A constructed HLA-A2-restricted pMAGE-A1 278–286 tetramer detects specific cytotoxic T lymphocytes in tumour tissues in situ

Abstract

Objective

To construct a human leucocyte antigen (HLA)-A2-restricted peptide 278–286 of melanoma-associated antigen family A, 1 (pMAGE-A1_278–286) tetramer to analyse the distribution of cytotoxic T lymphocytes (CTLs) in tumour tissue and tumour-adjacent normal tissue.

Methods

A HLA-A2-pMAGE-A1_278–286 tetramer was constructed. The distribution of pMAGE-A1_278–286-specific CTLs was investigated in tumour tissues and tumour-adjacent normal tissues from patients with hepatocellular carcinoma using in situ HLA-A2-pMAGE-A1_278–286 tetramer staining.

Results

Sodium dodecyl sulphate–polyacrylamide gel electrophoresis analysis indicated that HLA-A2 heavy and light chain proteins were successfully obtained. The successful construction of the HLA-A2-pMAGE-A1_278–286 monomer was confirmed with Western blot analysis using W6/32 antibody. Flow cytometry confirmed the specific binding of HLA-A2-pMAGE-A1_278–286 tetramer to pMAGE-A1_278–286-specific CTLs. In situ HLA-A2-pMAGE-A1_278–286 tetramer staining demonstrated that the number of pMAGE-A1_278–286-specific CTLs in tumour tissues was significantly higher than in tumour-adjacent normal tissues.

Conclusions

The HLA-A2-pMAGE-A1_278–286 tetramer was useful for the detection of pMAGE-A1_278–286-specific CTLs in both tumour tissues and tumour-adjacent normal tissues. In situ tetramer staining is a powerful tool for investigating the distribution of pMAGE-A1_278–286-specific CTLs in the tumour microenvironment.

Introduction

Cancer is a common disease that affects millions of people every year. ¹ Cytotoxic T lymphocytes (CTLs) play an important role in antitumour responses. ² Altman et al. ³ established the major histocompatibility complex (MHC) human leucocyte antigen (HLA)-A2-restricted tetramer technology in 1996, which permits the direct detection of antigen-specific CTLs with high sensitivity, specificity and efficiency. Tetramer technology provides a valuable tool for the detection and visualization of antigen-specific CTLs in in vivo immune responses.^4–6 However, previous studies primarily used tetramer staining in combination with flow cytometry to detect the frequency of tumour-specific CTLs in vitro.^7–9 In contrast, in situ tetramer staining can be used to evaluate the direct relationship between in situ antigen-specific CTLs and clinical outcome. ¹⁰ In situ tetramer staining has been developed to allow investigators to clearly detect the distribution of antigen-specific CTLs in the tissue microenvironment.^11,12

The present study constructed a HLA-A2-restricted peptide 278–286 of melanoma-associated antigen family A, 1 (pMAGE-A1_278–286) tetramer in order to investigate the distribution of antigen-specific CTLs in tumour tissues and tumour-adjacent normal tissues from patients with hepatocellular carcinoma.

Patients and methods

Patients and tissue samples

This study used tissue samples collected from patients with hepatocellular carcinoma who had undergone liver resection and tissue biopsy in the Department of General Surgery, The First Affiliated Hospital, Guangxi Medical University, Nanning, Guangxi, China between December 2012 and March 2013. The inclusion criteria for each patient were as follows: (i) detailed personal information including name, age, sex, and history of alcohol consumption were available; (ii) complete clinical and pathological data including lymphatic metastasis and pathological staging were available. The pathological diagnosis of hepatocellular carcinoma was confirmed in each patient postoperatively by experienced pathologists according to the World Health Organization International Histological Classification of Tumours. ¹³ A sample of tumour tissue and of tumour-adjacent normal tissue (2 mm from the tumour tissue) from 10 patients were collected during routine liver resection and processed as described below. Ethics approval for the study was obtained from the Ethics Committee of Guangxi Medical University and written consent was obtained from either the patient or the patient’s parent/legal guardian.

Plasmids, bacterial strains, cell lines and enzymes

JA-30 plasmids carrying the HLA-A*0201 heavy chain (HC) and light chain (LC; β2-microglobulin [β2m]) were kindly provided by Igor J. Koralnik (HIV/Neurology Center, Beth Israel Deaconess Medical Center, Harvard Medical School, Harvard, Cambridge, MA, USA). The JA-30 plasmids were kept in Escherichia coli strains BL21 and XA90 (also provided by Igor J. Koralnik). For HC use, E. coli BL 21 were transfected with plasmid JA-30; for LC use, E. coli XA90 were transfected with plasmid JA-30. The biotinylation enzyme, BirA (biotin-protein ligase), was purchased from Avidity (Aurora, CO, USA). The transporter associated with antigen processing-deficient B cell lymphoblastoid cell line T2,^14,15 in which the HLA class I molecules were inefficiently loaded with endogenous peptides, was purchased from Shanghai Fu Xiang Biotechnology (Shanghai, China). When T2 cells are incubated with HLA-A2-restricted high-affinity exogenous peptides, stable peptide–HLA-A2 complexes form.^16,17 The T2 cells were cultured in RPMI 1640 medium containing 5% heat-inactivated fetal calf serum, 100 µg/ml streptomycin and 100 U/ml penicillin (all cell culture reagents from Hyclone Laboratories, Logan, UT, USA) at 37℃ in an atmosphere containing 5% CO₂. Isopropyl-β-d-thiogalactoside (IPTG) and all other chemicals used were analytically pure (Sigma-Aldrich, St Louis, MO, USA). The antigenic peptide, pMAGE-A1_278–286 (KVLEYVIKV), ¹⁸ and the control peptide, pHIV-Gag_77–85 (SLYNTVATL),^19,20 were synthesized by China Peptides (Shanghai, China) with a purity of > 95%. These peptides were dissolved in dimethyl sulphoxide (Beijing Solarbio Science & Technology, Beijing, China) and stored at −80℃.

Antibodies and flow cytometric analysis

The mouse antihuman monoclonal antibody, W6/32, which recognizes the integral conformations of HLA class I molecules was prepared in-house (anti-HLA class I molecules). Fluorescein isothiocyanate (FITC)-labelled mouse antihuman cluster of differentiation (CD) 8 antibody was purchased from BD Biosciences (San Jose, CA, USA). Phycoerythrin (PE)-labelled streptavidin was purchased from eBioscience (San Diego, CA, USA). Horseradish peroxidase-labelled goat antimouse immunoglobulin (Ig)G was purchased from Beyotime Biotech (Jiangsu, China). FITC-labelled rabbit antihuman CD8 antibody (bs-4790R) was purchased from Beijing Biosynthesis Biotechnology (Beijing, China). Normal human serum was purchased from Wuhan Boster Biological Technology (Wuhan, Hubei Province, China). All flow cytometry experiments were performed using a Beckman Coulter® Epics XL™ flow cytometer and EXPO32 ADC v1.2 Analysis software (Beckman Coulter, Brea, CA, USA).

Construction of HLA-A2-pMAGE- A1_278–286 tetramer

Two subunits of HLA-A2 (HC and LC) were highly expressed in the inclusion bodies of E. coli after induction with 0.5 mM IPTG at 37℃ for 4 h when the optical density of the E. coli was approximately 600 nm. E. coli in 2-litre cultures were harvested by centrifugation at 2900  g for 15 min at 4℃ (Beckman Avanti® J-26XP High Performance Centrifuge; Beckman Coulter). The insoluble protein aggregates (inclusion bodies) were resuspended in 50 mM Tris-HCl buffer (pH 8.0) containing 0.1 mM phenylmethylsulfonyl fluoride (PMSF; Beyotime Biotech) and 10 mM dithiothreitol (DTT; Sigma-Aldrich), and then sonicated using a JY92-IIDN ultrasonic cell grinder (Ningboxinzhi, Beijing, China) for 20 s (10 times) on ice. The insoluble pellet was collected by centrifugation at 17 696  g for 20 min at 4℃ (Beckman Avanti® J-26XP High Performance Centrifuge; Beckman Coulter) and washed with washing buffer three times (50 mM Tris-HCl buffer [pH 8.0], 100 mM NaCl, 1 mM ethylenediaminetetra-acetic acid [EDTA; Ameresco, Framingham, MA, USA], 1 mM DTT and 5 g Triton™ X-100 [Beijing Solarbio Science & Technology]), followed by three cycles of sonication, centrifugation and collection as described previously. The insoluble inclusion bodies were washed three more times with 20 mM 2-(N-morpholino)ethanesulfonic acid (Beijing Solarbio Science & Technology), 8 M urea (Beijing Solarbio Science & Technology), 10 mM EDTA and 0.1 mM DTT in preparation for storage at −80℃. The purity of the protein was analysed using 12% sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE) and a 5% stacking gel. In brief, samples from each preparation were subjected to SDS–PAGE for 50 min at 140 V and then the gel was stained with Coomassie Brilliant Blue G250 (Ameresco). Protein molecular weight markers (Sigma-Aldrich) were also run on the gel and images were captured using Image Lab™ software version 3.0 (Bio-Rad, Hercules, CA, USA).

The antigen-specific peptide, pMAGE-A1_278–286 (40 µmol/l), LC (2 µmol/l) and HC (1 µmol/l) were added to 200 ml refolding buffer solution (100 mmol/l Tris-HCl [pH 8.3] containing 400 mmol/l l-arginine, 2 mmol/l EDTA, 0.5 mmol/l oxidized glutathione, 5 mmol/l reduced glutathione and 0.1 mmol/l PMSF). The mixture was incubated at 10℃ for 36 h to obtain the HLA-A2-pMAGE-A1_278–286 monomer. The same refolding reaction was performed to obtain the control refolded HC and LC, which were made from refolded HC and refolded LC, respectively. The refolded monomer mixture was concentrated from 200 ml to 5 ml using ultra-filtration apparatus with a 10 kDa molecular mass cut-off membrane (Centriprep-30 centrifugal filter unit with Ultracel-30 membrane; Millipore, Billerica, MA, USA). A solution of 10 mM Tris-HCl buffer (pH 8.0), containing 200 mM NaCl and 5 mM MgCl₂, was used for dialysis, followed by centrifugation at 800  g for 30 min at 4℃ using an Eppendorf® 5810 centrifuge (Eppendorf, Hamburg, Germany). The supernatant fraction was collected for biotinylation. A 700-µl aliquot of concentrated mixture was biotinylated overnight at room temperature in the presence of 100 µl Biomix-A (Avidity), 100 µl Biomix-B (Avidity), 100 µl d-biotin (Avidity), 1 µl PMSF, 1 µl leupeptin (Sigma-Aldrich), 1 µl aprotinin (Sigma-Aldrich), 3 µl pepstatin (Sigma-Aldrich) and 2 µl BirA enzyme (Avidity). The biotinylated mixture was then stored at 4℃ before analysis by Western blot analysis. The proteins were separated by non-denaturing PAGE using a 12% gel and a 5% stacking gel. In brief, samples from each preparation were subjected to PAGE for 50 min at 140 V using a Mini-PROTEAN® Tetra Cell vertical gel electrophoresis system (Bio-Rad). The gel was stained using Coomassie Brilliant Blue G 250 and the images were captured using Image Lab™ software version 3.0 (Bio-Rad). The proteins that were separated by PAGE were then electrotransferred to a nitrocellulose membrane (Bio-Rad) for 45 min at 50 V using the Mini-PROTEAN® Tetra Cell system (Bio-Rad). The membrane was blocked for 30 min at 37℃ with blocking buffer (0.01 mM phosphate-buffered saline [PBS; pH 7.4] containing 5% milk powder [Nestlé, Vevey, Switzerland]). The membrane was then incubated with mouse antihuman monoclonal antibody (W6/32; prepared in-house, diluted 1:1000 in 0.01 mM PBS [pH 7.4] containing 1% milk powder) for 8 h at 4℃. The membrane was then washed at room temperature three times (5 min per wash) with PBS-Tween® 20 (PBST; 0.01 mM PBS pH 7.4 containing 0.05% Tween® 20 [Beijing Solarbio Science & Technology]). The membrane was then incubated with the horseradish peroxidase-labelled goat antimouse IgG (Beyotime Biotech; diluted 1:2500 in 0.01 mM PBS [pH 7.4] containing 1% milk powder). After washing three times with PBST, commercially prepared diaminobenzidine solution (Beijing Solarbio Science & Technology) was added as the substrate to develop the band at room temperature following the manufacturer’s instructions. The images were captured using an Image Station 4000MM Molecular Imaging System (Bio-Rad) and Carestream Molecular Imaging Software Standard Edition v5.3.5.18701 (Carestream Health, Rochester, NY, USA). The HLA-A2-pMAGE-A1_278–286 tetramer was prepared by mixing the biotinylated HLA-A2-pMAGE-A1_278–286 monomer with PE–streptavidin (eBioscience) at a 4:1 molar ratio followed by incubation in the dark at 4℃ for 10–15 min. The control HLA-A2-pHIV-Gag_77–85 tetramer was prepared using the same method as described above. The prepared tetramer complexes were stored at 4℃.

Detecting the antigen-specific CTLs in vitro using the PE-pMAGE-A1_278–286 tetramer

The pMAGE-A1_278–286 (50 µg/ml) was added to T2 cells and incubated at 37℃ for 6 h to obtain T2:pMAGE-A1_278–286 complexes.^21,22 Peripheral blood mononuclear cells (PBMCs) were isolated from HLA-A2⁺ donor peripheral blood (Guangxi Nanning City Blood Bank, Nanning, Guangxi, China). Peripheral blood was diluted 1:1 with 0.01 mM PBS (pH 7.4) in preparation for Ficoll® (Beijing Solarbio Science & Technology) density gradient centrifugation. A 20-ml sample of diluted peripheral blood was gently layered onto 10 ml of Ficoll® followed by centrifugation at 289  g for 25 min at 25℃. The PBMC layer was collected from the diluted blood/Ficoll® interface. The PBMCs were diluted 1:5 with 0.01 mM PBS (pH 7.4) for washing, followed by centrifugation at 289  g for 25 min at 25℃, followed by a further wash with 0.01 mM PBS (pH 7.4). The PBMCs were subjected to centrifugation at 289  g for 25 min at 25℃ followed by the addition of cryopreservation medium (90% fetal calf serum plus 10% dimethyl sulphoxide) in preparation for storage in liquid nitrogen prior to use. PBMCs were plated at a density of 3 × 10⁶ cells/well in a 24-well plate and co-cultured with 3 × 10⁵/well of T2:pMAGE-A1_278–286 complexes in RPMI 1640 containing 10% fetal calf serum. On the first day of culture, 50 U/ml of interleukin (IL)-2 (R&D Systems, Minneapolis, MN, USA) was added to the cultures. On the fourth day, 10 ng/ml of IL-7 (R&D Systems) and 50 U/ml of IL-2 were added to the cultures. PBMCs were re-stimulated with T2:pMAGE-A1_278–286 complexes once per week and cultured for 3 weeks.^23–26 The PBMCs were collected and stained with 20 µg/ml of PE-labelled HLA-A2-pMAGE-A1_278–286 or 20 µg/ml of HLA-A2-pHIV-Gag_77–85 in combination with 20 µl FITC-labelled mouse antihuman CD8 monoclonal antibody (in 100 µl PBS [pH 7.4]) at 4℃ for 30 min in the dark. The cells were analysed by flow cytometry as described above. ²⁷

In situ tetramer staining for the detection of tumour HLA-A2-pMAGE-A1_278–286-specific CTL

Frozen biopsy samples of tumour tissue (n = 10) and tumour-adjacent normal tissue (n = 10) that were collected during surgical resection of the liver were routinely cut into 8 µm-thick sections. The sections were then mounted on glass slides and fixed with 4% paraformaldehyde for 3 min at room temperature and blocked with 20% human serum for 20 min at 4℃ (Wuhan Boster Biological Technology) before incubation with PE-labelled HLA-A2-pMAGE-A1_278–286 or PE-labelled HLA-A2-pHIV-Gag_77–85 (5 µM/l) overnight at 4℃. The sections were washed twice with 0.01 mM PBS (pH 7.4) and fixed with 4% paraformaldehyde for 20 min at room temperature. The slides were then stained with 2 µg/ml of FITC-labelled rabbit antihuman CD8 antibody for 30 min at room temperature to label CD8⁺ CTLs (stained with green fluorescence) and 0.5 µg/ml 2 -(4-amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI; eBioscience) for 30 min at room temperature to label the cell nuclei with blue fluorescent stain.

In order to analyse the distribution of pMAGE-A1_278–286-specific CTLs in the tumour tissue and tumour-adjacent normal tissue samples, the slides were then examined at ×400 magnification using a Nikon Eclipse 80i microscope (Nikon Corporation, Tokyo, Japan). A total of five sections for each tumour tissue and tumour-adjacent normal tissue sample were analysed and the number of cells that demonstrated immunoreactivity was determined in three fields per slide. Cells that showed double staining with HLA-A2-pMAGE-A1_278–286 tetramer (red fluorescence) and antihuman CD8 antibody (green fluorescence) with DAPI-stained nuclei (blue fluorescence) on overlaid images were considered to be pMAGE-A1_278–286-specific CTLs, and the total number of these immunoreactive cells per field was counted. The mean ± SD for each tumour tissue and tumour-adjacent normal tissue sample was determined.

Statistical analyses

Data are expressed as mean ± SD. All statistical analyses were performed using GraphPad Prism software, version 5.01 (GraphPad Software, San Diego, CA, USA). The mean numbers of pMAGE-A1_278–286-specific CD8⁺ CTLs per field in tumour tissue and tumour-adjacent normal tissue were compared using two-way analysis of variance. A P-value <0.05 was considered statistically significant.

Results

The molecular weight and purity of the HLA-A2 HC carrying biotin-binding sites and LC (β2 m) protein was identified by SDS–PAGE analysis. As shown in Figure 1, the HC protein had a molecular weight of 34 kDa and the LC β2 m protein had a molecular weight of 12 kDa, in agreement with their theoretical molecular weights. Following protein purification, 51.114 mg/ml of LC and 49.033 mg/ml of HC were obtained with purities of 83% and 62.9%, respectively. The actual concentration of the LC was 42.42 mg/ml and the actual concentration of the HC was 30.84 mg/ml.

Figure 1.

Sodium dodecyl sulphate–polyacrylamide gel electrophoresis analyses of recombinant human leucocyte antigen-A*0201 heavy chain (HC) carrying biotin-binding sites and light chain (LC) proteins expressed in Escherichia coli. (A) Lane 1: molecular weight markers; lane 2: HC protein with a molecular weight of 34 kDa (arrow). (B) Lane 1: molecular weight markers; lane 2: LC protein with a molecular weight of 12 kDa (arrow).

The HLA-A2-pMAGE-A1_278-286 monomer was constructed through in vitro refolding of HC, LC and pMAGE-A1_278-286 mixed together in a molar ratio of 3:2:30, respectively. The control refolded HC monomer and refolded LC monomer was prepared using the same molar ratio as above. Non-denaturing PAGE and Western blot analysis probed with W6/32 mouse antihuman monoclonal antibody showed that the HLA-A2-pMAGE-A1_278–286 monomer was successfully constructed (Figure 2). As shown in lane 4, one positive band corresponding to the refolded HLA-A2-pMAGE-A1_278–286 monomer was detectable. No positive band was found in lanes 5 or 6 for the LC and HC controls, respectively.

Figure 2.

Analyses of refolded human leucocyte antigen (HLA)-A2-restricted peptide 278–286 of melanoma-associated antigen family A, 1 (pMAGE-A1_278–286) monomer. (A) Non-denaturing polyacrylamide gel electrophoresis analyses of refolded monomer: lane 1: HLA-A2-pMAGE-A1_278–286 monomer; lane 2: light chain (LC) protein control; lane 3: heavy chain (HC) protein control. (B) Western blot analysis of refolded HLA-A2-pMAGE-A1_278–286 monomer (arrow) probed with mouse antihuman HLA class I molecules monoclonal antibody (W6/32): lane 4: HLA-A2-pMAGE-A1_278–286 monomer; lane 5: LC protein control; lane 6: HC protein control.

When T2:pMAGE-A1_278–286 antigen-presenting cells were co-cultured with HLA-A2⁺ PBMCs, which were the effector cells, they induced pMAGE-A1_278–286-specific CTLs. Using flow cytometry, antigen-specific CTLs were detected using the PE-HLA-A2-pMAGE-A1_278–286 tetramer and FITC-labelled mouse antihuman CD8 monoclonal antibody. As shown in Figure 3A and B, the proportion of HLA-A2⁺ PBMCs that were double-positively stained with HLA-A2-pMAGE-A1_278–286 tetramer and FITC-labelled mouse antihuman CD8 monoclonal antibody was higher in the HLA-A2⁺ PBMC group co-cultured with the antigen-presenting T2:pMAGE-A1_278–286 cells compared with the non-stimulated HLA-A2⁺ PBMC group (9.8% versus 0.2%, respectively). Binding of HLA-A2-pHIV-Gag_77–85 tetramer to HLA-A2-pMAGE-A1_278–286-specific CTLs was not detectable (Figure 3C and D). These results indicated that the HLA-A2-pMAGE-A1_278–286 tetramer had been successfully constructed and had the ability to specifically bind pMAGE-A1_278–286-specific CTL, so it was suitable for use in the in situ tetramer staining studies of tumour and tumour-adjacent normal tissue samples.

Figure 3.

Representative flow cytometry images showing specific binding of human leucocyte antigen (HLA)-A2-restricted peptide 278–286 of melanoma-associated antigen family A, 1 (pMAGE-A1_278–286) tetramer to antigen-specific cytotoxic T lymphocytes (CTLs). (A) The non-stimulated HLA-A2⁺-specific CTL group showing that the percentage of cells that were double-positively labelled for phycoerythrin (PE)-HLA-A2-pMAGE-A1_278–286 tetramer and fluorescein isothiocyanate (FITC)-labelled mouse antihuman cluster of differentiation (CD)8 antibody was 0.2%. (B) The HLA-A2⁺-specific CTL group co-cultured with the antigen-presenting T2:pMAGE-A1_278–286 cells showing that the percentage of cells that were double-positively labelled for PE-HLA-A2-pMAGE-A1_278–286 tetramer and FITC-labelled mouse antihuman CD8 was 9.8%. (C) The non-stimulated HLA-A2⁺-specific CTL group showing that the percentage of cells double-positively labelled for PE-HLA-A2-pHIV-Gag_77–85 tetramer and FITC-labelled mouse antihuman CD8 was 0.0%. (D) The HLA-A2⁺-specific CTL group co-cultured with the antigen-presenting T2:pMAGE-A1_278–286 cells showing that the percentage of cells that were double-positively labelled for PE-HLA-A2-pHIV-Gag_77–85 tetramer and FITC-labelled mouse antihuman CD8 was 2.6%.

To determine the localization of pMAGE-A1_278–286-specific CTL in tumour and tumour-adjacent normal tissue, in situ tetramer staining was performed using fresh tissue samples from 10 patients with hepatocellular carcinoma (Figure 4). pMAGE-A1_278–286-specific CTL were detected in both tumour and tumour-adjacent normal tissues in three of the patients. As shown in Figure 4A and B, the number of infiltrating pMAGE-A1_278–286-specific CTLs was significantly higher in tumour tissue compared with tumour-adjacent normal tissue (P < 0.01) in these three patients (Figure 4C).

Figure 4.

In situ tetramer staining of peptide 278–286 of melanoma-associated antigen family A, 1 (pMAGE-A1_278–286)-specific cluster of differentiation (CD)8⁺ cytotoxic T lymphocytes in tumour tissue (A) and tumour-adjacent normal tissue (B) from a representative patient with hepatocellular carcinoma. Tissue sections were prepared and stained with phycoerythrin (PE)–human leucocyte antigen (HLA)-A2-pMAGE-A1_278–286 tetramer, fluorescein isothiocyanate (FITC)-labelled rabbit antihuman CD8 antibody and 2-(4-amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI). The arrows indicate positive staining. (A) Staining of tumour tissue: (a) PE-HLA-A2-pMAGE-A1_278–286 tetramer staining; (b) FITC-labelled rabbit antihuman CD8 staining; (c) DAPI staining of nuclei; (d) overlaid image of the three-colour staining. (B) Staining of tumour-adjacent normal tissue: (a) PE- HLA-A2-pMAGE-A1_278–286 tetramer staining; (b) FITC-labelled rabbit antihuman CD8 staining; (c) DAPI staining of nuclei; (d) overlaid image of the three-colour staining. (C) The number of pMAGE-A1_278–286-specific CD8⁺ cytotoxic T lymphocytes per field in tumour tissue and tumour-adjacent normal tissue from three patients. Data presented as mean ± SD; *P < 0.01, **P < 0.001; two-way analysis of variance. The colour versions of these figures are available at: http://imr.sagepub.com.

Discussion

Antigen-specific CTLs play a critical role in preventing tumour growth, recurrence and metastasis by recognizing abnormal antigen expression on tumour cells through MHC class I presentation and the subsequent killing of tumour cells.^28,29 Therefore, it is important to study the role of tumour-specific CTLs at tumour onset and during progression in order to develop novel targets for tumour treatment.

Melanoma-associated antigen-A1 (MAGE-Al) is a member of the MAGE family and it is expressed on different tumour tissues, but not in normal tissues, except the testis and placenta.^10,30,31 MAGE-Al is an interesting target for cancer immunotherapy. Previous research demonstrated that the MAGE-Al protein contains the HLA-A2-restricted antigen epitope KVLEYVIKV (pMAGE-A1_278–286). ³² pMAGE-A1_278–286-specific CTLs were isolated from the peripheral blood of patients with breast cancer and they had specific killing activities against HLA-A2⁺ tumour cells expressing MAGE-Al. ¹⁸ However, the distribution of pMAGE-A1_278–286-specific CTLs in tumour tissue remains unknown.

MCH/peptide tetramer technology is an important tool for studying tumour antigen-specific CTLs.^33–36 The principle of the technology is to construct MHC–antigenic peptide complexes in vitro and label them with biotin and avidin to form tetramers.^3,37 Although the construction of an MHC–antigen peptide tetramer is complicated and impacted by various factors, tetramer technology is extremely useful for efficient binding, detection and separation of antigen-specific T cells.^38,39 Moreover, MHC–antigen peptide tetramer staining combined with a variety of functional experiments provides a powerful tool for antigen-specific CD8⁺ T cell detection, sorting and functional analysis. ⁴⁰ W6/32 is an antiHLA class I monoclonal antibody that recognizes a specific epitope of the α2 zone upon the formation of the complex between the HC and LC. W6/32 antibody binds the HC protein with extremely low affinity and does not bind to the LC protein.^41–43 As shown in Figure 2, W6/32 strongly bound to HLA-A2-pMAGE-A1_278–286 monomer molecules, but not LC or HC proteins alone. For the functional studies of the constructed tetramer in this present study, T2 cells pulsed with pMAGE-A1_278–286 were used as antigen-presenting cells to stimulate PBMCs from HLA-A2⁺ donor peripheral blood to induce pMAGE-A1_278–286-specific CTLs. The induced pMAGE-A1_278–286-specific CTLs were detected with the HLA-A2-pMAGE-A1_278–286 tetramer using flow cytometry, which indicated that the HLA-A2-pMAGE-A1_278–286 tetramer was successfully constructed in this present study.

Tetramer technology combined with flow cytometry has been widely used to study antigen-specific T lymphocytes.^44,45 However, this method cannot directly detect the interaction of the antigen-specific T lymphocytes with the local tissue immune response. As a consequence, in situ tetramer staining was developed and it has allowed for the clear visualization of the distribution of antigen-specific T cells in local tissues.^11,46 Moreover, the in situ tetramer staining method allows for the analysis of infiltrating T lymphocytes, cell membranes, as well as intracellular and nuclear markers. In situ tetramer staining is useful for tissue re-staining and retrospective analyses, which has led to it being widely used to study antigen-specific T cell responses.^11,46 Tetramer staining was used successfully to detect influenza virus-specific CD8⁺ T cells in the lung tissue of mice infected with influenza type A virus. ⁴⁷ Tumour-specific T lymphocytes were efficiently identified in the lymph nodes and the tumour-adjacent normal tissue using in situ tetramer staining. ⁴⁸ Simian immunodeficiency viruses (SIV) Gag-specific CD8⁺ T cells were successfully detected in the lymph nodes, spleen and vaginal tissue of SIV-infected rhesus macaques using in situ Gag and Tat tetramer staining. ⁴⁹ In this present study, the number of infiltrating pMAGE-A1_278–286-specific CTLs was significantly higher in the tumour tissue than the tumour-adjacent normal tissue using the constructed HLA-A2-pMAGE-A1_278-286 tetramer for in situ staining.

In conclusion, the HLA-A2-pMAGE-A1_278–286 tetramer that was constructed during this present study was able to detect pMAGE-A1_278–286-specific CTLs in both tumour tissues and tumour-adjacent normal tissues. Using in situ HLA-A2-pMAGE-A1_278–286 tetramer staining to investigate the distribution of pMAGE-A1_278–286-specific CTLs in the microenvironment of a tumour provides a powerful tool for the future study of the effects of pMAGE-A1_278–286-specific CTLs on tumours.