High-yielding and photolabile approaches to the covalent attachment of biomolecules to surfaces via hydrazone chemistry

Article abstract of DOI:10.1021/la500744s

The development of strategies to couple biomolecules covalently to surfaces is necessary for constructing sensing arrays for biological and biomedical applications. One attractive conjugation reaction is hydrazone formation - the reaction of a hydrazine w

Full text of DOI:10.1021/la500744s

Article
pubs.acs.org/Langmuir
High-Yielding and Photolabile Approaches to the Covalent
Attachment of Biomolecules to Surfaces via Hydrazone Chemistry
Ju Hun Lee,† Dylan W. Domaille,† Hyunwoo Noh,† Taeseok Oh,† Chulmin Choi,‡ Sungho Jin,‡
and Jennifer N. Cha*,†
†Department of Chemical and Biological Engineering, University of Colorado, Boulder, 596 UCB, Boulder, Colorado 80303-1904,
United States
‡Department of Mechanical and Aerospace Engineering, University of California, San Diego, 9500 Gilman Drive, M/C 0411, La Jolla,
California 92093-0411, United States
S
* Supporting Information
ABSTRACT: The development of strategies to couple biomolecules
covalently to surfaces is necessary for constructing sensing arrays for
biological and biomedical applications. One attractive conjugation
reaction is hydrazone formationthe reaction of a hydrazine with an
aldehyde or ketoneas both hydrazines and aldehydes/ketones are
largely bioorthogonal, which makes this particular reaction suitable
for conjugating biomolecules to a variety of substrates. We show that
the mild reaction conditions afforded by hydrazone conjugation
enable the conjugation of DNA and proteins to the substrate surface
in significantly higher yields than can be achieved with traditional
bioconjugation techniques, such as maleimide chemistry. Next, we
designed and synthesized a photocaged aryl ketone that can be
conjugated to a surface and photochemically activated to provide a
suitable partner for subsequent hydrazone formation between the surface-anchored ketone and DNA- or protein-hydrazines.
Finally, we exploit the latent functionality of the photocaged ketone and pattern multiple biomolecules on the same substrate,
effectively demonstrating a strategy for designing substrates with well-defined domains of different biomolecules. We expect that
this approach can be extended to the production of multiplexed assays by using an appropriate mask with sequential
photoexposure and biomolecule conjugation steps.
microarrays have been pursued, including the use of active
esters,19 epoxides,20,21 maleimides,21 and carbodiimides22 as
well as photochemical cross-linkers.23−26 These chemistries rely
on general nucleophile/electrophile interactions and conse-
quently are limited by lower chemoselectivity and are
susceptible to inactivation by water.7,9,19,22,27,28
To address these issues, aldehyde-based hydrazone chemistry
has been shown to be an excellent reaction for conjugating
biomolecules due to its high chemoselectivity and reactivity
under mild conditions. The number of reports accelerated after
Dirksen et al. reported that the small molecule aniline could
catalyze hydrazone and oxime formation and that these
conditions were suitable for biomolecule ligation and label-
ing.29−35 While hydrazones have previously been used as a
surface-patterning chemistry,9,28 because the hydrazone link-
ages formed typically arise from the reaction of a hydrazine with
an aliphatic aldehyde, these hydrazone linkages are less
thermodynamically stable than hydrazone linkages derived
from aryl aldehydes.30,31 As we36 and others37 have shown, bis-
INTRODUCTION
■
The development of techniques to immobilize biomolecules on
micropatterned substrates is necessary to producing biological
diagnostics and assays. For instance, DNA microarrays have
enabled the measurement of gene expression and allowed
researchers to diagnose particular genetic conditions.1−5 The
first DNA microarrays used UV cross-linking or relied on
simple electrostatic interactions between DNA and poly-L-
lysine-coated glass to monitor gene expression patterns in
human cancer or in plants.2,3 The extension of DNA arrays to
patterned protein surfaces has been used to quantify
biomolecules in solution, which has helped monitor and
study biological processes as well as screen for diseases.6−18 For
protein conjugation to surfaces, Macbeath et al. demonstrated
that parallel protein arrays could be produced using Schiff base
linkages between aldehyde surfaces and protein amino groups,7
while Zhu et al. utilized histidine−nickel interactions to anchor
proteins to surfaces.8 In addition, Robinson et al. described a
technique to analyze mutiple proteins (autoantibodies)
corresponding to eight distinct human autoimmune diseases
simultaneously by spotting the antigen microarrays on poly-L-
lysine-coated glass.18 Since these earlier reports, a variety of
covalent bioconjugation strategies for producing biomolecule
Received: February 24, 2014
Revised: June 8, 2014
Published: June 27, 2014
© 2014 American Chemical Society
8452
dx.doi.org/10.1021/la500744s | Langmuir 2014, 30, 8452−8460

Langmuir
Article
aliphatic hydrazones form exceptionally fast at physiological pH
with second-order rate constants on the order of 5−20 M−1 s−1,
yet the hydrolysis back reaction is also rapid, with first-order
rate constants on the order of 10−4 s−1. Thus, we sought to
pattern surfaces through aryl hydrazone linkages and, based on
previous reports,30 reasoned that the resulting linkage would be
more resistant to hydrolysis.
Here, we show that DNA, proteins, and enzymes can be
conjugated to aryl aldehyde-derivatized surfaces through
hydrazone linkages. The conditions are mild enough that
proteins retain their native conformation, as evidenced by the
persistence of antibody−antigen interactions, and enzymes
retain their enzymatic activity. Next, we designed and
synthesized a photocaged ketone compound that rapidly
converts to a reactive carbonyl compound in the presence of
light and can react with hydrazine nucleophiles under
organocatalysis conditions to form stable hydrazone linkages
between the surface and biomolecule. Finally, we demonstrate
that this approach enables the patterning of multiple
biomolecules on the same substrate in photolitographically
defined regions.
RESULTS AND DISCUSSION
■
Photoresist-capped micrometer-sized silicon pillars were
produced by lithography and a negative tone photoresist
(PR). After the hydroxylation of the non-PR-coated silicon
surfaces with nitric acid, the Si regions were passivated with
methoxypolyethylene glycol trimethylsilyl ether (PEG-silane)
(MW 5000) to minimize nonspecific biomolecule adsorp-
tion.10,38 The substrates were briefly washed with fresh toluene
and sonicated in acetone to remove the negative PR from the
tops of the Si pillars; these regions were then reacted with 3-
aminopropyltriethoxysilane (APTES) to install amine groups
on top of the Si pillars (Figure 1a).
Figure 1. (a) Schematic showing methods used to fabricate the PEG-
passivated aldehyde surface and conjugate biomoleulces via hydrazone
chemistry. (b) Fluorescent image of maleimide-functionalized
substrates reacted with FAM-labeled thiolated DNA. (c) Fluorescent
image of aldehyde-functionalized substrates reacted with FAM-labeled
hydrazide-DNA. (d) Comparing fluorescence intensity line plots
obtained with ImageJ. Red and blue plots corresponded to horizontal
fluorescence intensity line profiles of eight representative FAM-DNA
immobilized pillars in panel b and c, respectively.
maleimide-thiol chemistry (Figure 1d). It is likely that the
high conjugation efficiency of the hydrazone reaction is a result
of the high chemoselectivity of the reaction. While aldehyde
and hydrazine components selectively react with each other,
maleimides can be hydrolyzed by the aqueous buffer
conditions, and thiols can oxidize to disulfides, rendering
these reagents unreactive in the bioconjugation procedure.
We next explored the potential for attaching proteins
through aryl hydrazone chemistry. In this case, fluorescein-
labeled antigoat immunoglobulin (IgG) was derivatized with
pyridyl hydrazine groups (Materials and Methods). The
aldehyde-functionalized Si substrates were reacted with
fluorescent hydrazine-derivatized antigoat IgG in 100 mM
NH4OAc buffer at pH 4.8 for 24 h. Again, as a point of
comparison, thiolated fluorescent IgG was reacted with
maleimide-derivatized Si pillars. While the use of maleimide
chemistry yielded a negligible fluorescence signal from
fluorescein-labeled IgG (Figure S1a), using hydrazone for-
mation yielded a high amount of bound antibody (Figure S1b).
The competitive elution of the fluorescent IgG hydrazine with
pH 4.8 buffer and excess hydroxylamine allowed us to estimate
that ∼5.76 × 108 proteins were conjugated per pillar. Next, we
utilized antibody−antigen interactions to investigate whether
the hydrazone conjugation procedure would cause protein
denaturation. For this, antigoat IgG was derivatized with pyridyl
hydrazine groups and anchored to the aldehyde substrate
through hydrazone linkages. Next, goat IgG was fluorescently
labeled with NHS-fluorescein and incubated with the surface-
bound antigoat IgG. As shown in Figure S2, the antigoat IgG
microarrays were capable of binding the fluorescein-conjugated
We first sought to demonstrate the utility of using an aryl
aldehyde to conjugate biomolecules to the surface. To this end,
the amine-modified pillars were reacted with 4-formyl
succinimidyl benzoate to install aryl aldehyde moieties on the
pillar surfaces. The aldehyde density was approximated by
conjugating fluorescein hydrazide to the aldehydes, washing
away the excess dye, releasing the conjugated dye with pH 4.8
buffer, and measuring the concentration of liberated dye by
UV−visible spectroscopy. On the basis of this analysis, we
determined that the density of aldehyde groups on the Si pillars
was ∼6.4 nmol/cm2. With the aldehyde-functionalized
substrates in hand, we next sought to compare the efficiency
of hydrazone-mediated conjugation to traditional bioconjuga-
tion methodologies. To this end, the aldehyde substrates were
reacted with fluorescein-tagged DNA displaying a 5′-acyl
hydrazine (10 μM). We observed high conjugation of the
DNA-fluorescein to the aldehyde-functionalized pillars as
visualized by fluorescence microscopy and densitometric
analysis (Figure 1). As a mode of comparison, DNA
microarrays were produced by installing maleimide groups on
the Si pillars with succinimidyl-4-(N-maleimidomethyl)-
cyclohexane-1-carboxylate (SMCC) and then reacting these
substrates with fluorescein-tagged DNA displaying a 5′-thiol
group. Under these conditions, we observed low conjugation of
the DNA to the pillars, indicating that the hydrazone
conjugation proceeded more smoothly and efficiently than
this standard bioconjugation approach. Indeed, densitometric
analysis indicated that the hydrazone-mediated conjugation
resulted in 12.5-fold-higher DNA conjugation than the
8453
dx.doi.org/10.1021/la500744s | Langmuir 2014, 30, 8452−8460

Langmuir
Article
Scheme 1. Synthesis of a Photocaged Ketone Derivative
goat IgG; in the absence of conjugated antigoat IgG, no
appreciable fluorescein signal was detected. Because antibody−
antigen interactions are highly structure-specific, we concluded
that conjugating the antigoat IgG with pyridyl hydrazine groups
did not cause protein denaturation or activity. To verify this
further, we also conjugated the enzyme alkaline phosphatase to
the Si pillars using hydrazone chemistry and tested for
activity.39,40 For this, alkaline phosphatase was modified with
pyridyl hydrazine groups and conjugated to the aldehyde-
functionalized Si substrates. Upon incubation with phosphatase
substrate ELF-97, a reagent that forms an insoluble fluorogenic
product upon phosphate removal, we saw fluorescent
precipitation on the silicon pillars, indicating that the
phosphatase activity of the enzyme was retained through the
bioconjugation procedure (Figure S3). Taken together, these
antibody and enzyme results demonstrate the ability of this
conjugation chemistry to pattern proteins on surfaces while
preserving the innate biological activity of biomolecules.
With the ability to form stable bioconjugate linkages between
surface aldehydes and hydrazine-derivatized biomolecules
firmly demonstrated, we next sought to expand this patterning
methodology to include photolithography. We reasoned that by
using light to uncage a carbonyl that could react with hydrazine
nucleophiles, we could integrate this bioconjugation technique
with photopatterning to generate a photocontrolled method of
patterning DNA, peptides, and proteins on a single surface. In
order to achieve this, we first designed a photolabile system that
would reveal a hydrazine-reactive carbonyl upon light exposure.
For this, we chose to use an o-nitrobenzyl alcohol substrate.
Typically, this class of compounds is used to photocage an
alcohol or carboxylic acid: upon irradiation with UV light, the
compound releases the alcohol or carboxylate component with
concomitant production of an o-nitrosobenzaldehyde.41−43
Thus, if we could anchor the o-nitrobenzyl alcohol substrate
to the surface, we could generate a reactive electrophile upon
light exposure and use this photogenerated electrophile for
subsequent reaction with a hydrazine-derivatized biomolecule.
In our initial experiments, we examined the ability to use an
o-nitrobenzyl ether to photoconvert to its corresponding o-
nitrosobenzaldehyde 1. However, we observed that the
photogenerated aldehyde was unstable and rapidly decomposed
(Figure S4). This observation is consistent with previous
mechanistic investigations of o-nitrobenzyl alcohol/ether
photoconversions.42 In order to circumvent the instability of
the photochemically revealed aldehyde, we designed a
photocaged ketone 2 (Scheme 1). The resulting o-nitro-
soacetophenone derivative has been reported to be much less
susceptible to degradation than the corresponding benzalde-
hyde derivative.42 Upon irradiation with UV light, this
compound should liberate methanol to generate an acetophe-
none (ketone) derivative to react with hydrazine-derivatized
biomolecules. Initial experiments with acetophenone-modified
surfaces confirmed that surface-bound aryl ketones are reactive
with hydrazines and allowed us to identify optimal conditions
for bioconjugation to these functionalities (Figure S5).
The synthesis of NHS-ester 2 was achieved in six steps from
4-ethyl benzoic acid (Scheme 1, Materials and Methods) with
minimal chromatography. The nitration of 4-ethyl benzoic acid
in a mixture of H2SO4 and HNO3 installed a nitro group
adjacent to the ethyl substituent. After the precipitation of this
compound in ice water, 3 was converted to its corresponding
activated ester with EDC and N-hydroxysuccinimide and then
immediately reacted with NaOMe to yield the methyl ester in
79% over two steps. Radical bromination of 4 was carried out
with N-bromosuccinimide (NBS) and a benzoyl peroxide
initiator. After an overnight reaction, benzyl bromide 5 was
isolated, and we attempted to displace the bromide with the
methoxide ion; however, this approach led to an intractable
mixture of products. Thus, a methanolic solution of AgNO3 was
used to facilitate the replacement of the bromide with a
methoxy group under neutral conditions, which furnished
methyl ether 6 in good yield. Next, the methyl ester group was
hydrolyzed with NaOH in the presence of LiCl, and the crude
material was extracted and precipitated slowly from a
concentrated solution of ethyl acetate with petroleum ether.
This afforded carboxylic acid 7 as a white solid in high yield
(98%) and purity. Finally, 7 was converted to its corresponding
NHS-ester in the presence of EDC and N-hydroxysuccinimide,
providing the final photocaged ketone compound, 2.
1
We monitored the photoconversion of compound 7 by H
NMR to confirm that it would form the expected o-
nitrosoacetophenone derivative upon exposure to light (Figure
S6). A solution of 7 in MeOH/DI water (3:1) was exposed to
15 min of irradiation with a mercury vapor short arc lamp, and
8454
dx.doi.org/10.1021/la500744s | Langmuir 2014, 30, 8452−8460

Langmuir
Article
Figure 2. Time-course 1H NMR study for the conversion of the photocaged ketone with various degrees of light exposure. The first spectrum was
recorded without light exposure in MeOD/D2O (3:1), and the second and third spectra were recorded after 5 and 15 min of light exposure,
respectively.
Figure 3. Schematic showing how patterned biomolecules were generated from photocaged ketones. The photoresist was first removed with
acetone, and the Si pillars were derivatized with ATPES. Next, amine-functionalized Si pillars were reacted with the NHS ester of the photocaged
ketone. Upon photoillumination, ketone groups were exposed, and these could be reacted with hydrazine-derivatized biomolecules.
1
the progress of the reaction was monitored by H NMR; the
time-course NMR study is shown in Figure 2. We saw clean
photoconversion to the corresponding ketone as evidenced by
the disappearance of the benzylic and methoxy group protons
and the emergence of a singlet centered at 2.7 ppm,
corresponding to the ketone methyl group. In contrast to the
photogenerated aldehyde, the photogenerated ketone was
stable under these reaction conditions, indicating that this
reagent would be suitable for subsequent reaction with a
hydrazine-derivatized biomolecule. These data indicated that
not only does the photoconversion proceed cleanly without
side products but the compound also photolyzes rapidly with
complete conversion after only 15 min of light exposure.
The photocaged ketone groups were then installed on the Si
pillars by reacting NHS-ester 2 with the amine-derivatized
pillars (Figure 3). Next, the substrates were exposed to UV light
in pH 5.5 deionized water and MeOH and irradiated with a
mercury vapor arc lamp for 30 min, transferred to pH 8.0
phosphate buffer for 10 min, washed, and incubated overnight
at pH 4.8 with hydrazide-DNA bearing a fluorescein group. In
contrast to our initial expectations, we saw very low conjugation
of the DNA to the ketone substrate. We reasoned that the low
conjugation efficiency was a result of the lower reactivity of
ketones in forming hydrazones. While the ketone derivatives
are more stable and form fewer side products during
photogeneration, the reaction of an aryl ketone with a
hydrazide has slower reaction kinetics than the reaction
between an aldehyde and a hydrazide.44 Thus, we initiated a
search for an efficient organocatalyst that would facilitate the
formation of the hydrazone between the biomolecule hydrazine
and surface ketone.
Aniline is a common small-molecule organocatalyst often
used to enhance the rate of hydrazone formation with
aldehydes; however, ketones require more efficient catalysts
due to their lower reactivities.45 To increase the rate of
hydrazone formation between the photolyzed ketone and
hydrazine-functionalized biomolecules, we screened three
catalysts that have been reported to catalyze hydrazone
formation: aniline, anthranilic acid, and m-phenylenediamine
(m-PDA).45,46 Ketone-modified pillars were reacted with DNA-
hydrazide bearing a fluorescein tag in the presence of 10 mM
anthranilic acid or aniline or 150 mM m-PDA in pH 4.5 buffer
(Figure 4, Figure S7). We measured the fluorescence intensity
of each pillar for each experimental condition and observed that
m-PDA increased the DNA loading by ∼4.2-fold above that of
aniline and 3.0-fold above that of anthranilic acid. Though
8455
dx.doi.org/10.1021/la500744s | Langmuir 2014, 30, 8452−8460

Langmuir
Article
NaCNBH3 to form irreversible covalent bonds. Remaining
ketone groups were blocked with PEG-hydrazine, and the
remainder of the silicon surface, which was initially blocked by
the mask, was exposed to light to reveal reactive ketone groups.
These were then conjugated in a subsequent step with
TAMRA-labeled hydrazine antigoat IgG. As shown in Figure
6, by using the photocaged ketone in conjunction with a mask,
Figure 4. Comparing the different catalysts to promote hydrazone
formation between hydrazide DNA labeled with FAM and exposed
acetophenone (ketone) groups after photoillumination with (a) 10
mM aniline, (b) 10 mM anthranilic acid, and (c) 150 mM m-
phenylenediamine in pH 4.5 buffer. (d) Densitometric data comparing
the average fluorescence intensity per unit area on the pillars for the
three different catalysts. The error bars represent standard deviations
obtained from 10 pillars. The numbers are normalized to the intensity
of the fluorescence obtained from using 10 mM aniline (a).
anthranilic acid is a very efficient reagent for catalyzing
hydrazone formation at neutral pH, its solubility is limited in
acidic buffer, which likely decreased the concentration of
available catalyst. Control experiments confirmed that DNA
was unable to attach to the surfaces in the absence of
illumination (Figure 5), indicating that the reaction would
proceed only in areas that displayed a ketone moiety. These
experiments indicate that m-PDA is the most efficient catalyst
for hydrazone formation between DNA hydrazides and surface-
bound ketones, owing to its high solubility in acidic aqueous
buffer.
Finally, we reasoned that by utilizing our photopatterning
strategy, photoreactive carbonyls could be uncaged in a spatially
controlled manner. Thus, it may be possible to attach different
biomolecules to discrete regions of the same substrate. To
study this, a 500 μm2 mask was used to first reveal ketones in
one region of a substrate; these photogenerated ketones were
then reacted with FAM-conjugated hydrazide DNA. After DNA
conjugation, the hydrazone bonds were reduced with
Figure 6. (a) Schematic showing methods used to pattern two types of
biomolecules on the substrate via hydrazone chemistries. I. Light was
first exposed through a mask in MeOH/H2O. II. The photoexposed
regions were next conjugated with FAM-labeled DNA. III. After
reduction and PEG blocking, light was exposed in a second region of
the sample. IV. These newly exposed regions were reacted with
TAMRA-labeled hydrazine antigoat IgG. (b) Fluorescent images
obtained. I. Fluoresent image of the DNA-FAM region. II. Fluorescent
image of the TAMRA-antigoat IgG region. III. Fluorescent image
showing the boundary region between DNA-FAM and TAMRA-
antigoat IgG.
it was possible to pattern two different types of biomolecules on
a single substrate with high bioconjugation yields and no cross
reactivities.
Figure 5. Conjugation of FAM-labeled hydrazide DNA to photopatterned surfaces. Fluorescent images of FAM-labeled hydrazide DNA on Si pillars.
In all cases, 150 mM mPDA was used as a catalyst. (a) Fluorescent image showing no attachment of DNA to Si pillars displaying a photocaged
ketone. (b) Fluorescent image showing no attachment of DNA to Si pillars after incubation in the photolysis buffer. (c) Fluorescence image showing
DNA conjugated to the pillars after light exposure.
8456
dx.doi.org/10.1021/la500744s | Langmuir 2014, 30, 8452−8460

Langmuir
Article
acetone hydrazone (SANH, Solulink) for 3 h, followed by purification
with a 7k MWCO spin desalting column (Thermo Scientific).
Aldehyde−silicon pillars were then reacted with 2 μM hydrazine-
fluorescein-conjugated antigoat IgG in 100 mM NH4OAc buffer at pH
4.8 overnight, and the substrate was then washed with NH4OAc buffer.
Alkaline Phosphatase Immobilized on the Aldehyde-
Modified Pillars. Alkaline phosphatase (1 mg/mL) was first reacted
with 20 equiv of SANH in MOPS buffer (pH 8.1, 100 mM MOPS, 150
mM NaCl, 2 mM MgCl2, 0.1 mM ZnCl2) for 2.5 h, followed by
purification with a 7k MWCO spin desalting column (Thermo
Scientific). Aldehyde−silicon pillars were then reacted with 2 μM
hydrazine-conjugated alkaline phosphatase in 100 mM NH4OAc, 150
mM NaCl, 3 mM MgCl2 buffer at pH 4.8 overnight followed by
washing with NH4OAc buffer. The resulting alkaline phosphatase-
covered silicon pillars were incubated with fluorogenic phosphatase
substrate ELF-97 (20 μM) (Molecular Probes, OR) in glycine buffer
(100 mM, 1 mM MgCl2, 1 mM ZnCl2) at pH 10.
CONCLUSIONS
■
We have demonstrated a facile and high-yield approach for
covalently immobilizing DNA and proteins on photolitho-
graphically patterned silicon surfaces via hydrazone formation
chemistry between surface-bound aldehydes with hydrazide- or
hydrazine-derivatized biomolecules. DNA and proteins were
successfully conjugated to aldehyde-modified silicon pillars in
significantly higher yields than what could be achieved with
maleimide−thiol coupling, a traditional bioconjugation meth-
odology. Experiments with antibody−antigen pairs as well as
enzymes demonstrate that the structural integrity of the
proteins is preserved during and after conjugation to the
surfaces. We have also designed and synthesized a molecule
that can be anchored to a surface and prompted to reveal a
hydrazine-reactive carbonyl in the presence of UV light. Time-
course NMR studies establish that the compound undergoes
clean photoconversion to its correspond acetophenone
derivative. We demonstrate that this photorevealed ketone
can be coupled to hydrazide/hydrazine nucleophiles in the
presence of an m-PDA organocatalyst to achieve good coverage
biomolecules in photopatterned areas. Finally, we highlight the
ability of this approach for installing multiple biomolecules in
discrete regions of the same substrate. We expect that this
strategy will find applications in the develeopment of
multiplexed assay production.
Synthesis of a Photocaged Ketone Compound. The synthesis
route to photoactivate ketone compound 2 is shown in Scheme 1.
Detailed synthesis procedures and NMR spectra are provided as
follows.
4-Ethyl-3-nitrobenzoic Acid (3). A 50 mL round-bottomed flask
was immersed in a water bath at room temperature, and concentrated
sulfuric acid (5 mL) was carefully added to concentrated nitric acid
(60−70%, 5 mL). 4-Ethylbenzoic acid (1.00 g, 6.6 mmol) was added
in a single portion. The flask was capped with a septum, and a
precipitate formed after ca. 60 s. The reaction was allowed to proceed
for 3 h at room temperature, at which point it was poured into a
beaker with ice. The reaction was filtered, and the white solid was
washed with water, collected, and dried (1.26 g, 97%). 1H NMR (400
MHz, DMSO-d6) δ 8.34 (d, J = 1.8 Hz, 1H), 8.13 (dd, J = 8.0, 1.8 Hz,
1H), 7.65 (d, J = 8.1 Hz, 1H), 2.87 (q, J = 7.5 Hz, 2H), 1.22 (t, J = 7.5
Hz, 3H).
MATERIALS AND METHODS
■
Unless otherwise noted, all reagents were obtained from commercial
sources and used without further purification. UV−vis spectra were
acquired on a DU 730 spectrophotometer (Beckman Coulter, USA).
1H and 13C NMR spectra were recorded on 400 MHz NMR (Avance-
III 400 NMR, Bruker). Photoirradiation experiments were carried out
with a mercury lamp (200 W, X-cite exacte microscope illumination
system, Lumen Dynamics, USA).
Photoresist-Capped Micrometer-Sized Silicon Pillar. Micro-
patterned silicon substrates were fabricated by reactive ion etching
(RIE). Photoresist NR9-3000PY (Futurrex, Inc) was spun on the
silicon wafer at 4000 rpm for 40 s to obtain a layer thickness of 2.5 μm.
After a soft-bake step on a 150 °C hot plate for 60 s, the photoresist
layer was exposed to 22 s of UV light in a MA6 mask aligner through a
photomask (5 μm diameter and 10 μm spacing). Subsequently the
wafer was baked on a 150 °C hot plate for 60 s (postexposure bake,
PEB) and developed with an RD6 developer (Futurrex, Inc).
Poly(ethylene glycol) (PEG)-Passivated Amine-Modified
Pillars. The above substrates were first treated with 2 M nitric acid
for 1 h to hydroxylate the non-PR-coated areas. After being washed
with DI water, the substrates were reacted with 5 mg/mL
poly(ethylene glycol) (PEG)-silane (MW 5000) in toluene for 5 h
and then washed with toluene, acetone (with sonication for 4.5 min for
wahing and removing the photoresist), and ethanol. To form a
covalent bond between the PEG chains and the Si surface, the
substrates were next thermally treated at 120 °C for 30 min. The
revealed Si pillar surfaces were hydroxylated with a 1 h, 2 M nitric acid
treatment. The substrate was reacted with 5% v/v APTES ((3-
aminopropyl)triethoxysilane) in ethanol overnight followed by heat
treatment at 120 °C for 30 min to derivatize the pillars with amine
groups.
Methyl-4-ethyl-3-nitrobenzoate (4). A 20 mL scintillation vial was
charged with a stirbar, 2 (390 mg, 2.0 mmol), EDC (465 mg, 3.0
mmol, 1.5 equiv), N-hydroxysuccinimide (277 mg, 2.4 mmol, 1.2
equiv), and CH2Cl2 (10 mL). The reaction was allowed to proceed at
room temperature for 6 h, at which point it was washed with water
(2× 5 mL) and brine (1× 5 mL) and dried over MgSO4. The solvent
was removed by rotary evaporation to yield a light-yellow oil that
crystallized into lemon-yellow crystals upon standing. The crude
material was dissolved in MeOH (25 mL), and NaOMe (5.4 M
solution in MeOH) was added (840 uL, 2.0 equiv). The reaction was
monitored by TLC (CH2Cl2), and after 30 min, the reaction was
diluted with EtOAc (5 mL) and CH2Cl2 (20 mL) and washed with
water (2× 10 mL), and the organic phase was dried over MgSO4.
Concentration in vacuo provided pale-yellow oil (332 mg, 79% over
two steps). 1H NMR (400 MHz, CDCl3) δ 8.54 (d, J = 1.8, 1H), 8.19
(dd, J = 8.0, 1.8 Hz, 1H), 7.49 (d, J = 8.1, 1H), 3.98 (s, 3H), 2.97 (q, J
= 7.5 Hz, 2H), 1.33 (t, J = 7.5 Hz, 3H).
Methyl 4-(1-bromoethyl)-3-nitrobenzoate (5). N-Bromosuccini-
mide (309 mg, 1.75 mmol) and benzoyl peroxide (30 mg) were added
to a benzene solution (10 mL) of 3 (242 mg, 1.16 mmol). The
reaction was brought to reflux for 14 h in the dark, at which point it
was cooled to room temperature and the organic phase was washed
with water (2× 10 mL) and brine (1× 10 mL). The organic phase was
separated and dried over MgSO4. Concentration in vacuo furnished a
dark-orange oil. This compound coeluted with the only major
impurity, so the crude material was subjected to the next reaction
DNA and Antibody Immobilized on the Aldehyde-Modified
Pillars. The PEG-passivated amine-modified Si pillars were reacted
with 4-formyl succinimidyl benzoate (200 μM) for 12 h in 100 mM
NaHCO3 buffer, pH 8.2, and then washed with 1× PBS (pH 7.2)
buffer. The resulting aldehyde−silicon pillars were next reacted with
10 μM FAM-conjugated hydrazide A15 in 100 mM NH4OAc buffer at
pH 4.8 overnight, followed by washing with NH4OAc buffer. To
immobilize proteins, 1.5 mg/mL fluorescein-conjugated antigoat IgG
was first reacted with 20 equiv of succinimidyl-6-hydrazinonicotinate
8457
dx.doi.org/10.1021/la500744s | Langmuir 2014, 30, 8452−8460

Langmuir
Article
without any further purification. 1H NMR indicated that the material is
ca. 75% pure. H NMR (400 MHz, CDCl3) δ 8.50 (d, J = 1.8, 1H),
8.29 (dd, J = 8.0, 1.8 Hz, 1H), 8.00 (d, J = 8.3, 1H), 5.82 (q, J = 6.8
Hz, 1H), 4.00 (s, 3H), 2.12 (d, J = 6.8 Hz, 3H).
pH 8.0 10 mM phosphate buffer. The photolyzed ketone−Si pillars
were then reacted with 25 μM of FAM-conjugated hydrazide-DNA at
pH 4.5 and in the presence of 150 mM m-phenylenediamine (m-PDA)
overnight. The formed hydrazones between the photolyzed ketone
and FAM-modified hydrazide DNA were reduced with 5 mM
NaCNBH3 at pH 4.0 to form irreversible covalent bonds. Any
remaining unreacted ketones were reacted with 1 M hydrazide PEG
and 150 mM m-PDA at pH 4.5. For the DNA and protein arrays, the
DNA patterns were generated following the aforementioned steps.
Next, the rest of the surface was illuminated to expose ketone groups
which were then reacted with 2 μM TAMRA-conjugated hydrazine
antigoat IgG in 150 mM m-PDA at pH 4.5.
1
Methyl 4-(1-Methoxyethyl)-3-nitrobenzoate (6). A 20 mL
scintillation vial was charged with a stirbar, 4 (100 mg, 0.48 mmol),
AgNO3 (100 mg, 0.6 mmol), and MeOH (5 mL). The vial was
protected from light, and the reaction was stirred overnight at room
temperature. The precipitated AgBr was filtered off, and the filtrate was
concentrated. 1H NMR (400 MHz, CDCl3) δ 8.60 (dd, J = 1.7, 0.4 Hz,
1H), 8.31 (dd, J = 8.2, 1.7 Hz, 1H), 7.88 (d, J = 8.2 Hz, 1H), 4.98−
4.91 (m, 1H), 4.00 (s, 3H), 3.25 (s, 3H), 1.55 (d, J = 6.3 Hz, 3H).
ASSOCIATED CONTENT
* Supporting Information
■
S
Fluorescent images of fluorescein-labeled antigoat IgG reacted
silicon pillars, fluorescein-labeled goat IgG reacted antigoat IgG
immobilized silicon pillars, and alkaline phosphatase immobi-
lized silicon pillars. Photolysis to produce active aldehydes
1
upon photoillumination and H NMR spectra monitoring the
photoconversion of 1. Amine-modified pillars reacted with the
NHS-ester of 4-carboxyacetophenone (NHS-acetophenone) to
install acetophenone (ketone) moieties on the pillars.
Photolysis to produce active ketones upon photoillumination.
Comparison of fluorescence intensity line plots obtained with
ImageJ. This material is available free of charge via the Internet
at http://pubs.acs.org.
4-(1-Methoxyethyl)-3-nitrobenzoic Acid (7). A 4 mL vial was
charged with a stirbar, 5 (10 mg, 0.04 mmol), LiCl (8.8 mg, 0.2
mmol), NaOH (8.4 mg, 0.2 mmol), and 95% MeOH (3 mL). The
reaction was stirred in the dark at room temperature for 3 h, at which
point the solution was concentrated, diluted with H2O (5 mL), and
neutralized with HCl (2.5 M). The aqueous phase was extracted with
CH2Cl2 (2× 5 mL), and the organic phase was dried over MgSO4.
Filtration and concentration provided 6 as a light-brown solid (9 mg,
98%). The residue was dissolved in ethyl acetate, and petroleum ether
AUTHOR INFORMATION
Corresponding Author
*E-mail: Jennifer.Cha@colorado.edu.
■
1
slowly diffused into it to provide 6 as a white solid. H NMR (400
MHz, CDCl3) δ 8.66 (s, 1H), 8.37 (d, J = 8.2, 1H), 7.91 (d, J = 8.2 Hz,
1H), 4.96 (q, J = 6.3 Hz, 1H), 3.26 (s, 3H), 1.56 (d, J = 6.3 Hz, 3H).
Author Contributions
J.H.L. and D.W.D. contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
2,5-Dioxopyrolidin-1-yl 4-(1-methoxy-ethyl)-3-nitro-benzoate
(2). A 20 mL scintillation vial was charged with a stirbar, 6 (9 mg,
0.04 mmol), EDC (12.4 mg, 0.08 mmol), NHS (6.9 mg, 0.06 mmol),
and CH2Cl2 (2 mL). The reaction was protected from light, and the
solution was stirred at room temperature for 5 h. The reaction was
diluted with H2O (3 mL) and CH2Cl2 (5 mL), and the layers were
separated. The organic phase was washed with H2O (1× 2 mL) and
brine (1× 2 mL), and the organic phase was dried over MgSO4. The
reaction was concentrated by rotary evaporation to give a light-yellow
oil. 1H NMR (400 MHz, CDCl3) δ 8.72 (dd, J = 1.8, 0.4 Hz, 1H), 8.41
(dd, J = 8.2, 1.8 Hz, 1H), 7.99 (dt, J = 8.3, 0.5 Hz, 1H), 5.03−4.94 (m,
1H), 3.27 (s, 3H), 2.96 (s, 4H), 1.56 (d, J = 6.3 Hz, 3H).
This work was supported by an NSF Career Award (DMR-
1056808), the Office of Naval Research (award no. N00014-09-
01-0258), and a DOE Early Career Award (DE-SC0006398).
H.N. was supported by DOE DE-SC0006398. We thank Prof.
Andrew Goodwin for help in editing the article.
REFERENCES
■
(1) Hoopes, L. Genetic Diagnosis: DNA Microarrays and Cancer.
Nat. Educ. 2008, 1, 1.
(2) DeRisi, J.; Penland, L.; Brown, P. O.; Bittner, M. L.; Meltzer, P.
S.; Ray, M.; Chen, Y.; Su, Y. A.; Trent, J. M. Use of a cDNA
Microarray to Analyse Gene Expression Patterns in Human Cancer.
Nat. Genet. 1996, 14, 457−460.
(3) Schena, M.; Shalon, D.; Davis, R. W.; Brown, P. O. Quantitative
Monitoring of Gene Expression Patterns with A Complementary DNA
Microarray. Science 1995, 270, 467−470.
(4) Lockhart, D. J.; Winzeler, E. A. Genomics, Gene Expression and
DNA Arrays. Nature 2000, 405, 827−836.
Photolysis of 1 and 7 by UV Irradiation. Seven milligrams of 1
and 7 was dissolved in 500 μL of MeOH/DI water (3:1). Solutions
containing the photocaged carbonyl compounds were transferred to
NMR tubes and illuminated with a UV light source (200 W, mercury
lamp) (Figures 2 and S4). After different amounts of irradiation, the
(5) Eisen, M. B.; Spellman, P. T.; Brown, P. O.; Botstein, D. Cluster
Analysis and Display of Genome-Wide Expression Patterns. Proc. Natl.
Acad. Sci. U.S.A. 1998, 95, 14863−14868.
(6) Zhang, S.; Yan, L.; Altman, M.; Lassle, M.; Nugent, H.; Frankel,
̈
1
F.; Lauffenburger, D. A.; Whitesides, G. M.; Rich, A. Biological Surface
Engineering: A Simple System for Cell Pattern Formation. Biomaterials
1999, 20, 1213−1220.
(7) MacBeath, G.; Schreiber, S. L. Printing Proteins as Microarrays
for High-Throughput Function Determination. Science 2000, 289,
1760−1763.
degree of photolysis (1 to 1′, 7 to 7′) was monitored by H NMR.
DNA and Antibody Immobilized on the Photocaged
Ketone-Modified Pillars. The amine-modified Si pillars were reacted
with 10 mM NHS-ester 2 for 12 h in 1 mL of CH2Cl2/Me2SO (9:1)
and then washed with CH2Cl2. The photocaged ketone−silicon pillars
were next irradiated with UV light for 30 min and then incubated in
8458
dx.doi.org/10.1021/la500744s | Langmuir 2014, 30, 8452−8460

Langmuir
Article
(8) Zhu, H.; Bilgin, M.; Bangham, R.; Hall, D.; Casamayer, A.;
Bertone, P.; Lan, N.; Jansen, R.; Bidlingmaier, S.; Houfek, T.; Mitchell,
T.; Miller, P.; Dean, R. A.; Gerstein, M.; Snyder, M. Global Analysis of
Protein Activities Using Proteome Chips. Science 2001, 293, 2101−
2105.
(9) Christman, K. L.; Requa, M. V.; Enriquez-Rios, V. D.; Ward, S.
C.; Bradley, K. A.; Turner, K. L.; Maynard, H. D. Submicron
Streptavidin Patterns for Protein Assembly. Langmuir 2006, 22, 7444−
7450.
(25) Toh, C. R.; Fraterman, T. A.; Walker, D. A.; Bailey, R. C. Direct
Biophotolithographic Method for Generating Substrates with Multiple
Overlapping Biomolecular Patterns and Gradients. Langmuir 2009, 25,
8894−8898.
(26) Martin, T. A.; Herman, C. T.; Limpoco, F. T.; Michael, M. C.;
Potts, G. K.; Bailey, R. C. Quantitative Photochemical Immobilization
of Biomolecules on Planar and Corrugated Substrates: A Versatile
Strategy for Creating Functional Biointerfaces. ACS Appl. Mater.
Interfaces 2011, 3, 3762−3771.
(10) Kannan, B.; Castelino, K.; Chen, F. F.; Majumdar, A.
Lithographic Techniques and Surface Chemistries for the Fabrication
of PEG-Passivated Protein Microarrays. Biosens. Bioelectron. 2006, 21,
1960−1967.
(11) Bailey, R. C.; Kwong, G. A.; Radu, C. G.; Witte, O. N.; Heath, J.
R. DNA-Encoded Antibody Libraries: A Unified Platform for
Multiplexed Cell Sorting and Detection of Genes and Proteins. J.
Am. Chem. Soc. 2007, 129, 1959−1967.
(12) Kim, J.; Cho, J.; Seidler, P. M.; Kurland, N. E.; Yadavalli, V. K.
Investigations of Chemical Modifications of Amino-Terminated
Organic Films on Silicon Substrates and Controlled Protein
Immobilization. Langmuir 2010, 26, 2599−2608.
(27) Angenendt, P. Progress in Protein and Antibody Microarray
Technology. Drug Discovery Today 2005, 10, 503−511.
(28) O’Donovan, L.; De Bank, P. A. A Photocleavable Linker for the
Chemoselective Functionalization of Biomaterials. J. Mater. Chem.
2012, 22, 21878−21884.
(29) Dirksen, A.; Dirksen, S.; Hackeng, T. M.; Dawson, P. E.
Nucleophilic Catalysis of Hydrazone Formation and Transimination:
Implications for Dynamic Covalent Chemistry. J. Am. Chem. Soc. 2006,
128, 15602−15603.
(30) Dirksen, A.; Hackeng, T. M.; Dawson, P. E. Nucleophilic
Catalysis of Oxime Ligation. Angew. Chem., Int. Ed. 2006, 45, 7581−
7584.
(13) Zhang, C.-J.; Tan, C. Y. J.; Ge, J.; Na, Z.; Chen, G. Y. J.;
Uttamchandani, M.; Sun, H.; Yao, S. Q. Preparation of Small-Molecule
Microarrays by trans-Cyclooctene Tetrazine Ligation and Their
Application in the High-Throughput Screening of Protein−Protein
Interaction Inhibitors of Bromodomains. Angew. Chem., Int. Ed. 2013,
52, 14060−14064.
(14) Doh, J.; Irvine, D. J. Immunological Synapse Arrays: Patterned
Protein Surfaces That Modulate Immunological Synapse Structure
Formation in T Cells. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 5700−
5705.
(15) Wilson, D. S.; Nock, S. Recent Developments in Protein
Microarray Technology. Angew. Chem., Int. Ed. 2003, 42, 494−500.
(16) Langer, R.; Tirrell, D. A. Designing Materials for Biology and
Medicine. Nature 2004, 428, 487−492.
(17) Kingsmore, S. F. Multiplexed Protein Measurement: Tech-
nologies and Applications of Protein and Antibody Arrays. Nat. Rev.
Drug Discovery 2006, 5, 1474−1776.
(31) Kalia, J.; Raines, R. T. Hydrolytic Stability of Hydrazones and
Oximes. Angew. Chem., Int. Ed. 2008, 47, 7523−7526.
(32) Dirksen, A.; Dawson, P. E. Rapid Oxime and Hydrazone
Ligations with Aromatic Aldehydes for Biomolecular Labeling.
Bioconjugate Chem. 2008, 19, 2543−2548.
(33) Domaille, D. W.; Lee, J. H.; Cha, J. N. High Density DNA
Loading on the M13 Bacteriophage Provides Access to Colorimetric
and Fluorescent Protein Microarray Biosensors. Chem. Commun. 2013,
49, 1759−1761.
(34) Byeon, J.-Y.; Limpoco, F. T.; Bailey, R. C. Efficient
Bioconjugation of Protein Capture Agents to Biosensor Surfaces
Using Aniline-Catalyzed Hydrazone Ligation. Langmuir 2010, 26,
15430−15435.
(35) Lee, J. H.; Xu, P. F.; Domaille, D. W.; Choi, C.; Jin, S.; Cha, J. N.
M13 Bacteriophage as Materials for Amplified Surface Enhanced
Raman Scattering Protein Sensing. Adv. Funct. Mater. 2014, 24, 2079−
2084.
(18) Robinson, W. H.; Digennaro, C.; Hueber, W.; Haab, B. B.;
Kamachi, M.; Dean, E.; Fournel, S.; Fong, D.; Genovese, M. C.;
Neuman De Begvar, H. E.; Skriner, K.; Hirschberg, D. L.; Morris, R. I.;
Muller, S.; Prujin, G. J.; Van Verooij, W. J.; Smolen, J.S.; Brown, P. O.;
Steinman, L.; Utz, P. J. Antiantigen Microarrays for Multiplex
Chracterization of Autoantibody Responses. Nat. Med. 2002, 8,
295−301.
(36) McKinnon, D. D.; Domaille, D. W.; Cha, N. J.; Anseth, K. S.
Biophysically Defined and Cytocompatible Covalently Adaptable
Networks as Viscoelastic 3D Cell Culture Systems. Adv. Mater.
2014, 26, 865−872.
(37) Kool, E. T.; Park, D.; Pete Crisalli, P. Fast Hydrazone Reactants:
Electronic and Acid/Base Effects Strongly Influence Rate at Biological
pH. J. Am. Chem. Soc. 2013, 135, 17663−17666.
(19) Schweitzer, B.; Roberts, S.; Grimwade, B.; Shao, W.; Wang, M.;
Fu, Q.; Shu, Q.; Laroche, L.; Zhou, Z.; Tchernev, V. T.; Christiansen,
J.; Velleca, M.; Kingsmore, S. F. Mutiplexed Protein Profiling on
Microarrays by Rolling-Circle Amplification. Nat. Biotechnol. 2002, 20,
359−365.
(38) Zhu, Y.; Gupta, B.; Guan, B.; Ciampi, S.; Reece, P. J.; Gooding,
J. J. Photolithographic Strategy for Patterning Preformed, Chemically
Modified, Porous Silicon Photonic Crystal Using Click Chemistry.
ACS Appl. Mater. Interfaces 2013, 5, 6514−6521.
(39) Bernard, A.; Renault, J. P.; Michel, B.; Bosshard, H. R.;
Delamarche, E. Microcontact Printing of Proteins. Adv. Mater. 2000,
12, 1067−1070.
(40) Bernard, A.; Fitzli, D.; Sonderegger, P.; Delamarche, E.; Michel,
B.; Bosshard, H. R.; Biebuyck, H. Affinity Capture of Proteins from
Solution and Their Dissociation by Contacting Printing. Nat.
Biotechnol. 2001, 19, 866−869.
(20) Angenendt, P.; Golkler, J. r.; Sobek, J.; Lehrach, H.; Cahill, D. J.
̈
Next Generation of Protein Microarray Support Materials: Evaluation
for Protein and Antibody Microarray Applications. J. Chromatogr., A
2003, 1009, 97−104.
(21) Kusnezow, W.; Jacob, A.; Walijew, A.; Diehl, F.; Hoheisel, J. D.
Antibody Microarrays: An Evaluation of Production Parameters.
Proteomics 2003, 3, 254−264.
(22) Kimura, N.; Oda, R.; Inaki, Y.; Suzuki, O. Attachment of
Oligonucleotide Probes to Poly Carbodiimide-coated Glass for
Microarray Applications. Nucleic Acids Res. 2004, 32, e68.
(23) Rozsnyai, L. F.; Benson, D. R.; Fodor, S. P.; Schultz, P. G.
Photolithographic Immobilization of Biopolymers on Solid Supports.
Angew. Chem., Int. Ed. 1992, 31, 759−761.
(24) Herbert, C. B.; McLernon, T. L.; Hypolitel, C. L.; Adams, D. N.;
Pikusl, L.; Huang, C.-C.; Fields, G. B.; Letourneau, P. C.; Distefano,
M. D.; Hu, W. S. Micropatterning Gradients and Controlling Surface
Densities of Photoactivatable Biomolecules on Self-assembled
Monolayers of Oligo(ethylene glycol) Alkanethiolates. Chem. Biol.
1997, 4, 731−737.
(41) Corrie, J. E. T.; Barth, A.; Munasinghe, V. R. N.; Trentham, D.
R.; Hutter, M. C. Photolytic Cleavage of 1-(2-Nitrophenyl)ethyl
Ethers Involves Two Parallel Pathways and Product Release Is Rate-
Limited by Decomposition of A Common Hemiacetal Intermediate. J.
Am. Chem. Soc. 2003, 125, 8546−8554.
(42) Il’ichev, Y. V.; Schworer, M. A.; Wirz, J. Photochemical Reaction
̈
Mechanisms of 2-Nitrobenzyl Compounds: Methyl Ethers and Caged
ATP. J. Am. Chem. Soc. 2004, 126, 4581−4595.
(43) Yang, Y.; Shao, Q.; Deng, R.; Wang, C.; Teng, X.; Cheng, K.;
Cheng, Z.; Huang, L.; Liu, Z.; Liu, X.; Xing, B. In Vitro and In Vivo
Uncaging and Bioluminescence Imaging by Using Photocaged
Upconversion Nanoparticles. Angew. Chem., Int. Ed. 2012, 51, 3125−
3129.
8459
dx.doi.org/10.1021/la500744s | Langmuir 2014, 30, 8452−8460

Langmuir
Article
(44) Klein, D. Organic Chemistry, 2nd ed.; John Wiley and Sons:
Hoboken, N.J., 2011.
(45) Rashidian, M.; Mahmoodi, M. M.; Shah, R.; Dozier, J. K.;
Wagner, C. R.; Distefano, M. D. A. Highly Efficient Catalyst for Oxime
Ligation and Hydrazone-Oxime Exchange Suitable for Bioconjugation.
Bioconjugate Chem. 2013, 24, 333−342.
(46) Crisalli, P.; Kool, E. T. Water-Soluble Organocatalysts for
Hydrazone and Oxime Formation. J. Org. Chem. 2013, 78, 1184−1189.
8460
dx.doi.org/10.1021/la500744s | Langmuir 2014, 30, 8452−8460