Well-defined homopolypeptides, copolypeptides, and hybrids of poly(l-proline)

Article abstract of DOI:10.1021/bm200495n

l-Proline is the only, out of 20 essential, amino acid that contains a cyclized substituted α-amino group (is formally an imino acid), which restricts its conformational shape. The synthesis of well-defined homo- and copolymers of l-proline has been plagued either by the low purity of the monomer or the inability of most initiating species to polymerize the corresponding N-carboxy anhydride (NCA) because they require a hydrogen on the 3-N position of the five-member ring of the NCA, which is missing. Herein, highly pure l-proline NCA was synthesized by using the Boc-protected, rather than the free amino acid. The protection of the amine group as well as the efficient purification method utilized resulted in the synthesis of highly pure l-proline NCA. The high purity of the monomer and the use of an amino initiator, which does not require the presence of the 3-N hydrogen, led for the first time to well-defined poly(l-proline) (PLP) homopolymers, poly(ethylene oxide)-b-poly(l-proline), and poly(l-proline)-b-poly(ethylene oxide)-b-poly(l-proline) hybrids, along with poly(γ-benzyl-l-glutamate)-b- poly(l-proline) and poly(Boc-l-lysine)-b-poly(l-proline) copolypeptides. The combined characterization (NMR, FTIR, and MS) that results for the l-proline NCA revealed its high purity. In addition, all synthesized polymers exhibit high molecular and compositional homogeneity.

Full text of DOI:10.1021/bm200495n

ARTICLE
pubs.acs.org/Biomac
Well-Defined Homopolypeptides, Copolypeptides, and
Hybrids of Poly(L-proline)
Manos Gkikas,†,‡ Hermis Iatrou,*,† Nikolaos S. Thomaidis,† Paschalis Alexandridis,‡ and
Nikos Hadjichristidis†
†University of Athens, Chemistry Department, Panepistimiopolis, Zografou, 15771, Athens, Greece
‡Department of Chemical and Biological Engineering, University at Buffalo, The State University of New York, Buffalo,
New York 14260-4200, United States
S Supporting Information
b
ABSTRACT: L-Proline is the only, out of 20 essential, amino acid
that contains a cyclized substituted R-amino group (is formally an
imino acid), which restricts its conformational shape. The synth-
esis of well-defined homo- and copolymers of L-proline has been
plagued either by the low purity of the monomer or the inability of
most initiating species to polymerize the corresponding N-car-
boxy anhydride (NCA) because they require a hydrogen on the
3-N position of the five-member ring of the NCA, which is
missing. Herein, highly pure L-proline NCA was synthesized by using the Boc-protected, rather than the free amino acid. The
protection of the amine group as well as the efficient purification method utilized resulted in the synthesis of highly pure L-proline
NCA. The high purity of the monomer and the use of an amino initiator, which does not require the presence of the 3-N hydrogen,
led for the first time to well-defined poly(L-proline) (PLP) homopolymers, poly(ethylene oxide)-b-poly(L-proline), and poly-
(L-proline)-b-poly(ethylene oxide)-b-poly(L-proline) hybrids, along with poly(γ-benzyl-L-glutamate)-b-poly(L-proline) and poly-
(Boc-L-lysine)-b-poly(L-proline) copolypeptides. The combined characterization (NMR, FTIR, and MS) that results for the
L-proline NCA revealed its high purity. In addition, all synthesized polymers exhibit high molecular and compositional homogeneity.
’ INTRODUCTION
low-molecular-weight PLP II is considered to be one of the most
ideal and stiff helical polypeptidic chains with limited defects,
although this has not been verified for well-defined high-molecular-
weight PLPs. The most common route to high-molecular-weight
polypeptide synthesis is via the polymerization of the R-amino
acid N-carboxy anhydrides (NCAs). However, the living polym-
erization of NCAs was plagued, for several decades, by termina-
tion reactions.12 In 1948, Astbury et al. initially remarked on the
difficulty13a involved in the synthesis of poly-L-proline using
NCAs, whereas at about the same time Waley and Watson13b
carried out studies on the analogous N-substituted polysarcosine.
Block and Bamford14 attributed the controversy of various results
of the polymerization of L-proline NCA (LP-NCA) to the high
reactivity of LP-NCA to traces of water, whereas Goodman
et al.15 attributed the extensive racemization between equimolar
quantities of LP-NCA and sodium methoxide to a proton
abstraction from the R-carbon. Berger et al.16 synthesized low-
molecular-weight polypeptides of PLP, whereas the synthesis of
high-molecular-weight PLP was reported17 in the early 60s;
however, the high-molecular-weight nature of the polypeptides
was verified only via viscometry. Komoto et al. connected the two
Most polypeptides of R-amino acids form 3D structures
through hydrogen bonding. Exceptions are the N-substituted
polypeptides (NSP), which do not possess an amide (3-N)
hydrogen, and their secondary structure is due to only the
constraints induced in the main chain. The most characteristic
is poly(L-proline) (PLP). PLP exhibits tertiary amide groups that
lead to significant lowering of the barrier for cisꢀtrans amide
isomerization,1 resulting in the formation of right-handed PLP I
(cis conformation) and left-handed PLP II (trans conformation).2
PLP II is adopted in water and organic acids,3a,b whereas PLP I is
predominant in some organic solvents.3c PLP II is one of the
most interesting homopolypeptides, first discovered in collagen,4
because its conformation is highly extended with 3.1 Å axial
translation5 independent of pH and temperature up to 60 °C,6
rendering it an ideal molecular ruler7 or spacer with defined
dimensions. However, recent studies have shown that individual
PLP residues adopt the PLP I conformation even in water.8 It has
also been observed that the redox potential of the ferrocenyl
groupattachedtotheoligoprolinechainsdepends ontheir length,9
whereas PLP’s rigid backbone has been used as a scaffold-linker for
proteinꢀprotein recognition and cell penetration.10
It is well-documented11 that high-molecular-weight poly-
(γ-benzyl-L-glutamate) (PBLG) forms helices, which are not
considered long ideal tubes and present defects.11 In contrast,
Received: April 12, 2011
Revised:
May 13, 2011
Published: May 13, 2011
r
2011 American Chemical Society
2396
dx.doi.org/10.1021/bm200495n Biomacromolecules 2011, 12, 2396–2406
|

Biomacromolecules
ARTICLE
Scheme 1. Cyclization Reaction for the Synthesis of L-Proline N-Carboxy Anhydride
forms of PLP to the different crystallization behavior of the
growing chains18 during polymerization. Finally, Kricheldorf and
coworkers3c determined the ratio of the two forms obtained
during oligomerization of LP-NCA under different experimental
conditions.
Scheme 2. Ring-Opening Polymerization of L-Proline N-
Carboxy Anhydride
Recently, new initiating species have been developed for the
living polymerization of NCAs. In the pioneer work of Deming
et al.,19a followed later by Peng et al.,19b organometallic initiators
were used, but these initiating systems could not polymerize LP-
NCA because the polymerization mechanism requires the presence
of the 3-N hydrogen. Silazane derivatives were used by Lu et al.,20
but this mechanism also requires the presence of the 3-N hydrogen
and consequently cannot be applied to LP-NCA. The Schlaad,21a
Schuꢀe,21b and Zhang21c groups presented methodologies that
suppressed the termination reactions, allowing for the polymeriza-
tion of LP-NCA. We recently employed primary amines as
initiators along with high vacuum techniques (HVTs) to create
living conditions for the amino-initiated polymerization of NCAs.22
Our approach is general and can also be applied for LP-NCA.
Several methodologies have been presented for the synthesis
of LP-NCA.16ꢀ18 The most common is the reaction of LP with
phosgene or solid triphosgene, for the formation of the corre-
sponding N-carbamoyl chloride (NCC), followed by the cycliza-
tion reaction in the presence of a base. Recently, polystyrene
resins with different tertiary amines were used23 to suppress the
formation of side products. However, the purity of the synthe-
sized LP-NCA was not verified by the synthesis PLPs.
well as the diblock copolypeptides PBLG-b-PLP and poly(ε-tert-
butyloxycarbonyl-L-lysine)-b-poly(L-proline) (PBocLL-b-PLP).
’ MATERIALS AND METHODS
Materials. Boc-Pro-OH (>99%) and triphosgene (99%) were pur-
chased from Acros Organics. H-Pro-OH was purchased from Aldrich
(>99%). Triethylamine (>99%, Acros Organics) was dried over calcium
hydride for 1 day and then distilled and stored in the vacuum line over sodium.
The appropriate quantity needed was freshly distilled in a vacuum line.
Polystyrene resin bearing diethylamino groups (PS-DEAM) was purchased
from Aldrich. Purification of THF (dried, max 0.005% water, Merck) was
performed using standard HVTs reported elsewhere.25 Ethyl acetate
(>99.5%, Merck) was fractionally distilled over phosphorus pentoxide.
Hexane (>99%, Merck) was fractionally distilled over sodium. Acetonitrile
(ACN), the solvent for the synthesis of PLP (>99.5%, Merck), was left to
react with phosphorus pentoxide for 1 day, degassed, and distilled under high
vacuum to a previously flame-dried round-bottomed flask containing molec-
ular sieves. ACN (LC-MS grade), the solvent for mass spectra experiments,
was purchased from Labscan. Anhydrous dimethylamine (>99.9%, Aldrich,
bp 7 °C) was condensed under high vacuum at ꢀ78 °C and was treated with
sodium hydroxide pellets at room temperature for 1 day and was subsequently
distilled into precalibrated ampules with break-seals. It was then diluted with
ACN to the appropriate concentration in a sealed apparatus equipped with
precalibrated ampules and kept away from light. End-functionalized
We present here a novel methodology based on N-(Boc)24
protected rather than neat L-proline, which leads, in high yields,
to the synthesis of highly pure LP-NCA, as verified by NMR,
FTIR, and MS. The high purity of the monomer in combi-
nation with an amino initiator and HVTs led to the synthesis of
well-defined PLP homopolypeptides, the PLP hybrids poly(ethylene
oxide)-b-poly(L-proline) (PEO-b-PLP), and PLP-b-PEO-b-PLP as
2397
dx.doi.org/10.1021/bm200495n |Biomacromolecules 2011, 12, 2396–2406

Biomacromolecules
ARTICLE
Scheme 3. Synthesis of Well-Defined PEO-b-PLP Hybrids
Scheme 4. Synthesis of Well-Defined PEO-b-PLP-b-PEO Hybrids
Scheme 5. Synthesis of Well-Defined PBocLL-b-PLP
monoamino poly(ethylene oxide) having molecular weights (10.0 and 20.0)
ꢁ 103 and R,ω-diamino-functionalized poly(ethylene oxide) having molec-
ular weight 10.0 ꢁ 103 g/mol were purchased from Aldrich and were used as
mono- and difunctional macroinitiators, respectively.
FTIR. FTIR measurements were performed with a Perkin-Elmer
Spectrum One instrument in KBr pellets at room temperature in the
range 450ꢀ4000 cmꢀ1
.
Mass Spectrometry. Mass spectrometry was performed with a
Thermo TSQ Quantum Access equipped with an IonMax electrospray
(ESI) source. Purified L-proline-NCA was dissolved in ACN (LC-MS
grade) and directly infused in the ESI source in positive ionization mode.
Spray voltage ranged between 3000 and 4000 V; sheath gas and auxiliary
gas pressures were maintained at 25 and 15 (a.u.), respectively, whereas
two capillary temperatures were tested (150 and 270 °C). The full scan
spectrum was recorded in a mass range of m/z 50ꢀ200 with a scan time
of 0.5 s. Product ion scans and precursor ion scans were performed to
verify that the observed ions belonged to the same molecule.
Circular Dichroism. Circular dichroism was performed with a
Jasco J-815 model, featuring a peltier model PTC-423S/15 thermo-
stabilizing system. The cell used was 1 mm Quartz Suprasil cell. The
solutions of the PLP and hybrid polypeptides were made with Milli-Q
water. Typical concentrations were ∼3 ꢁ 10ꢀ4 g/mL.
Size Exclusion Chromatography. Size exclusion chromatogra-
phy (SEC) was used to determine the Mn and Mw/Mn values. The
analysis was performed using two SEC sets. The one was composed of a
Waters Breeze instrument equipped with a 2410 differential refract-
ometer and a Precision PD 2020 two angles (15°, 90°) light scattering
detector. The carrier solvent was a 0.1 M NaNO3 solution of water/
ACN (80/20 v/v) at a flow rate of 0.8 mL/min at 35 °C. Three linear
Waters hydrogel columns were used. For PBLG and PBocLL homo-
polymers, the analysis was performed using a second SEC equipment.
The system was composed of a Waters 600 high-pressure liquid chromato-
graphic pump, Waters Ultrastyragel columns (HR-2, HR-4, HR-5E, and
HR-6E), a Waters 410 differential refractometer detector, and a Precision
PD 2020 two angles (15°, 90°) light scattering detector at 60 °C. A 0.1 N
LiBr DMF solution was used as an eluent at a rate of 1 mL/min. For the
water-soluble deprotected copolypetide PLL-b-PLP, the first SEC instru-
ment was used, and the carrier solvent was 5% NaH2PO4 (w/v) solution of
water/ACN (97/3 v/v) at a flow rate of 0.8 mL/min at 35 °C (pH 4).
X-ray Diffraction. X-ray powder diffraction (XRD) was carried
out on a Rigaku Ultima IV diffractometer with a Cu KR radiation
(wavelength 1.54 Å) and was operated with θ/2θ mode. The operat-
ing conditions of the diffractometer were 40 kV, 44 mA, and 1.76 kW,
and data were collected at room temperature. The XRD data were
collected at a 0.030 interval, and the scan speed was 0.5 deg/min. The
technique used for measuring intensities was the focusing beam method
(BB method).
1
NMR Spectroscopy. H and 13C NMR spectroscopy (400 and
75 MHz, respectively) were performed using a Bruker DPX 400 or a
Varian Unity Plus 300/54 spectrometer. The spectra of the polymers
were performed in CF3COOD, whereas the spectra of the NCAs were
taken in CDCl3 at room temperature.
2398
dx.doi.org/10.1021/bm200495n |Biomacromolecules 2011, 12, 2396–2406

Biomacromolecules
ARTICLE
Scheme 6. Synthesis of Well-Defined PBLG-b-PLP
Thermogravimetric Analysis. Thermogravimetric analysis (TGA)
was performed using a TA Instruments TGA Q50 in nitrogen from 23 to
600 °C (10 °C/min).
Synthesis of L-Proline N-Carboxy Anhydride (LP-NCA). tert-
Butyloxycarbonyl-L-proline (Boc-L-proline) (30 g, 0.139 mol) was
dissolved in freshly distilled THF (600 mL) in a 1 L two-necked
round-bottomed flask with a slight flow of dry Ar gas. Then, triphosgene
(0.37 equiv, 15.2 g) was added under vigorous stirring. After 10 min,
freshly distilled triethylamine (1.1 equiv, 21.3 mL) was added dropwise
at 0 °C, leading to the formation of N-carboxy anhydride in a one-pot
procedure with instantaneous precipitation of TEA HCl salt. After
3
stirring for 6.5 h at room temperature under Ar atmosphere, the reaction
mixture was cooled at 0 °C to allow complete precipitation of the salt,
and the precipitate was removed by filtration. The filtrate was then
distilled under vacuum (to remove THF and excess phosgene), and the
crude mixture was dried overnight in the high vacuum line.
Crude LP-NCA was then dissolved in ethyl acetate, chilled, and
extracted26 with ice-cold water (100 mL) until neutral pH. The organic
phase was then separated, dried over MgSO4, and filtered, and the
solvent was distilled off in the vacuum line, yielding LP-NCA (8 g,
yield ∼50%).
Figure 1. FTIR of the crude product (red line) and of the purified LP-
NCA (green line) along with the 1H NMR (inset) spectra of LP-NCA.
Purification of LP-NCA. Crude LP-NCA was dissolved in the mini-
mum amount of THF and a large excess of hexane (400 mL) was added,
leading to a milky, thick precipitate which was cooled to ꢀ20 °C for
complete precipitation. The solvent mixture was then removed under Ar
filtration, using a microfilter candle for reversed filtration with a narrow
tube (Duran, d = 13 mm, por.3), and the precipitate was dried on the
high vacuum line. In this way, 1 (Scheme 1) was removed. The
purification was monitored with FTIR and 1H NMR spectroscopy.
The next step involved thermal dissolution of LP-NCA crystals in
hexane. In this additional purification step, the dried crystals were
suspended in a large amount of hexane (∼ 500 mL) and gradually
heated to 45 °C under vigorous stirring until the anhydride was
thoroughly dissolved. The solution was then filtered immediately with
a warm filter funnel (125 mL capacity, por.3) under vacuum in a new
two-necked 1 L round-bottomed flask, whereas nonsoluble oily con-
taminants were removed. Very long needle-like white LP-NCA crystals
were obtained when the flask was cooled to ꢀ20 °C. Cold hexane was
filtered under Ar using a microfilter candle for reversed filtration, and the
resulting LP-NCA crystals were dried under high vacuum. This proce-
dure resulted to removal of 2 of Scheme 1. The procedure was repeated if
contamination was indicated by FTIR spectroscopy. The purity of the
final product was verified by FTIR and 1H and 13C NMR spectroscopy,
and electrospray ionization mass spectroscopy (ESI-MS), and it was
stored in a glovebox. Overall yield: 6 g (30%) m.p.: 53ꢀ55 °C.
LP-NCA. 1H NMR (400 MHz, CDCl3) δ 4.27 (dd, J = 8.78, 7.84 Hz,
1H), 3.71 (dt, J = 11.4, 7.47, 7.47 Hz, 1H), 3.26 (ddd, J = 11.4, 8.32,
4.92 Hz, 1H), 2.25 (m, 1H), 2.11 (m, 2H), 1.88 (m, 1H). 13C NMR
(75 MHz, CDCl3) δ 169.0, 155.2, 63.3, 46.8, 28.0, 27.2. IR (thin film)
2994, 2958, 1845, 1824, 1772, 1364, 1328, 1270, 1202, 1185, 1150, 1095,
998, 952, 916 cmꢀ1. ESI-MS m/z 70, 114, 130, 141 (capillary tempera-
ture 150 °C, spray voltage 3000ꢀ4000 V).
Synthesis of Polypeptides. All manipulations were performed
under high vacuum in glass reactors and equipped with break seals,
glass-covered magnets, and constrictions for the addition of reagents and
removal of the intermediate products following well-established HVTs.25
Synthesis of Well-Defined Poly(L-proline). The reaction used
for the synthesis of well-defined PLP is given in Scheme 2.
LP-NCA (0.67 g) was pumped for 1 day to dryness in the high vacuum
line in a specially designed glass reactor (equipped with break-seal along
with a glass-covered magnet), which had been previously flame-dried
several times. The monomer was then dissolved in freshly distilled ACN,
and the glass magnet of the initiator ampule was subsequently ruptured,
allowing the solution of dimethylamine (5 ꢁ 10ꢀ5 mol, A/I = 94, for PLP
no. 2) to react with the monomer under vigorous stirring. After a few
minutes, the solution became turbid, and the heterogeneous polymeriza-
tion induced was left for 1 week, occasionally being degassed to remove
the generated CO2. The final polypeptide (mainly PLP I configuration)
2399
dx.doi.org/10.1021/bm200495n |Biomacromolecules 2011, 12, 2396–2406

Biomacromolecules
ARTICLE
Figure 2. (a) Positive electrospray mass spectrum (þESI-MS) spectra of purified LP-NCA. Capillary temperature 150 °C, spray voltage 3000 V. (b)
Fragmentation pattern of LP-NCA.
was precipitated in cold ether and dried in vacuo. The purity of the
To obtain the water-soluble poly-L-proline II (PLPII) form, the
product was verified by SEC-TALLS, FTIR, 1H NMR, and 13C NMR.
polypeptide was suspended in Milli-Q water (2 ꢁ 10ꢀ3 g/mL) for
2400
dx.doi.org/10.1021/bm200495n |Biomacromolecules 2011, 12, 2396–2406

Biomacromolecules
ARTICLE
Scheme 7. Homopolypeptides, Copolypeptides, and Hy-
brids of Poly(L-proline) Synthesized
reactor (equipped with break-seal and glass-covered magnet), which had
been previously flame-dried several times. The monomer was then
dissolved in freshly distilled DMF, and the glass magnet of the initiator
ampule was subsequently ruptured, allowing the solution of dimethyla-
mine (0.95 ꢁ 10ꢀ5 mol, A/I = 462) to react with the monomer under
vigorous stirring for 3 days, with occasional degassing. An aliquot
of the solution was removed after the completion of the polymerization
of ε-tert-butyloxycarbonyl-L-lysine-NCA for characterization. Then,
LP-NCA solution in DMF was cannula transferred (in an Ar/vacuum
line) in the specially designed glass reactor containing the PBocLL
macroinitiator, transferred to the high vacuum line, degassed, and left
1 week under vigorous stirring for the polymerization of the PLP branch.
The polymerization was heterogeneous, and the final copolypeptide was
precipitated in cold ether and dried in vacuo.
To obtain the water-soluble PLL-b-PLP, the polypeptide was sus-
pended in TFA and left to react until complete dissolution. Then, the
organic acid was distilled in the high vacuum line, and the product was
dissolved in Milli-Q water.
PBocLys. GPC (0.1 M LiBr) DMF at 65 °C, Mn = 99.1 K,
Mw/Mn = 1.18.
PLL-b-PLP. 1H NMR (300 MHz, CF3COOD) δ 5.25ꢀ4.35 (1H, CR
of PLP and 1H, Ca of PLL), 4.20ꢀ3.50 (2H, Cδ), 3.20 (2H, Ce),
2.75ꢀ1.1 (4H, Cβþγ of PLP and 6H, Cbþcþd of PLL). GPC (5%
NaH2PO4) water/ACN 97:3 at 35 °C, Mw/Mn = 1.10.
Synthesis of Poly(γ-benzyl-L-glutamate)-b-poly(L-proline)
(PBLG-b-PLP). The reactions used for the synthesis of the PBLG-b-
PLP are given in Scheme 6.
The procedure followed was the same as that in the case of PBocLL-b-
PLP, the only difference being that the LP-NCA solution that was
transferred with the cannula (in the Ar/vacuum line) was in THF and
the polymerization of the PLP branch was nearly homogeneous. The
final product was insoluble in DMF, and no SEC could be obtained.
TGA measurements were carried out for the two samples to determine
qualitatively the composition of the two blocks.
1 day under vigorous stirring to become soluble and centrifuged, and the
water-soluble part was freeze-dried. The conversion was about 60ꢀ80%,
in agreement with previous studies.16
PLP II. 1H NMR (300 MHz, CF3COOD) δ 4.76 (1H, CR), 3.78 (1H,
Cδ2), 3.58 (1H, Cδ1), 2.29 (1H, Cβ2), 1.99 (3H,, Cβ1þγ). 13C NMR
(75 MHz, CF3COOD): δ 175.1, 62.5, 50.9, 30.3, 27.0. IR (thin film) 3450,
2962, 2879, 1640, 1426, 1328, 1259, 1202, 1163, 1092, 1040, 916 cmꢀ1
.
GPC (0.1 M NaNO3) water/ACN 80:20 at 35 °C, Mw/Mn = 1.23.
Synthesis of Poly(ethylene oxide)-b-poly(L-proline) (PEO-
b-PLP). The reaction used for the synthesis of well-defined PEO-b-PLP
is given in Scheme 3.
Amino-end monofunctional PEO having molecular weights of either
10.0 or 20.0 ꢁ 103 g/mol was pumped to dryness in the high vacuum line
in a custom-made 100 mL glass apparatus (equipped with a filter), which
had been previously flame-dried several times. Benzene was then
distilled to remove azeotropically traces of water, the solvent was
distilled off, and the polymer was left to dryness overnight. ACN was
distilled to dissolve the polymer, which was finally filtrated in an ampule
and used as a macroinitiator. The heterogeneous polymerization pro-
cedure is the same as that presented for poly-L-proline in ACN. To
transform the PLP block to its PLP II form, the hybrid polypeptide was
suspended in Milli-Q water for 1 day under vigorous stirring, centri-
fuged, and then was freeze-dried.
PBLG No. 1. GPC (0.1 M LiBr) DMF at 65 °C, Mn = 20 K, Mw/Mn =
1.02.
PBLG-b-PLP No. 1. TGA(rangeoftemperature:23ꢀ600 °C, flow gas: N2)
215ꢀ360 °C for PBLG, 360ꢀ450 °C for PLP.
’ RESULTS AND DISCUSSION
Synthesis of the Monomer. The most critical factor in the
synthesis of well-defined PLP polymers is the purity of the
monomer.
PEO-b-PLP No. 1. 1H NMR (300 MHz, CF3COOD) δ 4.70 (1H, CR),
4.15ꢀ3.15 (4H, ꢀCH2CH2O and 2H, Cδ), 2.24 (1H, Cβ2), 1.93
(3H, Cβ1þγ). GPC (0.1 M NaNO3) water/ACN 80:20 at 35 °C,
Mw/Mn = 1.10.
In an early attempt, LP was reacted with triphosgene (details
of the experimental procedure are given in the Supporting
Information), followed by reaction of the resulting NCC
with an equimolar amount of triethylamine for the cyclization
reaction at room temperature, at 0 °C, or at ꢀ25 °C, according
to literature.16ꢀ18 In all cases, the LP-NCA yield was low
(Supporting Information, Table 1), and the main side products
were crystalline diketopiperazine (DKP), along with NCC,
which is oily and soluble in THF/hexane and can be effectively
removed by recrystallizations. However, after this procedure, the
product was enriched in DKP (Supporting Information, Scheme S1)
because it also crystallizes in hexane. The presence of DKP was
monitored by FTIR at 1654 cmꢀ1 (DKP characteristic peak)
(Supporting Information, Figure F1). The synthetic approach
that uses polystyrene resins bearing diethylamino groups (PS-
DEAM) was also followed (details of the experimental procedure
are given in Supporting Information); however, polymeric side
products were obtained (Supporting Information, Figure F2) at
1640 cmꢀ1, and the LP-NCA produced was not stable over time
Synthesis of Poly(L-proline)-b-poly(ethylene oxide)-b-
poly(L-proline) (PLP-b-PEO-b-PLP). The reaction used for the
synthesis of well-defined PEO-b-PLP-b-PEO is given in Scheme 4.
The synthetic procedure was the same as that in the previous case
except that the R,ω-diamino PEO having molecular weight of 10.0 ꢁ
103 g/mol was used.
1
PLP-b-PEO-b-PLP No. 2. H NMR (300 MHz, CF3COOD) δ 4.73
(2H, CR), 3.85ꢀ3.35 (4H, ꢀCH2CH2O and 4H, Cδ), 2.26 (2H, Cβ2),
1.96 (6H, C Cβ1þγ). GPC (0.1 M NaNO3) water/ACN 80:20 at 35 °C,
Mw/Mn = 1.09.
Synthesis of Poly(ε-tert-butyloxycarbonyl-L-lysine)-b-poly-
(L-proline) (PBocLL-b-PLP). The reactions used for the synthesis
of the PBocLL-b-PLP are given in Scheme 5.
ε-tert-Butyloxycarbonyl-L-lysine-NCA was synthesized according to
previously reported method.27 Boc-L-Lys-NCA (1.2 g) was pumped for
1 day to dryness in the high vacuum line in a specially designed glass
2401
dx.doi.org/10.1021/bm200495n |Biomacromolecules 2011, 12, 2396–2406

Biomacromolecules
ARTICLE
Figure 3. (a) XRD powder pattern of PLP no. 2 after precipitation, showing that is mainly PLP I form. (b) XRD powder pattern of PLP I no. 3 from
acetic acid/PrOH 1:9. (c) XRD powder pattern of PLP II no. 3 from formic acid.
(Supporting Information, Table 1). The above methods, there-
fore, were not appropriate for highly pure LP-NCA.
by the cyclization reaction. After filtration, the resulting LP-
NCA was formed in high yield. Approximately 5% of each
1 and 2 remained unreacted. LP-NCA was also contami-
It is well-documented that DKP formation is not favored if
the amine group is protected.28 Moreover, polymeric side
products can be formed when secondary cyclic unprotected
amines are present in the synthesis of NCA. We hypothesized
that N-(Boc)-protected proline (1, Scheme 1) would block
the formation of side products. The synthesis of LP-NCA was
a one-pot procedure involving the formation of the carbonyl
chlorocarbonate (2, Scheme 1) as an intermediate, followed
nated with traces of the Et3N HCl salt and excess phosgene,
3
which were removed in the aqueous phase by extraction with
chilled ethyl acetate/water mixture. The unreacted species
were removed by recrystallization in THF/hexane (removal of
unreacted 1), followed by dissolving the LP-NCA in warm
hexane (45 °C) and filtration (removal of the side product 2).
LP-NCA purity was monitored by NMR (Supporting Information,
2402
dx.doi.org/10.1021/bm200495n |Biomacromolecules 2011, 12, 2396–2406

Biomacromolecules
ARTICLE
polymerization of highly pure LP-NCA in ACN, using dimethy-
lamine as initiator and HVT. The use of HVT ensures the
effective purification of chemicals and the successful preparation
of well-defined polypeptides. The polymerization yield was
∼100%, the molecular weights close to that expected from
stoichiometry, and the polydispersity indices (I) were surpris-
ingly as low as 1.26 (Supporting Information, Figure F6). It was
possible to obtain PLP homopolymers only up to 13.0 ꢁ 103 g/mol
because of the heterogeneity of the polymerization. In the case of
the copolymers, the synthesis of higher-molecular-weight PLPs
(140.0 ꢁ 103 g/mol) was possible. The molecular characteristics
of the homopolypeptides are shown in Table 2 (Supporting
Information). The PLPs were characterized by GPC, 1H NMR,
13C NMR, and CD in water (Supporting Information, Figures
F6, F7 and F8), whereas comparisons between the two forms
were made in the solid state with IR and XRD.
Figure 4. SEC chromatograms of the difunctional macroinitiator H2N-
PEO-NH2 and the PLP-b-PEO-b-PLP no. 1 in water.
The PLP II form was also identified in water by circular
dichroism, showing a strong negative band29 at 206 nm and a
weak positive band at 228 nm (Supporting Information, Figure
F9). This left-handed helix is a common secondary structure in
natural proteins (gelatin, elastin, casein, collagen).
Figures F3 and F4), FTIR (Figure 1), and ESI-MS (Figure 2)
spectroscopies.
The characteristic signal of the LP-NCA R proton at 4.27
1
ppm is present in the H NMR spectrum (inset of Figure 1),
The two forms of PLP can be identified by FTIR and XRD in
the solid state. By FTIR, the identification of PLP I is confirmed
by the peaks at 960, 1355, and 1155 cmꢀ1, whereas these peaks
are significantly reduced in intensity or completely eliminated in
the PLP II form (Supporting Information, Figure F10). Accord-
ing to XRD data, the PLP II form gives multiple peaks because of
its well-organized macromolecular structure, whereas the PLP I
form with low crystallinity gives only characteristic peaks at 8.06
and 4.62 Å. As it is identified by FTIR and XRD measurements,
PLP is mainly form I after precipitation (Figure 3a). To adopt the
fully cis PLP I, a solvent mixture of acetic acid/PrOH 1:9 was
used30,31 (Figure 3b), whereas for the fully trans PLP II form,
water or organic acids were used (Figure 3c). The transition
between the two forms is cooperative and reversible. Peaks at
5.81, 4.93, 3.68, 3.17, 2.77, and 2.48 Å are characteristic of form
II, confirming previous XRD studies.5b,18,31 Our XRD measure-
ments, however, reveal that even after treatment with formic acid
(and centrifugation), a small fraction of cis-residues is still
present (Figure 3c), as proved by the peak at 8.15 Å. This is in
accordance with previous studies,8 where a small fraction of cis
prolines in PLP II oligoprolines was found. This was predicted in
a theoretical study by Schimmel and Flory,32 who pointed out
that even a small fraction of cis residues should be present in PLP
II, which would lower the end-to-end distance of the highly
organized chain.
whereas that of 2 appears at 4.59 and 4.48 ppm and that of 1
appears at 4.22 and 4.32 ppm. The high purity of the LP-NCA is
confirmed by the lack of peaks corresponding to side products.
The FTIR spectra of the crude product (red line) and the purified
LP-NCA (green line) are provided in Figure 1. The CdO stretch
of the carbonyl chlorocarbonate (2) appears at 1700 cmꢀ1
,
whereas that of 1 appears at 1638 cmꢀ1. The lack of a peak in
the 1740ꢀ1638 cmꢀ1 range verifies the absence of side materials.
Compared with FTIR and NMR spectroscopies, ESI-MS can
detect smaller quantities of impurities. The (þ)ESI-MS of the
purified LP-NCA at a spray voltage of 3000 V produced three
intense peaks at m/z 70, 114, and 130 (fragments of the same
molecule), with the absence of any impurity ions (Figure 2a,b).
The molecular ion (m/z 141), as well as the adducts (þ41) with
ACN of the observed ions (m/z 155 and 171), are also shown in
Figure 2a. The observed ions were sufficiently explained by the
fragmentation pattern proposed in Figure 2b. When the spray
voltage gradually increased, the peak of m/z 130 decreased and
disappeared at 4000 V, with a simultaneous increase in the other
two main ions (m/z 114 and 70). To verify that the three main
ions were part of the same molecule, product ion scans of m/z
114 and 130 were performed at collision energies of 10, 15, and
20 eV. In both cases, the only product was m/z 70. In an additional
experiment, precursor ion scans of product ion m/z 70 revealed two
ions at m/z 114 and 140, verifying that all of these ions are produced
from the molecular ion of L-proline-NCA (m/z 141) (Supporting
Information, Figure F5). This is, to our knowledge, the first report
confirming the high purity of LP-NCA (Supporting Information,
Scheme S2) with three spectroscopic methods.
Synthesis of PEO-b-PLP and PLP-b-PEO-b-PLP Hybrids.
Diblock and triblock water-soluble hybrid polypeptides contain-
ing PLP blocks were also synthesized in ACN using amino-
terminated and R,ω-diamino poly(ethylene oxide) as difunc-
tional macroinitiators. Initially, polymerization was homogeneous
because of the solubility of PEO in ACN, but when the molecular
weight of PLP increased, the polymerization became hetero-
geneous. From the SEC trace of PLP-b-PEO-b-PLP no. 1
(Figure 4), it is obvious that the polydispersity index (1.10) is
similar to that of the R,ω-diamino precursor, indicating that the
higher polydispersity of the PLP homopolypeptides was due to
the heterogeneity of the polymerization. The molecular char-
acteristics of the hybrid polypeptides were those expected from
stoichiometry (Supporting Information, Table 2). The composi-
The high purity of the synthesized LP-NCA has been achieved
by: (a) the use of N-protected proline rather than the unpro-
tected one, which on one hand prevents the formation of DKP
and on the other hand results in Boc-protected byproduct with
the appropriate solubility for easy removal, and (b) the efficient
purification methodology used, which quantitatively removed
the side products.
Synthesis of the Polymers. The synthesized polymers are
shown in Scheme 7.
PLP Homopolymers. A series of water-soluble PLPs with
rather narrow distributions was synthesized by heterogeneous
1
tion was verified by H NMR spectroscopy in CF3COOD
(Supporting Information, Figure F11 and F12). Jeon et al.33
2403
dx.doi.org/10.1021/bm200495n |Biomacromolecules 2011, 12, 2396–2406

Biomacromolecules
ARTICLE
Figure 5. (a) XRD powder pattern of PEO-b-PLP no. 1 directly after the synthesis without any treatment. (b) XRD powder pattern of PLP-b-PEO-
b-PLP no. 1 directly after the synthesis without any treatment.
synthesized a similar triblock; however, they reported that PEO
remained unreacted and no SEC was given. The diblock synthe-
sized here is reported for the first time.
The XRD data reveal two different pictures for the conforma-
tion of diblock and the triblock hybrid polypeptides after
precipitation. In the PEO-b-PLP no. 1, the fully cis PLP I form
is observed and PEO cannot crystallize, implying an interfacial
mixing (Figure 5a). However, after dissolution in formic acid, it
mutarotates to PLP II. The PLP-b-PEO-b-PLP no. 1 adopts
mainly the PLP II form after precipitation, and the crystallinity of
PEO is observable (Figure 5b). Dissolution in formic acid does
not change the spectra. The diblock no. 1 and triblock no. 1 have
different macromolecular architectures as well as different PLP
contents but the same size PEO block. From the data in
Figure 5a,b, it seems that the conformations of both the formed
PLP and the PEO crystallinity are affected by the macromole-
cular architecture and the PLP/PEO composition.
CD measurements of both the diblock and triblock hybrid
polypeptides in water reveal that after treatment in water, the
PLP II form is adopted (Figure 6). Despite a slight shift of the
peaks of the PLP II form in the hybrid triblock (205 and 227 nm),
the spectrum is still indicative of a left-handed helix.
Synthesis of PBLG-b-PLP and PBocLL-b-PLP Copolypep-
tides. Block copolypeptides of L-proline were synthesized
using the sequential addition methodology. Boc-L-Lys NCA or
Figure 6. CD spectra of hybrid polypeptides in water.
Bn-L-Glu NCA were both polymerized first in DMF using
dimethylamine as the initiator, with subsequent addition of the
LP-NCA solution, leading to the polymerization of the second
block. The protected copolypeptides formed aggregates in all
solvents. In organic solvents such as DMF, the PBocLL and
PBLG were soluble, whereas the PLP block was not. In water, the
2404
dx.doi.org/10.1021/bm200495n |Biomacromolecules 2011, 12, 2396–2406

Biomacromolecules
ARTICLE
Figure 8. TGA spectra of PBLG-b-PLP no. 1.
Figure 7. SEC chromatograms of the PBocLL in 0.1 N LiBr DMF
solution (red line), the corresponding deprotected PLL in 5% NaH2PO4
(w/v) solution of water/ACN (97/3 v/v) (blue line), and the diblock
PLL-b-PLP II in the same eluent system (green line).
decomposition, whereas the second transition between 325 and
450 °C is due to that of PLP. The total molecular weight of the
PBLG-b-PLP given in Table 2 of the Supporting Information is
that expected from stoichiometry. The obtained compositions of
these copolypeptides are in excellent agreement with those
expected from stoichiometry. It was not possible to measure
the polydispersity indices and the total molecular weight of the
PBLG-b-PLP copolypeptides by SEC because of the difference in
solubility of the two blocks. However, because the analysis of the
PBocLL-b-PLP showed the high degree of molecular and com-
positional homogeneity of the copolypeptides and because the
PBLG-b-PLPs were similarly synthesized, we can conclude that
these copolypeptides are also well-defined.
PBLG and PBocLL were insoluble, whereas the PLP became
soluble after it was transformed to the PLP II form. The
molecular weight and the polydispersity indices of the first
blocks, that is, PBLG and PBocLL, were obtained by SEC-
TALLS in DMF/LiBr, revealing their high molecular homoge-
neity and the molecular weights expected from stoichiometry.
In the case of PBocLL-b-PLP, the final copolypeptide was
dissolved in TFA, resulting in the deprotection of PBocLys,
simultaneously leading to the transformation to the PLP II form,
thus forming the water-soluble PLL-b-PLP II. The composition
of the deprotected copolypetide was obtained by NMR spec-
troscopy (Supporting Information, Figure F13). To obtain the
molecular weight and the polydispersity of the PBocLL-b-PLP
copolypeptide, the corresponding water-soluble PLL-b-PLP II
was characterized by SEC-TALLS (Figure 7).
The molecular weight of the PLP block of the PBocLL-b-PLP
copolypeptide was obtained as follows: The molecular weight of
the protected PBocLL precursor was 99.1 ꢁ 103 g/mol obtained
by SEC-TALLS in DMF, whereas the molecular weight of
corresponding deprotected block was 68.0 ꢁ 103 g/mol, ob-
tained in water/ACN system (pH 4). The total molecular weight
of the water-soluble PLL-b-PLP II copolypeptide was 208.0 ꢁ
103 g/mol. Therefore, the molecular weight of PLP block is the
resulting difference, 140.0 ꢁ 103 g/mol. It is obvious from
Figure 7 that even if the PLP block is heterogeneously polym-
erized, the final copolypeptide exhibits low polydispersity and has
the molecular weight expected from stoichiometry. Therefore, a
high-molecular-weight PLP (Mn = 140.0 ꢁ 103 g/mol) was
obtained, reported for the first time.
’ CONCLUSIONS
For many years, the synthesis of PLP homo- and copolymers
has been plagued by the inability to synthesize pure monomer
and to employ appropriate initiating systems. Our synthetic
approach led to highly pure LP-NCA due to: (a) the use of
N-protected proline rather than the unprotected one, which on
one hand prevents the formation of DKP and on the other hand
results in Boc-protected byproduct having the appropriate
solubility for easy removal, and (b) the efficient purification
methodology used, which quantitatively removed the side pro-
ducts. This novel route to highly pure LP-NCA, in combination
with an amine initiator and HVT, opens new avenues for the
synthesis of well-defined PLPs and PLP-containing polypeptides.
The synthesis of amphiphiles containing the PLP II, a hydro-
philic helix independent of pH and temperature within the
physiological range, is expected to have significant applications
on the formation of nanoconstructs for drug and gene delivery.
The synthesis of poly(hydroxyl-L-proline)-based materials will be
presented in a forthcoming paper.
The composition of the PBLG-b-PLP copolypeptides was
determined by TGA measurements. PBLG decomposes between
265 and 330 °C and leaves a residue of ∼20%, whereas PLP
decomposes at significantly higher temperatures, between 380
and 450 °C, leaving the same amount of residue. This difference
in decomposition temperature allowed a qualitative estimation of
the composition. In the case of PBLG-b-PLP no. 1, the first
transition between 215 and 360 °C is due to PBLG decomposi-
tion, whereas the second transition between 360 and 450 °C is
due to that of PLP (Figure 8). In the case of PBLG-b-PLP no. 2,
the first transition between 250 and 350 °C is due to PBLG
’ ASSOCIATED CONTENT
S
Supporting Information. Attempts leading to impure
b
LP-NCA, composition of crude products from the various
synthetic methods of LP-NCA, monitoring the purity of LP-
NCA by NMR and IR spectroscopy, molecular characteristics of
the synthesized polymers, SEC chromatogram of PLP homo-
polypeptides, circular dichroism of poly(L-proline) as a function
of temperature, IR spectra of PLP I and II, NMR spectra of PLP,
2405
dx.doi.org/10.1021/bm200495n |Biomacromolecules 2011, 12, 2396–2406

Biomacromolecules
ARTICLE
hybrids and polypeptidic samples. This material is available free
of charge via the Internet at http://pubs.acs.org.
X.; Oddon, M.; Giani, O.; Monge, S.; Robin, J. Macromolecules 2010,
43, 2654–2655.
(22) Aliferis, T.; Iatrou, H.; Hadjichristidis, N. Biomacromolecules
2004, 5, 1653–1656.
(23) Gulin, O. P.; Rabanal, F.; Giralt, E. Org. Lett. 2006, 23,
5385–5388.
’ ACKNOWLEDGMENT
(24) Wilder, R.; Mobashery, S. J. Org. Chem. 1992, 57, 2755–2756.
(25) Hadjichristidis, N.; Iatrou, H.; Pispas, S.; Pitsikalis, M. J. Polym.
Sci., Part A 2000, 38, 3211–3234.
(26) Poche, D.; Moore, M. J.; Bowles, J. L. Synth. Commun. 1999,
29, 843–854.
(27) Hanski, S.; Houbenov, N.; Ruokolainen, J.; Chondronicola, D.;
Iatrou, H.; Hadjichristidis, N.; Ikkala, O. Biomacromolecules 2006,
7, 3379–3384.
The Research Committee of the University of Athens is
acknowledged. The financial support of the Ministry of Educa-
tion through the HRAKLITOS II program, cofinanced from the
Operational Program and Initial Educational Vocational Train-
ing-EPEAEK and the European Social Funds, is greatly appre-
ciated. Mr. Biswajit Sarkar (University at Buffalo) is also
acknowledged for his help in collecting the XRD data.
(28) Fischer, P. J. Peptide Sci. 2003, 9, 9–35.
(29) (a) Bovey, F. A.; Hood, F. P. Biopolymers 1966, 5, 325–326.
(b) Rucker, A. L.; Creamer, T. P. Protein Sci. 2002, 11, 980–985.
(30) Steinberg, I. Z.; Berger, A; Katchalski, E. Biochem. Biophys. Acta
1958, 28, 647–648.
(31) Gornick, F.; Mandelkern, L.; Diorio, A. F.; Roberts, D. E. J. Am.
Chem. Soc. 1964, 86, 2549–2555.
(32) Schimmel, P. R.; Flory, P. J. Procl. Natl. Acad. Sci. U.S.A. 1967,
58, 52–59.
(33) Jeon, S.; Park, S.; Ree, T. J. Polym. Sci., Part A 1989, 27,
1721–1730.
’ REFERENCES
(1) Deber, C. M.; Bovey, F. A.; Carver, J. P.; Blout, E. R. J. Am. Chem.
Soc. 1970, 92, 6191–6198.
(2) Ramachandran, G. N.; Sasisekaran, V. Adv. Protein Chem. 1968,
23, 283–438.
(3) (a) Steinberg, I. Z.; Harrington, W. F.; Berger, A; Sela, M.;
Katchalski, E. J. Am. Chem. Soc. 1960, 82, 5263–5279. (b) Conti, F.;
Piatelli, M.; Viglino, P. Biopolymers 1969, 7, 411–415. (c) Muler, D.;
Stulz, J.; Kricheldorf, H. R. Macromol. Chem. 1984, 185, 1739–1749.
(4) Shoulders, M. D.; Raines, R. T. Annu. Rev. Biochem. 2009,
78, 929–958.
(5) (a) Cowan, P. M.; McGavin, S. Nature 1955, 176, 501–503.
(b) Sasisekharan, V. Acta Crystallogr. 1959, 12, 897–903.
(6) (a) Tooke, L.; Duitch, L; Measey, T. J.; Schweitzer-Stenner, R.
Biopolymers 2009, 93, 451–457. (b) Tiffany, M. L.; Krimm, S. Biopoly-
mers 1972, 11, 2309–2316.
(7) Stryer, L.; Haugland, R. P. Proc. Natl. Acad. Sci. U.S.A 1967,
58, 719–726.
(8) Doose, S.; Neuweiler, H.; Barsch, H.; Sauer, M. Proc. Natl. Acad.
Sci. U.S.A 2007, 104, 17400–17405.
(9) Kraatz, H. B.; Leek, D. M.; Houmam, A.; Enright, G. D.; Lusztyk,
J.; Wayner, D. D. M. J. Organomet. Chem. 1999, 589, 38–49.
(10) (a) Sato, S.; Kwon, Y.; Kamisuki, S.; Srivastava, N.; Mao, Q.;
Kawazoe, Y.; Uesugi, M. J. Am. Chem. Soc. 2007, 129, 873–880.
(b) Fillon, Y. A.; Anderson, J. P.; Chmielewski, J. J. Am. Chem. Soc.
2005, 127, 11798–11803. (c) Arora, P. S.; Ansari, A. Z.; Best, T. P.;
Ptashne, M.; Dervan, P. B. J. Am. Chem. Soc. 2002, 124, 13067–13071.
(d) Crespo, L.; Sanclimens, G.; Montaner, B.; Perez-Tomas, R.; Royo,
M.; Pons, M.; Albericio, F.; Giralt, E. J. Am. Chem. Soc. 2002,
124, 8876–8883.
(11) Papadopoulos, P.; Floudas, G.; Klok, H.-A.; Schnell, I.; Pakula,
T. Biomacromolecules 2004, 5, 81–91.
(12) Hadjichristidis, N.; Iatrou, H.; Pitsikalis, M.; Sakellariou, G.
Chem. Rev. 2009, 109, 5528–5578.
(13) (a) Astbury, W. T.; Dalgliesh, C. E.; Darmon, S. E.; Sutherland,
G. B. B. M. Nature 1948, 162, 596–600. (b) Waley, S. G.; Watson, J. Proc.
R. Soc. 1949, 199, 499–517.
(14) Bamford, C. H.; Block, H.; Pugh, A. C. P. J. Chem. Soc. 1961,
391, 2057–2063.
(15) Goodman, M.; Arnon, U. J. Am. Chem. Soc. 1964, 86, 3384–3390.
(16) Berger, A.; Kurtz, J.; Katchalski, E. J. Am. Chem. Soc. 1954,
76, 5552–5554.
(17) Fasman, G. D.; Blout, E. R. Biopolymers 1963, 1, 3–14.
(18) Kim, K. Y.; Komoto, T.; Kawai, T. Macromol. Chem. 1979,
180, 465–472.
(19) (a) Deming, T.; Curtin, A. S. J. Am. Chem. Soc. 2000,
122, 5710–5717. (b) Peng, Y.; Lai, S.; Lin, C. Macromolecules 2008, 41,
3455–3459.
(20) Lu, H.; Cheng, J. J. Am. Chem. Soc. 2007, 129, 14114–14115.
(21) (a) Dimitrov, I.; Schlaad, H. Chem. Commun. 2003, 23,
2944–2945. (b) Vayaboury, W.; Giani, O.; Cottet, H.; Deratani, A.;
Schuꢀe, F. Macromol. Rapid Commun. 2004, 25, 1221–1224. (c) Zhang,
2406
dx.doi.org/10.1021/bm200495n |Biomacromolecules 2011, 12, 2396–2406