Choline-derivate-modified nanoparticles for brain-targeting gene delivery

Article abstract of DOI:10.1002/adma.201101899

The choline transporter (ChT) is used to transporte choline-derivate- modified nanoparticles across the blood-brain barrier. It is demonstrated that ChT is an ideal Trojan horse. The choline-derivate-modified nanoparticles exhibit higher permeability across the brain capillary endothelial cells (BCECs) monolayer in vitro and higher gene distribution and expression in vivo. Copyright

Full text of DOI:10.1002/adma.201101899

www.advmat.de
www.MaterialsViews.com
Choline-Derivate-Modified Nanoparticles for
Brain-Targeting Gene Delivery
Jianfeng Li, Lu Zhou, Deyong Ye, Shixian Huang, Kun Shao, Rongqin Huang,
Liang Han, Yang Liu, Shuhuan Liu, Liya Ye, Jinning Lou, and Chen Jiang*
The major obstacle for drug delivery to the central nervous system
(CNS) is the blood–brain barrier (BBB), which protects the CNS
from potentially harmful xenobiotics and endogenous molecules
to ensure an optimal environment for brain function.[1] Despite
this natural barricade, small molecules and macromolecules
including peptides and proteins could be transported into the
CNS to maintain its normal physiological function via the endog-
enous BBB transporters. There are three families of endogenous
BBB transporters: carrier-mediated transporters (CMT), active
efflux transporters (AET), and receptor-mediated transporters
(RMT). The CMT and AET systems are mainly responsible for
the transport of small molecules, while the RMT systems are
responsible for endogenous large molecules.[2,3]
Drugs or drug-delivery systems can be modified with the sub-
strates of these transporters to realize their accumulation in the
CNS. The RMT mechanism has been mainly investigated and
utilized for brain-targeting drug delivery.[2] For example, the lig-
ands of transferrin receptor and insulin receptor have been used
to modify polymers for constructing brain-targeting drug-delivery
systems to transport small molecules, proteins, and gene drugs
into the CNS.[4,5] A number of specific transporters are expressed
on the brain capillary endothelial cells (BCECs), which are
responsible for the endogenous and exogenous nutrient supplies
for the brain. They possess inspiring features including high
transport capacity, selectivity, and an adequate transfer rate.[6]
Thus, a native BBB nutrient transporter could be utilized as an
alternative strategy to enable polar, water-soluble drugs to cross
the BBB via a CMT mechanism. Several successful applications
have been reported. For example, the BBB large neutral amino
acid transport system, the glucose transporter (GluT1), and the
BBB choline transporter (BBB ChT) could be used to deliver D,L-
2-amino-7-bis[(2-chloroethyl)amino]-1,2,3,4-tetrahydro-2-naph-
thoic acid (D,L-NAM), ketoprofen, and N-n-octylnicotinium iodide
(NONI), respectively.[7,8] Despite these successful applications,
the CMT mechanism remains an untapped resource for delivery
systems. In this work, CMT mechanism was challenged to facili-
tate the brain-targeting drug-delivery system across the BBB.
Choline is a biochemical precursor of the neurotransmitter
acetylcholine and other essential components of cell membrane
phospholipids such as phosphatidylcholine. The high choline
concentration in the brain demonstrates the amazing ability of the
BBB ChT to transport choline across the BBB.[9,10] Furthermore,
the low physiologic plasma concentration of choline (approxi-
mately 25% of the Michaelis–Menten constant, Km) makes the
BBB ChT free to transport choline derivates without interrupting
the supply of CNS choline.[11] Therefore, BBB ChT is considered
a potential target for active drug delivery into the CNS.
A quaternary ammonium group and a free hydroxyl
group were suggested as the key requirements for BBB ChT
substrates.[8] As a native BBB ChT substrate, choline was not
suitable for further modification. Recent studies revealed that
bis-quaternary ammonium compounds could also be applied as
BBB ChT substrates with even higher affinity.[12,13] Thus, bis-
quaternary ammonium compounds were challenged to modify
dendrimers for constructing the brain-targeting drug-delivery
system without affecting their BBB ChT affinity.
First, a series of novel bis-quaternary ammonium compounds
with high BBB ChT affinity were designed. There were two major
factors affecting the BBB ChT affinity of these compounds:
i) the lipophilicity of the quaternary ammonium moieties and
ii) the length and conformation rigidity of the linkers between
two quaternary ammonium moieties.[13] Given these factors, high
lipophilic isoquinoline (4a) and relatively low lipophilic 3-methyl-
pyridine (4b) were chosen as quaternary ammonium moieties.
(3,5-bis(3-bromopropoxy)phenyl)methanol (3a, 11 carbons) and
(3,5-bis(4-bromobutoxy)phenyl)methanol (3b, 13 carbons) were
selected as linkers (Scheme 1). High conformation rigidity of the
benzene ring could result in higher BBB ChT affinity and benzylic
hydroxyl could also provide a free reactive site for conjugation to a
brain-targeting drug-delivery system. Based on the above consider-
ations, four bis-quaternary ammonium compounds (5,6,7,8) have
been crossover designed, synthesized (as shown in Scheme 1),
J. F. Li, S. X. Huang, K. Shao, R. Q. Huang, L. Han, Y. Liu, S. H. Liu,
Prof. C. Jiang
Key Laboratory of Smart Drug Delivery
Ministry of Education and PLA Department of Pharmaceutics
School of Pharmacy
Fudan University
826 Zhangheng Road, Shanghai 201203, China
E-mail: jiangchen@shmu.edu.cn
Dr. L. Zhou, Prof. D. Y. Ye
Key Laboratory of Smart Drug Delivery
Ministry of Education and PLA Department of Medical Chemistry
School of Pharmacy
and characterized by mass spectrometry (MS), 1H NMR, and 13
C
Fudan University
826 Zhangheng Road, Shanghai 201203, China
NMR (see Supporting Information). The characterization showed
successful synthesis of compounds 5, 6, 7, and 8.
To verify the design strategy, the BBB ChT affinityiesof the syn-
thesized bis-quaternary ammonium compounds were compared
by evaluation of their ability to inhibit the [3H]-choline chloride
uptake by BCECs (Figure 1a). The IC50 value (concentration
L. Y. Ye, Prof. J. N. Lou
Institute of Clinical Medical Sciences
China–Japan Friendship Hospital
The Ministry of Health, Beijing, China
DOI: 10.1002/adma.201101899
©
4516 wileyonlinelibrary.com
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2011, 23, 4516–4520

www.advmat.de
www.MaterialsViews.com
HO
OH
Br
O
O
Br
a
n
n
Br
Br
+
n
OH
OH
2a =1
n
3a =1
n
1
2b =2
n
3b =2
n
R
R
N+
O
O
N+
Br
O
O
Br
b
Br-
Br-
n
n
n
n
R
+
N
OH
OH
5
=1 R=
=1 R=
=2 R=
=2 R=
n
n
n
n
N
N
6
7
8
4a
R=
N
N
4b
R=
N
N
Scheme 1. Synthesis of bis-quaternary ammonium compounds 5-8. Reagents and conditions: a) K2CO3, ACN, reflux 4 h and b) 65 °C, 1 h.
Figure 1. a) Dose–response relationship for the inhibition of [3H]-choline chloride uptake by various compounds. Uptake of [3H]-choline chloride
(10 nM) into BCECs was measured with a 10-min incubation in the presence of NaCl at pH 7.4 either in the absence (control) or presence of assigned
concentrations of various compounds. Data are expressed as mean standard deviation (S.D.) (number of samples, n = 4). b) Structure–activity
relationship indicated by IC50 values of various compounds. c) In vivo imaging of mice administrated with 50 μM10-BODIPY (right) and free BODIPY
of equal fluorescence (left). d) In vitro imaging of the main organs of mice administered with 50 μM10-BODIPY (low panel) and free BODIPY of equal
fluorescence (upper panel). Images were taken 2 h after administration. Intensity of the signal: dark red is the strongest and dark blue is the weakest,
as shown by the bar.
©
wileyonlinelibrary.com
Adv. Mater. 2011, 23, 4516–4520
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
4517

www.advmat.de
www.MaterialsViews.com
necessary for 50% inhibition) was calculated by One site-Fit
logIC50 (an equation in GraphPad Prism 5). For compound 5,
6, 7, and 8 and choline chloride, the IC50 values were 80.25 μM,
102.9 μM, 4.265 μM, 49.08 μM, and 14.50 μM, respectively. Com-
parison of compound 5 with 6 and 7 with 8 (Figure 1B) indicated
that compounds with isoquinoline as quaternary ammonium
moieties possessed higher affinity. The cationic binding site of
BBB ChT may locate inside a hydrophobic pocket, thus higher
lipophilicity of moieties could gain higher affinity. Comparison
of compound 5 with 7 and 6 with 8 (Figure 1b) suggested that
the optimal length of linker was 13 carbons. It may attribute
to the potential binding sites of quaternary ammonium moie-
ties with BBB ChT: i) one BBB ChT has several cationic binding
sites and two quaternary ammonium cations bind to two of
them and ii) one BBB ChT has only one cationic binding site
and two quaternary ammonium cations bind to two nearby
BBB ChTs. Independent of the binding arrangement, the
results verified that a distance of 13 carbons was more suit-
able. Comparison of the BBB ChT affinity of the synthesized
compounds revealed a simple structure–activity relationship.
The bis-quaternary ammonium compound with isoquinoline
as the quaternary ammonium moieties, which were connected
by a high conformation rigidity linker with 13 carbons (com-
pound 7), had the highest BBB ChT affinity. Among the four
synthesized compounds, compound 7 had the lowest IC50 value
(4.265 μM), which was even lower than that of choline chloride
(14.50 μM). This indicated that compound 7 could compete
efficiently with endogenous choline in binding to BBB ChT
in vivo. Thus compound 7, as a novel BBB ChT substrate,
was further conjugated to dendrimers for constructing the
brain-targeting drug-delivery system.
123 amino groups per molecule) have been utilized for gene
delivery to brain.[14,15] Compound 10 was conjugated to the
surface of the DGL via a-malemidyl-u-N-hydroxysuccinimidyl
polyethyleneglycol (NHS-PEG-MAL) and further constructed
the gene delivery system. Details of the synthesis and character-
ization of DGL-PEG/pDNA (plasmid DNA) and DGL-PEG-10/
pDNA nanoparticles (NPs) are shown in Figure S1 (Supporting
Information).
To determine whether BBB ChT could mediate the brain
accumulation of a gene-delivery system, the cellular uptake
mechanism of NPs with YOYO-1-labeled (Invitrogen, Y3601)
pDNA was investigated. The cellular uptake of DGL-PEG-10/
pDNA (Figure 2b) was much more than that of DGL-PEG/
pDNA (Figure 2a). Excess choline chloride could inhibit the
uptake of DGL-PEG-10/pDNA (Figure 2c). The uptake of DGL-
PEG-10/pDNA at 4 °C significantly decreased (Figure 2f).
These indicated that the uptake showed energy-dependent and
BBB-ChT-involving character. The uptake mechanism for the
NPs is endocytosis. There are two main kinds of endocytosis
on the BBB: i) receptor-mediated endocytosis (RME), in which a
In order to gain more efficient conjugation to dendrimers,
the benzylic hydroxyl of compound 7 was converted to benzylic
thiol. Compound 10 was the sulfhydrylation form of compound
7 (Scheme S1, Supporting Information) and its brain-targeting
efficiency was investigated in vivo by labeling with a fluorescent
agent, BODIPY (4,4-difluoro-5,7-dimethyl-4-bora-3a,4a-diaza-
s-indacene-3-propionic acid, sulfosuccinimidyl ester, sodium
salt). Compound 10-BODIPY and free BODIPY of equal fluores-
cence were injected via the caudal vein of nude mice. As shown
in Figure 1c, the fluorescence in the brain of the mice treated
with compound 10-BODIPY was significantly stronger than
that of the mice treated with free BODIPY. In vitro imaging
results showed that the accumulation of compound 10-BODIPY
in the brain and kidneys increased (Figure 1d). Compound 10-
BODIPY was more hydrophilic for the two cationic moieties
and led to faster excretion, thus it had more accumulation in the
kidneys. Traditionally it was thought that molecules with more
lipophilicity could cross the BBB more easily. It was amazing
that the decreased lipophilicity of compound 10-BODIPY did
not result in less, but instead more, accumulation in brain. This
was due to the high BBB ChT affinity of compound 10. Here the
BBB ChT acted as a Trojan horse to increase the accumulation
of compound 10-BODIPY in brain. This indicated the possibility
that BBB ChT could also mediate the brain accumulation of a
drug-delivery system modified with compound 10.
Figure 2. Cellular uptake of DGL-PEG/pDNA (a), DGL-PEG-10/pDNA
(b) with different inhibitors (c–e) were examined by fluorescent micro-
scopy after a 30-min incubation at 37 °C. BCECs were treated with different
inhibitors including choline chloride (c), filipin complex (d), phenylarsine
(e). The cellular uptake of DGL-PEG-10/pDNA was also carried out at
4 °C (f). Green: YOYO-1 labeled pDNA. Original magnification: ×100.
g) Results of transport studies of NPs across the BCECs monolayer.
Data is shown as Papp at assigned time points. Significance: +p < 0.05;
Next, compound 10 was conjugated to the drug-delivery
system and the brain-targeting efficiency was evaluated in vitro
and in vivo. Dendrigraft poly-L-lysines (DGLs, generation 3 with
++p
<
0.01, represents DGL-PEG-10/pDNA vs. DGL-PEG-10/
pDNA+inhibitor, while ∗∗∗p < 0.001, represents DGL-PEG-10/pDNA vs.
DGL-PEG/pDNA. Results are expressed as mean S.D. (n = 4).
©
4518 wileyonlinelibrary.com
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2011, 23, 4516–4520

www.advmat.de
www.MaterialsViews.com
ligand-modified system binds to its cognate receptor specifically
to elicit cellular internalization and can be inhibited by pheny-
larsine oxide and ii) adsorptive endocytosis (AE), in which the
drug-delivery system will bind to the cellular surface through
non-specific mechanisms, such as electrostatic interactions,
and can be blocked by a filipin complex.[1] It was reported that
the uptake of Angiopep (a high-affinity ligand of low-density
lipoprotein-receptor-related protein) modified NPs could sig-
nificantly be inhibited by phenylarsine oxide.[16] The modifica-
tion of Angiopep gave the NPs RME character. The uptake of
NPs decreased with addition of the filipin complex (Figure 2d)
but not phenylarsine oxide (Figure 2e), thus showing that the
main pathway of endocytosis of was AE not RME. The surface
amino groups were slightly cationic and this contributed to the
AE pathway. However, unlike ordinary AE, the modification of
compound 10 increased the capture possibility of NPs by the
cell membrane. Subsequently, the transport efficiency across
the BCECs monolayer was evaluated. As shown in Figure 2g,
the apparent permeability (Papp) of DGL-PEG-10/pDNA was
much higher than that of DGL-PEG/pDNA, especially in the
first 15 min (DGL-PEG-10/pDNA, 26.47 cm s−1 ; DGL-PEG/
pDNA, 6.61 cm s−1 ). The higher permeability was probably
due to the transporting character of BBB-ChT. When adding
the excess choline chloride as inhibitor, Papp decreased due to
the competition with compound 10. The incorporation of com-
pound 10 could lead to higher permeability across the BCECs
monolayer.
Inspired by the in vitro results, in vivo brain-targeting effi-
ciency of the NPs was investigated 2 h after injection of the
NPs. The accumulation of ethidium monoazide bromide (EMA)
labled pDNA in the brain treated with DGL-PEG-10/pDNA was
more than that treated with DGL-PEG/pDNA (Figure 3a). Then,
distribution of gene expression in mouse brain was detected.
For DGL-PEG/pEGFP, green fluorescent protein (GFP) expres-
sion was found mainly in fourth ventricle (Figure 3g) and
little was found in other regions (Figure 3c–f). It was reported
that exogenous gene expression could be observed in limited
regions around the cerebral ventricles due to their relatively
weak BBB.[17] DGL-PEG/pEGFP had high GFP expression in
fourth ventricle. It could be hypothesized that NPs modified
with compound 10 were likely to accumulate in high-choline-
demanding regions, thus resulting in more GFP expression in
these regions. For DGL-PEG-10/pEGFP, GFP expression was
mainly found in cortical layer (Figure 3h), caudate putamen
(Figure 3i), and fourth ventricle (Figure 3l). The cortical layer
and caudate putamen are rich in neurons and have high
Figure 3. a) In vivo imaging of mice administrated with DGL-PEG/pDNA (left) or DGL-PEG-10/pDNA (right). Images were taken 2 h after administra-
tion. b) Luciferase expression 48 h after intravenous administration of DGL-PEG/pGL3 and DGL-PEG-10/pGL3 into Balb/c mice at a dose of 50 μg
DNA per mouse. Luciferase expression is plotted as light units per mg protein. ∗∗p < 0.01 represents DGL-PEG-10/pGL3 vs. DGL-PEG/pGL3. Data are
expressed as mean S.D. (n = 4). c–l) Distribution of gene expression in brains of mice treated with DGL-PEG/pEGFP (c–g) or DGL-PEG-10/pEGFP
(h–l) 48 h after intravenous administration. Frozen sections (thickness of 20 μm) of cortical layer (c,h), caudate putamen (d,i), hippocampus (e, j),
and substantia nigra (f,k), and fourth ventricle (g,l) were examined by fluorescent microscopy. The sections were stained with 300 nM DAPI for 10 min
at room temperature. Green: GFP. Blue: cell nuclei. Original magnification: ×100.
©
wileyonlinelibrary.com
Adv. Mater. 2011, 23, 4516–4520
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
4519

www.advmat.de
www.MaterialsViews.com
Synthesis and Identify of DGL Derivatives and Derivative/DNA NPs
and Other Bioassay Parts: Experimental details for in vivo imaging of
compound 10-BODIPY, synthesis and identify of DGL derivatives and
derivative/DNA NPs, the cellular uptake mechanism of NPs, transport
studies of NPs across the BCECs monolayer, in vivo imaging of NPs,
the distribution of gene expression in the mouse brain, and in vivo gene
quantitative expression are described in the the Supporting Information.
demand for choline. High-affinity choline transporter-immu-
noreactive cell bodies were demonstrated in these regions.[18]
As a result, compound-10-modified NPs accumulated more in
these regions and had higher GFP expression. To examine the
transfection efficiencies quantitatively, the expression of pGL3
(luciferase reporter vector)-control vector in brain was meas-
ured (Figure 3b). The brain gene expression of DGL-PEG-10/
pGL3 was 8.63 0.61 × 103 light units per mg protein, which is
1.48-fold higher than that of DGL-PEG/pGL3 (5.85 0.26 × 103
light units per mg protein). The significant level (p) is smaller
than 0.01.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
In summary, a series of novel bis-quaternary ammonium
compounds with high BBB-ChT affinity have been synthesized.
Although BBB ChT has not been crystallized, the structure–
activity relationships shown here could inspire design of BBB
ChT substrates with higher affinity. The sulfhydrylation form
of the novel compound 7, i.e., compound 10, was utilized for
the construction of the brain-targeting gene-delivery system.
And DGL-PEG-10/pDNA NPs demonstrated higher uptake effi-
ciency in vitro and higher gene expression in vivo. The BBB
ChT-mediated brain-targeting strategy was first employed in a
drug-delivery system and proved to be an encouraging way to
overcome the BBB.
Acknowledgements
The authors thank Prof. Jianhua Zhu (School of Pharmacy, Fudan
University) for the 125I-labeled work. This work was supported by the
grant from National Basic Research Program (2007CB935802) of
China (973program), National Natural Science Foundation of China
(30973652), National Natural Science Foundation of China (81172993),
and the “Key new drug creation program” 2009ZX09310-006.
Received: May 23, 2011
Published online: September 5, 2011
Experimental Section
[1] M. W. Smith, M. Gumbleton, J. Drug Targeting 2006, 14, 191.
[2] R. Gabathuler, Neurobiol. Dis. 2010, 37, 48.
Synthesis of Compounds 5–8: Compounds 3a and 3b were prepared
by dissolving 3,5-dihydroxy-benzyl alcohol (1, 5 mmol) in acetonitrile,
adding 1,3-dibromopropane (2a, 30 mmol) or 1,4-dibromobutane (2b,
30 mmol), and K2CO3 (30 mmol), then stirring the the solution under
reflux for 4 h. The completion of the reactions was monitored by thin-
layer chromatography (TLC). After filtration and removal of the solvent
via rotary evaporation, the crude products were purified by flash
chromatography on silica (1:4 ethyl acetate/petroleum ether eluent) and
identified by 1H NMR. Compounds 5–8 were prepared by reacting an
excess of isoquinoline (4a) or 3-methylpyridine (4b) with compound 3a
or 3b for 1 h at 65 °C in the absence of solvent. The resulting solids were
collected by filtration and purified by recrystallisation in ethanol.
Cell Line and Animals: BCECs were kindly provided by Prof. J. N. Lou
(the Clinical Medicine Research Institute of the China–Japan Friendship
Hospital). Balb/c nude mice (male, age 4–5 weeks, 20–25 g) were
maintained under standard housing conditions. All animal experiments
were carried out in accordance with guidelines evaluated and approved
by the ethics committee of Fudan University.
[3] W. M. Pardridge, Drug Discovery Today 2007, 12, 54.
[4] C. C. Visser, L. H. Voorwinden, D. J. Crommelin, M. Danhof,
A. G. de Boer, Pharm. Res. 2004, 21, 761.
[5] Z. Laron, Arch. Physiol. Biochem. 2009, 115, 112.
[6] Q. R. Smith, Adv. Exp. Med. Biol. 1993, 331, 83.
[7] M. Gynther, J. Ropponen, K. Laine, J. Leppänen, P. Haapakoski,
L. Peura, T. Järvinen, J. Rautio, J. Med. Chem. 2009, 52, 3348.
[8] W. J. Geldenhuys, P.R. Lockman, T. H. Nguyen, C. J. Van der Schyf,
P. A. Crooks, L. P. Dwoskin, D. D. Allen, Bioorg. Med. Chem. 2005,
13, 4253.
[9] D. D. Allen, Q. R. Smith, J. Neurochem. 2001, 76, 1032.
[10] S. H. Zeisel, J. Am. Coll. Nutr. 2004, 23, 621S.
[11] D. D. Allen, P. R. Lockman, Life Sci. 2003, 73, 1609.
[12] J. T. Ayers, L. P. Dwoskin, A. G. Deaciuc, V. P. Grinevich, J. Zhu,
P. A. Crooks, Bioorg. Med. Chem. Lett. 2002, 12, 3067.
[13] G. Zheng, Z. Zhang, P. R. Lockman, W. J. Geldenhuys, D. D. Allen,
L. P. Dwoskin, P. A. Crooks, Bioorg. Med. Chem. Lett. 2010, 20,
3208.
[14] H. Cottet, M. Martin, A. Papillaud, E. Souaïd, H. Collet,
A. Commeyras, Biomacromolecules 2007, 8, 3235.
[15] R. Q. Huang, S. Liu, K. Shao, L. Han, W. Ke, Y. Liu, J. F. Li, S. Huang,
C. Jiang, Nanotechnology 2010, 21, 265101.
[16] W. L. Ke, K. Shao, R. Q. Huang, L. Han, Y. Liu, J. F. Li, Y. Y. Kuang,
L. Y. Ye, J. N. Lou, C. Jiang, Biomaterials 2009, 30, 6976.
[17] R. Q. Huang, Y. H. Qu, W. L. Ke, J. H. Zhu, Y. Y. Pei, C. Jiang, FASEB
J. 2007, 21, 1117.
[18] H. Misawa, K. Nakata, J. Matsuura, M. Nagao, T. Okuda, T. Haga,
Neuroscience 2001, 105, 87.
Inhibition of [3H]-Choline Chloride Uptake: Incubation buffer (0.25 mL)
containing 10 nmol [3H]-choline chloride dissolved in Hanks buffer with
different compounds at designated concentrations was added to each
well and incubated at 37 °C for 10 min. The unlabeled choline chloride
served as a comparison. Incubation buffer without inhibitions was the
total uptake as a control. The uptake was stopped by aspirating the
incubation buffer and washing the cells three times with ice-cold Hanks
buffer. 2 N NaOH (0.2 mL) was added in each well to lyse the cells and
neutralized with 4 N HCl (0.1 mL). Radioactivity was determined by
adding scintillant liquid (2 mL) to the solubilized cell solution (0.1 mL)
and counting it in a liquid scintillation counter (LS 6000SE, Beckman,
USA).
©
4520 wileyonlinelibrary.com
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2011, 23, 4516–4520