Impact of Molecular Architecture and Adsorption Density on Adhesion of Mussel-Inspired Surface Primers with Catechol-Cation Synergy

Article abstract of DOI:10.1021/jacs.9b04337

Marine mussels secrete proteins rich in residues containing catechols and cationic amines that displace hydration layers and adhere to charged surfaces under water via a cooperative binding effect known as catechol-cation synergy. Mussel-inspired adhesives containing paired catechol and cationic functionalities are a promising class of materials for biomedical applications, but few studies address the molecular adhesion mechanism(s) of these materials. To determine whether intramolecular adjacency of these functionalities is necessary for robust adhesion, a suite of siderophore analog surface primers was synthesized with systematic variations in intramolecular spacing between catechol and cationic functionalities. Adhesion measurements conducted with a surface forces apparatus (SFA) allow adhesive failure to be distinguished from cohesive failure and show that the failure mode depends critically on the siderophore analog adsorption density. The adhesion of these molecules to muscovite mica in an aqueous electrolyte solution demonstrates that direct intramolecular adjacency of catechol and cationic functionalities is not necessary for synergistic binding. However, we show that increasing the catechol-cation spacing by incorporating nonbinding domains results in decreased adhesion, which we attribute to a decrease in the density of catechol functionalities. A mechanism for catechol-cation synergy is proposed based on electrostatically driven adsorption and subsequent binding of catechol functionalities. This work should guide the design of new adhesives for binding to charged surfaces in saline environments.

Full text of DOI:10.1021/jacs.9b04337

Article
pubs.acs.org/JACS
Cite This: J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Impact of Molecular Architecture and Adsorption Density on
Adhesion of Mussel-Inspired Surface Primers with Catechol-Cation
Synergy
George D. Degen,† Parker R. Stow,‡ Robert B. Lewis,‡ Roberto C. Andresen Eguiluz,† Eric Valois,§
Kai Kristiansen,† Alison Butler,*,‡ and Jacob N. Israelachvili⊥,†,∥
‡
§
†Department of Chemical Engineering, Department of Chemistry and Biochemistry, Biomolecular Science and Engineering, and
∥Materials Department, University of California, Santa Barbara, California 93106, United States
S
* Supporting Information
ABSTRACT: Marine mussels secrete proteins rich in residues containing
catechols and cationic amines that displace hydration layers and adhere to
charged surfaces under water via a cooperative binding effect known as
catechol-cation synergy. Mussel-inspired adhesives containing paired
catechol and cationic functionalities are a promising class of materials for
biomedical applications, but few studies address the molecular adhesion
mechanism(s) of these materials. To determine whether intramolecular
adjacency of these functionalities is necessary for robust adhesion, a suite
of siderophore analog surface primers was synthesized with systematic
variations in intramolecular spacing between catechol and cationic
functionalities. Adhesion measurements conducted with a surface forces
apparatus (SFA) allow adhesive failure to be distinguished from cohesive
failure and show that the failure mode depends critically on the
siderophore analog adsorption density. The adhesion of these molecules to muscovite mica in an aqueous electrolyte solution
demonstrates that direct intramolecular adjacency of catechol and cationic functionalities is not necessary for synergistic
binding. However, we show that increasing the catechol-cation spacing by incorporating nonbinding domains results in
decreased adhesion, which we attribute to a decrease in the density of catechol functionalities. A mechanism for catechol-cation
synergy is proposed based on electrostatically driven adsorption and subsequent binding of catechol functionalities. This work
should guide the design of new adhesives for binding to charged surfaces in saline environments.
catechol-poly(ethylenimine),14,15 among others.16,17 Molecules
1. INTRODUCTION
containing Dopa and other catechols adhere to apatite18 and
Rational design of wet adhesives requires an understanding of
metal oxide19 surfaces found in bone and implant materials,
the intermolecular interactions between adhesives and
and have been shown to be biocompatible,20 making them
substrates in saline environments such as seawater or body
attractive alternatives to existing medical adhesives, many of
fluids. Designing adhesives for use in these environments is
which are ineffective21,22 or cytotoxic.23−25
challengingfor example, van der Waals forces are signifi-
Despite widespread scientific and applications-based re-
cantly reduced under water, and ions in solution compete with
search on materials containing catechol functionalities,20
adhesives for binding sites on charged surfaces. Despite these
challenges, marine mussels adhere to many inorganic and
specific adhesion mechanisms of these materials have only
recently begun to be explored.26 It has been demonstrated that
adhesives incorporating catechol and cationic amine function-
alities bind more strongly to muscovite mica in saline solutions
than adhesives incorporating either catechols or cations
alone,27 a cooperative effect known as catechol-cation synergy.
The adjacent pairing of Dopa and lysine in the Mfp-5
sequence5 has prompted speculation that intramolecular
proximity of catechol and cationic functionalities may be
necessary for adhesion, and that an intramolecular cutoff
distance may exist beyond which catechol-cation synergy no
organic surfaces,1−3 typically by relying on adhesive proteins
rich in the catecholic amino acid 3,4-dihydroxyphenylalanine
(Dopa).4 Due to the large proportion of cationic residues,
most commonly lysine, paired with Dopa in the most adhesive
interfacial mussel protein (Mfp-5),1,5 it has been hypothesized
that both these residues are important for adhesion.
Furthermore, other compounds containing catechol and
cationic functionalities have been shown to adhere to many
surfaces. Polydopamine, formed from the polymerization and
self-assembly of oxidized dopamine,6−9 adheres to a wide
variety of materials and has been proposed for many
applications.10 Other adhesives containing catechol and
cationic functionalities include catechol-chitosan11−13 and
Received: April 22, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.9b04337
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
longer operates.28 A recent study showed that the order of
catechol and cationic functionalities impacts the single-
molecule pull-off force,29 supporting the hypothesis that direct
adjacency of catechol and cationic functionalities may be
enhance adhesion. However, until now no study has directly
explored the impact of intramolecular spacing on catechol-
cation synergy.
Here, we present adhesion measurements of siderophore
analog surface primers with systematically varying intra-
molecular spacing between catechol and cationic amine
functionalities. We show that the pull-off force mediated by
the siderophore analogs depends critically on the adsorption
density and confirm that catechol-cation synergy enables
adhesion. Surprisingly, the results demonstrate that direct
intramolecular adjacency between catechol and cationic
functionalities is not necessary for catechol-cation synergy
and suggest that no intramolecular cutoff distance between
these functionalities exists for which catechol-cation synergy
will be abolished. To explain the results, we propose a
mechanism for catechol-cation synergy based on electrostati-
cally driven adsorption and support this mechanism with a
qualitative model.
2. EXPERIMENTAL SECTION
We synthesized a suite of seven siderophore analogs, synthetic mimics
of bacterial iron chelators called siderophores.30 Each analog has a
central tris(2-aminoethyl)amine (Tren) scaffold with three identical
peptide arms. Here, each peptide arm contained glycine (G) and
lysine (K) and were capped with either the catechol 2,3-
dihydroxybenzoyl (2,3-DHBA) or benzoyl functionality (Figure 1).
The intramolecular spacing between 2,3-DHBA and lysine in the
peptide arms was varied by changing the peptide sequence. Three of
the analogs were isomers: Tren(GGK-Cat)3, Tren(GKG-Cat)3, and
Tren(KGG-Cat)3. Siderophore analogs with even greater catechol-
cation spacing (Tren(KGGG-Cat)3 and Tren(KGGGGGG-Cat)3),
cationic amines but without catechols (Tren(GGK-Benz)3), and only
catechols (Tren(GGG-Cat)3) were also synthesized. Additional
details on materials, synthesis, and molecular characterization are
included in the Supporting Information, SI, (S1 and S2). Analogs
were dissolved at 1 mM in an aqueous salt solution (50 mM acetic
acid, 150 mM KNO3, pH = 3.3), chosen to mimic physiological ionic
strengths but avoid catechol oxidation by maintaining an acidic pH.
Adhesion measurements were performed using a surface forces
apparatus (SFA) model SFA2000 (SurForce, LLC), described in
detail in the SI (S3) and elsewhere.31 In the SFA, freshly cleaved
muscovite mica surfaces were arranged in a crossed cylinder geometry.
All experiments were performed at a constant temperature. Side-
rophore analogs were deposited via adsorption from solution into
films on either one (asymmetric deposition) or both (symmetric
deposition) mica surfaces. For asymmetric deposition, 50 μL of 400
μM siderophore analog solution was injected onto one of the mica
surfaces and incubated for at least 60 min. The surface was then
rinsed before adhesion measurements. For symmetric deposition,
siderophore analog solution was injected into a capillary meniscus
between both surfaces (final concentration 90−667 μM). The
surfaces were incubated for at least 60 min and were not rinsed
before adhesion measurements.
Figure 1. Suite of siderophore analogs investigated. (A) Tris(2-
aminoethyl)amine (Tren) scaffold. R-groups are shown in (B−H).
The intramolecular distance between 2,3-DHBA and lysine is varied
in (B) Tren(GGK-Cat)3, (C) Tren(GKG-Cat)3, (D) Tren(KGG-
Cat)3, (E) Tren(KGGGGGG-Cat)3, and (F) Tren(KGGG-Cat)3.
Analogs containing lysine without 2,3-DHBA, (G) Tren(GGK-
Benz)3, and 2,3-DHBA without lysine, (H) Tren(GGG-Cat)3 were
also synthesized.
during separation corresponding to the most negative force measured
during separation. Pull-off force did not depend on separation velocity
(Vout = 2−10 nm/s) (Figure S32) nor maximum compression (F/R =
9−108 mN/m). We use the term “pull-off” instead of “adhesion”
because the force required to separate the mica surfaces can
correspond to adhesive failure at the film−mica interface, cohesive
failure at the film−film interface, or a combination of the two failure
modes. The separation distance during compression of the surfaces at
which the force exceeded 1 mN/m was denoted the onset of
interaction (Donset). The surface separation distance measured at
maximum compression was denoted compressed film thickness (Dt).
The change in film thickness after waiting at maximum compression
(ΔDt) was reported as the difference between compressed film
thicknesses measured before and after waiting at maximum
compression. Error bars correspond to the standard deviation, with
an additional contribution to the error in Dt for 6 of the data points
from measuring Dt relative to the Dt measured in salt solution. To
characterize film coverage, atomic force microscopy (AFM) imaging
was performed on mica immersed in salt solution (50 mM acetic acid,
150 mM KNO3, pH = 3.3) with an MFP-3D Bio AFM (Asylum
Research, Goleta, CA), described in the SI (S4).
Adhesion measurements were performed with the mica surfaces
bridged by a capillary meniscus of ∼50 μL salt solution (50 mM acetic
acid, 150 mM KNO3, pH = 3.3) at a constant temperature (T = 22
1 °C). Normal force (F) and surface separation distance (D) were
measured during cycles of (i) approach and compression, (ii) waiting
at maximum compression (twait), and (iii) separation and jump from
contact were performed at constant approach and separation
velocities. Measured forces were normalized by the average radius
of curvature of the surfaces (R). Pull-off force (−Fad/R) was
calculated from the distance that the surfaces jumped from contact
B
DOI: 10.1021/jacs.9b04337
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
3. RESULTS AND DISCUSSION
Article
We used a surface forces apparatus (SFA) to directly test
whether intramolecular proximity is necessary for catechol-
cation synergy by measuring the adhesion to mica of
siderophore analogs with systematically varying spacing
between catechol and cationic amine functionalities. In
adhesion measurements, two failure modes can contribute to
the measured pull-off force: adhesive failure (separation at the
film−mica interface) and cohesive failure (separation at the
film−film interface). When adhesive failure occurs, the pull-off
force corresponds to intermolecular interactions between
siderophore analogs and mica, which can include bidentate
hydrogen bonds and coordinate covalent bonds.4 When
cohesive failure occurs, the pull-off force corresponds to
intermolecular interaction between siderophore analogs
adsorbed on each mica surface, which can include hydrogen
bonds, hydrophobic interactions, and cation-π interactions.4
Catechol−cation synergy refers to cooperative binding of
molecules to a substrate. Therefore, to assess the impact of
molecular structure on catechol−cation synergy, it is necessary
to measure pull-off forces corresponding to adhesive failure.
For the case of molecularly smooth mica surfaces (Figure S41),
a monolayer film between the surfaces guarantees adhesive
failure because each siderophore analog within the monolayer
can bind to both mica surfaces, and therefore separation must
occur at the film−mica interface. Below, we establish the
deposition conditions resulting in a monolayer of siderophore
analogs between the surfaces, and therefore adhesive failure.
We then confirm that catechol-cation synergy occurs and
discuss the impact of intramolecular catechol-cation spacing on
the synergy.
Compression and separation of films of siderophore analogs
in an SFA enabled measurement of pull-off force (−Fad/R),
onset of interaction (Donset), and compressed film thickness
(Dt). Figure 2 shows representative plots of normal force (F/
R) vs surface separation distance (D) measured for films of
Tren(GGK-Cat)3, Tren(GKG-Cat)3, and Tren(KGG-Cat)3.
Each plot corresponds to an experiment conducted using a
single pair of mica surfaces. First, bare mica surfaces were
compressed and separated in salt solution (black circles). Next,
a film of siderophore analogs was deposited onto one of the
mica surfaces via asymmetric deposition, followed by
compression and separation of the surfaces (red circles).
Finally, analogs were deposited symmetrically onto both
surfaces, and the surfaces were again compressed and separated
(blue circles). Open circles correspond to approach and
compression of the surfaces; closed circles correspond to
Figure 2. Plots of normal force (F/R) vs separation distance (D) for
bare mica surfaces (black circles) and after asymmetric (red circles)
and symmetric (blue circles) depositions of (A) Tren(GGK-Cat)3,
(B) Tren(GKG-Cat)3, and (C) Tren(KGG-Cat)3. Open circles show
approach and compression of the surfaces; closed circles show
separation and jump from contact.
separation and jump from contact. In each plot, Fad/R, Donset
,
and Dt are indicated.
Pull-off force, onset of interaction, and contact time
dependence measured during compression and separation of
films of siderophore analogs depended on the compressed film
thickness. Figure 3 shows plots of pull-off force (−Fad/R) as a
function of compressed film thickness (Dt). For each analog,
Dt < 10 Å corresponds to low pull-off force. Pull-off force is
maximized at Dt = 10 Å, and the maximum pull-off forces
mediated by each analog are not statistically significantly
different (α = 0.05) (Figure S33). As Dt increases 10 Å, pull-off
force decreases. Interestingly, the analog with the greatest
separation between catechol and cationic functionalities
(Tren(KGG-Cat)3) mediates larger pull-off forces than
Figure 3. Plots of pull-off force (−Fad/R) vs compressed film
thickness (Dt) for Tren(GGK-Cat)3 (red circles), Tren(GKG-Cat)3
(blue circles), and Tren(KGG-Cat)3 (black circles). Lines are
included to guide the eye.
Tren(GKG-Cat)3 and Tren(GGK-Cat)3 for 15 Å < Dt < 20
Å, discussed later.
C
DOI: 10.1021/jacs.9b04337
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
The compressed film thickness corresponding to maximum
pull-off force (Dt = 10 Å) also corresponds to a discontinuous
increase in the onset of interaction (Donset). Figure 4 shows
s (Ft = 10 s), as a function of Dt. For all analogs, increased
contact time generally results in increased pull-off force (Ft =
60 min/Ft = 60 min > 1). However, the increase in pull-off force is
minimized for Dt = 10 Å (Ft = 60 min/Ft = 10 s = 1). Increased
contact time also decreases Dt. Figure 5B shows the change in
compressed film thickness ΔDt as a function of Dt measured
before the wait time. Open circles correspond to short waiting
times (twait = 10 s) for which Dt did not decrease (ΔDt = 0). In
contrast, closed circles correspond to longer contact times (twait
= 60 min) and decreases in Dt (ΔDt > 0), with the largest
decreases occurring for Dt > 10 Å.
As shown above, the maximum pull-off force, discontinuous
increase in the onset of interaction, and minimum increase in
pull-off force with waiting time all occur for the same
compressed film thickness (Dt = 10 Å). Taken together, the
relationships between these quantities suggest that this film
thickness corresponds to a monolayer of siderophore analogs
between the surfaces. For Dt > 10 Å, a transition occurs from a
single monolayer between the mica surfaces to two
monolayers, one on each mica surface. This transition results
in a corresponding transition from adhesive failure to cohesive
failure. For reference, Figure 6 shows a schematic diagram of
mica surfaces before and after asymmetric or symmetric
deposition of siderophore analogs at different surface densities.
(i) and (iv) show mica surfaces prior to deposition of
siderophore analogs. (ii), (iii), (v), and (vi) show single
monolayers between the surfaces and adhesive failure. (vii)
shows a monolayer on each surface and cohesive failure.
Below, we relate our results to a transition from adhesive
failure to cohesive failure and interpret measured pull-off forces
in the context of catechol-cation synergy.
Figure 4. Plots of onset of interaction (Donset) vs compressed film
thickness (Dt) for Tren(GGK-Cat)3 (red circles), Tren(GKG-Cat)3
(blue circles), and Tren(KGG-Cat)3 (black circles).
plots of Donset as a function of Dt. As Dt increases from 0 to 10
Å, Donset increases from 5 to 15 Å. At Dt = 10 Å, Donset increases
discontinuously from 15 to 40 Å. As Dt increases further from
10 to 30 Å, Donset increases from 40 to 60 Å. Dt = 10 Å also
corresponds to a minimum increase in pull-off force with
increased contact time. Figure 5A shows pull-off force for twait
= 60 min (Ft = 60 min), normalized by pull-off force for twait = 10
We attribute the low pull-of force and onset of interaction
for Dt < 10 Å to a sparse monolayer between the mica surfaces.
A sparse monolayer contains relatively few siderophore analog
molecules binding to both surfaces, and therefore is expected
to mediate a low pull-off force, consistent with the data shown
in Figure 3. A sparse monolayer between the surfaces can be
established by either asymmetric deposition (Figure 6 (ii)) or
symmetric deposition (Figure 6 (v)) with a low concentration
of siderophore analogs in solution. We note that waiting at
maximum compression results in disproportionate increases in
the pull-off force mediated by Tren(GKG-Cat)3 relative to the
other analogs for Dt < 10 Å (Figure 5A, blue circles).
Separation of the glycine residues in Tren(GKG-Cat)3 may
reduce the conformational flexibility and inhibit initial binding
to both mica surfaces. Time in contact may enable rearrange-
ment and binding to both mica surfaces of individual
Tren(GKG-Cat)3 molecules, thus increasing the pull-off force.
The maximum pull-off force and minimum increase in pull-
off force with waiting time occur for Dt = 10 Å, suggesting that
this film thickness corresponds to a single densely packed
monolayer between the surfaces. The ability of siderophore
analogs to form films of varying density was confirmed with
AFM imagingincreasing the concentration of siderophore
analogs in solution during incubation increases the density of
the film on the mica surface (Figure S31). Like a sparse
monolayer, a densely packed monolayer can result from
asymmetric deposition of siderophore analogs onto one of the
surfaces (Figure 6 (iii)), or from symmetric deposition of a
sparse monolayer of analogs onto both surfaces (Figure 6 (vi)).
When these sparse monolayers are brought into contact, they
combine to form a densely packed monolayer. Regardless of
the deposition method, a densely packed monolayer is
Figure 5. (A) Plots of normalized pull-off force (Ft = 60 min/Ft = 10 s) vs
compressed film thickness (Dt) for Tren(GGK-Cat)3 (red circles),
Tren(GKG-Cat)3 (blue circles), and Tren(KGG-Cat)3 (black circles).
Lines are included to guide the eye. (B) Corresponding plots of
change in film thickness (ΔDt) vs Dt for twait = 10 s (open circles) and
twait = 60 min (closed circles).
D
DOI: 10.1021/jacs.9b04337
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 6. A transition from one monolayer (adhesive failure) to two monolayers (cohesive failure) decreases pull-off force (−Fad/R) and increases
compressed film thickness (Dt). (A) Asymmetric deposition of siderophore analogs on a single surface yields adhesive failure, shown in (ii) and
(iii). (B) Symmetric deposition of siderophore analogs on both surfaces results in either adhesive failure shown in (v) and (vi) or cohesive failure
shown in (vii). (ii) and (v) show configurations with the same pull-off force; configurations with maximum pull-off force are shown in (iii) and (vi).
Bare mica surfaces before deposition of siderophore analogs are shown in (i) and (iv).
expected to maximize the number of siderophore analogs
binding to both mica surfaces and therefore maximize the pull-
off force, consistent with the data shown in Figure 3. The
minimum increase in pull-off force with waiting time also
occurs at Dt = 10 Å, providing additional evidence that this
compressed film thickness corresponds to a densely packed
monolayer. Dense packing of analogs on a surface and a
corresponding low surface area per molecule may prevent
functional groups in each molecule from binding to the surface,
leaving the groups free to bind to the adjacent surface upon
contact. Therefore, increased contact time does not change the
distribution of functionalities binding to each surface and pull-
off force remains constant.
The discontinuous increase in onset of interaction for Dt >
10 Å indicates a transition from a single densely packed
monolayer to two monolayers between the surfaces. With a
monolayer on each surface, repulsive forces begin when the
films on each surface contact each other. This distance (Donset
= 40 Å) is slightly more than double the onset of interaction
for a single densely packed monolayer (Donset = 15 Å),
consistent with a transition from one to two monolayers. The
largest compressed film thickness (Dt = 30 Å) corresponds to
Donset = 60 Å. This value is larger than would be expected for
symmetric densely packed monolayers, suggesting that addi-
tional siderophore analogs can adsorb onto the monolayers on
each surface. However, no evidence of an adsorbed layer
beyond a monolayer is seen after asymmetric deposition,
suggesting that the siderophore analogs loosely adsorb to the
monolayer and are removed during the rinsing associated with
asymmetric deposition. We note that Dt is not expected to
increase discontinuously during a transition from one to two
monolayers because Dt depends on the adsorption density of
the monolayers. A densely packed monolayer is expected to
have a larger Dt than a sparse monolayer because individual
molecules occupy less area on the surface and therefore extend
further into solution. Similar behavior is also seen in adsorbed
surfactant monolayers.32 As a result, the discontinuous increase
in Donset reveals the transition from one to two monolayers in a
way that Dt does not.
The decreasing pull-off force with increasing film thickness
for Dt > 10 Å provides additional evidence of a transition from
a single densely packed monolayer to two monolayers between
the surfaces by indicating a transition from adhesive failure to
cohesive failure. An adhesive system is expected to fail at the
weakest interface. If the film−film interface were stronger than
the film−mica interface, then adhesive failure would continue
to occur at the film−mica interface, regardless of the value of
Dt. In that case, pull-off force would not decrease with
increasing Dt. Here, the pull-off force decreases with increasing
Dt, suggesting a transition from adhesive to cohesive failure.
E
DOI: 10.1021/jacs.9b04337
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
We propose that as the monolayers on each surface become
more densely packed, peptide arms on siderophore analogs in
each monolayer become unable to penetrate the adjacent
monolayer and therefore fail to bind to both mica surfaces.
Instead, pull-off forces correspond to cohesive failure and
interactions between analogs in opposite films (Figure 6 (vii)).
The decrease in molecular interdigitation with increasing film
density is analogous to the behavior of polymer brushes used
for antibiofouling surfaces: increasing polymer grafting density
decreases the ability of small peptide adhesives to penetrate the
brush layer and bind to the underlying substrate.33 Consistent
with this interpretation, the highest incubation concentration
of siderophore analogs during symmetric deposition resulted in
the lowest pull-off forces (Figure S31). We note that when a
bare mica surface is brought into contact with a monolayer of
siderophore analogs on the opposite surface (as occurs after
asymmetric deposition), siderophore analogs may transfer
from one surface to the other upon separation of the surfaces,
ultimately resulting in the same distribution of siderophore
analogs on each surface as the symmetric deposition
configuration shown in Figure 6 (vi).
The increase in pull-off force and decrease in film thickness
with waiting time for Dt > 10 Å suggest a partial transition
from cohesive failure to adhesive failure. We attribute this
transition to interdigitation of siderophore analogs in
monolayers on each surface. Here, interdigitation refers to
peptide arms of the siderophore analogs penetrating the
opposite film and binding to the underlying mica surface.
Increasing the fraction of individual siderophore analogs
binding to both mica surfaces results in an increased ratio of
adhesive failure to cohesive failure, and consequently an
increased pull-off force. Interdigitation likely also results in
partial coalescence of the two films, consistent with the
decrease in film thickness (ΔDt > 0) after increased waiting
time reported in Figure 5B for Dt > 10 Å.
As stated above, Tren(KGG-Cat)3, mediated larger pull-off
forces than Tren(GKG-Cat)3 and Tren(GGK-Cat)3 for 15 Å <
Dt < 20 Å (Figure 3). The increased pull-off force at these film
thicknesses suggests that the glycine residues in Tren(KGG-
Cat)3 give the catechols independent mobility from the
surface-bound lysines,34 enabling them to penetrate the film
on the adjacent mica surface and bind to both mica surfaces.
Films of Tren(KGG-Cat)3 also show the largest increases in
pull-off force (Ft = 60 min/Ft = 10 s > 1) and decreases in film
thickness (ΔDt), consistent with increased mobility of catechol
functionalities. Alternatively, because the lysine residues of
Tren(KGG-Cat)3 are close to the cationic Tren core, positive
charge is localized at the center of the molecule. This localized
charge density may enhance intermolecular cation-π inter-
actions and strengthen cohesion between symmetric mono-
layers, consistent with recent studies demonstrating the
importance of cation-π interactions in the adhesion and
cohesion of materials containing catechol and cationic
functionalities.35,36 Under this interpretation, the low pull-off
force mediated by Tren(GGK-Cat)3 at Dt = 30 Å suggests that
the monolayers become sufficiently densely packed to bury the
charged groups and inhibit cation-π interactions between
adjacent films.
tative force−distance plots shown in Figures S34 and S36.
Consistent with previous studies,27,28 these molecules medi-
ated much lower pull-off forces than siderophore analogs
containing catechol and cationic amine functionalities. Films of
Tren(GGK-Benz)3 showed a > 50% decrease in pull-off forces
relative to Tren(GGK-Cat)3 (Figure S33). The pull-off force
likely results from adhesive electrostatic interactions between
pendant cationic amines and the mica surfaces and cohesive
hydrophobic and cation−π interactions. Films of Tren(GGK-
Benz)3 also exhibited long-range repulsion on compression
that decreased over sequential compression and separation
cycles, indicating rearrangement of adsorbed aggregates.
Properties of films of Tren(GGK-Benz)3 are described in
Figure S35. Tren(GGG-Cat)3 showed no evidence of
adsorption in the SFA after a 5-h incubation, with onset of
interaction, compressed film thickness, and pull-off force
remaining identical to the values measured for bare mica
surfaces in salt solution. Over 144 h, pull-off force progressively
decreased from ∼3 mN/m (mica−mica adhesion in salt
solution) to zero as the compressed film thickness increased
from <1 to 50 nm. These changes indicate the adsorption of
multilayer aggregates on the mica surfaces, likely driven by
electrostatic attraction between the cationic Tren scaffold and
the negatively charged mica. The presence of adsorbed
aggregates after 24- and 48-h incubations of Tren(GGG-
Cat)3 on mica was confirmed with AFM (Figure S43). Pull-off
force mediated by the aggregates depends on the separation
velocity (Figure S36), suggesting that the aggregates are weakly
associated and that the pull-off force results from energy
dissipation. The delayed adsorption of Tren(GGG-Cat)3 (over
hours rather than minutes) suggests that cationic amines of the
lysine residues drive adsorption onto the mica surface.
Therefore, hydrogen bonds (involving the catechol function-
alities or the peptide backbones of each arm of the siderophore
analogs), electrostatic interactions involving the Tren core, and
nonspecific van der Waals interactions are insufficient to drive
rapid adsorption of siderophore analogs into monolayers on
the mica.
Above, we demonstrate the two criteria necessary for
confirming the presence of catechol-cation synergy. We
identify pull-off forces corresponding to a monolayer of
siderophore analogs, therefore guaranteeing adhesive failure.
We then show that siderophore analogs with catechol and
cationic amine functionalities mediate significantly larger
adhesion than analogs with either catechols or cationic amines
alone. Importantly, the molecular weight and density of
catechol and cationic amine functionalities were the same for
Tren(GGK-Cat)3, Tren(GKG-Cat)3, and Tren(KGG-Cat)3.
Since molecular weight37 and density of binding function-
alities28 influence the adsorption and adhesion of small
molecules, keeping these quantities constant enables direct
comparison of the adhesion forces to assess the impact of
intramolecular spacing on catechol-cation synergy. Surpris-
ingly, increasing the intramolecular catechol-cation spacing by
up to two glycine residues does not abolish catechol-cation
synergyTren(GGK-Cat)3, Tren(GKG-Cat)3, and Tren-
(KGG-Cat)3 all mediate the same maximum pull-off force
(Figure S33). Therefore, direct intramolecular proximity is not
necessary for catechol-cation synergy.
To confirm that catechol-cation synergy was occurring, we
measured forces mediated by siderophore analogs with cationic
amines but without catechols (Tren(GGK-Benz)3, Figure 1G)
and forces mediated by analogs with only catechol
functionalities (Tren(GGG-Cat)3, Figure 1H), with represen-
To further explore the impact of catechol-cation spacing on
adhesion, we synthesized siderophore analogs with three and
six glycine residues separating 2,3-DHBA and lysine, Tren-
(KGGG-Cat)3 and Tren(KGGGGGG-Cat)3, respectively. As
F
DOI: 10.1021/jacs.9b04337
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
expected, both molecules mediate substantial pull-off forces
despite the increased catechol-cation spacing (Figures S38 and
S40). However, the pull-off forces were lower than the pull-off
forces for Tren(GGK-Cat)3, Tren(GKG-Cat)3, and Tren-
(KGG-Cat)3. Figure 7 shows a plot of pull-off force vs
nm), the cationic amine is expected to detach first during
adhesive failure due to geometric considerations. For the case
where the catechol and cationic amine bind relatively far apart
on the mica (∼2 nm), simultaneous detachment is possible.
However, such distant binding is unlikely for entropic reasons,
and therefore sequential catechol−cation detachment from the
mica is expected to occur for the majority of Tren(KGG-Cat)3
molecules in the contact area. In contrast, simultaneous
detachment of catechol and cationic functionalities from the
mica surface is more likely for Tren(GGK-Cat)3 due to the
intramolecular adjacency of those functionalities. If detach-
ment order were necessary for catechol−cation synergy in our
experiments, then Tren(GGK-Cat)3 would be expected to
mediate larger adhesion forces than Tren(KGG-Cat)3. Since
Tren(GGK-Cat)3, Tren(GKG-Cat)3, and Tren(KGG-Cat)3
mediate the same adhesion forces, we conclude that detach-
ment order of catechol and cationic amine functionalities does
not contribute to catechol-cation synergy in this work.
On the basis of our results, we propose the following
mechanism for catechol-cation adhesion synergy: pendant
cationic amines of the siderophore analogs exchange with
adsorbed cations on the mica surface and drive adsorption
onto the mica, enabling subsequent binding of catechols to the
mica. With a monolayer of siderophore analogs on the surface,
the effective concentration of catechols within 1 nm of the
surface is ∼3 M, much greater than the bulk siderophore
analog concentration (90−667 μM). This effective catechol
concentration is calculated assuming that siderophore analogs
bind to every negative charge on the mica lattice (1 e− per 0.5
nm2). We suggest that the increased concentration of catechols
near the mica surface increases the probability of catechols
replacing surface-bound cations and binding to the mica.
To justify our mechanism for catechol-cation synergy we
developed a qualitative model based on Bell Theory38 to
predict the lifetime and fractional surface coverage of cationic
species adsorbed on the mica surface, further described in the
SI (S5). The model predicts that siderophore analogs can
adsorb to mica solely via their cationic amines. While the
assumptions made in the derivation of the model preclude
quantitative comparison with experiments, the predictions are
qualitatively consistent with our experimental results showing
that siderophore analogs lacking cationic functionalities do not
adsorb into adhesive monolayers on mica in an aqueous
electrolyte solution. As such, the role of cations in catechol−
cation synergy is to drive adsorption onto negatively charged
surfaces and enable subsequent binding of catechol function-
alities. The cooperative effect operates irrespective of the
intramolecular catechol−cation spacing. This result is surpris-
ing given that the majority of catechols in the most adhesive
interfacial mussel protein are located directly adjacent to
cationic amines and raises a fundamental biological question
about the evolutionary pressure(s) responsible for this residue
distribution.
Figure 7. Plot of pull-off force (−Fad/R) for twait = 10 s vs molecular
weight (MW) for siderophore analogs containing catechol and
cationic amine functionalities. Data for Tren(K-Cat)3 (twait = 2 min)
and Tren(KK-Cat)3 (twait = 10 min) reproduced from refs 27 and 28,
respectively. Lines included to guide the eye.
molecular weight for various siderophore analogs containing
catechol and cationic amine functionalities. The adhesion of
Tren(K-Cat)3 reported in a previous study27 remains the
highest of all the siderophore analogs studied thus far due to its
relatively low molecular weight and correspondingly large
catechol-cation density. As molecular weight increases, pull-off
force monotonically decreases, which we attribute to the
decreasing density of catechol functionalities within each
molecule. The gradually decreasing pull-off force due to
decreasing binding group density is fundamentally different
from an abrupt decrease in pull-off force at some intra-
molecular cutoff distance that abolishes catechol−cation
synergy. Since adhesion force decreases gradually with
catechol-cation spacing up to a spacing of six glycine residues,
our results suggest that no such cutoff distance exists, and that
further increases in catechol-cation spacing will continue to
gradually decrease the adhesion force even as catechol-cation
synergy persists.
The results also suggest that the simultaneous detachment of
catechol and cationic functionalities from the surface does not
contribute to catechol−cation synergy in our experiments.
Detachment order has been proposed to contribute to
catechol-cation synergy in single-molecule adhesion studies,
where the pulling geometry is precisely determined.29 Unlike
single-molecule studies, our experiments involve ∼108 side-
rophore analog molecules binding to mica (assuming each
analog occupies 1 nm2 on the mica surface) and a distribution
of binding geometries. For example, the adjacent glycine
residues in the peptide arms of Tren(KGG-Cat)3 are expected
to give conformational flexibility to the molecule34 and
consequently enable a range of distances between surface-
bound catechol and cationic amine functionalities in a single
siderophore analog arm. For the case where the catechol and
cationic amine bind to the mica surface in close proximity (<1
4. CONCLUSIONS
This work explores the effect of intramolecular separation of
catechol and cationic functionalities on adhesion mediated by
monolayers of siderophore analog surface primers. Our results
demonstrate that the pull-off force required to separate mica
surfaces depends critically on the siderophore analog
adsorption density, highlighting the importance of failure
mode on adhesive performance. Furthermore, direct intra-
molecular adjacency of catechol and cationic amine function-
G
DOI: 10.1021/jacs.9b04337
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(6) Lee, H.; Dellatore, S. M.; Miller, W. M.; Messersmith, P. B.
Mussel-Inspired Surface Chemistry for Multifunctional Coatings.
Science 2007, 318, 426.
(7) Hong, S.; Na, Y. S.; Choi, S.; Song, I. T.; Kim, W. Y.; Lee, H.
Non-Covalent Self-Assembly and Covalent Polymerization Co-
Contribute to Polydopamine Formation. Adv. Funct. Mater. 2012,
22 (22), 4711−4717.
(8) Liebscher, J.; Mrowczynski, R.; Scheidt, H. A.; Filip, C.; Hadade,
N. D.; Turcu, R.; Bende, A.; Beck, S. Structure of Polydopamine: A
Never-Ending Story? Langmuir 2013, 29 (33), 10539−10548.
(9) Delparastan, P.; Malollari, K. G.; Lee, H.; Messersmith, P. B.
Direct Evidence for the Polymeric Nature of Polydopamine. Angew.
Chem., Int. Ed. 2019, 58 (4), 1077−1082.
(10) Ryu, J. H.; Messersmith, P. B.; Lee, H. Polydopamine Surface
Chemistry: A Decade of Discovery. ACS Appl. Mater. Interfaces 2018,
10 (9), 7523−7540.
(11) Kim, K.; Kim, K.; Ryu, J. H.; Lee, H. Chitosan-Catechol: A
Polymer with Long-Lasting Mucoadhesive Properties. Biomaterials
2015, 52 (1), 161−170.
(12) Shin, M.; Park, S. G.; Oh, B. C.; Kim, K.; Jo, S.; Lee, M. S.; Oh,
S. S.; Hong, S. H.; Shin, E. C.; Kim, K. S.; Kang, S. W.; Lee, H.
Complete Prevention of Blood Loss with Self-Sealing Haemostatic
Needles. Nat. Mater. 2017, 16 (1), 147−152.
(13) Ryu, J. H.; Hong, S.; Lee, H. Bio-Inspired Adhesive Catechol-
Conjugated Chitosan for Biomedical Applications: A Mini Review.
Acta Biomater. 2015, 27, 101−115.
(14) Lee, H.; Park, T. G.; Messersmith, P. B.; Lee, Y.; Statz, A. R.;
Rho, J. Substrate-Independent Layer-by-Layer Assembly by Using
Mussel-Adhesive-Inspired Polymers. Adv. Mater. 2008, 20 (9), 1619−
1623.
(15) Kim, E.; Song, I. T.; Lee, S.; Kim, J. S.; Lee, H.; Jang, J. H.
Drawing Sticky Adeno-Associated Viruses on Surfaces for Spatially
Patterned Gene Expression. Angew. Chem., Int. Ed. 2012, 51 (23),
5598−5601.
(16) Statz, A. R.; Meagher, R. J.; Barron, A. E.; Messersmith, P. B.
New Peptidomimetic Polymers for Antifouling Surfaces. J. Am. Chem.
Soc. 2005, 127 (22), 7972−7973.
alities is not necessary for catechol−cation synergy. Instead,
increasing the intramolecular catechol-cation spacing in an
adhesive by the addition of nonbinding domains progressively
reduces adhesion due to the reduced density of binding groups.
In summary, the results presented here explain the synergistic
binding of catechol and cationic functionalities and should
guide the design of new adhesives for binding to negatively
charged surfaces in saline environments.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.9b04337.
■
S
Details on the synthesis and characterization of side-
rophore analogs, the experimental methods using a
surface forces apparatus (SFA) and atomic force
microscope (AFM), and the qualitative adsorption
model (PDF)
AUTHOR INFORMATION
Corresponding Author
*butler@chem.ucsb.edu
■
ORCID
George D. Degen: 0000-0002-7386-506X
Roberto C. Andresen Eguiluz: 0000-0002-5209-4112
Eric Valois: 0000-0002-6824-2933
Kai Kristiansen: 0000-0002-7555-9437
Alison Butler: 0000-0002-3525-7864
Jacob N. Israelachvili: 0000-0001-8915-8741
Author Contributions
All authors have given approval to the final version of the
manuscript.
(17) Ahn, B. K.; Das, S.; Linstadt, R.; Kaufman, Y.; Martinez-
Rodriguez, N. R.; Mirshafian, R.; Kesselman, E.; Talmon, Y.; Lipshutz,
B. H.; Israelachvili, J. N.; Waite, J. H. High-Performance Mussel-
Inspired Adhesives of Reduced Complexity. Nat. Commun. 2015, 6,
8663.
Notes
The authors declare no competing financial interest.
⊥Deceased September 20, 2018.
ACKNOWLEDGMENTS
(18) Chirdon, W. M.; O’Brien, W. J.; Robertson, R. E. Adsorption of
Catechol and Comparative Solutes on Hydroxyapatite. J. Biomed.
Mater. Res. 2003, 66B (2), 532−538.
■
The authors thank J. Herbert Waite for his insightful
comments. G.D.D. was supported by the National Science
Foundation Graduate Research Fellowship Program under
Grant No. 1650114. A.B. is grateful for support from NSF
CHE-171076. The research reported here made use of shared
facilities of the UCSB MRSEC (NSF DMR 1720256), a
member of the Materials Research Facilities Network (www.
mrfn.org).
(19) Li, Y.; Qin, M.; Li, Y.; Cao, Y.; Wang, W. Single Molecule
Evidence for the Adaptive Binding of DOPA to Different Wet
Surfaces. Langmuir 2014, 30 (15), 4358−4366.
(20) Lee, B. P.; Messersmith, P. B.; Israelachvili, J. N.; Waite, J. H.
Mussel-Inspired Adhesives and Coatings. Annu. Rev. Mater. Res. 2011,
41 (1), 99−132.
(21) Fortelny, R. H.; Petter-Puchner, A. H.; Walder, N.; Mittermayr,
̈
R.; Ohlinger, W.; Heinze, A.; Redl, H. Cyanoacrylate Tissue Sealant
REFERENCES
Impairs Tissue Integration of Macroporous Mesh in Experimental
Hernia Repair. Surg. Endosc. Other Interv. Technol. 2007, 21 (10),
1781−1785.
■
(1) Danner, E. W.; Kan, Y.; Hammer, M. U.; Israelachvili, J. N.;
Waite, J. H. Adhesion of Mussel Foot Protein Mefp-5 to Mica: An
Underwater Superglue. Biochemistry 2012, 51 (33), 6511−6518.
(2) Lu, Q.; Danner, E.; Waite, J. H.; Israelachvili, J. N.; Zeng, H.;
Hwang, D. S. Adhesion of Mussel Foot Proteins to Different Substrate
Surfaces. J. R. Soc., Interface 2013, 10 (79), 20120759−20120759.
(3) Yu, J.; Kan, Y.; Rapp, M.; Danner, E.; Wei, W.; Das, S.; Miller, D.
R.; Chen, Y.; Waite, J. H.; Israelachvili, J. N. Adaptive Hydrophobic
and Hydrophilic Interactions of Mussel Foot Proteins with Organic
Thin Films. Proc. Natl. Acad. Sci. U. S. A. 2013, 110 (39), 15680−
15685.
(22) Kocherov, S.; Lev, G.; Chertin, B. Use of BioGlue Surgical
Adhesive in Hypospadias Repair. Curr. Urol. 2013, 7 (3), 132−135.
(23) Montanaro, L.; Arciola, C.; Cenni, E.; Ciapetti, G.; Savioli, F.;
Filippini, F.; Barsanti, L. Cytotoxicity, Blood Compatibility and
Antimicrobial Activity of Two Cyanoacrylate Glues for Surgical Use.
Biomaterials 2000, 22 (1), 59−66.
(24) Furst, W.; Banerjee, A. Release of Glutaraldehyde from an
̈
Albumin-Glutaraldehyde Tissue Adhesive Causes Significant in Vitro
and in Vivo Toxicity. Ann. Thorac. Surg. 2005, 79 (5), 1522−1528.
(25) Klimo, P.; Khalil, A.; Slotkin, J. R.; Smith, E. R.; Scott, R. M.;
Goumnerova, L. C. Wound Complications Associated with the Use of
Bovine Serum Albumin-Glutaraldehyde Surgical Adhesive in Pediatric
Patients. Neurosurgery 2007, 60, 305−309.
(4) Waite, J. H. Mussel Adhesion − Essential Footwork. J. Exp. Biol.
2017, 220 (4), 517−530.
(5) Waite, J. H.; Qin, X. Polyphosphoprotein from the Adhesive
Pads of Mytilus Edulis. Biochemistry 2001, 40 (9), 2887−2893.
H
DOI: 10.1021/jacs.9b04337
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
́
(26) Saiz-Poseu, J.; Mancebo-Aracil, J.; Nador, F.; Busque, F.; Ruiz-
Molina, D. The Chemistry behind Catechol-Based Adhesion. Angew.
Chem., Int. Ed. 2019, 58 (3), 696−714.
(27) Maier, G. P.; Rapp, M. V.; Waite, J. H.; Israelachvili, J. N.;
Butler, A. Adaptive Synergy between Catechol and Lysine Promotes
Wet Adhesion by Surface Salt Displacement. Science 2015, 349
(6248), 628−632.
(28) Rapp, M. V.; Maier, G. P.; Dobbs, H. A.; Higdon, N. J.; Waite,
J. H.; Butler, A.; Israelachvili, J. N. Defining the Catechol-Cation
Synergy for Enhanced Wet Adhesion to Mineral Surfaces. J. Am.
Chem. Soc. 2016, 138 (29), 9013−9016.
(29) Li, Y.; Wang, T.; Xia, L.; Wang, L.; Qin, M.; Li, Y.; Wang, W.;
Cao, Y. Single Molecule Study of the Synergistic Effects of Positive
Charges and Dopa for Wet Adhesion. J. Mater. Chem. B 2017, 5,
4416−4420.
(30) Sandy, M.; Butler, A. Chrysobactin Siderophores Produced by
Dickeya Chrysanthemi Ec16. J. Nat. Prod. 2011, 74 (5), 1207−1212.
(31) Israelachvili, J. N.; Min, Y.; Akbulut, M.; Alig, A.; Carver, G.;
Greene, W.; Kristiansen, K.; Meyer, E.; Pesika, N.; Rosenberg, K.;
Zeng, H. Recent Advances in the Surface Forces Apparatus (SFA)
Technique. Rep. Prog. Phys. 2010, 73 (3), 036601.
(32) Paria, S.; Khilar, K. C. A Review on Experimental Studies of
Surfactant Adsorption at the Hydrophilic Solid-Water Interface. Adv.
Colloid Interface Sci. 2004, 110 (3), 75−95.
(33) Halperin, A. Polymer Brushes that Resist Adsorption of Model
Proteins: Design Parameters. Langmuir 1999, 15 (7), 2525−2533.
(34) Ramachandran, G. N.; Ramakrishnan, C.; Sasisekharan, V.
Stereochemistry of Polypeptide Chain Configurations. J. Mol. Biol.
1963, 7 (1), 95−99.
(35) Gebbie, M. A.; Wei, W.; Schrader, A. M.; Cristiani, T. R.;
Dobbs, H. A.; Idso, M.; Chmelka, B. F.; Waite, J. H.; Israelachvili, J.
N. Tuning Underwater Adhesion with Cation−π Interactions. Nat.
Chem. 2017, 9, 473−479.
(36) Hong, S.; Wang, Y.; Park, S. Y.; Lee, H. Progressive Fuzzy
Cation-π Assembly of Biological Catecholamines. Sci. Adv. 2018, 4
(9), eaat7475.
(37) Wei, W.; Yu, J.; Gebbie, M. A.; Tan, Y.; Martinez Rodriguez, N.
R.; Israelachvili, J. N.; Waite, J. H. Bridging Adhesion of Mussel-
Inspired Peptides: Role of Charge, Chain Length, and Surface Type.
Langmuir 2015, 31 (3), 1105−1112.
(38) Bell, G. I. Models for the Specific Adhesion of Cells to Cells.
Science 1978, 200, 618−627.
I
DOI: 10.1021/jacs.9b04337
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX