Synthesis of poly(ester-amide) dendrimers based on 2,2-Bis(hydroxymethyl) propanoic acid and glycine

Article abstract of DOI:10.1002/pola.27391

Water-soluble, biodegradable, and biocompatible poly(ester-amide) dendrimers with hydroxyl functional groups are synthesized from previously prepared AB₂ adduct of 2,2-bis(hydroxymethyl) propanoic acid (bis-MPA) and glycine as a repeating unit. Two esterification procedures using different coupling reagent/catalyst systems (DCC/DPTS or EDC/DMAP) are studied with respect to efficiency, ease of products purification, and quality of the final products. Both procedures have their own benefits and drawbacks, depending on dendrimer generation. The synthesized poly(ester-amide) dendrimers as well as commercially available bis-MPA dendrimers, poly(ester-amide) hyperbranched polymer, and poly(vinyl alcohol) are used for preparation of solid dispersions of sulfonylurea antidiabetic drug glimepiride to improve its poor water-solubility. In vitro dissolution studies show in comparison with pure glimepiride in crystalline or amorphous form, to the same extent improved glimepiride solubility for solid dispersions based on dendritic polymers, but not for poly(vinyl alcohol). The amount of glimepiride complexed with both dendrimer types increases with dendrimer generation.

Full text of DOI:10.1002/pola.27391

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Synthesis of Poly(ester-amide) Dendrimers based on
2,2-Bis(hydroxymethyl) Propanoic Acid and Glycine
1
2
3
1
ꢀ
ꢀ
David Pahovnik, Anja Cusak, Sebastjan Reven, Ema Zagar
1National Institute of Chemistry, Laboratory for Polymer Chemistry and Technology, Hajdrihova 19, SI-1000 Ljubljana, Slovenia
2EN-FIST Center of Excellence, Dunajska cesta 156, SI-1000 Ljubljana, Slovenia
3
ꢀ
Lek Pharmaceuticals d.d., Sandoz Development Center Slovenia, Verovskova 57, SI-1526 Ljubljana
ꢀ
Correspondence to: E. Zagar (E-mail: ema.zagar@ki.si)
Received 22 April 2014; accepted 28 August 2014; published online 19 September 2014
DOI: 10.1002/pola.27391
ABSTRACT: Water-soluble, biodegradable, and biocompatible
poly(ester-amide) dendrimers with hydroxyl functional groups
are synthesized from previously prepared AB2 adduct of 2,2-
bis(hydroxymethyl) propanoic acid (bis-MPA) and glycine as a
repeating unit. Two esterification procedures using different
coupling reagent/catalyst systems (DCC/DPTS or EDC/DMAP)
are studied with respect to efficiency, ease of products purifica-
tion, and quality of the final products. Both procedures have
their own benefits and drawbacks, depending on dendrimer
generation. The synthesized poly(ester-amide) dendrimers as
well as commercially available bis-MPA dendrimers, poly(es-
ter-amide) hyperbranched polymer, and poly(vinyl alcohol) are
used for preparation of solid dispersions of sulfonylurea anti-
diabetic drug glimepiride to improve its poor water-solubility.
In vitro dissolution studies show in comparison with pure gli-
mepiride in crystalline or amorphous form, to the same extent
improved glimepiride solubility for solid dispersions based on
dendritic polymers, but not for poly(vinyl alcohol). The amount
of glimepiride complexed with both dendrimer types increases
C
V
with dendrimer generation.
2014 Wiley Periodicals, Inc. J.
Polym. Sci., Part A: Polym. Chem. 2014, 52, 3292–3301
KEYWORDS: dendrimers; drug delivery systems; esterification;
hyperbranched; water-soluble polymers
tone), poly(trimethylene carbonate), etc. A group of neutral
aliphatic polyester dendrimers based on 2,2-bis(hydroxy-
methyl) propanoic acid (bis-MPA) and their hybrids with
poly(ethylene glycol), which are water soluble, nonimmuno-
genic, biodegradable, biocompatible and nontoxic,26,27 have
also been widely investigated as candidates for the develop-
ment of anticancer drug delivery systems.24,25,28–39
INTRODUCTION Dendrimers with
a precisely controlled
branched structure reveal several advantages over their lin-
ear counterparts of comparable molar mass, that is, uniform
molar mass distribution, multifunctionality, and in the case
of amphiphilic nature of dendrimers also unimolecular micel-
lar architecture.1,2 These intrinsic properties of dendrimers
translate to their unique chemical and physical properties
(e.g., solubility, viscosity, chemical reactivity, etc.), which
make them interesting as materials for biological application
(scaffolds for drug delivery systems, MRI contrast agent,
materials for tissue engineering, transfection).3–13 Although
many dendritic structures have been synthesized, only few
of them possess adequate water-solubility and biocompatibil-
ity to have a potential for biomedical applications.4,5,12–16 To
optimize the chemical and physical properties of dendrimers
for biomedical applications, various biocompatible monomers
have been applied for their synthesis,17–25 for example, natu-
ral metabolites (amino acids, sugars, a-hydroxy acids, fatty
acids), chemical intermediates in metabolic pathways (suc-
cinic acid, fumaric acid, citric acid, pyruvic acid), and mono-
mers currently used for the preparation of medical-grade
linear polymers, that is, poly(ethylene glycol), poly(caprolac-
The dendrimers based on bis-MPA can be prepared by
either convergent, divergent, or double exponential
approaches.40–44 Esterification reaction for preparation of
polyester dendrimers is most commonly accomplished by
utilizing a carbodiimide coupling reagent (usually N,N’-dicy-
clohexylcarbodiimide, DCC) in the presence of
a cata-
lyst.18,19,43–46 Unfortunately, the use of DCC has displayed
multiple disadvantages related primarily to a side reaction of
DCC with bis-MPA, resulting in formation of the N-acylurea
that have to be removed to ensure good quality of end prod-
ucts. This is of course accompanied by significantly
decreased reaction yields, especially when going to higher
dendrimer generation.41,44 To circumvent the aforementioned
problems related to carbodiimide based esterification, a
Additional Supporting Information may be found in the online version of this article.
C
V
2014 Wiley Periodicals, Inc.
3292
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 3292–3301

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
4-(dimethylamino)pyridinium 4-toluenesulfonate (DPTS) has
been utilized as a catalyst.19,30,44,46
coupling reagent/catalyst systems. The synthesized poly(es-
ter-amide) dendrimers were tested as solubility enhancers
for poorly water-soluble glimepiride drug and the results
were compared with those obtained for the solid dispersions
of glimepiride and commercially available bis-MPA based
dendrimers as well as the solid dispersion of glimepiride
and poly(vinyl alcohol) to explore the role of ester and
hydroxyl groups in complex formation. Additionally, the cyto-
toxicity of the synthesized poly(ester-amide) dendrimers was
evaluated.
Esterification reaction was performed also by
a pre-
activation of the bis-MPA either in the form of acid chloride40
or acid anhydride.41,42,44 The bis-MPA chloride and the bis-
MPA anhydride react with the hydroxyl groups of central
core or dendron/dendrimer under a catalytic amount of
4-dimethylaminopyridine (DMAP) in dichloromethane (DCM)
solution in the presence of triethylamine (TEA) and pyridine,
respectively. The divergent synthetic approach using anhy-
dride activated strategy belongs to one of the most robust
approaches for preparation of bis-MPA dendrimers on a large
scale, in high yields and with little byproducts that are sim-
ply purified by extraction and precipitation.
EXPERIMENTAL
All chemicals and solvents were used as purchased if not
stated otherwise. Commercial dendrimers (the third and the
fourth generations) with hydroxyl functional groups, synthe-
sized from bis-MPA as a repeating unit and 1,1,1-tris(hydrox-
ymethyl)propane as a core, were obtained from Polymer
Factory Sweden AB, Stockholm, Sweden. Poly(vinyl alcohol)
(98–99% hydrolyzed) with the Mn of 2.1 kDa and molar
mass dispersity of 2.2 was purchased from Sigma-Aldrich,
Germany.
Another approach involving precipitation as simple purifica-
tion method is a divergent synthetic route in which the bis-
MPA dendrons are grown up on a linear polystyrene support
in the form of dendritic hybrids.47
Recently, accelerated approaches involving highly selective
and efficient click reactions (copper-catalyzed azide–alkyne
cycloaddition (CuAAC), thiol-ene coupling (TEC), Diels–Alder
cycloaddition (DA), thiol/acrylate Michael addition) with48–50
or without51,52 traditional esterification reaction have been
proposed for the synthesis of bis-MPA type dendrimers with
high yields and purity as well as for dendrimer
functionalization.2,22,53–55
Detailed experimental procedures are available in the Sup-
porting information. In general, two coupling methods
either DCC/DPTS (method A) or EDC/DMAP (method B)
were used for amidation reaction of benzyl protected gly-
cine and acetonide protected bis-MPA to obtain the AB2
adduct. The same two coupling methods were used for the
esterification reaction to prepare the second-generation
dendron as well as the first (AB2 unit from the core), the
second (the second-generation dendron to the core), the
third (the second-generation dendron to the first-
generation dendrimer) and the fourth (the second-
generation dendron to the second-generation dendrimer)
generation dendrimers. The acetonide protection group
was removed by Dowex resin, whereas the benzyl pro-
tected focal group by catalytic hydrogenation.
Our previous work revealed that poly(ester-amide) hyper-
branched polymers enhance water-solubility of the poorly
water-soluble antidiabetic drug glimepiride by preparation of
solid dispersions56 that were finally incorporated into the
tablet formulation.57 Experimental results showed that the
improved glimepiride solubility is due to the formation of a
complex between the glimepiride drug and the hyper-
branched polymer which is stabilized by a hydrogen-bond
interaction between the slightly acidic proton of NH group of
the glimepiride sulfonylurea segment and the carbonyls of
the amide and ester bonds of the hyperbranched polymers,
while the hydroxyl groups did not seem to play a role in the
complex formation.58 This interaction allows molecularly dis-
persed glimepiride within the amorphous hyperbranched
polymers and thereby glimepiride solubility and dissolution
rate are significantly improved.
1H NMR and 13C NMR spectra were recorded in DMSO-d6 on
an Agilent Technologies DD2 spectrometer at 300 and 75
MHz, respectively and with delay time of 5 and 2 sec,
respectively.
Analytical thin layer chromatography (TLC) was per-
formed on Merck 60 F254 precoated silica gel plates. TLC
plates were analyzed by short-wave UV light and/or by
dipping in KMnO4 solution. Flash column chromatography
was performed using Zeochem Silica Gel (ZEOprep 60 Eco
40–63 lm).
Since hyperbranched polymers show distribution in molar
mass and branching pattern that can result in irreproducible
quality of the material from batch to batch synthesis and,
consequently, in variation of loading capacity and kinetics of
drug release, dendrimers are preferred over hyperbranched
polymers. Herein we thus report on a synthesis of different
generation, water-soluble poly(ester-amide) dendrimers with
hydroxyl functional groups from an AB2 adduct of bis-MPA
and glycine as a repeating unit and 1,1,1-tris(hydroxyme-
thyl)propane as a core molecule. The synthesis of den-
drimers was carried out by two parallel carbodiimide based
esterification reactions to determine the efficiency of both
The separations of protected and deprotected dendrimers by
SEC were carried out using an Agilent 1260 HPLC pump and
a PolarGel-L 8 lm analytical column (7.5 mm 3 300 mm)
with a precolumn (Polymer Laboratories, UK) in 0.1 M aque-
ous solution of NaNO3 with added azide. The PolarGel-L col-
umn covers the molar masses up to 30 kDa. For the
detection, we used a multi-angle light-scattering (MALS with
18 angles) detector (DAWN-HELEOS, Wyatt Technology
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 3292–3301
3293

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
cðHPLCÞ
Corp.) and an interferometric refractive index (RI) detector
(Optilab rEX, Wyatt Technology Corp.). The nominal eluent
flow rate was 0.8 mL/min, the injection volume was typically
100 lL, and the mass of the samples injected onto the col-
umn was 150 lg. The determination of absolute Mw and the
calculation of Mn values from MALS detector require a
sample-specific refractive-index increment (dn/dc), which
was determined from the RI response assuming 100% of
sample mass recovery from the column. For the data acquisi-
tion and evaluation, Astra 5.3.4 software (Wyatt Technology
Corp.) was utilized.
LCð%Þ 5
3 cðSDÞ
(1)
cðassayÞ
where LC: loading capacity in %, c(HPLC): concentration of
glimepiride (in lg/mL) determined by HPLC after 60 min
dissolution of solid dispersions in phosphate buffer
(pH 5 6.8), c(assay): theoretical 100% concentration of gli-
mepiride (4.4 lg/mL), calculated using the assay result and
volume of dissolution media, c(SD): concentration of glime-
piride in solid dispersions (5%, w/w).
In Vitro Dissolution Studies
Dissolution studies were performed in phosphate buffꢀer
solution at pH 5 6.8, physiologically relevant media, at 37 C
using an USP Dissolution Tester, Apparatus II (Paddle
method) at a rotation rate of 75 rpm. The tested samples
(glimepiride as well as solid dispersions) were added in the
correct amount directly to a 900 mL phosphate buffer solu-
tion to achieve a final concentration of 4.4 lg/mL (glimepir-
ide). The experiments were performed in triplicates.
Aliquots, each of 2 mL, were withdrawn from the dissolution
medium at a time interval of 60 min. The sample aliquots
were withdrawn through a syringe and filtered through the
Millipore filter (0.45 lm, PVDF). The sample aliquots were
analyzed for the dissolved glimepiride content using
reversed-phase HPLC method.
The mass spectra were acquired with a Bruker UltrafleXt-
reme MALDI-TOF mass spectrometer (Bruker Daltonik, Bre-
men, Germany) equipped with
a
nitrogen laser
(k 5 337 nm). The linear positive ion mode was used to
acquire the mass spectra. The calibration was made exter-
nally with a Peptide calibration standard (Bruker Daltonics).
The matrices used were super-DHB (9/1 mixture of 2,5-dihi-
droxybenzoic acid and 2-hydroxy-5-methoxy benzoic acid)
and trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylide-
ne]malononitrile (DCTB), which were supplied by Sigma-
Aldrich. Sodium trifluoroacetic acid was also supplied by
Sigma-Aldrich. A 5 mg/mL water solution of deprotected
dendrimer was mixed with a 20 mg/mL solution of sDHB
matrix and 2.2 mg/mL of cationic agent NaTFA dissolved in
a 50/50 mixture of acetonitrile/water. The volume ratio
between the sample, matrix solution and cationic agent was
3/3/0.5. Then, 1 lL of the final solution was spotted on the
target plate using the dried-droplet method. The protected
dendrimers were dissolved in THF at concentration of
5.0 mg/mL and mixed with a 20 mg/mL solution of DCTB
matrix in THF and a 2.2 mg/mL solution of NaTFA in THF.
The volume ratio between the solutions of protected den-
drimer, matrix and NaTFA was 3/3/0.5. Then, 0.5 lL of the
final solution was spotted on the target plate using the
dried-droplet method.
Assay
The tested samples were accurately weighted (5 mg) into a
10 mL volumetric flask. 5 mL of water/acetonitrile (20/80,
V/V) mixture was then added. The solution was stirred for 5
min and then completed to volume with water/acetonitrile
mixture.
HPLC Method
The assay analysis and amount of dissolved glimepiride was
estimated by reversed-phase HPLC (Waters Alliance) in a
binary mode with a photodiode array detector at 230 nm.
The analyses were performed on a C18 column (150 3
4.6 mm, 3.5 lm) placed in a column oven at 30 ꢀC and a
Preparation of Solid Dispersions
mobile phase:
A
(phosphate buffer, pH 5 2.5
:
A conventional solvent evaporation method was used for the
preparation of solid dispersions. The glimepiride substance
in crystalline form together with the carrier (dendrimers,
hyperbranched polymer, PVA) were weighted in a weight
ratio of 5/95 %, w/w, and dissolved in a mutual solvent—
ethanol at room temperature during continuous stirring with
magnetic stirrer (300 rpm) for 4 h to assure transparent sol-
utions. The solvent was then removed at slightly elevated
temperature (40 ꢀC) in vacuum. Such prepared solid disper-
sions were stored in a desiccated container until additional
study.
acetonitrile 5 72 : 28) and B (phosphate buffer, pH 5 2.5 :
acetonitrile 5 30 : 70) delivered at a flow-rate of 1.5 mL/min
under the following gradient conditions: 0–6 min (100% A–
0% A): 6–6.5 min (0% A–100% A). The column equilibration
time was 5 min. Retention time of the glimepiride was 4.0
min. The concentration of dissolved glimepiride was deter-
mined from the area of glimepiride peak using a preformed
calibration curve. Standard curve for glimepiride was meas-
ured over a range of 15 to 0.1 lg/mL and shown to be lin-
ear. The limit of detection was 0.005 lg/mL.
In Vitro Biocompatibility Testing—Hemolytic Activity
Freshly collected human blood samples supplemented with
EDTA were centrifuged at 2000 g for 10 min. The sediment
of red blood cells (RBC) was washed three times with 6 vol-
umes of phosphate buffered saline (PBS) and each time cen-
trifuged at 2000 g for 10 min. The buffy coat was removed
with each wash according to previous reports. The retrieved
Determination of Dendrimer Loading Capacity
Drug loading capacity (LC) of dendritic polymers was deter-
mined from the results of in vitro dissolution experiments
using reversed phase HPLC and calibration curve. The
amount of loaded glimepiride into carriers was determined
using eq 1.
3294
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 3292–3301

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
SCHEME 1 Synthesis of the AB2 adduct from bis-MPA and glycine and the second-generation dendrons: Reagents and conditions:
(a) 2,2-dimethoxypropane, acetone, 85%; (b) BnOH, DCC, DPTS, DCM, 87%; (c) TFA, DCM; (d) Method A: TEA, DCC, DPTS, DCM,
79%; Method B: TEA, EDC, DMAP, DCM, 99%; (e) Dowex, CH3CN, 99%; (f) H2, Pd/C, EtOAc, 99%; (g) Method A: DCC, DPTS, DCM,
90%; Method B: EDC, DMAP, DCM, 79%; (h) H2, Pd/C, EtOAc, 99%.
RBC was re-suspended in PBS at the dilution of 1:15. Dendri-
tic polymers were added to the 96-well plate, followed by
addition of RBC suspension resulting in final RBC dilutiꢀon
factor of 1 : 30. Then the plate was incubated for 3 h (37 C,
400 rpm). After the incubation period, the plate was centri-
fuged at 2000 g for 10 min and 10 lL of supernatant was
transferred into a 96-half well transparent plate, followed by
20-times dilution with PBS. Hemolysis degree was estimated
from the extent of hemoglobin release into the supernatant
as measured by absorbance at k 5 450 nm using Safire
reader (Tecan, Switzerland). The results are expressed as
percent hemolysis. The absorbance of the supernatant in the
absence of tested compounds was taken as zero hemolysis
(0%) and the total hemolysis (100%) was assigned when
sodium lauryl sulphate (0.1%, w/v) was added to the RBC
suspension.
drimers, it is essential that esterification as well as deprotec-
tion reactions proceed quantitatively and selectively. The
synthesis of the bis-protected AB2 repeating unit 6 alongside
the synthesis of the second-generation dendron 10 is
described in Scheme 1. The protected bis-MPA acid 2 and
the TFA-salt of 5 were reacted at room temperature in DCM
under basic conditions (TEA) according to two different pro-
cedures, either DCC/DPTS (Procedure A) or N-(3-dimethyla-
minopropyl)-N0-ethylcarbodiimide
hydrochloride (EDC)/
DMAP (Procedure B) to form the bis-protected AB2 repeating
unit 6. Both procedures provided the desired product 6,
however, the 6 prepared by procedure A necessarily required
purification by column chromatography, whereas the 6 pre-
pared by procedure B required no additional product purifi-
cation, only a simple extraction workup. Thus, the procedure
B is less time consuming and resulted in significantly higher
yield. Highly pure AB2 adduct is necessary for preparation of
well-defined dendron/dendrimer in the next step. The aceto-
nide protected group was easily removed under mild condi-
tions in the presence of acidic resin to provide the diol 7 in
quantitative yield. The benzyl protected group was also
quantitatively removed by catalytic hydrogenolysis utilizing
10% Pd/C to produce the acid 8.
RESULTS AND DISCUSSION
Synthesis of Poly(ester-amide) Dendrimers of Different
Generation
First the AB2 adduct of bis-MPA and glycine was prepared to
easily introduce an amide bond into the dendrimer structure
and to avoid a side reaction of an amine group with one of
the many dendron/dendrimer ester groups that could take
place in the case of direct amidation reaction. In order to
obtain a good overall yield and highly pure dendron/den-
Subsequently, the diol 7 and the acid 8 were coupled again
by two different coupling procedures to give the benzyl- and
acetonide-protected second-generation dendron 9 (Scheme 1).
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 3292–3301
3295

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 2 Schematic presentation of poly(ester-amide) dendrimers synthesis up to the fourth generation: reagents and condi-
tions: (a) Method A: DCC, DPTS, DCM/CH3CN, 50 ꢀC (protected dendrimers: 12 (1G), 81%; 14 (2G), 81%; 16 (3G), 69%); Method B:
R
V
EDC, DMAP, DCM/CH3CN, rt (protected dendrimers: 12 (1G), 98%; 14 (2G), 99%; 16 (3G), 83%); (b) Dowex , CH3CN (deprotected
dendrimers: 13 (1G), 15 (2G), 17 (3G), 19 (4G), 99%); (c) Method A: DCC, DPTS, DCM/CH3CN, 50 ꢀC (protected dendrimer: 18 (4G),
42%); Method B: EDC, DMAP, DCM/DMF, rt (protected dendrimer: 18 (4G), 45%).
Both coupling reactions were performed in DCM under
similar mild conditions and both provided comparable
results. Yet again, the procedure utilizing EDC coupling
reagent proved to be less time consuming due to a simple
extraction workup as the only purification method. After
removing of 1-ethyl-3-(3-dimethyl aminopropyl)urea dihy-
drochloride (EDU) and excess reagents, the obtained prod-
uct was analytically pure.
Since the preparation of the AB2 adduct of bis-MPA and gly-
cine, which is analogue of the bis-MPA monomer, requires
four synthetic steps (two for benzyl protected glycine,
3296
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 3292–3301

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 1 Comparison between the DCC/DPTS and the EDC/
congestion of hydroxyl functional groups resulting in their
decreased accessibility.
DMAP Reaction Conditions
Method A: DCCa
Method B: EDCb
The second esterification procedure utilized EDC/DMAP cou-
pling at room temperature in DCM/acetonitrile mixture for
the 1G–3G dendrimers and in DCM/DMF mixture for the 4G
dendrimer. For the 4G dendrimer, the DMF instead of aceto-
nitrile had to be used to ensure complete solubility of the
2G deprotected dendrimer at room temperature. This proce-
dure offered very mild conditions with significantly reduced
reaction time and showed considerably improved yields, that
is, 98, 99, 83, and 45% for the 1G, 2G, 3G, and 4G den-
drimers, respectively. For DCC/DPTS coupling method, the
amount of catalyst did not change significantly with den-
drimer generation, however, for EDC/DMAP coupling we had
to significantly enlarge the amount of DMAP with dendrimer
molar mass in order to achieve complete conversion. In
EDC/DMAP esterification procedure, the simple aqueous
workup method as the only purification procedure was
enough to obtain 1G and 2G dendrimers of good quality.
However, a detailed analysis of the raw 3G and 4G den-
drimers prepared by this procedure showed some side prod-
ucts that were not able to be removed solely by extraction
workup and thus a column chromatography had to be
applied. Besides, the 4G dendrimer showed encapsulated
Compound DPTS (Equiv) Yield (%)c DMAP (Equiv) Yield (%)c
6
0.05
0.2
79b
90b
81
0.25
1
99
79
99
99
83
45
9
12
14
16
18
0.2
3
0.2
81
3
0.4
69
12
24
0.44
42
a
Reactions were performed at 50 ꢀC.
b
c
Reactions were performed at room temperature.
Isolated yield.
followed by coupling with protected bis-MPA and subsequent
deprotection of benzyl group) we did not apply well estab-
lished, robust anhydride activated strategy (reported for
preparation of the bis-MPA dendrimers) for esterification
reaction to avoid the loss of starting material. Instead, the
syntheses of poly(ester-amide) dendrimers of the first (1G),
the second (2G), the third (3G), and the fourth (4G) genera-
tions were carried out by the divergent, convergent, mixed
divergent/convergent, and double stage convergent proto-
cols,59 respectively, using two different carbodiimide based
esterification reactions (Scheme 2). The first synthetic proce-
durꢀe (A) was performed in DCM/acetonitrile mixture at
50 C using DCC/DPTS coupling procedure, whereas the sec-
ond procedure (B) was performed in the same solvent mix-
ture but with EDC/DMAP coupling system at room
temperature. In the case of procedure A, the reaction had to
be carried out at 50 ꢀC in order to achieve completion in a
reasonable time. Additionally, the preparation of well-defined
3G and 4G dendrimers from the 1G and 2G dendrimers,
regardless of the type of synthetic procedure, required isola-
tion of the raw products with a quick workup in which the
resulting urea was removed (filtration in procedure A and
extraction in procedure B). The raw product was then re-
dissolved and coupling reaction repeated with a small
amount of acid 10 added.
The differences in yields and the amount of catalysts used
for each generation dendrimer for both esterification proce-
dures are summarized in Table 1. In both esterification pro-
cedures, the reaction time required for completion increases
with increasing dendrimer molar mass. Although reaction
preceded at a reasonable rate for lower generation den-
drimers, it slowed down for higher generation dendrimers.
By utilizing the DCC/DPTS esterification procedure, the pro-
tected 1G, 2G, 3G, and 4G poly(ester-amide) dendrimers
were obtained in 81, 81, 69, and 42% yields, respectively.
The purification of dendrimers by chromatographic techni-
ques was required for all generations, and it got harder at
higher dendrimer generation due to increase in dendrimer
polarity. The reaction yield of the fourth generation den-
drimer was significantly lower most probably due to steric
FIGURE 1 1H NMR spectra of deprotected 1G and 4G den-
drimers with the assignation of signals.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 3292–3301
3297

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
drimer interior might be the reason for higher amount of
catalyst needed for efficient reaction to higher generation
dendrimers.
The structure of synthesized poly(ester-amide) dendrimers
was investigated by 1H and 13C NMR and MALDI-TOF MS.
Integration of the assigned signals in proton NMR spectra
confirms high purity of dendrimers (Figs. 1 and 2, Support-
ing Information Figs. S1–S4). MALDI-TOF mass spectra of
protected and deprotected 1G and 2G (Fig. 3, Supporting
Information Fig. S5) confirm high structural uniformity of
synthesized dendrimers. On the other hand, mass spectra of
the 3G and 4G reveal some defect structures in trace amount
for the protected 3G and somewhat higher amount for the
protected 4G (Fig. 3). The content of defect structures is
larger in deprotected 3G and 4G (Supporting Information
Fig. S5). Since contrary the results of NMR and SEC-MALS
measurements indicate highly pure dendrimers, it is possible
that some degradation due to hydrolysis occurred during
sample preparation for mass spectroscopic measurements
rather than during hydroxyl group deprotection.
The molar mass characteristics of dendrimers were deter-
mined by SEC-MALS (Fig. 4, Table 2). SEC traces (responses
of RI and LS detectors) together with flat curves, represent-
ing the molar mass as a function of elution volume con-
firmed uniform molar mass distributions of deprotected
dendrimers, which are reflected in their low dispersity val-
-
ues (DM), (Table 2). The determined molar mass values of
dendrimers are very close to theoretical ones (Table 2).
FIGURE 2 13C NMR spectra of deprotected 1G and 4G den-
drimers with the assignation of signals.
Solid Dispersions of Glimepiride and Particular Polymer
Glimepiride is one of the third generation sulfonylurea drugs
used for control of diabetis mellitus, type 2.60 It shows poor
water-solubility and slow dissolution rate, which can cause
irreproducible clinical response or therapeutic failure due to
DMAP, which was used in a large amount, in the dendrimer
interior, and thus an additional acidic workup step was
required to remove it. Encapsulation of DMAP in the den-
FIGURE 3 MALDI-TOF MS spectra of protected 1G–4G dendrimers. The calculated exact mass of 1G sodium adduct is 796.38 Da,
and of 2G sodium adduct it is 1954.89 Da. The calculated molecular weight apex of 3G sodium adduct is 4273.99 Da, and of 4G
sodium adduct it is 8911.19 Da.
3298
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 3292–3301

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 3 Solubility and Loading Capacity (LC) of Glimepiride in
Synthesized Poly(ester-amide) Dendrimers Versus Poly(ester-
amide) hyperbranched polymer,56 Bis-MPA Dendrimers, and
PVA
c(HPLC)
w/w after 60 min c(theor.)
(%) (mg/mL) (mg/mL)
LC
Sample
(%)
Crystalline Glimepiride
Amorphous Glimepiride
Hybrane S1200
/
0.2
/
/
/
0.5
4.0
1.9
2.4
4.1
2.7
4.2
0.7
/
/
5
5
5
5
5
5
5
4.4
4.5
4.8
4.6
4.8
4.7
4.4
4.6
2.2
2.4
4.5
2.8
4.5
0.8
FIGURE 4 SEC-MALS chromatograms (solid curves 5 RI detec-
tor responses, dotted curves 5 LS detector responses at angle
90ꢀ) together with molar mass as a function of elution volume
(spotted curves) for deprotected 1G–4G dendrimers. [Color fig-
ure can be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
Dendrimer 2G
Dendrimer 3G
Dendrimer 4G
Bis-MPA dendrimer 3G
Bis-MPA dendrimer 4G
PVA
sub therapeutic plasma drug levels. In vitro dissolution stud-
ies of solid dispersions of glimepiride and poly(ester-amide)
dendrimers of different generations showed improved glime-
piride solubility in aqueous media as compared with pure
glimepiride in crystalline and amorphous form (Table 3).
The glimepiride solubility increased with increasing den-
drimer generation and reached the highest value when gli-
mepiride had been incorporated in the solid dispersion with
4G poly(ester-amide) dendrimer (Table 3). After 60 min, this
solid dispersion showed comparable glimepiride solubility as
the glimepiride incorporated in solid dispersion with hyper-
branched poly(ester-amide) polymer bearing hydroxyl func-
tional groups (Table 3).56
NH group of the glimepiride sulfonylurea segment as car-
bonyls of the ester and the amide groups in the structure of
poly(ester-amide) dendritic polymers. On the contrary, the
PVA proved to be ineffective. These results thus confirm the
findings of our previous studies, which revealed that
hydroxyl functional groups are not involved in complex for-
mation between glimepiride and poly(ester-amide) hyper-
branched polymer.56,58
From the results of in vitro dissolution measurements, we
estimated the amount of glimepiride complexed with poly-
mers in solid dispersions, that is, loading capacity (LC). LC
increased with dendrimer generation in both dendrimer
types and reached maximum values (4.5%, w/w) for the 4G
dendrimers which is comparable to LC of poly(ester-amide)
hyperbranched polymer (Table 3). On the other hand, the LC
of PVA was significantly lower as expected (Table 3).
In order to exactly evaluate the role of ester and amide
bonds as well as hydroxyl functional groups in synthesized
poly(ester-amide) dendrimers the in vitro dissolution studies
were performed also for the solid dispersions of glimepiride
and commercially available bis-MPA dendrimers (3G and 4G)
containing ester bonds and hydroxyl functional groups as
well as the solid dispersion of glimepiride and linear poly(vi-
nyl alcohol) (PVA) bearing only hydroxyl groups in the struc-
ture (Table 3). Surprisingly, the 4G bis-MPA dendrimer was
equally effective in improving the glimepiride solubility as
the 4G poly(ester-amide) dendrimer or poly(ester-amide)
hyperbranched polymer. These results reveal that carbonyl
groups of the ester bonds of bis-MPA dendrimer are as
effective in H-bond formation with slightly acidic proton of
-
TABLE 2 Molar Mass Characteristics (Mn, Mw, DM) of 1G–4G
Deprotected Dendrimers as Determined by SEC-MALS
Measurements
Deprotected
Dendrimers
Mtheor
Mn
Mw
-
(g/mol)
(g/mol)
(g/mol)
DM
FIGURE 5 The degree of hemolysis for G2–G4 poly(ester-
amide) dendrimers and Hybrane S1200 hyperbranched poly-
mer at polymer concentrations between 0.02 and 20 mg/mL.
[Color figure can be viewed in the online issue, which is avail-
able at wileyonlinelibrary.com.]
1G
2G
3G
4G
653.7
710
770
1.08
1.02
1.01
1.02
1692.7
3770.7
7926.7
1750
3830
7900
1780
3860
8060
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 3292–3301
3299

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Glimepiride has molar mass of 491 g/mol and rather rigid
structure and as such it is very unlikely to be encapsulated in
the interior of 4G bis-MPA dendrimer containing small cav-
ities. Since both dendrimer types reveal similar LC we infer
that glimepiride is most probably not encapsulated neither
into the interior of the 4G poly(ester-amide) dendrimer with
larger cavities. Instead, glimepiride is more likely simply
adsorbed on the dendrimers’ surface, where it interacts with
carbonyls of ester/amide bonds. Since the probability of
ester/amide bonds to be exposed on the dendrimer’s surface
increases with dendrimer generation number as a conse-
quence of greater probability for back-folding, the loading
capacity of both dendrimer types increases in the same order.
The synthesized poly(ester-amide) dendrimers as well as
commercially available bis-MPA dendrimers, poly(ester-
amide) hyperbranched polymer, and poly(vinyl alcohol) were
used for preparation of solid dispersions of sulfonylurea
antidiabetic drug glimepiride to improve its poor water-
solubility. In vitro dissolution studies show in comparison
with pure glimepiride in crystalline or amorphous form, sig-
nificantly improved glimepiride solubility for all solid disper-
sions based on dendritic polymers, but not for poly(vinyl
alcohol). These results indicate that the ester groups in den-
drimer structure are equally effective as the amide groups in
the formation of glimepiride/polymer complex, however, the
hydroxyl groups do not play a role in complex formation.
The amount of glimepiride complexed with both dendrimer
types increases with dendrimer generation and reaches the
maximum value for the solid dispersions of glimepiride with
the fourth generation dendrimers, that is, 4.5%, w/w. From
the red blood cell lysis test, the occurrence of mediated cyto-
toxicity of synthesized poly(ester-amide) dendrimers is esti-
mated to be very unlikely.
Hemolytic Activity of Poly(ester-amide) Dendrimers
The red blood cell (RBC) lysis is a cytotoxicity assay that is based
on hemoglobin release from RBC due to membrane rupture
caused by investigated compounds. Although the mechanistic
perspective of cell lysis is not investigated by such an approach, it
gives a clear indication whether the investigated compounds pos-
sess a cytotoxic potential. The positive control solution of sodium
lauryl sulphate (0.01%) caused a complete RBC lysis.
ACKNOWLEDGMENTS
Hemolytic activity was tested at different concentration of
poly(ester-amide) dendrimers in the range from 0.02 to
20 mg/mL. Since the use of synthesized 4G dendrimer is
suggested for oral pharmaceutical products in the range of
100 mg/formulation/250 mL, the expected gastrointestinal
luminal concentration is 0.4 mg/mL. At this concentration
none of the poly(ester-amide) dendritic polymers exhibit
hemolytic effect (Fig. 5). The 3G and the 4G poly(ester-
amide) dendrimers caused marginal RBC lysis only at the
highest tested concentration (20 mg/mL), which is 50-times
higher from the expected gastrointestinal luminal concentra-
tion. Therefore, the occurrence of mediated cytotoxicity of
these dendrimers is estimated to be very unlikely.
The authors gratefully acknowledge the financial support of
the Ministry of Higher Education, Science and Technology of
the Republic of Slovenia through the Slovenian Research
Agency (Program P2-0145 and Project L2-4166), the EN-FIST
Centre of Excellence, the Centre of Excellence—Polymer Mate-
rials and Technologies for access to MALDI-TOF MS, and
Dr. Luka Peternel for biocompatibility testing.
REFERENCES AND NOTES
1 K. Inoue, Prog. Polym. Sci. 2000, 25, 453–571.
2 A. Carlmark, C. Hawker, A. Hult, M. Malkoch, Chem. Soc.
Rev. 2009, 38, 352–362.
CONCLUSIONS
3 S. Svenson, D. A. Tomalia, Adv. Drug Delivery Rev. 2012, 64,
102–115.
We synthesized poly(ester-amide) dendrimers from the AB2
adduct of 2,2-bis(hydroxymethyl) propanoic acid (bis-MPA)
and glycine using two different esterification procedures,
either the DCC/DPTS or the EDC/DMAP coupling reagent/
catalyst systems. The synthetic procedure using DCC/DPTS
coupling was performed in DCM/acetonitrile at 50 ꢀC,
whereas the second one using EDC/DMAP was performed at
room temperature in DCM/acetonitrile mixture for the 1G to
3G dendrimers and in DCM/DMF mixture for the 4G den-
drimer. A comparison of both synthetic procedures reveals
that EDC/DMAP coupling offers reduced reaction time and
improved reaction yields, however, the amount of required
catalyst is larger, especially for the synthesis of higher gener-
ation dendrimers. In EDC/DMAP esterification procedure, the
simple aqueous workup method as the only purification pro-
cedure is enough to obtain the 1G and 2G dendrimers of
good quality, however, for isolation of highly pure 3G and 4G
dendrimers purification by column chromatography is neces-
sary. The DCC/DPTS procedure requires purification by col-
umn chromatography for all generation dendrimers.
ꢁ
4 C. L. Lee, J. A. MacKay, J. M. J. Frechet, F. C. Szoka, Nat. Bio-
technol. 2005, 23, 1517–1526.
5 S. Svenson, Eur. J. Pharm. Biopharm. 2009, 71, 445–462.
6 Y. Wang, S. M. Grayson, Adv. Drug Delivery Rev. 2012, 64,
852–865.
7 U. Gupta, H. B. Agashe, A. Asthana, N. K. Jain, Biomacromo-
lecules 2006, 7, 649–658.
8 C. M. Paleos, D. Tsiourvas, Z. Sideratou, Mol. Pharm. 2006, 4,
169–188.
9 N. K. Jain, U. Gupta, Expert Opin. Drug Met. 2008, 4, 1035–
1052.
ꢁ
10 S. M. Grayson, J. M. J. Frechet, Chem. Rev. 2001, 101,
3819–3867.
11 P. Kesharwani, K. Jain, N. K. Jain, Prog. Polym. Sci. 2014,
39, 268–307.
ꢁ
12 E. R. Gillies, J. M. J. Frechet, Drug Discov. Today 2005, 10,
35–43.
13 R. Duncan, L. Izzo, Adv. Drug Delivery Rev. 2005, 57, 2215–2237.
14 K. Jain, P. Kesharwani, U. Gupta, N. K. Jain, Int. J. Pharm.
2010, 394, 122–142.
3300
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 3292–3301

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
15 S. H. Medina, M. E. H. El-Sayed, Chem. Rev. 2009, 109,
3141–3157.
38 M. C. Parrott, S. R. Benhabbour, C. Saab, J. A. Lemon, S.
Parker, J. F. Valliant, A. Adronov, J. Am. Chem. Soc. 2009, 131,
2906–2916.
16 M. A. Mintzer, M. W. Grinstaff, Chem. Soc. Rev. 2011, 40,
ꢁ
€
173–190.
39 Y. Hed, Y. Zhang, O. C. J. Adren, X. Zeng, A. M. Nystrom,
M. Malkoch, J. Polym. Sci. Part A: Polym. Chem. 2013, 51,
3992–3996.
17 M. J. Cloninger, Curr. Opin. Chem. Biol. 2002, 6, 742–748.
18 M. W. Grinstaff, Chem. Eur. J. 2002, 8, 2838–2846.
€
40 H. Ihre, A. Hult, E. Soderlind, J. Am. Chem. Soc. 1996, 118,
19 M. A. Carnahan, M. W. Grinstaff, Macromolecules 2001, 34,
7648–7655.
6388–6395.
ꢁ
ꢁ
41 H. Ihre, O. L. Padilla De Jesus, J. M. J. Frechet, J. Am.
20 M. A. Carnahan, M. W. Grinstaff, J. Am. Chem. Soc. 2001,
123, 5905–5906.
Chem. Soc. 2001, 123, 5908–5917.
€
€
42 M. Malkoch, H. Claesson, P. Lowenhielm, E. Malmstrom, A.
21 M. A. Carnahan, M. W. Grinstaff, Macromolecules 2006, 39,
609–616.
Hult, J. Polym. Sci. Part A: Polym. Chem. 2004, 42, 1758–1767.
ꢁ
43 H. Ihre, A. Hult, J. M. J. Frechet, I. Gitsov, Macromolecules
22 M. Lo Conte, M. J. Robb, Y. Hed, A. Marra, M. Malkoch, C.
J. Hawker, A. Dondoni, J. Polym. Sci. Part A: Polym. Chem.
2011, 49, 4468–4475.
1998, 31, 4061–4068.
€
44 M. Malkoch, E. Malmstrom, A. Hult, Macromolecules 2002,
35, 8307–8314.
~
23 A. C. Fonseca, M. H. Gil, P. N. Simoes, Prog. Polym. Sci.
2013, 39, 1291–1311.
ꢁ
45 D. Soto-Castro, J. A. Cruz-Morales, M. T. Ramırez Apan, P.
Guadarrama, Molecules 2010, 15, 8082–8097.
24 D. G. van der Poll, H. M. Kieler-Ferguson, W. C. Floyd, S. J.
46 J. S. Moore, S. I. Stupp, Macromolecules 1990, 23, 65–70.
ꢁ
Guillaudeu, K. Jerger, F. C. Szoka, J. M. J. Frechet, Bioconju-
gate Chem. 2010, 21, 764–773.
47 S. -H. Yim, J. Huh, C. -H. Ahn, T. G. Park, Macromolecules
2007, 40, 205–210.
25 W. C. Floyd, P. J. Klemm, D. E. Smiles, A. C. Kohlgruber, V.
ꢁ
€
C. Pierre, J. L. Mynar, J. M. J. Frechet, K. N. Raymond, J. Am.
48 P. Antoni, D. Nystrom, C. J. Hawker, A. Hult, M. Malkoch,
Chem. Soc. 2011, 133, 2390–2393.
Chem. Commun. 2007, 69, 2249–2251.
~
26 N. Feliu, M. V. Walter, M. I. Montanez, A. Kunzmann, A.
49 M. I. Montanez, L. M. Campos, P. Antoni, Y. Hed, M. V.
Walter, B. T. Krull, A. Khan, A. Hult, C. J. Hawker, M. Malkoch,
Macromolecules 2010, 43, 6004–6013.
€
Hult, A. Nystrom, M. Malkoch, B. Fadeel, Biomaterials 2012, 33,
1970–1981.
ꢁ
27 O. L. P. De Jesus, H. R. Ihre, L. Gagne, J. M. J. Frechet, F. C.
Szoka, Bioconjugate Chem. 2002, 13, 453–461.
50 X. Ma, Z. Zhou, E. Jin, Q. Sun, B. Zhang, J. Tang, Y. Shen,
Macromolecules 2013, 46, 37–42.
ꢁ
28 E. R. Gillies, J. M. J. Frechet, J. Am. Chem. Soc. 2002, 124,
14137–14146.
51 P. Antoni, M. J. Robb, L. Campos, M. Montanez, A. Hult, E.
Malmstrom, M. Malkoch, C. J. Hawker, Macromolecules 2010,
43, 6625–6631.
ꢁ
29 E. R. Gillies, J. M. J. Frechet, Bioconjugate Chem. 2005, 16,
361–368.
52 A. Vieyres, T. Lam, R. Gillet, G. Franc, A. Castonguay, A.
Kakkar, Chem. Commun. 2010, 46, 1875–1877.
ꢁ
30 A. P. Goodwin, S. S. Lam, J. M. J. Frechet, J. Am. Chem.
€
Soc. 2007, 129, 6994–6995.
53 A. Carlmark, E. Malmstrom, M. Malkoch, Chem. Soc. Rev.
2013, 42, 5858–5879.
31 S. J. Guillaudeu, M. E. Fox, Y. M. Haidar, E. E. Dy, F. C.
ꢁ
Szoka, J. M. J. Frechet, Bioconjugate Chem. 2008, 19, 461–469.
54 M. V. Walter, M. Malkoch, Chem. Soc. Rev. 2012, 41, 4593–4609.
ꢁ
32 H. R. Ihre, O. L. Padilla De Jesus, F. C. Szoka, J. M. J.
ꢁ
55 K. L. Killops, L. M. Campos, C. J. Hawker, J. Am. Chem.
Soc. 2008, 130, 5062–5064.
Frechet, Bioconjugate Chem. 2002, 13, 443–452.
ꢀ
~
33 P. Lundberg, M. V. Walter, M. I. Montanez, D. Hult, A. Hult,
56 S. Reven, J. Grdadolnik, J. Kristl, E. Zagar, Int. J. Pharm.
€
A. Nystrom, M. Malkoch, Polym. Chem. 2011, 2, 394–402.
2010, 396, 119–126.
ꢀ
34 P. A. Ledin, F. Friscourt, J. Guo, G. -J. Boons, Chemistry
2011, 17, 839–846.
57 S. Reven, M. Homar, L. Peternel, J. Kristl, E. Zagar, Pharm.
Dev. Technol. 2013, 18, 323–332.
ꢁ
ꢀ
35 M. E. Fox, F. C. Szoka, J. M. J. Frechet, Acc. Chem. Res.
2009, 42, 1141–1151.
58 D. Pahovnik, S. Reven, J. Grdadolnik, R. Borstnar, J. Mavri,
E. Zagar, J. Pharm. Sci. 2011, 100, 4700–4709.
ꢀ
ꢁ
36 C. C. Lee, E. R. Gillies, M. E. Fox, S. J. Guillaudeu, J. M. J.
59 K. L. Wooley, C. J. Hawker, J. M. J. Frechet, J. Am. Chem.
ꢁ
Frechet, E. E. Dy, F. C. Szoka, Proc Natl Acad Sci USA 2006,
Soc. 1991, 113, 4252–4261.
103, 16649–16654.
60 I. Kouichi, W. Masaki, N. Youhei, S. Takahiro, T. Nobuki, T.
Masahiro, K. Hideyuki, Y. Kensuke, S. Masao, K. Susumu, A.
Takuya, K. Shigehiro, Diabetes Res. Clin. Pract. 2005, 68, 250–257.
37 M. C. Parrott, E. B. Marchington, J. F. Valliant, A. Adronov,
J. Am. Chem. Soc. 2005, 127, 12081–12089.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 3292–3301
3301