Combinatorial drug conjugation enables nanoparticle dual-drug delivery

Article abstract of DOI:10.1002/smll.201000631

A new approach to loading multiple drugs onto the same drug-delivery nanocarrier in a precisely controllable manner, by covalently preconjugating multiple therapeutic agents through hydrolyzable linkers to form drug conjugates, is reported. In contrast to loading individual types of drugs separately, this drug-conjugates strategy enables the loading of multiple drugs onto the same carrier with a predefined stoichiometric ratio. The cleavable linkers allow the therapeutic activity of the individual drugs to be resumed after the drug conjugates are delivered into the target cells and unloaded from the delivery vehicle. As a proof of concept, the synthesis and characterization of paclitaxel-gemcitabine conjugates are demonstrated. The time-dependent hydrolysis kinetics and cytotoxicity of the combinatorial drug conjugates against human pancreatic cancer cells are examined. It is shown that the synthesized drug conjugates can be readily encapsulated into a lipid-coated polymeric drug-delivery nanoparticle, which significantly improves the cytotoxicity of the drug conjugates as compared to the free drug conjugates.

Full text of DOI:10.1002/smll.201000631

full papers
Drug delivery
Combinatorial Drug Conjugation Enables Nanoparticle
Dual-Drug Delivery
Santosh Aryal, Che-Ming Jack Hu, and Liangfang Zhang*
A new approach to loading multiple drugs onto the same drug-delivery
Keywords:
ꢁ cancer therapy
nanocarrier in a precisely controllable manner, by covalently preconju-
gating multiple therapeutic agents through hydrolyzable linkers to form
drug conjugates, is reported. In contrast to loading individual types of drugs
separately, this drug-conjugates strategy enables the loading of multiple
drugs onto the same carrier with a predefined stoichiometric ratio. The
cleavable linkers allow the therapeutic activity of the individual drugs to be
resumed after the drug conjugates are delivered into the target cells and
unloaded from the delivery vehicle. As a proof of concept, the synthesis and
characterization of paclitaxel–gemcitabine conjugates are demonstrated.
The time-dependent hydrolysis kinetics and cytotoxicity of the combina-
torial drug conjugates against human pancreatic cancer cells are examined.
It is shown that the synthesized drug conjugates can be readily encapsulated
into a lipid-coated polymeric drug-delivery nanoparticle, which significantly
improves the cytotoxicity of the drug conjugates as compared to the free
drug conjugates.
ꢁ conjugation
ꢁ drug delivery
ꢁ lipid–polymer hybrids
ꢁ nanoparticles
kinetics and cellular uptake of various drug molecules, which
willallowtheprecisecontrolofthedosageandschedulingofthe
multiple drugs, thereby maximizing the combinatorial effects.
One of the most popular approaches to overcoming this
challenge is to load multiple types of therapeutic agents onto a
singledrug-deliveryvehicleandthenconcurrentlydeliverthem
to the sites of action.[4–7] Several drug-delivery systems, such as
polymeric nanoparticles and liposomes, have shown the ability
to co-deliver multiple drugs, but fine-controlling the compara-
tive loading yield and release kinetics of the multiple-drug
payloads remains an unmet need. Successfully addressing such
a need will advance current nanomedicine research and enable
more applications of combinatorial drug therapy.
1. Introduction
Combined therapy with two or more drugs provides a
promising strategy to suppress cancer-drug resistance, as
different drugs may attack cancer cells at varying stages of
their growth cycles.[1] It has been reported in clinics that a
variety of drug combinations can induce synergisms among
them and prevent disease recurrence.[2,3] However, one major
challenge of combinatorial therapy is to unify the pharmaco-
ꢀ
[ ] Prof. L. Zhang, Dr. S. Aryal
Department of Nanoengineering
University of California San Diego, La Jolla, CA 92093 (USA)
E-mail: zhang@ucsd.edu
Herein, we report a combinatorial drug-conjugation
strategy to meet the aforementioned need by covalently
conjugating multiple therapeutic agents through hydrolyzable
linkersto formdrug conjugatesprior toloading the drugsontoa
delivery vehicle. In contrast to loading individual types of drugs
separately, this drug-conjugates approach enables multiple
drugs to be loaded onto the same drug carrier with a predefined
stoichiometric ratio. The cleavable linkers allow the therapeu-
tic activity of the individual drugs to be resumed after the drug
conjugates aredeliveredintothe targetcellsandunloadedfrom
the delivery vehicles. In this study, paclitaxel (PTXL) and
Prof. L. Zhang, Dr. S. Aryal, C.-M. J. Hu
Moores Cancer Center
University of California San Diego, La Jolla, CA 92093 (USA)
C.-M. J. Hu
Department of Bioengineering
University of California San Diego, La Jolla, CA 92093 (USA)
: Supporting Information is available on the WWW under http://
www.small-journal.com or from the author.
DOI: 10.1002/smll.201000631
1442
ß 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2010, 6, No. 13, 1442–1448

Nanoparticle Dual-Drug Delivery
gemcitabinehydrochloride(GEM), twowidelyusedanticancer and
4-(N,N-dimethylamino)pyridinium-4-toluenesulfonate
chemotherapy drugs with completely different mechanisms of (DPTS), which resulted in the formation of compound 2.
1
action, were chosen as model drugs.[8,9] The vast difference in
The production of compound 2 was first confirmed by H
hydrophobicity between PTXL (water insoluble) and GEM NMR spectroscopy with all the characteristic peaks and their
(water soluble) also makes them a pair of representative integration values of PTXL and GEM, respectively, as
hydrophobic and hydrophilic drugs, which imposes some indicated in Figure 2A. The 20-OH reaction was confirmed
additional difficulties in the conjugation reaction as compared by the integration value of 14H for the resonance peaks at
to two drugs with close hydrophobicity. We achieved the d ¼ 2.7–2.2 ppm. These peaks correspond to the methyl protons
synthesis, characterization, and delivery of PTXL–GEM drug of acetate groups at C-4 and C-10, the methylene protons at the
conjugates and evaluated their concentration and time- C-14 position of the PTXL, and the methylene protons of the
dependent cytotoxicity against human pancreatic cancer cells GA linker. The resonance at d ¼ 2.7–2.2 ppm of unmodified
and their hydrolysis process. Note that the emphasis of this PTXL was integrated as 8H, which increased to 14H after the
article is to present a unique chemical approach that enables conjugation with GA because of the addition of 6H of the
nanoparticle-based combinatorial drug delivery, rather than methylene group from the GA moiety. In addition, the
demonstrating the synergistic effects between the two drugs d ¼ 4.4 ppm peak of the protons at the C-7 position of PTXL
chosen.
remained intact during the conjugation. This further indicated
that the PTXL–GA reaction only occurred at the 20-OH group,
as a downfield shifting of the C-7 proton would have appeared if
the 7-OH reaction had taken place. In contrast, a significant
downfield shifting from d ¼ 4.7 to 5.5 ppm was observed for the
protons at the C-20 position. On the other hand, the use of GEM
as its hydrochloride salt gives exclusive access to its hydroxyl
group, which is thus prone to couple with the carboxyl group in
the PTXL–GA to form an ester bond. In addition, it has been
reported that DIPC and DPTS are effective esterification
reagentswithhighreactionyield.[10] Furthermore, thechemical
shift associated with the NH2 protons of the pyrimidine ring at
9.0 ppm was intact after the reaction. This further confirms that
thePTXL–GEMconjugationoccurredviaesterformation.The
resulting compound 2 was further examined by high-resolution
mass spectrometry to determine its mass and molecular
formula. As shown in Figure 2B, the results were precisely
consistent with the expected formula of PTXL–GEM
conjugates.
2. Results and Discussion
Figure1illustratesthesynthesisschemeofthePTXL–GEM
conjugate (compound 2). We first took advantage of the steric
hindrance structural chemistry of PTXL to selectively convert
its 20 hydroxyl group (20-OH) to a carboxyl moiety (compound
1). PTXL has three hydroxyl groups, of which two are
secondary and one is tertiary. It has been reported that the
tertiary hydroxyl group is highly hindered and unreactive.[10–12]
The secondary hydroxyl group at the 7-position (7-OH) is less
reactive than that at the 20-position. Typically, one has to
protect the 20-OH to make any modification to the 7-OH
group.[10,13] Here, we used glutaric anhydride (GA) in a
reaction with PTXL in the presence of a catalytic amount of
N,N-dimethylaminopyridine (DMAP) for 3 h at room tem-
perature to selectively modify the 20-OH group, which resulted
in compound 1 as characterized in Figures S1–S3 (see
Supporting Information). We observed that the reaction had
tobelimitedto3 hwithaGA/PTXLmolarratioof3:1forthe20-
OH reaction, otherwise (longer reaction time or higher GA/
PTXL ratio) a 7-OH reaction occurred. Compound 1 was then
reacted with GEM using 1,3-diisopropyl carbodiimide (DIPC)
As the ultimate goal of this research is to concurrently
deliver two drugs to the same cancer cells for combinatorial
therapy, it is crucial to ascertain that the linker bridging the
drugs can be effectively hydrolyzed, thereby releasing the
individual drugs to allow them to arrest cancer cells in their
Figure 1. Synthesis scheme of the paclitaxel (PTXL) and gemcitabine hydrochloride (GEM) conjugate (compound 2).
small 2010, 6, No. 13, 1442–1448
ß 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 1443

full papers
L. Zhang et al.
Figure 2. Characterization of PTXL–GEM conjugates by A) 1H NMR
spectroscopy showing the characteristic protons, and B) high-resolution
mass spectrometry determining the exact mass and corresponding
molecular formula of the drug conjugates. HR-ESI-FT-MS ¼ high-
resolution electrospray-ionization Fourier transform mass spectrometry.
independent pathways. The hydrolysis of PTXL–GEM con-
jugates was evaluated and confirmed by high-performance
liquid chromatography (HPLC) and high-resolution mass
spectrometry. As shown in Figure 3A, the chromatogram
clearly indicates that after 24 h of incubation in water/
acetonitrile (75:25, v/v) solution at pH 7.4, a portion of the
PTXL–GEM conjugate was hydrolyzed to free PTXL and free
GEM with characteristic HPLC retention times of 6.2 and
1.8 min, respectively, which were confirmed by measuring the
mass of the compounds collected at these two retention times
(see Figures S4 and S5 in the Supporting Information for the
corresponding mass spectra).
Theformationof freePTXLandfreeGEM uponhydrolysis
provided further evidence that PTXL–GEM conjugation
occurred via the coupling of the hydroxyl and carboxyl groups
to form an ester bond. If the reaction had occurred via amide
formation between the NH2 group of the pyrimidine ring and
the carboxyl group, free PTXL and free GEM would not have
been released upon hydrolysis within only 24 h. We hypothe-
sized that when these PTXL–GEM conjugates are delivered to
target cells by a drug carrier through endocytosis, the
conjugates can be hydrolyzed at a faster rate in the mild acidic
Figure 3. Hydrolysis and cellular cytotoxicity of PTXL–GEM conjugates.
A) HPLC traces of PTXL–GEM conjugates a) before and b) after 24 h of
incubation in water/acetonitrile (75:25, v/v) solution at pH 7.4. B)
Hydrolysis kinetics of PTXL–GEM conjugates at pH 6.0 and 7.4.
C) Time-dependent comparative cytotoxicity of PTXL–GEM conjugates
with the corresponding mixture of free PTXL and free GEM at 100 nM
concentration against XPA3 human pancreatic cancer cell line (n ¼ 8).
endosomalenvironment(pH ꢂ 6).[14–17] Totestthishypothesis, drug conjugates were hydrolyzed to free PTXL and free GEM
we measured the hydrolysis kinetics of the PTXL–GEM at pH 6.0 within the first 10 h, whereas less than 25% were
conjugates at pH 6.0 and 7.4. As shown in Figure 3B, the cleaved at pH 7.4.
hydrolysis rate was significantly faster in acidic environments
Next we examined the in vitro cellular cytotoxicity of free
(pH 6.0) than in neutral solution (pH 7.4). Nearly 80% of the PTXL–GEM conjugates. As both PTXL and GEM are potent
1444 www.small-journal.com
ß 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2010, 6, No. 13, 1442–1448

Nanoparticle Dual-Drug Delivery
chemotherapy drugs against pancreatic cancer, we chose drugs, and their cytotoxicity against human pancreatic cancer
human pancreatic cancer cell line XPA3 for this study.[18] cell line XPA3, we next loaded the PTXL–GEM conjugates
Since it has been documented that the 20-OH group is essential into a recently developed lipid-coated polymeric nanoparticle
for the high cytotoxicity of PTXL,[13] it is natural to expect that to validate the feasibility of using this preconjugation approach
the cytotoxicity profile of PTXL–GEM conjugates will rely on to enable nanoparticle dual-drug delivery. The PTXL–GEM
their hydrolysis process. To test this, we evaluated the conjugates were mixed with poly(lactic-co-glycolic acid)
cytotoxicity of the drug conjugates (100 nM concentration) at (PLGA) in an acetonitrile solution, which was subsequently
different hydrolysis durations, with a mixture of free PTXL added to an aqueous solution containing lipid and lipid–
(100 nM)andfreeGEM(100 nM)asapositivecontrol. Asshown polyethylene glycol conjugates to prepare lipid-coated PLGA
in Figure 3C, a large cytotoxicity difference was observed nanoparticles following a previously published protocol.[19]
between the drug conjugates andthe free drug mixtures after24
Figure 4A shows a schematic representation of nanopar-
and 48 h of incubation, during which the drug conjugates were ticles loaded with PTXL–GEM conjugates, which are spherical
partially hydrolyzed. For example, the drug conjugates killed particles as imaged by scanning electron microscopy (SEM;
ꢂ15% of XPA3 cells whereas the drug mixtures killed ꢂ55% of Figure 4B). Dynamic light scattering (DLS) measurements
the cells after 24 h of incubation. However, after 72 h of showed that the resulting conjugate-loaded nanoparticles had a
incubation, the cytotoxicity of the PTXL–GEM conjugates was unimodal size distribution with an average hydrodynamic
nearly at the same level as that of the free PTXL and free GEM diameter of 70 ꢃ 1.5 nm (Figure 4C), which was consistent with
mixtures; over 80% of the cells were killed with both systems. the findings from the SEM image (Figure 3B). The surface zeta
This time-dependent cytotoxicity is consistent with the potential of the drug-loaded nanoparticles in water was about
temporal hydrolysis profile of the PTXL–GEM conjugates ꢄ53 ꢃ 2 mV (Figure 4C). We further found that the size and
at pH 7.4 measured by HPLC, as shown in Figure 3B. It is worth surface zeta potential of the PTXL–GEM conjugate-loaded
noting that small-molecule drugs such as PTXL, GEM, and nanoparticles were similar to those of the corresponding empty
PTXL–GEM conjugate can usually diffuse across the cell nanoparticles, that is, 70 ꢃ 1 nm and ꢄ51 ꢃ 2 mV, respectively.
membranes to the inside of cells without going through the ThissuggeststhattheencapsulationofPTXL–GEMconjugates
endocytosis mechanism. Therefore, the hydrolysis process of has a negligible effect on the formation of the lipid-coated
PTXL–GEM conjugates follows the pH ¼ 7.4 profile when the polymeric nanoparticles.
drug conjugates are administered directly without using a drug-
delivery vehicle.
The encapsulation yield and loading yield of PTXL–GEM
conjugates in the nanoparticles were quantified by HPLC after
After having demonstrated the formation of PTXL–GEM dissolving the particles in organic solvents to free all
drug conjugates, their spontaneous hydrolysis to individual encapsulated drugs. When the initial PTXL–GEM conjugate
Figure 4. Characterization of PTXL–GEM conjugate-loaded lipid-coated polymeric nanoparticles (NPs). A) Schematic illustration of a PTXL–GEM
conjugate-loadednanoparticle.B)RepresentativeSEMimageofPTXL–GEMconjugate-loadednanoparticles.C)Diameterandsurfacezetapotentialof
PTXL–GEM conjugate-loaded nanoparticles and empty nanoparticles measured by DLS.
small 2010, 6, No. 13, 1442–1448
ß 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 1445

full papers
L. Zhang et al.
input was 5, 10, and 15 wt% of the total nanoparticle weight, the
The cytotoxicity of PTXL–GEM conjugate-loaded nano-
drug encapsulation yield was 22.8 ꢃ 2.0, 16.2 ꢃ 0.5, and particles against XPA3 cell lines was then examined in
10.8 ꢃ 0.7%, respectively, which can be converted to the comparison with free PTXL–GEM conjugates. Figure 5B
corresponding final drug loading yield of 1.1, 1.6, and 1.6 wt%, summarizes the results of IC50 measurements of the conjugate-
respectively (Figure 5A). Here, the drug encapsulation yield is loaded nanoparticles and free PTXL–GEM conjugates for 24 h
defined as the weight ratio of the encapsulated drugs to the of incubation with the cancer cells. It was found that the IC50
initialdruginput. Thedrugloadingyieldisdefinedastheweight value of the PTXL–GEM conjugates decreased by a factor of
ratio of the encapsulated drugs to the entire drug-loaded 200 for XPA3 cells after loading the drug conjugates into the
nanoparticles including both excipients and bioactive drugs. It lipid-coated polymeric nanoparticles. This enhanced cytoto-
seemed that the maximum PTXL–GEM conjugate loading xicity of the PTXL–GEM conjugates upon nanoparticle
yield was about 1.6 wt% for the lipid-coated polymeric encapsulation can be explained, at least partially, by the fact
nanoparticles. This value can be converted to roughly 1700 that nanoparticle drug delivery can suppress cancer-drug
PTXL–GEM drug conjugate molecules per nanoparticle, by resistance.[20,21] Small-moleculechemotherapydrugsthatenter
calculations with the diameter of the nanoparticle (70 nm), the cells through either passive diffusion or membrane transloca-
PLGA density (1.2 g mLꢄ1), and the molecular weight of the tors are rapidly vacuumed out of the cells before they can take
PTXL–GEM conjugate (1212 Da).
effect by transmembrane drug efflux pumps, such as P-
glycoprotein (P-gp).[22] Drug-loaded nanoparticles, however,
can partially bypass the efflux pumps as they are internalized
through endocytosis.[23] Once engulfed by the plasma mem-
brane, nanoparticles are transported by endosomal vesicles
before unloading their drug payloads. Thus, drug molecules are
released farther away from the membrane-bound drug efflux
pumps and therefore are more likely to reach and interact with
their targets. The endocytic uptake mechanism is particularly
favorable to the combinatorial drug-delivery system presented
in this study. The pH drop upon endosomal maturation into
lysosomes[14] will subject the drug conjugates to a more acidic
environment and more hydrolase enzymes,[24,25] which will
facilitate the cleavage of the hydrolyzable linkers. Moreover,
the degradation of the polymer PLGA will also contribute to
lowering of the pH value surrounding the nanoparticles,[26]
which can accelerate the hydrolysis process of the drug
conjugates as well. The enhanced hydrolysis of the conjugate
linkers may also be partially responsible for the near 200-fold
cytotoxicity increase of PTXL–GEM conjugates after being
encapsulated in the nanoparticles.
While the focus of this article is to report a novel chemical
approach to loading two chemotherapy drugs into a single
nanoparticle for combinatorial drug delivery, it would be
interesting to compare the cytotoxicity of PTXL–GEM
conjugate-loaded nanoparticles with that of a cocktail mixture
of the same type of nanoparticles containing either free PTXL
or free GEM. However, the vast difference in hydrophobicity
(or solubility) between PTXL and GEM makes it practically
impossibletoloadthemintothesametypeofnanoparticle, such
as the lipid-coated polymeric nanoparticles used in this study.
These nanoparticles can encapsulate hydrophobic drugs such
as PTXL with high encapsulation and loading yields but can
barely encapsulate hydrophilic drugs such as GEM. In fact,
the inability to load different drugs into the same type of
nanoparticle represents a generic challenge to many pairs of
drugs for combination therapy. The work presented herein may
offer a new way to overcome this challenge.
Figure 5. A) PTXL–GEM conjugate loading yield at various initial weight
ratios of PTXL–GEM conjugates/excipient (PLGA polymer). B) Cellular
cytotoxicity of PTXL–GEM conjugate-loaded nanoparticles and free
PTXL–GEM conjugates (compound 2) at various drug-conjugate
concentrations against XPA3 human pancreatic cancer cell line. All
samples were incubated with cells for 24 h, and the cells were
subsequently washed and incubated in media for a total of 72 h before
assessment of cell viability in each group (n ¼ 8).
3. Conclusions
We have demonstrated the conjugation of PTXL and GEM
with a stoichiometric ratio of 1:1 via a hydrolyzable ester linker,
1446 www.small-journal.com
ß 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2010, 6, No. 13, 1442–1448

Nanoparticle Dual-Drug Delivery
and have subsequently loaded the drug conjugates into lipid- (d, 1H), 7.3–7.4(m, 7H), 7.5 (m, 3H), 7.6 (m, 1H), 7.74 (d, 2H),
coated polymeric nanoparticles. The cytotoxicity of the
resulting combinatorial drug conjugates against human cancer
cells was comparable to that of the corresponding free PTXL
and GEM drug mixtures after the conjugates were hydrolyzed.
The cytotoxicity of the drug conjugates was significantly
improved after their encapsulation into drug-delivery nano-
particles. This work provides a new method to load dual drugs
into the same drug-delivery vehicle in a precisely controllable
manner, whichholdsgreat promise for suppressingcancer-drug
resistance. A similar strategy may be generalized to other drug
combinations. The synthesis of combinatorial drug conjugates
with a broad range of stoichiometric ratios is currently ongoing
in our laboratory.
8.13 (d, 2H), 8.6 ppm (s, 1H); ESI-MS (positive): m/z: 990.29
[M þ Na]þ (Figure S3, Supporting Information).
Synthesis of PTXL–GEM conjugate (compound 2): Compound 1
(5 mg, 5.2 mmol) was dissolved in dry DCM (0.5 mL) containing
DPTS (4.6 mg, 15.6 mmol). A solution of GEM (1.5 mg, 5.2 mmol)
dissolved in dry N,N-dimethylformamide (DMF; 0.5 mL) was added
and the solution was stirred for 15 min. Then DIPC (5 mg, 39 mmol)
in pyridine (0.1 mL) was added slowly to the solution and reaction
was carried out at room temperature for 24 h. The reaction was
monitored by TLC with CHCl3/MeOH (9.2:0.8, v/v) as eluent
(product Rf ¼ 0.22). The complete disappearance of the starting
compound 1 (Rf ¼ 0.42) occurred after 24 h. The reaction was then
quenched by diluting the solution with DCM, followed by
extracting DPTS, DIPC, DMF, and pyridine with DI water. The
remaining DCM solution was concentrated and precipitated in
hexane, which resulted in compound 2 as a white powder (6.1 mg,
yield about 86%). The crude product was purified by HPLC using
acetonitrile/water (50:50, v/v) as eluent. NMR spectroscopy was
carried out to characterize the produced compound 2 (Figure 2A).
1H NMR (CDCl3): d ¼ 0.91 (s, 1H), 1.14 (s, 3H), 1.22 (s, 3H), 1.27(s,
3H), 1.62 (s, 7H), 1.67 (s, 3H), 1.9–1.2 (br, 8H), 2.2–2.7 (br, 14H),
2.89 (d, 2H), 3.7 (d, 2H), 3.85 (d, 2H), 3.9 (d, 1H), 4.32 (d, 1H),
4.48 (t, 1H), 5.0 (d, 1H), 5.5 (d, 1H), 5.69 (d, 1H), 6.0 (d, 1H), 6.3
(br, 3H) 7.28 (s, 3H), 7.4 (m, 5H), 7.5 (m, 3H), 7.6 (m, 1H), 7.74 (d,
2H), 8.13 (d, 2H), 8.75 (d, 1H), 9.1 ppm (NH2, pyrimidine ring). The
mass and molecular formula of compound 2 were determined by
HR-ESI-FT-MS (orbit-trap MS, positive): m/z: 1213.4327 [M þ H]þ,
1235.4140 [M þ Na]þ; calcd for C61H66F2N4O20: 1213.4311;
found: 1213.4327 (Figure 2B).
4. Experimental Section
Materials and instrumentation: Paclitaxel (PTXL) and gemcita-
bine hydrochloride (GEM) were purchased from ChemiTek Co. and
used without further purification. All other materials including
solvents were purchased from Sigma–Aldrich Co., USA. A single-
addition luminescence adenosine 50-triphosphate (ATP) detection
assay for cytotoxicity measurement was purchased from
PerkinElmer Inc. 1H NMR spectra were recorded in CDCl3 using a
Varian Mercury 400 MHz spectrometer. Electrospray ionization
mass spectrometry (ESI-MS, Thermo LCQdeca spectrometer) and
mass spectrometry (Thermo Fisher Scientific LTQ-XL Orbitrap
spectrometer) were used to determine the mass and molecular
formula of the compounds, respectively. Reversed-phase HPLC
purification was performed on a Varian HPLC system equipped
Hydrolysis of PTXL–GEM conjugate (compound 2): A hydrolysis
study of PTXL–GEM conjugate was performed to confirm that it
could be hydrolyzed to free PTXL and free GEM, and to measure
the hydrolysis kinetics at different pH values. PTXL–GEM
conjugate was incubated in aqueous solutions with a pH value
of 6.0 or 7.4 at 37 8C. At each predefined time interval, an aliquot
of the conjugate solution was collected and subjected to HPLC
(mobile phase: acetonitrile/water, 50:50, v/v) to determine the
amount of free PTXL, free GEM, and the remaining PTXL–GEM
conjugate.
with
a
m-Bonapack C18 column (4.6 ꢅ 150 mm, Waters
Associates, Inc.) with acetonitrile/water (50:50, v/v) as mobile
phase. Thin-layer chromatography (TLC) measurements were
carried out using precoated silica-gel HLF250 plates (Advenchen
Laboratories, LLC, USA). DPTS was prepared by mixing saturated
tetrahydrofuran (THF) solutions of DMAP (1 equiv) and p-
toluenesulfonic acid monohydrate (1 equiv) at room temperature.
The precipitate was isolated by filtration, washed three times with
THF, and dried under vacuum.
Preparation of drug-loaded nanoparticles: Drug-loaded nano-
particles were prepared by a nanoprecipitation process. In a
typical experiment, lecithin (0.12 mg, Alfa Aesar Co.) and 1,2-
distearoyl-sn-glycero-3-phosphoethanolamine-N-[carboxy(po-
lyethylene glycol)-2000] (0.259 mg, DSPE-PEG-COOH, Avinti Polar
Lipids Inc.) was dissolved in 4% ethanol, homogenized to
combine the components, and heated at 68 8C for 3 min. PLGA
(1 mg, Mn ¼ 40 kDa) and a calculated amount of drug dissolved in
acetonitrile were added dropwise to the solution while heating
and stirring. Then the vial was vortexed for 3 min followed by the
addition of water (1 mL). The solution mixture was stirred at room
temperature for 2 h, washed using an Amicon Ultra centrifugal
filter (Millipore, Billerica, MA) with a molecular-weight cutoff of
10 kDa, and drug-loaded nanoparticles (1 mL) were collected. Bare
nanoparticles were prepared similarly in the absence of drugs. The
nanoparticle size and surface zeta potential were obtained from
three repeat measurements by DLS (Malvern Zetasizer, ZEN 3600)
Synthesis of compound 1: PTXL (5 mg, 5.8 mmol) and GA (2 mg,
17.5 mmol) were dissolved in dry pyridine (200 mL). DMAP
(0.57 mmol) dissolved in pyridine (10 mL) was added and the
solution was stirred at room temperature for 3 h. The reaction was
monitored by TLC using CHCl3/MeOH (9.2:0.8, v/v) as eluent
(product Rf ¼ 0.42). The complete disappearance of the starting
PTXL (Rf ¼ 0.54) occurred after 3 h of reaction. Then the reaction
was quenched by diluting the solution with dichloromethane
(DCM), followed by extracting DMAP and pyridine with deionized
(DI) water. The remaining DCM solution was concentrated and
precipitated in hexane, which resulted in compound 1 as a white
powder (5.1 mg, yield about 90%). NMR spectroscopy was carried
out to characterize compound
1 (Figure S2, Supporting
Information). 1H NMR (CDCl3): d ¼ 1.14 (s, 3H), 1.25 (s, 3H),
1.69 (s, 3H), 1.9–2.06 (broad (br), 7H), 2.16–2.27 (br, 4H), 2.2–
2.7 (br, 14H), 3.82 (d, 1H), 4.21 (d, 1H), 4.32 (d, 1H), 4.48 (t, 1H),
5.0 (d, 1H), 5.5 (d, 1H), 5.69 (d, 1H), 6.0 (d, 1H), 6.3 (br, 2H), 7.09 with a backscattering angle of 1738. The morphology and particle
small 2010, 6, No. 13, 1442–1448
ß 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 1447

full papers
L. Zhang et al.
size were further characterized by SEM. Samples for SEM were
prepared by dropping nanoparticle solution (5 mL) onto a polished
silicon wafer. After drying the droplet at room temperature
overnight, the sample was coated with chromium and then
imaged by SEM. The drug loading yield was determined by HPLC.
Cellular viability assay: The cytotoxicity of compound 2 and
PTXL–GEM conjugate-loaded nanoparticles was assessed against
XPA3 human pancreatic carcinoma cell line by using the ATP
assay. First, cells were seeded (2 ꢅ 104) in 96-well plates and
incubated for 24 h. Next, the medium was replaced with fresh
medium (150 mL) and incubated with a different concentration of
compound 2 dissolved in dimethyl sulfoxide (DMSO). The final
concentration of DMSO in each well was kept constant at 2%. The
plates were then incubated for 72 h and measured by ATP
reagents following a protocol provided by the manufacturer. Fresh
cell medium with 2% DMSO was used as negative control. Similar
procedures were applied to compare the cytotoxicity of 100 nM
compound 2 with that of a mixture of free PTXL and GEM at the
corresponding drug concentrations and incubation times of 24,
48, and 72 h. Here, the use of DMSO was only for solubilizing the
free drugs. For the measurement of the cytotoxicity of PTXL–GEM
conjugate-loaded nanoparticles, the experiments were carried out
without using DMSO.
[2] F. Calabro, V. Lorusso, G. Rosati, L. Manzione, L. Frassineti, T. Sava,
E. D. Di Paula, S. Alonso, C. N. Sternberg, Cancer 2009, 115, 2652–
2659.
[3] H. M. McDaid, P. G. Johnston, Clin. Cancer Res. 1999, 5, 215–220.
[4] S. Sengupta, D. Eavarone, I. Capila, G. Zhao, N. Watson,
T. Kiziltepe, R. Sasisekharan, Nature 2005, 436, 568–572.
[5] Y. Wang, S. Gao, W.-H. Ye, H. S. Yoon, Y.-Y. Yang, Nat. Mater. 2006,
5, 791–796.
[6] L. Zhang, A. F. Radovic-Moreno, F. Alexis, F. X. Gu, P. A. Basto,
V. Bagalkot, S. Jon, R. S. Langer, O. C. Farokhzad, ChemMedChem
2007, 2, 1268–1271.
[7] F. Ahmed, R. I. Pakunlu, A. Brannan, F. Bates, T. Minko,
D. E. Discher, J. Controlled Release 2006, 116, 150–158.
[8] S. B. Horwitz, Ann. Oncol. 1994, 5(Suppl 6), S3–S6.
[9] S. B. Horwitz, D. Cohen, S. Rao, I. Ringel, H. J. Shen, C. P. Yang, J.
Natl. Cancer Inst. Monogr. 1993, 15, 55–61.
[10] J. D. Gibson, B. P. Khanal, E. R. Zubarev, J. Am. Chem. Soc. 2007,
129, 11653–11661.
[11] R. Tong, J. Cheng, Angew. Chem. 2008, 120, 4908–4912; Angew.
Chem. Int. Ed. 2008, 47, 4830–4834.
[12] R. Tong, J. Cheng, J. Am. Chem. Soc. 2009, 131, 4744–4754.
[13] N. F. Magri, D. G. Kingston, J. Nat. Prod. 1988, 51, 298–306.
[14] S. Modi, M. G. Swetha, D. Goswami, G. D. Gupta, S. Mayor,
Y. Krishnan, Nat. Nanotechnol. 2009, 4, 325–330.
[15] K. Remaut, B. Lucas, K. Braeckmans, J. Demeester, S. C. De Smedt,
J. Controlled Release 2007, 117, 256–266.
[16] X. Guo, F. C. Szoka, Jr., Bioconjugate Chem. 2001, 12, 291–300.
[17] C. Masson, M. Garinot, N. Mignet, B. Wetzer, P. Mailhe,
D. Scherman, M. Bessodes, J. Controlled Release 2004, 99,
423–434.
Supporting Information: 1H NMR spectra of the starting PTXL
and intermediate compound 1, and the mass spectra of
compound 1, free PTXL, and free GEM recovered from the
hydrolyzed PTXL–GEM conjugate can be found in the Supporting
Information.
[18] S. Kaushal, M. K. McElroy, G. A. Luiken, M. A. Talamini,
A. R. Moossa, R. M. Hoffman, M. Bouvet, J. Gastrointest. Surg.
2008, 12, 1938–1950.
[19] L. Zhang, J. M. Chan, F. X. Gu, J.-W. Rhee, A. Z. Wang, A. F. Radovic-
Moreno, F. Alexis, R. Langer, O. C. Farokhzad, ACS Nano 2008, 2,
1696–1702.
[20] C.-M. J. Hu, L. Zhang, Curr. Drug Metab. 2009, 10, 836–841.
[21] E. M. Leslie, R. G. Deeley, S. P. Cole, Toxicol. Appl. Pharmacol.
2005, 204, 216–237.
Acknowledgements
[22] M. M. Gottesman, Annu. Rev. Med. 2002, 53, 615–627.
[23] J. Rejman, V. Oberle, I. S. Zuhorn, D. Hoekstra, Biochem. J. 2004,
377, 159–169.
This work was supported by the National Institute of Health
(grant U54CA119335) and the University of California San
Diego.
[24] M. Masquelier, R. Baurain, A. Trouet, J. Med. Chem. 1980, 23,
1166–1170.
[25] G. M. Dubowchik, M. A. Walker, Pharmacol. Ther. 1999, 83, 67–
123.
[26] H. Liu, E. B. Slamovich, T. J. Webster, Int. J. Nanomedicine 2006, 1,
541–545.
[1] J. Lehar, A. S. Krueger, W. Avery, A. M. Heilbut, L. M. Johansen,
E. R. Price, R. J. Rickles, G. F. Short, 3rd, J. E. Staunton, X. Jin,
M. S. Lee, G. R. Zimmermann, A. A. Borisy, Nat. Biotechnol. 2009,
27, 659–666.
Received: April 15, 2010
Published online: June 17, 2010
1448 www.small-journal.com
ß 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2010, 6, No. 13, 1442–1448