Suppression of Host Immune Response by the
Core Protein of Hepatitis C Virus: Possible
Implications for Hepatitis C Virus Persistence
This information is current as
of June 18, 2017.
Subscription
Permissions
Email Alerts
J Immunol 1999; 162:931-938; ;
http://www.jimmunol.org/content/162/2/931
This article cites 35 articles, 22 of which you can access for free at:
http://www.jimmunol.org/content/162/2/931.full#ref-list-1
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 1999 by The American Association of
Immunologists All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
References
Mary Kathryn Large, David J. Kittlesen and Young S. Hahn
Suppression of Host Immune Response by the Core Protein of
Hepatitis C Virus: Possible Implications for Hepatitis C Virus
Persistence1
Mary Kathryn Large,* David J. Kittlesen,* and Young S. Hahn2*†
H
epatitis C virus (HCV)3 is a positive strand RNA virus
belonging to the family Flaviviridae (1, 2). HCV was a
major cause of transfusion-associated hepatitis before
the development of serologic tests for HCV-contaminated blood
products, and it remains a major agent of community-acquired
hepatitis (3, 4). Characteristic features of HCV infection include a
high incidence of persistent infection and progression to chronic
hepatitis (3). Chronic HCV infection of the liver is a strong risk
factor for development of hepatocellular carcinoma (5).
The high incidence of HCV persistence after infection suggests
that this virus has evolved a mechanism to evade the host response,
probably by inhibiting the immune response necessary for viral
clearance during acute infection. One proposed mechanism of immune evasion is the generation of viral variants during infection
that could escape from Ab or CTL recognition. This strategy is
consistent with the high error rate of the viral polymerase during
viral RNA replication and may account for the high frequency of
HCV quasispecies detected in HCV-infected patients (6). Alternatively, as observed for many of the large DNA viruses, such as
adenovirus and herpes simplex virus (7–9), HCV may encode one
*Beirne Carter Center for Immunology Research and †Department of Pathology, University of Virginia, Charlottesville, VA 22908
Received for publication May 4, 1998. Accepted for publication September 29, 1998.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1
This work was supported by grants from the Public Health Service (AI37569) and
the American Liver Foundation (to Y.S.H.).
2
Address correspondence and reprint requests to Dr. Young S. Hahn, Beirne Carter
Center for Immunology Research, Box MR4 – 4012, University of Virginia, Charlottesville, VA 22908. E-mail address: [email protected]
3
Abbreviations used in this paper: HCV, hepatitis C virus; CTLp, precursor of CTL;
C, core; NS, nonstructural; VV, vaccinia virus; pfu, plaque-forming units.
Copyright © 1999 by The American Association of Immunologists
or more products that act to inhibit viral clearance by the host and
can lead to progressive or persistent viral infection.
The induction of virus-specific CTLs during infection is a wellestablished mechanism for virus elimination during infection. In
the case of HCV, CTLs have been recovered from both the liver
(10) and the periphery (11) of chronically infected patients. However, the frequency of HCV-specific CTL precursor (CTLp) and
effectors is much lower than that observed during infection with
other persistent viruses, such as hepatitis B virus or HIV (12, 13).
It is currently unknown whether this low CTLp in chronically infected patients reflects an immunosuppressive mechanism causally
linked to the ability of the HCV to establish and maintain viral
persistence after infection.
The HCV RNA genome is approximately 9.5 kb in length and
encodes a long polyprotein. Ten discrete proteins are produced
from this polyprotein by proteolytic processing. The structural proteins include the core (C) and the E1 and E2 glycoproteins and are
found in the N-terminal portion of the polyprotein. The nonstructural (NS) proteins involved in RNA replication are found in the
remainder of polyprotein. Among the HCV gene products, only the
highly conserved HCV core protein has been suggested to have an
immunomodulatory function. This is based on the in vitro demonstration that the core protein binds to the cytoplasmic domain of
certain members of the TNF receptor superfamily (14) and can
modulate the sensitivity of cells expressing HCV core to TNFmediated lysis in culture (15).
Since HCV infects only humans and chimpanzees and replicates
inefficiently in cell culture, studies aimed at assessing the abilities
of individual HCV-encoded proteins to modulate host immune responses during HCV infection have been difficult to perform. To
examine this issue in a model system, we have used infection of
mice with recombinant vaccinia viruses (VV) expressing HCV
gene products as a way to assess possible immunomodulatory
0022-1767/99/$02.00
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
Hepatitis C virus (HCV) is a major human pathogen causing mild to severe liver disease worldwide. This positive strand RNA
virus is remarkably efficient at establishing chronic infections. Although a high rate of genetic variability may facilitate viral escape
and persistence in the face of Ag-specific immune responses, HCV may also encode proteins that facilitate evasion of immunological surveillance. To address the latter possibility, we examined the influence of specific HCV gene products on the host immune
response to vaccinia virus in a murine model. Various vaccinia/HCV recombinants expressing different regions of the HCV
polyprotein were used for i.p. inoculation of BALB/c mice. Surprisingly, a recombinant expressing the N-terminal half of the
polyprotein (including the structural proteins, p7, NS2, and a portion of NS3; vHCV-S) led to a dose-dependent increase in
mortality. Increased mortality was not observed for a recombinant expressing the majority of the nonstructural region or for a
negative control virus expressing the b-galactosidase protein. Examination of T cell responses in these mice revealed a marked
suppression of vaccinia-specific CTL responses and a depressed production of IFN-g and IL-2. By using a series of vaccinia/HCV
recombinants, we found that the HCV core protein was sufficient for immunosuppression, prolonged viremia, and increased
mortality. These results suggest that the HCV core protein plays an important role in the establishment and maintenance of HCV
infection by suppressing host immune responses, in particular the generation of virus-specific CTLs. The Journal of Immunology,
1999, 162: 931–938.
932
HOST IMMUNE RESPONSE SUPPRESSION BY HCV CORE
FIGURE 1. VV recombinants expressing HCV gene products. A diagram of the HCV genome is shown with the noncoding regions (lines) and the open
reading frame (box) indicated. In the polyprotein, the portions encoding the structural proteins and nonstructural proteins are shown as hatched or open
boxes, respectively, and the individual cleavage products produced by co- and post-translational processing are indicated. Below, the various vHCV
recombinants used in the study and the regions of the polyprotein they express are indicated by lines.
Materials and Methods
Cell lines
The P815 (H-2d) mastocytoma and BSC40 monkey kidney cell lines were
maintained in DMEM supplemented with 10% (v/v) FBS and 1% (w/v)
glutamine.
Plasmid constructions and generation of VV recombinants
VV/HCV recombinants are designated by the region of the HCV (H strain;
genotype 1a) polyprotein sequence expressed. Two recombinants, vHCV-S
and vHCV-NS, have been described previously (18, 19). For mapping studies, a nested set of C-terminal deletions was constructed using PCR to
introduce a stop codon at or near the end of each individual protein. VV
insertion vectors were either pTM3 (20) or pBRTM (19), and the HCVspecific portions amplified by PCR were verified by sequence analysis. The
resulting plasmids were then used to rescue the corresponding VV/HCV
recombinants by standard methods (21). The salient features of the six VV
constructs used in this study are summarized in Fig. 1. The vHCV-S expresses HCV C, E1, E2, p7, NS2, and a portion of NS3. The vHCV-C/p7,
vHCV-C/E2, vHCV-C/E1, and vHCV-C express the C to p7, C to E2, C to
E1, and C proteins, respectively. The vHCV-NS expresses the majority of
NS2 (which begins at residue 809) through the end of polyprotein (NS5B).
VV recombinants were rescued, plaque purified, and expanded in BSC40
cells. Their titers were determined by standard plaque assay using BSC40
cells. Various VV/HCV recombinants were no different in their viral
growth rates (data not shown). The production of the expected HCV-specific proteins was verified by infection of BSC40 cells, radiolabeling, and
immunoprecipitation with specific antisera (18, 19) (data not shown). The
negative control VV recombinant expressing b-galactosidase, vSC11, has
been described (22).
The expression of core protein in mice inoculated with vHCV-C or
vHCV-S was confirmed by Western blot analysis in tissues obtained from
VV recombinant-infected mice. The core protein was expressed in liver
tissue from mice inoculated with vHCV-C and vHCV-S, but was not expressed in liver tissue from mice inoculated with a control VV recombinant
(vSC11; data not shown).
Infection of mice and virus titration
Four- to six-week-old BALB/c mice were purchased from Taconic Farms
(Germantown, NY). Mice were inoculated i.p. with various doses of the
VV recombinants as indicated in the text and figure legends. On days 1–5
after virus inoculation, mice were sacrificed, and liver tissue was harvested
to determine recombinant VV titers. Tissue samples were weighed, homogenized in Iscove’s DMEM containing 10% newborn calf serum, and
centrifuged at 500 3 g for 10 min, and pellets were resuspended in PBS
containing 1% FCS and 1 mM MgCl2. Virus was released from the pellets
by three freeze-thaw cycles followed by sonication.
To assay for plaque-forming virus, 200 ml of 1021–1023 dilutions were
incubated with confluent BSC40 monolayers in six-well plates for 1 h at
37°C. Inocula were removed and replaced with DMEM containing 10
mg/ml penicillin/streptomycin, 10% FCS, and 2 mM L-glutamine followed
by incubation for 2 days at 37°C. Monolayers were fixed with 7% formaldehyde and stained with crystal violet. Plaques were counted to determine the viral titer of each liver sample, and the viral titer was calculated
as log plaque-forming units (pfu) per gram of tissue.
Measurement of primary VV-specific CTL responses
To determine the primary VV-specific CTL response, mice were infected
i.p. with 5 3 107 pfu of a vHCV-S recombinant or control recombinant
vHCV-NS. On day 5 postinoculation, mice were sacrificed, and spleens
were harvested. A single cell suspension of splenocytes was prepared and
purified through Isopaque-Ficoll. Cells were harvested from the layer between the Isopaque-Ficoll and medium, and washed twice with medium to
remove the residual Ficoll. To measure VV-specific CTL responses, these
purified splenocytes were tested using 51Cr-labeled P815 target cells previously infected with wild-type VV. The percentage of specific 51Cr release
was determined by standard procedures (23). Values for 51Cr release are
the mean of quadruplicate samples; SDs were typically ,5%.
Quantitation of VV-specific primary CTL
To determine CTLp frequency by limiting dilution (24), three BALB/c
mice per group were infected with 5 3 107 pfu of vHCV-S or vHCV-NS
(as a control). At 5 days postinfection, spleen cells from infected mice were
pooled and diluted to 5 3 104 to 4 3 106 cells/well. Diluted splenocytes
were cultured with VV-infected syngeneic splenocytes as stimulators (105
cells/well). Cells in individual wells were harvested after 5 days, and CTL
activity was measured on VV-infected P815 target cells. P815 (H-2d) target
cells were infected for 1 h at 37°C with the wild-type VV and were labeled
with 150 mCi of 51Cr for 2 h at 37°C. After washing to remove free label,
target cells were incubated with VV-specific CTL effectors for 6 h at 37°C,
as described previously (23, 25). Calculations using the Taswell method
generate a frequency estimate (1/f), a corresponding 95% confidence interval, and the x2 estimate of probability ( p) for the frequency estimate. In
these studies, p . 0.05 indicates that the frequency estimate is statistically
acceptable.
Cytokine analysis of bulk cultures
Primary mixed lymphocyte cultures were prepared using spleens harvested
on day 5 from mice infected with vHCV or vSC11 VV recombinants as
described above. Splenocytes were restimulated with irradiated VV-infected splenocytes. Responder cells were cultured at 4 3 106 cells/ml with
stimulators at a 5:1 responder:stimulator ratio in Iscove’s DMEM containing 10 mg/ml penicillin/streptomycin, 10% heat-inactivated FCS, 2 mM
L-glutamine, and 0.05 mM 2-ME. Cultures were incubated at 37°C, and a
portion of each culture (culture supernatant and cell pellet) was harvested
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
effects of specific HCV polypeptides on VV infection. The contributions of specific components of the immune response to VV
infection in mice have been well defined, and recovery from VV
infection has been demonstrated to be strongly correlated with the
generation of a virus-specific CTL response (16, 17).
In the study described here, we examined the virulence of various VV/HCV recombinants in mice and studied the host response
induced by recombinant viruses. Surprisingly, we found that a recombinant VV expressing the structural protein of HCV produced
a lethal disseminated infection in mice and concomitantly suppressed the VV-specific CTL response and the production of
proinflammatory cytokines as well. Using a series of VV recombinants expressing various C-terminally truncated polyproteins,
this immunosuppressive effect was mapped to the core protein.
These results suggest that expression of HCV core during HCV
infection could account for the low frequency of CTLp observed in
chronically infected patients and that HCV core may play a critical
role in establishing and maintaining persistent HCV infection.
The Journal of Immunology
933
FIGURE 2. Increased mortality in vHCV-S-infected mice. A, Percent survival of mice infected with vHCV recombinants. Five- to six-week-old BALB/c
mice (n 5 8) were inoculated with 108 pfu of vHCV-S, vHCV-NS, vSC11 virus. Infected mice were examined daily for morbidity and mortality. B, Viral
dose-dependent mortality in vHCV-S-infected mice. Groups of BALB/c mice (n 5 8) were inoculated with various doses (106, 107, 5 3 107, and 108
pfu/mouse) of vHCV-S virus. The percent survival is plotted up to day 10 postinoculation. Similar results were obtained in at least three experiments. The
p value was calculated by the Mann-Whitney U test based on paired comparison between mice that died and surviving mice. The mortality observed in
vHCV-S-infected mice was statistically significant.
Assay for target cell recognition by VV-specific CTL
VV-specific CTL were generated by a standard method. Two BALB/c mice
(6 – 8 wk old) were infected with 107 pfu of recombinant virus (vSC11)
encoding a b-galactosidase gene. Spleens were harvested from vSC11primed mice and were stimulated in vitro with vSC11-infected and irradiated (2000 rad) naive splenocytes. The level of target cell lysis was determined by a standard 51Cr release assay using P815 (H-2d) as target cells.
P815 (H-2d) cells were infected with 10 multiplicity of infection of vHCVNS, vHCV-S, or vHCV-C or were mock infected and were labeled with
51
Cr. CTL assay was performed as previously described (23). The spontaneous release was ,10%. The lysis of 51Cr-labeled VV-infected target
cell by VV-specific CTL was inhibited by unlabeled VV-infected target
cells (data not shown), suggesting that the VV-specific CTL activity on
P815 targets is mediated by class I MHC-restricted CD81 T cells.
Results
To assess the relationship of viral inoculum size to disease severity and mortality, groups of eight mice were infected i.p. with
varying doses of the structural protein expressing vHCV-S virus.
Lethal infection was demonstrable in a virus dose-dependent fashion. All recipients of 108 pfu of vHCV-S virus died by day 5
postinfection, and recipients of 5 3 107 pfu of virus also succumbed to lethal infection with 100% mortality evident by day 8
of infection (Fig. 2B). In addition, mice receiving smaller viral
inocula, i.e., 107 or 106 pfu, recovered from infection (Fig. 2B) and
showed neither morbidity nor mortality up to 21 days postinfection
(data not shown).
To determine the pathological changes associated with lethal
infection of vHCV-S-inoculated mice, recipients of 108 pfu of
vHCV-S virus were sacrificed on days 4 –5 postinfection, and various tissues were harvested for histologic studies. We found histologic evidence of both focal and confluent necrosis in the
spleens, livers, and ovaries of these mice (data not shown). These
findings suggest progressive lytic infection by the vHCV-S virus in
these primary target organs of VV replication.
HCV structural gene expression enhances viral virulence
To evaluate the possible effect of expression of HCV gene products on the virulence of and the host response to VV in the mouse,
we infected BALB/c (H-2d) mice i.p. with 108 pfu of a recombinant VV (rVV) expressing either the 59 half (vHCV-S) or the 39
half of the HCV genome (vHCV-NS). The vHCV-S virus expresses the HCV structural proteins, i.e., C, E1, E2, and p7, as well
as the nonstructural protein, NS2, and a portion of the NS3 protein,
while the vHCV-NS expresses the nonstructural proteins of HCV
exclusively (Fig. 1).
Mice infected with vHCV-S became clinically moribund by
days 1–2 postinfection, and 90% of these mice died by 5–7 days
after infection (Fig. 2A). By contrast, mice infected with 108 pfu of
the HCV-NS virus, which expresses the HCV nonstructural protein, survived, showed no clinical signs of infection or mortality,
and recovered from infection as efficiently as mice infected with
the control vSC11 virus (Fig. 2A). This latter virus is a rVV expressing the pSC11 recombination vector that is used for foreign
gene insertion into the VV genome (see Materials and Methods).
These findings suggested that expression of one or more HCV
structural proteins dramatically increased the virulence of the normally avirulent VV.
HCV structural gene expression is associated with sustained VV
replication in vivo
In view of the necropsy findings described above, we next determined if the HCV structural proteins expressed by the vHCV-S
virus were associated with sustained virus replication and elevated
virus titers in vivo. Three mice per group were infected with 108
pfu of vHCV-S, vHCV-NS, or the control recombinant vSC11
virus. Recipient livers were harvested daily over 5 days, and virus
titers in the infected livers were determined. As shown in Table I,
after an initial burst of virus replication on day 1 postinfection,
recipients of the control vSC11 virus and the vHCV-NS virus expressing the HCV nonstructural proteins rapidly cleared virus from
their livers, and liver viral titers from these mice were below detectable levels by day 4 of infection. This finding is consistent with
the uniform survival of mice infected with these viruses and suggests a vigorous host response to infection with these rVV.
By contrast, mice infected with the vHCV-S virus expressing
the HCV structural proteins failed to clear virus and maintained
high titers of infectious virus up to 5 days after inoculation, the
time period when we observed the lethality during vHCV-S infection. These results are in agreement with our necropsy findings and
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
after 24, 48, 72, and 96 h of stimulation. Culture supernatants were assayed
for cytokine levels by ELISA (26, 27). A standard ELISA protocol
(PharMingen, San Diego, CA) was used to measure the quantities of IL-2
and IFN-g in culture supernatants at each time point. For the standard curve
to determine the amount of produced cytokines, IL-2 and IFN-g purchased
from PharMingen were serially diluted and used for ELISA assay.
934
HOST IMMUNE RESPONSE SUPPRESSION BY HCV CORE
Table I. VV titers on BSC40 cells (log pfu/g tissue)a
Day
D
D
D
D
D
1
2
3
4
5
vSC11
vHCV-S
vHCV-NS
4.41 6 0.44
3.80 6 0.04
3.00 6 0.54
, 1.40
, 1.40
5.80 6 0.14
5.84 6 0.49
5.98 6 0.13*
5.83 6 0.50
6.10 6 0.15
4.56 6 0.28
4.32 6 0.64
2.83 6 0.43*
, 1.40
, 1.40
a
Three mice per group were inoculated with recombinant viruses expressing HCV
gene products. Liver tissues were harvested from mice each day after virus inoculation and pooled from three mice per group. VV titer from liver tissues was determined
on BSC 40 cells. The value (6 SD) represents a SD calculated from two separate
assays for VV titer. Experiments were reproducible from three independent experiments. *, Statistically significant difference between indicated pairs (p , 0.01).
Suppression of the VV-specific CTL response by HCV protein(s)
The finding that expression of the HCV structural proteins by rVV
resulted in the lack of viral clearance suggested that expression of
one or more HCV structural proteins drastically inhibited the host
response to viral infection. Since virus-specific CTL have been
demonstrated to play a critical role in virus clearance and recovery
from experimental viral infections, including VV infection (16,
17), it was of interest to determine whether the CTL response to
primary VV infection was suppressed in mice infected with
vHCV-S virus.
When mice were infected with 5 3 107 pfu of the vHCV-NS
virus, their immune splenocytes, harvested on day 5 postinfection,
exhibited significant cytolytic activity on VV-infected targets in a
standard in vitro cytotoxicity assay (Fig. 3A). The magnitude of the
in vivo primary VV-specific CTL response in vHCV-NS infected
mice on day 5 postinfection was comparable to that in mice infected with the control vSC11 virus (data not shown). As expected,
infectious virus was also not detectable in the livers of vHCV-NSinfected mice on day 5 postinfection (Fig. 3B).
By contrast, the in vivo primary CTL response to VV virus was
markedly reduced (.5-fold) on day 5 postinfection in the spleens
of mice infected with 5 3 107 pfu of the vHCV-S virus (Fig. 3A).
A comparable inhibition of in vivo primary CTL activity was also
observed on day 7 postinfection in vHCV-S-infected recipients
that survived to day 7 (data not shown). Commensurate with the
low levels of VV-specific CTL activity in the spleens of vHCVS-infected recipients, the livers of vHCV-S-infected mice had high
levels of infectious virus on day 5 postinfection (Fig. 3B). Liver
virus titers were at least 1000-fold greater than those of the vHCVNS-infected recipients on day 5 postinfection. These results suggest that the enhanced virulence of the vHCV-S virus may be due
in part to the inhibition of a VV-specific CTL response in infected
recipients through the action of one or more HCV structural proteins expressed in vHCV-S-infected cells.
To more precisely determine the degree of suppression of VVspecific CTL activity induced by the vHCV-S virus, we determined the frequency of CTLp in spleens of mice on day 5 postinfection with vHCV-S or vHCV-NS using limiting dilution analysis
according to the methods of Taswell (4). We found that mice infected with the rVV expressing the HCV structural proteins had
approximately a 10-fold lower frequency of CTLp on day 5 postin-
FIGURE 3. The VV-specific CTL response is suppressed in vHCV-Sinfected mice. A, Recognition of VV-infected target cells by primary VVspecific CTL. BALB/c mice were inoculated with 5 3 107 pfu of vHCV-S
or vHCV-NS. On day 5 after virus inoculation, spleens were harvested
from infected mice, and splenocytes were purified on a Isopaque-Ficoll
gradient. Purified splenocytes from vHCV-S-infected mice (n 5 3) and
vHCV-NS-infected mice (n 5 2) were added to VV-infected P815 target
cells prelabeled with 51Cr. The E:T cell ratio was 100:1. The percentage of
spontaneous lysis on uninfected target cells was 4 –5% (data not shown).
The spontaneous release was ,10%. B, Determination of VV titer. Liver
tissues were harvested from rVV-infected mice on day 5 after virus inoculation. Tissues were homogenized, and VV from liver tissue was released
by freezing and thawing of the cell pellet. VV titer was determined on
BSC40 monkey kidney cells and is presented as log plaque-forming units
per gram of tissue. Error bars indicate the SD of quadruplicate wells (A) or
the SD of duplicate virus titers (B). Experiments were repeated with three
times with similar results. p, Statistically significant difference between
indicated pairs (p , 0.01).
fection than recipients infected with virus expressing the nonstructural proteins (Fig. 4). The frequency of CTLp was 1.6 CTLp/106
immune splenocytes for the virulent vHCV-S virus and 12.5
CTLp/106 immune splenocytes for the avirulent vHCV-NS virus.
This result is in keeping with the above findings on the magnitude
of the in vivo primary CTL response to these two viruses.
Taken together, these results reinforce the view that the expression of one or more HCV structural proteins in infected cells in
vivo leads to inhibition of the host response to viral infection,
which is reflected in the suppression of the induction of virusspecific CTL. As discussed below, we also observed a marked
inhibition of IFN-g production by immune splenocytes from
vHCV-S-infected mice in response to antigenic stimulation in
vitro. Again, this finding suggests that one or more HCV gene
products can profoundly alter the host response to VV infection.
Identification of HCV core protein as the gene product involved
in immune modulation
Since the HCV-S virus expresses four putative HCV structural
proteins, C, E1, E2, and p7, it was important to determine whether
the immune suppression induced by infection with the vHCV-S
virus was mediated by a specific HCV structural gene product. To
approach this question, we constructed a panel of rVV expressing
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
strongly suggest that these recipients of vHCV-S died from overwhelming infection with this rVV. It is also noteworthy that the
effect of HCV structural protein expression on virus replication
and presumably the host response to infection was evident early in
the infectious process, as virus titers in the livers of vHCV-S recipients were 10-fold higher than titers of vHCV-NS virus or the
control VV on day 1 postinfection (Table I).
The Journal of Immunology
935
progressively larger deletions in the cDNA encoding the HCV
structural proteins (see Fig. 1). To evaluate the impact of the incremental loss of HCV structural genes on the host response to
infection, we evaluated the in vivo primary VV-specific CTL responses to each of these deletion mutants, to vHCV-S virus, and to
the vHCV-NS virus on day 5 after infection. As shown in Fig. 5A,
the immunosuppressive activity exhibited by the vHCV-S virus
expressing the full complement of HCV structural proteins was
also demonstrable in mice infected with a rVV expressing HCV
residues 1–192, which encodes the full length of the HCV core
protein exclusively. This result strongly suggests that the HCV
core protein is the viral gene product that suppresses the host response to infection and enhances the virulence of VV. In agreement with this cytotoxicity data, we also found that mice infected
with the rVV expressing only the HCV core protein (vHCV-C) had
elevated virus titers in their livers on day 5 of infection (Fig. 5B)
and succumbed to lethal infection (data not shown). Virus titers in
vHCV-C-infected mice were comparable to those in mice infected
with the vHCV-S virus expressing the full complement of HCV
structural proteins.
Alteration of cytokine synthesis by HCV gene product(s)
As noted above, along with the assessment of in vivo primary CTL
responses to infection, we also compared the in vitro production of
IFN-g and IL-2 in response to wild-type vaccinia-infected APCs of
immune splenocytes taken on day 5 after infection with the
vHCV-NS virus or the vHCV-S and vHCV-C viruses. IFN-g production by immune splenocytes from mice infected with the coreexpressing vHCV-C virus was profoundly suppressed (Fig. 6). The
degree of inhibition of IFN-g production in vitro was comparable
to that observed in splenocytes of mice infected with the vHCV-S
virus expressing the full complement of HCV structural proteins.
Infection with these core-expressing rVV also resulted in a modest,
but significant, inhibition of Ag-stimulated IL-2 production by immune splenocytes from vHCV-C- and vHCV-S-infected mice. Immune cells from mice infected with the vHCV-NS virus produced
high levels of both IFN-g and IL-2 (Fig. 6).
These results suggested that the HCV core protein was responsible for the diminished virus-specific CTL activity and IFN-g production as well as the enhanced virulence observed in mice in-
FIGURE 5. Identification of the HCV gene product involved in suppression of the VV-specific CTL response. A, Primary VV-specific CTL
response in mice inoculated with various vHCV recombinants. Two
BALB/c mice per group were inoculated with 5 3 107 pfu of vHCV-S,
vHCV-NS, and vHCV recombinants expressing the deleted polyproteins
shown in Fig. 1A. Purified splenocytes were prepared from spleens harvested on day 5 after virus inoculation. Primary VV-specific CTL from
pooled splenocytes was determined as described in Fig. 3. B, Determination of VV titer. The VV titer was determined on BSC40 monkey kidney
cells from liver tissues harvested on day 5 after virus inoculation. Error
bars represent the SD of quadruplicate wells (A) or the SD of duplicate
virus titers (B). Experiments were reproducible from two independent experiments. p, Statistically significant difference between the indicated pairs
(p , 0.01).
fected with the vHCV-S virus. Since activated CD81 T
lymphocytes effectors have been reported to produce high levels of
IFN-g in response to Ag stimulation (28 –30), inhibition of the
induction of virus-specific CTL by an HCV core-dependent suppressive mechanism would lead to decreased numbers of CTL effectors and diminished in vitro IFN-g production in response to
viral antigenic stimulation.
One potential mechanism by which HCV core could inhibit
CTL induction is that HCV core may interfere with viral Ag presentation at a step along the MHC class I processing pathway in
cells expressing core protein. To determine the impact of HCV
core on viral Ag presentation by MHC class I molecules, we examined the capacity of VV-specific CD81 CTL generated in response to infection with wild-type VV to lyse target cells displaying VV peptides after infection with the core-expressing vHCV-S,
vHCV-C virus, or with the core-negative vHCV-NS virus. The
VV-specific CTL do not lyse uninfected P815 target cells, and the
effects of VV-specific CTL are MHC restricted. As shown in Fig.
7, core protein expression had no significant effect on the recognition of VV epitopes by VV-specific CTL. This result suggests
that core protein does not disrupt Ag processing and/or presentation of viral Ag to CD81 T lymphocytes as shown by adenovirus
E3/19K protein (8). Therefore, the inhibitory effect of core on the
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 4. The CTLp frequency in vHCV-S-infected mice. On day 5
after virus inoculation, spleens were harvested from mice inoculated with
vHCV-S or vHCV-NS (n 5 3). Purified splenocytes were diluted severalfold and cultured with VV-infected splenocytes as stimulators. The CTL
activity for stimulated responder CTL was determined on VV-infected target cells. The frequency of CTLp was calculated by the Taswell method.
The p value was determined by x2 minimization, p . 0.10 for all estimates
(p . 0.05 indicates significance).
936
host response to viral infection is unlikely to be at the level of
CD81 T lymphocyte recognition of virally infected APCs.
Discussion
In this report we examined the effect of expression of HCV gene
products in a rVV on the in vivo virulence of this rVV and on the
regulation of the host response to VV in a murine model. We found
that the selective expression of the HCV structural proteins in
rVV(vHCV-S) dramatically enhanced the virulence of an otherwise avirulent VV strain. This enhanced virulence led to an increase in mortality of mice infected with this VV recombinant and
persistently elevated VV titers in primary targets of vaccinia replication in vivo, i.e., liver, spleen, and ovaries. We also observed
that this enhanced virulence and the sustained elevation of virus
titers in recipients of the vHCV-S were associated with a suppressed in vivo primary cytotoxic T lymphocyte response to VV
FIGURE 7. No inhibition effect on Ag processing and presentation by
HCV core protein. Recognition of target cells expressing the HCV gene
products by VV-specific CTL. 51Cr-labeled P815 (H-2d) target cells were
infected with 10 pfu/cell of the indicated vHCV or control recombinants
for 3 h. VV-specific CTL were added to infected or uninfected P815 target
cells at an E:T cell ratio of 50:1. The percent specific lysis was determined
as described in Materials and Methods. Error bars indicate the SD from
quadruplicate wells. Data are representative of three independent
experiments.
Ags in these recipients. In addition, immune cells from the spleens
of mice undergoing primary infection with this rVV (vHCV-S)
produced markedly lower levels of IFN-g in vitro in response to
VV Ags than immunocytes from animals infected with a rVV expressing vHCV-NS. Cells from these animals also exhibited a
modest, but significant, decrease in in vitro production of IL-2.
Using a series of rVV expressing various deletions in the gene
complex encoding the HCV structural proteins, we identified the
HCV core protein as the structural gene product that both accounted for the suppression of the host response to infection and
enhanced VV virulence.
Viruses have evolved a variety of mechanisms to ensure their
replication and survival and to circumvent the host immune response (31). One of the best characterized mechanisms is the inhibition of CD81 CTL response to the virus by the ability of viral
gene products to inhibit the processing of viral proteins and/or the
presentation of viral peptide/MHC complexes on the surface of
virus-infected target cells (32–34). Indeed, as noted above, expression of the HCV core protein does suppress CTL responses in this
model system. However, we could not detect any effect of core
protein on the presentation and recognition of processed viral Ag
by activated virus-specific CD81 CTL effectors. Rather, our data
from the analysis of the cytolytic activity of primary CD81 cytolytic effectors taken directly from infected animals as well as from
limiting dilution analysis of the frequency of activated CD81 T
lymphocytes giving rise to cytolytic effectors argue for an effect of
core protein primarily at the level of CD81 CTL induction. Therefore, it should be emphasized that one or more HCV gene product(s) may modulate the host response during natural HCV infection for the inhibition of viral Ag processing and presentation by
MHC class I molecules. In our murine model, however, we have
no evidence that the core protein enhances virulence and suppresses the host response by this mechanism.
Available evidence suggests that core protein has several functions in HCV replication, including viral RNA encapsidation into
viral nucleocapsid (35, 36). Because of the localization of the core
gene at the 59 end of the HCV polyprotein transcript, the core
protein is likely to be the first viral gene product produced in the
virus-infected cells. If, as our data suggest, core protein expression
may profoundly suppress the host response to virus infection, the
effect of core on the host immune response would be evident in an
earlier phase of virus infection before primary virus-specific CD81
T lymphocyte precursors give rise to CTL effectors. In this connection, it is noteworthy that mice infected with vHCV-C already
demonstrate a 10-fold higher titer of virus in the liver as early as
day 1 postinfection compared with mice comparably infected with
control vaccinia (vSC11) or vHCV-NS (Table I). This finding reenforces the view that HCV core protein acts to suppress the host
immune response at an early point in the process of viral infection.
Multiple mechanisms could account for the increase in viral
titers observed within 24 h of infection. Both type 1 IFNs, a and
b, have been shown to have important and nonredundant roles in
the control of VV infection (37). NK cells are also significant to
the early control of viral infections, and are greatly stimulated by
IL-12 or IFN-a and -b. IL-12 produced by macrophages or NK
cells synergize with TNF-a to stimulate the production of IFN-g
by NK cells. IL-12 also appears to be critical for the development
of Th1 responses, which can, in turn, augment CTL responses.
Therefore, a disruption in the expression or function of any of
these cytokines could explain not only the increased early viral
titers observed, but also the deficiencies in IL-2, IFN-g, and CTL
as shown here. Precedent for such a mechanism has been demonstrated with the inhibition of IL-12 synthesis by measles virus (38).
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 6. Down-regulated IFN-g and IL-2 production by the HCV
core protein. To examine the production of IFN-g and IL-2 in mice expressing the HCV core protein, Two BALB/c mice per group were inoculated with 5 3 107 pfu of vHCV-S, vHCV-C, or vHCV-NS as a control.
On day 5 after virus inoculation, splenocytes were harvested from two
infected mice per group, pooled, and stimulated in vitro with VV-infected
splenocytes from syngeneic mice. Supernatants from in vitro splenocyte
bulk cultures were assayed for the production of IFN-g and IL-2 by ELISA.
Error bars indicate the SD of quadruplicate wells for cytotoxic assay. Experiments were repeated with twice with similar results. p and #, Statistically significant difference between the indicated pairs (p , 0.01).
HOST IMMUNE RESPONSE SUPPRESSION BY HCV CORE
The Journal of Immunology
Acknowledgments
We thank our colleagues for many helpful discussions. We particularly
acknowledge Dr. Charles Rice and Jian Xu for providing us with recombinant vaccinia viruses expressing HCV gene products and constant support during this study.
References
1. Houghton, M., M. Selby, A. Weiner, and Q.-L. Choo. 1994. Hepatitis C virus:
structure, protein products and processing of the polyprotein precursor. In Current Studies in Hematology and Blood Transfusion: Hepatitis C Virus.
H. W. Reesink, ed. Karger, New York, pp. 1–11.
2. Choo, Q. L., G. Kuo, A. J. Weiner, L. R. Overby, D. W. Bradley, and
M. Houghton. 1989. Isolation of a cDNA clone derived from a blood-borne
non-A, non-B viral hepatitis genome. Science 244:359.
3. Hollinger, F. B. 1990. Non-A, non-B hepatitis viruses. In Virology. B. N. Fields,
ed. Raven Press, New York, pp. 2239 –2273.
4. Houghton, M. A., A. Weiner, J. Han, G. Kuo, and Q.-L. Choo. 1991. Molecular
biology of the hepatitis C viruses: implications for diagnosis, development and
control of viral disease. Hepatology 14:381.
5. Saito, I., T. Miyamura, A. Ohbayashi, H. Harada, T. Katayama, S. Kikuchi,
Y. Watanabe, S. Koi, M. Onji, Q.-L. Choo, et al. 1990. Hepatitis C virus infection
is associated with the development of hepatocellular carcinoma. Proc. Natl. Acad.
Sci. USA 87:6547.
6. Weiner, A. J., H. M. Geysen, C. Christopherson, J. E. Hall, T. J. Mason,
G. Saracco, F. Bonino, K. Crawford, C. D. Marion, K. A. Crawford, et al. 1992.
Evidence for immune selection of hepatitis C virus (HCV) putative envelope
glycoprotein variants: potential role in chronic HCV infections. Proc. Natl. Acad.
Sci. USA 89:3468.
7. Hill, A., P. Jugovic, I. York, G. Ross, J. Bennink, J. Yewdell, H. Ploegh, and
D. Johnson.1995. Herpes simplex virus turns off the TAP to evade host immunity.
Nature 375:411.
8. Marrack, P., and J. Kappler. 1994. Subversion of the immune system by pathogens. Cell 76:323.
9. York, I. A., C. Roop, D. W. Andrews, S. R. Riddell, F. L. Graham, and
D. C. Johnson. 1994. A cytosolic herpes simplex virus protein inhibits antigen
presentation to CD81 T lymphocytes. Cell 77:525.
10. Koziel, M. J., D. Dudley, J. T. Wong, J. Dienstag, M. Houghton, R. Ralston, and
B. D. Walker. 1992. Intrahepatic cytotoxic T lymphocytes specific for hepatitis C
virus in persons with chronic hepatitis. J. Immunol. 149:3339.
11. Koziel, M. J., D. Dudley, N. Afdhal, Q. L. Choo, M. Houghton, R. Ralston,
and B. D. Walker. 1993. Hepatitis C virus (HCV)-specific cytotoxic T lymphocytes recognize epitopes in the core and envelope proteins of HCV. J. Virol. 7:7522.
12. Rehermann, B., K. M. Chang, J. McHutchinson, R. Kokka, M. Houghton,
C. M. Rice, and F. V. Chisari. 1996. Differential cytotoxic T-lymphocyte responsiveness to the hepatitis B and C viruses in chronically infected patients. J. Virol.
70:7092.
13. Nelson, D. R., C. G. Marousis, G. L. Davis, C. M. Rice, J. Wong, M. Houghton,
and J. Y. Lau. 1997. The role of hepatitis C virus-specific cytotoxic T lymphocytes in chronic hepatitis C. J. Immunol. 158:1473.
14. Matsumoto, M., T. Hsieh, N. Zhu, T. VanArsdale, S. B. Hwang, K. S. Jeng,
A. E. Gorbalenya, S. Y. Lo, J. H. Ou, C. F. Ware, et al. 1997. Hepatitis C virus
core protein interacts with the cytoplasmic tail of lymphotoxin-B receptor. J. Virol. 71:1301.
15. Chen, C.-M., L.-R. You, L.-H. Hwang, and Y.-H. Lee. 1997. Direct interaction of hepatitis C virus core protein with the cellular lymphotoxin-b receptor
modulates the signal pathway of the lymphotoxin-b receptor. J. Virol. 71:
9417.
16. Blanden, R. V. 1971a. Mechanisms of recovery from a generalized virus infection: mousepox, II: passive transfer of recovery mechanisms with immune lymphoid cells. J. Exp. Med. 133:1074.
17. Blanden, R. V. 1971b. Mechanisms of recovery from a generalized virus infection: mousepox, III: regression of infectious foci. J. Exp. Med. 133:1090.
18. Dubuisson, J., H. H. Hsu, R. C. Cheung, H. B. Greenberg, D. G. Russell, and
C. M. Rice. 1994. Formation and intracellular localization of hepatitis C virus
envelope glycoprotein complexes expressed by recombinant vaccinia and Sindbis
viruses. J. Virol. 68:6147.
19. Grakoui, A., C. Wychowski, C. Lin, S. M. Feinston, and C. M. Rice. 1993.
Expression and identification of hepatitis C virus polyprotein cleavage products.
J. Virol. 67:1385.
20. Elroy-Stein, O., T. R. Fuerst, and B. Moss. 1989. Cap-independent translation of
mRNA conferred by encephalomyocarditis virus 59 sequence improves the performance of the vaccinia virus/bacteriophage T7 hybrid expression system. Proc.
Natl. Acad. Sci. USA 86:6126.
21. Mackett, M., and G. L. Smith. 1986. Vaccinia virus expression vectors. J. Gen.
Virol. 67:2067.
22. Moss, B., O. Elroy-Stein, T. Mizukami, W. A. Alexander, and T. R. Fuerst. 1990.
New mammalian expression vectors. Nature 348:91.
23. Braciale, T. J., M. E. Andrew, and V. L. Braciale. 1991. Heterogeneity and
specificity of cloned lines of influenza virus-specific cytotoxic T lymphocytes.
J. Exp. Med. 163:910.
24. Skinner, M. A., and J. Marbrook. 1976. An estimation of the frequency of precursor cells which generate cytotoxic lymphocytes. J. Exp. Med. 143:1562.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
The induction of an effective CD81 CTL response and the production of IFN-g during infection have been implicated as important factors in resistance to as well as recovery from VV infection
in the mouse (16, 17). Available evidence suggests that an effective
CD81 CTL response to virus infection can be mounted in the
absence of either IFN-g production or a functional IFN-gR (27).
Therefore, in the murine model of HCV core-mediated immune
suppression described here, it is unlikely that core protein acts by
inhibiting IFN-g production early in infection and thereby suppresses an IFN-g-dependent step in CD81 T lymphocyte differentiation into activated effectors. Since CD81 CTL effectors produce high levels of IFN-g in response to Ag (28 –30), it is more
likely that the diminished IFN-g production by immune splenocytes after infection with core expressing rVV reflects a core protein-dependent inhibition of activation of virus-specific CD81 T
cells into IFN-g-producing CTL effectors. One intriguing mechanism that could account for both the enhanced replication of core
expressing VV early in infection and the suppression of the virusspecific CTL response reported here is suggested by the recent
report of binding of HCV core protein to the cytoplasmic tail of the
human lymphotoxin-b receptor (14, 15). If HCV core can directly
interact with the intracellular signaling domains of one or more
members of the murine TNF receptor family in infected cells and
inhibit the function of these receptors, then this inhibitory effect on
TNF receptor signaling might account for the suppression of both
the early innate response to virus infection and the induction of a
specific CTL response observed here.
Although we observed a profound effect of HCV core expression on IFN-g production by immune splenocytes, the effect of
core on IL-2 production was less dramatic. One possible explanation for this result is suggested by the recent finding by Dr. Charles
Rice’s laboratory (unpublished observation) that HCV core can
up-regulate transcription of the IL-2 promoter. It is not yet clear
whether this activity reflects another strategy employed by HCV to
dysregulate the host response, but it does point out the potential for
this viral gene product to alter multiple steps in the host response
to infection with HCV.
In the studies reported here, enhanced virulence of VV was used
as a functional readout of HCV core protein-dependent suppression of the host response. Although rVV have been used by other
investigators to evaluate the effect of expression of foreign genes
on the immune response to VV infection, it was of concern that the
immunosuppressive effect of core observed by us in this model
system may reflect a suppressive effect dependent upon an interaction between HCV core and one or more VV gene products
unique to this class of DNA viruses. Although this possibility cannot as yet be formally excluded, it appears unlikely, since we have
been able to demonstrate that HCV core expression enhances the
replication and persistence of Sindbis virus (a member of the positive strand RNA alphavirus) in vivo in mice when the core gene
is expressed in a recombinant Sindbis virus (Y. S. Hahn, unpublished observation). This result adds support to our view that core
protein may likewise have an immunomodulatory effect during
HCV infection in humans.
One of the unique features of human HCV infection is the high
incidence of development of persistent HCV infection (3). This
fact strongly suggests that HCV has evolved one or more strategies
to suppress the host response during acute infection and thereby
facilitate the development of viral persistence. The results reported
here provide compelling evidence that the expression of the
HCV core protein may play a critical role in the establishment
and possibly the maintenance of persistence during HCV infection in humans.
937
938
HOST IMMUNE RESPONSE SUPPRESSION BY HCV CORE
25. Hahn, Y. S., C. S. Hahn, V. L. Braciale, T. J. Braciale, and C. M. Rice. 1992.
CD81 T cell recognition of an endogenously processed epitope is regulated primarily by residues within the epitope. J. Exp. Med. 176:1335.
26. Schreiber, R. D. 1991. Measurement of mouse and human interferon g. In Current Protocols in Immunology, Vol. 1, J. E. Cologan, A. M. Kruisbeek,
D. H. Margulies, E. M. Shevach, and W. Strober, eds. John Wiley & Sons, New
York.
27. Graham, M. B., D. K. Dalton, D. Giltinan, V. Braciale, T. A. Stewart, and
T. J. Braciale. 1993. Response to influenza infection in mice with a targeted
disruption in the interferon g gene. J. Exp. Med. 178:1725.
28. Morris, A. G., Y. L. Lin, and B. A. Askonas. 1982. Immune interferon release
when a cloned cytotoxic T cell line meets its correct influenza-infected target cell.
Nature 295:150.
29. Klein, J. R., D. H. Raulet, M. S. Pasternack, and M. J. Bevan. 1982. Cytotoxic T
lymphocytes produce immune interferon in response to antigen or mitogen.
J. Exp. Med. 155:1198.
30. Zinkernagel, R. M., and A. Althage. 1977. Antiviral protection by virus-immune
cytotoxic T lymphocytes: infected target cells are lysed before infectious virus
progeny is assembled. J. Exp. Med. 145:644.
31. Gooding, L. R. 1992. Virus proteins that counteract host immune defenses. Cell
71:5.
32. Cox, J. H., J. W. Yewdell, L. C. Eisenlohr, P. R. Johnson, and J. R. Bennink.
1990. Antigen presentation requires transport of MHC class I molecules from the
endoplasmic reticulum. Science 247:715.
33. Beck, S., and B. G. Barrell. 1988. Human cytomegalovirus encodes a glycoprotein homologous to MHC class I antigens. Nature 331:269.
34. Browne, H., G. Smith, S. Beck, and T. Minson. 1990. A complex between the
MHC class I homologue encoded by human cytomegalovirus and b2 microglobulin. Nature 347:770.
35. Santolini, E., G. Migliaccio, and M. N. La. 1994. Biosynthesis and biochemical
properties of the hepatitis C virus core protein. J. Virol. 68:3631.
36. Matsumoto, M., S. B. Hwang, K.-S. Jeng,, N. Zhu, and M. C. Lai. 1996. Homotypic interaction and multimerization of hepatitis C virus core protein. Virology
218:43.
37. Van Den Broek, M. F., U. Muller, S. Huang, M. Aguet, and R. M. Zinkernagel.
1995. Antiviral defense in mice lacking both a/b and g interferon receptors.
J. Virol. 69:4792.
38. Karp, C. L., M. Wysocka, L. M. Wahl, J. M. Ahearn, P. J. Cuomo, B. Sherry,
G. Trinchieri, and D. E. Griffin. 1996. Mechanism of suppression of cell mediated
immunity by measles virus. Science 273:228.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017