Biomimetic anchor for surface-initiated polymerization from metal substrates

Article abstract of DOI:10.1021/ja0532638

In this paper, we demonstrate the first use of a catecholic initiator for surface-initiated polymerization (SIP) from metal surfaces to create antifouling polymer coatings. A new bifunctional initiator inspired by mussel adhesive proteins was synthesized,

Full text of DOI:10.1021/ja0532638

Published on Web 10/22/2005
Biomimetic Anchor for Surface-Initiated Polymerization from
Metal Substrates
Xiaowu Fan, Lijun Lin, Jeffrey L. Dalsin, and Phillip B. Messersmith*
Contribution from the Department of Biomedical Engineering, Robert R. McCormick School of
Engineering and Applied Sciences, Northwestern UniVersity, 2145 Sheridan Road,
EVanston, Illinois 60208
Received May 18, 2005; Revised Manuscript Received September 16, 2005; E-mail: philm@northwestern.edu
Abstract: In this paper, we demonstrate the first use of a catecholic initiator for surface-initiated
polymerization (SIP) from metal surfaces to create antifouling polymer coatings. A new bifunctional initiator
inspired by mussel adhesive proteins was synthesized, which strongly adsorbs to Ti and 316L stainless
steel (SS) substrates, providing an anchor for surface immobilization of grafted polymers. Surface-initiated
atom transfer radical polymerization (SI-ATRP) was performed through the adsorbed biomimetic initiator
to polymerize methyl methacrylate macromonomers with oligo(ethylene glycol) (OEG) side chains. X-ray
photoelectron spectroscopy, surface FT-IR, and contact angle analysis confirmed the sequential grafting
of initiator and polymer, and ellipsometry indicated the formation of polymer coatings of up to 100 nm
thickness. Cell adhesion experiments performed with 3T3-Swiss albino fibroblasts showed substantially
reduced cell adhesion onto polymer grafted Ti and 316L SS substrates as compared to the unmodified
metals. Moreover, micropatterning of grafted polymer coatings on Ti surfaces was demonstrated by
combining SI-ATRP and molecular assembly patterning by lift-off (MAPL), creating cell-adhesive and cell-
resistant regions for potential use as cell arrays. Due to the ability of catechols to bind to a large variety of
inorganic surfaces, this biomimetic anchoring strategy is expected to be a highly versatile tool for polymer
thin film surface modification for biomedical and other applications.
Introduction
mers. We recently described the use of DOPA-containing
peptides for adsorption of the antifouling polymer poly(ethylene
Enediol-containing compounds (e.g., catechol, dopamine) are
common in nature, playing key roles in many biological
reactions and in medical treatment. For example, the catecholic
amino acid L-3,4-dihydroxyphenylalanine (L-DOPA) is found
in the adhesive pad proteins secreted by marine mussels.
Although its function is not fully understood, the presence of
L-DOPA at elevated concentrations in proteins found closest to
the adhesive-substrate interface suggests a possible role in
adhesion, a finding that has stimulated great interest in exploiting
catechols for enhancing interfacial adhesion of materials.
7,8
glycol) (PEG) onto Ti and Au surfaces. Although the resulting
surfaces exhibited excellent protein and cell fouling resistance,
adsorption of polymer from solution (so-called “graft-to”
approach) is generally not capable of achieving high polymer
1
2
9
surface densities and thick coatings. In contrast, surface-initiated
polymerization (SIP) of monomers from surface bound initiators
(
“graft-from”) is a more versatile technique capable of producing
3
1
0
polymer brush layers of greater density and thickness.
Although different types of functional groups have been reported
for immobilizing initiators onto metals for SIP, catechols have
never been used in this manner.
4
Catechol derivatives have recently been used to anchor small
functional biomolecules onto ferromagnetic nanoparticles (Fe2O3)
for protein separations, and for linking of DNA and dyes to
We describe in this Article a general approach to SIP of
antifouling polymers from metal surfaces via a mussel adhesive
protein (MAP) mimetic initiator. Our strategy takes advantage
of a bifunctional initiator (2 in Scheme 1) containing a catechol
end group for surface anchoring and an alkyl bromine to activate
SIP. We demonstrate this approach by grafting PEG-based
antifouling coatings on Ti and stainless steel (SS), two important
metals commonly used in the fabrication of medical devices
5
6
surfaces of semiconductor nanoparticles (TiO2).
However, catechols have been underutilized as tools for
modification of metal surfaces with synthetic functional poly-
(
1) Goldstein, D. S., Eisenhofer, G., McCarty, R., Eds.; Catecholamines:
Bridging Basic Science with Clinical Medicine; Adv. Pharmacol.; Academic
Press: San Diego, 1998; Vol. 42.
(
2) Waite, J. H.; Tanzer, M. L. Science 1981, 212, 1038-1040.
(
3) (a) Floriolli, R. Y.; von Langen, J.; Waite, J. H. Mar. Biotechnol. 2000, 2,
3
52-363. (b) Waite, J. H.; Qin, X. Biochemistry 2001, 40, 2887-2893.
(7) Dalsin, J. L.; Hu, B.-H.; Lee, B. P.; Messersmith, P. B. J. Am. Chem. Soc.
2003, 125, 4253-4258.
(
(
4) Yu, M.; Deming, T. J. Macromolecules 1998, 31, 4739-4745.
5) Xu, C.; Xu, K.; Gu, H.; Zheng, R.; Liu, H.; Zhang, X.; Guo, Z.; Xu, B. J.
Am. Chem. Soc. 2004, 126, 9938-9939.
(8) (a) Dalsin, J.; Tosatti, S.; V o¨ r o¨ s, J.; Textor, M.; Messersmith, P. B. Langmuir
2005, 21, 640-646. (b) Statz, A. R.; Meagher, R. J.; Barron, A. E.;
Messersmith, P. B. J. Am. Chem. Soc. 2005, 127, 7972-7973.
(9) Prucker, O.; Ruhe, J. Langmuir 1998, 14, 6893-6898.
(10) (a) Advincula, R. C. J. Dispersion Sci. Technol. 2003, 24, 343-361. (b)
Zhao, B.; Brittain, W. J. Prog. Polym. Sci. 2000, 25, 677-710.
(6) (a) Rajh, T.; Saponjic, Z.; Liu, J.; Dimitrijevic, N. M.; Scherer, N. F.; Vega-
Arroyo, M.; Zapol, P.; Curtiss, L. A.; Thurnauer, M. C. Nano Lett. 2004,
4
, 1017-1023. (b) Rice, C. R.; Ward, M. D.; Nazeeruddin, M. K.; Gratzel,
M. New J. Chem. 2000, 24, 651-652.
10.1021/ja0532638 CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 15843-15847
9
15843

A R T I C L E S
Fan et al.
Scheme 1. Synthesis and Anchoring of the DOPA Mimetic Initiator and Subsequent SI-ATRP
Table 1. Surface Analysis Results of Bare Ti, Initiator Modified Ti,
and 4-9 Grafted Ti (3 h SI-ATRP) Samples
due to their high specific strength, corrosion resistance, and
1
1
biocompatibility. The polymers are grown in situ from the
initiator modified surfaces by surface-initiated atom transfer
radical polymerization (SI-ATRP), which is a versatile and
XPS atomic concentration (%)
static water
sample
[Ti]
[O]
[C]
[N]
[Br]
contact angle (deg)
1
2
bare Ti
59
30
11
0
0
<10
56 ( 2
48 ( 2
robust approach widely used in “graft-from” strategies.
initiator modified 21.1 41.6 33.4 2.5 1.4
-9 30.4 69.6
Moreover, to demonstrate the potential applications of our
approach for patterning purposes, we also prepared patterned
thin films and subsequent cellular arrays using the combination
of SI-ATRP and a simple photolithography method, molecular
assembly patterning by lift-off (MAPL).13
4
0
0
0
Results and Discussion
As illustrated in Scheme 1, dopamine was protected in situ
as the borate ester using sodium borate decahydrate and then
reacted under mildly basic conditions with 2-bromopropionyl
bromide to yield initiator 2. In the design of our biomimetic
initiator, rather than using the hydrolytically labile bromoester
functionality, which is widely employed in ATRP initiators, we
coupled the catechol anchor and Br terminus using an amide
linkage that should remain stable under the aqueous conditions
of initiator adsorption, SI-ATRP, and in vitro cell adhesion test.
1
4
Initiator 2 was immobilized on Ti and 316L SS surfaces via
adsorption from aqueous solution. Initiator modified substrate
3
was then subjected to SI-ATRP of oligo ethylene glycol
methyl ether methacrylate (OEGMEMA) macromonomers with
OEG side chains of length 4 or 9 EG repeats (OEGMEMA-
Figure 1. XPS survey spectra of bare, initiator, and 4-9 coated Ti surfaces.
A high-resolution spectrum of the C 1s region of 4-9 coated Ti is shown in
the inset.
4
/-9). The grafted poly(OEGMEMA) coatings 4-4 (4 EG
repeats) and 4-9 (9 EG repeats) were analyzed by X-ray
photoelectron spectroscopy (XPS), ellipsometry, contact angle
goniometry, and polarization-modulation infrared reflection-
adsorption spectroscopy (PM-IRRAS).
Successful initiator immobilization and polymer grafting were
first ascertained by XPS and contact angle measurements. Figure
contamination. Successful anchoring of initiator 2 was indicated
by a significant increase in C concentration (33.4%) and the
presence of N and Br signals not detected for the bare Ti
substrate. After 3 h SI-ATRP of OEGMEMA-9, polymer
grafting was confirmed by a substantial increase in C concentra-
tion (69.6%) and the complete disappearance of Ti signal,
indicating the thickness of the grafted 4-9 layer (∼80 nm, see
ellipsometry results below) was greater than the escape depth
1
displays XPS survey spectra of unmodified, initiator-treated,
and polymer grafted Ti substrates. Surface chemical composition
and static water contact angle data of these surfaces are
summarized in Table 1. The pristine Ti surface was mainly
composed of Ti and O from the native surface oxide as well as
a small amount of C resulting from adventitious hydrocarbon
9
of the photoelectrons (∼10 nm). It should be noted that the
[O]/[C] ratio determined by XPS (0.44) was very close to the
theoretical value (0.48) of 4-9 polymer. Furthermore, high-
resolution XPS spectral analysis demonstrated good agreement
with the structure of the grafted polymer coating. As shown in
the inset of Figure 1, the C 1s high-resolution scan of 4-9 grafted
Ti surface was fitted with aliphatic (C-C/C-H), ether (C-O),
and ester (OdC-O) carbons corresponding to binding energies
at 284.6, 286.1, and 288.6 eV, respectively. The observed area
(
11) Brunette, D. M.; Tengvall, P.; Textor, M.; Thomsen, P. Titanium in
Medicine-Material Science, Surface Science, Engineering, Biological
Responses and Medical Applications; Springer-Verlag: New York, 2001.
12) Pyun, J.; Kowalewski, T.; Matyjaszewski, K. Macromol. Rapid Commun.
(
(
(
2
003, 24, 1043-1059 and references therein.
13) Falconnet, D.; Pasqui, D.; Park, S.; Eckert, R.; Schift, H.; Gobrecht, J.;
Barbucci, R.; Textor, M. Nano Lett. 2004, 4, 1909-1914.
14) Note that the native oxide layer is the outmost surface.
15844 J. AM. CHEM. SOC.
9
VOL. 127, NO. 45, 2005

Anchor for Polymerization from Metal Substrates
A R T I C L E S
Figure 4. 4 h cell adhesion assay results on bare and 4-9 grafted Ti and
Figure 2. PM-IRRAS spectrum of a 4-4 sample produced by 12 h SI-
ATRP of OEGMEMA-4.
316L SS substrates (*, p < 0.001 versus bare Ti).
Scheme 2. MAPL Process for the Preparation of Patterned
Polymer Thin Film by SI-ATRP13
Figure 3. Dry ellipsometric thickness of 4-4 thin films on Ti substrates as
a function of polymerization time. Samples were prepared from separate
yet identical reactions, and data points represent mean values of at least
three different measurements.
peak at 1738 cm- from CdO stretch in the ester group, and
peaks at 1360-1470 cm from CH2 and CH3 deformation.
1
ratio of these components, 3.1:17.2:1, is consistent with the
theoretical ratio 3:19:1 for 4-9 polymer. Similarly consistent
XPS results were also obtained from 4-9 coatings grafted on
-
1
Other features associated with the OEG moiety include bands
-
1
at 1070-1260 cm from C-O-C stretches in both ether and
3
16L SS and 4-4 coatings grafted on Ti. Contact angle changes
-1
ester groups, and a small peak at 2825 cm due to C-H stretch
also confirmed the sequential formation of the initiator and
polymer layers. Prior to initiator adsorption and SI-ATRP,
freshly cleaned Ti substrates were hydrophilic with a static water
contact angle <10°, which increased to an average value of
1
7
in the methyl ether end group. Moreover, the absence of the
-
1
absorption from the vinyl group (a peak at 1640 cm in the
spectrum of the monomer, not shown) further excluded the
possibility of simple monomer physisorption on the substrate.
5
6° after immobilization of 2. After 3 h SI-ATRP of OEG-
Ellipsometry was used to follow the increase in polymer thin
film thickness during SI-ATRP. Figure 3 shows the growth of
4-4 as a function of polymerization time. Under the conditions
of our experiments, the initial kinetics of aqueous SI-ATRP was
very fast, resulting in a polymer thin film of ∼30 nm thickness
after only 5 min of polymerization, after which the thickness
further increased with time to a limiting value of roughly 100
nm for extended reaction times up to 12 h. Growth of 4-9
coatings on Ti showed very similar behavior. The rapid initial
rate can be attributed to the solvent (water), which is believed
to increase the activity of the catalyst and accelerate ATRP.18
MEMA-9, the contact angle decreased to ∼48°, indicating an
increase in hydrophilicity due to the grafted polymer. The
observed XPS and water contact angle data of grafted polymer
surfaces are very consistent with those reported for similar PEG-
based polymer coatings grafted from Si substrates.1
5,16
PM-IRRAS was employed to further confirm the presence
of grafted coatings by observation of the absorptions of typical
functional groups contained in the polymer. Figure 2 demon-
strates the PM-IRRAS spectrum of 4-4 after 12 h SI-ATRP of
OEGMEMA-4. Typical features of the methyl methacrylate
polymer backbone include strong absorptions at 2850-2990
-1
cm due to C-H symmetric and asymmetric stretching, a sharp
(
17) (a) Mougin, K.; Lawrence, M. B.; Fernandez, E. J.; Hillier, A. C. Langmuir
2004, 20, 4302-4305. (b) Lambert, B. J.; Shurvell, H. F.; Lightner, D. A.;
Cooks, R. G. Organic Structural Spectroscopy; Prentice Hall: Upper Sandle
River, NJ, 1998; Chapter 8, pp 189-192.
(
15) Xu, F. J.; Zhong, S. P.; Yung, L. Y.; Kang, E. T.; Neoh, K. G.
Biomacromolecules 2004, 5, 2392-2403.
(16) Yu, W. H.; Kang, E. T.; Neoh, K. G.; Zhu, S. J. Phys. Chem. B 2003, 107,
1
0198-10205.
(18) Wang, X.-S.; Armes, S. P. Macromolecules 2000, 33, 6640-6647.
J. AM. CHEM. SOC.
9
VOL. 127, NO. 45, 2005 15845

A R T I C L E S
Fan et al.
Figure 5. Patterned polymer thin film created by MAPL and SI-ATRP. (a) ToF-SIMS chemical map of Ti+ (m/z ) 47.89) signal collected from a patterned
4
-9 thin film produced by MAPL and 12 h SI-ATRP. Scale bar ) 100 µm. (b) Fluorescence micrograph of fibroblast attachment on a patterned 4-9 surface
after 4 h of cell culture. Scale bar ) 50 µm. It should be noted that these two images were taken at different locations from samples that contained patterns
of variable sizes and shapes.
Similar kinetic results for SI-ATRP in pure aqueous or aqueous
alcoholic media have been observed by other investigators.
Efforts to achieve polymer coatings with better thickness control
are underway.
compared (data not shown) indicated that, for polymer coatings
with thickness ∼30 nm or higher, short term cell resistance did
not strongly depend on either the thickness or the OEG length
of grafted polymer layers.
1
9,20
The polymer coatings reported here are significantly thicker
than those formed by adsorption of DOPA functionalized
polymers from solution (“graft-to” approach) onto similar
The short-term antifouling behavior of our polymer grafted
metal substrates is similar to that of PEG-based polymer films
19
15
formed by SI-ATRP on Au and Si surfaces. The observed
97% reduction in cell adhesion density on Ti is also
7
,8
substrates. However, the nature of the chemical interaction
between grafted polymer and underlying substrate is expected
to be similar for both “graft-from” and “graft-to” approaches.
On Ti, the interaction is likely to take the form of a bidentate
coordination complex between the catechol oxygens and a Ti
∼
comparable to that of our previously reported antifouling PEG-
DOPA thin films (∼94% reduction in cell adhesion density)
formed by polymer adsorption onto Ti (“graft-to” approach).7
Although a more detailed study looking at long-term perfor-
mance of these coatings is in progress, a preliminary long-term
cell adhesion experiment with 4-9 showed promising extended
antifouling performance, with 4-9 coatings of approximately 100
nm thickness exhibiting roughly 90% less cell attachment than
uncoated Ti at 5 weeks of culture (data not shown).
21,22
atom at the native oxide surface.
Binding energies between
TiO2 and dopamine have recently been estimated by density
2
3
functional theory to be on the order of 25-30 kcal/mol,
suggesting that grafted polymer chains are strongly bound to
the metal surface. Although previous studies demonstrated that
similar interactions occur between catechols and Fe2O3 sur-
Finally, the patterning of polymer thin films was ac-
2
2
faces, the situation on 316L SS is potentially much more
complicated, as the SS surface is typically composed of mixed
Fe and Cr oxides/hydroxides.24
complished by the combination of SI-ATRP and MAPL.13
A
pattern containing circular domains of photoresist on Ti was
created using standard photomask lithography, after which
initiator was adsorbed to bare Ti regions (Scheme 2). Following
removal of the photoresist by rinsing with solvent, 12 h SI-
ATRP yielded circular regions of bare Ti within a background
of grafted 4-9. Patterned surfaces were analyzed by time-of-
flight secondary ion mass spectroscopy (ToF-SIMS). Figure 5a
shows a ToF-SIMS chemical map of a patterned polymer sample
where the bright circular regions represent bare Ti substrate that
was protected by photoresist during initiator adsorption, whereas
the dark matrix is the 4-9 polymer thin film grafted from the
initiator layer after photoresist lift-off. Patterned Ti samples were
then subjected to cell culture experiments as described above.
Figure 5b shows a typical florescence image of fibroblast
attachment on a sample patterned with 4-9 polymer. It was
observed that a cell pattern was produced in which 1-4 cells
were confined within the circular regions of bare Ti, while the
grafted polymer region resisted cell attachment in a spatially
controlled manner. This result demonstrates the potential of
using our biomimetic strategy to generate cellular arrays on
metal substrates.
The biological response to polymer grafted metal surfaces
was assessed by studying resistance to mammalian cell adhesion
in the presence of serum. Figure 4 shows the comparative results
from cell adhesion assays in which 3T3-Swiss albino fibroblasts
were cultured in the presence of unmodified and 4-9 grafted
metal substrates after 12 h SI-ATRP. While cells readily attached
after 4 h on bare Ti and 316L SS at average densities of
2
approximately 40 and 50 cells/mm , respectively, 4-9 grafted
2
surfaces supported attachment of around 1 cell/mm on Ti, and
no cells were found on 316L SS. Results from additional
experiments in which both thickness and OEG chain length were
(
19) Ma, H.; Hyun, J.; Stiller, P.; Chilkoti, A. AdV. Mater. 2004, 16, 338-341.
20) (a) Huang, W.; Kim, J.-B.; Bruening, M. L.; Baker, G. L. Macromolecules
(
2
002, 35, 1175-1179. (b) Jones, D. M.; Huck, W. T. S. AdV. Mater. 2001,
1
3, 1256-1259.
(
21) (a) Connor, P. A.; Dobson, K. D.; McQuillan, A. J. Langmuir 1995, 11,
4
193-4195. (b) Rajh, T.; Chen, L. X.; Lukas, K.; Liu, T.; Thurnauer, M.
C.; Tiede, D. M. J. Phys. Chem. B 2002, 106, 10543-10552.
22) Chen, L. X.; Liu, T.; Thurnauer, M. C.; Csencsits, R.; Rajh, T. J. Phys.
Chem. B 2002, 106, 8539-8546.
(
(
(
23) Vega-Arroyo, M.; LeBreton, P. R.; Rajh, T.; Zapol, P.; Curtiss, L. A. Chem.
Phys. Lett. 2005, 406, 306-311.
24) Olsson, C.-O. A.; Landolt, D. Electrochim. Acta 2003, 48, 1093-1104.
15846 J. AM. CHEM. SOC.
9
VOL. 127, NO. 45, 2005

Anchor for Polymerization from Metal Substrates
A R T I C L E S
The vast majority of previous SI-ATRP studies on inorganic
materials have utilized silane or thiol initiator designs.
into cell-adhesive and cell-resistant regions. We believe that
the antifouling properties of the thick polymer films, combined
with the structural and chemical versatility of the SI-ATRP
approach, suggest enormous potential in a wide variety of
biomedical applications where cell fouling resistance and cell
arrays are desired. Recent preliminary results demonstrate
successful polymer grafting from several other metal substrates
including Ti alloys (Ti-6Al-4V, NiTi) and Au, which suggests
this biomimetic strategy represents a general approach to
polymer functionalization of many substrate materials. For
nonmedical applications, this facile and versatile synthetic
strategy should open a new route toward functional thin films
and polymer-metal oxide nanocomposites via SIP.
2
5
Although well-known for strong interactions with inorganic
surfaces, catechols have not been used for SIP before. Thus,
the catechol-based initiator introduced in this paper adds a new
tool that can be used to modify inorganic surfaces with polymers
grafted by SI-ATRP. Although we have demonstrated this with
OEG-based methacrylate monomers for antifouling coatings,
simple extension of this approach to any monomer and solvent
system capable of being used in ATRP will expand the practical
uses to include polymer functionalization for many other
applications and types of materials, including inorganic nano-
particles. It should also be straightforward to generalize the
design of the catechol initiator from the 2-bromoamide used
here to functional groups suitable for other types of radical
polymerization techniques.
Acknowledgment. We thank NASA (NCC-1-02037) and the
NIH (DE14193) for financial support. Dr. N. Wu is greatly
appreciated for his assistance and discussions on XPS, PM-
IRRAS, and ToF-SIMS analysis performed at KeckII/NUANCE
at Northwestern University. Prof. R. P. H. Chang and H. Zhang
at Northwestern University are acknowledged for their help on
ellipsometry measurements.
Summary
In summary, we have demonstrated a new approach to SIP
from metal surfaces that takes advantage of strong interactions
between a catechol-based initiator and native metal oxides. The
grafted polymer coatings were up to ∼100 nm thick, resistant
to fouling by mammalian cells, and capable of being patterned
Supporting Information Available: Experimental details.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(
25) Pyun, J.; Kowalewski, T.; Matyjaszewski, K. Polymer Brushes by Atom
Transfer Radical Polymerization. In Polymer Brushes: Synthesis, Char-
acterization, Applications; Advincula, R. C., Brittain, W. J., Caster, K. C.,
Ruhe, J., Eds.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim,
Germany, 2004; pp 51-68.
JA0532638
J. AM. CHEM. SOC.
9
VOL. 127, NO. 45, 2005 15847