| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Centenary Institute of Cancer Medicine and Cell Biology, Newtown, Australia
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In contrast to most nonlymphoid tissues, the liver possesses an unusual endothelial barrier consisting of a single layer of fenestrated endothelial cells that do not form tight junctions (3, 4, 5, 6, 7). Moreover, hepatocytes and endothelial cells are not separated by a basement membrane, but define the space of Disse, a compartment communicating directly with the blood vessel lumen. It is therefore possible that naive CD8+ T lymphocytes circulating via the blood can interact directly with hepatocytes through T cell cytoplasmic extensions penetrating the space of Disse. Hepatocyte/T cell contact could occur either via the gap between two adjacent endothelial cells, or through the endothelial cell fenestrae. Experimental data suggest that liver permeability to naive (8) and differentiated CD8+ T cells (5) differs from that of other organs. However, none of these studies has addressed whether naive CD8+ T cells can be directly activated in the liver. This question has wide relevance with regard to liver immunobiology, viral hepatitis, and liver transplantation, as the liver has been implicated in inducing T cell tolerance (7, 9, 10, 11).
To address this issue, we investigated the early events of activation and proliferation of TCR transgenic CD8+ T cells injected into transgenic mice expressing the relevant alloantigen in the liver. Selective retention of autoreactive T cells was seen in the liver within minutes of transfer. Expression of the very early activation marker CD69 could be detected in the liver, but not blood, as early as 2 h posttransfer. Activated cells started to divide and recirculate via the blood at 24 h. To our knowledge, this is the first report demonstrating that naive T cells can undergo primary activation outside lymphoid organs. This study further challenges the concept that danger and inflammation associated with tissue damage are essential for T cell activation and migration into the liver (12, 13).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All mice were maintained in the Centenary Institute animal facility under specific pathogen-free conditions. Met-Kb transgenic mice expressing the H-2Kb gene in hepatocytes under the control of the sheep metallothionein promoter have been described previously (14) and were a gift of Grant Morahan and J. F. A. P. Miller (Walter and Eliza Hall Institute (WEHI), Melbourne, Australia). No zinc induction of Met-Kb mice was performed. The 178.3 transgenic mice expressing the H-2Kb gene under the control of its own promoter were provided by William Heath (WEHI) and have been described previously (15). The Des-TCR transgenic mice expressing an H-2Kb-specific TCR (16) identifiable by a clonotypic Ab, Désiré (17), were kindly provided by B. Arnold, G. Schönrich, and G. J. Hämmerling (Deutsches Krebsforschungs Zentrum, Heidelberg, Germany). Met-Kb, 178.3, and Des-TCR transgenic mice were maintained by breeding with B10.BR mice obtained from the Animal Resources Center (Perth, Western Australia, Australia). A Ly-5.1 allele was introduced into Met-Kb mice by crossing with a line of H-2k Ly-5.1 homozygous mice bred in house from F2 crosses between B6.SJLPtprca (H-2b, Ly-5.1) and B10.BR (H-2k, Ly-5.2) mice purchased from the Animal Resources Center.
Bone marrow chimeras and adoptive transfer of CFSE-labeled Des-TCR T cells
Tibiae and femora from B10.BR and Met-Kb mice were flushed with RPMI 1640 (JRH Biosciences, Lenexa, KS) supplemented with 10% FCS (Life Technologies, Grand Island, NY), 2 mM L-glutamine (JRH Biosciences), and 50 µM 2-ME (ICN Biomedical, Aurora, OH) (referred to as T cell medium). Cells were washed twice and incubated at 5 x 107/ml for 30 min at 4°C with a mixture of anti-CD4 (RL172.4) (18) and anti-CD8 (3.168) (19) hybridoma supernatants. Cells were then incubated at 5 x 107cells/ml in T cell medium plus young rabbit complement (C-SIX Diagnostics, Mequon, WI) for an additional 30 min at 37°C. Bone marrow cells were washed three times before being counted, and 5 x 106 to 107 cells were injected i.v. into 7- to 8-wk-old control nontransgenic B10.BR or Met-Kb mice that had been lethally irradiated at 1300 rad. Bone marrow chimeric mice were used in experiments 6–8 wk after reconstitution.
Single cell suspensions of pooled lymph node cells from Des-TCR transgenic mice were made by pressing the tissue through an 80-µm mesh sieve and washing twice with T cell medium. In some experiments, CD8+ T cells were purified using magnetic beads, as previously described (20). Cells were incubated for 10 min at 37°C with 5 µM of CFSE, purchased from Molecular Probes (Eugene, OR) in RPMI 1640 without FCS, as described elsewhere (20). A total of 1.5 x 107 CFSE-labeled lymph node cells, of which 30% were CD8+ T cells expressing Des-TCR, was injected into the lateral tail vein of recipient mice.
DC preparation and immunization protocol
DCs were prepared as previously described (21)
using a modification of the protocol of Vremec and Shortman
(22). Briefly, spleens from 178.3 mice were digested with
collagenase/EDTA and centrifuged over a Nycodenz density gradient
(
= 1.077). Cells in the lighter fraction
were positively selected for CD11c using N418 mAb (23),
anti-hamster FITC (Caltag, San Francisco, CA), and anti-FITC
Multisort microbeads (Miltenyi Biotech, Auburn, CA), followed by
passage over a MACS column (Miltenyi Biotech). The cell fraction
retained by the column contained 40–50% DCs. The rest comprised a
mixture of B cells, macrophages, and monocytes. Cells were resuspended
in medium, and 3 x 106 cells in 25 µl
were injected in each hind footpad. Recipients were B10.BR mice that
had been injected with 1.5 x 107
CFSE-labeled Des-TCR lymph node cells 3 days previously.
Isolation of lymph node, spleen, peripheral blood, lung, and liver lymphocytes for flow cytometry
Single cell suspensions of lymph nodes (celiac, popliteal, inguinal, axillary, subscapsular, cervical, aortic, and mesenteric) and spleens were prepared as described above. Blood was collected into Alsever’s solution (114 mM dextrose, 27 mM sodium citrate, 71 nM NaCl, pH 6.1) by cardiac puncture after killing the mouse. PMBC were prepared by lysing RBC with red cell removal buffer (145 mM NH4Cl, 0.1 mM EDTA, 12 mM NaHCO3; two washes). PBMC were then washed once with T cell medium and analyzed by flow cytometry.
To prepare lung lymphocytes, the lungs were first perfused by injecting 25 ml of PBS into the heart before removing the lungs. They were then gently pressed through an 80-gauge mesh sieve. The cells were centrifuged at 400 x g for 10 min at 4°C, and the pellet was resuspended in Percoll solution (36% of isotonic Percoll in PBS). The suspension was centrifuged at 500 x g for 10 min at 4°C, and the supernatant was gently removed. The lymphomyeloid and RBC contained in the pellet were resuspended in red cell removal buffer before being counted.
Liver, kidney, and pancreas lymphocytes were purified according to a protocol similar to the one used for lungs, as described previously (8). The lymphomyeloid and RBC contained in the pellet were resuspended in red cell removal buffer before being counted.
Abs and flow cytometry analysis
Anti-CD69 (H1.2F3 (24)), anti-CD44 (IM7.81 (25)), clonotypic anti-Des TCR (Désiré, a gift of A. M. Schmitt-Verhulst, Center d’Immunologie de Marseille-Luminy, Marseille, France (17)), and anti-Ly-5.1 (A20.1, originally provided by E. A. Boyse, Memorial Sloan-Kettering Cancer Center, New York, NY) mAbs were purified using a protein G affinity column and conjugated to either biotin or FITC. PE- and FITC-conjugated CD8 and CD4 mAb, streptavidin PE, and streptavidin tricolor were purchased from Caltag, while streptavidin Texas Red was purchased from Molecular Probes. For flow cytometry, 106 to 2 x 106 cells were incubated with 2.4G2 (26) supernatant to block Fc-mediated binding before adding Abs in a final volume of 50 µl for 30 min. Stained cells were washed with PBS containing 5% FCS and 5 mM sodium azide.
For annexin V staining, 4 x 106 cells were stained with anti-CD8 PE, Des-biotin plus streptavidin Texas Red (Molecular Probes), and anti-Ly-5.1 conjugated with allophycocyanin. Following further washing as above, annexin V staining (27) was conducted using a commercial kit (annexin V-fluos; Boehringer Mannheim, Indianapolis, IN), according to the manufacturer’s instructions.
Flow cytometry acquisition was performed using a FACScan (BD Biosciences, Franklin Lakes, NJ) or a FACStarplus (BD Biosciences). Analysis was performed using CellQuest (BD Biosciences) or Flow Jo software (TreeStar, San Carlos, CA) on a Macintosh computer (Apple Computers, Cupertino, CA).
Measurement of serum transaminase levels
Serum alanine aminotransferase levels were measured with an N Hitachi 917 automatic chemistry analyzer (Roche Diagnostics Corporation Laboratory Systems, Indianapolis, IN).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We previously demonstrated that adoptively transferred Des-TCR T
cells infiltrate the liver of irradiated, thymectomized
Met-Kb mice reconstituted with B10.BR bone marrow
(manipulated Met-Kb mice) (8).
Des-TCR T cells then undergo cell division in response to
H-2Kb, and are subsequently deleted. To determine
whether irradiation is a prerequisite for this response, lymph node
cells from Des-TCR mice were CFSE labeled and injected i.v. into
unmanipulated Met-Kb mice. Multiply divided donor
T cells could be detected in the liver, but not in other nonlymphoid
sites (kidney, pancreas, lung), 2 days after transfer (Fig. 1
A), consistent with previous
observations in manipulated Met-Kb mice
(8 , and P. Bertolina, unpublished results). At this time
point, donor cells had undergone up to four cell divisions, as
indicated by the stepwise loss of CFSE. The CFSE vs CD8 profiles in
Fig. 1A also showed a population of
CD8+ CFSE-negative cells derived from the host
mice. These cells were Des-TCR negative, whereas all the donor
CD8+ T cells expressed the Des-TCR. For this
reason, CD8 was used to define autoreactive CFSE+
donor T cells in some subsequent experiments. Consistent with previous
data from manipulated Met-Kb mice, unmanipulated
Met-Kb mice injected with Des-TCR T cells also
developed a transient hepatitis peaking at day 5–6 (Fig. 1B) and most adoptively transferred T cells were deleted
following proliferation (Fig. 1C). Deletion was observed 14
days after transfer in both lymph nodes and liver (Fig. 1D)
and was almost total at day 30 (Fig. 1E). B10.BR control
mice injected with the same number of Des-TCR lymph node cells did not
develop hepatitis (Fig. 1B) and transferred cells neither
proliferated (Fig. 1A) nor underwent deletion, because they
could still be detected 30 days after transfer (Fig. 1, C-E). These results suggested that preirradiation or
inflammation was not required for specific T cell activation and
proliferation in the liver of Met-Kb mice, which
resulted in transient hepatitis and subsequent T cell deletion.
However, because it is known that H-2Kb is
expressed at low levels on bone marrow-derived cells in lymph nodes and
spleen of unmanipulated Met-Kb mice
(8), we investigated the kinetics of T cell activation to
distinguish between direct activation in the liver and migration to the
liver after primary activation in lymphoid tissues.
|
To determine whether CFSE+
CD8+ cells were specifically retained in liver of
Met-Kb mice, CD8+ Des-TCR T
cells were purified, CFSE labeled, and transferred to unmanipulated
Met-Kb or control B10.BR mice. The total number
of CFSE+ CD8+ cells in each
organ was calculated at time points between 10 and 70 min. The liver
was the only organ showing a relative increase in the number of
CFSE+ CD8+ cells in
Met-Kb mice compared with syngeneic B10.BR
controls (Fig. 2A). Despite
expression of the H-2Kb transgene in the pancreas
and kidney of Met-Kb mice (14), very
few transgenic T cells were detected in these organs, or in the lungs,
suggesting that they were not capable of Ag-specific retention of
transgenic T cells. Total CD8+ Des-TCR T cell
numbers in the spleen, blood, lung, kidney, and pancreas of
Met-Kb and B10.BR mice were similar, whereas
numbers of CD8+ Des-TCR T cells in the lymph
nodes of Met-Kb mice were lower than in B10.BR
controls, possibly due to retention of transgenic cells in the
Met-Kb liver.
|
B and data not shown).
The livers did, however, contain a higher percentage (10–15%) of
CD8+ Des-TCR annexin V+
apoptotic T cells than lymph nodes (5%) (Fig. 2B),
suggesting that some injected cells died nonspecifically in the liver.
Nevertheless, the vast majority of autoreactive T cells migrating to
the liver of Met-Kb mice 1 h posttransfer
were not undergoing apoptosis. Furthermore, expression of CD8 and
Des-TCR was not down-regulated within the liver at that time (Fig. 2C), in contrast to the phenotype previously ascribed to
apoptotic T cells in the liver (28). These results were
confirmed by immunochemistry using the TUNEL technique, which showed no
significant difference in the number of apoptotic cells in
Met-Kb and B10.BR recipients at 1, 2, 5, and
24 h posttransfer (data not shown). In addition,
CD8+ Des-TCR T cells purified from the liver of
Met-Kb mice 1 h after transfer and
reinjected into a second nontransgenic host survived and recirculated
similarly to naive Des-TCR T cells for at least 24 h (data not
shown), indicating that they had not accumulated in the liver as a
result of irreversible commitment to apoptosis. These results suggest that viable Des-TCR T cells were retained specifically in the liver of Met-Kb mice within minutes of transfer, as a result of contacting the Ag in the liver without being preactivated in the lymph nodes.
Naive autoreactive CD8+ T cells in the liver and lymph nodes express activation markers within 2 h of transfer
To confirm that Des-TCR T cells found in the liver of
Met-Kb mice at early time points were undergoing
activation in situ, we investigated their expression of the very early
activation marker CD69. The donor autoreactive T cells (identified as
CFSE+ CD8+ cells)
increased expression levels of CD69 in the liver between 2 and
16 h of transfer, confirming that autoreactive cells migrating
to the liver were undergoing activation (Fig. 3A). Activation was specific,
because no increase in expression of CD69 was detected in control
B10.BR mice (Fig. 3A), although the mean levels of CD69
expression by Des-TCR T cells were always higher in the liver of
control animals compared with lymphoid organs.
|
A). Interestingly, the
appearance of CD69+ T cells in the spleen was
delayed 
2 h when compared with the liver and lymph nodes. This delay
was confirmed in several independent experiments.
As in most TCR transgenic mice, some Des-TCR T cells express an
activated/memory phenotype due to incomplete allelic exclusion of
endogenous 
-chains (20, 29). It was therefore important
to determine whether the activated Des-TCR T cells found in the liver
and lymph nodes of Met-Kb mice 2 h after
transfer were derived from CD44low or
CD44high cells. Both naive
(CD44low) and activated
(CD44high) cells were specifically retained in
the liver at 2 h posttransfer (data not shown). However, CD69 was
preferentially expressed by the CD44low subset of
transferred CD8+ CFSE+
cells (Fig. 3B). These results confirmed previous in vitro
data demonstrating that CD44low Des-TCR T cells
were preferentially activated by H-2Kb-expressing
hepatocytes (20).
To test whether Des-TCR CD8+
CD69+ T cells within the liver 2 h after
transfer had previously been activated in lymph nodes and then
recirculated via efferent lymph and blood, the expression of CD69 by
autoreactive T cells in the blood was measured at 1 and 2 h
posttransfer. There was no increase in expression of CD69 by Des-TCR
CD8+ T cells in the blood of
Met-Kb compared with control B10.BR mice (Fig. 3C). As a direct test of whether cells activated in lymphoid
organs could recirculate within 2 h of activation, expression of
H-2 Kb was restricted to lymphoid organs by using
B10.BR mice that were lethally irradiated and reconstituted with
Met-Kb bone marrow (Met-Kb
bm
BR) (Fig. 3, D-E). Control chimeras were B10.BR and
Met-Kb mice reconstituted with B10.BR bone
marrow. The CD69 profiles in Fig. 3D (expressed as MFI in
Fig. 3E) indicate that increased expression of CD69 by
Des-TCR T cells was detected in lymph nodes, and to a lesser extent
spleen, of both Met-Kb bmBR and BR
bmMet-Kb chimeras, whereas up-regulation of
CD69 in the liver was restricted to BR
bmMet-Kb chimeras, confirming that H-2
Kb expression by the liver itself was required
for Des-TCR T cell activation in that organ. The CD69 profiles
also indicated that a larger percentage of Des-TCR T cells expressed
high levels of CD69 in the liver than the lymph nodes, explaining why
the liver MFI values in Met-Kb mice were
consistently higher than those of the lymph nodes (Fig. 3A).
Collectively, these experiments suggested that adoptively transferred Des-TCR T cells found in the liver of unmanipulated Met-Kb mice within 2 h of transfer had undergone activation in situ.
Dividing cells in the liver and blood 1 day after transfer
Comparison of T cell activation in the liver, blood, lymph nodes,
and spleen 24 h after adoptive transfer (Fig. 4A) indicated that the liver
had the highest percentage of blasts and divided cells at that time.
Activation-dependent up-regulation of CD8 was also most apparent in the
liver (Fig. 4B). The blood contained a significant number of
recirculating CD69-expressing blasts, although the level of CD69
expression per cell was lower than in either the liver or lymph nodes
(compare Figs. 3E and 4C). By 48 h
posttransfer, high numbers of cells, which had divided up to four
times, were detected within the liver, blood, spleen, and lymph nodes
(Fig. 4D). Beyond 48 h, the total number of transgenic
CD8+ T cells increased exponentially in all
tissues in which activation and division of transgenic cells were
observed at earlier times (Fig. 1D and data not shown). As
mentioned above, mice developed hepatitis peaking at day 5 and
resolving after a few days, when autoreactive T cells were deleted (see
Fig. 1).
|
The above data did not address the question of whether the
activated self-reactive T cells found in the blood within 1 day of
transfer were derived from the liver or peripheral lymphoid tissues.
To determine when Des-TCR T cells activated in lymph nodes could
first enter the bloodstream, CFSE-labeled Des-TCR T cells were
adoptively transferred into B10.BR mice 3 days before footpad
immunization with purified DCs from 178.3 transgenic mice expressing
the H-2Kb molecule under the control of its
endogenous promoter. Previous results from our laboratory have shown
that myeloid DCs injected into the footpads migrated rapidly to the
draining popliteal nodes and generated a local T cell response
(21). Consistent with these observations, expression of
CD69 at 24 h was restricted to the lymph nodes draining the site
of injection of H-2Kb-bearing DCs (Fig. 5). The first cell divisions were
observed 48 h after DC injection and were also localized to the
draining lymph nodes (Fig. 5). By day 3, large numbers of proliferating
CFSElow cells were found in the blood, liver, and
other sites, suggesting that T cells activated in popliteal lymph nodes
began to recirculate via the blood to the liver and spleen between 2
and 3 days after initial activation (Fig. 5). Moreover, most cells
recirculating on day 3 had undergone more than five divisions (Fig. 5),
consistent with division at a sequestered site before recirculation.
This pattern of division and recirculation was quite distinct from that
seen in unmanipulated Met-Kb mice (Fig. 4) and,
although we cannot exclude that higher H-2Kb
expression by 178.3 DCs may have affected sequestration within lymph
nodes, these results suggest that the dividing cells in the blood of
Met-Kb mice at 24–48 h were derived from cells
activated in the liver, but not retained efficiently at that
site.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In addition to its postulated role as a site of extrathymic
differentiation (30, 31), it has been proposed that, under
some circumstances, the liver can act as a preferential disposal site
for activated CD8+ T cells undergoing apoptosis,
either because apoptotic cells migrate to the liver to die, or because
activated T cells receive an apoptotic signal within the liver tissue
itself. This process has been postulated to require signaling via ICAM,
but not TCR (32). The hypothesis was originally based on
experiments showing that i.v. injection of specific peptide into TCR
transgenic mice led to the accumulation of apoptotic peptide-specific T
cells in the liver (28). Results presented in this work do
not contradict this hypothesis, but suggest that in addition to its
function as a "T cell graveyard," the liver may play a role in
primary activation of CD8+ T cells specific for
Ags presented within the liver, including peptides delivered by the
i.v. route. In the present study, selective retention of self-reactive
CD8+ T cells by the liver was apparent within
minutes of transfer (Fig. 2A) and did not require T cells to
be apoptotic, as judged by absence of annexin V (Fig. 2B)
and TUNEL staining (data not shown). Moreover, cells isolated from the
liver were viable for at least 24 h after transfer into a second
nontransgenic host (data not shown), indicating that they were not
committed to early apoptotic death. However, our previous data
indicated that CD8+ T cells activated by purified
hepatocytes in vitro underwent premature death by neglect, consistent
with the high number of apoptotic cells at later time points in the
original studies (28). Moreover, the final fate of
autoreactive CD8+ T cells in unmanipulated
Met-Kb hosts was deletion (Fig. 1C).
Although the earliest appearance of activated
CD8+ Des-TCR cells occurred simultaneously in the
liver and lymph nodes, our data suggest that autoreactive T cells
isolated from the liver were not simply recirculating cells previously
activated in the lymph nodes. Such a scenario would require cells to
leave the nodes within 2 h of activation. Not only were activated
Des-TCR T cells not found in the blood within 2 h of transfer
(Fig. 3, C and E), but T cells activated in the
nodes of mice not expressing H-2Kb in the liver
(Met-Kb bmBR chimeras) were not detected in
either the blood or liver 2 h after transfer (Fig. 3, D
and E). When activation of T cells was localized to draining
lymph nodes by injection of myeloid DCs into the footpad, the earliest
evidence of recirculation was seen on day 3 (Fig. 5). Viewed in toto,
these results indicate that T cells activated in the lymph nodes of
Met-Kb mice are most unlikely to have contributed
to the blood T cell pool during the first 2 days after transfer. Thus,
we conclude that activation of T cells occurs independently in liver
and lymph nodes of unmanipulated Met-Kb
mice.
Interestingly, a few divided CD69+ Des-TCR T
cells were detected in the blood 24 h posttransfer (Fig. 4, A-C). Because Des-TCR T cells activated in the lymph nodes
begin to recirculate only after day 2 (Fig. 5), these results suggest
that T cells activated in the liver were not efficiently retained in
that organ, but recirculated via the blood within 24 h of
activation. This interpretation is supported by the presence in the
blood of cells that had undergone up to four divisions 48 h after
transfer into Met-Kb mice (Fig. 4D),
in contrast to their absence in B10.BR mice given
H-2Kb-expressing DCs in the footpad (Fig. 5).
The nature of the APC responsible for the activation of T cells in the liver was not addressed in the current study. Although we cannot exclude the involvement of other hepatic cells, including DCs or Kupffer cells, the most likely candidate is the hepatocyte. Not only is the metallothionein promoter preferentially expressed within hepatocytes, but we have also recently demonstrated the capacity of hepatocytes to act as efficient APC for Des-TCR T cells in vitro (20). In vivo interactions between hepatocytes and naive T cells would be facilitated by the unique structure of the hepatic endothelium, in which sinusoidal endothelial cells form only an incomplete barrier between hepatocytes and PBMC. The space of Disse, defined by hepatocytes and endothelial cells, opens directly into the sinusoidal lumen, and electron microscopic pictures indicate that cytoplasmic extensions of mononuclear cells, in particular Kupffer cells, penetrate this space and directly contact hepatocytes (3). It is therefore possible that naive T cells may contact hepatocytes in a similar way, as has been previously suggested by others for CTL (5). Likewise, a recent report suggested that lymphocytes recirculating via the blood enter the parenchymal space of the liver and, conversely, that extrathymic T cells generated in the parenchymal space migrate into the blood (33). Direct Ag presentation to T cells by hepatocytes expressing MHC class I and ICAM-1 (20), but not MHC class II, may explain why CD8+ T cells are more prone than CD4+ T cells to retention and death within the liver (32). Clearly, the ability of cytotoxic CD8+ T cells to recognize MHC class I may be central to their ability to induce hepatic damage (34).
The nature of the APC involved in intranodal activation of transgenic T
cells in Met-Kb mice also remains unknown. Bone
marrow-derived APC seem to play a role, because B10.BR mice
reconstituted with Met-Kb bone marrow exhibited
H-2Kb-specific activation in lymph nodes (but not
liver) (Fig. 3, D and E). Although expression of
H-2Kb in Met-Kb lymphoid
tissues could not be detected by flow cytometry of either total lymph
node cells or purified splenic myeloid DCs (8), or by
activation of Des-TCR T cells in vitro (data not shown), it has been
established that high avidity H-2Kb-specific T
cells are deleted in the thymus of unmanipulated
(Met-Kb x Des-TCR) double-transgenic mice,
consistent with low level expression by bone marrow-derived APC
(8).
The findings described in this work are at variance with those of Limmer et al. (13), in which Des-TCR T cells required either exogenous IL-2 or preexisting inflammation to cause hepatitis in recipients expressing H-2Kb in the liver. The data in that study were interpreted as indicating that inflammation is crucial for entry of T cells into the liver. However, the mice were maintained on an H-2kxd background in which cross-reaction of the Des-TCR with the H-2Kd molecule in the thymus or peripheral lymphoid organs deleted high avidity cells (35 and unpublished results). Thus, the study demonstrated only that low avidity cells are not activated directly in the liver, but entered the liver in response to liver inflammation. Moreover, T cell differentiation outside the liver was required, as indicated by hepatitis peaking at day 10, 5 days later than the peak of hepatitis in the model described in this work. Our study made use of Des-TCR mice maintained on an H-2k background in which high avidity cells were not deleted. Although other factors, such as different levels and sites of H-2Kb transgene expression, may have contributed to the different findings in the two models, we favor the view that differences in T cell avidity are principally responsible for the divergent conclusions reached in these two models. This important question is being currently addressed by testing the responses of high and low avidity cells in the Met-Kb model.
Primary activation in the liver challenges a major immunological dogma
postulating that naive T cell activation occurs only in lymphoid organs
(1) and that danger and/or inflammation, following tissue
damage, is essential for primary T cell activation (12, 13). To our knowledge, this is the first report demonstrating
that primary activation of naive T cells can be initiated in a
nonlymphoid organ. We predict that this type of activation will be
restricted to the liver, and will apply only to MHC class I-restricted
T cells. Indeed, it has already been established that Des-TCR T cells
ignore H-2Kb Ag when it is expressed by the 
cells of the pancreas, an organ with an intact blood endothelial
barrier (36). In this case, infiltration and subsequent
diabetes required local IL-2 production (36) or prior
priming of T cells by contact with
H-2Kb-expressing splenocytes
(15).
Taken together with our previous in vitro data indicating that CD8+ T cells can be directly activated by hepatocytes, but then die by neglect due to a lack of costimulation (20, 37), the in vivo data reported in the present study suggest the possibility that primary recognition of Ag within the liver may commit class I-restricted T cells to deletion, inducing a state of tolerance in the animal (8). This mechanism may contribute to the known tolerogenic capacity of histoincompatible liver transplants, which are accepted spontaneously in several species (9, 11). Moreover, acceptance of liver allografts induces specific tolerance to nonliver grafts from the same donor strain (10). Although it is likely that tolerance to liver allografts is a complex and multifactorial phenomenon, possibly involving direct Ag presentation by donor-derived nonhepatocyte APC to both recipient CD4+ and CD8+ T cells, Ag uptake or cross-presentation of liver Ags by recipient-derived APC to recipient CD4+ or CD8+ T cells, and interactions between donor and recipient lymphocytes, our data demonstrate that direct Ag presentation by liver cells to recipient CD8+ T cells could also potentially play an important role. By selectively activating and deleting high avidity CD8+ T cells very early after transplantation, the liver would dispose of the majority of host cells responsible for early acute rejection. Regulatory mechanisms, probably involving both CD4+ and CD8+ T cells, could supervene at later times.
The current findings also raise intriguing possibilities regarding the immunopathogenesis of viral hepatitis. The pathological manifestations of chronic hepatitis B and C appear to be mediated, at least in part, by an ongoing immune response that fails to clear the virus (38, 39, 40, 41). Although CTL directed to putative viral epitopes may be detected in the blood and the liver of patients with chronic hepatitis C (39, 42, 43, 44, 45, 46), the avidity of such clones has not been investigated. The possibility exists that higher avidity clones may be deleted as a result of primary activation in the liver, leading to ongoing immune reactivity mediated by cells of insufficient avidity to clear the virus. This hypothesis requires further examination as a possible mechanism in the pathogenesis of these major causes of morbidity.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Barbara Fazekas de St. Groth, Centenary Institute of Cancer Medicine and Cell Biology, Locked Bag Number 6, Newtown NSW 2042, Australia.
3 Abbreviations used in this paper: DC, dendritic cell; FSC, forward scatter; MFI, mean fluorescence intensity; PI, propidium iodide.
Received for publication September 28, 2000. Accepted for publication February 22, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
+ dendritic cells generate an immune response after subcutaneous injection without homing to the draining lymph nodes. J. Exp. Med. 189:593. chains on the surface of T cells in T cell receptor transgenic mice. J. Exp. Med. 178:1807. chain expression by adult human hepatic T cells: evidence for extrathymic T cell maturation. Eur. J. Immunol. 26:3114.[Medline]
This article has been cited by other articles:
|
|
 |
|
  E. D. Reynoso and S. J. Turley Unconventional antigen-presenting cells in the induction of peripheral CD8+ T cell tolerance J. Leukoc. Biol., October 1, 2009; 86(4): 795 - 801. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. S. Guy, P. M. Mulrooney-Cousins, N. D. Churchill, and T. I. Michalak Intrahepatic Expression of Genes Affiliated with Innate and Adaptive Immune Responses Immediately after Invasion and during Acute Infection with Woodchuck Hepadnavirus J. Virol., September 1, 2008; 82(17): 8579 - 8591. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. K. Polakos, I. Klein, M. V. Richter, D. M. Zaiss, M. Giannandrea, I. N. Crispe, and D. J. Topham Early Intrahepatic Accumulation of CD8+ T Cells Provides a Source of Effectors for Nonhepatic Immune Responses J. Immunol., July 1, 2007; 179(1): 201 - 210. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Morimoto, X. Tan, R. M. Teague, C. Ohlen, and P. D. Greenberg Induction of Tolerance in CD8+ T Cells to a Transgenic Autoantigen Expressed in the Liver Does Not Require Cross-Presentation J. Immunol., June 1, 2007; 178(11): 6849 - 6860. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. D. Klonowski, A. L. Marzo, K. J. Williams, S.-J. Lee, Q.-M. Pham, and L. Lefrancois CD8 T Cell Recall Responses Are Regulated by the Tissue Tropism of the Memory Cell and Pathogen J. Immunol., November 15, 2006; 177(10): 6738 - 6746. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Oikawa, Y. Kamimura, H. Akiba, H. Yagita, K. Okumura, H. Takahashi, M. Zeniya, H. Tajiri, and M. Azuma Preferential Involvement of Tim-3 in the Regulation of Hepatic CD8+ T Cells in Murine Acute Graft-versus-Host Disease J. Immunol., October 1, 2006; 177(7): 4281 - 4287. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Bertoletti and A. J. Gehring The immune response during hepatitis B virus infection J. Gen. Virol., June 1, 2006; 87(6): 1439 - 1449. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M B De Biasio, N Periolo, A Avagnina, M T Garcia de Davila, M Ciocca, J Goni, E de Matteo, C Galoppo, M C Canero-Velasco, H Fainboim, et al. Liver infiltrating mononuclear cells in children with type 1 autoimmune hepatitis J. Clin. Pathol., April 1, 2006; 59(4): 417 - 423. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Klein and I. N. Crispe Complete differentiation of CD8+ T cells activated locally within the transplanted liver J. Exp. Med., February 21, 2006; 203(2): 437 - 447. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Lim, R. M. Sutherland, Y. Zhan, G. Deliyannis, L. E. Brown, and A. M. Lew Targeting CD45RB alters T cell migration and delays viral clearance Int. Immunol., February 1, 2006; 18(2): 291 - 300. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Mohrs, D. P. Harris, F. E. Lund, and M. Mohrs Systemic Dissemination and Persistence of Th2 and Type 2 Cells in Response to Infection with a Strictly Enteric Nematode Parasite J. Immunol., October 15, 2005; 175(8): 5306 - 5313. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Sun, B. Tumurbaatar, J. Jia, H. Diao, F. Bodola, S. M. Lemon, W. Tang, D. G. Bowen, G. W. McCaughan, P. Bertolino, et al. Parenchymal Expression of CD86/B7.2 Contributes to Hepatitis C Virus-Related Liver Injury J. Virol., August 15, 2005; 79(16): 10730 - 10739. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Chen, P. Tabaczewski, S. M. Truscott, L. Van Kaer, and I. Stroynowski Hepatocytes Express Abundant Surface Class I MHC and Efficiently Use Transporter Associated with Antigen Processing, Tapasin, and Low Molecular Weight Polypeptide Proteasome Subunit Components of Antigen Processing and Presentation Pathway J. Immunol., July 15, 2005; 175(2): 1047 - 1055. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Golden-Mason, D. C. Douek, R. A. Koup, J. Kelly, J. E. Hegarty, and C. O'Farrelly Adult Human Liver Contains CD8pos T Cells with Naive Phenotype, but Is Not a Site for Conventional {alpha}{beta} T Cell Development J. Immunol., May 15, 2004; 172(10): 5980 - 5985. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Djilali-Saiah, P. Lapierre, S. Vittozi, and F. Alvarez DNA Vaccination Breaks Tolerance for a Neo-Self Antigen in Liver: A Transgenic Murine Model of Autoimmune Hepatitis J. Immunol., November 1, 2002; 169(9): 4889 - 4896. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Reignat, G. J.M. Webster, D. Brown, G. S. Ogg, A. King, S. L. Seneviratne, G. Dusheiko, R. Williams, M. K. Maini, and A. Bertoletti Escaping High Viral Load Exhaustion: CD8 Cells with Altered Tetramer Binding in Chronic Hepatitis B Virus Infection J. Exp. Med., May 6, 2002; 195(9): 1089 - 1101. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
