Multivalent benzoboroxole functionalized polymers as gp120 glycan targeted microbicide entry inhibitors

Article abstract of DOI:10.1021/mp900159n

Microbicides are women-controlled prophylactics for sexually transmitted infections. The most important class of microbicides target HIV-1 and contain antiviral agents formulated for topical vaginal delivery. Identification of new viral entry inhibitors that target the HIV-1 envelope is important because they can inactivate HIV-1 in the vaginal lumen before virions can come in contact with CD4+ cells in the vaginal mucosa. Carbohydrate binding agents (CBAs) demonstrate the ability to act as entry inhibitors due to their ability to bind to glycans and prevent gp120 binding to CD4+ cells. However, as proteins they present significant challenges in regard to economical production and formulation for resource-poor environments. We have synthesized water-soluble polymer CBAs that contain multiple benzoboroxole moieties. A benzoboroxole-functionalized monomer was synthesized and incorporated into linear oligomers with 2-hydroxypropylmethacrylamide (HPMAm) at different feed ratios using free radical polymerization. The benzoboroxole small molecule analogue demonstrated weak affinity for HIV-1BaL gp120 by SPR; however, the 25 mol % functionalized benzoboroxole oligomer demonstrated a 10-fold decrease in the K_D for gp120, suggesting an increased avidity for the multivalent polymer construct. High molecular weight polymers functionalized with 25, 50, and 75 mol % benzoboroxole were synthesized and tested for their ability to neutralize HIV-1 entry for two HIV-1 clades and both R5 and X4 coreceptor tropism. All three polymers demonstrated activity against all viral strains tested with EC₅₀s that decrease from 15000 nM (1500 μg mL ^-1) for the 25 mol % functionalized polymers to 11 nM (1 μg mL^-1) for the 75 mol % benzoboroxole-functionalized polymers. These polymers exhibited minimal cytotoxicity after 24 h exposure to a human vaginal cell line.

Full text of DOI:10.1021/mp900159n

articles
Multivalent Benzoboroxole Functionalized Polymers as
gp120 Glycan Targeted Microbicide Entry Inhibitors
†
‡
§
†
Julie I. Jay, Bonnie E. Lai, David G. Myszka, Alamelu Mahalingam,
|
‡
,|
Kris Langheinrich, David F. Katz, and Patrick F. Kiser*
Departments of Pharmaceutics and Pharmaceutical Chemistry and Bioengineering and the
Center for Biomolecular Interaction Analysis, UniVersity of Utah, Salt Lake City,
Utah 84112-5820, and Department of Biomedical Engineering, Duke UniVersity,
Durham, North Carolina 27708-0281
Received June 25, 2009; Revised Manuscript Received October 25, 2009; Accepted
November 23, 2009
Abstract: Microbicides are women-controlled prophylactics for sexually transmitted infections. The
most important class of microbicides target HIV-1 and contain antiviral agents formulated for topical
vaginal delivery. Identification of new viral entry inhibitors that target the HIV-1 envelope is important
because they can inactivate HIV-1 in the vaginal lumen before virions can come in contact with
CD4+ cells in the vaginal mucosa. Carbohydrate binding agents (CBAs) demonstrate the ability to
act as entry inhibitors due to their ability to bind to glycans and prevent gp120 binding to CD4+
cells. However, as proteins they present significant challenges in regard to economical production
and formulation for resource-poor environments. We have synthesized water-soluble polymer CBAs
that contain multiple benzoboroxole moieties. A benzoboroxole-functionalized monomer was
synthesized and incorporated into linear oligomers with 2-hydroxypropylmethacrylamide (HPMAm)
at different feed ratios using free radical polymerization. The benzoboroxole small molecule analogue
demonstrated weak affinity for HIV-1BaL gp120 by SPR; however, the 25 mol % functionalized
D
benzoboroxole oligomer demonstrated a 10-fold decrease in the K for gp120, suggesting an
increased avidity for the multivalent polymer construct. High molecular weight polymers functionalized
with 25, 50, and 75 mol % benzoboroxole were synthesized and tested for their ability to neutralize
HIV-1 entry for two HIV-1 clades and both R5 and X4 coreceptor tropism. All three polymers
demonstrated activity against all viral strains tested with EC50s that decrease from 15000 nM (1500
-1
-1
µg mL ) for the 25 mol % functionalized polymers to 11 nM (1 µg mL ) for the 75 mol %
benzoboroxole-functionalized polymers. These polymers exhibited minimal cytotoxicity after 24 h
exposure to a human vaginal cell line.
Keywords: Synthetic carbohydrate binding agent; gp120; benzoboroxole; multivalency; entry
inhibitor, HIV-1
2
Introduction
Rising rates of HIV infection among women, especially
in the pandemic regions of sub-Saharan Africa, combined
with recent HIV-1 vaccine failures have motivated the
development of women-controlled prophylactics, or micro-
3
1
bicides, for preventing HIV-1 infection. Microbicides
contain anti-HIV-1 or other antiviral agents formulated for
topical delivery in the vaginal lumen. However, successful
inactivation of HIV-1 in the vaginal lumen has been hindered
by the limited number of microbicide candidates that target
the HIV-1 envelope in comparison to antiretroviral agents
that are delivered to the cervicovaginal tissue such as
coreceptor antagonists or reverse transcriptase inhibitors.
The viral envelope protein plays a critical role in HIV-1 entry
into cells, with the surface unit, gp120, initiating the binding
*
To whom correspondence should be addressed. Mailing address:
University of Utah, Department of Bioengineering 20 S. 2030
E., Rm. 108, Salt Lake City, UT 84112-9458. Tel: (801) 505-
6
881. Fax: (801) 585-5151. E-mail: patrick.kiser@utah.edu.
†
Department of Pharmaceutics and Pharmaceutical Chemistry,
University of Utah.
Duke University.
Center for Biomolecular Interaction Analysis, University of Utah.
Department of Bioengineering, University of Utah.
4
‡
§
|
116 MOLECULAR PHARMACEUTICS VOL. 7, NO. 1, 116–129
10.1021/mp900159n  2010 American Chemical Society
Published on Web 12/16/2009

MultiValent Benzoboroxole Polymer gp120 Entry Inhibitor
articles
events to susceptible cells containing CD4 receptors on
immune system cells. Antiviral agents that act as entry
inhibitors by binding to gp120 have the capability to
inactivate HIV-1 virions in the vaginal lumen before they
reach susceptible CD4+ cells and thus prevent the first step
synthetic-CBAs which may provide the broad-spectrum
targeting capabilities of the CBAs without the cost of protein
production, isolation, and subsequent chemical and structural
stability issues.
Phenylboronic acid (PBA) forms reversible covalent
5
1
3
in the male-to-female heterosexual transmission of HIV-1.
complexes with diols, a common chemical moiety in
saccharides like fructose, glucose and mannose. Boronic
acids undergoes a well-known condensation reaction with
The recent failure of sulfonated anionic polymer entry
inhibitors to demonstrate efficacy in phase III clinical trials
3
exemplifies the need for new HIV-1 entry inhibitors.
1
,2- or 1,3-diols to form five-or six-membered cyclic
Carbohydrate-binding agents (CBAs) capable of binding
to the multiple glycans on the N-linked glycosylated regions
of gp120 represent a promising class of entry inhibitors.
boronate esters. This condensation reaction is reversible and
is highly influenced by the pH and chemical structure of the
6
1
4,15
boronic acid and diols.
Conversion of the boronic acid
Of these, the CBA protein, cyanovirin-N (CV-N), which
binds specifically to R(1,2)mannose glycans, has shown
consistent suppression of HIV-1 and HIV-2 replication
irrespective of the virus, the cell type, or the coreceptor
to the charged boronate tetrahedral conformation yields a
more stable complex that resists hydrolysis compared to its
trigonal form of the complex which is more readily revers-
ible. Investigations by B e´ rub e´ et al. reveal that the water-
soluble, o-hydroxymethylphenylboronic acid (benzoboroxole,
7
tropism of the viral strain. However, production, formulation
and stability of protein-based lectins may economically
impede their use by women in resource-poor pandemic
1
), demonstrates binding at physiological pH to nonreducing
sugars like methyl R-D-mannopyranoside and methyl ꢀ-D-
8
1
6
regions like sub-Saharan Africa. Investigations are therefore
galactopyranoside which are structurally similar to the
terminal sugar moieties found on the high mannose and
complex-type N-linked glycans of gp120. It is hypothesized
that the o-hydroxymethyl substituent is able to stabilize the
arylboronic acid in its tetrahedral conformation at neutral
pH thus preventing rapid hydrolysis of the arylboronic
acid-diol complex at physiological pH. Benzoboroxole-
containing constructs are currently being exploited in gly-
croprotein detection.
9-12
ongoing to identify non-peptide or protein based CBAs.
We are interested in developing multivalent polymer based
(
(
1) UNAIDS Report on the Global HIV/Aid Epidemic; The Joint
United Nations Program on HIV/AIDS: Geneva, Switzerland,
2
008; pp 31-64.
16
2) Sekaly, R.-P. The Failed HIV Merck Vaccine Study: a Step Back
or a Launching Point for Future Vaccine Development. J. Exp.
Med. 2008, 205 (1), 7–12.
1
7,18
(
(
(
(
3) Cutler, B.; Justman, J. Vaginal Microbicides and the Prevention
of HIV Transmission. Lancet Infect. Dis. 2008, 8 (11), 685–697.
4) Blazarini, J.; Damme, L. V. Microbicide Drug Candidates to
Prevent HIV Infection. Lancet 2007, 369 (9563), 787–797.
5) Hladik, F.; Hope, T. J. HIV Infection of the Genital Mucosa in
Women. Curr. HIV/AIDS Rep. 2009, 6 (1), 20–28.
There are 24 N-linked glycosylation sites (HIV-1/III ) on
B
gp120 of which 30-50% are high mannose glycans with
three terminal nonreducing mannose residues, and 50-63%
are complex glycans terminating to some degree with
6) Balzarini, J. Targeting the Glycans of Glycoproteins: a Novel
(11) Mazik, M.; Hartmann, A. Phenanthroline Unit as a Building Block
for Carbohydrate Receptors. J. Org. Chem. 2008, 73 (19), 7444–
7450.
(12) Abe, H.; Horii, A.; Matsumoto, S.; Shiro, M.; Inouye, M. D3h-
Symmetrical Hydrogen-Bonding Unit as a Saccharide Recognition
and Self-Assembling Module. Org. Lett. 2008, 10 (13), 2685–
2688.
Paradigm for Antiviral Therapy. Nat. ReV. Microbiol. 2007, 5 (8),
5
83–597.
(
7) Boyd, M. R.; Gustafson, K. R.; McMahon, J. B.; Shoemaker,
R. H.; O’Keefe, B. R.; Mori, T.; Gulakowski, R. J.; Wu, L.;
Rivera, M. I.; Laurencot, C. M.; Currens, M. J.; Cardellina, J. H.,
2
nd; Buckheit, R. W., Jr.; Nara, P. L.; Pannell, L. K.; Sowder,
R. C., 2nd; Henderson, L. E. Discovery of Cyanovirin-N, a Novel
Human Immunodeficiency Virus-Inactivating Protein That Binds
Viral Surface Envelope Glycoprotein gp120: Potential Applica-
tions to Microbicide Development. Antimicrob. Agents Chemother.
(13) Springsteen, G.; Wang, B. A Detailed Examination of Boronic
Acid-Diol Complexation. Tetrahedron 2002, 58 (26), 5291–5300.
(14) Yan, J.; Springsteen, G.; Deeter, S.; Wang, B. The Relationship
among pKa, pH, and Binding Constants in the Interactions
between Boronic Acids and Diols--It Is Not as Simple as It
Appears. Tetrahedron 2004, 60 (49), 11205–11209.
(15) Singhal, R. p.; Ramamurhy, B.; Govindraj, N.; Sarwar, Y. New
Ligands for Boronte Affinity Chromatography: Synthesis and
Properties. J. Chromatogr. A 1991, 543, 17–38.
(16) B e´ rub e´ , M.; Dowlut, M.; Hall, D. G. Benzoboroxoles as Efficient
Glycopyranoside-Binding Agents in Physiological Conditions:
Structure and Selectivity of Complex Formation. J. Org. Chem.
2008, 73 (17), 6471–6479.
(17) Ricardo, A.; Frye, F.; Carrigan, M. A.; Tipton, J. D.; Powell, D. H.;
Benner, S. A. 2-Hydroxymethylboronate as a Reagent to Detect
Carbohydrates: Application to the Analysis of the Formose
Reaction. J. Org. Chem. 2006, 71 (25), 9503–5.
(18) Mads D., orensen, R. M. Assessing the Terminal Glycosylation
of a Glycoprotein by the Naked Eye. Angew. Chem., Int. Ed. 2007,
46 (14), 2403–2407.
1
997, 41 (7), 1521–1530.
(
(
8) Balzarini, J.; Hatse, S.; Vermeire, K.; Princen, K.; Aquaro, S.;
Perno, C. F.; De Clercq, E.; Egberink, H.; Vanden Mooter, G.;
Peumans, W.; Van Damme, E.; Schols, D. Mannose-Specific Plant
Lectins from the Amaryllidaceae Family Qualify as Efficient
Microbicides for Prevention of Human Immunodeficiency Virus
Infection. Antimicrob. Agents Chemother. 2004, 48 (10), 3858–
3
870.
9) Balzarini, J.; Van Laethem, K.; Daelemans, D.; Hatse, S.; Bugatti,
A.; Rusnati, M.; Igarashi, Y.; Oki, T.; Schols, D. Pradimicin a, a
Carbohydrate-Binding Nonpeptidic Lead Compound for Treatment
of Infections with Viruses with Highly Glycosylated Envelopes,
Such as Human Immunodeficiency Virus. J. Virol. 2007, 81 (1),
3
62–373.
(
10) Striegler, S.; Dittel, M. A Sugar Discriminating Binuclear Cop-
per(II) Complex. J. Am. Chem. Soc. 2003, 125 (38), 11518–11524.
VOL. 7, NO. 1 MOLECULAR PHARMACEUTICS 117

articles
Jay et al.
a
Scheme 1
a
(
a) o-Hydroxymethylphenylboronic acid (benzoboroxole, 1). (b) Hypothetical scheme of the binding chemistry: multivalent polymer-bound 1 binding due
16
to complexation with a high mannose region of gp120 through interaction with the 4,6-diols of mannopyranose residues. (c) Graphical depiction of the
multivalent benzoboroxole-functionalized polymer (III) interacting with the gp120 complex (II) of HIV-1 (I). The polymer-gp120 interaction creates a
sterically constrained surface. (d) The interaction drawn to scale with multivalent polymers (III) interacting with the gp120 heterotrimer’s (II) many glycosylated
regions (IV).
1
9,20
nonreducing galactose moieties.
Considering that up to
a sterically hindered polymer layer on HIV-1 that prevents
interaction of gp120 with CD4 and/or CCR5 receptors
(Scheme 1). Multivalent conjugate approaches have been
utilized to improve the affinity for ligand binding, especially
1
4 gp120-trimers have been identified on the envelope of
21
HIV-1, the HIV-1 envelope presents multiple binding sites
for benzoboroxole-containing ligands. Incorporating the
benzoboroxole ligand into a flexible and water-soluble
polymer backbone would provide multivalent interaction with
abundant sites on the HIV-1 envelope, thus reducing the
impact of hydrolysis for a single benzoboroxole-diol
complex. Furthermore, binding of the polymer could create
22
in cases of binding to surfaces that have repeated epitopes.
Numerous studies have demonstrated the improvement of
multivalent affinity over monovalent interactions by conju-
23,24
25
26
gating small molecules,
polymeric architectures.
Fab, and peptides into
(
19) Gallaher, W. R.; Ball, J. M.; Garry, R. F.; Martin-Amedee, A. M.;
Montelaro, R. C. A General Model for the Surface Glycoproteins
of HIV and Other Retroviruses. AIDS Res. Hum. RetroViruses
(22) Mammen, M.; Choi, S.-K.; Whitesides, G. M. Polyvalent Interac-
tions in Biological Systems: Implications for Design and Use of
Multivalent Ligands and Inhibitors. Angew. Chem., Int. Ed. 1998,
37 (20), 2754–2794.
(23) Hernandez-Rocamora, V. M.; Maestro, B.; de Waal, B.; Morales,
M.; Garcia, P.; Meijer, E. W.; Merkx, M.; Sanz, J. M. Multivalent
Choline Dendrimers as Potent Inhibitors of Pneumococcal Cell-
Wall Hydrolysis. Angew. Chem., Int. Ed. 2009, 48 (5), 948–951.
(24) Wang, S.-K.; Liang, P.-H.; Astronomo, R. D.; Hsu, T.-L.; Hsieh,
S.-L.; Burton, D. R.; Wong, C.-H. Targeting the Carbohydrates
on HIV-1: Interaction of Oligomannose Dendrons with Human
Monoclonal Antibody 2g12 and Dc-Sign. Proc. Natl. Acad. Sci.
U.S.A. 2008, 105 (10), 3690–3695.
1
995, 11 (2), 191–202.
(
20) Leonard, C. K.; Spellman, M. W.; Riddle, L.; Harris, R. J.;
Thomas, J. N.; Gregory, T. J. Assignment of Intrachain Disulfide
Bonds and Characterization of Potential Glycosylation Sites of
the Type 1 Recombinant Human Immunodeficiency Virus Enve-
lope Glycoprotein (gp120) Expressed in Chinese Hamster Ovary
Cells. J. Biol. Chem. 1990, 265 (18), 10373–10382.
21) Zhu, P.; Liu, J.; Bess, J., Jr.; Chertova, E.; Lifson, J. D.; Grise,
H.; Ofek, G. A.; Taylor, K. A.; Roux, K. H. Distribution and
Three-Dimensional Structure of AIDS Virus Envelope Spikes.
Nature 2006, 441 (7095), 847–852.
(
118 MOLECULAR PHARMACEUTICS VOL. 7, NO. 1

MultiValent Benzoboroxole Polymer gp120 Entry Inhibitor
articles
The aim of this study is to investigate the binding of 1 to
gp120 and to determine the impact that its incorporation into
a polymer backbone at different mole ratios has on binding
affinity and antiviral activity. To this end, we synthesized a
benzoboroxole-functionalized monomer and incorporated it
at different feed ratios into a linear water-soluble, biocom-
patible backbone poly[N-2-hydroxypropyl]methacrylamide]
Laboratories, Amherst, MA), a differential refractive index
detector (BI-DNDC, Brookhaven Instruments, Holtsville,
NY) and a multiangle light scattering detector (BI-MwA,
Brookhaven Instruments, Holtsville, NY). A standard poly-
mer (PEO, Varian Inc., Palo Alto, CA) was prepared at 1
-
1
mg mL in DMF and used for calibration.
SPR Benzoboroxole (1) Affinity to gp120. 2-Hydroxy-
methylphenylboronic acid (Frontier Scientific, Logan, UT)
was prepared at 50 mM in pH 7.5 25 mM phosphate buffer
or a 25 mM carbonate buffer. HIV-1BaL gp120, recombinant,
produced in human embryonic kidney 293 cells (HEK), was
obtained from the NIH AIDS Research and Reference
Reagent Program (Division of AIDS, NIH). HIV-1BaL gp120
was attached to a CM4 or C1 sensor chip using standard
amine coupling chemistry at two to three densities (9,000
and 3,000 or 1600, 1400, and 900 RU respectively for the
two types of chips). Briefly, using HBS-P as the running
buffer the carboxy surface of a surface plasmon resonance
27
(pHPMAm). In this study binding affinity of benzoboroxole
(
1) and benzoboroxole-functionalized pHPMAm oligomers
to gp120 were assessed by SPR. Higher molecular weight
polymers were then tested for antiviral activity against three
strains of HIV-1. Because these polymers are intended to
be used in the vaginal lumen as an inhibitor of HIV-1 entry,
the polymers’ cytotoxicity was tested in the Vk2/E6E7
2
8
human vaginal cell line.
Experimental Section
Materials. Methacrylic acid (MAA) (stabilized with 250
ppm MeHQ) was purchased from Acros. O-Benzotriazole-
N,N,N′,N′-tetramethyl-uronium-hexafluoro-phosphate (HBTU)
was purchased from AK Scientific, Inc. (Mountain View,
CA). 5-Amino-2-hydroxymethylphenylboronic acid, HCl,
dehydrate and N-acetylneuraminic acid (sialic acid) were
purchased from Combi-Blocks, Inc. (Chester, PA). Methyl
(
SPR) chip (CM4 (carboxydextran) or C1 (carboxymethyl)
sensor chip, Biacore Life Sciences, Piscataway, NJ) was
activated with an injection of a 1:1 ratio of 0.4 M 1-ethyl-
3
-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC):
.1 M N-hydroxysuccinimide. HIV-1BaL gp120 in 10 mM
0
sodium acetate (pH 5.5) was then coupled to the surface.
The remaining activated groups were then blocked by an
injection of 1 M ethanolamine (pH 8.5). All coupling and
affinity analysis were performed at 25 °C using the Biacore
ꢀ-D-galactopyranoside was purchased from Alfa Aaesar. All
other solvents and reagents were purchased from Sigma-
Aldrich (St. Louis, MO) including 2,2′-azobisisobutyronitrile
3
000 (Biacore Life Sciences), which has improved signal-
(
AIBN), which was then recrystallized from chloroform in
to-noise ratio for samples that exhibit large bulk index of
refraction changes. Affinity for the gp120 functionalized
surface was determined by comparing raw data to three
reference cells: two of the unactivated CM4 or C1 surfaces,
and an activated CM4 or C1 surface capped completely by
ethanolamine. The small molecule ligand was injected at the
highest concentration prepared, followed by a 2-fold dilution
series in the appropriate buffer to produce a concentration
series. The association and dissociation phases were moni-
tored for 40-60 s and 3-10 min respectively. All conditions
were repeated at least twice. The reference surface was used
to subtract out any bulk refractive index change. Data was
fitted to a one-to-one model using local maxima using
Scrubber2 software (BioLogic Software, Cambell, Australia)
2
a chloroform-N (l) bath. All other reagents were used as
purchased unless noted. When necessary, solvents such as
tetrahydrofuran (THF) and ethyl acetate were dried over
activated 4Å sieves, as indicated by the solvent name
followed by 4Å. Thin layer chromatography plates were
purchased from Whatman (aluminum backing, UV fluores-
cence 254 nm, silica gel, Kent, U.K.). Flash chromatography
was performed using silica gel (200-400 mesh, 60 Å,
1
13
Aldrich, St. Louis, MO). All H and C NMR spectra were
1
acquired on a Varian Mercury 400 MHz spectrometer. H
chemical shifts are reported as δ referenced to solvent, and
coupling constants (J) are reported in Hz. Polymer molecular
distributions were determined in HPLC grade DMF using
GPC (GPC 1100, Agilent Technologies, Santa Clara, CA)
equipped with an organic column (PLgel mixed-B, Polymer
D D
to determine K and k .
Chemical Synthesis of 5-Methacrylamido-2-hydroxym-
ethylphenylboronic Acid (MAAm-OHMPBA, 4). MAAm-
OHMPBA (4) was synthesized as follows. First the HBTU
activation of methacrylic acid (MAAm-HBTU, i, Scheme
(
25) Johnson, R. N.; Kope cˇ kov a´ , P.; Kope cˇ ek, J. i. Synthesis and
Evaluation of Multivalent Branched Hpma Copolymer-Fab’
Conjugates Targeted to the B-Cell Antigen CD20. Bioconjugate
Chem. 2009, 20 (1), 129–137.
2
) was synthesized as follows. In the first step, i, methacrylic
acid (2, MAAm, 0.5256 mL, 6.20 mmol) was dissolved in
25 mL of tetrahydrofuran (THF) 4Å. HBTU (2.9456 g, 7.78
mmol) and diisopropylethylamine (DIPEA, 1.127 mL, 6.47
mmol) were added to the solution. The reaction was flushed
with N
that reaction had gone to completion (MAAm, R
MAAm-HBTU, R ) 0.5; HBTU, R ) 0; 2:1 hexane:ethyl
(
26) Liu, Z.; Deshazer, H.; Rice, A. J.; Chen, K.; Zhou, C.; Kallenbach,
N. R. Multivalent Antimicrobial Peptides from a Reactive Polymer
Scaffold. J. Med. Chem. 2006, 49 (12), 3436–3439.
(
27) Haag, R.; Kratz, F. Polymer Therapeutics: Concepts and Applica-
tions. Angew. Chem., Int. Ed. 2006, 45 (8), 1198–1215.
28) Fichorova, R. N.; Rheinwald, J. G.; Anderson, D. J. Generation
of Papillomavirus-Immortalized Cell Lines from Normal Human
Ectocervical, Endocervical, and Vaginal Epithelium That Maintain
Expression of Tissue-Specific Differentiation Proteins. Biol.
Reprod. 1997, 57 (4), 847–855.
2
(g) and ran for 4 h at room temperature. TLC showed
(
f
) 0.7;
f
f
acetate 1% acetic acid). Reaction material was moved
forward without any additional workup. In a second round-
VOL. 7, NO. 1 MOLECULAR PHARMACEUTICS 119

articles
Jay et al.
Scheme 2. Synthetic Conditions for Synthesis of
Scheme 3. Free Radical Polymer Synthesis of
5
-Methacrylamido-2-hydroxymethylphenylboronic Acid
5-Methacrylamido-2-hydroxymethylphenylboronic Acid (4)
with 2-Hydroxypropylmethacrylamide (5)
a
a
(
4, MAAm-OHMPBA)
a
(
2
i) Methacrylic acid (2, MAAm), HBTU, DIPEA, THF 4Å, N (g), 4 h,
a
RT. (ii) 5-amino-2-hydroxymethylphenylboronic acid, HCl, dehydrate (3),
(iii) MAAm-OHMPBA (4, 0 M, 0.125 M, 0.25 M, or 0.375 M) and
DIPEA, N (g), 0 °C during addition of i, warmed to RT, overnight.
2
HPMAm (5, 0.5 M, 0.375 M, 0.25, or 0.125 M) respectively, AIBN (0.025
2
M), DMF, 65 °C, 24 h, N (g). Polymer is drawn in the hemiboronic ester
tetrahedral conformation (1, pK ) 7.2) creating the stable form for
a
bottom flask, 5-amino-2-hydroxymethylphenylboronic acid
HCl dehydrate (3, 1.00 g, 5.39 mmol) was combined with
THF 4Å (15 mL). DIPEA (1.88 mL, 10.79 mmol) was added
hexopyranoside binding.
2
chamber was flushed with N (g) while the bulk reaction was
and the solution flushed with N
2
(g) while being chilled in
cooled to -20 °C in an ethanol-dry ice bath. The meth-
an ice bath. To this chilled solution, the reaction solution, i,
was added dropwise via syringe while vigorously stirring.
After addition 3 dissolved. The reaction was allowed to warm
to ambient temperature and reacted overnight (ii, Scheme
acryloyl chloride solution was added dropwise at a rate of 1
-
1
drop s . After addition, the reaction was warmed to room
temperature and stirred for an additional 3 h. Upon comple-
tion of reaction, the white solid (1-amino-2-propanol hydro-
chloride) was filtered off and rinsed with ethyl acetate (100
mL). The combined reaction supernatant and rinse were
placed in a -20 °C freezer. Crystals were collected after
24 h via vacuum filtration and rinsed with chilled (4 °C)
ethyl acetate and dried under high vacuum. The remaining
white solid which still showed HPMAm by TLC was rinsed
with acetone (3 times 100 mL) and concentrated. The white
solid was redissolved in ethyl acetate (500 mL) and
recrystallized for 24 h at -20 °C. Crystals were filtered and
2
). TLC confirmed completion (MAAm-OHMPBA, R
HBTU, R ) 0, MAAm-HBTU, R ) 0.7, 90:8:2 chloroform:
methanol:acetic acid). Reaction was condensed to a brown
oil and redissolved in 4 mL of CHCl containing 200 µL of
acetic acid and loaded onto a silica column slurry packed in
5:5 CHCl :MeOH 1% acetic acid, with an initial 20 mL of
CHCl eluted through (dry silica, 150 g; column dimensions,
inner diameter 4.5 cm, height, 23 cm). An additional 20 mL
of CHCl was eluted through after loading, followed by
elution in 95:5 CHCl :MeOH 1% acetic acid. Five milliliter
fractions were collected. Compound began eluting in fraction
7. Fractions 17-30 were collected and stripped to yield 4
f
) 0.5,
f
f
3
9
3
3
3
3
dried under hi-vac for a combined yield of 44 g, 307 mmol,
1
65% yield and characterized by H NMR (400 MHz, CDCl
3
)
1
δ 6.51 (br, 1H), 5.69 (s, 1H), 5.30 (s, 1H), 3.90 (m, 1H),
3.45 (m, 1H), 3.13 (m, 1H), 1.92 (s, 3H), 1.15 (d, 3H,
J ) 6).
1
as a brown solid (1.484 g, 98% yield): H NMR (400 MHz,
MeOD with NaOD to aid in dissolution) 7.4 (m, 2H), 7.0
(
2
d, 1H, J ) 9 Hz), 5.77 (s, 1H), 5.45 (s, 1H), 4.82 (s, 1H),
.02 (s, 3H); C NMR (400 MHz, MeOD with NaOD to
Oligomer Synthesis. Oligomers were synthesized by free
radical polymerization (Scheme 3) in the presence of a chain
transfer agent, 2-aminoethanethiol, with feed ratios of 100:
0, 75:25, and 50:50 of either HPMAm:MAAm-OHMPBA
or HPMAm:acrylic acid (AA, distilled), which was used as
a control to determine if nonspecific electrostatic interactions
occurred with the protein due to a negatively charged
backbone. Polymerizations were performed using 2.5 M
solution of monomer at appropriate molar feed ratios with
2,2′-azobisisobutyronitrile (AIBN, 1 mol %) and 2-amino-
ethanethiol (8 mol %) in MeOH 4Å at 60 °C for 20 h under
1
3
aid in dissolution) δ 168.8, 150.1, 140.9, 137.5, 124.3, 122.5,
1
21,3, 119.5, 71.0, 37.7, 17.2.
Synthesis of 2-Hydroxypropylmethacrylamide (HP-
MAm, 5). The synthesis was modified from a previously
2
9
published procedure. In short 1-amino-2-propanol (73.8
mL, 957 mmol) was added to a 1 L straight three neck round-
bottom flask followed by 500 mL of ethyl acetate 4Å. The
flask was equipped with an overhead stirrer (Caframo RZR
2
000, Digital 2000), thermometer and a 250 mL constant
30
addition funnel. To the addition funnel was added meth-
acryloyl chloride (Fluka) from a new, unopened container
nitrogen atmosphere. All oligomer reactions were triturated
3 times in 40 mL of acetone, centrifuged at 500 rpm for
5-15 min and supernatant decanted. HPMAm:acrylic acid
(
48 mL, 478 mmol). Dry ethyl acetate (70 mL) was added
-
1
to the addition funnel to dilute the methacryloyl chloride and
mixed to create a homogeneous solution. The reaction
oligomers were further purified by dialysis at 50 mg mL
using 500 MWCO membrane (Spectrum) against DDI water
(
29) Kopecek, J.; Bazilov a´ , H. Poly[N-(2-Hydroxypropyl)Methacry-
lamide]--I. Radical Polymerization and Copolymerization. Eur.
Polym. J. 1973, 9 (1), 7–14.
(30) Chen, G. H.; Hoffman, A. S. Graft-Copolymers That Exhibit
Temperature-Induced Phase- Transitions over a Wide-Range of
pH. Nature 1995, 373 (6509), 49–52.
120 MOLECULAR PHARMACEUTICS VOL. 7, NO. 1

MultiValent Benzoboroxole Polymer gp120 Entry Inhibitor
articles
a
Table 1. Oligomer and Polymer Compositions and Molecular Weight Distributions
b
c
actual mole ratio (mol %)
molecular weight (kDa)
oligomer/polymer
oligo(HPMAm100) (oligo-6
HPMAm
MAAm-OHMPBA
AA
M
w
M
n
0
)
100
74
1.5
oligo(HPMAm75-(MAAm-OHMPBA25)) (oligo-625
)
)
26
43
0.75
1.24
1.9
oligo(HPMAm50-(MAAm-OHMPBA50)) (oligo-650
57
oligo(HPMAm75-(AA25)) (oligo-AA25
)
)
65
35
51
oligo(HPMAm50-(AA50)) (oligo-AA50
49
1.8
p(HPMAm100) (6
0
)
100
71
118.1
65.9
88.4
73.6
49.6
p(HPMAm75-(MAAm-OHMPBA25)) (625
p(HPMAm50-(MAAm-OHMPBA50)) (650
p(HPMAm25-(MAAm-OHMPBA75)) (675
)
)
)
29
52
70
103.9
113.9
74.5
48
30
a
HPMAm: 2-hydroxypropylmethacrylamide. MAAm-OHMPBA: 5-methacrylamido-2-hydroxymethylphenylboronic acid. AA: acrylic acid.
b
1
c
Actual molar ratio was determined by H NMR.
w w n
M for oligomers was determined by MALDI-TOF, and M and M for polymers were
determined by GPC equipped with multiangle light scattering and differential refractive index detectors and are represented as means of at
least triplicate measurements.
followed by lyophilization. Purity and actual molar feed ratios
of polymers were determined in DMSO by H NMR (Mercury
index change. The response units were normalized by molecular
weight of the analyte, 152 and 750 Da for 1 and oligo-625
respectively. Data were fit to a one-to-one model using a global
maximum in Scrubber2 software to determine the KD.
1
400 MHz spectrometer, Varian) (Table 1). Molecular weight
distributions were determined using MALDI-TOF in the pres-
ence of NaCl under alkaline conditions (Table 1).
Oligomer SPR. Experiments were run identically to the
small molecule SPR but on a C1 flat carboxylated sensor
Polymer Synthesis. Polymers were synthesized by free
radical polymerizations using 100:0 (6 ), 75:25 (6 ), 50:50
0
25
(6 ) or 25:75 (6 ) mole ratios of HPMAm (5) with MAAm-
50
75
0
chip. Oligo(HPMAm100) (oligo-6 ), oligo(HPMAm75-(MAAm-
OHMPBA (4). Polymerizations were performed using 0.5
M solution of monomer with 2,2′-azobisisobutyronitrile
OHMPBA25)) (oligo-625), and oligo(HPMAm75-(AA25)) (oligo-
AA25) were prepared at 15 mg mL- in 25 mM phosphate
buffer, while the oligo(HPMAm50-(MAAm-OHMPBA50))
1
(AIBN, 5 mol %) in DMF 4Å at 65 °C for 24 h under
nitrogen atmosphere. All polymers reactions were triturated
into 40 mL of ether, centrifuged at 500 rpm for 5-15 min
and supernatant decanted. Polymers were placed under high
vacuum overnight to remove residual DMF. Polymers were
further purified by dissolving in 100 mM PBS buffer (50
(
oligo-650) and oligo(HPMAm50-(AA50)) (oligo-AA50) were
-
1
prepared at 1 mg mL in 2% DMSO (v/v) in 25 mM
phosphate buffer. SPR was also performed in the presence
of fructose and glucose to determine how incubation with
physiological sugar concentration impacts the benzoboroxole-
functionalized oligomers binding to gp120. For these experi-
ments oligo-625 and oligo-650 were prepared at 10 or 1 mg
-1
mM NaCl, iso-osmolar) to a concentration of 10 mg mL
and centrifuging through a 3000 MWCO membrane (Amicon
Ultra, 4 mL, 3000 MWCO, Millipore, Bellerica, MA). The
-
1
mL respectively in 2% DMSO, 25 mM phosphate buffer
containing 16 mM fructose and 6 mM glucose, the physi-
7
5 mol % sample required adjusting the buffer to pH 11 in
order for complete dissolution to occur. Samples were
centrifuged at 3000 rpm for 45 min, filtrate removed and
retentate diluted up to initial volume in buffer. This was
repeated three times with buffer, followed by three times
with DDI H O. Retentate was lyophilized. The resulting
2
yields after purification were around 60% for each polymer.
31
ological sugar concentrations found in seminal plasma. To
more specifically determine how the type of sugar impacted
the binding of oligo-625 to the gp120-functionalized surface
a solution of 7.5 mg mL- oligo-625 was preincubated with
fructose, N-acetylneuraminic acid (sialic acid), methyl R-D-
mannopyranoside or methyl ꢀ-D-galactopyranoside at 0.5,
1
Actual molar feed ratios of polymers were determined in
1
, or 10 times the ratio of sugar to oligomer-bound
1
D
2
O by H NMR (Mercury 400 MHz spectrometer, Varian)
benzoboroxole. Solutions were prepared in a pH 7.5 25 mM
phosphate buffer. The pH 7.5 25 mM phosphate buffer was
also used as the running buffer. The concentration of
benzoboroxole in the oligomer (oligomer-bound 1) solutions
in DMF was determined by UV/vis spectroscopy (USB-ISS-
UV/vis USB4000 spectrometer, Ocean Optics, Dunedin, FL)
at 320 nm using a calibration curve created from the MAAm-
OHMPBA monomer ensuring to not measure at a wavelength
where the alkene absorbed. For all SPR experiments the
reference surface was used to subtract out any bulk refractive
(
Table 1). Polymer molecular distributions were determined
in DMF using GPC (GPC 1100, Agilent Technologies, Santa
Clara, CA) equipped with an organic column (PLgel mixed-
B, Polymer Laboratories, Amherst, MA), a differential
refractive index detector (BI-DNDC, Brookhaven Instru-
ments, Holtsville, NY) and a multiangle light scattering
detector (BI-MwA, Brookhaven Instruments, Holtsville, NY)
(Table 1). The concentration of polymer-bound 1 in each
polymer solution was determined in the same manner as that
described for the oligomers.
Neutralization Assay. We used two R5-tropic, well-
characterized, reference strains of HIV-1 isolated from acute,
sexually transmitted infections and grown in peripheral blood
(
31) Owen, D. H.; Katz, D. F. A Review of the Physical and Chemical
Properties of Human Semen and the Formulation of a Semen
Simulant. J. Androl. 2005, 26 (4), 459–469.
VOL. 7, NO. 1 MOLECULAR PHARMACEUTICS 121

articles
Jay et al.
3
2
mononuclear cells (PBMCs): HIV-1 DU156 (Clade C) and
the viral solution. Additionally a dilution of DMEM diluted
with 2xDMEM was performed, followed by a similar dilution
series as the polymers with normal DMEM to determine if
this sample preparation impacted cytotoxicity or viral activity.
Care was taken to ensure that all RLU values were within
the linear range of the TZM-bl assay. After 48 h incubation
at 37 °C, 150 µL was removed from each well and 100 µL
of Britelite Reagent (PerkinElmer, Waltham, MA) was added.
After two minute room temperature incubation, the wells
were mixed by pipet action and 150 µL was transferred to a
96-well black plate and read in a luminometer (Wallac 1420
Victor, Perkin-Elmer, Waltham, MA). Relative luminescence
values were compared to negative (cells-only, no virus) and
positive (cells with virus) controls. Both polymer solutions
and all media solutions alone were also analyzed for relative
cytotoxicity using the same assay without any virus. In these
cytotoxicity assays, the cellular luminescence from each well
was compared to cell controls. A drop in well luminescence
suggests a sample-mediated reduction in cell number.
Concentrations showing a drop in cellular luminescence are
not reported in the neutralization dose response plots due to
potential cytotoxic effects on the cells used during antiviral
neutralization. No effect was noted for the dilution of DMEM
with 2xDMEM either in the cytotoxicity or viral samples.
The positive control, Chicago Sky Blue successfully inac-
0
3
3
TORNO (Clade B) as well as a pseudotyped X4-tropic
strain, WEAU (Clade B). Polymers were tested for
neutralization activity using a previously established and
validated assay for testing neutralizing antibodies.
Polymers and Chicago Sky Blue, a sulfate antiviral control,
were prepared in DMEM (4.5 g L D-glucose, 110 mg mL
sodium pyruvate and L-glutamine, Invitrogen, Carlsbad, CA)
alone, with the pH adjusted to 7.5 as necessary, and
pasteurized for 5 min at 70 °C, followed by 2-fold dilution
with 2X growth media in the first row of the 96-well plate:
DMEM supplemented with 20% fetal bovine serum (FBS,
heat activated, Hyclone, Logan, UT), 50 mM HEPES (Gibco/
Invitrogen, Carlsbad, CA) and 100 µg mL gentamicin
Sigma, St. Louis, MO), referred to as 2xDMEM. Preparation
was designed to pasteurize the polymers without proteins
being present in the media. The highest concentration for
3
4
3
5
-1
-1
-1
(
-
1
-1
each polymer was 6
6
0
at 20 mg mL , 625 at 20 mg mL ,
50 at 10 mg mL , and 675 at 3 mg mL , and Chicago Sky
Blue at 0.08 mg mL in DMEM with 10% FBS, 25 mM
-
1
-1
-
1
HEPES, and 50 µg mL- gentamicin.
1
A 7.5-fold dilution series was prepared using normally
prepared DMEM with 10% FBS, 25 mM HEPES, and 50
-
1
µg mL gentamicin (referred to as normal media). The
normal media alone was also used as a control. For the cell
control wells, 150 µL of normal DMEM media was added.
The viral control wells contained 100 µL of normal DMEM,
and for the neutralization assay the wells contained 100 µL
of the appropriate polymer concentration from the dilution
series. Each solution was incubated with 50 µL of 350-650
TCID virus for 60 min. Then 10,000 TZM-bl cells (100 µL
tivated HIV-1 while the negative control 6 showed activity
-
1
only at the highest concentration tested, 20 mg mL .
Antiviral activity is reported as the sample concentration at
which the RLUs were reduced by 50% compared to virus
control wells after subtraction of background RLUs. These
EC values were determined using GraphPad Prism software
50
5
-1
-1
of 1 × 10 cells mL in normal media with 10 µg mL
(Graph Pad, LaJolla, CA).
DEAE dextran (∼500,000 Da, Sigma, St. Louis, MO) were
added to each well. Cytotoxicity was also studied. These
experiments were set up identically to the neutralization
plates except 50 µL of normal media was added rather than
Cell Toxicity. Polymer samples were prepared for cyto-
toxicity evaluation in keratintocyte serum-free media (KSF,
Invitrogen, Carlsbad, CA) and pasteurized at 70 °C for five
minutes. A 2-fold dilution was made with 2X growth media
(
KSF supplemented with 2-fold nutrients and pen-strep (P/
(
32) Bures, R.; Morris, L.; Williamson, C.; Ramjee, G.; Deers, M.;
Fiscus, S. A.; Abdool-Karim, S.; Montefiori, D. C. Regional
Clustering of Shared Neutralization Determinants on Primary
Isolates of Clade C Human Immunodeficiency Virus Type 1 from
South Africa. J. Virol. 2002, 76 (5), 2233–2244.
33) Bures, R.; Gaitan, A.; Zhu, T.; Graziosi, C.; McGrath, K. M.;
Tartaglia, J.; Caudrelier, P.; El Habib, R.; Klein, M.; Lazzarin,
A.; Stablein, D. M.; Deers, M.; Corey, L.; Greenberg, M. L.;
Schwartz, D. H.; Montefiori, D. C. Immunization with Recom-
binant Canarypox Vectors Expressing Membrane-Anchored Gly-
coprotein 120 Followed by Glycoprotein 160 Boosting Fails to
Generate Antibodies That Neutralize R5 Primary Isolates of
Human Immunodeficiency Virus Type 1. AIDS Res. Hum.
RetroViruses 2000, 16 (18), 2019–2035.
-1
S): 0.2 ng mL EGF (epidermal growth factor, Invitrogen),
1
v v P/S (Invitrogen), 0.8 mM CaCl , referred to as 2xKSF)
2
resulting in the same polymer concentrations as those used
in the neutralization assay. Immortalized Vk2/E6E7 human
vaginal cells were seeded in 96-well plates, 5000 cells/
well, in 250 µL of normal cell growth medium. The cells
were cultured for 24 h; the growth medium was then removed
and replaced with growth medium containing the polymers,
and the cells were grown for an additional 24 h. A solution
-1
00 µg mL BPE (bovinepituitary extract, Invitrogen), 2%
-1
(
2
8
-1
0
of N-9 (1 mg mL ) and a solution of 6 , prepared identically
as the polymers, served as toxic and nontoxic controls
respectively. Metabolic activity of each well was determined
relative to no-treatment control wells (dilution of the KSF)
using the MTS cell proliferation assay (Promega, Madison,
WI). Briefly, 30 µL of the MTS reagent was added to each
well and the plate was incubated at 37 °C for three hours.
After incubation, the absorbance was measured at 490 nm
in a plate reader (Spectramax M2, Molecular Devices,
Sunnyvale, CA). Absorbance of each well was subtracted
(
(
34) Wei, X.; Decker, J. M.; Wang, S.; Hui, H.; Kappes, J. C.; Wu,
X.; Salazar-Gonzalez, J. F.; Salazar, M. G.; Kilby, J. M.; Saag,
M. S.; Komarova, N. L.; Nowak, M. A.; Hahn, B. H.; Kwong,
P. D.; Shaw, G. M. Antibody Neutralization and Escape by HIV-
1
. Nature 2003, 422 (6929), 307–312.
35) Hammonds, J.; Chen, X.; Fouts, T.; DeVico, A.; Montefiori, D.;
Spearman, P. Induction of Neutralizing Antibodies against Human
Immunodeficiency Virus Type 1 Primary Isolates by Gag-Env
Pseudovirion Immunization. J. Virol. 2005, 79 (23), 14804–14814.
122 MOLECULAR PHARMACEUTICS VOL. 7, NO. 1

MultiValent Benzoboroxole Polymer gp120 Entry Inhibitor
articles
from the background (blank KSF media with MTS reagent)
and percentage viability was calculated as: % viability )
absorbance of sample well at 490 nm/absorbance of the
corresponding media at 490 nm × 100.
Statistical Analysis. Statistical analysis was performed on
the EC50 of the polymer concentration. A one-way ANOVA
was performed to evaluate the significance the EC50 on each
strain of HIV-1 tested. For comparisons of two data point sets
a two-tailed t test was performed. A p-value of 0.05 was used.
Results
SPR Assays. We initially determined whether the weak
binding of benzoboroxole, 1, for hexopyranoside as deter-
mined by B e´ rub e´ et al. correlated to an affinity for gp120
by performing surface plasmon resonance (SPR). HIV-1BaL
gp120, recombinant, produced in HEK cells, was conjugated
to a carboxydextran sensor chip. The binding affinity at pH
9.5 and pH 7.5 was assessed at two densities of gp120 (9,000
and 3,000 RU) (Figure 1). The response corresponded with
the density of gp120 conjugated to the sensor chip with the
higher density yielding a greater response at both pHs (see
SFigure 1 in the Supporting Information). At pH 7.5 there
was a higher response upon binding of 1 but more rapid
dissociation compared to the pH 9.5 condition where a lower
response was observed, but the dissociation was slower. The
raw SPR suggests that at pH 7.5 there may be more binding
sites for 1 on gp120, but the affinity is lower; at pH 9.5 there
are correspondingly fewer binding sites but 1 appears to
exhibit a stronger affinity for them. The kinetics observed
during the washout phase exhibited dissociation rates with
-
1
-1
D
k ’s of 4.7 s at pH 7.5 compared to 0.32 s at pH 9.5.
Using a one-to-one binding model, the average affinity
determined from binding to the two densities of gp120
revealed a 4-fold increase in affinity at pH 9.5 compared to
pH 7.5 with respective K ’s of 46.5 mM and 187.5 mM. No
D
nonspecific binding to the carboxydextran reference sensor
chip was detected.
Figure 1. Responses for benzoboroxole (1, 75, 37.5,
1
8.75, 9.4, 4.7, 2.3, 1.2, 0.6, 0.3, 0.15 mM) binding to
HIVBAL gp120 captured on a carboxydextran surface at
the highest density conjugated (9,000 RU) at pH 7.5 (a)
and pH 9.5 (b) in a 25 mM phosphate or carbonate
buffer, respectively. The arrow indicates the start of the
dissociation phase.
Based on weak affinity of the benzoboroxole 1 for diols
on gp120 as measured by SPR we synthesized linear
benzoboroxole-functionalized oligomers to determine if
multivalency could improve affinity. Oligomers were chosen
to reduce the bulk refractive index changes in the SPR assay
as compared to large molecular weight polymers used later
in our studies. We first synthesized a benzoboroxole-
functionalized monomer (MAAm-OHMPBA, 4) by reacting
methacrylic acid (2) with the commercially available precur-
sor, 5-amino-2-hydroxymethylphenylboronic acid HCl de-
hydrate (3) using standard amidation chemistry. Oligomers
were then synthesized by reacting 4 with HPMAm (5) using
free radical polymerization in the presence of the chain
transfer agent 2-aminoethanethiol at feed ratios of 0:100, 25:
to correlate with feed ratio (Table 1). The molecular weight
was determined by MALDI-TOF and ranged from 750 to
1
,900 Da (Table 1).
A flat carboxymethyl sensor chip was used to assess
benzoboroxole-containing-oligomer binding to HIVBaL gp120
as well as to prevent any nonspecific interactions of the
benzoboroxole-functionalized oligomers with the dextran
surface used for initial assessment of the small molecule 1.
The flat sensor chip reduces the overall density of gp120
conjugated to the surface, decreases the assay sensitivity,
and modifies the surface presentation of gp120. These factors
likely impact the K determined for benzoboroxole 1 on this
D
7
5 and 50:50 respectively. Acrylic acid based HPMAm
sensor chip compared to that determined on the carboxy-
oligomers were similarly prepared to determine if nonspecific
electrostatic interactions affected binding affinity. The degree
of incorporation was determined from H NMR and found
dextran sensor chip (see SFigure 2 in the Supporting
Information). The concentration of the oligomer-bound 1
present in the oligomer solutions tested in the SPR assay
1
VOL. 7, NO. 1 MOLECULAR PHARMACEUTICS 123

articles
Jay et al.
was determined by UV/vis spectroscopy at 320 nm to allow
direct comparison to the results for the small molecule 1. A
set of benzoboroxole-functionalized oligomers was prepared,
but insolubility of oligo-650 prevented quantitative analysis
for binding affinity by SPR. Figure 2 displays the raw
response units of the control oligo-AA25, the small molecule
benzoboroxole 1, and the oligo-625 at approximately 6 mM
free or oligomer-bound benzoboroxole concentrations, as
well as the SPR response units normalized for the differences
in molecular weight for each molecule as a function of 1 or
0
the oligomer-bound 1 concentration. The oligo-6 and oligo-
AA25 showed no affinity for the gp120 surface at any
concentration (see SFigures 3-5 in the Supporting Informa-
tion).
In comparing the dissociation phase in the raw SPR data
for both molecules, it was observed that 1 returned rapidly
to the baseline (see SFigure 2 in the Supporting Information),
while oligo-625 returns to baseline slowly (see SFigure 4 in
the Supporting Information). The formation of multiple cross-
links due to the multivalent interactions between the ben-
zoboroxole ligands on the oligomer and terminal glycosylated
residues on single or multiple gp120 units on the sensor chip
may contribute to the slow dissociation. The normalized
D
response curves yielded K values of 70 mM and 8.5 mM
for the small molecule 1 and the oligo-625-bound 1 respec-
tively when fitted to a one-to-one model with a global
maximum.
To understand how exposure to physiological levels of
sugars affects affinity of oligo-625 we studied the binding
response in the presence of fructose and glucose which are
present in seminal plasma at a concentration of approximately
3
1
1
6 mM and 6 mM respectively. B e´ rub e´ et al. determined
the K ’s of benzoboroxole 1 to be 1.65 mM and 32 mM for
fructose and glucose respectively compared to a K of 42
D
D
Figure 2. (a) Raw SPR analysis of interaction between the
small molecule benzoboroxole (1, 6.25 mM), oligo-625
mM for the methyl R-D-mannopyranoside. Binding of the
oligo-625 to the gp120 surface was efficiently disrupted after
preincubation with physiological amounts of sugars. The
fructose concentration was over four times higher than the
highest oligomer-bound 1 concentration and outcompeted
interaction for hexopyranoside residues on gp120. The
dissociation phase correspondingly revealed a much more
rapid return to baseline similar to that occurring in the refer-
ence cell for oligo-625 without sugar present (see SFigures
(
oligomer-bound
1
concentration 6.7 mM; oligomer
-1
concentration 15 mg mL ), and the control oligo-AA25
-1
(oligomer concentration 15 mg mL ) and immobilized
HIVBAL gp120 grown in HEK cells (1600 RU) on a flat
carboxymethyl sensor chip at approximately equivalent
functional group concentrations. The arrow indicates the
start of the dissociation phase. (b) Representative SPR
equilibrium binding curves for 1 and oligo-625 as a function
4
and 5 in the Supporting Information).
of benzoboroxole concentration. Oligomer-bound
1
concentration was determined using UV/vis at 320 nm.
The data was normalized by the molecular weight of the
analytes, 152 and 750 Da respectively The K was
. D
determined using a one-to-one model fit to a global
maximum, yielding values of 70 mM and 8.5 mM for 1 and
oligo-625 respectively. The solid lines represent the fit to
the model. All experiments were conducted at 25 °C in 25
mM phosphate buffer at pH 7.5.
We were also interested in elucidating how the type of
terminal N-linked glycan moiety impacted the binding
specificity of oligo-625 for gp120. The oligo-625 at 7.5 mg
-
1
mL was preincubated with increasing ratio of either
fructose, sialic acid, methyl R-D-mannopyranoside or methyl
ꢀ
(
-D-galactopyranoside to the oligomer-bound 1 concentration
Figure 3). The binding affinity was then tested by SPR As
expected, due to the high affinity of benzoboroxole for
16
fructose, the increasing concentrations of fructose resulted
in a more rapid decrease in binding affinity to gp120. At 1
times the oligomer-bound 1 concentration, fructose reduced
binding to gp120 to 24%. Complete inhibition of binding
occurred at a fructose concentration of 10 times the oliogmer-
bound 1 concentration. The negative SPR response signal
at this concentration of fructose is due to the large mismatch
between the injected oligomer-plus-fructose solution com-
pared to the running buffer. The presence of methyl R-D-
mannopyranoside reduced binding affinity to 88%, 85% and
124 MOLECULAR PHARMACEUTICS VOL. 7, NO. 1

MultiValent Benzoboroxole Polymer gp120 Entry Inhibitor
articles
-
1
Figure 3. (a) Raw SPR analysis of the interaction between oligo-625 (7.5 mg mL ) and immobilized HIVBAL gp120
1300 RU) on a C1, flat carboxymethyl sensor chip, in the presence of increasing ratios (0.5×, 1×, and 10×) of
fructose (FR), methyl R-D-mannopyranoside (MP), N-acetylneuraminic acid (sialic acid, SA), or methyl
-D-galactopyranoside (GP) compared to the concentration of oligomer-bound benzoboroxole (1) at pH 7.5. A
(
ꢀ
representative of four injections is displayed. (b) The average response for the affinity of oligo-625 for gp120 at 30 s
after injection is given as a function of the ratio of sugar concentration to oligomer-bound 1. N ) 4, mean ( SD.
2
9% at 0.5, 1, and 10 times the concentration of oligomer-
ized, reference strains of HIV-1 isolated from acute, sexually
transmitted infections and grown in peripheral blood mono-
nuclear cells (PBMCs), HIV-1 DU156 (Clade C) and
TORNO (TRO, Clade B), as well as the pseudotyped X4-
tropic Clade B strain, WEAU. In North America and Western
Europe, the HIV-1 subtype Clade B predominates, while
Clade C dominates the southern and eastern regions of
Africa. The highest rates of HIV-1 infection occur in southern
bound 1, respectively. Methyl ꢀ-D-galactopyranoside de-
creased the binding affinity to 87% at both 0.5 and 1 times
the concentration of oligomer-bound 1. At 10 times the
concentration of oligomer-bound 1 the affinity was further
reduced to 46%. Sialic acid decreased binding affinity to
82%, 73% and 17% at 0.5, 1, and 10 times the concentration
of oligomer-bound 1, respectively. The ability of sialic acid
to reduce the binding affinity suggests that the sialic acid
terminal residue moieties of the N-linked complex glycans
of gp120 also play a role in the overall affinity of the
benzoboroxole-functionalized oligomer.
1
Africa. Both of the coreceptor tropisms were studied to
determine if the tropic strain impacts activity. The assay was
performed in single-cycle infection experiments using the
37
TZM-bl luciferase reporter gene assay system. These cells
are CXCR4-positive HeLa cell clones that have been
engineered to express high levels of CD4 and CCR5 and
contain the integrated reporter gene for luciferase under
control of the HIV long-term repeat sequence reporter, which
is induced in trans by the viral protein Tat soon after single
The results of decreased binding affinity of the benzoborox-
ole-functionalized oligomer in the presence of seminal plasma
sugar concentrations, as well as from competition with fructose,
methyl R-D-mannopyranoside, methyl ꢀ-D-galactopyranoside
and sialic acid further demonstrate that the affinity of the
benzoboroxole-functionalized oligomers is due to boronic-diol
interactions forming with the terminal N-linked glycan residues
on gp120. The SPR results in the presence of sugars also
indicate that concentrations of the benzoboroxole ligand will
likely need to be increased in the in vivo environment to
maintain affinity for the target ligands in the presence of
physiological sugars and competing glycoproteins present in
the vaginal lumen and seminal plasma.
3
7
round infection. Forty-eight hours post infection the
luciferase activity in the cells was measured. The media used
in the neutralization screen contained 25 mM glucose as well
as the standard 10% fetal bovine serum.
In this HIV-1 entry assay the dose response curves for all
three viral strains revealed consistent suppressed activity in
the presence of all three benzoboroxole-functionalized
polymers (Figure 4). The nonfunctionalized poly(HPMAm),
Viral Neutralization Assays. To determine whether
multivalent benzoboroxole ligands incorporated into a poly-
meric backbone exhibit antiviral activity we synthesized
higher molecular weight polymers. Increasing the molecular
weight augments the potential for sterically hindering the
interaction between gp120 and CD4 and/or the CCR5
coreceptor by creating polymers with a larger radius of
gyration compared to the approximately ten- to twenty-
monomer-containing oligomers used in the SPR binding
assay. Larger molecular weight polymers also decrease
0 0
6 , was tested as a nonactive control. 6 showed activity only
-
1
at 20 mg mL (see Figure 4), the highest concentration
tested against all three strains, suggesting that high polymer
concentration impedes cellular entry in this assay. 625 (∼150
benzoboroxoles per chain) showed only slightly higher
(
36) Duncan, R. The Dawning Era of Polymer Therapeutics. Nat. ReV.
Drug DiscoVery 2003, 2 (5), 347–360.
37) Wei, X.; Decker, J. M.; Liu, H.; Zhang, Z.; Arani, R. B.; Kilby,
J. M.; Saag, M. S.; Wu, X.; Shaw, G. M.; Kappes, J. C. Emergence
of Resistant Human Immunodeficiency Virus Type 1 in Patients
Receiving Fusion Inhibitor (T-20) Monotherapy. Antimicrob.
Agents Chemother. 2002, 46 (6), 1896–1905.
(
3
6
cytotoxicity by preventing cellular uptake.
The in vitro entry inhibitor activity of the functionalized
polymers was determined for two R5-tropic, well-character-
VOL. 7, NO. 1 MOLECULAR PHARMACEUTICS 125

articles
Jay et al.
Figure 5. EC50 determined from the single-cycle HIV-1
infectivity inhibition in TZM-bl cells for 625, 650, and 675
against R5 DU156 (Clade B), R5 TRO (Clade C),
isolated from acute sexually transmitted infections, and
the pseudotyped X4 WEAU (Clade B) viral strains
versus polymer concentration (nM). By ANOVA there
was no statistical difference in the EC50 of each polymer
compared across the three strains (p < 0.05). N ) 3,
mean ( SD.
to 15,000 nM) against all three strains (Figure 5). However
increasing the degree of functionalization to 75 mol % (675
,
∼
450 benzoboroxoles per chain) decreased the EC50 by 3
orders of magnitude against all three strains to ∼10 nM (∼1
-
1
µg mL ) (see SFigure 6 in the Supporting Information),
with no statistically significant differences in the EC50
between the three strains and two coreceptor tropisms.
Determining the concentration of total benzoboroxole in
the polymer solution (polymer-bound 1) provides a means
to compare the binding of the oligomers and polymer
-
1
antiviral activity. A 1.0 mg mL solution of 625 contains
-
1
0
.588 mM of polymer-bound 1 compared to a 1.0 mg mL
solution of oligo-625 that has 0.447 mM of oligomer-bound
. Solutions of the polymers and oligomers with the same
1
feed ratio of 4 contained similar concentrations of bound 1.
However, the oligo-625 was on average composed of only a
few benzoboroxole moieties per chain whereas the polymers
had as many as ∼150 benzoboroxole moieties per chain. This
may account for the weak affinity of oligo-625 for the
immobilized gp120 (K ) 8.5 mM, concentration units in
D
terms of bound 1) as determined by SPR. However, the 25
mol % benzoboroxole functionalized polymer (625) had an
EC50 against HIV-1 that corresponds to a polymer-bound 1
concentration of ∼200 nM (see SFigure 6 in the Supporting
Information). It is interesting to note that the concentration
of polymer-bound 1 is 125,000 fold lower than the 25 mM
glucose concentration present in the DMEM media, indicat-
ing that glucose is not out competing the benzoboroxole
groups’ affinity for the terminal diol residues on gp120
N-linked glycans in the antiviral assay measurements.
Cytotoxicity Assay. Investigations of the individual
polymer cytotoxic profiles were explored using Vk2/E6E7
Figure 4. Results from the single-cycle HIV-1 infectivity
inhibition in TZM-bl cells of polymer solutions 625, 650, and
6
75 against (a) R5 DU156 (Clade B), (b) R5 TRO (Clade
C), isolated from acute sexually transmitted infections, and
c) the pseudotyped X4 WEAU (Clade B). The
nonfunctionalized backbone, 6 , showed antiviral activity at
(
0
-1
the highest concentration tested, 20 mg mL , but at no
other concentrations tested. N ) 3, mean ( SD.
activity, with a 50% effective concentration for inhibiting
viral activity (EC50) of approximately 1 mg mL- (13,000
1
126 MOLECULAR PHARMACEUTICS VOL. 7, NO. 1

MultiValent Benzoboroxole Polymer gp120 Entry Inhibitor
articles
Discussion
In this study we have shown that the small molecule
benzoboroxole, 1, exhibited weak affinity to gp120 by SPR.
In accord with other investigations using boronic acid
14
moieties for sensing diols the affinity increased at alkaline
pH likely due to further stabilization of a tetrahedral
boronate-diol complex above the molecules pK
a
of 7.2
suggesting that the mechanism of polymer binding is
1
6
benzoboroxole-sugar mediated.
The SPR of 1 at pH 7.5 and pH 9.5 indicated that the
number and affinity to the binding sites changes as a function
of pH. This result may likely be explained by the pH sensitive
affinity displayed by arylboronic acids. In general arylboronic
acids exhibit increasing affinity above pH 7 for most sugar
4
1
moieties except sialic acid, a terminating moiety on some
of gp120s complex glycans. Sialic acid binds to the non-
functionalized phenylboronic acid at acidic pH, with decreas-
Figure 6. Results from cytotoxicity assay of polymer
solutions 6 , 625, 650, and 675 in Vk2/E6E7 human
4
1
ing affinity above pH 7. While the affinity of 1 has not
0
vaginal cells after 24 h exposure. N ) 3, mean ( SD.
been tested for sialic acid or for hexopyranosides above pH
7
.8 yet, it may display a similar affinity profile as phenyl-
human vaginal cells (Figure 6). Nonoxynol-9 (N-9) was
selected as a toxic control as it has demonstrated cytotoxicity
in in vitro studies and in vivo clinical studies and was
boronic acid. Thus at pH 7.5, 1 may have a weak affinity
for the terminal nonreducing mannose, galactose and sialic
acid providing the SPR signal that indicates a large number
of binding sites with weak affinity. At pH 9.5, 1 may then
exhibit a higher binding affinity for only the terminal
nonreducing mannose and galactose while the affinity for
sialic acid dramatically drops. A dramatically weak affinity
for sialic acid would decrease the number of binding sites
while an increase in affinity for mannopyranoside and
galactopyranoside due to the more alkaline pH would yield
an SPR signal indicating a higher affinity. This would provide
one explanation for the apparent low response signal,
indicating fewer binding sites, with higher affinity at pH 9.5
compared to the greater response signal with weaker affinity
displayed at pH 7.5 by the SPR signal of 1 binding to the
gp120 functionalized surface.
3
8
39
tested at concentrations ranging from 0.025-0.00125 mg
-1
mL resulting in 40% cell viability at the lowest concentra-
tion tested. The unfunctionalized pHPMAm, 6 , polymer was
tested as the negative control due to a history of noncyto-
0
2
7
toxicity in in vitro assays. A dose profile identical to that
performed in the antiviral study was conducted. 6 exhibited
no cytotoxicity up to 20 mg mL . 625 exhibited cytotoxicity
0
-1
-
1
at 20 mg mL with cell viability near 80%. Concentrations
higher than 0.18 mg mL- for 650 and 0.4 mg mL for 675
could not be tested due to gelation of the polymer after
addition to the cells that impeded readouts in the MTS assay.
Greater than 90% cell viability occurred in the presence of
1
-1
-
1
6
50 up to 0.18 mg mL in the vaginal cell line, two orders
of magnitude higher than its EC50 against all three viral
strains. Monolayer cell based toxicity assays are more
sensitive than tissue based assays that use normal stratified
epithelia found in ex vivo and in vivo tissue studies. The
lack of cytotoxicity in this assay suggests that these polymers
may have minimal toxicity in vivo.
Our results complement those of B e´ rub e´ et al. by
demonstrating affinity of benzoboroxoles for glycoprotein
diols at physiological pH. The synthetic incorporation of the
benzoboroxole moiety by free radical polymerization of a
benzoboroxole containing vinylic monomer with the water-
soluble and nontoxic HPMAm monomer created multivalent
oligomers. The 25 mol % functionalized oligomer, oligo-
4
0
(
38) Krebs, F. C.; Miller, S. R.; Catalone, B. J.; Fichorova, R.;
Anderson, D.; Malamud, D.; Howett, M. K.; Wigdahl, B.
Comparative in Vitro Sensitivities of Human Immune Cell Lines,
Vaginal and Cervical Epithelial Cell Lines, and Primary Cells to
Candidate Microbicides Nonoxynol 9, C31g, and Sodium Dodecyl
Sulfate. Antimicrob. Agents Chemother. 2002, 46 (7), 2292–2298.
39) Stafford, M. K.; Ward, H.; Flanagan, A.; Rosenstein, I. J.; Taylor-
Robinson, D.; Smith, J. R.; Weber, J.; Kitchen, V. S. Safety Study
of Nonoxynol-9 as a Vaginal Microbicide: Evidence of Adverse
Effects. J. Acquired Immune Defic. Syndr. Hum. RetroVirol. 1998,
6
25, resulted in an approximate 10-fold improved of affinity
for gp120 over the small molecule 1. Incorporation beyond
5 mol % to 50 mol % rapidly decreased solubility of the
2
resulting oligomer and prevented us from measuring affinity
as a function of benzoboroxole incorporation in the oligomers.
Exposure to seminal sugar concentrations of fructose (16
mM) and glucose (6 mM) efficiently reduced binding of
oligo-625 to surface-bound gp120. Additionally, the reduced
binding affinity that occurred in the presence of increasing
(
(
1
7 (4), 327–331.
40) Catalone, B. J.; Kish-Catalone, T. M.; Neely, E. B.; Budgeon,
L. R.; Ferguson, M. L.; Stiller, C.; Miller, S. R.; Malamud, D.;
Krebs, F. C.; Howett, M. K.; Wigdahl, B. Comparative Safety
Evaluation of the Candidate Vaginal Microbicide C31g. Antimi-
crob. Agents Chemother. 2005, 49 (4), 1509–1520.
(41) Otsuka, H.; Uchimura, E.; Koshino, H.; Okano, T.; Kataoka, K.
Anomalous Binding Profile of Phenylboronic Acid with N-
Acetylneuraminic Acid (Neu5ac) in Aqueous Solution with
Varying pH. J. Am. Chem. Soc. 2003, 125 (12), 3493–3502.
VOL. 7, NO. 1 MOLECULAR PHARMACEUTICS 127

articles
Jay et al.
concentration of fructose, sialic acid, methyl R-D-mannopy-
ranoside or methyl ꢀ-D-galactopyranoside demonstrates that
the interaction of benzoboroxole with diols is responsible
for the oligomer’s affinity to the glycoprotein. The presence
of sialic acid reduced binding more extensively than either
methyl R-D-mannopyranoside or methyl ꢀ-D-galactopyrano-
side indicating that this terminal residue on the complex
N-linked glycans of gp120 impacts affinity. Additionally, the
reduced affinity of the benzoboroxole-functionalized oligo-
mer in the presence of sialic acid likely indicates that sialic
acid may have a higher affinity for benzoboroxole compared
to other two pyranoside sugar moieties tested. While the
small molecule benzoboroxole had similar affinity for methyl
compared to 625. Increasing the degree of functionalization
from 50 to 75 mol % likely displays only a gradual increase
in activity owing to an improved probability of binding
events due to the increased number of benzoboroxole ligands
present on the polymer backbone.
Like CV-N the synthetic CBA behavior of the ben-
zoboroxole-functionalized polymers demonstrated similar
activity against the two tested clades and the R5 and X4
coreceptor tropisms of HIV-1 suggesting that these polymers
could have topical broad spectrum activity across a breadth
of sexually transmitted viral strains. This level of activity
against both coreceptor tropisms was not exhibited by the
anionic sulfate polymers initially tested in microbicides,
which displayed a higher activity against the X4 tropic strains
1
6
R-D-mannopyranoside and methyl ꢀ-D-galactopyranoside,
a more rapid reduction of affinity occurred in the presence
of methyl R-D-mannopyranoside compared to methyl ꢀ-D-
galactopyranoside. It may be that the difference in structure
between the high mannose and complex N-lined glycans
impacts affinity. The polymer-bound concentration of ben-
zoboroxole in a microbicide formulation will thus need to
take into account competing glycoprotein and monosaccha-
ride concentrations present in the vaginal lumen and seminal
plasma to maintain activity against HIV-1. Further testing
of the antiviral activity in the presence of seminal plasma
will need to be undertaken to further understand and address
these concerns.
43
compared to the R5. The N-linked glycosylated sites are
conserved to a large extent between different viral isolates
potentially providing a mechanism for the observed broad
44
spectrum activity.
The EC50 of CN-V against HIV-1 R5 Clade B (BR/92/
0
03) and C (ZA/97) isolates in TZM-bl cells was 3.9 nM
-
1
45
and 15.6 nM (0.04 to 0.17 µg mL ) respectively. These
values compare favorably to the EC50 of ∼10 nM for 675
-
1
(∼1 µg mL ) against the R5 Clade B and C HIV-1 strains
we screened. The antiviral activity of these polymers in the
TZM-bl luciferase reporter gene assay indicates that they
act as entry-inhibitors although we have not conclusively
determined whether they prevent binding to CD4 or the
CRR5 coreceptor or both. Based on these results we have
thus demonstrated the successful use of multivalent-ben-
zoboroxoles to create a polymer based synthetic CBA entry
inhibitor with broad spectrum activity and similar efficacy
in Vitro as CN-V. We also believe that linear polymer-based
CBA produced by free radical polymerization can more likely
be economically scaled up and formulated. Even considering
the recent breakthrough in the production of griffithsin, a
protein based CBA, which required a 5,000 square-foot green
We suspect that the capacity of benzoboroxole to bind
terminal hexopyranosides present on the N-linked glycosy-
lated sites of gp120 as demonstrated by SPR provides the
broad spectrum entry inhibition exhibited by the ben-
zoboroxole-functionalized polymers against the two clades
and two coreceptor tropic strains of HIV-1 tested. The
dramatic increase in activity from 625 to 650 may occur in
part due to the pattern and orientation of the benzoboroxole
moiety on the polymer chain as well as antagonistic binding
with glucose and glycoproteins from the fetal bovine serum
1
6
present in the DMEM media. However, 650 revealed
activities against all three strains even in the presence of a
large excess of glucose compared to the concentration of
polymer-bound 1, indicating that an increase in avidity may
occur due to incorporation of 50 mol % benzoboroxole
compared to the 25 mol % benzoboroxole-functionalized
polymer. Thus the increased multivalent presentation of the
benzoboroxole ligand in 650 appears to improve the affinity
4
6
house to produce 60 g of recombinant protein, industrial
scale up for polymer synthesis has been readily accomplished
on a multi-kilogram scale. These polymers may therefore
act as a potential broad-spectrum entry inhibitor at economi-
(
(
42) Kaur, G.; Fang, H.; Gao, X.; Li, H.; Wang, B. Substituent Effect
on Anthracene-Based Bisboronic Acid Glucose Sensors. Tetra-
hedron 2006, 62 (11), 2583–2589.
43) Moulard, M.; Lortat-Jacob, H.; Mondor, I.; Roca, G.; Wyatt, R.;
Sodroski, J.; Zhao, L.; Olson, W.; Kwong, P. D.; Sattentau, Q. J.
Selective Interactions of Polyanions with Basic Surfaces on
Human Immunodeficiency Virus Type 1 gp120. J. Virol. 2000,
for gp120 compared to 625
.
The low polymer concentration required to inhibit HIV-1
entry for 650 and 675 may be due to random incorporation of
1
into the polymer backbone during free radical polymeri-
zation that yields a diverse library of spatial arrangement
between two or more benzoboroxole moietiessout of the
more than 100 that are present in an average chainsthat
augment affinity for unique geometric conformation of the
terminal residues of the N-linked glycans on gp120. The
nature and length of linker separating two arylboronic acid
moieties has demonstrated great influence in the boronic acid
7
4 (4), 1948–1960.
(44) Willey, R. L.; Rutledge, R. A.; Dias, S.; Folks, T.; Theodore, T.;
Buckler, C. E.; Martin, M. A. Identification of Conserved and
Divergent Domains within the Envelope Gene of the Acquired
Immunodeficiency Syndrome Retrovirus. Proc. Natl. Acad. Sci.
U.S.A. 1986, 83 (14), 5038–5042.
45) Buffa, V.; Stieh, D.; Mamhood, N.; Hu, Q.; Fletcher, P.; Shattock,
R. J. Cyanovirin-N Potently Inhibits Human Immunodeficiency
Virus Type 1 Infection in Cellular and Cervical Explant Models.
J. Gen. Virol. 2009, 90 (Part 1), 234–243.
(
4
2
binding selectivity and affinity and may account in some
degree for the 3-fold increase in antiviral activity of 650
128 MOLECULAR PHARMACEUTICS VOL. 7, NO. 1

MultiValent Benzoboroxole Polymer gp120 Entry Inhibitor
articles
5
3
cal prices required for microbicide distribution in resource-
poor pandemic regions.
resistant to other entry inhibitors. These results may be
pertinent for investigation of future constructs targeted toward
systemic delivery for treatment of HIV-1 infection.
The use of benzoboroxole binding to terminal nonreducing
hexopyranoside residues has several potential benefits for
use as a microbicide entry inhibitor. First, they may be
capable of targeting other glycosylated sexually transmitted
Our investigation into cytotoxicity of the benzoboroxole-
functionalized polymers against a human vaginal cell line
revealed that no gross cytotoxicity occurred for the 50 mol
% functionalized polymer (650) at 2 orders of magnitude
above the EC50. Testing at higher concentrations was
prevented due to gel formation upon polymer interaction with
the cell media. Cell-based cytotoxicity assays are sensitive
indicators of topical toxicity in an intact vaginal epithelia,
therefore suggesting that these polymers may be safe.
However, CBAs have the potential to induce inflammatory
4
7
virions like herpes-simplex virus against which several
4
8
protein CBA are inactive, and the human papillomavirus.
Protein CBAs have also demonstrated the potential to block
the binding of HIV-1 to disseminating cells like DC-SIGN-
4
9
expressing dendritic cells which transfer HIV-1 to T-cell
lymphocytes as a dissemination pathway in the male-to-
5
0
female heterosexual transmission of HIV-1. Similarly the
benzoroboxole based synthetic CBA may also be capable
of preventing this dissemination pathway of HIV-1. Also,
recent research into the drug resistance profiles of CBAs like
CV-N demonstrate that these compounds select for mutant
HIV-1 strains that predominantly contain deleted N-glyco-
5
3
activity and to stimulate various differential markers
warranting further investigation into the cellular response
induced by these polymers. We envision these polymers
delivered to the vaginal lumen, rather than systemically, and
that they would likely not be absorbed by the tissue
potentially preventing adverse side effects.
5
1-53
sylated sites on the envelope gp120.
These deletions
expose previously hidden immunogenic epitopes and promote
the possibility of mutant strains that are more susceptible to
immunogenic attack. These mutant strains were also not
Conclusion
The strategy presented herein has potential advantages over
current protein CBAs. Specifically we have demonstrated
the potential of a polymer based synthetic lectin to act as an
entry inhibitor against HIV-1. Incorporation of benzoborox-
oles into a multivalent polymer architecture improved affinity
to gp120 by a factor of 10. Incorporation of benzoboroxole
moieties into polymers with approximately ten times the
molecular weight of the oligomers tested in the SPR assay
generated antiviral activity across two clades and two
corecepter tropisms of HIV-1 with EC50’s. The activity
increased by 3 orders of magnitude from a 25 mol %
benzoboroxole-functionalized polymer to a 75 mol % ben-
zoboroxole-functionalized polymer with an EC50 of ∼10 nM.
We envision that benzoboroxole-functionalized polymers
with this or other polymer architectures will lead to a new
class of polymer-based entry inhibitors for use as vaginal
microbicides that will exhibit broad spectrum activity against
HIV-1, with economical production and ease of formulation.
(
46) O’Keefe, B. R.; Vojdani, F.; Buffa, V.; Shattock, R. J.; Montefiori,
D. C.; Bakke, J.; Mirsalis, J.; d’Andrea, A. L.; Hume, S. D.;
Bratcher, B.; Saucedo, C. J.; McMahon, J. B.; Pogue, G. P.;
Palmer, K. E. Scaleable Manufacture of HIV-1 Entry Inhibitor
Griffithsin and Validation of Its Safety and Efficacy as a Topical
Microbicide Component. Proc. Natl. Acad. Sci. U.S.A. 2009, 106
(15), 6099–6104.
(
47) Liljeqvist, J. A.; Svennerholm, B.; Bergstrom, T. Herpes Simplex
Virus Type 2 Glycoprotein G-Negative Clinical Isolates Are
Generated by Single Frameshift Mutations. J. Virol. 1999, 73 (12),
9
796–9802.
(
48) Bertaux, C.; Daelemans, D.; Meertens, L.; Cormier, E. G.; Reinus,
J. F.; Peumans, W. J.; Van Damme, E. J.; Igarashi, Y.; Oki, T.;
Schols, D.; Dragic, T.; Balzarini, J. Entry of Hepatitis C Virus
and Human Immunodeficiency Virus Is Selectively Inhibited by
Carbohydrate-Binding Agents but Not by Polyanions. Virology
2
007, 366 (1), 40–50.
(
49) Balzarini, J.; Van Herrewege, Y.; Vermeire, K.; Vanham, G.;
Schols, D. Carbohydrate-Binding Agents (Cba) Efficiently Prevent
Dc-Sign-Directed Hiv-1 Transmission to T-Lymphocytes. Mol.
Pharmacol. 2006, 71 (1), 3-11.
Acknowledgment. We thank David Montefiori (Duke
University) for providing lab space, viral strains and the
TZM-bl cell line for the neutralization studies. We also thank
Michael Kay (University of Utah) for his assistance in
experimental design and for reviewing the manuscript, and
Rajan Kochambilli for help with the synthesis of the
oligomers. This work was supported by the NIH, Grant No.
R21-AI062445 (P.F.K), NSF Graduate Research Fellowship
(
50) Nguyen, D. G.; Hildreth, J. E. Involvement of Macrophage
Mannose Receptor in the Binding and Transmission of HIV by
Macrophages. Eur. J. Immunol. 2003, 33 (2), 483–493.
(
51) Hu, Q.; Mahmood, N.; Shattock, R. J. High-Mannose-Specific
Deglycosylation of HIV-1 gp120 Induced by Resistance to
Cyanovirin-N and the Impact on Antibody Neutralization. Virology
2
007, 368 (1), 145–154.
(
52) Balzarini, J.; Van Laethem, K.; Hatse, S.; Froeyen, M.; Peumans,
W.; Van Damme, E.; Schols, D. Carbohydrate-Binding Agents
Cause Deletions of Highly Conserved Glycosylation Sites in HIV
gp120: a New Therapeutic Concept to Hit the Achilles Heel of
HIV. J. Biol. Chem. 2005, 280 (49), 41005–41014.
53) Balzarini, J.; Van Laethem, K.; Peumans, W. J.; Van Damme, E. J.;
Bolmstedt, A.; Gago, F.; Schols, D. Mutational Pathways, Resistance
Profile, and Side Effects of Cyanovirin Relative to Human Immu-
nodeficiency Virus Type 1 Strains with N-Glycan Deletions in Their
gp120 Envelopes. J. Virol. 2006, 80 (17), 8411–8421.
(
(
B.E.L), and the University of Utah, Graduate Fellowship
J.I.J.).
Supporting Information Available: Complete profile of
raw SPR responses for 1 and oligomers binding to gp120 HIV-
(
-1
1
BaL functionalized surfaces; EC50 in µg mL and polymer-
bound 1 (nM). This material is available free of charge via the
Internet at http://pubs.acs.org.
MP900159N
VOL. 7, NO. 1 MOLECULAR PHARMACEUTICS 129