Hepatitis C Virus E2 Has Three Immunogenic Domains Containing Conformational Epitopes with Distinct Properties and Biological Functions
Mechanisms of virion attachment, interaction with its receptor, and cell entry are poorly understood for hepatitis C virus (HCV) because of a lack of an efficient and reliable in vitro system for virus propagation. Infectious HCV retroviral pseudotype particles (HCVpp) were recently shown to express native E1E2 glycoproteins, as defined in part by HCV human monoclonal antibodies (HMAbs) to conformational epitopes on E2, and some of these antibodies block HCVpp infection (A. Op De Beeck, C. Voisset, B. Bartosch, Y. Ciczora, L. Cocquerel, Z. Y. Keck, S. Foung, F. L. Cosset, and J. Dubuisson, J. Virol. 78:2994-3002, 2004). Why some HMAbs are neutralizing and others are nonneutralizing is looked at in this report by a series of studies to determine the expression of their epitopes on E2 associated with HCVpp and the role of antibody binding affinity. Antibody cross-competition defined three E2 immunogenic domains with neutralizing HMAbs restricted to two domains that were also able to block E2 interaction with CD81, a putative receptor for HCV. HCVpp immunoprecipitation showed that neutralizing and nonneutralizing domains are expressed on E2 associated with HCVpp, and affinity studies found moderate-to-high-affinity antibodies in all domains. These findings support the perspective that HCV-specific epitopes are responsible for functional steps in virus infection, with specific antibodies blocking distinct steps of virus attachment and entry, rather than the perspective that virus neutralization correlates with increased antibody binding to any virion surface site, independent of the epitope recognized by the antibody. Segregation of virus neutralization and sensitivity to low pH to specific regions supports a model of HCV E2 immunogenic domains similar to the antigenic structural and functional domains of other flavivirus envelope E glycoproteins.
JOURNAL OF VIROLOGY, Sept. 2004, p. 9224–9232 Vol. 78, No. 17
0022-538X/04/$08.00⫹0 DOI: 10.1128/JVI.78.17.9224–9232.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Hepatitis C Virus E2 Has Three Immunogenic Domains Containing
Conformational Epitopes with Distinct Properties
and Biological Functions
Anne Op De Beeck,
Kenneth G. Hadlock,
and Steven K. H. Foung
Department of Pathology, Stanford University School of Medicine, Stanford, California,
Institut de Biologie de Lille and Institut Pasteur de Lille, Lille, France
Received 5 February 2004/Accepted 21 April 2004
Mechanisms of virion attachment, interaction with its receptor, and cell entry are poorly understood for
hepatitis C virus (HCV) because of a lack of an efﬁcient and reliable in vitro system for virus propagation.
Infectious HCV retroviral pseudotype particles (HCVpp) were recently shown to express native E1E2 glyco-
proteins, as deﬁned in part by HCV human monoclonal antibodies (HMAbs) to conformational epitopes on E2,
and some of these antibodies block HCVpp infection (A. Op De Beeck, C. Voisset, B. Bartosch, Y. Ciczora, L.
Cocquerel, Z. Y. Keck, S. Foung, F. L. Cosset, and J. Dubuisson, J. Virol. 78:2994-3002, 2004). Why some
HMAbs are neutralizing and others are nonneutralizing is looked at in this report by a series of studies to
determine the expression of their epitopes on E2 associated with HCVpp and the role of antibody binding
afﬁnity. Antibody cross-competition deﬁned three E2 immunogenic domains with neutralizing HMAbs re-
stricted to two domains that were also able to block E2 interaction with CD81, a putative receptor for HCV.
HCVpp immunoprecipitation showed that neutralizing and nonneutralizing domains are expressed on E2
associated with HCVpp, and afﬁnity studies found moderate-to-high-afﬁnity antibodies in all domains. These
ﬁndings support the perspective that HCV-speciﬁc epitopes are responsible for functional steps in virus
infection, with speciﬁc antibodies blocking distinct steps of virus attachment and entry, rather than the
perspective that virus neutralization correlates with increased antibody binding to any virion surface site,
independent of the epitope recognized by the antibody. Segregation of virus neutralization and sensitivity to low
pH to speciﬁc regions supports a model of HCV E2 immunogenic domains similar to the antigenic structural
and functional domains of other ﬂavivirus envelope E glycoproteins.
Hepatitis C virus (HCV) infects over 170 million individuals
worldwide. Although acute infection is usually silent, most
HCV infections progress to chronicity that is not cleared by an
apparently robust immune response (3, 24). The virus is a
member of the family Flaviviridae (37), with a 9.5-kb positive-
strand RNA genome that encodes three structural proteins,
the capsid and viral envelope proteins E1 and E2, and at least
six nonstructural proteins, NS2 to NS5b (29). The envelope
proteins are thought to be the primary mediators of virion
attachment and cell entry (13). HCV E2 is a ⬃70-kDa glyco-
protein that shows large variations among HCV genotypes and
contains a 27-amino-acid (aa) sequence at its amino terminus
that is highly variable and is designated the hypervariable re-
gion 1, or HVR1 (reviewed in references 3 and 6). This linear
region on E2 is likely to be involved in virus infection, since
neutralizing antisera to HVR1 have been reported in in vitro
and in vivo models, although other studies showed that HCV
with HVR1 deleted remains infectious (15, 19, 34, 41, 47).
Unfortunately, a leading contributor to disease progression is
the emergence of new viral mutants or “quasispecies” in
HVR1 induced by immune selection. Increased diversity or
mutations in HVR1 correlate with progressive disease, and
decreased diversity correlates with resolving disease (14). HCV
E2 is thought to mediate attachment to target cells and binds
to human CD81, a member of the tetraspannin family of pro-
teins (28). Interaction of E2 with CD81 on B or T cells has
been reported to result in B-cell aggregation and a lowering of
the threshold for T- and B-cell activation (17, 43). Other al-
ternative receptors that have been proposed include the low-
density lipoprotein receptor (1, 44), two receptors on HepG2
cells, the scavenger receptor type B class I (5, 40), and two
closely related membrane-associated C-type mannose-binding
lectins, DC-SIGN and L-SIGN (20, 30, 33).
Mechanisms of virion attachment, entry, and virus replica-
tion have been difﬁcult to study because of difﬁculties in having
an efﬁcient and reliable in vitro system for virus propagation.
The development of infectious HCV retroviral pseudotype
particles expressing E1E2 (HCVpp) has permitted a more
detailed characterization of functional envelope glycoproteins
involved in virion attachment and entry (4, 25). HCVpp pref-
erentially infect human hepatocytes and hepatocellular cell
lines and express noncovalent E1E2 heterodimers as deﬁned in
part by HCV human monoclonal antibodies (HMAbs) to con-
formational epitopes on E2 (32). Production of HMAbs pro-
vides information on the immune response to native E1 and E2
proteins, as they are recognized during natural infection and
should be useful in determining the function and structure of
speciﬁc immunogenic domains of E1 or E2. HMAbs and re-
* Corresponding author. Mailing address: Stanford Medical School
Blood Center, 800 Welch Rd., Palo Alto, CA 94304. Phone: (650)
723-6481. Fax: (650) 725-6610. E-mail: email@example.com.
combinant antibodies to E2 have been isolated to conforma-
tional epitopes that are conserved between subtypes 1a and 1b
(2, 7, 9, 22) and genotypes 1 and 2 (23). These HMAbs include
antibodies that are effective and ineffective in inhibiting the
binding of E2 to CD81 and HCVpp entry into target cells (2,
7, 22, 23, 32). Our investigators developed a panel of HMAbs
to HCV E2, of which the majority were to conformational
epitopes (23). Each of the HCV HMAbs was secreted from a
human hybridoma expressing a unique immunoglobulin G1
(IgG1) gene that had undergone afﬁnity maturation in vivo
(10). Some of the epitopes recognized by the HMAbs were
broadly conserved across different HCV genotypes and were
able to inhibit the binding of E2 to human CD81, but only
three blocked HCVpp entry into Huh-7 cells (32). Why only a
subset of the HCV HMAbs to E2 was able to block HCVpp
entry remains unclear. One explanation is the availability of
the different antibody binding epitopes on the surface of
HCVpp. A second possibility is the binding afﬁnity of the
antibodies, and a third possibility is that virus entry is mediated
by only speciﬁc epitopes on the virus surface. For some viruses,
the prevailing view is that inhibition of virus entry or virus
neutralization correlates with increased antibody binding to
any virion surface site, independent of the epitope recognized
by the antibody. Neutralization is then the result of a critical
number of binding sites being occupied and virus entry being
prevented through steric hindrance (8). Higher-afﬁnity anti-
bodies will have higher neutralizing activities. Nonneutralizing
antibodies either do not bind to the virion surface or are poor
binders with low afﬁnity. In contrast, the role of speciﬁc
epitopes responsible for functional steps in virus entry has
been documented for other viruses, with speciﬁc antibodies
blocking distinct steps of virus attachment, interaction with
receptor and coreceptor, and initiation of viral envelope fusion
with the cellular membrane (26).
In this report, antibody competition studies showed three
immunogenic domains on HCV E2 that contained conserved
conformational epitopes. The lack of HVR1 involvement with
these domains was determined by binding studies to HVR1
deletion mutants. Expression of these epitopes on native pro-
teins was analyzed by the ability of HCV HMAbs to immuno-
precipitate HCVpp. Afﬁnity studies of HCV HMAbs were
performed to correlate antibody binding afﬁnity and blocking
of HCVpp entry to target cells. Also, a collective analysis of
previous and present ﬁndings suggests that the three immuno-
genic domains are associated with distinct properties similar to
the antigenic structure and function of other ﬂavivirus enve-
lope E glycoproteins.
MATERIALS AND METHODS
Cell culture and viruses. HeLa and HEK293T cells were from the American
Type Culture Collection (Manassas, Va.). Cells were grown in minimal essential
medium (MEM; Invitrogen, Carlsbad, Calif.) and Dulbecco’s MEM (Invitro-
gen), respectively, supplemented with 10% fetal calf serum (Gemini Bioproducts
Inc., Calabasas, Calif.). A cell line constitutively expressing sf1b-E2 on the cell
surface was derived by transfecting Chinese hamster ovary (CHO) cells with
plasmid expressing sf1b-E2. After selection with Geneticin (Invitrogen), CHO
cells expressing E2 were identiﬁed via ﬁxed-cell immunoﬂuorescence using
CBH-5 and hemagglutinin (HA) MAbs, essentially as described elsewhere (23).
Cells expressing high levels of E2 were subjected to single-cell cloning. Recom-
binant vaccinia virus expressing HCV envelope proteins was constructed and
grown as described previously (23). Vaccinia virus 1488 expressing the structural
proteins of HCV 1a strain H (21) was generously provided by Charles Rice
MAbs. The production, puriﬁcation, and biotinylation of the HCV HMAbs
were performed as described previously (23). Rat MAb 3/11 to HCV E2 was
cultured as described elsewhere (17) and generously provided by Jane McKeat-
ing (Rockefeller University). Rat MAb to the inﬂuenza virus HA epitope was
from Roche Applied Sciences (Indianapolis, Ind.). Murine MAb to the c-myc
epitope was from Santa Cruz Biotechnology (Santa Cruz, Calif.).
Competition assays. Monolayers of HeLa cells were grown to 80% conﬂuence
and infected at 5 PFU/cell with both wild-type virus and recombinant vaccinia
virus or wild-type virus only. Cells were harvested after 1 day of infection.
Extracts were prepared by washing the cells with phosphate-buffered saline
(PBS) and then resuspending ⬃25 ⫻ 10
cells in 1 ml of lysis buffer. Extracts
prepared in this manner contained approximately 25 g of E2 protein/ml. Nuclei
were pelleted by centrifugation at 18,000 ⫻ g at 4°C for 10 min, and resulting
cytoplasmic extracts were stored at 4°C and used for enzyme-linked immunosor-
bent assays within 24 h of preparation. Microtiter plates were prepared by
coating wells with 500 ng of puriﬁed Galanthus nivalis lectin (GNA; Sigma, St.
Louis, Mo.) in 100 lofPBSfor1hat37°C. Wells were washed with Tris-
buffered saline (TBS; 150 mM NaCl, 20 mM Tris-HCl; pH 7.5) and then blocked
with BLOTTO (TBS, 0.1% Tween 20, 2.5% normal goat serum, and 2.5% nonfat
dry milk) by incubation for1hatroom temperature (RT). Plates were washed
twice with TBS, followed by the addition to each well of 15 l of cytoplasmic
extract containing E2 diluted with 85 l of BLOTTO. After 1.5 h at RT, plates
were washed three times with TBS followed by the addition to each well of 50 l
of BLOTTO containing competing antibodies at various concentrations. After 30
min, 50 lofa2-g/ml solution of the biotinylated test antibody was added.
After incubation for 1.5 h at RT, the plates were washed three times with TBS,
and 100 l of 1/1,000-diluted alkaline phosphatase-conjugated streptavidin (Am-
ersham-Pharmacia Biotech, Piscataway, N.J.) was added. After1hatRT,the
plates were washed four times with TBS followed by a 30-min incubation with a
1-mg/ml solution of p-nitrophenyl phosphate. Absorbance was measured at 405
nm with a multiwell plate reader (BioTek Instruments, Winooski, Vt.). Each test
HMAb at 2 g/ml with competing HMAb ranging from 0.2 to 50 g/ml was
tested in duplicate in at least two different experiments. To develop a cross-
competition matrix for percentage of test antibody bound to E2, the mean signal
with biotinylated test antibody to E2 with competing antibody at 20 g/ml was
divided by the signal in the absence of the competing HMAb, followed by
multiplying by 100.
Phylogenetic grouping of HCV HMAbs. Without knowing the exact confor-
mational epitopes targeted by these antibodies, we attempted to determine their
spatial relationship based on competition study results. When two antibodies
cross-competed, the extent of bidirectional inhibition was interpreted as the
extent of epitope overlap by the competing antibodies. For unidirectional inhi-
bition or enhancement, effects were interpreted as proximal, but not overlapping
epitopes (31). Using the principles of UPGMA (unweighted pair-group meth-
od using arithmetic averages) to perform a sequential cluster analysis, spatial
relationships were developed as a phylogenetic tree to correlate the related-
ness of the epitopes as identiﬁed by this panel of antibodies (16, 42). In this
analysis, antibodies with the highest bidirectional inhibition were placed next
to each other (see Fig. 1C). The paired antibodies were averaged and used to
compare other antibodies according to the degree of their cross-competition
with the paired antibodies. The (third) identiﬁed antibody with the strongest
bidirectional inhibition was placed next to the ﬁrst pair of antibodies, and a
new average was obtained between the third antibody and the average of the
ﬁrst pair. The new average was then used for another cycle of comparison
with the other antibodies in this panel until the matrix was completely re-
Isolation and cloning of HCV E2 HVR1 deletion constructs. The vaccinia virus
recombinant Q1b (GenBank accession no. AF348705) (23), sf1B-E2, and the
HCV 1b deletion constructs were derived from the same HCV genotype 1b-
positive serum. DNA encoding the E2 protein was prepared by reverse tran-
scription-PCR using Pfu Taq polymerase (Stratagene, La Jolla, Calif.) with
HCV-speciﬁc oligonucleotide primers (forward
AAG-3⬘). HCV 1a constructs were ampliﬁed by PCR from viral stocks of
vaccinia virus construct vv1488 (21) with the HCV-speciﬁc oligonucleotide prim-
(5⬘-AGATCTATCCAACTGATCAACACCAAC-3⬘), and reverse
AGCTCGGACCTGTCCCTGTCTTC-3⬘). Flanking BglII or PstI restriction
sites in the primer sequences are underlined. Ampliﬁed DNA fragments were
ligated into the pDisplay vector (Invitrogen) in frame with HA and c-myc as tags,
VOL. 78, 2004 IMMUNOGENIC DOMAINS IN HCV E2 9225
FIG. 1. Competition analysis of HCV HMAbs. (A) Binding of biotinylated test antibody (as indicated on top of each panel) to HCV Q1b E2
protein captured onto GNA lectin-coated microtiter plates was competed by various concentrations of competing antibodies, as described in
Materials and Methods. The bound biotinylated test antibody was detected by using alkaline phosphatase-conjugated streptavidin and p-
nitrophenyl phosphate as the substrate. The y axis value (mean ⫾ standard deviation; n ⫽ 2) is the absorbance reading obtained in the presence
of competing antibody as a percentage of the reading in the absence of competing antibody. Competing antibodies were CBH-2 (Œ), CBH-5 (■),
CBH-8C (⽧), CBH-8E (), CBH-11 (F), CBH-7 (䊐), CBH-4G (ƒ), CBH-4B (‚), and R04 (*). R04 is an isotype-matched control antibody to
a cytomegalovirus protein. (B) Cross-competition matrix. (C) Phylogenetic grouping of HCV HMAbs based on the competitive binding assay. Solid
lines with numbers indicate the relatedness of the two adjoining antibodies. Circles are clusters of antibodies in a speciﬁc domain. Competition
results used for calculation are the mean values obtained from two to ﬁve separate experiments.
9226 KECK ET AL. J. VIROL.
which were used for puriﬁcation and detection of expressed proteins. In-frame
HCV inserts and deletion sites were conﬁrmed by DNA sequencing (PE-Applied
Biosystems, Foster City, Calif.).
Analyses of HCV E2 deletion constructs. HEK293T cells were seeded to
obtain 60 to 70% conﬂuence by the following day. For transfection of a T-75
ﬂask, a mixture of 40 g of the plasmid DNA and 240 g of PerFect Lipid Pfx-2
(Invitrogen) were combined in 1 ml of serum-free Dulbecco’s MEM. After4hof
incubation at 37°C, the transfection solution was replaced with 20 ml of complete
medium and cells were grown for 24 h. Cell extracts were prepared by washing
cells with PBS and resuspending them in 1 ml of lysis buffer. Nuclei were pelleted
by centrifugation at 18,000 ⫻ g at 4°C for 10 min. For microtiter plate assays, the
plates were prepared by coating wells with 500 ng of puriﬁed GNA lectin in 100
lofPBSfor1hat37°C. Wells were washed with TBS and then blocked with
150 l of BLOTTO by incubation for1hatRT.Wells were washed twice with
TBS, followed by the addition of 25 l of extract from HEK293 cells transfected
with E2 deletion constructs diluted in 75 l of BLOTTO. After 1.5 h at RT,
plates were washed three times with TBS followed by the addition of 100 lof
BLOTTO containing various MAbs at 10 g/ml. Plates were incubated for 1.5 h
and washed three times with TBS, and then 100 l of alkaline phosphatase-
conjugated secondary antibody, diluted in BLOTTO as recommended by the
manufacturer, was added (for anti-human and anti-mouse antibodies [Promega,
Madison, Wis.] and for anti-rat antibody [Kirkegard & Perry, South San Fran-
cisco, Calif.]). Bound secondary antibody was detected and quantiﬁed as de-
For Western blot analysis, 293T cells were transfected with either full-length
E2 or constructs with HVR1 deleted overnight using a calcium phosphate trans-
fection kit (Clontech, Palo Alto, Calif.). After washing once with PBS, the cells
were lysed in lysis buffer. Ten micrograms of the proteins was denatured in
Laemmli sodium dodecyl sulfate (SDS) sample buffer and loaded onto SDS-
polyacrylamide gels. Proteins were separated by electrophoresis and transferred
to nitrocellulose membranes. Blots were blocked for1hin5%(wt/vol) nonfat
dry milk dissolved in TBS–0.1% Tween 20 (TBST). Blots were then probed with
rat anti-HA antibody overnight at 4°C in blocking buffer. After washing with
TBST three times, the blots were probed with a secondary antibody (horseradish
peroxidase-conjugated anti-mouse IgG from Santa Cruz Biotech) for1hatRT.
Blots were washed three times in TBST and then developed with enhanced
chemiluminescence. Western blot images were captured using ChemiDoc imager
system (Bio-Rad, Richmond, Calif.).
Immunoprecipitation. Production of HCVpp was carried out as described
previously (4). Brieﬂy, 293T cells were transfected with expression vectors en-
coding the viral components, i.e., E1E2 glycoproteins, retroviral core proteins,
and packaging-competent green ﬂuorescent protein-containing retroviral trans-
fer vectors, by using a calcium-phosphate transfection protocol. 293T cells were
metabolically labeled from 16 to 40 h posttransfection with 50 Ci of
protein labeling mix (Amersham Biosciences)/ml. Metabolically labeled 293T
cells and supernatant containing HCVpp were lysed with 0.5% Igepal CA-630 in
TBS (50 mM Tris-HCl [pH 7.5], 150 mM NaCl). Approximately 10
pseudotype particles were used per immunoprecipitation reaction mixture. Im-
munoprecipitations were carried out as described elsewhere (12, 13). Brieﬂy, 7
g of MAb was incubated with protein A-Sepharose (Sigma) for1hat4°Cin
TBS containing 0.2% Igepal CA-630. Beads were then incubated with the anti-
gen for 1 h at 4°C. Between each step, beads were washed once with TBS-Igepal.
After the last step, they were washed three times with this buffer and once with
distilled water. The precipitates were then heated at 70°C for 5 min in SDS-
polyacrylamide gel electrophoresis sample buffer and run on a polyacrylamide
gel. After electrophoresis, gels were treated with sodium salicylate, dried, and
exposed at ⫺80°C to an autoradiograph (Amersham).
Antibody afﬁnity measurements. A range of 0.001 to 100 g of each HCV
HMAb/ml was incubated with either genotype 1b E2 constitutively expressed in
CHO cells or transiently expressed 1a E2 in 293T cells for 45 min and washed
twice, followed by incubation with ﬂuorescein isothiocyanate-labeled goat anti-
human IgG (4 g/ml; Jackson Immunoresearch, West Grove, Pa.) for 45 min on
ice. Cells were then washed in PBS containing 1% fetal calf serum at 4°C and
resuspended in ﬁxative solution. Fluorescence of HMAb-bound cells was ana-
lyzed by ﬂow cytometry using a FACSCalibur (Becton-Dickinson, San Jose,
Calif.), and the mean ﬂuorescence intensity (MFI) values of cell populations
were obtained. The MFI value of nonspeciﬁc ﬂuorescence was measured by using
an isotype-matched control HMAb (RO4), the ﬂuorescence of which was sub-
tracted from the MFI values of the speciﬁc HMAbs. The MFI values of the cell
populations incubated with different amounts of antibodies were analyzed using
Prism software, and the saturation binding curves were ﬁt by nonlinear regres-
Competition analyses of HCV HMAbs. Competition studies
were performed to determine the spatial proximity of each
conformational epitope to other epitopes on HCV E2 as de-
ﬁned by a panel of HCV HMAbs (Table 1) (23). These anti-
bodies were derived from peripheral B cells of an individual
who had an asymptomatic HCV genotype 1b infection. Se-
quence analysis of the IgG1 genes of the HMAbs conﬁrmed
that they were derived from independent B cells (10, 23). As
described previously, the antibodies varied in the breadth of
their reactivities with different genotypes of HCV E2, with
some broadly reactive with genotypes 1a, 1b, 2a, and 2b and in
their ability to inhibit the interaction of HCV E2 with human
CD81 (Table 1). Genotype-speciﬁc plasmids used to produce
HCV E2 in the vaccinia virus expression system in this analysis
were derived from sera of other patients infected with different
HCV genotypes. Complete inhibition of antibody reactivity to
denatured E2 protein conﬁrmed that all HMAbs were to con-
formational epitopes except for CBH-17, which retained reac-
tivity. Competition studies were performed with genotype 1b
E2, since the original B cells were from an individual infected
with HCV 1b. Each HMAb was puriﬁed and biotinylated, and
the binding of the antibodies in increasing concentrations of
competing antibody was determined. Representative binding
curves are presented in Fig. 1A. The cross-competition matrix
(Fig. 1B) shows the mean signals of biotinylated test HMAb
with competing HMAb divided by signals without competing
HMAb from multiple experiments. The antibodies are approx-
imately ordered on both axes to reﬂect bidirectional inhibition
along the diagonal axis of the matrix. Strong inhibition is in-
dicated by values of ⱕ50%, and enhancement is indicated by
values of ⱖ125%. HMAbs CBH-2, -5, -8C, -8E, and -11 (as
shown for CBH-2 and -5 in Fig. 1A) form one cluster, where
each antibody showed strong bidirectional inhibition with all
other antibodies in the cluster (Fig. 1B). No signiﬁcant inhibi-
tion was observed with a control HMAb, R04, or CBH-4B -4D,
-4G, and -7. HMAbs CBH-4B, -4D, and -4G form another
cluster with strong bidirectional inhibition to each other (as
shown for CBH-4B in Fig. 1A). CBH-7 is to itself and is not
signiﬁcantly inhibited by CBH-2, -5, -8C, -8E, -11, or the con-
trol antibody. The relationship between CBH-7 and the CBH-
4B, -4D, and -4G cluster is unusual, with either strongly inhib-
TABLE 1. Selected characteristics of HCV HMAbs
CBH-2 1a 1b 2a 2b ⫹
CBH-4B 1a 1b ⫺
CBH-4D 1a 1b ⫺
CBH-4G 1a 1b 2a 2b ⫺
CBH-5 1a 1b 2a 2b ⫹
CBH-7 1a 1b 2a 2b ⫹
CBH-8C 1a 1b 2a 2b ⫹
CBH-8E 1a 1b 2a 2b ⫹
CBH-11 1b 2a 2b ⫹
CBH-17 1a 1b ⫹
Designation of HMAbs generated previously.
Reactivity with separate isolates of the indicated genotypes.
Ability of antibody to block HCV E2 binding to CD81-LEL. ⫹, inhibitory; ⫺,
VOL. 78, 2004 IMMUNOGENIC DOMAINS IN HCV E2 9227
itory (CBH-4B) or enhancing (CBH-4G) effects on binding.
While uneven bidirectional inhibition was observed within each
cluster, a predominately unidirectional effect was observed be-
tween CBH-7 and the CBH-4B, -4D, and -4G cluster (Fig. 1B).
Many techniques have been developed for determining phy-
logenetic relationships in comparative biology. The UPGMA
algorithm is a straightforward approach for cluster analysis,
weighting each data point in the cluster equally (16, 42). The
competition data were analyzed and spatially placed by pairing
antibodies with the highest bidirectional inhibition as most
related (shown next to each other in Fig. 1C). For example, the
binding of CBH-5 to E2 was reduced to 25% by CBH-11, and
the binding of CBH-11 was reduced to 9% by CBH-5. The
average of these two values is 17%, or 0.17 (as shown above the
line in the ﬁgure). A relationship tree was generated with the
closest two antibodies (e.g., CBH-5 and -11) paired together.
Their inhibition percentages against each of the other antibod-
ies were averaged and added to the matrix in proximity to the
original pair. This cycle was repeated until all HMAbs were
assigned in this tree. This approach identiﬁed three distinct
immunogenic domains. HMAbs CBH-4G, -4B and -4D consti-
tute domain A. A second domain, B, includes CBH-2, -5, -8C,
-8E, and -11; CBH-7 is in a separate domain, C. Within each
domain, strong bidirectional inhibition was observed, with val-
ues generally less than 0.5. Virtually no cross-competition was
observed between domains A and B, with a relational value of
0.84. While domain C had a theoretical relational value with
the other two domains of 0.97, domain C has a more complex
relationship with domain A, as CBH-7 cross-competes with
CBH-4B but enhances CBH-4G binding. Because CBH-7 af-
fects these antibodies in opposite directions, it is possible that
CBH-7 is either in proximity to domain A or induces structural
changes that affect domain A. CBH-17 was the only antibody
to a linear epitope and did not inﬂuence the binding of the
other antibodies (data not shown). These studies suggest that
eight HCV HMAbs deﬁne three immunogenic domains con-
taining conformational epitopes on the HCV E2 glycoprotein.
HVR1 is not involved in conformational epitopes on E2. To
determine whether HVR1 is involved in the conformational
epitopes as deﬁned by domain A to C antibodies, constructs of
1b (pDN-411) and 1a strain H (pDNH-411) E2 without their
HVR (from aa 384 to 410) (Fig. 2A) were produced with HA
and c-myc tags at the N and C termini, respectively. To test
whether tag proteins have an effect on antibody binding, intra-
cellularly expressed 1a and 1b E2s without tags were tested and
shown to be without detectable differences by immunoﬂuores-
cence assay (data not shown). DNA sequencing conﬁrmed the
junction of deletion, and the expected sequences resulted in no
frameshifts or premature terminations. The expression of the
E2 deletion constructs was veriﬁed by Western blot analysis of
cytoplasmic extracts of transiently transfected HEK293 cells by
using a MAb to the tag-HA epitope (Fig. 2B). Wild-type and
HCV E2 deletion constructs were then transfected into
HEK293T cells, and intracellular forms of E2 and E2 deletions
were captured onto GNA lectin-coated microtiter plates. The
reactivities of the HCV HMAbs with E2 and the E2 deletion
were then determined. Anti-tag antibodies (HA and c-myc)
were used as positive controls for protein expression and for
GNA capture (Fig. 2C and D). All HCV HMAbs reacted with
wild-type genotype 1b sf1b E2 protein, and none reacted with
proteins captured from extracts of mock-transfected HEK293
cells (Fig. 2C). All HCV HMAbs retained reactivity with E2
produced by the pDN-411 deletion construct, indicating that
the epitopes recognized by the antibodies did not include
HVR1, although CBH-2 and -8E were reduced. No reactivity
was observed with a control antibody (R04) to either wild-type
E2 or the protein with E2 deleted. Antibody 3/11 was used as
a positive control in these studies, since the epitope of this
antibody has been deﬁned as being outside of HVR1 (aa 412 to
423) (17). Next, the same panel of antibodies was tested
against analogous genotype 1a E2 derived from strain H with
(sfH1a-E2) or without (pDNH-411) HVR1 (Fig. 2D). HMAbs
CBH-8C and CBH-11 did not recognize either sfH1a-E2 or
pDNH-411; CBH-2 had a signiﬁcant reduction. For CBH-11
this was expected, since this antibody does not react to 1a E2.
However, the reduced or lack of reactivity for CBH-2 and -8C
was more isolate speciﬁc, as previous studies showed these
antibodies binding equally well to other 1a E2 proteins (Table
1) (23). The other HCV HMAbs and control antibodies had
equivalent reactivities with the strain H-derived E2 proteins
and the genotype 1b E2 proteins. All HMAbs reactive with
sfH1a-E2 retained reactivity with the HVR1-deﬁcient con-
struct pDNH-411, conﬁrming that the epitopes recognized by
these HMAbs were outside HVR1. The reduction in binding of
CBH-2 and -8E to genotype 1b and not to 1a HVR1 proteins
with E2 deleted may reﬂect structural differences between ge-
notypes and the involvement of HVR1 in these two epitopes in
1b E2 but not with the other E2 HMAbs.
Conformational epitopes on E2 associated with HCVpp.
Only eight of nine HCV HMAbs to conformational epitopes
were tested with HCVpp, because of an inadequate amount of
CBH-8E antibodies. The hybridoma producing this HMAb
was unstable and will require alternative production for further
studies. Of the remaining eight antibodies, only CBH-5 and -7
had strong activity, and CBH-2 had weak activity, to block
HCVpp entry to target cells as shown previously (32). The
other antibodies had no neutralizing activity. To assess wheth-
er virus neutralization is caused by the expression or lack of
expression of their respective epitopes on E2 associated with
HCVpp, immunoprecipitation studies were performed with
HCVpp and cell lysate-associated E1E2 glycoproteins (Fig. 3).
It is worth noting that HCVpp-associated E2 had a slower and
more diffuse migration pattern than the cell-associated form.
This is due to modiﬁcations of the glycans by Golgi enzymes, as
previously shown (32). With HCVpp (Fig. 3A), CBH-4B, -4D,
and -7 showed strong binding, CBH-5 showed moderate bind-
ing, CBH-2 and -4G showed weak binding, and no detectable
binding was seen with CBH-8C or -11. The lack of reactivity
with CBH-8C and -11 was expected, since the HCVpp were
derived from the 1a H strain, which is not recognized by these
antibodies (Fig. 2D). For CBH-4G, a lower antibody afﬁnity
could be a contributing factor, as discussed below. Another
explanation is the masking of this epitope on the surface of
HCVpp. From these studies, it is reasonable to conclude that
conformational epitopes as identiﬁed by antibodies to domains
A, B, and C, CBH-2, -4B, -4D, -5, and -7, are present to some
degree on the virion surface, although virus neutralization is
restricted to domains B and C, CBH-2, -5, and -7. The strong
binding of domain A antibodies CBH-4D and -4B, while being
nonneutralizing, supports the view that virus neutralization for
9228 KECK ET AL. J. VIROL.
HCV is mediated in part by restricted virion surface E2
epitopes in speciﬁc domains.
Immunoprecipitation of intracellular E1E2 showed binding
with all conformation-sensitive antibodies except for CBH-8C
and -11. Interestingly, differences in reactivity were observed
between HCVpp-associated envelope proteins and their intra-
cellular forms. Domain B antibodies, CBH-2 and -5, had stron-
ger reactivities against cell-associated E1E2. Indeed, when
compared to CBH-7, these antibodies precipitated 4 to 5 times
less HCVpp-associated E2 protein. Since HCVpp-associated
envelope proteins are modiﬁed by Golgi enzymes (32), the
glycans added to these proteins in this compartment might
potentially reduce the accessibility of the epitopes recognized
by CBH-2 and -5. Alternatively, we cannot exclude some local
structural changes induced by modiﬁed glycans in domain B.
Antibody afﬁnity and virus neutralization. To estimate
afﬁnity, saturation of HCV HMAb binding to cell surface-
expressed genotype 1a and 1b E2 was measured by ﬂow cytom-
etry, and the data were analyzed using Prism software (Graph-
Pad). Saturation proﬁles for domain A-, B-, and C-speciﬁc
HMAbs to 1a E2 are shown in Fig. 4. As summarized in Table
2, the antibodies displayed a wide range of dissociation con-
). In general, afﬁnity tended to be higher to 1b E2
than to 1a E2, except with CBH-4G and -7. Moderate-to-high-
HCV Amino Acid
384 483 583
HCV anti body
Mean OD 405 nm
HCV anti body
Mean OD 405 nm
FIG. 2. Epitopes recognized by HCV HMAbs are located outside of HVR1. (A) Construct designation (on the left) showing the ﬁrst amino
acid sequence included in the two E2 deletion constructs employed in this study. All constructs were expressed in the pDisplay vector, which
includes a heterologous signal sequence and TM domain (solid black lines) as well as epitopes recognized by MAbs to inﬂuenza virus HA (vertical
bars) and c-myc (horizontal bars). HCV 1b E2 sequences are indicated in white, and strain H E2 sequences are shaded. (B) Western blot analysis
showing the expression of HCV E2 proteins. Wild-type 1b and 1a (strain H) E2 proteins are shown as sf1b and sfH1a, respectively. The HVR1
1b and 1a deletions (aa 384 to 410 were deleted) are shown as pDN-411 and pDNH-411, respectively. A protein size marker is indicated in
kilodaltons. (C) Reactivity of HCV HMAbs with E2 deletion constructs. HEK293 cells were mock transfected (white bars) or transfected with the
HCV E2 constructs sf1b (stippled bars) or pDN-411 (black bars). Twenty-four hours posttransfection, cytoplasmic extracts were prepared and
equivalent aliquots were captured onto GNA lectin-coated microtiter plates as described in Materials and Methods. The captured E2 proteins were
then incubated with 10 g of the indicated HCV HMAb/ml (x axis), and the amount of bound antibody was determined. Bars represent the mean
absorbance values obtained from duplicate wells. Error bars indicate one standard deviation from the mean. (D) Same experiment as in panel C,
except that HEK293 cells were transfected with sfH1a E2 (stippled bars) or pDNH-411 (black bars).
VOL. 78, 2004 IMMUNOGENIC DOMAINS IN HCV E2 9229
afﬁnity antibodies of ⬍5 ⫻ 10
were observed in all three
groups to 1b E2 and in domains A and C to 1a E2. In domain
B, CBH-5 had a similar afﬁnity with CBH-2 to 1a E2 but higher
neutralizing activity for HCVpp derived from strain H77. The
difference suggests that the number of CBH-5 epitopes is
greater than the number of CBH-2 epitopes on the surface of
HCVpp. This is supported by the observation that CBH-5 has
a higher total binding than CBH-2 to 1a E2, as shown in
Fig. 2D. CBH-4G weakly precipitated E1E2 associated with
HCVpp and weakly bound to intracellular E1E2 compared to
the other two antibodies in domain A. A possible explanation
is that this epitope is partly masked on the surface of HCVpp.
Collectively, there are distinct biological activities between
the antibody clusters (Table 2). Domain A antibodies, CBH-
4B, -4D, and -4G, had no neutralization activity but showed
greater recognition of HCVpp than intracellular E1E2, except
for CBH-4G. One of the antibodies, CBH-4D, in previous
studies was low-pH sensitive, with a 40% reduction in binding
(32). Domain B antibodies, CBH-5 and CBH-2, have neutral-
izing activities but show reduced recognition of HCVpp com-
pared to intracellular E1E2. The domain C antibody, CBH-7,
has neutralization activity and equal accessibility of its epitope
on E2 in HCVpp and intracellular E1E2.
Precise information on the mechanisms of virion attachment
and entry will be critical in the successful development of
newer therapeutics and an effective vaccine for HCV. Studies
with MAbs on some viruses tend to support two different views
on virus neutralization. One perspective is that virus neutral-
ization correlates with antibody binding afﬁnity to any virion
surface site and is irrespective of the epitopes recognized by
these antibodies. Antibody binding of a critical number of sites
on the virion surface prevents virus entry through steric hin-
drance (8). The other perspective is that speciﬁc antibodies
blocking distinct steps of virus attachment, interaction with
receptor and coreceptor, and initiation of viral envelope fusion
lead to virus neutralization (26). The development of HCVpp
provides a useful tool to study the functional roles of speciﬁc
immunogenic domains on envelope glycoproteins in virion
binding and entry. We and other colleagues recently showed that
HCVpp-associated envelope proteins are noncovalent E1E2 het-
erodimers, recognized by a panel of MAbs to conformational
epitopes and CD81, and some epitopes are sensitive to low-pH
treatment. Consequently, HCVpp are likely to contain enve-
lope proteins in a similar structure as native virions (32).
Of eight HCV HMAbs to conformational epitopes on E2,
only three antibodies, CBH-2, -5, and -7, were able to inhibit
HCVpp attachment and entry to Huh-7 cells (Table 2) (32).
Competition analyses of these HMAbs showed that conforma-
tional epitopes on E2 were clustered into three distinct do-
mains. Domain A consisted of CBH-4B, -4D, and -4G; domain
B contained CBH-2, -5, -8C, -8E, and -11; and domain C con-
tained HMAb CBH-7. In domain B, CBH-2 and -5 were neutral-
izing. The lack of neutralizing activity of the other two tested
antibodies, CBH-8C and -11, was explained by their lack of rec-
ognition of the genotype 1a H strain, which was used to construct
the HCVpp. The neutralizing antibodies in domains B and C
(CBH-2, -5, and -7) are to conformational epitopes on proteins
associated with the virion surface and do not involve HVR1, as
shown by HVR1 deletion studies. All antibodies in domains B
and C inhibited the E2-CD81 interaction (Table 1), supporting
the involvement of CD81 in virus entry (46). Domain A anti-
FIG. 3. Epitopes exposed on HCV glycoproteins associated with
HCVpp. HEK293T cells transfected to produce HCVpp were labeled
for 24 h. Supernatants (HCVpp) (A) and cell lysates (B) were immu-
noprecipitated with HCV HMAbs as noted.
FIG. 4. Saturation binding of HMAbs to HCV E2. Transiently
transfected 293T cells with 1a E2 were incubated with the HCV
HMAbs in increasing concentrations as indicated. Staining and ﬂow
cytometry analysis were performed as described in Materials and
Methods. The data points are means of two determinations and are
representative of three independent experiments. Binding afﬁnity data
shown in Table 2 were analyzed using Prism software.
9230 KECK ET AL. J. VIROL.
bodies were nonneutralizing and did not block the E2-CD81
interaction. Their ability to precipitate E1E2 associated with
HCVpp suggests that their epitopes are also on the surface of
virions, supporting the perspective that HCV virion attachment
and entry are restricted to speciﬁc virion surface domains.
This perspective is further supported by antibody afﬁnity
studies. CBH-4B has a moderate afﬁnity to 1a E2 that is higher
than that of CBH-2 or -5 (Table 2). But CBH-4B in domain A,
whose epitope is on E2 associated with the virion surface, is
nonneutralizing. Nonetheless, higher-afﬁnity antibodies will
tend to have higher neutralizing activities, with CBH-7 having
higher activity than CBH-2 or -5 (32). More studies of anti-
bodies in a speciﬁc domain, such as domain B, using HCVpp
constructed with genotype 1b are required to further support this
relationship of antibody afﬁnity and virus neutralization activity.
The clustering of these epitopes on HCV E2 into three
antigenic domains is in agreement with topological mapping by
similar MAb cross-competition studies with other ﬂavivirus E
glycoproteins and the crystal structure of tick-borne encepha-
litis virus E glycoprotein (35, 39). The ﬂavivirus E glycoprotein
is the dominant antigen inducing neutralizing antibodies and is
the protein responsible for virus attachment to cell receptors
and initiation of viral envelope fusion leading to cell entry (36,
38). Epitopes on ﬂavivirus E protein are clustered into three
structural domains. A central domain I containing nonneutral-
izing epitopes is felt to be a hinge region involved in low-pH-
induced conformational changes (39). Our HCV E2 domain A
has similar properties for nonneutralizing epitopes, and at least
one of the epitopes, CBH-4D, was shown to be pH sensitive,
with 40% binding reduction under low pH (Table 2). The
varied effects of CBH-7 on CBH-4G (increase) and CBH-4B
(decrease) in binding to E2 suggest that a conformational
change is possibly induced with CBH-7 binding, consistent with
domain A as a possible hinge region. Flavivirus E protein
domain II is involved in dimerization and membrane fusion
and is able to elicit neutralizing and nonneutralizing antibodies
(35, 39). For HCV, the identity of the fusion protein is unclear.
At ﬁrst, a putative fusion peptide in E1 led to the proposal that
this protein is the fusion protein (18). But, other homology
studies suggested that E2 is responsible for virus-induced fu-
sion (27, 45). Flavivirus E protein domain III containing distal
projecting loops from the virion surface elicits the strongest
neutralizing antibodies, is minimally affected by low pH, and is
felt to be the receptor binding motif (35, 38, 39). Currently, it
is not possible to correlate our HCV domain B and C antibod-
ies in a similar manner. Both domains contain neutralizing
antibodies and are able to inhibit the E2-CD81 interaction.
One possible clue is that previous studies showed that CBH-2
is uniquely able to recognize noncovalent E1E2 heterodimers
and not high-molecular-weight E1E2 aggregates that are mis-
folded. Furthermore, CBH-2 will only recognize 1a H strain E2
when complexed with E1 (11). These ﬁndings suggest that the
HCV domain B antibodies are potentially correlated with the
ﬂavivirus E protein domain II involved in envelope protein
dimerization. Further studies are required to substantiate this
model for HCV. Differences are to be expected in the struc-
tural organization of HCV E2 and ﬂavivirus E glycoproteins,
since a major determinant is the number of disulfate bonds,
which are different between these glycoproteins. Epitope map-
ping and more detailed investigation on structure-function
properties of this panel of HCV HMAbs will be useful to
advance our knowledge on E2 immunogenic structures, which
in turn should facilitate effective vaccine design.
We thank J. Rowe and S. Rajyaguru for technical assistance.
This work was supported in part by NIH grants HL079381 and
AI47355 to S.K.H.F. J.D. was supported by EU grant QLRT-2001-
01329 and grants from the Agence Nationale de Recherche sur le Sida
and the Association pour la Recherche sur le Cancer.
1. Agnello, V., G. Abel, M. Elfahal, G. V. Knight, and Q. X. Zhang. 1999.
Hepatitis C virus and other ﬂaviviridae viruses enter cells via low density
lipoprotein receptor. Proc. Natl. Acad. Sci. USA 96:12766–12771.
2. Allander, T., K. Drakenberg, A. Beyene, D. Rosa, S. Abrignani, M. Hough-
ton, A. Widell, L. Grillner, and M. A. A. Persson. 2000. Recombinant human
monoclonal antibodies against different conformational epitopes of the E2
envelope glycoprotein of hepatitis C virus that inhibit its interaction with
CD81. J. Gen. Virol. 81:2451–2459.
3. Alter, H. J. 1995. To C or not to C: these are the questions. Blood 85:1681–
4. Bartosch, B., J. Dubuisson, and F. L. Cosset. 2003. Infectious hepatitis C
pseudoparticles containing functional E1E2 envelope protein complexes. J.
Exp. Med. 197:633–642.
5. Bartosch, B., A. Vitelli, C. Granier, C. Goujon, J. Dubuisson, S. Pascale, E.
Scarselli, R. Cortese, A. Nicosia, and F. L. Cossett. 2003. Cell entry of
hepatitis C virus requires a set of co-receptors that include the CD81 tet-
raspanin and the SR-B1 scavenger receptor. J. Biol. Chem. 278:41624–41630.
TABLE 2. Comparison of biological activities of HCV HMAbs
IP-E1E2 associated with:
1b 1a Lysate
A CBH-4B 1.62 ⫻ 10
4.50 ⫻ 10
34 55 3 ⫺
CBH-4D 4.58 ⫻ 10
1.67 ⫻ 10
64 77 40 ⫺
CBH-4G 1.22 ⫻ 10
9.03 ⫻ 10
Weak ⫺ NT ⫺
B CBH-2 1.27 ⫻ 10
1.10 ⫻ 10
77 16 17 ⫹ (weak)
CBH-5 9.11 ⫻ 10
2.22 ⫻ 10
82 26 10 ⫹
CBH-8C 2.72 ⫻ 10
No binding ⫺⫺ NT ⫺
CBH-11 6.60 ⫻ 10
No binding ⫺⫺ NT ⫺
C CBH-7 1.88 ⫻ 10
1.11 ⫻ 10
87 93 3 ⫹
Grouping of HCV HMAbs based on results of competition and phylogenetic analyses.
Afﬁnity disassociation of antibodies with genotypes 1b (GenBank AF348705) and 1a (AF009606).
Reported in arbitrary quantitative units. “Weak”, low reactivity; ⫺, no reactivity.
HMAb binding ability to HCVpp after low-pH treatment. Percentage denotes reduction in binding (32). NT, not tested.
Ability of HMAbs to neutralize HCVpp infectivity of Huh-7 cells. “⫹” or “⫺” denotes neutralization or no neutralization (32).
VOL. 78, 2004 IMMUNOGENIC DOMAINS IN HCV E2 9231
6. Bukh, J., R. H. Miller, and R. H. Purcell. 1995. Genetic heterogeneity of
hepatitis C virus: quasispecies and genotypes. Semin. Liver Dis. 15:41–62.
7. Burioni, R., P. Plaisant, A. Manin, D. Rosa, V. Delli Carri, F. Bugli, L.
Solforosi, S. Abrignani, P. E. Varaldo, G. Fadda, and M. Clementi. 1998.
Dissection of human humoral immune response against hepatitis C virus E2
glycoprotein by repertoire cloning and generation of recombinant Fab frag-
ments. Hepatology 28:810–814.
8. Burton, D. R., E. Q. Saphire, and P. W. H. I. Parren. 2001. A model for
neutralization of viruses based on antibody coating of the virion surface.
Curr. Top. Microbiol. Immunol. 260:109–144.
9. Cardoso, M. S., K. Siemoneit, D. Sturm, C. Krone, D. Moradpour, and B.
Kubanek. 1998. Isolation and characterization of human monoclonal antibodies
against hepatitis C virus envelope glycoproteins. J. Med. Virol. 55:28–34.
10. Chan, H. C., K. G. Hadlock, S. K. H. Foung, and S. Levy. 2001. VH1–69 gene
is preferentially used by hepatitis C virus-associated B cell lymphomas and by
normal B cells responding to the E2 viral antigen. Blood 97:1023–1026.
11. Cocquerel, L., E. R. Quinn, M. Flint, K. G. Hadlock, S. K. H. Foung, and S.
Levy. 2003. Recognition of native hepatitis C virus E1E2 heterodimers by a
human monoclonal antibody. J. Virol. 77:1604–1609.
12. Deleersnyder, V., A. Pillez, C. Wychowski, K. Blight, J. Xu, Y. S. Hahn, C. M.
Rice, and J. Dubuisson. 1997. Formation of native hepatitis C virus glyco-
protein complexes. J. Virol. 71:697–704.
13. Dubuisson, J., and C. M. Rice. 1996. Hepatitis C virus glycoprotein folding:
disulﬁde bond formation and association with calnexin. J. Virol. 70:778–786.
14. Farci, P., A. Shimoda, A. Coiana, G. Diaz, G. Peddis, J. C. Melpolder, A.
Strazzera, D. Y. Chien, S. J. Munoz, A. Balestrieri, R. H. Purcell, and H. J.
Alter. 2000. The outcome of acute hepatitis C predicted by the evolution of
the viral quasispecies. Science 288:339–344.
15. Farci, P., A. Shimoda, D. Wong, T. Cabezon, D. De Gioannis, A. Strazzera,
Y. Shimizu, M. Shapiro, H. J. Alter, and R. H. Purcell. 1996. Prevention of
hepatitis C virus infection in chimpanzees by hyperimmune serum against
the hypervariable region 1 of the envelope 2 protein. Proc. Natl. Acad. Sci.
16. Fitch, W. M., and E. Margoliash. 1967. Construction of phylogenetic trees.
17. Flint, M., C. Maidens, L. D. Loomis-Price, C. Shotton, J. Dubuisson, P.
Monk, A. Higginbottom, S. Levy, and J. A. McKeating. 1999. Characteriza-
tion of hepatitis C virus E2 glycoprotein interaction with a putative cellular
receptor, CD81. J. Virol. 73:6235–6244.
18. Flint, M., J. M. Thomas, C. M. Maidens, C. Shotton, S. Levy, W. S. Barclay,
and J. A. McKeating. 1999. Functional analysis of cell surface-expressed
hepatitis C virus E2 glycoprotein. J. Virol. 73:6782–6790.
19. Forns, X., R. Thimme, S. Govindarajan, S. U. Emerson, R. H. Purcell, F. V.
Chisari, and J. Bukh. 2000. Hepatitis C virus lacking the hypervariable
region 1 of the second envelope protein is infectious and causes acute
resolving or persistent infection in chimpanzees. Proc. Natl. Acad. Sci. USA
20. Gardner, J. P., R. J. Durso, R. R. Arrigale, G. P. Donovan, P. J. Maddon, T.
Dragic, and W. C. Olson. 2003. L-SIGN (CD 209L) is a liver-speciﬁc capture
receptor for hepatitis C virus. Proc. Natl. Acad. Sci. USA 100:4498–4503.
21. Grakoui, A., C. Wychowski, C. Lin, S. M. Feinstone, and C. M. Rice. 1993.
Expression and identiﬁcation of hepatitis C virus polyprotein cleavage prod-
ucts. J. Virol. 67:1385–1395.
22. Habersetzer, F., A. Fournillier, J. Dubuisson, D. Rosa, S. Abrignani, C.
Wychowski, I. Nakano, C. Treppo, C. Desgranges, and G. Inchauspe. 1998.
Characterization of human monoclonal antibodies speciﬁc to the hepatitis C
virus glycoprotein E2 with in vitro binding neutralization properties. Virol-
23. Hadlock, K. G., R. E. Lanford, S. Perkins, J. Rowe, Q. Yang, S. Levy, P.
Pileri, S. Abrignani, and S. K. H. Foung. 2000. Human monoclonal antibod-
ies that inhibit the binding of hepatitis C virus E2 protein to CD81 and
recognizing conserved conformational epitopes. J. Virol. 74:10407–10416.
24. Houghton, M. 1996. Hepatitis C viruses, p. 1035–1058. In B. N. Fields, D. M.
Knipe, and P. M. Howley (ed.). Fields virology, 3rd ed. Lippincott-Raven,
25. Hsu, M., J. Zhang, M. Flint, C. Logvinoff, C. Cheng-Mayer, C. M. Rice, and
J. A. McKeating. 2003. Hepatitis C virus glycoproteins mediate pH-depen-
dent cell entry of pseudotyped retroviral particles. Proc. Natl. Acad. Sci.
26. Klasse, P. J., and Q. J. Sattentau. 2001. Mechanism of virus neutralization
by antibody. Curr. Top. Microbiol. Immunol. 260:87–108.
27. Lescar, J., A. Roussel, M. W. Wien, J. Navaza, S. D. Fuller, G. Wengler, and
F. A. Rey. 2001. The fusion glycoprotein shell of Semliki Forest virus: an
icosahedral assembly primed for fusogenic activation at endosomal pH. Cell
28. Levy, S., S. C. Todd, and H. T. Maecker. 1998. CD81 (TAPA-1): a molecule
involved in signal transduction and cell adhesion in the immune system.
Annu. Rev. Immunol. 16:89–109.
29. Lindenbach, B. D., and C. M. Rice. 2001. Flaviviridae: the viruses and their
replication, p. 991–1042. In D. M. Knipe, P. M. Howley, D. E. Grifﬁn, R. A.
Lamb, M. A. Martin, B. Roizman, and S. E. Straus (ed.), Fields virology, 4th
ed. Lippincott Williams & Wilkins, Philadelphia, Pa.
30. Lozach, P. Y., H. Lortat-Jacob, A. De Lacroix De Lavalette, I. Staropoli, S.
Foung, A. Amara, C. Houles, F. Fieschi, O. Schwartz, J. L. Virelizier, F.
Arenzana-Seisdedos, and R. Altmeyer. 2003. DC-SIGN and L-SIGN are
high-afﬁnity binding receptors for hepatitis C virus glycoprotein E2. J. Biol.
31. Moore, J. P., and J. Sodroski. 1996. Antibody cross-competition analysis of
the human immunodeﬁciency virus type 1 gp120 exterior envelope glycopro-
tein. J. Virol. 70:1863–1872.
32. Op De Beeck, A., C. Voisset, B. Bartosch, Y. Ciczora, L. Cocquerel, Z. Y.
Keck, S. Foung, F. L. Cosset, and J. Dubuisson. 2004. Characterization of
functional hepatitis C virus envelope glycoproteins. J. Virol. 78:2994–3002.
33. Pohlmann, S., J. Zhang, F. Baribaud, Z. Chen, G. J. Leslie, G. Lin, A.
Granelli Piperno, R. W. Doms, C. M. Rice, and J. A. McKeating. 2003.
Hepatitis C virus glycoproteins interact with DC-SIGN and DC-SIGNR.
J. Virol. 77:4070–4080.
34. Ray, S. C., Y. M. Wang, O. Laeyendecker, J. R. Ticehurst, S. A. Villano, and
D. L. Thomas. 1999. Acute hepatitis C virus structural gene sequences as
predictors of persistent viremia: hypervariable region 1 as a decoy. J. Virol.
35. Rey, F. A., F. X. Heinz, C. Mandl, C. Kunz, and S. G. Harrison. 1995. The
envelope glycoprotein from tick-borne encephalitis virus at 2 A resolution.
Nature (London) 375:291–298.
36. Rice, C.M. 1996. Flaviviridae: the viruses and their replication, p. 931–959. In
B. N. Fields, D. M. Knipe, and P. M. Howley (ed.), Fields virology, 3rd ed.
Lippincott-Raven, Philadelphia, Pa.
37. Robertson, B., G. Meyers, C. Howard, T. Brettin, J. Bukh, B. Gaschen, T.
Gojobori, G. Maertens, M. Mizokami, O. Nainan, S. Netesov, K. Nishioka,
T. Shini, P. Simmonds, D. Smith, L. Stuyver, and A. Weiner. 1988. Classi-
ﬁcation, nomenclature, and database development for hepatitis C virus
(HCV) and related viruses: proposals for standardization. Arch. Virol. 143:
38. Roehrig, J. T. 1997. Immunochemistry of the dengue viruses, p. 199–219. In
D. J. Gubler and G. Kuno (ed.), Dengue and dengue hemorrhagic fever.
CAB International, New York, N.Y.
39. Roehrig, J. T., R. A. Bolin, and R. G. Kelly. 1998. Monoclonal antibody
mapping of the envelope glycoprotein of the dengue 2 virus, Jamaica. Vi-
40. Scarselli, E., H. Ansuini, R. Cerino, R. M. Roccasecca, S. Acali, G. Filocamo,
C. Traboni, A. Nicosia, R. Cortese, and A. Vitelli. 2002. The human scaven-
ger receptor class B type I is a novel candidate receptor for the hepatitis C
virus. EMBO J. 21:5017–5025.
41. Shimizu, Y. K., M. Hijikata, A. Iwamoto, H. J. Alter, R. H. Purcell, and H.
Yoshikura. 1994. Neutralizing antibodies against hepatitis C virus and the
emergence of neutralization escape mutant viruses. J. Virol. 68:1494–1500.
42. Sneath, P. H. A., and R. R. Sokal. 1973. Numerical taxonomy, p. 230–234.
W.H. Freeman and Company, San Francisco, Calif.
43. Wack, A., E. Soldaini, C.-T. K. Tseng, S. Nuti, G. R. Klimpel, and S. Abrignani.
2001. Binding of the hepatitis C virus envelope protein E2 to CD81 provides
co-stimulatory signal for human T cells. Eur. J. Immunol. 31:166–175.
44. Wunschmann, S., J. D. Medh, D. Klinzmann, W. N. Schmidt, and J. T.
Stapleton. 2000. Characterization of hepatitis C virus (HCV) and HCV E2
interactions with CD81 and the low density lipoprotein receptor. J. Virol.
45. Yagnik, A. T., A. Lahm, A. Meola, R. M. Roccasecca, B. B. Ercole, A. Nicosia,
and A. Tramontano. 2000. A model for the hepatitis C virus envelope gly-
coprotein E2. Proteins 40:355–366.
46. Zhang, J., G. Randall, A. Higginbottom, P. Monk, C. M. Rice, and J. A.
McKeating. 2004. CD81 is required for hepatitis C virus glycoprotein-medi-
ated viral infection. J. Virol. 78:1448–1455.
47. Zibert, A., E. Schreier, and M. Roggendorf. 1995. Antibodies in human sera
speciﬁc to hypervariable region 1 of hepatitis C virus can block viral attach-
ment. Virology 208:653–661.
9232 KECK ET AL. J. VIROL.