| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Sir William Dunn School of Pathology, University of Oxford, Oxford, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
DC in peripheral tissues are functionally immature, but in culture acquire potent immunostimulatory capacity and modulate expression of surface markers (3, 4, 5). Thus, fresh murine Langerhans cells (LC) are phagocytic and endocytic (6), and can acquire and process protein Ags, but are poor stimulators of T cells (7). After culture, they become weakly endocytic and lose the ability to process native Ags, but develop potent immunostimulatory capacity. During this period they modulate expression of MHC class II, FcR, CR3, and F4/80 (3, 8). Murine cardiac and renal DC are weak stimulators of a MLR, but after overnight culture become potent stimulators (4). Their ability to process and present protein Ags has not been tested. Freshly isolated lamina propria DC can stimulate a moderate MLR, but after culture increase their potency to that of lymph or lymph node DC (5). Although these changes are thought to represent normal DC differentiation, it is not clear that all the functional and phenotypic changes observed in cultured DC occur in vivo under normal conditions. Thus in culture, L-DC rapidly express surface markers not seen on DC isolated freshly from mesenteric nodes (9, 10), and there is evidence that Ag presentation by DC in situ can be tolerogenic (11).
There is increasing evidence for heterogeneity among DC in terms of phenotype, lineage, and functions. Heterogeneity in DC surface marker expression has been shown in several situations (12, 13, 14, 15, 16, 17, 18). There is also evidence that cells with DC characteristics may arise from more than one lineage (reviewed in 19 . Thus human monocytes cultured in GM-CSF and IL-4 develop into cells with the phenotype and function of typical DC (20), but human bone marrow cultures can give rise to both CD14+ and CD14- DC (21), and the precursors of DC grown from murine blood do not have monocyte characteristics (22). A third DC lineage is suggested by the observation that T cell precursors can develop into DC (23). Little is known, however, of the functional significance of DC heterogeneity. It has been suggested that rat pulmonary DC vary in their ability to stimulate CD45RC+ and CD45RC- T cells (15). There is evidence that human blood DC vary in their ability to be infected by HIV (24), and this may correlate with CD4 expression (25). It has been shown that human dermis contains functionally distinct DC populations (26). Recently, a population of CD8+ murine splenic DC has been identified that can induce apoptosis in allogeneic CD4+ T cells (27).
We have shown that DC isolated from different sites in the rat (small intestine, pseudo-afferent lymph, and lymph nodes) are heterogeneous in their expression of several surface markers, including CD2, Thy1, and CD11b/c (5). We have previously shown that L-DC are themselves heterogeneous in terms of morphology, expression of surface markers, and enzyme content (2, 9, 10, 28). To examine the functional significance of L-DC heterogeneity we have separated rat lymph L-DC into subpopulations expressing different markers and tested the ability of the separated cells to stimulate a MLR, to process and present Ags to sensitized T cells, and to activate resting T cells in vivo. We show that L-DC expressing CD4 and OX41, a mAb made against rat macrophages (29), differ morphologically and in enzyme content from their negative counterparts. CD4+ DC are markedly more potent in stimulation of an allogeneic MLR, presentation of Ag to sensitized T cells and Ag-specific activation of naive T cells. These differences are accentuated after culture and do not reflect simply the age of the L-DC.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Rats were specific pathogen-free inbred strains bred at the Medical Research Council Cellular Immunology Unit, Sir William Dunn School of Pathology (Oxford, U.K.). The strains used were PVG-RT1c and DA-RT1a.
Surgical procedures
Mesenteric lymphadenectomy and thoracic duct cannulation were carried as described previously (28). Rats were not irradiated.
Antigen
Keyhole limpet hemocyanin (KLH) was obtained from Calbiochem (La Jolla, CA).
Immunization with Ag-pulsed L-DC
L-DC were incubated with KLH (1 mg/ml) in RPMI 1640 (Life Technologies, Irvine, U.K.) containing 10% FCS, 2 mM L-glutamine, 50 U/ml penicillin, 50 µg/ml streptomycin, 1 mM sodium pyruvate, and 25 mM ME (complete medium) at 37°C for 2.5 to 3 h. The pulsed cells were washed three times with RPMI 1640 plus 10% FCS (washing medium) and resuspended in complete medium. Different numbers of Ag-pulsed L-DC were resuspended in PBS and injected into the hind footpads of naive PVG rats (100 µl/footpad). Ten days later, cell preparations were made from the popliteal lymph nodes and used as a source of primed lymph node cells in an Ag presentation assay.
Immunization with protein Ag
KLH in PBS (2 mg/ml) was emulsified with an equal volume of CFA (Sigma, Poole, U.K.). Rats were injected into the hind footpads with 100 µl of emulsion containing 100 µg Ag. Twelve days later spleens were removed, and cells were prepared by teasing. These were used as a source of Ag-primed cells for Ag presentation assays.
L-DC
L-DC were enriched from lymph cells collected from the thoracic duct of mesenteric-lymphadenectomized rats (XTDL) by single step density separation. XTDL resuspended in washing medium (at a concentration of 5–10 x 106/ml) were overlaid over NycoPrep solution (NycoMed, Oslo, Norway) and centrifuged at 400 x g for 20 min. The interface cells contained 40 to 60% L-DC. The major contaminating cells were B lymphocytes. There were 1% or less macrophages present as identified by morphology and the ability to phagocytose opsonized SRBC.
CD4+ and CD4- L-DC separation
L-DC were incubated with W3/25 (anti-CD4) mAb tissue culture supernatant for 30 min at 4°C. After washing twice they were mixed with a 5% SRBC suspension coated with goat anti-mouse IgG (Sigma), rotated at 4°C for 25 min, and then separated over Isopaque-Ficoll. Both interface and pelleted cells were harvested. SRBC in the pelleted cells were lysed with Tris-NH4Cl. Both cell populations were washed three times before use. L-DC separation using other mAbs was performed in the same way.
Popliteal lymph node cells
Popliteal lymph node cells were obtained by teasing and were washed twice with RPMI 1640 plus 0.1% BSA before being resuspended in complete medium with 5% rat serum replacing FCS.
Ag presentation assay
Assays were performed in triplicate in 96-well round-bottom tissue culture plates (Flow Laboratories, Irvine, U.K.). KLH-primed spleen cells (2 x 105) in complete medium (containing 10% FCS) were cultured for 108 h in a total volume of 0.2 ml in the presence of Ag-pulsed L-DC. To measure the ability of Ag-pulsed cells to prime naive recipients, 2 x 105 popliteal lymph node cells from rats injected into the footpad with Ag-pulsed L-DC were cultured for 120 h in complete culture medium (10% FCS was replaced with 5% DA rat serum) in the presence of Ag without exogenous APC. All the cultures were conducted at 37°C in an atmosphere of 5% CO2 in air. Tritiated thymidine (Amersham, Aylesbury, U.K.; 0.5 µCi/well) was added to cultures 16 h before harvesting (Skatron, Lier, Norway), and uptake was measured by scintillation counting. Data are expressed as mean gross counts per minute.
Ab blocking of MLR or Ag presentation
To test whether CD4+ L-DC and CD4- L-DC stimulate different T cells subpopulations, i.e., CD4+ or CD8+ T cells, different concentrations of W3/25 (anti-CD4) or OX8 (anti-CD8) were added to cultures.
Flow cytometry
Partially enriched L-DC (40–60%) were prepared as described
above. L-DC were incubated with mouse anti-rat cell surface Abs for
25 min at 4°C. OX21 (anti-human factor I)) was always used as a
negative control. After washing twice, the cells were incubated with
phycoerythrin-conjugated rabbit anti-mouse IgG (RAM-PE) for 25 min.
After two washes they were incubated with 10 µl of mouse IgG (2
mg/ml) for 10 min. Without washing, FITC-conjugated OX6 (Serotec,
Kidlington, U.K.; anti-MHC class II, 1/100 dilution) was added to
each sample and incubated for 25 min. Finally, cells were washed twice
and analyzed in a FACScan flow cytometer (Becton Dickinson, Mountain
View, CA). Two-color analysis was used. To identify L-DC, cells were
labeled with OX6 (anti-MHC class II) and OX62, which only labels
L-DC in XTDL (30). OX6-FITC labeling reveals a very bright peak that
contains 96 to 99% L-DC as judged by OX62 labeling. Analysis of
CD4- L-DC was conducted on DC-enriched XTDL depleted of
CD4+ cells using a gate that included only the OX6-bright
peak. When analyzing CD4+ L-DC, whole L-DC were first
depleted of T cells with R73 (anti-
ß TCR; T cells are the only
cells apart from DC in TDL that express CD4) and then were incubated
with different mouse anti-rat mAb. They were incubated with rabbit
anti-mouse FITC (RAM-FITC), then with mouse Ig followed by
incubation with biotinylated W3/25 (anti-CD4). Finally, they were
incubated with avidin-PE and analyzed in a FACScan. CD4+
cells (PE-positive) were gated and analyzed against FITC-positive
cells.
Immunocytochemistry
Cytospin preparations (Shandon Scientific, Runcorn, U.K.) were fixed for 10 min in cold 100% ethanol and incubated with tissue culture supernatants of mouse anti-rat mAbs for 40 to 60 min at 4°C. OX21 (mouse anti-human factor I) was always used as a negative control. The second layer was peroxidase-conjugated rabbit anti-mouse Ig (Dako, Glostrup, Denmark; 1/30 dilution plus 5% DA rat serum). The reaction product was developed using 0.005% H2O2 in 0.1% 3,3'-diaminobenzidine tetrahydrochloride (Polyscience, Warrington, PA) made up in 50 mM Tris-HCl. Invariant chain (Ii) was detected by the RG11 mAb (a gift from Dr. K. Reske) (31). For cell counting, slides were coded and counted "blind." Intensity of labeling was assessed on a 0 to +++ scale by comparison with OX21 (0) and MHC class II expression (+++) on L-DC.
5-Bromo-2'-deoxyuridine (BrdUrd) labeling of DC
PVG rats were cannulated and injected i.v. with 5 mg BrdUrd (Sigma) in PBS. Lymph samples were collected at intervals thereafter, and DC were enriched by centrifugation over NycoPrep. Cytospin preparations were double labeled for BrdUrd and CD4. Briefly, cells were fixed in cold ethanol and labeled for CD4 using W3/25 and OX35. The second layer was alkaline phosphatase-coupled goat anti-mouse Ab, and the substrate was 5-bromo-4-chloro-3-indolyl phosphate and the blue reaction product developed with nitro blue tetrazolium (Sigma Fast BCIP/NBT). Slides were placed in 60 ml of N-N-dimethyl formamide (Sigma), 2 ml of H2O, and 1 ml of 20x sodium citrate buffer for 25 min at 70°C to hydrolyze DNA. They were then incubated with an anti-BrdUrd mAb, Bu20a (a gift from Dr. D. Y. Mason, John Radcliffe Hospital, Oxford, U.K.), and bound Ab was detected using immunoperoxidase, as described above, before mounting in Aquamount (BDH, Poole, U.K.). For counting, slides were coded and examined under oil immersion using Nomarski interference microscopy. At least 300 DC were counted for each time interval.
Electron microscopy
Partially purified L-DC were separated into CD4+ and CD4- subpopulations by rosetting. Cells were fixed in 2% glutaraldehyde and processed by standard methods for electron microscopy. Sections were examined on a Zeiss 912 electron microscope (New York, NY).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
L-DC separated into CD4+ and CD4-
populations were examined by Nomarski interference microscopy and on
cytospins. CD4+ L-DC were mostly large cells with round
cell bodies and multiple small spiky processes, with relatively few
cells possessing long blunt pseudopodia. In contrast, many
CD4- L-DC possessed irregular cell bodies and displayed
long pseudopodia and/or veils as well as some small spiky processes
(Fig. 1
, A–D). These
distinctions were not absolute, and DC with intermediate morphology
were present in both populations. Separated CD4+ and
CD4- L-DC cultured overnight in medium containing rGM-CSF
and IL-4 retained their original morphology; in particular,
CD4+ L-DC did not develop the long blunt pseudopodia
characteristic of CD4- cells (Fig. 1, E and
F). L-DC show heterogeneous staining for nonspecific
esterase (28). Nonspecific esterase staining of separated L-DC showed
that 96% of CD4- DC were strongly or moderately positive
with diffuse cytoplasmic staining, whereas only 12% of
CD4+ DC showed similar staining. Approximately 77% of
CD4+ DC showed a small perinuclear patch of stain, and 11%
were negative (Fig. 1, G and H, and Table I). CD4- DC were generally
larger than CD4+ cells, and many CD4- DC
possessed conspicuous cytoplasmic inclusions (Fig. 2A). Counts on stained
preparations showed that >75% of CD4- DC contained such
inclusions compared with <7% of CD4+ DC (Table I). When
separated populations were cultured for up to 48 h and stained for
nonspecific esterase, the expression of the enzyme by both populations
remained similar to that in fresh cells. Electron microscopy of
separated populations showed that the cytoplasm of CD4-,
but not that of CD4+, L-DC contained a variety of
inclusions of differing sizes and densities (Fig. 2, B–D).
We have previously shown that many of these inclusions are acid
phosphatase positive, and that some contain DNA (28). These results
show that CD4+ L-DC differ from CD4- L-DC
morphologically and that this difference is a stable characteristic in
culture.
|
|
|
L-DC enriched to 40 to 60% purity by separation over NycoPrep were examined by flow cytometry. MHC class II labeling revealed two positive peaks. The strongly positive peak was absent from normal TDL. Ninety-six to ninety-nine percent of cells in the MHC class II-bright peak are also OX62+ (not shown) and are therefore DC (DC are the only OX62+ cells in XTDL). The weakly MHC class II+ cells are B lymphocytes. For subsequent analysis, XTDL cells were gated to include only the bright peak.
Double labeling confirmed that L-DC are heterogeneous for several
markers including Thy1 (OX7), CD11b/c (OX42), CD4, and OX41. Labeling
for CD4 and OX41 revealed two discrete and largely nonoverlapping
peaks, whereas expression of CD11b/c or Thy1 did not reveal discrete
peaks (Fig. 3). Examination of
subpopulations (Figs. 4 and
5; see Materials and Methods)
showed that CD4+ and CD4- L-DC express similar
amounts of MHC class II, CD11b/c, ICAM-1, and OX62. The expression of
other markers by the two populations differs. Thus, 80% of
CD4+ L-DC but only 42% of CD4- LDC express
Thy1. Almost all CD4- L-DC show strong expression of B7
identified by CTLA4-Ig binding, but expression by CD4+ L-DC
is weaker, with some L-DC apparently negative. Strikingly, the great
majority of CD4+ L-DC express OX41, whereas
CD4- L-DC are almost negative for OX41, suggesting that
CD4 and OX41 may be coexpressed on L-DC.
|
|
). Similarly,
depletion using OX41 removed CD4+ L-DC (data not shown).
When XTDL MHC class II-bright cells were labeled with OX41, W3/25, or a
mixture of OX41 and W3/25, the three profiles were very similar. The
number of DC labeled by the mixture of mAbs was not significantly
different from the number labeled by W3/25 or OX41 alone, but the
intensity of labeling was increased, confirming that the populations of
L-DC expressing the two markers are largely overlapping (data not
shown). CD4+ and CD4- L-DC are functionally distinct
CD4+ L-DC are more potent APC for sensitized T cells.
Separated CD4+ L-DC and CD4- L-DC were pulsed
in vitro with KLH for 2.5 to 3 h and cultured with KLH-sensitized
spleen cells for 4 days, and proliferation was measured. Unfractionated
L-DC were efficient APC, but CD4- L-DC were considerably
weaker. CD4+ L-DC showed slightly lower APC potency than
unfractionated L-DC but much higher potency than CD4- L-DC
(Fig. 6a). The function of CD4
on DC is unknown, but it was possible that residual anti-CD4 Ab on
the positively selected CD4+ L-DC was interfering with Ag
presentation. This hypothesis was tested using L-DC fractionated with
other Abs. L-DC positively selected using OX41 showed a potency similar
to that of unfractionated L-DC and a markedly higher potency than
OX41- L-DC (Fig. 6b). In experiments
in which L-DC were fractionated using either W3/25 or OX41 and tested
with responder cells from the same animal, OX41+ L-DC
consistently gave higher counts than W3/25+ L-DC (data not
shown).
|
a), perhaps reflecting the
less clear distinction between positive and negative cells. In
contrast, CD11b/c- L-DC were more potent APC than
CD11b/c+ L-DC (Fig. 7b). This experiment
excludes the possibility that the high potency of CD4+ or
OX41+ L-DC is due to effects of the rosetting procedure.
|
a). OX41+ cells
had a slightly higher potency than unfractionated L-DC, but a much
higher potency than OX41- L-DC (Fig. 8b).
|
).
|
Culture differentially affects the APC activity of CD4+ and CD4- L-DC
Culture of unfractionated L-DC does not affect their ability to
process and present native Ag (32), but it was not clear that different
subpopulations behaved similarly. Fractionated CD4+ and
CD4- L-DC were cultured for different periods in the
presence of rGM-CSF and were tested as APC (Fig. 10). The recovery of cells cultured for
different periods showed that DC survival was similar up to 20 h,
but that by 48 and 72 h, a higher percentage of CD4+
L-DC were recovered (Fig. 10A). In allogeneic MLRs,
CD4+ L-DC did not change their immunostimulatory potency
over 48 h, but CD4- L-DC increased their potency to a
level equivalent to that of CD4+ cells (Fig. 10, B–D).
|
, E–H). CD4+ and CD4- L-DC express Ii
Cytospins of separated fresh or cultured L-DC were labeled for Ii
with the RG11 mAb (Fig. 11).
Seventy-five percent of fresh CD4+ L-DC were positive, but
the intensity of labeling was weak or moderate. In contrast, all fresh
CD4- L-DC expressed the marker, and levels of expression
were both higher than those in CD4+ cells and relatively
uniform. In culture, expression by CD4+ DC remained stable
for up to 48 h, but the proportion of positive CD4-
DC and the levels of expression decreased steadily over 48 h.
|
It was possible that CD4+ DC were the precursors of
CD4- DC. If this were so, then we would predict that, on
the average, CD4+ DC would have spent less time in the
intestinal wall before entering lymph. We have shown that bone marrow
precursors of DC can be labeled in vivo with thymidine, and that the
average time spent by DC in the intestine before entering lymph is 3 to
4 days (28). To determine whether CD4+ and
CD4- L-DC had different kinetic properties, rats were
injected i.v. with BrdUrd, and DC were collected at intervals. Cytospin
preparations of enriched DC were double labeled for CD4 and BrdUrd, and
the proportions of labeled DC were counted on coded slides. DC were
identified by their morphology (size and irregularity) using Nomarski
interference microscopy. The results (Fig. 12) show that at the earliest time
points a slightly larger proportion of CD4+ DC was labeled,
but the numbers of labeled cells were too small for the differences to
be significant. By 30 h after BrdUrd administration, almost equal
proportions of CD4+ and CD4- DC were labeled.
These results show that the minimum times required for both populations
of DC to enter lymph after their final bone marrow division are
similar, and that CD4- DC spend less time in the
intestinal wall before entering lymph. These observations suggest that
CD4- DC do not develop from CD4+ DC.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The best understood cause of DC heterogeneity is maturation. Peripheral DC are poor stimulators of naive T cells, are actively endocytic, and express macrophage markers (3, 4, 5, 33). Cultured peripheral DC acquire the characteristics of splenic or lymph node DC, and similar changes may be induced in DC cultured from bone marrow or blood by cytokines (20). Caution in this interpretation is necessary, because we have shown that cultured lymph DC express markers not seen in DC freshly extracted from nodes (5, 9, 10).
DC heterogeneity does not only relate to maturation/differentiation. DC
may derive from different lineages. Thus, human blood- or bone
marrow-derived DC may arise from monocytes or from a different
precursor (20, 34, 35), whereas some murine thymic and spleen DC may
arise from lymphoid precursors (23). In murine lung, distinct DC
populations exist in epithelium and connective tissues (14), and
functionally distinct rat lung DC can be identified by FcR expression
(15). In murine Peyer’s patch, phenotypically distinct DC reside in
the subdome and T cell areas (36), and in murine spleen distinct
populations of DC reside in marginal zones and T cell areas (12). A
population of mouse spleen DC that expresses CD8 can induce
apoptosis in responding T cells during a MLR (27). The relationships of
these different DC populations are obscure.
Here we show that two populations of DC migrate in intestinal lymph.
They differ in expression of CD4 and OX41, morphology, enzyme content,
surface marker expression, survival in culture, and as APC for T cells.
One explanation for this is that CD4+ DC are precursors of
CD4- DC. CD4+ DC do have features suggesting
relative immaturity. They have fewer cytoplasmic inclusions, lower
esterase content, and lower B7 expression. We think that this
explanation is unlikely. Flow cytometry with anti-CD4 or OX41
defines discrete DC populations with few cells expressing intermediate
levels. In contrast to LC (37), which express low levels of MHC class
II, both populations of L-DC express similar levels. If
CD4- DC spent an obligatory period as CD4+ DC,
we would predict that they would take longer to exit the intestinal
wall. However, kinetic studies show that both populations spend similar
times in the intestinal wall. In addition, cultured
CD4+/OX41+ DC do not acquire the
characteristics of CD4- DC. CD4+ DC do not
change their expression of nonspecific esterase after 48 h in
culture and do not acquire cytoplasmic inclusions, and the differences
in the processes displayed by the two populations are clearer after
culture than in fresh cells. The two populations respond differently as
APC in culture. Thus, CD4- DC rapidly lose the ability to
process/present native Ag to T cells, but CD4+ DC retain
this ability for up to 72 h. Importantly, fresh CD4-
DC are weaker stimulators of a MLR than CD4+ DC, but in
culture become as strong as CD4+ cells. If CD4+
DC were precursors of CD4- DC, we would expect them to
express higher levels of Ii, but they do not. CD4- DC lose
Ii expression in culture, correlating with their loss of Ag processing,
but CD4+ DC do not change Ii expression. These observations
are difficult to reconcile with CD4+ DC being precursors of
CD4- DC. We cannot exclude the possibility that
CD4+ DC might develop the properties of CD4-
DC if they were given different stimuli, e.g., LPS or TNF-, but our
results show that under steady state conditions, some DC will reach the
draining node still able to process Ag, and these DC would not be
stimulated by LPS or TNF- in the node.
CD4- DC are not macrophages because they constitutively express high levels of MHC class II and B7, are nonphagocytic, and do not adhere to glass or plastic. They can stimulate a MLR, and after culture they become as potent as CD4+ DC. Importantly, they can sensitize naive T cells in vivo, and this cannot be explained by contamination with CD4+ DC. Nonactivated rat macrophages do not stimulate a significant MLR or sensitize naive T cells in vivo (2). CD4- DC are also distinct from Ag-laden cells described in rat TDL (38), as those cells are adherent, phagocytic, and MHC class II-. CD4- DC could derive from a phagocytic monocyte that differentiates into a DC under the influence of local stimuli, as human DC can arise from monocytes (20, 39). We have no evidence of the relationship of either population of DC to the CD8+ DC identified in murine thymus and spleen (23, 27).
The central function of DC is Ag presentation, and the two populations in lymph show striking differences. CD4+ L-DC are at least 10-fold more potent than negative cells in presentation of Ags to sensitized T cells and in priming of naive T cells in vivo. Separated CD4+ L-DC are less potent than unseparated cells as APC and in MLR stimulation. This is not due to the rosetting procedure, as OX41-selected L-DC are as potent as unseparated cells. Perhaps CD4 on L-DC interacts with MHC class II, expressed on activated rat CD4+ T cells (40). It may be significant that murine DC are CD4-, and activated T cells do not express MHC class II.
At least two factors may contribute to the low APC activity of CD4- L-DC. Cultured CD4- L-DC rapidly lose the ability to process native Ag, and some fresh L-DC may have already lost this ability. In addition, CD4- L-DC survive poorly in culture, and more may have died before or during interaction with T cells. It is unlikely that CD4- DC induce apoptosis in T cells (27), as after culture, CD4- DC become as potent as CD4+ DC in MLR stimulation. Both populations express similar levels of B7, but we cannot assess the expression of B7.1 and B7.2 in the rat.
Several reports have shown that cultured DC shut off Ag processing (3, 41, 42, 43). In contrast, we have shown that cultured L-DC can process Ag for at least 72 h in culture (32). Our results now show that Ag handling by DC subpopulations differs in culture. CD4+ L-DC remain able to process and present Ag efficiently for up to 72 h, and their ability to stimulate a MLR does not change. In contrast, CD4- DC cannot process and present native Ag after 20 h in culture, whereas their ability to stimulate a MLR increases, and they become as potent as CD4+ cells. The changes in CD4- DC are strikingly similar to those in cultured LC (44), but the stability of Ag processing by CD4+ L-DC in culture is clear-cut. In contrast to cultured murine LC, which rapidly shut down synthesis of Ii (45), cultured CD4+ L-DC continue to express Ii for at least 48 h, and levels of expression remain constant. This suggests that the maintenance of Ag processing by CD4+ DC is not due to a small number of incompletely matured DC (46), but is a stable characteristic of this population. This contrasts with a current concept of DC differentiation. It is hypothesized that down-regulation of Ag processing in cultured LC permits retention of an "image" of Ags encountered in the periphery and prevents replacement by self Ags in the node. It might, however, be advantageous if some DC remained able to process Ags in nodes. It is probable that DC carry viruses, bacteria, and parasites to nodes (47), and continued processing of contained Ags would ensure the maintenance of the immune response.
We conclude that the intestine and perhaps other peripheral tissues release two distinct populations of DC into lymph that have different functional and phenotypic characteristics and that may have differential roles in the initiation and the regulation of immune responses.
Note added in proof. Since the submission of this paper, we have become aware of a publication in which similar populations of dendritic cells have been described in bovine skin afferent lymph (48).
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Center for Neurologic Diseases, Brigham and Women’s Hospital, Harvard Medical School, 221 Longwood Ave., Boston, MA 02115.
3 Address correspondence and reprint requests to Dr. G. G. MacPherson, Sir William Dunn School of Pathology, University of Oxford, South Parks Rd., Oxford, United Kingdom OX1 3RE.
4 Abbreviations used in this paper: DC, dendritic cell; LC, Langerhans cell; GM-CSF, granulocyte-macrophage CSF; L-DC, lymph dendritic cell; KLH, keyhole limpet hemocyanin; XTDL, thoracic duct lymph from a mesenteric-lymphadenectomized rat; TDL, thoracic duct lymph; RAM, rat anti-mouse; PE, phycoerythrin; Ii, invariant chain; BrdUrd, bromodeoxyuridine; Cr3, type 3 complement receptor.
Received for publication July 29, 1997. Accepted for publication March 30, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. J. Exp. Med. 179:1109.. Adv. Exp. Med. Biol. 378:17.[Medline]
-chain: detection by monoclonal antibodies directed against a C-terminal -chain segment. Eur. J. Immunol. 22:1413.[Medline]
. J. Exp. Med. 182:1111. for the development of human dendritic/Langerhans cells from cord blood CD34+ hematopoietic progenitor cells. Blood 87:2376. exhibit phenotype and function of blood dendritic cells. Adv. Exp. Med. Biol. 378:75.[Medline]
maintains the viability of murine epidermal Langerhans cells in culture, but in contrast to granulocyte/macrophage colony-stimulating factor, without inducing their functional maturation. J. Exp. Med. 171:159.This article has been cited by other articles:
|
|
 |
|
  K. Naito, T. Anzai, Y. Sugano, Y. Maekawa, T. Kohno, T. Yoshikawa, K. Matsuno, and S. Ogawa Differential Effects of GM-CSF and G-CSF on Infiltration of Dendritic Cells during Early Left Ventricular Remodeling after Myocardial Infarction J. Immunol., October 15, 2008; 181(8): 5691 - 5701. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Yrlid, V. Cerovic, S. Milling, C. D. Jenkins, J. Zhang, P. R. Crocker, L. S. Klavinskis, and G. G. MacPherson Plasmacytoid Dendritic Cells Do Not Migrate in Intestinal or Hepatic Lymph J. Immunol., November 1, 2006; 177(9): 6115 - 6121. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F.-X. Hubert, C. Voisine, C. Louvet, J.-M. Heslan, A. Ouabed, M. Heslan, and R. Josien Differential Pattern Recognition Receptor Expression but Stereotyped Responsiveness in Rat Spleen Dendritic Cell Subsets J. Immunol., July 15, 2006; 177(2): 1007 - 1016. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. De Bernardis, R. Lucciarini, M. Boccanera, C. Amantini, S. Arancia, S. Morrone, M. Mosca, A. Cassone, and G. Santoni Phenotypic and Functional Characterization of Vaginal Dendritic Cells in a Rat Model of Candida albicans Vaginitis Infect. Immun., July 1, 2006; 74(7): 4282 - 4294. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Yrlid, C. D. Jenkins, and G. G. MacPherson Relationships between Distinct Blood Monocyte Subsets and Migrating Intestinal Lymph Dendritic Cells In Vivo under Steady-State Conditions J. Immunol., April 1, 2006; 176(7): 4155 - 4162. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Bonneau, M. Epardaud, F. Payot, V. Niborski, M.-I. Thoulouze, F. Bernex, B. Charley, S. Riffault, L. A. Guilloteau, and I. Schwartz-Cornil Migratory monocytes and granulocytes are major lymphatic carriers of Salmonella from tissue to draining lymph node J. Leukoc. Biol., February 1, 2006; 79(2): 268 - 276. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. L. Turnbull, U. Yrlid, C. D. Jenkins, and G. G. MacPherson Intestinal Dendritic Cell Subsets: Differential Effects of Systemic TLR4 Stimulation on Migratory Fate and Activation In Vivo J. Immunol., February 1, 2005; 174(3): 1374 - 1384. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Epardaud, M. Bonneau, F. Payot, C. Cordier, J. Megret, C. Howard, and I. Schwartz-Cornil Enrichment for a CD26hi SIRP- subset in lymph dendritic cells from the upper aero-digestive tract J. Leukoc. Biol., September 1, 2004; 76(3): 553 - 561. [Abstract] [Full Text] [PDF] |
) suggestive
of secondary lysosomes. The nuclei of the CD4+ and
CD4- DC are similar.
