Tunable, Functional Diblock Copolypeptide Hydrogels Based on Methionine Homologs

Article abstract of DOI:10.1002/mabi.201900243

The preparation of new diblock copolypeptide hydrogels derived from homologs of l-methionine, that is, l-homomethionine and l-6-(methylthio)-l-norleucine is described. Compared to l-methionine residues, use of l-methionine homologs allow improved copolymerization with l-leucine residues to give well-defined block copolypeptides. These copolypeptides are subsequently modified using robust thioether alkylation reactions employing a variety of functional epoxides, which yield samples capable of forming transparent, self-healing hydrogels in water. The facile variation of different functional epoxides for postpolymerization modification is found to allow predictable functionalization and tuning of hydrogel properties by the modification of simple precursors.

Full text of DOI:10.1002/mabi.201900243

Full PaPer
Hydrogels
www.mbs-journal.de
Tunable, Functional Diblock Copolypeptide Hydrogels
Based on Methionine Homologs
Graciela E. Negri, Eric G. Gharakhanian, and Timothy J. Deming*
a means to readily introduce different
The preparation of new diblock copolypeptide hydrogels derived from
homologs of l-methionine, that is, l-homomethionine and l-6-(methylthio)-
l-norleucine is described. Compared to l-methionine residues, use of
l-methionine homologs allow improved copolymerization with l-leucine
residues to give well-defined block copolypeptides. These copolypeptides are
subsequently modified using robust thioether alkylation reactions employing
a variety of functional epoxides, which yield samples capable of forming
transparent, self-healing hydrogels in water. The facile variation of different
functional epoxides for postpolymerization modification is found to allow pre-
dictable functionalization and tuning of hydrogel properties by the modifica-
tion of simple precursors.
chemical functionalities into DCH using
diblock copolypeptides based on a small
set of amino acid monomers.
The objective of this study was the
preparation of diblock copolypeptides that
utilize homologs of l-methionine as pre-
cursors to a variety of different functional
[
8]
DCH (Scheme 1). l-Methionine residues
are emerging as versatile components of
biomimetic polypeptide materials since
they can be readily modified in different
ways with both high selectivity and high
[
9,10]
conversion.
These modifications also
mimic natural l-methionine biochemical
processes, that is, oxidation and alkyla-
[
10]
tion reactions.
nine sulfoxide residues in proteins are a natural product of
For example, l-methio-
1
. Introduction
[
11]
oxidative stress within cells and tissues,
and hydrophobic
There is a need for mechanically tunable, degradable hydrogels
that contain diverse functionality for applications as carriers for
drug delivery and as scaffolds for tissue repair.[1–3] Our group
has developed a family of diblock copolypeptide hydrogels
poly(l-methionine), M, can likewise be oxidized to give non-
ionic, hydrophilic, and minimally toxic poly(l-methionine sul-
O [12]
foxide), M .
Alkylation of l-methionine residues using a
variety of reagents also mimics the natural products S-methyl-
[4]
[13,14]
(
DCH) to address these needs, and have reported formula-
l-methionine and S-adenosyl-l-methionine,
and provides
tions that can be charged, non-ionic,[6] or thermoresponsive
[
5]
[7]
a means to convert hydrophobic l-methionine to hydrophilic
R
with tunable mechanical properties. Hydrophobically assem-
bled DCH contain long, hydrophilic segments with disordered
conformations, and shorter hydrophobic segments with either
α-helical or β-sheet ordered conformations, for example, poly(l-
poly(S-alkyl-l-methionine sulfonium), M , functionalized
[
10]
polypeptides.
Our group has developed and utilized these
methods to prepare l-methionine-based functional copolypep-
[
15]
[6]
tide vesicles and hydrogels.
lysine-HCl)180-block-poly(l-leucine) , K180L20, where the confor-
The preparation of l-methionine-based DCH was recently
reported, where l-methionine residues were either oxidized
to non-ionic sulfoxides, or alkylated to give methyl sulfonium
20
mation directed assembly of ordered hydrophobic segments in
water forms tape-like assemblies that branch and entangle to
give 3D hydrogel networks.[
4,5]
However, the functionality of
derivatives. One challenge in the development of these mate-
[6]
most DCH formulations are based on the specific amino acid
monomers used in their preparation, for example, lysine or glu-
tamic acid,[ requiring use of different monomers to prepare
DCH with different functionality. It would be useful to have
rials was copolypeptide synthesis, where the long M segments
required for hydrogel formation were found to aggregate during
polymerization, which in turn compromised the addition of
the hydrophobic poly(l-leucine), L, blocks. Consequently, the
5]
O
R
resulting modified copolypeptides, M 170L23 and M 170L24 were
[
6]
unable to form good hydrogels. Our solution was to incor-
porate ≈10 mol% l-alanine residues during methionine poly-
merization, giving poly(l-methionine-stat-l-alanine) segments
that did not aggregate, thus facilitating copolypeptide synthesis,
Dr. G. E. Negri, Dr. E. G. Gharakhanian, Prof. T. J. Deming
Department of Chemistry and Biochemistry
University of California, Los Angeles
6
07 Charles E Young Dr. E, Los Angeles, CA 90095–1600, USA
O
R
E-mail: demingt@seas.ucla.edu
for example, (M A)170L24 and (M A)170L23, and enabling DCH
formation.
[
6]
Prof. T. J. Deming
Department of Bioengineering
University of California, Los Angeles
Here, we report an alternative strategy utilizing l-methio-
nine homologs instead of l-methionine for preparation of
DCH. The advantage of these homologs, l-homomethionine
and l-6-(methylthio)-l-norleucine (Scheme 1), is that they
can be used to prepare long homopolypeptide segments
5
121 Engineering 5, Los Angeles, CA 90095–1600, USA
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/mabi.201900243.
[
8]
DOI: 10.1002/mabi.201900243
without aggregation, eliminating the need to incorporate a
Macromol. Biosci. 2019, 1900243
1900243 (1 of 7)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
2
.2. Experimental Procedures
.2.1. Representative Synthesis of Hmt180Lx Block Copolypeptides:
2
Hmt180L30
In a dinitrogen-filled glovebox, Hmt NCA (400 mg, 2.1 mmol)
−
1
was dissolved in anhydrous THF (25 mg mL ). To the stirred
solution was added (PMe ) Co (0.80 mL of a 50 mm solution in
3
4
Scheme 1. Structures of homologous thioether containing polypep-
tides: poly(l-methionine), M; poly(l-homomethionine), Hmt; and poly(6-
[
16]
THF) via syringe. After 2 h, an aliquot (20 µL) was removed
for FTIR analysis that confirmed all of the NCA had been
consumed. Another aliquot (0.20 mL) of the polymerization
solution was removed and reacted with mPEG -NCO (70 µL,
(methylthio)-l-norleucine), Mtn.
comonomer such as l-alanine. We also wanted to evaluate if
these homologs provide a means to adjust DCH properties via
their differences in side-chain length. These homologs pos-
sess the same thioether functionality found in l-methionine,
and so it was expected they could be modified using the
same alkylation or oxidation methods.[ To form DCH, the
thioether groups in l-homomethionine and l-6-(methylthio)-
l-norleucine copolypeptides were alkylated using a variety
of epoxides, giving the corresponding sulfonium deriva-
4
4
−
1
3.0 equiv. per Co, 40 mg mL in THF) to yield end-capped
1
[17]
polymer for molecular weight analysis by H NMR. From the
remaining polymerization reaction mixture, a sample (17 mL,
2.1 mmol Hmt residues) was removed, and to this was added
10]
−1
l-leucine NCA (Leu NCA; 1.1 mL, 0.35 mmol, 50 mg mL in
THF). After complete consumption of the second NCA was
confirmed by FTIR, the reaction mixture was removed from the
glove box and the copolymer was precipitated into H O (10 mL),
2
[
9]
tives bearing different functional groups. We have studied
the ability of these copolypeptides to form hydrogels, and
the effects of side-chain length and epoxide functionality on
hydrogel properties. These results provide a starting platform
for the development of self-healing, tunable copolypeptide
hydrogels prepared from simple precursors that can possess a
range of different functionality.
washed three times with H O (15 mL), and then isolated by lyo-
2
philization to obtain the product copolypeptide (310 mg, 89%
1
yield). H NMR (400 MHz, trifluoroacetic acid [TFA-d], 25 °C):
δ 4.76–4.63 (m, 1H), 2.73–2.54 (m, 2H), 2.26–2.10 (m, 3H),
2.08–1.86 (m, 2H), 1.85–1.56 (m, 5H), 1.03–0.86 (m, 6H).
2
.2.2. Representative Synthesis of Mtn180Lx Block Copolypeptides:
Mtn180L30
2
. Experimental Section
The procedure followed was identical to the synthesis of
Hmt180Lx copolymers, except that for the initial segment an
equivalent amount of Mtn NCA was used in place of Hmt
NCA. H NMR (400 MHz, TFA-d, 25 °C): δ 4.73–4.55 (m, 1H),
2.71–2.51 (m, 2H), 2.25–2.06 (m, 3H), 1.99–1.77 (m, 2H), 1.76–
1.61 (m, 5H), 1.60–1.41 (m, 2H), 1.02–0.84 (m, 6H).
2
.1. Materials and Methods
1
Ambient temperature reactions were performed at ≈22 °C.
Hexanes and tetrahydrofuran (THF) were degassed by sparging
with N , then dried by passing through alumina columns. All
2
other solvents were used as received from Fisher Scientific.
Dialysis was performed using deionized water (18.2 MΩ-cm)
prepared by passing in-house deionized water through a Milli-
pore Milli-Q Biocel A10 unit, and regenerated cellulose dialysis
tubing was obtained from Spectrum Labs (Rancho Dominguez,
CA). For other experiments, in-house deionized water was
used. NMR spectra were recorded on a Bruker AV400 instru-
ment with chemical shifts reported relative to the solvent
signal. Infrared spectra of solution samples were collected
using a Perkin Elmer 1605 FTIR Spectrophotometer. Tandem
gel permeation chromatography/light scattering (GPC/LS) was
performed on an SSI Accuflow Series III liquid chromatograph
pump equipped with Wyatt DAWN EOS LS and Optilab rEX
refractive index (RI) detectors. Separations were achieved using
2.2.3. Diblock Copolypeptide Alkylation
The thioether groups in diblock copolypeptides were alkylated
[
9]
with epoxides using previously reported procedures. Alkyla-
tions typically gave greater than 98% functionalization of
thioether groups, and isolated yields above 95%.
2.2.4. Alkylation of Hmt180L30 Using Epoxide Mixtures
For mixed alkylations, the ratio of alkylating agents, glycidyl
n-butyl ether, nBu-epoxide, glycidyl azide, N -epoxide, and
3
xStream H O 500 Å Jordi Labs 5 µm mixed bed column with
2-(2,5,8,11-tetraoxadodecyl)oxirane, EG -epoxide, were varied to
2
3
0
3
.5 wt% potassium trifluoroacetate in HFiP as the eluent at
determine the relative rate of alkylation due to steric hindrance
differences. In general, alkylating agents were combined in
a vial (0.61 mmol, 10 equiv. per each Hmt residue). To the
epoxide mixture was added Hmt180L30 (10 mg, 0.061 mmol of
Hmt residues) and acetic acid (AcOH; 0.40 mL). The reaction
was stirred for 72 h at 37 °C, then transferred to 2000 molecular
weight cutoff (MWCO) dialysis tubing. The sample was dia-
0 °C. All GPC/LS samples were prepared at concentrations
−
1
[16]
of 10 mg mL . Amino acid N-carboxyanhydrides (NCAs)
including l-homomethionine-N-carboxyanhydride (Hmt NCA)
and
6-(methylthio)-l-norleucine-N-carboxyanhydride
(Mtn
NCA),[ copolypeptide K180L , PEG-NCO,
8]
[5]
[17]
and non-com-
were prepared using
30
[9,18]
mercial epoxides (EG , EG and N )
1
3
3
previously described procedures.
lyzed against 3 mm HCl (aq) for 48 h, then H O for 24 h with
2
Macromol. Biosci. 2019, 1900243
1900243 (2 of 7)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
dialysate changes three times daily. The product copolypeptide
was isolated as a white solid by lyophilization. The mol% EG3
was determined by comparing the unobscured 2H peak of nBu
2.2.8. Hydrogel Formulation
The desired amount of lyophilized copolypeptide was weighed
out in a 1/2-dram glass vial. To this solid was added Milli-Q
H O, 25 mm NaCl (aq) or 1x PBS (aq) to achieve the desired
2
(
1.65 ppm) to the 3H peak of Hmt (3.06 ppm) and subtracting
1
the resulting mol% nBu from 100. H NMR (400 MHz, TFA-d,
25 °C): δ 4.89–4.65 (m, 1H), 4.64–4.48 (m, 1H), 4.02–3.29 (m,
29H), 3.19–2.94 (m, 3H), 2.40–1.92 (m, 4H), 1.61–1.54 (m, 2H),
1.44–1.31 (m, 4H), 1.01–0.83 (m, 9H).
final copolypeptide concentration, and the sample was allowed
to form a hydrogel. All further studies were performed on
hydrogels prepared at least 16 h in advance of analysis.
2
.2.5. Synthesis of 2,5,8,11-Tetraoxatetradec-13-yne (EG -alkyne)
2.2.9. Rheology Measurements
3
A solution of triethylene glycol monomethyl ether (1.0 mL,
.1 mmol) in anhydrous (anh.) THF (30 mL) was prepared and
Dynamic rheology analyses were conducted on an Anton Paar
Physica MCR 301 rheometer using an 8 mm diameter parallel
plate geometry. Oscillatory strain amplitude sweeps (0.0–5.0)
6
stirred under N . To this solution was added NaH (60% in min-
2
−
1
eral oil, 370 mg, 9.1 mmol) and NaI (45 mg, 0.30 mmol), and
then propargyl bromide (80% in toluene, 1.0 mL, 9.1 mmol).
The mixture was stirred for 16 h at room temperature, then con-
centrated. The crude product was dispersed in water (60 mL)
and extracted into ethyl acetate (EtOAc; 3 × 40 mL), and then
concentrated to give a yellow oil (1.0 g, 85% yield). Spectra were
were conducted at fixed frequency (1 Hz, or 6.3 rad s ) to estab-
lish the linear regime. Next, oscillatory frequency sweeps (ω =
−
1
1.0–100 rad s ) were conducted to measure G′ (storage mod-
ulus) and G″ (loss modulus) versus frequency. Samples were
allowed to rest for several minutes between analyses to allow
sample recovery. Analysis of hydrogel recovery (self-healing)
consisted of two alternating conditions: large amplitude oscil-
lation to break down the gel structure (strain amplitude of 5.0
at 1 Hz) for 250 s, followed by linear small-deformation oscilla-
tion to monitor recovery of mechanical strength (strain ampli-
tude of 0.10 at 1 Hz) for 250 s.
[19]
in accordance with data reported in literature.
2
.2.6. Synthesis of 5(6)-Carboxamide-N-(propargyl)-fluorescein
5
(6)-Carboxyfluorescein (0.25 g, 0.66 mmol) was dissolved in
anh. N,N-dimethylformamide (10 mL). To the solution was
added anh. N,N-diisopropylethylamine (0.58 mL, 3.3 mmol)
followed by propargylamine (0.13 mL, 0.11 mmol). Next,
N,N,N′,N′-tetramethyl-O-(1H-benzotriazol-1-yl)uronium hex-
afluorophosphate (0.38 g, 1.0 mmol) was added in one portion.
The foil-wrapped reaction was stirred for 16 h, then concen-
trated under vacuum to give a thick oil. The crude product was
purified using column chromatography using 73:2:25% (EtOAc:
methanol:hexanes). The recovered solute was triturated three
2.2.10. Laser Scanning Confocal Microscopy of Fluorescently
Labeled Hydrogel
The laser scanning confocal microscopy hydrogel image
(3.0 wt% in DI water) was taken on a Leica TCS-SP1 MP-
inverted confocal and multiphoton microscope equipped with
an argon laser (476 and 488 nm blue lines), a diode (DPSS) laser
(561 nm yellow-green line). The fluorescein labeled hydrogel
sample was visualized on a glass slide with a spacer between
the slide and the cover slip (double-sided tape) allowing the
hydrogel to be minimally disturbed during focusing. A Z-slice
thickness of 0.78 µm was used. Sample imaging was performed
at the Advanced Light Microscopy/Spectroscopy Center (ALMS)
at the UCLA California NanoSystems Institute (CNSI).
times with H O and lyophilized to give a yellow solid (210 mg,
2
7
7% yield). Spectra were in accordance with that reported in
[20]
literature.
2
.2.7. Procedure for CuAAC Conjugations
Example procedure for conjugation of alkyne functionalized fluo-
rescein probes to hydrogels. In a foil-wrapped vial, (R-Hmt)180L30
3
. Results and Discussion
(
R = N :EG in a 5:95 ratio (prepared using mixed epoxide
3 3
alkylation procedure), 10 mg, 0.025 mmol), 5(6)-carboxamide-
N-(propargyl)-fluorescein (0.60 mg, 0.00014 mmol), and sodium
ascorbate (2.6 mg, 0.013 mmol) were combined in 0.50 mL
H O. In a separate vial, CuSO ·5H O (0.70 mg, 0.0026 mmol)
The preparation of l-homomethionine and l-6-(methylthio)-
l-norleucine-based diblock copolypeptides utilized the cor-
responding N-carboxyanhydride (NCA) monomers of these
amino acids, Hmt NCA and Mtn NCA, respectively (Scheme 2),
2
4
2
[
8]
and N,N,N′,N″,N″-pentamethyldiethylenetriamine (1.1 µL) were
that were recently reported. Hmt NCA and Mtn NCA were
separately polymerized using Co(PMe3)4 initiator in THF to
give homopolypeptide segments of ≈180 residues in length
(Scheme 2; Table S1 and Figures S1 and S2, Supporting Infor-
dissolved in 0.25 mL H O. Solutions were stirred under N2
2
(g) for 1.0 h before combining in foil-wrapped vial. The reac-
tion was stirred under N (g) for 24 h, then transferred to 2000
2
[
16]
MWCO tubing. The product was dialyzed against 0.50 mm diso-
dium EDTA (aq) for 48 h, 3 mm HCl (aq) for 48 h, and then
water for 24 h with dialysate changed four times daily. Product
was isolated by lyophilization (11 mg, 98% yield, yellow-orange
solid).
mation). To each Hmt or Mtn segment was added Leu NCA
in different amounts to give diblock copolypeptides with L
segments ≈20 or 30 residues in length. These parent copoly-
peptides were obtained in high overall yields with low disper-
sities, and compositions that closely matched expected values
Macromol. Biosci. 2019, 1900243
1900243 (3 of 7)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
Scheme 2. Synthesis of hydrogel forming diblock copolypeptides.
(Table S1, Supporting Information). These fully hydrophobic
copolypeptides there then separately reacted with one of six
[
9]
different functional epoxides using established methods to
give the corresponding amphiphilic Hmt or Mtn sulfonium
containing samples (R-Hmt)180Ln and (R-Mtn)180Ln, respec-
tively (Scheme 2). Using excess epoxide (≈3–10 equivalents per
thioether group), the conversion of thioether groups to S-alkyl
sulfonium groups was found to be greater than 98%, quanti-
1
fied using H NMR spectroscopy (see Supporting Information).
All sulfonium containing block copolypeptide derivatives were
converted to chloride ion salts by dialysis against aqueous HCl,
followed by purification by dialysis against DI water. Samples
were isolated as fluffy white solids after lyophilization.
Samples of (R-Hmt)180L20 and (R-Hmt)180L30, where R = Me,
EG , EG , nBu, tBu, and Bn, were separately dissolved in DI
1
3
water at 3.0 wt% of copolypeptide, and all were found to form
self-supporting, transparent hydrogels, which were expected to
possess branched, tape-like fibrillar networks as observed for all
Figure 2. Rheology data for 3.0 wt% (R-Hmt)180Ln, n = 20 or 30, hydrogels
in DI water for different side-chain substituents, R. A) Storage modulus,
G′ (Pa), of (R-Hmt)180L20 (black) and (R-Hmt)180L30 (gray). B) Loss mod-
ulus, G″ (Pa), of (R-Hmt)180L20 (white) and (R-Hmt)180L30 (striped). All G′
and G″ values are measured at a frequency of 1 Hz and strain amplitude
of 0.10, and are averages of triplicate runs with bars indicating standard
deviations.
other DCH (Figure 1).[4,21] Rheological analysis of these sam-
ples revealed that both L segment length and side-chain func-
tionality, R, affected hydrogel mechanical properties (Figure 2).
All samples with longer L30 segments were found to give hydro-
gels with greater stiffness (i.e., increased storage modulus
G′) compared to the corresponding L20 [samples, as expected
5]
from prior studies on amphiphilic DCH. Samples with more
stiffer hydrogels compared to samples containing more hydro-
phobic side-chain groups (i.e., R = nBu, tBu, Bn). It was some-
what surprising that the R-Hmt segments containing these
hydrophobic side-chains were hydrophilic enough to promote
hydrogel formation at all. It appears the presence of charged
sulfonium ions and adjacent hydroxyl groups in the side-chains
derived from epoxide alkylation is sufficient to overcome the
hydrophobic influences of these R groups, albeit at the cost of
weaker hydrogel formation. Due to superior hydrogel forma-
tion with the (R-Hmt)180L30 samples, subsequent studies were
focused on samples with longer L30 segments.
hydrophilic side-chain groups (i.e., R = Me, EG , EG ) formed
1
3
To compare samples containing different side-chain lengths,
the mechanical properties of (R-Hmt)180L30 samples were com-
pared with the corresponding series of (R-Mtn)180L30 samples at
3
.0 wt% in DI water. All (R-Mtn)180L30 samples were found to
form transparent hydrogels, and rheological analysis revealed
that (R-Mtn)180L30 hydrogels possessed similar stiffness to cor-
responding (R-Hmt)180L30 hydrogels, but were overall weaker
especially for samples with more hydrophilic side-chains
Figure 1. Schematic showing proposed self-assembly of amphiphilic
R-Hmt)180Ln and (R-Mtn)180Ln in water into fibrils that branch and
entangle to form hydrogel networks.
(
Figure 3). The trends in G′ as functions of different R groups
(
were found to be similar between both sets of hydrogels. The
Macromol. Biosci. 2019, 1900243
1900243 (4 of 7)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
Figure 4. Rheology data for different concentrations (wt%) of (R-MH)180L30
hydrogels in DI water, R = nBu or EG . A) Storage modulus, G′ (Pa), when
3
R = nBu (black) and R = EG (gray). B) Loss modulus, G″ (Pa), when
3
R = nBu (white) and R = EG (striped). All G′ and G″ values are measured
3
Figure 3. Rheology data for 3.0 wt% (R-Hmt)180L30 and (R-Mtn)180L30
hydrogels in DI water for different side-chain substituents, R. A) Storage
modulus, G′ (Pa), of (R-Hmt)180L30 (black) and (R-Mtn)180L30 (gray).
B) Loss modulus, G″ (Pa), of (R-Hmt)180L30 (white) and (R-Mtn)180L30
at a frequency of 1 Hz and strain amplitude of 0.10, and are averages of
triplicate runs with bars indicating standard deviations.
(
striped). All G′ and G″ values measured at a frequency of 1 Hz and strain
chains. To illustrate the potential of this strategy, samples of
Hmt180L30 were alkylated with mixtures containing varying
amplitude of 0.10, and are averages of triplicate runs with bars indicating
standard deviations.
ratios of EG -epoxide and nBu-epoxide, yielding fully modi-
3
fied (R-Hmt)180L30 samples with EG content varying from 29
3
increased hydrophobic side-chain length in (R-Mtn)180L30
versus (R-Hmt)180L30 appears to have a modest weakening
effect on hydrogel stiffness, especially for the samples where
to 90 mol% (Table S2, Supporting Information). Rheology data
for these samples at 3.0 wt% in DI water, including pure (EG3-
Hmt)180L30 and (nBu-Hmt)180L30 for comparison (Figure S3,
Supporting Information), show that even small percentages
of nBu content can be used to finely tune hydrogel properties.
Beyond side-chain modification, copolypeptide concentration
was also found to be a convenient means to independently
adjust hydrogel stiffness (Figure 4). Similar to amphiphilic
R = Me and EG . This side-chain length effect is also seen by
3
M
comparison to previously reported DCH, (M A) 70L27, con-
1
taining methylated methionine residues.[ At 3.0 wt% in DI
6]
M
water, G′ for (Me-Mtn)180L30, (Me-Hmt)180L , and (M A)170L27
30
[
6]
hydrogels were 110, 195, and 230 Pa, respectively. While these
DCH possess similar mechanical properties overall, stiffness
was found to decrease with increasing side-chain lengths. For
all samples studied, those with the greatest EG content, R =
[
5]
DCH, increased sample concentrations in DI water resulted
in increased hydrogel stiffness for all samples studied. Both
(EG -Hmt)180L30 and (nBu-Hmt)180L30 hydrogels remained
3
EG , always gave the stiffest hydrogels at equivalent copolypep-
transparent at all concentrations.
3
tide concentration, likely due to the high water solubility of the
To study the effect of solution ionic strength on hydrogel
properties, (EG -Hmt) L , (EG -Mtn)180L30, and a K180L30
[
22]
EG groups that help solubilize the polypeptide chains. How-
3
3
180 30
5]
3
[
ever, variation of the R groups, even to substituents with much
less hydrophilicity, also resulted in hydrogel formation and pro-
vides a convenient means to adjust hydrogel stiffness.
The versatility of thioether alkylation methodology is not lim-
ited to uniform modification of polypeptides, but also allows
statistical incorporation of different R groups on individual
(K = poly(l-lysine·HCl)) sample for comparison were each
dissolved at 3.0 wt% in either DI water, 25 mm NaCl (aqueous)
or 1x PBS buffer (aqueous, ≈150 mm ionic strength). The
K180L30 sample forms a stiffer hydrogel compared to the
(EG -Hmt)180L30 and (EG -Mtn)180L30 samples, and similar
3
3
to earlier findings was found to weaken significantly in the
Macromol. Biosci. 2019, 1900243
1900243 (5 of 7)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
Table 1. Average storage modulus, G′, for K 80L30, (R-Hmt)180L30, and
thinning properties of the hydrogel allow it to liquefy during
injection and then quickly recover elastic network properties
and form a hydrogel deposit upon exiting the needle.
To illustrate the ability to incorporate chemically reac-
tive functionality into the (R-Hmt)180L30 hydrogels, glycidyl
1
(R-Mth)
L hydrogels in different media.
180 30
[
6,23]
Hydrogel stiffness, G′ [Pa]a)
Solvent
K180L30
1423
797
(R-Hmt)180L30
(R-Mtn)180L30
azide, N -epoxide, was also used for thioether alkylation to
DI water
321
115
94
222
71
3
highlight the potential for secondary functionalization of
hydrogels. Azide groups are well known to undergo copper-
catalyzed azide-alkyne cycloaddition (CuAAC) reactions with
alkynes to selectively install functional modifications such as
targeting ligands and fluorescent labels under mild condi-
2
5 mm NaCl
pH 7.4 PBS
623
47
a)All hydrogels were prepared at 3.0 wt% in DI water, 25 mm NaCl, or pH 7.4 PBS.
R = EG3. All G′ values were measured at a frequency of 1 Hz and strain amplitude
[
24]
of 0.10, and are averages of triplicate runs.
tions in aqueous media.
To illustrate the potential of this
strategy, a sample of Hmt180L30 was alkylated with a mix-
ture of EG -epoxide and N -epoxide, to yield a fully modified
presence of added ions,[4] with G′ decreasing more than 50%
in 1x PBS (Table 1; Figure S4, Supporting Information). The
3
3
(R-Hmt)180L30 sample with EG content of 95 mol% and N
3
3
(
EG -Hmt)180L30 and (EG -Mtn)180L30 samples showed similar
content of 5 mol% (see Supporting Information). In a proof
3
3
weakening in increased ionic strength media, yet all sam-
ples remained transparent hydrogels under these condi-
tions. In charged DCH, it is believed that added salt weakens
these hydrogels by both screening like charges on the poly-
electrolyte segments thereby weakening repulsive interactions
that influence self-assembly, and by increasing hydrophobic
attraction leading to decreased hydration of the copolypeptide
assemblies.[
of concept study, EG -alkynes were efficiently grafted onto
the azide groups of the N3 functionalized (R-Hmt)180L30
sample in water. Rheological analysis of this sample before
3
and after CuAAC conjugation with EG -alkyne at 2.5 wt%
3
in DI water showed that there was a small improvement in
hydrogel stiffness after conjugation of the hydrophilic EG3
groups (Figure S5, Supporting Information). Finally, an alkyne
functionalized fluorescein probe was also conjugated to the
4]
Amphiphilic DCH possess self-healing properties that allow
rapid recovery of hydrogel mechanical properties after being
broken down under high shear.[5] To see if the new hydrogels
reported here also possess the ability to self-heal, the mechan-
N functionalized (R-Hmt)180L30 sample using CuAAC which
allowed labeling of the hydrogel for fluorescence imaging
(Figure S6, Supporting Information).
3
ical recovery of a (EG -Hmt)180L30 hydrogel after breakdown
3
under high strain was studied (Figure 5). Exposure of a 3.0 wt%
4
. Conclusions
(
EG -Hmt)180L30 in water to high-amplitude oscillatory strain
3
resulted in complete loss of elastic properties and liquid like
behavior (G″ > G′). Upon removal of high-amplitude strain, the
sample was found to rapidly (i.e., <10 s) recover its elastic prop-
erties without any detectable hysteresis, which is a hallmark
feature of DCH.[ This self-healing behavior is beneficial for
minimally invasive applications requiring delivery of hydrogel
depots via small catheter or syringe needle, where the shear
We have described the preparation of new DCH derived from
homologs of l-methionine, that is, l-homomethionine and
l-6-(methylthio)-l-norleucine, that are self-healing and allow
fine tuning of properties by choice of side-chain length and by
reaction with different functional epoxides. This methodology
was found to eliminate the need for use of comonomers in the
preparation of thioether containing DCH, and allows for the
preparation of a diverse range of functional hydrogels from
simple precursors via high yield postpolymerization thioether
alkylation. These results provide a starting platform for the
development of self-healing, tunable copolypeptide hydro-
gels for potential use as depots for local delivery of bioactive
molecules.
5,6]
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
Acknowledgements
Figure 5. Storage modulus (G′ [Pa], solid line) and loss modulus (G″ [Pa],
This work was supported by the NSF under award No. CHE-1412367.
dotted line) for a 3.0 wt% (EG -Hmt)
L hydrogel as a function of step-
3
180 30
wise large-amplitude oscillatory strain in DI water. For the initial 250 s,
the hydrogel was broken down at a large strain amplitude of 5 at 1 Hz.
This was followed by low strain linear recovery at a small strain amplitude
of 0.10 at 1 Hz for 250 s. This cycle was then repeated to demonstrate
hydrogel self-healing after breakdown.
Conflict of Interest
The authors declare no conflict of interest.
Macromol. Biosci. 2019, 1900243
1900243 (6 of 7)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
[
[
[
11] N. Brot, H. Weissbach, Trends Biochem. Sci. 1982, 7, 137.
12] J. Pitha, L. Szente, J. Greenberg, J. Pharm. Sci. 1983, 72, 665.
13] R. McRorie, G. L. Sutherland, M. S. Lewis, A. D. Barton, M. R. Glazener,
Keywords
block copolymer, hydrogel, methionine, polypeptide
W. Shive, J. Am. Chem. Soc. 1954, 76, 115.
14] A.-W. Struck, M. L. Thompson, L. S. Wong, J. Micklefield, ChemBio-
Chem 2012, 13, 2642.
Received: July 8, 2019
Revised: August 10, 2019
Published online:
[
[
15] A. R. Rodriguez, J. R. Kramer, T. J. Deming, Biomacromolecules 2013,
1
4, 3610.
[
[
16] T. J. Deming, Macromolecules 1999, 32, 4500.
17] K. R. Brzezinska, S. A. Curtin, T. J. Deming, Macromolecules 2002,
35, 2970.
[
[
1] T. R. Hoare, D. S. Kohane, Polymer 2008, 49, 1993.
2] N. A. Peppas, Y. Huang, M. Torres-Lugo, J. H. Ward, J. Zhang,
Annu. Rev. Biomed. Eng. 2000, 2, 9.
[18] J. S. Jungk, J. A. Moore, R. D. Gandour, J. Org. Chem. 1983, 48,
1116.
[
[
[
3] A. Altunbas, D. J. Pochan, Top. Curr. Chem. 2012, 310, 135.
4] T. J. Deming, Soft Matter 2005, 1, 28.
[19] P. J. Jervis, M. Moulis, J-P. Jukes, H. Ghadbane, L. R. Cox,
V. Cerundolo, G. S. Besra, Carbohydr. Res. 2012, 356, 152.
[20] Y. L. Jiang, C. A. McGoldrick, D. Yin, J. Zhao, V. Patel, M. F. Brannon,
J. W. Lightner, K. Krishnan, W. L. Stone, Bioorg. Med. Chem. Lett.
2012, 22, 3632.
[21] D. J. Pochan, L. Pakstis, B. Ozbas, A. P. Nowak, T. J. Deming, Macro-
molecules 2002, 35, 5358.
[22] C. Chen, Z. Wang, Z. Li, Biomacromolecules 2011, 12, 2859.
[23] C. Y. Yang, B. Song, Y. Ao, A. P. Nowak, R. B. Abelowitz, R. A. Korsak,
L. A. Havton, T. J. Deming, M. V. Sofroniew, Biomaterials 2009, 30,
2881.
5] A. P. Nowak, V. Breedveld, L. Pakstis, B. Ozbas, D. J. Pine,
D. J. Pochan, T. J. Deming, Nature 2002, 417, 424.
6] A. L. Wollenberg, T. M. O’Shea, J. H. Kim, A. Czechanski, L. G. Reinholdt,
M. V. Sofroniew, T. J. Deming, Biomaterials 2018, 178, 527.
7] S. Zhang, D. J. Alvarez, M. V. Sofroniew, T. J. Deming, Biomacromol-
ecules 2015, 16, 1331.
[
[
[
[
8] E. G. Gharakhanian, E. Bahrun, T. J. Deming, J. Amer. Chem. Soc.
2019, https://doi.org/10.1021/jacs.9b07223
9] E. G. Gharakhanian, T. J. Deming, Biomacromolecules 2015, 16, 1802.
[
10] T. J. Deming, Bioconjugate Chem. 2017, 28, 691.
[24] J. E. Moses, A. D. Moorhouse, Chem. Soc. Rev. 2007, 36, 1249.
Macromol. Biosci. 2019, 1900243
1900243 (7 of 7)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim