| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Clinical Cooperation Group Hyperthermie,
Institute of Molecular Immunology, GSF National Research Center for Environment and Health, and
Medizinische Klinik III, Klinikum Grosshadern, Ludwig Maximilians University, Munich, Germany;
Institute of Pathology, GSF National Research Center for Environment and Health, Neuherberg, Germany; and
¶ Department of Radiation Oncology, Dana-Farber Cancer Institute and Harvard Medical School, Boston, MA 02115
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
), suggesting that signaling DC maturation is an intrinsic
property of the HSP70-PC itself and related to receptor-mediated
binding. The cross-presentation of a shared human tumor Ag together
with the exquisite efficacy are important new aspects for HSP70-based
immunotherapy in clinical anti-cancer vaccination strategies, and
suggest a potential extension of HSP70-based vaccination protocols from
a patient-individual treatment modality to its use in an allogeneic
setting. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Immunization with HSP peptide complexes (HSP-PC) is exquisitely dependent on the presence of functional APCs in the immunized host, since depletion of such cells renders the host incapable of mounting immune responses after injection of HSP-PC preparations (8). Dendritic cells (DCs) are very effective activators of CTLs, a process that requires the presentation of Ag bound to MHC molecules, together with expression of adhesion and costimulatory molecules (9, 10). In particular, the ability to present exogenous Ags through "cross-presentation" is a key feature of DCs. This term was introduced to describe the representation of exogenous cell-associated Ags by MHC class I (11) and II molecules (12). It became evident that GP96- and HSP70-chaperoned peptides can be presented to CTLs by DCs in the context of MHC class I molecules, and uptake of GP96 or HSP70 requires receptor-mediated endocytosis (2, 13, 14). Recent data provide evidence for the existence of distinct receptors for HSPs (e.g., GP96, HSP90, HSP70) on murine APCs (15), and one of them has been identified as CD91 (16, 17).
The mechanisms involved in the stimulation of T cell responses via HSP70 and GP96 were studied in murine systems using induced tumors and model Ags (18, 19, 20), OVA (21), and viral Ags (22, 23). More recently, immunization with tumor-derived HSPs has also been demonstrated for spontaneous tumors (24). These analyses provided the principle knowledge of HSP-mediated cross-presentation and the involvement of APCs and have been the basis for the first clinical trials involving tumor-derived GP96 (5, 6, 25).
Different from most of the studies described so far, we investigated the role of HSP70 family members in a system that more closely resembles the patient situation. Most studies investigating mechanistic events related to HSP-mediated cross-presentation involved highly immunogenic Ags either induced by mutagenesis or overexpressed by transfection, a situation that does not reflect most human cancers. We selected a human melanoma system and investigated the role of HSP70 in the cross-presentation of the tyrosinase Ag, an endogenously expressed, nonmutated tumor-associated differentiation Ag of low immunogenicity (26) that is shared among tumors of the melanocytic lineage.
We demonstrate that HSP70-PC purified from tyrosinase-positive
(HSP70-PC/tyr+) but not from tyrosinase-negative
(HSP70-PC/tyr-) melanoma cells deliver the
tyrosinase Ag to immature DCs for MHC class I-restricted T cell
recognition. Activation of the tyrosinase-specific T cell clone was
inhibited when binding of HSP70-PC/tyr+ to DCs
was competitively blocked by HSP70-PC not carrying the tyrosinase Ag. T
cell stimulation by immature DCs incubated with
HSP70-PC/tyr+ was very efficient even without
additional DC maturation signals (e.g., exogenous TNF-),
demonstrating the ability of tumor-derived HSP70-PC to act as a
chaperone for peptides and a signal for DC maturation. Using a human
tumor system, it can be concluded that endogenously expressed
nonmutated and shared tumor Ags with low immunogenicity
(26) are chaperoned by HSP70 and efficiently presented by
DCs for T cell recognition.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
SPA-810 mAb (clone C92F3A-5, specific for the inducible protein HSP70) and rat mAb SPA-815 (clone 1B5, specific for the protein HSC70) were purchased from StressGen Biotechnologies (Victoria, British Columbia, Canada). Anti-HSP70 mAb (clone BRM-22, specific for both HSP70 and HSC70) and anti-FITC mAb (clone FL-D6) were obtained from Sigma-Aldrich (Taufkirchen, Germany). Rat mAb HSP-6B3, specific for the inducible isoform of HSP70 was generated by our in-house facility. Approximately 50 µg of recombinant human (rh)-HSP70 expressed in Escherichia coli (protein SPP-755; StressGen Biotechnologies) were injected i.p. and s.c. into LOU/C rats. After a 2-mo interval, a final boost was given i.p. and s.c. 3 days before fusion. Fusion of the myeloma cell line P3X63-Ag8.653 with the rat immune spleen cells was performed according to standard procedure. Hybridoma supernatants were tested in a solid-phase immunoassay using recombinant HSP70 adsorbed to polystyrene microtiter plates. Following incubation with culture supernatants for 1 h, bound mAbs were detected using peroxidase-labeled goat anti-rat IgG + IgM Abs (Dianova, Hamburg, Germany) and o-phenylenediamine as chromogene in the peroxidase reaction. Crude E. coli extract was used as a negative control. Positive-reacting hybridomas were further tested on HSC70 (constitutive form of HSP70, SPP-750; StressGen Biotechnologies). mAb HSP-6B3 (rat IgG1) specific for the inducible HSP70 was used in this study.
The C-19 goat polyclonal Ab specific for tyrosinase was purchased from
Santa Cruz Biotechnology (Santa Cruz, CA). Anti-
-tubulin mAb (Santa
Cruz Biotechnology) was used in Western blotting procedures to control
for protein loading. Abs used for FACS analysis included PE-conjugated
anti-CD14, anti-CD83, and goat anti-mouse IgG (Beckman
Coulter, Krefeld, Germany). Cytokines used for cell culture included
IL-2 (Cetus, Emerville, CA), IL-4, TNF- (Biomol, Hamburg,
Germany), and GM-CSF (Hölzel Diagnostika, Köln, Germany).
The tyrosinase peptide (YMNGTMSQV; aa 368–376) was synthesized and
purified by Dr. M. Eulitz in our in-house facilities at the GSF
Institute of Molecular Immunology (Munich, Germany).
Cells
The human melanoma cell lines, 624.38-MEL and SK23-MEL
(tyrosinase-positive; HLA-A*0201-positive) (27, 28), were
kind gifts from Dr. M. C. Panelli (National Institute of Health,
Bethesda, MD), and A375-MEL (tyrosinase-negative; HLA-A*0201-positive)
(27, 28) was purchased from the American Type Culture
Collection (Rockville, MD). HLA-A2 and tyrosinase expression were
verified by FACS analysis using Ab HB54 (Ref. 29 and data
not shown) or by Western blot analysis using the goat polyclonal
antiserum C-19 (see Fig. 2
B), respectively. As a second
tyrosinase-negative cell line, the EBV-transformed B lymphoblastoid
cell line (B-LCL) of a local donor was used. All tumor cell lines were
cultured in RPMI supplemented with 10% FCS. Cell lines were routinely
tested for mycoplasma using the manufacturer’s directions (Biochrom,
Berlin, Germany). PBMCs were prepared from leukapheresis samples
or from peripheral blood of healthy donors by density gradient
centrifugation over Ficoll/Hypaque (Amersham Pharmacia Biotech,
Freiburg, Germany). To obtain CD14+ monocytes,
PBMCs were incubated in 75-cm2 plastic flasks
(Nunc, Wiesbaden-Biebrich, Germany) for 2 h and the nonadherent
cells were washed off. Adherent cells were cultured in VLE-RPMI culture
medium (Biochrom) supplemented with 2 mM glutamine, 100 U/ml
penicillin/streptomycin, and 1% autologous serum. GM-CSF (1000 U/ml)
and IL-4 (800 U/ml) were admixed to the monocytes on day 0 to generate
DCs. GM-CSF alone was again added on day 4 of culture. On day 7, the
DCs were tested for their ability to stimulate peptide-specific T cells
in the cross-presentation assay. For binding experiments with
FITC-labeled proteins, DC maturation was induced by the addition of a
cytokine mixture (IL-1, 10 ng/ml; IL-6, 1000 U/ml; TNF-, 10
ng/ml; PGE2, 1 µg/ml; a kind gift from Dr. E.
Kaempgen, Department of Dermatology, Julius Maximilians University,
Würzburg, Germany) on day 7. Mature DCs were used on day
9.
|
Isolation of HSP70-PC from melanoma cells
The HSP70-PC were isolated as previously described (31) with minor modifications. A total of 10 ml of packed cell pellet of either A375-MEL, 624.38-MEL, SK23-MEL, or B-LCL cells was homogenized in 40 ml of hypotonic buffer A (10 mM NaHCO3, 0.5 mM PMSF, pH 7.1) by repeated freeze and thaw cycles and sonication, and a 100,000 x g supernatant was obtained. The sample buffer was exchanged for buffer B (20 mM Tris-acetate, 20 mM NaCl, 15 mM 2-ME, 3 mM MgCl2, 0.5 mM PMSF, pH 7.5) via gel filtration chromatography using Sephadex G-25. The sample was applied directly to an ADP-agarose column equilibrated with buffer B. The column was washed extensively with buffer B until protein was undetectable in the eluate by absorbance at 280 nm. Finally, the column was incubated with buffer B containing 3 mM ADP at room temperature for 30 min and subsequently eluted with the same buffer (25 ml). The buffer of the eluate was exchanged again with Sephadex G-25 for buffer C (20 mM Na2HPO4, 20 mM NaCl, pH 7.0). The protein was separated by anion exchange chromatography using MonoQ Sepharose and eluted over a 20–600 mM NaCl gradient. Fractions containing HSP70-PC as a single protein, as determined by SDS-PAGE silver stain and by Western blot with anti-HSP70 mAb (clone BRM-22) were pooled. Sephadex G-25, ADP-agarose, MonoQ Sepharose, ADP, and all buffer ingredients were obtained from Sigma-Aldrich. Protein content was determined by the Lowry method using the protein assay kit of Bio-Rad (Munich, Germany).
SDS-PAGE, two-dimensional (2D) isoelectric focusing (IEF)/SDS-PAGE, and Western blot analysis
For analysis of protein content in various fractions of the HSP70-PC isolation steps, samples were denatured by boiling for 5 min in SDS sample buffer and separated on 10% SDS-PAGE. After electrophoresis, proteins were either silver stained or transferred to nitrocellulose membranes (Sartorius, Goettingen, Germany) and probed with appropriate Ab as described in the figure legends. Detection of proteins was achieved by the ECL system (Amersham Life Science, Karlsruhe, Germany).
The natural expression levels for HSC70, HSP70, and tyrosinase in
cultured cell lines were determined using cell lysates (lysis buffer:
2% 3-((3-cholamidopropyl)dimethylammonio)-1-propanesulfonate in
50 mM HEPES and 200 mM NaCl (pH 7.5), and protease inhibitors). Equal
amounts of protein (40 µg/lane measured with the Bio-Rad assay kit)
were boiled for 5 min in SDS-sample buffer and separated on a 10%
SDS-PAGE. After electrophoresis, Western blot analysis was performed as
described above. For detection of tyrosinase, the polyclonal goat serum
C-19 was used. HSP70 and HSC70 were detected using the rat mAbs HSP-6B3
and SPA-815, respectively. Equal protein loading was verified by
reprobing the blots with anti--tubulin Ab (Santa Cruz
Biotechnology; data not shown).
For 2D-IEF/SDS-PAGE, the first dimension used IEF using a Multiphor II electrophoresis unit and ImmunobilineDryStrip (IPG) (pH 3–10 NL), all from Amersham Pharmacia Biotech. A total of 100 ng of purified HSP70-PC were solubilized in lysis buffer (10 M urea, 1% DTT, 4% 3-((3-cholamidopropyl)dimethylammonio)-1-propanesulfonate, 2.5 mM EDTA, 2.5 mM EGTA) supplemented with 0.05% bromophenol blue and IPG buffer (pH 3–10 NL; Amersham Pharmacia Biotech), loaded onto the IPG (pH 3–10 NL), and covered with paraffin. The running conditions were 91 kVh. IEF was followed by a vertical 10% SDS-PAGE. The IPG strip was equilibrated and placed on top of the vertical 10% SDS-PAGE. After separation, Western blotting and immune detection using the anti-HSP70 and HSC70 mAb (BRM-22) and the ECL kit were performed.
Fluorescence labeling of proteins
Purified HSP70-PC (from A375-MEL melanoma cells) or BSA (1 mg/ml) was incubated with FITC (50 µg; Sigma-Aldrich) or FluoroLink Cy5 (Amersham Pharmacia Biotech) in carbonate-bicarbonate buffer (0.1 M; pH 9.5) overnight at 4°C, or 5 h at room temperature for Cy5. Free unconjugated FITC was removed by passing the mixture over a gel filtration column (Sephadex G-25; Amersham Pharmacia Biotech). For Cy5-labeling, protein was dialyzed (12–14 kDa dialysis membrane) for 12–14 h. FITC-conjugated proteins were analyzed by SDS-PAGE and immunoblotting using anti-HSP70 mAb (clone BRM-22) and anti-FITC Ab. Labeled proteins were centrifuged at 100,000 x g before use to remove any particulate matter.
FACS analysis
For phenotypic characterization, 2 x 105 cells were stained with PE-labeled mAb to CD14 and CD83, FITC-labeled BSA, and FITC-labeled HSP70-PC/tyr+ or FITC-HSP70-PC/tyr- (each 100 ng/ml) in PBS supplemented with 5% FCS for 30 min at 4°C. For competition studies, labeled (100 ng/ml) and unlabeled proteins (in concentrations to reach indicated ratios) were mixed and added to the cells. Cells were washed with ice-cold PBS and analyzed by flow cytometry using a FACScan with CellQuest software (BD Biosciences, Mountain View, CA). The analysis gate was set on propidium iodide negative cells and cell type characteristic forward (forward light scatter) and orthogonal scatter (side light scatter). For intracellular staining of HSC70 and HSP70, cells were permeabilized using the FIX and PERM kit (Dianova). Then, mAb SPA-815 (HSC70) and SPA-810 (HSP70), respectively, in combination with FITC-labeled secondary reagents were applied.
Confocal microscopy
Immature DCs were incubated with 5 µg Cy5-labeled HSP70-PC or
Cy5-labeled BSA for 30 min at 4°C (surface staining) or at 37°C
(uptake studies). After washing, cells were settled on
poly-L-lysine-coated glass slides, fixed with 4%
paraformaldehyde (PFA) in PBS for 30 min at room temperature, and
mounted with VECTASHIELD (Vector Laboratories, Burlingame, CA). Samples
were analyzed for transmission and fluorescence using a Zeiss model
LSM510 confocal laser-scanning microscope (Zeiss, Oberkochen, Germany)
equipped with an external Helium-Neon Laser 633 nm lens CAPO 63x/1.2 W
as well as equipment for digital interference contrast (DIC). Confocal
section of specific fluorescence and DIC contrast image were taken
simultaneously. Fluorescence of the Cy5-labeled probe was detected
using an excitation wavelength of 633 nm, and a 650-nm LP emission
filter. Cy5 fluorescence is visualized in yellow (see Fig. 4B). DIC images were treated by multiplicative shading
correction using software KS 400 (Zeiss).
|
Activation of the T cell clone TyrF8 was assessed by measurement
of the amount of IFN-
released into the culture supernatant. DCs
derived from monocytic precursors were seeded on day 7 at a
concentration of 104 cells/well in 96-well round
bottom plates in 100 µl of DC culture medium, and indicated
concentrations of the HSP70-PC were added. After 24 h, DCs were
induced to mature by addition of 200 U/ml TNF- and incubated for a
further 24 h. As a control for the T cell stimulation capacity of
the DCs, tyrosinase peptide (aa 368–376; YMNGTMSQV) was added
exogenously at concentrations ranging from 1–10 µg/ml to
TNF--matured DCs. Before T cell addition, the DCs were irradiated
(5,000 rad). TyrF8 cells were added (2 x
104 cells/well) in 100 µl of medium to give
final concentrations of 25 U/ml IL-2, 5% FCS, and 5% human serum.
After 24 h, culture supernatants were harvested and the content of
IFN- was measured by OptEIA (BD Biosciences). When the
cross-presentation assay was performed to investigate the ability of
HSP70-PC to mature DCs, we added polymyxin B (Sigma-Aldrich) at a
concentration of 1 µg/ml together with HSP70-PC to rule out the
possible influence of LPS contamination. For Ab-blocking experiments,
before adding the T cells, DCs were incubated for 1 h at 37°C
with the mAb HB54 (20 µg/ml) recognizing an antigenic determinant
shared by HLA-A*02 and B17 (29). In some experiments,
immature DCs were fixed with PFA (1%) for 10 min at room temperature,
then they were washed three times with excess of VLE medium before
being used in cross-presentation assays (32).
Statistical analyses
The statistical significance of experimental values was assessed by means of Student’s t test. *, **, and *** represent values of p < 0.5, p < 0.05, and p < 0.005, respectively. For the calculation of the dose response value Y50, the statistics menu of the Origin 4.1 program and the logistic nonlinear curve fitting for the dose response in pharmacology and biology were used. Y = (A1 - A2)/(1 + (x/x0) exp(p) + A2) with A1 = -2.845; A2 = 556,25; x0 = 74,710; p = 2,2351. The second derivation of the above equation is set to zero and the resulting turning point of the sigmoid curve represents Y50.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We previously showed that rh-HSP70 binds to human monocytes
(33) and DCs (34). In this study, we
demonstrate that HSP70-PC isolated from human melanoma cell lines
exhibit a differential binding pattern to human APCs derived from the
monocytic lineage. Monocytes (CD14+,
CD83-), immature DCs
(CD14-, CD83-) and mature
DCs (CD14-, CD83+) were
coincubated with FITC-conjugated HSP70-PC at 4°C to exclude
endocytosis, and surface binding characteristics were analyzed by flow
cytometry (Fig. 1A). BSA-FITC
used as a control did not result in significant binding to any of the
APCs tested. HSP70-PC bound to immature and mature DCs with similar
high intensity (mean fluorescence of 40), whereas monocytes bound
HSP70-PC at a low intensity (mean fluorescence of 8). The binding
characteristics of HSP70-PC thus differed from our previous results
with rh-HSP70, which did not bind to mature DCs (34).
|
A). This composition was
found to reflect the natural HSP70 expression pattern of the melanoma
cell lines, 624.38-MEL, SK23-MEL, and A375-MEL, that were used for
HSP70-PC isolation. As demonstrated by intracellular FACS analysis and
Western blot, these cell lines expressed the HSC70 and HSP70 isoforms
at physiological growth conditions (Fig. 2B).
To negate the possibility that binding of HSP70-PC to DCs was a
consequence of nonspecific events, competition assays between unlabeled
HSP70-PC and FITC-labeled HSP70-PC were performed (Fig. 1B).
Unlabeled HSP70-PC competitively inhibited the binding of FITC-labeled
HSP70-PC to immature DCs. Inhibition was shown to be dependent on the
ratio of FITC-labeled to nonlabeled HSP70-PC and displayed a saturation
kinetics characteristic for receptor-mediated binding. The inhibition
of binding was highly significant at a 10-fold excess of unlabeled
HSP70-PC (p < 0.005). In contrast, the surface
binding was not affected using a 1:10 ratio of HSP70-PC-FITC to
unlabeled BSA.
HSP70 from melanoma cells chaperones the tyrosinase peptide and delivers it to DCs for MHC class I-restricted cross-presentation
To determine the functional consequence of HSP70-PC binding to
APCs, experiments were performed in which we tested whether peptides
chaperoned by HSP70-PC are effectively cross-presented by MHC class I
molecules to T cells. HSP70-PC were isolated from two
tyrosinase-positive (HSP70-PC/tyr+) and one
tyrosinase-negative (HSP70-PC/tyr-)
melanoma cell lines, 624.38-MEL, SK23-MEL, and A375-MEL, respectively
(Refs. 27 and 28 , and Fig. 2B),
and incubated with immature DCs from HLA-A*0201-positive donors in
vitro. The expression of HLA-A*0201 by DCs was required because the
tyrosinase peptide-specific T cell clone (TyrF8) is
HLA-A*0201-restricted. After incubation with HSP70-PC, the DCs were
treated with TNF- to induce maturation and to switch function from
Ag uptake and processing to presentation. Pulsed DCs were then assayed
for their capacity to stimulate the tyrosinase peptide-specific and
HLA-A*0201-restricted T cell clone TyrF8. TyrF8 stimulation was
measured by its ability to release IFN-.
DCs pulsed with HSP70-PC/tyr+ stimulated
TyrF8 to secrete IFN- in an HSP70-PC dose-dependent manner (Fig. 3A, 
). IFN- secretion
reached a plateau at 600 ng/ml, with a half-maximal value
(Y50) of IFN- achieved at 74.76
ng/ml of HSP70-PC/tyr+. The amount of IFN-
secreted was comparable to the maximum stimulation obtained using the
HLA-A*0201- and tyrosinase-positive melanoma cell lines 624.38-MEL and
SK23-MEL (data not shown). TyrF8 activation was Ag-dependent,
since HSP70-PC purified from melanoma cells not expressing the
tyrosinase Ag (HSP70-PC/tyr-) were unable to
stimulate TyrF8 (Fig. 3A, 
). IFN- secretion was
further dependent on the interaction of both DCs and T cells, since
neither DCs alone (Fig. 3B) nor T cells alone (Fig. 3C) incubated with HSP70-PC-secreted IFN-. To confirm
that the stimulation of TyrF8 by DCs pulsed with HSP70-PC from
tyrosinase-positive melanoma cells was indeed specific and
HLA-A*02-dependent, we performed blocking experiments using the
anti-HLA-A*02-specific Ab HB54 (29). The release of
IFN- by TyrF8 was inhibited by preincubation of HSP70-PC-pulsed DCs
with HB54 (p < 0.05 at 10 ng/ml of
HSP70-PC/tyr+ and p < 0.005 at
100 ng/ml of HSP70-PC/tyr+; Fig. 3D).
|
To test that receptor-dependent binding of HSP70-PC is required
for T cell activation, the T cell stimulation capacity of DCs pulsed
with HSP70-PC/tyr+ was measured in the presence
of increasing amounts of HSP70-PC/tyr- in a
cross-presentation assay. IFN- secretion induced by
HSP70-PC/tyr+ was inhibited by 80% when DCs were
coincubated with a 10-fold excess of
HSP70-PC/tyr- (Fig. 4A, p <
0.005). However, excess of BSA did not influence IFN- secretion.
The binding of HSP70-PC to DCs and early downstream consequences of
binding were analyzed by confocal laser scanning microscopy. Immature
DCs were incubated for 30 min with Cy5-conjugated HSP70-PC at 4°C to
exclude endocytosis, or at 37°C to induce uptake. Cy5-labeled BSA was
used as a negative control. After staining, cells were settled on
poly-L-lysine-coated glass slides, fixed and analyzed for
transmission and fluorescence. Consistent with the observed FACS
analysis (see Fig. 1), BSA did not result in detectable staining of DCs
(Fig. 4B, c, d, g,
h, k, and l), while all DCs were
stained strongly positive with HSP70-PC (Fig. 4B,
a, b, e, f, i,
and j). Different staining patterns were observed at 4°C
and 37°C. At 4°C (Fig. 4B, a, b,
e, and f), the fluorescence signal was localized
to the cell surface. In contrast, at 37°C (Fig. 4B,
i and j) this surface staining was replaced by a
vesicular staining at two distinct subcellular locations. Fluorescent
signals were localized to perinuclear areas and clusters of focal
staining near the cell surface, presumably early endosomes.
These staining patterns suggested that HSP70-PC, after binding to the
cell surface, was translocated into the cell interior, a process
requiring active cell metabolism. To prove that HSP70-PC does not
deliver its peptide cargo to cell surface MHC directly but chaperones
it through an intracellular pathway for loading onto newly synthesized
MHC class I molecules, immature DCs were fixed with PFA before being
used in cross-presentation assays. PFA fixation completely abrogated T
cell stimulation, while exogenously added tyrosinase peptide was still
efficiently presented (Fig. 4C).
HSP70-PC-dependent cross-presentation and T cell stimulation do not require additional external DC maturation signals
For our cross-presentation assays, we used immature DCs because
they demonstrated strongest binding for HSP70-PC and because they are
highly efficient in Ag uptake and processing (10). Based
on the rationale that for the process of T cell stimulation mature DCs
are the most efficient, we added TNF- exogenously after HSP70-PC had
bound to immature DCs and had delivered the peptide cargo. In the
meantime, others and we had observed that rh-HSP70 stimulates secretion
of inflammatory cytokines, including TNF-, from monocytes and DCs
(33, 35, 36), and induces maturation of DCs (34, 37, 38). Therefore, we reasoned that DCs through binding HSP70-PC
might be stimulated to release TNF- and induce their maturation by
an autocrine loop. To investigate whether DCs after incubation with
HSP70-PC need exogenous TNF- for efficient cross-presentation, we
incubated immature DCs with HSP70-PC/tyr+ and
either left them untreated (intrinsic DC maturation by HSP70-PC) or
gave TNF- (external DC maturation) before addition of the T cells.
As shown in Fig. 5, cross-presentation by
HSP70-PC-treated DCs without exogenously added TNF- was even
stronger than that with additional TNF- (p
< 0.005). Polymyxin B, a potent inhibitor of LPS, was included in the
cross-presentation assay. No inhibitory effect on IFN- secretion by
the T cells was observed, ruling out the possibility that endotoxin
contamination within the HSP70-PC preparations was responsible for DC
maturation and their ability to efficiently stimulate the T cells.
Immature DCs treated with HSP70-PC only consistently performed better
in Ag-specific and allogeneic T cell stimulation assays than those
treated with HSP70-PC and TNF- (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Using HSP70-PC/tyr+ and HSP70-PC/tyr-, and by focusing on the nature of the APCs that mediate cross-presentation as well as the biochemical composition of the tumor-derived HSP70-PC, we were able to demonstrate that the immunogenic potential of HSP70 as a tool to induce anti-tumor immune responses can be extended to naturally expressed nonmutated human tumor Ags of low immunogenicity (26). Binding of HSP70-PC to DCs and intracellular events are required for HSP70-mediated cross-presentation. Although HSP70-PC bound to immature and mature DCs with similar intensity, immature DCs were more efficient in cross-presentation than mature DCs (data not shown). This makes sense from an immunological point of view, since immature DCs are better in Ag uptake than monocytes or mature DCs and have a strong capacity to process Ag (10, 41). Ag processing ability is potentially useful if peptides bind to HSP70 as longer precursors.
Our results demonstrate further that HSP-PC-mediated cross-presentation
by immature DCs does not require external maturation signals, such as
TNF-. This finding is consistent with previous observations that
HSPs represent natural danger signals to the immune system. When
released by stressed cells (42, 43), they stimulate
monocytes and DCs to secrete proinflammatory cytokines (TNF-, IL-12)
(33, 35, 36, 44), and are maturation signals for immature
DCs (34, 37, 38, 45, 46). Our results demonstrating
efficient cross-presentation without external DC maturation signals
indicate that the two properties—the chaperoning of antigenic peptides
and the induction of DC maturation—are intimately linked within
tumor-derived HSP70-preparations. A similar conclusion can be reached
for GP96 (47). The significance for the clinical use of
tumor-derived HSP preparations in stimulating anti-tumor immune
responses is discussed below.
The biochemical analysis of our HSP70 preparations revealed that they consisted of both the HSC70 and the HSP70. This was found to reflect the natural expression pattern of HSP70 and HSC70 in the melanoma cell lines used for HSP70-PC isolation. Similar constitutive expression of HSP70 has also been described for surgical specimens of primary and metastatic human melanoma (48). Our previous findings that rh-HSP70 but not recombinant HSC70 is able to deliver the DC maturation signal (34) indicate that the heterogeneous composition of tumor-derived HSP70-PC might be of functional relevance for the cross-presentation.
The ability to cross-present HSP-bound Ag has also been shown for APCs
other than DCs, including monocytes (49) and macrophages
(50). Using blood monocytes, Castelli et al.
(49) demonstrated HSP70-mediated cross-presentation for
melanoma Ags other than tyrosinase. However, in their system a much
higher number of APCs and significantly more HSP70-PC were required for
T cell stimulation than in our system. Possibly, the poor binding of
HSP70 to monocytes is one explanation for this difference (see Fig. 1A). Furthermore, it is our own experience that experimental
variations in APCs, i.e., DCs grown in FCS vs autologous serum, not
fully differentiated DCs still expressing residual CD14, or matured DCs
expressing CD83, influence the efficiency of cross-presentation (data
not shown).
The studies presented in this report are of special clinical interest
for HSP70-based vaccinations. First, HSP70-PC-mediated Ag presentation
by DCs is very efficient, requiring low amounts (in the nanogram range)
of HSP70-PC (Fig. 3; and Ref. 6). This can be explained by
the receptor-mediated uptake for HSP70-PC vs fluid phase uptake and
surface peptide exchange mechanisms for exogenous peptides,
respectively. In addition, Binder et al. (51) described
that HSP70 positively influences cross-presentation of chaperoned
peptides by efficiently directing them to an endoplasmic
reticulum/Golgi compartment where loading onto the MHC class I
molecules occurs. The observation that an endoplasmic reticulum/Golgi
localization is important for efficient cross-presentation is
consistent with our confocal microscopic data that show a perinuclear
staining of immature DCs incubated with Cy5-labeled HSP70-PC at 37°C
(see Fig. 4B). An additional explanation for the efficacy of
HSP preparations to induce anti-tumor immune responses might be
related to the dual function of HSPs delivering Ag and inducing DC
maturation. Linking these two properties is one possibility to ensure
that Ag presentation occurs in an environment optimal for T cell
stimulation. Second, initial vaccination studies using murine tumor
models demonstrated that the anti-tumor immunity achieved by
vaccination with GP96 or HSP70 preparations is restricted to the tumor
from which the GP96/HSP70 was isolated (1, 6). Therefore,
HSP-based vaccination strategies are considered patient-individual
treatment modalities. This view is challenged by our observation, and
that of Castelli et al. (49), demonstrating that HSP70
isolated from human melanoma cells chaperone naturally expressed
nonmutated and shared human melanoma Ags and transfer them to APCs for
T cell recognition. If cross-presentation of shared human tumor Ags by
HSP70-PC followed by efficient T cell stimulation is routinely achieved
with HSP70-PC, the clinical application of HSP70-based vaccines may be
extended from a patient-individual treatment to use in an allogeneic
vaccination setting. Third, our novel insights into the mechanistic
events responsible for HSP70-mediated cross-presentation by DCs are of
additional interest for treatments involving hyperthermia. Clinical
hyperthermia has been found to be effective as a locoregional treatment
modality for certain solid tumors (52, 53). For melanoma
in particular, a randomized phase III study in humans has been
completed showing improvement of local tumor control and survival
benefits in patients with multiple lesions after hyperthermia treatment
(54). During clinical hyperthermia, peak temperatures of
up to 42°C can be achieved in the tumor tissue, and at this
temperature the de novo synthesis of heat inducible HSP70 is
up-regulated (55). Within this temperature range, local
necrosis occurs which might result in the release of HSP (56, 57) and uptake by DCs (38, 47) with subsequent
processing and presentation of the associated peptides. Since
HSP70-mediated cross-presentation is efficient without additional DC
maturation signals, an environment optimal for T cell stimulation is
ensured. Hyperthermia, by up-regulating HSP70 or other stress proteins
(e.g., HSP110 and grp170; Ref. 58) and by causing local
necrosis (57) in tumor tissue, has the potential to
directly activate the immune system against tumors.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 E.N., R.G., and V.M. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Elfriede Noessner, Institute for Molecular Immunology, GSF National Research Center for Environment and Health, 81377 Munich, Germany. E-mail address: Noessner{at}gsf.de
4 Current address: Department of Hematology and Oncology, Klinikum University of Regensburg, 93053 Regensburg, Germany.
5 Abbreviations used in this paper: HSP, heat shock protein; B-LCL, B lymphoblastoid cell line; 2D, two dimensional; DC, dendritic cell; HSC70, constitutively expressed HSP; HSP70, heat-inducible HSP; HSP-PC, HSP peptide complexes; HSP70-PC/tyr+, HSP70-PC isolated from tyrosine-positive melanoma cell lines; HSP70-PC/tyr-, HSP70-PC isolated from tyrosinase-negative melanoma cell lines; IEF, isoelectric focusing; rh, recombinant human; PFA, paraformaldehyde; DIC, digital interference contrast; GP, glucose-regulated protein.
Received for publication October 10, 2001. Accepted for publication September 9, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
12,14-prostaglandin J2 in transgenic adenocarcinoma of mouse prostata cells. Cancer Res. 60:4714.v5 and CD36, and cross-present antigens to cytotoxic T lymphocytes. J. Exp. Med. 188:1359.
B pathway. Int. Immunol. 12:1539.This article has been cited by other articles:
|
|
 |
|
  H. Bendz, B.-C. Marincek, F. Momburg, J. W. Ellwart, R. D. Issels, P. J. Nelson, and E. Noessner Calcium Signaling in Dendritic Cells by Human or Mycobacterial Hsp70 Is Caused by Contamination and Is Not Required for Hsp70-mediated Enhancement of Cross-presentation J. Biol. Chem., September 26, 2008; 283(39): 26477 - 26483. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Valentinis, A. Capobianco, F. Esposito, A. Bianchi, P. Rovere-Querini, A. A. Manfredi, and C. Traversari Human recombinant heat shock protein 70 affects the maturation pathways of dendritic cells in vitro and has an in vivo adjuvant activity J. Leukoc. Biol., July 1, 2008; 84(1): 199 - 206. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. Binder, J. B. Kelly III, R. E. Vatner, and P. K. Srivastava Specific Immunogenicity of Heat Shock Protein gp96 Derives from Chaperoned Antigenic Peptides and Not from Contaminating Proteins J. Immunol., December 1, 2007; 179(11): 7254 - 7261. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kettner, F. Kalthoff, P. Graf, E. Priller, F. Kricek, I. Lindley, and T. Schweighoffer EWI-2/CD316 Is an Inducible Receptor of HSPA8 on Human Dendritic Cells Mol. Cell. Biol., November 1, 2007; 27(21): 7718 - 7726. [Abstract] [Full Text] [PDF]
|
|
|
|
mean fluorescence 1). Error bars indicate SEM. Expression of
HSC70, HSP70, and tyrosinase was also analyzed by Western blotting.
Gels were run in parallel and probed with Abs to HSC70 (SPA-815), HSP70
(HSP-6B3), and tyrosinase (C-19).
), or a mixture of labeled HSP70-PC and unlabeled BSA,
ratio 1:10 (
