GASTROENTEROLOGY 2007;132:667– 678
Hepatitis C Virus Continuously Escapes From Neutralizing Antibody and
T-Cell Responses During Chronic Infection In Vivo
THOMAS VON HAHN,* JOO CHUN YOON,‡ HARVEY ALTER,§ CHARLES M. RICE,* BARBARA REHERMANN,‡
PETER BALFE,储,¶ and JANE A. MCKEATING*,¶
*Center for the Study of Hepatitis C, The Rockefeller University, New York, New York; ‡Immunology Section, Liver Diseases Branch, and the §Department of
Transfusion Medicine, Warren Grant Magnuson Clinical Center, National Institutes of Health, Bethesda, Maryland; 储Division of Infectious Diseases, Columbia
University, New York, New York; and the ¶Division of Immunity and Infection, Institute of Biomedical Research, Medical School, University of Birmingham,
Birmingham, United Kingdom
Background & Aims: Broadly reactive neutralizing
antibodies (nAbs) and multispecific T-cell responses are generated during chronic hepatitis C
virus (HCV) infection and yet fail to clear the virus.
This study investigated the development of autologous nAb and HCV-glycoprotein–specific T-cell responses and their effects on viral sequence evolution during chronic infection in order to
understand the reasons for their lack of effectiveness. Methods: Numerous E1E2 sequences were
amplified and sequenced from serum samples collected over a 26-year period from patient H, a
uniquely well-characterized, chronically infected
individual. HCV pseudoparticles (HCVpp) expressing the patient-derived glycoproteins were generated and tested for their sensitivity to neutralization by autologous and heterologous serum
antibodies. Results: A strain-specific nAb response
developed early in infection (8 weeks postinfection), whereas cross-reactive antibodies able to neutralize HCVpp-bearing heterologous glycoproteins
developed late in infection (>33 wk postinfection).
The humoral response continuously failed to neutralize viruses bearing autologous glycoprotein sequences that were present in the serum at a given
time. The amplified glycoprotein sequences displayed high variability, particularly in regions corresponding to defined linear B-cell epitopes. Mutations in defined neutralizing epitopes were
associated with a loss of recognition by monoclonal
antibodies against these epitopes and with decreased neutralization of corresponding HCVpp.
Viral escape from CD4 and CD8 T-cell responses
also was shown for several novel epitopes throughout the glycoprotein region. Conclusions: During
chronic infection HCV is subjected to selection
pressures from both humoral and cellular immunity, resulting in the continuous generation of escape variants.
H
epatitis C virus (HCV) is an important human
pathogen infecting about 170 million people
worldwide. In the United States, it is the single most
common cause of chronic liver disease requiring liver
transplantation.1,2 Cellular and humoral responses are
generated during acute infection, however, they are insufficient to achieve viral clearance in the majority of
individuals, with approximately 60%– 80% of new infections becoming persistent.
In vivo, HCV replicates to high levels using an errorprone viral RNA polymerase, which leads to a spectrum
of related but distinct sequences within infected individuals, often referred to as a quasispecies.3 The immune
system is thought to exert unequal selective pressure on
variants within the circulation, favoring the rapid emergence of escape mutants.4 Escape from CD8 T-cell responses by mutation is well documented and an important predictor of progression to chronic HCV
infection.5– 8 In keeping with this, a vigorous and broad
CD8 T-cell response during the acute phase of infection
is associated with viral clearance.9 –11 Studies of acute
HCV infection in chimpanzees previously exposed to the
virus provide compelling evidence that protective CD8
T-cell–mediated immunity exists.12 CD4 T cells also are
required for control of HCV on re-exposure, but their role
is less well defined.13
Even less is known about the impact of the humoral
immune response on HCV pathobiology. Without the
ability to culture HCV, there was, until recently, no simple in vitro method to evaluate viral escape from the
antibody-mediated immune response. The development
of HCV glycoprotein-bearing retroviral pseudoparticles
(HCVpp) has made it possible to assess antibody-dependent neutralization of HCV entry.14 –18 Neutralizing anAbbreviations used in this paper: EIA, enzyme-linked immunosorbent assay; gp, glycoprotein; HCVpp, HCV pseudoparticle; HVR, hypervariable region; MAb, monoclonal antibody; nAb, neutralizing antibody;
PBMC, peripheral blood mononuclear cell; PCR, polymerase chain
reaction; sE2, soluble HCV E2.
© 2007 by the AGA Institute
0016-5085/07/$32.00
doi:10.1053/j.gastro.2006.12.008
BASIC–LIVER,
PANCREAS, AND
BILIARY TRACT
See editorial on page 801.
668
VON HAHN ET AL
BASIC–LIVER,
PANCREAS, AND
BILIARY TRACT
tibody (nAb) responses often provide the first-line adaptive defense against infection by limiting virus spread.
Serum antibodies from chronically HCV-infected individuals show broadly reactive neutralizing properties and yet
fail to clear viral infection.15,17,19 The reasons for their
lack of effect are understood poorly, however, one possible explanation is that Ab response(s) are less able to
neutralize autologous glycoprotein species circulating
within an individual at the time of sampling.
In setting out to address these questions we were
fortunate to have access to sequential serum samples
from patient H, an individual who was infected with
HCV in 1977 and has been meticulously followed-up
since then. Moreover, patient H was the source of the
prototype HCV strains H and H77, and thus a wide range
of tailor-made reagents are available for virologic and
immunologic analyses.20 –22 For these reasons this patient
is a unique subject for studies into the immunologic
history of chronic HCV infection.
To assess the impact of the HCV quasispecies on the
nAb response we generated HCVpp-bearing glycoprotein
variants cloned from sequential samples from chronically
infected patient H. By using a series of samples obtained
between 3 weeks postinfection and throughout the 26
years thereafter, we sought to investigate the process of
antigenic escape in the viral glycoproteins from humoral
and cellular immune surveillance. We found compelling
evidence for repeated mutational change resulting in loss
of recognition of the HCV glycoprotein by the cognate
antibody response and escape from antibody-mediated
neutralization. Similarly, mapping of T-cell responses to
the E1E2 region identified 4 novel T-cell epitopes in
which mutations occurred, leading to escape from CD4
and CD8 T-cell recognition.
Materials and Methods
Antibodies, Cells, and Reagents
Both 293T and Hep3B cells were propagated in
Dulbecco’s modified Eagle medium with 10% fetal bovine
serum. Antibodies against HCV E2 have been described
previously.14,23 To generate soluble CD81 an expression
plasmid encoding the CD81 large extracellular loop fused
to Glutathione S-transferase (GST) was transformed into
Rosetta-gami Escherichia coli (Novagen, La Jolla, CA). Fusion proteins were prepared by lysis with an Avestin
(Avestin, Ottawa, Canada) air emulsifier (3 passages at
15,000 psi) and subsequent centrifugation (25,000 ⫻ g
for 30 min at 4°C). Cleared lysates were purified over a
GSTrap FF affinity column according to the manufacturer’s instructions (Amersham Biosciences, Piscataway, NJ).
HCV Antigens
For the initial screening of T-cell responses
against the HCV E1 and E2 proteins, 15-mer peptides
(total, 111) (Mimotopes, Clayton, Australia), overlapping
GASTROENTEROLOGY Vol. 132, No. 2
by 10 amino acids and covering the complete HCV H77
(genotype 1a) E1E2 sequence, were resuspended at 20
mg/mL in dimethyl sulfoxide and further diluted with
phosphate-buffered saline (PBS) solution to obtain a
final concentration of 1 g/mL. For direct comparison of
responses against the mutant epitopes, the corresponding mutant peptides were synthesized at Rockefeller University (New York, NY) to more than 95% purity.
Polymerase Chain Reaction Amplification,
Cloning, and Sequence Analysis of HCV E1E2
The plasmid encoding H77 E1E2 has been described previously.14,21,24 Total RNA was prepared from
HCV-infected plasma using commercial kits (Qiagen, Valencia, CA).25 Briefly, complementary DNA (cDNA) was
synthesized in a reaction volume of 20 L, containing
2–5 L of template RNA, 2.5 U of Multiscribe MMuLV
reverse transcriptase with 400 mol/L each of the 4
deoxynucleoside triphosphates and 200 nmol/L of the
antisense primer p7-2710 (AGC AGG AGG AGN GGC
CAY ATC CCR TAG A, Y ⫽ C/T mixture, R ⫽ A/G, N ⫽
A/G/C/T) in the manufacturer’s recommended buffer
(N808-0234; ABI, Foster City, CA) for 2 hours at 42°C.
This cDNA was used as the template for polymerase
chain reaction (PCR) amplification of the E1E2 region as
previously described.24 Briefly, a 50-L PCR was set up
containing 2.5 L of cDNA, 2.5 U of the proofreading
Expand polymerase mixture (1 681 834; Roche, Mannheim, Germany) in 1⫻ Expand buffer 3 (2.25 mmol/L
Mg2⫹), with 400 mol/L each of the 4 deoxynucleoside
triphosphates and 200 nmol/L each of the primers
core⫹813 (GAG GAC GGY RTR AAY TAY GCA ACA GG;
sense) and p7-2710. The PCR consisted of 30 cycles at
92°C for 45 seconds, 45°C for 45 seconds, and 68°C for
300 seconds, and was performed in an Eppendorf (Westbury, NY) thermal cycler. A total of 2 L of the completed
reaction was used as template for a second amplification,
containing the same reaction components as described
previously with 200 nmol/L of the primers core⫹843
(CACC ATG GGT TGC TCT TTC TCT ATC TT; sense)
and E2-2580H (CTA CTA CGC CTC CGC TTG GGA
TAT GAG TAA CAT CAT CCA, antisense). This second
round of PCR comprised 25 cycles at 92°C for 35 seconds, 55°C for 35 seconds, and 68°C for 150 seconds. In
those cases in which the input RNA was more than 2000
viral copies, full-length E1E2 was amplified readily. PCR
products were cloned into pcDNA3.1D-TOPO (K490001; Invitrogen, Carlsbad, CA) and the sense and antisense
strands were sequenced (Big Dye 3.1 Terminator Kit;
ABI). All sequences were deposited with Genbank and
have the accession numbers DQ897773–DQ897818. Several sequences for the E1E2 region of HCV within this
patient have been deposited in Genbank previously. For
comparison with the new sequences described here we
included the following sequences from 1977 in our analyses: H77C (AF01175120), HPCST77 (M6238122),
H77IMC (AF00960621), H21 (AF01175320), H
(M6746326), and H11 (AF01175220). An additional clone,
H90, obtained in 1990, was available and was included in
the analysis (M6238222). In addition to the sequences
recorded in Genbank, several sequences for the E2 hypervariable region (HVR) were reported by Ogata et al,22 and
several matched the HVR sequences of clones obtained in
this study (data not shown).
Phylogenetic Analyses
The nucleotide sequences were aligned and translated using the SeAL2.0 program (A. Rambaut, Oxford
University, available at: http://evolve.zoo.ox.ac.uk). Synonymous and nonsynonymous distances were estimated
using the Nei and Gojobori27 method implemented in
the PAML3.14 program.28 Phylogenetic analyses were
performed using the PAUP 4.0 software package29 using
a modified HKY85 evolutionary model for the data, selected by hierarchic likelihood testing (program Modeltest 3.630), the transition/transversion ratio (7.04), proportion of invariable sites (.46), and gamma distribution
shape parameter (␣) for variable sites (.86) were estimated
by maximum likelihood methods; distance estimates
were averaged within and between groups using Excel
2001 (Microsoft, Redmond, WA).
Pseudoparticle Production, Infection, and
Neutralization Assays
Pseudoparticles were generated by transfection of
293T cells with pNL4-3.Luc.R⫺E⫺ plasmid containing
the env-defective proviral genome and an expression plasmid encoding the HCV glycoproteins (gps), as previously
described.14 The virus containing extracellular media was
collected 48 and 72 hours after transfection. Heat-inactivated sera, monoclonal antibodies (MAbs), or soluble
CD81 were incubated with virus at their appropriate
dilution in 3% fetal bovine serum/Dulbecco’s modified
Eagle medium plus 4 g/mL polybrene at 37°C for 1
hour. The virus-Ab mixture was transferred to Hep3B
cells seeded in 96-well plates (8 ⫻ 103 cells/well) and
infections were centrifuged at 400g for 1 hour, incubated
at 37°C for 6 hours, and unbound virus was removed and
incubated for a total of 72 hours. Cells were lysed with
cell lysis buffer (Promega, Madison, WI) and tested for
luciferase activity as previously described.14 The percentage neutralization was determined by comparing pseudoparticle infectivity (luciferase relative light units) in
the presence of a test serum or MAb with infection in the
presence of a control HCV-negative serum or an irrelevant isotype-matched immunoglobulin (Ig)G at the same
dilution.
Quantitative Enzyme-Linked Immunosorbent
Assay
GNA lectin (Sigma, St. Louis, MO) was used to
coat Immulon II enzyme-linked immunosorbent assays
HCV ESCAPES NEUTRALIZING ANTIBODIES AND CTL
669
(EIA) plates (Nunc, Rochester, NY) at 1 g/mL for 4
hours at 37°C. After washing with PBS, the plates were
blocked with 5% bovine serum albumin/PBS and lysates
of cells expressing HCV E1E2 or pelleted virus allowed to
bind overnight at 4°C. A preparation of truncated E2661
was used as an internal calibrator in all EIAs and allowed
comparison of data between different assays. Bound antigen was visualized with MAbs specific for E2 or pooled
HCV-positive human sera, an antispecies IgG– horseradish-peroxidase conjugate (Jackson, West Grove, PA) and
tetramethylbenzidene (BioFX Laboratories, Owings
Mills, MD). Absorbance values were measured at 450 nm
(fusion plate reader; Perkin Elmer, Boston, MA).
Isolation of CD4 and CD8 T Cells
Peripheral blood mononuclear cells (PBMCs) were
isolated by density gradient centrifugation as previously
described.31 For selected experiments, CD8 T cells were
isolated from PBMCs using CD8 microbeads and the
autoMACS separator (Miltenyi, Bergisch-Gladbach, Germany). Subsequently, CD4 T cells were isolated from the
CD8-negative population using CD4 microbeads. The
purity of the CD8 T-cell population was 95% and
the purity of the CD4 T-cell population was 92%, as
analyzed by flow cytometry. The negatively selected
CD4⫺CD8⫺ cells were irradiated (3000 rad) and 105 cells
per well were used as feeder cells in enzyme-linked immunospot assays.
Enzyme-Linked Immunospot Assay
Interferon-␥ enzyme-linked immunospot assays
were performed as described32 using duplicate cultures of
either the indicated number of CD4 and CD8 T cells and
105 irradiated (3,000 rad) CD4⫺CD8⫺ cells, or 3 ⫻ 105
CD25-depleted PBMCs. Cells were stimulated with either
1 g/mL of the individual HCV E1E2 peptides, 10 g/mL
of either wild-type or mutant epitope peptides, 1 g/mL
phytohemagglutinin (Murex Biotech Limited, Dartford,
England), or dimethyl sulfoxide control, respectively. The
number of spot-forming cells was determined with a KS
enzyme-linked immunospot Reader (Zeiss, Thornwood,
NY). Numbers of antigen-specific spot-forming cells (in
the presence of antigen minus spot-forming cells in dimethyl sulfoxide controls) are shown.
Results
Clinical Information
Patient H, the subject of this study, was infected
with HCV through a blood transfusion in 1977 while
undergoing cardiac surgery. After an initial high alanine
aminotransferase peak (2112 IU/L), indicative of severe
acute hepatitis, he went on to develop mild chronic
hepatitis with persistently low alanine aminotransferase
levels and HCV RNA levels in the 104 and 106 genome
copies/mL range over the following 26 years (Figure 1A).
BASIC–LIVER,
PANCREAS, AND
BILIARY TRACT
February 2007
670
VON HAHN ET AL
GASTROENTEROLOGY Vol. 132, No. 2
Development of Antiglycoprotein Responses
The appearance of the early strain-specific nAb
response was associated with the detection of antibodies
specific for H77 E1E2 and the HVR in an EIA assay
(Figure 1B). However, high levels of antibodies capable of
binding H77-soluble E2 (sE2) or J6 E1E2 were first detected at 111 weeks, coincident with the appearance of
cross-reactive nAbs and an increase in the titer of H77specific nAbs (Figure 1A and B). In contrast to the other
steadily increasing anti-gp responses, the Ig response
specific for the H77 HVR peaked at 9 weeks after infection and decreased to undetectable levels by 33 weeks
(Figure 1B), suggesting a dynamic and rapidly evolving
immune response to the viral gps in the early phase of
infection. As previously reported by others,34 the anti-gp
response was restricted to the IgG1 isotype (data not
shown).
The Early Neutralizing Response Targets
the HVR
BASIC–LIVER,
PANCREAS, AND
BILIARY TRACT
Figure 1. Evolving anti-HCV E1E2 response in patient H over time. (A)
Lines show serum ALT (- - -) and viral load (〫 and grey line) over 26
years of chronic infection. Bars represent the serum dilution capable of
neutralizing HCVpp-H77 (□) or HCVpp -J6 () infectivity by 90% (ID90).
Arrows indicate samples from which E1E2 sequences were amplified
and included in the phylogenetic analysis. (B) The reactivity of sequential
serum samples with antigens representing HCV E1E2 sequences by
EIA. Antigens tested were as follows: 〫, H77 HVR peptide; ⽧, H77
E1E2; ⌬, J6 E1E2; Œ, H77 sE2. (C) HCVpp H77 infection of Hep3B cells
in the presence of neutralizing E2 antibodies 9/27 and 11/20 (at
1 g/mL) or H serum (week 9 at 1:300; week 244 at 1:6000, and week
759 at 1:10,000 dilution) with or without H77 sE2 (5 mol/L) or HVR
peptide (5 mol/L; mean of n ⫽ 3 ⫾ SD). MOCK, mock infected cells; NO
ENV, cells infected with an envelope-deficient pseudoparticle; RLU, relative light units.
Patient H underwent liver biopsy procedures on 4 occasions, each showing minimal inflammation with no fibrosis and no discernable progression over time. As expected, liver function was well preserved. Thus, patient H
is representative of many chronically HCV-infected patients with mild, nonprogressive disease.33
Development of Strain-Specific and
Cross-Reactive nAb Responses
Serial serum samples from patient H covering
26 years of chronic HCV infection were used for this
study. All samples were characterized for their ability
to neutralize HCVpp-bearing autologous H77 gps (a
sequence cloned at 3 weeks after infection) and heterologous gps of the closely related genotype 1b (Con1,
OH8) and the distant 2a (JFH, J6). In keeping with our
previous report,17 HCVpp-H77–specific nAbs were observed at seroconversion (8 weeks after infection) (Figure 1A and data not shown). Neutralization of HCVppbearing heterologous gps was first detected at 111
weeks after infection, at a time when chronic infection
had been established.
To characterize epitopes targeted by the polyclonal nAb response, sE2 and peptide pools covering the
entire H77 E1E2 region were screened for their ability to
compete with acute-phase serum neutralization of
HCVpp-H77 (data not shown). The only peptide able to
inhibit serum neutralization represented the HVR. As
controls, increasing concentrations of H77 sE2 and HVR
peptide did not affect HCVpp-H77 infectivity but specifically reduced the neutralizing activity of MAbs targeting
the HVR (MAb 9/27) or a non-HVR E2 epitope (MAb
11/20) (Figure 1C). sE2 and HVR peptide prevented the
neutralization of HCVpp-H77 by serum antibodies collected after 9 weeks, but not after 244 or 759 weeks,
consistent with the early strain-specific neutralizing response targeting predominantly the HVR and later responses evolving to a breadth and/or titer that could not
be blocked by H77 HVR or sE2.
E1E2 Diversity and Generation of Autologous
HCVpp
Patient H developed an nAb response to HCVppH77 that increased in breadth and titer over time, acquiring the ability to neutralize distantly related HCV genotypes. Yet, paradoxically, this failed to control HCV
replication, as evidenced by fluctuating yet robust HCV
RNA levels throughout the course of infection. One possible explanation is that Ab neutralization of heterologous isolates does not reflect the response against the
virus population present within the serum at the same
time point. To address this issue we cloned the E1E2
region from plasma samples collected over a 26-year
period. Although no reverse-transcription PCR products
were obtained from serum samples between 1977 and 1991,
we were able to generate and sequence multiple clones
originating from 1977, 1991, 1992, 1993, 1994, 1995, 2002,
and 2003. These were used to determine the consensus
February 2007
HCV ESCAPES NEUTRALIZING ANTIBODIES AND CTL
671
sequence at the time of sampling and the quasispecies
diversity. Four E1E2 clones included stop codons and were
not studied further. Clones subsequently are referred to as
Haa.bb, where aa indicates the year of the serum sample and
bb indicates the clone number. Phylogenetic analysis of the
sequences identified 3 major groups (Figure 2A). The diver-
sity within a single sample from 1977 was 1.1% based on
the entire 1734-bp sequence. This was comparable with
that seen between 1991 and 1995 (1.3%) and 2002 and
2003 (1.5%). When only phylogenetically informative
sites (ie, those with changes occurring in at least 2 independent clones) were analyzed, the intrasample diversity
was 12.5%, 10.4%, and 13.3% for 1977, 1991–1995, and
2002–2003, respectively.
To assess the autologous nAb response in patient H, all
E1E2 clones that were free of stop codons were coexpressed individually with pNL4-3.Luc.R⫺E⫺ in 293T
cells to generate HCVpp. All clones expressed E1E2 at
comparable levels (data not shown), however, only 9 of 42
produced infectious HCVpp (Figure 2B and data not
shown). Comparison of sequences generating infectious
HCVpp with their nearest neighbors revealed that a single amino acid change often could distinguish between
viable and nonviable sequences (eg, S234L in H77 vs
H77.16). Often, the nonviable clones possessed unique,
noninformative changes that were atypical of the group.
Figure 2. Generation of HCVpp-bearing glycoprotein clones from H
serum samples. (A) The phylogenetic tree displays the relationships
between E1E2 sequences amplified from H serum samples. The 2
digits after H define the year of the serum sample from which the clone
was obtained. *Clones that gave rise to infectious HCVpp. (B) Infectivity of
functional HCVpp generated using sequences amplified from H serum
samples. This constitutes the set of HCVpp that was used for subsequent
analyses (mean of 3 experiments ⫾ SD).
Development of Autologous nAb Responses
All gp sequences cloned from patient H in 1977,
1991, 1992, 1995, and 2002 that gave rise to functional
HCVpp and the H77 sequence were tested for their sensitivity to neutralization by autologous serum antibodies.
As previously noted, the serum samples collected before
BASIC–LIVER,
PANCREAS, AND
BILIARY TRACT
Figure 3. Loss of antigen recognition by monoclonal antibodies raised
against the H77 sequence over time. Lysates from 293T-cells expressing the H-derived E1E2 clones were immobilized on GNA-coated EIA
plates and the ability of 4 neutralizing anti-E2 MAbs raised against the
H77 sequence (3/11, 7/16, 9/27, 11/20) to bind the respective glycoproteins was tested. To control for variation in antigen expression level,
optical density readings were normalized relative to the signal obtained
with a nonneutralizing antibody, 9/75, that recognizes an epitope that
was fully conserved among all H isolates. The value obtained with H77
E1E2 as the capture antigen was normalized to 1.
672
VON HAHN ET AL
GASTROENTEROLOGY Vol. 132, No. 2
suggest that the nAb response fails to efficiently neutralize gp sequences that are dominant in vivo at a given time
point. Interestingly, all of the patient H-derived HCVpp
were neutralized by serum antibodies from 2 unrelated,
chronically HCV-infected individuals, confirming that
autologous glycoprotein sequences are specifically resistant to antibodies generated by their own host, but not
unrelated hosts.
Sequence Polymorphism Associates With Sites
of Immune Recognition
Figure 4. Identification of T-cell epitopes in the E1E2. (A) Patient H’s
PBMCs were stimulated with 15-mer peptides corresponding to the
H77 E1E2 sequence. The number of peptide-specific interferon-␥–producing cells was determined by enzyme-linked immunospot analysis.
(B) The 3 most vigorously recognized peptides indicated in A were
retested on purified CD4 and CD8 cells isolated from patient H.
BASIC–LIVER,
PANCREAS, AND
BILIARY TRACT
seroconversion in 1977 failed to neutralize any of the
viruses tested, whereas subsequent samples showed hightiter neutralization of HCVpp-bearing H77.19, H77.20,
and H77 gps (Table 1). In contrast, serum antibodies in
1991 neutralized HCVpp bearing some autologous gps
from that same sampling time, but with variable efficiency, with some clones being resistant to neutralization
(H91.A11 and H91.B1). Similarly, serum antibodies in
1992, 1995, and 2002 showed reduced neutralization of
HCVpp-bearing concurrent and later gps, although generally showing high-titer neutralization of HCVpp-bearing gps cloned from earlier time points in infection. The
last clone, obtained in 2002 (H02.E10), was resistant to
neutralization by all of the serum samples. These data
An alignment of patient H-derived E1E2 sequences
with a map of known epitopes taken from the HCV
Immunology Database (hcv.lanl.gov and Yusim et al35)
revealed that of 52 residues showing phylogenetically
informative changes, 30 were located within defined Bcell epitopes (Table 2). To assess the impact of sequence
polymorphism on antibody neutralization we studied the
sensitivity of HCVpp bearing the set of patient H-derived
gps to neutralization by MAbs that were raised against a
recombinant form of H77 E2 and have been reported to
neutralize HCVpp-H77.14,23 HCVpp-bearing gps cloned
from 1977 and 1991 were neutralized equally by all of the
MAbs. However, with the exception of MAb 11/20,
viruses bearing gps from later time points displayed increasing resistance to neutralization by all MAbs
(Table 3). Although all viruses were neutralized by a
soluble form of the large extracellular loop of CD81,
HCVpp-bearing gps cloned in 1992, 1995, and 2002
showed a 3- to 4-fold reduced sensitivity. Analysis of
epitope diversity in the patient H-derived E1E2 sequences
identified several amino acid changes in linear MAb
epitopes that may partly explain the neutralization patterns observed. For example, mutation of amino acid 417
N/K within the 3/11 epitope in clones 92.C1 and 95.C8
may reduce their sensitivity to neutralization. However,
clone 02.E10, which is resistant to 3/11 neutralization,
shows no amino acid changes within the epitope, suggesting that changes outside the linear epitope modulate
3/11 activity. To analyze the effect these amino acid
Table 1. Sensitivity of HCVpp to Autologous and Heterologous Serum Neutralization Over Time
HCVpp
1977
1991
1992
1995
2002
HCV⫹ 1
HCV⫹ 2
cH77
H77.19
H77.20
H91.64
H91.A6
H91.A11
H91.B1
H92.C1
H95.C8
H02.E10
⬍50
⬍50
⬍50
⬍50
⬍50
⬍50
⬍50
⬍50
⬍50
⬍50
10,000
4,000
5,000
12,000
7,000
500
500
⬍50
⬍50
⬍50
10,000
5,000
5,000
20,000
1,000
3,000
⬍50
500
⬍50
⬍50
10,000
6,000
8,000
5,000
2,000
1,500
⬍50
500
1,000
⬍50
28,000
5,000
5,000
23,000
25,000
500
⬍50
1,000
2,000
⬍50
16,000
12,000
15,000
12,000
14,500
15,000
12,000
15,500
14,500
14,000
18,500
16,000
16,500
12,000
14,500
13,000
15,500
16,000
15,500
16,500
H serum samples from 1977, 1991, 1992, 1995, and 2002, as well as 2 sera from unrelated chronically HCV infected individuals (HCV⫹ 1, HCV⫹
2), were tested for their ability to neutralize HCVpp bearing E1E2 clones that had been amplified from the same serum samples. Values represent
ID90 titers, ie, the dilution of serum that gave a 90% inhibition of the respective HCVpp.
February 2007
HCV ESCAPES NEUTRALIZING ANTIBODIES AND CTL
673
Region
E1
HVR
E2
Amino acid position1
mAb epitope2
T-cell epitope2
22222333
33455037
46436814
––––––––
||
|––
—— –––—
333333334444444444
889999990000000001
782456780123456781
||||||||||||||||||
——————————————————
––––––––––––––––––
44444444444455555556666677
13333445678802233380123501
62689574761322928913177494
|||||||
|–——
||––——
||––––––––––
——————— –—
|||||–––––––––––––––––––
––—————
1977 sequences
H77C
H77.16
H77.18
H77.19
H77.20
H77.21
H77.22
H77.23
H77.24
H77.26
H77.25
H77.27
H77.28
HPCST77
H77IMC
pH21
H
PCVH11
H90
GATKTSAV
––––––––
––––––––
––––––––
––––––––
––––––––
––––––––
––––––––
––––––––
––––––––
––––––––
––––––––
––––––––
––––––––
––––––––
––––––––
––––––––
––––––––
ST–T––––
HVNGRTTALVGLLTPAKN
––––––––––––––––––
––––––––––S–––––––
––––––––––S–––––––
––––––––––S–––––––
––––––––––S–––––––
––––––––––S–––––––
–A––––––––S–––––––
––S–H–A–IA––F–L–––
––––––––––S–––––––
––––––––––S–––––––
––––––––––––––––––
––––––––––––––––––
––S–––––––––––––––
––S–––––––––––––––
––––––––––––––––––
––––––––––––––––––
––S–H–––––––––––––
––S––SVLIASF––RP––
NETWLQKPAALESRSAFVLWDIVDSA
––––––––––––––––––––––––––
–D––VY–S––P––K–E––R––––––T
––––––––––P––––––––N––––––
––––––––––––––––––––––––––
–––R––––––––––––––––––––––
–G––––––––––––––––––––––––
–––––––––––G––––––––––––––
–D––VY–S––P––K–E––R––––––T
–D––VY–S––P––K–E––R–––––––
––––––––––––––––––––––––––
––––––––––––––––––––––––––
––––––––––––G–––––––––––––
–––––H––––––––––––––––––––
––––––––––––––––––––––––––
––––––––––––––––––––––––––
–––––––––––––––––––R––––––
–D–––R––––––––––––––––––––
KDA–IHG–D–P–––NELIR–H–I––T
1991-2003 sequences
H90
H91.58
H91.60
H91.61
H91.64
H92.72
H92.73
H92.74
H92.76
H93.85
H93.86
H93.87
H93.88
H94.A2
H94.A4
H94.A5
H94.A7
H95.C2
H95.C3
H95.C4
H95.C7
H02.E1
H02.E4
H02.E7
H02.E3
H03.G1
H03.G5
H03.G10
H03.G11
STTKTTAV
––––––T–
––––––––
––––––––
––––––––
–––R––––
–––––––I
G–––––TI
––––––––
––––––––
––––––––
–––R––––
––––––––
–––R––––
––AR––––
–––R––––
–––R––––
––––––––
–––R––––
G––R––––
–––R––––
–––RA–––
–––RA–––
–––R–––A
–––RA–––
––ARA–––
–––RA–––
–––RA–––
–––RA–––
HVSGRSVLIASFLTRPKN
––––––––F––L––––––
–––A–––––––L––––––
–––––––FV–––––––––
––––––––F–––––K––K
––––––––F–––––Q–––
––––––––F––L––––––
––––––––F––L––––––
–––––––––––IF–––––
Y–––––––F–––––K–––
Y–––––––F–––––K–––
Y–––––––F–––––K–––
Y–––––––F–––––K–––
–––A––––L––––SQ–––
–––A––––L––––SQ–––
–––A––––L––––SQ–––
–––––––FV–––––L–––
–––––––––T–L–SL–––
–––A––––L––––SQ–––
R––A––––L––––SQ–––
–––A––––L––––SQ–––
Y––A––A–F–N–F–P–QR
Y––A––A–F–N–F–P–QR
Y––A––A–F–N–F–P–QR
Y––A––A–F–N–F–P–QR
YA–A––A–F–N–F–P–QK
Y––A––A–F–N–F–P–QK
Y––A––A–F–N–––P–QK
Y––A––A–F–N–F–P–QK
KDAWIHGPDAPESRNELIRWHIIDST
N––––––––––––––––––––V––––
–––––––S––––––––P––––V––––
N–––L––––––––K––––––––––––
N–––L––––––––––––––––––––A
–––––––––––––––––––––––GG–
N–––––––––––––––––––––––G–
N––––––T––––––––––––––––––
–––––––T––––––––––––R–––––
N–––L––––––––––––––––V–N––
N–––L––––––––––––––––V–N––
N–––L––––––––––––––––V–N––
N–––L––––––––––––––––V–N––
––––––––––––––––––––––––––
––––––––––––––––––––––––––
––––––––––––––––––––––––––
N–––L–––––––––––––––––––––
–––––––T––––––––––––––––––
–––––––S–––––––––T––––––––
–––––––S–––––––––––R––––––
–––––––S––––––––––––––––––
N–––LYR––––QN–––F––––V––––
N–––LYR––––Q–––––––––––N––
N–––LYR––V–Q––––F––––V––––
N–––LYR––––QN–––F––––V––––
N–––LYK––V–Q––––F–––––––––
N––RLYK––V–Q––––F–––––––––
N–––LYK––––QN–––F––––V––––
N––RLYK––V–Q––––F–––––––––
All residues in the E1E2 sequence that showed phylogenetically informative aa changes are represented. Changes are indicated relative to the
H77 or H90 (a clone that was previously obtained from patient H in 1990; accession M62382) sequence with “–” indicating conserved sequence.
1 - aa coordinates (shown in columns) are relative to the strain H sequence; 2 - “⫹” indicates that the respective site is within a known nAb
epitope (information from own data as well as the HCV Immunology database at hcv.lanl.gov) or within a T-cell epitope identified in this study
(Figures 4 and 5).
BASIC–LIVER,
PANCREAS, AND
BILIARY TRACT
Table 2. Amino Acid Changes in the H Derived E1E2 Sequence Set and Their Relation to Defined B- and T-cell Epitopes
674
VON HAHN ET AL
GASTROENTEROLOGY Vol. 132, No. 2
Table 3. Sensitivity of HCVpp Bearing H E1E2 Glycoproteins to Neutralization by mAbs and Soluble CD81
Monoclonal antibodies
Location
Epitope1
cH77
77.19
77.20
91.64
91.A6
E1E2
clones
91.A11
91.B1
92.C1
95.C8
02.E10
9/27
396–407
TAGLVGLLTPG
3/11
412–423
NIQLINTNGSWHIN
2/69a
432–443
ESLNTGWLAGLP
1/39
432–443
ESLNTGWLAGLF
7/16b
436–447
TGWLAGLFYQHK
11/20
436–447
sCD81
TGWLAGLFYQHK undefined
0.8
–––––––––––
0.7
–––––––––––
0.7
–––––––––––
0.7
VL–FASF––K–
1.1
VL–FASF––K–
1.2
VF–VASF––R–
1.4
VL–FASF––K–
⬎20
VL–IASIF–R–
⬎20
VL–ITSL–SL–
⬎20
AL–FANFF–––
1.1
––––––––––––––
2.0
––––––––––––––
1.3
––––––––––––––
2.0
k–––––––––––––
1.1
k–––––––––––––
5.0
––––––––––––––
2.0
k–––––––––––––
⬎20
–––––K––––––––
⬎20
–––––K––––––––
⬎20
k–––––––––––––
2.0
––––––––––––
2.5
––––––––––––
2.5
––––––––––––
2.3
D–––A–––––––
2.3
D–––A–––––––
2.4
D–––A–––––––
2.5
D–––A–––––––
⬎20
D–––A––I––––
⬎20
D–––A––I––––
⬎20
D–––A–––––––
3.0
––––––––––––
3.5
––––––––––––
4.0
––––––––––––
3.6
D–––A–––––––
4.0
D–––A–––––––
3.8
D–––A–––––––
3.7
D–––A–––––––
⬎20
D–––A––I––––
⬎20
D–––A––I––––
⬎20
D–––A–––––––
1.0
––––––––––––
1.5
––––––––––––
1.8
––––––––––––
2.5
A––––––––H–G
1.1
A––––––––H–G
1.3
A––––––––H–G
1.6
A––––––––H–G
⬎20
A––I–––––H–G
⬎20
A––I–––––H–G
⬎20
A––––––––Y–R
0.4
––––––––––––
0.2
––––––––––––
0.4
––––––––––––
0.3
A––––––––H–G
0.4
A––––––––H–G
0.7
A––––––––H–G
0.4
A––––––––H–G
1.5
A––I–––––H–G
1.5
A––I–––––H–G
2.0
A––––––––Y–R
0.4
0.2
0.4
0.3
0.4
0.7
0.4
1.5
1.5
2.5
Monoclonal antibodies against HCV E2 and soluble CD81 (sCD81) were tested for neutralization of HCVpp bearing sequential E1E2 glycoproteins
cloned from patient H. Numbers indicate the concentration of inhibitor in g/mL required to neutralize viral infectivity by 90% (ID90). Below each
ID90 value changes in the linear epitope sequence compared to cH77 are indicated, with “-” denoting conserved residues.
BASIC–LIVER,
PANCREAS, AND
BILIARY TRACT
changes may have on antigen recognition we studied
MAb binding to the diverse E1E2 gps by EIA. Four of the
6 neutralizing MAbs (3/11, 7/16, 9/27, 11/20) bound
lysates from 293T expressing H77 E1E2 in a GNA lectin
capture EIA. All of the MAbs showed reduced binding to
H gps cloned late in infection (Figure 3). In contrast, 2
nonneutralizing MAbs (6/1a and 9/75) bound all H gps
cloned over the course of the infection (data not shown).
In the case of 3/11 and 9/27 the loss of reactivity in the
EIA paralleled a decrease in neutralizing activity against
viruses bearing gps cloned in 1992 and at later time
points, whereas 11/20 showed reduced reactivity for 1992
and later E1E2 gps while maintaining its neutralizing
ability, suggesting differences between MAbs in the
amount of Ab required to bind and neutralize a virus
particle and that required to saturate antigen in an EIA.
Because we hypothesized that patient H serum antibodies selected for escape variants, we screened all serum
samples for reactivity to a panel of overlapping peptides
representing the H77 E1E2 sequence, some of which
represent epitopes recognized by the neutralizing MAbs.
Reactivity was observed only for the HVR peptide (data
not shown), suggesting that the selecting Ab response
was specific for conformational epitopes overlapping the
defined MAb neutralization epitopes.
HCV Escape From T-Cell Responses
Given the association between sequence polymorphism in nAb epitopes and viral escape we were inter-
ested to know whether any of the remaining polymorphic
sites corresponded to sites recognized by T cells. To
probe for escape mutations that may have arisen due to
selection pressure exerted by HCV-specific T-cell responses, we screened PBMCs from patient H in 2003 for
reactivity to a series of 15-mer overlapping peptides representing the H77 E1E2 sequence in an interferon-␥ enzyme-linked immunospot assay (Figure 4A). Three peptides, located at residues E1 226-240, E1 296-310, and E2
436-450, were recognized and represent new T-cell
epitopes. Both E1 peptides were recognized by CD4 T
cells, whereas the E2 peptide was recognized by both CD4
and CD8 T cells (Figure 4B). Interestingly, all 3 epitopes
contained between 1 and 4 sites that showed changes
within the H sequence set. The consensus sequence of the
E1 226-240 peptide changed from CVPCVREGNASRCWV in 1977 to CVPCVRESNTSRCWV in 2003; similarly E1 296-310 changed from RRHWTTQDCNCSIYP
to RRHWTTQDCNCTIYP and E2 436-450 from GWLAGLFYQHKFNSS to GWLAGLFYYHRFNSS. Furthermore, these mutations abrogated CD4 T-cell recognition
of one of the E1 peptides and CD4 and CD8 recognition
of the E2 peptide (Figure 5). These changes in the E1
T-cell epitopes had appeared between 1977 and 1990
(Table 2). The E2 T-cell epitope had accumulated 2
changes by 1990, these residues changed again between
1990 and 2002, with one of them showing a further
change between 2002 and 2003 (Table 2 and Figure 5).
Figure 5. Escape from CD4 and CD8 T-cell recognition owing to mutations in the H 2003 E1E2 consensus sequence. CD4 or CD8 T cells
obtained in 2003 were stimulated with peptides representing the E1
226-240, E1 296-310, and E2 436-450 epitopes in their original 1977
(WT) form or with mutations that were present in the 2003 consensus
(MUT). Interferon-␥ responses were assessed by enzyme-linked immunospot assay.
Collectively, these data show that the HCV E1E2 sequence not only is targeted by antibodies, but by CD4
and CD8 T cells and that ongoing HCV sequence change
mediates viral escape from these T-cell responses.
Discussion
This study is an in-depth analysis of the nAb
response against autologous HCV E1E2 sequences and
shows ongoing evasion of the virus-specific adaptive immune response during persistent infection. Patient H is a
uniquely well-suited subject for this study because he is
representative of many patients with mild, chronic hepatitis C and clinically is well characterized with serum
samples spanning a long period of time. Furthermore,
specific research tools such as sequences, infectious
clones, established HCVpp, and well-characterized antibodies are available.11,22,36
Although several studies have investigated the CD8
T-cell responses during HCV infection and their impact
on viral evolution and disease outcome, the role(s) of
HCV ESCAPES NEUTRALIZING ANTIBODIES AND CTL
675
CD4 T cells and nAbs are much less well defined. Several
earlier reports have suggested that nAbs develop late in
primary HCV infection and have a minimal role in controlling acute HCV replication.15,17,37,38 However, these
studies measured antibody neutralization of HCVppbearing gps of prototype laboratory strains and not the
virus population within the patient. The generation of
HCVpp-bearing gps cloned from serial serum samples
from patient H (Figure 2) allowed us to study the dynamics of autologous and heterologous nAb responses
over time (Figure 1A; Table 1).
HCVpp provide an ideal tool for the study of nAb
responses because they can be made to display a range of
related glycoprotein sequences, allowing for direct comparison of their neutralization behavior. The percentage
of patient H– derived E1E2 clones that produced infectious HCVpp, albeit sufficient for our purposes, was
small (21.4%). Inspection of the 146 polymorphic sites in
the sequence set showed a marked bias in the distribution of transitional changes (17 G to A vs 26 A to G and
11 C to T vs 32 T to C, P ⬍ .0001, Fisher exact test) that
was indicative of reverse-transcriptase errors. However, it
certainly is feasible that significant numbers of E1E2
sequences that are present in infected serum are either
defective or, despite being functional in vivo, are not
capable of generating infectious HCVpp in vitro.
Our data show that nAbs against HCVpp-bearing gps
representing the infecting strain are detected at seroconversion and that these early strain-specific nAbs appear to
target the HVR (Figure 1C). However, the response
against the H77 HVR, a sequence representative of the
initially infecting virus, as detected by EIA, is short lived
and undetectable by 28 weeks (Figure 1B), suggesting
that additional strain-specific epitopes account for the
low-titer anti-H77 nAb response observed at this point
(Figure 1A). It is intriguing to speculate that the short
half-life of the antibody response against this HVR sequence that was present and triggered a humoral response early in infection could be the result of a lack of
CD4 T-cell help required for the establishment of memory B-cell responses.
Cross-reactive nAb responses were first detected at 111
weeks after infection and increased in titer and breadth
to recognize distant HCV genotypes. Despite this crossreactivity, the nAb response efficiently neutralized viruses
bearing ancestral E1E2 sequences yet was less able to
neutralize viruses expressing E1E2 sequences that were
dominant in the serum at the time of sampling (Table 1).
These data parallel what has been reported for human
immunodeficiency virus39,40 and suggest that the nAb
response lags behind the rapidly evolving gp sequences
present within the quasispecies population. Sera from 2
unrelated chronically HCV-infected individuals could
neutralize all H-derived E1E2 clones tested in this study,
underscoring the conclusion that nAb responses elicited
during chronic infection are broadly reactive to most
BASIC–LIVER,
PANCREAS, AND
BILIARY TRACT
February 2007
676
VON HAHN ET AL
BASIC–LIVER,
PANCREAS, AND
BILIARY TRACT
E1E2 sequences, including ones that the host immune
system has not encountered. Thus, viruses bearing autologous E1E2 sequences are not insensitive to serum antibody–mediated neutralization per se. The observed resistance of viruses expressing gps cloned from samples
collected in 1992 and later sequences to Abs present in
earlier serum samples suggests that viral escape from the
humoral immune response continues after decades of
chronic infection. HCV continuously generates new sequences that are poorly neutralized by serum Abs. The
dominance of neutralization-resistant sequences in the
viral population at all time points (Table 1) suggests that
viral evolution is driven by strong antibody-mediated
selective pressures. Because serum antibodies from
chronically HCV-infected individuals can neutralize diverse E1E2 sequences (Table 1 and Bartosch et al,15
Logvinoff et al,17 and Meunier et al19), it seems that E1E2
sequences that dominate during the course of chronic
HCV infection must exploit specific gaps in the Ab repertoire of the individual host to evade the polyclonal
humoral immune response. This strongly suggests the
presence of a mechanism that removes nAb-sensitive sequences from the quasispecies pool.
The high rate of amino acid substitutions in defined
linear neutralization epitopes (Tables 2 and 3) suggests
that these regions are under selective pressure(s). Rapid
mutation in the HVR has long been suspected to be a
consequence of antibody-mediated selection.41,42 Compelling evidence that antibodies exert selective pressure
on the HVR in vivo stems from the observation that HVR
evolution is reduced in hypogammaglobulinemic patients.43,44 Sequence analysis of envelope clones from
patient H shows the expected high rate of amino acid
substitutions in the HVR. HCVpp-bearing gps cloned
from samples obtained in 1992, 1995, and 2002 were
resistant to neutralization by an MAb targeting the HVR
of the 1977 sequence (Table 3). Our data suggest that
HCV can readily escape anti-HVR nAbs, consistent with a
model that this region may act as an immune decoy.45,46
It should be noted, however, that changes in defined
epitopes may only partly predict the ability of an antibody to neutralize a given E1E2 sequence. For example,
clones 91.B1 and 02.E10 are identical in the 3/11 epitope,
however, HCVpp-02.E10 escapes 3/11 neutralization, illustrating the complexity of the 3/11-gp interaction and
its modulation by amino acid changes outside the minimally defined linear epitope.
Several publications have reported that a broad and
robust virus-specific CD4 and CD8 response is important
for controlling the acute phase of HCV replication and
achieving clearance of infection.9 –11 The emergence of
viral variants bearing escape mutations in CD8 T-cell
epitopes is associated with the development of chronic
infection.7,47– 49 Moreover, with the development of
chronic infection, T-cell responses seem to become narrower and less vigorous.50 –52 However, little is known
GASTROENTEROLOGY Vol. 132, No. 2
about the impact of T-cell responses on the evolution of
HCV E1E2. An analysis of T-cell reactivity against the
consensus H77 sequence in PBMCs obtained from H in
2003 revealed strong reactivity against 4 epitopes, 2 of
them located in a single peptide (Figures 4 and 5). All of
these epitopes coincide with residues that show phylogenetically informative changes over time, changes that
completely abrogated T-cell reactivity in 3 of 4 epitopes
(Figure 5). Escape from cellular immune responses previously has been well documented only for CD8 T cells,
yet our study uncovered 2 instances of viral escape in a
CD4 epitope. These data strongly suggest that CD4 and
CD8 T-cell– as well as B-cell–mediated selective pressures
impact on HCV glycoprotein evolution.
The quasispecies nature of HCV in vivo has long been
viewed as important for the pathobiology of this disease.
Several reports have shown that quasispecies complexity
is associated with failure to clear acute infection7,36 and is
associated inversely with the severity of HCV-related disease, in which patients with mild liver disease have a
diverse repertoire of viral sequences.53,54 The latter observation has led several investigators to suggest that the
HCV-specific immune responses that limit viral damage
to the liver may lead to selective expansion of antigenically diverse viral variants.54 –58 This model of viral escape
is analogous to that reported for human immunodeficiency virus.39,40,59 Here, we extend the earlier-described
data by showing that amino acid changes in the gps
directly reduce autologous serum antibody and MAb
neutralization (Tables 1 and 3), and that the majority of
sequence changes in E1E2 are located in regions identified to contain T-cell or antibody epitopes (Table 2 and
Yusim et al35). These data provide a compelling explanation for the failure of the immune response in this
patient to resolve HCV infection. Our study highlights
the importance of studying autologous and heterologous
nAb responses in individuals with different disease outcomes.
References
1. Alter M. Hepatitis C virus infection in the United States. J Hepatol
1999;31(Suppl 1):88 –91.
2. Alter M, Margolis H, Krawczynski K, Judson F, Mares A, Alexander
W, Hu P, Miller J, Gerber M, Sampliner R, et al. The natural history
of community-acquired hepatitis C in the United States. The
Sentinel Counties Chronic non-A, non-B Hepatitis Study Team.
N Engl J Med 1992;327:1899 –1905.
3. Martell M, Esteban J, Quer J, Genesca J, Weiner A, Esteban R,
Guardia J, Gomez J. Hepatitis C virus (HCV) circulates as a
population of different but closely related genomes: quasispecies
nature of HCV genome distribution. J Virol 1992;66:3225–3229.
4. Brown R, Juttla V, Tarr A, Finnis R, Irving W, Hemsley S, Flower D,
Borrow P, Ball J. Evolutionary dynamics of hepatitis C virus envelope genes during chronic infection. J Gen Virol 2005;86:
1931–1942.
5. Weiner A, Erickson A, Kansopon J, Crawford K, Muchmore E,
Hughes A, Houghton M, Walker C. Persistent hepatitis C virus
infection in a chimpanzee is associated with emergence of a
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
cytotoxic T lymphocyte escape variant. Proc Natl Acad Sci U S A
1995;92:2755–2759.
Chang K, Rehermann B, McHutchison J, Pasquinelli C, Southwood S, Sette A, Chisari F. Immunological significance of cytotoxic T lymphocyte epitope variants in patients chronically infected by the hepatitis C virus. J Clin Invest 1997;100:2376 –
2385.
Erickson A, Kimura Y, Igarashi S, Eichelberger J, Houghton M,
Sidney J, McKinney D, Sette A, Hughes A, Walker C. The outcome
of hepatitis C virus infection is predicted by escape mutations in
epitopes targeted by cytotoxic T lymphocytes. Immunity 2001;15:
883– 895.
Seifert U, Liermann H, Racanelli V, Halenius A, Wiese M, Wedemeyer H, Ruppert T, Rispeter K, Henklein P, Sijts A, Hengel H,
Kloetzel PM, Rehermann B. Hepatitis C virus mutation affects
proteasomal epitope processing. J Clin Invest 2004;114:250 –
259.
Cooper S, Erickson A, Adams E, Kansopon J, Weiner A, Chien D,
Houghton M, Parham P, Walker C. Analysis of a successful immune response against hepatitis C virus. Immunity 1999;10:
439 – 449.
Lechner F, Wong D, Dunbar P, Chapman R, Chung R, Dohrenwend
P, Robbins G, Phillips R, Klenerman P, Walker B. Analysis of
successful immune responses in persons infected with hepatitis
C virus. J Exp Med 2000;191:1499 –1512.
Thimme R, Oldach D, Chang K, Steiger C, Ray S, Chisari F.
Determinants of viral clearance and persistence during acute
hepatitis C virus infection. J Exp Med 2001;194:1395–1406.
Shoukry NH, Grakoui A, Houghton M, Chien DY, Ghrayeb J,
Reimann KA, Walker CM. Memory CD8⫹ T cells are required for
protection from persistent hepatitis C virus infection. J Exp Med
2003;197:1645–1655.
Grakoui A, Shoukry NH, Woollard DJ, Han JH, Hanson HL,
Ghrayeb J, Murthy KK, Rice CM, Walker CM. HCV persistence and
immune evasion in the absence of memory T cell help. Science
2003;302:659 – 662.
Hsu M, Zhang J, Flint M, Logvinoff C, Cheng-Mayer C, Rice C,
McKeating J. Hepatitis C virus glycoproteins mediate pH-dependent cell entry of pseudotyped retroviral particles. Proc Natl Acad
Sci U S A 2003;100:7271–7276.
Bartosch B, Bukh J, Meunier J, Granier C, Engle R, Blackwelder
W, Emerson S, Cosset F, Purcell R. In vitro assay for neutralizing
antibody to hepatitis C virus: evidence for broadly conserved
neutralization epitopes. Proc Natl Acad Sci U S A 2003;100:
14199 –14204.
Drummer HE, Maerz A, Poumbourios P. Cell surface expression
of functional hepatitis C virus E1 and E2 glycoproteins. FEBS Lett
2003;546:385–390.
Logvinoff C, Major M, Oldach D, Heyward S, Talal A, Balfe P,
Feinstone S, Alter H, Rice C, McKeating J. Neutralizing antibody
response during acute and chronic hepatitis C virus infection.
Proc Natl Acad Sci U S A 2004;101:10149 –10154.
McKeating J, Zhang L, Logvinoff C, Flint M, Zhang J, Yu J, Butera
D, Ho D, Dustin L, Rice C, Balfe P. Diverse hepatitis C virus
glycoproteins mediate viral infection in a CD81-dependent manner. J Virol 2004;78:8496 – 8505.
Meunier JC, Engle RE, Faulk K, Zhao M, Bartosch B, Alter H,
Emerson SU, Cosset FL, Purcell RH, Bukh J. Evidence for crossgenotype neutralization of hepatitis C virus pseudo-particles and
enhancement of infectivity by apolipoprotein C1. Proc Natl Acad
Sci U S A 2005;102:4560 – 4565.
Yanagi M, Purcell R, Emerson S, Bukh J. Transcripts from a single
full-length cDNA clone of hepatitis C virus are infectious when
directly transfected into the liver of a chimpanzee. Proc Natl Acad
Sci U S A 1997;94:8738 – 8743.
HCV ESCAPES NEUTRALIZING ANTIBODIES AND CTL
677
21. Kolykhalov A, Agapov E, Blight K, Mihalik K, Feinstone S, Rice C.
Transmission of hepatitis C by intrahepatic inoculation with transcribed RNA. Science 1997;277:570 –574.
22. Ogata N, Alter H, Miller R, Purcell R. Nucleotide sequence and
mutation rate of the H strain of hepatitis C virus. Proc Natl Acad
Sci U S A 1991;88:3392–3396.
23. Flint M, Maidens C, Loomis-Price LD, Shotton C, Dubuisson J,
Monk P, Higginbottom A, Levy S, McKeating JA. Characterization
of hepatitis C virus E2 glycoprotein interaction with a putative
cellular receptor, CD81. J Virol 1999;73:6235– 6244.
24. Flint M, Logvinoff C, Rice C, McKeating J. Characterization of
infectious retroviral pseudotype particles bearing hepatitis C virus glycoproteins. J Virol 2004;78:6875– 6882.
25. Hammond A, Lewis J, May J, Albert J, Balfe P, McKeating J.
Antigenic variation within the CD4 binding site of human immunodeficiency virus type 1 gp120: effects on chemokine receptor
utilization. J Virol 2001;75:5593–5603.
26. Inchauspe G, Zebedee S, Lee D, Sugitani M, Nasoff M, Prince A.
Genomic structure of the human prototype strain H of hepatitis C
virus: comparison with American and Japanese isolates. Proc
Natl Acad Sci U S A 1991;88:10292–10296.
27. Nei M, Gojobori T. Simple methods for estimating the numbers of
synonymous and nonsynonymous nucleotide substitutions. Mol
Biol Evol 1986;3:418 – 426.
28. Yang Z. Maximum likelihood estimation on large phylogenies and
analysis of adaptive evolution in human influenza virus A. J Mol
Evol 2000;51:423– 432.
29. Rogers J, Swofford D. A fast method for approximating maximum
likelihoods of phylogenetic trees from nucleotide sequences.
Syst Biol 1998;47:77– 89.
30. Posada D, Crandall K. MODELTEST: testing the model of DNA
substitution. Bioinformatics 1998;14:817– 818.
31. Takaki A, Wiese M, Maertens G, Depla E, Seifert U, Liebetrau A,
Miller J, Manns M, Rehermann B. Cellular immune responses
persist and humoral responses decrease two decades after recovery from a single-source outbreak of hepatitis C. Nat Med
2000;6:578 –582.
32. Rahman F, Heller T, Sobao Y, Mizukoshi E, Nascimbeni M, Alter
H, Herrine S, Hoofnagle J, Liang T, Rehermann B. Effects of
antiviral therapy on the cellular immune response in acute hepatitis C. Hepatology 2004;40:87–97.
33. Afdhal NH. The natural history of hepatitis C. Semin Liver Dis
2004;24(Suppl 2):3– 8.
34. Chen M, Sallberg M, Sonnerborg A, Weiland O, Mattsson L, Jin L,
Birkett A, Peterson D, Milich DR. Limited humoral immunity in
hepatitis C virus infection. Gastroenterology 1999;116:135–
143.
35. Yusim K, Richardson R, Tao N, Dalwani A, Agrawal A, Szinger J,
Funkhouser R, Korber B, Kuiken C. Los alamos hepatitis C immunology database. Appl Bioinformatics 2005;4:217–225.
36. Farci P, Shimoda A, Coiana A, Diaz G, Peddis G, Melpolder J,
Strazzera A, Chien D, Munoz S, Balestrieri A, Purcell R, Alter H.
The outcome of acute hepatitis C predicted by the evolution of
the viral quasispecies. Science 2000;288:339 –344.
37. Lavillette D, Morice Y, Germanidis G, Donot P, Soulier A, Pagkalos E, Sakellariou G, Intrator L, Bartosch B, Pawlotsky J, Cosset
F. Human serum facilitates hepatitis C virus infection, and neutralizing responses inversely correlate with viral replication kinetics at the acute phase of hepatitis C virus infection. J Virol
2005;79:6023– 6034.
38. Netski D, Mosbruger T, Depla E, Maertens G, Ray S, Hamilton R,
Roundtree S, Thomas D, McKeating J, Cox A. Humoral immune
response in acute hepatitis C virus infection. Clin Infect Dis
2005;41:667– 675.
39. Frost SD, Wrin T, Smith DM, Kosakovsky Pond SL, Liu Y, Paxinos
E, Chappey C, Galovich J, Beauchaine J, Petropoulos CJ, Little SJ,
Richman DD. Neutralizing antibody responses drive the evolution
BASIC–LIVER,
PANCREAS, AND
BILIARY TRACT
February 2007
678
40.
41.
42.
43.
44.
45.
46.
47.
BASIC–LIVER,
PANCREAS, AND
BILIARY TRACT
48.
49.
50.
VON HAHN ET AL
of human immunodeficiency virus type 1 envelope during recent
HIV infection. Proc Natl Acad Sci U S A 2005;102:18514 –
18519.
Richman DD, Wrin T, Little SJ, Petropoulos CJ. Rapid evolution of
the neutralizing antibody response to HIV type 1 infection. Proc
Natl Acad Sci U S A 2003;100:4144 – 4149.
Kato N, Ootsuyama Y, Sekiya H, Ohkoshi S, Nakazawa T, Hijikata
M, Shimotohno K. Genetic drift in hypervariable region 1 of the
viral genome in persistent hepatitis C virus infection. J Virol
1994;68:4776 – 4784.
Kato N, Sekiya H, Ootsuyama Y, Nakazawa T, Hijikata M, Ohkoshi
S, Shimotohno K. Humoral immune response to hypervariable
region 1 of the putative envelope glycoprotein (gp70) of hepatitis
C virus. J Virol 1993;67:3923–3930.
Kumar U, Monjardino J, Thomas HC. Hypervariable region of
hepatitis C virus envelope glycoprotein (E2/NS1) in an agammaglobulinemic patient. Gastroenterology 1994;106:1072–1075.
Booth JC, Kumar U, Webster D, Monjardino J, Thomas HC. Comparison of the rate of sequence variation in the hypervariable
region of E2/NS1 region of hepatitis C virus in normal and
hypogammaglobulinemic patients. Hepatology 1998;27:223–
227.
Ray S, Wang Y, Laeyendecker O, Ticehurst J, Villano S, Thomas
D. Acute hepatitis C virus structural gene sequences as predictors of persistent viremia: hypervariable region 1 as a decoy.
J Virol 1999;73:2938 –2946.
Mondelli M, Cerino A, Segagni L, Meola A, Cividini A, Silini E,
Nicosia A. Hypervariable region 1 of hepatitis C virus: immunological decoy or biologically relevant domain? Antiviral Res 2001;
52:153–159.
Cox A, Mosbruger T, Mao Q, Liu Z, Wang X, Yang H, Sidney J,
Sette A, Pardoll D, Thomas D, Ray S. Cellular immune selection
with hepatitis C virus persistence in humans. J Exp Med 2005;
201:1741–1752.
Tester I, Smyk-Pearson S, Wang P, Wertheimer A, Yao E, Lewinsohn D, Tavis J, Rosen H. Immune evasion versus recovery after
acute hepatitis C virus infection from a shared source. J Exp Med
2005;201:1725–1731.
Timm J, Lauer G, Kavanagh D, Sheridan I, Kim A, Lucas M, Pillay
T, Ouchi K, Reyor L, Schulze zur Wiesch J, Gandhi R, Chung R,
Bhardwaj N, Klenerman P, Walker B, Allen T. CD8 epitope escape
and reversion in acute HCV infection. J Exp Med 2004;
200:1593–1604.
Cox AL, Mosbruger T, Lauer GM, Pardoll D, Thomas DL, Ray SC.
Comprehensive analyses of CD8⫹ T cell responses during longitudinal study of acute human hepatitis C. Hepatology 2005;42:
104 –112.
GASTROENTEROLOGY Vol. 132, No. 2
51. Ulsenheimer A, Gerlach JT, Gruener NH, Jung MC, Schirren CA,
Schraut W, Zachoval R, Pape GR, Diepolder HM. Detection of
functionally altered hepatitis C virus-specific CD4 T cells in acute
and chronic hepatitis C. Hepatology 2003;37:1189 –1198.
52. Wedemeyer H, He XS, Nascimbeni M, Davis AR, Greenberg HB,
Hoofnagle JH, Liang TJ, Alter H, Rehermann B. Impaired effector
function of hepatitis C virus-specific CD8⫹ T cells in chronic
hepatitis C virus infection. J Immunol 2002;169:3447–3458.
53. Qin H, Shire NJ, Keenan ED, Rouster SD, Eyster ME, Goedert JJ,
Koziel MJ, Sherman KE. HCV quasispecies evolution: association
with progression to end-stage liver disease in hemophiliacs infected with HCV or HCV/HIV. Blood 2005;105:533–541.
54. Sheridan I, Pybus O, Holmes E, Klenerman P. High-resolution
phylogenetic analysis of hepatitis C virus adaptation and its
relationship to disease progression. J Virol 2004;78:3447–
3454.
55. Chien D, Choo Q, Ralston R, Spaete R, Tong M, Houghton M, Kuo
G. Persistence of HCV despite antibodies to both putative envelope glycoproteins. Lancet 1993;342:933.
56. Farci P, Alter H, Govindarajan S, Wong D, Engle R, Lesniewski R,
Mushahwar I, Desai S, Miller R, Ogata N, et al. Lack of protective
immunity against reinfection with hepatitis C virus. Science
1992;258:135–140.
57. Farci P, Alter H, Wong D, Miller R, Govindarajan S, Engle R,
Shapiro M, Purcell R. Prevention of hepatitis C virus infection in
chimpanzees after antibody-mediated in vitro neutralization. Proc
Natl Acad Sci U S A 1994;91:7792–7796.
58. Farci P, Orgiana G, Purcell R. Immunity elicited by hepatitis C
virus. Clin Exp Rheumatol 1995;13(Suppl 13):S9 –S12.
59. Ross H, Rodrigo A. Immune-mediated positive selection drives
human immunodeficiency virus type 1 molecular variation and
predicts disease duration. J Virol 2002;76:11715–11720.
Received August 29, 2006. Accepted November 13, 2006.
Address requests for reprints to: Thomas von Hahn, MD, Center for
the Study of Hepatitis C, The Rockefeller University, 1230 York Avenue,
New York, New York 10021. e-mail: vonhaht@rockefeller.edu; fax:
(212) 327-7048.
Supported by PHS grants CA57973, AI50798, U19 AI40034, Medical Research Council, UK (G0400802), the National Institute of Diabetes and Digestive and Kidney Diseases intramural research program, the Greenberg Medical Research Institute, and the Starr
Foundation. T.v.H. is supported by a postdoctoral fellowship from the
Deutsche Forschungsgemeinschaft.
The authors thank Merna Torres and Ke Hu for expert technical
assistance.