An electroactive catalytic dynamic substrate that immobilizes and releases patterned ligands, proteins, and cells

Article abstract of DOI:10.1002/anie.200800166

(Figure Presented) Dynamic release: A quinone-terminated self-assembled monolayer captures and releases ligands and cells in situ through a pH-dependent electrochemical potential. The surface is catalytic for multiple rounds of immobilization/release. When combined with a photochemical approach, peptide ligands that mediate cell attachment are captured and released on defined gradient patterns (see picture).

Full text of DOI:10.1002/anie.200800166

Angewandte
Chemie
DOI: 10.1002/anie.200800166
Electroactive Monolayers
An Electroactive Catalytic Dynamic Substrate that Immobilizes and
Releases Patterned Ligands, Proteins, and Cells**
Eugene W. L. Chan, Sungjin Park, and Muhammad N. Yousaf*
The ability to spatially and temporally control the attachment
and detachment of molecules and cells on solid supports is
important to research areas ranging from fundamental cell
biology to biomaterial development and heterogeneous
catalysis.[1–5] Various approaches have been developed to
promote or inhibit cell attachment by altering the macro-
scopic properties of the materials in situ, including mechan-
ical stretching,[6–8] photochemical illumination,[9–11] electro-
chemical modulation,[12,13] and thermal activation.[14–16] Alter-
natively, other strategies have focused on manipulating the
cell–surface interactions at the molecular level by incorpo-
rating specific chemistries that can alter ligand presentation
for attached cell culture through a noninvasive external
switch.[17–23] To access more sophisticated and complex cell
behavior studies, a surface strategy that can dynamically
modulate attached cell culture at the molecular level catalyti-
cally, with spatial and temporal control in patterns and
gradients, would greatly extend the utility of these model
surfaces for a variety of cell motility, cell signaling, and cell–
cell communication studies. These surfaces may also lead to
the development of renewable surfaces for synthetic organic
chemistry applications ranging from new solid-phase peptide
synthesis resins to heterogeneous catalysis.
Herein, we report an electroactive quinone-terminated
self-assembled monolayer (SAM) on gold that captures and
subsequently releases ligands, proteins, and cells in situ
through an electrochemical potential. We also show that the
surface is catalytic for multiple rounds of immobilization and
release that are pH dependent. From a synthetic organic
chemistry perspective, a clean and quantitative functional-
group transformation occurs from an oxyamine group to a
primary alcohol upon mild electrochemically applied poten-
tial. Furthermore, by extending this strategy with a photo-
chemical approach, we demonstrate the immobilization and
release of peptide ligands that mediate cell attachment in
defined gradient patterns on inert surfaces.
Our approach is based on a redox-active hydroquinone-
terminated SAM that can be electrochemically oxidized to
the corresponding quinone (Figure 1). The resulting quinone
Figure 1. Interfacial reaction between soluble oxyamine and quinone-
terminated SAMs. Electrochemical oxidation [O] of the mixed mono-
layers presenting hydroquinone and tetraethylene glycol groups con-
verts the hydroquinone into the corresponding quinone. This quinone
then reacts selectively with a soluble oxyamine-tagged ligand (R-
ONH2) to give the redox-active oxime conjugate on the surface. The
oxime is chemically stable and undergoes reversible redox coupling in
HClO4 (1m, pH 0). Electrochemical reduction [R] of the monolayer in
buffer solution at pH>0 spontaneously reverts the oxime to the
hydroquinone by release of the surface-bound ligand as a hydroxy
group.
monolayer permits the selective coupling of soluble oxyamine
ligands to the surface via an oxime conjugate. The oxime
conjugate is chemically stable at pH 1–14, but upon applica-
tion of an applied potential it can undergo a subsequent
reversible redox reaction at low pH (1m HClO4, pH 0) with
the ligand still covalently bound. However, electrochemical
reduction of the monolayer at a higher pH value (phosphate-
buffered saline (PBS), pH 7) reverts the oxime to the
hydroquinone and releases the ligand from the surface. The
oxyamine terminal group on the ligand is converted to a
hydroxy group and the surface is regenerated to the catalytic
hydroquinone form for subsequent immobilization and
release.
This strategy possesses several unique features that are
important in the preparation of model substrates for modu-
lating cell attachment. First, the immobilization and subse-
quent release of ligands is controlled by mild electrochemical
potentials under physiological conditions (PBS, pH 7), and
therefore permits the modulation of ligands in the presence of
attached cell culture.[24] Second, the electroactive quinone and
oxime monolayers permit quantitative characterization of the
[*] Dr. E. W. L. Chan, S. Park, Prof. M. N. Yousaf
Department of Chemistry and The Carolina Center for Genome
Science
University of North Carolina at Chapel Hill
Chapel Hill, NC 27599-3290 (USA)
Fax: (+1)919-962-2388
E-mail: mnyousaf@email.unc.edu
Homepage: http://www.chem.unc.edu/people/faculty/yousafmn/
mnyindex.ht
[**] This work was supported by the Carolina Center for Cancer
Nanotechnology Excellence, The National Cancer Institute, and the
Burroughs Wellcome Foundation (Interface Career Award).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200800166.
Angew. Chem. Int. Ed. 2008, 47, 6267 –6271
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6267

Communications
extent of ligand immobilization and release by cyclic voltam-
metry (CV).[25] Third, oxyamine moieties can be introduced
into a variety of ligands through straightforward solution- and
solid-phase synthesis to further extend the utility of these
model substrates for other biological studies.[26] A potentially
wide range of soluble aminooxy-terminated ligands can be
coupled to the quinone surface by the oximation reaction, and
subsequently released from the surface electrochemically.
Finally, the hydroquinone surface acts as a catalyst for the
immobilization and release of ligands and also provides a
route for a mild functional-group transformation of oxyamine
groups to hydroxy groups. Because the surface is renewable,
this approach greatly simplifies the preparation of monolayer
substrates for both presenting biologically active ligands and
dynamically modulating the activity of immobilized ligands.
We first prepared monolayer surfaces presenting both
hydroquinone (1) and tetraethylene glycol (2) groups
(Scheme 1), and characterized the interfacial immobilization
Figure 2. Electrochemical characterization of the reaction between the
quinone monolayer and soluble aminooxyacetic acid by CV. A) The
hydroquinone-terminated monolayer undergoes electrochemical oxida-
tion [O] at 539 and 600 mV and reduction [R] at 322 and 95 mV in 1m
HClO4 and PBS(pH 7.4), respectively. B) The quinone monolayer
reacts with soluble aminooxyacetic acid to form the chemically stable
oxime on the surface. The oxime undergoes reversible redox coupling
at 624 and 484 mV in 1m HClO4. C) Consecutive cyclic voltammo-
grams of the oxime monolayer in PBSshow the breakdown of the
oxime, release of ligand, and formation of the hydroquinone in situ.
D) The resulting hydroquinone is electrochemically oxidized to the
quinone. Subsequent immobilization of the monolayer with soluble
aminooxyacetic acid regenerates the oxime conjugate on the surface.
All voltammograms were recorded at a scan rate of 50 mVsꢀ1; the
intersection of the crosshairs represents zero current.
Scheme 1. Surface groups used in this study.
and subsequent release of aminooxyacetic acid by CV.
Figure 2A shows the cyclic voltammograms of a mixed
monolayer presenting both the hydroquinone and tetraethy-
lene glycol groups (1:1) in 1m HClO4 and PBS. The hydro-
quinone undergoes a reversible two-electron, two-proton
process to give the corresponding quinone at different redox
potentials in the two buffer solutions. Selective immobiliza-
tion of soluble aminooxyacetic acid (0.2m in H2O, 2 h) to the
quinone monolayer resulted in the formation of a redox-
active oxime on the surface. The changes in the redox-active
peaks for the cyclic voltammogram in 1m HClO4 correspond
to the oxidation and reduction of the oxime monolayer
(Figure 2B). The oxime is stable in 1m HClO4 and can be
cycled between the oxidized and reduced forms at least 50
times with no change in the voltammograms.
When the same electrochemical experiment was repeated
in PBS, consecutive voltammograms showed a decrease in
peak currents for the oxime monolayer, and an increase in
peak currents corresponding to the hydroquinone–quinone
redox couples (Figure 2C). This result suggests that the oxime
undergoes electrochemical reduction to the corresponding
hydroquinone by selectively releasing the ligand under
physiological conditions (pH 7). To verify that the resulting
peaks are characteristics of the oxidation of the hydroquinone
and reduction of the quinone, we reimmobilized soluble
aminooxyacetic acid (0.2m in water, 2 h) on the monolayer. A
cyclic voltammogram showed diagnostic peaks at 624 and
484 mV in 1m HClO4, and confirmed the oxime conjugate on
the surface (Figure 2D).
6268 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 6267 –6271

Angewandte
Chemie
To determine the chemical nature of the ligand released
from the surface, we synthesized a soluble oxime conjugate
(3) by reacting benzoquinone with n-decanyl oxyamine. The
oxime was chemically reduced in 1m ascorbic acid. Character-
ization by 1H NMR and IR spectroscopy identified the
released ligand as n-decanol. This result suggests that electro-
chemical reduction of the oxime monolayer in PBS regener-
ates the hydroquinone by releasing the ligand as a hydroxy
moiety. X-ray photoelectron spectroscopy of the surface after
release shows no nitrogen present on the surface. We believe
that nitrogen in the form of ammonia (NH3) is also released
during the electrochemical reduction, but have not yet been
able to identify this species.
To demonstrate the utility of this methodology for the
release of biological ligands, we used the association of FLAG
antibody to a monolayer presenting surface-immobilized
FLAG peptides.[27] We prepared a mixed monolayer present-
ing 1% hydroquinone groups and 99% tetraethylene glycol
groups. A high percentage of background tetraethylene glycol
density ensures that the monolayer surface completely resists
nonspecific protein adsorption.[28] The hydroquinone mono-
layer was electrochemically oxidized at 750 mV for 10 s in
PBS to give the corresponding quinone. We next immobilized
an aminooxy-terminated FLAG peptide (4) onto the mono-
layer (0.1m solution in PBS, 4 h). Association of FLAG
antibody to the peptide-immobilized monolayer substrate
was characterized by surface plasmon resonance (SPR)
spectroscopy.
To further characterize the kinetics of oxime degradation
in PBS, we examined the cyclic voltammograms obtained for
the interfacial reaction. Figure 3 shows a plot of the normal-
Figure 4 (top) shows the SPR sensorgram for the binding
of anti-FLAG (0.1 mgmLꢀ1) onto the monolayer presenting
immobilized FLAG peptides. When the peptide-functional-
ized monolayer substrate was electrochemically reduced at
Figure 3. Normalized peak currents (Ip/Io) at 425 mV versus time for
the consecutive cyclic voltammograms shown in Figure 2C. Inset: Plot
of pseudo-first-order rate constants versus pH, which shows the
relative rates of oxime degradation in various buffer solutions.
ized peak currents at 425 mV versus time for the data that
correspond to the loss of oxime on the surface illustrated in
Figure 2C. The data were fitted to an exponential decay
[Eq. (1)] to give a pseudo-first-order rate constant of 0.01 sꢀ1
in PBS, where It is the peak current at time t, I0 is the initial
peak current, and If is the residual nonfaradaic current.
Figure 4. SPR spectroscopy demonstrating the association and release
of anti-FLAG. A monolayer presenting surface-immobilized FLAG
peptide through oxime conjugation before (top) and after (bottom)
electrochemical (EChem) reduction, which releases the bound peptide.
The 700 R.U. in anti-FLAG binding from the baseline before release of
peptide shows biospecific association of the anti-FLAG. After release
of the peptide no anti-FLAG is bound to the surface.
It ¼ If þ ðI0ꢀIfÞ expðꢀk0tÞ
ð1Þ
We repeated the same electrochemical experiment in
various buffer solutions to determine the effect of pH on the
rate of oxime degradation and therefore release from the
surface. Figure 3 (inset) shows a plot of the observed first-
order rate constants (k’) versus pH. The data show a pH
dependence on the rate of oxime degradation. Interestingly,
the observed first-order rate constants increase from pH 0 to
3, and then decrease from pH 3 to 7. Although the reaction
mechanism for the oxime degradation is complex and remains
unclear, a break in the observed first-order rate constants at
pH 3 suggests a change in the rate-determining step at this
buffer pH. Note that the oxime conjugate is stable in these pH
ranges and only becomes unstable, and therefore releases,
upon application of a reductive potential, thus enhancing its
ability for dynamic surface applications.
ꢀ50 mV for 5 min in PBS causing release of peptide, a SPR
sensorgram showed that the anti-FLAG did not bind to the
surface (Figure 4, bottom). This result confirmed that electro-
chemical reduction of the monolayer released the immobi-
lized FLAG peptide ligands from the surface through oxime
degradation.
We next extended this methodology to cell biological
applications by releasing peptide ligands that support cell
adhesion in patterns on monolayer surfaces. To prepare
surfaces that can capture and release patterned cells, we
introduced a photochemical strategy that permitted the
selective immobilization and release of ligands in defined
patterns. Monolayers presenting nitroveratryloxycarbonyl
(NVOC) hydroquinone groups undergo photochemical
Angew. Chem. Int. Ed. 2008, 47, 6267 –6271
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6269

Communications
deprotection to reveal the hydroquinone in the selected
After the hexadecanethiols were printed in line patterns, the
remaining bare gold region was backfilled with a mixed
monolayer presenting 1% NVOC-protected hydroquinone
(5) and 99% tetraethylene glycol groups. A photomask
consisting of gradient patterns was placed in direct contact
with the monolayer. Subsequent UV illumination of the
monolayer through a photomask revealed the hydroquinone
in select regions on the surface. The substrate was electro-
chemically oxidized and then treated with soluble RGD
oxyamine (0.1m, 4 h) to form the corresponding peptide
oxime conjugate on the surface. Swiss 3T3 fibroblasts were
added to the resulting substrate and attached exclusively to
both the microcontact-printed region (hydrophobic and
therefore cell-adhesive) and the photoactivated gradient
region of the monolayer (presenting adhesive RGD peptides;
Figure 5H).
region upon UV illumination through a photomask. This
photochemical approach permits the patterning of a variety of
soluble oxyamines onto the quinone monolayer.[29]
Figure 5 shows the photopatterning of attached cell
culture on surface-immobilized RGD gradients, and subse-
To selectively release the surface-immobilized RGD
ligands, a reductive potential of ꢀ50 mV was applied for
1 min in serum-free medium. Cells patterned on the gradient
began to adopt a more rounded morphology and then
detached from the surface. Figure 5I shows a phase-contrast
image of patterned cells after the electrochemical treatment.
Cells on the gradient pattern were released, whereas cells on
the microcontact-printed hydrophobic lines remained
attached. This result confirmed that electrochemical reduc-
tion of the monolayer released the RGD ligands in situ, and
therefore caused the cells to detach from the otherwise inert
surface as a result of a lack of surface-bound adhesive RGD
ligands. We also showed the release of patterned cells on
gradients overlapping with microcontact-printed hydrophobic
regions, to demonstrate that this methodology can also be
used for the spatial and temporal control of cell–cell
interactions and co-cultures (see the Supporting Informa-
tion).
In conclusion, we have developed a general methodology
to immobilize oxyamine ligands on an electroactive quinone
monolayer, and subsequently release the same ligands from
the surface to regenerate the hydroquinone monolayer by
electrochemical reduction. The redox activity between the
quinone and oxime groups permits characterization of each
step of the interfacial immobilization and release quantita-
tively by CV in real time. The hydroquinone surface is
catalytic, can perform several rounds of immobilization and
release of ligands, and also converts the oxyamine functional
group to a hydroxy group by a mild electrochemical potential.
This selective functional-group transformation may be used
for applications ranging from solid-phase peptide synthesis
and heterogeneous catalysis to chemical-based sensor am-
plification. Furthermore, we have extended this methodology
to modulate the activity of immobilized peptide ligands to
promote or inhibit the selective binding of protein. Finally, by
combining this electrochemical strategy with a photochemical
approach, we have demonstrated the immobilization and
subsequent release of peptide ligands that mediate cell
attachment in defined gradient patterns on inert surfaces.
The examples demonstrated herein present a molecular level
of control over ligand presentation for modulating cell
behavior on model surfaces. We believe this methodology
will provide a broad range of tailored substrates for new
Figure 5. Electrochemical release of photopatterned cells adhered to
RGD gradient SAMs. A) Hexadecanethiols were microcontact printed
to generate hydrophobic line patterns on the gold-coated glass
substrate. B) The remaining bare gold region was backfilled with a
mixed monolayer presenting both the NVOC hydroquinone and tetra-
ethylene glycol groups. C) UV illumination through a photomask
deprotected the NVOC groups to reveal the hydroquinone in select
regions on the monolayer surface. D) The substrate was oxidized to
convert the hydroquinone to the corresponding quinone. Addition of
soluble RGD-oxyamine installs the peptide on the quinone monolayer
through oxime formation. The resulting peptide oxime conjugate alters
the inert photopattern area to biospecific cell adhesive. E) Addition of
fibroblasts to the monolayer substrate resulted in cells adhering to
both microcontact-printed and photopatterned regions. F) Mild elec-
trochemical reduction of the gold substrate causes selective release of
cells from only the RGD-defined gradient, whereas cells attached to
the hydrophobic SAMs remain adherent. G) Micrograph of a photo-
mask with a gradient used in the preparation of the photopatterned
RGD peptide ligands. H) Image showing patterned fibroblasts on a
RGD gradient and on microcontact-printed line patterns. I) Electro-
chemical reduction of the monolayer leads to cell detachment on the
gradient by release of the peptide ligand, while cells patterned on
hydrophobic lines remain attached.
quent release of the patterned cells by an electrochemical
potential.[30] To demonstrate that electrochemical treatment
of the monolayer is noncytotoxic to the attached cell culture,
we used microcontact printing (mCp) to pattern hexadecane-
thiols that promote adhesion of another subset of cells
through hydrophobic interactions on the same substrate.[31]
6270 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 6267 –6271

Angewandte
Chemie
[18] W.-S. Yeo, M. Mrksich, Langmuir 2006, 22, 10816 – 10820.
fundamental studies in attached cell culture and applications
in synthetic chemistry and biotechnology.[32]
[19] W.-S. Yeo, M. Mrksich, Adv. Mater. 2004, 16, 1352 – 1356.
[20] W.-S. Yeo, C. D. Hodneland, M. Mrksich, ChemBioChem 2001,
2, 590 – 593.
[21] W.-S. Yeo, M. N. Yousaf, M. Mrksich, J. Am. Chem. Soc. 2003,
125, 14994 – 14995.
[22] M. N. Yousaf, B. T. Houseman, M. Mrksich, Angew. Chem. 2001,
113, 1127 – 1130; Angew. Chem. Int. Ed. 2001, 40, 1093 – 1096; .
[23] M. N. Yousaf, B. T. Houseman, M. Mrksich, Proc. Natl. Acad.
Sci. USA 2001, 98, 5992 – 5996.
[24] a) D. K. Hoover, E.-J. Lee, E. W. L. Chan, M. N. Yousaf,
ChemBioChem 2007, 8, 1920 – 1923; b) D. K. Hoover, E. W. L.
Chan, M. N. Yousaf, J. Am. Chem. Soc. 2008, 130, 3280 – 3281.
[25] a) E. W. L. Chan, M. N. Yousaf, J. Am. Chem. Soc. 2006, 128,
15542 – 15546; b) E. W. L. Chan, M. N. Yousaf, ChemPhysChem
2007, 8, 1469 – 1472.
[26] T. Horn, B. B. Lee, K. A. Dill, R. N. Zuckermann, Bioconjugate
Chem. 2004, 15, 428– 435.
[27] FLAG peptide is an eight amino acid (DYKDDDDK) fusion tag
commonly used for immunoaffinity chromatography: see A.
Einhauer, A. J. Jungbauer, J. Biochem. Biophys. Methods 2001,
49, 455 – 465.
[28] K. L. Prime, G. M. Whitesides, J. Am. Chem. Soc. 1993, 115,
10714 – 10721.
[29] a) W. S. Dillmore, M. N. Yousaf, M. Mrksich, Langmuir 2004, 20,
7223 – 7231; b) N. P. Westcott, M. N. Yousaf, Langmuir 2008, 24,
2261 – 2265.
[30] The RGD peptide is found in the central binding domain of the
extracellular matrix protein fibronectin and is known to facilitate
adhesion through cell-surface integrin receptors: see E. Ruo-
slahti, Annu. Rev. Cell Dev. Biol. 1996, 12, 697 – 748.
[31] Polydimethylsiloxane stamps were fabricated according to the
literature: see Y. Xia, J. Tien, D. Qin, G. M. Whitesides,
Langmuir 1996, 12, 4033 – 4038.
[32] a) L. O. Hodgson, E. W. L. Chan, K. M. Hahn, M. N. Yousaf, J.
Am. Chem. Soc. 2007, 129, 9264 – 9265; b) D. G. Barrett, M. N.
Yousaf, Angew. Chem. 2007, 119, 7581 – 7583; Angew. Chem. Int.
Ed. 2007, 46, 7437 – 7439; c) E. W. L. Chan, M. N. Yousaf,
Angew. Chem. 2007, 119, 3955 – 3958; Angew. Chem. Int. Ed.
2007, 46, 3881 – 3884.
Received: January 12, 2008
Revised: April 16, 2008
Published online: June 20, 2008
Keywords: cell adhesion · electrochemistry ·
.
heterogeneous catalysis · monolayers · redox chemistry
[1] S. R. Khetani, S. N. Bhatia, Curr. Opin. Biotechnol. 2006, 17,
524 – 531.
[2] R. Langer, D. A. Tirrell, Nature 2004, 428, 487 – 492.
[3] U. Hersel, C. Dahmen, H. Kessler, Biomaterials 2003, 24, 4385 –
4415.
[4] J. A. Hubbell, Curr. Opin. Biotechnol. 1999, 10, 123 – 129.
[5] M. Mrksich, Chem. Soc. Rev. 1995, 10, 73 – 78.
[6] G. Bao, S. Suresh, Nat. Mater. 2003, 2, 715 – 725.
[7] J. L. Tan, J. Tien, D. M. Pirone, D. S. Gray, K. Bhadriraju, C. S.
Chen, Proc. Natl. Acad. Sci. USA 2003, 100, 1484 – 1489.
[8] N. Q. Balaban, U. S. Schwarz, D. Riveline, P. Goichberg, G. Tzur,
I. Sabany, D. Mahalu, S. Safran, A. Bershadsky, L. Addadi, B.
Geiger, Nat. Cell Biol. 2001, 3, 466 – 472.
[9] L. A. Haines, K. Rajapopal, B. Ozbas, D. A. Salick, D. J. Pochan,
J. P. Schneider, J. Am. Chem. Soc. 2005, 127, 17025 – 17029.
[10] D. L. Elbert, J. A. Hubbell, Biomacromolecules 2001, 2, 430 –
441.
[11] D. L. Hern, J. A. Hubbell, J. Biomed. Mater. Res. 1998, 39, 266 –
276.
[12] X. Jiang, D. A. Bruzewicz, A. P. Wong, M. Piel, G. M. White-
sides, Proc. Natl. Acad. Sci. USA 2005, 102, 975 – 978.
[13] X. Jiang, R. Ferrigno, M. Mrksich, G. M. Whitesides, J. Am.
Chem. Soc. 2003, 125, 883 – 886.
[14] T. Okano, N. Yamata, J. Sakai, Y. Sakurai, Biomaterials 1995, 16,
297 – 303.
[15] T. O. Collier, J. M. Anderson, A. Kikuchi, T. J. Okano, J.
Biomed. Mater. Res. 2002, 59, 136 – 143.
[16] T. Shimizu, M. Yamato, A. Kikuchi, T. Okano, Biomaterials
2003, 24, 2309 – 2316.
[17] M. Mrksich, Curr. Opin. Chem. Biol. 2002, 6, 794 – 797.
Angew. Chem. Int. Ed. 2008, 47, 6267 –6271
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6271