Efficient Innate Immune Killing of Cancer Cells Triggered by Cell-Surface Anchoring of Multivalent Antibody-Recruiting Polymers

Article abstract of DOI:10.1002/anie.201905093

Binding of monoclonal antibodies (mAbs) onto a cell surface triggers antibody-mediated effector killing by innate immune cells through complement activation. As an alternative to mAbs, synthetic systems that can recruit endogenous antibodies from the blood stream to a cancer cell surface could be of great relevance. Herein, we explore antibody-recruiting polymers (ARPs) as a novel class of immunotherapy. ARPs consist of a cell-binding motif linked to a polymer that contains multiple small molecule antibody-binding motifs along its backbone. As a proof of concept, we employ a lipid anchor that inserts into the phospholipid cell membrane and make use of a polymeric activated ester scaffold onto which we substitute dinitrophenol as an antibody-binding motif. We demonstrate that ARPs allow for high avidity antibody binding and drive antibody recruitment to treated cells for several days. Furthermore, we show that ARP-treated cancer cells are prone to antibody-mediated killing through phagocytosis by macrophages.

Full text of DOI:10.1002/anie.201905093

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Efficient innate immune killing of cancer cells triggered by cell
surface anchoring of multivalent antibody-recruiting polymers
Authors: Annemiek Uvyn, Ruben De Coen, Mandy Gruijs, Cees Tuk,
Jana De Vrieze, Marjolein van Egmond, and Bruno de Geest
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.201905093
Angew. Chem. 10.1002/ange.201905093
Link to VoR: http://dx.doi.org/10.1002/anie.201905093
http://dx.doi.org/10.1002/ange.201905093

10.1002/anie.201905093
Angewandte Chemie International Edition
COMMUNICATION
Efficient innate immune killing of cancer cells triggered by cell
surface anchoring of multivalent antibody-recruiting polymers
Annemiek Uvyn,# [a] Ruben De Coen,# [a] Mandy Gruijs, [b] Cees W. Tuk, [b] Jana De Vrieze,[a] Marjolein
van Egmond, [b] Bruno G. De Geest*[a]
Abstract: Rewiring innate immune defence to eradicate cancer is one
of the mechanisms of action of several monoclonal antibodies (mAbs)
used in the clinic. Binding of mAbs onto a cell surface triggers
antibody-mediated effector killing by innate immune cells and through
complement activation. As an alternative to mAbs, which face inherent
issues with costs of production and stability, synthetic systems that
can recruit endogenous antibodies from the blood stream to a cancer
cell surface could be of great relevance. Here we explore antibody-
recruiting polymers (ARPs) as a novel class of immunotherapy. ARPs
consist of a cell binding motif linked to a polymer that contains multiple
small molecule antibody-binding motifs along its backbone. As a proof
of concept, we employ a lipid anchor that inserts into the phospholipid
cell membrane and make use of a polymeric activated ester scaffold
onto which we substitute dinitrophenol as antibody-binding motif. We
demonstrate that ARPs allow for high avidity antibody binding and are
able to drive antibody recruitment to treated cells for up to several
days. Further we show that ARP-treated cancer cells are prone to
antibody-mediated killing of cancer cells through phagocytosis by
macrophages.
questions are (1) which Abs to exploit and (2) how to recruit these
to a target cell surface. The first question has been addressed by
several groups and mainly focuses on the presence of antibodies
in
human
blood
against
galactose--1,3-galactose,
phosphorylcholine, rhamnose or dinitrophenol (DNP).[8] The
nature of these endogenous antibodies varies from the
consumption of meat (galactose--1,3-galactose), commensal
bacteria in the gut (phosphorylcholine and rhamnose) and
exposure to pesticides (DNP). The second question relates to
antibody recruitment, which involves the process that drives
binding of endogenous antibodies to a target cell surface. In this
regard, hydrophobic interaction, which typically relies on a lipid
motif conjugated to an antibody-binding motif, has been
explored.[9] The lipid part anchors the cell membrane and exposes
the antibody-binding motif to the external medium. This approach
is intended for local application, i.e. by direct intratumoral injection,
due to the lack in selectivity towards cancer cells. Another
strategy involves conjugates composed of an antibody-binding
motif and a target cell binding motif. For the latter, several cell
surface receptors that are overexpressed by cancer cells have
been explored, including the folate receptor,[10] PSMA,[11]
uPAR,[12] and others. These approaches all involve mono- or at
best bi-valent molecules, both on the level of the antibody-binding
as well as the target cell binding motifs.
Exploiting innate immune effector mechanisms against the Fc
domain of monoclonal antibodies (Abs) is gaining immense
popularity in anti-cancer therapy and also holds promise for the
treatment of infectious diseases.[1][2] Its mechanism of action relies
on clustering of Abs onto a target (cancer) cell surface by binding
with its Fab fragments to a protein expressed on the target cell
surface. Subsequent exposure of the Fc domain[3] towards the
external medium flags the cell for destruction by a combination of
complement activation, NK-cell mediated killing and phagocytosis
by macrophages.[4-6] The latter two mechanisms directly depend
on recognition of Fc domains on the target cell surface by Fc
receptors on the innate effector cells.[3]
As an alternative to the use of externally administered monoclonal
Abs, exploiting the presence of endogenous antibodies that are
present in the serum of every human being,[7] could be beneficial
as it circumvents the need for recombinant production and genetic
engineering to make the Ab compatible with the host as well as
avoiding stability issues often encountered with long term storage
of mAbs. In the context of using endogenous Abs, the main
The question we aim to address in this paper is whether polymeric
scaffolds that contain multiple motifs of an antibody-recruiting
motif allow for a more efficient antibody recruitment to the cell
surface with prolonged cell surface persistence and efficient
induction of innate effector killing. Hereto we explore the concept
of so-called antibody-recruiting polymers (ARPs). These consist
of multiple copies of on the one hand motifs that bind to
endogenous antibodies and on the other hand a motif that
anchors to the cell surface, thereby tagging the cell for attack by
the innate immune system (Scheme 1).
[a]
[b]
#
Annemiek Uvyn, Dr. Ruben De Coen, Jana De Vrieze,Prof.Dr.
Bruno G. De Geest
Department of Pharmaceutics, Ghent University, Belgium
Mandy Gruijs, Cees W. Tuk, Prof.Dr. Marjolein van Egmond
Department of Molecular Cell Biology and Immunology, Amsterdam
UMC, location VUmc, Amsterdam, the Netherlands.
These authors contributed equally to the work.
Scheme 1 Antibody-recruiting polymer (ARP) concept. ARPs anchor to
the cell surface by hydrophobic insertion of a lipid tail at the ARP chain end into
the phospholipid cell membrane. Cell surface anchoring of ARPs triggers
recruitment of endogenous antibodies to the cell, thereby flagging the cell for
destruction by innate immune effector mechanisms.
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.201905093
Angewandte Chemie International Edition
COMMUNICATION
Scheme 2 Antibody-recruiting polymer (ARP) synthesis. A functionalized (i.e. biotin, cholesterol, monoalkyl- or dialkyl-lipid) chain transfer agent is used for
RAFT polymerization of pentafluorophenyl acrylate followed by removal of the trithiocarbonate RAFT end group by treatment with an excess of ACVA. The polymer
backbone is subsequently substituted with DNP antibody-recruiting motifs and 2-aminoethanol to render the polymer backbone hydrophilic. For fluorescent labeling
rhodamine cadaverine is used. Control polymers are synthesized using a non-functionalized chain transfer agent (HO*) and/or omitting DNP substitution.
Whereas polymeric constructs that contain multiple hapten copies
have been reported earlier,[13-14] it remains elusive whether these
are capable to efficiently recruit endogenous antibodies to a
cancer cell surface and trigger innate immune killing.
Subsequently, biotin-polyPFPA was substituted with an amine-
functionalized dinitrophenol (DNP) derivative (Scheme 2),
targeting a degree of substitution (DS; amount of DNP motifs per
100 PFPA repeating units) of respectively 2, 5 and 10. To
generate a hydrophilic polymer backbone, all unreacted PFP-
esters were converted into hydrophilic repeating units by
treatment of the polymer with an excess of 2-aminoethanol. UV-
Vis analysis of the resulting polymers after purification revealed
that the DNP content was in good agreement with the targeted
DS. For comparison, we also synthesized a biotin-polymer
without DNP and a monovalent biotin-DNP compound by
conjugating biotin to DNP via a diethylene glycol linker (cfr.
Supporting Information for the synthesis route). Note that the
reason for choosing DNP as antibody-recruiting motif in the
context of this work was fostered by the commercial availability of
anti-DNP antibodies, which strongly facilitates experimental
readout compared to the use of full human serum as a source of
endogenous antibodies.
Biolayer interferometry (BLI) was used to investigate the influence
of multiple DNP motifs on the avidity of the antibody binding.
Hereto, streptavidin-coated sensors were loaded with the
monovalent biotin-DNP conjugate to test whether the chemistry
used for conjugation still allows for antibody binding. When
confirmed this was indeed the case, we also used this system to
elucidate an optimal concentration window for sensor loading and
subsequent kinetic analysis (Figure S7) of anti-DNP binding,
which was then used to measure the avidity of (polyclonal rabbit)
anti-DNP binding to DNP-polymers. Figure 1 shows sensorgrams
of anti-DNP binding to DNP-polymer-functionalized sensors.
These data demonstrate low non-specific antibody binding to
control sensors loaded with control polymers that did not contain
DNP. By contrast,
Here we focus on ARPs bearing a lipid moiety at one of the chain
ends that anchors to cell membranes through hydrophobic
interaction. The therapeutic relevance of such ARPs lies within
context of intratumoral injection where one aims to engineer the
tumor microenvironment to become infiltrated with innate immune
cells that induce effector killing. The thus created tumor debris can
subsequently serve as a tumor antigen pool which offers
opportunities to mount tumor-specific adaptive immunity to fight
distal tumors as well as metastases. Of note, intratumoral
administration also reduces the risks of unwanted systemic side
reaction due to potential immunogenicity of multiple hapten
copies onto a polymer backbone.
To start, we focused on how to optimally design ARPs for antibody
recruitment. Hereto, pentafluorophenyl acrylate (PFPA)[15] was
polymerized by reversible addition-fragmentation chain transfer
(RAFT),[16] targeting a final degree of polymerization (DP) of 100,
followed by removal of the trithiocarbonate RAFT end-group by
treatment with an excess of 4,4-azobis(4-cyanovaleric acid)
(ACVA). For immobilization purpose (vide infra), a biotin-
functionalized chain transfer agent (CTA) was used. Hereto, biotin
was conjugated to 2,2′-(ethylenedioxy)diethylamine followed by
conjugation to 2-propanoic acid butyl trithiocarbonate (PABTC).
The synthesis route of the polymers is depicted in Scheme 2 and
the polymer properties are summarized in Table 1, showing
narrow dispersity and thus good control over RAFT
polymerization. Experimental characterization data are depicted
in the Supporting Information section.
This article is protected by copyright. All rights reserved.

10.1002/anie.201905093
Angewandte Chemie International Edition
COMMUNICATION
Table 1 Characteristics of the synthesized polymers.
Polymer
DPtheo
110
Conv.calc., a [%]
DPcalc., b
98
Mntheo, b [Da]
23 800
Mnc [Da]
15 200
13 200
12 500
17 600
16 300
Mwc [Da]
20 100
15 400
15 900
20 400
18 100
ᴆ
biotin-polyPFPA
polyPFPA
89
85
93
87
90
1.32
1.18
1.27
1.16
1.11
100
85
20 400
monoalkyl-polyPFPA
dialkyl-polyPFPA
cholesterol-polyPFPA
120
112
87
27 200
100
21 400
120
108
26 300
a: conversion determined by 19F-NMR (282 MHz; CDCl3).
b: based on conversion and molecular weight of monomer and CTA.
c: determined by SEC (THF)
DNP-polymers showed strong adsorption and low desorption of
anti-DNP. Curve fitting and calculation of the binding avidity KD
(Table 2), reveals a dramatic increase in binding avidity with
increasing DS of DNP on the polymer backbone. Notably, more
than a 4-log increase in KD of a DS 10 DNP-polymer compared to
monovalent DNP is observed. This can likely be attributed to
multivalent binding,[17] as once an antibody dissociates from one
DNP motif, it can rapidly bind to a neighboring DNP motif located
on the same or adjacent polymer backbone.
These findings prompted us to design ARPs based on a DP 100
polyPFPA backbone substituted with a DS 10 of DNP. As a proof-
of-concept, we elaborated on ARPs that can anchor to the cell
surface via hydrophobic insertion into the phospholipid cell
membrane, inspired by earlier studies that macromolecular
amphiphiles containing a lipid motif at their chain end are prone
to spontaneous cell membrane anchoring.[18,19] Hereto we
synthesized 3 different RAFT CTAs bearing respectively a
monoalkyl, dialkyl and cholesterol motif (Scheme 2). In brief, a
cholesterol-CTA was synthesized by simple esterification of
cholesterol and PABTC. A mono-alkyl-CTA was synthesized by
conjugation of stearyl chloride to aminopropanol followed by
conjugation to PABTC. A dialkyl-CTA was synthesized by
conjugating tetradecylamine to aspartic acid followed by
conjugation to PABTC. Note that the choice for a 14 carbon alkyl
chain was fostered by earlier reports suggesting that these show
improved cell membrane anchoring relative to longer 18 carbon
lipid tail based structures.[19] Lipid-CTAs were subsequently used
for RAFT polymerization of polyPFPA, followed by DNP
modification as detailed above for biotinylated polymers. Note that
following trithiocarbonate end-group removal, the polymer was
fluorescently labeled with tetramethylrhodamine-cadaverine (DS
1) to allow for detection by fluorescence-based techniques. The
resulting polymer properties are summarized in Table 1, while
experimental characterization is shown in the Supporting
Information section. In all cases well defined narrowly dispersed
polymers were obtained, pointing at good control over RAFT
polymerization using these lipid-functionalized CTAs.
Figure 1 Multivalency effect on antibody binding to ARPs. Biolayer
interferometry sensorgrams of (polyclonal rabbit) anti-DNP binding to
streptavidin-coated sensors functionalized with biotinylated polyDNP. A blank
biotinylated polymer was used as a control to assess non-specific binding. A
dilution series of 22.5, 11.25, 5.6, 2.8 and 0 nM of anti-DNP was measured to
calculate KD values. Sensors were dipped into the anti-DNP solution at the 500
s time point, which marks the onset of adsorption. At the 1125 s time point
(dotted line), sensors were dipped into buffer, which marks the onset of
desorption.
Table 2 Calculated KD values and KD error measured by biolayer
interferometry (BLI).
Ligand
KD (M)
KD error (M)
3.18·10.11
2.73·10-12
2.54·10-12
< 1.00·10-12
biotin-DNP
1.12·10-8
1.23·10-10
6.80·10-11
< 1.00·10-12
biotin-polyDNP DS2
biotin-polyDNP DS5
biotin-polyDNP DS10
To investigate the influence of the lipid tail on cell membrane
anchoring, CT26 mouse colon cancer cells were pulsed in vitro
with the lipid-polymer conjugates for 2h at 37 °C, followed by flow
This article is protected by copyright. All rights reserved.

10.1002/anie.201905093
Angewandte Chemie International Edition
COMMUNICATION
cytometry and confocal microscopy analysis. Note that in these
experiments, the sample concentration was slightly adjusted
based on the fluorescence emission intensity to ensure that the
cells received equal doses of fluorescence as the latter parameter
was used for quantification purpose. The mean fluorescence
intensity (MFI; Figure 2A) values measured by flow cytometry
indicated that all lipid-polymer conjugates anchor to cells, unlike
the control polymer that lacks a lipid tail. These findings are
confirmed by confocal microscopy showing the localization of the
fluorescently labeled lipid-polymer conjugates at the cell surface.
(Figure 2B) Interestingly, the dialkyl-polymer conjugate shows
the highest extent of cellular association. Next, we investigated
the ability of the lipid-polymer conjugates to recruit anti-DNP
antibodies to the cell surface. Hereto, polymer-treated cells were
washed to remove unbound polymer, and subsequently pulsed
with AF488-labeled (polyclonal rabbit) anti-DNP followed by flow
cytometry (Figure 2C) and confocal microscopy (Figure 2B)
analysis. These data unambiguously demonstrate that only
conjugates that contain both a lipid tail and DNP motifs are able
to recruit anti-DNP to the cell surface. In accordance to the
binding of the lipid-polymer conjugate itself, the dialkyl-containing
construct is the most proficient in antibody recruitment.
Importantly, control polymers that do not contain DNP motifs
induce low non-specific binding in the same range as the
autofluorescence of the cell. Furthermore, as shown in Figure 2D,
a monovalent DNP-dialkyl lipid control conjugate (cfr. Scheme S1
in supporting information for the chemical structure) had very poor
anti-DNP recruiting capacity. This could be attributed to the DNP
motif being buried too deep into the glycocalyx to allow for the
anti-DNP to bind its target epitope, and because of the much
lower avidity of anti-DNP towards monovalent DNP-constructs
relative to multivalent DNP-constructs as evidenced by biolayer
interferometry (vide supra).
In analogy with earlier reports by the Bertozzi group,[19] both
dialkyl- and cholesterol-based ARP conjugates show efficient cell
membrane anchoring, which is translated into efficient antibody
Figure 2Antibody recruitment to ARP-treated cells. (A) Binding of ARPs through lipid-mediated anchoring to CT26 cells.(n=3; 1 way Anova **** : p<0,0001;
*** : p<0,001; ** : p<0,0001) (B) Confocal microscopy images of cells treated with ARPs (red color) and anti-DNP (green color). Scale bar represents 10 m. (C)
Antibody (anti-DNP) recruitment to ARP-treated CT26 cells. Left panel shows a representative flow cytometry histogram of the dialkyl-polyDNP ARP with controls.
Right panel compares all ARPs for their ability to recruit anti-DNP antibody to the surface of treated cells. (n=3; 1 way Anova **** : p<0,0001; *** : p<0,001; ** :
p<0,0001) (D) Flow cytometry analysis comparing antibody recruiting capacity between a monovalent dialkyl-DNP conjugate and dialkyl-polyDNP. (n=3; 1 way
Anova **** : p<0,0001) (E) Flow cytometry analysis of cell surface persistence of ARPs. Cells were treated with ARP and pulsed with anti-DNP at different time
points. (n=3; t test **** p<0,0001; *** : p<0,001). Note that in all experiments AF488-labeled (polyclonal rabbit) anti-DNP was used.
This article is protected by copyright. All rights reserved.

10.1002/anie.201905093
Angewandte Chemie International Edition
COMMUNICATION
recruitment. Monoalkyl lipids have been reported to exhibit less
efficient cell membrane anchoring,[20] likely due to their lower
hydrophobicity and less structural similarity to lipid compounds in
the cell membrane.[21] Cholesterol-based conjugates on their
hand have been reported to undergo fast exchange from the cell
membrane, which might account for the less efficient antibody-
recruiting capacity of cholesterol-based ARPs.[22] Using the
dialkyl-polyDNP conjugate, we subsequently tested its
persistence on the cell surface over prolonged periods of time by
treating cells with polymers followed by pulsing with anti-DNP at
different time points prior to flow cytometry and confocal
microscopy analysis. Note that in these experiments, cells were
carefully washed to remove unbound polymer to avoid continuous
replenishing of lipid-polymer on the cell surface from lipid-polymer
in solution. These experiments (Figure 2E) demonstrate that for
at least up to 4 days post dialkyl-polyDNP treatment, anti-DNP
can bind to the cell surface. Whereas antibody-dependent effector
mechanisms occur fairly rapidly (vide infra for ADCP data), we
hypothesize that prolonged cell surface display in vivo of ARPs
allows for a higher chance of Fc-recognition to occur and innate
effector killing to take place. Moreover, our data show that despite
the fact that indeed ARPs become endocytosed to a certain extent
(as evidenced by the dotted red fluorescent intracellular pattern in
the confocal images in Figure 2B), a very significant fraction of
ARPs remains on the cell surface and is capable of recruiting anti-
DNP to the cell surface for up to 4 days post pulsing of cells with
ARPs.
Interestingly, earlier reports suggest the cell surface persistence
of dialkyl-polymer conjugates to be no more than 24h,[19,20]
whereas at present we demonstrate antibody recruitment for at
least 4 days on dialkyl-polyDNP treated cells. However, in these
previous works, typically phospholipid-based conjugates where
used whereas in our present work we make us of neutral charged
amide-linked lipids. These findings highlight the complex
influence of lipid-polymer amphiphile structure and membrane
anchoring and internalization kinetics and the underlying
pathways. Our data indicates that the dialkyl-polyDNP conjugate
was endocytosed, but at a rate that is sufficiently slow to still allow
for antibody recruitment to last over several days. This could be a
crucial feature in vivo to allow for prolonged exposure of ARP-
treated tumors to antibody recruitment and subsequent innate
immune attack. Moreover, our ARP strategy excels in simplicity
and dialkyl-polyDNP does not require formulation into higher
ordered structures such as fusionogenic liposomes to drive cell
surface display.[23]
Figure 3 Innate effector killing of ARP-treated cells (A) Schematic
illustration of the ADCP assay. A431 cancer cells are fluorescently labeled with
eFluor450 (green color) and pulsed with polymer samples and monoclonal
human IgG1 anti-DNP. Macrophages prepared from human peripheral blood
monocytes (PBMCs) are fluorescently labeled with DiO (red color) and co-
cultured for 4h with the treated cancer cells. (B) Phagocytosis of cancer cells by
macrophages provokes a population of double positive cells in the upper right
quadrant of flow cytometry scatter plot. (C) Quantification of ADCP efficiency by
calculating the percentage of double positive cells in the upper right quadrant in
the flow cytometry scatter plots in panel B (n=3). Cetuximab concentration was
10 ng/mL. (n=3; 1 way Anova **** : p<0,0001)
To assess ADCP, cells were pulsed with ARPs and anti-DNP,
fluorescently labeled with eFluor 450 and co-cultured with
macrophages prepared from human peripheral blood
mononuclear cells (PBMCs), labeled with DiO. Flow cytometry
analysis (Figure 3) clearly shows the presence of a double-
positive quadrant, indicating phagocytosis of cancer cells by
macrophages, only when cancer cells were treated with dialkyl-
polyDNP ARP + anti-DNP. Controls that include lipid-polymer
conjugates without DNP or without lipid only induce a low
background phagocytosis. Note that the minor background
phagocytosis caused by the lipid-poly control could be attributed
to a slight extent of toxicity of this amphiphile, thereby triggering
macrophage to clear dead/dying cells.
Interestingly, ARPs induce a similar extent of phagocytosis as
Cetuximab, a monoclonal antibody that recognizes the epidermal
growth factor receptor (EGFR) which is strongly over-expressed
by A431 cells and that is known to exert its mechanism of action
through Fc-mediated innate effector killing. [4] The ability of ARPs,
within a similar experimental context, to be equally potent as a
clinically effective drug highlights the potential of ARPs. Moreover,
ARPs operate in a receptor-independent way, albeit requiring
As a final part of this work, we aimed at demonstrating the ability
of ARPs to trigger cancer cell killing by innate effector immune
cells. Hereto we performed a functional assay to analyze ARP-
mediated phagocytosis (ADCP) by macrophages, triggered by
Fc-receptor mediated recognition of clustered antibodies on the
surface of an ARP-anchored cancer cell. Hereto we made use of
A431 human squamous carcinoma cells and monoclonal human
IgG1 anti-DNP to match species and because the IgG1 isotype is
the most potent inducer of Fc-mediated effector killing in human.[4]
Figure S8 in Supporting Information confirms that this antibody
can also robustly be recruited onto the surface of ARP-treated
cells.
This article is protected by copyright. All rights reserved.

10.1002/anie.201905093
Angewandte Chemie International Edition
COMMUNICATION
[7]
[8]
[9]
P. J. McEnaney, C. G. Parker, A. X. Zhang, D. A. Spiegel, ACS Chem.
Biol. 2012, 7, 1139–51.
direct intratumoral injection when envisioning therapeutic
translation.
C. E. Jakobsche, C. G. Parker, R. N. Tao, M. D. Kolesnikova, E. F.
Douglass, D. A. Spiegel, ACS Chem. Biol. 2013, 8, 2404–11.
R. T. C. Sheridan, J. Hudon, J. A. Hank, P. M. Sondel, L. L. Kiessling,
Chembiochem 2014, 15, 1393–8.
In summary, we have demonstrated the ability of polymer-lipid
conjugates containing multiple DNP motifs to highly efficiently
anchor to cell membranes and provoke robust recruitment of anti-
DNP antibodies to the cell surface in a selective way. The
antibody recruitment efficacy of polymeric constructs composed
of multiple antibody-recruiting motifs dramatically surpasses
control monovalent constructs. The feasibility of this approach
was demonstrated to be highly efficient in vitro and proved
capable to induce innate effector killing against ARP-treated cells.
[10] S. Li, B. Yu, J. Wang, Y. Zheng, H. Zhang, M. J. Walker, Z. Yuan, H. Zhu,
J. Zhang, P. G. Wang, et al., ACS Chem. Biol. 2018, 13, 1686–1694.
[11] R. P. Murelli, A. X. Zhang, J. Michel, W. L. Jorgensen, D. A. Spiegel, J.
Am. Chem. Soc. 2009, 131, 17090–17092.
[12] A. F. Rullo, K. J. Fitzgerald, V. Muthusamy, M. Liu, C. Yuan, M. Huang,
M. Kim, A. E. Cho, D. A. Spiegel, Angew. Chemie Int. Ed. 2016, 55,
3642–3646.
[13] S.J. Metallo, R.S. Kane, R.E. Holmlin, G.M. Whitesides, J. Am. Chem.
Soc. 2003, 125, 4534-40.
Acknowledgements
[14] J. Li, S. Zacharek, X. Chen, J. Wang, W. Zhzng, A. Janczuk, P.G. Wang,
Bioorg. Med. Chem. 1999, 7, 1549-58.
[15] A. Das, P. Theato, Chem. Rev. 2016, 116, 1434-95.
[16] C. Boyer, V. Bulmus, T. P. Davis, V. Ladmiral, J. Liu, S. Perrier, Chem.
Rev. 2009, 109, 5402–36.
BGDG acknowledges the FWO-EOS program for funding. RDC
acknowledges Kom Op Tegen Kanker for funding.
[17] M. Mammen, S. Choi, G. Whitesides, Angew. CHEMIE-
INTERNATIONAL Ed. 1998, 37, 2755–2794.
Keywords: immune therapy • polymer amphiphiles • antibodies
• cancer • multivalency
[18] H. Liu, B. Kwong, D. J. Irvine, Angew. Chemie Int. Ed. 2011, 50, 7052–
7055.
[19] E. C. Woods, N. A. Yee, J. Shen, C. R. Bertozzi, Angew. Chem. Int. Ed.
Engl. 2015, 54, 15782–8.
[1]
A. M. Scott, J. D. Wolchok, L. J. Old, Nat. Rev. Cancer 2012, 12, 278–
287.
[20] K. Kato, C. Itoh, T. Yasukouchi, T. Nagamune, Biotechnol. Prog. 2004,
20, 897–904.
[2]
[3]
L. M. Walker, D. R. Burton, Nat. Rev. Immunol. 2018, 18, 297–308.
A. Pincetic, S. Bournazos, D. J. DiLillo, J. Maamary, T. T. Wang, R.
Dahan, B.-M. Fiebiger, J. V Ravetch, Nat. Immunol. 2014, 15, 707–716.
N. Gül, L. Babes, K. Siegmund, R. Korthouwer, M. Bögels, R. Braster, G.
Vidarsson, T. L. M. ten Hagen, P. Kubes, M. van Egmond, J. Clin. Invest.
2014, 124, 812–823.
[21] G. van Meer, D. R. Voelker, G. W. Feigenson, Nat. Rev. Mol. Cell Biol.
2008, 9, 112–24.
[4]
[22] T. Fritz, M. Voigt, M. Worm, I. Negwer, S. S. Müller, K. Kettenbach, T. L.
Ross, F. Roesch, K. Koynov, H. Frey, et al., Chem. - A Eur. J. 2016, 22,
11578–11582.
[5]
[6]
W. Wang, A. K. Erbe, J. A. Hank, Z. S. Morris, P. M. Sondel, Front.
Immunol. 2015, 6, DOI 10.3389/FIMMU.2015.00368.
C.-H. Lee, G. Romain, W. Yan, M. Watanabe, W. Charab, B. Todorova,
J. Lee, K. Triplett, M. Donkor, O. I. Lungu, et al., Nat. Immunol. 2017, 18,
889–898.
[23] A. Pulsipher, M. E. Griffin, S. E. Stone, J. M. Brown, L. C. Hsieh-Wilson,
J. Am. Chem. Soc. 2014, 136, 6794–6797.
This article is protected by copyright. All rights reserved.

10.1002/anie.201905093
Angewandte Chemie International Edition
COMMUNICATION
COMMUNICATION
Annemiek Uvyn, Ruben De Coen,
Mandy Gruijs, Cees W. Tuk, Marjolein
van Egmond, Bruno G. De Geest*
Page No. – Page No.
Efficient innate immune killing of
cancer cells triggered by cell surface
anchoring of multivalent antibody-
recruiting polymers
Antibody-recruiting polymers mark cancer cells for destruction by the innate
immune sysyem.
This article is protected by copyright. All rights reserved.