RAFT polymerization and thio-bromo substitution: An efficient way towards well-defined glycopolymers

Article abstract of DOI:10.1002/pola.28745

Despite an increasing effort to design well-defined glycopolymers, the convenient synthesis of polymers with higher DPs (>100) and without tedious protection and deprotection steps remains a challenge. Combining the reversible addition fragmentation transfer (RAFT) polymerization and the efficient substitution of primary bromo groups by thiols, we were able to synthesize a set of well-defined glycopolymers with DPs of up to 115. With the polymerization of the highly reactive monomer (2-bromoethyl)-acrylate polymers with low dispersities were obtained that could efficiently be functionalized with various sugar thiol(ate)s. In particular, derivatives of d-glucose, d-galactose, and d-mannose gave excellent degrees of functionalization close to quantitative conversion using only a slight excess of the thiol. This atom efficient synthesis can even be applied for copolymers with acid or base labile components due to the use of unprotected sugar moieties and, hence, the lack of further deprotection steps. Binding studies with the lectin concanavalin A and the subsequent competition studies with α-d-methyl-mannopyranose (αMeMan) proved the effective binding of these derivatives and revealed a DP- and carbohydrate-dependent clustering and dissolution.

Full text of DOI:10.1002/pola.28745

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
RAFT Polymerization and Thio-Bromo Substitution: An Efficient Way
towards Well-Defined Glycopolymers
1,2
Michael Prohl, Christoph Englert,1,2 Michael Gottschaldt,1,2 Johannes C. Brendel
1,2
,
€
Ulrich S. Schubert1,2
1Laboratory of Organic and Macromolecular Chemistry (IOMC), Friedrich Schiller University Jena, Humboldtstraße 10, Jena
07743, Germany
2Jena Center for Soft Matter (JCSM), Friedrich Schiller University Jena, Philosophenweg 7, Jena 07743, Germany
Correspondence to: U. S. Schubert (E-mail: ulrich.schubert@uni-jena.de) or J. C. Brendel (E-mail: johannes.brendel@uni-jena.de)
Received 15 June 2017; accepted 18 July 2017; published online 00 Month 2017
DOI: 10.1002/pola.28745
ABSTRACT: Despite an increasing effort to design well-defined
glycopolymers, the convenient synthesis of polymers with
higher DPs (>100) and without tedious protection and depro-
tection steps remains a challenge. Combining the reversible
addition fragmentation transfer (RAFT) polymerization and the
efficient substitution of primary bromo groups by thiols, we
were able to synthesize a set of well-defined glycopolymers
with DPs of up to 115. With the polymerization of the highly
reactive monomer (2-bromoethyl)-acrylate polymers with low
dispersities were obtained that could efficiently be functional-
ized with various sugar thiol(ate)s. In particular, derivatives of
D-glucose, D-galactose, and D-mannose gave excellent degrees
of functionalization close to quantitative conversion using only
a slight excess of the thiol. This atom efficient synthesis can
even be applied for copolymers with acid or base labile com-
ponents due to the use of unprotected sugar moieties and,
hence, the lack of further deprotection steps. Binding studies
with the lectin concanavalin A and the subsequent competition
studies with a-D-methyl-mannopyranose (aMeMan) proved the
effective binding of these derivatives and revealed a DP- and
C
V
carbohydrate-dependent clustering and dissolution.
2017
Wiley Periodicals, Inc. J. Polym. Sci., Part A: Polym. Chem.
2017, 00, 000–000
KEYWORDS: carbohydrates; glycosylation; lectins; nucleophilic
substitution; polymers; postmodification; RAFT
Early glycopolymer synthesis was conducted by Kobayashi
et al. in 1985, who prepared lactose and D-maltose conju-
gated styrene polymers, offering strong interactions with
hepatocytes and D-galactose selective lectins.5 Since then,
various polymers with defined architectures can be obtained
due to the use of modern polymerization techniques and
available sugar derivatives. In general, there are two ways to
obtain carbohydrate-conjugated polymers: Polymerization of
carbohydrate-conjugated monomers (glycomonomers) or
postpolymerization conjugation of a suitable, reactive poly-
mer. Many attempts were performed to polymerize glycomo-
nomers in a controlled manner by anionic6 and cationic,7
controlled radical,8 ring-opening,9 and other polymerization
techniques.10 However, the synthesis of glycomonomers usu-
ally requires various steps and the functional groups of
unprotected glycomonomers reveal incompatibility with
most controlled polymerization techniques.11 Additionally,
high molar mass polymers with a narrow distribution are
still challenging to obtain by polymerizing glycomonomers.
To postmodify polymers with carbohydrates, a reactive
INTRODUCTION Synthetic, carbohydrate-functionalized poly-
mers gained increased interest in biological sciences due to
the crucial role of natural glycoconjugates in cell-cell recog-
nition processes. The main interactions of living organisms
are saccharide-protein, protein-protein, and protein-antibody
interactions and include binding possibilities in a multivalent
manner caused by hydrophobic and electrostatic interactions
as well as by hydrogen bonding.1 The saccharide-protein
interactions were found to play important roles in cell adhe-
sion and cell differentiation, but also in inflammations, viral
replications, and parasitic infections.2
Biological events are often related to the interaction between
carbohydrates and lectins. Lectins are binding proteins with high
stereo-specificity for carbohydrates. A single saccharide reveals
only a low affinity for its natural ligand, whereas multivalent
interactions between a single or more lectins with one or more
of the corresponding carbohydrate units are highly prevalent in
nature.3 The so-called “cluster-effect”4 strongly influences the
design of well-defined glycopolymer architectures.
Additional Supporting Information may be found in the online version of this article.
C
V
2017 Wiley Periodicals, Inc.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000
1

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 1 Schematic representation of the polymers P1 to P12 obtained via RAFT with subsequent postglyosylation.
EXPERIMENTAL
polymer backbone is required. For this purpose, many
functional groups are reported, including 4-nitrophenyl car-
Materials and Methods
bonate,12 para-fluoro-phenyl,13 alkinyl-,14 alkenyl,15 and
All reagents and solvents were commercial products pur-
others.11
chased from Sigma-Aldrich (triethylamine, dioxan, AIBN, Con
A), Roth (DMSO, Zellutrans dialysis tube), or Carbosynth (a-
D-thiomannose sodium salt) and were used without further
purification. (4-Cyanopentanoic acid)ylethyl trithiocarbonate
(CPAETC) was synthesized as previously reported.19,20 HBS
buffer (0.10 M HEPES, 0.9 M NaCl, 1 mM CaCl2, 1 mM MgCl2,
and 1 mM MnCl2) was prepared with Milli-Q water as the
solvent.
A relatively unexplored reaction for the postmodification of
polymers represents the nucleophilic substitution of various
alkyl halides with suitable nucleophiles. Mainly used in
metal-catalyzed polymerizations, the use of halide-containing
polymers obtained by reversible deactivation radical poly-
merization (RDRP) techniques is limited. This is primarily
attributed to the sensitivity of alkyl halides towards abstrac-
tion by radicals, which is intentionally used in iodine trans-
fer polymerization.16 However, only a few examples using
RDRP for alkyl halide monomers are reported and even less
are applied for the synthesis of glycopolymers. Stenzel and
coworkers reported the synthesis of star shaped glycopoly-
mers derived from poly(vinyl benzylchloride) (PVBC) via
reversible addition-fragmentation polymerization (RAFT).17
The polymers were reacted with equimolar amounts of 1-
deoxy-1-thio-b-D-glucopyranose sodium salt in DMSO for
Equipment
1H NMR spectra were measured with a Bruker spectrometer
(300 MHz). Elemental analysis was performed with a Leco
CHN-932. Size-exclusion chromatography (SEC) of polymers
P1 to P3 was performed on a Agilent system (series 1200)
equipped with a G1310A pump, a G1362A refractive index
detector and a PSS GRAM column with DMAc (1 0.21 wt %
LiCl) as eluent. The column oven was set to 40 8C and a
polystyrene (PS) standard was used for calibration. SEC of
polymers P4 to P12 was performed on a Jasco system
equipped with a PU-980 pump, a RI-2031 Plus refractive
index detector and a PSS SUPREMA column with H2O (1
0.1 M NaNO3 and 0.05% NaN3) as eluent. The column oven
was set to 30 8C and a pullulan standard was used for cali-
bration. UV/Vis absorbance spectra were measured with an
Analytik Jena AG Specord250 spectrometer. Quartz cuvettes
were purchased from Hellma Analytics.
110
h without any catalyst. Also poly(epichlorhydrin)
obtained via cationic ring-opening polymerization (CROP)
was postmodified with 1-deoxy-1-thio-b-D-glucopyranose to
obtain bristle-like polymers with moderate dispersity
(Ð 5 1.68).18 Recently, the Perrier group demonstrated the
versatility of poly(bromoethyl acrylate) (PBEA), which was
polymerized via RAFT.19 Various nucleophiles were used to
yield efficient substitution of the bromides, which resulted in
polymers with a wide range of functionalities including
already a D-glucose derivative. However, a rather large excess
of the nucleophile 1-deoxy-1-thio-b-D-glucopyranose sodium
salt was used to form the D-glucosylated acrylate. To the best
of our knowledge, there is no systematic study, which shows
the synthesis of glyco-homopolymers by controlled radical
polymerization techniques with low dispersities and DPs
exceeding 100 units comparing different sugar residues.
Syntheses of the Thio-Sugars
1-Deoxy-1-thio-b-D-glucopyranose and 1-deoxy-1-thio-b-D-gal-
actopyranose were synthesized according to literature
procedures.21
Synthesis of 2-Bromoethyl Acrylate (BEA)
The BEA monomer was synthesized according to a modified
literature procedure.19 Thirty-eight milliliter of acryloyl chlo-
ride (0.54 mol) were dissolved in 250 mL dry CH2Cl2 under
an Ar atmosphere and 82 mL of Et3N (0.59 mol) were
added. The mixture was cooled to 0 8C and 53 mL of 2-
bromoethanol (0.66 mol) dissolved in 40 mL CH2Cl2 were
added dropwise. The reaction mixture was allowed to stir
for 18 h at room temperature. Subsequently, the solution
was filtered and the organic layer was washed thrice with
saturated, aqueous NaHCO3 solution. Additionally, the
organic layer was stirred with 250 mL of 0.1 M NaOH(aq)
Based on these results, we attempted to further exploit the
reaction based on PBEA to synthesize a whole library of
various glycopolymers and to test their suitability for selec-
tive binding to well-known lectins (Fig. 1). For this pur-
pose, polymers with lengths ranging from 45 to 115
repeating units were prepared by RAFT polymerization and
subsequently modified with various sugar residues (D-glu-
cose, D-galactose, D-mannose). All these polymers were
finally tested for their affinity to the binding-protein conca-
navalin A (Con A).
2
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
solution for 24 h separated and dried with Na2SO4. The solu-
tion was filtered and the organic solvent was evaporated in
vacuo to obtain the crude product as a colorless oil. Ten mil-
ligram of 2,6-di-tert-butyl-4-methylphenol were added and
the product was subjected a fractionated destillation under
vacuum to yield 34.97 g (0.2 mol) of the desired product in
highest purity and 37% yield. bp 55 8C (4.5 mmHg). dH (300
MHz; CDCl3) 6.49 (m, 1H, CH2@CH), 6.19 (m, 1H, CH2@CH),
5.90 (m, 1H, CH2@CH), 4.48 (t, 2H, 3J 6.2, CH2CH2Br), and
g mol21, Ð 5 1.19. Anal. calcd for C503H824O317S48: C, 44.81;
H, 6.16; S, 11.41; Br, 0. Found: C, 43.87; H, 6.17; S, 10.81; Br,
0.89.
P5
3
1H NMR (300 MHz; D2O, d ppm) 4.44 (n H, d, J 9.3, HGal-1),
4.25 (2n H, br s), 3.89 (n H, s), 3.82–3.41 (5n H, m), 3.16–
2.72 (2n H, m), 2.36 (2n H, br s), 2.04–1.00 (2n 1 1 H, m).
SEC (H2O, 0.1 M NaNO3, 0.05% NaN3, pullulan standard):
Mn 5 9,900 g mol21, Mw 5 12,000 g mol21, Ð 5 1.22. Anal.
calcd for C503H824O317S48: C, 44.81; H, 6.16; S, 11.41; Br, 0.
Found: C, 41.88; H, 6.13; S, 11.99; Br, 1.99.
3
4.48 (t, 2H, J 6.2, CH2CH2Br).
General Procedure for the Syntheses of PBEA
Homopolymers
P7
CPAETC, BEA, and AIBN were dissolved in dioxane and the
reaction mixture was deoxygenated with Ar for 10 min and
stirred at 65 8C for 4 h. After completion, the solution was
cooled to room temperature, opened to air, and precipitated
in diethyl ether to give PBEA homopolymers P1 to P3.
3
1H NMR (300 MHz; D2O, d ppm) 4.49 (n H, d, J 9.7, HGlc-1),
4.24 (2n H, br s), 3.85–3.70 (n H, m), 3.70–3.56 (n H, m),
3.55–3.18 (4n H, m), 3.11–2.70 (2n H, m), 2.37 (2n H, br s),
2.11–1.00 (2n 1 1 H, m). SEC (H2O, 0.1 M NaNO3, 0.05%
NaN3, pullulan standard): Mn 5 17,400 g mol21, Mw 5 21,800
g mol21, Ð 5 1.25. Anal. calcd for C866H1418O548S81: C, 44.84;
H, 6.16; S, 11.20; Br, 0. Found: C, 44.15; H, 6.18; S, 9.48; Br,
2.14.
P1
1H NMR (300 MHz; CDCl3, d ppm) 4.99–4.88 (1 H, m), 4.42
3
(2n H, br s), 3.57 (2n H, br s), 3.40 (2 H, t, J 7.4), 2.49 (2n
H, br s), 2.05 (2n H, br s), 1.94–1.50 (2n 1 1 H, m), 1.48–
1.41 (3 H, m), 0.95 (3 H, t, 3J 7.3). SEC (DMAc 1 0.21 wt %
P8
LiCl, PS standard): Mn 5 8,800 g mol21, Mw 5 9,600 g mol21
Ð 5 1.10.
,
1H NMR (300 MHz; D2O, d ppm) 4.44 (n H, d, J 8.9, HGal-1),
3
4.24 (2n H, br s), 3.90 (n H, s), 3.83–3.41 (5n H, m), 3.17–
2.71 (2n H, m), 2.38 (2n H, br s), 2.04–0.99 (2n 1 1 H, m).
SEC (H2O, 0.1 M NaNO3, 0.05% NaN3, pullulan standard):
Mn 5 16,400 g mol21, Mw 5 21,000 g mol21, Ð 5 1.28. Anal.
calcd for C866H1418O548S81: C, 44.84; H, 6.16; S, 11.20; Br, 0.
Found: C, 43.28; H, 6.18; S, 9.31; Br, 3.09.
P2
1H NMR (300 MHz; CDCl3, d ppm) 4.98–4.88 (1 H, m), 4.43
3
(2n H, br s), 3.57 (2n H, br s), 3.39 (2 H, t, J 7.2), 2.47 (2n
H, br s), 2.15–1.97 (2n H, br s), 1.90–1.39 (2n H, m), 0.96
(3 H, t, 3J 7.4). SEC (DMAc 1 0.21 wt % LiCl, PS standard):
Mn 5 11,200 g mol21, Mw 5 13,500 g mol21, Ð 5 1.21.
P10
1H NMR (300 MHz; D2O, d ppm) 4.49 (n H, d, J 9.7, HGlc-1),
3
P3
4.24 (2n H, br s), 3.86–3.69 (n H, m), 3.69–3.55 (n H, m),
3.55–3.17 (4n H, m), 3.08–2.71 (2n H, m), 2.36 (2n H, br s),
2.06–1.00 (2n 1 1 H, m). SEC (H2O, 0.1 M NaNO3, 0.05%
NaN3, pullulan standard): Mn 5 23,300 g mol21, Mw 5 29,300
g mol21, Ð 5 1.26. Anal. calcd for C1273H2084O807S118: C,
44.86; H, 6.16; S, 11.10; Br, 0. Found: C, 43.53; H, 6.20; S,
10.20; Br, 1.54.
1H NMR (300 MHz; CDCl3, d ppm) 4.98–4.88 (1 H, m), 4.42
3
(2n H, br s), 3.57 (2n H, br s), 3.38 (2 H, t, J 6.4), 2.47 (2n
H, br s), 2.14–1.96 (2n H, m), 1.92–1.39 (2n H, m), 0.95
(3 H, t, 3J 7.3). SEC (DMAc 1 0.21 wt % LiCl, PS standard):
Mn 5 18,000 g mol21, Mw 5 20,000 g mol21, Ð 5 1.11.
General Procedure for the Postpolymerization
Modification with Glc and Gal
P11
One hundred milligram of the precursor polymer was dis-
solved in 1 mL DMSO and a solution of the desired carbohy-
drate (DPꢀ1.1 equiv.) in 1 mL DMSO was added. The solution
was deoxygenated with Ar for 30 min and 1 equiv. (based
on the carbohydrate) dry triethylamine was added dropwise.
The reaction mixture was stirred at room temperature for
one day and dialyzed against water for one week (CE,
MWCO: 3.5 kDa). The dialyzed solution was freeze-dried to
obtain the desired polymer.
3
1H NMR (300 MHz; D2O, d ppm) 4.44 (n H, d, J 9.1, HGal-1),
4.24 (2n H, br s), 3.89 (n H, s), 3.83–3.41 (5n H, m), 3.22–
2.72 (2n H, m), 2.37 (2n H, br s), 2.09–1.00 (2n 1 1 H, m).
SEC (H2O, 0.1 M NaNO3, 0.05% NaN3, pullulan standard):
Mn 5 23,300 g mol21, Mw 5 29,900 g mol21, Ð 5 1.28. Anal.
calcd for C1273H2084O807S118: C, 44.86; H, 6.16; S, 11.10; Br,
0. Found: C, 41.73; H, 6.05; S, 10.58; Br, 2.52.
General Procedure for the Postpolymerization
Modification with Man
P4
1H NMR (300 MHz; D2O, d ppm) 4.49 (n H, d, J 9.8, HGlc-1),
3
One hundred milligram of the precursor polymer was dis-
solved in 1 mL DMSO and a solution of Man-SNa (DPꢀ1.1
equiv.) in 0.5 mL H2O was added. The solution was deoxygen-
ated with Ar for 30 min, stirred at room temperature for one
day and dialyzed against water for one week (CE, MWCO: 3.5
4.24 (2n H, br s), 3.85–3.69 (n H, m), 3.69–3.55 (n H, m),
3.47–3.17 (4n H, m), 3.07–2.74 (2n H, m), 2.33 (2n H, br s),
2.03–1.00 (2n 1 1 H, m). SEC (H2O, 0.1 M NaNO3, 0.05%
NaN3, pullulan standard): Mn 5 10,000 g mol21, Mw 5 11,800
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000
3

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 1 Summary of BEA Polymerization
Conv.a (%)
Mn,th (g/mol)
Mn,NMR (g/mol)
Mn,SEC (g/mol)
Ð
b
c
d
Abbrev.
[M]0/CTA
[CTA]/I0
P1
P2
P3
60
10
10
5
69
58
42
7,600
8,300
8,800
1.10
1.21
1.11
150
300
15,700
22,700
14,200
20,800
11,200
18,000
a
Determined from 1H NMR of the polymerization mixture before
(d 5 4.93 ppm) in comparison to the proton signal of the C1 atom of the
acryl ester (d 5 4.43 ppm) before postmodification]. These ratios were
used to calculate the DP.
precipitation.
b
Calculated from monomer conversion.
Determined from 1H NMR end-group analysis [calculated from signal
c
d
SEC: DMAc 1 0.21 wt % LiCl, PS calibration.
intensity of the proton of the tertiary C-atom next to trithiocarbonate
kDa). The dialyzed solution was freeze-dried to obtain the
desired polymer.
Reversal Aggregation Assay
The competition experiments with a-D-methyl-mannopyra-
nose were carried out according to literature procedure.22
The solutions of the turbidimetry assay were allowed to rest
for 2 h at room temperature. Following the addition of
0.2 mL of a 54 mM stock solution of methyl-a-D-mannopyra-
noside in HBS buffer, the absorbance at k 5 420 nm was
immediately recorded for 60 min every 0.5 s.
P6
1H NMR (300 MHz; D2O, d ppm) 5.30 (n H, s, HMan-1), 4.49–
4.07 (3n H, m), 3.97 (n H, s), 3.93–3.49 (4n H, m), 2.88 (2n
H, s), 2.38 (2n H, br s), 2.11–0.93 (2n 1 1 H, m). SEC (H2O,
0.1 M NaNO3, 0.05% NaN3, pullulan standard): Mn 5 3,300
g mol21, Mw 5 4,800 g mol21, Ð 5 1.44. Anal. calcd for
C503H824O317S48: C, 44.81; H, 6.16; S, 11.41; Br, 0. Found: C,
43.76; H, 6.27; S, 10.36; Br, 0.
RESULTS AND DISCUSSION
RAFT Polymerization of BEA
Similar to the previously published procedure, the BEA
monomer was polymerized using CPAETC as chain transfer
agent (CTA) and AIBN as radical initiator for the RAFT pro-
cess. Optimizing the synthesis and purification procedure for
the monomer (see experimental section for details) we were
able to increase the final DP of the polymers without the
loss of control. As a consequence, various DPs could be
addressed by changing the ratio [monomer]/[CTA] (Table 1).
P9
1H NMR (300 MHz; D2O, d ppm) 5.30 (n H, s, HMan-1), 4.50–
4.06 (3n H, m), 3.97 (n H, s), 3.92–3.47 (4n H, m), 2.88 (2n
H, s), 2.38 (2n H, br s), 2.12–0.93 (2n 1 1 H, m). SEC (H2O,
0.1 M NaNO3, 0.05% NaN3, pullulan standard): Mn 5 9,400
g mol21, Mw 5 12,600 g mol21, Ð 5 1.34. Anal. calcd for
C866H1418O548S81: C, 44.84; H, 6.16; S, 11.20; Br, 0. Found: C,
43.80; H, 6.14; S, 10.45; Br, 0.
For all polymers, low dispersity (Ð ꢂ 1.2) and narrow
mono-modal distributions were obtained, emphasizing the
capabilities to polymerize halide-containing monomers by
RDRP. In addition, we analyzed the end group fidelity using
NMR analysis. The intensities of the signals at d 5 0.95 ppm
for the terminal CH3 group and of the signal at d 5 3.38 ppm
for the CH2 group next to the trithiocarbonate match well
with the expectations proving the excellent retention of the
RAFT end group. Recent studies showed the synthesis of a
similar polymer named poly(2-bromoethyl methacrylate) by
postmodification of poly(2-hydroxyethyl methacrylate) with
a mixture of various chemicals.23 Our study represents a
more effective approach due to the lack of postbromination
procedures and the introduction of the bromide directly to
the monomer circumventing the requirement for additional
tedious modification steps.
P12
1H NMR (300 MHz; D2O, d ppm) 5.30 (n H, s, HMan-1), 4.47–
4.06 (3n H, m), 3.98 (n H, s), 3.93–3.46 (4n H, m), 2.88 (2n
H, s), 2.38 (2n H, br s), 2.15–0.97 (2n 1 1 H, m). SEC (H2O,
0.1 M NaNO3, 0.05% NaN3, pullulan standard): Mn 5 15,900
g mol21, Mw 5 21,200 g mol21, Ð 5 1.33. Anal. calcd for
C1273H2084O807S118: C, 44.86; H, 6.16; S, 11.10; Br, 0. Found:
C, 43.83; H, 6.16; S, 10.42; Br, 0.36.
Turbidimetry Assay
The aggregation studies with Con A were conducted as pre-
viously reported.22 Con A was fully dissolved in HBS buffer
(ꢁ 1 mg mL21) and diluted to a 1 lM stock solution. 1 mL
of the Con A stock solution was added to a quarz glass
cuvette, which was placed in the UV–Vis spectrometer. One
milliliter of a 50 lM (50 lM per sugar unit) stock solution
of the polymer in HBS buffer was appended with a pipette
to the ground of the glass cuvette and the absorbance of the
mixture was immediately recorded at k 5 420 nm for 30 min
every 0.5 s. The interaction rate was calculated by using the
slope of the linear fit of the steepest portion. Every experi-
ment was conducted thrice.
Postglycosylation
With the defined precursor polymers at hand, the various car-
bohydrates were examined for the next step. For this purpose,
the thiolated carbohydrates (1-deoxy-1-thio-b-D-glucopyra-
nose,21 1-deoxy-1-thio-b-D-galactopyranose,21 and commer-
cially available 1-deoxy-1-thio-b-D-mannopyranose sodium
salt) were reacted with precursor polymers P1 to P3 in a SN2
reaction (Table 2). These carbohydrates were chosen for
4
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 2 Overview of the Glycopolymers P4 to P12 Prepared
reactions with polymers P1 to P3 were conducted in DMSO
and the thiol functionalities of the carbohydrates were
deprotonated with one equivalent of triethyl amine. The
reactions were stopped after 24 h and purified by dialysis
against H2O (MWCO: 3.5 kDa) to remove low-molar mass
impurities and side-products such as DMSO and formed
triethyl ammoniumbromide. In the case of D-mannose attach-
ment, 1.1 equivalents of the commercially available sodium
salt of the a-1-thiol derivative of D-mannose were used.
Therefore, no base is required and the formation of sodium
bromide as a side-product represents the driving force for
the substitution reaction. The mannosylated polymers were
purified in a similar manner to the gluco- and galactosylated
ones. The success of the reaction can be shown by various
NMR techniques as depicted for polymer P12 in Figure 2,
representative for all other polymers. All spectra of polymers
P1 to P12 are available in the Supporting Information (Fig.
S1–S47).
and Tested in This Study
Attached
Sugar (R1)
a
b
Abbrev.
DP
Mn,th
Mn,SEC
Ð
P4
45
Glc
13,500
13,500
13,500
23,200
23,200
23,200
34,100
34,100
34,100
10,000
9,900
1.19
1.22
1.44
1.25
1.28
1.34
1.26
1.28
1.33
P5
45
Gal
Man
Glc
P6
45
3,300
P7
78
17,400
16,400
9,400
P8
78
Gal
Man
Glc
P9
78
P10
P11
P12
115
115
115
23,300
23,300
15,900
Gal
Man
a
Calculated by assuming polymers with the respective DP and full sub-
stitution of the bromide groups.
b
SEC: H2O, 0.1 M NaNO3, 0.05% NaN3, Pullulan calibration.
The substitution of the bromides in P3 against the sugar-
thiolate results in the formation of a thio-ether. This is
clearly visible in the shift of the ethyl acrylate signals from
d 5 4. 42 ppm in P3 to d 5 4.22 ppm in P12, respectively,
from d 5 3.57 ppm in P3 to d 5 2.89 ppm in P12 in the 1H
NMR spectra. Additionally, the signal of the C1 proton
appears nicely isolated at d 5 5.30 ppm, whereas the other
D-mannose proton signals are visible between d 5 3.5 and
4.5 ppm. As also previously shown,28 the attachment of car-
bohydrates result in water-soluble polymers P4 to P12,
whereas precursor polymers P1 to P3 revealed nonsolubility
in water. Additionally, by binding D-mannose S-glycosidically
to the polymer backbone, a thioether is formed, which pos-
sesses a lower rate of hydrolysis of the thioglycosidic bond
by enzymatic cleavage relative to O-glycosydically coupled
sugar residues.29
different reasons. D-galactose (Gal) conjugated polymers were
shown to bind to hepatocytes via a receptor-mediated mecha-
nism and, therefore, representing useful targeting moieties for
the development of polymers used in various biomedical
fields, such as tissue engineering and drug-loaded nanopar-
ticles.24 D-Mannose (Man) is in the focus of the scientific com-
munity due to its strong interactions with the lectin Con A
and with macrophages, which overexpress mannose-receptors
on their surface.25 D-Glucose (Glc) interacts with transport
proteins like with the GLUT1 transporter, which is shown to
be overexpressed in various types of cancers, such as renal
cell carcinoma26 or non-small cell lung carcinoma.27
D-Galactose and D-glucose were attached using only 1.1
equivalents of the deprotected b-1-thiol derivatives. The SN2
FIGURE 2 Left: Comparison of 1H NMR of compounds P3 (measured in CDCl3) and P12 (measured in D2O) between d 5 0 and 6
ppm. Intermittent axis between d 5 4.6 and 5 ppm was used due to solvent residue signal in the spectrum of P12. Right: 1H
diffusion-ordered NMR (DOSY) at T 5 25 8C of P12. [Color figure can be viewed at wileyonlinelibrary.com]
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000
5

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 3 Elemental Compositions of Polymers P4 to P12; the Calculated DFs are Based on Theoretical Br Content of Polymers P1
to P3
Abbrev.
DP
Saccharide
Elemental Compositiona (%)
DF (%)
C
H
S
Br
P1
45
–
33.74
43.87
41.88
43.76
33.66
44.15
43.28
43.80
33.63
43.53
41.73
43.83
4.00
6.17
6.13
6.27
3.97
6.18
6.18
6.14
3.96
6.20
6.05
6.16
1.16
43.35
0.89
1.99
0
–
P4
45
Glc
Gal
Man
–
10.81
11.99
10.36
0.68
97.9
95.4
100
–
P5
45
P6
45
P2
78
43.89
2.14
3.09
0
P7
78
Glc
Gal
Man
–
9.48
95.1
93
P8
78
9.31
P9
78
10.45
0.46
100
–
P3
115
115
115
115
44.12
1.54
2.52
0.36
P10
P11
P12
Glc
Gal
Man
10.20
10.58
10.42
96.5
94.3
99.2
a
Elemental composition of starting material polymers P1 to P3 was calculated assuming one polymer species with the depicted DP.
1H diffusion-ordered NMR (DOSY) investigations of P12
revealed the appearance of all signals (except the signal for
the solvent residue) at high diffusions constants typical for
polymers (D 5 5ꢀ10211 m2 s21), indicating the successful
attachment of D-mannose units without the appearance of
any side-products.
in comparison to the solubility of the formed triethyl ammo-
niumbromide of the other polymers in DMSO. Additionally,
the DF with D-galactose is generally decreased in comparison
to D-glucose, which might be caused by the existence of
small amounts of 1-deoxy-1-thio-a-D-galactopyranose, which
could be sterically hindered during the substitution reaction.
Due to the well resolved signals from the C1 protons of the
1
In contrast to the previous polymers P1 to P3 the end group
fidelity of the RAFT group could not be determined after
postpolymerization functionalization due to the overlap of
the important signals in the NMR with the signals of the
sugar moiety. Reinitiation experiments, however, indicated a
degradation of the trithiocarbonate as no block formation is
observed.
D-mannose units in H NMR, they can be used to validate the
DF by comparing their intensity to that of the backbone sig-
nal at d 5 2.9 ppm. The obtained data is in good agreement
with the results obtained from the elemental analysis of the
polymers.
The polymers P4 to P12 were also investigated via SEC to
obtain information about the increase of the molar mass of
the polymers and their dispersities (Fig. 3).
To further evaluate the quality of the substitution reaction,
the elemental composition of polymers P4 to P12 was ana-
lyzed. The determined halogen content represents the
remaining bromoethyl ester content and can be used to cal-
culate the degree of functionalization (DF) with the following
equation (Table 3).
The traces still show a mono-modal distribution and the dis-
persity remains narrow below 1.5, indicating no formation of
macromolecular side-products during the SN2 reaction and
the absence of previously observed chain-chain coupling due
to the removal of the CTA and the subsequent disulfide for-
mation or other side reactions.19 The latter can certainly be
attributed to the reduced excess of thio-sugar molecules
used for the substitution. The symmetric broadening of the
SEC traces and the slight increase of the dispersities relative
to the precursor polymers P1 to P3 is most probably related
to the difference in the SEC system applied for the analysis
of the polymers and potential interactions of the attached
sugar moieties with the column material. An indication for
this phenomenon is further the later elution of D-mannosy-
lated polymers P6, P9, and P12 in comparison to the corre-
sponding D-glucosylated or D-galactosylated polymers,
respectively, although the absolute molar masses should be
very similar. An additional effect for this shift in the elution
Br content ðsugar polymerÞ
DF5 12
3100
Br content ðPBEAÞ
The DF ranges from 93 to 100% and some general tenden-
cies are noticeable. D-Mannose functionalized polymers P6,
P9, and P12 showed remaining Br content between 0 and
the lower detection limit (0.36%), whereas the other poly-
mers revealed Br content between 0.89 and 3.09%. There-
fore, the use of the sodium salt of the corresponding thiol-
sugars, instead of the thiol-sugars and a suitable base, seems
to be beneficial for the substitution reaction. This might be
attributed to the decreased solubility of the formed sodium
bromide in the solvent mixture of D-mannosylated polymers
6
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
using the lectin Con A. Con A consists of aggregates of
25,000 g mol21 size. While existing at pH range 5–5.6 as a
dimer, Con A is predominantly aggregating into tetramers
above pH 5 7.30 Each monomer unit is known to selectively
bind to one unit of a-gluco- or a-mannopyranose, but no
binding should be observed in the case of the galactopyra-
nose.31 The rate of clustering was monitored in real-time by
measuring the absorbance at k 5 420 nm over time after
mixing the lectin and polymer solutions. The change in tur-
bidity is related to the rate of receptor-receptor associations
caused by the sugar-units of the polymers.32 The slope of
the steepest portion of the initial curve was used to repre-
sent the clustering rate, expressed in arbitrary units per sec-
ond (a.u. s21).22 The initial values of the curves are
correlated to the formation of isolated Con A polymer clus-
ters, whereas interactions between the clusters occur at later
points. The formation of cross-linked clusters of higher order
increases over the time and, therefore, the analysis is limited
to the initial portion of the curve.33 The experimental results
are summarized in Figure 4. All spectra of the triplicate
measurements including the linear fits are available in the
Supporting Information (Fig. S9, S14, S18, S23, S28, S32,
S37, S42, and S46).
FIGURE 3 SEC traces of polymers P4 to P12 (H2O, 0.1 M
NaNO3, 0.05% NaN3, Pullulan standard). [Color figure can be
viewed at wileyonlinelibrary.com]
volume might be the interaction between the axial C2
hydroxyl groups of the D-mannose units and the carboxylic
ester functionalities on the polymer backbone, which
decrease the resulting hydrodynamic radius, whereas D-glu-
cose and D-galactose possess equatorial hydroxyl groups at
their C2 atoms. Another reason could be the stronger inter-
action of D-mannose with the column material in comparison
to the other sugars. D-Glucose conjugated polymers roughly
eluted after the same volume than the D-galactolysated poly-
mers with the same DP.
Polymers P5, P8, and P11 bearing D-galactose residues did
not show any aggregation with Con A over a time period of
30 min, which is in accordance to previously reported D-gal-
actolysated polymers.14 D-Glucosylated and D-mannosylated
polymers exhibited the formation of Con A clusters with
varying clustering rates. As a general trend, the polymers
grafted with D-mannose residues revealed higher rates of
Con A clustering than their D-glucosylated polymers with the
same DP. This might be attributed to the higher binding
affinity of the Con A tetramers to a-D-mannose in comparison
to D-glucose residues. Another reason could be the decreased
hydrodynamic radius of D-mannosylated polymers relative to
the D-glucosylated ones, which could offer beneficial shape
or length of the polymers in solution in terms of Con A bind-
ing. The absorbance of D-mannose bearing polymers reached
Lectin Binding
One critical requirement for the application of glycopoly-
mers, for example, in targeted drug delivery, is their ability
to selectively bind to lectins for the respective type of sugar.
To examine the binding efficacy of the glycopolymers, turbi-
dimetry assays were performed for the polymers P4 to P12
FIGURE 4 Results of turbidimetry measurements. Left: Absorbance (k 5 420 nm) curves after adding 1 mL solution of polymers P4
to P12 (50 lM per sugar unit) to 1 lM solutions of Con A in HBS buffer. Right: Calculated rates of clustering between Con A tet-
ramers and D-glucosyl- or D-mannosylated polymers obtained by a linear fit of the steepest portions of the curves, k values repre-
sent the average of three replicates. [Color figure can be viewed at wileyonlinelibrary.com]
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000
7

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 5 Results of reversal aggregation measurements. Left: Absorbance (k 5 420 nm) curves after adding 0.2 mL solution of
aMeMan (54 mM) to the polymer solutions (50 lM per sugar unit) in HBS buffer. Right: Calculated rates of the reverse interaction
between Con A aggregates and the competitor aMeMan obtained by a linear fit of the steepest portions of the curves. [Color figure
can be viewed at wileyonlinelibrary.com]
very fast a plateau and stayed almost constant at this level
for the remaining measurement, which is attributed to a
rapid precipitation of most of the Con A tetramers (Fig. 4).
In contrast, D-glucosylated polymers exhibited a continuous,
but slower increase of the absorbance, indicating secondary
interactions such as cross-linked clusters or partially soluble
conjugates. In particular, the shortest D-glucose bearing poly-
mer P4 exhibited different clustering rates compared to all
other polymers. One reason is the general decreased affinity
of D-glucose to Con A relative to D-mannose. Another one is
certainly the overall length of the polymer. Considering a
fully stretched chain the length of P4 can be estimated to be
ꢁ80 Å taking into account the binding angles. The distance
between two binding sites of the Con A tetramer is around
72 Å and,34 therefore, the length of the polymer, which is
just above the distance between the binding sites, paired
with the low affinity of Con A towards D-glucose did not
result in fast agglutination of the Con A clusters. In the light
of this, the existence of a critical polymer length for high
clustering rates of Con A aggregates is assumed, which is
also in accordance to literature reports.35
higher DPs are necessary. The described polymerization and
postglycosylation could be used to synthesize these polymers
in future studies.
Reversal Aggregation
In addition to the previous binding assays, the strengths and
efficiencies of the interaction between glycopolymer and lec-
tin can be evaluated by competition experiments. For this
purpose, the aggregates formed during turbidimetry assays
were allowed to rest for around 2 h to finish the formation
of higher-order aggregates and were subsequently treated
with an excessive amount of a-D-methyl-mannopyranose
(aMeMan), a competitor for the binding sites of Con A. The
absorbance at k 5 420 nm over the time was monitored and
the results are summarized in Figure 5 for D-manno- and D-
glucosylated polymers. The D-galactose bearing polymers
were omitted due to the inefficient binding to the lectin Con
A and all spectra are available in the Supporting Information
(Fig. S10, S19, S24, S33, S38, and S47).
The results of the reversal aggregation assay revealed
carbohydrate-specific and length-dependent tendencies. The
rates of the dissolution of the D-glucosylated polymers are
strongly enhanced relative to those of the D-mannosylated
polymers when treated with the monovalent competitor
aMeMan in large excess. The turbidity was rapidly reduced
to a constant level, indicating a rather weak interaction
between Con A and the D-glucose bearing polymers. Addi-
tionally, the dissolution of the D-glucose Con A clusters hap-
pened faster with increasing DP of the polymers. The
dissolution of the cluster of D-mannosylated polymers was
slow but continuous to a constant level, indicating higher
binding affinity of the Con A tetramers to D-mannosylated
polymers. Other tendencies are observable when comparing
the difference of the highest and the lowest absorbance
(k 5 420 nm) after addition of Con A to the difference
Another trend is obtained by comparing the rate of cluster-
ing caused by the polymers with the same sugar residues,
but with different DPs. With higher DPs (for D-glucose as
well as for D-mannose bearing polymers) the precipitation of
Con A clusters is promoted more rapidly. Therefore, we
assume a dependency of the speed of clustering with the
epitope density, which was also reported for other glycopoly-
mers.33,35 Additionally, the rate of clustering for polymers
with D-mannose residues seems to approximate a constant
level in dependency of the DP. A lower increase of the clus-
tering rates for D-mannosylated polymers with higher DPs
relative to P12 in comparison to the increase between P9 to
P12 is expected. To exactly determine the required DP, more
turbidimetry investigations with D-mannosylated polymers of
8
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
repeating units. The bromides were readily substituted in a
post-glycosylation procedure by addition of almost equal
amounts of sugar-thiol(ate)s, in particular D-glucose, D-galac-
tose, and D-mannose, in an SN2 reaction. The substitution
was performed directly with unprotected carbohydrates,
emphasizing the versatility of this method towards potential
copolymers or linker structures, which are labile under the
basic or acidic conditions usually applied for the deprotec-
tion of sugar units. Furthermore, the reactions are highly
atom efficient and create only nontoxic side-products
(triethyl ammoniumbromide, sodium bromide). The analytics
of the novel polymers revealed mono-modal distributions
with narrow dispersities and excellent DF of up to 100%,
which was confirmed by elemental analysis. Binding studies
of Con A with the synthesized polymers as multivalent
ligands prove the selective binding ability of the sugar moie-
ties as expected for the small molecules. D-Mannosylated
polymers revealed higher clustering rates than the D-glucosy-
lated polymers, whereas D-galactose bearing polymers
showed no formation of cluster with Con A at all. With the
linear, high molar mass polymers at hand, we could also
demonstrate that the precipitation of Con A clusters is pro-
moted more rapidly with increasing DP of the polymers.
Reverse results were obtained by treating the Con A clusters
with monovalent competitor a-D-methyl-mannopyranose
(aMeMan). The turbidity of the D-glycosylated polymer clus-
ters was rapidly reduced to a constant level, whereas the D-
mannosylated ones decreased slower but continuous. This
indicates a weak interaction between the Con A binding site
and the multivalent D-glucose bearing polymers. Additionally,
the remaining undissolved clusters of D-glucosylated and D-
mannosylated polymers indicate an excellent binding affinity
of these new types of polymers, which surpasses most other
reported polymers with respective sugar groups attached.
FIGURE 6 Ratio of the dissolved Con A polymer clusters after
adding aMeMan relative to the DP. The values were calculated
by the ratio of the moduli of the difference between the high-
est and lowest absorbance after adding aMeMan respectively
Con A to the polymer solutions. [Color figure can be viewed at
wileyonlinelibrary.com]
obtained after addition of aMeMan. First of all, the Con A
aggregates of D-mannosylated polymers (P6, P9, P12) were
not completely dissolved after adding the competitor and a
clear trend is that with higher DP, the amount of undissolved
Con A clusters remaining in solution increases (Fig. 6, Sup-
porting Information Table S1).
Comparing the different sugar moieties (except for the D-gal-
actolysated polymers) all aggregates with D-glucosylated pol-
ymers revealed decreased stability relative to the respective
clusters of the D-mannosylated polymers with the same DP.
This is in accordance with the results we obtained for the
reversal aggregation assay showing higher dissolution rates
for increasing DPs of the polymers (P4 < P7 < P10). It is fur-
ther noteworthy to mention that we observe an increased total
stability of the D-glucosylated polymers P7 (69%) and P10
(62%) although Con A clusters of D-glucose are commonly fully
dissolved after adding aMeMan.22 For the D-mannosylated poly-
mers an even higher stability is observed despite the addition
of almost 1000 equivalents of the competitor aMeMan, which
has the same binding motif. This could indicate beneficial prop-
erties of the presented polymers in terms of their length, the
flexibility of the polymeric backbone and way of attachment of
the pendant sugar moieties, which may yield in an improved
binding strength to the lectin.
This work demonstrated the use of RAFT for the polymeriza-
tion of bromide-containing monomer BEA with high control
and DPs higher than 100 as well as the subsequent post-
glycosylation with almost quantitative DFs by versatile SN2
reaction with thiol(ate) derivatives of various sugars. The D-
mannose bearing polymers via thioether functionality revealed
the formation of very stable clusters with the lectin Con A.
ACKNOWLEDGMENTS
The authors gratefully thank Gabi Sentis and Peter Bellstedt for
the conducted DOSY NMR measurements and Beate Lentvogt
€
and Sandra Kohn for the performed determination of the ele-
mental compositions. The funding of the collaborative research
center ChemBioSys (SFB 1127) by the Deutsche Forschungsge-
meinschaft (DFG) is highly acknowledged. J.C. Brendel further
thanks the DFG for support (Return Grant, BR 4905/2–1).
CONCLUSION
In summary, this work demonstrates the high potential of
the combination of RAFT polymerization and the bromo-thio
substitution to create different, well-defined sugar-conju-
gated polymers. The basis for the reactive scaffold is the (2-
bromo-ethyl)-acrylate monomer that enables excellent con-
trol in the RAFT polymerization and high DPs of up to 115
REFERENCES
1 Y. Miura, Y. Hoshino, H. Seto, Chem. Rev. 2016, 116, 1673.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000
9

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
2 (a) Y. Miura, Polym. J. 2012, 44, 679; (b) R. A. Dwek, Chem.
Rev. 1996, 96, 683.
20 C. J. Ferguson, R. J. Hughes, D. Nguyen, B. T. T. Pham, R.
G. Gilbert, A. K. Serelis, C. H. Such, B. S. Hawkett, Macromole-
cules. 2005, 38, 2191.
3 (a) R. J. Pieters, Org. Biomol. Chem. 2009, 7, 2013; (b) A.
Ghadban, L. Albertin, Polymers 2013, 5, 431.
21 A. Bruneau, M. Roche, A. Hamze, J.-D. Brion, M. Alami, S.
Messaoudi, Chem. Eur. J. 2015, 21, 8375.
4 J. J. Lundquist, E. J. Toone, Chem. Rev. 2002, 102, 555.
22 Y. Gou, J. Geng, S.-J. Richards, J. Burns, C. Remzi Becer, D.
M. Haddleton, J. Polym. Sci. Polym. Chem. 2013, 51, 2588.
5 (a) K. Kobayashi, H. Sumitomo, Y. Ina, Polym. J. 1985, 17,
567; (b) A. Kobayashi, T. Akaike, K. Kobayashi, H. Sumitomo,
Macromol. Rapid Commun. 1986, 7, 645.
23 H. Zhou, Y. Chen, C. M. Plummer, H. Huang, Y. Chen,
Polym. Chem. 2017, 8, 2189.
6 S. Loykulnant, A. Hirao, Macromolecules. 2000, 33, 4757.
24 (a) Y. Wang, G. Jiang, T. Qiu, F. Ding, Drug Dev. Ind. Pharm.
2012, 38, 1039; (b) C. S. Cho, S. J. Seo, I. K. Park, S. H. Kim, T.
H. Kim, T. Hoshiba, I. Harada, T. Akaike, Biomaterials 2006, 27,
576.
7 M.-P. Labeau, H. Cramail, A. Deffieux, Macromol. Chem.
Phys. 1998, 199, 335.
8 (a) D. M. Haddleton, R. Edmonds, A. M. Heming, E. J. Kelly,
D. Kukulj, New J. Chem. 1999, 23, 477; (b) K. Ohno, Y. Tsujii, T.
Fukuda, J. Polym. Sci. Part A: Polym. Chem. 1998, 36, 2473.
25 (a) S. Kawakami, A. Sato, M. Nishikawa, F. Yamashita, M.
Hashida, Gene Ther. 2000, 7, 292; (b) H. Kitano, Y. Takahashi,
K. Mizukami, K. Matsuura, Colloids Surf. B 2009, 70, 91.
9 K. Aoi, K. Tsutsumiuchi, M. Okada, Macromolecules. 1994,
27, 875.
26 A. Lidgren, A. Bergh, K. Grankvist, T. Rasmuson, B.
Ljungberg, BJU Int. 2008, 101, 480.
10 V. Ladmiral, E. Melia, D. M. Haddleton, Eur. Polym. J. 2004,
40, 431.
27 M. Younes, R. W. Brown, M. Stephenson, M. Gondo, P. T.
Cagle, Cancer. 1997, 80, 1046.
11 J. A. Burns, M. I. Gibson, C. R. Becer, Functional Polymers
by Post-Polymerization Modification; Wiley-VCH Verlag GmbH
& Co. KGaA: Weinheim, 2012; pp. 237.
28 E. L. Dane, S. L. Chin, M. W. Grinstaff, ACS Macro Lett.
2013, 2, 887.
ꢀ
ꢀ
12 M. C. Garcıa-Oteiza, M. Sanchez-Chaves, F. Arranz, Macro-
mol. Chem. Phys. 1997, 198, 2237.
29 J. R. Rich, A. Szpacenko, M. M. Palcic, D. R. Bundle, Angew.
Chem. Int. Ed. 2004, 43, 613.
13 C. R. Becer, K. Babiuch, D. Pilz, S. Hornig, T. Heinze, M.
Gottschaldt, U. S. Schubert, Macromolecules. 2009, 42, 2387.
30 S. M. Dimick, S. C. Powell, S. A. McMahon, D. N. Moothoo, J.
H. Naismith, E. J. Toone, J. Am. Chem. Soc. 1999, 121, 10286.
14 V. Ladmiral, G. Mantovani, G. J. Clarkson, S. Cauet, J. L.
Irwin, D. M. Haddleton, J. Am. Chem. Soc. 2006, 128, 4823.
31 K. H. Mortell, R. V. Weatherman, L. L. Kiessling, J. Am.
Chem. Soc. 1996, 118, 2297.
15 G. Chen, S. Amajjahe, M. H. Stenzel, Chem. Commun. 2009,
1198.
32 A. M. Puertas, F. J. de las Nieves, J. Phys.: Condens. Matter.
1997, 9, 3313.
16 G. David, C. Boyer, J. Tonnar, B. Ameduri, P. Lacroix-
Desmazes, B. Boutevin, Chem. Rev. 2006, 106, 3936.
33 C. W. Cairo, J. E. Gestwicki, M. Kanai, L. L. Kiessling, J. Am.
Chem. Soc. 2002, 124, 1615.
17 Y. Chen, G. Chen, M. H. Stenzel, Macromolecules. 2010, 43,
8109.
34 P. N. Kanellopoulos, K. Pavlou, A. Perrakis, B. Agianian, C.
E. Vorgias, C. Mavrommatis, M. Soufi, P. A. Tucker, S. J.
Hamodrakas, J. Struct. Biol. 1996, 116, 345.
18 J. C. Kim, Y. Rho, G. Kim, M. Kim, H. Kim, I. J. Kim, J. R.
Kim, M. Ree, Polym. Chem. 2013, 4, 2260.
35 C. Xiao, C. Zhao, P. He, Z. Tang, X. Chen, X. Jing, Macro-
mol. Rapid Commun. 2010, 31, 991.
19 T. R. Barlow, J. C. Brendel, S. Perrier, Macromolecules.
2016, 49, 6203.
10
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000