Polymer Cancerostatics Targeted with an Antibody Fragment Bound via a Coiled Coil Motif: In Vivo Therapeutic Efficacy against Murine BCL1 Leukemia

Article abstract of DOI:10.1002/mabi.201700173

A BCL1 leukemia-cell-targeted polymer–drug conjugate with a narrow molecular weight distribution consisting of an N-(2-hydroxypropyl)methacrylamide copolymer carrier and the anticancer drug pirarubicin is prepared by controlled radical copolymerization fo

Full text of DOI:10.1002/mabi.201700173

Full PaPer
Drug Targeting
www.mbs-journal.de
Polymer Cancerostatics Targeted with an Antibody
Fragment Bound via a Coiled Coil Motif: In Vivo
Therapeutic Efficacy against Murine BCL1 Leukemia
Michal Pechar,* Robert Pola, Olga Janoušková, Irena Sieglová, Vlastimil Král,
Milan Fábry, Barbora Tomalová, and Marek Kováˇr
Dedicated to Professor Karel Ulbrich on the occasion of his 70th birthday.
polymer conjugates in solid tumors
resulting from the enhanced permea-
tion and retention (EPR) effect further
A BCL1 leukemia-cell-targeted polymer–drug conjugate with a narrow
molecular weight distribution consisting of an N-(2-hydroxypropyl)meth-
acrylamide copolymer carrier and the anticancer drug pirarubicin is prepared
by controlled radical copolymerization followed by metal-free click chemistry.
A targeting recombinant single chain antibody fragment (scFv) derived from a
B1 monoclonal antibody is attached noncovalently to the polymer carrier via
a coiled coil interaction between two complementary peptides. Two pairs of
coiled coil forming peptides (abbreviated KEK/EKE and KSK/ESE) are used as
linkers between the polymer–pirarubicin conjugate and the targeting protein.
The targeted polymer conjugate with the coiled coil linker KSK/ESE exhibits
4× better cell binding activity and 2× higher cytotoxicity in vitro compared
with the other conjugate. Treatment of mice with established BCL1 leukemia
using the scFv-targeted polymer conjugate leads to a markedly prolonged
survival time of the experimental animals compared with the treatment using
the free drug and the nontargeted polymer–pirarubicin conjugate.
improves the tumor selectivity of the
polymer therapeutics.[1] Additionally, the
polymer carrier enables the attachment
of various targeting ligands that actively
target cancer cells.[2–6]
Unfortunately, the EPR effect can be
utilized only in the treatment of solid vas-
cularized tumors. In other cases, such as
the treatment of blood malignancies or
metastases in early stages, the active tar-
geting of the polymer therapeutics is highly
desirable to improve the overall therapeutic
efficiency. Among the various ligands
that have been described to actively target
cancer cells, antibodies and their fragments
thus far appear to be the most efficient.[2,7–9]
However, the well-defined covalent con-
jugation of proteins to polymers is not
1. Introduction
easily accomplished. The reaction between proteins and reactive
polymer precursors often results in a mixture of poorly defined
products with a compromised biological activity. Therefore, there
is an urgent need for conjugation methods that provide well-
defined products with fully preserved biological activities.
Although several sophisticated methods to achieve the site-
specific covalent modification of proteins have been recently
described,[10–14] the formation of a specific noncovalent bond
between two peptide tags is an equally attractive approach.
Among the various noncovalent methods, the utilization of
coiled coil heterodimers,[15–20] the hybridization of complemen-
tary morpholino oligonucleotides[21–23] and the formation of a
complex between bungarotoxin and a bungarotoxin-binding
peptide[24] for the attachment of biologically active proteins to
polymer carriers are particularly notable.
The use of coiled coil heterodimers for preparation of
polymer–drug conjugates containing either a biologically active
protein (FosWC peptide)[16] or a low molecular weight cytostatic
drug (methotrexate)[18] emerged in the literature after 2010.
Recently, we have reported the synthesis and in vitro evalu-
ation of an N-(2-hydroxypropyl)methacrylamide (HPMA)-based
polymer conjugate with an anticancer drug doxorubicin (Dox)
targeted to murine leukemia BCL1 via a recombinant single
The application of polymer–cancerostatic conjugates for neo-
plastic treatment provides several significant advantages
compared with conventional chemotherapy. The polymer thera-
peutics usually exhibit much lower nonspecific toxicity against
healthy cells and tissues as the biologically active molecules
are preferentially released from the conjugates into the target
tumor tissue or cells. The increased accumulation of the
Dr. M. Pechar, Dr. R. Pola, Dr. O. Janoušková
Institute of Macromolecular Chemistry
Czech Academy of Sciences
Heyrovského nám. 2, 162 06 Prague 6, Czech Republic
E-mail: pechar@imc.cas.cz
I. Sieglová, Dr. V. Král, Dr. M. Fábry
Institute of Molecular Genetics
Czech Academy of Sciences
Flemingovo nám. 2, 166 10 Prague 6, Czech Republic
B. Tomalová, Dr. M. Kovárˇ
Institute of Microbiology
Czech Academy of Sciences
Vídenˇská 1083, 142 20 Prague 4, Czech Republic
DOI: 10.1002/mabi.201700173
©
1700173 (1 of 11)
Macromol. Biosci. 2017, 1700173
2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
chain fragment (scFv) of the monoclonal antibody B1.[25] The
targeting protein was attached to the polymer–drug conjugate
via a noncovalent interaction between two peptides that formed
a coiled coil heterodimer. The scFv-targeted polymer conjugate
exhibited almost 100 times higher cytotoxicity against BCL1
cells compared with the corresponding nontargeted polymer
conjugate.
We have further optimized the structure of the targeted mac-
romolecular therapeutic using a modified method to synthesize
the polymer carrier, employing an improved structure of the
coiled coil heterodimer between the polymer and the targeting
protein[26] and introducing pirarubicin (Pir) instead of doxoru-
bicin as a cytostatic drug.[27] In this paper, we describe the syn-
thesis, results of physicochemical characterization and both in
vitro and in vivo biological evaluations of the optimized HPMA-
based polymer system using a BCL1 leukemia model.
Table 1. Basic characteristics of the copolymers.
a)
a)
Copolymer
MW
MW/Mn
TT
[mol%]b)
Pir
[wt%]c)
Peptide
[wt%]d)
Biotin
[wt%]d)
PTT
46 300
59 700
61 500
62 000
1.28
1.19
1.15
1.13
6.8
–
–
–
–
PPir
9.6
8.0
8.1
–
–
PEKE
PESE
–
14.2
13.8
2.0
2.1
–
a)Molecular weights determined by SEC using RI and LS detection; b)TT determined
by UV–vis spectrophotometry in methanol (ε305 = 10 300 L·mol−1·cm−1); c)Pir deter-
mined by UV–vis spectrophotometry in methanol (ε488 = 11 300 L·mol−1·cm−1);
d)Determined by HPLC analysis.
2.4. Size-Exclusion Chromatography (SEC)
The molecular weights and dispersity values of the polymers
and polymer–Pir conjugates were determined by SEC on a Shi-
madzu HPLC system equipped with UV–vis diode array detector
(Shimadzu, Japan), refractive index Optilab-rEX, and multi-
angle light scattering DAWN EOS detectors (Wyatt Technology
Corp., Santa Barbara, CA). TSK-Gel SuperAW3000 column and
80% methanol/20% sodium acetate buffer (0.3 m, pH 6.5) as an
eluent at a flow rate of 0.6 mL min−1 were used in all experi-
ments. A method based on the known total injected mass with
an assumption of 100% recovery was used to calculate of the
molecular weights from the light scattering data. The number-
and weight-average molecular weights for the polymer precur-
sors and the polymer–Pir conjugates are summarized in Table 1.
2. Experimental Section
2.1. Materials and Methods
3-[2-[2-[2-(2-Azidoethoxy)ethoxy]ethoxy]ethoxy]propanoic
acid (N3-PEG4-COOH) and N-[2-[2-[2-(2-azidoethoxy)ethoxy]-
ethoxy]ethyl]-biotinamide (N3-PEG3-biotin) were purchased
from Click Chemistry Tools, USA. (RS)-1-Aminopropan-2-ol,
2,2′-azobis(2-methylpropionitrile) (AIBN), 4-cyano-4-thioben-
zoylsulfanylpentanoic acid (CTP), 1-ethyl-3-(3-dimethyl-
amino-propyl)carbodiimide (EDC), N,N-dimethylacetamide
(DMA), 4-dimethylaminopyridine (DMAP), dimethyl sulfoxide
(DMSO), methacryloyl chloride, tert-butanol, triisopropylsilane
(TIPS), and trifluoroacetic acid (TFA) were purchased from
Sigma-Aldrich (Sigma-Aldrich, Czech Republic). 3-Amino-
1-(11,12-didehydrodibenzo[b,f]azocin-5(6H)-yl)propan-1-one
(DBCO-NH2) was purchased from Click Chemistry Tools (AZ,
USA). Pirarubicin (Pir) was obtained from Meiji Seika Pharma
Co., Ltd. (Japan). All other chemicals and solvents were of ana-
lytical grade. Solvents were dried and purified by conventional
procedures and distilled before use.
2.5. UV–Vis Spectrophotometry
The spectrophotometric analyses were carried out in quartz glass
cuvettes on a Helios Alpha UV–vis spectrophotometer (Ther-
mospectronic, UK). The content of dithiobenzoate (DTB) end
groups in the polymers were determined at 302 nm in methanol
using the molar absorption coefficient εDTB = 12 100 L mol−1 cm−1.
The results are summarized in Table 1. The determination of the
Pir content in the polymer–Pir conjugates (without fluorophore)
was performed at 488 nm in methanol using the molar absorp-
tion coefficient εPir = 11 300 L mol−1 cm−1. The Pir contents are
summarized in Table 2. The contents of carbonylthiazolidine-
2-thione (TT) reactive groups in the polymer precursors were
determined at 305 nm in methanol using the molar absorption
2.2. HPLC Monitoring of Polymer–Analogous Reactions
Monitoring of the conjugation reactions of DBCO-NH2, piraru-
bicin, biotin, and peptides to the reactive polymer precursors was
performed by HPLC using a 100 × 4.6 mm Chromolith Perfor-
mance RP-18e column (Merck, Germany) and a linear gradient
of water–acetonitrile (0–100% acetonitrile) in the presence of
0.1% TFA with a UV–vis diode array detector (Shimadzu, Japan).
Table 2. Cytostatic activity of the scFv-targeted and nontargeted
polymer–Pir conjugates and free Pir.
IC50 ( SD)a)
Sample
PESE/scFvKSK
PEKE/scFvKEK
PESE/scFv0
PEKE/scFv0
Pir
9
18
2
2.3. Cell Lines
3
ˇ
BCL1 cell line was obtained from Prof. Blanka Ríhová (Insti-
89
146
1
2
tute of Microbiology, Czech Academy of Sciences). The cells
were cultivated in RPMI medium (Thermo Scientific, Czech
Republic) supplemented with heat inactivated 10% fetal bovine
serum (FCS), 100 U mL−1 penicillin, 100 µg mL−1 streptomycin,
and 0.05 × 10−3 m 2-sulfanylethanol.
2
0.2
a)IC50 (µg L−1), concentration of Pir equivalent in the sample inhibiting growth of
the 50% cells compared with the untreated control.
©
1700173 (2 of 11)
Macromol. Biosci. 2017, 1700173
2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
coefficient εTT = 10 300 L mol−1 cm−1. The contents of aza-diben-
zocyclooctyne (DBCO) groups in the copolymers were deter-
mined at 292 nm using the absorption coefficient for DBCO in
methanol, ε292 = 13 000 L mol−1 cm−1.
pirarubicin (10 mg, 0.016 mmol) in 0.5 mL DMA. After 2 h,
no remaining free drug was detected by HPLC. DBCO-NH2
(5.8 mg, 0.021 mmol) was added and the reaction was left over-
night. HPLC revealed a small amount of remaining DBCO-
NH2 and no TT groups on the polymer. The reaction mixture
was precipitated into ethyl acetate; the crude polymer was dis-
solved in methanol and re-precipitated into ethyl acetate to yield
the polymer–drug conjugate PPir. The polymer was character-
ized using UV–vis to determine the amount of Pir (9.6 wt%).
The conjugation of the EKE or ESE peptide (1 mol%) to PPir
in DMA was monitored by HPLC, and the reaction was com-
pleted in 5 min. Then, N3-PEG3-biotin (1 mol%) was attached to
the polymer and the remaining DBCO groups were end-capped
with two molar excesses of N3-PEG4-COOH. The mixture was
precipitated into acetone; the crude product was dissolved in
methanol and re-precipitated into acetone to yield the polymer–
2.6. Dynamic Light Scattering (DLS)
The hydrodynamic radii and scattering intensities of the polymer
precursors and polymer conjugates were measured using the DLS
technique at a scattering angle of θ = 173° on a Nano-ZS instru-
ment (Model ZEN3600, Malvern Instruments, UK) equipped with
a 632.8 nm laser. The measurements were performed in 0.012 M
phosphate buffer with 0.138 M NaCl (PBS) (1.0 mg mL−1, pH 7.4)
solutions. For the evaluation of the dynamic light scattering data,
the DTS (Nano) program was used. The mean of at least three
independent measurements was calculated.
drug–peptide conjugates PEKE and PESE
.
The contents of unbound Pir and peptides in the polymer
conjugates (measured by HPLC analysis) were below 0.2% and
0.4% w/w of their total amount, respectively.
2.7. Synthesis of Peptides EKE and ESE
The EKE and ESE peptides were prepared as described
previously.[26]
2.11. Preparation of Recombinant Proteins and Scfv-Targeted
Polymer Conjugates
2.8. Synthesis of Monomers
The scFv B1 fragment with a C-terminal KSK tag was obtained
using a similar method as that used previously[17,25] for scFv B1
tagged with KEK. The difference between the KEK and KSK
tags was that the design of KSK was improved; namely, IAALK-
SKIAALKSE-(IAALKSK)2 ensured the formation of an antipar-
allel coiled coil with a suitable polymer-bound counterpart.[26]
Briefly, the 90 bp oligonucleotide duplex was prepared from
four oligonucleotides:
HPMA and Ma-GFLG-OH were prepared as described ear-
lier.[28,25] Ma-GFLG-TT was prepared by reacting Ma-GFLG-OH
(92 mg, 0.2 mmol) with 4,5-dihydrothiazole-2-thiol (29 mg, 0.24
mmol) using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide
hydrochloride (57.5 mg, 0.3 mmol) in DMF in the presence
of 4-dimethylaminopyridine at 4 °C overnight. After DMF was
evaporated, the reaction mixture was dissolved in DCM and
the water-soluble urea derivative was removed by subsequent
washing of the organic solution with an aqueous solution of
KHSO4, with a solution of NaCl and with water. After DCM was
evaporated, the product was dried to yield 80 mg of the mon-
omer, which was then characterized by HPLC (single peak) and
MS ESI (calculated 561.7, found 562.9 M+H).
SK1 = 5′-gtactatcgcagcgctgaaatctaagattgcggccttgaaa,
SK2 = 5′-tccgagatcgcggcactgaaatctaagatcgccgctctgaaaagcaagg,
SK3 = 5′-tgccgcgatctcggatttcaaggccgcaatcttagatttcagcgctgcgata,
and
SK4 = 5′-gtacccttgcttttcagagcggcgatcttagatttcag,
where SK2 and SK3 partly overlapped and had their 5′ ends
phosphorylated, and where the 5′ ends of SK1 and SK4 con-
tained four-base overhangs to allow for cloning into the Acc65I
site. The oligonucleotides SK1 + SK3 and SK2 + SK4 were
annealed, the resulting duplexes were ligated, and the 90 bp
KSK oligonucleotide duplex was gel-purified and used to replace
the KEK tag in scFv B1. The final construct thus encoded scFv
B1 in the format of VH-(gly4ser)4-VL-myc-KSK tag-His5.
2.9. Synthesis of Polymer Precursor
Reversible addition-fragmentation chain transfer (RAFT)
polymerization was performed as described earlier.[29] A mon-
omer/CTA/initiator molar ratio of 1000:2:1 was used. HPMA
(90 mol%, 100 mg), Ma-GFLG-TT (10 mol%, 43.6 mg) treated
with the initiator ABIN (1.3 mg), and the chain transfer agent
4-cyano-4-thiobenzoylsulfanylpentanoic acid (1.06 mg) were
mixed in DMA and tert-butyl alcohol (50/50 v/v). After removal
of dithiobenzoate (DTB) ω-end groups,[30] the polymer precursor
2.12. Expression and Purification of the Fusion
Protein scFv B1-KSK
P
TT was characterized by SEC (MW = 46 300, MW/Mn = 1.28) and
the content of TT groups (6.8 mol%) was determined by UV–vis.
For expression in E. coli BL21(DE3) cells, a modified pET-22(b)
vector was used. In this vector, the scFv coding sequence is
preceded by the PelB signal sequence, which allows for trans-
location of the product into the periplasmic space. The His5
tag at the C-terminus of the polypeptide was used for product
isolation and purification by IMAC chromatography on
2.10. Synthesis of Polymer Conjugates
The polymer precursor PTT (90 mg, 0.033 mmol TT, i.e.,
6.3 mol%) was dissolved in 1.5 mL of DMA and mixed with
©
1700173 (3 of 11)
Macromol. Biosci. 2017, 1700173
2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
Ni-CAM (Sigma). The final purification was achieved by ion
exchange chromatography on a MonoS column.[17,25]
Academy of Sciences. Mice were used at 9–15 weeks of age,
and food and water were given ad libitum. In all animal works,
institutional guidelines for the care and use of laboratory ani-
mals were strictly followed under a protocol approved by the
Institutional Animal Care and Use Committee of the Czech
Academy of Sciences and compliant with local and European
guidelines.
2.13. In Vitro Cytotoxic Activity of the Polymer Conjugates
The cells (5 × 103) were seeded into 100 µL of media in
96-well flat-bottom plates 24 h before the addition of free
Pir or before the polymer conjugates PESE or PEKE were dis-
solved in the solution of recombinant scFv fragments of the
B1 antibody. These antibody fragments varied based on the
presence of the coiled coil tags (KEK or KSK) or the absence
of the tags (scFvKEK, scFvksk, and scFv0, respectively, concentra-
tion 3.16 mg mL−1). The polymer/protein weight ratio of 2:1,
which corresponded to an ESE/KSK (or EKE/KEK) molar ratio
of 3:1, was used. The concentrations of the PESE and PEKE con-
jugates dissolved in the solutions of scFv fragments for the
cytotoxicity testing varied from 0.02 to 100 µg mL−1. The drug
concentrations of free Pir varied from 0.001 to 5 µg mL−1 for
the cytotoxicity testing. The cells were subsequently cultivated
for 72 h in 5% CO2 at 37 °C. Then, 10 µL of Alamar Blue
cell viability reagent was added to each well, and the plates
were incubated for 4 h at 37 °C. The metabolic activity was
measured according to the protocol for the Synergy Neo plate
reader (Bio-Tek, Czech Republic) using an excitation wave-
length of 570 nm and an emission wavelength of 600 nm. As
a control, the cells cultivated in medium without any treat-
ment were employed. The assay was conducted in triplicate
and repeated three times independently.
2.16. Monoclonal Antibodies
The following monoclonal antibodies were used to stain sur-
face antigens: anti-Myc-fluorescein (ExBio), CD3-biotin, CD3-
eF450, CD4-PE, CD4-APC, CD4-FITC, CD80-APC, MHC II-PE,
and STP-eF450 (eBiosciences). Live and dead cells were distin-
guished by propidium iodide staining.
2.17. Blood Clearance of the Polymer Conjugates
BALB/c mice were i.v. injected with the polymer–pirarubicin
conjugate PESE/scFvKSK, the polymer–pirarubicin conjugate
P
ESE/scFv0, or the same volume of PBS (220 µL). The dose of
the conjugate corresponded to a dose of 75 µg Pir per mouse.
Samples of blood were taken from experimental mice at
1 min, 1, 6, 12, 24, and 48 h postinjection. Samples taken at
1 min, 12, and 48 h were taken from the carotid arteries and
samples taken at 1, 6, and 24 h were taken from the tail vein.
Blood samples were collected in heparinized microtubes and
the plasma was separated. The concentration of Pir in the
plasma samples was determined using HPLC analysis. The
amount of Pir released from the polymer conjugate was deter-
mined after its extraction from the plasma into chloroform.
Mixtures of 50 µL of blood plasma samples and 50 µL of 6 m
HCl were heated to 50 °C for 1 h followed by extraction with
0.4 mL of chloroform for 15 min. The chloroform extract was
evaporated to dryness and the residue was diluted with 0.5 mL
of methanol. The Pir content was determined using an HPLC
Shimadzu system equipped with a fluorescence detector with
an excitation wavelength of 488 nm and an emission wave-
length of 560 nm. The calibration was carried out using Pir
standards dissolved in DMSO that were diluted with blood
plasma, hydrolyzed with 6 m HCl and extracted with chloro-
form as described above.
2.14. In Vitro Cell Binding Studies
To determine the binding of the scFvKEK/KSK-targeted PEKE/ESE
conjugates to the cell membrane, the cells were washed with
0.5% bovine serum albumin in PBS (BSA–PBS) and ≈2.5 × 105
of cells in a 50 µL volume were incubated for 30 min at 25 °C
with PESE/scFvKSK, PEkE/scFvKEK or PESE/scFv0, PEkE/scFv0 (as a
control). The conjugates with scFv were prepared as described
above using a polymer/protein weight ratio of 2:1. The final con-
centration of scFv for the binding studies was 50 µg mL−1. Then,
the cells were washed with 0.5% BSA–PBS, diluted in 50 µL,
and labeled for 30 min in the dark at 25 °C with streptavidin-
Alexa 405 (Thermo Scientific, Czech Republic), which recognizes
biotin on conjugates, and anti-c-Myc-fluorescein (Exbio, Czech
Republic), which recognizes the Myc tag sequence in scFv. After-
ward, the cells were washed with 0.5% BSA–PBS and diluted in
0.5 mL of 0.5% BSA–PBS containing 1 µg mL−1 7-AAD to detect
dead cells. The median fluorescence intensity of the polymer con-
jugates labeled with streptavidin-Alexa fluor 405 and scFv labeled
with anti-c-Myc-fluorescein was determined. The samples were
analyzed by FACS Verse (Becton Dickinson) and FlowJo software
(TreeStar). The FACS analysis of the cell binding of the polymer
conjugates was performed five times in triplicates.
2.18. In Vivo Binding Studies
BALB/c mice were i.p. inoculated with 5 × 105 BCL1 cells in
250 µL of PBS on day 0. The targeted polymer–pirarubicin con-
jugate PESE/scFvKSK and the nontargeted control polymer con-
jugate PESE dissolved in PBS at concentrations of 6.2 mg mL−1
(polymer conjugate) and 3.1 mg mL−1 (scFv). These treatments
were i.v. administered on day 30, and one dose contained
5 mg kg−1 of polymer-bound Pir. The spleens were harvested
1, 2.5, 6, and 24 h after the administration of conjugates. Naïve
mice, mice inoculated with BCL1 cells, and mice i.v. injected
with PBS alone were used as controls. Each experimental group
contained two mice.
2.15. Mice
Inbred BALB/c (H-2d) mice (females) were obtained from the
animal breeding facility of the Institute of Physiology, Czech
©
1700173 (4 of 11)
Macromol. Biosci. 2017, 1700173
2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
2.19. Staining for Surface Antigens and Flow Cytometry
heterooligomers with antiparallel orientations of the peptides
and stronger binding activities. Though it is not possible to
describe the formation of the coiled coil oligomers using a
single binding constant due to the multi-step characteristics of
the process, the higher stability of the latter pair was evident
from the following observations. While the individual peptides
EKE and KEK had random coil conformations and adopted the
coiled coil conformation only upon mixing of the two peptides
(α-helix melting Tm > 61 °C), the peptides ESE and KSK already
exhibited high helical contents as individual peptides, and upon
mixing of the two components, the stability of the coiled coil
was further increased (Tm > 95 °C). We hypothesized that the
latter pair of peptides would be more suitable for attachment of
a targeting ligand to the polymer drug carrier due to the higher
stability of the coiled coil and due to the antiparallel orienta-
tion of the peptide chains that would minimize eventual steric
hindrance.
Compared with our previous work, we have improved the
synthesis of the polymer precursors in order to obtain polymer
carriers with low dispersity (<1.2). Specifically, we used RAFT
copolymerization of HPMA with 3-(N-methacryloylglycylphe-
nylalanylleucylglycyl)thiazolidine-2-thione (Ma-GFLG-TT). The
resulting reactive copolymer PTT was submitted to reaction
with a cytostatic drug pirarubicin (Pir), and the remaining TT
groups were aminolyzed with an amino derivative of dibenzo-
azacyclooctyne (DBCO-NH2) to yield the polymer precursor PPir
(Scheme 1).
The coiled coil-forming peptides with N-terminal azide
groups were designed and synthesized as described in our
previous publications.[17,25,26] The peptide azides (EKE and
ESE) were bound to the polymer precursor PPir via metal-free
azide-alkyne cycloaddition (“click” chemistry) utilizing the reac-
tion between DBCO groups of the polymer precursor and the
azide functions of the peptides. Another part of DBCO groups
was modified with an azide derivative of biotin to enable moni-
toring of the fate of the polymer carrier both in vitro and in
vivo. The remaining DBCO groups were end-capped with N3-
PEG4-COOH to yield the polymer–drug–peptide conjugates
PEKE and PESE, which contained the peptides EKE and ESE,
respectively (Scheme 1).
The spleens were harvested and homogenized using a gen-
tleMACS Dissociator (Miltenyi Biotech) in flow cytometry
buffer (PBS, 2% FCS, 2 × 10−3 m EDTA). The cell suspensions
were filtered using a 70 µm cell strainer (BD Biosciences),
resuspended in flow cytometry buffer after red blood cell lysis
with ACK lysing buffer (GIBCO), and filtered again using a
30 µm cell strainer (BD Biosciences). The resultant cell suspen-
sions were blocked by 20% mouse serum for 30 min on ice and
stained with anti-c-Myc-fluorescein (Exbio, Czech Republic)
for 30 min on ice in the dark. The cells were then washed
twice in flow cytometry buffer, stained with streptavidin-eF450
(eBioscience, Czech Republic) for 20 min on ice in the dark and
washed twice in flow cytometry buffer. Finally, the cells were
stained with propidium iodide shortly before analysis on an
LSR II flow cytometer (BD Biosciences). The data were ana-
lyzed using FlowJo software (Tree Star).
2.20. Treatment of Established BCL1 Leukemia In Vivo
BALB/c mice were i.p. inoculated with 5 × 105 BCL1 cells in
250 µL of PBS on day 0. The targeted polymer–Pir conjugate
P
ESE/scFvKSK and the nontargeted control polymer conjugate
PESE were dissolved in PBS at concentrations of 6 mg mL−1
(polymer conjugate) and 3 mg mL−1 (scFv). The treatments
were then i.v. administered in three doses on days 11, 14,
and 17. One dose contained 5 mg kg−1 of polymer-bound Pir.
Another group of BALB/c mice were inoculated with BCL1 cells
and treated with free Pir on days 11, 14, and 17. One dose con-
tained 3.5 mg kg−1 Pir (estimated as equitoxic to 5 mg kg−1 of
polymer-bound Pir) in PBS with 2% of DMSO at a concentra-
tion of 0.34 mg mL−1. BALB/c mice inoculated with BCL1 cells
and injected with PBS on days 11, 14, and 17 were used as the
control group. Tumor progression and general fitness of the
mice were checked every second day, and the body weight and
survival of the mice were recorded. Each experimental group
included eight mice.
The basic physicochemical characteristics of all of the pre-
pared copolymers are summarized in Table 1. SEC chromato-
gram of the polymer–pirarubicin conjugate PESE is shown in
Figure S2 (Supporting Information) as an example.
2.21. Statistical Analysis
Statistical analysis was performed using the Log-rank test (sur-
vival graphs) or ANOVA, followed by Tukey’s multiple com-
parison test; *, ** and *** represent p-values <0.05, 0.01, and
0.001, respectively. The data were representative of at least two
experiments.
The low dispersity value of the copolymers is an important
feature of the presented polymer drug delivery system. It has
been repeatedly reported that the pharmacokinetic behavior
of a polymer therapeutic is significantly influenced by both its
molecular weight and dispersity. Polymers with a broad dis-
tribution of molecular weights contain a fraction of smaller
macromolecules with shorter blood circulation times and lower
levels of accumulation in solid tumors, whereas a fraction of
larger macromolecules show longer blood circulation times and
higher levels of tumor accumulation due to the EPR effect. If
the molecular weight of the largest macromolecules exceeds
the renal threshold, they cannot be eliminated via glomerular
filtration and may remain in organism for extended periods
of time, with unknown physiological effects.[31] Consequently,
only a limited fraction of such polydispersed polymeric drugs
3. Results and Discussion
3.1. Synthesis of the Peptides and Polymer Conjugates
In our previous work,[26] we compared the associative behavior
of two pairs of coiled coil peptides: (VAALEKE)4/(VAALKEK)4
(EKE/KEK) formed coiled coil heterodimers with randomly ori-
ented peptide chains, whereas (IAALESE)2-IAALESKIAALESE/
IAALKSKIAALKSE-(IAALKSK)2 (ESE/KSK) formed higher
©
1700173 (5 of 11)
Macromol. Biosci. 2017, 1700173
2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
Scheme 1. Synthesis of the polymer–drug–peptide conjugates PEKE and PESE.
have the optimal pharmacokinetic and therapeutic profiles. In
contrast, we can speculate that macromolecular therapeutics
with low dispersity (e.g., originating from RAFT polymeriza-
3.3. In Vitro Binding Studies
The in vitro binding efficacy of the targeted polymer conjugates
tion) should exhibit more uniform pharmacokinetics and, con-
sequently, better therapeutic efficacy. Nevertheless, it should be
noted that this is just a hypothesis, and there is not yet enough
experimental in vivo data available in the literature to support
this statement.
P
EKE/scFvKEK and PESE/scFvKSK or the polymer conjugate PEKE
or PESE mixed together with the targeting protein without the
coiled coil-forming tag (PEKE/scFv0; PESE/scFv0) was evaluated
in BCL1 cells using flow cytometry. Figure 2A,B shows a rep-
resentative example of the detection of scFv and the polymer
backbone in the samples incubated with the targeted conju-
gates. We were able to detect similar anti-Myc-FITC signals
originating from the targeting scFv in both targeted polymer
conjugates (PEKE/scFvKEK and PESE/scFvKSK). The binding of
3.2. Preparation of Recombinant Proteins and scFv-Targeted
Polymer Conjugates
streptavidin-Alexa fluor 405 to the biotin-labeled targeted PESE
scFvKSK conjugate showed a significantly higher median fluo-
rescence intensity (MFI) of Alexa 405 (MFI 16 306) than PEKE
/
The recombinant scFv fragments of the B1 antibody with
either the KEK or KSK coiled coil tag or without the tag
(scFvKEK, scFvESE, and scFv0, respectively) were expressed and
isolated from E. coli, as described earlier. The purified proteins
in PBS buffer (pH 7.4) were mixed with the corresponding
complementary polymer conjugates to yield the scFv-targeted
/
scFvKEK (MFI 3967). Figure 2C shows quantitative difference in
MFI of Alexa 405 after binding streptavidin-Alexa fluor 405 to
the biotin-labeled targeted PEKE/scFvKEK and PESE/scFvKSK con-
jugates. In accordance with our original hypothesis and previ-
ously published[26] results, we attributed this difference in cell
binding efficiency to the formation of antiparallel coiled coil
heterodimers and to the stronger interactions between ESE/
KSK peptides compared with the weaker, randomly oriented
EKE/KEK coiled coils.
supramolecular complexes PEKE/scFvKEK and PESE/scFvKSK
.
The molar ratio of EKE/KEK (and ESE/KSK) was set to 3:1,
which corresponds to a polymer/protein weight ratio of 2:1. In
addition to the physicochemical methods (size-exclusion chro-
matography and sedimentation analysis) described in our pre-
vious papers, the formation of the supramolecular complexes
polymer–protein was also confirmed by dynamic light scat-
tering (Figure 1).
The cells incubated with the nontargeted polymers mixed
with the control protein without the coiled coil tag (PEKE+scFv0;
PESE+scFv0) exhibited only the signal of scFv binding to the
©
1700173 (6 of 11)
Macromol. Biosci. 2017, 1700173
2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
Figure 1. Particle size distribution and hydrodynamic diameters obtained by DLS analysis. Green line, scFvKSK; blue line, PESE; red line, PESE/scFvKSK
complex.
cells, and no signal corresponding to the polymer was detected
(data not shown).
exhibited cytotoxicities that were ten times and eight times
higher, respectively, than those of the nontargeted conjugates
PESE/scFv0 and PEKE/scFv0.
In general, the targeted polymer conjugates show signifi-
3.4. Cytostatic Activity of the Polymer Conjugates In Vitro
cantly higher cytotoxic effects in vitro than the corresponding
nontargeted conjugates. The cytotoxicities of both the non-
In accordance with the results of flow cytometry cell binding
studies, the in vitro evaluation of the cytotoxicity of the tar-
geted conjugates revealed that the cytotoxicity of PESE/scFvKSK
was two times greater than that of PEKE/scFvKEK (Table 2). The
targeted polymer conjugates PESE/scFvKSK and PEKE/scFvKEK
targeted polymer conjugates differed only slightly, as PESE
scFv0 exhibited somewhat higher cytotoxic effects than PEKE
scFv0. We speculated that this difference might be explained
by different interactions between the polymer-borne peptide
sequences and the cell membrane. The dependence of the
/
/
Figure 2. Flow cytometry analysis of BCL1 cell binding by the targeted conjugates A) PEKE/scFvKEK and B) PESE/scFvKSK. Along the y-axis, the dot plot
shows anti-Myc-FITC signals (orange) indicating the binding of scFv to the cells; along the x-axis, the streptavidin-Alexa 405 signal (blue) indicates the
binding of the polymer to the cells and nonlabeled population of cells (red). C) Quantification of MFI of the cell binding experiment.
©
1700173 (7 of 11)
Macromol. Biosci. 2017, 1700173
2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
cell viability on the concentration of the polymer conjugates
is shown in Figure S1 (Supporting Information). Free piraru-
bicin shows the highest cytotoxicity, which is a well-known fact
from in vitro studies.[25,32,33] The advantages of polymeric drugs
become evident in vivo due to their longer circulation times,
decreased side effects and the possible presence of targeting
ligands.
Based on the results of both the cell binding and cytotox-
icity studies, the following in vivo experiments were performed
using only the most efficient targeted polymer conjugate,
with the biexponential functions used for the calculations are
shown in Figure S3 (Supporting Information).
3.6. In Vivo Binding Studies
The ability of the targeted polymeric conjugate to bind to
tumor cells in vivo was determined in BALB/c mice with
developed BCL1 leukemia. Mice were i.v. injected with the
targeted polymer–Pir conjugate PESE/scFvKSK or the non-
targeted polymer conjugate PESE mixed with scFv 30 d after
inoculation with BCL1 cells, which ensured that the spleens
of the experimental mice contained significant BCL1 tumor
cell counts. BCL1 cells were identified by double positivity
for MHC II and CD80, as BCL1 cells are known to strongly
express these markers in vivo.[34] At selected time points after
conjugate administration, the polymeric conjugates and scFv
bound to the surface of BCL1 cells were detected in spleen cell
suspensions using streptavidin-eF450 (as the polymer con-
jugates contained biotin) and anti-c-Myc mAb-FITC (as scFv
contained the Myc tag), respectively. The BCL1 cells in mice
injected with the polymer conjugate PESE mixed with scFv0
showed gradually increasing scFv0 binding for up to 6 h and
much lower binding at 24 h (Figure 4). In contrast, the BCL1
cells in mice injected with the targeted polymer–Pir conju-
gate PESE/scFvKSK showed very strong binding of both scFv
and the polymer at 1 h postinjection. The binding was some-
what lower at 2.5 and 6 h postinjection and very low at 24 h
postinjection (Figure 4). The sharp decrease in binding of the
polymer conjugate targeted with scFv in comparison with free
scFv after 24 h likely reflected the much stronger internaliza-
tion of the multivalent polymer–Pir conjugate PESE/scFvKSK
complexes compared with that of monovalent scFv. Overall,
we clearly demonstrated that scFvKSK tightly binds to the
polymer–Pir conjugate PESE and that the resulting targeted
polymer–Pir conjugate PESE/scFvKSK binds to the targeted
BCL1 cells in vivo upon i.v. administration. Our targeted
polymeric carrier bearing the cytostatic drug Pir is thus able to
selectively deliver Pir to tumor cells in mice.
PESE/scFvKSK.
3.5. Blood Clearance of the Polymer Conjugates
The persistence of the conjugates in circulation was determined
in BALB/c mice i.v. injected with the polymer–pirarubicin
conjugate PESE/scFvKSK or the polymer–pirarubicin conjugate
PESE + scFv0. The data showed that the majority (≈90%) of
injected Pir disappeared from circulation within 6 h after the
administration of polymeric conjugates (Figure 3). However, a
small fraction of injected Pir (about 1%) was still detectable in
the blood even 48 h after administration. At the 6 h time point,
a slightly slower elimination rate of Pir from the circulation
was seen when the polymeric conjugate associated with the tar-
geting scFv was used in comparison to the use of the conjugate
without scFv; however, this effect was diminished at later time
points. The slower elimination of the scFv-targeted polymer
conjugate can be most likely attributed to its higher molecular
weight (and hydrodynamic volume) compared with that of the
nontargeted polymer. The corresponding half-times T1/2α and
T1/2β characterizing the absorption and elimination phases of
the pharmacokinetics of the both polymer conjugates together
3.7. Treatment of Established BCL1 Leukemia In Vivo
The antitumor activities of the targeted polymer–pirarubicin
conjugate PESE/scFvKSK and the nontargeted control polymer
conjugate PESE mixed with scFv was tested in a BCL1 leukemia
mouse model with diffuse malignancy that did not form solid
tumors. The mice were injected with tumor cells and the con-
jugates or free pirarubicin were administered in three sepa-
rate doses on days 11, 14, and 17. Only the targeted polymeric
conjugate therapy impeded the increase in body weight of
the experimental mice, which is a sign of disease progression
(Figure 5A). This result showed that only the targeted poly-
meric conjugate was capable of inhibiting the massive outbreak
of the disease within the recorded time period (up to day 40).
The median survival times were 36.5, 42.5, 45.5, and 66.5 d
for the untreated mice, the mice treated with free pirarubicin,
the mice injected with the nontargeted polymer and the mice
injected with the targeted conjugate, respectively.
Figure 3. Pharmacokinetics of Pir in the blood of mice injected with tar-
geted and nontargeted polymer–pirarubicin conjugates. BALB/c mice
were i.v. injected with the targeted polymer–pirarubicin conjugate PESE
/
scFvKSK (red line) or the nontargeted polymer conjugate PESE (blue line)
mixed with scFv0 at doses equivalent to 75 µg of pirarubicin. Blood was
collected 1 min, 1, 6, 12, 24, and 48 h postinjection and was analyzed
using HPLC. The concentration of Pir in the blood determined at 1 min
after administration was considered 100%. Each experimental group
included three mice.
©
1700173 (8 of 11)
Macromol. Biosci. 2017, 1700173
2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
Figure 4. In vivo binding of targeted and nontargeted polymer–pirarubicin conjugates. BALB/c mice were i.p. injected with 5 × 105 BCL1 cells on day
0. Mice were i.v. injected with the targeted polymer–pirarubicin conjugate PESE/scFvKSK (PESE/scFvKSK; 5 mg kg−1 of polymer-bound pirarubicin per
dose) or the nontargeted control polymer conjugate PESE (PESE; 5 mg kg−1 of polymer-bound pirarubicin per dose) mixed with scFv on day 30. BALB/c
mice bearing BCL1 leukemia (BCL1) and i.v. injected with PBS were used as controls. Spleens were harvested 1, 2.5, 6, and 24 h after the injection of
polymeric conjugates. BCL1 cells were gated as MHC II+ CD80+ double positive cells. A) Dot plots showing binding of scFv (anti-Myc labeling) and
polymeric conjugate (streptavidin labeling) to BCL1 cells. Each dot plot shows one representative mouse. The upper row shows a control mouse, the
middle row shows mice injected with the polymer conjugate PESE mixed with scFv0, and the lower row shows mice injected with the targeted polymer
conjugate PESE/scFvKSK. B) Mean fluorescence intensities of scFv binding (dark columns) and polymer conjugate binding (empty columns); each
column shows one representative mouse from the group.
Both conjugates significantly prolonged the survival times of
the experimental mice (Figure 5B) compared with both mice
treated with the free drug and the untreated control. Though
there was a statistically nonsignificant difference in the sur-
vival times of mice treated with the nontargeted and targeted
conjugates, therapy with the targeted conjugate led to a com-
plete cure in one experimental mouse and markedly prolonged
survival of another five animals. Thus, the targeted polymeric
conjugate proved to be the most efficient treatment modality,
and the results reflected the ability of the targeted polymeric
conjugate to selectively deliver the cytostatic drug to the cancer
cells.
4. Conclusions
We synthesized hydrophilic polymer conjugates with narrow
molecular weight distributions containing the anticancer drug
pirarubicin bound to the polymer backbone via an enzymatically
cleavable tetrapeptide spacer. A recombinant antibody fragment
©
1700173 (9 of 11)
Macromol. Biosci. 2017, 1700173
2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
Figure 5. Treatment of BCL1 leukemia with polymer conjugates. BALB/c mice were i.p. injected with 5 × 105 BCL1 cells on day 0. Mice were i.v. injected
with three doses of targeted polymer–pirarubicin conjugate PESE/scFvKSK (PESE/scFvKSK; 5 mg kg−1 of polymer-bound pirarubicin per dose), nontargeted
control polymer conjugate PESE (PESE; 5 mg kg−1 of polymer-bound pirarubicin per dose) mixed with scFv0, or free pirarubicin (Pir; 3.5 mg kg−1) on days
11, 14, and 17. BALB/c mice injected with BCL1 cells and treated with PBS were used as controls. A) Changes in relative body weights of experimental
mice. B) Survival of experimental mice. Groups were compared using ANOVA followed by A) Tukey’s multiple comparison test or B) Log-rank test; *, **,
and *** represent p-values <0.05, 0.01, and 0.001, respectively. Eight mice per group were used. The experiment was repeated twice with similar results.
that specifically binds to leukemia cells was attached to the
polymer–drug conjugate via a universal noncovalent coiled coil
cancer therapy, coiled coil, drug targeting, hydrophilic polymer, scFv
interaction. The major advantage of the coiled coil approach
Keywords
compared with traditional covalent conjugation methods lies
Received: May 16, 2017
in the well-defined and absolutely nondestructive preparation
of the polymer–protein complex. It was demonstrated that the
choice of the coiled coil linker between the protein and the pol-
ymer can significantly affect both the cell binding efficiency of
the targeted polymer–drug conjugate and its cytotoxic activity
against the target malignant cells. The superior therapeutic
efficiency of the scFv-targeted polymer cancerostatic com-
pared with the low-molecular weight drug and the nontargeted
polymer–drug conjugate was demonstrated in vivo using a
murine BCL1 leukemia model.
Revised: July 17, 2017
Published online:
[1] H. Maeda, H. Nakamura, J. Fang, Adv. Drug Delivery Rev. 2013, 65,
71.
[2] J. Kopecˇek, Z.-R. Lu, P. Kopecˇková, Nat. Biotechnol. 1999, 17, 1101.
[3] C. M. Ward, M. Pechar, D. Oupický, K. Ulbrich, L. W. Seymour,
J. Gene Med. 2002, 4, 536.
[4] R. Pola, M. Studenovský, M. Pechar, K. Ulbrich, O. Hovorka,
ˇ
D. Veˇtvicˇka, B. Ríhová, J. Drug Targeting 2009, 17, 763.
We believe that targeted polymer cancerostatics utilizing
noncovalent interactions of the two complementary peptides
between the polymer carrier and the targeting protein ligand
represent a highly promising new type of nanomedicine. The
approach used herein might help to overcome not only the
drawbacks of current chemotherapies, such as the general non-
specific toxicity, but also frequent problems with the clinical
approval of nanomedicines by regulatory authorities due to the
low uniformity and poorly defined structures of polymer–pro-
tein conjugates prepared by more traditional covalent methods.
[5] Z.-R. Lu, S.-Q. Gao, P. Kopecˇková, J. Kopecˇek, Bioconjugate Chem.
2000, 11, 3.
[6] J. Hongrapipat, P. Kopecˇková, J. Liu, S. Prakongpan, J. Kopecˇek,
Mol. Pharmaceutics 2008, 5, 696.
ˇ
[7] B. Ríhová, P. Kopecˇková, J. Strohalm, P. Rossmann, V. Veˇtvicˇka,
J. Kopecˇek, Clin. Immunol. Immunopathol. 1988, 46, 100.
[8] M. Kovárˇ, T. Mrkvan, J. Strohalm, T. Etrych, K. Ulbrich, M. Štastný,
ˇ
B. Ríhová, J. Controlled Release 2003, 92, 315.
[9] T.-W. Chu, J. Kopecˇek, Biomater. Sci. 2015, 3, 908.
[10] K. S. Palla, L. S. Witus, K. J. Mackenzie, C. Netirojjanakul,
M. B. Francis, J. Am. Chem. Soc. 2015, 137, 1123.
[11] F. Sun, W.-B. Zhang, A. Mahdavi, F. H. Arnold, D. A. Tirrell, Proc.
Natl. Acad. Sci. USA 2014, 111, 11269.
[12] R. J. Spears, M. A. Fascione, Org. Biomol. Chem. 2016, 14, 7622.
[13] L. S. Witus, T. Moore, B. W. Thuronyi, A. P. Esser-Kahn,
R. A. Scheck, A. T. Iavarone, M. B. Francis, J. Am. Chem. Soc. 2010,
132, 16812.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
[14] S. Ulrich, D. Boturyn, A. Marra, O. Renaudet, P. Dumy, Chem. – A
Eur. J. 2014, 20, 34.
Acknowledgements
This work was supported by the Ministry of Education, Youth and Sports
of the Czech Republic within the National Sustainability Program I
(project POLYMAT LO1507) and II (project BIOCEV-FAR LQ1604), by
the Ministry of Health of the Czech Republic (project 16-28594A), and
by the Czech Science Foundation (projects 16-17207S and 13-12885S).
[15] B. Apostolovic, M. Danial, H.-A. Klok, Chem. Soc. Rev. 2010, 39,
3541.
[16] S. P. E. Deacon, B. Apostolovic, R. J. Carbajo, A. Schott, K. Beck,
M. J. Vicent, A. Pineda-Lucena, H. Klok, R. Duncan, Biomacromol-
ecules 2011, 12, 19.
[17] M. Pechar, R. Pola, R. Laga, K. Ulbrich, L. Bednárová, P. Malonˇ,
I. Sieglová, V. Král, M. Fábry, O. Vaneˇk, Biomacromolecules 2011, 12,
3645.
[18] B. Apostolovic, S. P. E. Deacon, R. Duncan, H.-A. Klok, Macromol.
Rapid Commun. 2011, 32, 11.
Conflict of Interest
The authors declare no conflict of interest.
©
1700173 (10 of 11)
Macromol. Biosci. 2017, 1700173
2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.mbs-journal.de
[19] K. Wu, J. Yang, J. Liu, J. Kopecˇek, J. Controlled Release 2012, 157, 126.
[20] K. Wu, J. Liu, R. N. Johnson, J. Yang, J. Kopecˇek, Angew. Chem.,
Int. Ed. 2010, 49, 1451.
[27] H. Nakamura, E. Koziolová, T. Etrych, P. Chytil, J. Fang, K. Ulbrich,
H. Maeda, Eur. J. Pharm. Biopharm. 2014, 90, 90.
ˇ
[28] K. Ulbrich, V. Šubr, J. Strohalm, D. Plocová, M. Jelínková, B. Ríhová,
[21] T.-W. Chu, R. Zhang, J. Yang, M. P. Chao, P. J. Shami, J. Kopecˇek,
J. Kopecˇek, Theranostics 2015, 5, 834.
[22] T.-W. Chu, K. M. Kosak, P. J. Shami, J. Kopecˇek, Drug Delivery Transl.
Res. 2014, 4, 389.
[23] T.-W. Chu, J. Yang, R. Zhang, M. Sima, J. Kopecˇek, ACS Nano 2014,
8, 719.
J. Controlled Release 2000, 64, 63.
[29] R. Pola, O. Janoušková, T. Etrych, Physiol. Res. 2016, 65, 225.
[30] S. Perrier, P. Takolpuckdee, J. Westwood, D. M. Lewis, Macromol-
ecules 2004, 37, 2709.
ˇ
[31] T. Etrych, V. Šubr, J. Strohalm, M. Šírová, B. Ríhová, K. Ulbrich,
J. Controlled Release 2012, 164, 346.
[24] R. A. Willemsen, M. Pechar, R. C. Carlisle, E. Schooten, R. Pola,
A. J. Thompson, L. W. Seymour, K. Ulbrich, Pharm. Res. 2010, 27, 2274.
[25] R. Pola, R. Laga, K. Ulbrich, I. Sieglová, V. Král, M. Fábry,
M. Kabešová, M. Ková , M. Pechar, Biomacromolecules 2013, 14,
881.
[26] M. Pechar, R. Pola, R. Laga, A. Braunová, S. K. Filippov,
A. Bogomolova, L. Bednárová, O. Vaneˇk, K. Ulbrich, Biomacromol-
ecules 2014, 15, 2590.
[32] H. Nakamura, E. Koziolová, P. Chytil, K. Tsukigawa, J. Fang,
M. Haratake, K. Ulbrich, T. Etrych, H. Maeda, Mol. Pharmaceutics
2016, 13, 4106.
[33] R. Laga, O. Janoušková, K. Ulbrich, R. Pola, J. Blažková,
S. K. Filippov, T. Etrych, M. Pechar, Biomacromolecules 2015, 16, 2493.
[34] M. Kováˇr, J. Tomala, H. Chmelová, L. Kováˇr, T. Mrkvan, R. Josková,
Z. Zákostelská, T. Etrych, J. Strohalm, K. Ulbrich, M. Šírová,
ˇ
B. Ríhová, Cancer Res. 2008, 68, 9875.
©
1700173 (11 of 11)
Macromol. Biosci. 2017, 1700173
2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim