Self-Assembled Thermoresponsive Polymeric Nanogels for 19F MR Imaging

Article abstract of DOI:10.1021/acs.biomac.8b00812

Magnetic resonance imaging using fluorinated contrast agents (¹⁹F MRI) enables to achive highcontrast in images due to the negligible fluorine background in living tissues. In this pilot study, we developed new biocompatible, temperature-responsive, and easily synthesized polymeric nanogels containing a sufficient concentration of magnetically equivalent fluorine atoms for ¹⁹F MRI purposes. The structure of the nanogels is based on amphiphilic copolymers containing two blocks, a hydrophilic poly[N-(2-hydroxypropyl)methacrylamide] (PHPMA) or poly(2-methyl-2-oxazoline) (PMeOx) block, and a thermoresponsive poly[N(2,2difluoroethyl)acrylamide] (PDFEA) block. The thermoresponsive properties of the PDFEA block allow us to control the process of nanogel self-assembly upon its heating in an aqueous solution. Particle size depends on the copolymer composition, and the most promising copolymers with longer thermoresponsive blocks form nanogels of suitable size for angiogenesis imaging or the labeling of cells (approximately 120 nm). The in vitro ¹⁹F MRI experiments reveal good sensitivity of the copolymer contrast agents, while the nanogels were proven to be noncytotoxic for several cell lines.

Full text of DOI:10.1021/acs.biomac.8b00812

Subscriber access provided by READING UNIV
Article
19
Self-assembled Thermoresponsive Polymeric Nanogels for F MR Imaging
Kristyna Kolouchova, Ondrej Sedlacek, Daniel Jirak, David Babuka, Jan Blahut, Jan Kotek, Martin Vít,
Jiri Trousil, Rafal Konefal, Olga Janoušková, Bohumila Podhorska, Miroslav Slouf, and Martin Hruby
Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.8b00812 • Publication Date (Web): 16 Jul 2018
Downloaded from http://pubs.acs.org on July 22, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 42
Biomacromolecules
1
2
3
4
5
6
7
8
9
Selfꢀassembled Thermoresponsive Polymeric
Nanogels for 19F MR Imaging
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Kristyna Kolouchova, 1 Ondrej Sedlacek,1,2 Daniel Jirak, 3,5 David Babuka, 1 Jan Blahut, 4 Jan
Kotek, 4 Martin Vit, 3, 6 Jiri Trousil, 1,7 Rafał Konefał, 1 Olga Janouskova, 1 Bohumila Podhorska,
1 Miroslav Slouf,1 Martin Hruby 1*
1 Institute of Macromolecular Chemistry AS CR, v.v.i., Heyrovského sq. 2, Prague 6, 162 06,
Czech Republic
2 Department of Organic and Macromolecular Chemistry, Ghent University, Krijgslaan 281ꢀS4,
9000 Ghent, Belgium.
3 Institute for Clinical and Experimental Medicine, Vídeňská 9, Prague 4, 140 21, Czech
Republic
4 Department of Inorganic Chemistry, Faculty of Science, Charles University, Hlavova 8, Prague
2, 128 00, Czech Republic
5 Institute of Biophysics and Informatics, 1st Medicine Faculty, Charles University, Salmovská 1,
Prague, 120 00, Czech Republic
ACS Paragon Plus Environment
1

Biomacromolecules
Page 2 of 42
1
2
3
4
5
6 TU Liberec, Faculty of mechatronics, informatics and interdisciplinary studies, Studentská
1402/2 Liberec 1,461 17, Czech Republic
6
7
8
9
7 Department of Analytical Chemistry, Charles University, Faculty of Science, Hlavova 8, 128 43
Prague 2, Czech Republic
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
*ꢀcorresponding author, eꢀmail: mhruby@centrum.cz
Keywords: Self-assembly, Thermoresponsive, Nanogel, 19F MRI, Contrast agent,
Angiogenesis
Graphical Abstract
ACS Paragon Plus Environment
2

Page 3 of 42
Biomacromolecules
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Abstract
Magnetic resonance imaging using fluorinated contrast agents (19F MRI) enables enhanced
resolution due to the negligible fluorine background in living tissues. In this pilot study, we
developed new biocompatible, temperatureꢀresponsive and easily synthetized polymeric
nanogels containing a sufficient concentration of magnetically equivalent fluorine atoms
for 19F MRI purposes. The structure of the nanogels is based on amphiphilic copolymers
containing
two
blocks:
a
hydrophilic
poly[Nꢀ(2ꢀhydroxypropyl)methacrylamide]
thermoresponsive
(PHPMA) or poly(2ꢀmethylꢀ2ꢀoxazoline) (PMeOx) block, and
a
poly[N(2,2difluoroethyl)acrylamide] (PDFEA) block. The thermoresponsive properties
of the PDFEA block allow us to control the process of nanogel selfꢀassembly upon its heating
in an aqueous solution. Particle size depends on the copolymer composition, and the most
promising copolymers with longer thermoresponsive blocks form nanogels of suitable size
for angiogenesis imaging or the labelling of pancreatic islets (approx. 120 nm). The in vitro
ACS Paragon Plus Environment
3

Biomacromolecules
Page 4 of 42
1
2
3
4
5
19F MRI experiments reveal good sensitivity of the copolymer contrast agents, while
the nanogels were proven to be nonꢀcytotoxic for several cell lines.
6
7
8
9
Introduction
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Contrast agents for medical imaging methods such as magnetic resonance imaging (MRI),
positron emission tomography (PET), computed tomography (CT), photon emission computed
tomography (SPECT) or ultrasound techniques are crucial for the diagnosis of cancer and other
diseases.1 The widely used 1H MRI possesses advantages such as noninvasiveness,
highꢀresolution anatomical imaging and practically no limitation due to sample penetration.
1
However, the main disadvantage of H MRI is a low capability to differentiate healthy tissue
from diseased due to the small differences in relaxation times of protons within these tissue
types. This disadvantage used to be at least partly resolved by using T1ꢀcontrast
agents, including paramagnetic species, which contain paramagnetic metal (mostly GdIII) that
accelerates the spin relaxation of surrounding water protons,2,3,4 or superparamagnetic iron oxide
nanoparticles (SPIONs), serving as T2ꢀcontrast agents.
1
Our approach is focused on 19F MRI, which has similar benefits as H MRI. In addition,
19F MRI using fluorinated contrast agents enables enhanced resolution (“hotꢀspot” imaging)
1
due to the very low (negligible) MRꢀactive fluorine background in living tissues. H MRI
and 19F MRI can be performed simultaneously, as fluorine atoms have similar resonance
frequencies as protons,5,6,7 combining anatomical imaging (1H) with precise and selective
positioning of the contrast agent (19F). Another benefit of fluorinated contrast agents over the use
of standard paramagnetic species is their nonꢀtoxicity and stability with no risk of decomposition
or release of toxic metal ions.8
ACS Paragon Plus Environment
4

Page 5 of 42
Biomacromolecules
1
2
3
4
5
6
7
8
9
Currently, numerous contrast agents suitable for 19F MRI are available for applications
including cell targeting,9,10 plaque detection or suppression,11,12 pO2 sensitivity,13,14 monitoring
of inflammation15 or antiproliferative drug delivery systems.16 The currently used contrast agents
are based on various designs such as polyfluorinated nanoemulsions,17 drug delivery systems
with fluorinated hydrocarbons as liposomal fluorinated particles,18 and fluorinated dendrimers,
hyperbranched polymers19,20 or micelles.21 These systems suffer from various disadvantages,
such as the distribution of signals due to multiple equivalent atoms of fluorine or the low
solvation of fluorine atoms, either leading to low resolution or long processing time.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The main principle of the imaging of angiogenesis is based on the Enhanced Permeability and
Retention (EPR) effect.22 Due to the tumorꢀstimulated angiogenesis, the primary tumor and
metastatic growth generate new blood vessels to better supply nutrients to tumor cells.23 The
newly formed vessels are more permeable for larger particles (up to approx. 200 nm). In
addition, these particles are retained in the tumor due to the lack of lymphatic vessels in the
tumor mass. This effect is also observed in some inflammation states (where it helps the immune
system to reach such places). Nanosized particles such as nanogels could be also used for
labeling of pancreatic islets by microporation or endocytosis.24
Herein, we describe for the first time selfꢀassembled thermoresponsive polymer nanogels
suitable for 19F MRI with a high loading of magnetically equivalent fluorine atoms. The structure
of the nanogels is based on amphiphilic copolymers containing two blocks: a hydrophilic
biocompatible poly[Nꢀ(2ꢀhydroxypropyl)methacrylamide] (PHPMA) or poly(2ꢀmethylꢀ2ꢀ
oxazoline) (PMeOx) block, sterically stabilizing the system in aqueous media and protecting the
thermoresponsive fluorinated core against unwanted nonspecific interaction with biological
environments, and a thermoresponsive poly[Nꢀ(2,2ꢀdifluoroethyl)acrylamide] (PDFEA) block.
ACS Paragon Plus Environment
5

Biomacromolecules
Page 6 of 42
1
2
3
4
5
6
7
8
9
According to our method of preparation, these nanogels are physically prepared via
selfassembly, provided by nonꢀcovalent bonding, such as hydrophobic interactions and hydrogen
bonding, instead of the more common chemical crossꢀlinking. Moreover, these physically selfꢀ
assembled particles show better biodegradability, due to reversibility of the phase transition,
which is not the case for crosslinking by covalent bonds.25 The existence of one very sharp signal
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
19
as the outcome from F MRI, due to the high content of magnetically equivalent fluorine atoms
and their relatively high solvation compared to various systems with overly crystalline cores,
makes these polymers very promising for further applications in MR imaging techniques.
Another benefit of the improved solvation of the polymer chains in nanogels is presupposing
physical degradability.26 The thermoresponsive properties of the PDFEA block allow us to
control the process of nanoparticle selfꢀassembly upon its heating in an aqueous solution. At
laboratory temperatures, the system remains soluble as unimers, whereas the nanoparticles
assemble upon heating to body temperature, which is the most controlled way of producing wellꢀ
defined nanoparticles without using any nanoprecipitation method before injection.27, 28 The
molecular weight of the polymer chains (i.e., unimers in dynamic equilibrium with the assembled
structures) forming nanogels is kept below the threshold of renal elimination to allow for the
clearing of the polymer system after the completion of its task.
Experimental section
Materials. (RS)ꢀ1ꢀAminoꢀpropanꢀ2ꢀol, methacryloyl chloride (≥97%), tertꢀbutanol (tꢀBuOH,
anhydrous, ≥99.5%), 2,2’ꢀazobis(2ꢀmethylbutyronitrile) (AIBN, ≥98%), dimethyl sulfoxide
(DMSO, anhydrous, ≥99.9%), 2ꢀmethylꢀ2ꢀoxazoline (MeOx, 98%), allyl alcohol (≥99%),
4ꢀnitrobenzenesulfonyl chloride (97%), triethylamine (TEA, ≥99%), acetonitrile (ACN,
ACS Paragon Plus Environment
6

Page 7 of 42
Biomacromolecules
1
2
3
4
anhydrous, 99,8%), 4ꢀcyanoꢀ4ꢀ(phenylcarbonothioylthio)pentanoic acid
(CTA1,>97%),
5
6
7
8
4ꢀcyanoꢀ4ꢀ[(dodecylsulfanylthiocarbonyl)sulfanyl]pentanoic acid (CTA2, 97%), acryloyl
chloride (97,0%), N,Nꢀdimethylformamide (anhydrous, 99,8%) and silica gel 60 were purchased
from SigmaꢀAldrich spol. s.r.o. (Czech Republic). 2,2ꢀDifluoroethylamine (97%) was purchased
from Fluorochem (Chempur Feinchemikalien GmbH, Germany). All other chemicals and
solvents were of analytical grade. Solvents and liquid chemicals were dried and purified by
conventional procedures or distillation before use.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of monomers and initiators. Nꢀ(2ꢀHydroxypropyl)methacrylamide (HPMA) was
prepared by the reaction of methacryloyl chloride with 1ꢀaminoꢀpropanꢀ2ꢀol and sodium
carbonate in dichloromethane according to the reference.29
Nꢀ(2,2ꢀDifluoroethyl)acrylamide (DFEA) was synthesized by the reaction of acryloyl chloride
with 2,2ꢀdifluoroethylamine in tetrahydrofuran (THF) in the presence of triethylamine (TEA),
according to the reference.30
Allylꢀ4ꢀnitrophenylꢀsulfonate was synthesized by
a
reaction of allyl alcohol
with 4ꢀnitrobenzenesulfonyl chloride, in the presence of triethylamine. 4ꢀNitrobenzenesulfonyl
chloride (1.79 g, 8.09 mmol, 1.1 eq.) was dissolved in dry dichloromethane (40 mL) under an
argon atmosphere and cooled in a dry ice/acetonitrile bath to ꢀ30 °C. Triethylamine (1.27 mL,
8.09 mmol, 1.1 eq) was added, followed by the dropwise addition of allyl alcohol (0.50 mL,
7.35 mmol) in dry dichloromethane (5 mL). The reaction mixture was allowed to warm to 0 °C
using an ice bath and kept at that temperature for 2 h. Afterwards, the reaction products were
washed with water, dried with anhydrous magnesium sulfate and purified
by column chromatography on silica (mobile phase hexane:DCM 1:1 v/v).
ACS Paragon Plus Environment
7

Biomacromolecules
Page 8 of 42
1
2
3
4
5
1H NMR (CDCl3, 300 Hz): δ 4.66 (m, CH2), 5.34 (m, CH2), 5.83 (m, CH2), 8.12 (d, CH),
8.41 (d, CH)
6
7
8
9
13C NMR (CDCl3, 300 Hz): δ 92.0 (CH2), 121.4 (CH2), 124.6 (CH), 129.4 (CH), 129.7 (CH),
142.3 (C), 150.1 (C)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of homopolymers. Poly[Nꢀ(2ꢀhydroxypropyl)methacrylamide] (PHPMA) was
prepared
consisted
by
of
RAFT
polymerization.
The
polymerization
g, 20.97
mixture
mmol),
Nꢀ(2ꢀhydroxypropyl)methacrylamide
(3
4ꢀcyanoꢀ4ꢀ(phenylcarbonothioylthio)pentanoic acid (CTA1, 48 mg, 0.17 mmol) and AIBN
(18 mg, 0.11 mmol) in tertꢀbutanol (24 mL). The reaction mixture was flushed with argon
in a dried Schlenk flask and polymerized for 6 h in an oil bath heated to 70 °C. Afterwards,
the resulting polymer was precipitated in diethyl ether and purified by gel filtration using
a Sephadex LHꢀ20 column with methanol as the eluent. The polymerꢀcontaining fractions were
evaporated under reduced pressure, and the polymer was isolated by freezeꢀdrying, yielding 1.9 g
(63 %) of macroCTA1. The concentration of CTA1 in macroCTA1 was determined by UV/VIS
to be 0.04 mmol g−1 CTA1 in PHPMA (λ = 334 nm, ε = 3.71 × 103 mol−1 cm−1).
Poly(2ꢀmethylꢀ2ꢀoxazoline) (PMeOx) was prepared by ringꢀopening living cationic
polymerization. 2ꢀMethylꢀ2ꢀoxazoline was purified by distillation from calcium hydride.
The mixture of 2ꢀmethylꢀ2ꢀoxazoline (4 mL, 47.01 mmol) with allylꢀ4ꢀnitrophenylꢀsulfonate
(120 mg, 0.49 mmol) in dry acetonitrile (12 mL) was purged with argon in a dried
Ace pressure tube and heated to 120 °C in an oil bath for 30 min.
After
polymerization,
the
reaction
temperature
was
lowered
to
70
°C
and cyano4[(dodecylsulfanylthiocarbonyl)sulfanyl]pentanoic acid (CTA2, 218 mg, 0.54 mmol,
1.1 eq.) and anhydrous triethylamine (0.17 mL, 1.18 mmol, 2.2 eq.) were added under an argon
ACS Paragon Plus Environment
8

Page 9 of 42
Biomacromolecules
1
2
3
4
5
6
7
8
atmosphere, and the mixture was stirred for an additional 2 h at room temperature. The resulting
polymer was precipitated in diethyl ether and purified by gel filtration using a Sephadex LHꢀ20
with methanol as the eluent. Methanol was evaporated under reduced pressure, and the polymer
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
was
lyophilized
from
an
aqueous
solution
yielding
2.21
g
(55
%)
of macroCTA2. The concentration of CTA2 in macroCTA2 was determined by UV/VIS and was
found to be 0.03 mmol g−1 CTA2 in PMeOx (λ = 308 nm, ε = 10.9 × 103 mol−1 cm−1).
Synthesis of diblock copolymers. The three diblock copolymers of different block ratios
of poly[Nꢀ(2ꢀhydroxypropyl)methacrylamide] (PHPMA) were prepared by RAFT
polymerization. The three reaction mixtures containing macroCTA1 (0.5 g), AIBN (3 mg,
0.02 mmol) and different amounts of Nꢀ(2,2ꢀdifluoroethyl)acrylamide (0.38 g, 0.75 g or 1.50 g,
i.e., 2.81 mmol, 5.56 mmol and 11.11 mmol, respectively) in dry DMF (4.5 mL) were flushed
with argon and polymerized overnight in an oil bath heated to 70 °C.
After polymerization, the resulting polymers were precipitated into diethyl ether and purified
by gel filtration using a Sephadex LHꢀ20 column with methanol. Methanol was evaporated under
reduced pressure, and the polymers were lyophilized, yielding 0.42 g (48 %) of PHPMAꢀDFEA
2:1 (HF1), 0.71 g (56 %) of PHPMAꢀPDFEA 1:1 (HF2) and 1.10 g (55 %) of PHMAꢀPDFEA
1:2 (HF3).
The three diblock copolymers with different block ratios of poly(2ꢀmethylꢀ2ꢀoxazoline)
(PMeOx) were also prepared by RAFT polymerization. The three reaction mixtures of
macroCTA2 (0.5 g), AIBN (2.5 mg, 0.02 mmol) and different amounts of Nꢀ(2,2ꢀ
difluoroethyl)acrylamide (0.32 g, 0.64 g and 1.28 g, i.e., 2.37 mmol, 4.74 mmol and 9.48 mmol,
respectively) in dry DMF (4.7 mL) were flushed with argon and polymerized overnight in an oil
bath at 70 °C. After polymerization, the resulting polymers were precipitated in diethyl ether and
ACS Paragon Plus Environment
9

Biomacromolecules
Page 10 of 42
1
2
3
4
5
6
7
8
9
purified by gel filtration using a Sephadex LHꢀ20 column with methanol. Methanol was
evaporated under reduced pressure, and the polymers were lyophilized, yielding 0.39 g (47 %)
of PMeOxꢀPDFEA 2:1 (MF1), 0.81 g (71 %) of PMeOxꢀPDFEA 1:1 (MF2) and 1.15 g (64 %)
PMeOxꢀPDFEA 1:2 (MF3).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Removal of copolymer chain-end groups. The end groups of macroCTA1 were removed
by treating the polymer with an excess of AIBN in methanol heated to 50 °C for 3 h.
The polymer was precipitated in diethyl ether and purified by gel filtration using Sephadex LH20
in methanol (yield ≥98%).
The end groups of macroCTA2 were removed by aminolysis, followed by Michael addition,
according to the described procedure.31 Then, the polymer was precipitated in diethyl ether
and purified by gel filtration using Sephadex LHꢀ20 in methanol (yield ≥98%).
Figure S1 shows the disappearance of the CTA1 and CTA2 characteristic signals
from the NMR spectra.
13
1
13
1
H and C NMR. H and C NMR spectra were acquired using a Bruker Avance III 600
(or Avance DPX 300) spectrometer operating at 600.2ꢁMHz (300.1 MHz). The width
1
of the H NMR 90° pulse was 10ꢁꢂs (15.6 ꢂs), with a relaxation delay of 10ꢁs and acquisition
time of 2.18 s (4.95 s). All spectra are provided in the supplementary information (Figures
1
S2S7). Block ratios in the copolymers were determined by H NMR (Figures S2ꢀS7), using
the methylene protons of the CHF2 group of PDFEA (δ = 5.9), the proton coupled
with the carbon atom of the CHOH group of PHPMA (δ = 3.9) and protons of the CH3 group
of MeOx (δ = 2.1), calculated by following equations (1, 2).
ꢇ ꢈ
ꢀꢁꢀꢂꢃ ꢁ 4
=
ꢀꢄꢅꢆꢃ ꢁꢇ9ꢈ
(1)
ACS Paragon Plus Environment
10

Page 11 of 42
Biomacromolecules
1
2
3
4
5
ꢇ
ꢈ
ꢀꢂꢉꢊꢋ ꢁ 2 + 4 − ꢁꢇ6ꢈ
=
ꢀꢄꢅꢆꢃ
3 ∙ ꢁꢇ6ꢈ
6
7
8
(2)
9
Size exclusion chromatography (SEC). The numberꢀaverage molecular weight (Mn), weightꢀ
average molecular weight (Mw) and polydispersity (Đ = Mw/Mn) of polymers were analyzed
by SEC, using an HPLC Ultimate 3000 system (Dionex, USA) equipped with an SEC column
(TSKgel SuperAW3000 150 × 6 mm, 4 ꢂm). Three detectors, UV/Vis, refractive index (RI)
Optilab®ꢀrEX and multiꢀangle light scattering (MALS) DAWN EOS (Wyatt Technology Co.,
USA), were employed with a methanol and sodium acetate buffer (0.3 M, pH 6.5) mixture
(80:20 w %, flow rate of 0.5 mL min−1) as the mobile phase. RI chromatograms are shown
in the supplementary information (Figure S8).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
UV/VIS Spectrophotometry. Concentrations of endꢀgroups in the polymers with attached
CTA groups were measured using a UV/VIS spectrophotometer (Evolution 220 Spectrometer
(Thermo
Scientific,
USA).
Cyanoꢀ4ꢀ(phenylcarbonothioylthio)pentanoic
acid
endꢀgroups were determined at 334 nm (ε = 3.71 × 103 mol−1 cm−1), whereas
cyano4[(dodecylsulfanylthiocarbonyl)sulfanyl] pentanoic acid endꢀgroups were found at 308 nm
(ε = 10.9 × 103 mol−1 cm−1).
Dynamic light scattering (DLS). The hydrodynamic radii (Rh) and scattering intensity
as a function of polymer temperature were determined by the DLS method using a Zetasizer
NanoZS instrument, Model ZEN3600 (Malvern Instruments, UK). The apparent volumeꢀ
weighted hydrodynamic diameter of particles, Rh, was measured at a scattering angle
of θ = 173 °, and the DTS (Nano) program was used to evaluate the data. The measurement was
conducted with a polymer concentration of 1, 5 or 10 mg mL−1 in phosphate buffered saline
(150 mM, pH 7.4) at 8 – 50 °C; all solutions were filtered prior to measurement through a
ACS Paragon Plus Environment
11

Biomacromolecules
Page 12 of 42
1
2
3
4
5
6
7
8
9
0.22ꢂm PVDF syringe filter. The measurement step was 1 °C, and at each temperature,
the sample was equilibrated for 5 min before measurement. The acquisition time was set
to 5 min, which helped to thoroughly smooth out any fluctuations affecting the correlation curve.
Static light scattering (SLS). The radii of gyration (Rg) and the molecular weights of particles
(M) at body temperature were determined by the SLS method using an ALVꢀ6000 (ALVꢀGmbH,
Germany). The zꢀaverage radii of gyration of particles, <Rg>z, and their weight average
molecular weights, Mw, were determined from measurements at several angles
and concentrations through the analysis of Zimmꢀlike plots created by the
ALV / Static & Dynamic FIT and PLOT 4.31 10/01 programs. The measurements were
conducted in phosphate buffered saline (150 mM, pH 7.4) with polymer concentrations of 2.5,
5.0, 7.5 and 12 mg mL−1 for HF3 and 1.0, 1.25, 1.5 and 2.0 mg mL−1 for MF3 at 37 °C; all
solutions were filtered prior to measurement through a 0.22ꢀꢂm PVDF syringe filter.
The measurement angle range was 40° ꢀ 50° with a step of 10°. The experiment used 3 averaged
measurements for each angle with the acquisition time set to 1 minute. This setup reduces
the influence of random dust particles and other fluctuations on measurement and provides
us with statistical error data.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Transmission electron microscopy (TEM) and cryo-transmission electron microscopy
(Cryo-TEM). Transmission electron microscopy (TEM) of nanoparticles was performed using
a Tecnai G2 Spirit Twin 12 (FEI Company; Czech Republic) microscope equipped
with a Cryoꢀholder (Gatan; CA, USA). The samples were prepared above their LCST
temperatures as follows: In the first step, all solutions (polymer concentration 5 mg mL−1)
and tools for sample preparation (tweezers, microscopic grids, etc.) were placed in an oven
heated to 50 °C and left there for >15 min. In the second step, the solution at 50 °C was
ACS Paragon Plus Environment
12

Page 13 of 42
Biomacromolecules
1
2
3
4
5
6
7
8
9
processed in one of three ways: (a) 2 ꢂL of the solution was deposited onto a TEM carbonꢀcoated
copper grid and left to evaporate in the heated oven at 50 °C; (b) 2 ꢂL of the solution was
deposited onto a TEM carbonꢀcoated copper grid, left in the oven at 50 °C for 1 min, removed
from the oven, and then the excess solvent was removed by touching the bottom
of the copper grid with a small piece of filter paper; or (c) 3 ꢂL of the solution was deposited
onto a TEM holey carbonꢀcoated copper grid, flash frozen in liquid ethane and kept under liquid
nitrogen. More technical details about all of the aboveꢀlisted preparation methods can be found
in our previous papers.32, 33, 34 The dried samples from the first two methods (a, b) were observed
using TEM microscopy at room temperature. The frozen samples from the third method (c) were
observed using CryoꢀTEM microscopy. All micrographs are bright field images taken at 120 kV.
Critical association concentration (CAC). Solutions of the copolymers were prepared
by direct dissolution in cold phosphate buffered saline (PBS, 150 mM, pH 7.4) at a concentration
of 1 mg mL−1 PBS and subjected to serial dilution with concentrations down to 10–5 mg mL−1.
Each sample was then prepared by the addition of 10 µL of pyrene solution (1.2 × 10–4 mol L−1
in methanol) to 2 mL of the diluted solution. The final concentration of pyrene in water
thus reached 6 × 10–7 mol L−1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fluorescence spectra of the samples were recorded (90 ° angle geometry, 1 × 1 cm quartz cell)
with an FPꢀ6200 spectrofluorometer (Jasco Europe, Italy), equilibrated for 10 min at 37 °C,
using excitation at 333 nm. The intensities of the emission bands I1 at 372 nm and I3 at 383 were
then evaluated, and their ratio was plotted versus the polymer concentration. The I1/I3 ratio
remained constant at a certain polymer concentration, below which the pyrene remained
in the aqueous phase. When pyrene was sequestered into the nanoparticle core, I1/I3 decreased.
ACS Paragon Plus Environment
13

Biomacromolecules
Page 14 of 42
1
2
3
4
5
6
7
8
9
The CAC was determined as the intersection between the plateau at I1/I3 ≈ 1 and the tangent
of the curve where the I1/I3 ratio decreased with an increase in the copolymer concentration.
T1 19F NMR relaxation time. The T1 measurements were performed by a Bruker Avance III
400 MHz spectrometer using an inversion recovery pulse sequence. The temperature
of the sample was calibrated using 99.8 % CD3OD by measuring the chemical shift differences
between the two 1H NMR signals.35 Particular care was taken to ensure that the sample reached
thermal equilibrium before the data acquisition started (Table S1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cytotoxicity. Cell lines: HeLa ꢀ human cervical carcinoma cell line (kindly provided
by Dr. Melkova, First Faculty of Medicine, Charles University in Prague); J774A.1 ꢀ murine
monocyte/macrophage cell line (ECACC General Cell collection, Public Health England, UK);
HF ꢀ primary fibroblasts, SUꢀDHLꢀ5 ꢀ human B lymphoblast cell line (ATCC, LGC Standards
Sp. z.o.o. Poland). The cells were cultivated in DMEM and IMDM for SUꢀDHLꢀ5 cells
(Thermo Scientific, CR). The media were supplemented with heatꢀinactivated 10% fetal bovine
serum (FBS), 100 U mLꢀ1 penicillin and 100 ꢂg mLꢀ1 streptomycin.
Cytotoxicity of polymer conjugates: in total, 5 × 103 cells for HeLa and 10 × 103 cells
for J77A4A.1, SUꢀDHLꢀ5 and HF per well, respectively, were seeded in 100 µL of media
in 96ꢀwell flatꢀbottom plates (TPP, Czech Republic) 24 h before the addition of polymer
conjugates. The concentrations of conjugates varied in the range of 0.002 – 1 mg L−1. The cells
were cultivated for 72 h in 5 % CO2 at 37 °C. After that, 10 µL of AlamarBlue® cell viability
reagent (Life Technologies, Czech Republic) was added to each well and incubated for 4 h
at 37 °C. The active component of the AlamarBlue reagent, resazurin, was reduced to the highly
fluorescent compound resorufin only in viable cells. Its fluorescence was detected using
a Synergy Neo plate reader (BioꢀTek, Czech Republic) with excitation at 570 nm and emission
ACS Paragon Plus Environment
14

Page 15 of 42
Biomacromolecules
1
2
3
4
5
6
at 600 nm. As a control, cells cultivated in the medium without polymer conjugates were
examined.
7
8
The assay was repeated two times independently.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Hemolysis. Fresh human blood was collected from a healthy volunteer into heparinꢀcoated
vacutainers (Becton Dickinson Czechia Ltd., Prague, Czech Republic). Plasma was removed
by centrifugation at 3000 rpm for 10 min. The red blood cells (RBCs) were washed three times
with Dulbecco's PhosphateꢀBuffered Saline (DPBS). For the hemolysis experiment, the RBC
suspension was diluted with DPBS to the final concentration corresponding to full blood (1:49).
To 0.3 mL of diluted RBCs, 1.2 mL of the appropriately diluted formulation in PBS was added.
By this procedure, several dilutions of polymers were prepared (100, 200, 400 and 800 ꢂg mL−1).
Triton Xꢀ100 (1%, DPBS) and DPBS served as positive and negative controls, respectively.
After incubation at 37 °C for 3, 6 or 24 h, the samples were centrifuged (3000 rpm, 10 min).
Subsequently, supernatants were collected, and their absorbance values at 541 nm were
measured on a Synergy H1 Hybrid Reader instrument (Biotek, Winooski, USA) using Nunc Cell
Culture Microplates (Thermo Scientific Nunc, USA). The percent hemolysis of RBCs in each
sample was calculated by the following equation (3):
sample absorbance – negative control
Hemolysis =
ꢌ 100
positive control – negative control
(3)
Hemolysis assays were expressed as a percentage of the positive control (1% Triton Xꢀ100
in DPBS), which was considered to be 100 %. Hemoglobin release up to 2 % F was classified
as nonꢀhemolytic, according to the ASTM F756ꢀ08 standard.36 The result of this assay was
obtained as a mean value from three parallel samples.
ACS Paragon Plus Environment
15

Biomacromolecules
Page 16 of 42
1
2
3
4
1
19
H and F MR imaging and spectroscopy (in vitro study). All imaging and spectroscopy
5
6
7
8
9
experiments were performed on a 4.7 T MR Bruker spectrometer Bruker Biospec (Bruker
BioSpin, Germany).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MR imaging of HF3 and MF3 phantoms was performed using a wideꢀtuned
1Hꢀ19F halfꢀsaddle homemade coil. Tubes containing polymer solutions with different
19
concentrations (3.3, 5, 10 and 50 g L−1) were examined. A F turbo spin echo sequence Rapid
Acquisition with Relaxation Enhancement (RARE) for phantom imaging was used with these
sequence parameters: repetition time (TR) = 2000 ms, effective echo time (TE) = 43.5 ms,
turbo factor =16, number of acquisitions (NA) = 16, 128 and 256, field of view 5 cm, matrix
64 x 64, spatial resolution 0.6 x 0.6 mm2, slice thickness 4 mm, acquisition time (TA) = 2.8,
17.4 and 34.8 min).
1
The temperature dependence of HF3 and MF3 was measured using a wideꢀtuned Hꢀ19F
19
fiveturn homemade solenoid coil. For the temperature dependence experiment, a single pulse F
spectroscopy sequence was used with these parameters: TR = 1069 ms, AC = 32, flip angle (FA)
= 30°, TA = 34 s. The wire coil was twisted with a heating pipe with flowing hot water
for precise phantom temperature adjustment. The temperature range was 30 – 45 °C.
Each measurement was followed with a 5ꢀmin pause to achieve temperature equilibrium within
the whole phantom volume. For verification of the results, measurements at 37 °C and 45 °C
were repeated after 30 minutes.
Results and discussion
Block
copolymers
poly[Nꢀ(2ꢀhydroxypropyl)methacrylamide]ꢀblockꢀpoly[Nꢀ(2,2ꢀ
difluoroethyl)acrylamide] (HF1ꢀ3) were prepared by two successive RAFT polymerizations
ACS Paragon Plus Environment
16

Page 17 of 42
Biomacromolecules
1
2
3
4
5
6
7
8
9
of the corresponding monomers with dithiobenzoate as the chain transfer agent (CTA1).
Poly(2ꢀmethylꢀ2ꢀoxazoline) (PMeOx) was prepared by living cationic ring opening
polymerization using allylꢀ4ꢀnitrophenylꢀsulfonate as the initiator. The initiatorꢀbased allyl group
at the polymer chainꢀend allows its further modification, e.g., with targeting units, another
imaging modality or chelators.37,38 After the polymerization was complete, the cationic living
chainꢀends were treated with triethylammonium salt of trithiocarbonate. In this reaction,
the chain transfer agent is introduced to the polymer to form macroCTA2, which was used
for the RAFT polymerization of the second poly[Nꢀ(2,2ꢀdifluoroethyl)acrylamide] block
(MF1ꢀ3 copolymers). The concentration of CTA in macroCTA1 and macroCTA2 was detected
by UV/VIS at 334 nm and 308 nm, respectively. After the polymerization
of poly[N(2,2difluoroethyl)acrylamide] block, both CTAs were removed to avoid the obvious
effect of CTA endꢀgroups on polymer thermoresponsivity. The dithiobenzoate was removed
by radical reaction with AIBN, and trithiocarbonate was removed by aminolysis, followed
by Michael addition (Scheme 1). Figure S1 shows the disappearance of the CTA1 and CTA2
characteristic signals from the NMR spectra.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
17

Biomacromolecules
Page 18 of 42
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Synthesis of diblock copolymers with PHPMA or PMeOx as hydrophilic blocks
and PDFEA as the thermoresponsive block.
1
Block ratios in the copolymers were determined by H NMR (Figures S2ꢀS7), using
the methylene protons of the CHF2 group of PDFEA (δ = 5.9), the proton coupled
with the carbon atom of the CHOH group of PHPMA (δ = 3.9) and protons of the CH3 group
of MeOx (δ = 2.1). Molecular weights and polydispersity were obtained using SEC (Table 1,
Figure S8).
Table 1. Physicochemical characteristics of the prepared first blocks and corresponding diblock
copolymers and physical properties of assembled particles.
Sample
name
Block
w % of
fluorine
Rh ± σ
(nm)c
Ɖ (Mw/Mn)
Polymer
Mw (kDa)b Mn (kDa)b
CPT (°C)c
ratiosa
PHPMA
ꢀ
13.3
17.6
33.7
51.5
12.7
15.8
31.9
48.3
1.05
1.11
1.06
1.07
MacroCTA1
HF1
2:1
1:1
1:2
6.7
37
31
23
123 ± 19
35 ± 5
PHPMAꢀ
PDFEA
16.9
20.7
HF2
77 ± 52
HF3
ACS Paragon Plus Environment
18

Page 19 of 42
Biomacromolecules
1
2
3
4
5
6
7
8
9
PMeOx
ꢀ
8.1
7.2
1.12
1.07
1.06
1.08
MacroCTA2
2:1
1:1
1:2
10.1
14.0
18.7
12.7
16.2
24.5
11.9
15.3
22.8
31
30
33
670 ± 203
31 ± 12
MF1
MF2
MF3
PMeOxꢀ
PDFEA
47 ± 14
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aDetermined by NMR spectroscopy. bDetermined by SEC. cDetermined by DLS.
The thermal response of the prepared copolymers in aqueous solutions was investigated
by dynamic light scattering. Hydrodynamic radii (Rh) of the particles were measured
in the temperature range from 10 °C to 50 °C.
All copolymers exhibited temperatureꢀdependent solution behavior; above a certain
temperature, their individual molecules dissolved in PBS buffer started to assemble and form
larger particles. This cloudꢀpoint temperature (CPT) was highly dependent on the ratio between
homopolymer blocks for copolymers with PHPMA, while the copolymers with PMeOx showed
no such trend (Table 1). The higher content of the more hydrophobic block should result
in the lower CPT, but this effect is obviously also dependent on the type of hydrophilic
monomer, as it was previously described for other systems.28 The concentration of the polymer
seemed to have only a marginal influence on the CPT, even for tenꢀfold diluted samples, which
is essential for the intended use where the sample is diluted with body fluids. The values of CPT,
as well as Rh of the most prevalent particles in the solution at 37 °C, are listed in Table 1.
The representative examples of size distributions are presented in Figures 1ꢀ2, while the other
CPTs are illustrated in Figure S10. For the size distributions and dependence of CPTs
on concentration for all samples and concentrations, see the supplementary information (Figure
S9).
ACS Paragon Plus Environment
19

Biomacromolecules
Page 20 of 42
1
2
3
4
5
6
7
8
9
The reversibility of these thermally dependent changes was investigated as well.
After obtaining the data from the heating process, the polymer solution was cooled from 50 °C
back to 10 °C. When the temperature dropped below the CPT, the particles disassemble rapidly
into single molecules. Therefore, the nanoprecipitation was found to be fully reversible.
This is important for the unimerꢀnanoparticle equilibriumꢀbased degradability because polymers
that crystallize after the initial phase separation from water (e.g., poly(2ꢀisopropylꢀ2ꢀoxazoline))
may become kinetically frozen and, in fact, become nonꢀdissolvable even after cooling
below the phase separation temperature.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Selfꢀassembly of HF3 (1% polymer solution in 150 mM phosphate buffered saline,
pH 7.4): 3D graph of the dependence of the size distribution on temperature (left) and CryoTEM
image (right).
ACS Paragon Plus Environment
20

Page 21 of 42
Biomacromolecules
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Selfꢀassembly of MF1 (1% polymer solution in 150 mM phosphate buffered saline,
pH 7.4): 3D graph of the dependence of the size distribution on temperature (left) and TEM
image (right).
Samples of HF3 and MF3 (block copolymers with a block ratio of 1:2
hydrophobic/fluorinated) were further analyzed by static light scattering to obtain the molecular
weights, radii of gyration and densities of the nanoparticles formed at 37 °C. The obtained data
are listed in Table 2.
Table 2. SLS data for selected samples. Molecular weight (Mw), hydrodynamic radius (Rh)
and calculated density (ρ) of particles assembled from each block copolymer.
Sample
name
Block
Polymer
Mw ꢀ103 (kDa)b
66.8
Rg (nm)b
106
ρ
(g cmꢁ3)b
0.022
ratiosa
PHPMAꢀ
PDFEA
HF3
1:2
PMeOxꢀ
PDFEA
MF3
1:2
10.4
67
0.013
aDetermined by NMR spectroscopy. bDetermined by SLS
ACS Paragon Plus Environment
21

Biomacromolecules
Page 22 of 42
1
2
3
4
5
6
7
8
9
The DLS and SLS results together with the transmission electron microscope (TEM) images
provide comprehensive insight into the structures for each copolymer, HF3 and MF3,
and suggest that the internal structure of the nanoparticles is that of a nanogel. The first
indication is the size of the particles (by DLS and SLS), which cannot be so high that the system
would be a micelle. Hence, we determined the density of the core (Table 2), which was very low,
indicating that the system could be a nanogel or vesicle. Nevertheless, the samples cannot
be vesicular particles because of the uniform polydispersity and uniform distribution of matter
(by TEM). Another proof is the molecular weight of the particle after selfꢀassembly (Table 2),
which indicates that approx. 500 – 1000 polymer chains comprise one nanoparticle,
while a micelle contains approx. 10 – 100 polymer chains. The morphology of the selfassembled
polymer particles is summarized in Figures 1, 2 and S11. Both polymers (PMeOxꢀPDFEA and
PHPMAꢀPDFEA) formed larger particles at a ratio of 2:1 (copolymers MF1, HF1) and smaller
particles at a ratio of 1:2 (copolymers MF3, HF3). The particles of both polymers with a ratio of
1:1 were unstable during the TEM sample preparation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The size ratio between hydrophobic and hydrophilic blocks of amphiphilic copolymers has
obviously significant influence on the morphology of particles they form in aqueous solutions.
In the case of highly organized micelles and vesicles the block copolymers with smaller
hydrophobic blocks form smaller micellar particles, whereas block copolymers with larger
hydrophobic blocks form larger vesicular particles, as it has been previously described.39 This is
in direct contrast to the behavior of our nanogel particles, which exhibit only limited degree
of organization and form larger particles for polymers with smaller hydrophobic blocks
and smaller particles for polymers with larger hydrophobic blocks. The formation of stable
particles from amphiphilic block copolymers is a matter of equilibrium between forces holding
ACS Paragon Plus Environment
22

Page 23 of 42
Biomacromolecules
1
2
3
4
5
6
7
8
9
the polymer chains together and forces acting to solvate the chains with solvent molecules (e.g.
water) and cause molecular dissolution.40 Based on our experimental results we hypothesize
that our polymers with smaller hydrophobic blocks form particles, which could be described
in a similar way as the plum pudding model of an atom. Most of the volume of the particles is
formed by hydrated hydrophilic blocks whose chains are swollen and extended in aqueous
environment, and within this hydrophilic gel (the “pudding”) smaller pockets of hydrophobic
blocks (the “plums”) nonꢀcovalently bound together are placed, while their chains are contracted
to minimize their interaction with water. Similar behavior can be observed in other copolymers.41
This morphological arrangement is made possible only by the favorable size ratio of hydrophilic
to hydrophobic blocks in the polymer molecule. In the case of our other polymers in which
the hydrophobic blocks are larger than the hydrophilic blocks there is not enough hydrophilic
“pudding” covering the hydrophobic “plums” formed by the fluorinated blocks aggregating
together to form an interconnecting net with the hydrophilic domains of other particles. In result,
the hydrophilic blocks only cover the surface of the hydrophobic domains, which is why we
obtain smaller particles for these polymers and block size ratios.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The smaller particles of MF3 exhibited lower average sizes in comparison to the particles
of HF3, which was in perfect agreement with the light scattering experiments. To summarize,
the TEM and DLS measurements reveal that the copolymers with higher contents
of thermoresponsive blocks form nanoparticles (Figure 1) in a size range that is well suited
for imaging, e.g., of tumors or inflamed tissue (20 – 200 nm, approx. 80 nm in this case),
due to the EPR effect.
ACS Paragon Plus Environment
23

Biomacromolecules
Page 24 of 42
1
2
3
4
5
6
7
8
9
In order to verify the correctness of the TEM results showing the studied polymer systems
after fast drying, we also performed CryoꢀTEM microscopy visualizing the particles after fast
freezing the solution. Figure 1 shows a CryoꢀTEM micrograph of one selected system, HF3.
The DLS measurements reveal that at temperatures below the cloud point temperature (CPT),
there are always (with the exception of MF2) two dominant species present in the sample.
Smaller entities (Rh about 3ꢀ7 nm) correspond to free polymers in solution, while the larger ones
correspond to particles that are always present even in lower temperatures (Figures 1, 2, S10),
resulting from the partial agglomeration between copolymers. However, the larger particles
scatter light much more than smaller ones, so the content of larger particles can be neglected.
At the CPT, the peak representing polymers in solution immediately converts into peaks
corresponding to the nanoparticles due to nanoprecipitation, while the intensities of scattered
light show an abnormal increase. This increase corresponds to phenomenon described
as anomalous micellization behavior in diblock copolymer solutions.42, 43
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The critical association concentration (CAC) is indispensable information about the stability
of the selfꢀassembled particles. Individual CACs are deduced from the charts shown
in Figure S12. The resulting CAC for HF3 is 51.6 mg L−1 and for MF3 is 26.2 mg L−1.
19
As nanogels are more appropriate for the bioimaging purposes due to their size, the F NMR,
MRI, cytotoxicity and blood compatibility (i.e., hemolytic properties) were further studied
for the micellar samples (PHPMAꢀPDFEA and PMeOxꢀPDFEA at a ratio of 1:2, HF3 and MF3).
19
The F NMR T1 relaxation times (Table S2) were measured before and after nanoprecipitation.
The T1 relaxation time dependence on the temperature is relatively low, showing only minor
19
effects of nanoprecipitation on the relaxation behavior of the F nuclei. This information is vital
for the potential use of nanogels as contrast agents for 19F MR imaging.
ACS Paragon Plus Environment
24

Page 25 of 42
Biomacromolecules
1
2
3
4
5
6
After the nanoprecipitation, the mobility of fluorine atoms remains high, so the fluorine signals
in 19F NMR are still clearly visible.
7
8
9
To assess the biological properties of synthesized polymers, their cytotoxicity and hemolytic
properties were studied. The polymer conjugates did not show any significant cytotoxicity
in the range of polymer concentrations tested (0.002 – 1 mg mL−1). The viability of cells
incubated with even the highest polymer concentration (1 g L−1, corresponding to 5 g of polymer
per human blood pool of approx. 5 L) was not decreased below 80% for the most sensitive cell
line SuꢀDHLꢀ5 and not below 90% for the rest of the cell lines (Figure S13). These data point
towards the possible application of these polymers for their intended biomedical purposes.
The interaction of the formulations with red blood cells (RBCs) as major blood components
and one of the first contact partners after systemic administration was studied with regard to red
blood cell hemolysis. The HF3 copolymer induced no hemolysis after a 3ꢀ, 6ꢀ or 24ꢀhour
incubation at 37 °C within the tested concentration range (i.e., 100 – 800 µg mL−1).
The formulation MF3 caused no hemolysis after a 3ꢀhour incubation within the tested
concentrations. Longꢀterm hemolytic studies (6 and 24 hours), however, revealed the beginning
of hemolysis (i.e., values ranging from 2 to 4 %) in the case of samples with higher polymer
concentrations (200, 400 and 800 µg mL−1). This effect is probably due to the amphiphilic
character of the tested formulation. However, for comparison, negative controls revealed a mean
hemolysis of approx. 1 %. Moreover, the MF3 formulation at the biologically relevant
concentration of 100 µg mL−1 caused no hemolysis, even during the 6ꢀ and 24ꢀhour incubation,
suggesting that it is safe for RBCs. Hemolysis assayꢀrelated results are shown in Figure S14.
The MR spectroscopy experiment showed the dependence of the 19F MR signal on temperature
for both copolymer (HF3 and MF3) probes. HF3 possesses a stronger temperature dependence
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
25

Biomacromolecules
Page 26 of 42
1
2
3
4
5
6
7
8
9
(Figure 3) in that the MR signal rises with increasing temperature. Compared to HF3,
the temperature dependency for MF3 was not so significant (Figure 3); for 30 °C and higher
temperatures, the MR signal was practically constant, with a difference lower than 1 %.
For lower temperatures, the signal was higher. Repeated measurements at 37 °C and 45 °C
confirmed the stability of probes, as the signal intensity did not change significantly.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The resonance frequency for each measurement was stable, with frequency differences lower
than 0.5 ppm. Signal was received from 550 µl solutions with polymer concentrations of 50 g
L−1. For copolymer MF3, the abnormal signal level at 25 °C was due to a phase change.
For imaging use, an appropriate MR image was collected for each temperature.
Figure 3. Temperature dependence. MR spectra and corresponding images of copolymer HF3
and MF3 phantoms are shown for various temperatures.
ACS Paragon Plus Environment
26

Page 27 of 42
Biomacromolecules
1
2
3
4
5
6
7
8
9
The MRI sensitivity of the probes was checked by measurement with various scan times.
The MRI experiments proved the feasibility of in vitro visualization of these probes. Therefore,
we supposed that these fluorine probes might also be used in in vivo experiments. Only phantoms
with the highest concentration of polymers were detected within 2 minutes of scanning
(Figure 4). For 17ꢀminute scans, we were able to detect phantoms of all used concentrations;
however, the 19F signal from HF3 was higher than that of MF3, which corresponds
with the higher concentration of fluorine in the HF3 copolymer. The temperature change could
have a high impact on T2 relaxation times due to the possible change in mobility of fluorine
atoms, which is reflected in the unfavorable broadening of the fluorine signal.44 However
in Figure 3, it can be seen that the broadening of signal, that is, the T2 change, is insignificant
(e.g., 284 Hz at 30 °C and 255 Hz at 40 °C), due to sufficient solvation of fluorine atoms
in the nanogels.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. MR imaging of phantoms of HF3 and MF3 copolymers at 20 °C (polymer solution
in 150 mM phosphate buffered saline, pH 7.4). In the first column, there are 2 min 8 s
1
19F acquisition time images only. At the top of the second column, there is a H navigator image
19
overlapped with the 17 min 4 s F red image. At the bottom of the second column, there is
ACS Paragon Plus Environment
27

Biomacromolecules
Page 28 of 42
1
2
3
4
1
a 19F image after 17 min 4 s only. At the bottom of the third column, there is a H navigator
5
6
image overlapped with the 34 min 8 s 19F red image.
7
8
9
The nonꢀuniformity of signals shown in Figure 4 is mainly caused by a surface radiofrequency
coil that was used in the MR measurements. This coil has a high sensitivity but low
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
19
homogeneity. Nonꢀuniformity of H and F (HF3; t = 17 min 4 s; signal on the right side) is
possibly caused by the sedimentation of the 19F polymer at the bottom of the tube, while the time
for settling between protons and fluorine MRI is quite long and, in the case of HF3, which has
a CPT very close to room temperature, sedimentation could appear. In addition, Figure S15
shows graphs with the signal to noise ratio (SNR) dependence on concentration. They show
a linear dependence (that is, with higher concentrations, we observe higher 19F MR signals).
Since the thermoresponsive nanogel shows fair sensitivity for 19F MRI, mainly due to low
T2 changes and high concentrations of fluorine, it could be considered a universal system where
a variety of further modification could create new applications. There are many previously
19
published contrast agents for F MRI with various types of stimuli responsivity, for example
temperatureꢀsensitive liposomal contrast agents,18 pHꢀresponsive star polymer nanoparticles45
or fluorinated dendrimers,46 ionꢀresponsive contrast agents,47 etc. These systems have similar
functions as smart drug carrier for simultaneous diagnostic and therapy (theranostics).
In our approach, thermoresponsive function is a very simple way of preparing well defined
nanoparticles in aqueous solution, without need of any other nanoparticle preparation technique.
Conclusions
We describe for the first time selfꢀassembled thermoresponsive polymer nanogels suitable
for 19F MRI with a high loading of magnetically equivalent fluorine atoms. The structure
of the nanogels is based on amphiphilic copolymers containing two blocks: one hydrophilic
biocompatible block and one fluorinated thermoresponsive block. Depending on the block ratio,
ACS Paragon Plus Environment
28

Page 29 of 42
Biomacromolecules
1
2
3
4
5
6
7
8
9
the size of the nanogels formed after heating from room to body temperature may be fineꢀtuned
in a wide range. The in vitro 19F MRI experiments reveal the excellent sensitivity
of the copolymer contrast agents, while the nanogels were proven to be nonꢀcytotoxic for several
cell lines.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Associated content
1
1
Supporting Information. Charts with H NMR spectra of diblock copolymers and H NMR
spectra of CTA groups; summary from size exclusion chromatography (SEC); 3D graphs of the
dependence of the size distribution on temperature measured by dynamic light scattering (DLS);
charts of the dependence of relative count rates on temperature; transmission electron
microscopy (TEM) micrographs; summary tables from 19F NMR (T1 relaxation time for different
temperatures); charts showing the critical association concentration (CAC), summary of the
cytotoxicity and hemolysis assays and the signal to noise ratio dependence on polymer
concentration.
Acknowledgements
The authors acknowledge financial support from the Czech grant foundation (grant #
16ꢀ03156S). Author Jiri Trousil acknowledges financial support from Charles University (project
no. SVV260440).
References
(1)
Nam, S. Y.; Ricles, L. M.; Suggs, L. J.; Emelianov, S. Y. Imaging Strategies for Tissue
Engineering Applications. Tissue Eng. Part B. Rev. 2015, 21, 88–102.
(2)
Bruckman, M. A.; Yu, X.; Steinmetz, N. F. Engineering Gdꢀloaded nanoparticles to
enhance MRI sensitivity via T1 shortening. Nanotechnology 2013, 24, 462001–462021.
ACS Paragon Plus Environment
29