Received: 10 August 2016 Revised: 2 August 2017 Accepted: 7 September 2017

DOI: 10.1002/rmv.1952

REVIEW

Systematic screening of viral entry inhibitors using surface

plasmon resonance
Penmetcha K.R. Kumar

National Institute of Advanced Industrial

Science and Technology, Tsukuba City, Ibaraki,
Japan
Correspondence
Dr. P. K. R. Kumar, Biomedical Research
Institute, National Institute of Advanced
Industrial Science and Technology, Central 6,
1‐1‐1 Higashi, Tsukuba City 305‐8566, Ibaraki,
Japan.
Email: pkr‐kumar@aist.go.jp
Funding information
National Institute of Advanced Industrial Sci-
ence and Technology (AIST)
Summary
Viral binding and entry into host cells for various viruses have been studied extensively, yielding a
detailed understanding of the overall viral entry process. As cell entry is an essential and requisite
process by which a virus initiates infection, it is an attractive target for therapeutic intervention.
The advantages of targeting viral entry are an extracellular target site, relatively easy access for
biological interventions, and lower toxicity. Several cell‐based strategies and biophysical tech-
niques have been used to screen compounds that block viral entry. These studies led to the dis-
covery of inhibitors against HIV, HCV, influenza, Ebola, and RSV. In recent years, several
compounds screened by fragment‐based drug discovery have been approved as drugs or are in
the final stages of clinical trials. Among fragment screening technologies, surface plasmon reso-
nance has been widely used because it provides accurate information on binding kinetics, allows
real‐time monitoring of ligand‐drug interactions, requires very small sample amounts to perform
analyses, and requires no modifications to or labeling of ligands. This review focuses on surface
plasmon resonance–based schemes for screening viral entry inhibitors.

KEY W ORDS

screening inhibitors, surface plasmon resonance, viral entry

1 | I N T RO D U CT I O N enveloped viral proteins undergo a series of conformational changes

and bind either to the same host receptors or a different set of

1.1 | Viral entry host receptor proteins. After these changes are triggered in both the
membrane proteins of the virus and the host, the viral and host mem-
Viral envelope proteins initiate an important process by which a virus
branes undergo further structural changes, such as deformation,
enters a host cell, referred to as viral entry. Viral entry is a complex,
which force the 2 membranes into close proximity and result in the
multistep process that involves the remarkable choreography of spe-
formation of a stalk followed by hemifusion intermediates and the
cific and sequential interactions between enveloped viral proteins
formation of a fusion pore. Representations of membrane fusion pro-
and receptor proteins in host cells. The process can be separated into
cesses (sequentially shown as pre‐fusion, pre‐hairpin, collapse of
2 steps: (a) viral envelope protein specific binding to host cell recep-
intermediate, and post‐fusion processes) that are mediated by viral
tors Figure 1A and B activation and membrane fusion. Each of these
surface proteins are shown in Figure 1B. The host and viral proteins
steps contributes variedly to the viral entry process. Viral surface pro-
involved in the entry steps of commonly known viruses are listed in
tein binding to cognate receptors on the host cell allows the accumu-
Table 1. Upon the completion of membrane fusion, the viral genome
lation of viruses at the host cell surface, which in turn allows the virus
enters the cytoplasm in complex with viral proteins such as
to recognize receptors on the host cell that allow entry. This step is
nucleocapsids.
followed by viral fusion, wherein the same or a different set of
During the process of viral entry, viruses encounter the
challenge of penetrating a cell and releasing their viral genomes
Abbreviations used: CCR5, chemokine receptor type 5; gp41, glycoprotein 41; inside it, which initiates the infection cycle. Targeting the viral entry
GPCRs, G‐protein–coupled receptors; HA, hemagglutinin; HVEM, herpes virus step is an attractive strategy for therapeutic intervention, not only
entry mediator; NTA chip, nitrilotriacetic acid chip; NMR, nuclear magnetic
resonance; RU, response unit; SA chip, streptavidin chip; SPR, surface plasmon
because the target site is extracellular and easy to access for biolog-
resonance ical interventions but also because therapeutics targeting viral entry

Rev Med Virol. 2017;e1952. wileyonlinelibrary.com/journal/rmv Copyright © 2017 John Wiley & Sons, Ltd. 1 of 12
https://doi.org/10.1002/rmv.1952
2 of 12
FIGURE 1 A schematic view of viral protein binding to host receptor and membrane fusion mediated by viral proteins (hemagglutinin, HA). A, Viral
surface protein binding to the host cell receptor. B, Viral protein initiates membrane fusion process with host cell membrane
are likely to be less toxic as they do not require to permeate the
membrane. Therapeutic interventions that block viral entry can also
prevent damage to cellular machinery that commonly occurs during
viral replication in the host cell. Viral entry inhibitors could play an
important role as topical microbicides or prophylactic drugs to pre-
vent virus transmission between individuals and promote the con-
tainment of viral infections. The best examples of such viral entry
inhibitors are 2 approved drugs for HIV treatment, enfuvirtide
(Fuzeon) and maraviroc (Selzentry); the former targets the viral glyco-
protein 41 (gp41), and the latter targets the host CCR5 coreceptor.
Enfuvirtide binds to the HR1 region of gp41 and prevents HR1 and
HR2 from folding properly, which in turn prevents the conforma-
tional change of gp41 required to complete the fusion process.
Maraviroc binds selectively to the CCR5 coreceptor and inhibits its
interaction with the V3 loop region of glycoprotein 120, thereby
blocking the fusion of cellular membranes. With these new classes
of antivirals targeting the entry step, along with previously available
antivirals that target reverse transcriptase, protease, and integrase
enzymes, it is possible to contain HIV epidemics using combination
antiretroviral therapy, transforming HIV from a nearly always‐fatal
disease into a medically managed scenario. These success stories
against HIV provide momentum to the search for antiviral com-
pounds targeting the viral entry step of HIV and other enveloped
KUMAR
viruses. Some promising compounds are in active clinical trials and
are listed in Table 2.
1.2 | Drug discovery strategies
Pharmaceutical companies routinely use high‐throughput screening
strategies to identify drug candidates. However, false positives are a
common problem, yielding candidate drugs that bind sub‐optimally to
the target or lack other important medicinal properties such as solubil-
ity, absorption, and bioavailability.1 To circumvent these limitations in
the drug discovery process, fragment‐based drug discovery is com-
monly used, a technique in which fragments or a set of fragments bind
to the target of interest (purified proteins). This strategy has identified
several compounds that are either currently approved drugs or are in
the final stages of clinical trials.2,3 In fragment‐based drug discovery,
several biophysical techniques are used to evaluate ligand‐drug inter-
actions, including nuclear magnetic resonance (NMR),4,5 X‐ray crystal-
lography,6 mass spectroscopy,7 isothermal titration calorimetry,8
protein thermal shift,9 affinity capillary electrophoresis,10 weak and
other affinity chromatography,11 and label‐free biosensor methods.12
In contrast to cell‐based methods such as recombinant reporter virus
systems,13-16 virus infection reporter cell systems,17-21 pseudotyped
virus systems,22-25 or virus free cell‐cell fusion systems,26,27
KUMAR 3 of 12

TABLE 1 Viral proteins and host receptors involved in the viral entry steps of different viruses

Virus Attachment Stage Activation and Fusion Stage

Viral Proteins Host Proteins Viral Proteins Host Proteins

HSV gC, gB Heparin sulfate gD, gB, gH/gL Nectin1, Nectin2

Herpes virus entry mediator, 3O‐S
HCV E1, E2 Heparin sulfate, LDL E2 Paired immunoglobulin‐like type‐2
receptor
alpha, CD81, scavenger receptor class
B
member 1, Claudin‐1, Occludin

EV71 VP1 Sialic acid VP1, VP (4) Scavenger receptor class B member,
P‐selectin glycoprotein ligand‐1

HIV‐1 gp120 CD4 gp120, gp41 Chemokine (C‐C Motif) receptor

5/chemokine (C‐X‐C Motif) receptor 4

Influenza HA N‐Acetylneuraminic acid HA N‐Acetylneuraminic acid

DC‐SIGN
Marburg and Ebola Phosphotidylserine, C‐type lectin, phosphotidylserine GP, Niemann‐Pick Disease type
GP receptors phosphotidylserine C1, C‐type lectins
Receptor family (TIM and TAM)
SARS‐CoV S Trans membrane protease serine 2, S Trans membrane protease serine 2
ACE2
Hendra/Nipha Cedar G EphrinB2, EphrinB3 F
Dengue/ E Heparin sulfate, DC‐SIGN, E Heparin sulfate, DC‐SIGN,
West Nile/ Mannose receptor, Mannose receptor,
Japanese encephalitis HSP90/HSP70, Tim/TAM HSP90/HSP70, TIM/TAM

DC‐SIGN, dendritic cell‐specific intracellular adhesion molecule‐3‐grabbing non‐integrin; HSP70/90, heat shock protein 70/heat shock protein 90; TAM,
constitutes 3 proteins (Tyro3, Axl, and Mer); TIM, T‐cell immunoglobulin and mucin.

TABLE 2 Compounds that inhibit viral entry and are in active clinical biophysical methods enable the performance of assays in cell‐free
trials environments. Among commonly used biophysical methods, surface
Virus Inhibitor Action Against the Entry Stage plasmon resonance (SPR) has emerged as one of the most widely used
fragment screening technologies because it provides accurate informa-
HIV BMS‐488043 Binds to region with CD4
BMS‐663068 Binds to the pocket of gp120 tion on binding kinetics. SPR also allows the real‐time monitoring of
Ibalizumab Interfere with CD4 binding
ligand‐drug interactions, requires very small amounts of sample for
to the gp120 to coreceptors
(CCR5 and CXCR4) analyses (usually micrograms), allows the simultaneous use of refer-
Aplaviroc Binds to CCR5 and block ence ligands, and more importantly, requires no modifications or ligand
interaction with gl120
Maraviroc Binds to CCR5 and block labeling. The use of SPR in drug discovery has been reviewed,28-30 and
interaction with gl120 its applications in the primary and secondary screening of drug leads
Vicriviroc Binds to CCR5 and block
interaction with gl120 against various targets have been described.31-35 These studies sug-
Cenicriviroc Binds to CCR5 and block gest the potential of SPR‐based approaches to the screening and iden-
interaction with gp120,
tification of viral entry inhibitors.
additionally binds and
inhibits CCR2
AMD‐3100 Binds to CXCR4 and block
interaction with gp120 1.3 | SPR platform and analyses of biomolecular
Enfuvirtide Binds to HR1 and HR2 interactions
region of gp41
HCV Lactoferrin Prevents attachment In an SPR system, one interacting biomolecule is typically immobilized
Ant‐CD81mAbs Binds to CD81 on the surface of a sensor using a matrix such as dextran. Beneath
Anti‐SRB1mAb Binds to SRB1
ITX5061 Binds to SRB1 the dextran surface, a thin gold layer is attached to a glass surface
Erlotinib Binds to EGFR Figure 2A. The composite surface is placed within a flow cell, and a
Ezetimibe Binds to NPC1L1
cognate biomolecule (analyte) that interacts with an immobilized bio-
Influenza Sialidase Remove sialic acid and thus
reduce the viral attachment molecule is injected into the flow cell in the presence of a suitable
EBOV Cyanovirin‐N Binds to the carbohydrate buffer. Surface plasmons are generated when an incident beam of light
molecules of GP and prevents is directed toward the gold surface at a critical angle. The critical angle
Binding to the host receptor
depends on the refractive index near the surface, which changes when
RSV Motavizumab Binds to the F protein
an analyte binds to the immobilized biomolecule Figure 2A. The
4 of 12
FIGURE 2 Surface plasmon resonance biosensing platform. A, SPR
biosensing system. B, Sensogram response observed upon ligand
interaction with immobilized biomolecules
resulting change in refractive index at the surface is measured, and the
extent of that change correlates with the mass concentration near the
surface; this is displayed as a plot (sensorgram) of resonance units RU;
Figure 2B. The following 3 phases are observed in the sensorgram
when an analyte is injected on the immobilized protein surface: (1) an
association phase, in which the analyte binds to the immobilized pro-
tein during the injection process; (2) an equilibrium phase, which
occurs at the end of the analyte injection process and in which the ana-
lyte binding rate reaches a steady state; and (3) a dissociation phase
that is observed when the analyte dissociates from the immobilized
protein. This occurs during the continuous flow of buffer over the
immobilized surface Figure 2B. Importantly, all of these events can
be monitored in real time. The residual analyte remaining on the sensor
surface can be removed, and the surface can be regenerated without
damaging the activity of the immobilized protein. The regeneration of
an active ligand surface, including the removal of bound analytes from
previous screens and the maintenance of ligand activity, is crucial,
especially in high‐throughput screens, as the ligand surface must be
used multiple times in each successive round of screening. There are
2 common approaches to stably immobilizing ligands on the sensor
chip surface: covalent coupling and noncovalent capturing. In the for-
mer, the sensor surface is coated with different functional groups that
react with an appropriate functional group from the ligand to form sta-
ble linkages. In the latter, capture involves the use of high‐affinity
recombinant tags or antibodies Figure 3. In the covalent method of
immobilization of ligand, all NH2 functional groups of amino acids on
the surface of proteins (ligands) are accessible for amino coupling reac-
tion, and thus, it suffers from the random orientation of immobilized
ligands and the possibility that some ligands will lose their activity
under the coupling conditions Figure 3B. The noncovalent method
for immobilizing the ligand uses either fusion protein (His‐tag or
KUMAR
FIGURE 3 Strategies for the immobilization of viral or host proteins
on the sensor chip. A, The covalent strategy (such as amino coupling
reaction). The noncovalent strategies includes B, affinity tag‐mediated
strategy (such as His‐tagged or GST fusion proteins). C, Antibody‐
mediated strategy
GST‐Tag proteins) or ligand‐antibodies Figure 3B,C. This method has
the advantages of achieving properly oriented immobilized ligands
and the means to efficiently remove immobilized ligands and their
binding partners to regenerate a clean surface for the next round of
screening. However, this strategy faces other issues such as high ligand
consumption and the need to prepare tagged ligands. Thus, the selec-
tion of an immobilization method must be done carefully with consid-
eration of the strategy selected for the fragment‐based screening of
drug leads. In addition to the immobilization methods used, assay for-
mat is important when considering the primary or secondary screening
of drug leads.
After the ligand is immobilized on the sensor chip, it is important
to consider the assay format. Both direct and indirect assays are com-
monly used in SPR screens. In direct assays, the ligand is immobilized
on the sensor surface, and analytes (drug leads) are injected into the
flow cell over the sensor surface Figure 4A. The binding of an analyte
to the ligand causes a variation in the refractive index and provides an
analytical signal (RU) according to the concentration of mass at the sur-
face. These changes are detected in real time using surface plasmon
phenomena and are represented as a sensorgram, a plot showing
SPR responses over time. These plots are generated during the entire
biomolecular interaction process, from association through dissocia-
tion Figure 4A. All of the response curve data for different test com-
pounds obtained at different concentrations can be analyzed using
Biacore evaluation software after fitting with an appropriate binding
model to estimate binding kinetics (ka, association constant; kd, dissoci-
ation constant; and KD, equilibrium dissociation constant). Using this
approach, a number of target protein and drug interactions have been
evaluated, involving viral36,37 and human proteins.38,39 In indirect
assays, the ligand is immobilized on the sensor chip in a similar manner
to the direct assay. The advantage of an indirect assay is the ability to
evaluate whether the analyte is a competitor or an inhibitor that inter-
feres with ligand‐analyte interactions. Screening of compounds that
interfere or block interaction of ligand‐analyte can be evaluated on
KUMAR 5 of 12

FIGURE 4 Schematic views for the screening of inhibitors/competitors on the SPR platform. A, Schematic view of direct SPR analysis for analyzing
inhibitor/competitor's ability to block ligand‐analyte interactions. B, Schematic view of inhibitor/competitor indirect screening strategy where
inhibitor/competitor is injected over the surface of ligand‐analyte complex. C, Schematic view of inhibitor/competitor indirect screening strategy
where inhibitor/competitor is incubated with analyte before injecting over the immobilized surface of ligand
6 of 12
SPR platforms using 2 kinds of assays Figure 4B,C. In one format, ini-
tially, the analyte is injected over the immobilized ligand surface step
2 in Figure 4B and the response signal collected Figure 4B. If the
injected analyte is able to reduce the response signal, it is possible that
such ligands have the potential to block the ligand‐analyte interactions
step 3 in Figure 4B. The ligand surface can be regenerated by removing
both analyte or analyte‐competitor complex step 4 in Figure 4B. Alter-
natively, a competitor or inhibitor ligand (that competes with the
immobilized ligand) is incubated at equilibrium in solution with increas-
ing concentrations of analyte, and the resulting complexes are injected
into the sensing surface step 2 in Figure 4C. Only free analyte binds to
the immobilized receptor, providing a signal inversely proportional to
its concentration in the sample Figure 4C. The ligand surface can be
regenerated by removing both analyte or analyte‐competitor complex
step 3 in Figure 4C. A number of inhibitors have been analyzed using
indirect assays, including small molecules40 and macromolecules.41-46
1.4 | SPR platform for screening drug leads
An SPR‐based drug lead screening against chemokine receptor type 5
(CCR5) has been described. 47
A total of 200 compounds were
screened for binding to CCR5 from a compound chemical library using
a Biacore 4000 SPR machine. The fluidic system of the Biacore 4000
has 4 flow cells, each of which contains 5 detection spots, allowing
the analysis of at least 4 compounds with 1 as a reference. A tagged
monoclonal antibody against C‐9 (1D4) was covalently linked on the
CM4 sensor chip using amino‐coupling reactions followed by the cap-
ture of C‐9 tagged CCR5 receptor protein on the sensor chip of the
Biacore 4000 instrument. To screen drug‐lead compounds, flow cell
spots 1 and 5 were captured by the CCR5 receptor and spots 2 to 4
were used as reference spots Figure 547. All 200 compounds selected
for CCR5 binding evaluation likely overlap with the maraviroc binding
site (an efficient binder of CCR5); therefore, an additional reference
control, CCR5 blocked with maraviroc, was used. A clear difference
in binding modes was observed between active, blocked CCR5, and
the control flow cell spots, which facilitates the identification of both
orthosteric and allosteric compounds that interact with the CCR5
receptor. Each compound was analyzed using 3 concentrations (0.1,
1.0, and 10μM) and takes less than 2 minutes per run. This study
KUMAR
clearly shows that native receptors can be efficiently immobilized on
the sensor chips and facilitates screening of inhibitors compared with
other expensive and labor‐intensive biophysical methods.47
1.5 | SPR platform for screening viral entry inhibitors
The activity and conformation of proteins on the surface of the sensor
during the screening of inhibitors is critical to the success of the
screening process. Invariably, assuring these conditions is tedious and
not straightforward. To address this issue, we report 2 assays used to
evaluate viral entry inhibitors: one is used for proteins sensitive to
regeneration conditions (eg, the influenza hemagglutinin (HA) protein)
and the other is suitable for use with proteins resistant to regeneration
conditions (eg, the HSV‐1 gD protein).
1.5.1 | Influenza virus
Hemagglutinin is a homotrimer glycoprotein composed of HA1 and
HA2 disulfide‐linked polypeptide chains and expressed on the influ-
enza virus membrane. A key step in the process of infection and
transmission is the recognition of the HA of influenza viruses by
the host cell surface via terminal sialic acid (Sia), with α2‐3, α2‐6,
or both linkages. The glycan‐binding domain is located within the
HA1 globular domain. HA derived from avian‐ and human‐adapted
influenza viruses appears to be specific for α2‐3 Sia glycans and
α2‐6 Sia glycans, respectively. Thus, to understand the potential abil-
ity for interspecies transmission, it is important to monitor and eval-
uate the receptor (glycan) binding preferences of HAs derived from
influenza A viruses, especially those originating from birds and swine.
Previously, agglutination and solid‐phase fetuin capture assays were
routinely used; however, both methods fail to provide detailed bind-
ing rates. A protocol for analyzing glycan‐HA interactions efficiently
and kinetically, based on SPR, was developed after considering that
glycan‐HA interactions are multivalent in nature.48 Structures of the
HA‐glycan complex reveal that each trimeric HA possesses 3 gly-
can‐binding sites, which are estimated to be approximately 5 nm
apart. A synthetic glycan that displays multiple glycans on its surface
more appropriately mimics the distribution of natural glycans on the
cell surface. Synthetic biotinylated multivalent glycans are available
from a commercial source, including 4 Sia glycan moieties located
FIGURE 5 Non‐covalent methods for
immobilizing proteins
KUMAR 7 of 12

at the distal end of a biotinylated tetravalent glycan Figure 6A, blue
box48. This tetravalent glycan, based on our building model, predicts
that the distance between the Sia glycan moieties is approximately
4 nm. We assume, under this scenario, that the biotinylated tetrava-
lent glycan would bind to HA monovalently but would capture other
HA trimers using the 3 remaining unbound Sia glycan moieties. Thus,
each biotinylated tetravalent glycan is expected to bind a maximum
of 4 HA trimers, resulting in a higher avidity than a monovalent
glycan. These synthetic multivalent glycans (tetravalent‐glycan α2‐3
Sia glycan) were immobilized on flow cells 3 and 4 Figure 6B, and
glycan‐free biotinylated compound was immobilized on flow cell 2
Figure 6B on a streptavidin chip (SA chip) before the injection of
HA protein into the flow cells. The glycan‐free compound Figure 6
B, flow cell 2 and reference surface flow cell 1 Figure 6B served as
negative controls. The HA used in this study was a recombinant pro-
tein expressed in insect cells and derived from the A/Vietnam/1203/

FIGURE 6 The HA of influenza protein interacts with synthetic glycan, mimicking the host cell surface. A, Biotinylated tetravalent glycan. B,
Schematic view of the surface of the sensor chip after immobilizing biotinylated glycans and analyte (HA) flow during analysis. C, A typical
sensorgram showing responses to the HA concentrations (10 to 160nM) in single‐cycle kinetics analyses
8 of 12 KUMAR

2004(H5N1) virus. Different concentrations of HA (10, 20, 40, 80,
and 160nM) were injected into the flow cells. A clear signal or
response (RU) was observed with increasing HA concentrations from
the glycans bearing α2‐3 Sia glycan linkages Figure 6C after
deducting the signal from the reference flow cell (flow cell 1). How-
ever, in the absence of glycans on a biotinylated compound with a
similar backbone, no signal was detected green line, Figure 6C. We
performed single‐cycle kinetics to avoid the complete regeneration
of the biosensor surface between dissociation and the next associa-
tion phase. Single‐cycle kinetics has been successfully used to ana-
49-
lyze different biomolecular interactions in SPR (Biacore) systems.
51
These preliminary studies showed that the synthetic multivalent
glycan is suitable for the analysis of the specificity of HA on a sensor
chip (SA chip) using a Biacore T100, as well as for deriving binding
rate constants.
Although the protocol described above is optimized for the analy-
sis of the glycan‐HA interaction, it requires additional optimization to
regenerate the SA chip surface without destroying streptavidin bind-
ing activity for further analyses. We tried various conditions for the
regeneration of the SA sensor chip but failed to obtain reproducible
glycan‐HA interaction results after regeneration. It is important to
develop an alternative protocol that allows the sensor chips to be used
repeatedly for glycan‐HA analysis, as well as for screening inhibitors.
To achieve this goal, GE Healthcare recently designed the Biotin
CAP chip, a sensor chip with reversibly immobilized biotinylated mol-
ecules that facilitate its regeneration. A schematic representation of
immobilization, glycan‐HA interactions, and regeneration of the sensor
surface is shown in Figure 7. The rate constants obtained for the gly-
can‐HAs (α2‐3 Sia glycan and HA derived from A/Vietnam/1203/
2004(H5N1) virus on the SA chip and the CAP chip are comparable,
suggesting that these interactions could be reproducible in SPR analy-
ses. We have successfully performed multiple rounds (at least 90) of
glycan‐HA analyses and regeneration without affecting the quality of
the data.48 Thus, we believe that the SPR‐based protocol described
above (Figure 7) is suitable not only for influenza surveillance to define
pandemic scenarios but also for the screening of synthetic glycans and
other compounds that may interfere with glycan‐HA interactions.48 As
a proof of principle, we used our selected aptamers (high‐affinity RNA
ligands isolated from a completely random pool of RNA by iterative
rounds of selection and amplification) that bind efficiently to the HA
of highly pathogenic influenza viruses such as H5N1 and H7N7. A
schematic representation of the competitive assay is shown in
Figure 8A. In this assay, a multivalent α2‐3 glycan was immobilized
on the streptavidin surface (SA) after the capture of the SA‐containing
reagent on the sensor chip CAP [(GE Healthcare), step 2]. On this gly-
can surface, the HA protein (80 nM) from A/Indonesia/05/
2005(H5N1) was injected into the flow cells to obtain an initial binding
response. We assumed that this HA binding response would decrease
if the HA was bound by the selected aptamer (8‐3). To test this sce-
nario, we mixed different concentrations (25, 50, 100, 200, and
400nM) of aptamer 8‐3 with HA from A/Indonesia/05/2005(H5N1)
and incubated this mixture for 1 hour before flowing it over the glycan
surface Figure 8B. We observed that the binding of HA to the glycan
surface decreased with increasing aptamer concentration Figure 8C.

FIGURE 7 Schematic flow of steps 1 to 4 in the regeneration cycles of glycan‐HA interactions on the CAP chip and observed responses, using a
Biacore T100. Initially, the SA‐ssDNA conjugate was immobilized on the CAP chip (step 1). Subsequently, a biotinylated multivalent glycan was
allowed to bind to this conjugate (step 2). Once this surface was generated, various concentrations of HAs derived from different viruses were
passed over this surface (step 3). Once the glycan‐HA interactions were completed, the whole surface could be regenerated by stripping the entire
complex (streptavidin‐biotin‐glycan‐HA), using the capture reagent (step 4)
KUMAR
FIGURE 8 Overview of analyses of inhibitors that interfere with the interaction between HA of the influenza virus and glycans, mimicking the host
surface. A, SPR analyses. B, Schematic view of HA interactions with glycan at different inhibitor (aptamer) concentrations. C, Schematic view of SPR
responses observed in the presence of different inhibitor (aptamer) concentrations
Using these response signals, the effective concentration of aptamer
8‐3 necessary for 50% inhibition (EC50) of HA‐glycan interactions
was calculated and was found to be in the nanomolar range
(EC50 ~ 25nM). These analyses suggest that the binding of this
aptamer to HA interferes efficiently with HA‐glycan interactions.
However, a complementary sequence failed to reduce HA‐glycan
48
interactions. Similarly, compounds that interfere with HA‐glycan
interactions can be evaluated using the above protocol by program-
ming the analyses into either BiacoreT100 or Biacore 4000 machines.
If 2 flow cells are used for screening and 2 for controls, a total of 96
compounds can be screened per day by the Biacore T100 and 384
compounds by the Biacore 400, as each analysis cycle lasts approxi-
mately 30 minute. Thus, our described method can facilitate the
screening of compounds from chemical libraries.
1.5.2 | HSV‐1
The gD proteins of herpes simplex viruses play an important role in
viral entry by binding to specific cellular coreceptors and mediating
viral entry into host cells. The gD protein primarily recognizes 2 pro-
52 53
tein receptors, herpes virus entry mediator (HVEM) and nectin‐1,
which belong to the tumor necrosis factor family and the cell adhe-
sion molecules of the immunoglobulin superfamily, respectively. In
addition to these receptor proteins, HSV‐1 entry can be mediated
by 3‐O‐sulfated‐modified heparan sulfate.54 Structural studies of
gD‐HVEM and gD‐Nectin‐1 indicate that both the N‐ and C‐terminal
regions of the gD protein play important roles in receptor binding (on
9 of 12
the target cell) and in the membrane fusion process. The gD protein is
expressed on the viral surface and is an easily accessible target for
biological interventions, making gD an attractive target for the devel-
opment of therapeutics. As an alternative to plaque assays, biological
reporter assays have been reported to facilitate the efficient screen-
ing of inhibitors of HSV infection.14,17,18,27,55 Importantly, some of
these reporter assays were readily adapted for automated high‐
throughput screening.55,56 However, all of these assays require a des-
ignated cell culture facility, as these assays rely on using either the
native or recombinant whole herpes simplex virus. To overcome
these limitations, we report a molecular assay based on SPR that
allows continuous monitoring of the interaction between the host
receptors (HVEM and nectin‐1) and the gD protein (Figure 9).43
In our molecular‐level assay, cell receptors (either HVEM or
nectin‐1) are immobilized on a sensor chip, and gD is allowed to pass
over the receptor surface as an analyte. Upon gD binding to the
immobilized receptor, a response signal is generated that can be
integrated to evaluate binding kinetics Figure 9A. This SPR assay can
be modified into a competitive assay in which increasing concentra-
tions of gD‐specific binders (inhibitors) progressively decrease the
binding of gD to its cognate receptor Figure 9B. We used this
approach to test a selected aptamer that binds efficiently to the gD
of HSV‐1 and found that the aptamer specifically blocks the interac-
tions of gD with both HVEM and nectin‐1 receptors.43 Using the
response signals, the effective concentrations of aptamer‐1 required
for a 50% inhibition (EC50) of gD‐HVEM or gD‐nectin‐1 interactions
were calculated to be in the nanomolar range. These analyses indicated
10 of 12
FIGURE 9 Screening of inhibitors that interfere with interactions between HSV‐1 gD and human receptors (HVEM or Nectin1). A, SPR analyses of
HSV‐1 gD‐human receptor interaction analyses. B, SPR analyses of HSV‐1 gD‐human receptor interaction analyses in the presence of inhibitors
(boxed)
that aptamer‐1 efficiently blocks the interaction between the gD pro-
tein of HSV‐1 and the cognate receptors HVEM and nectin‐1, suggest-
ing that aptamer‐1 may have potent anti‐HSV‐1 activity. These results
correlate well with those of the plaque assay.43
Heparan sulfate has been shown to block the initial phase of HSV
57,58
infection. More recent investigations revealed that glycoproteins
from the HSV‐1 and HSV‐2 viruses are known to interact directly
with 3‐O‐sulfated heparin, which is abundantly expressed on host
cells.54,59-63 These studies suggest that the binding of heparan sulfate
to gD prevents its interactions with host cell receptors. However,
because of the absence of a structure for the gD‐heparan sulfate
complex, heparan sulfate binding sites within gD have not been iden-
tified. Another sulfated polysaccharide isolated from the sporophyll of
Undaria pinnatifida (Mekuba) was found to efficiently block HSV‐1
and HSV‐2 infection.64,65 Similarly, polysaccharides isolated from Spi-
rulina platensis and Sargassum horneri were found to have anti‐HSV‐1
activity.66,67 These studies indicate that sulfated polysaccharides have
the potential to bind to the gD proteins of HSV‐1 and HSV‐2 and
interfere with their interactions with host cell receptors such as
HVEM. The reported molecular assay43 was used to determine
whether sulfated polysaccharides, including heparan sulfate, have
the ability to interfere with gD‐HVEM interactions. Our studies
revealed that heparan sulfate binds to gD and efficiently interferes
with the gD‐HVEM interaction. However, when these studies were
extended to shorter synthetic sulfated saccharides, we did not
observe any interference with gD‐HVEM interactions, suggesting that
higher molecular weights or longer chain sulfated glycans are favor-
able for blocking viral glycoprotein‐host receptor interactions. The
sulfated polysaccharides isolated from S. horneri and fractioned are
longer chain sulfated saccharides.67 We evaluated these fractionated
sulfated polysaccharides using our molecular assay,43 to determine
KUMAR
their ability to interfere with gD‐HVEM interactions. A total of 14
natural polysaccharides were isolated, fractionated, and tested for
their ability to interfere with gD‐HVEM interactions using both our
molecular assay and the plaque assay. Among the evaluated polysac-
charides, 6 were found to interfere efficiently, 3 moderately, and 5
weakly.45 The polysaccharides that efficiently interfered with the
gD‐HVEM interaction were also found to be efficient inhibitors of
HSV‐1 replication in plaque assays.45 These analyses show that sul-
fated polysaccharides isolated from an edible alga, S. horneri, effi-
ciently inhibit the HSV‐1 gD‐HVEM interaction, most likely by
binding to the gD of HSV‐1. Molecular assays based on SPR are suit-
able for the screening of a wide range of compounds, including sul-
fated glycans and other specific inhibitors such as aptamers, to
identify inhibitors that target the initial phase of HSV infection and
the interaction of viral glycoproteins with host cell receptors.
SUMMARY REMARKS
The SPR platform has been shown to be useful for fragment‐based
drug discovery, producing several currently approved drugs and other
compounds in the final stages of clinical trials. Therefore, an SPR plat-
form for screening viral entry inhibitors could also be used to produce
more efficient inhibitors. This review covers different strategies for the
efficient screening of entry inhibitors using SPR. Some of these
improved strategies can be automated to facilitate the continuous
screening of entry inhibitors while requiring minimal amounts of recep-
tor and viral surface proteins compared with other screening strate-
gies. SPR could therefore be an appropriate platform to meet the
challenges posed by newly emerging viruses, enabling the identifica-
tion of suitable inhibitors to prevent pandemics. The strategies
reviewed above will undoubtedly find applications in the evaluation
and discovery of new viral entry inhibitors against diverse viruses.
KUMAR
ACKNOWLEDGEMEN TS
The author would like to acknowledge those who participated in the
viral protein and host receptor interaction studies (Drs. S.C.B.
Gopinath, T. Misono, E. Suenaga, and K. Hayashi). Funding from the
National Institute of Advanced Industrial Science and Technology
(AIST) to P.K.R.K supported this work.
CONF LICT OF INTE R ES T
The authors have no competing interest.
ORCID
Penmetcha K.R. Kumar http://orcid.org/0000-0002-3512-704X
RE FE R ENC E S
1. Bembenek SD, Tounge BA, Reynolds CH. Ligand efficiency and
fragment‐based drug discovery. Drug Discov Today. 2009;14:278‐283.
2. Schulz MN, Hubbard RE. Recent progress in fragment‐based lead
discovery. Curr Opin Pharmacol. 2009;9:615‐621.
3. Chessari G, Woodhead AJ. From fragment to clinical candidates—a
historical perspective. Drug Discov Today. 2009;14:668‐675.
4. Shuker SB, Hajduk PJ, Meadows RP, Fesik SW. Discovering high‐affin-
ity ligands for proteins: SAR by NMR. Science. 1996;274:1531‐1534.
5. Mayer M, Meyer B. Characterization of ligand binding by saturation
transfer difference NMR spectroscopy. Angew Chem Int Ed. 1999;
38:1784‐1788.
6. Blundell TL, Jhoti H, Abell C. High‐throughput crystallography for lead
discovery in drug design. Nat Rev Drug Discov. 2002;1:45‐54.
7. Duong‐Thi MD, Bergström M, Fex T, Isaksson R, Ohlson S. High‐
throughput fragment screening by affinity LC‐MS. J Biomol Screen.
2013;18:160‐171.
8. Ladbury JE, Klebe G, Freire E. Adding calorimetric data to decision mak-
ing in lead discovery: a hot tip. Nat Rev Drug Discov. 2010;9:23‐27.
9. Kranz JK, Schalk‐Hihi C. Protein thermal shifts to identify low molecular
weight fragments. Methods Enzymol. 2011;493:277‐298.
10. Lewis LM, Engle LJ, Pierceall WE, Hughes DE, Shaw KJ. Affinity
capillary electrophoresis for the screening of novel antimicrobial
targets. J Biomol Screen. 2004;9:303‐308.
11. Duong‐Thi MD, Meiby E, Bergstrom M, et al. Weak affinity chromatog-
raphy as a new approach for fragment screening in drug discovery. Anal
Biochem. 2011;414:138‐146.
12. Proll F, Fechner P, Proll G. Direct optical detection in fragment‐based
screening. Anal Bioanal Chem. 2001;393:1557‐1562.
13. Grondin B, DeLuca N. Herpes simplex virus type 1 ICP4 promotes tran-
scription preinitiation complex formation by enhancing the binding of
TFIID to DNA. J Virol. 2000;74:11504‐11510.
14. Tanaka M, Kodaira H, Nishiyama Y, Sata T, Kawaguchi Y. 2004. Con-
struction of recombinant herpes simplex virus type I expressing green
fluorescent protein without loss of any viral genes. Microbes Infect.
2004;6:485‐493.
15. Tscherne DM, Jones CT, Evans MJ, Lindenbach BD, McKeating JA, Rice
CM. Time and temperature‐dependent activation of hepatitis C virus
for low‐pH‐triggered entry. J Virol. 2006;80:1734‐1741.
16. Han Q, Xu C, Wu C, Zhu W, Yang R, Chen X. Compensatory mutations
in NS3 and NS5A proteins enhance the virus production capability of
hepatitis C reporter virus. Virus Res. 2009;145:63‐73.
17. Wang YC, Kao CL, Liu WT, Sun JR, Tai YE, Kung SH. A cell line that
secretes inducibly a reporter protein for monitoring herpes simplex
virus infection and drug susceptibility. J Med Virol. 2002;68:599‐605.
11 of 12
18. Stranska R, Schuurman R, Scholl DR, et al. ELVIRA HSV, a yield reduc-
tion assay for rapid herpes simplex virus susceptibility testing.
Antimicrob Agents Chemother. 2004;48:2331‐2333.
19. Wymer JP, Chung TD, Chang YN, Hayward GS, Aurelian L. Identifica-
tion of immediate‐early‐type cis‐response elements in the promoter
for the ribonucleotide reductase large subunit from herpes simplex
virus type 2. J Virol. 1989;63:2773‐2784.
20. Lee JC, Shih YF, Hsu SP, Chang TY, Chen LH, Hsu JT. Development of a
cell‐based assay for monitoring specific hepatitis C virus NS3/4A pro-
tease activity in mammalian cells. Anal Biochem. 2003;316:162‐170.
21. Lee JC, Yu MC, Lien TW, Chang CF, Hsu JT. High throughput cell‐based
screening for hepatitis C virus NS3/4A protease inhibitors. Assay Drug
Dev Technol. 2005;3:385‐392.
22. Codran A, Royer C, Jaeck D, et al. Entry of hepatitis C virus
pseudotypes into primary human hepatocytes by clathrin‐dependent
endocytosis. J Gen Virol. 2006;87:2583‐2593.
23. Netski DM, Mosbruger T, Depla E, et al. Humoral immune response in
acute hepatitis C virus infection. Clin Infect Dis. 2005;41:667‐675.
24. Basu A, Kanda T, Beyene A, Saito K, Meyer K, Ray R. Sulfated homo-
logues of heparin inhibit hepatitis C virus entry into mammalian cells.
J Virol. 2007;81:3933‐3941.
25. Tani H, Komoda Y, Matsuo E, et al. Replication competent recombinant
vesicular stomatitis virus encoding hepatitis C virus envelope proteins.
J Virol. 2007;81:8601‐8612.
26. Mann AM, Rusert P, Berlinger L, Kuster H, Günthard HF, Trkola A. HIV
sensitivity to neutralization is determined by target and virus producer
cell properties. AIDS. 2009;23:1659‐1667.
27. Taylor TJ, Knipe DM. 2009. The use of green fluorescent fusion pro-
teins to monitor herpes simplex virus replication. Methods Mol Biol.
2009;515:239‐248.
28. Myszka DG, Rich RL. Implementing surface plasmon resonance biosen-
sors in drug discovery. Pharm Sci Technol Today. 2000;3:310‐317.
29. Cooper MA. Label‐free screening of biomolecular interactions. Anal
Bioanal Chem. 2003;377:834‐842.
30. Lofas S. Optimizing the hit‐to‐lead process using SPR analysis. Assay
Drug Dev Technol. 2004;2:407‐416.
31. Huber W. A new strategy for improved secondary screening and lead
optimization using high‐resolution SPR characterization of compound‐
target interactions. J Mol Recognit. 2005;18:273‐281.
32. Geschwindner S, Olsson LL, Albert JS, et al. Discovery of a novel
warhead β‐secretase through fragment‐based lead generation. J Med
Chem. 2007;50:5903‐5911.
33. Godemann R, Madden J, Kramer J, et al. Fragment‐based discovery of
BACE1 inhibitors using functional assays. Biochemistry. 2009;
48:10743‐10751.
34. Cole DC, Olland AM, Jacob J, et al. Identification and characteriza-
tion of acidic mammalian chitinase inhibitors. J Med Chem. 2010;
53:6122‐6128.
35. Giannetti AM. From experimental design to validation hits a compre-
hensive walk‐through of fragment lead identification using surface
plasmon resonance. Methods Enzymol. 2011;493:169‐218.
36. Shuman CF, Markgren PO, Hämäläinen M, Danielson UH. Elucidation of
HIV‐1 protease resistance by characterization of interaction kinetics
between inhibitors and enzyme variants. Antiviral Res. 2003;58:235‐243.
37. Geitmann M, Danielson UH. Studies of substrate‐induced conforma-
tional changes in human cytomegalovirus protease using optical
biosensor technology. Anal Biochem. 2004;332:203‐214.
38. Rich RL, Hoth LR, Geoghegan KF, et al. Kinetic analysis of estrogen
receptor/ligand interactions. Proc Natl Acad Sci. 2002;99:2562‐2567.
39. Davidson W, Frego L, Peet GW, et al. Discovery and characterization
of a substrate selective p38alpha inhibitor. Biochemistry. 2004;
43:11658‐11671.
40. De Mol NJ. Affinity constants for small molecules from SPR competi-
tion experiments. Methods Mol Biol. 2010;627:101‐111.
12 of 12
41. Gonzales NR, Schuck P, Schlom J, Kashmiri SV. Surface plasmon reso-
nance based competition assay to assess the sera reactivity of
variants of humanized antibodies. J Immunol Methods. 2002;268:197‐
210.
42. Maurer MF, Garrigues U, Jaspers SR, et al. Generation and characteri-
zation of human anti‐human IL‐21 neutralizing monoclonal antibodies.
Monoclonal Antibodies. 2012;4:69‐83.
43. Gopinath SC, Hayashi K, Kumar PKR. Aptamer that binds to the
gD protein of herpes simplex virus 1 and efficiently inhibits viral entry.
J Virol. 2012;86:6732‐6744.
44. Gamsjaeger R, Kariawasam R, Bang LH, et al. Semiquantitative and
quantitative analysis of protein‐DNA interactions using steady‐state
measurements in surface plasmon resonance competition experiments.
Anal Biochem. 2013;440:178‐185.
45. Gopinath SC, Hayashi K, Lee JB, et al. Analysis of compounds that
interfere with herpes simplex virus‐host receptor interactions using
surface plasmon resonance. Anal Chem. 2013;85:10455‐10462.
46. Suenaga E, Kumar PKR. An aptamer that binds efficiently to the hemag-
glutinins of highly pathogenic avian influenza viruses (H5N1 and H7N7)
and inhibits hemagglutinin‐glycan interactions. Acta Biomater. 2014;
10:1314‐1323.
47. Navratilova I, Besnard J, Hopkins AL. Screening for GPCR ligands using
surface plasmon resonance. ACS Med Chem Lett. 2011;2:549‐554.
48. Suenaga E, Mizuno H, Kumar PKR. Monitoring influenza hemagglutinin
and glycan interactions using surface plasmon resonance. Biosens
Bioelectron. 2012;15:195‐201.
49. Karlsson R, Katsamba PS, Nordin H, Pol E, Myszka DG. Analyzing a
kinetic titration series using affinity biosensors. Anal Biochem.
2006;349:136‐147.
50. Trutnau HH. New multi‐step kinetics using common affinity biosensors
saves time and sample at full access to kinetics and concentration.
J Biotechnol. 2006;124:191‐195.
51. Gluck JM, Koenig BW, Willbold D. Nanodiscs allow the use of integral
membrane proteins as analytes in surface plasmon resonance studies.
Anal Biochem. 2011;408:46‐52.
52. Whitbeck JC, Peng C, Lou H, et al. Glycoprotein D of herpes simplex
virus (HSV) binds directly to HVEM, a member of the tumor necrosis
factor receptor superfamily and a mediator of HSV entry. J Virol.
1997;71:6083‐6093.
53. Montgomery RI, Warner MS, Lum BJ, Spear PG. Herpes simplex virus‐1
entry into cells mediated by a novel member of the TNF/NGF receptor
family. Cell. 1996;87:427‐436.
54. Shukla D, Liu J, Blaiklock P, et al. A novel role for 3‐O‐sulfated heparin
sulfate in herpes simplex virus 1 entry. Cell. 1999;99:13‐22.
55. Dicker IB, Blasecki JW, Seetharam S. Herpes simplex type 1: lacZ
recombinant viruses. II. Microtiter plate‐based colorimetric assays for
the discovery of new antiherpes agents and the points at which
KUMAR
such agents disrupt the viral replication cycle. Antiviral Res.
1995;28:213‐224.
56. Zhu QC, Wang Y, Peng T. Herpes simplex virus (HSV) immediate‐
early (IE) promoter‐directed reporter system for the screening of anti‐
herpetics targeting the early stage of HSV infection. J Biomol Screen.
2010;15:1016‐1020.
57. Nahmias AJ, Kibrick S. Inhibitory effect of heparin on herpes simplex
virus. J Bacteriol. 1964;87:1060‐1066.
58. WuDunn D, Spear PG. Initial interaction of herpes simplex virus with
cells is binding to heparin sulfate. J Virol. 1989;63:52‐58.
59. Xia G, Chen J, Tiwari V, et al. Heparan sulfate 3‐O‐sulfotransferase
isoform 5 generates both an anti‐thrombin‐binding site and an entry
receptor for herpes simplex virus type 1. J Biol Chem. 2002;277:
37912‐37919.
60. Xu D, Tiwari V, Xia G, Clement C, Shukla D, Liu J. Characterization of
heparan sulphate 3‐O‐sulphotransferase isoform 6 and its role in
assisting the entry of herpes simplex virus type 1. Biochem J. 2005;
385:451‐459.
61. O'Donnell CD, Tiwari V, Oh MJ, Shukla D. A role for heparan sulfate
3‐O‐sulfotransferase isoform 2 in herpes simplex virus type 1 entry
and spread. Virology. 2006;346:452‐459.
62. Liu J, Shriver Z, Pope RM, et al. Characterization of a heparan sulfate
octasaccharide that binds to herpes simplex virus type 1 glycoprotein
D. J Biol Chem. 2002;277:33456‐33467.
63. Antoine TE, Park PJ, Shukla D. Glycoprotein targeted therapeutics: a
new era of anti‐herpes simplex virus‐1 therapeutics. Rev Med Virol.
2013;23:194‐208.
64. Lee JB, Hayashi K, Hashimoto M, Nakano T, Hayashi T. Novel antiviral
fucoidan from sporophyll of Undaria pinnatifida (Mekabu). Chem Pharm
Bull. 2004;52:1091‐1094.
65. Hayashi K, Nakano T, Hashimoto M, Kanekiyo K, Hayashi T. Defensive
effects of a fucoidan from brown alga Undaria pinnatifida against her-
pes simplex virus infection. Int Immunopharmacol. 2008;8:109‐116.
66. Lee JB, Srisomporn P, Hayashi K, Tanaka T, Sankawa U, Hayashi T.
Effects of structural modification of calcium spirulan, a sulfated poly-
saccharide from Spirulina platensis, on antiviral activity. Chem Pharm
Bull. 2001;49:108‐110.
67. Preeprame S, Hayashi K, Lee JB, Sankawa U, Hayashi T. A novel antivi-
ral active fucan sulfate derived from an edible brown alga, Sargassum
horneri. Chem Pharm Bull. 2001;49:484‐485.
How to cite this article: Kumar PKR. Systematic screening of
viral entry inhibitors using surface plasmon resonance. Rev
Med Virol. 2017;e1952. https://doi.org/10.1002/rmv.1952