Elaboration of densely functionalized polylactide nanoparticles from N-acryloxysuccinimide-based block copolymers

Article abstract of DOI:10.1002/pola.24553

Poly(N-acryloxysuccinimide) (PNAS) and poly(N-acryloxysuccinimide-co-N- vinylpyrrolidone) (P(NAS-co-NVP)) of adjustable molecular weights and narrow polydispersities were prepared by nitroxide-mediated polymerization (NMP) in N,N-dimethylformamide in the presence of free SG1 (N-tert-butyl-N-1- diethylphosphono-(2,2-dimethylpropyl) nitroxide), with MAMA-SG1 (N-(2-methylpropyl)-N-(1-diethylphosphono-2,2-dimethylpropyl) -O-(2-carboxylprop-2-yl)hydroxylamine) alkoxyamine as initiator. The reactivity ratios of NAS and NVP were determined to be r_NAS = 0.12 and r _NVP = 0, indicating a strong alternating tendency for the P(NAS-co-NVP) copolymer. NAS/NVP copolymerization was then performed from a SG1-functionalized poly(D,L-lactide) (PLA-SG1) macro-alkoxyamine as initiator, leading to the corresponding PLA-b-P(NAS-co-NVP) block copolymer, with similar NAS and NVP reactivity ratios as mentioned above. The copolymer was used as a surface modifier for the PLA diafiltration and nanoprecipitation processes to achieve nanoparticles in the range of 450 and 150 nm, respectively. The presence of the functional/hydrophilic P(NAS-co-NVP) block, and particularly the N-succinimidyl (NS) ester moieties at the particle surface, was evidenced by ethanolamine derivatization and zeta potential measurements.

Full text of DOI:10.1002/pola.24553

Elaboration of Densely Functionalized Polylactide
Nanoparticles from N-Acryloxysuccinimide-Based Block Copolymers
1
NADEGE HANDKE, THOMAS TRIMAILLE,1 ELSA LUCIANI,2 MARION ROLLET,1
`
´
THIERRY DELAIR,3 BERNARD VERRIER,2 DENIS BERTIN,1 DIDIER GIGMES1
1Laboratoire Chimie Provence (LCP), UMR 6264, Universite´s d’Aix-Marseille I, II et III - CNRS, Equipe Chimie Radicalaire
Organique et Polyme`res de Spe´cialite´, Case 542, Av. Escadrille Normandie-Niemen, 13397 Marseille Cedex 20, France
2Institut de Biologie et de Chimie des Prote´ines (FRE 3310), Universite´ Lyon 1 - CNRS, 7 Passage du Vercors,
69367 Lyon, France
3Laboratoire des Mate´riaux Polyme`res et Biomate´riaux (UMR 5223), Universite´ Lyon 1 - CNRS, 15 Boulevard Latarjet,
69622 Villeurbanne, France
Received 12 November 2010; accepted 20 December 2010
DOI: 10.1002/pola.24553
Published online 18 January 2011 in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: Poly(N-acryloxysuccinimide) (PNAS) and poly(N-
acryloxysuccinimide-co-N-vinylpyrrolidone) (P(NAS-co-NVP)) of
adjustable molecular weights and narrow polydispersities were
prepared by nitroxide-mediated polymerization (NMP) in N,N-
dimethylformamide in the presence of free SG1 (N-tert-butyl-
N-1-diethylphosphono-(2,2-dimethylpropyl) nitroxide), with
MAMA-SG1 (N-(2-methylpropyl)-N-(1-diethylphosphono-2,2-di-
block copolymer, with similar NAS and NVP reactivity ratios as
mentioned above. The copolymer was used as a surface modi-
fier for the PLA diafiltration and nanoprecipitation processes to
achieve nanoparticles in the range of 450 and 150 nm, respec-
tively. The presence of the functional/hydrophilic P(NAS-co-
NVP) block, and particularly the N-succinimidyl (NS) ester moi-
eties at the particle surface, was evidenced by ethanolamine
C
V
methylpropyl)-O-(2-carboxylprop-2-yl)hydroxylamine)
alkoxy-
derivatization and zeta potential measurements.
2011 Wiley
amine as initiator. The reactivity ratios of NAS and NVP were
determined to be rNAS ¼ 0.12 and rNVP ¼ 0, indicating a strong
alternating tendency for the P(NAS-co-NVP) copolymer. NAS/
NVP copolymerization was then performed from a SG1-func-
tionalized poly(D,L-lactide) (PLA-SG1) macro-alkoxyamine as ini-
tiator, leading to the corresponding PLA-b-P(NAS-co-NVP)
Periodicals, Inc. J Polym Sci Part A: Polym Chem 49: 1341–
1350, 2011
KEYWORDS: block copolymers; functionalized poly(D,L-lactide)
nanoparticles; nitroxide-mediated polymerization; N-acryloxy-
succinimide; N-vinylpyrrolidone
INTRODUCTION Because of their properties of biocompatibil-
ity and biodegradability, polylactide (PLA)-based nanopar-
ticles (NP) have a long history in the field of drug delivery1,2
and are still attractive candidates.3 The now classically used
PLA-based NP carriers are surface coated with hydrophilic
‘‘stealth’’ polymers such as poly(ethylene oxide) (PEO) that
are efficient for reducing NP uptake by the reticular endothe-
lial system (RES) and thus prolonging their circulation in the
blood stream.4 By far, this was mainly achieved by the use of
PEO-b-PLA or PEO-b-poly(propylene oxide) (PPO)-b-PEO
(Pluronics) amphiphilic block copolymers as surface modi-
fiers during the fabrication process of the nanoparticles. One
major remaining challenge for drug carrier engineering is
the possibility to further couple cell specific ligands on this
hydrophilic corona to envision ‘‘active’’ targeting drug deliv-
ery. However, PEO, as a polyether, can hardly be functional-
ized along its backbone for further ligand coupling and up to
now, the most used approach has been to exploit the PEO
end group of the PLA-PEO block copolymer to introduce the
ligand moieties through various functionalization chemis-
tries.5,6,7,8 As a result, the surface density of ligands at the
micelle or particle surface is somehow limited (one ligand
per chain). In this context, we aimed at developing block
copolymers containing a PLA block and a block bringing
both hydrophilicity and functional group along the backbone
for biomolecule coupling, namely a poly(N-vinylpyrrolidone-
co-N-acryloxysuccinimide) (P(NVP-co-NAS)) block. The NAS
units present reactive N-succinimidyl (NS)-activated esters
that can indeed efficiently react with the amino-bearing mol-
ecules,9 whereas NVP ones have to confer hydrophilic/steal-
thy properties. NVP is gaining increasing attention, because
of the biocompatibility and low toxicity of its corresponding
polymer, which makes it a viable alternative to poly(ethylene
gycol) regarding stealth properties.10
Our envisioned PLA-based block copolymer requires to com-
bine both ring-opening polymerization (ROP) and controlled
radical polymerization (CRP) techniques. It is to point out
Additional Supporting Information may be found in the online version of this article. Correspondence to: T. Trimaille (E-mail: thomas.trimaille@
univ-provence.fr)
C
V
Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 49, 1341–1350 (2011) 2011 Wiley Periodicals, Inc.
´
DENSELY FUNCTIONALIZED POLYLACTIDE NANOPARTICLES, HANDKE ET AL.
1341

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
that such combination has strongly emerged over the last
years to achieve well-defined block copolymers made of ali-
phatic polyester and vinyl polymer segments, as reviewed by
Dove,11 either through ROP/atom transfer radical polymer-
ization (ATRP),12 ROP/reversible addition fragmentation
chain transfer (RAFT),13 or ROP/nitroxide-mediated poly-
merization (NMP)14 combined techniques. We particularly
reported in a previous article the convenient access to PLA-
b-poly(hydroxyethyl (meth)acrylate) and PLA-b-polystyrene
block copolymers through the ROP/NMP combination.15 Our
strategy is based on the use of a SG1 nitroxide functionalized
PLA as macro-alkoxyamine to initiate NMP of the vinyl
monomer(s) of interest. Regarding the NAS/NVP vinyl mono-
mer system envisioned here for the ‘‘radical’’ made block,
CRP techniques have not been reported to our knowledge.
Indeed, whereas copolymerization of NAS with other como-
nomers has been extensively reported via the CRP techni-
ques, principally by RAFT (N-isopropylacrylamide (NIPAM),16
NIPAM and fluorescein O-methacrylate,17 N,N-dimethylacryla-
mide (DMA),18,19,20 hydroxyethylacrylate (HEA),21 or N-acryl-
oylmorpholine (NAM)22,23,24) and to a less extent by ATRP
(styrene25) and NMP (DMA),26 its copolymerization with
NVP has only been reported by conventional radical poly-
merization.27 We here successively describe the synthesis by
NMP of P(NAS-co-NVP), PLA-b-P(NAS-co-NVP) block copoly-
mers and their use as surface modifier in PLA nanoprecipita-
tion and diafiltration processes to achieve novel PLA nano-
particles with advanced functional/hydrophilic surface
properties.
(80:12 mL). The product was filtered and washed succes-
sively with 200 mL of hexane:ethyl acetate mixture (4:1),
then (9:1) and with 400 mL of pure hexane. The final prod-
uct was dried under vacuum to a constant weight (24.6 g,
74% yield).
1H NMR (300 MHz, CDCl3, d, ppm): 2.87 (s, 4H), 6.16–6.20
(dd, 1H), 6.29–6.38 (dd, 1H) 6.68–6.74 ppm (dd, 1H). Anal.
Calcd for C7H7NO4: C, 49.71; H, 4.17; N, 8.28. Found: C,
49.79; H, 4.23; N, 8.30.
Polymer Synthesis
The homopolymerization of NAS from the MAMA-SG1 alkoxy-
amine was carried out in DMF (25 wt % monomer content)
in the presence of free SG1 (0, 2.5, 5, or 7.5 mol % to initia-
tor). 1,3,5-Trioxane was used as internal 1H NMR reference
in the mixture to follow the conversion of the monomer. Typ-
ically, NAS (1 g, 5.92 mmol) and MAMA-SG1 alkoxyamine
(38.5 mg, 0.1 mmol, for a targeted Mn of 10,000 g molꢁ1
)
were introduced in a 25-mL two-neck round-bottomed flask
fitted with a reflux condenser and an argon inlet and deoxy-
genated for 20 min by argon bubbling. The mixture was then
heated to 110 ꢀC under argon with vigorous stirring. After
polymerization, the mixture was cooled down and precipi-
tated in diethyl ether. During the polymerization, samples
were withdrawn periodically for determining conversion and
1
molecular weight by H NMR and GPC, respectively. The final
polymer was purified by precipitation in diethyl ether and
dried under vacuum at room temperature to a constant
weight.
Copolymerization of NAS and NVP was performed under a
similar procedure from the MAMA-SG1 alkoxyamine, with an
overall monomer concentration of 1 mol Lꢁ1 in DMF. The
mixture was heated to 100 ꢀC. Different Mn values (10,000,
20,000, and 40,000 g molꢁ1) were targeted using 5 mol %
of free SG1 to initiator. The final polymer was purified by
precipitation in diethyl ether and dried under vacuum at
room temperature to a constant weight. NAS/NVP copoly-
merization from the PLA-SG1 macro-alkoxyamine was per-
formed under a similar procedure, except that polymeriza-
tion temperature was 120 ꢀC and precipitation was
performed in diethyl ether/methanol (1/1) mixture.
EXPERIMENTAL
Materials
N-Hydroxysuccinimide (NHS, 98%), acryloyl chloride (96%),
triethylamine (TEA, 99%), N-vinylpyrrolidone (NVP, 99%),
and 1,3,5-trioxane (99%) were purchased from Aldrich. The
MAMA-SG1 alkoxyamine (BlocBuilder MA, 99%) and the SG1
nitroxide (82%) were kindly provided by Arkema. Poly(D,L-
lactide) (PLA50 Mn ¼ 30,000 g/mol, PDI ¼ 1.7) with a car-
boxylic end group was purchased from Phusis (Grenoble,
France). PLA-SG1 macro-alkoxyamine was synthesized as
previously described.15 All NMR solvents were purchased
from Eurisotop. All the solvents were used as received.
Characterization
1H NMR (300 MHz) spectra were recorded on a Bruker
Avance DPX-300 at room temperature. For organic com-
pounds analysis were performed with 10 mg in 500 lL of
CDCl3. The polymerization kinetics was studied with 60 lL
of mixture in 500 lL of DMSO-d6. For determination of NAS
and NVP conversions, the peak integrals of the vinyl protons
(6.3–6.8 ppm and 6.9–7.05 ppm for NAS and NVP, respec-
tively) and 1,3,5-trioxane (5.13 ppm) were used. Elemental
analyses on NAS were performed on a Thermo Finnigan EA
1112 (Spectropole, Marseille).
N-Acryloxysuccinimide Synthesis
N-acryloxysuccinimide (NAS) was synthesized according to a
previously reported procedure.28 In brief, it consists in direct
reaction of NHS and acryloyl chloride in presence of TEA. A
solution of acryloyl chloride (18 mL, 0.22 mol) in chloroform
(30 mL) was added dropwise to a stirred solution of NHS
(22.65 g, 0.196 mol) and TEA (30 mL, 0.22 mol) in chloro-
ꢀ
ꢀ
form (300 mL) at 0 C. After stirring 3 h at 0 C, the reaction
mixture was washed with ice-cold saturated sodium bicar-
bonate solution three times and dried over MgSO4. Then, the
solution was filtered and chloroform was removed under
reduced pressure until a residual volume of 60 mL. The
The number-average molar molecular weight (Mn) and the
polydispersity index (PDI) were determined using gel perme-
ation chromatography (GPC). The homo- or co-polymer sam-
ples were prepared using 5 mg of the polymer in 2 mL of
ꢀ
product was precipitated overnight at ꢁ18 C after the drop-
wise addition of a mixture of hexane and ethyl acetate
1342
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
DMF and then filtered on nylon Millipore Millex GN syringe
filter. The GPC chain consisted of a Waters 717 plus auto-
sampler, a Waters 600 system controller and a Waters 600
fluid unit. A waters 2414 differential refractometer was used
hour during these 6 h and, subsequently, dialysis was carried
out overnight. The day after, water was renewed one last
time and after 1 h of dialysis, the nanoparticles were col-
lected from the dialysis tube. The final dispersions had a
solid content of typically 1 wt %. No residual DMSO was
detected from proton NMR spectra of dissolved particle sam-
ples. For nanoprecipitation process, 1.1 g of PLA were dis-
solved in 50 mL acetone and 0 or 0.11 g of block copolymer
in 5-mL acetonitrile. The two organic solutions were mixed
and added dropwise through an addition funnel to 35 mL of
an aqueous continuous phase under moderate magnetic stir-
ring. The mass transfer of acetone/acetonitrile into the con-
tinuous water phase resulted in the formation of PLA nano-
spheres. After completion, the solvents were removed by
evaporation under reduced pressure without heating. No
residual acetone was detected from proton NMR spectra of
dissolved particle samples, to the sensitivity limits of this
method. The final dispersions had a solid content of typically
2.5 wt %.
ꢀ
at 40 C as detector. DMF was used as eluent at a flow rate
of 0.7 mL minꢁ1 after filtration on Alltech nylon membranes
with a porosity of 0.2 lm. The column oven was kept at 70
ꢀC, and the injection volume was 20 lL. Two ResiPore col-
umns (600 mm ꢂ 7.5 mm), Polymer Laboratories, were used
in series. The system was calibrated using polystyrene (PS),
poly(methyl methacrylate) (PMMA) or polyethylene glycol
(PEG) narrow standards, in the range 1000–400,000, pur-
chased from Polymer Laboratories.
Reactivity Ratios
Several copolymerizations of NAS and NVP were carried out
by NMP targeting constant total degree of polymerization
(300). Different initial feed in NAS and NVP monomers were
used: 0.2/0.8, 0.3/0.7, 0.5/0.5, 0.7/0.3, and 0.8/0.2. The
reactivity ratios r1 and r2 were determined using a nonlinear
least square method, as previously reported,29,30,31 based on
fitting experimental values of conversion and theoretical
ones obtained from the integrated copolymerization eq 1
developed by Lowry and Meyer.32
Nanoparticle Characterization
Nanoparticle size was determined by dynamic light scatter-
ing (DLS) at 25 ꢀC, using a Zeta Sizer NanoZS (Malvern
instruments, UK). Highly diluted colloidal dispersions in 1-
mM NaCl solution were used, and each value is at least the
average of three measurements. Zeta potentials were meas-
ured with Zeta Sizer NanoZS using the same highly diluted
samples. The measurements of the electrophoretic mobili-
ties were carried out at 25 ꢀC, and the data were converted
to the Zeta potentials using Smoluchowski equation. The
values were the average of five measurements. The nano-
particles were observed by scanning electron microscopy
(SEM). Few droplets of dispersion (0.1% solid content)
were deposited on a Thermanox coverslip and allowed to
dry overnight. After mounting on stub and sputtering with
gold-palladium, the samples were observed using a FEI
Quanta 250 FEG ESEM microscope. UV spectroscopy was
used to quantify the available NS groups at the particle sur-
face through the absorbance at 260 nm of the released
NHS anion on derivatization with ethanolamine. Typically,
to a 2 mL of NP dispersion at 0.024% in NaCl 1 mM was
added 26 lL of an ethanolamine solution (6.1 mg mLꢁ1).
After 15 min, the size and zeta potential of the particles
were measured, and the particles were centrifuged at
11,000 rpm for 20 min (in some cases additional centrifu-
gation was necessary). Then, the UV spectrum of the super-
natant was recorded with Cary 50 Varian spectrometer and
the absorbance (OD) at 260 nm gave access to the NS
available groups, through the use of a calibration curve
OD ¼ f([NHS]) built up in the same conditions. The surface
density (NS groups/nm2) could be deduced using the spe-
cific surface of the particles. The amount of hydrolyzed NS
esters could be also evaluated by UV using the same proce-
dure, with the exception that ethanolamine was added in
the supernatants after centrifugation. Colloidal stability of
the dispersions was evaluated by measuring particle size by
DLS in solutions presenting increasing sodium chloride
concentrations.
ꢀ
ꢁ ꢀ
ꢁ ꢀ
ꢁ
a
b
c
½Mꢃ
f1
1 ꢁ f1
ðf1Þ0ꢁd
f1 ꢁ d
1 ꢁ
¼ 1 ꢁ
(1)
½Mꢃ0
ðf1Þ0
1 ꢁ ðf1Þ0
with
r2
r1
1 ꢁ r1r2
1 ꢁ r2
2 ꢁ r1 ꢁ r2
a ¼
b ¼
c ¼
d ¼
1 ꢁ r2
1 ꢁ r1
ð1 ꢁ r1Þð1 ꢁ r2Þ
where 1 ꢁ [M]/[M]0 is the conversion, f1 is the feed compo-
sition of NAS at that value of conversion and (f1)0 is the ini-
tial feed composition. The reactivity ratios r1 and r2 were
obtained for the minimum of the sum of squares of the
weighted residuals developed in eq 2:
n
X
2
1
2
ssðr1; r2Þ ¼
wiðyi ꢁ fiðr1; r2ÞÞ w ¼ =
(2)
yi
i
i¼1
where yi is the experimental value of conversion at the ith
experimental data point and fi(r1, r2) is the theoretical one.
This method has the advantage to be not restricted to low
conversion values.
Preparation of the Copolymer-Functionalized
PLA Nanoparticles
PLA nanoparticles with or without block copolymer as sur-
face modifier were prepared by the dialysis method (diafil-
tration), initially described by Jeon et al.33 or nanoprecipita-
tion (solvent diffusion), initially described by Fessi et al.34
For diafiltration: briefly, 400-mg PLA and 40 or 0 mg of PLA-
b-P(NAS-co-NVP) block copolymer were dissolved in 20 mL
of dimethylsulfoxide (DMSO). The solution was introduced
into a dialysis tube having a molecular weight cut-off of
6000–8000 g/mol and dialyzed against 4 L of milliQ water
for 6 h at room temperature. The water was changed every
´
DENSELY FUNCTIONALIZED POLYLACTIDE NANOPARTICLES, HANDKE ET AL.
1343

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
FIGURE 1 Chemical structures of poly(N-acryloxysuccinimide) (PNAS) (a) and poly(N-acryloxysuccinimide-co-N-vinylpyrrolidone)
(P(NAS-co-NVP)) (b) obtained by nitroxide mediated polymerization (NMP) of NAS and NAS/NVP from the MAMA-SG1
alkoxyamine.
RESULTS AND DISCUSSION
ular weights was observed along conversion, even if it was
less marked as theoretically expected, indicating a relative
control of the polymerization (Fig. 2, Supporting Information
NMP of NAS and NAS/NVP from the MAMA-SG1
Alkoxyamine
Fig. S1b for GPC traces). Poly(methyl methacrylate) (PMMA)
Before investigating the NMP of NAS/NVP to obtain our
appeared to be the most appropriate standard for GPC
desired hydrophilic/functional copolymer, we first focused
calibration. Finally, the polydispersity index values were
on the NMP of the NAS alone. Homopolymerization of this
quite low, supporting the controlled character of the
monomer by this technique has not been studied up to now,
polymerization.
and PNAS is a preactivated polymer of interest in the field of
conjugation with biomolecules. NMP was performed in DMF
at 110 ꢀC, using the MAMA-SG1 alkoxyamine targeting a Mn
of 10,000 g molꢁ1 [Fig. 1(a), for PNAS structure]. The effect
of free SG1 nitroxide on the polymerization process was
evaluated (0, 2.5, 5, and 7.5 mol % to initiator), because it
has been previously reported that for monomers presenting
high-propagation rate constant (which is the case for NAS),
free SG1 is usually needed to force the equilibrium between
the active and dormant species and thus to ensure better
control of the polymerization.35 The introduction of free SG1
in the polymerization medium yielded an improved linearity
for ln([M]0/[M]) versus time (Supporting Information Fig.
S1a), indeed suggesting a better polymerization control, even
for the lowest SG1 amounts introduced (2.5 mol % to
MAMA-SG1). The conditions retained for further studies
were 5 mol % SG1 to alkoxyamine.
We then focused on the synthesis by NMP of P(NAS-co-NVP)
copolymers [Fig. 1(b), for the copolymer structure], in which
the NVP units are suitable to improve hydrophilicity for drug
delivery applications. The NMP was performed using MAMA-
SG1 as initiator, in DMF at 100 ꢀC. Higher polymerization
temperatures were reported to induce NVP degradation.36
Different NAS molar feed for the same targeted total degree
of polymerization were investigated to determine the reactiv-
ity ratios. The most common procedures for calculating reac-
tivity ratios such as those proposed by Fineman and Ross37
and by Kelen and Tu¨dos38 can only be applied to experimen-
tal data at sufficiently low conversion (<5%), because calcu-
lation is based on the differential copolymerization equa-
tion.39,40 The extended Kelen–Tu¨dos method involves a
rather more complex calculation, and it can be applied to
medium conversion experimental data (ꢄ40ꢁ50%). With
regards to rather high conversion of our copolymerization,
In these selected conditions, whatever the standards used
for GPC calibration, an increase of the PNAS apparent molec-
we used
a non linear least-squares procedure to fit
FIGURE 2 Apparent Mn (plain symbols) and PDI (empty sym-
bols) versus conversion for nitroxide mediated polymerization
(NMP) of N-acryloxysuccinimide (NAS) in DMF at 110 ꢀC, using
different polymer standards: PS (squares), PMMA (triangles),
and PEG (rounds); thick line is the theoretical curve.
FIGURE 3 N-Acryloxysuccinimide (NAS) molar fraction in feed
(fNAS) vs global molar conversion for copolymerizations of N-
acryloxysuccinimide (NAS) and N-vinylpyrrolidone (NVP) with
5% of free SG1 to MAMA-SG1 initiator (DMF, 100 ꢀC).
1344
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
FIGURE 4 Copolymerization of N-acryloxysuccinimide (NAS) and N-vinylpyrrolidone (NVP) at 100ꢀC, initiated by the MAMA-SG1
alkoxyamine initiator, in presence of free SG1 (5% to initiator): (a) global conversion and ln([M]0/[M]) vs time; (b) Mn (plain sym-
bols) and PDI (empty symbols) vs conversion for different targeted Mn: 10,000 g molꢁ1 (triangles), 20,000 g molꢁ1 (rounds) and
40,000 g molꢁ1 (squares).
experimental and theoretical values of conversion obtained
with the integrated copolymerization equations developed by
Lowry and Meyer32 by minimizing weighted residuals, as
previously reported and described in the experimental sec-
tion. The reactivity ratios were determined to be rNAS ¼ 0.15
and rNVP ¼ 0 when no free SG1 was added in the polymer-
ization medium, and rNAS ¼ 0.12 and rNVP ¼ 0 when using 5
mol % SG1 to MAMA-SG1 alkoxyamine (Fig. 3), indicating a
very limited impact of the extra free nitroxide on the mono-
mer addition process during the polymerization. The
obtained reactivity ratio values were indicative of a strong
alternating tendency for the copolymer, similarly as previ-
ously reported in an earlier study on copolymerization of the
two monomers by conventional radical polymerization (rNAS
¼ 0.27 and rNVP ¼ 0.01).27 Interestingly, this induces a ho-
mogenous distribution and quite high density of the reactive
NS ester functions of the NAS for further biomolecule cou-
pling. Based on the obtained reactivity ratios (rNAS ¼ 0.12
and rNVP ¼ 0), the azeotropic composition was determined
to be at 53% of NAS.
of 5 mol % free SG1 to initiator to favor control. Despite
lower monomer concentrations and lower temperature used,
the copolymerization of NAS with NVP appeared to be much
faster than that observed for the NAS homopolymerization
[Fig. 4(a)]. The ln([M]0/[M]) expression relying to the global
conversion increased linearly with time for different targeted
molecular weights (10,000, 20,000, and 40,000 g molꢁ1). As
shown in Figure 4(b), the molecular weights tended to
increase with global conversion and were relatively close to
the expected ones (calculated from the relation Mn,th
¼
[NAS]0 ꢂ MNAS ꢂ convNAS þ [NVP]0 ꢂ MNVP ꢂ convNVP]/
[MAMA-SG1]0, with MNAS ¼ 169.1 g molꢁ1 and MNVP
¼
111.1 g molꢁ1). In addition, the PDI values tended to
decrease with conversion and did not exceed 1.4, indicating
a relative control of the polymerization.
NMP of NAS/NVP from the PLA-SG1 Macro-Alkoxyamine
We then focused on the synthesis of PLA-b-P(NAS-b-NVP)
block copolymers, to use them as surface modifiers in the
fabrication process of PLA nanoparticles. To obtain them, our
approach was based on the use of a SG1 functionalized PLA
macro-alkoxyamine as initiator for the NMP of NAS/NVP
(Scheme 1). This macro-alkoxyamine was obtained as previ-
Further studies on the NAS/NVP copolymerization process
were then performed using a starting molar fraction in NAS
of 0.6. Again, the polymerization was performed in presence
ously described in our group by performing
a
1,2-
SCHEME 1 Strategy of synthesis for PLA-
b-P(NAS-co-NVP) copolymer, using a PLA-
SG1 macro-alkoxyamine as initiator for
NMP of NAS/NVP, and further nanopar-
ticle preparation.
´
DENSELY FUNCTIONALIZED POLYLACTIDE NANOPARTICLES, HANDKE ET AL.
1345

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
TABLE 1 Characteristics of the Different PLA-b-P(NAS-co-NVP)
Block Copolymers Synthesized Through NMP of NAS/NVP
from PLA-SG1 Macro-Alkoxyamine (NAS Molar Composition
Feed of 0.6)
known molar mass of the PLA block. In Figure 6 are repre-
sented the 1H NMR spectra of the crude mixtures in the
course of NMP of NAS and NVP from the PLA-SG1. Besides
the progressive decreasing of the monomer proton signals
concomitant with increasing ones from the forming polymer,
it is to point out (in the range of 1–1.5 ppm) the shift of the
tBu protons of the SG1 moiety along the polymerization pro-
cess, as a result of its environment change (the SG1 initially
linked to the PLA a-end group becomes adjacent to the NAS/
NVP, Scheme 1). This supports a quite good initiation effi-
ciency of the NMP from PLA-SG1. The polydispersity index
obtained by SEC was quite acceptable (Table 1).
Mn PLA
Blocka
(g molꢁ1
Mn P(NAS-co-NVP)b
Copolymer
Code
Block
)
(g molꢁ1
)
PDI
BC1
BC2
BC3
a
19,000
19,000
10,500
17,000
12,000
11,000
1.64
1.34
1.62
Determined by GPC from universal calibration.15
Determined by 1H NMR integration.
b
Functionalized PLA Nanoparticles
The block copolymer was then used with PLA homopolymer
(Mn ¼ 30,000 g molꢁ1) in the nanoparticle formation pro-
cess, to achieve both hydrophilic and densely NS ester-func-
tionalized nanoparticles at their surface (Scheme 1, last
step). Nanoprecipitation and diafiltration processes were
used to reach size ranges of about 150 and 500 nm, respec-
tively. For nanoprecipitation, both polymers were first dis-
solved in acetone/acetonitrile (the copolymer was indeed in-
soluble in pure acetone, which is the solvent classically used
in the PLA nanoprecipitation process). The nanoparticles
formed instantaneously on addition of this polymer organic
phase into water, as a result of rapid diffusion of the ace-
tone/acetonitrile solvents in the water phase. The organic
solvents were then removed under reduced pressure. As for
diafiltration, PLA and the copolymer were dissolved in
DMSO, and the solution was dialyzed against water for sev-
eral hours, with periodic changes of water bath. The nano-
particles formed slowly as a result of progressive replace-
ment of DMSO by water in the dialysis bag. As both methods
are known to be surfactant free, reference-naked PLA par-
ticles were also prepared using the same procedure for both
intermolecular radical addition (IRA) of the MAMA-SG1
alkoxyamine onto acrylate end-capped PLA previously pre-
pared by ROP.15 The NMP of NAS/NVP from this macroinitia-
tor was performed in DMF at 120 ꢀC. This polymerization
temperature, higher than that used for the NMP previously
run from the MAMA-SG1, was chosen to favor the initiation
efficiency of NMP, because we showed in earlier studies that
the activation energy for dissociation of PLA-SG1 macro-
alkoxyamine is higher than that of MAMA-SG1.15 In addition,
this higher temperature was found not to be prejudicial
regarding NVP integrity due to the short polymerization
times. The found NAS and NVP reactivity ratios were 0.17
and 0.02, respectively, very close to those previously
obtained, as expected, confirming the alternating tendency
for the P(NAS-co-NVP) block. In Table 1 are compiled the dif-
ferent molar masses of block copolymers prepared (with a
molar feed in NAS of 0.6). The molecular weight of the
1
P(NAS-co-NVP) block was obtained from H NMR integration
(Fig. 5) on the spectrum of the purified block copolymer
after precipitation in diethyl ether/methanol, based on the
FIGURE 5 1H NMR spectrum (CDCl3) of the PLA-b-P(NAS-co-NVP) block copolymer after precipitation in 1/1 diethylether/methanol
mixture.
1346
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
FIGURE
6
1H NMR spectra of
crude polymerization mixtures
along NAS/NVP conversion for
the NMP of NAS/NVP from PLA-
SG1 macro-alkoxyamine (15, 30,
45, and 60 min polymerization
times, from top to bottom).
processes, but in the absence of block copolymer. The char-
acteristics of the obtained nanoparticles are presented in
Table 2. All the nanoparticles presented a rather homogene-
ous distribution in size, as shown by the polydispersity index
values given by DLS, close to 0.1 in most cases, as well as no
aggregation (Supporting Information Fig. S2 for the size dis-
tribution profiles). These results were confirmed by SEM
analysis of the nanoparticles (Fig. 7). Interestingly, naked
PLA particles [NP0, Fig. 7(b)] tended to more collapse than
functionalized nanoparticles [NP1, Fig. 7(a)] under exposi-
tion to the electron beam heating, strongly suggesting the
contribution of the NAS/NVP-based copolymer at the particle
surface to prevent PLA filmification.
by size measurements as a function of sodium chloride con-
centration. DIA0 flocculated at less than 0.5M salt concentra-
tion, whereas DIA1 remained stable until at least 1.5M. Simi-
lar results were obtained for the bigger NP1 and NP0
particles. Such higher colloidal stability for NP1 and DIA1
can clearly be attributed to the influence of the copolymer at
the interface, responsible for a steric contribution to the sta-
bilization of the particles, which corroborates the zeta poten-
tial measurements. To further evidence the presence and
availability of the activated esters at the particle surface for
further coupling of biomolecules, ethanolamine as a model
reactive compound was added in excess (compared with
The particles prepared in the presence of copolymer were
slightly bigger than the naked particles for the nanoprecipi-
tation process, whereas no significant differences were
observed between naked and copolymer-functionalized par-
ticles obtained by diafiltration (Table 2). Nevertheless, for
both processes, the zeta potential significantly changed, and
was much less negative for nanoparticles prepared in the
presence of copolymer than for the naked ones, indicating
the contribution of the hydrophilic/functional neutral copoly-
mer at the particle surface. The still negative zeta potential
for the copolymer-functionalized nanoparticles was ascribed
to the partial hydrolysis of the NS esters in water, as previ-
ously described18 and pointed out later.
TABLE 2 Characteristics of the Functionalized PLA Nanoparticles
(10 wt % Copolymer with Respect to Homo-PLA) and Naked
PLA Particles
Average
Diametera
(nm)
Zeta
NP
Potential
n (mV)
Code Copolymer Process
PIb
NP0
NP1
NP2
NP3
DIA0
–
Nanoprec.
Nanoprec.
Nanoprec.
Nanoprec.
126
156
180
204
0.099 ꢁ69
0.095 ꢁ60
0.160 ꢁ45
0.130 ꢁ45
0.098 ꢁ75
0.195 ꢁ40
0.086 ꢁ55
0.088 ꢁ35
BC1
BC2
BC3
–
Diafiltration 461
Diafiltration 435
Diafiltration 451
Diafiltration 452
DIA1 BC1
DIA2 BC2
DIA3 BC3
a
NP1 and DIA1 copolymer-functionalized nanoparticles were
compared with NP0 and DIA0 naked nanoparticles as refer-
ence, respectively, by means of colloidal stability in the pres-
ence of salt. A marked difference in colloidal stability was
observed between both kinds of nanoparticles, as evidenced
Average hydrodynamic particle size.
Polydispersity index given by DLS.
b
´
DENSELY FUNCTIONALIZED POLYLACTIDE NANOPARTICLES, HANDKE ET AL.
1347

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
FIGURE 7 Scanning electron microscopy images of the NP1 (a) and NP0 (b) nanoparticles.
activated NS ester functions) in both naked and copolymer-
functionalized PLA nanoparticle dispersions (in NaCl 0.001M).
After ethanolamine adding, the diameter of the reference na-
ked NP0 particles remained constant (127 nm) and the zeta
potential became even more negative (ꢁ80 mV), as a result of
the fully deprotonation of carboxylic acid moities at the parti-
cle surface (Table 3). In contrast, the diameter of the copoly-
mer-functionalized particles (NP1) significantly increased
(from ꢄ20 nm), and the zeta potential increased to ꢁ40 mV,
compared to ꢁ60 mV before addition. This was due to the
reaction of amine of ethanolamine with the NS-activated
esters of the surface of the particle, which renders the particle
corona highly hydrophilic and, as a result, favor its expansion
toward aqueous medium. Consequently, the zeta potential was
logically increased (reduced in absolute value) due to the
shielding by this newly formed neutral corona based on
N-vinylpyrrolidone and hydroxyethyl pendant groups of the
copolymer. The still negative value was explained by the car-
boxylates arising from the NAS prehydrolysis and competitive
hydrolysis by water, due to the drastically increased pH of the
continuous phase of the dispersion.
The reaction of ethanolamine with the activated esters was
finally confirmed by UV measurements. Indeed, the UV spec-
trum of the supernatant after ethanolamine adding to the
NP1 dispersion and centrifugation, clearly evidenced the
presence of the released N-hydroxysuccinimide (NHS) anion,
with its typical maximum absorption wavelength at 260 nm
(Supporting Information Fig. S3), as a result of ethanolamine
nucleophilic attack on the activated NS ester. The naked PLA
particles after ethanolamine adding did not exhibit any UV
absorption, as expected (Supporting Information Fig. S3).
Based on a calibration curve built up in the same conditions
(OD as a function of NHS concentration in ethanolamine ba-
sic conditions), the surface density of the NS ester groups
available at the particle surface for coupling could be deter-
mined to be about 4 lmol mꢁ2, that is, 2.4 groups nmꢁ2
,
which was qꢀuite high. The particles remained stable under
storage at 4 C in terms of size. Indeed, after 3 months, the
TABLE 3 Effect of Ethanolamine Addition in the NP Dispersion on the Size and Zeta Potential
Before Adding Ethanolamine
Average
After Adding Ethanolamine
Average
Diameter
(nm)
Diameter
NP Code
PI
n (mV)
(nm)
PI
n (mV)
NP0
NP1
126
156
0.099
0.095
ꢁ69
ꢁ60
127
176
0.105
0.080
ꢁ80
ꢁ40
1348
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
Chem 2010, 48, 3244–3254; (d) Jubeli, E.; Moine, L.; Barratt, G.
J Polym Sci Part A: Polym Chem 2010, 48, 3178–3187.
particle mean diameter and polydispersity index given by
dynamic light scattering remained quasi the same. However,
pronounced hydrolysis of the activated esters was shown to
proceed with time (as also pointed out earlier18). The
amount of hydrolyzed esters could be obtained by measuring
the OD by adding the ethanolamine in the supernatant of the
particle dispersions after centrifugation. After 3 months, the
fraction of hydrolyzed esters on the total available was
determined to be more than 75%. This was confirmed by
the poor colloidal stability of the dispersion with a floccula-
tion observed at less than 0.5M NaCl, showing that the esters
were hydrolyzed to carboxyl ionic groups thus reducing the
steric contribution to stabilization. This underlines that the
reactive groups of these promising novel functionalized PLA
nanoparticles should be used within a limited time after
preparation, if not freeze-dried.
4 Avgoustakis, K. Curr Drug Deliv 2004, 1, 321–333.
5 Rieger, J.; Freichels, H.; Imberty, A.; Putaux, J. L.; Delair, T.,
Jerome, C.; Auzely-Velty, R. Biomacromolecules 2009, 10,
651–657.
6 Nagasaki, Y.; Yasugi, K.; Yamamoto, Y.; Harada, A.; Kataoka,
K. Biomacromolecules 2001, 2, 1067–1070.
7 Yamamoto, Y.; Nagasaki, Y.; Kato, Y.; Sugiyama, Y.; Kataoka,
K. J Control Release 2001, 77, 27–38.
8 Betancourt, T.; Byrne, J. D.; Sunaryo, N.; Crowder, S. W.;
Kadapakkam, M.; Patel, S.; Casciato, S.; Brannon-Peppas, L.
J Biomed Mater Res A 2009, 91, 263–276.
9 Theato, P.
J Polym Sci Part A: Polym Chem 2008, 46,
6677–6687.
CONCLUSIONS
10 Knop, K.; Hoogenboom, R.; Fischer, D.; Schubert, U. S.
Angew Chem Int Ed 2010, 49, 6288–6308.
In this work, MAMA-SG1 alkoxyamine-initiated NMP was suc-
cessfully applied to synthesis of well-defined PNAS and
P(NAS-co-NVP) polymers. The homopolymerization of NAS
was controlled in the presence of SG1. A strong alternating
tendency was observed for the copolymerization of NAS with
NVP and the obtained copolymers exhibited acceptable poly-
dispersities and adjustable molecular weights, consistent
with a controlled process. The NMP of NAS/NVP was then
initiated with a PLA-SG1 macro-alkoxyamine previously pre-
pared, leading to PLA-b-P(NAS-co-NVP) block copolymer. The
latter was involved as a particle surface modifier in the PLA
nanopreciptation and diafiltration processes, to achieve
nanoparticles of targeted size with a hydrophilic and highly
functionalized interface. The presence of the NS reactive
groups was clearly evidenced by zeta potential, colloidal sta-
bility measurements and UV-based determination of the
releasing NHS anion upon ethanolamine derivatization. We
are currently focusing on the freeze-drying process of these
particles to limit NS ester hydrolysis and investigating the
coupling on these novel particles of ligands such as mannos-
amine or immunostimulating molecules for further vaccine
delivery applications.
11 Dove, A. P. Chem Commun 2008, 6446–6470.
12 (a) Sha, K.; Li, D.; Wang, S.; Qin, L.; Wang, J. Polym Bull
2005, 55, 349–355; (b) de Geus, M.; Peeters, J.; Wolffs, M.; Her-
mans, T.; Palmans, A. R. A.; Koning, C. E.; Heise, A. Macromo-
lecules 2005, 38, 4220–4225; (c) Likhitsup, A.; Parthiban, A.;
Chai, C. L. L. J Polym Sci Part A: Polym Chem 2008, 46,
102–116; (d) Meyer, U.; Palmans, A. R. A.; Loontjens, T.; Heise,
A. Macromolecules 2002, 35, 2873–2875; (e) Smith, A. P.;
Fraser, C. L. Macromolecules 2003, 36, 2654–2660.
13 (a) You, Y. Z.; Hong, C. Y.; Wang, W. P.; Lu, W. Q.; Pan, C.
Y. Macromolecules, 2004, 37, 9761–9767; (b) Li, J. B.; Ren, J.;
Cao, Y.; Yuan, W. Z. Polymer 2010, 51, 1301–1310; (c) Ozturk,
T.; Goktas, M.; Hazer, B.
1638–1645.
J App Polym Sci 2010, 117,
14 (a) van As, B. A. C.; Thomassen, P.; Kalra, B.; Gross, R. A.;
Meijer, E. W.; Palmans, A. R. A.; Heise, A. Macromolecules
2004, 37, 8973–8977; (b) Pratt, R. C.; Lohmeijer, B. G. G.; Long,
D. A.; Pontus Lundberg, P. N.; Dove, A. P.; Li, H.; Wade, C. G.;
Waymouth, R. M.; Hedrick, J. L. Macromolecules 2006, 39,
7863–7871; (c) Nederberg, F.; Lohmeijer, B.; G. G.; Leibfarth, F.;
Pratt, R. C.; Choi, J.; Dove, A. P.; Waymouth, R. M.; Hedrick, J.
L. Biomacromolecules 2007, 8, 153–160.
The authors are grateful to Arkema, Universite´ de Provence and
CNRS for financial support and Marilyne Malbouyres (IBCP)
and Be´atrice Burdin from the Centre Technologique des Micro-
structures (Lyon, France) for SEM experiments. Part of this
work has been funded through two grants from European
Union (Europrise and Cuthivac) to E. Luciani and B. Verrier.
15 Cle´ment, B.; Trimaille, T.; Alluin, O.; Gigmes, D.; Mabrouk,
K.; Fe´ron, F.; Decherchi, P.; Marqueste, T.; Bertin, D. Biomacro-
molecules 2009, 10, 1436–1445.
16 Zhang, J.; Jiang, X.; Zhang, Y.; Li, Y.; Liu, S. Macromole-
cules 2007, 40, 9125–9132.
17 Rowe, M. D.; Thamm, D. H.; Kraft, S. L.; Boyes, S. G. Bioma-
cromolecules 2009, 10, 983–993.
REFERENCES AND NOTES
18 Li, Y.; Lokitz, B. S.; McCormick, C. L. Macromolecules 2006,
1 Jalil, R.; Nixon, J. R. J Microencapsul 1990, 7, 297–325.
39, 81–89.
2 Mundargi, R. C.; Babu, V. R.; Rangaswamy, V.; Patel, P.; Ami-
19 Relogio, P.; Charreyre, M.-T.; Farinha, J. P. S.; Martinho, J.
nabhavi, T. M. J Control Release 2008, 125, 193–209.
M. G.; Pichot, C. Polymer 2004, 45, 8639–8649.
3 (a) Zhuang, X.; Xiao, C.; Oyaizu, K.; Chikushi, N.; Chen, X.;
20 Pascual, S.; Monteiro, M. J. Eur Polym
J 2009, 45,
Nishide, H.
J Polym Sci Part A: Polym Chem 2010, 48,
2513–2519.
5404–5410; (b) Sun, L.; Shen, L.-J.; Zhu, M.-Q.; Dong, C.-M.;
Wei, Y. 2010, 48, 4583–4593; (c) Nottelet, B.; Di Tommaso, C.;
21 Kakwere, H.; Perrier, S.
J
Am Chem Soc 2009, 131,
Mondon, K.; Gurny, R.; Moller, M. J Polym Sci Part A: Polym
1889–1895.
¨
´
DENSELY FUNCTIONALIZED POLYLACTIDE NANOPARTICLES, HANDKE ET AL.
1349

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
22 Favier, A.; D’Agosto, F.; Charreyre, M.-T.; Pichot, C. Polymer
31 Van Herk, A.; Droge, T. Macromol Theor Simul 1997, 6,
¨
2004, 45, 7821–7830.
1263–1276.
23 Li, Y.; Akiba, I.; Harrisson, S.; Wooley, K. L. Adv Funct Mater
32 Meyer, V. E.; Lowry, G. G. J Polym Sci Part A: Polym Chem
2008, 18, 551–559.
1965, 3, 2843–2851.
24 De Lambert, B.; Chaix, C.; Charreyre, M.-T.; Martin, T.;
Aigoui, A.; Perrin-Rubens, A.; Pichot, C.; Mandrand, B. Anal
Biochem 2008, 373, 229–238.
33 Jeon, H. J.; Jeong, Y. I.; Jang, M. K.; Park, Y. H.; Nah, J. W.
Int J Pharm 2000, 207, 99–108.
34 Fessi, H.; Devissaguet, J.-P.; Puisieux, F.; Thies, C. U.S. Patent
25 Yu, X.; Tang, X. Z.; Pan, C. Y. Polymer 2005, 46, 11149–11156.
5,118,528, European Patent EP0275796B2, December 31, 1986.
26 Fukukawa, K.; Rossin, R.; Hagooly, A.; Pressly, E. D.; Hunt,
J. N.; Messmore, B. W.; Wooley, K. L.; Welch, M. J.; Hawker, C.
J. Biomacromolecules 2008, 9, 1329–1339.
35 (a) Gigmes, D.; Bertin, D.; Lefay, C.; Guillaneuf, Y. Macromol
Theor Simul 2009, 18, 402–419; (b) Chauvin, F.; Dufils, P.-E.;
Gigmes, D.; Guillaneuf, Y.; Marque, S. R. A.; Tordo, P.; Bertin,
D. Macromolecules 2006, 39, 5238–5250.
27 Erout, M.-N.; Elaissari, A.; Pichot, C.; Llauro, M.-F. Polymer
1996, 37, 1157–1165.
36 Bilalis, P.; Pitsikalis, M.; Hadjichristidis, N. J Polym Sci Part
A: Polym Chem 2005, 44, 659–665.
28 Yamanaka, H.; Yoshizako, K.; Akiyama, Y.; Sota, H.; Hase-
gawa, Y.; Shinohara, Y.; Kikuchi, A.; Okano, T. Anal Chem
2003, 75, 1658–1663.
37 Fineman, M.; Ross, S. D. J Polym Sci 1950, 5, 259–265.
38 Kelen, T.; Tudos, F. J Macromol Sci Chem A 1975, 9, 1–27.
¨
¨
29 Arehart, S. V.; Matyjaszewski, K. Macromolecules 1999, 32,
39 Mayo, F. R.; Lewis, F. M.
J Am Chem Soc 1944, 66,
2221–2231.
1594–1601.
30 Van Herk, A. J Chem Educ 1995, 72, 138–140.
40 Alfrey, T.; Goldfinger, G. J Chem Phys 1944, 12, 205–209.
1350
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA