Non-covalent hitchhiking on endogenous carriers as a protraction mechanism for antiviral macromolecular prodrugs

Article abstract of DOI:10.1016/j.jconrel.2018.12.016

Albumin is a highly successful tool of drug delivery providing drastically extended body and blood residence time for the associated cargo, but it only traffics single drug copies at a time. In turn, macromolecular prodrugs (MP) are advantaged in carrying

Full text of DOI:10.1016/j.jconrel.2018.12.016

Accepted Manuscript
Non-covalent hitchhiking on endogenous carriers as a protraction
mechanism for antiviral macromolecular prodrugs
Camilla Kaas Frich, Franziska Krüger, Raoul Walther, Cecilie
Domar, Anna H.F. Andersen, Anne Tvilum, Frederik Dagnæs-
Hansen, Paul W. Denton, Martin Tolstrup, Søren R. Paludan, Jan
Münch, Alexander N. Zelikin
PII:
S0168-3659(18)30719-3
DOI:
https://doi.org/10.1016/j.jconrel.2018.12.016
COREL 9566
Reference:
To appear in:
Journal of Controlled Release
Received date:
Revised date:
Accepted date:
12 October 2018
26 November 2018
11 December 2018
Please cite this article as: Camilla Kaas Frich, Franziska Krüger, Raoul Walther, Cecilie
Domar, Anna H.F. Andersen, Anne Tvilum, Frederik Dagnæs-Hansen, Paul W. Denton,
Martin Tolstrup, Søren R. Paludan, Jan Münch, Alexander N. Zelikin , Non-covalent
hitchhiking on endogenous carriers as a protraction mechanism for antiviral
macromolecular prodrugs. Corel (2018), https://doi.org/10.1016/j.jconrel.2018.12.016
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.

ACCEPTED MANUSCRIPT
Non-covalent hitchhiking on endogenous carriers as a protraction mechanism for
antiviral macromolecular prodrugs
#
a
#b
a
a
Camilla Kaas Frich, Franziska Krüger, Raoul Walther, Cecilie Domar, Anna H. F.
c,d
a
e
c,d
Andersen, Anne Tvilum, Frederik Dagnæs-Hansen, Paul W. Denton, Martin
Tolstrup, Søren R. Paludan, Jan Münch, * Alexander N. Zelikin *
c,d
e
b,
a
a: Department of Chemistry, Aarhus University, 8000 Aarhus C, Denmark
*
Corresponding author Email : zelikin@chem.au.dk
b: Institute of Molecular Virology, Ulm University Medical Center, 89081, Ulm, Germany
Corresponding author Email : Jan.Muench@uni-ulm.de
*
c: Department of Infectious Diseases, Aarhus University Hospital, 8000 Aarhus C, Denmark
d: Department of Clinical Medicine, Aarhus University, 8000 Aarhus N, Denmark
e: Department of Biomedicine, Aarhus University, 8000 Aarhus C, Denmark
#
These authors contributed equally
Abstract
Albumin is a highly successful tool of drug delivery providing drastically extended body and blood
residence time for the associated cargo, but it only traffics single drug copies at a time. In turn,
macromolecular prodrugs (MP) are advantaged in carrying a high drug payload but offering only a
modest extension of residence time to the conjugated drugs. In this work, we engineer MP to contain
terminal groups that bind to albumin via non-covalent association and reveal that this facile measure
affords a significant protraction for the associated polymers. This methodology is applied to MP of
acyclovir, a successful drug against herpes simplex virus infection but with poor pharmacokinetics.
Resulting albumin-affine MP were efficacious agents against herpes simplex virus type 2 (HSV-2)
both in vitro and in vivo. In the latter case, sub-cutaneous administration of MP resulted in local
(vaginal) antiviral effects and a systemic protection. Presented benefits of non-covalent association
with albumin are readily transferrable to a wide variety of MP in development for drug delivery as
anticancer, anti-inflammatory, and anti-viral measures.
Keywords
acyclovir; macromolecular prodrugs; albumin; herpes simplex virus
1

ACCEPTED MANUSCRIPT
Introduction
Endogenous carriers such as antibodies and albumin are the most successful tools of drug delivery. [1-
] Indeed, antibody-drug conjugates have re-shaped the landscape of medicinal chemistry as they
3
afford site-specific drug delivery. In turn, albumin conjugation affords phenomenal drug circulation
times for associated drugs, [2, 4-7] and research activities around this phenomenon recently
culminated in an approval of a once-weekly injectable drug formulation for diabetes therapy
(Semaglutide). Albumin persistence in the body, in large part, is due to its recognition by the neonatal
FcRn receptor through which albumin avoids intracellular degradation. [8] Both covalent attachment
of the drugs and non-covalent association with this protein are state-of-the-art.[2] This is also true for
strategies for in situ covalent ligation to albumin upon drug administration.[9] Of these strategies,
covalent conjugation may lead to a loss of albumin recognition by the FcRn, especially upon
excessive modification of the protein globule. [10] Non-covalent association is advantageous from
this standpoint, and also due to the flexibility of drug administration and most importantly, reduced
handling of the biomolecule. However, non-covalent association with albumin has only been achieved
with individual drug molecules thus affording minimal payload trafficking capacity of the carrier. [2,
4]
We hypothesized that macromolecular prodrugs (MP) as a high payload carrier[11] can drastically
increase the deliverable payload trafficked by albumin. Indeed, we have previously achieved this via
covalent ligation of polymers with albumin[12, 13] and documented that at equimolar conjugation, the
resulting albumin-polymer-drug conjugates have affinity to FcRn similar to that of the pristine
albumin[12]. In contrast, non-covalent association of polymers with albumin is challenging and
typically leads to pronounced protein aggregation.[14] [15] In vivo, this may lead to a fast elimination
of the polymer-protein adduct from circulation via hepatic clearance. [16] Albumin-based
nanoparticles[14] represent a subject of investigation successful in their own right, but the highest
degree of success for albumin as a drug carrier is documented for the molecularly dissolved albumin
adducts. [2, 6] In this work, we specifically aimed to develop a methodology for non-covalent
association of MP with albumin.
One of the prime physiological functions of albumin is trafficking hydrophobic solutes. Drawing
inspiration from nature, non-covalent binding to albumin has previously been documented for a range
of synthetic molecules, namely fatty acids, [6] cholesterol, [17] 4-(p-iodophenyl)butyric acid [18, 19]
and many others. Albumin binding motifs can be conjugated to specific drugs such as peptide
hormones[6] and oligonucleotides[17, 20] to achieve extended blood and body residence time
(protraction). We hypothesized that using albumin binding domains as terminal groups for polymers
would afford the desired architecture of albumin-affine macromolecular prodrugs. Specifically, as a
polymer backbone, we chose poly(N-(2-hydroxypropyl) methacrylamide), PHPMA, due to its prior
history of use as a drug carrier. [21] In this work, we present the design of MP containing albumin-
binding ligands, investigate the non-covalent association between the polymers and albumin, conduct
a comparative in vivo analysis of the pharmacokinetics for PHPMA, its albumin-affine derivative, and
the covalent conjugate between albumin and PHPMA (albumin-co-PHPMA). Finally, developed
methodology is applied to the macromolecular prodrugs of acyclovir as an innovative systemic
therapy against herpes simplex virus type 2 (HSV-2).
2

ACCEPTED MANUSCRIPT
Results and discussion
Literature survey identified a suitable non-covalent albumin binding handle, 1,2-distearoyl-sn-
glycero-3-phosphoethanolamine polyethylene glycol, DSPE-PEG. [22] This molecule is conveniently
available from several commercial sources. We used DSPE-PEG to obtain a novel chain-transfer
reagent used in the Reversible Addition-Fragmentation chain-Transfer polymerization technique
(RAFT). [23] RAFT methodology is well-adapted towards the synthesis of polymer samples with
narrow dispersity, that is, syntheses to obtain controlled molar mass of the polymer. This
methodology relies on the use of the RAFT agent such that repetitive addition-fragmentation of the
radical leads to equilibration of the growing polymer chains and an ensuing relative uniformity of
chains by molar mass. The latter characteristic is what typically defines the properties of the polymer
chains such as the hydrodynamic radius and therefore the blood circulation time. [24] However, in the
design of non-covalent albumin binders, control over pharmacokinetics of the molecule is transferred
from the polymer to albumin and as such, the polymer molar mass may not be a decisive factor.
Nevertheless, RAFT is still highly important in that it enables the preparation of albumin-affine MP
with the use of a DSPE-PEG containing RAFT agent, (Figure 1). The function of this RAFT agent is
to ensure that the polymer chains grow from the DSPE-PEG to afford the desired structure. In this
work, DSPE-PEG RAFT agent was used to polymerize HPMA into its polymer containing albumin
binding handle, DSPE-PHPMA, as well as the corresponding macromolecular prodrugs (vide infra).
Macromolecular characteristics of the polymers are listed in Table 1.
Figure 1. (A) Structural formulas of the DSPE RAFT agent, HPMA, and acyclovir. (B) Synthesis of DSPE RAFT
agent in two steps and synthesis of DSPE-PHPMA polymers; (C) Schematic illustration of the DSPE-PHPMA-ACV
MP with affinity to albumin and the non-covalent association of MP with albumin to comprise a high payload drug
delivery vehicle with albumin as the tool of protraction.
3

ACCEPTED MANUSCRIPT
Albumin binding
Albumin binding for PHPMA and DSPE-
PHPMA was analyzed in vitro using a probe
displacement assay. [16, 25] Human serum
albumin was mixed with probes known to bind
one of the two most well-described binding sites
for albumin, the Sudlow I and Sudlow II sites
(
used for binding by warfarin and salicylic acid,
or naproxen and ibuprofen, respectively)[26, 27].
The adduct formed by the probe and albumin is
fluorescent whereas both albumin and the probe
taken individually are non-fluorescent. Thus,
competitive displacement of the probe by the
added albumin binder would result in a decrease
of solution fluorescence and thus provide a
Figure 2. Albumin competition assay showing the
displacement of the Sudlow I and Sudlow II probes (dansyl
means to quantify the binding. In this study asparagine and dansylsarcosine, respectively) from albumin
dansyl-L-asparagine was used as a probe for the by DSPE-HPMA (P5), HPMA (P1) and PEAA (for PEAA,
data are taken from Ref. 16.)
Sudlow I site, whereas dansylsarcosine was used
as a probe for the Sudlow II site. [25] Pristine
PHPMA revealed no measurable binding to
Sudlow I site and only a minor affinity to the
Sudlow II site, in the latter case only at polymer
to albumin ratio exceeding 10 (Figure 2). In
contrast, addition of polyethylacrylic acid
(PEAA), the most potent polymer binder we
have identified in our prior work, [16] afforded
probe displacement at ratios only just over unity
and a complete fluorescence quenching at 10-
fold polymer excess over albumin, for both
Sudlow I and II sites (Figure 2). DSPE-PHPMA
revealed an intermediate level of binding to
albumin and specifically to the Sudlow I site.
Binding was more pronounced than for the
pristine PHPMA but less potent than for PEAA
Figure 3. Size exclusion chromatography elugrams (MALS
detector) illustrating the interaction between albumin and
(Figure 2). This data set thus reveals that DSPE DSPE-PHPMA. PanelB is a close-up view of thealbumin
modification of the polymers indeed renders monomer elution region from panel A.
these albumin-affine, with
binding to the Sudlow I site.
a
predominant
Albumin binding was also analyzed by size-exclusion chromatography with multi-angle laser light
scattering detection (SEC-MALS), using varied polymer-to-albumin ratios (1, 5 and 10 equivalents of
polymer). The SEC-MALS analysis revealed a decrease in retention time for the albumin monomer
peak from 18 min to lower retention times, which is attributed to the non-covalent protein interaction
with the polymer. By increasing the amount of polymer, the albumin peak is further left-shifted. It is
noted that addition of the polymer also affords a minor increase in the amount of aggregates from
albumin (peak at 12 min, about 0.1 % of the total sample when quantified using the UV-vis detector),
but the majority of albumin is molecularly dissolved. Thus, SEC-MALS and probe displacement
assays confirmed that DSPE-PEG terminal groups on the polymer chain produce a polymer with
affinity to albumin.
4

ACCEPTED MANUSCRIPT
Figure 4. Quantification of PHPMA, DSPE-PHPMA, and covalent conjugate between PHPMA and albumin (denoted
Albumin-co-PHPMA) in mice blood following i.v. or s.c. administration, over 7 days. (A) Quantification of DSPE-PHPMA
in blood; (B) Concentration of DSPE-PHPMA in blood normalized to thehighest value observed for therespective route of
administration; (C) Quantification of the solutes in blood upon s.c. administration. Fractions were compared with an
unpaired t-test, and * represents p<0.05 and ** is p<0.01 For all graphs, each data point represents and average of 4
individual mice ± SEM. Data for PHPMA and Albumin-co-PHPMA are taken from Ref. 12
Biodistribution/pharmacokinetics. DSPE-PHPMA was labelled with a Cy7 fluorophore and
administered into Balb/C mice via sub-cutaneous (s.c.) or intravenous (i.v.) routes. Concentrations of
the administered polymer in blood were quantified in periodically drawn blood samples via
fluorescent measurements on an In Vivo Imaging System. In agreement with recent reports,[7] s.c.
administration expectedly afforded a lower absolute concentrations of the polymer in the blood than
the i.v. route (Figure 4A, B). I.V. administration resulted in a gradual decrease in the DSPE-PHPMA
content with a blood residence half-life of 138 min. In turn, s.c. route showed a significant advantage
in that it afforded an extended period of time during which the solute concentration in blood was
fairly constant, a pharmacokinetics profile that is highly desired for long-acting formulations. This is
likely due to the superposition of polymer elimination process with supplementation of blood polymer
levels from the administered solute present in the subcutaneous space. [7] This observation is very
similar to the previously recorded behavior of PHPMA and its covalent conjugate with albumin
(denoted as Albumin-co-PHPMA). [12] However, the time period characterized by a steady
concentration of the solute in blood is vastly enhanced for the DSPE-PHPMA compared to the
PHPMA (Figure 4C). In fact, the DSPE-PHPMA blood concentration over time profile is near-
identical to that of the covalent conjugate Albumin-co-PHPMA and the sustained blood concentration
of the solute is observed between 2 and 12 h post administration. Protraction of the administered dose
is also highly significant at later points and at 24 h, at least 30% of the maximum and at least 20% at
48 h after dose administration. Thus, non-covalent protraction using albumin is successfully
engineered into synthetic polymers. Furthermore, we found that using DSPE-PEG, albumin binding is
non-inferior to the covalent binding to albumin with regards to the body and blood residence time of
the conjugate.
5

ACCEPTED MANUSCRIPT
Figure 5. Quantification of PHPMA, DSPE-PHPMA, and albumin-PHPMA covalent conjugate in mice organs following
i.v. or s.c. administration of thesamples and distribution time up to 7 days, presented as the fold-change compared to the 1
h timepoint. LN: lymph nodes. Data for PHPMA and Albumin-co-PHPMA aretaken from Ref. 12
We also analyzed organ distribution of the solutes. Figure 5 shows organ content for the polymers
normalized to the respective content registered at 1 h after drug administration such as to illustrate
accumulation or depletion of the polymer from the organs. Upon i.v. administration, over time we
observed a depletion of DSPE-PHPMA from all organs, including the liver. One notable exception
from this observation is the relatively constant content of DSPE-PHPMA in the lymph nodes for at
least 7 days, which may be favorable for lymphatic drug delivery (antiretroviral therapy, vaccination,
etc). In contrast, s.c. administration afforded a pronounced organ accumulation for DSPE-PHPMA
during the first 24 h after administration (note the Y-axis scale difference between the panels).
Somewhat surprisingly, organ distribution profile for DSPE-PHPMA was rather similar to that of the
pristine PHPMA polymer and varied significantly from that of the covalent Albumin-co-PHPMA
conjugate. The latter shows the most significant organ accumulation, specifically in the lymph nodes
at day 7, a feature we previously reported in the context of lymphatic delivery of antiretroviral
therapy[12]. DSPE-PHPMA does not show a comparable organ accumulation. This observation
suggests [7] that for the non-covalent albumin binder used in this work, translocation from s.c. space
to blood is facilitated for DSPE-PHPMA by albumin (Figure 4), whereas subsequent body and organ
distribution is dictated by the polymer (Figure 5).
6

ACCEPTED MANUSCRIPT
Macromolecular prodrugs ofacyclovir.
Having validated a favorable in vivo biodistribution profile for the DSPE-PHPMA, we aimed to use
this polymer as a carrier for systemic delivery of antiviral drugs and specifically acyclovir. [28, 29]
The latter is a highly successful drug against infections with herpes simplex virus (HSV) type 1 and 2
but suffers from poor bioavailability, largely because of poor water solubility. Oral administration of
this drug is therefore practiced seldom, and it requires high daily doses. More commonly, acyclovir is
used as a topical formulation which only has a localized effect and is not suited for systemic treatment
of HSV. We hypothesized that macromolecular prodrugs with an affinity to albumin would become a
tool for systemic delivery of acyclovir.
Pursuit of macromolecular prodrugs of acyclovir required de novo design of the suitable monomers.
To our knowledge, acyclovir conjugates have hardly been investigated with the sole example we are
aware of being an ester-based ligation of the drug to the carrier polymer. [30] In our prior studies, we
also used this method [31] but subsequently developed an alternative methodology for conjugation of
nucleoside analogues, with significantly improved drug release profile. [32] [33] For the latter, the key
design feature was the presence of disulfide as the scissile bond in the monomer structure. Disulfides
are stable in blood but undergo a fast thiol-disulfide reshuffling in the presence of thiols such as the
natural tripeptide, glutathione, GSH. [34] This tripeptide is present in mammalian cells in millimolar
concentrations and efficiently triggers disulfide cleavage upon the drug cell entry. Acyclovir does not
have a thiol to conjugate to a carrier via a disulfide. To use the disulfide chemistry, we used a self-
immolative linker that undergoes spontaneous decomposition upon the disulfide-to-thiol conversion,
and releases the parent drug. [15, 35] We have previously used this design in the syntheses of
macromolecular prodrugs of nucleoside analogues[15] and HIV latency reversal agents[13, 36].
The full synthesis scheme leading to the acyclovir-containing monomer is depicted in Figure 6. 2-
hydroxyethyl disulfide was mono-functionalized with a methacrylate functionality. Independently,
acyclovir was reacted to afford a 4-nitrophenyl-activated carbonate. Coupling of the activated
carbonate and the disulfide derivative produced the desired monomer. This synthesis strategy was
used in the initial studies but was sub-optimal in that it contained a preparative HPLC purification
step due to insolubility of the ACV-SIL monomer in common organic solvents. An alternative
strategy was developed for a scaled up synthesis of MP, starting with protection of the primary amine
functionality of the guanine with a 4-monomethoxytrityl-group (MMTr). The intermediate products
were soluble in common organic solvents such as dichloromethane or chloroform and only required
common purification methods. Activation of ACV-MMTr was accomplished with bis(4-
nitrophenol)carbonate in presence of a mild base. Notably, activation with the more reactive 4-
nitrophenol chloroformate showed considerable loss of the MMTr-group, presumably due to the loss
of HCl during the reaction. Reacting 4 with 1 in the presence of a base yielded the ACV-MMTr SIL
monomer 5 in 34% yield overall.
7

ACCEPTED MANUSCRIPT
Figure 6. Synthesis of the disulfide containing ACV/ACV-MMTr SIL monomer. a) p-nitrophenylchloroformate and
TEA in DMF/CH
bis(4-nitrophenyl)carbonate, TEA, CH
,2'-Azobis(4-methoxy-2,4-dimethylvaleronitrile) and g) azobisisobutyronitrile. RAFT agent R groups (OH/TA/DSPE-
PEG) are specified in Experimental section and in Table 1. Removal of the MMTr protectinggroup h) TFA, Et O.
2 2
Cl , 52 %, b) 1, DIEA/DMAP in DMF/THF; c) TMS-Cl, pyr, 15 min; MMTr-Cl, DMAP, 59%;d)
2
Cl , 0°C, 71% e) 1, TEA/DMAP, CH Cl , 80%. Polymerization initiator was f)
2
2
2
2
2
The acyclovir monomer was used to obtain a series of MP using HPMA as a comonomer. Polymers of
the latter have extensive prior characterization in biomedicine and endow the conjugated drug with
enhanced aqueous solubility, highly desirable for ACV which suffers from poor aqueous solubility.
MP were synthesized using DSPE-PEG RAFT agents as well as the commercially available RAFT
agents for control purposes. To obtain fluorescently labelled polymers, we also used
azidomethacrylate monomer with subsequent coupling of the fluorophore (Cy7 alkyne) via the click
reaction. Macromolecular characteristics of the synthesized polymers are listed in Table 1.
Release of acyclovir from the macromolecular
prodrugs was quantified using HPLC. In
agreement with the design considerations, MP
release insignificant quantities of acyclovir in
physiological buffer (phosphate buffered saline)
even over 24 h of incubation. In contrast, in the
presence of GSH taken at a physiologically
relevant concentration of 5 mM, the polymer
readily released acyclovir (Figure 7), thus
validating the proposed design of the monomer.
We note that detailed quantitative investigation
of kinetics of drug release was impeded by the
low aqueous solubility of the drug (precipitation
upon activity of GSH).
Figure 7: (A) Illustration of themechanism of drug release
for ACV prodrugs upon disulfide reshuffling; (B) HPLC
traces illustrating the release of ACV from its macromolecular
prodrug in the presence of 5 mM GSH in PBS (10 mM, pH =
7
.4) after 3 h. Quantitativedata is represented as mean ± SD
(N=3). * p < 0.01
8

ACCEPTED MANUSCRIPT
Table 1. RAFT agent, conversions, and macromolecular characteristics for the polymers synthesized and used in this work.
Type
RAFT agent (R group)
Conv. HPMA
Conv. mon. 2 DP HPMA
DP mon. 2
M
n (NMR)
M
n (SEC-
% mon. 2 (NMR)
Ð
(
NMR)
(NMR)
-
kDa
8.5
MALS), kDa
8.9
PHPMA (P1)
OH
OH
33%
67%
42%
39%
68%
44%
40%
-
59
-
1.1
1.1
1.1
1.1
1.3
1.3
1.2
PHPMA-co-ACV (P2)
PHPMA-co-AzMA (P5)
PHPMA-co-AzMA (P6)
PHPMA-co-ACV (P7)
PHPMA-co-ACV (P8)
PHPMA-co-ACV (P9)
67%
-
60
3
10.1
8.7
14.3
12.1
5.1
6.5%
2%
DSPE-PEG
37.8
35.6
68.7
38.7
35.2
0.7
0.8
4.0
6.9
7.7
OH
-
5.1
1.5%
7%
DSPE-PEG
79%
78%
88%
14.9
12.0
8.7
14.5
13.5
7.1
DSPE-PEG
12%
11%
thiazolidine-2-thione (TA)
9

ACCEPTED MANUSCRIPT
In vitro antiviral activity against HSV-2. To assess the in vitro anti-viral activity of the synthesized
prodrugs of ACV, ELVIS reporter cells were used [37]. These cells are genetically engineered baby
hamster kidney cells containing an HSV inducible promoter gene chimerically linked to an E. coli
lacZ "reporter" gene. Induction of the promoter by HSV leads to the production of the LacZ product,
β-galactosidase. Cells were incubated with the respective prodrugs at concentrations ranging from 0 to
100 mg/L for one hour, and were then infected with HSV-2. Read-out of infection was performed
after a period of time that corresponds to ca. 2 replication cycles of the virus via quantification of the
β-galactosidase activity using a sensitive chemiluminescence based assay. This common model of
HSV-2 infectivity is characterized by an early expression of β-galactosidase (as early as 6 h post-
infection with HSV [38]) and a relatively late expression of the viral thymidine kinase,[39-41] the
latter being necessary to phosphorylate acyclovir and to observe the associated antiviral effects. [42]
This means that inhibition of HSV-2 infectivity is only complete for inhibitors of viral cell entry such
as heparin [43] whereas intracellular ACV that acts at a later stage of the viral replication cycle is
expected to exert pronounced but not exhaustive decrease of viral infectivity. In agreement with this,
HSV-2 infection was entirely inhibited at elevated concentrations of polyanionic heparin (Figure 8).
In contrast, pristine acyclovir revealed antiviral activity with efficacy not exceeding 50% inhibition.
Noteworthy, treatment with acyclovir was associated with a significant anti-proliferative effect too, as
Figure 8. Dose responsecurves reflecting activity of heparin, ACV, and MPs in suppressingHSV-2 replication. ELVIS cells
were treated with heparin, ACV and MPs for one hour. Thereafter, cells were infected with HSV-2 eGFP and infection rates
were determined 48 h post infection by quantifying β-galactosidase activity in cellular lysates. Cell viability was determined
by MTT assay of ELVIS cells incubated with heparin, ACV and MPs for 48 h. Values shown are means ± SEM derived from
three independent experiments performed in triplicates.
is well documented. [44]
Pristine, drug-free PHPMA (P1) and its conjugates with albumin expectedly had no antiviral activity.
In contrast, MP of ACV revealed strong antiviral activity characterized by a broad range of
concentration with antiviral effect devoid of associated cellular toxicity. The same was true for the
DSPE-PHPMA-ACV conjugates (P7 and P8 differ by drug loading, 7 mol % and 11 mol % drug
content, respectively) and especially for P8, where the associated cytotoxicity was minimal. Finally,
activity of the covalent conjugate between PHPMA-ACVand albumin was also potent and efficacious
as an anti-HSV-2 agent in vitro. We note that DSPE-PHPMA-ACV and the covalent conjugate of
10

ACCEPTED MANUSCRIPT
PHPMA-ACV with albumin appear to be significantly superior to ACV in that the prodrugs afford a
strong antiviral effect with drastically decreased toxicity as compared to ACV.
The inhibitory effect was further confirmed using fluorescence microscopy of HeLa cells infected
with an eGFP expressing HSV-2. This combination is convenient in that HeLa cells do not exhibit
strong cytopathic effects upon HSV-2 infection in comparison to other cell lines[45], and the GFP
signal is not interfering with the expression of the thymidine kinase [46]. Cells were treated with
heparin, acyclovir, or macromolecular prodrugs of acyclovir, in the same concentration range as in
Figure 8. After ~ 3 replication cycles of the virus, cells were fixed, stained for cell nuclei with
Hoechst 33342, and imaged using a fluorescent microscope (Figure 9). Microscopy imaging produced
results that fully confirmed the quantitative data shown in Figure 8, and while PHPMA and its
conjugates with albumin were devoid of antiviral activity, macromolecular prodrugs containing ACV
decreased infectivity of HSV-2. This includes the covalent conjugate of PHPMA-ACV with albumin
as well as the DSPE-PHPMA-ACV copolymers. Together, results presented in Figures 8 and 9
Figure 9. Fluorescence microscopy images of HeLa cells infected with HSV-2 eGFP in thepresence or absence of heparin,
ACV, and MPs of ACV based on PHPMA, DSPE-PHPMA, and covalent conjugate between PHPMA-ACVand albumin.
HeLa cells were incubated with the indicated concentrations of thecompounds for one hour and were then infected with
HSV-2 eGFP. 72 h post infection, cells were fixed in 4% PFA and stained with Hoechst 33342. Microscopy image
acquisition was performed using a Cell Imaging Multi-ModeReader and Gen 5 softwarefor image processing. Experiment
was performed in triplicates, one representativeimage is shown for each condition.
11

ACCEPTED MANUSCRIPT
present the macromolecular prodrugs of ACV as efficacious inhibitors of HSV-2 infection.
Antiviral activity of ACV-prodrugs in vivo:
Encouraged by the favorable in vitro activity of the prodrugs, these polymers were next tested as
systemic agents to prevent HSV infection and disease in mice. Specifically, in vivo activity of the MP
was quantified using PHPMA and DSPE-PHPMA containing acyclovir in an HSV-2 mouse model.
Prodrugs were administered s.c. at a dose matching 60 mg/kg dosing for acyclovir (being a standard
orally administered dose for this model). To contain the envisioned administered dose within 100 µL
of administered liquid, efforts were made to ensure complete solubility of the polymer in phosphate
buffered saline. While PHPMA-ACV proved to be fully soluble at the required concentration, DSPE-
PHPMA-ACV was less soluble and could not be reconstituted to the required concentration. To
overcome this, the polymer sample was dissolved in dimethylsulfoxide, diluted in PBS in the presence
of albumin, dialyzed, and concentrated to the required volume in PBS. This procedure afforded a fully
soluble preparation of the MP, apparently due to the solubilizing power of albumin. Pristine acyclovir
dose was very insoluble at the required concentration and was administered into mice as a suspension.
This approach to drug administration is similar to that pursued for administration of antiretroviral
therapy whereby gradual drug dissolution affords sustained drug concentration. [47, 48] However,
from the practical standpoint, administration of suspensions is disadvantaged as poorly reproducible,
and was used here only to match the administered dose to the literature reported values. [49]
Acyclovir and the prodrugs were administered s.c. following which mice underwent vaginal infection
with HSV-2 (5 mice per group). Animals were observed for the subsequent seven days while keeping
clinical scoring of the disease progression daily (animals with clinical disease score over three were
Figure 10. Analysis of the anti-HSV-2 in
vivo activity for acyclovir, PHPMA-ACV
and DSPE-PHPMA-ACV, all taken at 60
mg/kg equivalent dose of ACV. (A)
Kaplan-Meier plot (% disease free
animals, treatments were compared to
non-treated using a log-rank (Mantel-Cox)
test) and (B) average disease scores
(means were compared to non-treated by a
two-way ANOVA at each time point). (C)
Virus titer in vaginal lavage obtained at
day 2 and medulla spinalis at day 7 post
infection. Treatments were compared to
the non-treated group using the Kruskal-
Wallis tests with Dunn’s multiple
comparisons tests. Dashed lines indicate
the detection limit.
12

ACCEPTED MANUSCRIPT
euthanized). Virus titer was quantified in vaginal wash at day 2 and in medulla spinalis at day 7.
Animals in the control (no treatment) group revealed clinical manifestations of HSV progression and
all animals reached the maximum clinical disease score by day 5. In contrast, animals that received
suspension of ACV remained disease-free and exhibited reduction of the viral titers below the
respective levels of detection both in the vaginal wash and in medulla spinalis. Most importantly,
animals that received PHPMA-ACV and DSPE-PHPMA-ACV were also protected from the HSV
disease progression. In the prodrug treatment groups, only one animal in five revealed a mild HSV
progression. The same animals exhibited virus titers above the detection limit in both vaginal wash
and in medulla spinalis, whereas four out of five animals were disease-free with no detectable viral
loads in both compartments. Thus, MP containing ACV prevent HSV-2 infection and disease
development in vivo. We acknowledge that in the presented form, in vivo experiments only addressed
the non-inferiority of the PHPMA-ACV and DSPE-PHPMA-ACV. Experiments are now underway to
investigate superiority of the prodrug system developed in this work, specifically through monitoring
the duration of antiviral protection afforded by the (pro)drugs.
Conclusions.
In this work, we developed macromolecular prodrugs for acyclovir and demonstrated their systemic
efficacy in protecting mice from HSV-2 infections. Specifically, prodrugs were developed to contain
ACV conjugated via a disulfide linkage, for which purpose we used a self-immolative linker to
accommodate the use of thiol-disulfide chemistry in thiol-free molecules. Polymers were designed to
bind albumin such as to create injectable formulations of acyclovir. Prodrugs were efficacious in vitro
and in vivo. S.c. administration of the prodrug prior to vaginal administration of the virus prevented
disease progression which highlights this formulation’s systemic antiviral activity. Further
experiments are now underway to optimize the dosing of MP and specifically DSPE-PHPMA-ACV
and to validate the superiority of this drug administration approach over other available options in
terms of duration of treatment.
Acknowledgements. We wish to acknowledge financial support from the Danish Council for
Independent Research, Technology and Production Sciences, Denmark (ANZ; project DFF – 4184-
00177); the Deutsche Forschungsgemeinschaft (JM; CRC1279); the Aarhus University Research
Foundation (PWD, project AUFF-E-2016-FLS-8-9), the Novo Nordisk Foundation (SP, project
NNF18OC0030274) and The Lundbeck Foundation (SP, R198-2015-171). We thank Lise Sørensen
and Kirsten S. Petersen (Aarhus University) for experimental assistance, and Drs .Nikolaj L.
Villadsen and Thomas B. Poulsen (Aarhus University) for providing access to a preparative HPLC.
13

ACCEPTED MANUSCRIPT
Experimental section.
Synthesis
All chemicals were purchased from Sigma-Aldrich and used without further purification unless stated
otherwise. DSPE-PEG(2000) amine was purchased from Laysan Bio. Acyclovir was purchased from
ApiChemistry. N-(3-azidopropyl) methacrylamide (AzMA) was synthesized as described in the
literature. [50] Dansylsarcosine and dansyl-L-asparagine were synthesized as previously
described.[16]
Deuterated solvents were purchased from Euriso-Top. All moisture and air sensitive reactions were
performed in flame-dried glassware under a positive pressure of argon or N
2
. Dichloromethane
(
CH Cl ) was dried over aluminium oxide via an MBraun SPS-800 solvent purification system.
2
2
Dimethylformamide (DMF) and pyridine were purchased in an anhydrous state from Sigma-Aldrich.
Thin layer chromatography (TLC) analysis was carried out on silica coated aluminum foil plates
(
Merck Kieselgel 60 F254). The TLC plates were visualized by UV irradiation and/or by staining with
KMnO -stain. Flash column chromatography was carried out using silica gel – high-purity grade
w/Ca, ~0.1%, 230-400 mesh particle size, 60 Å pore size) as the stationary phase acquired by Sigma-
4
(
Aldrich. Mass spectra (High Resolution Mass Spectrometry - HRMS) were recorded on a Micromass
LC-TOF spectrometer with positive electrospray ionization. Nuclear magnetic resonance (NMR)
spectra were recorded on a Varian Mercury 400 MHz spectrometer, running at 400 and 101 MHz for
1
13
H and C, respectively. Chemical shifts (δ) are reported in ppm relative to the residual solvent
signals: chloroform 7.26 ppm and 77.16 ppm, dimethylsulfoxide 2.50 ppm and 39.52 ppm, deuterium
1
13
oxide 4.79 ppm for H- and C-NMR, respectively. Multiplicities are indicated using the following
abbreviations: s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, bs = broad signal.
Analytical HPLC experiments were carried out on Shimadzu-LC-2010A HT with a Ascentis®
Express Peptide ES-C18 column with 2.7 μm particles, a length of 150 mm and an internal diameter
of 3.0 mm from Supelco Analytical. Mobile phases were ultrapure H
supplemented with 0.1 % TFA (v/v). Ultrapure water (MQ) was received from Milli Q direct 8 system
Millipore). Size exclusion chromatography was performed on a LC-20AD Shimadzu HPLC system
2
O
and acetonitrile
(
with a Wyatt Dawn Heleos 8-angle light scattering detector, a RID-10 Shimadzu refractive index
detector, and a SPD-M20A UV/VIS detector. Polymers were analyzed on the following column
systems: Samples run in DMF were analyzed on either a Mz-Gel SDplus Linear column from MZ
Analysentechnik (5m particels, length 300 mm, diameter 8 mm) at 40°C at a flow rate of 1 mL/min
or on a three column system with PLgel 1000Å, a PLgel 500Å, and a PLgel 50Å from Agilent (5m,
length 300 mm and diameter 7.5 mm) at 40°C at a flow rate of 1 mL/min.
Synthesis ofthe C12H25S/thiazolidine-2-thione (Z/R) RAFT agent was performed according to the
previously published protocols. [13] 4-Cyano-4-[(dodecylsulfanylthiocarbonyl)sulfanyl]pentanoic
acid (205 mg, 0.508 mmol, 1 equiv.), 4-dimethylaminopyridine (DMAP) (8 mg, 0.0655 mmol, 0.12
equiv.), and 2-mercaptothioazoline (68 mg, 0.57 mmol, 1.12 eqiuv.) were dissolved in dry DCM (15
mL). N,N'-Dicyclohexylcarbodiimide (DCC) (120 mg, 0.582 mmol, 1.14 equiv.) was added to the
mixture. The reaction was stirred at room temperature until full conversion at 6.5 hours (monitored by
thin-layer chromatography). The reaction mixture was filtered and concentrated in vacuo, and the
product purified by flash chromatography (EtOAc/Pentane 1:3) yielding the pure product as a yellow
solid (64 mg, 0.508 mmol, 25%).
1
H NMR (400 MHz, CDCl
H), 2.61 (m, 1H), 2.48 (m, 1H), 1.89 (s, 3H), 1.69 (p, J = 7.4 Hz, 2H), 1.49 – 1.35 (m, 3H), 0.88 (t, J
7.1 Hz, 1H).
3
) δ 4.58 (t, J = 7.5 Hz, 2H), 3.69 – 3.43 (m, 2H), 3.32 (td, J = 7.4, 1.5 Hz,
4
=
14

ACCEPTED MANUSCRIPT
Synthesis ofC12
H
25S/DSPE-PEG (Z/R) RAFT agent. DSPE-PEG-amine (210 mg, 0.0753 mmol, 1
equiv.) was dissolved in DCM (2 mL), upon addition of a solution of Z/R (C12H25S/thiazolidine-2-
thione) RAFT agent (64 mg, 0.127 mmol, 1.68 equiv.) in DCM (5 mL) the reaction was stirred for 24
hours at room temperature. The solvent was removed in vacuo leaving a sticky oil. Addition of cold
ether caused formation of a solid precipitate, which was washed with ether 4 times yielding the pure
1
product as a yellow solid (162 mg, 0.0753 mmol, 68%). H NMR (400 MHz, CDCl3) δ 7.90 (s, 0.5H),
6
3
2
.43 (s, 0.5H), 5.21 (s, 1H), 4.36 (dd, J = 12.0, 3.6 Hz, 1H), 4.16 (dd, J = 12.0, 6.5 Hz, 1H), 4.10 –
.97 (m, 5H), 3.64 (m, 166H), 2.58 – 2.43 (m, 2H), 2.28 (m, 4H) 1.89 (s, 3H), 1.68 (p, J = 7.8 Hz,
H), 1.45 (s, 2H), 1.39 (s, 2H), 1.25 (s, 72H), 0.87 (t, J = 6.8 Hz, 9H).
Synthesis ofC H /thiazolidine-2-thione (Z/R) RAFT agent. 4-Cyano-4-
6
5
(phenylcarbonothioylthio)pentanoic acid (1.031 g, 3.69 mmol, 1 equiv.), DMAP (0.052 g, 0.426
mmol, 0.12 equiv.) and 2-mercaptothioazoline (0.483 g, 4.05 mmol, 1.1 equiv.) was dissolved in
DCM (150 mL). DCC (0.860 g, 4.17 mmol, 1.13 equiv.) was added to the mixture. The reaction was
followed with TLC. After 19 hours TLC showed full conversion. The crude was concentrated in
vacou, followed by purification with flash column chromatography (3:7, EtOAc:pentane) yielding the
1
product as a red oil (0.346 g, 0.909 mmol, 25%) Rf (3:7, EtOAc:pentane) = 0:24 H-NMR (400 MHz,
CDCl
3
): δH (ppm) 7.90 (d, J= 8.0 Hz, 2H, H3), 7.56 (t, J=7.4 Hz, 1H, H1), 7.39 (t, J= 7.8 Hz, 2H),
4
.58 (t, J=7.5 Hz, 2H), 3.76-3.51 (m, 2H), 3.31 (t, J= 7.5 Hz, 2H), 2.76-2.48 (m, 2H), 1.95 (s, 3H).
1
3
C-NMR (101 MHz, CDCl
3
): δC (ppm) 201.87,172.56, 144.73, 133.15, 128.72, 126.86, 118.80,
+
56.14, 45.84, 34.59, 33.46, 28.60, 24.42. HRMS (ESI+): calcd. for C16
H
16
N
2
OS + H :381.0173;
4
+
found 381.338, calcd. For C16
H
16
N
2
OS + Na :402.9993; found 403.0038
4
Synthesis ofC
equiv.) was dissolved in DCM (14 mL). Z/R C
6
H
5
/DSPE-PEG (Z/R) RAFT agent. DSPE-PEG-amine (1427 mg, 0.512 mmol, 1
/thiazolidine-2-thione RAFT agent (327 mg, 0.86
6
H
5
mmol, 1.68 equiv.) was dissolved in DCM (35 mL) and added. After 24 hours the solvent was
removed leaving a red, sticky oil. After washing with cold ether a pink solid precipitate formed. The
solid was washed with ether four times. The product was isolated as pink solid (1339 mg, 0.439
1
mmol, 81%) H-NMR (400 MHz, CDCl ): δH (ppm) 7.90 (d, J= 7.4 Hz, 1H), 7.83 (s, 1H), 7.56 (t,
3
J=7.3 Hz, 0.5 H), 7.41 (t, J= 8.3 Hz, 1H), 5.21 (p, J= 4.9 Hz, 1H), 4.35 (dd, J=11.9; 3.9 Hz, 1H),
4.21 - 4.09 (m, 3H), 3.64 (s, 172H), 3.46 (t, J=5.1 Hz, 2H) , 2.30 (q, J= 8.2 Hz, 18H), 1.59 (bs, 3H),
1.46 (t, J=7.3 Hz, 2H), 1.25 (s, 51H), 0.87 (t, J= 6.8 Hz, 6H)
+
HRMS (ESI+): calcd. for C145
H
277
N
4
O
55PS
2
+ H :3050.8273; found 3050.3319, calcd. for
+
C
145
H
277
N
4
O
55PS
2
+ Na :3072.8092; found 3072.7094
Monomer synthesis.
Synthesis ofSIL monomer (compound 1) was performed as described previously. [32]
Triethylamine (21.5 mL, 155 mmol, 2 equiv.) was added dropwise to a solution of 2-hydroxyethyl
disulfide (23.9 mg, 155 mmol, 2 equiv.) in dry DCM (350 mL) at 0 °C. Methacryloyl chloride (8.1
mg, 77.5 mmol, 1 equiv.) was added dropwise to the mixture, upon which the reaction mixture was
allowed to heat up to room temperature. The reaction mixture was stirred for 23 hours, quenched with
saturated NH
4
Cl(aq.), washed with brine and NH
4
Cl(aq.), dried over MgSO , and concentrated in
4
vacuo. The product was purified by flash chromatography (Pentane/EtOAc from 9:1 to 6:4) yielding
the pure product as a yellow oil (10.9 mg, 77.5 mmol, 60%).
1
H NMR (400 MHz, CDCl ) δ 6.14 (dd, J = 1.6, 1.0 Hz, 1H), 5.60 (t, J = 1.6 Hz, 1H), 4.43 (t, J = 6.7
3
Hz, 3H), 3.89 (t, J = 5.7 Hz, 3H), 2.97 (t, J = 6.7 Hz, 2H), 2.89 (t, J = 5.7 Hz, 2H), 1.95 (t, J = 1.3 Hz,
H).
3
Synthesis ofactivated acyclovir (compound 2)
15

ACCEPTED MANUSCRIPT
Triethylamine (15 mL, 108 mmol, 8 equiv.) was added dropwise to a solution of acyclovir (3.04 g,
13.5 mmol, 1 equiv.) in dry DMF (400 mL) at 0 °C. p-nitrophenyl chloroformate (6.26 g, 31.1 mmol,
2.3 equiv.) dissolved in dry DCM (20 mL) was added dropwise to the reaction mixture at 0 °C, upon
which the reaction was allowed to heat up to room temperature. The reaction was stirred at room
temperature for two days, the volume was reduced in vacuo, and the solution filtered. The precipitate
was washed with DCM and the product isolated as an off-white solid (2.73 g, 7.01 mmol, 52%)
1
H NMR (400 MHz, CDCl
3
) δ 8.31 (d, J = 9.1 Hz, 2H), 7.87 (s, 1H), 7.53 (d, J = 9.1 Hz, 2H), 6.54 (s,
2
H), 5.40 (s, 2H), 4.33 (s, 2H), 3.77 (s, 2H).
1
3
C NMR (100 MHz, CDCl
3
) δ 156.78, 155.21, 153.96, 152.01, 151.45, 145.18, 137.75, 125.48,
122.60, 116.47, 71.84, 67.80, 66.16.
+
+
HRMS (ESI): calc. [C15
H
16
N
16
6
O
6
7
+H ]:391.0997; found 391.0997, calc. [C15
7
H
16
N
6
O +Na ]:413.0816;
7
+
found 413.0814, calc. [C H N O +K ]:429.0556; found 429.0561
1
5
Synthesis ofACVSIL monomer (compound 3).
A solution of activated acyclovir (2) (1.72 g, 4.21 mmol, 1 equiv.) in dry DMF (120 mL) was added
to a solution of SIL monomer (1) (972 mg, 4.37 mmol, 1 equiv.), DIEA (2.4 mL, 14 mmol, 3 equiv.)
and DMAP (305 mg, 2.5 mmol, 0.6 equiv.) in dry THF (17 mL). After 10 days more DIEA was added
(
1 mL) and after 11 days more DMAP (102 mg) was added. After 13 days the crude was concentrated
in vacuo. The crude was dissolved in DMSO (10 mL) and purified by preparative HPLC.
HPLC:1x25cm, C18 column, Luna 5u 100Å, 250x10mm, phenomenex, MeCN/H O, 254nm,
mL/min, from 5% MeCN to 80% over 20min, 1 min plateu at 80% MeCN, then 80% to 100%
2
7
MeCN over 1 min and 1 min. plateu at 100 % MeCN. Product eluted after 15 min.
1
H NMR (400 MHz, dDMSO) δ 10.72 (s, 1H), 7.82 (s, 1H), 6.56 (s, 2H), 6.04 (s, 1H), 5.70 (s, 1H),
.35 (s, 2H), 4.31 (dt, J = 18.9, 6.2 Hz, 4H), 4.21 – 4.14 (m, 2H), 3.71 – 3.64 (m, 2H), 3.01 (dt, J =
4.8, 6.2 Hz, 4H), 1.88 (s, 3H).
5
1
1
3
C NMR (100 MHz, dDMSO) δ 166.80, 157.27, 154.72, 154.40, 151.87, 138.11, 136.09, 126.61,
116.94, 72.22, 66.90, 66.79, 65.61, 62.61, 36.71, 36.58, 18.43.
+
+
HRMS (ESI): calc. [C17
H
23
N
5
O
7
S
2
+H ]:474.1112; found 474.1109, calc. [C17
H
23
N
5
O
7
S +Na ]:
2
4
96.0931; found 496.0925
Synthesis ofMMTr protected acyclovir ACV-MMTr (compound 4).
The compound was synthesized according to literature procedure.[51, 52] Acyclovir (1.0 g,
4.44 mmol, 1.0 equiv.) was suspended in anhydrous pyridine (25 mL) and the suspension was cooled
to 0ºC. Chlorotrimethylsilane (3.4 mL, 26.6 mmol, 6.0 equiv.) were added dropwise and the
suspension was left to stir for 15 min. Then, DMAP (0.1 g, 0.9 mmol, 0.2 equiv.) and 4-methoxytrityl
chloride (1.65 g, 5.3 mmol, 1.2 equiv.) were added in one scoop. The reaction was left to stir at r.t.
overnight. The reaction mixture is cooled to 0C, water (10 mL) and a 25% NH4OH solution (10 mL)
is added and the mixture is left to stir for 30min. The crude mixture is filtered, and the solvent is
removed under reduced pressure. The crude product is purified via column chromatography over
silica gel with CH
2
2
Cl :MeOH (100:00 to 100:07) and isolated as colorless solid (1.3 g, 2.6 mmol,
59%). The spectral data matches the reported.
Synthesis ofactivated MMTr protected acyclovir (compound 5)
To an ice-cold solution of MMTrACV (4) (0.2 g, 0.4 mmol, 1.0 equiv.) and TEA (0.08 mL, 0.6 mmol,
1
(
.5 equiv.) in anhydrous DCM (3 mL) was added dropwise a solution of bis(4-nitrophenyl) carbonate
183 mg, 0.6 mmol, 1.5 equiv.) in DCM (2 mL). The reaction was left to stir overnight under an
atmosphere of argon. The reaction solution was diluted with DCM, and washed with sat. NaHCO (3
x 20 mL), brine (1 x 20mL), dried over Na SO . The organic phase was evaporated to dryness and the
3
2
4
16

ACCEPTED MANUSCRIPT
crude product was subjected to flash column chromatography over silica gel with DCM:MeOH
1
1
00:00 to 100:04. The product was isolated as faint yellow solid (190mg, 0.29 mmol, 71%). H NMR
(
400 MHz, Chloroform-d) δ 8.23 (d, J = 9.2 Hz, 2H), 7.43 – 7.33 (m, 8H), 7.33 – 7.28 (m, 4H), 7.23 –
13
7
(
1
.06 (m, 4H), 6.72 (d, J = 8.9 Hz, 2H), 4.90 (s, 2H), 4.06 (s, 2H), 3.70 (s, 3H), 3.20 (s, 2H). C NMR
101 MHz, CDCl ) δ 158.24, 155.49, 152.45, 151.88, 151.01, 145.47, 144.90, 137.54, 136.90, 130.42,
29.06, 128.03, 127.96, 127.79, 126.73, 125.45, 121.86, 113.05, 72.08, 70.56, 67.86, 67.10, 55.30.
3
+
HRMS (ESI): [M+H ] calcd. For C35
H
30
N
6
8
O 663.2198, found 663.2220.
Synthesis of MMTr protected acyclovir SIL monomer (compound 6)
To a solution of SIL monomer (1) (81 mg, 0.36 mmol, 1.2 equiv.), TEA (0.06 mL, 0.45 mmol, 1.5
equiv.), and DMAP (41 mg, 0.33 mmol, 1.1 equiv.) in anhydrous DCM (4 mL) was added dropwise a
solution of activated ACV (4) (200 mg, 0.30 mmol, 1.0 equiv.) in DCM (2 mL). The reaction was left
to stir overnight under an atmosphere of argon. The reaction solution was adsorbed onto celite and
subjected to column chromatography over silica gel with DCM:MeOH 100:00 to 100:04. The product
1
was isolated as faint yellow solid (181mg, 0.30 mmol, 80%). H NMR (400 MHz, DMSO-d
6
) δ 10.66
(
s, 1H, NH), 7.74 (s, 1H, NH), 7.70 (s, 1H, Ar-H), 7.37 - 7.25 (m, 8H, mTr-), 7.26 – 7.10 (m, 4H,
mTr), 6.87 (d, J = 8.9 Hz, 2H, mTr), 6.05 (s, 1H, olefin-H), 5.69 (m, 1H, olefin-H), 4.88 (s, 2H, CH2),
.38 - 4.27 (m, 4H, 2 x CH2), 3.93 – 3.76 (m, 2H, CH2), 3.71 (s, 3H, OCH3), 3.14 - 2.97 (m, 6H, 3 x
4
1
3
CH2), 1.88 (d, J = 1.2 Hz, 3H, CH3). C NMR (101 MHz, DMSO) δ 166.37, 157.72, 156.56, 154.23,
51.09, 150.05, 144.84, 137.85, 136.65, 135.66, 129.90, 128.50, 127.67, 126.57, 126.16, 116.54,
1
1
+
12.93, 71.32, 69.57, 66.88, 66.46, 65.15, 62.19, 54.99, 54.96, 36.31, 36.24, 18.00. HRMS [M+H ]
calcd. for C37 746.2313, found 746.2332.
H
39
N
5
O
S
8 2
Polymer syntheses.
PHPMA P1
HPMA (199 mg, 1.39 mmol) and C12
H25S/OH RAFT agent (4-Cyano-4-
[(dodecylsulfanylthiocarbonyl)sulfanyl]pentanoic acid) (3.12 mg, 0.00773 mmol) were dissolved in
DMF (280 μL) and transferred to a glass ampoule. A stock solution of initiator, 2,2'-Azobis(4-
methoxy-2,4-dimethylvaleronitrile) (V-70), (19.4 μL, 12.7 mg/mL) was added to the reaction mixture.
The reaction mixture was degassed by five cycles of freeze-pump-thaw and the ampoule flame-sealed.
1
The polymerization was performed for 24 h at 30 °C. A sample was taken out for crude H NMR, and
the reaction mixture precipitated in ether, filtered
and washed with ether and DCM. The pure product was isolated as a white solid (16 mg).
1
H NMR (400 MHz, DMSO-d6) δ 7.22 (bs, 1H), 4.73 (bs, 1H), 3.67 (bs, 1H), 2.90 (bs, 2H), 1.85 –
0
.49 (m, 9H)
PHPMA-co-ACVP2
25S/OH RAFT agent (7.37 mg, 0.0183 mmol), V-70 (1.14 mg, 0.0037 mmol), HPMA (234 mg,
C H
12
1.63 mmol) and ACV SIL-monomer (40.9 mg, .0864 mmol) were dissolved in DMF (550 μL) and
transferred to a glass ampoule. The reaction mixture was degassed by nine cycles of freeze-pump-
thaw and the ampoule was flame-sealed under vacuum. The polymerization was performed for 24 h at
1
3
0 °C. A sample was taken out for crude H NMR, and the reaction mixture precipitated in DCM,
filtered and washed with DCM. The pure product was isolated as a white solid (154.6 mg).
1
H NMR (400 MHz, dDMSO) δ 7.83 (s, 1H), 7.21 (bs, 1H), 6.53 (s, 2H), 5.37 (s, 2H), 4.72 (bs, 1H),
4.32 (m, 4H) 4.19 (bs, 2H), 3.68 (bs, 1H), 2.91 (bs, 1H), 1.99 – 0.51 (m, 9H).
PHPMA-co-fluorescein methacylate P3
17

ACCEPTED MANUSCRIPT
HPMA (202 mg, 1.41 mmol, 113 equiv.), C12
H
25S/TA RAFT agent (6.34 mg, 0.0126 mmol, 1 equiv.),
V-70 (0.386 mg, 0.00125 mmol, 0.1 equiv.) and fluorescein methacrylate (10.1 mg, 0.0252 mmol, 2
equiv.) were dissolved in DMF (450 µL). The solution was transferred to a glass ampule. The reaction
mixture was degassed by 3 cycles of freeze-pump-thaw and the ampule flame-sealed under vacuum.
1
The reaction was performed at 40 °C for 22 hours. A sample was taken out for crude H NMR, and
the reaction mixture precipitated in DCM, filtered and washed with DCM. The pure product was
isolated as a white solid (49.7 mg).
1
H NMR (400 MHz, dDMSO) δ 10.22 (s, 1H), 8.03 (s, 2H), 7.75 (d, 4H), 7.20 (s, 1H), 6.73 (s, 1H),
6.58 (s, 2H), 4.71 (s, 1H), 3.66 (s, 1H), 2.90 (s, 2H), 1.89 – 0.56 (m, 9H).
DSPE-PHPMA-co-fluorescein P4
HPMA (201 mg, 1.4 mmol, 112 equiv.), C12H25S/DSPE-PEGRAFT agent (39.6 mg, 0.0125 mmol, 1
equiv.), V-70 (0.386 mg, 0.00125 mmol, 0.1 equiv.), and fluorescein methacrylate (10 mg, 0.025
mmol, 2 equiv.) were dissolved in DMF (0.5 mL). The solution was transferred to an ampule and
degassed by 9 cycles of freeze-pump-thaw. The ampule was flame-sealed under vacuum,and the
1
polymerization performed at 40°C for 22 hours. A sample was taken out for crude H NMR, and the
reaction mixture precipitated in DCM, filtered and washed with DCM. The pure product was isolated
as a white solid (127.65 mg).
1
H NMR (400 MHz, dDMSO) δ 10.26 (s, 1H), 8.02 (s, 2H), 7.77 (s, 4H), 7.19 (m, 1H), 6.86 (s, 2H),
6.75 (s, 1H), 6.60 (s, 2H), 4.72 (s, 1H), 3.67 (s, 1H), 3.50 (s), 2.90 (s, 2H), 1.94 – 0.40 (m, 9H)
DSPE-PHPMA-co-AzMA P5
HPMA (100 mg, 0.698 mmol, 90 equiv.), C12H25S/DSPE-PEGRAFT agent (24.7 mg, 0.00778 mmol,
1
equiv.) and V-70 (0.239 mg, 0.000775 mmol, 0.1 equiv.) were dissolved in DMF (265 µL). AzMA
(4.5 mg, 0.0226 mmol, 3.5 equiv., in 10 µL DMF) was added to the solution. The solution was
transferred to a glass ampule and degassed by 9 cycles of freeze-pump-thaw. The ampule was flame-
sealed under vacuum and the polymerization performed at 30 °C for 23.5 hours. A sample was taken
1
out for crude H NMR, and the reaction mixture precipitated in Et
2
O, filtered and washed with Et O.
2
The pure product was isolated as a white solid (43.11 mg).
1
H NMR (400 MHz, D
H).
2
O) δ 4.04 (s), 3.87 (s, 1H), 3.66 (s), 3.35 (s, 2H), 3.08 (s, 2H), 2.02 – 0.59 (m,
8
PHPMA-co-AzMA P6
HPMA (101 mg, 0.705 mmol, 91 equiv.), C12H25S/TA RAFT agent (3.12 mg, 0.00773 mmol, 1
eqiuv.) and V-70 (0.239 mg, 0.000775 mmol, 0.1 equiv.) were dissolved in DMF (265 µL). AzMA
4.5 mg, 0.0226 mmol, 3.4 equiv., in 10 µL DMF) was added to the solution. The solution was
(
transferred to a glass ampule and degassed by 8 cycles of freeze-pump-thaw. The ampule was flame-
sealed under vacuum and the polymerization was performed at 30 °C for 23.5 hours. A sample was
1
taken out for crude H NMR, and the reaction mixture precipitated in Et
2
O, filtered and washed with
Et
2
O. The pure product was isolated as a white solid (31.85 mg).
1
H NMR (400 MHz, D O) δ 3.87 (s, 1H), 3.34 (s, 2H), 3.08 (s, 2H), 2.26 (s, 2H), 2.02 – 0.50 (m, 8H).
2
DSPE-PHPMA-co-ACV P7
25S/DSPE-PEGRAFT agent (31.6 mg, 0.00995 mmol, 1 equiv.), V-70 (0.62 mg, 0.00201 mmol,
C H
12
0.2 equiv.), HPMA (144 mg, 1.01 mmol, 101 equiv.), and ACV-SIL-monomer (23.6 mg, 0.0499
mmol, 5 equiv.) were dissolved in DMF (0.55 mL) and added to a glass ampoule. The reaction
mixture was degassed by 9 cycles of freeze-pump-thaw, and the ampule flame-sealed under vacuum.
1
The polymerization was performed at 30 °C for 24 hours. A sample was taken out for crude H NMR,
and the reaction mixture precipitated in Et
isolated as a white solid (81.66 mg).
2
O, filtered and washed with Et O. The pure product was
2
18

ACCEPTED MANUSCRIPT
1
H NMR (400 MHz, dDMSO) δ 7.84 (s, 1H) , 7.22 (bs, 1H), 6.55 (s, 2H), 5.37 (s, 2H), 4.73
(bs, 1H), 4.31 (m, 4H) , 4.19 (bs, 2H), 3.68 (bs, 1H), 3.51 (s), 2.92 (bs, 2H), 2.13-0.82 (m, 9H).
DSPE-PEG PHPMA-co-ACV P8
25S/DSPE-PEGRAFT agent (35.9 mg, 0.0113 mmol, 1 equiv.), V-70 (0.703 mg, 0.00228 mmol,
C H
12
0.2 equiv.), HPMA (143mg, 0.999 mmol, 88 equiv.), and ACV-SIL-monomer (47.1 mg, 0.0994
mmol, 8.8 equiv.) were dissolved in DMF (750 µL) and added to a glass ampoule. The reaction
mixture was degassed by 15 cycles of freeze-pump-thaw. The ampule was flame-sealed under
vacuum and the polymerization performed at 30 °C for 24 hours. A sample was taken out for crude H
1
NMR, and the reaction mixture precipitated in Et
was isolated as a white solid (81.72 mg)
2
O, filtered and washed with Et O. The pure product
2
1
H NMR (400 MHz, dDMSO) δ 7.83 (s, 1H), 7.26 (s, 1H), 6.56 (s, 2H), 5.37 (s, 2H), 4.73
s, 1H), 4.42 – 4.03 (m, 6H), 3.69 (s, 1H), 3.51 (s), 3.11 – 2.75 (m, 2H), 1.97 – 0.41 (m, 9H).
(
TA PHPMA-co-ACV P9
25S/TA RAFT agent (5.76 mg, 0.0114 mmol, 1 equiv.), V-70 (0.697 mg, 0.00226 mmol, 0.2
C H
12
equiv.), HPMA (141 mg, 0.985 mmol, 88 equiv.), and ACV-SIL-monomer (46.7 mg, 0.0986 mmol,
.8 equiv.) were dissolved in DMF (750 µL) and added to a glass ampoule. The reaction mixture was
8
degassed by 19 cycles of freeze-pump-thaw. The ampule was flame-sealed under vacuum and the
polymerization was performed at 30 °C for 24 hours. Crude 1H-NMR
1
showed conversion of 40%. A sample was taken out for crude H NMR, and the reaction mixture
precipitated in Et
47.49 mg).
2
O, filtered and washed with Et O. The pure product was isolated as a white solid
2
(
1
H NMR (400 MHz, dDMSO-d6) δ 10.66 (s), 7.83 (s, 1H), 7.22 (s, 1H), 6.54 (m, 2H), 5.34 (m, 2H),
.71 (s, 1H), 3.71 (s, 1H), 2.94 (s, 2H), 2.08 – 0.47 (m, 9H).
4
DSPE-PHPMA-co-Cy7
DSPE-PEGPHPMA-co-AzMA (P5) (9.39 mg, 1 equiv.) was dissolved in 500 uL mili Q. 100 µL of a
stock solution containing CuBr (1.07 mg/mL), THPTA (6.48 mg/mL) and sodium ascorbate (29.6
mg/mL) in PBS was added followed by Cy7 alkyne (0.828 mg, 1.2 equiv.) dissolved in DMSO (60
µL). The reaction was left at room temperature overnight and purified by NAP25 column, the eluate
was freeze dried giving the pure Cy7 labeled polymer (7.87 mg)
PHPMA-co-Cy7
Was synthesized as previously described.[13] Murine serum albumin (10.6 mg, 0.2 equiv.) was
dissolved in 0.5 mL PBS buffer. TA-PHPMA-co-AzMA (5.56 mg, 1 equiv.) dissolved in 0.35 mL
PBS buffer was added and the reaction allowed to react at room temperature for 30 min. Cy7 alkyne
(0.552 mg, 1.1 equiv.) in 50 uL DMSO was added followed by the addition of 100 µL of stock
solution containing CuBr (1.07 mg/mL), THPTA (6.48 mg/mL) and sodium ascorbate (29.6 mg/mL)
in PBS. The reaction was allowed to react overnight at room temperature. The reaction mixture was
transferred to two 30 kDa spin filters and concentrated for 30 min at 10000 rpm on a centrifuge. 0.4
mL milliQ was added to each spin filters and the procedure repeated 4 times. The sample was diluted
with water, transfered to a dialysis tube with MWCO 14 kDa and dialyzed against pure water at room
temperature for 2x3 hours and 1 x 60 hours at 4 °C. The solution was lyophilized to obtain the Cy7
labeled protein-polymer conjugate (12.58 mg).
DSPE-PHPMA-co-ACV P10
19

ACCEPTED MANUSCRIPT
ACV MMTr SIL (5) (83 mg, 0.111 mmol, 10 equiv.), HPMA (143 mg, 1.0 mmol, 90 equiv.),
C
6
H
5
/DSPE-PEGRAFT agent (34 mg, 0.011 mmol, 1 equiv.)) and AIBN (0.37 mg, 0.002 mmol, 0.2
equiv.) were dissolved in 1.5 mL DMF, and the reaction solution was transferred into glass ampoules.
rounds of freeze-pump-thaw were performed prior to sealing the ampoules under vacuum. The
polymerization was performed at 60°C for 18-20 h. The reaction mixture was precipitated into
4
Et
2
O/acetone (50:50 v/v%) yielding a slightly red solid (75 mg)
1
H NMR (400 MHz, DMSO-d6) δ 10.65 (s, 1H, ACV-H), 7.71 (d, J = 12.4 Hz, 2H, ACV-H), 7.45 – 6.93 (m,
2H, ACV-H, 1H, PHPMA), 6.87 (d, J = 7.7 Hz, 2H, ACV-H), 4.87 (s, 2H, ACV-H), 4.71 (s, 1H, HPMA), 4.31
s, 4H, ACV-H), 4.11 (bs, 2H, ACV-H), 3.84 (s, 2H, ACV-H), 3.76 – 3.57 (m, 3H, ACV-H + 1H, HPMA), 3.50
s, PEG), 3.14 – 2.66 (m, 6H, ACV-H + 2H PHPMA), 1.38 – 0.38 (polymer backbone).
1
(
(
The MMTr protected polymer was suspended in Et
mL TFA was added dropwise to the suspension. The suspension turns yellow after addition of TFA.
The suspension was left to stir overnight. The suspension turned into a yellow solution. The polymer
2
O (1 mL), the suspension was cooled to 0°C and
1
was purified via precipitation into a cooled mixture of acetone/Et
a off red solid (53 mg)
2
O (50/50 v/v%) and dried, yielding
1
H NMR (400 MHz, DMSO-d6) δ 10.67 (s, 1H, ACH-H), 7.85 (s, 1H, ACV-H), 7.55 (bs, 1H, PHMPA), 6.54 (s,
H, ACV-H), 5.36 (s, 2H, ACV-H), 5.06 (s, 1H, PHPMA), 4.43 – 3.99 (m, 4H, ACV-H), 3.67 (s, 1H, PHPMA),
2
3.50 (s, PEG), 3.40 – 2.80 (m, 6H, ACV-H + 2H, PHPMA), 1.41 – 0.40 (polymer backbone).
DSPE-PHPMA-co-ACV P11
ACV MMTr SIL (5) (39 mg, 0.111 mmol, 5 equiv.), HPMA (143 mg, 1.0 mmol, 95 equiv.),
C
6
H
5
/DSPE-PEGRAFT agent (32 mg, 0.011 mmol, 1 equiv.)) and AIBN (0.35 mg, 0.002 mmol, 0.2
equiv.) were dissolved in 1.5 mL DMF, and the reaction solution was transferred into glass ampoules.
rounds of freeze-pump-thaw were performed prior to sealing the ampoules under vacuum. The
polymerization was performed at 60°C for 18-20 h. The reaction mixture was precipitated into
4
Et
2
O/acetone (50:50 v/v%) yielding an off red solid (64 mg)
1
H NMR (400 MHz, DMSO-d6) δ 10.61 (s, 1H, ACV-H), 7.71 (m, 2H, ACV-H), 7.52 – 6.97 (m, 12H, ACV-H,
1
H, PHPMA), 6.88 (d, J = 8.2 Hz, 2H, ACV-H), 4.87 (s, 2H, ACV-H), 4.72 (s, 1H, PHPMA), 4.32 (s, 4H,
ACV-H), 4.12 (bs, 2H, ACV-H), 3.85 (s, 2H, ACV-H), 3.80 – 3.60 (m, 3H, ACV-H, 1H, PHPMA), 3.51 (s,
PEG), 2.97 (bs, 6H, ACV-H, 2H, PHPMA), 2.03 – 0.49 (polymer backbone).
The MMTr protected polymer was suspended in Et
mL TFA was added dropwise to the suspension. The suspension turns yellow after addition of TFA.
The suspension was left to stir overnight. The suspension turned into a yellow solution. The polymer
2
O (1 mL), the suspension was cooled to 0°C and
1
was purified via precipitation into a cooled mixture of acetone/Et O (50/50 v/v%) and dried, yielding
2
an off red precipitate (43 mg)
1
H NMR (400 MHz, DMSO-d6) δ 10.66 (s, 1H, ACV-H), 7.84 (s, 1H, ACV-H), 7.38 – 7.03 (m, 1H, PHPMA),
.54 (s, 2H, ACV-H), 5.37 (s, 2H, ACV-H), 4.71 (s, 1H, PHPMA), 4.31 (s, 2H, ACV-H), 4.19 (s, 2H, ACV-H),
.11 (s, 2H), 3.70 (s, 1H, PHPMA), 3.50 (s, PEG), 3.08 – 2.75 (m, 6H, ACV-H + 2H, PHPMA), 1.69 – 0.40
polymer backbone).
6
4
(
20

ACCEPTED MANUSCRIPT
Albumin competition
The competition assay was performed in a 96-well plate format and fluorescence analyzed (excitation
wavelength 360 nm, emission wavelength 465 nm) on an Enspire 2300 Multilabel Reader (Perkin
Elmer®). Polymers P1 and P4 were dissolved in 0.01 M PBS, pH 7.4 and diluted to obtain two stock
solutions. Stock solutions of human serum albumin (HSA) and fluorescent probes (dansyl-L-
asparagine/dansylsarcosine) were prepared in 0.01 M PBS, pH 7.4. The components were mixed and
PBS added to give a total volume of 100 µL in each well, with the following concentrations: HSA 5
µM, dansylsarcosine 5 µM or dansyl-L-asparagine 15 µM. Polymer concentrations were varied to
study the range from 0.1 to 100 equivalents of polymer relative to albumin. Upon mixing the plates
were incubated at 37 ° C for 1 hour to allow the system to equilibrate. The data represents three
independent experiments, which each included three technical replicates.
HPLC release studies
Polymer stock solution (RWACV2d / P12) in DMSO (4 g/L) was diluted to a final concentration of
1
g/L in PBS (pH 7.4, 10 mM) and incubated at 37°C for 3 h in presence or absence of 5.0 mM
glutathione. The polymer was removed via spin filtration (MWCO = 3k) and the filtrate was analyzed
via HPLC. The AUC (254nm) corresponding to ACV was converted in molar concentration via a
standard curve.
In vivo – biodistribution
For evaluation of the biodistribution and blood residence time we used female BALB/cJRjmice (8
weeks old, Janvier-labs.com, France). We performed subcutaneous and intraveneous injections in
different experiments, but with identical protocols, except for the route of injection.
12 mice received injection either i.v. injection in the tail vein or s.c. injection in the scruff of the neck
with a fluorescent molecule or PBS: DSPE-HPMA-Cy7 (2 mg/kg), HPMA-Cy7 (0.4 mg/kg),
PHPMA-Cy7-albumin (4 mg/kg) or vehicle (PBS). Blood samples were drawn for 4 mice from each
group at time points up to 7 days post injection, either from tail tip or retroorbital plexus into 20 µL
heparin-coated glass capillaries (Vitrex Medical AS, Denmark) and centrifuged briefly to separate
plasma and cells. Mice were anaesthetized with 3.75% isoflurane at the same time points and scanned
with in vivo imaging system (IVIS, PerkinElmer, MA, USA). Images were acquired using 745/820
nm filterset for Cy7 and analyzed with Living Image software v4.3 (PerkinElmer). After subtracting
background fluorescence signal from vehicle, the fraction of compound in the plasma was estimated
based on the fluorescence intensity in the plasma. The amount of compound in plasms is expressed as
2
the mean radiance (photons/sec/cm /sr) relative to the maximum fluorescence of the specified
injection route for the specific compound. 4 mice were terminated under anesthesia at 1 hour, 24
hours and 7 days after injection by PBS perfusion through the heart to displace blood from organs.
Organs were then removed and scanned with IVIS.
In Vitro antiviral studies.
Effect of polymers on HSV-2 infection using ELVIS cells
The generation of HSV-2 strain 333 eGFP virus stocks was done by infecting ELVIS cells. [37] 48 h
later collected culture supernatant containing progeny virus was centrifuged for 3 min at 330×g to
®
remove cell debris. Virus stocks were stored at -80°C. ELVIS cells were cultured in Dulbecco’s
Modified Essential Medium (DMEM) (Lonza, Basel, Switzerland) supplemented with 100 units/ml
penicillin, 120 μg/ml streptomycin, 2 mM glutamine (Invitrogen, Glostrup, Denmark) and 10% heat-
®
inactivated fetal calf serum (FCS) (Life Technologies, Carlsbad, USA). ELVIS cells are genetically
engineered baby hamster kidney cells. The DNA harbors a promoter gene that is chimerically linked
21

ACCEPTED MANUSCRIPT
to an Escherichia coli LacZ reporter gene. Upon infection the promotor is induced by HSV and β-
galactosidase is expressed.
®
5
,000 ELVIS cells were seeded one day prior infection in 96 well plates (Cell culture 96-well plates,
TC-treated (F), Sarstedt, Nürmbrecht, GER). Afterwards, the cells were incubated with triplicates of
polymer concentration ranging from 0-100 µg/ml for 1 h at 37°C. Subsequently the cells were
infected with HSV-2 eGFP. As controls, Heparin with the same concentration as the polymers and
acyclovir with concentrations ranging from 0-1000 µM were used. Infection rates were determined
48 h post virus inoculation by quantifying β-galactosidase activity in cellular lysates using the Gal-
Screen™ β-Galactosidase Reporter Gene Assay System for Mammalian Cells (Thermo Fisher
Scientific) and the Orion II microplate luminometer (Titertek-Berthold, Pforzheim, Germany).
Reporter gene activities were obtained relative light units per second (RLU/s). Values derived from
triplicate infection and background activities derived from uninfected cells were subtracted. Percent
infection rates relative to the untreated infected control (100%) are shown.
Effect of polymers on HSV-2 infection using HeLa cells
®
HSV-2 strain 333 eGFP was produced by infecting HeLa (ATCC CCL-2™) cells and harvesting
supernatant 96 hours later. For the removal of cell debris the collected culture supernatant was
centrifuged for 3 min at 330×g. Virus stocks were stored at -80°C. HeLa cells were cultured in
DMEM (Lonza, Basel, Switzerland) containing 10% heat-inactivated fetal calf serum (FCS), 100
units/ml penicillin and 120 μg/ml streptomycin (Invitrogen, Glostrup, Denmark).
10,000 HeLa cells were seeded one day prior infection in black 96 well plates (Costar® 96-well cell
culture plates, black with clear flat bottom, TC-treated, Corning, New York, USA). Afterwards, the
cells were incubated with triplicates of the drugs, heparin or acyclovir for 1 h at 37°C.Thereafter cells
were infected, 96 h post infection media was removed and cells fixed in 4% PFA solution for 10 min
at 4°C. Subsequently, cells were washed twice with PBS and afterwards nucleus was stained with
Hoechst 33342 (ThermoFisher Scientific) (1:2,000 (v/v)) for 45 min. Cells were stored in PBS, kept
TM
in the dark and left at 4°C until microscopy image acquisition was performed using a BioTek
Cytation 3 Cell Imaging Multi-Mode Reader (BioTek, Winooski, USA) and Gen 5 software (BioTek,
Winooski, USA) for image processing.
®
Cell Viability. To access cellviability, applied cell lines (10,000 HeLa cells, 5,000 ELVIS cells)
were seeded into 96-well plates and incubated with the highest polymers concentrations used in the
®
respective assays. After 48 h for ELVIS cells and 96 h for HeLa cells of incubation, the medium was
discarded and 100 μl of 0.5 mg/ml MTT-PBS (3-[4,5-dimethyl-2-thiazolyl]-2,5-diphenyl-2H-
tetrazolium bromide) solution was added to the cells. The yellow MTT is reduced by NAD(P)H-
dependent oxidoreductase system of living cells causing the formation of insoluble purple formazan
crystals. 2,5 h of incubation at 37°C the cell-free supernatant was discarded and formazan crystals
were dissolved in 100 μl DMSO:Ethanol (1:2). Absorption was detected at 490 nm and corrected by
the background absorption at 650 nm by Orion II microplate luminometer (Titertek-Berthold,
Pforzheim, Germany).
In Vivo anti-HSV-2 studies.
Genital HSV-2 infection in mice
The mice were pretreated by subcutaneous (s.c.) injection of 2 mg of Depo-Provera (Pfizer). Five
days later mice were anesthetized with Isoflurane (Abbott) and inoculated intravaginally with 20 μL
HSV-2 (6.7 × 104 p.f.u.) suspended in IMDM. Mice were then placed on their backs and maintained
under anesthesia for 10 min. Five hours post infection, mice were treated s.c. with the compounds.
22

ACCEPTED MANUSCRIPT
Free acyclovir was administered as a suspension in PBS, 60 mg/kg mouse. Polymers were
administered such that the molar amount of acyclovir was identical in each formulation (60 mg
ACV/kg mouse). Polymer solutions were administered as premixed solutions with equimolar amounts
of polymer and murine serum albumin, both for the DSPE-HPMA-ACV polymer and the HPMA-
ACV polymer. Vaginal washes were collected 48 h post infection by washing with 2 × 40 μL of
IMDM and diluted to a final volume of 250 μL. Vaginal washes were collected from mice infected
intravaginally with HSV-2 or uninfected (UN) mice. The washes were immediately placed on dry ice
and then stored at −70 °C until analyzed. On day seven post infection, mice were euthanized by
cervical dislocation, and Medulla spinalis were isolated and homogenized for viral titer determination.
Infected mice were observed and examined daily. At this time, they were scored for vaginal
inflammation, neurological illness, and death. The severity of disease was scored according to the
following scale: 0, healthy; 1, genital erythema; 2, moderate genital inflammation; 3, purulent genital
lesion/or generally bad conditions; 4, hindlimb paralysis or generally very poor condition. Mice were
sacrificed by cervical dislocation at score exceeding 3.
Virus plaque assay
Vaginal washes were collected from mice infected intravaginally with HSV-2 or not (UN) and
immediately placed on dry ice and then stored at −70 °C until used for analysis. Viral titers of vaginal
washes were determined on monolayers of Vero cells seeded in MEM supplemented with 5% FCS at
6
a density of 1.2 × 10 cells per well and left overnight to settle. The next day the cells were infected
with 100 μL serially diluted vaginal washes. The cells were incubated for 1 h, then 5 mL MEM
supplemented with 0.2% human immunoglobulin (ZLB Behring GmBH) was added. The plates were
further incubated for 2–3 d and stained with 0.03% methylene blue in order to count the plaques. The
results were expressed as p.f.u. per vaginal wash.
Ethics statement
All animal work was performed according to national and international recognized guidelines and the
animal experimental protocols were approved by The Experimental Animal Inspectorate in Denmark
under The Danish Veterinary and Food Administration, Ministry of Food, Agriculture and Fisheries
(License file number 2014-15-0201-00377). All mice were housed in certified animal facilities in a
12- hours dark:light cycle, with temperature 22 ± 2°C and humidity of 55 ± 10 %. The animals had
free accessibility to water and standard chow and prior to treatment the animals were housed 5 per
cage in 7 type III plastic cages (Techniplast, Italy) with wooden asp bedding (Finn Tapvei, Finland)
enriched with hide shelters and gnawing softwood sticks. Experienced animal caretakers performed
daily welfare monitoring and bodyweight measurement.
HSV infectivity animal studies were reviewed and approved by Dyreforsøgstilsynet under the Danish
Ministry for Veterinary and Food Administration (permission#: 2015-15-0201-00686). The study was
carried out in strict accordance with the recommendations in the Guide for the Care and Use of
Laboratory Animals, EEC Council Directive 2010/63/EU.
23

ACCEPTED MANUSCRIPT
REFERENCES
1.
Beck, A., et al., Strategies and challenges forthe nextgeneration of antibody–drug
conjugates. Nature Reviews DrugDiscovery, 2017. 16: p. 315.
2
.
Zelikin, A.N., C. Ehrhardt, and A.M. Healy, Materials and methods fordelivery of biological
drugs. Nature Chemistry, 2016. 8: p. 997.
Mullard, A., Maturing antibody–drug conjugatepipelinehits 30. Nature Reviews Drug
Discovery, 2013. 12: p. 329.
3.
4.
Hoogenboezem, E.N. and C.L. Duvall, Harnessing albumin as a carrier for cancertherapies.
Advanced DrugDelivery Reviews, 2018. 130: p. 73-89.
5.
Yu, M., et al., Battle of GLP-1 delivery technologies. Advanced DrugDelivery Reviews, 2018.
130: p. 113-130.
6
.
Lau, J., et al., Discovery of the Once-Weekly Glucagon-LikePeptide-1(GLP-1) Analogue
Semaglutide. Journal of Medicinal Chemistry, 2015. 58(18): p. 7370-7380.
Ramírez-Andersen, H.S., etal., Long-Acting Human Growth HormoneAnalogueby
NoncovalentAlbumin Binding. Bioconjugate Chemistry, 2018.
7.
8.
Sockolosky, J.T. and F.C. Szoka, The neonatalFcreceptor, FcRn, as a targetfor drug delivery
and therapy. Adv DrugDeliv Rev, 2015. 91: p. 109-24.
9.
Renoux, B., etal., Targeting the tumourmicroenvironmentwith an enzyme-responsivedrug
delivery system forthe efficienttherapy of breastand pancreaticcancers. Chemical Science,
2
017. 8(5): p. 3427-3433.
1
0.
1.
2.
3.
4.
5.
Stehle, G., etal., The loading ratedetermines tumortargeting properties of methotrexate-
albumin conjugates in rats. Anti-CancerDrugs, 1997. 8(7): p. 677-685.
Riber, C.F. and A.N. Zelikin, Recentadvances in macromolecularprodrugs. CurrentOpinion in
Colloid &Interface Science, 2017. 31: p. 1-9.
Andersen, A.H.F., etal., Long-Acting, PotentDelivery of Combination AntiretroviralTherapy.
ACS Macro Letters, 2018. 7(5): p. 587-591.
Smith, A.A.A., etal., Albumin–Polymer–Drug Conjugates:Long Circulating, High Payload
Drug Delivery Vehicles. ACS Macro Letters, 2016. 5(10): p. 1089-1094.
Jiang, Y. and M. Stenzel, Drug Delivery Vehicles Based on Albumin–PolymerConjugates.
MacromolecularBioscience, 2016. 16(6): p. 791-802.
Zuwala, K., etal., Macromolecularprodrugs of ribavirin:Polymerbackbonedefines blood
safety, drug release, and efficacy of anti-inflammatory effects. Journal of Controlled Release,
1
1
1
1
1
2
018. 275: p. 53-66.
1
6.
7.
8.
Riber, C.F., etal., SyntheticPolymerwith a Structure-Driven HepaticDeposition and Curative
PharmacologicalActivity in Hepatic Cells. ACS Macro Letters, 2017. 6(9): p. 935-940.
Bienk, K., etal., An albumin-mediated cholesteroldesign-basedstrategy fortuning siRNA
pharmacokinetics and genesilencing. Journal of Controlled Release, 2016. 232: p. 143-151.
Müller, C., etal., Synthesis, Radiolabeling, and Characterizationof Plasma Protein-Binding
Ligands:PotentialTools for Modulation of thePharmacokineticProperties of
1
1
(
Radio)Pharmaceuticals. Bioconjugate Chemistry, 2017. 28(9): p. 2372-2383.
1
9.
0.
1.
2.
3.
Dumelin, C.E., etal., A PortableAlbumin Binderfrom a DNA-Encoded ChemicalLibrary.
Angewandte ChemieInternationalEdition, 2008. 47(17): p. 3196-3201.
Purdie, L., etal., Alkyl-Modified Oligonucleotides as Intercalating Vehicles forDoxorubicin
Uptakevia Albumin Binding. MolecularPharmaceutics, 2018. 15(2): p. 437-446.
Kopecek, J. and P. Kopeckova, HPMA copolymers:origins, early developments, present,and
future. Adv Drug Deliv Rev, 2010. 62(2): p. 122-49.
Liu, H., et al., Structure-based programming of lymph-nodetargeting in molecular vaccines.
Nature, 2014. 507: p. 519.
Moad, G., E. Rizzardo, and S.H. Thang, RAFT Polymerization and Someof its Applications.
Chemistry –An Asian Journal, 2013. 8(8): p. 1634-1644.
2
2
2
2
24

ACCEPTED MANUSCRIPT
24.
Fox, M.E., F.C. Szoka, and J.M.J. Frechet, SolublePolymerCarriers forthe Treatmentof
Cancer:The Importanceof MolecularArchitecture. Accounts of Chemical Research, 2009.
4
2(8): p. 1141-1151.
25.
Caravan, P., etal., The Interaction of MS-325 with Human Serum Albumin and Its Effect on
Proton Relaxation Rates. Journal of the American Chemical Society, 2002. 124(12): p. 3152-
3
162.
2
6.
Sudlow, G., D.J. Birkett, and D.N. Wade, Further Characterization of SpecificDrug Binding
Sites on Human Serum Albumin. MolecularPharmacology, 1976. 12(6): p. 1052-1061.
SJÖHOLM, I., et al., Binding of Drugs to Human Serum Albumin. XI. The Specificity of Three
BindingSites as Studied with Albumin Immobilized in Microparticles, 1979. 16(3): p. 767-
27.
7
77.
2
8.
9.
0.
Brigden, D., etal., Acyclovir - a review of thepreclinical and early clinical data of a new
antiherpes drug. Antiviral Res, 1981. 1(4): p. 203-12.
De Clercq, E. and G. Li, Approved AntiviralDrugs overthePast50 Years. Clin Microbiol Rev,
2
2
016. 29(3): p. 695-747.
3
Li, X., etal., Novel designed polymer-acyclovirconjugates with linker-controlled drug release
and hepatoma cell targeting. Journal of PolymerScience Parta-PolymerChemistry, 2008.
4
6(1): p. 117-126.
31.
Smith, A.A.A., etal., MacromolecularProdrugs of Ribavirin:Concerted Efforts of theCarrier
and the Drug. Advanced Healthcare Materials, 2014. 3(9): p. 1404-1407.
32.
Kock, A., et al., Disulfide reshuffling triggers therelease of a thiol-freeanti-HIV agentto make
up fast-acting, potentmacromolecularprodrugs. Chemical Communications, 2014. 50(93): p.
1
4498-14500.
3
3.
4.
5.
Ruiz-Sanchis, P., etal., Highly ActiveMacromolecularProdrugs InhibitExpression of the
Hepatitis C Virus Genomein theHost Cells. Advanced Healthcare Materials, 2015. 4(1).
Lee, M.H., etal., Disulfide-Cleavage-Triggered Chemosensors and TheirBiological
Applications. Chemical Reviews, 2013. 113(7): p. 5071-5109.
Riber, C.F., A.A.A. Smith, and A.N. Zelikin, Self-ImmolativeLinkers Literally Bridge Disulfide
Chemistry and theRealm of Thiol-FreeDrugs. Advanced Healthcare Materials, 2015. 4(12): p.
3
3
1
887-1890.
3
6.
Zuwala, K., etal., HIV anti-latency treatmentmediated by macromolecularprodrugs of
histonedeacetylaseinhibitor, panobinostat. Chemical Science, 2016. 7(3): p. 2353-2358.
Proffitt, M.R. and S.A. Schindler, Rapid detection of HSV with an enzyme-linked virus
inducible system (ELVIS) employing a genetically modified cell line. Clin Diagn Virol, 1995.
37.
4
(2): p. 175-82.
38.
Stabell, E.C. and P.D. Olivo, Isolation of a cell line forrapid and sensitive histochemicalassay
forthe detection of herpes simplex virus. Journal of VirologicalMethods, 1992. 38(2): p. 195-
204.
3
9.
0.
1.
2.
3.
4.
Lehman, I.R. and P.E. Boehmer, Replication of Herpes Simplex Virus DNA. Journal of
Biological Chemistry, 1999. 274(40): p. 28059-28062.
Weir, J.P., Regulation of herpes simplex virus gene expression. Gene, 2001. 271(2): p. 117-
4
1
30.
4
Wachsman, M.B., et al., Enterocin CRL35 inhibits late stages of HSV-1and HSV-2 replication
in vitro. Antiviral Research, 2003. 58(1): p. 17-24.
4
Elion, G.B., Mechanism of action and selectivity of acyclovir. The American Journal of
Medicine, 1982. 73(1): p. 7-13.
Shieh, M.T., etal., Cell surfacereceptors for herpes simplex virus are heparan sulfate
proteoglycans. The Journal of Cell Biology, 1992. 116(5): p. 1273-1281.
Elion, G.B., etal., Selectivity of action of an antiherpeticagent, 9-(2-
hydroxyethoxymethyl)guanine. Proceedings of the NationalAcademy of Sciences, 1977.
4
4
74(12): p. 5716-5720.
25

ACCEPTED MANUSCRIPT
4
5.
6.
7.
Buthod, L.M., N.L. Schindler, and H.W. Rogers, Determination of cell line suitability for rapid
isolation of herpes simplex virus. Journal of Virological Methods, 1987. 18(1): p. 37-46.
Taylor, J.M., etal., AlternativeEntry Receptors forHerpes Simplex Virus and Their Roles in
Disease. Cell Host& Microbe, 2007. 2(1): p. 19-28.
4
4
Landovitz, R.J., R. Kofron, and M. McCauley, Thepromise and pitfalls of long-acting
injectableagents forHIV prevention. CurrentOpinion in Hiv and Aids, 2016. 11(1): p. 122-
128.
48.
Edagwa, B., et al., Long-acting slow effectivereleaseantiretroviraltherapy. ExpertOpinion
on Drug Delivery, 2017. 14(11): p. 1281-1291.
49.
Sutton, D. and M.R. Boyd, ComparativeActivity of Penciclovirand Acyclovir in Mice Infected
Intraperitoneally with Herpes-Simplex Virus Type-1Sc16. AntimicrobialAgents and
Chemotherapy, 1993. 37(4): p. 642-645.
50.
Ebbesen, M.F., etal., Synthesis of click‐reactiveHPMA copolymers using RAFTpolymerization
fordrug delivery applications. Journal of PolymerScience PartA:PolymerChemistry, 2013.
5
1(23): p. 5091-5099.
51.
Martin, J.C., etal., Synthesis and anti-herpes virus activity of acyclic 2'-deoxyguanosine
analogs related to 9-[(1,3-dihydroxy-2-propoxy)methyl]guanine. Journal of Medicinal
Chemistry, 1986. 29(8): p. 1384-1389.
52.
Beadle, J.R., etal., Synthesis and AntiviralEvaluation of 1-O-Hexadecylpropanediol-3-P-
acyclovir:Efficacy AgainstHSV-1Infection in Mice. Nucleosides, Nucleotides and Nucleic
Acids, 2000. 19(1-2): p. 471-479.
26

ACCEPTED MANUSCRIPT
Graphical abstract
27