, with the
its position.
ule, while for
cross‐coupl
risk.
8
onjugation o
0‐6 M in DMS
MSO/TEA 0.1
m.
and 4B, we
pportunity to
h present al
d already pre
uence, Fab’ a
a disulphide
v preparation
form a disu
he disulphid
should offer a
ce thiols tha
a statistic a
use of Fab’
. Furthermor
r the thiolate
ing reaction
of antibodies
SO/TEA 0.1 M
M). Spectra
e were ready
o use this c
lready one t
esent in the
and scFv exis
e bond betw
ns contain a
ulphide bond
de bonds ha
an advantag
at where not
pproach. In
and scFv, an
re, in the ca
ed IgG the r
ns during con
s to CNTs
85
M); b)
were
y for the
hemistry
thiol per
full IgG,
st in part
ween two
lso some
d with its
ad to be
ge for the
t present
this way
n already
se of the
ratio was
njugation
Chapter
86
T
cysteam
monitor
the dim
efficacy
(MW = 3
both for
KDa (th
cleavage
Figure
simply m
room te
was cen
(absorpt
this way
the reac
centrifug
7.4) usin
unbound
construc
or scFv.
3
The reductio
mine (Schem
red the react
er from the
of the reduc
30 KDa), and
r Fab’, where
e presence
e of heavy an
13. Electrop
(lane 2)
As we alread
mixing the t
emperature
ntrifuged, to
tion peak at
y we could f
ction, and w
gation and r
ng a membra
d proteic m
cts depicted
on step was
e 8), follow
tion via elect
monomer,
ction. In the
d, more impo
e it was at m
of some m
nd light chain
phoretic gels
reduction an
dy did for th
two compon
(Scheme 9).
precipitate
280 nm), co
ollow the co
e thoroughly
removal of s
ane with a c
material from
in Scheme 9
s performed,
wed by a pu
trophoresis (
according to
gel we can s
ortantly, we c
ore or less 1
minor bands
ns during the
under non‐r
nd of scFv be
he DWCNTs a
nents in PBS
At certain t
CNTs, and t
orresponding
oupling react
y washed th
upernatant.
ut‐off of 300
m the MWC
9, each one
, for both F
urification w
(Figure 13). W
o their differ
see the Fab’
can apprecia
100 KDa, and
at smaller
e heating ste
reducing con
efore (lane 3)
and the IgG,
S‐EDTA (pH
time points,
he concentr
g to the unr
tions. When
he crude wit
Afterwards
0 KDa, to ma
CNTs. In thi
bearing, alt
ab’ and scFv
ith size exc
With this tec
rent size, an
band (MW
ate the disap
for scFv, wh
MW is due
ep in the pre
nditions of Fa
) and after (l
, we perform
6.6), and ge
a small aliq
ration of Fab
reacted Ab f
the couplin
h fresh PBS
we perform
ake sure of t
s way we p
ernatively, f
v, through a
lusion chrom
chnique we c
d thus we c
= 48 KDa) an
ppearance of
here it was at
e to some s
paration of t
ab’ before (la
ane 4) reduc
med the coup
ently shaking
quot of the
b’ or scFv in
ragments, w
g was comp
(pH 7.4) by
ed a dialysis
the complete
prepared th
luorescein o
a treatment
matography.
could disting
could contro
nd the scFv
f the dimer b
t more or les
slight proteo
the gel).
ane 1) and af
ction.
pling reactio
g the mixtur
reaction mix
the superna
was measure
plete we stop
several cycle
s against PBS
e removal of
he four diffe
or DTPA and
with
. We
guish
l the
band
band,
ss 60
olytic
fter
on by
re at
xture
atant
ed. In
pped
es of
S (pH
f any
erent
Fab’
3.2.2
alre
conj
all th
cond
2 Characteri
For the
ady used fo
jugates obta
he synthetic
Figure 14
An ele
ditions (Figu
Schem
ization
characteriz
or the Ab‐DW
ained, and im
steps (Figur
. Representa
ctrophoretic
re 15). In th
me 9. (Fab’/
ation of the
WCNTs const
mages clearly
e 14).
ative TEM im
c analysis w
e first case w
scFv and CN
e final conj
tructs. TEM
y showed th
mages of s‐MW
was perform
we assessed
C
T are not in s
ugates we
was used t
hat CNT stru
WCNT‐5A (a
med both
the lack of
onjugation o
size scale)
exploited th
o assess the
cture was pr
), 5B (b), 6A
in reducing
unbound pro
of antibodies
he same te
e quality of
reserved thr
(c) and 6B (d
g and non‐
oteic materi
s to CNTs
87
chniques
the final
roughout
d).
‐reducing
al, since,
Chapter
88
when ru
could m
could ob
did not
related
located
in fact t
reductio
the two
interacti
coupling
through
reducing
Figure
this case
that the
lack of p
bound t
and scF
coupling
with ele
electrop
conditio
3
unning the g
ove. On the
bserve the p
see anything
to the light
in the heavy
the contribu
on step of th
o chains was
ions and th
g different t
the thiol o
g electropho
15. Electrop
(lane 2
A very impo
e, to confirm
e washing tre
proteins in t
to CNTs, but
v and then
g reactions.
ectrophoresi
phoretic gel,
ons.
gel, the MWC
other hand,
pattern of pr
g. The prese
and heavy
y chain, we s
ution from t
he Fab’ dime
s reduced. N
hey could su
thiols. There
on the light
oretic gel.
horetic gels
), s‐MWCNT
ortant piece
m that the b
eatments w
he non‐redu
not that th
we washed
The produc
s. As clearly
, for none
CNTs stayed
in reducing
oteic fragme
nce of two b
chains of th
hould expec
this band is
er with cyste
Nevertheless
ubsequently
efore it is po
chain, there
under non‐r
‐5A (lane 3),
of informati
bond betwee
ere actually
ucing gels of
e bond is co
d them follo
ts obtained
y shown in F
of the two
d in the load
conditions, t
ents present
bands in the
he Fab’. Sinc
ct to observe
stronger. H
eamine, also
s, they rem
form dime
ossible that
efore leaving
reducing or r
, 5B (lane 4),
ion was give
en the Ab fr
able to rem
f the final co
ovalent. Ther
owing exactl
with these
Figure 16, n
o controls,
ding well and
the disulphid
t in Fab’, wh
e reducing ge
ce the Fab’ w
e only the lig
However, it
a part of th
ained coupl
rs again bu
part of the
g the heavy
reducing con
6A (lane 5)
en by a nega
agments and
move any ad
onjugates pro
refore we tr
y the same
control rea
no proteic m
both under
d only the u
de bonds are
ile of course
el of s‐MWC
was bound
ht chain ban
is possible
e disulphide
ed because
t in a rearr
Fab’ finally
y chain free
ditions of Fa
and 6B (lane
ative control
d the CNT w
sorbed prot
ove that the
eated s‐MW
protocol u
ctions were
material was
r non‐reduc
nbound mat
e broken, an
e for the scF
NT‐ 5A and
through its
nd in the gel,
that, during
e bonds betw
of hydroph
ranged way
y bound to C
to move in
ab’ (lane 1), s
e 6).
l we prepare
was covalent
teins. In fact
e Ab are stro
WCNT‐1 with
sed for the
finally anal
detected in
cing or redu
terial
d we
v we
5B is
thiol
, and
g the
ween
hobic
, i.e.
CNTs
n the
scFv
ed in
t and
, the
ongly
Fab’
real
lysed
n the
ucing
well
disp
bind
prot
strat
esta
Figure 16. E
The dete
l by a calib
persions used
ding of Fab’
tein in the c
tegy as cal
ablishment o
Electrophore
reactions be
ection limit o
bration perf
d and the vo
or scFv to s
conjugate), o
culated from
f the desired
Fi
tic gels unde
etween s‐MW
of the electro
ormed with
olume loade
s‐MWCNT, i
one order of
m TGA valu
d covalent bo
igure 17. TGA
er non‐reduc
WCNT‐1 and
ophoretic ge
h a standar
ed for each w
if any, was a
f magnitude
ues (Table
ond between
A (N2) of MW
C
cing or reduc
Fab’ (lane 1)
el was estima
d. Consider
well, this va
at most of 0
less than th
1). Thus, w
n CNTs and t
WCNTs deriva
onjugation o
cing conditio
) or scFv (lan
ated to be 0
ing the con
lue implies t
0.8% (expres
he loading o
we indirectly
he Ab fragm
atives.
of antibodies
ns of the con
ne 2).
.12 µg of pro
ncentration
that the non
ssed as weig
obtained by
y proved th
ments.
s to CNTs
89
ntrol
otein per
of CNTs
n‐specific
ght % of
covalent
e actual
Chapter 3
90
The degree of loading related to Fab’ and scFv was evaluated using TGA (Figure 17). The
increase in the weight loss for the final conjugates relative to their precursors (at the
temperature at which, according to the first derivative, the weight loss was complete)
corresponds to the Fab’ or the scFv bound to the MWCNTs.
Loading of Fab’ or scFv s‐MWCNT‐5A s‐MWCNT‐5B s‐MWCNT‐6A s‐MWCNT‐6B
Weight % in the conjugate 6.8 6.3 9 8.4
µmol/g of CNT precursor 1.5 1.4 3.3 3.1
Table 1. Loading values relative to Fab’ and scFv for s‐MWCNT‐5A, 5B, 6A and 6B.
Thus, it was possible to calculate how much protein was attached on the nanotubes
(Table 1). These values are in good agreement with the decrease of Fab’ or scFv concentration in
the supernatants of the coupling reactions, considering that part of the Ab fragments, especially
scFv, was simply adsorbed on CNT and therefore it was washed off after the reaction. Moreover,
they correlate well also with the relative intensities of electrophoretic spots, considering the
amount of material loaded for each well (Figure 15).
As for the previous conjugates, the most important characterization of the material was
SPR. All the compounds gave a positive response (Figure 18), even if with lower intensity than
the response obtained with the Ab‐DWCNTs constructs. This is probably due to the lower degree
of functionalization obtained for these conjugates, and to the lower affinity for the antigen
possessed by Fab’ and scFv, with respect to the full IgG.
In conclusion, we prepared different kinds of covalent Ab‐CNT conjugates, using
DWCNTs and MWCNTs and either the whole IgG or smaller fragments, i.e. Fab’ and scFv. In the
case of MWCNTs we used a double functionalization strategy in order to introduce on the CNTs
also a probe, necessary for biological studies. We proved the effectiveness of the covalent
strategy by means of different complementary techniques, and, importantly, we assessed the
capability of all the constructs to specifically recognize the antigen.
Figu
3.2.3
CNT
the
cont
cons
com
3B,
injec
h th
wer
dilut
ure 18. SPR a
3 Preliminar
We stud
Ts, that let us
DTPA‐MWC
taining buffe
structs, we
mputed tomo
we could a
cted, as in th
he animals w
e sampled, e
tion of the in
analyses of h
ry biological
died the in vi
s couple them
NTs with 111
er. In this w
could follow
ography/com
lso assess q
he previous e
were killed, a
each sample
njected dose
hCTM01 Fab’
results
ivo behaviou
m with a rad
InCl3 for 30
way, by inje
w their in v
mputed tomo
uantitatively
experiments
nd blood wa
e being weig
e, in order to
’ (a), scFv (b)
ur of our con
dionuclide, 11
min at r.t. a
ecting BALB/
vivo fate thr
ography) sca
y the organ
s, via the tail
as collected.
hted and co
o calculate t
C
), s‐MWCNT‐
njugates exp
11In. The cou
and then wa
/c mice via
rough a SPE
anner (Figure
biodistribut
vein with th
. The differe
unted on a G
he percenta
onjugation o
‐5A (c), 5B (d
loiting the D
pling was ma
ashing unbou
the tail vein
ECT/CT (sing
es 19, 20 an
tion (Figure
he [111In]DTPA
nt organs an
Gamma Cou
ge of injecte
of antibodies
d), 6A (e) and
DTPA attache
ade by simp
und 111In wit
n with the
gle photon
d 21). Only
22). The m
A‐MWCNTs.
nd tissues of
unter togethe
ed dose per
s to CNTs
91
d 6B (d).
ed on the
ly mixing
th EDTA‐
different
emission
for CNTs
ice were
After 24
f interest
er with a
organ or
Chapter
92
per gram
well.
Figure
Figure
Figure
3
m of tissue.
19. Whole B
20. Whole B
21. Whole B
Quantitative
ody 3D Imag
in rats a
ody 3D Imag
in rats a
ody 3D Imag
in rats a
e organ biod
ging (SPECT/C
after 30 min,
ging (SPECT/C
after 30 min,
ging (SPECT/C
after 30 min,
distribution w
CT) showing
4 h, and 24
CT) showing
4 h, and 24
CT) showing
4 h, and 24
will be evalu
distribution
h post‐inject
distribution
h post‐inject
distribution
h post‐inject
uated for CN
of [111In]DTP
tion.
of [111In]DTP
tion.
of [111In]DTP
tion.
NTs 5B and 6
PA‐MWNCTs
PA‐MWNCTs
PA‐MWNCTs
6B as
s 3B
s 5B
s 6B
Fig
or
early
22).
thes
and
accu
bear
agen
sele
fluo
7) an
with
fluo
5A)
Cy3
imag
cells
med
deri
inte
cond
ure 22. Orga
r per gram o
MWCNTs 3
The orga
y accumulat
Comparing
se observatio
6B, we cou
umulation an
A furthe
ring mouse m
nts for the w
ctive cancer
Also the
rescein‐tagg
nd a PEM‐ne
h Fab’ alone,
rescence giv
was localize
(hCy3). The
ges show (F
s, and for the
diated endo
vatives, ins
rnalization w
ditions (data
an biodistribu
of tissue (b) a
3B, quantifie
an biodistrib
ion in liver, s
the three c
ons need to
ld state that
nd reduced k
er important
model, to as
whole conjug
treatment.
e internaliza
ged ones) wa
egative (hum
, s‐MWCNT‐
ven by the flu
ed using a se
scFv is not r
igures 23a a
e three CNTs
cytosis, due
tead, also
was observe
a not shown)
ution expres
at 1 h, 4 h an
ed by gamma
bution for s‐
spleen and lu
onstructs th
be confirme
t Fab’ and sc
kidney elimin
step will be
sess whethe
gate, leading
ation and t
as assessed,
man lung carc
3A, 5A or 6A
uorescein, w
econdary ant
recognized b
and 24a), we
s constructs i
e to the rec
an energy‐
ed as well af
.
ssed as % of
d 24 h after
a counting (n
‐MWCNT‐3B
ung, with gra
hrough the w
ed by quantit
cFv conjugat
nation.
e the study o
er the Fab’ a
to accumula
the cellular
in both a PE
cinoma, Calu
A. The locali
while the Fab
ti‐human an
by hCy3, sinc
e observed c
in both cell l
cognition of
‐independen
fter incubat
C
injected dos
intravenous
n = 3, error b
(i.e. not be
adual cleara
whole body
tative organ
tion to CNTs
of our conju
nd the scFv w
ation in the t
trafficking
EM‐positive
u‐6) cell line.
zation of CN
b’ (both alon
tibody label
ce it is too sm
cellular upta
ines. Fab’ wa
PEM on ce
nt pathway
ion at 4°C,
onjugation o
e (ID) radioa
administrat
ars for stand
earing any A
nce from the
SPECT/CT im
biodistribut
increased t
gates in a P
would be ab
tumour and
of the com
(human bre
Cells were t
NTs was dete
e, and in the
led with the
mall. As the
ke for Fab’ o
as internalize
ell surface.
was proba
i.e. under e
of antibodies
activity per o
tion of [111In]
dard deviatio
Ab fragment)
e liver in 24
maging, even
tion for s‐MW
their liver an
EM‐positive
ble to act as t
therefore a
mpounds (u
east carcinom
reated, alter
ected with t
e Fab’‐CNT c
e red fluores
confocal mi
only in PEM
ed through r
In the case
ably involve
ndocytosis‐i
s to CNTs
93
organ (a)
]DTPA‐
on).
showed
h (Figure
n though
WCNT‐5B
nd spleen
tumour‐
targeting
llowing a
sing the
ma, MCF‐
rnatively,
he green
onjugate
cent dye
croscopy
M‐positive
receptor‐
e of CNT
ed, since
nhibiting
Chapter
94
Figu
µg
interfe
Cy3 (
Figu
µg
interfe
Cy3 (
media fo
fact the
the cell
lysosom
3
ure 23. Confo
g/mL), s‐MW
erence contr
red fluoresce
sam
ure 24. Confo
g/mL), s‐MW
erence contr
red fluoresce
sam
Furthermore
or 24 h befo
Fab’, accord
s, probably
mal degradat
ocal microsco
WCNT‐3A, 5A
rast (DIC) cha
ence) chann
me treatment
ocal microsco
WCNT‐3A, 5A
rast (DIC) cha
ence) chann
me treatment
e, it is worth
re the confo
ding to the f
because, a
tion. Differe
opy images o
or 6A (10 µg
annel, third l
el. Panel a sh
t followed by
opy images o
or 6A (10 µg
annel, third l
el. Panel a sh
t followed by
h noting that
ocal analysis,
fluorescence
fter the en
ntly, the CN
of MCF‐7 cel
g/mL). First l
line: FITC (gr
hows treatm
y incubation
of Calu‐6 cel
g/mL). First l
line: FITC (gr
hows treatm
y incubation
t when cells
, the scenari
e given by th
docytotic in
NT derivative
ls untreated
ine: nuclei, s
reen fluoresc
ment for 3 h a
with full me
ls untreated
ine: nuclei, s
reen fluoresc
ment for 3 h a
with full me
were treate
o was differe
e hCy3, was
nternalization
es could sti
or treated w
second line: d
cence) chann
at 37°C. Pane
edia for 24 h.
or treated w
second line: d
cence) chann
at 37°C. Pane
edia for 24 h.
ed and then
ent (Figures
not present
n, its intrac
ll be found
with Fab’ (0.7
differential
nel, fourth lin
el b shows th
.
with Fab’ (0.7
differential
nel, fourth lin
el b shows th
.
incubated in
23b and 24b
t anymore in
cellular fate
inside the
7
ne:
he
7
ne:
he
n full
b). In
nside
was
cells
Conjugation of antibodies to CNTs
95
according to the fluorescein green signal and, for s‐MWCNT‐5A, to the red fluorescence of hCy3
bound to the Fab’.
It is possible to conclude that the uptake and the intracellular trafficking of the Fab’‐CNT
and of the scFv‐CNT conjugates were driven, at least in part, by CNTs. In fact, all the constructs
were internalized even by PEM negative cells (while Fab’ alone was not) and the fate of Fab’
after 24 h was different if it was alone or bound to the CNTs. Considering these preliminary in
vitro results we could not state if these conjugates would be specifically targeted in vivo towards
PEM‐positive cells, but we can conclude that CNTs were able to deliver their Ab cargo inside
cells, following also an energy‐independent pathway, and therefore dramatically changing the
intracellular fate of the material. This important observation paves the way to the use of CNTs to
deliver antibodies inside cells, thus leading to a variety of unexplored therapeutic possibilities.
3.3 Acknowledgments
The work described in this Chapter was part of the FP7 European Project ANTICARB,
which involves different collaborations. I wish to thank the group of Dr. Alberto Bianco (CNRS,
Institut de Biologie Moléculaire et Cellulaire, Laboratoire d’Immunologie et Chimie
Thérapeutiques in Strasbourg, France) and in particular Enrica Venturelli for the syntheses of the
antigen and the “scrambled” antigen peptides, and for some of the electrophoretic gels, and Dr.
Olivier Chaloin for the SPR analyses.
For what concerns the biological data, I wish to thank the group of Prof. Kostas
Kostarelos (Nanomedicine Laboratory, Centre for Drug Delivery, The School of Pharmacy in
London, UK), and in particular Antonio Nunes and Dr. Khuloud Al‐Jamal for the in vivo studies,
and Dr. Chang Guo for the in vitro ones.
3.4 References
1. J. R. Adair, P. R. Hamann, R. J. Owens, T. S. Baker, A. H. Lyons, L. M. Hinman, and A. T.
Menendez. WO/1993/006231.
2. E. Kaiser, R. L. Colescott, C. D. Bossinger, and P. I. Cook, Anal. Biochem. 1970, 34, 595‐
598
3. R. R. Traut, A. Bollen, T. T. Sun, J. W. B. Hershey, J. Sundberg, and L. R. Pierce,
Biochemistry 1973, 12, 3266‐3273.
4. G. L. Ellman, Arch. Biochem. Biophys. 1959, 82, 70‐77.
5. R. Singh, L. Kats, W. A. Blattler, and J. M. Lambert, Anal. Biochem., 1996, 114‐125.
Chapter 3
96
6. a) R. Karlsson, J. Mol. Recognit., 2004, 17, 151‐161; b) M. A. Cooper, Nat. Rev., 2002, 1,
515‐528.
7. A. P. Spicer, G. Parry, S. Patton, and S. J. Gendler, J. Biol. Chem., 1991, 266, 15099‐15109.
97
Conjugation of doxorubicin to CNSs
Doxorubicin is an antineoplastic drug that exerts its activity mainly at the level of DNA,
by inhibition of topoisomerase II, an essential enzyme for DNA replication. Though being widely
used, its clinical utility is hindered by the onset of a severe cardiomyopathy, due to its lack of
selectivity, and by multi‐drug resistance (MDR). Both these limitations, and the way nanocarriers
could help in overcoming them, have been already analysed in the Introduction (see Chapter
1.5). Many studies have been published so far in which CNT‐based systems were used for the
delivery of doxorubicin, making it a promising tool (see Chapter 1.5.2 and in particular Table 1).1
Nevertheless, none of these works used a covalent approach for the conjugation of the
drug to CNTs. The major limitation of a non‐covalent construct is its lack of stability in vivo, as
already explained (see Chapter 1.2.1). In fact, most of the above papers that addressed this
question reported the undesired partial release of non‐covalently bound doxorubicin from the
carrier.1c,g,h For this reason we decided to prepare covalent doxorubicin‐CNT conjugates,
exploiting the primary amine present on the sugar of the drug. However, it is known from the
literature that substituents in that position strongly affect its activity.2 Therefore, also another
construct was prepared, introducing a cleavable linker between the carrier and doxorubicin, to
trigger its release once the system is internalized. To this end we choose a tetrapeptide, glycine‐
phenylalanine‐leucine‐glycine (GFLG), known to be cleaved by lysosomal cathepsins and already
used as a linker to release polymer‐bound doxorubicin.3
In order to better study the chemistry of the conjugates we first prepared fullerene
derivatives and then we translated the same chemistry to CNTs. Additionally, this strategy led to
an expansion in the kinds of constructs to be tested, thus allowing a comparison between the
biological behaviour of the different carbon nanostructures (CNSs).
Chapter
98
4.1 Full
4.1.1 Sy
T
fullero‐p
of fuller
monoad
and by
obtained
nitrogen
fulleropy
doxorub
probe fo
of the ve
illustrate
fullerene
yield).
required
4
erene deriv
nthesis
The first ste
pyrrolidine 1
rene C60, as
dduct and di
subsequent
d in 29% yie
n was then
yrrolidine 2.
Once obtain
bicin is fluor
or in vitro loc
ector with a
ed in Schem
e 2, followe
On the oth
d, to introdu
vatives
ep for the pr
1 by 1,3‐dipo
s depicted i
ifferent poly
chromatogr
eld. The ter
n Boc‐depro
ned the full
rescent, with
calization, b
non‐toxic m
me 2, by cou
ed by precip
er hand, to
ce a carboxy
reparation o
olar cycloadd
in Scheme
y‐adducts. Ne
raphic purifi
minal amine
otected by
S
lerene mono
h the conse
ut a differen
molecule. The
upling fluore
pitation of re
S
o prepare d
yl group. Ful
of the fullere
dition of azom
1. The reac
evertheless,
ication of th
e of the trie
treatment
Scheme 1
oadduct, it
equent adva
nt fluoropho
e fluorescein
escein isothi
eaction crud
Scheme 2
oxorubicin‐b
llerene 2 wa
ene derivativ
methine ylid
ction actuall
by use of 1
he mixture,
ethylene glyc
with trifluo
was derivat
ntage of be
re was used,
‐derivatized
ocyanate (F
de (in DMF)
bearing deriv
s treated wi
ves was the
es to the [6,
y affords a
1 equivalent
the desired
col chain on
roacetic ac
tized with f
ehaving simu
, to first stud
fullerene 3 w
ITC) with th
by addition
vatives, a fu
th succinic a
synthesis o
,6] double b
mixture of
t of the reag
mono‐addu
n the pyrroli
id to yield
fluorescein.
ultaneously
dy the behav
was prepare
he free amin
n of MeOH
urther step
anhydride, w
f the
onds
f the
gents
uct is
dinic
the
Also
as a
viour
ed, as
ne of
(63%
was
which
reac
carb
fulle
activ
chro
grou
unsu
nam
perf
puri
thou
cted with its
boxylic acid.
erene 4 in 75
The ami
vation of th
omatography
up was attem
uccessful, pr
Table
Therefo
mely allyloxy
formed in th
fication. The
ugh in low
s terminal a
The reaction
5% yield (Sch
ino Boc‐prot
e carboxylic
y purification
mpted using
robably due t
e 1. Condition
re a differen
ycarbonyl (A
e same cond
e deprotectio
yield. Treat
amine, with
n crude (in D
eme 3).
tected tetra
acid via ED
n (79% yield
g different c
to degradatio
a HC
b TF
c M
ns tested for
nt protectin
Alloc). The
ditions used
on of the te
ting Alloc‐G
consequent
DMF) was pu
Scheme
peptide Boc
DC∙HCl/HOBt
d, Scheme 4)
onditions, a
on of the pro
Scheme
Cl(g)/CH2Cl2, r
FA/CH2Cl2 (1:1
Me3SiCl, pheno
r the deprote
g group for
coupling b
for 8, afford
rminal amin
FLG‐doxorub
Co
t opening of
urified by pre
3
c‐GFLG was
, and produ
). Subsequen
s depicted i
oduct trigger
4
r.t.
1), r.t.
ol, CH2Cl2, r.t.4
ection reactio
the peptide
between All
ding 9 with 7
e from the A
bicin 9 with
njugation of
f the ring, a
ecipitation w
then couple
ct 8 was ob
ntly, the dep
n Table 1, w
red by the ac
on of Boc‐GF
e’s terminal
oc‐GFLG an
74% yield, af
Alloc group w
h tetrakis (t
f doxorubicin
and exposur
with MeOH, t
ed to doxoru
btained after
protection of
which result
cidic conditio
FLG‐doxorub
amine was
nd doxorub
fter chromat
was success
triphenyl ph
n to CNSs
99
re of the
to afford
ubicin by
r column
f the Boc
ed to be
ons.
bicin 8.
s chosen,
icin was
tographic
ful, even
osphine)
Chapter
100
palladiu
chromat
dimethy
allyl gro
product,
Howeve
conjugat
The fina
chromat
4
m(0) in the
tography, wi
ylbarbituric a
oup scavenge
, by reactio
er, results did
Finally, in o
ted with bot
al products
tography, wi
presence o
ith 38% yield
acid (see inse
er, in fact, s
on of the d
d not improv
rder to obta
th free doxor
5 and 6
ith 42% and
of dimedone
d (Scheme 5
et in Scheme
should avoid
deprotected
ve.
S
ain the targ
rubicin and G
were obtai
48% yield, re
S
e,5 we obtai
5). An attem
e 5) instead
d the undes
amine with
Scheme 5
get doxorubi
GFLG‐doxoru
ned, after
espectively (
Scheme 6
ned 10, wh
pt to increas
of dimedon
sired format
h one of di
cin‐bearing
ubicin 10, us
purification
Schemes 6 a
ich was pur
se reaction y
e was done.
ion of the k
imedone’s c
derivatives,
ing PyAOP as
by prepara
and 7).
rified by col
yield, using N
.6 The use of
ketoenamine
carbonyl gro
fullerene 4
s coupling ag
ative thin
lumn
N,N’‐
f this
e by‐
oups.
was
gent.
layer
4.1.2
shap
gave
cont
in al
be a
solu
inte
Fig
M)
2 Spectrosco
Absorpt
pe of fullere
e absorption
tribution of f
Fluoresc
ll cases a stro
appreciated
ution of an e
nsity.
gure 1. a) Ab
; b) fluoresc
opic characte
ion spectrum
ne monoadd
n profiles si
fullerene 2.
cence spectr
ong quenchi
in Figures 1
equimolar am
bsorption spe
ence spectra
M). Spe
erization
m of fulleren
ducts, with a
milar to the
oscopy of th
ng of the flu
1b and 2b.
mount of ful
ectra of FITC
a of FITC, FIT
ectra were re
Scheme
ne 2, as Figu
a maximum
e ones of F
he final fluor
orescence w
On the othe
lerene 2 did
C, fullerene 2
TC + fullerene
ecorded upo
Co
7
ures 1a and
at 433 nm.7
ITC and dox
ophore‐bear
with respect t
er hand, the
not cause s
and fulleren
e 2 and fuller
on excitation
njugation of
2a show, p
Fullerene d
xorubicin, re
ring fullerene
to the fluoro
e presence
strong decre
ne 3 (2∙10‐5 M
rene 3 (10‐6 M
n at 488 nm.
f doxorubicin
presented th
derivatives 3,
espectively,
e 3, 5 and 6
ophores alon
in the fluor
eases in the
M in DMSO/T
M in DMSO/
n to CNSs
101
e typical
, 5 and 6
plus the
revealed
ne, as can
ophores’
emission
TEA 0.1
/TEA 0.1
Chapter
102
Figure 2
DMS
f
the case
shorter.
T
the fluo
Transien
data are
concern
4.2 CNT
and we
cycloadd
(120°C)
normally
treatme
function
with res
rates a
function
4
2. a) Absorpt
SO); b) fluore
fullerene 6 (5
Quite intere
e of the first
The reason
orescent mo
nt spectrosco
e not sufficie
However, co
was related
T derivative
Once prepar
e choose p
dition on the
of a DMF d
y working fo
ent in hydro
nalization occ
pect to the o
Microwave
and improve
nalization wa
tion spectra
escence spec
5∙10‐6 M in D
estingly, com
t compound
for the quen
olecule, once
opy would h
nt to fully el
onsidering t
d to the decre
es: preparat
red the fulle
pristine MW
ese CNTs was
dispersion of
or oxidized C
ochloric acid,
curred, prob
oxidized tube
(MW)‐assist
ed product
as recently r
of doxorubic
ctra of doxor
DMSO). Spect
mparing fuller
d, where the
nching of th
e excited, a
help in bette
ucidate it.
he intended
ease in the f
tion and ch
erene derivat
WCNTs. How
s unsuccessf
f CNTs, alde
NTs (see Cha
, a very low
bably due to
es.
ted organic
yields. M
reported by
cin, fullerene
rubicin, doxo
tra were rec
rene 5 and f
e chain betw
e fluorescen
and the fulle
er understan
d biological
fluorescent s
aracterizat
tives, we mo
wever, the
ful. We perfo
ehyde and α
apters 3.1.1
w kaiser tes
the lower d
synthesis is
Moreover, th
our group.9
e 2, fullerene
orubicin + fu
orded upon
fullerene 6, t
ween the flu
nce could lie
erene, as al
nding the me
application
ignal during
ion
oved to CNT
first attem
ormed the re
α‐aminoacid,
and 3.2.1).
t (15 µmol/
dispersibility
s known to
he use of
9 It seems th
e 5 and fuller
llerene 2, ful
excitation at
the quenchi
orophore an
in energy t
ready obser
echanism, sin
of the const
in vitro expe
s, using the
pt to perf
eaction by th
i.e. in the
However, we
/g), meaning
in DMF of p
provide en
MW irrad
hat CNTs, in
rene 6 (10‐4 M
llerene 5 and
t 482 nm.
ng was high
nd fullerene
transfer betw
rved by oth
nce the pre
structs, the
eriments.
same chem
form 1,3‐dip
hermal treatm
same condit
e obtained,
g that very
pristine MWC
hanced reac
diation for
n the absenc
M in
d
er in
was
ween
hers.8
esent
main
istry,
polar
ment
tions
after
little
CNTs
ction
CNT
ce of
solv
prop
met
cond
in C
micr
cond
solv
a
b
c
d
e
f
g
h
tem
solv
prob
lowe
para
that
this
unst
tem
ent, are stro
posed also th
tal in the mic
Therefo
ditions. In al
H2Cl2, and s
rowave reac
ditions teste
ent, temper
a.a.
a 3
b 3
c 3
d 3
e 3
f 7
g 7
h 7
Table 2
As can c
perature (co
ent with re
bably becaus
er local he
aformaldehy
t could be ex
Importa
behaviour w
table, and
peratures eq
ongly heated
hat the heat
crowave field
re we decide
l cases the r
ubsequently
ctor where
ed, in terms
ature and re
aldehy
p‐HCH
y
p‐HCH
p‐HCH
p‐HCH
p‐HCH
p‐HCH
z
2. Different
Figure 3.
clearly be se
onditions a,
spect to ne
se MW irrad
eating of t
yde (conditio
xplained cons
ntly, a big im
was not unex
could the
qual to or ab
d by microw
ing of the sa
d.11
ed to perfor
reactants we
y the solvent
the reaction
s of aminoa
eaction time.
yde solve
HO ‐
‐
HO ‐
HO DM
HO ‐
HO ‐
HO ‐
‐
MW experim
azomethine
. Structures o
een by comp
c and e) res
eat condition
iation was a
the sample
ons b vs. a, a
sidering ster
mprovement
xpected. In fa
erefore und
bove 140°C, e
wave irradiat
ample is the
rm cycloaddi
ere mixed in
t was evapo
n mixture w
acid (a.a.) an
ent T (°C
120
120
140
F 140
160
160
160
160
mental condi
e ylides on pr
of aminoacid
parison of th
sulted in inc
ns (conditio
bsorbed by
. Moreover
and conditio
ric hindrance
t was obtain
act, the Boc
dergo depro
especially co
Co
tion due to
result of ma
ition under M
a quartz mic
orated. The t
was irradiate
nd aldehyde
C) Time
0 1
0 1
0 1
0 1
0 1
0 1
0 2
0 1
tions tested
ristine MWC
ds 3 and 7, a
he results ob
creased yield
ns d vs. c)
solvent, and
r, the use
ons h vs. f) d
e effect.
ed when usi
protecting g
otection du
onsidering th
njugation of
resonance p
agnetic reson
MW irradiat
crowave tub
tube was the
ed. Table 2
e used, pres
(h) k
depr
for 1,3‐dipo
NTs.
nd aldehyde
btained (Tab
ds, as expect
led to a de
not by CNT
of differe
did not impr
ing aminoaci
group of ami
uring reacti
e formation
f doxorubicin
phenomena.
nance of the
ion, testing
be by little so
en placed w
shows the
sence or ab
kaiser test aftrotection (µm
35
20
50
15
70
100
100
90
olar cycloadd
es y and z.
ble 2), an inc
ted. The pre
ecrease in t
Ts, with a con
ent aldehyd
rove the res
id 7 instead
noacid 3 is t
ions perfor
of “hot spot
n to CNSs
103
10 It was
e residual
different
onication
within the
different
sence of
ter mol/g)
ition of
crease of
esence of
he yield,
nsequent
es than
ults, fact
of 3 and
hermally
rmed at
ts” in the
Chapter
104
neat rea
aminoac
it is pro
deprote
part of t
consequ
Figure 4
thermal
improve
conditio
a treatm
confirme
fulleropy
4
action mixtu
cid 3 (Figure
obably due
ction for rea
the primary
uent lower ef
4. TGA (air) o
On the othe
ly stable ph
ement was o
ons f were se
ment with
ed the prese
yrrolidine 1.
Figure
ure during C
4), the therm
to loss of
actions c and
amine could
fficiency in t
of aminoacid
er hand, usi
hthalimide, t
observed pr
elected to pre
hydrazine, a
ence of the
e 5. TGA‐MS
CNT microw
mal decomp
Boc and de
d e, indeed, w
d become av
he formation
3. In the ins
i
ing aminoac
this problem
rolonging rea
epare MWCN
afforded M
desired cha
analyses (He
wave irradiat
osition of th
ecarboxylatio
was positive
vailable to t
n of the azom
set the enlarg
is shown.
cid 7, whose
m was avoid
action time
NT‐7, which
WCNT‐8 (Sc
ain on the C
e) of fulleren
tion. As can
e molecule s
on. Kaiser t
e, confirming
the aldehyde
methine ylid
gement of te
e terminal a
ed. Therefo
to 2 h (con
after deprot
cheme 8). T
CNTs, by com
ne 1 (a) and
n be seen b
starts early a
test perform
g this hypoth
e during the
e.
emperature r
amine is pro
re, consider
nditions g v
tection of the
TGA‐MS ana
mparison wit
MWCNT‐7 (b
by TGA in a
above 140°C,
med before
hesis. In this
reaction, w
range 100‐2
otected with
ring also tha
vs. f in Tabl
e phthalimid
alysis (Figur
th the profi
b).
ir of
, and
Boc‐
way,
with a
00°C
h the
at no
e 1),
de by
re 5)
le of
appe
inte
exce
MW
TGA
prep
obta
To
func
coup
anhy
reac
raisi
amin
gave
In fact,
ended mole
nsity in the
eptions were
WCNT‐7. The
A‐MS analyse
The foll
pared by rea
ained by ami
monitor the
ctionalization
pling (amine
ydride could
ction yield (in
ing tempera
nes were pro
e positive res
the same
ecules during
case of CNTs
e given by th
peaks ascrib
es of fulleren
lowing step
action of MW
idic coupling
e formation
n of 55 µm
e loading was
d possible cro
ncreasing mo
tures up to
obably hinde
sponse at Ka
pattern of
g the therm
s, consistent
he protecting
bable to the p
e 2 and MW
s were per
WCNT‐8 with
g with doxoru
n of MWCN
mol/g. Some
s 100 µmol/g
oss react dur
olar equivale
50°C) did no
ered to react
aiser test, be
peaks, give
ogravimetric
tly with thei
g groups, be
protecting gr
WCNT‐8, resp
rformed as
h succinic an
ubicin and G
NT‐9 we us
concern m
g) , since the
ring the follo
ents of succi
ot succeed.
t in that con
ing it perfor
Scheme
Co
n by fragm
c experimen
r lower degr
eing Boc for
roups were e
ectively (dat
for fulleren
nhydride, an
FLG‐doxorub
sed kaiser t
may arise fr
e free amine
owing step. H
nic anhydrid
Therefore it
nditions, due
med at 120°
8
njugation of
entation an
t, appeared
ree of functio
fullerene 1,
extrapolated
a not shown
ne derivative
d then MWC
bicin 10, resp
test, which
om the inc
s that did no
However, the
de, prolongin
was conclu
to aggregat
C.
f doxorubicin
nd ionization
d, even if wi
onalization.
and phthali
d by compari
n).
es: MWCNT
CNT‐10 and
pectively (Sc
gave a de
ompletenes
ot react with
e attempt to
ng reaction t
uded that pa
tion phenom
n to CNSs
105
n of the
th lower
The only
mide for
ison with
T‐9 were
11 were
heme 8).
egree of
s of the
h succinic
improve
ime, and
rt of the
mena, but
Chapter
106
T
Figure 6
10 and
doxorub
doxorub
be expla
kaiser te
derivativ
directly
describe
mixture
toward s
(see Cha
MWCNT
4
The loading
6a. The differ
2.4% for M
bicin for the
bicin in MWC
ained consid
est and TGA,
In addition
ves, perform
with the am
ed in Chapte
of aqueous
However, si
smaller valu
apter 3.2.1),
T with an ave
of doxorub
rence in the
MWCNT‐11, a
e two conju
CNT‐10 and
dering that t
which are n
Figur
to these co
ming an oxida
mine of doxo
er 3.2.1 was
sulphuric an
nce the len
es (average
we perform
erage length
icin and GFL
weight loss
and these v
ugates, respe
the amount
the two valu
ot always in
re 6. TGA (N2
njugates we
ation step to
orubicin and
used, name
nd nitric acids
S
gth distribut
length: 311
ed a shorter
of 215 ± 148
LG‐doxorubi
with respec
values corres
ectively. The
of carboxyl
ues were ga
perfect agre
2) of MWCNT
e prepared a
o introduce c
GFLG‐doxo
ely sonicatin
s.
Scheme 9
tion of this
nm ± 236 nm
r treatment,
8 nm (Schem
cin was eva
t to MWCNT
spond to 75
e discrepanc
ic groups ca
thered from
eement.
Ts derivative
also other d
carboxyl gro
rubicin 10. T
ng the nanot
batch of pr
m) with resp
i.e. 10 h long
me 9 and Figu
luated by TG
T‐9 was of 4.
5 µmol/g an
cy between
lculated for
m different m
s.
doxorubicin‐b
oups and the
The same ox
tubes in a so
ristine MWC
ect to the ba
g (instead of
ure 7).
GA, as show
.1% for MW
nd 45 µmol/
the amoun
MWCNT‐9 c
methods, na
bearing MW
en coupling t
xidation prot
onic bath w
CNTs was sh
atch used be
f 24 h), yieldi
wn in
CNT‐
/g of
nt of
could
mely
WCNT
them
tocol
with a
ifted
efore
ng s‐
Fi
leng
sam
case
incre
foun
doxo
nano
igure 7. a) Re
gth distributi
Afterwa
me protocol u
e, TGA was
ease with re
nd to be 3.0
orubicin was
TEM an
otubes, with
epresentativ
ons of pristin
rds, an amid
used for full
used to as
espect to s‐M
0%, correspo
s 2.2%, i.e. 40
nalyses conf
h a preserved
ve TEM imag
ne MWCNTs
indicate t
dic coupling w
erene 4 and
sess the de
MWCNT (Figu
onding to 55
0 µmol/g.
firmed the
d morpholog
ges of pristine
s (count = 12
the average
was perform
d MWCNT‐9,
egree of fun
ure 6b). For s
5 µmol/g, w
Scheme 1
quality of
gical structur
Co
e MWCNTs (
9) and s‐MW
length value
med, as descr
affording s‐
nctionalizatio
s‐MWCNT‐12
while for s‐M
10
the four fi
e (Figure 8).
njugation of
left) and s‐M
WCNT (count
es.
ibed in Sche
‐MWCNT‐12
on, consider
2 the loading
MWCNT‐13 t
nal conjuga
f doxorubicin
MWCNTs (rig
t = 146). Vert
eme 10, follo
2 and 13. Als
ring the we
g of doxorub
the loading o
ates, showin
n to CNSs
107
ght); b)
tical lines
owing the
so in this
ight loss
bicin was
of GFLG‐
ng clean
Chapter
108
Figure 8
T
the fulle
the pres
fluoresc
function
Figure
DMSO,
to doxo
upon ex
4
8. Represent
The fluoresc
erene deriva
sence itself o
ence with
nalization.
e 9. Fluoresce
relative to d
orubicin), co
xcitation at 4
tative TEM im
cence of DM
tives, a stro
of CNTs in th
respect to
ence spectra
doxorubicin)
mpared to t
482 nm. Note
mages of MW
MSO dispersio
ng quenchin
he dispersion
o fullerene
a of DMSO di
and of s‐MW
he one of do
e that CNT sa
intensity r
WCNT‐10 (a)
(d).
ons of the f
ng of emissio
n gave a grea
derivatives
ispersions of
WCNT‐12 and
oxorubicin (1
amples inten
refers to righ
and 11 (b),
four final con
on was obser
ater contribu
s, because
f MWCNT‐10
d s‐MWCNT‐
10‐5 M in DM
nsity refers to
ht axes.
and s‐MWCN
njugates wa
rved. Moreo
ution in redu
of the lo
0 and MWCN
13 (10‐6 M in
SO). Spectra
o left axes, w
NT‐12 (c) and
s studied. A
over, in this c
ucing doxoru
ower degree
NT‐11 (10‐5 M
n DMSO, rela
a were record
while doxoru
d 13
As for
case,
bicin
e of
M in
ative
ded
bicin
CNT
cons
fluo
the
oxid
colo
stud
drug
rele
well
4.3
hum
fulle
7) an
efflu
inte
Fig
fulle
conc
(Figu
In fact d
Ts, would be
structs are a
Therefo
rescence of
dispersions
dized, their d
our, and CNT
In conc
dying doxoru
g was attach
ase inside c
l.
Preliminary
First of
man breast c
erene conjug
nd one resist
ux pump ab
rnalized, thu
ure 10. Cell v
erene 2 at va
The res
centrations
ure 10).
oxorubicin lo
658 µmol/g
bout ten tim
re CNT cont
these conju
were diluted
dispersibility
s were proba
lusion, diffe
ubicin deliver
hed either d
ells. Further
y biological
all, we want
carcinoma ce
gates 5 and 6
tant to the d
ble to recog
us preventing
viability (ass
arious conce
sults showe
up to 25 µM
oading of ful
for fullerene
mes less funct
tent in the
ugates. It sho
d ten times,
y in DMSO w
ably complet
erent carbon
ry. Both fulle
directly or in
rmore, fuller
l results wit
ted to assess
ells, i.e. the
6. The cell lin
drug (MCF‐7/
gnize the dr
g it from exe
essed by MT
ntrations for
sextu
d a lack o
M, thus confi
llerene deriv
e 5 and 528 µ
tionalized th
dispersion
ould also be
in order to
was higher,
tely absorbin
n nanostruct
erene and C
ntroducing a
rene was fun
th fullerene
s the lack of
model to s
nes studied w
/ADR), due t
rug and tran
erting its cyto
TT assay) of M
r different tim
uplicate exp
of cytotoxici
irming the fe
Co
vatives, if exp
µmol/g for fu
han the corre
is most like
e pointed ou
detect an e
resulting in
ng doxorubic
tures‐based
NTs, pristine
a cleavable
nctionalized
e derivative
f cytotoxicity
tudy the an
were two, on
o overexpre
nsport it ou
otoxic action
MCF7 and M
me (data are
eriment).
ity in both
easibility of
njugation of
pressed in th
ullerene 6, m
esponding fu
ly responsib
t that, for s‐
mission sign
n dispersions
cin emission.
conjugates
e or oxidized
peptidic seq
with a non‐
es
y of the fulle
ticancer act
ne sensitive t
ssion of P‐gly
ut of the ce
.
CF7/ADR ce
e shown as m
cell lines
using fullere
f doxorubicin
he same unit
meaning that
ullerene deriv
ble for the v
‐MWCNT‐12
nal. Being the
s with deep
.
were prep
d, were used
quence to tr
‐toxic fluoro
erene carrie
tivity of dox
to doxorubic
ycoprotein (
ell once it h
lls after expo
mean values
for fulleren
ene as a dru
n to CNSs
109
used for
t the CNT
vatives.
very low
2 and 13,
ese CNTs
per black
pared for
d and the
rigger its
phore as
r toward
orubicin‐
cin (MCF‐
P‐gp), an
has been
osure to
± SD of a
ne 2, at
ug carrier
Chapter
110
dihexylo
cell heal
penetrat
the met
respect
the two
cells, ap
treatme
substant
the com
shown).
Figure
(25 µM,
7/ADR c
exploite
the free
MCF‐7/A
h, at a
intensity
(TB), wh
cells, thu
non‐inte
the fluo
behavio
4
Another im
oxacarbocyan
lth status. In
te cell mem
tabolic activ
to a healthy
probes, a ce
poptotic cells
ent with ful
tial changes
mpound did
11. Cell popu
, 48 h) (b). L1
Once assess
cells, the fo
d the fluore
fluorophore
ADR cells wa
concentratio
y (yellow line
hich is a dye
us quenchin
ernalized co
orescence int
ur means th
mportant inf
nine iodide (
fact PI is a D
branes of dy
vity of mitoc
y cell corresp
ellular popula
s and viable
lerene 2 w
occurred, w
not alter ce
ulation distri
1: dead cells
cells (P
sed the lack
llowing step
escence of fu
e due to que
as studied by
on of 25 µM
es in Figure 1
used to que
g only fluore
mponent du
tensity decre
hat probably
formation w
(DiOC6) test
DNA intercal
ying or dead
chondria. A
ponds to an a
ation could b
cells (Figure
with the unt
with the majo
ell viability.
ibution for M
(PI+/DiOC6‐
PI‐/DiOC6‐); L
of toxicity
p was studyi
ullerene der
nching phen
y flow cytoflu
M and then
12). This valu
ench extrace
escence give
uring cytoflu
eases to the
y fullerene 3
was obtain
. These two
ating probe,
cells. On the
decrease in
apoptotic ph
be divided in
e 11). By com
treated one
ority of the c
MCF‐7/ADR
MCF‐7 cells u
‐); L2: late ap
L4: viable cel
of the carri
ing the inte
ivative 3, ev
nomena, as a
uorimetry. Ce
analysed, s
ue did not de
ellular fluore
en by extrace
uorimetric an
e values of t
3, 1 h after
ed by a p
fluorescent
, which is exc
e other hand
n the level
ase. Combin
nto four grou
mparison of
, it is poss
cells in the “
R cells gave
untreated (a)
poptotic cells
lls (PI‐/DiOC6
er fullerene
rnalization o
ven though i
already ment
ells were inc
howing an i
ecrease after
escence. In fa
ellular dyes,
nalysis. Neve
he control (g
treatment, w
propidium
dyes are use
cluded by via
d, DiOC6 give
of DiOC6 flu
ning the infor
ups: dead cel
MCF‐7 cell
sible to app
“viable” zone
the same r
, or treated w
s (PI+/DiOC6
6+).
2 toward M
of the syste
it was lower
tioned. Upta
ubated with
ncrease in t
r treatment w
act, TB is no
and it is use
ertheless, w
grey lines in
was interact
iodide (PI)/
ed to invest
able cells but
es indication
uorescence
rmation give
lls, late apop
population
preciate tha
e, indicating
results (data
with fulleren
6+); L3: apopt
MCF‐7 and M
m. To do so
r with respe
ke by MCF‐7
h fullerene 3
the fluoresc
with Trypan
ot internalize
ed to exclude
washing the
n Figure 12).
ting with ce
/3,3′‐
igate
t can
ns on
with
en by
ptotic
after
t no
that
a not
ne 2
totic
MCF‐
o we
ct to
7 and
for 1
ence
Blue
ed by
e the
cells,
This
llular
mem
inte
cy
fulle
fulle
trea
requ
Figu
14 f
the
mbrane to s
rnalized and
Figure 12.
ytofluorimetr
before (yel
We ther
erene deriva
erene 3, reac
tments, cell
uirement of t
ure 13. Upta
Moreove
for the 48 h
25 μM treat
such an ext
d therefore it
Count of fluo
ry analysis of
low line) and
refore prolo
ative. As ex
ching soon a
ls were alwa
the experim
ake of fullere
er, we obser
treatment, w
ment with re
tent that it
t was then w
orescent cell
f MCF‐7 and
d after (grey
onged expos
pected, for
plateaux (F
ays washed
ental proced
ene 3 (25 μM
(with addit
rved a conce
with values o
espect to the
t was hinde
washed off.
ls (one repre
MCF‐7/ADR
line) washin
sure time, to
12 h or lo
igure 13). It
before the
dure.
M) by MCF‐7 c
ion of Trypa
entration‐de
of mean fluo
e 10 μM one
Co
ered from T
esentative ex
R cells treated
ng. Black line
o let cells t
nger treatm
should be p
flow cytofl
cells, plotted
n Blue) vs. ti
ependence o
orescence in
e.
njugation of
TB action, b
xperiment is
d with fullere
e represents
he time to
ments, cells
ointed out t
uorimetry a
d as mean flu
me.
f the uptake
tensity (MFI
f doxorubicin
but it was
shown) by f
ene 3 (25 μM
untreated ce
fully intern
actually inte
that, for thes
analysis, bein
uorescence i
e, as shown
) almost dou
n to CNSs
111
not fully
low
M, 1 h),
ells.
alize the
ernalized
se longer
ng this a
ntensity
in Figure
ubled for
Chapter
112
Figure 1
(10 or 2
unt
fluoresc
assess th
Also the
peptidic
conjugat
4.4 Ack
Laborato
the Boc‐
Antonia
de Casti
4
14. Flow cyto
25 μM, 48 h)
reated cells
ence intensi
Tryp
According to
he kinetics o
Furthermore
e CNT series
c sequence w
tes prepared
knowledgm
I wish to tha
oire d’Immu
‐protected a
I would also
Herrero (De
lla‐La Manch
ofluorimetry
. Upper pane
(one represe
ty (MFI) with
pan Blue (me
o these preli
of the interna
e, the cytoto
s of compou
will be asses
d.
ents
ank Dr. Albe
nologie et C
nd Alloc‐pro
like to than
epartamento
ha in Ciudad
analyses of
els show the
entative exp
h (striped his
ean MFI ± SD
minary data
alization.
oxicity of do
nds will be
sed, to verif
erto Bianco
himie Thérap
otected GFLG
k Dr. Luigi F
o de Química
Real, Spain)
MCF‐7 and M
e count of flu
eriment is sh
stograms) or
D of a duplica
a, uptake in t
oxorubicin‐be
evaluated. I
fy the impor
(CNRS, Instit
peutiques in
G tetrapeptid
eruglio for t
a Orgánica,
for the MAL
MCF‐7/ADR c
uorescent cel
hown). Lowe
r without (fu
ate experime
the time fram
earing fullere
n particular
rtance of dru
tut de Biolog
n Strasbourg,
des.
he 500 MHz
Facultad de
LDI‐TOF‐MS a
cells treated
lls, with blac
er panels sho
ll histograms
ent is shown)
me 1‐12 h w
ene 5 and 6
, the effect
ug release fr
gie Molécula
, France) for
NMR analys
Químicas‐IR
analyses.
with fullere
ck line relativ
ow the mean
s) the additio
).
will be studie
6 will be stud
of the cleav
rom the cova
aire et Cellu
the synthes
ses, and Pro
RICA, Univers
ne 3
ve to
n
on of
d, to
died.
vable
alent
laire,
es of
f. M.
sidad
Conjugation of doxorubicin to CNSs
113
For the biological studies, I wish to thank Marianna Lucafò, Dr. Sabrina Pacor, Dr. Sonia
Zorzet and Prof. Gianni Sava (Dipartimento di Scienze della vita, Università di Trieste).
4.5 References
1. a) H. Ali‐Boucetta, K. T. Al‐Jamal, D. McCarthy, M. Prato, A. Bianco, and K. Kostarelos,
Chem. Commun., 2008, 459‐461; b) Liu, X. Sun, N. Nakayama‐Ratchford, and H. Dai, ACS
Nano, 2007, 1, 50‐56; c) Z. Liu, A. C. Fan, K. Rakhra, S. Sherlock, A. Goodwin, X. Chen, Q.
Yang, D. W. Felsher, and H. Dai, Angew. Chem. Int. Ed., 2009, 48, 7668‐7672; d) Z. Liu, W.
Cai, L. He, N. Nakayama, K. Chen, X. Sun, X. Chen, and H. Dai, Nat. Nanotechnol., 2007, 2,
47‐52; e) X. Zhang, L. Meng, Q. Lu, Z. Fei, and P. J. Dyson, Biomaterials, 2009, 30, 6041‐
6047; f) E. Heister, V. Neves, C. Tîlmaciu, K. Lipert, V. S. Beltran, H. M. Coley, S. R. P. Silva,
and J. McFadden, Carbon, 2009, 47, 2152‐2160; g) P. Chaudhuri, S. Soni, and S. Sengupta,
Nanotechnology, 2010, 21, 025102 (11pp); h) R. Li, R. Wu, L. Zhao, M. Wu, L. Yang, and
H. Zou, ACS Nano, 2010, 4, 1399‐1408.
2. M. Binaschi, R. Farinosi, M. E. Borgnetto, and G. Capranico, Cancer Res., 2000, 60, 3770‐
3776.
3. S. R. Wedge, R. Duncan, and P. Kopeckova, Br. J. Cancer, 1991, 63, 546‐549.
4. E. Kaiser, J. P. Tam, T. M. Kubiak, and R. B. Merrifield, Tetrahedron Lett., 1988, 29, 303‐
306.
5. H. Kunz and C. Unverzagt, Angew. Chem. Int. Ed., 1984, 23, 436‐437.
6. H. Kunz and J. März, Angew. Chem. Int. Ed., 1988, 27, 1375‐1377.
7. K. Kordatos, T. Da Ros, M. Prato, R. V. Bensasson, and S. Leach, Chem. Phys., 2003, 293,
263‐280.
8. a) S. A. Vail, P. J. Krawczuk, D. M. Guldi, A. Palkar, L. Echegoyen, J. P. C. Tomé, M. A.
Fazio, and D. I. Schuster, Chem. Eur. J., 2005, 11, 3375‐3388; b) D. M. Guldi, M. Maggini,
G. Scorrano, and M. Prato, J. Am. Chem. Soc., 1997, 119, 974‐980.
9. F. G. Brunetti, M. A. Herrero, J. D. M. Munoz, S. Giordani, A. Diaz‐Ortiz, S. Filippone, G.
Ruaro, M. Meneghetti, M. Prato, and E. Vazquez, J. Am. Chem. Soc., 2007, 129, 14580‐
14581.
10. Z. Ye, W. Deering, A. Krokhin, and J. Roberts, Phys. Rev. B, 2006, 74, 075425 (1‐5).
11. A. Wadhawan, D. Garrett, and J. M. Perez, Appl. Phys. Lett., 2003, 83, 2683‐2685.
114
115
Experimental part
5.1 Materials and Methods
Chemicals were purchased from Sigma‐Aldrich or Acros Organics and used as received,
when not differently specified. Solvents were purchased from Sigma‐Aldrich, and deuterated
solvents from Cambridge Isotope Laboratories.
Single‐Walled carbon nanotubes (HiPCO), were purchased from Carbon
Nanotechnologies (lot #R0510C).
Double‐walled carbon nanotubes (CCVD), were provided both as pristine and as
shortened and oxidized by Nanocyl.
Multi‐walled carbon nanotubes (CCVD), were purchased from Nanostructured &
Amorphous Materials (stock# 1240XH and 1237YJS).
Antibody hCTM01 IgG, Fab’ and scFv were obtained from UCB.
Column chromatography was carried out with Merck silica gel 60 (particle size 40‐63
µm).
Preparative thin layer chromatography was carried out with Analtech pre‐coated glass
plates with 1.5 mm thick silica gel.
Thin layer chromatography was carried out on Merck pre‐coated aluminium plates with
silica gel 60 F254.
When anhydrous reaction conditions were required, reaction flasks were dried with a
heating gun (300‐500 °C), placed under high vacuum using a Schlenk line and purged with Ar. To
keep the atmosphere dry and inert balloons filled with Ar where used.
CNTs washing procedure by filtration consisted in dispersing the nanotubes in the
specified solvent (at a concentration of about 1mg/mL), followed by sonication for 15‐20 min
and vacuum filtration on polytetrafluoroethylene (PTFE) Millipore filters of the specified pore
size. The procedure was normally repeated twice or three times for each solvent. Et2O was
subsequently poured on the filtered CNTs, and vacuum was applied for 30 min after emptying
the filtration flask. Finally, CNTs were scratched from the filter and dries under high vacuum.
Kaiser test was performed using Fluka Kaiser test kit. In a typical test, 0.3‐0.5 mg of CNTs
were weighted. 75 µL of the kit solution of phenol (80% in EtOH) and 100 µL of the kit solution of
Chapter 5
116
KCN (in H2O/pyridine) were added and the dispersion was sonicated for 2 minutes in a sonic
bath. Subsequently, 75 µL of the kit solution of ninhydrin (6% in ethanol) were added, and the
mixture was heated at 120 °C for 10 minutes. Then, it was cooled and diluted with EtOH/H2O
(60%) to a final volume of 3mL. After centrifugation, the absorption spectrum of the supernatant
was measured, using as a blank a solution obtained in the same way as above but without CNTs.
The value of the absorption maximum at 570 nm was used to calculate amine loading in the CNT
sample (Ɛ = 15000 M‐1cm‐1). Reported values are average of at least two separate
measurements.
Dialysis of Ab or Ab‐CNTs samples was carried out with Spectrum Laboratories
membranes (regenerated cellulose membrane with a 12‐14 KDa molecular weight cutoff and
cellulose ester membranes with a 300 KDa molecular weight cutoff).
Size‐exclusion chromatography of Fab’ and scFv samples was performed using PD‐10
Desalting Columns pre‐packed with Sephadex G‐25 Medium, equipped with a LabMate PD‐10
Buffer Reservoir (GE Healthcare).
Gel electrophoresis was performed with a mini‐vertical electrophoresis system
(Invitrogen XCell SureLock), using a Novex®8‐16% Tris‐Glycine gel (Invitrogen) and run under
either non‐reducing or reducing conditions (by addition of 5% β‐mercaptoethanol). The gels
were stained with Coomassie blue.
Ellman’s assay was performed as follows: a stock 0.01 M solution of 5,5'‐Dithio‐bis‐(2‐
nitrobenzoic acid) (DTNB) in PBS‐EDTA‐NaHCO3 (pH 7.8) was prepared; 10 µL of DTNB stock
solution were diluted with 500 µL of buffer, and 50 µL of analyte solution was added. After
incubation at r.t. for 15 min, the absorption spectrum of the solution was measured, using as a
blank a solution obtained in the same way, adding 50 µL of buffer in place of the analyte
solution. The value of the absorption maximum at 412 nm was used to quantify free sulfhydryl
groups in the analyte solution (Ɛ = 14150 M‐1cm‐1).
5.2 Instrumentation
Ball milling treatments were carried out in a PM100 Planetary Mill (Retsch).
Microwave‐assisted reactions were carried out in a CEM Discover reactor.
Nuclear magnetic resonance (NMR) 200 MHz 1H‐NMR and 50 MHz 13C‐NMR spectra
were obtained on a Varian Gemini 200 spectrometer. 500 MHz 1H‐NMR spectra were obtained
on a Varian Inova. Chemical shifts are reported in ppm using the solvent residual signal as an
internal reference (CDCl3: δ H = 7.26 ppm, CD3OD: δ H = 3.31 ppm, DMSO‐d6: δ H = 2.50 ppm).
Experimental part
117
The resonance multiplicity is described as s (singlet), d (doublet), t (triplet), q (quartet), dd
(doublet of doublets), ddd (doublet of doublets of doublets) m (multiplet), br (broad signal).
Mass spectrometry: Electrospray Ionization (ESI) mass analysis was performed on a
Perkin‐Elmer API1. Matrix‐assisted laser desorption‐time of flight (MALDI‐TOF) mass analysis was
performed on a Bruker Daltonics instrument, using a matrix of 1,8‐dihydroxy‐9(10H)‐
anthracenone (dithranol)/silver trifluoroacetate. Mass isotopic distribution simulation were
performed with IsoPro 3.1 software.
UV–vis–NIR spectra were recorded on a Cary 5000 Spectrophotometer (Varian), using 1
cm path quartz or optical glass cuvettes.
Fluorescence spectra were recorded on a Cary Eclipse Fluorescence Spectrophotometer
(Agilent Technologies), using 1 cm path quartz cuvettes.
Thermogravimetric analyses were performed using a TGA Q500 (TA Instruments), with
the following procedure: isotherm at 100°C for 20 min (to remove residual solvent), ramp from
100 to 1000°C at 10°C/min, under N2 or air with a flow rate on the sample of 60 mL/min.
Reported graphs are average of at least two separate measurements. Ab‐CNT samples (in PBS)
were extensively washed with milliQ water by several cycles of centrifugation and removal of
the supernatant, followed by a dialysis (12‐14 kDa cutoff membrane), and subsequently
lyophilized, prior to TGA analysis, in order to remove the salts present in the buffer.
TGA‐MS experiments were performed on the same TGA instrument coupled with a
ThermoStar Mass Spectrometer (Pfeiffer Vacuum) with the following procedure: isotherm at
100°C for 20 min, ramp from 100 to 800°C at 20°C/min, under He with a flow rate on the sample
of 60 mL/min.
AFM analyses were carried out in Tapping‐Mode (TM‐AFM), in air at r.t., using a
Nanoscope IIIa, MMAFMLN (Veeco). Surfaces were imaged with phosphorus‐doped silicon tips
(cantilever: thickness = 3.5–4.5 μm, length = 115–135 μm, frequency f0 = 245–279 kHz, force
constant k = 20–80 N/m; Veeco). Statistical analysis of the AFM images was carried out using
Gwyddion 2.14 software. The surfaces for AFM analyses were prepared as follows: s‐SWCNTs
were dispersed in 1% SDBS aqueous solution at a concentration of 0.02 mg/mL, by 5 h sonication
in a sonic bath. Nanotubes were then deposited on mica surfaces by spin‐coating (3000 rpm, 3
min) 100–200 μL of these solutions. Finally, the surfaces were heated for 4 h in an oven at 180
°C, to remove the surfactant.
Raman spectroscopy analyses were performed on an inVia Raman microscope
(Renishaw), equipped with lasers at 532 nm, 633 nm or 785 nm, on solid samples deposited onto
Chapter
118
a glass
differen
T
voltage
were typ
of sonic
3.00 mm
vacuum
The sen
purchas
HEPES, 1
used as
surface
perform
formate
ethanola
Biotinyla
streptav
dissolve
constant
min, fol
each exp
5.3 Exp
5.3.1 Or
N‐Boc‐a
T
solution
the mixt
under va
celite to
5
coverslip. Re
t areas of th
TEM analyse
of 100 kV (im
pically suspe
cation, and t
m, 200 mes
overnight, p
Surface plas
nsor chip CM
ed from Bia
150 mM sod
running bu
of a sensor
med by inject
e buffer, pH
amine hydr
ated antigen
vidin until
d/dispersed
t flow rate o
lowed by a
periment by
erimental P
rganic compo
mino‐dietho
To a solutio
of Boc2O (1
ture at 0°C. T
acuum, and
o remove the
eported spe
e sample, an
es were per
mages were
ended in DM
these suspen
sh, coated w
prior to the T
smon resona
M5, surfacta
acore and st
dium acetate
uffer. The se
r CM5 chip w
tion onto th
4.3), which
ochloride (p
n and scramb
a response
in the runn
of 20 μL/min
dissociation
injection of
Procedures
ounds
oxy‐ethyl am
on of amino‐
11.79 g, 54 m
The reaction
the resulting
e white prec
ectra are ave
nd they are n
formed with
acquired us
MF (apart from
nsions were
with carbon
TEM analysis
ance measur
nt P20, ami
treptavidin f
e, 3 mM mag
ensor chip fo
was activate
he activated
gave a sign
pH 8.5) to
bled antigen
e of 700 R
ing buffer. B
n. Different c
phase of 3m
10 μL of 10 m
ine (1)
‐diethoxy‐eth
mmol) in CH2
was then st
g white semi
ipitate corre
erage of at
normalized to
h a Philips E
sing an Olym
m Ab‐CNT sa
drop‐casted
film), whic
s.
rements wer
ine coupling
from Sigma‐
gnesium acet
or the meas
ed by EDC/N
d surface wit
al of approx
saturate th
(1 μM in He
RU was ob
Binding expe
concentratio
min. The sen
mM HCl.
hyl amine (2
2Cl2 (100 mL
irred at r.t. f
i‐solid was d
esponding to
least 5 diffe
o the G‐band
M 208 micr
pus Morada
ample, whic
d on copper
h were sub
re performe
g kit contain
‐Aldrich. HEP
tate, 0.005%
surement w
NHS. Immob
th 35 μL of
ximately 500
he free acti
pes buffer) w
btained. Th
eriments wer
ons of Ab or
nsor chip su
20 g, 134.9 m
L) was added
for 24 hours.
dissolved in H
o the doubly
erent analys
d.
oscope with
CCD camera
h were in PB
r or nickel g
sequently d
d on a Biaco
ning NHS an
PES‐buffered
% surfactant
as prepared
ilization of s
streptavidin
00 RU, follow
ivated sites
were allowed
e different
re carried ou
Ab‐CNTs we
rface was re
mmol) in CH
d drop‐wise o
The solvent
H2O (70 mL)
substituted
ses performe
h an accelera
a). CNTs sam
BS) with the
grids (diamet
dried under
ore 3000 sys
nd EDC∙HCl w
d saline (10
P20, pH 7.4)
d as follows
streptavidin
n (100 μg/m
wed by 20 μ
of the ma
d to interact
analytes w
ut at 25°C w
ere injected
egenerated
H2Cl2 (100 m
over 3 h, sti
t was evapor
and filtered
by‐product
ed in
ating
mples
help
ter =
high
stem.
were
mM
) was
: the
was
mL in
μL of
atrix.
with
were
with a
for 3
after
mL), a
rring
rated
over
. The
aque
H2O
solv
Cha
N‐Bo
adde
drop
evap
H2O
rem
the
with
N‐Bo
adde
The
prod
Cha
N‐Ph
16.1
Star
colu
oil. C
N‐Ph
eous filtrate
(50 mL) to
ent under
racterization
oc‐amino‐die
To a solu
ed. Then, a s
p‐wise over 3
porated und
(70 ml x 3)
oved under
desired com
h literature.1
oc‐amino‐die
To a deo
ed, and the
catalyst was
duct was t
racterization
ht‐N‐Boc‐am
N‐Boc‐a
1 mmol) wer
rk apparatus
umn purificat
Characteriza
ht‐amino‐die
was extract
remove the
vacuum yi
ns were in ac
ethoxy‐ethyl
ution of 1 (7
solution of b
3 h at 0°C , a
der vacuum,
). The comb
vacuum. Pu
mpound 2 (5.
ethoxy‐ethyl
oxygenated M
reaction fla
s removed by
triturated in
ns were in ac
mino‐diethoxy
mino‐dietho
re dissolved
for 20 h. Th
tion (toluene
tions were in
ethoxy‐ethyl
ted with CH2
excess of d
elded comp
ccordance wi
lamino‐aceti
g, 28.2 mol)
benzyl brom
nd the react
and the res
bined organic
urification by
.52 g, 49%) a
lamino‐aceti
MeOH soluti
sk was purg
y filtration o
n diethyl e
ccordance wi
y‐ethyl amin
oxy‐ethyl am
in toluene (8
he solvent wa
e/AcOEt 7:3)
n accordance
amine (5)
2Cl2 (50 mL x
iamine, and
pound 1 (7
ith literature
ic acid benzy
) in CH2Cl2 (3
oacetate (6.
tion mixture
sidue was di
c phases we
y column ch
as a colourle
ic acid (3)
ion (90 mL) o
ged with H2 t
over celite, an
ether to g
ith literature
ne (4)
mine 1 (4 g,
80 mL), and
as then evap
) afforded th
e with literat
3). The orga
then dried
7.9 g, 58%
e.1
yl ester (2)
30 mL) at 0 °
46 g, 28.2 m
was stirred
ssolved in C
ere dried ov
romatograph
ess oil. Chara
of 2 (5 g, 12
three times
nd the solve
ive a whit
e.1
16.1 mmol)
the mixture
porated und
he pure prod
ture.1
anic phase w
with Na2SO4
) as a col
C, TEA (4.69
mol) in CH2Cl
at r.t. overni
CH2Cl2 (70 m
ver Na2SO4, a
hy (CH2Cl2/M
acterizations
.6 mmol), 10
and then sti
nt was evapo
e solid in
and phthali
e was stirred
er vacuum a
duct (4.33 g,
Experime
was backwas
4. Evaporatio
ourless visc
mL, 33.8 m
2 (40 mL) wa
ight. The sol
mL) and was
and the solv
MeOH 97:3)
s were in acc
0% Pd/C (90
irred at r.t.
orated. The
quantitativ
ic anhydride
d at 120°C in
and chromat
71%) as a co
ental part
119
hed with
on of the
cous oil.
mol) was
as added
vent was
hed with
vent was
afforded
cordance
mg) was
for 24 h.
resulting
ve yield.
e (2.39 g,
n a Dean‐
tographic
olourless
Chapter
120
T
in CH2Cl
reaction
removed
diethyl e
with lite
N‐Pht‐am
T
was add
added d
was eva
H2O (50
removed
the desi
literatur
N‐Pht‐am
T
was add
h. The c
resulting
Characte
Boc‐GFL
T
(contain
the solu
5
The N‐phtha
l2 (15 mL), a
n mixture w
d under vacu
ether to giv
erature.1
mino‐dietho
To a solutio
ded. Then, a
drop‐wise ov
porated und
0 ml x 3). Th
d under vac
red product
re.1
mino‐dietho
To a deoxyg
ded, and the
catalyst was
g product w
erizations we
LG‐doxorubic
To a solutio
ning 12% H2O
ution at 0°C.
alimido‐N‐Bo
nd TFA (15
as then allo
uum and the
e a white so
xy‐ethylamin
n of 5 (3.41
a solution of
er 3 h at 0°C
der vacuum,
he combined
uum. Purific
6 (1.48 g, 40
xy‐ethylamin
genated MeO
reaction fla
s removed b
was triturate
ere in accord
cin (8)
on of Boc‐G
O, 31.7 mg, 2
Then TEA (
oc‐amino‐die
mL) was slow
owed to reac
e resulting p
olid in quan
no‐acetic aci
g, 8.7 mmo
f benzyl bro
C , and the re
and the resi
d organic ph
cation by col
0%) as a colo
no‐acetic aci
OH solution
sk was purg
by filtration
ed in diethy
dance with li
FLG peptide
207 µmol) an
(29 µL, 207
ethoxy‐ethyl
wly added t
ch r.t. and
product, as t
titative yield
id benzyl est
ol) in THF (40
omoacetate
eaction mixtu
due was diss
hases were d
lumn chrom
ourless oil. Ch
id (7)
(40 mL) of 6
ed with H2 t
over celite,
l ether to g
terature.1
e (84.7 mg,
nd EDC∙HCl (
µmol) was
amine 4 (4
o the solutio
it was stirre
trifluoroaceti
d. Character
ter (6)
0 mL) at 0 °C
(1.99 g, 8.7
ure was stirr
solved in CH2
dried over N
atography (A
haracterizati
6 (1.48 g, 3.
hree times a
, and the so
give a white
172 µmol)
(39.7 mg, 20
added, and
g, 10.6 mmo
on, while sti
ed for 2 h. T
ic acid salt, w
izations wer
C, TEA (3.64
mmol) in T
red at r.t. for
2Cl2 (50 mL) a
Na2SO4, and
AcOEt/MeOH
ons were in
5 mmol), 10
and then stir
olvent was e
e solid in qu
in dry DMF
7 µmol) wer
the mixture
ol) was disso
rring at 0°C.
The solvent
was triturate
re in accord
4 mL, 26.1 m
THF (60 mL)
r 6 h. The so
and washed
the solvent
H 95:5) affo
accordance
0% Pd/C (40
rred at r.t. fo
evaporated.
uantitative y
F (1.5 mL), H
re added, sti
e was allowe
olved
. The
was
ed in
ance
mmol)
was
lvent
with
was
orded
with
0 mg)
or 24
The
yield.
HOBt
rring
ed to
reac
µmo
r.t.
was
chro
1H‐N
Hz, 1
1H),
1H),
(s, 3
= 18
(dd,
2H),
1040
Allo
(con
the
reac
µmo
r.t.
was
chro
(127
= 7.7
(br,
22.7
(s, 1
5.8 H
Hz,
ch r.t., and i
ol) and TEA (
under Ar, in
hed twice w
omatography
NMR (500 M
1H), 7.39 (d,
, 6.67 (d, J =
, 4.78 (dd, J
3H), 4.06‐4.0
8.8 Hz, 2H), 3
J = 14.6, 3.7
, 1.35 (s, 9H)
0.41], 1056.5
c‐GFLG‐doxo
To a so
ntaining 12%
solution at
ch r.t., and i
ol) and TEA (
under Ar, in
hed twice w
omatography
7.5 mg, 74%)
7 Hz, 1H), 7.
1H), 6.88 (d
7, 10.9, 5.7 H
1H), 4.78 (dd,
Hz, 1H), 4.38
1H), 3.93 (d,
t was stirred
(29 µL, 207 µ
n the dark,
with Et2O, to
y (CH2Cl2/Me
Hz, CDCl3): δ
J = 8.4 Hz, 1
8.8 Hz, 1H),
= 4.8, 2.8 Hz
4 (m, 1H), 4.
3.25 (br, 1H)
7 Hz, 1H), 2.
), 1.33 (d, J =
5 (M+K)+[cal
orubicin (9)
olution of Al
% H2O, 31.7 m
0°C. Then T
t was stirred
(29 µL, 207 µ
n the dark,
with Et2O, to
y (CH2Cl2/Me
) as a red sol
78 (t, 1H), 7
d, J = 4.9 Hz,
Hz, 1H), 5.57
, J = 4.9, 2.1
8‐4.31 (m, 1H
, J = 4.3 Hz,
d under Ar f
µmol) in dry
overnight. T
remove DM
eOH, 9:1), af
δ 13.96 (s, 1
1H), 7.36‐7.2
6.63 (d, J = 6
z, 2H), 4.71 (
.04‐3.99 (m,
), 3.10 (br, 1
.07‐1.95 (m,
= 6.6 Hz, 3H),
culated = 10
loc‐GFLG pe
mg, 207 µmo
TEA (29 µL, 2
d under Ar f
µmol) in dry
overnight. T
remove DM
eOH, 9:1), af
id. 1H‐NMR
.38 (d, J = 8.6
1H), 6.69 (d
(br, 1H), 5.5
Hz, 2H), 4.71
H), 4.24‐4.15
1H), 3.76‐3.6
for 15 min.
y DMF (2 mL
The crude r
MF, TEA and
ffording the
1H), 13.29 (s,
28 (m, 3H), 7
6.0 Hz, 1H), 5
(s, 1H), 4.57
1H), 3.92 (d
H), 2.99 (t, J
1H), 1.83 (d
, 0.92‐0.85 (m
056.52], 1016
eptide (82.2
ol) and EDC∙H
207 µmol) w
for 15 min.
y DMF (4 mL
The crude r
MF, TEA and
ffording, aft
(500 MHz, C
6 Hz, 1H), 7.
d, J = 7.7 Hz
3 (d, J = 3.7
1 (s, 1H), 4.5
5 (m, 1H), 4.0
61 (m, 4H), 3
A solution o
) was added
eaction mix
HOBt, and
pure produ
, 1H), 8.05 (d
7.18 (d, J = 7.
5.53 (d, J = 3
(dd, J = 11.3
d, J = 3.3 Hz,
J = 5.0 Hz, 1H
d, J = 8.0 Hz
m, 6H); ESI‐M
6.3 (M‐1)‐ [ca
mg, 172 µ
HCl (39.7 mg
was added, a
A solution o
) was added
eaction mix
HOBt, and
er precipitat
CDCl3): δ 13.9
37‐7.28 (m,
z, 1H), 6.64 (
Hz, 1H), 5.31
56 (dd, J = 10
08 (br, 1H), 4
3.32 (s, 1H),
of doxorubic
, and the re
ture was pr
it was finally
ct (138 mg,
d, J = 7.6 Hz
5 Hz, 2H), 7.
.6 Hz, 1H), 5
3, 5.7 Hz, 1H
1H), 3.77‐3.
H), 2.38 (d, J
, 1H), 1.70 (
MS: 1040.5 (
alculated = 1
µmol) in dry
g, 207 µmol)
and the mix
of doxorubic
, and the re
ture was pr
it was finally
tion with Et2
96 (s, 1H), 13
3H), 7.17 (d,
(d, J = 5.3 Hz
1 (br, 1H), 5.2
.9, 5.5 Hz, 1H
4.07 (s, 3H), 4
3.30‐3.23 (m
Experime
cin∙HCl (100
eaction was s
recipitated a
y purified by
79%) as a r
z, 1H), 7.78 (
.13 (br, 1H),
.32 (br, 1H),
H), 4.25 (br, 1
58 (m, 4H), 3
J = 14.7 Hz, 1
(s, 1H), 1.45‐
M+Na)+ [calc
1016.41].
y DMF (3 m
were added
xture was all
cin∙HCl (100
eaction was s
recipitated a
y purified by
2O, the pure
3.28 (s, 1H), 8
, J = 6.8 Hz, 2
z, 1H), 5.83
29‐5.26 (m,
H), 4.44 (dd,
4.01 (dd, J =
m, 1H), 3.10‐
ental part
121
mg, 172
stirred at
and then
y column
red solid.
(t, J = 8.1
6.88 (br,
5.20 (br,
1H), 4.08
3.30 (d, J
1H), 2.14
‐1.39 (m,
culated =
L), HOBt
d, stirring
lowed to
mg, 172
stirred at
and then
y column
e product
8.05 (d, J
2H), 7.05
(ddd, J =
1H), 5.24
J = 13.0,
16.3, 6.4
‐3.02 (m,
Chapter
122
2H), 3.00
1H), 1.8
1.27 (m,
GFLG‐do
mL). Dim
stirred a
cyclohex
excess d
95:5, the
product
7.89‐7.8
2.8 Hz, 1
5.6 Hz, 1
(d, J = 1
8.3 Hz, 1
(m, 4H),
5.3.2 Fu
Fulleren
ethylam
120 °C f
column
dissolve
brown s
5
0 (t, J = 4.9 H
1 (dd, J = 13
, 2H), 0.87 (d
oxorubicin (1
Alloc‐GFLG‐d
medone (28
at r.t., unde
xane, and th
dimedone, a
en 9:1), follo
(10 mg, 54%
81 (m, 1H), 7
1H), 5.19 (br
1H), 4.26‐4.2
6.9 Hz, 2H),
1H), 2.38 (d,
1.27 (d, J = 6
llerene deriv
ne 1
A toluene
mino‐acetic a
for 20 min. T
chromatogr
d in CH2Cl2 a
olid. Charact
Hz, 1H), 2.38
3.2, 4.6 Hz, 1
dd, J = 12.4, 6
10)
doxorubicin
mg, 200 µmo
r Ar and in
hen washed
nd once wit
owed by re‐p
%) as a dark
7.59 (d, J = 8
r, 1H), 4.74 (d
20 (m, 1H), 4
3.62 (br, 1H
J = 14.0 Hz,
6.5 Hz, 3H), 0
vatives
solution (30
cid 3 (170 m
Then, the re
raphy (tolue
and precipita
terizations w
8 (d, J = 15.0
1H), 1.63 (dd
6.5 Hz, 6H).
9 (20 mg, 2
ol) and Pd(P
the dark, fo
twice by pre
th Et2O. Puri
precipitation
red solid. 1H
.4 Hz, 1H), 7
d, J = 3.6 Hz,
.13 (d, J = 12
H), 3.25 (d, J
1H), 2.19 (d
0.90 (dd, J =
00 mL) of
mg, 560 µmo
action mixtu
ene, followe
ated by addit
were in accor
Hz, 1H), 2.1
d, J = 16.9, 7
20 µmol) wa
Ph3)4 (2.3 mg
or 2 h. Then
ecipitation fr
ification by
n from CH2C
H‐NMR (500
7.20 (d, J = 4
, 2H), 4.62 (d
2.6 Hz, 1H), 4
= 17.1 Hz, 1
dd, J = 14.7,
20.2, 5.3 Hz
C60 (400 m
ol) and p‐HC
ure was allo
d by toluen
tion of MeO
rdance with
4 (dd, J = 14
7.1 Hz, 1H), 1
s dissolved i
g, 2 µmol) w
n, the crude
rom CH2Cl2 w
column chro
Cl2/MeOH wi
MHz, CD3OD
.0 Hz, 3H), 7
dd, J = 8.1, 5
4.04 (s, 3H), 3
1H), 3.15‐3.0
4.8 Hz, 1H),
, 6H).
mg, 560 µm
CHO (167 mg
wed to reac
ne/AcOEt 8:2
H, to yield fu
literature.2
.7, 4.0 Hz, 1
1.32 (d, J = 6
in dry, deoxy
were added. T
e was precip
with cyclohe
omatography
ith Et2O, yie
D): δ 8.01 (d
.16‐7.10 (m,
.9 Hz, 1H), 4
3.90 (d, J = 1
8 (m, 2H), 2
2.05‐1.96 (m
mol), N‐Boc‐
g, 5.56 mmo
h r.t., and it
2). Finally, t
ullerene 1 (16
H), 2.02‐1.93
6.5 Hz, 3H), 1
xygenated TH
The reaction
pitated with
exane, to rem
y (CH2Cl2/Me
elded the de
d, J = 7.7 Hz,
, 2H), 5.42 (d
4.28 (dd, J =
16.8 Hz, 2H),
.92 (dd, J =
m, 1H), 1.67‐
‐amino‐dieth
ol) was heate
t was purifie
the product
60 mg, 29%)
3 (m,
1.31‐
HF (4
n was
cold
move
eOH,
sired
1H),
d, J =
12.3,
3.68
14.0,
‐1.52
hoxy‐
ed at
ed by
was
) as a
Fulle
slow
reac
resid
prod
liter
Fulle
FITC
dark
was
desi
1H),
6.53
(br,
(M+
= 12
Fulle
soni
mL)
over
erene 2
Fulleren
wly added to
ch r.t. and it
due was was
duct, affordi
rature.2
erene 3
To a solu
C (9.7 mg, 25
k, overnight.
hed by re‐pr
red product
, 10.04 (br, 1
3 (s, 4H), 4.55
4H), 2 H of t
+1)+ [calculat
282.16].
erene 4
Fulleren
ic bath. To th
and TEA (4
rnight. Purifi
e 1 (140 mg
o the solutio
t was stirre
shed by re‐p
ng fullerene
ution of fulle
5 µmol) were
. Finally, the
recipitation f
(20 mg, 63%
1H), 8.35 (s,
5 (s, 4H), 4.0
the ethylene
ed = 1284.1
e 2 (100 mg
his suspensio
1 µL, 298 µm
ication of re
g, 141 µmol)
on, while stir
d for 4 h. T
precipitation
e 2 in quant
erene 2 (25 m
e added, and
e crude mixt
from toluene
%) as an ochr
1H), 7.98 (b
01 – 3.94 (m,
e glycol chain
8], 1306.2 (M
g, 99 µmol)
on, a solution
mol) were a
action crude
) was dissolv
rring at 0°C.
The solvent
with CH2Cl2
itative yield
mg, 25 µmol)
d the reactio
ture was pre
e, with MeO
re‐brown sol
r, 1H), 7.71
, 2H), 3.80 (d
n are probab
M+Na)+ [calc
was suspend
n of succinic
added. The r
e by precipit
ved in CH2Cl
. The reactio
was remove
in order to
. Characteriz
) in dry DMF
n mixture w
ecipitated w
OH and from
lid. 1H‐NMR
(br, 1H), 7.1
d, J = 4.5 Hz,
bly covered b
culated = 130
ded in dry C
c anhydride (
reaction mix
tation and w
l2 (25 mL), a
on mixture w
ed under va
remove any
zations were
(4 mL), DIEA
as stirred at
with MeOH a
CH2Cl2 with
(500 MHz, D
3 (d, J = 8.3
2H), 3.75 (d,
by the H2O si
06.16], 1282
CH2Cl2 (30 m
19.9 mg, 198
ture was sti
washing in M
Experime
and TFA (25
was then all
acuum and t
trace of the
e in accorda
A (9 µL, 50 µ
t r.t., under A
and then th
MeOH, to a
DMSO‐d6): δ
Hz, 1H), 6.6
, J = 4.2 Hz, 2
ignal; ESI‐MS
2.1 (M‐1)‐ [ca
mL) by sonica
8 µmol) in C
irred at r.t.
eOH, to rem
ental part
123
mL) was
lowed to
the solid
e starting
nce with
mol) and
Ar, in the
oroughly
fford the
10.09 (s,
5 (s, 2H),
2H), 3.69
S: 1284.2
alculated
ation in a
H2Cl2 (10
under Ar
move TEA
Chapter
124
and suc
(500 MH
3.62‐3.5
[calculat
Fulleren
T
(11.8 mg
stirred a
DIEA (4
in the d
with Me
afforded
d6): δ 1
7.49 (d,
(d, J = 5
(m, 2H),
3.26‐3.2
(m, Hz, 4
6.1 Hz,
[calculat
5
cinic anhydr
Hz, CDCl3) δ
57 (m, 2H),
ted = 995.16
ne 5
To a solutio
g, 23 µmol) i
at r.t. under
µL, 23 µmol
dark overnig
eOH and Et2
d the desired
4.05 (s, 1H),
J = 7.6 Hz, 1
.8 Hz, 1H), 4
3.82 (br, 1H
20 (m, 2H), 3
4H), 2.15 (d,
3H); MAL
ted = 1106.2
ride, yielded
4.64 (s, 4H)
3.55‐3.45 (
6], 1017.1 (M
on of fulleren
in dry DMF (
Ar for 20 m
) in dry DMF
ht. The crud
2O. Preparat
d product (9.
, 13.30 (s, 1H
1H), 5.39 (s, 1
4.55 (d, J = 6
H), 3.75‐3.68
.16‐3.06 (m,
, J = 4.0 Hz, 2
DI‐TOF‐MS:
23], 800.32 (M
the desired
), 4.10 (t, J =
(m, 4H), 2.6
M+Na)+ [calcu
ne 4 (15 mg
1 mL) and D
min. Then, a
F (2 mL) was
de was prec
ive thin laye
.7 mg, 42%)
H), 7.98‐7.92
1H), 5.23 (br
.0 Hz, 2H), 4
(m, 2H), 3.6
2H), 3.02 (d
2H), 1.86‐1.7
1519.29, M
M‐C60+1) [ca
d product (7
= 5.5 Hz, 2H
69 (m, 2H),
lated = 1017
g, 15 µmol)
DIEA (4 µL, 23
solution of
s added, and
ipitated upo
er chromatog
as a reddish
2 (m, 2H), 7.
r, 1H), 4.98‐4
4.49 (s, 4H), 4
66‐3.58 (m, 2
d, J = 18.0 Hz
73 (m, 1H), 1
M‐1 [calcula
lculated = 80
3.8 mg, 75%
H), 3.88‐3.83
2.58 (m, 2
7.14], 993.1 (
in dry DMF
3 µmol) were
doxorubicin
d the reactio
on addition
graphy purif
/brown solid
.82 (t, 1H), 7
4.92 (m, 1H),
4.10‐4.03 (m
2H), 3.45 (t, J
z, 1H), 2.93 (
1.38 (dd, J =
ated = 151
00.33].
%) as a brow
(m, 2H), 3.
2H); ESI‐M
(M‐1)‐ [calcu
(3 mL), a so
e added and
∙HCl (10.5 m
n was stirred
of MeOH, a
fication (CH2
d. 1H‐NMR (5
7.69 (dd, J =
, 4.83 (t, J =
m, 1H), 4.00 (
J = 5.8 Hz, 2
d, J = 18.2 H
12.5, 4.0 Hz,
9.32], 1106
wn solid. 1H‐
77‐3.72 (m,
S: 995.2 (M
lated = 993.
olution of Py
the mixture
mg, 18 µmol)
d at r.t. unde
nd then wa
2Cl2/MeOH, 9
500 MHz, DM
7.6, 2.1 Hz,
6.0 Hz, 1H),
(s, 3H), 3.97‐
H), 3.43 (br,
Hz, 1H), 2.27‐
, 1H), 1.11 (d
6.23, (M‐413
NMR
2H),
M+1)+
14].
yAOP
e was
) and
er Ar
shed
92:8)
MSO‐
1H),
4.67
‐3.90
1H),
‐2.18
d, J =
3.09)
Foun
Fulle
mg,
stirr
and
und
was
92:8
DMS
8.03
7.45
4.95
4.46
(br,
(t, J
nd and calcu
erene 6
To a solu
15 µmol) in
red at r.t. un
DIEA (3 µL,
er Ar in the
hed with Me
8) afforded t
SO‐d6): δ 14
3 (s, 1H), 7.94
5 (d, J = 7.8 H
5 (t, 1H), 4.8
6‐4.39 (m, 1H
1H), 3.93 (t,
= 6.0 Hz, 2H
ulated mass i
ution of fulle
n dry DMF (0
nder Ar for 2
15 µmol) in
dark overni
eOH and Et2O
the desired p
4.05 (s, 1H),
4‐7.91 (m, 2
Hz, 1H), 7.26
2 (t, J = 6.1
H), 4.20 (dd,
J = 5.6 Hz, 2
H), 3.41 (br, 1
sotopic distr
erene 4 (10 m
0.5 mL) and
20 min. Then
n dry DMF (
ight. The cru
O. Preparativ
product (9.2
13.27 (s, 1H
H), 7.84 (d, J
‐7.19 (m, 3H
Hz, 1H), 4.74
J = 14.4, 8.3
2H), 3.72 (dd
1H), 3.28 (t,
ribution for f
mg, 10 µmol
DIEA (3 µL,
n, a solution
(1.5 mL) was
ude was pre
ve thin layer
mg, 48%) a
H), 8.29 (dd,
J = 8.1 Hz, 1H
H), 7.18‐7.12
4 (d, J = 5.9
3 Hz, 1H), 4.1
d, J = 5.7, 3.7
2H), 3.21 (dd
fullerene 5:
) in dry DMF
15 µmol) w
of GFLG‐do
s added, and
ecipitated up
chromatogr
as a reddish/
J = 5.8, 5.4
H), 7.76 (q, J
(m, 2H), 5.4
Hz, 1H), 4.56
14 (dd, J = 12
7 Hz, 2H), 3.6
d, J = 11.5, 5
F (2 mL), a so
were added a
xorubicin 10
d the reactio
pon addition
raphy purific
/brown solid
Hz, 1H), 8.04
= 5.3 Hz, 1H
44 (s, 1H), 5.2
6 (d, J = 6.0
2.7, 6.5 Hz, 1
65‐3.60 (m, 4
5.6 Hz, 2H), 3
Experime
olution of Py
and the mix
0 (11.1 mg, 1
on was stirre
of MeOH, a
cation (CH2Cl
d. 1H‐NMR (5
04 (d, J = 3.3
H), 7.70‐7.63
23 (d, J = 3.0
Hz, 2H), 4.49
1H), 3.99 (s, 3
4H), 3.59 (t, 2
3.04 (dd, J =
ental part
125
yAOP (7.9
ture was
12 µmol)
ed at r.t.
and then
2/MeOH,
500 MHz,
Hz, 1H),
3 (m, 1H),
0 Hz, 1H),
9 (s, 4H),
3H), 3.95
2H), 3.46
14.2, 3.9
Chapter
126
Hz, 1H),
1H), 2.1
(d, J = 6
1893.51
Found a
5.3.3 An
IgG‐SH
T
prepare
buffer w
20 equiv
the exce
6.5) at 4
be appro
5
2.97 (s, 2H)
2 (dd, J = 14
6.5 Hz, 3H),
1], 1480.44, (
nd calculate
ntibodies
To an Ab so
d solution o
was added, to
valents with
ess of 2‐IT w
4°C. The num
oximately 5
, 2.84 (dd, J
4.2, 5.5 Hz, 1
0.83 (dd, J =
M‐413.09) [c
ed mass isoto
lution (1 mL
of 2‐iminothi
o reach a fin
respect to t
was removed
mber of free
per Ab.
= 14.0, 10.3
1H), 1.89‐1.7
= 30.3, 6.4 H
calculated =
opic distribut
L, 67 μM) in
iolane∙HCl (2
al 2‐IT conce
the Ab. The
d by dialysis
sulfhydryl g
Hz, 1H), 2.4
9 (m, 1H), 1
Hz, 6H); MA
1480.42], 11
tion for fulle
PBS (5 mM E
2‐IT, Traut’s
entration of
mixture was
(MWCO = 1
groups introd
5‐2.23 (m, 4
.58‐1.48 (m,
LDI‐TOF‐MS
174.53 (M‐C6
rene 6:
EDTA, 50 mM
reagent) (90
1.3 mM, cor
s shaken for
12‐14 kDa) a
duced was a
4H), 2.19 (dd,
2H), 1.48‐1.
: 1893.61, M
60+1) [calcula
M NaHCO3, p
0 μL, 14.5 m
responding t
1 h at r.t. a
against PBS (
ssessed by E
, J = 14.3, 2.
.39 (m, 2H),
M‐1 [calculat
ated = 1174.
pH 7.8), a fre
mM) in the s
to approxim
and subseque
(5 mM EDTA
Ellman’s assa
7 Hz,
1.12
ted =
52].
eshly
same
ately
ently
A, pH
ay to
Fab’
fres
fina
resp
by s
scFv
fres
fina
resp
obta
5.3.4
s‐SW
for 1
was
dilut
time
vacu
s‐SW
stirr
’‐SH
To a Fab
hly prepared
l cysteamine
pect to Fab’.
ize exclusion
v‐SH
To a scF
hly prepared
l cysteamine
pect to Fab’.
ained by size
4 CNT deriva
WCNT‐1
HiPCO S
12 h under A
added, allo
ted in water
es with H2O,
uum, yielding
WCNT‐2
s‐SWCNT
red at 100°C
b’ solution (1
d cysteamine
e concentrat
The mixture
n chromatog
Fv solution (
d cysteamine
e concentrat
The mixture
e exclusion ch
atives
SWCNTs (100
Ar atmosphe
owing air in,
r and filtere
up to a neu
g 770 mg of
Ts‐1 (100 m
for 48h und
1.8 mL, 123
e solution (3
tion of 4 mM
e was shaken
raphy purific
7.5 mL, 20 µ
e solution (15
tion of 4 mM
e was shake
hromatograp
00 g) were d
re. Then, of
and stirring
d (pore size
utral pH in t
s‐SWCNT‐1.
g) were disp
er Ar atmosp
µM) in 50 m
6 µL, 200 m
M, correspo
n for 2 h at
cation.
µM) in 50 m
50 µL, 200 m
M, correspon
en for 2 h at
phy purificat
ispersed in o
mixture of o
g the mixtur
= 5 µm). A
the filtrate, M
persed in aqu
phere (in a P
mM sodium
M) in the sa
nding to app
r.t. and then
mM sodium
mM) in the sa
nding to app
r.t. and sub
tion.
oleum (500 m
oleum and aq
re at 65°C fo
fterwards, t
MeOH and E
ueous NaOH
PTFE flask). A
acetate (2 m
me buffer w
proximately
n the pure F
acetate (2 m
ame buffer w
roximately 2
bsequently th
mL), and the
queous nitric
or 2h. The re
he material
Et2O and fina
H 8M (100 m
Afterwards, C
Experime
mM EDTA, p
was added, to
32 equivale
Fab’‐SH was
mM EDTA, p
was added, to
200 equivale
he pure scFv
e mixture wa
c acid (65%)
eaction mixt
was washed
ally dried un
mL) by sonica
CNTs were se
ental part
127
H 5.6), a
o reach a
ents with
obtained
H 5.6), a
o reach a
ents with
v‐SH was
as stirred
(500 mL)
ture was
d several
nder high
ation and
eparated
Chapter
128
from th
washed
In order
in 0.1 M
and Et2O
s‐SWCNT
steel gri
the plan
Chapter
sonicatio
cycles o
dissolve
size = 0
SWCNT‐
s‐SWCNT
T
differen
supplied
conditio
s‐DWCN
min. A s
µmol) in
The resu
5
e dissolved
with H2O, M
r to perform
M HCl and su
O.
T‐3 (A, B, C a
In a typical e
inding jar wi
netary mill.
2.2. After t
on for 5 min
of sonication
d Fe3+ speci
0.1 μm) with
‐3A, B, C and
T‐4 (B, C, E a
The prepara
ce that the
d with N2, an
on and final w
NT‐1
A suspensio
solution of E
n DMF (5 mL
ulting NHS‐a
amorphous
MeOH and Et2
Raman anal
bsequently f
and D)
experiment,
ith eight stai
The differen
the correspo
n, and filtere
n and centr
es disappea
h H2O, MeO
d D. The weig
and F)
ation followe
starting wei
nd that the
weight yields
n of s‐DWCN
DC∙HCl (19.2
L) was added
activated s‐D
material by
2O to afford,
ysis, the aqu
filtering it (p
30 mg of pr
inless steel b
nt rotationa
onding treat
ed (pore size
rifugation w
red. Afterwa
OH and Et2O
ght yields are
ed the same
ighting oper
grinding jar
s are reporte
NT (10 mg) i
2 mg, 100 µm
d and the re
DWCNT were
y filtration (
, after high v
ueous filtrate
ore size = 0.
ristine SWCN
balls (1 cm d
l speed and
tment, the s
= 0.1 μm). T
with HCl 37%
ards, the s‐S
O and finally
e reported in
e protocol a
ration was p
was sealed
ed in Table 1
in DMF (5 m
mol), NHS (1
eaction mixtu
e then wash
(pore size =
vacuum dryin
e was neutra
.45 µm), and
NTs were int
diameter), w
d times used
s‐SWCNTs we
The collecte
% (w/w) unt
SWCNTs wer
y dried unde
n Table 1 of C
s the one fo
erformed in
before placi
of Chapter 2
mL) was sonic
11.5 mg, 100
ure was stirr
hed several
0.45 µm),
ng, 40 mg of
alized by stir
d washing it w
roduced in a
which was the
d are detaile
ere suspend
d black solid
til the yello
e washed by
er high vacu
Chapter 2.2.
or s‐SWCNT‐
side a glove
ing it within
2.2.
cated in a w
0 µmol) and
red for 24 h
times with
and thorou
s‐SWCNT‐2.
rring the solu
with H2O, M
a 25 mL stai
en placed w
ed in Table
ded in CH2Cl
d was purifie
ow color du
y filtration (
uum affordin
‐3, with the
ebox worksta
the mill. M
water bath fo
DIEA (17 µL,
at r.t. unde
isopropanol
ughly
ution
MeOH
nless
within
1 of
2, by
ed by
ue to
pore
ng s‐
only
ation
illing
or 20
, 100
er Ar.
and
diet
sequ
μM)
decr
reac
thor
spec
and
gram
s‐DW
adde
reac
redu
ethy
bath
wer
drie
diox
was
vacu
s‐DW
neut
male
for 2
was
hyl ether by
uences, and
NHS act
) in PBS (pH
rease of IgG
ction. Finally
roughly with
ctroscopy. Th
stored at 4
m of CNTs.
WCNT‐2
s‐DWCN
ed and the
ction was sti
uced pressur
yl amine 1 (1
h for 30 min.
e washed se
d under high
Then, th
xane (100 m
hed by filtra
uum. The deg
WCNT‐3
s‐DWCN
tralized wi
eimidopropi
20 min and t
hed by filtra
y successive
they were fi
ivated s‐DW
7.4) and the
concentrati
y, the suspen
h PBS (pH
he resulting
°C. The degr
NTs (100 mg)
reaction wa
irred at 75°C
re. The CNTs
100 mg, 403
. Then the re
everal times
h vacuum.
he Boc group
L), by sonica
ation (pore
gree of funct
NTs‐2 (20 m
ith dry D
onate (12.8
then stirred
tion (pore si
sonication (
nally dried u
CNT (10 mg)
e mixture wa
on in the su
nsion was ce
7.4) until
Ab conjugat
ree of functi
were suspe
s sonicated
C under Ar f
s were susp
µmol) was
eaction mixtu
by filtration
p was cleave
ation for 30
size = 0.45
tionalization
g, 2.4 μmo
IEA (21 µ
mg, 48 μmo
at r.t. under
ize = 0.45 μm
water bath)
under high va
) were dispe
as shaken fo
upernatant, a
entrifuged an
no Ab was
te s‐DWCNT‐
onalization,
ended in thio
in a water b
for 24 h. Af
ended again
added to th
ure was heat
n (pore size
ed by disper
min, and st
μm) with D
n, determine
l of amine
µL, 120 μ
ol) in DMF (2
r Ar for 48 h.
m) with DMF
, centrifugat
acuum.
ersed in 20 m
r 60 h at r.t.
after centrif
nd the precip
s detected
‐2 was dialys
determined
onyl chloride
bath for 30
ter this peri
n in dry pyrid
e mixture, fo
ted at 90°C u
= 0.45 μm)
sing CNTs (1
irring it at r
DMF, MeOH
d by Kaiser t
groups) we
μmol). A
2 mL) was ad
. The obtaine
F, MeOH and
tion and rem
mL of IgG solu
., monitoring
ugation of a
pitate was co
in the supe
sed against P
by TGA, wa
(20 mL), the
min under A
od the solve
dine and N‐
ollowed by s
under Ar for
with DMF, M
100 mg) in a
.t. overnight
and Et2O an
est, was 120
ere suspende
solution o
ded. The rea
ed s‐DWCNT
d Et2O, and th
Experime
moval of sup
ution (0.45 m
g the reactio
a small aliquo
ollected and
ernatant (by
PBS (pH 7.4)
as 2.1 µmol o
en DMF (0.4
Ar. Subseque
ent was drie
Boc‐amino‐d
sonication in
96 h. The na
MeOH and E
a 4 M HCl so
t. The DWCN
nd dried un
0 μmol/g.
ed in dry D
of N‐succin
action was s
Ts‐3 were ex
hen dried un
ental part
129
pernatant
mg/mL, 3
on by the
ot of the
d washed
y UV‐vis
for 24 h
of Ab per
mL) was
ently the
ed under
diethoxy‐
n a water
anotubes
Et2O, and
olution in
NTs were
der high
DMF and
imidyl‐3‐
sonicated
tensively
nder high
Chapter
130
vacuum
the one
s‐DWCN
T
solution
r.t., mo
centrifug
thoroug
spectros
mg/mL i
per gram
s‐DWCN
s
20 min
paraform
and the
filtration
T
dioxane
washed
vacuum
s‐DWCN
5
. The degree
of s‐DWCNT
NT‐4
The maleim
(0.45 mg/m
nitoring the
gation of a
ghly with PB
scopy). The r
in PBS (pH 7
m of CNTs.
NT‐5
s‐DWCNTs (1
n. N‐Boc‐am
maldehyde (
reaction mi
n (pore size =
Then, the Bo
(100 mL),
by filtration
. The degree
NT‐6
e of functiona
T‐2, was 60 μ
mido‐derivatiz
mL, 3 μM) in P
e reaction by
small aliqu
BS (pH 7.4)
resulting Ab
.4). The deg
100 mg) wer
mino‐dietho
294 mg, 9.8
ixture was h
= 0.45 μm) w
oc group wa
by sonicatio
n (pore size
e of functiona
alization, det
μmol/g (Yield
zed s‐DWCN
PBS (5 mM E
y the decrea
uot of the r
) until no
‐conjugate s
ree of functi
re suspended
xy‐ethylamin
mmol) were
eated at 120
with DMF, Me
as cleaved by
on for 30 m
= 0.45 μm)
alization, det
termined by
d: 50%).
NT‐3 (10 mg
EDTA, pH 6.5
ase of IgG c
reaction. Th
Ab was de
s‐DWCNT‐4 w
ionalization,
d in 100 mL o
no‐acetic a
e added port
0 °C for 2 da
eOH and Et2
y dispersing
in, and stirr
) with DMF,
termined by
y comparison
g) were disp
5) and the m
concentratio
he CNTs we
etected in t
were stored
determined
of DMF and
acid 3 (30
tion‐wise ove
ays. CNTs we
O and dried
CNTs (100
ring it at r.t
MeOH and
y the Kaiser t
n of the Kaise
persed in 2
ixture was sh
on in the su
re centrifug
he superna
at 4 °C as d
by TGA, wa
sonicated in
00 mg, 98
er 2 days (1 a
ere washed
under high v
mg) in a 4 M
. overnight.
d Et2O and d
est, was 80 μ
er test value
20 mL of Ig
haken for 60
upernatant,
ged and wa
atant (by U
dispersions o
as 2.2 µmol o
a water bat
80 µmol)
addition per
several time
vacuum.
M HCl solutio
The CNTs w
dried under
μmol/g.
with
G‐SH
0 h at
after
shed
V‐vis
of 0.5
of Ab
h for
and
day)
es by
on in
were
high
neut
(8.5
stirr
(por
of f
DWC
s‐DW
solu
r.t.,
cent
thor
spec
mg/
per
s‐MW
mixt
24 h
volu
H2O
high
s‐MW
s‐DWCN
tralized with
mg, 32 μm
red at r.t. un
re size = 0.45
functionaliza
CNT‐5, was 3
WCNT‐7
The ma
ution (0.45 m
monitoring
trifugation o
roughly with
ctroscopy). T
/mL in PBS (p
gram of CNT
WCNT
Pristine
ture (3:1 v/v
h, keeping te
ume of 150 m
until filtrate
h vacuum (we
WCNT‐1
NT‐5 (20 mg
h dry DIEA (1
ol) in DMF (
der Ar for 48
5 μm) with D
tion, determ
30 μmol/g (Y
leimido‐deri
mg/mL, 3 μM
the reactio
of a small aliq
h PBS (pH
The resulting
pH 7.4). The
Ts.
MWCNTs (
v, 98% and 6
emperature
mL and the m
e reached a
eight yield: 7
g, 1.6 μmol
14 µL, 80 μm
(2 mL) was a
8 h. The obt
DMF, MeOH a
mined by co
Yield: 38%).
vatized s‐DW
) in PBS (5m
on by the de
quot of the r
7.4) until
g Ab‐conjuga
degree of fu
100 mg) we
5%, respecti
below 50°C.
mixture was
neutral pH,
70%).
of amine
mol). A soluti
added. The r
ained s‐DWC
and Et2O, an
omparison o
WCNT‐6 (10
mM EDTA, pH
ecrease of I
reaction. Fin
no Ab was
ate s‐DWCNT
unctionalizat
ere disperse
ively), and th
Deionized w
filtered (por
then in Me
groups) we
on of N‐succ
reaction was
CNTs‐6 were
nd then dried
of the Kaiser
0 mg) were
H 6.5) and th
gG concent
nally, the CNT
s detected
T‐7 were sto
tion, determ
ed in 50 mL
he mixture w
water was th
re size = 0.1
OH and Et2O
re suspende
cinimidyl‐3‐m
s sonicated f
e extensively
d under high
r test value
dispersed i
e mixture wa
ration in the
Ts were cent
in the supe
ored at 4 °C
ined by TGA
L of aq. sulf
was sonicated
hen carefully
μm), and wa
O. Finally CN
Experime
ed in dry D
maleimidopr
for 20 min a
y washed by
vacuum. Th
with the o
in 20 mL o
as shaken fo
e supernata
trifuged and
ernatant (by
as dispersio
A, was 2.6 µm
furic acid/ni
d in a water
y added up t
ashed by filt
NTs were drie
ental part
131
DMF and
opionate
and then
filtration
e degree
one of s‐
of IgG‐SH
or 60 h at
ant, after
d washed
y UV‐vis
ns of 0.5
mol of Ab
itric acid
r bath for
to a final
tration in
ed under
Chapter
132
s
added a
reaction
reduced
amine 5
for 30 m
washed
under hi
20 min
paraform
and the
filtration
was add
were wa
vacuum,
was 170
s‐MWCN
s
DIEA (74
sonicate
the dark
and Et2
function
1, was 6
s‐MWCN
5
s‐MWCNTs (
and the reac
n was stirred
d pressure. T
5 (150 mg, 3
min. Then the
several time
igh vacuum.
CNTs (100 m
n. N‐Boc‐am
maldehyde (
reaction mi
n (pore size =
Finally, CNTs
ded. After so
ashed by filt
, to afford s
0 μmol/g.
NT‐2A
s‐MWCNT‐1
4 µL, 425 µm
ed for 20 min
k. Reaction m
2O and fina
nalization, de
60 μmol/g (Yi
NT‐2B
(100 mg) we
ction was son
d at 60°C un
The CNTs we
82 µmol) wa
e reaction mi
es by filtratio
mg) were the
mino‐dietho
294 mg, 9.8
ixture was h
= 0.1 μm) wit
s (100 mg) w
onication for
ration (pore
s‐MWCNT‐1.
(50 mg, 8.5
mol) and FITC
n under Ar a
mixture was
ally dried u
etermined by
ield: 35%).
ere suspende
nicated in a
nder Ar for
ere suspende
as added to
ixture was he
on (pore size
n suspended
xy‐ethylamin
mmol) were
eated at 120
th DMF, MeO
were dispers
r 20 min, th
size = 0.1 μ
The degree
μmol of am
C (99.3 mg, 2
nd in the da
filtered (por
under high
y compariso
ed in oxalyl c
water bath
24 h. After
ed again in d
the mixture
eated at 70°
e = 0.1 μm) w
d in 100 mL o
no‐acetic a
e added port
0 °C for 2 da
OH and Et2O
ed in EtOH (
e dispersion
μm) with EtO
of function
mine groups)
255 µmol) w
ark, and then
re size = 0.1
vacuum to
n of the Kais
chloride (20
for 30 min
this period t
dry THF and
e, followed b
C under Ar f
with DMF, M
of DMF and
acid 3 (30
tion‐wise ove
ays. CNTs we
O and dried u
(100 mL) and
n was stirred
OH, MeOH an
alization, de
were disper
were added a
n stirred at r
μm) and the
afford s‐M
ser test valu
mL), then D
under Ar. Su
the solvent
N‐Pht‐amin
by sonication
or 48 h. The
eOH and Et2
sonicated in
00 mg, 98
er 2 days (1 a
ere washed
nder high va
d hydrazine
d overnight a
nd Et2O and
etermined by
rsed in dry D
and the react
.t. overnight
en washed w
MWCNT‐2A.
e with the o
DMF (0.4 mL)
ubsequently
was dried u
no‐diethoxy‐e
n in a water
nanotubes w
2O, the and d
a water bat
80 µmol)
addition per
several time
acuum.
hydrate (600
at r.t.. The C
dried under
y the Kaiser
DMF (10 mL)
tion mixture
t under Ar an
with DMF, M
The degre
one of s‐MW
) was
y, the
nder
ethyl
bath
were
dried
h for
and
day)
es by
0 µL)
CNTs
high
test,
. Dry
e was
nd in
MeOH
e of
CNT‐
DIEA
soni
filte
high
of th
s‐MW
soni
filtra
degr
one,
s‐MW
and
male
for 2
size
func
35 μ
s‐MWCN
A (74 µL, 425
icated for 20
red (pore siz
h vacuum to
he Kaiser tes
WCNT‐3A an
s‐MWCN
ication for 3
ation (pore
ree of functi
, was 35 μmo
WCNT‐4A an
s‐MWCN
neutralize
eimidopropi
20 min and
= 0.1 μm) w
ctionalization
μmol/g (Yield
NT‐1 (50 mg,
5 µmol) and
0 min under
ze = 0.1 μm)
afford s‐MW
st value with
nd 3B
NT‐2A and 2B
0 min, and
size = 0.1 μ
onalization,
ol/g in both
nd 4B
NT‐3A and 3
d with dr
onate (3.7 m
then stirred
with DMF, M
n, determine
d: 100%) in b
, 8.5 μmol of
DTPA (91.1 m
Ar, and then
and then w
WCNT‐2B. The
the one of s
B (40 mg) we
stirring it at
μm) with DM
determined
cases.
3B (20 mg, 0
ry DIEA (6
mg, 14 μmol
at r.t. unde
MeOH and E
ed by compa
both cases.
f amine grou
mg, 255 µmo
n stirred at r
ashed with D
e degree of f
s‐MWCNT‐1,
ere dispersed
r.t. overnig
MF, MeOH a
d by compari
0.7 μmol of
6 µL, 35
) in DMF (2
er Ar for 48 h
Et2O, and the
arison of the
ups) were di
ol) were add
.t. overnight
DMF, MeOH
functionaliza
was 50 μmo
d in a 4 M HC
ht. The CNT
nd Et2O and
son of the K
amine group
μmol). A
mL) was ad
h. The CNTs
en dried und
e Kaiser test
spersed in d
ded and the r
t under Ar. R
and Et2O an
ation, determ
ol/g (Yield: 29
Cl solution in
Ts were subs
d dried unde
Kaiser test va
ps) were sus
solution o
ded. The rea
were washe
der high vac
value with t
Experime
dry DMF (10
reaction mix
Reaction mix
nd finally drie
mined by com
9%).
n dioxane (40
sequently wa
er high vacu
alue with the
spended in
of N‐succin
action was s
ed by filtrati
cuum. The d
the starting o
ental part
133
mL). Dry
xture was
xture was
ed under
mparison
0 mL), by
ashed by
uum. The
e starting
dry DMF
imidyl‐3‐
sonicated
on (pore
degree of
one, was
Chapter
134
s‐MWCN
T
Fab’‐SH
for 37
superna
centrifug
superna
as dispe
was 1.5
s‐MWCN
T
scFv‐SH
for 22
superna
centrifug
superna
as dispe
was 3.3
5
NT‐5A and 5B
The maleim
solution (0.4
h at r.t., m
tant, after
ged and wa
tant (by UV‐
rsions of 0.5
and 1.4 µmo
NT‐6A and 6B
The maleim
solution (0.1
h at r.t., m
tant, after
ged and wa
tant (by UV‐
rsions of 0.5
and 3.1 µmo
B
ido‐derivatiz
4 mg/mL, 8.
monitoring
centrifugatio
ashed thoro
‐vis spectros
5 mg/mL in P
ol of Fab’ per
B
ido‐derivatiz
16 mg/mL, 5
monitoring
centrifugatio
ashed thoro
‐vis spectros
5 mg/mL in P
ol of scFv per
zed s‐MWCN
4 µM) in PB
the reaction
on of a sma
oughly with
copy). The re
PBS (pH 7.4).
r gram of CN
zed s‐MWCN
5.2 µM) in PB
the reaction
on of a sma
oughly with
copy). The re
PBS (pH 7.4).
r gram of CN
NT‐4A and 4
S (5 mM ED
n by the d
all aliquot o
PBS (pH 7
esulting Fab’
The degree
NTs, respectiv
NT‐4A and 4
BS (5 mM ED
n by the d
all aliquot o
PBS (pH 7
esulting scFv
The degree
NTs, respectiv
4B (10 mg) w
TA, pH 6.5)
decrease of
of the react
.4) until no
’‐MWCNT co
of functiona
vely.
4B (10 mg) w
DTA, pH 6.5)
decrease of
of the react
.4) until no
v‐MWCNT co
of functiona
vely.
were disper
and the mix
Fab’ conce
ion. Finally,
Fab’ was
onjugates we
alization, det
were disper
and the mix
scFv conce
ion. Finally,
scFv was
onjugates we
alization, det
rsed in 20 m
xture was sh
entration in
the CNTs w
detected in
ere stored at
termined by
rsed in 20 m
xture was sh
entration in
the CNTs w
detected in
ere stored at
termined by
mL of
aken
the
were
n the
t 4 °C
TGA,
mL of
aken
the
were
n the
t 4 °C
TGA,
MW
297
soni
reac
proc
‐ 1 m
‐ 2 m
‐ 60
Then
DMF
MW
adde
was
und
Kais
MW
succ
soni
Reac
and
WCNT‐7
Pristine
µmol), and
ication in a q
ction was pe
cedure, in or
min, 120°C
min, 140°C
min, 160°C
n, CNTs were
F, MeOH and
WCNT‐8
MWCNT
ed. After so
hed by filtra
er high vacu
er test, was
WCNT‐9
MWCNT
cinic anhydr
icated for 20
ction mixtur
finally drie
MWCNTs (1
d p‐HCHO (8
quartz micro
rformed in t
rder to caref
e re‐disperse
d Et2O, to aff
T‐7 (90 mg) w
nication for
tion (pore si
uum, to affo
100 μmol/g.
T‐8 (50 mg,
ide (5 mg, 5
0 min unde
e was filtere
ed under hig
100 mg), N‐
89.3 mg, 2.
owave tube.
he microwav
ully control t
ed in DMF (2
ford, after hi
were dispers
20 min, the
ize = 0.45 μm
ord MWCNT
.
5 µmol of a
50 µmol) an
r Ar, and th
ed (pore size
gh vacuum
‐Boc‐amino‐d
97 mmol) w
Then, the so
ve reactor, w
temperature
200 mL) and
igh vacuum d
sed in EtOH (
e dispersion
m) with EtOH
T‐8. The deg
amine group
nd TEA (7 µ
hen the reac
= 0.45 μm)
to afford M
diethoxy‐eth
were suspen
olvent was e
with a 50 W
e increase:
washed by f
drying the fin
(100 mL) and
was stirred
H, aq. HCl 1N
ree of funct
ps) were dis
µL, 50 µmol)
ction was st
and then wa
MWCNT‐9. T
hylamino‐ace
nded in CH2
evaporated b
power, using
filtration (po
nal CNTs.
d hydrazine h
overnight at
N, H2O, MeOH
tionalization,
persed in dr
) were adde
tirred at r.t.,
ashed with D
The degree
Experime
etic acid 7 (
2Cl2 (4 mL)
by a N2 flow,
g a step‐wise
ore size = 0.4
hydrate (540
t r.t.. The CN
H and Et2O a
, determine
ry DMF (20
ed. The mixt
, under Ar f
DMF, MeOH
of functiona
ental part
135
(100 mg,
by little
, and the
e heating
45 μm) in
0 µL) was
NTs were
and dried
d by the
mL) and
ture was
for 24 h.
and Et2O
alization,
Chapter
136
determi
(Yield: 5
MWCNT
T
solution
µmol) in
min und
subsequ
detected
were dr
µmol of
MWCNT
T
solution
mg, 1.65
under A
washed
filtrate (
under h
doxorub
s‐MWCN
5
ned by com
5%).
T‐10
To a dispers
of PyAOP (2
n dry DMF (2
der Ar, and
uently washe
d in the filtra
ried under h
doxorubicin
T‐11
To a dispers
of PyAOP (
5 µmol) and
Ar, and then
by filtration
(by UV‐vis s
high vacuum
bicin per gram
NT
parison of t
ion of MWC
2.6 mg, 5 µm
2mL) and DIE
then it wa
ed by filtrat
ate (by UV‐v
high vacuum
n per gram of
ion of MWC
2.6 mg, 5 µ
d DIEA (6 µL
it was stirred
n (pore size
spectroscopy
m. The degre
m of CNTs.
he Kaiser te
NT‐9 (15 mg
mol) in dry D
EA (6 µL, 33 µ
as stirred at
ion (pore si
vis spectrosc
m. The degre
f CNTs.
NT‐9 (15 mg
mol) in dry
L, 33 µmol) w
d at r.t. unde
= 0.45 μm)
y), and then
ee of functi
est value wit
g, 0.83 µmol
DMF (1 mL),
µmol) were a
t r.t. under
ze = 0.45 μ
opy), and th
ee of functio
g, 0.83 µmol
DMF (1 mL)
were added
er Ar in the
with DMF u
n with MeOH
ionalization,
th the one o
of carboxyl
a solution o
added. The m
Ar in the
μm) with DM
hen with MeO
onalization, d
of carboxyl
, a solution
. The mixtur
dark overnig
until no doxo
H and Et2O
determined
of MWCNT‐8
groups) in d
of doxorubici
mixture was
dark overni
MF until no
OH and Et2O
determined
groups) in d
of GFLG‐dox
re was sonic
ght. CNTs we
orubicin was
and finally
d by TGA, w
8, was 55 μm
ry DMF (3 m
in∙HCl (2.4 m
sonicated fo
ight. CNTs w
doxorubicin
O and finally
by TGA, wa
ry DMF (3 m
xorubicin 10
cated for 20
ere subseque
s detected in
they were d
was 45 µmo
mol/g
mL), a
mg, 4
or 20
were
was
they
as 75
mL), a
0 (1.5
0 min
ently
n the
dried
ol of
mixt
10 h
volu
H2O
high
s‐MW
µmo
und
µmo
over
doxo
and
TGA
s‐MW
µmo
und
27 µ
over
doxo
and
TGA
Pristine
ture (3:1 v/v
h, keeping te
ume of 150 m
until filtrate
h vacuum (we
WCNT‐12
To a disp
ol) in dry DM
er Ar for 20
ol) in dry DM
rnight. CNTs
orubicin was
finally they
A, was 55 µm
WCNT‐13
To a disp
ol) in dry DM
er Ar for 20
µmol) in dry
rnight. CNTs
orubicin was
finally they
A, was 40 µm
MWCNTs (
v, 98% and 6
emperature
mL and the m
e reached a
eight yield: 8
persion of s‐
MF (1 mL) and
min. Then,
MF (2 mL) w
were subse
s detected in
were dried
mol of doxoru
persion of s‐
MF (1 mL) and
min. Then, a
DMF (2 mL)
were subse
s detected in
were dried
mol of doxoru
100 mg) we
5%, respecti
below 50°C.
mixture was
neutral pH,
80%).
MWCNT (15
d DIEA (5 µL,
a solution o
was added, a
equently was
n the filtrate
under high v
ubicin per gra
MWCNT (15
d DIEA (5 µL,
a solution of
was added,
equently was
n the filtrate
under high v
ubicin per gra
ere disperse
ively), and th
Deionized w
filtered (por
then in Me
5 mg) in dry D
, 27 µmol) w
of doxorubici
and the reac
shed by filtra
(by UV‐vis s
vacuum. The
am of CNTs.
5 mg) in dry D
, 27 µmol) w
f GFLG‐doxor
, and the rea
shed by filtra
(by UV‐vis s
vacuum. The
am of CNTs.
ed in 50 mL
he mixture w
water was th
re size = 0.1
OH and Et2O
DMF (3 mL),
were added a
in∙HCl (1.6 m
ction was sti
ation (pore s
pectroscopy
e degree of f
DMF (3 mL),
were added a
rubicin 10 (2
action was st
ation (pore s
pectroscopy
e degree of f
L of aq. sulf
was sonicated
hen carefully
μm), and wa
O. Finally CN
a solution o
nd the mixtu
mg, 2.7 µmol
rred at r.t. u
size = 0.1 μm
y), and then w
functionaliza
a solution o
nd the mixtu
2.5 mg, 2.7 µ
tirred at r.t.
size = 0.1 μm
y), and then w
functionaliza
Experime
furic acid/ni
d in a water
y added up t
ashed by filt
NTs were drie
of PyAOP (4.3
ure was stirr
l) and DIEA
under Ar in
m) with DMF
with MeOH
ation, determ
of PyAOP (4.3
ure was stirr
µmol) and DI
under Ar in
m) with DMF
with MeOH
ation, determ
ental part
137
itric acid
r bath for
to a final
tration in
ed under
3 mg, 8.2
red at r.t.
(5 µL, 27
the dark
F until no
and Et2O
mined by
3 mg, 8.2
red at r.t.
EA (5 µL,
the dark
F until no
and Et2O
mined by
Chapter 5
138
5.4 References
1. G. Pastorin, W. Wu, S. Wieckowski, J.‐Paul Briand, K. Kostarelos, M. Prato, and A. Bianco,
Chem. Commun., 2006, 1182‐1184.
2. K. Kordatos, T. D. Ros, S. Bosi, E. Vazquez, M. Bergamin, C. Cusan, F. Pellarini, V.
Tomberli, B. Baiti, D. Pantarotto, V. Georgakilas, G. Spalluto, and M. Prato, J. Org. Chem.,
2001, 66, 4915‐4920.