key: cord-0001859-fvoke2yc authors: TANG, BAISHAN; ZHOU, WEIJUN; DU, JINGWEN; HE, YANJIE; LI, YUHUA title: Identification of human leukemia antigen A*0201-restricted epitopes derived from epidermal growth factor pathway substrate number 8 date: 2015-04-23 journal: Mol Med Rep DOI: 10.3892/mmr.2015.3673 sha: 4b2ef217cb3fedccaffa0ecdd43344810630987b doc_id: 1859 cord_uid: fvoke2yc T-cell-mediated immunotherapy of hematological malignancies requires selection of targeted tumor-associated antigens and T-cell epitopes contained in these tumor proteins. Epidermal growth factor receptor pathway substrate 8 (EPS8), whose function is pivotal for tumor proliferation, progression and metastasis, has been found to be overexpressed in most human tumor types, while its expression in normal tissue is low. The aim of the present study was to identify human leukemia antigen (HLA)-A*0201-restricted epitopes of EPS8 by using a reverse immunology approach. To achieve this, computer algorithms were used to predict HLA-A*0201 molecular binding, proteasome cleavage patterns as well as translocation of transporters associated with antigen processing. Candidate peptides were experimentally validated by T2 binding affinity assay and brefeldin-A decay assay. The functional avidity of peptide-specific cytotoxic T lymphocytes (CTLs) induced from peripheral blood mononuclear cells of healthy volunteers were evaluated by using an enzyme-linked immunosorbent spot assay and a cytotoxicity assay. Four peptides, designated as P455, P92, P276 and P360, had high affinity and stability of binding towards the HLA-A*0201 molecule, and specific CTLs induced by them significantly responded to the corresponding peptides and secreted IFN-γ. At the same time, the CTLs were able to specifically lyse EPS8-expressing cell lines in an HLA-A*0201-restricted manner. The present study demon-strated that P455, P92, P276 and P360 were CTL epitopes of EPS8, and were able to be used for epitope-defined adoptive T-cell transfer and multi-epitope-based vaccine design. Even though chemotherapy, radiotherapy and hematopoietic stem cell transplantation (HSCT) have been used to treat hematological malignancies, several of these are still considered incurable. Harnessing the host immune system towards the targeting of cancer cells for their destruction has been shown to be effective (1) , and therapeutic vaccines against leukemia-associated antigens (LAAs) are among the approaches to trigger tumor-specific immune responses. Identification of LAA and its cytotoxic T lymphocytes (CTL) epitopes are of major importance for the development of tumor antigen-specific therapeutic and prophylactic vaccines. A number of T-cell epitopes from LAAs have been identified, and peptide vaccines targeting LAAs based on their T-cell epitopes have been shown to be effective in clinical trials (2) . Epidermal growth factor receptor pathway substrate 8 (Eps8) contains 822 amino acids and is located in chromosome 12p12-p13 in humans. The EPS8 gene encoded for the expression of proteins with molecular weights of 97 and 68 kDa, referred to as the two EPS8 isoforms. As a signaling adapter, EPS8 regulates actin cytoskeleton dynamics and architecture (3) and participates in transduction of signals from Ras to Rac by activating the Rac-specific guanine nucleotide exchange factor activity in the form of a trimeric complex together with SOS1 and ABI1 (4) . It involves in the regulation of processes including dendritic cell migration as well as cancer cell migration and invasion (5) . Strong evidence demonstrated that EPS8 had oncogenic potential, and its overexpression was reported in a range of human malignancies, including breast cancer (6) , pituitary tumors (7) , pancreatic cancer (8) , cervical cancer (9) , colon cancer (10), head and neck squamous cell carcinoma (11) , esophageal cancers (12) and human gliomas (13) , and its amplification is often associated with tumor progression, acquired drug resistance and poor prognosis (14) . EPS8 upregulation is thus expected to be involved in human carcinogenesis, implying that this gene may be used as a target in tumor immunotherapy. A previous study by our group demonstrated an elevated expression of EPS8 in patients with acute myeloid leukemia (AML), its overexpression was inversely correlated with overall survival of patients (15) , and EPS8 protein as a vaccine reagent induced a CTL response in a murine breast carcinoma model (16) . From the aforementioned facts, it was speculated that this gene may be a prognostic marker and could be a target for immunotherapy of hematological malignancies. To the best of our knowledge, no T-cell epitopes of EPS8 have been reported to date. The aim of the present study was to identify human leukemia antigen (HLA)-A*0201-restricted epitopes for EPS8. To achieve this aim, the expression of EPS8 in a range of tumor cell lines was detected using western blot analysis and reverse transcription quantitative polymerase chain reaction (RT-qPCR), and their phenotypes were detected using a direct immunofluorescence method. Furthermore, screening for the EPS8 protein sequence was performed using various algorithms to predict the peptide-binding ability to the HLA-A*0201 molecule and its proteasome cleavage sites at the C-terminus to identify T-cell epitopes. The findings were validated in vitro by peptide binding affinity assay and brefeldin-A (BFA) decay assay. The functional avidity of candidate peptide-specific CTLs was evaluated using an enzyme-linked immunosorbent spot (ELISPOT) assay and a cytotoxicity assay. Four peptides which were CTL epitopes of EPS8 were identified, and which may be used in vaccine design and tumor immunotherapy. Cell lines. A lymphoblast cell line, designated as T2 (174xCEM. T2), was purchased from the American Type Culture Collection (cat. no. CRL-1992™; Manassas, VA, USA). This cell line is transporter associated with antigen processing (TAP)-deficient and expresses HLA-A2. In the absence of exogenous antigen peptide load, major histocompatibility complex (MHC) class I expression levels on its surface are very low due to the poor stability of non-peptide-loaded MHC class I molecules. Epitope prediction. To identify the potential T-cell epitopes for EPS8, the HLA-A*0201 allele was screened, covering a wide range of populations (17) . First, the protein sequence was screened for the best binding epitopes using the algorithms SYFPEITHI (http://www.syfpeithi.de/) and Bioinformatics and Molecular Analysis Section (BIMAS; http://www-bimas. cit.nih.gov/). The cut-off score was adjusted to >20 for SYFPEITHI and T 1/2 >100 for BIMAS, and peptides matching these two criteria and which were shared by the two algorithms were selected. Second, the NetChop algorithm (http://www. cbs.dtu.dk/services/NetChop/) was used to predict whether the selected peptides in the first step would be cleaved at the C-terminus. Third, protein sequences were analyzed with five different algorithms for HLA-A*0201, including immune epitope database (IEDB; http://www.iedb.org/), NetMHC (http://www.cbs.dtu.dk/services/NetMHC/), SVMHC (http://www.sbc.su.se/~pierre/svmhc/), EpiJen (http://www. ddg-pharmfac.net/epijen/EpiJen/EpiJen.htm) and Rankpep (http://imed.med.ucm.es/Tools/rankpep.html). Peptides predicted by at least four algorithms were selected. The known HLA-A*0201-restricted carcinoembryonic antigen peptide-1 (CAP-1, YLSGANLNL) was used as a positive control peptide (18) , while the HBcAg-derived HLA-A*2402-restricted peptide HBc117e125 (EYLVSFGVW) was used as a negative control peptide (19) , and all peptides were synthesized using standard 9-fluorenylmethyl-oxycarbonyl chemistry by the Chinese Peptide Company (Hangzhou, China). They were of >95% purity as determined by reverse-phase chromatography and their identity was confirmed by mass spectrometry. The lyophilized peptide powder was dissolved according to the manufacturer's instructions in ultrapure water or in dimethylsulfoxide (BioLegend), diluted in phosphate-buffered saline (PBS; Invitrogen Life Technologies) to a final concentration of 1 mg/ml and stored in aliquots at -80˚C until use. Peptide binding affinity assay. To evaluate the binding affinity of each candidate peptide to the HLA-A*0201 molecule, a classic T2 peptide-binding assay was performed as described by Dervillez et al (20) with certain modifications. T2 cells are TAP-deficient and HLA-A*0201-positive, but due to the poor stability of non-peptide-loaded HLA-A*0201 molecules, its HLA-A*0201 expression levels on the surface are low. Exogenous peptides are able to induce the accumulation of HLA-A*0201 molecules, and thus, upregulation of HLA-A*0201 molecules on T2 cells may be detected by florescent intensity exchange which reflects peptide binding ability to HLA-A*0201 molecules. Briefly, T2 cells were incubated overnight with peptides (final concentration, 100 µM) in IMDM serum-free medium containing human β2-microglobulin (final concentration, 3 µg/ml) in a humidified 26˚C incubator with 5% CO 2 . Following 18 h, the temperature was raised to 37˚C for 2 h. Cells were harvested by gently transferring into a sterile centrifuge tube and centrifuged at 200 x g for 5 min. Subsequently, cells were washed twice, firstly with serum free IMDM and then with cell staining buffer (BioLegend; cat. no 420201). The cells were then stained directly with anti-HLA-A*0201 monoclonal antibody conjugated to FITC for 30 min at 4˚C, and then analyzed using an Elite flow cytometer (Beckman Coulter, Miami, FL, USA). Three duplicate wells were set to each group and experiments were repeated three times; CAP-1 was used as a positive control, HBc117e125 was used as negative control, and T2 cells without any added peptide was used as a background control. The fluorescence index (FI) was calculated using the following formula: FI = [mean fluorescence intensity (MFI) sample -MFI background ] / MFI backgro und , where MFI background represents the value without peptide. FI>1.5 indicated that the peptide had a high affinity for HLA-A*0201 molecules, 1.08 h, followed by peptide 276 with DC 50 <8 h, while the stability of peptide 92/MHC complex was lowest with a DC 50 of 2-4 h (Table I) . EPS8 expression detection and phenotypic analysis of target cell lines. To identify candidate target cells, RT-qPCR was used to detect the mRNA expression of EPS8; furthermore, the protein expression of EPS8 was detected by western blot analysis. Among the eight cell lines, MCF-7, SW480, KG1a and K562 were EPS8-positive at the mRNA expression level ( Fig. 2A) and the protein expression level (Fig. 2B) . The HLA-A phenotype of the target cells was detected by direct immunofluorescence antibody staining and flow cytometric analysis. Of the eight cell lines, only MCF-7 and T2 were HLA-A*0201-positive ( Fig. 3A and B) . To determine the immunogenicity of peptides, PBMCs were stimulated in a 10-day procedure as described in the Materials and methods section, followed by ELISPOT analysis. PBMCs separated from two HLA-A*0201-positive healthy volunteers were repeatedly stimulated with peptides, and the IFN-γ secretion of the stimulated PBMCs in response to the peptides was assessed for the two donors (Fig. 4) . The PBMCs of the two donors responded positively to stimulation with all of the four predicted peptides, as indicated by an increased number of SFCs following stimulation. Of note, PBMCs responded strongly to stimulation with the relatively weak binder peptide 92 (218±30.4 peptide-specific spots/2x10 5 cells), but elicited a response of lower intensity to strong binder peptide 455 (73.0±14.2 spots/2x10 5 cells) (Fig. 4C) (P<0.001) . To characterize the functional avidity of peptide-specific CTLs, an LDH release assay was performed to evaluate their cytotoxic effects on MCF-7 and T2 cells pulsed with cognate peptides. As shown in Fig. 5 , the four peptide-specific CTLs efficiently lysed MCF-7 as well as T2 cells pulsed with corresponding peptides but could almost not lyse T2 cells pulsed with irrelevant peptide, T2 only, K562 or THP-1 cells at an effector to target cell ratio of 100:1 (P<0.05). Of them, peptide 92-specific CTLs had the highest lysis rates of MCF-7 and T2 cells pulsed with the corresponding peptide, and the average lysis rate of MCF-7 cells was 51.28±1.54% (P=0.001; donor 1) (Fig. 5A ) and 54.22±3.01% (P=0.001; donor 2) (Fig. 5B) , respectively, while the lysis rate of T2 cells loaded with the corresponding peptide was 45.41±4.16% (P=0.023; donor 1) and 35.23±1.50%, (P=0.001; donor 2), respectively. The cytotoxic effects of peptide-induced CTLs on target cells were EPS8 expression-specific and restricted by the HLA-A*0201 molecule as the effectors only target HLA-A*0201-positive cells. Concepts for treating hematologic malignancies by harnessing the immune system have been established due to defects of current treatment methods (27) . HSCT is the only potential method to cure hematological malignancies, but relapse rates remain high, particularly in patients transplanted during relapse or in second or subsequent remission (28) . Therefore, efforts are required to seek for novel avenues which are tumor-specific and have synergy with current treatment methods; from this viewpoint, therapeutic vaccines against LAAs are an optimal choice. Mounting evidence demonstrated the existence of LAAs (29) ; as most LAAs are not muted, their immunogenicity is often poor; however, at the same time, the immunity evoked by current therapeutic vaccines targeting LAAs is far from optimal. Accordingly, isolating specific components that have immunogenicity and can induce a strong immune response are rational choices for vaccines. Therefore, the most important task in the discovery of cancer vaccines is to identify novel TAAs and their CTL epitopes in order to make additional therapeutic targets available which are closely associated with tumor incidence and progression. Reverse immunology has been commonly used to identify epitopes of LAAs (30) . The present study was the first, to the best of our knowledge, to report EPS8 as a potential TAA and identify its HLA-A*0201-restricted epitopes. EPS8 was reported to be expressed at elevated levels in numerous human malignancies and its overexpression has been reported to be sufficient to transform non-tumorigenic human cells into a tumorigenic phenotype (31) . Its expression in normal tissue types is relatively low, and it has a pivotal role in tumorigenesis, tumor proliferation, invasion and metastasis; furthermore, the expression levels of EPS8 correlated with disease severity and overall survival of patients (32) . Thus, EPS8 is a TAA and may be utilized as a target in immunotherapy of cancer as well as in hematological malignancies. In the present study, a reverse immunology method was used to identify HLA-A*0201-restricted epitopes from EPS8. In the prediction phase, to appropriately select the T-cell epitopes contained in this protein, the focus was on three main events of the intracellular generation of peptides presented in HLA class I molecules; these HLA class I ligands are called CTL epitopes only when immunogenic (33) . The first event is enzymatic digestion of protein by cytosolic peptidases, leading to the release of the epitope or epitope precursor in the protein. The second important event is required to prevent peptides from further cytosolic degradation via the TAP, only peptides that escape from cytosolic degradation and finally enter into endoplasmic reticulum can bind with HLA class I molecules. The third event is the assembly of peptides with HLA class I heavy and light chains in the ER for cell surface presentation (34) (35) (36) (37) (38) (39) . Peptides processed and presented through these three steps have the highest potential to be CTL epitopes. In the final step of the in silico analysis, five different popular algorithms were used to further confirm the selected peptides, leading to the final selection of four 9-mer peptides. It is worth mentioning that the aim of the present study was to identify dominant T-cell epitopes. For this, a combined prediction using SYFPEITHI and BIMAS was utilized, resulting in only four peptides, which significantly saved time by narrowing the validation scope. However, this does not indicate that there are only four CTL epitopes within the EPS8 protein sequence; due to stringent restrictions arising from the selection criteria, several potential subdominant epitopes may have been ignored. BIMAS and SYFPEITHY are based on the principle of binding motifs/quantitative matrices (39) , and in spite of numerous novel algorithms that have emerged, they remain to be commonly used in current studies. The main reason for this may be the large dataset they comprise, which is important, as most predictions are dataset-driven -the bigger the dataset, the more accurate the prediction. One weakness of the motifs or quantitative matrices is the ignorance of the contribution of the overall peptide structure to binding. To overcome the limitations of motif-based predictions, more powerful methods based on tools including ANN, PSSM, SMM and SVMHC were used. It was reported that a sensitivity of 80% and a specificity of 80% can be achieved with ANN (40) . Together with proteasomal cleavages and TAP translocation prediction, it was speculated that the anchor residues of selected peptides could chemically complement the main pockets of the peptide-binding groove of the HLA molecule. The predicted peptides are not always true binders, and several false-positive peptides may be selected (41) ; therefore, experiments are required to validate their immunogenicity. Peptide affinity to HLA molecules is a key event, and a correlation between HLA binding and immunogenicity is often observed (42) . The eradication of tumor cells by T cells requires high-affinity targeted peptide-MHC interactions, which lead to efficient cross-presentation of antigens. Their binding affinity to the HLA-A*0201 molecule was evaluated using the TAP-deficient cell line T2. When exogenous peptides were added, the HLA-A*0201 molecule on the surface of T2 accumulated, resulting in a change in fluorescent intensity, which changes in the MFI reflecting their binding affinity to the HLA-A*0201 molecule. Of the four peptides, peptide 455 and 360 had the highest affinity to the HLA-A*0201 molecule, while peptide 276 and 92 had a relatively low binding affinity to the HLA-A*0201 molecule. This meant that peptide 455 and 360 were most likely to be CTL epitopes. However, certain peptides with high binding affinity to HLA class I molecules may be non-immunogenic (43) , as they fail to form stable complexes with HLA class I molecules (44) . Although peptide immunogenicity was found to correlate significantly with high affinity, the stability of peptide interaction with MHC-I correlated better with immunogenicity than affinity (45) . Following the identification of the CTL epitopes, the present study assessed the peptide/MHC complex stability by using the BFA decay assay. BFA inhibits protein synthesis in cultured cells and reversibly inhibits the intracellular translocation of proteins to the cell surface for secretion or expression (46, 47) ; therefore, the accumulated complex remained stable on the cells. When peptide-pulsed T2 cells were continuously cultured in the presence of BFA at 37˚C, their MFI on the cells decreased, the peptide/HLA-A*0201 molecule complex dissociated naturally, and DC 50 -values indicated that the 455 and 360/HLA-A*0201 molecule complexes were more stable than those of peptide 276 and peptide 92. This meant that peptide 455 and 360 may be more immunogenic than peptide 276 and peptide 92. High binding affinity and stable binding of peptides to MHC in vitro does not necessarily imply that these peptides can be naturally presented. To test the natural presentation of the candidate epitopes, the target cells were first screened by RT-qPCR and western blot to detect the EPS8 expression, and their HLA-A phenotype was detected by direct immune fluorescence antibody staining and flow cytometric analysis. Three tumor cell lines and T2 were selected as target cells. Most of epitope identification methods which have been used to date require the induction of peptide-specific CTLs, followed by their functional avidity evaluation; PBMCs separated from healthy donors are commonly used to evaluate the immunogenicity of a peptide (48) . Two HLA-A*0201-positive donors were identified among healthy volunteers using an external high-resolution method. Peptide-specific CTLs were induced by immunizing the PBMCs of two healthy HLA-A*0201-positive donors through at least two rounds of weekly stimulations with peptides and in the presence of IL-2 and IL-7, the number of peptide-specific CTLs was sufficiently increased. Following successful induction of specific CTLs, an IFN-γ ELISPOT assay was used to quantify the occurrence of T-lymphocyte cells secreting IFN-γ after stimulation with cognate peptide. The results demonstrated that the four peptide-specific CTLs all responded to the corresponding peptides and secreted IFN-γ; their effective SFCs met the criteria of the positive standard, with the number of SFCs of peptide 92 specific CTL being the highest. The results for the two donors were consistent, which meant that peptide 92 was more immunogenic than the other three peptides. Finally, the recognition efficiency of peptide-specific CTLs to target cells was evaluated using a non-radioactive cytotoxicity assay, and in the two volunteers, all peptide-specific CTLs could lyse MCF-7 (HLA-A*0201+, EPS8+) and T2 cells pulsed with the same peptides, but could almost not lyse either HLA-A*0201-negative or EPS8-negative cells at the effectors to targets ratio of 100:1 (P<0.05). This indicated that their cytotoxicity was EPS8 expression-specific and in an HLA-A*0201-restricted fashion, the immunogenicity of peptide 92 was the strongest. Together with the ELISPOT results, it was demonstrated that the four predicted epitopes were naturally presented, and that they were CTL epitopes of EPS8. In the present study it was observed that peptide binding affinity and peptide/HLA complex stability did not correlate well with their intensity of immunogenicity; for example, irrespective of its low affinity or complex stability with HLA-A*0201 molecule, the relatively weak binder peptide 92 stimulated PBMCs more strongly to secret the highest amount of IFN-γ than the strong binder peptide 455, which stimulated the least amount of IFN-γ secretion by the PBMCs. At the same time, compared to the mean lysis rates of CTLs induced by other peptides, the mean lysis rate of peptide 92-induced CTLs was the highest to MCF-7 and T2 pulsed with corresponding peptide. The reason of this may be based on the interaction dynamics of the T-cell receptor (TCR)-peptide/MHC complex, which is closely associated with T-cell activation and subsequently its differential fate (49) . There is a physiological affinity range, which is optimal to activate T-cells, and a plateau of T-cell functional activity above a defined affinity threshold has been reported, suggesting that all clustered TCRs will be occupied above the defined threshold, and further increases in affinity do not contribute to T-cell functions; therefore, above a defined threshold, TCRs with higher avidity may not be at an advantage over lower-avidity TCRs in the exertion of cytotoxic function (50) . Furthermore, negative feedback mechanisms, including programmed death 1 and lymphocyte activation gene-3 (CD223), would further curtail the potency of high-avidity TCRs (51) (52) (53) . The discrepancy observed in the present study may be ascribed to the peptide affinity of high-affinity binding peptides above the defined affinity threshold. Another reason may be the supra-optimal antigen dose. Evidence suggested that culturing of CTLs in vitro in the presence of high-or low-dose antigen leads to polarization of low-and high-avidity responses, respectively (54, 55) . In regard to low-or intermediate-affinity peptides, the peptide dose used for CTL induction in the present study may have been supra-optimal for high-affinity peptides, with the optimal TCR stimulation peptide dose for maintenance of high-avidity T cells ex vivo being 1 ng/ml and resulting in retained avidity, proliferation and ability to kill specific targets. By contrast, the supra-optimal TCR stimulation peptide dose (10 µg/ml peptide) for high-avidity T-cell maintenance ex vivo resulted in reduced avidity and failure to kill tumor cells (56) . The present study highlights the importance of optimal stimulation for the induction and maintenance of high-avidity CTL. Due to certain limitations, the present study did not further explore the best affinity to HLA molecules, which is in favor of peptide-specific CTL induction and maintenance. Furthermore, the experiments of the present study had additional drawbacks: First, due to limitations to the amount of blood samples, the optimal cytokine concentration and their dosing intervals for best induction and maintenance of peptide-specific CTL were not explored. Second, background signals of target cells that lacked tumor antigens or restriction were observed, implying that non-specific killing effects still existed. The reason for this may be ascribed to the use of bulk CTL instead of CD8+ T cell from the colon as effectors. Furthermore, CD8+ T cells were not sorted from PBMCs for CTL induction at the beginning and designated antigen-presenting cells such as dendritic cells (DCs) were not used; this may have resulted in inadequate antigen processing and presentation. Despite these drawbacks, from an economic and practical point of view, the methods used in the present study are an economical and rapid method for epitope validation. Furthermore, total PBMCs include a variety of cell types, including monocytes and DCs, which can be used as antigen-presenting cells (APC) (57) . It is worth mentioning that certain CTL epitopes identified in vitro may be less immunogenic in vivo, which may be due to inadequate antigen processing and presentation by APC (58) . Therefore, a subsequent study by our group will test the immunogenicity of the epitopes in vivo using transgene mice. In conclusion, the present study identified four HLA-A*0201-restricted, EPS8-derived epitopes, which are located at the starting positions of 92, 276, 360 and 455 in the EPS8 protein sequence. They had good binding affinity for the HLA-A*0201 molecule and their peptide/HLA-A*0201 complex stability was high. At the same time, these peptides promoted lymphocyte proliferation, and their specific CTLs responded to the corresponding peptides and secreted IFN-γ. They were also able to recognize and lyse ESP8-expressing target tumor cells in an HLA-A*0201-restricted manner, and they were naturally processed and presented on tumor cells. Among them, peptide 92 was the most promising as an antigenic epitope to target tumor cells due to its higher immunogenicity. Future studies will be required to investigate the clinical utility of the identified epitopes by using blood samples from patients with hematological malignancies. Targeted cancer immunotherapy Clinical peptide vaccination trials for leukemia patients The Eps8/IRSp53/VASP network differentially controls actin capping and bundling in filopodia formation Eps8 controls actin-based motility by capping the barbed ends of actin filaments SCFFbxw5 mediates transient degradation of actin remodeller Eps8 to allow proper mitotic progression Combined cDNA array comparative genomic hybridization and serial analysis of gene expression analysis of breast tumor progression Epidermal growth factor receptor pathway substrate 8 is overexpressed in human pituitary tumors: Role in proliferation and survival Eps8 is increased in pancreatic cancer and required for dynamic actin-based cell protrusions and intercellular cytoskeletal organization Eps8 decreases chemosensitivity and affects survival of cervical cancer patients Eps8 facilitates cellular growth and motility of colon cancer cells by increasing the expression and activity of focal adhesion kinase Role for EPS8 in squamous carcinogenesis P66shc and its downstream Eps8 and Rac1 proteins are upregulated in esophageal cancers Eps8 promotes cellular growth of human malignant gliomas EPS8 inhibition increases cisplatin sensitivity in lung cancer cells Real-time quantitative polymerase chain reaction assay for detecting the eps8 gene in acute myeloid leukemia Eps8 vaccine exerts prophylactic antitumor effects in a murine model: A novel vaccine for breast carcinoma Identification of novel HLA-A 0201-restricted cytotoxic T lymphocyte epitopes from Zinc Transporter 8 Identification of an HLA-A*0201-restricted CD8+ T-cell epitope SSp-1 of SARS-CoV spike protein Identification of an HLA-A*0201-restricted CD8(+) T-cell epitope encoded within Leptospiral immunoglobulin-like protein A Asymptomatic HLA-A*02:01-restricted epitopes from herpes simplex virus glycoprotein B preferentially recall polyfunctional CD8+ T cells from seropositive asymptomatic individuals and protect HLA transgenic mice against ocular herpes Dipeptides promote folding and peptide binding of MHC class I molecules Identification of HLA-A2-restricted CTL epitopes of a novel tumour-associated antigen, KIF20A, overexpressed in pancreatic cancer A novel cytotoxic T lymphocyte epitope analogue with enhanced activity derived from cyclooxygenase-2 Identification of HLA-A*0201-restricted cytotoxic T lymphocyte epitope from TRAG-3 antigen Identification of a novel HLA-A*24:02-restricted adenovirus serotype 11-specific CD8+ T-cell epitope for adoptive immunotherapy HLA class I alleles are associated with peptide-binding repertoires of different size, affinity, and immunogenicity Generation of multi-leukemia antigen-specific T cells to enhance the graft-versus-leukemia effect after allogeneic stem cell transplant Cancer immunoediting: Antigens, mechanisms, and implications to cancer immunotherapy Immunoinformatics and modeling perspective of T cell epitope-based cancer immunotherapy: A holistic picture Constitutive phosphorylation of eps8 in tumor cell lines: Relevance to malignant transformation Novel oncoprotein EPS8: A new target for anticancer therapy Identification of T-cell epitopes for cancer immunotherapy Insights into the processing of MHC class I ligands gained from the study of human tumor epitopes Integrated modeling of the major events in the MHC class I antigen processing pathway Peptide-based vaccinology: Experimental and computational approaches to target hypervariable viruses through the fine characterization of protective epitopes recognized by monoclonal antibodies and the identification of T-cell-activating peptides The validity of predicted T-cell epitopes Predicting affinity and specificity of antigenic peptide binding to major histocompatibility class I molecules Major histocompatibility complex class I binding predictions as a tool in epitope discovery Reverse immunology approach for the identification of CD8 T-cell-defined antigens: Advantages and hurdles Competition-based cellular peptide binding assays for 13 prevalent HLA class I alleles using fluorescein-labeled synthetic peptides Modification of MHC anchor residues generates heteroclitic peptides that alter TCR binding and T cell recognition A quantitative analysis of the variables affecting the repertoire of T cell specificities recognized after vaccinia virus infection NetMHCstab -predicting stability of peptide-MHC-I complexes; impacts for cytotoxic T lymphocyte epitope discovery Peptide-MHC class I stability is a better predictor than peptide affinity of CTL immunogenicity Differential effects of brefeldin A on transport of secretory and lysosomal proteins Novel blockade by brefeldin A of intracellular transport of secretory proteins in cultured rat hepatocytes Immunological responses to a multi-peptide vaccine targeting cancer-testis antigens and VEGFRs in advanced pancreatic cancer patients Different T cell receptor affinity thresholds and CD8 coreceptor dependence govern cytotoxic T lymphocyte activation and tetramer binding properties T-cell receptor affinity and avidity defines antitumor response and autoimmunity in T-cell immunotherapy Attenuated T cell responses to a high-potency ligand in vivo The Goldilocks model for TCR -too much attraction might not be best for vaccine design SHP-1 phosphatase activity counteracts increased T cell receptor affinity Selective expansion of high-or low-avidity cytotoxic T lymphocytes and efficacy for adoptive immunotherapy High avidity CTLs for two self-antigens demonstrate superior in vitro and in vivo antitumor efficacy High avidity cytotoxic T lymphocytes can be selected into the memory pool but they are exquisitely sensitive to functional impairment In vitro methods for generating CD8+ T-cell clones for immunotherapy from the naïve repertoire Identification of human leucocyte antigen (HLA)-A*0201-restricted cytotoxic T lymphocyte epitopes derived from HLA-DOβ as a novel target for multiple myeloma The present study was supported by the National Natural