| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Microbiology and Immunology, Loyola University Chicago Medical Center, Maywood, IL 60153
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
is not
involved in LCMV-induced wasting disease and show that IFN-
contributes to the disease. Consistent with a role for IFN- in
wasting, we find that IFN- is necessary for LCMV-specific CD4 T cell
responses in the CNS, most likely because it is required to induce MHC
class II expression. Our data also indicate that IL-1 is required for
LCMV-induced wasting and that IL-6 contributes to the wasting disease.
Additionally, our results identify -melanocyte-stimulating hormone
as a potential mediator of the disease. Overall, this work defines the
critical role of virus-primed CD4 T cells and of proinflammatory
cytokines in the pathogenesis of wasting disease induced by LCMV
infection. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2-microglobulin (2m)
(14, 16, 17), CD8 (17), or perforin
(15), or of wild-type mice depleted of CD8 T cells
(16) leads to wasting disease. This wasting disease does
not occur after i.p. or i.v. infection with LCMV (18) (C.
Kamperschroer, unpublished observations), suggesting that wasting is a
pathology of the CNS. The wasting disease is not an effect of LCMV
replication, but rather results from the immune response against the
virus (17), and requires CD4 T cells (13, 14, 16, 17). However, the mechanism whereby the antiviral immune
response leads to wasting disease is unknown.
In this study, we investigated which components of the immune response
against LCMV contribute to the wasting disease. CD4 T cells are
required for wasting (13, 14, 16, 17), but it is not clear
whether CD4 T cells of any specificity allow for wasting or whether
wasting disease is caused by CD4 T cells specific for LCMV. We
therefore tested whether CD4 T cells activated by LCMV infection are
sufficient to induce wasting disease. We also assessed the role of
cytokines in the disease. A number of proinflammatory cytokines, such
as IFN-, TNF-, IL-1, and IL-6, are involved in wasting syndromes
(1, 2, 3). Infection of mice with LCMV generates T cells that
produce IFN- (19), and both IL-1 (18) and
IL-6 (20) are expressed in the CNS during persistent LCMV
infection. In these studies, we addressed the hypothesis that
LCMV-induced wasting disease results from the action of proinflammatory
cytokines produced in response to the virus.
Our studies suggest that virus-specific CD4 T cells are critical for
LCMV-induced wasting disease. We also show that the proinflammatory
cytokines IFN-, IL-1, and IL-6 contribute to the wasting disease,
whereas TNF- does not. Our data further identify an
appetite-suppressing factor, -melanocyte-stimulating hormone
(-MSH), as a potential mediator of wasting. This work provides
insight into how immune responses against pathogens can cause severe
weight loss, a debilitating or even fatal symptom of a number of
disease states.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The Armstrong-3 strain of LCMV was propagated in BHK-21 cells (American Type Culture Collection, Manassas, VA), and viral titers of infectious supernatants were determined by plaque assay on Vero cell monolayers. Virus was diluted in serum-free DMEM to obtain appropriate titers for use in experiments.
Treatment of mice
B6.129P2-B2mtm1Unc
(2m-/-)
(21), B6.129S2-Cd4tm1Mak
(CD4-/-) (22),
B6.129S7-Ifngtm1Ts
(IFN--/-) (23),
B6;129S-Tnftm1Gkl
(TNF--/-) (24),
B6.129-Tnfrsf1atm1Mak
(TNFR-1-/-) (25),
B6.129-Tnfrsf1btm1Mwm
(TNFR-2-/-) (26),
B6.129S6-Il6tm1Kopf
(IL-6-/-) (27), or
B6;129S-Il1r1tm1Roml
(IL-1R-/-) (28) mice were
obtained from The Jackson Laboratory (Bar Harbor, ME). C57BL/6 (B6)
mice were obtained from Taconic (Germantown, NY).
B6.129-B2mtm1 or
B6.SJL-Ptprca/BoAiTac-B2mtm1
(2m-/-)
(29) mice obtained from Taconic were used for some
experiments. Strains of
2m-/- mice from either
vendor gave similar results. Mice were maintained in specific
pathogen-free, American Association for the Accreditation of Laboratory
Animal Care-accredited facilities at Loyola University Medical Center
on a 12-h light-dark cycle with continuous access to food and water.
Mice were infected via the i.p. route with 4 x
105 PFU LCMV in a volume of 0.4 ml. For i.c.
infection, mice were anesthetized by inhalation of Halothane
(Sigma-Aldrich, St. Louis, MO) or by i.p. injection of Avertin (2, 2,
2-tribromoethanol), and were then injected i.c. with 2 x
104 PFU LCMV in a volume of 20 µl.
Intracranial infection of wild-type mice with LCMV causes lethal meningitis due to CD8 CTL-mediated killing of brain cells (30, 31). To prevent this lethal meningitis in experiments with IL-1R-/- or IL-6-/- and B6 control mice, CD8 T cells were removed in vivo by injecting animals i.p. with the CD8-depleting Ab 2.43 (32) on days -2, +2, and +7 relative to infection, and at weekly intervals thereafter. The effective dose of 2.43 was determined before experiments, and depletion of CD8 T cells was periodically confirmed on individual mice during experiments by flow cytometry.
Peptide synthesis
Peptides corresponding to aa 61–80 of the LCMV glycoprotein (gp61) or 309–328 of the LCMV nucleoprotein (np309) were synthesized by the Loyola University Macromolecular Synthesis Facility using a Synergy 432A automated synthesizer. The gp61 and np309 are restricted by I-Ab (33).
Cell preparations
Spleen cell suspensions were prepared as previously described (19). Mononuclear cells were isolated from brain tissue over a Percoll gradient using a modification of described methods (34). Mice were bled by cardiac puncture and perfused with PBS. Brains were disrupted in complete medium (HEPES-buffered RPMI 1640 (Mediatech, Herndon, VA) containing 10% (v/v) FCS (Atlanta Biologicals, Norcross, GA), 2 mM L-glutamine, and 50 µM 2-ME) by passage through a sterile nylon mesh (100 µm pore size; BD Biosciences, San Jose, CA), followed by sequential passages through 19- and 23-gauge needles (BD Biosciences). Cell suspensions were centrifuged at 700 x g for 7 min at 4°C, and pellets were resuspended in a Percoll solution comprised of a 10:1 mixture of Percoll (Sigma-Aldrich) and 10x RPMI 1640 (Sigma-Aldrich). Cell suspensions were then overlaid with 60%, 40%, and 0% dilutions of this Percoll solution. Following centrifugation at 1000 x g for 15 min at 4°C, mononuclear cells were harvested from the 40–60% interface. Cells were washed in complete medium before use in experiments.
Flow cytometry
Aliquots containing 1 x 106 spleen
cells in ice-cold PBS supplemented with 1% (v/v) FCS and 15 mM sodium
azide (FACS buffer) were incubated with flourochrome-conjugated
anti-CD4 (clone RM4-5; BD PharMingen, San Diego, CA) or
anti-CD8 (clone 53-6.7; BD PharMingen) Abs at a concentration of 1
µg/106 cells, for 30 min at 4°C. Samples were
washed with FACS buffer and were analyzed using a FACSCalibur flow
cytometer (BD Biosciences). Dead cells were excluded on the basis of
forward and side scatter, and data analyses were performed on 
10,000
acquired events using the CellQuest software (BD Biosciences).
ELISPOT and intracellular cytokine staining to enumerate LCMV-reactive cells
ELISPOT assays to detect cells producing IFN- in response to
antigenic peptides from LCMV were performed, as described previously
(19), using cells isolated from spleen or brain, as
described above.
Intracellular cytokines were detected, as previously described
(19), using the Cytofix/Cytoperm intracellular staining
kit (BD PharMingen), according to the manufacturer’s instructions.
Briefly, test cells were stimulated for 5 h with antigenic
peptides in the presence of brefeldin A (Sigma-Aldrich) and IL-2
(Sigma-Aldrich) in wells of 96-well tissue culture plates. Where
indicated, we also added 1 x 106 spleen
cells from naive CD4-/- mice as APC to each
well during the stimulation period. Cell surface CD4 was stained using
the Ab RM4-5. Cells were fixed and permeabilized, then Abs XMG1.2,
MP6-XT22, and MP5-20F3 were used to detect IFN-, TNF-, and IL-6,
respectively. Isotype-matched Abs were used as controls. All Abs for
intracellular cytokine staining were purchased from BD PharMingen.
Adoptive transfer
Donor spleen cells were obtained from mice 10 days after i.p.
infection with LCMV, except where indicated in figure legends. In these
cases, donor cells were obtained 9 days after i.p. infection and were
cultured overnight at 5 x 106 cells/ml
along with 10 U/ml human rIL-2 (Sigma-Aldrich). In our experience, this
treatment enhances cell survival during culture. To prevent transfer of
lethal meningitis, CD8 T cells were removed from donor populations
before cell transfer by in vitro treatment with the anti-CD8 Ab
31M-6 (32) and baby rabbit complement (Cedarlane
Laboratories, Hornby, Ontario, Canada), as previously described
(19). Similarly treated control donor cells were prepared
from noninfected mice. Where indicated, donor spleen cells were
additionally separated into CD4+ and
CD4- fractions using anti-CD4-coated
magnetic beads before transfer, according to instructions supplied by
the manufacturer (Miltenyi Biotec, Auburn, CA). Alternatively, CD4
cells were sorted using a FACStar cell sorter (BD Biosciences) after
labeling with PE-conjugated anti-CD4 Ab (clone RM4-5; BD
PharMingen). Isolated CD4+ fractions were 98%
pure, and CD4- fractions were >99% CD4
deficient, as assessed by flow cytometry (data not shown).
Recipient mice were infected i.c. with LCMV 2 days before cell transfer. Recipients were also given 650 rad gamma radiation using a GammaCell-40 irradiator 3 days before cell transfer. This dose of irradiation prevented immune responses against LCMV, as assessed by ELISPOT, and did not cause weight loss (data not shown). Gamma-irradiated mice were used as recipients instead of T cell-deficient RAG-/- or SCID mice so that certain strains of gene-targeted mice could be used as recipients. To remove any residual transferred CD8 T cells capable of causing lethal meningitis (30, 31), recipient mice were injected i.p. with the anti-CD8 Ab 2.43 (32) on the day of cell transfer. Depletion of CD8 T cells was confirmed by flow cytometry.
Donor cells were injected i.v. into recipient mice (= day 0) in 400 µl PBS. Following adoptive transfer, recipient mice were weighed daily. Experiments were typically terminated at or before 8–10 days after transfer to exclude the possibility that naive LCMV-specific CD4 T cells in various donor populations would become activated, expand, and contribute to weight loss.
Western blotting
Lysates were prepared from brain homogenates and from the cell
lines CH.B2 (H-2b) and CH.12
(H-2a) using previously described methods
(35). Equal amounts of protein from each sample were
separated on a 12% SDS-PAGE gel, and I-Ab was
detected by Western blot using culture supernatant from the
anti-I-Ab-secreting hybridoma KL295, as
described previously (35).
Measuring food and water intake
2m-/- mice were
housed in metabolism cages (Nalgene Company, Rochester, NY), and given
continuous access to food and water. At 24-h intervals, food and water
containers were weighed, and their masses were subtracted from those of
the respective containers at the previous time point. This difference
was taken to be the amount of food or water consumed during each 24-h
interval. The mice also were weighed at each time point. Six days after
the start of the experiment, mice were infected i.c. with LCMV. All
measurements were taken at hour 11:00 of the light cycle.
Serum leptin quantification
2m-/- mice were
infected i.c. with LCMV, injected i.c. with PBS, or left untreated. At
hour 11:00 of the light cycle on various days after i.c. infection,
mice were bled from the retroorbital sinus. Serum was collected and
stored at -80°C until assayed. Leptin concentrations were measured
by ELISA, according to the instructions provided with the Quantikine
immunoassay for mouse leptin (R&D Systems, Minneapolis, MN).
Quantification of neuropeptides
2m-/- mice were
infected i.c. with LCMV, injected i.c. with PBS, or left untreated. At
hour 11:00 of the light cycle on days 6, 12, or 16 after infection,
mice were bled by cardiac puncture and perfused with PBS. Brains were
removed, and tissue blocks containing the hypothalamus were prepared.
Neuropeptides were isolated from tissues according to published methods
(36). Samples were lyophilized and stored at -80°C
until subjected to RIA specific for each neuropeptide. Concentrations
of the neuropeptides -MSH, neuropeptide Y (NPY), and neuromedin U
were measured using RIA kits, according to instructions provided by the
manufacturer (Pheonix Pharmaceuticals, Belmont, CA).
Statistical analyses
The statistical significance of differences observed between groups or of changes within individual groups over time was assessed by an ANOVA using the program Prism, version 3.0 (Graphpad Software, San Diego, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
If wasting disease induced by LCMV infection results from the CD4
T cell response against the virus, then we would expect weight loss to
correlate with the generation of the LCMV-specific CD4 T cell response.
To determine whether this is the case, we enumerated LCMV-reactive CD4
T cells using IFN- ELISPOT at various time points following
infection of 2m-/-
mice. By 10–12 days after infection, the number of splenic cells that
reacted against the dominant class II-restricted LCMV epitope, gp61,
peaked at 
2 x 106 (Fig. 1
A). This peak in the CD4 T
cell response correlated with the onset of wasting (Fig. 1B). After the antiviral response peaked, it rapidly
declined so that by 20 days after infection the numbers of
gp61-reactive CD4 T cells decreased nearly 45-fold from their peak
numbers (Fig. 1A), which corresponded to the time when mice
began to recover from the wasting disease (Fig. 1B). Mice
generally regain the lost body weight by 40–50 days after infection
(16) (data not shown), and by this time gp61-reactive
cells were nearly undetectable (Fig. 1A). CD4 T cell
responses against the subdominant MHC class II-restricted LCMV epitope
np309 followed the same trend, but at all time points examined, the
numbers of np309-reactive cells were approximately one-half of the
numbers of gp61-reactive cells (data not shown). These data indicate
that the LCMV-specific CD4 T cell response correlates with the timing
of wasting disease.
|
We found that if we adoptively transferred unfractionated spleen
cells from LCMV-infected mice into infected and irradiated recipient
mice, the recipients lost 20% of their initial body weight within 6
days, and removal of donor CD4 T cells before injection prevented
transfer of weight loss (Fig. 2).
Transfer of equal numbers of cells from noninfected mice did not cause
weight loss. These results indicate that the wasting disease can be
transferred using cells primed during LCMV infection and confirm that
CD4 T cells are required for the wasting disease. We reasoned that if
antiviral CD4 T cells are responsible for inducing wasting disease,
then purified CD4 T cells from LCMV-infected mice should cause wasting
if adoptively transferred to another LCMV-infected mouse. When we
transferred purified CD4 T cells from LCMV-infected mice, weight loss
of recipients was comparable with that induced by transfer of whole
spleen cells from infected mice (Fig. 2). The transfer of wasting
required CD4 T cells primed during LCMV infection because transfer of
purified CD4 T cells from noninfected mice did not cause weight loss.
These data indicate that CD4 T cells primed during LCMV infection are
sufficient to cause wasting disease.
|
and TNF- in LCMV-induced wasting disease
A major function of CD4 T cells is to secrete cytokines, so we
examined whether cytokines contribute to LCMV-induced wasting disease.
We first needed to identify cytokines produced by LCMV-specific CD4 T
cells under conditions that lead to wasting. To do so, we isolated
cells from brains or spleens of
2m-/- mice after i.c.
infection with LCMV. Cytokines produced by CD4 T cells following
epitope-specific stimulation were detected using intracellular
staining. CD4 T cells from spleens (Fig. 3A) and from brains (Fig. 3B) produced IFN- and TNF- following stimulation with
gp61 (Fig. 3) or with np309 (data not shown), whereas unstimulated CD4
T cells did not produce either cytokine. We did not detect CD4 T cells
producing IL-4 or IL-6 by intracellular cytokine staining following
LCMV infection of
2m-/- mice (data not
shown).
|
or IFN- contributes to the LCMV-induced
wasting disease, we adoptively transferred spleen cells from
LCMV-infected TNF--/- or
IFN--/- donor mice and assessed whether the
transferred cells caused weight loss. Recipients of cells from
LCMV-infected TNF--/- mice lost 25–30% of
their starting body weight within 6 days (Fig. 4A), which was
indistinguishable from the weight loss induced by cells from infected
wild-type B6 mice. Weight loss comparable with wild-type controls was
observed regardless of whether the TNF--/-
donor cells were transferred into wild-type recipients or into TNF
receptor-deficient TNFR-1-/- or
TNFR-2-/- recipients (Fig. 4A).
These results strongly suggest that TNF- is not required for the
LCMV-induced wasting disease. Fig. 4B shows the results of
experiments to assess the role of IFN-. As expected, transfer of
cells from noninfected mice did not cause weight loss. Mice that
received cells from infected wild-type mice lost 25% of their body
weight. In contrast, mice that received cells from infected
IFN--/- mice lost 10% of their starting
body weight (Fig. 4B). This indicates that IFN-
contributes to the LCMV-induced wasting disease. We additionally
transferred IFN--/- donor cells into
TNFR-1-/- recipient mice, but the weight loss
of recipients observed under these conditions was indistinguishable
from that observed when IFN--/- donor cells
were transferred into wild-type recipients (data not shown). The
results further suggest that TNF- does not contribute to the wasting
disease and confirm that IFN- contributes to the wasting disease
following LCMV infection.
|
-/- cells
We have shown that transfer of LCMV-primed
IFN--/- cells leads to less severe weight
loss than does transfer of primed wild-type cells (Fig. 4
B).
One possible explanation for this is that without IFN-, insufficient
LCMV-specific CD4 T cells are generated in the donor mice. However,
when we compared spleen cells from wild-type B6 donor mice with
IFN--/- donor mice, we did not observe any
consistent differences in the percentages or the numbers of donor CD4 T
cells reacting against gp61 following infection (data not shown). This
indicates that IFN- deficiency does not impair the generation of
LCMV-reactive CD4 T cells.
It is also possible that IFN--/- cells are
not maintained following i.v. transfer or that
IFN--/- donor cells do not migrate to
appropriate sites after transfer. Finally, transferred
IFN--/- CD4 T cells may be unable to respond
to Ag. To test these possibilities, we enumerated virus-reactive CD4 T
cells in spleen and brains of each recipient mouse following transfer.
LCMV-specific CD4 T cells produce TNF- (Fig. 3), so we detected
IFN--/- antiviral CD4 T cells by
intracellular cytokine staining for TNF-. In the spleens of
recipient mice 5 days after cell transfer, no significant differences
were observed in the total number of cells or in the number of CD4 T
cells between mice that received wild-type cells and mice that received
IFN--/- cells. By intracellular staining, we
found that 5% of CD4 T cells in the spleen were gp61 specific when
IFN--/- cells were transferred compared with
4% if wild-type cells were transferred (Fig. 5A). This shows that there is
no defect in the ability of LCMV-specific
IFN--/- CD4 T cells to survive i.v. transfer
or to populate the spleen following transfer. The number of mononuclear
cells as well as the number and proportion of CD4 cells isolated from
brains of recipients receiving either wild-type or
IFN--/- cells were similar (Fig. 5B and data not shown). In contrast, we did not observe CD4
T cells infiltrating the brains of recipient mice if cells from
noninfected mice were transferred. This indicates that LCMV-specific
CD4 T cells migrate to the brains of infected recipient mice following
transfer, and that this migration is not grossly affected by IFN-
deficiency. When we measured LCMV-specific responses, greater than 12%
of CD4 T cells from brains of recipients receiving wild-type cells were
gp61 specific. In contrast, few (2.4%) of the CD4 T cells in brains of
mice receiving IFN--/- cells were able to
respond against gp61 (Fig. 5B). Taken together, these
results show that IFN--/- donor cells are
capable of responding against LCMV in the spleen and are able to
migrate to the brain, yet very few of these cells respond against LCMV
when they reach the brain.
|
dependence of LCMV-specific CD4 T cell responses in the CNS
during wasting disease
IFN- is a potent inducer of class II MHC expression in the CNS
(37), so one possible explanation for the failure of
transferred IFN--/- T cells to respond in
the CNS is that there is insufficient expression of
I-Ab to present LCMV Ag to these T cells. To
investigate this, we examined levels of I-Ab in
the brains of recipient mice by Western blot after adoptive transfer of
cells from infected IFN--/- mice, infected
wild-type B6 mice, or noninfected wild-type mice, as described above.
When no cells were transferred, I-Ab was almost
undetectable, whereas I-Ab was dramatically
up-regulated in brains after transfer of wild-type cells (Fig. 6A). In recipients of
IFN--/- cells, levels of
I-Ab in brains were comparable with those in
controls receiving noninfected spleen cells or no cells at all (Fig. 6A). This shows that production of IFN- from the
transferred cells is required for up-regulation of
I-Ab in the brain following i.c. infection
with LCMV.
|
-/- CD4 T cells are not
responding in the brain because of insufficient
I-Ab expression, then their ability to respond
should be restored by adding I-Ab-expressing
cells during the intracellular staining assay. This was done by adding
spleen cells from CD4-/- mice so that we could
distinguish brain-derived CD4 T cells from the added spleen cells. When
we added I-Ab-expressing spleen cells as APC
(from CD4-/- mice) to brain-derived
IFN--/- donor cells, the percentage of
gp61-reactive CD4 T cells increased from 1.3% to 18.1% (Fig. 6B). Addition of APC had a modest effect on the proportion
of transferred wild-type CD4 T cells that reacted against gp61. These
data indicate that LCMV-specific CD4 T cells from brains of mice
receiving IFN--/- cells are capable of
responding against viral Ag, and suggest that they do not respond due
to insufficient Ag presentation within the brain. Together, the data
shown in Fig. 6 suggest that IFN- promotes LCMV-specific CD4 T cell
responses in the CNS during wasting disease by up-regulating MHC class
II expression. Food and water intake during LCMV-induced wasting disease
To this point, we had determined that CD4 T cells activated by
LCMV infection are sufficient to cause wasting disease and that
IFN-, which is produced by LCMV-specific CD4 T cells, contributes to
the disease. However, it was still unclear how the CD4 T cells promote
wasting. In an attempt to identify factors that could mediate
LCMV-induced wasting, we investigated the physiological basis for the
weight loss. One possible explanation is that mice decrease their
consumption of food or water. To address whether wasting correlates
with decreased intake of food or water, we measured changes in body
weight and in the amount of food and water consumed daily before and
after i.c. infection of
2m-/- mice with LCMV.
Before infection, mice consumed 3–4 g of food per day (Fig. 7B). Daily food intake was
fairly constant until approximately day 6 after infection. By 8 days
after infection, the amount of food consumed per day decreased 3- to
4-fold (Fig. 7B), and mice had lost >15% of their starting
body weight (Fig. 7A). Daily food intake increased after
this time, but it took 10 additional days to return to preinfection
levels of food intake (Fig. 7B). During these 10 days, mice
reached their peak weight loss, which was maintained until 19 days
after infection (Fig. 7A). We also observed a similar trend
in water consumption (Fig. 7C), but the decrease of water
intake was less pronounced. Together, these data show that daily intake
of food, and to a lesser degree water, decreases as mice succumb to
wasting disease.
|
The decrease of food and water intake observed as LCMV-infected
mice lose weight (Fig. 7) suggests that the wasting disease results at
least partly from a loss of appetite, or anorexia. IL-1 and IL-6 can
cause anorexia (1, 2, 3), and both cytokines are expressed in
the CNS during persistent LCMV infection (18, 20).
Therefore, we determined whether IL-1 or IL-6 contributes to
LCMV-induced wasting disease.
If IL-1 is required for the wasting disease, then IL-1R-deficient
(IL-1R-/-) mice that cannot respond to IL-1
should not lose weight after infection. We infected CD8 T cell-depleted
IL-1R-/- mice i.c. with LCMV and measured
subsequent changes in body weight. Infected
IL-1R-/- mice lost little or no weight compared
with noninfected wild-type B6 control mice (Fig. 8A). In contrast, infected B6
control mice lost 20–25% of their initial body weight. These results
show that the IL-1R is required for wasting disease induced by
LCMV.
|
B). After day 14, however,
IL-6-/- mice quickly regained the weight lost,
whereas wild-type mice did not begin to regain the weight lost until at
least 24 days after infection. These results indicate that IL-6
contributes to the wasting disease, particularly in the later stages of
the disease. Levels of factors that control appetite during LCMV-induced wasting disease
If anorexia contributes to the LCMV-induced wasting disease, as
our data suggest (Fig. 7), then there should be a change in the amounts
of one or more of the factors that control appetite as mice lose
weight. The hypothalamus is the main appetite control center of the
brain, so we measured levels of several factors that act within the
hypothalamus to control appetite. We chose to measure levels of leptin,
-MSH, NPY, and neuromedin U because these factors strongly influence
appetite (38, 39).
Leptin is produced systemically and is a potent suppressor of appetite
(38). Thus, if leptin were mediating the weight loss,
leptin levels should increase as mice lose weight. Serum concentrations
of leptin in untreated
2m-/- mice were 7–8
ng/ml before infection (Fig. 9A). Conversely, by 8–10 days
after i.c. infection with LCMV, serum leptin concentrations were
20-fold lower than preinfection concentrations and 10-fold lower
compared with those of mock-infected
2m-/- mice (Fig. 9A). The observation that serum leptin levels drastically
decrease following infection is inconsistent with leptin being a
mediator of LCMV-induced wasting disease.
|
-MSH are neuropeptides that act within the
hypothalamus to affect appetite (38, 39). To assess the
levels of these neuropeptides during wasting, we measured the amounts
of each neuropeptide present in brain tissues containing the
hypothalamus isolated at various time points after i.c. infection of
2m-/- mice. For each
time point, these amounts were compared with the amounts of weight lost
during the preceding 2 days. In this way, we assessed whether mice were
in the process of losing weight when the neuropeptide was measured. We
detected no change in the levels of NPY or neuromedin U following
infection, suggesting that neither of these factors mediates
LCMV-induced wasting disease. -MSH suppresses appetite
(38). Thus, if -MSH mediates wasting, then we would
expect hypothalamic -MSH levels to increase at times when mice are
losing weight. We detected 2 ng -MSH/mg protein isolated from
untreated (day 0) brain tissues before infection (Fig. 9B).
Amounts of -MSH remained low (from 4 to 7 ng/mg protein) at days 6
and 12 after infection and at day 16 after control injection with PBS.
In this particular experiment, we also did not observe weight loss
during the 2 days preceding these time points. In contrast, a 14-fold
increase in -MSH from preinfection levels to 27 ng/mg protein was
observed at 16 days after LCMV infection, and this correlated with
significant loss of body weight (2.3 g) from day 14 to day 16 (Fig. 9B). These data are consistent with -MSH being a mediator
of wasting disease induced by LCMV infection. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, IL-1, and IL-6
contribute to wasting and suggest that the disease is at least partly
due to anorexia, possibly via the action of -MSH. Although CTL-deficient mice cannot control LCMV infection (13, 14, 15), viral replication itself does not cause wasting disease (17). Rather, the observed weight loss results from the immune response against the virus (17). We therefore focused on what components of the immune response against LCMV contribute to the disease. The requirement for CD4 T cells for wasting suggests that the CD4 T cell responses against LCMV may induce disease. Following infection, we found that the expansion of large numbers of LCMV-reactive CD4 T cells correlates with induction of wasting disease, and that the loss of these LCMV-reactive CD4 T cells correlates with recovery from wasting. Further evidence that the LCMV-specific CD4 T cell response leads to wasting comes from our adoptive transfer studies. We showed that the wasting disease could be adoptively transferred with purified CD4 T cells taken from LCMV-primed donor mice, but not with those taken from noninfected donors. These results indicate that CD4 T cells activated by LCMV infection are sufficient to cause wasting disease. At 9–10 days after infection, when we isolated donor cells for adoptive transfer, LCMV is undetectable by plaque assay (data not shown). However, it is possible that in our adoptive transfer experiments, residual virus is transferred along with the donor cells. Regardless, the transfer of virus to recipient mice would not affect these experiments because the recipients are already infected with LCMV at the time of transfer and because LCMV replication by itself does not cause wasting disease (17). It is possible that CD4 T cells that are not specific for LCMV become activated in a bystander fashion during LCMV infection, and that they are the cells that are responsible for inducing wasting disease. However, a recent report has estimated that greater than 80% of the CD4 T cells that proliferate in response to LCMV infection react against either gp61 or np309 (40). This estimate argues against substantial bystander activation of CD4 T cells during LCMV infection. Therefore, we think it is unlikely that bystander-activated CD4 T cells cause wasting disease following LCMV infection.
To elucidate functions of LCMV-specific CD4 T cells that induce wasting
disease, we first determined what cytokines are produced by
LCMV-specific CD4 T cells during the disease. We found that
LCMV-specific CD4 T cells from brains and spleens of
2m-/- mice produce
IFN- and TNF- during wasting. Hildeman et al. (41)
have shown using ELISA that cells from the cerebrospinal fluid of
2m-/- mice 7 days
after i.c. infection produce IFN-, but these authors were unable to
detect TNF- production by cells from spleen or from cerebrospinal
fluid of these animals (18, 41). A recent report showed
that the percentage of CD4 T cells producing TNF- in response to
gp61 increases from 0.9 to 3.3% from day 7 to day 9 following
infection (40). We find that by day 10–12 after i.p.
infection of both B6 and
2m-/- mice, 6–8% of
splenic CD4 T cells produce TNF- in response to gp61 (data not
shown), so it is likely that the time points examined by Hildeman et
al. (3 and 7 days after i.c. infection) were too early in the antiviral
CD4 T cell response to detect TNF- production by CD4 T cells. From
our data, we conclude that under conditions that lead to wasting
disease, LCMV-specific CD4 T cells produce IFN- and TNF-.
We next determined whether IFN- or TNF- contributes to wasting.
In our adoptive transfer studies, the weight loss induced by
TNF--/- cells was equivalent to that induced
by wild-type cells. This was the case regardless of whether the
recipient mice lacked the ability to respond to TNF- through TNFR-1
or TNFR-2. These results strongly suggest that TNF- is not required
for wasting. It is formally possible, however, that TNF- produced by
radioresistant recipient cells induces the transferred cells to produce
another factor that causes wasting.
IFN--/- cells adoptively transferred minimal
weight loss compared with that transferred by wild-type cells, so we
conclude that IFN- contributes to wasting. Removing IFN- did not
provide complete protection from wasting disease because recipients of
IFN--/- cells lost more weight than did
recipients of noninfected cells. It is unlikely that recipient cells
produced sufficient IFN- to induce the minimal weight loss observed
because we detected no IFN--producing cells in recipients after
transfer of noninfected cells. The most likely explanation, then, is
that another factor acts independently of IFN- to promote weight
loss. Regardless, little weight loss was observed in the absence of
IFN-, so IFN- is an important contributing factor to the wasting
disease.
To determine how IFN- contributes to the wasting disease, we
examined whether removing IFN- prevents CD4 T cells from responding
against LCMV in recipient mice after adoptive transfer. To detect
LCMV-reactive CD4 T cells in these experiments, we relied on their
ability to produce TNF- after antigenic stimulation. We have
observed that 10 days or more after infection, the ratio of the
percentage of TNF--positive CD4 T cells to the percentage of
IFN--positive CD4 T cells is constant (Fig. 3 and data not shown).
This indicates that for our experiments, TNF- production is a
reliable indicator of LCMV-reactive CD4 T cells. Our data indicate that
while the transferred IFN--/- CD4 T cells
are capable of responding against LCMV in the spleen, few do so once
they reach the brain. Fewer LCMV-responsive CD4 T cells in the CNS
correlate with relatively little weight loss in recipients of
IFN--/- cells, and is consistent with the
idea that LCMV-induced wasting disease is a pathology of the CNS. Upon
further investigation, we found that levels of
I-Ab were substantially lower in brains of mice
receiving IFN--/- cells when compared with
those receiving wild-type cells. Additionally, we could restore
antiviral responses of brain-derived IFN--/-
CD4 T cells if we added spleen cells as APC. Together, these results
suggest that IFN- contributes to the LCMV-induced wasting disease by
inducing sufficient class II MHC expression to drive LCMV-specific CD4
T cell responses within the CNS.
Although we identified IFN- as a factor produced by CD4 T cells that
contributes to the LCMV-induced wasting disease, the mechanism by which
the weight loss occurs was still unclear. One possible explanation is
that mice decrease food or water intake, so we measured food and water
intake over the course of wasting. We observed that
2m-/- mice decreased
food intake at the onset of wasting, in agreement with a recent report
(18). By restricting uninfected mice to the same amount of
food intake as LCMV-infected mice, the authors of this report found it
possible to account for weight loss by the observed decrease in
feeding. Although this result does not demonstrate that the wasting
disease is due to decreased food intake, it suggests, as our data do,
that decreased food intake contributes to the weight loss. We
additionally found that water intake also decreased during wasting,
although this decrease was less pronounced than the decrease in food
intake. Decreased water intake is consistent with the severe
dehydration we observe during LCMV-induced wasting (C. Kamperschroer,
unpublished observations). Taken together, our data show that intake of
food and water decreases during wasting disease, suggesting that
anorexia (appetite suppression) is at least partly responsible for the
wasting disease.
IL-1 can induce anorexia (1, 2, 3), and we found that
IL-1R-/- mice did not lose weight after i.c.
infection, demonstrating that the IL-1R is required for wasting. A
recent study by Hildeman and Muller (18) reported that
treatment of mice via intracerebroventricular (i.c.v.) cannulae with
neutralizing Ab against IL-1 lessened the severity of LCMV-induced
wasting. However, there was no statistically significant difference in
weight loss between mice treated with anti-IL-1 and untreated
controls at any time points analyzed, except at day 8 after infection,
and both groups ultimately succumbed to wasting. IL-1 also binds
IL-1R, so one explanation consistent with both our data and that of
Hildeman and Muller is that IL-1 is the major mediator of
LCMV-induced wasting disease, whereas IL-1 plays a minor role in the
disease. Another possibility is that these authors did not achieve
complete neutralization of IL-1. Regardless, because
IL-1R-/- mice were completely protected from
weight loss after i.c. infection with LCMV, we infer that IL-1 is
essential for the LCMV-induced wasting disease. IL-1 directly acts
upon neurons in the hypothalamus during the weight loss that results
from i.c.v. injection of this cytokine (42). It is
possible that IL-1 mediates LCMV-induced wasting in this manner.
Our results demonstrate that IL-6 contributes to the wasting disease,
and that its effect is most prominent beyond the first 2 wk after
infection. IL-6 is induced in the brain later than IFN- or IL-1
following i.c. infection of B6 mice (43), consistent with
IL-6 having its major effect later in the disease course. During LCMV
infection, the IL-6 produced in the CNS is largely derived from
microglia and astrocytes (44), so these cell types may
produce the IL-6 that contributes to wasting disease.
The involvement of IL-1 and IL-6 in LCMV-induced wasting is consistent
with the CD4 T cell dependence of this disease. It has been shown that
production of IL-1 and IL-6 in the LCMV-infected CNS requires T cells
(43, 44). We know that LCMV-specific CD4 T cells produce
IFN- and that IFN- can induce cells to produce IL-1 (45, 46), so it is possible that the IFN- secreted by
LCMV-specific CD4 T cells induces IL-1 production. This could explain
why removing IFN- prevents severe weight loss. However, another CD4
T cell-derived factor must also induce IL-1 production in this scenario
because the protection afforded by removing IFN- is not
complete.
The observation that LCMV-infected mice decrease food and water
consumption at the onset of wasting (18) prompted us to
examine factors involved in the regulation of appetite. We observed
increased amounts of the appetite-suppressing neuropeptide -MSH in
brain tissue containing the hypothalamus only at a time when mice were
losing weight. This is consistent with the idea that LCMV-induced
weight loss is mediated by -MSH. The increase in -MSH is unlikely
to directly result from viral replication because at all time points
examined in this experiment, brain viral titers are equivalent (data
not shown). There is evidence that i.c.v. injection of IL-1 causes
weight loss through the -MSH pathway (47), so during
LCMV-induced wasting disease, IL-1 may mediate weight loss by directly
stimulating hypothalamic neurons (41) to produce -MSH.
Additional experiments are needed to determine whether this is the
case.
With the data presented in this study, we propose a model for the
pathogenesis of LCMV-induced wasting disease (Fig. 10). Following i.c. infection with
LCMV, virus replicates within the CNS. Activated CD4 T cells migrate to
the CNS, where they encounter LCMV Ag. These antiviral CD4 T cells
react against LCMV by producing IFN- and TNF-. IFN-
up-regulates MHC class II on cells within the CNS, which in turn drives
the CD4 T cell response. The antiviral CD4 T cell response cannot
control the virus, and instead initiates the wasting disease. IFN-
or other CD4 T cell-derived factors induce other cells within the CNS,
such as microglia or infiltrating macrophages, to produce IL-1 and
IL-6. IL-1 directly mediates weight loss by inducing neurons in the
hypothalamus to produce -MSH or other anorectic neuropeptides. IL-6
also contributes to the wasting disease, particularly later in the
disease. Through the actions of IL-1 and IL-6, mice succumb to severe
weight loss due at least partly to anorexia. Despite continued viral
replication, antiviral CD4 T cell responses wane by 2–3 wk after
infection, possibly due to overwhelming Ag load (48).
Following the loss of the CD4 T cell response against LCMV, the
inflammatory response in the brain subsides, and production of IL-1 and
IL-6 within the brain decreases. In the absence of these cytokines, the
mouse begins to recover from disease and regains the body weight
lost.
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Daniel G. Quinn, Department of Microbiology and Immunology, Loyola University Chicago Medical Center, Maywood, IL 60153. E-mail address: dquinn{at}lumc.edu
3 Abbreviations used in this paper: LCMV, lymphocytic choriomeningitis virus; -MSH, -melanocyte-stimulating hormone; 2m, 2-microglobulin; i.c., intracranial; i.c.v., intracerebroventricular; NP, nucleoprotein; NPY, neuropeptide Y.
Received for publication October 23, 2001. Accepted for publication April 29, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-herpesvirus-specific CD8+ T cells but confers minimal survival advantage on CD4-deficient mice. Proc. Natl. Acad. Sci. USA 97:2725. in the feeding responses to endotoxin and influenza virus infection in mice. Brain Behav. Immun. 13:252.[Medline]
2-microglobulin gene. J. Virol. 67:332.2-microglobulin-deficient mice by CD4+ T cells. Int. Immunol. 5:1193.. Viral Immunol. 13:273.[Medline]
interferon. J. Virol. 65:1364.2M, MHC class I proteins, and CD8+ T cells. Science 248:1227. genes. Science 259:1739.-deficient mice: a critical requirement for TNF in the formation of primary B cell follicles, follicular dendritic cell networks and germinal centers, and in the maturation of the humoral immune response. J. Exp. Med. 184:1397.
B6,
n = 4; B6
), or left untreated
(
). Leptin concentrations in serum samples taken at various time
points after infection were measured by ELISA. Each set of filled
symbols represents an individual mouse. Data points indicated by open
symbols represent the mean ± SEM. Statistically significant
differences were observed between LCMV (n = 6) vs
PBS (n = 3) and untreated (n =
5) groups, p 
0.001. B, Levels of
). Values shown represent the mean ±
SEM.