Polyvalent Diazonium Polymers Provide Efficient Protection of Oncolytic Adenovirus Enadenotucirev from Neutralizing Antibodies while Maintaining Biological Activity in Vitro and in Vivo

Article abstract of DOI:10.1021/acs.bioconjchem.9b00189

Oncolytic viruses offer many advantages for cancer therapy when administered directly to confined solid tumors. However, the systemic delivery of these viruses is problematic because of the host immune response, undesired interactions with blood components, and inherent targeting to the liver. Efficacy of systemically administered viruses has been improved by masking viral surface proteins with polymeric materials resulting in modulation of viral pharmacokinetic profile and accumulation in tumors in vivo. Here we describe a new class of polyvalent reactive polymer based on poly(N-(2-hydroxypropyl)methacrylamide) (polyHPMA) with diazonium reactive groups and their application in the modification of the chimeric group B oncolytic virus enadenotucirev (EnAd). A series of six copolymers with different chain lengths and density of reactive groups was synthesized and used to coat EnAd. Polymer coating was found to be extremely efficient with concentrations as low as 1 mg/mL resulting in complete (>99%) ablation of neutralizing antibody binding. Coating efficiency was found to be dependent on both chain length and reactive group density. Coated viruses were found to have reduced transfection activity both in vitro and in vivo, with greater protection against neutralizing antibodies resulting in lower transgene production. However, in the presence of neutralizing antibodies, some in vivo transgene expression was maintained for coated virus compared to the uncoated control. The decrease in transgene expression was found not to be solely due to lower cellular uptake but due to reduced unpackaging of the virus within the cells and reduced replication, indicating that the polymer coating does not cause permanent inactivation of the virus. These data suggest that virus activity may be modulated by the appropriate design of coating polymers while retaining protection against neutralizing antibodies.

Full text of DOI:10.1021/acs.bioconjchem.9b00189

Subscriber access provided by UNIV OF DURHAM
Article
Polyvalent diazonium polymers provide efficient protection
of oncolytic adenovirus Enadenotucirev from neutralising
antibodies while maintaining biological activity in vitro and in vivo
Nora Francini, Daniel Cochrane, Sam Illingworth, Laura Purdie, Giuseppe Mantovani,
Kerry Fisher, Leonard W. Seymour, Sebastian Guy Spain, and Cameron Alexander
Bioconjugate Chem., Just Accepted Manuscript • DOI: 10.1021/acs.bioconjchem.9b00189 • Publication Date (Web): 15 Mar 2019
Downloaded from http://pubs.acs.org on March 16, 2019
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 40
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
Polyvalent Diazonium Polymers Provide Efficient Protection
of Oncolytic Adenovirus Enadenotucirev from Neutralising
Antibodies while Maintaining Biological Activity In Vitro and
In Vivo
Nora Francinia,‡, Daniel Cochraneb, Sam Illingworthb, Laura Purdiea, Giuseppe
Mantovania, Kerry Fisherb,c, Leonard W. Seymourc*, Sebastian G. Spaind* and
Cameron Alexandera*
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
a.
cameron.alexander@nottingham.ac.uk; Tel: +44 (0)115 846 7678.
b. PsiOxus Therapeutics Limited, 4-10, The Quadrant, Abingdon Science Park, Abingdon,
Oxfordshire, OX14 3YS, UK.
School of Pharmacy, University of Nottingham, NG7 2RD, UK. E-mail:
c.
Department of Oncology, Old Road Campus Research Building, Roosevelt Drive, Oxford,
OX3 7DQ, UK. Email: len.seymour@oncology.ox.ac.uk.
d.
Department of Chemistry, University of Sheffield, Sheffield, S3 7HF. E-mail
s.g.spain@sheffield.ac.uk; Tel: +44 (0)114 222 9362; Twitter: @sebspain.
‡ Present address. IIT Central Research Laboratory, Istituto Italiano di Tecnologia, 16163 Genova,
Italy.
Page 1 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 2 of 40
1
2
3
4
Abstract.
5
6
7
8
9
Oncolytic viruses offer many advantages for cancer therapy when administered directly to
confined solid tumors. However, the systemic delivery of these viruses is problematic due to
host immune response, undesired interactions with blood components and inherent targeting to
the liver. Efficacy of systemically administered viruses has been improved by masking viral
surface proteins with polymeric materials resulting in modulation of viral pharmacokinetic
profile and accumulation in tumors in vivo. Here we describe a new class of polyvalent reactive
polymer based upon poly(N-(2-hydroxypropyl)methacrylamide) (polyHPMA) with diazonium
reactive groups and their application in the modification of the chimeric group B oncolytic
virus enadenotucirev (EnAd). A series of six copolymers with different chain lengths and
density of reactive groups was synthesised and used to coat EnAd. Polymer coating was found
to be extremely efficient with concentrations as low as 1 mg/mL resulting in complete (>99%)
ablation of neutralising antibody binding. Coating efficiency was found to be dependent on
both chain length and reactive group density. Coated viruses were found to have reduced
transfection activity both in vitro and in vivo, with greater protection against neutralising
antibodies resulting in lower transgene production. However, in the presence of neutralising
antibodies some in vivo transgene expression was maintained for coated virus compared to the
uncoated control. The decrease in transgene expression was found not to be solely due to lower
cellular uptake but due to reduced unpackaging of the virus within the cells and reduced
replication indicating that polymer coating does not cause permanent inactivation of the virus.
These data suggest that virus activity may be modulated by appropriate design of coating
polymers while retaining protection against neutralising antibodies.
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
Page 2 of 40
ACS Paragon Plus Environment

Page 3 of 40
Bioconjugate Chemistry
1
2
3
4
Introduction
5
6
7
8
9
Virus-based therapies for cancer were first proposed over 100 years ago but the first clinical
trials involving administration of attenuated viruses were of mixed success.(1),(2) As our
understanding of virology has improved conditionally-replicating viruses that can act as
anticancer agents (oncolytic viruses)(3),(4),(5) have been developed. These have led to the first
product, Oncorine, an engineered oncolytic adenovirus approved for the treatment of head and
neck cancer, being released in China in 2005.(6),(7) More recently, Imlygic, an oncolytic
herpes virus, has been approved in US and EU, for the treatment of unresectable melanoma
lesions.(8)
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 widespread use of oncolytic viruses for disseminated cancers via systemic (i.e.
intravenous) administration remains limited by virus immunogenicity, and the pharmacokinetic
and biodistribution profiles of injected nanoparticles.(9-11) Accordingly, there have been
efforts to target viruses to cancer sites, in a manner analogous to those adopted for targeting
synthetic nanoparticles.(12-15) The conjugation of synthetic polymers to provide ‘stealth’
coatings for oncolytic viruses is one means by which extended circulation and tumour targeting
might be achieved. A number of papers have reported successful conjugation of polymers to
synthetic viruses,(16-18) with concomitant changes in viral biodistribution and targeting.(19-
22) PEGylation of adenovirus (Ad) was reported in the pioneering work of O’Riordan et al.(23)
and Croyle et al.(24) Adenovirus (Ad) vectors were modified with PEG5000 and were shown to
retain infectivity and to transduce cells in vitro and in vivo in the presence of neutralizing
antibodies. Although this was the first example of successfully chemically modified Ad, the
study was limited to intranasal delivery and a limited degree of PEGylation, due to aggregation
and precipitation of more highly functionalized viruses. Other polymers have been used to coat
viruses,(25)
of
which
the
most
widely
used
has
been
poly(N-(2-
hydroxypropyl)methacrylamide) (polyHPMA), covalently bound to the amine groups on the
Page 3 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 4 of 40
1
2
3
4
5
6
7
8
9
viral surface. polyHPMA with modified side-chains is a polyvalent polymer, thus the coating
of viral surfaces is more dense with this polymer and lower concentrations are required when
compared to PEG.(26) polyHPMA for Ad shielding was first developed as a random copolymer
with pendant amino-reactive 4-nitrophenolate ester linked to polyHPMA by a gly-gly dipeptide
spacer.(27),(28) In these studies, coating resulted in ablation of Coxsackie-Adenovirus
Receptor (CAR) binding and conferred partial resistance to neutralizing antibodies. Extended
plasma circulation was observed for polyHPMA coated virus with higher doses achieving
longer half-life, to the extent that a 10-fold higher concentration of polyHPMA-Ad was found
in plasma 30 minutes after injection compared to naked Ad virus. A marked decrease in virus
hepatotoxicity of Ad upon coating with HPMA polymers, was reported, and additional in vivo
studies showed accumulation of the conjugate in tumor masses after injection and a 40-fold
increase in transgene expression compared to unmodified Ad.(29) Coating with this copolymer
prevented cellular uptake but transfection was restored and boosted when retargeting agents
such as fibroblast growth factor 2 (FGF2),(30) epidermal growth factor (EGF)(31) or cell-
penetrating peptides(32) were covalently attached. The ability to retarget Ad was evaluated in
vitro and in vivo where a highly efficient and ligand-dependent transduction was observed.
However, while a degree of protection against neutralizing antibodies(27),(33) and improved
pharmacokinetic and biodistribution profiles compared to unmodified-Ad viruses(34) have
been obtained, complete neutralization of antibody responses by polymer conjugation has still
to be accomplished.
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
Accordingly, we set out to develop new methods for polymer conjugation to viruses in order
to increase the reaction at the virus surface and therefore enhance the protection against
neutralising antibodies. Conjugation strategies of polymers to surface viral proteins usually
involves reaction with the ε-amino groups of lysine residues. The natural abundance of this
amino acid on viral surfaces, which for adenovirus has been reported to number over 1800
Page 4 of 40
ACS Paragon Plus Environment

Page 5 of 40
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
copies(35) make it an important target. However, though the use of this strategy has shown
relatively high coverage of surface proteins(22-24, 27) as demonstrated by 50 to 80% lower
anti-adenovirus antibody recognition compared to naked virus, it has failed to produce the
complete viral antigen masking that is required for the application of viral therapeutics in vivo.
Herein, we report the synthesis of new polymers with broadened reactivity towards a variety
of amino acid residues – potentially overcoming the limitation suffered from specific lysine
targeting – and the effects of these polymers on viral vectors in vitro and in vivo.
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
Results and Discussion
Synthesis of functional poly(N-(2-hydroxypropyl)methacrylamide)
derivatives
The intial aims were to i) utilise a reactive functional group with broad reactivity and ii)
determine the effects of polymer chain length and reactive group density on the efficiency of
polymer coating of an adenoviral vector. EnAd (and reporter transgene derivatives(36)) was
selected for these studies since intravenous delivery to tumours has been demonstrated with
this virus,(37) so modifications that improved delivery efficiency or avoidance of neutralizing
antibodies could have direct therapeutic relevance. The synthetic approach is outlined in Fig.
1. PolyHPMA was synthesised via reversible addition-fragmentation chain transfer (RAFT)
polymerization which allowed polymers of well-defined chain length to be synthesised. In
previous work, we demonstrated that the pendant secondary hydroxyl group of polyHPMA
could be modified via a one-step procedure with allyl isocyanate to yield allyl-functional
reactive polyHPMA derivatives.(38) Here, aniline groups were introduced at different densities
along the polyHPMA backbone by reaction with Boc-protected aminobenzoic acid followed
by deprotection of amino groups. The aniline groups could then be activated to form highly
reactive diazonium groups that would react with the virus surface. The low selectivity of
Page 5 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 6 of 40
1
2
3
4
5
6
7
8
9
diazonium coupling for specific amino acids at neutral pH was identified as a possible
advantage that may allow reaction with a wider range of sites on the virus surface leading to
improved coverage.
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. Synthetic outline for the production of polymer-coated viruses from polyvalent diazonium salts.
Reagents: i. N-Boc-4-aminobenzoic acid, EDC, DMAP, DMSO; ii. TBAF; iii. CF3CO2H, NaNO2, sulfamic acid.
A series of polyHPMA homopolymers was synthesized as described by Scales et al.(39) The
chain transfer agent (CTA) to initiator ratios ([CTA]:[I]) were kept constant at 3:1, while the
HPMA monomer:CTA ratios were varied, (100, 250, 350) to obtain polymers with different
chain lengths, P1–P3 (Table 1). The end-groups introduced during RAFT polymerization are
a potential source of toxicity.(40) Consequently they were removed as described by Perrier et
al.(41) End-group removal was confirmed by UV-Vis spectrophotometry (Figure S1) and
Ellman’s assay (Table 1), performed before and after treatment.
Table 1. Poly[N-(2-hydroxypropyl)methacrylamide] precursors
Mn, (kDa)
Theora
SECb
17.0
30.3
45.9
Ðb
%Thiolc
Sample
DPn
d
P1
P2
P3
85
165
240
12.2
23.7
34.3
1.11
1.14
1.11
—
—
d
0.32
a
b
Calculated considering the initial [HPMA]/[CTA] ratio and final conversion;
Determined by SEC using DMF + 0.1 % LiBr as the mobile phase calibrated with PMMA
standards; c Residual thiol after CTA removal using AIBN, as determined by Ellman’s
assay.d Below detection limit.
Pendent aniline groups were attached to the polymer backbone via the coupling reaction of N-
t-butoxycarbonyl-4-aminobenzoic acid with the hydroxyl groups of polyHPMA, using EDC
Page 6 of 40
ACS Paragon Plus Environment

Page 7 of 40
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
and DMAP, as reported by Jones et al. for the esterification of hydroxy-terminated PEG.(42)
The reactivities of diazonium salts have been demonstrated to increase when electron-
withdrawing groups are introduced in the para position of the aromatic ring, therefore, in
analogy to data reported by Jones et al.,(42) 4-aminobenzoic acid (PABA) was selected here
to generate a polymeric diazonium precursor (hereafter referred to as polyvalent diazonium).
Standard Boc protection of commercially available 4-aminobenzoic acid was carried out to
afford the pure product (PABA-Boc, Figure S2). The Steglich esterification of alcohols
mediated by carbodiimides is often carried out in the presence of catalytic amounts of DMAP.
However, under these conditions, a large excess of DMAP was found to be needed to promote
the reaction. The optimized reaction was monitored by 1H NMR spectroscopy for 48 h. when
no further conversion of the alcohol into the corresponding benzoic esters was observed.
Coupling of PABA-Boc to the PHPMA precursors was tailored to target degrees of
functionalization of 0.15 and 0.30 aniline groups per initial OH group. Higher modifications
(0.5) resulted in unusable water-insoluble polymers after deprotection while lower
modifications (0.10 or single terminal modification) failed to undergo later diazo coupling
when bioconjugation was tested with a model protein (data not shown).
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
Benzoic acid content in the final products was quantified by H NMR spectroscopy by
comparing the reference integral values of the propyl CHO, whose summed value was assigned
to 1 proton, and the integrals of the aromatic protons or the NH proton of the Boc protecting
group. To validate the results, alternatively, the signal of the Boc NH was used as a reference
and compared to values of polyHPMA signals. Polyvalent diazonium precursors were treated
with 8 equivalents of tetrabutylammonium fluoride (Bu4NF) to aniline groups at 80 °C for 16
h in DMSO in order to remove the protecting group. The reaction yielded the desired aniline-
modified polymers with only limited hydrolysis of the ester groups. 1H NMR spectroscopy was
used to assess the removal of Boc groups and to determine the final aniline content (Figure 2).
Page 7 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 8 of 40
1
2
3
4
5
6
Six polyvalent diazonium precursors were generated using this method with 3 different chain
lengths and 2 degrees of functionalization (Table 2).
7
8
9
As these polymers bear multiple activated aromatic rings in close proximity to each other, there
is significant potential inter- and intra-molecular side reactions that would effectively limit the
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
number of diazo groups available for coupling to the desired target. Consequently, the ability
of polydiazonium polymers to react completely to give the required diazo-adducts was tested
1
Figure 2. H NMR spectrum of deprotected P9 in DMSO-d6. All polymers in the
P4–P9 library gave analogous spectra, where only the relative intensity of the
PABA-Boc and HPMA repeating units was found to change.
and monitored by 1H NMR spectroscopy. The shift in aromatic signals of the aromatic groups
in the 1H NMR spectra were used to monitor conversion at each step of the synthesis (Figure
3).
Table 2. Degree of functionalization of PHPMA precursors after post-polymerization
modification with N-Boc-4-aminobenzoic acid and subsequent deprotection.
Page 8 of 40
ACS Paragon Plus Environment

Page 9 of 40
Bioconjugate Chemistry
1
2
3
4
Composition
PABAtarg (%)a
PABAcalc (%)b,c
PABAdep (%)c,d
5
6
7
8
P4
P5
P6
P7
P8
P9
HPMA85-r-PABA9
HPMA85-r-PABA20
HPMA165-r-PABA18
HPMA165-r-PABA40
HPMA240-r-PABA22
HPMA240-r-PABA58
15
30
15
30
15
30
11
24
12
25
12
26
10
24
11
25
9
9
24
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
a targeted based on the initial HPMA:PABA ratio; b after esterification; c calculated using
1H NMR spectroscopy in DMSO-d6; d after deprotection with Bu4NF.
1
Figure 3. Stacked aromatic region of H NMR spectra of the
various stages of diazonium salt formation: aniline precursor
(black, top), protonated aniline (green, middle) and diazonium
salt (blue, bottom).
Diazonium salts of P7 were generated by sequential addition of TFA and NaNO2 to an aqueous
solution of the polymer at 0 °C to form the nitrous acid required to convert the PABA repeating
units into their corresponding diazonium salts in situ. Excess nitrous acid was quenched and
with sulfamic acid and the pH adjusted to 7.4 for subsequent protein conjugation. Control of
the reaction temperature was essential for successful diazonium formation with small increases
due to addition of mild/warm solutions resulting in decomposition.
Polyvalent diazonium polymers provide efficient shielding of EnAd
against neutralizing antibodies in vitro
Efficient shielding of the viral surface is essential to prevent recognition, and subsequent
destruction, by the immune system. Consequently, the ability of polyvalent diazonium
polymers to effectively shield EnAd from neutralizing antibodies was evaluated. The series of
Page 9 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 10 of 40
1
2
3
4
5
6
7
8
9
polymers synthesized allowed three variables in the coating to be assessed: the density of
diazonium moieties on the polymer chain (i.e. 10 and 24 mol %); the copolymer degree of
polymerization (i.e. 85, 165 and 240 units) and the concentration of the copolymers in the
conjugation reaction (0.5 to 5.0 mg/mL). The EnAd concentration was kept constant
throughout.
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. EnAd shielding from neutralizing antibodies as a
means of determining polymer coating efficiency determined
via ELISA. Values are normalized to uncoated virus control.
The bioconjugation of polyvalent diazonium polymers to was first tested using BSA as a model
protein (Figure S3). For virus bionconjugation, stock solutions of P4 to P9 were activated in
situ and added to the required virus aliquot on ice. Three independent experiments were
performed with triplicate samples of each formulation. The efficiency of coating was estimated
using an enzyme-linked immunosorbent assay (ELISA) with antibodies for EnAd. Briefly,
anti-Ad antibody (polyclonal rabbit) was immobilised on an ELISA plate to act as a capture
antibody then solutions of unmodified and polymer-coated EnAd were added. The amount of
captured EnAd was determined by addition of anti-Ad antibody (polyclonal mouse) which was
detected using an anti-mouse HRP secondary antibody with 3,3’,5,5’-tetramethylbenzidine
(TMB) as the substrate. The absorbance relative to unmodified EnAd was then used to estimate
Page 10 of 40
ACS Paragon Plus Environment

Page 11 of 40
Bioconjugate Chemistry
1
2
3
4
5
6
shielding efficiency i.e. a lower absorbance indicates less bound EnAd and therefore more
efficient coating . Represntative results for the six polymers are shown in Figure 4.
7
8
9
At a polymer concentration of 5 mg/mL, the formation of a yellow precipitate was observed
after coating with the polymers carrying the higher diazonium content (~24 mol%) indicating
the formation of undesired cross-linked viral particle aggregates. Conversely, the lower
diazonium content polymers (~10 mol%) did not produce sufficient viral shielding, with the
intermediate chain length (DP = 165) achieving around 60% protection against neutralising
antibodies. Unexpectedly, the absorbance of EnAd-P8 exceeded the value observed for the
uncoated viral control which would indicate an increase in antibody binding. This was
attributed to the combined effect of inefficient polymer coating, resulting in significant
antibody binding, and the absorbance in the same spectral region as the ELISA substrate (450
nm)(43) of any azobenzene linkages present.
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
As cross-linked aggregates are potentially too large for systemic delivery, polymer
concentrations of 1, 0.75 and 0.5 mg/mL were investigated to reduce the possibility of polymer-
mediated crosslinking. The optimal coating concentration was found to be 1 mg/mL at which
all polymers provided viral shielding without the formation of precipitates or aggregates, as
confirmed by dynamic light scattering (DLS, table 3). Lower polymer concentrations resulted
in insufficient and highly variable coating efficiency (data not shown). At 1 mg/mL polymer
concentration, the coating efficiency of polymers with low aniline content (10 mol%) increased
with polymer Mn, indicating that larger polymers do improve shielding, with viral shielding as
high as 90%. High coating efficiency – defined as reaching 99% viral shielding from
neutralising antibodies – was obtained with polymers with higher aniline content (24 mol%),
independent of the degree of polymerization. It is worth noting, that EnAd-P4, EnAd-P6 and
EnAd-P8 (10 mol % diazonium) showed higher variability in the outcome of the coating within
Page 11 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 12 of 40
1
2
3
4
5
6
the triplicate experiments, indicating that a higher diazonium content might be required to
increase the reproducibility of the coating.
7
8
9
Table 3. Hydrodynamic diameter of polymer-coated virus at 1mg/mL coating concentration.
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
diameter (nm)
99 ± 1.7
PDI
EnAd
0.1 ± 0.0
0.1 ± 0.0
0.1 ± 0.0
0.1 ± 0.0
0.2 ± 0.0
0.1 ± 0.1
0.1 ± 0.0
EnAd-P4
EnAd-P5
EnAd-P6
EnAd-P7
EnAd-P8
EnAd-P9
100 ± 1.6
107 ± 1.5
110 ± 2.9
133 ± 1.0
108 ± 7.2
153 ± 6.9
Dynamic light scattering (DLS) analysis of the polymer-coated virus formulations, indicated
an increase in the hydrodynamic diameter upon coating that can be related to the polymer chain
length and capsid shielding (table 3). EnAd-P4, EnAd-P6 and EnAd-P8, which displayed less
antibody shielding by ELISA, did not present significant increases in their sizes compared to
unmodified virus. EnAd-P5, EnAd-P7 and EnAd-P9, displayed an increase in the diameter
that was polymer chain length-dependent, in agreement with the polymer shielding observed
by ELISA.
Previous reports of polymer-coating viruses with polyHPMA have relied upon active esters
(e.g. nitrophenolates) as the reactive moiety with surface amines (lysines) as the target on the
viral capsid. Typically, active ester chemistries have required high polymer concentrations (20
mg/mL) to achieve antibody shielding with 50–70% reduction in binding as estimated by
ELISA.(33),(34),(31) Consequently, achieving high levels of shielding (up to 99%) at low polymer
concentrations (1 mg/mL) demonstrates the benefit of using diazonium groups as the reactive
species.
In vitro and in vivo luciferase expression of polymer-coated EnAd
Although polymer coating can prevent antibody binding, it can also affect the biological
activity of the virus by preventing interactions with cell surfaces (and thus viral uptake) and
Page 12 of 40
ACS Paragon Plus Environment

Page 13 of 40
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
viral unpacking (and thus gene expression and virus replication). To determine whether EnAd
activity was affected by coating, a luciferase-expressing EnAd (EnAd(luc)) was coated as
before and luciferase expression determined in HT-29 (epithelial colorectal adenocarinoma)
cells. EnAd has previously been shown to display high oncolytic potency against this cell line
with a 2-fold increase compared to Ad5(44) and it was therefore selected as an optimal model
to assess its biological activity. Prior to these studies, cell metabolic activity was determined
using an MTS assay 24 h. after exposure to the polymers alone (Figure S4). As no significant
toxicity was found for these polymers at the concentration used and there was relatively little
free polymer in a model reaction (Figure S3d), all the experiments were performed with virus-
polymer conjugates without extensive purification from unreacted polymer.
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
Cells were treated with polymer-coated EnAd(luc) and an uncoated control and the luciferase
expression was measured 24 h. post infection (Figure 5a). Transduction efficiency was
markedly reduced for coated EnAd(luc) with up to 4-orders of magnitude reduction in
luciferase expression compared to the uncoated control. There is no clear correlation between
the shielding from neutralizing antibodies and the transduction efficiency, with virus coated
with P8, which displayed high shielding efficiency at 1 mg/mL coating concentration, having
one of the highest luciferase levels. Polymers with the higher level of diazonium modification,
which showed more consistent and efficient coating, all reduced luciferase expression to the
same level.
As in vitro results do not necessarily translate in vivo, EnAd(luc)-P4 to P9 were screened on
naïve Balb/C mice to test gene transduction in vivo in the presence or absence of neutralizing
serum (Figure 5b–d). Groups of three mice were injected intramuscularly with 1 × 109
vp/mouse. Luciferase expression was imaged using an IVIS imaging system and quantification
performed at the site of injection, (Figure 5d, red circle).
Page 13 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 14 of 40
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 5. Transduction efficiency in vitro (a) and in vivo (b–d) of EnAd(luc) at 1 mg/mL polymer coating. a)
Luciferase expression of HT-29 cells infected with 500 vp/cell of EnAd(luc) or polymer-coated EnAd(luc),
measured 24 hours post infection. b, c and d) In vivo luciferase expression after intramuscular injection in Balb/C
mice of 1 × 109 virus samples in PBS (b) and in neutralizing murine serum (c, d), measured at 24 hours post-
injection using an IVIS imaging system. (Means of triplicate values and standard deviation, one-way ANOVA
statistical analysis; graph a: p < 0.0001 vs EnAd(luc), ns vs uninfected cells (nil))(n=3),graph b; p=0.0255, graph
c: p=0.0392.
When polymer-coated viruses were administered using phosphate-buffered saline (PBS) as the
vehicle (Figure 5b), the luciferase expression was reduced in comparison to the uncoated virus
in a similar manner to that seen in vitro albeit to a lesser extent. When neutralizing serum was
used as a vehicle (Figure 5c and d), luciferase expression of the uncoated virus was reduced
approximately 1000-fold compared to administration in PBS, demonstrating the high
susceptibility of EnAd to neutralisation. Under these conditions, all formulations except
EnAd(luc)-P9 showed a reduction in transfection compared to administration in PBS.
EnAd(luc)-P9 maintained a significant level of luciferase expression, while expression levels
similar to that of the uncoated control were observed for formulations EnAd(luc)-P5 to
EnAd(luc)-P7. EnAd(luc)-P4 and EnAd(luc)-P8, which showed poor shielding in the ELISA,
maintained more expression than the uncoated control but not to a significant level. These
results indicate that polymeric features play an important role in imparting protection against
Page 14 of 40
ACS Paragon Plus Environment

Page 15 of 40
Bioconjugate Chemistry
1
2
3
4
neutralising antibodies in vivo, with both longer polymer chain length and higher analine
5
6
content being required to ensure sufficient protection.
7
8
9
Polymer coating delays EnAd cellular uptake and transgene expression
EnAd(luc)-P9 was selected for further investigations in order to obtain a better understanding
of the intracellular mechanisms involved in the loss of viral activity after polymer coating. To
allow separation of the effects of polymer coating on cell uptake and later downstream
processes such as virus unpacking and trafficking, a “retargeted” virus was also produced.
Poly(lysine) sequences of 6–10 residues are known to interact with cellular membranes via
electrostatic interactions of the positively charged pendant amino groups and the negatively
charged phospholipids and increase cell uptake.(45) Therefore, P9 was functionalized with
1 mol% of oligolysine consisting of 7 amino acids (K7) (Figure S5) which was then used to
coat EnAd.
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
Uptake of EnAd(luc), EnAd(luc)-P9 and retargeted virus, EnAd(luc)-P9-K7, viruses was
quantified by qPCR. Cells were washed three times with PBS adjusted to pH 3 in order to
detach any membrane-associated virus that could contribute to false positive results. The
different uptake profiles of the three formulations are shown in Figure 6 where the coated virus,
EnAd(luc)-P9, demonstrated a reduction in uptake compared to both EnAd(luc) and
EnAd(luc)-P9-K7.
Re-targeting not only restored uptake compared to EnAd(luc)-P9 but was able to markedly
improve it, most likely due to an increased interaction with the cell membrane that facilitated
uptake through endocytic pathways other than receptor-mediated.(45) However, these results
indicate that, although cellular uptake was involved in loss of transduction observed, this may
not have been the only factor involved as the level of uptake was still high.
Page 15 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 16 of 40
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. Viral uptake 1.5 hours post-infection. Viral genome copies were quantified by
qPCR after three acid washes (PBS pH 3) to detach any membrane associated virus from
the cells. Samples were prepared and analyzed in triplicate. Statistical analysis
performed by one-way ANOVA, **** p < 0.0001.
Later time points were investigated to determine if the polymer coating was causing a delay in
the unpackaging, a reduction of viral infectivity or a permanent inhibition due to hampered
intracellular trafficking. HT-29 cells were infected for 1.5 h., washed with acidic PBS as
described for the uptake studies, and incubated for 24, 48 and 72 h. Cells were then lysed and
viral genome copies (qPCR) and transgene expression (luciferase assay) determined (Figure 7a
and b). EnAd(luc) was able to replicate within the one day post infection, which corresponded
to a high transgene expression (103 increase). Conversely, no significant replication was
determined up to 3 days post-infection with EnAd(luc)-P9 and EnAd(luc)-P9-K7. However,
during the same time-course a steady increase of transgene expression became evident, with a
faster expression mediated by EnAd(luc)-P9-K7. Luciferase expression is controlled by the
immediate-early CMV promoter and requires partially disassembled virus escaping the
endosome and reaching the cytoplasm.(46) These results indicate that the polymer-coating
induces endosomal acculmation, or a delay in the unpackaging of the viral capsid, rather than
a permanent inhibition. Accumulation in localised organelles at an early stage of infection was
further confirmed by confocal fluorescence microscopy imaging (Figure S6).
Page 16 of 40
ACS Paragon Plus Environment

Page 17 of 40
Bioconjugate Chemistry
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 7. Viral disassembly and replication. Cells were infected with 500 vp/cell for 1.5 hours, washed three times
with PBS pH 3 and, subsequently incubated at 37 °C, 5% CO2. Cells were lysed with 1x reporter lysis buffer at
1.5, 24, 48 and 72 h. post-infection. 25 µL of lysate was used to quantify viral genome copies by qPCR (a) and
25 µL to quantify transgene expression by luciferase reporter assay (b). C) “Specific transduction efficiency”
calculated as RLU/genome copies ratio.
The “specific transduction efficiency”, defined as the ratio of transgene expression per genome
copies of internalised virus, seems to corroborate this hypothesis (Figure 7c). For EnAd, this
increased in time until 72 h, when a drop was observed due to cell death. EnAd(luc)-P9 showed
a small increase in transduction efficiency in the same time-frame. Up to 48 h.,
EnAd(luc)-P9-K7 transduction efficiency was comparable to that of unmodified EnAd(Luc).
However, a spike in luciferase activity was observed at 72 hours which may have arisen through
different mechanisms such as the increased uptake of EnAd-P9-K7 and the possibility of K7
aiding endosomal escape through membrane disruption by the charged alkyl amines.(47)
To investigate the effect of polymer-coating on replication, cell monolayers were infected with
EnAd engineered to express green fluorescent protein (GFP) after replication has started
(endogenous promoter SA, EnAd(SA)). EnAd(SA) was coated as before. The ability of
EnAd(SA) to produce GFP positive cells allowed for visual observation of replication by
fluorescence microscopy. Cells were infected with EnAd(SA), EnAd(SA)-P9 and EnAd(SA)-
P9-K7 and imaged for up to 5 days post infection (Figure 8a).
For EnAd(SA), GFP expression was detectable at 1 day post infection and achieved the highest
levels at day 2; limited fluorescence was detected at day 3 due to significant cell death.
Page 17 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 18 of 40
1
2
3
4
5
6
7
8
9
EnAd(SA)-P9 and EnAd(SA)-P9-K7 did not induce any significant fluorescence until 5 days
post infection. Quantification of green fluorescence in the samples was performed by flow
cytometry (Figure 8b). For EnAd, 60% of cells were GFP positive after two days. As observed
by fluorescent microscopy, both the polymer-coated formulations, EnAd(SA)-P9 and
EnAd(SA)-P9-K7, showed a slow onset of fluorescence which markedly increased at 5 days
post infection at which point 80% of cells treated EnAd(SA)-P9-K7 were GFP positive. This
enchancement of transgene expression is attributed to enhanced cell uptake provided by the K7
targeting.
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
Page 18 of 40
ACS Paragon Plus Environment

Page 19 of 40
Bioconjugate Chemistry
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 8. Green fluorescence micrographs (a) and percentage of fluorescent cellls (b) for HT-29 cells
infected with EnAd(SA) GFP. Cells were infected with 100 vp/cell of EnAd(SA)-GFP, EnAd(SA)-
P9, EnAd(SA)-P9-K7. Images were recorded every 24 h. for 5 days using a widefield inverted
microscope equipped with a green fluorescence filter. After imaging, cells were fixed for
quantification using flow cytometry. Cells were detached from the well and fixed in 4 % PFA for 10
minutes at 4 C. After centrifugation, the pellet was resuspended in PBS and the percentage of GFP
positive cells were quantified by flow cytometry.
Page 19 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 20 of 40
1
2
3
4
5
6
7
8
9
These data together confirmed that the ability of EnAd(SA)-P9 to replicate is not permanently
hampered. The delayed onset of expression is in agreement with the delay in unpackaging, and
suggestive that only a small number of viruses participated in the infection at this stage.
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
Consideration of the biological mechanisms underlying viral activity provide some
explanations for the transfection and replication data observed. In EnAd(luc) is regulated under
a cytomegalovirus (CMV) promoter which, when inserted in exogenous genetic materials, is
constitutively active and allows luciferase to be expressed even when replication does not
occur. In naïve mice, EnAd(luc) is replication-incompetent and luciferase expression thus
reflects only the ability of the virus to enter cells and unpackage intracellularly. It is also
important to recognise that the receptor used by EnAd, and in general by adenoviruses of
species B, CD46, is not expressed in rodent cells.(48) For this reason, rodents are not ideal in
vivo models when employed for testing with adenoviruses that enter cells through interactions
with this receptor. However, in this study, the use of a rodent in vivo model was designed to
serve as a ‘background assay’ to detect any uptake which might occur through the plethora of
endocytic pathways. In particular, adenoviruses have been shown to induce cellular activation
of macropinocytosis as a means to enter host cells, as reported by Amstutz et al. and Kalin et
al. (49),(50),(51) for macropinocytic uptake of Ad3 and Ad35 in epithelial cells, respectively.
These viruses also use CD46 as a receptor for epithelial cellular uptake, and thus can be
considered to be analogous to EnAd in their cell internalisation properties. Both transgene
expression and virus internalization assays in this study supported the hypothesis that different
mechanisms, other than CD46 clathrin-dependent endocytosis, are involved in the
internalization of EnAd. In addition, luciferase expression in vivo following polymer-coated
EnAd administration was partly restored compared to that observed in the in vitro experiments.
Interestingly, the effect of the different polymers resembled the trend observed for the in vitro
screening and ELISA previously described. These data suggested that ligand-receptor
Page 20 of 40
ACS Paragon Plus Environment

Page 21 of 40
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
interactions play a pivotal role in ablating the activity of polymer-coated viruses in vitro but
are not essential when the latter are administered in vivo. Polymers with higher diazonium
content (EnAd-P5, EnAd-P7 and EnAd-P9) and EnAd-P6 could only partially restore
transduction in vivo, with EnAd-P9 recovering higher levels of luciferase expression compared
to the other formulations. In a similar fashion, polydiazoniums of higher Mn (EnAd-P8 and
EnAd-P9) imparted enhanced protection against neutralizing mouse serum as compared to
uncoated EnAd (Figure 5c and d). It is important to notice that although similar capsid-
shielding and in vitro activity profiles were observed for EnAd-P5, EnAd-P7 and EnAd-P9
formulations, lower luciferase expression was induced by EnAd-P5 and EnAd-P7. Taken
together, these data allowed an insight into a structure-activity relationship of the investigated
polymers which suggest that for the same content of reactive side-chain functionality, an
increase in polymer size is favourable for both shielding and de-targeting purposes. It could be
hypothesized that polymer coating interferes at the intracellular level, altering or limiting its
natural trafficking. Successful infection by a virus relies on the delivery of viral genetic
material to the host cell nuclei. A polymer that is too tightly bound to the viral surface could
disrupt the tertiary structure of capsid proteins or sterically prevent capsid disassembly or
endosomal escape, thus reducing the overall progress of the infection. When considering
EnAd-P5 and EnAd-P9 (chain lengths 85 and 240, respectively) it might be expected that
EnAd-P9 would produce lower transgene expression in vivo compared to EnAd-P5 due to the
larger volume of the polymer. The result obtained could be attributed to the fact that a shorter
polymer chain (polyHPMA P1, Mn ~ 12 kDa) can facilitate diazonium coupling by an increased
exposure of the active groups compared to a more coiled and sterically hindered conformation
of a longer polymer chain (pHPMA P3, Mn ~ 35 kDa). This would result in fewer azobenzene
bonds being formed on EnAd-P9 and a looser polymer coating, which is in agreement with the
DLS results, through which prevention of antibody recognition would be mediated by steric
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
Page 21 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 22 of 40
1
2
3
4
5
6
7
8
9
hindrance of this flexible and coiled structure as well as chemical blocking recognition sites.
Consequently, EnAd-P9 might allow increased viral disassembly compared to EnAd-P5 or
EnAd7 as this looser coating will be more flexible. In contrast, if the polymer chains were
tightly bound to the virus surface, capsid disassembly would be hindered. Since the DLS data
demonstrated a marked increase in hydrodynamic volume of virus formulations as the polymer
Mn increased, it is therefore likely that the longer polymers (EnAd-P6 and EnAd-P9) bound
in a more open structure to the virus surface. In turn, this higher entropy structure enhanced the
steric shielding against large antibodies but did not completely prevent viral replication
processes which arose through disassembly of smaller components at the viral capsid.
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
Conclusions
In this study, a novel set of functional polyHPMA-based materials bearing multiple diazonium
salts was readily generated by simple and well-established coupling techniques via post-
polymerization modifications. The broadened reactivity spectrum of diazonium salts towards
various amino acids enabled optimized shielding of EnAd adenoviral capsid against
neutralising antibody at polymer concentrations of 1 mg/mL, 10–20 fold lower than that
reported in literature, as demonstrated in vitro and in vivo. Polymer chain length were found to
be a key factor in both virus shielding and modulation of activity. Importantly, irreversible
polymer conjugation with larger polymers did not result in permanent inactivation, as
highlighted by time-course transgene expression and viral genome quantification, but likely a
delayed unpackaging caused by impaired cellular trafficking. These data suggest that by further
tailoring the polymer macromolecular features, it may be possible to tune key pharmaceutical
properties of polymer-virus conjugates, such as circulation time, resistance to neutralising
antibodies, internalisation and intracellular trafficking at target cells and to preserve viral
activity.
Page 22 of 40
ACS Paragon Plus Environment

Page 23 of 40
Bioconjugate Chemistry
1
2
3
4
5
6
Materials and Methods
Materials
7
8
9
1-amino-2-propanol (>99%), methacryloyl chloride (97%), (4-cyanopentanoic acid)-4-
dithiobenzoate (CPADB), 4,4'-azobis(4-cyanovaleric acid) (V-501, >90%),
azobis(isobutyronitrile) (AIBN, 98%), Fmoc-lysine(Boc), EDC (1-ethyl-3-(3-
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
dimethylaminopropyl) carbodiimide), DMAP (4-dimethylamino pyridine), trifluoroacetic acid
(TFA, 99%), sulfamic acid and triethylamine (TEA, >98%), Tris- Ethylenediaminetetraacetic
acid (Tris-EDTA) were purchased from Sigma-Aldrich (Poole, UK). Sodium nitrite was
purchased from Fluka. Di-tert-butyl-dicarbonate (97+%) were purchased from Alfa Aesar.
Piperidine (>99%), anhydrous (<50 ppm water) N,N-dimethylacetamide (DMAc) and
anhydrous (<50 ppm water) dimethylsulfoxide (DMSO) were purchased from Acros Organics
(Geel,
Belguim).
2-(7-Aza-1H-benzotriazole-1-yl)-1,1,3,3-tetra-methyluronium
hexafluorophosphate (HATU, 98%) was purchased from Fluorochem Ltd (Hadfield, UK). H-
NovaSyn®TGT resin was purchased from Novabiochem. High glucose Dulbecco’s modified
Eagle’s Medium (DMEM) with L-glutamine, Dulbecco’s phosphate buffer saline (DPBS),
Hank’s balanced salt solution (HBSS), fetal bovine serum (FBS) were purchased from Fisher
scientific UK as well as all other chemicals and solvents (analytical or HPLC grade). AIBN
was recrystallized from MeOH and methacryloyl chloride was distilled under Ar flow before
use, all other chemicals were used as received unless otherwise stated. EnAd adenoviruses were
supplied by PsiOxus Therapeutics: EnAd, EnAd(luc) (expressing firefly luciferase under CMV
promoter), EnAd(SA) GFP (expressing green fluorescent protein under control of splicing
acceptor).
Analytical Methods
Nuclear magnetic resonance (NMR) spectra were recorded on a Bruker Avance III 400 MHz
spectrometer. All chemical shifts (δ) are reported in ppm relative to tetramethylsilane or
Page 23 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 24 of 40
1
2
3
4
5
6
7
8
9
referenced to the chemical shifts of residual solvent resonances. Multiplicities are described
with the following abbreviations: s = singlet, br = broad, d = doublet, t = triplet, m = multiplet.
Mass spectrometry was carried out using a Micromass LCT ToF with electrospray ionization
and OpenLynx software. Samples were prepared in 1:1 MeCN/water containing 0.1% v/v
formic acid. Size exclusion chromatography (SEC) was performed on a Polymer Laboratories
GPC 50 system equipped with a refractive index detector. Separations were achieved with a
pair of PLgel Mixed-D (5 μm bead, 7.8 × 300 mm) columns with a matching guard (7.8 × 50
mm) and DMF containing 0.1% LiBr as eluent at a flow rate of 1 mL min-1. Calibration was
performed using narrow PMMA standards (Polymer Labs) in the molecular weight range 1-
1200 kDa. Molecular weights and dispersity values were calculated using Cirrus GPC 3.0.
Aqueous SEC was performed on Shimadzu Prominence UPLC system fitted with a DGU-20A5
degasser, LC-20AD low pressure gradient pump, CBM-20A LITE system controller, SIL-20A
autosampler and an SPD-M20A diode array detector. Separations were performed on a PL
Aquagel-OH 30 column (7.5 × 300 mm) with a mixture of 3:7 MeOH:PBS at a flow rate of 1.0
mL min-1 as the mobile phase.
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 N-(2-hydroxypropyl)methacrylamide (HPMA)
HPMA was prepared as previously described.(52, 53) Briefly, 1-amino-2-propanol (30 g, 0.40
mol) was dissolved in CH2Cl2 (800 mL) followed by the addition of solid NaHCO3 (36 g, 0.44
mol). Freshly distilled methacryloyl chloride (40 g, 0.38 mol) was dissolved in CH2Cl2 (60
mL) and added to the suspension dropwise over the course of 1 h, at 0 °C. The reaction was
allowed to proceed for 3 h. at room temperature. The suspension was filtered and the solvent
removed under reduced pressure. The crude residue was recrystallized from acetone twice and
the crystals dried under vacuum (42 g, 0.29 mol, 73%). 1H NMR (400 MHz, CDCl3): δ = 6.48
(br, 1H, NH), 5.73–5.69 (m, 1H, vinyl Z to Me-C=C), 5.37–5.30 (m, 1H, vinyl E to Me-C=C),
4.00–3.85 (m, 1H, CHOH), 3.53–3.43 (m, 1H, CHaN), 3.28–2.95 (m, 2H, CHbN + OH), 1.95
Page 24 of 40
ACS Paragon Plus Environment

Page 25 of 40
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
13
(s, 3H, C=C-CH3), 1.18 (dd, J = 6.3, 1.9 Hz, 3H, CH3). C NMR (100 MHz, CDCl3): δ =
169.55 (1C, C=O), 139.62 (1C, C=CH2), 120.13 (1C, C=CH2), 67.03 (1C, CHOH), 47.19 (1C,
CH2), 20.89 (1C, CHCH3), 18.66 (1C, CCH3). FT-IR: ν (cm-1) = 3337, 2973, 1655, 1614, 1541.
ESI-TOF-MS: m/z found 144.1011, [M+H]+ requires 144.0980.
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 polyHPMA homopolymers
PolyHPMA homopolymers were synthesised by RAFT polymerization of HPMA following a
known standard protocol.(39) Briefly, for the synthesis of pHPMA P1: HPMA (4.0 g, 28 mmol)
was dissolved in acetate buffer (40 mL, pH 5.2, 1.0 M) along with CPADB (79 mg, 0.28 mmol)
and V-501 (26 mg, 0.092 mmol). The solution was bubbled with nitrogen for 40 min before
heating to 70 °C. Aliquots were taken periodically to monitor monomer conversion by 1H NMR
spectroscopy. Once a conversion of 70–85% was reached, the polymerisation was stopped by
cooling the solution in an ice bath and exposing it to air. Polymers were subsequently purified
by dialysis against water acidified with HCl (pH 3−4) using a 3.5 kDa MWCO membrane for
1
24 h and isolated by lyophilization. H NMR (400 MHz, DMSO-d6): δ = 7.16 (broad s, 1H,
CONH), 4.75 (broad s, 1H, CHOH), 3.76 (broad s, 1H, CHOH), 3.02 (broad s, 2H, CH2),
1.74−1.00 (broad m, 8H, CH2 and CH3 of backbone and CHCH3).
Chain Transfer Agent (CTA) end-group removal
Removal of the end group was performed according to the procedure described by Perrier et
al.(41) A 30-fold molar excess of AIBN with respect to the chain transfer agent was employed.
polyHPMA and AIBN were dissolved in dry DMSO and the solutions were bubbled with
nitrogen for 35 min then heated to 80 °C for 3 h under continuous stirring. The resulting
solutions were dialyzed against water for 24 h and subsequently lyophilized. The final thiol
1
concentration was calculated using a standard Ellman’s assay.(54) H NMR (400 MHz,
DMSO-d6): δ = 7.16 (broad s, 1H, CONH), 4.75 (broad s, 1H, CHOH), 3.76 (broad s, 1H,
Page 25 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 26 of 40
1
2
3
4
CHOH), 3.02 (broad s, 2H, CH2), 1.74−1.00 (broad m, 8H, CH2 and CH3 of backbone and
5
6
CHCH3).
7
8
9
Synthesis of N-t-butoxycarbonyl-4-aminobenzoic acid
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
N-t-butoxycarbonyl-4-aminobenzoic acid was synthesized according to the protocol reported
by Jones et al.(42) In a round bottomed flask, 4-aminobenzoic acid (2.5 g, 18 mmol) was
dissolved in a mixture of 1,4-dioxane (35 mL) and water (20 mL). Triethylamine (5.0 mL, 37
mmol) was added to the solution and left under stirring for 10 minutes before addition of solid
di-tert-butyl dicarbonate (8.0 g, 37 mmol). The reaction was allowed to proceed for 24 h. at
room temperature. The solvent was removed under vacuum, then HClaq (3 M, 35 mL) was
added to the residue to obtain a white precipitate that was recovered by filtration and freeze
dried. The product was crystallized from hot methanol to yield N-t-butoxycarbonyl-4-
aminobenzoic acid as a white solid (3.9 g, 91%). M.p. found 197-199 °C, reported 192-194
°C.(55) .1H NMR (400 MHz, DMSO-d6): δ = 9.73 (s, 1H, CONH), 7.83 (d, 2H, J = 8.9 Hz,
CHarom), 7.55 (d, 2H, J = 8.9 Hz, CHarom), 1.47 (s, 9H, 3 CCH3). 13C NMR (100 MHz, DMSO-
d6) δ (ppm) 167.1 (1C, COOH), 152.6 (1C, CONH), 143.8 (1C, CCOOH), 130.4 (2C, CH),
124.0 (1C, CHNH), 117.2 (2C, CH), 79.7 (1C, C(CH3)3, 28.1 (3C, 3 CH3). ESI-TOF-MS: m/z
found 237.1529, [M+H]+ requires 237.1001.
General procedure for the synthesis of Poly[HPMA]-r-[N-(2-
hydroxypropyl)methacrylamide–4-(tert-butoxycarbonylamino)benzoate
poly(HPMA-r-Boc-PABA)
Synthesis of HPMA-r-Boc-PABA P9: N-t-butoxycarbonyl-4-aminobenzoic acid (268 mg, 1.13
mmol) was dissolved in dry DMSO (5 mL). DMAP (276 mg, 2.22 mmol) and EDC (433 mg,
2.22 mmol) were subsequently added and allowed to dissolve under stirring for 10 min, after
which polyHPMA (540 mg, 3.77 mmol) was added. The reaction was left to proceed for 36 h.
Page 26 of 40
ACS Paragon Plus Environment

Page 27 of 40
Bioconjugate Chemistry
1
2
3
4
5
6
at room temperature; conversion was monitored by 1H NMR spectroscopy. The crude product
was first precipitated from Et2O twice and, subsequently, purified by dialysis against water,
7
8
9
1
using a 3.5 kDa cut-off, for 2 days and freeze dried. H NMR (400 MHz, DMSO-d6): δ= 9.75
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
(br s, CONHBoc), 7.87 (br s, CHarom), 7.59 (br s, CHarom), 7.13 (br s, CONHHPMA), 4.96 (br s,
1H, CHOC(O)), 4.71 (br s, 1H, CHOH), 3.68 (br s, 1H, CHOH), 2.91 (br s, 2H, CH2CH), 2.02-
0.26 (br m, 17 H, CH2CCH3 backbone, CHCH3 and 3 CH3 Boc).
General procedure for the synthesis of poly[HPMA]-r-[N-(2-
hydroxypropyl) methacrylamide–4-amino benzoate] poly(HPMA-r-PABA)
Poly(HPMA-r-Boc-PABA) P9 (450 mg, 0.31 mmol of aniline) was dissolved in dry DMSO (7
mL) and to the resulting solution tetrabutylammonium fluoride (1.6 mL of 1M solution in THF,
1.6 mmol) was added. The reaction was allowed to proceed overnight at 80 °C under reflux.
After cooling to room temperature, the final product was precipitated twice from Et2O and,
subsequently, purified by dialysis (3.5 kDa MWCO) against water for 1 day, and freeze-dried.
1H NMR (400 MHz, DMSO-d6): δ= 7.66 (br s, 2CHarom), 7.16 (br s, CONHHPMA), 6.58 (br s,
2CHarom), 5.92 (br s, 2H, NH2aniline) 4.94 (br s, 1H, CHOCO), 4.73 (br s, 1H, CHOH), 3.68 (br
s, 1H, CHOH), 2.91 (br s, 2H, CH2CH), 2.15- 0.25 (br m, 17 H, CH2CCH3 backbone, CHCH3 and
3 CH3Boc).
Synthesis of oligolysine-N-Boc (K7)
A standard solid phase peptide synthesis procedure was used to synthesize oligolysine-N-Boc
(K7) peptide.(56) Seven N-Fmoc-Lys(Boc) were coupled in sequence to Fmoc-Lys(Boc)
NovaSyn®TGT resin (750 mg, 0.13 mmol) using HATU as the coupling agent. The N-terminus
was acetylated using 20% acetic anhydride in dry DMAc for 45 minutes. The peptide was
cleaved from the resin using 0.1% TFA, 0.5% TIPS in dry DCM, the product dried under
Page 27 of 40
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 28 of 40
1
2
3
4
5
6
reduced pressure and analyzed by HPLC and ESI-MS-ToF. ESI-TOF-MS: m/z found
1679.3453, 851.5477. [M+Na]+ requires 1680.0421, [M+2Na]2+ requires 851.5477.
7
8
9
Synthesis of re-targeting of polymer P9 with K7 (P9-K7)
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
Polymer P9 (10 mg, 69.8 µmol of monomer units) was dissolved in DMSO-d6 (0.7 mL) along
with K7 (2.3 mg, 1.4 µmol), EDC (27 mg, 0.140 mmol) and DMAP (4.3 mg, 34.9 µmol). The
solution was allowed to react under stirring for 24 hours at room temperature. The product was
purified by precipitation in mixture Et2O: DCM 80:20 and twice in water before freeze drying
the product. 1H NMR (400 MHz, DMSO-d6): δ= 8.08-7.82 (s, 7NHBoc) 7.66 (br s, 2CHarom),
7.16 (br s, CONHHPMA), 6.58 (br s,2CHarom), 5.92 (br s, 2H, NH2aniline) 4.94 (br s, 1H, CHOCO),
4.73 (br s, 1H, CHOH), 3.68 (br s, 1H, CHOH), 2.91 (br s, 2H, CH2CH), 2.15- 0.25 (br m, 17
H, CH2CCH3 backbone, CHCH3 and 3 CH3Boc).
General procedure for polymer coating of EnAd
Coating experiments were performed using a 10 mg/mL polymer stock solution. Typically, P9
(3.0 mg, 5.0 µmol of aniline) was dissolved or suspended in HPLC grade water (150 µL) and
10% TFA in water (7.7 µL, 10.0 µmol) under stirring. The mixture was allowed to react for 45
min at room temperature, then cooled with an ice bath. 200 mg/mL NaNO2 and of 150 mg/mL
sulfamic acid stock solutions in water were placed in ice. Under vigorous stirring, ice cold
NaNO2 stock solution (3.5 µL) was added and incubated at 0 °C for 1 hour. Ice cold sulfamic
acid stock solution (6.5 µL) was then added, and after 1 hour at 0 °C, the pH of the solution
was adjusted to 7.4 with ice cold 500 mM HEPES pH 7.6 (132 µL). Polymer-coated virus stock
solutions were prepared with a typical final titre of 5 × 1011 vp/mL in 50 mM HEPES buffer
pH 7.4 at final polymer concentrations of 1 or 5 mg/mL. After addition of the polymer solution
to the virus suspension, the conjugation reaction was allowed to proceed overnight at 4 °C.
Formulations were aliquoted, flash frozen in liquid nitrogen and stored at -80°C.
Page 28 of 40
ACS Paragon Plus Environment

Page 29 of 40
Bioconjugate Chemistry
1
2
3
4
Dynamic Light Scattering (DLS)
5
6
7
8
Particles size analyses were carried out at 25 °C using a Zetasizer Nano spectrometer (Malvern
Instruments Ltd) equipped with a 633 nm laser at a fixed angle of 173°. Samples were
suspended in HEPES buffer pH 7.4 10 mM.
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
Enzyme-Linked Immunosorbent Assay (ELISA)
A MaxiSorp™ NUNC ELISA plate was prepared by coating with 50 µL of 1:500 dilution of
polyclonal rabbit anti-Ad capturing antibody (prepared by Morendum Scientific, Penicuik,
UK) in 0.05 M pH 9.6 carbonate/ bicarbonate buffer overnight at 4°C. After incubation, the
plate was washed four times with PBS, as for each incubation step of the procedure. Non-
specific binding was prevented by incubation of 200 µL blocking solution, consisting of 5%
BSA in PBS, for 1 hour. Next, 50 µL of 1 × 109 virus particles (vp)/mL virus formulations in
blocking solution were incubated for 2 hours at room temperature. 50 µL of a 1:500 dilution of
polyclonal mouse anti-Ad antibody (prepared in house by PsiOxus Therapeutics, Oxford, UK)
was added to the wells. After 1 hour incubation at room temperature, 50 µL of 1: 300 dilution
of goat anti-mouse HRP antibody (secondary antibody) in blocking solution were incubated
for 1 hour. TMB-ELISA substrate was added to the plate ad incubated 20-30 min before
stopping the reaction with 1N HCl. Absorbance was read at 450 nm.
Cell culturing conditions
For the in vitro studies, HT-29 human colorectal adenocarcinoma cells well cultured according
to the procedures to the following procedures. Manufacturers or suppliers of reagents and
equipments are stated where they are first mentioned. Laboratory plastic-ware was received by
Fisher Scientific or Corning Inc. Cell culture media was high glucose Dulbecco’s modified
Eagle’s Medium (DMEM) with L-glutamine and supplemented with 10% FBS (“growth
media”) or 2% FBS (“assay media”).
Page 29 of 40
ACS Paragon Plus Environment