Immunity, Vol. 20, 293–303, March, 2004, Copyright 2004 by Cell Press
Functional Impairment of CD8?T Cells
by Regulatory T Cells
during Persistent Retroviral Infection
long-term chronic (persistent) infections. Once a virus
becomes persistent it is very difficult to eliminate. In
most cases an equilibrium is established between virus
proliferation and antiviral immunity such that the virus
persists with relatively little damage to the host. While a
host-virus de ´tente is an acceptable outcome with many
viruses, persistent infections with viruses such as hu-
man immunodeficiency virus (HIV) or the hepatitis vi-
ruses lead to life-threatening diseases. Even persistent
infections with normally innocuous viruses may lead to
severe disease in cases where the host becomes immu-
nocompromised. For example, latent infections with hu-
ening disease in AIDS patients or patients receiving
immunosuppressive drugs to treat cancer or prevent
We have been investigating persistent Friend virus
(FV) infections in mice as a model to study basic mecha-
nisms of immunological control of a persistent and im-
munosuppressive retroviral infection. FV infection of re-
of acute infection, but the mice are never able to com-
pletely eradicate the virus, and a lifelong chronic infec-
tion ensues. FV levels are low during the chronic or
one to ten infected cells per one hundred thousand
et al., 2002). The predominant reservoir for persistent
FV is B cells (Hasenkrug et al., 1998), although other cell
types also harbor persistent virus to a much lesser extent.
sion, including suppression of CD8?T cell-mediated
antitumor responses (Iwashiro et al., 2001a). This immu-
nosuppression is transferable to naive mice by adoptive
transfer of CD4?T cells, indicating that a subpopulation
of CD4?T cells, most likely adaptive regulatory T cells
(Bluestone and Abbas, 2003), mediates suppression of
theantitumor response(Iwashiroet al., 2001a). Conversely,
there isalso asubpopulation of CD4?T cells,most likely
Th1 type cells (Iwashiro et al., 2001b; Stromnes et al.,
2002), that is required to keep persistent FV levels at a
stable, low level (Hasenkrug et al., 1998). Depletion of
CD8?T cells are critical for recovery from the acute
phase ofFV infection (Hasenkrug etal., 1995; Robertson
et al., 1992), and even highly resistant mouse strains
such as C57BL/6 fail to control acute infections in the
absence of a CD8?T cell response (Hasenkrug, 1999).
However, in the persistent phase of infection, depletion
of CD8?T cells has no effect on virus levels (Hasenkrug
et al., 1998). One explanation of this transition is that
some form of immunological escape has occurred. Vari-
described in which infected cells are not recognized by
CD8?T cells (Johnson and Desrosiers, 2002; Lucas et
al., 2001; Xu et al., 2001). In addition, loss of CD8?T cell
function can also occur and has been described for
viruses like HIV (Andersson et al., 2002; Appay et al.,
2000; Geertsma et al., 1999; Migueles et al., 2002; Trim-
Ulf Dittmer,2Hong He,1Ronald J. Messer,1
Simone Schimmer,2Anke R.M. Olbrich,2Claes Ohlen,3
Philip D. Greenberg,3Ingunn M. Stromnes,1,5
Michihiro Iwashiro,1,6Shimon Sakaguchi,4
Leonard H. Evans,1Karin E. Peterson,1Guojun Yang,1
and Kim J. Hasenkrug1,*
1Laboratory of Persistent Viral Diseases
Rocky Mountain Laboratories
National Institute of Allergy
and Infectious Diseases
National Institutes of Health
Hamilton, Montana 59840
2Institut fu ¨r Virologie des Universita ¨tsklinikums
3Department of Medicine and Immunology
University of Washington
Seattle, Washington 98195
4Department of Experimental Pathology
Institute for Frontier Medical Sciences
The establishment of viral persistence generally re-
quires evasion of the host CD8?T cell response. Here
are fully capable of recognizing their cognate antigen
but their effector functions are suppressed by regula-
tory T cells. Virus-specific CD8?T cells adoptively
transferred into mice persistently infected with Friend
virus proliferated and appeared activated, but failed
to produce IFN? or reduce virus loads. Cotransfer ex-
periments revealed that a subpopulation of CD4?
T cells from persistently infected mice suppressed
IFN? production by the CD8?T cells. Treatment of
persistently infected mice with anti-GITR antibody to
ameliorate suppression by regulatory T cells signifi-
cantly improved IFN? production by transferred CD8?
T cells and allowed a significant reduction in viral
loads. The results indicate thatCD4?regulatory T cells
contribute to viral persistence and demonstrate an
immunotherapy for treating chronic retroviral infec-
Many human viruses such as herpes viruses, hepatitis
viruses, and retroviruses are able to evade immunologi-
cal destruction during acute infection and establish
5Present address: Department of Immunology, University of Wash-
ington, Seattle, Washington 98195.
6Present address: Tango Furusato Hospital, Kyoto-Fu, 629-3313
Table 1. Infectious Centers in Different Organs from Persistently Infected Mice
Blood CellsMouse Liver Cells
0/3 ? 107
0/3 ? 107
0/3 ? 107
0/3 ? 107
0/3 ? 107
0/3 ? 107
0/3 ? 107
0/3 ? 107
0/3 ? 107
The values are expressed as number of infectious centers (IC) per number of cells from the indicated tissue. The limit of detection for the
assay was one IC per 3 ? 107cells.
ant, not tested.
ble and Lieberman, 1998) and HCV (Gruener et al., 2001;
Urbani et al., 2002; Wedemeyer et al., 2002). However,
the reasons for CD8 dysfunction remain unclear. The
current studies investigate the failure of CD8?T cell
responses to eradicate persistent infection in the Friend
virus model. Results indicated that recognition by CD8?
T cells was intact, but effector function was impaired.
Adoptive transfer experiments indicated that CD4?
T cells from persistently infected mice could actively
suppress the effector function of CD8?T cells in vivo.
Furthermore, ameliorating the suppressive activity of
regulatory T cells by in vivo treatment with anti-GITR
antibody allowed adoptively transferred CD8?T cells to
produce IFN? and eliminate a significant proportion of
the persistent virus.
mately ten PCR cycles) during acute infection than the
respective signals during persistent infection (data not
shown), but the average env RNA to DNA ratios were
very similar during both phases of infection (27.0and 27.6,
respectively). Therefore, viral transcription appeared to
be just as active during persistent infection as during
Neither Mutation of the GagL CTL Epitope nor Clonal
Deletion of GagL-Specific CD8?T Cells Account
for Virus Escape
The finding of active virus transcription prompted us
to investigate mutation of a CD8?T cell epitope as a
mechanism of immunological escape. To date, one FV-
specific CD8 epitope has been described, an H-2Db-
restricted peptide from the leader portion of the F-MuLV
glycosylated gag protein (GagL) (Chen et al., 1996).
H-2Db/GagL tetramers (Schepers et al., 2002) were used
epitope by staining CD8?T cells from mice at 3 weeks
postinfection, the peak of the CTL response (Robertson
paring CD8?T cells from naive and acutely infected
mice are shown in Figures 1A and 1B, respectively. An
average of 2.7% of the CD8?T cells from the infected
mice stained tetramer-positive, indicating that GagL
was a significant CD8 epitope during acute FV infection.
To determine whether the GagL CD8?T cell epitope
was mutated in persistently infected mice, PCR primers
flanking the coding sequence were used to amplify pro-
viral DNA, which was then directly sequenced. Results
from 12 persistently infected mice all showed wild-type
sequences with no mutation of the CD8?T cell epitope
or sequences within 50 base pairs of both the 3? and
5? ends of the epitope sequence (data not shown). In
addition, viral cDNAs from five of the same mice were
sequenced to investigate whether only transcriptionally
type sequences were found, indicating that mutations
in the GagL CD8?T cell epitope did not account for FV
persistence (data not shown).
We tested for deletion of the T cells (Moskophidis et
al., 1993; Rich and Green, 2000) specific for the GagL
epitope by staining spleen cells from persistently in-
fected mice with H-2Db/GagL tetramers. Results from
three representative mice are shown in Figure 1C. In
seven persistently infected mice tested, the percentage
that the H-2Db/GagL-specific cell subset had not been
Persistent FV Infections Are Not Latent
We first sought to determine whether the mechanism
by which persistent FV escaped killing by CD8?T cells
was related to a loss of active provirus transcription in
chronically infected cells (latency). Previously we showed
mice (Hasenkrug et al., 1998), but it was possible that
in vitro culture may have activated transcription that
was silent in vivo. We therefore used amplification of
RNA by PCR as a sensitive method to detect transcrip-
tion from cells taken directly ex vivo. Tissues from six
persistently infected mice were tested for the presence
of viral env RNA. Positive signals were obtained from
all spleens, lymph nodes,bone marrows, and thymuses.
Five of the six mice had positive signals from nucleated
blood cells, and four of six liver samples were positive
(data not shown). In addition, infectious virus was de-
tected in cell suspensions from spleens, lymph nodes,
and bone marrows (Table 1). These results indicated
a wider tissue distribution of persistent FV than was
previously known and also that persistent FV was not
completely latent. However, they did not reveal whether
transcription was at a low level or a high level compared
to an acute infection. Since the spleen is one of the
most active sites of virus replication during acute FV
infection, we chose this tissue to compare transcription
levels during acute versus persistent infection using
real-time PCR. The ratios between viral RNA and DNA
were determined using host GAPDH signals as internal
controls. On average, both the viral RNA and DNA sig-
nals were about one thousand times higher (approxi-
CD8?T Cell Dysfunction
and naive control mice, and at 3 days posttransfer,
spleen cells from the host mice were analyzed by flow
cytometry. In uninfected control recipients we found an
average of 2.4% of the splenic CD8?T cells to be CFSE-
labeled donor cells. The donor cells were low in expres-
sion of the CD43 activation marker and had retained full
CFSE signal, indicating that no proliferation had oc-
T cells were adoptively transferred into persistently in-
fected mice, virtually all of the cells upregulated the
CD43 activation marker, and the bulk of the donor cells
underwent six cell divisions within 3 days (Figure 2B).
Proliferating CD8?T cells also expressed CD44 (data
not shown). By 7 days posttransfer almost all the CFSE
label was gone, indicating that more than seven cell
divisions had occurred (Figure 2C). At 1 week posttrans-
fer the CD8?donor cells had proliferated to an average
of 2.0 ? 107donor cells per spleen in the persistently
infected mice. These results indicated that the virus-
specific CD8?T cells had recognized and responded to
their cognate viral antigen.
FV-Specific CD8?T Cells Reduce Acute Infections
but Not Persistent Infections
Given the rapid proliferation and activation of the adop-
mine whether the transferred cells had any effect on
levels of persistent virus. At 1 week posttransfer there
was no reduction in spleen infectious center levels; in
fact, there was a slight increase (Figures 3A and 3B).
This increase could be accounted for by the infection
of the CD8?donor cells, which had an average of 1.5%
infection at 1 week posttransfer (as determined by de-
tection of viral antigen using flow cytometry, data not
shown). By 2 weeks posttransfer, infectious center lev-
els were still not reduced (Figure 3C). To confirm that
the TCR transgenic CD8?T cells had the capacity to
reduce viral loads in vivo, an adoptive transfer experi-
ment was performed into mice during acute infection.
By 1 week posttransfer the average infectious center
levels were reduced from 107.0per spleen to 105.3per
spleen, a reduction of approximately 50-fold or almost
ten million infectious centers per spleen (Figures 3D and
3E). Thus, the TCR Tg CD8?T cells were effective at
reducing acute Friend virus infection but ineffective at
pearance and ability to proliferate.
To determine whether higher levels of TCR Tg CD8?
T cell engraftment could account for the higher reduc-
tion of infection in the acutely infected mice, real num-
bers of cells were determined by repeating the adoptive
transfers using green fluorescent CD8?T cells obtained
from a cross between TCR transgenic mice and green
fluorescent protein (GFP) transgenic mice. Interestingly,
there was an average of 5-fold fewer donor CD8?T cells
in the acutely infected mice than in the persistently in-
fected mice at 1 week posttransfer (Table 2A). Lower
engraftment levels in the acutely infected mice were
likely due to less antigenic stimulation during the first
few days after transfer when virus levels were still low.
Thus, although the persistently infected mice became
better engrafted with virus-specific donor cells than the
acutely infected mice, the cells were still not effective
Figure 1. Tetramer Staining of CD8?T cells in Naive, Acutely In-
fected, and Persistently Infected Mice
All cells are CD8 gated and stained with DbGagL tetramers.
(A) CD8 T cells from three naive mice.
(B) CD8 T cells from three acutely infected mice (21 days post-
infection [dpi]). The geometric mean side scatter signal was
190.7 ? 3.6.
(C) CD8 T cells from persistently infected mice at 16 or 26 weeks
postinfection (wpi) as indicated.
The geometric mean side scatter signal was 174.7 ? 2.3. The differ-
ence in side scatter was statistically significant by Student’s t test,
p ? .0001.
deleted. Interestingly, mostof the tetramer-postive cells
from the persistently infected mice had a significantly
lower side scatter profile compared to tetramer-postive
cells fromthe acutely infectedmice (Figures 1Band 1C).
Side scatter is a measure of granularity and a reflection
ing that the tetramer-postive CD8?T cells from the per-
sistently infected mice had lower levels of cytotoxic
granules than the cells in the acutely infected mice.
Persistent Friend Virus Does Not “Hide”
from CD8?T Cells
To determine whether persistent Friend virus was utiliz-
regulation of MHC class I molecules, interference with
cealmentmechanisms, we tested whether virus-specific
CD8?T cells would become activated upon adoptive
transfer into persistently infected mice. CD8?T cells
from B6 mice carrying a transgenic T cell receptor (TCR)
specific for the H-2Db-restricted GagL epitope (Ohlen et
al., 2002) were adoptively transferred into persistently
infected mice. The TCR Tg CD8?T cells were labeled
eration could be measured. Four million CD8?T cells
were injected intravenously into persistently infected
Figure 2. Proliferation and Activation of Vi-
fer into Persistently Infected Mice
Flow cytometry was used to detect upregula-
tion of the CD43 activation marker and loss
of CFSE staining on CD8-gated cells taken
from spleens of recipient mice at the indi-
cated time points.
(A) CD8-gated T cells 3 days after adoptive
transfer into naive mice.
(B) CD8-gated T cells 3 days after adoptive
transfer into persistently infected mice.
(C) CD8-gated T cells 7 days after adoptive
transfer into persistently infected mice.
The numbers at the top indicate the number
of cell divisions of the CFSE-labeled cells di-
rectly below each box. The number of cell
divisions was calculated empirically.
in reducing virus levels. These results suggested the
possibility that the effector functions of the CD8 cells
preparation of cells inoculated into acutely infected
Suppression of CD8?T Cells Producing IFN? by CD4?
T Cells from Persistently Infected Mice
Previously we found that mice persistently infected with
Friend virus had increased levels of CD4?regulatory
tion of tumors (Iwashiro et al., 2001a). To determine
whether CD4?regulatory T cells were inhibiting the anti-
viral functions of virus-specific CD8?T cells, we per-
formed a cotransfer experiment in which acutely in-
fected mice received an infusion of GFP-labeled, TCR
Tg CD8?T cells and also received CD4?T cells from
either naive mice or persistently infected mice. The
CD4?T cells were genetically labeled (CD45.1), which
allowed us to analyze both the host and donor CD4?
were analyzed for cytokine expression by intracellular
Virus-Specific CD8?T Cells Fail to Produce IFN? after
Transfer into Persistently Infected Mice
As one measure of CD8?T cell effector function that
correlates with recovery from FV infection, intracellular
cytokine staining for interferon ? (IFN?) was determined
in TCR transgenic CD8?T cells adoptively transferred
into acutely infected or persistently infected mice. At 1
week posttransfer an average of 15.0% of the donor
cells in acutely infected mice produced IFN? whereas
only 2.4% of the CD8?T cells recovered from persis-
tently infected mice produced IFN? (p ? .0286). Repre-
sentative flow cytometry plots are shown in Figure 4.
Thus, although the virus-specific CD8?T cells prolifer-
ing adoptive transfer into persistently infected mice, the
cells were functionally impaired relative to the same
Figure 3. Infectious Center Levels after Adop-
tive Transfer of Virus-Specific TCR Tg CD8?
The data from experiments (A)–(C) were ob-
tained from (B10 x A.BY)F1 mice that were
infected with FV at least 8 weeks prior to
adoptive transfer (persistentlyinfected mice).
The data from experiments (D) and (E) are
mice. The mice were infected with FV 1 day
prior to adoptive transfer. (B6.SJL x A.BY)F1
mice were used so that the Ly5 (CD45.1)
marker could be used to distinguish host and
donor cells to obtain the data in Table 2B. No
strain-associated differences were seen in
these experiments. The difference in means
(log10) between (D) and (E) is statistically sig-
nificant by the Mann-Whitney test, p ?
CD8?T Cell Dysfunction
Table 2. Effects of Infection Status and Anti-GITR Treatment on Adoptively Transferred TCR Tg CD8?T Cells
A. Reconstitution Levels of TCR Tg CD8?T Cells in Acutely Infected and Persistently Infected Micea
1 Week Posttransfer into
Acutely Infected Mice
1 Week Posttransfer into
Persistently Infected Mice
Percentage of total spleen cells that were of donor origin (mean)
Percentage of CD8?T cells that were of donor origin (mean)
Number of donor CD8?T cells per spleen (mean)
3.9 ? 106
2.0 ? 107
B. Effects of Anti-GITR on Cytokine Production in Persistently Infected Miceb
% Donor CD8?T Cells % Donor CD8?T Cells
Treatment Producing IFN?
Adoptive transfer of 8 ? 106TCR Tg CD8?T cells plus rat Ig
Adoptive transfer of 8 ? 106TCR Tg CD8?T cells plus anti-GITR
aFor acute infections the CD8?T cells were transferred 1 day after infection, and the data were collected 1 week later (n ? 6 mice). For
persistent infections the mice were infected 2 months before adoptive transfer of CD8?T cells, and the data were collected 1 week later (n ?
5 mice). To distinguish host from donor cells, the TCR transgenic mice were mated with green fluorescent protein transgenic mice to obtain
donor cells with green fluorescence.
bB6.PL (Thy 1.1 congenic) mice persistently infected with FV for 2 months were adoptively transferred with 8 ? 106CD8?T cells from B6
TCR Tg mice. The anti-GITR group received 70 ?g of rat monoclonal anti-GITR every other day for a total of six injections beginning on the
day of adoptive transfer. The control group was treated with nonspecific rat immunoglobulin with the same dose and schedule. Statistical
analyses were done using the Mann-Whitney test (n ? 8 mice per group).
cytokine staining. There were significantly more IL-10-
producing CD4?T cells from persistently infected do-
nors than from naive donors (Figure 5B). Furthermore,
transfer of CD4?T cells from persistently infected mice
was associated with an increase in the percentage of
host CD4?T cells producing IL-10 (Figure 5B). Interest-
ingly, the cotransfer of CD4?T cells from persistently
infected mice significantly diminished the percentage
of IFN?-producing CD8?T cells of both host and donor
origin (Figure 5A). There was no significant difference
in the level of host or donor CD4?T cells producing
IFN? (Figure 5C), and there were no detectable IL-4
producing CD4?T cells in the host or donor T cells of
from persistently infected mice produced immunosup-
pressive IL-10 and diminished CD8?T cell IFN? re-
sponses in vivo.
Since CD4?T cells constitutively expressing CD25
are important regulatory T cells in the prevention of
autoimmune disease, the cotransfer experiment was re-
peated with CD4?T cells enriched for CD25-positive
or -negative subpopulations. Both the CD25-postitive
and -negative subpopulations of CD4?T cells sup-
T cells (Figure 5D). Thus, the immunosuppressive CD4?
T cells in this model are not confined to the subset of
CD4?T cells expressing high levels of CD25.
Figure 4. Intracellular Cytokine Staining for CD8?T Cells Produc-
At 1 week posttransfer of 4 ? 106Friend virus (DbGagL)-specific
TCR transgenic CD8?T cells, spleens were harvested, and donor
CD8?T cells were stained by intracellular cytokine staining for IFN?.
Spleen cells were from either uninfected mice (A), acutely infected
mice at 1 week postinfection (B), or persistently infected mice (C).
The cells were stimulated with 1 micromolar DbGagL peptide during
the 5 hr incubation with monensin. The difference in the percentage
cally significant by the Mann-Whitney test, p ? .0286. Very similar
results were obtained in a second experiment with 12 acutely in-
fected and 8 persistently infected mice. The virus-specific CD8?
donor T cells had significantly higher side scatter in acutely infected
mice (mean ? 220) than in the persistently infected mice (mean ?
186, p ? .0001 by t test).
Treatment of Mice with Anti-GITR Antibody Allows
CD8?T Cells to Produce IFN? and Reduce
We next sought to determine whether we could improve
the function of CD8?T cells by modulating the sup-
pressive effects of regulatory T cells in persistently in-
fected mice. Persistently infected B6 congenic mice
were treated with a monoclonal antibody against the
GITR molecule, a treatment which has been shown to
diminish the suppressive function of regulatory T cells
in vivo (Shimizu et al., 2002). Anti-GITR monoclonal anti-
Figure 5. Intracellular Cytokine Staining after Cotransfers of T Cells into Acutely Infected Mice
Percentages of T cells producing the indicated cytokines at 1 week post-adoptive transfer are indicated by the bars. The host mice were
acutely infected (A.BY x B10)F1. The donor TCR Tg CD8?T cells were from (B6.GFP x B6.TCR Tg)F1 mice and were gated by GFP expression.
Donor CD4?T cells were from naive or persistently infected (A.BY x B6.SJL) and were gated by CD45 expression. The cotransfers were done
with 2 ? 107CD4?T cells plus 4 ? 106CD8?TCR Tg T cells. P values for (A)–(C) were determined by Mann-Whitney tests, and the p value
for donor CD8?T cells in (A) was corrected for multiple comparisons using the Bonferroni posttest. (D) CD25-posititive and -negative
subpopulations of CD4?T cells from persistently infected mice were cotransferred with TCR Tg CD8?T cells and compared with cotransfers
of unfractionated CD4?T cells from naive mice (A) and two mice that received only TCR Tg CD8?T cells (*). P values were determined by
ering an agonistic signal to cell surface GITR, which
is constitutively expressed by CD4?regulatory T cells
no effect on levels of persistent infection (Figure 6C).
However, when anti-GITR therapy was combined with
adoptive transfer of 4 ? 106TCR Tg CD8?T cells, there
was a significant reduction in virus loads (Figure 6D).
Interestingly,when thenumberof adoptivelytransferred
TCR Tg CD8?T cells was increased to 8 ? 106cells per
mouse, a significant decrease in spleen virus levels was
T cells could be partially overcome by adding sufficient
numbers of CD8?T cells. An even greater decrease was
observed when anti-GITR therapy was combined with
adoptive transfer of 8 ? 106TCR Tg CD8?T cells per
mouse (Figure 6H). Compared with control mice (Figure
6E), combined anti-GITR and adoptive transfer therapy
produced a striking reduction from an average of 28,000
infectious centers per spleen to an average of only 500
per spleen (Figure 6H). In agreement with the results
from the cotransfer studies which showed immunosup-
pressive activity in both CD25-positive and -negative
subsets of CD4?T cells, depletion of CD25?T cells from
persistently infected mice did not produce a consistent
improvement in the ability of CD8?T cells to decrease
in spleen virus loads (data not shown).
kine production by donor TCR Tg CD8?T cells, the
combination therapy was repeated, and spleen cells
lular cytokine staining. At 2 weeks post-adoptive trans-
fer the anti-GITR group had almost twice as many donor
cells producing IFN? and triple the number producing
TNF? (Table 2B). Thus, the effect of anti-GITR therapy
correlated with improved production of antiviral cyto-
kines by the adoptively transferred CD8?T cells.
To our knowledge this is the first description of a virus
utilizing the regulatory T cell system to escape CD8?
T cell responses. It will be of interest to determine
whether this escape is an inadvertent consequence of
the infection or dependent on specific viral genes that
actively induce regulatory T cells. The generation of
adaptive regulatory T cells may be a normal process
that occurs to prevent immunopathological damage as
antigen levels decrease over the course of an infection
(MacDonald et al., 2002; Sakaguchi, 2003). Such activity
CD8?T Cell Dysfunction
McGuirk and Mills, 2002; Uraushihara et al., 2003), and
it appears that function rather than cell surface markers
distinguish regulatory T cells from helper T cells. One
function we associated with immunosuppression was
correlated with immunosuppression we do not believe
it is the sole effector in vivo. In preliminary experiments,
blockade of the IL-10 receptor was not sufficient to
restore function in CD8?T cells transferred into persis-
tently infected mice (data not shown). Thus, IL-10 may
be either a surrogate marker for immunosuppressive
T cells or just one of multiple factors required for the
The CD8?T cell dysfunction in persistent FV infection
is different from the progressive loss of effector func-
lymphocytic choriomeningitis virus (LCMV). In LCMV in-
fections, the CD8?T cells become “exhausted” due to
continuous stimulation, fail to develop into memory
cells, and lose their effector function over several weeks
of time (Onami et al., 2002). However, in the current FV
experiments, transferred naive CD8?T cells showed
dysfunction already by 1 week posttransfer, a time point
when antigen stimulation should be eliciting maximum
responsiveness rather than exhaustion. Our experi-
ments do not exclude a contributory role for exhaustion
in CD8?T cell dysfunction during persistent FV infec-
tions. In fact, the reason that anti-GITR therapy in the
absenseof CD8?Tcell transferwasineffective inreduc-
ing virus loads could be attributable to functional ex-
haustion of the endogenous population of virus-specific
CD8?T cells. Alternatively, the functional impairment of
CD8?T cells by regulatory T cells could be long-lived
or even irreversible. The adoptive transfers of CD8?
T cells into persistently infected mice provided us with
sufficient numbers of cells to study the effects of this very
interesting in vivo environment. Although other factors
may contribute to the immunosuppressive environment
in persistently infected mice, the current experiments
demonstrate that the presence of immunosuppressive
CD4?T cells is sufficient to account for dysfunction of
virus-specific CD8?T cells.
It is worth noting that it was recently shown that GITR
ligand (Kim et al., 2003) could act as a costimulatory
molecule for effector T cells (Tone et al., 2003). It is
unclear whether anti-GITR antibody works in the same
manner, but the possibility exists. Thus, costimulation
of the CD8?T cells could contribute to the therapeutic
effects of the anti-GITR antibody treatments. However,
it is very unlikely that costimulation of the CD8?T cells
could account for all of the increased CD8?T cell func-
tion in our experiments. First of all, preactivation of the
CD8?T cells with anti-CD3 antibody combined with
costimulation by anti-CD28 antibody prior to adoptive
transfer did not improve their activity (data not shown).
Second, lackof costimulation cannot explainwhy adop-
tive transfer of twice as many CD8?T cells overcame
demonstrate that activity by regulatory CD4?T cells can
inhibit CD8?T cell function. Thus, it appears that the
anti-GITR effect is predominantly on the regulatory
T cells but that costimulation of the CD8?T cells may
contribute to the observed effects.
It was unexpected that suppression of the CD8 re-
Figure 6. Reduction in Spleen Virus Loads after Anti-GITR Therapy
Combined with Adoptive Transfer of Virus-Specific TCR Tg CD8?
Each dot represents spleen virus loads from a single mouse. In
(A)–(D), B6.SJL (CD45.1 congenic) mice were treated as indicated.
The difference between the geometric means (log10) of groups (B)
and (D) is statistically significant (p ? .0027). B6.PL (Thy 1.1 con-
genic) mice were used in experiments (E)–(H). The difference be-
significant (p ? .0482). Statistical analyses were done by unpaired
t tests. A repeat of this experiment gave similar results with an
average decrease to less than 500 infectious centers per spleen
following combined treatment (data not shown).
by regulatory cells could lead to incomplete clearance
of pathogens and contribute to persistent infections
(Belkaid et al., 2001, 2002; MacDonald et al., 2002;
McGuirk et al., 2002). Alternatively, epitopes encoded
by viral genes may be specifically inducing regulatory
Tcells. Forexample, aviral epitopemight triggerregula-
tory T cells that normally function to maintain tolerance
to an endogenous retroviral protein. Endogenous ret-
roviral elements make up a significant proportion of the
mouse genome and a minimum of 8% of the human
genome (Hughes and Coffin, 2001; International Human
Genome Sequencing Consortium, 2001; Venter et al.,
The immunosuppressive CD4?T cells demonstrated
in this study may be the same cells we previously found
to be associated with depressed antitumor responses
in mice persistently infected with FV (Iwashiro et al.,
2001a). As yet we have not identified a single or set of
cell surface markers that conclusively distinguishes the
immunsuppressive subpopulation of CD4?T cells in
vivo. Treatments with anti-GITR, which had a clear bio-
logical effect, were not associated with detectable
changes in CD25 or other activation markers such as
CD69 and CD38 (data not shown). Our results indicate
that immunosuppressive CD4?T cells are not solely in
the CD25-positive subset (Figure 5D). CD25 negative
CD4?T cells with suppressive function have recently
been described in other studies (Bynoe et al., 2003;
on a C57BL/6 or B6.GFP background as indicated, and greater than
90% of the CD8?T cells contained a TCR specific for the DbGagL
FV epitope (Chen et al., 1996; Ohlen et al., 2002). All mice were
females of 12–24 weeks of age at the beginning of the experiments
and were treated in accordance with the regulations and guidelines
of the Animal Care and Use Committee of the Rocky Mountain
Laboratories and the National Institutes of Health.
sponse by regulatory T cells occurred at the level of
effector functionrather thanproliferation andactivation.
Given the massive proliferation of the adoptively trans-
ferred cells, it is quite remarkable that no effect on the
virus load was produced. In experiments where this
mainly at the level of proliferation (Ohlen et al., 2002).
However, there was likely functional impairment of the
CD8?T cells from the double transgenic mice as well,
since GagL-specific CD8?T cells in the periphery were
not cytotoxic when taken directly ex vivo and did not
cause pathological damage in the mice (Ohlen et al.,
2002). Peripheral tolerance may occur at different levels
on the same CD8?T cells depending on whether they
encounter antigen in the context of an inflammatory
infection or the context of a “self”-antigen during devel-
opment. It is quite interesting and revealing that it is
appear highly activated based on cell surface markers.
lation of CD43 are not necessarily coupled to IFN? pro-
duction or the ability to eliminate virus-infected cells.
There are indications in the literature that induction
of regulatory T cells may be more than an idiosyncratic
mechanism peculiar to Friend virus. MacDonald et al.
have reported a strong association between the estab-
lishment of chronic hepatitis C infections in humans and
the expansion of CD4?regulatory T cells (MacDonald
et al., 2002). It has also been shown that virus-specific
CD8?T cells in chronic HCV infections are functionally
impaired (Gruener et al., 2001). Thus, it is possible that
CD4?regulatory T cells may be causally involved in
the functional impairment of CD8?T cells specific for
hepatitis C virus. HIV infections are also associated with
dysfunctional CD8?T cells (Lieberman et al., 2001; Mi-
hoets et al., 1998), and it has been shown that HIV pa-
tients withdisease progession or activevirus replication
have increased frequencies of CD4?T cells producing
IL-10 (Ostrowski et al., 2001). Recently, it was also
trolled by CD4?regulatory T cells (Aandahl et al., 2004).
Thus, a link between CD4?regulatory T cells and CD8?
T cell dysfunction is feasible in chronic HIV infections
as well. A high viral setpoint during the asymptomatic
period of HIV infection is a strong predictor of fast pro-
gression to AIDS (Gandhi and Walker, 2002; Mellors et
al., 1995, 1996). Thus, a practical intervention that could
reduce virus loads during chronic HIV infection would
likely be an invaluable tool in postponing the onset of
AIDS. While it remains to be seen whether an interven-
tion such as described in this paper would work in HIV
infections, our experiments open up new possibilities of
elements of retroviral infections.
Virus and Virus Infection
The FV stock used in these experiment was an FV complex con-
taining B-tropic Friend murine leukemia helper virus (F-MuLV) and
Steeves, 1973). The stock was prepared as a 10% spleen cell ho-
mogenate from BALB/c mice infected 14 days previously with 3000
spleen focus-forming units of uncloned virus stock. For virus chal-
lenge experiments, mice were injected intravenously with 0.5 ml of
phosphate-buffered balanced salt solution containing 1500 spleen
focus-forming units of Friend virus complex. Persistently infected
mice were mice that had been infected at least 8 weeks prior to
experimentation. Virus levels are generally stable at approximately
104infectious centers per spleen by 6–8 weeks postinfection.
focus-forming virus (Lilly and
Intracellular Cytokine Staining and Flow Cytometry
Cell surface and intracellular cytokine staining was performed using
Becton Dickinson/Pharmingen reagents (except where noted) and
the Cytofix/Cytoperm intracellular cytokine staining kit. T cell anti-
bodies were: FITC-anti-CD43 (1B11), PE-anti-CD25 (PC61), PE-anti-
CD44 (IM7), APC- anti-CD4(RM4-5), and APC-anti-CD8(53-6.7).
analyses. During the 5 hr incubation with monensin (2 ?g/ml), cells
were incubated either with 1 micromolar DbGagL peptide (Chen et
al., 1996) to specifically stimulate TCR Tg CD8?T cells or with
plate-bound anti-CD3 and anti-CD28 (1 ?g/ml) to obtain a broad
perspective of the full range of T cell activity. Cells were washed
twice, incubated with anti- Fc? 2/3 receptor (2.4G2) to block Fc
receptors, and stained with APC-labeled anti-CD4 or anti-CD8 and
FITC-labeled anti-CD43 in round-bottom 96-well plates. The cells
were then washed, fixed, permeabilized, and incubated with PE-
conjugated monoclonal antibodies specific for IL-2 (CalTag, Burl-
ingame, CA), IL-4, IL-10, IFN?, or TNF?. (Becton Dickinson). Data
were acquired on a FACSCalibur flow cytometer (Becton Dickinson)
from CD4- or CD8-gated events. Analyses were done using BD
Cellquest Pro software (Version 4.0.1, Becton Dickinson).
Virus-Specific DNA- and RNA-PCR
One microgram template DNA was prepared from single-cell sus-
pensions of different mouse organs as indicated. F-MuLV env-spe-
cific proviral DNA was amplified using the upstream primer 5?-AAGT
CTCCCCCCGCCTCTA-3? and the downstream primer 5?-TGCCAG
CCATAGTTAGCC-3?. The PCR cycle profile was: 45 s 95?C, 45 s
55?C, and 45 s 72?C. Total RNA for PCR was extracted with Trizol
(Invitrogen) from 106cells of the different organs from the same
persistently infected mice that were used for DNA preparation.
mRNA was isolated from total RNA by immunomagnetic beads (Dy-
nal) specific for poly(A) sequences. Purified mRNA was then tran-
scribed into cDNA using Moloney murine leukemia virus reverse
transcriptase (Invitrogen). A 632 base pair-long F-MuLV env frag-
ment was amplified from the cDNA with the same PCR protocol
described above. Control samples subjected to PCR amplification
without prior reverse transcription did not yield PCR signals indicat-
ments the amplified fragments were directly sequenced to verify
the F-MuLV env specificity of the PCR products. Uninfected mice
were always negative by RNA and DNA PCR.
Sequencing of the FV GagL T Cell Epitope
Total RNA and DNA were extracted with Trizol (Invitrogen) from
spleencellsuspensions ofmicepersistentlyinfected withFV.mRNA
was isolated from total RNA and reverse transcribed into cDNA as
described above. Proviral DNA and the cDNA were then used to
amplify a 481 base pair-long fragment of the F-MuLV gag gene,
including the gag leader sequence. The primers were: sense primer,
5?-CGGCCAGAGTCCAACCATCC-3?; antisense primer, 3?-TCCCAG
Experimentswere donewith (C57BL/10x A.BY)F1mice (H-2b/b,Fv1b,
Fv2r/s), C57BL/6, the two B6 congenic strains B6.PL (Thy 1.1 con-
genic), and B6.SJL (CD45.1 congenics) as noted in the figure leg-
ends. The DbGagL TCR Tg (T cell receptor transgenic) mice were
CD8?T Cell Dysfunction
GTCACGATGTAGGGG-5?. The fragment from each sample was di-
rectly sequenced to determine the FV GagL epitope sequence and
its flanking sequences at the 3? and 5? ends. The primer used for
tide sequence was determined with a Prism 3100 (Applied Biosys-
tems) automated sequencer.
70 ?g of antibody was injected i.p. every other day beginning on
the day of adoptive transfer. The mice were injected a total of six
times. Control mice were treated by i.p. injection of nonspecific rat
immunoglobulin (Sigma, St. Louis, MO) on the same schedule as
the experimental groups. Mice were depleted of CD25?T cells by
i.p. injection of 1 mg of purified anti-CD25 monoclonal antibody
(Suvas et al., 2003) from clone PC61.5.3 from Accurate Chemical
and Scientific Corp. (Westbury, NY). Adoptive transfers were done
on day 3 following depletion when 0.2% or fewer CD25?cells re-
i.p. injection of anti-IL-10R (mAb IB1.3a, a gift from Robert Coffman,
DNAX, Palo Alto, CA) as previously described (Dittmer et al., 2002).
In the anti-IL-10R experiment there were no significant differences
anti-IL-10R, anti-IL-10R with 4 ? 106CD8?T cells, or only 4 ? 106
Serial dilutions of spleen cells from infected mice were plated onto
susceptible Mus dunni cells, cocultivated for 3 days, fixed with
oped with peroxidase-conjugated goat anti-mouse IgG and sub-
strate to detect foci of infected cells (Dittmer et al., 1998).
Tetramers and Tetramer Staining
For detection of Db-GagL-specific CD8?T cells, nucleated spleen
mingen) and PE-labeled MHC class I H2-Dbtetramers specific for
FV GagL peptide (Db-GagL tetramers) for 15 min at room tempera-
facility using a peptide in which all three cysteine residues were
replaced with amino-butyric acid to prevent interpeptide disulfide
bonding as has been described (Schepers et al., 2002). This variant
peptide is recognized by polyclonal GagL-specific CD8?T cells as
determined by intracellular interferon-? staining. Cells were washed
two times, resuspended in buffer with propidium iodide, and ana-
lyzed by flow cytometry.
This work was supported by the NIAID intramural research program
and by a grant to U.D. from the Deutsche Forschungsgemeinschaft
Received: September 20, 2003
Revised: January 7, 2004
Accepted: January 27, 2004
Published: March 16, 2004
An F-MuLV env-specific fluorogenic PCR probe (5?-[6FAM] ACTCC
fluorescein and TAMRA is 6-carboxy-tetramethyl-rhodamine) (Ap-
plied Biosystems, Foster City, CA) was used for real-time PCR. The
upstream and downstream primers were 5?-AAGTCTCCCCCCGCC
TCTA-3? and 5?-AGTGCCTGGTAAGCTCCCTGT-3?, respectively.
Real-time PCR and RT-PCR amplifications were performed in a 10
ul reaction mixture with either Taqman Universal PCR Master Mix
or Taqman One-Step RT-PCR Master Mix Reagants Kit (Applied
Biosystems), respectively, using an ABI 7900 sequence detector
system (Perkin-Elmer). Each sample was run in triplicate. The
housekeeping gene, GAPDH (Taqman Rodent GAPDH Control
Reagants, Applied Biosystems), was amplified from each sample to
normalize template concentrations. Comparisons between groups
was made using differences in critical threshold values.
Aandahl, E.M., Michaelsson, J., Moretto, W.J., Hecht, F.M., and
Nixon, D.F. (2004). Human CD4?CD25? regulatory T cells control
T cell responses to HIV and CMV Antigens. J. Virol. 78, 2454–2459.
Andersson, J., Kinloch, S., Sonnerborg, A., Nilsson, J., Fehniger,
T.E., Spetz, A.L., Behbahani, H., Goh, L.E., McDade, H., Gazzard, B.,
granules in lymphoid tissue during acute human immunodeficiency
virus type 1 infection. J. Infect. Dis. 185, 1355–1358.
Appay, V., Nixon, D.F., Donahoe, S.M., Gillespie, G.M., Dong, T.,
King, A., Ogg, G.S., Spiegel, H.M., Conlon, C., Spina, C.A., et al.
(2000). HIV-specific CD8(?) T cells produce antiviral cytokines but
are impaired in cytolytic function. J. Exp. Med. 192, 63–75.
Belkaid, Y., Hoffmann, K.F., Mendez, S., Kamhawi, S., Udey, M.C.,
Wynn, T.A., and Sacks, D.L. (2001). The role of interleukin (IL)-10 in
the persistence of Leishmania major in the skin after healing and
the therapeutic potential of anti-IL-10 receptor antibody for sterile
cure. J. Exp. Med. 194, 1497–1506.
Belkaid, Y., Piccirillo, C.A., Mendez, S., Shevach, E.M., and Sacks,
D.L. (2002). CD4?CD25? regulatory T cells control Leishmania ma-
jor persistence and immunity. Nature 420, 502–507.
Bluestone, J.A., and Abbas, A.K. (2003). Natural versus adaptive
regulatory T cells. Nat. Rev. Immunol. 3, 253–257.
Bynoe, M.S., Evans, J.T., Viret, C., and Janeway, C.A., Jr. (2003).
Epicutaneous immunization with autoantigenic peptides induces T
suppressor cells that prevent experimental allergic encephalomy-
elitis. Immunity 19, 317–328.
tion of a gag-encoded cytotoxic T-lymphocyte epitope from FBL-3
virus-induced tumors. J. Virol. 70, 7773–7782.
Dittmer, U., Brooks, D.M., and Hasenkrug, K.J. (1998). Characteriza-
tion of a live-attenuated retroviral vaccine demonstrates protection
via immune mechanisms. J. Virol. 72, 6554–6558.
Dittmer, U., Brooks, D.M., and Hasenkrug, K.J. (1999). Requirement
for multiple lymphocyte subsets in protection against retroviral in-
fection by a live attenuated vaccine. Nat. Med. 5, 189–193.
Dittmer, U., Race, B., Peterson, K.E., Stromnes, I.M., Messer, R.J.,
and Hasenkrug, K.J. (2002). Essential roles for CD8? T cells and
gamma interferon in protection of mice against retrovirus-induced
immunosuppression. J. Virol. 76, 450–454.
Enrichment of Lymphocyte Subsets
and Adoptive Cell Transfers
bach, Germany) was used for the purification of T lymphocyte sub-
sets aspreviously described (Dittmeret al., 1999).In brief,CD4?and
CD8?T cells were labeled with paramagnetic microbeads (Miltenyi
Biotech) and purified by immunomagnetic separation. The purity of
the cell fractions was monitored by flow cytometry and was greater
than 90% in all cases. All cell fractions contained less than 5% dead
cells after the separation procedure. Adoptive transfers were done
by i.v. injection of 0.5 ml of PBBS containing 15 U/ml heparin, and
4 ? 106or 8 ? 106purified CD8?T cells as indicated. For CD25?
subpopulation purifications, cells were labeled with anti-CD25-bio-
tin and purified using streptavidin microbeads. The cells were 87%
CD4 positive, and the CD4?T cells were 98% CD25 positive. To
obtain the CD25-negative population, the flowthrough from the
above experiment was purified using anti-CD4 microbeads yeilding
96% CD4?T cells with 0.2% CD25-positive cells. Recipient mice
received 8 ? 106TCR Tg CD8?T cells plus 2 ? 106CD4?T cells
from either the CD25-positive or -negative fractions.
Treatment of Mice with Antibodies
The antiglucocorticoid induced tumor necrosis factor receptor
(GITR)-producing hybridoma (DTA-1) (Shimizu et al., 2002) was
grown in RPMI medium containing 10% FCS. Supernatant was har-
vested, spun at 500 ? g for 10 min to remove cells and debris, and
frozenat?20?C. Supernatantswerepooled,quantitated, andstored
Gandhi, R.T., and Walker, B.D. (2002). Immunologic control of HIV-1.
Annu. Rev. Med. 53, 149–172.
Geertsma, M.F., van Wengen-Stevenhagen, A., van Dam, E.M., Ris-
berg, K., Kroon, F.P., Groeneveld, P.H., and Nibbering, P.H. (1999).
Decreased expression of zeta molecules by T lymphocytes is corre-
lated with disease progression in human immunodeficiency virus-
infected persons. J. Infect. Dis. 180, 649–658.
Gruener, N.H., Lechner, F., Jung, M.C., Diepolder, H., Gerlach, T.,
man, P. (2001). Sustained dysfunction of antiviral CD8? T lympho-
cytes after infection with hepatitis C virus. J. Virol. 75, 5550–5558.
Hasenkrug, K.J. (1999). Lymphocyte deficiencies increase suscepti-
bility to Friend virus-induced erythroleukemia in Fv-2 genetically
resistant mice. J. Virol. 73, 6468–6473.
Hasenkrug, K.J., Brooks, D.M., and Chesebro, B. (1995). Passive
immunotherapy for retroviral disease: influence of major histocom-
patibility complex type and T-cell responsiveness. Proc. Natl. Acad.
Sci. USA 92, 10492–10495.
Hasenkrug, K.J., Brooks, D.M., and Dittmer, U. (1998). Critical role
for CD4? T cells in controlling retrovirus replication and spread in
persistently infected mice. J. Virol. 72, 6559–6564.
Hughes, J.F., and Coffin, J.M. (2001). Evidence for genomic re-
arrangements mediated by human endogenous retroviruses during
primate evolution. Nat. Genet. 29, 487–489.
International Human Genome Sequencing Consortium (2001). Initial
sequencing and analysis of the human genome. Nature 409,
Iwashiro, M., Messer, R.J., Peterson, K.E., Stromnes, I.M., Sugie,
latory T cells induced by chronic retroviral infection. Proc. Natl.
Acad. Sci. USA 98, 9226–9230.
Iwashiro, M., Peterson, K., Messer, R.J., Stromnes, I.M., and Hasen-
krug, K.J. (2001b). CD4(?) T cells and gamma interferon in the long-
term control of persistent friend retrovirus infection. J. Virol. 75,
Johnson, W.E., and Desrosiers, R.C. (2002). Viral persistance: HIV’s
strategies of immune system evasion. Annu. Rev. Med. 53, 499–518.
Kim, J.D., Choi, B.K., Bae, J.S., Lee, U.H., Han, I.S., Lee, H.W., Youn,
B.S., Vinay, D.S., and Kwon, B.S. (2003). Cloning and characteriza-
tion of GITR ligand. Genes Immun. 4, 564–569.
Lieberman, J., Shankar, P., Manjunath, N., and Andersson, J. (2001).
Dressed to kill? A review of why antiviral CD8 T lymphocytes fail
to prevent progressive immunodeficiency in HIV-1 infection. Blood
Lilly, F., and Steeves, R.A. (1973). B-tropic Friend virus: a host-
range pseudotype of spleen focus-forming virus (SFFV). Virology
Lucas, M., Karrer, U., Lucas, A., and Klenerman, P. (2001). Viral
escape mechanisms—escapology taught by viruses. Int. J. Exp.
Pathol. 82, 269–286.
MacDonald, A.J., Duffy, M., Brady, M.T., McKiernan, S., Hall, W.,
Hegarty, J., Curry, M., and Mills, K.H. (2002). CD4 T helper type 1
and regulatory T cells induced against the same epitopes on the
core protein in hepatitis C virus-infected persons. J. Infect. Dis.
McGuirk, P., and Mills, K.H. (2002). Pathogen-specific regulatory
T cells provoke a shift in the Th1/Th2 paradigm in immunity to
infectious diseases. Trends Immunol. 23, 450–455.
McGuirk, P., McCann, C., and Mills, K.H. (2002). Pathogen-specific
T regulatory 1 cells induced in the respiratory tract by a bacterial
a novel strategy for evasion of protective T helper type 1 tesponses
by Bordetella pertussis. J. Exp. Med. 195, 221–231.
Mellors, J.W., Kingsley, L.A., Rinaldo, C.R., Jr., Todd, J.A., Hoo,
B.S., Kokka, R.P., and Gupta, P. (1995). Quantitation of HIV-1 RNA
in plasma predicts outcome after seroconversion. Ann. Intern. Med.
Mellors, J.W., Rinaldo, C.R., Jr., Gupta, P., White, R.M., Todd, J.A.,
and Kingsley, L.A. (1996). Prognosis in HIV-1 infection predicted by
the quantity of virus in plasma. Science 272, 1167–1170.
Migueles, S.A., Laborico, A.C., Shupert, W.L., Sabbaghian, M.S.,
Rabin,R., Hallahan,C.W., VanBaarle,D., Kostense,S., Miedema,F.,
Nat. Immunol. 3, 1061–1068.
Moskophidis, D., Lechner, F., Pircher, H., and Zinkernagel, R.M.
by exhaustion of antiviral cytotoxic effector T cells. Nature 362,
Ohlen, C., Kalos, M., Cheng, L.E., Shur, A.C., Hong, D.J., Carson,
B.D., Kokot, N.C., Lerner, C.G., Sather, B.D., Huseby, E.S., and
Greenberg, P.D. (2002). CD8(?) T cell tolerance to a tumor-associ-
ated antigen is maintained at the level of expansion rather than
effector function. J. Exp. Med. 195, 1407–1418.
cytes: molecular parameters and players. Virology 234, 179–185.
Onami, T.M., Harrington, L.E., Williams, M.A., Galvan, M., Larsen,
C.P., Pearson, T.C., Manjunath, N., Baum, L.G., Pearce, B.D., and
Ahmed, R. (2002). Dynamic regulation of T cell immunity by CD43.
J. Immunol. 168, 6022–6031.
Ostrowski, M.A., Gu, J.X., Kovacs, C., Freedman, J., Luscher, M.A.,
and MacDonald, K.S. (2001). Quantitative and qualitative assess-
ment of human immunodeficiency virus type 1 (HIV-1)-specific
ential disease progression: reciprocal interferon-gamma and in-
terleukin-10 responses. J. Infect. Dis. 184, 1268–1278.
Rich, R.F., and Green, W.R. (2000). Inhibition of antiviral CTL re-
sponses by virus-infected cells: line item veto (cells) revisited. Virol-
ogy 272, 237–243.
Robertson, M.N., Spangrude, G.J., Hasenkrug, K., Perry, L., Nishio,
J., Wehrly, K., and Chesebro, B. (1992). Role and specificity of T-cell
subsets in spontaneous recovery from Friend virus-induced leuke-
mia in mice. J. Virol. 66, 3271–3277.
Sakaguchi, S. (2003). Regulatory T cells: mediating compromises
between host and parasite. Nat. Immunol. 4, 10–11.
Schepers, K., Toebes,M., Sotthewes, G., Vyth-Dreese,F.A., Dellem-
ijn, T.A., Melief, C.J., Ossendorp, F., and Schumacher, T.N. (2002).
Differential kinetics of antigen-specific CD4? and CD8? T cell re-
sponses in the regression of retrovirus-induced sarcomas. J. Immu-
nol. 169, 3191–3199.
Shimizu, J., Yamazaki, S., Takahashi, T., Ishida, Y., and Sakaguchi,
S. (2002). Stimulation of CD25(?)CD4(?) regulatory T cells through
GITR breaks immunological self-tolerance. Nat. Immunol. 3,
R.J., Evans, L.H., Peterson, K.E., Race, B., and Hasenkrug, K.J.
(2002). Temporal effects of gamma interferon deficiency on the
course of Friend retrovirus infection in mice. J. Virol. 76, 2225–2232.
Suvas, S., Kumaraguru, U., Pack, C.D., Lee, S., and Rouse, B.T.
(2003). CD4?CD25? T cells regulate virus-specific primary and
memory CD8? T cell responses. J. Exp. Med. 198, 889–901.
Tone, M., Tone, Y., Adams, E., Yates, S.F., Frewin, M.R., Cobbold,
S.P., and Waldmann, H. (2003). Mouse glucocorticoid-induced tu-
Natl. Acad. Sci. USA 100, 15059–15064.
Trimble, L.A., and Lieberman, J. (1998). Circulating CD8 T lympho-
cytes in human immunodeficiency virus-infected individuals have
impaired function and downmodulate CD3 zeta, the signaling chain
of the T-cell receptor complex. Blood 91, 585–594.
Uraushihara, K., Kanai, T., Ko, K., Totsuka, T., Makita, S., Iiyama,
R., Nakamura, T., and Watanabe, M. (2003). Regulation of murine
J. Immunol. 171, 708–716.
Urbani, S., Boni, C., Missale, G., Elia, G., Cavallo, C., Massari, M.,
Raimondo, G., and Ferrari, C. (2002). Virus-specific CD8? lympho-
cytes share the same effector-memory phenotype but exhibit func-
CD8?T Cell Dysfunction Download full-text
tional differences in acute hepatitis B and C. J. Virol. 76, 12423–
Venter, J.C., Adams, M.D., Myers, E.W., Li, P.W., Mural, R.J., Sutton,
G.G., Smith, H.O., Yandell, M., Evans, C.A., Holt, R.A., et al. (2001).
The sequence of the human genome. Science 291, 1304–1351.
Vingerhoets, J., Bisalinkumi, E., Penne, G., Colebunders, R., Bos-
mans, E., Kestens, L., and Vanham, G. (1998). Altered receptor ex-
kines IL-2, IL-7 and IL-12 in HIV infection. Immunol. Lett. 61, 53–61.
Wedemeyer, H., He, X.S., Nascimbeni, M., Davis, A.R., Greenberg,
H.B., Hoofnagle, J.H., Liang, T.J., Alter, H., and Rehermann, B.
(2002). Impaired effector function of hepatitis C virus-specific CD8?
T cells in chronic hepatitis C virus infection. J. Immunol. 169, 3447–
Xu, X.N., Screaton, G.R., and McMichael, A.J. (2001). Virus infec-
tions: escape,resistance, and counterattack. Immunity15, 867–870.