Polymer Cancerostatics Targeted by Recombinant Antibody Fragments to GD2-Positive Tumor Cells

Article abstract of DOI:10.1021/acs.biomac.8b01616

A water-soluble polymer cancerostatic actively targeted against cancer cells expressing a disialoganglioside antigen GD2 was designed, synthesized and characterized. A polymer conjugate of an antitumor drug doxorubicin with a N-(2-hydroxypropyl)methacrylamide-based copolymer was specifically targeted against GD2 antigen-positive tumor cells using a recombinant single chain fragment (scFv) of an anti-GD2 monoclonal antibody. The targeting protein ligand was attached to the polymer-drug conjugate either via a covalent bond between the amino groups of the protein using a traditional nonspecific aminolytic reaction with a reactive polymer precursor or via a noncovalent but highly specific interaction between bungarotoxin covalently linked to the polymer and the recombinant scFv modified with a C-terminal bungarotoxin-binding peptide. The GD2 antigen binding activity and GD2-specific cytotoxicity of the targeted noncovalent polymer-scFv complex proved to be superior to the covalent polymer-scFv conjugate.

Full text of DOI:10.1021/acs.biomac.8b01616

Subscriber access provided by University at Buffalo Libraries
Article
POLYMER CANCEROSTATICS TARGETED BY RECOMBINANT
ANTIBODY FRAGMENTS TO GD2-POSITIVE TUMOR CELLS
Robert Pola, Vlastimil Král, Sergey K. Filippov, Leonid Kaberov, Tomáš
Etrych, Irena Sieglova, Juraj Sedlacek, Milan Fábry, and Michal Pechar
Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.8b01616 • Publication Date (Web): 28 Nov 2018
Downloaded from http://pubs.acs.org on December 3, 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 37
Biomacromolecules
1
2
3
4
5
6
7
8
9
POLYMER CANCEROSTATICS TARGETED BY
RECOMBINANT ANTIBODY FRAGMENTS TO
GD2-POSITIVE TUMOR CELLS
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
Robert Pola1‡, Vlastimil Král2‡, Sergey K. Filippov1, Leonid Kaberov1, Tomáš Etrych1, Irena
Sieglová2, Juraj Sedláček2, Milan Fábry2, Michal Pechar1*
1Institute of Macromolecular Chemistry, Czech Academy of Sciences, Heyrovského nám. 2, 162
06 Prague 6, Czech Republic
2Institute of Molecular Genetics, Czech Academy of Sciences, Flemingovo nám. 2, 166 10
Prague 6, Czech Republic
KEYWORDS
polymer cancerostatics, drug targeting, scFv, bungarotoxin, doxorubicin, GD2 antigen
ABSTRACT
A water-soluble polymer cancerostatic actively targeted against cancer cells expressing a
disialoganglioside antigen GD2 was designed, synthesized and characterized. A polymer
conjugate of an anti-tumor drug doxorubicin with a N-(2-hydroxypropyl)methacrylamide-based
copolymer was specifically targeted against GD2 antigen-positive tumor cells using a
1
ACS Paragon Plus Environment

Biomacromolecules
Page 2 of 37
1
2
3
4
5
6
7
8
9
recombinant single chain fragment (scFv) of an anti-GD2 monoclonal antibody. The targeting
protein ligand was attached to the polymer-drug conjugate either via a covalent bond between the
amino groups of the protein using a traditional non-specific aminolytic reaction with a reactive
polymer precursor or via a non-covalent but highly specific interaction between bungarotoxin
covalently linked to the polymer and the recombinant scFv modified with a C-terminal
bungarotoxin-binding peptide. The GD2 antigen binding activity and GD2-specific cytotoxicity
of the targeted non-covalent polymer-scFv complex proved to be superior to the covalent
polymer-scFv conjugate.
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. INTRODUCTION
Conjugates of hydrophilic polymers with cytostatic drugs offer numerous advantages in the
treatment of malignancies compared to low-molecular weight cancerostatics.1 The polymer
conjugates generally exhibit higher solubility, prolonged blood circulation and increased
accumulation in solid tumors. The choice of a specific targeting ligand can further increase the
therapeutic potential of these polymer-drug conjugates. Among the numerous targeting ligands
described, antibodies and antibody fragments are the most efficient. However, well-defined
covalent attachment of a complex protein molecule to a multivalent polymer carrier is
challenging. Therefore, we have investigated various covalent and non-covalent methods of
polymer-protein conjugation. Recently, we developed2,3 an anchoring system based on the
formation of a coiled-coil heterodimer between two complementary peptide sequences. The
peptide (IAALESE)2-IAALESKIAALESE was covalently attached to a synthetic hydrophilic
copolymer based on N-(2-hydroxypropyl)methacrylamide (HPMA) bearing a cytostatic drug
doxorubicin (Dox) and/or a diagnostic label; peptide IAALKSKIAALKSE-(IAALKSK)2 was
incorporated into the C-terminal part of a recombinant single chain antibody fragment (scFv) of
2
ACS Paragon Plus Environment

Page 3 of 37
Biomacromolecules
1
2
3
4
5
6
7
8
9
the B1 monoclonal antibody that specifically binds to murine BCL1 leukemia cells. The
polymer-protein complex formed upon mixing of the polymer-Dox conjugate with scFv
exhibited significantly higher specific cytotoxicity against murine BCL1 leukemia in vitro and
markedly increased anti-leukemia efficacy in vivo compared with the non-targeted polymer-Dox
conjugate (without the scFv ligand).
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
Tumor associated carbohydrate antigens represent attractive targets for therapy.4 One such
antigen is disialoganglioside GD2, which poses restricted expression in normal tissue and is
primarily expressed on the cell surface of various tumor tissues including neuroblastoma, most
melanoma lesions, sarcoma, glioma, retinoblastoma, small cell lung cancer, brain tumors,
osteosarcoma, rhabdomyosarcoma, Ewing’s sarcoma in children and adolescents, as well as
liposarcoma, fibrosarcoma, leiomyosarcoma and other soft tissue sarcomas in adults.5–7
Moreover, its presence in breast cancer stem cells has recently been reported.8,9 The total
incidence of GD2-positive tumors in the USA is >200 000 annually. These cancers have a high
mortality rate (20–80%), where 60–100% are possible candidates for anti-GD2 targeted
therapy.10
In view of the GD2 therapeutic potential, this work focused on actively targeted polymer
anticancer drugs based on HPMA copolymers as a polymer carrier (P), Dox as an anticancer
drug and the recombinant scFv fragment of anti-GD2 antibody as a targeting moiety. We utilize
a non-covalent interaction between α-bungarotoxin (BTX) covalently attached to the HPMA-
based copolymer with Dox and bungarotoxin-binding peptide WRYYESSLEPYPD (Bp) linked
to the C-terminus of the recombinant scFv of anti-GD2 antibody. The antigen-specific cell
binding and cytotoxicity of this drug delivery system was investigated and compared with
corresponding covalent polymer-drug-protein conjugates (P-Dox-scFv) prepared via the
3
ACS Paragon Plus Environment

Biomacromolecules
Page 4 of 37
1
2
3
4
5
6
traditional aminolytic reaction between a reactive polymer precursor and the ε-Lys amino groups
of the protein.
7
8
9
It was hypothesized that the non-covalent conjugate P-Dox-BTX / scFv-Bp might exhibit
superior binding activity to GD2 antigen and increased cytotoxicity against malignant GD2-
positive cell lines compared with the covalent conjugate P-Dox-scFv and with the non-targeted
P-Dox conjugate. The approach utilizing formation of the non-covalent complex between the
recombinant scFv-Bp and P-Dox-BTX leads to a structurally better-defined product compared
with the covalent conjugate P-Dox-scFv, which may substantially facilitate approval of targeted
nanotherapeutics for eventual clinical applications by regulatory authorities.
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
2. EXPERIMENTAL SECTION
2.1 Materials
Methacryloyl chloride, 1-aminopropan-2-ol, 3-aminopropanoic acid, 4,5-dihydrothiazole-2-
thiol, dimethylaminopyridine (DMAP), 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide
hydrochloride (EDC), N,N'-dicyclohexylcarbodiimide (DCC), 2,2'-azobis(isobutyronitrile)
(AIBN),
4-cyano-4-thiobenzoylsulfanylpentanoic
acid,
N-(2-aminoethyl)biotinamide
hydrobromide (biotin-NH2), N,N-dimethylacetamide (DMA), dimethyl sulfoxide (DMSO) and
tert-butyl alcohol were purchased from Sigma-Aldrich. 2,2'-Azobis(4-methoxy-2,4-
dimethylvaleronitrile) (V-70) was purchased from Wako Chemicals Europe GmbH, Germany. 2-
Chlorotrityl chloride resin, protected amino acid derivatives and were purchased from Iris
Biotech, GmbH, Germany. 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). Doxorubicin
hydrochloride was purchased from Meiji Seika, Japan and BTX was from Alomone Labs, Israel.
4
ACS Paragon Plus Environment

Page 5 of 37
Biomacromolecules
1
2
3
4
5
6
All other reagents and solvents were purchased from Sigma-Aldrich, Prague, Czech Republic.
All chemicals and solvents were analytical grade.
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
2.2 HPLC Monitoring of Polymer-Analogous Reactions
Monitoring of the conjugation reactions of Dbco-NH2, biotin-NH2 and N-(5-azidopentanoyl-
glycylphenylalanylleucylglycyl)doxorubicin (Az-GFLG-Dox) to the reactive polymer precursors
was performed by HPLC using a 100 × 4.6 mm Chromolith Performance 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).
2.3 Mass spectrometry and Size-Exclusion Chromatography (SEC)
The molecular mass of the monomers and peptide derivatives was determined using mass
spectrometry performed on an LCQ Fleet mass analyzer with electrospray ionization (ESI MS)
(Thermo Fisher Scientific, Inc., MA, USA). The determination of the molecular weights and
polydispersity of the copolymers was performed by SEC on a HPLC system (Shimadzu, Japan)
equipped with UV, differential refractive index, and multi-angle light scattering (LS) DAWN
Heleos II (Wyatt Technology Corp., USA) detectors using a TSK 3000 SWXL column (Tosoh
Bioscience, Japan) (80% methanol, 20% 0.3 M acetate buffer pH 6.5) at a flow rate of 0.5
mL/min. The association behavior of the BTX and polymer-BTX conjugates with scFv-Bp and
conjugation of the proteins to copolymers was studied using a MicroSuperose® 12 column (30
cm × 3.2 mm, 0.01 M phosphate buffer with 0.15 M NaCl, pH 7.4) at a flow rate of 0.1 mL/min.
2.4 UV/VIS Spectrophotometry
5
ACS Paragon Plus Environment

Biomacromolecules
Page 6 of 37
1
2
3
4
5
6
7
8
9
The contents of thiazolidine-2-thione (TT) groups, doxorubicin, dithiobenzoate (DTB) end
groups and aza-dibenzocyclooctyne (Dbco) groups were determined spectrophotometrically on a
Helios Alpha UV/VIS spectrophotometer (Thermospectronic, UK) using the following
absorption coefficients εTT = 10 300 L mol−1 cm−1 (305 nm, methanol), εDox = 8200 L mol−1 cm−1
(488 nm, water), εDTB = 12 100 L mol-1 cm-1 (302 nm, methanol), εDbco = 13 000 L mol−1 cm−1
(292 nm, methanol).
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
2.5 Synthesis of the Monomers
2.5.1 N-(2-hydroxypropyl)methacrylamide (HPMA)
N-(2-Hydroxypropyl)methacrylamide (HPMA) was prepared by the reaction of methacryloyl
chloride with 1-aminopropan-2-ol in dichloromethane.11 Methacrylamidopropanoic acid (Ma-
AP-OH) was prepared by the reaction of methacryloyl chloride with 3-aminopropanoic acid in
an aqueous alkaline medium.12 3-Methacrylamidopropanoylthiazolidine-2-thione (Ma-AP-TT)
was synthesized by the reaction of Ma-AP-OH with 4,5-dihydrothiazole-2-thiol in the presence
of dimethylaminopyridine. The synthesis was performed according to a previously described
method13 using EDC instead of DCC allowing removal of the water-soluble urea derivative by
extraction of the organic solution with water. The monomer was characterized using HPLC
(single peak) and ESI MS (calculated 258.3, found 259.1 M+H).
2.5.2 5-Azidopentanoylglycylphenylalanylleucylglycine (Az-GFLG-OH)
Az-GFLG-OH was prepared via manual solid phase peptide synthesis on 2-chlorotrityl
chloride resin (0.5 g, loading 1 mmol/g) using 0.2 M solutions of AA (5-azidopentanoic acid,
Fmoc-Gly-OH, Fmoc-Phe-OH, Fmoc-Leu-OH). The product was cleaved from the resin with a
6
ACS Paragon Plus Environment

Page 7 of 37
Biomacromolecules
1
2
3
4
5
6
30% solution of HFIP in DCM yielding 121 mg (0.23 mmol, 46%) of the peptide azide. ESI MS
(calculated 531.6, found 554.6 M+Na).
7
8
9
2.5.3 N-(5-azidopentanoylglycylphenylalanylleucylglycyl)doxorubicin (Az-GFLG-Dox)
Az-GFLG-OH (50 mg, 0.094 mmol), 1-hydroxysuccinimide (12 mg, 0.104 mmol) and DCC
(23 mg, 0.113 mmol) was reacted in THF (2 mL) at 4°C overnight. A drop of acetic acid was
added to the mixture, then the precipitated N,N’-dicyclohexylurea was removed by filtration 30
min later and the filtrate was evaporated to dryness. The solid was dissolved in DMA (0.5 mL)
and added to a solution of doxorubicin hydrochloride in DMA (0.5 mL), followed by DIPEA
(0.094 mmol, 16 µL). The progress of the reaction was followed by TLC (silica gel,
chloroform/methanol, 9:1). After 1 h, the reaction mixture was chromatographed on a column
(silica gel, chloroform/methanol, 9:1). The collected fractions with the required product
(according to HPLC) were evaporated to dryness; the residue was dissolved in methanol and
precipitated to diethyl ether to yield 80 mg (0.077 mmol, 82%) of the title compound. ESI MS
(calculated 1042.4, found 1065.7 M+Na).
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
2.5.4 Reactive Polymer Precursor P-TT
Copolymer poly(HPMA-co-Ma-AP-TT) (P-TT) was prepared by reversible addition-
fragmentation chain transfer (RAFT) copolymerization of HPMA (1.5 g, 10.5 mmol, 88 mol%)
and Ma-AP-TT (369 mg, 1.43 mmol, 12 mol%) using V70 (0.03 mmol, 9.17 mg) as an azo
initiator and (1-cyano-1-methyl-ethyl) benzenecarbodithioate (0.06 mmol, 13.2 mg) as a chain
transfer agent (CTA). The molar ratio of monomers/CTA/initiator used was 400:2:1. The
polymerization mixture was dissolved in tert-butyl alcohol with 15% of DMSO (17 mL, 0.9 M
solution of monomers), transferred into a glass ampule, bubbled with Ar and sealed. After 16 h at
40°C, the product was isolated by precipitation with acetone, then washed with diethyl ether and
7
ACS Paragon Plus Environment

Biomacromolecules
Page 8 of 37
1
2
3
4
5
6
7
8
9
dried under vacuum. The copolymer was reacted with AIBN (10 molar excess) in DMSO (15%
w/w solution of polymer) under Ar for 3 h at 70°C in a sealed ampule to remove dithiobenzoate
(DTB) ω-end groups.14 The reaction mixture was isolated by precipitation with acetone, the
precipitate was washed with diethyl ether and dried under vacuum to yield copolymer P-TT (1.1
g, 59%). Molecular parameters of the product were Mw = 27200, Mw/Mn = 1.03. Content of the
reactive TT groups was 11.6 mol%.
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
2.5.5 P-TT-biotin-Dbco
Copolymer P-TT (100 mg, 75 µmol TT) in DMA (0.8 mL) and N-(2-aminoethyl)biotinamide
hydrobromide (2.3 mg, 6.3 µmol) in DMA (0.2 mL) were mixed together and reacted for 3 h at
25°C. Then Dbco-NH2 (8.8 mg, 32 µmol) was dissolved in the reaction mixture and DIPEA (5
µL, 32 µmol) was added. The reaction was completed in 40 min (as indicated by HPLC) and the
polymer product was twice precipitated into acetone/diethyl ether (1:1) to yield 102 mg of the
title copolymer P-TT-biotin-Dbco. The biotin content was 1.0 mol% as determined by HABA
assay.15
2.5.6 P-Dox
A solution of Az-GFLG-Dox (4.2 mg, 4 µmol) in DMA (50 µL) was added to P-TT-biotin-
Dbco (18 mg, 5.7 µmol Dbco) dissolved in DMA (150 µL). After 2 h at 25 °C, the crude
polymer was precipitated into acetone/diethyl ether (1:1), re-dissolved in PBS (2 mL) containing
1-aminopropan-2-ol (1 µL), purified on Sephadex G-25 in water and lyophilized to yield 16.4
mg of the title copolymer. The polymer conjugate contained 10.6% w/w of Dox.
2.5.7 P-Dbco-BTX
BTX (1 mg, 0.125 µmol) was dissolved in 40 µL of 0.01 M phosphate buffer with 0.15 M
NaCl pH 7.4 (PBS) and mixed with equal volume of PBS solution of copolymer P-TT-biotin-
8
ACS Paragon Plus Environment

Page 9 of 37
Biomacromolecules
1
2
3
4
5
6
7
8
9
Dbco (2 mg, 0.64 µmol TT). The aminolytic reaction was terminated after 16 h by addition of 1-
aminopropan-2-ol (0.08 µL in 16 µL of PBS). The title polymer conjugate was purified via
Sephadex G-25 in PBS and concentrated to final volume of 150 µL using an Amicon Ultra-0.5
Centrifugal Filter Unit with Ultracel-10 membrane (cut off 10 kDa).
2.5.8 P-BTX-Dox
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
Az-GFLG-Dox (0.4 mg, 0.38 µmol) in DMSO (6 µL) was added to the solution of polymer
conjugate P-Dbco-BTX (3 mg in 150 µL of PBS) under stirring. A small fraction of undissolved
Dox derivative was removed by centrifugation 16 h later. The polymer conjugate contained 7.7%
w/w of Dox.
2.5.9 P-Dbco-scFv
A solution of scFv (preparation is described below) in PBS (9 mg in 2.5 mL) was mixed with a
PBS solution of polymer P-TT-biotin-Dbco (18 mg, 0.5 mL). The reaction mixture was kept for
16 h at 25°C and the solution was used for subsequent reactions.
2.5.10 P-scFv-Dox
Az-GFLG-Dox (3.5 mg, 3.35 µmol) in DMSO (50 µL) was added to the solution of polymer
conjugate P-Dbco-scFv (27 mg in 3 mL of PBS) under stirring. A small fraction of undissolved
Dox derivative was removed by centrifugation 16 h later. The polymer conjugate contained 9.1%
w/w of Dox.
2.5.11 Non-Covalent Complex P-BTX-Dox / scFv-Bp
The polymer conjugate P-BTX-Dox (1.8 mg in 0.3 mL PBS) and equimolar amount (molar
ratio BTX/Bp = 1/1) of recombinant protein scFv-Bp (1.2 mg in 0.3 mL) were mixed at 25°C to
yield the non-covalent polymer-protein complex within 10 min. The resulting solution was used
for flow cytometry and cytotoxicity studies.
9
ACS Paragon Plus Environment

Biomacromolecules
Page 10 of 37
1
2
3
4
2.5.12 Non-covalent complex P-Dbco-BTX / scFv-Bp
5
6
The complex was prepared by the same procedure as described above.
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
2.6 Cloning, Expression and Purification of scFv Molecules
DNA encoding variable domains of heavy and light chains of anti-GD2 mAb hu3F8 were
chemically synthesized according to a published sequence16 and assembled into the scFv
fragment as described previously.3,17 The scFv anti-GD2 molecule, in the format (VH)-
(Gly4Ser)4-(VL), contains 118 N-terminal residues of the heavy chain linked to 105 N-terminal
residues of the light chain, followed by the C-myc tag sequence EQKLISEEDL and His5 tag. A
DNA fragment coding for 13 amino acid residues long BTX binding peptide
WRYYESSLEPYPD (Bp), which displays strong non-covalent interaction with α-bungarotoxin
(BTX), was introduced between the myc and his tags in the scFv (scFv-Bp). As a control we
used recombinant scFv of B1 mAb, which recognizes the idiotype of surface IgM expressed on
mouse BCL1 leukemia cells. The appropriate DNA fragment coding the scFv B1 was used to
replace anti-GD2 sequence in the above anti-GD2 scFv molecule. This mock scFv-Bp-B1 was
expressed in a bacterial system as described earlier.3
For a covalent mode of attachment to the polymer, the anti-GD2 scFv fragment was simply
bound to polymer by aminolytic reaction.2
For expression, the anti-GD2 scFv-Bp coding sequence, fused to the antibody heavy chain
leader sequence to allow secretion of the product into the medium, was cloned into the vector
pcDNA3.1. Transfection and cultivation using the ExpiCHO system (Thermofisher) was
performed according to the high titre manufacturer’s manual, with minor modifications.
10
ACS Paragon Plus Environment

Page 11 of 37
Biomacromolecules
1
2
3
4
5
6
7
8
9
Purification of the recombinant product was performed on a HisTRAP 5 mL column (GE
Healthcare) with the following buffers: buffer A (50 mM NaH2PO4 .2H2O; 300 mM NaCl; pH
8.4) and buffer B (50 mM NaH2PO4 .2H2O; 300 mM NaCl; 500 mM imidazole; pH 8.4).
Conditioned media with scFv-Bp was dialyzed 3 times against 3 L of buffer A. The column was
equilibrated by 10 column volumes of buffer A and washed with 10 column volumes of buffer A
and 10 column volumes of buffer A + 1% buffer B. For the elution, buffer A + 10% buffer B,
buffer A + 50% buffer B and 100% buffer B were used. All steps were performed at the rate of 1
ml/min at 4°C (except loading speed 0.5 ml/min). After elution, the fractions were analyzed by
SDS-PAGE and western blot. The concentration of eluted protein was determined by A280
measurement using an extinction coefficient obtained from ExPASy, tool ProtParam.
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
2.7 Isothermal Titration Calorimetry (ITC)
ITC measurements were performed on a MicroCal-200 isothermal titration calorimeter. The
experiment was conducted with consecutive injections of BTX or polymer-BTX conjugate
solution into the calorimeter cell. The cell contained 280 μL of the binding peptide or scFv-Bp
protein solution. The BTX solution was added with a 40-μL injection syringe with a tip modified
to act as a stirrer. The chosen stirring speed was 750 rpm. The injection volume was set for the
first injection to 0.5 μL, and 2 μL for the rest of the 19 injections. The time between injections
was varied from 300 to 600 seconds to reach a baseline. The measurements were conducted at
37°C. The data were analyzed using the Microcal ORIGIN software.
2.8 Dynamic Light Scattering (DLS)
11
ACS Paragon Plus Environment

Biomacromolecules
Page 12 of 37
1
2
3
4
5
6
7
8
9
DLS was performed to characterize the individual solutions of polymer conjugate and protein,
and their mixture after ITC. For this purpose, the hydrodynamic diameter of the particles, Dh,
and the scattering intensity were measured at a scattering angle of θ = 173° with a Zetasizer
Nano-ZS instrument, model ZEN3600 (Malvern Instruments, UK). The DTS (Nano) program
was used to evaluate the data. The volume weighted value of the apparent Dh was chosen to
characterize the dispersity of the solutions. Solutions were filtered prior to DLS experiments
using 0.22 µm PVDF filter (Millipore).
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
2.9 Cells and Media
UKF-NB-3 and EL-4 cells were obtained from ATCC. The NIH-3T3 cell line was a kind gift
from Dr Jan Závada (IMG). The cell line UKF-NB-3 was cultured in IMDM media (Sigma-
Aldrich), while EL-4 and NIH-3T3 in DMEM media (Sigma-Aldrich), supplemented with 10%
of heat inactivated FCS (Gibco, South America origin) and antibiotics Anti-Anti (Gibco). These
cell lines were incubated under standard conditions of 5% CO2 at 37°C. Maternal ExpiCHO cell
culture was maintained in ExpiCHO expression medium at 37°C, 8% CO2, 150 rpm. All cell
lines tested negatively for the presence of mycoplasma.
2.10 Flow Cytometry
Cells (4 × 105) in 20 µL PBS with 1% BSA (Sigma-Aldrich) were incubated with targeted
conjugate, non-targeted conjugates and controls at similar concentrations of targeting scFv either
polymer-bound or in control. Non-targeted conjugates were diluted to the same concentration
which was used in the targeted conjugates. Solutions of the polymer conjugates contained 16–
130 µg mL-1 of Dox equivalent. The commercially obtained secondary antibody anti-His/FITC
12
ACS Paragon Plus Environment

Page 13 of 37
Biomacromolecules
1
2
3
4
5
6
7
8
9
(Exbio) for detection of the recombinant scFv (containing His-tag) was diluted 100 times, while
streptavidin/FITC (Sigma-Aldrich) for detection of the biotinylated polymer was diluted 200
times. Incubation of samples with the cells in each step was for 30 min at 4°C. Between
incubations of the cells with conjugates and staining molecules, the samples were washed twice
by ice cold 1% BSA in PBS. Stained samples were analyzed by flow-cytometer (LSRFortessa,
BD), with the gating strategy FSC-A, SSC-A; FSC-A, FSC-H; 561-610/20, SSC-A; 488-530/30,
count. Obtained data were evaluated with FlowJo software, version 10.0 (Tree Star, Inc.).
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
2.11 In Vitro Cell Cytotoxicity Assay
EL-4 cells were collected from an exponential growth phase culture. Cells were centrifuged for
5 min at 300 g, washed, counted in Trypan blue assay and re-suspended in fresh culture medium.
The cell density was adjusted to 2 × 105 cells/ml, and 0.05 mL of this cell suspension was loaded
by a repeating pipette into each well of Nunc 96-well round-bottom plates. The tested samples
dissolved in growth medium were then added to the wells (50 µL/well) to achieve the required
concentrations. UKF-NB-3 and NIH-3T3 cell lines were assayed similarly, with the exception
that the loading cell density was 1 × 105 cells/mL and the cells were pipetted into wells 20 hours
prior to adding the assayed samples. Just prior to adding the tested samples, the conditioned
medium was aspirated from wells and samples dissolved in fresh growth medium in the desired
concentrations were added (100 µL/well). The final well volume was 0.1 mL and cell density
was 1 × 104 cells/well. Each condition was measured in triplicate. The final concentration of Dox
equivalent in tested samples for EL-4 cells were in the range 0.01–144 µg/mL. The plates were
cultured in 5% CO2 for 24 and 72 h at 37°C, then 100 µL of XTT cell assay solution (Sigma-
Aldrich) was added to each well, and the cell viability was measured by spectrometer at several
13
ACS Paragon Plus Environment

Biomacromolecules
Page 14 of 37
1
2
3
4
5
6
7
8
9
time points (0, 1, 2, 4 and 8 hours). The optimal time point was chosen based on the best
difference between the strongest and the weakest signal of absorbance. Shown numbers of
absorbance were obtained as difference data measured at two different wavelengths (490 and 655
nm) on the Infinite M200 (Tecan) and collected by software Magellan 6.0 (Tecan). Evaluation
and IC50 determination was performed using Excel (Microsoft).
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
3. RESULTS AND DISCUSSION
3.1 Synthesis of the Polymer Conjugates
The multivalent polymer precursor P-TT with approximately 12 mol% of reactive TT groups
used for preparation of the targeted polymer conjugates was synthesized via a RAFT
copolymerization of HPMA and Ma-AP-TT providing a copolymer with very low dispersity
(Mw/Mn = 1.03). One portion of the TT groups (about 1 mol%) was modified with biotin-NH2 to
enable monitoring of the polymer using flow cytometry. Another portion of the TT groups (5
mol%) reacted with Dbco-NH2 to yield polymer P-TT-biotin-Dbco (Scheme 1).
14
ACS Paragon Plus Environment

Page 15 of 37
Biomacromolecules
1
2
3
4
1. RAFT copolymerization
t-BuOH, 12 h, 40°C
1. biotin-NH2
2. Dbco-NH2
5
6
2. ABIN, DMA, 2 h, 80°C
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
HPMA
Ma-AP-TT
P-TT
P-TT-biotin-Dbco
Scheme 1
The remaining TT groups (5–6 mol%) of the polymer were used for reaction with primary
amino groups of α-bungarotoxin or scFv yielding the corresponding polymer-protein conjugates.
In the last step, the azide derivative of Dox, Az-GFLG-Dox, containing an enzymatically
cleavable tetrapeptide spacer was bound to the polymer-protein conjugates via a metal-free
strain-promoted azide-alkyne cycloaddition (Scheme 2).
15
ACS Paragon Plus Environment

Biomacromolecules
Page 16 of 37
1
2
3
4
5
6
1. BTX or scFv
7
2. Az-GFLG-Dox
8
PBS, pH 7.4
BTX or scFv
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
P-BTX-Dox
or
P-scFv-Dox
P-TT-biotin-Dbco
GFLG
Scheme 2
The initial attempts to bind Az-GFLG-Dox to P-TT-biotin-Dbco and then conjugate the protein
in PBS led to precipitation of the hydrophobic polymer in the aqueous solution providing
practically zero yield of the polymer-protein conjugate. Addition of a large amount of organic
solvent, such as DMSO or DMA, which was necessary for the dissolution of the precipitate
resulted in total loss of biological activity (including antigen binding) of the protein.
Generally, polymers prepared via RAFT polymerization with low dispersity exhibit more
favorable pharmacokinetic behavior,18 consequently, the corresponding nanotherapeutics also
have a higher therapeutic efficacy.19 Furthermore, such polymers are more likely to have better
clinical potential than more polydisperse polymers. Consequently, the polymer precursors
prepared via RAFT polymerization were selected for the present study.
3.2 Recombinant Proteins and scFv-Targeted Polymer Conjugates
Non-covalent attachment of targeting protein molecules to polymer conjugates has many
advantages.20 In our previous studies,3 we used coiled-coil interaction for non-covalent
16
ACS Paragon Plus Environment

Page 17 of 37
Biomacromolecules
1
2
3
4
5
6
7
8
9
attachment of scFv moieties to polymers. In the present study, an alternative approach was used,
namely a much stronger BTX – Bp interaction.21 This BTX-Bp pair was used for the first time in
an arrangement where a recombinant PSMA-specific scFv fragment C-terminally fused to BTX
was used for targeting the polymer-Bp modified adenovirus to prostate-specific membrane
antigen.22 When we attempted to produce an analogous recombinant antiGD2 scFv-BTX fusion
protein, the expression yields were very low, irrespective of the expression system (bacterial or
mammalian cells), so antiGD2 scFv-Bp and polymer modified with BTX were prepared as
described above in the methods. As shown below, the polymer conjugate assembly realized
through the non-covalent interaction BTX-Bp proved very efficient in specific targeting GD2-
positive cells and was superior to the covalent conjugate.
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 quality of every new batch of scFv was checked by SDS electrophoresis where no
evidence of impurities was observed.
3.3 Association Behavior of BTX and Polymer-BTX Conjugates with Bp and scFv-Bp
Previously, we demonstrated that unmodified BTX and a recombinant protein consisting of
scFv and BTX is able to bind strongly to unmodified Bp and to HPMA-based copolymers
containing multiple Bp chains covalently bound to the polymer backbone.22 However, prior to
the biological evaluation of our BTX-based drug delivery system, it was necessary to verify if
chemical modification of BTX with copolymers containing reactive TT groups does not destroy
the binding activity of the resulting polymer-BTX conjugate to Bp and/or scFv-Bp. At the same
time, it was important to check if the Bp incorporated in the recombinant protein scFv-Bp did not
lose its binding activity with BTX and polymer-BTX conjugates by performing SEC, ITC and
DLS experiments.
17
ACS Paragon Plus Environment

Biomacromolecules
Page 18 of 37
1
2
3
4
3.3.1 Size-Exclusion Chromatography
5
6
7
8
The association behavior of the polymer conjugates containing BTX (P-Dbco-BTX and P-
BTX-Dox) with recombinant protein scFv-Bp was studied by SEC using MicroSuperose 12
column in aqueous PBS. First, we confirmed formation of the complex between unmodified
BTX and recombinant protein scFv-Bp. The original peaks corresponding to BTX (tR = 20 min)
and scFv-Bp (tR = 17.5 min) disappeared upon mixing of equimolar amounts of the two
components and a new peak corresponding to a higher molecular weight adduct was observed
(Figure 1). The surprisingly short retention time of the adduct (tR = 12 min) indicates the
formation of stable high-molecular weight associates with quite narrow molecular weight
distribution, however, with molecular weight exceeding 106. Nevertheless, this result clearly
demonstrated that the Bp sequence incorporated in the structure of recombinant protein scFv-Bp
retained its binding affinity to BTX. The molecular characteristics of the copolymers and non-
covalent complexes are summarized in Table 1.
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
Table 1. Basic characteristics of the copolymers
a
a
Copolymer
Mw
Mw/Mn
1.03
1.10
1.14
3.15
1.46
1.54
TT mol% b
Dox wt% c
P-TTd
27 200
56 000
87 000
145 000
400 000
500 000
11.6
5.5
0
0
P-TT-biotin-Dbcod
P-Dbco-BTX
0
0
P-Dbco-BTX/scFv-Bp
P-BTX-Dox
0
0
0
7.7
4.6
P-BTX-Dox/scFv-Bp
0
18
ACS Paragon Plus Environment

Page 19 of 37
Biomacromolecules
1
2
3
4
5
6
7
8
9
P-Dbco-scFv
P-scFv-Dox
P-Dox
207 000
1.95
0
0
>1 000 000 >3.0
9.1
79 000
1.08
10.6
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
BTX/scFv-Bp
>1 000 000 1.45
a Molecular weight determined by SEC in PBS using RI and LS detection.
b TT determined by UV/VIS spectrophotometry in methanol (ε305 = 10 300 L·mol−1·cm−1).
c Dox determined by UV/VIS spectrophotometry in methanol (ε488 = 8 200 L·mol−1·cm−1).
d SEC measurement in 80 % methanol 20 % 0.3 M sodium acetate, pH 6.5.
(x10,000)
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0.0
5.0
10.0
15.0
20.0
25.0
30.0
min
Figure 1. SEC chromatograms of BTX (blue), scFv-Bp (red) and equimolar mixture of scFv-Bp
+ BTX (black). MicroSuperose 12 column, eluent PBS, UV detector 280 nm, flow rate 0.1
mL/min and sample concentration of 3 mg/mL was used.
After the injection of the polymer-BTX conjugate P-Dbco-BTX (tR = 16.5 min) and its
equimolar mixture (molar ratio BTX/Bp = 1/1) with scFv-Bp, the formation of the non-covalent
complex between the polymer and the protein was proved by a new broad peak with a maximum
19
ACS Paragon Plus Environment

Biomacromolecules
Page 20 of 37
1
2
3
4
5
6
7
8
9
around 15 min (Figure 2). The relatively wide molecular weight distribution (Mw/Mn = 3.15) is
probably caused by statistical distribution of the various BTX molecules per individual polymer
chain resulting in a variable number of scFv-Bp molecules per polymer. The formation of the
polymer-protein complex clearly demonstrated that the binding affinity of BTX in the polymer
conjugate P-Dbco-BTX was preserved.
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
(x10,000)
10.0
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
0.0
5.0
10.0
15.0
20.0
25.0
30.0
min
Figure 2. SEC chromatograms of scFv-Bp (red), P-Dbco-BTX (green) and equimolar mixture of
the two components (blue). MicroSuperose 12 column, eluent PBS, UV detector 280 nm, flow
rate 0.1 mL/min and sample concentration of 3 mg/mL was used.
Finally, an analogous experiment with polymer-BTX conjugate P-BTX-Dox containing the
anticancer drug doxorubicin was performed. In this case, the addition of an equimolar amount of
scFv-Bp to the polymer resulted in the disappearance of the protein peak; however, only a
negligible shift of the newly formed peak compared with the peak of P-BTX-Dox was observed.
A shoulder on the peak corresponding to unbound scFv-Bp appeared only after addition of 4-fold
excess of the protein to the polymer (Figure 3).
This may be explained with the dramatic decrease of hydrophobicity of the original polymer-
Dox conjugate upon complexation with the hydrophilic scFv protein. Most probably, the P-BTX-
20
ACS Paragon Plus Environment

Page 21 of 37
Biomacromolecules
1
2
3
4
5
6
7
8
9
Dox formed micellar structures in the aqueous media, which disassembled after the
complexation with the scFv protein. The apparent weight-average molecular weight Mw =
400 000 of the original P-BTX-Dox increased to Mw = 500 000 after the addition of scFv-Bp,
also indicating formation of the complex. The large difference between the measured Mw of the
polymer precursor P-TT and the corresponding polymer conjugate P-BTX-Dox can be attributed
to self-aggregation of the hydrophobic polymer conjugate in aqueous medium. While the SEC
measurements of the polymer conjugates in methanol as a mobile phase provide molecular
parameters of individual macromolecules, SEC in aqueous solution represents a more realistic
result, reflecting their behavior in a physiologically relevant medium.
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
5.0(x100,000)
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
5.0
7.5
10.0
12.5
15.0
17.5
20.0
22.5
min
Figure 3. SEC chromatograms of scFv-Bp (green), P-BTX-Dox (blue), P-BTX-Dox + scFv-Bp
(red, equimolar mixture) and P-BTX-Dox + 4×scFv-Bp (black, 4-fold excess of scFv-Bp).
MicroSuperose 12 column, eluent PBS, UV detector 280 nm, flow rate 0.1 mL/min and sample
concentration of 3 mg/mL was used.
In comparison with the non-covalent polymer-scFv complexes, covalent conjugate P-scFv-Dox
has a higher molecular weight and dispersity resulting from the reaction of the multiple amino
groups of the protein with the multiple reactive TT groups of the polymer. Consequently, the
21
ACS Paragon Plus Environment

Biomacromolecules
Page 22 of 37
1
2
3
4
5
6
non-covalent polymer-scFv complex containing Dox seems to be a good candidate for further
investigation.
7
8
9
3.3.2 Dynamic Light Scattering
The conclusions from the previous section were further supported by DLS experiments
conducted for individual components and their mixture (Figure 4).
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. Volume weighted distribution function over hydrodynamic diameter Dh of scFv-Bp
(green), P-BTX-Dox (blue), P-BTX-Dox + scFv-Bp (red, 2-fold excess of P-BTX-Dox).
The distribution function of the scFv-Bp protein manifests in the presence of a dominant peak
corresponding to 5 nm that could be attributed to molecular dissolved proteins. The minor peak
is related to a small fraction of protein aggregates that are also visible on the SEC chromatogram
22
ACS Paragon Plus Environment

Page 23 of 37
Biomacromolecules
1
2
3
4
5
6
7
8
9
(Figure 3). In contrast, the inspection of the distribution function for the polymer conjugate P-
BTX-Dox reveals a single mode with a hydrodynamic diameter Dh of 20 nm. Keeping in mind
that the molecular weight of the polymer precursor P-TT measured in 80% methanol (preventing
aggregation) is 27 000 g/mol, this mode could be assigned to the aggregates of polymer
conjugate, since the estimated hydrodynamic diameter Dh for a single chain of such molecular
weight should in the range of 5–7 nm. Thus, the DLS results corroborate the SEC data,
suggesting self-aggregation of the hydrophobic polymer conjugate in PBS. Complex formation
between P-BTX-Dox and scFv-Bp protein is clearly visible on the distribution curve (Figure 4,
red line); a bimodal distribution with a left peak corresponding to the particles with
hydrodynamic diameter of 40 nm. This is indicative of complexation and subsequent association
of the protein and the polymer conjugate. No peak related to individual protein or conjugate is
manifested on the distribution function. The second peak might be attributed to a very small
amount of large aggregates.
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
23
ACS Paragon Plus Environment

Biomacromolecules
Page 24 of 37
1
2
3
4
5
6
7
3.3.3 Isothermal Titration Calorimetry
8
9
To quantify the strength of binding of BTX or its polymer conjugate to the binding peptide Bp
or scFv-Bp protein, ITC experiments were performed. ITC provided complete thermodynamic
parameters of the binding constant (Kb), stoichiometry (N), enthalpy change of binding (∆H),
and entropy change of binding (∆S) from the thermograms produced after integration of each
titration followed by subtraction of the dilution heat. Gibbs free energies of each reaction were
then calculated using the enthalpy and entropy results. Thus, ITC offered a direct tool for
characterizing the thermodynamic properties of a binding process. The binding of BTX or its
polymer conjugate to the binding peptide or scFv-Bp protein was studied by ITC in phosphate
buffer solution at 37°C. The integrated heats of BTX or its polymer conjugate per mole of
injectant is shown in Figure 5.
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
0.0
0.0
-5.0x103
-2.0x103
-1.0x104
-1.5x104
-2.0x104
-4.0x103
-6.0x103
P-biotin-BTX-Dox to BP
P-biotin-BTX-Dox to ScFv-BP
unmodified BTX to BP
unmodified BTX to scFv-Bp
0
1
2
0
2
4
6
8
molar ratio
24
ACS Paragon Plus Environment

Page 25 of 37
Biomacromolecules
1
2
3
4
5
6
7
8
9
Figure 5. Calorimetric titration curves obtained for titration of 10 µM of BTX or its BTX-
polymer conjugate solution into 1µM binding peptide or scFv-Bp protein in PBS solution at
37°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
The results proved the complexation of BTX or its BTX-polymer conjugate with the binding
peptide Bp or scFv-Bp protein. It is exothermic process and the thermodynamic parameters
calculated using one-set-of-sites model are listed in Table 2. On the basis of the thermodynamic
data, it is concluded that the complexation of BTX or BTX-polymer conjugate with Bp or scFv-
Bp is enthalpy driven, not entropy driven. Indeed, both values of ∆H and T∆S are negative
making the -T∆S term in the Gibbs equation positive. The values of stoichiometry for the
complexation of BTX with binding peptide and scFv-Bp protein are 0.66 and 0.78, and the
binding constant (Kb) of 7.3 and 12.2×106 mol-1, respectively. The stoichiometry value lower
than 1 is often due to inherent aggregation of proteins, which are easily detectable as shown in
Figure 1, 2, and 3 for the SEC data. A general thermodynamic profile for the binding of pure
BTX to binding peptide or scFv-Bp protein is relatively similar. The overall discrepancy in ∆H,
T∆S, N, and Kb values could be explained by small differences in the binding site between the
binding peptide Bp or protein scFv-Bp. In contrast, the covalent binding of BTX to the pHPMA
main chain has a significant influence on a binding event. Although the ∆H values for BTX and
P-BTX-Dox conjugate are only two-fold different for the same type of peptide, indicating that
complexation of P-BTX-Dox with protein is energetically a less favorable process, the binding
constant for P-BTX-Dox conjugate is one order of magnitude lower in comparison with pure
BTX (Table 2). The possible explanation for such discrepancy is decreased accessibility of BTX
moiety due to the steric hindrance caused by the polymer. Another peculiar observation for the
25
ACS Paragon Plus Environment

Biomacromolecules
Page 26 of 37
1
2
3
4
5
6
7
8
binding of P-BTX-Dox conjugate is the shift of stoichiometry to the values above 2, as
evidenced by the SEC and DLS data where large aggregates of P-BTX-Dox conjugate were
observed in solution.
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
Table 2. Thermodynamic parameters for the complexation of BTX or polymer-BTX conjugate
with binding peptide Bp or scFv protein in PBS at 37°C
Ligand
Substrate
∆H,
T∆S,
N, sites
Kb, 106 mol-1
kcal/mol
kcal/mol
BTX
Bp
-19.1±0.4
-16.4±0.5
-11.7±0.6
-4.3±0.5
-9.3
-6.4
-3.1
-3.7
0.66±0.01
0.78±0.02
2.1±0.2
7.3±1.6
12.2±0.5
0.5±0.3
1.4±0.4
scFv-Bp
P-BTX-Dox Bp
scFv-Bp
2.1±0.1
Taken together, the ITC experiments explicitly provide evidence of complexation of BTX to
binding peptide or scFv-Bp protein regardless of its existing in solution as a single moiety or
bound to a pHPMA chain. The binding was observed in both cases, although these binding
processes are different in some features.
3.4 Cell Binding Studies of GD2-Targeted Polymer-Drug Conjugates
The aim of this study was to design, synthesize and evaluate polymer-Dox conjugates targeted
against GD2-positive tumor cells, murine T-cell lymphoma EL423 and human neuroblastoma
UKF-NB3.5 Murine fibroblast cells NIH 3T3 were used as a negative control. Two different anti-
GD2-targeted polymer-Dox systems were compared: a) covalent conjugate P-scFv-Dox prepared
via a repeatedly described24,25 and often used aminolytic reaction between the reactive TT groups
26
ACS Paragon Plus Environment

Page 27 of 37
Biomacromolecules
1
2
3
4
5
6
7
8
9
of the polymer with the primary lysine residues of scFv of anti-GD2 monoclonal antibody, and
b) non-covalent complex P-BTX-Dox / scFv-Bp between polymer-bungarotoxin-Dox conjugate
and a recombinant protein consisting of scFv and bungarotoxin-binding peptide. It was
hypothesized that the latter system P-BTX-Dox / scFv-Bp would exhibit superior binding
activity and, consequently, also cytotoxicity against GD2-positive cell lines due to its better-
defined structure compared with the covalent conjugate.
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
In the flow cytometry experiments, anti-His/FITC was used for detection of the recombinant
scFv protein (containing His tag) on the cells, streptavidin/FITC detected presence of the
biotinylated polymer on the cells. If the both probes provided a strong positive signal a
conclusion could be made that both components of the polymer-scFv complex were bound to the
cells.
Although the exact number and positions of the bonds between the polymer and the protein in
the covalent conjugate P-scFv-Dox are not known and the resulting polymer-protein conjugate is
probably a complex mixture of macromolecules, the product still exhibited specific binding to
both GD2-positive cell lines as demonstrated by flow cytometry (Figure 6). No binding to GD2-
negative cell line NIH 3T3 was observed (Figure S1A, Supporting Information).
27
ACS Paragon Plus Environment

Biomacromolecules
Page 28 of 37
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 6. Flow cytometry analysis of (A) EL-4 and (B) UKF-NB3 cell binding by the targeted
covalent conjugate P-scFv-Dox. Along the x-axis, the histograms show fluorescence intensity of
the cells alone (red), cells incubated with scFv and anti-His-FITC (blue), cells incubated with P-
scFv-Dox and anti-His-FITC (orange) and cells incubated with P-scFv-Dox and streptavidin-
FITC (green).
Figure 7. Flow cytometry analysis of (A) EL-4 and (B) UKF-NB3 cell binding by the targeted
non-covalent complex P-BTX-Dox / scFv-Bp. Along the x-axis, the histograms show
28
ACS Paragon Plus Environment

Page 29 of 37
Biomacromolecules
1
2
3
4
5
6
7
8
9
fluorescence intensity of the cells alone (red), cells incubated with scFv and anti-His-FITC
(blue), cells incubated with P-BTX-Dox / scFv-Bp and anti-His-FITC (orange) and cells
incubated with P-BTX-Dox / scFv-Bp and streptavidin-FITC (green).
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 non-covalent complex P-BTX-Dox / scFv-Bp exhibits somewhat better binding activity
against the GD2 positive cell lines compared with the covalent conjugate P-BTX-Dox, as
indicated by the fluorescence intensity of the cells incubated with the targeted polymer
conjugates and visualized with streptavidin-FITC (Figure 7, green histograms). Surprisingly, the
cells treated with P-BTX-Dox / scFv-Bp and detected with anti-His-FITC (orange histograms)
exhibited even higher fluorescence intensity than the cells incubated with scFv alone (blue
histograms). This may be attributed to the presence of multiple scFv ligands attached to one P-
BTX-Dox conjugate (containing in average about 4 BTX binding sites) as well as the less
destructive character of the non-covalent scFv binding to the polymer compared to the covalent
aminolytic approach.
Interestingly, the enhanced targeting ability, which is directly connected with the increased
cytotoxicity, was also observable from the cell counts during the cell binding experiment. The
number of the living cells was substantially decreased after the incubation of the GD2-positive
cells with the scFv-targeted samples. When the cell binding experiment was performed with a
higher concentration (130 µg mL-1 Dox equivalent) of the targeted non-covalent complex, it led
to a significant cytotoxic effect resulting in a very low number of living cells (using both GD2-
positive cell lines) compared with the untreated control (Figure S2, Table S1, Supporting
Information). No such toxicity was observed against the GD-2 negative control cell line NIH
3T3 (Figure S1B, Table S1, Supporting Information).
29
ACS Paragon Plus Environment