High control, fast growth OEG-based dendron synthesis via a sequential two-step process of copper-free diazo transfer and click chemistry

Article abstract of DOI:10.1021/ma500166e

Dendrons up to generation 3 were quickly synthesized using two sequential steps of diazo transfer and click chemistry reactions. The building blocks for dendron synthesis were made by incorporating monodisperse oligoethylene glycol (OEG) chains of exact length onto diethyelenetriaminepentaacetic acid (DTPA)-derived moieties to obtain first-generation dendrons. Afterward, the amine groups of the first-generation core dendron were converted to azides by means of the diazo transfer reaction. The second- and third-generation dendrons were constructed using the copper-catalyzed azide-alkyne cycloaddition (CuAAC). Hemocompatibility studies including hemolysis, red blood cell morphology, cell count and size distribution, hemostasis, and complement system activation were performed to evaluate their safety in blood.

Full text of DOI:10.1021/ma500166e

Article
pubs.acs.org/Macromolecules
High Control, Fast Growth OEG-Based Dendron Synthesis via a
Sequential Two-Step Process of Copper-Free Diazo Transfer and
Click Chemistry
Peter Fransen,†,§ Daniel Pulido,‡,§ Chantal Sevrin,∥ Christian Grandfils,∥ Fernando Albericio,*,†,§,⊥,#
and Miriam Royo*,‡,§
†Chemistry & Molecular Pharmacology Institute for Research in Biomedicine, Baldiri Reixac 10, 08028 Barcelona, Spain
‡Combinatorial Chemistry Barcelona Science Park, Baldiri Reixac 10, 08028 Barcelona, Spain
∥
§CIBER-BBN, Networking Center on Bioengineering, Biomaterials and Nanomedicine and Interfacultary Center of Biomaterials
́
̂
(CEIB), Liege University, Allee du 6 aout, 11,4000 Liege, Belgium
⊥Department of Organic Chemistry, University of BarcelonaMartí, i Franques
̀
1-11, 08028 Barcelona, Spain
#School of Chemistry & Physics, University of KwaZulua-Natal, 4001 Durban, South Africa
S
* Supporting Information
ABSTRACT: Dendrons up to generation 3 were quickly synthesized using two
sequential steps of diazo transfer and click chemistry reactions. The building
blocks for dendron synthesis were made by incorporating monodisperse
oligoethylene glycol (OEG) chains of exact length onto diethyelenetriaminepenta-
acetic acid (DTPA)-derived moieties to obtain first-generation dendrons.
Afterward, the amine groups of the first-generation core dendron were converted
to azides by means of the diazo transfer reaction. The second- and third-
generation dendrons were constructed using the copper-catalyzed azide−alkyne
cycloaddition (CuAAC). Hemocompatibility studies including hemolysis, red
blood cell morphology, cell count and size distribution, hemostasis, and
complement system activation were performed to evaluate their safety in blood.
in complex systems such as dendrimers4 and entire proteins.5 Our
INTRODUCTION
■
laboratory has previous experience in the synthesis and application
of OEG (oligoethylene glycol)-based dendrons functionalized with
amines.6 These dendritic structures are formed by means of amide
bond formation; however, this approach has the drawback of
incomplete reactions and tedious purification and is therefore not
the most effective method. Given that click chemistry has been
proven to be an effective tool for constructing dendrimers in other
systems and that the diazo transfer reaction has been used effectively
to decorate the surfaces of higher generation dendrimers via amine−
azide transformation and subsequent biofunctionalization by click
chemistry, we hypothesized that the combination of these two
synthetic methods would allow us to construct higher generation
OEG-based dendrons with greater efficacy (see Scheme 1).
Dendrimers show precise architecture, a branched globular
structure, and the capacity to host an abundance of functional
groups. These features have led to an increased interest in their
use for biomedical applications.1 Dendrimers consist of a central
core from which branching units proliferate in a treelike fashion
yielding molecules of higher generations after each synthetic step.
Since its introduction in 2001 by Sharpless et al., the Huisgen
1,3-dipolar cycloaddition between azides and alkynes (CuAAC)
has been widely applied in the fields of chemical biology,
combinatorial chemistry, and material science.2 The capacity of
this specific type of click chemistry to produce high (almost
quantitive) yields without producing difficult-to-remove byproducts
makes the CuAAC an efficient tool for the construction of complex
macromolecules such as dendrimers.3 The click reaction has a high
thermodynamic driving force, and therefore, when applied to
dendrimers, it favors their growth in spite of the steric hindrance
caused by the sheer size of the molecules. Also, because of the lack of
byproducts produced, the formed dendrimers are easier to purify. In
order to decorate the surfaces of a dendritic structure via click
chemistry, azide functional groups need to be introduced to serve as
handles onto which to click the decorative structures.
EXPERIMENTAL SECTION
■
Synthesis of Building Block Dendrons (1 and 2). The synthesis
of compounds 1 and 2 has been described in previous work performed
by our research group.6,7 Nevertheless, the experimental procedure of
the synthesis can be found in the Supporting Information.
Received: January 21, 2014
Revised: March 17, 2014
Published: April 1, 2014
The mild and facile diazo transfer reaction applied to amines is
an effective method to convert amines selectively to azides even
© 2014 American Chemical Society
2585
dx.doi.org/10.1021/ma500166e | Macromolecules 2014, 47, 2585−2591

Macromolecules
Article
Scheme 1. Sequential Steps for Dendrimer Construction: (a) Deprotection with TFA; (b) Diazo Transfer Reaction; (c) CuAAC
Synthesis of Diazo Transfer Agent. The diazo transfer agent was
synthesized as described in the literature by Goddard-Borger and Stick8
except for the final acidic treatment which was performed with 4 M HCl
in dioxane.
Synthesis of Bn-G1-N3 (3). The Boc-protected G1 dendron 1
(500 mg, 0.2955 mmol) was dissolved in a 5 mL solution of TFA/H2O
(95:5) and stirred for 1 h. Subsequently, the TFA was evaporated using a
flow of N2, and the product was precipitated in methyl tert-butyl ether.
After decanting the methyl tert-butyl ether, the pellet containing the
deprotected dendron was dissolved in 10 mL of DMF and the diazo
transfer agent (255 mg, 1.21 mmol) was added. The basicity was
adjusted to pH 8−8.5 at regular intervals during the reaction by the
addition of 1 M NaOH, and the solution was allowed to stir for 1−2 h at
room temperature; meanwhile, the reaction was monitored using
2586
dx.doi.org/10.1021/ma500166e | Macromolecules 2014, 47, 2585−2591

Macromolecules
Article
HPLC-PDA. Afterward, the reaction mixture was added to CH2Cl2
(50 mL) and washed with aqueous brine (3 × 50 mL) acidified to pH
3−4. The organic phase was evaporated, and the resulting crude was
transferred to a 50 mL tube dissolved in CH2Cl2 (10 mL) and hexane
(40 mL) was added. The mixture was stirred vigorously and centrifuged.
The supernatant was discarded, and the remaining oily precipitate
corresponded to the crude compound. The crude product was purified
with flash chromatography over basis alumina oxide and was eluted with
1% MeOH in DCM as the eluent to yield the azide terminated dendron
3 (367 mg, 89%). 1H NMR (400 MHz, CDCl3): δ = 1.77 (q, J = 6.35 Hz,
8 H), 1.84 (q, J = 6.44 Hz, 8 H), 2.64 (m, 8 H), 3.14 (s, 8 H), 3.26−3.33
(m, 8 H), 3.33−3.41 (m, 10 H), 3.45−3.53 (m, 16 H), 3.56 (m, 16 H),
3.59 (m, 16 H), 5.11 (s, 2 H), 7.33 (m, 5 H), and 7.57 (bs, NH). 13C
NMR (100 MHz, CDCl3): δ = 29.04, 29.42, 36.99, 48.37, 52.46, 53.38,
55.44, 59.08, 66.50, 67.80, 69.27, 70.10, 70.24, 70.39, 70.45, 128.27,
128.45, 128.62, 135.374, 170.66, and 171.26. Theoretical mass for
[C61H109N19O18]+: 1396.63; found by LC-MS: 699.12 (M + 2)/2 and by
HRMS (ES+): 1396.8271.
170.56. Theoretical mass for [C281H521N95O82]+: 6538.9519; found by
LC-MS: 936.48 (M + 7)/7, 1092.41 (M + 6)/6, 1310.76 (M + 5)/5,
1638.32 (M + 4)/4, and by HRMS (ES+): 6538.9725.
Synthesis of Bn-G3-Boc (6). The azide-terminated dendrimer
Bn-G2-N3 5 (20 mg, 0.031 mmol) and the alkyne dendron 4 (82 mg,
0.53 mmol) were dissolved in 10 mL of anhydrous CH2Cl2 in a two-
necked round-bottom flask purged with N2. DIEA (94 μL, 0.53 mmol)
was added followed by the addition of CuCl (52 mg, 0.53 mmol). The
reaction mixture slowly turned blue. The solution was stirred
magnetically for several hours, and the disappearance of the alkyne-
functionalized building block dendron was monitored by HPLC-MS.
When the reaction was finished, 40 mL of an aqueous 0.05 M EDTA 5%
NaHCO3 solution was added. The mixture was vigorously stirred for a
couple of minutes, and the two layers were separated. The organic phase
was collected, and this process was repeated twice in order to remove the
free copper as well the copper entrapped in the core of the dendron.
Afterward, the organic phase was evaporated, and the crude was dialyzed
in EtOH/H2O (1:1) using a 3 kDa MWCO membrane to remove the
excess building block and the Boc-terminated third-generation
dendrimer 6 was obtained (85 mg, 84%). 1H NMR (400 MHz, CDCl3):
δ = 1.42 (s, 576 H), 1.71−1.82 (m, 296 H), 2.10−2.19 (m, 40 H), 2.50−
2.62 (m, 84 H), 2.62−2.76 (m, 84 H), 3.15−3.24 (m, 336 H), 3.31 (m, 168
H), 3.38−3.53 (m, 336 H), 3.59 (m, 336 H), 3.63 (m, 336 H), 3.76 (s, 20
H), 4.46 (t, J = 6.81 Hz, 40 H), 7.34 (m, 5 H), 7.61 (s, 20 H), and 7.85 (bs).
13C NMR (100 MHz, CDCl3): δ = 28.57, 29.62, 29.80, 37.04, 38.54, 47.45,
48.01, 51.53, 52.84, 59.03, 67.64, 69.38, 69.56, 70.27, 70.58, 70.60, 78.95,
123.24, 128.39, 128.77, 143.56, 156.17, 170.96.
Synthesis of Bn-G2-Boc (4). The azide-terminated dendron Bn-
G1-N3 3 (0.200 g, 0.14 mmol) and the Boc-terminated dendron Alkyne-
G1-Boc 2 (0.905 g, 0.57 mmol) were dissolved in 40 mL of anhydrous
CH2Cl2 in a two-necked round-bottom flask purged with N2. DIEA
(0.1 mL, 0.57 mmol) was added followed by the addition of CuCl
(42 mg, 0.42 mmol). The reaction mixture slowly turned blue. The
solution was allowed to stir for 1 h at room temperature; meanwhile, the
reaction was monitored using HPLC-PDA and LC-MS. When the
reaction was finished, 40 mL of an aqueous 0.05 M EDTA 5% NaHCO3
solution was added. The mixture was vigorously stirred for a couple of
minutes, and the two layers were separated. The organic phase was
collected, and this washing process was repeated twice in order to
remove the free copper as well the copper entrapped in the core of the
dendron. Afterward, the organic phase was evaporated to yield the Boc-
Synthesis of Dendron Library. The synthesis and characterization
of the small dendron library used in the hemocompatibility studies are
described in the Supporting Information.
Dynamic Light Scattering (DLS). For dendron size determination,
DLS measurements were performed on a Malvern Zetasizer Nano-S
Zen1600 using an He−Ne 125 mW 633 nm laser. The refractive index of
the material and the dispersant were set to 1.33. The dendrons were
dissolved in a 0.2 M PBS 0.05 M NaCl buffer at a concentration of 5 mg/
mL and filtered with a 0.2 μm filter to remove dust particles. Twenty
runs per measurement were performed, and the measurements were
carried out in triplicate.
ζ-Potential Experiments. Measurements for ζ-potentials were
made with a Malvern ZetaSizer Nano ZS instrument. Dendrons
dissolved in distilled water at a 10 mg/mL concentration and passed
through a 0.2 μm filter were placed in a specific cuvette, and certain
parameters of the refractive index and the absorption coefficient of the
material and the viscosity of the solvent were introduced into the
software.
Size Exclusion Chromatography (SEC). SEC was carried out on a
Waters instrument comprising of a Ultrahydrogel 250 column, 6 μm,
7.8 × 300 mm, a separation module (Waters 2695), an automatic
injector, and a photodiode array detector (Waters 2298). Data were
managed with Empower 2 software. UV detection was performed at
210 nm, and a 0.2 M PBS, 0.05 M NaCl buffer (pH 7.2) was used as the
mobile phase to perform SEC at a flow rate of 1.0 mL/min.
General Considerations Hemocompatibility Studies. In-vitro
hemocompatibility tests of the liposomal formulations were performed
according to ISO standards (10993-4). Normal human blood from
healthy volunteer donors was collected in Terumo Venosafe citrated
tubes (Terumo Europe N. V., Belgium). Experiments were done within 2 h
after blood collection. All tests were performed with the agreement of the
1
terminated second generation dendron 4 (1.010 g, 93%). H NMR
(400 MHz, CDCl3): δ = 1.43 (s, 144 H), 1.71−1.82 (m, 72 H), 2.16 (m,
8 H), 2.57 (t, J = 5.4 Hz, 20 H), 2.68 (t, J = 5.4 Hz, 20 H), 3.15−3.24 (m,
72 H), 3.28−3.35 (m, 40 H), 3.41 (s, 2 H), 3.42−3.46 (m, 8 H), 3.48−
3.53 (m, 72 H), 3.56−3.61 (m, 80 H), 3.6−3.65 (80 H), 3.72 (s, 8 H),
4.46 (t, J = 13.7 Hz, 8 H), 5.12 (s, 2 H), 7.34 (m, 5 H), 7.61 (s, 4 H), and
7.86 (bs, NH). 13C NMR (100 MHz, CDCl3): δ = 28.56, 29.62, 29.79,
30.43, 37.06, 38.54, 47.44, 48.02, 51.53, 52.84, 59.01, 66.06, 67.44, 69.33,
69.39, 59.56, 70.26, 70.41, 70.51, 70.57, 70.60, 78.93, 123.38, 128.38, 128.59,
128.76, 123.52, 143.58, 156.16, 170.84, and 170.93. Theoretical mass for
[C361H681N63O114]+: 7723.9422; found by LC-MS: 1106.19 (M + 7)/7,
1290.41 (M + 6)/6, 1548.31 (M + 5)/5 and by HRMS (ES+): 7723.9012.
Synthesis of Bn-G2-N3 (5). The Boc-protected G2 dendron 4 (200
mg, 0.026 mmol) was dissolved in a solution of TFA/H2O (95:5) and
stirred for 1 h. Subsequently, the TFA was evaporated using a flow of N2,
and the product was precipitated in methyl tert-butyl. After decanting
the methyl tert-butyl ether, the pellet containing the deprotected
dendron was dissolved in 10 mL of DMF, and the diazo transfer agent
(93 mg, 0.44 mmol) was added. The basicity was adjusted to pH 8−8.5
at regular intervals during the reaction by the addition of 1 M NaOH,
and the solution was allowed to stir for 3 h at room temperature;
meanwhile, the reaction was monitored using HPLC-PDA. Afterward,
the reaction mixture was added to CH2Cl2 (50 mL) and washed three
times with an aqueous brine (50 mL) acidified to pH 3−4. The organic
phase was evaporated, and the resulting crude was transferred to a 50 mL
tube dissolved in CH2Cl2 (10 mL) and hexane (40 mL) was added. The
mixture was stirred vigorously and centrifuged. The supernatant was
discarded, and the remaining oily precipitate corresponded to the crude
compound. The pure compound was isolated by means of semiprep
purification using a Sunfire C18 column with a gradient of 20−80% CH3CN
̀
local ethical committee of the Medicine Faculty of the University of Liege.
Blood Smears for the Control of the RBC (Red Blood Cell)
Morphology. Dendrons were dissolved in PBS (or DMSO/PBS (1:9))
at a 1000 μM concentration and further diluted in whole blood in order
to obtain final dendron concentrations of 100, 10, 1, and 0.1 μM.
Samples were incubated for 15 min at 37 °C under lateral agitation
(250 rpm). After blood incubation, 5 μL of the blood was withdrawn and
spread on a microscopy glass slide. Blood cells were observed with an
Olympus Provis microscope at 20× and 50× magnification in
transmission mode. At least two representative pictures were acquired
per sample with a digital camera (VisiCam (5 megapixels), VWR
International).
1
in H2O, rendering compound 5 (43 mg, 25%). H NMR (400 MHz,
CDCl3): δ = 1.74 (q, J = 6.55 Hz, 40 H), 1.81 (q, J = 6.26, 32 H), 2.14 (m, 8
H), 3.11 (bs, 20 H), 3.19−3.31 (m, 60 H), 3.35 (t, J = 6.70 Hz, 32 H), 3.43
(bs, 40 H), 3.45−3.52 (m, 82 H), 3.52−3.62 (m, 160 H), 3.85 (s, 8 H), 4.47
(m, 8 H), 5.12 (s, 2 H), 7.31 (m, 5 H), 7.88 (bs, NH), and 8.20 (s, 4 H). 13C
NMR (100 MHz, CDCl3): δ = 29.01, 29.17, 29.97, 37.13, 45.99, 47.58,
48.38, 49.82, 50.89, 53.41, 57.76, 67.09, 67.81, 68.76, 69.10, 70.01, 70.20,
70.33, 70.40, 115.04, 117.95, 127.28, 128.36, 128.62, 135.46, 160.89, and
2587
dx.doi.org/10.1021/ma500166e | Macromolecules 2014, 47, 2585−2591

Macromolecules
Article
Figure 1. Molecular structure of first-generation dendrons with either a benzylester carboxylic acid (1) or an alkyne (2) in the focal point.
Scheme 2. Dendron Synthesis: (a) TFA/H2O (95:5); (b) NaOH (aq, 1 M), DMF; (c) Dendron 2, CuCl, DIEA, DCM
Hemolysis. Dendron solutions and blood were prepared and
incubated as described above. The hemolytic test was performed
following Standard Practice for Assessment of Hemolytic Properties of
Materials (ASTM designation F 756-00). Briefly, after incubation, the
samples were centrifuged at 600g for 5 min at rt, and supernatants were
collected and mixed with the cyanmethemoglobin reagent. The
hemoglobin released was measured by reading the absorbance of
100-fold dilution of whole blood in Drabkin’s reagent at 540 nm in a
microplate reader (Anthos HT III, type 12600, Anthos, Salzburg, AU). A
calibration curve was established using bovine hemoglobin as the
standard. Saponine (0.8 mg mL−1), and PBS were used as positive and
negative controls, respectively. Hemolysis was expressed as the
percentage of hemoglobin released to total hemoglobin content, taking
the positive control as 100% of hemolysis. The tests were done in
triplicate. In view to assess the influence of plasma protein on the
hemolytic action of the dendrons, this test was also performed using
washed RBCs (4% RBC suspension) instead of whole blood.
Count and Size Distribution of RBCs, Platelets, and White
Blood Cells. Dendron solutions and blood were prepared and
incubated as described before. After 15 min of incubation, blood cells
were counted and their size distribution was determined with CELL-
DYN 18 Emerald (Abbott Diagnostics). Three analyses were conducted
per sample.
Complement Activation. Complement activation was assesed
using the Human C3a ELISA kit for quantification of Human C3a-des-
Arg (Becton Dickinson). After a 15 min incubation of blood and
dendron mixtures, EDTA (1 mM final) was added to block any future
complement activation. Samples were centrifuged at 2000g for 5 min at
rt, and supernatants were used for the analysis of complement activation
following the kit protocol (BD OptEIA, Human C3a ELISA, Instruction
Manual. Cat. No.550499). Absorbance was measured at 450 nm with a
microplate reader (Anthos HT III, type 12600). Plasma containing
2 mg mL−1 of Zymosan was taken as a positive control and plasma
without additives as a negative control. The concentration of C3a was
expressed as a percentage of activation by reference to the negative
control incubated set at a value of 100% of complement activation.
Measurements were done in duplicate.
supplied with the specific activators of coagulation (thromboplastin).
Prothrombin time (PT), to evaluate the extrinsic pathway, and activated
partial thromboplastin time (aPTT), to evaluate the intrinsic pathway,
were measured directly with a Dade Behring coagulation timer analyzer
(BCT) (Siemens Healthcare Diagnostics NV/SA, Belgium) using
commercial reagents (Thromborel S, Dade Behring/Siemens, for PT
determination and C.K. PREST kit, Roche Diagnostics, France, for
aPTT). Kaolin reagent was used as a positive control and PBS as a
negative control. Clotting time was measured for each sample, and
coagulation capacity was expressed as a percentage, taking the value of
standard human plasma (Dade Behring/Siemens) as 100%. Measure-
ments were done in duplicate.
RESULTS AND DISCUSSION
■
The building of dendrons of generation 2 and 3 consisted of two
main steps: (1) conversion of amines to azides via the diazo
transfer reaction and (2) incorporation of the alkyne building
block dendron by CuAAC. To achieve this goal, the synthesis of
two types of building block dendrons was required: the core
dendron (1) and the alkyne-functionalized building block
dendron (2) (see Figure 1). The two OEG-based dendrons
consisted of a branching unit derived from diethylene triamine
pentaacetic acid (DTPA), to which monodisperse OEG-chains,
Boc-TOTA (1-(tert-butoxycarbonylamino)-4,7,10-trioxa-13-tri-
decanamine) were incorporated. The DTPA derivatives, and the
corresponding first-generation dendrons, contain either an
alkyne or a benzyl ester carboxylic acid in their focal point and
were synthesized as described in previous work.6,7
The conversion of amines to azides was carried out with
imidazole-1-sulfonyl azide hydrochloride as the diazo transfer
agent (see Scheme 2). This reagent was developed by Goddard-
Borger and Stick as a nonexplosive, shelf-stable, and efficient
alternative to TfN3.8 With the aforementioned reagent, the diazo
transfer reaction is usually performed in a protic solvent such as
methanol, ethanol, or 2-propanol, using K2CO3 or NEt3 as base
and CuSO4 as the source of CuII catalyst. At first, in our case, the
solvent of choice was either ethanol or 2-propanol because
methanol leads to transesterification of the benzylic ester of the
Coagulation Experiments. Whole blood and dendrimer solutions
were mixed and incubated as described before. Samples were
centrifuged at 2000g for 5 min at rt, and the supernatants were
collected, recalcified to reverse the effect of citrate anticoagulant, and
2588
dx.doi.org/10.1021/ma500166e | Macromolecules 2014, 47, 2585−2591

Macromolecules
Article
Figure 2. 1H NMR spectra of first-, second-, and third-generation amine-terminated dendrons (in D2O).
core dendron under basic conditions. Given that the dendron
contains a chelating DTPA derivative, a portion of the CuSO4
added to the reaction mixture was trapped by coordination in the
DTPA derivative, which was detectable by HPLC-MS, thereby
impeding its catalytic performance. Therefore, higher amounts of
CuSO4 were required to drive the reaction to completion. Other
metals, such as ZnII, CoII, and NiII, were tested for the catalysis of
the diazo transfer reaction, but none gave any improvement. It
came to our attention that this new diazo transfer agent is so
effective that no CuII catalyst is needed to convert the amines to
azides, as reported in the literature for the introduction of azides
into proteins.9 In fact, the choice of solvent and the pH are the
factors that influence the rate of the diazo transfer reaction the
most. Polar solvents such as DMF and H2O turn out to be more
suitable solvents than the conventionally used protic solvents,
such as MeOH. Maintaining the pH of the reaction mixture
between 8 and 8.5 was crucial for clean conversion of the amines
to azides. The azide dendrons 3 and 5 were obtained with 89%
and 25% yields in the diazotransfer reaction, respectively. Yield
differences can be attributed to the high ratio of amine groups
that are transformed to azides to obtain dendron 5 (4 times more
than dendron 3) and required longer reaction time, causing the
appearance of more undesired impurities. Therefore, dendron 5
could only be isolated by reverse phase chromatography, in contrast
to dendron 3, which was easily purified by flash chromatography.
Despite avoiding Cu to catalyze the diazo transfer reaction, the
chelating capacity of the DTPA derivative dendrons still
complicates the CuAAC as the CuI required for the click
reaction is sequestered. In order to make the reaction proceed,
0.5 equiv of CuI catalyst per DTPA derivative dendron was
needed. This amount was determined by performing a series of
trial experiments in which the amount of CuI was increased in a
stepwise manner until the click reaction occurred. The most
conventional procedure for click chemistry involves the use of
CuSO4 as the source of copper and sodium ascorbate as the agent
to reduce in situ the CuII to CuI species which actually serves as
the catalyst. Two or more equivalents of ascorbic acid with
respect to CuSO4 are generally used to reduce CuII sufficiently. In
our case, the minimal required amounts of CuSO4 and sodium
ascorbate were 2.5 and 5 equiv (with respect to the core dendron
with azides), respectively, in order to build the second-generation
dendrimer. Using these conditions, the dendron G2 was formed
but was accompanied by other impurities (60% purity as
determined by HPLC). The best results were achieved using
3 equiv of CuCl, a direct source of CuI, together with DIEA,
which makes the alkyne more accessible,10 in dichloromethane
under a N2 atmosphere. The entrapped copper was removed
afterward by washes with an EDTA solution. Using these
conditions, the dendron G2 (4) was obtained at a scale of 1 g at
high purity (>95%) and with a CuAAC reaction yield of 93%. To
build the third-generation dendron (6), the same CuAAC
conditions were used with similar results (84% CuAAC reaction
yield). The global yield of G2 (4) and G3 dendron (6) from Boc
protected G1 (1) was 87% and 35%, respectively. The formation
of the second-generation G2 was confirmed by NMR and ES-
MS. The use of mass spectrometry to confirm the formation of
the third-generation dendron G3 was not possible; therefore, this
molecule was characterized by NMR (see Figure 2). Integration
of the peaks indicated a total of 20 protons corresponding to the
triazole signal. Integration of the signals corresponding to the
protons adjacent to the triazole ring also gave the correct values.
This corresponding increase in molecular weight of the
dendrons was also confirmed by aqueous SEC of the free-amine
surface dendrons G1, G2, and G3 which eluted at retention times
2589
dx.doi.org/10.1021/ma500166e | Macromolecules 2014, 47, 2585−2591

Macromolecules
Article
anticipated cytotoxicity of large dendrons is the result of their size
or of their surface charge. Furthermore, the acetyl analogues
could be suitable mimics of the potential structures to be
synthesized and used for further applications. Dendrons
functionalized with azides were evaluated for their biocompat-
ibility because they might serve as cross-linkers for biomaterials.
Hemocompatibility was assessed following ISO-10.993, and
these assays should be considered as prescreening tests. They
consisted of microscopic evaluation of red blood cell morphology,
hemolysis testing, blood cell counting and size distribution,
assaying the activation of the complement immune system, and
finally evaluating hemostasis (intrinsic and extrinsic pathways).
The results of the blood cell counting and size distribution testing
showed that no cell type (erythrocytes, leukocytes, and platelets)
was affected by the presence of the dendrimers whatever the
concentration. The total count number and size distribution did
not show any deviation from the control values (results in
Supporting Information). In thecaseofthe erythrocytes, this result
was also confirmed by the hemolysis assay and microscopic
observation of cell morphology (see Figure 4).
Figure 3. SEC traces of dendrons G1-NH2, G2-NH2, and G3-NH2.
of 12.0, 10.4, and 9.6 min, respectively (see Figure 3). The
presence of a small shoulder in SEC traces of the second- and
third-generation dendron indicates that to a certain degree there
are defects in the formed structures.
First-, second-, and third-generation dendrons with free or
acetylated surface amines (G1-NH2 (7), G1-Ac (8), G2-NH2
(9), G2-Ac (10), G3-NH2 (11), and G3-Ac (12)) were chosen to
determine their size and the electrical potential at the dendron
surface. Dynamic light scattering (DLS) measurements (see
Table 1) showed the relative difference in size between amine-
functionalized or acetylated dendrons G1, G2, and G3 with no
apparent variation between the surface functionalities. The size
of the OEG-dendrons G1 and G2 was consistent with PAMAM
dendrimers of generations G0 (1.5 nm) and G3 (3.6 nm),
respectively. The third-generation OEG-dendron has a molec-
ular weight (25−28 kDa) comparable to a G5 PAMAM
dendrimer (28 kDa) but in terms of hydrodynamic diameter is
similar to a G7 PAMAM dendrimer (8.1 nm). This means that
with an increase in generation the OEG-dendrons are becoming
less dense than the PAMAM dendrimers.
Figure 4. Hemolysis induced by dendrons in whole blood, expressed as
percent to whole blood hemoglobin content. Inset: microscopic images
of erythrocytes after dendron incubation.
The electrical potential at the dendron surface was determined
by ζ-potential measurements. As expected, the influence of the
surface functionalization was clear as the dendrons with the free
amines in the exterior had a positive ζ-potential ranging from 12
to 43 mV, and the number of positive charges is directly related to
an increase in the ζ-potential. The dendrons with the acetyl
groups, on the other hand, were practically neutral.
To evaluate whether OEG-based dendrons synthesized in this
study can be used for further biomedical applications, their
biocompatibilty was assessed by means of hemocompatibility
assays. For this purpose, we prepared a small library of eight
dendrons of generation 1−3 with different functional groups in
the periphery: G1-NH2 (7), G1-Ac (8), G1-N3 (3); G2-NH2 (9),
G2-Ac (10), G2-N3 (5); G3-NH2 (11), G3-Ac (12). Despite its
successful synthesis, the dendron G3-NH2 was omitted from
these studies because of its insolubility in the isotonic buffer at
the assessed doses. Previous results in the literature show that
large cationic dendrimers with a high number of free amines can
negatively interfere with human blood.11 Therefore, we also
included acetylated dendrons in order to verify whether the
No significant increase in the hemolysis rate was detected. In
addition, the erythrocytes remained round after being in contact
with the dendrimers. These results indicate that the dendrimers
did not damage the integrity of the cellular components of
blooda result that is particularly interesting regarding platelets,
which are known to be particularly sensitive to polycations. The
coagulation cascade leading to the formation of blood clots
involves two separate, but linked, pathways: the intrinsic and the
extrinsic pathway. The former, as assayed by the PT assay, was
not inhibited by any type of dendrimer. The intrinsic pathway
was assayed by the aPTT assay. In this case the polycationic
dendrimers G2-NH2 caused up to 50% inhibition at the highest
dose of 100 μM. This inhibition, also found with other synthetic
polycations,12 was most probably caused by electrostatic
interference with the signaling cascade as a result of electrostatic
interactions between the polycationic dendrimers and several
negatively charged coagulation factors involved in this pathway.
Finally, activation of the complement immune system was tested
by measuring the concentration of the anaphylatoxin C3a by
Table 1. Zeta-Potential and Measured Diameters of the Dendrons
G1
G2
G3
G1-NH2
G1-Ac
G2-NH2
G2-Ac
G3-NH2
G3-Ac
ξ-potential (mV)
12.3 1.5
1.2 0.1
−1.6 0.6
1.2 0.12
33.9 3.0
4.2 0.2
−5.65 1.3
4.2 0.1
43.5 0.6
8.7 1.2
−4.2 0.2
8.4 1.9
diameter (nm)
2590
dx.doi.org/10.1021/ma500166e | Macromolecules 2014, 47, 2585−2591

Macromolecules
Article
Applications of Dendrimers: A Tutorial. Chem. Soc. Rev. 2011, 40, 173−
190.
ELISA. In similar fashion to coagulation, the complement system
was activated by the polycationic dendrimers G2-NH2, showing
activation at 100 μM. The other types of dendrimers did not
cause any significant activation In comparison to the typical
hemoreactivity of polycations, these dendrimers showed better
hemocompatibility in in vitro tests at a concentration range
typically used for pharmacological purposes (0.1−100 μM).
In conclusion, the platform described here allows the synthesis
of dendrons of low (G1), medium (G2), and high (G3)
molecular weight with potential use in biomedical applications,
including intravenous administration. The dendrons can be adapted
to specific requirements by incorporation of OEG-chains of variable
length and distinct functional moieties due to the orthogonality of
the click reaction. To the best of our knowledge, this is the first study
to report the use of the diazo transfer reaction to build dendritic
molecules. Also, this reaction allows switching between azides and
amines, a feature enabling the functionalization of the dendrons
through click chemistry (for G1 and G2) and amide bond formation
(G1, G2 and G3). Finally, the DTPA-derived units that form part of
the dendritic scaffold offer metal chelation sites, a feature that
deserves further attention for potential use in medical imaging.
(2) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Click Chemistry: Diverse
Chemical Function from a Few Good Reactions. Angew. Chem., Int. Ed.
2001, 40, 2004−2021.
(3) (a) Wu, P.; Feldman, A. K.; Nugent, A. K.; Hawker, C. J.; Scheel, A.;
́
Voit, B.; Pyun, J.; Frechet, J. M. J.; Sharpless, K. B.; Fokin, V. V.
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. 2004, 43, 3928−3932. (b) Franc, G.;
Kakkar, A. K. “Click” Methodologies: Efficient, Simple and Greener
Routes to Design Dendrimers. Chem. Soc. Rev. 2010, 39, 1536−1544.
(c) Franc, G.; Kakkar, A. Dendrimer Design Using CuI-Catalyzed
Alkyne-Azide “Click-Chemistry”. Chem. Commun. 2008, 5267−5276.
(4) Papadopoulos, A.; Shiao, T. C.; Roy, R. Diazo Transfer and Click
Chemistry in the Solid Phase Syntheses of Lysine-Based Glycoden-
drimers as Antagonists against Escherichia coli FimH. Mol. Pharmaceu-
tics 2011, 9, 394−403.
(5) van Dongen, S. F. M.; Teeuwen, R. L. M.; Nallani, M.; van Berkel, S.
S.; Cornelissen, J. J. L. M.; Nolte, R. J. M.; van Hest, J. C. M. Single-Step
Azide Introduction in Proteins via an Aqueous Diazo Transfer.
Bioconjugate Chem. 2008, 20, 20−23.
(6) Simon-Gracia, L.; Pulido, D.; Sevrin, C.; Grandfils, C.; Albericio, F.;
Royo, M. Biocompatible, Multifunctional, and Well-Defined OEG-
Based Dendritic Platforms for Biomedical Applications. Org. Biomol.
Chem. 2013, 11, 4109−4121.
ASSOCIATED CONTENT
■
S
* Supporting Information
(7) San
́
chez-Nieves, J.; Fransen, P.; Pulido, D.; Lorente, R.; Munoz-
̃
́
Fernandez, M. A.; Albericio, F.; Royo, M.; Gomez, R.; Javier de la Mata,
́
́
Experimental procedures and characterization data for all the
new compounds, i.e., branching units and dendrons, and
hemocompatibility assays results for all dendrons. This material
is available free of charge via the Internet at http://pubs.acs.org.
F. Amphiphilic Cationic Carbosilane-PEG Dendrimers: Synthesis and
Applications in Gene Therapy. Eur. J. Med. Chem. 2014, 76, 43−52.
(8) Goddard-Borger, E. D.; Stick, R. V. An Efficient, Inexpensive, and
Shelf-Stable Diazotransfer Reagent: Imidazole-1-sulfonyl Azide Hydro-
chloride. Org. Lett. 2007, 9, 3797−3800.
AUTHOR INFORMATION
■
(9) (a) Schoffelen, S.; van Eldijk, M. B.; Rooijakkers, B.; Raijmakers, R.;
Heck, A. J. R.; van Hest, J. C. M. Metal-Free and pH-Controlled
Introduction of Azides in Proteins. Chem. Sci. 2011, 2, 701−705.
Corresponding Author
*E-mail: mroyo@pcb.ub.es (M.R.).
(b) Hansen, M. B.; van Gurp, T. H. M.; van Hest, J. C. M.; Lowik, D. W.
̈
Notes
P. M. Simple and Efficient Solid-Phase Preparation of Azido-peptides.
Org. Lett. 2012, 14, 2330−2333.
The authors declare no competing financial interest.
(10) Meldal, M.; Tornøe, C. W. Cu-Catalyzed Azide−Alkyne
Cycloaddition. Chem. Rev. 2008, 108 (8), 2952−3015.
ACKNOWLEDGMENTS
■
(11) (a) Ziemba, B.; Matuszko, G.; Bryszewska, M.; Klajnert, B.
Influence of Dendrimers on Red Blood Cells. Cell. Mol. Biol. Lett. 2012,
17, 21−35. (b) Ziemba, B.; Halets, I.; Shcharbin, D.; Appelhans, D.;
Voit, B.; Pieszynski, I.; Bryszewska, M.; Klajnert, B. Influence of Fourth
Generation Poly(propyleneimine) Dendrimers on Blood Cells. J. Biom.
This work was partially supported by (CTQ2008-00177,
SAF2011-30508-C02-01, CTQ2009-07758, CTQ2012-30930).
Institute Carlos III (CB06_01_0074CFA), the Generalitat de
Catalunya (2009SGR1024), CIBER-BBN, and the la Caixa social
program are acknowledged for their financial support.
Mater. Res., Part A 2012, 100A, 2870−2880. (c) Domanski, D. M.;
́
Klajnert, B.; Bryszewska, M. Influence of PAMAM Dendrimers on
Human Red Blood Cells. Bioelectrochemistry 2004, 63, 189−191.
(12) Cerda-Cristerna, B. I.; Flores, H.; Pozos-Guillen, A.; Perez, E.;
Sevrin, C.; Grandfils, C. Hemocompatibility Assessment of Poly(2-
dimethylamino ethylmethacrylate) (PDMAEMA)-Based Polymers. J.
Controlled Release 2011, 153, 269−77.
ABBREVIATIONS
■
aPTT, activated partial thromboplastin assay; BOC-TOTA, 1-
(tert-butoxycarbonylamino)-4,7,10-trioxa-13-tridecanamine;
CuAAC, copper-catalyzed azide−alkyne cycloaddition; DIEA,
N,N-diisopropylethylamine; DLS, dynamic light scattering;
DMF, dimethylformamide; DTPA, diethylenetriaminepenta-
acetic acid; EDTA, ethylenediaminetetraactic acid; ELISA,
enzyme-linked immunosorbent assay; ES-MS, electrospray
mass spectrometry; EtOH, ethanol; HPLC-PDA, high perform-
ance liquid chromatography−photodiode array; HRMS, high
resolution mass spectrometry; LC-MS, liquid chromatography−
mass spectrometry; OEG, oligoethylene glycol; PBS, phosphate
buffered saline; PT, prothrombin time; SEC, size exclusion
chromatography; TFA, trifluoroacetic acid.
REFERENCES
■
(1) (a) Svenson, S.; Tomalia, D. A. Dendrimers in Biomedical
ApplicationsReflections on the Field. Adv. Drug Delivery Rev. 2005,
57, 2106−2129. (b) Mintzer, M. A.; Grinstaff, M. W. Biomedical
2591
dx.doi.org/10.1021/ma500166e | Macromolecules 2014, 47, 2585−2591