| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Molecular Genetics and Biochemistry and
Cell Biology and Physiology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15261
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cancer patients develop immune responses to many tumor Ags, but these responses are typically weak and not therapeutic. CTL, if found, are present at low frequency, and Ab responses are characterized by low titers. The poor quality of T cell responses to tumor Ags in patients can be attributed to several factors that can vary with the type of the tumor or the nature of the tumor Ag. Tumors themselves are not good APCs (7, 8), and in some instances have been shown to induce T cell anergy (9, 10, 11). Similarly, many identified tumor Ags are differentially expressed self proteins to which high affinity T cells may have been eliminated (12, 13). We report here another mechanism by which cancer patients’ T cells are kept unresponsive to an abundant circulating tumor Ag MUC1 by the inability of APCs to efficiently process this potentially immunogenic tumor protein.
Even though it has been identified as a tumor Ag, MUC1 is also found on the apical surfaces of normal ductal epithelial cells (14). It consists of numerous tandemly repeated 20-aa segments (up to 200/molecule), each of which contains five sites for O-linked glycosylation. When expressed on normal cells, MUC1 is extensively glycosylated at all sites with long, branched, O-linked oligosaccharide chains. Furthermore, the extracellular membrane-proximal domain (preceding the tandem repeat region) is glycosylated at five sites with N-linked carbohydrates. This heavily glycosylated protein is cleaved off the epithelial cell surface and becomes a component of ductal secretions. MUC1 expression on tumor cells is much higher than that on normal cells and it is not polarized. The glycosylation pattern also changes to linear rather than branched oligosaccharides and to much shorter chains added to fewer glycosylation sites (15). These tumor-specific forms of MUC1 are also cleaved off the tumor cells, drain to regional lymph nodes and enter the peripheral circulation. They can be purified from sera and ascites fluid of patients with late stage tumors.
Immune responses against MUC1 found in cancer patients are characterized by a low frequency of CTL and a low titer of Ab of the IgM isotype (16, 17, 18, 19, 20, 21). The CTL responses consist of a low frequency of class I MHC-restricted CTL and a low frequency of MHC-unrestricted CTL (16, 17, 18, 19, 20, 21, 22, 23, 24). No evidence has been reported of in vivo elicited MUC1-specific Th activity. We recently reported that MHC class II-restricted, CD4+ Th cells can be primed in vitro to recognize MUC1, but only if synthetic, completely unglycosylated, MUC1 protein core peptide is used as Ag to load dendritic cells (DC)5 (25). Tumor-derived MUC1 glycosylated with short linear carbohydrates, isolated from patients’ sera or ascites, failed to prime Th cells. Moreover, in vitro priming of class I-restricted CTL by DC loaded with tumor-derived MUC1 was also much less efficient than priming with DC loaded with the synthetic MUC1 peptide (26). Others have also shown that in mice immunization with the human MUC1 (which shares little homology with mouse MUC1) elicits immune responses only when unglycosylated protein is used (27, 28, 29).
In this paper we examine the uptake and processing of two forms of MUC1, the circulating form of MUC1 with tumor-specific glycosylation that is available to patient APC in vivo, and the unglycosylated synthetic polypeptide (140 aa long) composed of seven MUC1 tandem repeats that can be provided through vaccination. By directly fluorescently labeling each protein, we examined by flow cytometry the uptake of these proteins by DC and followed them intracellularly by confocal microscopy. We show that each protein was readily internalized by DC, but that the transport of tumor MUC1 and that of synthetic MUC1 were completely divergent. The glycosylated tumor MUC remained stranded in the early endosomes, while the synthetic peptide was transported to late endosomes for processing. We also explored the mechanism involved in the internalization of both forms of MUC1 by DC and what effect (if any) uptake of these proteins had on their ability to process and present other Ags. Furthermore, to address the possibility that inhibited processing of glycosylated proteins might be a common phenomenon, we examined uptake by DC and intracellular transport of another glycoprotein tumor Ag, HER-2/neu. Like MUC1, HER-2/neu fails to elicit Th cells in vivo, while vaccination with short synthetic peptides derived from this molecule elicits strong helper responses (30). We show that its transport is also blocked at early endosomes.
The strategy we determined to be useful for circumventing inefficient processing of MUC1 was to provide APCs with an unglycosylated form of the Ag, which we show can be efficiently taken up and processed by DC. These latest studies explain our previously reported observations that tumor MUC1 is not able to elicit class I- and class II-restricted T cell responses, but that these responses are efficiently elicited with a synthetic MUC1 peptide (25, 26).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Tumor-derived MUC1 (glycosylated form) was purified from ascites
fluid pooled from several patients with metastatic breast or pancreatic
tumors by density gradient centrifugation (31), followed
by immunoaffinity chromatography using MUC1-specific Ab 3C6 (gift from
Dr. Jo Hilgers, Free University, Amsterdam, The Netherlands). This form
of MUC1 (
200 kDa) is cleaved off the tumor cell surface and does not
contain the transmembrane and the cytoplasmic domain. The recombinant,
baculovirus-derived form of MUC1 contains 22 tandem repeats as well as
the transmembrane and cytoplasmic domains (31). The
synthetic 140-mer peptide representing seven repeats of the MUC1
protein sequence GVTSAPDTRPAPGSTAPPAH was synthesized at the Peptide
Synthesis Facility of the Department of Molecular Genetics and
Biochemistry, University of Pittsburgh School of Medicine (Pittsburgh,
PA). All forms of MUC1 were directly conjugated to either Cy3 or Cy5
fluorescent dye according to the manufacturer’s instructions (Amersham
Pharmacia Biotech, Piscataway, NJ). FITC-conjugated mannosylated BSA
was purchased from Sigma (St. Louis, MO). Human HER-2/neu extracellular
domain (ECD) was a gift from Dr. Tom Vedvick (Corixa, Seattle, WA). It
was produced as a secreted protein in a mammalian L cell expression
system by cloning HER-2/neu cDNA into the plasmid vector pMTK-bGH-C.
The secreted protein was extensively purified through several standard
chromatographic methods. Following purification, the protein
preparation showed only one peak on reverse phase HPLC and a single
band of 110 kDa on Coomassie-stained gel. CMV lysate and mock cell
lysates were gifts from Dr. Penelope Morel (University of Pittsburgh
School of Medicine). The CMV lysate was generated by sonication of MRC5
human fetal lung fibroblast cells infected with the AD169 strain of CMV
following 1 wk of infection at a multiplicity of infection of 10:1.
Mock lysate was generated under the same conditions from uninfected
fibroblasts.
Antibodies
Purified, mouse anti-human transferrin receptor (CD71) was purchased from Immunotech, (Westbrooke, ME). PE-conjugated mouse anti-human LAMP-1, HLA-DM, and mouse anti-human mannose receptor (blocking Ab clone 19 and PE-labeled marker Ab) were purchased from PharMingen (San Diego, CA). PE-conjugated mouse anti-human HLA-DR, CD14, and CD86 were purchased from Becton Dickinson (San Jose, CA). Alexa-546-conjugated goat anti-mouse IgG was purchased from Molecular Probes (Eugene, OR).
Generation of DC
Monocyte-derived DC were generated in vitro according to a previously described protocol (32). Following removal of RBC from a leukopheresis product from a healthy donor, PBMC were plated in T-75 flasks for adherence for 2 h. After 2 h, nonadherent cells were removed by gentle washing with warm medium. The remaining loosely adherent cells were cultured for 8 days in serum-free AIM V medium (Life Technologies, Gaithersburg, MD) containing GM-CSF (10 ng/ml; Schering-Plough, Kenilworth, NJ) and IL-4 (25 ng/ml; Schering-Plough). The DC cultures were fed with additional medium and cytokines on days 4 and 6 of culture. DC were then harvested and used in the uptake and trafficking experiments without further purification. The purity of DC in the cultures at this time point was >60% as measured by flow cytometry for DC markers.
Ag uptake/inhibition of uptake assays
DC were harvested, washed, and resuspended in serum-free AIMV medium at a concentration of 2.5 x 106/ml. DC were then added to sterile 4-ml polypropylene pop-capped tubes (Falcon, Becton Dickinson, Franklin Lakes, NJ) at 100 µl/tube (2.5 x 105/tube). Fluorescently labeled proteins were added to the tubes at a final concentration of 10 µg/ml. The DC were allowed to internalize the protein at 37°C for various times. The cells were then washed three times with cold medium and fixed with 1% paraformaldehyde. Uptake of the proteins was quantitated by flow cytometric analysis using a Becton Dickinson FACScalibur and CellQuest software. To exclude contaminating non-DC, analysis gates were set on cells with the highest forward scatter. These cells typically represented >60% of the total cell population.
Cytochalasin D (used at 10 µM), wortmannin (used at 100 nM), LY294002 (used at 50 µM), and mannan (used at 200 µg/ml) were purchased from Sigma. Sucrose (used at 500 µM) was purchased from Life Technologies/BRL (Grand Island, NY). Inhibition assays were performed by pretreating the DC for 30 min with an inhibitor followed by the addition of fluorescently labeled Ag. Unless otherwise indicated, the DC were allowed to take up the labeled proteins for 1 h at 37°C in the presence of inhibitor. Following this uptake period, the DC were washed and fixed with 1% paraformaldehyde, and the amount of internalized fluorescence was quantitated by flow cytometry. The percent inhibition was calculated by first subtracting the mean fluorescence intensity (MFI) of DC incubated with labeled protein on ice from both inhibitor-treated and untreated samples. These normalized numbers were then used in this equation: % uptake = [((MFI of untreated DC + protein) - (MFI of inhibitor-treated DC + protein))/(MFI of untreated DC + protein)] x 100%; % inhibition = 100 - % uptake.
Competition assays to determine cell surface binding of Ags were performed in a similar manner as those described above with the following modifications: binding was conducted at 4°C to eliminate internalization, and competitor used in these assays was a 10-fold excess of the same protein, unlabeled, which was added to the DC for 15 min before the addition of labeled protein. Competitor was maintained throughout the 30-min binding period. The percent inhibition of binding was calculated as described above.
Pulse-chase analyses for retention of MUC1 were performed by allowing the DC to take up Ag for 1 h at 37°C, washing extensively, then maintaining the DC at 37°C for the indicated periods of chase to allow processing of Ag. DC, which were fixed immediately following the pulse period, were used as a positive control and were counted as 100%. The MFIs of chased samples were then divided by the MFI of positive control DC to calculate the percentage of retained fluorescence.
Electron microscopy
Glycosylated tumor MUC1 and synthetic 140-mer peptide were directly conjugated to 20-nm colloidal gold particles according to a published protocol (33). The gold-labeled Ags were then fed to 5 x 106 immature day 8 DC at a 20 µg/ml final concentration in 1 ml of AIM V medium for 15 min or 2 h as indicated. The cells were then washed to remove excess Ag, pelleted, and fixed in 2.5% glutaraldehyde in PBS. The 1-mm3 pellets were postfixed in 1% OsO4 and 1% K3Fe(CN)6 for 1 h. After three washes the pellet was dehydrated through a graded series of 30–100% ethanol and 100% propylene polypropylene oxide and infiltrated in a 1/1 mixture of propylene oxide/Polybed 812 epoxy resin (Polysciences, Warrington, PA) for 1 h. After several changes of 100% resin over 24 h, cells were embedded in molds and cured at 37°C overnight, followed by additional hardening at 65°C for 2 more days. Ultrathin (60-nm) sections were collected 200-mesh copper grids and stained with 2% uranyl acetate in 50% methanol for 10 min, followed by 1% lead citrate for 7 min. Sections were viewed with a CX100 transmission electron microscope (Jeol, Peabody, MA) at 60 or 80 kV.
Confocal laser scanning microscopy
DC were harvested after 8 days of culture and plated on Lab-Tek chamber slides (Nalge Nunc, Naperville, IL) at a density of 5 x 103–104 DC/chamber and allowed to adhere for 2 h. DC were then washed with warm medium, and adherent DC were exogenously pulsed with 10 µg/ml fluorescently labeled protein for the indicated time periods. Following uptake and processing, DC were fixed with 2% formaldehyde, permeabilized with 0.1% saponin, and counterstained with PE-labeled Abs to visualize intracellular compartments. To amplify the signal of the primary Abs and compensate for the bleaching of PE by the confocal laser, an Alexa-546-labeled goat anti-mouse IgG was used as a secondary Ab for staining. Following staining, DC were fixed a second time with 1% paraformaldehyde, the chambers were removed, and the slides were mounted using Vectashield mounting medium (Vector, Burlingame, CA) and analyzed by confocal laser scanning microscopy. For these analyses, we used a Leica TCS NT confocal LSM microscope (Rockleigh, NJ); images were collected using the x40 objective. Micrographs shown are projections of serial sections scanning the majority of the cell. Each image is representative of at least 50 fields scanned, with an average of 10 cells/field; 0.5-µm-thick optical sections (typically eight per cell) were simultaneously scanned for excitation of Alexa-546 and Cy5. Cy5 is a far red fluorescing dye to which any color can be assigned by computer.
CMV-specific T cell proliferation
DC were generated from a CMV-seropositive normal donor for 8 days in culture with GM-CSF and IL-4, as described above. The cells were harvested and plated in 96-well plates at a density of 104 DC/well. They were pulsed with either glycosylated tumor MUC1 (100 µg/ml), keyhole limpet hemocyanin (KLH; 100 µg/ml) or no Ag for 1 h. A lysate of CMV-infected cells or mock lysate (described above) was then added to the DC in increasing doses in 96-well plates. Autochthonous T cells were added to the plates at a concentration of 10 T cells/DC. These cultures were maintained for 4 days, and proliferation was measured in a standard tritiated thymidine incorporation assay.
Detection of recycled glycosylated tumor MUC1
Immature day 8 GM-CSF/IL-4 human DC were pulsed at 4 or 37°C with the fluorescent Ag (as described above) for 30 min. The uptake was stopped with extensive washes in cold PBS, and the cells were subsequently treated with 0.15 M glycine for 30 min at 4°C. As controls, untreated cells were similarly handled. After the acid treatment, the cells were extensively washed and returned to normal medium conditions (200-µl final volume) and chased for 90 min under the temperature conditions of the pulse step (4 or 37°C, respectively). At the end of the chase period the cells were centrifuged, and 100 µl of the supernatant was used for protein detection by ELISA. The pelleted cells were analyzed for retention of fluorescence by FACS.
MUC1 detection in the supernatant was performed using an Ab-sandwich ELISA. Immulon 4 (Dynex Technologies, Chantilly, VA) ELISA plates were coated overnight at 4°C with a mixture of two anti-MUC1 mAbs, Ma552 (34) and HMPV (35), at a final concentration of 10 µg/ml in a volume of 50 µl/well. After extensive washes and blocking in Tween-20 (0.05%) blocking buffer, 30 µl of the saved supernatant was added to the coated wells in triplicate and incubated at room temperature for 2 h. The same mixture of Abs was used to detect the bound Ag. The reaction was developed by adding a secondary anti-mouse IgG, peroxidase conjugated (Sigma), in the presence of o-phenylenediamine substrate solution (1 mg o-phenylenediamine/ml of 0.05 M citrate buffer, pH 4.0, and 0.05% H2O2; Sigma). The reaction was stopped at 20 min with 2.5 M H2SO4. Absorbencies were read at 490 nm using the automatic MRX Revelation plate reader (Dynex Technologies) and Revelation 4.21 software to calculate means and SDs. Data were plotted using Cricket Graph software (Computer Associates International, Islandia, NY).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We considered that one potential explanation for the inability of DC to prime MUC1-specific T cells when loaded with the glycosylated tumor form of MUC1 (25, 26) could be that DC were simply unable to take up this protein. To analyze and quantitate the uptake of this form of MUC1 by DC and to compare it to the uptake of unglycosylated synthetic peptide, we directly labeled the two forms with fluorescent dyes, fed them to 8-day-old DC in vitro, and monitored by flow cytometry the increase in DC fluorescence intensity over time. The day 8 DC expressed high levels of class II MHC, mannose receptor, and costimulatory molecules and had lost CD14 expression (data not shown). These cells were not fully mature DC, however, as they were still able to take up large amounts of Ag through macropinocytosis and receptor-mediated endocytosis and contained numerous intracellular MHC class II compartments.
Depicted in Fig. 1
is the uptake of two
forms of MUC1, 140-mer synthetic peptide and glycosylated tumor MUC1.
The histograms (Fig. 1, A and B) show uptake of
each fluorescently (Cy5)-labeled protein at 37°C compared with the
amount of surface-bound protein at 4°C (on ice). Both forms of MUC1
were efficiently taken up by DC after 30 min and in even greater amount
after 2 h. This increase in fluorescence intensity was due to
internalization of both forms, as can be seen in the electron
micrographs. Glycosylated tumor MUC1 can be seen at the cell surface in
a clathrin-coated pit (Fig. 1C) as well as inside the cell
in a small endocytic vesicle (Fig. 1D). The MUC1 peptide is
found predominantly in large, fluid-filled, pinocytic vesicles (Fig. 1E). Both forms were directly conjugated to 20-nm gold
beads.
|
Peptide epitopes presented by class II MHC are typically derived from proteins acquired by APCs from exogenous sources (36, 37, 38, 39). Following internalization, proteins are transported through a series of early and late endosomes or lysosomes in which acidification and degradative activity progressively intensify (37). These vesicles can be identified microscopically using fluorescently labeled Abs against specific resident proteins, such as transferrin receptor, a marker of early endosomes, and LAMP-1, a lysosomal marker. Endocytic vesicles with degraded protein fragments fuse with class II MHC-rich MIIC vesicles (38), where binding of peptides to class II molecules is facilitated by HLA-DM (39, 40). Both HLA-DM and HLA-DR are markers of the MIIC compartments.
We considered that the lack of presentation of glycosylated tumor MUC1
by class II MHC might be the result of this form not being properly
transported to late endosomal and MIIC Ag-processing compartments. DC
cultured on chamber slides were simultaneously fed fluorescently
labeled tumor MUC1 and synthetic MUC1. After 2 h to allow for both
uptake and processing to take place, the DC were fixed, permeabilized,
and stained with an Ab against HLA-DR. The slides were then examined by
confocal laser scanning microscopy. The results of this experiment are
depicted in the first four panels of Fig. 2, one for each separate fluorescence
channel (A, B, and C) and the last one
(D), an overlay of all three, demonstrating areas of
colocalization. FITC-anti-HLA-DR Ab staining (in green,
A) shows intense perinuclear, vesicular staining of DR-rich
late endosomal compartments (MIIC). Cy3-labeled 140-mer peptide (in
red, B) also gives intense fluorescence in perinuclear
vesicles, indicative of late endosomes. In contrast, Cy5-labeled
glycosylated MUC1 (in blue, C) is located in large,
membrane-proximal vesicles quite distant from the perinuclear region.
The three-color overlay (D) shows almost complete
colocalization (in yellow) of the peptide and HLA-DR and no
colocalization of glycosylated MUC1 (in blue) with HLA-DR. These
results clearly show that following uptake, glycosylated tumor MUC1 is
not transported to the MIIC compartments where processing and binding
to class II MHC would occur. In contrast, the same DC transports the
unglycosylated MUC1 synthetic peptide to these compartments very
efficiently. This difference is in the strong correlation with
our published functional studies, which demonstrated that only the
synthetic peptide form of MUC1 was presented to T cells, while the
glycosylated tumor form was not (25, 26).
|
E, this form of
MUC1 colocalizes with HLA-DR in the same way as the unglycosylated
peptide. Within 2 h after uptake, all the recombinant MUC1 was
found in a compartment rich in HLA-DR. These data suggest that
glycosylation rather than protein size is the determining factor in the
retention of tumor MUC1 in early endosomes.
In a more comprehensive microscopic analysis, DC were exogenously
pulsed with Cy5-labeled glycosylated tumor MUC1 (indicated in green),
fixed, permeabilized, and costained (red) with Abs against markers of
either early or late endosomal compartments. As expected, glycosylated
tumor MUC1 colocalized with the transferrin receptor (Fig. 2F), a marker of early endosomes (37), and the
mannose receptor, present predominantly in early endosomes and less
frequently in later endosomes and MIIC (41).
Colocalization with the mannose receptor was seen only in peripheral
vesicles and was not present in the more perinuclear, mannose
receptor-containing, later endosomes (Fig. 2G). There was
little or no colocalization with late endosomal markers HLA-DM (Fig. 2H) and LAMP1 (Fig. 2I).
Long-term retention of glycosylated MUC1 by DC
To address the fate of glycosylated MUC1 and the 140-mer peptide
inside DC, we followed these proteins by confocal microscopy over a
period of 24 h and quantitated by flow cytometry the amount of
each form that was retained. For confocal analyses, DC were exogenously
pulsed with either form of Cy5-labeled MUC1 (assigned a green color)
and then chased for indicated times, when they were fixed,
permeabilized, and stained with a PE-labeled anti-HLA-DR Ab (red).
As in the previous micrographs, colocalization of green and red appears
in yellow. The 140-mer peptide (Fig. 3, upper panels) is trafficked to HLA-DR-rich compartments
within 15 min after uptake, remains exclusively associated with
perinuclear DR-rich endosomes at 2 h, and is difficult to detect
at 24 h, probably due to almost complete degradation. In contrast,
glycosylated tumor MUC1 (lower panels of Fig. 3) is
retained in large vesicles adjacent to the plasma membrane, apparently
undegraded even after 24 h following uptake. The appearance of
some yellow color in this figure is a consequence of the way the images
are displayed. The green MUC1 and red HLA-DR actually reside in
different vesicles in separate planes, but some of these vesicles
become superimposed when the images are shown as stacked projections.
The images are displayed this way to enable visualization of the entire
DC and to demonstrate the peripheral localization of glycosylated MUC1.
We have also performed FACS analysis to evaluate loss of fluorescence
over time and have found fluorescence retained only in DC loaded with
the glycosylated MUC1. A high level of fluorescence was maintained even
after 48 h and was not diminished after activation of DC through
ligation of cell surface CD40 by soluble CD40 ligand trimer (data not
shown).
|
Carbohydrate moieties have been shown to inhibit presentation of
either Ags to which they are bound or of other Ags taken up by the same
APC (42). Because glycosylated tumor MUC1 is taken up and
retained by APCs long term, we questioned whether this may inhibit the
ability of these cells to process and present other Ags to class
II-restricted Th cells. To address this question, we pretreated DC
generated from a CMV-seropositive donor with 100 µg/ml of
glycosylated MUC1, control Ag KLH, or no Ag, before addition of CMV Ag,
a whole cell lysate from CMV-infected fibroblasts. Autochthonous T
cells were then added to the cultures, and the proliferative responses
to CMV Ags were evaluated in a 4-day proliferation
assay. T cells responded in a dose-dependent manner to the CMV lysate
and had similar responses to this Ag in the presence of either excess
MUC1 glycoprotein (10 times the amount used in the confocal studies) or
KLH (Fig. 4).
|
DC internalize Ag through receptor-mediated endocytosis, phagocytosis, and macropinocytosis (43, 44). The most efficient mechanisms for delivery of Ags to MIIC compartments for presentation via class II MHC are receptor-mediated endocytosis and macropinocytosis (45, 46, 47). Thus, we compared the mechanisms responsible for uptake of the two forms of MUC1 by determining the binding specificity (if any) of either form to DC and using inhibitors of either the receptor-mediated endocytosis or macropinocytosis.
As a confirmation of the active endocytic uptake, preincubation of DC
with cytochalasin D (48) for 30 min prior to addition of
the Ag led to a dramatic (90%) inhibitory effect on uptake when the
glycosylated tumor MUC1 was added for an additional 30 min, suggesting
that active actin-based cytoskeletal rearrangements are required for
the internalization of this molecular form (Fig. 5A). The uptake inhibition of
the synthetic peptide was only about 70% (Fig. 5B),
suggesting that although the majority of increased fluorescence is due
to active uptake, this smaller molecular form may still be internalized
through small pinocytic vesicles.
|
) while strongly inhibiting uptake of
the 140-mer peptide. Sucrose (0.5 M), an inhibitor of receptor-mediated
endocytosis that interferes with clathrin-coated pit formation
(51), caused significant inhibition of the uptake of
glycosylated form and had no effect on the synthetic peptide (data not
shown). In addition, uptake of glycosylated tumor MUC1 is inhibitable
by cold, unlabeled protein and shows a saturable uptake curve, whereas
the synthetic peptide is uninhibitable by its own unlabeled control
(data not shown). Taken together, the results of the above experiments
correlated with the electron micrographs findings suggested that the
uptake of glycosylated MUC1 by DC is regulated by a receptor-mediated
mechanism and that internalization of the 140-mer peptide is primarily
mediated through macropinocytosis.
|
Having determined that glycosylated MUC1 was taken up by DC
through a receptor-mediated mechanism, we addressed the question of
specifically which receptor was involved. One particularly efficient
receptor expressed at high levels on DC (expression is highest on
immature DC and drops as they mature) is the mannose receptor (MR).
This receptor has high specific affinity for both mannose and fucose
residues. Furthermore, the MR has been characterized as a pattern
recognition receptor mediating the uptake of many microbial products,
the best defined of these being lipoarabinomannan, from mycobacteria
(41, 52). The mannose and fucose residues for which
mannose receptor is specific are present on glycosylated tumor
MUC1,6 raising the possibility that MR may
mediate the uptake of this molecule. To address the binding of MUC1 to
MR, we used mannan (polymannose), which competes for binding to MR, and
a blocking mAb specific for MR. DC were preincubated with either
inhibitor for 10 min before the addition of labeled MUC1 (glycosylated
tumor protein or peptide). Mannosylated BSA was included as a positive
control. Following a 1-h uptake period at 37°C, DC were washed,
fixed, and analyzed by flow cytometry. Uptake of glycosylated MUC1 was
significantly inhibited both by excess mannan and by the blocking Ab
against the MR (Table II). Both these
reagents also strongly blocked uptake of mannosylated BSA, but had
little or no inhibitory effect on uptake of the 140-mer peptide. These
findings strongly suggest that uptake of glycosylated MUC1 by DC is in
large part mediated by the MR. We cannot rule out the contribution of
macropinocytosis or another receptor, however, which may account for
the inability of MR blockers to completely inhibit MUC1 uptake.
|
Due to numerous glycosylated tandem repeats on each molecule, MUC1
uptake via MR probably results in massive cross-linking of the
receptors. The high avidity of this interaction may not allow the
uncoupling of the MR from the MUC1 ligand that is supposed to occur in
endosomal sorting compartments or CURL (compartments for uncoupling of
receptors and ligands) (53, 54, 55). CURL are specialized
endosomes where receptors disengage from ligands and are selectively
removed and concentrated in tubular extensions that are then recycled
to the plasma membrane. CURL vesicles containing concentrated free
ligand are transported to lysosomes where ligand is degraded. The
extended colocalization of glycosylated MUC1 with MR in early endosomes
suggests that the receptors are unable to dissociate from the ligand
and leave the endosomes; therefore, further progression down the
endocytic pathway is halted. To address this possibility, we allowed
the DC to internalize MUC1 protein for 1 h, extensively washed the
cells, then added excess mannose to compete with MUC1 for the MR in the
early endosomes. We postulated that this may facilitate MUC1
dissociation from the MR and release the block on its further
transport. In the absence of free mannose, 44% of glycosylated MUC1
was retained after a 24-h chase period. With the addition of mannose to
the chase, however, the amount of retained MUC1 was substantially
reduced at 24 h, and this reduction was dependent on the dose of
mannose (Fig. 6). The greatest decrease
in retained fluorescence (to 18%, a level similar to that of the
control protein, mannosylated BSA) resulted from the addition of 50
µM mannose. This dose of mannose was not toxic to the DC, and there
was no effect on the processing of the 140-mer peptide and only a
slight reduction in the retention of mannosylated BSA. We interpret
these results to mean that mannose, which diffuses into the cell and
early endosomes due to its small size, competes with MUC1 for binding
to MR. The loss of fluorescence may indicate that after being
disengaged from the MR, MUC1 goes on to be properly sorted and
processed.
|
Difficulty in intracellular trafficking and inefficient processing as causes of a diminished immune response to tumor Ags are not likely to be unique to MUC1. HER-2/neu is an oncogenic glycoprotein overexpressed by adenocarcinomas of the breast, ovary, stomach, uterus, and lung. Patients with HER-2/neu-expressing tumors demonstrate weak immune responses to HER-2/neu, including low titer Abs, limited T cell proliferative activity, and some CTL activity (56). Interestingly, strong Th cell responses and Abs were elicited to this protein in a vaccination model in which rats were immunized with short synthetic peptides derived from the rat HER-2/neu sequence in Freund’s adjuvant. In the same model, immunization with the full-length protein in adjuvant failed to elicit any immune response (30). HER-2/neu protein contains numerous carbohydrates on its ECD. The ECD is shed from tumor cells and can be found in the serum of adenocarcinoma patients. We postulated that the glycosylated, soluble form of HER2/neu, much like MUC1, may not be properly trafficked by DC, resulting in the lack of processing for Th cell generation.
Our confocal microscopic analyses support this hypothesis by
demonstrating that, like glycosylated tumor MUC1, HER-2/neu is retained
in early endosomes and is not trafficked to MIIC compartments (Fig. 7). For these experiments, DC were
exogenously pulsed on chamber slides with glycosylated Cy5-labeled
HER-2/neu protein for 1 h, then washed and allowed to process the
protein for an additional 1 h. The cells were fixed,
permeabilized, and stained with PE-labeled Abs against markers of
either early or late endosomes. In the top six panels of
Fig. 7, it is clear that HER-2/neu was colocalized with the early
endosomal markers, transferrin receptor, and MR. HER-2/neu was,
however, not colocalized with late endosomal markers HLA-DR, HLA-DM, or
LAMP-1, as demonstrated in the lower nine panels of Fig. 7.
HER-2/neu staining was proximal to the plasma membrane, with little or
no protein present in perinuclear endosomes, showing the same hindered
intracellular trafficking as that seen for MUC1.
|
MR has been shown to recycle back to the plasma membrane after
unloading its ligand in early endosomes (52). Mannose and
sucrose residues recognized by the MR on the glycosylated tumor MUC1
are positioned in a highly tandemly repeated array, possibly leading to
a high avidity receptor-ligand cross-linking. We hypothesize that as a
consequence of this interaction many MUC1 protein molecules will be
unable to dissociate from MR in early endosomes and will be eventually
recycled back (for possibly several rounds) to the cell surface where,
at a different pH, some MUC1 molecules may ultimately detach and be
released in the extracellular environment. To test this hypothesis we
exposed the DC, following a 30-min pulse with Ag, to acidic conditions
(0.15 M glycine, pH 4.0, for 30 min on ice), which favor the disruption
of receptor-ligand interactions at the cell surface. Following the acid
stripping of all ligand still on the cell surface, the cells were
returned to physiologic conditions, for 90 min after which the cells
were pelleted, and the supernatants were saved for detection by ELISA
of internalized MUC1 molecules that were recycled back to the cell
surface and detached from the receptors. Fig. 8 shows that following acid stripping,
there was loss of fluorescence indicating loss of internalized Ag.
There was an 80% decrease in the MFI for the glycosylated tumor
MUC1 (Fig. 8A) and a 55% decrease for the MUC1 140-mer
peptide (Fig. 8B). The control FITC-conjugated mannosylated
BSA (a ligand for the MR that dissociates in early endosomes) showed a
decrease of only 35% (Fig. 8C). We assayed the supernatants
by ELISA for the presence of either surface-bound MUC1, released by
acid treatment, or soluble MUC1, recycled from intracellular
compartments (Fig. 8D). We found greater amounts of released
Ag in the case of glycosylated MUC1 compared with the MUC1 peptide.
Furthermore, in the case of glycosylated MUC1, protein is clearly
recycled back to the cell surface bound to the receptor. When the bound
Ag is stripped off the receptor with acid treatment first, less MUC1 is
detected in the supernatant. The acid treatment of DC did not affect
their viability, as observed by forward-side scatter analysis on
FACS.
|
).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Internalization of Ags through the MR typically results in at least a 100-fold increase in efficiency of presentation by class II MHC over uptake via macropinocytosis (45, 62). In the case of MUC1, we show that binding and internalization via the MR do not deliver this Ag to the MIIC compartments. We propose that this is due to the exceptional avidity of binding of glycosylated MUC1 to MR via numerous tandemly repeated carbohydrates that does not allow dissociation from the receptor in the early endosomal sorting compartments (CURL). Therefore, MUC1 cannot be transported to later endosomal compartments for degradation and binding to class II MHC.
Long-term retention of Ag by immature DC has been also observed by other investigators (42). It is thought that long-term retention of Ag internalized in tissues enables the immature DC to travel from peripheral locations to lymph nodes before degrading its internalized Ag. DC then process and present those Ags to T cells in the lymph node. The authors of this study employed a murine fetal skin-derived DC line exhibiting the characteristics of immature DC. In this model the endosomal compartment in which the Ag was retained was characterized as a class II-MHC rich compartment that also contained the lysosomal marker LAMP1 and cathepsin D. There are, however, significant differences between the long-term retention compartment identified in immature DC and the compartment in which glycosylated MUC1 is retained. The compartment in which MUC1 is stored contains no markers of late endosomes or class II MHC, only early endosomal markers. Therefore, it is unlikely that MUC1 is being stored by DC for later degradation, but instead, DC appear unable to properly sort and transport MUC1 to late endosomes and lysosomes for processing.
While much is known about MUC1 glycosylation, there is a lot less information about the specific sugar residues present on HER-2/neu ECD, and more will need to be done to completely understand its processing and presentation by DC. Our findings presented here on the similarity in intracellular trafficking of MUC1 and HER-2/neu do show, however, that in addition to MUC1, other tumor glycoproteins may be inefficiently handled by APCs, which is expected to result in the lack of presentation by class II MHC and failure to stimulate Th cells (63).
Understanding how an Ag of interest is handled by the immune system, especially by APC at the priming step of the immune response, is critical for any effort directed toward manipulating the immune response against that Ag. In the case of MUC1, we now believe that the data presented here provide a reasonable explanation for our previously published observations that Th cell responses are not elicited against the soluble form of the Ag available either in vivo in the cancer patient or in vitro. We also know from studies presented here and from our previous studies (25, 26, 64) that we can overcome this problem by providing APCs with an unglycosylated synthetic form of MUC1. The long synthetic MUC1 peptide (100 aa) is currently in use in several cancer vaccine clinical trials.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 E.M.H. and A.M.V. contributed equally to this manuscript.
3 Current address: Division of Basic Sciences, Experimental Immunology Branch, National Cancer Institute, National Institutes of Health, Building 10, Room 4B36, 10 Center Drive, Bethesda, MD 20892.
4 Address correspondence and reprint requests to Dr. Olivera J. Finn, Department of Molecular Genetics and Biochemistry, W1142 Biomedical Science Tower, University of Pittsburgh, Pittsburgh, PA 15261.
5 Abbreviations used in this paper: DC, dendritic cell; ECD, extracellular domain; MFI, mean fluorescence intensity; KLH, keyhole limpet hemocyanin; MR, mannose receptor; CURL, compartments for uncoupling of receptors and ligands; MIIC, MHC class II compartment.
6 P. Beatty, F.-G. Hanisch, D. Beer Stolz, O. J. Finn, and P. Ciborowski. Biochemical characterization of the soluble form of tumor antigen MUCI isolated from sera and ascites fluid of breast and pancreatic cancer patients. Submitted for publication.
Received for publication January 19, 2000. Accepted for publication July 7, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. J. Exp. Med. 179:1109.ß dimers and facilitates peptide loading. Cell 82:155.[Medline]
This article has been cited by other articles:
|
|
 |
|
  T. L. Tinder, D. B. Subramani, G. D. Basu, J. M. Bradley, J. Schettini, A. Million, T. Skaar, and P. Mukherjee MUC1 Enhances Tumor Progression and Contributes Toward Immunosuppression in a Mouse Model of Spontaneous Pancreatic Adenocarcinoma J. Immunol., September 1, 2008; 181(5): 3116 - 3125. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Kadaoui and B. Corthesy Secretory IgA Mediates Bacterial Translocation to Dendritic Cells in Mouse Peyer's Patches with Restriction to Mucosal Compartment J. Immunol., December 1, 2007; 179(11): 7751 - 7757. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Napoletano, A. Rughetti, M. P. Agervig Tarp, J. Coleman, E. P. Bennett, G. Picco, P. Sale, K. Denda-Nagai, T. Irimura, U. Mandel, et al. Tumor-Associated Tn-MUC1 Glycoform Is Internalized through the Macrophage Galactose-Type C-Type Lectin and Delivered to the HLA Class I and II Compartments in Dendritic Cells Cancer Res., September 1, 2007; 67(17): 8358 - 8367. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ninkovic and F.-G. Hanisch O-Glycosylated Human MUC1 Repeats Are Processed In Vitro by Immunoproteasomes J. Immunol., August 15, 2007; 179(4): 2380 - 2388. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Sangha and C. Butts L-BLP25: A Peptide Vaccine Strategy in Non Small Cell Lung Cancer Clin. Cancer Res., August 1, 2007; 13(15): 4652s - 4654s. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. L. Beatty, S. E. Plevy, A. R. Sepulveda, and O. J. Finn Cutting Edge: Transgenic Expression of Human MUC1 in IL-10-/- Mice Accelerates Inflammatory Bowel Disease and Progression to Colon Cancer J. Immunol., July 15, 2007; 179(2): 735 - 739. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. S. Turner, P. A. Cohen, and O. J. Finn Lack of Effective MUC1 Tumor Antigen-Specific Immunity in MUC1-Transgenic Mice Results from a Th/T Regulatory Cell Imbalance That Can Be Corrected by Adoptive Transfer of Wild-Type Th Cells J. Immunol., March 1, 2007; 178(5): 2787 - 2793. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Meltzer and D. Goldblatt Pneumococcal Polysaccharides Interact with Human Dendritic Cells Infect. Immun., March 1, 2006; 74(3): 1890 - 1895. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. G. Wahrenbrock and A. Varki Multiple Hepatic Receptors Cooperate to Eliminate Secretory Mucins Aberrantly Entering the Bloodstream: Are Circulating Cancer Mucins the "Tip of the Iceberg"? Cancer Res., February 15, 2006; 66(4): 2433 - 2441. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. von Mensdorff-Pouilly, L. Kinarsky, K. Engelmann, S. E. Baldus, R. H. Verheijen, M. A. Hollingsworth, V. Pisarev, S. Sherman, and F.-G. Hanisch Sequence-variant repeats of MUC1 show higher conformational flexibility, are less densely O-glycosylated and induce differential B lymphocyte responses Glycobiology, August 1, 2005; 15(8): 735 - 746. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Carlos, H. F. Dong, O. M. Z. Howard, J. J. Oppenheim, F.-G. Hanisch, and O. J. Finn Human Tumor Antigen MUC1 Is Chemotactic for Immature Dendritic Cells and Elicits Maturation but Does Not Promote Th1 Type Immunity J. Immunol., August 1, 2005; 175(3): 1628 - 1635. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. E. Napper and M. E. Taylor The mannose receptor fails to enhance processing and presentation of a glycoprotein antigen in transfected fibroblasts Glycobiology, October 1, 2004; 14(10): 7C - 12C. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Monti, B. E. Leone, A. Zerbi, G. Balzano, S. Cainarca, V. Sordi, M. Pontillo, A. Mercalli, V. Di Carlo, P. Allavena, et al. Tumor-Derived MUC1 Mucins Interact with Differentiating Monocytes and Induce IL-10highIL-12low Regulatory Dendritic Cell J. Immunol., June 15, 2004; 172(12): 7341 - 7349. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Jackaman, C. S. Bundell, B. F. Kinnear, A. M. Smith, P. Filion, D. van Hagen, B. W. S. Robinson, and D. J. Nelson IL-2 Intratumoral Immunotherapy Enhances CD8+ T Cells That Mediate Destruction of Tumor Cells and Tumor-Associated Vasculature: A Novel Mechanism for IL-2 J. Immunol., November 15, 2003; 171(10): 5051 - 5063. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Schlienger, C. S. Chu, E. Y. Woo, P. M. Rivers, A. J. Toll, B. Hudson, M. V. Maus, J. L. Riley, Y. Choi, G. Coukos, et al. TRANCE- and CD40 Ligand-matured Dendritic Cells Reveal MHC Class I-restricted T Cells Specific for Autologous Tumor in Late-Stage Ovarian Cancer Patients Clin. Cancer Res., April 1, 2003; 9(4): 1517 - 1527. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Vlad, S. Muller, M. Cudic, H. Paulsen, L. Otvos Jr., F.-G. Hanisch, and O. J. Finn Complex Carbohydrates Are Not Removed During Processing of Glycoproteins by Dendritic Cells: Processing of Tumor Antigen MUC1 Glycopeptides for Presentation to Major Histocompatibility Complex Class II-restricted T Cells J. Exp. Med., December 2, 2002; 196(11): 1435 - 1446. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Higashi, A. Morikawa, K. Fujioka, Y. Fujita, Y. Sano, M. Miyata-Takeuchi, N. Suzuki, and T. Irimura Human macrophage lectin specific for galactose/N-acetylgalactosamine is a marker for cells at an intermediate stage in their differentiation from monocytes into macrophages Int. Immunol., June 1, 2002; 14(6): 545 - 554. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Sarkar, V. Kalia, M. Murphey-Corb, and R. C. Montelaro Detailed Analysis of CD4+ Th Responses to Envelope and Gag Proteins of Simian Immunodeficiency Virus Reveals an Exclusion of Broadly Reactive Th Epitopes from the Glycosylated Regions of Envelope J. Immunol., April 15, 2002; 168(8): 4001 - 4011. [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]
|
|
|
|
|
|
J. Schmielau and O. J. Finn Activated Granulocytes and Granulocyte-derived Hydrogen Peroxide Are the Underlying Mechanism of Suppression of T-Cell Function in Advanced Cancer Patients Cancer Res., June 1, 2001; 61(12): 4756 - 4760. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Soares, V. Mehta, and O. J. Finn Three Different Vaccines Based on the 140-Amino Acid MUC1 Peptide with Seven Tandemly Repeated Tumor-Specific Epitopes Elicit Distinct Immune Effector Mechanisms in Wild-Type Versus MUC1-Transgenic Mice with Different Potential for Tumor Rejection J. Immunol., June 1, 2001; 166(11): 6555 - 6563. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Nelson, S. Mukherjee, C. Bundell, S. Fisher, D. van Hagen, and B. Robinson Tumor Progression Despite Efficient Tumor Antigen Cross-Presentation and Effective "Arming" of Tumor Antigen-Specific CTL J. Immunol., May 1, 2001; 166(9): 5557 - 5566. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
