| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Emory Vaccine Center and Department of Microbiology and Immunology and
Department of Surgery, Emory University School of Medicine, Atlanta, GA 30322; and
Department of Pathobiological Sciences, University of Wisconsin, Madison, WI 53706
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2- to 3-fold lower in
CD28-/- mice compared with +/+ mice; the lack of
CD28-mediated costimulation did not lead to preferential suppression of
CD8 T cell responses to the weaker subdominant epitopes. As seen in
CD28-/- mice, blockade of B7-mediated costimulation by
CTLA4-Ig treatment of +/+ mice also resulted in a 2-fold reduction in
the anti-LCMV CD8 T cell responses. Loss of CD28/B7 interactions
did not significantly affect the generation and maintenance of CD8 T
cell memory; the magnitude of CD8 T cell memory was 2-fold lower in
CD28-/- mice as compared with +/+ mice. Further, in
CD28-/- mice, LCMV-specific memory CD8 T cells showed
normal homeostatic proliferation in vivo and also conferred protective
immunity. Therefore, CD28 signaling is not necessary for the
proliferative renewal and maintenance of memory CD8 T
cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Many acute viral infections in humans and mice elicit potent CD8+ T cell responses that are instrumental in clearing virus from the tissues. The requirement for CD28/B7 interactions in eliciting anti-viral CD8+ CTL responses has been examined in several models (9, 10, 11, 12, 13, 14). Interactions between CD28 and B7 are obligatory to generate optimal CD8+ CTL responses following influenza and vesicular stomatitis virus (VSV)4 infections (10, 11). In contrast, CD28/B7 interactions are not essential to elicit primary CD8+ CTL responses following vaccinia virus (VV) and lymphocytic choriomeningitis virus (LCMV) infection (9, 10, 12, 13). CD8+ T cell responses during viral infections are directed against dominant and subdominant epitopes. It has been reported that CD28 deficiency did not affect the activation of CD8 T cells directed against the dominant CTL epitopes during LCMV infection in mice (9, 13). However, the role of CD28/B7 interactions in activation and expansion of subdominant vs dominant epitope-specific CD8 T cells is not known. Also, the requirement for CD28 signaling in the generation and maintenance of long-term T cell memory is unclear.
Infection of immunocompetent mice with the Armstrong strain of LCMV (LCMV-Arm) elicits a potent CD8 T cell response that eliminates virus within 8–10 days (15, 16, 17). The generation of primary anti-LCMV CD8 T cell response is not dependent either on the presence of CD4 T cells (15, 18) or B cells (16). The CD8 T cell response to LCMV is directed against well-characterized dominant and subdominant epitopes (19, 20). Furthermore, the techniques of intracellular cytokine staining and MHC class I tetramers allows precise quantitation of CD8+ T cells specific to multiple epitopes of LCMV within the same mouse (9, 13, 20). These features of the LCMV-specific T cell response allowed us to ask the following questions in this study: Is there a differential requirement for costimulation in LCMV-specific CD8 T cell responses against dominant and subdominant CTL epitopes? Is CD28/B7 interaction required for the generation and maintenance of T cell memory? What is the role of CD28/B7 interactions in regulating homeostatic proliferation of memory CD8 T cells? To address these questions, we compared LCMV-specific CD8 T cell responses between wild-type (+/+) and CD28-deficient (CD28-/-) mice. Our studies revealed that the generation of primary LCMV-specific CD8 T cell responses (against dominant and subdominant epitopes) does not require CD28 signaling. The generation and maintenance of CD8 T cell memory against both dominant and subdominant epitopes was unperturbed under conditions of CD28 deficiency. Also, loss of CD28/B7 interactions did not affect the homeostatic proliferation of LCMV-specific memory CD8 T cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
C57BL/6 (H-2b) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). The CD28-/- mice (C57BL/6 background) used in these experiments were created by targeted gene disruption, which abrogates surface expression of the CD28 (12). The C57BL/6 LCMV carrier colony was established and bred at Emory University (Atlanta, GA) as described previously (20). Spleen cell preparations from these mice contain LCMV-infected cells that present viral Ag with both MHC class I and II molecules, but lack LCMV-specific T cells.
Virus
LCMV-Arm was used in these studies for infection of mice (21). Mice were infected with LCMV-Arm by i.p. injection (2 x 105 PFU/mouse). To measure protective immunity, LCMV-immune mice were challenged with 102 PFU of LCMV-Arm (intracranial (i.c.) infection) or 2 x 106 PFU of LCMV-clone 13 (i.v. injection; 15, 16, 17, 21). Infectious virus in serum and tissues was quantitated by plaque assay on Vero cell monolayers as previously described (21).
CTL assay
MHC class I-restricted LCMV-specific CTL activity was determined by 51Cr release assay as previously described (21).
Determination of the LCMV-specific CTLp frequency
CTLp frequency was determined by a limiting dilution assay as described previously (15, 16, 17). Spleen cells from LCMV-immunized mice were cultured in graded doses in 96-well flat-bottom plates (12 wells per dose). Syngeneic feeder spleen cells (8 x 105) from uninfected mice and syngeneic stimulator spleen cells (2 x 105) from LCMV carrier mice were irradiated (1200 rad) and added to each well. Recombinant human IL-2 was purchased from BD PharMingen (La Jolla, CA) and was added at a final concentration of 50 U per ml. After 8 days, the contents from each well were split to test CTL activity against LCMV-infected and uninfected MC57 targets in a 6-h 51Cr-release assay.
Flow cytometry
The number of CD8 T cells in the spleen and lymph nodes was
determined by staining with specific mAbs followed by FACS analysis, as
previously described (16, 20). For FACS analysis,
PE-conjugated anti-CD8
(7), and
FITC-conjugated anti-mouse CD44 (IM7) were purchased from BD
PharMingen and were used at concentrations recommended by the
manufacturer.
Visualization of LCMV-specific CD8 T cells by MHC I tetramer staining
Construction of the MHC I Db tetramers that contain the LCMV CTL epitope peptides nucleoprotein (NP) 396–404 or gp33–41 has been described previously (20). Spleen cells were surface stained with either FITC- or PE-conjugated anti-CD8 (BD PharMingen) and fluorochrome-labeled MHC I tetramer for 1 h at 4°C. Spleen cells from uninfected mice were always stained in parallel with cells from infected mice as a negative control. To analyze CD28 expression on LCMV-specific memory CD8 T cells, single cell suspensions of splenocytes were stained with anti-CD8, anti-CD28 (BD PharMingen), and MHC I tetramers, followed by three-color flow cytometry.
Quantitation of T cell responses by intracellular staining
for IFN-
Intracellular staining for IFN- in CD8 T cells following
stimulation with the specific peptide was done as described previously
(20). Spleen cells were incubated in vitro for 5 h at
37°C in medium containing brefeldin A and recombinant human IL-2 (50
U/ml). Cells were either left unstimulated during the culture period or
were stimulated with MHC class I-restricted epitope peptides. After the
incubation period, the cells were surface stained for CD8 using
allophycocyanin- or PE-conjugated Abs and were then stained
intracellularly with anti-IFN--FITC using the Cytofix/Cytoperm
kit from BD PharMingen. Generally, detection of LCMV-specific CD8 T
cells by either intracellular cytokine staining or MHC class I
tetramers gives comparable results (20).
Quantitation of virus-specific IFN--secreting T cells
Virus-specific T cell responses were measured by ELISPOT assay
as described previously (20). The capture Ab,
anti-mouse IFN- (clone R4-6A2) and detection Ab, biotinylated
anti-mouse IFN- (clone XMG1.2) were purchased from BD
PharMingen. The ELISPOT plates were purchased from Millipore
(Bedford, MA). MHC I-restricted peptides LCMV
NP396–404,gp33–41, or gp276–286 were
used to stimulate CD8 T cells. Uninfected spleen cells contain
IFN--producing cells at a frequency of <2 per
106 cells with or without stimulation.
Quantitation of LCMV-specific CD8 T cells by intracellular cytokine
staining or ELISPOT gives comparable results (20).
5-Bromo-2'-deoxyuridine (BrdU) incorporation studies
To monitor proliferation of LCMV-specific memory CD8 T cells in vivo, LCMV-immune mice were given BrdU in drinking water (0.8 mg/ml) for 8 days. After the BrdU pulse, splenocytes were isolated and stained with anti-CD8 and anti-CD44 or fluorochrome-labeled MHC I tetramers. Following surface staining, cells were stained with anti-BrdU Abs, as described previously (20). Flow cytometry and data analysis were performed as described above. Spleen cells from mice that were not exposed to BrdU were always used as negative controls for BrdU staining.
CTLA4-Ig treatment in vivo
The CTLA4-Ig fusion protein used in this study was provided by
Bristol-Myers Squibb (New York, NY) and has been described elsewhere
(22, 23, 24). CTLA4-Ig is a fusion protein that contains the
extracellular domain of CTLA4 and is fused to the C region of human
Ig. The human Ig control Ab was purchased from BD PharMingen. CTLA4-Ig
was injected i.p. at a dose of 200 µg/mouse/injection on days 0, 2,
4, and 6 relative to infection with LCMV-Arm.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We determined the role of CD28/B7 interactions in generating
primary CD8 T cell responses by comparing the CD8 CTL responses between
+/+ and CD28-/- mice following an acute
infection with LCMV-Arm. On day 8 postinfection (PI), we quantitated
the direct ex vivo MHC class I-restricted LCMV-specific CD8
CTL-mediated cytotoxicity in the spleens of +/+ and
CD28-/- mice. Consistent with previous reports
(10, 12), the LCMV-specific cytotoxic activity in the
spleens of CD28-/- mice was comparable with
that of +/+ mice (Fig. 1
). These data
show that generation of LCMV-specific effector CD8 CTL response is not
dependent upon CD28/B7 interactions. CD28-/-
mice cleared LCMV by day 8 PI similar to +/+ mice (data not shown).
|
2-fold lower as compared with +/+ mice (Fig. 2a). Uninfected +/+ and
CD28-/- mice showed no differences in the total
number of naive or activated CD8 T cells (data not shown). We next
performed more precise analysis of the activation and expansion of
LCMV-specific CD8 T cells using MHC class I tetramers bearing the LCMV
NP CTL epitope peptide, NP396–404. On the day 8
PI, spleen cells from LCMV-infected +/+ and
CD28-/- mice were stained with anti-CD8 Abs
and fluorochrome-labeled MHC class I (Db)
tetramers followed by flow cytometric analysis. Representative FACS
profiles of staining for CD8+ tetramer-binding
cells are shown in Fig. 2b. As shown in Fig. 2b,
the relative proportions of NP396–404-specific
cells in the spleens of CD28-/- mice was lower
compared with +/+ mice. On day 8 PI, the spleens of +/+ and
CD28-/- mice had 12–18 x
106 and 6–9 x 106
NP396–404-specific CD8 T cells, respectively.
Although high numbers of NP396–404-specific CD8
T cells were generated in CD28-/- mice, the
overall magnitude was 2-fold lower compared with +/+ mice. Similar
results were obtained following staining with MHC class I tetramers
specific to the other LCMV immunodominant epitope in the viral
glycoprotein, gp33–41 (data not shown). The effect of B7-CD28 blockade
was also examined in +/+ mice treated with CTLA4-Ig. Treatment of +/+
mice with CTLA4-Ig blocks B7 interactions, and this reagent has been
shown to block alloantigen-induced T cell responses and certain
allograft reactions (22, 23, 24). The total number of
NP396–404-specific CD8 T cells was reduced in
+/+ mice treated with CTLA4-Ig to levels similar to
CD28-/- mice (Fig. 2c).
|
The MHC class I-restricted LCMV-specific CD8 T cell epitopes can
be divided into dominant and subdominant epitopes. In
H-2b mice, the dominant epitopes are
NP396–404 and gp33–41, and the subdominant
epitopes are gp276–286, NP205–212, and
gp92–101 (19, 20). It was of interest to determine the
requirement for CD28-mediated costimulation for optimal activation of
CD8 T cells, specific to dominant epitopes vs the weaker subdominant
epitopes. The epitope-specific CD8 T cell responses in +/+ and
CD28-/- mice were analyzed by intracellular
staining for IFN- on day 8 PI (Fig. 3). Upon LCMV-Arm infection,
CD28-/- mice mounted readily detectable CD8 T
cell responses to both dominant and subdominant epitopes. At the peak
of immune response, the total number of CD8 T cells specific to all the
epitopes was 2-fold lower in the CD28-/-
mice compared with +/+ mice. A previous study has indicated that CD8 T
cells specific to the immunodominant epitopes undergo cell division
15 times during the expansion phase of the anti-LCMV T cell
response (20). Therefore, a 2-fold reduction in the number
of LCMV-specific CD8 T cells in CD28-/- mice
may reflect one less cell division as compared with +/+ mice during the
proliferation phase of the CD8 T cell response. In summary, these data
suggested that following an acute LCMV infection, the activation of CD8
T cells specific to both dominant and subdominant epitopes is largely
independent of CD28-mediated costimulation. Furthermore, these data
show that the development of CD8 T cell effector functions, namely
cell-mediated cytotoxicity (Fig. 1
) and production of IFN- (Fig. 3),
did not require CD28 signaling.
|
Memory CD8 T cells in humans exhibit heterogeneity with respect to
CD28 expression; only a subpopulation of memory CD8 T cells in humans
express CD28 on their surface (25, 26, 27, 28, 29). Although it has
been shown that all murine T cells express CD28 constitutively
(3), the expression of CD28 on Ag-specific memory CD8 T
cells has not been studied. We examined the expression of CD28 on the
surface of LCMV-specific memory CD8 T cells by flow cytometry. As
illustrated in Fig. 4, LCMV-specific
memory CD8 T cells expressed readily detectable levels of CD28 on their
surface with little or no heterogeneity. Currently, it is not known
whether CD28-B7 interactions are necessary to maintain
CD8+ T cell memory. To examine the role of
CD28-B7 interactions in generating LCMV-specific CD8 T cell memory,
CD28-/- mice were immunized with LCMV-Arm, and
memory CD8 T cell responses were analyzed by IFN- ELISPOT assay at
various time points after infection. Fig. 5 shows the kinetics of LCMV-specific CD8
T cell response in +/+ and CD28-/- mice. As
shown in Fig. 5, the peak of LCMV-specific CD8 T cell response was
attained on day 8 PI, which was followed by a
contraction/down-regulatory phase (days 8–15 PI). During the
contraction phase of the T cell response, the magnitude of loss of
LCMV-specific CD8 T cells in the CD28-/- mice
was comparable with that of +/+ mice. Approximately 90% of the
LCMV-specific CD8+ T cells were lost
(20), presumably by apoptosis (30) during the
contraction phase in both +/+ and CD28-/- mice.
There is evidence that activation in the absence of CD28-mediated
costimulation may lead to T cell apoptosis in vitro (6).
However, the data shown in Fig. 5 suggested that in vivo, lack of
CD28-mediated costimulation did not lead to an exaggerated death phase
of LCMV-specific CD8 T cell response. The contraction phase of the CD8
T cell response is followed by the phase of memory, when a stable pool
of memory T cells are maintained indefinitely. Analysis of
LCMV-specific CD8 T cell responses revealed that the number of memory
cells was 2-fold lower in CD28-/- than in
+/+ mice. However, this did not change over time and the lower number
of CD8 memory T cells in CD28-/- mice most
likely reflected the 2- to 3-fold lower expansion seen in these mice
during the acute phase of the CD8 response (day 8). These data were
confirmed by quantitating the number of LCMV-specific CTL precursors in
the spleens of LCMV-immune +/+ and CD28-/- mice
by limiting dilution analysis (data not shown).
|
|
a, consistent with published
data, the spleens of LCMV-immune +/+ mice contained high frequencies of
memory CD8 T cells specific to the immunodominant epitopes,
NP396–404 and gp33–41 (20).
LCMV-specific memory CD8 T cells were also readily detected in the
spleens of LCMV-immune CD28-/- mice (Fig. 6a), and the frequencies were comparable with those in +/+
mice. Memory CD8 T cells specific to dominant and subdominant epitopes
were also quantitated in the spleens of LCMV-immune +/+ and
CD28-/- mice by staining for intracellular
IFN-. Data in Fig. 6b illustrate that irrespective of
epitope specificity, the percentages of memory CD8 T cells in the
spleens of LCMV-immune CD28-/- mice were
comparable with +/+ mice. Thus, B7-CD28 interaction is not necessary to
generate and maintain memory CD8 T cells following an acute LCMV
infection.
|
We examined the effect of CD28 deficiency on the activation
threshold of memory CD8 T cells in LCMV-immune mice. The activation
threshold of memory CD8 T cells (specific to both dominant and
subdominant epitopes) was compared between LCMV-immune +/+ and
CD28-/- mice by measuring IFN- production as
a function of the concentration of antigenic peptide. As shown in Fig. 7, in both +/+ and
CD28-/- mice, the number of IFN--producing
LCMV-specific memory CD8 T cells varied in a peptide dose-dependent
fashion. As shown in Fig. 7, memory CD8 T cells in
CD28-/- mice exhibited a slight difference in
the activation threshold at one peptide dilution for three of the four
epitopes as compared with +/+ mice; 0.0001 µg/ml for
NP396–404, 0.001 µg/ml for gp276–286, and
0.0001 µg/ml for NP205–211. Taken together, these data suggest that
lack of CD28/B7 interactions did not significantly affect the
sensitivity of LCMV-specific memory CD8 T cells to antigenic
stimulation in vitro. Further, these data show that memory CD8 T cells
generated under conditions of CD28 deficiency may not be qualitatively
different as compared with memory CD8 T cells that were generated in
the presence of CD28/B7 interactions.
|
It is well established that memory T cells undergo homeostatic
proliferation, which is believed to be an important mechanism promoting
survival by avoiding cell attrition over time. Studies have indicated
that cytokines IL-15 and IL-7 may be important for homeostatic
proliferation of memory T cells (31, 32, 33). Although
homeostatic proliferation of memory CD8 T cells is not dependent upon
TCR/MHC interactions (34), the role of costimulatory
molecules has not been examined. In this study, we determined the
requirement for CD28/B7 interactions in the homeostatic proliferation
of LCMV-specific memory CD8 T cells. At 60 days PI with LCMV, +/+ and
CD28 -/- mice were given BrdU in drinking water
for 8 days. At the end of the BrdU pulse, the percentage of
BrdU+ cells among LCMV-specific memory CD8 T
cells was determined by flow cytometry. As shown in Fig. 8, 23% and 20% of
CD8+CD44high T cells
(activated/memory) incorporated BrdU in +/+ and
CD28-/- mice, respectively. About 1.5–2% of
naive CD8 T cells (CD44low) in both +/+ and
CD28-/- mice incorporated BrdU over a period of
8 days (data not shown). Importantly, the percentages of
BrdU+ memory CD8 T cells specific to two
different LCMV CD8 CTL epitopes were comparable between +/+ and
CD28-/- mice. Further, blocking interactions
between B7 and CD28/CTLA-4 molecules using CTLA-4-Ig fusion proteins
did not affect the homeostatic proliferation of LCMV-specific memory
CD8 T cells (data not shown). Taken together, these data suggest that
CD28/B7 interactions are not obligatory for homeostatic proliferation
of memory CD8 T cells. Also, the proliferation rate of memory CD8 T
cells was not affected by antigenic specificity: within the same mouse,
memory CD8 T cells specific to two different epitopes had similar
proliferation rates in both +/+ and
CD28-/- mice.
|
T cell memory to LCMV infection is characterized by the ability of
memory CD8+ T cells to generate an accelerated
response upon re-exposure, leading to viral clearance more rapidly than
they do during the primary exposure (17, 20). We examined
protective immunity in CD28-/- mice using two
challenge models: 1) protection against a lethal i.c.
infection and 2) protection against a persistent LCMV
infection. Groups of +/+ and CD28-/- mice were
immunized by i.p. infection with LCMV-Arm. To evaluate protection
against lethal choriomeningitis, 35 days after immunization, these mice
were challenged with LCMV-Arm by i.c. injection. Naive +/+ and
CD28-/- mice were infected with LCMV-Arm (i.c.)
as controls. As shown in Fig. 9a, all of the naive +/+ and
CD28-/- mice succumbed to an i.c. LCMV-Arm
infection by 7–8 days postchallenge. In striking contrast, all of the
LCMV-immune +/+ and CD28-/- mice were
completely protected against a lethal LCMV challenge, and survived at
least up to 4 mo. The clone 13 strain of LCMV establishes persistent
infections in naive immunocompetent mice, which is characterized by low
CTL responses and disseminated infection of several tissues (15, 21). However, wild-type mice that have recovered from an acute
infection with LCMV-Arm are protected against a persistent infection
with LCMV-clone 13 (17). To examine protective immunity
against a persistent LCMV infection, +/+ and
CD28-/- mice that were previously immunized
with LCMV-Arm (100 days PI) were challenged with 2 x
106 PFU LCMV-clone 13 by i.v. injection. Naive
+/+ and CD28-/- were also infected with
LCMV-clone 13 for comparison. LCMV titer in the serum was determined 5
days after challenge with LCMV clone-13. Data in Fig. 9b
show that both +/+ and CD28-/- LCMV-immune mice
had undetectable levels of infectious virus in the serum; serum from
naive +/+ and CD28-/- mice infected with
LCMV-clone 13 contained high levels of infectious virus. In summary,
these data provide convincing evidence that memory CD8 T cells in
LCMV-immune CD28-/- mice confer protective
immunity in both peripheral (i.c. infection with LCMV-Arm) and systemic
(i.v. infection with LCMV-clone 13) challenge experiments.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous studies have documented that CD28-B7 interactions are not
essential for generating primary LCMV-specific CTL responses
(9, 10, 11, 12, 13). These studies primarily examined CD8 T cell
responses against dominant epitopes of LCMV. The present study confirms
and extends these findings. It is well established that CD8 T cell
responses during an acute LCMV infection are directed against dominant
and subdominant epitopes (19, 20). It is believed that the
strength of TCR signaling determines the requirement for costimulation
during T cell activation (3, 35). Dominant epitopes most
likely are presented at higher densities on the cell surface, thereby
delivering a potent signal through the TCR, obviating a need for
costimulation. In contrast, activation of T cells by weaker subdominant
epitopes may be costimulation dependent. To address this issue, we
compared the CD8 T cell responses to dominant and subdominant LCMV CTL
epitopes for their dependence on CD28-mediated costimulation.
Interestingly, CD28-/- mice generated high
numbers of CD8 T cells specific to both dominant and subdominant
epitopes, albeit 2-fold lower in magnitude compared with +/+ mice (Fig. 3). Furthermore, the hierarchy of immunodominance among LCMV CTL
epitopes was not altered in the absence of CD28-B7 interactions.
Although the mechanistic basis of LCMV CTL epitope hierarchy is not
known, these data suggested that CD8 T cell responses against dominant
and subdominant epitopes are largely independent of CD28-mediated
costimulation. These data are in agreement with the studies done by Tan
et al. (36), who blocked CD28/CTLA-4-B7 interactions by
treating mice with CTLA-4-Ig fusion proteins during an acute LCMV
infection. Similar to LCMV infection, CTL responses to VV are not
significantly altered by abrogation of CD28 signaling
(10). However, B7-CD28 interactions are essential for
generating effector CD8 CTL following VSV and influenza virus infection
in mice (10, 11). In contrast to LCMV and VV, which
replicate efficiently in the lymphoid system, the replication of
influenza virus and VSV is very limited in lymphoid tissues. Thus, one
common theme that has emerged from these studies is that the
CD28-mediated costimulation seems to be dispensable for CTL responses
against viruses that replicate to very high levels in the lymphoid
system. The rules defining the costimulatory requirements of CD8 T
cells may also depend on whether the infection is localized or
systemic, and the ability of virus to infect APC, particularly the
dendritic cells. Primary CD8 T cell responses to LCMV were normal in
the absence of CD40-CD40L interactions (9, 37, 38). In the
absence of 41BB-41BBL interactions, LCMV-specific CD8 T cell responses
were reduced by 2-fold (36). It remains to be
determined whether CD40-CD40L- and 41BB-41BBL-mediated costimulatory
interactions play redundant and/or compensatory roles in activating CD8
T cells under conditions of CD28 deficiency.
The resolution of an acute LCMV infection is dependent upon
CD8+ CTLs (39, 40) and does not
require CD4 T cells (15, 18). Nonetheless, mice acutely
infected with LCMV mount a strong CD4 T cell response and develop
humoral immunity (41, 42). In contrast to the development
of a potent CD8 T cell response, the induction of LCMV-specific CD4 T
cell response was compromised in CD28-/- mice
(Ref. 13 and M. Suresh, J. K. Whitmire, J. D. Altman, and
R. Ahmed, manuscript in preparation). It is likely that lack of
CD4 T cell activation in CD28-/- mice reduced
the CD8 T cell response by one-half compared with +/+ mice.
CD4-deficient mice also exhibit a similar phenotype: the magnitude of
LCMV-specific CD8 T cell response was 2-fold lower in comparison
with what was generated in +/+ mice.
In addition to providing costimulatory signals necessary for activation
of naive T cells, CD28-B7 interaction has also been shown to enhance
survival of activated T cells by inducing the expression of the
antiapoptotic gene Bcl-xL (6, 7). Furthermore, Bcl-xL induction
prevented Fas- and anti-CD3-induced apoptosis in activated T cells
(7, 8). These data suggested that CD28-mediated signaling
may be important in the survival of memory T cells. It was of interest
to examine the role of CD28-B7 interactions in the generation and
maintenance of LCMV-specific memory CD8 T cells. In this study, the
maintenance of LCMV-specific CD8 T cell memory was unaffected by lack
of CD28/B7 interactions (Figs. 5 and 6). The initial expansion of CD8 T
cells during the primary response (clonal burst size) has been shown to
be one of the determinants of the magnitude of T cell memory
(20). Therefore, a slight reduction in the total number of
LCMV-specific memory CD8 T cells in CD28-/-
mice most likely reflect a 2-fold lower expansion of virus-specific
CD8 T cells during the primary response.
Studies have shown that Ag dose, duration of TCR stimulation, and
number of TCRs engaged can determine the requirement for costimulation
during T cell activation. During activation of naive T cells,
CD28-mediated costimulation reduces the number of TCRs that need to be
triggered by the Ag (43). It was of interest to examine
the activation threshold of memory CD8 T cells in the absence of
CD28/B7 interactions. To this end, we compared the activation
thresholds of memory CD8 T cells specific to multiple epitopes between
+/+ and CD28-/- mice. These experiments
revealed that loss of CD28/B7 interactions did not significantly affect
the activation thresholds of memory CD8 T cells (Fig. 7). One
intriguing finding was that the activation threshold of memory CD8 T
cells specific to the subdominant epitope
NP205–212 was lower as compared with memory CD8
T cells specific to the dominant epitopes
NP396–404 and gp33–41 (even in the presence of
CD28/B7 interactions in +/+ mice). The increased sensitivity of
NP205–212-specific memory CD8 T cells to peptide
stimulation cannot be explained based on the differences in the binding
affinities of peptides to the MHC I molecule. This is because the
MHC-binding affinity of peptide NP396–404 is
substantially greater than for NP205–212
(19). One possibility is that during the primary response,
low-level presentation of NP205–212 subdominant
epitope by the APCs may selectively activate CD8 T cells with high
affinity. Alternatively, the repertoire of CD8 T cells that recognize
NP205–212 are inherently of high affinity and/or
avidity. Nevertheless, taken together, these data suggested that
CD28-deficient memory CD8 T cells may be qualitatively similar to +/+
memory CD8 T cells.
According to the current axiom, maintenance of memory T cells is dependent upon homeostatic proliferation, which prevents T cell attrition over time (31, 32, 33). We examined whether homeostatic proliferation of LCMV-specific memory CD8 T cells is affected in the absence of CD28-B7 interactions. In vivo BrdU labeling studies indicated that the rate of homeostatic proliferation of LCMV-specific memory CD8 T cells in LCMV-immune CD28-/- mice was comparable with that of immune +/+ mice. These data are consistent with normal maintenance of CD8 T cell memory in CD28-/- mice.
Protective immunity is a definitive marker of T cell memory. +/+ mice
that have recovered from an acute LCMV infection are protected against
lethal choriomeningitis resulting from an i.c. LCMV challenge. Akin to
+/+ mice, memory CD8 T cells in LCMV-immune
CD28-/- mice successfully protected against a
lethal i.c. infection with LCMV (Fig. 9a). LCMV-clone 13 is
a highly virulent strain of LCMV that establishes persistent infections
in immunocompetent mice (21). However, accelerated CD8 T
cell responses in LCMV-immune mice promptly controls LCMV-clone 13
infection, preventing viral persistence (17). LCMV-immune
CD28-/- mice were completely protected against
a persistent infection with LCMV-clone 13, presumably due to memory CD8
T cell-dependent accelerated viral clearance (Fig. 9b). In
summary, generation and maintenance of CD8 T cell memory, as assessed
by quantitation of virus-specific CD8 T cells by phenotypic and
functional assays and protective immunity, is not dependent on CD28/B7
interactions.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 M.S. and J.K.W. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Rafi Ahmed, Department of Microbiology and Immunology, Emory Vaccine Center, G211 Rollins Research Center, 1510 Clifton Road, Atlanta, GA 30322. E-mail address: ra{at}microbio.emory.edu
4 Abbreviations used in this paper: VSV, vesicular stomatitis virus; VV, vaccinia virus; LCMV, lymphocytic choriomeningitis virus; LCMV-Arm, the Armstrong CA 1371 strain of LCMV; BrdU, 5-bromo-2'-deoxyuridine; PI, postinfection; NP, nucleoprotein; i.c., intracranial.
Received for publication January 30, 2001. Accepted for publication September 4, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  Z.-Y. Wang, E. B. Olson Jr., D. E. Bjorling, G. S. Mitchell, and G. E. Bisgard Sustained hypoxia-induced proliferation of carotid body type I cells in rats J Appl Physiol, March 1, 2008; 104(3): 803 - 808. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Fuse, W. Zhang, and E. J. Usherwood Control of Memory CD8+ T Cell Differentiation by CD80/CD86-CD28 Costimulation and Restoration by IL-2 during the Recall Response J. Immunol., January 15, 2008; 180(2): 1148 - 1157. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. G. Thebeau, S. P. Vagvala, Y. M. Wong, and L. A. Morrison B7 Costimulation Molecules Expressed from the Herpes Simplex Virus 2 Genome Rescue Immune Induction in B7-Deficient Mice J. Virol., November 15, 2007; 81(22): 12200 - 12209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. B. Borowski, A. C. Boesteanu, Y. M. Mueller, C. Carafides, D. J. Topham, J. D. Altman, S. R. Jennings, and P. D. Katsikis Memory CD8+ T Cells Require CD28 Costimulation J. Immunol., November 15, 2007; 179(10): 6494 - 6503. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Tewari, Y. Nakayama, and M. Suresh Role of Direct Effects of IFN-{gamma} on T Cells in the Regulation of CD8 T Cell Homeostasis J. Immunol., August 15, 2007; 179(4): 2115 - 2125. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Valujskikh and X. C. Li Frontiers in Nephrology: T Cell Memory as a Barrier to Transplant Tolerance J. Am. Soc. Nephrol., August 1, 2007; 18(8): 2252 - 2261. [Full Text] [PDF]
|
|
|
|
 |
|
 H.-P. Raue and M. K. Slifka Pivotal Advance: CTLA-4+ T cells exhibit normal antiviral functions during acute viral infection J. Leukoc. Biol., May 1, 2007; 81(5): 1165 - 1175. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. P. Ndejembi, J. R. Teijaro, D. S. Patke, A. W. Bingaman, M. R. Chandok, A. Azimzadeh, S. G. Nadler, and D. L. Farber Control of Memory CD4 T Cell Recall by the CD28/B7 Costimulatory Pathway J. Immunol., December 1, 2006; 177(11): 7698 - 7706. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Fang and L. J. Sigal Direct CD28 Costimulation Is Required for CD8+ T Cell-Mediated Resistance to an Acute Viral Disease in a Natural Host J. Immunol., December 1, 2006; 177(11): 8027 - 8036. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Prlic, G. Hernandez-Hoyos, and M. J. Bevan Duration of the initial TCR stimulus controls the magnitude but not functionality of the CD8+ T cell response J. Exp. Med., September 4, 2006; 203(9): 2135 - 2143. [Abstract] [Full Text] [PDF]
|
|
