| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Pharmaceutical Sciences, School of Pharmacy, Texas Tech University Health Sciences Center, Amarillo, TX 79106;
Receptor Logic, Ltd., Amarillo, TX 79106;
Department of Microbiology and Immunology, Texas Tech University Health Sciences Center, Lubbock, TX 79430;
Department of Microbiology and Immunology, University of Oklahoma Health Sciences Center, Oklahoma City, OK 73104; and
¶ Sidney Kimmel Cancer Center, San Diego CA 92121
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
and TNF-
for Her2/neu protein and HLA-A2 molecule expression did not reveal a direct correlation between the levels of Her2/neu Ag, HLA-A2 molecule, and Her2(369)-A2 complex expression. However, compared with untreated cells, cytokine-treated cell lines showed an increase in Her2(369)-A2 epitope density that directly correlated with enhanced tumor cell death (p = 0.05). Although a trend was observed between tumor cell lysis and the level of the Her2(369)-A2 complex for untreated cells, the association was not significant. These findings suggest that tumor cell susceptibility to CTL-mediated lysis may be predicted based on the level of specific peptide-MHC class I expression rather than on the total level of TAA expression. Further, these studies demonstrate the potential of the TCRm mAb for validation of endogenous HLA-peptide epitopes on tumor cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
The Her-2/neu protein is an intensively investigated TAA and a member of the tyrosine kinase family of growth factor receptors (4, 5). The gene is frequently amplified and the protein product often overexpressed in many cancers, including breast and ovarian cancer (6, 7). The Her-2/neu protein is also a target for the therapeutic Ab trastuzumab. In stark contrast to what might be predicted for the presentation of MHC class I epitopes on the tumor cell surface due to the expression pattern of Her-2, Herrmann et al. (8) reported that overexpression actually resulted in a reduction in the level of several components of the Ag presentation machinery. Further, they demonstrated that Her2/neu-overexpressing cells transduced with vaccinia virus encoding for the tyrosinase protein displayed reduced tyrosinase-specific CTL-mediated lysis as compared with control cells expressing tyrosinase but not the Her2/neu protein. These findings raise several questions regarding TAA expression in general and Her2/neu expression in particular as well as questions about the relationship of these molecules to CTL epitope presentation and CTL-mediated killing of tumor cells. Unfortunately, quantitative determination of specific peptide-HLA complex density on the surface of tumor cells has been hindered by the limited availability of reagents to directly visualize these molecules. Direct visualization of peptide-MHC complexes on tumor cells would provide a number of advantages over current methods for CTL epitope identification and confirmation.
In recent years Abs have been used for direct detection and visualization of specific peptide-MHC complexes on the surfaces of cells (9, 10). Early attempts to produce such Abs relied on the immunization of mice with cells displaying MHC complexes loaded with a specific peptide and resulted in a low frequency of Abs that were, in general, characterized by weak binding affinity for specific peptide-MHC complexes (11). More recently, Reiter and coworkers (12, 13) isolated anti-peptide-MHC mAbs from both mouse and human Ab phage display libraries. These recombinant Abs bind peptide-pulsed cells but weakly stain tumor cells expressing endogenously processed and presented peptide-MHC complexes (14). This weak staining may be due to the affinity of phage-derived recombinant Abs that are suboptimal for staining tumor cells with low expression of specific peptide-MHCs. For the current study we generated a high affinity MHC-peptide specific mAb using a new methodology that we recently developed. This Ab, called a TCR mimic (TCRm), has high affinity and avidity for MHC-peptide complexes and is capable of detecting low densities of the specific MHC-peptide complex present on tumor cells (J. A. Weidanz and T. Woodburn, manuscript in preparation).
As discussed above, the relationship between TAA expression, MHC-peptide density, the recognition of tumor cells by CTL, and eventual tumor cell lysis has not yet been elucidated. The continued development of diagnostic and therapeutic protocols for cancer would benefit from a better understanding of the relationships between these critical mechanisms. The Her2/neu protein was selected for the current studies because of its biologic and clinical relevance. Total cellular levels of the Her2/neu protein were determined by ELISA and correlated with Her2/neu receptor expression on the tumor cell surface. For direct visualization of peptide-MHC class I complexes, we used the TCRm Ab 1B8, which specifically recognizes the dominant Her2/neu peptide 369–377 (designated Her2(369)) in the context of the HLA-A2 complex (15, 16) Our results indicate that a direct correlation does not exist between the Her2/neu protein level and the presentation of the dominant epitope Her2(369)-A2 on tumor cell lines. The CTL killing of tumor cells pretreated with IFN- and TNF-, however, was dependent on the density of the Her2(369)-A2 complexes present on tumor cells. Importantly, these results suggest an association between CTL recognition and the number of MHC-peptide complexes present on the surface of tumor cells and show that the TCRm is a useful tool for directly determining the susceptibility of tumor cells to CTL killing.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Anti-c-ErbB2/c-Neu (TA-1; mouse IgG1) Ab was purchased from Calbiochem. Polyclonal Abs goat anti-mouse-IgG (H + L chains)-HRP and goat anti-mouse IgG H chain-PE were purchased from Caltag Laboratories. The isotype control Abs mouse IgG1 and IgG2b were purchased from Southern Biotech. The peptides designated Her-2(369) (from Her2/neu; KIFGSLAFL, residues 369–377), eIF4G(720) (from eukaryotic initiation translation factor 4; VLMTEDIKL, residues 720–728), TMT(40) (from human chorionic gonadotropin-
; TMTRVLQGV, residues 40–48), and VLQ(44) (from human chorionic gonadotropin-; VLQGVLPAL, residues 44–53) as well as the control peptides were synthesized at the University of Oklahoma Health Science Center (Oklahoma City, OK) using a solid phase method and purified by HPLC to >90%.
Cell lines and human PBMCs
The human lymphoblastoid cell line T2 (HLA-A*0201) was purchased from the American Type Culture Collection. The mouse myeloma cell line P3X.63Ag8.653 (fusion partner), the BB7.2 anti-HLA A2.1 mAb-expressing mouse hybridoma cell line, the normal human mammary epithelial cell line (NHMEC 184A; HLA-A*0201+), and human tumor cell lines (MDA-MB-231, Saos-2, MCF-7, SW620, COLO205, BT-20, and SKOV3) were also purchased from the American Type Culture Collection. All human tumor cell lines were HLA typed using a DNA sequence-based technique described by Ellexson-Turner et al. (17). The MDA-MB-231, Saos-2, MCF-7, SW620, and COLO205 cell lines were identified as HLA-A*0201+ (positive), and the BT-20 and SKOV3 cell lines were typed as HLA-A*0201– (negative).
Human PBMCs from anonymous donors were obtained from separation cones of discarded apheresis units from the Coffee Memorial Blood Bank (Amarillo, TX) after platelet harvest. Cells were separated on a Ficoll gradient and then washed, counted, and resuspended in AIM-V medium before being stained with 3F9, 1B8, and anti-Her2/neu (TA-1) Abs.
Generation of HLA-A2-peptide complexes
Soluble HLA-A2 was prepared from inclusion bodies essentially as described by Altman et al. (18). The human HLA-A*0201 H chain gene, a kind gift from Dr. William Hildebrand (University of Oklahoma, Oklahoma City, OK), was amplified by PCR and cloned into the pAC4 plasmid containing the birA amino acid sequence (Avidity). The human -2 microglobulin gene was previously cloned into an expression vector for production in Escherichia coli strain BL-21 (19). The refolded monomer was concentrated and purified on an S-75 size exclusion column by fast protein liquid chromatography (Pharmacia) and then biotinylated using the biotin ligase enzyme according to the manufacturer’s instructions (Avidity). Tetramers were formed by mixing the biotin-tagged, refolded HLA-A2-peptide complex with streptavidin at a molar ratio of 4:1, respectively. Tetramers were purified on an S-200 Sephadex size exclusion column, and the protein concentration was determined by BCA protein assay (Pierce).
Generation of 1B8 TCRm mAb
Female BALB/c mice 6–8 wk of age were immunized with a solution containing 50 µg of a purified HLA-A2-Her2/neu (369–377) complex and a Quil-A adjuvant (Sigma-Aldrich) at 15-day intervals. One week after the third immunization, splenocytes were fused with the P3X63.Ag8.653 myeloma cell line using a ClonaCell-HY kit (StemCell Technologies). After 2 wk in semisolid medium, single clones were picked, transferred to 96-well tissue culture plates, and grown for 3–4 days in RPMI 1640 medium supplemented with 10% FBS and penicillin-streptomycin, and the supernatant was tested for the appropriate mAb production by ELISA as described below in the paragraph titled "TCRm specificity ELISA." All hybridomas selected are of the IgG1 isotype.
Growth and purification of 1B8 TCRm
The 1B8 hybridoma line was cultured in RPMI 1640 medium supplemented with 10% FBS and penicillin-streptomycin in T-175 flasks. The TCRm mAb was purified from supernatant by affinity chromatography with protein A-Sepharose (Amersham Biosciences/GE Health). The purity of the TCRm was assessed by SDS-PAGE.
TCRm specificity ELISA
Reactivity of supernatants and purified TCRm was assessed by ELISA as follows. Ninety six-well plates (Maxisorb; Nunc) were coated overnight at 4°C with 0.1 µg of purified HLA-A2-peptide complexes in PBS. After blocking with 5% nonfat milk in PBS, TCRm mAb at the appropriate concentration was incubated for 1 h at room temperature. After washing three times in PBS buffer, bound Ab was detected by incubation with a HRP-goat anti-mouse IgG, and color was developed with ABTS substrate (Pierce). OD was measured at 405 nm with a Victor II microplate reader (PerkinElmer). The BB7.2 mAb was used for evaluating HLA-A2 expression, the TCRm clone 3F9 (anti-TMT(40)-A2 mAb, murine IgG1) served as isotype and specificity control, and the murine Abs IgG1 and IgG2b from Southern Biotechnology were used as isotype control Abs.
Her2/neu ELISA
Cellular levels of Her2/neu were determined by preparing tumor cell lysates and quantifying Her2/neu with the c-ErbB2/c-Neu Rapid Format ELISA (Calbiochem) according to the manufacturer’s instructions. Her2/neu protein was detected in a sandwich ELISA using two mouse mAbs. The detector Ab was bound to HRP-conjugated streptavidin, and color was developed by incubation with the tetramethylbenzidine substrate (Pierce). The concentration of Her2/neu in the samples was quantified by generating a standard curve using known concentrations of Her2/neu provided in the kit.
T2 binding assay
T2 cells are HLA-A2+ and are deficient in TAP 1 and 2 proteins (20). T2 cells (5 x 105) were incubated in AIM-V medium (Invitrogen Life Technologies) and loaded with each of the peptides Her2/neu(369), TMT(40), or eIF4G(720) (50 µg/ml). After 4 h the cells were washed to remove excess peptide and stained with either BB7.2 Ab (0.5 µg/ml) to detect the level of HLA-A2 molecules present on the surface or 1B8 TCRm mAb (0.5 µg/ml). Pulsing the cells with peptides that have high affinity for HLA-A2 stabilizes the bound peptide complex and leads to an increase in the BB7.2 staining of the cells that received peptide vs those that did not (negative controls) (21, 22). Bound Ab was detected using the PE-conjugated goat anti-mouse IgG H chain-specific polyclonal Ab. In all experiments the IgG1 and IgG2b isotype controls were included to determine nonspecific staining.
Acid stripping and reloading of T2 cells with peptides
T2 cells (5 x 106/ml) were acid stripped (0.131 M citric acid and 0.067 M Na2HPO4 (pH 3.3)) for 45 s, washed twice with 50 ml of RPMI 1640 supplemented with 2 mM HEPES, and resuspended at 3.3 x 106/ml in 30 µg/ml 2-microglobulin (Fitzgerald Industries) (23, 24). Cells were then incubated for 3.5 h in a 20°C water bath with each peptide (2 µM), washed, stained with Abs, and evaluated on a BD FACScan. Subsequent analysis was performed using CellQuest software version 3.3 (BD Biosciences).
Tumor cell staining
All adherent tumor cell lines were grown in medium specified by the American Type Culture Collection and detached using 1x trypsin/EDTA (0.25% trypsin plus 2.21 mM EDTA in HBSS without sodium bicarbonate, calcium, and magnesium (Mediatech)). Cells were washed and then stained with 5 µg/ml 1B8 TCRm in PBS, 0.5% FBS, and 2 mM EDTA (staining/wash buffer), and the bound TCRm was detected by subsequent incubation with PE-labeled goat anti-mouse IgG. FACS analysis was performed on a FACScan (BD Biosciences). The results from flow cytometric studies are expressed either as mean fluorescence intensity (MFI) in histogram plots or as the mean fluorescence intensity ratio (MFIR), which is the ratio between the MFI of cells stained with the selected mAb and the MFI of cells stained with the isotype-matched mouse Ig. Generation of MFRI values normalizes background staining between the cell lines.
Tetramer competition assays
MDA-MB-231 cells (5 x 105) were incubated (30 min at room temperature) in 100 µl of staining/wash buffer (PBS plus 1% FBS and 2 mM EDTA) containing 0.5 µg/ml 1B8 TCRm in the presence of a competitor or noncompetitor tetramer or without tetramer addition. The competitive Her2(369)-HLA-A2 tetramer was tested using 0.1 µg or 1.0 µg, and the noncompetitive TMT(40)-HLA-A2 tetramer using 1.0 µg. After staining, the reactions were washed once and resuspended in 100 µl of wash buffer containing 0.5 µg of PE-conjugated goat anti-mouse IgG. Cells were washed as described previously and resuspended in 0.5 ml of wash buffer for characterization on a FACScan. Subsequent analysis was performed using CellQuest software.
Treatment with cytokines
IFN- and TNF- (R&D Systems) were added to culture medium at 20 ng/ml and 3 ng/ml, respectively, for 24 h before flow cytometric staining and lysis assays.
CTL line
The CTL line was generated as described by Lustgarten et al. (16). The Her2(369)-specific CTL line was maintained in vitro by weekly restimulation. Briefly, CTLs (1 x 106) were restimulated in 2 ml cultures with 0.2 x 106 irradiated Jurkat-A2.1 cells (20,000 rad) that were preincubated with Her-2/neu peptide (15 µM). Irradiated (3000 rad) C57BL/6 spleen cells (5 x 105) were added as fillers. Restimulation medium was complete RPMI 1640 containing 2% (v/v) supernatant from concanavalin-A-stimulated rat spleen cells.
Cytotoxic assays
T2 cells pulsed with peptides and tumor cells (MDA-MB-231, Saos-2, MCF-7, SW620, and COLO205) were incubated with 150 µCi of 51Cr-labeled sodium chromate for 1 h at 37°C. Cells were washed three times and resuspended in complete RPMI 1640 medium. For the cytotoxicity assay, 51Cr-labeled target cells (104) were incubated at a 10:1 CTL/target ratio in a final volume of 200 µl in U-bottom 96-well microtiter plates. Previous studies have shown optimal killing at a 10:1 CTL/tumor cell ratio (16). Supernatants were recovered after 4–7 h of incubation. The percent specific lysis was determined by the following formula: percentage of specific lysis = 100 x [(experimental release – spontaneous release)/(maximum release – spontaneous release)].
Anti-Her2(369)-A2 (1B8) and anti-A2.1 mAb (BB7.2) (25) were added to the assay to determine that the CTL lysis was specific for the HLA-A2.1/Her2/neu peptide (369–377) complex and A2.1 restricted, respectively. Before the addition of the effector cells, tumor cells were incubated in the presence or absence of 0.5 µg/ml 1B8, BB7.2, or murine IgG1 and IgG2b isotype control Abs.
Statistical analysis
The relationship between two parameters was tested using regression analysis, and p 
0.05 was considered significant. In the presence of a significant relationship, the coefficient of determination (R2) was calculated to express the degree of correlation. We used linear regression analysis and the staining and cytotoxic data from MDA-MB-231, Saos-2, MCF-7, SW620, and Colo205 tumor cell lines to determine significance.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
We generated the 1B8 TCRm mAb by immunizing mice with soluble recombinant HLA-A*0201 loaded with the Her2/neu(369) peptide epitope. The soluble H chains of HLA-A*0201 (hereafter designated A2+) and the 2-microglobulin were produced in the form of inclusion bodies in E. coli, purified, and then refolded in the presence of the Her2 KIFGSLAFL peptide. The conformation of the refolded protein was assessed using an anti-HLA class I Ab (W632) and the anti-HLA-A2-specific mAb BB7.2 (data not shown). The refolded protein served as the immunogen and as the positive control in screening assays of hybridoma supernatants. The eIF4G(720), TMT(40), and VLQ(44) peptide-loaded A2+ molecules served as negative controls. Over 2000 hybridomas were screened, and the 1B8 TCRm hybridoma was selected because it specifically recognized the recombinant HLA-A2 protein loaded with the p369 (aa 369–377 from He2/neu) peptide but did not bind recombinant HLA-A2 proteins loaded with irrelevant peptides (Fig. 1A). As a control for specificity, we used the 3F9 TCRm mAb, which is specific for the TMT(40) peptide-HLA-A2 complex. As shown in Fig. 1B, the 3F9 TCRm mAb binds specifically to the TMT(40)-A2 complex without binding to the Her2(369)-A2 complex. To demonstrate that recombinant HLA-A2 proteins were properly folded after being loaded with the peptide, they were stained with the BB7.2 anti-A2.1 mAb (Fig. 1C). These data demonstrate that the TCRm Abs recognize a specific MHC-peptide complex and that they do not have detectable cross-reactivity with either A2+ molecules or HLA complexes loaded with irrelevant peptides.
|
Although 1B8 TCRm recognizes the recombinant Her2(369)-A2 complex target in coated wells, it was unclear whether this mAb would recognize the specific peptide when loaded into HLA-A*0201 complexes expressed on a cell surface. To ensure that 1B8 recognized the Her2(369) peptide in the context of the native HLA-A2, we analyzed its binding to T2 cells pulsed with 10 µM p369 peptide, irrelevant peptides (TMT and eIF4G),or no peptide. As shown in Fig. 2A, 1B8 TCRm only stains T2 cells pulsed with the Her2/neu peptide but does not bind T2 cells not pulsed or pulsed with irrelevant peptides. These results confirm the fine and unique specificity of the 1B8 TCRm for the Her2/neu(369) peptide present in the binding pocket of the HLA-A2 complex.
|
Detection of endogenously processed and presented Her2/neu peptide by 1B8 TCRm on human tumor cell lines
We observed that the 1B8 TCRm mAb recognizes recombinant HLA-A2 protein or T2 cells pulsed with the p369 peptide. We next evaluated whether this Ab would recognize the Her2(369)-A2 complex presented by tumor cells using five HLA-A2+/neu+ cell lines, MDA-MB-231, Saos-2, MCF-7, SW620 and COLO205. We have previously demonstrated that the p369 epitope is processed and presented in MDA-MB-231 and MCF-7 breast carcinoma cells (16). As negative controls we used the HLA-A2–/neu+ cell lines BT-20 and SKOV3. In the first series of experiments, cells were stained with 0.5 µg of IgG1 isotype control mAb and 3F9 or 1B8 TCRm mAbs and, as shown in Fig. 3, all tumor cells except the BT-20 and SKOV3 cells (Fig. 3A) were stained with the 1B8 TCRm mAb (thick gray line). In contrast, only human chorionic gonadotropin-expressing cells, COLO205 cells, were weakly positive when stained with 3F9 TCRm mAb (solid black line). In the second series of experiments, the cell lines were pretreated overnight with IFN- and TNF- and then stained with the same panel of Abs used in Fig. 3A. As shown in Fig. 3B, the same five cell lines were stained with the 1B8 TCR mAb. In addition, with the exception of Saos-2, four cell lines showed enhanced staining with 1B8, suggesting an increase in levels of the Her2(369)-A2 complex. No staining was detected on SKOV3 cells, and a low background signal was detected on BT-20 cells (Fig. 3B). These results indicate that TCRm mAb can be used in the validation of epitopes that are endogenously processed and presented on the surface of tumor cells.
|
To further demonstrate that the 1B8 TCRm mAb binds specifically to endogenously processed Her2(369)-A2 complex on human tumor cells, the Ab was evaluated in two different competition assays. In the first system, HLA-A2 tetramer complexes were loaded with either a Her-2/neu peptide that would compete with specific binding to Her2(369)-A2 or an irrelevant TMT(40) peptide that would not compete for binding sites and then added to the staining reactions. MDA-MB-231 tumor cells were stained with 0.5 µg of 1B8 in the presence of the Her2(369)-A2 tetramer or TMT(40)-A2 tetramer complex. The results, shown in Fig. 4A, reveal that 1B8 TCRm mAb binding was reduced by >50% in the presence of 0.1 µg of the Her2(369)-A2-tetramer and was completely blocked by 1.0 µg of the Her2(369)-A2-tetramer. In contrast, when TMT(40)-A2 tetramer was added (1.0 µg) there was no inhibition of 1B8 TCRm mAb staining.
|
(20 ng/ml) plus TNF- (3 ng/ml) by adding 1B8 TCRm (anti-Her2(369)-A2) or BB7.2 (anti-HLA 2.1) mAb (Fig. 4C). In contrast, isotype control Abs (IgG1 and IgG2b) did not inhibit the CTL activity (Fig. 4C). Collectively, these data illustrate that the 1B8 TCRm mAb can specifically recognize the Her2(369)-A2 immunodominant epitope on the surface of tumor cells. Her2/neu Ag and HLA-A2.1 expression does not correlate with levels of peptide-HLA-A2 complex on tumor cell lines or tumor cell lysis
Expression of peptide-HLA class I on the cell surface depends on multiple parameters, including the quantity and quality of the peptide supplied. The supply of peptide is also dependent on the availability of protein and the rate at which the protein is processed. It is not clear, however, whether tumor Ag expression and MHC expression are directly linked with the level of expression of MHC-peptide complexes. We assessed the expression of Her-2/neu molecules, HLA-A2.1 molecules, and Her2(369)-A2 complexes on the surface of different tumor cell lines. Tumor cell lines were stained for Her-2/neu, and the expression of this Ag was variable among the cell lines (Fig. 5). For example, the COLO205 cell line revealed noticeably higher levels of Her2/neu protein than the MDA-MB-231, Saos-2, MCF-7, and SW620 tumor cell lines. The BT-20 (HLA-A2–) cell line had an intermediate level of Her2/neu protein expression. Detection of Her2/neu protein expression by two different methods revealed that the level of cell surface expression directly correlates (p < 0.05) with the cellular level of Her2 protein expression (R2 = 0.82) as evaluated by ELISA (Fig. 5, A and B).
|
|
To further examine the relationship between the levels of MHC-peptide complexes present on the cell surface and the levels of Ag and MHC molecules expressed, we pretreated the cell lines with IFN- (20 ng/ml) plus TNF- (3 ng/ml). Treating tumor cells in this way is known to increase the expression of adhesion molecules (e.g., ICAM) and MHC class I H chain (26). These cytokines also enhance protein processing and peptide presentation by HLA class I through the activation of the immunoproteasome (27), which has been hypothesized to cause an increase in the expression of specific MHC-peptide complexes, especially in cells with greater availability of Ag. We tested this hypothesis by treating the tumor cell lines for 24 h with cytokines and then staining with the BB7.2 mAb (Fig. 7A) and the 1B8 TCR mimic (Fig. 7B). We observed that, after cytokine treatment, all tumor cell lines except Saos-2 displayed greater 1B8 TCRm staining intensity (see also Fig. 3B), indicating that more of the specific complex was expressed on the cell surface. When comparing cell surface levels of the Her2(369)-A2 complexes between the differently treated cell lines, we found that the 1B8 staining intensity for COLO205 (MFIR = 9.5) was markedly lower than that of MDA-MB-231 (MFIR = 38) and MCF-7 (MFIR = 27). This observation suggests that stimulation of cellular machinery for Ag processing and presentation did not favor higher levels of specific HLA-peptide complex in cells that, as demonstrated previously (Fig. 5A), expressed significantly more of the tumor Ag. Validation of cytokine-induced effects on the MHC class I system was demonstrated by the increase observed in HLA-A2 expression (Fig. 7A). Interestingly, in this group of cell lines the surface levels of HLA-A2 were equivalent in all but MCF-7 cells, which had noticeably lower HLA-A2 expression. We conclude from these data that TAA expression does not correlate with levels of specific MHC-peptide complexes.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
The Her2/neu receptor is considered a good target for Ab-mediated immunotherapeutic approaches because it is expressed at high levels in many tumors but at low levels in normal cells (28). Within the Her2/neu protein there are multiple HLA-A2 binding peptide epitopes that can induce CTL responses against autologous tumor cells (15, 29). Conventional wisdom would suggest that the overexpression of Her2/neu would lead to higher protein turnover, increase the supply of Her2/neu peptides, and enhance the levels of Her2/neu-specific peptides that are presented by HLA class I molecules. We tested this assumption by visualizing levels of the dominant Her2(369)-A2 epitope on the surface of tumor cell lines using the TCRm mAb 1B8. Except for the MDA-MB-231 cells, which were stained by 1B8 TCRm with significantly greater intensity, all cell lines exhibited similar MFIR values even though COLO205 cells expressed almost 5-fold higher levels of Her2/neu Ag. Our findings suggest that Her2/neu protein overexpression does not guarantee that endogenous HLA-A2 will present more of the dominant Her2(369) peptide. In fact, a possible relationship has been reported between Her2 Ag expression and the function of MHC class I Ag processing and presentation machinery components. Herrmann et al. (8) revealed that the Her2/neu protein regulated MHC expression in transfected cells by modulation of components of the Ag-processing pathway. In particular, they found that a transfectant overexpressing Her2/neu significantly down-regulated TAP1, TAP2, PA28, LMP2, LMP7, LMP10, and tapasin mRNA and/or protein levels when compared with Her2/neu– transfectants. In contrast, transcription of the constitutive proteasome subunits such as the chaperones calnexin, calreticulin, and ER60, the protein disulfide isomerase, and the MHC class I H chain and -2-microglobulin chain was not influenced by Her2/neu transformation (8).
In this study, we attempted to counter the potential inhibitory effects mediated by Her2/neu protein on MHC class I expression and components of the Ag-processing pathway by pretreating cells with the cytokines IFN- and TNF-. As a proinflammatory cytokine and potent regulator of cell-mediated immune responses, IFN- acts by up-regulating a large number of genes including HLA class I (30, 31), endoplasmic reticulum peptide transporters (e.g., TAP1,2) (32, 33), proteasome subunits (e.g., LMP2,7,10) (34), and proteasome regulators (e.g., PA28), all of which contribute to Ag processing and presentation. In general, IFN- treatment is believed to enhance the presentation of peptides in the context of HLA class I molecules on the surface of target cells, leading to more efficient recognition by CTL. In our study we showed that all five tumor lines expressed at least 2-fold more HLA-A2 after cytokine treatment. Further, we confirmed that four of the five cell lines in our study demonstrated a significant increase in staining with the Abs 1B8 TCRm and BB7.2, thus indicating that cytokine treatment was effective in enhancing Ag processing and presentation and HLA-A2 expression. However, cytokine treatment alone did not reverse the poor 1B8 TCRm staining profile for the COLO205 tumor cell line that overexpressed Her2/neu Ag. The levels of the Her2(369)-A2 epitope on COLO205 (MFIR = 9.5) cells were significantly lower than those reported for MDA-MB-231 (MFIR = 37.5) and MCF-7 (MFIR = 27) cells, even though COLO205 cells have up to 5-fold more Her2 Ag. Our findings indicate involvement of other factors independent of the MHC class I processing and presentation pathways that control Her2/neu degradation. For example, stability of membrane associated Her2/neu is provided, at least in part, by the chaperones Grp94 and HSP70 (35). Previous studies showed that SKOV3-A2 cells treated with geldanamycin increased expression of HLA-A2 molecules and promptly degraded Her2/neu by the proteasome pathway, leading to enhanced lysis by Her2(369)-A2-specific CTL (36, 37). We used this drug to address whether drug-treated SKOV3-A2 cells demonstrate an increase in 1B8 staining. In our study, drug-treated cells showed a 2-fold increase in HLA-A2 surface expression and a 15% increase in 1B8 staining (data not shown), suggesting that factors such as chaperones may provide protection from the proteasomal degradation pathway and that dissociation of these complexes may lead to enhanced lysis by CTLs. Other studies have shown that the internalization of Her2/neu mediated by trastuzumab can enhance tumor cell sensitivity to CTL-mediated lysis, and it is hypothesized that this is due to an increase in Her2/neu peptide-HLA-A2 complexes on the cell surface (28). This theory can now be tested directly using the 1B8 TCRm to assess the presentation of the dominant Her2(369)-A2 epitope.
The number of peptide-HLA complexes required for a T lymphocyte response is not definitively known (38, 39). Direct visualization of CTL epitope presentation on tumor cells and its relationship to CTL-mediated tumor cell lysis was a major finding of this study. The current methods in use for CTL epitope validation rely on peptides that are acid eluted from tumor cells, enriched, concentrated, and then tested by pulsing cells and performing 51Cr-release assays (40). By using this approach we showed previously that the MCF-7 and MDA-MB-231 tumor cell lines present the Her2(369) in the context of HLA-A2 (16). However, this approach is neither direct nor quantitative and cannot be used to effectively examine the relationship between CTL epitope level and tumor lysis by specific CTLs. In this study we directly compared the intensity of 1B8 staining to tumor cell lysis by using a high avidity CTL line generated in HLA-A2.1 transgenic mice (16). We stained tumor cell lines that were either treated or not treated with cytokine and found a trend between 1B8 staining intensity and tumor cell lysis. The relationship between CTL epitope presentation and tumor cell lysis directly correlated (p = 0.05) with cell lines pretreated with cytokines (R2 = 0.75). Although we have shown a good correlation between tumor cell killing by CTL and Her2(369)-A2 expression, it cannot be ignored that other variables (e.g., tumor-mediated suppression) play a role in determining tumor cell lysis susceptibility by CTL. An example of this can be seen with the pretreated tumor cell lines. Both Saos-2 and SW620 cells showed similar Her2(369)-A2 levels after staining with 1B8 TCRm but observed almost 4-fold greater tumor cell lysis of Saos-2 cells than of SW620 cells.
The 1B8 TCRm mAb, with its unique TCR-like binding specificity and an affinity that allows detection of physiological levels of peptide-HLA complexes on tumor cell surfaces, is an excellent example of a growing class of powerful reagents that can be used to measure HLA class I-tumor-associated peptide targets (9, 11, 41). The ability to directly validate or confirm specific peptide-HLA epitopes on tumor cells could widely impact vaccine development. Still, these tools can also prove useful in understanding the mechanisms that regulate tumor Ag processing and presentation through their ability to detect and quantify the number of specific complexes. Several other groups have reported on Abs with similar specificity for peptide-MHC targets (9, 11, 42, 43). Recently, the groups of Hoogenboom and Reiter (13, 14, 44, 45, 46) have created Ab libraries using phage display to isolate single-chain variable fragments that specifically recognize HLA class I molecules presenting peptide fragments from several TAAs, including telomerase, gp100, and MAGE-1. It would be interesting to examine the specific CTL epitope levels by using these single-chain variable fragment Abs with tumor cell lysis for comparison with the findings reported in this study for Her2/neu. The value of TCRms for vaccine discovery, validation of targets, and characterization of vaccines in potency assays is obvious. These tools will also be applicable for investigation of factors that affect Ag processing and MHC class I presentation in health and disease.
In conclusion, we have demonstrated the fine specificity of a TCRm and its ability to discriminate between peptides in the context of the HLA-A2 molecule and, thus, the utility of TCRms for directly detecting peptide-HLA complexes on tumor cells. Our studies included five HLA-A2+ tumor cell lines and two HLA-A2– tumor cell lines. We should, of course, be cautious and not extrapolate beyond the Her2/neu case to other overexpressed TAAs and primary/metastatic tumors. However, our finding that Her2/neu receptor expression does not correlate with the density of Her2/neu peptide-HLA-A2 complexes and does not accurately predict tumor cell susceptibility to lysis by CTLs, raises interesting questions about the assumptions currently used for the selection of future CTL epitopes for the development of cancer vaccines. We offer an alternative for selection of CTL epitopes that makes use of TCRm to directly visualize tumor-associated peptide presented in the context of HLA class I molecules.
|
Acknowledgments |
|---|
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 This work was supported by the Advanced Technology Program National Institute of Standards and Technology Grant 70NANB4H3048. 
2 Address correspondence and reprint requests to Dr. Jon A. Weidanz, Department of Pharmaceutical Sciences, School of Pharmacy, Texas Tech University Health Sciences Center, 1300 Coulter Drive, Amarillo, TX 79106. E-mail address: jon.weidanz{at}ttuhsc.edu
3 Abbreviations used in this paper: TAA, tumor-associated Ag; p369, aa 369–377 from He2/neu; TCRm, TCR mimic; MFI, mean fluorescence intensity; MFIR, MFI ratio.
Received for publication June 22, 2006. Accepted for publication August 1, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
2-microglobulin in Escherichia coli and its purification. Gene 83: 117-124. [Medline] and interleukin-1 on major histocompatibility complex antigen and intercellular adhesion molecule-1 expression by 9L gliosarcoma: relevance to its cytolysis by alloreactive cytotoxic T lymphocytes. J. Interferon Cytokine Res. 22: 1209-1216. [Medline]. J. Exp. Med. 183: 1807-1816. differentially regulates susceptibility of lung cancer cells to telomerase-specific cytotoxic T lymphocytes. Int. J. Cancer 110: 403-412. [Medline]This article has been cited by other articles:
|
|
 |
|
  H. Conrad, K. Gebhard, H. Kronig, J. Neudorfer, D. H. Busch, C. Peschel, and H. Bernhard CTLs Directed against HER2 Specifically Cross-React with HER3 and HER4 J. Immunol., June 15, 2008; 180(12): 8135 - 8145. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. G. Schuster, D. H. Busch, E. Eppinger, E. Kremmer, S. Milosevic, C. Hennard, C. Kuttler, J. W. Ellwart, B. Frankenberger, E. Nossner, et al. Allorestricted T cells with specificity for the FMNL1-derived peptide PP2 have potent antitumor activity against hematologic and other malignancies Blood, October 15, 2007; 110(8): 2931 - 2939. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Purbhoo, Y. Li, D. H. Sutton, J. E. Brewer, E. Gostick, G. Bossi, B. Laugel, R. Moysey, E. Baston, N. Liddy, et al. The HLA A*0201-restricted hTERT540-548 peptide is not detected on tumor cells by a CTL clone or a high-affinity T-cell receptor Mol. Cancer Ther., July 1, 2007; 6(7): 2081 - 2091. [Abstract] [Full Text] [PDF]
|
|
| ||
