pst283

pst283.xml

wileyml

June 23, 2005

18:38

Encyclopedia of Polymer Sceince and Technology
Copyright c

2005 John Wiley & Sons, Inc. All rights reserved.

POLYPEPTIDE SYNTHESIS, RING-OPENING
POLYMERIZATION OF α-AMINO ACID
N-CARBOXYANHYDRIDES

Introduction

α-Amino acid N-carboxyanhydrides (NCAs) were first described in 1906 by
Hermann Leuchs (1). Leuchs discovered that heating N-ethoxycarbonyl and
N-methoxycarbonyl

α-amino acid chlorides under reduced pressure resulted in

the formation of the corresponding NCAs. Upon exposure to water, he observed
that these NCAs reacted and were transformed into an insoluble mass. The con-
cept of polymers not being established at that time, Leuchs described this product
as cyclic oligopeptides. It was not until the 1920s, when the existence of macro-
molecules became widely accepted, that systematic studies on the synthesis and
polymerization of NCAs were started (2–3). Since then, the NCA polymeriza-
tion has matured and has become the most commonly applied technique for the
large-scale (ie, multigram) synthesis of high molecular weight polypeptides and
polypeptide-based block copolymers. A survey of the number of publications de-
scribing polypeptide synthesis via NCA ring-opening polymerization reveals that
after a strong increase in the 1960s and 1970s, the interest in this technique de-
creased until the mid 1990s (Fig. 1). The renewed interest in the NCA ring-opening
polymerization since then may be attributed to two factors, which are closely re-
lated to each other: (1) the development of novel methods for the polymerization
of NCAs and (2) the notion that polypeptides and polypeptide-based hybrid block
copolymers may be potentially interesting materials for a variety of applications
in biomedicine and biomineralization, and analysis.

Much of the interest in the development of novel materials based on polypep-

tides is driven by their ability to fold in a variety of stable secondary, tertiary, and
quaternary structures (see P

ROTEIN

F

OLDING

). Polypeptides can self-assemble into

complex, hierarchically organized nanoscale morphologies, which cannot be pro-
duced from conventional nonbiological polymers. These characteristics may allow
the preparation of ordered films or organic/inorganic composites that could be
of interest for a multitude of advanced applications. Since the folding properties
of natural peptide polymers (proteins) are, amongst others, due to their uniform
chain lengths and well-defined monomer sequences, synthetic methods that allow
the polymerization of NCAs with precise control of molecular weight, molecular
weight distribution, and composition are important to explore the full potential
of novel synthetic polypeptide materials. This article provides an overview of the

1

pst283

pst283.xml

wileyml

June 23, 2005

18:38

2

POLYPEPTIDE SYNTHESIS

Fig. 1.

Survey of the number of reports describing polypeptide synthesis via NCA ring-

opening polymerization. Source: SciFinder Scholar, Dec. 2003.

different methodologies that have been developed for the polymerization of NCAs
and discusses the scope and limitations of each of these techniques for the prepa-
ration of polypeptides with controlled molecular weights and narrow molecular
weight distributions.

NCA Synthesis

One obvious method for the preparation of NCAs is of course by cyclization of
N-alkoxycarbonyl amino acid chlorides, as originally discovered by Leuchs (eq. 1)
(1). Over the years, various improvements of this so-called Leuchs method have
been made. While the original procedure used thionylchloride for the chlorination
of N-alkoxycarbonyl

α-amino acids, it was found that phosphorus bromide is more

convenient, allowing cyclization to take place at lower temperatures (4). The rate of
cyclization depends on the nature of the alkoxycarbonyl group and decreases in the
following order: ethyl

< methyl < allyl < benzyl. Equation 1 shows a mechanism

that is in line with these observations.

(1)

pst283

pst283.xml

wileyml

June 23, 2005

18:38

POLYPEPTIDE SYNTHESIS

3

One of the most widely applied methods for the preparation of NCAs in-

volves the direct phosgenation of free

α-amino acids (5). This synthesis, which is

referred to as the Fuchs–Farthing method, is usually carried out in inert, low boil-
ing solvents such as tetrahydrofuran, dioxane, or ethyl acetate. Mixtures of these
solvents with methylene chloride have also been reported and have the advantage
of a lower solubility for the hydrochloric acid that is released during the reaction.
A low concentration of hydrochloric acid is important in order to prevent cleav-
age of the NCA ring. The liberated hydrochloric acid may also lead to problems
in the synthesis of NCAs derived from trifunctional amino acids that carry acid-
sensitive protective groups. To avoid premature cleavage of such protective groups,
hydrochloric acid scavengers such as AgCN (6) or triethylamine (7) are added.
Another method that prevents the formation of hydrochloric acid and allows the
synthesis of NCAs carrying acid-sensitive protective groups starts with the prepa-
ration of the corresponding silylated amino acid derivatives. Upon treatment with
phosgene, thionylchloride, or phosphorous trihalides, silylated amino acids react
to produce the NCA and trimethylsilylchloride (8). The original Fuchs–Farthing
procedure used phosgene, but modifications of this method have been developed
in which phosgene is replaced by diphosgene (trichloromethyl chloroformate) (9)
or triphosgene (bis(trichloromethyl)carbonate) (10). Instead of the gaseous phos-
gene, which can be difficult to handle, diphosgene and triphosgene are liquid and
solid, respectively, which makes them much easier to handle and allows the use
of stoichiometric quantities.

Monomer purity is of critical importance for the success of the NCA polymer-

ization; the presence of traces of hydrochloric acid or other contaminants can lead
to premature termination of chain growth and result in broadening of the molec-
ular weight distribution. Advanced procedures have been developed, which allow
the preparation of NCAs in high purities (11). These procedures involve wash-
ing an organic solution containing the NCA with chilled, aqueous bicarbonate.
Polymerization of NCAs purified following this method by primary amine func-
tionalized polystyrene macroinitiators was found to result in synthetic-peptide
hybrid block copolymers with polydispersities that were smaller than those ob-
tained using NCAs prepared and isolated according to the traditional procedures
(12).

Traditional NCA Polymerization Methods

Traditionally, the most important classes of initiators for the NCA ring-opening
polymerization included a broad range of nucleophiles and bases, such as primary
amines, tertiary amines, and alkoxide and hydroxide ions (4,13). Other initiating
systems, such as temperature, metal salts, and organometallic compounds, have
also been used, but these have not found widespread application (4). Polymer-
ization of NCAs using nucleophilic or basic initiators is thought to proceed along
two main pathways, the amine mechanism (eq. 2) and the activated-monomer
mechanism (Fig. 2) (4,13). The amine mechanism is a nucleophilic ring-opening
chain growth process. The activated-monomer mechanism, on the other hand,
starts with deprotonation of an NCA, which subsequently acts as the nucleophile
that initiates chain growth. If decarboxylation is slow, chain growth may proceed

pst283

pst283.xml

wileyml

June 23, 2005

18:38

4

POLYPEPTIDE SYNTHESIS

Fig. 2.

NCA polymerization via the activated monomer mechanism.

along a third pathway, the carbamate mechanism (eq. 3). The preferred pathway
of polymerization for a given initiator is determined by its nucleophilicity/basicity
ratio. Generally, NCA polymerizations initiated by primary amines, which have
a high nucleophilicity/basicity ratio, predominantly follow the primary amine
mechanism. NCA polymerizations initiated by tertiary amines or alkox-
ide/hydroxide ions, which are more basic than nucleophilic, mainly proceed along
the activated monomer mechanism. This classification, however, is by no means
unambiguous, and it is important to note that a given polymerization can switch
back and forth many times between the different mechanisms during the course
of the reaction. In other words, a propagation step for one mechanism is a side
reaction for the other, and vice versa.

(2)

(3)

The existence of multiple, competitive mechanisms, which may act simul-

taneously during the course of a polymerization reaction, restricts control over
the molecular weight and the molecular weight distribution of the resulting

pst283

pst283.xml

wileyml

June 23, 2005

18:38

POLYPEPTIDE SYNTHESIS

5

polypeptide and also hampers the formation of well-defined block copolymers.
Using very nucleophilic or basic initiators, the NCA polymerization can be bi-
ased toward the amine and activated-monomer mechanism, respectively. These
initiators allow a certain level of control over the polymerization and can produce
polypeptides with molecular weights which roughly correlate with the monomer-
to-initiator ratio.

Especially primary amines and primary amine functionalized macroinitia-

tors have been extensively used for the preparation of a broad range of polypep-
tides and polypeptide-containing hybrid polymers. In addition to simple polypep-
tide homopolymers, random copolymers have been prepared by co- or terpolymer-
ization of different NCAs. Polypeptide copolymers have attracted attention, eg,
as polymeric drug carriers (14). Furthermore, random copolypeptides have been
used to study the conformational properties of amino acids (15). Polypeptides and
polypeptide-containing block copolymers have been obtained by sequential copoly-
merization of different NCAs and by using primary amine end-functionalized non-
peptidic macroinitiators for the NCA ring-opening polymerization, respectively.
Amongst other, triblock copolypeptides have been used to investigate the helix-
forming propensities of amino acids (16). The solid-state structures of synthetic-
peptide hybrid block copolymers have been studied in considerable detail (17).
Most of these block copolymers form lamellar superstructures, with the polypep-
tide chains folded and arranged in an hexagonal array. Furthermore, peptide-
synthetic hybrid block copolymers can possess good antithrombogenic properties,
which make them of interest for medical applications.

In addition to simple, linear polypeptides, the NCA polymerization has

also been used to prepare branched macromolecular architectures. Star polypep-
tides, probably the simplest example of branched polypeptides, can be obtained
via NCA polymerization using multiple, primary amine functionalized initia-
tors. Examples of multifunctional initiators include hexakisamines derived from
1,3,5-tris(bromomethyl)benzene (18), tetra-amino functionalized perylene chro-
mophores (19) and cyclotriphosphazenes (20). Interestingly, Teflon-supported
films of hexaarmed, ethylene glycol substituted polyglutamic acid star polymers
based on a cyclotriphosphazene core were found to be excellent enantioselective
membranes for amino acids (21). A further class of initiators that has been fre-
quently used to prepare star-shaped polypeptides are poly(amido amine) (PA-
MAM) dendrimers (22-23). PAMAM dendrimers have been used as initiators for
the ring-opening polymerization of sugar-substituted NCAs (eg, O-(tetra-O-acetyl-
β-

D

-glucopyranosyl)-

L

-serine N-carboxyanhydride and O-(2-acetamido-3,4,6-tri-

O-acetyl-2-deoxy-

β-

D

-glucopyranosyl)-

L

-serine N-carboxyanhydride) (23) as well

as

γ -benzyl-

L

-glutamate N-carboxyanhydride (23). It was found that the helicity

of the PAMAM-based star polypeptides was significantly enhanced compared to
their linear analogues. This effect was attributed to the aggregation of peptide
segments on the dendrimer surface.

More complex branched polypeptides can be obtained following iterative re-

action schemes involving several NCA ring-opening polymerization steps. Graft
copolypeptides, which have also been termed multichain polypeptides, have been
prepared using a linear poly(

L

-lysine) as initiator for the NCA ring-opening poly-

merization (24). These graft polypeptides were designed with the aim of de-
veloping polymer–drug and polymer–epitope conjugates (25–26). This synthetic

pst283

pst283.xml

wileyml

June 23, 2005

18:38

6

POLYPEPTIDE SYNTHESIS

approach can be further extended by utilizing these graft copolymers as initiators
for a next NCA ring-opening polymerization step. Repetition of this graft-on-graft
strategy leads to highly branched, so-called dendritic-graft, polypeptides (27). In
addition to the graft-on-graft strategy, branched polypeptides can also be prepared
by end-functionalization of a peptide chain with an appropriate N

α

,N

ε

-diprotected

lysine derivative (28). Deprotection of the lysine amino groups doubles the num-
ber of end groups that can be used to initiate a subsequent NCA ring-opening
polymerization step. Repetition of this NCA ring-opening polymerization–end-
functionalization–deprotection sequence has been successfully used to prepare
highly branched poly(

L

-lysine)s with molecular weights up to 33 kDa in small

number of reaction steps.

Primary amine initiated NCA polymerizations have not only been carried

out in homogeneous solution but also under heterogeneous conditions. The use
of primary amine functionalized surfaces allows the preparation of thin, direc-
tionally aligned polypeptide films (29). If the density of the peptide chains at the
surface is sufficiently high, the chains are mainly oriented normal to the sur-
face. Such polypeptide films have been grown from a variety of surfaces, includ-
ing silicon, aluminum, gold, indium-tin-oxide, quartz, and polytetrafluoroethylene
(Teflon). Surface-initiated NCA polymerizations can be carried out either by ex-
posing the amino-functionalized substrate to a solution containing the appropriate
NCA monomer or by vapor phase deposition of the NCA monomer (30). Sequential
addition of two different NCA monomers allows the preparation of surface-grafted
polypeptide block copolymers (31). Surface-grafted polypeptide films obtained via
NCA polymerization have been found to possess piezoelectric properties, (32) in-
dicative for the parallel alignment of the peptide helices. Furthermore, the water
permeability of porous Teflon membranes, which were modified with a surface-
grafted layer of poly(

L

-glutamic acid) could be modulated by pH-induced variations

in the secondary structure of the peptide chains (33).

Transition-Metal-Mediated NCA Polymerization

Transition-metal complexes have already been explored as initiators for the NCA
ring-opening polymerization since the 1960s (34–35). Unfortunately, most of these
complexes were characterized by relatively low initiator efficiencies and did not
permit the synthesis of polypeptides with precise control of molecular weight and
narrow molecular weight distributions in high yields. This situation changed dra-
matically, however, with the discovery of zero-valent nickel and cobalt complexes,
which were found to be able to promote very efficient “living” polymerization of
NCAs (36–37). The putative mechanism for the polymerization of NCAs using
these initators, bipyNi(COD) and (PMe

3

)

4

Co, is outlined in Figure 3 (35). These

transition-metal initiators were found to offer unprecedented control over molec-
ular weight and allowed the high yield synthesis of polypeptides with molecular
weights between 500 and 500,000 Da and polydispersities

<1.20. Sequential ad-

dition of two different NCA monomers allowed the formation of well-defined block
copolypeptides.

Peptide block copolymers prepared via transition-metal-mediated NCA poly-

merization have been investigated, eg, to direct the biomimetic synthesis of

pst283

pst283.xml

wileyml

June 23, 2005

18:38

POLYPEPTIDE SYNTHESIS

7

Fig. 3.

Proposed mechanism for NCA polymerization using the initiators bipyNi(COD)

and (PMe

3

)

4

Co.

ordered silica structures (38) or to prepare self-assembled polypeptide hydrogels
(39). Furthermore, such block copolypeptides have found application to generate
hollow, organic/inorganic hybrid microspheres, which are composed of a thin in-
ner layer of gold nanoparticles surrounded by a thick layer of silica nanoparticles
(40). Alternatively, hollow spheres could be prepared that consisted of a thick inner
layer of core–shell CdSe/CdS nanoparticles that are surrounded by a thicker silica
nanoparticle layer (41). The latter spheres allow for microcavity lasing without
the use of additional mirrors, substrate spheres, or gratings.

Initially, a major drawback of the transition-metal-mediated NCA polymer-

ization, as it is outlined in Fig. 3, was the lack of control over the chain end
functionality of the polypeptide chains. This is in great contrast to, eg, primary
amine initiated NCA polymerizations, which are widely explored for the prepara-
tion of synthetic-peptide block copolymers and other peptide-hybrid architectures.
This limitation could be overcome by the direct synthesis of the amido amidate
metallacycle propagating species (eq. 4) (42).

(4)

These complexes allowed controlled polymerization of NCAs, which also pro-

vided additional support for their validity as polymerization intermediates. The
initiating ligand was found to be quantitatively incorporated as a C-terminal

pst283

pst283.xml

wileyml

June 23, 2005

18:38

8

POLYPEPTIDE SYNTHESIS

Fig. 4.

Synthesis of poly(

γ -benzyl-

L

-glutamate)-polyethylene-poly(

γ -benzyl-

L

-glutamate)

triblock copolymers from metallacyle-substituted macroinitiators.

group, which would allow the introduction of a broad variety of functional
end groups. Interestingly, the direct synthesis of functional amido amidate
metallacycles is not restricted to low molecular weight substituents, but can
also be extended to polymeric macroinitiators (43). This is outlined in Fig. 4,
which shows the transformation of

α,ω-amino polyoctenamers obtained via AD-

MET into metallacycle-substituted macroinitiators. These macroinitiators were
used to polymerize

γ -benzyl-

L

-glutamate N-carboxyanhydride (Bn-Glu NCA).

Selective hydrogenation of the polyoctenamer double bonds subsequently af-
forded poly(

γ -benzyl-

L

-glutamate)–polyethylene–poly(

γ -benzyl-

L

-glutamate) tri-

block copolymers. Transition-metal-mediated NCA polymerization is not re-
stricted to homogeneous solutions, but can also be adapted to allow controlled
polymerization of NCAs from surfaces (44).

Replacement of the achiral bipyridyl ligand in bipyNi(COD) by an optically

active 2-pyridinyloxazoline ligand results in a metal complex (1) that can act

pst283

pst283.xml

wileyml

June 23, 2005

18:38

POLYPEPTIDE SYNTHESIS

9

Fig. 5.

Proposed mechanism for the polymerization of Bn-Glu NCA by 2.

as an initiator for the stereoselective polymerization of Bn-Glu NCA (45). Com-
plex 1 has polymerization properties identical to bipyNi(COD), ie, it can produce
narrow polydispersity polypeptides with molecular weights corresponding to the
monomer–initiator ratio in high yields. For the chiral 2-pyridinyloxazoline com-
plexes, however, the polymerization rate constant for the

D

enantiomer (k

D

) was

found to be five times larger than that of the

L

enantiomer. These enantioselectiv-

ities are higher than those reported for many other chiral initiators and catalysts
that have been explored for the stereoselective polymerization of NCAs (46). Even
better results were obtained with chiral ruthenium initiators derived from the
Noyori complex (2) (47).

The presence of 1 equiv 1,2-bis(dimethylphosphino)ethane (dmpe) resulted

in an initiator that can promote living NCA polymerization of Bn-Glu NCA with
a rate constant (k) that was found to be eight times larger for the

L

enantiomer in

comparison with the

D

antipode. A putative model for the polymerization of Bn-

Glu NCA by ruthenium complex (2) is shown in Fig. 5. It has been proposed that
the observed enantioselectivity is due to the rigid, chiral coordination geometry
at the metal center, which is located at the propagating chain end.

pst283

pst283.xml

wileyml

June 23, 2005

18:38

10

POLYPEPTIDE SYNTHESIS

Fig. 6.

Proposed mechanism for NCA polymerization in the presence of primary amine

hydrochlorides.

Polymerization of NCAs Using Primary Amine Hydrochlorides

The inherent problem of the use of any of the traditional initiators for the NCA
ring-opening polymerization is that the amino group at the growing peptide’s N-
terminus can act as both a nucleophile and a base. As a result, chain growth
may proceed via both the amine and the activated-monomer mechanism. In the
previous paragraph, the use of transition-metal complexes to control the reactiv-
ity of the growing polymer chain end has been described. An interesting alter-
native approach, which allows chain growth to take place preferentially via the
amine mechanism and suppresses the activated-monomer pathway is to use pri-
mary amine hydrochlorides as initiators for the NCA ring-opening polymerization
(48). The use of primary amine hydrochlorides is based on early work by Knobler
and co-workers (49) who investigated stoichiometric reactions between primary
amine hydrochlorides and NCAs. Their investigations revealed that this reaction
proceeds smoothly and without the formation of polypeptides. When anionically
prepared polystyrene macroinitiators with a primary amine end group were used
for the ring-opening polymerization of

ε-benzyloxycarbonyl-

L

-lysine NCA, hybrid

block copolymers with very narrow polydispersities (

<1.03) were obtained (48).

These narrow polydispersities, which are close to a Poisson distribution, suggest
that primary amines promote polymerization of NCAs in a well-controlled fash-
ion. The mechanism proposed for the polymerization of NCAs in the presence of
primary amine hydrochlorides is outlined in Figure (6). As illustrated in the fig-
ure, the primary amine hydrochloride is regarded as a dormant species, which
is in equilibrium with the corresponding free amine and H

+

(Cl

−

). While the pri-

mary amine end group is responsible for chain growth via the amine mechanism,
the liberated H

+

(Cl

−

) serves to protonate any NCA anion that may be present,

thereby preventing polymerization via the activated monomer mechanism.

Polymerization of NCAs under High Vacuum Conditions

The two approaches that have been discussed so far, viz transition-metal-mediated
and primary amine hydrochloride initiated NCA ring-opening polymerization,

pst283

pst283.xml

wileyml

June 23, 2005

18:38

POLYPEPTIDE SYNTHESIS

11

Fig. 7.

GPC traces illustrating the evolution of polypeptide molecular weight

and monomer consumption during the polymerization of

γ -benzyl-

L

-glutamate N-

carboxyanhydride (Bn-Glu NCA). Targeted number-average degree of polymerization was

∼440. After 17 h, monomer consumption is >99%.

suggest that the limited control over molecular weight and molecular weight dis-
tribution of polypeptides produced from NCAs using traditional initiators such as
primary amines is due to the polymerization mechanism. However, polypeptide
homopolymers and block copolypeptides with molecular weights corresponding
to the monomer–initiator ratio and with narrow molecular weight distributions
can also be prepared using high vacuum techniques, which have traditionally
been used for the anionic polymerization of vinyl monomers (50). High vacuum
techniques allow careful purification of all reagents and can provide a reaction
environment that is virtually free of any impurities that can interfere with poly-
merization. Under these conditions, using monomers purified via high vacuum
techniques and thoroughly dried solvents, primary amine initiated NCA polymer-
izations can produce polypeptide homo and block copolymers with precise control
of molecular weight and composition and narrow polydispersities (51). These pri-
mary amine initiated NCA ring-opening polymerizations fulfill all the require-
ments to be considered truly living polymerizations. For example, poly(benzyl-

L

-glutamate) homopolymers with a number-average degree of polymerization of

∼440 (M

n

≈ 100,000 Da) could be produced with a polydispersity of 1.18. Monomer

consumption could be monitored using gel-permeation chromatography and was
found to be quantitative after 17 h (Fig. 7).

Summary and Conclusions

Since its initial discovery by Leuchs in the early twentieth century, the NCA
ring-opening polymerization has grown to become the most convenient strategy

pst283

pst283.xml

wileyml

June 23, 2005

18:38

12

POLYPEPTIDE SYNTHESIS

for the large-scale preparation of polypeptides and synthetic-polypeptide hybrid
materials. Traditionally, initiators for the NCA ring-opening polymerization were
nucleophiles and bases, such as primary amines and KOtBu, which, however, only
provided limited control over the molecular weight and molecular weight distribu-
tion of the resulting polypeptides. This is due to the existence of multiple, compet-
ing reaction pathways, which may act simultaneously during polymerization. To
overcome these problems and enhance control over molecular weight, molecular
weight distribution, and polymer composition, which is required for the prepa-
ration of advanced, complex macromolecular architectures, several alternative
strategies have been developed. One approach involves the use of transition-metal
complexes to control the reactivity of the growing polymer chain end. An alter-
native strategy, which has been successfully used to bias the NCA ring-opening
polymerization to proceed exclusively via one pathway relies on the use of primary
amine hydrochlorides as initiators. While these two examples suggest that the lim-
ited control over molecular weight, molecular weight distribution, and composition
of polypeptides produced via NCA ring-opening polymerization using traditional
nucleophilic or basic initiators can be solely attributed to the polymerization mech-
anism, the picture may be more complex. When high vacuum techniques are used
to purify the NCA monomer and to conduct the polymerization, primary amines
can also generate polypeptide homopolymers and block copolypeptides with nar-
row polydispersities and predictable molecular weights. These primary amine
initiated NCA ring-opening polymerizations were shown to fulfill all the require-
ments to be characterized as truly living polymerizations. This last example un-
derlines the importance of monomer purity and the absence of other contaminants
that can interfere with the polymerization. Almost 100 years after its discovery,
the mechanistic details of the NCA polymerization have not yet been fully unrav-
elled. Nevertheless, great advances have been made in the development of novel
synthetic methodologies, which allow controlled polymerization of NCAs, ie, the
production of peptide polymers with controlled architecture, predictable molecu-
lar weight, and narrow molecular weight distributions. These advanced synthetic
methodologies will prove useful to further exploit the potential of these materials,
in particular for applications in biomedicine and biomineralization.

BIBLIOGRAPHY

“Polypeptides” in EPSE 2nd ed., Vol. 12, pp. 786–811, by E. Peggion, University of Padua,
Italy, and M. Goodman, University of California at San Diego.

1. H. Leuchs, Ber. Dtsch. Chem. Ges. 39, 857 (1906);

H. Leuchs and W. Manasse, Ber.

Dtsch. Chem. Ges. 40, 3235 (1907); H. Leuchs and W. Geiger, Ber. Dtsch. Chem. Ges.
41, 1721 (1908).

2. F. Wessely, Z. Physiol. Chem. 146, 72 (1925); F. Sigmund and F. Wessely, Z. Physiol.

Chem. 157, 91 (1926); F. Wessely and F. Sigmund, Z. Physiol. Chem. 159, 102 (1926);
F. Wessely and M. John, Z. Physiol. Chem. 170, 38 (1927);

F. Wessely and J. Mayer,

Monatsh. Chem. 50, 439 (1928).

3. E. Katchalski, Adv. Protein Chem. 6, 123 (1951).
4. H. R. Kricheldorf,

α-aminoacid-N-carboxyanhydrides and Related Heterocycles—

Synthesis, Properties, Peptide Synthesis, Polymerization, Springer-Verlag, Berlin, 1987.

pst283

pst283.xml

wileyml

June 23, 2005

18:38

POLYPEPTIDE SYNTHESIS

13

5. F. Fuchs, Chem. Ber. 55, 2943 (1922);

C. J. Brown, D. Coleman, and A. C. Farthing,

Nature 163, 834 (1949); A. C. Farthing, R. J. W. Reynolds, Nature 165, 647 (1950); A.
C. Farthing, J. Chem. Soc. 3213 (1950).

6. R. Hirschmann, H. Schwam, R. G. Strachan, E. F. Schoenewaldt, H. Barkemeyer, S. M.

Miller, J. B. Conn, V. Garsky, D. F. Veber, and R. G. Denkewalter, J. Am. Chem. Soc.
93, 2746 (1971).

7. R. Wilder and S. Mobashery, J. Org. Chem. 57, 2755 (1992).
8. H. R. Kricheldorf, Chem. Ber. 103, 3353 (1970); Chem. Ber. 104, 87 (1971).
9. M. Oya, R. Katakai, H. Nakai, and Y. Iwakura, Chem. Lett. 1143 (1973); R. Katakai

and Y. Iizuka, J. Org. Chem. 50, 715 (1985).

10. W. H. Daly and D. S. Poch´e, Tetrahedron Lett. 29, 5859 (1988).
11. D. S. Poch´e, M. J. Moore, and J. L. Bowles, Synth. Commun. 29, 843 (1999).
12. I. Dimitrov, H. Kukula, H. C¨olfen, and H. Schlaad, Macromol. Symp. (in press).
13. T. J. Deming, Adv. Mater. 9, 299 (1997);

T. J. Deming, J. Polym. Sci., Part A: Polym.

Chem. 38, 3011 (2000).

14. J. M. Anderson, D. F. Gibbons, R. L. Martin, A. Hiltner, and R. Woods, J. Biomed. Mater.

Res. Symp. 5, 197 (1974); R. Duncan, CRC Crit. Rev. Biocompat. 2, 127 (1986); H. C.
Chiu, P. Kopeckova, S. D. Deshmane, and J. Kopecek, J. Biomed. Mater. Res. 34, 381
(1997).

15. R. K. Scheule, F. Cardinaux, G. T. Taylor, and H. A. Scheraga, Macromolecules 9, 23

(1976).

16. F. Cardinaux, J. C. Howard, G. T. Taylor, and H. A. Scheraga, Biopolymers 16, 2005

(1997);

D. M. Rothwarf, V. G. Davenport, P.-T. Shi, J.-L. Peng, and H. A. Scheraga,

Biopolymers 39, 531 (1996).

17. B. Gallot, Prog. Polym. Sci. 21, 1035 (1996); H.-A. Klok and S. Lecommandoux, Adv.

Mater. 13, 1217 (2001);

H. Schlaad and M. Antonietti, Eur. Phys. J. E 10, 7 (2003),

and references cited therein.

18. W. H. Daly, D. Poche, P. S. Russo, and I. Negulescu, Polym. Prepr. 33(1), 188 (1992);

W. H. Daly, D. S. Poche, P. S. Russo, and I. I. Negulescu, J. Macromol. Sci., A 31(Suppl.
6/7), 795 (1994).

19. H.-A. Klok, J. Rodriguez-Hernandez, S. Becker, and K. M ¨

ullen, J. Polym. Sci., Part A:

Polym. Chem. 39, 1572 (2001).

20. K. Inoue, H. Sakai, S. Ochi, T. Itaya, and T. Tanigaki, J. Am. Chem. Soc. 116, 10783

(1994); K. Inoue, S. Horibe, M. Fukae, T. Muraki, E. Ihara, and H. Kayama, Macromol.
Biosci. 3, 26 (2003).

21. K. Inoue, A. Miyahara, and T. Itaya, J. Am. Chem. Soc. 119, 6191 (1997).
22. K. Aoi, K. Tsutsumiuchi, A. Yamamoto, and M. Okada, Tetrahedron 53, 15415 (1997).
23. N. Higashi, T. Koga, N. Niwa, and M. Niwa, Chem. Commun. 361 (2000); N. Higashi,

T. Koga, and M. Niwa, Adv. Mater. 12, 1373 (2000).

24. M. Sela, S. Fuchs, and R. Arnon, Biochem. J. 85, 223 (1962);

M. Sakamoto and Y.

Kuroyanagi, J. Polym. Sci., Part A: Polym. Chem. 16, 1107 (1978); G. Mezo, J. Remenyi,
J. Kajtar, K. Barna, D. Gaal, and F. Hudecz, J. Control. Release 63, 81 (2000).

25. F. Hudecz, Anti-Cancer Drug 6, 171 (1995).
26. F. Hudecz, in A. Aggeli, N. Boden, and S. Zhang, eds., Self-Assembling Peptide Systems

in Biology, Medicine and Engineering, Kluwer Academic Publishers, Boston, Mass.,
2001, Chapt. 11.

27. H.-A. Klok and J. Rodr´ıguez-Hern ´andez, Macromolecules 35, 8718 (2002).
28. J. Rodr´ıguez-Hern ´andez, M. Gatti, and H.-A. Klok, Biomacromolecules 4, 249 (2003).
29. J. K. Whitesell and H. K. Chang, Science 261, 73 (1993);

R. H. Wieringa and A. J.

Schouten, Macromolecules 29, 3032 (1996).

30. Y.-C. Chang and C. W. Frank, Langmuir 14, 326 (1998);

Y. Wang and Y.-C. Chang,

Langmuir 18, 9859 (2002).

pst283

pst283.xml

wileyml

June 23, 2005

18:38

14

POLYPEPTIDE SYNTHESIS

31. R. H. Wieringa, E. A. Siesling, P. J. Werkman, E. J. Vorenkamp, and A. J. Schouten,

Langmuir 17, 6491 (2001).

32. T. Jaworek, D. Neher, G. Wegner, R. H. Wieringa, and A. J. Schouten, Science 279, 57

(1998).

33. Y. Ito, Y. Ochiai, Y. S. Park, and Y. Imanishi, J. Am. Chem. Soc. 119, 1619 (1997).
34. T. J. Deming, Adv. Mater. 9, 299 (1997), and references therein.
35. T. J. Deming, J. Polym. Sci., Part A: Polym. Chem. 38, 3011 (2000), and references

therein.

36. T. J. Deming, Nature 390, 386 (1997).
37. T. J. Deming, Macromolecules 32, 4500 (1999).
38. J. N. Cha, G. D. Stucky, D. E. Morse, and T. J. Deming, Nature 403, 289 (2000).
39. A. P. Novak, V. Breedveld, L. Pakstis, B. Ozbas, D. J. Pine, D. Pochan, and T. J. Deming,

Nature 417, 424 (2002).

40. M. S. Wong, J. N. Cha, K.-S. Choi, T. J. Deming, and G. D. Stucky, Nano Lett. 2, 583

(2002).

41. J. N. Cha, M. H. Bartl, M. S. Wong, A. Popitsch, T. J. Deming, and G. D. Stucky, Nano

Lett. 3, 907 (2003).

42. S. A. Curtin and T. J. Deming, J. Am. Chem. Soc. 121, 7427 (1999).
43. K. R. Brzezinska and T. J. Deming, Macromolecules 34, 4348 (2001).
44. P. Witte and H. Menzel, Polym. Prepr. 44, 423 (2003).
45. J. Cheng and T. J. Deming, Macromolecules 32, 4745 (1999).
46. T. Tsuruta, S. Inoue, and K. Matsuura, Makromol. Chem. 63, 219 (1963); K. Matsuura,

S. Inoue, and T. Tsuruta, Makromol. Chem. 80, 149 (1964); T. Makino, S. Inoue, and T.
Tsuruta, Makromol. Chem. 150, 137 (1971); H. G. B ¨

uhrer and H. G. Elias, Makromol.

Chem. 169, 145 (1973); S. Yamashita, N. Yamawaki, and H. Tani, Macromolecules 7,
724 (1974); N. Oguni, H. Kuboyama, and A. Nakamura, J. Polym. Sci., Part A: Polym.
Chem. 21, 1559 (1983).

47. S. W. Seidel and T. J. Deming, Macromolecules 36, 969 (2003).
48. I. Dimitrov and H. Schlaad, Chem. Commun. 2944 (2003).
49. Y. Knobler, S. Bittner, and M. Frankel, J. Chem. Soc. 3941 (1964);

Y. Knobler, S.

Bittner, D. Virov, and M. Frankel, J. Chem. Soc. 1821 (1969).

50. N. Hadjichristidis, H. Iatrou, S. Pispas, and M. Pitsikalis, J. Polym. Sci., Part A: Polym.

Chem. 38, 3211 (2000).

51. T. Aliferis, H. Iatrou, and N. Hadjichristidis, Biomacromolecules (in press).

H

ARM

-A

NTON

K

LOK

Ecole Polytechnique F´ed´erale de Lausanne