The defining characteristics of cancer/testis (CT) antigens are high levels of expression in germ line tissues, lack of expression in other normal adult tissues, and aberrant expression in a wide range of different tumor types.1 CT antigens have become promising targets for immunotherapy for patients with various tumors.2 However, the expression frequency of each CT antigen varies with tumor types. Many CT antigens show heterogeneous expression patterns in the same tumor tissue.2, 3 Additional CT antigens are needed for the development of polyvalent cancer vaccines designed to overcome the limited frequency and heterogeneity of expression of CT antigens.
To date more than 47 genes or gene families coding for CT antigens have been identified by several methodologies,4 such as T-cell epitope cloning, SEREX (serological identification of antigens by recombinant expression cloning), representational difference analysis (RDA), DNA microarray analysis, and bioinformatics. The T-cell epitope cloning method developed by Boon and coworkers in 1991 led to the discovery of MAGE-A1, BAGE, and GAGE1,5–7 and RDA led to the cloning of LAGE-1, MAGE-E1 and SAGE.8–10 Using DNA microarray analysis, MAA-1A was identified.11 More recently, bioinformatics based analysis resulted in the cloning of BRDT, OY-TES-1, PAGE5, LDHC and TPTE.12–16 Among these methods, SEREX seems to be effective for the identification of CT antigens. Using cDNA libraries from cancer or normal testis tissues, SSX2, SYCP-1, NY-ESO-1 and XAGE-1 were isolated.17–20
In our study, we performed SEREX analysis to screen a testicular cDNA library with the aim of isolating novel CT antigens. In addition to a previously defined CT antigen, SYCP-1, a novel CT antigen, CCDC62-2, was identified.
Material and methods
Sera and tumor tissues
The serum used in this SEREX study was obtained from a 75-yr-old male patient admitted to Goto Central Hospital, Nagasaki, Japan. The patient was diagnosed with metastatic poorly differentiated gastric adenocarcinoma. Because of his age, general condition, and the prognosis of advanced gastric cancer, he was followed in the outpatient department of the hospital without any treatment for cancer with his and his family's agreement. After 5 months of follow-up, the primary gastric cancer had regressed, and most of the liver metastatic regions had also disappeared. Tumor recurrence was observed after around 1 yr and the patient died eventually.
Tumor tissue specimens were obtained surgically from patients at Okayama University Hospital. Sera samples were obtained from 191 patients with various cancers and 41 healthy donors. Written informed consent was obtained from all patients and healthy donors in our study.
Preparation of a testicular cDNA library
mRNA was purified from normal testiclar total RNA using a Quick Prep mRNA purification kit (Amersham Pharmacia, Piscataway, NJ). A cDNA library was prepared in a λZAP Express vector using a cDNA library kit (Stratagene, La Jolla, CA).
Immunoscreening of a testicular cDNA library
A cDNA expression library from normal testis was screened with serum from a gastric adenocarcinoma patient. The screening procedure was described previously.20 In brief, serum samples diluted 1:10 were preabsorded with lysate from Escherichia coli Y1090/Y1089, and coupled to sepharose 4B (BioDynamics Lab, Tokyo, Japan). Recombinant phages (∼4,000 pfu) on agar in a plastic dish (140 mm diameter) were amplified for 8 hr, and transferred to a 135 mm diameter nitrocellulose membrane (Schleicher & Schuell, Dassel, Germany) for 15 hr at 37°C. Membranes were blocked with 5% nonfat milk, and prescreened by incubating with peroxidase-conjugated Fc fragment-specific goat antihuman IgG (Jackson ImmunoResearch, West Grove, PA) (1:2000 dilution) for 1 hr at room temperature. Signals were developed using the chromogenic agent 3, 3′ diaminobenzidine (Sigma, St. Louis, MO), and IgG-encoding clones were marked to be excluded from further study. The membranes were then incubated overnight at room temperature with the preabsorbed serum diluted 1:200 then with peroxidase-conjugated Fc fragment-specific goat antihuman IgG (Jackson ImmunoResearch) (1:2000 dilution) for 1 hr at room temperature, and the chromogenic reaction was repeated. Positive clones were picked and subcloned to monoclonality by 2nd and 3rd screenings using 82 and 47 mm-diameter membranes, respectively. A randomly chosen negative clone was included in each assay as a negative control.
Sequence analysis of cDNA
Positive clones were subcloned to monoclonality, purified, and excised in vivo into pBK-CMV plasmid forms (Stratagene). Plasmid DNA was prepared using a Quantum Prep Plasmid Miniprep Kit (Bio-Rad, Hercules, CA). The nucleotide sequences of the cDNA inserts were determined using an ABI PRISM R310 Genetic Analyzer (PerkinElmer, Foster City, CA). Sequence aligments were performed using BLAST software and compared to sequences in the Genbank database.
Total RNA isolation and cDNA synthesis
Total RNA was isolated from tumor tissues using an RNeasy Mini Kit (Qiagen, Hilden, Germany). Total RNA from normal tissues was obtained commercially (BD Bioscience Clonetch, Palo Alto, CA) and treated with TURBO DNA-free (Ambion, Austin, TX) to remove any trace amounts of genomic DNA. RNA (2 μg) was reverse-transcripted into single-strand cDNA using Moloney murine leukemia virus reverse transcriptase (Ready-To-Go You-Prime First-Strand Beads, GE Healthcare, Buckinghamshire, UK) and oligo(DT)15 as a primer. cDNA samples were tested for integrity by amplification of G3PDH in a 30-cycle reaction.
To amplify cDNA segments from tumor and normal tissues, primers for EID3, GKAP1 and CCDC62 transcripts was designed. The primer pairs were as follows: EID3, 5′-TCGTGGG TCTGAATTGGATG-3′ (forward), 5′-TAGGACACAGGAGTA TCAGG-3′ (reverse); GKAP1, 5′-CATTGTCAAACCCAGTACAG-3′ (forward), 5′-CTTTCAGAACCACTTCCTGG-3′ (reverse); CCDC62-1, (variant 1 specific) 5′-TCCCCGGCAAGTGAGCTAAT-3′ (forward), 5′-ATACATCCCCATTCCCGAGG-3′ (reverse); CCDC62-2, (variant 2 specific) 5′-AAGTCAGAGGTCCCAGAAGA-3′ (forward), 5′-CTATGCAGGGGTTC TTTCTC-3′ (reverse). RT-PCR was performed at an annealing temperature of 60°C using 35 amplification cycles followed by a 10 min elongation step at 72°C. The PCR products were analyzed by agarose gel electrophoresis.
Quantitative real-time RT-PCR
Two-step real-time RT-PCR was performed using a StepOne Real-Time PCR System (Applied Biosystems, Foster City, CA). cDNA was synthesized using a High-capacity cDNA Reverse Transcription kit (Applied Biosystems). TaqMan Gene Expression Assays (Applied Biosystems) were used to measure the mRNA levels of GKAP1 (Assay ID: Hs01091282_m1) and CCDC62 (Assay ID: Hs00261486_m1). The TaqMan Gene Expression Assays for CCDC62 was designed to target the exon 10 and 11 boundary region where the sequence is conserved between CCDC62-1 and CCDC62-2. mRNA levels were expressed as n-fold differences relative to G3PDH (internal standard) and the levels in normal testis (calibrater). PCR was performed using TaqMan PCR Master Mix (Applied Biosystems) and the thermal cycling conditions comprised an initial denaturation at 95°C for 10 min then 50 cycles at 95°C for 15 sec and 60°C for 1 min. The parameter Ct was defined as the threshold cycle number at which the fluorescence generated by cleavage of the probe passed above the baseline. The GKAP1 and CCDC62 target message was quantified by measuring the Ct value, and transcripts of G3PDH were quantified as an endogenous RNA control using TaqMan human G3PDH control regents (Applied Biosystems).
Recombinant CCDC62-2 protein
CCDC62-2 was expressed in E. coli BL21 using the GST-containing vector pGEX-6P-1 (Amersham Biosciences). cDNA amplification primers were designed to encompass the partial coding sequence of the gene corresponding to amino acid position 366-684 for the C-terminal half of CCDC62-2, which includes an amino acid sequence different from CCDC62-1. The isolated GST fusion protein was purified on a GSTrap FF column according to the manufacturer's instructions.
Recombinant CCDC62-2 protein (1 μg/ml) in 0.05 M carbonate buffer (pH 9.6) was absorbed onto 96-well plates (Nunc) at 4°C overnight. GST protein was used as a negative control. Plates were washed with PBS/Tween and blocked with 5% FCS/PBS at room temperature for 1 hr. After washing, serum dilutions (100 μl) in 5% FCS/PBS were added and incubated at room temperature for 2 hr. Plates were washed and incubated with secondary antibody (peroxidase-conjugated Fc fragment-specific goat antihuman IgG, Jackson ImmunoResearch) at 1/5000 dilution for 1 hr at room temperature. Plates were washed and incubated with the substrate solution (1,2-phenylenediamine dihydrochloride) for 20 min at room temperature. After the addition of 6 M H2SO4 (50 μl), the absorbance was determined with a microplate reader (BioRad). A positive reaction was defined as an optical density (OD) value for 1:400 diluted serum that exceeded the mean OD value of sera from healthy donors by four standard deviations.
The purified recombinant CCDC62-2 protein was fractionated with 12% SDS-PAGE and transferred to a nitrocellulose membrane. After blocking with 5% FCS/PBS for 1 hr, the membrane was incubated with patient's sera (1:200 dilution) for 2 hr at room temperature. After washing, the membrane was incubated with alkaline phosphatase-conjugated AffiniPure Goat antihuman IgG, Fcγ fragment specific (Jackson ImmunoResearch, West Grove, PA) for 1 hr at room temperature and color was developed.
Immunoscreening of a testicular cDNA library with serum from a gastric cancer patient
A cDNA expression library of 5 × 105 clones was prepared from normal testicular total RNA. Approximately 1 × 105 clones were immunoscreened with serum from a gastric adenocarcinoma patient using SEREX methodology. In a preliminary experiment, the patient's serum was confirmed to react with a normal testis lysate by ELISA (data not shown). As shown in Table I, 55 clones representing 23 genes were isolated and the antigens were designated as OY-ST-1 through OY-ST-23, respectively. These included 4 genes with testis specific expression as reported previously (OY-ST-3 and 6)18, 21 or in the UniGene database (OY-ST-2 and OY-ST-18). The other genes, except for OY-ST-8 and OY-ST-13, showed ubiquitous expression in normal adult tissues as demonstrated by the UniGene database. Eight genes, OY-ST-2, 4, 5, 6, 11, 13, 17 and 20, were in the SEREX database (http://ludwig-sun5.unil.ch/CancerImmunomeDB/).
Table I. Genes Identified From a Testicular cDNA Library with Serum From a Gastric Cancer Patient
Actin related protein 2/3 complex, subunit 2, 34 kDa (ARPC2)
Suppressor of Ty 5 homolog (S.cerevisiae) (SUPT5H)
Polymerase (RNA) III (DNA directed) polypeptide H (22.9KD) (POLR3H)
Centrosomal protein 290 kDa (CEP290)
Coiled-coil domain containing 62 (CCDC62)
Dihydrouridine synthase 1-like (S.cerevisiae)
Leiomodin 1 (smooth muscle) (LMOD1)
Activating signal cointegrator 1 complex subunit 2 (ASCC2)
Chromosome 1 open reading frame 57 (C1orf57)
F-box and leucine-rich repeat protein 5 (FBXL5), transcript variant 1
Of the 4 genes with testis specific profiles, OY-ST-2, corresponding to EP300 interacting inhibitor of differentiation 3 (EID3), represented by 8 overlapping clones, was isolated as the 2nd most frequent gene. OY-ST-3, represented by 2 clones, was identical to G kinase anchoring protein 1 (GKAP1). OY-ST-6 was identical to a CT gene, SYCP-1. OY-ST-18, represented by 2 clones, was identical to coiled-coil domain containing 62 (CCDC62). The two CCDC62 clones were found to consist of two splice variants (variant 1 and variant 2).
mRNA expression in normal adult tissues
To investigate the restricted expression of EID3, GKAP1, CCDC62 variant 1 (CCDC62-1) and CCDC62 variant 2 (CCDC62-2) mRNAs in normal adult tissues, qualitative RT-PCR was performed at 35 cycles using gene specific primer pairs. As shown in Figure 1a, expression of the 2 CCDC62 splice variant mRNAs was restricted to the testis. Expression of EID3 mRNA was observed in all tissues tested. Expression of GKAP1 mRNA was observed in some normal tissues, but highest in the testis. To further investigate GKAP1 and CCDC62 mRNA expression, we performed quantitative real-time RT-PCR analysis using a GKAP1-specific TaqMan probe and a common TaqMan probe for the 2 CCDC62 variants. For comparison, a prototype CT antigen, NY-ESO-1 (TaqMan Gene Expression Assays: Hs00265824_m1), was also analyzed. As shown in Figure 1b, real-time RT-PCR analysis revealed considerably lower levels of CCDC62 gene transcripts in normal, nongametogenetic tissues compared to normal testis, as in the case of NY-ESO-1. Expression over 1% of that in normal testis was observed in several normal tissues with GKAP1.
CCDC62 mRNA expression in tumors
Because of the restricted expression in normal testis, CCDC62 was chosen for mRNA expression analysis in malignant tissues. By qualitative RT-PCR, no expression was observed with the CCDC62-1 mRNA in any of the cancer tissues tested (data not shown). On the other hand, CCDC62-2 mRNA expression was observed in various types of tumors at different frequencies, including stomach cancer (Table II). We performed quantitative real-time RT-PCR analysis using a common TaqMan probe for the 2 CCDC62 variants. As shown in Figure 2, expression was detected in 1 of 9 colon cancers, 2 of 12 liver cancers, 4 of 19 lung cancers, 5 of 8 prostate cancers and 3 of 15 stomach cancers at >1% of the testicular expression level. Corresponding normal tissues showed less than 0.1% of the testicular expression level. Furthermore, 1 of 8 prostate cancers and 1 of 15 stomach cancers expressed at >10% of the testicular level.
Table II. Qualitative RT-PCR Analysis of CCDC62-2 mRNA in Tumor
Head and neck cancer
Immunogenicity of CCDC62-2 in cancer patients
We then investigated the immunogenicity of CCDC62-2. Sera from 191 cancer patients and 41 healthy donors were tested for IgG antibody by ELISA using recombinant CCDC62-2 protein. As shown in Table III, 2/11 (18.2%) sera from colon cancer patients, 5/76 (6.6%) sera from lung cancer patients, and 6/104 (5.8%) sera from stomach cancer patients were reactive against CCDC62-2. No CCDC62-2 antibody was detected in the sera from 41 healthy donors. Figure 3a illustrates titration curves with sera from selected lung and stomach cancer patients. Western blot analysis showed that sera were reactive with the recombinant CCDC62-2 protein (Fig. 3b).
Table III. Antibody Response to the CCDC62-2 Recombinant Protein in Cancer Patients by ELISA
In our study, we identified a novel CT antigen, CCDC62-2, by screening a testicular cDNA library with serum from a gastric adenocarcinoma patient. Using a testicular instead of a tumor-derived cDNA library facilitated the detection of antigens with CT antigen characteristics. By immunoscreening, we isolated 55 positive cDNA clones consisting of 23 different genes, designated OY-ST-1 to OY-ST-23. There were 4 genes with testis-specific expression in the UniGene database and in the literature, including SYCP-1,18 a CT antigen defined by SEREX previously. CT antigens are now classified as CT-X and CT-non X based on whether the gene is located on the X chromosome.22 All 4 genes including SYCP-1, which resides on chromosome 1,18 detected in our study are not located on the X chromosome.
OY-ST-2 (EID3) was isolated as the 2nd most frequent gene (8 of 55). EID3 was reported as a 3rd member of the EID family and suggested to inhibit cellular differentiation by forming a homodimer or a heterodimer with EID2.23EID3 is an intron-less gene and its expression is restricted to the testis according to the database. However, in our RT-PCR analysis using normal tissue RNAs after removal of genomic DNA, EID3 mRNA expression was observed in all normal tissues examined.
GKAP1 has been isolated by a yeast two-hybrid system using a mouse embryo library and reported as a male germ cell-specific 42-kDa protein, GKAP42.21 In our study, we cloned a human GKAP1 cDNA and demonstrated that GKAP1 mRNA expression was highest in the testis in normal adult tissues. GKAP1 functions as an anchoring protein for cGK-Iα and seems to be similar to AKAPs (A kinase anchoring protein),24 which binds to cAK via its regulatory subunits. We showed previously that AKAP3 (A kinase anchoring protein 3) was a CT-like antigen. AKAP3 mRNA expression is an independent and favorable prognosis factor in patients with poorly differentiated ovarian cancer.25
The CCDC62 gene consists of 13 exons and is located on chromosome 12q24.31. It has 2 splice variants of 2481 bp (CCDC62-1, NM_032573) and 3044 bp (CCDC62-2, NM_201435), which encode for proteins of 682 amino acids and 684 amino acids, respectively. RT-PCR analysis revealed that the expression of both variants was restricted to the testis in normal adult tissues. In tumors, no expression was observed with CCDC62-1 mRNA, but CCDC62-2 mRNA expression was observed in several types of tumors. The results of quantitative and qualitative RT-PCR on the same set of specimens suggested higher sensitivity of quantitative real-time RT-PCR than qualitative RT-PCR. It appeared to result in the discrepancy of the expression frequency. In a large scale ELISA survey of 191 sera samples from patients with several types of cancer, 13 patients produced antibody to CCDC62-2 protein. Western blot analysis revealed the reaction against the recombinant CCDC62-2 molecules.
Among CT antigens, NY-ESO-1 was shown to induce antibody responses in cancer patients. Around 50% of patients with melanoma expressing NY-ESO-1 mRNA showed antibody responses.26, 27 In the Japanese population, around 7% of bladder cancer patients and 4% of esophageal cancer patients showed NY-ESO-1 antibody responses.28, 29 In antibody positive patients, strong CD4 and CD8 responses against NY-ESO-1 were also elicited.30, 31 We have shown similar antibody responses in patients with XAGE-1b-expressing lung cancer.32 However, antibody responses were observed in less than 1% of patients against other CT antigens, e.g., MAGE-1.26 Our study demonstrated the immunogenicity of CCDC62-2 by examining serum reactivity by ELISA using recombinant protein. Strong reactions were observed in 5/76 (6.6%) sera from lung cancer patients and 6/104 (5.8%) sera from stomach cancer patients. None of the 41 sera samples from healthy donors was reactive. The significant frequency of antibody responses against CCDC62-2 suggested the strong immunogenicity and the CD4 and CD8 T-cell responses against the antigen should be investigated.
The expression frequency of tumor antigens, including CT antigens, is in a range between 5% and 40% depending on the cancer type,2, 3 and expression is heterogeneous with frequent antigen loss.33, 34 The development of polyvalent cancer vaccine containing epitopes derived from different CT antigens could overcome the disadvantage of these tumor antigen characteristics. We have isolated and identified several CT and CT-like antigens, such as OY-TES-1, XAGE-1, AKAP3, RFX4 and TSGA10.13, 20, 25, 35, 36 Our study added CCDC62-2 to the list of CT antigens for polyvalent cancer vaccines.
The authors thank Ms. M. Isobe for excellent technical assistance and Ms. J. Mizuuchi for preparation of the article. The authors also thank Mr. T. Iwasa (General Research Laboratory, Okayama University Medical School) for DNA sequencing.