Inhibition of S. aureus Infection of Human Umbilical Vein Endothelial Cells (HUVECs) by Trehalose- and Glucose-Functionalized Gold Nanoparticles

Article abstract of DOI:10.1002/anie.202106544

Microbial adhesion to host cells represents the initial step in the infection process. Several methods have been explored to inhibit microbial adhesion including the use of glycopolymers based on mannose, galactose, sialic acid and glucose. These sugar re

Full text of DOI:10.1002/anie.202106544

Angewandte
Communications
Chemie
How to cite:
International Edition: doi.org/10.1002/anie.202106544
German Edition: doi.org/10.1002/ange.202106544
Glycopolymers
Inhibition of S. aureus Infection of Human Umbilical Vein Endothelial
Cells (HUVECs) by Trehalose- and Glucose-Functionalized Gold
Nanoparticles
Yimeng Li, Nicholas Ariotti, Behnaz Aghaei, Elvis Pandzic, Sylvia Ganda, Mark Willcox,
Manuel Sanchez-Felix, and Martina Stenzel*
Abstract: Microbial adhesion to host cells represents the initial
step in the infection process. Several methods have been
explored to inhibit microbial adhesion including the use of
glycopolymers based on mannose, galactose, sialic acid and
glucose. These sugar receptors are, however, abundant in the
body, and are not unique to bacteria. Trehalose, in contrast, is
a unique disaccharide that is widely expressed by microbes.
This carbohydrate has not yet been explored as an anti-
adhesive agent. Herein, gold nanoparticles (AuNPs) coated
with trehalose-based polymers were prepared and compared to
glucose-functionalized AuNPs and examined for their ability
to prevent binding to endothelial cells. Acting as anti-adhesive
agents, trehalose-functionalized NPs decreased the binding of
S. aureus to HUVECs, while outperforming the control NPs.
Microscopy revealed that trehalose-coated NPs bound strongly
to S. aureus compared to the controls. In conclusion, nano-
particles based on trehalose could be a non-toxic alternative to
inhibit S. aureus infection.
resistance. For pathogenic invasion to occur, pathogens must
first adhere to host cells. Anti-adhesion treatment inhibits the
invasion of pathogens by preventing the adherence of
pathogens with host cells. Unlike conventional antibiotic
treatment, anti-adhesion treatment is less susceptible to
antibacterial resistance.[1] Adhesion can often occur via
lectin-glycan interactions. Lectins are a type of protein
located on the surface of cells and functions by binding to
specific glycans.[2] Glycomimetics have been employed as
anti-adhesive that block interactions with lectins.[3] In partic-
ular, well-defined macrostructures of the glycomimetics are
attractive. Owing to the rapid development of nanotechnol-
ogy, well-defined glycopolymers can be synthesized with
proper functional groups, sizes and morphologies.[4–6] As
a result, glycopolymers have attracted interest as potent
glycomimetic candidates that could serve as anti-adhe-
sive.[3,7,8]
A number of factors have been investigated in the
synthesis of anti-adhesives including scaffolds, multivalency
and size.[5,9,10] Notably, the glycan specificity to lectins plays
a crucial role in the binding behavior. For instance, the
hemagglutinin of Influenza virus and Cholera toxins are well-
known sialic acid-binding lectins.[11] FimH of certain strain of
pathogenic E. coli is a mannose-specific lectin.[12] Human
lectin Dendritic Cell-Specific Intercellular adhesion mole-
cule-3-Grabbing Non-integrin (DC-SIGN), the target of HIV,
is capable of specifically binding mannose and fucose.[13,14]
Accordingly, numerous glycopolymers employing different
sugars, such as mannose, galactose, sialic acid and glucose,
have been investigated depending on the targeted patho-
gens.[7,10,15–20] However, diverse sugar receptors are also
abundant in human cells, such as the expression of mannose
receptors in liver and macrophage cells as well as sialic acid
receptor on endothelial cells and leukocytes.[11,21] These
sugars on human cells play significant biological roles as
they participate in multiple biological events other than
adhesion. Therefore, promising inhibitors for anti-adhesion
treatments are expected to possess the ability to bind
specifically to bacteria but does not interfere with host
systems.
A
nti-adhesion treatment has emerged as a promising anti-
infection strategy due to the global emergency of antibiotics
[*] Y. Li, Dr. S. Ganda, Prof. M. Stenzel
Centre for Advanced Macromolecular Design, School of Chemistry,
University of New South Wales
Sydney, NSW 2052 (Australia)
E-mail: m.stenzel@unsw.edu.au
Dr. N. Ariotti
Electron Microscope Unit, Mark Wainwright Analytical Centre,
University of New South Wales, Sydney, NSW 2052 (Australia)
Dr. B. Aghaei
Inventia Life Science Pty Ltd
Sydney, NSW 2015 (Australia)
and
School of Biotechnology and Biomolecular Sciences, University of
New South Wales, Sydney, NSW 2052 (Australia)
Dr. E. Pandzic
Katharina Gaus Light Microscopy Facility, Mark Wainwright
Analytical Centre, University of New South Wales, Sydney, NSW 2052
(Australia)
Prof. M. Willcox
School of Optometry and Vision Science, University of New South
Wales, Sydney, NSW 2052 (Australia)
Trehalose is a non-reducing disaccharide composed from
two D-glucose molecules. As a result of its distinctive
structure, trehalose displays impressive physical and chemical
properties; including having high hydrophilicity and chemical
stability.[22] Trehalose is able to improve cell tolerance against
extreme conditions, such as temperature and pH change.
Inspired by its bio-functional role, trehalose-based polymers
Dr. M. Sanchez-Felix
Novartis Institutes for BioMedical Research
Cambridge, MA 02139 (USA)
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.202106544.
Angew. Chem. Int. Ed. 2021, 60, 1 – 8
ꢀ 2021 Wiley-VCH GmbH
1
These are not the final page numbers!

Angewandte
Communications
Chemie
have been employed in the effort to protect, stabilize and
even deliver proteins and nucleic acid.[23–25] Notably, trehalose
occurs in prokaryotes, lower eukaryotes, plants and inverte-
brates.[26] The biosynthesis of trehalose is absent in mammals.
However, the specific hydrolytic enzyme trehalase is
expressed in the microvillus of the intestinal mucosa and
renal brush border membrane in humans.[27] Trehalose is
a fundamental carbohydrate in bacteria, not only as energy
source. Trehalose dimycolate is a glycolipid that exists in the
cell wall of Mycobacterium tuberculosis.[28] This trehalose
derivative specifically binds to the C-type lectin (Mincle) on
macrophages and mediates bacterial adhesion.[29] Accord-
ingly, trehalose-based nanoparticles have been developed to
detect mycobacteria.[30,31] As trehalose glycopolymers are
non-toxic to mammalian cells,[25,32,33] these polymers could
potentially be used in infection treatments, particularly in
anti-adhesion treatment.
Herein, we explored the potential role of trehalose-based
nanoparticles as an anti-adhesive, to prevent adhesion of
bacteria to endothelial cells. Three glyco-gold nanoparticles
(glyco-AuNP) and their performance as inhibitors of bacterial
infections is described and investigated. As trehalose is
a disaccharide based on two a-glucose, connected by 1,1-
glycosidic bond, we choose to use glycopolymers based on
glucose as control. In these glycopolymers, the glucose was
conjugated to the polymer backbone either in C1 or C6
position. While other glycopolymers can be used as control,
we try to specifically answer the question if the disaccharide
structure is essential or if one glucose unit is sufficient.
Therefore, AuNPs were coated with different glycopolymers
(Scheme 1) and the resulting glyco-AuNPs were evaluated in
regard to their cell toxicity, action as anti-adhesives and their
binding to bacteria.
Acrylate-based monomers were polymerized via RAFT
polymerization resulting in homoglycopolymers with similar
degree of polymerizations (DP) of 30 (ESI, Table S1, Fig-
ure S9), followed by deprotection (ESI, Figure S6–S8).
AuNPs were synthesized via the seeded-growth method
based on previous work.[37] To evaluate the effect of nano-
particle (NP) size on inhibition of adhesion, AuNPs of two
different sizes, 62 nm (Au62NP) and 16 nm (Au16NP), were
prepared and analyzed with transmission electron microscopy
(TEM) (Figure 1a and b, ESI, Figure S10). The resulting
AuNPs were coated with the prepared glycopolymers via
thiol-gold interaction by mixing glycopolymers with AuNPs in
water. UV/Vis analysis revealed a red shift (from 537 nm to
539 nm) confirming the successful polymer grafting (Fig-
ure 1c). The increase of the AuNP size after coating was
observed by dynamic light scattering (DLS) measurement
(Figure 1d, ESI, Table S2), and the zeta potential of all
AuNPs ranged between À25 mVand À30 mV, most likely due
to the carboxylic group of the R-group of the RAFT agent.
The negative charge can assist in repelling the adsorption of
blood proteins. In addition, thermal gravimetric analysis
(TGA) was employed to measure the grafting density of
glyco-AuNPs (ESI, Figure S11 and Table S3). In general, the
measured grafting densities ranged from 0.17 to 0.37 chain per
nm2 with PGlu-6 displaying the highest grafting density whilst
PTre displayed the lowest grafting density. Although the chain
lengths of the three glycopolymers were similar, the molec-
ular weights of the glycopolymers were different and hence
the bulkier side groups led to a reduced grafting density.
Endothelial cells function as a physical barrier between
blood and tissue and the ability of microbes to adhere to them
plays a significant role in a number of infection patho-
genesis.[38] For example, bacterial adhesion to endocardial
surfaces results in acute infective endocarditis.[39] Therefore,
HUVECs were chosen as the host cells in this project.
The synthetic strategy to the glycopolymers and detailed
discussed on the synthesis can be found in the supporting
information. Protected C1 and C6 functionalized glucoside
monomers and trehalose monomer were synthesized based
on previously published procedures,[34–36] (ESI, Figure S1–S4).
Prior to studying the ability of nanoparticles to prevent
adhesion, the cytotoxicity of three 62 nm glyco-AuNPs were
Scheme 1. a) Schematic illustration of the synthesis of glyco-AuNPs
and b) schematic representation of anti-adhesion function of glyco-
AuNPs.
Figure 1. TEM images of a) Au62NP and b) Au16NP. c) UV/Vis spectra
of uncoated Au62NP and coated glyco-AuNPs. d) Size increase of
uncoated Au62NP and coated glyco-AuNPs as measured by DLS.
2
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 1 – 8
These are not the final page numbers!

Angewandte
Communications
Chemie
evaluated by WST-1 assay using two healthy cell lines: mouse
macrophage RAW 264.7 and human umbilical vein endothe-
lial cells (HUVECs). After 48 h incubation with glyco-
AuNPs, the cytotoxicity against RAW 264.7 and HUVECs
was determined (ESI, Figure S12). At concentrations up to
0.75 mgmLÀ1, none of the tested glyco-AuNPs displayed
significant toxicity to these cell lines. In fact, glyco-AuNPs
appeared to enhance the proliferation of RAW 264.7 cells in
the range from 0.19 mgmLÀ1 to 0.75 mgmLÀ1. Although
glycopolymers, including trehalose-based polymers, are often
considered non-toxic,[24,40,41] high levels of toxicity is occa-
sionally observed by trehalose-based materials such as nano-
particles prepared from trehalose 6,6’-dimycolate.[42] Based
upon the cytotoxicity results, the highest concentration of
glyco-AuNPs used in the following bio-work was set to
0.6 mgmLÀ1
.
Gram-positive bacteria Staphylococcus aureus 38 served
as the infecting bacteria, as S. aureus is known to strongly
adhere to HUVECs.[43] The Gram-negative bacteria Pseudo-
monas aeruginosa PAO1 was used as a further model. To
better understand the efficiency of glyco-AuNPs as anti-
adhesives, flow cytometry was used as a quantitative tech-
nique.[44] In this established technique, the host cells and
bacteria are stained separately to determine their co-local-
ization. HUVECs and bacteria were stained with Hoechst
33342 and VybrantTM DiD, respectively, before co-incubation
for 2 h with nanoparticles. Afterwards, unbound bacteria and
nanoparticles were removed from the HUVECs by washing
with Hankꢀs balanced salt solution (HBSS). The monolayer of
cells was trypsinized and collected for flow cytometry. The
measured median fluorescence intensity (MFI) of DiD was
correlated directly to the level of bacterial infection. A
sample that was not treated with nanoparticles and a sample
containing only HUVECs were employed as control and
background, respectively. The MFI of the background was
subtracted from the MFI of the control sample, and this was
regarded as 100% infection. The presence of nanoparticles is
expected to decrease the MFI if they inhibit infection. A
multiplicity of infection (MOI), the ratio of bacteria to
HUVECs, was kept at 50:1.
Figure 2. Inhibition of adhesion of bacteria to HUVECs by glyco-
AuNPs. a) Inhibition of S. aureus 38 infection with 62 nm glyco-AuNPs
at 0.15, 0.3, 0.6 mgmLÀ1. b) Inhibition of P. aeruginosa PAO1 infection
by 62 nm glyco-AuNPs at 0.15, 0.3, 0.6 mgmLÀ1. c) Inhibition of
S. aureus 38 infection by 16 nm glyco-AuNPs at 0.15, 0.3, 0.6 mgmLÀ1
d) Inhibition of S. aureus 38 infection using glyco-AuNPs and uncon-
.
jugated glycopolymers with equivalent glycopolymer contents (PGlu-6,
1.8 mM; PTre, 0.8 mM). (C represents control sample containing
HUVEC and bacteria.) Data shown as mean Æ SD; ****P<0.0001. All
P values were from one-way ANOVA followed by Turkey’s multiple
comparisons tests.
may be the interaction of transporters in the surface of S.
aureus. This may include ABC permease LpqY/SugABC
which is specific to trehalose,[48] or other trehalose PTS
permease (TreB) which also act as trehalose transporters in
Gram-positive bacteria.[49,50] Moreover, some surface-associ-
ated enzymes involved in the carbohydrate metabolism[51] can
interact with trehalose.[52,53]
Similarly, the effects of glyco-AuNPs on inhibiting adhe-
sion were studied on Gram-negative bacteria, using P.
aeruginosa PAO1 as a model (Figure 2b). In this case, no
signs of inhibition were observed for any nanoparticles at
concentrations up to 0.6 mgmLÀ1. This implies a lack of
nanoparticle binding interaction with P. aeruginosa. While it
can only be speculated as to why P. aeruginosa does not
respond to the presence of glyco-AuNPs, there are significant
differences in the structure of transporter molecules between
Gram-negative and Gram-positive bacteria. For example, the
trehalose-specific LpqY/SugABC transporter has a substrate
binding protein (SBP), LpqY, on the bacterial surface. In
Gram-positive bacteria Mycobacterium smegmatis (M. smeg-
matis), the arrangement of LpqY (SBP) is significantly
different relative to the rest of the transporter and to the
SBPs found on Gram-negative E. coil.[54] Furthermore, Gram-
negative bacteria possess an outer membrane in the cell wall,
a feature which is not displayed by Gram-positive bacteria.
This outer membrane may also be suspected to hinder the
interaction of nanoparticles with certain transporters.[50] The
All three 62 nm glyco-AuNPs were evaluated at concen-
trations of 0.15 mgmLÀ1, 0.3 mgmLÀ1 and 0.6 mgmLÀ1 using
the S. aureus—HUVEC system. It is evident from Figure 2a
that the percentage of co-associated cells decreased with an
increasing amount of nanoparticles (ESI, Table S4). Most
notable is the inhibition by Au62-PTre which reduced the
association to 54.9 Æ 10.9% at
a
concentration of
0.6 mgmLÀ1. Interestingly, Au-PGlu-6 and Au-PTre were
both capable in inhibiting infection, however, the underlying
mechanism as to how this occurs remains unclear. Bacteria
use carbohydrates in a range of ways and also have means of
taking up trehalose.[45] In the case of S. aureus this bacterium
has a large amount of carbohydrate transporters, including
several phosphotransferase system (PTS) transporters, ATP-
binding cassette (ABC) transporters and others, which
distinguishes it from other Staphylococcus types.[46,47] These
transporters make it possible to metabolize an array of sugars
as many transporters are not specific to one type of sugar.
Therefore, a possible mechanism for the observed inhibition
Angew. Chem. Int. Ed. 2021, 60, 1 – 8
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
outer membrane proteome of P. aeruginosa PAO1 has
a glucoside-sensitive porin but there is no evidence of
surface-associated enzymes involved in carbohydrate metab-
olism.[55]
trehalose-based nanoparticles are promising inhibitors of S.
aureus infection. Nanoparticles based on PGlu-6 still show
remarkable activity although it needs to be taken into account
that the glycopolymer concentration in each particle is higher
due to a higher grafting density.
Next, the inhibitory effect of S. aureus adhesion by the
16 nm glyco-AuNPs was studied to evaluate the effect of the
nanoparticle size (Figure 2c). Interestingly, Au16-PGlu-6 and
Au16-PTre led to similar inhibition effects (43.2 Æ 1.2% and
36.9 Æ 4.4% infection) at 0.6 mgmLÀ1 while Au16-PGlu-
1 resulted to weak inhibition (71.0 Æ 1.8% infection). Au16-
PGlu-6 and Au16-PTre clearly showed a trend in concentra-
tion-dependent inhibition. This result confirmed that the
activity of the inhibitors is dependent on the substitution
position of sugar. Glycopolymers with glucose modified on
the C6 position is significantly more effective as inhibitors
than the analogues based on substitution at the C1 position.
Furthermore, at same NP concentrations, the smaller NPs
Au16-PGlu-6 and Au16-PTre appear to display stronger
inhibition capacity than their larger-sized counterparts. This
is however deceiving as the size of the NPs are now smaller,
and therefore more nanoparticles and a higher fraction of
glycopolymers are present at the same concentration in
mgmLÀ1 (Table 1). Also, variations in grafting densities need
to be taken into account that may lead to different amounts of
sugar per AuNP. Although Au-PGlu-6 and Au-PTre were
similar in performance at the same working concentration,
the amount of trehalose was lower in these samples as the
bulky PTre allowed less glycopolymer attachment onto the
nanoparticle. It can therefore be argued that PGlu-6 is a less
effective inhibitor.
The effect of the size of NPs was reevaluated maintaining
a constant glycopolymer concentration for the nanoparticles.
Free glycopolymers at the same carbohydrate concentration
were used as controls. The concentrations were set to PGlu-6:
1.8 mM, PTre: 0.8 mM, respectively, which equated to
0.6 mgmLÀ1 of the two 62 nm NPs and was adjusted accord-
ingly for the smaller NPs or the free polymers. Upon
comparison of equivalent glycopolymer contents, it became
evident that the inhibition activity of both glyco-AuNPs was
equivalent, whilst the free glycopolymers failed to act as
inhibitors (Figure 2d). This result revealed that the inhibition
is affected by the glycopolymer concentration and not directly
the size of NPs, but the presence of nanoparticles is essential.
Similar results were previously reported where linear gluco-
side polymers showed a lack of binding ability to bacteria,
whereas glucose-based polymeric micelles exhibited a remark-
able affinity to E. coli.[20] It can be concluded at this stage that
The measured activity needs to be contrasted to the
activity displayed by mannose containing polymers. Here, to
achieve 50% inhibition, around 600 mgmLÀ1 Au62-PTre is
required, which equates to a nanoparticle concentration of
0.8 nM or 24.4 mM of trehalose molecules. For comparison,
heptyl a-D-mannoside has been identified as an excellent
mannose-based inhibitors which is a nanomolar antagonist of
FimH.[56] Bacterial adhesion to epithelial cells was decreased
to 22% in the presence of 1 nM (equivalent to 188 nM)
heptylmannoside-based glycopolymer while this was
decreased to around 70% in the presence of 0.1 nM
glycopolymer (equivalent to 18.8 nM mannose).[57] However,
it should be noted the multiplicities of infection (MOI) in our
work was 50 which is higher than the aforementioned work
(MOI = 10). Other examples include Gold manno-glycona-
noparticles as an inhibitor of HIV infection that had IC50
values between 15 and 200 nM per mannose depending on the
architecture,[58] while small molecule a-mannoside inhibitors
that prevent adhesion of E.coli to HT-29 have IC50 values of
around 1 mM.[59] Although mannose based polymers are still
superior, the disadvantage of mannose-based polymers is the
large number of competing receptors in the body.
In the next step, the inhibitory role of the 62 nm glyco-
AuNPs was investigated by confocal laser scanning micro-
scope (CLSM). To identify the distinctive function of sugar,
particularly trehalose, poly(2-hydroxyethyl acrylate) (PHEA)
was synthesized to coat Au62NP (ESI, Table S1). As a result of
bearing neutral and non-bioactive hydroxyl groups, Au62-
PHEA was introduced in the visualization experiments as
negative control. Using the aforementioned infection system,
HUVECs, labelled with Hoechst 33342, were co-incubated
for 2 h with DiD-labeled S. aureus in the presence of 62 nm
glyco-AuNPs. Unbound bacteria and free NPs were also
removed prior to the measurement by CLSM. Non-infected
HUVECs and non-treated infected HUVECs were involved
as background and control separately. Images were collected
after 2 h of incubation (ESI, Figure S13). NPs were observed
using the reflective light, which allows quantitative analysis of
the interaction of different NPs with cells. Significant cellular
uptake of NPs by HUVECs was observed in the case of Au62-
PGlu-1, suggesting that the free hydroxyl group on C6 of
glucose facilitated the cellular interaction, which effectively
removed the NPs from exposure to
bacteria. This significantly reduced
Table 1: Molar concentration of nanoparticles and glycopolymers used in the inhibition assay.
the inhibitory efficiency of Au62-
PGlu-1 and provides a reasonable
explanation for its low perfor-
mance. As shown earlier, the treat-
ment with Au62-PTre led to reduced
adhesion compared to other sam-
ples. To further emphasize the
importance of trehalose as struc-
tural feature, another CLSM study
was conducted to compare Au62-
Sample AuNPs
0.15 mgmLÀ1
0.3 mgmLÀ1
0.6 mgmLÀ1
Chains per Sugars per [Au] [Polymer] [Au] [Polymer] [Au] [Polymer]
AuNP
AuNP
(nM)
0.1
(mM)
(nM)
0.2
(mM)
(nM)
0.4
(mM)
Au62-PGlu-1
Au62-PGlu-6
Au62-PTre
Au16-PGlu-1
Au16-PGlu-6
Au16-PTre
3049
4449
2040
217
269
147
97568
124572
61200
6944
7532
4410
0.3
0.4
0.2
1.3
1.6
0.9
0.6
0.9
0.4
2.6
3.2
1.7
1.2
1.8
0.8
5.1
6.3
3.5
5.9
11.7
23.5
4
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 1 – 8
These are not the final page numbers!

Angewandte
Communications
Chemie
PTre with Au62-PHEA. Infected HUVECs with or without NP
To better understand the mechanism of inhibition of
treatments were fixed followed by staining with Wheat Germ
Agglutinin (WGA) and DAPI to label the plasma membrane
and nucleic acid separately (Figure 3a). In this case, the
invading S. aureus were also stained with DAPI, but the
nucleic acid belonging to the bacteria and the one belonging
to HUVECs could be distinguished by the size as shown by
the highlighted bacteria (red circle) in Figure 3a. The DAPI
channels show a large number of small bacteria, coexisting
with HUVEC cells in the case of the control and Au62-PHEA.
The merged channels highlight the co-localization of both.
Notably, a lower number of bacteria was observed in the
DAPI channel for HUVECs treated with Au62-PTre. This
indicates successful infection inhibition as the Au62-PTre
treatment prevented bacterial adhesion, resulting in the
particles and bacteria being washed away prior to analysis.
glyco-AuNPs, S. aureus were fixed onto coverslips and
allowed to incubate with NPs for 20 min prior to SEM and
TEM analysis (see supporting information). The resulting
SEM images (Figure 3b, ESI, Figure S14a) revealed direct
binding of Au62-PTre to bacteria in comparison to a lack of
interactions for the other nanoparticles. Complementary
TEM imaging provided a close-up showing the nanoparticles
in close vicinity of the surface in the case of Au62-PTre while
no particles were found with the other NP samples (Figure 3b,
ESI, Figure S14a). Counting the number of NPs per bacteria
revealed the remarkable binding ability of Au62-PTre to S.
aureus (Figure 3c). Next, the HUVECs were infected with S.
aureus and treated with Au62-PTre and Au62-PHEA, respec-
tively, for 2 hours (Figure 3d). TEM analysis confirmed that
the number of S. aureus that has invaded HUVEC was
significantly reduced. It was interesting to observe that a few
Figure 3. a) CLSM images of infected HUVECs with or without treatment of nanoparticles. (Examples of bacteria was highlighted by labelling)
b) SEM and TEM images showing interaction of S. aureus with Au62-PTre and Au62-PHEA. c) Quantitative analysis on the binding of nanoparticles
with S. aureus by SEM. d) TEM images of infective HUVECs with treatment of Au62-PTre and Au62-PHEA. Scale bars equal to 200 mm in (a);
200 nm in (c) and 2 mm in (d).
Angew. Chem. Int. Ed. 2021, 60, 1 – 8
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
S. aureus which invaded the cells were still tightly associated
with Au62-PTre. This is in contrast to the treatment with Au62-
PHEA as the invading S. aureus is devoid of any associating
Au62-PHEA within HUVECs. Based on Figure 3d, it is
proposed that trehalose-functionalized NPs adhere to S.
aureus hindering the infection efficiency towards HUVECs.
Although the majority of bacteria surrounded by trehalose-
functionalized nanoparticles were inhibited, a small portion
of the bacteria-nanoparticle assembly was taken up by
HUVECs by its arrangement on the plasma membrane. As
a result, some S. aureus may still be able to enter HUVECs,
but most likely in a random, non-specific manner. However,
the partial inhibition of S. aureus by the trehalose-function-
alized NPs remains sufficient in affecting colonization.
To further examine the favored binding of trehalose with
S. aureus, a competitive experiment evaluated by SEM was
conducted: S. aureus was pre-incubated with free trehalose
(40 mM and 400 mM; 1000- and 10000-fold of PTre) and free
glucose (400 mM; 10000-fold of PTre) for 20 min prior to
incubation with Au62-PTre (0.02 nM AuNP; 0.04 mM PTre).
Incubation in PBS without added sugar was employed as
control sample. Figure 4a shows the SEM images of the
resulting interaction between S. aureus and Au62-PTre,
revealing that decreased binding of nanoparticles to S.
aureus occurred after pre-treatment with free trehalose. A
quantitative analysis based on SEM (Figure 4b) displayed
a trehalose-dose dependent disruption on the binding of Au62-
PTre to S. aureus (ESI, Table S5), while the pretreatment of
free glucose at concentration up to 10000-fold of PTre had
little effects on disrupting the subsequent binding of Au62-
PTre to S. aureus. This competitive experiment demonstrated
the distinctive role of trehalose as an anti-adhesive in S.
aureus infection.
glucose. This could be tested in future work that looks at
various glucose-based polymers that contain spacers of
various nature between the carbohydrate and the polymer
backbone.
In summary, we developed three glyco-AuNPs in two
sizes, based on glucose and trehalose glycopolymers. The
resulting 62 nm glyco-AuNPs did not present significant
toxicity against to RAW 264.7 cells and HUVECs at concen-
tration up to 0.75 mgmLÀ1. The 62 nm NPs were successful to
inhibit the S. aureus infection to HUVECs albeit failed to
inhibit the P. aeruginosa infection to HUVECs. Au-PGlu-
1 acted less efficient as inhibitor, comparing with Au-PGlu-6
and Au-PTre. This implied the advantageous substitution
position of sugar as anti-adhesive. Meanwhile, Au-PGlu-6 and
Au-PTre showed the concentration-dependent inhibitory
activity. Small sized Au-PGlu-6 and Au-PTre achieved
stronger inhibition than the larger AuNPs. However, this
difference could be assigned to the presented amount of
glycopolymer on the NPs. In this work, two sized trehalose-
functionalized NPs displayed impressive inhibition to S.
aureus infection, with performance surpassing the two
glucose-functionalized nanoparticles. The preferential bind-
ing of trehalose-functionalized NPs to S. aureus was identified
by SEM and TEM. This study indicated trehalose as
a promising candidate in anti-adhesion treatment. Future
work could involve tests with different pathogenic system. In
addition, it is also worth exploring if and how trehalose-based
therapeutic agents will interfere with the host system.
Acknowledgements
MS would like to thank the Australian Research Council for
funding (ARC DP200101918). The authors would like to
thank the Mark Wainwright Analytical Centre for their
support.
While the better activity of PTre is evident, it is still
interesting to see that PGlu-6 has a good performance. It is
intriguing to think that it is not the attachment of trehalose
that causes the anti-adhesive behavior, but simply the
presence of glucose connected at 6-position to the polymer.
This raises the question if the second glucose unit in PTre
might just act as a stiff spacer that directs the binding of
Conflict of Interest
The authors declare no conflict of interest.
Keywords: anti-adhesive agents · bacteria inhibition ·
glycopolymers · gold nanoparticles · trehalose
[1] I. Ofek, D. L. Hasty, N. Sharon, FEMS Immunol. Med. Micro-
biol. 2003, 38, 181 – 191.
[2] N. Sharon, H. Lis, Glycobiology 2004, 14, 53R – 62R.
[3] S. Sattin, A. Bernardi, Trends Biotechnol. 2016, 34, 483 – 495.
[4] S. R. S. Ting, G. Chen, M. H. Stenzel, Polym. Chem. 2010, 1,
1392 – 1412.
[5] X. Li, G. Chen, Polym. Chem. 2015, 6, 1417 – 1430.
[6] I. Pramudya, H. Chung, Biomater. Sci. 2019, 7, 4848 – 4872.
[7] M. W. Jones, L. Otten, S.-J. Richards, R. Lowery, D. J. Phillips,
D. M. Haddleton, M. I. Gibson, Chem. Sci. 2014, 5, 1611 – 1616.
[8] S. M. Rele, W. Cui, L. Wang, S. Hou, G. Barr-Zarse, D. Tatton, Y.
Gnanou, J. D. Esko, E. L. Chaikof, J. Am. Chem. Soc. 2005, 127,
10132 – 10133.
Figure 4. a) SEM images showing S. aureus interaction with Au62-PTre
with pre-treatment. (Left top: pre-treatment with PBS; right top: pre-
treatment with 40 mM free trehalose; left bottom: pre-treatment with
400 mM free trehalose; right bottom: pre-treatment with 400 mM free
glucose.) Scale bar equals to 500 nm. b) Quantitative analysis on the
binding of nanoparticles per S. aureus cell with different pre-treatment
by SEM images.
[9] M. A. Mintzer, E. L. Dane, G. A. OꢀToole, M. W. Grinstaff, Mol.
Pharm. 2012, 9, 342 – 354.
6
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 1 – 8
These are not the final page numbers!

Angewandte
Communications
Chemie
[10] S. Bhatia, L. C. Camacho, R. Haag, J. Am. Chem. Soc. 2016, 138,
[35] M. Nasiri, T. M. Reineke, Polym. Chem. 2016, 7, 5233 – 5240.
8654 – 8666.
[36] A. Miyagawa, R. Tomita, K. Kurimoto, H. Yamamura, Synth.
Commun. 2016, 46, 556 – 562.
[37] H. Lu, J. Su, R. Mamdooh, Y. Li, M. H. Stenzel, Macromol.
Biosci. 2020, 20, 1900221.
[38] S. K. Sahni, Thromb. Res. 2007, 119, 531 – 549.
[39] H. L. Bleich, E. S. Boro, L. Weinstein, J. J. Schlesinger, N. Engl.
J. Med. 1974, 291, 1122 – 1126.
[40] M. Ahmed, B. F. L. Lai, J. N. Kizhakkedathu, R. Narain,
Bioconjugate Chem. 2012, 23, 1050 – 1058.
[41] J. Lee, E. W. Lin, U. Y. Lau, J. L. Hedrick, E. Bat, H. D.
Maynard, Biomacromolecules 2013, 14, 2561 – 2569.
[42] B. J. Spargo, L. M. Crowe, T. Ionedaf, B. L. Beaman, J. H. Crowe,
Cord Factor (a,a-Trehalose 6,6’-Dimycolate) Inhibits Fusion
between Phospholipid Vesicles (Trehalose/Membrane Fusion/
Liposomes/Tuberculosis/Nocardiosis), 1991.
[43] S. K. Ogawa, E. R. Yurberg, V. B. Hatcher, M. A. Levitt, F. D.
Lowy, Bacterial Adherence to Human Endothelial Cells In Vitro,
1985.
[44] B. Hara-Kaonga, T. G. Pistole, J. Microbiol. Methods 2007, 69,
37 – 43.
[45] M. Vanaporn, R. W. Titball, Virulence 2020, 11, 1192 – 1202.
[46] J. M. Jeckelmann, B. Erni, Pfluegers Arch. Eur. J. Physiol. 2020,
472, 1129 – 1153.
[11] F. Lehmann, E. Tiralongo, J. Tiralongo, Cell. Mol. Life Sci. 2006,
63, 1331 – 1354.
[12] K. A. Krogfelt, H. Bergmans, P. Klemm, Infect. Immun. 1990, 58,
1995 – 1998.
[13] T. B. H. Geijtenbeek, D. S. Kwon, R. Torensma, S. J. Van Vliet,
G. C. F. Van Duijnhoven, J. Middel, I. L. M. H. A. Cornelissen,
H. S. L. M. Nottet, V. N. KewalRamani, D. R. Littman, C. G.
Figdor, Y. Van Kooyk, DC-SIGN, a Dendritic Cell-Specific HIV-
1-Binding Protein That Enhances Trans-Infection of T Cells,
Steinman And Inaba, 2000.
[14] Y. Guo, H. Feinberg, E. Conroy, D. A. Mitchell, R. Alvarez, O.
Blixt, M. E. Taylor, W. I. Weis, K. Drickamer, Nat. Struct. Mol.
Biol. 2004, 11, 591 – 598.
[15] Q. Zhang, J. Collins, A. Anastasaki, R. Wallis, D. A. Mitchell,
C. R. Becer, D. M. Haddleton, Angew. Chem. Int. Ed. 2013, 52,
4435 – 4439; Angew. Chem. 2013, 125, 4531 – 4535.
[16] M. Andreini, D. Doknic, I. Sutkeviciute, J. J. Reina, J. Duan, E.
Chabrol, M. Thepaut, E. Moroni, F. Doro, L. Belvisi, J. Weiser, J.
Rojo, F. Fieschi, A. Bernardi, Org. Biomol. Chem. 2011, 9, 5778 –
5786.
[17] S. Beloin, A. Barras, F. A. Martin, O. Bande, J.-S. Baumann, J.-
M. Ghigo, R. Boukherroub, C. Beloin, A. Siriwardena, S.
Szunerits, Nanoscale 2013, 5, 2307 – 2316.
[47] N. P. Vitko, M. R. Grosser, D. Khatri, T. R. Lance, A. R.
Richardson, MBio 2016, 7, e00296-16.
[48] V. G. Lewis, M. P. Ween, C. A. Mcdevitt, Protoplasma 2012, 249,
919 – 942.
[49] H. Tournu, A. Fiori, P. Van Dijck, PLoS Pathog. 2013, 9,
e1003447.
[50] U. Andersson, F. Levander, P. Rꢂdstrçm, J. Biol. Chem. 2001,
276, 42707 – 42713.
[18] B. D. Polizzotti, K. L. Kiick, Biomacromolecules 2006, 7, 483 –
490.
[19] S. Bhatia, M. Hilsch, J. L. Cuellar Camacho, K. Ludwig, C. Nie,
B. Parshad, M. Wallert, S. Block, D. Lauster, C. Bçttcher, A.
Herrmann, R. Haag, Angew. Chem. Int. Ed. 2020, 59, 12417 –
12422; Angew. Chem. 2020, 132, 12517 – 12522.
[20] S. R. S. Ting, E. H. Min, P. B. Zetterlund, M. H. Stenzel, Macro-
molecules 2010, 43, 5211.
[51] S. Khan, N. Cole, E. B. H. Hume, L. L. Garthwaite, T. Nguyen-
Khuong, B. J. Walsh, M. D. P. Willcox, Exp. Eye Res. 2016, 151,
171 – 178.
[21] L. East, C. M. Isacke, Biochim. Biophys. Acta Gen. Subj. 2002,
1572, 364 – 386.
[22] A. B. Richards, S. Krakowka, L. B. Dexter, H. Schmid, A. P. M.
Wolterbeek, D. H. Waalkens-Berendsen, A. Shigoyuki, M.
Kurimoto, Food Chem. Toxicol. 2002, 40, 871 – 898.
[23] R. J. Mancini, J. Lee, H. D. Maynard, J. Am. Chem. Soc. 2012,
134, 8474 – 8479.
[24] Y. Liu, J. Lee, K. M. Mansfield, J. H. Ko, S. Sallam, C.
Wesdemiotis, H. D. Maynard, Bioconjugate Chem. 2017, 28,
836 – 845.
[25] A. Sizovs, L. Xue, Z. P. Tolstyka, N. P. Ingle, Y. Wu, M. Cortez,
T. M. Reineke, J. Am. Chem. Soc. 2013, 135, 15417 – 15424.
[26] J. C. Argꢁelles, J. Mol. Evol. 2014, 79, 111 – 116.
[27] B. Sacktor, Proc. Natl. Acad. Sci. USA 1968, 60, 1007 – 1014.
[28] H. Noll, H. Bloch, J. Asselineau, E. Lederer, Biochim. Biophys.
Acta 1956, 20, 299 – 309.
[29] E. Ishikawa, T. Ishikawa, Y. S. Morita, K. Toyonaga, H. Yamada,
O. Takeuchi, T. Kinoshita, S. Akira, Y. Yoshikai, S. Yamasaki, J.
Exp. Med. 2009, 206, 2879 – 2888.
[52] Y. J. Kim, Acta Crystallogr. Sect. F 2018, 74, 277 – 282.
[53] D. Araiza-Olivera, J. G. Sampedro, A. Mffljica, A. PeÇa, S. Uribe-
Carvajal, FEMS Yeast Res. 2010, 10, 282 – 289.
[54] F. Liu, J. Liang, B. Zhang, Y. Gao, X. Yang, T. Hu, H. Yang, W.
Xu, L. W. Guddat, Z. Rao, Sci. Adv. 2020, 6, eabb9833.
[55] A. S. Nouwens, M. D. P. Willcox, B. J. Walsh, S. J. Cordwell,
Proteomics 2002, 2, 1325 – 1346.
[56] J. Bouckaert, J. Berglund, M. Schembri, E. De Genst, L. Cools,
M. Wuhrer, C.-S. Hung, J. Pinkner, R. Slättegꢂrd, A. Zavialov, D.
Choudhury, S. Langermann, S. J. Hultgren, L. Wyns, P. Klemm,
S. Oscarson, S. D. Knight, H. De Greve, Mol. Microbiol. 2005,
55, 441 – 455.
[57] A. Sivignon, X. Yan, D. A. Dorta, R. Bonnet, J. Bouckaert, E.
Fleury, J. Bernard, S. G. Gouin, A. Darfeuille-Michaud, N.
Barnich, MBio 2015, 6, e01298-15.
[58] O. Martínez-vila, L. M. Bedoya, M. Marradi, C. Clavel, J.
Alcamí, S. PenadØs, ChemBioChem 2009, 10, 1806 – 1809.
[59] M. Hartmann, H. Papavlassopoulos, V. Chandrasekaran, C.
Grabosch, F. Beiroth, T. K. Lindhorst, C. Rçhl, FEBS Lett. 2012,
586, 1459 – 1465.
[30] N. Hao, X. Chen, S. Jeon, M. Yan, Adv. Healthcare Mater. 2015,
4, 2797 – 2801.
[31] X. Chen, B. Wu, K. W. Jayawardana, N. Hao, H. S. N. Jayawar-
dena, R. Langer, A. Jaklenec, M. Yan, Adv. Healthcare Mater.
2016, 5, 2007 – 2012.
[32] S. Srinivasachari, Y. Liu, L. E. Prevette, T. M. Reineke, Bioma-
terials 2007, 28, 2885 – 2898.
[33] K. M. Mansfield, H. D. Maynard, ACS Macro Lett. 2018, 7, 324 –
329.
[34] A. Piloni, A. Walther, M. H. Stenzel, Polym. Chem. 2018, 9,
4132 – 4142.
Manuscript received: May 15, 2021
Revised manuscript received: July 19, 2021
Accepted manuscript online: August 13, 2021
Version of record online: && &&, &&&&
Angew. Chem. Int. Ed. 2021, 60, 1 – 8
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
7
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Glycopolymers
Due to increased resistance against anti-
biotics, the fight against bacteria is
becoming increasingly difficult. Here, we
used a sugar that is unique to the bacteria
metabolism, trehalose, and prepared tre-
halose-based polymer-coated gold nano-
particles. These nanoparticles were
shown to inhibit the infection of healthy
cells by S. aureus and could serve as
a non-toxic alternative to prevent adhe-
sion of certain bacteria.
Y. Li, N. Ariotti, B. Aghaei, E. Pandzic,
S. Ganda, M. Willcox, M. Sanchez-Felix,
M. Stenzel*
&&&& — &&&&
Inhibition of S. aureus Infection of
Human Umbilical Vein Endothelial Cells
(HUVECs) by Trehalose- and Glucose-
Functionalized Gold Nanoparticles
8
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 1 – 8
These are not the final page numbers!