α-Mannosyl-Functionalized Cationic Nanohydrogel Particles for Targeted Gene Knockdown in Immunosuppressive Macrophages

Article abstract of DOI:10.1002/mabi.201900162

Immunosuppressive M2 macrophages govern the immunophathogenic micromilieu in many severe diseases including cancer or fibrosis, thus, their re-polarization through RNA interference is a promising concept to support combinatorial therapies. For targeted si

Full text of DOI:10.1002/mabi.201900162

Full PaPer
Nanohydrogel Particles
www.mbs-journal.de
α-Mannosyl-Functionalized Cationic Nanohydrogel
Particles for Targeted Gene Knockdown in
Immunosuppressive Macrophages
Nadine Leber, Leonard Kaps, Aiting Yang, Misbah Aslam, Mariacristina Giardino,
Adrian Klefenz, Niklas Choteschovsky, Sebastian Rosigkeit, Asmaa Mostafa, Lutz Nuhn,*
Detlef Schuppan,* and Rudolf Zentel*
This paper is dedicated to Prof. Helmut Ringsdorf on occasion of his 90th birthday.
1. Introduction
Immunosuppressive M2 macrophages govern the immunophathogenic
micromilieu in many severe diseases including cancer or fibrosis, thus,
their re-polarization through RNA interference is a promising concept to
support combinatorial therapies. For targeted siRNA delivery, however,
safe and stable carriers are required that manage cell specific transport to
M2 macrophages. Here, siRNA-loaded cationic nanogels are reported with
α-mannosyl decorated surfaces that target and modify M2 macrophages
selectively. Via amphiphilic precursor block copolymers bearing one single
α-mannosyl moiety at their chain end mannosylated cationic nanohydrogel
particles (ManNP) were obtained of 20 nm diameter determined by dynamic
light scattering and cryogenic electron transmission microscopy. α-Mannosyl
surface modification is confirmed by agglutination with concanavalin A.
SiRNA-loaded ManNP preferentially targets the overexpressed mannose
receptor CD206 on M2 macrophages, as shown by in vitro cell uptake
studies in M2 polarized primary macrophages. This specificity is confirmed,
since ManNP uptake could be reduced by blocking of CD206 with mannan.
Effective ManNP-guided siRNA delivery is confirmed by sequence-specific
gene knockdown of CSF-1R in M2-type macrophages exclusively, while the
expression levels in M1-polarized macrophages is not affected. In conclu-
sion, α-mannosyl-functionalized ManNPs are promising universal siRNA
carriers for targeted immunomodulatory treatment of immunosuppressive
macrophages.
The awarding of 2018’s Nobel Prize in
Physiology and Medicine to J.P. Allison
and T. Honjo for their work on immune
checkpoint inhibition exemplifies the tre-
mendous potential of immune interven-
tions to revolutionize classical therapies
including cancer.[1–3] Immunotherapy
has become a promising alternative or
additive therapeutic anti-tumor strategy
compared to classical chemotherapy or
radiation, as well as a novel approach
to halt or even reverse advanced organ
fibrosis.[4–6] Beside T-cells, other immune
cells including macrophages play an
important role during tumor progres-
sion.[7] Macrophages are present in almost
all tissues throughout the body and func-
tion as central regulators of tissue homeo-
stasis as well as innate immunity. Thereby,
they can be classified into two extreme
subsets: A typically highly activated phe-
notype (M1) expresses proinflammatory
cytokines like interleukin (IL)-12 which
promotes differentiation of T-helper
cells (Th1) and, thereby, prevents tumor
Dr. N. Leber, Prof. R. Zentel
M. Aslam
Institutes of Organic Chemistry
Department of Microbiology
Johannes Gutenberg-University of Mainz
Duesbergweg 10-14, 55128 Mainz, Germany
E-mail: zentel@uni-mainz.de
Shaheed Benazir Bhutto Women University
LARAMA, Charsadda Road, Peshawar, Pakistan
Dr. L. Nuhn
Dr. L. Kaps, Dr. A. Yang, M. Aslam, M. Giardino, A. Klefenz,
N. Choteschovsky, Dr. S. Rosigkeit, A. Mostafa, Prof. D. Schuppan
Institute of Translational Immunology and Research Center for
Immunotherapy
University Medical Center of the Johannes Gutenberg-University Mainz
Obere Zahlbacher Str. 63, 55131 Mainz, Germany
Max-Planck-Institute for Polymer Research
Ackermannweg 10, 55128 Mainz, Germany
E-mail: lutz.nuhn@mpip-mainz.mpg.de
Prof. D. Schuppan
Division of Gastroenterology
Beth Israel Deaconess Medical Center, Harvard Medical School
330 Brookline Avenue, Boston, MA 02215, USA
E-mail: Detlef.Schuppan@unimedizin-mainz.de
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/mabi.201900162.
DOI: 10.1002/mabi.201900162
©
1900162 (1 of 12)
Macromol. Biosci. 2019, 1900162
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
progression and metastasis.[8] In vitro, M1 macrophages can be
induced by proinflammatory mediators and cytokines such as
bacterial lipopolysaccharide (LPS) or interferon-γ. M2 macro-
phages are induced by regulatory cytokines such as IL-4 and
IL-13. This phenotype is involved in wound healing processes
while promoting the production of extracellular matrix leading
to progressive organ fibrosis during persistent damaging
stimuli.[4–6,9,10] Overall, M2 macrophages also favor an immuno-
suppressive and anti-inflammatory state in the cancer microen-
vironment.[4–6,11,12] As so-called tumor-associated macrophages
(TAMs—resembling M2-type macrophages), they generally
correlate with a poor prognosis toward aggressive tumor pro-
gression.[13,14] TAMs shape their tumor immunoenvironment
by releasing anti-inflammatory, proangiogenic, and profibrotic
growth factors and cytokines like vascular endothelial growth
factor (VEGF), IL-10, and transforming growth factor (TGF)-β,
undermining tumor surveillance by the immune system.
M2 polarized macrophages are also overrepresented in
chronically damaged organs including fibrotic or cirrhotic
livers. Here, they are associated with excessive deposition of
extracellular matrix, while M1 polarized macrophages are linked
to its degradation and, thus, fibrosis regression.[15] During
fibrosis resolution, the quantity of M2 macrophages decreases,
while their M1 counterparts increase. With protracted damage,
fibrosis can progress toward liver cirrhosis, the major cause of
morbidity and mortality of patients with chronic liver disease.[6]
The prevention or reversal of cirrhosis has become one of the
major endpoints in clinical trials.
Thus, suppressing the activity of M2 macrophages, regarding
their maturation, differentiation, and activation state is of high
importance as immunotherapeutic strategies for many disease
types.[16]
Previous studies have demonstrated that macrophages can
be re-educated or re-polarized to change their phenotype by
various stimuli in the local microenvironment.[17,18] Thus, tar-
geting of TAMs/M2 macrophages and repolarizing toward a
proinflammatory M1-type phenotype could be an effective tool
to control tumor growth. In a comprehensive proteomic study
of 174 ovarian tumors, Zhang et al. provided evidence that the
M1/M2 ratio is a crucial prognostic factor for a 5 year survival
of patients.[8]
Although RNA interference mediated by siRNA is a robust
method for sequence-specific gene knockdown, siRNA, as a
polyanionic oligonucleotide, has poor pharmacokinetics as evi-
denced by rapid degradation of nucleases in the blood stream
and low cell penetration rates. Consequently, a suitable carrier
system is required to ensure efficient cellular siRNA delivery.
Furthermore, a targeted delivery of siRNA into M2 macro-
phages is essential to ensure gene knock down and repolariza-
tion only in this cell population.
TAM/M2 macrophages express high levels of the macro-
phage mannose receptor CD206, which effectively binds
α-mannosyl structures.[23] Instead of addressing this receptor
by nanobodies,[24] we took advantage of the macrophage man-
nose receptor’s intrinsic binding affinity to mannose and
developed a densely α-mannosyl-functionalized siRNA carrier
system for effective targeting of M2 macrophages. To protect
the sensitive cargo from nuclease degradation and promote
cellular uptake, the polyanionic siRNA has to be loaded into
a cationic nanohydrogel particle system[25–30] via electrostatic
interactions.[28,31] Such systems can be derived from amphi-
philic reactive ester block copolymers which self-assemble
into micellar structures. The core of these micelles can be
covalently cross-linked by bifunctional oligoamines leading to
a polarity switch from hydrophobic to cationic/hydrophilic.
Thus, complexation of siRNA into the core of these cationic
nanohydrogel particles can be achieved. In addition, such car-
riers can further be equipped with either non-biodegradable or
biodegradable crosslinkers to enhance siRNA release after cel-
lular internalization.[30,32]
For addressing the macrophage mannose receptor CD206 on
M2 macrophages, a densely mannose-functionalized cationic
nanohydrogel particle is required. Therefore, we developed here
a well-defined α-chain end functionalized amphiphilic block
copolymer. To this respect, a new chain transfer agent (CTA)
was equipped with an α-mannosyl functionality and applied
to RAFT block copolymerization. The obtained α-chain end
mannose-functionalized amphiphilic block copolymer could
then be utilized in the fabrication process of α-mannosyl modi-
fied cationic nanohydrogel particles (ManNP). The resulting
α-mannose surface modification enables cell specific targeting
of TAM/M2 macrophages via their mannose receptor (CD206).
Furthermore, a targeted immunomodulatory effect (CSF-1R
gene knockdown) induced by ManNP complexed anti-CSF-1R
siRNA was found exclusively in M2 macrophages while the
expression status in M1 macrophages remained unaffected.
The results confirm that α-mannosyl-functionalized cationic
nanohydrogel particles can be used as universal siRNA carrier
for targeted immunomodulatory treatment of various macro-
phage-guided immunosuppressive diseases.
As a key regulator for differentiation and survival of bone
marrow–derived macrophages (BMDM), the macrophage
colony-stimulating factor (MCSF-1) is an interesting target
for cancer therapy.[19–21] Joyce et al. evinced that inhibition of
the CSF-1 receptor (CSF-1R) via the tyrosine kinase inhibitor
BLZ945 significantly hindered metastasis and tumor progres-
sion in glioblastoma multiforme diseased mice by repolarizing
M2 macrophages toward pro-inflammatory M1 macrophages.[22]
BLZ945 is currently being tested in Phase II clinical trials
(https://clinicaltrials.gov/ct2/show/NCT02829723).
Beside small molecules or anti-CSF-1R antibodies, CSF-1R
inhibition can be achieved by RNA interference using anti-CSF-
1R siRNA.[16,20] Anti-CSF-1R siRNA mediated knockdown could
afford a selective repression of CSF-1R with fewer side-effects
compared to tyrosine kinase inhibitors or monoclonal anti-
bodies, which address several targets globally or bear the risk
of fatal systemic allergic reactions and, thus, may cause severe
side reactions.
2. Results and Discussion
2.1. α-Mannosyl-Modified Cationic Nanohydrogel Particles
Design and Preparation
The preparation of α-mannosyl-functionalized cationic nano-
hydrogel particles (ManNP) was achieved according to the con-
cept presented in Figure 1.[25,26]
©
1900162 (2 of 12)
Macromol. Biosci. 2019, 1900162
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
Figure 1. Preparation of α-mannosyl-functionalized cationic nanohydrogel. An α-mannosyl-CTA (Man CTA) is applied to RAFT polymerization of an
amphiphilic reactive ester block copolymer, followed by self-assembly forming micellar structures with reactive ester cores. Cross-linking of these pre-
cursor micelles is obtained by a bifunctional oligoamine giving α-mannosyl-functionalized cationic nanohydrogel particles with a cationic core. Loading
of siRNA into carrier for targeted delivery is achieved by electrostatic interactions.
A stepwise RAFT polymerization, using a novel α-mannosyl-
functionalized chain transfer agent (Man CTA) yielded a
precisely α-chain end mannose-functionalized amphiphilic
block copolymer. The amphiphilic diblock copolymer, com-
posed of a hydrophilic PEG-like and a hydrophobic reactive
ester block from polymerized triethylene glycol methyl ether
methacrylate (MEO3MA) or pentafluorophenyl methacrylate
(PFPMA), respectively, was dissolved in polar aprotic solvents
(e.g., DMSO) and spontaneously self-assembled into nano-
sized micellar structures. Stabilization of the inner core was
achieved by cross-linking using spermine as bifunctional oli-
goamine cross-linker.[26] During this post-polymerization modi-
fication step, the primary amines of the cross-linker react with
the reactive ester moieties of the polymer and form covalently
stabilized nanohydrogel particles. The secondary amines of the
cross-linker are protonated under physiological pH (pKa ≈ 8.4)
and cause a change of the inner core’s polarity, turning it from
hydrophobic to cationic, thus, hydrophilic.[33] The crosslinked
α-mannosyl-functionalized cationic nanohydrogel particles
efficiently complex anionic siRNA due to electrostatic interac-
tions,[26] while the mannose functionalities on the hydrophilic
chain ends are exposed to the particle surface. They facilitate
cell specific siRNA delivery to M2-type macrophages via interac-
tion with the mannose receptor CD206.
end-functionalized block copolymer and chose to introduce the
α-mannose functionalization already at the beginning of the
polymer synthesis utilizing a novel α-mannose-modified chain
transfer agent (Man CTA) for the RAFT block copolymerization.
As shown in Figure 2A, the commercially available CTA
4-cyano-4-(phenylcarbonothioylthio)pentanoic acid was esteri-
fied by pentafluorophenyl trifluoracetate yielding pentafluo-
rophenyl 4-cyano-4-((phenylcarbonothioyl)thio)pentanoate (PFP
CTA). The introduced activated ester group ensures selective
reactivity to primary amines.[34,35] Subsequently, PFP CTA was
converted selectively with an α-mannosylated ethylene glycol
amine spacer to afford an α-mannosyl-functionalized CTA
(Man CTA).[36] Successful synthesis was confirmed by 1H,
13C, and further 2D NMR spectroscopy as well as ESI mass
spectroscopy (see Figure 2B and Figures S1–S5, Supporting
Information).
Applying this novel Man CTA in a RAFT polymerization the
required α-mannosyl α-chain end-functionalized amphiphilic
block copolymer was obtained (see Figure 2C).
In general, the polymerization of the hydrophilic and hydro-
phobic block is feasible in both directions.[26] However, when
using the Man CTA, the hydrophilic block, consisting of the
PEG-like monomer MEO3MA monomers, must be polymer-
ized first to ensure exposure of the mannose functionalization
at the hydrophilic end of the block copolymer. The P(MEO3MA)
block is essential for block copolymer self-assembly as well as
solubility and steric shielding of the later cationic nanohydrogel
particles. MEO3MA conversion during RAFT-polymerization
was monitored by 1H NMR (see Figures S6 and S7, Supporting
Information). Here, the novel Man CTA could guarantee suc-
cessful linear monomer conversion, indicating a controlled
To ensure an effective targeting of the macrophage mannose
receptor CD206 on M2 macrophages a highly functionalized
carrier system is required. Following the well-established azide-
alkyne click approach such a high degree of functionalization
would not be guaranteed due to limitations in the accessibility
of the reacting moieties.[24] Thus, to construct such highly
functionalized ManNPs, we designed an amphiphilic α-chain
©
1900162 (3 of 12)
Macromol. Biosci. 2019, 1900162
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
Figure 2. A) Synthesis of mannose-functionalized chain transfer agent (Man CTA) and B) its corresponding HSQC (1H, 13C) NMR. C) Applied during
RAFT polymerization it affords α-mannosyl α-chain end functionalized amphiphilic block copolymers Man-P(MEO3MA)-b-P(PFPMA), D) as demon-
strated by the corresponding SEC elugrams with THF as eluent.
quasi-living character in line with the conditions for RAFT
polymerization. Afterward, the hydrophobic block was attached,
using PFPMA as reactive ester monomer. This fluorophilic
block enables the formation of micelles in polar aprotic sol-
vents. Furthermore, it can be substituted selectively by primary
amines during post-polymerization cross-linking. After removal
of the dithiobenzoate endgroup by addition of an excess of
the initiator 2,2-azobis(4-methoxy-2,4-dimethylvaleronitrile)
(AMDVN), the final amphiphilic α-mannosyl α-chain end
modified block copolymers α-mannosyl-poly(tri(ethylene
glycol)methyl ether methacrylate)-block-poly(pentafluorophenyl
methacrylate) (Man-P(MEO3MA)-b-P(PFPMA) were obtained.
Monomodal elution profiles with narrow size distributions
were found for these polymers by size exclusion chromato-
graphy (Figure 2D). H NMR spectroscopy further confirmed
successful synthesis of Man-P(MEO3MA)-b-P(PFPMA) (see
Figures S8 and S9, Supporting Information).
As reported earlier, a hydrophilic block ratio of ≈40%
is necessary to facilitate reactive ester block copolymer
self-assembly into micellar structures in polar aprotic
solvent.[25,26,30] At the same time nanoparticles with a thicker
hydrophilic shell are obtained which might have better pro-
tein repelling stealth-like properties conferred by the poly-
merized MEO3MA monomers. The composition of both
mannose- and non-functionalized diblock copolymer used in
this study met these requirements and favored nanoparticle
formation in DMSO (see Table 1).
α-Mannosylated cationic nanohydrogel particles, derived
from α-mannosyl α-chain end modified amphiphilic reactive
ester block copolymers, were prepared according to Figure 1. As
reported earlier, the fluorophilic reactive ester block promotes
micellar self-assembly when the polymer is dispersed in DMSO
as polar aprotic solvent.[25,26,37] The formed micelles were cova-
lently cross-linked, using stoichiometric amounts of spermine,
as bifunctional cross-linker (1 eq. of spermine bearing two pri-
mary amines to 2 eq. of PFP reactive ester) (see Figure S10,
Supporting Information). Complete conversion of the PFP
esters was confirmed by 19F NMR spectroscopy as shown in
Figure S11, Supporting Information. Covalently polymer-bound
PFP and liberated PFP differ in their chemical shift (polymer
1
©
1900162 (4 of 12)
Macromol. Biosci. 2019, 1900162
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
Table 1. Physiochemical properties of the carriers.
b)
Rha) (NP only) [nm] (and µ2
9.5 (0.19)
)
ζ (NP only)c) [mV]
ζ (NP:siRNA)c) 10:1 [mV]
Name
Precursor polymer
ManNP
NonNP
Man-P(MEO3MA)18-b-P(PFPMA)30
38.4 1.8 (H2O) 20.1 3.1 (HEPES)
38.2 0.6 (H2O) 22.8 0.7 (HEPES)
1.7 0.1 (H2O) 14.2 1.6 (HEPES)
[30]
P(MEO3MA)24-b-P(PFPMA)32
11.9 (0.13)
−1.1 0.2 (H2O) 20.0 1.6 (HEPES)
a)Hydrodynamic radius (Rh) determined by multi-angle DLS measurements in PBS (see corresponding 30° measurement and angle-dependency in Figure S11, Supporting
Information); b)Polydispersity determined by DLS via cumulant fit at 90°; c)Zeta potential given as mean and standard deviation of three measurements determined by
zetasizer under the given buffer conditions.
bound PFP from −152 to −163 ppm, free PFP −172 to 194 ppm)
when analyzed by 19F NMR.
2.1.1. Nanoparticle Characterization
Complete conversion of PFPMA, even below the NMR detec-
tion limit, was promoted by adding additional 2 eq. of sper-
mine, yielding stable amide bonds. Subsequently, the covalently
cross-linked nanohydrogel particles were dialyzed for several
days against pure (millipore) H2O to remove all small mole-
cular by-products. After lyophilization, both ManNP as well as
non-functionalized particles (NonNP) were obtained as volumi-
nous solid. The particles could easily be resuspended in phos-
phate buffered saline (PBS) as physiological medium compat-
ible in view of the planned biological experiments.
The size of the NPs was determined by multi-angle dynamic
light scattering (DLS) measurements providing hydrodynamic
radii (Rh) for ManNP and NonNP of 9.5 and 11.9 nm, respec-
tively (see Figure 3A; Figure S12, Supporting Information;
Table 1), a comparable size for both species. Consequently,
NonNPs can be considered as a valid control for ManNP for all
further chemical and biological experiments.
Cryogenic transmission electron microscopy images of sur-
face modified ManNP, as shown in Figure 3B, confirmed their
spherical shape as well as narrow dispersity. These findings
Figure 3. Characterization of ManNP: A) DLS autocorrelation curve of ManNP (red dots) and NonNP (blue dots) at 30° and corresponding fits (colored
straight lines); B) cryoTEM image of ManNP demonstrating spherical particles. C) Agarose gel electrophoresis to determine optimal complexation
mass ratios of ManNP to siRNA by increasing amounts of ManNP to siRNA. D) DLS measurements over 30 min showing agglutination behavior of
ManNP in presence of Concanavalin A (red), E) while no agglutination is found for NonNP with Con A (blue) or Con A alone (green); corresponding
visual observation of agglutination.
©
1900162 (5 of 12)
Macromol. Biosci. 2019, 1900162
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
correspond to the µ2 parameter of 0.19 determined by multi-
angle DLS measurements (see Table 1).
after addition of Con A. Continuous monitoring over 5 min
revealed stable nanoparticles in aqueous solution before
Con A was added. Addition of Con A triggered a sudden
increase in size for ManNP, resulting in macroscopic aggre-
gates (Figure 3E), whereas NonNP did not aggregate at all
(Figure 3D). The minor increase in size for NonNP after addi-
tion of Con A was caused by impurities of Con A at around
100 nm.
siRNA complexation capacity of ManNP was determined by
agarose gel electrophoresis. siRNA complexed by NPs cannot
migrate into the agarose gel during electrophoresis, while
free siRNA is attracted to diffuse into the gel according to its
size and negative charge. A mass ratio of 10:1 of ManNP to
siRNA yielded complete complexation of siRNA, as indicated
by the absence of free siRNA (Figure 3C, first lane). Similarly
to ManNP, NonNP exhibited identical complexation capacities
(10:1 for NonNP:siRNA) as reported previously.[30] Furthermore,
the complexation stability of these siRNA carrier systems was
confirmed under biologically relevant conditions using blood
serum (see Figure S13, Supporting Information). These experi-
ments show that only a minor fraction of siRNA was released
from the carriers, while the majority of siRNA remained inside
the particles. Consequently, both carriers were suitable carriers
for siRNA delivery.
The zeta potential value (ζ) of NPs corresponds to their elec-
trokinetic potential at the slipping/shear plane of moving par-
ticles in an electric field.[38] Measurements of non-complexed
and complexed (10:1 for NP:siRNA) particles were performed
in pure H2O as well as in 10 mm HEPES (4-(2-hydroxyethyl)-
1-piperazineethanesulfonic acid) buffer, at pH 7.4 as non-ionic
buffer. Both particles (ManNP and NonNP) without siRNA
cargo yielded positive zeta potentials of around +38 mV in
H2O in accordance with previous studies[27] (see Table 1 and
Figure S14, Supporting Information). After particle compl-
exation with siRNA at a mass ratio of 10:1 (NP:siRNA), they
displayed almost neutral zeta potentials. The near neutraliza-
tion of the particles’ cationic core with anionic charged siRNA
confirms successful electrostatic complexation between car-
rier and cargo. Neutral siRNA/NP complexes are crucial to
avoid unspecific interactions with blood components, the
extracellular matrix and cell membranes and to facilitate bio-
compatibility as well as colloidal stability in biological experi-
ments.[27] Additional zeta potential measurements of siRNA
ManNP/NonNP complexes (mass ratio 10:1 NP:siRNA)
were conducted in HEPES buffer at neutral pH (Table 1 and
Figure S14, Supporting Information). Interestingly, no neutral
potential for siRNA ManNP/NonNP complexes was obtained
after complexation in the HEPES buffered solution, in con-
trast to complexation in pure H2O. However, complexes of
ManNP/siRNA—in comparison to NonNP/siRNA—demon-
strated a reduced cationic character (+14.2 mV vs +20.0 mV)
suggesting a superior electrostatic shielding of the cationic
core through the introduction of the hydrophilic α-mannosyl
functionalization.
In brief, these results confirm that ManNPs were success-
fully functionalized with surface accessible mannose moieties
for interaction with mannose-binding receptors.
2.2. Targeting of Immunosuppressive M2-Type Macrophages
Prior to targeting studies of the particles, NonNP and ManNP
loaded with scrambled siRNA (scsiRNA) (weight-to-weight w/w
ratio of NP to siRNA 10:1) were tested for in vitro cytotoxicity
in two different cell lines relevant for liver carcinogenesis as
well as fibrogenesis.[4] A significant cytotoxicity was neither
detectable in RAW-macrophages nor in 3T3-fibroblasts up to
corresponding high siRNA concentrations of 300 nm siRNA
(Figure 4A). The results confirm the excellent biocompatibility
of the novel ManNP equal to their non-functionalized ana-
logues as reported earlier.[29,30]
Fluorescently labeled (Oregon Green 488 cadaverine)
NonNP/ManNP loaded with scsiRNA demonstrated a robust
dose-dependent in vitro uptake in RAW-macrophages, HepG2
human hepatoma cells and SVEC4-10 endothelial cells that
reflect characteristics of, for example, parenchymal and non-
parenchymal cells in the liver (Figure S15, Supporting Informa-
tion). Here, the cellular uptake was equally efficient in all three
cell lines, although baseline mean fluorescent intensity (MFI)
was elevated for endothelial cells due to intrinsic autofluores-
cence compared to hepatocytes and macrophages (Figure S15,
Supporting Information; endothelial cells treated with PBS
control).
Next, we studied if the α-mannose functionalization of
ManNP would affect their cellular uptake in CD206 overex-
pressing M2- versus M1-polarized primary murine macro-
phages (BMDM). As demonstrated in Figure 4B, in vitro
M2-macrophage polarization induced the expression of the
M2-type marker CD206 in contrast to M1-polarized macro-
phages, while unpolarized (M0) macrophages showed an
intermediate expression level. We observed that siRNA
loaded ManNP demonstrated a preferential cellular uptake
in M2 polarized macrophages, while uptake was significantly
lower in M1 polarized macrophages. Non-functionalized
NonNP loaded with scsiRNA displayed, however, a rather
unspecific uptake regardless of the polarized phenotype
(Figure 4C).
Additionally, we tested the specificity of the uptake of
α-mannosyl functionalized particles in M2 macrophages with
a mannan blocking study.[40] Mannan as a linear poly-mannose
can be used to block the corresponding mannose receptor
(CD206) on M2 macrophages. By incubating M2 polarized
macrophages with mannan prior to the addition of ManNP, the
majority of CD206 receptors were saturated which led to a sig-
nificant decrease in cellular uptake of ManNPs (see Figure S16,
Effective targeting and specific uptake of these ManNP by
mannose receptor bearing M2-polarized macrophages can
only be expected when the mannose moieties are exposed at
the NP’s surface. Therefore, accessibility of the mannose moi-
eties was verified by a Concanavalin A (Con A) binding assay.
Con A is a lectin that binds specifically to α-D-mannosyl or
α-D-glucosyl residues at terminal, accessible positions. Due
to its tetrameric structure at physiological pH, Con A can
only cause agglutination by binding to carbohydrate residues
on multiple particles.[39] Agglutination of ManNP as well as
NonNP, as control, was investigated by DLS measurements
©
1900162 (6 of 12)
Macromol. Biosci. 2019, 1900162
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
Figure 4. In vitro performance of ManNP and NonNP. A) Cell viability of RAW-macrophages and 3T3 fibroblasts exposed to escalating doses of
scramble siRNA (scsiRNA)-loaded ManNP/NonNP showing no significant cytotoxic effect versus buffer (PBS-control) treated cells, as determined by
the MTT assay (10:1 weight-to-weight ratio w/w of NP:siRNA, incubation time 72 h, n = 2). B) Mannose receptor CD206 expression of polarized M1,
M2, or native M0 primary macrophages (BMDM). Macrophages were polarized toward M2 or M1 phenotype by incubation with IL-4/13 (20 ng mL−1
)
or LPS/Int-γ (1 µg mL−1, 50 ng mL−1) for 24 h, respectively. M0 macrophages as non-polarized native phenotype were obtained by incubation with
equivalent amounts of PBS as control. C) Cellular uptake of scsiRNA loaded, Oregon Green labeled ManNP/NonNP in M2- and M1-polarized primary
macrophages at 25, 50, and 100 nm siRNA concentrations after 1 h incubation and analyzed by flow cytometry (MFI—mean fluorescence intensity of
the near infrared dye-labeled NPs, n = 3, *p < 0.05, **p < 0.001).
Supporting Information). For NonNPs, no such effect in cel-
lular uptake could be observed once M2 macrophages had been
incubated with mannan.
2.3. Targeted Gene Knockdown in Immunosuppressive M2-Type
Macrophages
Conclusively, the introduction of α-mannosyl functionalities
on the surface of ManNPs leads to a preferential cellular uptake
into M2 macrophages and at the same time shows a reduced
unspecific uptake in non-targeted phenotypes.
CSF-1 (colony-stimulating factor) drives the differentiation and
proliferation of M2-type macrophages by binding to its receptor
CSF-1R. Targeting the CSF-1R is attractive in combination with
both standard treatment modalities and immunotherapeutic
©
1900162 (7 of 12)
Macromol. Biosci. 2019, 1900162
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
qualify as promising carriers for any kind of
siRNA relevant in treatments of immunosup-
pressive diseases characterized by a M2 mac-
rophage overexpression.
3. Conclusion
In this study siRNA nanohydrogel carriers could
successfully be equipped with an α-mannosyl
surface coating for cell specific targeting of M2
polarized macrophages through the overex-
pressed mannose receptor CD206. To achieve
a high degree of α-mannosyl functionaliza-
tion exposed to the carrier’s surface, α-chain
end α-mannosyl functionalized amphiphilic
diblock copolymers were generated by applying
a α-mannosylated CTA. Stepwise RAFT polym-
erization, first of the hydrophilic PEG-like block
containing MEO3MA and subsequent polym-
erization of the hydrophobic fluorinated block
PFPMA yielded well defined α-mannosyl func-
tionalized amphiphilic precursor diblock copoly-
mers which could subsequently be used for
cationic nanohydrogel synthesis. For this pur-
pose, block copolymers were self-assembled into
nanosized micelles in DMSO as aprotic organic
solvents. The nanosized aggregates were cova-
Figure 5. In vitro gene knockdown of anti-CSF-1R siRNA or scsiRNA loaded nanohydrogel
particles in M2- or M1-polarized primary macrophages. The amount of CSF-1R transcripts
was determined by quantitative real-time polymerase chain reaction (qPCR) normalized
by GAPDH (glyceraldehyde phosphate dehydrogenase) expression as housekeeping gene.
Expression of CSF-1R was significantly lowered in M2- but not M1-polarized macrophages
when treated with 100 nm anti-CSF-1R siRNA/ManNP complexes, while M1 polarized mac-
rophages remained unaffected (*p < 0.05 vs scsiRNA/ManNP, means SD, n = 1). Statis-
tical significance was determined by one-way ANOVA. Complete statistical evaluation can
be found in Figure S17, Supporting Information.
agents for various cancers and immunosuppressive diseases.[41]
Furthermore, induction of repolarization of immunosuppres-
sive M2 macrophages toward proinflammatory M1 macro-
phages by inhibition of the CSF-1R has great potential in
hindering metastasis and tumor progression.[22]
Thus, ManNPs/NonNPs were loaded with anti-CSF-1R
siRNA to assess their targeted knockdown efficacy in polar-
ized primary macrophages. In line with our prior data for cel-
lular uptake, anti-CSF-1R siRNA loaded ManNP induced a
significant knockdown for CSF-1R transcripts in M2-polarized
macrophages, while M1-macrophages remained unaffected
(Figure 5). Interestingly, the anti-CSF-1R loaded NonNP
(without α-mannosyl functionalization on the nanoparticle sur-
face) did not exert any significant knockdown for the targeted
transcript regardless of the macrophage phenotype.
Conclusively, we demonstrated that the α-mannosyl func-
tionalized cationic nanoparticles enabled a targeted delivery
of siRNA to immunosuppressive M2 macrophages. The func-
tionalization did not affect the biocompatibility of the carriers.
ManNP were preferentially taken up by M2 polarized macro-
phages, while their uptake was low in M1 macrophages, likely
mediated by the mannose receptor CD206+ that is upregulated
in M2 versus M1 macrophages.
Furthermore, dense α-mannosyl coatings render an addi-
tional steric shielding, in synergism with the PEG corona,
thereby, reducing unspecific interactions. In line with the
improved cellular uptake of ManNP in CD206 expressing
macrophages, anti-CSF-1R siRNA loaded ManNP induced a
significant and sequence-specific knockdown for the thera-
peutically relevant target transcript CSF-1R predominantly in
M2 polarized primary macrophages. Consequently, ManNPs
lently stabilized by crosslinking the reactive ester groups of their
inner core with spermine, a bifunctional oligoamine, resulting
in stable α-mannosyl functionalized nanohydrogel particles. Suc-
cessful surface exposure of α-mannose moieties was confirmed
by Concanavalin A (Con A) agglutination. Besides, α-mannosyl
functionalization provided additional shielding for the nanohy-
drogel particles, synergistically to their hydrophilic PEG corona.
ManNP were thoroughly validated in in vitro experiments
for cell toxicity, cellular uptake, and siRNA delivery to M2-type
macrophages. α-Mannosyl functionalization did not compro-
mise the biocompatibility, as shown by cell viability assays in
RAW-macrophages and 3T3-fibroblasts. M2 polarized primary
macrophages showed an efficient cellular uptake of ManNP
via macrophage mannose receptor (CD206+) mediated uptake.
The specificity of the targeting could be confirmed by blocking
CD206 with mannan.
Finally, 100 nm anti-CSF-1R siRNA loaded ManNP performed
a robust and dose-dependent in vitro gene knockdown of the
CSF-1R mRNA transcript in CD206 high M2-macrophages
exclusively, while the expression status was not altered in
M1-polarized (CD206 low) primary macrophages. Currently
we are working to translate these in vitro results into a relevant
in vivo model. Therefore, we are focusing on liver fibrosis and
liver cancer models, since it is known that the progression of
liver fibrosis as well as of liver cancer is promoted by overex-
pression of M2 macrophages.
To conclude, the α-mannosyl functionalized nanohydrogel
particles represent a promising vehicle platform for cell specific
delivery of any kind of therapeutic siRNA to immunosuppres-
sive M2 polarized macrophages that are key players in tumor
and fibrosis progression.
©
1900162 (8 of 12)
Macromol. Biosci. 2019, 1900162
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
purchased from IBA GmbH (Göttingen, Germany) with the following
sequence:
Sense strand: 5′-AGG UAG UGU AAU CGC CUU C TT-3′
4. Experimental Section
Materials: All chemicals were obtained from common commercial
sources and used as received unless otherwise indicated. Oregon
Green 488 cadaverine was obtained from Invitrogen and IRDye 800
RS cadaverine from Li-Cor Biosciences GmbH (Germany). PBS was
obtained from Fisher BioReagents containing 137 mm NaCl, 11.9 mm
phosphates and 2.7 mm KCl. 1 mN-(2-hydroxyethyl)piperazine-N-(2-
ethanesulfonic acid) (HEPES) buffer pH 7.0–7.6 was obtained from
Sigma-Aldrich and prepared at 10 mm in Millipore water. Anhydrous
dimethyl sulfoxide (DMSO) was obtained from Acros and stored over
a molecular sieve (4 Å). Anhydrous dioxane as well as tetrahydrofuran
(THF) were freshly distilled from sodium and anhydrous dichlormethane
from calcium hydride. Column chromatography was done using silica
obtained from Machery-Nagel (0.063–0.2 mm/20–230 mesh). For
dialysis ZelluTrans membranes from Roth with a molecular weight cutoff
of 6000–8000 g mol−1 were used.
Anti-sense strand: 3′-TT UUC AUC ACA UUA GCG GAA C-5′
A 0.75 wt% agarose gel was prepared with TBE buffer and 0.02%
GelRed (10000x in water) for DNA staining. Samples were prepared
by mixing the nanoparticles in PBS buffer with siRNA in endotoxin free
water at given weight-to-weight ratios and following incubation at room
temperature for 30 min. After addition of loading buffer, all samples
were applied onto the gel and analyzed at 120 V for 30 min. Visualization
occurred in a dark hood under UV light (365 nm) and documentation
was done by a DCC camera (scientific grade, Intas).
Lectin Binding Assay: As proof of successful preparation of
functionalized nanohydrogel particles with accessible α-mannosyl
moieties on the surface, a Concanavalin A (Con A, 104 kDa) agglutination
assay was performed to investigate the ability to bind mannose-
binding lectins. From a 1.0 g L−1 solution of NonNP and ManNP in
phosphate buffering solution (DPBS, pH 7.4) containing CaCl2 and
MgCl2 (0.901 mm and 0.493 mm, respectively) 250 µL were transferred
into a DLS cuvette (semi-micro, polystyrene, 1.5 mL, BRAND). DLS
measurements were recorded each for 30 s using a Malvern Zetasizer
NanoZS. After 5 min, 250 µL of a 1.0 g L−1 ConA in DPBS was added
to the microcuvette under continuous recording of the DLS data. As
control, pure ConA (1.0 g L−1, 250 µL) was analyzed.
NMR Spectroscopy: 1H, 13C, and 19F NMR spectra were recorded on a
Bruker 400 MHz FT NMR spectrometer and the chemical shifts (δ) are
given in parts per million (ppm) relative to TMS. Samples were prepared
in deuterated solvents and their signal referenced to residual non-
deuterated solvent signals (δ = 2500 ppm for DMSO-d6, δ = 7260 ppm
for CDCl3).
Size Exclusion Chromatography (SEC): SEC measurements were
performed in THF as solvent using the following setup: pump PU
1580, auto sampler AS1555, UV-detector UV 1575 (detection at
254 nm), and RI-detector RI 1530 from JASCO. Columns were used
from MZ-Analysetechnik: MZ-Gel SDplus 102 Å and MZ-Gel SDplus
106 Å. Calibration was done using polystyrene standard purchased
from Polymer Standard Services and toluene was used as internal
standard.
Monomer Synthesis: Tri(ethylene glycol) methyl ether methacrylate
(MEO3MA) and pentafluorophenyl methyacrylate (PFPMA) were
synthesized as previously reported.[26]
Synthesis
of
11-Amino-3,6,9-trioxa-undecyl-α-d-mannopyranoside
(Amine-Mannose): An α-mannosylated ethylene glycol amine spacer
was synthesized according to an earlier reported six-step reaction
procedure.[36]
Dynamic Light Scattering (DLS): For multi-angle DLS measurements
cylindrical quartz cuvettes (Hellma, Mühlheim, Germany) were cleaned
by dust-free distilled acetone and transferred to a dust free flow box.
Nanoparticle samples were prepared at 0.1 g L−1 in PBS and filtered
into the cuvettes through Pall GHP filters, 0.2 µm pore size. DLS
measurements were performed using the following setup at 20 °C: a
Uniphase He/Ne Laser (25 mW output power at λ = 632.8 nm) and
an ALV-CGS 8F SLS/DLS 5022F goniometer with eight simultaneously
working ALV 7004 correlators and eight QEAPD avalanche photodiode
detectors. The correlation functions of the particles were fitted using
a sum of two exponentials. The z-average diffusion coefficient Dz was
Synthesis of Pentafluorophenyl 4-cyano-4-((phenylcarbonothioyl)
thio)pentanoate (PFP CTA): In a round bottom flask under nitrogen
atmosphere equipped with a magnetic stir bar and a rubber septum
4-cyano-4-(phenylcarbonothioylthio)pentanoic acid (1.5 g; 5.37 mmol)
and triethylamine (1.50 mL; 10.74 mmol) were dissolved in 50 mL of
anhydrous THF. Via syringe through the septum pentafluorophenyl
trifluoroacetate (1.84 mL; 10.74 mmol) was slowly added and the reaction
continued for 18h at room temperature. The solvent was evaporated in
vacuo and the residue was dissolved in 50 mL dichloromethane. After
extraction with water (three times) the organic layer was dried over
magnesium sulfate, filtered and evaporated to dryness in vacuo. The
crude product was purified by column chromatography with cyclohexane:
calculated by extrapolating Dapp for q = 0. By formal application of
−1 −1
Stokes law, the inverse z-average hydrodynamic radius is Rh = 〈Rh
〉
.
ethylacetate 15:1 (Rf
4-cyano-4-((phenylcarbonothioyl)thio)pentanoate as pink solid (1.68 g;
3.77 mmol; 70%).
= 0.3) as eluent yielding pentafluorophenyl
z
Single-angle DLS measurements were performed using a Malvern
Zetasizer NanoZS with a 633 nm He/Ne Laser at a fixed scattering
angle of 173°. Nanoparticles were again prepared at 0.1 g L−1 in PBS and
filtered with Pall GHP filters, 0.2 µm pore size.
1H NMR (CDCl3, 400 MHz): δ [ppm] = 7.93 (dt, J = 8.6, 1.6 Hz, 2H,
m-ArH); 7.59 (ddt, J = 8.7, 7.1, 1.2 Hz, 1H, p-ArH); 7.46–7.37 (m, 2H,
o-ArH); 3.06 (dt, J = 9.4, 6.0 Hz, 2H, CH2CH2CO), 2.77 (ddd,
J = 14.3, 9.6, 6.4 Hz, 1H, CH2CH2CO), 2.56 (ddd, J = 14.3, 9.9,
6.3 Hz, 1H, CH2CH2CO), 1.99 (s, 3H, CH3).
Cryogenic Transmission Electron Microscopy (cryoTEM): 5 µL of the
nanohydrogel solution (0.1 mg mL−1, in PBS) were applied to freshly
glow-discharged carbon grids with a copper 200 mesh (Quantfoil Micro
Tools GmbH). Excess fluid was removed by direct blotting (2.5 s) and
the grids were individually plunge-frozen in liquid ethane. Grids were
cryotransferred in liquid nitrogen using a Gatan cryoholder (model 626
DH) to a Technai T12 transmission electron microscope equipped with
a field emission electron source and operating at 120 kV accelerating
voltage. Images were recorded using a TemCam-F416 (TVIPS, Gauting,
Germany).
Synthesis of α-Mannosyl Chain Transfer Agent (Man CTA): In a round
bottom flask under argon atmosphere pentafluorophenyl 4-cyano-
4-((phenylcarbonothioyl)thio)pentanoate (388 mg; 0.87 mmol)
was dissolved in 35 mL anhydrous THF and triethylamine (139 µL;
1.00 mmol) was added. α-Mannosylated ethylene glycol amine spacer
(237 mg; 0.67 mmol) was dissolved in 3 mL anhydrous DMSO and
added slowly through a septum via syringe. The reaction was continued
for 18 h at room temperature. The solvent was evaporated under
reduced pressure. The residue was dissolved in 30 mL of water and
lyophilized to remove the remaining DMSO. After lyophilization the
product was purified by column chromatography. First, the non-reacted
PFP CTA was removed using petroleum ether: ethyl acetate (10:1) as
eluent. Immediately after complete removal of PFP CTA the eluent was
switched to chloroform: ethanol (3:1) and pure Man CTA (Rf = 0.3) could
be isolated as a pink oil (282 mg; 0.46 mmol; 67%).
Zeta Potential: Samples were prepared at 1 g L−1 in Millipore water
or 10 mm HEPES buffer and used as such for uncomplexed samples
or they were incubated with siRNA for 30 min for measurements of
complexed nanoparticles. Samples were transferred into folded capillary
zeta cells (DTS 1070, Malvern) and measured using a Malvern Zetasizer
NanoZS. The effective zeta potential was determined by applying the the
Smoluchowski equation. Data is given as mean standard deviation of
three repeated measurements per sample.
1H NMR (CDCl3, 400 MHz): δ [ppm] = 7.90 (d, J = 7.3 Hz, 2H,
m-ArH); 7.55 (t, J = 7.4 Hz, 1H, p-ArH); 7.38 (t, J = 7.8 Hz, 2H, o-ArH);
Agarose Gel Electrophoresis: Complexation experiments were
performed with scrambled (random sequence) siRNA which was
©
1900162 (9 of 12)
Macromol. Biosci. 2019, 1900162
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
1H NMR (CDCl3, 400 MHz): δ [ppm] = 4.93 (s, 1H, Man-C1),
4.14–4.02 (m, 2·XnH, COOCH2CH2O), 3.72–3.62 (m, 8·XnH,
OCH2CH2O), 3.59–3.52 (m, 2·XnH, CH2CH2OCH3),
3.42–3.33 (m, 3·XnH, OCH3), 2.71–1.91 (m, 2·XnH, CH2
backbone PFPMA), 1.91–1.73 (m, 2·XnH, CH3 backbone MEO3MA),
1.72–1.14 (m, 3·XnH, CH3 backbone PFPMA), 1.14–0.68 (m, 3·XnH,
CH3 backbone MEO3MA).
6.97 (t, J = 4.6 Hz, 1H, NH-CO); 4.89 (s, 1H, Man-C1); 4.02–3.82 (m,
4H, Man-C2-C5); 3.77 (d, J = 10.9 Hz, 2H, Man-C6); 3.62 (d, J = 3.4 Hz,
12H, OCH2CH2O); 3.56 (t, J = 5.2 Hz, 2H, NHCH2CH2O);
3.48–3.39 (m, 2H, NHCH2CH2O); 2.60–2.34 (m, 4H,
NHCOCH2CH2Cquat; 1.93 (s, 3H, CH3).
13C NMR (CDCl3, 100 MHz): δ [ppm] = 222.82 (CS); 171.09
(CO); 144.65 (ArCquat); 133.18 (p-ArC); 128.73 (o-ArC); 126.83
(m-ArC); 118.98 (CN); 100.17 (Man-C1); 74.01–68.39 (m, Man-
C2, C3, C5 +CH2OCH2CH2); 66.97 (Man-C4); 66.82
(CH2O-Man); 61.44 (Man-C6); 46.27 (Cquat); 39.61 (NHCH2); 34.29
(NHCOCH2CH2Cquat); 31.71(NHCOCH2CH2Cquat); 24.17
(CH3).
Synthesis of α-Mannosyl Modified Cationic Nanohydrogel Particles: The
synthesis of non-functionalized cationic nanohydrogel particles followed
the previously reported procedure.[25,26] For α-mannosyl modified
particles the procedure was performed in analogy and is described below
for fluorescent labeled particles with Oregon Green 488. For in vivo
biodistribution experiments the particles were synthesized under similar
conditions using IR800 RS cadaverine as fluorescent near-infrared dye.
All steps were carefully carried out under argon atmosphere. In
ESI-MS: [M+H]+ = 617.2197 g mol−1; [M+Na]+ = 639.2016 g mol−1
Synthesis of Mannose-poly(tri(ethylene glycol)methyl ether methacrylate)
(Man-P(MEO3MA)): In a Schlenk tube equipped with a stir bar MEO3MA
(1 g; 4.31 mmol), α-mannosylated CTA (92.4 mg; 0.15 mmol) and
2,2-azobis(4-methoxy-2,4-dimethylvaleronitrile) (AMDVN as initiator)
(9.2 mg; 0.03 mmol) were dissolved in anhydrous dioxane (2 mL).
After three freeze–pump–thaw cycles the tube was immersed in an oil
a
round bottom flask Man-P(MEO3MA)18-b-P(PFPMA)30 (40 mg;
3.24 or 97.2 µmol reactive ester) as α-mannosyl end group modified
amphiphilic block copolymer was dissolved in anhydrous DMSO (4 mL;
10 g L−1) supported by sonication for 1 h. After transfer of the clear,
slightly foamy solution into a Schlenk tube equipped with a stir bar
triethylamine (7.3 µL; 10% of total 0.54 mmol) and Oregon Green 488
cadaverine (89.8 µL of a 2.5 g L−1 stock solution in anhydrous DMSO;
0.45 µmol) were added and the reaction vigorously stirred at room
temperature for 16 h. Afterward, spermine (9.2 mg; 45.2 µmol) and
additional triethylamine (68 µL; 90% of total 0.54 mmol) were added
for crosslinking, and the reaction was continuously stirred at 50 °C in
an oil bath for 24 h. Complete conversion of the reactive ester units
was confirmed by 19F NMR. To assure conversion of all PFPMA units
even below the NMR detection limit additional spermine (18.3 mg;
90.5 µmol) was added and the reaction continued for 18 h. Subsequently
the reaction mixture was dialyzed against Millipore water for several
days (with water exchange twice per day) to remove all small molecular
by-products. After lyophilization of the aqueous solution the cationic
nanohydrogel particles were obtained as voluminous, slightly orange
powder (25.9 mg; 2.63 µmol; 81%), which could be easily re-dissolved in
aqueous buffer at given concentrations.
Cell Culture: HepG2 human hepatocellular carcinoma cells, RAW
264.7 macrophages, and SVEC4-10 small vascular endothelial cells
were cultured in Dulbecco’s modified Eagle’s medium (Sigma-Aldrich,
Taufkirchen, Germany) supplemented with 10% fetal bovine serum
(FBS), 1% penicillin/streptomycin, 1% l-glutamine in 5% CO2 at 37 °C.
The medium was routinely changed every 2 days and the cells were
separated by cell scraping (only for RAW macrophages) or with trypsin-
EDTA (0.05%) plus phenol red from (Life Technologies, Darmstadt,
Germany) before reaching confluency.
As in vitro model for liver macrophages, myeloid precursor cells were
harvested from bone marrow of mice and differentiated in the presence of
macrophage colony-stimulating factor (M-CSF), a lineage-specific growth
factor, into primary macrophages. Once matured to macrophages, they exert
numerous phenotypic characteristics of monocyte-derived macrophages.
They exhibit a homogenous and reproducibly elevated expression of
CD206 (macrophage mannose receptor) when polarized toward M2
(Figure 4B). While M1 macrophages, as the anti-fibrotic phenotype, and
native M0 macrophages exhibit a low or intermediate expression of CD206,
respectively (Figure 4B). Thus, primary macrophages were an ideal cell
model to study mannose receptor mediated cellular uptake of ManNP.
Primary BMDMs were harvested from 6 to 9 weeks old Balb/c mice as
described in detail elsewhere.[42] All animal studies were approved by the
local ethics committee on animal care (number 23 177-07/G 17-1-030,
Government of Rhineland Palatinate, Germany). In brief, mice were
sacrificed by cervical dislocation, and femurs and tibias were isolated
under sterile conditions. A 21G needle and a 10 mL syringe were
used to flush out the marrow into cold PBS substituted with 2% heat
inactivated FBS and 1% penicillin/streptomycin. Subsequently, single
cell suspensions were passed through a 70 µm cell strainer to exclude
cell clumps and other impurities. Contaminating erythrocytes were
removed with red cell lysis buffer (eBioscience, San Diego, USA). Cells
were resuspended in Iscove’s Modified Dulbecco’s Medium (IMDM),
1
bath at 37 °C and stirred for 5 h. Hourly H NMR control showed 45%
monomer conversion after 5 h and the resulting polymer was isolated
by precipitation in n-hexane. After centrifugation the polymer was
re-dissolved in dioxane and the precipitation step was repeated three
times. The polymer was dried overnight in vacuo at room temperature
affording Man-P(MEO3MA)18 (336 mg; 47%) as a pink oil.
SEC (THF(RI), PS-Std.): Mn = 3200 g mol−1; Mw = 3800 g mol−1
;
PDI = 1.20.
1H NMR (CDCl3, 400 MHz): δ [ppm] = 7.93–7.81 (m, 2H, m-ArH),
7.58–7.47 (m, 1H, p-ArH), 7.42–7.30 (m, 2H, o-ArH), 4.92 (s, 1H,
Man-C1), 4.19–4.02 (m, 2·XnH, COOCH2CH2O), 3.69–3.60
(m, 8·XnH, OCH2CH2O), 3.55 (t, J = 4.5 Hz, 2·XnH, CH2C
H2OCH3), 3.38 (s, 3·XnH, OCH3), 2.02–1.73 (m, 2·XnH, CH2
backbone), 1.13–0.74 (m, 3·XnH, CH3 backbone).
Synthesis of α-Mannosyl-poly(tri(ethylene glycol)methyl ether methacrylate)-
block-poly(pentafluorophenyl methacrylate) (Man-P(MEO3MA)-b-P(PFPMA)):
In a Schlenk tube equipped with a stir bar PFPMA (454 mg; 1.80 mmol),
Man-P(MEO3MA)18 as macro CTA (334 mg; 66.8 µmol) and AMDVN
(4.1 mg; 13.4 µmol) were dissolved in anhydrous dioxane (1.5 mL). After
three freeze–pump–thaw cycles the tube was immersed in an oil bath
at 37 °C and stirred for 15 h. 1H NMR control showed 77% monomer
conversion and the resulting polymer was isolated by precipitation in
n-hexane. After centrifugation the polymer was re-dissolved in dioxane and
the precipitation step was repeated three times. The polymer was dried
overnight in vacuo at room temperature affording Man-P(MEO3MA)18-b-
P(PFPMA)34 (534 mg; 40.0 µmol; 60%) as a pink solid.
SEC (THF(RI), PS-Std.): Mn = 6800 g mol−1; Mw = 8900 g mol−1
PDI = 1.31.
1H NMR (CDCl3, 400 MHz):
[ppm] = 7.96–7.77 (m, 2H,
;
δ
m-ArH), 7.59–7.49 (m, 1H, p-ArH), 7.45–7.28 (m, 2H, o-ArH), 4.92
(s, 1H, Man-C1), 4.12–3.99 (m, 2·XnH, COOCH2CH2O),
3.72–3.58 (m, 8·XnH, OCH2CH2O), 3.57–3.49 (m, 2·XnH,
CH2CH2OCH3), 3.41–3.26 (m, 3·XnH, OCH3), 2.77–2.22
(m, 2·XnH, CH2 backbone PFPMA), 2.22–1.60 (m, 2·XnH, CH3
backbone MEO3MA), 1.60–1.13 (m, 3·XnH, CH3 backbone PFPMA),
1.13–0.55 (m, 3·XnH, CH3 backbone MEO3MA).
Removal of Dithiobenzoate Endgroup for α-Mannosyl-poly(tri(ethylene
glycol)methyl ether methacrylate)-block-poly(pentafluorophenyl methacrylate)
((Man-P(MEO3MA)-b-P(PFPMA)): In a Schlenk tube equipped with a stir
bar under argon atmosphere Man-P(MEO3MA)18-b-P(PFPMA)34 (530 mg;
39.7 µmol) was dissolved in 5 mL of anhydrous dioxane and AMDVN
(376 mg; 1.22 mmol) was added. The reaction was stirred for 20 h at 37 °C
in an oil bath while the solution became colorless. The resulting polymer
was precipitated in ice-cold n-hexane: ethyl ether (2:1) and centrifuged. After
re-dissolving the precipitation step was repeated three times. Afterward
the polymer was dried overnight in vacuo affording Man-P(MEO3MA)18-b-
P(PFPMA)30 (437 mg; 35.4 µmol; 89%) as a colorless solid.
SEC (THF(RI), PS-Std.): Mn = 6800 g mol−1; Mw = 9800 g mol−1
PDI = 1.44.
;
©
1900162 (10 of 12)
Macromol. Biosci. 2019, 1900162
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
10% FBS, 1% penicillin/streptomycin, 1% L-glutamine containing
25 ng mL−1 monocyte colony-stimulating factor (M-CSF, ImmunoTools,
Friesoythe, Germany). 5 million BMDM in 10 mL media were seeded into
bacterial culture plate (Sigmal-Aldrich/Merck, Darmstadt, Germany),
to facilitate better detachment and improved viability of the cells
during passaging. On day 3, half of the media in the petri dishes was
replaced with fresh IMDM containing the same supplements. Immature
BMDM were incubated for an additional 4 days M-CSF for further
differentiation into mature primary macrophages. For polarization
toward an M2- or M1-phenotype, primary macrophages were incubated
with fully supplemented IMDM containing 20 ng mL−1 IL-4 and IL-13, or
50 ng mL−1 interferon-γ and 0.1 µg mL−1 LPS (all from ImmunoTools,
Friesoythe, Germany), respectively, for 24 h. For sub-culturing and cell
harvest, cells were preincubated in PBS (calcium and magnesium free),
supplemented with 10 mm EDTA, for 15 min on ice and subsequently
gently detached by cell scraping.
In Vitro Cytotoxicity: The MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium-bromide) assay was used to assess the cytotoxic
effects of scrambled siRNA-loaded (non)-ManNP on RAW-macrophages
and 3T3-fibroblasts as described before.[30] In brief, cells were seeded
into 96-well plates at a density of approximately 6000–7000 cells per
well and incubated in supplement containing medium. After 24 h
increasing concentrations of siRNA/NP (10:1 weight-to-weight ratio
of (non)-ManNP corresponding to 5, 10, 20, 40, 60, 80, 100, 200, and
300 nm siRNA were added. After 48 h of incubation, 20 µL of MTT in
PBS (4 mg mL−1) was added to the culture medium and incubation
was carried out for another 5 h. Afterward, the medium was carefully
removed and 150 µL DMSO was added. Plates were placed on an
orbital shaker for 1 h under exclusion of light for homogenization.
Finally, the OD at 570 nm were measured on an Infinite M200Pro
spectrofluorometer (TECAN, Männerdorf, Switzerland). As reference the
OD570 of nanoparticle-treated cells was normalized by the OD570 of
cells exposed to equal volumes of PBS alone.
In Vitro Cellular Uptake by Fluorescent Activated Cell Sorting Flow
Cytometry (FACS): Primary macrophages, RAW macrophages,
hepatocellular carcinoma HepG2 and endothelial SVEC4-10 cells were
seeded in 6-well culture plates at a density of 700000 cells per well.
Primary macrophages were incubated with interferon-γ/LPS or IL-4/13
for polarization toward an M1 or M2 phenotype, respectively, as
described above. 24 h after cells were seeded, plates were washed with
PBS and incubated with a final concentration of 25, 50, and 100 nm Cy5-
labeled scrambled siRNA complexed with (non)-ManNP (10:1 weight-to-
weight ratio of non-ManNP:siRNA) for 1 h at 37 °C.
High-performance liquid chromatography purified Cy5 labeled
scrambled siRNA (Cy-5 NIR dye attached to the 5′ by a C6 amino spacer)
with the following sequence was purchased from Dharmacon (Freiburg,
Germany):
5′-Cy5 GCAUCUGGCUUAAGGUGAAUU-3′
5′-PGUAUCUCUUCAUAGCCUUAUU-3′
above and seeded in a 6-well plate at a density of 700000 cells per well.
Cells were incubated with mannan (2 g mL−1) for 1 h at 37 °C followed
by addition of 100 nm Cy5-labeled scrambled siRNA complexed with
NonNP/ManNP (10:1 weight-to-weight ratio of NP:siRNA). After 1 h
incubation, culture medium was removed, and the cells were repetitively
washed with cold PBS, detached as described above and stained with a
viability dye (Fixable Viability Dye eFluor 506, eBioscience, San Diego,
USA) according to the manufacture’s protocol and stored at 4 °C under
the exclusion of light. On the same day, cells were subjected to flow
cytometry (BD FACS Canto II, BD Bioscience, Mississauga, Canada).
Compensation was performed by BD FACSDiva software version 7.0
with single stained OneComp eBeads from eBioscience (Frankfurt,
Germany). 50000 cells were measured per sample. Further data analysis
was carried out using open source Flowing Software 2.5.0 (Perttu Terho,
Turku Centre for Biotechnology, Finland).
In Vitro Target Knockdown: Anti-CSFR-1 (colony-stimulating factor 1
receptor) siRNA with the following sequence
5′-GCAUCUGGCCUUAAGGUGAAUU-3′
5′-PUUCACCUUAAGCCAGAUGCUU-3′
was purchased from Dharmacon (Freiburg, Germany). As off target
control the following siRNA (scrambled siRNA) siGENOME non-
targeting siRNA #2 (Dharmacon, Freiburg, Germany) was applied
5′-UAAGGCUAUGAAGAGAUACUU-3′
5′-GUAUCUCUUCAUAGCCUUAUU-3′.
The specific modification pattern of siGENOME is an intellectual
property of Dharmacon.
Primary BMDM were seeded in 12-well culture plates at a density of
250000 cells per well and allowed to adhere overnight. 24 h before the
knockdown cells were polarized toward M1- or M2-phenotype as described
above. They were then incubated with final concentrations of 25, 50, and
100 nm anti-CSFR1 or scrambled siRNA complexed to ManNP or non-
ManNP, respectively, (mass ration NP:siRNA 10:1) for 48 h at 37 °C.
RNA Isolation and Quantitative RT-PCR: After 48 h incubation of
primary macrophages with anti-CSFR-1 siRNA loaded in (non)-ManNP
(described above), plates were carefully evacuated, and cells were
homogenized in 1 mL per well Ribozol (VWR, Wiesbaden, Germany).
The cells’ RNA was isolated by guanidinium thiocyanate-phenol-
chloroform extraction[43] as reported before.[29,30]
For qPCR, 1 µg of total RNA was reverse-transcribed into cDNA using
the qScript cDNA SuperMix (Quantas, Beverly, USA). The following
SYBR green primers for CSFR-1 and GAPDH were synthesized by
Eurofins (Mannheim, Germany):
GAPDH: forward 5′-AGGTCGGTGTGAACGGATTT-3′
reverse 5′-GGGGTCGTTGATGGCAACA-3′
CSFR-1: forward 5′-TGTCATCGAGCCTAGTGGC-3′
reverse 5′-CGGGAGATTCAGGGTCCAAG-3′
SYBR reaction mixtures were purchased from Applied Biosystems
(Darmstadt, Germany). Transcript levels of glyceraldehyde phosphate
dehydrogenase (GAPDH) were used to normalize data and to control
for RNA integrity. Samples were amplified by the SYBR technology and
analyzed using a Step One Plus sequence amplification system from
LifeTechnologies (Darmstadt, Germany). Results were calculated by the
ΔΔCT method as reported in detail before.[29,30]
To normalize for potential autofluorescence of siRNA and nanoparticles,
cells were incubated with a final concentration of 100 nm unlabeled
scrambled siRNA complexed with unlabeled (non)-ManNP (10:1 weight-
to-weight ratio of NP:siRNA) for 1 h in the cell incubator before harvest
for flow cytometry. In addition, control cells were incubated with equal
volumes of PBS for 1 h at 37 °C. After 1 h incubation, culture medium
was removed, and the cells were repetitively washed with cold PBS,
detached as described above, and stained with a viability dye (Fixable
Viability Dye eFluor 506, eBioscience, San Diego, USA) according to
the manufacture’s protocol and stored at 4 °C under the exclusion of
light. On the same day, cells were subjected to flow cytometry (BD FACS
Canto II, BD Bioscience, Mississauga, Canada). Compensation was
performed by BD FACSDiva software version 7.0 with single stained
OneComp eBeads from eBioscience (Frankfurt, Germany). 50000 cells
were measured per sample. Further data analysis was carried out using
open source Flowing Software 2.5.0 (Perttu Terho, Turku Centre for
Biotechnology, Finland).
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
Acknowledgements
N.L. and L.K. contributed equally to this work. The data of the NPs’
biological evaluation is part of Leonard Kaps’ medical doctoral thesis.
N.L. and L.K. conceived the experiments. N.L., L.K., L.N., D.S., and R.Z.
interpreted the experiments and wrote the manuscript. M.A., A.Y., M.G.,
In Vitro Mannan Blocking Study with M2 Polarized Macrophages:
Primary macrophages were polarized into M2 phenotype as described
©
1900162 (11 of 12)
Macromol. Biosci. 2019, 1900162
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
and M.D. assisted with the animal experiments. All authors are grateful
to the German Research Foundation (DFG) Collaborative Research
Center SFB 1066 (project B3+B4) for support. Meike Schinnerer and Prof.
Manfred Schmidt (Institutes of Physical Chemistry, Johannes Gutenberg
University of Mainz) are acknowledged for providing access to the DLS
set up and DLS measurements. Benjamin Weber (Institute of Organic
Chemistry, Johannes Gutenberg University of Mainz) is acknowledged for
the cryo-TEM images. Dr. Hajime Yurugi (Paul-Klein-Center for Immune
Intervention, JGU, University Medical Center) is acknowledged for
assisting in the fluorescence confocal laser scanning microscopy.
[17] A. Sica, P. Larghi, A. Mancino, L. Rubino, C. Porta, M. G. Totaro,
M. Rimoldi, S. K. Biswas, P. Allavena, A. Mantovani, Semin. Cancer
Biol. 2008, 18, 349.
[18] A. Sica, V. Bronte, J. Clin. Invest. 2007, 117, 1155.
[19] E. R. Stanley, J. Exp. Med. 1976, 143, 631.
[20] D. A. Hume, K. P. A. MacDonald, Blood 2012, 119, 1810.
[21] D. Hanahan, R. A. Weinberg, Cell 2011, 144, 646.
[22] S. M. Pyonteck, L. Akkari, A. J. Schuhmacher, R. L. Bowman,
L. Sevenich, D. F. Quail, O. C. Olson, M. L. Quick, J. T. Huse,
V. Teijeiro, M. Setty, C. S. Leslie, Y. Oei, A. Pedraza, J. Zhang,
C. W. Brennan, J. C. Sutton, E. C. Holland, D. Daniel, J. A. Joyce,
Nat. Med. 2013, 19, 1264.
[23] M. Stein, J. Exp. Med. 1992, 176, 287.
Conflict of Interest
[24] L. Nuhn, E. Bolli, S. Massa, I. Vandenberghe, K. Movahedi,
B. Devreese, J. A. Van Ginderachter, B. G. De Geest, Bioconjugate
Chem. 2018, 29, 2394.
[25] N. Leber, L. Nuhn, R. Zentel, Macromol. Biosci. 2017, 17,
1700092.
The authors declare that the research was conducted in the absence of
any commercial or financial relationships that could be construed as a
potential conflict of interest.
[26] L. Nuhn, M. Hirsch, B. Krieg, K. Koynov, K. Fischer, M. Schmidt,
M. Helm, R. Zentel, ACS Nano 2012, 6, 2198.
[27] L. Nuhn, S. Gietzen, K. Mohr, K. Fischer, K. Toh, K. Miyata,
Y. Matsumoto, K. Kataoka, M. Schmidt, R. Zentel, Biomacromole-
cules 2014, 15, 1526.
Keywords
gene knock-down, M2 macrophages, mannose targeting, nanohydrogels,
siRNA delivery
[28] L. Nuhn, S. Tomcin, K. Miyata, V. Mailänder, K. Landfester,
K. Kataoka, R. Zentel, Biomacromolecules 2014, 15, 4111.
[29] L. Kaps, L. Nuhn, M. Aslam, A. Brose, F. Foerster, S. Rosigkeit,
P. Renz, R. Heck, Y. O. Kim, I. Lieberwirth, D. Schuppan, R. Zentel,
Adv. Healthcare Mater. 2015, 4, 2809.
Received: May 8, 2019
Published online:
[30] N. Leber, L. Kaps, M. Aslam, J. Schupp, A. Brose, D. Schäffel,
K. Fischer, M. Diken, D. Strand, K. Koynov, A. Tuettenberg, L. Nuhn,
R. Zentel, D. Schuppan, J. Controlled Release 2017, 248, 10.
[31] B. Krieg, M. Hirsch, E. Scholz, L. Nuhn, I. Tabujew, H. Bauer,
S. Decker, A. Khobta, M. Schmidt, W. Tremel, R. Zentel, K. Peneva,
K. Koynov, A. J. Mason, M. Helm, Pharm. Res. 2015, 32, 1957.
[32] L. Nuhn, L. Kaps, M. Diken, D. Schuppan, R. Zentel, Macromol.
Rapid Commun. 2016, 37, 924.
[1] Nobel Assembly at Karolinska Institutet, The Nobel Prize in
Physiology or Medicine 2018. https://www.nobelprize.org/prizes/
medicine/2018/press-release/ 2018, 1–5.
[2] J. G. Egen, M. S. Kuhns, J. P. Allison, Nat. Immunol. 2002, 3, 611.
[3] T. Okazaki, S. Chikuma, Y. Iwai, S. Fagarasan, T. Honjo, Nat.
Immunol. 2013, 14, 1212.
[4] D. Schuppan, Y. O. Kim, J. Clin. Invest. 2013, 123, 1887.
[5] W. Mehal, D. Schuppan, Semin. Liver Dis. 2015, 35, 184.
[6] D. Schuppan, M. Ashfaq-Khan, A. T. Yang, Y. O. Kim, Matrix Biol.
2018, 68-69, 435.
[33] A. R. Khomutov, N. A. Grigorenko, S. G. Skuridin, A. V. Demin,
J. Vepsalainen, R. A. Casero, P. M. Woster, Russ. J. Bioorg. Chem.
2005, 31, 271.
[7] N. B. Hao, M. H. Lü, Y. H. Fan, Y. L. Cao, Z. R. Zhang, S. M. Yang,
Clin. Dev. Immunol. 2012, 2012, 1.
[34] P. J. Roth, D. Kessler, R. Zentel, P. Theato, Macromolecules 2008, 41,
8316.
[8] M. Zhang, Y. He, X. Sun, Q. Li, W. Wang, A. Zhao, W. Di, J. Ovarian
Res. 2014, 7, 1.
[35] P. J. Roth, K. T. Wiss, R. Zentel, P. Theato, Macromolecules 2008, 41,
8513.
[9] S. L. Friedman, D. Sheppard, J. S. Duffield, S. Violette, Sci. Transl.
Med. 2013, 5, 167sr1.
[36] N. Mohr, C. Kappel, S. Kramer, M. Bros, S. Grabbe, R. Zentel,
Nanomedicine. 2016, 11, 2679.
[10] O. Krenkel, F. Tacke, Nat. Rev. Immunol. 2017, 17, 306.
[11] E. Song, N. Ouyang, M. Hörbelt, B. Antus, M. Wang, M. S. Exton,
Cell. Immunol. 2000, 204, 19.
[12] A. Mantovani, S. Sozzani, M. Locati, P. Allavena, A. Sica, Trends
Immunol. 2002, 23, 549.
[37] L. Nuhn, N. Vanparijs, A. De Beuckelaer, L. Lybaert, G. Verstraete,
K. Deswarte, S. Lienenklaus, N. M. Shukla, A. C. D. Salyer,
B. N. Lambrecht, J. Grooten, S. A. David, S. De Koker,
B. G. De Geest, Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 8098.
[38] S. Bhattacharjee, J. Controlled Release 2016, 235, 337.
[39] D. A. Mann, M. Kanai, D. J. Maly, L. L. Kiessling, J. Am. Chem. Soc.
1998, 120, 10575.
[13] A. Mantovani, B. Bottazzi, F. Colotta, S. Sozzani, L. Ruco, Immunol.
Today 1992, 13, 265.
[14] C. D. Mills, K. Kincaid, J. M. Alt, M. J. Heilman, A. M. Hill,
J. Immunol. 2000, 164, 6166.
[40] P. Heller, N. Mohr, A. Birke, B. Weber, A. Reske-Kunz, M. Bros,
M. Barz, Macromol. Biosci. 2015, 15, 63.
[15] L. Beljaars, M. Schippers, C. Reker-Smit, F. O. Martinez,
L. Helming, K. Poelstra, B. N. Melgert, Front. Immunol. 2014, 5, 13.
[16] Y. Kitoh, M. Saio, N. Gotoh, N. Umemura, K. Nonaka, J. Bai,
L. Vizkeleti, D. Torocsik, M. Balazs, R. Adany, T. Takami, Int. J.
Oncol. 2011, 38, 1409.
[41] M. A. Cannarile, M. Weisser, W. Jacob, A.-M. Jegg, C. H. Ries,
D. Rüttinger, J. Immunother. Cancer 2017, 5, 53.
[42] W. Ying, P. S. Cheruku, F. W. Bazer, S. H. Safe, B. Zhou, J. Vis. Exp.
2013.
[43] P. Chomczyński, N. Sacchi, Anal. Biochem. 1987, 162, 156.
©
1900162 (12 of 12)
Macromol. Biosci. 2019, 1900162
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim