| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Thomas E. Starzl Transplantation Institute and Department of Surgery, and
Department of Molecular Genetics and Biochemistry, University of Pittsburgh, Pittsburgh, PA 15213
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Intracellular expression
of IL-10 was very low in both BM DC- and liver DCp-stimulated
CD4+ T cells. Only stimulation by liver DCp was associated
with IL-10 secretion in primary MLR. Notably, these liver DCp
cocultured with allogeneic T cells stained strongly for IL-10.
Following local (s.c.) injection in allogeneic recipients, both BM DC
and liver DCp homed to T cell areas of draining lymph nodes and spleen,
where they were readily detected by immunohistochemistry up to 2 wk
postinjection. Liver DCp induced clusters of IL-10- and IL-4-secreting
mononuclear cells, whereas Th2 cytokine-secreting cells were not
detected in mice injected with mature BM DC. By contrast, comparatively
high numbers of IFN-+ cells were induced by BM DC.
Modulation of Th2 cytokine production by donor-derived DCp may
contribute to the comparative immune privilege of hepatic
allografts. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Following organ transplantation, donor interstitial DC migrate to secondary lymphoid tissue, where they interact with specific, donor-reactive T cells (12, 13, 14, 15, 16, 17). Based on observations in rodent kidney or heart transplantation, these donor APC have been regarded historically, as instigators of rejection (14, 18). More recent evidence suggests that persistence of donor-derived DC in recipient lymphoid and nonlymphoid tissue, as occurs in nonimmunosuppressed murine liver allograft recipients, may be linked to the development of donor-specific tolerance (11, 15, 16, 17, 19). Immature myeloid DC propagated in GM-CSF from normal mouse liver, a hemopoietic organ, are deficient in cell surface costimulatory molecules. They migrate in vivo to T cell areas of secondary lymphoid tissue, where they persist for weeks in allogeneic recipients (20, 21). These liver-derived DC progenitors (DCp) can prolong allograft survival (22), a property shared with immature myeloid DC propagated from rodent BM (23, 24, 25).
DC tolerogenicity was first reported in the context of central (intrathymic) tolerance (26, 27, 28). More recently, evidence has accumulated for a role of DC in peripheral tolerance in various experimental models (2). In addition, several agents, including UVB radiation (29), IL-10 (30, 31, 32, 33), TGF-ß (34, 35), or the chimeric fusion protein cytotoxic T lymphocyte Ag 4-Ig (36, 37), which blocks the B7-CD28 costimulatory pathway of T cell activation, have been shown to confer tolerogenic properties on DC. Mechanisms reported to underlie the capacity of DC to subvert T cell responses include the induction of anergy, activation-induced cell death, regulatory cells, veto function, and immune deviation toward a predominant Th2 cell response (2, 38). It is accepted that alloantigen-specific Th1 cells initiate allograft rejection, and that Th2 cells exert an inhibitory influence on the development of Th1 clones. It has thus been proposed that preferential induction of alloantigen-specific Th2 lymphocytes could suppress the development of Ag-specific Th1 cells, and as a consequence, inhibit allograft rejection. Evidence that supports or refutes this hypothesis has recently been reviewed (39, 40).
In this study, we have analyzed the T cell stimulatory capacity, tissue trafficking, and influence on Th1 and Th2 cytokine production of immature liver-DCp compared with functionally mature BM-derived DC (BM DC) propagated under similar conditions. The findings suggest that modulation of Th2 cytokine (IL-10) production by donor-derived DCp might be a mechanism underlying the capacity of liver allografts to subvert host immune responses, and contribute to tolerance induction.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ten- to 12-wk-old C57BL/10J (B10; H2Kb, I-Ab, I-E-) and C3H/HeJ (C3H; H2Kk, I-Ak, I-Ek) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Animals were maintained in the specific pathogen-free facility of the University of Pittsburgh Medical Center (Pittsburgh, PA).
Cytokines
rmGM-CSF was a gift from the Schering-Plough Research Institute (Kenilworth, NJ).
Propagation and purification of DC populations
BM cell suspensions were prepared in RPMI 1640 (Life Technologies, Grand Island, NY), supplemented with 10% v/v heat-inactivated FCS (Life Technologies), glutamine, nonessential amino acids, sodium pyruvate, 2-ME, and penicillin/streptomycin (complete medium), using conventional procedures. The method for in vitro culture of DC was modified after that described initially by Inaba et al. (41). Briefly, 2 x 106 cells were cultured in 24-well plates in RPMI 1640 complete medium, supplemented with 1000 U/ml rmGM-CSF. Nonadherent cells released spontaneously from proliferating cell clusters were collected after 6–8 days of culture, and resuspended in complete medium. This low density cell population (DC) was washed twice before final resuspension in complete medium. Purity of the DC was verified by morphologic appearance (Giemsa staining), flow-cytometric analysis, and immunocytochemical staining of cytospin preparations. An extensive panel of mAbs was used, including those specific for leukocytic lineage (CD45), lymphoid, and myeloid cell markers, and mouse DC-restricted cell surface and intracellular Ags. Mature DC were also characterized by the absence of phagocytic activity, using both carboxylated fluorescence latex microspheres (Fluoresbrite Carboxy YG, 0.5–3.0 µm diameter; Polysciences, Warrington, PA) and opsonized SRBC (Remel, Lenexa, KS). Liver DCp were propagated in rmGM-CSF from nonparenchymal cells isolated from collagenase-digested normal liver tissue, as described (20), and characterized by mAb staining and flow-cytometric analysis, as outlined above.
Flow cytometry
Cells were incubated with the following primary mAbs (each from PharMingen, San Diego, CA, unless specified): mouse anti-H2Kb (clone AF6-88.5); rat anti-DEC 205/NLDC145 (42); hamster anti-mouse CD40 (clone HM40-3); rat anti-Gr-1 (clone RB6-8C5); rat anti-mouse CD11b (clone M1/70); hamster anti-mouse CD11c (clone HL3); rat anti-mouse CD80 (clone 1G10); and rat anti-mouse CD86 (clone GL1). Incubation with primary mAbs was followed by FITC-conjugated secondary Abs, as described (20). MHC class II Ag (I-Ab) expression was identified using biotin-conjugated mouse anti-mouse mAb (clone 25-9-3) with FITC streptavidin (Jackson ImmunoResearch, West Grove, PA) as the secondary reagent. After staining, the cells were fixed in 2% paraformaldehyde in PBS before analysis. Cytometric analysis was performed using an EPICS Elite flow cytometer (Coulter, Hialeah, FL). Appropriate fluorochrome-conjugated, isotype-matched, irrelevant mAbs were used as negative controls.
Intracellular cytokine staining
Intracellular cytokines were detected in
CD4+ responder T cells after 72-h MLR. To
increase the intracellular concentration of cytokines, cells were
restimulated with plate-bound anti-CD3
mAb (clone 145.2C11;
PharMingen) and soluble anti-CD28 mAb (clone 37.51;
PharMingen), for 5 h at 37°C, in the presence of Brefeldin A
(Sigma, St. Louis, MO) (10 µg/ml; 5 h at 37°C) before
staining. Thereafter, the cells were washed with 1% FCS/PBS, fixed in
4% paraformaldehyde (20 min, 4°C), and permeabilized with 0.1%
saponin/1% FCS/PBS. Cells were incubated with FITC-conjugated rat
anti-mouse CD4 mAb (clone GK1.5) and with PE-labeled anti-mouse
IFN- (clone XMG1.2), anti-mouse IL-10 (clone JES5-16E3), or
anti-mouse IL-4 (clone BVD4-1D11) (all mAbs from PharMingen). The
cells were then washed in 1% FCS/PBS, resuspended in 1% formaldehyde,
and analyzed by flow cytometry.
Allostimulatory activity (MLR)
The stimulatory activity of the DC in 72-h primary MLR was determined using purified naive allogeneic splenic T cells as responders (20).
Induction of CTLs
Generation of CTLs was quantified using 51Cr-labeled specific, third party, and syngeneic target cells, as described (24).
Cytokine quantitation
ELISA kits (Biosource International, Camarillo, CA) were used to
quantify mouse IFN-, IL-4, and IL-10 in supernatants of cocultures
of C3H T cells (responders) and either allogeneic liver DCp, allogeneic
BM DC, or control splenocytes.
Immunohistochemistry and cytochemistry
Tissue samples were embedded in Tissue-Tek OCT (Miles
Laboratories, Elkhart, IN), snap frozen in isopentane, chilled in
liquid nitrogen, and stored at -80°C until use. Cryostat sections (8
µm) were mounted on slides treated with Vectabound (Vector,
Burlingame, CA), air dried, and fixed in cold acetone (4°C) for 10
min. Cells harvested from cocultures of C3H T lymphocytes and either
allogeneic liver DCp or allogeneic BM DC were spun onto glass slides (5
min at 230 x g) using a Shandon cytocentrifuge, then
air dried and fixed in cold acetone for 5 min. For cytokine staining,
sections or cytospins were incubated successively with: 1) normal goat
serum (1/10; 20 min at RT) to inhibit nonspecific binding by blocking
FcR; 2) avidin-blocking solution (Vector) (15 min at RT); 3) optimal
concentrations (1/50–1/100) of primary rat mAbs (PharMingen) specific
for mouse IFN- (clone R4-6A2), mouse IL-10 (clone JES5-16E3), or
mouse IL-4 (clone BVD4-1D11) (2 h at RT); 4) biotinylated polyclonal
anti-rat Ig (1/100; PharMingen) (30 min at RT); and 5)
avidin-biotin complex-alkaline phosphatase (ABC-AP; Vector) (30 min
at RT).
For detection of donor cells in recipient tissues, sections were incubated in a 1/100 dilution of biotinylated mouse anti-I-Ab ß-chain mAb (clone 25-9-17; PharMingen), followed by ABC-peroxidase (ABC-Px; Vector). AP activity was detected by incubation with the substrate Vector Blue (Vector). To increase the sensitivity of the assay for detection of IL-10 and IL-4 in some samples, the AP activity was developed with the substrate BCIP/NBT (Vector). Peroxidase (Px) activity was developed with 3,3'-diaminobenzidine tetrahydrochloride (Sigma). Endogenous AP activity was inhibited by addition of levamisole (Vector) in the substrate solution. Endogenous Px activity was blocked by successive passages in 70% ethanol, 1% H2O2 in methanol, and 70% ethanol. Tissue sections immunostained with ABC-AP (end product blue) were counterstained with Fast Red (Vector), and mounted first in Crystal/mount (Biomeda, Foster City, CA), and then in Permount (Fisher Scientific Company, Pittsburgh, PA). Sections stained with ABC-Px (end product brown) were counterstained with hematoxylin, dehydrated, and mounted in Crystal/mount. Rat or mouse irrelevant Igs of the same isotype as the primary mAb were used as controls.
Statistical analysis
Results are expressed as means ± 1 SD. Comparisons between different means were performed by ANOVA, followed by the Student Newman Keuls test. Comparison between two means was performed by the Student t test. A p value <0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
After 6–8 days of culture, most cells released from proliferating aggregates of GM-CSF-stimulated liver- or BM-derived cells exhibited typical DC morphology, with prominent cytoplasmic processes, and absence of prominent cytoplasmic granules. The surface phenotype of nonadherent or loosely adherent cells from liver or BM-derived cultures was analyzed by flow cytometry after 6–8 days of culture. Staining for cells of T (CD3), B (CD45/B220), NK (NK1.1), and granulocytic (Gr-1) lineages was absent. As detailed previously (20), liver DCp expressed CD45 (leukocyte common Ag), CD24 (heat-stable protein), CD54 (ICAM-1), CD11b (MAC-1), and CD44 (nonpolymorphic determinant of Pgp.1 glycoprotein). These liver-derived cells also showed weak positivity for the mouse DC-restricted markers CD11c, DEC205, and 33D1, and were positive for the macrophage marker F4/80 (data not shown). Similar results were obtained for BM DC, except that they expressed CD11c and DEC205 strongly. Whereas liver DCp displayed low amounts of MHC and costimulatory molecules (CD40, CD80, CD86), BM DC expressed moderate to high levels of these markers. As reported previously (20), the low expression of MHC Ag and the absence or low levels of costimulatory molecules on liver DCp, cultured under similar conditions to BM DC, were indicative of cells at an immature stage of differentiation.
Liver DCp induce only minimal T cell proliferation
To test their allostimulatory activity, in vitro generated B10 BM
DC or liver DCp were irradiated, and set up in 72-h primary MLR
cultures with naive C3H T cells. In comparison with BM DC, which were
potent inducers of DNA synthesis, and consistent with their surface
phenotype, the liver DCp induced only minimal levels of T cell
proliferation (Fig. 1
). The poor
stimulatory capacity of liver DCp remained unchanged after longer
incubation times with allogeneic T cells (4- or 5-day MLR; data not
shown).
|
We next determined the ability of the cells to induce generation
of CTL in 6-day primary MLR. As shown in Fig. 2, effector T cells induced by allogeneic
BM DC exhibited a progressive increase in specific cytotoxic activity
at increasing E:T ratios. By contrast, identical numbers of liver DCp
induced only minimal levels of CTL activity, which were lower than
those induced by bulk allogeneic spleen cells.
|
The intracytoplasmic expression of IFN-, IL-10, and IL-4 was
analyzed by flow cytometry in responder CD4+ T
cells (C3H) after in vitro stimulation in 3-day MLR with either
allogeneic mature B10 BM DC, or immature liver DCp. Before staining for
intracytoplasmic cytokines, T cells were restimulated by immobilized
mAb to CD3 and soluble mAb to CD28 in the presence of Brefeldin A,
which causes newly synthesized proteins to accumulate within the
endoplasmic reticulum. Without Brefeldin A, the expression of cytokines
could not be detected by flow cytometry. When Brefeldin A was added to
the cultures, without restimulation with anti-CD3 and
anti-CD28 mAbs, a similar pattern of cytokine expression was
detected, but at a much lower level. The cells were harvested, fixed,
permeabilized, and processed for double staining using primary PE-
conjugated cytokine-specific mAb, followed by FITC-coupled anti-CD4
mAb. Based on their high SSC compared with T cells, and their lack of
expression of CD4 (41), DC were gated out of the mixed
cell population (composed of T cells and DC at a ratio 20:1) harvested
from the MLR. The intracellular expression of cytokines in
CD4+ T cells stimulated either with liver DCp or
with BM DC is illustrated in Fig. 3. In a
typical experiment, a substantial fraction of
CD4+ T cells (>15%) was induced to produce
IFN-, but few cells expressed IL-10 or IL-4, when allogeneic BM DC
are used as stimulators (Fig. 3). By contrast, very few
CD4+ T cells (
2%) produced detectable levels
of either IFN-, IL-10, or IL-4 after 3 days allostimulation with
liver DCp (Fig. 3). C3H T cells maintained for 72 h, with or
without IL-2 (2 U/ml), then stimulated with anti-CD3 mAb and
anti- CD28 mAb, treated with Brefeldin A, and immunostained as
described above, were included as controls. Less than 0.2% of these
control CD4+ T cells were positive for the
cytokines analyzed (data not shown).
|
, but comparatively high levels
of IL-10, are present in cocultures of liver-derived DCp and allogeneic
T cells
To ascertain the influence of mature BM DC vs liver DCp on
allogeneic Th0 differentiation into Th1 or Th2 clones, we quantified
the production of IFN-, IL-10, and IL-4 by ELISA in supernatants
obtained from 72-h cocultures of allogeneic naive T cells with either
BM DC or liver DCp (DC:T cell ratio = 1:20). IFN- production
was detected in BM DC-stimulated cultures, but was substantially lower
(
10% of the former) when liver-derived DCp were employed as
stimulators (Fig. 4A). To
confirm whether DC or T cells were the source(s) of cytokine
production, cytospin preparations of cells collected from MLRs were
stained for IFN- by immunocytochemistry. Both BM DC and some T
cells, grouped in small clusters, were positive for IFN- (data not
shown). By contrast, IL-10 production was detected only in cocultures
of liver DCp and allogeneic T cells. In these latter cultures, the
concentration of IL-10 at 72 h was 25-fold that in cocultures
of BM DC and allogeneic T cells (Fig. 4B). Immunostained
cytospins revealed that liver DCp were significant IL-10 producers
(Fig. 5A). In addition, very
few clusters of T cells, most of them in close apposition to DC, were
also positive for intracytoplasmic IL-10 (Fig. 5A). There
was much less evidence of IL-10 production in corresponding cytospin
preparations from BM DC-T cell cultures (Fig. 5B). IL-4 was
not detected by ELISA, or by immunocytochemistry in cytospin
preparations.
|
|
To assess their ability to home to recipient lymphoid tissues, and
to survive in an allogeneic environment in vivo, 5 x
105 liver DCp or BM DC (B10) were injected s.c.
in one hind footpad of normal allogeneic mice (C3H). At days 1, 2, 7,
and 14, animals were sacrificed, and donor cells detected in cryostat
sections of recipient lymphoid tissues (draining lymph nodes, spleen,
and thymus) by staining for donor MHC class II
(I-Ab). Donor cells with dendritic morphology
were present in the subcapsular and paracapsular sinuses, and in the
T-dependent areas (the interfollicular area of the cortex, and
paracortex) of popliteal lymph nodes, 2 days after their
administration, but could not be detected thereafter (Fig. 6, A and D). No
donor MHC class II+ cells were observed in the
inguinal lymph nodes. Donor BM DC and liver DCp showed a similar
pattern of tissue homing in the spleen and thymus (Fig. 6, B, C, E, and F). During the
first 2 days after injection, both types of donor DC were located
principally in the marginal zones of the spleen (Fig. 7A). At later time points
(days 7 and 14), DC mobilized into the periarteriolar lymphatic sheaths
(PALS, Fig. 6, B and E). In general, both donor
BM DC and liver DCp were detected as single cells; however, less
frequently, liver DCp formed clusters of 5–10 cells. These clusters
were located mainly in marginal zones (Fig. 7A). The number
of donor BM DC present in spleen peaked at day 7, and decreased
significantly thereafter (p < 0.05). By
contrast, the number of liver-derived DCp remained constant throughout
the 14-day follow-up period (p = 0.65). In the
thymus, a few donor BM DC or liver DCp were detected at the
corticomedullary junction on days 1, 2, 7, and 14 (Fig. 6, C
and F).
|
|
, IL-10, and IL-4
in allogeneic recipients
Analysis of in situ cytokine production by recipient lymphocytic
cells in T cell areas of spleens of animals injected either with
allogeneic liver DCp or BM DC was performed at different times (days 1,
2, 7, and 14) after their administration. The local synthesis of
cytokines was quantified as the number of lymphocytic clusters (more
than three positive cells with round cell morphology in close
apposition) per cm2 of recipient spleen. As can
be seen in Figs. 8 and
9, the number of clusters producing
IFN- at day 7 was significantly higher (p =
0.04) in those animals injected with BM DC compared with mice injected
with liver DCp. Conversely, clusters expressing IL-10 or IL-4 were only
detected in spleens of mice given liver DCp (Figs. 7
B and
8). Immunohistochemical staining of serial sections with a mAb specific
for the T cell marker CD3 demonstrated that the areas positive for
specific cytokines corresponded to T cell areas of the spleen.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Therefore, passenger leukocytes (most likely DC) may have a dualistic
role with potential to elicit T cell activation and graft rejection, or
induce T cell tolerance and graft acceptance (11, 14). The
sustained release from the transplanted liver of immature APC,
deficient in costimulatory molecules, with poor comparative ability to
stimulate Th1 and the development of alloantigen-specific CTLs,
delayed-type hypersensitivity, and IgG2a production (that promotes
Ab-dependent cellular cytotoxicity), may contribute to tolerance
induction. Alternatively, liver-derived DCp might induce the
proliferation of Th2 clones with capacity to inhibit Th1 responses, or
facilitate the development of Th cells with regulatory/suppressor
functions, such as T regulatory-1 (Tr1) cells (50, 51).
Recent findings suggest that in mice, lymphoid-related DC
(CD11c+, CD11b-,
CD8
+) elicit only Th1 responses, whereas
myeloid DC (CD11c+, CD11b+,
CD8-) induce either Th1 or Th2 responses in
vivo (52, 53).
In the present study, we have confirmed that, in comparison with mature
BM DC, liver DCp display an immature phenotype with absence of CD40 and
CD86 surface expression, low levels of MHC class I and II, and as a
consequence, low stimulatory capacity for naive allogeneic T cells.
Unlike mature BM DC, these liver DCp did not induce detectable levels
of intracytoplasmic IFN- in allogeneic CD4+
cells in 72-h MLR, and elicited very low levels of CTLs in vitro. These
observations suggest that liver DCp have a more restricted capacity
than BM DC to stimulate the differentiation of Th0 cells to Th1 cells.
Accumulating evidence indicates that the type of cytokine secreted
during the early stages of a primary immune response determines the
sort of T cell clone to be generated. Thus, in an IL-4-rich
microenvironment, DC induce T cells to differentiate into
IL-4/IL-5/IL-10/IL-13-producing Th2 cells (50). DC are
able to produce and secrete IL-12, a cytokine that turns naive,
CD4+ Th0 cells into IFN--secreting Th1
lymphocytes (54, 55). Although we were able to detect
secretion of IL-12 p70 by BM DC using ELISA (49.9 ± 2.4 pg/ml/24
h/106 DC), IL-12 p70 secretion by liver DCp could
not be detected. This finding could explain the poor capacity of liver
DCp to generate Th1 cells in vitro. Interestingly, we found that liver
DCp produced IL-10, and induced a low incidence of
IL-10+ allogeneic responder
CD4+ cells, a phenomenon that may be related to
the induction of either Th2 or Tr1 cell clones (50, 51).
The very low incidence of IL-10+ T cells induced,
and the low proliferative capacity of Th2 or Tr1 cell clones that has
been reported (10, 51) could explain the poor
incorporation of [3H]TdR observed in
proliferation assays in this study. Stumbles et al. (10)
have reported that in the rat respiratory tract, resident DC also show
an immature phenotype (MHC class Ilow, class
IIlow, CD80-,
CD86-) and express high levels of IL-10 mRNA.
Although freshly isolated respiratory tract immature DC are poor
stimulators in primary MLR, they elicit a predominant Th2 response in
vivo after being pulsed ex vivo with OVA (10). By
contrast, maturation of respiratory tract DC with GM-CSF up-regulates
the expression of MHC molecules, B7, and the capacity to stimulate both
Th1 and Th2 responses, a phenomenon that seems to be associated with
increased production of IL-12 p35 chain mRNA, and down-regulation of
IL-10 mRNA synthesis (10).
Previous reports have demonstrated that IL-10-treated DC, or viral
IL-10 gene-transduced DC, exhibit tolerogenic properties
(30, 31, 32, 33). In the present study, we have confirmed that
IL-10 is produced by putative tolerogenic liver-derived DCp, an
observation that agrees with the high levels of IL-10 mRNA reported in
immature DC in the rat respiratory tract (10). There is
also recent evidence that in humans, the subpopulation of
CD14+ CD1a- DC generated
in vitro in the presence of GM-CSF and TNF- produces IL-10, a fact
suggesting that other subpopulations of myeloid DC that reside in
peripheral tissues might secrete this cytokine (56). In
this regard, it was demonstrated recently that DC isolated from
Peyer’s patches secreted IL-10, and were able to induce
differentiation of Th2 cells (57). IL-10, produced by
immature DC, Th2, and Tr1 cells, down-modulates the expression of CD80
and CD86 on DC, accelerates their apoptotic death, and skews the
Th1/Th2 balance to Th2 by inhibiting IL-12 synthesis by DC
(30, 31, 32, 58, 59, 60, 61). Thus, IL-10 may play a key role in
exhibition of the tolerogenic properties of liver DCp in vivo.
A specialized ability of DC is the capacity to migrate in vivo to T cell areas of peripheral lymphoid tissues (20, 21, 62). As reported herein, both donor BM DC and liver DCp were detected in the subcapsular and paracortical sinuses of the draining lymph node after s.c. administration, a fact that might represent donor DC in transit to the interfollicular T cell area where they were also found. Using a similar model of DC trafficking, others have observed that the number of donor DC in the popliteal lymph node decreases rapidly 48 h after their administration (63). In the spleen, donor DC were present 24 h after administration. Whereas the number of donor liver DCp detected in spleen remained constant throughout the 14-day follow-up period, the density of BM DC increased steadily until day 7, and decreased thereafter. A greater capacity of donor liver DCp, in comparison with BM DC, to shut down generation of a Th1 response, or alternatively, their inability to trigger a cellular immune response, might contribute to the extended survival of donor liver DC that has been observed after liver transplantation (16, 19).
In the present study, injection of liver DCp mimicked the trafficking pattern to the spleen described by Demetris et al. after orthotopic liver transplantation in rats (16). Donor DC were concentrated initially in the marginal zones of the spleen, and later mobilized to the T cell-dependent PALS. The marginal zone has been considered a site populated by immature myeloid DC, B cells, and specialized macrophages (64, 65). Specialized gates in the marginal zone appear to be the main routes to the white pulp employed by circulating DC in physiological conditions (66, 67). However, the present and previous trafficking studies (16) indicate that donor-derived DC may also enter the white pulp through the marginal zone after liver transplantation. Whether immature donor DC remain as precursors in the marginal zone, and replicate to maintain the population of immature DC long-term (late chimerism) remains unknown. However, the presence of donor MHC class II+ proliferating cells (detected by others using 5-bromo-2'-deoxyuridine incorporation, or by the presence of mitotic figures), and our observation of clusters of donor cells with DC morphology in the marginal zone, are evidence for the replicative potential of donor liver DCp that colonize the recipient spleen (16, 21). As reported previously after liver transplantation, or after administration of liver DCp, very few donor DC were detected in the recipient thymus throughout the 14-day follow-up (16, 21). Interestingly, those that were found were located at the corticomedullary junction. Whether these donor DC may play a role in acquired thymic tolerance to alloantigens (68) is still unknown.
Administration of mature BM DC was associated with a higher number of
IFN--positive cells in spleen than observed after injection of
immature liver DCp. A certain degree of DCp maturation and limited
IL-12p70 secretion may have occurred in vivo in response to
proinflammatory cytokines released at the injection site, and following
interaction of DCp with extracellular matrix proteins (20)
during cell trafficking. Clusters of cells with lymphocyte morphology,
located in T cell areas, and positive for IL-10 or IL-4, were only
detected after injection of liver DCp. These results suggest that in
vivo, allogeneic immature liver DCp have more restricted capacity than
BM DC to induce the proliferation of IFN--producing recipient T
cells. On the other hand, unlike BM DC, they appear to promote the
proliferation of recipient allogeneic T cells with the ability to
produce IL-10 and IL-4. Such cells may either shift the Th balance
toward a Th2 response, or induce the proliferation of regulatory T cell
clones with the capability to produce IL-10 and/or IL-4, but
comparatively little IFN- (50, 51). The very low number
of IL-10- or IL-4-positive clusters detected in spleens of liver DCp
recipients, in proportion to the number of IFN--positive areas
observed when BM DC were injected, might be ascribed to the lower
replicative potential of Th2 and T regulatory cell clones compared with
Th1 cells (10, 51).
Although generation of a Th1 response, determined by cytokine mRNA
expression after liver transplantation (69, 70, 71), may
appear to contradict the present observation that liver DCp
preferentially induce Th2 cytokines in vivo, the two phenomena may
represent different phases of the tolerance induction process. During
the early phase (days 1 to 2), migration of mature DC from the liver
may induce a Th1 response in recipient lymphoid tissue (peak of IL-2
and IFN- mRNA synthesis in spleen and celiac lymph nodes (69, 70, 72)). However, during the following 14 days, the number of
infiltrating recipient T cells and their cytotoxic activity rapidly
decrease, due to apoptosis (72). Meanwhile, migration of
donor liver DCp to secondary lymphoid organs may promote, in a second
phase, generation of Th2 and/or regulatory T cell clones with the
ability to down-regulate the Th1 allo-response, and maintain the state
of alloantigen-specific tolerance.
There is evidence to support or refute the hypothesis that organ transplant tolerance involves a dominant Th2 response (39, 40, 73, 74). Moreover, it has been observed that administration of IL-10 to heart allograft recipients can inhibit or exacerbate rejection, depending on cytokine dosage and timing (75, 76). The present observations suggest that immature and mature myeloid DC may differentially modulate Th1 and Th2 cytokine production in vitro and in vivo. An implication of the finding that immature donor (liver) DCp secrete IL-10 and/or modulate host responses toward Th2 cytokine predominance is that these cells may contribute toward the comparative immunologic privilege of hepatic allografts.
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 A.K. and A.E.M. contributed equally to this work and should be considered as co-first authors.
3 Address correspondence and reprint requests to Dr. Lina Lu or Dr. Angus W. Thomson, Department of Surgery, University of Pittsburgh Medical Center, W1540, Biomedical Science Tower, 200 Lothrop Street, Pittsburgh, PA 15213. E-mail addresses:
4 Abbreviations used in this paper: BM, bone marrow; ABC, avidin-biotin complex; AP, alkaline phosphatase; DC, dendritic cell; DCp, dendritic cell progenitor; m, mouse; PALS, periarteriolar lymphatic sheaths; Px, peroxidase; RT, room temperature; Tr1, T regulatory-1.
Received for publication March 26, 1999. Accepted for publication November 18, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
on the development of dendritic cells from progenitors in mouse bone marrow. Stem Cells 15:144.+ and CD8- subclasses of dendritic cells direct the development of distinct T helper cells in vivo. J. Exp. Med. 189:587. production by T helper 1 cells. Eur. J. Immunol. 26:659.[Medline]
, but strongly enhanced by interleukin-10. Eur. J. Immunol. 25:1943.[Medline]
production. J. Immunol. 150:3755.[Abstract]
in long-surviving mouse heart allografts after brief CD4-monoclonal antibody therapy. Transplantation 59:559.[Medline]
This article has been cited by other articles:
|
|
 |
|
  D H Adams, B Eksteen, and S M Curbishley Immunology of the gut and liver: a love/hate relationship Gut, June 1, 2008; 57(6): 838 - 848. [Full Text] [PDF]
|
|
|
|
|
|
T. Watanabe, H. Katsukura, T. Chiba, T. Kita, and Y. Wakatsuki Periportal and sinusoidal liver dendritic cells suppressing T helper type 1-mediated hepatitis Gut, October 1, 2007; 56(10): 1445 - 1451. [Abstract] [Full Text] [PDF]
|
|
|
), or mature B10 BM DC (
),
assessed using naive C3H (H2k) splenic T cells as
responders. The stimulator cells were propagated as described in
Materials and Methods, harvested after 6- to 8-day
culture, and purified by metrizamide centrifugation. They were set up
at various concentrations with 2 x 105 responder T
cells, and the cultures maintained for 72 h. [3H]TdR
was added 18 h before harvesting. The MLR-stimulatory activity of
freshly isolated syngenic (C3H 
) or allogenic (B10 •) bulk spleen
cells is also shown. The results are expressed as mean cpm ± 1
SD, and are representative of at least three separate
experiments.
), or liver DCp (
, dotted line). Results are means ± 1 SD
obtained from groups of three mice at each time point.