Bioreducible Poly- l -Lysine-Poly[HPMA] Block Copolymers Obtained by RAFT-Polymerization as Efficient Polyplex-Transfection Reagents

Article abstract of DOI:10.1002/mabi.201500212

Polylysine-b-p[HPMA] block copolymers containing a redox-responsive disulfide bond between both blocks are synthesized by RAFT polymerization of pentafluorphenyl-methacrylate with a macro-CTA from N_∈-benzyloxycarbonyl (Cbz) protected polylysine

Full text of DOI:10.1002/mabi.201500212

Full Paper
Bioreducible Poly-L-Lysine–Poly[HPMA] Block
Copolymers Obtained by RAFT-Polymerization
as Efficient Polyplex-Transfection Reagentsa
Kristof Tappertzhofen, Simone Beck, Evelyn Montermann, David Huesmann,
Matthias Barz, Kaloian Koynov, Matthias Bros,* Rudolf Zentel*
Polylysine-b-p[HPMA] block copolymers containing a redox-responsive disulfide bond between
both blocks are synthesized by RAFT polymerization of pentafluorphenyl-methacrylate with a
macro-CTA from Ne-benzyloxycarbonyl (Cbz) protected polylysine (synthesized by NCA
polymerization). This polylysine-b-p[PFMA] precursor block copolymer is converted to
polylysine(Cbz)-b-p[HPMA] by postpolymerization
modification with 2-hydroxypropylamine. After
removal of the Cbz protecting group, cationic poly-
lysine-b-p[HPMA] copolymers with a biosplittable
disulfide moiety became available, which can be used
as polymeric transfection vectors. These disulfide
linked polylysine-S-S-b-p[HPMA] block copolymers
show low cytotoxicity and increased transfection
efficiencies (HEK-293T cells) compared to analogous
blockcopolymers without disulfide group making
them interesting for the transfection of sensitive
immune cells.
R. Zentel, K. Tappertzhofen, S. Beck, D. Huesmann, M. Barz
Institute of Organic Chemistry, Johannes Gutenberg-University,
1. Introduction
Duesbergweg 10-14, 55128 Mainz, Germany
E-mail: zentel@uni-mainz.de
Cancer immunotherapy is considered
a promising
K. Koynov
approach, aimed to recognize and eliminate malignant
cells by exploiting the effectiveness and potency of the
adaptiveimmunesystem.[1–3] Besidesclassicalvaccination,
the activation of dendritic cells (DCs) offers a promising
strategy to induce cellular immune response.[4–6]
Max Planck Institute for Polymer Research, Ackermannweg 10,
55128 Mainz, Germany
M. Bros, E. Montermann
Department of Dermatology, University Medical Center of the
Johannes Gutenberg-University, Langenbeckstrasse 1, 55131
Mainz, Germany
Especially, the transcriptional activation of DCs with
antigen-encoding pDNA seems to be a very powerful
method.[7,8] It facilitates the parallel activation of both T
helper (CD4þ) and cytotoxic T (CD8þ) cells.[9,10] Moreover,
pDNA vectors display a useful platform to incorporate
additional moieties such as the DC-specific fascin promoter,
which ensures that transcription of the encoding gene
occurs selectively in DCs.[11,12]
E-mail: mbros@uni-mainz.de
S. Beck
MAINZ Graduate School of Excellence (Materials Science in
Mainz), Johannes Gutenberg-University, Staudingerweg 9, 55128
Mainz, Germany
aSupporting Information is available from the Wiley Online Library or
from the author.
Macromol. Biosci. 2016, 16, 106–120
106
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
DOI: 10.1002/mabi.201500212

Bioreducible Poly-L-Lysine–Poly[HPMA] Block Copolymers . . .
www.mbs-journal.de
It gets only activated in DCs and ensures, thus, the
selective read-out only in these cells. In addition, in case of
DCs, onlyasfewasabout1000DCsneedtobetransfectedto
induce a profound immune response in mice.[13] However,
itisstillchallengingtofindappropriatetransfectionvectors
for sensitive immune cells and especially for in vivo
applications.
substituted (meth) acrylates[35–37] in terms of cytotoxicity.
This is especially important when dealing with sensitive
immune cells. Thus PEG, or poly[sarcosine] derivatives of
polylysine seem most promising to transfect immune
cells.[38,39] In this context, the use of clinically investigated
poly[2-(hydroxypropyl methacrylamide)] (p[HPMA]) as a
biocompatible, non-ionic hydrophilic corona appears quite
reasonable too, since it features low immunogenicity[40,41]
and has been used in clinical studies for anticancer
vaccines.[42] Furthermore, p[HPMA] can be modified with
targeting ligands like mannose or aDEC-205 antibodies to
address DCs.[24,25]
Several drawbacks of using recombinant virus as trans-
fection vectors, such as a risk in therapeutical applications,
limitation in passenger DNA size, and difficulties in large
scale production,[14–16] have led to intensive investigations
on liposomal, lipidic, and polymeric transfection agents in
the last decades. Especially polymer-based systems, more
precisely polyplexes derived from pDNA and block copoly-
mers consisting of a cationic block for pDNA complexation
and a hydrophilic, biocompatible block for efficient
shielding, seem most promising. This has nicely been
shown by the innovating work of Wagner et al., and
Kataoka and co-workers.[17–20] A biocompatible corona of
the polyplexes, e.g., from PEG or p[HPMA] for stealth like
properties, is thereby indispensable for systemic in vivo
applications to (i) avoid unspecific interactions with serum
proteins, to (ii) prevent extra- and intracellular degradation
of the pDNA, and to (iii) prevent unspecific immune
response due to stimulation of toll like receptors.[21–23] A DC
specific uptake can be achieved by the incorporation of
targeting ligands to the polyplex’ surface to induce receptor
mediated uptake. For DCs, a specific targeting of polymeric
systems has been shown by the use of mannose or the
monoclonal antibody aDEC-205.[24–26]
The combination of reversible addition-fragmentation
chain transfer polymerization (RAFT-polymerization) and
reactive ester monomers allows the synthesis of multi-
functional p[HPMA] based block copolymers of various
topologies.[43,44] By this synthetic approach, we previously
obtained polylysine-b-p[HPMA] block copolymers, which
efficiently form polyplexes in the presence of pDNA and
display reasonable transfection efficiencies at minimal
cytotoxicity.[45] To increase transfection efficiencies, the
incorporation of redox cleavable groups—like disulfide
groups—seems suitable.
To achieve this, polylysine with carboxybenzyl (Cbz)
blocked amino groups was selectively modified at the N-
terminus with a chain transfer agent (CTA) including a
disulfide bond. Then the synthesis of disulfide-linked
polylysine-b-p[HPMA] block copolymers was realized via
RAFT-polymerizatzion, that is—to some extent—compat-
ible with disulfid bonds.[46,47] The disulfide containing
polylsine-CTAcouldbesuccessfullyusedforRAFT-polymer-
ization of pentafluorophenyl methacrylate. The resulting
block copolymers were finally converted to the desired
polylysine-S-S-b-p[HPMA] with disulfide bonds. The for-
mation and physicochemical characterization of the
polyplexes as well as in vitro studies with HEK-293T cells
to ascertain cellular uptake, transfection efficiencies, and
toxicity are carefully investigated to analyze the influence
of the bioreducible disulfide group.
To improve the intracellular pDNA release, stimulus-
responsive groups between the cationic core and the
hydrophilic corona are needed. In this context, chemical
moieties that respond to intracellular pH changes in the
endosome or which are sensitive to the cytoplasmatic
redox potential are highly suitable.[27,28] In the latter
case, redox-stimuli responsive disulfide bonds seem
applicable to achieve intracellular deshielding of poly-
plexes. They are quite stable under extracellular con-
ditions. Intracellulary, however, they are rapidly cleaved
due to a 50- to 1 000-fold increased concentration of
glutathione.[29–31]
2. Experimental Section
Especially in antigen presenting cells (APCs) like DCs
the conditions for reductive splitting are already fulfilled
in the endosome.[32] The removal of the shielding corona
leads potentially to enhanced interactions of the cationic
part of the polyplex with the endosomal membrane,
which finally results in a facilitated endosomal escape.
Therefore, disulfide groups are attractive to increase
intracellular release of cargo, particularly of pDNA from
polyplexes.[33,34]
2.1. Materials
All chemicals (reagent grade) were obtained from Sigma–Aldrich
(Deisenhofen, Germany). Chemicals were used without further
purification unless otherwise indicated. Trifluoroethanol (TFE) was
purchased from Acros Organics (Niederau, Germany). Protected L-
lysine(NH2-L-Lys(Cbz)-OH wasobtainedfrom Orpegen(Heidelberg,
Germany). Oregon green (OG) cadaverine 488 was obtained from
Invitrogen (Carlsbad, CA). Pentafluorophenol was purchased from
Fluorochem (Derbyshire, UK). Dioxane and tetrahydrofuran used
for synthesis of polymer were freshly distilled from sodium. 2,2-
Azobis(isobutyronitrile) (AIBN) was recrystallized from diethyl
ether, and was stored at ꢀ20 8C. Dialysis was performed using
Due to their biodegradability, peptidic polylysine as
cationic core seems advantageous in comparison to poly
[ethylenimine] (PEI), poly[amidoamidine] (PAMAM) and
Macromol. Biosci. 2016, 16, 106–120
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
107
www.MaterialsViews.com

K. Tappertzhofen et al.
www.mbs-journal.de
Spectra/Por 3 membranes (MWCO 14 000 or 25 000 g ꢁ molꢀ1
)
filtered product, and the mixture was stirred for 30 min at room
temperature. This procedure was repeated twice, yielding 6.5 g
(28.4 mmol, 80%) of 2-((2-nitrophenyl) disulfanyl) ethanamine as a
yellow solid.
obtained from Carl Roth (Karlsruhe, Germany). Sephadex Hi Trap
desalting columns were purchased from GE Healthcare (Buc,
France). The vector pGL3-Basic (agarose gel electrophoresis, FCS-
measurements, CLSM) was obtained from Promega GmbH
(Mannheim, Germany). pDNA encoding for Enhanced Green
Fluorescent Protein (pEGFP-N1; transfection efficiency and toxicity
by FACS) was purchased from Clonetech (Mountain View, CA). Gel
Red was purchased from VWR (Darmstadt, Germany).
1
300 MHz H-NMR (dmso-d6): [d/ppm] ¼ 8.33–8.27 (m, 2H), 8.24
(br. s, 2H), 7.95–7.89 (t, 1H), 7.59–7.54 (t, 1H), 3.05 (s, 4H).
FD-MS: [m/z] ¼ 230.0 ([M þ H], calc. 230.0[M þ H]).
2.5. Synthesis of 3-((2-Aminoethyl) disulfanyl)
propanoic Acid
2.2. Characterization
The synthesis of 3-((2-aminoethyl) disulfanyl) propanoic acid
was adapted from the literature.[50] In brief, 4 g (17.4 mmol, 1 eq.)
of 2-((2-nitrophenyl)-disulfanyl) ethanamine and 1.84 g
(17.4 mmol, 1 eq.) of 3-mercaptopropanoic acid were dissolved
in 50 mL of water in a round bottom flask. 4.8 mL (34.8 mmol, 2
eq.) triethylamine were added and the red colored reaction
mixture was stirred for 30 min at room temperature. Afterward,
the solution was acidified (pH 5) using 1M hydrochloric acid
(change in color from red to yellow) and extracted five times with
ethyl acetate. The combined aqueous phases were evaporated
under reduced pressure to obtain the colorless crude product,
which was purified by flush column chromatography (EtOAc/
AcOH/H2O 8:2:1; Rf ¼ 0.28). After additional recrystallization
from a solution of 90% n-pentanol in water, 580 mg (2.5 mmol,
14.5%) of 3-((2-aminoethyl) disulfanyl) propanoic acid were
obtained as a colorless solid.
1H-, 19F-NMR spectra were obtained at 300 or 400 MHz using a FT
spectrometer from Bruker (Billerica, MA) and analyzed using
MestReNova 6.0.2. IR spectra were recorded on a Bruker FT/IR-4100
using an ATR unit. SEC of hydrophilic HPMA-copolymers was
performed in hexafluoroisopropanol (HFIP) containing 3 g ꢁ Lꢀ1
potassium trifluoroacetate as eluent and with the following
components: column packed with modified silica (PFG columns
̊
particle size: 7 mm; porosity: 100 and 1 000 A, respectively),
refractive index detector G1362A RID and UV-detector UV-2075
plus from Jasco. Calibration was done using PMMA standards (PSS,
Mainz, Germany). The flow rate was set to 0.8 mL ꢁ minꢀ1 at a
temperatureof40 8C. AllhydrophilicHPMApolymerswerepurified
by preparative SEC chromatography using Sephadex Hi Trap
desalting columns as eluent with the following parts: pump (pU-
2086 Plus series), UV/Vis detector (UV-2077 Plus), and a Jasco RI-
detector (Jasco RI 2031 Plus series) from Jasco. The flow rate was set
to 1.0 mL ꢁ minꢀ1. All elution diagrams were evaluated with PSS
WinGPC.
300 MHz 1H-NMR (D2O): [d/ppm] ¼ 3.37–3.31 (t, 2H), 2.99–2.92 (q,
4H), 2.70–2.66 (t, 2H).
ESI-MS: [m/z] ¼ 182.04 ([M þ H], calc. 182.02 [M þ H]).
IRy[cmꢀ1]:2950(br),2619(w),2350(w),2031(w),1698(s),1620
(s), 1 510 (s), 1 407 (vs), 1 247 (s), 1 092 (s), 947 (w), 903 (w), 772 (w),
656 (w).
2.3. Synthesis of 4-Cyano-4-((thiobenzoyl) sulfanyl)
pentanoic Acid (Acid CTA)
mp: 152.8 8C.
4-Cyano-4-((thiobenzoyl) sulfanyl) pentanoic acid was synthesized
according to the literature.[48] 300 MHz 1H-NMR (CDCl3): [d/
ppm] ¼ 7.92–7.90 (d, 2H), 7.59–7.55 (t, 1H), 7.42–7.38 (t, 2H),
2.81–2.40 (m, 4H), 1.94 (s, 3H).
2.6. Synthesis of 3-((2-(4-Cyano-4-
((phenylcarbonothioyl) thio) pentanamido)-ethyl)-
disulfanyl) propanoic Acid (Disulfide CTA)
Synthesis of Pentafluorophenyl-(4-phenylthiocarbonylthio-4-
cyanovalerate) (PFP-CTA). Pentafluorophenyl-(4-phenylthiocarbo-
nylthio-4-cyanovalerate) was synthesized according to the liter-
ature.[49] 300 MHz 1H-NMR (CDCl3): [d/ppm] ¼ 7.95–7.92 (d, 2H),
7.61–7.57 (t, 1H), 7.44–7.30 (t, 2H), 3.10–2.51 (m, 4H), 1.99 (s, 3H).
376 MHz 19F-NMR (CDCl3): [d/ppm] ¼ ꢀ152.85 (d, 2F), ꢀ157.60 (t,
1F), ꢀ162.21 (t, 2F).
The disulfide CTA was obtained by reacting 370 mg (0.83 mmol, 1.3
eq.) PFP-CTA, 0.22 mL (1.6 mmol, 2.5 eq.) of triethylamine and
116 mg (0.64 mmol, 1 eq.) of 3-((2-aminoethyl)-disulfanyl) prop-
anoic acid in 10 mL of THF. The reaction was stirred at room
temperature for 4 h under argon atmosphere and light exclusion.
The solvent was removed by evaporation under reduced pressure
and the crude was dissolved in 50 mL of dichloromethane. The
solution was extracted 3 times with pure water and once with
brine. Theorganicphasewasdriedovermagnesiumsulfateandthe
solvent was removed under reduced pressure. Further purification
was carried out via flash chromatography (chloroform/ethanol/
acetic acid 40:1:0.1; Rf ¼ 0.30) yielding 70 mg (0.16 mmol; 25%) of
the disulfide CTA as a red solid.
2.4. Synthesis of 2-((2-Nitrophenyl) disulfanyl)
ethanamine
4 g (355 mmol; 1 eq) of 2-mercaptoethylamine hydrochloride and
10 g (53 mmol, 1.5 eq.) of 2-nitrophenylsulfanyl chloride were
suspended in 100 mL of glacial acetic acid and stirred for 3.5 h at
70 8C. The solvent was evaporated under reduced pressure and the
crude product was dissolved in 12 mL of dimethylformamide
(DMF).Afterdilutingwith100 mLofchloroformthedesiredproduct
crystallizedasayellowsolid. 50 mLofchloroformwereaddedtothe
300 MHz 1H-NMR (CDCl3): [d/ppm] ¼ 7.92–7.89 (d, 2H,), 7.59–7.55
(t, 1H), 7.42–7.37 (t, 2H), 3.64–3.58 (q, 2H), 2.98–2.94 (t, 2H), 2.85–
2.77 (m, 4H), 2.67–2.35 (m, 4H), 1.93 (s, 3H).
ESI-MS: [m/z] ¼ 465.06 ([M þ Na], calc. 465.05 [M þ Na]); 907.14
([M þ M þ Na], calc. 907.10 [M þ M þ Na]).
Macromol. Biosci. 2016, 16, 106–120
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
108
www.MaterialsViews.com

Bioreducible Poly-L-Lysine–Poly[HPMA] Block Copolymers . . .
www.mbs-journal.de
3.3 mL of abs. DMF. Afterward, 5.73 mg (0.036 mol) of the initiator
neopentylammonuim tetrafluoroborate were added through a
stocksolutionin DMF. Theresultingsolutionwasstirred3d at40 8C
under a constant nitrogen flow. The polymer was precipitated in
cold ethyl ether, dissolved in dioxane/water mixture andlyophi-
lized yielding 241 mg of the polymer as a colorless powder.
400 MHz 1H-NMR (dmso-d6): [d/ppm] ¼ 8.49–7.89 (m, 30H), 7.26–
7.16 (m, 180H), 4.97–4.94 (m, 61 H), 4.33–3.66 (m, 30H), 2.93 (br. s,
61H), 2.04–1.15 (m, 180H), 0.82–0.79 (d, 9H).
2.7. Synthesis of 3-((2-(4-Cyano-4-
((phenylcarbonothioyl) thio) pentanamido) ethyl)-
disulfanyl) propanoic Acid-pentafluorophenyl-ester
(Disulfide-PFP-CTA)
A round bottom flask was loaded with 60mg (0.13 mmol, 1 eq.) of 3-
((2-(4-cyano-4-((phenylcarbonothioyl) thio) pentanamido) ethyl)
disulfanyl) propanoic acid in 5 ml of abs. THF and 37mL (0.27 mmol,
2 eq.)of triethylamine. Underargon atmosphere 46mL (0.27 mmol, 2
eq.)of pentaflourophenyl triflouroacetate were added dropwise and
the reaction was stirred for 4 h at room temperature under light
exclusion. The solvent was removed by evaporation under reduced
pressure, the crude was dissolved in 25mL of dichloromethane and
extracted with water. The organic phase was driedover magnesium
sulfateandthesolventwasremovedbyusingtherotaryevaporator.
Further purification was conducted via flash chromatography
(chloroform/ethanol/acetic acid 40:1:0.1%; Rf ¼ 0.43) yielding
74mg (0.12 mmol, 94%) of the disulfide-PFP-CTA as a red solid.
300 MHz 1H-NMR (CDCl3): [d/ppm] ¼ 7.92–7.89 (d, 2H), 7.59–7.54
(t, 1H), 7.42–7.37 (t, 2H), 3.66–3.60 (q, 2H), 3.15–3.01 (m, 4H), 2.86–
2.81 (t, 2H), 2.67–2.38 (m, 4H), 1.94(s, 3H).
2.12. Synthesis of Polylysine-Chain Transfer Agents
(Polylysine-CTAs)
Inatypicalreactionprocedure500 mg(0.063mol,1eq.)ofHOOC-poly
[Lys(Ne-Cbz)]30-NH2were dissolved in dry NMP. In a separate flask
55.6mg (0.13 mmol, 2 eq.) of pentafluorophenyl-(4-phenylthiocar-
bonylthio-4-cyanovalerate) (PFP-CTA)and 55.8 mg (0.25 mmol, 4 eq.)
1
1
6
6
of N , N , N , N -tetramethylnaphthalene-1,8-diamine(‘‘proton
sponge’’) were dissolved in 1.5 mL of dry NMP and added to the
polylysine with free amino terminus. The resulting mixture was
stirred overnight under argon atmosphere in the dark. The NMP was
removed under reduced pressure, 1 mL of TFE was added and the
polymer was precipitated in cold ethyl ether. The product was
dissolved with 5 mL TFE and purified by silica gel flash chromatog-
raphy(eluent:ethylacetate/methanol10:1!1:1!1:2).Theproduct
wasfinallydissolvedinaTFE/watermixtureandlyophilizedyielding
414 mg (0.05mmol) of the product as a pink powder.
376 MHz 19F-NMR (CDCl3): [d/ppm] ¼ ꢀ162.30 (t, 2F), ꢀ157.76 (t,
1F), ꢀ152.77 (d, 2F).
2.8. Synthesis of Pentafluorophenyl Methacrylate
(PFMA)
400 MHz1H-NMR(dmso-d6):[d/ppm] ¼ 7.89–7.87(d,1.63H),7.68–
7.65 (t, 0.82H), 7.49–7.45 (t, 1.62H), 7.31–7.14 (m, 170H), 4.98–4.95
(m, 68H), 4.21–3.74 (m, 34H), 2.94 (s, br, 68H), 1.92–1.18 (m, 216H),
0.83–0.80 (m, 9H).
Pentafluorophenyl methacrylate (PFPMA) was prepared according
to literature.[51] 300 MHz 1H-NMR (CDCl3): [d/ppm] ¼ 6.45 (s, 1H),
5.91 (s, 1H), 2.09 (s, 3H). 376 MHz 19F-NMR (CDCl3): [d/ppm] ¼
ꢀ152.71 (d, 2F), ꢀ158.12 (t, 1F), ꢀ162.41 (t, 2F)
GPC (HFIP): Mn ¼ 11 300 g ꢁ molꢀ1. ð ¼ 1.11.
2.9. Synthesis of Neopentyl-Ammonium
Tetrafluoroborate
2.13. Synthesis of Polylysine-Disulfide-Chain
Transfer Agent (Polylysine-Disulfide-CTA)
Neopentyl-ammonium tetrafluoroborate was synthesized as
previously described.[52]
500 mg (0.063mmol, 1 eq.) of HOOC-poly[Lys(Ne-Cbz)]30-NH2were
dissolved in 5 mL of abs. N-methyl-2-pyrrolidone (NMP). A separate
flask was loaded with 76mg (0.13 mmol. 2 eq.) of 3-((2-(4-cyano-4-
((phenylcarbonothioyl) thio) pentanamido) ethyl)-disulfanyl) prop-
anoic acid-pentafluoro-phenyl ester (disulfide-PFP-CTA) and 55.8mg
2.10. Synthesis of Ne-Benzyloxycarbonyl-L-Lysine-N-
Carboxyanhydride (Lys(Cbz) NCA)
1
1
6
6
(0.25 mmol, 4 eq.) of N ,N ,N ,N -tetramethylnaphthalene-1,8-dia-
minein1.5 mLofabs.NMP.Thissolutionwasaddedtothepolylysine
solution under argon atmosphere. The reaction was stirred over
nightatroomtemperatureand lightexclusion. Thenthesolventwas
removed by evaporation under reduced pressure, the crude was
dissolved in 1 mL of trifluoroethanol, and finally the polymer was
precipitated in cold ethyl ether. Afterward, the product was purified
byflashchromatography(eluent:ethylacetate/methanol10:1!1:1
! 1:2), dissolved in trifluoroethanol again and slowly added into
water dropwise. The solution was lyophilized and 428 mg
(0.054mmol, 80%) of the product were obtained as a pink solid.
400 MHz1H-NMR(dmso-d6):[d/ppm] ¼ 7.90–7.88(d,1.72H),7.69–
7.65 (t, 0.86H), 7.51–7.47 (t, 1.72H), 7.31–7.14 (m, 150H), 4.97–4.94
(m, 60H), 4.19–3.81 (m, 30H), 2.93(s, br, 60H), 1.99–1.23 (m, 180H),
0.82–0.79 (m, 9H).
Ne-Benzyloxycarbonyl-L-lysine-N-carboxyanhydride was synthe-
sized as previously described.[52]
2.11. Synthesis of Polylysine with Ne-
Benzyloxycarbonyl (Ne-Cbz) Protecting Group with
NCA-Polymerization (HOOC-poly[L-Lys(Ne-Cbz)]30-NH2)
HOOC-poly[L-Lys(Ne-Cbz)]30-NH2 was synthesized by polymeriza-
tion of a-aminoacid-N-carboxy anhydrides (NCA-polymerization)
with Ne-benzyloxycarbonyl-L-lysine-N-carboxyanhydride (Lys
(Cbz) NCA) as monomer and neopentyl-ammonuim tetrafluor-
oborate as initiator as previously described.[52] In brief, 333 mg
(1.09 mmol) of the monomer (Lys(Ne-Cbz) NCA) were transferred
into a Schlenk tube in a nitrogen counterflow and dissolved in
GPC (HFIP): Mn ¼ 11 000 g ꢁ molꢀ1, ð ¼ 1.13.
Macromol. Biosci. 2016, 16, 106–120
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
109
www.MaterialsViews.com

K. Tappertzhofen et al.
www.mbs-journal.de
acetic acid for 90 min at room temperature. 1.5 mL of water was
added and stirred for 90 min at room temperature. After
codestillation with toluene and dichloromethane the cationic
block copolymerwas dialyzedagainstpurewater(MWCO14 000 or
25 000 g ꢁ molꢀ1) and lyophilized.
2.14. Synthesis of Polymers
PolymerizationswerecarriedoutinSchlenktubeswithaTFE/dioxane
mixture(1:8)assolventafterfourfreeze–thawcyclesat608Cfor18h
in an oil bath. AIBN was used as initiator with a ratio of (disulfide)
polylysine-CTA/initiator of 6:1. Block copolymers were precipitated
from TFE/dioxane in hexane. The end group was removed with an
access of 4,4-azobis(4-cyanovaleric acid) (ACVA). Conversion of poly
[PFMA] to poly[HPMA] was carried out by postpolymerization
modification with2-hydroxypropylamineand completeconversion
was monitored by 19F-NMR spectroscopy. Each polymer was
synthesized with and without the fluorescence dye Oregon Green
488. Labeled polymers were used for cell uptake experiments and
FCS-measurements. Unlabeled polymers for agarose gel electro-
phoresis and determination of transfection and toxicity.
2.16. Formation of Polymer–pDNA Complexes
(Polyplex Micelles)
DesiredamountsofpDNAweredissolvedinwaterbioreagentgradein
an Eppendorf tube. The cationic block copolymers P2 and P4 were
dissolvedin 10 mMTris–HCl buffer (c¼ 1mgꢁ mLꢀ1)andslowlyadded
to the pDNA in solution in different amounts as indicated. The
resulting polyplexes were mixed by frequently pipetting and
vortexting. The polyplexes were formed one day before the experi-
ments. Complexes of pDNA and Jet-PEI were prepared according to
manufacture instructions. pGL3-Basic vector (Promega GmbH Man-
nheim, Germany) was used for formation of polyplexes for agarose gel
electrophoresis, FCS-measurements and CLSM. pEGFP-N1pDNA (Clo-
netech, Mountain View, CA) was used for determination of trans-
fection efficiency and toxicity by FACS analysis.
2.15. Synthesis of Block Copolymers Polylysine-b-
Poly[HPMA] (P1–P3) and Polylysine-S-S-b-p[HPMA]
(P4–P6) by RAFT-Polymerization
In a typical reaction procedure a Schlenk tube was loaded with
125 mg (107 mg, 0.012 mmol with CTA end-group) of disulfide-
polylysine-CTA were diluted in a mixture of 600 mL TFE and 5 mL
abs. dioxane, 500 mg (2 mmol) of the monomer PFMA and 0.34 mg
(0.002 mmol) of the initiator AIBN were added via stock solution
(molar ratio: AIBN/CTA/monomer 1/6/1 000). After four freeze–
thaw cycles the resulting solution was stirred for 2 d at 60 8C in an
oil bath. The polymer was precipitated from TFE/dioxane (1:8) in
hexane three times yielding 394 mg of a pink powder. The polymer
was dissolved in TFE/dioxane (1:8) and stirred for 4 h with 120 mg
(0.44 mmol, 25 eq.) of 4,4-azobis(4-cyanovaleric acid) (ACVA) at
85 8C for removal of end groups. The polymer with removed end
groups was precipitated three times from dioxane/TFE (1:8) in
hexane/ethyl ether (2:1) obtaining 366 mg of a colorless powder.
Subsequently, 50 mg of the block copolymer were dissolved in 1 mL
abs. DMSO with 3 mL abs. dioxane, 1 mg (0.002 mmol) of Oregon
Green488cadaverine,71 mL(0.52 mmol)NEt3 and30 mg(0.4 mmol)
2-hydroxypropylaminewereaddedandstirredfor3dat35 8C.After
complete conversion the HPMA block copolymer was precipitated
in ethyl ether and dialyzed against pure water (MWCO 14
000 g ꢁ molꢀ1) obtaining 27 mg of the disulfide polylysine-b-p
[HPMA] block copolymer with Ne-Cbz-protected polylysine.
2.17. Agarose Gel Electrophoresis
Agarose gel electrophoresis was performed to determine optimal
pDNA polymer mixing ratios. In brief, 160 ng pGL3-Basic
vectorpDNA was mixed with the different polymersin different
ratiosas described earlier and filled into the slots of a 0.5 wt%
agarose gel with 6ꢂ loading buffer. The experiments were
conducted at 120 V for 20 min and the polyplexes were visualized
by staining the pDNA with GelRed and detected with UV light at
365 nm.
2.18. Fluorescence Correlation Spectroscopy
Fluorescence correlation spectroscopy (FCS) was performed using a
commercial setup (Zeiss, Oberkochen, Germany) that consisted of
the module ConfoCor 2 and an inverted microscope (Axiovert 200),
equipped with a Zeiss C-Apochromat 40ꢂ/1.2 W water immersion
objective. For excitation the 488 nm line of an argon laser was used,
and collected fluorescence was filtered through a LP505 long pass
emission filter before reaching an avalanche photodiode detector
that enables single-photon counting. Eight-well polystyrene-
chambered coverglass (Laboratory-Tek, Nalge Nunc International,
Penfield, NY) was used as a sample cell. For each solution, a series of
10 measurements with a total duration of 5 min was performed.
The confocal observation volume was calibrated using a reference
dye with a known diffusion coefficient (i.e., Alexa Fluor 488).
̵
GPC (HFIP): Mn ¼ 32 100,D ¼ 1.48
In case of the disulfide linked block copolymer (polylsysine-S-S-
b-p[HPMA] A–C), deprotection was carried out by reaction with
200 mL methane sulfonic acid (MeSO3H) in 1.5 mL TFA for 3 h at
room temperature. 1.5 mL of waterwas addedand the reactionwas
stirred for additional 3 h at room temperature. After codestillation
with toluene and dichloromethane the cationic block copolymer
was first dialyzed against 10 mM NaCl solution and then against
pure water (MWCO 14 000 or 25 000 g ꢁ molꢀ1) and finally
lyophilized obtaining 15 mg of an orange powder. The same
polymer was synthesized without Oregon Green 488.
2.19. Cell Culture
400 MHz1H-NMR(dmso-d6):[d/ppm] ¼ 7.40(br. s1H),4.71(br. s,1H),
4.23 (br. s,0.12H), 3.67 (br. s, 1H), 2.91 (br. s, 2H), 1.69–0.84 (m, 12H).
GPC (HFIP): Mw /Mn ¼ 1.43
Adherent HEK 293-T cells were grown in culture medium (DMEM,
supplemented with 10% FBS, 4 mM l-glutamine, 0.1 M 2-mercap-
toethanol, 100 U ꢁ mLꢀ1 penicilin and 100 mg ꢁ mLꢀ1streptomycin)
and maintained at 37 8C at a humidified atmosphere with 10% CO2
under sterile conditions. Prior to experiments, cells were harvested,
Deprotection of the Ne-benzyloxycarbonyl protecting group (Ne-
Cbz)ofthepolymerwithoutdisulfidebond(polylsysine-b-p[HPMA]
A–C) was carried out by reaction with 100 mL 33 wt% HBr in 1.5 mL
Macromol. Biosci. 2016, 16, 106–120
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
110
www.MaterialsViews.com

Bioreducible Poly-L-Lysine–Poly[HPMA] Block Copolymers . . .
www.mbs-journal.de
counted and the required number of cells (cells ꢁ mLꢀ1) was
adjusted. All experiments with polyplexes formed by pDNA and
blockcopolymerwithbioreducibledisulfidegroup(P4)werecarried
out without 0.1 M 2-mercaptoethanol.
postpolymerization with 2-hydroxypropylamine can afford
multifunctionalp[HPMA]basedblockcopolymersofvarious
topologies.[44,53] These p[HPMA] based block copolymers are
usually obtained by sequential RAFT polymerization of two
different methacrylate monomers, including pentafluoro-
phenyl methacrylate, with low molecular weight chain
transfer agents (CTAs).
2.20. Cellular Toxicity Determined by Flow
Cytometry
We modified this synthetic strategy to obtain semi-
peptidic polypeptide-b-p[HPMA] block copolymers, more
precisely cationic polylysine-b-p[HPMA] block copolymers,
which can be used for pDNA delivery (transfection).
Therefore, we use macromolecular-polylysine chain trans-
fer agents (polylysine-CTAs) for RAFT-polymerization of
PFMA.[45] Postpolymerization modification with 2-hydrox-
ypropylamine and deprotection of the polylysine block
under acidic conditions lead to cationic polylysine-b-p
[HPMA] block copolymers. We now further extended this
synthetic approach to obtain cationic polylysine-b-p
[HPMA] block polymers with a bioreducible disulfide group
between the blocks to increase transfection properties.
Assessment of cytotoxicity on the level of single cells, HEK 293-T
cells were seeded onto 12-well cell culture plates (1.5 ꢂ 105 cells in
1 mL) one day before experiments. The culture media was changed
and cells were incubated with naked pDNA (pEGFP-N1, 5 mg per
well, negative control), polyplexes derived from polymer P2 or
P4 (5 mg pDNApEGFP-N1at N/P 7) or complexes with Jet-PEI (2.5 mg
pEGFP-N1). Theculturemediawaschangedafter24 handcellswere
harvested (PBS)/2 mM EDTA). After resuspension in HBSS buffer,
the cells were coincubated with Alexa Fluor 647-conjugated
Annexin V (BioLegend, San Diego, CA) for identification of early
apoptotic cells and 7-AAD (BioLegend, San Diego, CA) to detect
necrotic cells. Samples where analyzed by flow cytometry using a
FACS Canto II flow cytometer quipped with BD FACS Diva software
(BD Biosciences, San Diego, CA). Data were analyzed using FlowJo
software (FLOWJO, Ashland, OR).
3.1. Polymer Synthesis
2.21. Transfection Efficiency Determined by Flow
Cytometry
Cationic polylysine-b-p[HPMA] block copolymers with a
bioreducible disulfide bond were synthesized by RAFT-
polymerization of the reactive ester monomer pentafluor-
ophenyl methacrylate with macromolecular, peptidic
disulfide-polylysine chain transfer agents with a blocked
Ne-side group (Disulfide-Polylysine(Cbz)-CTA). The p[PFMA]
block of the resulting hydrophobic block copolymers
polylysine(Cbz)-S-S-b-poly[PFMA] A–C was reacted with
2-hydroxypropylamine (HPA) to prepare poly[N-(2-hydrox-
ypropyl methacrylate (p[HPMA]). After final removal of the
lysine’s Ne-protecting group under acidic conditions,
cationic polylysine-S-S-b-p[HPMA] block copolymers A–C
(P4–P6) are obtained. Simultaneously, the same block
copolymers were synthesized without a disulfide bond
(polylysine-b-p[HPMA] block copolymers A–C (P1–P3)
(Scheme 1–3).
Assessmentoftransfectionefficiencyonthelevelofsinglecellswas
performed using pDNA encoding for Enhanced Green Fluorescent
Protein (EGFP) (pEGFP-N1). In brief, HEK 293-T cells were seeded
onto 12-well cell culture plates (1.5 ꢂ 105 cells in 1 mL) 1 d before
experiments. The culture media was changed and cells were
incubated with naked pDNA (5 mg pEGFP-N1 per well, negative
control), polyplexesderived from polymersP2 or P4 (5 mg pEGFP-N1
at N/P 7) or complexes with Jet-PEI (2.5 mg pEGFP-N1). The culture
media was changed after 24 h, cells were harvested (PBS)/2 mM
EDTA) and analyzed by flow cytometry (see above) for determi-
nation of EGFP positive cells.
2.22. Confocal Microscopy
HEK-293T cells were seeded on a m-slide (cover glass with
chambers) in a 8-well (Ibidi) format with 2 ꢂ 105 cells ꢁ mLꢀ1. 24 h
afterincubationat37 8CthemediumwaschangedtoOptiMEMand
the cells were incubated with different polyplexes composed of
pGL3-Basicvectorand polymersP2 or P4(N/P 7) at 37 8C. Afterward,
the cells were analyzed in a 8-well format by confocal microscopy.
Pictures were recorded with a Leica Microscope (with ‘‘live cell
Unit’’) and analyzed with Leica Application suite software.
For the synthesis of bioreducible structures, an asym-
metric disulfide-CTA (PFP-disulfide-CTA) was synthesized
in multiple steps by reaction with the conventional PFP-
CTA (Scheme 1). The resulting PFP-disulfide CTA contains
both, a dithioester group for RAFT-polymerization and an
activated pentafluorophenyl ester moiety for furher
modification.
In parallel, poly-L-lysine with Ne-benzyloxycarbonyl
protecting groups (i.e., (polylysine(Cbz) was obtained by
polymerization of corresponding a-amino acid-N-carboxy
anhydrides(Scheme2).NCA-polymerizationwasusedsince
well-defined polylsine(Cbz) can be synthesized with a high
degree of polymerization and in good yields (g scale)[52]
compared to solid phase peptide synthesis (SPPS). In
3. Results and Discussion
The combination of controlled radical polymerization
(RAFT-polymerization) of reactive ester monomers, like
pentafluorophenyl
methacrylate
and
sequential
Macromol. Biosci. 2016, 16, 106–120
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
111
www.MaterialsViews.com

K. Tappertzhofen et al.
www.mbs-journal.de
Scheme 1. Multistep synthesis of low molecular weight chain transfer agent with disulfide bond between the dithioester and the activated
ester group (PFP-disulfide-CTA).
α-aminoacid-N-
previousworkofKataokaetal., itcouldbe
shown that the optimal length of the
polylysine block for pDNA complexation
varies between a degree of polymer-
ization of 30 ꢃ Xn ꢃ 50.[39] Moreover, we
achieved the highest transfection effi-
ciency for polylysine-b-p[HPMA] poly-
mers with polylysine obtained by NCA-
polymerization with Xn ¼ 30.[45] That is
why we chose polylysine witha degree of
polymerization of 30 to preparedisulfide-
polylysine(Cbz)-CTA. For the synthesis of
macromolecularpolylysineCTAwithand
without redox cleavable disulfide bond
polylysine(Cbz)30 was modified with PFP-
CTA or rather PFP-disulfide-CTA to obtain
polylysine(Cbz) CTA and disulfide-poly-
lysine(Cbz)-CTA (Scheme 2). Both CTAs
were characterized by GPC in HFIP
(Table 1).
carboxyanhydridepolymerization
(NCA polymerization)
O
O
O
NH
O
HN
O
BF4
NH3
H
NH
O
R
N
O
H
O
30
DMF, 3d, 40 °C
Nε-Cbz-L-
Lysine-N-Carboxy-anhydride
Polylysine(Cbz) with
free N-terminus
(NCA)
O
HN
O
1,8-Bis(N,N-dimethyl-
amino)-naphthalin
proton sponge
PFP-Disulfide-CTA
O
O
(Scheme 1)
R
S
S
N
H
S
N
H
O
30
NC
S
N
N
Disulfide-Polylysine(Cbz)-CTA
NMP, 20h
O
PFP-CTA
HN
R
O
The macromolecular, peptidic chain
transfer agents polylysine(Cbz)-CTA and
disulfide-polylysine(Cbz)-CTA have then
been applied for RAFT-polymerization of
pentafluorophenyl methacrylate (PFMA)
leading initially to hydrophobic block
copolymers polylysine(Cbz)-b-p[PFMA]
A–C and polylysine(Cbz)-S-S-b-p[PFMA]
A–C (Scheme 3). A mixture of dioxane/
(Scheme 1)
N
O
S
N
H
30
O
S
Polylysine(Cbz)-CTA
Scheme 2. Synthesis of poly-L-lysine with Ne-benzyloxycarbonyl (Ne-Cbz) protecting
group by polymerization of a-amino acid-N-carboxy anhydrides (NCA-
polymerization) and subsequent modification at the N-terminus with PFP-disulfide-
CTA or PFP-CTA obtaining disulfide-polylsine(Cbz)-CTA and polylysine(Cbz)-CTA.
Table 1. Characterization of disulfide-polylysine-CTA and polylysine(Cbz)-CTA with GPC in HFIP.
Mn (HFIP-GPC) [g ꢁ molꢀ1
]
̵
CTA
XN (NMR)
D (HFIP-GPC)
Polylysine(Cbz)-CTA
30
30
11 300
11 000
1.11
1.13
Disulfide-polylysine(Cbz)-CTA
Macromol. Biosci. 2016, 16, 106–120
112
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com

Bioreducible Poly-L-Lysine–Poly[HPMA] Block Copolymers . . .
www.mbs-journal.de
undesired side reactions.[54] Noteworthily, the disulfide
bond showed stability both toward the polymerization
conditions and during end-group removal (this observation
follows in partresults reported in ref.[55]). Postpolymeriza-
tion modification with 2-hydroxypropylamine (HPA)
yielded block copolymers without (polylysine(Cbz)-b-p
[HPMA] A–C) and with disulfide bond between the two
blocks (polylysine(Cbz)-S-S-b-p[HPMA] A–C) (Scheme 3,
Table 2).
O
O
HN
R
O
HN
R
O
N
O
O
O
S
S
S
N
H
S
N
H
N
H
30
O
30
NC
S
O
S
Polylysine(Cbz)-CTA
Disulfide-Polylysine(Cbz)-CTA
1.) RAFT-polymerization
O
F
O
F
AIBN
F
F
2.) Removal of end-groups
F
For each CTA, three different polymers have been
synthesized. The AIBN/polylysine CTA/monomer (PFMA)
ratioshavethereby been chosento obtainblockcopolymers
with large p[PFMA] or rather p[HPMA] blocks after
polymeranalogous modification, since the size of the
hydrophilic p[HPMA] block should be much bigger than
the cationic part for efficient shielding of the charged
polyplexes.[39,56] Supporting Information S1 and S2 display
the elution diagrams of both polylysine(Cbz)-CTA and
disulfide-polylysine(Cbz)-CTA and the corresponding block
copolymers polylysine(Cbz)-b-poly[HPMA] A–C without
disulfide group and polylysine(Cbz)-S-S-b-poly[HPMA] A–
C with disulfide, respectively. It shows the strong increase
in molecular weight during successful RAFT polymer-
ization. However, measurements of Ne-benzyloxycarbonyl-
protected polylysine block copolymers by GPC in HFIP
(Table 2) may give incorrect results due to secondary
structures of protected polylysine.[52] A more precise
characterization of the molecular weight of polylysine-b-
excess
ACVA
Dioxane/TFE,
48 h, 60 °C
Dioxane/ TFE,
4h, 85 °C
O
O
O
HN
R
HN
R
O
N
N
N
O
O
O
COOH
S
N
H
COOH
N
H
S
N
H
N
30
30
O
O
O
O
O
F5
F5
Polylysine(Cbz)-b-p[PFMA]
Polylysine(Cbz)-S-S-b-p[PFMA]
3. Postpolymerization modification
OH
Dioxane/ DMSO,
3d, 35 °C
NH2
TEA
O
O
O
HN
R
O
HN
R
O
N
N
O
N
O
O
COOH
N
H
COOH
1
S
30
N
H
S
N
H
poly[HPMA] block copolymers was achieved by H-NMR-
N
30
HN
O
O
HN
O
OH
OH
measurements after deprotection of the polylysine(Cbz)
block (P1–P6) (Table 3).
Polylysine(Cbz)-b-p[HPMA]
Polylysine(Cbz)-S-S-b-p[HPMA]
To verify that the redox cleavable disulfide bond is still
existing (non reacted) after all reaction steps, the
protected block copolymers polylysine(Cbz)-S-S-b-poly
[HPMA] have been treated with tris(2-carboxy ethyl-
phosphin) (TCEP) in aqueous solution overnight and
directly submit to GPC measurements with HFIP as
solvent. Figure 1 shows the elution profile for every block
copolymer before and after treatment with TCEP. In all
cases it can nicely been seen that reduction leads to a
decrease in molecular weight due to cleavage of the
disulfide bond. The shift to lower molecular weights is
relatively small, since the shown p[HPMA] block is quite
larger than the polylysine block. In addition, a signal
corresponding to the size of the cleaved polylysine block
increases in size with an elution volume of about 18 ml.
This clearly proves the presence of an intact disulfide
bond between the polylysine and the p[HPMA] block
(Figure 1).
4. Removal of protecting groups
N
ε−Cbz: 33 w% HBr
Nε−Cbz: MeSO3H
in AcOH
in TFA
NH2
NH2
N
N
O
N
R
O
O
N
O
N
H
COOH
COOH
30
R
O
S
S
N
H
HN
O
OH
N
H
30
HN
O
OH
Polylysine-b-p[HPMA]
Polylysine-S-S-b-p[HPMA]
S
S
Scheme 3. Synthesis of Polylysine-b-poly[HPMA] A–C P1–P3 and
polylysine-S-S-b-p[HPMA] A–C (P4–P6) by RAFT polymerization of
PFMA with disulfide-polylysine-CTA or polylysine CTA,
subsequent end-group removal with ACVA, postpolymerization
modification with 2-hydroxypropylamine, and protecting group
cleavage under acidic conditions.
In a final step the Ne-benzyloxycarbonyl-protecting
group of the polylysine block was removed under acidic
conditions (Scheme 3). In case of the polylysine(Cbz)-b-
poly[HPMA] A–C 33 wt% HBr in acetic acid can be used for
an efficient deprotection without side reactions.[45,52] In
trifluoroethanol (TFE) (8:1) was thereby used to ensure
solubility of both the polylysine-CTAs and PFMA monomer.
Afterward, the dithioester end-group was removed with an
excess of 4,4-azobis(4-cyanovaleric acid) (ACVA) to prevent
Macromol. Biosci. 2016, 16, 106–120
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
113
www.MaterialsViews.com

K. Tappertzhofen et al.
www.mbs-journal.de
Table 2. Characterization of polylysine(Cbz)-b-p[HPMA] block copolymers A–C without disulfide and polylysine(Cbz)-S-S-b-p[HPMA] A–C
with disulfide by GPC in HFIP.
̵
Polymer
Polylysine-
CTA used
Ratio AIBN/
Polylysine-
CTA/monomer
(PFMA)
D
Mn Polylysine
Mn
(HFIP-
GPC)
CTA (HFIP-GPC)
Polylysine(Cbz)-b-
poly[HPMA]
(HFIP-GPC)
[g ꢁ molꢀ1
]
[g ꢁ molꢀ1
]
Polylysine(Cbz)-b-p[HPMA] A
Polylysine(Cbz)-b-poly[HPMA] B
Polylysine(Cbz)-b-poly[HPMA] C
Polylysine(Cbz) 30-CTA
Polylysine(Cbz) 30-CTA
Polylysine(Cbz) 30-CTA
11 300
11 300
11 300
11 000
1/6/708
1/6/2067
1/6/2638
1/6/1000
18 700
38 500
55 100
32 100
1.40
1.47
1.35
1.48
Polylysine(Cbz)-S-S-b-p[HPMA] A Disulfide-polylysine(Cbz)
30-CTA
Polylysine(Cbz)-S-S-b-p[HPMA] B Disulfide-polylysine(Cbz)
11 000
11 000
1/6/1714
1/6/2224
41 500
47 100
1.50
1.54
30-CTA
Polylysine(Cbz)-S-S-b-p[HPMA] C Disulfide-polylysine(Cbz)
30-CTA
Table 3. Characteristics of polylysine-b-poly[HPMA] A–C (P 1–3) and polylysine-S-S-b-p[HPMA] A–C (P4–6).
̵
Polymer
D (HFIP-GPC)
Ratios Xn (poly[HPMA])
to Xn (polylysine) (NMR)
Mn (NMR)
polylysine-b-
poly[HPMA]
polylysine(X)-b-
poly[HPMA]
Polylysine-b-poly[HPMA] A P1
Polylysine-b-poly[HPMA] B P2
Polylysine-b-poly[HPMA] C P3
Polylysine-S-S-b-p[HPMA] A P4
Polylysine-S-S-b-p[HPMA] B P5
Polylysine-S-S-b-p[HPMA] C P6
9 400
30 500
57 100
36 800
46 900
65 000
1.57
1.48
1.32
1.43
1.49
1.52
61:30 (2:1)
200:30 (6:1)
390:30 (13:1)
230:30 (7:1)
300:30 (10:1)
430:30 (14:1)
case of the polylysine(Cbz)-S-S-b-poly[HPMA] A–C block
copolymers, deprotection with 33 wt% HBr leads to
cleavage of disulfide bond, probably due to readily
oxidation of bromine (data not shown). That is why
deprotection was carried out with methane sulfonic acid
in trifluoroacetic acid (TFA).[57] (Supporting information
S3 shows the 1H-NMR spectra of successful removal of the
Ne-benzyloxycarbonyl-protecting group of polylysine-S-S-
b-poly[HPMA]). Figure 2 additionally displays the GPC
elution profile of the block copolymer polylysine(Cbz)-S-
S-b-poly[HPMA] A, before and after deprotection (poly-
lysine-S-S-b-poly[HPMA] A, P4) as well as after reduction
(¼cleavage) with TCEP. Since, the elution profile shows no
significant difference between the blue and the red line,
deprotection with methane sulfonic acid does not lead to
disulfide bond cleavage compared to treatment with TCEP
(green line) (Figure 2).
The ratio of the p[HPMA] block to the polylysine
segments and the resulting molecular weight of the block
copolymers were calculated by 1H-NMR-spectroscopy,
since GPC measurement might give false results due the
cationic character of polylysine (rod shape rather than
random coil). Table 3 displays the characteristics of the
final cationic block copolymers without (P 1–3) and with
disulfide bond (P 4–6). For in vitro studies P2 and P4 have
been chosen, since they exhibit similarand most optimal
block ratios of cationicpolylysine to the shielding p
[HPMA] block (1:6 for P2 and 1:7 for P4; Table 3). Figure 3
displays the elution profiles for the (Cbz)-polylysine-CTA
and disulfide (Cbz)-polylysine-CTA and the corresponding
Macromol. Biosci. 2016, 16, 106–120
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
114
www.MaterialsViews.com

Bioreducible Poly-L-Lysine–Poly[HPMA] Block Copolymers . . .
www.mbs-journal.de
cationic block copolymers P2 and P4
derived from RAFT polymerization and
several polymeranalogues reaction steps.
These results nicely show that an asym-
metric disulfide could be synthesized in a
multistep reaction, which contains both a
dithioester group for RAFT-polymerization
and an activated PFP group (PFP-disulfide-
CTA) (Scheme 1). Additionally,
a well
defined polylysine (Xn:30) with Ne-benzy-
loxycarbonyl protecting group (polylsine
(Cbz)) obtained by NCA-polymerization
could be modified with this PFP-disulfide
(Cbz)-CTA selectively at the N-terminus
obtaining a macromolecular, peptidic chain
transfer agents with a biocleavable disul-
fide group (disulfide-polylysine(Cbz)-CTA)
(Scheme 2). Controlled RAFT polymeriza-
tion utilizing both the CTA with disulfide
(disulfide-polylysine(Cbz)-CTA) and with-
out disulfide bond (polylysine(Cbz)-CTA)
and PFMA monomer (rather hydrophobic),
and subsequent postpolymerization mod-
ification with 2-hydroxypropylamine leads
to polylysine(Cbz)-S-S-b-p[HPMA] A–C and
polylysine(Cbz)-b-p[HPMA] A–C. Different
molecular weights of the block copolymers
could be obtained by varying polylysine
(Cbz)-CTA to monomer ratios (Table 2). Most
importantly, the presence of the disulfide
bond for the polylysine(Cbz)-S-S-b-p
[HPMA] A–C block copolymers was proved
by direct reduction (¼cleavage) with TCEP,
thus showing that the disulfide is stable
during synthesis. After removal of the Ne-
benzyloxycarbonyl protecting group of the
polylysine block under acidic conditions
cationic block copolymers P 1–3 and P 4–6
were obtained. Noteworthily, for the depro-
tection of disulfide containing polymers,
methane sulfonic acid in TFA had to be used
instead of HBr in acetic acid to prevent
undesired cleavage of the disulfide bond.
Cationic polymers with disulfide (P4) and
without disulfide (P2) have been chosen for
detailed in vitro transfection experiments
with HEK-293T cells due to their similar and
most optimal block lengths ratio of cationic
polylysine to the (p[HPMA]) shielding block
(1:6 for P2 and 1:7 for P4) in order to test the
significance of the disulfide bond.
Figure 1. Proof of presence of a disulfide bond by directed reduction of block
copolymers polylysine(Cbz)-S-S-b-poly[HPMA] A–C through overnight treatment
with TCEP.
Macromol. Biosci. 2016, 16, 106–120
115
www.MaterialsViews.com
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

K. Tappertzhofen et al.
www.mbs-journal.de
it is especially suitable to study the
formation of complexes between the
relatively small fluorescent polymers and
the significantly larger non-fluorescent
pDNA. Thereby, we found a hydrodynamic
radius of 58 nm for the polyplex formed by
pDNA and P2 and of 40 nm for the
polyplexes formed bypDNA and P4(bothat
N/P 7). For comparison the hydrodynamic
radii of P2 and P4, were found to be 4.1 and
3.0 nm respectively (Figure S4). Thus, the
polymers P2 and P4 show similar pDNA
encapsulation properties and hydrody-
namic diameters as expected.
3.3. Biological Evaluation of Polyplexes
Derived from pDNA and Polymers P2
and P4 with HEK 293-T Cells
Figure 2. GPC elution profile of polylysine(Cbz)-S-S-b-poly[HPMA]
A
(blue),
deprotected polylysine-S-S-b-poly[HPMA] A (red) and cleaved block copolymer
after reduction with TCEP (green).
3.3.1. Cellular Binding and Uptake
3.2. Polyplex Formation and Physicochemical
Characterization
In order to determine cellular binding and successful
internalization, polyplexes composed of Oregon Green
488-labeled polymers P2 and P4 and pDNA (pGL3-Basic
vector) (N/P 7) were subjected to in vitro experiments
using HEK 293-T cells. Confocal laser scanning microscopy
(CLSM) was performed after incubating the cells with the
polyplexes for 24 h (Figure 5). The images reveal that both
polyplexes P2 and P4 are successfully internalized by
HEK-293T cells. Thus, the precondition for a transfection
is given.
PolyplexeswereformedbysimplemixingofpDNAwiththe
different cationic block copolymers P2 and P4. Thereafter,
the corresponding polymers were dissolved in TRIS-buffer
and added to the pDNA (pGL3-Basic vector or pEGFP-N1) in
water at the desired concentrations, thoroughly mixed and
incubatedfor 18 h (Scheme 4).
InordertotestthecapabilityofpDNAcondensationandto
determine suitable polymer to pDNA ratios we performed
agarose gel electrophoresis (Figure 4). As to be expected
increasedretardationofpDNAwasobservedwithincreasing
amounts of polymer added. For both block copolymers (P2
andP4)completeencapsulationofthepDNAwasobserved.A
more precise expression of polymer to pDNA ratio is the N/P
value. The N/P value of 1 implies equal amounts of negative
chargesonpDNAandpositivechargesonthepolymer.1 mgof
pDNA represents 3.07 ꢂ 10ꢀ9 mol of negative charges. With
the known molecular weights of block copolymers and the
numberofpositivechargesoneachchain(equaltothedegree
of polymerization of polylysine), the amount of polymer
needed for N/P 1, etc. can be calculated. It can nicely be seen
that both polymers P2 and P4 mediate pDNA encapsulation
at N/P ratios of higher or equal to one. For transfection, N/P
ratios of 7 have been used to ensure absence of free, non
complexed pDNA (Figure 4, white arrows).
3.3.2. Transfection Efficiencies
For a successful transfection, the pDNA needs to escape
from endosomal compartments and it needs to be released
from the polyplexes after entering the cell in order to be
expressed. To determine transfection efficiencies, we
employed fluorescence activated cell sorting (FACS) to
detect expression of a pDNA-encoded fluorescence reporter.
Thismethodisadvantageouscomparedtotheconventional
Luciferase assay, since transfection efficiencies can be
determined on single cell level.[38,45] Thus, a deeper insight
in transfection efficiencies is obtained. Thereto, polyplexes
wereformedwithpDNA, whichencodesfor enhancedgreen
fluorescent protein (pEGFP-N1). Hence, transfected cells are
detectable by FACS measurements due to EGFP expression
(this cannot be done with Luciferase assays due to fast
disappearance of bioluminescence). HEK-293T cells were
incubated with polyplexes formed by pEGFP-N1 and
polymers P2 and P4, eachat N/P values of 7. Additionally,
cells were treated with naked pDNA (neg. control) and
polyplexes with commercial, polymeric transfection
reagent Jet-PEI (pos. control). Frequencies of transfected
For a more detailed characterization of the polyplexes we
performed fluorescence correlation spectroscopy (FCS)
studies of Oregon Green 488-labeled polymers P2 and P4
before and after mixing with pDNA (pGL3-Basic vector)
(supporting information, Figure S4). FCS is a very sensitive
technique that allows precise measurement of the hydro-
dynamic radii of fluorescent species in solution[58] and thus
Macromol. Biosci. 2016, 16, 106–120
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
116
www.MaterialsViews.com

Bioreducible Poly-L-Lysine–Poly[HPMA] Block Copolymers . . .
www.mbs-journal.de
for the positive control (JetPEI-derived
polyplexes). Interestingly, in case of
JetPEI either dose of pDNA resulted in
comparable transfection of HEK-293 T
cells. This shows—first of all—that the
disulfide bond mediates a very strong
increase in efficiency (since the poly-
mers are almost identical, except for the
disulfide bond in P4), due to an intra-
cellular reductive cleavage of the p
[HPMA] corona and improved release
of cargo pDNA. Importantly, bioreduci-
ble polyplexes with P4 are as effective
as the commercial transfection reagent
Jet-PEI. This very nicely shows, that the
redox-stimuli responsive, intracellular
cleavage of the p[HPMA] corona leads to
a
strong increase in transfection
efficiency.
3.3.3. Cytoxicity of Polyplexes
For efficient systems it remains most
challenging to increase the transfection
efficiency, but to keep a low cytoxicity at
thesametime. Infact, toxicityisanatural
result of the polycationic nature of the
vector, which is—on the other side—
needed for DNA condensation and endo-
somal escape. The need for low toxicity
becomes especially important when
dealing with sensitive immune cells such
as DCs and it is an essential precondition
for in vivo applications of transfection
agents.
Figure 3. GPC elution profiles for the polylysine(Cbz)-CTA and disulfide(Cbz)-polylysine-
A precise determination of cytotox-
icity can be achieved via FACS meas-
urements. By staining the cells with
Annexin V, which binds to phosphati-
CTA and the corresponding cationic block copolymers P2 and P4, respectively, in HFIP.
cells were determined by FACS measurements 24 h after
incubation (Figure 6). The mean fluorescence intensities
of EGFP of one representative experiment are given in
Figure S5.
dylserine at the extracellular membrane, early apoptotic
cells are detected. The dye 7-AAD, which enters
disintegrated cell membranes only, intercalates with
chromosomal DNA, and serves to stain necrotic cells.
Cells that are double-positive for Annexin V and 7-AAD
We aimed to increase the transfection efficiency by
introducing a bioreducible disulfide bond between the
complexing polylysine and the shielding p[HPMA] block.
While treatment with naked pDNA resulted in no
detectable transfection (as to be expected), low trans-
fection efficiencies were obtained with polyplexes
composed of pDNA and polymer P2. In contrast, poly-
plexes composed of pDNA and polymer P4 yielded strong
transfection, in a pDNA dose-dependent manner. At
higher dose of pDNA (5 mg), P4-derived polyplexses
mediated a higher frequency of EGFPþ cells than obtained
represent
a late apoptotic state. This approach is
advantageous to conventional MTT assays, since more
detailed information on cytotoxicity is obtained on
single cell-level.
To assess potential cytotoxicity of polyplexes com-
posed of pDNA and P2 and P4, we performed according
FACS measurements. 24 h after transfection of HEK-293
T cells with naked pDNA (pEGFP-N1) as a negative
control, and according polyplexes using Jet-PEI (pos.
control).
Macromol. Biosci. 2016, 16, 106–120
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
117
www.MaterialsViews.com

K. Tappertzhofen et al.
www.mbs-journal.de
Scheme 4. Formation of bioreducible polyplexes with negatively charged pDNA and cationic, redox-stimuli responsive polymer polylysine-
S-S-b-p[HPMA] (P4) for improved transfection properties, due to intracellular polyplex deshielding (cleavage of the p[HPMA] corona).
As shown in Figure 7, transfection of
HEK-293 T cells with polyplexes yielded
no enhanced apoptosis or necrosis as
compared with the negative control.
These results clearly indicate that intro-
duction of a disulfide group mediates a
strong increase in transfection efficiency
without increasing cytotoxicity and,
thus, makes polyplexes composed of
pDNA and P4 highly suitable transfec-
tion reagents for sensitive cells.
4. Conclusion
The multistep synthesis of a disulfide
CTA allows the preparation of Ne-
benzyloxycarbonyl (Cbz) protected pol-
ylysine chain transfer agents (CTAs)
with an internal disulfide group. Sub-
sequent RAFT polymerization experi-
ments with pentafluorphenyl-metha-
crylate demonstrate that this disulfide
Figure 4. Agarose gel electrophoresis with pDNA and polyplexes composed of pDNA and
polymers P2 and P4 at different N/P ratios as indicated.
group is inert under the conditions of
Macromol. Biosci. 2016, 16, 106–120
118
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com

Bioreducible Poly-L-Lysine–Poly[HPMA] Block Copolymers . . .
www.mbs-journal.de
radical polymerization. It gets, thus, pos-
sible to obtain block copolymers from Cbz
protected polylysine and poly(PFMA). Con-
version of the reactive ester block with 2-
hydroxypropylamine and deprotection of
the polylysine block makes block copoly-
mers from the cationic poly(lysine) and the
biocompatible
poly(HPMA)
available,
which are linked by a disulfide group.
They can thus be split under reductive
conditions.
The resulting cationic block copolymer
forms polyplexes with pDNA in a size
range of 40 nm in radius at optimized N/P
values of 7. Transfection experiments
with HEK-cells demonstrate that poly-
plexes containing the bioreducible disul-
fide groups possess a better transfection
efficiency than polyplexes from analogues
blockcopolymer, but without disulfide
group. Finally transfection efficiencies of
polylysine-S-S-b-p[HPMA] block copoly-
mers are comparable to these of Jet-PEI
and in addition they show lower cyto-
toxicity. This makes the new polylysine-
Figure 5. CLSM of HEK-293 T cells 24 h after incubation with polyplexes composed of
pDNA and polymers P2 and P4 (N/P 7) (green: polymer dye Oregon Green 488).
S-S-b-p[HPMA biosplittable block copolymers interesting
for the transfection of sensitive immune cells.
Figure 6. Frequencies of transfected HEK-293T cells as
determined with FACS using pDNA encoding for EGFP (for
polyplexes derived from or P2 and P4 (N/P 7)). Two different
pDNA concentrations (1 mg: white bars, 5 mg: black bars) were
used in parallel. FACS analysis was performed 24 h after the onset
of incubation. Naked pDNA served as a negative control, and
polyplexes with Jet-PEI as a positive control (y-axis: Frequencies
of GFP-positive cells; data represent the mean ꢄ SEM of 3
independent experiments). Statistical significant difference:
versus negative control (ꢀ). ꢅP < 0.05.
Figure 7. Frequencies of apoptotic/necrotic HEK-293 T cells as
assessed 24 h after transfection with naked-pDNA as negative
control (ꢀ), polyplexes derived from Jet-PEI, and polyplexes
composed of P2 and P4 (each N/P 7). Frequencies of early
apoptotic (Annexin Vþ), late apoptotic (Annexin Vþ7-AADþ) and
necrotic (7-AADþ) cells are indicated. Untreated HEK-293 T cells
left untreated and unstained were used for gating. Graphs are
representative for one of two independent experiments.
Macromol. Biosci. 2016, 16, 106–120
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
119
www.MaterialsViews.com

K. Tappertzhofen et al.
www.mbs-journal.de
Acknowledgement: The authors would like to acknowledge
funding from DFG (SFB 1066) and from Excellence Initiative
DFG/GSC 266.
[26] L. Nuhn, Antibody Drug Conj. 2014, DOI: 10.14229/
jadc.2014.8.15.002.
[27] E. Fleige, M. A. Quadir, R. Haag, Adv. Drug Deliv. Rev. 2012, 64,
866.
[28] H. Wei, R. X. Zhuo, X. Z. Zhang, Prog. Polym. Sci. 2013, 38, 503.
[29] R. Mandel, H. J. P. Ryser, F. Ghani, M. Wu, D. Peak, Proc. Natl.
Acad. Sci. USA 1993, 90, 4112.
[30] A. Meister, M. E. Anderson, Annu. Rev. Biochem. 1983, 52, 711.
[31] D. J. Morre, D. M. Morre, Free Radic. Res. 2003, 37, 795.
[32] G. Saito, J. A. Swanson, K. D. Lee, Adv. Drug Deliv. Rev. 2003, 55,
199.
Received: May 31, 2015; Published online: July 29, 2015; DOI:
10.1002/mabi.201500212
Keywords: bioreducible polyplexes; disulfide block copolymers;
RAFT-polymerization; transfection
[33] S. Takae, K. Miyata, M. Oba, T. Ishii, N. Nishiyama, K. Itaka, Y.
Yamasaki, H. Koyama, K. Kataoka, J. Am. Chem. Soc. 2008, 130,
6001.
[34] Y. H. Wang, M. Zheng, F. H. Meng, J. Zhang, R. Peng, Z. Y. Zhong,
Biomacromolecules 2011, 12, 1032.
[1] J. N. Blattman, P. D. Greenberg, Science 2004, 305, 200.
[2] R. M. Steinman, I. Mellman, Science 2004, 305, 197.
[3] T. A. Waldmann, Nat. Med. 2003, 9, 269.
[35] M. Ahmed, R. Narain, Prog. Polym. Sci. 2013, 38, 767.
[36] W. T. Godbey, K. K. Wu, A. G. Mikos, Proc. Natl. Acad. Sci. USA
1999, 96, 5177.
[37] S. Richardson, P. Ferruti, R. Duncan, J. Drug Target 1999, 6, 391.
[38] P. Heller, A. Birke, D. Huesmann, B. Weber, K. Fischer, A. Reske-
Kunz, M. Bros, M. Barz, Macromol. Biosci. 2014, 14, 1380.
[39] K. Itaka, K. Osada, K. Morii, P. Kim, S. H. Yun, K. Kataoka, J.
Control. Release 2010, 143, 112.
[40] B. Rihova, M. Bilej, V. Vetvicka, K. Ulbrich, J. Strohalm, J.
Kopecek, R. Duncan, Biomaterials 1989, 10, 335.
[41] B. Rihova, M. Kovar, Adv. Drug Deliv. Rev. 2010, 62, 184.
[42] L. Nuhn, S. Hartmann, B. Palitzsch, B. Gerlitzki, E. Schmitt, R.
Zentel, H. Kunz, Angew. Chem. Int. Ed. 2013, 52, 10652.
[43] N. Mohr, M. Barz, R. Forst, R. Zentel, Macromol. Rapid
Commun. 2014, 35, 1522.
[44] L. Nuhn, M. Barz, R. Zentel, Macromol. Biosci. 2014, 14, 607.
[45] K. W. F. Tappertzhofen, E. Montermann, A. Reske-Kunz, M.
Bros, R. Zentel, Macromol. Biosci. 2015, 15, 1158.
[46] P. J. Roth, C. Boyer, A. B. Lowe, T. P. Davis, Macromol. Rapid
Commun. 2011, 32, 1123.
[47] Z. Cao, H. Wu, J. Dong, G. Wang, Macromolecules 2014, 47,
8777.
[48] S. H. Thang, Y. K. Chong, R. T. A. Mayadunne, G. Moad, E.
Rizzardo, Tetrahedron Lett. 1999, 40, 2435.
[49] F. D. Jochum, L. zur Borg, P. J. Roth, P. Theato, Macromolecules
2009, 42, 7854.
[50] J. Marchand-Brynaert, M. Bouchet, R. Touillaux, C. Beauve, J.
Fastrez, Tetrahedron 1996, 52, 5591.
[51] M. Eberhardt, R. Mruk, R. Zentel, Eur. Polym. J. 2005, 41,
1569.
[52] D. Huesmann, A. Birke, K. Klinker, S. Turk, H. J. Rader, M. Barz,
Macromolecules 2014, 47, 928.
[4] P. Guermonprez, J. Valladeau, L. Zitvogel, C. Thery, S.
Amigorena, Annu. Rev. Immunol. 2002, 20, 621.
[5] O. Joffre, M. A. Nolte, R. Sporri, C. R. E. Sousa, Immunol. Rev.
2009, 227, 234.
[6] J. M. Timmerman, R. Levy, Annu. Rev. Med. 1999, 50, 507.
[7] M. D. Joshi, W. J. Unger, G. Storm, Y. van Kooyk, E.
Mastrobattista, J. Control. Release 2012, 161, 25.
[8] C. Wang, Q. Ge, D. Ting, D. Nguyen, H. R. Shen, J. Z. Chen, H. N.
Eisen, J. Heller, R. Langer, D. Putnam, Nat. Mater. 2004, 3, 190.
[9] H. L. Robinson, T. M. Pertmer, Adv. Virus Res. 2000, 55, 1.
[10] S. Sudowe, S. Dominitzki, E. Montermann, M. Bros, S. Grabbe,
A. B. Reske-Kunz, Immunology 2009, 128, e 193.
[11] M. Bros, X. L. Ross, A. Pautz, A. B. Reske-Kunz, R. Ross, J.
Immunol. 2003, 171, 1825.
[12] R. Ross, S. Sudowe, J. Beisner, X. L. Ross, I. Ludwig-Portugall, J.
Steitz, T. Tuting, J. Knop, A. B. Reske-Kunz, Gene Ther. 2003, 10,
1035.
[13] A. S. Irvine, P. K. E. Trinder, D. L. Laughton, H. Ketteringham, R.
H. McDermott, S. C. H. Reid, A. M. R. Haines, A. Amir, R. Husain,
R. Doshi, L. S. Young, A. Mountain, Nat. Biotechnol. 2000, 18,
1273.
[14] X. P. Huang, Y. P. Yang, Hum. Gene Ther. 2009, 20, 293.
[15] E. Marshall, Science 1999, 286, 2244.
[16] J. Tian, Z. L. Xu, J. S. Smith, S. E. Hofherr, M. A. Barry, A. P.
Byrnes, J. Virol. 2009, 83, 5648.
[17] R. J. Christie, N. Nishiyama, K. Kataoka, Endocrinology 2010,
151, 466.
[18] I. Martin, C. Dohmen, C. Mas-Moruno, C. Troiber, P. Kos, D.
Schaffert, U. Lachelt, M. Teixido, M. Gunther, H. Kessler, E.
Giralt, E. Wagner, Org. Biomol. Chem. 2012, 10, 3258.
[19] K. Miyata, N. Nishiyama, K. Kataoka, Chem. Soc. Rev. 2012, 41,
2562.
[53] M. Barz, R. Luxenhofer, R. Zentel, A. V. Kabanov, Biomaterials
2009, 30, 5682.
[54] D. Pissuwan, C. Boyer, K. Gunasekaran, T. P. Davis, V. Bulmus,
Biomacromolecules 2010, 11, 412.
[55] C. H. Zhu, M. Zheng, F. H. Meng, F. M. Mickler, N. Ruthardt, X. L.
Zhu, Z. Y. Zhong, Biomacromolecules 2012, 13, 769.
[56] K. Itaka, K. Yamauchi, A. Harada, K. Nakamura, H. Kawaguchi,
K. Kataoka, Biomaterials 2003, 24, 4495.
[20] E. Wagner, Adv. Genet. 2014, 88, 231.
[21] H. C. Chiou, M. V. Tangco, S. M. Levine, D. Robertson, K.
Kormis, C. H. Wu, G. Y. Wu, Nucleic Acids Res. 1994, 22, 5439.
[22] A. M. Krieg, A. K. Yi, S. Matson, T. J. Waldschmidt, G. A. Bishop,
R. Teasdale, G. A. Koretzky, D. M. Klinman, Nature 1995, 374,
546.
[23] M. Ogris, S. Brunner, S. Schuller, R. Kircheis, E. Wagner, Gene
Ther. 1999, 6, 595.
[57] A. Harada, S. Cammas, K. Kataoka, Macromolecules 1996, 29,
6183.
[24] P. Heller, N. Mohr, A. Birke, B. Weber, A. Reske-Kunz, M. Bros,
M. Barz, Macromol. Biosci. 2015, 15, 63.
[58] K. Koynov, H. J. Butt, Curr. Opin. Colloid Interface Sci. 2012, 17,
377
[25] K. Tappertzhofen, M. Bednarczyk, K. Koynov, M. Bros, S.
Grabbe, R. Zentel, Macromol. Biosci. 2014, 14, 1444.
Macromol. Biosci. 2016, 16, 106–120
ß 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
120
www.MaterialsViews.com