| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Melbourne Tumor Biology Branch, Ludwig Institute for Cancer Research, Austin and Repatriation Medical Centre, Heidelberg, Victoria, Australia;
University of Heidelberg, Heidelberg, Germany; and
Department of Molecular Biology and Immunology, CSL Ltd., Parkville, Melbourne, Australia
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2-microglobulin to stabilize the MHC complex on the cell
surface. In contrast, presentation of high affinity peptides was not
sensitive to cycloheximide or brefeldin A, which suggests that
different mechanisms may operate for presentation of high and low
affinity peptides by TAP-competent cells. High affinity peptides can
apparently compete with peptides in preloaded MHC class I molecules at
the cell surface, whereas low affinity peptides require empty MHC
molecules within cells. Accordingly, very high concentrations of
exogenous low affinity peptides in conjunction with active MHC class I
metabolism were required to allow successful presentation against a
background of competing intracellular high affinity peptides in
TAP-competent cells. These findings have implications for the design of
peptide and protein-based vaccines. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have chosen Melan-A as a model Ag system because its antigenicity is
well characterized and has multiple HLA-A2-binding analogs
(2, 3, 4, 5). The immunodominant HLA-A*0201-restricted peptide
is the 9-mer peptide AAGIGILTV comprising amino acids
27–35 (AAG) (6, 7). In addition, the native 10-mer
EAAGIGILTV (EAA) has also been described (8).
It has been demonstrated that an A27
L substitution in the
anchor residue at position 2, ELAGIGILTV (ELA),
resulted in increased affinity and immunogenicity of the peptide
without altering the specificity of the T cells to recognize the native
9-mer peptide (5, 9). This study investigates the
presentation of peptide Ags by APCs. The use of the native (low
affinity) AAG peptide and the modified (high affinity) ELA peptide has
enabled us to define two distinct mechanisms, or pathways, for
processing and loading these Ags into MHC class I molecules.
Recognition of HLA-A*0201-restricted peptides by CTL is often assessed by pulsing the peptides onto the TAP-deficient T2 cell line. However, clinical trials with peptide vaccines rely on APCs that express TAP to take up and present peptides in vivo. TAP-competent cells translocate endogenously generated peptides into the endoplasmic reticulum (ER),3 where they are loaded onto MHC class I molecules. These peptides are derived from cellular proteins degraded in the proteasomal complex (10, 11). TAP-deficient cells lack the transporter for MHC class I-restricted peptides to enter the ER and therefore present only peptides derived from leader sequences (12) or exogenous peptides. The absence of ER-localized peptides in TAP-deficient cells results in the accumulation of empty MHC class I molecules in the ER-Golgi intermediate compartments as well as on the cell surface. These empty MHC molecules can be rapidly loaded by available peptides (13). Therefore, exogenous peptides can potentially be loaded onto MHC complexes at the cell surface, a mechanism that would not require peptide uptake into the cell, or be internalized and loaded onto MHC molecules in the ER. It is known that exogenous peptides can enter intracellular compartments directly in a TAP-independent way (14). Within these ER compartments, high affinity peptides can bind to de novo-synthesized MHC class I molecules. The relative role of these two mechanisms for the loading of MHC molecules is not yet understood.
This study investigates these mechanisms for the presentation of high and low affinity peptide analogs by TAP-deficient cells (T2) and TAP-competent cells (EBV cell line, immature and mature dendritic cells (DC)). We report that in the presence of competing high affinity peptides, exogenously derived low affinity peptides (AAG) required MHC class I synthesis and vesicular transport for presentation at the cell surface. In contrast, the presentation of exogenous high affinity peptides (ELA) did not. This suggests that only high affinity peptides can replace peptides from already preformed MHC-peptide complexes. To further evaluate the relative roles of these mechanisms, we studied the duration of peptide presentation when cells were loaded with peptides of high or low affinity in the presence or absence of TAP. The high affinity peptide (ELA) was recognized for the longest period of time when presented by TAP-deficient cells or mature DC, indicating that peptide-MHC complexes are most stable when made up of high affinity peptides under conditions where newly formed, empty complexes are available.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Peripheral blood of normal laboratory volunteers was obtained in accordance with the guidelines of the Ludwig Institute for Cancer Research (Melbourne, Australia) and used to produce peptide-specific CTL lines and monocyte-derived DC (McDC). The human T2 cell line was kindly provided by Dr. P. Romero (Ludwig Institute, Lausanne, Switzerland). The HLA-A*0201+Melan-A+ tumor cell line LAR-5 was established and characterized in our laboratory from a patient with melanoma. Human K562 cells were a gift from H. Zogos (Rotary Bone Marrow Research Institute, Melbourne, Australia). The HLA-A*0201+Melan-A- tumor cell line SK-Mel37 was a gift from Dr. E. Stockert (Ludwig Institute, New York, NY). The EBV cell line (LCL-HV) was produced in our laboratory by transformation of B lymphocytes from a normal HLA-A*0201+ individual.
Media
Cell lines were maintained in RPMI 1640 (Trace Biosciences, Melbourne, Australia) supplemented with 60 mg/l penicillin G, 12.6 mg/l streptomycin, 2 mM L-glutamine, 1% nonessential amino acids, 5 x 10-5M 2-ME, and 10% heat-inactivated FCS (CSL, Melbourne, Australia). CTL lines were cultured in IMDM (Life Technologies, Grand Island, NY) and 5% pooled normal human serum (gift of the Victorian Tissue Typing Service, Royal Melbourne Hospital, Melbourne, Australia).
The mAbs and other reagents
The W6/32 anti-HLA class I Ab was purchased from DAKO
(Carpinteria, CA). The secondary sheep anti-mouse (FITC-conjugated)
Ab was purchased from SILENUS Labs (Boronia, Australia). The capture
and detection Abs for IFN-
ELISPOT assays were a kind gift from CSL.
MACS beads conjugated with anti-CD14 and used to purify blood
monocytes were purchased from Miltenyi Biotec (Sunnyvale, CA). All
peptides used in this study were produced by Multiple Peptide Systems
(San Diego, CA). Two Melan-A peptide analogs were used: AAGIGILTV (AAG,
native 9-mer) and ELAGIGILTV (ELA, modified 10-mer). In addition, the
influenza matrix protein peptide GILGFVFTL (FLU) was used as a
control peptide. To control for technical issues relating to peptide
reconstitution, quantity, and stability (15), experiments
were performed using two different batches of all peptides tested.
Peptides were reconstituted in 30% DMSO and stored as small aliquots
at -70°C. After thawing, peptides were only used once and then
discarded. Cycloheximide (Sigma-Aldrich, Castle Hill, Australia) was
used at 1 mg/ml. Brefeldin A (ICN Pharmceuticals, Seven Hills,
Australia) was used at a concentration of 10 µg/ml.
Cell separation, morphology, and flow cytometry
PBMC were separated by density fractionation using Ficoll-Hypaque (Pharmacia Biotech, Uppsala, Sweden), RBCs were lysed using NH4Cl, and CD14+ cells were separated with the MACS CD14 isolation kit according to the manufacturer’s instructions (Miltenyi Biotec). Cells for flow cytometry were resuspended in PBS plus 10% human serum, labeled with the primary Abs, washed, and labeled with the secondary Ab. Cells were fixed in PBS-2% formaldehyde-0.01% sodium azide. The immunophenotype was determined using a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA).
Establishment of a T2 cell line expressing endogenous FLU (T2*FLU)
Complementary oligonucleotides coding for the HLA-A*0201-restricted FLU and a preceding leader sequence matched for the +1 amino acid of the mature peptide (ApoH) were produced by CyberSyn (Lennis, PA) (sense, ctcatgattt ctccagtgct catcttgttc tcgagttttc tctgccatgt tgctattgca gggattttag gatttgtgtt cacgctctaa gctagct; and antisense, ctagagctag cttagagcgt gaacacaaat cctaaaatcc ctgcaatagc aacatggcag agaaaactcg agaacaagat gagcactgga gaaatcatga ggtac).
Oligonucleotides were annealed and ligated into the Apex-3 vector (16), the kind gift of B. Loveland (Austin Research Institute, Melbourne, Australia). Vector fragments containing the CMV promoter, minigene, SV40-poly(A) tail, and a hygromycin resistance gene were electroporated into T2 cells (210 mV, 500 µF, 20 µg DNA/2 x 107 cells). Cells were selected with hygromycin B (320 µg/ml, 120 U/ml) and cloned. Clones were screened for peptide expression in the 51Cr-release assay and the ELISPOT assay. Recognition of T2*FLU cells by a pre-established, peptide-specific CTL line was identical with the recognition of T2 cells pulsed with 10 µg/ml exogenous FLU over a range of E:T ratios in both assays (data not shown).
Culture of McDC
CD14+ monocytes
(105/well) from the PBMC of normal volunteers
were cultured in 24-well plates (Falcon; BD Biosciences) for 7 days in
1 ml RPMI 1640 and 10% FCS in the presence of GM-CSF (40 ng/ml;
Schering-Plough, Sydney, Australia) and IL-4 (1000 U/ml; kindly
provided by Schering-Plough). After 7 days, cultures were pooled,
adjusted to 105 DC/ml, and split into two samples
in a GM-CSF and IL-4-containing medium. A mixture of
maturation-inducing factors (10 ng/ml TNF-
(R&D Systems,
Minneapolis, MN); 1000 U/ml IFN- (RoferonA; Roche, Sydney,
Australia) and 1 µM PGE2 (ICN Pharmaceticals))
was added to one half of the DC cultures to induce DC maturation
(17), whereas the other half was used as a source of
immature DC. Both immature and mature McDC were evaluated in peptide
presentation assays on day 10.
Establishment of Melan-A-specific CTL lines
CTL lines specific for the Melan-A (27–35 aa) 9-mer peptide were established from PBMC of normal volunteers. Briefly, 106 PBMC were pulsed with the Melan-A 9-mer (AAG) peptide (10 µg/ml for 2 h at room temperature) and irradiated (3,000 rad). After two washes in RPMI 1640, peptide-pulsed, irradiated PBMC were incubated with equal numbers of nonirradiated PBMC in 0.5 ml IMDM and 5% human serum in 48-well plates. At day 3 of the culture, 5 U/ml IL-2 (PeproTech, Rocky Hill, NJ) was added. After 7 days, cells were restimulated with irradiated, peptide-pulsed, autologous PBMC, and cultures were continued in the presence of 25 U/ml IL-2 in 24-well plates. Thereafter, cells were restimulated weekly with 104 peptide-pulsed, irradiated (10,000 rad) T2 cells and 2 x 104 allogeneic, irradiated (10,000 rad) EBV cells/105 CTL. One week after the last restimulation, cells were used in the experiments described.
51Cr-release assay
51Cr-release assays were used weekly to test the CTL lines for specificity. T2 cells were labeled with 150 µCi of Na51Cr04 (DuPont/NEN, Boston, MA) for 1 h at 37°C. After three washes in RPMI 1640, cells were incubated with peptide in X-Vivo 20 (BioWhittaker, Walkersville, MD) for 1 h at 37°C, washed once, and resuspended in RPMI 1640 with 10% FCS at 104 cells/ml. Responder cells were preincubated at different concentrations (from 103 to 3 x 104 cells/ml) with unlabeled 5 x 104 K562 for at least 30 min. K562 is a NK cell target and was added to inhibit nonspecific lysis during the assay by contaminating NK cells. 51Cr-labeled target cells (1000/well) with or without peptide were cocultured with the CTL line for 4 h. Control wells containing K562 and labeled targets with or without peptide in medium alone or with 10% SDS were used to assess spontaneous release and maximal release, respectively. Plates were centrifuged (300 rpm for 5 min), and 50 µl of supernatant (SN) was loaded onto a LumaPlate containing solid scintillant (Packard Instrument, Meriden, CT) and air-dried. Plates were counted in a TopCount NXT (Packard). To calculate peptide-specific lysis, background lysis of targets pulsed with an irrelevant control peptide (FLU) was subtracted from lysis of targets pulsed with the Melan-A peptides, as published previously (18).
IFN- ELISPOT assays
ELISPOT assays were performed to quantify APC-target cell
interactions. Briefly, Millipore multiscreen plates (Millipore,
Molsheim, France) were coated with anti-IFN- Ab (5 µg/ml in
0.1 M NaHCO3 buffer (pH8.3) for 2 h at room
temperature) and blocked for 1 h at room temperature with PBS plus
10% FCS. A total of 5000 peptide-pulsed target cells were coincubated
overnight with 15,000 CTL. After lysing cells with
H2O for 30 min, the secondary HRP-labeled
anti-IFN- Ab was added for 2 h (10 µg/ml PBS, 3% FCS,
and 0.05% Tween 20). Spots were developed with
3-amino-9-ethyl-carbazole in acetate buffer for 8 min, washed under tap
water, and air-dried. To assess the numbers of IFN- spots
objectively, a video camera (TK-1280E; Zeiss, Oberkochen, Germany) and
the VideoPro software (Olympus, Melbourne, Australia) were used.
Image analysis included: 1) subtraction of background; 2) increase of
contrast; 3) definition of positive spots according to gray values; 4)
binary functions (clear and centers); 5) definition of area with the
draw and fill functions; and 6) analysis of the field. Background
images were taken from wells containing CTL alone. Target cells pulsed
with an irrelevant peptide (FLU) were used as negative controls for
peptide-specific cytokine release.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
51Cr-release assay: recognition and cell lysis.
Peptide-specific CTL lines were generated to detect functional
peptide-MHC molecules on the surface of target cells. Five different
CTL lines from three separate donors were established by weekly
restimulations with the native Melan-A AAG peptide. In addition, one
CTL line was raised against the high affinity ELA peptide. To
demonstrate the specificity of these lines for the naturally produced
(native) Melan-A epitope, cytotoxicity assays were performed using
melanoma cell lines. Fig. 1
shows that
the CTL lines generated with these peptides were capable of recognizing
and killing the
HLA-A2+Melan-A+ tumor cell
line (LAR-5), but did not recognize the
HLA-A2+Melan-A- control
tumor cell line (SK-Mel37). Interestingly, the CTL line raised against
the high affinity ELA peptide demonstrated the highest tumor-specific
lysis (Fig. 1).
|
A
shows the means ± SEM of five experiments in the 60:1 E:T ratio.
Lysis of T2 cells pulsed with the native AAG peptide could only be
detected at the 10 ng/ml or higher range. In contrast, the high
affinity ELA peptide could be detected at all peptide concentrations in
both E:T ratios. In this regard, the ELA peptide induced detectable
lysis at 3000-fold lower concentrations of peptide than that required
for the native Melan-A peptide (Fig. 2A). Furthermore, the
maximal levels of lysis achieved by the AAG peptide were seen at the 10
µg/ml concentration, which was equivalent to the level of lysis
induced by the high affinity ELA peptide at the 3 pg/ml concentration
(Fig. 2A). Similar results were obtained in the two lower
E:T ratios tested (data not shown). This indicates that the ELA peptide
is at least 106-fold more potent at inducing
CTL-mediated lysis than the native Melan-A peptide. Having established
this recognition pattern for five different CTL lines, all following
experiments were performed using the CTL line no. 3.
|
ELISPOT assay: recognition and IFN-
secretion–titration of analog peptides.
In vitro-generated peptide-specific CTL lines have been shown to
secrete IFN- in response to the appropriate MHC-peptide complex. The
ELISPOT assay demonstrates IFN- secretion by individual cells or
cell clusters and can therefore be used to numerically evaluate T
cell-target cell interactions. To investigate whether the Melan-A
peptide analogs differed in their capacity to induce IFN- secretion
by CTL lines, 5,000 peptide-pulsed T2 cells were cocultured overnight
with 15,000 CTL on nitrocellulose filter plates coated with
anti-IFN- Ab. To objectively assess the numbers of IFN-
spots, a video camera and the VideoPro software were used. Fig. 2
B shows the titration of the peptide analogs presented by
T2 cells to the Melan-A-specific CTL line (no. 3). The high affinity
ELA peptide was 
10,000-fold more efficient at inducing IFN-
spot-forming cells than the native AAG low affinity peptide. The
recognition of the ELA peptide induced maximal IFN- spot-forming
cells at concentrations between 1 and 100 ng/ml. ELISPOT assay–time course of peptide presentation
The IFN- ELISPOT assay was next used to determine the
kinetics of peptide-MHC complex expression on the surface of APC. After
2 h of exposure to peptide in serum-free medium, T2 cells were
washed twice and resuspended at 50,000 cells/ml in IMDM plus 5% human
serum. Cells were either used immediately as APC (0-h time point) or
incubated in 10-ml tubes at 37°C for a further 24, 48, or 84 h
before being used in the ELISPOT assays in an attempt to evaluate the
duration of surface peptide-MHC expression. During the period of the
experiment, cells remained viable without any significant
proliferation. Fig. 3A shows a
time course of peptide presentation on T2 cells relative to the ELA
peptide (10 µg/ml, 0-h time point). The AAG peptide (10
µg/ml) induced IFN- spot formation at the 0-h time point, but this
was only 20–30% of that seen with the high affinity ELA
peptide. Furthermore, T2 cells pulsed with the lower affinity AAG lost
50% of their spot-forming activity by 48 h and did not express
detectable peptide-MHC class I molecules by 84 h. In contrast, the
modified ELA peptide (10 µg/ml) continued to associate with class I
molecules and induced 33% of the maximal IFN- spot formation even
84 h after pulsing. Interestingly, 10-fold lower concentrations of
the high affinity ELA peptide (1 µg/ml) resulted in equivalent
IFN- spot-forming responses, highlighting the superiority of this
modified peptide analog as compared with the two lower affinity native
peptides (Fig. 3A).
|
To investigate the possibility that insufficient washing of free
peptide from the pulsed T2 cells contributed to the results, we
compared T2 cells exposed to 10 µg/ml ELA peptide, washed either two
or six times, and rested over a period of 24 and 48 h.
Peptide-pulsed T2 cells continued to induce equivalent levels of
IFN- secretion 48 h after removal of peptide even after six
washes (data not shown). In addition, SNs of T2 cells washed two or six
times contained comparable levels of intact peptides 24 h after
exposure (data not shown). This suggests that intact peptides were
released into the SN within 24 h after removal of exogenous
peptides (also see Fig. 3B).
We next examined whether the release of free peptides from an
intracellular location was responsible for the prolonged presentation
by T2 cells. T2 cells were pulsed with high (10 µg/ml) and low (0.1
µg/ml) concentrations of ELA peptide, and peptide presentation was
measured over a 48-h time period. In parallel samples, supernatants of
both peptide concentrations were examined for the presence of released
peptides 4, 24, and 48 h after peptide removal (Fig. 3B). In this experiment, both concentrations of ELA peptide
induced equivalent numbers of IFN- spots over a 48-h time period.
However, only T2 cells pulsed with the high peptide concentration (10
µg/ml) released free peptide into the SN that was detectable 4 and
24 h after peptide removal. In contrast, no peptide was detected
in the SN of T2 cells pulsed with the lower peptide concentration (0.1
µg/ml) (Fig. 3B). These results demonstrate that at the
lower peptide concentration, the release of free peptide into the SN is
not responsible for the prolonged presentation by T2 cells. However,
the release of free bioactive peptide within 24 h after pulsing
with high peptide concentrations was clearly demonstrated.
Intact intracellular peptides recovered 24 h after peptide pulsing
To assess whether peptides were being released from an
intracellular store, T2 cells were pulsed for 1 h with 10 µg/ml
ELA or a control peptide (FLU), washed three times, and recultured for
a further 24 h at 37°C. Following culture, cells were washed,
and MHC class I-associated peptides were stripped off the cell surface
by washing the cells twice for 5 min in citrate buffer (50 mM, pH 3.2).
Cells were then washed an additional three times in PBS. The
supernatant of the third wash was pulsed onto fresh T2 cells, and
ELISPOT assays were performed to measure free peptide. No free peptides
were found after the acid elution (data not shown). Cells were then
resuspended in 200 µl PBS. A small fraction of the cells was examined
for surface expression of conformational MHC class I complexes by FACS
using the W6/32 Ab (Fig. 4A).
The remaining peptide-stripped T2 cells (2 x
106) were freeze-thawed, the cell lysates passed
through a small-gauge needle, and nuclei and cell debris removed by
centrifugation. The cell lysate was then pulsed onto fresh T2 cells,
and the presence of peptides within the intracellular material was
evaluated using 51Cr-release and IFN- ELISPOT
assays. Fig. 4A clearly demonstrates that MHC class I
molecules were not detectable on the surface of T2 cells following acid
elution as assessed by FACS. However, the T2 lysates contained
measurable quantities of intact ELA peptide as detected by the
51Cr-release (Fig. 4B) and IFN-
ELISPOT assays (Fig. 4C). These results demonstrate that the
high affinity Melan-A peptide was stored inside pulsed T2 cells for at
least 24 h and suggest that one mechanism for prolonged
presentation of MHC class I-restricted peptides may involve loading of
MHC complexes with peptides from internal stores.
|
The observed recovery of peptides from acid-stripped cell lysates suggests that exogenous peptides may load onto intracellular MHC class I. For instance, they might compete inside the ER for peptide-binding grooves of de novo-synthesized MHC molecules. Alternatively, exogenous peptides may also displace endogenous peptides from established MHC complexes on the cell surface. Both mechanisms were examined by blocking protein biosynthesis using cycloheximide or blocking vesicular transport by brefeldin A. To mimic the abundance of other competing peptides in TAP-competent cells, we saturated HLA-A2-binding sites of T2 cells by adding an irrelevant, HLA-A2-restricted high-affinity peptide (FLU) before the relevant Melan-A peptides. T2 cells were cultured for 1 h in 1 mg/ml cycloheximide or in 10 µg/ml brefeldin A. FLU (10 µg/ml) was added for 2 h more to one half of the cells to saturate all peptide-receptive HLA-A*0201 molecules. Finally, cells were split again, and AAG or ELA peptide (10 µg/ml) was added for an additional hour without intermittent washing steps. Thus, cells had been exposed to cycloheximide or brefeldin A for a total of 4 h, to the irrelevant FLU peptide for 3 h, and to the relevant Melan-A peptide for 1 h. The viability of cycloheximide- or brefeldin A-treated cells was 80% compared with the untreated controls (100%). Overnight ELISPOT assays were performed as described without further addition of cycloheximide or brefeldin A.
Fig. 5 shows remarkable differences
between the presentation of low affinity and high affinity peptides
under these conditions. Competing irrelevant high affinity peptides
(FLU, brefeldin A, and cycloheximide) all reduced presentation of the
low affinity AAG peptide independently and synergistically (Fig. 5A). In contrast, no major effect of these factors on the
presentation of ELA was observed (Fig. 5B). To investigate
the possibility that the inhibition of presentation of AAG by
cycloheximide and brefeldin A might be caused be the inhibition of
soluble 2-microglobulin production and
secretion, exogenous 2-microglobulin was added
in two more experiments together with the AAG peptide during the last
hour of the experiment. Fig. 5C demonstrates that exogenous
2-microglobulin did not antagonize the
inhibitory effect of cycloheximide and brefeldin A.
|
Presentation of high and low affinity peptides in the presence of competing endogenous peptides by T2*FLU cells
To investigate whether the superior peptide presentation by TAP-deficient cells can be explained solely by competition of endogenous peptides, we established a T2 cell line expressing a minigene of FLU after a leader sequence (T2*FLU). These peptides are transported into the ER, bypassing TAP, and provide a source of endogenous peptides competing for HLA-A2 molecules. We have produced a stable and clonal T2*FLU line using a peptide-specific CTL line to control peptide presentation. This T2*FLU line was recognized by peptide-specific CTL at least as well as or better than T2 cells pulsed with 10 µg/ml exogenous FLU peptide (T2 plus FLU). These experiments were performed by 1) titrating CTL in the 51Cr-release assay with 103 T2*FLU or T2 plus FLU and by 2) titrating T2*FLU or T2 plus FLU cells in the ELISPOT assay with 104 CTL (data not shown).
To investigate the effect of endogenously produced high affinity
peptides on the presentation of exogenous peptides, the capacity of the
T2*FLU cells and T2 cells to present varying concentrations of the AAG
and ELA peptides was examined in ELISPOT assays. Fig. 6 demonstrates that the presence of
endogenous irrelevant high affinity peptides in T2*FLU cells markedly
reduced the capacity of these cell lines to present exogenous low
affinity peptides at all concentrations (103-fold
lower half-maximal activity). In contrast, the presentation of
exogenous high affinity ELA peptide was only reduced at the three
lowest concentrations (10-fold lower half-maximal activity). These
results support the hypothesis that high and low affinity peptides
compete at different levels for MHC peptide-binding grooves. The
presence of high affinity peptides inside the ER has little affect on
the ability of exogenous high affinity peptides to bind MHC class I
complexes on the cell surface. In contrast, exogenous low affinity
peptides are unable to displace high affinity peptides in this
situation. Therefore, competition with endogenous peptides can result
in poor presentation of exogenous low affinity peptides by
TAP-competent cells.
|
To examine peptide presentation by cells with functional TAP, an
HLA-A*0201+ EBV cell line was used in both the
51Cr-release and ELISPOT assays. DC were not used
initially because of insufficient numbers of cells. Subsequently
HLA-A*0201+ DC were studied in the ELISPOT assay.
Fig. 7 shows the means of four separate
experiments comparing peptide-pulsed EBV and T2 cells at three E:T
ratios (60:1, 20:1, and 7:1). Peptides were the AAG peptide and ELA
peptide. It is seen that T2 cells presented both peptides more
efficiently than EBV cells, as assessed by induction of CTL lysis.
Although EBV cells were capable of presenting the ELA peptide as
efficiently as T2 cells at high concentrations (Fig. 7B),
they were not efficient at presenting the low concentration of peptide
(0.0001 µg/ml) (Fig. 7C). Furthermore, the AAG peptide
stimulated only low levels of lytic activity
(<20\% specific lysis at the highest E:T ratio and peptide concentration) (Fig. 7>A).
Equivalent findings were seen in ELISPOT assays (Fig. 8). The titration of the high affinity
ELA peptide on T2 cells and EBV cells is shown in Fig. 8A.
TAP-competent EBV cells required higher concentrations of ELA peptide
to stimulate CTL responses. Surface detection of ELA peptide/MHC was
also lost more rapidly on EBV cells, with little or no specific
activity remaining by 48 h (Fig. 8B). Furthermore, EBV
cells presenting the native AAG peptide induced small numbers of
IFN- spots in only one of three experiments and failed to induce
IFN- spots in two of three experiments. These results confirm that
TAP-competent cells present exogenous peptides less efficiently than
TAP-deficient cells.
|
|
-ELISPOT assay
DC perform different functions according to their stage of
maturation. Immature DCs are best viewed as Ag uptake cells, which
mature to become APCs when exposed to proinflammatory stimuli such as
TNF-, IFN-, and PGE2, or to T cell-derived
factors such as CD40 ligand. Consequently, the effectiveness in
presentation of exogenous peptides might be expected to differ for DC
at different stages of maturation. To evaluate peptide presentation by
DC at each of these stages, Ag presentation to CTL lines by
peptide-pulsed immature and mature McDC was compared using the ELISPOT
assay (Fig. 9A). Immature McDC
expressed medium levels of MHC class I, and this was further
up-regulated upon maturation using the combination of TNF- (10
ng/ml), IFN- (1000 U/ml), and PGE2 (1 µM)
(data not shown). These mature DC also up-regulated CD80, CD83, and
CD86 (data not shown). In accordance with the findings of other groups,
our results showed that micropinocytosis of FITC-labeled dextran (38.2
kDa) was more active in immature DC as compared with mature DC (data
not shown). Fig. 9A shows that both immature and mature DC
presented the high affinity ELA peptide, whereas the native AAG peptide
could only be presented by mature DC. The dramatic differences between
peptide presentation by TAP-competent and TAP-deficient cells are
further highlighted in Fig. 9B, which examines the high
affinity ELA peptide (10 µg/ml). Unlike DC, it was found that
TAP-deficient T2 cells continued to stimulate peptide-specific IFN-
secretion 84 h after peptide pulsing, whereas a substantial number
of IFN- spots was still induced by mature DC 48 h after peptide
exposure. Immature DC failed to present peptide beyond 24 h. This
demonstrates that the stage of DC maturation influences the efficiency
and stability of peptide presentation, with mature DC being capable of
more prolonged peptide presentation. Furthermore, these TAP-competent
cells lost exogenously derived peptide-MHC complexes more rapidly than
TAP-deficient T2 cells.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Day et al. (14) have previously demonstrated that exogenous peptides can be directly delivered to the ER, a pathway that was affected by expression of functional TAP and blocked by pinocytosis inhibitors. Our finding that bioactive, exogenous high affinity ELA peptide was still present in lysates of T2 cells 24 h after peptide exposure extends these observations. It also provides reason to believe that exogenously derived peptides can access de novo-synthesized MHC molecules for a prolonged period of time following Ag uptake.
The use of peptide analogs of different affinities revealed two
mechanisms of peptide loading into MHC class I molecules. Exogenous low
affinity peptides (AAG) required de novo synthesis of MHC in the
presence of competing high affinity peptides. This was demonstrated by
the ability of cycloheximide (an inhibitor of protein synthesis) and
brefeldin A (an inhibitor of vesicular transport) to block the
presentation of AAG peptide in the presence of saturating levels of a
competing peptide. In contrast, the presentation of high affinity
peptides (ELA) was largely insensitive to these agents. Here, the
predominant mechanism appears to be displacement of bound peptides from
established MHC complexes on the cell surface, even if exogenous high
affinity peptides are also able to access de novo MHC molecules. Our
results are in accordance with a previous report comparing MHC peptide
loading of TAP-/- and
TAP+/+ cells using a
Kd-restricted high affinity peptide
(OVA257–264) (19). Similar to our
results in Fig. 5B, this group showed that at 26°C,
brefeldin A did not inhibit the presentation of high affinity peptides
in both TAP-/- and TAP+/+
cells. However, analog low affinity peptides were not investigated in
that study.
This argument is also substantiated by our results using EBV cells and immature DC where the mechanisms for endogenous peptide trafficking were intact. For both of these cell types, presentation of the exogenous low affinity peptide AAG was reduced, whereas the high affinity peptides were presented. Similarly, T2 cells transfected with a minigene containing FLU showed reduced levels of presentation, particularly of the AAG peptide. In contrast, mature DC were capable of presenting the AAG peptides. This may reflect the increased synthesis of MHC class I molecules by mature DC (20), which improves the chances of high concentrations of low affinity peptides to compete for peptide-binding grooves inside the ER. All TAP-competent cells presented the high affinity peptide analog (ELA), again suggesting that this peptide did not require de novo-synthesized MHC molecules to be presented. Furthermore, prolonged presentation of detectable surface MHC-peptide complexes also appeared to correlate with the activity of MHC class I metabolism in DC or the amount of peptide-receptive MHC in T2 cells.
In summary, our study demonstrates that the availability of peptide-receptive MHC class I influences the presentation of low affinity peptides if other peptides compete for binding sites. This availability of peptide-receptive MHC is highest in T2 cells because the lack of peptides keeps a large proportion of receptive MHC inside the cells. If T2 cells are exposed to competing high affinity peptides (either extracellular peptides or peptides derived from minigenes), the availability of MHC class I is decreased and presentation of low affinity peptides is reduced. Likewise, TAP-competent cells such as EBV-transfected B cells or DC have less empty, peptide-receptive HLA-A2 molecules than T2 cells and therefore present low affinity peptides less efficiently. Thirdly, if DC are matured, de novo synthesis of MHC class I is increased (20). Thus, mature DC will have more peptide-receptive HLA-A2 molecules within the ER than immature DC.
These findings have several implications for clinical vaccination
strategies that target either endogenous, resident APC at the site of
vaccination or in vitro-generated DC for peptide presentation. In this
regard, the efficacy of intradermal peptide injections may increase
when combined with DC-activating stimuli such as IFN-
(21), immunostimulating complexes
(22), or nucleic acid (23). Similarly,
optimal peptide presentation will likely require that in vitro-expanded
DC be matured or activated before Ag loading and injection into
patients to prolong their peptide-presenting capacity and to ensure
effective presentation at the site of T cell interactions. These
results highlight the importance of how understanding the mechanisms of
Ag processing and presentation by APC can impact on the application of
vaccine-based immunotherapy.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Thomas Luft, University of Heidelberg, Medizinische Klinik und Poliklinik, Hospitalstrasse 3, 69221 Heidelberg, Germany. E-mail address: Thomas_luft{at}med.uni-heidelberg.de
3 Abbreviations used in this paper: ER, endoplasmic reticulum; DC, dendritic cells; McDC, monocyte-derived DC; SN, supernatant; FLU, influenza matrix protein, peptide GILGFVFTL.
Received for publication December 5, 2000. Accepted for publication June 21, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  D.-W. Liu, Y.-C. Yang, H.-F. Lin, M.-F. Lin, Y.-W. Cheng, C.-C. Chu, Y.-P. Tsao, and S.-L. Chen Cytotoxic T-Lymphocyte Responses to Human Papillomavirus Type 16 E5 and E7 Proteins and HLA-A*0201-Restricted T-Cell Peptides in Cervical Cancer Patients J. Virol., March 15, 2007; 81(6): 2869 - 2879. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Dunn, P. J. Rizkallah, E. Baston, T. Mahon, B. Cameron, R. Moysey, F. Gao, M. Sami, J. Boulter, Y. Li, et al. Directed evolution of human T cell receptor CDR2 residues by phage display dramatically enhances affinity for cognate peptide-MHC without increasing apparent cross-reactivity. Protein Sci., April 1, 2006; 15(4): 710 - 721. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Bettosini, M. T. Fiorillo, A. Magnacca, L. Leone, M. R. Torrisi, and R. Sorrentino The C Terminus of the Nucleoprotein of Influenza A Virus Delivers Antigens Transduced by Tat to the trans-Golgi Network and Promotes an Efficient Presentation through HLA Class I J. Virol., December 15, 2005; 79(24): 15537 - 15546. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Elkord, P. E. Williams, H. Kynaston, and A. W. Rowbottom Differential CTLs specific for prostate-specific antigen in healthy donors and patients with prostate cancer Int. Immunol., October 1, 2005; 17(10): 1315 - 1325. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X.-L. Huang, Z. Fan, B. A. Colleton, R. Buchli, H. Li, W. H. Hildebrand, and C. R. Rinaldo Jr. Processing and Presentation of Exogenous HLA Class I Peptides by Dendritic Cells from Human Immunodeficiency Virus Type 1-Infected Persons J. Virol., March 1, 2005; 79(5): 3052 - 3062. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Gnjatic, D. Atanackovic, M. Matsuo, E. Jager, S. Y. Lee, D. Valmori, Y.-T. Chen, G. Ritter, A. Knuth, and L. J. Old Cross-Presentation of HLA Class I Epitopes from Exogenous NY-ESO-1 Polypeptides by Nonprofessional APCs J. Immunol., February 1, 2003; 170(3): 1191 - 1196. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. D. Davis, M. Jefford, P. Parente, and J. Cebon Rational approaches to human cancer immunotherapy J. Leukoc. Biol., January 1, 2003; 73(1): 3 - 29. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Luft, P. Luetjens, H. Hochrein, T. Toy, K.-A. Masterman, M. Rizkalla, C. Maliszewski, K. Shortman, J. Cebon, and E. Maraskovsky IFN-{alpha} enhances CD40 ligand-mediated activation of immature monocyte-derived dendritic cells Int. Immunol., April 1, 2002; 14(4): 367 - 380. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||
) and 0.1
µg/ml (
) ELA 10-mer peptide. Columns, T2 cells were pelleted 4,
24, and 48 h after the three initial washes, and SN were pulsed
onto fresh T2 cells. Only T2 cells pulsed with 10 µg/ml peptide (
