| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Pathology and Laboratory Medicine, University of Wisconsin Medical School, and
Department of Pathobiological Sciences, School of Veterinary Medicine, University of Wisconsin, Madison, WI 53706; and
Department of Internal Medicine, University of Iowa, Iowa City, IA 52242
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-specific primers and CDR3 analysis. Each granuloma has a
very diverse TCR repertoire indicating that most of the T cells
recruited to these lesions are activated systemically. At the same
time, sequence analysis of individually sized CDR3 products from single
granuloma indicate that a fraction of T cells expand locally at the
lesion site. Using TCR transgenic mice containing a pigeon cytochrome
c-specific T cell population or lymphocytic
choriomeningitis virus infection tracked with lymphocytic
choriomeningitis virus-specific tetramers, we demonstrated that
nonspecific T cells home to the granuloma if they are activated.
However, recombinase-activating gene 2-/- pigeon
cytochrome c-specific TCR transgenic mice fail to
form granulomas in response to S. mansoni ova even after
T cell activation, suggesting a requirement for egg-specific T cells in
the initiation of these inflammatory lesions. Understanding the
mechanism of T cell recruitment into granulomas has important
implications for the rational design of immunotherapies for
granulomatous diseases. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CD4+ T cells are required for the initiation,
regulation, and possibly closure of granulomatous inflammation
(2, 3, 4, 5). Because of this, granulomas represent a
microenvironment for studying the regulation of
CD4+ T cells and their requirements for
protective function. T cell responses to Schistosoma mansoni
in mice have been extensively studied after parasite-induced immune
reactions (3, 6, 7, 8, 9). After 8 wk of infection, the liver
contains a large number of mature granulomas surrounding deposited
eggs. Approximately 50% of the granuloma-infiltrating cells are
eosinophils, 30% are macrophages, and 
10% are T cells. These T
cells produce IL-4, IL-5, IL-6, and IL-10, and are characteristically
of the TH2 phenotype during chronic infection
(8, 10, 11, 12). The formation of granulomas is largely
dependent upon the CD4+ T cells that represent
80% of the total infiltrating T cells (12), and in their
absence the liver tissue becomes rapidly necrotic.
Our approach in these studies focused on the question of how T cells initiate granuloma formation. The initial recruitment of T cells to granulomas may occur either with the arrival of limited numbers of T cells from the periphery which expand locally in response to the presence of parasite Ag, or systemically expanded T cells may localize to granulomas in a nonspecific way. It is also possible that both systemically and locally expanded T cells are present at local inflammatory lesions. Each one of these alternatives makes specific predictions about the TCR repertoire present in individual lesions which can be tested by the analysis of single granulomas (SG).3 Because ova are quite large, single liver granulomas induced by schistosome can be easily isolated free of liver tissue (11) for study by flow cytometry and RT-PCR to examine TCR repertoire usage on an individual basis. In our analysis, we observed a very heterogeneous repertoire in individual granulomas, but with substantial lesion-to-lesion variation and evidence for local clonal expansion.
A related question is whether T cell specificity is important for recruitment of T cells to the granuloma. T cells homing to granulomas may be limited to those with parasite-specific TCR or may encompass a larger repertoire of T cells marked by an activated phenotype. To test whether nonschistosome-specific T cells can home to S. mansoni-induced granulomas, we used flow cytometry of granuloma-infiltrating cells to examine the homing of nonparasite-specific T cells after immunization of infected TCR transgenic mice, and after lymphocytic choriomeningitis virus (LCMV) coinfection of infected wild-type mice. Both sets of data suggested that while Ag specificity is an important determinant of granuloma localization, peripheral activation is sufficient to enable T cells of any specificity to home to local inflammatory sites.
Finally, since our experiments had shown that peripheral activation allowed the entry of non-Ag-specific T cells into S. mansoni-induced granulomas, we questioned whether TCR specificity was an absolute requirement for granuloma formation. To address this question, we infected recombinase-activating gene 2 (RAG2)-/- TCR transgenic mice containing a single T cell population lacking specificity for S. mansoni Ags. Even after peripheral activation with cognate Ag, granuloma formation was absent, indicating that a large population of activated nonspecific T cells is not sufficient to induce granulomas in response to schistosome ova, and that parasite-specific T cells are essential in the initiation of granuloma formation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In these studies, we used CBA/J mice (Cumberland View Farms, Clinton, TN) and a TCR transgenic strain, AND (gift of S. Hedrick, University of California, San Diego, La Jolla, CA). B10.D2 RAG2-/- and 5C.C7/RAG2-/-(H-2k) (13) (Taconic Emerging Models Program) mice were purchased from Taconic Farms (Germantown, NY).
Infections with S. mansoni and histology
CBA/J mice were infected s.c. at 7–8 wk of age with 35 cercariae of the Puerto Rican strain of S. mansoni (14). RAG2-/- and 5C.C7/RAG2-/- infections with the Naval Medical Research Institute strain were performed under National Institute of Allergy and Infectious Diseases contract by the Biomedical Research Institute (Rockville, MD). Infection was monitored by the presence of visible granulomas on the liver surface 8 wk after injection of cercariae. Additionally, for microscopic inspection of granulomas, pieces of liver were fixed in 10% formalin, before being imbedded in paraffin for thin sectioning (10 µm). H&E staining was done by the Histopathology Service of the Department of Pathology of University of Wisconsin (Madison, WI).
Coinfections with LCMV
LCMV-Armstrong (15) was used in these studies for infection of C57BL/6 mice previously infected with S. mansoni for 8 wk. Infection was by i.p. injection (2 x 10 5 PFU/mouse). Nine days later, mice were sacrificed and their organs isolated for study by histology and flow cytometry.
Isolation of splenocytes and granuloma-infiltrating cells
Isolation of bulk granulomas has been previously described (11, 16, 17). Spleens were removed aseptically from 8- to 16-wk-old mice, and viable cells separated by centrifugation through Lympholyte M solution (Cedarlane Laboratories, Hornsby, Ontario, Canada). SGs were isolated from the preparative suspension before dispersal with collagenase under x10 magnification using a Pasteur pipet flame-drawn to a finer tip and then processed in bullet tubes in a similar manner to bulk granulomas.
Polymerase chain reaction
RT-PCR was performed as previously described (16) using published primers for CDR3 length analysis (18). Primers were titered using positive (spleen cDNA) and negative (endothelial cell cDNA) templates. Products of these reactions were analyzed on either 1 or 2% agarose gels in 1x TBE as indicated and visualized by ethidium bromide staining.
Size separation of PCR products and sequencing of individual bands
The products of the secondary CDR3 PCR were end-labeled with
[
-32P]-ATP using T4 polynucleotide
kinase (NEB, Beverly, MA) by standard methods. End-labeled products
were separated on 6% acrylamide 8 M urea sequencing gels alongside
control samples amplified from bulk spleen RNA and visualized by
autoradiography. Individual bands were cut out of dried gels and
sequenced by the University of Wisconsin Biotechnology Center using the
indicated V primer.
Flow cytometry and Abs
Splenocytes or granuloma cell suspensions were incubated for 30 min at 4°C with different labeled Abs at saturation, then washed and analyzed. Unlabeled 50 µg/ml 2.4G2 anti-FcR Ab was used to block binding of labeled Abs to FcRs. Cell surface staining on 10,000 events was measured using a FACSCalibur (BD Biosciences, Mountain View, CA) and analyzed using the CellQuest computer program (Power Macintosh version 3.0; BD Biosciences).
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 (19). For detection of LCMV-specific CD8+ T cells, single-cell suspensions from spleen or granuloma were surface stained with Abs and fluorochrome-labeled MHC I tetramer for 1 h at 4°C, followed by four-color flow cytometry.
mAbs used were purified from hybridoma cell lines obtained from the
American Type Culture Collection (Manassas, VA) as indicated. Hybridoma
cells were cultured in HB-101 serum-free medium, and the Abs were
precipitated from supernatants by 45% saturated ammonium sulfate. The
purified Abs were labeled with biotin, FITC, or Cy5. Abs used included
those specific for murine CD4 (GK1.5), CD44 (Pgp-1; IM 7.8.1), and
L-selectin (CD62L; Mel-14). The KJ25 hybridoma specific for V3 was a
gift from P. Marrack (University of Colorado Heath Sciences Center,
Denver, CO). Quantum Red-labeled anti-
TCR and
PE-labeled anti-V3 Abs were purchased from Sigma-Aldrich (St.
Louis, MO). PE-labeled anti-murine CD8 Ab was purchased from BD
PharMingen (San Diego, CA).
Pigeon cytochrome C (PCC)-coated latex beads
Polystyrene latex microspheres (diameter 5 µm) were purchased from Polysciences (Warrington, PA). PCC protein (Sigma-Aldrich) was chemically attached to activated beads as previously described (20). Briefly, 2 mg of purified protein were mixed with 107 polystyrene beads in a final volume of 1 ml PBS, and incubated for 90 min at room temperature with constant tumbling. Beads were then blocked in 1.5 ml of PBS/1% BSA for 30 min. After three washes in PBS, latex beads were resuspended in PBS and stored at 4°C.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We began our analysis of TCR usage in granulomas with the
isolation of individual lesions from CBA/J mice infected with S.
mansoni for 8 wk. Granulomas formed in chronic schistosomiasis are
readily identified due to the large size of the inducing schistosome
egg. Fig. 1
A shows the
granulomatous inflammatory reaction surrounding a schistosome egg
lodged in the murine liver, while Fig. 1B illustrates the
relative purity of a typical bulk granuloma preparation from liver
assessed by flow cytometry. CD4+ T cells are
around 40% of the cells in the lymphocyte gate and express an effector
phenotype (L-selectinlow,
CD44high). The granuloma preparation was diluted
in a petri dish (the preparation is illustrated in Fig. 1C),
and granulomas were isolated individually under a stereoscope for
analysis by RT-PCR or flow cytometry. After collagenase digestion, each
granuloma yields between 2,000 and 20,000 cells. Fig. 1D
shows flow cytometric analysis of three SGs showing that each contains
CD4+ T cells. The proportion of
CD4+ T cells in each individual granuloma is
similar to the bulk preparation (Fig. 1B), indicating a
similar recovery of CD4+ T cells from single and
bulk preparations.
|
Ten individual granulomas were prepared from a single infected
mouse liver. From those granulomas, isolated mRNA was used to
synthesize cDNA for RT-PCR analysis of TCR -chain V gene
usage. Fig. 2 shows results using
primers specific for V2, V14, and V15 in combination with a
common C primer, indicating that each individual granuloma contains
a unique pattern of V chain usage and suggesting the presence of a
diverse TCR repertoire at individual sites. Similar results arose from
the use of V1 and V4 primers (data not shown and Fig. 3). Although the TCRs from the
different SGs contain many similarities, they are also distinct from
each other, indicating their heterogeneous nature. Secondary PCR was
performed on products of the V/C primer-directed PCR, using 12
different J gene-specific primers in combination with the indicated
V. Fig. 3 shows the results of that analysis using V1, V2,
V4, and V14 primers. Products similar to V14-specific products
were seen with V15 (data not shown and Fig. 4). Reading across the lanes for each
individual granuloma can be compared with reading a "bar
code" of TCR repertoire. These data clearly show that each granuloma
contains a large and diverse repertoire and although there are many
similarities, each granuloma has a different TCR repertoire. This
argues against the presence of a few immunodominant T cell epitopes
creating a T cell repertoire at the local level. Although these data do
not rule out selection at individual sites by locally expressed Ags, it
is clear that a large repertoire of T cells are recruited from the
periphery into the individual inflammatory sites. This result excludes
the hypothesis that the local expansion of a few recruited T cells is
the dominant pathway of T cell accumulation in granulomas.
|
|
|
1 + J2.2 and V15 + J2.7) were radioactively
labeled with [-32P]ATP using T4
polynucleotide kinase and displayed on sequencing gels alongside
reactions derived from uninfected spleens as controls for banding (Fig. 4
). The end-labeled reactions from the spleen clearly show a
nonspecific or Gaussian pattern of band distribution expected from a
randomly distributed pattern of untemplated nucleotide additions in the
CDR3 region arising during the assembly of the V-D-J -chain regions.
In contrast, all of the reaction products from the SGs show an
asymmetric banding pattern in which one or two bands strongly dominate.
Differences in size and numbers of bands suggest antigenic selection
for CDR3 regions of a particular size for a given V/J
combination, while differences in band intensity of 5- to 10-fold
(compare lane 19 to 18) are consistent with the
local clonal expansion of individual T cell clones.
To eliminate the possibility that differences in band intensity were
the result of a random dilution of T cells from a broad repertoire,
individual bands were cut out of the dried sequencing gels and
subjected to sequence analysis using the appropriate primers. In the
case of dilution, heterogeneity of joint sequences would result in
unreadable sequence reactions. However, six of nine attempts yielded
enough material for readable sequence (Fig. 5, SG5 upper band, SG7, SG18,
SG19, SG20 upper band, and SG20 lower band with
the V1/J2.2 primer combination). All sequences obtained
were clearly unique with readily identifiable V/D and D/J junctions
(Fig. 5). The evidence points toward clonality of a population of
individual RT-PCR products and the possibility of local proliferation
of parasite-specific T cells.
|
We predicted that if Ag specificity plays an important
role in the localization of T cells to S. mansoni-induced
granulomas, then TCR transgenic mice should show an altered
distribution ratio of transgenic to nontransgenic T cells between the
spleen and the granuloma. We used the AND strain carrying a
V11/V3 TCR transgene specific for PCC in association with
I-Ek to examine the role of Ag specificity in the
granuloma. In uninfected AND mice, typically only 2–5% of T cells
bear nontransgenic TCR arising from incomplete allelic exclusion. Upon
infection, the proportion of nontransgenic (V3-negative) cells in
the spleen and periphery rises to 15–30% (Fig. 6A, upper plot).
Thus, the nontransgenic T cells likely expand under antigenic pressure,
while the transgenic TCR marks the nonspecific population. Neither
schistosome eggs nor schistosomal egg Ag (SEA) activate
transgenic cells for in vitro proliferation (data not shown). Granuloma
lymphocytes show a substantial increase in the ratio of nontransgenic
(V3-negative) to transgenic cells (V3-positive), implying that
there is an active selection against transgenic cells in the granuloma
T cell population (Fig. 6A, lower plot). In
addition, gating of the granuloma cells on V3-positive and -negative
populations shows that the activation stages of the two populations are
very different. A significant proportion of the V3-positive
population exhibits a resting or virgin phenotype (Fig. 6B,
upper plot), while almost all of the nontransgenic
population displays an L-selectinlow,
CD44high phenotype characteristic of activated T
cells (Fig. 6B, lower plot).
Granuloma-infiltrating CD4+ T cells from
wild-type mice do not express a CD44low,
L-selectinhigh phenotype (Fig. 1B,
right panel). It is likely that in wild-type mice, small
numbers of naive T cells enter granulomatous inflammatory sites, but
are rapidly activated so that a naive phenotype is not normally
detected among granuloma-infiltrating cells. In situations where
nonactivated T cells are in excess, naive T cells may enter granulomas
without becoming activated (21, 22).
|
11-positive) to nontransgenic (V11-negative) T cells in spleen
and granuloma was compared. Noninfected animals show significantly
depleted numbers of transgenic T cells in the spleen demonstrating that
activation was effective and led to the elimination of part of the
population through apoptosis (data not shown). In the granuloma, PCC
bead-immunized animals have greater numbers of transgenic T cells
(V11-positive) present after systemic activation (Fig. 6C, compare upper and lower plots).
Thus, specific immunization can promote the homing of
nonschistosome-specific T cells to granulomas and alter distribution of
T cells at local inflammatory sites.
To track the homing of peripherally activated nonlocal
Ag-specific T cells in a non-TCR transgenic model, C57BL/6 mice
chronically infected with S. mansoni were coinfected with
LCMV-Armstrong 9 days before sacrifice of animals and organ harvest. As
expected, LCMV infection induced a significant expansion of
LCMV-specific CD8+ T cells in the periphery
(19). Fig. 7 shows the
CD4+ to CD8+ T cell ratio
after flow cytometric analysis of splenocytes and
granuloma-infiltrating cells from either control infections or LCMV
coinfections. In splenocytes from schistosome-infected mice, the ratio
is 2:1, increasing to >3:1 in the S. mansoni-induced
granuloma. In LCMV coinfected animals, the relative proportion of
CD8+ T cells rises among splenocytes in response
to infection similar to the rise seen during control LCMV infections
(19). The granuloma-infiltrating cells from those
LCMV-infected animals contain five times more
CD8+ T cells proportionately, outnumbering
CD4+ T cells 4:1. H&E-stained thin liver sections
show large, well-formed granulomas similar to those seen in control
infections (data not shown). This result indicates that activation of T
cells in the periphery can induce their localization to granulomas. The
CD4+ T cell predominance in schistosome
granulomas may reflect the dominance of CD4+ T
cells among the activated T cells after S. mansoni
infection. MHC class I tetramers specific for
CD8+ T cells recognizing a dominant LCMV epitope
np396–404 were used to further characterize the Ag specificity of
CD8+ T cell-gated granuloma-infiltrating cells.
After LCMV superinfection, 8.5% of the CD8+ T
cells were np396 tetramer+ and
LFA-1+, indicating that LCMV-specific T cells are
recruited to S. mansoni-induced granulomas. This model also
completely underscores that peripheral activation of non-Ag-specific T
cells is sufficient to redirect their localization to the granulomatous
inflammatory site irrespective of their TCR specificity. Furthermore,
the influx of CD8+ T cells into a granuloma
normally dominated by CD4+ T cells did not result
in compromised granuloma structure at the level of histopathology.
|
, A and
C). Likewise, granulomas were not formed in
5C.C7/RAG2-/- mice bearing a single TCR
specific for PCC after infection with S. mansoni (data not
shown). Systemic activation of 5C.C7 RAG2-/-
mice by s.c. injection of 100 µg PCC in CFA 7 wk after infection had
no protective effect either (Fig. 8, B and D).
The presence of a large population of monoclonal activated T cells not
specific for schistosome ova is not sufficient for granuloma formation.
It is possible that a heterogeneous T cell population is required to
induce granulomas; however, it is more likely that T cells specific for
local ova-specific Ags are needed for granuloma formation.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production (30, 31). One of the functions of IFN- may
be to enhance the role of IL-12 in discriminating between
Th1 and Th2 subset
development (27, 32). Among the Th2
type cytokines, IL-4 (33), IL-13 (9, 34), and
IL-5 each have distinct and important roles. T cells are required, not
only for the host reaction but also for parasite development
(35). Although cytokine regulation in schistosome
granulomas is a well-established model of local cytokine circuits, much
less is known about the role of TCR specificity in the recruitment of T
cells into granulomas. Our interest in granuloma formation and regulation focuses upon requirements for T cell Ag specificity. Our data simultaneously demonstrate both the diverse and heterogeneous nature of the TCR on granuloma-infiltrating cells in response to S. mansoni infection and the capacity of those cells to clonally expand locally in response to parasite Ag. Several explanations for the diversity of the TCR repertoire can be given. Despite the probable existence of an immunodominant S. mansoni epitope to which some cells expand locally, many other parasite-specific Ags may be present at the chronic stage in the granuloma. Alternatively, a single dominant Ag could invoke a response by T cells with diverse TCRs. Finally, genetic variation of the uncloned parasite might lead to epitope differences at the granuloma level.
The evidence for Ag specificity as a requirement for T cell
localization to the granuloma lies in the unique nature of the TCR
repertoire in each granuloma both at the level of V and J chain
usage (Figs. 2 and 3), and at the CDR3 length level (Fig. 4). Previous
analysis of the TCR repertoire in the spleen of schistosome-infected
CBA/J mice had not detected any superantigen type shifts in TCR usage
of V4, V8.2, or V14 TCR- chains using flow cytometry with
V-specific Abs (10), and we have extended this analysis
to the bulk and SG level. The unique sequences arising from individual
size CDR3 bands and differences in their relative abundance (Fig. 5)
suggest local proliferative capacity. This is powerfully reinforced by
infection of AND TCR transgenic mice in which parasite-specific
nontransgenic cells are preferentially localized to the site of
inflammation (Fig. 6A) where they have a highly activated
phenotype (Fig. 6B). This is similar to work showing local
expansion at the inflammatory site in sarcoidosis (36),
and a highly activated and proliferative phenotype (IL-2R-positive) of
T cells within human periapical granulomas (37) and
Leishmania panamensis-induced granulomas
(38).
At the same time, the TCR repertoire present in individual lesions and
in the population of lesions throughout the liver is demonstrably very
heterogeneous. We cannot exclude the possibility that a fraction of
that heterogeneity arises from temporal alterations in Ag availability
arising from the nonsynchronized nature of the granulomas isolated and
the genetic heterogeneity of the S. mansoni parasites. As a
technical matter, one tends to pick the largest lesions out of the pool
under the microscope, and they may be somewhat synchronized. Another
fraction of the heterogeneity likely arises from the presence of
non-Ag-specific T cells unresponsive to SEA or other parasite egg Ags.
Our infection of AND mice suggests that the nonparasite-specific
component is composed of both activated and nonactivated T cells (Fig. 6B). We also observed that systemic activation of T cells by
a nonparasite Ag promotes their localization to granulomatous lesions
(Fig. 6C). This was consistent with previous observations of
the development of SEA-reactive lymphocytes in the spleen before their
detection in granuloma-infiltrating cells (39).
Phillips and coworkers (40) have also shown that activated
cells home to granulomas, although they did not address the role of Ag
specificity. Both groups found that splenic lymphocytes have a higher
proliferative capacity, suggesting that both systemic expansion and
activation play a role in the recruitment of lymphocytes to
granulomatous sites (39, 40).
Although there is clearly a strong role for dominant local Ags, and T cells responsive to those Ags are significantly more than a minority of the T cells present, nonparasite-specific T cells are also readily found in the granuloma. Several mechanisms are likely operative. On the one hand, Ag-specific retention, survival, and local expansion ensures the availability of parasite reactive T cells in the lesion. This is supplemented by populations of T cells recruited independently of Ag specificity. Those will include systemically activated nonparasite-specific T cells, random traffic of naive T cell populations, random and nonrandom traffic of regulatory T cells contributing to pathogenesis, lymphokine-induced traffic of T cells, and T cell populations activated by non-TCR-mediated mechanisms or by bystander activation via parasite-specific T cells.
The work of Stadecker and colleagues (22) has shown that a
single Th1 cell clone specific for SEA can induce
granuloma formation in response to embolized eggs after transfer into
non-RAG hosts and that single altered peptide epitopes can
down-regulate granuloma formation (31). We systemically
activated single Ag-specific T cells lacking parasite specificity to
test whether an abundant systemically activated monoclonal T cell
population that does not react with the parasite can induce or fail to
induce granulomas. Our data indicate that T cells recognizing local
parasitic Ags are required for initiating granulomas. Activated
nonschistosome-specific T cells home to schistosome-induced granulomas
in AND mice (Fig. 6), but alone are not sufficient to induce typical
lesions around the schistosome ova (Fig. 8). The requirement for a
single Ag specificity is currently being studied in our laboratory
using RAG-deficient hosts.
Although skews in the repertoire at local inflammatory sites have been
repeatedly reported, definitive conclusions have been hard to draw
(reviewed in Ref. 41). On the one hand, Ag-specific T
cells may dominate locally. Systemic and local Ag leads to activation
and expansion of parasite-specific T cells (39, 40)
which home to the granuloma and are preferentially retained and
expanded. This model is supported by data showing that SEA-reactive
lymphocytes are recruited to granulomas in vivo (42) and
by in vivo identification of immunodominant T cell epitopes from SEA
(30). An alternative model derives from the work of
Steinman (43) on autoimmunity. In this scenario, a minor
population of parasite-specific T cells are activated and expand at the
inflammatory site where they control a much larger randomly recruited
nonspecific infiltrate. The existence of non-Ag-specific lymphocytes in
the granuloma is indicated by the finding that SEA induces a large
fraction, but not all, of the IL-5, IL-10, or IFN- response from
granuloma-infiltrating cells in vitro assessed by the comparison of SEA
stimulation vs anti-CD3 stimulation (44). Recent
studies have also suggested that highly homogeneous populations of
eosinophils in the granuloma are primarily responsible for production
of the granuloma cytokine microenvironment (26). Our data
argue most convincingly for a broadly based T cell expansion from
outside the local inflammatory site and a broad array of selecting Ags
in the granuloma. Local expansion cannot account for all the
heterogeneity; T cells must be systemically activated and recruited. At
the same time, our data do also show clear evidence for local
expansion.
The finding that systemic activation can promote accumulation of
nonparasite-specific T cells in granulomas (Figs. 6 and 7) provides
clues for modulating granulomatous diseases. Our results suggest that
it may be feasible to induce anti-inflammatory T cells to home to
granulomas (i.e., IL-10 producing regulatory T cells). Interestingly,
in granulomatous lesions the ratio of accumulated
CD4+ and CD8+ T cells is
very consistent and characteristic for different granuloma-inducing
agents. In S. mansoni-induced granulomas the
CD4+:CD8+ ratio is 3–10:1,
in Leishmania chagasi the
CD4+:CD8+ ratio is 3–7:1
(45), and in mycobacterial granulomas the
CD4+:CD8+ ratio is 1:1
(46). These differences may reflect the different spectrum
of chemokines induced by these agents (47, 48). However,
it is also likely that during chronic infection, the ratio of activated
CD4+:CD8+ T cells is skewed
and the composition of the granuloma T cell repertoire reflects that
systemic ratio. This is supported by our LCMV coinfection experiments
(Fig. 7). LCMV induces a strong CD8+ T cell
expansion. Coinfection by LCMV of S. mansoni-infected mice
resulted in the transient appearance of a CD8+ T
cell-dominated S. mansoni granuloma. Further studies of
coinfections will be necessary to see how different infections
interfere with granulomatous responses.
In this report, we showed that SGs contain extensive and varied TCR
repertoires that differ from granuloma to granuloma. A small fraction
of T cells expand locally, but overall, peripheral T cell activation is
the major factor defining the granuloma T cell repertoire. Our ability
to change the TCR repertoire in granulomas by changing the peripheral T
cell repertoire through immunization or coinfection (Figs. 6 and 7)
indicates that systemic manipulation can regulate local inflammatory
regions. Additionally, T cells that are not specific for local Ags
accumulate in granulomas, but lesion formation requires local
parasite-specific T cells. The implications of our data have important
consequences both for understanding autoimmunity as modeled by
granulomatous lesions and for design of rational immunotherapies to
treat chronic granulomatous diseases in which Ag is present for
extended periods of time.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Laura H. Hogan, Department of Pathology, University of Wisconsin Medical School, Room 5580 Medical Science Center, 1300 University Avenue, Madison, WI 53706. E-mail address: lhhogan{at}facstaff.wisc.edu
3 Abbreviations used in this paper: SG, single granuloma; LCMV, lymphocytic choriomeningitis virus; PCC, pigeon cytochrome c; RAG2, recombinase-activating gene 2; SEA, schistosomal egg Ag; np, nucleoprotein.
Received for publication July 16, 2002. Accepted for publication September 20, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cells expressing different V genes are recruited into schistosome-induced liver granulomas. J. Immunol. 155:275.[Abstract]
chains vary as a function of the recombined germ-line segments. Proc. Natl. Acad. Sci. USA 90:4319. usage during the immune response to Schistosoma mansoni. Infect. Immun. 61:5381. treatment. Infect. Immun. 67:2201. (IFN-) receptor-deficient mice, but requires IFN- signalling to downregulate T-helper 2 responses. Immunology 97:588.[Medline]
-chain and cells expressing major histocompatibility complex class II antigen in chronic human periapical lesions. Oral Microbiol. Immunol. 13:259.[Medline]
in hepatic granulomas correlates with tissue-specific replication of Leishmania chagasi. J. Immunol. 156:2231.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  L. H. Hogan, D. O. Co, J. Karman, E. Heninger, M. Suresh, and M. Sandor Virally Activated CD8 T Cells Home to Mycobacterium bovis BCG-Induced Granulomas but Enhance Antimycobacterial Protection Only in Immunodeficient Mice Infect. Immun., March 1, 2007; 75(3): 1154 - 1166. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. O. Co, L. H. Hogan, J. Karman, E. Heninger, S. Vang, K. Wells, Y. Kawaoka, and M. Sandor Interactions between T Cells Responding to Concurrent Mycobacterial and Influenza Infections J. Immunol., December 15, 2006; 177(12): 8456 - 8465. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Heninger, L. H. Hogan, J. Karman, S. Macvilay, B. Hill, J. P. Woods, and M. Sandor Characterization of the Histoplasma capsulatum-Induced Granuloma. J. Immunol., September 1, 2006; 177(5): 3303 - 3313. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Wen, C. M. Hogaboam, J. Gauldie, and S. L. Kunkel Severe Sepsis Exacerbates Cell-Mediated Immunity in the Lung Due to an Altered Dendritic Cell Cytokine Profile Am. J. Pathol., June 1, 2006; 168(6): 1940 - 1950. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Edwards, O. Buchatska, M. Ashton, M. Montoya, Q. D. Bickle, and P. Borrow Reciprocal Immunomodulation in a Schistosome and Hepatotropic Virus Coinfection Model J. Immunol., November 15, 2005; 175(10): 6275 - 6285. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F Chen, G Sethi, R Goldin, A R Wright, and C J Lacey Concurrent granulomatous Pneumocystis carinii and Mycobacterium xenopi pneumonia: an unusual manifestation of HIV immune reconstitution disease Thorax, November 1, 2004; 59(11): 997 - 999. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. B. Stavitsky Regulation of Granulomatous Inflammation in Experimental Models of Schistosomiasis Infect. Immun., January 1, 2004; 72(1): 1 - 12. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
