Amphiphilic Polyphenylene Dendron Conjugates for Surface Remodeling of Adenovirus 5

Article abstract of DOI:10.1002/anie.201913708

Amphiphilic surface groups play an important role in many biological processes. The synthesis of amphiphilic polyphenylene dendrimer branches (dendrons), providing alternating hydrophilic and lipophilic surface groups and one reactive ethynyl group at the

Full text of DOI:10.1002/anie.201913708

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Amphiphilic Polyphenylene Dendron Conjugates for Surface
Remodeling of Adenovirus 5
Authors: Jessica Wagner, Longjie Li, Johanna Simon, Katharina
Landfester, Volker Mailänder, Klaus Müllen, David Y.W. Ng,
Yuzhou Wu, Tanja Weil, and Lea Krutzke
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.201913708
Angew. Chem. 10.1002/ange.201913708
Link to VoR: http://dx.doi.org/10.1002/anie.201913708
http://dx.doi.org/10.1002/ange.201913708

10.1002/anie.201913708
Angewandte Chemie International Edition
RESEARCH ARTICLE
Amphiphilic Polyphenylene Dendron Conjugates for Surface
Remodeling of Adenovirus 5
Jessica Wagner,[a,b] Longjie Li,[c] Johanna Simon,[a] Lea Krutzke,[d] Katharina Landfester,[a] Volker
Mailänder,[a,e] Klaus Müllen,[a] David Y.W. Ng,*[a] Yuzhou Wu*[a,c] and Tanja Weil*[a]
Abstract: The arrangement of amphiphilic surface groups plays an
important role in many biological processes such as protein folding or
biorecognition. We report the synthesis of amphiphilic polyphenylene
dendrimer branches (dendrons) providing alternating hydrophilic and
lipophilic surface groups and one reactive enthynyl group at the core.
The amphiphilic surface groups serve as biorecognition units that bind
to the surface of adenovirus 5 (Ad5), which is one of the most common
vectors in gene therapy. The Ad5/dendron complexes showed high
gene transduction efficiencies in coxsackie-adenovirus receptor
(CAR)-negative cells. Moreover, the dendrons offer incorporating new
functions at the dendron core by in situ post-modifications even when
bound to the Ad5 surface. Surfaces coated with these amphiphilic
dendrons were analyzed for their blood protein binding capacity,
which is essential to predict their performance in the blood stream. In
this way, we provide a new platform for introducing bioactive groups
to the Ad5 surface without the necessity to chemically modify the virus
particles.
nanoparticles that also directs their transport in vivo.[3] It has been
demonstrated that the variation of surface charges[4] or coating of
nanoparticles e.g. with polymers like polyethylene glycol[5] has an
impact on the protein corona and often controls their
aggregation[6], biodistribution[7] as well as cellular uptake
properties[5]. By employing amphiphilic surface patterns on
nanoparticles, their influence on biological systems was studied.[8]
It is still very challenging to control the surface contour of
nanoparticles[8b] and to impart distinct amphiphilic surface
patterns with molecular precision that maintain their perfect
nanosize definition in various biological environments[9].
Therefore, highly branched macromolecules with precise
structures and molecular weights such as dendrimers have
emerged as a monodisperse platform providing characteristic
features of proteins.[10] Hence, they are often referred to as
“artificial proteins”[11] and their applications in biomedicine range
from drug delivery of serum albumin mimicking polyphenylene
dendrimers[12] to multivalent dendrimers as antiviral drugs[13] and
gene delivery agents[14]. For example, it has been demonstrated
that dendrons bind to a virus capsid via supramolecular
interactions leading to an electrostatically driven self-assembly
into dendron-virus complexes. These complexes could be
disassembled by an optical trigger to release the virus.[15]
Amphiphilic polyphenylene dendrimers (PPDs) are macro-
molecules with given surface patterns consisting of e.g.
alternating sulfonic acid and n-propyl groups.[16] These
dendrimers are internalized into cells while showing low toxicity
both in vitro and in vivo and they possess the ability to transport
lipophilic drugs within their unpolar inner cavities.[12] PPDs are
unique due to the rigidity of their sterically demanding and space
filling pentaphenyl-benzene scaffold and therefore provide
persistent three-dimensional structures.[17] This class of
dendrimer has the advantage that surface patterns can be exactly
positioned since no backfolding of single dendritic arms
(dendrons) can occur.[18] Furthermore, we have shown previously
that out of a set of amphiphilic PPDs, only one type of PPDs with
high density of amphiphilic surface patterns was able to bind to
adenovirus 5 (Ad5).[19] Less branched amphiphilic PPDs showed
a significantly lower binding to Ad5 and a negatively charged
PPD-surface did not lead to any binding. These findings indicated
that the dense amphiphilic surface motif is required for Ad5
binding.[19]
Introduction
Amphiphilicity plays an important role in the formation of
biological architectures such as the structure of proteins, the self-
assembly of peptides, or the build-up of biological membranes.[1]
Due to the characteristics of amphiphiles to organize into higher
ordered structures[2], their interactions with other biomolecules is
a complex process of high biological relevance, which is still not
fully understood. For example, the exposure of nanomaterials like
polymers, liposomes, or nanoparticles to biological fluids, such as
human blood plasma, gives rise to a protein corona around
[a]
J. Wagner, Dr. J. Simon, Prof. Dr. K. Müllen, Prof. Dr. K. Landfester,
Prof. Dr. V. Mailänder, Prof. Dr. Y. Wu, Dr. D.Y.W. Ng, Prof. Dr. T.
Weil
Max Planck Institute for Polymer Research
Ackermannweg 10, 55128 Mainz (Germany)
Email: david.ng@mpip-mainz.mpg.de
tanja.weil@mpip-mainz.mpg.de
[b]
[c]
J. Wagner
Graduate School Materials Science in Mainz,
Staudingerweg 9, 55128 Mainz (Germany)
L. Li, Prof. Dr. Wu
Hubei Key Laboratory of Bioinorganic Chemistry and Materia
Medica, School of Chemistry and Chemical Engineering
Huazhong University of Science and Technology
1037 Luoyu Road, 430074 Wuhan (China)
Email: wuyuzhou@hust.edu.cn
Adenovirus (Ad) is a non-enveloped double stranded DNA
virus with an icosahedral capsid infecting respiratory epithelial
cells.[20] Ads are the most common vectors in gene therapy due to
their significant advantages such as genetic stability, well-
characterized biology, and high transduction efficiency in cells.[20a,
[d]
[e]
Dr. L. Krutzke
University clinic, Department of Gene Therapy,
Helmholtzstr. 8/1, 89081 Ulm (Germany)
Prof. Dr. V. Mailänder
21]
They enter cells by specific interaction with the coxsackie-
Department of Dermatology, University Medical Center of the
Johannes Gutenberg-University Mainz, Langenbeckstr. 1, 55131
Mainz (Germany)
adenovirus receptor (CAR) and integrins limiting applications to
such cell types.[22] Moreover, the three major capsid proteins —
hexon protein, penton base, and fiber — bind to antibodies, which
lead to immunogenic responses or neutralization, which needs to
Supporting information for this article is given via a link at the end of
the document
This article is protected by copyright. All rights reserved.

10.1002/anie.201913708
Angewandte Chemie International Edition
RESEARCH ARTICLE
be reduced for in vivo applications.[20a] One strategy focuses on
shielding the Ad surface from antibody binding by covalent
attachment of polymers like polyethylene glycol, which could also
lead to reduced transduction efficacies.[23]
PPDs in biomedicine. Therefore, a dendritic structure is required
that enables the incorporation of additional functionalities (Fig. 1).
Consequently, we combine the defined amphiphilic pattern
of PPD3 for biorecognition with a novel synthetic handle enabling
post-modifications (Fig. 1). Since PPDs are symmetric
macromolecules, the incorporation of an additional feature is
challenging while retaining the desired amphiphilic surface
structures. Therefore, we desymmetrized the PPD3 structure and
synthesized the first amphiphilic polyphenylene dendron, i.e. a
dendrimer branch corresponding to one quarter of the entire
PPD3 that should combine Ad5 binding features with the potential
We have discovered recently that the formation of an
amphiphilic PPD corona promotes cellular internalization into
CAR-negative cells that cannot be intrinsically targeted by Ad5.[19]
In human blood serum, neutralization by antibodies and binding
of coagulation factor X (FX), the primary transport mechanism of
Ad5 to the liver, have been altered significantly after PPD
adsorption. We could show that there are no electrostatic
interactions between the positively charged fibers of Ad5 and
negatively charged sulfonic acids of the PPD. In addition, FX
could not bind to Ad5 when shielded with the dendrimer indicating
that the PPD blocks the binding site for FX. As the amphiphilic
PPDs bind to the virus capsid proteins, they also impart a novel
surface pattern onto Ad5 controlling their various interactions with
other blood serum proteins.[19] In consequence, reduced gene
transduction in liver tissue and an enhanced transduction in heart
tissue was observed in vivo.[19] Thus, amphiphilic PPDs provide a
novel platform for virus redirection into different cells and tissues
due to their ability to coat and protect Ad5 from FX binding, which
influences cellular uptake and biodistribution of Ad5 in vivo.
Enhanced structural variability of the dendrimer scaffold is a
prerequisite to further advance the applicability of amphiphilic
for post-modifications. We demonstrate
a novel multistep
synthesis of an amphiphilic polyphenylene dendron with a
propargyl-modified triethylene glycol linker at the core. By
applying this linker, high water solubility as well as the possibility
to introduce functional units like a fluorophore for imaging or a bio-
orthogonal
group via copper(I)-catalyzed
azide-alkyne
cycloaddition (CuAAC) are envisioned. Additionally, upon com-
plexation with Ad5, the alkyne group can act as a functional
handle for in situ CuAAC to serve as a versatile platform for
introducing chemical modifications on the viral surface.
Furthermore, we analyzed the binding of blood proteins in order
to understand the influence of amphiphilic dendron-coated
surfaces in the blood stream.
Figure 1. Structural design of an amphiphilic polyphenylene dendron by desymmetrization of amphiphilic dendrimer PPD3. By the employment of a “click”-able
ethynyl group, the introduction of a second function by post-modification was achieved. The amphiphilic pattern of these dendron-conjugates interacts with biological
structures like proteins which was verified by interactions of dendron-coated liposomes with blood serum proteins (protein structures: PDB-files 4NHH[24], 1FZC[25]
5Z0B[26]) as well as binding to Ad5 for re-direction into CAR-negative cells.
,
This article is protected by copyright. All rights reserved.

10.1002/anie.201913708
Angewandte Chemie International Edition
RESEARCH ARTICLE
neopentyl-protected sulfonic acid group was applied as end-
capping unit and was synthesized based on modified protocols
from Stangenberg et al.[16] The syntheses (Fig. S1B) and reaction
conditions are summarized in the SI.
In this study, neopentyl-protected amphiphilic poly-
phenylene dendron 7 was synthesized and post-modified with a
propargyl-TEG-linker followed by attachment of D-biotin or
Cyanine 5 (Cy5)-moiety via CuAAC (“Click” reaction) (Scheme 1).
Cy5 was introduced for cellular uptake experiments and co-
visualization of dendron uptake and Ad5 gene delivery. D-Biotin
was attached as an example of a bio-orthogonal group to study
the influence of a functional group on Ad5 binding.
Results and Discussion
The divergent synthesis of a second-generation amphiphilic
dendron (Scheme 1) is based on iterative [4+2] Diels–Alder
cycloadditions of an ethynylated core with tetraphenylcyclopenta-
dienones (CPs). The CP either determines the branching or
serves as an end-capping unit introducing the surface patterns.[27]
As previously reported, a high density of amphiphilic surface
groups resulting from a highly branched dendrimer scaffold leads
to Ad5 binding.[19] In order to integrate this dense surface pattern
in a dendron scaffold, we synthesized a dendritic branch with
similar amphiphilic surface groups, representing one quarter of
the entire dendrimer. Therefore, the AB4 building block 1 with four
branching points was used, which was synthesized based on
modified protocols from Morgenroth et al.[28] The synthesis
scheme of the AB4 building block 1 (SI, Fig. S1A) as well as all
reaction conditions are summarized in the Supporting Information
(SI). The bifunctionalized CP 2 with an n-propyl group and
The detailed structure of the dendron core is crucial since its
accessibility as well as peripheral functionality after dendron
growth are required. Dendron 7 was synthesized from
a
bifunctional biphenyl dendron core 3, which allowed minimizing
steric hindrance at the post-modification step. The iodo group of
Scheme 1. Synthesis of amphiphilic polyphenylene dendron conjugates. a) Benzophenone, toluene, molecular sieve 4Å, reflux, 15 h; b) TIPS-acetylene CuI,
Pd(Ph3P)2Cl2, THF/ NEt3 (5:1), RT, 15 h; c) TBAF, THF, 0 °C, 0.5 h; d) AB4 building block, o-xylene, 160 °C, 24 h; c) TBAF, THF, 0 °C, 0.5 h; e) Surface building
block, o-xylene, 145 °C, 48 h; f) HCl, THF, 20 min; g) propargyl-TEG-COOH, EDC·HCl, DMAP, DMF, RT, 24 h; h) DMF, 180 °C, 36 h; i) R–N3, CuSO4, sodium
ascorbate, TBTA, DMF/H2O, RT, 24 h.
This article is protected by copyright. All rights reserved.

10.1002/anie.201913708
Angewandte Chemie International Edition
RESEARCH ARTICLE
dendron core 3 enabled the coupling with an ethynyl group that is
required for dendron growth while the aniline group offered the
possibility for post-modification at the focal point after dendron
synthesis. In a condensation reaction, the aniline group was
protected with benzophenone resulting in an imine to prevent side
reactions of the amine group during the harsh conditions of PPD
synthesis. In the next step, the triisopropylsilyl (TIPS) protected
ethynyl group was introduced by Sonogashira–Hagihara coupling
followed by removal of the silyl groups to afford dendron core 4 in
good yields (75% over three steps). Further dendron growth was
conducted by utilizing branching unit 1 (AB4[28]) in a [4+2] Diels–
Alder cycloaddition under standard conditions. Since dendron
core 4 only features one dienophile (ethynyl group), the reaction
time was reduced from 48 h[12] to 24 h. After deprotection of the
ethynyl groups, the first generation dendron 5 was obtained in
65% yield over two steps. Subsequently, the next [4+2] Diels–
Alder reaction with amphiphilic end capping unit 2 was performed
for 48 h at reduced temperatures of 145 °C to avoid deprotection
of the sulfonic acid groups.[16] Under these conditions, the imine
protective group was partially cleaved so that imine-dendron 6
(62%) as well as amine-dendron 7 (34%) were isolated by column
chromatography. The remaining imine protective group of
dendron 6 was removed by acidic treatment to afford amine-
dendron 7. After ligation of 7 to a triethylene glycol (TEG)
derivative (propargyl-TEG-linker) by amide coupling, the sulfonic
acid groups were deprotected thermally to obtain propargyl-TEG-
dendron 8. The removal of the neopentyl group requires high
temperatures so that the heat-sensitive bioactive groups were
attached after deprotecting the sulfonic acid groups. D-Biotin and
Cy5 derivatives were ligated by ligand-accelerated CuAAC,
applying tris((1-benzyl-4-triazolyl)methyl)amine (TBTA) as ligand.
The crude products were purified by gel permeation chromato-
Figure 2. Dendron-coated liposomes form a protein corona in blood serum/plasma. (A) Coating of liposomes with dendron 8 and incubation with blood serum or
blood plasma leading to protein corona (protein structures: PDB-files 4NHH[24], 1FZC[25], 5Z0B[26]). Heat map of adsorbed proteins to dendron 8 and dendrimer
PPD3-coated liposomes in (B) blood serum and (C) blood plasma. The amount of each protein is given in % based on all identified corona proteins. A list of all
identified proteins is provided in the SI (Fig. S34-37).
This article is protected by copyright. All rights reserved.

10.1002/anie.201913708
Angewandte Chemie International Edition
RESEARCH ARTICLE
graphy (GPC) in dimethylformamide to remove CuAAC reagents
and unreacted starting materials to afford biotinylated dendron 9a
and Cy5-dendron 9b. A detailed reaction scheme is provided in
the Supporting Information (SI) (Fig. S2).
the blood circulation time and hereby, also reduce the interaction
with targeted cells.[35] As reported in the literature, the protein
source additionally shapes the protein corona formation.[36] This
is in line with our findings as we have observed differences in the
protein corona composition after serum and plasma incubation
(Fig. 2B versus C).
For serum preparation, blood was clotted and then centrifuged to
remove the clot. The resulting supernatant no longer contained all
proteins. Due to this preparation, fibrinogen and other clotting
factors were removed. In contrast, for plasma preparation, an
anticoagulant was added to prevent blood clotting and therefore
plasma contained all blood proteins including the clotting factors.
For dendron 8 and PPD3-covered liposomes, the amount of
fibrinogen was higher than for the unfunctionalized ones after
plasma incubation. Comparable results were obtained for
dendron 8 and PPD3-coated PS (Fig. S16-18) indicating that the
protein interactions are governed by the dendron 8 and PPD3
coating and not by the underlying base material. In addition, minor
differences were observed for the protein corona composition of
Due to the asymmetry of surface building block 2, second
generation dendrons were obtained as constitutional isomers, as
29]
reported previously.[16,
The constitutional isomers were
confirmed by 1H NMR spectroscopy, where they are most notably
visible in the spectra of the neopentyl-protected dendrons (SI, Fig.
S5). NMR spectroscopy and MALDI-TOF spectrometry
demonstrate the successful synthesis of propargyl-modified
dendron 8 (SI Fig. S6-7 and S13) and post-modification by
CuAAC to achieve biotinylated dendron 9a and Cy5 labelled
dendron 9b (SI Fig. S8-S12 and S14-S15). The signals in the 1H
NMR spectrum can be assigned to propargyl-modified dendron 8,
biotinylated dendron 9a and Cy5 labelled dendron 9b. The
detailed synthesis description as well as characterization of
dendron intermediates and final products are summarized in the
SI.
It is well known that Ads are involved in several protein
interactions in the blood stream.[30] However, after PPD3
complexation, blood coagulation factor X (FX) could not bind to
Ad5.[30] In order to shed light on the potential interaction partners
of amphiphilic dendrons in the blood stream, we analyzed proteins
binding to surfaces coated with 8. The virus capsid was simplified
by applying nanocarriers as already validated model systems with
less complexity.[31] Thus, liposomes[31a] prepared from 1,2-
dioleoyl-sn-glycero-3-phosphoethanolamine
(DOPE),
L-α-
phosphatidylcholine (egg PC) and cholesterol (Chol)
(PC:DOPE:Chol = 1:1:1, Ø = 242 ± 6 nm) as well as polystyrene
particles (PS-NH2, Ø= 98 ± 10 nm)[31b] with comparable sizes to
Ad5 were coated with the amphiphilic dendrimer PPD3 as well as
dendron 8 (for details see SI). Then, the protein adsorption in
blood plasma as well as in blood serum was analyzed. As
mentioned above, changes of surface charges as well as polymer
coatings can influence the protein corona of nanoparticles.[4b] As
previously reported, amphiphilic patterned PPDs seem to bind to
a lipid monolayer via electrostatic interactions between negatively
charged sulfonic acid groups and positively charged headgroups
of the lipid (SI, Fig. S22).[32] Here, the dendron/dendrimer-coated
liposomes and polystyrene (PS) particles were prepared following
a standard protocol.[5, 33] Briefly, liposomes or nanoparticles were
incubated with excess dendron or PPD3 and purified by
centrifugation. The coating was verified by DLS and zeta potential
(SI, Table S1). Then, blood serum or plasma was added (Fig. 2A).
After centrifugation, washing steps and desorption of corona
proteins, the isolated proteins were analyzed qualitatively by
SDS-PAGE (SI, Fig. S17-20) and quantitatively by Pierce Assay
(Fig. S21) and LC-MS/MS (Fig. 2B, C and SI Fig. S16).
Coating with dendron 8 or PPD3 had a major impact on
protein corona formation of liposomes (Fig. 2) and PS
nanoparticles (SI, Fig. S16). For unfunctionalized liposomes and
PS nanoparticles, we detected high amounts of albumin
(~26 ± 3%) and immunoglobulins (e.g. Ig kappa, ~6 ± 1%) in the
protein corona after serum incubation (Fig. 2B). However, after
coverage with dendron 8 (lipo-dendron) the amount of Ig kappa is
significantly lower (~2 ± 0.2%). Immunoglobulins belong to the
protein class of opsonins and can mediate the interaction with
phagocytic cells[34]. The adsorption of IgG can dramatically reduce
Figure 3. Cellular internalization of amphiphilic dendrons and cell viability on
CHO-K1 (A) Confocal image of CHO-K1 cells incubated with 1 μM Cy5-Dendron
9b for 24 h and nucleus staining with Hoechst 33258 (scale = 20 µM) (B)
Cellular viability of propargyl-dendron 8 and biotin-dendron 9a compared to
PPD3 by applying four times higher dendron concentration to achieve
approximate same quantities of surface patterns. Cell viability was tested with
CellTiter-Glo®-Assay. Data from three independent experiments with
quadruplets (total n = 12) is shown.
This article is protected by copyright. All rights reserved.

10.1002/anie.201913708
Angewandte Chemie International Edition
RESEARCH ARTICLE
lipo-dendron and lipo-dendrimer after serum and plasma
incubation. This effect was most prominent for β-2 glycoprotein 1,
also known as Apolipoprotein H (ApoH).[4b] The corona of lipo-
dendron was enriched with ApoH (8 ± 0.5% serum, 11 ± 1.5%
plasma), whereas lower amounts were detected for lipo-
dendrimer (5 ± 0.3% serum, 4 ± 0.3% plasma) or unfunctionalized
liposomes (0.5 ± 0.2% serum, 0.3 ± 0.5% plasma). It has been
previously shown that due to coating of PS nanoparticles with
ApoH, the interaction with human mesenchymal stem cells is
favored.[4b] A similar trend was observed for the adsorption of
vitronectin, which increased in case of lipo-dendron compared to
PPD3-coated or unfunctionalized liposomes. Here, we found that
the protein corona of lipo-dendron was enriched with vitronectin
(9 ± 0.8% serum, 6 ± 0.5% plasma), whereas lower amounts were
adsorbed on lipo-dendrimer (6 ± 0.5% serum, 3 ± 0.2% plasma)
or unfunctionalized liposomes (1 ± 0.7% serum, 0.2 ± 0.3%
plasma). Interestingly, vitronectin was also detected in the protein
corona of DOTAP/DNA lipoplexes and it was demonstrated that
vitronectin could mediate a selective uptake of the lipoplexes
Since amphiphilic dendrimers with alternating sulfonic acid
and n-propyl groups are internalized into cells and transported via
vesicles,[12] we tested the cellular uptake of the dendron to assess
whether the surface pattern of a desymmetrized dendron is similar
to the symmetric PPD3 dendrimer. Chinese Hamster ovary cell
line CHO-K1 was selected, since it possesses low CAR-
expression rendering it also suitable for gene transduction
experiments within this study. The cells were incubated with 1 µM
Cy5-dendron 9b for 24 h, washed with phosphate buffered saline
(PBS), and the nucleus was stained with Hoechst 33258. The
cellular internalization was followed by confocal laser scanning
microscopy. As depicted in Fig. 3A, Cy5-dendron 9b was
internalized into CHO-K1 cells. We observed that dendron 9b was
located in vesicles in a similar way reported for amphiphilic
PPDs.[12] The comparison with the blank control is provided in the
SI (Fig. S23).
Cell compatibility of dendron conjugates 8 and 9a was compared
to PPD3 in CHO-K1 cells by a cell viability assay. Cells were
treated with 1–40 µM dendron conjugates 8 and 9a as well as 1–
10 µM PPD3 for 24 h. We used four equivalents of dendron
conjugates compared to PPD3 to adjust the numbers of surface
patterns to approximately similar quantities. The cell viability was
determined by quantification of ATP levels applying CellTiter-
Glo®-Assay. Both dendron conjugates and PPD3 displayed no
towards MDA-MB-435S cancer cells, which have
a high
expression level of the vitronectin ανβ3 integrin receptor.[37] Taken
together, these results indicate that coating with dendron 8 favors
the interactions with specific blood proteins, which eventually also
determine the interactions with cells and cellular uptake behavior.
Figure 4. Dendron-Ad5 complex formation leads to EGFP-transduction in CAR-negative cells. (A) Coating of Ad5 by dendrons and the potential for cellular uptake
into CAR-negative cells resulting in gene transduction; (B) TEM images show binding of dendrons to Ad5 (Ad5: Dendron = 1: 100k); (C) Fluorescent microscopy
image of EGFP-transduction in CAR-negative CHO-K1 cells (Ad5: Dendron = 1: 500k); (D) Flow cytometry of CHO-K1 cells incubated with Ad5-dendron complexes
with ratios of Ad5: Dendron = 1:20k–1:500k and Ad5: PPD3 = 1:5k–1:125k respectively (n = 3, * represents p-value <0.05, ** represents p-value <0.01, ***
represents p-value <0.001, **** represents p-value <0.0001).
This article is protected by copyright. All rights reserved.

10.1002/anie.201913708
Angewandte Chemie International Edition
RESEARCH ARTICLE
significant cytotoxicity up to 20 µM and 5 µM, respectively.
Next, we studied the performance of the dendron
conjugates in Ad5-assisted gene transduction (Fig. 4A). The
formation of a dendron corona was studied by transmission
electron microscopy (TEM), dynamic light scattering (DLS) and
zeta potential, and we performed a functional assay to assess its
capability to transport Ad5 into CAR-negative cells monitored by
fluorescence microscopy. The influence of surface patterns on
was observed when infecting the cell with Cy5-dendron 9b/Ad5
(Fig. S29). We found a significantly higher gene transduction of
Ad5 when coated with four molar equivalents of propargyl-
dendron 8 and biotin-dendron 9a compared to coating with one
molar equivalent of sterically more demanding PPD3. Cy5-
labelled dendron 9b showed lower gene transduction efficiency
than dendrons 8 and 9a, which could be due to the attachment of
the fluorophore that might also influence Ad5 binding and its
gene transduction was compared with PPD3 using flow cytometry. cellular uptake. These results indicate that four molar equivalents
Consequently, dendron 8, 9a and 9b as well as PPD3 were mixed
with Ad5 at the ratios of 1:20k to 1:500k (Ad5:dendron) and 1:5k
to 1:125k (Ad5:PPD3), respectively, in PBS and complex
formation as well as gene transduction of Ad5 in low CAR
expressing cell line CHO-K1 were tested. For all experiments, four
equivalents of dendron related to PPD3 were used to compare
their properties while maintaining approximately the same amount
of surface patterns. Here we used an eGFP-expressing Ad5 and
analyzed its interaction with the dendrons. TEM analysis of vector
particles incubated with the dendron for 40 min at a molar ratio of
1:100K was performed. Results clearly confirmed that dendrons
8, 9a and 9b bound to and formed complexes with Ad5 (Fig. 4B,
Fig. S24). To further analyze this interaction, DLS measurements
were conducted. Our data shows that dendrons alone self-
assemble in solution due to their intrinsic amphiphilicity (Table S5
and Fig. S26B). Nevertheless, in presence of Ad5, all dendrons
demonstrably bound to the vector particles, which was clearly
indicated by an increase of the vector particle size (Table S3,
Table S4, Fig. S25 and Fig. S26A). To further confirm this
interaction, we measured the surface charge of dendrons alone,
Ad5 alone and dendron-coated Ad5 by zeta potential analysis.
Dendron-coated vector particles showed a ratio-dependent
increase of the negative surface charge compared to dendrons
alone or Ad5 particles alone (Table S3, Table S4, Table S5). By
this means, we ancillary confirmed the binding of Ad5 and
analyzed dendrons. Additionally, we could show that at ratios
beyond 1:1000K for Ad5/propargyl-dendron 8 and 1:200K for
Ad5/biotin-dendron 9a, saturation of Ad5 particles is reached,
which results in free, unbound dendron molecules in solution
(Table S4, Fig S26).
of dendrons 8 and 9a increase Ad5 transport into these CAR-
negative cells compared to one molar equivalent of PPD3. As
reported for amphiphilic PPDs[19], distinct amphiphilic surface
patterns of the dendrons are crucial for biorecognition of Ad5.
Even though the amphiphilic dendrons only represent one quarter
of the full dendrimer PPD3, they retain both Ad5 binding capacity
and gene transduction into CAR-negative cells, while providing a
second functionality for post modifications (Fig. 4A, Fig. 5A).
In the next step, we studied whether the functionality at the focal
point of the dendron was still accessible after complexation with
Ad5. Consequently, we covalently modified the dendrons in situ
By biolayer interferometry (BLI) we previously studied the
binding strength of an amphiphilic dendrimer binding motif to Ad5
[19]. This method required the immobilization of the dendrimer on
a streptavidin-coated surface (SI, Fig. S32). Thus, we used biotin-
dendron 9a and an equilibrium dissociation constant value of KD
= 1.27 pM was determined (SI, Fig. S33 and Table S6). Since the
dendron binds to Ad5 in a multivalent way, this value does not
Figure 5. CuAAC on dendrons bound to Ad5. (A) CuAAC of propargyl-dendron
8 with 3-azido-7-hydroxycoumarin on the Ad5 surface leads to fluorescence of
the dendron-fluorophore conjugate (λabs= 404 nm, λem= 477 nm after click
reaction). (B) Fluorescence spectra of dendron conjugates after treatment with
CuAAC reagents. Ad5 was mixed with propargyl-dendron 8 or biotin-dendron
9a (negative control). After incubation for 40 min, unbound dendron was
removed by ultrafiltration. Propargyl-dendron 8 alone was treated under same
conditions. Then, CuAAC reagents were added and fluorescence spectra were
recorded after incubation for 1 h. 3-azido-7-hydroxycoumarin was subtracted as
background. (C) The change in relative fluorescence of Ad5+Biotin-D 9a group,
Propargyl-D 8 group (with ultrafiltration) and Ad5+Biotin-D 9a group at 477nm
(emission) after treatment are described in (B). 3-azido-7-hydroxycoumarin was
subtracted as background and the change in relative fluorescence of
Ad5/propargyl-dendron after CuAAC (blue column) is relative to the Ad5+Biotin-
D 9a group (black column) that serves as negative control (n=3, * represents
p<0.05, ns means not significant).
present a single binding event of the dendron to Ad5, but point
[19]
towards strong interactions between Ad5 and dendron.
We
observed green fluorescence for all cells treated with
Ad5/dendron complexes by fluorescence microscopy whereas
Ad5 alone led to low gene transduction (Fig. 4C and Fig. S27).
Therefore, all dendron conjugates as well as the control
dendrimer PPD3 bound to Ad5, and they transported Ad5 into
cells via
a CAR-independent pathway leading to EGFP-
expression. We quantified the gene transduction of Ad5 in low
CAR cell line CHO-K1 via flow cytometry by measuring the
fluorescence intensity of EGFP. We found that gene transduction
was enhanced when increasing the molar ratio of dendron to Ad5
(Fig.4D and Fig. S28). A double-positive signal of EGFP and Cy5
This article is protected by copyright. All rights reserved.

10.1002/anie.201913708
Angewandte Chemie International Edition
RESEARCH ARTICLE
by CuAAC on the Ad5 surface. The reaction was performed using
propargyl-dendron 8 with 3-azido-7-hydroxycoumarin that is
known to become highly fluorescent when forming a 1,2,3-triazole
product (Fig. 5A).[38] As negative control, we used biotin-dendron
9a attached to Ad5 that cannot undergo a CuAAC. We first
investigated if the CuAAC proceeded in the micromolar range as
well as the necessity of a ligand like TBTA to stabilize the
copper(I) species[39]. After 1 h of incubation of propargyl-dendron
8 and 3-azido-7-hydroxycoumarin with click reagents at a
concentration of 10 μM dendron, we only found high fluorescence
for the TBTA treated sample. This indicates the importance of
adding a ligand like TBTA to the reaction mixture (Fig. S30).
Additionally, it is reported that this class of ligands protect
biomolecules from reactive oxygen species (ROS) during the
ligation.[40] For CuAAC on the Ad5 surface, we used the water-
soluble (4-{[bis-(1-tert-butyl-1H-[1,2,3]triazol-4-ylmethyl)-amino]-
methyl}-[1,2,3]triazol-1-yl)-acetic acid (BTTAA), which is even
more efficient in aqueous solutions than TBTA.[41] Briefly,
propargyl-dendron 8 was incubated with Ad5 for 1 h, and unbound
dendrons were removed by ultrafiltration. Subsequently, the
CuAAC reagents were added. The negative control, biotin-
dendron 9a coated Ad5, was treated under the same conditions.
In addition, ultrafiltered propargyl-dendron 8 without Ad5 was
prepared and click reagents were added to the supernatant to
demonstrate that free dendrons could be removed by
ultrafiltration. After incubation for 1 h, fluorescence spectra were
recorded (Fig. 5B). Since free 3-azido-7-hydroxycoumarin is
slightly fluorescent, we subtracted it as background. We observed
a 21-fold increase in relative fluorescence intensity at 477 nm
compared to the biotin-dendron coated Ad5 that served as
negative control (Fig. 5C). These results indicate successful
CuAAC at the focal points of the propargyl-dendron 8 after
formation of Ad5/dendron complexes. Therefore, we demonstrate
that the functionalities of the dendron core are still accessible for
post-modifications even after complexation with Ad5. This model
reaction proves that our structural dendron concept represents a
promising tool for future applications in terms of in situ attachment
of cell targeting groups or drug molecules.
By desymmetrization of the dendrimer structure, we opened up a
new platform for introducing bioactive groups to the Ad5 surface
without the necessity to covalently modify the virus particles.
These reactive groups were well accessible at the Ad5 surface as
shown by CuAAC reactions. Moreover, this new concept of
forming a supramolecular dendron corona at virus surfaces opens
up exciting opportunities for attaching e.g. cell targeting groups or
drug molecules and paves the way to rational control of Ad5
biodistribution to ultimately improve its capacity in virus-assisted
gene therapy.
Acknowledgements
T.W and K.M. thank the Volkswagen Foundation (Project
No.88396) for financial support. The project is supported by the
Deutsche Forschungsgemeinschaft (DFG, German Research
Foundation) under project number 213555243 – SFB 1066. J.W.
is a recipient of a DFG-funded position through the Excellence
Initiative by the Graduate School Materials Science in Mainz
(GSC 266). We thank Sarah Chagri for technical assistance,
Manfred Wagner and Stefan Spang for NMR measurements as
well as Hans-Joachim Räder, Jutta Schnee and Stephan Türk for
MALDI-TOF MS measurements. Y.W. and L.L. thank the
Analytical and Testing Center of Huazhong University of Science
and Technology (HUST) and the Instrument Center at College of
Life Science and Technology of HUST for instrument support.
Support by the IMB Microscopy Core Facility is gratefully
acknowledged. We would like to thank Meiyu Gai and Katja Klein
for the synthesis and characterization of the liposomes and
polystyrene nanoparticles. Open Access charges were covered
by the Max Planck Society.
Keywords: Adenovirus-dendron complex • adenovirus coating •
amphiphilic surfaces • protein corona • gene delivery
[1]
a) R. Lipowsky, Nature 1991, 349, 475-481; b) D. Eisenberg, W. Wilcox,
A. D. McLachlan, J. Cell. Biochem. 1986, 31, 11-17; c) A. Dehsorkhi, V.
Castelletto, I. W. Hamley, J. Pept. Sci. 2014, 20, 453-467.
S. Zhang, Biotechnol. Adv. 2002, 20, 321-339.
[2]
[3]
Conclusion
a) T. Cedervall, I. Lynch, S. Lindman, T. Berggård, E. Thulin, H. Nilsson,
K. A. Dawson, S. Linse, Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 2050-
2055; b) J. Müller, K. N. Bauer, D. Prozeller, J. Simon, V. Mailänder, F.
R. Wurm, S. Winzen, K. Landfester, Biomaterials 2017, 115, 1-8.
a) S. Tenzer, D. Docter, J. Kuharev, A. Musyanovych, V. Fetz, R. Hecht,
F. Schlenk, D. Fischer, K. Kiouptsi, C. Reinhardt, K. Landfester, H. Schild,
M. Maskos, S. K. Knauer, R. H. Stauber, Nat. Nanotechnol. 2013, 8, 772-
781; b) S. Ritz, S. Schöttler, N. Kotman, G. Baier, A. Musyanovych, J.
Kuharev, K. Landfester, H. Schild, O. Jahn, S. Tenzer, V. Mailänder,
Biomacromolecules 2015, 16, 1311-1321.
Ad5 is a common vector in gene therapy but its clinical usage
has limitations due to mistargeting of plasma protein-coated Ad5
and acute toxicity. We present the synthesis of amphiphilic
polyphenylene dendrons that bind to the surface of Ad5 by their
polar and unpolar surface groups and facilitate transport of the
Ad5/dendron complexes into CAR-negative cells. In this way,
these dendrons maintained the crucial biorecognition features of
the full dendrimer PPD3. As the dendrons form a new, outer layer
at the Ad5 surface, their interaction with blood plasma proteins
might be crucial for future applications. Dendron-coated
liposomes were found to interact with specific proteins of the
blood serum and plasma proteins such as vitronectin and ApoH,
which could promote uptake into cancer and mesenchymal stem
cells. In contrast to dendrimer PPD3, the dendrons provide an
[4]
[5]
[6]
[7]
[8]
S. Schöttler, G. Becker, S. Winzen, T. Steinbach, K. Mohr, K. Landfester,
V. Mailänder, F. R. Wurm, Nat. Nanotechnol. 2016, 11, 372-377.
Y. T. Ho, N. A. Azman, F. W. Y. Loh, G. K. T. Ong, G. Engudar, S. A.
Kriz, J. C. Y. Kah, Bioconjugate Chem. 2018, 29, 3923-3934.
M. P. Monopoli, C. Åberg, A. Salvati, K. A. Dawson, Nat. Nanotechnol.
2012, 7, 779-786.
a) O. Uzun, Y. Hu, A. Verma, S. Chen, A. Centrone, F. Stellacci, Chem.
Comm. 2008, 196-198; b) A. Verma, O. Uzun, Y. Hu, Y. Hu, H.-S. Han,
N. Watson, S. Chen, D. J. Irvine, F. Stellacci, Nat. Mater. 2008, 7, 588-
595.
additional functionality for post-modifications such as
a
fluorophore for imaging and D-biotin as a bio-orthogonal group.
This article is protected by copyright. All rights reserved.

10.1002/anie.201913708
Angewandte Chemie International Edition
RESEARCH ARTICLE
[9]
L. Guerrini, R. A. Alvarez-Puebla, N. Pazos-Perez, Materials (Basel)
2018, 11, 1154.
[37] G. Caracciolo, F. Cardarelli, D. Pozzi, F. Salomone, G. Maccari, G. Bardi,
A. L. Capriotti, C. Cavaliere, M. Papi, A. Laganꢀ, ACS applied materials
& interfaces 2013, 5, 13171-13179.
[10] B. K. Nanjwade, H. M. Bechra, G. K. Derkar, F. V. Manvi, V. K. Nanjwade,
Eur. J. Pharm. Sci. 2009, 38, 185-196.
[38] K. Sivakumar, F. Xie, B. M. Cash, S. Long, H. N. Barnhill, Q. Wang, Org.
Lett. 2004, 6, 4603-4606.
[11] a) R. Esfand, D. A. Tomalia, Drug Discov. Today 2001, 6, 427-436; b) S.
Svenson, D. A. Tomalia, Adv. Drug Deliver. Rev. 2012, 64, Supplement,
102-115.
[39] V. O. Rodionov, S. I. Presolski, S. Gardinier, Y.-H. Lim, M. G. Finn, J.
Am. Chem. Soc. 2007, 129, 12696-12704.
[12] R. Stangenberg, Y. Wu, J. Hedrich, D. Kurzbach, D. Wehner, G.
Weidinger, S. L. Kuan, M. I. Jansen, F. Jelezko, H. J. Luhmann, D.
Hinderberger, T. Weil, K. Müllen, Adv. Healthcare Mater. 2015, 4, 377-
384.
[40] V. Hong, S. I. Presolski, C. Ma, M. G. Finn, Angew. Chem., Int. Ed. 2009,
48, 9879-9883.
[41] a) C. Besanceney-Webler, H. Jiang, T. Zheng, L. Feng, D. Soriano del
Amo, W. Wang, L. M. Klivansky, F. L. Marlow, Y. Liu, P. Wu, Angew.
Chem., Int. Ed. 2011, 50, 8051-8056; b) C. Uttamapinant, A.
Tangpeerachaikul, S. Grecian, S. Clarke, U. Singh, P. Slade, K. R. Gee,
A. Y. Ting, Angew. Chem., Int. Ed. 2012, 51, 5852-5856.
[13] Z. Mhlwatika, B. A. Aderibigbe, Molecules 2018, 23, 2205.
[14] J. Yang, Q. Zhang, H. Chang, Y. Cheng, Chem. Rev. 2015, 115, 5274-
5300.
[15] a) M. A. Kostiainen, O. Kasyutich, J. J. L. M. Cornelissen, R. J. M. Nolte,
Nat. Chem. 2010, 2, 394-399; b) G. Doni, M. A. Kostiainen, A. Danani,
G. M. Pavan, Nano Lett. 2011, 11, 723-728.
[16] R. Stangenberg, I. Saeed, S. L. Kuan, M. Baumgarten, T. Weil, M.
Klapper, K. Müllen, Macromol. Rapid Commun. 2014, 35, 152-160.
[17] B. A. G. Hammer, K. Müllen, J. Nanopart. Res. 2018, 20, 262.
[18] P. Carbone, F. Negri, F. Müller-Plathe, Macromolecules 2007, 40, 7044-
7055.
[19] Y. Wu, L. Li, L. Frank, J. Wagner, P. Andreozzi, B. Hammer, M.
D’Alicarnasso, M. Pelliccia, W. Liu, S. Chakrabortty, S. Krol, J. Simon, K.
Landfester, S. L. Kuan, F. Stellacci, K. Müllen, F. Kreppel, T. Weil, ACS
Nano 2019, 13, 8749-8759.
[20] a) W. S. M. Wold, K. Toth, Curr. Gene Ther. 2013, 13, 421-433; b) R. G.
Crystal, Hum. Gene Ther. 2014, 25, 3-11.
[21] C. S. Lee, E. S. Bishop, R. Zhang, X. Yu, E. M. Farina, S. Yan, C. Zhao,
Z. Zheng, Y. Shu, X. Wu, J. Lei, Y. Li, W. Zhang, C. Yang, K. Wu, Y. Wu,
S. Ho, A. Athiviraham, M. J. Lee, J. M. Wolf, R. R. Reid, T.-C. He, Genes
Dis. 2017, 4, 43-63.
[22] J. N. Glasgow, M. Everts, D. T. Curiel, Cancer Gene Ther. 2006, 13, 830-
844.
[23] a) C. R. O'Riordan, A. Lachapelle, C. Delgado, V. Parkes, S. C.
Wadsworth, A. E. Smith, G. E. Francis, Hum. Gene Ther. 1999, 10, 1349-
1358; b) M. A. Croyle, Q.-C. Yu, J. M. Wilson, Hum. Gene Ther. 2000,
11, 1713-1722.
[24] Y. Wu, Anthony P. West, Helen J. Kim, Matthew E. Thornton, Andrew B.
Ward, Pamela J. Bjorkman, Cell Rep. 2013, 5, 1443-1455.
[25] S. J. Everse, G. Spraggon, L. Veerapandian, M. Riley, R. F. Doolittle,
Biochemistry 1998, 37, 8637-8642.
[26] J. Park, M. S. Kim, D. H. Shin, 2018, doi: 10.2210/pdb5Z0B/pdb.
[27] U. M. Wiesler, A. J. Berresheim, F. Morgenroth, G. Lieser, K. Müllen,
Macromolecules 2001, 34, 187-199.
[28] F. Morgenroth, Chem. Comm. 1998, 1139-1140.
[29] G. Mihov, I. Scheppelmann, K. Müllen, J. Org. Chem. 2004, 69, 8029-
8037.
[30] A. H. Baker, J. H. McVey, S. N. Waddington, N. C. Di Paolo, D. M.
Shayakhmetov, Mol. Ther. 2007, 15, 1410-1416.
[31] a) C. Weber, M. Voigt, J. Simon, A.-K. Danner, H. Frey, V. Mailänder, M.
Helm, S. Morsbach, K. Landfester, Biomacromolecules 2019, 20, 2989-
2999; b) M. Kokkinopoulou, J. Simon, K. Landfester, V. Mailänder, I.
Lieberwirth, Nanoscale 2017, 9, 8858-8870.
[32] M. Okuno, M. Mezger, R. Stangenberg, M. Baumgarten, K. Müllen, M.
Bonn, E. H. G. Backus, Langmuir 2015, 31, 1980-1987.
[33] M. Kokkinopoulou, J. Simon, K. Landfester, V. Mailander, I. Lieberwirth,
Nanoscale 2017, 9, 8858-8870.
[34] J. Simon, T. Wolf, K. Klein, K. Landfester, F. R. Wurm, V. Mailänder,
Angew. Chem., Int. Ed. 2018, 57, 5548-5553.
[35] D. E. Owens, N. A. Peppas, Int. J. Pharm. 2006, 307, 93-102.
[36] a) S. Schöttler, K. Klein, K. Landfester, V. Mailänder, Nanoscale 2016, 8,
5526-5536; b) V. Mirshafiee, R. Kim, M. Mahmoudi, M. L. Kraft, Int. J.
Biochem. Cell Biol. 2016, 75, 188-195.
This article is protected by copyright. All rights reserved.

10.1002/anie.201913708
Angewandte Chemie International Edition
RESEARCH ARTICLE
RESEARCH ARTICLE
Jessica Wagner, Longjie Li, Johanna
Simon, Lea Krutzke, Katharina
Landfester, Volker Mailänder, Klaus
Müllen, David Y.W. Ng*, Yuzhou Wu*
and Tanja Weil*
Page No. – Page No.
Amphiphilic polyphenylene dendrons were prepared that recognize gene vector
adenovirus 5 (Ad5) due to polar and unpolar surface groups that also control its
cellular uptake. The new dendron layer at the Ad5 surface provides reactive groups
accessible for post-modifications at the virus surface.
Amphiphilic Polyphenylene Dendron
Conjugates for Surface Remodeling
of Adenovirus 5
This article is protected by copyright. All rights reserved.