Gastroenterology 2020;158:1058–1071

Antibody Responses to Immunization With HCV Envelope

Glycoproteins as a Baseline for B-Cell–Based Vaccine
Development
Fang Chen,1 Kenna Nagy,1 Deborah Chavez,2 Shelby Willis,3 Ryan McBride,3 Erick Giang,1
Andrew Honda,1 Jens Bukh,4,5 Phillip Ordoukhanian,3 Jiang Zhu,6 Sharon Frey,7
Robert Lanford,2 and Mansun Law1
1
Department of Immunology and Microbiology, The Scripps Research Institute, La Jolla, California; 2Southwest National
Primate Research Center at Texas Biomedical Research Institute, San Antonio, Texas; 3Next-Generation Sequencing and
Microarray Research Cores, The Scripps Research Institute, La Jolla, California; 4Copenhagen Hepatitis C Program,
Department of Infectious Diseases, Copenhagen University Hospital, Hvidovre, Denmark; 5Department of Immunology and
Microbiology, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark; 6Department of
Integrative Structural and Computational Biology, The Scripps Research Institute, La Jolla, California; and 7Saint Louis
University Center for Vaccine Development, St Louis, Missouri

Immunization Antibody function and specificity Future HCV vaccine

development
¾ Antibody function
HCV E1E2 Non-nAb
Non nAb
¾ Preclinical animal model
Serum
S erum
Plasma
Plasma nAb The NHP model recapitulates
the human antibody responses
Polyclonal
olyclona to vaccination by targeting
antibodies ¾ Epitope mapping similar neutralizing epitopes
P AS192 746
74

¾ Rational vaccine design

Epitope-based strategy
PBMC
TRANSLATIONAL LIVER

Antibody competition B cell ontology-based strategy

1
BASIC AND

Single cell Antigen-specific VH1-69 encoded

sorting Memory B cells Monoclonal ¾ AR3-specific bnAbs
Germline gene
Plasmablasts antibodies CDR region
SHM

non-neutralizing epitopes on the E1 N-terminus and E2 hy-

See editorial on page 820.
pervariable region 1 did not differ significantly. Antibodies
from volunteers and NHPs that neutralized heterologous
BACKGROUND & AIMS: We investigated antibody responses to strains of HCV primarily interacted with epitopes in the
hepatitis C virus (HCV) antigens E1 and E2 and the relevance of antigen region 3. However, the neutralizing antibodies
animal models for vaccine development. We compared anti- were not produced in sufficient levels for broad neutrali-
body responses to vaccination with recombinant E1E2 complex zation of diverse HCV isolates. Broadly neutralizing anti-
in healthy volunteers, non-human primates (NHPs), and mice. bodies similar to the human VH1-69 class antibody specific
METHODS: We analyzed 519 serum samples from participants for antigen region 3 were produced in the immunized
in a phase 1 vaccine trial (ClinicalTrials.gov identifier NHPs. CONCLUSIONS: In an analysis of vaccinated volun-
NCT00500747) and compared them with serum or plasma teers, NHPs, and mice, we found that recombinant E1E2
samples from C57BL/6J mice (n ¼ 28) and rhesus macaques vaccine antigen induces high-antibody titers that are insuf-
(n ¼ 4) immunized with the same HCV E1E2 antigen. Blood ficient to neutralize diverse HCV isolates. Antibodies from
samples were collected at different time points and analyzed volunteers and NHPs bind to the same neutralizing epitopes
for antibody binding, neutralizing activity, and epitope speci- for virus neutralization. NHPs can therefore be used as a
ficity. Monoclonal antibodies from the immunized NHPs were preclinical model to develop HCV vaccines. These findings
isolated from single plasmablasts and memory B cells, and their also provide useful baseline values for development of vac-
immunogenetic properties were characterized. RESULTS: cines designed to induce production of neutralizing
Antibody responses of the volunteers, NHPs, and mice to the antibodies.
March 2020 Antibody Responses to HCV E1E2 Immunization 1059

Keywords: HVR1; AR3; Animal Model; mAb.

WHAT YOU NEED TO KNOW

H epatitis C virus (HCV) chronically infects an esti-

mated 71 million people worldwide, predisposing
them to risk of developing life-threatening liver cirrhosis
BACKGROUND AND CONTEXT
Studies are needed of antibody responses to hepatitis C
virus (HCV) antigens E1 and E2 in animal models and
and hepatocellular carcinoma. Although direct-acting anti- clinical trials.
viral drugs to HCV have dramatically improved the rate of
NEW FINDINGS
cure, treatment alone will unlikely be sufficient to achieve
HCV elimination. Limited access to care, high cost of ther- The authors compared antibody responses among
volunteers, non-human primates (NHPs), and mice given
apy, poor awareness of infection status, drug resistance,
the Chiron HCV vaccine candidate E1E2. These
reinfection after cure, and the rising rate of new infections antigens induced high antibody titers, but were
illustrate a pressing need for a broadly effective vaccine to insufficient to neutralize diverse HCV isolates. The
stop global HCV transmission.1,2 immunized NHPs used a similar antibody gene found in
HCV vaccine development has been stymied by the humans to produce cross-neutralizing antibodies.
extreme genetic diversity of circulating viruses, the LIMITATIONS
numerous mechanisms through which HCV evades the im-
B cells from participants in the phase 1 trial were
mune response, and the lack of a preclinical animal model
unavailable for immunogenetic comparison with B cells
for vaccination-challenge trials. The chimpanzee is the only from the NHPs.
non-human species susceptible to persistent HCV infection,
but financial, practical, and especially ethical constraints IMPACT
have urged the exploration of alternative animal models. NHPs can be used as a model for development of HCV
Nevertheless, cumulative evidence of protective cellular and vaccines. These findings provide useful baseline values
humoral immunity during HCV infection suggests that for development of vaccines designed to elicit
neutralizing antibodies.
vaccination to prevent viral persistence is feasible.3,4
Several strategies to elicit neutralizing antibodies (nAbs),
T-cell responses, or both, have demonstrated immunoge- Most potent and bnAbs isolated so far have been mapped to
nicity of vaccine antigens and, in some cases, even protec- the E2 neutralizing face, a predominantly hydrophobic
tive immunity in the chimpanzee model.5–9 Yet there surface formed by the E2 front layer and the tip of the CD81
remains the open question of whether animals can faithfully binding loop,12 and to the E1E2 heterodimer.13 Early im-
recapitulate human immune responses to HCV vaccination. munization studies of the E1E2 candidate in animal models
Two major promising vaccine candidates had been assessed (chimpanzee, guinea pig, and mouse) and humans have
previously in humans. The first one is based on recombinant yielded various results on its ability to induce antiviral
full-length HCV envelope glycoproteins E1 and E2 derived antibody responses, ranging from cross-neutralizing10,14,15
to neutralizing14,16,17 to virus-enhancing18 to epitope-

TRANSLATIONAL LIVER
from a genotype 1a isolate, HCV-1. In a phase 1 clinical trial
(ClinicalTrials.gov identifier NCT00500747), immunization interfering.19 The discrepancy could be a result of the

BASIC AND
of healthy human volunteers with the E1E2 candidate eli- variability of the assays and reagents used in different
cited broadly nAbs (bnAbs) in only a few subjects.10 The studies. Here, we immunized rhesus macaques and mice
second one composed of nonstructural proteins NS3–NS5 with the same antigens and compared their antibody re-
has been evaluated recently in phase 1/2 clinical trials sponses directly with those of immunized humans.
(ClinicalTrials.gov identifier NCT01436357). According to
the announcement by the National Institute of Allergy
and Infectious Diseases at an HCV vaccine development Methods
workshop (https://www.niaid.nih.gov/news-events/trial- Human Serum Samples
evaluating-experimental-hepatitis-c-vaccine-concludes), and A total of 519 human serum samples were obtained from a
the results recently released at ClinicalTrials.gov, it failed to completed phase 1, placebo-controlled, dose-escalation clinical
elicit protective immunity in high-risk people who inject trial (DMID 01-002; ClinicalTrials.gov identifier NCT00500747)
drugs, although high frequencies of virus-specific poly- in which a candidate HCV vaccine constituting a recombinant
functional CD4 and CD8 T cells were elicited in healthy E1E2 immunogen derived from a genotype 1a isolate (HCV-1)
volunteers.11 The next-generation HCV vaccine strategies
will likely require rationally designed antigens targeting Abbreviations used in this paper: AR, antigenic region; AS, antigen site;
conserved epitopes to overcome viral variability. A reliable ASC, antibody-secreting cell; bnAb, broadly neutralizing antibody; CDR,
animal model and a better understanding of vaccine-elicited complementarity-determining region; CDRH, heavy chain
complementarity-determining region; HCV, hepatitis C virus; HCVpp,
immune responses will be essential to facilitate future HCV hepatitis C virus pseudoparticles; HVR1, hypervariable region 1; mAb,
vaccine development. monoclonal antibody; nAb, neutralizing antibody; NHP, non-human pri-
mate; PBMC, peripheral blood mononuclear cell; PCR, polymerase chain
In this study, we sought to gain a comprehensive insight reaction.
into antibody response to HCV glycoproteins and to evaluate Most current article
preclinical animal models for vaccine development. We took
© 2020 by the AGA Institute
advantage of the recombinant E1E2 vaccine candidate uti- 0016-5085/$36.00
lized in the clinical trial. E1 and E2 are targets of HCV nAbs. https://doi.org/10.1053/j.gastro.2019.11.282
 1060 Chen et al Gastroenterology Vol. 158, No. 4

formulated with the MF59 adjuvant (Novartis Vaccines and corresponding expression plasmids encoding the E1E2 from
Diagnostics, Basel, Switzerland) was tested for safety and isolate HCV-1 (genotype 1a), H77 (genotype 1a), UKN1B 12.6
immunogenicity in healthy human volunteers.16 All samples (genotype 1b), J6 (genotype 2a), S52 (genotype 3a), UKN4.11.1
were heat-inactivated at 56 C for 30 minutes before all assays (genotype 1a), or SA13 (genotype 5a) at a 4:1 ratio by poly-
to inactivate complement. ethylenimine (Polysciences, Warrington, PA). Neutralization
was carried out on Huh-7 cells with diluted sera or monoclonal
Non-Human Primate Immunization antibodies (mAbs), as described previously.20 For HCV cell
Four male rhesus macaques (Macaca mulatta) of India culture neutralization, adapted HCV recombinant HCV-1, H77,
origin, 4–5 years of age, and weighted between 8 and 11 kg, and SA13(Core-NS2)/JFH1 were propagated in Huh-7.5.1 cells,
designated 30734, 31782, 31859, and 31881, were housed at as described previously.21 Virus neutralization was determined
Southwest National Primate Research Center at Texas by measurement of the percentage reduction of the number of
Biomedical Research Institute. Animals were immunized infectious foci. See the Supplementary Materials for further
intramuscularly 5 times at 0, 4, 12, 25 or 28, and 42 weeks with details.
the same immunogen utilized in the clinical trial. Each immu-
nization consisted of 2 intramuscular injections in the quadri-
Enzyme-Linked Immunospot
ceps of each leg with a total of 50 mg HCV E1E2 formulated in
E1E2-specific antibody-secreting cells (ASCs) were detected
AddaVax (InvivoGen, San Diego, CA) or Adjuplex (Advanced
by enzyme-linked immunospot assay on fresh PBMCs, as
Bioadjuvants, Omaha, NE and Sigma-Aldrich, St Louis, MO)
described previously, with modifications.22 See the
adjuvant. Whole blood was collected and processed for plasma
Supplementary Materials for further details.
and peripheral blood mononuclear cells (PBMCs) using Ficoll-
Paque Plus (GE Healthcare, Chicago, IL) and Leucosep
(Greiner Bio-One, Kremsmünster, Austria) tubes according to Flow Cytometry and Single B-Cell Sorting
manufacturers’ instructions. Plasma samples were heat- For analytical flow cytometry, PBMCs were surface-stained
inactivated at 56 C for 30 minutes before all assays. All pro- with a panel of antibodies (Supplementary Table 1) for 30
cedures and experiments in NHPs were performed in accor- minutes in the dark at 4 C and washed twice with fluorescence-
dance with protocols reviewed and approved by the activated cell sorting buffer composed of 2% fetal bovine serum
Institutional Animal Care and Use Committees of Texas and 2 mM EDTA in phosphate-buffered saline. Samples were
Biomedical Research Institute and The Scripps Research acquired with LSR-II flow cytometer (BD Bioscience, Franklin
Institute. Lakes, NJ) and FlowJo software, version 10 (Tree Star, Ashland,
OR) was used for analysis. For sorting, 25–50 million fresh
Mouse Immunization PBMCs were stained and sorted on the BD FACSAria III (BD
Twenty-eight female mice (C57BL/6) approximately 6–8 Biosciences). Single plasmablasts (CD3–CD20–CD80þHLA-DRþ)
weeks old were housed at The Scripps Research Institute. An- and E2-specific memory B cells (CD3–CD20þCD27þIgGþ E2þ)
imals were immunized subcutaneously 4 times in 4-week in- were collected into 96-well polymerase chain reaction (PCR)
TRANSLATIONAL LIVER

tervals with the same immunogen utilized in humans and NHPs plates with 4 mL lysis buffer containing RNaseOUT (Invitrogen,
and formulated in AddaVax or Adjuplex adjuvant (25 mg im- Carlsbad, CA) and dithiothreitol. Plates were flash-frozen on
BASIC AND

munogens for prime immunization and 5 mg immunogens for dry ice and immediately transferred to –80 C.
boosters). Serum samples were collected at 0, 9, and 13 weeks,
and heat-inactivated at 56 C for 30 minutes before analyses. All
procedures and experiments in mice were performed in Generation of Monoclonal Antibodies
accordance with protocols reviewed and approved by the Immunoglobulin variable genes from single-sorted B cells
Institutional Animal Care and Use Committees of The Scripps were amplified by reverse-transcription PCR and nested PCR
Research Institute. reactions, as described previously.23 PCR products were sent
for sequencing (Retrogen, San Diego, CA) before cloning into
human Igg1, Igk, and Igl expression vectors.24 Plasmids con-
Enzyme-Linked Immunosorbent Assay taining paired antibody heavy- and light-chain genes were
Binding enzyme-linked immunosorbent assay and compe- cotransfected (1:1 ratio) into HEK293T or ExpiCHO cells using
tition enzyme-linked immunosorbent assay were performed to polyethylenimine (Polysciences) or ExpiFectamine CHO Trans-
assess antibody titers and to map discontinuous epitopes, fection Kit (Thermo Fisher Scientific, Waltham, MA), respec-
respectively. See the Supplementary Materials for further tively. Antibody-containing supernatants were harvested 3–14
details. days after transfection. Antibodies produced in HEK293T cell
were quantified by anti-human IgG Fc and used directly in
Pepscan Analysis neutralization assays for screening purposes. Antibody super-
A library of peptides consisting of 15 amino acids in length natants produced in ExpiCHO cells were purified over Protein
and overlapped by 12 residues spanning the full length of HCV-1 A-Sepharose 4 Fast Flow (GE Healthcare) columns per manu-
envelope glycoproteins was used in pepscan to map continuous facture’s instruction.
epitopes. See the Supplementary Materials for further details.
Bioinformatics
Hepatitic C Virus Neutralization Assays Antibody sequences were submitted to IgBLAST (https://
HCV pseudoparticles (HCVpp) were generated by co- www.ncbi.nlm.nih.gov/igblast/) and ImMunoGeneTics infor-
transfection of 293T cells with pNL4-3.lucR-E- and the mation system (http://www.imgt.org/) for gene identification
March 2020 Antibody Responses to HCV E1E2 Immunization 1061

and genetic assignment. Multiple-sequence alignments were In contrast, 23% of immunized humans and NHP 31859 had
performed using MegAlign Pro program in Lasergene, version a transient response that waned precipitously and became
15.3 (DNAstar, Madison, WI). Maximum-likelihood phylogenetic non-neutralizing at the final time point. No significant dif-
tree was constructed using MEGA X software. ference was found in the magnitude of the anti-E1E2 anti-
body titers among the prolonged neutralizers, transient
neutralizers, and non-neutralizers (Figure 1E and F, left).
Statistics There is no difference in the proportion of responders
Graphpad Prism, version 8.02 (GraphPad, San Diego, CA) among the human dosage groups for prolonged nAb re-
was used for all statistical analyses. The significance of differ-
sponses, but there appears to be a trend for more transient
ences in antibody response between groups was calculated
nAb responders in the 20-mg and 100-mg dosage groups
using unpaired, 2-tailed Mann-Whitney U tests. Antibody re-
than in the 4-mg dosage group (Figure 1G and
sponses by dose in immunized humans were compared using 1-
way analysis of variance test. Correlations in data were
Supplementary Figure 1B). In the mouse model, most ani-
assessed using a 2-tailed Pearson correlation coefficient. An r mals reached peak antibody responses after the fourth im-
value between 0 and 0.3 indicates a negligible positive corre- munization (Figure 1H). Overall, the animals mounted an
lation, between 0.3 and 0.5 indicates a weak correlation, be- antibody response that was comparable to, or even stronger
tween 0.5 and 0.7 indicates a moderate correlation, and than, that of the immunized humans (Figure 1B). An overall
between 0.7 and 1.0 indicate a strong correlation. P values <.05 weak-to-moderate correlation between autologous E1E2-
were considered significant. binding and nAb response was observed for humans (r ¼
0.48, P < .0001) and NHPs (r ¼ 0.55, P < .0001), but not for
mice (r ¼ 0.26, P ¼ .0531) (Figure 1I).
Results
Immunization With Hepatitis C Virus E1E2 Elicits Human and Non-Human Primate Neutralizing
Comparable Antibody Responses in Humans and Antibodies Exhibit Cross-Binding Reactivity and
Animal Models Neutralization Potential Against Heterologous
NHP rhesus macaques (n ¼ 4) and C57BL/6J mice (n ¼ Viral Strains
28) were immunized with the same E1E2 antigen formu- To assess the antibody functions, selected human, NHP,
lated in AddaVax,25 a squalene-based oil-in-water emulsion and mouse immune samples with autologous neutralization
similar to MF59 that was utilized in the clinical trial, or were tested for cross-reactivity against a panel of viral
Adjuplex,26,27 a non-oil–based adjuvant that has demon- isolates representing 6 major HCV genotypes at a dilution of
strated superiority to alum, Freund, and numerous other 1:50. Similar profiles of cross-reactivity were observed for
experimental adjuvants (Figure 1A). In parallel, longitudinal humans and NHPs (Figure 2). In both cases, all samples
serum samples collected from 56 human subjects in the tested bound to native and denatured E1E2 from diverse

TRANSLATIONAL LIVER
clinical trial were re-analyzed and compared to the animal HCV genotypes (Figure 2A). As expected, the highest reac-
samples (plasma for NHPs and serum for mice) using the tivity was observed against E1E2 from the autologous ge-

BASIC AND
same experimental methods and controls. Initial screening notype 1a isolate HCV-1, followed by viral strains H77 and
of the immune samples at a dilution of 1:50 demonstrated UKN1B12.6 from the same genotype. Notably, the NHP
that all immunized subjects developed high titers of anti- plasma had a broader reactivity across the genotypes
E1E2 antibodies, with the peak end-point titers ranging compared to the human sera, likely caused by their higher
from 19,107 to 321,951; 108,710 to 397,754; and 12,255 to binding titers.
604,246 for humans, NHPs, and mice, respectively Most human and NHP autologous neutralizers also
(Figure 1B and Supplementary Tables 2–4). Approximately neutralized a heterologous genotype 5a strain, SA13, in both
66% of humans and all animals displayed autologous serum HCVpp and HCV cell culture systems (Figure 2B). Of note, a
neutralizing activity for at least 1 time point in the study natural variation L442 was found in the E2 of SA13
(Supplementary Tables 2–4). There was no significant dif- (Supplementary Figure 2), which may render the virus more
ference for different antigen doses in immunized humans sensitive to antibody neutralization.28 Strikingly, the pro-
except the autologous neutralization at week 26 totypic genotype 1a strain H77, which shares 95% of E1E2
(Supplementary Figure 1) and the results were consistent amino acid identity with HCV-1 E1E2 (Supplementary
with a previous report.16 In animal models, the E1E2/ Figure 2), was resistant to neutralization by the majority
Adjuplex formulation induced apparently stronger antibody of human and NHP samples. In the mouse model, a weak
responses than the E1E2/AddaVax formulation (Figure 1C heterologous nAb response against H77 was also elicited,
and D). but no strong response against SA13 was observed for most
Longitudinal analysis of the human and NHP antibody animals (Figure 2B). Together, these data indicate that nAb
responses showed that most E1E2-specific antibodies responses to E1E2 immunization were very similar in NHPs
appeared after the second immunization and peaked after and humans, but were qualitatively different in mice.
the third immunization (Figure 1E and F). Forty-one percent
of immunized humans and 3 of 4 NHPs (30734, 31782, and Mapping of Antibody Specificity
31881) generated a prolonged autologous nAb response To understand the virus-neutralizing activities of anti-
persisting for more than 3 months (Figure 1E and F, right). body responses and the targeted epitopes, we performed
 1062 Chen et al Gastroenterology Vol. 158, No. 4
TRANSLATIONAL LIVER
BASIC AND
March 2020 Antibody Responses to HCV E1E2 Immunization 1063

Figure 2. Cross-reactivity of serum nAbs. Selected autologous neutralizing samples from human (week 52), NHP (week 27 or
30), and mouse (week 13) were tested for cross-binding and neutralization to a panel of HCV strains representing the major 6
genotypes at 1:50. (A) Binding of immune sera to native and denatured E1E2. The E1E2 amino-acid sequences of the HCV
isolates (represented by different color dots) are provided in Supplementary Figure 2 and compared here in the maximum-
likelihood phylogenic tree. The genotype of each isolate is shown in brackets. (B) Neutralization of HCVpp and HCV cell
culture by the immune sera.

pepscan and antibody competition analyses on selected map discontinuous epitopes. AR1 is proximal to the CD81
neutralizing and non-neutralizing samples from the post binding site. The AR1-specific mAbs only bind genotype 1
third or fourth immunization (at 1:50). At these time points, HCV and do not have significant neutralizing activity. AR2 is
most immunized subjects reached peak nAb activities for distal from the CD81 binding site and mAb AR2A can

TRANSLATIONAL LIVER
epitope mapping. Due to the limited availability of the hu- neutralize several HCV isolates. AR3, as well as the over-
man sera at week 50 and mouse sera at week 13, the human lapping AS434 and the nearby AS412, cluster on an E2

BASIC AND
week-52 and mouse week-9 sera were studied instead. In antigenic surface collectively known as the E2 neutralizing
pepscan, a library of overlapping peptides covering the HCV- face.12 Antibodies to AR3 (eg, AR3A) usually exhibit broadly
1 E1E2 sequence was used to map continuous epitopes. neutralizing activity against diverse HCV genotypes. AR4
Here, continuous epitopes and antigen sites (AS) are named and AR5 are present only on the E1E2 complex and are
based on their first amino acid position on the HCV poly- adjacent to each other. The mAbs AR4A and AR5A also
peptide, for example, E2 antigenic site 412-423 is called mediated cross-neutralization.
“AS412.” In antibody competition, soluble large external An overall similar set of immune epitopes were recog-
loop of CD81 and 5 well-characterized human mAbs tar- nized by humans and NHPs (Figure 3A). In both cases, the
geting distinct antigenic regions (AR)1–529–31 were used to most dominant continuous epitopes were mapped to the E2

=
Figure 1. Antibody responses elicited by E1E2 immunization in humans and animal models. Immune samples were tested for
endpoint titer of anti-E1E2 IgG and HCVpp neutralization against the autologous HCV-1 isolate at a dilution of 1:50. (A) Im-
munization and sampling schedules for humans, NHPs, and mice. (B) Comparison of peak antibody responses in humans,
NHPs, and mice. P values were calculated by 2-tailed Mann-Whitney test. (C, D) NHP (C) and mouse (D) antibody responses to
E1E2 formulated with AddaVax or Adjuplex adjuvant. (E, F) Kinetics of antibody response in humans (E) and NHPs (F). Positive
binding (green shading) was defined as binding titers 3 SDs above the mean of non-immune healthy human or NHP samples.
Positive neutralization (orange shading) was defined as >50% neutralization by immune samples at 1:50. Arrows indicate
immunization time points. Green and orange arrows indicate the fourth immunization on NHPs 31782 and 31859, and on NHPs
30734 and 31881, respectively. Human prolonged nAb responses, transient nAb responses, and non-nAb responses were
compared using 1-way analysis of variance tests. *P < .05, ***P < .0001. ns, not significant. (G) The proportion of human
responders immunized with different E1E2 dosage. (H) Mouse antibody responses following the third (week 9) and fourth (week
13) immunizations. (I) Correlation analysis (2-tailed Pearson correlation) between autologous binding and neutralization of
human (n ¼ 387), NHP (n ¼ 50), and mouse (n ¼ 56) immune sera. Placebo and pre-immunization human samples are
excluded in this analysis.
 1064 Chen et al Gastroenterology Vol. 158, No. 4
TRANSLATIONAL LIVER
BASIC AND

Figure 3. Epitope specificities of serum antibody responses. Selected neutralizing and non-neutralizing samples from human
(week 52, n ¼ 26), NHP (week 27 or 30, n ¼ 4), and mouse (week 9, n ¼ 12) were analyzed for specificity to continuous and
discontinuous epitopes at 1:50. (A) Left: mapping of continuous epitopes by pepscan analysis. Neutralizing activity against
HCVpp-HCV-1 of each sample is shown on the left. Right: mapping of discontinuous epitopes by antibody competition
analysis. (B–D) Comparison of specificities of neutralizing (orange) and non-neutralizing (gray) responses in human and NHP
samples. (B) Binding to HCV-1 E1E2 peptides. (C) Competition of binding to HCV-1 E1E2 by mAbs targeting AR1-5 and
soluble large external loop of CD81. P values were calculated by 2-tailed Mann-Whitney tests. *P < .05; **P < .001. (D)
Immunodominant epitopes targeted by neutralizers and non-neutralizers.
March 2020 Antibody Responses to HCV E1E2 Immunization 1065

Figure 4. Neutralizing epi-

TRANSLATIONAL LIVER
topes targeted by human,
NHP, and mouse immune

BASIC AND
sera. (A, B) Correlation
between neutralization and
antibody responses to
HVR1, AR3, and AR5. (A)
Human antibody
response. (B) Mouse anti-
body response. P values
were calculated using a 2-
tailed Pearson correlation.
(C) Inhibition of neutraliza-
tion against HCV-1 (geno-
type 1a) and SA13
(genotype 5a) by peptides
corresponding to known
continuous epitopes. See
Supplementary Table 5 for
further information of the
peptides.

HVR1 AS396 (100% responders for both humans and respectively. Although the polyclonal antibodies had less
NHPs), followed by the E1 N-terminus AS192 (88% human reactivity toward the discontinues epitopes on the AR1,
and 100% NHP responders) and AS210 (69% human and AR2, and AR4, most human and NHP neutralizers competed
75% NHP responders). Antibody responses to the with mAbs AR3A and AR5A, and blocked CD81 binding to
conserved AS412 and the E2 front layer AS434 were E1E2. Compared to non-neutralizers, the neutralizers
observed for 35% and 27% humans vs 25% and 75% NHPs, exhibited more frequent and stronger reactivities toward
 1066 Chen et al Gastroenterology Vol. 158, No. 4
TRANSLATIONAL LIVER
BASIC AND
March 2020 Antibody Responses to HCV E1E2 Immunization 1067

epitopes on AR3, AR5, the E2 variable region 2 (VR2) AS462 N-terminus and E2 HVR1, but they appeared to be mostly
and stalk region AS654 (Figure 3B–D), indicating that the non-neutralizing.
autologous nAbs likely targeted these regions rather than
the well-described immune decoy HVR1.32–34 Kinetics of Antigen-Specific B-Cell Responses to
In the mouse model, the immunodominant epitopes
E1E2 Immunization
were directed against the E1 N-terminus AS246, AS192,
Given that immune cells are not available from the
AS210, and AS204 (83%–100% responders), E2 HVR1
clinical trial, we investigated the antigen-specific B-cell
AS396 (83% responders), as well as the stalk region AS645
responses in the NHPs after the fourth immunization. With
and AS705 (100% responders) (Figure 3A, left). A few mice
enzyme-linked immunospot, we observed a rapid and
developed a weak response to AR5, but none competed
robust ASC response that peaked on day 4 and contracted
strongly with mAb AR3A (Figure 3A, right). Collectively,
rapidly on day 7 (Figure 5A). Seventy percent of these
E1E2 immunization elicited antibody responses primarily
cells were E1E2-specific, with IgG being the predominant
targeting the E1 N-terminus and E2 HVR1. Most human and
isotype, followed by IgA and IgM. Similar dynamics of
NHP neutralizers, but not mice, generated a detectable AR3-
plasmablasts (CD3–CD20–CD80þHLA-DRþ) were detected
specific antibody response.
using flow cytometric analysis (Figure 5B, left). The re-
sponses are in line with previous findings from NHP
Serum Neutralization Is More Related to AR3- studies35 and earlier than human ASC/plasmablast re-
and AR5- Instead of HVR1-Specific Antibody sponses to influenza, which peaked around 7 days post-
vaccination.36,37
Responses
A rapid expansion of antigen-specific IgGþ memory B
To map the neutralizing epitopes, we carried out cor-
cells (CD3–CD20þCD27þIgGþ E2þ) were also observed on
relation analysis between neutralization and specificity of
day 4 and accounted for approximately 10% of total IgGþ
antibody responses to HVR1, AR3, and AR5 in humans and
memory B cells in circulation. The frequency reached
mice. NHP samples were not analyzed here due to the small
approximately to 20% on day 10 post-boost (Figure 5B,
number of animals studied. A weak correlation between
right). These findings indicate the optimal time points for
autologous neutralization and HVR1-binding antibodies was
cloning antibodies from plasmablasts and memory B cells
observed in humans (r ¼ 0.39, P ¼ .048), but not in mice
from the immunized NHPs.
(r ¼ 0.26, P ¼ .4145) (Figure 4A and B). However, the
correlations between neutralization (both autologous and
heterologous) and antibody responses to AR3 or AR5 in Identification and Characterization of Monoclonal
humans were significantly higher (r  0.68, P  .0002, Antibodies From Non-Human Primates
Figure 4A). Ninety-three E1E2-reactive mAbs were isolated from
Next, we performed virus neutralization with the single plasmablasts and memory B cells from NHPs 30734

TRANSLATIONAL LIVER
neutralizing samples pre-complexed with a saturating con- and 31881. Of these, only 30% displayed autologous
centration of peptides (Supplementary Table 5) to investi- neutralizing activity in the preliminary screening. Using

BASIC AND
gate nAb specificity (Figure 4C). The antigenic sites studied purified IgG, we determined the specificity and function of
included the E1 N-terminus (AS192 and AS210), E2 HVR1 22 neutralizing mAbs. Most nAbs (59%) targeted AR3, while
(AS396), AS412, front layer (AS434), VR2 (AS462), and the others targeted AR5, HVR1, AR2, AR4, or AS434
CD81 binding loop (AS519). A few samples exhibited partial (Figure 5C–E). Heterologous neutralization against SA13
inhibition by peptide AS396, AS434, AS412, AS192, or (genotype 5a), H77 (genotype 1a), UKN4 11.1 (genotype
AS210 in HCVpp-HCV-1, and by AS434, AS462 or AS519 in 4a), and/or J6 (genotype 2a) were detected for all the nAbs
HCVpp-SA13. However, for most samples the presence of (except for those targeting HVR1), albeit at relatively low
antibody-blocking peptides had a negligible impact on potency (Figure 5D). Consistent with the plasma analysis
neutralizing activity. These data confirmed that, despite data above, a moderate correlation was observed between
being immunodominant, HVR1-specific antibodies elicited autologous or heterologous neutralization and antibody
by the recombinant E1E2 antigen were not the main factor response to AR3 (Figure 5F).
responsible for autologous neutralization. Taken together, To explore genetic features underlying the nAbs, we
immunization of humans and the 2 animal models with investigated the heavy-chain and light-chain germline gene
E1E2 elicited dominant antibody responses against the E1 usage, complementarity-determining region (CDR) 3 length

=
Figure 5. Characterization of NHP B-cell responses. (A, B) Kinetics of antigen-specific B responses in NHPs after E1E2 im-
munization. (A) Total and E1E2-specific ASC responses measured by enzyme-linked immunospot assay. The wells shown
were plated with 5  105 PBMCs. Percentage of E1E2-specific ASCs for each antibody isotype is shown in the bar charts.
(B) Flow cytometry analysis of plasmablast (CD3–CD20–CD80þHLA-DRþ) and E2-specific IgGþ memory B-cell responses
(CD3–CD20þCD27þIgGþ E2þ). Cells from red boxes were sorted for generation of mAbs. (C–F) Characterization of neutralizing
mAbs. (C) Specificity of mAbs targeting the continuous epitopes measured by pepscan. (D) Upper: specificity of mAbs to
discontinuous epitopes measured by antibody competition analysis. Lower: neutralization breadth and potency of mAbs using
HCVpp. (E) Distribution of nAb specificities. ND, not defined. (F) Correlation between autologous (HCV-1) or heterologous
(SA13) neutralization and antibody response to AR3 (2-tailed Pearson correlation).
 1068 Chen et al Gastroenterology Vol. 158, No. 4

Figure 6. Summary of NHP neutralizing mAbs. (A) Specificity, function, and genetic features of NHP neutralizing mAbs. Clonal
lineages were assigned based on the following criteria: 1) matching of V and J gene usage, 2) identical CDR3 length, and 3)
CDR3 nucleotide sequence homology >80%. NA, not available; ND, not defined. CDR3 length is based on Kabat numbering,
in which the CDRH3 is 2 amino acids shorter than in the ImMunoGeneTics information system definition. (B) Alignment of
amino acid sequences of human VH1-69 gene and its homologues in NHPs and mice. Human VH1-69 sequences were
TRANSLATIONAL LIVER

exemplified by VH1-69*01 and *06, the most frequently used alleles by human HCV AR3-specific antibodies, and VH1-69*10,
the closest human corresponding gene of to NHP VH1.36. The amino acid identity of each germline gene/allele compared to
BASIC AND

human VH1-69*01 allele is square-bracketed in red. FR, heavy-chain framework. (C) Comparison of the CDRH3 length and
somatic hypermutation (SHM) rate between cross-nAbs and strain-specific or limited-breadth nAbs. P values were calculated
by 2-tailed Mann-Whitney test. *P < .05. ns, not significant.

and somatic hypermutation rate of the antibodies. Like hu- gene, the NHP cross-nAbs tend to carry a longer CDRH3 loop
man AR3-specific bnAbs, which are predominantly derived but similar levels of somatic hypermutation compared to
from an immunoglobulin heavy-chain variable gene VH1-69 strain-specific nAbs or nAbs with limited breadth
(mainly VH1-69*01 and VH1-69*06 alleles),30,38–42 most (Figure 6C).
(90%) NHP AR3-specific cross-nAbs identified in this study
were encoded by VH1.36 (Figure 6A), a rhesus VH gene that
shares 90% amino acid identity and similar genetic features Discussion
(eg, the hydrophobic tip of heavy-chain complementarity- Overall, this comparative study provides important in-
determining region [CDRH] 2) with the human VH1-69 sights into antibody responses elicited by an E1E2-based
(Figure 6B). By contrast, the closest murine VH gene is VH1- subunit vaccine candidate. We demonstrated that all sub-
81*01, which differs from human VH1-69 by 33% amino acid jects developed an anti-E1E2 antibody response after im-
identity, and murine VH1-69 gene (VH1-69*01 and VH1- munization, but most antibodies induced are directed against
69*02) displays an even higher divergence (33%–34%, epitopes that have no or low antiviral function (eg, the E1 N-
Figure 6B). These data suggest that the NHP VH1.36 gene is terminus). Surprisingly, serum antibodies to HVR1 are mostly
highly homologous to human VH1-69 gene both genetically non-neutralizing (Figure 4), implying that, instead of forming
and functionally in response to HCV vaccination. Of the a flexible, exposed loop on the virus, the highly immunogenic
VH1.36 cross-neutralizing mAbs, RM2-01 and RM9-93 HVR1 appears to be mostly inaccessible to nAbs, by packing
belong to the same clonal lineage, based on the V and J tightly on the viral surface, adopting a conformation not
gene usage, CDR3 length, and CDR3 nucleotide sequence recognizable by most of the anti-HVR1 antibodies, or both.
identity (Figure 6A). In addition to the biased usage of VH HVR1 has recently been reported to work in concert with N-
March 2020 Antibody Responses to HCV E1E2 Immunization 1069

glycans on E2 for antibody escape.34,43 In contrast, neutral- conserved neutralizing epitopes. Several studies have
izing antibody responses to the known conserved antigenic shown that AR3-targeting antibodies are relevant in natural
sites, for example, AS412, AS434, and AR3, were relatively clearance and protection against re-infection in HCV-
rare (Figure 3), indicating the intrinstically low immunoge- infected individuals.40,41,50 However, as shown here the
nicity of these regions, which could explain, in part, the fail- subunit vaccine based on an authentic recombinant E1E2
ure of induction of sufficient bnAbs by this vaccine candidate. complex elicited only weak to moderate nAb response in a
The failure may also be caused by the presence of immuno- fraction of the subjects. A broadly effective vaccine would
dominant regions outcompeting conserved neutralizing epi- require the production of potent and broad nAbs, particu-
topes on the same antigen to elicit bnAb response.34,44 Future larly those targeting AR3. There are several technical chal-
HCV vaccine development should include rational antigen lenges for such endeavor. First, the HCV E2 neutralizing face
design to direct the antibody responses from non- containing AR3 appears to be conformationally flexible on
neutralizing immunodominant regions to conserved recombinant E2 antigens,51 potentially limiting the ability of
neutralizing epitopes. the antigens to properly present the neutralizing epitopes to
In humans and NHPs, E1E2 immunization elicited nAbs the immune system in vaccination. Second, as evidenced in
predominantly focused on AR3. At the polyclonal level, the this and other studies, not all binding antibodies to AR3 are
antibody responses are mostly against the autologous HCV- broadly neutralizing (Figures 2 and 6).41,52 This can be
1 (1a) and heterologous SA13 (5a) isolates. At the mono- caused partly by E2 flexibility and genetic diversity within
clonal level, several AR3-specific mAbs also exhibited this conserved region (Supplementary Figure 2).53–56 A
neutralization against a broad spectrum of heterologous better understanding of the subtle difference between nAb-
strains H77 (1a), UKN1B12.6 (1b), UKN4.11.1 (4a), and sensitive and nAb-resistant HCV isolates will be crucial in
even J6 (2a) and S52 (3a), albeit at relatively low potency. the development of a broad effective HCV vaccine. This
These findings implicate that cross-nAbs targeting AR3 were study provides novel information on the antibody responses
able to be elicited by E1E2 immunization, but their levels in of humans, NHPs and mice immunized with the same E1E2
blood were insufficient for broad neutralization. We did not vaccine candidate. In addition to a general description of the
observe a strong AR3-specific antibody response in mice, antigen-binding and virus-neutralizing activities of the
although non-nAbs to E2 front layer and CD81 binding loop antibody responses, we have extensively mapped the anti-
regions were isolated previously.20 Strikingly, of the isolated body specificities and identified similarities and differences
NHP mAbs, the majority of cross-nAbs were derived from in the antibody responses between the different in vivo
rhesus macaque VH1.36 gene, the closest macaque ortho- models to understand the virus-neutralizing activities. The
logue to the human VH1-69 gene (90% homology). Such a data offer a useful reference and baseline for studying nAb
preferential usage of a specific VH gene has previously been production and epitope specificity of future envelope
observed in antiviral antibody responses against HIV-1 and glycoprotein-based vaccine antigens, particularly the
influenza.42 These antibodies share germline-encoded ge- development of epitope-focused vaccine designed to elicit

TRANSLATIONAL LIVER
netic and structural features to recognize the same antigenic bnAbs targeting conserved E1E2 neutralizing epitopes. We
site, and often develop broadly neutralizing activity through are currently studying the genetic and structural properties

BASIC AND
a similar pathway. Understanding the developmental path- of the NHP neutralizing mAbs to understand how bnAbs are
ways and structural biology of these antibodies will faciliate generated. The results will inform whether human and NHP
the B-cell ontogeny vaccine strategy to elicit bnAbs.45 use similar structural features to neutralize HCV. Never-
Chimpanzee is the best animal model for studies of HCV theless, here we demonstrated that AR3-targeting anti-
infection and vaccine development because of its ability to bodies can be readily elicited in human and NHP
support persistent HCV infection and its close genetic rela- immunization studies.
tionship to humans.46,47 Immunization of chimpanzees with
E1E2 has previously been shown to elicit nAbs and protective
immunity in the animals.5,15,48 Intriguingly, chimpanzees
Supplementary Material
Note: To access the supplementary material accompanying
immunized with HCV-like particles consisting of the viral
this article, visit the online version of Gastroenterology at
core, E1 and E2 failed to generate antibodies to E1 and E2.7
www.gastrojournal.org, and at https://doi.org/10.1053/
This raises an important question regarding E1E2 immuno-
j.gastro.2019.11.282.
genicity when the antigens are presented on different plat-
forms. With the moratorium on chimpanzee research, use of
this animal model is no longer feasible.49 In this study, we References
demonstrate that the NHP rhesus macaque model produced 1. Li DK, Chung RT. Overview of direct-acting antiviral
nAbs sharing several important features with human nAbs, drugs and drug resistance of hepatitis C virus. Methods
and thus is potentially a useful preclinical model for studying Mol Biol 2019;1911:3–32.
HCV vaccine candidates designed to elicit nAbs. In addition, 2. Walker CM, Grakoui A. Hepatitis C virus: why do we
given the scarcity of human samples from the clinical trial, need a vaccine to prevent a curable persistent infection?
the NHP immune samples from this study may serve as a Curr Opin Immunol 2015;35:137–143.
substitute of E1E2-immune sera for future vaccine studies. 3. Freeman ZT, Cox AL. Lessons from nature: under-
A rational approach for HCV vaccine development is to standing immunity to HCV to guide vaccine design.
design immunogens that focus the antibody responses on PLoS Pathog 2016;12:e1005632.
 1070 Chen et al Gastroenterology Vol. 158, No. 4

4. Liang TJ. Current progress in development of hepatitis C 20. Ruwona TB, Giang E, Nieusma T, et al. Fine mapping of
virus vaccines. Nat Med 2013;19:869–878. murine antibody responses to immunization with a novel
5. Choo QL, Kuo G, Ralston R, et al. Vaccination of chim- soluble form of hepatitis C virus envelope glycoprotein
panzees against infection by the hepatitis C virus. Proc complex. J Virol 2014;88:10459–10471.
Natl Acad Sci U S A 1994;91:1294–1298. 21. Zhong J, Gastaminza P, Cheng G, et al. Robust hepa-
6. Elmowalid GA, Qiao M, Jeong SH, et al. Immunization titis C virus infection in vitro. Proc Natl Acad Sci U S A
with hepatitis C virus-like particles results in control of 2005;102:9294–9299.
hepatitis C virus infection in chimpanzees. Proc Natl 22. Smith K, Garman L, Wrammert J, et al. Rapid generation
Acad Sci U S A 2007;104:8427–8432. of fully human monoclonal antibodies specific to a
7. Folgori A, Capone S, Ruggeri L, et al. A T-cell HCV vaccinating antigen. Nat Protoc 2009;4:372–3784.
vaccine eliciting effective immunity against heterologous 23. Sundling C, Phad G, Douagi I, et al. Isolation of antibody
virus challenge in chimpanzees. Nat Med 2006;12:190– V(D)J sequences from single cell sorted rhesus macaque
197. B cells. J Immunol Methods 2012;386:85–93.
8. Park SH, Shin EC, Capone S, et al. Successful vacci- 24. Tiller T, Meffre E, Yurasov S, et al. Efficient generation of
nation induces multifunctional memory T-cell precursors monoclonal antibodies from single human B cells by
associated with early control of hepatitis C virus. single cell RT-PCR and expression vector cloning.
Gastroenterology 2012;143:1048–1060. J Immunol Methods 2008;329:112–124.
9. Youn JW, Hu YW, Tricoche N, et al. Evidence for pro- 25. Calabro S, Tritto E, Pezzotti A, et al. The adjuvant effect
tection against chronic hepatitis C virus infection in of MF59 is due to the oil-in-water emulsion formulation,
chimpanzees by immunization with replicating recombi- none of the individual components induce a comparable
nant vaccinia virus. J Virol 2008;82:10896–10905. adjuvant effect. Vaccine 2013;31:3363–3369.
10. Law JL, Chen C, Wong J, et al. A hepatitis C virus (HCV) 26. Sastry M, Zhang B, Chen M, et al. Adjuvants and the
vaccine comprising envelope glycoproteins gpE1/gpE2 vaccine response to the DS-Cav1-stabilized fusion
derived from a single isolate elicits broad cross- glycoprotein of respiratory syncytial virus. PLoS One
genotype neutralizing antibodies in humans. PLoS One 2017;12:e0186854.
2013;8:e59776. 27. Zhou T, Doria-Rose NA, Cheng C, et al. Quantification
11. Swadling L, Capone S, Antrobus RD, et al. A human of the impact of the HIV-1-glycan shield on antibody
vaccine strategy based on chimpanzee adenoviral and elicitation. Cell Rep 2017;19:719–732.
MVA vectors that primes, boosts, and sustains functional 28. Velazquez-Moctezuma R, Galli A, Law M, et al. Hepatitis
HCV-specific T cell memory. Sci Transl Med 2014; C virus-escape studies for human monoclonal antibody
6:261ra153. AR4A reveal isolate-specific resistance and a high barrier
12. Tzarum N, Wilson IA, Law M. The neutralizing face of to resistance. J Infect Dis 2019;219:68–79.
hepatitis C virus E2 envelope glycoprotein. Front 29. Law M, Maruyama T, Lewis J, et al. Broadly neutralizing
Immunol 2018;9:1315. antibodies protect against hepatitis C virus quasispecies
TRANSLATIONAL LIVER

13. Kinchen VJ, Cox AL, Bailey JR. Can broadly neutralizing challenge. Nat Med 2008;14:25–27.
BASIC AND

monoclonal antibodies lead to a hepatitis C virus vac- 30. Giang E, Dorner M, Prentoe JC, et al. Human broadly
cine? Trends Microbiol 2018;26:854–864. neutralizing antibodies to the envelope glycoprotein
14. Stamataki Z, Coates S, Evans MJ, et al. Hepatitis C virus complex of hepatitis C virus. Proc Natl Acad Sci U S A
envelope glycoprotein immunization of rodents elicits 2012;109:6205–6210.
cross-reactive neutralizing antibodies. Vaccine 2007; 31. Gopal R, Jackson K, Tzarum N, et al. Probing the an-
25:7773–7784. tigenicity of hepatitis C virus envelope glycoprotein
15. Meunier JC, Gottwein JM, Houghton M, et al. Vaccine- complex by high-throughput mutagenesis. PLoS Pathog
induced cross-genotype reactive neutralizing antibodies 2017;13:e1006735.
against hepatitis C virus. J Infect Dis 2011;204:1186–1190. 32. Prentoe J, Velazquez-Moctezuma R, Foung SK, et al.
16. Frey SE, Houghton M, Coates S, et al. Safety and Hypervariable region 1 shielding of hepatitis C virus is a
immunogenicity of HCV E1E2 vaccine adjuvanted with main contributor to genotypic differences in neutraliza-
MF59 administered to healthy adults. Vaccine 2010; tion sensitivity. Hepatology 2016;64:1881–1892.
28:6367–6373. 33. Velazquez-Moctezuma R, Law M, Bukh J, et al. Applying
17. Ray R, Meyer K, Banerjee A, et al. Characterization of antibody-sensitive hypervariable region 1-deleted hepa-
antibodies induced by vaccination with hepatitis C virus titis C virus to the study of escape pathways of
envelope glycoproteins. J Infect Dis 2010;202:862–866. neutralizing human monoclonal antibody AR5A. PLoS
18. Meyer K, Banerjee A, Frey SE, et al. A weak neutralizing Pathog 2017;13:e1006214.
antibody response to hepatitis C virus envelope glyco- 34. Prentoe J, Bukh J. Hypervariable region 1 in envelope
protein enhances virus infection. PLoS One 2011;6: protein 2 of hepatitis C virus: a linchpin in neutralizing
e23699. antibody evasion and viral entry. Front Immunol 2018;
19. Kachko A, Frey SE, Sirota L, et al. Antibodies to an 9:2146.
interfering epitope in hepatitis C virus E2 can mask 35. Silveira EL, Kasturi SP, Kovalenkov Y, et al. Vaccine-
vaccine-induced neutralizing activity. Hepatology 2015; induced plasmablast responses in rhesus macaques:
62:1670–1682. phenotypic characterization and a source for generating
March 2020 Antibody Responses to HCV E1E2 Immunization 1071

antigen-specific monoclonal antibodies. J Immunol 51. Kong L, Lee DE, Kadam RU, et al. Structural flexibility at a
Methods 2015;416:69–83. major conserved antibody target on hepatitis C virus E2
36. Wrammert J, Smith K, Miller J, et al. Rapid cloning of antigen. Proc Natl Acad Sci U S A 2016;113:12768–12773.
high-affinity human monoclonal antibodies against 52. Kinchen VJ, Zahid MN, Flyak AI, et al. Broadly neutral-
influenza virus. Nature 2008;453:667–671. izing antibody mediated clearance of human hepatitis C
37. Li GM, Chiu C, Wrammert J, et al. Pandemic H1N1 virus infection. Cell Host Microbe 2018;24:717–730.
influenza vaccine induces a recall response in humans 53. Tarr AW, Urbanowicz RA, Hamed MR, et al. Hepatitis C
that favors broadly cross-reactive memory B cells. Proc patient-derived glycoproteins exhibit marked differences
Natl Acad Sci U S A 2012;109:9047–9052. in susceptibility to serum neutralizing antibodies: genetic
38. Keck ZY, Saha A, Xia J, et al. Mapping a region of subtype defines antigenic but not neutralization sero-
hepatitis C virus E2 that is responsible for escape from type. J Virol 2011;85:4246–4257.
neutralizing antibodies and a core CD81-binding region 54. Urbanowicz RA, McClure CP, Brown RJ, et al. A diverse
that does not tolerate neutralization escape mutations. panel of hepatitis C virus glycoproteins for use in vaccine
J Virol 2011;85:10451–10463. research reveals extremes of monoclonal antibody
39. Keck ZY, Wang Y, Lau P, et al. Affinity maturation of a neutralization resistance. J Virol 2015;90:3288–3301.
broadly neutralizing human monoclonal antibody that 55. Wasilewski LN, Ray SC, Bailey JR. Hepatitis C virus
prevents acute hepatitis C virus infection in mice. Hep- resistance to broadly neutralizing antibodies measured
atology 2016;64:1922–1933. using replication-competent virus and pseudoparticles.
40. Merat SJ, Molenkamp R, Wagner K, et al. Hepatitis C J Gen Virol 2016;97:2883–2893.
virus broadly neutralizing monoclonal antibodies isolated 56. El-Diwany R, Cohen VJ, Mankowski MC, et al. Extra-epit-
25 years after spontaneous clearance. PLoS One 2016; opic hepatitis C virus polymorphisms confer resistance to
11:e0165047. broadly neutralizing antibodies by modulating binding to
41. Bailey JR, Flyak AI, Cohen VJ, et al. Broadly neutralizing scavenger receptor B1. PLoS Pathog 2017;13:e1006235.
antibodies with few somatic mutations and hepatitis C
virus clearance. JCI Insight 2017;2:e92872.
42. Chen F, Tzarum N, Wilson IA, et al. VH1-69 antiviral Author names in bold designate shared co-first authorship.
broadly neutralizing antibodies: genetics, structures, and Received August 12, 2019. Accepted November 12, 2019.
relevance to rational vaccine design. Curr Opin Virol
2019;34:149–159. Correspondence
Address correspondence to: Mansun Law, DPhil, Department of Immunology
43. Prentoe J, Velazquez-Moctezuma R, Augestad EH, et al. and Microbiology, The Scripps Research Institute, 10550 North Torrey Pines
Hypervariable region 1 and N-linked glycans of hepatitis Road, La Jolla, California 92037. e-mail: [email protected]; fax: (858) 784-
7842.
C regulate virion neutralization by modulating envelope
conformations. Proc Natl Acad Sci U S A 2019; Acknowledgments
116:10039–10047. The authors thank Rajen Koshy for help with clinical sample coordination;

TRANSLATIONAL LIVER
Philip Dormitzer, Timothy Pepini, and Novartis Vaccines and Diagnostics for
44. Farci P, Shimoda A, Coiana A, et al. The outcome of recombinant E1E2 antigen; Emily Carrow and Advanced Bioadjuvants for

BASIC AND
acute hepatitis C predicted by the evolution of the viral Adjuplex adjuvant; and Normal Blood Donor Service at The Scripps
Research Institute for procurement of normal blood samples. This is
quasispecies. Science 2000;288:339–344. manuscript 29802 from The Scripps Research Institute.
45. Kwong PD, Mascola JR. HIV-1 vaccines based on anti- Author contributions: Fang Chen, Kenna Nagy, and Mansun Law designed
body identification, B cell ontogeny, and epitope struc- and conceived the study. Kenna Nagy and Andrew Honda coordinated
mouse immunization, assessed antibody titers and neutralization. Deborah
ture. Immunity 2018;48:855–871. Chavez and Robert Lanford coordinated non-human primate experiments.
46. Bukh J. Animal models for the study of hepatitis C virus Fang Chen performed antibody competition, cross-binding enzyme-linked
immunosorbent assay, enzyme-linked immunospot, fluorescence-activated
infection and related liver disease. Gastroenterology cell sorting, monoclonal antibody cloning, and bioinformatics processing.
2012;142:1279–1287. Kenna Nagy, Shelby Willis, Ryan McBride, and Phillip Ordoukhanian
performed pepscan analysis. Kenna Nagy and Erick Giang performed virus
47. Burm R, Collignon L, Mesalam AA, et al. Animal models neutralization, antibody expression, and reagents preparation. Sharon Frey
to study hepatitis C virus infection. Front Immunol 2018; provided clinical samples. Jens Bukh provided hepatitis C virus cell culture
9:1032. reagents. Fang Chen and Kenna Nagy analyzed and interpreted data.
Mansun Law, Jiang Zhu, and Robert Lanford obtained funding support. Fang
48. Houghton M. Prospects for prophylactic and therapeutic Chen and Mansun Law wrote the manuscript. All authors read, edited, and
vaccines against the hepatitis C viruses. Immunol Rev approved the manuscript.
2011;239:99–108. Conflicts of interest
49. Kaiser J. Biomedical research. An end to US chimp The authors disclose no conflicts.
research. Science 2015;350:1013.
Funding
50. Merat SJ, Bru C, van de Berg D, et al. Cross-genotype This work was funded by National Institutes of Health grants AI079031,
AR3-specific neutralizing antibodies confer long-term Al106005, and AI123861. This investigation used resources that were
supported by the Southwest National Primate Research Center grant P51
protection in injecting drug users after HCV clearance. OD011133 from the Office of Research Infrastructure Programs, National
J Hepatol 2019;71:14–24. Institutes of Health.
1071.e1 Chen et al Gastroenterology Vol. 158, No. 4

Supplementary Methods Hepatitis C Virus Neutralization Assays

The methods for determining antibody neutralization of
HCV have been described recently.5,6 HCVpp were gener-
Enzyme-Linked Immunosorbent Assay ated by co-transfection of 293T cells with pNL4-3.lucR-E-
The methods for measuring anti-E1E2 antibody have and the corresponding expression plasmids encoding the
been described recently.1 In brief, Costar High Binding E1E2 from isolate HCV-1, H77, UKN1B12.6, J6, S52,
Half-Area 96-well plates (Corning, Corning, NY) were UKN4.11.1, or SA132,3,7 at a 4:1 ratio by polyethylenimine
coated overnight at 4 C with 5 mg/mL of Galanthus nivalis (Polysciences). Neutralization was carried out on Huh-7
lectin (Vector Laboratories, Burlingame, CA). After cells with diluted sera or mAbs as described previously.8
blocking with 4% nonfat milk (Bio-Rad, Hercules, CA) in Virus infectivity was detected with Bright-Glo luciferase
phosphate-buffered saline þ 0.05% Tween-20 (PBS-T), assay system (Promega), and percent neutralization was
plates were incubated with batch diluted cell lysates from calculated as the virus infectivity inhibited at the antibody
293T cells expressing E1E2 for isolates HCV-1 (genotype concentrations or serum/plasma dilutions indicated divided
1a), H77 (genotype 1a), UKN1B12.6 (genotype 1b), J6 by the infectivity without antibody after background sub-
(genotype 2a), S52 (genotype 3a), UKN4.11.1 (genotype traction. The background infectivity of the pseudotype virus
4a), SA13 (genotype 5a), or HK6a (genotype 6a) at room was defined by infecting cells with virus made only with
temperature for 1 hour. The E1E2 expression plasmids pNL4-3.lucR-E-. Pseudoparticles displaying the vesicular
have been described previously.2,3 Serial dilutions of stomatitis virus envelope glycoprotein G were used as
serum/plasma samples or mAbs were then added and control for nonspecific neutralizing activity. The virus was
incubated at room temperature for at least 1 hour. incubated with the antibodies for 1 hour at 37  C before
Horseradish peroxidase–conjugated goat anti-human, adding to Huh-7 cell monolayers and incubated for 5 hours
anti-monkey, or anti-mouse IgG Fc fragment antibody at 37  C. Expression of the luciferase reporter gene in the
(1:2,000; Jackson ImmunoResearch, West Grove, PA) was infected cells was measured with a luminometer on day 3
used for detection. End-point titers of anti-E1E2 IgG were post infection.
expressed as the reciprocal dilution that produced a For HCV cell culture neutralization, HCV isolates HCV-1,
signal 3-fold above background (293T cell lysate with no H77, and SA139,10 were propagated in Huh-7.5.1 cells as
envelope glycoprotein). described previously.11 Diluted serum/plasma samples or
For antibody competition analysis, G nivalis–captured mAbs were preincubated with HCV cell culture for 1 hour at
HCV-1 E1E2 was blocked with CD81 or mAbs AR1-5 (20 37 C and then added to cells plated at 6.5  103 per well
mg/mL) and incubated for 30 minutes at room temper- the night before. After 3 days of incubation, cell monolayers
ature, followed by addition of biotinylated mAb diluted were fixed and permeabilized with 100% methanol at
to 75% maximal binding level. Binding was detected –20 C for 15 minutes. Infectious foci were visualized by
using horseradish peroxidase–conjugated streptavidin indirect staining using mAb AR3A (5 mg/mL) and alkaline
(1:2,000; Thermo Fisher Scientific). Results are phosphatase–conjugated anti-human IgG Fc secondary
expressed as the percentage of inhibition without antibody. The plate was scanned and counted by Immuno-
blocking antibody. spot CTL counter and Image Acquisition 4.5 software
(Cellular Technology). Virus neutralization was determined
Pepscan Analysis by measurement of the percentage reduction of the number
Pepscan was performed as described previously, with of infectious foci.
modifications.4 Briefly, a library of peptides consisting of
15 amino acids in length and overlapped by 12 residues Enzyme-Linked Immunospot
spanning the full length of HCV-1 envelope glycoproteins E1E2-specific ASCs were detected by ELISPOT assay on
were custom synthesized in-house and printed in fresh PBMCs as described previously, with modifications.12
quadruplicate on N-hydroxysuccinimide ester–derived Multiscreen HTS IP filter plates (96-well) (MSIPS4510;
glass slides. After blocking with 5% fetal bovine serum, Millipore, Billerica, MA) were coated with 10 mg/mL of goat
PBS-T, serum/samples (1:50), or mAbs (10 mg/mL) were anti-monkey IgG, IgA, or IgM (H&L) antibody (Novus Bi-
added and incubated in humidified chamber for 1 hour at ologicals, Littleton, CO) or with 20 mg/mL of E1E2 protein
room temperature followed by 3 washes in PBS-T. The (Novartis Vaccines and Diagnostics) overnight at 4 C for
arrays were then incubated for 1 hour with Alexa Fluor enumeration of total or antigen-specific ASCs. Plates were
647–conjugated goat anti-human, anti-monkey, or anti- washed 4 times with PBS-T and 4 times with PBS, and
mouse secondary antibody, followed by washing and blocked with complete RPMI medium (supplemented with
another incubation with Alexa Fluor 647–conjugated 10% fetal bovine serum and penicillin/streptomycin) for 2
donkey anti-goat secondary antibody to amplify signal. hours in a 5% CO2 incubator at 37 C. PBMCs were diluted in
The processed slides were scanned using Innoscan 1100 complete RPMI medium, plated in serial 2-fold dilutions,
AL microarray scanner (Innopsys, Carbonne, France). The and incubated overnight in a 5% CO2 incubator at 37 C.
median feature and background pixel intensities for each Wells were again washed 4 times with PBS and 4 times with
antigen spot were determined by Mapix microarray PBS-T, followed by incubation with anti-monkey IgG-, IgA-,
analysis software (Innopsys). or IgM- horseradish peroxidase–conjugated antibodies
March 2020 Antibody Responses to HCV E1E2 Immunization 1071.e2

(1:1000; Novus) for 2 hours at room temperature. After 6. Prentoe J, Bukh J. In vitro neutralization assay using
washing, spots were developed with TMB substrate (Mab- cultured hepatitis C virus. Methods Mol Biol 2019;
tech, Stockholm, Sweden). To stop the reaction, wells were 1911:433–439.
washed with running water. Spots were documented and 7. Meunier JC, Engle RE, Faulk K, et al. Evidence for
counted using the Immunospot CTL counter and Image cross-genotype neutralization of hepatitis C virus
Acquisition 4.5 software (Cellular Technology). The results pseudo-particles and enhancement of infectivity by
were reported as the number of total or antigen-specific apolipoprotein C1. Proc Natl Acad Sci U S A 2005;
ASCs of each Ig isotype per million PBMCs. 102:4560–4565.
8. Ruwona TB, Giang E, Nieusma T, et al. Fine mapping of
murine antibody responses to immunization with a novel
soluble form of hepatitis C virus envelope glycoprotein
References complex. J Virol 2014;88:10459–10471.
1. Major M, Law M. Detection of antibodies to HCV E1E2 by 9. Li YP, Ramirez S, Mikkelsen L, et al. Efficient infectious
lectin-capture ELISA. Methods Mol Biol 2019;1911:421– cell culture systems of the hepatitis C virus (HCV)
432. prototype strains HCV-1 and H77. J Virol 2015;89:811–
2. Bartosch B, Dubuisson J, Cosset FL. Infectious hepatitis C 823.
virus pseudo-particles containing functional E1-E2 envelope 10. Jensen TB, Gottwein JM, Scheel TK, et al. Highly efficient
protein complexes. J Exp Med 2003;197:633–642. JFH1-based cell-culture system for hepatitis C virus
3. Lavillette D, Tarr AW, Voisset C, et al. Characterization of genotype 5a: failure of homologous neutralizing-antibody
host-range and cell entry properties of the major geno- treatment to control infection. J Infect Dis 2008;
types and subtypes of hepatitis C virus. Hepatology 198:1756–1765.
2005;41:265–274. 11. Zhong J, Gastaminza P, Cheng G, et al. Robust hepa-
4. Ruwona TB, McBride R, Chappel R, et al. Optimization titis C virus infection in vitro. Proc Natl Acad Sci U S A
of peptide arrays for studying antibodies to hepatitis C 2005;102:9294–9299.
virus continuous epitopes. J Immunol Methods 2014; 12. Smith K, Garman L, Wrammert J, et al. Rapid
402:35–42. generation of fully human monoclonal antibodies spe-
5. Bailey JR, Urbanowicz RA, Ball JK, et al. Standardized cific to a vaccinating antigen. Nat Protoc 2009;4:372–
method for the study of antibody neutralization of HCV 384.
pseudoparticles (HCVpp). Methods Mol Biol 2019;
1911:441–450. Author names in bold designate shared co-first authorship.
1071.e3 Chen et al Gastroenterology Vol. 158, No. 4

Supplementary Figure 1. Human antibody responses to different doses of E1E2 immunization. (A) Upper: serum anti-E1E2
IgG titer. Lower: HCVpp-HCV-1 neutralization. Green and orange shading indicate positive binding and positive neutraliza-
tion, respectively. One-way analysis of variance (ANOVA) test was used to compare the antibody responses by dose in
immunized humans. *P < .05 (1-way ANOVA test). (B) Distribution of prolonged nAb responses, transient nAb responses, and
non-nAb responses in human subjects immunized by dose.
March 2020 Antibody Responses to HCV E1E2 Immunization 1071.e4

Supplementary Figure 2. Alignment of E1E2 amino acid sequences from the HCV strains used in this study. The amino acid
identity of each HCV isolate compared to HCV-1 is square-bracketed in red. Yellow shading indicates the discontinuous
regions forming the E2 antigenic region AR3. HVR1, hypervariable region 1; TM, transmembrane domain; VR, variable region.
1071.e5 Chen et al Gastroenterology Vol. 158, No. 4

Supplementary Table 1.Non-Human Primate Plasmablast and Memory B-Cell Staining Panel

Antibody or probe Clone Fluorophore Source

CD3 SP34-2 PE-Cy7 BD Biosciences

CD20 2H7 APC-Cy7 BioLegend
CD80 L307.4 PE-Cy5 BD Biosciences
HLA-DR L234 BV421 Novus Biologicals
CD27 O323 PE BioLegend
IgG G18-145 FITC BD Biosciences
Live/Dead Aqua — — ThermoFisher Scientific
HCV-1 E2DTMa — Alexa Fluor 647 This study

APC, allophycocyanin; Cy5, cyanine 5; Cy7, cyanine 7; FITC, fluorescein isothiocyanate; PE, phycoerythrin.
a
E2DTM, full-length E2 without the transmembrane domain.

Supplementary Table 3.E1E2 Binding and Autologous Neutralization by Non-Human Primate Immune Plasma

End-point titer of anti-E1E2 IgGa % HCVpp-HCV-1 neutralizationb

Week 30,734 31,859 31,782 31,881 30,734 31,859 31,782 31,881

0 93 71 75 50 –83 –85 –54 –72

1 100 120 55 <50 –93 –93 –79 -66
4 2341 1999 728 2094 –48 –67 –56 –13
5 48,655 33,650 11,979 40,565 38 16 9 48
12 12,073 12,033 8,215 10,133 12 –23 –26 4
13 114,901 90,630 75,922 47,457 46 4 37 27
14 159,133 78,818 71,223 40,320 58 13 26 35
16 18,993 10,842 41,684 11,190 86 63 74 51
25 NA 8918 26,969 NA NA –53 –3 NA
26 NA 3430 7549 NA NA –50 4 NA
26 NA 66,924 378,416 NA NA –13 74 NA
27 NA 108,710 397,754 NA NA 37 87 NA
28 5983 NA NA 2215 44 NA NA 11
29 21,326 NA NA 7805 –25 NA NA –75
30 265,855 NA NA 132,777 67 NA NA 64
42 11,276 5560 40,878 6751 63 5 63 49
44 114,625 34,195 307,676 122,541 54 –10 78 69

NA, not available.

a
Green shading indicates positive binding was set as 3 SDs above the mean of control non-immune NHP plasma.
b
Red shading indicates positive neutralization defined as >50% of neutralization by immune plasma at a dilution of 1:50.
March 2020 Antibody Responses to HCV E1E2 Immunization 1071.e6

Supplementary Table 4.E1E2 Binding and Hepatitis C Virus Neutralization by Mouse Immune Sera

Week 9 Week 13

End-point % HCVpp neutralizationb End-point % HCVpp neutralization

titer of titer of
anti-E1E2 anti-E1E2
Animal ID IgGa HCV-1 (1a) H77 (1a) SA13 (5a) IgG HCV-1 (1a) H77 (1a) SA13 (5a)

221 33,497 75 15 –1 28,631 75 25 3

222 27,181 82 19 50 11,097 67 18 21
223 2812 30 6 –29 12,255 88 40 39
224 11,430 16 3 –157 12,367 73 6 5
225 18,220 –32 –4 –121 12,491 67 12 –39
226 25,352 76 –13 –13 12,990 85 –25 21
227 12,627 34 –19 –15 39,826 90 28 25
228 14,014 –43 15 –58 37,267 55 14 –20
229 49,585 –17 19 –5 48,106 52 30 43
230 9906 –32 8 –49 50,797 60 7 17
201 51,200 72 –28 –51 96,478 87 44 –4
202 47,256 18 –10 –18 183,338 67 25 15
203 49,423 37 –4 –18 105,633 76 12 –12
204 51,498 51 12 4 111,306 57 29 –39
205 11,657 19 –8 –45 93,742 79 34 30
206 42,383 73 51 17 169,353 81 70 –7
207 38,428 86 3 –61 37,262 54 34 –34
208 49,616 26 18 –24 104,628 94 42 71
209 47,310 –5 –1 –47 188,308 80 25 –35
210 30,833 44 –3 14 115,933 91 48 –7
317 30,545 89 76 –68 177,575 96 83 –41
318 18,467 59 55 –215 140,293 76 25 –48
319 38,664 45 54 –173 168,199 95 28 –69
320 15,979 68 49 –80 50,317 86 43 –79
321 116,315 27 25 –204 604,246 70 56 –55
322 49,309 54 39 –137 165,770 34 56 –67
323 25,229 23 21 –124 15,005 51 29 –56
324 76,116 37 43 –86 156,465 92 39 –53

a
Green shading indicates positive binding was set as 3 SDs above the mean of naïve mouse sera.
b
Red shading indicates positive neutralization defined as >50% of neutralization by immune sera at a dilution of 1:50.

Supplementary Table 5.Peptides Used in Neutralization

Inhibition Assaysa

Peptide Residues Region Sequence

AS192 192-206 E1 N-terminus YQVRNSTGLYHVTND

AS210 210-224 E1 N-terminus SSIVYEAADAILHTP
AS396 396-410 E2 HVR1 VSGFVSLLAPGAKQN
AS412 411-425 E2 AS412 VQLINTNGSWHLNST
AS434 435-449 E2 AS434 TGWLAGLFYHHKFNS
AS462 462-476 E2 VR2 LTDFDQGWGPISYAN
AS519 519-533 E2 CD81 binding loop TDRSGAPTYSWGEND

a
Reference table for Figure 4C.