| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Immunology, Max Planck Institute for Infection Biology, Berlin, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionNote added in proof.References |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionNote added in proof.References |
|---|
, kill infected
target cells, and can also kill mycobacteria directly (8, 9). These findings suggest an important role for group I CD1
molecules in immunity to tuberculosis and other mycobacterial
infections.
Mycobacteria reside inside host macrophages (M
) in phagosomes that
are nonacidified due to a paucity in the vacuolar H+ATPase and are
therefore restricted in their capacity to fuse with late
endosomes/lysosomes (10). Consequently, mycobacterial
phagosomes are positive for the early endosomal markers rab5,
transferrin, and transferrin receptor (TfR) (11, 12, 13).
Therefore, we consider it important to clarify whether, how, and where
mycobacterial glycolipid Ags intersect with CD1 molecules during
infection. Prigozy et al. (14) have shown that the M
mannose receptor (MMR), a pattern recognition/scavenger receptor (PRR)
expressed on DC, is involved in delivery of exogenously added LAM into
CD1b-bearing vesicles. These vesicles are similar to those late
endosomal/lysosomal compartments where MHC class II molecules are
loaded with antigenic peptides, i.e., the MHC class II compartments
(MIIC) (15, 16). The studies presented here suggest that
during infection all three group I CD1 molecules have access to
mycobacterial phagosomes, although at different stages of maturation.
In addition, CD1b and CD1c can intersect with glycolipids such as LAM
and PIM, which are released from the mycobacterial surface and
transported into late endosomes/lysosomes. This indicates distinct
intracellular pathways for CD1a, CD1b, and CD1c similar to those
established for MHC class I and class II, respectively. Our findings
suggest that CD1a, CD1b, and CD1c can sample mycobacterial cell wall
glycolipids from distinct intracellular sites of the infected host cell
and, hence, that they contribute to protection against tuberculosis by
broadening the range of T cell Ags available to the host immune
system.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionNote added in proof.References |
|---|
Murine mAbs against CD1a (10H3.9.3) and CD1b (4A7.6.5) were provided by Dr. Daniel Olive (Marseilles, France) (17), and additional mAb against CD1b and CD1c were supplied by Dr. Walter Knapp (University of Vienna, Vienna, Austria) (18). The CD1a-specific mAb OKT6 was purchased from American Type Culture Collection (Manassas, VA), and additional mAb to CD1b and CD1c were obtained from Serotech (Oxford, U.K.). The murine mAb to human CD14 (M5E2), CD64, and MMR were obtained from PharMingen (Hamburg, Germany). Those to human MHC class II (L243); human HLA-A, -B, and -C (MHC class I; W6/32); and human transferrin receptor (TfR; L5.1) were obtained from American Type Culture Collection. Those against human lysosome-associated membrane protein-1 (LAMP-1) (H4A3) were a gift from Dr. John August (provided through Developmental Hybridoma Bank, University of Iowa, Ames, IA). The murine mAb against PIM (3D6) (19) and LAM (CS35) (20) were provided by Dr. Mario Ehlers (University of Cape Town, Cape Town, South Africa) and Dr. John Belisle (University of Colorado, Fort Collins, CO), respectively. The mAb to lysobiphosphatic acid (LBPA; 2C6) was a gift from Drs. Toshihide Kobayashi and Jean Gruenberg, (University of Geneva, Geneva, Switzerland) (21). Rabbit polyclonal Abs to human cathepsins B and D were purchased from Calbiochem (Bad Soden, Germany), and the affinity-purified rabbit polyclonal Abs to the cation-independent mannose-6-phosphate receptor (ciM6PR) was provided by Dr. Albert Haas (University of Wurzburg, Wurzburg, Germany). Species-specific secondary Abs labeled with peroxidase, Texas Red, FITC, Cy2, Cy3, or Cy5 were purchased from Dianova (Hamburg, Germany) and absorbed against Mycobacterium bovis BCG lysates before use. Cells were fixed in 4% paraformaldehyde and embedded in moviol.
Bacteria and labeling
M. bovis BCG (Copenhagen) or M.
tuberculosis (H37Rv) were grown in 7H10 Middlebrook medium
supplemented with oleic acid albumin dextrose complex (OADC; Difco,
Detroit, MI), harvested during logarithmic growth phase (
1–2
x 108/ml), and washed three times in PBS.
M. bovis BCG carrying the green fluorescent protein (GFP)
gene under the 60-kDa heat shock protein promotor were provided by Dr.
Camille Locht (Unit 447, Institut Nationale d la Santé de la
Recherche Médicale, Institute Pasteur de Lille, Lille,
France).
For labeling with N-hydroxysuccinimidyl ester
carboxyfluorescein (NHS-FITC; Boehringer Mannheim, Mannheim, Germany),
bacteria were resuspended in PBS containing 1 mg/ml of the dye and
shaken for 1 h at room temperature. To label mycobacteria with
hydrazide compounds, bacteria were agitated in acetate buffer (pH 5.5)
containing 15 mM sodium periodate for 15 min. The reaction was stopped
by adding 15 mM sodium bisulfite for 5 min, and the sugar moieties on
the bacterial surface were labeled with Alexa568-hydrazide (50 mM;
Molecular Probes, Leiden, The Netherlands) at room temperature for
3 h. Labeled bacteria were washed five times in PBS and twice in
medium before infection. Bacteria retained 95% viability after
labeling. Confocal microscopy using this dye was performed on live DC
and M cultured on coverslips. The cells were kept in Ringer solution
at 37°C for the time of observation.
Cell culture
PBMC from healthy donors (Bloodbank, Charite, Berlin) were
purified from buffy coats using consecutive Ficoll and Percoll
gradients. Cells were resuspended in RPMI (10% FCS) and either
directly plated to produce M or further cultured for 2 or 6 days in
the presence of human rGM-CSF (100 µg/ml) and rIL-4 (100 µg/ml; R&D
Systems, Wiesbaden, Germany) in petri dishes, on multichamber slides
(Nunc, Roskilde, Denmark), or on coverslips to produce DC (37°C, 7%
CO2). Where appropriate, cells were infected with
mycobacteria at a multiplicity of infection of 10:1 for 2 h. Cells
were washed and further incubated for the time periods indicated. Cells
cultured for 6 days in GM-CSF/IL-4 were positive for MHC class I, MHC
class II, CD1a, CD1b, and CD1c, but were negative for CD14 and CD64
(FcI). Cells showing typical DC-like morphology were clustered
around adherent cells (data not shown). Although the latter cells are
more M-like, they nevertheless expressed all three group I CD1
molecules. In the text, all GM-CSF/IL-4-treated cells are termed
DC.
Intracellular tracers
Texas Red-labeled dextran (dex-TR) was purchased from Molecular Probes (Leiden, The Netherlands), and FITC-labeled mannosylated BSA (manBSA) was obtained from Sigma (Munich, Germany). FITC- or Cy3-labeled OVA or transferrin (Tf) were labeled with either NHS-FITC (Boehringer Mannheim) or Cy3 (Dianova) in carbonate buffer. Tagged proteins were purified using Pd10 columns (Pharmacia, Freiburg, Germany). Competition experiments using unlabeled Tf confirmed that the labeled compound retained its receptor-specific binding capacity (12). Tracers were applied either in cell culture medium (dex-TR, OVA-FITC, OVA-Cy3) or Ringer solution (manBSA-FITC, Tf-FITC). Viable cells were fluorescently labeled using 2.5 ng/ml of the vital stain carboxyfluorescein diacetate-NHS (CFDA; Molecular Probes) for 5 min at room temperature in PBS and washed three times in PBS. Cells maintained fluorescence for up to 4 days. Tosyl-activated magnetic beads (Dynal, Hamburg, Germany) were labeled with human IgG (Sigma) and incubated with APC in a ratio of 5:1 for 2 h. For fluorescence microscopy either a conventional Leica DMIRB or a Leica TCS-NT confocal scanner (Leica, Bensheim, Germany) was used. To control the confocal data for false colocalization, independent sequential scans were performed and compared/overlayed thereafter. Confocal data were further processed using the Leica TCS-NT software and Adobe Photoshop.
Western blotting and TLC
To characterize the mycobacterial material that has been labeled with Alexa568-hydrazide, labeled mycobacteria were submitted to SDS-PAGE and Western blotting or to TLC. For SDS-PAGE, 107 mycobacteria/lane were separated on 4–20% polyacrylamide gradient gels containing 4 M urea. Fluorescence labeling was visualized by UV light before Western blotting onto nitrocellulose. Western blots were probed with mAb to LAM and PIM and species-specific secondary Abs labeled with peroxidase (HRP). Blots were developed using an enhanced chemiluminescence procedure (New England Nuclear, Boston, MA). For TLC, 108 mycobacteria were extracted twice for 24 h in chloroform/methanol/water (16/6/1), and the extract was loaded onto high performance TLC silica plates. Plates were developed in the first dimension in chloroform/methanol (80/20; alkaline) and in the second dimension in propanol/water/acetic acid (80/10/10; acidic) and were visualized using UV light. For immunostaining, TLC plates were fixed in Plexigum (Röhm, Darmstadt, Germany), blocked in 10% BSA/0.05% Tween 20 in PBS, stained with mAb and HRP-labeled secondary Abs, and developed by enhanced chemiluminescence.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionNote added in proof.References |
|---|
Uninfected DC expressed CD1a, CD1b, and CD1c at 2 and 6 days after
cytokine stimulation, although the percentage of positive cells varied
among donors (data not shown). The distribution of the three CD1
molecules in uninfected and infected DC as well as the tracers/markers
used in this study, with respect to their specificity for intracellular
compartments, are summarized in Table I
.
CD1a was exclusively found on the plasma membrane of DC and did not
colocalize with tracers/markers for intracellular vesicles (Fig. 1A and Table I). This
distribution is reminiscent of the cellular location of MHC class I
molecules (15). In contrast, CD1b was less intensively
expressed on the plasma membrane, but was strongly expressed in
intracellular compartments. These CD1b-positive compartments
colocalized with late endosome/lysosome-specific tracers/markers, but
not with those specific for early endosomes (Fig. 1B and
Table I). CD1b also partially overlapped with manBSA-FITC internalized
for 2 h but not chased, which labels the continuum between early
and late endosomes (Table I). As previously shown, CD1b distribution
partially colocalized with the late endosomal/lysosomal proteins
cathepsin D, LAMP-1, and MHC class II (14). We have
further extended these observations by showing partial overlap of
CD1b with ciM6PR (Fig. 1D) and LBPA (Table I), which are
markers for late endosomes (15, 21, 22).
|
|
C). Furthermore, CD1c was additionally found
in small vesicles, which in part colocalized with late
endosomal/lysosomal tracers/markers such as manBSA-FITC, OVA-Cy3,
cathepsin D, LAMP-1, MHC class II, LBPA, and ciM6PR (Fig. 1, C and D, and Table I). CD1c was also found in a
small percentage of Tf-FITC-labeled early endosomes (Fig. 1C
and Table I). None of the three group I CD1 molecules colocalized with
a Golgi-specific marker (Table I). This was probably due to their slow
turnover/long half-life and consequently the small amount of newly
synthesized CD1 protein (1). These experiments were
repeated with three different mAb to either CD1a or CD1b and with two
different mAb to CD1c and revealed similar results (data not shown).
Additionally, we also used dex-TR at different incubation times to
trace early endosomes (5-min pulse), late endosomes (30-min pulse), or
lysosomes (2-h pulse, 1-h chase). These experiments gave similar
results with respect to colocalization with the three CD1 molecules
(data not shown). Taken together, the plasma membrane restricted
localization of CD1a differed from additional late endosomal/lysosomal
localization of CD1b and CD1c. Intracellular trafficking of CD1a, CD1b, and CD1c in mycobacteria-infected DC
Infection of DC with M. bovis BCG or M.
tuberculosis for 2–48 h did not alter the distribution of any of
the group I CD1 molecules. However, the overall amount of
surface-expressed CD1a, CD1b, and CD1c was slightly diminished 1–2
days after infection as previously described (data not shown)
(23). All three CD1 molecules were present in phagosomes
containing mycobacteria (Fig. 2A). In DC, mycobacteria
resided in phagosomes that either did not fuse (early endosomal
phagosomes; 66% for BCG phagosomes) or fused with late
endosomes/lysosomes (phagolysosomes; 44%) that were labeled with
OVA-FITC or OVA-Cy3 (16-h pulse, 2-h chase). Interestingly, CD1a was
exclusively present in the early endosomal subset of the mycobacterial
phagosomes negative for OVA labeling (Fig. 2BA). In
contrast, CD1b was only found in the OVA-positive late
endosomal/lysosomal subset (Fig. 2BB). As previously
reported, phagosomes with late endosomal/lysosomal characteristics most
probably represent mature phagolysosomes containing nonviable
mycobacteria (24). Similar to CD1a, CD1c was found in
mycobacterial phagosomes with characteristics for early endosomes (Fig. 2BC). These results were further corroborated by the finding
that CD1b associated strongly with IgG-bead phagosomes, CD1c associated
faintly, and CD1a not at all (Fig. 2C). Thus, although all
three CD1 molecules intersect with intracellular mycobacteria they do
so in phagosomes at different stages of maturation: early endosomal for
CD1a and CD1c vs late endosomal/lysosomal for CD1b.
|
As previously shown by immunoelectron microscopy, LAM is exported
from the mycobacterial phagosome and delivered into lysosomes
(25). To analyze the overall trafficking of mycobacterial
glycolipids from the phagosomes into the host cell vesicular system,
terminal sugar moieties of cell wall components of M. bovis
BCG, either wild type or expressing GFP, were labeled using
Alexa568-hydrazide. The labeled material was analyzed by
SDS-PAGE/Western blot as well as TLC as shown in Fig. 3. The two main fluorescent species
observed by SDS-PAGE were a smear of 30–40 kDa and a band of 12
kDa running close to the dye front (Fig. 3A). Western
blotting using specific mAb revealed that these bands corresponded to
LAM (a smear between 30 and 40 kDa) (20) and PIM (12
kDa) (19), respectively (Fig. 3A). By TLC, we
were able to separate six Alexa568-hydrazide-labeled compounds of
mycobacterial origin (Fig. 3B, arrows, spots 1–6).
Furthermore, TLC analysis revealed that Alexa568-labeled mycobacteria
were devoid of any uncoupled dye (Fig. 3, B vs
C). The compounds separated by TLC are currently under
investigation, but are distinct from LAM (data not shown).
Immunostaining of the TLC revealed that three of these spots react with
an mAb specific for PIM (gray arrows, spots 3, 5, and 6, Fig. 3, B vs D) (19). The presence of three
differently migrating spots staining with this mAb probably reflects
PIM with various amounts of mannose. When analyzed 24 h p.i. in
isolated phagosomes and nonphagosomal vesicles, spots 3–6 were still
associated with the bacteria, whereas spots 1 and 2 were only found in
other vesicles (data not shown).
|
A). This picture changed at
10 h p.i. and was even more pronounced 24 and 48 h p.i. At
these time points the Alexa568-labeled glycolipids were widely
distributed through the intracellular network, including tubular as
well as large perinuclear lysosomal structures (Fig. 4A).
This transfer of labeled glycolipids was not observed when mycobacteria
were heat killed before labeling with Alexa568-hydrazide (data not
shown). Following additional labeling of infected cells with
intracellular tracers, we found that exported glycolipids colocalized
predominantly with the late endosomal/lysosomal tracers, OVA-FITC and
manBSA-FITC, but not with the early endosomal tracer Tf-FITC (Fig. 4B). Similar results were obtained in DC and M (data not
shown). Fluorescence labeling of mycobacterial surface proteins with
NHS-FITC did not lead to this vast export of labeled material from the
phagosome, suggesting that glycoproteins are only a small part of the
exported material labeled with Alexa568-hydrazide (data not shown).
|
infected with Alexa568-labeled M.
bovis BCG and uninfected DC labeled with CFDA. By 24 h most
of the bystander DC contained significant amounts of mycobacterially
derived label (Fig. 4C). Longer than 48-h coculture resulted
in bystander DC becoming infected (data not shown). These data reveal
that mycobacterial glycolipids are exported from phagosomes containing
live mycobacteria into late endosomes/lysosomes and to uninfected APC
in their vicinity. MMR expression in early endosomal compartments
The MMR has been proposed as a putative receptor and transport
molecule to deliver mycobacterial glycolipids to CD1b-bearing vesicles
(14). In our experiments the intracellular distribution of
the MMR in DC was mainly restricted to small peripheral vesicles and
colocalized with the early endosome-specific tracer Tf-FITC and
strongly with manBSA-FITC, which labels the continuum between early and
late endosomes (2-h pulse, no chase; Fig. 5). Localization of the MMR did not
overlap with late endosome/lysosome-specific tracers (Fig. 5) or
markers (cathepsin D, ciM6PR, MHC class II; Table I). The intracellular
distribution of the MMR was not altered upon infection of the DC with
mycobacteria, and the MMR was not observed in mycobacterial phagosomes
(Fig. 5). These data suggest that the MMR does not participate in
intracellular glycolipid transfer from the phagosome but, rather, is
involved in the uptake of free mycobacterial glycolipids by uninfected
DC (14).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionNote added in proof.References |
|---|
/
T cells
(27). Recently, T cells recognizing glycolipids in
conjunction with CD1 have been added to the antimycobacterial armory
(9). Of the three group I CD1 molecules, only CD1b has
been studied in detail with respect to its intracellular localization
(14). These experiments were performed with purified LAM,
which leaves open how, or even if, mycobacteria intersect with CD1
molecules during natural infection. The data presented here show that
all three group I CD1 molecules intersect with mycobacterial
phagosomes, although at different stages; whereas CD1a and CD1c gain
access to phagosomes arrested at the early endosomal stage, CD1b
trafficks to phagolysosomes. Furthermore, CD1b and CD1c molecules
intersect with late endosomes/lysosomes carrying exported
glycolipids. The intracellular distribution of CD1b is similar to that of MHC class II, i.e., it has been localized together with HLA-DR and the invariant chain to late endosomal/lysosomal multilamellar vesicles reminiscent of the MIIC (14, 15). The MIIC also contains CD63, cathepsin D, and LAMP-1 and accesses the fluid phase marker dex-TR (14). In addition to these late endosomal/lysosomal markers, we show that CD1b associates with vesicles accessing manBSA. This tracer is taken up via the MMR and is released from its receptor in an early compartment to be delivered to late endosomes/lysosomes. Furthermore, we identified CD1b in vesicles carrying LBPA (21) and the ciM6PR (22), markers for late endosome.
In contrast to CD1b, CD1a colocalized neither with markers for early endosomes nor with those for late endosomes/lysosomes. Rather, CD1a was exclusively expressed on the plasma membrane, similar to MHC class I molecules (15). This observation is consistent with previous findings that the cytoplasmic tail of CD1a does not contain the endosomal targeting sequence YXXZ (Y = tyrosine, X = any amino acid, Z = bulky hydrophobic residue) present in other CD1 molecules (1, 28). Mutations introduced into the endosomal targeting motif of CD1b prevent these molecules from trafficking to the MIIC and inhibit presentation of mycobacterial glycolipids such as mycolic acids to T cells (29).
Presentation of glycolipid Ags by CD1b depends on the integrity of acidified compartments as a prerequisite for processing of Ags and their loading into the Ag binding groove of the CD1 molecules (27, 28). Similar to processing/presentation of MHC class II-restricted protein Ags, chloroquine inhibits CD1b-mediated T cell stimulation (5, 28, 29). Moreover, the CD1b molecule accommodates its ligands more efficiently when loading is performed at low pH, which probably facilitates widening of the groove (30). It can also be speculated that glycolipids are processed by host cell-derived lysosomal enzymes, which have lower pH optima. Preliminary data show that mycobacterial glycolipids are modified in infected APC (K. Fischer et al., unpublished observations). The data presented here, in combination with the presence of the YXXZ motif in the cytoplasmic tail of CD1c, suggests that CD1c is also trafficked through late endosomes/lysosomes and may pick up glycolipid Ags in an acidified environment similar to CD1b, although direct evidence is not yet available. The intracellular distribution of CD1c was not completely identical with that of CD1b. Compared with CD1b, CD1c was expressed at higher levels on the plasma membrane and less intensely in intracellular vesicles. CD1c only partially overlapped with late endosomes/lysosomes and was additionally present in mycobacterial phagosomes arrested at the early endosomal stage. In addition, CD1c was also found in a few Tf-FITC-labeled early endosomes. Compared with CD1b, CD1c is less efficiently trafficked into phagolysosomes containing beads. This indicates that CD1c can also pick up glycolipid ligands in earlier compartments, such as the early endosomal mycobacterial phagosome similar to CD1a. Although we observed a slight decrease in the overall expression of all three CD1 molecules in mycobacteria-infected cells, as has been described by others (23), their intracellular distribution was not significantly altered during infection. The only apparent difference was the presence in mycobacterial phagosomes of CD1a, which otherwise was plasma membrane associated. The fact that we do not see a substantial decrease in CD1 expression in infected APC as described by Stenger et al. (23) could be due to experimental differences such as mycobacterial strains and different infection rates. The presence of CD1a and CD1c in the mycobacterial phagosome is not surprising in light of earlier reports showing 1) that surface-derived MHC class I and II molecules are present in the mycobacterial phagosome (31), and 2) that there is a continuous exchange of material between the phagosome and the plasma membrane (32). The significant amount of CD1a and CD1c on the cell surface compared with CD1b may facilitate simultaneous uptake or continuous import of these molecules into mycobacterial phagosomes arrested in the early endosomal stage. Although CD1a was present in early mycobacterial phagosomes that accumulated Tf, it was not detected in early endosomes of the transferrin pathway. This could be due to the following explanations. Either endosomes of the early recycling pathway engulf distinct receptors such as the TfR and exclude other surface molecules such as CD1a, or CD1a is recycled at an earlier endosomal stage than Tf.
Mycobacterial survival is facilitated in an early endosomal compartment characterized by an almost neutral pH due to a paucity in the vacuolar H+ATPase (10). The mycobacterial phagosome has access to the Tf/TfR pathway and also carries the early endosomal marker rab5 (11, 12, 13). It is, therefore, an intriguing question how mycobacterial glycolipids are trafficked from the phagosome into CD1-bearing compartments. Our data reveal that mycobacterial glycolipids are exported from the phagosome into late endosomes/lysosomes. A previous immunoelectron microscopy study has described LAM in lysosomes of infected murine macrophages (25). We show that LAM, PIM (19, 20), and three other as yet unidentified glycolipids, fluorescently labeled via their terminal sugar moieties, were released from the mycobacterial surface and transported out of the phagosome into late endosomes/lysosomes. These glycolipids reached compartments carrying CD1b and CD1c. Both LAM and PIM have been described as T cell Ags presented by CD1b and CD1c (3, 5, 7). Trafficking of fluorescent glycolipids has only been observed with live mycobacteria, which usually reside in early endosomal compartments. The early stage of the phagosome probably facilitated export of material into the downstream lysosomal pathway.
The observation that labeled glycolipids were transferred to uninfected
bystander cells may have important implications for CD1-mediated immune
responses. Although CD1-positive DC have been demonstrated in leprosy
lesions and can harbor mycobacteria, M represent the major habitat
for mycobacteria in vivo (24, 33, 34). Macrophages, on the
other hand, do not express CD1 in appreciable amounts. Furthermore,
data by others revealed that CD1b expression is down-regulated upon
infection with M. tuberculosis (23). Therefore,
transfer of Ags from infected to noninfected APC would allow induction
of an immune response even when the Ag-presenting capacity of the
infected APC is impaired. The data presented here suggest that
mycobacterial glycolipids can be transferred from infected M to
CD1-expressing DC to facilitate presentation to T cells. Preliminary
data suggest that the transfer to bystander cells is mediated by
extracellular vesicles, which are released into the culture medium.
These vesicles partly resemble exosomes (35) and blebs
probably generated by apoptosis (U. E. Schaible, V. Brinkmann, and
K. Fischer, unpublished observations), which is a well described
outcome of mycobacterial infection in APC (36, 37). This
transfer was not inhibited by Abs against the MMR (data not shown) and
may therefore require additional receptors. At present, we cannot
exclude the possibility that Alexa568-hydrazide labeling may alter the
trafficking behavior of cell wall glycolipids. However, the observation
that uptake by macrophages of Alexa568-hydrazide-labeled purified LAM
can be competed by nonlabeled LAM would argue against this idea (data
not shown).
It is of interest that in DC we found relatively high numbers of mycobacteria in phagolysosomes. This could be due to the cell type investigated, i.e., human IL-4/GM-CSF-induced DC, and could account for the fact that in these cells CD1b can intersect with the late endosomal/lysosomal subset of phagosomes probably containing nonviable mycobacteria (24). The presence of CD1b, but not CD1a, in mycobacterial phagosomes of the late endosomal/lysosomal stage is consistent with the presence of CD1b in phagolysosomes containing IgG-coupled beads. These data together with a recent report showing that phagolysosomes represent competent compartments to process protein Ags for MHC class II presentation to T cells (38) suggest that CD1b can pick up and present glycolipids from dying mycobacteria in phagolysosomes.
Intracellular trafficking and intercellular transfer of glycolipids
probably involve receptor molecules of the PRR family, such as the MMR.
The MMR participates in delivery of exogenously added LAM to
CD1b-bearing vesicles, thereby facilitating presentation of LAM to
CD1b-restricted T cells (14). However, the MMR is
predominantly present in early endosomes. As shown here, the MMR
resides in some Tf- or manBSA-carrying vesicles. The latter ones are
probably compartments where mannosylated material is separated from the
MMR and trafficks further down the lysosomal pathway (39).
These manBSA-labeled vesicles can also carry CD1b and CD1c and may
represent the meeting point for MMR-delivered material and these
molecules. To date, the MMR has not been localized to mycobacterial
phagosomes, and therefore may not be involved in direct transfer of
glycolipids from the phagosome into late endosomes/lysosomes. Because
the MMR is predominantly expressed in early endosomes and on the plasma
membrane, it may, rather, participate in the engulfment of free
glycolipids by bystander cells, as shown for purified LAM
(14). This raises the question of whether other PRR are
involved in glycolipid transfer from phagosomes to lysosomes and
bystander cells. CD14, which binds LAM directly (40), is
present in mycobacterial phagosomes as well as in late
endosomes/lysosomes (U. E. Schaible et al., manuscript in
preparation). Hence, CD14 is a candidate for glycolipid trafficking in
mycobacteria-infected M. Furthermore, CD14 has been shown to be
involved in the uptake of apoptotic blebs by human M
(41). Moreover, we cannot exclude participation of other
PRR or the possibility that glycolipids such as LAM, which can directly
integrate into the host cell membrane, are trafficked independently of
specific receptors, similar to glycosyl phosphatidyl inositol-anchored
cellular proteins (42).
Purified mycobacterial glycolipids are presented by group I CD1 molecules to specialized T cells (43). This finding has been taken as evidence that this T cell population plays an important role in protection against tuberculosis. Here we identified distinct intracellular compartments where mycobacteria, mycobacterial glycolipids, and the three group I CD1 molecules intersect during mycobacterial infection. This suggests that CD1a, CD1b, and CD1c can sample mycobacterial glycolipid species (shed vs bacteria-associated) from distinct intracellular sites and bacterial stages (growing vs nongrowing). These findings will allow further dissection of the processing pathways for mycobacterial glycolipids that lead to stimulation of CD1-restricted T cells. Further investigations are directed at determining the value of such glycolipid Ags for a novel antituberculosis vaccine.
|
Note added in proof. |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionNote added in proof.References |
|---|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Ulrich E. Schaible, Department of Immunology, Max Planck Institute for Infection Biology, Monbijoustrasse 2, D-10117 Berlin, Germany.
3 Abbreviations used in this paper: DC, dendritic cells; LAM, lipoarabinomannan; PIM, phosphatidylinositol mannosides; M, macrophages: TfR, transferrin receptor; MMR, M mannose receptor; PRR, pattern recognition/scavenger receptor; MIIC, MHC class II compartments; LBPA, lysobiphosphatic acid; BCG, bacillus Calmette-Guérin; GFP, green fluorescent protein; NHS, N-hydroxysuccinimidyl ester carboxyfluorescein; dex-TR, Texas Red-labeled dextran; Tf, transferrin; CFDA, carboxyfluorescein diacetate; p.i., postinfection; LAMP-1, lysosome-associated membrane protein-1; PFA, paraformaldehyde.
Received for publication September 27, 1999. Accepted for publication February 22, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionNote added in proof.References |
|---|
ß+ T cells. Nature 372:691.[Medline]
ß T cell pool. J. Immunol. 162:366./ and other unconventional T lymphocytes: what do they see and what do they do?. Proc. Natl. Acad. Sci. USA 93:2272.This article has been cited by other articles:
|
|
 |
|
  I. Van Rhijn, D. C. Young, A. De Jong, J. Vazquez, T.-Y. Cheng, R. Talekar, D. C. Barral, L. Leon, M. B. Brenner, J. T. Katz, et al. CD1c bypasses lysosomes to present a lipopeptide antigen with 12 amino acids J. Exp. Med., June 8, 2009; 206(6): 1409 - 1422. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Zajonc, G. D. Ainge, G. F. Painter, W. B. Severn, and I. A. Wilson Structural Characterization of Mycobacterial Phosphatidylinositol Mannoside Binding to Mouse CD1d J. Immunol., October 1, 2006; 177(7): 4577 - 4583. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Torres, L. Ramachandra, R. E. Rojas, K. Bobadilla, J. Thomas, D. H. Canaday, C. V. Harding, and W. H. Boom Role of Phagosomes and Major Histocompatibility Complex Class II (MHC-II) Compartment in MHC-II Antigen Processing of Mycobacterium tuberculosis in Human Macrophages Infect. Immun., March 1, 2006; 74(3): 1621 - 1630. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. A. Lang, P. A. Illarionov, A. Glatman-Freedman, G. S. Besra, and M. L. Lang BCR targeting of biotin-{alpha}-galactosylceramide leads to enhanced presentation on CD1d and requires transport of BCR to CD1d-containing endocytic compartments Int. Immunol., July 1, 2005; 17(7): 899 - 908. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Bergman, L. A. Cummings, S. L. R. Barrett, K. D. Smith, J. C. Lara, A. Aderem, and B. T. Cookson CD4+ T Cells and Toll-Like Receptors Recognize Salmonella Antigens Expressed in Bacterial Surface Organelles Infect. Immun., March 1, 2005; 73(3): 1350 - 1356. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. A. Sieling, J. B. Torrelles, S. Stenger, W. Chung, A. E. Burdick, T. H. Rea, P. J. Brennan, J. T. Belisle, S. A. Porcelli, and R. L. Modlin The Human CD1-Restricted T Cell Repertoire Is Limited to Cross-Reactive Antigens: Implications for Host Responses against Immunologically Related Pathogens J. Immunol., March 1, 2005; 174(5): 2637 - 2644. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Martino, A. Sacchi, N. Sanarico, F. Spadaro, C. Ramoni, A. Ciaramella, L. P. Pucillo, V. Colizzi, and S. Vendetti Dendritic cells derived from BCG-infected precursors induce Th2-like immune response J. Leukoc. Biol., October 1, 2004; 76(4): 827 - 834. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Fischer, E. Scotet, M. Niemeyer, H. Koebernick, J. Zerrahn, S. Maillet, R. Hurwitz, M. Kursar, M. Bonneville, S. H. E. Kaufmann, et al. Mycobacterial phosphatidylinositol mannoside is a natural antigen for CD1d-restricted T cells PNAS, July 20, 2004; 101(29): 10685 - 10690. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Gilleron, S. Stenger, Z. Mazorra, F. Wittke, S. Mariotti, G. Bohmer, J. Prandi, L. Mori, G. Puzo, and G. De Libero Diacylated Sulfoglycolipids Are Novel Mycobacterial Antigens Stimulating CD1-restricted T Cells during Infection with Mycobacterium tuberculosis J. Exp. Med., March 1, 2004; 199(5): 649 - 659. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Elewaut, A. P. Lawton, N. A. Nagarajan, E. Maverakis, A. Khurana, S. Honing, C. A. Benedict, E. Sercarz, O. Bakke, M. Kronenberg, et al. The Adaptor Protein AP-3 Is Required for CD1d-Mediated Antigen Presentation of Glycosphingolipids and Development of V{alpha}14i NKT Cells J. Exp. Med., October 20, 2003; 198(8): 1133 - 1146. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Cernadas, M. Sugita, N. van der Wel, X. Cao, J. E. Gumperz, S. Maltsev, G. S. Besra, S. M. Behar, P. J. Peters, and M. B. Brenner Lysosomal Localization of Murine CD1d Mediated by AP-3 Is Necessary for NK T Cell Development J. Immunol., October 15, 2003; 171(8): 4149 - 4155. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Porcaro, M. Vidal, S. Jouvert, P. D. Stahl, and J. Giaimis Mannose receptor contribution to Candida albicans phagocytosis by murine E-clone J774 macrophages J. Leukoc. Biol., August 1, 2003; 74(2): 206 - 215. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H.E. Kaufmann and U. E. Schaible A Dangerous Liaison between Two Major Killers: Mycobacterium tuberculosis and HIV Target Dendritic Cells through DC-SIGN J. Exp. Med., January 6, 2003; 197(1): 1 - 5. [Full Text] [PDF]
|
|
|
|
|
|
L. Tailleux, O. Schwartz, J.-L. Herrmann, E. Pivert, M. Jackson, A. Amara, L. Legres, D. Dreher, L. P. Nicod, J. C. Gluckman, et al. DC-SIGN Is the Major Mycobacterium tuberculosis Receptor on Human Dendritic Cells J. Exp. Med., January 6, 2003; 197(1): 121 - 127. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Goffin and J.-M. Ghuysen Biochemistry and Comparative Genomics of SxxK Superfamily Acyltransferases Offer a Clue to the Mycobacterial Paradox: Presence of Penicillin-Susceptible Target Proteins versus Lack of Efficiency of Penicillin as Therapeutic Agent Microbiol. Mol. Biol. Rev., December 1, 2002; 66(4): 702 - 738. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Sirkar and S. Majumdar Lipoarabinomannan-Induced Cell Signaling Involves Ceramide and Mitogen-Activated Protein Kinase Clin. Vaccine Immunol., November 1, 2002; 9(6): 1175 - 1182. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Howard and C. M. Isacke The C-type Lectin Receptor Endo180 Displays Internalization and Recycling Properties Distinct from Other Members of the Mannose Receptor Family J. Biol. Chem., August 23, 2002; 277(35): 32320 - 32331. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Fischer, H. Collins, M. Taniguchi, S. H. E. Kaufmann, and U. E. Schaible IL-4 and T Cells Are Required for the Generation of IgG1 Isotype Antibodies Against Cardiolipin J. Immunol., March 15, 2002; 168(6): 2689 - 2694. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Giuliani, S. P. Prete, G. Graziani, A. Aquino, A. Balduzzi, M. Sugita, M. B. Brenner, E. Iona, L. Fattorini, G. Orefici, et al. Influence of Mycobacterium bovis Bacillus Calmette Guerin on In Vitro Induction of CD1 Molecules in Human Adherent Mononuclear Cells Infect. Immun., December 1, 2001; 69(12): 7461 - 7470. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. d. C. Salamone, A. K. Mendiguren, G. V. Salamone, and L. Fainboim Membrane trafficking of CD1c on activated T cells J. Leukoc. Biol., October 1, 2001; 70(4): 567 - 577. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Fischer, D. Chatterjee, J. Torrelles, P. J. Brennan, S. H. E. Kaufmann, and U. E. Schaible Mycobacterial Lysocardiolipin Is Exported from Phagosomes upon Cleavage of Cardiolipin by a Macrophage-Derived Lysosomal Phospholipase A2 J. Immunol., August 15, 2001; 167(4): 2187 - 2192. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Neyrolles, K. Gould, M.-P. Gares, S. Brett, R. Janssen, P. O'Gaora, J.-L. Herrmann, M.-C. Prevost, E. Perret, J. E. R. Thole, et al. Lipoprotein Access to MHC Class I Presentation During Infection of Murine Macrophages with Live Mycobacteria J. Immunol., January 1, 2001; 166(1): 447 - 457. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
