Saccharide-peptide hybrid copolymers as biomaterials

Article abstract of DOI:10.1002/anie.200501944

Au natural: A new class of synthesized biomaterials (saccharide-peptide hybrid copolymers shown) are biodegradable, nontoxic, and nonimmunogenic. The cationic saccharide-peptide hybrid copolymers were also shown to be effective in compacting and transferr

Full text of DOI:10.1002/anie.200501944

Angewandte
Chemie
Biomaterials
DOI: 10.1002/anie.200501944
Saccharide–Peptide Hybrid Copolymers as
Biomaterials**
Mark Metzke, Naphtali OꢀConnor, Soumen Maiti,
Edward Nelson, and Zhibin Guan*
Biomaterials are important for many biomedical applications
such as implants and prosthetics, pharmaceutical formula-
tions, drug and gene-delivery agents, DNA and protein
microarrays, and tissue engineering.[1] Synthetic polymers
remain the most versatile class of biomaterials because of the
ease in controlling their compositions, structures, and proper-
ties.[2,3] There is currently a great demand for novel polymeric
biomaterials with tailored structures and more-defined func-
tions. Ideally, desired biomaterials can be chemically synthe-
sized in a ground-up approach to exhibit precisely the needed
chemical, biological, and engineering properties for the
targeted medical application.
An important strategy for designing new biomaterials is to
construct synthetic polymers from natural building blocks.
The premise is to combine the advantages of both bio- and
synthetic polymers, thereby gaining the biocompatibility and
biodegradability of natural materials and the versatility of
synthetic structural design. Examples of biomaterials made
[*] M. Metzke, N. O’Connor, Dr. S. Maiti, Prof. Dr. Z. Guan
Department ofChemistry
University ofCaliof rnia
Irvine, CA 92697-2025 (USA)
Fax: (+1)949-824-2210
E-mail: zguan@uci.edu
Prof. Dr. E. Nelson
School ofMedicine
University ofCaliof rnia
Irvine, CA 92697 (USA)
[**] We thank the Arnold and Mabel Beckman Foundation and the
University of California at Irvine (UCI) for partial financial support.
We thank Professor Ken Longmuir and Sherry Haynes for generous
help with cell culturing and gene transfection studies, Professor
Gregory Weiss and Sara Avrantinis for assistance in DNA electro-
phoresis study, and Jane Z. Bai and Derek Weisel for the AFM study.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2005, 44, 6529 –6533
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6529

Communications
from natural building blocks include poly(lactic acid),[4]
chemical, biological, and mechanical properties of these
polymers. Although much effort has been devoted to
making oligomeric glycopeptides that can mimic the structure
and function of natural glycoproteins,[11,12] little has been
reported on the synthesis of high-molecular-weight saccha-
ride–peptide polymers.[13,14] In our novel design, both the
saccharide and peptide building blocks are incorporated into
the main chain of the polymer. Herein we report our general
design concept, synthesis, and the initial testing of a new class
of biomaterials. We also discuss our studies of these saccha-
ride-peptide hybrid copolymers in one specific biomedical
application, gene delivery.
poly(glycolic acid),[5] poly(anhydride)s,[6] poly(amino
acid)s,[7] pseudo-poly(amino acid)s,[8] carbohydrate-derived
polyesters,[9] and artificial proteins.[10] Whereas these poly-
mers generally exhibit good biocompability, and some are
used in important clinical applications, their structural
diversity and functional properties are relatively limited,
warranting further development of more versatile biomate-
rials.
Herein we describe a new class of biomaterials derived
from natural saccharide and amino acid building blocks
(Figure 1). Our design is based on the following consider-
We chose oligolysines and a galactose-derived monomer 1
for copolymerization. Three hybrid polymers (poly-1, poly-2,
and poly-3) were synthesized by interfacial polymerization
with various peptide monomers 2–4 (Scheme 1). 1H and
13C NMR spectroscopic analyses were used to confirm the
structure of each polymer. The number-averaged molecular
weights (Mn) of the three prepolymers for poly-1, poly-2, and
poly-3 were measured by gel-permeation chromatography
with polystyrene calibration standards to be 15000, 9085, and
13200 gmolÀ1 respectively. Global cleavage of the acetonide
and tert-butoxycarbonyl (Boc) protecting groups afforded the
final carbohydrate–peptide hybrid copolymers, poly-1, poly-2,
and poly-3 (details of the synthesis and characterization can
be found in the Supporting Information).
After the successful synthesis of the hybrid copolymers,
we investigated their general properties as biomaterials
including the biodegradability, cytotoxicity, and immunolog-
ical properties. Enzymatic-degradation studies with serine
proteases (subtilisin A and trypsin) showed that the polymers
are indeed biodegradable. MALDI-TOF mass spectra were
measured periodically to monitor qualitatively the decrease
Figure 1. Formation ofthe saccharide–peptide hybrid copolymer.
ations; 1) saccharides and amino acids are abundant and
readily available natural monomers, 2) these natural building
blocks are likely to yield polymers with inherent biocompat-
ibility that then degrade into natural, nontoxic, and biosorb-
able species, 3) their rich functionalities allow convenient
modification/tailoring of materials for desired applications,
and 4) the modular synthesis offers advantages in combining
structural precision with design flexibility.
The combination of various modules gives rise to numer-
ous polymers suitable for combinatorial screening of the
Scheme 1. Synthesis ofgalactaro–oligolysine hybrid copolymers (poly- 1, poly-2, poly-3). a) Interfacial polymerization, H2O/CCl4; b) 30% trifluoro-
acetic acid in THF/water.
6530
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6529 –6533

Angewandte
Chemie
in the molecular weight of the polymer over time. The
resultant profile indicated that the polymers were nearly
completely degraded after 5–7 days of exposure to an enzyme
(see Supporting Information).
The cytotoxicity of the polymers was then assayed at
various concentrations by means of a standard MTT (3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) test.
The hybrid copolymers exhibited minimal cytotoxicity to
Cos 7 cells at a range of concentrations (Figure 2). Poly-l-
lysine (PLL) with Mn = 8500 gmolÀ1 was used as a control
membrane. Other studies have shown that the introduction
of carbohydrate units into other polymers also results in a
lower cytoxicity.[16]
As our carbohydrate–peptide copolymers are new com-
pounds, it is important to test whether they generate immune
responses in vivo. As a representative example, the immuno-
genicity of poly-3 was evaluated by ELISA with Fisher 344
rats as models. Based on a standard protocol, 100 mg of poly-3
was administrated in the first, third, and sixth weeks by either
subcutaneous (SC) injections in the footpad or by intravenous
(IV) injections into tail veins. The animals underwent a
phlebotomy 21 days after each administration of the polymer.
The serum for ELISA testing was obtained at three-week
intervals. If a positive immunogenic response is elicited by the
polymer, then antibodies generated in the rat serum would
bind to the polymer-coated wells. An anti-rat immunoglobu-
lin (IgG) conjugated HRP (horseradish peroxidase) then
binds to the adsorbed polymer antibodies, catalyzing oxida-
tion of a substrate that can be detected by UV/Vis spectrom-
etry (see Supporting Information for experimental details).
Figure 3 summarizes the ELISA data, which show no
evidence of antibody response. All rats were healthy (no
weight loss, normal activity, good hygiene/quality fur), which
suggests that there is no adverse immune response or toxicity.
After the assays to test their biodegradability, cytotoxicity,
and immunogenicity, the hybrid copolymers were evaluated
for a specific biomedical application. Synthetic cationic
Figure 2. Cytotoxicity data acquired through MTT testing at various
polymer concentrations with PLL as the control. Comparison was
made with cationic amino groups at equivalent molar concentrations,
consistent with those used in gene transfection studies. Standard devi-
ations are shown with error bars (n=4). The symbol *** indicates
statistical significance at levels of p<0.001 for the experimental poly-
mers versus each concentration ofPLL (indicated by brackets). The
p values were obtained by using Student–Newman–Keuls multiple
comparisons testing.
because our polymers are derived partially from l-lysine and
are of comparable molecular weight, which affects both
cytotoxicity and transfection levels.[15] Comparisons between
the polymers and the control were made at equivalent molar
concentrations of cationic amino groups (the same as those
used in later gene-transfection studies). The contrast is
striking: whereas the PLL homopolymer exhibits high
cytotoxicity at relatively low concentrations (5–
100 mgmLÀ1), our hybrid copolymers show lower cytotoxicity
even at higher concentrations (82–330 mgmLÀ1 for poly-1; 56–
226 mgmLÀ1 for poly-2; and 46–183 mgmLÀ1 for poly-3). The
placement of saccharide spacers on the main chain of the
polymers lowered their cytotoxicity to levels approaching the
blank controls in the absence of polymer. Although the exact
mechanism for this decrease in toxicity remains to be
investigated, we believe that the breakage of the cationic
polypeptide into short segments with saccharide spacers
lowers continuous charge density while the hydrophilic
saccharide fragments shield the surface charge of the poly-
plexes. Both effects are hypothesized to alleviate disruptive
coulombic interactions of the polyplexes with the cell
Figure 3. Anti-polymer ELISA data, which include the results for back-
ground (Bkgd), pre-immune serum, subcutaneous (SC), intravenous
(IV), and normal rat serum (positive control: the wells were coated
with rat serum enriched in antibodies, hence resulting in an expected
positive signal). The y axis is the optical density (OD) at l=490 nm
for the oxidation product of ELISA substrate, tetramethylbenzidine
(TMB), which represents the level ofimmunogenic response ofthe
polymer. The first, second, and third sera were taken at the third, sixth,
and ninth week, respectively (n=3). All experimental details are
included in the Supporting Information.
Angew. Chem. Int. Ed. 2005, 44, 6529 –6533
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6531

Communications
polymers are currently being actively pursued as gene
delivery vectors because they can neutralize and condense
DNA into particles capable of undergoing endocytosis.[17]
A
critical issue in developing effective synthetic polymeric
vectors is that competent gene carriers, such as PLL and
polyethyleneimine (PEI), are often cytotoxic.[17,18] As our
hybrid copolymers carry cationic charges at physiological pH
values, and have minimal cytotoxicity, they were evaluated as
vectors for gene delivery. Electrophoretic mobility-shift
assays (EMSAs) indicated that poly-1, poly-2, and poly-3
efficiently complex pSV-b-gal plasmid DNA under physio-
logical conditions. For poly-1 and poly-3, an N/P (ammonium
positive charge on polymer/phosphate negative charge on
DNA) ratio of 1.5 completely retarded the DNA; for poly-2
this ratio was 2(see Supporting Information). The difference
in DNA/polymer binding efficiency is presumably due to the
difference in the molecular weight of the polymer as poly-2
has a slightly shorter chain length than poly-1 and poly-3.
Complex formation occurs largely because of entropic gains
owing to the liberation of smaller counterions along the
macromolecular chains.[19] Thus, as the chain is shortened (as
for poly-2), there is less entropic gain during DNA complex-
ation, which results in slightly weaker binding. The physical
characteristics of the polymer/DNA complexes were then
investigated by using AFM. Each polymer condensed DNA
into spherical nanoparticles with typical diameters of 50–
200 nm (Supporting Information), which is within the normal
size range for cellular internalization.[20]
The transfection efficiency of the three hybrid polymers
was tested and compared with PLL by using a luciferase-assay
kit under serum-free conditions. As PLL and poly-1, poly-2,
and poly-3 have only primary amines and lack other amino
residues to afford proton sponge effects, chloroquin (which is
known to disrupt the membrane of the endosome) was used in
all gene-transfection studies to enhance the endosomal
release after entrance into the cell. Figure 4 summarizes the
gene-transfection efficiency (normalized to the total cellular
protein). Poly-2 and poly-3 showed a significantly higher
transfection ability than PLL at similar N/P ratios. This is due
primarily to the high toxicity of PLL at those concentrations.
The lower transfection efficiency of poly-1 compared to poly-
2 or poly-3 is presumably due to lower local charge density on
poly-1 and the varied nature of the amino groups on the
polymer chain. Poly-2 and poly-3 have both a-amino and
more-flexible e-amino functionalities, whereas poly-1 has
only a-amine groups. It has been reported that very subtle
changes in polymer structure can result in significant changes
in gene–transfection efficiency.[16a] Further structure–property
correlation will be investigated in the future, which will
provide information for structural optimization to improve
the transfection efficiency.
Figure 4. Luciferase gene-transfection data for the hybrid copolymers
with PLL as control. Standard deviations are shown by the error bars
(n=3). Relative light units (RLU) were normalized by using the total
cellular protein in each well. The symbols * and *** indicate statistical
significance at levels of p<0.05 and p<0.001, respectively, for the
experimental polymers and PLL at corresponding N/P ratios. The
p values were obtained by using Student—Newman–Keuls multiple
comparisons testing.
mers are biodegradable, nontoxic, and nonimmunogenic. The
hybrid copolymers were tested as vectors for possible
application in gene delivery. EMSA, AFM, and luciferase-
transfection studies demonstrate that the hybrid copolymers
can efficiently compact plasmid DNA into soluble nano-
particles and be used as safe gene carriers. Given the natural
abundance and functional diversity of saccharides and amino
acids, their biodegradability, low cytotoxicity, and nonimmu-
nogenicity, a diverse family of saccharide–peptide hybrid
polymers are currently under development in our laboratory
for various biomedical applications including gene/drug
delivery and tissue engineering.
Received: June 6, 2005
Revised: July 21, 2005
Published online: September 15, 2005
Keywords: biomaterials · gene technology · polymers · peptides ·
.
saccharides
[1] Biomaterials Science:An Introduction to Materials in Medicine
(Eds.: B. D. Ratner, A. S. Hoffman, F. J. Schoen, J. E. Lemons),
Elsevier, London, England, 1996.
[2] Polymeric Biomaterials, 2nd ed. (Ed.: S. Dumitriu), Marcel
Dekker, Inc., New York, USA, 2002.
In summary, we have described our concept for the design
of saccharide–peptide hybrid copolymers as a new class of
biomaterials. As examples, galactaro–dilysine (poly-1), tri-
lysine (poly-2), and tetralysine(poly-3) hybrid copolymers
were synthesized through interfacial polymerization of a
galactose-derived monomer and corresponding l-lysine-
derived peptide monomers. Enzymatic degradation, MTT
tests, and immunological assays show that the hybrid copoly-
[3] R. Langer, D. A. Tirrell, Nature 2004, 428, 487.
[4] M. Vert, J. Mauduit, S. Li, Biomaterials 1994, 15, 1209.
[5] J. Mauduit, M. Vert, S.T.P. Pharma Sci. 1993, 3, 197.
[6] A. J. Domb, O. Elmalak, V. R. Shastri, Z. Ta-Shma, D. M.
Masters, I. Ringel, D. Teomim, R. Langer, Drug Targeting
Delivery 1997, 7, 135.
[7] F. Rypacek, M. Dvorak, I. Stefko, L. Machova, V. Skarda, D.
Kubies, ACS Symp. Ser. 2001, 786, 258.
[8] J. Kohn, Drugs Pharm. Sci. 1990, 45, 195.
6532
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6529 –6533

Angewandte
Chemie
[9] M. Metzke, J. Z. Bai, Z. Guan, J. Am. Chem. Soc. 2003, 125,
7760.
[10] J. G. Tirrell, D. A. Tirrell, M. J. Fournier, T. L. Mason, Protein-
Based Mater. 1997, 61.
[11] H. Herzner, T. Reipen, M. Schultz, H. Kunz, Chem. Rev. 2000,
100, 4495.
[12] S. A. W. Gruner, E. Locardi, E. Lohof, H. Kessler, Chem. Rev.
2002, 102, 491.
[13] K. Aoi, K. Tsutsumiuchi, M. Okada, Macromolecules 1994, 27,
875.
[14] K. Aoi, K. Tsutsumiuchi, E. Aoki, M. Okada, Macromolecules
1996, 29, 4456.
[15] A. Zelikin, D. Putman, P. Shastri, R. Langer, V. Izumrudov,
Bioconjugate Chem. 2002, 13, 548.
[16] a) S. J. Hwang, N. C. Bellocq, M. E. Davis, Bioconjugate Chem.
2001, 12, 280; b) Y. Liu, L. Wenning, M. Lynch, T. M. Reineke, J.
Am. Chem. Soc. 2004, 126, 7422.
[17] S. C. De Smedt, J. Demeester, W. E. Hennink, Pharm. Res. 2000,
17, 113.
[18] W. Zauner, M. Ogris, E. Wagner, Adv. Drug Delivery Rev. 1998,
30, 97.
[19] Z. Kakizawa, K. Kataoka, Adv. Drug Delivery Rev. 2002, 54, 203.
[20] T. Friedmann, Sci. Am. 1997, 276, 96.
Angew. Chem. Int. Ed. 2005, 44, 6529 –6533
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6533