Design, synthesis, and biological functionality of a dendrimer-based modular drug delivery platform

Article abstract of DOI:10.1021/bc100360v

A modular dendrimer-based drug delivery platform was designed to improve upon existing limitations in single dendrimer systems. Using this modular strategy, a biologically active platform containing receptor mediated targeting and fluorescence imaging modules was synthesized by coupling a folic acid (FA) conjugated dendrimer with a fluorescein isothiocyanate (FITC) conjugated dendrimer. The two different dendrimer modules were coupled via the 1,3-dipolar cycloaddition reaction ("click" chemistry) between an alkyne moiety on the surface of the first dendrimer and an azide moiety on the second dendrimer. Two simplified model systems were also synthesized to develop appropriate "click" reaction conditions and aid in spectroscopic assignments. Conjugates were characterized by ¹H NMR spectroscopy and NOESY. The FA-FITC modular platform was evaluated in vitro with a human epithelial cancer cell line (KB) and found to specifically target the overexpressed folic acid receptor.

Full text of DOI:10.1021/bc100360v

ARTICLE
pubs.acs.org/bc
Design, Synthesis, and Biological Functionality of a Dendrimer-Based
Modular Drug Delivery Platform
Douglas G. Mullen,†,‡ Daniel Q. McNerny,‡,^ Ankur Desai,‡ Xue-min Cheng,‡ Stassi C. DiMaggio,§
||
Alina Kotlyar,‡ Yueyang Zhong,‡,# Suyang Qin,‡ Christopher V. Kelly,‡, Thommey P. Thomas,‡
||
||,#
Istvan Majoros,‡ Bradford G. Orr,‡, James R. Baker, Jr.,‡ and Mark M. Banaszak Holl*,†,‡,
†Programs in Macromolecular Science and Engineering and Applied Physics, #Departments of Chemistry and ^Chemical Engineering,
‡Michigan Nanotechnology Institute for Medicine and Biological Sciences, University of Michigan, Ann Arbor, Michigan 48019,
United States
§Department of Chemistry, Xavier University, New Orleans, Louisiana 70125, United States
S Supporting Information
b
ABSTRACT: A modular dendrimer-based drug delivery platform
was designed to improve upon existing limitations in single den-
drimer systems. Using this modular strategy, a biologically active
platform containing receptor mediated targeting and fluorescence
imaging modules was synthesized by coupling a folic acid (FA)
conjugated dendrimer with a fluorescein isothiocyanate (FITC)
conjugated dendrimer. The two different dendrimer modules were
coupled via the 1,3-dipolar cycloaddition reaction (“click”
chemistry) between an alkyne moiety on the surface of the first
dendrimer and an azide moiety on the second dendrimer. Two
simplified model systems were also synthesized to develop appro-
priate “click” reaction conditions and aid in spectroscopic assign-
ments. Conjugates were characterized by 1H NMR spectroscopy and NOESY. The FA-FITC modular platform was evaluated in
vitro with a human epithelial cancer cell line (KB) and found to specifically target the overexpressed folic acid receptor.
’ INTRODUCTION
toward or through clinical trials for cancer treatments with promising
results.2 Each approach, however, is not without limitations and the
potential for widespread application of these platforms in their
present design is unclear.
The high toxicity of conventional cytotoxic anticancer drugs
often forces these agents to be given at suboptimal dosages, and this
can result in treatment failure.1 To resolve this problem, delivery
platforms that can discriminate between healthy and malignant cells
have been developed.1,2 Actively targeted therapeutic delivery plat-
forms consist of three different components: a targeting component
comprising targeting ligands with affinities for molecules expressed
on cancer cells; a payload consisting of drug and/or imaging agents;
and a nanoscale structure to which the targeting and payload moieties
are attached. This platform targeting of anticancer drugs with cancer
cell-specific ligands can dramatically improve a drug’s therapeutic
index. Conjugating multiple targeting ligands to a single platform
molecule further increases the potential for specifictargetingofcancer
cells by allowing the possibility of multivalent interactions.3,4
The structural design of these types of delivery platforms is
critical to the success of the delivery device. Numerous classes of
targeted drug delivery platforms have been developed that potentially
meet the requirements needed to combine targeting ligands, imaging
agents, and drug molecules together to deliver the therapeutic
payload to a desired location in the body. These include drug-target
conjugates, linear polymers, lipid-based carriers (liposomes and
micelles), carbon nanotubes, inorganic nanoparticles, and dendri-
mers. Several of these different delivery platforms are progressing
Dendrimer-based platforms have a unique branching structure
which results in exceptionally high degrees of monodispersity
and well-defined terminal groups that provide the ability to form
soluble conjugates containing multiple copies of hydrophobic drug
and/or targeting molecules. The compact, branched structures
appear to enhance the ability of the targeting molecules to interact
in a fashion conducive to multivalent binding to cell membrane
receptors.3 The dendrimer’s small size enables efficient diffusion
across the vascular endothelium to find tumors and also allows the
rapid clearance of these molecules from the bloodstream. This
clearance avoids potential long-term toxicities and reduces the
necessity of a rapidly degradable platform. The most widely used
dendrimer in biomedical applications, poly(amidoamine) (PAM-
AM), is nonimunogenetic and nontoxic once the surface primary
amines have been modified.5ꢀ10 There have been numerous, recent
examples describing the development of dendrimer-based targeted
Received: October 15, 2010
Revised:
February 3, 2011
Published: March 22, 2011
r
2011 American Chemical Society
679
dx.doi.org/10.1021/bc100360v Bioconjugate Chem. 2011, 22, 679–689
|

Bioconjugate Chemistry
ARTICLE
delivery systems using a wide variety of targeting ligands including
monoclonal antibodies,11ꢀ15 peptides,16 T-antigens,17ꢀ19 and folic
acid.20ꢀ29
co-workers have detailed the synthesis of multimodule platforms
using both unfunctionalized PAMAM dendrons35ꢀ37 and unfunc-
tionalized Frechet-type dendrons38 for each of the modules. In all of
these systems, the focal point of the dendron possessed either an
azide or alkyne moiety. Wu and co-workers developed a 2,2-bis-
(hydroxymethyl)propionic acid (bis-MPA) based asymmetric mod-
ular dendron with 16 mannose units and 2 coumarin chromophores,
and demonstrated binding in a hemagglutination assay.39 Goyal,
Yoon, and Weck have also developed a poly(amine) dendrimer that
possessed a single aldehyde or azide moiety on the dendrimer
periphery capable of orthogonal functionalization by small molecule
functional groups.40,41 These approaches appear promising, but only
the bis-MPA dendron system has been demonstrated to have func-
tion as a targeted drug delivery platform. In all cases, the dendrons
used were G3 or smaller. This has significant limitations for wide-
spread therapeutic use because of the limited carrying capacity of low
generation dendrons. To date, click chemistry has not been applied
to a modular dendrimer system.
Despite the success of these dendrimer-based platforms, this
approach has several challenges associated with its implementation.
First, the synthesis of dendrimers with different functional groups for
targeted delivery (including targeting, drug, and imaging agents)
requires a laborious chemical process that is unique for each different
molecular combination. Second, the carrying capacity of a single
dendrimer, although significantly better than other types of delivery
platforms, is finite due to limits in surface molecule density and
solubility. This becomes a problem when one attempts to conjugate
multiple copies of the target, drug, and/or dye molecules to the same
dendrimer. Third, significant increases in heterogeneity occur with
the conjugation of each additional molecule to a single dendrimer
platform due to the stochastic nature of these chemical reactions.30
This has limited the flexibility of these systems.
To address the drawbacks of the single dendrimer platforms,
several groups have sought to apply modular design concepts to
dendrimer systems.31ꢀ41 The common premise of this new approach
is to use dendrimers or dendrons as modular units, each with multiple
copies of a single functional molecule. Multifunctional platforms can
be generated by combining different modules through a universal
coupling mechanism. The main benefits of this design are twofold:
First, segregating each functional group (drug/target/dye) to a
different dendrimer module avoids the need to develop a new
orthogonal coupling chemistry for each new combination of func-
tional groups. This advantage should not be underestimated. Sig-
nificant time is spent developing new orthogonal coupling strategies
for desired functional combinations because many of the component
drug molecules and targeting ligands (Taxol and RGD, for example)
are susceptible to a loss of activity due to undesired cross reactions as
well as degradation by hydrolysis. The second benefit of the modular
strategy is a greater carrying capacity of the delivery platform, because
the different functional molecules are localized on separate dendrimer
units and the water solubilizing dendrimer backbone is effectively
double the mass of the single dendrimer.
In the modular dendrimer systems synthesized to date, two
coupling strategies have been used to combine different dendri-
mer-based modules together: oligonucleotide self-assembly and
the 1,3-dipolar cycloaddition between an azide and an alkyne,
commonly called “click” chemistry. Oligonucleotide self-assembly
has been used to link both dendron and dendrimer modular units.
Choi and co-workers demonstrated this strategy by using two
complementary oligonucleotides to link two PAMAM dendrimers
together.31,32 DeMattie, Huang, and Tomalia used the same method
to connect two unfunctionalized PAMAM dendrons together.33
Characterization of these systems was accomplished by gel electro-
phoresis, AFM, MALDI, and UV/vis due to the small synthetic scales
employed. For the dendrimer-based system, isolated dendrimer
samples were not obtained; rather, the samples were generated in
solution. Choi et al. did demonstrate biological functionality of the
modular system in a cell culture assay32 and an in vivo model.34
The use of “click” chemistry to create dendritic modular
systems has mainly involved dendrons. “Click” chemistry is a
particularly attractive coupling method because it can be performed
with a wide variety of solvent conditions including aqueous environ-
ments. The stable triazole ring bridge, resulting from coupling alkyne
with azide moieties, is frequently achieved at near quantitative yields
and is considered to be biologically stable.42ꢀ44 Furthermore, the
“click” coupling chemistry is orthogonal to the coupling chemistries
typically used to attach functional groups to the dendrimer. Lee and
In this paper, we report on a new modular approach based
upon “clicking” together generation 5 (G5) PAMAM dendrimers
containing either the targeting agent folic acid (FA) (13) or the
dye fluorescein isothiocyanate (FITC) (14). The molecules were
1
characterized by H NMR spectroscopy and two-dimensional
Nuclear Overhauser Effect Spectroscopy (NOESY). To the best
of our knowledge, this is the first example of the successful “click”
chemistry conjugation of dendrimers (as opposed to dendrons).
Linking together two dendrimer-based modules was achieved by
first conjugating one dendrimer module to an alkyne linker (2b)
and conjugating a second dendrimer module to an azide linker
(3c). The ability of this modular system to specifically target folic
acid receptor (FAR) expressing cells has been verified by flow
cytometric documentation of the uptake of the clicked FA-FITC
modules into KB cells.
This dendrimer-based modular system has several potential
benefits over predecessor systems. In using “click” chemistry rather
than oligonucleotide linking, the modular system can be scaled up
with far greater ease and at a substantially lower cost. Oligonucleo-
tides are typically purchased in nanogram quantities, whereas the
“click” linkers in this study were produced at the gram scale. Addi-
tionally, because the clicked dendrimers are covalently linked rather
than joined via the hydrogen bond base-pairing oligonucleotide
bridge, the platform is less likely to become unlinked. This character-
istic proves beneficial when attempting to isolate and characterize
multimodule platforms. In using generation 5 dendrimers with
diameters of approximately 5 nm and over 630 hydrogen bonding
sites, the carrying capacity is substantially greater than the previously
used dendrons, which were approximately 2 nm in diameter and
possess approximately 150 hydrogen bonding sites. In fact, efforts in
our group to generate G2 and G3 PAMAM dendrons functionalized
with folic acid, failed due to a loss of platform solubility.
’ EXPERIMENTAL PROCEDURES
Reagents and Materials. Biomedical grade, generation 5
PAMAM (poly(amidoamine)) dendrimer was purchased from
Dendritech Inc. and purified by dialysis, as previously described,30 to
remove lower molecular weight impurities including trailing gen-
eration dendrimer defect structures.
MeOH (99.8%), acetic anhydride (99.5%), triethylamine
(99.5%), dimethyl sulfoxide (99.9%), fluorescein isothiocyanate
(98%), dimethylformamide (99.8%), 1-[3-(dimethylamino)-pro-
pyl-3-ethylcarbodiimide HCl (EDC) (98%), acetone (ACS reagent
680
dx.doi.org/10.1021/bc100360v |Bioconjugate Chem. 2011, 22, 679–689

Bioconjugate Chemistry
ARTICLE
grade g99.5%), methyl 3-(4-hydroxyphenyl)propanoate (97%),
sodium azide (99.99%), 1-bromo-2-chloroethane (98%), ethyl
acetate (EtOAc) (99.5%), copper sulfate pentahydrate (99%),
sodium ascorbate, 18-crown-6, K2CO3, tetrahydrofuran (99.9%),
N,N-diisopropylethylamine, benzotriazol-1-yl-oxytripyrrolidinopho-
sphonium hexafluorophosphate (98%), D2O, NaCl, 1 N HCl,
2 M KOH, and volumetric solutions (0.1 M HCl and 0.1 M
NaOH) for potentiometric titration were purchased from Sigma
Aldrich Co. and used as received. Hexanes (HPLC grade) and
10 000 molecular weight cutoff centrifugal filters (Amicon Ultra)
were from Fisher Scientific. 1ꢁ phosphate buffer saline (PBS)
(Ph = 7.4) without calcium or magnesium was purchased from
Invitrogen. Sephadex G-25 and Sephadex LH-20 were purchased
from GE Lifesciences.
Table 1. Good Correlation is Found between the Small
Molecule Model System (2a, 3b, and 4) and the Dendrimer
Model System (5, 6, and 7) for the Chemical Shifts (ppm) of
Triazole Related Protons (aꢀh) both before and after the
“Click” Reactiona
1
Nuclear Magnetic Resonance Spectroscopy. All H NMR
experiments were conducted using a Varian Inova 500 MHz instru-
ment. For all dendrimer samples, the delay time was 10s. No delay
time was used for small molecule samples. NOESY experiments on
the dendrimer samples found in Figure 1 and utilized for Table 1
were also conducted using a Varian Inova 500 MHz instrument. For
these experiments, the mixing time was 250 ms, the relaxation time
was 1 s, and the number of increments was 128 with 32 scans per
increment. Temperature was controlled at 25 ꢀC. The NOESY
experiments on the small molecule model system found in Figure 1
and utilized in Table 1 were conducted using a Varian MR400 (400
MHz) instrument. For the experiments on the small molecules, the
mixing time was 800 ms, the relaxation time was 1.2 s, and the
number of increments was 200 with 4 scans per increment. On
the basis of work published by Hoffman, the internal solvent
reference peak for all experiments in CDCl3 was set to 7.26 ppm.
For experiments conducted in D2O, the internal reference peak was
set to 4.72 ppm.45
Gel Permeation Chromatography. GPC experiments were
performed on an Alliance Waters 2690 separation module equipped
with a 2487 dual wavelength UV absorbance detector (Waters
Corporation), a Wyatt Dawn DSP laser photometer, and an Optilab
DSP interferometric refractometer (Wyatt Technology Corpora-
tion). Columns employed were TosoHaas TSK-Gel Guard PHW
06762 (75 mm ꢁ 7.5 mm, 12 μm), G 2000 PW 05761 (300 mm ꢁ
7.5 mm, 10 μm), G 3000 PW 05762 (300 mm ꢁ 7.5 mm, 10 μm),
and G 4000 PW (300 mm ꢁ 7.5 mm, 17 μm). Column temperature
was maintained at 25 ( 0.1 ꢀC with a Waters temperature control
module. The isocratic mobile phase was 0.1 M citric acid and 0.025
wt % sodium azide, pH 2.74, at a flow rate of 1 mL/min. The sample
concentration was 10 mg/5 mL with an injection volume of 100 μL.
The weight average molecular weight, Mw, was determined by GPC,
and the number average molecular weight, Mn, was calculated with
Astra 4.7 software (Wyatt Technology Corporation) based on the
molecular weight distribution.
Reverse Phase High Performance Liquid Chromatogra-
phy. HPLC analysis was carried out on a Waters Delta 600
HPLC system equipped with a Waters 2996 photodiode array
detector, a Waters 717 Plus auto sampler, and Waters Fraction
collector III. The instrument was controlled by Empower 2
software. For analysis of the conjugates, a C5 silica-based RP-
HPLC column (250 ꢁ 4.6 mm, 300 Å) connected to a C5 guard
column (4 ꢁ 3 mm) was used. The mobile phase for elution of
the conjugates was a linear gradient beginning with 90:10 (v/v)
water/acetonitrile and ending with 10:90 (v/v) water/acetoni-
trile over 25 min at a flow rate of 1 mL/min. Trifluoroacetic acid
(TFA) at 0.14 wt % concentration in water as well as in
acetonitrile was used as a counterion.
a Chemical shifts for protons in the model dendrimer system were
detected primarily via NOESY experiments.
Cell Culture and Treatment. The uptake of the synthesized
conjugates was performed using FA-receptor-expressing KB cells
(ATCC #CCL-17) as we have described previously.23 Cells were
maintained in FA-free Roswell Park Memorial Institute-1640 (RPMI
1640) medium supplemented with 10% fetal bovine serum (FBS), 2
μML-glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin
in 5% CO2 at 37 ꢀC. Cells were planted into 24 well plates at density
of 250 000 per well and allowed to reach ∼90% confluence on the
day of the experiment. They were rinsed and incubated in serum free
medium with conjugates for 1 h at 37 ꢀC in 5% CO2. In some wells, a
20-fold excess of free folic acid or the folic acid conjugated dendrimer
was added 30 min prior to the addition of the dendrimer platform for
the blocking of folate receptors.
Flow Cytometric Analysis. The cellular fluorescence was quanti-
fied on a Beckman-Coulter EPICS-XL MCL flow cytometer, and the
data were analyzed using Expo32 software (Beckman-Coulter,
681
dx.doi.org/10.1021/bc100360v |Bioconjugate Chem. 2011, 22, 679–689

Bioconjugate Chemistry
ARTICLE
Figure 1. (a) NOESY of the small-molecule model system after the “click” reaction (4). NOE cross-peaks between triazole related protons (b, c, e, f, and
g) are labeled. (b) NOESY of the model dendrimer system after the “click” reaction (7). NOE cross-peaks for the triazole related protons are similarly
labeled. The cross-peaks in the 2D spectra reveal proton chemical shifts for the several of the triazole related protons that are otherwise obscured by
overlapping dendrimer peaks. (c) Chemical structure and proton labels for the clicked small molecule model system (4) and the clicked dendrimer
model system (7). The G5 PAMAM dendrimer is represented by a gray sphere.
Miami, FL). Cells were trypsinized, rinsed, and suspended in PBS
containing 0.1% bovine serum albumin (PBSB). The viable cells
were gated, and the mean FL1-fluorescence of 10 000 cells was
quantified. Error bars are calculated using the half-peak coefficient of
variation (HPCV).46
Synthesis. Dendrimers are identified by the core dendrimer
(G5) and conjugated groups: Ac, Alkyne, Azide, FA, and FITC.
In the cases where dendrimers are linked together via the triazole
ring, alkyne and azide labels are replaced with “L.” Ac refers to the
acetamide termination, alkyne to linker 2b, azide to linker 3c, FA
to folic acid, and FITC to fluorescein isothiocyanate.
G5Ac67%:1H NMR integration determined the degree of acetylation
to be 67%. Number average molecular weight (30 473 g/mol) was
computed based on the addition of mass to the dendrimer from the
1
acetyl groups as determined by NMR. 10000 G5Ac58%: H NMR
integration determined the degree of acetylation to be 58%. Number
average molecular weight (30 064 g/mol) was computed based on
the addition of mass to the dendrimer from the acetyl groups as
determined by NMR. 100000 G5Ac68.8% 1H NMR integration
:
determined the degree of acetylation to be 69%. Number average
molecular weight (30 572 g/mol) was computed based on the addi-
tion of mass to the dendrimer from the acetyl groups as determined
1
by NMR. 1000000 G5Ac64%: H NMR integration determined the
Synthesis of Partially Acetylated Dendrimer 1. Partial acet-
ylation of G5 PAMAM dendrimer was previously described.5,30
The number average molecular weight (27 336 g/mol) and PDI
(1.018) of the unacetylated dendrimer was determined by GPC.
Potentiometric titration was conducted to determine the average
number of primary amines per dendrimer (112 ( 5). Three batches
of partially acetylated dendrimer were synthesized for further
degree of acetylation to be 64%. Number average molecular weight
(30 347 g/mol) was computed based on the addition of mass to the
dendrimer from the acetyl groups as determined by NMR.
Synthesis of Alkyne Linker (3-(4-(Prop-2-ynyloxy)phenyl)-
propanoic Acid) 2. Synthesis of methyl 3-(4-(prop-2-ynyloxy)-
phenyl)propanoate 2a has been previously reported.30 Synthesis
of (3-(4-(prop-2-ynyloxy)phenyl)propanoic acid) 2b has also
been previously reported.30
Synthesis of Azide Linker (3-(4-(2-Azidoethoxy)phenyl)-
propanoic Acid) 3. To a solution of methyl 3-(4-hydroxyphe-
nyl)propanoate (1.699 g, 9.43 mmol) in dry acetone (47.5 mL)
was added anhydrous K2CO3 (3.909 g, 0.0283 mol) followed by
1-bromo-2-chloroethane (1.56 mL, 0.0189 mol). The resulting
suspension was refluxed for 43 h with vigorous stirring. The
reaction mixture was cooled to room temperature and the salt
was removed by filtration followed by washing with portions of
1
modification. 10 G5Ac72%: H NMR integration determined the
degree of acetylation to be 72%. Number average molecular weight
(30 722 g/mol) was computed based on the addition of mass to the
dendrimer from the acetyl groups as determined by NMR. PDI
(1.019) of the purified acetylated dendrimer was determined by
GPC. 100 G5Ac65%: 1H NMR integration determined the degree of
acetylation to be 65%. Number average molecular weight (30 394 g/
mol) was computed based on the addition of mass to the dendrimer
from the acetyl groups as determined by NMR. PDI (1.060) of
the purified acetylated dendrimer was determined by GPC. 1000
682
dx.doi.org/10.1021/bc100360v |Bioconjugate Chem. 2011, 22, 679–689

Bioconjugate Chemistry
ARTICLE
EtOAc (3 ꢁ 70 mL). The crude material was purified by silica
chromatography (25:75 EtOAc/Hexane) and the solvent was
removed under vacuum to give the desired product, methyl
3-(4-(2-chloroethoxy)phenyl)propanoate 3a, as an oil (0.75 g,
33%). 1H NMR (500 MHz, CDCl3) δ 7.12 (d, J = 8.7, 2H), 6.84
(d, J = 8.7, 2H), 4.21 (t, J = 5.9, 2H), 3.80 (t, J = 5.9, 2H), 3.66 (s,
3H), 2.90 (t, J = 7.8, 2H), 2.60 (t, J = 7.8, 2H).
the dendrimer/alkyne linker solution. The resulting solution was
stirred for 24 h. All reaction steps were carried out in glass flasks
at room temperature under nitrogen.
The reaction mixture was purified using 10 000 MWCO
centrifugal filtration devices. Purification consisted of two cycles
using 1ꢁ PBS and eight cycles using DI water. All cycles were 10
min at 5000 rpm. The resulting product 5 was lyophilized for
1
To a solution of methyl 3-(4-(2-chloroethoxy)phenyl)pro-
panoate 3a (0.75 g 3.1 mmol) in anhydrous DMF (6.1 mL) was
added 18-crown-6 (3.4 mg, 0.013 mmol) and sodium azide (0.44
g, 6.8 mmol). The resulting solution was heated at 78 ꢀC for 11 h.
The reaction mixture was cooled to room temperature, diluted
with ethyl acetate (50 mL), washed with a saturated NaHCO3
solution (4 ꢁ 70 mL), and then dried over MgSO4. The solvent
was removed under vacuum to give methyl 3-(4-(2-azidoethoxy)-
three days to yield a white solid (41.5 mg, 75%). H NMR
integration determined an average of 1.4 alkyne linkers coupled
to the dendrimer. The quantification of the number of linkers by
NMR integration is described previously.30 See Supporting
Information for 1H NMR spectrum of 5.
Synthesis of G5-Ac72%-Azide1.3 6. A solution of partially
acetylated dendrimer 10 (60.5 mg, 1.97 μmol) was prepared
with anhydrous DMSO (13.444 mL). The azide linker 3c (1.2
mg, 4.9 μmol) was dissolved in DMSO (578 μL) and added to
the dendrimerꢀDMSO solution. To this mixture was added N,
N-diisopropylethylamine (5.1 μL, 30 μmol) and the resulting
solution was stirred for 45 min. Benzotriazol-1-yl-oxytripyrroli-
dinophosphonium hexafluorophosphate (2.6 mg, 4.9 μmol) was
dissolved in DMSO (511 μL) and added in a dropwise manner to
the dendrimer/azide linker solution. The resulting solution was
stirred for 24 h. All reaction steps were carried out in glass flasks
at room temperature under nitrogen.
1
phenyl)propanoate 3b as a yellow oil (0.58 g, 75%) H NMR
(500 MHz, CDCl3) δ 7.12 (d, J = 8.6, 2H), 6.85 (d, J = 8.6, 2H),
4.13 (t, J = 5.0 2H), 3.67 (s, 3H), 3.58 (t, J = 5.0, 2H), 2.90 (t, J =
7.8, 2H), 2.60 (t, J = 7.8, 2H).
To a solution of methyl 3-(4-(2-azidoethoxy)phenyl)pro-
panoate 3b (3.88 g, 0.0156 mol) in methanol (102 mL) was
added potassium hydroxide (2 M, 28.3 mL, 0.0566 mol). The
resulting solution was refluxed at 70 ꢀC for 3 h. The solution was
cooled to room temperature and condensed under reduced pres-
sure. The residue was dissolved in water (30 mL) and was acidified
by addition of 1 N HCl to pH 1. The white cloudy solution was
diluted with EtOAc. Layers were separated and the aqueous layer
was extracted with EtOAc (2 ꢁ 70 mL). The combined organic
extracts were washed with a saturated NaCl solution and dried over
MgSO4. Solvent was evaporated under vacuum to give the (3-(4-(2-
azidoethoxy)phenyl)propanoic acid) 3c as a white solid (3.44 g,
94%). 1H NMR (500 MHz, CDCl3) δ 7.14 (d, J = 8.5, 2H), 6.86 (d,
J = 8.5, 2H), 4.13 (t, J = 5.0 2H), 3.58 (t, J = 5.0, 2H), 2.91 (t, J = 7.7,
2H), 2.65 (t, J = 7.7, 2H).
Synthesis of Small Molecule Model System 4. The methyl
ester forms of the alkyne linker 2a (448.0 mg, 1.80 mmol) and
azide linker 3b (371.5 mg, 1.70 mmol) were dissolved in a mixture of
THF (1.6 mL) and water (0.4 mL). To this mixture was added
copper sulfate pentahydrate (43.1 mg, 86.0 μmol) and sodium
ascorbate (170.9 mg, 431 μmol). The resulting reaction mixture
was stirred at room temperature for 24 h. The solution was then
diluted in EtOAc (60 mL) and water (60 mL). Layers were
separated, and the aqueous layer was extracted twice with EtOAc
solution (2 ꢁ 70 mL). The combined organic extracts were washed
with a saturated NaHCO3 solution (3 ꢁ 70 mL) and then with
saturated NaCl solution (2 ꢁ 70 mL). The organic extracts were
finally dried over MgSO4. Solvent was evaporated under reduced
pressure to give the desired product 4 as a white solid (0.54 g, 95%).
1H NMR (500 MHz, CDCl3) δ 7.80 (s, 1H), 7.108 (overlapping d,
J = 8.4, 4H), 6.91 (d, J = 8.6, 2H), 6.78 (d, J = 8.6, 2H), 5.18 (s, 2H),
4.75 (t, J = 5.0, 2H), 4.33 (t, J = 5.0, 2H), 3.66 (s, 3H), 3.657 (s, 3H),
2.890 (t, J = 7.8, 2H), 2.88 (t, J = 7.7, 2H), 2.59 (t, J = 7.8, 2H), 2.59
(t, J = 7.7, 2H).
The reaction mixture was purified using 10 000 MWCO
centrifugal filtration devices. Purification consisted of two cycles
using 1ꢁ PBS and eight cycles using DI water. All cycles were 10
min at 5000 rpm. The resulting product 6 was lyophilized for
1
three days to yield a white solid (50 mg, 91%). H NMR
integration determined an average of 1.3 alkyne linkers coupled
1
to the dendrimer. See Supporting Information for H NMR
spectrum of 6.
Synthesis of Model Dendrimer System G5-Ac72%-L-G5Ac72%
7. Partially acetylated dendrimer with an average of 1.4 alkyne
linkers 5 (15.30 mg, 0.493 μmol) and partially acetylated dendrimer
with an average of 1.3 azide linkers 6 (15.4 mg, 0.496 μmol) was
dissolved in deuterium oxide (0.820 mL) and placed in a glass
microwave reactor vessel. Sodium ascorbate (6.9 mg, 35 μmol) and
copper sulfate pentahydrate (5.9 mg, 24 μmol) was added to the
dendrimer solution. The resulting solution was placed in a microwave
reactor for 40 s at 100 W with a cutoff temperature of 100 ꢀC.
The microwave conditions were repeated for an additional 40 s. The
cutoff temperature was then increased to 110 ꢀC and the microwave
was run at 100 W for 2 min. The resulting crude product was a turbid
yellow. The crude product was transferred to an NMR tube and
analyzed by NOESY and 1H NMR. After two days, the crude product
in solution turned to a redꢀbrown solution with a precipitate at the
bottom of the NMR tube. NOESY and 1H NMR experiments were
repeated at this time point. The supernatant was separated from
the precipitate and lyophilized to yield 4.9 mg of a brown solid. See
Supporting Information for 1H NMR spectrum of 7.
Synthesis of G5-Ac65%-Alkyne1.6 8. The alkyne linker was
conjugated to the partially acetylated dendrimer in two consecutive
reactions. First, a stock solution of the alkyne linker 2b (9.5 mg,
0.047 mmol) was generated with a mixture of DMF (6.198 mL) and
DMSO (3.099 mL). To this mixture was added EDC (124.9 mg,
0.651 mmol). The resulting solution was stirred for 2 h at room
temperature to create the active ester form of the alkyne linker.
A solution of partially acetylated dendrimer 100 (58.8 mg, 1.930
μmol) was prepared with DI water (13.099 mL). The active ester
form of the alkyne linker (5.784 mL, 0.0289 mmol) in DMF/
DMSO was added in a dropwise manner (0.13 mL/min) to the
Synthesis of G5-Ac72%-Alkyne1.4 5. A solution of partially
acetylated dendrimer 10 (54.6 mg, 1.78 μmol) was prepared
with anhydrous DMSO (12.133 mL). The alkyne linker 2b (0.9
mg, 4.4 μmol) was dissolved in DMSO (453 μL) and added to
the dendrimerꢀDMSO solution. To this mixture was added N,
N-diisopropylethylamine (4.6 μL, 26 μmol) and the resulting
solution was stirred for 45 min. Benzotriazol-1-yl-oxytripyrroli-
dinophosphonium hexafluorophosphate (2.3 mg, 4.4 μmol) was
dissolved in DMSO (462 μL) and added in a dropwise manner to
683
dx.doi.org/10.1021/bc100360v |Bioconjugate Chem. 2011, 22, 679–689

Bioconjugate Chemistry
ARTICLE
dendrimerꢀwater solution. The resulting reaction mixture was
stirred for 2 days. All reaction steps were carried out in glass flasks
at room temperature under nitrogen. The reaction mixture was
purified using 10000 MWCO centrifugal filtration devices. Pur-
ification consisted of five cycles using 1ꢁ PBS and five cycles using
DI water. All cycles were 30 min at 5000 rpm. The resulting
product 8 was lyophilized for three days to yield a white solid (55
cycles using 1ꢁ PBS and six cycles using DI water. All 1ꢁ PBS
cycles were 15 min at 5 000 rpm and all DI water cycles were 15
min at 5000 rpm. HPLC analysis of the lyophilized product
detected unconjugated FITC remaining in the sample. To
remove the remaining unreacted FITC, the conjugate was
purified by size exclusion chromatography using Sephadex
G-25 beads in 1ꢁ PBS. The dendrimer fraction was collected
and the elution buffer was exchanged with DI water using 10 000
MWCO centrifugal filtration devices (four cycles of 10 min at
5000 rpm). The purified product 11 was lyophilized to yield a
1
mg, 92%). H NMR integration determined an average of 1.6
alkyne linkers coupled to the dendrimer. See Supporting Informa-
tion for 1H NMR spectrum of 8.
1
Synthesis of G5-Ac65%-Azide2.5 9. The azide linker was con-
jugated to the partially acetylated dendrimer in two consecutive
reactions. First, a stock solution of the azide linker 3c (7.6 mg,
0.032 mmol) was generated with a mixture of DMF (4.96 mL)
and DMSO (2.48 mL). To this mixture was added EDC (86.7
mg, 0.452 mmol). The resulting solution was stirred for 1.75 h at
room temperature to create the active ester form of the azide
linker.
yellowꢀorange solid (10.1 mg, 46%). H NMR integration
determined an average of 3.2 FITC coupled to the dendrimer.
See Supporting Information for 1H NMR spectrum of 11.
Synthesis of G5-Ac65%-Alkyne1.6-FA3.5 12. Additional folic
acid was conjugated to the partially acetylated dendrimer with
an average of 1.6 alkyne linkers and 1.7 folic acid molecules 10 in
two consecutive reactions. First, a solution of folic acid (1.1 mg,
2.4 μmol) was generated with a mixture of DMF (0.69 mL) and
DMSO (0.34 mL). To this mixture was added EDC (6.3 mg, 33
μmol). The resulting solution was stirred for 1 h at room
temperature to create the active ester form of the folic acid.
A solution of partially acetylated dendrimer with an average
number of 1.6 alkyne linkers and 1.7 folic acid molecules 10 (8.5
mg, 0.30 μmol) was prepared with DI water (1.90 mL). The
active ester form of folic acid (1.03 mL, 2.4 μmol) in DMF/
DMSO was added in a dropwise manner to the dendrimerꢀ
water solution. The resulting reaction mixture was stirred for 3
days. All reaction steps were carried out in glass flasks at room
temperature under nitrogen. The reaction mixture was purified
by size exclusion chromatography using Sephadex G-25 in 1ꢁ
PBS. The dendrimer fraction was collected and the elution buffer
was exchanged with DI water using 10 000 MWCO centrifugal
filtration devices (four cycles of 10 min at 5000 rpm). The
purified product 12 was lyophilized for three days to yield a
yellow solid (7.0 mg, 80%). 1H NMR integration determined an
average of 3.5 folic acid molecules coupled to the dendrimer. See
Supporting Information for 1H NMR spectrum of 12.
A solution of partially acetylated dendrimer 100 (58.8 mg, 1.930
μmol) was prepared with DI water (13.10 mL). The active ester
form of the azide linker (6.66 mL, 0.0289 mmol) in DMF/
DMSO was added in a dropwise manner (0.13 mL/min) to the
first dendrimerꢀwater aliquot. The resulting mixture was stirred
for 2 days. All reaction steps were carried out in glass flasks at
room temperature under nitrogen. The reaction mixture was
purified using 10 000 MWCO centrifugal filtration devices.
Purification consisted of five cycles using 1ꢁ PBS and five cycles
using DI water. All cycles were 10 min at 5000 rpm. The resulting
product 9 was lyophilized for three days to yield a white solid (55
1
mg, 89%). H NMR integration determined an average of 2.5
azide linkers coupled to the dendrimer. See Supporting Informa-
tion for 1H NMR spectrum of 9.
Synthesis of G5-Ac65%-Alkyne1.6-FA1.7 10. Folic acid was
conjugated to the alkyne linker-conjugated dendrimer 8 in two
consecutive reactions. First, a solution of folic acid (1.9 mg, 4.26
μmol) was generated with a mixture of DMF (1.23 mL) and
DMSO (0.62 mL). To this mixture was added EDC (11.4 mg,
59.7 μmol). The resulting solution was stirred for 1 h at room
temperature to create the active ester form of folic acid.
Synthesis of G5-Ac107-Alkyne1.6-FA3.5 13. Partially acetylated
dendrimer with an average number of 1.6 alkyne linkers and 3.5
folic acid 12 (7.0 mg, 0.22 μmol) was dissolved in anhydrous
methanol (1.124 mL). Triethylamine (1.7 μL, 0.012 mmol) was
added to this mixture and stirred for 30 min. Acetic anhydride
(0.9 μL, 9.6 μmol) was added in a dropwise manner to the
dendrimer solution. The reaction was carried out in a glass flask,
under nitrogen, at room temperature for 24 h. Methanol was
evaporated from the resulting solution and the product was
purified by size exclusion chromatography using Sephadex LH-
20 in methanol. The purified dendrimer 13 was lyophilized for
A solution of partially acetylated dendrimer with an average
number of 1.6 alkyne linkers 8 (20.8 mg, 0.752 μmol) was
prepared with DI water (4.638 mL). The active ester form of folic
acid (1.85 mL, 4.26 μmol) in DMF/DMSO was added in a
dropwise manner to the dendrimerꢀwater solution. The result-
ing reaction mixture was stirred for 3 days. All reaction steps were
carried out in glass flasks at room temperature under nitrogen.
The reaction mixture was purified using 10 000 MWCO cen-
trifugal filtration devices. Purification consisted of five cycles
using 1ꢁ PBS and four cycles using DI water. All cycles were 10
min at 5000 rpm. The resulting product 10 was lyophilized for
1
three days to yield a yellow solid (6.6 mg, 90%). H NMR
integration determined the degree of acetylation to be 100%. See
1
three days to yield a yellow solid (15.6 mg, 73%). H NMR
Supporting Information for 1H NMR spectrum of 13.
integration determined an average of 1.7 folic acid molecules
coupled to the dendrimer. See Supporting Information for H
NMR spectrum of 10.
Synthesis of G5-Ac106-Azide2.5-FITC3.2 14. Partially acetylated
dendrimer with an average number of 2.5 azide linkers and 3.2
FITC 11 (7.5 mg, 0.23 μmol) was dissolved in anhydrous
methanol (1.21 mL). Triethylamine (1.8 μL, 0.013 mmol) was
added to this mixture and stirred for 30 min. Acetic anhydride
(1.0 μL, 10.0 μmol) was added in a dropwise manner to the
dendrimer solution. The reaction was carried out in a glass flask,
under nitrogen, at room temperature for 24 h. Methanol was
evaporated from the resulting solution, and the product was
purified by size exclusion chromatography using Sephadex LH-
20 in methanol. The purified dendrimer 14 was lyophilized for
1
Synthesis of G5-Ac65%-Azide2.5-FITC3.2 11. Partially acetylated
dendrimer with an average number of 2.5 azide linkers 9 (21.5
mg, 0.694 μmol) was dissolved in DMSO (1.5 mL). Fluorescein
isothiocyanate (1.4 mg, 3.5 μmol) was dissolved in DMSO
(0.54 mL) and added in a dropwise fashion to the dendrimer
solution. The resulting mixture was stirred for 24 h at room
temperature. The reaction mixture was purified using 10 000
MWCO centrifugal filtration devices. Purification consisted of six
684
dx.doi.org/10.1021/bc100360v |Bioconjugate Chem. 2011, 22, 679–689

Bioconjugate Chemistry
ARTICLE
1
three days to yield a yellow solid (7.1 mg, 91%). H NMR
integration determined the degree of acetylation to be 100%. See
Supporting Information for 1H NMR spectrum of 14.
Synthesis of Folic Acid Targeted Dendrimer System FA3.5-G5-
Ac107-L-G5-Ac106-FITC3.2 15. Dendrimer with an average of 1.6
alkyne linkers and 3.5 folic acid molecules 13 (3.1 mg, 91 nmol)
and dendrimer with an average of 2.5 azide linkers and 3.2 FITC
molecules 14 (3.0 mg, 88 nmol) were dissolved in deuterium
oxide (0.65 mL) and placed in a glass microwave reactor vessel.
Sodium ascorbate (1.1 mg, 4.5 μmol) and copper sulfate
pentahydrate (1.1 mg, 5.4 μmol) was added to the dendrimer
solution. The resulting solution was placed in a microwave
reactor for 6.5 min at 100 W with a cutoff temperature of
100 ꢀC. The reaction mixture was transferred to an NMR tube
and analyzed by NOESY and 1H NMR spectroscopy. Lyophili-
zation yielded 6.7 mg of a red solid. See Supporting Information
for 1H NMR spectrum of 15.
Synthesis of G5-Ac67%-Alkyne1.3 16. A solution of partially
acetylated dendrimer 1000 (176.7 mg, 5.8 μmol) was prepared
with anhydrous DMSO (39.27 mL). The alkyne linker 2b (2.6
mg, 13 μmol) was dissolved in DMSO (1.31 mL) and added to
the dendrimerꢀDMSO solution. To this mixture was added N,
N-diisopropylethylamine (13.4 μL, 76.7 μmol) and the resulting
solution was stirred for 45 min. Benzotriazol-1-yl-oxytripyrroli-
dinophosphonium hexafluorophosphate (6.7 mg, 13 μmol) was
dissolved in DMSO (1.331 mL) and added in a dropwise manner
to the dendrimer/alkyne linker solution. The resulting solution
was stirred for 24 h. All reaction steps were carried out in glass
flasks at room temperature under nitrogen.
Figure 2. Proton NMR spectra of the small-molecule model system and
model dendrimer system both pre- and post-”click” reaction. An upfield
shift is observed for aromatic proton g as a result of the “click” reaction.
(a) Spectrum of the small molecule model system before the “click”
reaction (2a and 3b), taken in CDCl3. Proton g has a chemical shift of
6.85 ppm. (b) Spectrum of the small molecule model system after the
“click” reaction (4), taken in CDCl3. As a result of the “click” reaction, g
has experienced an upfield change to 6.78 ppm. (c) Spectrum of the
model dendrimer system before the “click” reaction (5 and 6), taken in
D2O. Proton g overlaps proton b at 6.90 ppm. (d) Spectrum of the
model dendrimer system after the “click” reaction (7), taken in D2O.
Similar to the small molecule model system, proton g experiences an
upfield change as a result of the “click” reaction. In the model dendrimer
system, the new chemical shift is 6.74 ppm.
The reaction mixture was purified using 10 000 MWCO
centrifugal filtration devices. Purification consisted of two cycles
using 1ꢁ PBS and eight cycles using DI water. All cycles were 10
min at 5000 rpm. The resulting product 5 was lyophilized for
1
three days to yield a white solid (116.2 mg, 65.2%). H NMR
integration determined an average of 1.3 alkyne linkers coupled
1
to the dendrimer. See Supporting Information for H NMR
spectrum of 16.
Synthesis of G5-Ac110.7-Alkyne1.3 17. Partially acetylated
dendrimer with an average number of 1.3 alkyne linkers 16
(22.6 mg, 0.737 μmol) was dissolved in anhydrous methanol
(3.0 mL). Triethylamine (5.8 μL, 0.042 mmol) was added to this
mixture and stirred for 30 min. Acetic anhydride (3.2 μL, 34
μmol) was added in a dropwise manner to the dendrimer
solution. The reaction was carried out in a glass flask, under
nitrogen, at room temperature for 24 h. Methanol was evapo-
rated from the resulting solution and the product was purified by
size exclusion chromatography using Sephadex LH-20 in metha-
nol. The purified dendrimer 17 was lyophilized for three days to
Lyophilization yielded 2.4 mg of a red solid. See Supporting
Information for 1H NMR spectrum of 18.
’ RESULTS AND DISCUSSION
Synthesis and Characterization of the Small-Molecule
Model System. A small molecule model system (4) was first
synthesized to facilitate the spectroscopic assignment of triazole-
related atoms resulting from successful “click” reactions between
dendrimer modules with an azide linker (3c) and dendrimer modules
possessing an alkyne linker (2b). This model system utilized the
methyl ester forms of the two linkers (2a and 3b). Proton assign-
1
yield a white solid (19.1 mg, 80.5%). H NMR integration
determined the degree of acetylation to be 100%. See Supporting
Information for 1H NMR spectrum of 17.
Synthesis of Untargeted Dendrimer System G5-Ac110.7-L-G5-
Ac106-FITC3.2 18. Dendrimer with an average of 1.3 alkyne linkers
17 (1.0 mg, 31 nmol) and dendrimer with an average of 2.5 azide
linkers and 3.2 FITC molecules 14 (1.0 mg, 29 nmol) were
dissolved in deuterium oxide (0.74 mL) and placed in a glass
microwave reactor vessel. Sodium ascorbate (0.36 mg, 1.8 μmol)
and copper sulfate pentahydrate (0.38 mg, 1.5 μmol) were added
to the dendrimer solution. The resulting solution was placed in a
microwave reactor for 6.5 min at 100 W with a cutoff temperature
of 100 ꢀC. The reaction mixture was transferred to an NMR
1
ments were based upon H NMR and NOESY experiments in
CDCl3. Figure 1, panel a, displays the cross-peaks for the triazole
related protons in the clicked product (4), and Table 1 contains the
chemical shifts for these in both the pre- and post-”click” reaction
states. Protons c, e, and f experience the greatest change in chemical
shift as a result of the “click” reaction. Also of interest is the region
between 6.4 and 8.5 ppm (Figure 2, panels a and b), which shows the
upfield shift for peak g from 6.85 ppm to 6.78 ppm.
Synthesis and Characterization of the Model Dendrimer
System. Dendrimers without target or dye functionalities, possessing
1
tube and analyzed by NOESY and H NMR spectroscopy.
685
dx.doi.org/10.1021/bc100360v |Bioconjugate Chem. 2011, 22, 679–689

Bioconjugate Chemistry
ARTICLE
Figure 3. Synthetic scheme for the model dendrimer system (7). This
simplified modular platform was developed to assist with the spectro-
scopic characterization of the folic acid targeted dendrimer system. The
G5 dendrimer, used in this study, had an average of 112 end groups as
determined by GPC and potentiometric titration.
only the “click” reaction functional groups (5 and 6), were employed
to develop “click” reaction conditions (Figure 3). Because many of
the proton peaks associated with the dendrimer-conjugated alkyne
and azide linkers (particularly those closest to the “click” reaction
sites) overlap in the 1H NMR spectra with other protons belonging
to the PAMAM dendrimer, NOESY experiments were used to
document proton chemical shifts via the resolution of cross peaks
in the 2-D spectra (Figure 1, panel b). The chemical shifts for the
triazole related protons in the dendrimer system both pre- and post-
”click” reaction can be found in Table 1. The region between 6.4 and
8.5 ppm in the proton spectra for the pre- and post-”click”ed
dendrimer samples can be found in Figure 2, panels c and d. In the
spectra for the pre-”click” reaction mixture (panel c), both sets of
aromatic protons overlap at 6.90 ppm (b and g) and 7.13 ppm (a and
h). In the sample post-”click” reaction (panel d), the aromatic
protons no longer overlap. Protons a and h partially overlap at 7.09
ppm and 7.06 ppm, and protons b and g are found at 6.90 ppm and
6.74 ppm, respectively.
A comparison of the chemical shifts in Table 1 for the small
molecule system and the dendrimer system reveals good correla-
tions for both the pre- and post-reaction states. This indicates
that a successful “click” reaction has occurred between the azide
and alkyne conjugated dendrimers. It is important to note that,
whereas chemical shifts for the small molecule model system are
determined in CDCl3, the chemical shifts for the dendrimer
sample were detected in D2O. Although the different solvents
could influence the proton chemical shifts, this does not appear
to be an issue for these particular molecules.
Figure 4. Synthetic scheme for the folic acid targeted modular den-
drimer-based platform (15). A module possessing a terminal alkyne
moiety and folic acid (13) is coupled to a second module possessing a
terminal azide moiety and FITC (14) via the Cu-catalyzed 1,3-dipolar
cycloaddition reaction. Both modules were fully acetylated to avoid
nonspecific cellular interactions.
overlapped by several peaks between 7.80 ppm and 8.20 ppm
associated with the dendrimer (Figure 2, panel d). The NOESY
spectra did not expose any cross peaks in this region with other
protons in the linkers that are in close proximity to the triazole ring.
The cross-peaks associated with the triazole proton appear to be
below the intensity required for NMR detection.
Synthesis and Characterization of the Folic Acid Targeted
Dendrimer System. Synthesis of the folic acid targeted modular
dendrimer platform is outlined in Figure 4. Dendrimers with an
average of 1.6 alkyne linkers (8) were functionalized with the
targeting molecule folic acid. The formation of an amide linkage
between one of the remaining primary amines on the dendrimer
and one of the carboxylic acid groups in folic acid was achieved by
EDC coupling chemistry as previously reported.22 This reaction
conjugated an average of 1.7 folic acid molecules per dendrimer
as determined by NMR (10). Because this average was below the
optimal range for multivalent binding,3 the reaction was repeated
using the dendrimer with an average of 1.6 alkyne linkers and 1.7
folic acid (10). The second reaction resulted in the addition of
1.8 folic acid molecules per dendrimer bringing the final average
to 3.5 folic acid molecules per dendrimer (12). The remaining
dendrimer primary amines were then fully acetylated to avoid
positive charge-based cellular interactions (13).6,9,10 Dendrimers
A peak for the single proton in the triazole ring is absent from
both the NOESY and 1D experiments. In the small molecule
system, this peak is found at 7.80 ppm (Figure 2, panel b). Working
with a similar system using PAMAM dendrons that contained the
alkyne and azide moieties at the dendron focal point, Lee et al. found
that the triazole proton peak gradually shifted downfield from 7.77
ppm to 7.93 ppm as the generation of the clicked dendrons increased
from 1 to 3.47 If this downfield change in chemical shift also holds
for the generation 5 dendrimer case, the triazole proton would be
686
dx.doi.org/10.1021/bc100360v |Bioconjugate Chem. 2011, 22, 679–689

Bioconjugate Chemistry
ARTICLE
Figure 5. Binding and uptake of the fluorescent modular targeted dendrimer platform in KB cells as measured by flow cytometry. (a) Uptake of FA3.5
-
G5-Ac107-L-G5Ac106-FITC3.2 (15). (b) Uptake of G5-Ac106-Azide2.5-FITC3.2 (14) is not observed for 30 nM, 100 nM, and 300 nM. Very minimal
uptake of this untargeted module is observed at 1000 nM. (c) Similarly, no uptake is observed for an uncoupled mixture of G5-Ac106-Azide2.5-FITC3.2
(14) and G5-Ac107-Alkyne1.6-FA3.5 (13). (d) Uptake of FA3.5-G5-Ac107-L-G5Ac106-FITC3.2 (15) is successfully blocked using a 20-fold excess of free
folic acid. (e) Uptake of FA3.5-G5-Ac107-L-G5Ac106-FITC3.2 (15) is also successfully blocked using a 20-fold excess (with respect to the folic acid
content) of G5-Ac107-Alkyne1.6-FA3.5 (13). (f) Summary of mean fluorescence values for aꢀe. Uptake of FA3.5-G5-Ac107-L-G5Ac106-FITC3.2 (15) is
displayed in blue. Uptake of the targeted platform (15) blocked by a 20-fold excess of G5-Ac107-Alkyne1.6-FA3.5 (13) is shown in orange. Uptake of the
targeted platform (15) blocked by a 20-fold excess of free folic acid can be found in green. Uptake of a mixture of G5-Ac106-Azide2.5-FITC3.2 (14) and
G5-Ac106-Alkyne1.6-FA3.5 (13) can be found in teal. Finally, uptake of G5-Ac106-Azide2.5-FITC3.2 (14) can be found in purple. Error bars indicate
standard deviation as computed from half-peak coefficient of variation (HPCV) values.
with 2.5 azide linkers (9) were functionalized with the dye
molecule FITC. Using conditions similar to those in previously
published work,25 an average of 3.2 FITC molecules were
coupled to the dendrimer via the formation of a thiourea bond
between the primary amine on the dendrimer and the isothio-
cyanate group in FITC (11). This dendrimer conjugate also was
fully acetylated (14). Reverse phase HPLC confirmed that any
unreacted FITC or folic acid molecules had been removed by
purification of dendrimer 13 and 14.
The two dendrimer modules (13 and 14) were coupled
together using the Cu-catalyzed 1,3-dipolar cycloaddition reac-
tion under conditions similar to those used with the model
dendrimer system. NOESY experiments provided direct spectro-
scopic proof that the functionalized dendrimers had been clicked
together. Specifically, the AA0BB0 pattern was observed to shift in
a fashion identical to that observed for the two model systems
previously described (Table 1 and Figure 2).
In Vitro Testing of the Folic Acid Targeted Dendrimer
System with KB Cells. Cellular uptake of the folic acid targeted
dendrimer system (15) at four different concentrations (30 nM,
100 nM, 300 nM, and 1000 nM of 15) was measured in KB cells
that express a high cellular membrane concentration of folic acid
receptor (FAR). Fluorescence uptake was quantified by flow
cytometry. As seen in Figure 5a and f-blue, a dose-dependent
uptake was observed with saturation occurring at 100 nM. This
binding affinity is consistent with previous studies on single
dendrimer platforms possessing multiple FITC and multiple FA
molecules.23
A series of control experiments were performed in order to
ensure that uptake of the folic acid targeted dendrimer system
(15) was occurring via receptor-mediated endocytosis and not
nonspecific membrane interactions. The first set of controls
measured uptake of single dendrimers possessing the azide linker
and multiple FITC (14) at 30 nM, 100 nM, 300 nM, and 1000
nM (Figure 5b and f-purple). No uptake was observed for this
sample above the background level. The second control sample
contained a nonconjugated (unclicked) mixture of the two
dendrimers functionalized with either FITC or folic acid (13
and 14). Uptake of this mixture was quantified at 30 nM, 100 nM,
and 300 nM (Figure 5c and f-teal). At all three concentrations, no
florescence uptake was observed. This control eliminates the
possibility that the dendrimer modules could form a noncova-
lently linked complex that would be internalized. A third control
sample (18) was composed of an untargeted dendrimer module
(17) coupled to the FITC conjugated imaging module (14). The
untargeted dual module platform (18) was assembled under the
same conditions used to form the folic acid targeted platform
(15). Mean fluorescence uptake of the untargeted platform can
687
dx.doi.org/10.1021/bc100360v |Bioconjugate Chem. 2011, 22, 679–689

Bioconjugate Chemistry
ARTICLE
be found in the Supporting Information. At concentrations up to
300 nM, no uptake was observed beyond the background level.
This untargeted dendrimer platform is an essential control,
because it matches the molecular weight and size of the folic
acid targeted dendrimer system (15). This control will become
even more important as evaluation of this system moves to an
in vivo model.
’ AUTHOR INFORMATION
Corresponding Author
*Corresponding author. Mark M. Banaszak Holl, mbanasza@
umich.edu, 930 N. University Avenue, Ann Arbor, MI 48109-
1055. TEL: 734-763-2283, FAX: 734-936-2307.
The final set of controls investigated active blocking of the folic
acid receptor by either free folic acid (Figure 5d and f-green) or a
folic acidꢀdendrimer conjugate without a fluorescent dye (13)
(Figure 5e and f-orange) to prevent the specific uptake of the
folic acid targeted dendrimer system. A 20-fold excess of blocking
agent was employed relative to the targeted platform. For the
blocking experiment using the single dendrimerꢀfolic acid
conjugate (13), molar equivalence was based on the folic acid
content of the sample rather than the dendrimer content. Both
blocking agents were evaluated at 30 nM, 100 nM, 300 nM, and
1000 nM. Complete blocking is achieved using free folic
acid concentrations up to 100 nM. While the dendrimerꢀfolic
acid conjugate is not as effective at blocking as the free folic acid,
approximately 75% blocking is achieved. These binding data
indicate that the cellular association of the folic acid targeted
dendrimer system occurs through the folic acid receptor rather
than via nonspecific interactions. On a more fundamental level,
the biological results prove again that the folic acid conjugate
dendrimer module is covalently linked by “click” chemistry to
dendrimer module functionalized with FITC.
’ ACKNOWLEDGMENT
This project was supported with Federal funds from the
National Cancer Institute, National Institutes of Health, under
award 1 R01 CA119409.
’ REFERENCES
(1) Allen, T. M. (2002) Ligand-targeted therapeutics in anticancer
therapy. Nat. Rev. Cancer 2, 750–763.
(2) Peer, D., Karp, J. M., Hong, S., FaroKhzad, O. C., Margalit, R.,
and Langer, R. (2007) Nanocarriers as an emerging platform for cancer
therapy. Nat. Nanotechnol. 2, 751–760.
(3) Hong, S., Leroueil, P. R., Majoros, I. J., Orr, B. G., Baker, J. R., and
Holl, M. M. B. (2007) The binding avidity of a nanoparticle-based
multivalent targeted drug delivery platform. Chem. Biol. 14, 105–113.
(4) Mammen, M., Choi, S. K., and Whitesides, G. M. (1998) Polyvalent
interactions in biological systems: Implications for design and use of
multivalent ligands and inhibitors. Angew. Chem., Int. Ed. 37, 2755–2794.
(5) Majoros, I. J., Keszler, B., Woehler, S., Bull, T., and Baker, J. R.
(2003) Acetylation of poly(amidoamine) dendrimers. Macromolecules
36, 5526–5529.
Design Analysis. One of the limitations of the modular
platform design in its current iteration is that each module is
composed of a heterogeneous distribution of dendrimerꢀligand
components.30,48ꢀ50 To a first approximation, these distribu-
tions are consistent with Poissonian statistics. Not only are the
modules composed of a distribution of dendrimer particles with
different numbers of the biologically active ligand (folic acid,
FITC), but the modules are also composed of a distribution of
dendrimer with different numbers of the coupling ligands (azide
and alkyne). The implication of these distributions is that the
final “dual” module platform is most likely composed of a large
number of different platforms with different numbers of modules
linked together and different numbers of ligands per module.
Future designs should incorporate characterization and control
of these distributions in order to make a more controlled,
reproducible material.
(6) Hong, S. P., Bielinska, A. U., Mecke, A., Keszler, B., Beals, J. L., Shi,
X. Y., Balogh, L., Orr, B. G., Baker, J. R., and Holl, M. M. B. (2004)
Interaction of poly(amidoamine) dendrimers with supported lipid bilayers
and cells: Hole formation and the relation to transport. Bioconjugate Chem.
15, 774–782.
(7) Lee, C. C., MacKay, J. A., Frechet, J. M. J., and Szoka, F. C.
(2005) Designing dendrimers for biological applications. Nat. Biotech-
nol. 23, 1517–1526.
(8) Svenson, S., and Tomalia, D. A. (2005) Commentary - Den-
drimers in biomedical applications - reflections on the field. Adv. Drug
Delivery Rev. 57, 2106–2129.
(9) Hong, S. P., Leroueil, P. R., Janus, E. K., Peters, J. L., Kober, M. M.,
Islam, M. T., Orr, B. G., Baker, J. R., andHoll, M. M. B. (2006)Interactionof
polycationic polymers with supported lipid bilayers and cells: Nanoscale
hole formation and enhanced membrane permeability. Bioconjugate Chem.
17, 728–734.
(10) Leroueil, P. R., Hong, S. Y., Mecke, A., Baker, J. R., Orr, B. G.,
and Holl, M. M. B. (2007) Nanoparticle interaction with biological
membranes: Does nanotechnology present a janus face?. Acc. Chem. Res.
40, 335–342.
(11) Patri, A. K., Myc, A., Beals, J., Thomas, T. P., Bander, N. H., and
Baker, J. R. (2004) Synthesis and in vitro testing of J591 antibody-dendrimer
conjugates for targeted prostate cancer therapy. Bioconjugate Chem.
15, 1174–1181.
(12) Shukla, R., Thomas, T. P., Peters, J. L., Desai, A. M., Kukowska-
Latallo, J., Patri, A. K., Kotlyar, A., and Baker, J. R. (2006) HER2 specific
tumor targeting with dendrimer conjugated anti-HER2 mAb. Bioconjugate
Chem. 17, 1109–1115.
(13) Wu, G., Barth, R. F., Yang, W. L., Kawabata, S., Zhang, L. W.,
and Green-Church, K. (2006) Targeted delivery of methotrexate to
epidermal growth factor receptor-positive brain tumors by means of
cetuximab (IMC-C225) dendrimer bioconjugates. Mol. Cancer Ther.
5, 52–59.
’ CONCLUSION
A delivery platform covalently linking two different PAMAM
dendrimer modules through “click” chemistry was successfully
designed and synthesized. Two model systems were also gener-
ated to assist in the NMR spectroscopy characterization of the
modular platform. The folic acid targeted system was evaluated
for functional activity in vitro with KB cells and found to
selectively target the cancer cell line through the high-affinity
folate receptor.
’ ASSOCIATED CONTENT
(14) Wu, G., Barth, R. F., Yang, W. L., Chatterjee, M., Tjarks, W.,
Ciesielski, M. J., and Fenstermaker, R. A. (2004) Site-specific conjugation of
boron-containing dendrimers to anti-EGF receptor monoclonal antibody
cetuximab (IMC-C225) and its evaluation as a potential delivery agent for
neutron capture therapy. Bioconjugate Chem. 15, 185–194.
Supporting Information. 1H NMR characterization of
S
b
dendrimer conjugates and a plot of mean fluorescence for the
untargeted modular platform (18) control. This material is available
free of charge via the Internet at http://pubs.acs.org.
688
dx.doi.org/10.1021/bc100360v |Bioconjugate Chem. 2011, 22, 679–689

Bioconjugate Chemistry
ARTICLE
(15) Backer, M. V., Gaynutdinov, T. I., Patel, V., Bandyopadhyaya, A. K.,
Thirumamagal, B. T. S., Tjarks, W., Barth, R. F., Claffey, K., and Backer, J. M.
(2005) Vascular endothelial growth factor selectively targets boronated
dendrimers to tumor vasculature. Mol. Cancer Ther. 4, 1423–1429.
(16) Shukla, R., Thomas, T. P., Peters, J., Kotlyar, A., Myc, A., and
Baker, J. R. (2005) Tumor angiogenic vasculature targeting with
PAMAM dendrimer-RGD conjugates. Chem. Commun. 5739–5741.
(17) Sheng, K. C., Kalkanidis, M., Pouniotis, D. S., Esparon, S., Tang,
C. K., Apostolopoulos, V., and Pietersz, G. A. (2008) Delivery of antigen
using a novel mannosylated dendrimer potentiates immunogenicity in
vitro and in vivo. Eur. J. Immunol. 38, 424–436.
(18) Baek, M. G., and Roy, R. (2002) Synthesis and protein binding
properties of T-antigen containing GlycoPAMAM dendrimers. Bioorg.
Med. Chem. 10, 11–17.
(19) Taite, L. J., and West, J. L. (2006) Poly(ethylene glycol)-lysine
dendrimers for targeted delivery of nitric oxide. J. Biomater. Sci., Polym.
Ed. 17, 1159–1172.
(20) Kono, K., Liu, M. J., and Frechet, J. M. J. (1999) Design of
dendritic macromolecules containing folate or methotrexate residues.
Bioconjugate Chem. 10, 1115–1121.
(21) Shukla, S., Wu, G., Chatterjee, M., Yang, W. L., Sekido, M.,
Diop, L. A., Muller, R., Sudimack, J. J., Lee, R. J., Barth, R. F., and Tjarks,
W. (2003) Synthesis and biological evaluation of folate receptor-
targeted boronated PAMAM dendrimers as potential agents for neutron
capture therapy. Bioconjugate Chem. 14, 158–167.
(22) Majoros, I. J., Myc, A., Thomas, T., Mehta, C. B., and Baker, J. R.
(2006) PAMAM dendrimer-based multifunctional conjugate for cancer
therapy: Synthesis, characterization, and functionality. Biomacromole-
cules 7, 572–579.
(33) DeMattei, C. R., Huang, B. H., and Tomalia, D. A. (2004)
Designed dendrimer syntheses by self-assembly of single-site, ssDNA
functionalized dendrons. Nano Lett. 4, 771–777.
(34) Choi, Y. (2005) in Biomedical Engineering, p 191, University of
Michigan, Ann Arbor, MI.
(35) Lee, J. W., Kim, J. H., Kim, H. J., Han, S. C., Kim, J. H., Shin,
W. S., and Jin, S. H. (2007) Synthesis of symmetrical and unsymmetrical
PAMAM dendrimers by fusion between azide- and alkyne-functiona-
lized PAMAM dendrons. Bioconjugate Chem. 18, 579–584.
(36) Lee, J. W., Kim, H. J., Han, S. C., Kim, J. H., and Jin, S. H. (2008)
Designing poly(amido amine) dendrimers containing core diversities by
click chemistry of the propargyl focal point poly(amido amine) den-
drons. J. Polym. Sci., Part A: Polym. Chem. 46, 1083–1097.
(37) Lee, J. W., Kim, B. K., Kim, H. J., Han, S. C., Shin, W. S., and Jin,
S. H. (2006) Convergent synthesis of symmetrical and unsymmetrical
PAMAM dendrimers. Macromolecules 39, 2418–2422.
(38) Lee, J. W., Kim, J. H., Kim, B. K., Shin, W. S., and Jin, S. H.
(2006) Synthesis of Frechet type dendritic benzyl propargyl ether and
Frechet type triazole dendrimer. Tetrahedron 62, 894–900.
(39) Wu, P., Malkoch, M., Hunt, J. N., Vestberg, R., Kaltgrad, E.,
Finn, M. G., Fokin, V. V., Sharpless, K. B., and Hawker, C. J. (2005)
Multivalent, bifunctional dendrimers prepared by click chemistry. Chem.
Commun. 5775–5777.
(40) Goyal, P., Yoon, K., and Weck, M. (2007) Multifunctionaliza-
tion of dendrimers through orthogonal transformations. Chem.—Eur. J.
13, 8801–8810.
(41) Yoon, K., Goyal, P., and Weck, M. (2007) Monofunctionaliza-
tion of dendrimers with use of microwave-assisted 1,3-dipolar cycload-
ditions. Org. Lett. 9, 2051–2054.
(23) Thomas, T. P., Majoros, I. J., Kotlyar, A., Kukowska-Latallo,
J. F., Bielinska, A., Myc, A., and Baker, J. R. (2005) Targeting and
inhibition of cell growth by an engineered dendritic nanodevice. J. Med.
Chem. 48, 3729–3735.
(42) Rostovtsev, V. V., Green, L. G., Fokin, V. V., and Sharpless, K. B.
(2002) A stepwise Huisgen cycloaddition process: Copper(I)-catalyzed
regioselective “ligation” of azides and terminal alkynes. Angew. Chem.,
Int. Ed. 41, 2596–2599.
(24) Myc, A., Douce, T. B., Ahuja, N., Kotlyar, A., Kukowska-Latallo,
J., Thomas, T. P., and Baker, J. R. (2008) Preclinical antitumor efficacy
evaluation of dendrimer-based methotrexate conjugates. Anti-Cancer
Drugs 19, 143–149.
(25) Majoros, I. J., Thomas, T. P., Mehta, C. B., and Baker, J. R.
(2005) Poly(amidoamine) dendrimer-based multifunctional engineered
nanodevice for cancer therapy. J. Med. Chem. 48, 5892–5899.
(26) Kukowska-Latallo, J. F., Candido, K. A., Cao, Z. Y., Nigavekar,
S. S., Majoros, I. J., Thomas, T. P., Balogh, L. P., Khan, M. K., and Baker,
J. R. (2005) Nanoparticle targeting of anticancer drug improves
therapeutic response in animal model of human epithelial cancer. Cancer
Res. 65, 5317–5324.
(27) Myc, A., Patri, A. K., and Baker, J. R. (2007) Dendrimer-based
BH3 conjugate that targets human carcinoma cells. Biomacromolecules
8, 2986–2989.
(28) Myc, A., Majoros, I. J., Thomas, T. P., and Baker, J. R. (2007)
Dendrimer-based targeted delivery of an apoptotic sensor in cancer cells.
Biomacromolecules 8, 13–18.
(29) Landmark, K. J., DiMaggio, S., Ward, J., Kelly, C., Vogt, S.,
Hong, S., Kotlyar, A., Myc, A., Thomas, T. P., Penner-Hahn, J. E., Baker,
J. R., Holl, M. M. B., and Orr, B. G. (2008) Synthesis, characterization,
and in vitro testing of superparamagnetic iron oxide nanoparticles
targeted using folic acid-conjugated dendrimers. ACS Nano 2, 773–783.
(30) Mullen, D. G., Desai, A. M., Waddell, J. N., Cheng, X.-M., Kelly,
C. V., McNerny, D. Q., Majoros, I. J., Baker, J. R., Sander, L. M., Orr,
B. G., and Banaszak Holl, M. M. (2008) The implications of stochastic
synthesis for the conjugation of functional groups to nanoparticles.
Bioconjugate Chem. 19, 1748–52.
(43) Wu, P., Feldman, A. K., Nugent, A. K., Hawker, C. J., Scheel, A.,
Voit, B., Pyun, J., Frechet, J. M. J., Sharpless, K. B., and Fokin, V. V.
(2004) Efficiency and fidelity in a click-chemistry route to triazole
dendrimers by the copper(I)-catalyzed ligation of azides and alkynes.
Angew. Chem., Int. Ed. 43, 3928–3932.
(44) Wu, P., and Fokin, V. V. (2007) Catalytic azide-alkyne
cycloaddition: Reactivity and applications. Aldrichim. Acta 40, 7–17.
(45) Hoffman, R. E. (2006) Standardization of chemical shifts of
TMS and solvent signals in NMR solvents. Magn. Reson. Chem.
44, 606–616.
(46) Marie, D., and Brown, S. C. (1993) A Cytometric exercise in
plant DNA Hhistograms, with 2c-values for 70 species. Biol. Cell
78, 41–51.
(47) Lee, J. W., Kim, J. H., Kim, B. K., Kim, J. H., Shin, W. S., and Jin,
S. H. (2006) Convergent synthesis of PAMAM dendrimers using click
chemistry of azide-functionalized PAMAM dendrons. Tetrahedron
62, 9193–9200.
(48) Mullen, D. G., Fang, M., Desai, A., Baker, J. R., Orr, B. G., and
Holl, M. M. B. (2010) A quantitative assessment of nanoparticle-ligand
distributions: implications for targeted drug and imaging delivery in
dendrimer conjugates. ACS Nano 4, 657–670.
(49) Mullen, D. G., Borgmeier, E. L., Fang, M., McNerny, D. Q.,
Desai, A., Baker, J. R., Orr, B. G., and Banaszak Holl, M. M. (2010) Effect
of mass transport in the synthesis of partially acetylated dendrimer:
Implications for functional ligand-nanoparticle distributions. Macromo-
lecules 43, 6577ꢀ6587.
(50) Mullen, D. G., Borgemeier, E. L., Desai, A., van Dongen, M. A.,
Barash, M., Cheng, X., Baker, J. R., and Banaszak Holl, M. M. (2010)
Isolation and characterization of dendrimer with precise numbers of
functional groups, Chem. Eur. J. 16, 10675ꢀ10678.
(31) Choi, Y. S., Mecke, A., Orr, B. G., Holl, M. M. B., and Baker, J. R.
(2004) DNA-directed synthesis of generation 7 and 5 PAMAM
dendrimer nanoclusters. Nano Lett. 4, 391–397.
(32) Choi, Y., Thomas, T., Kotlyar, A., Islam, M. T., and Baker,
J. R. (2005) Synthesis and functional evaluation of DNA-assembled
polyamidoamine dendrimer clusters for cancer cell-specific targeting.
Chem. Biol. 12, 35–43.
689
dx.doi.org/10.1021/bc100360v |Bioconjugate Chem. 2011, 22, 679–689