| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Neurology,
Molecular Microbiology and Immunology, and
Pathology, Keck School of Medicine, University of Southern California, Los Angeles, CA 90033; and
§
Departments of Neurology and Medical Microbiology, University of Wisconsin and William S. Middleton Veterans Hospital, Madison, WI 53792
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-secreting CD8+ T cells were
recovered from the brains of asymptomatic mice compared with mice
undergoing demyelination, suggesting that IFN- ameliorates rather
than perpetuates JHM strain of mouse hepatitis virus-induced
demyelination. The present data thus indicate that in immunocompetent
mice, effector CD8+ T cells control infection without
mediating either clinical disease or demyelination. In contrast,
demyelination correlated with early and sustained infection of the
spinal cord. Rapid viral spread, attributed to determinants within the
spike protein and possibly perpetuated by suboptimal CD8+ T
cell effector function, thus ultimately leads to the process of
immune-mediated demyelination. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cellular as well as humoral immunity contribute to viral control at multiple, discrete functional levels, indicating the necessity of an orchestrated immune response in controlling JHMV infection (1, 2). The major effectors eliminating infectious virus from the CNS are CD8+ T cells (9, 10), of which up to 50% can be virus specific within the CNS (11). The potent local effector function of this population is evidenced by Ag-specific ex vivo lytic activity during acute infection (11, 12, 13) and delayed viral clearance in perforin-deficient mice (14). However, CD8+ T cells require CD4+ T cells for optimal function and survival (15). Furthermore, virus-specific Ab control the recurrence of infectious virus during persistent CNS infection (16).
The multicomponent nature of immune protection from fatal JHMV disease (1, 2) has kept the determinants involved in demyelination elusive and controversial (1, 3, 4, 17, 18, 19). Although JHMV-induced demyelination was initially attributed to direct viral cytopathic effects (20), immune components have been implicated by little or no demyelination in severely immunosuppressed mice (3, 4, 5). Thus, infected irradiated or SCID mice lacking functional T and B cells exhibit neither virus clearance nor demyelination. Restoration of demyelination by adoptive transfer of JHMV immune splenocytes implicated T lymphocytes as the responsible mediators (5, 21). Subsequent studies revealed that infected nude mice having intact B cells but lacking functional T cells also develop demyelinating disease (3); however, immune cells from nude mice were incapable of restoring demyelination in SCID recipients. This dichotomy was confounded even more by analysis of JHMV infection of mice deficient in class I or class II expression, which revealed a bimodal distribution between demyelination-positive and -negative mice (3). Similarly, studies of selectively T cell-deficient mice infected with the more hepatotropic A59 strain of mouse hepatitis virus (MHV), questioned the role of CD4+ and CD8+ T cells as effectors of demyelination (17, 18). Lastly, depletion of blood-borne macrophages failed to prevent JHMV-induced demyelination despite blocking demyelination in other experimental models (19). It is also unclear whether persistently infected cells contribute to demyelination directly via disruption of cellular functions or indirectly via chronic immune activation.
JHMV-induced pathogenesis is directly controlled by properties of the viral spike (S) protein. The S protein determines tropism, mediates cell fusion, and elicits both B and T cell responses (1, 2, 22). Parental JHMV has tropisms for both neurons and glial cells, causing a uniformly fatal encephalitis associated with demyelination (2, 23). By contrast, mAb-selected JHMV variants, characterized by point mutations and deletions in the S protein, infect predominantly glial cells of the brain and spinal cord white matter (24, 25). These attenuated mutants are generally associated with reduced lethality, but increased subacute demyelination. Efforts to correlate S gene mutations and neurological disease suggested that at least two regions, one comprising the hypervariable domain (nt 1500–1950) (24, 26) and the other surrounding nt 3340, are critical for pathogenesis (27). Although alteration of either site alone limits infection to glial cells, causing extensive demyelination, variants with mutations in both sites display an asymptomatic phenotype (25, 27). This was interpreted to reflect limited distribution in the white matter concomitant with early containment by the immune system. The inability to attribute demyelination to any single cell population implies complex interactions between the virus and immune components, which may be confounded by compensatory immune functions in selectively immune-deficient mice used to identify distinct effector populations.
To assess the relative contribution of immune vs viral factors in JHMV-induced chronic pathology, immunocompetent mice were infected with two JHMV variants with similar growth characteristics, but vastly different subacute diseases (25). In C57BL/6 mice variant 2.2-V-1 (designated V-1 for brevity) is associated with subacute paralysis and extensive demyelination (25, 27). By contrast, the variant derived from V-1 (2.2/7.2-V-2; designated V-2) is nonpathogenic, causing little or no encephalitis or demyelination (25, 27). A potential contribution to the asymptomatic phenotype of V-2 in infected C57BL/6 mice is a deletion in the S protein that spans the immunodominant S510 CTL epitope (28, 29). Infected BALB/c mice, capable of mounting a CD8+ T cell response to the dominant nucleocapsid (N) epitope shared between both virus variants (30), were therefore chosen to compare pathogenesis. The data indicate that CD8+ T cells recruited to the brain by acute infection are not sufficient to initiate or to sustain demyelination in immunocompetent mice. Rather, concomitant with delayed local immune effector function, pathology correlated with the relative kinetics of viral spread to the spinal cords, presumably mediated by determinants in the S protein.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Male BALB/c (H-2d) mice were purchased from the National Cancer Institute (Frederick, MD) at 6 wk of age and were certified naive to prior MHV exposure. Mice were housed in an accredited animal facility at the University of Southern California and were infected within 1 wk of arrival. Except where noted otherwise, CNS infections were induced by intracranial (i.c.) injection of 33 µl containing 1000 PFU of JHMV mAb-selected variants 2.2-V-1 and 2.2/7.2-V-2, designated V-1 and V-2, respectively, for brevity (25). V-1 was derived from parental JHMV by in vitro selection with a neutralizing S protein-specific mAb, J.2.2 (25). V-2 was selected from V-1 using a second neutralizing mAb, J.7.2, specific for a different portion of the S protein (25). V-1 contains a single point mutation at nt 3340 of the S gene, whereas V-2 has a 100-base deletion (nt 1523–1624) in addition to the point mutation (27). Viruses were propagated and quantitated by plaque assay using the murine DBT astrocytoma cell line as previously described (23). Clinical disease was graded as previously described (16, 31, 32): 0, healthy; 1, ruffled fur and hunched back; 2, hind limb weakness, inability to right itself if turned on back; and 3, paralysis.
Determination of virus titers
Virus titers were determined from brain homogenates as described previously (23). One half-brain from three individual mice per time point was washed once in Dulbecco’s PBS (D-PBS; pH 7.4) and homogenized using a Tenbroeck tissue homogenizer (Wheaton Science Products, Millville, NJ) in a volume of 4.0 ml of D-PBS. Homogenates were centrifuged for 7 min at 800 x g at 4°C. Supernatants were stored at -70°C until assayed for infectious virus by plaque assay using DBT cells as previously described (23).
Histology
Brain and spinal cord tissues were fixed in Clark’s solution (75% ethanol and 25% glacial acetic acid) for 3 h, embedded in paraffin, and stained with hematoxylin and eosin. Viral Ag was identified by incubating sections with the N protein-specific mAb J.3.3 (33, 34), followed by immunoperoxidase staining (Vectastain-ABC kit, Vector, Burlingame, CA).
Isolation of CNS-derived CD8+ T cells
Mononuclear cells were isolated from the CNS of JHMV-infected mice as described previously (11, 35). Briefly, brains were removed, washed twice in RPMI 1640 supplemented with 25 mM HEPES (pH 7.2) and 1% FCS, and minced on ice before homogenization in Tenbroeck tissue homogenizers. Cells were suspended in 30% Percoll (Pharmacia, Piscataway, NJ) and concentrated onto a 1-ml cushion of 70% Percoll by centrifugation at 1300 x g for 20 min at 4°C. Mononuclear cells were collected from the interphase and washed twice before analysis.
FACS analysis
Expression of cell surface markers was determined by staining
cells with mAb specific for CD8 (53.67), CD4 (GK1.5), CD44 (IM7), CD62L
(MEL-14), and CD69 (H1.2F3). All mAb were purchased from PharMingen
(San Diego, CA). The Ld MHC class I tetramer
associated with pN318–326 peptide (N-peptide)
has been described previously (11). Nonspecific binding
was minimized by blocking with rat anti-mouse FcIII/IIR mAb
(2.4G2, PharMingen) and naive mouse serum (1 µl of each per
106 cells stained) in PBS containing 0.5% BSA
for 10 min before staining. Cells were stained for 1 h on ice,
washed twice in PBS with 0.5% BSA, and analyzed by flow cytometry on a
FACStar (Becton Dickinson, Mountain View, CA).
CTL assays and peptides
Brain-derived CD8+ T cells were analyzed
for lytic activity directly ex vivo. Splenocytes from infected animals
were stimulated in vitro with 1 µM N-peptide in RPMI 1640 medium
supplemented with 2 mM glutamine, 25 µg/ml gentamicin, 1 mM sodium
pyruvate, 5 x 10-5 M 2-ME, nonessential
amino acids (RPMI complete), 10% FCS, and 5% rat Con A supernatant
for 6 days before determination of lytic activity. CTL assays were
performed as previously described (30, 36). Briefly, BC10
ME (H-2d) target cells labeled with
Na51CrO4 (New England
Nuclear, Boston, MA) were preincubated with peptide at a concentration
of 100 nM before addition of CTL. 51Cr release
was determined in 100 µl of supernatant after 4 h of incubation,
and specific lysis was defined as 100 x [(experimental
release - spontaneous release)/(detergent release -
spontaneous release)]. Maximum spontaneous release values were <25%
of the total detergent release values in all experiments. The 9-mer
N-peptide (APTAGAFFF), purchased from Chiron Mimotopes (Clayton,
Australia), was solubilized at 
1 mM in DMSO and diluted in PBS as
previously described (30).
IFN- ELISPOT assays
ELISPOT assays to measure the frequency of Ag-specific
IFN--secreting cells were conducted as described previously
(11, 35). Briefly, 96-well plates supporting cellulose
ester membranes (MultiScreen HA, Millipore, Bedford, MA) were coated
with 10 µg/ml R4-6A2 mAb (PharMingen). Nonspecific binding was
blocked with RPMI 1640 medium containing 10% FCS. Mononuclear cells
were resuspended in Iscove’s DMEM supplemented with 2 mM glutamine, 25
µg/ml gentamicin, 1 mM sodium pyruvate, 5 x
10-5 M 2-ME, nonessential amino acids (Iscove’s
complete), and 10% FCS. Serial 3.3-fold dilutions of CNS and splenic
mononuclear cells were plated in triplicate and stimulated with
irradiated (25 Gy) splenocytes from naive mice (4 x
105/well) in the presence or the absence of 1
µM N-peptide. EL-4 supernatant was added as a source of IL-2 to a
final 2.5% concentration, and cultures were incubated for 40 h at
37°C. Bound IFN- was detected by 8-h incubation at 4°C with
biotinylated anti-IFN- mAb (5 µg/ml; XMG1.2, PharMingen),
followed by consecutive incubations with streptavidin/peroxidase
(Sigma, St. Louis, MO) and diaminobenzidene as a substrate (Sigma).
Spots from two mononuclear cell dilutions (n = 6) were
counted.
Statistical analysis
Results are represented as the mean ± the SD or SEM and were analyzed using paired Student’s t test where indicated.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The differential disease outcomes caused by V-1 compared with V-2
infection in C57BL/6 (H-2b) mice could not be
attributed to altered virus growth (25). The pathology and
immune components controlling V-1 infection are well characterized
(14, 16, 32); however, the immune components clearing V-2
infection are largely unknown (37). Specifically, it is
unclear how the absence of the immunodominant S510 CTL epitope, present
within V-1, but absent from V-2, affects the vastly different
pathologies in infected H-2b mice
(27). To confirm viral growth characteristics and
pathogenesis in the presence of CD8+ T cells
specific for the conserved immunodominant N epitope (30),
BALB/c mice (H-2d) were infected with an equal
dose of each variant and monitored for clinical symptoms and viral
titers (Fig. 1
). Clinical symptoms in
V-1-infected mice lasted from 7 to 15 days postinfection (p.i.), with
varying degrees of recovery observed from day 15 onward (Fig. 1A). V-1-infected mice began to exhibit symptoms of clinical
relapse between days 25 and 30, with increasing signs of encephalitis,
weakness in the hind limbs, and weight loss. Mice infected with lower
doses of V-1 generally recover from acute clinical symptoms (16, 25, 32, 38). Therefore, relapses of clinical symptoms were
attributed to the higher inoculation dose. However, neither infectious
virus nor increased viral Ag could be detected during disease relapse,
suggesting that recurrent symptoms were not associated with increased
viral replication (data not shown). By contrast, mice infected with the
identical inoculum of V-2 developed barely detectable clinical symptoms
and remained essentially asymptomatic throughout the course of
infection (Fig. 1A).
|
B). Although infectious V-1
was still detected in one mouse on day 15 p.i., virus titers from
most V-1- and all V-2-infected mice were reduced below the limit of
detection by day 15 p.i. Overall, slightly increased and prolonged
V-1 replication was considered insufficient to cause highly divergent
disease severity, thus confirming the results observed in infected
C57BL/6 mice (25). These data suggest that differences in
the rate of acute virus replication are not sufficient to account for
differences in virus-induced pathogenesis. Both paralytic and nonparalytic variants of JHMV induce potent CD8+ T cell responses
Due to their prominent role in reducing infectious virus via both
lytic and cytokine-mediated mechanisms (14, 32), CNS
infiltrating CD8+ T cells were considered likely
candidates for initiating tissue destruction and immunopathology
(39). BALB/c mice infected with parental JHMV or V-1 mount
a dominant CD8+ T cell response to the N protein
epitope (30, 35). To initially ensure
CD8+ T cell immunogenicity of this conserved
epitope during V-2 infection, peripheral N epitope-specific
CD8+ T cells were enumerated by IFN- ELISPOT.
V-1 and V-2 infection elicited similar frequencies of N-responsive
CD8+ T cells, amounting to 1/2000 splenocytes
(Table I). Furthermore, in vitro
stimulated splenocytes from V-2-infected mice lysed target cells with
similar efficiency compared with splenocytes from V-1-infected mice
(Table I). Similar levels of expansion and function of peripheral
CD8+ T cells following either infection thus
confirmed the presence of an immunogenic N epitope within V-2 and
demonstrated that differences in pathogenesis were not due to the
inability of V-2 infection to induce a CTL response.
|
). During infection with both viruses
peak T cell infiltration occurred on day 7 p.i., when
CD8+ T cells accounted for 30–40% of
brain-derived mononuclear cells (Fig. 2A). The increased
percentage of CD8+ T cells within the CNS of
V-1-infected mice coincided with a higher percentage of
tetramer+ cells within the
CD8+ T cell population (37 vs 24%). On day
14 p.i. the percentage of tetramer+ cells
within the CD8+ T cell population increased to
40–42% during either infection. However, after day 14 p.i.
slightly higher percentages of CD8+ T cells were
retained within the CNS of V-2- compared with V-1-infected mice (Fig. 2A). Although total CD8+ T cells
declined following the clearance of both infectious viruses,
tetramer+ CD8+ T cells were
retained to at least day 35 p.i. Whereas the
tetramer+ cells in the CD8+
T cell population declined to 23% by day 35 following V-1 infection,
they remained relatively stable (between 37 and 42%) following
resolution of V-2 infection.
|
B. During both infections the
percentage and total number of CD8+ T cells were
maximal on day 7 p.i., declined most sharply between days 7 and
14 p.i., and dropped more slowly thereafter (Fig. 2B,
two left panels). The pattern of
tetramer+ T cells residing within the CNS during
both infections was more distinct (Fig. 2B, two right
panels). Although tetramer+
CD8+ T cells were maximal on day 7 p.i.
during both infections, the relative percentage and absolute numbers of
tetramer+ CD8+ T cells were
twice as high in V-1-infected mice compared with V-2-infected mice.
During V-1 infection the percentage and total number of
tetramer+ CD8+ T cells
declined sharply between days 7 and 14 p.i. and continued to
decline thereafter. By contrast, the decline of
tetramer+ CD8+ T cells
during V-2 clearance was less pronounced between days 7 and 14
p.i., and absolute numbers of tetramer+
CD8+ T cells exceeded those maintained during
persistent V-1 infection from day 14 p.i. onward. By day 35
p.i. absolute numbers of tetramer+
CD8+ T cells were at least 2-fold higher in the
brains of V-2 compared with V-1-infected mice, thereby presenting an
inverse ratio compared with that on day 7 p.i. The increased number of V-2-induced CD8+ T cells retained in the brain in the absence of clinical disease suggested that CD8+ T cells do not play a direct role in pathogenesis. As maintenance of JHMV-specific CD8+ T cells in the parenchyma is supported by CD4+ T cells (15), preferential recruitment or retention of CD4+ T cells during V-2 infection was assessed. Similar to CD8+ T cells, infiltration by CD4+ T cells was maximal on day 7 p.i., constituting 18–19% of CNS-derived mononuclear cells (data not shown). The percentages of CD4+ T cells gradually declined, but were similar at each time point throughout either infection, with the exception of day 14 p.i., when the brains of V-2-infected mice contained a slightly increased percentage (19 vs 13%). However, these appeared to be insufficient to account for the differences observed in retention of Ag-specific CD8+ T cells.
CD8+ T cells retained within the CNS following acute viral infection may comprise populations expressing different functional capabilities (11, 35, 40). To determine whether differential activation states of retained CD8+ T cells influence the suppression of infectious virus and pathogenesis, CD8+ T cells were analyzed for surface expression of the activation/memory markers CD44 and CD62L (41) and the very early activation marker CD69 (11, 40). The majority of CNS-derived tetramer+ CD8+ T were CD44high (92–99%) and CD62Llow (95–99%) throughout the course of both infections, confirming previous analysis of CNS-infiltrating CD8+ T cells (11). Although staining with mAb to CD69 on day 7 p.i. revealed increased numbers of CD69+ cells in the tetramer+ CD8+ population in V-1- compared with V-2-infected animals (57 vs 45%; data not shown), CD69 expression by tetramer+ CD8+ T cells at all later times was comparable between the two infections, peaking at 89–93% on day 14 p.i. A similar trend was observed in tetramer- CD8+ T cells; although this population demonstrated overall lower CD69 expression, it increased with time from 20 to 70% from days 7–35 p.i. (data not shown). These data suggest slightly enhanced activation of CD8+ T cells at the site of higher virus replication during the peak of CNS infiltration following V-1 infection. However, similar populations of CD69-expressing cells after day 14 p.i. indicate that CD8+ T cell regulation is independent of disease status.
CD8+ T cell lytic activity does not account for differential pathogenesis
Cytolytic activity is down-regulated during the transition from
acute to persistent V-1 infection (11). To determine
whether differences in CD8+ T cell effector
functions are involved in V-1-induced disease, the functional activity
of brain-derived mononuclear cells was tested. On day 7 p.i. V-1-
and V-2-induced CNS-infiltrating CD8+ T cells
both exhibited similar levels of Ag specific lytic activity, which was
severely reduced by day 14 p.i. and was undetectable at later
times p.i. in both groups (Fig. 3).
Adjustment of total E:T cell ratios to E:T cells ratios based on
tetramer+ CD8+ T cells
revealed that the loss in CD8+ T cell lytic
activity may only partially be attributed to the decline in Ag-specific
cells. This suggests that differential regulation of
CD8+ T cell lytic function within the CNS, either
early or late, does not account for differences in viral clearance or
pathogenesis.
|
secretion
Clearance of infectious JHMV from the CNS is dependent upon both
cytolytic function and IFN- secretion (14, 32).
Perforin-mediated cytolysis eliminates virus from infected microglia
and astrocytes, whereas IFN- reduces virus in oligodendrocytes, the
dominant cellular reservoir for persistent V-1 infection (14, 32). The capacity of CNS-derived CD8+ T
cells to secrete IFN- in response to N-peptide stimulation was
therefore compared by ELISPOT during the course of both infections
(Fig. 4). The frequency of CNS-derived T
cells capable of secreting IFN- was similar among both infected
groups during acute infection on day 7 p.i. and following
reduction of infectious virus on day 14 p.i. By contrast, at days
21 and 35 p.i. V-2-infected animals contained almost twice as many
N-responsive T cells within the brain as V-1-infected mice.
IFN--secreting cells were most abundant on day 14 p.i.
following V-1 infection and on day 21 p.i. following V-2
infection, indicating a delay in the optimal IFN-. Remarkably, these
data also indicate an overall delay between peak frequencies of
CD8+ T cells assessed by IFN- secretion vs
peak frequencies assessed by lytic activity and tetramer staining on
day 7 p.i. (Figs. 2 and 3) during both infections. The fact that
frequencies of IFN--secreting CD8+ T cells
were equivalent during initial clearance of either virus suggests that
the absence of clinical symptoms and demyelination following V-2
infection was not due to increased IFN--dependent virus clearance
from oligodendrocytes. Furthermore, the higher frequency of
IFN--secreting T cells within the CNS of asymptomatic V-2-infected
mice on days 21 and 35 p.i. (Fig. 4) also suggests that IFN-
does not play a key role in chronic, ongoing demyelination observed in
V-1-infected mice.
|
The above data demonstrate that neither lytic potential nor
IFN- secretion by CD8+ T cells correlates with
disease. To assess whether altered tropism or viral spread contributes
to disease severity, brains and spinal cords of V-1- and V-2-infected
mice were examined histologically. On day 7 p.i. viral Ag in the
brains of V-1-infected mice was primarily detected in white matter
glia, with rare infection of neurons (Fig. 5A), confirming the biased
tropism of V-1 for white matter (25). Similar to V-1
infection, V-2 viral Ag in the brain also localized mainly to glia,
with few neurons staining positive on day 7 p.i. (Fig. 5B). Following viral clearance, viral Ag began to decline
after day 21 p.i. (data not shown), but was still detectable on
day 35 p.i. in the brains of both V-1- and V-2-infected mice (Fig. 5, C and D). These data implied no significant
differences in the distribution of viral Ag in the brains during either
acute or subacute infection. Spread of both viruses to the spinal cord
was apparent by day 3 p.i., but was initially restricted to
ependymal cells lining the central canal (data not shown). Although Ag
was evident in the both the gray and the white matter of the spinal
cord of V-1-infected mice by day 5 p.i., (Fig. 5E), it
was not detected outside of ependymal cells until day 7 p.i. in
V-2-infected mice (Fig. 5, F and H). Furthermore,
on day 7 p.i. Ag-positive cells in the spinal cord of V-2-infected
mice appeared reduced in frequency compared with those in the spinal
cords from V-1-infected mice and were more prevalent in the gray
matter, with rare infection of the white matter (Fig. 5H).
These data indicated delayed centripetal dissemination of V-2 through
the spinal cord. Consistent with less effective infection of spinal
cords by V-2, Ag was only detected to day 15 p.i. (data not
shown), but not on days 21 or 35 p.i. (Fig. 5J). By
contrast, spinal cords of V-1-infected mice still exhibited multiple
Ag-positive cells in the spinal cord white matter on day 35 p.i.
(Fig. 5I). Although no significant differences in Ag
distribution were evident in the brains during either the acute phase
or following clearance of infectious virus, the viruses appeared to
differ significantly in their ability to initiate persistent infections
of spinal cord.
|
).
V-1 infection induced extensive infiltration by mononuclear cells in
both brain and spinal cord by day 7 p.i. (Fig. 6, A and
E), supporting peak infiltration determined by flow
cytometry (see above). Infiltrating mononuclear cells responding to V-1
infection were frequently found in regions undergoing demyelination
(Fig. 6E). Following V-2 infection focal areas of
demyelination remained extremely rare, and most mononuclear cells
remained localized to perivascular regions despite similar levels of
mononuclear cell infiltration within both the brain and spinal cords
compared with V-1 infection (Fig. 6, B and F).
Although levels of persisting viral Ag were similar in brains from V-1-
and V-2-infected mice on day 35 p.i., chronic V-1 infection
coincided with increasingly numerous and severe foci of demyelination
in both brain and spinal cord (Fig. 6, C and G).
These foci were marked by the presence of infiltrating mononuclear
cells. In contrast to the distribution of mononuclear cell infiltrates
throughout the parenchyma in V-1-infected mice, mononuclear cells
retained within the brains of V-2-infected mice remained localized to
perivascular regions (Fig. 6D). This localized distribution
was consistent with the virtual absence of focal areas of demyelination
(Fig. 6D). In further contrast to V-1 infection, a gradual
reduction in the number of mononuclear cell infiltrates from day
21 p.i. onward, with barely detectable mononuclear cells by day
35 p.i., were noted in the spinal cords of mice infected with V-2
(Fig. 6H). Furthermore, no demyelinating lesions in spinal
cord were observed at any time point up to day 35 p.i. in
V-2-infected BALB/c mice (Fig. 6H), consistent with findings
in V-2-infected C57BL/6 mice (25). These data suggested
that V-2 infection results in a clinically unapparent infection despite
the presence of immune infiltrates similar to V-1 infection, including
a potent CD8+ T cell response. Furthermore, in
contrast to early and prolonged infection of the spinal cord by V-1,
transient infection of the spinal cord by V-2, associated with
mononuclear cell infiltration, is not sufficient to induce either acute
or chronic demyelination. Early replication in the spinal cord combined
with apparently inefficient or delayed immune control, therefore,
appear to be the major determinants leading to V-1 pathogenesis.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Similar peak virus titers in BALB/c mice confirmed that asymptomatic infection by V-2 could not be attributed to diminished V-2 replication, as demonstrated by infected C57BL/6 mice (25). However, both increased early replication and delayed clearance of V-1 suggested less efficient immune control due to more rapid viral spread within the CNS and/or inferior induction of antiviral effector functions within the CD8+ T cell population. Nearly identical levels of virus-specific ex vivo lytic activity by brain-derived mononuclear cells during both acute infections indicated that pathogenesis did not result from enhanced cytolytic activity. Consistent with the peak of cytolytic function, CNS infiltration by CD8+ T cells was maximal at early times following either infection. It can be argued that initially higher absolute numbers of virus-specific CD8+ T cells in the brains of V-1-infected mice may result in more expansive tissue destruction. However, V-2 infection using a 10-fold higher inoculum still did not induce clinical disease or demyelination, despite enhancing virus-specific CD8+ T cell infiltration to a level equivalent to that induced by V-1 infection (data not shown). These data indicate that neither differential numbers nor cytolytic function of CD8+ T cells during acute infection accounted for the dramatic differences in viral pathogenesis.
Virus-specific CD8+ T cells within the brain
began to decline as either infectious virus was cleared, consistent
with previous analysis of V-1-infected BALB/c and
CB6F1 mice (11, 35). A more dramatic
decline in V-1-induced compared with V-2-induced
CD8+ T cells during days 7–35 p.i. implied a
higher turnover of CD8+ T cells during V-1
infection. However, there was no evidence for differential
regulation of effector function during the course of either infection.
Both virus-specific CD8+ T cell populations
isolated from the brain lost ex vivo lytic function after 14 days p.i.,
retained their ability to secrete IFN- and expressed nearly
identical phenotypic surface markers. These data excluded the idea that
chronically activated CD8+ T cells, which were
found within the brains of both V-1- and V-2-infected groups, are
responsible for ongoing demyelination characteristic of V-1
infection.
IFN- secretion comprises a crucial function of
CD8+ T cells for controlling JHMV infection in
oligodendroglia (32), but has also been implicated in
perpetuating demyelination (4, 47, 48, 49). Both the
frequencies and the absolute numbers of Ag-specific IFN--secreting
CD8+ T cells recruited into the CNS were similar
during the first 2 wk following either V-1 or V-2 infection. In
addition, at least 2-fold higher frequencies were found at later time
points in the brains of V-2-infected mice despite similar viral Ag load
compared with V-1-infected brains. These data imply that IFN-
secretion by CD8+ T cells does not contribute to
pathogenesis, but, rather, enhances protection. Similar frequencies of
IFN--secreting cells relative to tetramer+
cells further indicate that the differential disease outcomes are not
due to preferential induction of T cell anergy and/or exhaustion, as
observed following sustained T cell exposure to high Ag levels
(50, 51, 52). TNF-
, another potent inflammatory mediator,
released not only by activated astrocytes and microglia (53, 54), but also by activated CD8+ T cells,
has been implicated in CNS pathologies (54). However,
similar magnitudes of CD8+ T cell effector
function during both infections indirectly confirm previous findings
that TNF- is unlikely to play a role in the differences in the
pathogenesis of V-1 and V-2 (55). Similarly, in light of
the strong type 1 response in the CD8 compartment and comparable
recruitment of CD4+ T cells, it is unlikely that
a dominant type 2 cytokine response plays a role in inhibiting
pathogenesis in V-2-infected mice.
In addition to a significant number of tetramer+ CD8+ T cells, a large population of tetramer- CD8+ T cells of unknown specificity remain in the brain for at least 1 mo following infection with either virus. The ratio of tetramer- to tetramer+ CD8+ T cells within the brain was fairly consistent throughout the course of JHMV infection, similar to results from CB6F1 mice (11). The absence of CNS pathology following V-2 infection therefore suggests that these tetramer- CD8+ T cells are unlikely to act as autoimmune or active bystander components in the ongoing demyelinating process (4).
The present results appear to exclude Ag-specific cytotoxicity, nonspecific bystander effects, and initiation of autoimmunity by CD8+ T cells as major determinants of JHMV-induced immunopathology. A nonpathogenic role of CD8+ T cells in this disease is supported by the presence of demyelination in mice infected with the related A59 strain of MHV, which were either physically or genetically depleted of functional CD8+ T cells before infection (17, 18). However, as MHC class II-restricted cytolytic CD4+ T cells are induced following infection with MHV strain A59 (56), these cells may compensate for CD8+ T cells in initiating pathogenic processes. Nevertheless, ongoing demyelination in mice deficient in perforin during both V-1 and Theiler’s murine encephalomyelitis virus infection also suggest that CD8+ T cell-mediated cytotoxicity does not contribute directly to CNS pathology (14, 57), although it may enhance neurological deficits (57). The increasing evidence for a dominantly protective role of CD8+ T cells during JHMV infection (9) contrasts with their image as predominant mediators of CNS pathology during lymphocytic choriomeningitis, Semliki Forest and Borna virus infections of rodents, and human T cell leukemia virus-1 (HTLV-1) infection in humans (39, 58, 59, 60, 61, 62). Although the evidence is overwhelming that immune components generally play a prominent role in virus-induced demyelination and neurologic disease, discrete contributions made by individual immune and/or viral components may vary significantly between different infection models (48, 57, 58, 62).
Interestingly, a major difference between V-1 and V-2 infection was the differential detection of Ag-positive cells within the spinal cord despite similar levels in the brains of both V-1- and V-2-infected mice. The early detection of Ag-positive cells in nonependymal cells of the spinal cord of V-1-infected mice before extensive immune infiltration is consistent with previously published data (63). However, whereas Ag was only detected transiently in the spinal cord of V-2-infected mice, predominantly in the gray matter, vastly higher numbers of Ag-positive cells persisted in the spinal cord white matter of V-1-infected mice. This suggests that sequences within the S protein may regulate the ability to cross the gray matter barrier surrounding the central canal of the spinal cord (63), thereby contributing to the establishment of persistent infection of the spinal cord white matter (25, 63). Another distinguishing characteristic of V-2 infection was the perivascular location of cell infiltrates throughout the course of infection. This implies that differential chemokine, cytokine, or protease activities induced by either infection may play a critical role in homing patterns of mononuclear cells and thus pathogenic outcome. Nevertheless, viral replication of both V-1 and V-2 is effectively controlled within the brain. However, the more rapid spread of V-1 throughout the spinal cord may allow the virus to surpass a threshold of infection preventing immune-mediated clearance, while delayed dissemination of V-2 throughout the spinal cord allows effective clearance. In summary, these data demonstrate for the first time that the vastly different disease outcomes induced by two related JHMV variants are the result not of a differential overall immune response within the CNS, but, rather, of a rapid and persistent establishment of infection within the spinal cord. By swaying the balance between replication and immune control, viral factors determining cellular tropism and spread within the CNS thus control these processes, ultimately leading to both clinical symptoms and immune-mediated demyelination.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Cornelia C. Bergmann, 1333 San Pablo Street, MCH 142, Los Angeles, CA 90033.
3 Abbreviations used in this paper: JHMV, JHM strain of MHV; MHV, mouse hepatitis virus; S protein, spike protein; nt, nucleotides; V-1, JHMV variant 2.2-V-1; V-2, JHMV variant 2.2/7.2-V-2; N, nucleocapsid protein; i.c. intracranial; D-PBS, Dulbecco’s-PBS; DBT, delayed brain tumor; N-peptide, JHMV pN318–326 peptide; ELISPOT, enzyme-linked immunospot; p.i., postinfection.
Received for publication October 18, 1999. Accepted for publication February 8, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
is required for viral clearance from central nervous system oligodendroglia. J. Immunol. 162:1641. in progression to chronic demyelination and neurological deficit following acute EAE. Mol. Cell. Neurosci. 12:376.[Medline]
. Nat. Med. 3:1037.[Medline]
transgenic and knockout models of CNS inflammation and degeneration. J. Neuroimmunol. 72:137.[Medline]
This article has been cited by other articles:
|
|
 |
|
  K. B. Walsh, L. L. Lanier, and T. E. Lane NKG2D Receptor Signaling Enhances Cytolytic Activity by Virus-Specific CD8+ T Cells: Evidence for a Protective Role in Virus-Induced Encephalitis J. Virol., March 15, 2008; 82(6): 3031 - 3044. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. D. C. Ireland, S. A. Stohlman, D. R. Hinton, R. Atkinson, and C. C. Bergmann Type I Interferons Are Essential in Controlling Neurotropic Coronavirus Infection Irrespective of Functional CD8 T Cells J. Virol., January 1, 2008; 82(1): 300 - 310. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Gonzalez, C. C. Bergmann, C. Ramakrishna, D. R. Hinton, R. Atkinson, J. Hoskin, W. B. Macklin, and S. A. Stohlman Inhibition of Interferon-{gamma} Signaling in Oligodendroglia Delays Coronavirus Clearance without Altering Demyelination Am. J. Pathol., March 1, 2006; 168(3): 796 - 804. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. van Lint, L. Kleinert, S. R. M. Clarke, A. Stock, W. R. Heath, and F. R. Carbone Latent Infection with Herpes Simplex Virus Is Associated with Ongoing CD8+ T-Cell Stimulation by Parenchymal Cells within Sensory Ganglia J. Virol., December 1, 2005; 79(23): 14843 - 14851. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. C. MacNamara, M. M. Chua, J. J. Phillips, and S. R. Weiss Contributions of the Viral Genetic Background and a Single Amino Acid Substitution in an Immunodominant CD8+ T-Cell Epitope to Murine Coronavirus Neurovirulence J. Virol., July 15, 2005; 79(14): 9108 - 9118. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. C. MacNamara, M. M. Chua, P. T. Nelson, H. Shen, and S. R. Weiss Increased Epitope-Specific CD8+ T Cells Prevent Murine Coronavirus Spread to the Spinal Cord and Subsequent Demyelination J. Virol., March 15, 2005; 79(6): 3370 - 3381. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Ramakrishna, S. A. Stohlman, R. A. Atkinson, D. R. Hinton, and C. C. Bergmann Differential Regulation of Primary and Secondary CD8+ T Cells in the Central Nervous System J. Immunol., November 15, 2004; 173(10): 6265 - 6273. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Moniuszko, C. Brown, R. Pal, E. Tryniszewska, W.-P. Tsai, V. M. Hirsch, and G. Franchini High Frequency of Virus-Specific CD8+ T Cells in the Central Nervous System of Macaques Chronically Infected with Simian Immunodeficiency Virus SIVmac251 J. Virol., November 15, 2003; 77(22): 12346 - 12351. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. W. Marten, S. A. Stohlman, J. Zhou, and C. C. Bergmann Kinetics of Virus-Specific CD8+-T-Cell Expansion and Trafficking following Central Nervous System Infection J. Virol., February 15, 2003; 77(4): 2775 - 2778. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. G. Glass and T. E. Lane Functional Expression of Chemokine Receptor CCR5 on CD4+ T Cells during Virus-Induced Central Nervous System Disease J. Virol., December 6, 2002; 77(1): 191 - 198. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Zhou, S. A. Stohlman, R. Atkinson, D. R. Hinton, and N. W. Marten Matrix Metalloproteinase Expression Correlates with Virulence following Neurotropic Mouse Hepatitis Virus Infection J. Virol., June 27, 2002; 76(15): 7374 - 7384. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-I. Tschen, C. C. Bergmann, C. Ramakrishna, S. Morales, R. Atkinson, and S. A. Stohlman Recruitment Kinetics and Composition of Antibody-Secreting Cells Within the Central Nervous System Following Viral Encephalomyelitis J. Immunol., March 15, 2002; 168(6): 2922 - 2929. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Ramakrishna, S. A. Stohlman, R. D. Atkinson, M. J. Shlomchik, and C. C. Bergmann Mechanisms of Central Nervous System Viral Persistence: the Critical Role of Antibody and B Cells J. Immunol., February 1, 2002; 168(3): 1204 - 1211. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Eo, U. Kumaraguru, and B. T. Rouse Plasmid DNA Encoding CCR7 Ligands Compensate for Dysfunctional CD8+ T Cell Responses by Effects on Dendritic Cells J. Immunol., October 1, 2001; 167(7): 3592 - 3599. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Kumaraguru, I. A. Davis, S. Deshpande, S. S. Tevethia, and B. T. Rouse Lymphotoxin {{alpha}}-/- Mice Develop Functionally Impaired CD8+ T Cell Responses and Fail to Contain Virus Infection of the Central Nervous System J. Immunol., January 15, 2001; 166(2): 1066 - 1074. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. W. Marten, S. A. Stohlman, and C. C. Bergmann Role of Viral Persistence in Retaining CD8+ T Cells within the Central Nervous System J. Virol., September 1, 2000; 74(17): 7903 - 7910. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
0.002 for
differences between scores for all V-1- and V-2-infected animals.
B, Virus titers were determined by plaque assay, and
each time point represents the average titer from three infected mice.
The dashed line in B represents the limit of detection
for infectious virus, and error bars represent the SD.
