| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Receptor I on Dendritic Cells Delivers IgE-Bound Multivalent Antigens into a Cathepsin S-Dependent Pathway of MHC Class II Presentation1
*
Division of Immunology, Allergy and Infectious Diseases,
General Dermatology, Department of Dermatology, and
Institute of General and Experimental Pathology, University of Vienna Medical School, Vienna, Austria; and
§
Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA 02115
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RI-mediated Ag uptake and
presentation mechanisms of dendritic cells (DC). We found that
FcRI-bound IgE, after polyvalent but not after monovalent ligation,
is efficiently internalized into acidic, proteolytic compartments,
degraded, and delivered into organelles containing MHC class II,
HLA-DM, and lysosomal proteins. To follow the fate of the fragmented
ligand, we sought to interfere with invariant chain (Ii) degradation, a
process critical for peptide loading of nascent MHC class II molecules.
We found DC to express cathepsin (Cat) S, a cysteine protease involved
in li processing by B cells. Exposure of DC to a specific, active-site
inhibitor of Cat S resulted in the loss of anti-Cat S
immunoreactivity, led to the appearance of an N-terminal Ii remnant,
and decreased the export of newly synthesized MHC class II to the DC
surface. Furthermore, inactivation of Cat S as well as blockade of
protein neosynthesis by cycloheximide strongly reduced
IgE/FcRI-mediated Ag presentation by DC. Thus, multimeric ligands of
FcRI, instead of being delivered into a recycling MHC class II
pathway, are channeled efficiently into MIIC (MHC class II
compartment)-like organelles of DC, in which Cat S-dependent li
processing and peptide loading of newly synthesized MHC class II
molecules occur. This IgE/FcRI-dependent signaling pathway in DC may
be a particularly effective route for immunization and a promising
target for interfering with the early steps of allergen
presentation. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RII-mediated Ag
uptake by cytokine-stimulated human monocyte-derived DC are the
interiorization pathways best investigated to date (3, 4). The nature
and functionality of Ag uptake receptors used by virgin human DC (i.e.,
DC that have not been exposed to robust activation stimuli) are less
well characterized. Freshly isolated, virgin tissue DC are clearly
inferior to monocyte-derived DC in their nonselective fluid-phase
(macro)pinocytotic capacity (5, 6, 7). They may, therefore, rely on the
expression of selective Ag binding and uptake receptors to present the
relevant determinants in quantities needed to ensure vigorous T cell
responses. Among others, the high affinity IgE receptor (FcRI) is a
candidate pattern recognition and Ag uptake moiety expressed by DC of
the skin (i.e., epidermal Langerhans cells (8, 9) and dermal DC (10))
and DC circulating in the peripheral blood (PB-DC) (11). In fact, we
have found previously that targeting of allergen-IgE complexes to
FcRI-bearing PB-DC results in a much stronger allergen-specific T
cell response than that elicited following DC exposure to allergen in
the absence of IgE (11). To unravel the biologic role of this Ag
presentation mechanism and, perhaps, to define the targets for possible
interference strategies, we investigated the cellular and molecular
mechanisms that allow FcRI-bearing DC to specifically recognize,
internalize, and present IgE-defined Ags. |
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Chimeric (human Fc/mouse Fab) IgE
anti-4-hydroxy-3-nitrophenacetyl (cIgE, IgE anti-NP) was from
Serotec (Oxford, U.K.). Biotinylation and PE labeling of cIgE (cIgE
biot, cIgE PE) were performed according to standard protocols.
Purified, monoclonal human myeloma IgG1, IgG2, IgG3, IgG4, IgA2, and
polyclonal human serum IgG were from Chemicon (Temecula, CA). The mouse
anti-FcRI
mAb 15-1 has been described (8). mAbs used for
immunodepletion experiments included anti-CD3, anti-CD11b,
anti-CD16, anti-CD19, anti-CD34, and anti-CD56 (all
IgG1; Immunotech, Marseille, France). Fab fragments of the anti-IgE
mAb (IgG1, clone B3102E8; Southern Biotechnology, Birmingham, AL) were
prepared using a kit (ImmunoPure IgG1 Fab and F(ab')2
Preparation Kit; Pierce, Rockford, IL) following the manufacturer’s
recommendations. FITC-conjugated mouse mAbs used were clone CR3/43
recognizing the ß-chain of HLA-DR, HLA-DP, and HLA-DQ (Boehringer
Mannheim, Mannheim, Germany); clone L243 recognizing HLA-DR/ß, and
anti-CD3 (Leu4; both from Becton Dickinson, Mountain View, CA),
anti-CD11b (Immunotech), anti-CD20 (Leu16; Becton Dickinson),
anti-CD54 (PharMingen, Hamburg, Germany), mAb VICY1 recognizing the
N terminus of invariant chain (Ii, CD74; kindly provided by Dr. O
Majdic, University of Vienna Medical School, Austria), anti-LAMP-1
(CD107a), and anti-LAMP-2 (CD107b; both from PharMingen).
PE-conjugated mAbs included mAb LN-2 (directed against the Ii C
terminus) and IT2.2 (anti-CD86; both from PharMingen). Streptavidin
PE (SA-PE) and anti-HLA-DR PerCP were from Becton Dickinson.
Affinity-purified rabbit F(ab')2 anti-human IgE was
purchased from Central Laboratory of The Netherlands Red Cross Blood
Transfusion Service (CLB; Amsterdam, The Netherlands). The
generation of affinity-purified rabbit anti-cathepsin (Cat) S Abs
has been previously reported (12). The rabbit anti-HLA-DMß (R.DMB/c)
Abs were kindly provided by Dr. P. Cresswell (Yale University School of
Medicine, New Haven, CT). Affinity-purified rhodamine
(TRITC)-conjugated and unlabeled mouse F(ab')2-specific
goat F(ab')2, goat IgG (H+L)-specific donkey
F(ab')2 TRITC, and FITC-labeled goat F(ab')2
anti-rabbit IgG (H+L) were from Jackson ImmunoResearch
(West Grove, PA). Control mAbs were FITC- or PE-labeled mouse IgG1,
IgG2a, and IgG2b (Becton Dickinson). The Cat S inhibitor
N-morpholinurea-leucine-homophenylalanine-vinylsulfone-phenyl
(LHVS; generous gift of Dr. J. T. Palmer, Arris Pharmaceuticals,
South San Francisco, CA) was dissolved at 10 mM in DMSO. Bafilomycin A1
and cycloheximide (CHX) were from Calbiochem (La Jolla, CA).
Cell preparations and lines
E--mononuclear cells (E--PBMC;
2–4 x 108 cells per donor) from healthy individuals
and birch pollen-allergic patients (as defined by clinical criteria,
skin prick test, and radioallergosorbent test) were prepared by
Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) density-gradient
centrifugation of heparinized venous blood, followed by depletion of
SRBC-binding (E+) T lymphocytes. Thereafter, residual T
cells, monocytes/basophils, NK cells, B, and progenitor cells were
labeled with anti-CD3, anti-CD11b, anti-CD16/CD56,
anti-CD19, and anti-CD34 mAbs (2 µg/ml each), respectively,
allowed to bind anti-mouse IgG1-coated paramagnetic particles
(MACS; Milteny Biotec, Bergisch-Gladbach, Germany) and, finally, were
depleted using a magnet (MiniMACS; Milteny Biotec). Non-bead-bound
cells were recovered by two sequential washing steps. This procedure
typically yielded 2 to 4 x 106 cells, 90 to 95% of
which displayed an HLA-DRhigh, CD3-,
CD20-, and CD14-/low immunophenotype and,
thus, fulfill the phenotypic criteria of PB-DC (11). The great
majority of these cells bound monomeric cIgE and reacted with the
anti-FcRI mAb 15-1. The remaining non-DC population consisted
of T cells, B cells, FcRI+ basophils (together <2% of
the total cell population), and some as yet unidentified cells. In
certain experiments, the anti-CD11b mAb was omitted from the mAb
mixture to obtain a lymphocyte-depleted cell population
(Ly-PBMC) that contains not only
HLA-DRhighCD11b-DC, but also
HLA-DR+CD11b+ monocytes and
FcRI+HLA-DR-CD11b+ basophils. T
cells were procured from the E+ PBMC fraction by
NH4Cl-mediated lysis of SRBC and were purified by
anti-HLA-DR-, anti-CD16-, anti-CD19-, anti-CD20-,
anti-CD56-based immunomagnetic depletion.
Flow-cytometry analysis
To release in vivo bound IgE molecules, isolated cells were
subjected to lactic acid treatment (13). Ly-PBMC (5
x 105 in 50 µl) were stained at 4°C with biotinylated
cIgE, IgG1–4, or IgA (each 10 µg/ml), and exposed to anti-CD11b
FITC and anti-HLA-DR PerCP (5 µg/ml each). Cell surface-bound
biotin groups were visualized by SA-PE (1 µg/ml). Aliquots of cell
samples were preincubated for 30 min with purified mAb 15-1 (50
µg/ml) or human IgG (200 µg/ml) before staining. In all other
experiments, isolated acid-stripped cells were allowed to rest for 30
min at 37°C before their further processing. In one experimental
protocol, cIgE biot-loaded Ly-PBMC, either left untreated
or exposed to anti-IgE Fab (10 µg/ml, 30 min at 4°C), were
dispensed in Ca2+/Mg2+-containing HBSS and
cultured at 37°C in the presence of anti-IgE Fab or goat
F(ab')2 anti-mouse F(ab')2 (10 µg/ml
each). After the indicated 37°C culture periods, samples were placed
on ice until the final time point of each experiment was reached and,
thereafter, exposed to anti-CD11b FITC, anti-HLA-DR PerCP, and
SA-PE. For internalization studies, cIgE-PE-loaded Ly-PBMC
were cultured at 37°C in the presence or absence of IgE-specific
rabbit F(ab')2 (30 µg/ml) and, after the indicated
culture periods, were placed on ice until the last sample was
harvested. An aliquot of each sample was subjected to acid elution of
cell surface FcRI-bound cIgE PE. Thereafter, all samples were
exposed to anti-CD11b FITC and anti-HLA-DR PerCP mAbs and
analyzed by flow cytometry. Purified PB-DC were either kept on ice or
cultured at 37°C for 4 to 12 h in granulocyte-macrophage CSF
(100 U/ml; Novartis, Basel, Switzerland)-supplemented, serum-free
UltraCulture medium (BioWhittaker, Walkersville, MD), and then analyzed
by immunostaining or Western blotting (see below). Intact and/or
permeabilized PB-DC were stained with anti-HLA-DR FITC, VICY1 FITC,
LN-2 PE, anti-ICAM-1 FITC, and anti-B7-2 PE. Permeabilization
of PB-DC was performed using a kit (Fix & Perm; An der Grub, Kaumberg,
Austria), according to the manufacturer’s recommendations. Cellular
fluorescence and light scatter parameters were analyzed on a FACScan
(Becton Dickinson). Within the Ly-PBMC population, PB-DC
were gated as HLA-DRhighCD11b- cells,
monocytes as HLA-DR+CD11bhigh, and basophils as
HLA-DR-CD11b+, as previously described (11).
The specific reactivity of biotinylated cIgE or human Ig and of FITC-
or PE-labeled mAbs with a given cell type was expressed as the 
-mean
fluorescence channel number (-MFC; biotinylated cIgE/human Ig,
MFC of the SA-PE reactivity in the presence of the biotinylated Ab
minus the MFC in its absence; FITC/PE-labeled mAbs, MFC of the
reactivity of the FITC/PE-labeled mAb minus MFC of the reactivity of an
isotype- and fluorochrome-matched control mAb). The percentage of
cIgE-PE internalized by HLA-DRhighCD11b-PB-DC
was calculated using the formula: specific PE reactivity of DC upon
acid stripping divided by the specific PE reactivity of
non-acid-exposed DC multiplied by 100. Statistical significancies were
determined using the paired Student’s t test, and
p values of <0.05 were considered to be statistically
significant.
Confocal laser microscopy
A total of 1 to 2 x 105 purified PB-DC was mounted onto individual fields of adhesion slides (Bio-Rad, Richmond, CA) and exposed sequentially to cIgE, mouse anti-IgE Fab, TRITC-conjugated goat F(ab')2, anti-mouse F(ab')2, and donkey F(ab')2 anti-goat IgG (each 10 µg/ml, 30 min at 4°C). Slides were then overlayered with Ca2+/Mg2+-containing HBSS and cultured in a humidified atmosphere at 37°C, 5% CO2. After the indicated culture periods, cells were fixed in 3% paraformaldehyde (Fluka Chemie AG, Buchs, Switzerland)/Ca2+/Mg2+-free PBS, washed, and quenched by washing with 50 mM NH4Cl/PBS. For immunofluorescence-based staining of cytoplasmic Ags, cells were permeabilized with 0.3% saponin (Sigma, St. Louis, MO)/PBS, washed, and preincubated with 10% mouse serum/10% goat serum for 30 min at 4°C. Thereafter, cells were exposed to FITC-labeled anti-HLA-DR (clone CR3/43, diluted 1/10), anti-LAMP-1/2 (diluted 1/5), or isotype-matched control mAbs (2 µg/ml), or were incubated with rabbit anti-HLA-DMß (diluted 1/1000) or control rabbit serum for 30 min at 4°C. All Ab dilutions and the washing buffer were supplemented with 0.1% saponin. After two washings, the binding of rabbit Abs was detected by an incubation step with FITC-labeled goat F(ab')2 anti-rabbit IgG (H+L) (10 µg/ml). After rinsing with PBS/saponin and embedding in Fluoprep medium (BioMerieux, Marcy L’Etoile, France), slides were examined using a confocal laser-scanning microscope system (LSM 410; Zeiss, Oberkochen, Germany) fitted with lasers emitting light at 488 and 543 nm for the excitation of FITC and TRITC, respectively. Images corresponding to fluorescence signals originating from 0.5-µm sections of individual cells were acquired. To demonstrate the relative positional distribution of FITC- and TRITC-labeled Ags, the images collected in the two channels were merged.
Western blot analysis
Native or cIgE/anti-IgE Fab-bound PB-DC were either kept on ice or cultured at 37°C in the presence of the indicated agents before their solubilization in Tris lysis buffer (20 mM, pH 8.2) containing 1% Nonidet P-40 (Sigma), 140 mM NaCl, 2 mM EDTA, 1 mM iodoacetamide, 1 mM PMSF, aprotinin, and leupeptin (both 10 µg/ml; all from Sigma). Lysates were kept on ice for 30 min, reacted with SDS sample buffer, and analyzed by immunoblotting, as previously described (14). Briefly, aliquots of lysates were submitted to 7 to 15% SDS-PAGE and blotted onto Hybond-P nitrocellulose membranes (Amersham Life Science, Buckinghamshire, U.K.). Membranes were blocked with 10% dry milk/0.1% Tween-20 (Sigma)/PBS for at least 6 h. Thereafter, membranes were exposed to mAbs against the Ii N terminus (VICY1), CD45 (mAb MEM-28; kindly provided by Dr. V. Horejsi, Academy of Natural Sciences, Prague, Czech Republic), Cat D (Transduction Laboratories, Lexington, KY), or rabbit anti-Cat S Abs. The binding of peroxidase-conjugated rabbit anti-human IgE (Accurate Chemicals & Scientific, Westbury, NY), goat anti-mouse Ig (1:40,000), or goat anti-rabbit Ig (1:50,000; both from Bio-Rad) was detected using ECLplus developing solution (Amersham). Films were digitalized by high resolution scanning (S-12; UMAX Technologies, Fremont, CA), and integrated band densities were expressed as relative OD values using the VistaScan software package (UMAX Technologies).
Ag-specific T cell proliferation
Lactic acid-exposed, purified PB-DC from atopic individuals were used as stimulator cells and dispensed into 96-well round-bottom microtiter plates (1–2 x 104/well; Costar, Cambridge, MA). Autologous Bet v I-specific T cell clones (TCC) were established and characterized by techniques described previously (15). TCCs used in this study included GZ11 M (peptide specificity Bet v 1, amino acids (aa) 82–93) and GZ15R (aa 119–125); HR1217, HR2131, and HR2149 (aa 142–156); RR10/5 (aa 34–48), RR161/III (aa 76–90), and RR4R (aa 142–156); and HPR 17/II, HPR31/II, HPR 49/II, and HPR 77 (aa 142–156). Before the DC-TCC (5 x 104 cells/well) coculture, PB-DC were subjected to either of two experimental protocols. In the first experimental group, PB-DC were exposed for 2 h at 37°C to 5 nM LHVS or medium before their exposure to 1 µg/ml NP-specific IgE and/or the indicated concentrations of NP-coupled recombinant birch pollen allergen rBet v 1 (Mayerhofer, Linz, Austria) (16). LHVS remained in the culture medium throughout the subsequent DC-TCC coculture period. In the second experimental group, DC were cultured in solvent- or CHX (5 µg/ml)-conditioned medium for a total of 5 h at 37°C. After the initial 2-h incubation period, cultures were substituted with NP-specific IgE (1 µg/ml final concentration). DC were allowed to bind IgE for 30 min and then exposed to rBet v 1-NP at the indicated final concentrations. After a 2.5-h Ag pulse, non-DC-bound culture additives were removed by several washings. DC-TCC cocultures and the appropriate controls were supplemented with stimulatory anti-CD28 mAbs (Leu28, 0.5 µg/ml; Becton Dickinson) to correct for the CHX-induced inhibition of costimulatory molecule expression of DC. All proliferation assays were performed in granulocyte-macrophage CSF (100 U/ml)-supplemented serum-free UltraCulture medium. DC-TCC cocultures were incubated for 48 h before pulsing with 0.5 µCi [3H]thymidine (Amersham) per well for additional 16 h. Incorporation of the radionucleotide was measured by liquid scintillation spectroscopy. Background counts of controls (DC, DC plus Ag, TCC, DC plus TCC, TCC plus Ag, or anti-CD28 mAbs) were always <500 cpm. The results are expressed as the mean cpm (±SD) obtained in duplicate cultures.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In a first series of experiments, we investigated the Ig isotype-
and the IgG subtype-binding capacity of PB-DC
(HLA-DRhighCD11b-Ly-PBMC (11);
referred to as DC thoughout this work) and monocytes. As shown in
Figure 1
A, DC avidly bind
monomeric IgE, but display no or only negligible affinity to the other
isotypes/subtypes. The FcRI dependency of IgE binding by DC was
confirmed by the ability of a mouse anti-FcRI mAb (mAb 15-1)
to prevent the IgE loading of DC. In contrast, monocytes bound IgE less
efficiently than the other monomeric Ig isotypes/IgG subtypes tested
(IgG3, IgG1 > IgG4 >> IgA, IgG2 > IgE, Fig. 1B). To study the fate of FcRI-bound IgE, DC were
monitored for surface-bound IgE upon exposure to monomeric (mouse
anti-IgE Fab) and multimeric (anti-IgE Fab, followed by mouse
F(ab')2-specific goat F(ab')2) ligands. Figure 1C shows that, at 37°C, cross-linking led to a rapid and
almost complete disappearance of DC surface-bound IgE, whereas
monomeric ligands of IgE hardly affected the amount of IgE at the
plasma membrane. Basophils of the same donor and in the same experiment
bound IgE 10-fold more efficiently than DC, but showed only modest
down-regulation of this molecule upon polyvalent ligation (Fig. 1D). To determine whether the disappearance of surface-bound
IgE moieties was due to preferential internalization or shedding of
receptor-ligand complexes, DC were acid stripped and labeled with
defined concentrations of IgE-PE. Following culture at 37°C in medium
alone or in the presence of a multimeric ligand, they were again
exposed to lactic acid to dissociate IgE-PE from surface-bound FcRI.
Figure 1E shows that in the presence, but not in the
absence, of a multimeric ligand, the fluorescent moieties become
rapidly resistant to acid elution, indicating that the receptor-ligand
complexes had moved to an intracellular location. Multimerization of
IgE had no effect on the amount of the total cell-associated PE
fluorescence, reflecting the detectability of this dye irrespective of
its membraneous or intracellular localization (Fig. 1E).
Based on the ratio of lactic acid-resistant and total cell-associated
PE fluorescence, we calculated the internalization efficacy and
kinetics of FcRI-bound IgE. As demonstrated in Figure 1F,
multimerization of DC-bound IgE-IgE receptor complexes results in their
effective (>70%) and rapid (6- to 7-min t1/2
on the cell surface) internalization. Virtually no uptake was seen in
the absence of the cross-linking agent (Fig. 1F).
|
RI complexes on purified DC, while restricted to the cell
surface when the cells were kept on ice, progressively accumulated in
vesicular intracellular compartments after the DC had been shifted to
37°C (Fig. 2). We observed pronounced
patching, capping, and partial internalization at 5 min and,
thereafter, the continuous translocation of the complexes to certain
distinctive cytosolic DC organelles that occur in a perinuclear
location as well as adjacent to the cell membrane. Little or no
fluorescent material remained on the cell surface.
|
RI are subject to
acidification-dependent degradation and selective targeting into
lysosomal, MIIC-like compartments of DC
We then studied, by Western blotting, the fate of FcRI ligands
upon internalization. Lysates of IgE-loaded DC that had been cultured
in the presence or absence of an anti-IgE-directed cross-linking
agent were blotted, and membranes were probed for anti-IgE
reactivity. As shown in Figure 3A, the lysates of DC that had
been consecutively reacted with cIgE/anti-IgE and, then, were
exposed to goat anti-mouse F(ab')2 progressively
lost the anti-IgE-reactive 180-kDa band (representative experiment,
n = 5). No decrease in the amount of intact IgE was
seen in the absence of the cross-linking stimulus (Fig. 3A),
and the 180-kDa anti-IgE reactivity was still discernible even
after a 12-h culture period (not shown). When the amount of DC-bound
IgE was adjusted to the total cellular content of CD45 at any given
time point, it appeared that intact IgE (180 kDa) had a 
20-min
t1/2 upon cross-linking (Fig. 3B). In
two of five experiments, the fast disappearance kinetics was followed
by a slower phase, during which anti-IgE-reactive moieties with
m.w. characteristics of free IgE heavy chains emerged (Fig. 3, A and B). To investigate whether internalized IgE
is subject to acidification-dependent proteolysis, DC were exposed to
bafilomycin A1, a drug that affects lysosomal function by increasing
the lysosomal pH (17). While this compound did not interfere with the
rate of FcRI-mediated internalization of IgE (two independent
experiments, data not shown), it did inhibit the loss of intact IgE and
the appearance of IgE fragments (Fig. 3A). Altogether, these
results demonstrate that FcRI on DC can efficiently and rapidly
direct multimeric ligands into subcellular acidic compartments
containing the machinery for protein degradation.
|
RI,
DC were subjected to immunofluorescence double-labeling procedures and
analyzed by confocal laser microscopy. IgE/FcRI complexes on
purified DC were cross-linked with secondary Abs conjugated to a
pH-stable fluorescent dye (TRITC), washed, and cultured for 40 min at
37°C. Cells were then permeabilized and stained with Abs recognizing
HLA-DR, HLA-DMß, or LAMP-1 (CD107a) and LAMP-2 (CD107b). As shown in
Figure 4, the intracellular vesicles that
are accessible for IgE/FcRI complexes in DC stained prominently for
HLA-DR and HLA-DMß, and were, at least partly, positive for LAMP-2.
These data demonstrate that Ags taken up via FcRI on DC are
preferentially targeted to compartments that display the same antigenic
profile as the lysosomal/late endosomal MIIC previously identified in
immature DC and various other APC (18, 19, 20, 21). In fact, li degradation and
the loading of antigenic peptides onto class II molecules have been
shown to take place in these organelles (21).
|
In view of the critical role of Cat S in Ii processing by
EBV-transformed human B cells (22) and murine splenocytes (23), we
searched for the presence of this cysteine protease in DC. Using
affinity-purified rabbit anti-Cat S Abs and Western blotting, we
detected anti-Cat S-reactive, 28-kDa moieties in lysates prepared
from freshly isolated DC (Fig. 5A). Bands of identical and
also different m.w. were identified in EBV-transformed human B cells
(Fig. 5A). Short-term culture of DC in the presence of 5 nM
LHVS, a specific and irreversible active-site inhibitor of Cat S
(22, 23, 24), but not in solvent-conditioned medium only, resulted in a
complete loss of the DC- and in a reduction of the B cell-expressed
28-kDa anti-Cat S reactivity (Fig. 5A). (Control
experiments revealed a 43-kDa anti-Cat D immunoreactivity that was
not sensitive to LHVS in either of the two cell types (data not
shown).) The other anti-Cat S immunoreactive moieties of B cells
were not reduced by LHVS treatment. The inhibitor-induced loss of Cat S
is not due to toxic effects since culture in the presence of this
compound affected neither DC viability (data not shown) nor the level
of CD45 expression by these cells (Fig. 5A). The further
observation that the exposure of DC lysates to even high concentrations
of LHVS did not alter the level of anti-Cat S immunoreactivity (not
shown) demonstrated that this inhibitor does not interfere with the
immunodetection of Cat S. One may therefore assume that LHVS not only
blocks the function, but also leads to the elimination of Cat S in
viable DC.
|
B, the elimination of Cat S
activity resulted in a significantly decreased culture-induced HLA-DR
surface expression by DC. This observation combined with the further
finding that the total cellular content of HLA-DR was essentially
unaffected by LHVS suggests that, in the absence of functionally intact
Cat S, DC accumulate HLA class II molecules in an intracellular
location. Since the inhibitory effect of CHX on HLA-DR synthesis could
not be enhanced by the addition of LHVS, it appears that de novo
synthesized rather than preexisting HLA class II molecules require Cat
S-mediated, lysosomal Ii degradation for their efficient transport to
the cell surface. In line with previous findings in human B cells (22),
the functional inactivation of Cat S in DC led to the accumulation of
intracellular, anti-N terminus-reactive Ii. These moieties
apparently represent newly synthesized Ii, since the LHVS-induced
effect was not observed in the presence of CHX. In contrast, the
culture-induced surface expression of mAb LN-2 (anti-Ii C
terminus)-reactive, intact Ii was not changed significantly by LHVS
(Fig. 5B). It is not yet clear whether these LN-2-reactive
Ii moieties are MHC class II ß associated. In DC, such complexes
have been shown to exist and to undergo Ii degradation after their
internalization from the cell surface (25). In contrast to the profound
effects on HLA-DR and Ii expression, the inhibition of Cat S activity
did not impair the culture-induced up-regulation of the costimulatory
molecules B7-2 and ICAM-1 (Fig. 5B).
The molecular nature of the LHVS-inducible anti-Ii-reactive
moieties was investigated by Western blotting using a mAb directed
against the N terminus of Ii. Purified PB-DC (and B cells) express the
p33 (Fig. 5B) and, to a lesser extent, p41 Ii isoforms
(visible after prolonged exposure time only; data not shown). DC
culture in LHVS- but not in solvent-conditioned medium led to the
appearance and pronounced accumulation of an N-terminal Ii breakdown
intermediate of approximately 13 kDa both at early (4 h, data not
shown) and later time points of the culture (12 h, Fig. 5C),
while the steady state level of the Ii p33 was not influenced by LHVS.
These findings demonstrate that Cat S is essential for complete Ii
degradation by DC and, thus, perhaps for the proper Ii-dependent Ag
presentation.
Essential role of Cat S-mediated Ii processing for conventional and receptor-mediated Ag presentation by DC
To explore the relative contribution of the Ii-dependent vs the
Ii-independent pathway for FcRI-mediated Ag presentation, we used
NP-conjugated recombinant birch pollen allergen (rBet v
1-NP) as well as synthetic Bet v 1 peptides as model Ags,
NP-specific IgE, and HLA-DR-restricted T cell clones (TCC) to assess
peptide presentation by purified, autologous DC (11, 16). Bet
v 1 peptides, which bind to cell surface-expressed HLA-DR and,
conceivably, are presentated in an Ii-independent manner, were added to
DC-T cell cocultures with or without LHVS. As exemplified in Figure 6A, DC can present synthetic
peptides in a largely Cat S-independent fashion (mean inhibition of
peptide-induced TCC proliferation by LHVS: 4.1 ± 16.1%; seven
independent experiments). Expectedly, IL-2-stimulated TCC proliferation
was not affected by LHVS treatment. In striking contrast, functionally
active Cat S was essential for the optimal presentation of epitopes
generated from intact protein (Fig. 6B). When DC were
exposed to high Ag concentrations (10 µg/ml), the blockade of Cat S
activity resulted in proliferative TCC responses that were reduced
drastically, albeit not entirely abrogated (mean inhibition: 63%,
results obtained with six TCC from three donors recognizing five
different Bet v 1 epitopes, i.e., aa 34–48, 76–90, 82–93
(Fig. 6B), 118–129, 142–156). This indicates that, in the
presence of saturating Ag concentrations, DC can load MHC class II in a
Cat S-dependent and, perhaps, also Cat S-independent fashion. Figure 6B also demonstrates that, at lower Ag concentrations (100
ng/ml), significant peptide-specific TCC responses occur selectively in
the presence of NP-specific IgE. This IgE-dependent Ag presentation was
almost entirely abolished (inhibition 80–90%) when the Cat S activity
of DC was eliminated (Fig. 6B, one of three experiments with
TCC recognizing Bet v 1 aa 82–93, 118–129, and 142–156).
In Western blotting experiments, we addressed whether Cat S plays a
major role in the proteolysis of rBet v 1-NP taken up via
FcRI and found that the anti-Bet v 1-reactive
material disappeared within a 4-h incubation period whether or not Cat
S activity was inhibited (data not shown). This finding supports the
notion that Cat S has no exclusive function during the earliest steps
of exogenous protein degradation. However, the possibility cannot be
ruled out that Cat S activity is of relevance for the liberation of
certain peptides from partially digested Ags. Based on our observations
that FcRI-mediated Ag presentation depends on Cat S activity and
that ongoing protein production was required for LHVS-induced Ii
accumulation, we reasoned that FcRI-dependent Ag delivery could
result in the peptide loading of newly synthesized MHC class II. To
explore this possibility, IgE-loaded and, for control purposes,
nonloaded DC were Ag pulsed in the presence or absence of CHX. These
experiments showed that FcRI-dependent Ag presentation indeed
critically depends on protein neosynthesis (Fig. 6C). Thus,
uptake via FcRI results in the sorting of Ags into DC compartments,
where nascent class II molecules acquire peptides in a Cat S-dependent
manner.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RI (30, 31), FcRII (3, 31, 32), ILT3 (33), and
FcRI (11, 16). The possibility that some of these receptors can
acquire a special function when expressed by DC is exemplified by our
finding that DC- rather than basophil-expressed FcRI can be
efficiently internalized. This may be due to the fact that DC express
FcRI complexes that differ in subunit composition from those
encountered on basophils (11). We further show that DC display affinity
for monomeric IgE, but are largely unable to bind monomeric IgA or IgG.
Thus, Ags/pathogens against which IgE responses have been mounted may
be preferentially retargeted to DC, whereas IgG-defined Ags may be
predominantly trapped by monocytic cells and other cell types that
display abundant high and low affinity IgG receptors. Prolonged
expression of FcRI-bound monomeric IgE should enable DC to acquire
and continuously display Ag-specific memory receptors and, thus, to
efficiently interact with Ags/pathogens both within the circulation
and, depending on their migratory properties, within the tissue.
We further demonstrate that the critical signal to internalize
antigenic moieties in an IgE-dependent manner is conferred by the Ag’s
ability to cross-link FcRI. Thus, the decision to take up and,
consequently, to present Ags in an IgE-dependent fashion is made on the
basis of whether or not Ags/pathogens carry repetitive and/or multiple
different IgE epitopes. We have also found that FcRI-mediated
internalization results in the proteolysis of the ligand and the
sorting of the resulting fragments into lysosome-related vesicles that
display the same antigenic profile (HLA-DR+,
HLA-DM+, LAMP+) as the MIICs previously
identified in various types of APC (18, 19, 20, 21). FcRI expressed by DC is
composed of an FcRI-chain linked to a pair of immunotyrosine
activation motif-bearing FcRI-chains (11). It, thus, has a
structural design similar to that of the BCR complex containing the
immunotyrosine activation motif-bearing molecules Ig and Igß. As
opposed to the BCR, which apparently requires Ig for being channeled
to organelles, where peptide loading of nascent MHC class II occurs
(34), the molecules directing FcRI to these compartments have yet to
be clarified. Conceivably, FcRI and Ig, but not Igß, will
associate with common but yet unknown adaptor molecules that allow
their trafficking to MIICs and/or prevent their entry into recycling
endosomes.
The cysteine protease Cat S critically contributes to Ii degradation to class II-associated invariant peptide (CLIP) and, thus, to the formation of peptide-bound MHC class II dimers in B cells (22). We have now demonstrated that human DC express Cat S and require this enzyme for optimal presentation of protein allergens taken up by fluid-phase and receptor-mediated endocytosis. Exposure of DC to the active-site Cat S inhibitor LHVS significantly decreased the surface export of newly synthesized HLA-DR and led to the intracellular accumulation of N-terminal Ii fragments with migratory properties similar to those of the LHVS-inducible Ii remnants previously detected in human B cells (22). As a predicted Cat S cleavage site in Ii is located in closest proximity to the N terminus of CLIP, the 13-kDa li intermediate in DC most likely contains the abluminal localization/retention signal as well as the peptide competitor CLIP (22, 35, 36, 37). This LHVS-mediated effect in DC occurs at inhibitor concentrations that do not interfere with the expression and function of other cathepsins (22–24; data not shown). Another observation was that LHVS not merely inhibited Cat S function, but further led to a disappearance of DC-expressed anti-Cat S immunoreactivity. We have evidence that LHVS does not interfere with the detection of Cat S by polyclonal Abs and, therefore, assume that enhanced lysosomal degradation of and/or decreased autocatalytic activation of procathepsin S by LHVS-bound Cat S is/are responsible for the observed phenomenon.
In preliminary experiments, we have comparatively analyzed the effects of the multiple cysteine/serine protease inhibitor leupeptin and of the Cat S-specific compound LHVS, and found both substances to display a similar inhibitory potency for Ag presentation by PB-DC (data not shown). Together with the observations of nonaltered Ii degradation/Ag presentation by bone marrow-derived APC of Cat D-/- (23) and Cat L-/- mice (38), this finding suggests that Cat S, rather than other leupeptin-sensitive proteases, appears to be the most potent catalyst of complete Ii degradation in DC. Thus, the observation that protein-pulsed DC require Cat S activity for the elicitation of robust peptide-specific T cell responses is apparently attributable to a unique role of Cat S for Ii degradation to CLIP. However, it should be taken into account that, at least in murine B cells, delayed Ii degradation by leupeptin leads to the mistargeting of Ii remnant-associated MHC class II complexes to high density lysosomes, from where compact dimers move only slowly to the cell surface (39). Our demonstration of decreased MHC class II surface transport in PB-DC together with the observation of delayed, rather than prevented, appearance of newly synthesized MHC class II on monocyte-derived DC in the presence of LHVS (manuscript in preparation), suggests that Cat S inhibition also leads to an altered trafficking of MHC class II molecules in DC. Although to be tested, it is possible that part of the inhibitory activity of LHVS on Ag presentation by DC may be related to the preferential loading of nascent MHC class II with peptides that are not derived from exogeneous proteins.
The requirement of Cat S is particularly pronounced under conditions
when DC depend on FcRI-mediated Ag uptake to mount a proliferative T
cell response but, less so, when high Ag concentrations are applied. In
this context, it is conceivable that DC express proteases other than
Cat S, which allow for sufficient MHC class II loading only if peptides
are generated in high copy numbers or, alternatively, that Cat
S-independent Ag presentation via mature, recycling MHC class II ß
dimers operates particularly efficiently at high Ag concentrations
(40, 41, 42). Another candidate Cat S-independent presentation mechanism
relates to surface-targeted class II ßIi. In DC, HLA-DR ßIi
can directly reach the plasma membrane, internalize into early
endosomes, and rapidly recycle to the cell surface as compact dimer
(25). While the specific protease requirement of this early endosomal
pathway remains to be established, the observation of a stringent Cat S
dependency for the presentation of FcRI-targeted Ags strongly
suggests that Cat S plays a major role for the maturation and peptide
loading of, at least, those MHC class II molecules that access
MIIC-like organelles of DC.
Peripheral blood DC, based on the differential anti-CD11c
reactivity, contain two major subpopulations, both of which express and
internalize FcRI-bound IgE complexes (Refs. 11 and 43; unpublished
observations). Counterparts of the CD11c- and
CD11c+ DC have been identified in T and B cell-rich areas
of lymphoid tissues, respectively (6, 7). It is feasible that PB-DC as
well as epidermal Langerhans cells at peripheral sites bind and
internalize Ag/pathogen in an FcRI/IgE-dependent manner and, upon
homing to lymphoid tissues, fulfill their T cell-priming capacity. The
harsh proteolytic conditions operative in the FcRI-dependent
internalization pathway may allow the efficient liberation not only of
previously recognized, but also of cryptic peptides of IgE-defined Ags,
thus widening the repertoire of allergen/pathogen-specific T cells.
This could be of particular importance for atopy since multiple Ags of
the house dust mite, a major source of allergen in atopic dermatitis
and asthma, have revealed to be proteases and, thus, are presumably
resistant to rapid endosomal degradation (44). Similarly, a common
biochemical feature of food allergens is their remarkable resistance to
protease destruction (45). In a scenario of limited IL-12 production by
APC, as has been demonstrated in atopy (46), FcRI-dependent allergen
degradation and presentation may amplify and widen allergen-specific
Th2 cell responses (47) and even induce IgE-specific Th responses.
These T cells may be able to provide help for any B cell that
internalizes and presents IgE in a CD23-dependent manner as well as for
IgE-specific B cells, resulting in the occurrence of
anti-IgE-directed immune responses frequently observed in atopy
(48).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Dieter Maurer, Division of Immunology, Allergy and Infectious Diseases, Department of Dermatology, University of Vienna Medical School, Waehringer Guertel 18-20, A-1090 Vienna, Austria. E-mail address:
3 Abbreviations used in this paper: DC, dendritic cell; aa, amino acid; BCR, B cell receptor; biot, biotin; Cat, cathepsin; CHX, cycloheximide; cIgE, chimeric human Fc/mouse Fab immunoglobulin E anti-4-hydroxy-3-nitrophenacetyl; CLIP, class II-associated invariant peptide; Ii, invariant chain; LAMP, lysosome-associated membrane protein; LHVS, N-morpholinurea-leucine-homophenylalanine-vinylsulfone-phenyl; Ly-, lymphocyte-depleted; MFC, mean fluorescence channel; MIIC, major histocompatibility complex class II compartment; NP, 4-hydroxy-3-nitrophenacetyl; PB-DC, dendritic cells circulating in peripheral blood; PE, phycoerythrin; PerCP, peridinine chlorophyll protein; SA, streptavidin; TCC, T cell clone; TRITC, tetramethylrhodamine isothiocyanate.
Received for publication January 28, 1998. Accepted for publication May 13, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. J. Exp. Med. 179:1109.RI. J. Exp. Med. 175:1353.RI). J. Exp. Med. 175:1285.RI-bearing cells within the human dermis. J. Invest. Dermatol. 102:315.[Medline]
RI as a complex composed of FcRI- and FcRI-chains and can use this receptor for IgE-mediated allergen presentation. J. Immunol. 157:607.[Abstract]
RI) on monocytes of atopic individuals. J. Exp. Med. 179:745.RI) mediates IgE-dependent allergen presentation. J. Immunol. 154:6285.[Abstract]
/ß sheath and occurs independently of its cytoplasmic tyrosines. Cell 74:939.[Medline]
RI (CD64) expressed on human blood dendritic cells. J. Immunol. 158:3090.[Abstract]
receptor (monocyte/macrophage)-specific immunogen. J. Immunol. 149:3477.[Abstract]
and Igß subunits in MHC class II-restricted antigen presentation. Immunity 3:335.[Medline]
release in patients with allergic asthma. J. Immunol. 158:5560.[Abstract]
RI autoantibodies in clinical allergy. Curr. Opin. Immunol. 8:784.[Medline]
This article has been cited by other articles:
|
|
 |
|
  C. von Garnier, M. E. Wikstrom, G. Zosky, D. J. Turner, P. D. Sly, M. Smith, J. A. Thomas, S. R. Judd, D. H. Strickland, P. G. Holt, et al. Allergic Airways Disease Develops after an Increase in Allergen Capture and Processing in the Airway Mucosa J. Immunol., November 1, 2007; 179(9): 5748 - 5759. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Amatschek, E. Kriehuber, W. Bauer, B. Reininger, P. Meraner, A. Wolpl, N. Schweifer, C. Haslinger, G. Stingl, and D. Maurer Blood and lymphatic endothelial cell-specific differentiation programs are stringently controlled by the tissue environment Blood, June 1, 2007; 109(11): 4777 - 4785. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. C. Strunk and G. R. Bloomberg Omalizumab for asthma. N. Engl. J. Med., June 22, 2006; 354(25): 2689 - 2695. [Full Text] [PDF]
|
|
|
|
 |
|
 O. Fainaru, D. Shseyov, S. Hantisteanu, and Y. Groner Accelerated chemokine receptor 7-mediated dendritic cell migration in Runx3 knockout mice and the spontaneous development of asthma-like disease PNAS, July 26, 2005; 102(30): 10598 - 10603. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bayry, S. Lacroix-Desmazes, M. D. Kazatchkine, O. Hermine, D. F. Tough, and S. V. Kaveri Modulation of Dendritic Cell Maturation and Function by B Lymphocytes J. Immunol., July 1, 2005; 175(1): 15 - 20. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. O'Neill, M. J. Shlomchik, T. T. Glant, Y. Cao, P. D. Doodes, and A. Finnegan Antigen-Specific B Cells Are Required as APCs and Autoantibody-Producing Cells for Induction of Severe Autoimmune Arthritis J. Immunol., March 15, 2005; 174(6): 3781 - 3788. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. On, J. M. Billingsley, M.-H. Jouvin, and J.-P. Kinet Molecular Dissection of the FcR{beta} Signaling Amplifier J. Biol. Chem., October 29, 2004; 279(44): 45782 - 45790. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ebihara, S. Endo, K. Ito, Y. Ito, K. Akiyama, M. Obinata, and T. Takai Immortalized Dendritic Cell Line with Efficient Cross-Priming Ability Established from Transgenic Mice Harboring the Temperature-Sensitive SV40 Large T-Antigen Gene J. Biochem., September 1, 2004; 136(3): 321 - 328. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Jaksits, W. Bauer, E. Kriehuber, M. Zeyda, T. M. Stulnig, G. Stingl, E. Fiebiger, and D. Maurer Lipid Raft-Associated GTPase Signaling Controls Morphology and CD8+ T Cell Stimulatory Capacity of Human Dendritic Cells J. Immunol., August 1, 2004; 173(3): 1628 - 1639. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Ellinger-Ziegelbauer, B. Stuart, B. Wahle, W. Bomann, and H.-J. Ahr Characteristic Expression Profiles Induced by Genotoxic Carcinogens in Rat Liver Toxicol. Sci., January 1, 2004; 77(1): 19 - 34. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. C. Heystek, C. Moulon, A. M. Woltman, P. Garonne, and C. van Kooten Human Immature Dendritic Cells Efficiently Bind and Take up Secretory IgA Without the Induction of Maturation J. Immunol., January 1, 2002; 168(1): 102 - 107. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Reali, J. W. Greiner, A. Corti, H. J. Gould, F. Bottazzoli, G. Paganelli, J. Schlom, and A. G. Siccardi IgEs Targeted on Tumor Cells: Therapeutic Activity and Potential in the Design of Tumor Vaccines Cancer Res., July 1, 2001; 61(14): 5517 - 5522. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. B. Oriss, P. Q. Hu, and T. M. Wright Distinct Autoreactive T Cell Responses to Native and Fragmented DNA Topoisomerase I: Influence of APC Type and IL-2 J. Immunol., May 1, 2001; 166(9): 5456 - 5463. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Santambrogio, A. K. Sato, G. J. Carven, S. L. Belyanskaya, J. L. Strominger, and L. J. Stern Extracellular antigen processing and presentation by immature dendritic cells PNAS, December 21, 1999; 96(26): 15056 - 15061. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
