Polyphenol-Mediated Assembly of Proteins for Engineering Functional Materials

Article abstract of DOI:10.1002/anie.202002089

Functional materials composed of proteins have attracted much interest owing to the inherent and diverse functionality of proteins. However, establishing general techniques for assembling proteins into nanomaterials is challenging owing to the complex physicochemical nature and potential denaturation of proteins. Here, a simple, versatile strategy is introduced to fabricate functional protein assemblies through the interfacial assembly of proteins and polyphenols (e.g., tannic acid) on various substrates (organic, inorganic, and biological). The dominant interactions (hydrogen-bonding, hydrophobic, and ionic) between the proteins and tannic acid were elucidated; most proteins undergo multiple noncovalent stabilizing interactions with polyphenols, which can be used to engineer responsiveness into the assemblies. The proteins retain their structure and function within the assemblies, thereby enabling their use in various applications (e.g., catalysis, fluorescence imaging, and cell targeting).

Full text of DOI:10.1002/anie.202002089

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Polyphenol-Mediated Assembly of Proteins for Engineering
Functional Materials
Authors: Yiyuan Han, Zhixing Lin, Jiajing Zhou, Gyeongwon Yun, Rui
Guo, Joseph J. Richardson, and Frank Caruso
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202002089
Angew. Chem. 10.1002/ange.202002089
Link to VoR: http://dx.doi.org/10.1002/anie.202002089
http://dx.doi.org/10.1002/ange.202002089

10.1002/anie.202002089
Angewandte Chemie International Edition
RESEARCH ARTICLE
Polyphenol-Mediated Assembly of Proteins for Engineering
Functional Materials
Yiyuan Han, Zhixing Lin, Jiajing Zhou, Gyeongwon Yun, Rui Guo, Joseph J. Richardson and Frank
Caruso*
[*]
Y. Han, Z. Lin, Dr. J. Zhou, Dr. G. Yun, R. Guo, Dr. J. J. Richardson, Prof. F. Caruso
ARC Centre of Excellence in Convergent Bio-Nano Science and Technology, Department of Chemical Engineering, The University of Melbourne, Parkville,
Victoria 3010, Australia
E-mail: fcaruso@unimelb.edu.au
Supporting information for this article is given via a link at the end of the document.
Abstract: Functional materials composed of proteins have attracted
much interest owing to the inherent and diverse functionality of
proteins. However, establishing facile and general techniques for
assembling proteins into nanomaterials is challenging owing to the
complex physicochemical nature and potential denaturation of
proteins. Here a simple, versatile strategy is introduced to fabricate
functional protein assemblies through the interfacial assembly of
proteins (>10 studied herein) and polyphenols (e.g., tannic acid) on
various substrates (organic, inorganic, and biological). The dominant
interactions (hydrogen bonding, hydrophobic and ionic interactions)
between the proteins and tannic acid are elucidated—most proteins
transport proteins, and fluorescent proteins) are assembled onto
diverse substrates of different composition, size, and shape
(Figure 1a) with different phenolics (e.g., TA, gallic acid, and
epigallocatechin gallate) (Figure S1, Supporting Information).
Polyphenols exhibit multiple interactions with proteins including
hydrogen bonding, and hydrophobic and ionic interactions (Figure
1b).[7] These interactions allow for robust and responsive protein–
polyphenol networks to be deposited, where, for instance, free-
standing protein–polyphenol capsules are obtained after coating
and dissolution of spherical sacrificial templates (Figure 1c).[8]
Owing to the purely organic nature, these proteins-based systems
can be exploited for various biomedical applications.[9] We
demonstrate that their bioactive functions can be applied to
enzyme catalysis, cell targeting, and tandem reactions (involving
multicomponent functional materials) (Figure 1d). Moreover,
these protein–polyphenol capsules enable the elucidation of the
dominant interactions between the different proteins and TA,
which is not possible by examining unstructured polyphenol–
protein complexes in solution, because individual proteins in
solution can be difficult to handle.[10] Applying this knowledge, pH-
responsive capsules with different shrinking–swelling behaviors,
which are dependent on the isoelectric point (pI) of the proteins,
can be engineered. This work introduces i) a toolbox for protein
assembly, ii) a model to study the interactions between proteins
and other macromolecules, and iii) a versatile platform for diverse
applications in biotechnology.[4,11]
undergo
multiple
noncovalent-stabilizing
interactions
with
polyphenols, which can be used to engineer responsiveness into the
assemblies. As demonstrated, the proteins retain their structure and
function within the assemblies, thereby enabling their use in various
applications (e.g., catalysis, fluorescent imaging, and cell targeting).
Introduction
Naturally occurring building blocks that can assemble into
functional materials have attracted scientific interest.[1] Proteins,
as biologically functional macromolecules involved in most
fundamental processes of living organisms, can mediate specific
interactions with diverse macromolecules for various applications
including catalysis, biosensing, diagnostics, and therapy.[2]
However, protein assembly is often achieved on a case-by-case
basis owing to the sensitivity of proteins to denaturation and the
chemical specificity of assembly routes.[3] In addition, simple and
versatile strategies for assembling proteins on substrates are
limited, as the noncovalent interactions between proteins and
substrates are often insufficiently strong to effectively tether them
to surfaces.[4] Therefore, developing versatile protein assembly
strategies are important to: i) deposit proteins onto various
surfaces of different sizes, aspect ratios, and material classes; ii)
assemble a range of proteins; iii) preserve the structure and
function of the proteins after assembly; and iv) allow for various
physicochemical properties to be introduced into the
assemblies.[5]
Results and Discussion
To demonstrate the versatility of the present method, protein–
polyphenol assemblies were deposited on substrates of different
composition, size, and shape (Figure 1e–h). Bovine serum
albumin (BSA), a globular protein commonly found in blood
plasma, was used as a model protein for protein–polyphenol
assembly. Fluorescein isothiocyanate (FITC)-labeled BSA (FITC-
BSA) was also used to visualize the assemblies. Simply, by
depositing TA on
a
planar substrate (e.g., glass,
polydimethylsiloxane (PDMS), or plastic on a centimeter scale)
and subsequently adding FITC-BSA in one pot without washing,
protein–polyphenol complexes (FITC-BSA–TA) were rapidly
deposited, as visualized by a change in the color of the substrate
after coating (Figure 1e). BSA–TA was also deposited on micro-
and nanoparticles. As observed from the transmission electron
microscopy (TEM) image in Figure 1f, the gold nanorods featured
a distinct shell after coating. From the confocal laser scanning
In the present study, we develop a general platform for
assembling functional protein materials via a simple, versatile
polyphenol-mediated coating strategy. Polyphenols (e.g., tannic
acid (TA)) are ubiquitous, universally adherent natural
compounds that can form conjugates with various
macromolecules, particularly proteins.[6] Therefore, various
proteins (more than 10 types, including enzymes, antibodies,
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202002089
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 1. Interfacial protein–polyphenol assembly on various substrates. a–d) Schematic of the interfacial assembly process on different substrates (a), the different
interactions between proteins and polyphenols (representative amino acids: Lys, lysine; Ser, serine; Phe, phenylalanine) (b), coating of sacrificial template and
subsequent capsule formation (c), and examples of biotechnological applications of protein–polyphenol capsules (d). e) Photographs of planar substrates before
(left) and after (right) coating with FITC-BSA–TA assemblies; PDMS, polydimethylsiloxane. f–h) Microscopy images of BSA–TA assemblies on various substrates:
gold nanorods (TEM image) (f); Escherichia coli (CLSM image) (g); and TRITC-dextran-loaded CaCO3 particles (h, red), FITC-BSA–TA coating shown in green. i–
l) Microscopy images of BSA–TA capsules formed using sacrificial CaCO3 templates: DIC (i), SEM (j), TEM (k), and AFM (l) images.
microscopy (CLSM) image in Figure 1g,h, FITC-BSA–TA coatings
were observed after deposition on live Escherichia coli (E. coli)
bacteria or tetramethylrhodamine (TRITC)-dextran-loaded
calcium carbonate (CaCO3) particles. Other organic and inorganic
particles were coated with FITC-BSA–TA including polystyrene
and silica (Figure S2). Collectively, these results suggest that the
present protein assembly strategy is applicable to diverse
substrates across multiple length scales (20 nm to 1 cm).
To study the physicochemical properties of the assemblies,
CaCO3 particles (3 µm in diameter) were used as sacrificial
templates for preparing free-standing protein–polyphenol
capsules.[12] The surface of the uncoated CaCO3 particles was
rough and porous. In contrast, after BSA–TA assembly, the
particle surfaces were relatively smooth and appeared less
porous, demonstrating that the interfacial assembly process was
successful (Figure S3). Removal of the CaCO3 template by
ethylenediaminetetraacetic acid (EDTA) generated spherical and
monodisperse BSA–TA microcapsules, as observed from the
differential interference contrast (DIC) microscopy image in
Figure 1i, highlighting the robust nature of the assemblies.
Compared with the BSA–TA-coated CaCO3 core–shell particles
(before template removal), the BSA–TA capsules demonstrated
lower light scattering, indicating the successful removal of the
solid CaCO3 core (Figure S4a). Both the FITC-BSA–TA-coated
CaCO3 core–shell particles and FITC-BSA–TA capsules showed
comparable fluorescence, suggesting that the template removal
process did not cause significant loss of BSA (Figure S4b).
Scanning electron microscopy (SEM), TEM, and atomic force
microscopy (AFM) showed collapsed (air-dried) capsules, and
excess polyphenol–protein aggregates, unbound or attached to
the BSA–TA capsules, were not detected (Figure 1j–l). The shell
thickness of the BSA–TA capsules was ~40 nm, as measured by
AFM (Figure S5), and circular dichroism demonstrated that the
conformation of BSA did not change after assembly[13] (Figure S6),
indicating no loss of protein functionality.
A
library of functional protein–polyphenol assemblies was
subsequently established using TA and a variety of proteins with
different pI, molecular weight (Mw), and aliphatic index (Figure 2a
and Table S1),[14] which allowed for an in-depth characterization
of the influence of the physicochemical properties of the proteins
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202002089
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 2. Different protein–polyphenol capsules. a) Physicochemical properties of different proteins. (x-axis: isoelectric point; y-axis: molecular weight; and z-axis:
aliphatic index.) b) DIC microscopy images of different protein–TA capsules and a fluorescence microscopy image of GFP–TA capsules.
on the properties of the protein–polyphenol assemblies. The
functional proteins studied include lysozyme (LYZ), glucose
oxidase (GOx), pepsin, horseradish peroxidase (HRP),
prepared from proteins other than BSA were between 50 and 90
nm (Figure S11). Importantly, all the proteins studied herein
successfully assembled at interfaces and formed conformal and
hemoglobin (Hgb), cytochrome C (CYC), immunoglobulin G (IgG), robust films, likely due to the diverse interactions with TA.[17]
fibrinogen (FGN), insulin, and green fluorescent protein (GFP)
(Figure 2b). LYZ, GOx, HRP, and pepsin are enzymes. CYC and
Hgb are essential transport proteins. IgG is a class of antibodies
and FGN is responsible for blood clotting (protection proteins).
Insulin is a peptide hormone and GFP is a fluorescent protein,
often used as a gene expression marker (regulatory proteins).
Hemeproteins (i.e., CYC and HRP) were first chosen to
characterize the possible interactions between TA and the metal
centers of the heme group, as polyphenols are known to chelate
metals.[15] Interestingly, the mixture of hemeproteins with TA did
not show the characteristic ligand-to-metal charge transfer band
for polyphenol/FeIII complexes as observed by UV–vis
spectroscopy (Figure S7), where the CYC–TA capsules featured
a heme group peak at ~410 nm similarly to free CYC (Figure S8a),
suggesting that TA did not competitively chelate the bound iron.[16]
The Fourier transform infrared spectrum of the CYC–TA capsules
demonstrated peak overlaps with the spectra of free TA and free
CYC (Figure S8b). Energy-dispersive X-ray mapping results
demonstrated that the CYC-TA capsules comprised four
elements making up CYC and TA, namely C, N, O, and Fe (Figure
S9). The loading efficiency of HRP in the HRP–TA capsules was
~80% (Figure S10 and Table S2), as determined from the iron
content in HRP measured by inductively coupled plasma-optical
emission spectrometry. The thicknesses of protein–TA capsules
The structural units of proteins are amino acids, which possess
different electrically charged, polar uncharged, and hydrophobic
side chains. These functional side chains of amino acids can
potentially interact with polyphenols via hydrogen bonding,
hydrophobic interactions, and/or ionic interactions (Figure 3a). To
investigate how these three interactions govern the stability of
different protein–TA assemblies, urea, Tween 20, or sodium
chloride (NaCl) was added to each of the five model protein–
polyphenol capsules (LYZ–TA, IgG–TA, Hgb–TA, GOx–TA, and
CYC-TA) to study disassembly over time.[6b] These proteins were
chosen as model molecules, as they have different pI, Mw, and
aliphatic index. Urea, which can participate in the formation of
strong hydrogen bonds (and therefore break weaker hydrogen
bonds),[18] effectively disassembled the IgG–TA and GOx–TA
capsules (within 2 h), indicating that these two types of protein–
polyphenol assemblies were mainly stabilized by hydrogen bonds
(Figure S12a). In contrast, 90, 89, and 69% of the LYZ–TA, CYC–
TA, and Hgb–TA capsules, respectively, remained after
incubation with 100 mM urea for 42 h, suggesting that different or
additional intermolecular forces stabilize these systems. Tween
20 is
a nonionic surfactant that can form hydrophobic
interactions.[19] CYC–TA, LYZ–TA, and IgG–TA capsules were
completely disassembled by 100 mM Tween 20 within 5 min of
incubation as a result of the hydrophobic interactions being
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202002089
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 3. Dominant interactions between proteins and TA in protein–polyphenol assemblies. a) Schematic diagram of the possible interactions between the
functional groups of polyphenols and proteins. Ser: serine; Asn: asparagine; Phe: phenylalanine; Val: valine; Asp: aspartic acid; Lys: lysine. b) Percentage of
protein–polyphenol capsules (LYZ, IgG, Hgb, GOx, and CYC) degraded after 1 h of incubation with 100 mM of urea, Tween 20, or NaCl, highlighting the dominant
interactions between the different proteins and TA. c–e) Relationship between the dominant interactions and physiochemical properties of the proteins, as analyzed
from the capsule degradability of CYC–TA, LYZ–TA, Hgb–TA, GOx–TA, and IgG–TA: hydrogen bonding vs molecular weight, incubation for 1 h with 100 mM urea
(c); hydrophobic interactions vs aliphatic index, incubation for 2 min with 100 mM Tween 20 (d); and ionic interactions vs isoelectric point, incubation for 19 h with
100 mM NaCl (e). Data are shown as the mean ± standard deviation (SD) (N = 3).
disturbed. In contrast, 50 and 75% of the GOx–TA and Hgb–TA
capsules remained intact in Tween 20 solution over 40 h of
incubation (Figure S12b). Incubation of the capsules in 100 mM
NaCl resulted in the disassembly of all five protein–polyphenol
capsules to varying extent over 90 h owing to ionic shielding—
among the capsules studied, GOx–TA capsules showed the
highest disassembly rate or the lowest percentage of capsules
remaining (37%) after 90 h (Figure S12c). Generally, the type of
protein determines which interactions are present in the
assemblies (Figure 3b and Figure S13). For instance,
hydrophobic interactions are the dominant stabilizers for LYZ–TA
and CYC–TA, whereas they are the least dominant interactions
between Hgb and TA. In IgG–TA, the most dominant interactions
are hydrogen bonding and hydrophobic interactions. In GOx–TA,
the interactions are governed by all three types of interactions.
To gain a deeper insight into the nature of the protein–polyphenol
interactions in the assemblies, we further analyzed the
relationship between the dominant interactions and the
physiochemical properties of the proteins such as the size (Mw),
aliphatic index, and pI. The three key findings are as follows. i)
The strength of hydrogen bonding between polyphenols and
proteins is likely based on the Mw or size of the proteins as the
peptide backbone and the majority of the exposed protein surface
(polar amino acids) are hydrophilic. For example, the molecular
size of the proteins increases in the order of CYC < LYZ < Hgb <
GOx < IgG, and the disassembly rate of the respective protein–
TA capsules also increases in the same order as hydrogen bonds
between the proteins and hydroxyl groups of the catechol/gallol
moieties in polyphenols are disrupted (Figure 3c). ii) Proteins with
high aliphatic indexes containing more aliphatic and hydrophobic
side chains (such as alanine, valine, isoleucine, and leucine) can
form stronger hydrophobic interactions with the aromatic groups
of polyphenols compared with proteins with low aliphatic indexes
(low hydrophobicity). For example, the interactions of polyphenols
with high aliphatic index proteins, such as Hgb and GOx, are more
difficult to disrupt using surfactants when compared with those of
polyphenols with low aliphatic index proteins such as CYC and
IgG (Figure 3d). iii) The strength of ionic interactions between
proteins and polyphenols depends on the charge of the protein.
For instance, the pI of the proteins increases in the order of GOx
< Hgb < IgG < CYC < LYZ, which correlates to the relatively more
positively charged side chains in the protein and thereby
enhanced ionic interactions with the deprotonated hydroxyls of
the catechol/gallol moieties of polyphenols (Figure 3e). These
results highlight the complex nature in which polyphenols interact
with different proteins at the molecular level, suggesting that TA
is essential for helping to construct amino acid-based materials.
The protein–polyphenol assemblies exhibited distinct and tunable
pH-responsive properties, owing to the various pKa values of the
hydroxyl groups of TA and the pIs of the different proteins. For
example, BSA–TA capsules swelled from 2.1 to 4.2 µm when the
environmental pH increased from 4.0 to 7.4 and shrank to their
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202002089
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 4. pH-Responsiveness of protein–polyphenol assemblies. a) Fluorescence microscopy images of the BSA–TA capsules at pH 4.0 and pH 7.4. b) Changes
in the diameters of the pepsin–TA, BSA–TA, and CYC–TA capsules as a function of pH wash cycle. The diameters of 20 capsules were measured using DIC
microscopy, and data are shown as the mean ± SD. c) Zeta potential values of the pepsin–TA, BSA–TA, and CYC–TA capsules at different pH. Data are shown as
the mean ± SD. d) AFM images and corresponding height measurements of the BSA–TA capsules at pHs 4.0 and 7.4. e) Percentage of permeable capsules plotted
against FITC-dextran of different molecular weight at pHs 8.5 and 6.0. The permeability of 100 capsules was measured using CLSM.
original size when the pH was lowered to 4.0 (Figure 4a). This
swelling/shrinking process could be repeated reproducibly over
multiple cycles (Figure 4b). TA and BSA both possess a net
negative charge when the pH rises above the pI of BSA (4.6),
which leads to electrostatic repulsion, and thus swelling, and
eventual dissociation. The other protein–TA capsules also
showed reversible pH-responsive swelling and shrinking
behaviors when the pI values of the proteins (pepsin: 2.9; CYC:
9.6) were within the swelling/shrinking values (pepsin: pH 2.8–6.0
and CYC: pH 4.0–10.0), which supports our proposed mechanism
(Figure 4b and Figure S14). In addition, the zeta potential of the
pepsin–TA, BSA–TA, and CYC–TA capsules changed under
different pH, further suggesting the importance of the protonation
state of the proteins (Figure 4c). The thickness of the BSA–TA
capsules decreased from ~50 nm at pH 4.0 to ~30 nm at pH 7.4,
which demonstrates that the swelling behavior stretches the films
to accommodate for the increase in volume of the capsule (Figure
4d and Figure S15). As such pH-responsiveness can be desirable
for drug delivery applications, the molecular permeability of the
CYC–TA capsules was assessed using FITC-dextran of different
molecular weights as the model cargo. As observed in Figure 4e,
the permeability of the CYC–TA capsules was pH dependent,
suggesting that these assemblies could be used to load and
release drugs under specific pH environments. For instance,
~25% of the CYC–TA capsules were permeable to 250 kDa FITC-
dextran at pH 6.0, whereas nearly 100% of the capsules were
permeable at pH 8.5 (Figure 4f and Figure S16), which
corresponds to higher permeability when the capsules are
swollen. When considering physiological pH changes, such as the
stomach (pH 1.0–3.0), blood (pH 7.4), or duodenum (pH 4.8–
8.2),[20] we envision that our polyphenol–protein assemblies could
serve as smart delivery systems for various biomedical
applications.
The potential biotechnological applications of the protein–
polyphenol assemblies were examined. The functionality of the
protein–polyphenol assemblies for catalysis was first investigated.
HRP is a common enzyme that catalyzes the oxidation of various
organic substrates in the presence of H2O2.[21] Specifically, the
conversion of amplex red (AR) into resorufin (RS), which has a
strong absorbance at 560 nm,[22] in the presence of the HRP–TA
assemblies (core–shell particles and capsules) was examined
(Figure 5a). The HRP–TA capsules showed comparable catalytic
activity to free HRP, suggesting that the enzymatic activity was
not affected after assembly (Figure 5b and Figure S17).[23]
Moreover, these capsules showed higher catalytic activity (~25%)
than the HRP–TA core–shell particles, which is attributed to the
likely improved mass transfer in the hollow (capsule) system. In
addition, the enzymes in the HRP–TA capsules displayed
excellent recyclability, where more than 68% of the initial HRP
activity was retained after five cycles of catalysis and collection
(Figure 5c).
Multicomponent systems composed of GOx and HRP
(GOx&HRP–TA) were engineered and investigated to conduct a
cascade reaction.[24] Specifically, GOx can catalyze the oxidation
and hydrolysis of β-D-glucose into gluconic acid and H2O2, and
HRP can catalyze AR oxidation to RS in the presence of the
generated H2O2 (Figure 5d). The GOx&HRP–TA capsules
showed successful colorimetric reaction with glucose in contrast
to the (single-component) GOx–TA and HRP–TA capsules
(Figure 5e). Therefore, the current approach adopted to prepare
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202002089
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 5. Protein–polyphenol assemblies for diverse applications. a–c) Enzymatic activity of the HRP–TA capsules. Schematic of the HPR-catalyzed oxidation of
AR to RS over the HRP–TA capsules (a). Time-dependent absorbance changes as a result of the oxidation of AR by different catalytic systems: free HRP, HRP–
TA capsules, and HRP–TA core–shell particles (b). Changes in the relative catalytic activity of the HRP–TA capsules as a function of cycle number (c). d–e) Cascade
reaction using multicomponent GOx&HRP–TA capsules: schematic of glucose-triggered AR oxidation to produce RS in the presence of GOx and HRP (d) and UV–
vis spectra of the cascade reaction using different single- and multicomponent capsule systems (e). Inset: color changes using the different systems. f–j) Cell
targeting of antibody–TA capsules: schematic of the targeting specificity of antiCD44–TA capsules toward CD44 overexpressing (CD44+) cells (MDA-MB-231 cells)
and CD44 minimal expressing (CD44−) cells (BT-474 cells) (f); flow cytometry analysis of the antiCD44–TA and IgG–TA capsules binding to MDA-MB-231 and BT-
474 cells after incubation for 1 h at 4 °C (g); percentage of MDA-MB-231 cells associated with antiCD44–TA and IgG–TA capsules after 1 or 3 h incubation at 37 °C
(h); and CLSM images of antiCD44–TA (i) and IgG–TA capsules (j) incubated with MDA-MB-231 cells for 3 h at 37 °C. The capsule-to-cell ratio was 50:1 in all cell
experiments. Data are shown as the mean ± SD (N = 3).
the multicomponent and multifunctional protein–polyphenol
assemblies presents a simple and versatile strategy to design
intricate biomimetic systems.[25]
significant—96% of the MDA-MB-231 cells associated with the
antiCD44–TA capsules, whereas only 4% of the MDA-MB-231
cells associated with the IgG–TA capsules (Figure 5h and Figure
S18), as supported visually by CLSM (Figure 5i,j). These above
results reveal that the formation of antibody–TA assemblies does
not negatively affect the inherent targeting ability of the antibodies.
Additionally, the BSA–TA and LYZ–TA capsules displayed
excellent stability in serum (Figure S19), negligible cytotoxicity
and were internalized in cells (~100% capsules internalized by
cells after 14 h, Figure S20), highlighting that this assembly
strategy shows potential for intracellular protein delivery
applications.[29] Moreover, protein–TA capsules did not have
significant interactions with live E. coli, indicating that these
coatings have the potential to endow materials with low-fouling
properties (Figure S21).[30] To improve reproducibility, reporting,
and re-analysis, this study conforms to the Minimum Information
Reporting in Bio–Nano Experimental Literature (MIRIBEL)
standard,[31] and a companion checklist is provided in the
Supporting Information.
Proteins also plays a significant role in recognition, such as cell
targeting by antibodies.[26] Antibody–TA capsules were thus
synthesized to study their cell binding and targeting ability. CD44,
a cell-surface glycoprotein involved in cell–cell interactions, is
overexpressed on many types of cancer cell membranes,[27] and
anti-CD44 antibodies can specifically target CD44 overexpressing
(CD44+) cells, but do not readily interact with CD44 minimal
expressing (CD44−) cells (Figure 5f). The cell binding and
targeting ability of antiCD44–TA capsules were investigated using
two human breast cancer cell lines, MDA-MB-231 (CD44+) cells
and BT-474 (CD44−) cells; IgG–TA (anti-Human IgG) capsules
were used as the negative control. Cell binding was performed at
4 °C after 1 h incubation to prevent internalization.[28] The
antiCD44–TA capsules showed high binding specificity toward
CD44+ cells over CD44− cells, whereas the IgG–TA capsules
demonstrated nonspecific cell binding toward both cells (Figure
5g and Figure S18). In addition, cell association experiments with
CD44+ cells were performed at 37 °C after 1 and 3 h incubation
to compare the antiCD44–TA capsules with control IgG–TA
capsules. The flow cytometry results showed that after 1 h
incubation, 58% of the MDA-MB-231 (CD44+) cells associated
with the antiCD44–TA capsules, whereas only 2% of the MDA-
MB-231 cells associated with the IgG–TA capsules (Figure 5h).
After 3 h, the difference in cell association became more
Conclusion
We have introduced interfacial polyphenol-mediated protein
assembly as a versatile and strategy to create a library of
functional
protein-based
materials.
Protein–polyphenol
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202002089
Angewandte Chemie International Edition
RESEARCH ARTICLE
[2]
a) J. A. Shadish, G. M. Benuska, C. A. DeForest, Nat. Mater. 2019, 18,
assemblies can be deposited on various substrates of different
composition, size, and shape, affording new possibilities for
surface modification. This technique is applicable to a wide range
of proteins of diverse physicochemical properties and biological
functions, thus enabling bioactive surface coating and functional
capsule formation. Moreover, the protein–polyphenol capsules
can be used to elucidate the dominant interaction(s) between
different proteins and polyphenols; the results reveal that the
molecular weight (size), hydrophobicity (aliphatic index), and
isoelectric point (electrical charge) of the protein determine the
strength of hydrogen bonding, hydrophobic interactions, and ionic
interactions, respectively. These various noncovalent-stabilizing
interactions between proteins and polyphenols offer possibilities
to engineer assemblies with desirable responsiveness to multiple
stimuli such as changes in pH, as exemplified. Finally, we have
demonstrated various applications for the protein–polyphenol
assemblies, as a range of functionalities can be engineered into
the assemblies through judicious choice of the constituent
proteins. This platform is expected to advance both fundamental
and applied research owing to its simplicity and versatility.
1005-1014; b) C. D. Spicer, C. Jumeaux, B. Gupta, M. M. Stevens, Chem. Soc.
Rev. 2018, 47, 3574-3620.
[3]
a) X. Huang, M. Li, D. C. Green, D. S. Williams, A. J. Patil, S. Mann, Nat.
Commun. 2013, 4, 2239; b) Q. Luo, C. Hou, Y. Bai, R. Wang, J. Liu, Chem. Rev.
2016, 116, 13571-13632; c) T. P. Knowles, T. W. Oppenheim, A. K. Buell, D. Y.
Chirgadze, M. E. Welland, Nat. Nanotechnol. 2010, 5, 204-207.
[4]
R. Liu, J. Zhao, Q. Han, X. Hu, D. Wang, X. Zhang, P. Yang, Adv. Mater.
2018, 30, 1802851.
[5]
[6]
E. V. Skorb, D. V. Andreeva, Adv. Funct. Mater. 2013, 23, 4483-4506.
a) M. A. Rahim, S. L. Kristufek, S. Pan, J. J. Richardson, F. Caruso,
Angew. Chem. 2019, 131, 1920-1945; Angew. Chem. Int. Ed. 2019, 58, 1904-
1927; b) J. E. Chung, S. Tan, S. J. Gao, N. Yongvongsoontorn, S. H. Kim, J. H.
Lee, H. S. Choi, H. Yano, L. Zhuo, M. Kurisawa, J. Y. Ying, Nat. Nanotechnol.
2014, 9, 907-912.
[7]
S. Quideau, D. Deffieux, C. Douat-Casassus, L. Pouysegu, Angew.
Chem. 2011, 123, 610-646; Angew. Chem. Int. Ed. 2011, 50, 586-621.
[8]
[9]
F. Caruso, R. A. Caruso, H. Möhwald, Science 1998, 282, 1111-1114.
a) U. Shimanovich, G. J. Bernardes, T. P. Knowles, A. Cavaco-Paulo,
Chem. Soc. Rev. 2014, 43, 1361-1371; b) M. Yan, J. Du, Z. Gu, M. Liang, Y.
Hu, W. Zhang, S. Priceman, L. Wu, Z. H. Zhou, Z. Liu, Nat. Nanotechnol. 2010,
5, 48-53.
[10] a) P. Bandyopadhyay, A. K. Ghosh, C. Ghosh, Food Funct. 2012, 3, 592-
605; b) A. Papadopoulou, R. J. Green, R. A. Frazier, J. Agric. Food Chem. 2005,
53, 158-163.
[11] a) M. Shin, H.-A. Lee, M. Lee, Y. Shin, J.-J. Song, S.-W. Kang, D.-H.
Nam, E. J. Jeon, M. Cho, M. Do, S. Park, M. S. Lee, J.-H. Jang, S.-W. Cho, K.-
S. Kim, H. Lee, Nat. Biomed. Eng. 2018, 2, 304-317; b) C. Liu, T. Wan, H. Wang,
S. Zhang, Y. Ping, Y. Cheng, Sci. Adv. 2019, 5, eaaw8922.
[12] J. J. Richardson, J. W. Maina, H. Ejima, M. Hu, J. Guo, M. Y. Choy, S. T.
Gunawan, L. Lybaert, C. E. Hagemeyer, B. G. De Geest, Adv. Sci. 2015, 2,
1400007.
Acknowledgements
[13] a) N. J. Greenfield, G. D. Fasman, Biochemistry 1969, 8, 4108-4116; b)
N. J. Greenfield, Nat. Protoc. 2006, 1, 2876-2890.
[14] P. G. Righetti, G. Tudor, K. Ek, J. Chromatogr. A 1981, 220, 115-194.
[15] J. Guo, Y. Ping, H. Ejima, K. Alt, M. Meissner, J. J. Richardson, Y. Yan,
K. Peter, D. von Elverfeldt, C. E. Hagemeyer, F. Caruso, Angew. Chem. 2014,
126, 5652-5657; Angew. Chem. Int. Ed. 2014, 53, 5546-5551.
[16] M. A. Rahim, K. Kempe, M. Müllner, H. Ejima, Y. Ju, M. P. van Koeverden,
T. Suma, J. A. Braunger, M. G. Leeming, B. F. Abrahams, F. Caruso, Chem.
Mater. 2015, 27, 5825–5832.
[17] K. J. Siebert, N. V. Troukhanova, P. Y. Lynn, J. Agric. Food Chem. 1996,
44, 80-85.
[18] L. B. Sagle, Y. Zhang, V. A. Litosh, X. Chen, Y. Cho, P. S. Cremer, J.
Am. Chem. Soc. 2009, 131, 9304-9310.
[19] N. B. Bam, J. L. Cleland, T. W. Randolph, Biotechnol. Prog. 1996, 12,
801-809.
[20] D. Schmaljohann, Adv. Drug Delivery Rev. 2006, 58, 1655-1670.
[21] N. C. Veitch, Phytochemistry 2004, 65, 249-259.
[22] M. Zhou, Z. Diwu, N. Panchuk-Voloshina, R. P. Haugland, Anal. Biochem.
1997, 253, 162-168.
[23] a) H. Tan, S. Guo, N. D. Dinh, R. Luo, L. Jin, C. H. Chen, Nat. Commun.
2017, 8, 663; b) K. Li, G. Xiao, J. J. Richardson, B. L. Tardy, H. Ejima, W. Huang,
J. Guo, X. Liao, B. Shi, Adv. Sci. 2019, 6, 1801688.
This research was conducted and funded by the Australian
Research Council (ARC) Centre of Excellence in Convergent Bio-
Nano Science and Technology (project number CE 140100036)
and an ARC Discovery Project (DP170103331). F. C.
acknowledges the award of a National Health and Medical
Research Council Senior Principal Research Fellowship
(GNT1135806). This work was performed in part at the Materials
Characterization and Fabrication Platform and the Bio21
Advanced Microscopy Facility at The University of Melbourne and
the Victorian Node of the Australian National Fabrication Facility.
We thank Dr. Shuaijun Pan, Yutian Ma, Jingqu Chen, Yingjie Hu,
and Jiaying Song for assistance with experiments, Wenjie Zhang
for providing the green fluorescent proteins, Dr. Christina Cortez-
Jugo, Dr. Yi Ju, and Dr. Md. Arifur Rahim for helpful discussions,
and Dr. Yee-Foong Mok (Macromolecular Interactions Facility
Platform, Bio21 Institute, The University of Melbourne) for
assistance with the circular dichroism analysis.
[24] I. Wheeldon, S. D. Minteer, S. Banta, S. C. Barton, P. Atanassov, M.
Sigman, Nat. Chem. 2016, 8, 299-309.
[25] a) N. P. King, J. B. Bale, W. Sheffler, D. E. McNamara, S. Gonen, T.
Gonen, T. O. Yeates, D. Baker, Nature 2014, 510, 103-108; b) B. O. Okesola,
A. Mata, Chem. Soc. Rev. 2018, 47, 3721-3736.
[26] a) A. M. Scott, J. D. Wolchok, L. J. Old, Nat. Rev. Cancer 2012, 12, 278-
287; b) L. M. Weiner, R. Surana, S. Wang, Nat. Rev. Immunol. 2010, 10, 317-
327.
[27] H. Ponta, L. Sherman, P. A. Herrlich, Nat. Rev. Mol. Cell Biol. 2003, 4,
33-45.
Competing interests
The authors declare no competing interests.
[28] A. Lesniak, A. Salvati, M. J. Santos-Martinez, M. W. Radomski, K. A.
Dawson, C. Åberg, J. Am. Chem. Soc. 2013, 135, 1438-1444.
[29] a) M. P. Stewart, A. Sharei, X. Ding, G. Sahay, R. Langer, K. F. Jensen,
Nature 2016, 538, 183-192; b) F. Scaletti, J. Hardie, Y. W. Lee, D. C. Luther, M.
Ray, V. M. Rotello, Chem. Soc. Rev. 2018, 47, 3421-3432.
[30] W. Luo, G. Xiao, F. Tian, J. J. Richardson, Y. Wang, J. Zhou, J. Guo, X.
Liao, B. Shi, Energy Environ. Sci. 2019, 12, 607–614.
Keywords: biocatalysis • cancer target • particle surface
engineering • polyphenols • proteins
[31] M. Faria, M. Björnmalm, K. J. Thurecht, S. J. Kent, R. G. Parton, M.
Kavallaris, A. P. Johnston, J. J. Gooding, S. R. Corrie, B. J. Boyd, P.
Thordarson, A. K. Whittaker, M. M. Stevens, C. A. Prestidge, C. J. H. Porter,
W. J. Parak, T. P. Davis, E. J. Crampin, F. Caruso, Nat. Nanotechnol. 2018,
13, 777-785.
[1]
a) H. Ejima, J. J. Richardson, K. Liang, J. P. Best, M. P. van Koeverden,
G. K. Such, J. Cui, F. Caruso, Science 2013, 341, 154-157; b) J. Guo, B. L.
Tardy, A. J. Christofferson, Y. Dai, J. J. Richardson, W. Zhu, M. Hu, Y. Ju, J.
Cui, R. R. Dagastine, I. Yarovsky, F. Caruso, Nat. Nanotechnol. 2016, 11, 1105-
1111.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202002089
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
A simple and versatile strategy for assembling functional nanomaterials is established through the interfacial assembly of proteins and
polyphenols on various substrates. Protein–polyphenol capsules are used to elucidate the dominant interaction(s) between different
proteins and polyphenols. The assembled proteins retain their structure and function, thereby enabling their use in various applications
(e.g. biocatalysis or cell targeting).
8
This article is protected by copyright. All rights reserved.